From 91949d64548ead31df51c9fb6f7201ca8ffc9b51 Mon Sep 17 00:00:00 2001
From: Alexander Belyakov <abelyako@mail.ru>
Date: Sun, 4 May 2008 14:32:58 +0400
Subject: [PATCH 0001/1171] [MTD] [NOR] Remove cfi_cmdset_0001.c erase suspend
 fixup typo

Fix typo in erase suspend while write fixup code leading to compile time
error if CMDSET0001_DISABLE_ERASE_SUSPEND_ON_WRITE was defined.

drivers/mtd/chips/cfi_cmdset_0001.c: In function 'fixup_intel_strataflash':
drivers/mtd/chips/cfi_cmdset_0001.c:212: error: 'struct cfi_pri_amdstd' has no member named 'SuspendCmdSupport'

Signed-off-by: Alexander Belyakov <abelyako@googlemail.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/chips/cfi_cmdset_0001.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/chips/cfi_cmdset_0001.c b/drivers/mtd/chips/cfi_cmdset_0001.c
index fcd1aeccdf93..235830b9e994 100644
--- a/drivers/mtd/chips/cfi_cmdset_0001.c
+++ b/drivers/mtd/chips/cfi_cmdset_0001.c
@@ -204,7 +204,7 @@ static void fixup_intel_strataflash(struct mtd_info *mtd, void* param)
 {
 	struct map_info *map = mtd->priv;
 	struct cfi_private *cfi = map->fldrv_priv;
-	struct cfi_pri_amdstd *extp = cfi->cmdset_priv;
+	struct cfi_pri_intelext *extp = cfi->cmdset_priv;
 
 	printk(KERN_WARNING "cfi_cmdset_0001: Suspend "
 	                    "erase on write disabled.\n");

From deb1a5f1134e7da0e3dacd37b5d32b7fe0600a7f Mon Sep 17 00:00:00 2001
From: Nate Case <ncase@xes-inc.com>
Date: Tue, 13 May 2008 14:45:29 -0500
Subject: [PATCH 0002/1171] [MTD] [NOR] Support for M50FLW080A and M50FLW080B

Add support for M50FLW080A and M50FLW080B revisions of LPC flash
devices.

Signed-off-by: Aaron Lindner <alindner@xes-inc.com>
Signed-off-by: Nate Case <ncase@xes-inc.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/chips/cfi_cmdset_0001.c |  4 ++++
 drivers/mtd/chips/jedec_probe.c     | 32 +++++++++++++++++++++++++++++
 2 files changed, 36 insertions(+)

diff --git a/drivers/mtd/chips/cfi_cmdset_0001.c b/drivers/mtd/chips/cfi_cmdset_0001.c
index 235830b9e994..cc6b7bb6de02 100644
--- a/drivers/mtd/chips/cfi_cmdset_0001.c
+++ b/drivers/mtd/chips/cfi_cmdset_0001.c
@@ -50,6 +50,8 @@
 #define I82802AC	0x00ac
 #define MANUFACTURER_ST         0x0020
 #define M50LPW080       0x002F
+#define M50FLW080A	0x0080
+#define M50FLW080B	0x0081
 #define AT49BV640D	0x02de
 
 static int cfi_intelext_read (struct mtd_info *, loff_t, size_t, size_t *, u_char *);
@@ -301,6 +303,8 @@ static struct cfi_fixup jedec_fixup_table[] = {
 	{ MANUFACTURER_INTEL, I82802AB,   fixup_use_fwh_lock, NULL, },
 	{ MANUFACTURER_INTEL, I82802AC,   fixup_use_fwh_lock, NULL, },
 	{ MANUFACTURER_ST,    M50LPW080,  fixup_use_fwh_lock, NULL, },
+	{ MANUFACTURER_ST,    M50FLW080A, fixup_use_fwh_lock, NULL, },
+	{ MANUFACTURER_ST,    M50FLW080B, fixup_use_fwh_lock, NULL, },
 	{ 0, 0, NULL, NULL }
 };
 static struct cfi_fixup fixup_table[] = {
diff --git a/drivers/mtd/chips/jedec_probe.c b/drivers/mtd/chips/jedec_probe.c
index aa07575eb288..99f9ed21f660 100644
--- a/drivers/mtd/chips/jedec_probe.c
+++ b/drivers/mtd/chips/jedec_probe.c
@@ -141,6 +141,8 @@
 #define M50FW080	0x002D
 #define M50FW016	0x002E
 #define M50LPW080       0x002F
+#define M50FLW080A	0x0080
+#define M50FLW080B	0x0081
 
 /* SST */
 #define SST29EE020	0x0010
@@ -1590,6 +1592,36 @@ static const struct amd_flash_info jedec_table[] = {
 		.nr_regions	= 1,
 		.regions	= {
 			ERASEINFO(0x10000,16),
+		},
+	}, {
+		.mfr_id		= MANUFACTURER_ST,
+		.dev_id		= M50FLW080A,
+		.name		= "ST M50FLW080A",
+		.devtypes	= CFI_DEVICETYPE_X8,
+		.uaddr		= MTD_UADDR_UNNECESSARY,
+		.dev_size	= SIZE_1MiB,
+		.cmd_set	= P_ID_INTEL_EXT,
+		.nr_regions	= 4,
+		.regions	= {
+			ERASEINFO(0x1000,16),
+			ERASEINFO(0x10000,13),
+			ERASEINFO(0x1000,16),
+			ERASEINFO(0x1000,16),
+		}
+	}, {
+		.mfr_id		= MANUFACTURER_ST,
+		.dev_id		= M50FLW080B,
+		.name		= "ST M50FLW080B",
+		.devtypes	= CFI_DEVICETYPE_X8,
+		.uaddr		= MTD_UADDR_UNNECESSARY,
+		.dev_size	= SIZE_1MiB,
+		.cmd_set	= P_ID_INTEL_EXT,
+		.nr_regions	= 4,
+		.regions	= {
+			ERASEINFO(0x1000,16),
+			ERASEINFO(0x1000,16),
+			ERASEINFO(0x10000,13),
+			ERASEINFO(0x1000,16),
 		}
 	}, {
 		.mfr_id		= MANUFACTURER_TOSHIBA,

From a29ccf6f823a84d89e1c7aaaf221cf7282022024 Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw2@infradead.org>
Date: Tue, 3 Jun 2008 14:59:40 +0100
Subject: [PATCH 0003/1171] Make console charset translation optional

By turning off the new CONSOLE_TRANSLATIONS option and dropping the
associated code and tables from the kernel, we can save about 7KiB.

Taken from linux-tiny project by Tim Bird and mangled further by dwmw2.

Signed-off-by: Tim Bird <tim.bird@am.sony.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/char/Kconfig       |  8 ++++++++
 drivers/char/Makefile      |  4 ++--
 drivers/char/vt.c          |  2 +-
 include/linux/consolemap.h | 14 ++++++++++++++
 include/linux/vt_kern.h    | 19 +++++++++++++++++++
 5 files changed, 44 insertions(+), 3 deletions(-)

diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig
index 595a925c62a9..b7f7371dee73 100644
--- a/drivers/char/Kconfig
+++ b/drivers/char/Kconfig
@@ -36,6 +36,14 @@ config VT
 	  If unsure, say Y, or else you won't be able to do much with your new
 	  shiny Linux system :-)
 
+config CONSOLE_TRANSLATIONS
+	depends on VT
+	default y
+	bool "Enable character translations in console" if EMBEDDED
+	---help---
+	  This enables support for font mapping and Unicode translation
+	  on virtual consoles.
+
 config VT_CONSOLE
 	bool "Support for console on virtual terminal" if EMBEDDED
 	depends on VT
diff --git a/drivers/char/Makefile b/drivers/char/Makefile
index 4c1c584e9eb6..6ef173cab144 100644
--- a/drivers/char/Makefile
+++ b/drivers/char/Makefile
@@ -12,8 +12,8 @@ obj-y	 += mem.o random.o tty_io.o n_tty.o tty_ioctl.o
 obj-$(CONFIG_LEGACY_PTYS)	+= pty.o
 obj-$(CONFIG_UNIX98_PTYS)	+= pty.o
 obj-y				+= misc.o
-obj-$(CONFIG_VT)		+= vt_ioctl.o vc_screen.o consolemap.o \
-				   consolemap_deftbl.o selection.o keyboard.o
+obj-$(CONFIG_VT)		+= vt_ioctl.o vc_screen.o selection.o keyboard.o
+obj-$(CONFIG_CONSOLE_TRANSLATIONS) += consolemap.o consolemap_deftbl.o
 obj-$(CONFIG_HW_CONSOLE)	+= vt.o defkeymap.o
 obj-$(CONFIG_AUDIT)		+= tty_audit.o
 obj-$(CONFIG_MAGIC_SYSRQ)	+= sysrq.o
diff --git a/drivers/char/vt.c b/drivers/char/vt.c
index fa1ffbf2c621..18b7fb06dace 100644
--- a/drivers/char/vt.c
+++ b/drivers/char/vt.c
@@ -2208,7 +2208,7 @@ rescan_last_byte:
 			c = 0xfffd;
 		    tc = c;
 		} else {	/* no utf or alternate charset mode */
-		    tc = vc->vc_translate[vc->vc_toggle_meta ? (c | 0x80) : c];
+		    tc = vc_translate(vc, c);
 		}
 
 		param.c = tc;
diff --git a/include/linux/consolemap.h b/include/linux/consolemap.h
index e2bf7e5db39a..c4811da1338b 100644
--- a/include/linux/consolemap.h
+++ b/include/linux/consolemap.h
@@ -3,6 +3,9 @@
  *
  * Interface between console.c, selection.c  and consolemap.c
  */
+#ifndef __LINUX_CONSOLEMAP_H__
+#define __LINUX_CONSOLEMAP_H__
+
 #define LAT1_MAP 0
 #define GRAF_MAP 1
 #define IBMPC_MAP 2
@@ -10,6 +13,7 @@
 
 #include <linux/types.h>
 
+#ifdef CONFIG_CONSOLE_TRANSLATIONS
 struct vc_data;
 
 extern u16 inverse_translate(struct vc_data *conp, int glyph, int use_unicode);
@@ -18,3 +22,13 @@ extern int conv_uni_to_pc(struct vc_data *conp, long ucs);
 extern u32 conv_8bit_to_uni(unsigned char c);
 extern int conv_uni_to_8bit(u32 uni);
 void console_map_init(void);
+#else
+#define inverse_translate(conp, glyph, uni) ((uint16_t)glyph)
+#define set_translate(m, vc) ((unsigned short *)NULL)
+#define conv_uni_to_pc(conp, ucs) ((int) (ucs > 0xff ? -1: ucs))
+#define conv_8bit_to_uni(c) ((uint32_t)(c))
+#define conv_uni_to_8bit(c) ((int) ((c) & 0xff))
+#define console_map_init(c) do { ; } while (0)
+#endif /* CONFIG_CONSOLE_TRANSLATIONS */
+
+#endif /* __LINUX_CONSOLEMAP_H__ */
diff --git a/include/linux/vt_kern.h b/include/linux/vt_kern.h
index 9448ffbdcbf6..14c0e91be9b5 100644
--- a/include/linux/vt_kern.h
+++ b/include/linux/vt_kern.h
@@ -12,6 +12,7 @@
 #include <linux/mutex.h>
 #include <linux/console_struct.h>
 #include <linux/mm.h>
+#include <linux/consolemap.h>
 
 /*
  * Presently, a lot of graphics programs do not restore the contents of
@@ -54,6 +55,7 @@ void redraw_screen(struct vc_data *vc, int is_switch);
 struct tty_struct;
 int tioclinux(struct tty_struct *tty, unsigned long arg);
 
+#ifdef CONFIG_CONSOLE_TRANSLATIONS
 /* consolemap.c */
 
 struct unimapinit;
@@ -71,6 +73,23 @@ void con_free_unimap(struct vc_data *vc);
 void con_protect_unimap(struct vc_data *vc, int rdonly);
 int con_copy_unimap(struct vc_data *dst_vc, struct vc_data *src_vc);
 
+#define vc_translate(vc, c) ((vc)->vc_translate[(c) |			\
+					(vc)->vc_toggle_meta ? 0x80 : 0])
+#else
+#define con_set_trans_old(arg) (0)
+#define con_get_trans_old(arg) (-EINVAL)
+#define con_set_trans_new(arg) (0)
+#define con_get_trans_new(arg) (-EINVAL)
+#define con_clear_unimap(vc, ui) (0)
+#define con_set_unimap(vc, ct, list) (0)
+#define con_set_default_unimap(vc) (0)
+#define con_copy_unimap(d, s) (0)
+#define con_get_unimap(vc, ct, uct, list) (-EINVAL)
+#define con_free_unimap(vc) do { ; } while (0)
+
+#define vc_translate(vc, c) (c)
+#endif
+
 /* vt.c */
 int vt_waitactive(int vt);
 void change_console(struct vc_data *new_vc);

From 8fd310a1cc3aadb7a17d844beeefae66b1a169c6 Mon Sep 17 00:00:00 2001
From: Mike Rapoport <mike@compulab.co.il>
Date: Tue, 27 May 2008 11:19:57 +0300
Subject: [PATCH 0004/1171] [MTD] [NOR] Add support for AMD AM29SL800D[BT] NOR
 flash chips

Signed-off-by: Mike Rapoport <mike@compulab.co.il>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/chips/jedec_probe.c | 32 ++++++++++++++++++++++++++++++++
 1 file changed, 32 insertions(+)

diff --git a/drivers/mtd/chips/jedec_probe.c b/drivers/mtd/chips/jedec_probe.c
index 99f9ed21f660..b717f1a5d6b2 100644
--- a/drivers/mtd/chips/jedec_probe.c
+++ b/drivers/mtd/chips/jedec_probe.c
@@ -58,6 +58,8 @@
 #define AM29LV040B	0x004F
 #define AM29F032B	0x0041
 #define AM29F002T	0x00B0
+#define AM29SL800DB	0x226B
+#define AM29SL800DT	0x22EA
 
 /* Atmel */
 #define AT49BV512	0x0003
@@ -523,6 +525,36 @@ static const struct amd_flash_info jedec_table[] = {
 			ERASEINFO(0x02000,2),
 			ERASEINFO(0x04000,1),
 		}
+	}, {
+		.mfr_id		= MANUFACTURER_AMD,
+		.dev_id		= AM29SL800DT,
+		.name		= "AMD AM29SL800DT",
+		.devtypes	= CFI_DEVICETYPE_X16|CFI_DEVICETYPE_X8,
+		.uaddr		= MTD_UADDR_0x0AAA_0x0555,
+		.dev_size	= SIZE_1MiB,
+		.cmd_set	= P_ID_AMD_STD,
+		.nr_regions	= 4,
+		.regions	= {
+			ERASEINFO(0x10000,15),
+			ERASEINFO(0x08000,1),
+			ERASEINFO(0x02000,2),
+			ERASEINFO(0x04000,1),
+		}
+	}, {
+		.mfr_id		= MANUFACTURER_AMD,
+		.dev_id		= AM29SL800DB,
+		.name		= "AMD AM29SL800DB",
+		.devtypes	= CFI_DEVICETYPE_X16|CFI_DEVICETYPE_X8,
+		.uaddr		= MTD_UADDR_0x0AAA_0x0555,
+		.dev_size	= SIZE_1MiB,
+		.cmd_set	= P_ID_AMD_STD,
+		.nr_regions	= 4,
+		.regions	= {
+			ERASEINFO(0x04000,1),
+			ERASEINFO(0x02000,2),
+			ERASEINFO(0x08000,1),
+			ERASEINFO(0x10000,15),
+		}
 	}, {
 		.mfr_id		= MANUFACTURER_ATMEL,
 		.dev_id		= AT49BV512,

From 5c9c11e1c47c2101253a95c54ef72e13edcc728a Mon Sep 17 00:00:00 2001
From: Mike Rapoport <mike@compulab.co.il>
Date: Tue, 27 May 2008 11:20:03 +0300
Subject: [PATCH 0005/1171] [MTD] [NOR] Add support for flash chips with ID in
 bank other than 0

According to JEDEC "Standard Manufacturer's Identification Code"
(http://www.jedec.org/download/search/jep106W.pdf)
several first banks of NOR flash can contain 0x7f instead of actual ID.
This patch adds support for reading manufacturer ID from banks other than 0.

Signed-off-by: Mike Rapoport <mike@compulab.co.il>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/chips/jedec_probe.c | 19 ++++++++++++++++---
 1 file changed, 16 insertions(+), 3 deletions(-)

diff --git a/drivers/mtd/chips/jedec_probe.c b/drivers/mtd/chips/jedec_probe.c
index b717f1a5d6b2..279fe60b7855 100644
--- a/drivers/mtd/chips/jedec_probe.c
+++ b/drivers/mtd/chips/jedec_probe.c
@@ -37,6 +37,7 @@
 #define MANUFACTURER_ST		0x0020
 #define MANUFACTURER_TOSHIBA	0x0098
 #define MANUFACTURER_WINBOND	0x00da
+#define CONTINUATION_CODE	0x007f
 
 
 /* AMD */
@@ -1760,9 +1761,21 @@ static inline u32 jedec_read_mfr(struct map_info *map, uint32_t base,
 {
 	map_word result;
 	unsigned long mask;
-	u32 ofs = cfi_build_cmd_addr(0, cfi_interleave(cfi), cfi->device_type);
-	mask = (1 << (cfi->device_type * 8)) -1;
-	result = map_read(map, base + ofs);
+	int bank = 0;
+
+	/* According to JEDEC "Standard Manufacturer's Identification Code"
+	 * (http://www.jedec.org/download/search/jep106W.pdf)
+	 * several first banks can contain 0x7f instead of actual ID
+	 */
+	do {
+		uint32_t ofs = cfi_build_cmd_addr(0 + (bank << 8),
+						  cfi_interleave(cfi),
+						  cfi->device_type);
+		mask = (1 << (cfi->device_type * 8)) - 1;
+		result = map_read(map, base + ofs);
+		bank++;
+	} while ((result.x[0] & mask) == CONTINUATION_CODE);
+
 	return result.x[0] & mask;
 }
 

From 1b0b30acf3fc76e5d4d278fa5a8c9c6ac9898745 Mon Sep 17 00:00:00 2001
From: Mike Rapoport <mike@compulab.co.il>
Date: Tue, 27 May 2008 11:20:07 +0300
Subject: [PATCH 0006/1171] [MTD] [NOR] Add support for Eon EN29SL800B[BT] NOR
 flash chips

This patch add support for non-CFI Eon EN29SL800B[BT] NOR flash chips.
The Eon chips have manufacturer ID in the first bank, therefore this patch
depends on support for flash chips with ID in bank other than 0.

Signed-off-by: Mike Rapoport <mike@compulab.co.il>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/chips/jedec_probe.c | 35 +++++++++++++++++++++++++++++++++
 1 file changed, 35 insertions(+)

diff --git a/drivers/mtd/chips/jedec_probe.c b/drivers/mtd/chips/jedec_probe.c
index 279fe60b7855..b229da8060dd 100644
--- a/drivers/mtd/chips/jedec_probe.c
+++ b/drivers/mtd/chips/jedec_probe.c
@@ -26,6 +26,7 @@
 /* Manufacturers */
 #define MANUFACTURER_AMD	0x0001
 #define MANUFACTURER_ATMEL	0x001f
+#define MANUFACTURER_EON	0x001c
 #define MANUFACTURER_FUJITSU	0x0004
 #define MANUFACTURER_HYUNDAI	0x00AD
 #define MANUFACTURER_INTEL	0x0089
@@ -70,6 +71,10 @@
 #define AT49BV32X	0x00C8
 #define AT49BV32XT	0x00C9
 
+/* Eon */
+#define EN29SL800BB	0x226B
+#define EN29SL800BT	0x22EA
+
 /* Fujitsu */
 #define MBM29F040C	0x00A4
 #define MBM29F800BA	0x2258
@@ -633,6 +638,36 @@ static const struct amd_flash_info jedec_table[] = {
 			ERASEINFO(0x10000,63),
 			ERASEINFO(0x02000,8)
 		}
+	}, {
+		.mfr_id		= MANUFACTURER_EON,
+		.dev_id		= EN29SL800BT,
+		.name		= "Eon EN29SL800BT",
+		.devtypes	= CFI_DEVICETYPE_X16|CFI_DEVICETYPE_X8,
+		.uaddr		= MTD_UADDR_0x0AAA_0x0555,
+		.dev_size	= SIZE_1MiB,
+		.cmd_set	= P_ID_AMD_STD,
+		.nr_regions	= 4,
+		.regions	= {
+			ERASEINFO(0x10000,15),
+			ERASEINFO(0x08000,1),
+			ERASEINFO(0x02000,2),
+			ERASEINFO(0x04000,1),
+		}
+	}, {
+		.mfr_id		= MANUFACTURER_EON,
+		.dev_id		= EN29SL800BB,
+		.name		= "Eon EN29SL800BB",
+		.devtypes	= CFI_DEVICETYPE_X16|CFI_DEVICETYPE_X8,
+		.uaddr		= MTD_UADDR_0x0AAA_0x0555,
+		.dev_size	= SIZE_1MiB,
+		.cmd_set	= P_ID_AMD_STD,
+		.nr_regions	= 4,
+		.regions	= {
+			ERASEINFO(0x04000,1),
+			ERASEINFO(0x02000,2),
+			ERASEINFO(0x08000,1),
+			ERASEINFO(0x10000,15),
+		}
 	}, {
 		.mfr_id		= MANUFACTURER_FUJITSU,
 		.dev_id		= MBM29F040C,

From 437d0d299f0e08954c3cb680221d7e888e1a857f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Toralf=20F=C3=B6rster?= <toralf.foerster@gmx.de>
Date: Mon, 26 May 2008 20:35:46 +0200
Subject: [PATCH 0007/1171] [MTD] [NAND] fix 2 "unused variable" warnings in
 cafe_nand.c
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Toralf Förster <toralf.foerster@gmx.de>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/nand/cafe_nand.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/nand/cafe_nand.c b/drivers/mtd/nand/cafe_nand.c
index da6ceaa80ba1..95345d051579 100644
--- a/drivers/mtd/nand/cafe_nand.c
+++ b/drivers/mtd/nand/cafe_nand.c
@@ -626,10 +626,12 @@ static int __devinit cafe_nand_probe(struct pci_dev *pdev,
 {
 	struct mtd_info *mtd;
 	struct cafe_priv *cafe;
-	struct mtd_partition *parts;
 	uint32_t ctrl;
-	int nr_parts;
 	int err = 0;
+#ifdef CONFIG_MTD_PARTITIONS
+	struct mtd_partition *parts;
+	int nr_parts;
+#endif
 
 	/* Very old versions shared the same PCI ident for all three
 	   functions on the chip. Verify the class too... */

From ff0de61c3612410dc84a8de28761ef840d5d35ac Mon Sep 17 00:00:00 2001
From: "Robert P. J. Day" <rpjday@crashcourse.ca>
Date: Sun, 25 May 2008 07:35:17 -0400
Subject: [PATCH 0008/1171] [MTD] [NAND] excite_nandflash: simplify code using
 ARRAY_SIZE() macro.

Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/nand/excite_nandflash.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/nand/excite_nandflash.c b/drivers/mtd/nand/excite_nandflash.c
index bed87290decc..ced14b5294d5 100644
--- a/drivers/mtd/nand/excite_nandflash.c
+++ b/drivers/mtd/nand/excite_nandflash.c
@@ -209,7 +209,7 @@ static int __init excite_nand_probe(struct device *dev)
 	if (likely(!scan_res)) {
 		DEBUG(MTD_DEBUG_LEVEL2, "%s: register partitions\n", module_id);
 		add_mtd_partitions(&drvdata->board_mtd, partition_info,
-				   sizeof partition_info / sizeof partition_info[0]);
+				   ARRAY_SIZE(partition_info));
 	} else {
 		iounmap(drvdata->regs);
 		kfree(drvdata);

From 2606c79759e83fd8b1e45bc99b10e65a1dcf1602 Mon Sep 17 00:00:00 2001
From: "matthias@kaehlcke.net" <matthias@kaehlcke.net>
Date: Sat, 31 May 2008 15:28:09 +0200
Subject: [PATCH 0009/1171] [MTD] use list_for_each_entry() in add_mtd_device()

Signed-off-by: Matthias Kaehlcke <matthias@kaehlcke.net>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/mtdcore.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/mtd/mtdcore.c b/drivers/mtd/mtdcore.c
index f7e7890e5bc6..4c102f123463 100644
--- a/drivers/mtd/mtdcore.c
+++ b/drivers/mtd/mtdcore.c
@@ -53,7 +53,7 @@ int add_mtd_device(struct mtd_info *mtd)
 
 	for (i=0; i < MAX_MTD_DEVICES; i++)
 		if (!mtd_table[i]) {
-			struct list_head *this;
+			struct mtd_notifier *not;
 
 			mtd_table[i] = mtd;
 			mtd->index = i;
@@ -72,8 +72,7 @@ int add_mtd_device(struct mtd_info *mtd)
 			DEBUG(0, "mtd: Giving out device %d to %s\n",i, mtd->name);
 			/* No need to get a refcount on the module containing
 			   the notifier, since we hold the mtd_table_mutex */
-			list_for_each(this, &mtd_notifiers) {
-				struct mtd_notifier *not = list_entry(this, struct mtd_notifier, list);
+			list_for_each_entry(not, &mtd_notifiers, list) {
 				not->add(mtd);
 			}
 

From 856613c98c2f864994d5fb33a62b7a468f68ab9b Mon Sep 17 00:00:00 2001
From: "matthias@kaehlcke.net" <matthias@kaehlcke.net>
Date: Sat, 31 May 2008 15:28:10 +0200
Subject: [PATCH 0010/1171] [MTD] use list_for_each_entry() in del_mtd_device()

Signed-off-by: Matthias Kaehlcke <matthias@kaehlcke.net>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/mtdcore.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/mtd/mtdcore.c b/drivers/mtd/mtdcore.c
index 4c102f123463..8c61035b968b 100644
--- a/drivers/mtd/mtdcore.c
+++ b/drivers/mtd/mtdcore.c
@@ -112,12 +112,11 @@ int del_mtd_device (struct mtd_info *mtd)
 		       mtd->index, mtd->name, mtd->usecount);
 		ret = -EBUSY;
 	} else {
-		struct list_head *this;
+		struct mtd_notifier *not;
 
 		/* No need to get a refcount on the module containing
 		   the notifier, since we hold the mtd_table_mutex */
-		list_for_each(this, &mtd_notifiers) {
-			struct mtd_notifier *not = list_entry(this, struct mtd_notifier, list);
+		list_for_each_entry(not, &mtd_notifiers, list) {
 			not->remove(mtd);
 		}
 

From e9d42227bdc96238676bd28feca5815fcff2d6a8 Mon Sep 17 00:00:00 2001
From: Michael Hennerich <michael.hennerich@analog.com>
Date: Tue, 3 Jun 2008 12:26:05 +0800
Subject: [PATCH 0011/1171] [MTD] DataFlash: fix bug - ATMEL AT45DF321D spi
 flash card fails to be copied to (v2)

 - Add support for binary page size DataFlashes.
 - The driver now prints out pagesize and erasesize.
   Printout valuable information for creating flash filesystems.

Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/devices/mtd_dataflash.c | 117 ++++++++++++++++++++++++++--
 1 file changed, 112 insertions(+), 5 deletions(-)

diff --git a/drivers/mtd/devices/mtd_dataflash.c b/drivers/mtd/devices/mtd_dataflash.c
index b35e4813a3a5..acc3728872e3 100644
--- a/drivers/mtd/devices/mtd_dataflash.c
+++ b/drivers/mtd/devices/mtd_dataflash.c
@@ -487,7 +487,9 @@ add_dataflash(struct spi_device *spi, char *name,
 	device->write = dataflash_write;
 	device->priv = priv;
 
-	dev_info(&spi->dev, "%s (%d KBytes)\n", name, device->size/1024);
+	dev_info(&spi->dev, "%s (%d KBytes) pagesize %d bytes, "
+		"erasesize %d bytes\n", name, device->size/1024,
+		 pagesize, pagesize * 8);	/* 8 pages = 1 block */
 	dev_set_drvdata(&spi->dev, priv);
 
 	if (mtd_has_partitions()) {
@@ -521,7 +523,7 @@ add_dataflash(struct spi_device *spi, char *name,
  *
  *   Device      Density         ID code          #Pages PageSize  Offset
  *   AT45DB011B  1Mbit   (128K)  xx0011xx (0x0c)    512    264      9
- *   AT45DB021B  2Mbit   (256K)  xx0101xx (0x14)   1025    264      9
+ *   AT45DB021B  2Mbit   (256K)  xx0101xx (0x14)   1024    264      9
  *   AT45DB041B  4Mbit   (512K)  xx0111xx (0x1c)   2048    264      9
  *   AT45DB081B  8Mbit   (1M)    xx1001xx (0x24)   4096    264      9
  *   AT45DB0161B 16Mbit  (2M)    xx1011xx (0x2c)   4096    528     10
@@ -529,9 +531,114 @@ add_dataflash(struct spi_device *spi, char *name,
  *   AT45DB0642  64Mbit  (8M)    xx111xxx (0x3c)   8192   1056     11
  *   AT45DB1282  128Mbit (16M)   xx0100xx (0x10)  16384   1056     11
  */
+
+struct flash_info {
+	char		*name;
+
+	/* JEDEC id zero means "no ID" (most older chips); otherwise it has
+	 * a high byte of zero plus three data bytes: the manufacturer id,
+	 * then a two byte device id.
+	 */
+	u32		jedec_id;
+
+	/* The size listed here is what works with OPCODE_SE, which isn't
+	 * necessarily called a "sector" by the vendor.
+	 */
+	unsigned	nr_pages;
+	u16		pagesize;
+	u16		pageoffset;
+
+	u16		flags;
+#define	SUP_POW2PS	0x02
+#define	IS_POW2PS	0x01
+};
+
+static struct flash_info __devinitdata dataflash_data [] = {
+
+	{ "at45db011d",  0x1f2200, 512, 264, 9, SUP_POW2PS},
+	{ "at45db011d",  0x1f2200, 512, 256, 8, SUP_POW2PS | IS_POW2PS},
+
+	{ "at45db021d",  0x1f2300, 1024, 264, 9, SUP_POW2PS},
+	{ "at45db021d",  0x1f2300, 1024, 256, 8, SUP_POW2PS | IS_POW2PS},
+
+	{ "at45db041d",  0x1f2400, 2048, 264, 9, SUP_POW2PS},
+	{ "at45db041d",  0x1f2400, 2048, 256, 8, SUP_POW2PS | IS_POW2PS},
+
+	{ "at45db081d",  0x1f2500, 4096, 264, 9, SUP_POW2PS},
+	{ "at45db081d",  0x1f2500, 4096, 256, 8, SUP_POW2PS | IS_POW2PS},
+
+	{ "at45db161d",  0x1f2600, 4096, 528, 10, SUP_POW2PS},
+	{ "at45db161d",  0x1f2600, 4096, 512, 9, SUP_POW2PS | IS_POW2PS},
+
+	{ "at45db321c",  0x1f2700, 8192, 528, 10, },
+
+	{ "at45db321d",  0x1f2701, 8192, 528, 10, SUP_POW2PS},
+	{ "at45db321d",  0x1f2701, 8192, 512, 9, SUP_POW2PS | IS_POW2PS},
+
+	{ "at45db641d",  0x1f2800, 8192, 1056, 11, SUP_POW2PS},
+	{ "at45db641d",  0x1f2800, 8192, 1024, 10, SUP_POW2PS | IS_POW2PS},
+};
+
+static struct flash_info *__devinit jedec_probe(struct spi_device *spi)
+{
+	int			tmp;
+	u8			code = OP_READ_ID;
+	u8			id[3];
+	u32			jedec;
+	struct flash_info	*info;
+	int status;
+
+
+	/* JEDEC also defines an optional "extended device information"
+	 * string for after vendor-specific data, after the three bytes
+	 * we use here.  Supporting some chips might require using it.
+	 */
+	tmp = spi_write_then_read(spi, &code, 1, id, 3);
+	if (tmp < 0) {
+		DEBUG(MTD_DEBUG_LEVEL0, "%s: error %d reading JEDEC ID\n",
+			spi->dev.bus_id, tmp);
+		return NULL;
+	}
+	jedec = id[0];
+	jedec = jedec << 8;
+	jedec |= id[1];
+	jedec = jedec << 8;
+	jedec |= id[2];
+
+	for (tmp = 0, info = dataflash_data;
+			tmp < ARRAY_SIZE(dataflash_data);
+			tmp++, info++) {
+		if (info->jedec_id == jedec) {
+			if (info->flags & SUP_POW2PS) {
+				status = dataflash_status(spi);
+				if (status & 0x1)
+					/* return power of 2 pagesize */
+					return ++info;
+				else
+					return info;
+			}
+		}
+	}
+	return NULL;
+}
+
 static int __devinit dataflash_probe(struct spi_device *spi)
 {
 	int status;
+	struct flash_info	*info;
+
+	/*
+	 * Try to detect dataflash by JEDEC ID.
+	 * If it succeeds we know we have either a C or D part.
+	 * D will support power of 2 pagesize option.
+	 */
+
+	info = jedec_probe(spi);
+
+	if (info != NULL)
+		return add_dataflash(spi, info->name, info->nr_pages,
+				 info->pagesize, info->pageoffset);
+
 
 	status = dataflash_status(spi);
 	if (status <= 0 || status == 0xff) {
@@ -551,16 +658,16 @@ static int __devinit dataflash_probe(struct spi_device *spi)
 		status = add_dataflash(spi, "AT45DB011B", 512, 264, 9);
 		break;
 	case 0x14:	/* 0 1 0 1 x x */
-		status = add_dataflash(spi, "AT45DB021B", 1025, 264, 9);
+		status = add_dataflash(spi, "AT45DB021B", 1024, 264, 9);
 		break;
 	case 0x1c:	/* 0 1 1 1 x x */
-		status = add_dataflash(spi, "AT45DB041x", 2048, 264, 9);
+		status = add_dataflash(spi, "AT45DB041B", 2048, 264, 9);
 		break;
 	case 0x24:	/* 1 0 0 1 x x */
 		status = add_dataflash(spi, "AT45DB081B", 4096, 264, 9);
 		break;
 	case 0x2c:	/* 1 0 1 1 x x */
-		status = add_dataflash(spi, "AT45DB161x", 4096, 528, 10);
+		status = add_dataflash(spi, "AT45DB161B", 4096, 528, 10);
 		break;
 	case 0x34:	/* 1 1 0 1 x x */
 		status = add_dataflash(spi, "AT45DB321x", 8192, 528, 10);

From 271c5c59e00302494ffc299741e7fa2d63103e76 Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw2@infradead.org>
Date: Wed, 4 Jun 2008 17:43:22 +0100
Subject: [PATCH 0012/1171] [MTD] DataFlash: use proper types

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/devices/mtd_dataflash.c | 32 ++++++++++++++---------------
 1 file changed, 16 insertions(+), 16 deletions(-)

diff --git a/drivers/mtd/devices/mtd_dataflash.c b/drivers/mtd/devices/mtd_dataflash.c
index acc3728872e3..54e36bfc2c3b 100644
--- a/drivers/mtd/devices/mtd_dataflash.c
+++ b/drivers/mtd/devices/mtd_dataflash.c
@@ -82,7 +82,7 @@
 
 
 struct dataflash {
-	u8			command[4];
+	uint8_t			command[4];
 	char			name[24];
 
 	unsigned		partitioned:1;
@@ -150,7 +150,7 @@ static int dataflash_erase(struct mtd_info *mtd, struct erase_info *instr)
 	struct spi_transfer	x = { .tx_dma = 0, };
 	struct spi_message	msg;
 	unsigned		blocksize = priv->page_size << 3;
-	u8			*command;
+	uint8_t			*command;
 
 	DEBUG(MTD_DEBUG_LEVEL2, "%s: erase addr=0x%x len 0x%x\n",
 			spi->dev.bus_id,
@@ -182,8 +182,8 @@ static int dataflash_erase(struct mtd_info *mtd, struct erase_info *instr)
 		pageaddr = pageaddr << priv->page_offset;
 
 		command[0] = do_block ? OP_ERASE_BLOCK : OP_ERASE_PAGE;
-		command[1] = (u8)(pageaddr >> 16);
-		command[2] = (u8)(pageaddr >> 8);
+		command[1] = (uint8_t)(pageaddr >> 16);
+		command[2] = (uint8_t)(pageaddr >> 8);
 		command[3] = 0;
 
 		DEBUG(MTD_DEBUG_LEVEL3, "ERASE %s: (%x) %x %x %x [%i]\n",
@@ -234,7 +234,7 @@ static int dataflash_read(struct mtd_info *mtd, loff_t from, size_t len,
 	struct spi_transfer	x[2] = { { .tx_dma = 0, }, };
 	struct spi_message	msg;
 	unsigned int		addr;
-	u8			*command;
+	uint8_t			*command;
 	int			status;
 
 	DEBUG(MTD_DEBUG_LEVEL2, "%s: read 0x%x..0x%x\n",
@@ -274,9 +274,9 @@ static int dataflash_read(struct mtd_info *mtd, loff_t from, size_t len,
 	 * fewer "don't care" bytes.  Both buffers stay unchanged.
 	 */
 	command[0] = OP_READ_CONTINUOUS;
-	command[1] = (u8)(addr >> 16);
-	command[2] = (u8)(addr >> 8);
-	command[3] = (u8)(addr >> 0);
+	command[1] = (uint8_t)(addr >> 16);
+	command[2] = (uint8_t)(addr >> 8);
+	command[3] = (uint8_t)(addr >> 0);
 	/* plus 4 "don't care" bytes */
 
 	status = spi_sync(priv->spi, &msg);
@@ -311,7 +311,7 @@ static int dataflash_write(struct mtd_info *mtd, loff_t to, size_t len,
 	size_t			remaining = len;
 	u_char			*writebuf = (u_char *) buf;
 	int			status = -EINVAL;
-	u8			*command;
+	uint8_t			*command;
 
 	DEBUG(MTD_DEBUG_LEVEL2, "%s: write 0x%x..0x%x\n",
 		spi->dev.bus_id, (unsigned)to, (unsigned)(to + len));
@@ -539,16 +539,16 @@ struct flash_info {
 	 * a high byte of zero plus three data bytes: the manufacturer id,
 	 * then a two byte device id.
 	 */
-	u32		jedec_id;
+	uint32_t	jedec_id;
 
 	/* The size listed here is what works with OPCODE_SE, which isn't
 	 * necessarily called a "sector" by the vendor.
 	 */
 	unsigned	nr_pages;
-	u16		pagesize;
-	u16		pageoffset;
+	uint16_t	pagesize;
+	uint16_t	pageoffset;
 
-	u16		flags;
+	uint16_t	flags;
 #define	SUP_POW2PS	0x02
 #define	IS_POW2PS	0x01
 };
@@ -582,9 +582,9 @@ static struct flash_info __devinitdata dataflash_data [] = {
 static struct flash_info *__devinit jedec_probe(struct spi_device *spi)
 {
 	int			tmp;
-	u8			code = OP_READ_ID;
-	u8			id[3];
-	u32			jedec;
+	uint8_t			code = OP_READ_ID;
+	uint8_t			id[3];
+	uint32_t		jedec;
 	struct flash_info	*info;
 	int status;
 

From 83973b87938a06a2af7e2a7fd1b630c35f8baff4 Mon Sep 17 00:00:00 2001
From: Kyungmin Park <kyungmin.park@samsung.com>
Date: Thu, 29 May 2008 14:52:40 +0900
Subject: [PATCH 0013/1171] [MTD] [OneNAND] Check the ECC status first instead
 of controller

To get the correct information in case of power off recovery,
it should read ECC status first
Also remove previous workaround method.

Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/onenand/onenand_base.c | 54 ++++++++++++++----------------
 1 file changed, 25 insertions(+), 29 deletions(-)

diff --git a/drivers/mtd/onenand/onenand_base.c b/drivers/mtd/onenand/onenand_base.c
index 5d7965f7e9ce..926cf3a4135d 100644
--- a/drivers/mtd/onenand/onenand_base.c
+++ b/drivers/mtd/onenand/onenand_base.c
@@ -325,28 +325,11 @@ static int onenand_wait(struct mtd_info *mtd, int state)
 
 	ctrl = this->read_word(this->base + ONENAND_REG_CTRL_STATUS);
 
-	if (ctrl & ONENAND_CTRL_ERROR) {
-		printk(KERN_ERR "onenand_wait: controller error = 0x%04x\n", ctrl);
-		if (ctrl & ONENAND_CTRL_LOCK)
-			printk(KERN_ERR "onenand_wait: it's locked error.\n");
-		if (state == FL_READING) {
-			/*
-			 * A power loss while writing can result in a page
-			 * becoming unreadable.  When the device is mounted
-			 * again, reading that page gives controller errors.
-			 * Upper level software like JFFS2 treat -EIO as fatal,
-			 * refusing to mount at all.  That means it is necessary
-			 * to treat the error as an ECC error to allow recovery.
-			 * Note that typically in this case, the eraseblock can
-			 * still be erased and rewritten i.e. it has not become
-			 * a bad block.
-			 */
-			mtd->ecc_stats.failed++;
-			return -EBADMSG;
-		}
-		return -EIO;
-	}
-
+	/*
+	 * In the Spec. it checks the controller status first
+	 * However if you get the correct information in case of
+	 * power off recovery (POR) test, it should read ECC status first
+	 */
 	if (interrupt & ONENAND_INT_READ) {
 		int ecc = this->read_word(this->base + ONENAND_REG_ECC_STATUS);
 		if (ecc) {
@@ -364,6 +347,15 @@ static int onenand_wait(struct mtd_info *mtd, int state)
 		return -EIO;
 	}
 
+	/* If there's controller error, it's a real error */
+	if (ctrl & ONENAND_CTRL_ERROR) {
+		printk(KERN_ERR "onenand_wait: controller error = 0x%04x\n",
+			ctrl);
+		if (ctrl & ONENAND_CTRL_LOCK)
+			printk(KERN_ERR "onenand_wait: it's locked error.\n");
+		return -EIO;
+	}
+
 	return 0;
 }
 
@@ -1135,22 +1127,26 @@ static int onenand_bbt_wait(struct mtd_info *mtd, int state)
 	interrupt = this->read_word(this->base + ONENAND_REG_INTERRUPT);
 	ctrl = this->read_word(this->base + ONENAND_REG_CTRL_STATUS);
 
-	/* Initial bad block case: 0x2400 or 0x0400 */
-	if (ctrl & ONENAND_CTRL_ERROR) {
-		printk(KERN_DEBUG "onenand_bbt_wait: controller error = 0x%04x\n", ctrl);
-		return ONENAND_BBT_READ_ERROR;
-	}
-
 	if (interrupt & ONENAND_INT_READ) {
 		int ecc = this->read_word(this->base + ONENAND_REG_ECC_STATUS);
-		if (ecc & ONENAND_ECC_2BIT_ALL)
+		if (ecc & ONENAND_ECC_2BIT_ALL) {
+			printk(KERN_INFO "onenand_bbt_wait: ecc error = 0x%04x"
+				", controller error 0x%04x\n", ecc, ctrl);
 			return ONENAND_BBT_READ_ERROR;
+		}
 	} else {
 		printk(KERN_ERR "onenand_bbt_wait: read timeout!"
 			"ctrl=0x%04x intr=0x%04x\n", ctrl, interrupt);
 		return ONENAND_BBT_READ_FATAL_ERROR;
 	}
 
+	/* Initial bad block case: 0x2400 or 0x0400 */
+	if (ctrl & ONENAND_CTRL_ERROR) {
+		printk(KERN_DEBUG "onenand_bbt_wait: "
+			"controller error = 0x%04x\n", ctrl);
+		return ONENAND_BBT_READ_ERROR;
+	}
+
 	return 0;
 }
 

From af3deccfa67341a9df39b6f734afcc85998ad4b7 Mon Sep 17 00:00:00 2001
From: Adrian Hunter <ext-adrian.hunter@nokia.com>
Date: Fri, 30 May 2008 15:56:18 +0300
Subject: [PATCH 0014/1171] [MTD] [NAND] nandsim: fix size bug

Signed-off-by: Adrian Hunter <ext-adrian.hunter@nokia.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/nand/nandsim.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/nand/nandsim.c b/drivers/mtd/nand/nandsim.c
index bb885d1fcab5..b28d76013ab6 100644
--- a/drivers/mtd/nand/nandsim.c
+++ b/drivers/mtd/nand/nandsim.c
@@ -511,7 +511,7 @@ static int init_nandsim(struct mtd_info *mtd)
 	}
 
 	if (ns->options & OPT_SMALLPAGE) {
-		if (ns->geom.totsz < (32 << 20)) {
+		if (ns->geom.totsz <= (32 << 20)) {
 			ns->geom.pgaddrbytes  = 3;
 			ns->geom.secaddrbytes = 2;
 		} else {

From 07293b20083cb66df35bf2041f0c554eaac43e8c Mon Sep 17 00:00:00 2001
From: Adrian Hunter <ext-adrian.hunter@nokia.com>
Date: Fri, 30 May 2008 15:56:23 +0300
Subject: [PATCH 0015/1171] [MTD] [NAND] nandsim: fix overridesize

Signed-off-by: Adrian Hunter <ext-adrian.hunter@nokia.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/nand/nandsim.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/mtd/nand/nandsim.c b/drivers/mtd/nand/nandsim.c
index b28d76013ab6..b48051684733 100644
--- a/drivers/mtd/nand/nandsim.c
+++ b/drivers/mtd/nand/nandsim.c
@@ -2022,6 +2022,7 @@ static int __init ns_init_module(void)
 		nsmtd->size = new_size;
 		chip->chipsize = new_size;
 		chip->chip_shift = ffs(new_size) - 1;
+		chip->pagemask = (chip->chipsize >> chip->page_shift) - 1;
 	}
 
 	if ((retval = setup_wear_reporting(nsmtd)) != 0)

From 6eda7a55f786b75e7d3d636a9431e6c850b20d72 Mon Sep 17 00:00:00 2001
From: Adrian Hunter <ext-adrian.hunter@nokia.com>
Date: Fri, 30 May 2008 15:56:26 +0300
Subject: [PATCH 0016/1171] [MTD] [NAND] nandsim: allow for 64-bit size

Amend nandsim so that it does not assume 32-bit flash size.

Signed-off-by: Adrian Hunter <ext-adrian.hunter@nokia.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/nand/nandsim.c | 35 +++++++++++++++++++++--------------
 1 file changed, 21 insertions(+), 14 deletions(-)

diff --git a/drivers/mtd/nand/nandsim.c b/drivers/mtd/nand/nandsim.c
index b48051684733..68c150c8ff9d 100644
--- a/drivers/mtd/nand/nandsim.c
+++ b/drivers/mtd/nand/nandsim.c
@@ -298,11 +298,11 @@ struct nandsim {
 
 	/* NAND flash "geometry" */
 	struct nandsin_geometry {
-		uint32_t totsz;     /* total flash size, bytes */
+		uint64_t totsz;     /* total flash size, bytes */
 		uint32_t secsz;     /* flash sector (erase block) size, bytes */
 		uint pgsz;          /* NAND flash page size, bytes */
 		uint oobsz;         /* page OOB area size, bytes */
-		uint32_t totszoob;  /* total flash size including OOB, bytes */
+		uint64_t totszoob;  /* total flash size including OOB, bytes */
 		uint pgszoob;       /* page size including OOB , bytes*/
 		uint secszoob;      /* sector size including OOB, bytes */
 		uint pgnum;         /* total number of pages */
@@ -459,6 +459,12 @@ static char *get_partition_name(int i)
 	return kstrdup(buf, GFP_KERNEL);
 }
 
+static u_int64_t divide(u_int64_t n, u_int32_t d)
+{
+	do_div(n, d);
+	return n;
+}
+
 /*
  * Initialize the nandsim structure.
  *
@@ -469,8 +475,8 @@ static int init_nandsim(struct mtd_info *mtd)
 	struct nand_chip *chip = (struct nand_chip *)mtd->priv;
 	struct nandsim   *ns   = (struct nandsim *)(chip->priv);
 	int i, ret = 0;
-	u_int32_t remains;
-	u_int32_t next_offset;
+	u_int64_t remains;
+	u_int64_t next_offset;
 
 	if (NS_IS_INITIALIZED(ns)) {
 		NS_ERR("init_nandsim: nandsim is already initialized\n");
@@ -487,8 +493,8 @@ static int init_nandsim(struct mtd_info *mtd)
 	ns->geom.oobsz    = mtd->oobsize;
 	ns->geom.secsz    = mtd->erasesize;
 	ns->geom.pgszoob  = ns->geom.pgsz + ns->geom.oobsz;
-	ns->geom.pgnum    = ns->geom.totsz / ns->geom.pgsz;
-	ns->geom.totszoob = ns->geom.totsz + ns->geom.pgnum * ns->geom.oobsz;
+	ns->geom.pgnum    = divide(ns->geom.totsz, ns->geom.pgsz);
+	ns->geom.totszoob = ns->geom.totsz + (uint64_t)ns->geom.pgnum * ns->geom.oobsz;
 	ns->geom.secshift = ffs(ns->geom.secsz) - 1;
 	ns->geom.pgshift  = chip->page_shift;
 	ns->geom.oobshift = ffs(ns->geom.oobsz) - 1;
@@ -537,15 +543,16 @@ static int init_nandsim(struct mtd_info *mtd)
 	remains = ns->geom.totsz;
 	next_offset = 0;
 	for (i = 0; i < parts_num; ++i) {
-		unsigned long part = parts[i];
-		if (!part || part > remains / ns->geom.secsz) {
+		u_int64_t part_sz = (u_int64_t)parts[i] * ns->geom.secsz;
+
+		if (!part_sz || part_sz > remains) {
 			NS_ERR("bad partition size.\n");
 			ret = -EINVAL;
 			goto error;
 		}
 		ns->partitions[i].name   = get_partition_name(i);
 		ns->partitions[i].offset = next_offset;
-		ns->partitions[i].size   = part * ns->geom.secsz;
+		ns->partitions[i].size   = part_sz;
 		next_offset += ns->partitions[i].size;
 		remains -= ns->partitions[i].size;
 	}
@@ -573,7 +580,7 @@ static int init_nandsim(struct mtd_info *mtd)
 	if (ns->busw == 16)
 		NS_WARN("16-bit flashes support wasn't tested\n");
 
-	printk("flash size: %u MiB\n",          ns->geom.totsz >> 20);
+	printk("flash size: %llu MiB\n",        ns->geom.totsz >> 20);
 	printk("page size: %u bytes\n",         ns->geom.pgsz);
 	printk("OOB area size: %u bytes\n",     ns->geom.oobsz);
 	printk("sector size: %u KiB\n",         ns->geom.secsz >> 10);
@@ -583,7 +590,7 @@ static int init_nandsim(struct mtd_info *mtd)
 	printk("bits in sector size: %u\n",     ns->geom.secshift);
 	printk("bits in page size: %u\n",       ns->geom.pgshift);
 	printk("bits in OOB size: %u\n",        ns->geom.oobshift);
-	printk("flash size with OOB: %u KiB\n", ns->geom.totszoob >> 10);
+	printk("flash size with OOB: %llu KiB\n", ns->geom.totszoob >> 10);
 	printk("page address bytes: %u\n",      ns->geom.pgaddrbytes);
 	printk("sector address bytes: %u\n",    ns->geom.secaddrbytes);
 	printk("options: %#x\n",                ns->options);
@@ -825,7 +832,7 @@ static int setup_wear_reporting(struct mtd_info *mtd)
 
 	if (!rptwear)
 		return 0;
-	wear_eb_count = mtd->size / mtd->erasesize;
+	wear_eb_count = divide(mtd->size, mtd->erasesize);
 	mem = wear_eb_count * sizeof(unsigned long);
 	if (mem / sizeof(unsigned long) != wear_eb_count) {
 		NS_ERR("Too many erase blocks for wear reporting\n");
@@ -2013,7 +2020,7 @@ static int __init ns_init_module(void)
 	}
 
 	if (overridesize) {
-		u_int32_t new_size = nsmtd->erasesize << overridesize;
+		u_int64_t new_size = (u_int64_t)nsmtd->erasesize << overridesize;
 		if (new_size >> overridesize != nsmtd->erasesize) {
 			NS_ERR("overridesize is too big\n");
 			goto err_exit;
@@ -2021,7 +2028,7 @@ static int __init ns_init_module(void)
 		/* N.B. This relies on nand_scan not doing anything with the size before we change it */
 		nsmtd->size = new_size;
 		chip->chipsize = new_size;
-		chip->chip_shift = ffs(new_size) - 1;
+		chip->chip_shift = ffs(nsmtd->erasesize) + overridesize - 1;
 		chip->pagemask = (chip->chipsize >> chip->page_shift) - 1;
 	}
 

From 59018b6d2acabb114ab58637e6ab95ba424a89d0 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Tue, 20 May 2008 01:03:52 +0300
Subject: [PATCH 0017/1171] MTD/JFFS2: remove CVS keywords

Once upon a time, the MTD repository was using CVS.

This patch therefore removes all usages of the no longer updated CVS
keywords from the MTD code.

This also includes code that printed them to the user.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/Kconfig                 | 2 --
 drivers/mtd/afs.c                   | 2 --
 drivers/mtd/chips/cfi_cmdset_0001.c | 2 --
 drivers/mtd/chips/cfi_cmdset_0002.c | 3 ---
 drivers/mtd/chips/cfi_cmdset_0020.c | 2 --
 drivers/mtd/chips/cfi_probe.c       | 1 -
 drivers/mtd/chips/cfi_util.c        | 3 ---
 drivers/mtd/chips/chipreg.c         | 2 --
 drivers/mtd/chips/gen_probe.c       | 1 -
 drivers/mtd/chips/jedec_probe.c     | 1 -
 drivers/mtd/chips/map_absent.c      | 1 -
 drivers/mtd/chips/map_ram.c         | 1 -
 drivers/mtd/chips/map_rom.c         | 1 -
 drivers/mtd/cmdlinepart.c           | 2 --
 drivers/mtd/devices/Kconfig         | 1 -
 drivers/mtd/devices/Makefile        | 1 -
 drivers/mtd/devices/block2mtd.c     | 6 ------
 drivers/mtd/devices/doc2000.c       | 2 --
 drivers/mtd/devices/doc2001.c       | 2 --
 drivers/mtd/devices/doc2001plus.c   | 2 --
 drivers/mtd/devices/docecc.c        | 2 --
 drivers/mtd/devices/docprobe.c      | 3 ---
 drivers/mtd/devices/lart.c          | 2 --
 drivers/mtd/devices/ms02-nv.c       | 2 --
 drivers/mtd/devices/ms02-nv.h       | 2 --
 drivers/mtd/devices/mtdram.c        | 1 -
 drivers/mtd/devices/phram.c         | 2 --
 drivers/mtd/devices/pmc551.c        | 2 --
 drivers/mtd/devices/slram.c         | 2 --
 drivers/mtd/ftl.c                   | 3 ---
 drivers/mtd/inftlcore.c             | 5 -----
 drivers/mtd/inftlmount.c            | 4 ----
 drivers/mtd/maps/Kconfig            | 1 -
 drivers/mtd/maps/Makefile           | 1 -
 drivers/mtd/maps/amd76xrom.c        | 1 -
 drivers/mtd/maps/autcpu12-nvram.c   | 2 --
 drivers/mtd/maps/bast-flash.c       | 2 --
 drivers/mtd/maps/cdb89712.c         | 1 -
 drivers/mtd/maps/ceiva.c            | 1 -
 drivers/mtd/maps/cfi_flagadm.c      | 2 --
 drivers/mtd/maps/dbox2-flash.c      | 2 --
 drivers/mtd/maps/dc21285.c          | 2 --
 drivers/mtd/maps/dilnetpc.c         | 2 --
 drivers/mtd/maps/dmv182.c           | 2 --
 drivers/mtd/maps/ebony.c            | 2 --
 drivers/mtd/maps/edb7312.c          | 2 --
 drivers/mtd/maps/fortunet.c         | 1 -
 drivers/mtd/maps/h720x-flash.c      | 2 --
 drivers/mtd/maps/ichxrom.c          | 1 -
 drivers/mtd/maps/impa7.c            | 2 --
 drivers/mtd/maps/integrator-flash.c | 2 --
 drivers/mtd/maps/ipaq-flash.c       | 2 --
 drivers/mtd/maps/ixp2000.c          | 2 --
 drivers/mtd/maps/ixp4xx.c           | 2 --
 drivers/mtd/maps/l440gx.c           | 2 --
 drivers/mtd/maps/map_funcs.c        | 2 --
 drivers/mtd/maps/mbx860.c           | 2 --
 drivers/mtd/maps/mtx-1_flash.c      | 2 --
 drivers/mtd/maps/netsc520.c         | 2 --
 drivers/mtd/maps/nettel.c           | 2 --
 drivers/mtd/maps/octagon-5066.c     | 1 -
 drivers/mtd/maps/omap-toto-flash.c  | 2 --
 drivers/mtd/maps/pci.c              | 2 --
 drivers/mtd/maps/pcmciamtd.c        | 5 +----
 drivers/mtd/maps/physmap.c          | 2 --
 drivers/mtd/maps/plat-ram.c         | 2 --
 drivers/mtd/maps/redwood.c          | 2 --
 drivers/mtd/maps/rpxlite.c          | 2 --
 drivers/mtd/maps/sa1100-flash.c     | 2 --
 drivers/mtd/maps/sbc8240.c          | 3 ---
 drivers/mtd/maps/sbc_gxx.c          | 2 --
 drivers/mtd/maps/sc520cdp.c         | 2 --
 drivers/mtd/maps/scb2_flash.c       | 1 -
 drivers/mtd/maps/scx200_docflash.c  | 2 --
 drivers/mtd/maps/sharpsl-flash.c    | 2 --
 drivers/mtd/maps/solutionengine.c   | 2 --
 drivers/mtd/maps/sun_uflash.c       | 2 +-
 drivers/mtd/maps/tqm8xxl.c          | 2 --
 drivers/mtd/maps/ts5500_flash.c     | 2 --
 drivers/mtd/maps/tsunami_flash.c    | 1 -
 drivers/mtd/maps/uclinux.c          | 2 --
 drivers/mtd/maps/vmax301.c          | 1 -
 drivers/mtd/maps/walnut.c           | 2 --
 drivers/mtd/maps/wr_sbc82xx_flash.c | 2 --
 drivers/mtd/mtd_blkdevs.c           | 2 --
 drivers/mtd/mtdblock.c              | 2 --
 drivers/mtd/mtdblock_ro.c           | 2 --
 drivers/mtd/mtdchar.c               | 2 --
 drivers/mtd/mtdconcat.c             | 2 --
 drivers/mtd/mtdcore.c               | 2 --
 drivers/mtd/mtdpart.c               | 2 --
 drivers/mtd/nand/Kconfig            | 1 -
 drivers/mtd/nand/Makefile           | 1 -
 drivers/mtd/nand/au1550nd.c         | 2 --
 drivers/mtd/nand/autcpu12.c         | 2 --
 drivers/mtd/nand/diskonchip.c       | 2 --
 drivers/mtd/nand/edb7312.c          | 2 --
 drivers/mtd/nand/h1910.c            | 2 --
 drivers/mtd/nand/nand_bbt.c         | 2 --
 drivers/mtd/nand/nand_ecc.c         | 2 --
 drivers/mtd/nand/nand_ids.c         | 2 --
 drivers/mtd/nand/nandsim.c          | 2 --
 drivers/mtd/nand/ppchameleonevb.c   | 2 --
 drivers/mtd/nand/rtc_from4.c        | 2 --
 drivers/mtd/nand/s3c2410.c          | 2 --
 drivers/mtd/nand/sharpsl.c          | 2 --
 drivers/mtd/nand/spia.c             | 2 --
 drivers/mtd/nand/toto.c             | 2 --
 drivers/mtd/nand/ts7250.c           | 2 --
 drivers/mtd/nftlcore.c              | 5 -----
 drivers/mtd/nftlmount.c             | 4 ----
 drivers/mtd/redboot.c               | 2 --
 drivers/mtd/rfd_ftl.c               | 2 --
 include/linux/jffs2.h               | 3 ---
 include/linux/mtd/blktrans.h        | 2 --
 include/linux/mtd/cfi.h             | 1 -
 include/linux/mtd/cfi_endian.h      | 5 -----
 include/linux/mtd/concat.h          | 2 --
 include/linux/mtd/doc2000.h         | 2 --
 include/linux/mtd/flashchip.h       | 3 ---
 include/linux/mtd/ftl.h             | 2 --
 include/linux/mtd/gen_probe.h       | 1 -
 include/linux/mtd/inftl.h           | 4 ----
 include/linux/mtd/map.h             | 1 -
 include/linux/mtd/mtd.h             | 2 --
 include/linux/mtd/nand.h            | 2 --
 include/linux/mtd/nand_ecc.h        | 2 --
 include/linux/mtd/nftl.h            | 2 --
 include/linux/mtd/partitions.h      | 2 --
 include/linux/mtd/physmap.h         | 2 --
 include/linux/mtd/plat-ram.h        | 2 --
 include/linux/mtd/pmc551.h          | 4 +---
 include/linux/mtd/xip.h             | 2 --
 include/mtd/inftl-user.h            | 2 --
 include/mtd/jffs2-user.h            | 2 --
 include/mtd/mtd-abi.h               | 2 --
 include/mtd/mtd-user.h              | 2 --
 include/mtd/nftl-user.h             | 2 --
 138 files changed, 3 insertions(+), 279 deletions(-)

diff --git a/drivers/mtd/Kconfig b/drivers/mtd/Kconfig
index eed06d068fd1..14f11f8b9e5f 100644
--- a/drivers/mtd/Kconfig
+++ b/drivers/mtd/Kconfig
@@ -1,5 +1,3 @@
-# $Id: Kconfig,v 1.11 2005/11/07 11:14:19 gleixner Exp $
-
 menuconfig MTD
 	tristate "Memory Technology Device (MTD) support"
 	depends on HAS_IOMEM
diff --git a/drivers/mtd/afs.c b/drivers/mtd/afs.c
index 52d51eb91c16..d072ca5be689 100644
--- a/drivers/mtd/afs.c
+++ b/drivers/mtd/afs.c
@@ -21,8 +21,6 @@
    This is access code for flashes using ARM's flash partitioning
    standards.
 
-   $Id: afs.c,v 1.15 2005/11/07 11:14:19 gleixner Exp $
-
 ======================================================================*/
 
 #include <linux/module.h>
diff --git a/drivers/mtd/chips/cfi_cmdset_0001.c b/drivers/mtd/chips/cfi_cmdset_0001.c
index cc6b7bb6de02..324ff82a3cd9 100644
--- a/drivers/mtd/chips/cfi_cmdset_0001.c
+++ b/drivers/mtd/chips/cfi_cmdset_0001.c
@@ -4,8 +4,6 @@
  *
  * (C) 2000 Red Hat. GPL'd
  *
- * $Id: cfi_cmdset_0001.c,v 1.186 2005/11/23 22:07:52 nico Exp $
- *
  *
  * 10/10/2000	Nicolas Pitre <nico@cam.org>
  * 	- completely revamped method functions so they are aware and
diff --git a/drivers/mtd/chips/cfi_cmdset_0002.c b/drivers/mtd/chips/cfi_cmdset_0002.c
index f7fcc6389533..a972cc6be436 100644
--- a/drivers/mtd/chips/cfi_cmdset_0002.c
+++ b/drivers/mtd/chips/cfi_cmdset_0002.c
@@ -16,9 +16,6 @@
  * Occasionally maintained by Thayne Harbaugh tharbaugh at lnxi dot com
  *
  * This code is GPL
- *
- * $Id: cfi_cmdset_0002.c,v 1.122 2005/11/07 11:14:22 gleixner Exp $
- *
  */
 
 #include <linux/module.h>
diff --git a/drivers/mtd/chips/cfi_cmdset_0020.c b/drivers/mtd/chips/cfi_cmdset_0020.c
index 1b720cc571f3..d4714dd9f7ab 100644
--- a/drivers/mtd/chips/cfi_cmdset_0020.c
+++ b/drivers/mtd/chips/cfi_cmdset_0020.c
@@ -4,8 +4,6 @@
  *
  * (C) 2000 Red Hat. GPL'd
  *
- * $Id: cfi_cmdset_0020.c,v 1.22 2005/11/07 11:14:22 gleixner Exp $
- *
  * 10/10/2000	Nicolas Pitre <nico@cam.org>
  * 	- completely revamped method functions so they are aware and
  * 	  independent of the flash geometry (buswidth, interleave, etc.)
diff --git a/drivers/mtd/chips/cfi_probe.c b/drivers/mtd/chips/cfi_probe.c
index a4463a91ce31..c418e92e1d92 100644
--- a/drivers/mtd/chips/cfi_probe.c
+++ b/drivers/mtd/chips/cfi_probe.c
@@ -1,7 +1,6 @@
 /*
    Common Flash Interface probe code.
    (C) 2000 Red Hat. GPL'd.
-   $Id: cfi_probe.c,v 1.86 2005/11/29 14:48:31 gleixner Exp $
 */
 
 #include <linux/module.h>
diff --git a/drivers/mtd/chips/cfi_util.c b/drivers/mtd/chips/cfi_util.c
index 72e0022a47bf..0ee457018016 100644
--- a/drivers/mtd/chips/cfi_util.c
+++ b/drivers/mtd/chips/cfi_util.c
@@ -6,9 +6,6 @@
  * Copyright (C) 2003 STMicroelectronics Limited
  *
  * This code is covered by the GPL.
- *
- * $Id: cfi_util.c,v 1.10 2005/11/07 11:14:23 gleixner Exp $
- *
  */
 
 #include <linux/module.h>
diff --git a/drivers/mtd/chips/chipreg.c b/drivers/mtd/chips/chipreg.c
index 2174c97549f0..c85760968227 100644
--- a/drivers/mtd/chips/chipreg.c
+++ b/drivers/mtd/chips/chipreg.c
@@ -1,6 +1,4 @@
 /*
- * $Id: chipreg.c,v 1.17 2004/11/16 18:29:00 dwmw2 Exp $
- *
  * Registration for chip drivers
  *
  */
diff --git a/drivers/mtd/chips/gen_probe.c b/drivers/mtd/chips/gen_probe.c
index d338b8c92780..e53a58ae384f 100644
--- a/drivers/mtd/chips/gen_probe.c
+++ b/drivers/mtd/chips/gen_probe.c
@@ -2,7 +2,6 @@
  * Routines common to all CFI-type probes.
  * (C) 2001-2003 Red Hat, Inc.
  * GPL'd
- * $Id: gen_probe.c,v 1.24 2005/11/07 11:14:23 gleixner Exp $
  */
 
 #include <linux/kernel.h>
diff --git a/drivers/mtd/chips/jedec_probe.c b/drivers/mtd/chips/jedec_probe.c
index b229da8060dd..afb35e3a3cee 100644
--- a/drivers/mtd/chips/jedec_probe.c
+++ b/drivers/mtd/chips/jedec_probe.c
@@ -1,7 +1,6 @@
 /*
    Common Flash Interface probe code.
    (C) 2000 Red Hat. GPL'd.
-   $Id: jedec_probe.c,v 1.66 2005/11/07 11:14:23 gleixner Exp $
    See JEDEC (http://www.jedec.org/) standard JESD21C (section 3.5)
    for the standard this probe goes back to.
 
diff --git a/drivers/mtd/chips/map_absent.c b/drivers/mtd/chips/map_absent.c
index fc478c0f93f5..494d30d0631a 100644
--- a/drivers/mtd/chips/map_absent.c
+++ b/drivers/mtd/chips/map_absent.c
@@ -1,7 +1,6 @@
 /*
  * Common code to handle absent "placeholder" devices
  * Copyright 2001 Resilience Corporation <ebrower@resilience.com>
- * $Id: map_absent.c,v 1.6 2005/11/07 11:14:23 gleixner Exp $
  *
  * This map driver is used to allocate "placeholder" MTD
  * devices on systems that have socketed/removable media.
diff --git a/drivers/mtd/chips/map_ram.c b/drivers/mtd/chips/map_ram.c
index 5cb6d5263661..072dd8abf33a 100644
--- a/drivers/mtd/chips/map_ram.c
+++ b/drivers/mtd/chips/map_ram.c
@@ -1,7 +1,6 @@
 /*
  * Common code to handle map devices which are simple RAM
  * (C) 2000 Red Hat. GPL'd.
- * $Id: map_ram.c,v 1.22 2005/01/05 18:05:12 dwmw2 Exp $
  */
 
 #include <linux/module.h>
diff --git a/drivers/mtd/chips/map_rom.c b/drivers/mtd/chips/map_rom.c
index cb27f855074c..821d0ed6bae3 100644
--- a/drivers/mtd/chips/map_rom.c
+++ b/drivers/mtd/chips/map_rom.c
@@ -1,7 +1,6 @@
 /*
  * Common code to handle map devices which are simple ROM
  * (C) 2000 Red Hat. GPL'd.
- * $Id: map_rom.c,v 1.23 2005/01/05 18:05:12 dwmw2 Exp $
  */
 
 #include <linux/module.h>
diff --git a/drivers/mtd/cmdlinepart.c b/drivers/mtd/cmdlinepart.c
index e472a0e9de9d..68782ab2f0de 100644
--- a/drivers/mtd/cmdlinepart.c
+++ b/drivers/mtd/cmdlinepart.c
@@ -1,6 +1,4 @@
 /*
- * $Id: cmdlinepart.c,v 1.19 2005/11/07 11:14:19 gleixner Exp $
- *
  * Read flash partition table from command line
  *
  * Copyright 2002 SYSGO Real-Time Solutions GmbH
diff --git a/drivers/mtd/devices/Kconfig b/drivers/mtd/devices/Kconfig
index 35ed1103dbb2..9c613f06623c 100644
--- a/drivers/mtd/devices/Kconfig
+++ b/drivers/mtd/devices/Kconfig
@@ -1,5 +1,4 @@
 # drivers/mtd/maps/Kconfig
-# $Id: Kconfig,v 1.18 2005/11/07 11:14:24 gleixner Exp $
 
 menu "Self-contained MTD device drivers"
 	depends on MTD!=n
diff --git a/drivers/mtd/devices/Makefile b/drivers/mtd/devices/Makefile
index 0f788d5c4bf8..0993d5cf3923 100644
--- a/drivers/mtd/devices/Makefile
+++ b/drivers/mtd/devices/Makefile
@@ -1,7 +1,6 @@
 #
 # linux/drivers/devices/Makefile
 #
-# $Id: Makefile.common,v 1.7 2004/12/22 17:51:15 joern Exp $
 
 obj-$(CONFIG_MTD_DOC2000)	+= doc2000.o
 obj-$(CONFIG_MTD_DOC2001)	+= doc2001.o
diff --git a/drivers/mtd/devices/block2mtd.c b/drivers/mtd/devices/block2mtd.c
index 519d942e7940..303ea9b8cfe4 100644
--- a/drivers/mtd/devices/block2mtd.c
+++ b/drivers/mtd/devices/block2mtd.c
@@ -1,6 +1,4 @@
 /*
- * $Id: block2mtd.c,v 1.30 2005/11/29 14:48:32 gleixner Exp $
- *
  * block2mtd.c - create an mtd from a block device
  *
  * Copyright (C) 2001,2002	Simon Evans <spse@secret.org.uk>
@@ -20,9 +18,6 @@
 #include <linux/mutex.h>
 #include <linux/mount.h>
 
-#define VERSION "$Revision: 1.30 $"
-
-
 #define ERROR(fmt, args...) printk(KERN_ERR "block2mtd: " fmt "\n" , ## args)
 #define INFO(fmt, args...) printk(KERN_INFO "block2mtd: " fmt "\n" , ## args)
 
@@ -451,7 +446,6 @@ MODULE_PARM_DESC(block2mtd, "Device to use. \"block2mtd=<dev>[,<erasesize>]\"");
 static int __init block2mtd_init(void)
 {
 	int ret = 0;
-	INFO("version " VERSION);
 
 #ifndef MODULE
 	if (strlen(block2mtd_paramline))
diff --git a/drivers/mtd/devices/doc2000.c b/drivers/mtd/devices/doc2000.c
index 846989f292e3..50de839c77a9 100644
--- a/drivers/mtd/devices/doc2000.c
+++ b/drivers/mtd/devices/doc2000.c
@@ -3,8 +3,6 @@
  * Linux driver for Disk-On-Chip 2000 and Millennium
  * (c) 1999 Machine Vision Holdings, Inc.
  * (c) 1999, 2000 David Woodhouse <dwmw2@infradead.org>
- *
- * $Id: doc2000.c,v 1.67 2005/11/07 11:14:24 gleixner Exp $
  */
 
 #include <linux/kernel.h>
diff --git a/drivers/mtd/devices/doc2001.c b/drivers/mtd/devices/doc2001.c
index 6413efc045e0..e32c568c1145 100644
--- a/drivers/mtd/devices/doc2001.c
+++ b/drivers/mtd/devices/doc2001.c
@@ -3,8 +3,6 @@
  * Linux driver for Disk-On-Chip Millennium
  * (c) 1999 Machine Vision Holdings, Inc.
  * (c) 1999, 2000 David Woodhouse <dwmw2@infradead.org>
- *
- * $Id: doc2001.c,v 1.49 2005/11/07 11:14:24 gleixner Exp $
  */
 
 #include <linux/kernel.h>
diff --git a/drivers/mtd/devices/doc2001plus.c b/drivers/mtd/devices/doc2001plus.c
index 83be3461658f..d853f891b586 100644
--- a/drivers/mtd/devices/doc2001plus.c
+++ b/drivers/mtd/devices/doc2001plus.c
@@ -6,8 +6,6 @@
  * (c) 1999 Machine Vision Holdings, Inc.
  * (c) 1999, 2000 David Woodhouse <dwmw2@infradead.org>
  *
- * $Id: doc2001plus.c,v 1.14 2005/11/07 11:14:24 gleixner Exp $
- *
  * Released under GPL
  */
 
diff --git a/drivers/mtd/devices/docecc.c b/drivers/mtd/devices/docecc.c
index fd8a8daba3a8..874e51b110a2 100644
--- a/drivers/mtd/devices/docecc.c
+++ b/drivers/mtd/devices/docecc.c
@@ -7,8 +7,6 @@
  * Author: Fabrice Bellard (fabrice.bellard@netgem.com)
  * Copyright (C) 2000 Netgem S.A.
  *
- * $Id: docecc.c,v 1.7 2005/11/07 11:14:25 gleixner Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
diff --git a/drivers/mtd/devices/docprobe.c b/drivers/mtd/devices/docprobe.c
index d8cc94ec4e50..6e5d811ae83a 100644
--- a/drivers/mtd/devices/docprobe.c
+++ b/drivers/mtd/devices/docprobe.c
@@ -4,9 +4,6 @@
 /* (C) 1999 Machine Vision Holdings, Inc.			*/
 /* (C) 1999-2003 David Woodhouse <dwmw2@infradead.org>		*/
 
-/* $Id: docprobe.c,v 1.46 2005/11/07 11:14:25 gleixner Exp $	*/
-
-
 
 /* DOC_PASSIVE_PROBE:
    In order to ensure that the BIOS checksum is correct at boot time, and
diff --git a/drivers/mtd/devices/lart.c b/drivers/mtd/devices/lart.c
index 1d324e5c412d..f4bda4cee495 100644
--- a/drivers/mtd/devices/lart.c
+++ b/drivers/mtd/devices/lart.c
@@ -2,8 +2,6 @@
 /*
  * MTD driver for the 28F160F3 Flash Memory (non-CFI) on LART.
  *
- * $Id: lart.c,v 1.9 2005/11/07 11:14:25 gleixner Exp $
- *
  * Author: Abraham vd Merwe <abraham@2d3d.co.za>
  *
  * Copyright (c) 2001, 2d3D, Inc.
diff --git a/drivers/mtd/devices/ms02-nv.c b/drivers/mtd/devices/ms02-nv.c
index 9cff119a2024..6a9a24a80a6d 100644
--- a/drivers/mtd/devices/ms02-nv.c
+++ b/drivers/mtd/devices/ms02-nv.c
@@ -5,8 +5,6 @@
  *	modify it under the terms of the GNU General Public License
  *	as published by the Free Software Foundation; either version
  *	2 of the License, or (at your option) any later version.
- *
- *	$Id: ms02-nv.c,v 1.11 2005/11/14 13:41:47 macro Exp $
  */
 
 #include <linux/init.h>
diff --git a/drivers/mtd/devices/ms02-nv.h b/drivers/mtd/devices/ms02-nv.h
index 8a6eef7cfee3..04deafd3a771 100644
--- a/drivers/mtd/devices/ms02-nv.h
+++ b/drivers/mtd/devices/ms02-nv.h
@@ -9,8 +9,6 @@
  *	modify it under the terms of the GNU General Public License
  *	as published by the Free Software Foundation; either version
  *	2 of the License, or (at your option) any later version.
- *
- *	$Id: ms02-nv.h,v 1.3 2003/08/19 09:25:36 dwmw2 Exp $
  */
 
 #include <linux/ioport.h>
diff --git a/drivers/mtd/devices/mtdram.c b/drivers/mtd/devices/mtdram.c
index 0399be178620..3aaca88847d3 100644
--- a/drivers/mtd/devices/mtdram.c
+++ b/drivers/mtd/devices/mtdram.c
@@ -1,6 +1,5 @@
 /*
  * mtdram - a test mtd device
- * $Id: mtdram.c,v 1.37 2005/04/21 03:42:11 joern Exp $
  * Author: Alexander Larsson <alex@cendio.se>
  *
  * Copyright (c) 1999 Alexander Larsson <alex@cendio.se>
diff --git a/drivers/mtd/devices/phram.c b/drivers/mtd/devices/phram.c
index c7987b1c5e01..088fbb7595b5 100644
--- a/drivers/mtd/devices/phram.c
+++ b/drivers/mtd/devices/phram.c
@@ -1,6 +1,4 @@
 /**
- * $Id: phram.c,v 1.16 2005/11/07 11:14:25 gleixner Exp $
- *
  * Copyright (c) ????		Jochen Schäuble <psionic@psionic.de>
  * Copyright (c) 2003-2004	Joern Engel <joern@wh.fh-wedel.de>
  *
diff --git a/drivers/mtd/devices/pmc551.c b/drivers/mtd/devices/pmc551.c
index bc9981749064..d38bca64bb15 100644
--- a/drivers/mtd/devices/pmc551.c
+++ b/drivers/mtd/devices/pmc551.c
@@ -1,6 +1,4 @@
 /*
- * $Id: pmc551.c,v 1.32 2005/11/07 11:14:25 gleixner Exp $
- *
  * PMC551 PCI Mezzanine Ram Device
  *
  * Author:
diff --git a/drivers/mtd/devices/slram.c b/drivers/mtd/devices/slram.c
index cb86db746f28..a425d09f35a0 100644
--- a/drivers/mtd/devices/slram.c
+++ b/drivers/mtd/devices/slram.c
@@ -1,7 +1,5 @@
 /*======================================================================
 
-  $Id: slram.c,v 1.36 2005/11/07 11:14:25 gleixner Exp $
-
   This driver provides a method to access memory not used by the kernel
   itself (i.e. if the kernel commandline mem=xxx is used). To actually
   use slram at least mtdblock or mtdchar is required (for block or
diff --git a/drivers/mtd/ftl.c b/drivers/mtd/ftl.c
index 4a79b187b568..3fed8f94ac6f 100644
--- a/drivers/mtd/ftl.c
+++ b/drivers/mtd/ftl.c
@@ -1,5 +1,4 @@
 /* This version ported to the Linux-MTD system by dwmw2@infradead.org
- * $Id: ftl.c,v 1.59 2005/11/29 14:48:31 gleixner Exp $
  *
  * Fixes: Arnaldo Carvalho de Melo <acme@conectiva.com.br>
  * - fixes some leaks on failure in build_maps and ftl_notify_add, cleanups
@@ -1082,8 +1081,6 @@ static struct mtd_blktrans_ops ftl_tr = {
 
 static int init_ftl(void)
 {
-	DEBUG(0, "$Id: ftl.c,v 1.59 2005/11/29 14:48:31 gleixner Exp $\n");
-
 	return register_mtd_blktrans(&ftl_tr);
 }
 
diff --git a/drivers/mtd/inftlcore.c b/drivers/mtd/inftlcore.c
index b0e396504e67..c4f9d3378b24 100644
--- a/drivers/mtd/inftlcore.c
+++ b/drivers/mtd/inftlcore.c
@@ -7,8 +7,6 @@
  * (c) 1999 Machine Vision Holdings, Inc.
  * Author: David Woodhouse <dwmw2@infradead.org>
  *
- * $Id: inftlcore.c,v 1.19 2005/11/07 11:14:20 gleixner Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
@@ -953,9 +951,6 @@ static struct mtd_blktrans_ops inftl_tr = {
 
 static int __init init_inftl(void)
 {
-	printk(KERN_INFO "INFTL: inftlcore.c $Revision: 1.19 $, "
-		"inftlmount.c %s\n", inftlmountrev);
-
 	return register_mtd_blktrans(&inftl_tr);
 }
 
diff --git a/drivers/mtd/inftlmount.c b/drivers/mtd/inftlmount.c
index c551d2f0779c..9113628ed1ef 100644
--- a/drivers/mtd/inftlmount.c
+++ b/drivers/mtd/inftlmount.c
@@ -8,8 +8,6 @@
  * Author: Fabrice Bellard (fabrice.bellard@netgem.com)
  * Copyright (C) 2000 Netgem S.A.
  *
- * $Id: inftlmount.c,v 1.18 2005/11/07 11:14:20 gleixner Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
@@ -39,8 +37,6 @@
 #include <linux/mtd/inftl.h>
 #include <linux/mtd/compatmac.h>
 
-char inftlmountrev[]="$Revision: 1.18 $";
-
 /*
  * find_boot_record: Find the INFTL Media Header and its Spare copy which
  *	contains the various device information of the INFTL partition and
diff --git a/drivers/mtd/maps/Kconfig b/drivers/mtd/maps/Kconfig
index 1bd69aa9e22a..b2c180245618 100644
--- a/drivers/mtd/maps/Kconfig
+++ b/drivers/mtd/maps/Kconfig
@@ -1,5 +1,4 @@
 # drivers/mtd/maps/Kconfig
-# $Id: Kconfig,v 1.61 2005/11/07 11:14:26 gleixner Exp $
 
 menu "Mapping drivers for chip access"
 	depends on MTD!=n
diff --git a/drivers/mtd/maps/Makefile b/drivers/mtd/maps/Makefile
index a9cbe80f99a0..5444eaf4026f 100644
--- a/drivers/mtd/maps/Makefile
+++ b/drivers/mtd/maps/Makefile
@@ -1,7 +1,6 @@
 #
 # linux/drivers/maps/Makefile
 #
-# $Id: Makefile.common,v 1.34 2005/11/07 11:14:26 gleixner Exp $
 
 ifeq ($(CONFIG_MTD_COMPLEX_MAPPINGS),y)
 obj-$(CONFIG_MTD)		+= map_funcs.o
diff --git a/drivers/mtd/maps/amd76xrom.c b/drivers/mtd/maps/amd76xrom.c
index 728aed6ad722..948b86f35ef4 100644
--- a/drivers/mtd/maps/amd76xrom.c
+++ b/drivers/mtd/maps/amd76xrom.c
@@ -2,7 +2,6 @@
  * amd76xrom.c
  *
  * Normal mappings of chips in physical memory
- * $Id: amd76xrom.c,v 1.21 2005/11/07 11:14:26 gleixner Exp $
  */
 
 #include <linux/module.h>
diff --git a/drivers/mtd/maps/autcpu12-nvram.c b/drivers/mtd/maps/autcpu12-nvram.c
index 7ed3424dd959..cf32267263df 100644
--- a/drivers/mtd/maps/autcpu12-nvram.c
+++ b/drivers/mtd/maps/autcpu12-nvram.c
@@ -2,8 +2,6 @@
  * NV-RAM memory access on autcpu12
  * (C) 2002 Thomas Gleixner (gleixner@autronix.de)
  *
- * $Id: autcpu12-nvram.c,v 1.9 2005/11/07 11:14:26 gleixner Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
diff --git a/drivers/mtd/maps/bast-flash.c b/drivers/mtd/maps/bast-flash.c
index 1f492062f8ca..ca5414880341 100644
--- a/drivers/mtd/maps/bast-flash.c
+++ b/drivers/mtd/maps/bast-flash.c
@@ -9,8 +9,6 @@
  *	20-Sep-2004  BJD  Initial version
  *	17-Jan-2005  BJD  Add whole device if no partitions found
  *
- * $Id: bast-flash.c,v 1.5 2005/11/07 11:14:26 gleixner Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
diff --git a/drivers/mtd/maps/cdb89712.c b/drivers/mtd/maps/cdb89712.c
index 9f17bb6c5a9d..cb507da0a87d 100644
--- a/drivers/mtd/maps/cdb89712.c
+++ b/drivers/mtd/maps/cdb89712.c
@@ -1,7 +1,6 @@
 /*
  * Flash on Cirrus CDB89712
  *
- * $Id: cdb89712.c,v 1.11 2005/11/07 11:14:26 gleixner Exp $
  */
 
 #include <linux/module.h>
diff --git a/drivers/mtd/maps/ceiva.c b/drivers/mtd/maps/ceiva.c
index 629e6e2641a8..6464d487eb1a 100644
--- a/drivers/mtd/maps/ceiva.c
+++ b/drivers/mtd/maps/ceiva.c
@@ -11,7 +11,6 @@
  *
  * (C) 2000 Nicolas Pitre <nico@cam.org>
  *
- * $Id: ceiva.c,v 1.11 2004/09/16 23:27:12 gleixner Exp $
  */
 
 #include <linux/module.h>
diff --git a/drivers/mtd/maps/cfi_flagadm.c b/drivers/mtd/maps/cfi_flagadm.c
index 65e5ee552010..0ecc3f6d735b 100644
--- a/drivers/mtd/maps/cfi_flagadm.c
+++ b/drivers/mtd/maps/cfi_flagadm.c
@@ -1,8 +1,6 @@
 /*
  *  Copyright © 2001 Flaga hf. Medical Devices, Kári Davíðsson <kd@flaga.is>
  *
- *  $Id: cfi_flagadm.c,v 1.15 2005/11/07 11:14:26 gleixner Exp $
- *
  *  This program is free software; you can redistribute  it and/or modify it
  *  under  the terms of  the GNU General  Public License as published by the
  *  Free Software Foundation;  either version 2 of the  License, or (at your
diff --git a/drivers/mtd/maps/dbox2-flash.c b/drivers/mtd/maps/dbox2-flash.c
index 92a9c7fac993..e115667bf1d0 100644
--- a/drivers/mtd/maps/dbox2-flash.c
+++ b/drivers/mtd/maps/dbox2-flash.c
@@ -1,6 +1,4 @@
 /*
- * $Id: dbox2-flash.c,v 1.14 2005/11/07 11:14:26 gleixner Exp $
- *
  * D-Box 2 flash driver
  */
 
diff --git a/drivers/mtd/maps/dc21285.c b/drivers/mtd/maps/dc21285.c
index b32bb9347d71..3aa018c092f8 100644
--- a/drivers/mtd/maps/dc21285.c
+++ b/drivers/mtd/maps/dc21285.c
@@ -4,8 +4,6 @@
  * (C) 2000  Nicolas Pitre <nico@cam.org>
  *
  * This code is GPL
- *
- * $Id: dc21285.c,v 1.24 2005/11/07 11:14:26 gleixner Exp $
  */
 #include <linux/module.h>
 #include <linux/types.h>
diff --git a/drivers/mtd/maps/dilnetpc.c b/drivers/mtd/maps/dilnetpc.c
index 1c3b34ad7325..0713e3a5a22c 100644
--- a/drivers/mtd/maps/dilnetpc.c
+++ b/drivers/mtd/maps/dilnetpc.c
@@ -14,8 +14,6 @@
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA
  *
- * $Id: dilnetpc.c,v 1.20 2005/11/07 11:14:26 gleixner Exp $
- *
  * The DIL/Net PC is a tiny embedded PC board made by SSV Embedded Systems
  * featuring the AMD Elan SC410 processor. There are two variants of this
  * board: DNP/1486 and ADNP/1486. The DNP version has 2 megs of flash
diff --git a/drivers/mtd/maps/dmv182.c b/drivers/mtd/maps/dmv182.c
index e0558b0b2fe6..d171674eb2ed 100644
--- a/drivers/mtd/maps/dmv182.c
+++ b/drivers/mtd/maps/dmv182.c
@@ -4,8 +4,6 @@
  *
  * Flash map driver for the Dy4 SVME182 board
  *
- * $Id: dmv182.c,v 1.6 2005/11/07 11:14:26 gleixner Exp $
- *
  * Copyright 2003-2004, TimeSys Corporation
  *
  * Based on the SVME181 flash map, by Tom Nelson, Dot4, Inc. for TimeSys Corp.
diff --git a/drivers/mtd/maps/ebony.c b/drivers/mtd/maps/ebony.c
index 1488bb92f26f..d92b7c70d3ed 100644
--- a/drivers/mtd/maps/ebony.c
+++ b/drivers/mtd/maps/ebony.c
@@ -1,6 +1,4 @@
 /*
- * $Id: ebony.c,v 1.16 2005/11/07 11:14:26 gleixner Exp $
- *
  * Mapping for Ebony user flash
  *
  * Matt Porter <mporter@kernel.crashing.org>
diff --git a/drivers/mtd/maps/edb7312.c b/drivers/mtd/maps/edb7312.c
index 1c5b97c89685..9433738c1664 100644
--- a/drivers/mtd/maps/edb7312.c
+++ b/drivers/mtd/maps/edb7312.c
@@ -1,6 +1,4 @@
 /*
- * $Id: edb7312.c,v 1.14 2005/11/07 11:14:27 gleixner Exp $
- *
  * Handle mapping of the NOR flash on Cogent EDB7312 boards
  *
  * Copyright 2002 SYSGO Real-Time Solutions GmbH
diff --git a/drivers/mtd/maps/fortunet.c b/drivers/mtd/maps/fortunet.c
index 7c50c271651c..a8e3fde4cbd5 100644
--- a/drivers/mtd/maps/fortunet.c
+++ b/drivers/mtd/maps/fortunet.c
@@ -1,6 +1,5 @@
 /* fortunet.c memory map
  *
- * $Id: fortunet.c,v 1.11 2005/11/07 11:14:27 gleixner Exp $
  */
 
 #include <linux/module.h>
diff --git a/drivers/mtd/maps/h720x-flash.c b/drivers/mtd/maps/h720x-flash.c
index 6dde3182d64a..ef8915474462 100644
--- a/drivers/mtd/maps/h720x-flash.c
+++ b/drivers/mtd/maps/h720x-flash.c
@@ -2,8 +2,6 @@
  * Flash memory access on Hynix GMS30C7201/HMS30C7202 based
  * evaluation boards
  *
- * $Id: h720x-flash.c,v 1.12 2005/11/07 11:14:27 gleixner Exp $
- *
  * (C) 2002 Jungjun Kim <jungjun.kim@hynix.com>
  *     2003 Thomas Gleixner <tglx@linutronix.de>
  */
diff --git a/drivers/mtd/maps/ichxrom.c b/drivers/mtd/maps/ichxrom.c
index 2c884c49e84a..aeb6c916e23f 100644
--- a/drivers/mtd/maps/ichxrom.c
+++ b/drivers/mtd/maps/ichxrom.c
@@ -2,7 +2,6 @@
  * ichxrom.c
  *
  * Normal mappings of chips in physical memory
- * $Id: ichxrom.c,v 1.19 2005/11/07 11:14:27 gleixner Exp $
  */
 
 #include <linux/module.h>
diff --git a/drivers/mtd/maps/impa7.c b/drivers/mtd/maps/impa7.c
index a0b4dc7155dc..2682ab51a367 100644
--- a/drivers/mtd/maps/impa7.c
+++ b/drivers/mtd/maps/impa7.c
@@ -1,6 +1,4 @@
 /*
- * $Id: impa7.c,v 1.14 2005/11/07 11:14:27 gleixner Exp $
- *
  * Handle mapping of the NOR flash on implementa A7 boards
  *
  * Copyright 2002 SYSGO Real-Time Solutions GmbH
diff --git a/drivers/mtd/maps/integrator-flash.c b/drivers/mtd/maps/integrator-flash.c
index 325c8880c437..ee361aaadb1e 100644
--- a/drivers/mtd/maps/integrator-flash.c
+++ b/drivers/mtd/maps/integrator-flash.c
@@ -22,8 +22,6 @@
    This is access code for flashes using ARM's flash partitioning
    standards.
 
-   $Id: integrator-flash.c,v 1.20 2005/11/07 11:14:27 gleixner Exp $
-
 ======================================================================*/
 
 #include <linux/module.h>
diff --git a/drivers/mtd/maps/ipaq-flash.c b/drivers/mtd/maps/ipaq-flash.c
index f27c132794c3..a806119797e0 100644
--- a/drivers/mtd/maps/ipaq-flash.c
+++ b/drivers/mtd/maps/ipaq-flash.c
@@ -4,8 +4,6 @@
  * (C) 2000 Nicolas Pitre <nico@cam.org>
  * (C) 2002 Hewlett-Packard Company <jamey.hicks@hp.com>
  * (C) 2003 Christian Pellegrin <chri@ascensit.com>, <chri@infis.univ.ts.it>: concatenation of multiple flashes
- *
- * $Id: ipaq-flash.c,v 1.5 2005/11/07 11:14:27 gleixner Exp $
  */
 
 #include <linux/module.h>
diff --git a/drivers/mtd/maps/ixp2000.c b/drivers/mtd/maps/ixp2000.c
index c8396b8574c4..c2264792a20b 100644
--- a/drivers/mtd/maps/ixp2000.c
+++ b/drivers/mtd/maps/ixp2000.c
@@ -1,6 +1,4 @@
 /*
- * $Id: ixp2000.c,v 1.9 2005/11/07 11:14:27 gleixner Exp $
- *
  * drivers/mtd/maps/ixp2000.c
  *
  * Mapping for the Intel XScale IXP2000 based systems
diff --git a/drivers/mtd/maps/ixp4xx.c b/drivers/mtd/maps/ixp4xx.c
index 01f19a4714b5..9c7a5fbd4e51 100644
--- a/drivers/mtd/maps/ixp4xx.c
+++ b/drivers/mtd/maps/ixp4xx.c
@@ -1,6 +1,4 @@
 /*
- * $Id: ixp4xx.c,v 1.13 2005/11/16 16:23:21 dvrabel Exp $
- *
  * drivers/mtd/maps/ixp4xx.c
  *
  * MTD Map file for IXP4XX based systems. Please do not make per-board
diff --git a/drivers/mtd/maps/l440gx.c b/drivers/mtd/maps/l440gx.c
index 67620adf4811..9e054503c4cf 100644
--- a/drivers/mtd/maps/l440gx.c
+++ b/drivers/mtd/maps/l440gx.c
@@ -1,6 +1,4 @@
 /*
- * $Id: l440gx.c,v 1.18 2005/11/07 11:14:27 gleixner Exp $
- *
  * BIOS Flash chip on Intel 440GX board.
  *
  * Bugs this currently does not work under linuxBIOS.
diff --git a/drivers/mtd/maps/map_funcs.c b/drivers/mtd/maps/map_funcs.c
index 9105e6ca0aa6..3f268370eeca 100644
--- a/drivers/mtd/maps/map_funcs.c
+++ b/drivers/mtd/maps/map_funcs.c
@@ -1,6 +1,4 @@
 /*
- * $Id: map_funcs.c,v 1.10 2005/06/06 23:04:36 tpoynor Exp $
- *
  * Out-of-line map I/O functions for simple maps when CONFIG_COMPLEX_MAPPINGS
  * is enabled.
  */
diff --git a/drivers/mtd/maps/mbx860.c b/drivers/mtd/maps/mbx860.c
index 06b118727846..706f67394b07 100644
--- a/drivers/mtd/maps/mbx860.c
+++ b/drivers/mtd/maps/mbx860.c
@@ -1,6 +1,4 @@
 /*
- * $Id: mbx860.c,v 1.9 2005/11/07 11:14:27 gleixner Exp $
- *
  * Handle mapping of the flash on MBX860 boards
  *
  * Author:	Anton Todorov
diff --git a/drivers/mtd/maps/mtx-1_flash.c b/drivers/mtd/maps/mtx-1_flash.c
index 2a8fde9b92f0..a3b651904127 100644
--- a/drivers/mtd/maps/mtx-1_flash.c
+++ b/drivers/mtd/maps/mtx-1_flash.c
@@ -1,8 +1,6 @@
 /*
  * Flash memory access on 4G Systems MTX-1 boards
  *
- * $Id: mtx-1_flash.c,v 1.2 2005/11/07 11:14:27 gleixner Exp $
- *
  * (C) 2005 Bruno Randolf <bruno.randolf@4g-systems.biz>
  * (C) 2005 Joern Engel <joern@wohnheim.fh-wedel.de>
  *
diff --git a/drivers/mtd/maps/netsc520.c b/drivers/mtd/maps/netsc520.c
index 95dcab2146ad..c0cb319b2b70 100644
--- a/drivers/mtd/maps/netsc520.c
+++ b/drivers/mtd/maps/netsc520.c
@@ -3,8 +3,6 @@
  * Copyright (C) 2001 Mark Langsdorf (mark.langsdorf@amd.com)
  *	based on sc520cdp.c by Sysgo Real-Time Solutions GmbH
  *
- * $Id: netsc520.c,v 1.14 2005/11/07 11:14:27 gleixner Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
diff --git a/drivers/mtd/maps/nettel.c b/drivers/mtd/maps/nettel.c
index 0c9b305a72e0..965e6c6d6ab0 100644
--- a/drivers/mtd/maps/nettel.c
+++ b/drivers/mtd/maps/nettel.c
@@ -5,8 +5,6 @@
  *
  *      (C) Copyright 2000-2001, Greg Ungerer (gerg@snapgear.com)
  *      (C) Copyright 2001-2002, SnapGear (www.snapgear.com)
- *
- *	$Id: nettel.c,v 1.12 2005/11/29 14:30:00 gleixner Exp $
  */
 
 /****************************************************************************/
diff --git a/drivers/mtd/maps/octagon-5066.c b/drivers/mtd/maps/octagon-5066.c
index a6642db3d325..43e04c1d22a9 100644
--- a/drivers/mtd/maps/octagon-5066.c
+++ b/drivers/mtd/maps/octagon-5066.c
@@ -1,4 +1,3 @@
-// $Id: octagon-5066.c,v 1.28 2005/11/07 11:14:27 gleixner Exp $
 /* ######################################################################
 
    Octagon 5066 MTD Driver.
diff --git a/drivers/mtd/maps/omap-toto-flash.c b/drivers/mtd/maps/omap-toto-flash.c
index e6e391efbeb6..0a60ebbc2175 100644
--- a/drivers/mtd/maps/omap-toto-flash.c
+++ b/drivers/mtd/maps/omap-toto-flash.c
@@ -4,8 +4,6 @@
  * jzhang@ti.com (C) 2003 Texas Instruments.
  *
  *  (C) 2002 MontVista Software, Inc.
- *
- * $Id: omap-toto-flash.c,v 1.5 2005/11/07 11:14:27 gleixner Exp $
  */
 
 #include <linux/module.h>
diff --git a/drivers/mtd/maps/pci.c b/drivers/mtd/maps/pci.c
index d2ab1bae9c34..5c6a25c90380 100644
--- a/drivers/mtd/maps/pci.c
+++ b/drivers/mtd/maps/pci.c
@@ -7,8 +7,6 @@
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
  *
- *  $Id: pci.c,v 1.14 2005/11/17 08:20:27 dwmw2 Exp $
- *
  * Generic PCI memory map driver.  We support the following boards:
  *  - Intel IQ80310 ATU.
  *  - Intel EBSA285 (blank rom programming mode). Tested working 27/09/2001
diff --git a/drivers/mtd/maps/pcmciamtd.c b/drivers/mtd/maps/pcmciamtd.c
index 1912d968718b..8f7ca863f89d 100644
--- a/drivers/mtd/maps/pcmciamtd.c
+++ b/drivers/mtd/maps/pcmciamtd.c
@@ -1,6 +1,4 @@
 /*
- * $Id: pcmciamtd.c,v 1.55 2005/11/07 11:14:28 gleixner Exp $
- *
  * pcmciamtd.c - MTD driver for PCMCIA flash memory cards
  *
  * Author: Simon Evans <spse@secret.org.uk>
@@ -48,7 +46,6 @@ static const int debug = 0;
 
 
 #define DRIVER_DESC	"PCMCIA Flash memory card driver"
-#define DRIVER_VERSION	"$Revision: 1.55 $"
 
 /* Size of the PCMCIA address space: 26 bits = 64 MB */
 #define MAX_PCMCIA_ADDR	0x4000000
@@ -790,7 +787,7 @@ static struct pcmcia_driver pcmciamtd_driver = {
 
 static int __init init_pcmciamtd(void)
 {
-	info(DRIVER_DESC " " DRIVER_VERSION);
+	info(DRIVER_DESC);
 
 	if(bankwidth && bankwidth != 1 && bankwidth != 2) {
 		info("bad bankwidth (%d), using default", bankwidth);
diff --git a/drivers/mtd/maps/physmap.c b/drivers/mtd/maps/physmap.c
index 183255fcfdcb..1f6b9066b63e 100644
--- a/drivers/mtd/maps/physmap.c
+++ b/drivers/mtd/maps/physmap.c
@@ -1,6 +1,4 @@
 /*
- * $Id: physmap.c,v 1.39 2005/11/29 14:49:36 gleixner Exp $
- *
  * Normal mappings of chips in physical memory
  *
  * Copyright (C) 2003 MontaVista Software Inc.
diff --git a/drivers/mtd/maps/plat-ram.c b/drivers/mtd/maps/plat-ram.c
index 3eb2643b2328..e7dd9c8a965e 100644
--- a/drivers/mtd/maps/plat-ram.c
+++ b/drivers/mtd/maps/plat-ram.c
@@ -6,8 +6,6 @@
  *
  * Generic platfrom device based RAM map
  *
- * $Id: plat-ram.c,v 1.7 2005/11/07 11:14:28 gleixner Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
diff --git a/drivers/mtd/maps/redwood.c b/drivers/mtd/maps/redwood.c
index 4d858b3d5f82..de002eb1a7fe 100644
--- a/drivers/mtd/maps/redwood.c
+++ b/drivers/mtd/maps/redwood.c
@@ -1,6 +1,4 @@
 /*
- * $Id: redwood.c,v 1.11 2005/11/07 11:14:28 gleixner Exp $
- *
  * drivers/mtd/maps/redwood.c
  *
  * FLASH map for the IBM Redwood 4/5/6 boards.
diff --git a/drivers/mtd/maps/rpxlite.c b/drivers/mtd/maps/rpxlite.c
index 809a0c8e7aaf..14d90edb4430 100644
--- a/drivers/mtd/maps/rpxlite.c
+++ b/drivers/mtd/maps/rpxlite.c
@@ -1,6 +1,4 @@
 /*
- * $Id: rpxlite.c,v 1.22 2004/11/04 13:24:15 gleixner Exp $
- *
  * Handle mapping of the flash on the RPX Lite and CLLF boards
  */
 
diff --git a/drivers/mtd/maps/sa1100-flash.c b/drivers/mtd/maps/sa1100-flash.c
index c7d5a52a2d55..e177a43dfff0 100644
--- a/drivers/mtd/maps/sa1100-flash.c
+++ b/drivers/mtd/maps/sa1100-flash.c
@@ -2,8 +2,6 @@
  * Flash memory access on SA11x0 based devices
  *
  * (C) 2000 Nicolas Pitre <nico@cam.org>
- *
- * $Id: sa1100-flash.c,v 1.51 2005/11/07 11:14:28 gleixner Exp $
  */
 #include <linux/module.h>
 #include <linux/types.h>
diff --git a/drivers/mtd/maps/sbc8240.c b/drivers/mtd/maps/sbc8240.c
index b8c1331b7a04..6e1e99cd2b59 100644
--- a/drivers/mtd/maps/sbc8240.c
+++ b/drivers/mtd/maps/sbc8240.c
@@ -4,9 +4,6 @@
  * Carolyn Smith, Tektronix, Inc.
  *
  * This code is GPLed
- *
- * $Id: sbc8240.c,v 1.5 2005/11/07 11:14:28 gleixner Exp $
- *
  */
 
 /*
diff --git a/drivers/mtd/maps/sbc_gxx.c b/drivers/mtd/maps/sbc_gxx.c
index 7cc4041d096d..1b1c0b7e11ef 100644
--- a/drivers/mtd/maps/sbc_gxx.c
+++ b/drivers/mtd/maps/sbc_gxx.c
@@ -17,8 +17,6 @@
    along with this program; if not, write to the Free Software
    Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA
 
-   $Id: sbc_gxx.c,v 1.35 2005/11/07 11:14:28 gleixner Exp $
-
 The SBC-MediaGX / SBC-GXx has up to 16 MiB of
 Intel StrataFlash (28F320/28F640) in x8 mode.
 
diff --git a/drivers/mtd/maps/sc520cdp.c b/drivers/mtd/maps/sc520cdp.c
index 4045e372b90d..85c1e56309ec 100644
--- a/drivers/mtd/maps/sc520cdp.c
+++ b/drivers/mtd/maps/sc520cdp.c
@@ -16,8 +16,6 @@
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA
  *
- * $Id: sc520cdp.c,v 1.23 2005/11/17 08:20:27 dwmw2 Exp $
- *
  *
  * The SC520CDP is an evaluation board for the Elan SC520 processor available
  * from AMD. It has two banks of 32-bit Flash ROM, each 8 Megabytes in size,
diff --git a/drivers/mtd/maps/scb2_flash.c b/drivers/mtd/maps/scb2_flash.c
index 0fc5584324e3..21169e6d646c 100644
--- a/drivers/mtd/maps/scb2_flash.c
+++ b/drivers/mtd/maps/scb2_flash.c
@@ -1,6 +1,5 @@
 /*
  * MTD map driver for BIOS Flash on Intel SCB2 boards
- * $Id: scb2_flash.c,v 1.12 2005/03/18 14:04:35 gleixner Exp $
  * Copyright (C) 2002 Sun Microsystems, Inc.
  * Tim Hockin <thockin@sun.com>
  *
diff --git a/drivers/mtd/maps/scx200_docflash.c b/drivers/mtd/maps/scx200_docflash.c
index 5e2bce22f37c..b5391ebb736e 100644
--- a/drivers/mtd/maps/scx200_docflash.c
+++ b/drivers/mtd/maps/scx200_docflash.c
@@ -2,8 +2,6 @@
 
    Copyright (c) 2001,2002 Christer Weinigel <wingel@nano-system.com>
 
-   $Id: scx200_docflash.c,v 1.12 2005/11/07 11:14:28 gleixner Exp $
-
    National Semiconductor SCx200 flash mapped with DOCCS
 */
 
diff --git a/drivers/mtd/maps/sharpsl-flash.c b/drivers/mtd/maps/sharpsl-flash.c
index 917dc778f24e..026eab028189 100644
--- a/drivers/mtd/maps/sharpsl-flash.c
+++ b/drivers/mtd/maps/sharpsl-flash.c
@@ -4,8 +4,6 @@
  * Copyright (C) 2001 Lineo Japan, Inc.
  * Copyright (C) 2002  SHARP
  *
- * $Id: sharpsl-flash.c,v 1.7 2005/11/07 11:14:28 gleixner Exp $
- *
  * based on rpxlite.c,v 1.15 2001/10/02 15:05:14 dwmw2 Exp
  *          Handle mapping of the flash on the RPX Lite and CLLF boards
  *
diff --git a/drivers/mtd/maps/solutionengine.c b/drivers/mtd/maps/solutionengine.c
index d76ceef453ce..0eb41d9c6786 100644
--- a/drivers/mtd/maps/solutionengine.c
+++ b/drivers/mtd/maps/solutionengine.c
@@ -1,6 +1,4 @@
 /*
- * $Id: solutionengine.c,v 1.15 2005/11/07 11:14:28 gleixner Exp $
- *
  * Flash and EPROM on Hitachi Solution Engine and similar boards.
  *
  * (C) 2001 Red Hat, Inc.
diff --git a/drivers/mtd/maps/sun_uflash.c b/drivers/mtd/maps/sun_uflash.c
index 001af7f7ddda..0d7c88396c88 100644
--- a/drivers/mtd/maps/sun_uflash.c
+++ b/drivers/mtd/maps/sun_uflash.c
@@ -1,4 +1,4 @@
-/* $Id: sun_uflash.c,v 1.13 2005/11/07 11:14:28 gleixner Exp $
+/*
  *
  * sun_uflash - Driver implementation for user-programmable flash
  * present on many Sun Microsystems SME boardsets.
diff --git a/drivers/mtd/maps/tqm8xxl.c b/drivers/mtd/maps/tqm8xxl.c
index 521734057314..a5d3d8531faa 100644
--- a/drivers/mtd/maps/tqm8xxl.c
+++ b/drivers/mtd/maps/tqm8xxl.c
@@ -2,8 +2,6 @@
  * Handle mapping of the flash memory access routines
  * on TQM8xxL based devices.
  *
- * $Id: tqm8xxl.c,v 1.15 2005/11/07 11:14:28 gleixner Exp $
- *
  * based on rpxlite.c
  *
  * Copyright(C) 2001 Kirk Lee <kirk@hpc.ee.ntu.edu.tw>
diff --git a/drivers/mtd/maps/ts5500_flash.c b/drivers/mtd/maps/ts5500_flash.c
index b47270e850bc..e2147bf11c88 100644
--- a/drivers/mtd/maps/ts5500_flash.c
+++ b/drivers/mtd/maps/ts5500_flash.c
@@ -22,8 +22,6 @@
  * - Drive A and B use the resident flash disk (RFD) flash translation layer.
  * - If you have created your own jffs file system and the bios overwrites
  *   it during boot, try disabling Drive A: and B: in the boot order.
- *
- * $Id: ts5500_flash.c,v 1.5 2005/11/07 11:14:28 gleixner Exp $
  */
 
 #include <linux/init.h>
diff --git a/drivers/mtd/maps/tsunami_flash.c b/drivers/mtd/maps/tsunami_flash.c
index 0f915ac3102e..77a8bfc02577 100644
--- a/drivers/mtd/maps/tsunami_flash.c
+++ b/drivers/mtd/maps/tsunami_flash.c
@@ -2,7 +2,6 @@
  * tsunami_flash.c
  *
  * flash chip on alpha ds10...
- * $Id: tsunami_flash.c,v 1.10 2005/11/07 11:14:29 gleixner Exp $
  */
 #include <asm/io.h>
 #include <asm/core_tsunami.h>
diff --git a/drivers/mtd/maps/uclinux.c b/drivers/mtd/maps/uclinux.c
index c42f4b83f686..bac000a88313 100644
--- a/drivers/mtd/maps/uclinux.c
+++ b/drivers/mtd/maps/uclinux.c
@@ -4,8 +4,6 @@
  *	uclinux.c -- generic memory mapped MTD driver for uclinux
  *
  *	(C) Copyright 2002, Greg Ungerer (gerg@snapgear.com)
- *
- * 	$Id: uclinux.c,v 1.12 2005/11/07 11:14:29 gleixner Exp $
  */
 
 /****************************************************************************/
diff --git a/drivers/mtd/maps/vmax301.c b/drivers/mtd/maps/vmax301.c
index b3e487395435..5a0c9a353b0f 100644
--- a/drivers/mtd/maps/vmax301.c
+++ b/drivers/mtd/maps/vmax301.c
@@ -1,4 +1,3 @@
-// $Id: vmax301.c,v 1.32 2005/11/07 11:14:29 gleixner Exp $
 /* ######################################################################
 
    Tempustech VMAX SBC301 MTD Driver.
diff --git a/drivers/mtd/maps/walnut.c b/drivers/mtd/maps/walnut.c
index ca932122fb64..e243476c8171 100644
--- a/drivers/mtd/maps/walnut.c
+++ b/drivers/mtd/maps/walnut.c
@@ -1,6 +1,4 @@
 /*
- * $Id: walnut.c,v 1.3 2005/11/07 11:14:29 gleixner Exp $
- *
  * Mapping for Walnut flash
  * (used ebony.c as a "framework")
  *
diff --git a/drivers/mtd/maps/wr_sbc82xx_flash.c b/drivers/mtd/maps/wr_sbc82xx_flash.c
index ac5b8105b6ef..413b0cf9bbd2 100644
--- a/drivers/mtd/maps/wr_sbc82xx_flash.c
+++ b/drivers/mtd/maps/wr_sbc82xx_flash.c
@@ -1,6 +1,4 @@
 /*
- * $Id: wr_sbc82xx_flash.c,v 1.8 2005/11/07 11:14:29 gleixner Exp $
- *
  * Map for flash chips on Wind River PowerQUICC II SBC82xx board.
  *
  * Copyright (C) 2004 Red Hat, Inc.
diff --git a/drivers/mtd/mtd_blkdevs.c b/drivers/mtd/mtd_blkdevs.c
index 839eed8430a2..a0ada45672d8 100644
--- a/drivers/mtd/mtd_blkdevs.c
+++ b/drivers/mtd/mtd_blkdevs.c
@@ -1,6 +1,4 @@
 /*
- * $Id: mtd_blkdevs.c,v 1.27 2005/11/07 11:14:20 gleixner Exp $
- *
  * (C) 2003 David Woodhouse <dwmw2@infradead.org>
  *
  * Interface to Linux 2.5 block layer for MTD 'translation layers'.
diff --git a/drivers/mtd/mtdblock.c b/drivers/mtd/mtdblock.c
index 952da30b1745..208c6faa0358 100644
--- a/drivers/mtd/mtdblock.c
+++ b/drivers/mtd/mtdblock.c
@@ -1,8 +1,6 @@
 /*
  * Direct MTD block device access
  *
- * $Id: mtdblock.c,v 1.68 2005/11/07 11:14:20 gleixner Exp $
- *
  * (C) 2000-2003 Nicolas Pitre <nico@cam.org>
  * (C) 1999-2003 David Woodhouse <dwmw2@infradead.org>
  */
diff --git a/drivers/mtd/mtdblock_ro.c b/drivers/mtd/mtdblock_ro.c
index f79dbb49b1a2..852165f8b1c3 100644
--- a/drivers/mtd/mtdblock_ro.c
+++ b/drivers/mtd/mtdblock_ro.c
@@ -1,6 +1,4 @@
 /*
- * $Id: mtdblock_ro.c,v 1.19 2004/11/16 18:28:59 dwmw2 Exp $
- *
  * (C) 2003 David Woodhouse <dwmw2@infradead.org>
  *
  * Simple read-only (writable only for RAM) mtdblock driver
diff --git a/drivers/mtd/mtdchar.c b/drivers/mtd/mtdchar.c
index 5d3ac512ce16..4b3156f9b36f 100644
--- a/drivers/mtd/mtdchar.c
+++ b/drivers/mtd/mtdchar.c
@@ -1,6 +1,4 @@
 /*
- * $Id: mtdchar.c,v 1.76 2005/11/07 11:14:20 gleixner Exp $
- *
  * Character-device access to raw MTD devices.
  *
  */
diff --git a/drivers/mtd/mtdconcat.c b/drivers/mtd/mtdconcat.c
index d563dcd4b264..2972a5edb73d 100644
--- a/drivers/mtd/mtdconcat.c
+++ b/drivers/mtd/mtdconcat.c
@@ -6,8 +6,6 @@
  * NAND support by Christian Gan <cgan@iders.ca>
  *
  * This code is GPL
- *
- * $Id: mtdconcat.c,v 1.11 2005/11/07 11:14:20 gleixner Exp $
  */
 
 #include <linux/kernel.h>
diff --git a/drivers/mtd/mtdcore.c b/drivers/mtd/mtdcore.c
index 8c61035b968b..4373790401d3 100644
--- a/drivers/mtd/mtdcore.c
+++ b/drivers/mtd/mtdcore.c
@@ -1,6 +1,4 @@
 /*
- * $Id: mtdcore.c,v 1.47 2005/11/07 11:14:20 gleixner Exp $
- *
  * Core registration and callback routines for MTD
  * drivers and users.
  *
diff --git a/drivers/mtd/mtdpart.c b/drivers/mtd/mtdpart.c
index 07c701169344..11b803cc405b 100644
--- a/drivers/mtd/mtdpart.c
+++ b/drivers/mtd/mtdpart.c
@@ -5,8 +5,6 @@
  *
  * This code is GPL
  *
- * $Id: mtdpart.c,v 1.55 2005/11/07 11:14:20 gleixner Exp $
- *
  * 	02-21-2002	Thomas Gleixner <gleixner@autronix.de>
  *			added support for read_oob, write_oob
  */
diff --git a/drivers/mtd/nand/Kconfig b/drivers/mtd/nand/Kconfig
index 5076faf9ca66..7dea6c3a6603 100644
--- a/drivers/mtd/nand/Kconfig
+++ b/drivers/mtd/nand/Kconfig
@@ -1,5 +1,4 @@
 # drivers/mtd/nand/Kconfig
-# $Id: Kconfig,v 1.35 2005/11/07 11:14:30 gleixner Exp $
 
 menuconfig MTD_NAND
 	tristate "NAND Device Support"
diff --git a/drivers/mtd/nand/Makefile b/drivers/mtd/nand/Makefile
index a6e74a46992a..d95a10c51866 100644
--- a/drivers/mtd/nand/Makefile
+++ b/drivers/mtd/nand/Makefile
@@ -1,7 +1,6 @@
 #
 # linux/drivers/nand/Makefile
 #
-# $Id: Makefile.common,v 1.15 2004/11/26 12:28:22 dedekind Exp $
 
 obj-$(CONFIG_MTD_NAND)			+= nand.o nand_ecc.o
 obj-$(CONFIG_MTD_NAND_IDS)		+= nand_ids.o
diff --git a/drivers/mtd/nand/au1550nd.c b/drivers/mtd/nand/au1550nd.c
index 09e421a96893..22ad9f367760 100644
--- a/drivers/mtd/nand/au1550nd.c
+++ b/drivers/mtd/nand/au1550nd.c
@@ -3,8 +3,6 @@
  *
  *  Copyright (C) 2004 Embedded Edge, LLC
  *
- * $Id: au1550nd.c,v 1.13 2005/11/07 11:14:30 gleixner Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
diff --git a/drivers/mtd/nand/autcpu12.c b/drivers/mtd/nand/autcpu12.c
index dd38011ee0b7..553dd7e9b41c 100644
--- a/drivers/mtd/nand/autcpu12.c
+++ b/drivers/mtd/nand/autcpu12.c
@@ -6,8 +6,6 @@
  *  Derived from drivers/mtd/spia.c
  *	 Copyright (C) 2000 Steven J. Hill (sjhill@realitydiluted.com)
  *
- * $Id: autcpu12.c,v 1.23 2005/11/07 11:14:30 gleixner Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
diff --git a/drivers/mtd/nand/diskonchip.c b/drivers/mtd/nand/diskonchip.c
index 0e72153b3297..cd4393ce2562 100644
--- a/drivers/mtd/nand/diskonchip.c
+++ b/drivers/mtd/nand/diskonchip.c
@@ -15,8 +15,6 @@
  * converted to the generic Reed-Solomon library by Thomas Gleixner <tglx@linutronix.de>
  *
  * Interface to generic NAND code for M-Systems DiskOnChip devices
- *
- * $Id: diskonchip.c,v 1.55 2005/11/07 11:14:30 gleixner Exp $
  */
 
 #include <linux/kernel.h>
diff --git a/drivers/mtd/nand/edb7312.c b/drivers/mtd/nand/edb7312.c
index ba67bbec20d3..387e4352903e 100644
--- a/drivers/mtd/nand/edb7312.c
+++ b/drivers/mtd/nand/edb7312.c
@@ -6,8 +6,6 @@
  *  Derived from drivers/mtd/nand/autcpu12.c
  *       Copyright (c) 2001 Thomas Gleixner (gleixner@autronix.de)
  *
- * $Id: edb7312.c,v 1.12 2005/11/07 11:14:30 gleixner Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
diff --git a/drivers/mtd/nand/h1910.c b/drivers/mtd/nand/h1910.c
index 2d585d2d090c..9e59de501c2e 100644
--- a/drivers/mtd/nand/h1910.c
+++ b/drivers/mtd/nand/h1910.c
@@ -7,8 +7,6 @@
  *       Copyright (C) 2002 Marius Gröger (mag@sysgo.de)
  *       Copyright (c) 2001 Thomas Gleixner (gleixner@autronix.de)
  *
- * $Id: h1910.c,v 1.6 2005/11/07 11:14:30 gleixner Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
diff --git a/drivers/mtd/nand/nand_bbt.c b/drivers/mtd/nand/nand_bbt.c
index 5e121ceaa598..0b1c48595f12 100644
--- a/drivers/mtd/nand/nand_bbt.c
+++ b/drivers/mtd/nand/nand_bbt.c
@@ -6,8 +6,6 @@
  *
  *  Copyright (C) 2004 Thomas Gleixner (tglx@linutronix.de)
  *
- * $Id: nand_bbt.c,v 1.36 2005/11/07 11:14:30 gleixner Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
diff --git a/drivers/mtd/nand/nand_ecc.c b/drivers/mtd/nand/nand_ecc.c
index 9003a135e050..918a806a8471 100644
--- a/drivers/mtd/nand/nand_ecc.c
+++ b/drivers/mtd/nand/nand_ecc.c
@@ -9,8 +9,6 @@
  *
  * Copyright (C) 2006 Thomas Gleixner <tglx@linutronix.de>
  *
- * $Id: nand_ecc.c,v 1.15 2005/11/07 11:14:30 gleixner Exp $
- *
  * This file is free software; you can redistribute it and/or modify it
  * under the terms of the GNU General Public License as published by the
  * Free Software Foundation; either version 2 or (at your option) any
diff --git a/drivers/mtd/nand/nand_ids.c b/drivers/mtd/nand/nand_ids.c
index a3e3ab0185d5..69ee2c90eb0b 100644
--- a/drivers/mtd/nand/nand_ids.c
+++ b/drivers/mtd/nand/nand_ids.c
@@ -3,8 +3,6 @@
  *
  *  Copyright (C) 2002 Thomas Gleixner (tglx@linutronix.de)
  *
- * $Id: nand_ids.c,v 1.16 2005/11/07 11:14:31 gleixner Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
diff --git a/drivers/mtd/nand/nandsim.c b/drivers/mtd/nand/nandsim.c
index 68c150c8ff9d..add975a229a7 100644
--- a/drivers/mtd/nand/nandsim.c
+++ b/drivers/mtd/nand/nandsim.c
@@ -21,8 +21,6 @@
  * You should have received a copy of the GNU General Public License
  * along with this program; if not, write to the Free Software
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307, USA
- *
- * $Id: nandsim.c,v 1.8 2005/03/19 15:33:56 dedekind Exp $
  */
 
 #include <linux/init.h>
diff --git a/drivers/mtd/nand/ppchameleonevb.c b/drivers/mtd/nand/ppchameleonevb.c
index 082073acf20f..cc8658431851 100644
--- a/drivers/mtd/nand/ppchameleonevb.c
+++ b/drivers/mtd/nand/ppchameleonevb.c
@@ -6,8 +6,6 @@
  *  Derived from drivers/mtd/nand/edb7312.c
  *
  *
- * $Id: ppchameleonevb.c,v 1.7 2005/11/07 11:14:31 gleixner Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
diff --git a/drivers/mtd/nand/rtc_from4.c b/drivers/mtd/nand/rtc_from4.c
index 26f88215bc47..a033c4cd8e16 100644
--- a/drivers/mtd/nand/rtc_from4.c
+++ b/drivers/mtd/nand/rtc_from4.c
@@ -6,8 +6,6 @@
  *  Derived from drivers/mtd/nand/spia.c
  *       Copyright (C) 2000 Steven J. Hill (sjhill@realitydiluted.com)
  *
- * $Id: rtc_from4.c,v 1.10 2005/11/07 11:14:31 gleixner Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
diff --git a/drivers/mtd/nand/s3c2410.c b/drivers/mtd/nand/s3c2410.c
index b34a460ab679..91f42e485520 100644
--- a/drivers/mtd/nand/s3c2410.c
+++ b/drivers/mtd/nand/s3c2410.c
@@ -20,8 +20,6 @@
  *	20-Oct-2005  BJD  Fix timing calculation bug
  *	14-Jan-2006  BJD  Allow clock to be stopped when idle
  *
- * $Id: s3c2410.c,v 1.23 2006/04/01 18:06:29 bjd Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
diff --git a/drivers/mtd/nand/sharpsl.c b/drivers/mtd/nand/sharpsl.c
index 033f8800b1e6..6dba2fb66ae5 100644
--- a/drivers/mtd/nand/sharpsl.c
+++ b/drivers/mtd/nand/sharpsl.c
@@ -3,8 +3,6 @@
  *
  *  Copyright (C) 2004 Richard Purdie
  *
- *  $Id: sharpsl.c,v 1.7 2005/11/07 11:14:31 gleixner Exp $
- *
  *  Based on Sharp's NAND driver sharp_sl.c
  *
  * This program is free software; you can redistribute it and/or modify
diff --git a/drivers/mtd/nand/spia.c b/drivers/mtd/nand/spia.c
index 1f6d429b1583..0cc6d0acb8fe 100644
--- a/drivers/mtd/nand/spia.c
+++ b/drivers/mtd/nand/spia.c
@@ -8,8 +8,6 @@
  *			to controllines	(due to change in nand.c)
  *			page_cache added
  *
- * $Id: spia.c,v 1.25 2005/11/07 11:14:31 gleixner Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
diff --git a/drivers/mtd/nand/toto.c b/drivers/mtd/nand/toto.c
index f9e2d4a0ab8c..bbf492e6830d 100644
--- a/drivers/mtd/nand/toto.c
+++ b/drivers/mtd/nand/toto.c
@@ -14,8 +14,6 @@
  *  Overview:
  *   This is a device driver for the NAND flash device found on the
  *   TI fido board. It supports 32MiB and 64MiB cards
- *
- * $Id: toto.c,v 1.5 2005/11/07 11:14:31 gleixner Exp $
  */
 
 #include <linux/slab.h>
diff --git a/drivers/mtd/nand/ts7250.c b/drivers/mtd/nand/ts7250.c
index f40081069ab2..807a72752eeb 100644
--- a/drivers/mtd/nand/ts7250.c
+++ b/drivers/mtd/nand/ts7250.c
@@ -9,8 +9,6 @@
  * Derived from drivers/mtd/nand/autcpu12.c
  *   Copyright (c) 2001 Thomas Gleixner (gleixner@autronix.de)
  *
- * $Id: ts7250.c,v 1.4 2004/12/30 22:02:07 joff Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
diff --git a/drivers/mtd/nftlcore.c b/drivers/mtd/nftlcore.c
index 0c9ce19ea27a..320b929abe79 100644
--- a/drivers/mtd/nftlcore.c
+++ b/drivers/mtd/nftlcore.c
@@ -1,7 +1,6 @@
 /* Linux driver for NAND Flash Translation Layer      */
 /* (c) 1999 Machine Vision Holdings, Inc.             */
 /* Author: David Woodhouse <dwmw2@infradead.org>      */
-/* $Id: nftlcore.c,v 1.98 2005/11/07 11:14:21 gleixner Exp $ */
 
 /*
   The contents of this file are distributed under the GNU General
@@ -803,12 +802,8 @@ static struct mtd_blktrans_ops nftl_tr = {
 	.owner		= THIS_MODULE,
 };
 
-extern char nftlmountrev[];
-
 static int __init init_nftl(void)
 {
-	printk(KERN_INFO "NFTL driver: nftlcore.c $Revision: 1.98 $, nftlmount.c %s\n", nftlmountrev);
-
 	return register_mtd_blktrans(&nftl_tr);
 }
 
diff --git a/drivers/mtd/nftlmount.c b/drivers/mtd/nftlmount.c
index 345e6eff89ce..ccc4f209fbb5 100644
--- a/drivers/mtd/nftlmount.c
+++ b/drivers/mtd/nftlmount.c
@@ -4,8 +4,6 @@
  * Author: Fabrice Bellard (fabrice.bellard@netgem.com)
  * Copyright (C) 2000 Netgem S.A.
  *
- * $Id: nftlmount.c,v 1.41 2005/11/07 11:14:21 gleixner Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by
  * the Free Software Foundation; either version 2 of the License, or
@@ -31,8 +29,6 @@
 
 #define SECTORSIZE 512
 
-char nftlmountrev[]="$Revision: 1.41 $";
-
 /* find_boot_record: Find the NFTL Media Header and its Spare copy which contains the
  *	various device information of the NFTL partition and Bad Unit Table. Update
  *	the ReplUnitTable[] table accroding to the Bad Unit Table. ReplUnitTable[]
diff --git a/drivers/mtd/redboot.c b/drivers/mtd/redboot.c
index 47474903263c..5afa268c02f1 100644
--- a/drivers/mtd/redboot.c
+++ b/drivers/mtd/redboot.c
@@ -1,6 +1,4 @@
 /*
- * $Id: redboot.c,v 1.21 2006/03/30 18:34:37 bjd Exp $
- *
  * Parse RedBoot-style Flash Image System (FIS) tables and
  * produce a Linux partition array to match.
  */
diff --git a/drivers/mtd/rfd_ftl.c b/drivers/mtd/rfd_ftl.c
index c84e45465499..e538c0a72abb 100644
--- a/drivers/mtd/rfd_ftl.c
+++ b/drivers/mtd/rfd_ftl.c
@@ -3,8 +3,6 @@
  *
  * Copyright (C) 2005  Sean Young <sean@mess.org>
  *
- * $Id: rfd_ftl.c,v 1.8 2006/01/15 12:51:44 sean Exp $
- *
  * This type of flash translation layer (FTL) is used by the Embedded BIOS
  * by General Software. It is known as the Resident Flash Disk (RFD), see:
  *
diff --git a/include/linux/jffs2.h b/include/linux/jffs2.h
index 6b563cae23df..da720bc3eb15 100644
--- a/include/linux/jffs2.h
+++ b/include/linux/jffs2.h
@@ -7,9 +7,6 @@
  *
  * For licensing information, see the file 'LICENCE' in the
  * jffs2 directory.
- *
- * $Id: jffs2.h,v 1.38 2005/09/26 11:37:23 havasi Exp $
- *
  */
 
 #ifndef __LINUX_JFFS2_H__
diff --git a/include/linux/mtd/blktrans.h b/include/linux/mtd/blktrans.h
index 9a6e2f953cba..310e61606415 100644
--- a/include/linux/mtd/blktrans.h
+++ b/include/linux/mtd/blktrans.h
@@ -1,6 +1,4 @@
 /*
- * $Id: blktrans.h,v 1.6 2005/11/07 11:14:54 gleixner Exp $
- *
  * (C) 2003 David Woodhouse <dwmw2@infradead.org>
  *
  * Interface to Linux block layer for MTD 'translation layers'.
diff --git a/include/linux/mtd/cfi.h b/include/linux/mtd/cfi.h
index b0ddf4b25862..d6fb115f5a07 100644
--- a/include/linux/mtd/cfi.h
+++ b/include/linux/mtd/cfi.h
@@ -1,7 +1,6 @@
 
 /* Common Flash Interface structures
  * See http://support.intel.com/design/flash/technote/index.htm
- * $Id: cfi.h,v 1.57 2005/11/15 23:28:17 tpoynor Exp $
  */
 
 #ifndef __MTD_CFI_H__
diff --git a/include/linux/mtd/cfi_endian.h b/include/linux/mtd/cfi_endian.h
index 25724f7d3867..d802f7736be3 100644
--- a/include/linux/mtd/cfi_endian.h
+++ b/include/linux/mtd/cfi_endian.h
@@ -1,8 +1,3 @@
-/*
- * $Id: cfi_endian.h,v 1.11 2002/01/30 23:20:48 awozniak Exp $
- *
- */
-
 #include <asm/byteorder.h>
 
 #ifndef CONFIG_MTD_CFI_ADV_OPTIONS
diff --git a/include/linux/mtd/concat.h b/include/linux/mtd/concat.h
index ed8dc6755219..c02f3d264ecf 100644
--- a/include/linux/mtd/concat.h
+++ b/include/linux/mtd/concat.h
@@ -4,8 +4,6 @@
  * (C) 2002 Robert Kaiser <rkaiser@sysgo.de>
  *
  * This code is GPL
- *
- * $Id: concat.h,v 1.1 2002/03/08 16:34:36 rkaiser Exp $
  */
 
 #ifndef MTD_CONCAT_H
diff --git a/include/linux/mtd/doc2000.h b/include/linux/mtd/doc2000.h
index 9addd073bf15..0a6d516ab71d 100644
--- a/include/linux/mtd/doc2000.h
+++ b/include/linux/mtd/doc2000.h
@@ -6,8 +6,6 @@
  * Copyright (C) 2002-2003 Greg Ungerer <gerg@snapgear.com>
  * Copyright (C) 2002-2003 SnapGear Inc
  *
- * $Id: doc2000.h,v 1.25 2005/11/07 11:14:54 gleixner Exp $
- *
  * Released under GPL
  */
 
diff --git a/include/linux/mtd/flashchip.h b/include/linux/mtd/flashchip.h
index 39e7d2a1be9a..08dd131301c1 100644
--- a/include/linux/mtd/flashchip.h
+++ b/include/linux/mtd/flashchip.h
@@ -5,9 +5,6 @@
  * Contains information about the location and state of a given flash device
  *
  * (C) 2000 Red Hat. GPLd.
- *
- * $Id: flashchip.h,v 1.18 2005/11/07 11:14:54 gleixner Exp $
- *
  */
 
 #ifndef __MTD_FLASHCHIP_H__
diff --git a/include/linux/mtd/ftl.h b/include/linux/mtd/ftl.h
index d99609113307..0be442f881dd 100644
--- a/include/linux/mtd/ftl.h
+++ b/include/linux/mtd/ftl.h
@@ -1,6 +1,4 @@
 /*
- * $Id: ftl.h,v 1.7 2005/11/07 11:14:54 gleixner Exp $
- *
  * Derived from (and probably identical to):
  * ftl.h 1.7 1999/10/25 20:23:17
  *
diff --git a/include/linux/mtd/gen_probe.h b/include/linux/mtd/gen_probe.h
index 256e7342ed1e..df362ddf2949 100644
--- a/include/linux/mtd/gen_probe.h
+++ b/include/linux/mtd/gen_probe.h
@@ -1,7 +1,6 @@
 /*
  * (C) 2001, 2001 Red Hat, Inc.
  * GPL'd
- * $Id: gen_probe.h,v 1.4 2005/11/07 11:14:54 gleixner Exp $
  */
 
 #ifndef __LINUX_MTD_GEN_PROBE_H__
diff --git a/include/linux/mtd/inftl.h b/include/linux/mtd/inftl.h
index 85fd041d44ad..64ee53ce95a9 100644
--- a/include/linux/mtd/inftl.h
+++ b/include/linux/mtd/inftl.h
@@ -2,8 +2,6 @@
  *	inftl.h -- defines to support the Inverse NAND Flash Translation Layer
  *
  *	(C) Copyright 2002, Greg Ungerer (gerg@snapgear.com)
- *
- *	$Id: inftl.h,v 1.7 2005/06/13 13:08:45 sean Exp $
  */
 
 #ifndef __MTD_INFTL_H__
@@ -52,8 +50,6 @@ struct INFTLrecord {
 int INFTL_mount(struct INFTLrecord *s);
 int INFTL_formatblock(struct INFTLrecord *s, int block);
 
-extern char inftlmountrev[];
-
 void INFTL_dumptables(struct INFTLrecord *s);
 void INFTL_dumpVUchains(struct INFTLrecord *s);
 
diff --git a/include/linux/mtd/map.h b/include/linux/mtd/map.h
index a9fae032ba81..85e3939cf487 100644
--- a/include/linux/mtd/map.h
+++ b/include/linux/mtd/map.h
@@ -1,6 +1,5 @@
 
 /* Overhauled routines for dealing with different mmap regions of flash */
-/* $Id: map.h,v 1.54 2005/11/07 11:14:54 gleixner Exp $ */
 
 #ifndef __LINUX_MTD_MAP_H__
 #define __LINUX_MTD_MAP_H__
diff --git a/include/linux/mtd/mtd.h b/include/linux/mtd/mtd.h
index 245f9098e171..31ed234b2a74 100644
--- a/include/linux/mtd/mtd.h
+++ b/include/linux/mtd/mtd.h
@@ -1,6 +1,4 @@
 /*
- * $Id: mtd.h,v 1.61 2005/11/07 11:14:54 gleixner Exp $
- *
  * Copyright (C) 1999-2003 David Woodhouse <dwmw2@infradead.org> et al.
  *
  * Released under GPL
diff --git a/include/linux/mtd/nand.h b/include/linux/mtd/nand.h
index c42bc7f533a5..1288be7b7740 100644
--- a/include/linux/mtd/nand.h
+++ b/include/linux/mtd/nand.h
@@ -5,8 +5,6 @@
  *                     Steven J. Hill <sjhill@realitydiluted.com>
  *		       Thomas Gleixner <tglx@linutronix.de>
  *
- * $Id: nand.h,v 1.74 2005/09/15 13:58:50 vwool Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
diff --git a/include/linux/mtd/nand_ecc.h b/include/linux/mtd/nand_ecc.h
index 12c5bc342ead..090da505425d 100644
--- a/include/linux/mtd/nand_ecc.h
+++ b/include/linux/mtd/nand_ecc.h
@@ -3,8 +3,6 @@
  *
  *  Copyright (C) 2000 Steven J. Hill (sjhill@realitydiluted.com)
  *
- * $Id: nand_ecc.h,v 1.4 2004/06/17 02:35:02 dbrown Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
diff --git a/include/linux/mtd/nftl.h b/include/linux/mtd/nftl.h
index 001eec50cac6..dcaf611ed748 100644
--- a/include/linux/mtd/nftl.h
+++ b/include/linux/mtd/nftl.h
@@ -1,6 +1,4 @@
 /*
- * $Id: nftl.h,v 1.16 2004/06/30 14:49:00 dbrown Exp $
- *
  * (C) 1999-2003 David Woodhouse <dwmw2@infradead.org>
  */
 
diff --git a/include/linux/mtd/partitions.h b/include/linux/mtd/partitions.h
index 7c37d7e55abc..5014f7a9f5df 100644
--- a/include/linux/mtd/partitions.h
+++ b/include/linux/mtd/partitions.h
@@ -4,8 +4,6 @@
  * (C) 2000 Nicolas Pitre <nico@cam.org>
  *
  * This code is GPL
- *
- * $Id: partitions.h,v 1.17 2005/11/07 11:14:55 gleixner Exp $
  */
 
 #ifndef MTD_PARTITIONS_H
diff --git a/include/linux/mtd/physmap.h b/include/linux/mtd/physmap.h
index 0dc07d5f3354..c8e63a5ee72e 100644
--- a/include/linux/mtd/physmap.h
+++ b/include/linux/mtd/physmap.h
@@ -2,8 +2,6 @@
  * For boards with physically mapped flash and using
  * drivers/mtd/maps/physmap.c mapping driver.
  *
- * $Id: physmap.h,v 1.4 2005/11/07 11:14:55 gleixner Exp $
- *
  * Copyright (C) 2003 MontaVista Software Inc.
  * Author: Jun Sun, jsun@mvista.com or jsun@junsun.net
  *
diff --git a/include/linux/mtd/plat-ram.h b/include/linux/mtd/plat-ram.h
index 0e37ad07bce2..e07890aff1cf 100644
--- a/include/linux/mtd/plat-ram.h
+++ b/include/linux/mtd/plat-ram.h
@@ -6,8 +6,6 @@
  *
  * Generic platform device based RAM map
  *
- * $Id: plat-ram.h,v 1.2 2005/01/24 00:37:40 bjd Exp $
- *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
diff --git a/include/linux/mtd/pmc551.h b/include/linux/mtd/pmc551.h
index 5cc070c24d88..27ad40aed19f 100644
--- a/include/linux/mtd/pmc551.h
+++ b/include/linux/mtd/pmc551.h
@@ -1,6 +1,4 @@
 /*
- * $Id: pmc551.h,v 1.6 2005/11/07 11:14:55 gleixner Exp $
- *
  * PMC551 PCI Mezzanine Ram Device
  *
  * Author:
@@ -17,7 +15,7 @@
 
 #include <linux/mtd/mtd.h>
 
-#define PMC551_VERSION "$Id: pmc551.h,v 1.6 2005/11/07 11:14:55 gleixner Exp $\n"\
+#define PMC551_VERSION \
        "Ramix PMC551 PCI Mezzanine Ram Driver. (C) 1999,2000 Nortel Networks.\n"
 
 /*
diff --git a/include/linux/mtd/xip.h b/include/linux/mtd/xip.h
index e9d40bdde48c..36efcba15ecd 100644
--- a/include/linux/mtd/xip.h
+++ b/include/linux/mtd/xip.h
@@ -11,8 +11,6 @@
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
- *
- * $Id: xip.h,v 1.5 2005/11/07 11:14:55 gleixner Exp $
  */
 
 #ifndef __LINUX_MTD_XIP_H__
diff --git a/include/mtd/inftl-user.h b/include/mtd/inftl-user.h
index 9b1e2526b45e..e17eda302b2d 100644
--- a/include/mtd/inftl-user.h
+++ b/include/mtd/inftl-user.h
@@ -1,6 +1,4 @@
 /*
- * $Id: inftl-user.h,v 1.2 2005/11/07 11:14:56 gleixner Exp $
- *
  * Parts of INFTL headers shared with userspace
  *
  */
diff --git a/include/mtd/jffs2-user.h b/include/mtd/jffs2-user.h
index d508ef0ae091..001685d7fa88 100644
--- a/include/mtd/jffs2-user.h
+++ b/include/mtd/jffs2-user.h
@@ -1,6 +1,4 @@
 /*
- * $Id: jffs2-user.h,v 1.1 2004/05/05 11:57:54 dwmw2 Exp $
- *
  * JFFS2 definitions for use in user space only
  */
 
diff --git a/include/mtd/mtd-abi.h b/include/mtd/mtd-abi.h
index 615072c4da04..c6c61cd5a254 100644
--- a/include/mtd/mtd-abi.h
+++ b/include/mtd/mtd-abi.h
@@ -1,6 +1,4 @@
 /*
- * $Id: mtd-abi.h,v 1.13 2005/11/07 11:14:56 gleixner Exp $
- *
  * Portions of MTD ABI definition which are shared by kernel and user space
  */
 
diff --git a/include/mtd/mtd-user.h b/include/mtd/mtd-user.h
index 713f34d3e62e..170ceca3b2d0 100644
--- a/include/mtd/mtd-user.h
+++ b/include/mtd/mtd-user.h
@@ -1,6 +1,4 @@
 /*
- * $Id: mtd-user.h,v 1.2 2004/05/05 14:44:57 dwmw2 Exp $
- *
  * MTD ABI header for use by user space only.
  */
 
diff --git a/include/mtd/nftl-user.h b/include/mtd/nftl-user.h
index b2bca18e7311..390d21c080aa 100644
--- a/include/mtd/nftl-user.h
+++ b/include/mtd/nftl-user.h
@@ -1,6 +1,4 @@
 /*
- * $Id: nftl-user.h,v 1.2 2005/11/07 11:14:56 gleixner Exp $
- *
  * Parts of NFTL headers shared with userspace
  *
  */

From 71a928c0e52cedc43747c64b96a5f74592ab678f Mon Sep 17 00:00:00 2001
From: Chris Malley <mail@chrismalley.co.uk>
Date: Mon, 19 May 2008 20:11:50 +0100
Subject: [PATCH 0018/1171] [MTD] Use list_for_each_entry[_safe] where
 appropriate.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Janitorial work to remove temporary pointers and make some functions a bit
more readable.

Signed-off-by: Chris Malley <mail@chrismalley.co.uk>
Reviewed-By: Jörn Engel <joern@logfs.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/mtd_blkdevs.c | 32 ++++++++++----------------------
 drivers/mtd/mtdcore.c     |  6 ++----
 drivers/mtd/mtdpart.c     | 23 +++++++----------------
 3 files changed, 19 insertions(+), 42 deletions(-)

diff --git a/drivers/mtd/mtd_blkdevs.c b/drivers/mtd/mtd_blkdevs.c
index a0ada45672d8..9ff007c4962c 100644
--- a/drivers/mtd/mtd_blkdevs.c
+++ b/drivers/mtd/mtd_blkdevs.c
@@ -210,7 +210,7 @@ static struct block_device_operations mtd_blktrans_ops = {
 int add_mtd_blktrans_dev(struct mtd_blktrans_dev *new)
 {
 	struct mtd_blktrans_ops *tr = new->tr;
-	struct list_head *this;
+	struct mtd_blktrans_dev *d;
 	int last_devnum = -1;
 	struct gendisk *gd;
 
@@ -219,8 +219,7 @@ int add_mtd_blktrans_dev(struct mtd_blktrans_dev *new)
 		BUG();
 	}
 
-	list_for_each(this, &tr->devs) {
-		struct mtd_blktrans_dev *d = list_entry(this, struct mtd_blktrans_dev, list);
+	list_for_each_entry(d, &tr->devs, list) {
 		if (new->devnum == -1) {
 			/* Use first free number */
 			if (d->devnum != last_devnum+1) {
@@ -307,33 +306,24 @@ int del_mtd_blktrans_dev(struct mtd_blktrans_dev *old)
 
 static void blktrans_notify_remove(struct mtd_info *mtd)
 {
-	struct list_head *this, *this2, *next;
-
-	list_for_each(this, &blktrans_majors) {
-		struct mtd_blktrans_ops *tr = list_entry(this, struct mtd_blktrans_ops, list);
-
-		list_for_each_safe(this2, next, &tr->devs) {
-			struct mtd_blktrans_dev *dev = list_entry(this2, struct mtd_blktrans_dev, list);
+	struct mtd_blktrans_ops *tr;
+	struct mtd_blktrans_dev *dev, *next;
 
+	list_for_each_entry(tr, &blktrans_majors, list)
+		list_for_each_entry_safe(dev, next, &tr->devs, list)
 			if (dev->mtd == mtd)
 				tr->remove_dev(dev);
-		}
-	}
 }
 
 static void blktrans_notify_add(struct mtd_info *mtd)
 {
-	struct list_head *this;
+	struct mtd_blktrans_ops *tr;
 
 	if (mtd->type == MTD_ABSENT)
 		return;
 
-	list_for_each(this, &blktrans_majors) {
-		struct mtd_blktrans_ops *tr = list_entry(this, struct mtd_blktrans_ops, list);
-
+	list_for_each_entry(tr, &blktrans_majors, list)
 		tr->add_mtd(tr, mtd);
-	}
-
 }
 
 static struct mtd_notifier blktrans_notifier = {
@@ -404,7 +394,7 @@ int register_mtd_blktrans(struct mtd_blktrans_ops *tr)
 
 int deregister_mtd_blktrans(struct mtd_blktrans_ops *tr)
 {
-	struct list_head *this, *next;
+	struct mtd_blktrans_dev *dev, *next;
 
 	mutex_lock(&mtd_table_mutex);
 
@@ -414,10 +404,8 @@ int deregister_mtd_blktrans(struct mtd_blktrans_ops *tr)
 	/* Remove it from the list of active majors */
 	list_del(&tr->list);
 
-	list_for_each_safe(this, next, &tr->devs) {
-		struct mtd_blktrans_dev *dev = list_entry(this, struct mtd_blktrans_dev, list);
+	list_for_each_entry_safe(dev, next, &tr->devs, list)
 		tr->remove_dev(dev);
-	}
 
 	blk_cleanup_queue(tr->blkcore_priv->rq);
 	unregister_blkdev(tr->major, tr->name);
diff --git a/drivers/mtd/mtdcore.c b/drivers/mtd/mtdcore.c
index 4373790401d3..a9d246949820 100644
--- a/drivers/mtd/mtdcore.c
+++ b/drivers/mtd/mtdcore.c
@@ -70,9 +70,8 @@ int add_mtd_device(struct mtd_info *mtd)
 			DEBUG(0, "mtd: Giving out device %d to %s\n",i, mtd->name);
 			/* No need to get a refcount on the module containing
 			   the notifier, since we hold the mtd_table_mutex */
-			list_for_each_entry(not, &mtd_notifiers, list) {
+			list_for_each_entry(not, &mtd_notifiers, list)
 				not->add(mtd);
-			}
 
 			mutex_unlock(&mtd_table_mutex);
 			/* We _know_ we aren't being removed, because
@@ -114,9 +113,8 @@ int del_mtd_device (struct mtd_info *mtd)
 
 		/* No need to get a refcount on the module containing
 		   the notifier, since we hold the mtd_table_mutex */
-		list_for_each_entry(not, &mtd_notifiers, list) {
+		list_for_each_entry(not, &mtd_notifiers, list)
 			not->remove(mtd);
-		}
 
 		mtd_table[mtd->index] = NULL;
 
diff --git a/drivers/mtd/mtdpart.c b/drivers/mtd/mtdpart.c
index 11b803cc405b..56a760a736a9 100644
--- a/drivers/mtd/mtdpart.c
+++ b/drivers/mtd/mtdpart.c
@@ -300,22 +300,15 @@ static int part_block_markbad (struct mtd_info *mtd, loff_t ofs)
 
 int del_mtd_partitions(struct mtd_info *master)
 {
-	struct list_head *node;
-	struct mtd_part *slave;
+	struct mtd_part *slave, *next;
 
-	for (node = mtd_partitions.next;
-	     node != &mtd_partitions;
-	     node = node->next) {
-		slave = list_entry(node, struct mtd_part, list);
+	list_for_each_entry_safe(slave, next, &mtd_partitions, list)
 		if (slave->master == master) {
-			struct list_head *prev = node->prev;
-			__list_del(prev, node->next);
+			list_del(&slave->list);
 			if(slave->registered)
 				del_mtd_device(&slave->mtd);
 			kfree(slave);
-			node = prev;
 		}
-	}
 
 	return 0;
 }
@@ -511,18 +504,16 @@ static LIST_HEAD(part_parsers);
 
 static struct mtd_part_parser *get_partition_parser(const char *name)
 {
-	struct list_head *this;
-	void *ret = NULL;
+	struct mtd_part_parser *p, *ret = NULL;
+
 	spin_lock(&part_parser_lock);
 
-	list_for_each(this, &part_parsers) {
-		struct mtd_part_parser *p = list_entry(this, struct mtd_part_parser, list);
-
+	list_for_each_entry(p, &part_parsers, list)
 		if (!strcmp(p->name, name) && try_module_get(p->owner)) {
 			ret = p;
 			break;
 		}
-	}
+
 	spin_unlock(&part_parser_lock);
 
 	return ret;

From 2e3c22f57029ce04d55c07b8332ae405005456d9 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier.adi@gmail.com>
Date: Mon, 19 May 2008 18:32:24 +0800
Subject: [PATCH 0019/1171] [MTD] [MAPS] Blackfin Async Flash Maps
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Handle the case where flash memory and ethernet mac/phy are mapped onto
the same async bank

[try #4]
 - drop superfluous casts
 - drop SSYNC() when reading from the flash and rewrite
   bfin_copy_from() to be like bfin_copy_to() so that
   we dont have to handle all the aligned/unaligned cases

[try #3] rename bf5xx-flash to bfin-async-flash
 - move all kconfig board settings into board resources
 - fixup casting style according to lkml feedback
 - rewrite driver so that it can handle arbitrary of
   instances according to the declared platform resources

[try #2]
Remove useless SSYNC() as Will said

[try #1]
The BF533-STAMP does this for example.
All board-specific configuration goes in your board resources file.

Signed-off-by: Mike Frysinger <vapier.adi@gmail.com>
Acked-By: Jörn Engel <joern@logfs.org>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/maps/Kconfig            |  11 ++
 drivers/mtd/maps/Makefile           |   1 +
 drivers/mtd/maps/bfin-async-flash.c | 219 ++++++++++++++++++++++++++++
 3 files changed, 231 insertions(+)
 create mode 100644 drivers/mtd/maps/bfin-async-flash.c

diff --git a/drivers/mtd/maps/Kconfig b/drivers/mtd/maps/Kconfig
index b2c180245618..02c45e1e24ee 100644
--- a/drivers/mtd/maps/Kconfig
+++ b/drivers/mtd/maps/Kconfig
@@ -523,6 +523,17 @@ config MTD_PCMCIA_ANONYMOUS
 
 	  If unsure, say N.
 
+config MTD_BFIN_ASYNC
+	tristate "Blackfin BF533-STAMP Flash Chip Support"
+	depends on BFIN533_STAMP && MTD_CFI
+	select MTD_PARTITIONS
+	default y
+	help
+	  Map driver which allows for simultaneous utilization of
+	  ethernet and CFI parallel flash.
+
+	  If compiled as a module, it will be called bfin-async-flash.
+
 config MTD_UCLINUX
 	tristate "Generic uClinux RAM/ROM filesystem support"
 	depends on MTD_PARTITIONS && !MMU
diff --git a/drivers/mtd/maps/Makefile b/drivers/mtd/maps/Makefile
index 5444eaf4026f..d6cfce490583 100644
--- a/drivers/mtd/maps/Makefile
+++ b/drivers/mtd/maps/Makefile
@@ -67,3 +67,4 @@ obj-$(CONFIG_MTD_PLATRAM)	+= plat-ram.o
 obj-$(CONFIG_MTD_OMAP_NOR)	+= omap_nor.o
 obj-$(CONFIG_MTD_MTX1)		+= mtx-1_flash.o
 obj-$(CONFIG_MTD_INTEL_VR_NOR)	+= intel_vr_nor.o
+obj-$(CONFIG_MTD_BFIN_ASYNC)	+= bfin-async-flash.o
diff --git a/drivers/mtd/maps/bfin-async-flash.c b/drivers/mtd/maps/bfin-async-flash.c
new file mode 100644
index 000000000000..6fec86aaed7e
--- /dev/null
+++ b/drivers/mtd/maps/bfin-async-flash.c
@@ -0,0 +1,219 @@
+/*
+ * drivers/mtd/maps/bfin-async-flash.c
+ *
+ * Handle the case where flash memory and ethernet mac/phy are
+ * mapped onto the same async bank.  The BF533-STAMP does this
+ * for example.  All board-specific configuration goes in your
+ * board resources file.
+ *
+ * Copyright 2000 Nicolas Pitre <nico@cam.org>
+ * Copyright 2005-2008 Analog Devices Inc.
+ *
+ * Enter bugs at http://blackfin.uclinux.org/
+ *
+ * Licensed under the GPL-2 or later.
+ */
+
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/mtd/mtd.h>
+#include <linux/mtd/map.h>
+#include <linux/mtd/partitions.h>
+#include <linux/mtd/physmap.h>
+#include <linux/platform_device.h>
+#include <linux/types.h>
+
+#include <asm/blackfin.h>
+#include <linux/gpio.h>
+#include <linux/io.h>
+#include <asm/unaligned.h>
+
+#define pr_devinit(fmt, args...) ({ static const __devinitconst char __fmt[] = fmt; printk(__fmt, ## args); })
+
+#define DRIVER_NAME "bfin-async-flash"
+
+struct async_state {
+	struct mtd_info *mtd;
+	struct map_info map;
+	int enet_flash_pin;
+	uint32_t flash_ambctl0, flash_ambctl1;
+	uint32_t save_ambctl0, save_ambctl1;
+	unsigned long irq_flags;
+};
+
+static void switch_to_flash(struct async_state *state)
+{
+	local_irq_save(state->irq_flags);
+
+	gpio_set_value(state->enet_flash_pin, 0);
+
+	state->save_ambctl0 = bfin_read_EBIU_AMBCTL0();
+	state->save_ambctl1 = bfin_read_EBIU_AMBCTL1();
+	bfin_write_EBIU_AMBCTL0(state->flash_ambctl0);
+	bfin_write_EBIU_AMBCTL1(state->flash_ambctl1);
+	SSYNC();
+}
+
+static void switch_back(struct async_state *state)
+{
+	bfin_write_EBIU_AMBCTL0(state->save_ambctl0);
+	bfin_write_EBIU_AMBCTL1(state->save_ambctl1);
+	SSYNC();
+
+	gpio_set_value(state->enet_flash_pin, 1);
+
+	local_irq_restore(state->irq_flags);
+}
+
+static map_word bfin_read(struct map_info *map, unsigned long ofs)
+{
+	struct async_state *state = (struct async_state *)map->map_priv_1;
+	uint16_t word;
+	map_word test;
+
+	switch_to_flash(state);
+
+	word = readw(map->virt + ofs);
+
+	switch_back(state);
+
+	test.x[0] = word;
+	return test;
+}
+
+static void bfin_copy_from(struct map_info *map, void *to, unsigned long from, ssize_t len)
+{
+	struct async_state *state = (struct async_state *)map->map_priv_1;
+
+	switch_to_flash(state);
+
+	memcpy(to, map->virt + from, len);
+
+	switch_back(state);
+}
+
+static void bfin_write(struct map_info *map, map_word d1, unsigned long ofs)
+{
+	struct async_state *state = (struct async_state *)map->map_priv_1;
+	uint16_t d;
+
+	d = d1.x[0];
+
+	switch_to_flash(state);
+
+	writew(d, map->virt + ofs);
+	SSYNC();
+
+	switch_back(state);
+}
+
+static void bfin_copy_to(struct map_info *map, unsigned long to, const void *from, ssize_t len)
+{
+	struct async_state *state = (struct async_state *)map->map_priv_1;
+
+	switch_to_flash(state);
+
+	memcpy(map->virt + to, from, len);
+	SSYNC();
+
+	switch_back(state);
+}
+
+#ifdef CONFIG_MTD_PARTITIONS
+static const char *part_probe_types[] = { "cmdlinepart", "RedBoot", NULL };
+#endif
+
+static int __devinit bfin_flash_probe(struct platform_device *pdev)
+{
+	int ret;
+	struct physmap_flash_data *pdata = pdev->dev.platform_data;
+	struct resource *memory = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	struct resource *flash_ambctl = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+	struct async_state *state;
+
+	state = kzalloc(sizeof(*state), GFP_KERNEL);
+	if (!state)
+		return -ENOMEM;
+
+	state->map.name       = DRIVER_NAME;
+	state->map.read       = bfin_read;
+	state->map.copy_from  = bfin_copy_from;
+	state->map.write      = bfin_write;
+	state->map.copy_to    = bfin_copy_to;
+	state->map.bankwidth  = pdata->width;
+	state->map.size       = memory->end - memory->start + 1;
+	state->map.virt       = (void __iomem *)memory->start;
+	state->map.phys       = memory->start;
+	state->map.map_priv_1 = (unsigned long)state;
+	state->enet_flash_pin = platform_get_irq(pdev, 0);
+	state->flash_ambctl0  = flash_ambctl->start;
+	state->flash_ambctl1  = flash_ambctl->end;
+
+	if (gpio_request(state->enet_flash_pin, DRIVER_NAME)) {
+		pr_devinit(KERN_ERR DRIVER_NAME ": Failed to request gpio %d\n", state->enet_flash_pin);
+		return -EBUSY;
+	}
+	gpio_direction_output(state->enet_flash_pin, 1);
+
+	pr_devinit(KERN_NOTICE DRIVER_NAME ": probing %d-bit flash bus\n", state->map.bankwidth * 8);
+	state->mtd = do_map_probe(memory->name, &state->map);
+	if (!state->mtd)
+		return -ENXIO;
+
+#ifdef CONFIG_MTD_PARTITIONS
+	ret = parse_mtd_partitions(state->mtd, part_probe_types, &pdata->parts, 0);
+	if (ret > 0) {
+		pr_devinit(KERN_NOTICE DRIVER_NAME ": Using commandline partition definition\n");
+		add_mtd_partitions(state->mtd, pdata->parts, ret);
+
+	} else if (pdata->nr_parts) {
+		pr_devinit(KERN_NOTICE DRIVER_NAME ": Using board partition definition\n");
+		add_mtd_partitions(state->mtd, pdata->parts, pdata->nr_parts);
+
+	} else
+#endif
+	{
+		pr_devinit(KERN_NOTICE DRIVER_NAME ": no partition info available, registering whole flash at once\n");
+		add_mtd_device(state->mtd);
+	}
+
+	platform_set_drvdata(pdev, state);
+
+	return 0;
+}
+
+static int __devexit bfin_flash_remove(struct platform_device *pdev)
+{
+	struct async_state *state = platform_get_drvdata(pdev);
+	gpio_free(state->enet_flash_pin);
+#ifdef CONFIG_MTD_PARTITIONS
+	del_mtd_partitions(state->mtd);
+#endif
+	map_destroy(state->mtd);
+	kfree(state);
+	return 0;
+}
+
+static struct platform_driver bfin_flash_driver = {
+	.probe		= bfin_flash_probe,
+	.remove		= __devexit_p(bfin_flash_remove),
+	.driver		= {
+		.name	= DRIVER_NAME,
+	},
+};
+
+static int __init bfin_flash_init(void)
+{
+	return platform_driver_register(&bfin_flash_driver);
+}
+module_init(bfin_flash_init);
+
+static void __exit bfin_flash_exit(void)
+{
+	platform_driver_unregister(&bfin_flash_driver);
+}
+module_exit(bfin_flash_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("MTD map driver for Blackfins with flash/ethernet on same async bank");

From 451d33993b13174d27474ad2ce7a2f10ce2e31ad Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Tue, 20 May 2008 17:32:14 +0100
Subject: [PATCH 0020/1171] [MTD] [NAND] S3C2410: Change printk() into
 dev_dbg()

Fix a minor problem with what should have been
debug output by changing printk() to dev_dbg()
inside s3c2410_nand_update_chip().

Thanks to David Woodhouse for pointing this out.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/nand/s3c2410.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/mtd/nand/s3c2410.c b/drivers/mtd/nand/s3c2410.c
index 91f42e485520..c9726bd2c64a 100644
--- a/drivers/mtd/nand/s3c2410.c
+++ b/drivers/mtd/nand/s3c2410.c
@@ -689,7 +689,8 @@ static void s3c2410_nand_update_chip(struct s3c2410_nand_info *info,
 {
 	struct nand_chip *chip = &nmtd->chip;
 
-	printk("%s: chip %p: %d\n", __func__, chip, chip->page_shift);
+	dev_dbg(info->device, "chip %p => page shift %d\n",
+		chip, chip->page_shift);
 
 	if (hardware_ecc) {
 		/* change the behaviour depending on wether we are using

From 7e74a5076edb3555dc6c96dc91b155706515bb4c Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Tue, 20 May 2008 17:32:27 +0100
Subject: [PATCH 0021/1171] [MTD] [NAND] S3C2410: Remove changelog and tidy
 header

The changelog on the driver is superflous given this
is being kept under revision control. Remove the other
cruft in the header and update the copyright and the
supported device list.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/nand/s3c2410.c | 20 +++-----------------
 1 file changed, 3 insertions(+), 17 deletions(-)

diff --git a/drivers/mtd/nand/s3c2410.c b/drivers/mtd/nand/s3c2410.c
index c9726bd2c64a..35c6db54c985 100644
--- a/drivers/mtd/nand/s3c2410.c
+++ b/drivers/mtd/nand/s3c2410.c
@@ -1,24 +1,10 @@
 /* linux/drivers/mtd/nand/s3c2410.c
  *
- * Copyright (c) 2004,2005 Simtec Electronics
- *	http://www.simtec.co.uk/products/SWLINUX/
+ * Copyright © 2004-2008 Simtec Electronics
+ *	http://armlinux.simtec.co.uk/
  *	Ben Dooks <ben@simtec.co.uk>
  *
- * Samsung S3C2410/S3C240 NAND driver
- *
- * Changelog:
- *	21-Sep-2004  BJD  Initial version
- *	23-Sep-2004  BJD  Multiple device support
- *	28-Sep-2004  BJD  Fixed ECC placement for Hardware mode
- *	12-Oct-2004  BJD  Fixed errors in use of platform data
- *	18-Feb-2005  BJD  Fix sparse errors
- *	14-Mar-2005  BJD  Applied tglx's code reduction patch
- *	02-May-2005  BJD  Fixed s3c2440 support
- *	02-May-2005  BJD  Reduced hwcontrol decode
- *	20-Jun-2005  BJD  Updated s3c2440 support, fixed timing bug
- *	08-Jul-2005  BJD  Fix OOPS when no platform data supplied
- *	20-Oct-2005  BJD  Fix timing calculation bug
- *	14-Jan-2006  BJD  Allow clock to be stopped when idle
+ * Samsung S3C2410/S3C2440/S3C2412 NAND driver
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License as published by

From 4474573a90bae41351fad576fa80c424d62be567 Mon Sep 17 00:00:00 2001
From: Randy Dunlap <randy.dunlap@oracle.com>
Date: Thu, 5 Jun 2008 09:43:03 -0700
Subject: [PATCH 0022/1171] [MTD] [NAND] nandsim: missing header for do_div

Fix nandsim build error, missing #include:

linux-next-20080605/drivers/mtd/nand/nandsim.c: In function 'divide':
linux-next-20080605/drivers/mtd/nand/nandsim.c:462: error: implicit declaration of function 'do_div'

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/nand/nandsim.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/mtd/nand/nandsim.c b/drivers/mtd/nand/nandsim.c
index add975a229a7..ecd70e2504f6 100644
--- a/drivers/mtd/nand/nandsim.c
+++ b/drivers/mtd/nand/nandsim.c
@@ -37,6 +37,7 @@
 #include <linux/delay.h>
 #include <linux/list.h>
 #include <linux/random.h>
+#include <asm/div64.h>
 
 /* Default simulator parameters values */
 #if !defined(CONFIG_NANDSIM_FIRST_ID_BYTE)  || \

From aa83570e23e626fe8dd1253f17e6d175507025f1 Mon Sep 17 00:00:00 2001
From: Anton Vorontsov <avorontsov@ru.mvista.com>
Date: Fri, 6 Jun 2008 18:59:40 +0400
Subject: [PATCH 0023/1171] [MTD] [NAND] fsl_elbc_nand: fix section mismatch
 between probe and remove

WARNING: drivers/mtd/nand/built-in.o(.devinit.text+0x114): Section mismatch
in reference from the function fsl_elbc_ctrl_probe() to the function
.devexit.text:fsl_elbc_ctrl_remove()

__devinit functions should not call functions with __devexit. Since probe
function calls remove in case of errors, we want to remove __devexit
attribute from it.

Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Acked-by: Scott Wood <scottwood@freescale.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/nand/fsl_elbc_nand.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/nand/fsl_elbc_nand.c b/drivers/mtd/nand/fsl_elbc_nand.c
index 4b69aacdf5ca..1b06d29dd06b 100644
--- a/drivers/mtd/nand/fsl_elbc_nand.c
+++ b/drivers/mtd/nand/fsl_elbc_nand.c
@@ -917,7 +917,7 @@ static int __devinit fsl_elbc_ctrl_init(struct fsl_elbc_ctrl *ctrl)
 	return 0;
 }
 
-static int __devexit fsl_elbc_ctrl_remove(struct of_device *ofdev)
+static int fsl_elbc_ctrl_remove(struct of_device *ofdev)
 {
 	struct fsl_elbc_ctrl *ctrl = dev_get_drvdata(&ofdev->dev);
 	int i;
@@ -1041,7 +1041,7 @@ static struct of_platform_driver fsl_elbc_ctrl_driver = {
 	},
 	.match_table = fsl_elbc_match,
 	.probe = fsl_elbc_ctrl_probe,
-	.remove = __devexit_p(fsl_elbc_ctrl_remove),
+	.remove = fsl_elbc_ctrl_remove,
 };
 
 static int __init fsl_elbc_init(void)

From 62fd71fe710886ba449e932ad7877f4a8340c2d4 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?H=C3=A5vard=20Skinnemoen?= <haavard.skinnemoen@atmel.com>
Date: Fri, 6 Jun 2008 18:04:51 +0200
Subject: [PATCH 0024/1171] [MTD] [NAND] at91_nand: Convert to generic GPIO API
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

No point in using an AT91-specific GPIO API when the generic API works
just as well.

Signed-off-by: Håvard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/nand/at91_nand.c | 12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

diff --git a/drivers/mtd/nand/at91_nand.c b/drivers/mtd/nand/at91_nand.c
index 0adb287027a2..2dcaeeae2068 100644
--- a/drivers/mtd/nand/at91_nand.c
+++ b/drivers/mtd/nand/at91_nand.c
@@ -31,12 +31,10 @@
 #include <linux/mtd/nand.h>
 #include <linux/mtd/partitions.h>
 
+#include <asm/gpio.h>
 #include <asm/io.h>
-#include <asm/sizes.h>
 
-#include <asm/hardware.h>
 #include <asm/arch/board.h>
-#include <asm/arch/gpio.h>
 
 #ifdef CONFIG_MTD_NAND_AT91_ECC_HW
 #define hard_ecc	1
@@ -99,7 +97,7 @@ struct at91_nand_host {
 static void at91_nand_enable(struct at91_nand_host *host)
 {
 	if (host->board->enable_pin)
-		at91_set_gpio_value(host->board->enable_pin, 0);
+		gpio_set_value(host->board->enable_pin, 0);
 }
 
 /*
@@ -108,7 +106,7 @@ static void at91_nand_enable(struct at91_nand_host *host)
 static void at91_nand_disable(struct at91_nand_host *host)
 {
 	if (host->board->enable_pin)
-		at91_set_gpio_value(host->board->enable_pin, 1);
+		gpio_set_value(host->board->enable_pin, 1);
 }
 
 /*
@@ -142,7 +140,7 @@ static int at91_nand_device_ready(struct mtd_info *mtd)
 	struct nand_chip *nand_chip = mtd->priv;
 	struct at91_nand_host *host = nand_chip->priv;
 
-	return at91_get_gpio_value(host->board->rdy_pin);
+	return gpio_get_value(host->board->rdy_pin);
 }
 
 /*
@@ -447,7 +445,7 @@ static int __init at91_nand_probe(struct platform_device *pdev)
 	at91_nand_enable(host);
 
 	if (host->board->det_pin) {
-		if (at91_get_gpio_value(host->board->det_pin)) {
+		if (gpio_get_value(host->board->det_pin)) {
 			printk ("No SmartMedia card inserted.\n");
 			res = ENXIO;
 			goto out;

From d4f4c0aa8e36f69e46360b3d3569dc15d6099894 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?H=C3=A5vard=20Skinnemoen?= <haavard.skinnemoen@atmel.com>
Date: Fri, 6 Jun 2008 18:04:52 +0200
Subject: [PATCH 0025/1171] [MTD] [NAND] rename at91_nand -> atmel_nand: file
 names and Kconfig
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The AT91 NAND driver needs just a few tiny modifications to work on
AVR32 as well. Rename it atmel_nand to reflect this.

Also move the ECC register definitions into drivers/mtd/nand since they
are only useful to the atmel_nand driver, and get rid of the useless
filename at the top of each file.

Signed-off-by: Håvard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/nand/Kconfig                        | 17 +++++++++--------
 drivers/mtd/nand/Makefile                       |  2 +-
 drivers/mtd/nand/{at91_nand.c => atmel_nand.c}  | 10 ++++------
 .../mtd/nand/atmel_nand_ecc.h                   |  6 ++----
 4 files changed, 16 insertions(+), 19 deletions(-)
 rename drivers/mtd/nand/{at91_nand.c => atmel_nand.c} (98%)
 rename include/asm-arm/arch-at91/at91_ecc.h => drivers/mtd/nand/atmel_nand_ecc.h (93%)

diff --git a/drivers/mtd/nand/Kconfig b/drivers/mtd/nand/Kconfig
index 7dea6c3a6603..cdd2952c1533 100644
--- a/drivers/mtd/nand/Kconfig
+++ b/drivers/mtd/nand/Kconfig
@@ -271,7 +271,7 @@ config MTD_NAND_CS553X
 
 	  If you say "m", the module will be called "cs553x_nand.ko".
 
-config MTD_NAND_AT91
+config MTD_NAND_ATMEL
 	bool "Support for NAND Flash / SmartMedia on AT91"
 	depends on ARCH_AT91
 	help
@@ -279,14 +279,15 @@ config MTD_NAND_AT91
 	  on Atmel AT91 processors.
 choice
 	prompt "ECC management for NAND Flash / SmartMedia on AT91"
-	depends on MTD_NAND_AT91
+	depends on MTD_NAND_ATMEL
 
-config MTD_NAND_AT91_ECC_HW
+config MTD_NAND_ATMEL_ECC_HW
 	bool "Hardware ECC"
 	depends on ARCH_AT91SAM9263 || ARCH_AT91SAM9260
 	help
-	  Uses hardware ECC provided by the at91sam9260/at91sam9263 chip
-	  instead of software ECC.
+	  Use hardware ECC instead of software ECC when the chip
+	  supports it.
+
 	  The hardware ECC controller is capable of single bit error
 	  correction and 2-bit random detection per page.
 
@@ -296,16 +297,16 @@ config MTD_NAND_AT91_ECC_HW
 
 	  If unsure, say Y
 
-config MTD_NAND_AT91_ECC_SOFT
+config MTD_NAND_ATMEL_ECC_SOFT
 	bool "Software ECC"
 	help
-	  Uses software ECC.
+	  Use software ECC.
 
 	  NB : hardware and software ECC schemes are incompatible.
 	  If you switch from one to another, you'll have to erase your
 	  mtd partition.
 
-config MTD_NAND_AT91_ECC_NONE
+config MTD_NAND_ATMEL_ECC_NONE
 	bool "No ECC (testing only, DANGEROUS)"
 	depends on DEBUG_KERNEL
 	help
diff --git a/drivers/mtd/nand/Makefile b/drivers/mtd/nand/Makefile
index d95a10c51866..d772581de573 100644
--- a/drivers/mtd/nand/Makefile
+++ b/drivers/mtd/nand/Makefile
@@ -23,7 +23,7 @@ obj-$(CONFIG_MTD_NAND_TS7250)		+= ts7250.o
 obj-$(CONFIG_MTD_NAND_NANDSIM)		+= nandsim.o
 obj-$(CONFIG_MTD_NAND_CS553X)		+= cs553x_nand.o
 obj-$(CONFIG_MTD_NAND_NDFC)		+= ndfc.o
-obj-$(CONFIG_MTD_NAND_AT91)		+= at91_nand.o
+obj-$(CONFIG_MTD_NAND_ATMEL)		+= atmel_nand.o
 obj-$(CONFIG_MTD_NAND_CM_X270)		+= cmx270_nand.o
 obj-$(CONFIG_MTD_NAND_BASLER_EXCITE)	+= excite_nandflash.o
 obj-$(CONFIG_MTD_NAND_PXA3xx)		+= pxa3xx_nand.o
diff --git a/drivers/mtd/nand/at91_nand.c b/drivers/mtd/nand/atmel_nand.c
similarity index 98%
rename from drivers/mtd/nand/at91_nand.c
rename to drivers/mtd/nand/atmel_nand.c
index 2dcaeeae2068..51b703155db6 100644
--- a/drivers/mtd/nand/at91_nand.c
+++ b/drivers/mtd/nand/atmel_nand.c
@@ -1,6 +1,4 @@
 /*
- * drivers/mtd/nand/at91_nand.c
- *
  *  Copyright (C) 2003 Rick Bronson
  *
  *  Derived from drivers/mtd/nand/autcpu12.c
@@ -36,13 +34,13 @@
 
 #include <asm/arch/board.h>
 
-#ifdef CONFIG_MTD_NAND_AT91_ECC_HW
+#ifdef CONFIG_MTD_NAND_ATMEL_ECC_HW
 #define hard_ecc	1
 #else
 #define hard_ecc	0
 #endif
 
-#ifdef CONFIG_MTD_NAND_AT91_ECC_NONE
+#ifdef CONFIG_MTD_NAND_ATMEL_ECC_NONE
 #define no_ecc		1
 #else
 #define no_ecc		0
@@ -54,7 +52,7 @@
 #define ecc_writel(add, reg, value)			\
 	__raw_writel((value), add + AT91_ECC_##reg)
 
-#include <asm/arch/at91_ecc.h> /* AT91SAM9260/3 ECC registers */
+#include "atmel_nand_ecc.h"	/* Hardware ECC registers */
 
 /* oob layout for large page size
  * bad block info is on bytes 0 and 1
@@ -588,5 +586,5 @@ module_exit(at91_nand_exit);
 
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Rick Bronson");
-MODULE_DESCRIPTION("NAND/SmartMedia driver for AT91RM9200 / AT91SAM9");
+MODULE_DESCRIPTION("NAND/SmartMedia driver for AT91 / AVR32");
 MODULE_ALIAS("platform:at91_nand");
diff --git a/include/asm-arm/arch-at91/at91_ecc.h b/drivers/mtd/nand/atmel_nand_ecc.h
similarity index 93%
rename from include/asm-arm/arch-at91/at91_ecc.h
rename to drivers/mtd/nand/atmel_nand_ecc.h
index 1e5a8caca2d1..170db869aacf 100644
--- a/include/asm-arm/arch-at91/at91_ecc.h
+++ b/drivers/mtd/nand/atmel_nand_ecc.h
@@ -1,6 +1,4 @@
 /*
- * include/asm-arm/arch-at91/at91_ecc.h
- *
  * Error Corrected Code Controller (ECC) - System peripherals regsters.
  * Based on AT91SAM9260 datasheet revision B.
  *
@@ -10,8 +8,8 @@
  * option) any later version.
  */
 
-#ifndef AT91_ECC_H
-#define AT91_ECC_H
+#ifndef ATMEL_NAND_ECC_H
+#define ATMEL_NAND_ECC_H
 
 #define AT91_ECC_CR		0x00			/* Control register */
 #define		AT91_ECC_RST		(1 << 0)		/* Reset parity */

From 3c3796cc32b6e53653a5eb868dc959b8c2779db9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?H=C3=A5vard=20Skinnemoen?= <haavard.skinnemoen@atmel.com>
Date: Fri, 6 Jun 2008 18:04:53 +0200
Subject: [PATCH 0026/1171] [MTD] [NAND] rename at91_nand -> atmel_nand:
 internal symbols
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This is basically s/at91_nand/atmel_nand/g with some manual inspection.

Signed-off-by: Håvard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 arch/arm/mach-at91/at91cap9_devices.c    |   8 +-
 arch/arm/mach-at91/at91rm9200_devices.c  |   8 +-
 arch/arm/mach-at91/at91sam9260_devices.c |   8 +-
 arch/arm/mach-at91/at91sam9261_devices.c |  12 +-
 arch/arm/mach-at91/at91sam9263_devices.c |   8 +-
 arch/arm/mach-at91/at91sam9rl_devices.c  |  12 +-
 arch/arm/mach-at91/board-cam60.c         |   2 +-
 arch/arm/mach-at91/board-cap9adk.c       |   2 +-
 arch/arm/mach-at91/board-dk.c            |   2 +-
 arch/arm/mach-at91/board-kb9202.c        |   2 +-
 arch/arm/mach-at91/board-sam9-l9260.c    |   2 +-
 arch/arm/mach-at91/board-sam9260ek.c     |   2 +-
 arch/arm/mach-at91/board-sam9261ek.c     |   2 +-
 arch/arm/mach-at91/board-sam9263ek.c     |   2 +-
 arch/arm/mach-at91/board-sam9rlek.c      |   2 +-
 arch/arm/mach-at91/board-yl-9200.c       |   2 +-
 drivers/mtd/nand/atmel_nand.c            | 152 +++++++++++------------
 drivers/mtd/nand/atmel_nand_ecc.h        |  34 ++---
 include/asm-arm/arch-at91/board.h        |   4 +-
 19 files changed, 133 insertions(+), 133 deletions(-)

diff --git a/arch/arm/mach-at91/at91cap9_devices.c b/arch/arm/mach-at91/at91cap9_devices.c
index be526746e01e..fe5148e9f0cc 100644
--- a/arch/arm/mach-at91/at91cap9_devices.c
+++ b/arch/arm/mach-at91/at91cap9_devices.c
@@ -278,7 +278,7 @@ void __init at91_add_device_mmc(short mmc_id, struct at91_mmc_data *data) {}
  * -------------------------------------------------------------------- */
 
 #if defined(CONFIG_MTD_NAND_AT91) || defined(CONFIG_MTD_NAND_AT91_MODULE)
-static struct at91_nand_data nand_data;
+static struct atmel_nand_data nand_data;
 
 #define NAND_BASE	AT91_CHIPSELECT_3
 
@@ -296,7 +296,7 @@ static struct resource nand_resources[] = {
 };
 
 static struct platform_device at91cap9_nand_device = {
-	.name		= "at91_nand",
+	.name		= "atmel_nand",
 	.id		= -1,
 	.dev		= {
 				.platform_data	= &nand_data,
@@ -305,7 +305,7 @@ static struct platform_device at91cap9_nand_device = {
 	.num_resources	= ARRAY_SIZE(nand_resources),
 };
 
-void __init at91_add_device_nand(struct at91_nand_data *data)
+void __init at91_add_device_nand(struct atmel_nand_data *data)
 {
 	unsigned long csa, mode;
 
@@ -346,7 +346,7 @@ void __init at91_add_device_nand(struct at91_nand_data *data)
 	platform_device_register(&at91cap9_nand_device);
 }
 #else
-void __init at91_add_device_nand(struct at91_nand_data *data) {}
+void __init at91_add_device_nand(struct atmel_nand_data *data) {}
 #endif
 
 
diff --git a/arch/arm/mach-at91/at91rm9200_devices.c b/arch/arm/mach-at91/at91rm9200_devices.c
index de19bee83f75..8ced9bc82099 100644
--- a/arch/arm/mach-at91/at91rm9200_devices.c
+++ b/arch/arm/mach-at91/at91rm9200_devices.c
@@ -369,7 +369,7 @@ void __init at91_add_device_mmc(short mmc_id, struct at91_mmc_data *data) {}
  * -------------------------------------------------------------------- */
 
 #if defined(CONFIG_MTD_NAND_AT91) || defined(CONFIG_MTD_NAND_AT91_MODULE)
-static struct at91_nand_data nand_data;
+static struct atmel_nand_data nand_data;
 
 #define NAND_BASE	AT91_CHIPSELECT_3
 
@@ -382,7 +382,7 @@ static struct resource nand_resources[] = {
 };
 
 static struct platform_device at91rm9200_nand_device = {
-	.name		= "at91_nand",
+	.name		= "atmel_nand",
 	.id		= -1,
 	.dev		= {
 				.platform_data	= &nand_data,
@@ -391,7 +391,7 @@ static struct platform_device at91rm9200_nand_device = {
 	.num_resources	= ARRAY_SIZE(nand_resources),
 };
 
-void __init at91_add_device_nand(struct at91_nand_data *data)
+void __init at91_add_device_nand(struct atmel_nand_data *data)
 {
 	unsigned int csa;
 
@@ -429,7 +429,7 @@ void __init at91_add_device_nand(struct at91_nand_data *data)
 	platform_device_register(&at91rm9200_nand_device);
 }
 #else
-void __init at91_add_device_nand(struct at91_nand_data *data) {}
+void __init at91_add_device_nand(struct atmel_nand_data *data) {}
 #endif
 
 
diff --git a/arch/arm/mach-at91/at91sam9260_devices.c b/arch/arm/mach-at91/at91sam9260_devices.c
index 393a32aefce5..3aa62b1151b8 100644
--- a/arch/arm/mach-at91/at91sam9260_devices.c
+++ b/arch/arm/mach-at91/at91sam9260_devices.c
@@ -283,7 +283,7 @@ void __init at91_add_device_mmc(short mmc_id, struct at91_mmc_data *data) {}
  * -------------------------------------------------------------------- */
 
 #if defined(CONFIG_MTD_NAND_AT91) || defined(CONFIG_MTD_NAND_AT91_MODULE)
-static struct at91_nand_data nand_data;
+static struct atmel_nand_data nand_data;
 
 #define NAND_BASE	AT91_CHIPSELECT_3
 
@@ -301,7 +301,7 @@ static struct resource nand_resources[] = {
 };
 
 static struct platform_device at91sam9260_nand_device = {
-	.name		= "at91_nand",
+	.name		= "atmel_nand",
 	.id		= -1,
 	.dev		= {
 				.platform_data	= &nand_data,
@@ -310,7 +310,7 @@ static struct platform_device at91sam9260_nand_device = {
 	.num_resources	= ARRAY_SIZE(nand_resources),
 };
 
-void __init at91_add_device_nand(struct at91_nand_data *data)
+void __init at91_add_device_nand(struct atmel_nand_data *data)
 {
 	unsigned long csa, mode;
 
@@ -351,7 +351,7 @@ void __init at91_add_device_nand(struct at91_nand_data *data)
 	platform_device_register(&at91sam9260_nand_device);
 }
 #else
-void __init at91_add_device_nand(struct at91_nand_data *data) {}
+void __init at91_add_device_nand(struct atmel_nand_data *data) {}
 #endif
 
 
diff --git a/arch/arm/mach-at91/at91sam9261_devices.c b/arch/arm/mach-at91/at91sam9261_devices.c
index 728bb8f39441..6b9e423ec474 100644
--- a/arch/arm/mach-at91/at91sam9261_devices.c
+++ b/arch/arm/mach-at91/at91sam9261_devices.c
@@ -199,7 +199,7 @@ void __init at91_add_device_mmc(short mmc_id, struct at91_mmc_data *data) {}
  * -------------------------------------------------------------------- */
 
 #if defined(CONFIG_MTD_NAND_AT91) || defined(CONFIG_MTD_NAND_AT91_MODULE)
-static struct at91_nand_data nand_data;
+static struct atmel_nand_data nand_data;
 
 #define NAND_BASE	AT91_CHIPSELECT_3
 
@@ -211,8 +211,8 @@ static struct resource nand_resources[] = {
 	}
 };
 
-static struct platform_device at91_nand_device = {
-	.name		= "at91_nand",
+static struct platform_device atmel_nand_device = {
+	.name		= "atmel_nand",
 	.id		= -1,
 	.dev		= {
 				.platform_data	= &nand_data,
@@ -221,7 +221,7 @@ static struct platform_device at91_nand_device = {
 	.num_resources	= ARRAY_SIZE(nand_resources),
 };
 
-void __init at91_add_device_nand(struct at91_nand_data *data)
+void __init at91_add_device_nand(struct atmel_nand_data *data)
 {
 	unsigned long csa, mode;
 
@@ -262,11 +262,11 @@ void __init at91_add_device_nand(struct at91_nand_data *data)
 	at91_set_A_periph(AT91_PIN_PC1, 0);		/* NANDWE */
 
 	nand_data = *data;
-	platform_device_register(&at91_nand_device);
+	platform_device_register(&atmel_nand_device);
 }
 
 #else
-void __init at91_add_device_nand(struct at91_nand_data *data) {}
+void __init at91_add_device_nand(struct atmel_nand_data *data) {}
 #endif
 
 
diff --git a/arch/arm/mach-at91/at91sam9263_devices.c b/arch/arm/mach-at91/at91sam9263_devices.c
index 719667e25c98..f1dfbfe094a2 100644
--- a/arch/arm/mach-at91/at91sam9263_devices.c
+++ b/arch/arm/mach-at91/at91sam9263_devices.c
@@ -353,7 +353,7 @@ void __init at91_add_device_mmc(short mmc_id, struct at91_mmc_data *data) {}
  * -------------------------------------------------------------------- */
 
 #if defined(CONFIG_MTD_NAND_AT91) || defined(CONFIG_MTD_NAND_AT91_MODULE)
-static struct at91_nand_data nand_data;
+static struct atmel_nand_data nand_data;
 
 #define NAND_BASE	AT91_CHIPSELECT_3
 
@@ -371,7 +371,7 @@ static struct resource nand_resources[] = {
 };
 
 static struct platform_device at91sam9263_nand_device = {
-	.name		= "at91_nand",
+	.name		= "atmel_nand",
 	.id		= -1,
 	.dev		= {
 				.platform_data	= &nand_data,
@@ -380,7 +380,7 @@ static struct platform_device at91sam9263_nand_device = {
 	.num_resources	= ARRAY_SIZE(nand_resources),
 };
 
-void __init at91_add_device_nand(struct at91_nand_data *data)
+void __init at91_add_device_nand(struct atmel_nand_data *data)
 {
 	unsigned long csa, mode;
 
@@ -421,7 +421,7 @@ void __init at91_add_device_nand(struct at91_nand_data *data)
 	platform_device_register(&at91sam9263_nand_device);
 }
 #else
-void __init at91_add_device_nand(struct at91_nand_data *data) {}
+void __init at91_add_device_nand(struct atmel_nand_data *data) {}
 #endif
 
 
diff --git a/arch/arm/mach-at91/at91sam9rl_devices.c b/arch/arm/mach-at91/at91sam9rl_devices.c
index 054689804e77..b21f33393269 100644
--- a/arch/arm/mach-at91/at91sam9rl_devices.c
+++ b/arch/arm/mach-at91/at91sam9rl_devices.c
@@ -100,7 +100,7 @@ void __init at91_add_device_mmc(short mmc_id, struct at91_mmc_data *data) {}
  * -------------------------------------------------------------------- */
 
 #if defined(CONFIG_MTD_NAND_AT91) || defined(CONFIG_MTD_NAND_AT91_MODULE)
-static struct at91_nand_data nand_data;
+static struct atmel_nand_data nand_data;
 
 #define NAND_BASE	AT91_CHIPSELECT_3
 
@@ -117,8 +117,8 @@ static struct resource nand_resources[] = {
 	}
 };
 
-static struct platform_device at91_nand_device = {
-	.name		= "at91_nand",
+static struct platform_device atmel_nand_device = {
+	.name		= "atmel_nand",
 	.id		= -1,
 	.dev		= {
 				.platform_data	= &nand_data,
@@ -127,7 +127,7 @@ static struct platform_device at91_nand_device = {
 	.num_resources	= ARRAY_SIZE(nand_resources),
 };
 
-void __init at91_add_device_nand(struct at91_nand_data *data)
+void __init at91_add_device_nand(struct atmel_nand_data *data)
 {
 	unsigned long csa;
 
@@ -164,11 +164,11 @@ void __init at91_add_device_nand(struct at91_nand_data *data)
 	at91_set_A_periph(AT91_PIN_PB5, 0);		/* NANDWE */
 
 	nand_data = *data;
-	platform_device_register(&at91_nand_device);
+	platform_device_register(&atmel_nand_device);
 }
 
 #else
-void __init at91_add_device_nand(struct at91_nand_data *data) {}
+void __init at91_add_device_nand(struct atmel_nand_data *data) {}
 #endif
 
 
diff --git a/arch/arm/mach-at91/board-cam60.c b/arch/arm/mach-at91/board-cam60.c
index b22a1a004055..af2c33aff1a8 100644
--- a/arch/arm/mach-at91/board-cam60.c
+++ b/arch/arm/mach-at91/board-cam60.c
@@ -142,7 +142,7 @@ static struct mtd_partition * __init nand_partitions(int size, int *num_partitio
 	return cam60_nand_partition;
 }
 
-static struct at91_nand_data __initdata cam60_nand_data = {
+static struct atmel_nand_data __initdata cam60_nand_data = {
 	.ale		= 21,
 	.cle		= 22,
 	// .det_pin	= ... not there
diff --git a/arch/arm/mach-at91/board-cap9adk.c b/arch/arm/mach-at91/board-cap9adk.c
index e5512d1ff217..7144e1a89a2a 100644
--- a/arch/arm/mach-at91/board-cap9adk.c
+++ b/arch/arm/mach-at91/board-cap9adk.c
@@ -175,7 +175,7 @@ static struct mtd_partition * __init nand_partitions(int size, int *num_partitio
 	return cap9adk_nand_partitions;
 }
 
-static struct at91_nand_data __initdata cap9adk_nand_data = {
+static struct atmel_nand_data __initdata cap9adk_nand_data = {
 	.ale		= 21,
 	.cle		= 22,
 //	.det_pin	= ... not connected
diff --git a/arch/arm/mach-at91/board-dk.c b/arch/arm/mach-at91/board-dk.c
index c1a813c7169b..ffecacb71f89 100644
--- a/arch/arm/mach-at91/board-dk.c
+++ b/arch/arm/mach-at91/board-dk.c
@@ -150,7 +150,7 @@ static struct mtd_partition * __init nand_partitions(int size, int *num_partitio
 	return dk_nand_partition;
 }
 
-static struct at91_nand_data __initdata dk_nand_data = {
+static struct atmel_nand_data __initdata dk_nand_data = {
 	.ale		= 22,
 	.cle		= 21,
 	.det_pin	= AT91_PIN_PB1,
diff --git a/arch/arm/mach-at91/board-kb9202.c b/arch/arm/mach-at91/board-kb9202.c
index 4b39b9cda75b..153450f168f2 100644
--- a/arch/arm/mach-at91/board-kb9202.c
+++ b/arch/arm/mach-at91/board-kb9202.c
@@ -102,7 +102,7 @@ static struct mtd_partition * __init nand_partitions(int size, int *num_partitio
 	return kb9202_nand_partition;
 }
 
-static struct at91_nand_data __initdata kb9202_nand_data = {
+static struct atmel_nand_data __initdata kb9202_nand_data = {
 	.ale		= 22,
 	.cle		= 21,
 	// .det_pin	= ... not there
diff --git a/arch/arm/mach-at91/board-sam9-l9260.c b/arch/arm/mach-at91/board-sam9-l9260.c
index 8f76af5e219a..57a6221943ed 100644
--- a/arch/arm/mach-at91/board-sam9-l9260.c
+++ b/arch/arm/mach-at91/board-sam9-l9260.c
@@ -141,7 +141,7 @@ static struct mtd_partition * __init nand_partitions(int size, int *num_partitio
 	return ek_nand_partition;
 }
 
-static struct at91_nand_data __initdata ek_nand_data = {
+static struct atmel_nand_data __initdata ek_nand_data = {
 	.ale		= 21,
 	.cle		= 22,
 //	.det_pin	= ... not connected
diff --git a/arch/arm/mach-at91/board-sam9260ek.c b/arch/arm/mach-at91/board-sam9260ek.c
index 4d1d9c777084..6a680795c3c8 100644
--- a/arch/arm/mach-at91/board-sam9260ek.c
+++ b/arch/arm/mach-at91/board-sam9260ek.c
@@ -178,7 +178,7 @@ static struct mtd_partition * __init nand_partitions(int size, int *num_partitio
 	return ek_nand_partition;
 }
 
-static struct at91_nand_data __initdata ek_nand_data = {
+static struct atmel_nand_data __initdata ek_nand_data = {
 	.ale		= 21,
 	.cle		= 22,
 //	.det_pin	= ... not connected
diff --git a/arch/arm/mach-at91/board-sam9261ek.c b/arch/arm/mach-at91/board-sam9261ek.c
index 08382c0df221..43dfbd0d543a 100644
--- a/arch/arm/mach-at91/board-sam9261ek.c
+++ b/arch/arm/mach-at91/board-sam9261ek.c
@@ -183,7 +183,7 @@ static struct mtd_partition * __init nand_partitions(int size, int *num_partitio
 	return ek_nand_partition;
 }
 
-static struct at91_nand_data __initdata ek_nand_data = {
+static struct atmel_nand_data __initdata ek_nand_data = {
 	.ale		= 22,
 	.cle		= 21,
 //	.det_pin	= ... not connected
diff --git a/arch/arm/mach-at91/board-sam9263ek.c b/arch/arm/mach-at91/board-sam9263ek.c
index b4cd5d0ed597..6605a0980117 100644
--- a/arch/arm/mach-at91/board-sam9263ek.c
+++ b/arch/arm/mach-at91/board-sam9263ek.c
@@ -187,7 +187,7 @@ static struct mtd_partition * __init nand_partitions(int size, int *num_partitio
 	return ek_nand_partition;
 }
 
-static struct at91_nand_data __initdata ek_nand_data = {
+static struct atmel_nand_data __initdata ek_nand_data = {
 	.ale		= 21,
 	.cle		= 22,
 //	.det_pin	= ... not connected
diff --git a/arch/arm/mach-at91/board-sam9rlek.c b/arch/arm/mach-at91/board-sam9rlek.c
index ffc0597aee8d..35e69e51f376 100644
--- a/arch/arm/mach-at91/board-sam9rlek.c
+++ b/arch/arm/mach-at91/board-sam9rlek.c
@@ -88,7 +88,7 @@ static struct mtd_partition * __init nand_partitions(int size, int *num_partitio
 	return ek_nand_partition;
 }
 
-static struct at91_nand_data __initdata ek_nand_data = {
+static struct atmel_nand_data __initdata ek_nand_data = {
 	.ale		= 21,
 	.cle		= 22,
 //	.det_pin	= ... not connected
diff --git a/arch/arm/mach-at91/board-yl-9200.c b/arch/arm/mach-at91/board-yl-9200.c
index b5717108991d..e642b3a8bd10 100755
--- a/arch/arm/mach-at91/board-yl-9200.c
+++ b/arch/arm/mach-at91/board-yl-9200.c
@@ -251,7 +251,7 @@ static struct mtd_partition * __init nand_partitions(int size, int *num_partitio
 	return yl_9200_nand_partition;
 }
 
-static struct at91_nand_data __initdata yl_9200_nand_data = {
+static struct atmel_nand_data __initdata yl_9200_nand_data = {
 	.ale= 6,
 	.cle= 7,
 	/*.det_pin	= AT91_PIN_PCxx,*/   /*we don't have a det pin because NandFlash is fixed to board*/
diff --git a/drivers/mtd/nand/atmel_nand.c b/drivers/mtd/nand/atmel_nand.c
index 51b703155db6..675a82ca77f5 100644
--- a/drivers/mtd/nand/atmel_nand.c
+++ b/drivers/mtd/nand/atmel_nand.c
@@ -48,9 +48,9 @@
 
 /* Register access macros */
 #define ecc_readl(add, reg)				\
-	__raw_readl(add + AT91_ECC_##reg)
+	__raw_readl(add + ATMEL_ECC_##reg)
 #define ecc_writel(add, reg, value)			\
-	__raw_writel((value), add + AT91_ECC_##reg)
+	__raw_writel((value), add + ATMEL_ECC_##reg)
 
 #include "atmel_nand_ecc.h"	/* Hardware ECC registers */
 
@@ -59,7 +59,7 @@
  * the bytes have to be consecutives to avoid
  * several NAND_CMD_RNDOUT during read
  */
-static struct nand_ecclayout at91_oobinfo_large = {
+static struct nand_ecclayout atmel_oobinfo_large = {
 	.eccbytes = 4,
 	.eccpos = {60, 61, 62, 63},
 	.oobfree = {
@@ -72,7 +72,7 @@ static struct nand_ecclayout at91_oobinfo_large = {
  * the bytes have to be consecutives to avoid
  * several NAND_CMD_RNDOUT during read
  */
-static struct nand_ecclayout at91_oobinfo_small = {
+static struct nand_ecclayout atmel_oobinfo_small = {
 	.eccbytes = 4,
 	.eccpos = {0, 1, 2, 3},
 	.oobfree = {
@@ -80,11 +80,11 @@ static struct nand_ecclayout at91_oobinfo_small = {
 	},
 };
 
-struct at91_nand_host {
+struct atmel_nand_host {
 	struct nand_chip	nand_chip;
 	struct mtd_info		mtd;
 	void __iomem		*io_base;
-	struct at91_nand_data	*board;
+	struct atmel_nand_data	*board;
 	struct device		*dev;
 	void __iomem		*ecc;
 };
@@ -92,7 +92,7 @@ struct at91_nand_host {
 /*
  * Enable NAND.
  */
-static void at91_nand_enable(struct at91_nand_host *host)
+static void atmel_nand_enable(struct atmel_nand_host *host)
 {
 	if (host->board->enable_pin)
 		gpio_set_value(host->board->enable_pin, 0);
@@ -101,7 +101,7 @@ static void at91_nand_enable(struct at91_nand_host *host)
 /*
  * Disable NAND.
  */
-static void at91_nand_disable(struct at91_nand_host *host)
+static void atmel_nand_disable(struct atmel_nand_host *host)
 {
 	if (host->board->enable_pin)
 		gpio_set_value(host->board->enable_pin, 1);
@@ -110,16 +110,16 @@ static void at91_nand_disable(struct at91_nand_host *host)
 /*
  * Hardware specific access to control-lines
  */
-static void at91_nand_cmd_ctrl(struct mtd_info *mtd, int cmd, unsigned int ctrl)
+static void atmel_nand_cmd_ctrl(struct mtd_info *mtd, int cmd, unsigned int ctrl)
 {
 	struct nand_chip *nand_chip = mtd->priv;
-	struct at91_nand_host *host = nand_chip->priv;
+	struct atmel_nand_host *host = nand_chip->priv;
 
 	if (ctrl & NAND_CTRL_CHANGE) {
 		if (ctrl & NAND_NCE)
-			at91_nand_enable(host);
+			atmel_nand_enable(host);
 		else
-			at91_nand_disable(host);
+			atmel_nand_disable(host);
 	}
 	if (cmd == NAND_CMD_NONE)
 		return;
@@ -133,10 +133,10 @@ static void at91_nand_cmd_ctrl(struct mtd_info *mtd, int cmd, unsigned int ctrl)
 /*
  * Read the Device Ready pin.
  */
-static int at91_nand_device_ready(struct mtd_info *mtd)
+static int atmel_nand_device_ready(struct mtd_info *mtd)
 {
 	struct nand_chip *nand_chip = mtd->priv;
-	struct at91_nand_host *host = nand_chip->priv;
+	struct atmel_nand_host *host = nand_chip->priv;
 
 	return gpio_get_value(host->board->rdy_pin);
 }
@@ -144,7 +144,7 @@ static int at91_nand_device_ready(struct mtd_info *mtd)
 /*
  * write oob for small pages
  */
-static int at91_nand_write_oob_512(struct mtd_info *mtd,
+static int atmel_nand_write_oob_512(struct mtd_info *mtd,
 		struct nand_chip *chip, int page)
 {
 	int chunk = chip->ecc.bytes + chip->ecc.prepad + chip->ecc.postpad;
@@ -172,7 +172,7 @@ static int at91_nand_write_oob_512(struct mtd_info *mtd,
 /*
  * read oob for small pages
  */
-static int at91_nand_read_oob_512(struct mtd_info *mtd,
+static int atmel_nand_read_oob_512(struct mtd_info *mtd,
 		struct nand_chip *chip,	int page, int sndcmd)
 {
 	if (sndcmd) {
@@ -192,11 +192,11 @@ static int at91_nand_read_oob_512(struct mtd_info *mtd,
  * dat:        raw data (unused)
  * ecc_code:   buffer for ECC
  */
-static int at91_nand_calculate(struct mtd_info *mtd,
+static int atmel_nand_calculate(struct mtd_info *mtd,
 		const u_char *dat, unsigned char *ecc_code)
 {
 	struct nand_chip *nand_chip = mtd->priv;
-	struct at91_nand_host *host = nand_chip->priv;
+	struct atmel_nand_host *host = nand_chip->priv;
 	uint32_t *eccpos = nand_chip->ecc.layout->eccpos;
 	unsigned int ecc_value;
 
@@ -207,7 +207,7 @@ static int at91_nand_calculate(struct mtd_info *mtd,
 	ecc_code[eccpos[1]] = (ecc_value >> 8) & 0xFF;
 
 	/* get the last 2 ECC bytes */
-	ecc_value = ecc_readl(host->ecc, NPR) & AT91_ECC_NPARITY;
+	ecc_value = ecc_readl(host->ecc, NPR) & ATMEL_ECC_NPARITY;
 
 	ecc_code[eccpos[2]] = ecc_value & 0xFF;
 	ecc_code[eccpos[3]] = (ecc_value >> 8) & 0xFF;
@@ -222,7 +222,7 @@ static int at91_nand_calculate(struct mtd_info *mtd,
  * chip:       nand chip info structure
  * buf:        buffer to store read data
  */
-static int at91_nand_read_page(struct mtd_info *mtd,
+static int atmel_nand_read_page(struct mtd_info *mtd,
 		struct nand_chip *chip, uint8_t *buf)
 {
 	int eccsize = chip->ecc.size;
@@ -281,11 +281,11 @@ static int at91_nand_read_page(struct mtd_info *mtd,
  *
  * Detect and correct a 1 bit error for a page
  */
-static int at91_nand_correct(struct mtd_info *mtd, u_char *dat,
+static int atmel_nand_correct(struct mtd_info *mtd, u_char *dat,
 		u_char *read_ecc, u_char *isnull)
 {
 	struct nand_chip *nand_chip = mtd->priv;
-	struct at91_nand_host *host = nand_chip->priv;
+	struct atmel_nand_host *host = nand_chip->priv;
 	unsigned int ecc_status;
 	unsigned int ecc_word, ecc_bit;
 
@@ -293,43 +293,43 @@ static int at91_nand_correct(struct mtd_info *mtd, u_char *dat,
 	ecc_status = ecc_readl(host->ecc, SR);
 
 	/* if there's no error */
-	if (likely(!(ecc_status & AT91_ECC_RECERR)))
+	if (likely(!(ecc_status & ATMEL_ECC_RECERR)))
 		return 0;
 
 	/* get error bit offset (4 bits) */
-	ecc_bit = ecc_readl(host->ecc, PR) & AT91_ECC_BITADDR;
+	ecc_bit = ecc_readl(host->ecc, PR) & ATMEL_ECC_BITADDR;
 	/* get word address (12 bits) */
-	ecc_word = ecc_readl(host->ecc, PR) & AT91_ECC_WORDADDR;
+	ecc_word = ecc_readl(host->ecc, PR) & ATMEL_ECC_WORDADDR;
 	ecc_word >>= 4;
 
 	/* if there are multiple errors */
-	if (ecc_status & AT91_ECC_MULERR) {
+	if (ecc_status & ATMEL_ECC_MULERR) {
 		/* check if it is a freshly erased block
 		 * (filled with 0xff) */
-		if ((ecc_bit == AT91_ECC_BITADDR)
-				&& (ecc_word == (AT91_ECC_WORDADDR >> 4))) {
+		if ((ecc_bit == ATMEL_ECC_BITADDR)
+				&& (ecc_word == (ATMEL_ECC_WORDADDR >> 4))) {
 			/* the block has just been erased, return OK */
 			return 0;
 		}
 		/* it doesn't seems to be a freshly
 		 * erased block.
 		 * We can't correct so many errors */
-		dev_dbg(host->dev, "at91_nand : multiple errors detected."
+		dev_dbg(host->dev, "atmel_nand : multiple errors detected."
 				" Unable to correct.\n");
 		return -EIO;
 	}
 
 	/* if there's a single bit error : we can correct it */
-	if (ecc_status & AT91_ECC_ECCERR) {
+	if (ecc_status & ATMEL_ECC_ECCERR) {
 		/* there's nothing much to do here.
 		 * the bit error is on the ECC itself.
 		 */
-		dev_dbg(host->dev, "at91_nand : one bit error on ECC code."
+		dev_dbg(host->dev, "atmel_nand : one bit error on ECC code."
 				" Nothing to correct\n");
 		return 0;
 	}
 
-	dev_dbg(host->dev, "at91_nand : one bit error on data."
+	dev_dbg(host->dev, "atmel_nand : one bit error on data."
 			" (word offset in the page :"
 			" 0x%x bit offset : 0x%x)\n",
 			ecc_word, ecc_bit);
@@ -341,14 +341,14 @@ static int at91_nand_correct(struct mtd_info *mtd, u_char *dat,
 		/* 8 bits words */
 		dat[ecc_word] ^= (1 << ecc_bit);
 	}
-	dev_dbg(host->dev, "at91_nand : error corrected\n");
+	dev_dbg(host->dev, "atmel_nand : error corrected\n");
 	return 1;
 }
 
 /*
  * Enable HW ECC : unsused
  */
-static void at91_nand_hwctl(struct mtd_info *mtd, int mode) { ; }
+static void atmel_nand_hwctl(struct mtd_info *mtd, int mode) { ; }
 
 #ifdef CONFIG_MTD_PARTITIONS
 static const char *part_probes[] = { "cmdlinepart", NULL };
@@ -357,9 +357,9 @@ static const char *part_probes[] = { "cmdlinepart", NULL };
 /*
  * Probe for the NAND device.
  */
-static int __init at91_nand_probe(struct platform_device *pdev)
+static int __init atmel_nand_probe(struct platform_device *pdev)
 {
-	struct at91_nand_host *host;
+	struct atmel_nand_host *host;
 	struct mtd_info *mtd;
 	struct nand_chip *nand_chip;
 	struct resource *regs;
@@ -372,21 +372,21 @@ static int __init at91_nand_probe(struct platform_device *pdev)
 #endif
 
 	/* Allocate memory for the device structure (and zero it) */
-	host = kzalloc(sizeof(struct at91_nand_host), GFP_KERNEL);
+	host = kzalloc(sizeof(struct atmel_nand_host), GFP_KERNEL);
 	if (!host) {
-		printk(KERN_ERR "at91_nand: failed to allocate device structure.\n");
+		printk(KERN_ERR "atmel_nand: failed to allocate device structure.\n");
 		return -ENOMEM;
 	}
 
 	mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	if (!mem) {
-		printk(KERN_ERR "at91_nand: can't get I/O resource mem\n");
+		printk(KERN_ERR "atmel_nand: can't get I/O resource mem\n");
 		return -ENXIO;
 	}
 
 	host->io_base = ioremap(mem->start, mem->end - mem->start + 1);
 	if (host->io_base == NULL) {
-		printk(KERN_ERR "at91_nand: ioremap failed\n");
+		printk(KERN_ERR "atmel_nand: ioremap failed\n");
 		kfree(host);
 		return -EIO;
 	}
@@ -403,14 +403,14 @@ static int __init at91_nand_probe(struct platform_device *pdev)
 	/* Set address of NAND IO lines */
 	nand_chip->IO_ADDR_R = host->io_base;
 	nand_chip->IO_ADDR_W = host->io_base;
-	nand_chip->cmd_ctrl = at91_nand_cmd_ctrl;
+	nand_chip->cmd_ctrl = atmel_nand_cmd_ctrl;
 
 	if (host->board->rdy_pin)
-		nand_chip->dev_ready = at91_nand_device_ready;
+		nand_chip->dev_ready = atmel_nand_device_ready;
 
 	regs = platform_get_resource(pdev, IORESOURCE_MEM, 1);
 	if (!regs && hard_ecc) {
-		printk(KERN_ERR "at91_nand: can't get I/O resource "
+		printk(KERN_ERR "atmel_nand: can't get I/O resource "
 				"regs\nFalling back on software ECC\n");
 	}
 
@@ -420,15 +420,15 @@ static int __init at91_nand_probe(struct platform_device *pdev)
 	if (hard_ecc && regs) {
 		host->ecc = ioremap(regs->start, regs->end - regs->start + 1);
 		if (host->ecc == NULL) {
-			printk(KERN_ERR "at91_nand: ioremap failed\n");
+			printk(KERN_ERR "atmel_nand: ioremap failed\n");
 			res = -EIO;
 			goto err_ecc_ioremap;
 		}
 		nand_chip->ecc.mode = NAND_ECC_HW_SYNDROME;
-		nand_chip->ecc.calculate = at91_nand_calculate;
-		nand_chip->ecc.correct = at91_nand_correct;
-		nand_chip->ecc.hwctl = at91_nand_hwctl;
-		nand_chip->ecc.read_page = at91_nand_read_page;
+		nand_chip->ecc.calculate = atmel_nand_calculate;
+		nand_chip->ecc.correct = atmel_nand_correct;
+		nand_chip->ecc.hwctl = atmel_nand_hwctl;
+		nand_chip->ecc.read_page = atmel_nand_read_page;
 		nand_chip->ecc.bytes = 4;
 		nand_chip->ecc.prepad = 0;
 		nand_chip->ecc.postpad = 0;
@@ -440,7 +440,7 @@ static int __init at91_nand_probe(struct platform_device *pdev)
 		nand_chip->options |= NAND_BUSWIDTH_16;
 
 	platform_set_drvdata(pdev, host);
-	at91_nand_enable(host);
+	atmel_nand_enable(host);
 
 	if (host->board->det_pin) {
 		if (gpio_get_value(host->board->det_pin)) {
@@ -463,22 +463,22 @@ static int __init at91_nand_probe(struct platform_device *pdev)
 		/* set ECC page size and oob layout */
 		switch (mtd->writesize) {
 		case 512:
-			nand_chip->ecc.layout = &at91_oobinfo_small;
-			nand_chip->ecc.read_oob = at91_nand_read_oob_512;
-			nand_chip->ecc.write_oob = at91_nand_write_oob_512;
-			ecc_writel(host->ecc, MR, AT91_ECC_PAGESIZE_528);
+			nand_chip->ecc.layout = &atmel_oobinfo_small;
+			nand_chip->ecc.read_oob = atmel_nand_read_oob_512;
+			nand_chip->ecc.write_oob = atmel_nand_write_oob_512;
+			ecc_writel(host->ecc, MR, ATMEL_ECC_PAGESIZE_528);
 			break;
 		case 1024:
-			nand_chip->ecc.layout = &at91_oobinfo_large;
-			ecc_writel(host->ecc, MR, AT91_ECC_PAGESIZE_1056);
+			nand_chip->ecc.layout = &atmel_oobinfo_large;
+			ecc_writel(host->ecc, MR, ATMEL_ECC_PAGESIZE_1056);
 			break;
 		case 2048:
-			nand_chip->ecc.layout = &at91_oobinfo_large;
-			ecc_writel(host->ecc, MR, AT91_ECC_PAGESIZE_2112);
+			nand_chip->ecc.layout = &atmel_oobinfo_large;
+			ecc_writel(host->ecc, MR, ATMEL_ECC_PAGESIZE_2112);
 			break;
 		case 4096:
-			nand_chip->ecc.layout = &at91_oobinfo_large;
-			ecc_writel(host->ecc, MR, AT91_ECC_PAGESIZE_4224);
+			nand_chip->ecc.layout = &atmel_oobinfo_large;
+			ecc_writel(host->ecc, MR, ATMEL_ECC_PAGESIZE_4224);
 			break;
 		default:
 			/* page size not handled by HW ECC */
@@ -503,7 +503,7 @@ static int __init at91_nand_probe(struct platform_device *pdev)
 
 #ifdef CONFIG_MTD_PARTITIONS
 #ifdef CONFIG_MTD_CMDLINE_PARTS
-	mtd->name = "at91_nand";
+	mtd->name = "atmel_nand";
 	num_partitions = parse_mtd_partitions(mtd, part_probes,
 					      &partitions, 0);
 #endif
@@ -512,7 +512,7 @@ static int __init at91_nand_probe(struct platform_device *pdev)
 							 &num_partitions);
 
 	if ((!partitions) || (num_partitions == 0)) {
-		printk(KERN_ERR "at91_nand: No parititions defined, or unsupported device.\n");
+		printk(KERN_ERR "atmel_nand: No parititions defined, or unsupported device.\n");
 		res = ENXIO;
 		goto release;
 	}
@@ -534,7 +534,7 @@ out:
 	iounmap(host->ecc);
 
 err_ecc_ioremap:
-	at91_nand_disable(host);
+	atmel_nand_disable(host);
 	platform_set_drvdata(pdev, NULL);
 	iounmap(host->io_base);
 	kfree(host);
@@ -544,14 +544,14 @@ err_ecc_ioremap:
 /*
  * Remove a NAND device.
  */
-static int __devexit at91_nand_remove(struct platform_device *pdev)
+static int __devexit atmel_nand_remove(struct platform_device *pdev)
 {
-	struct at91_nand_host *host = platform_get_drvdata(pdev);
+	struct atmel_nand_host *host = platform_get_drvdata(pdev);
 	struct mtd_info *mtd = &host->mtd;
 
 	nand_release(mtd);
 
-	at91_nand_disable(host);
+	atmel_nand_disable(host);
 
 	iounmap(host->io_base);
 	iounmap(host->ecc);
@@ -560,31 +560,31 @@ static int __devexit at91_nand_remove(struct platform_device *pdev)
 	return 0;
 }
 
-static struct platform_driver at91_nand_driver = {
-	.probe		= at91_nand_probe,
-	.remove		= at91_nand_remove,
+static struct platform_driver atmel_nand_driver = {
+	.probe		= atmel_nand_probe,
+	.remove		= atmel_nand_remove,
 	.driver		= {
-		.name	= "at91_nand",
+		.name	= "atmel_nand",
 		.owner	= THIS_MODULE,
 	},
 };
 
-static int __init at91_nand_init(void)
+static int __init atmel_nand_init(void)
 {
-	return platform_driver_register(&at91_nand_driver);
+	return platform_driver_register(&atmel_nand_driver);
 }
 
 
-static void __exit at91_nand_exit(void)
+static void __exit atmel_nand_exit(void)
 {
-	platform_driver_unregister(&at91_nand_driver);
+	platform_driver_unregister(&atmel_nand_driver);
 }
 
 
-module_init(at91_nand_init);
-module_exit(at91_nand_exit);
+module_init(atmel_nand_init);
+module_exit(atmel_nand_exit);
 
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("Rick Bronson");
 MODULE_DESCRIPTION("NAND/SmartMedia driver for AT91 / AVR32");
-MODULE_ALIAS("platform:at91_nand");
+MODULE_ALIAS("platform:atmel_nand");
diff --git a/drivers/mtd/nand/atmel_nand_ecc.h b/drivers/mtd/nand/atmel_nand_ecc.h
index 170db869aacf..1ee7f993db1c 100644
--- a/drivers/mtd/nand/atmel_nand_ecc.h
+++ b/drivers/mtd/nand/atmel_nand_ecc.h
@@ -11,26 +11,26 @@
 #ifndef ATMEL_NAND_ECC_H
 #define ATMEL_NAND_ECC_H
 
-#define AT91_ECC_CR		0x00			/* Control register */
-#define		AT91_ECC_RST		(1 << 0)		/* Reset parity */
+#define ATMEL_ECC_CR		0x00			/* Control register */
+#define		ATMEL_ECC_RST		(1 << 0)		/* Reset parity */
 
-#define AT91_ECC_MR		0x04			/* Mode register */
-#define		AT91_ECC_PAGESIZE	(3 << 0)		/* Page Size */
-#define			AT91_ECC_PAGESIZE_528		(0)
-#define			AT91_ECC_PAGESIZE_1056		(1)
-#define			AT91_ECC_PAGESIZE_2112		(2)
-#define			AT91_ECC_PAGESIZE_4224		(3)
+#define ATMEL_ECC_MR		0x04			/* Mode register */
+#define		ATMEL_ECC_PAGESIZE	(3 << 0)		/* Page Size */
+#define			ATMEL_ECC_PAGESIZE_528		(0)
+#define			ATMEL_ECC_PAGESIZE_1056		(1)
+#define			ATMEL_ECC_PAGESIZE_2112		(2)
+#define			ATMEL_ECC_PAGESIZE_4224		(3)
 
-#define AT91_ECC_SR		0x08			/* Status register */
-#define		AT91_ECC_RECERR		(1 << 0)		/* Recoverable Error */
-#define		AT91_ECC_ECCERR		(1 << 1)		/* ECC Single Bit Error */
-#define		AT91_ECC_MULERR		(1 << 2)		/* Multiple Errors */
+#define ATMEL_ECC_SR		0x08			/* Status register */
+#define		ATMEL_ECC_RECERR		(1 << 0)		/* Recoverable Error */
+#define		ATMEL_ECC_ECCERR		(1 << 1)		/* ECC Single Bit Error */
+#define		ATMEL_ECC_MULERR		(1 << 2)		/* Multiple Errors */
 
-#define AT91_ECC_PR		0x0c			/* Parity register */
-#define		AT91_ECC_BITADDR	(0xf << 0)		/* Bit Error Address */
-#define		AT91_ECC_WORDADDR	(0xfff << 4)		/* Word Error Address */
+#define ATMEL_ECC_PR		0x0c			/* Parity register */
+#define		ATMEL_ECC_BITADDR	(0xf << 0)		/* Bit Error Address */
+#define		ATMEL_ECC_WORDADDR	(0xfff << 4)		/* Word Error Address */
 
-#define AT91_ECC_NPR		0x10			/* NParity register */
-#define		AT91_ECC_NPARITY	(0xffff << 0)		/* NParity */
+#define ATMEL_ECC_NPR		0x10			/* NParity register */
+#define		ATMEL_ECC_NPARITY	(0xffff << 0)		/* NParity */
 
 #endif
diff --git a/include/asm-arm/arch-at91/board.h b/include/asm-arm/arch-at91/board.h
index dc189f01c5b3..6296922a6843 100644
--- a/include/asm-arm/arch-at91/board.h
+++ b/include/asm-arm/arch-at91/board.h
@@ -85,7 +85,7 @@ struct at91_usbh_data {
 extern void __init at91_add_device_usbh(struct at91_usbh_data *data);
 
  /* NAND / SmartMedia */
-struct at91_nand_data {
+struct atmel_nand_data {
 	u8		enable_pin;	/* chip enable */
 	u8		det_pin;	/* card detect */
 	u8		rdy_pin;	/* ready/busy */
@@ -94,7 +94,7 @@ struct at91_nand_data {
 	u8		bus_width_16;	/* buswidth is 16 bit */
 	struct mtd_partition* (*partition_info)(int, int*);
 };
-extern void __init at91_add_device_nand(struct at91_nand_data *data);
+extern void __init at91_add_device_nand(struct atmel_nand_data *data);
 
  /* I2C*/
 extern void __init at91_add_device_i2c(struct i2c_board_info *devices, int nr_devices);

From cc0c72e173db70a3a864994b05ebbe59b79b888f Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?H=C3=A5vard=20Skinnemoen?= <haavard.skinnemoen@atmel.com>
Date: Fri, 6 Jun 2008 18:04:54 +0200
Subject: [PATCH 0027/1171] [MTD] [NAND] atmel_nand: Clean up and fix probe()
 error path
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This fixes several bugs in the atmel_nand_probe() error path, including
at least one memory leak.

Signed-off-by: Håvard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/nand/atmel_nand.c | 41 +++++++++++++++++++----------------
 1 file changed, 22 insertions(+), 19 deletions(-)

diff --git a/drivers/mtd/nand/atmel_nand.c b/drivers/mtd/nand/atmel_nand.c
index 675a82ca77f5..325ce29f53fc 100644
--- a/drivers/mtd/nand/atmel_nand.c
+++ b/drivers/mtd/nand/atmel_nand.c
@@ -371,6 +371,12 @@ static int __init atmel_nand_probe(struct platform_device *pdev)
 	int num_partitions = 0;
 #endif
 
+	mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!mem) {
+		printk(KERN_ERR "atmel_nand: can't get I/O resource mem\n");
+		return -ENXIO;
+	}
+
 	/* Allocate memory for the device structure (and zero it) */
 	host = kzalloc(sizeof(struct atmel_nand_host), GFP_KERNEL);
 	if (!host) {
@@ -378,17 +384,11 @@ static int __init atmel_nand_probe(struct platform_device *pdev)
 		return -ENOMEM;
 	}
 
-	mem = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (!mem) {
-		printk(KERN_ERR "atmel_nand: can't get I/O resource mem\n");
-		return -ENXIO;
-	}
-
 	host->io_base = ioremap(mem->start, mem->end - mem->start + 1);
 	if (host->io_base == NULL) {
 		printk(KERN_ERR "atmel_nand: ioremap failed\n");
-		kfree(host);
-		return -EIO;
+		res = -EIO;
+		goto err_nand_ioremap;
 	}
 
 	mtd = &host->mtd;
@@ -446,14 +446,14 @@ static int __init atmel_nand_probe(struct platform_device *pdev)
 		if (gpio_get_value(host->board->det_pin)) {
 			printk ("No SmartMedia card inserted.\n");
 			res = ENXIO;
-			goto out;
+			goto err_no_card;
 		}
 	}
 
 	/* first scan to find the device and get the page size */
 	if (nand_scan_ident(mtd, 1)) {
 		res = -ENXIO;
-		goto out;
+		goto err_scan_ident;
 	}
 
 	if (nand_chip->ecc.mode == NAND_ECC_HW_SYNDROME) {
@@ -498,7 +498,7 @@ static int __init atmel_nand_probe(struct platform_device *pdev)
 	/* second phase scan */
 	if (nand_scan_tail(mtd)) {
 		res = -ENXIO;
-		goto out;
+		goto err_scan_tail;
 	}
 
 #ifdef CONFIG_MTD_PARTITIONS
@@ -514,7 +514,7 @@ static int __init atmel_nand_probe(struct platform_device *pdev)
 	if ((!partitions) || (num_partitions == 0)) {
 		printk(KERN_ERR "atmel_nand: No parititions defined, or unsupported device.\n");
 		res = ENXIO;
-		goto release;
+		goto err_no_partitions;
 	}
 
 	res = add_mtd_partitions(mtd, partitions, num_partitions);
@@ -526,17 +526,19 @@ static int __init atmel_nand_probe(struct platform_device *pdev)
 		return res;
 
 #ifdef CONFIG_MTD_PARTITIONS
-release:
+err_no_partitions:
 #endif
 	nand_release(mtd);
-
-out:
-	iounmap(host->ecc);
-
-err_ecc_ioremap:
+err_scan_tail:
+err_scan_ident:
+err_no_card:
 	atmel_nand_disable(host);
 	platform_set_drvdata(pdev, NULL);
+	if (host->ecc)
+		iounmap(host->ecc);
+err_ecc_ioremap:
 	iounmap(host->io_base);
+err_nand_ioremap:
 	kfree(host);
 	return res;
 }
@@ -553,8 +555,9 @@ static int __devexit atmel_nand_remove(struct platform_device *pdev)
 
 	atmel_nand_disable(host);
 
+	if (host->ecc)
+		iounmap(host->ecc);
 	iounmap(host->io_base);
-	iounmap(host->ecc);
 	kfree(host);
 
 	return 0;

From e1677ce4e41a7c1bbc35a0dee27d5c7e7d223a4b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?H=C3=A5vard=20Skinnemoen?= <haavard.skinnemoen@atmel.com>
Date: Fri, 6 Jun 2008 18:04:55 +0200
Subject: [PATCH 0028/1171] avr32: move hsmc_init() to core_initcall
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The board init code, typically running from postcore_initcall, may
need to set up SMC timings. We have to make sure the SMC driver is
ready before this happens.

Signed-off-by: Håvard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 arch/avr32/mach-at32ap/hsmc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/avr32/mach-at32ap/hsmc.c b/arch/avr32/mach-at32ap/hsmc.c
index fa427ed42787..b2d9bc61a35c 100644
--- a/arch/avr32/mach-at32ap/hsmc.c
+++ b/arch/avr32/mach-at32ap/hsmc.c
@@ -278,4 +278,4 @@ static int __init hsmc_init(void)
 {
 	return platform_driver_register(&hsmc_driver);
 }
-arch_initcall(hsmc_init);
+core_initcall(hsmc_init);

From 62090a08aba579e6c69319fac4d4a1f806f26400 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?H=C3=A5vard=20Skinnemoen?= <haavard.skinnemoen@atmel.com>
Date: Fri, 6 Jun 2008 18:04:56 +0200
Subject: [PATCH 0029/1171] [MTD] [NAND] avr32: atmel_nand platform code for
 AT32AP700x
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This function initializes and adds a platform_device for a NAND flash
interface on SMC chip select 3.

Signed-off-by: Håvard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 arch/avr32/mach-at32ap/at32ap700x.c   | 52 +++++++++++++++++++++++++++
 include/asm-avr32/arch-at32ap/board.h | 13 +++++++
 2 files changed, 65 insertions(+)

diff --git a/arch/avr32/mach-at32ap/at32ap700x.c b/arch/avr32/mach-at32ap/at32ap700x.c
index 0f24b4f85c17..b65d3e0667a8 100644
--- a/arch/avr32/mach-at32ap/at32ap700x.c
+++ b/arch/avr32/mach-at32ap/at32ap700x.c
@@ -1729,6 +1729,58 @@ fail:
 }
 #endif
 
+/* --------------------------------------------------------------------
+ * NAND Flash / SmartMedia
+ * -------------------------------------------------------------------- */
+static struct resource smc_cs3_resource[] __initdata = {
+	{
+		.start	= 0x0c000000,
+		.end	= 0x0fffffff,
+		.flags	= IORESOURCE_MEM,
+	}, {
+		.start	= 0xfff03c00,
+		.end	= 0xfff03fff,
+		.flags	= IORESOURCE_MEM,
+	},
+};
+
+struct platform_device *__init
+at32_add_device_nand(unsigned int id, struct atmel_nand_data *data)
+{
+	struct platform_device *pdev;
+
+	if (id != 0 || !data)
+		return NULL;
+
+	pdev = platform_device_alloc("atmel_nand", id);
+	if (!pdev)
+		goto fail;
+
+	if (platform_device_add_resources(pdev, smc_cs3_resource,
+				ARRAY_SIZE(smc_cs3_resource)))
+		goto fail;
+
+	if (platform_device_add_data(pdev, data,
+				sizeof(struct atmel_nand_data)))
+		goto fail;
+
+	set_ebi_sfr_bits(HMATRIX_BIT(CS3A));
+	if (data->enable_pin)
+		at32_select_gpio(data->enable_pin,
+				AT32_GPIOF_OUTPUT | AT32_GPIOF_HIGH);
+	if (data->rdy_pin)
+		at32_select_gpio(data->rdy_pin, 0);
+	if (data->det_pin)
+		at32_select_gpio(data->det_pin, 0);
+
+	platform_device_add(pdev);
+	return pdev;
+
+fail:
+	platform_device_put(pdev);
+	return NULL;
+}
+
 /* --------------------------------------------------------------------
  * AC97C
  * -------------------------------------------------------------------- */
diff --git a/include/asm-avr32/arch-at32ap/board.h b/include/asm-avr32/arch-at32ap/board.h
index a4e2d28bfb58..cc5906cd7419 100644
--- a/include/asm-avr32/arch-at32ap/board.h
+++ b/include/asm-avr32/arch-at32ap/board.h
@@ -85,4 +85,17 @@ struct platform_device *
 at32_add_device_cf(unsigned int id, unsigned int extint,
 		struct cf_platform_data *data);
 
+/* NAND / SmartMedia */
+struct atmel_nand_data {
+	int	enable_pin;	/* chip enable */
+	int	det_pin;	/* card detect */
+	int	rdy_pin;	/* ready/busy */
+	u8	ale;		/* address line number connected to ALE */
+	u8	cle;		/* address line number connected to CLE */
+	u8	bus_width_16;	/* buswidth is 16 bit */
+	struct mtd_partition *(*partition_info)(int size, int *num_partitions);
+};
+struct platform_device *
+at32_add_device_nand(unsigned int id, struct atmel_nand_data *data);
+
 #endif /* __ASM_ARCH_BOARD_H */

From 984290ded4ee3834ca913fe361afe3bf625cd9c0 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?H=C3=A5vard=20Skinnemoen?= <haavard.skinnemoen@atmel.com>
Date: Fri, 6 Jun 2008 18:04:57 +0200
Subject: [PATCH 0030/1171] [MTD] [NAND] atmel_nand: make available on AVR32
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Make the atmel_nand driver selectable on AVR32, and update the Kconfig
help text to reflect this.

Signed-off-by: Håvard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/nand/Kconfig | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/mtd/nand/Kconfig b/drivers/mtd/nand/Kconfig
index cdd2952c1533..d8c0d8698b49 100644
--- a/drivers/mtd/nand/Kconfig
+++ b/drivers/mtd/nand/Kconfig
@@ -272,18 +272,18 @@ config MTD_NAND_CS553X
 	  If you say "m", the module will be called "cs553x_nand.ko".
 
 config MTD_NAND_ATMEL
-	bool "Support for NAND Flash / SmartMedia on AT91"
-	depends on ARCH_AT91
+	bool "Support for NAND Flash / SmartMedia on AT91 and AVR32"
+	depends on ARCH_AT91 || AVR32
 	help
 	  Enables support for NAND Flash / Smart Media Card interface
-	  on Atmel AT91 processors.
+	  on Atmel AT91 and AVR32 processors.
 choice
-	prompt "ECC management for NAND Flash / SmartMedia on AT91"
+	prompt "ECC management for NAND Flash / SmartMedia on AT91 / AVR32"
 	depends on MTD_NAND_ATMEL
 
 config MTD_NAND_ATMEL_ECC_HW
 	bool "Hardware ECC"
-	depends on ARCH_AT91SAM9263 || ARCH_AT91SAM9260
+	depends on ARCH_AT91SAM9263 || ARCH_AT91SAM9260 || AVR32
 	help
 	  Use hardware ECC instead of software ECC when the chip
 	  supports it.

From e150d6e725c014d72106a8610be40b7b08beb77a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?H=C3=A5vard=20Skinnemoen?= <haavard.skinnemoen@atmel.com>
Date: Fri, 6 Jun 2008 18:04:58 +0200
Subject: [PATCH 0031/1171] avr32: Add support for ATSTK1006
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The ATSTK1006 is basically an upgraded version of ATSTK1002 with
128 MiB SDRAM and 256 MiB NAND flash on board.

Otherwise, the board is very similar to the ATSTK1002, so it uses the
same board support file.

Signed-off-by: Håvard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 arch/avr32/boards/atstk1000/Kconfig     |  4 ++
 arch/avr32/boards/atstk1000/Makefile    |  1 +
 arch/avr32/boards/atstk1000/atstk1002.c | 78 ++++++++++++++++++++++++-
 3 files changed, 81 insertions(+), 2 deletions(-)

diff --git a/arch/avr32/boards/atstk1000/Kconfig b/arch/avr32/boards/atstk1000/Kconfig
index af90b00100fd..8dc48214f0b7 100644
--- a/arch/avr32/boards/atstk1000/Kconfig
+++ b/arch/avr32/boards/atstk1000/Kconfig
@@ -18,6 +18,10 @@ config BOARD_ATSTK1004
 	bool "ATSTK1004"
 	select CPU_AT32AP7002
 
+config BOARD_ATSTK1006
+	bool "ATSTK1006"
+	select CPU_AT32AP7000
+
 endchoice
 
 
diff --git a/arch/avr32/boards/atstk1000/Makefile b/arch/avr32/boards/atstk1000/Makefile
index beead86462e8..edecee03742d 100644
--- a/arch/avr32/boards/atstk1000/Makefile
+++ b/arch/avr32/boards/atstk1000/Makefile
@@ -2,3 +2,4 @@ obj-y				+= setup.o flash.o
 obj-$(CONFIG_BOARD_ATSTK1002)	+= atstk1002.o
 obj-$(CONFIG_BOARD_ATSTK1003)	+= atstk1003.o
 obj-$(CONFIG_BOARD_ATSTK1004)	+= atstk1004.o
+obj-$(CONFIG_BOARD_ATSTK1006)	+= atstk1002.o
diff --git a/arch/avr32/boards/atstk1000/atstk1002.c b/arch/avr32/boards/atstk1000/atstk1002.c
index 000eb4220a12..5bfaf4d87b3f 100644
--- a/arch/avr32/boards/atstk1000/atstk1002.c
+++ b/arch/avr32/boards/atstk1000/atstk1002.c
@@ -1,7 +1,7 @@
 /*
- * ATSTK1002 daughterboard-specific init code
+ * ATSTK1002/ATSTK1006 daughterboard-specific init code
  *
- * Copyright (C) 2005-2006 Atmel Corporation
+ * Copyright (C) 2005-2007 Atmel Corporation
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -29,6 +29,74 @@
 #include "atstk1000.h"
 
 
+/*
+ * The ATSTK1006 daughterboard is very similar to the ATSTK1002. Both
+ * have the AT32AP7000 chip on board; the difference is that the
+ * STK1006 has 128 MB SDRAM (the STK1002 uses the 8 MB SDRAM chip on
+ * the STK1000 motherboard) and 256 MB NAND flash (the STK1002 has
+ * none.)
+ *
+ * The RAM difference is handled by the boot loader, so the only
+ * difference we end up handling here is the NAND flash.
+ */
+#ifdef CONFIG_BOARD_ATSTK1006
+#include <linux/mtd/partitions.h>
+#include <asm/arch/smc.h>
+
+static struct smc_timing nand_timing __initdata = {
+	.ncs_read_setup		= 0,
+	.nrd_setup		= 10,
+	.ncs_write_setup	= 0,
+	.nwe_setup		= 10,
+
+	.ncs_read_pulse		= 30,
+	.nrd_pulse		= 15,
+	.ncs_write_pulse	= 30,
+	.nwe_pulse		= 15,
+
+	.read_cycle		= 30,
+	.write_cycle		= 30,
+
+	.ncs_read_recover	= 0,
+	.nrd_recover		= 15,
+	.ncs_write_recover	= 0,
+	/* WE# high -> RE# low min 60 ns */
+	.nwe_recover		= 50,
+};
+
+static struct smc_config nand_config __initdata = {
+	.bus_width		= 1,
+	.nrd_controlled		= 1,
+	.nwe_controlled		= 1,
+	.nwait_mode		= 0,
+	.byte_write		= 0,
+	.tdf_cycles		= 2,
+	.tdf_mode		= 0,
+};
+
+static struct mtd_partition nand_partitions[] = {
+	{
+		.name		= "main",
+		.offset		= 0x00000000,
+		.size		= MTDPART_SIZ_FULL,
+	},
+};
+
+static struct mtd_partition *nand_part_info(int size, int *num_partitions)
+{
+	*num_partitions = ARRAY_SIZE(nand_partitions);
+	return nand_partitions;
+}
+
+struct atmel_nand_data atstk1006_nand_data __initdata = {
+	.cle		= 21,
+	.ale		= 22,
+	.rdy_pin	= GPIO_PIN_PB(30),
+	.enable_pin	= GPIO_PIN_PB(29),
+	.partition_info	= nand_part_info,
+};
+#endif
+
 struct eth_addr {
 	u8 addr[6];
 };
@@ -212,6 +280,12 @@ static int __init atstk1002_init(void)
 
 	at32_add_system_devices();
 
+#ifdef CONFIG_BOARD_ATSTK1006
+	smc_set_timing(&nand_config, &nand_timing);
+	smc_set_configuration(3, &nand_config);
+	at32_add_device_nand(0, &atstk1006_nand_data);
+#endif
+
 #ifdef	CONFIG_BOARD_ATSTK100X_SW2_CUSTOM
 	at32_add_device_usart(1);
 #else

From 90574d0a4d4b73308ae54a2a57a4f3f1fa98e984 Mon Sep 17 00:00:00 2001
From: David Woodhouse <dwmw2@infradead.org>
Date: Sat, 7 Jun 2008 08:49:00 +0100
Subject: [PATCH 0032/1171] [MTD] [NAND] Fix checkpatch warnings which showed
 up when atmel_nand.c moved

Some of them, at least.

Signed-off-by: David Woodhouse <dwmw2@infradead.org>
---
 drivers/mtd/nand/atmel_nand.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/mtd/nand/atmel_nand.c b/drivers/mtd/nand/atmel_nand.c
index 325ce29f53fc..50700ab5a57a 100644
--- a/drivers/mtd/nand/atmel_nand.c
+++ b/drivers/mtd/nand/atmel_nand.c
@@ -29,8 +29,8 @@
 #include <linux/mtd/nand.h>
 #include <linux/mtd/partitions.h>
 
-#include <asm/gpio.h>
-#include <asm/io.h>
+#include <linux/gpio.h>
+#include <linux/io.h>
 
 #include <asm/arch/board.h>
 
@@ -444,7 +444,7 @@ static int __init atmel_nand_probe(struct platform_device *pdev)
 
 	if (host->board->det_pin) {
 		if (gpio_get_value(host->board->det_pin)) {
-			printk ("No SmartMedia card inserted.\n");
+			printk("No SmartMedia card inserted.\n");
 			res = ENXIO;
 			goto err_no_card;
 		}

From fc68911ee379bff429c2f8dfc0a4d3277eb193ec Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier.adi@gmail.com>
Date: Wed, 25 Jun 2008 11:41:42 +0800
Subject: [PATCH 0033/1171] Blackfin arch: use the generic platform nand driver
 to support nand flash on bf53x board which do not have on-chip nand flash
 controller

Signed-off-by: Mike Frysinger <vapier.adi@gmail.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/mach-bf537/boards/stamp.c | 86 +++++++++++++++++++++++++
 1 file changed, 86 insertions(+)

diff --git a/arch/blackfin/mach-bf537/boards/stamp.c b/arch/blackfin/mach-bf537/boards/stamp.c
index 671f9d67f23a..7e4b9d95674b 100644
--- a/arch/blackfin/mach-bf537/boards/stamp.c
+++ b/arch/blackfin/mach-bf537/boards/stamp.c
@@ -29,9 +29,12 @@
  */
 
 #include <linux/device.h>
+#include <linux/kernel.h>
 #include <linux/platform_device.h>
 #include <linux/mtd/mtd.h>
+#include <linux/mtd/nand.h>
 #include <linux/mtd/partitions.h>
+#include <linux/mtd/plat-ram.h>
 #include <linux/mtd/physmap.h>
 #include <linux/spi/spi.h>
 #include <linux/spi/flash.h>
@@ -355,6 +358,84 @@ static struct platform_device net2272_bfin_device = {
 };
 #endif
 
+#if defined(CONFIG_MTD_NAND_PLATFORM) || defined(CONFIG_MTD_NAND_PLATFORM_MODULE)
+#ifdef CONFIG_MTD_PARTITIONS
+const char *part_probes[] = { "cmdlinepart", "RedBoot", NULL };
+
+static struct mtd_partition bfin_plat_nand_partitions[] = {
+	{
+		.name   = "linux kernel",
+		.size   = 0x400000,
+		.offset = 0,
+	}, {
+		.name   = "file system",
+		.size   = MTDPART_SIZ_FULL,
+		.offset = MTDPART_OFS_APPEND,
+	},
+};
+#endif
+
+#define BFIN_NAND_PLAT_CLE 2
+#define BFIN_NAND_PLAT_ALE 1
+static void bfin_plat_nand_cmd_ctrl(struct mtd_info *mtd, int cmd, unsigned int ctrl)
+{
+	struct nand_chip *this = mtd->priv;
+
+	if (cmd == NAND_CMD_NONE)
+		return;
+
+	if (ctrl & NAND_CLE)
+		writeb(cmd, this->IO_ADDR_W + (1 << BFIN_NAND_PLAT_CLE));
+	else
+		writeb(cmd, this->IO_ADDR_W + (1 << BFIN_NAND_PLAT_ALE));
+}
+
+#define BFIN_NAND_PLAT_READY GPIO_PF3
+static int bfin_plat_nand_dev_ready(struct mtd_info *mtd)
+{
+	return gpio_get_value(BFIN_NAND_PLAT_READY);
+}
+
+static struct platform_nand_data bfin_plat_nand_data = {
+	.chip = {
+		.chip_delay = 30,
+#ifdef CONFIG_MTD_PARTITIONS
+		.part_probe_types = part_probes,
+		.partitions = bfin_plat_nand_partitions,
+		.nr_partitions = ARRAY_SIZE(bfin_plat_nand_partitions),
+#endif
+	},
+	.ctrl = {
+		.cmd_ctrl  = bfin_plat_nand_cmd_ctrl,
+		.dev_ready = bfin_plat_nand_dev_ready,
+	},
+};
+
+#define MAX(x, y) (x > y ? x : y)
+static struct resource bfin_plat_nand_resources = {
+	.start = 0x20212000,
+	.end   = 0x20212000 + (1 << MAX(BFIN_NAND_PLAT_CLE, BFIN_NAND_PLAT_ALE)),
+	.flags = IORESOURCE_IO,
+};
+
+static struct platform_device bfin_async_nand_device = {
+	.name = "gen_nand",
+	.id = -1,
+	.num_resources = 1,
+	.resource = &bfin_plat_nand_resources,
+	.dev = {
+		.platform_data = &bfin_plat_nand_data,
+	},
+};
+
+static void bfin_plat_nand_init(void)
+{
+	gpio_request(BFIN_NAND_PLAT_READY, "bfin_nand_plat");
+}
+#else
+static void bfin_plat_nand_init(void) {}
+#endif
+
 #if defined(CONFIG_MTD_PHYSMAP) || defined(CONFIG_MTD_PHYSMAP_MODULE)
 static struct mtd_partition stamp_partitions[] = {
 	{
@@ -922,6 +1003,10 @@ static struct platform_device *stamp_devices[] __initdata = {
 
 	&bfin_gpios_device,
 
+#if defined(CONFIG_MTD_NAND_PLATFORM) || defined(CONFIG_MTD_NAND_PLATFORM_MODULE)
+	&bfin_async_nand_device,
+#endif
+
 #if defined(CONFIG_MTD_PHYSMAP) || defined(CONFIG_MTD_PHYSMAP_MODULE)
 	&stamp_flash_device,
 #endif
@@ -936,6 +1021,7 @@ static int __init stamp_init(void)
 				ARRAY_SIZE(bfin_i2c_board_info));
 #endif
 
+	bfin_plat_nand_init();
 	platform_add_devices(stamp_devices, ARRAY_SIZE(stamp_devices));
 	spi_register_board_info(bfin_spi_board_info, ARRAY_SIZE(bfin_spi_board_info));
 

From 72289824423655e67993c25c91a7a86a34917209 Mon Sep 17 00:00:00 2001
From: Michael Hennerich <michael.hennerich@analog.com>
Date: Thu, 3 Jul 2008 23:54:42 -0700
Subject: [PATCH 0034/1171] [MTD] m25p80: fix bug - ATmel spi flash fails to be
 copied to

Atmel serial flash tends to power up with the protection status bits set.
http://blackfin.uclinux.org/gf/project/uclinux-dist/tracker/?action=TrackerItemEdit&tracker_item_id=4089

[michael.hennerich@analog.com: remove duplicate code]
Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/devices/m25p80.c | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/drivers/mtd/devices/m25p80.c b/drivers/mtd/devices/m25p80.c
index b402269301f6..b35c3333e210 100644
--- a/drivers/mtd/devices/m25p80.c
+++ b/drivers/mtd/devices/m25p80.c
@@ -33,6 +33,7 @@
 /* Flash opcodes. */
 #define	OPCODE_WREN		0x06	/* Write enable */
 #define	OPCODE_RDSR		0x05	/* Read status register */
+#define	OPCODE_WRSR		0x01	/* Write status register 1 byte */
 #define	OPCODE_NORM_READ	0x03	/* Read data bytes (low frequency) */
 #define	OPCODE_FAST_READ	0x0b	/* Read data bytes (high frequency) */
 #define	OPCODE_PP		0x02	/* Page program (up to 256 bytes) */
@@ -112,6 +113,17 @@ static int read_sr(struct m25p *flash)
 	return val;
 }
 
+/*
+ * Write status register 1 byte
+ * Returns negative if error occurred.
+ */
+static int write_sr(struct m25p *flash, u8 val)
+{
+	flash->command[0] = OPCODE_WRSR;
+	flash->command[1] = val;
+
+	return spi_write(flash->spi, flash->command, 2);
+}
 
 /*
  * Set write enable latch with Write Enable command.
@@ -589,6 +601,16 @@ static int __devinit m25p_probe(struct spi_device *spi)
 	mutex_init(&flash->lock);
 	dev_set_drvdata(&spi->dev, flash);
 
+	/*
+	 * Atmel serial flash tend to power up
+	 * with the software protection bits set
+	 */
+
+	if (info->jedec_id >> 16 == 0x1f) {
+		write_enable(flash);
+		write_sr(flash, 0);
+	}
+
 	if (data && data->name)
 		flash->mtd.name = data->name;
 	else

From 5f6928378b165c4b0d57a711e1c1eb925ad33846 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Thu, 3 Jul 2008 23:40:13 -0700
Subject: [PATCH 0035/1171] [MTD] mtdchar.c silence sparse warning

The copy_to_user was casting away the address space to get the offset of
the length member.  Use offsetof() instead and add it to the void __user
*argp.

drivers/mtd/mtdchar.c:527:23: warning: cast removes address space of expression
drivers/mtd/mtdchar.c:527:23: warning: incorrect type in argument 1 (different address spaces)
drivers/mtd/mtdchar.c:527:23:    expected void [noderef] <asn:1>*to
drivers/mtd/mtdchar.c:527:23:    got unsigned int *<noident>

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/mtdchar.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/mtdchar.c b/drivers/mtd/mtdchar.c
index 4b3156f9b36f..5fc2c4216c03 100644
--- a/drivers/mtd/mtdchar.c
+++ b/drivers/mtd/mtdchar.c
@@ -479,6 +479,7 @@ static int mtd_ioctl(struct inode *inode, struct file *file,
 	{
 		struct mtd_oob_buf buf;
 		struct mtd_oob_ops ops;
+		struct mtd_oob_buf __user *user_buf = argp;
 	        uint32_t retlen;
 
 		if(!(file->f_mode & 2))
@@ -522,8 +523,7 @@ static int mtd_ioctl(struct inode *inode, struct file *file,
 		if (ops.oobretlen > 0xFFFFFFFFU)
 			ret = -EOVERFLOW;
 		retlen = ops.oobretlen;
-		if (copy_to_user(&((struct mtd_oob_buf *)argp)->length,
-				 &retlen, sizeof(buf.length)))
+		if (copy_to_user(&user_buf->length, &retlen, sizeof(buf.length)))
 			ret = -EFAULT;
 
 		kfree(ops.oobbuf);

From 175428b2b3eeacf90dcc171d5915d6b4dc86e917 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Thu, 3 Jul 2008 23:40:14 -0700
Subject: [PATCH 0036/1171] [MTD] mtdchar.c remove shadowed variable warnings

Use einfo, oinfo for the inner erase_info and otp_info structs used in
individual case statements.

drivers/mtd/mtdchar.c:582:26: warning: symbol 'info' shadows an earlier one
drivers/mtd/mtdchar.c:380:23: originally declared here
drivers/mtd/mtdchar.c:596:26: warning: symbol 'info' shadows an earlier one
drivers/mtd/mtdchar.c:380:23: originally declared here
drivers/mtd/mtdchar.c:704:19: warning: symbol 'info' shadows an earlier one
drivers/mtd/mtdchar.c:380:23: originally declared here

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/mtdchar.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/mtd/mtdchar.c b/drivers/mtd/mtdchar.c
index 5fc2c4216c03..f5061fe72e4c 100644
--- a/drivers/mtd/mtdchar.c
+++ b/drivers/mtd/mtdchar.c
@@ -577,29 +577,29 @@ static int mtd_ioctl(struct inode *inode, struct file *file,
 
 	case MEMLOCK:
 	{
-		struct erase_info_user info;
+		struct erase_info_user einfo;
 
-		if (copy_from_user(&info, argp, sizeof(info)))
+		if (copy_from_user(&einfo, argp, sizeof(einfo)))
 			return -EFAULT;
 
 		if (!mtd->lock)
 			ret = -EOPNOTSUPP;
 		else
-			ret = mtd->lock(mtd, info.start, info.length);
+			ret = mtd->lock(mtd, einfo.start, einfo.length);
 		break;
 	}
 
 	case MEMUNLOCK:
 	{
-		struct erase_info_user info;
+		struct erase_info_user einfo;
 
-		if (copy_from_user(&info, argp, sizeof(info)))
+		if (copy_from_user(&einfo, argp, sizeof(einfo)))
 			return -EFAULT;
 
 		if (!mtd->unlock)
 			ret = -EOPNOTSUPP;
 		else
-			ret = mtd->unlock(mtd, info.start, info.length);
+			ret = mtd->unlock(mtd, einfo.start, einfo.length);
 		break;
 	}
 
@@ -699,15 +699,15 @@ static int mtd_ioctl(struct inode *inode, struct file *file,
 
 	case OTPLOCK:
 	{
-		struct otp_info info;
+		struct otp_info oinfo;
 
 		if (mfi->mode != MTD_MODE_OTP_USER)
 			return -EINVAL;
-		if (copy_from_user(&info, argp, sizeof(info)))
+		if (copy_from_user(&oinfo, argp, sizeof(oinfo)))
 			return -EFAULT;
 		if (!mtd->lock_user_prot_reg)
 			return -EOPNOTSUPP;
-		ret = mtd->lock_user_prot_reg(mtd, info.start, info.length);
+		ret = mtd->lock_user_prot_reg(mtd, oinfo.start, oinfo.length);
 		break;
 	}
 #endif

From 23a346ca4a5a6f50f81062456af955155f68e313 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Thu, 3 Jul 2008 23:40:16 -0700
Subject: [PATCH 0037/1171] [MTD] [NAND] atmel_nand speedup via
 {read,write}s{b,w}()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This uses __raw_{read,write}s{b,w}() primitives to access data on NAND
chips for more efficient I/O.

On an arm926 with memory clocked at 100 MHz, this reduced the elapsed time
for a 64 MiB read by 16%.  ("dd" /dev/mtd0 to /dev/null, with an 8-bit
NAND using hardware ECC and 128KiB blocksize.)

Also some minor section tweaks:

  - Use platform_driver_probe() so no pointer to probe() lingers
    after that code has been removed at run-time.

  - Use __exit and __exit_p so the remove() code will normally be
    removed by the linker.

Since these buffer read/write calls are new, this increases the runtime
code footprint (by 88 bytes on my build, after the section tweaks).

[haavard.skinnemoen@atmel.com: rebase onto atmel_nand rename]
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Håvard Skinnemoen <haavard.skinnemoen@atmel.com>
Acked-by: Andrew Victor <linux@maxim.org.za>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/atmel_nand.c | 46 +++++++++++++++++++++++++++++++----
 1 file changed, 41 insertions(+), 5 deletions(-)

diff --git a/drivers/mtd/nand/atmel_nand.c b/drivers/mtd/nand/atmel_nand.c
index 50700ab5a57a..4814fc9b237b 100644
--- a/drivers/mtd/nand/atmel_nand.c
+++ b/drivers/mtd/nand/atmel_nand.c
@@ -141,6 +141,37 @@ static int atmel_nand_device_ready(struct mtd_info *mtd)
 	return gpio_get_value(host->board->rdy_pin);
 }
 
+/*
+ * Minimal-overhead PIO for data access.
+ */
+static void atmel_read_buf(struct mtd_info *mtd, u8 *buf, int len)
+{
+	struct nand_chip	*nand_chip = mtd->priv;
+
+	__raw_readsb(nand_chip->IO_ADDR_R, buf, len);
+}
+
+static void atmel_read_buf16(struct mtd_info *mtd, u8 *buf, int len)
+{
+	struct nand_chip	*nand_chip = mtd->priv;
+
+	__raw_readsw(nand_chip->IO_ADDR_R, buf, len / 2);
+}
+
+static void atmel_write_buf(struct mtd_info *mtd, const u8 *buf, int len)
+{
+	struct nand_chip	*nand_chip = mtd->priv;
+
+	__raw_writesb(nand_chip->IO_ADDR_W, buf, len);
+}
+
+static void atmel_write_buf16(struct mtd_info *mtd, const u8 *buf, int len)
+{
+	struct nand_chip	*nand_chip = mtd->priv;
+
+	__raw_writesw(nand_chip->IO_ADDR_W, buf, len / 2);
+}
+
 /*
  * write oob for small pages
  */
@@ -436,8 +467,14 @@ static int __init atmel_nand_probe(struct platform_device *pdev)
 
 	nand_chip->chip_delay = 20;		/* 20us command delay time */
 
-	if (host->board->bus_width_16)		/* 16-bit bus width */
+	if (host->board->bus_width_16) {	/* 16-bit bus width */
 		nand_chip->options |= NAND_BUSWIDTH_16;
+		nand_chip->read_buf = atmel_read_buf16;
+		nand_chip->write_buf = atmel_write_buf16;
+	} else {
+		nand_chip->read_buf = atmel_read_buf;
+		nand_chip->write_buf = atmel_write_buf;
+	}
 
 	platform_set_drvdata(pdev, host);
 	atmel_nand_enable(host);
@@ -546,7 +583,7 @@ err_nand_ioremap:
 /*
  * Remove a NAND device.
  */
-static int __devexit atmel_nand_remove(struct platform_device *pdev)
+static int __exit atmel_nand_remove(struct platform_device *pdev)
 {
 	struct atmel_nand_host *host = platform_get_drvdata(pdev);
 	struct mtd_info *mtd = &host->mtd;
@@ -564,8 +601,7 @@ static int __devexit atmel_nand_remove(struct platform_device *pdev)
 }
 
 static struct platform_driver atmel_nand_driver = {
-	.probe		= atmel_nand_probe,
-	.remove		= atmel_nand_remove,
+	.remove		= __exit_p(atmel_nand_remove),
 	.driver		= {
 		.name	= "atmel_nand",
 		.owner	= THIS_MODULE,
@@ -574,7 +610,7 @@ static struct platform_driver atmel_nand_driver = {
 
 static int __init atmel_nand_init(void)
 {
-	return platform_driver_register(&atmel_nand_driver);
+	return platform_driver_probe(&atmel_nand_driver, atmel_nand_probe);
 }
 
 

From d6248fddf717041f25781050e6392cc76525272d Mon Sep 17 00:00:00 2001
From: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Date: Thu, 3 Jul 2008 23:40:18 -0700
Subject: [PATCH 0038/1171] [MTD] [NAND] atmel_nand: Work around AT32AP7000 ECC
 erratum
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The ALE signal isn't correctly wired up to the ECC controller on the
AP7000, so it starts calculating ECC during the address cycles.

Work around this by resetting the ECC controller between the address and
data cycles.

Signed-off-by: Håvard Skinnemoen <haavard.skinnemoen@atmel.com>
Acked-by: Andrew Victor <linux@maxim.org.za>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/atmel_nand.c | 25 +++++++++++++++++++++++--
 1 file changed, 23 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/nand/atmel_nand.c b/drivers/mtd/nand/atmel_nand.c
index 4814fc9b237b..99aec46e2145 100644
--- a/drivers/mtd/nand/atmel_nand.c
+++ b/drivers/mtd/nand/atmel_nand.c
@@ -33,6 +33,7 @@
 #include <linux/io.h>
 
 #include <asm/arch/board.h>
+#include <asm/arch/cpu.h>
 
 #ifdef CONFIG_MTD_NAND_ATMEL_ECC_HW
 #define hard_ecc	1
@@ -264,6 +265,19 @@ static int atmel_nand_read_page(struct mtd_info *mtd,
 	uint8_t *ecc_pos;
 	int stat;
 
+	/*
+	 * Errata: ALE is incorrectly wired up to the ECC controller
+	 * on the AP7000, so it will include the address cycles in the
+	 * ECC calculation.
+	 *
+	 * Workaround: Reset the parity registers before reading the
+	 * actual data.
+	 */
+	if (cpu_is_at32ap7000()) {
+		struct atmel_nand_host *host = chip->priv;
+		ecc_writel(host->ecc, CR, ATMEL_ECC_RST);
+	}
+
 	/* read the page */
 	chip->read_buf(mtd, p, eccsize);
 
@@ -377,9 +391,16 @@ static int atmel_nand_correct(struct mtd_info *mtd, u_char *dat,
 }
 
 /*
- * Enable HW ECC : unsused
+ * Enable HW ECC : unused on most chips
  */
-static void atmel_nand_hwctl(struct mtd_info *mtd, int mode) { ; }
+static void atmel_nand_hwctl(struct mtd_info *mtd, int mode)
+{
+	if (cpu_is_at32ap7000()) {
+		struct nand_chip *nand_chip = mtd->priv;
+		struct atmel_nand_host *host = nand_chip->priv;
+		ecc_writel(host->ecc, CR, ATMEL_ECC_RST);
+	}
+}
 
 #ifdef CONFIG_MTD_PARTITIONS
 static const char *part_probes[] = { "cmdlinepart", NULL };

From bd5a43822b438f297f4088f1cfd3514e32e56328 Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Thu, 3 Jul 2008 23:40:19 -0700
Subject: [PATCH 0039/1171] [MTD] [NAND] atmel_nand can be modular
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

There's no reason to prevent the Atmel NAND driver from
building as a module.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Håvard Skinnemoen <haavard.skinnemoen@atmel.com>
Acked-by: Andrew Victor <linux@maxim.org.za>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/nand/Kconfig b/drivers/mtd/nand/Kconfig
index d8c0d8698b49..71406e517857 100644
--- a/drivers/mtd/nand/Kconfig
+++ b/drivers/mtd/nand/Kconfig
@@ -272,7 +272,7 @@ config MTD_NAND_CS553X
 	  If you say "m", the module will be called "cs553x_nand.ko".
 
 config MTD_NAND_ATMEL
-	bool "Support for NAND Flash / SmartMedia on AT91 and AVR32"
+	tristate "Support for NAND Flash / SmartMedia on AT91 and AVR32"
 	depends on ARCH_AT91 || AVR32
 	help
 	  Enables support for NAND Flash / Smart Media Card interface

From 452db2724351ff3d9416a183a7955e00ab4e6ab4 Mon Sep 17 00:00:00 2001
From: Anton Vorontsov <avorontsov@ru.mvista.com>
Date: Fri, 27 Jun 2008 23:04:04 +0400
Subject: [PATCH 0040/1171] [MTD] [NAND] fsl_elbc_nand: fix OOB workability for
 large page NAND chips

For large page chips, nand_bbt is looking into OOB area, and checking
for "0xff 0xff" pattern at OOB offset 0. That is, two bytes should be
reserved for bbt means.

But ELBC driver is specifying ecclayout so that oobfree area starts at
offset 1, so only one byte left for the bbt purposes.

This causes problems with any OOB users, namely JFFS2: after first mount
JFFS2 will fill all OOBs with "erased marker", so OOBs will contain:

  OOB Data: ff 19 85 20 03 00 ff ff ff 00 00 08 ff ff ff ff
  OOB Data: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
  OOB Data: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
  OOB Data: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff

And on the next boot, NAND core will rescan for bad blocks, then will
see "0xff 0x19" pattern, and will mark all blocks as bad ones.

To fix the issue we should implement our own bad block pattern: just one
byte at OOB start. Though, this will work only for x8 chips. For x16
chips two bytes must be checked. Since ELBC driver does not support x16
NANDs (yet), we're safe for now.

Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Acked-by: Scott Wood <scottwood@freescale.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/fsl_elbc_nand.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/drivers/mtd/nand/fsl_elbc_nand.c b/drivers/mtd/nand/fsl_elbc_nand.c
index 1b06d29dd06b..99dc2be620a6 100644
--- a/drivers/mtd/nand/fsl_elbc_nand.c
+++ b/drivers/mtd/nand/fsl_elbc_nand.c
@@ -116,6 +116,20 @@ static struct nand_ecclayout fsl_elbc_oob_lp_eccm1 = {
 	.oobavail = 48,
 };
 
+/*
+ * fsl_elbc_oob_lp_eccm* specify that LP NAND's OOB free area starts at offset
+ * 1, so we have to adjust bad block pattern. This pattern should be used for
+ * x8 chips only. So far hardware does not support x16 chips anyway.
+ */
+static u8 scan_ff_pattern[] = { 0xff, };
+
+static struct nand_bbt_descr largepage_memorybased = {
+	.options = 0,
+	.offs = 0,
+	.len = 1,
+	.pattern = scan_ff_pattern,
+};
+
 /*=================================*/
 
 /*
@@ -687,6 +701,7 @@ static int fsl_elbc_chip_init_tail(struct mtd_info *mtd)
 			chip->ecc.layout = (priv->fmr & FMR_ECCM) ?
 			                   &fsl_elbc_oob_lp_eccm1 :
 			                   &fsl_elbc_oob_lp_eccm0;
+			chip->badblock_pattern = &largepage_memorybased;
 			mtd->ecclayout = chip->ecc.layout;
 			mtd->oobavail = chip->ecc.layout->oobavail;
 		}

From ec6e0ea3bdf82ee9761d324c011c3627821f7410 Mon Sep 17 00:00:00 2001
From: Anton Vorontsov <avorontsov@ru.mvista.com>
Date: Fri, 27 Jun 2008 23:04:13 +0400
Subject: [PATCH 0041/1171] [MTD] [NAND] fsl_elbc_nand: implement support for
 flash-based BBT

This patch implements support for flash-based BBT for chips working
through ELBC NAND controller, so that NAND core will not have to re-scan
for bad blocks on every boot.

Because ELBC controller may provide HW-generated ECCs we should adjust
bbt pattern and bbt version positions in the OOB free area.

Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Acked-by: Scott Wood <scottwood@freescale.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/fsl_elbc_nand.c | 34 +++++++++++++++++++++++++++++++-
 1 file changed, 33 insertions(+), 1 deletion(-)

diff --git a/drivers/mtd/nand/fsl_elbc_nand.c b/drivers/mtd/nand/fsl_elbc_nand.c
index 99dc2be620a6..5f1bc5ea2a73 100644
--- a/drivers/mtd/nand/fsl_elbc_nand.c
+++ b/drivers/mtd/nand/fsl_elbc_nand.c
@@ -130,6 +130,34 @@ static struct nand_bbt_descr largepage_memorybased = {
 	.pattern = scan_ff_pattern,
 };
 
+/*
+ * ELBC may use HW ECC, so that OOB offsets, that NAND core uses for bbt,
+ * interfere with ECC positions, that's why we implement our own descriptors.
+ * OOB {11, 5}, works for both SP and LP chips, with ECCM = 1 and ECCM = 0.
+ */
+static u8 bbt_pattern[] = {'B', 'b', 't', '0' };
+static u8 mirror_pattern[] = {'1', 't', 'b', 'B' };
+
+static struct nand_bbt_descr bbt_main_descr = {
+	.options = NAND_BBT_LASTBLOCK | NAND_BBT_CREATE | NAND_BBT_WRITE |
+		   NAND_BBT_2BIT | NAND_BBT_VERSION,
+	.offs =	11,
+	.len = 4,
+	.veroffs = 15,
+	.maxblocks = 4,
+	.pattern = bbt_pattern,
+};
+
+static struct nand_bbt_descr bbt_mirror_descr = {
+	.options = NAND_BBT_LASTBLOCK | NAND_BBT_CREATE | NAND_BBT_WRITE |
+		   NAND_BBT_2BIT | NAND_BBT_VERSION,
+	.offs =	11,
+	.len = 4,
+	.veroffs = 15,
+	.maxblocks = 4,
+	.pattern = mirror_pattern,
+};
+
 /*=================================*/
 
 /*
@@ -767,8 +795,12 @@ static int fsl_elbc_chip_init(struct fsl_elbc_mtd *priv)
 	chip->cmdfunc = fsl_elbc_cmdfunc;
 	chip->waitfunc = fsl_elbc_wait;
 
+	chip->bbt_td = &bbt_main_descr;
+	chip->bbt_md = &bbt_mirror_descr;
+
 	/* set up nand options */
-	chip->options = NAND_NO_READRDY | NAND_NO_AUTOINCR;
+	chip->options = NAND_NO_READRDY | NAND_NO_AUTOINCR |
+			NAND_USE_FLASH_BBT;
 
 	chip->controller = &ctrl->controller;
 	chip->priv = priv;

From 0acf944c6853813ed19cdf46d4042a77dd878ab5 Mon Sep 17 00:00:00 2001
From: Anton Vorontsov <avorontsov@ru.mvista.com>
Date: Fri, 27 Jun 2008 23:04:20 +0400
Subject: [PATCH 0042/1171] [MTD] [NAND] fsl_elbc_nand: ecclayout cleanups

This patch deletes oobavail assignments, they're calculated by the nand
core code in nand_scan_tail, plus current oobavail values are wrong for
the LP NANDs.

Also remove mtd->ecclayout and mtd->oobavail assignments, mtd core
handles this all by itself.

Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/fsl_elbc_nand.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/drivers/mtd/nand/fsl_elbc_nand.c b/drivers/mtd/nand/fsl_elbc_nand.c
index 5f1bc5ea2a73..d6d1ff55c4e5 100644
--- a/drivers/mtd/nand/fsl_elbc_nand.c
+++ b/drivers/mtd/nand/fsl_elbc_nand.c
@@ -89,7 +89,6 @@ static struct nand_ecclayout fsl_elbc_oob_sp_eccm0 = {
 	.eccbytes = 3,
 	.eccpos = {6, 7, 8},
 	.oobfree = { {0, 5}, {9, 7} },
-	.oobavail = 12,
 };
 
 /* Small Page FLASH with FMR[ECCM] = 1 */
@@ -97,7 +96,6 @@ static struct nand_ecclayout fsl_elbc_oob_sp_eccm1 = {
 	.eccbytes = 3,
 	.eccpos = {8, 9, 10},
 	.oobfree = { {0, 5}, {6, 2}, {11, 5} },
-	.oobavail = 12,
 };
 
 /* Large Page FLASH with FMR[ECCM] = 0 */
@@ -105,7 +103,6 @@ static struct nand_ecclayout fsl_elbc_oob_lp_eccm0 = {
 	.eccbytes = 12,
 	.eccpos = {6, 7, 8, 22, 23, 24, 38, 39, 40, 54, 55, 56},
 	.oobfree = { {1, 5}, {9, 13}, {25, 13}, {41, 13}, {57, 7} },
-	.oobavail = 48,
 };
 
 /* Large Page FLASH with FMR[ECCM] = 1 */
@@ -113,7 +110,6 @@ static struct nand_ecclayout fsl_elbc_oob_lp_eccm1 = {
 	.eccbytes = 12,
 	.eccpos = {8, 9, 10, 24, 25, 26, 40, 41, 42, 56, 57, 58},
 	.oobfree = { {1, 7}, {11, 13}, {27, 13}, {43, 13}, {59, 5} },
-	.oobavail = 48,
 };
 
 /*
@@ -730,8 +726,6 @@ static int fsl_elbc_chip_init_tail(struct mtd_info *mtd)
 			                   &fsl_elbc_oob_lp_eccm1 :
 			                   &fsl_elbc_oob_lp_eccm0;
 			chip->badblock_pattern = &largepage_memorybased;
-			mtd->ecclayout = chip->ecc.layout;
-			mtd->oobavail = chip->ecc.layout->oobavail;
 		}
 	} else {
 		dev_err(ctrl->dev,

From 6f40470e745693ee4ad85edb441b0aad5cae3f00 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Wed, 9 Jul 2008 14:09:20 +0100
Subject: [PATCH 0043/1171] [MTD] [MAPS] Remove the bast-flash driver.

Remove the Simtec BAST flash driver as this has been replaced by using
the platform flash driver.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/maps/Kconfig      |  18 ---
 drivers/mtd/maps/Makefile     |   1 -
 drivers/mtd/maps/bast-flash.c | 224 ----------------------------------
 3 files changed, 243 deletions(-)
 delete mode 100644 drivers/mtd/maps/bast-flash.c

diff --git a/drivers/mtd/maps/Kconfig b/drivers/mtd/maps/Kconfig
index d2d339a7598d..ef1e29ea5a2c 100644
--- a/drivers/mtd/maps/Kconfig
+++ b/drivers/mtd/maps/Kconfig
@@ -556,24 +556,6 @@ config MTD_DMV182
         help
           Map driver for Dy-4 SVME/DMV-182 board.
 
-config MTD_BAST
-	tristate "Map driver for Simtec BAST (EB2410ITX) or Thorcom VR1000"
-	depends on ARCH_BAST || MACH_VR1000
-	select MTD_PARTITIONS
-	select MTD_MAP_BANK_WIDTH_16
-	select MTD_JEDECPROBE
-	help
-	  Map driver for NOR flash on the Simtec BAST (EB2410ITX), or the
-	  Thorcom VR1000
-
-	  Note, this driver *cannot* over-ride the WP link on the
-	  board, or currently detect the state of the link.
-
-config MTD_BAST_MAXSIZE
-	int "Maximum size for BAST flash area (MiB)"
-	depends on MTD_BAST
-	default "4"
-
 config MTD_SHARP_SL
 	tristate "ROM mapped on Sharp SL Series"
 	depends on ARCH_PXA
diff --git a/drivers/mtd/maps/Makefile b/drivers/mtd/maps/Makefile
index f3b0c5951663..b29ea5460657 100644
--- a/drivers/mtd/maps/Makefile
+++ b/drivers/mtd/maps/Makefile
@@ -9,7 +9,6 @@ endif
 # Chip mappings
 obj-$(CONFIG_MTD_CDB89712)	+= cdb89712.o
 obj-$(CONFIG_MTD_ARM_INTEGRATOR)+= integrator-flash.o
-obj-$(CONFIG_MTD_BAST)		+= bast-flash.o
 obj-$(CONFIG_MTD_CFI_FLAGADM)	+= cfi_flagadm.o
 obj-$(CONFIG_MTD_DC21285)	+= dc21285.o
 obj-$(CONFIG_MTD_DILNETPC)	+= dilnetpc.o
diff --git a/drivers/mtd/maps/bast-flash.c b/drivers/mtd/maps/bast-flash.c
deleted file mode 100644
index ca5414880341..000000000000
--- a/drivers/mtd/maps/bast-flash.c
+++ /dev/null
@@ -1,224 +0,0 @@
-/* linux/drivers/mtd/maps/bast-flash.c
- *
- * Copyright (c) 2004-2005 Simtec Electronics
- *	Ben Dooks <ben@simtec.co.uk>
- *
- * Simtec Bast (EB2410ITX) NOR MTD Mapping driver
- *
- * Changelog:
- *	20-Sep-2004  BJD  Initial version
- *	17-Jan-2005  BJD  Add whole device if no partitions found
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
-*/
-
-#include <linux/module.h>
-#include <linux/types.h>
-#include <linux/init.h>
-#include <linux/kernel.h>
-#include <linux/string.h>
-#include <linux/ioport.h>
-#include <linux/device.h>
-#include <linux/slab.h>
-#include <linux/platform_device.h>
-#include <linux/mtd/mtd.h>
-#include <linux/mtd/map.h>
-#include <linux/mtd/partitions.h>
-
-#include <asm/io.h>
-#include <asm/mach/flash.h>
-
-#include <asm/arch/map.h>
-#include <asm/arch/bast-map.h>
-#include <asm/arch/bast-cpld.h>
-
-#ifdef CONFIG_MTD_BAST_MAXSIZE
-#define AREA_MAXSIZE (CONFIG_MTD_BAST_MAXSIZE * SZ_1M)
-#else
-#define AREA_MAXSIZE (32 * SZ_1M)
-#endif
-
-#define PFX "bast-flash: "
-
-struct bast_flash_info {
-	struct mtd_info		*mtd;
-	struct map_info		 map;
-	struct mtd_partition	*partitions;
-	struct resource		*area;
-};
-
-static const char *probes[] = { "RedBoot", "cmdlinepart", NULL };
-
-static void bast_flash_setrw(int to)
-{
-	unsigned int val;
-	unsigned long flags;
-
-	local_irq_save(flags);
-	val = __raw_readb(BAST_VA_CTRL3);
-
-	if (to)
-		val |= BAST_CPLD_CTRL3_ROMWEN;
-	else
-		val &= ~BAST_CPLD_CTRL3_ROMWEN;
-
-	pr_debug("new cpld ctrl3=%02x\n", val);
-
-	__raw_writeb(val, BAST_VA_CTRL3);
-	local_irq_restore(flags);
-}
-
-static int bast_flash_remove(struct platform_device *pdev)
-{
-	struct bast_flash_info *info = platform_get_drvdata(pdev);
-
-	platform_set_drvdata(pdev, NULL);
-
-	if (info == NULL)
-		return 0;
-
-	if (info->map.virt != NULL)
-		iounmap(info->map.virt);
-
-	if (info->mtd) {
-		del_mtd_partitions(info->mtd);
-		map_destroy(info->mtd);
-	}
-
-	kfree(info->partitions);
-
-	if (info->area) {
-		release_resource(info->area);
-		kfree(info->area);
-	}
-
-	kfree(info);
-
-	return 0;
-}
-
-static int bast_flash_probe(struct platform_device *pdev)
-{
-	struct bast_flash_info *info;
-	struct resource *res;
-	int err = 0;
-
-	info = kmalloc(sizeof(*info), GFP_KERNEL);
-	if (info == NULL) {
-		printk(KERN_ERR PFX "no memory for flash info\n");
-		err = -ENOMEM;
-		goto exit_error;
-	}
-
-	memzero(info, sizeof(*info));
-	platform_set_drvdata(pdev, info);
-
-	res = pdev->resource;  /* assume that the flash has one resource */
-
-	info->map.phys = res->start;
-	info->map.size = res->end - res->start + 1;
-	info->map.name = pdev->dev.bus_id;
-	info->map.bankwidth = 2;
-
-	if (info->map.size > AREA_MAXSIZE)
-		info->map.size = AREA_MAXSIZE;
-
-	pr_debug("%s: area %08lx, size %ld\n", __func__,
-		 info->map.phys, info->map.size);
-
-	info->area = request_mem_region(res->start, info->map.size,
-					pdev->name);
-	if (info->area == NULL) {
-		printk(KERN_ERR PFX "cannot reserve flash memory region\n");
-		err = -ENOENT;
-		goto exit_error;
-	}
-
-	info->map.virt = ioremap(res->start, info->map.size);
-	pr_debug("%s: virt at %08x\n", __func__, (int)info->map.virt);
-
-	if (info->map.virt == 0) {
-		printk(KERN_ERR PFX "failed to ioremap() region\n");
-		err = -EIO;
-		goto exit_error;
-	}
-
-	simple_map_init(&info->map);
-
-	/* enable the write to the flash area */
-
-	bast_flash_setrw(1);
-
-	/* probe for the device(s) */
-
-	info->mtd = do_map_probe("jedec_probe", &info->map);
-	if (info->mtd == NULL)
-		info->mtd = do_map_probe("cfi_probe", &info->map);
-
-	if (info->mtd == NULL) {
-		printk(KERN_ERR PFX "map_probe() failed\n");
-		err = -ENXIO;
-		goto exit_error;
-	}
-
-	/* mark ourselves as the owner */
-	info->mtd->owner = THIS_MODULE;
-
-	err = parse_mtd_partitions(info->mtd, probes, &info->partitions, 0);
-	if (err > 0) {
-		err = add_mtd_partitions(info->mtd, info->partitions, err);
-		if (err)
-			printk(KERN_ERR PFX "cannot add/parse partitions\n");
-	} else {
-		err = add_mtd_device(info->mtd);
-	}
-
-	if (err == 0)
-		return 0;
-
-	/* fall through to exit error */
-
- exit_error:
-	bast_flash_remove(pdev);
-	return err;
-}
-
-static struct platform_driver bast_flash_driver = {
-	.probe		= bast_flash_probe,
-	.remove		= bast_flash_remove,
-	.driver		= {
-		.name	= "bast-nor",
-		.owner	= THIS_MODULE,
-	},
-};
-
-static int __init bast_flash_init(void)
-{
-	printk("BAST NOR-Flash Driver, (c) 2004 Simtec Electronics\n");
-	return platform_driver_register(&bast_flash_driver);
-}
-
-static void __exit bast_flash_exit(void)
-{
-	platform_driver_unregister(&bast_flash_driver);
-}
-
-module_init(bast_flash_init);
-module_exit(bast_flash_exit);
-
-MODULE_LICENSE("GPL");
-MODULE_AUTHOR("Ben Dooks <ben@simtec.co.uk>");
-MODULE_DESCRIPTION("BAST MTD Map driver");
-MODULE_ALIAS("platform:bast-nor");

From f63af11ddb508ce7b2a270515244d145248cad7f Mon Sep 17 00:00:00 2001
From: Milton Miller <miltonm@bga.com>
Date: Thu, 10 Jul 2008 16:14:18 -0500
Subject: [PATCH 0044/1171] [MTD] [NAND] remove __PPC__ hardcoded address from
 DiskOnChip drivers

Such a hardcoded address can cause a checkstop or machine check if
the driver is in the kernel but the address is not acknowledged.

Both drivers allow an address to be specified as either a module
parameter or config option.   Any future powerpc board should either
use one of these methods or find the address in the device tree.

Signed-off-by: Milton Miller <miltonm@bga.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/devices/docprobe.c | 2 --
 drivers/mtd/nand/diskonchip.c  | 2 --
 2 files changed, 4 deletions(-)

diff --git a/drivers/mtd/devices/docprobe.c b/drivers/mtd/devices/docprobe.c
index 6e5d811ae83a..6e62922942b1 100644
--- a/drivers/mtd/devices/docprobe.c
+++ b/drivers/mtd/devices/docprobe.c
@@ -76,8 +76,6 @@ static unsigned long __initdata doc_locations[] = {
 	0xe0000, 0xe2000, 0xe4000, 0xe6000,
 	0xe8000, 0xea000, 0xec000, 0xee000,
 #endif /*  CONFIG_MTD_DOCPROBE_HIGH */
-#elif defined(__PPC__)
-	0xe4000000,
 #else
 #warning Unknown architecture for DiskOnChip. No default probe locations defined
 #endif
diff --git a/drivers/mtd/nand/diskonchip.c b/drivers/mtd/nand/diskonchip.c
index cd4393ce2562..765d4f0f7c86 100644
--- a/drivers/mtd/nand/diskonchip.c
+++ b/drivers/mtd/nand/diskonchip.c
@@ -52,8 +52,6 @@ static unsigned long __initdata doc_locations[] = {
 	0xe0000, 0xe2000, 0xe4000, 0xe6000,
 	0xe8000, 0xea000, 0xec000, 0xee000,
 #endif /*  CONFIG_MTD_DOCPROBE_HIGH */
-#elif defined(__PPC__)
-	0xe4000000,
 #else
 #warning Unknown architecture for DiskOnChip. No default probe locations defined
 #endif

From 3a3688b6af103e2c86a7cfc2050988655e184ecc Mon Sep 17 00:00:00 2001
From: Jean Delvare <khali@linux-fr.org>
Date: Thu, 10 Jul 2008 13:37:08 +0200
Subject: [PATCH 0045/1171] [MTD] [NOR] gen_probe: No debug message when
 debugging is disabled

Use pr_debug(...) instead of printk(KERN_DEBUG ...) so that the message
is only printed when debugging is enabled.

Signed-off-by: Jean Delvare <khali@linux-fr.org>
Tested-by: John stoffel <john@stoffel.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/chips/gen_probe.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/chips/gen_probe.c b/drivers/mtd/chips/gen_probe.c
index e53a58ae384f..f061885b2812 100644
--- a/drivers/mtd/chips/gen_probe.c
+++ b/drivers/mtd/chips/gen_probe.c
@@ -70,8 +70,8 @@ static struct cfi_private *genprobe_ident_chips(struct map_info *map, struct chi
 	   interleave and device type, etc. */
 	if (!genprobe_new_chip(map, cp, &cfi)) {
 		/* The probe didn't like it */
-		printk(KERN_DEBUG "%s: Found no %s device at location zero\n",
-		       cp->name, map->name);
+		pr_debug("%s: Found no %s device at location zero\n",
+			 cp->name, map->name);
 		return NULL;
 	}
 

From 36560d255b017dbcaefbf0f8fee7ad4dd1f0fe0a Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Tue, 8 Jul 2008 17:09:03 +0100
Subject: [PATCH 0046/1171] [MTD] Fix const assignment in the MTD command line
 partitioning driver

Fix const to non-const pointer assignment in the MTD command line partitioning
driver.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/cmdlinepart.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/cmdlinepart.c b/drivers/mtd/cmdlinepart.c
index 68782ab2f0de..71bc07f149b7 100644
--- a/drivers/mtd/cmdlinepart.c
+++ b/drivers/mtd/cmdlinepart.c
@@ -306,7 +306,7 @@ static int parse_cmdline_partitions(struct mtd_info *master,
 	unsigned long offset;
 	int i;
 	struct cmdline_mtd_partition *part;
-	char *mtd_id = master->name;
+	const char *mtd_id = master->name;
 
 	/* parse command line */
 	if (!cmdline_parsed)

From 0533400b7813df6c22a171499434d30bd57e799c Mon Sep 17 00:00:00 2001
From: Stoyan Gaydarov <stoyboyker@gmail.com>
Date: Mon, 7 Jul 2008 07:45:59 -0500
Subject: [PATCH 0047/1171] [JFFS2] Use .unlocked_ioctl

This changes the .ioctl to the .unlocked_ioctl version.

Signed-off-by: Stoyan Gaydarov <stoyboyker@gmail.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 fs/jffs2/dir.c      | 2 +-
 fs/jffs2/file.c     | 2 +-
 fs/jffs2/ioctl.c    | 3 +--
 fs/jffs2/os-linux.h | 2 +-
 4 files changed, 4 insertions(+), 5 deletions(-)

diff --git a/fs/jffs2/dir.c b/fs/jffs2/dir.c
index c0c141f6fde1..cd219ef55254 100644
--- a/fs/jffs2/dir.c
+++ b/fs/jffs2/dir.c
@@ -38,7 +38,7 @@ const struct file_operations jffs2_dir_operations =
 {
 	.read =		generic_read_dir,
 	.readdir =	jffs2_readdir,
-	.ioctl =	jffs2_ioctl,
+	.unlocked_ioctl=jffs2_ioctl,
 	.fsync =	jffs2_fsync
 };
 
diff --git a/fs/jffs2/file.c b/fs/jffs2/file.c
index 5e920343b2c5..5a98aa87c853 100644
--- a/fs/jffs2/file.c
+++ b/fs/jffs2/file.c
@@ -46,7 +46,7 @@ const struct file_operations jffs2_file_operations =
  	.aio_read =	generic_file_aio_read,
  	.write =	do_sync_write,
  	.aio_write =	generic_file_aio_write,
-	.ioctl =	jffs2_ioctl,
+	.unlocked_ioctl=jffs2_ioctl,
 	.mmap =		generic_file_readonly_mmap,
 	.fsync =	jffs2_fsync,
 	.splice_read =	generic_file_splice_read,
diff --git a/fs/jffs2/ioctl.c b/fs/jffs2/ioctl.c
index e2177210f621..9d41f43e47bb 100644
--- a/fs/jffs2/ioctl.c
+++ b/fs/jffs2/ioctl.c
@@ -12,8 +12,7 @@
 #include <linux/fs.h>
 #include "nodelist.h"
 
-int jffs2_ioctl(struct inode *inode, struct file *filp, unsigned int cmd,
-		unsigned long arg)
+long jffs2_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 {
 	/* Later, this will provide for lsattr.jffs2 and chattr.jffs2, which
 	   will include compression support etc. */
diff --git a/fs/jffs2/os-linux.h b/fs/jffs2/os-linux.h
index 2cc866cf134f..5e194a5c8e29 100644
--- a/fs/jffs2/os-linux.h
+++ b/fs/jffs2/os-linux.h
@@ -167,7 +167,7 @@ int jffs2_fsync(struct file *, struct dentry *, int);
 int jffs2_do_readpage_unlock (struct inode *inode, struct page *pg);
 
 /* ioctl.c */
-int jffs2_ioctl(struct inode *, struct file *, unsigned int, unsigned long);
+long jffs2_ioctl(struct file *, unsigned int, unsigned long);
 
 /* symlink.c */
 extern const struct inode_operations jffs2_symlink_inode_operations;

From 1f2d18690f4390ac4cf75648a5bc18fc07b3aef6 Mon Sep 17 00:00:00 2001
From: Michael Hennerich <michael.hennerich@analog.com>
Date: Mon, 14 Jul 2008 16:31:22 +0800
Subject: [PATCH 0048/1171] Blackfin arch: Fix Bug - GPIO interrupts not
 disabled; edge sensitive interrupt hang system.

 - Implement irq_chip.enable and irq_chip.disable

Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/mach-common/ints-priority.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/arch/blackfin/mach-common/ints-priority.c b/arch/blackfin/mach-common/ints-priority.c
index f5fd768022ea..64d746114e4b 100644
--- a/arch/blackfin/mach-common/ints-priority.c
+++ b/arch/blackfin/mach-common/ints-priority.c
@@ -459,6 +459,8 @@ static struct irq_chip bfin_gpio_irqchip = {
 	.mask = bfin_gpio_mask_irq,
 	.mask_ack = bfin_gpio_mask_ack_irq,
 	.unmask = bfin_gpio_unmask_irq,
+	.disable = bfin_gpio_mask_irq,
+	.enable = bfin_gpio_unmask_irq,
 	.set_type = bfin_gpio_irq_type,
 	.startup = bfin_gpio_irq_startup,
 	.shutdown = bfin_gpio_irq_shutdown,
@@ -846,6 +848,8 @@ static struct irq_chip bfin_gpio_irqchip = {
 	.mask = bfin_gpio_mask_irq,
 	.mask_ack = bfin_gpio_mask_ack_irq,
 	.unmask = bfin_gpio_unmask_irq,
+	.disable = bfin_gpio_mask_irq,
+	.enable = bfin_gpio_unmask_irq,
 	.set_type = bfin_gpio_irq_type,
 	.startup = bfin_gpio_irq_startup,
 	.shutdown = bfin_gpio_irq_shutdown,

From 260d5d3517c67c5b68b4e28c5d3e1e3b73976a90 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier.adi@gmail.com>
Date: Mon, 14 Jul 2008 16:34:05 +0800
Subject: [PATCH 0049/1171] Blackfin arch: Fix bug - do not overflow the buffer
 given to us which tends to happen when CONFIG_L1_MAX_PIECE is increased past
 its default

Singed-off-by: Mike Frysinger <vapier.adi@gmail.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/mm/blackfin_sram.c | 38 ++++++++++++++++++++++----------
 1 file changed, 26 insertions(+), 12 deletions(-)

diff --git a/arch/blackfin/mm/blackfin_sram.c b/arch/blackfin/mm/blackfin_sram.c
index 3246f91c7baa..8f6fdc245330 100644
--- a/arch/blackfin/mm/blackfin_sram.c
+++ b/arch/blackfin/mm/blackfin_sram.c
@@ -549,13 +549,16 @@ EXPORT_SYMBOL(sram_alloc_with_lsl);
 /* Once we get a real allocator, we'll throw all of this away.
  * Until then, we need some sort of visibility into the L1 alloc.
  */
-static void _l1sram_proc_read(char *buf, int *len, const char *desc,
+/* Need to keep line of output the same.  Currently, that is 44 bytes
+ * (including newline).
+ */
+static int _l1sram_proc_read(char *buf, int *len, int count, const char *desc,
 		struct l1_sram_piece *pfree, const int array_size)
 {
 	int i;
 
-	*len += sprintf(&buf[*len], "--- L1 %-14s Size  PID State\n", desc);
-	for (i = 0; i < array_size; ++i) {
+	*len += sprintf(&buf[*len], "--- L1 %-14s Size   PID State     \n", desc);
+	for (i = 0; i < array_size && *len < count; ++i) {
 		const char *alloc_type;
 		switch (pfree[i].flag) {
 		case SRAM_SLT_NULL:      alloc_type = "NULL"; break;
@@ -563,31 +566,42 @@ static void _l1sram_proc_read(char *buf, int *len, const char *desc,
 		case SRAM_SLT_ALLOCATED: alloc_type = "ALLOCATED"; break;
 		default:                 alloc_type = "????"; break;
 		}
-		*len += sprintf(&buf[*len], "%p-%p %8i %4i %s\n",
+		/* if we've got a lot of space to cover, omit things */
+		if ((PAGE_SIZE - 1024) < (CONFIG_L1_MAX_PIECE + 1) * 4 * 44 &&
+		    pfree[i].size == 0)
+			continue;
+		*len += sprintf(&buf[*len], "%p-%p %8i %5i %-10s\n",
 			pfree[i].paddr, pfree[i].paddr + pfree[i].size,
 			pfree[i].size, pfree[i].pid, alloc_type);
 	}
+	return (i != array_size);
 }
 static int l1sram_proc_read(char *buf, char **start, off_t offset, int count,
 		int *eof, void *data)
 {
 	int len = 0;
 
-	_l1sram_proc_read(buf, &len, "Scratchpad",
-			l1_ssram, ARRAY_SIZE(l1_ssram));
+	if (_l1sram_proc_read(buf, &len, count, "Scratchpad",
+			l1_ssram, ARRAY_SIZE(l1_ssram)))
+		goto not_done;
 #if L1_DATA_A_LENGTH != 0
-	_l1sram_proc_read(buf, &len, "Data A",
-			l1_data_A_sram, ARRAY_SIZE(l1_data_A_sram));
+	if (_l1sram_proc_read(buf, &len, count, "Data A",
+			l1_data_A_sram, ARRAY_SIZE(l1_data_A_sram)))
+		goto not_done;
 #endif
 #if L1_DATA_B_LENGTH != 0
-	_l1sram_proc_read(buf, &len, "Data B",
-			l1_data_B_sram, ARRAY_SIZE(l1_data_B_sram));
+	if (_l1sram_proc_read(buf, &len, count, "Data B",
+			l1_data_B_sram, ARRAY_SIZE(l1_data_B_sram)))
+		goto not_done;
 #endif
 #if L1_CODE_LENGTH != 0
-	_l1sram_proc_read(buf, &len, "Instruction",
-			l1_inst_sram, ARRAY_SIZE(l1_inst_sram));
+	if (_l1sram_proc_read(buf, &len, count, "Instruction",
+			l1_inst_sram, ARRAY_SIZE(l1_inst_sram)))
+		goto not_done;
 #endif
 
+	*eof = 1;
+ not_done:
 	return len;
 }
 

From 68e2fc78e5055740126df8eab0d31005495756c9 Mon Sep 17 00:00:00 2001
From: Michael Hennerich <michael.hennerich@analog.com>
Date: Mon, 14 Jul 2008 16:51:57 +0800
Subject: [PATCH 0050/1171] Blackfin arch: Fix bug - Kernel does not boot if
 re-program clocks

Don't write conflicting data to EBIU_SDBCTL after the SDRAM is
configured. This can cause data corruption, since we might change SDRAM
row and column addressing modes.

Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/Kconfig                      | 14 -----------
 arch/blackfin/mach-bf527/head.S            | 12 +++-------
 arch/blackfin/mach-bf533/head.S            | 12 +++-------
 arch/blackfin/mach-bf537/head.S            | 12 +++-------
 arch/blackfin/mach-bf548/head.S            |  6 ++---
 arch/blackfin/mach-bf561/head.S            |  6 -----
 include/asm-blackfin/mach-bf527/mem_init.h | 27 ----------------------
 include/asm-blackfin/mach-bf533/mem_init.h | 27 ----------------------
 include/asm-blackfin/mach-bf537/mem_init.h | 27 ----------------------
 include/asm-blackfin/mach-bf561/mem_init.h | 27 ----------------------
 10 files changed, 12 insertions(+), 158 deletions(-)

diff --git a/arch/blackfin/Kconfig b/arch/blackfin/Kconfig
index b87634e75f20..d0f7ff3dc088 100644
--- a/arch/blackfin/Kconfig
+++ b/arch/blackfin/Kconfig
@@ -315,20 +315,6 @@ config MEM_SIZE
 	depends on BFIN_KERNEL_CLOCK
 	default 64
 
-config MEM_ADD_WIDTH
-	int "Memory Address Width"
-	depends on BFIN_KERNEL_CLOCK
-	depends on (!BF54x)
-	range 8 11
-	default  9 if BFIN533_EZKIT
-	default  9 if BFIN561_EZKIT
-	default  9 if H8606_HVSISTEMAS
-	default 10 if BFIN527_EZKIT
-	default 10 if BFIN537_STAMP
-	default 11 if BFIN533_STAMP
-	default 10 if PNAV10
-	default 10 if BFIN532_IP0X
-
 config PLL_BYPASS
 	bool "Bypass PLL"
 	depends on BFIN_KERNEL_CLOCK
diff --git a/arch/blackfin/mach-bf527/head.S b/arch/blackfin/mach-bf527/head.S
index 57bdb3ba2fed..fe05cc1ef174 100644
--- a/arch/blackfin/mach-bf527/head.S
+++ b/arch/blackfin/mach-bf527/head.S
@@ -32,7 +32,7 @@
 #include <asm/blackfin.h>
 #include <asm/trace.h>
 
-#if CONFIG_BFIN_KERNEL_CLOCK
+#ifdef CONFIG_BFIN_KERNEL_CLOCK
 #include <asm/mach-common/clocks.h>
 #include <asm/mach/mem_init.h>
 #endif
@@ -185,7 +185,7 @@ ENTRY(__start)
 
 	/* Put The Code for PLL Programming and SDRAM Programming in L1 ISRAM */
 	call _bf53x_relocate_l1_mem;
-#if CONFIG_BFIN_KERNEL_CLOCK
+#ifdef CONFIG_BFIN_KERNEL_CLOCK
 	call _start_dma_code;
 #endif
 
@@ -318,7 +318,7 @@ ENDPROC(_real_start)
 __FINIT
 
 .section .l1.text
-#if CONFIG_BFIN_KERNEL_CLOCK
+#ifdef CONFIG_BFIN_KERNEL_CLOCK
 ENTRY(_start_dma_code)
 
 	/* Enable PHY CLK buffer output */
@@ -398,12 +398,6 @@ ENTRY(_start_dma_code)
 	w[p0] = r0.l;
 	ssync;
 
-	p0.l = LO(EBIU_SDBCTL);
-	p0.h = HI(EBIU_SDBCTL);     /* SDRAM Memory Bank Control Register */
-	r0 = mem_SDBCTL;
-	w[p0] = r0.l;
-	ssync;
-
 	P2.H = hi(EBIU_SDGCTL);
 	P2.L = lo(EBIU_SDGCTL);
 	R0 = [P2];
diff --git a/arch/blackfin/mach-bf533/head.S b/arch/blackfin/mach-bf533/head.S
index 1295deac00a4..c671e8549b17 100644
--- a/arch/blackfin/mach-bf533/head.S
+++ b/arch/blackfin/mach-bf533/head.S
@@ -31,7 +31,7 @@
 #include <linux/init.h>
 #include <asm/blackfin.h>
 #include <asm/trace.h>
-#if CONFIG_BFIN_KERNEL_CLOCK
+#ifdef CONFIG_BFIN_KERNEL_CLOCK
 #include <asm/mach-common/clocks.h>
 #include <asm/mach/mem_init.h>
 #endif
@@ -186,7 +186,7 @@ ENTRY(__start)
 
 	/* Put The Code for PLL Programming and SDRAM Programming in L1 ISRAM */
 	call _bf53x_relocate_l1_mem;
-#if CONFIG_BFIN_KERNEL_CLOCK
+#ifdef CONFIG_BFIN_KERNEL_CLOCK
 	call _start_dma_code;
 #endif
 
@@ -319,7 +319,7 @@ ENDPROC(_real_start)
 __FINIT
 
 .section .l1.text
-#if CONFIG_BFIN_KERNEL_CLOCK
+#ifdef CONFIG_BFIN_KERNEL_CLOCK
 ENTRY(_start_dma_code)
 	p0.h = hi(SIC_IWR);
 	p0.l = lo(SIC_IWR);
@@ -390,12 +390,6 @@ ENTRY(_start_dma_code)
 	w[p0] = r0.l;
 	ssync;
 
-	p0.l = LO(EBIU_SDBCTL);
-	p0.h = HI(EBIU_SDBCTL);     /* SDRAM Memory Bank Control Register */
-	r0 = mem_SDBCTL;
-	w[p0] = r0.l;
-	ssync;
-
 	P2.H = hi(EBIU_SDGCTL);
 	P2.L = lo(EBIU_SDGCTL);
 	R0 = [P2];
diff --git a/arch/blackfin/mach-bf537/head.S b/arch/blackfin/mach-bf537/head.S
index 48cd58a410a0..6b019eaee0b6 100644
--- a/arch/blackfin/mach-bf537/head.S
+++ b/arch/blackfin/mach-bf537/head.S
@@ -32,7 +32,7 @@
 #include <asm/blackfin.h>
 #include <asm/trace.h>
 
-#if CONFIG_BFIN_KERNEL_CLOCK
+#ifdef CONFIG_BFIN_KERNEL_CLOCK
 #include <asm/mach-common/clocks.h>
 #include <asm/mach/mem_init.h>
 #endif
@@ -217,7 +217,7 @@ ENTRY(__start)
 
 	/* Put The Code for PLL Programming and SDRAM Programming in L1 ISRAM */
 	call _bf53x_relocate_l1_mem;
-#if CONFIG_BFIN_KERNEL_CLOCK
+#ifdef CONFIG_BFIN_KERNEL_CLOCK
 	call _start_dma_code;
 #endif
 
@@ -350,7 +350,7 @@ ENDPROC(_real_start)
 __FINIT
 
 .section .l1.text
-#if CONFIG_BFIN_KERNEL_CLOCK
+#ifdef CONFIG_BFIN_KERNEL_CLOCK
 ENTRY(_start_dma_code)
 
 	/* Enable PHY CLK buffer output */
@@ -430,12 +430,6 @@ ENTRY(_start_dma_code)
 	w[p0] = r0.l;
 	ssync;
 
-	p0.l = LO(EBIU_SDBCTL);
-	p0.h = HI(EBIU_SDBCTL);     /* SDRAM Memory Bank Control Register */
-	r0 = mem_SDBCTL;
-	w[p0] = r0.l;
-	ssync;
-
 	P2.H = hi(EBIU_SDGCTL);
 	P2.L = lo(EBIU_SDGCTL);
 	R0 = [P2];
diff --git a/arch/blackfin/mach-bf548/head.S b/arch/blackfin/mach-bf548/head.S
index f7191141a3ce..06b9178cfcfe 100644
--- a/arch/blackfin/mach-bf548/head.S
+++ b/arch/blackfin/mach-bf548/head.S
@@ -31,7 +31,7 @@
 #include <linux/init.h>
 #include <asm/blackfin.h>
 #include <asm/trace.h>
-#if CONFIG_BFIN_KERNEL_CLOCK
+#ifdef CONFIG_BFIN_KERNEL_CLOCK
 #include <asm/mach-common/clocks.h>
 #include <asm/mach/mem_init.h>
 #endif
@@ -130,7 +130,7 @@ ENTRY(__start)
 
 	/* Put The Code for PLL Programming and SDRAM Programming in L1 ISRAM */
 	call _bf53x_relocate_l1_mem;
-#if CONFIG_BFIN_KERNEL_CLOCK
+#ifdef CONFIG_BFIN_KERNEL_CLOCK
 	call _start_dma_code;
 #endif
 	/* Code for initializing Async memory banks */
@@ -288,7 +288,7 @@ ENDPROC(_real_start)
 __FINIT
 
 .section .l1.text
-#if CONFIG_BFIN_KERNEL_CLOCK
+#ifdef CONFIG_BFIN_KERNEL_CLOCK
 ENTRY(_start_dma_code)
 
 	/* Enable PHY CLK buffer output */
diff --git a/arch/blackfin/mach-bf561/head.S b/arch/blackfin/mach-bf561/head.S
index 5b8bd40851dd..cf1a2dff01e7 100644
--- a/arch/blackfin/mach-bf561/head.S
+++ b/arch/blackfin/mach-bf561/head.S
@@ -377,12 +377,6 @@ ENTRY(_start_dma_code)
 	w[p0] = r0.l;
 	ssync;
 
-	p0.l = LO(EBIU_SDBCTL);
-	p0.h = HI(EBIU_SDBCTL);     /* SDRAM Memory Bank Control Register */
-	r0 = mem_SDBCTL;
-	w[p0] = r0.l;
-	ssync;
-
 	P2.H = hi(EBIU_SDGCTL);
 	P2.L = lo(EBIU_SDGCTL);
 	R0 = [P2];
diff --git a/include/asm-blackfin/mach-bf527/mem_init.h b/include/asm-blackfin/mach-bf527/mem_init.h
index 008ca66719e2..cbe03f4a5698 100644
--- a/include/asm-blackfin/mach-bf527/mem_init.h
+++ b/include/asm-blackfin/mach-bf527/mem_init.h
@@ -146,33 +146,6 @@
 #define SDRAM_CL    CL_3
 #endif
 
-#if (CONFIG_MEM_SIZE == 128)
-#define SDRAM_SIZE      EBSZ_128
-#endif
-#if (CONFIG_MEM_SIZE == 64)
-#define SDRAM_SIZE      EBSZ_64
-#endif
-#if (CONFIG_MEM_SIZE == 32)
-#define SDRAM_SIZE      EBSZ_32
-#endif
-#if (CONFIG_MEM_SIZE == 16)
-#define SDRAM_SIZE      EBSZ_16
-#endif
-#if (CONFIG_MEM_ADD_WIDTH == 11)
-#define SDRAM_WIDTH     EBCAW_11
-#endif
-#if (CONFIG_MEM_ADD_WIDTH == 10)
-#define SDRAM_WIDTH     EBCAW_10
-#endif
-#if (CONFIG_MEM_ADD_WIDTH == 9)
-#define SDRAM_WIDTH     EBCAW_9
-#endif
-#if (CONFIG_MEM_ADD_WIDTH == 8)
-#define SDRAM_WIDTH     EBCAW_8
-#endif
-
-#define mem_SDBCTL      (SDRAM_WIDTH | SDRAM_SIZE | EBE)
-
 /* Equation from section 17 (p17-46) of BF533 HRM */
 #define mem_SDRRC       (((CONFIG_SCLK_HZ / 1000) * SDRAM_Tref) / SDRAM_NRA) - (SDRAM_tRAS_num + SDRAM_tRP_num)
 
diff --git a/include/asm-blackfin/mach-bf533/mem_init.h b/include/asm-blackfin/mach-bf533/mem_init.h
index f8f31901fca9..995c06b2b1ef 100644
--- a/include/asm-blackfin/mach-bf533/mem_init.h
+++ b/include/asm-blackfin/mach-bf533/mem_init.h
@@ -133,33 +133,6 @@
 #define SDRAM_CL    CL_3
 #endif
 
-#if (CONFIG_MEM_SIZE == 128)
-#define SDRAM_SIZE      EBSZ_128
-#endif
-#if (CONFIG_MEM_SIZE == 64)
-#define SDRAM_SIZE      EBSZ_64
-#endif
-#if (CONFIG_MEM_SIZE == 32)
-#define SDRAM_SIZE      EBSZ_32
-#endif
-#if (CONFIG_MEM_SIZE == 16)
-#define SDRAM_SIZE      EBSZ_16
-#endif
-#if (CONFIG_MEM_ADD_WIDTH == 11)
-#define SDRAM_WIDTH     EBCAW_11
-#endif
-#if (CONFIG_MEM_ADD_WIDTH == 10)
-#define SDRAM_WIDTH     EBCAW_10
-#endif
-#if (CONFIG_MEM_ADD_WIDTH == 9)
-#define SDRAM_WIDTH     EBCAW_9
-#endif
-#if (CONFIG_MEM_ADD_WIDTH == 8)
-#define SDRAM_WIDTH     EBCAW_8
-#endif
-
-#define mem_SDBCTL      (SDRAM_WIDTH | SDRAM_SIZE | EBE)
-
 /* Equation from section 17 (p17-46) of BF533 HRM */
 #define mem_SDRRC       (((CONFIG_SCLK_HZ / 1000) * SDRAM_Tref)  / SDRAM_NRA) - (SDRAM_tRAS_num + SDRAM_tRP_num)
 
diff --git a/include/asm-blackfin/mach-bf537/mem_init.h b/include/asm-blackfin/mach-bf537/mem_init.h
index 9ad979d416c6..f67698f670ca 100644
--- a/include/asm-blackfin/mach-bf537/mem_init.h
+++ b/include/asm-blackfin/mach-bf537/mem_init.h
@@ -139,33 +139,6 @@
 #define SDRAM_CL    CL_3
 #endif
 
-#if (CONFIG_MEM_SIZE == 128)
-#define SDRAM_SIZE      EBSZ_128
-#endif
-#if (CONFIG_MEM_SIZE == 64)
-#define SDRAM_SIZE      EBSZ_64
-#endif
-#if (CONFIG_MEM_SIZE == 32)
-#define SDRAM_SIZE      EBSZ_32
-#endif
-#if (CONFIG_MEM_SIZE == 16)
-#define SDRAM_SIZE      EBSZ_16
-#endif
-#if (CONFIG_MEM_ADD_WIDTH == 11)
-#define SDRAM_WIDTH     EBCAW_11
-#endif
-#if (CONFIG_MEM_ADD_WIDTH == 10)
-#define SDRAM_WIDTH     EBCAW_10
-#endif
-#if (CONFIG_MEM_ADD_WIDTH == 9)
-#define SDRAM_WIDTH     EBCAW_9
-#endif
-#if (CONFIG_MEM_ADD_WIDTH == 8)
-#define SDRAM_WIDTH     EBCAW_8
-#endif
-
-#define mem_SDBCTL      (SDRAM_WIDTH | SDRAM_SIZE | EBE)
-
 /* Equation from section 17 (p17-46) of BF533 HRM */
 #define mem_SDRRC       (((CONFIG_SCLK_HZ / 1000) * SDRAM_Tref) / SDRAM_NRA) - (SDRAM_tRAS_num + SDRAM_tRP_num)
 
diff --git a/include/asm-blackfin/mach-bf561/mem_init.h b/include/asm-blackfin/mach-bf561/mem_init.h
index 439a5895b346..e163260bca18 100644
--- a/include/asm-blackfin/mach-bf561/mem_init.h
+++ b/include/asm-blackfin/mach-bf561/mem_init.h
@@ -131,33 +131,6 @@
 #define SDRAM_CL    CL_3
 #endif
 
-#if (CONFIG_MEM_SIZE == 128)
-#define SDRAM_SIZE      EB0_SZ_128
-#endif
-#if (CONFIG_MEM_SIZE == 64)
-#define SDRAM_SIZE      EB0_SZ_64
-#endif
-#if ( CONFIG_MEM_SIZE == 32)
-#define SDRAM_SIZE      EB0_SZ_32
-#endif
-#if (CONFIG_MEM_SIZE == 16)
-#define SDRAM_SIZE      EB0_SZ_16
-#endif
-#if (CONFIG_MEM_ADD_WIDTH == 11)
-#define SDRAM_WIDTH     EB0_CAW_11
-#endif
-#if (CONFIG_MEM_ADD_WIDTH == 10)
-#define SDRAM_WIDTH     EB0_CAW_10
-#endif
-#if (CONFIG_MEM_ADD_WIDTH == 9)
-#define SDRAM_WIDTH     EB0_CAW_9
-#endif
-#if (CONFIG_MEM_ADD_WIDTH == 8)
-#define SDRAM_WIDTH     EB0_CAW_8
-#endif
-
-#define mem_SDBCTL      (SDRAM_WIDTH | SDRAM_SIZE | EB0_E)
-
 /* Equation from section 17 (p17-46) of BF533 HRM */
 #define mem_SDRRC       (((CONFIG_SCLK_HZ / 1000) * SDRAM_Tref) / SDRAM_NRA) - (SDRAM_tRAS_num + SDRAM_tRP_num)
 

From 96b86e94bcf7eadbd8c959c0474d2971b3695478 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier.adi@gmail.com>
Date: Mon, 14 Jul 2008 16:39:43 +0800
Subject: [PATCH 0051/1171] Blackfin arch: use the symbolic IRQ define rather
 than the backend number

Signed-off-by: Mike Frysinger <vapier.adi@gmail.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/mach-bf537/boards/stamp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/blackfin/mach-bf537/boards/stamp.c b/arch/blackfin/mach-bf537/boards/stamp.c
index 7e4b9d95674b..6dbc76fb080b 100644
--- a/arch/blackfin/mach-bf537/boards/stamp.c
+++ b/arch/blackfin/mach-bf537/boards/stamp.c
@@ -861,7 +861,7 @@ static struct platform_device bfin_sport1_uart_device = {
 #endif
 
 #if defined(CONFIG_PATA_PLATFORM) || defined(CONFIG_PATA_PLATFORM_MODULE)
-#define PATA_INT	55
+#define PATA_INT	IRQ_PF5
 
 static struct pata_platform_info bfin_pata_platform_data = {
 	.ioport_shift = 1,

From 99d95bbd48f43dafdcd0540eb0da26c5655d7f33 Mon Sep 17 00:00:00 2001
From: Michael Hennerich <michael.hennerich@analog.com>
Date: Mon, 14 Jul 2008 17:04:14 +0800
Subject: [PATCH 0052/1171] Blackfin arch: Remove redundant kernel option

use kernel command line mem and max_mem bootargs to limit
availabe memory instead.

Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/Kconfig                      |  7 +------
 arch/blackfin/kernel/cplb-nompu/cplbinit.c |  6 +-----
 arch/blackfin/kernel/setup.c               | 14 +++++---------
 arch/blackfin/mach-common/arch_checks.c    |  6 ------
 4 files changed, 7 insertions(+), 26 deletions(-)

diff --git a/arch/blackfin/Kconfig b/arch/blackfin/Kconfig
index d0f7ff3dc088..c602727d1a9a 100644
--- a/arch/blackfin/Kconfig
+++ b/arch/blackfin/Kconfig
@@ -310,11 +310,6 @@ config BFIN_KERNEL_CLOCK
 	  are also not changed, and the Bootloader does 100% of the hardware
 	  configuration.
 
-config MEM_SIZE
-	int "SDRAM Memory Size in MBytes"
-	depends on BFIN_KERNEL_CLOCK
-	default 64
-
 config PLL_BYPASS
 	bool "Bypass PLL"
 	depends on BFIN_KERNEL_CLOCK
@@ -376,7 +371,7 @@ config SCLK_DIV
 
 config MAX_MEM_SIZE
 	int "Max SDRAM Memory Size in MBytes"
-	depends on !BFIN_KERNEL_CLOCK && !MPU
+	depends on !MPU
 	default 512
 	help
 	  This is the max memory size that the kernel will create CPLB
diff --git a/arch/blackfin/kernel/cplb-nompu/cplbinit.c b/arch/blackfin/kernel/cplb-nompu/cplbinit.c
index 6be0c50122e8..224e7cc30bc5 100644
--- a/arch/blackfin/kernel/cplb-nompu/cplbinit.c
+++ b/arch/blackfin/kernel/cplb-nompu/cplbinit.c
@@ -26,11 +26,7 @@
 #include <asm/cplb.h>
 #include <asm/cplbinit.h>
 
-#ifdef CONFIG_MAX_MEM_SIZE
-# define CPLB_MEM CONFIG_MAX_MEM_SIZE
-#else
-# define CPLB_MEM CONFIG_MEM_SIZE
-#endif
+#define CPLB_MEM CONFIG_MAX_MEM_SIZE
 
 /*
 * Number of required data CPLB switchtable entries
diff --git a/arch/blackfin/kernel/setup.c b/arch/blackfin/kernel/setup.c
index 8efea004aecb..861a1db74df8 100644
--- a/arch/blackfin/kernel/setup.c
+++ b/arch/blackfin/kernel/setup.c
@@ -664,11 +664,8 @@ static __init void setup_bootmem_allocator(void)
 })
 static inline int __init get_mem_size(void)
 {
-#ifdef CONFIG_MEM_SIZE
-	return CONFIG_MEM_SIZE;
-#else
-# if defined(EBIU_SDBCTL)
-#  if defined(BF561_FAMILY)
+#if defined(EBIU_SDBCTL)
+# if defined(BF561_FAMILY)
 	int ret = 0;
 	u32 sdbctl = bfin_read_EBIU_SDBCTL();
 	ret += EBSZ_TO_MEG(sdbctl >>  0);
@@ -676,10 +673,10 @@ static inline int __init get_mem_size(void)
 	ret += EBSZ_TO_MEG(sdbctl >> 16);
 	ret += EBSZ_TO_MEG(sdbctl >> 24);
 	return ret;
-#  else
+# else
 	return EBSZ_TO_MEG(bfin_read_EBIU_SDBCTL());
-#  endif
-# elif defined(EBIU_DDRCTL1)
+# endif
+#elif defined(EBIU_DDRCTL1)
 	u32 ddrctl = bfin_read_EBIU_DDRCTL1();
 	int ret = 0;
 	switch (ddrctl & 0xc0000) {
@@ -694,7 +691,6 @@ static inline int __init get_mem_size(void)
 		case DEVWD_16: break;
 	}
 	return ret;
-# endif
 #endif
 	BUG();
 }
diff --git a/arch/blackfin/mach-common/arch_checks.c b/arch/blackfin/mach-common/arch_checks.c
index caaab49e9cfa..f9160d83b91f 100644
--- a/arch/blackfin/mach-common/arch_checks.c
+++ b/arch/blackfin/mach-common/arch_checks.c
@@ -53,9 +53,3 @@
 # endif
 
 #endif /* CONFIG_BFIN_KERNEL_CLOCK */
-
-#ifdef CONFIG_MEM_SIZE
-#if (CONFIG_MEM_SIZE % 4)
-#error "SDRAM mem size must be multible of 4MB"
-#endif
-#endif

From ed987319ed6a3f5e14c87141e210deb057a6c507 Mon Sep 17 00:00:00 2001
From: Sonic Zhang <sonic.zhang@analog.com>
Date: Sat, 26 Jul 2008 18:40:06 +0800
Subject: [PATCH 0053/1171] Blackfin arch: update board defconfigs

Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/configs/BF527-EZKIT_defconfig | 187 ++++++++++-
 arch/blackfin/configs/BF533-EZKIT_defconfig | 279 +++++-----------
 arch/blackfin/configs/BF533-STAMP_defconfig | 342 ++++++++------------
 arch/blackfin/configs/BF537-STAMP_defconfig | 337 ++++++++-----------
 arch/blackfin/configs/BF548-EZKIT_defconfig |  65 +++-
 arch/blackfin/configs/BF561-EZKIT_defconfig | 273 +++++-----------
 arch/blackfin/configs/CM-BF533_defconfig    |  14 +-
 arch/blackfin/configs/CM-BF537E_defconfig   |  21 +-
 arch/blackfin/configs/CM-BF537U_defconfig   |  18 +-
 arch/blackfin/configs/CM-BF548_defconfig    |  10 +-
 arch/blackfin/configs/CM-BF561_defconfig    |  14 +-
 arch/blackfin/configs/H8606_defconfig       |   2 +-
 arch/blackfin/configs/IP0X_defconfig        |   2 +-
 arch/blackfin/configs/PNAV-10_defconfig     |   4 +-
 arch/blackfin/configs/SRV1_defconfig        |   2 +-
 15 files changed, 707 insertions(+), 863 deletions(-)

diff --git a/arch/blackfin/configs/BF527-EZKIT_defconfig b/arch/blackfin/configs/BF527-EZKIT_defconfig
index 5e6fb9d8e50f..66854a83c0de 100644
--- a/arch/blackfin/configs/BF527-EZKIT_defconfig
+++ b/arch/blackfin/configs/BF527-EZKIT_defconfig
@@ -1,7 +1,6 @@
 #
 # Automatically generated make config: don't edit
 # Linux kernel version: 2.6.24.7
-# Fri May 16 10:02:29 2008
 #
 # CONFIG_MMU is not set
 # CONFIG_FPU is not set
@@ -290,7 +289,7 @@ CONFIG_SPLIT_PTLOCK_CPUS=4
 # CONFIG_RESOURCES_64BIT is not set
 CONFIG_ZONE_DMA_FLAG=1
 CONFIG_VIRT_TO_BUS=y
-# CONFIG_BFIN_GPTIMERS is not set
+CONFIG_BFIN_GPTIMERS=y
 CONFIG_BFIN_DMA_5XX=y
 # CONFIG_DMA_UNCACHED_4M is not set
 # CONFIG_DMA_UNCACHED_2M is not set
@@ -430,7 +429,58 @@ CONFIG_DEFAULT_TCP_CONG="cubic"
 #
 # CONFIG_NET_PKTGEN is not set
 # CONFIG_HAMRADIO is not set
-# CONFIG_IRDA is not set
+CONFIG_IRDA=m
+
+#
+# IrDA protocols
+#
+CONFIG_IRLAN=m
+CONFIG_IRCOMM=m
+# CONFIG_IRDA_ULTRA is not set
+
+#
+# IrDA options
+#
+# CONFIG_IRDA_CACHE_LAST_LSAP is not set
+# CONFIG_IRDA_FAST_RR is not set
+# CONFIG_IRDA_DEBUG is not set
+
+#
+# Infrared-port device drivers
+#
+
+#
+# SIR device drivers
+#
+CONFIG_IRTTY_SIR=m
+CONFIG_BFIN_SIR=m
+CONFIG_BFIN_SIR0=y
+CONFIG_SIR_BFIN_DMA=y
+# CONFIG_SIR_BFIN_PIO is not set
+
+#
+# Dongle support
+#
+# CONFIG_DONGLE is not set
+# CONFIG_KINGSUN_DONGLE is not set
+# CONFIG_KSDAZZLE_DONGLE is not set
+# CONFIG_KS959_DONGLE is not set
+
+#
+# Old SIR device drivers
+#
+# CONFIG_IRPORT_SIR is not set
+
+#
+# Old Serial dongle support
+#
+
+#
+# FIR device drivers
+#
+# CONFIG_USB_IRDA is not set
+# CONFIG_SIGMATEL_FIR is not set
+# CONFIG_MCS_FIR is not set
 # CONFIG_BT is not set
 # CONFIG_AF_RXRPC is not set
 
@@ -689,8 +739,11 @@ CONFIG_BFIN_OTP=y
 # CONFIG_BFIN_SPORT is not set
 # CONFIG_BFIN_TIMER_LATENCY is not set
 # CONFIG_TWI_LCD is not set
-# CONFIG_SIMPLE_GPIO is not set
-# CONFIG_VT is not set
+CONFIG_SIMPLE_GPIO=m
+CONFIG_VT=y
+CONFIG_VT_CONSOLE=y
+CONFIG_HW_CONSOLE=y
+# CONFIG_VT_HW_CONSOLE_BINDING is not set
 # CONFIG_SERIAL_NONSTANDARD is not set
 
 #
@@ -872,18 +925,136 @@ CONFIG_SSB_POSSIBLE=y
 #
 # CONFIG_VGASTATE is not set
 # CONFIG_VIDEO_OUTPUT_CONTROL is not set
-# CONFIG_FB is not set
-# CONFIG_BACKLIGHT_LCD_SUPPORT is not set
+CONFIG_FB=y
+# CONFIG_FIRMWARE_EDID is not set
+# CONFIG_FB_DDC is not set
+CONFIG_FB_CFB_FILLRECT=y
+CONFIG_FB_CFB_COPYAREA=y
+CONFIG_FB_CFB_IMAGEBLIT=y
+# CONFIG_FB_CFB_REV_PIXELS_IN_BYTE is not set
+# CONFIG_FB_SYS_FILLRECT is not set
+# CONFIG_FB_SYS_COPYAREA is not set
+# CONFIG_FB_SYS_IMAGEBLIT is not set
+# CONFIG_FB_SYS_FOPS is not set
+CONFIG_FB_DEFERRED_IO=y
+# CONFIG_FB_SVGALIB is not set
+# CONFIG_FB_MACMODES is not set
+# CONFIG_FB_BACKLIGHT is not set
+# CONFIG_FB_MODE_HELPERS is not set
+# CONFIG_FB_TILEBLITTING is not set
+
+#
+# Frame buffer hardware drivers
+#
+CONFIG_FB_BFIN_T350MCQB=y
+# CONFIG_FB_BFIN_7393 is not set
+# CONFIG_FB_S1D13XXX is not set
+# CONFIG_FB_VIRTUAL is not set
+CONFIG_BACKLIGHT_LCD_SUPPORT=y
+CONFIG_LCD_CLASS_DEVICE=m
+CONFIG_LCD_LTV350QV=m
+CONFIG_BACKLIGHT_CLASS_DEVICE=m
+# CONFIG_BACKLIGHT_CORGI is not set
 
 #
 # Display device support
 #
 # CONFIG_DISPLAY_SUPPORT is not set
 
+#
+# Console display driver support
+#
+CONFIG_DUMMY_CONSOLE=y
+CONFIG_FRAMEBUFFER_CONSOLE=y
+# CONFIG_FRAMEBUFFER_CONSOLE_DETECT_PRIMARY is not set
+# CONFIG_FRAMEBUFFER_CONSOLE_ROTATION is not set
+# CONFIG_FONTS is not set
+CONFIG_FONT_8x8=y
+CONFIG_FONT_8x16=y
+CONFIG_LOGO=y
+# CONFIG_LOGO_LINUX_MONO is not set
+# CONFIG_LOGO_LINUX_VGA16 is not set
+# CONFIG_LOGO_LINUX_CLUT224 is not set
+# CONFIG_LOGO_BLACKFIN_VGA16 is not set
+CONFIG_LOGO_BLACKFIN_CLUT224=y
+
 #
 # Sound
 #
-# CONFIG_SOUND is not set
+CONFIG_SOUND=m
+
+#
+# Advanced Linux Sound Architecture
+#
+CONFIG_SND=m
+CONFIG_SND_TIMER=m
+CONFIG_SND_PCM=m
+# CONFIG_SND_SEQUENCER is not set
+# CONFIG_SND_MIXER_OSS is not set
+# CONFIG_SND_PCM_OSS is not set
+# CONFIG_SND_DYNAMIC_MINORS is not set
+CONFIG_SND_SUPPORT_OLD_API=y
+CONFIG_SND_VERBOSE_PROCFS=y
+# CONFIG_SND_VERBOSE_PRINTK is not set
+# CONFIG_SND_DEBUG is not set
+
+#
+# Generic devices
+#
+# CONFIG_SND_DUMMY is not set
+# CONFIG_SND_MTPAV is not set
+# CONFIG_SND_SERIAL_U16550 is not set
+# CONFIG_SND_MPU401 is not set
+
+#
+# SPI devices
+#
+
+#
+# ALSA Blackfin devices
+#
+# CONFIG_SND_BLACKFIN_AD1836 is not set
+# CONFIG_SND_BLACKFIN_AD1836_TDM is not set
+# CONFIG_SND_BLACKFIN_AD1836_I2S is not set
+# CONFIG_SND_BLACKFIN_AD1836_MULSUB is not set
+# CONFIG_SND_BLACKFIN_AD1836_5P1 is not set
+# CONFIG_SND_BFIN_AD73311 is not set
+# CONFIG_SND_BFIN_AD73322 is not set
+
+#
+# USB devices
+#
+# CONFIG_SND_USB_AUDIO is not set
+# CONFIG_SND_USB_CAIAQ is not set
+
+#
+# System on Chip audio support
+#
+CONFIG_SND_SOC_AC97_BUS=y
+CONFIG_SND_SOC=m
+CONFIG_SND_BF5XX_SOC=m
+CONFIG_SND_MMAP_SUPPORT=y
+CONFIG_SND_BF5XX_SOC_I2S=m
+CONFIG_SND_BF5XX_SOC_AC97=m
+# CONFIG_SND_BF5XX_SOC_WM8750 is not set
+# CONFIG_SND_BF5XX_SOC_WM8731 is not set
+CONFIG_SND_BF5XX_SOC_SSM2602=m
+CONFIG_SND_BF5XX_SOC_BF5xx=m
+CONFIG_SND_BF5XX_SPORT_NUM=0
+# CONFIG_SND_BF5XX_HAVE_COLD_RESET is not set
+
+#
+# SoC Audio support for SuperH
+#
+CONFIG_SND_SOC_SSM2602=m
+# CONFIG_SND_SOC_SSM2602_SPI is not set
+CONFIG_SND_SOC_AD1980=m
+
+#
+# Open Sound System
+#
+# CONFIG_SOUND_PRIME is not set
+CONFIG_AC97_BUS=m
 CONFIG_HID_SUPPORT=y
 CONFIG_HID=y
 # CONFIG_HID_DEBUG is not set
diff --git a/arch/blackfin/configs/BF533-EZKIT_defconfig b/arch/blackfin/configs/BF533-EZKIT_defconfig
index 8d817ba01945..6bc11db12690 100644
--- a/arch/blackfin/configs/BF533-EZKIT_defconfig
+++ b/arch/blackfin/configs/BF533-EZKIT_defconfig
@@ -1,6 +1,6 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.22.16
+# Linux kernel version: 2.6.24.7
 #
 # CONFIG_MMU is not set
 # CONFIG_FPU is not set
@@ -13,35 +13,34 @@ CONFIG_GENERIC_FIND_NEXT_BIT=y
 CONFIG_GENERIC_HWEIGHT=y
 CONFIG_GENERIC_HARDIRQS=y
 CONFIG_GENERIC_IRQ_PROBE=y
-CONFIG_GENERIC_TIME=y
 CONFIG_GENERIC_GPIO=y
 CONFIG_FORCE_MAX_ZONEORDER=14
 CONFIG_GENERIC_CALIBRATE_DELAY=y
 CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
 
 #
-# Code maturity level options
+# General setup
 #
 CONFIG_EXPERIMENTAL=y
 CONFIG_BROKEN_ON_SMP=y
 CONFIG_INIT_ENV_ARG_LIMIT=32
-
-#
-# General setup
-#
 CONFIG_LOCALVERSION=""
 CONFIG_LOCALVERSION_AUTO=y
 CONFIG_SYSVIPC=y
-# CONFIG_IPC_NS is not set
 CONFIG_SYSVIPC_SYSCTL=y
 # CONFIG_POSIX_MQUEUE is not set
 # CONFIG_BSD_PROCESS_ACCT is not set
 # CONFIG_TASKSTATS is not set
-# CONFIG_UTS_NS is not set
+# CONFIG_USER_NS is not set
+# CONFIG_PID_NS is not set
 # CONFIG_AUDIT is not set
 CONFIG_IKCONFIG=y
 CONFIG_IKCONFIG_PROC=y
 CONFIG_LOG_BUF_SHIFT=14
+# CONFIG_CGROUPS is not set
+CONFIG_FAIR_GROUP_SCHED=y
+CONFIG_FAIR_USER_SCHED=y
+# CONFIG_FAIR_CGROUP_SCHED is not set
 CONFIG_SYSFS_DEPRECATED=y
 # CONFIG_RELAY is not set
 CONFIG_BLK_DEV_INITRD=y
@@ -64,32 +63,24 @@ CONFIG_EPOLL=y
 CONFIG_SIGNALFD=y
 CONFIG_EVENTFD=y
 CONFIG_VM_EVENT_COUNTERS=y
-CONFIG_BIG_ORDER_ALLOC_NOFAIL_MAGIC=3
-# CONFIG_NP2 is not set
 CONFIG_SLAB=y
 # CONFIG_SLUB is not set
 # CONFIG_SLOB is not set
+CONFIG_SLABINFO=y
 CONFIG_RT_MUTEXES=y
 CONFIG_TINY_SHMEM=y
 CONFIG_BASE_SMALL=0
-
-#
-# Loadable module support
-#
 CONFIG_MODULES=y
 CONFIG_MODULE_UNLOAD=y
 # CONFIG_MODULE_FORCE_UNLOAD is not set
 # CONFIG_MODVERSIONS is not set
 # CONFIG_MODULE_SRCVERSION_ALL is not set
 CONFIG_KMOD=y
-
-#
-# Block layer
-#
 CONFIG_BLOCK=y
 # CONFIG_LBD is not set
 # CONFIG_BLK_DEV_IO_TRACE is not set
 # CONFIG_LSF is not set
+# CONFIG_BLK_DEV_BSG is not set
 
 #
 # IO Schedulers
@@ -141,12 +132,12 @@ CONFIG_BF_REV_0_3=y
 # CONFIG_BF_REV_ANY is not set
 # CONFIG_BF_REV_NONE is not set
 CONFIG_BF53x=y
-CONFIG_BFIN_SINGLE_CORE=y
 CONFIG_MEM_MT48LC16M16A2TG_75=y
 CONFIG_BFIN533_EZKIT=y
 # CONFIG_BFIN533_STAMP is not set
 # CONFIG_BFIN533_BLUETECHNIX_CM is not set
 # CONFIG_H8606_HVSISTEMAS is not set
+# CONFIG_BFIN532_IP0X is not set
 # CONFIG_GENERIC_BF533_BOARD is not set
 
 #
@@ -189,12 +180,14 @@ CONFIG_WDTIMER=13
 # Board customizations
 #
 # CONFIG_CMDLINE_BOOL is not set
+CONFIG_BOOT_LOAD=0x1000
 
 #
 # Clock/PLL Setup
 #
 CONFIG_CLKIN_HZ=27000000
 # CONFIG_BFIN_KERNEL_CLOCK is not set
+CONFIG_MAX_MEM_SIZE=512
 CONFIG_MAX_VCO_HZ=750000000
 CONFIG_MIN_VCO_HZ=50000000
 CONFIG_MAX_SCLK_HZ=133333333
@@ -208,13 +201,17 @@ CONFIG_HZ_250=y
 # CONFIG_HZ_300 is not set
 # CONFIG_HZ_1000 is not set
 CONFIG_HZ=250
+CONFIG_GENERIC_TIME=y
+CONFIG_GENERIC_CLOCKEVENTS=y
+# CONFIG_CYCLES_CLOCKSOURCE is not set
+CONFIG_TICK_ONESHOT=y
+# CONFIG_NO_HZ is not set
+CONFIG_HIGH_RES_TIMERS=y
+CONFIG_GENERIC_CLOCKEVENTS_BUILD=y
 
 #
-# Memory Setup
+# Misc
 #
-CONFIG_MAX_MEM_SIZE=512
-CONFIG_MEM_ADD_WIDTH=9
-CONFIG_BOOT_LOAD=0x1000
 CONFIG_BFIN_SCRATCH_REG_RETN=y
 # CONFIG_BFIN_SCRATCH_REG_RETE is not set
 # CONFIG_BFIN_SCRATCH_REG_CYCLES is not set
@@ -250,12 +247,14 @@ CONFIG_FLATMEM_MANUAL=y
 CONFIG_FLATMEM=y
 CONFIG_FLAT_NODE_MEM_MAP=y
 # CONFIG_SPARSEMEM_STATIC is not set
+# CONFIG_SPARSEMEM_VMEMMAP_ENABLE is not set
 CONFIG_SPLIT_PTLOCK_CPUS=4
 # CONFIG_RESOURCES_64BIT is not set
 CONFIG_ZONE_DMA_FLAG=1
-CONFIG_LARGE_ALLOCS=y
+CONFIG_VIRT_TO_BUS=y
 # CONFIG_BFIN_GPTIMERS is not set
 CONFIG_BFIN_DMA_5XX=y
+# CONFIG_DMA_UNCACHED_4M is not set
 # CONFIG_DMA_UNCACHED_2M is not set
 CONFIG_DMA_UNCACHED_1M=y
 # CONFIG_DMA_UNCACHED_NONE is not set
@@ -293,17 +292,13 @@ CONFIG_C_AMBEN_ALL=y
 CONFIG_BANK_0=0x7BB0
 CONFIG_BANK_1=0x7BB0
 CONFIG_BANK_2=0x7BB0
-CONFIG_BANK_3=0xAAC3
+CONFIG_BANK_3=0xAAC2
 
 #
 # Bus options (PCI, PCMCIA, EISA, MCA, ISA)
 #
 # CONFIG_PCI is not set
 # CONFIG_ARCH_SUPPORTS_MSI is not set
-
-#
-# PCCARD (PCMCIA/CardBus) support
-#
 # CONFIG_PCCARD is not set
 
 #
@@ -321,7 +316,9 @@ CONFIG_BINFMT_ZFLAT=y
 CONFIG_PM=y
 # CONFIG_PM_LEGACY is not set
 # CONFIG_PM_DEBUG is not set
-# CONFIG_PM_SYSFS_DEPRECATED is not set
+CONFIG_PM_SLEEP=y
+CONFIG_SUSPEND_UP_POSSIBLE=y
+CONFIG_SUSPEND=y
 CONFIG_PM_BFIN_SLEEP_DEEPER=y
 # CONFIG_PM_BFIN_SLEEP is not set
 # CONFIG_PM_WAKEUP_BY_GPIO is not set
@@ -367,6 +364,7 @@ CONFIG_SYN_COOKIES=y
 CONFIG_INET_XFRM_MODE_TRANSPORT=y
 CONFIG_INET_XFRM_MODE_TUNNEL=y
 CONFIG_INET_XFRM_MODE_BEET=y
+# CONFIG_INET_LRO is not set
 CONFIG_INET_DIAG=y
 CONFIG_INET_TCP_DIAG=y
 # CONFIG_TCP_CONG_ADVANCED is not set
@@ -393,10 +391,6 @@ CONFIG_DEFAULT_TCP_CONG="cubic"
 # CONFIG_LAPB is not set
 # CONFIG_ECONET is not set
 # CONFIG_WAN_ROUTER is not set
-
-#
-# QoS and/or fair queueing
-#
 # CONFIG_NET_SCHED is not set
 
 #
@@ -428,6 +422,7 @@ CONFIG_IRDA_CACHE_LAST_LSAP=y
 # SIR device drivers
 #
 CONFIG_IRTTY_SIR=m
+# CONFIG_BFIN_SIR is not set
 
 #
 # Dongle support
@@ -457,6 +452,7 @@ CONFIG_IRTTY_SIR=m
 # CONFIG_MAC80211 is not set
 # CONFIG_IEEE80211 is not set
 # CONFIG_RFKILL is not set
+# CONFIG_NET_9P is not set
 
 #
 # Device Drivers
@@ -465,14 +461,11 @@ CONFIG_IRTTY_SIR=m
 #
 # Generic Driver Options
 #
+CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
 CONFIG_STANDALONE=y
 CONFIG_PREVENT_FIRMWARE_BUILD=y
 # CONFIG_FW_LOADER is not set
 # CONFIG_SYS_HYPERVISOR is not set
-
-#
-# Connector - unified userspace <-> kernelspace linker
-#
 # CONFIG_CONNECTOR is not set
 CONFIG_MTD=y
 # CONFIG_MTD_DEBUG is not set
@@ -492,6 +485,7 @@ CONFIG_MTD_BLOCK=y
 # CONFIG_INFTL is not set
 # CONFIG_RFD_FTL is not set
 # CONFIG_SSFDC is not set
+# CONFIG_MTD_OOPS is not set
 
 #
 # RAM/ROM/Flash chip drivers
@@ -548,20 +542,8 @@ CONFIG_MTD_COMPLEX_MAPPINGS=y
 # UBI - Unsorted block images
 #
 # CONFIG_MTD_UBI is not set
-
-#
-# Parallel port support
-#
 # CONFIG_PARPORT is not set
-
-#
-# Plug and Play support
-#
-# CONFIG_PNPACPI is not set
-
-#
-# Block devices
-#
+CONFIG_BLK_DEV=y
 # CONFIG_BLK_DEV_COW_COMMON is not set
 # CONFIG_BLK_DEV_LOOP is not set
 # CONFIG_BLK_DEV_NBD is not set
@@ -571,10 +553,8 @@ CONFIG_BLK_DEV_RAM_SIZE=4096
 CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024
 # CONFIG_CDROM_PKTCDVD is not set
 # CONFIG_ATA_OVER_ETH is not set
-
-#
-# Misc devices
-#
+CONFIG_MISC_DEVICES=y
+# CONFIG_EEPROM_93CX6 is not set
 # CONFIG_IDE is not set
 
 #
@@ -582,32 +562,29 @@ CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024
 #
 # CONFIG_RAID_ATTRS is not set
 # CONFIG_SCSI is not set
+# CONFIG_SCSI_DMA is not set
 # CONFIG_SCSI_NETLINK is not set
 # CONFIG_ATA is not set
-
-#
-# Multi-device support (RAID and LVM)
-#
 # CONFIG_MD is not set
-
-#
-# Network device support
-#
 CONFIG_NETDEVICES=y
+# CONFIG_NETDEVICES_MULTIQUEUE is not set
 # CONFIG_DUMMY is not set
 # CONFIG_BONDING is not set
+# CONFIG_MACVLAN is not set
 # CONFIG_EQUALIZER is not set
 # CONFIG_TUN is not set
+# CONFIG_VETH is not set
 # CONFIG_PHYLIB is not set
-
-#
-# Ethernet (10 or 100Mbit)
-#
 CONFIG_NET_ETHERNET=y
 CONFIG_MII=y
 CONFIG_SMC91X=y
 # CONFIG_SMSC911X is not set
 # CONFIG_DM9000 is not set
+# CONFIG_IBM_NEW_EMAC_ZMII is not set
+# CONFIG_IBM_NEW_EMAC_RGMII is not set
+# CONFIG_IBM_NEW_EMAC_TAH is not set
+# CONFIG_IBM_NEW_EMAC_EMAC4 is not set
+# CONFIG_B44 is not set
 CONFIG_NETDEV_1000=y
 # CONFIG_AX88180 is not set
 CONFIG_NETDEV_10000=y
@@ -624,15 +601,7 @@ CONFIG_NETDEV_10000=y
 # CONFIG_NETCONSOLE is not set
 # CONFIG_NETPOLL is not set
 # CONFIG_NET_POLL_CONTROLLER is not set
-
-#
-# ISDN subsystem
-#
 # CONFIG_ISDN is not set
-
-#
-# Telephony Support
-#
 # CONFIG_PHONE is not set
 
 #
@@ -647,7 +616,6 @@ CONFIG_INPUT=m
 #
 # CONFIG_INPUT_MOUSEDEV is not set
 # CONFIG_INPUT_JOYDEV is not set
-# CONFIG_INPUT_TSDEV is not set
 CONFIG_INPUT_EVDEV=m
 # CONFIG_INPUT_EVBUG is not set
 
@@ -672,13 +640,12 @@ CONFIG_INPUT_EVDEV=m
 #
 # CONFIG_AD9960 is not set
 # CONFIG_SPI_ADC_BF533 is not set
-# CONFIG_BF5xx_PFLAGS is not set
 # CONFIG_BF5xx_PPIFCD is not set
 # CONFIG_BFIN_SIMPLE_TIMER is not set
 # CONFIG_BF5xx_PPI is not set
 CONFIG_BFIN_SPORT=y
 # CONFIG_BFIN_TIMER_LATENCY is not set
-# CONFIG_AD5304 is not set
+CONFIG_SIMPLE_GPIO=m
 # CONFIG_VT is not set
 # CONFIG_SERIAL_NONSTANDARD is not set
 
@@ -706,28 +673,11 @@ CONFIG_UNIX98_PTYS=y
 # CAN, the car bus and industrial fieldbus
 #
 # CONFIG_CAN4LINUX is not set
-
-#
-# IPMI
-#
 # CONFIG_IPMI_HANDLER is not set
-CONFIG_WATCHDOG=y
-# CONFIG_WATCHDOG_NOWAYOUT is not set
-
-#
-# Watchdog Device Drivers
-#
-# CONFIG_SOFT_WATCHDOG is not set
-CONFIG_BFIN_WDT=y
 CONFIG_HW_RANDOM=y
 # CONFIG_GEN_RTC is not set
-CONFIG_BLACKFIN_DPMC=y
 # CONFIG_R3964 is not set
 # CONFIG_RAW_DRIVER is not set
-
-#
-# TPM devices
-#
 # CONFIG_TCG_TPM is not set
 # CONFIG_I2C is not set
 
@@ -748,22 +698,37 @@ CONFIG_SPI_BFIN=y
 #
 # CONFIG_SPI_AT25 is not set
 # CONFIG_SPI_SPIDEV is not set
-
-#
-# Dallas's 1-wire bus
-#
+# CONFIG_SPI_TLE62X0 is not set
 # CONFIG_W1 is not set
+# CONFIG_POWER_SUPPLY is not set
 CONFIG_HWMON=y
 # CONFIG_HWMON_VID is not set
-# CONFIG_SENSORS_ABITUGURU is not set
 # CONFIG_SENSORS_F71805F is not set
+# CONFIG_SENSORS_F71882FG is not set
+# CONFIG_SENSORS_IT87 is not set
 # CONFIG_SENSORS_LM70 is not set
+# CONFIG_SENSORS_PC87360 is not set
 # CONFIG_SENSORS_PC87427 is not set
 # CONFIG_SENSORS_SMSC47M1 is not set
 # CONFIG_SENSORS_SMSC47B397 is not set
 # CONFIG_SENSORS_VT1211 is not set
 # CONFIG_SENSORS_W83627HF is not set
+# CONFIG_SENSORS_W83627EHF is not set
 # CONFIG_HWMON_DEBUG_CHIP is not set
+CONFIG_WATCHDOG=y
+# CONFIG_WATCHDOG_NOWAYOUT is not set
+
+#
+# Watchdog Device Drivers
+#
+# CONFIG_SOFT_WATCHDOG is not set
+CONFIG_BFIN_WDT=y
+
+#
+# Sonics Silicon Backplane
+#
+CONFIG_SSB_POSSIBLE=y
+# CONFIG_SSB is not set
 
 #
 # Multifunction device drivers
@@ -780,72 +745,27 @@ CONFIG_DAB=y
 #
 # Graphics support
 #
+# CONFIG_VGASTATE is not set
+# CONFIG_VIDEO_OUTPUT_CONTROL is not set
+# CONFIG_FB is not set
 # CONFIG_BACKLIGHT_LCD_SUPPORT is not set
 
 #
 # Display device support
 #
 # CONFIG_DISPLAY_SUPPORT is not set
-# CONFIG_VGASTATE is not set
-# CONFIG_FB is not set
 
 #
 # Sound
 #
 # CONFIG_SOUND is not set
-
-#
-# HID Devices
-#
+CONFIG_HID_SUPPORT=y
 CONFIG_HID=m
 # CONFIG_HID_DEBUG is not set
-
-#
-# USB support
-#
-CONFIG_USB_ARCH_HAS_HCD=y
-# CONFIG_USB_ARCH_HAS_OHCI is not set
-# CONFIG_USB_ARCH_HAS_EHCI is not set
-# CONFIG_USB is not set
-
-#
-# Enable Host or Gadget support to see Inventra options
-#
-
-#
-# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support'
-#
-
-#
-# USB Gadget Support
-#
-# CONFIG_USB_GADGET is not set
+# CONFIG_HIDRAW is not set
+# CONFIG_USB_SUPPORT is not set
 # CONFIG_MMC is not set
-
-#
-# LED devices
-#
 # CONFIG_NEW_LEDS is not set
-
-#
-# LED drivers
-#
-
-#
-# LED Triggers
-#
-
-#
-# InfiniBand support
-#
-
-#
-# EDAC - error detection and reporting (RAS) (EXPERIMENTAL)
-#
-
-#
-# Real Time Clock
-#
 CONFIG_RTC_LIB=y
 CONFIG_RTC_CLASS=y
 CONFIG_RTC_HCTOSYS=y
@@ -861,10 +781,6 @@ CONFIG_RTC_INTF_DEV=y
 # CONFIG_RTC_INTF_DEV_UIE_EMUL is not set
 # CONFIG_RTC_DRV_TEST is not set
 
-#
-# I2C RTC drivers
-#
-
 #
 # SPI RTC drivers
 #
@@ -875,8 +791,10 @@ CONFIG_RTC_INTF_DEV=y
 # Platform RTC drivers
 #
 # CONFIG_RTC_DRV_DS1553 is not set
+# CONFIG_RTC_DRV_STK17TA8 is not set
 # CONFIG_RTC_DRV_DS1742 is not set
 # CONFIG_RTC_DRV_M48T86 is not set
+# CONFIG_RTC_DRV_M48T59 is not set
 # CONFIG_RTC_DRV_V3020 is not set
 
 #
@@ -885,22 +803,9 @@ CONFIG_RTC_INTF_DEV=y
 CONFIG_RTC_DRV_BFIN=y
 
 #
-# DMA Engine support
+# Userspace I/O
 #
-# CONFIG_DMA_ENGINE is not set
-
-#
-# DMA Clients
-#
-
-#
-# DMA Devices
-#
-
-#
-# PBX support
-#
-# CONFIG_PBX is not set
+# CONFIG_UIO is not set
 
 #
 # File systems
@@ -945,7 +850,6 @@ CONFIG_PROC_SYSCTL=y
 CONFIG_SYSFS=y
 # CONFIG_TMPFS is not set
 # CONFIG_HUGETLB_PAGE is not set
-CONFIG_RAMFS=y
 # CONFIG_CONFIGFS_FS is not set
 
 #
@@ -971,10 +875,12 @@ CONFIG_YAFFS_SHORT_NAMES_IN_RAM=y
 CONFIG_JFFS2_FS=m
 CONFIG_JFFS2_FS_DEBUG=0
 CONFIG_JFFS2_FS_WRITEBUFFER=y
+# CONFIG_JFFS2_FS_WBUF_VERIFY is not set
 # CONFIG_JFFS2_SUMMARY is not set
 # CONFIG_JFFS2_FS_XATTR is not set
 # CONFIG_JFFS2_COMPRESSION_OPTIONS is not set
 CONFIG_JFFS2_ZLIB=y
+# CONFIG_JFFS2_LZO is not set
 CONFIG_JFFS2_RTIME=y
 # CONFIG_JFFS2_RUBIN is not set
 # CONFIG_CRAMFS is not set
@@ -983,10 +889,7 @@ CONFIG_JFFS2_RTIME=y
 # CONFIG_QNX4FS_FS is not set
 # CONFIG_SYSV_FS is not set
 # CONFIG_UFS_FS is not set
-
-#
-# Network File Systems
-#
+CONFIG_NETWORK_FILESYSTEMS=y
 CONFIG_NFS_FS=m
 CONFIG_NFS_V3=y
 # CONFIG_NFS_V3_ACL is not set
@@ -1006,17 +909,12 @@ CONFIG_SMB_FS=m
 # CONFIG_NCP_FS is not set
 # CONFIG_CODA_FS is not set
 # CONFIG_AFS_FS is not set
-# CONFIG_9P_FS is not set
 
 #
 # Partition Types
 #
 # CONFIG_PARTITION_ADVANCED is not set
 CONFIG_MSDOS_PARTITION=y
-
-#
-# Native Language Support
-#
 CONFIG_NLS=m
 CONFIG_NLS_DEFAULT="iso8859-1"
 # CONFIG_NLS_CODEPAGE_437 is not set
@@ -1057,21 +955,16 @@ CONFIG_NLS_DEFAULT="iso8859-1"
 # CONFIG_NLS_KOI8_R is not set
 # CONFIG_NLS_KOI8_U is not set
 # CONFIG_NLS_UTF8 is not set
-
-#
-# Distributed Lock Manager
-#
 # CONFIG_DLM is not set
-
-#
-# Profiling support
-#
+CONFIG_INSTRUMENTATION=y
 # CONFIG_PROFILING is not set
+# CONFIG_MARKERS is not set
 
 #
 # Kernel hacking
 #
 # CONFIG_PRINTK_TIME is not set
+CONFIG_ENABLE_WARN_DEPRECATED=y
 CONFIG_ENABLE_MUST_CHECK=y
 # CONFIG_MAGIC_SYSRQ is not set
 # CONFIG_UNUSED_SYMBOLS is not set
@@ -1079,6 +972,7 @@ CONFIG_DEBUG_FS=y
 # CONFIG_HEADERS_CHECK is not set
 # CONFIG_DEBUG_KERNEL is not set
 # CONFIG_DEBUG_BUGVERBOSE is not set
+# CONFIG_SAMPLES is not set
 CONFIG_DEBUG_MMRS=y
 CONFIG_DEBUG_HUNT_FOR_ZERO=y
 CONFIG_DEBUG_BFIN_HWTRACE_ON=y
@@ -1098,11 +992,7 @@ CONFIG_ACCESS_CHECK=y
 # CONFIG_KEYS is not set
 CONFIG_SECURITY=y
 # CONFIG_SECURITY_NETWORK is not set
-CONFIG_SECURITY_CAPABILITIES=m
-
-#
-# Cryptographic options
-#
+# CONFIG_SECURITY_CAPABILITIES is not set
 # CONFIG_CRYPTO is not set
 
 #
@@ -1113,6 +1003,7 @@ CONFIG_CRC_CCITT=m
 # CONFIG_CRC16 is not set
 # CONFIG_CRC_ITU_T is not set
 CONFIG_CRC32=y
+# CONFIG_CRC7 is not set
 # CONFIG_LIBCRC32C is not set
 CONFIG_ZLIB_INFLATE=y
 CONFIG_ZLIB_DEFLATE=m
diff --git a/arch/blackfin/configs/BF533-STAMP_defconfig b/arch/blackfin/configs/BF533-STAMP_defconfig
index 20d598d17bd1..d77d991a1f61 100644
--- a/arch/blackfin/configs/BF533-STAMP_defconfig
+++ b/arch/blackfin/configs/BF533-STAMP_defconfig
@@ -1,6 +1,6 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.22.16
+# Linux kernel version: 2.6.24.7
 #
 # CONFIG_MMU is not set
 # CONFIG_FPU is not set
@@ -13,35 +13,34 @@ CONFIG_GENERIC_FIND_NEXT_BIT=y
 CONFIG_GENERIC_HWEIGHT=y
 CONFIG_GENERIC_HARDIRQS=y
 CONFIG_GENERIC_IRQ_PROBE=y
-CONFIG_GENERIC_TIME=y
 CONFIG_GENERIC_GPIO=y
 CONFIG_FORCE_MAX_ZONEORDER=14
 CONFIG_GENERIC_CALIBRATE_DELAY=y
 CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
 
 #
-# Code maturity level options
+# General setup
 #
 CONFIG_EXPERIMENTAL=y
 CONFIG_BROKEN_ON_SMP=y
 CONFIG_INIT_ENV_ARG_LIMIT=32
-
-#
-# General setup
-#
 CONFIG_LOCALVERSION=""
 CONFIG_LOCALVERSION_AUTO=y
 CONFIG_SYSVIPC=y
-# CONFIG_IPC_NS is not set
 CONFIG_SYSVIPC_SYSCTL=y
 # CONFIG_POSIX_MQUEUE is not set
 # CONFIG_BSD_PROCESS_ACCT is not set
 # CONFIG_TASKSTATS is not set
-# CONFIG_UTS_NS is not set
+# CONFIG_USER_NS is not set
+# CONFIG_PID_NS is not set
 # CONFIG_AUDIT is not set
 CONFIG_IKCONFIG=y
 CONFIG_IKCONFIG_PROC=y
 CONFIG_LOG_BUF_SHIFT=14
+# CONFIG_CGROUPS is not set
+CONFIG_FAIR_GROUP_SCHED=y
+CONFIG_FAIR_USER_SCHED=y
+# CONFIG_FAIR_CGROUP_SCHED is not set
 CONFIG_SYSFS_DEPRECATED=y
 # CONFIG_RELAY is not set
 CONFIG_BLK_DEV_INITRD=y
@@ -64,32 +63,24 @@ CONFIG_EPOLL=y
 CONFIG_SIGNALFD=y
 CONFIG_EVENTFD=y
 CONFIG_VM_EVENT_COUNTERS=y
-CONFIG_BIG_ORDER_ALLOC_NOFAIL_MAGIC=3
-# CONFIG_NP2 is not set
 CONFIG_SLAB=y
 # CONFIG_SLUB is not set
 # CONFIG_SLOB is not set
+CONFIG_SLABINFO=y
 CONFIG_RT_MUTEXES=y
 CONFIG_TINY_SHMEM=y
 CONFIG_BASE_SMALL=0
-
-#
-# Loadable module support
-#
 CONFIG_MODULES=y
 CONFIG_MODULE_UNLOAD=y
 # CONFIG_MODULE_FORCE_UNLOAD is not set
 # CONFIG_MODVERSIONS is not set
 # CONFIG_MODULE_SRCVERSION_ALL is not set
 CONFIG_KMOD=y
-
-#
-# Block layer
-#
 CONFIG_BLOCK=y
 # CONFIG_LBD is not set
 # CONFIG_BLK_DEV_IO_TRACE is not set
 # CONFIG_LSF is not set
+# CONFIG_BLK_DEV_BSG is not set
 
 #
 # IO Schedulers
@@ -141,12 +132,12 @@ CONFIG_BF_REV_0_3=y
 # CONFIG_BF_REV_ANY is not set
 # CONFIG_BF_REV_NONE is not set
 CONFIG_BF53x=y
-CONFIG_BFIN_SINGLE_CORE=y
 CONFIG_MEM_MT48LC64M4A2FB_7E=y
 # CONFIG_BFIN533_EZKIT is not set
 CONFIG_BFIN533_STAMP=y
 # CONFIG_BFIN533_BLUETECHNIX_CM is not set
 # CONFIG_H8606_HVSISTEMAS is not set
+# CONFIG_BFIN532_IP0X is not set
 # CONFIG_GENERIC_BF533_BOARD is not set
 
 #
@@ -189,12 +180,14 @@ CONFIG_WDTIMER=13
 # Board customizations
 #
 # CONFIG_CMDLINE_BOOL is not set
+CONFIG_BOOT_LOAD=0x1000
 
 #
 # Clock/PLL Setup
 #
 CONFIG_CLKIN_HZ=11059200
 # CONFIG_BFIN_KERNEL_CLOCK is not set
+CONFIG_MAX_MEM_SIZE=512
 CONFIG_MAX_VCO_HZ=750000000
 CONFIG_MIN_VCO_HZ=50000000
 CONFIG_MAX_SCLK_HZ=133333333
@@ -208,14 +201,17 @@ CONFIG_HZ_250=y
 # CONFIG_HZ_300 is not set
 # CONFIG_HZ_1000 is not set
 CONFIG_HZ=250
+CONFIG_GENERIC_TIME=y
+CONFIG_GENERIC_CLOCKEVENTS=y
+# CONFIG_CYCLES_CLOCKSOURCE is not set
+CONFIG_TICK_ONESHOT=y
+# CONFIG_NO_HZ is not set
+CONFIG_HIGH_RES_TIMERS=y
+CONFIG_GENERIC_CLOCKEVENTS_BUILD=y
 
 #
-# Memory Setup
+# Misc
 #
-CONFIG_MAX_MEM_SIZE=512
-CONFIG_MEM_ADD_WIDTH=11
-CONFIG_ENET_FLASH_PIN=0
-CONFIG_BOOT_LOAD=0x1000
 CONFIG_BFIN_SCRATCH_REG_RETN=y
 # CONFIG_BFIN_SCRATCH_REG_RETE is not set
 # CONFIG_BFIN_SCRATCH_REG_CYCLES is not set
@@ -251,12 +247,14 @@ CONFIG_FLATMEM_MANUAL=y
 CONFIG_FLATMEM=y
 CONFIG_FLAT_NODE_MEM_MAP=y
 # CONFIG_SPARSEMEM_STATIC is not set
+# CONFIG_SPARSEMEM_VMEMMAP_ENABLE is not set
 CONFIG_SPLIT_PTLOCK_CPUS=4
 # CONFIG_RESOURCES_64BIT is not set
 CONFIG_ZONE_DMA_FLAG=1
-CONFIG_LARGE_ALLOCS=y
+CONFIG_VIRT_TO_BUS=y
 # CONFIG_BFIN_GPTIMERS is not set
 CONFIG_BFIN_DMA_5XX=y
+# CONFIG_DMA_UNCACHED_4M is not set
 # CONFIG_DMA_UNCACHED_2M is not set
 CONFIG_DMA_UNCACHED_1M=y
 # CONFIG_DMA_UNCACHED_NONE is not set
@@ -294,17 +292,13 @@ CONFIG_C_AMBEN_ALL=y
 CONFIG_BANK_0=0x7BB0
 CONFIG_BANK_1=0x7BB0
 CONFIG_BANK_2=0x7BB0
-CONFIG_BANK_3=0xAAC3
+CONFIG_BANK_3=0xAAC2
 
 #
 # Bus options (PCI, PCMCIA, EISA, MCA, ISA)
 #
 # CONFIG_PCI is not set
 # CONFIG_ARCH_SUPPORTS_MSI is not set
-
-#
-# PCCARD (PCMCIA/CardBus) support
-#
 # CONFIG_PCCARD is not set
 
 #
@@ -322,7 +316,9 @@ CONFIG_BINFMT_ZFLAT=y
 CONFIG_PM=y
 # CONFIG_PM_LEGACY is not set
 # CONFIG_PM_DEBUG is not set
-# CONFIG_PM_SYSFS_DEPRECATED is not set
+CONFIG_PM_SLEEP=y
+CONFIG_SUSPEND_UP_POSSIBLE=y
+CONFIG_SUSPEND=y
 CONFIG_PM_BFIN_SLEEP_DEEPER=y
 # CONFIG_PM_BFIN_SLEEP is not set
 # CONFIG_PM_WAKEUP_BY_GPIO is not set
@@ -368,6 +364,7 @@ CONFIG_SYN_COOKIES=y
 CONFIG_INET_XFRM_MODE_TRANSPORT=y
 CONFIG_INET_XFRM_MODE_TUNNEL=y
 CONFIG_INET_XFRM_MODE_BEET=y
+# CONFIG_INET_LRO is not set
 CONFIG_INET_DIAG=y
 CONFIG_INET_TCP_DIAG=y
 # CONFIG_TCP_CONG_ADVANCED is not set
@@ -394,10 +391,6 @@ CONFIG_DEFAULT_TCP_CONG="cubic"
 # CONFIG_LAPB is not set
 # CONFIG_ECONET is not set
 # CONFIG_WAN_ROUTER is not set
-
-#
-# QoS and/or fair queueing
-#
 # CONFIG_NET_SCHED is not set
 
 #
@@ -429,6 +422,9 @@ CONFIG_IRDA_CACHE_LAST_LSAP=y
 # SIR device drivers
 #
 CONFIG_IRTTY_SIR=m
+CONFIG_BFIN_SIR=m
+CONFIG_SIR_BFIN_DMA=y
+# CONFIG_SIR_BFIN_PIO is not set
 
 #
 # Dongle support
@@ -458,6 +454,7 @@ CONFIG_IRTTY_SIR=m
 # CONFIG_MAC80211 is not set
 # CONFIG_IEEE80211 is not set
 # CONFIG_RFKILL is not set
+# CONFIG_NET_9P is not set
 
 #
 # Device Drivers
@@ -466,14 +463,11 @@ CONFIG_IRTTY_SIR=m
 #
 # Generic Driver Options
 #
+CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
 CONFIG_STANDALONE=y
 CONFIG_PREVENT_FIRMWARE_BUILD=y
 # CONFIG_FW_LOADER is not set
 # CONFIG_SYS_HYPERVISOR is not set
-
-#
-# Connector - unified userspace <-> kernelspace linker
-#
 # CONFIG_CONNECTOR is not set
 CONFIG_MTD=y
 # CONFIG_MTD_DEBUG is not set
@@ -493,6 +487,7 @@ CONFIG_MTD_BLOCK=y
 # CONFIG_INFTL is not set
 # CONFIG_RFD_FTL is not set
 # CONFIG_SSFDC is not set
+# CONFIG_MTD_OOPS is not set
 
 #
 # RAM/ROM/Flash chip drivers
@@ -524,11 +519,7 @@ CONFIG_MTD_ROM=m
 #
 CONFIG_MTD_COMPLEX_MAPPINGS=y
 # CONFIG_MTD_PHYSMAP is not set
-CONFIG_MTD_BF5xx=m
-CONFIG_BFIN_FLASH_BANK_0=0x7BB0
-CONFIG_BFIN_FLASH_BANK_1=0x7BB0
-CONFIG_BFIN_FLASH_BANK_2=0x7BB0
-CONFIG_BFIN_FLASH_BANK_3=0x7BB0
+CONFIG_MTD_BFIN_ASYNC=m
 # CONFIG_MTD_UCLINUX is not set
 # CONFIG_MTD_PLATRAM is not set
 
@@ -555,20 +546,8 @@ CONFIG_BFIN_FLASH_BANK_3=0x7BB0
 # UBI - Unsorted block images
 #
 # CONFIG_MTD_UBI is not set
-
-#
-# Parallel port support
-#
 # CONFIG_PARPORT is not set
-
-#
-# Plug and Play support
-#
-# CONFIG_PNPACPI is not set
-
-#
-# Block devices
-#
+CONFIG_BLK_DEV=y
 # CONFIG_BLK_DEV_COW_COMMON is not set
 # CONFIG_BLK_DEV_LOOP is not set
 # CONFIG_BLK_DEV_NBD is not set
@@ -578,10 +557,8 @@ CONFIG_BLK_DEV_RAM_SIZE=4096
 CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024
 # CONFIG_CDROM_PKTCDVD is not set
 # CONFIG_ATA_OVER_ETH is not set
-
-#
-# Misc devices
-#
+CONFIG_MISC_DEVICES=y
+# CONFIG_EEPROM_93CX6 is not set
 # CONFIG_IDE is not set
 
 #
@@ -589,32 +566,29 @@ CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024
 #
 # CONFIG_RAID_ATTRS is not set
 # CONFIG_SCSI is not set
+# CONFIG_SCSI_DMA is not set
 # CONFIG_SCSI_NETLINK is not set
 # CONFIG_ATA is not set
-
-#
-# Multi-device support (RAID and LVM)
-#
 # CONFIG_MD is not set
-
-#
-# Network device support
-#
 CONFIG_NETDEVICES=y
+# CONFIG_NETDEVICES_MULTIQUEUE is not set
 # CONFIG_DUMMY is not set
 # CONFIG_BONDING is not set
+# CONFIG_MACVLAN is not set
 # CONFIG_EQUALIZER is not set
 # CONFIG_TUN is not set
+# CONFIG_VETH is not set
 # CONFIG_PHYLIB is not set
-
-#
-# Ethernet (10 or 100Mbit)
-#
 CONFIG_NET_ETHERNET=y
 CONFIG_MII=y
 CONFIG_SMC91X=y
 # CONFIG_SMSC911X is not set
 # CONFIG_DM9000 is not set
+# CONFIG_IBM_NEW_EMAC_ZMII is not set
+# CONFIG_IBM_NEW_EMAC_RGMII is not set
+# CONFIG_IBM_NEW_EMAC_TAH is not set
+# CONFIG_IBM_NEW_EMAC_EMAC4 is not set
+# CONFIG_B44 is not set
 CONFIG_NETDEV_1000=y
 # CONFIG_AX88180 is not set
 CONFIG_NETDEV_10000=y
@@ -631,15 +605,7 @@ CONFIG_NETDEV_10000=y
 # CONFIG_NETCONSOLE is not set
 # CONFIG_NETPOLL is not set
 # CONFIG_NET_POLL_CONTROLLER is not set
-
-#
-# ISDN subsystem
-#
 # CONFIG_ISDN is not set
-
-#
-# Telephony Support
-#
 # CONFIG_PHONE is not set
 
 #
@@ -654,7 +620,6 @@ CONFIG_INPUT=y
 #
 # CONFIG_INPUT_MOUSEDEV is not set
 # CONFIG_INPUT_JOYDEV is not set
-# CONFIG_INPUT_TSDEV is not set
 CONFIG_INPUT_EVDEV=m
 # CONFIG_INPUT_EVBUG is not set
 
@@ -667,14 +632,8 @@ CONFIG_INPUT_EVDEV=m
 # CONFIG_INPUT_TABLET is not set
 # CONFIG_INPUT_TOUCHSCREEN is not set
 CONFIG_INPUT_MISC=y
-# CONFIG_INPUT_ATI_REMOTE is not set
-# CONFIG_INPUT_ATI_REMOTE2 is not set
-# CONFIG_INPUT_KEYSPAN_REMOTE is not set
-# CONFIG_INPUT_POWERMATE is not set
-# CONFIG_INPUT_YEALINK is not set
 # CONFIG_INPUT_UINPUT is not set
 CONFIG_TWI_KEYPAD=m
-CONFIG_BFIN_TWIKEYPAD_IRQ_PFX=39
 
 #
 # Hardware I/O ports
@@ -687,15 +646,13 @@ CONFIG_BFIN_TWIKEYPAD_IRQ_PFX=39
 #
 # CONFIG_AD9960 is not set
 # CONFIG_SPI_ADC_BF533 is not set
-# CONFIG_BF5xx_PFLAGS is not set
 # CONFIG_BF5xx_PPIFCD is not set
 # CONFIG_BFIN_SIMPLE_TIMER is not set
 # CONFIG_BF5xx_PPI is not set
 CONFIG_BFIN_SPORT=y
 # CONFIG_BFIN_TIMER_LATENCY is not set
 CONFIG_TWI_LCD=m
-CONFIG_TWI_LCD_SLAVE_ADDR=34
-# CONFIG_AD5304 is not set
+CONFIG_SIMPLE_GPIO=m
 # CONFIG_VT is not set
 # CONFIG_SERIAL_NONSTANDARD is not set
 
@@ -723,28 +680,11 @@ CONFIG_UNIX98_PTYS=y
 # CAN, the car bus and industrial fieldbus
 #
 # CONFIG_CAN4LINUX is not set
-
-#
-# IPMI
-#
 # CONFIG_IPMI_HANDLER is not set
-CONFIG_WATCHDOG=y
-# CONFIG_WATCHDOG_NOWAYOUT is not set
-
-#
-# Watchdog Device Drivers
-#
-# CONFIG_SOFT_WATCHDOG is not set
-CONFIG_BFIN_WDT=y
 CONFIG_HW_RANDOM=y
 # CONFIG_GEN_RTC is not set
-CONFIG_BLACKFIN_DPMC=y
 # CONFIG_R3964 is not set
 # CONFIG_RAW_DRIVER is not set
-
-#
-# TPM devices
-#
 # CONFIG_TCG_TPM is not set
 CONFIG_I2C=m
 CONFIG_I2C_BOARDINFO=y
@@ -764,6 +704,7 @@ CONFIG_I2C_ALGOBIT=m
 # CONFIG_I2C_OCORES is not set
 # CONFIG_I2C_PARPORT_LIGHT is not set
 # CONFIG_I2C_SIMTEC is not set
+# CONFIG_I2C_TAOS_EVM is not set
 # CONFIG_I2C_STUB is not set
 
 #
@@ -771,14 +712,15 @@ CONFIG_I2C_ALGOBIT=m
 #
 # CONFIG_SENSORS_DS1337 is not set
 # CONFIG_SENSORS_DS1374 is not set
+# CONFIG_DS1682 is not set
 # CONFIG_SENSORS_AD5252 is not set
 # CONFIG_SENSORS_EEPROM is not set
 # CONFIG_SENSORS_PCF8574 is not set
 # CONFIG_SENSORS_PCF8575 is not set
-# CONFIG_SENSORS_PCA9543 is not set
 # CONFIG_SENSORS_PCA9539 is not set
 # CONFIG_SENSORS_PCF8591 is not set
 # CONFIG_SENSORS_MAX6875 is not set
+# CONFIG_SENSORS_TSL2550 is not set
 # CONFIG_I2C_DEBUG_CORE is not set
 # CONFIG_I2C_DEBUG_ALGO is not set
 # CONFIG_I2C_DEBUG_BUS is not set
@@ -801,14 +743,11 @@ CONFIG_SPI_BFIN=y
 #
 # CONFIG_SPI_AT25 is not set
 # CONFIG_SPI_SPIDEV is not set
-
-#
-# Dallas's 1-wire bus
-#
+# CONFIG_SPI_TLE62X0 is not set
 # CONFIG_W1 is not set
+# CONFIG_POWER_SUPPLY is not set
 CONFIG_HWMON=y
 # CONFIG_HWMON_VID is not set
-# CONFIG_SENSORS_ABITUGURU is not set
 # CONFIG_SENSORS_AD7418 is not set
 # CONFIG_SENSORS_ADM1021 is not set
 # CONFIG_SENSORS_ADM1025 is not set
@@ -816,12 +755,12 @@ CONFIG_HWMON=y
 # CONFIG_SENSORS_ADM1029 is not set
 # CONFIG_SENSORS_ADM1031 is not set
 # CONFIG_SENSORS_ADM9240 is not set
-# CONFIG_SENSORS_ASB100 is not set
+# CONFIG_SENSORS_ADT7470 is not set
 # CONFIG_SENSORS_ATXP1 is not set
 # CONFIG_SENSORS_DS1621 is not set
 # CONFIG_SENSORS_F71805F is not set
-# CONFIG_SENSORS_FSCHER is not set
-# CONFIG_SENSORS_FSCPOS is not set
+# CONFIG_SENSORS_F71882FG is not set
+# CONFIG_SENSORS_F75375S is not set
 # CONFIG_SENSORS_GL518SM is not set
 # CONFIG_SENSORS_GL520SM is not set
 # CONFIG_SENSORS_IT87 is not set
@@ -836,13 +775,16 @@ CONFIG_HWMON=y
 # CONFIG_SENSORS_LM87 is not set
 # CONFIG_SENSORS_LM90 is not set
 # CONFIG_SENSORS_LM92 is not set
+# CONFIG_SENSORS_LM93 is not set
 # CONFIG_SENSORS_MAX1619 is not set
 # CONFIG_SENSORS_MAX6650 is not set
 # CONFIG_SENSORS_PC87360 is not set
 # CONFIG_SENSORS_PC87427 is not set
+# CONFIG_SENSORS_DME1737 is not set
 # CONFIG_SENSORS_SMSC47M1 is not set
 # CONFIG_SENSORS_SMSC47M192 is not set
 # CONFIG_SENSORS_SMSC47B397 is not set
+# CONFIG_SENSORS_THMC50 is not set
 # CONFIG_SENSORS_VT1211 is not set
 # CONFIG_SENSORS_W83781D is not set
 # CONFIG_SENSORS_W83791D is not set
@@ -852,6 +794,20 @@ CONFIG_HWMON=y
 # CONFIG_SENSORS_W83627HF is not set
 # CONFIG_SENSORS_W83627EHF is not set
 # CONFIG_HWMON_DEBUG_CHIP is not set
+CONFIG_WATCHDOG=y
+# CONFIG_WATCHDOG_NOWAYOUT is not set
+
+#
+# Watchdog Device Drivers
+#
+# CONFIG_SOFT_WATCHDOG is not set
+CONFIG_BFIN_WDT=y
+
+#
+# Sonics Silicon Backplane
+#
+CONFIG_SSB_POSSIBLE=y
+# CONFIG_SSB is not set
 
 #
 # Multifunction device drivers
@@ -863,24 +819,20 @@ CONFIG_HWMON=y
 #
 # CONFIG_VIDEO_DEV is not set
 # CONFIG_DVB_CORE is not set
-CONFIG_DAB=y
+# CONFIG_DAB is not set
 
 #
 # Graphics support
 #
-# CONFIG_BACKLIGHT_LCD_SUPPORT is not set
-
-#
-# Display device support
-#
-# CONFIG_DISPLAY_SUPPORT is not set
 # CONFIG_VGASTATE is not set
+# CONFIG_VIDEO_OUTPUT_CONTROL is not set
 CONFIG_FB=m
 CONFIG_FIRMWARE_EDID=y
 # CONFIG_FB_DDC is not set
 CONFIG_FB_CFB_FILLRECT=m
 CONFIG_FB_CFB_COPYAREA=m
 CONFIG_FB_CFB_IMAGEBLIT=m
+# CONFIG_FB_CFB_REV_PIXELS_IN_BYTE is not set
 # CONFIG_FB_SYS_FILLRECT is not set
 # CONFIG_FB_SYS_COPYAREA is not set
 # CONFIG_FB_SYS_IMAGEBLIT is not set
@@ -895,7 +847,7 @@ CONFIG_FB_DEFERRED_IO=y
 #
 # Frame buffer hardware drivers
 #
-CONFIG_FB_BFIN_7171=m
+# CONFIG_FB_BFIN_T350MCQB is not set
 CONFIG_FB_BFIN_7393=m
 CONFIG_NTSC=y
 # CONFIG_PAL is not set
@@ -905,9 +857,14 @@ CONFIG_NTSC=y
 # CONFIG_PAL_YCBCR is not set
 CONFIG_ADV7393_1XMEM=y
 # CONFIG_ADV7393_2XMEM is not set
-# CONFIG_FB_BFIN_T350MCQB is not set
 # CONFIG_FB_S1D13XXX is not set
 # CONFIG_FB_VIRTUAL is not set
+# CONFIG_BACKLIGHT_LCD_SUPPORT is not set
+
+#
+# Display device support
+#
+# CONFIG_DISPLAY_SUPPORT is not set
 # CONFIG_LOGO is not set
 
 #
@@ -940,6 +897,10 @@ CONFIG_SND_VERBOSE_PROCFS=y
 # CONFIG_SND_SERIAL_U16550 is not set
 # CONFIG_SND_MPU401 is not set
 
+#
+# SPI devices
+#
+
 #
 # ALSA Blackfin devices
 #
@@ -953,69 +914,43 @@ CONFIG_SND_BLACKFIN_SPI_PFBIT=4
 CONFIG_SND_BFIN_AD73311=m
 CONFIG_SND_BFIN_SPORT=0
 CONFIG_SND_BFIN_AD73311_SE=4
+CONFIG_SND_BFIN_AD73322=m
+CONFIG_SND_BFIN_AD73322_SPORT0_SE=10
+CONFIG_SND_BFIN_AD73322_SPORT1_SE=14
+CONFIG_SND_BFIN_AD73322_RESET=12
 
 #
 # System on Chip audio support
 #
-# CONFIG_SND_SOC is not set
+CONFIG_SND_SOC_AC97_BUS=y
+CONFIG_SND_SOC=m
+CONFIG_SND_BF5XX_SOC=m
+CONFIG_SND_MMAP_SUPPORT=y
+CONFIG_SND_BF5XX_SOC_AC97=m
+# CONFIG_SND_BF5XX_SOC_WM8750 is not set
+# CONFIG_SND_BF5XX_SOC_WM8731 is not set
+# CONFIG_SND_BF5XX_SOC_SSM2602 is not set
+CONFIG_SND_BF5XX_SOC_BF5xx=m
+CONFIG_SND_BF5XX_SPORT_NUM=0
+# CONFIG_SND_BF5XX_HAVE_COLD_RESET is not set
+
+#
+# SoC Audio support for SuperH
+#
+CONFIG_SND_SOC_AD1980=m
 
 #
 # Open Sound System
 #
 # CONFIG_SOUND_PRIME is not set
-
-#
-# HID Devices
-#
+CONFIG_AC97_BUS=m
+CONFIG_HID_SUPPORT=y
 CONFIG_HID=y
 # CONFIG_HID_DEBUG is not set
-
-#
-# USB support
-#
-CONFIG_USB_ARCH_HAS_HCD=y
-# CONFIG_USB_ARCH_HAS_OHCI is not set
-# CONFIG_USB_ARCH_HAS_EHCI is not set
-# CONFIG_USB is not set
-
-#
-# Enable Host or Gadget support to see Inventra options
-#
-
-#
-# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support'
-#
-
-#
-# USB Gadget Support
-#
-# CONFIG_USB_GADGET is not set
+# CONFIG_HIDRAW is not set
+# CONFIG_USB_SUPPORT is not set
 # CONFIG_MMC is not set
-
-#
-# LED devices
-#
 # CONFIG_NEW_LEDS is not set
-
-#
-# LED drivers
-#
-
-#
-# LED Triggers
-#
-
-#
-# InfiniBand support
-#
-
-#
-# EDAC - error detection and reporting (RAS) (EXPERIMENTAL)
-#
-
-#
-# Real Time Clock
-#
 CONFIG_RTC_LIB=y
 CONFIG_RTC_CLASS=y
 CONFIG_RTC_HCTOSYS=y
@@ -1035,6 +970,7 @@ CONFIG_RTC_INTF_DEV=y
 # I2C RTC drivers
 #
 # CONFIG_RTC_DRV_DS1307 is not set
+# CONFIG_RTC_DRV_DS1374 is not set
 # CONFIG_RTC_DRV_DS1672 is not set
 # CONFIG_RTC_DRV_MAX6900 is not set
 # CONFIG_RTC_DRV_RS5C372 is not set
@@ -1042,6 +978,7 @@ CONFIG_RTC_INTF_DEV=y
 # CONFIG_RTC_DRV_X1205 is not set
 # CONFIG_RTC_DRV_PCF8563 is not set
 # CONFIG_RTC_DRV_PCF8583 is not set
+# CONFIG_RTC_DRV_M41T80 is not set
 
 #
 # SPI RTC drivers
@@ -1053,8 +990,10 @@ CONFIG_RTC_INTF_DEV=y
 # Platform RTC drivers
 #
 # CONFIG_RTC_DRV_DS1553 is not set
+# CONFIG_RTC_DRV_STK17TA8 is not set
 # CONFIG_RTC_DRV_DS1742 is not set
 # CONFIG_RTC_DRV_M48T86 is not set
+# CONFIG_RTC_DRV_M48T59 is not set
 # CONFIG_RTC_DRV_V3020 is not set
 
 #
@@ -1063,22 +1002,9 @@ CONFIG_RTC_INTF_DEV=y
 CONFIG_RTC_DRV_BFIN=y
 
 #
-# DMA Engine support
+# Userspace I/O
 #
-# CONFIG_DMA_ENGINE is not set
-
-#
-# DMA Clients
-#
-
-#
-# DMA Devices
-#
-
-#
-# PBX support
-#
-# CONFIG_PBX is not set
+# CONFIG_UIO is not set
 
 #
 # File systems
@@ -1123,7 +1049,6 @@ CONFIG_PROC_SYSCTL=y
 CONFIG_SYSFS=y
 # CONFIG_TMPFS is not set
 # CONFIG_HUGETLB_PAGE is not set
-CONFIG_RAMFS=y
 # CONFIG_CONFIGFS_FS is not set
 
 #
@@ -1149,10 +1074,12 @@ CONFIG_YAFFS_SHORT_NAMES_IN_RAM=y
 CONFIG_JFFS2_FS=m
 CONFIG_JFFS2_FS_DEBUG=0
 CONFIG_JFFS2_FS_WRITEBUFFER=y
+# CONFIG_JFFS2_FS_WBUF_VERIFY is not set
 # CONFIG_JFFS2_SUMMARY is not set
 # CONFIG_JFFS2_FS_XATTR is not set
 # CONFIG_JFFS2_COMPRESSION_OPTIONS is not set
 CONFIG_JFFS2_ZLIB=y
+# CONFIG_JFFS2_LZO is not set
 CONFIG_JFFS2_RTIME=y
 # CONFIG_JFFS2_RUBIN is not set
 # CONFIG_CRAMFS is not set
@@ -1161,10 +1088,7 @@ CONFIG_JFFS2_RTIME=y
 # CONFIG_QNX4FS_FS is not set
 # CONFIG_SYSV_FS is not set
 # CONFIG_UFS_FS is not set
-
-#
-# Network File Systems
-#
+CONFIG_NETWORK_FILESYSTEMS=y
 CONFIG_NFS_FS=m
 CONFIG_NFS_V3=y
 # CONFIG_NFS_V3_ACL is not set
@@ -1184,17 +1108,12 @@ CONFIG_SMB_FS=m
 # CONFIG_NCP_FS is not set
 # CONFIG_CODA_FS is not set
 # CONFIG_AFS_FS is not set
-# CONFIG_9P_FS is not set
 
 #
 # Partition Types
 #
 # CONFIG_PARTITION_ADVANCED is not set
 CONFIG_MSDOS_PARTITION=y
-
-#
-# Native Language Support
-#
 CONFIG_NLS=m
 CONFIG_NLS_DEFAULT="iso8859-1"
 # CONFIG_NLS_CODEPAGE_437 is not set
@@ -1235,21 +1154,16 @@ CONFIG_NLS_DEFAULT="iso8859-1"
 # CONFIG_NLS_KOI8_R is not set
 # CONFIG_NLS_KOI8_U is not set
 # CONFIG_NLS_UTF8 is not set
-
-#
-# Distributed Lock Manager
-#
 # CONFIG_DLM is not set
-
-#
-# Profiling support
-#
+CONFIG_INSTRUMENTATION=y
 # CONFIG_PROFILING is not set
+# CONFIG_MARKERS is not set
 
 #
 # Kernel hacking
 #
 # CONFIG_PRINTK_TIME is not set
+CONFIG_ENABLE_WARN_DEPRECATED=y
 CONFIG_ENABLE_MUST_CHECK=y
 # CONFIG_MAGIC_SYSRQ is not set
 # CONFIG_UNUSED_SYMBOLS is not set
@@ -1257,6 +1171,7 @@ CONFIG_DEBUG_FS=y
 # CONFIG_HEADERS_CHECK is not set
 # CONFIG_DEBUG_KERNEL is not set
 # CONFIG_DEBUG_BUGVERBOSE is not set
+# CONFIG_SAMPLES is not set
 CONFIG_DEBUG_MMRS=y
 CONFIG_DEBUG_HUNT_FOR_ZERO=y
 CONFIG_DEBUG_BFIN_HWTRACE_ON=y
@@ -1276,11 +1191,7 @@ CONFIG_ACCESS_CHECK=y
 # CONFIG_KEYS is not set
 CONFIG_SECURITY=y
 # CONFIG_SECURITY_NETWORK is not set
-CONFIG_SECURITY_CAPABILITIES=m
-
-#
-# Cryptographic options
-#
+# CONFIG_SECURITY_CAPABILITIES is not set
 # CONFIG_CRYPTO is not set
 
 #
@@ -1291,6 +1202,7 @@ CONFIG_CRC_CCITT=m
 # CONFIG_CRC16 is not set
 # CONFIG_CRC_ITU_T is not set
 CONFIG_CRC32=y
+# CONFIG_CRC7 is not set
 # CONFIG_LIBCRC32C is not set
 CONFIG_ZLIB_INFLATE=y
 CONFIG_ZLIB_DEFLATE=m
diff --git a/arch/blackfin/configs/BF537-STAMP_defconfig b/arch/blackfin/configs/BF537-STAMP_defconfig
index b5189c8ba263..5fd7c4b143df 100644
--- a/arch/blackfin/configs/BF537-STAMP_defconfig
+++ b/arch/blackfin/configs/BF537-STAMP_defconfig
@@ -1,6 +1,6 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.22.16
+# Linux kernel version: 2.6.24.7
 #
 # CONFIG_MMU is not set
 # CONFIG_FPU is not set
@@ -13,35 +13,34 @@ CONFIG_GENERIC_FIND_NEXT_BIT=y
 CONFIG_GENERIC_HWEIGHT=y
 CONFIG_GENERIC_HARDIRQS=y
 CONFIG_GENERIC_IRQ_PROBE=y
-CONFIG_GENERIC_TIME=y
 CONFIG_GENERIC_GPIO=y
 CONFIG_FORCE_MAX_ZONEORDER=14
 CONFIG_GENERIC_CALIBRATE_DELAY=y
 CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
 
 #
-# Code maturity level options
+# General setup
 #
 CONFIG_EXPERIMENTAL=y
 CONFIG_BROKEN_ON_SMP=y
 CONFIG_INIT_ENV_ARG_LIMIT=32
-
-#
-# General setup
-#
 CONFIG_LOCALVERSION=""
 CONFIG_LOCALVERSION_AUTO=y
 CONFIG_SYSVIPC=y
-# CONFIG_IPC_NS is not set
 CONFIG_SYSVIPC_SYSCTL=y
 # CONFIG_POSIX_MQUEUE is not set
 # CONFIG_BSD_PROCESS_ACCT is not set
 # CONFIG_TASKSTATS is not set
-# CONFIG_UTS_NS is not set
+# CONFIG_USER_NS is not set
+# CONFIG_PID_NS is not set
 # CONFIG_AUDIT is not set
 CONFIG_IKCONFIG=y
 CONFIG_IKCONFIG_PROC=y
 CONFIG_LOG_BUF_SHIFT=14
+# CONFIG_CGROUPS is not set
+CONFIG_FAIR_GROUP_SCHED=y
+CONFIG_FAIR_USER_SCHED=y
+# CONFIG_FAIR_CGROUP_SCHED is not set
 CONFIG_SYSFS_DEPRECATED=y
 # CONFIG_RELAY is not set
 CONFIG_BLK_DEV_INITRD=y
@@ -64,32 +63,24 @@ CONFIG_EPOLL=y
 CONFIG_SIGNALFD=y
 CONFIG_EVENTFD=y
 CONFIG_VM_EVENT_COUNTERS=y
-CONFIG_BIG_ORDER_ALLOC_NOFAIL_MAGIC=3
-# CONFIG_NP2 is not set
 CONFIG_SLAB=y
 # CONFIG_SLUB is not set
 # CONFIG_SLOB is not set
+CONFIG_SLABINFO=y
 CONFIG_RT_MUTEXES=y
 CONFIG_TINY_SHMEM=y
 CONFIG_BASE_SMALL=0
-
-#
-# Loadable module support
-#
 CONFIG_MODULES=y
 CONFIG_MODULE_UNLOAD=y
 # CONFIG_MODULE_FORCE_UNLOAD is not set
 # CONFIG_MODVERSIONS is not set
 # CONFIG_MODULE_SRCVERSION_ALL is not set
 CONFIG_KMOD=y
-
-#
-# Block layer
-#
 CONFIG_BLOCK=y
 # CONFIG_LBD is not set
 # CONFIG_BLK_DEV_IO_TRACE is not set
 # CONFIG_LSF is not set
+# CONFIG_BLK_DEV_BSG is not set
 
 #
 # IO Schedulers
@@ -141,7 +132,6 @@ CONFIG_BF_REV_0_2=y
 # CONFIG_BF_REV_ANY is not set
 # CONFIG_BF_REV_NONE is not set
 CONFIG_BF53x=y
-CONFIG_BFIN_SINGLE_CORE=y
 CONFIG_MEM_MT48LC32M8A2_75=y
 CONFIG_IRQ_PLL_WAKEUP=7
 CONFIG_IRQ_RTC=8
@@ -197,12 +187,14 @@ CONFIG_IRQ_PROG_INTA=12
 # Board customizations
 #
 # CONFIG_CMDLINE_BOOL is not set
+CONFIG_BOOT_LOAD=0x1000
 
 #
 # Clock/PLL Setup
 #
 CONFIG_CLKIN_HZ=25000000
 # CONFIG_BFIN_KERNEL_CLOCK is not set
+CONFIG_MAX_MEM_SIZE=512
 CONFIG_MAX_VCO_HZ=600000000
 CONFIG_MIN_VCO_HZ=50000000
 CONFIG_MAX_SCLK_HZ=133333333
@@ -216,13 +208,17 @@ CONFIG_HZ_250=y
 # CONFIG_HZ_300 is not set
 # CONFIG_HZ_1000 is not set
 CONFIG_HZ=250
+CONFIG_GENERIC_TIME=y
+CONFIG_GENERIC_CLOCKEVENTS=y
+# CONFIG_CYCLES_CLOCKSOURCE is not set
+CONFIG_TICK_ONESHOT=y
+# CONFIG_NO_HZ is not set
+CONFIG_HIGH_RES_TIMERS=y
+CONFIG_GENERIC_CLOCKEVENTS_BUILD=y
 
 #
-# Memory Setup
+# Misc
 #
-CONFIG_MAX_MEM_SIZE=512
-CONFIG_MEM_ADD_WIDTH=10
-CONFIG_BOOT_LOAD=0x1000
 CONFIG_BFIN_SCRATCH_REG_RETN=y
 # CONFIG_BFIN_SCRATCH_REG_RETE is not set
 # CONFIG_BFIN_SCRATCH_REG_CYCLES is not set
@@ -258,12 +254,14 @@ CONFIG_FLATMEM_MANUAL=y
 CONFIG_FLATMEM=y
 CONFIG_FLAT_NODE_MEM_MAP=y
 # CONFIG_SPARSEMEM_STATIC is not set
+# CONFIG_SPARSEMEM_VMEMMAP_ENABLE is not set
 CONFIG_SPLIT_PTLOCK_CPUS=4
 # CONFIG_RESOURCES_64BIT is not set
 CONFIG_ZONE_DMA_FLAG=1
-CONFIG_LARGE_ALLOCS=y
+CONFIG_VIRT_TO_BUS=y
 # CONFIG_BFIN_GPTIMERS is not set
 CONFIG_BFIN_DMA_5XX=y
+# CONFIG_DMA_UNCACHED_4M is not set
 # CONFIG_DMA_UNCACHED_2M is not set
 CONFIG_DMA_UNCACHED_1M=y
 # CONFIG_DMA_UNCACHED_NONE is not set
@@ -301,17 +299,13 @@ CONFIG_C_AMBEN_ALL=y
 CONFIG_BANK_0=0x7BB0
 CONFIG_BANK_1=0x7BB0
 CONFIG_BANK_2=0x7BB0
-CONFIG_BANK_3=0x99B3
+CONFIG_BANK_3=0x99B2
 
 #
 # Bus options (PCI, PCMCIA, EISA, MCA, ISA)
 #
 # CONFIG_PCI is not set
 # CONFIG_ARCH_SUPPORTS_MSI is not set
-
-#
-# PCCARD (PCMCIA/CardBus) support
-#
 # CONFIG_PCCARD is not set
 
 #
@@ -329,7 +323,9 @@ CONFIG_BINFMT_ZFLAT=y
 CONFIG_PM=y
 # CONFIG_PM_LEGACY is not set
 # CONFIG_PM_DEBUG is not set
-# CONFIG_PM_SYSFS_DEPRECATED is not set
+CONFIG_PM_SLEEP=y
+CONFIG_SUSPEND_UP_POSSIBLE=y
+CONFIG_SUSPEND=y
 CONFIG_PM_BFIN_SLEEP_DEEPER=y
 # CONFIG_PM_BFIN_SLEEP is not set
 # CONFIG_PM_WAKEUP_BY_GPIO is not set
@@ -375,6 +371,7 @@ CONFIG_SYN_COOKIES=y
 CONFIG_INET_XFRM_MODE_TRANSPORT=y
 CONFIG_INET_XFRM_MODE_TUNNEL=y
 CONFIG_INET_XFRM_MODE_BEET=y
+# CONFIG_INET_LRO is not set
 CONFIG_INET_DIAG=y
 CONFIG_INET_TCP_DIAG=y
 # CONFIG_TCP_CONG_ADVANCED is not set
@@ -401,10 +398,6 @@ CONFIG_DEFAULT_TCP_CONG="cubic"
 # CONFIG_LAPB is not set
 # CONFIG_ECONET is not set
 # CONFIG_WAN_ROUTER is not set
-
-#
-# QoS and/or fair queueing
-#
 # CONFIG_NET_SCHED is not set
 
 #
@@ -436,6 +429,10 @@ CONFIG_IRDA_CACHE_LAST_LSAP=y
 # SIR device drivers
 #
 CONFIG_IRTTY_SIR=m
+CONFIG_BFIN_SIR=m
+CONFIG_BFIN_SIR1=y
+CONFIG_SIR_BFIN_DMA=y
+# CONFIG_SIR_BFIN_PIO is not set
 
 #
 # Dongle support
@@ -465,6 +462,7 @@ CONFIG_IRTTY_SIR=m
 # CONFIG_MAC80211 is not set
 # CONFIG_IEEE80211 is not set
 # CONFIG_RFKILL is not set
+# CONFIG_NET_9P is not set
 
 #
 # Device Drivers
@@ -473,14 +471,11 @@ CONFIG_IRTTY_SIR=m
 #
 # Generic Driver Options
 #
+CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
 CONFIG_STANDALONE=y
 CONFIG_PREVENT_FIRMWARE_BUILD=y
 # CONFIG_FW_LOADER is not set
 # CONFIG_SYS_HYPERVISOR is not set
-
-#
-# Connector - unified userspace <-> kernelspace linker
-#
 # CONFIG_CONNECTOR is not set
 CONFIG_MTD=y
 # CONFIG_MTD_DEBUG is not set
@@ -500,6 +495,7 @@ CONFIG_MTD_BLOCK=y
 # CONFIG_INFTL is not set
 # CONFIG_RFD_FTL is not set
 # CONFIG_SSFDC is not set
+# CONFIG_MTD_OOPS is not set
 
 #
 # RAM/ROM/Flash chip drivers
@@ -572,20 +568,8 @@ CONFIG_MTD_NAND_IDS=m
 # UBI - Unsorted block images
 #
 # CONFIG_MTD_UBI is not set
-
-#
-# Parallel port support
-#
 # CONFIG_PARPORT is not set
-
-#
-# Plug and Play support
-#
-# CONFIG_PNPACPI is not set
-
-#
-# Block devices
-#
+CONFIG_BLK_DEV=y
 # CONFIG_BLK_DEV_COW_COMMON is not set
 # CONFIG_BLK_DEV_LOOP is not set
 # CONFIG_BLK_DEV_NBD is not set
@@ -595,10 +579,8 @@ CONFIG_BLK_DEV_RAM_SIZE=4096
 CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024
 # CONFIG_CDROM_PKTCDVD is not set
 # CONFIG_ATA_OVER_ETH is not set
-
-#
-# Misc devices
-#
+CONFIG_MISC_DEVICES=y
+# CONFIG_EEPROM_93CX6 is not set
 # CONFIG_IDE is not set
 
 #
@@ -606,22 +588,18 @@ CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024
 #
 # CONFIG_RAID_ATTRS is not set
 # CONFIG_SCSI is not set
+# CONFIG_SCSI_DMA is not set
 # CONFIG_SCSI_NETLINK is not set
 # CONFIG_ATA is not set
-
-#
-# Multi-device support (RAID and LVM)
-#
 # CONFIG_MD is not set
-
-#
-# Network device support
-#
 CONFIG_NETDEVICES=y
+# CONFIG_NETDEVICES_MULTIQUEUE is not set
 # CONFIG_DUMMY is not set
 # CONFIG_BONDING is not set
+# CONFIG_MACVLAN is not set
 # CONFIG_EQUALIZER is not set
 # CONFIG_TUN is not set
+# CONFIG_VETH is not set
 CONFIG_PHYLIB=y
 
 #
@@ -635,21 +613,24 @@ CONFIG_PHYLIB=y
 # CONFIG_VITESSE_PHY is not set
 CONFIG_SMSC_PHY=y
 # CONFIG_BROADCOM_PHY is not set
+# CONFIG_ICPLUS_PHY is not set
 # CONFIG_FIXED_PHY is not set
-
-#
-# Ethernet (10 or 100Mbit)
-#
+# CONFIG_MDIO_BITBANG is not set
 CONFIG_NET_ETHERNET=y
 CONFIG_MII=y
-# CONFIG_SMC91X is not set
 CONFIG_BFIN_MAC=y
 CONFIG_BFIN_MAC_USE_L1=y
 CONFIG_BFIN_TX_DESC_NUM=10
 CONFIG_BFIN_RX_DESC_NUM=20
 # CONFIG_BFIN_MAC_RMII is not set
+# CONFIG_SMC91X is not set
 # CONFIG_SMSC911X is not set
 # CONFIG_DM9000 is not set
+# CONFIG_IBM_NEW_EMAC_ZMII is not set
+# CONFIG_IBM_NEW_EMAC_RGMII is not set
+# CONFIG_IBM_NEW_EMAC_TAH is not set
+# CONFIG_IBM_NEW_EMAC_EMAC4 is not set
+# CONFIG_B44 is not set
 CONFIG_NETDEV_1000=y
 # CONFIG_AX88180 is not set
 CONFIG_NETDEV_10000=y
@@ -666,15 +647,7 @@ CONFIG_NETDEV_10000=y
 # CONFIG_NETCONSOLE is not set
 # CONFIG_NETPOLL is not set
 # CONFIG_NET_POLL_CONTROLLER is not set
-
-#
-# ISDN subsystem
-#
 # CONFIG_ISDN is not set
-
-#
-# Telephony Support
-#
 # CONFIG_PHONE is not set
 
 #
@@ -689,7 +662,6 @@ CONFIG_INPUT=y
 #
 # CONFIG_INPUT_MOUSEDEV is not set
 # CONFIG_INPUT_JOYDEV is not set
-# CONFIG_INPUT_TSDEV is not set
 CONFIG_INPUT_EVDEV=m
 # CONFIG_INPUT_EVBUG is not set
 
@@ -702,14 +674,8 @@ CONFIG_INPUT_EVDEV=m
 # CONFIG_INPUT_TABLET is not set
 # CONFIG_INPUT_TOUCHSCREEN is not set
 CONFIG_INPUT_MISC=y
-# CONFIG_INPUT_ATI_REMOTE is not set
-# CONFIG_INPUT_ATI_REMOTE2 is not set
-# CONFIG_INPUT_KEYSPAN_REMOTE is not set
-# CONFIG_INPUT_POWERMATE is not set
-# CONFIG_INPUT_YEALINK is not set
 # CONFIG_INPUT_UINPUT is not set
 CONFIG_TWI_KEYPAD=m
-CONFIG_BFIN_TWIKEYPAD_IRQ_PFX=72
 
 #
 # Hardware I/O ports
@@ -722,15 +688,13 @@ CONFIG_BFIN_TWIKEYPAD_IRQ_PFX=72
 #
 # CONFIG_AD9960 is not set
 # CONFIG_SPI_ADC_BF533 is not set
-# CONFIG_BF5xx_PFLAGS is not set
 # CONFIG_BF5xx_PPIFCD is not set
 # CONFIG_BFIN_SIMPLE_TIMER is not set
 # CONFIG_BF5xx_PPI is not set
 CONFIG_BFIN_SPORT=y
 # CONFIG_BFIN_TIMER_LATENCY is not set
 CONFIG_TWI_LCD=m
-CONFIG_TWI_LCD_SLAVE_ADDR=34
-# CONFIG_AD5304 is not set
+CONFIG_SIMPLE_GPIO=m
 # CONFIG_VT is not set
 # CONFIG_SERIAL_NONSTANDARD is not set
 
@@ -766,28 +730,11 @@ CONFIG_CAN4LINUX=y
 # CONFIG_CAN_MCF5282 is not set
 # CONFIG_CAN_UNCTWINCAN is not set
 CONFIG_CAN_BLACKFIN=m
-
-#
-# IPMI
-#
 # CONFIG_IPMI_HANDLER is not set
-CONFIG_WATCHDOG=y
-# CONFIG_WATCHDOG_NOWAYOUT is not set
-
-#
-# Watchdog Device Drivers
-#
-# CONFIG_SOFT_WATCHDOG is not set
-CONFIG_BFIN_WDT=y
 CONFIG_HW_RANDOM=y
 # CONFIG_GEN_RTC is not set
-CONFIG_BLACKFIN_DPMC=y
 # CONFIG_R3964 is not set
 # CONFIG_RAW_DRIVER is not set
-
-#
-# TPM devices
-#
 # CONFIG_TCG_TPM is not set
 CONFIG_I2C=m
 CONFIG_I2C_BOARDINFO=y
@@ -809,6 +756,7 @@ CONFIG_I2C_BLACKFIN_TWI_CLK_KHZ=50
 # CONFIG_I2C_OCORES is not set
 # CONFIG_I2C_PARPORT_LIGHT is not set
 # CONFIG_I2C_SIMTEC is not set
+# CONFIG_I2C_TAOS_EVM is not set
 # CONFIG_I2C_STUB is not set
 
 #
@@ -816,14 +764,15 @@ CONFIG_I2C_BLACKFIN_TWI_CLK_KHZ=50
 #
 # CONFIG_SENSORS_DS1337 is not set
 # CONFIG_SENSORS_DS1374 is not set
+# CONFIG_DS1682 is not set
 CONFIG_SENSORS_AD5252=m
 # CONFIG_SENSORS_EEPROM is not set
 # CONFIG_SENSORS_PCF8574 is not set
 # CONFIG_SENSORS_PCF8575 is not set
-# CONFIG_SENSORS_PCA9543 is not set
 # CONFIG_SENSORS_PCA9539 is not set
 # CONFIG_SENSORS_PCF8591 is not set
 # CONFIG_SENSORS_MAX6875 is not set
+# CONFIG_SENSORS_TSL2550 is not set
 # CONFIG_I2C_DEBUG_CORE is not set
 # CONFIG_I2C_DEBUG_ALGO is not set
 # CONFIG_I2C_DEBUG_BUS is not set
@@ -846,14 +795,11 @@ CONFIG_SPI_BFIN=y
 #
 # CONFIG_SPI_AT25 is not set
 # CONFIG_SPI_SPIDEV is not set
-
-#
-# Dallas's 1-wire bus
-#
+# CONFIG_SPI_TLE62X0 is not set
 # CONFIG_W1 is not set
+# CONFIG_POWER_SUPPLY is not set
 CONFIG_HWMON=y
 # CONFIG_HWMON_VID is not set
-# CONFIG_SENSORS_ABITUGURU is not set
 # CONFIG_SENSORS_AD7418 is not set
 # CONFIG_SENSORS_ADM1021 is not set
 # CONFIG_SENSORS_ADM1025 is not set
@@ -861,12 +807,12 @@ CONFIG_HWMON=y
 # CONFIG_SENSORS_ADM1029 is not set
 # CONFIG_SENSORS_ADM1031 is not set
 # CONFIG_SENSORS_ADM9240 is not set
-# CONFIG_SENSORS_ASB100 is not set
+# CONFIG_SENSORS_ADT7470 is not set
 # CONFIG_SENSORS_ATXP1 is not set
 # CONFIG_SENSORS_DS1621 is not set
 # CONFIG_SENSORS_F71805F is not set
-# CONFIG_SENSORS_FSCHER is not set
-# CONFIG_SENSORS_FSCPOS is not set
+# CONFIG_SENSORS_F71882FG is not set
+# CONFIG_SENSORS_F75375S is not set
 # CONFIG_SENSORS_GL518SM is not set
 # CONFIG_SENSORS_GL520SM is not set
 # CONFIG_SENSORS_IT87 is not set
@@ -881,13 +827,16 @@ CONFIG_HWMON=y
 # CONFIG_SENSORS_LM87 is not set
 # CONFIG_SENSORS_LM90 is not set
 # CONFIG_SENSORS_LM92 is not set
+# CONFIG_SENSORS_LM93 is not set
 # CONFIG_SENSORS_MAX1619 is not set
 # CONFIG_SENSORS_MAX6650 is not set
 # CONFIG_SENSORS_PC87360 is not set
 # CONFIG_SENSORS_PC87427 is not set
+# CONFIG_SENSORS_DME1737 is not set
 # CONFIG_SENSORS_SMSC47M1 is not set
 # CONFIG_SENSORS_SMSC47M192 is not set
 # CONFIG_SENSORS_SMSC47B397 is not set
+# CONFIG_SENSORS_THMC50 is not set
 # CONFIG_SENSORS_VT1211 is not set
 # CONFIG_SENSORS_W83781D is not set
 # CONFIG_SENSORS_W83791D is not set
@@ -897,6 +846,20 @@ CONFIG_HWMON=y
 # CONFIG_SENSORS_W83627HF is not set
 # CONFIG_SENSORS_W83627EHF is not set
 # CONFIG_HWMON_DEBUG_CHIP is not set
+CONFIG_WATCHDOG=y
+# CONFIG_WATCHDOG_NOWAYOUT is not set
+
+#
+# Watchdog Device Drivers
+#
+# CONFIG_SOFT_WATCHDOG is not set
+CONFIG_BFIN_WDT=y
+
+#
+# Sonics Silicon Backplane
+#
+CONFIG_SSB_POSSIBLE=y
+# CONFIG_SSB is not set
 
 #
 # Multifunction device drivers
@@ -913,21 +876,15 @@ CONFIG_DAB=y
 #
 # Graphics support
 #
-CONFIG_BACKLIGHT_LCD_SUPPORT=y
-CONFIG_BACKLIGHT_CLASS_DEVICE=m
-CONFIG_LCD_CLASS_DEVICE=m
-
-#
-# Display device support
-#
-# CONFIG_DISPLAY_SUPPORT is not set
 # CONFIG_VGASTATE is not set
+# CONFIG_VIDEO_OUTPUT_CONTROL is not set
 CONFIG_FB=m
 CONFIG_FIRMWARE_EDID=y
 # CONFIG_FB_DDC is not set
 CONFIG_FB_CFB_FILLRECT=m
 CONFIG_FB_CFB_COPYAREA=m
 CONFIG_FB_CFB_IMAGEBLIT=m
+# CONFIG_FB_CFB_REV_PIXELS_IN_BYTE is not set
 # CONFIG_FB_SYS_FILLRECT is not set
 # CONFIG_FB_SYS_COPYAREA is not set
 # CONFIG_FB_SYS_IMAGEBLIT is not set
@@ -942,7 +899,8 @@ CONFIG_FB_DEFERRED_IO=y
 #
 # Frame buffer hardware drivers
 #
-CONFIG_FB_BFIN_7171=m
+# CONFIG_FB_HITACHI_TX09 is not set
+# CONFIG_FB_BFIN_T350MCQB is not set
 CONFIG_FB_BFIN_7393=m
 CONFIG_NTSC=y
 # CONFIG_PAL is not set
@@ -956,10 +914,18 @@ CONFIG_FB_BF537_LQ035=m
 CONFIG_LQ035_SLAVE_ADDR=0x58
 # CONFIG_FB_BFIN_LANDSCAPE is not set
 # CONFIG_FB_BFIN_BGR is not set
-# CONFIG_FB_BFIN_T350MCQB is not set
-# CONFIG_FB_HITACHI_TX09 is not set
 # CONFIG_FB_S1D13XXX is not set
 # CONFIG_FB_VIRTUAL is not set
+CONFIG_BACKLIGHT_LCD_SUPPORT=y
+CONFIG_LCD_CLASS_DEVICE=m
+# CONFIG_LCD_LTV350QV is not set
+CONFIG_BACKLIGHT_CLASS_DEVICE=m
+CONFIG_BACKLIGHT_CORGI=m
+
+#
+# Display device support
+#
+# CONFIG_DISPLAY_SUPPORT is not set
 # CONFIG_LOGO is not set
 
 #
@@ -992,6 +958,10 @@ CONFIG_SND_VERBOSE_PROCFS=y
 # CONFIG_SND_SERIAL_U16550 is not set
 # CONFIG_SND_MPU401 is not set
 
+#
+# SPI devices
+#
+
 #
 # ALSA Blackfin devices
 #
@@ -1005,6 +975,10 @@ CONFIG_SND_BLACKFIN_SPI_PFBIT=4
 CONFIG_SND_BFIN_AD73311=m
 CONFIG_SND_BFIN_SPORT=0
 CONFIG_SND_BFIN_AD73311_SE=4
+CONFIG_SND_BFIN_AD73322=m
+CONFIG_SND_BFIN_AD73322_SPORT0_SE=10
+CONFIG_SND_BFIN_AD73322_SPORT1_SE=14
+CONFIG_SND_BFIN_AD73322_RESET=12
 
 #
 # System on Chip audio support
@@ -1016,9 +990,14 @@ CONFIG_SND_MMAP_SUPPORT=y
 CONFIG_SND_BF5XX_SOC_AC97=m
 # CONFIG_SND_BF5XX_SOC_WM8750 is not set
 # CONFIG_SND_BF5XX_SOC_WM8731 is not set
+# CONFIG_SND_BF5XX_SOC_SSM2602 is not set
 CONFIG_SND_BF5XX_SOC_BF5xx=m
 CONFIG_SND_BF5XX_SPORT_NUM=0
 # CONFIG_SND_BF5XX_HAVE_COLD_RESET is not set
+
+#
+# SoC Audio support for SuperH
+#
 CONFIG_SND_SOC_AD1980=m
 
 #
@@ -1026,59 +1005,18 @@ CONFIG_SND_SOC_AD1980=m
 #
 # CONFIG_SOUND_PRIME is not set
 CONFIG_AC97_BUS=m
-
-#
-# HID Devices
-#
+CONFIG_HID_SUPPORT=y
 CONFIG_HID=y
 # CONFIG_HID_DEBUG is not set
-
-#
-# USB support
-#
-CONFIG_USB_ARCH_HAS_HCD=y
-# CONFIG_USB_ARCH_HAS_OHCI is not set
-# CONFIG_USB_ARCH_HAS_EHCI is not set
-# CONFIG_USB is not set
-
-#
-# Enable Host or Gadget support to see Inventra options
-#
-
-#
-# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support'
-#
-
-#
-# USB Gadget Support
-#
-# CONFIG_USB_GADGET is not set
+# CONFIG_HIDRAW is not set
+# CONFIG_USB_SUPPORT is not set
+# CONFIG_NO_DUMMY_DELAY is not set
+# CONFIG_DUMMY_DELAY_BANK0 is not set
+# CONFIG_DUMMY_DELAY_BANK1 is not set
+# CONFIG_DUMMY_DELAY_BANK2 is not set
+# CONFIG_DUMMY_DELAY_BANK3 is not set
 # CONFIG_MMC is not set
-
-#
-# LED devices
-#
 # CONFIG_NEW_LEDS is not set
-
-#
-# LED drivers
-#
-
-#
-# LED Triggers
-#
-
-#
-# InfiniBand support
-#
-
-#
-# EDAC - error detection and reporting (RAS) (EXPERIMENTAL)
-#
-
-#
-# Real Time Clock
-#
 CONFIG_RTC_LIB=y
 CONFIG_RTC_CLASS=y
 CONFIG_RTC_HCTOSYS=y
@@ -1098,6 +1036,7 @@ CONFIG_RTC_INTF_DEV=y
 # I2C RTC drivers
 #
 # CONFIG_RTC_DRV_DS1307 is not set
+# CONFIG_RTC_DRV_DS1374 is not set
 # CONFIG_RTC_DRV_DS1672 is not set
 # CONFIG_RTC_DRV_MAX6900 is not set
 # CONFIG_RTC_DRV_RS5C372 is not set
@@ -1105,6 +1044,7 @@ CONFIG_RTC_INTF_DEV=y
 # CONFIG_RTC_DRV_X1205 is not set
 # CONFIG_RTC_DRV_PCF8563 is not set
 # CONFIG_RTC_DRV_PCF8583 is not set
+# CONFIG_RTC_DRV_M41T80 is not set
 
 #
 # SPI RTC drivers
@@ -1116,8 +1056,10 @@ CONFIG_RTC_INTF_DEV=y
 # Platform RTC drivers
 #
 # CONFIG_RTC_DRV_DS1553 is not set
+# CONFIG_RTC_DRV_STK17TA8 is not set
 # CONFIG_RTC_DRV_DS1742 is not set
 # CONFIG_RTC_DRV_M48T86 is not set
+# CONFIG_RTC_DRV_M48T59 is not set
 # CONFIG_RTC_DRV_V3020 is not set
 
 #
@@ -1126,22 +1068,9 @@ CONFIG_RTC_INTF_DEV=y
 CONFIG_RTC_DRV_BFIN=y
 
 #
-# DMA Engine support
+# Userspace I/O
 #
-# CONFIG_DMA_ENGINE is not set
-
-#
-# DMA Clients
-#
-
-#
-# DMA Devices
-#
-
-#
-# PBX support
-#
-# CONFIG_PBX is not set
+# CONFIG_UIO is not set
 
 #
 # File systems
@@ -1186,7 +1115,6 @@ CONFIG_PROC_SYSCTL=y
 CONFIG_SYSFS=y
 # CONFIG_TMPFS is not set
 # CONFIG_HUGETLB_PAGE is not set
-CONFIG_RAMFS=y
 # CONFIG_CONFIGFS_FS is not set
 
 #
@@ -1212,10 +1140,12 @@ CONFIG_YAFFS_SHORT_NAMES_IN_RAM=y
 CONFIG_JFFS2_FS=m
 CONFIG_JFFS2_FS_DEBUG=0
 CONFIG_JFFS2_FS_WRITEBUFFER=y
+# CONFIG_JFFS2_FS_WBUF_VERIFY is not set
 # CONFIG_JFFS2_SUMMARY is not set
 # CONFIG_JFFS2_FS_XATTR is not set
 # CONFIG_JFFS2_COMPRESSION_OPTIONS is not set
 CONFIG_JFFS2_ZLIB=y
+# CONFIG_JFFS2_LZO is not set
 CONFIG_JFFS2_RTIME=y
 # CONFIG_JFFS2_RUBIN is not set
 # CONFIG_CRAMFS is not set
@@ -1224,10 +1154,7 @@ CONFIG_JFFS2_RTIME=y
 # CONFIG_QNX4FS_FS is not set
 # CONFIG_SYSV_FS is not set
 # CONFIG_UFS_FS is not set
-
-#
-# Network File Systems
-#
+CONFIG_NETWORK_FILESYSTEMS=y
 CONFIG_NFS_FS=m
 CONFIG_NFS_V3=y
 # CONFIG_NFS_V3_ACL is not set
@@ -1247,17 +1174,12 @@ CONFIG_SMB_FS=m
 # CONFIG_NCP_FS is not set
 # CONFIG_CODA_FS is not set
 # CONFIG_AFS_FS is not set
-# CONFIG_9P_FS is not set
 
 #
 # Partition Types
 #
 # CONFIG_PARTITION_ADVANCED is not set
 CONFIG_MSDOS_PARTITION=y
-
-#
-# Native Language Support
-#
 CONFIG_NLS=m
 CONFIG_NLS_DEFAULT="iso8859-1"
 # CONFIG_NLS_CODEPAGE_437 is not set
@@ -1298,21 +1220,16 @@ CONFIG_NLS_DEFAULT="iso8859-1"
 # CONFIG_NLS_KOI8_R is not set
 # CONFIG_NLS_KOI8_U is not set
 # CONFIG_NLS_UTF8 is not set
-
-#
-# Distributed Lock Manager
-#
 # CONFIG_DLM is not set
-
-#
-# Profiling support
-#
+CONFIG_INSTRUMENTATION=y
 # CONFIG_PROFILING is not set
+# CONFIG_MARKERS is not set
 
 #
 # Kernel hacking
 #
 # CONFIG_PRINTK_TIME is not set
+CONFIG_ENABLE_WARN_DEPRECATED=y
 CONFIG_ENABLE_MUST_CHECK=y
 # CONFIG_MAGIC_SYSRQ is not set
 # CONFIG_UNUSED_SYMBOLS is not set
@@ -1320,6 +1237,7 @@ CONFIG_DEBUG_FS=y
 # CONFIG_HEADERS_CHECK is not set
 # CONFIG_DEBUG_KERNEL is not set
 # CONFIG_DEBUG_BUGVERBOSE is not set
+# CONFIG_SAMPLES is not set
 CONFIG_DEBUG_MMRS=y
 CONFIG_DEBUG_HUNT_FOR_ZERO=y
 CONFIG_DEBUG_BFIN_HWTRACE_ON=y
@@ -1339,11 +1257,7 @@ CONFIG_ACCESS_CHECK=y
 # CONFIG_KEYS is not set
 CONFIG_SECURITY=y
 # CONFIG_SECURITY_NETWORK is not set
-CONFIG_SECURITY_CAPABILITIES=m
-
-#
-# Cryptographic options
-#
+# CONFIG_SECURITY_CAPABILITIES is not set
 # CONFIG_CRYPTO is not set
 
 #
@@ -1354,6 +1268,7 @@ CONFIG_CRC_CCITT=m
 # CONFIG_CRC16 is not set
 # CONFIG_CRC_ITU_T is not set
 CONFIG_CRC32=y
+# CONFIG_CRC7 is not set
 # CONFIG_LIBCRC32C is not set
 CONFIG_ZLIB_INFLATE=y
 CONFIG_ZLIB_DEFLATE=m
diff --git a/arch/blackfin/configs/BF548-EZKIT_defconfig b/arch/blackfin/configs/BF548-EZKIT_defconfig
index 1ff2ff4b49aa..390669e8668e 100644
--- a/arch/blackfin/configs/BF548-EZKIT_defconfig
+++ b/arch/blackfin/configs/BF548-EZKIT_defconfig
@@ -365,7 +365,7 @@ CONFIG_C_AMBEN_ALL=y
 CONFIG_BANK_0=0x7BB0
 CONFIG_BANK_1=0x5554
 CONFIG_BANK_2=0x7BB0
-CONFIG_BANK_3=0x99B3
+CONFIG_BANK_3=0x99B2
 CONFIG_EBIU_MBSCTLVAL=0x0
 CONFIG_EBIU_MODEVAL=0x1
 CONFIG_EBIU_FCTLVAL=0x6
@@ -468,7 +468,60 @@ CONFIG_DEFAULT_TCP_CONG="cubic"
 #
 # CONFIG_NET_PKTGEN is not set
 # CONFIG_HAMRADIO is not set
-# CONFIG_IRDA is not set
+CONFIG_IRDA=m
+
+#
+# IrDA protocols
+#
+CONFIG_IRLAN=m
+CONFIG_IRCOMM=m
+# CONFIG_IRDA_ULTRA is not set
+
+#
+# IrDA options
+#
+# CONFIG_IRDA_CACHE_LAST_LSAP is not set
+# CONFIG_IRDA_FAST_RR is not set
+# CONFIG_IRDA_DEBUG is not set
+
+#
+# Infrared-port device drivers
+#
+
+#
+# SIR device drivers
+#
+CONFIG_IRTTY_SIR=m
+CONFIG_BFIN_SIR=m
+# CONFIG_BFIN_SIR0 is not set
+# CONFIG_BFIN_SIR2 is not set
+CONFIG_BFIN_SIR3=y
+CONFIG_SIR_BFIN_DMA=y
+# CONFIG_SIR_BFIN_PIO is not set
+
+#
+# Dongle support
+#
+# CONFIG_DONGLE is not set
+# CONFIG_KINGSUN_DONGLE is not set
+# CONFIG_KSDAZZLE_DONGLE is not set
+# CONFIG_KS959_DONGLE is not set
+
+#
+# Old SIR device drivers
+#
+# CONFIG_IRPORT_SIR is not set
+
+#
+# Old Serial dongle support
+#
+
+#
+# FIR device drivers
+#
+# CONFIG_USB_IRDA is not set
+# CONFIG_SIGMATEL_FIR is not set
+# CONFIG_MCS_FIR is not set
 # CONFIG_BT is not set
 # CONFIG_AF_RXRPC is not set
 
@@ -575,6 +628,7 @@ CONFIG_MTD_NAND=y
 CONFIG_MTD_NAND_IDS=y
 CONFIG_MTD_NAND_BF5XX=y
 CONFIG_MTD_NAND_BF5XX_HWECC=y
+# CONFIG_MTD_NAND_BF5XX_BOOTROM_ECC is not set
 # CONFIG_MTD_NAND_DISKONCHIP is not set
 # CONFIG_MTD_NAND_NANDSIM is not set
 # CONFIG_MTD_NAND_PLATFORM is not set
@@ -766,7 +820,7 @@ CONFIG_BFIN_OTP=y
 # CONFIG_BFIN_SPORT is not set
 # CONFIG_BFIN_TIMER_LATENCY is not set
 # CONFIG_TWI_LCD is not set
-# CONFIG_SIMPLE_GPIO is not set
+CONFIG_SIMPLE_GPIO=m
 CONFIG_VT=y
 CONFIG_VT_CONSOLE=y
 CONFIG_HW_CONSOLE=y
@@ -1071,6 +1125,7 @@ CONFIG_SND_BF5XX_SOC_AC97=y
 CONFIG_SND_BF5XX_SOC_BF548_EZKIT=y
 # CONFIG_SND_BF5XX_SOC_WM8750 is not set
 # CONFIG_SND_BF5XX_SOC_WM8731 is not set
+# CONFIG_SND_BF5XX_SOC_SSM2602 is not set
 CONFIG_SND_BF5XX_SPORT_NUM=0
 CONFIG_SND_BF5XX_HAVE_COLD_RESET=y
 CONFIG_SND_BF5XX_RESET_GPIO_NUM=19
@@ -1133,7 +1188,7 @@ CONFIG_USB_MUSB_HOST=y
 # CONFIG_USB_MUSB_OTG is not set
 CONFIG_USB_MUSB_HDRC_HCD=y
 # CONFIG_MUSB_PIO_ONLY is not set
-# CONFIG_USB_INVENTRA_DMA is not set
+CONFIG_USB_INVENTRA_DMA=y
 # CONFIG_USB_TI_CPPI_DMA is not set
 CONFIG_USB_MUSB_LOGLEVEL=0
 
@@ -1312,7 +1367,7 @@ CONFIG_FS_MBCACHE=y
 CONFIG_INOTIFY=y
 CONFIG_INOTIFY_USER=y
 # CONFIG_QUOTA is not set
-CONFIG_DNOTIFY=y
+# CONFIG_DNOTIFY is not set
 # CONFIG_AUTOFS_FS is not set
 # CONFIG_AUTOFS4_FS is not set
 # CONFIG_FUSE_FS is not set
diff --git a/arch/blackfin/configs/BF561-EZKIT_defconfig b/arch/blackfin/configs/BF561-EZKIT_defconfig
index b4a20c890816..976a4d7ba175 100644
--- a/arch/blackfin/configs/BF561-EZKIT_defconfig
+++ b/arch/blackfin/configs/BF561-EZKIT_defconfig
@@ -1,6 +1,6 @@
 #
 # Automatically generated make config: don't edit
-# Linux kernel version: 2.6.22.16
+# Linux kernel version: 2.6.24.7
 #
 # CONFIG_MMU is not set
 # CONFIG_FPU is not set
@@ -13,35 +13,34 @@ CONFIG_GENERIC_FIND_NEXT_BIT=y
 CONFIG_GENERIC_HWEIGHT=y
 CONFIG_GENERIC_HARDIRQS=y
 CONFIG_GENERIC_IRQ_PROBE=y
-CONFIG_GENERIC_TIME=y
 CONFIG_GENERIC_GPIO=y
 CONFIG_FORCE_MAX_ZONEORDER=14
 CONFIG_GENERIC_CALIBRATE_DELAY=y
 CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
 
 #
-# Code maturity level options
+# General setup
 #
 CONFIG_EXPERIMENTAL=y
 CONFIG_BROKEN_ON_SMP=y
 CONFIG_INIT_ENV_ARG_LIMIT=32
-
-#
-# General setup
-#
 CONFIG_LOCALVERSION=""
 CONFIG_LOCALVERSION_AUTO=y
 CONFIG_SYSVIPC=y
-# CONFIG_IPC_NS is not set
 CONFIG_SYSVIPC_SYSCTL=y
 # CONFIG_POSIX_MQUEUE is not set
 # CONFIG_BSD_PROCESS_ACCT is not set
 # CONFIG_TASKSTATS is not set
-# CONFIG_UTS_NS is not set
+# CONFIG_USER_NS is not set
+# CONFIG_PID_NS is not set
 # CONFIG_AUDIT is not set
 CONFIG_IKCONFIG=y
 CONFIG_IKCONFIG_PROC=y
 CONFIG_LOG_BUF_SHIFT=14
+# CONFIG_CGROUPS is not set
+CONFIG_FAIR_GROUP_SCHED=y
+CONFIG_FAIR_USER_SCHED=y
+# CONFIG_FAIR_CGROUP_SCHED is not set
 CONFIG_SYSFS_DEPRECATED=y
 # CONFIG_RELAY is not set
 CONFIG_BLK_DEV_INITRD=y
@@ -64,32 +63,24 @@ CONFIG_EPOLL=y
 CONFIG_SIGNALFD=y
 CONFIG_EVENTFD=y
 CONFIG_VM_EVENT_COUNTERS=y
-CONFIG_BIG_ORDER_ALLOC_NOFAIL_MAGIC=3
-# CONFIG_NP2 is not set
 CONFIG_SLAB=y
 # CONFIG_SLUB is not set
 # CONFIG_SLOB is not set
+CONFIG_SLABINFO=y
 CONFIG_RT_MUTEXES=y
 CONFIG_TINY_SHMEM=y
 CONFIG_BASE_SMALL=0
-
-#
-# Loadable module support
-#
 CONFIG_MODULES=y
 CONFIG_MODULE_UNLOAD=y
 # CONFIG_MODULE_FORCE_UNLOAD is not set
 # CONFIG_MODVERSIONS is not set
 # CONFIG_MODULE_SRCVERSION_ALL is not set
 CONFIG_KMOD=y
-
-#
-# Block layer
-#
 CONFIG_BLOCK=y
 # CONFIG_LBD is not set
 # CONFIG_BLK_DEV_IO_TRACE is not set
 # CONFIG_LSF is not set
+# CONFIG_BLK_DEV_BSG is not set
 
 #
 # IO Schedulers
@@ -140,7 +131,6 @@ CONFIG_BF_REV_0_3=y
 # CONFIG_BF_REV_0_5 is not set
 # CONFIG_BF_REV_ANY is not set
 # CONFIG_BF_REV_NONE is not set
-CONFIG_BFIN_DUAL_CORE=y
 CONFIG_MEM_MT48LC16M16A2TG_75=y
 CONFIG_IRQ_PLL_WAKEUP=7
 CONFIG_IRQ_SPORT0_ERROR=7
@@ -233,12 +223,14 @@ CONFIG_IRQ_WDTIMER=13
 # Board customizations
 #
 # CONFIG_CMDLINE_BOOL is not set
+CONFIG_BOOT_LOAD=0x1000
 
 #
 # Clock/PLL Setup
 #
 CONFIG_CLKIN_HZ=30000000
 # CONFIG_BFIN_KERNEL_CLOCK is not set
+CONFIG_MAX_MEM_SIZE=512
 CONFIG_MAX_VCO_HZ=600000000
 CONFIG_MIN_VCO_HZ=50000000
 CONFIG_MAX_SCLK_HZ=133333333
@@ -252,13 +244,17 @@ CONFIG_HZ_250=y
 # CONFIG_HZ_300 is not set
 # CONFIG_HZ_1000 is not set
 CONFIG_HZ=250
+CONFIG_GENERIC_TIME=y
+CONFIG_GENERIC_CLOCKEVENTS=y
+# CONFIG_CYCLES_CLOCKSOURCE is not set
+CONFIG_TICK_ONESHOT=y
+# CONFIG_NO_HZ is not set
+CONFIG_HIGH_RES_TIMERS=y
+CONFIG_GENERIC_CLOCKEVENTS_BUILD=y
 
 #
-# Memory Setup
+# Misc
 #
-CONFIG_MAX_MEM_SIZE=512
-CONFIG_MEM_ADD_WIDTH=9
-CONFIG_BOOT_LOAD=0x1000
 CONFIG_BFIN_SCRATCH_REG_RETN=y
 # CONFIG_BFIN_SCRATCH_REG_RETE is not set
 # CONFIG_BFIN_SCRATCH_REG_CYCLES is not set
@@ -294,12 +290,14 @@ CONFIG_FLATMEM_MANUAL=y
 CONFIG_FLATMEM=y
 CONFIG_FLAT_NODE_MEM_MAP=y
 # CONFIG_SPARSEMEM_STATIC is not set
+# CONFIG_SPARSEMEM_VMEMMAP_ENABLE is not set
 CONFIG_SPLIT_PTLOCK_CPUS=4
 # CONFIG_RESOURCES_64BIT is not set
 CONFIG_ZONE_DMA_FLAG=1
-CONFIG_LARGE_ALLOCS=y
+CONFIG_VIRT_TO_BUS=y
 # CONFIG_BFIN_GPTIMERS is not set
 CONFIG_BFIN_DMA_5XX=y
+# CONFIG_DMA_UNCACHED_4M is not set
 # CONFIG_DMA_UNCACHED_2M is not set
 CONFIG_DMA_UNCACHED_1M=y
 # CONFIG_DMA_UNCACHED_NONE is not set
@@ -341,17 +339,13 @@ CONFIG_C_AMBEN_ALL=y
 CONFIG_BANK_0=0x7BB0
 CONFIG_BANK_1=0x7BB0
 CONFIG_BANK_2=0x7BB0
-CONFIG_BANK_3=0xAAC3
+CONFIG_BANK_3=0xAAC2
 
 #
 # Bus options (PCI, PCMCIA, EISA, MCA, ISA)
 #
 # CONFIG_PCI is not set
 # CONFIG_ARCH_SUPPORTS_MSI is not set
-
-#
-# PCCARD (PCMCIA/CardBus) support
-#
 # CONFIG_PCCARD is not set
 
 #
@@ -367,8 +361,14 @@ CONFIG_BINFMT_ZFLAT=y
 # Power management options
 #
 # CONFIG_PM is not set
+CONFIG_SUSPEND_UP_POSSIBLE=y
 # CONFIG_PM_WAKEUP_BY_GPIO is not set
 
+#
+# CPU Frequency scaling
+#
+# CONFIG_CPU_FREQ is not set
+
 #
 # Networking
 #
@@ -405,6 +405,7 @@ CONFIG_SYN_COOKIES=y
 CONFIG_INET_XFRM_MODE_TRANSPORT=y
 CONFIG_INET_XFRM_MODE_TUNNEL=y
 CONFIG_INET_XFRM_MODE_BEET=y
+# CONFIG_INET_LRO is not set
 CONFIG_INET_DIAG=y
 CONFIG_INET_TCP_DIAG=y
 # CONFIG_TCP_CONG_ADVANCED is not set
@@ -431,10 +432,6 @@ CONFIG_DEFAULT_TCP_CONG="cubic"
 # CONFIG_LAPB is not set
 # CONFIG_ECONET is not set
 # CONFIG_WAN_ROUTER is not set
-
-#
-# QoS and/or fair queueing
-#
 # CONFIG_NET_SCHED is not set
 
 #
@@ -466,6 +463,7 @@ CONFIG_IRDA_CACHE_LAST_LSAP=y
 # SIR device drivers
 #
 CONFIG_IRTTY_SIR=m
+# CONFIG_BFIN_SIR is not set
 
 #
 # Dongle support
@@ -495,6 +493,7 @@ CONFIG_IRTTY_SIR=m
 # CONFIG_MAC80211 is not set
 # CONFIG_IEEE80211 is not set
 # CONFIG_RFKILL is not set
+# CONFIG_NET_9P is not set
 
 #
 # Device Drivers
@@ -503,14 +502,11 @@ CONFIG_IRTTY_SIR=m
 #
 # Generic Driver Options
 #
+CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
 CONFIG_STANDALONE=y
 CONFIG_PREVENT_FIRMWARE_BUILD=y
 # CONFIG_FW_LOADER is not set
 # CONFIG_SYS_HYPERVISOR is not set
-
-#
-# Connector - unified userspace <-> kernelspace linker
-#
 # CONFIG_CONNECTOR is not set
 CONFIG_MTD=y
 # CONFIG_MTD_DEBUG is not set
@@ -530,6 +526,7 @@ CONFIG_MTD_BLOCK=y
 # CONFIG_INFTL is not set
 # CONFIG_RFD_FTL is not set
 # CONFIG_SSFDC is not set
+# CONFIG_MTD_OOPS is not set
 
 #
 # RAM/ROM/Flash chip drivers
@@ -590,20 +587,8 @@ CONFIG_MTD_PHYSMAP_BANKWIDTH=2
 # UBI - Unsorted block images
 #
 # CONFIG_MTD_UBI is not set
-
-#
-# Parallel port support
-#
 # CONFIG_PARPORT is not set
-
-#
-# Plug and Play support
-#
-# CONFIG_PNPACPI is not set
-
-#
-# Block devices
-#
+CONFIG_BLK_DEV=y
 # CONFIG_BLK_DEV_COW_COMMON is not set
 # CONFIG_BLK_DEV_LOOP is not set
 # CONFIG_BLK_DEV_NBD is not set
@@ -613,10 +598,8 @@ CONFIG_BLK_DEV_RAM_SIZE=4096
 CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024
 # CONFIG_CDROM_PKTCDVD is not set
 # CONFIG_ATA_OVER_ETH is not set
-
-#
-# Misc devices
-#
+CONFIG_MISC_DEVICES=y
+# CONFIG_EEPROM_93CX6 is not set
 # CONFIG_IDE is not set
 
 #
@@ -624,32 +607,29 @@ CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024
 #
 # CONFIG_RAID_ATTRS is not set
 # CONFIG_SCSI is not set
+# CONFIG_SCSI_DMA is not set
 # CONFIG_SCSI_NETLINK is not set
 # CONFIG_ATA is not set
-
-#
-# Multi-device support (RAID and LVM)
-#
 # CONFIG_MD is not set
-
-#
-# Network device support
-#
 CONFIG_NETDEVICES=y
+# CONFIG_NETDEVICES_MULTIQUEUE is not set
 # CONFIG_DUMMY is not set
 # CONFIG_BONDING is not set
+# CONFIG_MACVLAN is not set
 # CONFIG_EQUALIZER is not set
 # CONFIG_TUN is not set
+# CONFIG_VETH is not set
 # CONFIG_PHYLIB is not set
-
-#
-# Ethernet (10 or 100Mbit)
-#
 CONFIG_NET_ETHERNET=y
 CONFIG_MII=y
 CONFIG_SMC91X=y
 # CONFIG_SMSC911X is not set
 # CONFIG_DM9000 is not set
+# CONFIG_IBM_NEW_EMAC_ZMII is not set
+# CONFIG_IBM_NEW_EMAC_RGMII is not set
+# CONFIG_IBM_NEW_EMAC_TAH is not set
+# CONFIG_IBM_NEW_EMAC_EMAC4 is not set
+# CONFIG_B44 is not set
 CONFIG_NETDEV_1000=y
 # CONFIG_AX88180 is not set
 CONFIG_NETDEV_10000=y
@@ -666,15 +646,7 @@ CONFIG_NETDEV_10000=y
 # CONFIG_NETCONSOLE is not set
 # CONFIG_NETPOLL is not set
 # CONFIG_NET_POLL_CONTROLLER is not set
-
-#
-# ISDN subsystem
-#
 # CONFIG_ISDN is not set
-
-#
-# Telephony Support
-#
 # CONFIG_PHONE is not set
 
 #
@@ -689,7 +661,6 @@ CONFIG_INPUT=m
 #
 # CONFIG_INPUT_MOUSEDEV is not set
 # CONFIG_INPUT_JOYDEV is not set
-# CONFIG_INPUT_TSDEV is not set
 CONFIG_INPUT_EVDEV=m
 # CONFIG_INPUT_EVBUG is not set
 
@@ -714,13 +685,12 @@ CONFIG_INPUT_EVDEV=m
 #
 # CONFIG_AD9960 is not set
 # CONFIG_SPI_ADC_BF533 is not set
-# CONFIG_BF5xx_PFLAGS is not set
 # CONFIG_BF5xx_PPIFCD is not set
 # CONFIG_BFIN_SIMPLE_TIMER is not set
 # CONFIG_BF5xx_PPI is not set
 # CONFIG_BFIN_SPORT is not set
 # CONFIG_BFIN_TIMER_LATENCY is not set
-# CONFIG_AD5304 is not set
+CONFIG_SIMPLE_GPIO=m
 # CONFIG_VT is not set
 # CONFIG_SERIAL_NONSTANDARD is not set
 
@@ -748,27 +718,11 @@ CONFIG_UNIX98_PTYS=y
 # CAN, the car bus and industrial fieldbus
 #
 # CONFIG_CAN4LINUX is not set
-
-#
-# IPMI
-#
 # CONFIG_IPMI_HANDLER is not set
-CONFIG_WATCHDOG=y
-# CONFIG_WATCHDOG_NOWAYOUT is not set
-
-#
-# Watchdog Device Drivers
-#
-# CONFIG_SOFT_WATCHDOG is not set
-CONFIG_BFIN_WDT=y
 CONFIG_HW_RANDOM=y
 # CONFIG_GEN_RTC is not set
 # CONFIG_R3964 is not set
 # CONFIG_RAW_DRIVER is not set
-
-#
-# TPM devices
-#
 # CONFIG_TCG_TPM is not set
 # CONFIG_I2C is not set
 
@@ -789,22 +743,37 @@ CONFIG_SPI_BFIN=y
 #
 # CONFIG_SPI_AT25 is not set
 # CONFIG_SPI_SPIDEV is not set
-
-#
-# Dallas's 1-wire bus
-#
+# CONFIG_SPI_TLE62X0 is not set
 # CONFIG_W1 is not set
+# CONFIG_POWER_SUPPLY is not set
 CONFIG_HWMON=y
 # CONFIG_HWMON_VID is not set
-# CONFIG_SENSORS_ABITUGURU is not set
 # CONFIG_SENSORS_F71805F is not set
+# CONFIG_SENSORS_F71882FG is not set
+# CONFIG_SENSORS_IT87 is not set
 # CONFIG_SENSORS_LM70 is not set
+# CONFIG_SENSORS_PC87360 is not set
 # CONFIG_SENSORS_PC87427 is not set
 # CONFIG_SENSORS_SMSC47M1 is not set
 # CONFIG_SENSORS_SMSC47B397 is not set
 # CONFIG_SENSORS_VT1211 is not set
 # CONFIG_SENSORS_W83627HF is not set
+# CONFIG_SENSORS_W83627EHF is not set
 # CONFIG_HWMON_DEBUG_CHIP is not set
+CONFIG_WATCHDOG=y
+# CONFIG_WATCHDOG_NOWAYOUT is not set
+
+#
+# Watchdog Device Drivers
+#
+# CONFIG_SOFT_WATCHDOG is not set
+CONFIG_BFIN_WDT=y
+
+#
+# Sonics Silicon Backplane
+#
+CONFIG_SSB_POSSIBLE=y
+# CONFIG_SSB is not set
 
 #
 # Multifunction device drivers
@@ -821,91 +790,33 @@ CONFIG_DAB=y
 #
 # Graphics support
 #
+# CONFIG_VGASTATE is not set
+# CONFIG_VIDEO_OUTPUT_CONTROL is not set
+# CONFIG_FB is not set
 # CONFIG_BACKLIGHT_LCD_SUPPORT is not set
 
 #
 # Display device support
 #
 # CONFIG_DISPLAY_SUPPORT is not set
-# CONFIG_VGASTATE is not set
-# CONFIG_FB is not set
 
 #
 # Sound
 #
 # CONFIG_SOUND is not set
-
-#
-# HID Devices
-#
+CONFIG_HID_SUPPORT=y
 CONFIG_HID=m
 # CONFIG_HID_DEBUG is not set
-
-#
-# USB support
-#
-CONFIG_USB_ARCH_HAS_HCD=y
-# CONFIG_USB_ARCH_HAS_OHCI is not set
-# CONFIG_USB_ARCH_HAS_EHCI is not set
-# CONFIG_USB is not set
-
-#
-# Enable Host or Gadget support to see Inventra options
-#
-
-#
-# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support'
-#
-
-#
-# USB Gadget Support
-#
-# CONFIG_USB_GADGET is not set
+# CONFIG_HIDRAW is not set
+# CONFIG_USB_SUPPORT is not set
 # CONFIG_MMC is not set
-
-#
-# LED devices
-#
 # CONFIG_NEW_LEDS is not set
-
-#
-# LED drivers
-#
-
-#
-# LED Triggers
-#
-
-#
-# InfiniBand support
-#
-
-#
-# EDAC - error detection and reporting (RAS) (EXPERIMENTAL)
-#
-
-#
-# Real Time Clock
-#
 # CONFIG_RTC_CLASS is not set
 
 #
-# DMA Engine support
+# Userspace I/O
 #
-# CONFIG_DMA_ENGINE is not set
-
-#
-# DMA Clients
-#
-
-#
-# DMA Devices
-#
-
-#
-# PBX support
-#
-# CONFIG_PBX is not set
+# CONFIG_UIO is not set
 
 #
 # File systems
@@ -950,7 +861,6 @@ CONFIG_PROC_SYSCTL=y
 CONFIG_SYSFS=y
 # CONFIG_TMPFS is not set
 # CONFIG_HUGETLB_PAGE is not set
-CONFIG_RAMFS=y
 # CONFIG_CONFIGFS_FS is not set
 
 #
@@ -976,10 +886,12 @@ CONFIG_YAFFS_SHORT_NAMES_IN_RAM=y
 CONFIG_JFFS2_FS=m
 CONFIG_JFFS2_FS_DEBUG=0
 CONFIG_JFFS2_FS_WRITEBUFFER=y
+# CONFIG_JFFS2_FS_WBUF_VERIFY is not set
 # CONFIG_JFFS2_SUMMARY is not set
 # CONFIG_JFFS2_FS_XATTR is not set
 # CONFIG_JFFS2_COMPRESSION_OPTIONS is not set
 CONFIG_JFFS2_ZLIB=y
+# CONFIG_JFFS2_LZO is not set
 CONFIG_JFFS2_RTIME=y
 # CONFIG_JFFS2_RUBIN is not set
 # CONFIG_CRAMFS is not set
@@ -988,10 +900,7 @@ CONFIG_JFFS2_RTIME=y
 # CONFIG_QNX4FS_FS is not set
 # CONFIG_SYSV_FS is not set
 # CONFIG_UFS_FS is not set
-
-#
-# Network File Systems
-#
+CONFIG_NETWORK_FILESYSTEMS=y
 CONFIG_NFS_FS=m
 CONFIG_NFS_V3=y
 # CONFIG_NFS_V3_ACL is not set
@@ -1011,17 +920,12 @@ CONFIG_SMB_FS=m
 # CONFIG_NCP_FS is not set
 # CONFIG_CODA_FS is not set
 # CONFIG_AFS_FS is not set
-# CONFIG_9P_FS is not set
 
 #
 # Partition Types
 #
 # CONFIG_PARTITION_ADVANCED is not set
 CONFIG_MSDOS_PARTITION=y
-
-#
-# Native Language Support
-#
 CONFIG_NLS=m
 CONFIG_NLS_DEFAULT="iso8859-1"
 # CONFIG_NLS_CODEPAGE_437 is not set
@@ -1062,21 +966,16 @@ CONFIG_NLS_DEFAULT="iso8859-1"
 # CONFIG_NLS_KOI8_R is not set
 # CONFIG_NLS_KOI8_U is not set
 # CONFIG_NLS_UTF8 is not set
-
-#
-# Distributed Lock Manager
-#
 # CONFIG_DLM is not set
-
-#
-# Profiling support
-#
+CONFIG_INSTRUMENTATION=y
 # CONFIG_PROFILING is not set
+# CONFIG_MARKERS is not set
 
 #
 # Kernel hacking
 #
 # CONFIG_PRINTK_TIME is not set
+CONFIG_ENABLE_WARN_DEPRECATED=y
 CONFIG_ENABLE_MUST_CHECK=y
 # CONFIG_MAGIC_SYSRQ is not set
 # CONFIG_UNUSED_SYMBOLS is not set
@@ -1084,6 +983,7 @@ CONFIG_DEBUG_FS=y
 # CONFIG_HEADERS_CHECK is not set
 # CONFIG_DEBUG_KERNEL is not set
 # CONFIG_DEBUG_BUGVERBOSE is not set
+# CONFIG_SAMPLES is not set
 CONFIG_DEBUG_MMRS=y
 CONFIG_DEBUG_HUNT_FOR_ZERO=y
 CONFIG_DEBUG_BFIN_HWTRACE_ON=y
@@ -1104,11 +1004,7 @@ CONFIG_ACCESS_CHECK=y
 # CONFIG_KEYS is not set
 CONFIG_SECURITY=y
 # CONFIG_SECURITY_NETWORK is not set
-CONFIG_SECURITY_CAPABILITIES=m
-
-#
-# Cryptographic options
-#
+# CONFIG_SECURITY_CAPABILITIES is not set
 # CONFIG_CRYPTO is not set
 
 #
@@ -1119,6 +1015,7 @@ CONFIG_CRC_CCITT=m
 # CONFIG_CRC16 is not set
 # CONFIG_CRC_ITU_T is not set
 CONFIG_CRC32=y
+# CONFIG_CRC7 is not set
 # CONFIG_LIBCRC32C is not set
 CONFIG_ZLIB_INFLATE=y
 CONFIG_ZLIB_DEFLATE=m
diff --git a/arch/blackfin/configs/CM-BF533_defconfig b/arch/blackfin/configs/CM-BF533_defconfig
index 560890fe0d30..09deea44480b 100644
--- a/arch/blackfin/configs/CM-BF533_defconfig
+++ b/arch/blackfin/configs/CM-BF533_defconfig
@@ -39,7 +39,8 @@ CONFIG_SYSVIPC_SYSCTL=y
 # CONFIG_TASKSTATS is not set
 # CONFIG_UTS_NS is not set
 # CONFIG_AUDIT is not set
-# CONFIG_IKCONFIG is not set
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
 CONFIG_LOG_BUF_SHIFT=14
 CONFIG_SYSFS_DEPRECATED=y
 # CONFIG_RELAY is not set
@@ -291,7 +292,7 @@ CONFIG_C_AMBEN_ALL=y
 CONFIG_BANK_0=0x7BB0
 CONFIG_BANK_1=0x7BB0
 CONFIG_BANK_2=0x7BB0
-CONFIG_BANK_3=0xFFC3
+CONFIG_BANK_3=0xFFC2
 
 #
 # Bus options (PCI, PCMCIA, EISA, MCA, ISA)
@@ -617,8 +618,7 @@ CONFIG_SERIAL_CORE=y
 CONFIG_SERIAL_CORE_CONSOLE=y
 # CONFIG_SERIAL_BFIN_SPORT is not set
 CONFIG_UNIX98_PTYS=y
-CONFIG_LEGACY_PTYS=y
-CONFIG_LEGACY_PTY_COUNT=256
+# CONFIG_LEGACY_PTYS is not set
 
 #
 # CAN, the car bus and industrial fieldbus
@@ -778,7 +778,7 @@ CONFIG_FS_MBCACHE=y
 CONFIG_INOTIFY=y
 CONFIG_INOTIFY_USER=y
 # CONFIG_QUOTA is not set
-CONFIG_DNOTIFY=y
+# CONFIG_DNOTIFY is not set
 # CONFIG_AUTOFS_FS is not set
 # CONFIG_AUTOFS4_FS is not set
 # CONFIG_FUSE_FS is not set
@@ -866,11 +866,11 @@ CONFIG_MSDOS_PARTITION=y
 CONFIG_ENABLE_MUST_CHECK=y
 # CONFIG_MAGIC_SYSRQ is not set
 # CONFIG_UNUSED_SYMBOLS is not set
-# CONFIG_DEBUG_FS is not set
+CONFIG_DEBUG_FS=y
 # CONFIG_HEADERS_CHECK is not set
 # CONFIG_DEBUG_KERNEL is not set
 # CONFIG_DEBUG_BUGVERBOSE is not set
-# CONFIG_DEBUG_MMRS is not set
+CONFIG_DEBUG_MMRS=y
 CONFIG_DEBUG_HUNT_FOR_ZERO=y
 CONFIG_DEBUG_BFIN_HWTRACE_ON=y
 CONFIG_DEBUG_BFIN_HWTRACE_COMPRESSION_OFF=y
diff --git a/arch/blackfin/configs/CM-BF537E_defconfig b/arch/blackfin/configs/CM-BF537E_defconfig
index 9f66d2de1007..219fc345a5f5 100644
--- a/arch/blackfin/configs/CM-BF537E_defconfig
+++ b/arch/blackfin/configs/CM-BF537E_defconfig
@@ -39,7 +39,8 @@ CONFIG_SYSVIPC_SYSCTL=y
 # CONFIG_TASKSTATS is not set
 # CONFIG_UTS_NS is not set
 # CONFIG_AUDIT is not set
-# CONFIG_IKCONFIG is not set
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
 CONFIG_LOG_BUF_SHIFT=14
 CONFIG_SYSFS_DEPRECATED=y
 # CONFIG_RELAY is not set
@@ -299,7 +300,7 @@ CONFIG_C_AMBEN_ALL=y
 CONFIG_BANK_0=0x7BB0
 CONFIG_BANK_1=0x7BB0
 CONFIG_BANK_2=0x7BB0
-CONFIG_BANK_3=0xFFC3
+CONFIG_BANK_3=0xFFC2
 
 #
 # Bus options (PCI, PCMCIA, EISA, MCA, ISA)
@@ -351,7 +352,10 @@ CONFIG_INET=y
 # CONFIG_IP_MULTICAST is not set
 # CONFIG_IP_ADVANCED_ROUTER is not set
 CONFIG_IP_FIB_HASH=y
-# CONFIG_IP_PNP is not set
+CONFIG_IP_PNP=y
+# CONFIG_IP_PNP_DHCP is not set
+# CONFIG_IP_PNP_BOOTP is not set
+# CONFIG_IP_PNP_RARP is not set
 # CONFIG_NET_IPIP is not set
 # CONFIG_NET_IPGRE is not set
 # CONFIG_ARPD is not set
@@ -645,8 +649,7 @@ CONFIG_SERIAL_CORE=y
 CONFIG_SERIAL_CORE_CONSOLE=y
 # CONFIG_SERIAL_BFIN_SPORT is not set
 CONFIG_UNIX98_PTYS=y
-CONFIG_LEGACY_PTYS=y
-CONFIG_LEGACY_PTY_COUNT=256
+# CONFIG_LEGACY_PTYS is not set
 
 #
 # CAN, the car bus and industrial fieldbus
@@ -806,7 +809,7 @@ CONFIG_FS_MBCACHE=y
 CONFIG_INOTIFY=y
 CONFIG_INOTIFY_USER=y
 # CONFIG_QUOTA is not set
-CONFIG_DNOTIFY=y
+# CONFIG_DNOTIFY is not set
 # CONFIG_AUTOFS_FS is not set
 # CONFIG_AUTOFS4_FS is not set
 # CONFIG_FUSE_FS is not set
@@ -894,12 +897,12 @@ CONFIG_MSDOS_PARTITION=y
 CONFIG_ENABLE_MUST_CHECK=y
 # CONFIG_MAGIC_SYSRQ is not set
 # CONFIG_UNUSED_SYMBOLS is not set
-# CONFIG_DEBUG_FS is not set
+CONFIG_DEBUG_FS=y
 # CONFIG_HEADERS_CHECK is not set
 # CONFIG_DEBUG_KERNEL is not set
 # CONFIG_DEBUG_BUGVERBOSE is not set
-# CONFIG_DEBUG_MMRS is not set
-# CONFIG_DEBUG_HUNT_FOR_ZERO is not set
+CONFIG_DEBUG_MMRS=y
+CONFIG_DEBUG_HUNT_FOR_ZERO=y
 CONFIG_DEBUG_BFIN_HWTRACE_ON=y
 CONFIG_DEBUG_BFIN_HWTRACE_COMPRESSION_OFF=y
 # CONFIG_DEBUG_BFIN_HWTRACE_COMPRESSION_ONE is not set
diff --git a/arch/blackfin/configs/CM-BF537U_defconfig b/arch/blackfin/configs/CM-BF537U_defconfig
index 2694d06c5bde..9873d586fc77 100644
--- a/arch/blackfin/configs/CM-BF537U_defconfig
+++ b/arch/blackfin/configs/CM-BF537U_defconfig
@@ -39,7 +39,8 @@ CONFIG_SYSVIPC_SYSCTL=y
 # CONFIG_TASKSTATS is not set
 # CONFIG_UTS_NS is not set
 # CONFIG_AUDIT is not set
-# CONFIG_IKCONFIG is not set
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
 CONFIG_LOG_BUF_SHIFT=14
 CONFIG_SYSFS_DEPRECATED=y
 # CONFIG_RELAY is not set
@@ -298,8 +299,8 @@ CONFIG_C_AMBEN_ALL=y
 #
 CONFIG_BANK_0=0x7BB0
 CONFIG_BANK_1=0x7BB0
-CONFIG_BANK_2=0xFFC3
-CONFIG_BANK_3=0xFFC3
+CONFIG_BANK_2=0xFFC2
+CONFIG_BANK_3=0xFFC2
 
 #
 # Bus options (PCI, PCMCIA, EISA, MCA, ISA)
@@ -628,8 +629,7 @@ CONFIG_SERIAL_CORE=y
 CONFIG_SERIAL_CORE_CONSOLE=y
 # CONFIG_SERIAL_BFIN_SPORT is not set
 CONFIG_UNIX98_PTYS=y
-CONFIG_LEGACY_PTYS=y
-CONFIG_LEGACY_PTY_COUNT=256
+# CONFIG_LEGACY_PTYS is not set
 
 #
 # CAN, the car bus and industrial fieldbus
@@ -806,7 +806,7 @@ CONFIG_FS_MBCACHE=y
 CONFIG_INOTIFY=y
 CONFIG_INOTIFY_USER=y
 # CONFIG_QUOTA is not set
-CONFIG_DNOTIFY=y
+# CONFIG_DNOTIFY is not set
 # CONFIG_AUTOFS_FS is not set
 # CONFIG_AUTOFS4_FS is not set
 # CONFIG_FUSE_FS is not set
@@ -894,12 +894,12 @@ CONFIG_MSDOS_PARTITION=y
 CONFIG_ENABLE_MUST_CHECK=y
 # CONFIG_MAGIC_SYSRQ is not set
 # CONFIG_UNUSED_SYMBOLS is not set
-# CONFIG_DEBUG_FS is not set
+CONFIG_DEBUG_FS=y
 # CONFIG_HEADERS_CHECK is not set
 # CONFIG_DEBUG_KERNEL is not set
 # CONFIG_DEBUG_BUGVERBOSE is not set
-# CONFIG_DEBUG_MMRS is not set
-# CONFIG_DEBUG_HUNT_FOR_ZERO is not set
+CONFIG_DEBUG_MMRS=y
+CONFIG_DEBUG_HUNT_FOR_ZERO=y
 CONFIG_DEBUG_BFIN_HWTRACE_ON=y
 CONFIG_DEBUG_BFIN_HWTRACE_COMPRESSION_OFF=y
 # CONFIG_DEBUG_BFIN_HWTRACE_COMPRESSION_ONE is not set
diff --git a/arch/blackfin/configs/CM-BF548_defconfig b/arch/blackfin/configs/CM-BF548_defconfig
index 90207251c533..0e3605fdb7b0 100644
--- a/arch/blackfin/configs/CM-BF548_defconfig
+++ b/arch/blackfin/configs/CM-BF548_defconfig
@@ -363,7 +363,7 @@ CONFIG_C_AMBEN_ALL=y
 CONFIG_BANK_0=0x7BB0
 CONFIG_BANK_1=0x5554
 CONFIG_BANK_2=0x7BB0
-CONFIG_BANK_3=0x99B3
+CONFIG_BANK_3=0x99B2
 CONFIG_EBIU_MBSCTLVAL=0x0
 CONFIG_EBIU_MODEVAL=0x1
 CONFIG_EBIU_FCTLVAL=0x6
@@ -744,8 +744,8 @@ CONFIG_BFIN_OTP=y
 #
 CONFIG_SERIAL_BFIN=y
 CONFIG_SERIAL_BFIN_CONSOLE=y
-# CONFIG_SERIAL_BFIN_DMA is not set
-CONFIG_SERIAL_BFIN_PIO=y
+CONFIG_SERIAL_BFIN_DMA=y
+# CONFIG_SERIAL_BFIN_PIO is not set
 # CONFIG_SERIAL_BFIN_UART0 is not set
 CONFIG_SERIAL_BFIN_UART1=y
 # CONFIG_BFIN_UART1_CTSRTS is not set
@@ -1149,7 +1149,7 @@ CONFIG_RTC_DRV_BFIN=y
 CONFIG_INOTIFY=y
 CONFIG_INOTIFY_USER=y
 # CONFIG_QUOTA is not set
-CONFIG_DNOTIFY=y
+# CONFIG_DNOTIFY is not set
 # CONFIG_AUTOFS_FS is not set
 # CONFIG_AUTOFS4_FS is not set
 # CONFIG_FUSE_FS is not set
@@ -1332,7 +1332,7 @@ CONFIG_DEBUG_FS=y
 # CONFIG_DEBUG_KERNEL is not set
 # CONFIG_DEBUG_BUGVERBOSE is not set
 # CONFIG_SAMPLES is not set
-# CONFIG_DEBUG_MMRS is not set
+CONFIG_DEBUG_MMRS=y
 CONFIG_DEBUG_HUNT_FOR_ZERO=y
 CONFIG_DEBUG_BFIN_HWTRACE_ON=y
 CONFIG_DEBUG_BFIN_HWTRACE_COMPRESSION_OFF=y
diff --git a/arch/blackfin/configs/CM-BF561_defconfig b/arch/blackfin/configs/CM-BF561_defconfig
index daf00906c1ef..59c7cdbee904 100644
--- a/arch/blackfin/configs/CM-BF561_defconfig
+++ b/arch/blackfin/configs/CM-BF561_defconfig
@@ -35,7 +35,8 @@ CONFIG_SYSVIPC_SYSCTL=y
 # CONFIG_USER_NS is not set
 # CONFIG_PID_NS is not set
 # CONFIG_AUDIT is not set
-# CONFIG_IKCONFIG is not set
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
 CONFIG_LOG_BUF_SHIFT=14
 # CONFIG_CGROUPS is not set
 CONFIG_FAIR_GROUP_SCHED=y
@@ -341,7 +342,7 @@ CONFIG_C_AMBEN_ALL=y
 CONFIG_BANK_0=0x7BB0
 CONFIG_BANK_1=0x7BB0
 CONFIG_BANK_2=0x7BB0
-CONFIG_BANK_3=0xFFC3
+CONFIG_BANK_3=0xFFC2
 
 #
 # Bus options (PCI, PCMCIA, EISA, MCA, ISA)
@@ -631,8 +632,7 @@ CONFIG_SERIAL_CORE=y
 CONFIG_SERIAL_CORE_CONSOLE=y
 # CONFIG_SERIAL_BFIN_SPORT is not set
 CONFIG_UNIX98_PTYS=y
-CONFIG_LEGACY_PTYS=y
-CONFIG_LEGACY_PTY_COUNT=256
+# CONFIG_LEGACY_PTYS is not set
 
 #
 # CAN, the car bus and industrial fieldbus
@@ -756,7 +756,7 @@ CONFIG_FS_MBCACHE=y
 CONFIG_INOTIFY=y
 CONFIG_INOTIFY_USER=y
 # CONFIG_QUOTA is not set
-CONFIG_DNOTIFY=y
+# CONFIG_DNOTIFY is not set
 # CONFIG_AUTOFS_FS is not set
 # CONFIG_AUTOFS4_FS is not set
 # CONFIG_FUSE_FS is not set
@@ -830,12 +830,12 @@ CONFIG_ENABLE_WARN_DEPRECATED=y
 CONFIG_ENABLE_MUST_CHECK=y
 # CONFIG_MAGIC_SYSRQ is not set
 # CONFIG_UNUSED_SYMBOLS is not set
-# CONFIG_DEBUG_FS is not set
+CONFIG_DEBUG_FS=y
 # CONFIG_HEADERS_CHECK is not set
 # CONFIG_DEBUG_KERNEL is not set
 # CONFIG_DEBUG_BUGVERBOSE is not set
 # CONFIG_SAMPLES is not set
-# CONFIG_DEBUG_MMRS is not set
+CONFIG_DEBUG_MMRS=y
 CONFIG_DEBUG_HUNT_FOR_ZERO=y
 CONFIG_DEBUG_BFIN_HWTRACE_ON=y
 CONFIG_DEBUG_BFIN_HWTRACE_COMPRESSION_OFF=y
diff --git a/arch/blackfin/configs/H8606_defconfig b/arch/blackfin/configs/H8606_defconfig
index 679c7483ea71..ba0bee90b7e1 100644
--- a/arch/blackfin/configs/H8606_defconfig
+++ b/arch/blackfin/configs/H8606_defconfig
@@ -967,7 +967,7 @@ CONFIG_FS_MBCACHE=y
 CONFIG_INOTIFY=y
 CONFIG_INOTIFY_USER=y
 # CONFIG_QUOTA is not set
-CONFIG_DNOTIFY=y
+# CONFIG_DNOTIFY is not set
 # CONFIG_AUTOFS_FS is not set
 # CONFIG_AUTOFS4_FS is not set
 # CONFIG_FUSE_FS is not set
diff --git a/arch/blackfin/configs/IP0X_defconfig b/arch/blackfin/configs/IP0X_defconfig
index 4384a670a8b8..285d2241df26 100644
--- a/arch/blackfin/configs/IP0X_defconfig
+++ b/arch/blackfin/configs/IP0X_defconfig
@@ -1066,7 +1066,7 @@ CONFIG_FS_MBCACHE=y
 CONFIG_INOTIFY=y
 CONFIG_INOTIFY_USER=y
 # CONFIG_QUOTA is not set
-CONFIG_DNOTIFY=y
+# CONFIG_DNOTIFY is not set
 # CONFIG_AUTOFS_FS is not set
 # CONFIG_AUTOFS4_FS is not set
 # CONFIG_FUSE_FS is not set
diff --git a/arch/blackfin/configs/PNAV-10_defconfig b/arch/blackfin/configs/PNAV-10_defconfig
index 87622ad9df47..bffca7de65d4 100644
--- a/arch/blackfin/configs/PNAV-10_defconfig
+++ b/arch/blackfin/configs/PNAV-10_defconfig
@@ -294,7 +294,7 @@ CONFIG_C_AMBEN_ALL=y
 CONFIG_BANK_0=0x7BB0
 CONFIG_BANK_1=0x33B0
 CONFIG_BANK_2=0x33B0
-CONFIG_BANK_3=0x99B3
+CONFIG_BANK_3=0x99B2
 
 #
 # Bus options (PCI, PCMCIA, EISA, MCA, ISA)
@@ -1080,7 +1080,7 @@ CONFIG_FS_MBCACHE=y
 CONFIG_INOTIFY=y
 CONFIG_INOTIFY_USER=y
 # CONFIG_QUOTA is not set
-CONFIG_DNOTIFY=y
+# CONFIG_DNOTIFY is not set
 # CONFIG_AUTOFS_FS is not set
 # CONFIG_AUTOFS4_FS is not set
 # CONFIG_FUSE_FS is not set
diff --git a/arch/blackfin/configs/SRV1_defconfig b/arch/blackfin/configs/SRV1_defconfig
index 951ea0412576..b1309f878fcd 100644
--- a/arch/blackfin/configs/SRV1_defconfig
+++ b/arch/blackfin/configs/SRV1_defconfig
@@ -1067,7 +1067,7 @@ CONFIG_FS_MBCACHE=y
 CONFIG_INOTIFY=y
 CONFIG_INOTIFY_USER=y
 # CONFIG_QUOTA is not set
-CONFIG_DNOTIFY=y
+# CONFIG_DNOTIFY is not set
 # CONFIG_AUTOFS_FS is not set
 # CONFIG_AUTOFS4_FS is not set
 # CONFIG_FUSE_FS is not set

From c71b47835188d0c2a1e1f9590564f1b71c651710 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier.adi@gmail.com>
Date: Mon, 14 Jul 2008 17:10:50 +0800
Subject: [PATCH 0054/1171] Blackfin arch: set CROSS_COMPILE in our arch
 Makefile so it gets used by default

Signed-off-by: Mike Frysinger <vapier.adi@gmail.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/Makefile | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/arch/blackfin/Makefile b/arch/blackfin/Makefile
index 3cbe16caad4b..9564731ad3a8 100644
--- a/arch/blackfin/Makefile
+++ b/arch/blackfin/Makefile
@@ -6,8 +6,9 @@
 # for more details.
 #
 
-
-CROSS_COMPILE    ?= bfin-uclinux-
+ifeq ($(CROSS_COMPILE),)
+CROSS_COMPILE    := bfin-uclinux-
+endif
 LDFLAGS_vmlinux  := -X
 OBJCOPYFLAGS     := -O binary -R .note -R .comment -S
 GZFLAGS          := -9

From 1c0d20cd29aec11a3580cedf0bccec25052e8d4c Mon Sep 17 00:00:00 2001
From: Bryan Wu <cooloney@kernel.org>
Date: Tue, 15 Jul 2008 12:08:50 +0800
Subject: [PATCH 0055/1171] Blackfin arch: add TXDWA definition to enable new
 feature

Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 include/asm-blackfin/mach-bf527/anomaly.h  | 2 ++
 include/asm-blackfin/mach-bf527/defBF527.h | 1 +
 include/asm-blackfin/mach-bf537/defBF537.h | 1 +
 3 files changed, 4 insertions(+)

diff --git a/include/asm-blackfin/mach-bf527/anomaly.h b/include/asm-blackfin/mach-bf527/anomaly.h
index 4725268a5ada..b7b166f4f064 100644
--- a/include/asm-blackfin/mach-bf527/anomaly.h
+++ b/include/asm-blackfin/mach-bf527/anomaly.h
@@ -23,6 +23,8 @@
 #define ANOMALY_05000245 (1)
 /* Sensitivity To Noise with Slow Input Edge Rates on External SPORT TX and RX Clocks */
 #define ANOMALY_05000265 (1)
+/* New Feature: EMAC TX DMA Word Alignment */
+#define ANOMALY_05000285 (1)
 /* Errors when SSYNC, CSYNC, or Loads to LT, LB and LC Registers Are Interrupted */
 #define ANOMALY_05000312 (1)
 /* Incorrect Access of OTP_STATUS During otp_write() Function */
diff --git a/include/asm-blackfin/mach-bf527/defBF527.h b/include/asm-blackfin/mach-bf527/defBF527.h
index 82134f578f32..f1a70db70cb8 100644
--- a/include/asm-blackfin/mach-bf527/defBF527.h
+++ b/include/asm-blackfin/mach-bf527/defBF527.h
@@ -302,6 +302,7 @@
 #define	PHYIE             0x00000001    /* PHY_INT Interrupt Enable                               */
 #define	RXDWA             0x00000002    /* Receive Frame DMA Word Alignment (Odd/Even*)           */
 #define	RXCKS             0x00000004    /* Enable RX Frame TCP/UDP Checksum Computation           */
+#define	TXDWA             0x00000010    /* Transmit Frame DMA Word Alignment (Odd/Even*)          */
 #define	MDCDIV            0x00003F00    /* SCLK:MDC Clock Divisor [MDC=SCLK/(2*(N+1))]            */
 
 #define	SET_MDCDIV(x) (((x)&0x3F)<< 8)   /* Set MDC Clock Divisor                                 */
diff --git a/include/asm-blackfin/mach-bf537/defBF537.h b/include/asm-blackfin/mach-bf537/defBF537.h
index 3f455909c418..abde24c6d3b1 100644
--- a/include/asm-blackfin/mach-bf537/defBF537.h
+++ b/include/asm-blackfin/mach-bf537/defBF537.h
@@ -290,6 +290,7 @@
 #define	PHYIE		0x00000001	/* PHY_INT Interrupt Enable                                                     */
 #define	RXDWA		0x00000002	/* Receive Frame DMA Word Alignment (Odd/Even*)         */
 #define	RXCKS		0x00000004	/* Enable RX Frame TCP/UDP Checksum Computation         */
+#define	TXDWA		0x00000010	/* Transmit Frame DMA Word Alignment (Odd/Even*)        */
 #define	MDCDIV		0x00003F00	/* SCLK:MDC Clock Divisor [MDC=SCLK/(2*(N+1))]          */
 
 #define	SET_MDCDIV(x)	(((x)&0x3F)<< 8)	/* Set MDC Clock Divisor                                */

From 4f13f548cef5af1717cbbc341a1a3474f3e7466e Mon Sep 17 00:00:00 2001
From: Michael Hennerich <michael.hennerich@analog.com>
Date: Tue, 15 Jul 2008 16:38:28 +0800
Subject: [PATCH 0056/1171] Blackfin arch: Fix BUG - JUMP error in kernel
 (relocation truncated to fit: R_pcrel12_jump_s)

Use long jump

Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/kernel/cplb-nompu/cplbhdlr.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/blackfin/kernel/cplb-nompu/cplbhdlr.S b/arch/blackfin/kernel/cplb-nompu/cplbhdlr.S
index 2788532de72b..ecbabc0a1fed 100644
--- a/arch/blackfin/kernel/cplb-nompu/cplbhdlr.S
+++ b/arch/blackfin/kernel/cplb-nompu/cplbhdlr.S
@@ -125,6 +125,6 @@ ENTRY(__cplb_hdr)
 	SP += -12;
 	call _panic_cplb_error;
 	SP += 12;
-	JUMP _handle_bad_cplb;
+	JUMP.L _handle_bad_cplb;
 
 ENDPROC(__cplb_hdr)

From 1efc80b53eb54770139219f99657abd92595fc86 Mon Sep 17 00:00:00 2001
From: Michael Hennerich <michael.hennerich@analog.com>
Date: Sat, 19 Jul 2008 16:57:32 +0800
Subject: [PATCH 0057/1171] Blackfin arch: Functional power management support

Enable: PM_SUSPEND_MEM -> Blackfin Hibernate to SDRAM
This feature requires a special bootloader (u-boot)
supporting return from hibernate.

Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/Kconfig                  |  66 ++-
 arch/blackfin/kernel/bfin_dma_5xx.c    |  26 ++
 arch/blackfin/kernel/bfin_gpio.c       | 118 ++++-
 arch/blackfin/mach-common/dpmc_modes.S | 606 +++++++++++++++++++++++--
 arch/blackfin/mach-common/pm.c         | 225 ++++++++-
 include/asm-blackfin/dma.h             |   8 +
 include/asm-blackfin/dpmc.h            |  82 ++--
 include/asm-blackfin/gpio.h            |  19 +-
 include/asm-blackfin/mach-bf548/gpio.h |   8 +
 9 files changed, 1036 insertions(+), 122 deletions(-)

diff --git a/arch/blackfin/Kconfig b/arch/blackfin/Kconfig
index c602727d1a9a..8f21b78b2e6f 100644
--- a/arch/blackfin/Kconfig
+++ b/arch/blackfin/Kconfig
@@ -880,7 +880,7 @@ config ARCH_SUSPEND_POSSIBLE
 	depends on !SMP
 
 choice
-	prompt "Default Power Saving Mode"
+	prompt "Standby Power Saving Mode"
 	depends on PM
 	default PM_BFIN_SLEEP_DEEPER
 config  PM_BFIN_SLEEP_DEEPER
@@ -899,6 +899,8 @@ config  PM_BFIN_SLEEP_DEEPER
 	  normal during Sleep Deeper, due to the reduced SCLK frequency.
 	  When in the sleep mode, system DMA access to L1 memory is not supported.
 
+	  If unsure, select "Sleep Deeper".
+
 config  PM_BFIN_SLEEP
 	bool "Sleep"
 	help
@@ -906,15 +908,17 @@ config  PM_BFIN_SLEEP
 	  dissipation by disabling the clock to the processor core (CCLK).
 	  The PLL and system clock (SCLK), however, continue to operate in
 	  this mode. Typically an external event or RTC activity will wake
-	  up the processor. When in the sleep mode,
-	  system DMA access to L1 memory is not supported.
+	  up the processor. When in the sleep mode, system DMA access to L1
+	  memory is not supported.
+
+	  If unsure, select "Sleep Deeper".
 endchoice
 
 config PM_WAKEUP_BY_GPIO
-	bool "Cause Wakeup Event by GPIO"
+	bool "Allow Wakeup from Standby by GPIO"
 
 config PM_WAKEUP_GPIO_NUMBER
-	int "Wakeup GPIO number"
+	int "GPIO number"
 	range 0 47
 	depends on PM_WAKEUP_BY_GPIO
 	default 2 if BFIN537_STAMP
@@ -935,6 +939,58 @@ config  PM_WAKEUP_GPIO_POLAR_EDGE_B
 	bool "Both EDGE"
 endchoice
 
+comment "Possible Suspend Mem / Hibernate Wake-Up Sources"
+	depends on PM
+
+config PM_BFIN_WAKE_RTC
+	bool "Allow Wake-Up from RESET and on-chip RTC"
+	depends on PM
+	default n
+	help
+	  Enable RTC Wake-Up (Voltage Regulator Power-Up)
+
+config PM_BFIN_WAKE_PH6
+	bool "Allow Wake-Up from on-chip PHY or PH6 GP"
+	depends on PM && (BF52x || BF534 || BF536 || BF537)
+	default n
+	help
+	  Enable PHY and PH6 GP Wake-Up (Voltage Regulator Power-Up)
+
+config PM_BFIN_WAKE_CAN
+	bool "Allow Wake-Up from on-chip CAN0/1"
+	depends on PM && (BF54x || BF534 || BF536 || BF537)
+	default n
+	help
+	  Enable CAN0/1 Wake-Up (Voltage Regulator Power-Up)
+
+config PM_BFIN_WAKE_GP
+	bool "Allow Wake-Up from GPIOs"
+	depends on PM && BF54x
+	default n
+	help
+	  Enable General-Purpose Wake-Up (Voltage Regulator Power-Up)
+
+config PM_BFIN_WAKE_USB
+	bool "Allow Wake-Up from on-chip USB"
+	depends on PM && (BF54x || BF52x)
+	default n
+	help
+	  Enable USB Wake-Up (Voltage Regulator Power-Up)
+
+config PM_BFIN_WAKE_KEYPAD
+	bool "Allow Wake-Up from on-chip Keypad"
+	depends on PM && BF54x
+	default n
+	help
+	  Enable Keypad Wake-Up (Voltage Regulator Power-Up)
+
+config PM_BFIN_WAKE_ROTARY
+	bool "Allow Wake-Up from on-chip Rotary"
+	depends on PM && BF54x
+	default n
+	help
+	  Enable Rotary Wake-Up (Voltage Regulator Power-Up)
+
 endmenu
 
 menu "CPU Frequency scaling"
diff --git a/arch/blackfin/kernel/bfin_dma_5xx.c b/arch/blackfin/kernel/bfin_dma_5xx.c
index d54f19085f37..ad0e75845ac2 100644
--- a/arch/blackfin/kernel/bfin_dma_5xx.c
+++ b/arch/blackfin/kernel/bfin_dma_5xx.c
@@ -472,6 +472,32 @@ unsigned long get_dma_curr_addr(unsigned int channel)
 }
 EXPORT_SYMBOL(get_dma_curr_addr);
 
+#ifdef CONFIG_PM
+int blackfin_dma_suspend(void)
+{
+	int i;
+
+	for (i = 0; i < MAX_BLACKFIN_DMA_CHANNEL; i++) {
+		if (dma_ch[i].chan_status == DMA_CHANNEL_ENABLED) {
+			printk(KERN_ERR "DMA Channel %d failed to suspend\n", i);
+			return -EBUSY;
+		}
+
+		dma_ch[i].saved_peripheral_map = dma_ch[i].regs->peripheral_map;
+	}
+
+	return 0;
+}
+
+void blackfin_dma_resume(void)
+{
+	int i;
+
+	for (i = 0; i < MAX_BLACKFIN_DMA_CHANNEL; i++)
+		dma_ch[i].regs->peripheral_map = dma_ch[i].saved_peripheral_map;
+}
+#endif
+
 static void *__dma_memcpy(void *dest, const void *src, size_t size)
 {
 	int direction;	/* 1 - address decrease, 0 - address increase */
diff --git a/arch/blackfin/kernel/bfin_gpio.c b/arch/blackfin/kernel/bfin_gpio.c
index b6d89d1644cc..ecbd141e0ef2 100644
--- a/arch/blackfin/kernel/bfin_gpio.c
+++ b/arch/blackfin/kernel/bfin_gpio.c
@@ -186,7 +186,10 @@ static struct str_ident {
 	char name[RESOURCE_LABEL_SIZE];
 } str_ident[MAX_RESOURCES];
 
-#if defined(CONFIG_PM) && !defined(CONFIG_BF54x)
+#if defined(CONFIG_PM)
+#if defined(CONFIG_BF54x)
+static struct gpio_port_s gpio_bank_saved[gpio_bank(MAX_BLACKFIN_GPIOS)];
+#else
 static unsigned short wakeup_map[gpio_bank(MAX_BLACKFIN_GPIOS)];
 static unsigned char wakeup_flags_map[MAX_BLACKFIN_GPIOS];
 static struct gpio_port_s gpio_bank_saved[gpio_bank(MAX_BLACKFIN_GPIOS)];
@@ -206,7 +209,7 @@ static unsigned int sic_iwr_irqs[gpio_bank(MAX_BLACKFIN_GPIOS)] = {IRQ_PORTF_INT
 #ifdef BF561_FAMILY
 static unsigned int sic_iwr_irqs[gpio_bank(MAX_BLACKFIN_GPIOS)] = {IRQ_PROG0_INTB, IRQ_PROG1_INTB, IRQ_PROG2_INTB};
 #endif
-
+#endif
 #endif /* CONFIG_PM */
 
 #if defined(BF548_FAMILY)
@@ -667,7 +670,7 @@ static int bfin_gpio_wakeup_type(unsigned gpio, unsigned char type)
 	return 0;
 }
 
-u32 bfin_pm_setup(void)
+u32 bfin_pm_standby_setup(void)
 {
 	u16 bank, mask, i, gpio;
 
@@ -679,7 +682,7 @@ u32 bfin_pm_setup(void)
 		gpio_bankb[bank]->maskb = 0;
 
 		if (mask) {
-#ifdef BF537_FAMILY
+#if defined(BF527_FAMILY) || defined(BF537_FAMILY)
 			gpio_bank_saved[bank].fer   = *port_fer[bank];
 #endif
 			gpio_bank_saved[bank].inen  = gpio_bankb[bank]->inen;
@@ -715,7 +718,7 @@ u32 bfin_pm_setup(void)
 	return 0;
 }
 
-void bfin_pm_restore(void)
+void bfin_pm_standby_restore(void)
 {
 	u16 bank, mask, i;
 
@@ -724,7 +727,7 @@ void bfin_pm_restore(void)
 		bank = gpio_bank(i);
 
 		if (mask) {
-#ifdef BF537_FAMILY
+#if defined(BF527_FAMILY) || defined(BF537_FAMILY)
 			*port_fer[bank]   	= gpio_bank_saved[bank].fer;
 #endif
 			gpio_bankb[bank]->inen  = gpio_bank_saved[bank].inen;
@@ -743,8 +746,111 @@ void bfin_pm_restore(void)
 	AWA_DUMMY_READ(maskb);
 }
 
+void bfin_gpio_pm_hibernate_suspend(void)
+{
+	int i, bank;
+
+	for (i = 0; i < MAX_BLACKFIN_GPIOS; i += GPIO_BANKSIZE) {
+		bank = gpio_bank(i);
+
+#if defined(BF527_FAMILY) || defined(BF537_FAMILY)
+			gpio_bank_saved[bank].fer   = *port_fer[bank];
+#ifdef BF527_FAMILY
+			gpio_bank_saved[bank].mux   = *port_mux[bank];
+#else
+			if (bank == 0)
+				gpio_bank_saved[bank].mux   = bfin_read_PORT_MUX();
+#endif
+#endif
+			gpio_bank_saved[bank].data  = gpio_bankb[bank]->data;
+			gpio_bank_saved[bank].inen  = gpio_bankb[bank]->inen;
+			gpio_bank_saved[bank].polar = gpio_bankb[bank]->polar;
+			gpio_bank_saved[bank].dir   = gpio_bankb[bank]->dir;
+			gpio_bank_saved[bank].edge  = gpio_bankb[bank]->edge;
+			gpio_bank_saved[bank].both  = gpio_bankb[bank]->both;
+			gpio_bank_saved[bank].maska  = gpio_bankb[bank]->maska;
+	}
+
+	AWA_DUMMY_READ(maska);
+}
+
+void bfin_gpio_pm_hibernate_restore(void)
+{
+	int i, bank;
+
+	for (i = 0; i < MAX_BLACKFIN_GPIOS; i += GPIO_BANKSIZE) {
+			bank = gpio_bank(i);
+
+#if defined(BF527_FAMILY) || defined(BF537_FAMILY)
+#ifdef BF527_FAMILY
+			*port_mux[bank] = gpio_bank_saved[bank].mux;
+#else
+			if (bank == 0)
+				bfin_write_PORT_MUX(gpio_bank_saved[bank].mux);
+#endif
+			*port_fer[bank]   	= gpio_bank_saved[bank].fer;
+#endif
+			gpio_bankb[bank]->inen  = gpio_bank_saved[bank].inen;
+			gpio_bankb[bank]->dir   = gpio_bank_saved[bank].dir;
+			gpio_bankb[bank]->polar = gpio_bank_saved[bank].polar;
+			gpio_bankb[bank]->edge  = gpio_bank_saved[bank].edge;
+			gpio_bankb[bank]->both  = gpio_bank_saved[bank].both;
+
+			gpio_bankb[bank]->data_set = gpio_bank_saved[bank].data
+							| gpio_bank_saved[bank].dir;
+
+			gpio_bankb[bank]->maska = gpio_bank_saved[bank].maska;
+	}
+	AWA_DUMMY_READ(maska);
+}
+
+
 #endif
 #else /* BF548_FAMILY */
+#ifdef CONFIG_PM
+
+u32 bfin_pm_standby_setup(void)
+{
+	return 0;
+}
+
+void bfin_pm_standby_restore(void)
+{
+
+}
+
+void bfin_gpio_pm_hibernate_suspend(void)
+{
+	int i, bank;
+
+	for (i = 0; i < MAX_BLACKFIN_GPIOS; i += GPIO_BANKSIZE) {
+		bank = gpio_bank(i);
+
+			gpio_bank_saved[bank].fer  = gpio_array[bank]->port_fer;
+			gpio_bank_saved[bank].mux  = gpio_array[bank]->port_mux;
+			gpio_bank_saved[bank].data  = gpio_array[bank]->port_data;
+			gpio_bank_saved[bank].data  = gpio_array[bank]->port_data;
+			gpio_bank_saved[bank].inen  = gpio_array[bank]->port_inen;
+			gpio_bank_saved[bank].dir   = gpio_array[bank]->port_dir_set;
+	}
+}
+
+void bfin_gpio_pm_hibernate_restore(void)
+{
+	int i, bank;
+
+	for (i = 0; i < MAX_BLACKFIN_GPIOS; i += GPIO_BANKSIZE) {
+			bank = gpio_bank(i);
+
+			gpio_array[bank]->port_mux  = gpio_bank_saved[bank].mux;
+			gpio_array[bank]->port_fer  = gpio_bank_saved[bank].fer;
+			gpio_array[bank]->port_inen  = gpio_bank_saved[bank].inen;
+			gpio_array[bank]->port_dir_set   = gpio_bank_saved[bank].dir;
+			gpio_array[bank]->port_set = gpio_bank_saved[bank].data
+							| gpio_bank_saved[bank].dir;
+	}
+}
+#endif
 
 unsigned short get_gpio_dir(unsigned gpio)
 {
diff --git a/arch/blackfin/mach-common/dpmc_modes.S b/arch/blackfin/mach-common/dpmc_modes.S
index b7981d31c392..46ee77afed20 100644
--- a/arch/blackfin/mach-common/dpmc_modes.S
+++ b/arch/blackfin/mach-common/dpmc_modes.S
@@ -7,7 +7,7 @@
 #include <linux/linkage.h>
 #include <asm/blackfin.h>
 #include <asm/mach/irq.h>
-
+#include <asm/dpmc.h>
 
 .section .l1.text
 
@@ -56,26 +56,25 @@ ENTRY(_hibernate_mode)
 	[--SP] = ( R7:0, P5:0 );
 	[--SP] =  RETS;
 
+	R3 = R0;
+	R0 = IWR_DISABLE_ALL;
+	R1 = IWR_DISABLE_ALL;
+	R2 = IWR_DISABLE_ALL;
 	call _set_sic_iwr;
+	call _set_dram_srfs;
+	SSYNC;
 
 	R0 = 0xFFFF (Z);
 	call _set_rtc_istat;
 
 	P0.H = hi(VR_CTL);
 	P0.L = lo(VR_CTL);
-	R1 = W[P0](z);
-	BITSET (R1, 8);
-	BITCLR (R1, 0);
-	BITCLR (R1, 1);
-	W[P0] = R1.L;
-	SSYNC;
 
+	W[P0] = R3.L;
 	CLI R2;
 	IDLE;
-
-	/* Actually, adding anything may not be necessary...SDRAM contents
-	 * are lost
-	 */
+.Lforever:
+	jump .Lforever;
 
 ENTRY(_deep_sleep)
 	[--SP] = ( R7:0, P5:0 );
@@ -233,51 +232,70 @@ ENTRY(_sleep_deeper)
 	( R7:0, P5:0 ) = [SP++];
 	RTS;
 
+
 ENTRY(_set_dram_srfs)
 	/*  set the dram to self refresh mode */
-#if defined(CONFIG_BF54x)
+	SSYNC;
+#if defined(EBIU_RSTCTL)	/* DDR */
 	P0.H = hi(EBIU_RSTCTL);
 	P0.L = lo(EBIU_RSTCTL);
 	R2 = [P0];
-	R3.H = hi(SRREQ);
-	R3.L = lo(SRREQ);
-#else
-	P0.H = hi(EBIU_SDGCTL);
-	P0.L = lo(EBIU_SDGCTL);
-	R2 = [P0];
-	R3.H = hi(SRFS);
-	R3.L = lo(SRFS);
-#endif
-	R2 = R2|R3;
+	BITSET(R2, 3); /* SRREQ enter self-refresh mode */
 	[P0] = R2;
-	ssync;
-#if defined(CONFIG_BF54x)
-.LSRR_MODE:
+	SSYNC;
+1:
 	R2 = [P0];
 	CC = BITTST(R2, 4);
-	if !CC JUMP .LSRR_MODE;
+	if !CC JUMP 1b;
+#else 				/* SDRAM */
+	P0.L = lo(EBIU_SDGCTL);
+	P0.H = hi(EBIU_SDGCTL);
+	R2 = [P0];
+	BITSET(R2, 24); /* SRFS enter self-refresh mode */
+	[P0] = R2;
+	SSYNC;
+
+	P0.L = lo(EBIU_SDSTAT);
+	P0.H = hi(EBIU_SDSTAT);
+1:
+	R2 = w[P0];
+	SSYNC;
+	cc = BITTST(R2, 1); /* SDSRA poll self-refresh status */
+	if !cc jump 1b;
+
+	P0.L = lo(EBIU_SDGCTL);
+	P0.H = hi(EBIU_SDGCTL);
+	R2 = [P0];
+	BITCLR(R2, 0); /* SCTLE disable CLKOUT */
+	[P0] = R2;
 #endif
 	RTS;
 
+
 ENTRY(_unset_dram_srfs)
 	/*  set the dram out of self refresh mode */
-#if defined(CONFIG_BF54x)
+#if defined(EBIU_RSTCTL)	/* DDR */
 	P0.H = hi(EBIU_RSTCTL);
 	P0.L = lo(EBIU_RSTCTL);
 	R2 = [P0];
-	R3.H = hi(SRREQ);
-	R3.L = lo(SRREQ);
-#else
-	P0.H = hi(EBIU_SDGCTL);
-	P0.L = lo(EBIU_SDGCTL);
-	R2 = [P0];
-	R3.H = hi(SRFS);
-	R3.L = lo(SRFS);
-#endif
-	R3 = ~R3;
-	R2 = R2&R3;
+	BITCLR(R2, 3); /* clear SRREQ bit */
 	[P0] = R2;
-	ssync;
+#elif defined(EBIU_SDGCTL)	/* SDRAM */
+
+	P0.L = lo(EBIU_SDGCTL); /* release CLKOUT from self-refresh */
+	P0.H = hi(EBIU_SDGCTL);
+	R2 = [P0];
+	BITSET(R2, 0); /* SCTLE enable CLKOUT */
+	[P0] = R2
+	SSYNC;
+
+	P0.L = lo(EBIU_SDGCTL); /* release SDRAM from self-refresh */
+	P0.H = hi(EBIU_SDGCTL);
+	R2 = [P0];
+	BITCLR(R2, 24); /* clear SRFS bit */
+	[P0] = R2
+#endif
+	SSYNC;
 	RTS;
 
 ENTRY(_set_sic_iwr)
@@ -307,6 +325,11 @@ ENTRY(_set_rtc_istat)
 	P0.L = lo(RTC_ISTAT);
 	w[P0] = R0.L;
 	SSYNC;
+#elif (ANOMALY_05000371)
+	nop;
+	nop;
+	nop;
+	nop;
 #endif
 	RTS;
 
@@ -318,3 +341,508 @@ ENTRY(_test_pll_locked)
 	CC = BITTST(R0,5);
 	IF !CC JUMP 1b;
 	RTS;
+
+.section .text
+
+
+ENTRY(_do_hibernate)
+	[--SP] = ( R7:0, P5:0 );
+	[--SP] =  RETS;
+	/* Save System MMRs */
+	R2 = R0;
+	P0.H = hi(PLL_CTL);
+	P0.L = lo(PLL_CTL);
+
+#ifdef SIC_IMASK0
+	PM_SYS_PUSH(SIC_IMASK0)
+#endif
+#ifdef SIC_IMASK1
+	PM_SYS_PUSH(SIC_IMASK1)
+#endif
+#ifdef SIC_IMASK2
+	PM_SYS_PUSH(SIC_IMASK2)
+#endif
+#ifdef SIC_IMASK
+	PM_SYS_PUSH(SIC_IMASK)
+#endif
+#ifdef SICA_IMASK0
+	PM_SYS_PUSH(SICA_IMASK0)
+#endif
+#ifdef SICA_IMASK1
+	PM_SYS_PUSH(SICA_IMASK1)
+#endif
+#ifdef SIC_IAR2
+	PM_SYS_PUSH(SIC_IAR0)
+	PM_SYS_PUSH(SIC_IAR1)
+	PM_SYS_PUSH(SIC_IAR2)
+#endif
+#ifdef SIC_IAR3
+	PM_SYS_PUSH(SIC_IAR3)
+#endif
+#ifdef SIC_IAR4
+	PM_SYS_PUSH(SIC_IAR4)
+	PM_SYS_PUSH(SIC_IAR5)
+	PM_SYS_PUSH(SIC_IAR6)
+#endif
+#ifdef SIC_IAR7
+	PM_SYS_PUSH(SIC_IAR7)
+#endif
+#ifdef SIC_IAR8
+	PM_SYS_PUSH(SIC_IAR8)
+	PM_SYS_PUSH(SIC_IAR9)
+	PM_SYS_PUSH(SIC_IAR10)
+	PM_SYS_PUSH(SIC_IAR11)
+#endif
+
+#ifdef SICA_IAR0
+	PM_SYS_PUSH(SICA_IAR0)
+	PM_SYS_PUSH(SICA_IAR1)
+	PM_SYS_PUSH(SICA_IAR2)
+	PM_SYS_PUSH(SICA_IAR3)
+	PM_SYS_PUSH(SICA_IAR4)
+	PM_SYS_PUSH(SICA_IAR5)
+	PM_SYS_PUSH(SICA_IAR6)
+	PM_SYS_PUSH(SICA_IAR7)
+#endif
+
+#ifdef SIC_IWR
+	PM_SYS_PUSH(SIC_IWR)
+#endif
+#ifdef SIC_IWR0
+	PM_SYS_PUSH(SIC_IWR0)
+#endif
+#ifdef SIC_IWR1
+	PM_SYS_PUSH(SIC_IWR1)
+#endif
+#ifdef SIC_IWR2
+	PM_SYS_PUSH(SIC_IWR2)
+#endif
+#ifdef SICA_IWR0
+	PM_SYS_PUSH(SICA_IWR0)
+#endif
+#ifdef SICA_IWR1
+	PM_SYS_PUSH(SICA_IWR1)
+#endif
+
+#ifdef PINT0_ASSIGN
+	PM_SYS_PUSH(PINT0_ASSIGN)
+	PM_SYS_PUSH(PINT1_ASSIGN)
+	PM_SYS_PUSH(PINT2_ASSIGN)
+	PM_SYS_PUSH(PINT3_ASSIGN)
+#endif
+
+	PM_SYS_PUSH(EBIU_AMBCTL0)
+	PM_SYS_PUSH(EBIU_AMBCTL1)
+	PM_SYS_PUSH16(EBIU_AMGCTL)
+
+#ifdef EBIU_FCTL
+	PM_SYS_PUSH(EBIU_MBSCTL)
+	PM_SYS_PUSH(EBIU_MODE)
+	PM_SYS_PUSH(EBIU_FCTL)
+#endif
+
+	PM_SYS_PUSH16(SYSCR)
+
+	/* Save Core MMRs */
+	P0.H = hi(SRAM_BASE_ADDRESS);
+	P0.L = lo(SRAM_BASE_ADDRESS);
+
+	PM_PUSH(DMEM_CONTROL)
+	PM_PUSH(DCPLB_ADDR0)
+	PM_PUSH(DCPLB_ADDR1)
+	PM_PUSH(DCPLB_ADDR2)
+	PM_PUSH(DCPLB_ADDR3)
+	PM_PUSH(DCPLB_ADDR4)
+	PM_PUSH(DCPLB_ADDR5)
+	PM_PUSH(DCPLB_ADDR6)
+	PM_PUSH(DCPLB_ADDR7)
+	PM_PUSH(DCPLB_ADDR8)
+	PM_PUSH(DCPLB_ADDR9)
+	PM_PUSH(DCPLB_ADDR10)
+	PM_PUSH(DCPLB_ADDR11)
+	PM_PUSH(DCPLB_ADDR12)
+	PM_PUSH(DCPLB_ADDR13)
+	PM_PUSH(DCPLB_ADDR14)
+	PM_PUSH(DCPLB_ADDR15)
+	PM_PUSH(DCPLB_DATA0)
+	PM_PUSH(DCPLB_DATA1)
+	PM_PUSH(DCPLB_DATA2)
+	PM_PUSH(DCPLB_DATA3)
+	PM_PUSH(DCPLB_DATA4)
+	PM_PUSH(DCPLB_DATA5)
+	PM_PUSH(DCPLB_DATA6)
+	PM_PUSH(DCPLB_DATA7)
+	PM_PUSH(DCPLB_DATA8)
+	PM_PUSH(DCPLB_DATA9)
+	PM_PUSH(DCPLB_DATA10)
+	PM_PUSH(DCPLB_DATA11)
+	PM_PUSH(DCPLB_DATA12)
+	PM_PUSH(DCPLB_DATA13)
+	PM_PUSH(DCPLB_DATA14)
+	PM_PUSH(DCPLB_DATA15)
+	PM_PUSH(IMEM_CONTROL)
+	PM_PUSH(ICPLB_ADDR0)
+	PM_PUSH(ICPLB_ADDR1)
+	PM_PUSH(ICPLB_ADDR2)
+	PM_PUSH(ICPLB_ADDR3)
+	PM_PUSH(ICPLB_ADDR4)
+	PM_PUSH(ICPLB_ADDR5)
+	PM_PUSH(ICPLB_ADDR6)
+	PM_PUSH(ICPLB_ADDR7)
+	PM_PUSH(ICPLB_ADDR8)
+	PM_PUSH(ICPLB_ADDR9)
+	PM_PUSH(ICPLB_ADDR10)
+	PM_PUSH(ICPLB_ADDR11)
+	PM_PUSH(ICPLB_ADDR12)
+	PM_PUSH(ICPLB_ADDR13)
+	PM_PUSH(ICPLB_ADDR14)
+	PM_PUSH(ICPLB_ADDR15)
+	PM_PUSH(ICPLB_DATA0)
+	PM_PUSH(ICPLB_DATA1)
+	PM_PUSH(ICPLB_DATA2)
+	PM_PUSH(ICPLB_DATA3)
+	PM_PUSH(ICPLB_DATA4)
+	PM_PUSH(ICPLB_DATA5)
+	PM_PUSH(ICPLB_DATA6)
+	PM_PUSH(ICPLB_DATA7)
+	PM_PUSH(ICPLB_DATA8)
+	PM_PUSH(ICPLB_DATA9)
+	PM_PUSH(ICPLB_DATA10)
+	PM_PUSH(ICPLB_DATA11)
+	PM_PUSH(ICPLB_DATA12)
+	PM_PUSH(ICPLB_DATA13)
+	PM_PUSH(ICPLB_DATA14)
+	PM_PUSH(ICPLB_DATA15)
+	PM_PUSH(EVT0)
+	PM_PUSH(EVT1)
+	PM_PUSH(EVT2)
+	PM_PUSH(EVT3)
+	PM_PUSH(EVT4)
+	PM_PUSH(EVT5)
+	PM_PUSH(EVT6)
+	PM_PUSH(EVT7)
+	PM_PUSH(EVT8)
+	PM_PUSH(EVT9)
+	PM_PUSH(EVT10)
+	PM_PUSH(EVT11)
+	PM_PUSH(EVT12)
+	PM_PUSH(EVT13)
+	PM_PUSH(EVT14)
+	PM_PUSH(EVT15)
+	PM_PUSH(IMASK)
+	PM_PUSH(ILAT)
+	PM_PUSH(IPRIO)
+	PM_PUSH(TCNTL)
+	PM_PUSH(TPERIOD)
+	PM_PUSH(TSCALE)
+	PM_PUSH(TCOUNT)
+	PM_PUSH(TBUFCTL)
+
+	/* Save Core Registers */
+	[--sp] = SYSCFG;
+	[--sp] = ( R7:0, P5:0 );
+	[--sp] = fp;
+	[--sp] = usp;
+
+	[--sp] = i0;
+	[--sp] = i1;
+	[--sp] = i2;
+	[--sp] = i3;
+
+	[--sp] = m0;
+	[--sp] = m1;
+	[--sp] = m2;
+	[--sp] = m3;
+
+	[--sp] = l0;
+	[--sp] = l1;
+	[--sp] = l2;
+	[--sp] = l3;
+
+	[--sp] = b0;
+	[--sp] = b1;
+	[--sp] = b2;
+	[--sp] = b3;
+	[--sp] = a0.x;
+	[--sp] = a0.w;
+	[--sp] = a1.x;
+	[--sp] = a1.w;
+
+	[--sp] = LC0;
+	[--sp] = LC1;
+	[--sp] = LT0;
+	[--sp] = LT1;
+	[--sp] = LB0;
+	[--sp] = LB1;
+
+	[--sp] = ASTAT;
+	[--sp] = CYCLES;
+	[--sp] = CYCLES2;
+
+	[--sp] = RETS;
+	r0 = RETI;
+	[--sp] = r0;
+	[--sp] = RETX;
+	[--sp] = RETN;
+	[--sp] = RETE;
+	[--sp] = SEQSTAT;
+
+	/* Save Magic, return address and Stack Pointer */
+	P0.H = 0;
+	P0.L = 0;
+	R0.H = 0xDEAD;	/* Hibernate Magic */
+	R0.L = 0xBEEF;
+	[P0++] = R0;	/* Store Hibernate Magic */
+	R0.H = pm_resume_here;
+	R0.L = pm_resume_here;
+	[P0++] = R0;	/* Save Return Address */
+	[P0++] = SP;	/* Save Stack Pointer */
+	P0.H = _hibernate_mode;
+	P0.L = _hibernate_mode;
+	R0 = R2;
+	call (P0); /* Goodbye */
+
+pm_resume_here:
+
+	/* Restore Core Registers */
+	SEQSTAT = [sp++];
+	RETE = [sp++];
+	RETN = [sp++];
+	RETX = [sp++];
+	r0 = [sp++];
+	RETI = r0;
+	RETS = [sp++];
+
+	CYCLES2 = [sp++];
+	CYCLES = [sp++];
+	ASTAT = [sp++];
+
+	LB1 = [sp++];
+	LB0 = [sp++];
+	LT1 = [sp++];
+	LT0 = [sp++];
+	LC1 = [sp++];
+	LC0 = [sp++];
+
+	a1.w = [sp++];
+	a1.x = [sp++];
+	a0.w = [sp++];
+	a0.x = [sp++];
+	b3 = [sp++];
+	b2 = [sp++];
+	b1 = [sp++];
+	b0 = [sp++];
+
+	l3 = [sp++];
+	l2 = [sp++];
+	l1 = [sp++];
+	l0 = [sp++];
+
+	m3 = [sp++];
+	m2 = [sp++];
+	m1 = [sp++];
+	m0 = [sp++];
+
+	i3 = [sp++];
+	i2 = [sp++];
+	i1 = [sp++];
+	i0 = [sp++];
+
+	usp = [sp++];
+	fp = [sp++];
+
+	( R7 : 0, P5 : 0) = [ SP ++ ];
+	SYSCFG = [sp++];
+
+	/* Restore Core MMRs */
+
+	PM_POP(TBUFCTL)
+	PM_POP(TCOUNT)
+	PM_POP(TSCALE)
+	PM_POP(TPERIOD)
+	PM_POP(TCNTL)
+	PM_POP(IPRIO)
+	PM_POP(ILAT)
+	PM_POP(IMASK)
+	PM_POP(EVT15)
+	PM_POP(EVT14)
+	PM_POP(EVT13)
+	PM_POP(EVT12)
+	PM_POP(EVT11)
+	PM_POP(EVT10)
+	PM_POP(EVT9)
+	PM_POP(EVT8)
+	PM_POP(EVT7)
+	PM_POP(EVT6)
+	PM_POP(EVT5)
+	PM_POP(EVT4)
+	PM_POP(EVT3)
+	PM_POP(EVT2)
+	PM_POP(EVT1)
+	PM_POP(EVT0)
+	PM_POP(ICPLB_DATA15)
+	PM_POP(ICPLB_DATA14)
+	PM_POP(ICPLB_DATA13)
+	PM_POP(ICPLB_DATA12)
+	PM_POP(ICPLB_DATA11)
+	PM_POP(ICPLB_DATA10)
+	PM_POP(ICPLB_DATA9)
+	PM_POP(ICPLB_DATA8)
+	PM_POP(ICPLB_DATA7)
+	PM_POP(ICPLB_DATA6)
+	PM_POP(ICPLB_DATA5)
+	PM_POP(ICPLB_DATA4)
+	PM_POP(ICPLB_DATA3)
+	PM_POP(ICPLB_DATA2)
+	PM_POP(ICPLB_DATA1)
+	PM_POP(ICPLB_DATA0)
+	PM_POP(ICPLB_ADDR15)
+	PM_POP(ICPLB_ADDR14)
+	PM_POP(ICPLB_ADDR13)
+	PM_POP(ICPLB_ADDR12)
+	PM_POP(ICPLB_ADDR11)
+	PM_POP(ICPLB_ADDR10)
+	PM_POP(ICPLB_ADDR9)
+	PM_POP(ICPLB_ADDR8)
+	PM_POP(ICPLB_ADDR7)
+	PM_POP(ICPLB_ADDR6)
+	PM_POP(ICPLB_ADDR5)
+	PM_POP(ICPLB_ADDR4)
+	PM_POP(ICPLB_ADDR3)
+	PM_POP(ICPLB_ADDR2)
+	PM_POP(ICPLB_ADDR1)
+	PM_POP(ICPLB_ADDR0)
+	PM_POP(IMEM_CONTROL)
+	PM_POP(DCPLB_DATA15)
+	PM_POP(DCPLB_DATA14)
+	PM_POP(DCPLB_DATA13)
+	PM_POP(DCPLB_DATA12)
+	PM_POP(DCPLB_DATA11)
+	PM_POP(DCPLB_DATA10)
+	PM_POP(DCPLB_DATA9)
+	PM_POP(DCPLB_DATA8)
+	PM_POP(DCPLB_DATA7)
+	PM_POP(DCPLB_DATA6)
+	PM_POP(DCPLB_DATA5)
+	PM_POP(DCPLB_DATA4)
+	PM_POP(DCPLB_DATA3)
+	PM_POP(DCPLB_DATA2)
+	PM_POP(DCPLB_DATA1)
+	PM_POP(DCPLB_DATA0)
+	PM_POP(DCPLB_ADDR15)
+	PM_POP(DCPLB_ADDR14)
+	PM_POP(DCPLB_ADDR13)
+	PM_POP(DCPLB_ADDR12)
+	PM_POP(DCPLB_ADDR11)
+	PM_POP(DCPLB_ADDR10)
+	PM_POP(DCPLB_ADDR9)
+	PM_POP(DCPLB_ADDR8)
+	PM_POP(DCPLB_ADDR7)
+	PM_POP(DCPLB_ADDR6)
+	PM_POP(DCPLB_ADDR5)
+	PM_POP(DCPLB_ADDR4)
+	PM_POP(DCPLB_ADDR3)
+	PM_POP(DCPLB_ADDR2)
+	PM_POP(DCPLB_ADDR1)
+	PM_POP(DCPLB_ADDR0)
+	PM_POP(DMEM_CONTROL)
+
+	/* Restore System MMRs */
+
+	P0.H = hi(PLL_CTL);
+	P0.L = lo(PLL_CTL);
+	PM_SYS_POP16(SYSCR)
+
+#ifdef EBIU_FCTL
+	PM_SYS_POP(EBIU_FCTL)
+	PM_SYS_POP(EBIU_MODE)
+	PM_SYS_POP(EBIU_MBSCTL)
+#endif
+	PM_SYS_POP16(EBIU_AMGCTL)
+	PM_SYS_POP(EBIU_AMBCTL1)
+	PM_SYS_POP(EBIU_AMBCTL0)
+
+#ifdef PINT0_ASSIGN
+	PM_SYS_POP(PINT3_ASSIGN)
+	PM_SYS_POP(PINT2_ASSIGN)
+	PM_SYS_POP(PINT1_ASSIGN)
+	PM_SYS_POP(PINT0_ASSIGN)
+#endif
+
+#ifdef SICA_IWR1
+	PM_SYS_POP(SICA_IWR1)
+#endif
+#ifdef SICA_IWR0
+	PM_SYS_POP(SICA_IWR0)
+#endif
+#ifdef SIC_IWR2
+	PM_SYS_POP(SIC_IWR2)
+#endif
+#ifdef SIC_IWR1
+	PM_SYS_POP(SIC_IWR1)
+#endif
+#ifdef SIC_IWR0
+	PM_SYS_POP(SIC_IWR0)
+#endif
+#ifdef SIC_IWR
+	PM_SYS_POP(SIC_IWR)
+#endif
+
+#ifdef SICA_IAR0
+	PM_SYS_POP(SICA_IAR7)
+	PM_SYS_POP(SICA_IAR6)
+	PM_SYS_POP(SICA_IAR5)
+	PM_SYS_POP(SICA_IAR4)
+	PM_SYS_POP(SICA_IAR3)
+	PM_SYS_POP(SICA_IAR2)
+	PM_SYS_POP(SICA_IAR1)
+	PM_SYS_POP(SICA_IAR0)
+#endif
+
+#ifdef SIC_IAR8
+	PM_SYS_POP(SIC_IAR11)
+	PM_SYS_POP(SIC_IAR10)
+	PM_SYS_POP(SIC_IAR9)
+	PM_SYS_POP(SIC_IAR8)
+#endif
+#ifdef SIC_IAR7
+	PM_SYS_POP(SIC_IAR7)
+#endif
+#ifdef SIC_IAR6
+	PM_SYS_POP(SIC_IAR6)
+	PM_SYS_POP(SIC_IAR5)
+	PM_SYS_POP(SIC_IAR4)
+#endif
+#ifdef SIC_IAR3
+	PM_SYS_POP(SIC_IAR3)
+#endif
+#ifdef SIC_IAR2
+	PM_SYS_POP(SIC_IAR2)
+	PM_SYS_POP(SIC_IAR1)
+	PM_SYS_POP(SIC_IAR0)
+#endif
+#ifdef SICA_IMASK1
+	PM_SYS_POP(SICA_IMASK1)
+#endif
+#ifdef SICA_IMASK0
+	PM_SYS_POP(SICA_IMASK0)
+#endif
+#ifdef SIC_IMASK
+	PM_SYS_POP(SIC_IMASK)
+#endif
+#ifdef SIC_IMASK2
+	PM_SYS_POP(SIC_IMASK2)
+#endif
+#ifdef SIC_IMASK1
+	PM_SYS_POP(SIC_IMASK1)
+#endif
+#ifdef SIC_IMASK0
+	PM_SYS_POP(SIC_IMASK0)
+#endif
+
+	[--sp] = RETI;	/* Clear Global Interrupt Disable */
+	SP += 4;
+
+	RETS = [SP++];
+	( R7:0, P5:0 ) = [SP++];
+	RTS;
diff --git a/arch/blackfin/mach-common/pm.c b/arch/blackfin/mach-common/pm.c
index 0be805ca423f..4fe6a2366b13 100644
--- a/arch/blackfin/mach-common/pm.c
+++ b/arch/blackfin/mach-common/pm.c
@@ -38,8 +38,9 @@
 #include <linux/io.h>
 #include <linux/irq.h>
 
-#include <asm/dpmc.h>
 #include <asm/gpio.h>
+#include <asm/dma.h>
+#include <asm/dpmc.h>
 
 #ifdef CONFIG_PM_WAKEUP_GPIO_POLAR_H
 #define WAKEUP_TYPE	PM_WAKE_HIGH
@@ -61,16 +62,17 @@
 #define WAKEUP_TYPE	PM_WAKE_BOTH_EDGES
 #endif
 
+
 void bfin_pm_suspend_standby_enter(void)
 {
+	unsigned long flags;
+
 #ifdef CONFIG_PM_WAKEUP_BY_GPIO
 	gpio_pm_wakeup_request(CONFIG_PM_WAKEUP_GPIO_NUMBER, WAKEUP_TYPE);
 #endif
 
-	u32 flags;
-
 	local_irq_save(flags);
-	bfin_pm_setup();
+	bfin_pm_standby_setup();
 
 #ifdef CONFIG_PM_BFIN_SLEEP_DEEPER
 	sleep_deeper(bfin_sic_iwr[0], bfin_sic_iwr[1], bfin_sic_iwr[2]);
@@ -78,7 +80,7 @@ void bfin_pm_suspend_standby_enter(void)
 	sleep_mode(bfin_sic_iwr[0], bfin_sic_iwr[1], bfin_sic_iwr[2]);
 #endif
 
-	bfin_pm_restore();
+	bfin_pm_standby_restore();
 
 #if defined(CONFIG_BF54x) || defined(CONFIG_BF52x)  || defined(CONFIG_BF561)
 	bfin_write_SIC_IWR0(IWR_ENABLE_ALL);
@@ -93,6 +95,195 @@ void bfin_pm_suspend_standby_enter(void)
 	local_irq_restore(flags);
 }
 
+int bf53x_suspend_l1_mem(unsigned char *memptr)
+{
+	dma_memcpy(memptr, (const void *) L1_CODE_START, L1_CODE_LENGTH);
+	dma_memcpy(memptr + L1_CODE_LENGTH, (const void *) L1_DATA_A_START,
+			L1_DATA_A_LENGTH);
+	dma_memcpy(memptr + L1_CODE_LENGTH + L1_DATA_A_LENGTH,
+			(const void *) L1_DATA_B_START, L1_DATA_B_LENGTH);
+	memcpy(memptr + L1_CODE_LENGTH + L1_DATA_A_LENGTH +
+			L1_DATA_B_LENGTH, (const void *) L1_SCRATCH_START,
+			L1_SCRATCH_LENGTH);
+
+	return 0;
+}
+
+int bf53x_resume_l1_mem(unsigned char *memptr)
+{
+	dma_memcpy((void *) L1_CODE_START, memptr, L1_CODE_LENGTH);
+	dma_memcpy((void *) L1_DATA_A_START, memptr + L1_CODE_LENGTH,
+			L1_DATA_A_LENGTH);
+	dma_memcpy((void *) L1_DATA_B_START, memptr + L1_CODE_LENGTH +
+			L1_DATA_A_LENGTH, L1_DATA_B_LENGTH);
+	memcpy((void *) L1_SCRATCH_START, memptr + L1_CODE_LENGTH +
+			L1_DATA_A_LENGTH + L1_DATA_B_LENGTH, L1_SCRATCH_LENGTH);
+
+	return 0;
+}
+
+#ifdef CONFIG_BFIN_WB
+static void flushinv_all_dcache(void)
+{
+	u32 way, bank, subbank, set;
+	u32 status, addr;
+	u32 dmem_ctl = bfin_read_DMEM_CONTROL();
+
+	for (bank = 0; bank < 2; ++bank) {
+		if (!(dmem_ctl & (1 << (DMC1_P - bank))))
+			continue;
+
+		for (way = 0; way < 2; ++way)
+			for (subbank = 0; subbank < 4; ++subbank)
+				for (set = 0; set < 64; ++set) {
+
+					bfin_write_DTEST_COMMAND(
+						way << 26 |
+						bank << 23 |
+						subbank << 16 |
+						set << 5
+					);
+					CSYNC();
+					status = bfin_read_DTEST_DATA0();
+
+					/* only worry about valid/dirty entries */
+					if ((status & 0x3) != 0x3)
+						continue;
+
+					/* construct the address using the tag */
+					addr = (status & 0xFFFFC800) | (subbank << 12) | (set << 5);
+
+					/* flush it */
+					__asm__ __volatile__("FLUSHINV[%0];" : : "a"(addr));
+				}
+	}
+}
+#endif
+
+static inline void dcache_disable(void)
+{
+#ifdef CONFIG_BFIN_DCACHE
+	unsigned long ctrl;
+
+#ifdef CONFIG_BFIN_WB
+	flushinv_all_dcache();
+#endif
+	SSYNC();
+	ctrl = bfin_read_DMEM_CONTROL();
+	ctrl &= ~ENDCPLB;
+	bfin_write_DMEM_CONTROL(ctrl);
+	SSYNC();
+#endif
+}
+
+static inline void dcache_enable(void)
+{
+#ifdef CONFIG_BFIN_DCACHE
+	unsigned long ctrl;
+	SSYNC();
+	ctrl = bfin_read_DMEM_CONTROL();
+	ctrl |= ENDCPLB;
+	bfin_write_DMEM_CONTROL(ctrl);
+	SSYNC();
+#endif
+}
+
+static inline void icache_disable(void)
+{
+#ifdef CONFIG_BFIN_ICACHE
+	unsigned long ctrl;
+	SSYNC();
+	ctrl = bfin_read_IMEM_CONTROL();
+	ctrl &= ~ENICPLB;
+	bfin_write_IMEM_CONTROL(ctrl);
+	SSYNC();
+#endif
+}
+
+static inline void icache_enable(void)
+{
+#ifdef CONFIG_BFIN_ICACHE
+	unsigned long ctrl;
+	SSYNC();
+	ctrl = bfin_read_IMEM_CONTROL();
+	ctrl |= ENICPLB;
+	bfin_write_IMEM_CONTROL(ctrl);
+	SSYNC();
+#endif
+}
+
+int bfin_pm_suspend_mem_enter(void)
+{
+	unsigned long flags;
+	int wakeup, ret;
+
+	unsigned char *memptr = kmalloc(L1_CODE_LENGTH + L1_DATA_A_LENGTH
+					 + L1_DATA_B_LENGTH + L1_SCRATCH_LENGTH,
+					  GFP_KERNEL);
+
+	if (memptr == NULL) {
+		panic("bf53x_suspend_l1_mem malloc failed");
+		return -ENOMEM;
+	}
+
+	wakeup = bfin_read_VR_CTL() & ~FREQ;
+	wakeup |= SCKELOW;
+
+	/* FIXME: merge this somehow with set_irq_wake */
+#ifdef CONFIG_PM_BFIN_WAKE_RTC
+	wakeup |= WAKE;
+#endif
+#ifdef CONFIG_PM_BFIN_WAKE_PH6
+	wakeup |= PHYWE;
+#endif
+#ifdef CONFIG_PM_BFIN_WAKE_CAN
+	wakeup |= CANWE;
+#endif
+#ifdef CONFIG_PM_BFIN_WAKE_GP
+	wakeup |= GPWE;
+#endif
+#ifdef CONFIG_PM_BFIN_WAKE_USB
+	wakeup |= USBWE;
+#endif
+#ifdef CONFIG_PM_BFIN_WAKE_KEYPAD
+	wakeup |= KPADWE;
+#endif
+#ifdef CONFIG_PM_BFIN_WAKE_ROTARY
+	wakeup |= ROTWE;
+#endif
+
+	local_irq_save(flags);
+
+	ret = blackfin_dma_suspend();
+
+	if (ret) {
+		local_irq_restore(flags);
+		kfree(memptr);
+		return ret;
+	}
+
+	bfin_gpio_pm_hibernate_suspend();
+
+	dcache_disable();
+	icache_disable();
+	bf53x_suspend_l1_mem(memptr);
+
+	do_hibernate(wakeup);	/* Goodbye */
+
+	bf53x_resume_l1_mem(memptr);
+
+	icache_enable();
+	dcache_enable();
+
+	bfin_gpio_pm_hibernate_restore();
+	blackfin_dma_resume();
+
+	local_irq_restore(flags);
+	kfree(memptr);
+
+	return 0;
+}
+
 /*
  *	bfin_pm_valid - Tell the PM core that we only support the standby sleep
  *			state
@@ -101,7 +292,24 @@ void bfin_pm_suspend_standby_enter(void)
  */
 static int bfin_pm_valid(suspend_state_t state)
 {
-	return (state == PM_SUSPEND_STANDBY);
+	return (state == PM_SUSPEND_STANDBY
+#ifndef BF533_FAMILY
+	/*
+	 * On BF533/2/1:
+	 * If we enter Hibernate the SCKE Pin is driven Low,
+	 * so that the SDRAM enters Self Refresh Mode.
+	 * However when the reset sequence that follows hibernate
+	 * state is executed, SCKE is driven High, taking the
+	 * SDRAM out of Self Refresh.
+	 *
+	 * If you reconfigure and access the SDRAM "very quickly",
+	 * you are likely to avoid errors, otherwise the SDRAM
+	 * start losing its contents.
+	 * An external HW workaround is possible using logic gates.
+	 */
+	|| state == PM_SUSPEND_MEM
+#endif
+	);
 }
 
 /*
@@ -115,10 +323,9 @@ static int bfin_pm_enter(suspend_state_t state)
 	case PM_SUSPEND_STANDBY:
 		bfin_pm_suspend_standby_enter();
 		break;
-
 	case PM_SUSPEND_MEM:
-		return -ENOTSUPP;
-
+		bfin_pm_suspend_mem_enter();
+		break;
 	default:
 		return -EINVAL;
 	}
diff --git a/include/asm-blackfin/dma.h b/include/asm-blackfin/dma.h
index c0d5259e315b..3cd4b522aa3f 100644
--- a/include/asm-blackfin/dma.h
+++ b/include/asm-blackfin/dma.h
@@ -144,8 +144,16 @@ struct dma_channel {
 	void *data;
 	unsigned int dma_enable_flag;
 	unsigned int loopback_flag;
+#ifdef CONFIG_PM
+	unsigned short saved_peripheral_map;
+#endif
 };
 
+#ifdef CONFIG_PM
+int blackfin_dma_suspend(void);
+void blackfin_dma_resume(void);
+#endif
+
 /*******************************************************************************
 *	DMA API's
 *******************************************************************************/
diff --git a/include/asm-blackfin/dpmc.h b/include/asm-blackfin/dpmc.h
index 7f34cd384f12..de28e6e018b3 100644
--- a/include/asm-blackfin/dpmc.h
+++ b/include/asm-blackfin/dpmc.h
@@ -7,63 +7,18 @@
 #ifndef _BLACKFIN_DPMC_H_
 #define _BLACKFIN_DPMC_H_
 
-#define SLEEP_MODE		1
-#define DEEP_SLEEP_MODE		2
-#define ACTIVE_PLL_DISABLED	3
-#define FULLON_MODE		4
-#define ACTIVE_PLL_ENABLED	5
-#define HIBERNATE_MODE		6
-
-#define IOCTL_FULL_ON_MODE	_IO('s', 0xA0)
-#define IOCTL_ACTIVE_MODE	_IO('s', 0xA1)
-#define IOCTL_SLEEP_MODE	_IO('s', 0xA2)
-#define IOCTL_DEEP_SLEEP_MODE	_IO('s', 0xA3)
-#define IOCTL_HIBERNATE_MODE	_IO('s', 0xA4)
-#define IOCTL_CHANGE_FREQUENCY	_IOW('s', 0xA5, unsigned long)
-#define IOCTL_CHANGE_VOLTAGE	_IOW('s', 0xA6, unsigned long)
-#define IOCTL_SET_CCLK		_IOW('s', 0xA7, unsigned long)
-#define IOCTL_SET_SCLK		_IOW('s', 0xA8, unsigned long)
-#define IOCTL_GET_PLLSTATUS	_IOW('s', 0xA9, unsigned long)
-#define IOCTL_GET_CORECLOCK	_IOW('s', 0xAA, unsigned long)
-#define IOCTL_GET_SYSTEMCLOCK	_IOW('s', 0xAB, unsigned long)
-#define IOCTL_GET_VCO		_IOW('s', 0xAC, unsigned long)
-#define IOCTL_DISABLE_WDOG_TIMER _IO('s', 0xAD)
-#define IOCTL_UNMASK_WDOG_WAKEUP_EVENT _IO('s',0xAE)
-#define IOCTL_PROGRAM_WDOG_TIMER _IOW('s',0xAF,unsigned long)
-#define IOCTL_CLEAR_WDOG_WAKEUP_EVENT _IO('s',0xB0)
-#define IOCTL_SLEEP_DEEPER_MODE _IO('s',0xB1)
-
-#define DPMC_MINOR		254
-
-#define ON	0
-#define OFF	1
-
 #ifdef __KERNEL__
+#ifndef __ASSEMBLY__
 
-unsigned long calc_volt(void);
-int calc_vlev(int vlt);
-unsigned long change_voltage(unsigned long volt);
-int calc_msel(int vco_hz);
-unsigned long change_frequency(unsigned long vco_mhz);
-int set_pll_div(unsigned short sel, unsigned char flag);
-int get_vco(void);
-unsigned long change_system_clock(unsigned long clock);
-unsigned long change_core_clock(unsigned long clock);
-unsigned long get_pll_status(void);
-void change_baud(int baud);
-void fullon_mode(void);
-void active_mode(void);
 void sleep_mode(u32 sic_iwr0, u32 sic_iwr1, u32 sic_iwr2);
 void deep_sleep(u32 sic_iwr0, u32 sic_iwr1, u32 sic_iwr2);
 void hibernate_mode(u32 sic_iwr0, u32 sic_iwr1, u32 sic_iwr2);
 void sleep_deeper(u32 sic_iwr0, u32 sic_iwr1, u32 sic_iwr2);
-void program_wdog_timer(unsigned long);
-void unmask_wdog_wakeup_evt(void);
-void clear_wdog_wakeup_evt(void);
-void disable_wdog_timer(void);
+void do_hibernate(int wakeup);
+void set_dram_srfs(void);
+void unset_dram_srfs(void);
 
-extern unsigned long get_cclk(void);
-extern unsigned long get_sclk(void);
+#define VRPAIR(vlev, freq) (((vlev) << 16) | ((freq) >> 16))
 
 struct bfin_dpmc_platform_data {
 	const unsigned int *tuple_tab;
@@ -71,8 +26,33 @@ struct bfin_dpmc_platform_data {
 	unsigned short vr_settling_time; /* in us */
 };
 
-#define VRPAIR(vlev, freq) (((vlev) << 16) | ((freq) >> 16))
+#else
 
+#define PM_PUSH(x) \
+	R0 = [P0 + (x - SRAM_BASE_ADDRESS)];\
+	[--SP] =  R0;\
+
+#define PM_POP(x) \
+	R0 = [SP++];\
+	[P0 + (x - SRAM_BASE_ADDRESS)] = R0;\
+
+#define PM_SYS_PUSH(x) \
+	R0 = [P0 + (x - PLL_CTL)];\
+	[--SP] =  R0;\
+
+#define PM_SYS_POP(x) \
+	R0 = [SP++];\
+	[P0 + (x - PLL_CTL)] = R0;\
+
+#define PM_SYS_PUSH16(x) \
+	R0 = w[P0 + (x - PLL_CTL)];\
+	[--SP] =  R0;\
+
+#define PM_SYS_POP16(x) \
+	R0 = [SP++];\
+	w[P0 + (x - PLL_CTL)] = R0;\
+
+#endif
 #endif	/* __KERNEL__ */
 
 #endif	/*_BLACKFIN_DPMC_H_*/
diff --git a/include/asm-blackfin/gpio.h b/include/asm-blackfin/gpio.h
index ff95e9d88342..168f1251eb4d 100644
--- a/include/asm-blackfin/gpio.h
+++ b/include/asm-blackfin/gpio.h
@@ -376,8 +376,12 @@ struct gpio_port_t {
 #endif
 
 #ifdef CONFIG_PM
-unsigned int bfin_pm_setup(void);
-void bfin_pm_restore(void);
+
+unsigned int bfin_pm_standby_setup(void);
+void bfin_pm_standby_restore(void);
+
+void bfin_gpio_pm_hibernate_restore(void);
+void bfin_gpio_pm_hibernate_suspend(void);
 
 #ifndef CONFIG_BF54x
 #define PM_WAKE_RISING	0x1
@@ -392,17 +396,8 @@ void gpio_pm_wakeup_free(unsigned gpio);
 
 struct gpio_port_s {
 	unsigned short data;
-	unsigned short data_clear;
-	unsigned short data_set;
-	unsigned short toggle;
 	unsigned short maska;
-	unsigned short maska_clear;
-	unsigned short maska_set;
-	unsigned short maska_toggle;
 	unsigned short maskb;
-	unsigned short maskb_clear;
-	unsigned short maskb_set;
-	unsigned short maskb_toggle;
 	unsigned short dir;
 	unsigned short polar;
 	unsigned short edge;
@@ -411,10 +406,10 @@ struct gpio_port_s {
 
 	unsigned short fer;
 	unsigned short reserved;
+	unsigned short mux;
 };
 #endif /*CONFIG_BF54x*/
 #endif /*CONFIG_PM*/
-
 /***********************************************************
 *
 * FUNCTIONS: Blackfin GPIO Driver
diff --git a/include/asm-blackfin/mach-bf548/gpio.h b/include/asm-blackfin/mach-bf548/gpio.h
index cb8b0f15c9a6..bba82dc75f16 100644
--- a/include/asm-blackfin/mach-bf548/gpio.h
+++ b/include/asm-blackfin/mach-bf548/gpio.h
@@ -209,3 +209,11 @@ struct gpio_port_t {
 	unsigned short dummy7;
 	unsigned int port_mux;
 };
+
+struct gpio_port_s {
+	unsigned short fer;
+	unsigned short data;
+	unsigned short dir;
+	unsigned short inen;
+	unsigned int mux;
+};

From 2cfebf2bceff4645b403246b608b1bb6222e4deb Mon Sep 17 00:00:00 2001
From: Michael Hennerich <michael.hennerich@analog.com>
Date: Wed, 16 Jul 2008 16:59:05 +0800
Subject: [PATCH 0058/1171] Blackfin arch: Add ANOMALY_05000368 workaround

Possible RETS Register Corruption when Subroutine Is under 5 Cycles in Duration

Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/kernel/entry.S | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/arch/blackfin/kernel/entry.S b/arch/blackfin/kernel/entry.S
index 65f4e67a65c4..31bd9bf3efae 100644
--- a/arch/blackfin/kernel/entry.S
+++ b/arch/blackfin/kernel/entry.S
@@ -64,6 +64,11 @@ ENDPROC(_ret_from_fork)
 
 ENTRY(_sys_fork)
 	r0 = -EINVAL;
+#if (ANOMALY_05000371)
+	nop;
+	nop;
+	nop;
+#endif
 	rts;
 ENDPROC(_sys_fork)
 

From 6546eae4fd90ab11ca7ab6d6b9e1b243d1ce5fe6 Mon Sep 17 00:00:00 2001
From: Jie Zhang <jie.zhang@analog.com>
Date: Tue, 15 Jul 2008 16:15:40 +0800
Subject: [PATCH 0059/1171] Blackfin arch: Allow ptrace to peek and poke
 application data in L1 data SRAM.

Signed-off-by: Jie Zhang <jie.zhang@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/kernel/ptrace.c | 28 ++++++++++++++++++++++++++++
 1 file changed, 28 insertions(+)

diff --git a/arch/blackfin/kernel/ptrace.c b/arch/blackfin/kernel/ptrace.c
index f51ab088098e..bf1a51d8e608 100644
--- a/arch/blackfin/kernel/ptrace.c
+++ b/arch/blackfin/kernel/ptrace.c
@@ -219,6 +219,20 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 				safe_dma_memcpy (&tmp, (const void *)(addr + add), sizeof(tmp));
 				copied = sizeof(tmp);
 			} else
+#endif
+#if L1_DATA_A_LENGTH != 0
+			if (addr + add >= L1_DATA_A_START
+			    && addr + add + sizeof(tmp) <= L1_DATA_A_START + L1_DATA_A_LENGTH) {
+				memcpy(&tmp, (const void *)(addr + add), sizeof(tmp));
+				copied = sizeof(tmp);
+			} else
+#endif
+#if L1_DATA_B_LENGTH != 0
+			if (addr + add >= L1_DATA_B_START
+			    && addr + add + sizeof(tmp) <= L1_DATA_B_START + L1_DATA_B_LENGTH) {
+				memcpy(&tmp, (const void *)(addr + add), sizeof(tmp));
+				copied = sizeof(tmp);
+			} else
 #endif
 			if (addr + add >= FIXED_CODE_START
 			    && addr + add + sizeof(tmp) <= FIXED_CODE_END) {
@@ -289,6 +303,20 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 				safe_dma_memcpy ((void *)(addr + add), &data, sizeof(data));
 				copied = sizeof(data);
 			} else
+#endif
+#if L1_DATA_A_LENGTH != 0
+			if (addr + add >= L1_DATA_A_START
+			    && addr + add + sizeof(data) <= L1_DATA_A_START + L1_DATA_A_LENGTH) {
+				memcpy((void *)(addr + add), &data, sizeof(data));
+				copied = sizeof(data);
+			} else
+#endif
+#if L1_DATA_B_LENGTH != 0
+			if (addr + add >= L1_DATA_B_START
+			    && addr + add + sizeof(data) <= L1_DATA_B_START + L1_DATA_B_LENGTH) {
+				memcpy((void *)(addr + add), &data, sizeof(data));
+				copied = sizeof(data);
+			} else
 #endif
 			if (addr + add >= FIXED_CODE_START
 			    && addr + add + sizeof(data) <= FIXED_CODE_END) {

From 7dee62ac5a3e3f5aa7cc5069fa2d32cec5117229 Mon Sep 17 00:00:00 2001
From: Yi Li <yi.li@analog.com>
Date: Sat, 19 Jul 2008 20:53:46 +0800
Subject: [PATCH 0060/1171] Blackfin arch: Do not need this dualcore test
 module in kernel.

Signed-off-by: Yi Li <yi.li@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/Kconfig.debug          |  7 ----
 arch/blackfin/kernel/Makefile        |  1 -
 arch/blackfin/kernel/dualcore_test.c | 49 ----------------------------
 3 files changed, 57 deletions(-)
 delete mode 100644 arch/blackfin/kernel/dualcore_test.c

diff --git a/arch/blackfin/Kconfig.debug b/arch/blackfin/Kconfig.debug
index c61bdebb9974..c468624d55f0 100644
--- a/arch/blackfin/Kconfig.debug
+++ b/arch/blackfin/Kconfig.debug
@@ -154,13 +154,6 @@ config EARLY_PRINTK
 	  all of this lives in the init section and is thrown away after the
 	  kernel boots completely.
 
-config DUAL_CORE_TEST_MODULE
-	tristate "Dual Core Test Module"
-	depends on (BF561)
-	default n
-	help
-	  Say Y here to build-in dual core test module for dual core test.
-
 config CPLB_INFO
 	bool "Display the CPLB information"
 	help
diff --git a/arch/blackfin/kernel/Makefile b/arch/blackfin/kernel/Makefile
index 6140cd69c782..606adc78aa85 100644
--- a/arch/blackfin/kernel/Makefile
+++ b/arch/blackfin/kernel/Makefile
@@ -18,6 +18,5 @@ endif
 obj-$(CONFIG_BFIN_GPTIMERS)          += gptimers.o
 obj-$(CONFIG_MODULES)                += module.o
 obj-$(CONFIG_BFIN_DMA_5XX)           += bfin_dma_5xx.o
-obj-$(CONFIG_DUAL_CORE_TEST_MODULE)  += dualcore_test.o
 obj-$(CONFIG_KGDB)                   += kgdb.o
 obj-$(CONFIG_EARLY_PRINTK)           += early_printk.o
diff --git a/arch/blackfin/kernel/dualcore_test.c b/arch/blackfin/kernel/dualcore_test.c
deleted file mode 100644
index 0fcba74840b7..000000000000
--- a/arch/blackfin/kernel/dualcore_test.c
+++ /dev/null
@@ -1,49 +0,0 @@
-/*
- * File:         arch/blackfin/kernel/dualcore_test.c
- * Based on:
- * Author:
- *
- * Created:
- * Description:  Small test code for CoreB on a BF561
- *
- * Modified:
- *               Copyright 2004-2006 Analog Devices Inc.
- *
- * Bugs:         Enter bugs at http://blackfin.uclinux.org/
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, see the file COPYING, or write
- * to the Free Software Foundation, Inc.,
- * 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
- */
-
-#include <linux/init.h>
-#include <linux/module.h>
-
-static int *testarg = (int *)0xfeb00000;
-
-static int test_init(void)
-{
-	*testarg = 1;
-	printk(KERN_INFO "Dual core test module inserted: set testarg = [%d]\n @ [%p]\n",
-	       *testarg, testarg);
-	return 0;
-}
-
-static void test_exit(void)
-{
-	printk(KERN_INFO "Dual core test module removed: testarg = [%d]\n", *testarg);
-}
-
-module_init(test_init);
-module_exit(test_exit);

From 1a8caeebe3689ad4ef67d7ff5d4143f7748deedd Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier.adi@gmail.com>
Date: Wed, 16 Jul 2008 17:07:26 +0800
Subject: [PATCH 0061/1171] Blackfin arch: use local labels and ENDPROC()
 markings

Signed-off-by: Mike Frysinger <vapier.adi@gmail.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/mach-common/dpmc_modes.S | 19 +++++++++++++------
 1 file changed, 13 insertions(+), 6 deletions(-)

diff --git a/arch/blackfin/mach-common/dpmc_modes.S b/arch/blackfin/mach-common/dpmc_modes.S
index 46ee77afed20..5e3f1d8a4fb8 100644
--- a/arch/blackfin/mach-common/dpmc_modes.S
+++ b/arch/blackfin/mach-common/dpmc_modes.S
@@ -51,6 +51,7 @@ ENTRY(_sleep_mode)
 	RETS = [SP++];
 	( R7:0, P5:0 ) = [SP++];
 	RTS;
+ENDPROC(_sleep_mode)
 
 ENTRY(_hibernate_mode)
 	[--SP] = ( R7:0, P5:0 );
@@ -75,6 +76,7 @@ ENTRY(_hibernate_mode)
 	IDLE;
 .Lforever:
 	jump .Lforever;
+ENDPROC(_hibernate_mode)
 
 ENTRY(_deep_sleep)
 	[--SP] = ( R7:0, P5:0 );
@@ -130,6 +132,7 @@ ENTRY(_deep_sleep)
 	RETS = [SP++];
 	( R7:0, P5:0 ) = [SP++];
 	RTS;
+ENDPROC(_deep_sleep)
 
 ENTRY(_sleep_deeper)
 	[--SP] = ( R7:0, P5:0 );
@@ -231,7 +234,7 @@ ENTRY(_sleep_deeper)
 	RETS = [SP++];
 	( R7:0, P5:0 ) = [SP++];
 	RTS;
-
+ENDPROC(_sleep_deeper)
 
 ENTRY(_set_dram_srfs)
 	/*  set the dram to self refresh mode */
@@ -270,7 +273,7 @@ ENTRY(_set_dram_srfs)
 	[P0] = R2;
 #endif
 	RTS;
-
+ENDPROC(_set_dram_srfs)
 
 ENTRY(_unset_dram_srfs)
 	/*  set the dram out of self refresh mode */
@@ -297,6 +300,7 @@ ENTRY(_unset_dram_srfs)
 #endif
 	SSYNC;
 	RTS;
+ENDPROC(_unset_dram_srfs)
 
 ENTRY(_set_sic_iwr)
 #if defined(CONFIG_BF54x) || defined(CONFIG_BF52x)  || defined(CONFIG_BF561)
@@ -318,6 +322,7 @@ ENTRY(_set_sic_iwr)
 
 	SSYNC;
 	RTS;
+ENDPROC(_set_sic_iwr)
 
 ENTRY(_set_rtc_istat)
 #ifndef CONFIG_BF561
@@ -332,6 +337,7 @@ ENTRY(_set_rtc_istat)
 	nop;
 #endif
 	RTS;
+ENDPROC(_set_rtc_istat)
 
 ENTRY(_test_pll_locked)
 	P0.H = hi(PLL_STAT);
@@ -341,10 +347,10 @@ ENTRY(_test_pll_locked)
 	CC = BITTST(R0,5);
 	IF !CC JUMP 1b;
 	RTS;
+ENDPROC(_test_pll_locked)
 
 .section .text
 
-
 ENTRY(_do_hibernate)
 	[--SP] = ( R7:0, P5:0 );
 	[--SP] =  RETS;
@@ -593,8 +599,8 @@ ENTRY(_do_hibernate)
 	R0.H = 0xDEAD;	/* Hibernate Magic */
 	R0.L = 0xBEEF;
 	[P0++] = R0;	/* Store Hibernate Magic */
-	R0.H = pm_resume_here;
-	R0.L = pm_resume_here;
+	R0.H = .Lpm_resume_here;
+	R0.L = .Lpm_resume_here;
 	[P0++] = R0;	/* Save Return Address */
 	[P0++] = SP;	/* Save Stack Pointer */
 	P0.H = _hibernate_mode;
@@ -602,7 +608,7 @@ ENTRY(_do_hibernate)
 	R0 = R2;
 	call (P0); /* Goodbye */
 
-pm_resume_here:
+.Lpm_resume_here:
 
 	/* Restore Core Registers */
 	SEQSTAT = [sp++];
@@ -846,3 +852,4 @@ pm_resume_here:
 	RETS = [SP++];
 	( R7:0, P5:0 ) = [SP++];
 	RTS;
+ENDPROC(_do_hibernate)

From 4a5e3638b11978262ab76bbb2062e57fefaaedba Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Tue, 15 Jul 2008 09:42:57 -0600
Subject: [PATCH 0062/1171] ACPI: stop complaints about interrupt link End Tags
 and blank IRQ descriptors

Silently ignore _PRS End Tags.  We already ignore Start Dependent Functions in
_PRS, and we already ignore End Tags in _CRS, so we might as well ignore End
Tags in _PRS as well.

Silently ignore _PRS IRQ descriptors that mention no interrupts.  The spec
allows this (section 6.4.2.1 in ACPI 3.0b spec), and it probably means the
interrupt link can't be configured at all.

This patch doesn't change any functional behavior; it just removes confusing
complaints like these:

    ACPI: Blank IRQ resource
    ACPI: Resource is not an IRQ entry

when parsing _PRS data "23 00 00 18 79 00" from an IBM xSeries 335 dual
Pentium IV Xeon 2.40 GHz machine.  For more details, see
    http://bugzilla.kernel.org/show_bug.cgi?id=11049

The "23 00 00 18" part is a three-byte-long small IRQ resource with no bits set
in the IRQ mask ("00 00"), and level-triggered, active low, shareable ("18").

The "79 00" is an End Tag (type 0x7).  It is superfluous since there is no
Start Dependent Function tag and there are no resources after it, but it is
harmless.

Thanks to Gabriele Trombetti <g.trombetti.lkrnl1213@logicschema.com>
(aka Kurk) for reporting this and testing the patch.

Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
---
 drivers/acpi/pci_link.c | 31 ++++++++++++++++++++-----------
 1 file changed, 20 insertions(+), 11 deletions(-)

diff --git a/drivers/acpi/pci_link.c b/drivers/acpi/pci_link.c
index 233c40c51684..89f3b2abfdc7 100644
--- a/drivers/acpi/pci_link.c
+++ b/drivers/acpi/pci_link.c
@@ -113,20 +113,23 @@ acpi_pci_link_check_possible(struct acpi_resource *resource, void *context)
 
 	switch (resource->type) {
 	case ACPI_RESOURCE_TYPE_START_DEPENDENT:
+	case ACPI_RESOURCE_TYPE_END_TAG:
 		return AE_OK;
 	case ACPI_RESOURCE_TYPE_IRQ:
 		{
 			struct acpi_resource_irq *p = &resource->data.irq;
 			if (!p || !p->interrupt_count) {
-				printk(KERN_WARNING PREFIX "Blank IRQ resource\n");
+				ACPI_DEBUG_PRINT((ACPI_DB_INFO,
+						  "Blank _PRS IRQ resource\n"));
 				return AE_OK;
 			}
 			for (i = 0;
 			     (i < p->interrupt_count
 			      && i < ACPI_PCI_LINK_MAX_POSSIBLE); i++) {
 				if (!p->interrupts[i]) {
-					printk(KERN_WARNING PREFIX "Invalid IRQ %d\n",
-						      p->interrupts[i]);
+					printk(KERN_WARNING PREFIX
+					       "Invalid _PRS IRQ %d\n",
+					       p->interrupts[i]);
 					continue;
 				}
 				link->irq.possible[i] = p->interrupts[i];
@@ -143,15 +146,16 @@ acpi_pci_link_check_possible(struct acpi_resource *resource, void *context)
 			    &resource->data.extended_irq;
 			if (!p || !p->interrupt_count) {
 				printk(KERN_WARNING PREFIX
-					      "Blank EXT IRQ resource\n");
+					      "Blank _PRS EXT IRQ resource\n");
 				return AE_OK;
 			}
 			for (i = 0;
 			     (i < p->interrupt_count
 			      && i < ACPI_PCI_LINK_MAX_POSSIBLE); i++) {
 				if (!p->interrupts[i]) {
-					printk(KERN_WARNING PREFIX "Invalid IRQ %d\n",
-						      p->interrupts[i]);
+					printk(KERN_WARNING PREFIX
+					       "Invalid _PRS IRQ %d\n",
+					       p->interrupts[i]);
 					continue;
 				}
 				link->irq.possible[i] = p->interrupts[i];
@@ -163,7 +167,8 @@ acpi_pci_link_check_possible(struct acpi_resource *resource, void *context)
 			break;
 		}
 	default:
-		printk(KERN_ERR PREFIX "Resource is not an IRQ entry\n");
+		printk(KERN_ERR PREFIX "_PRS resource type 0x%x isn't an IRQ\n",
+		       resource->type);
 		return AE_OK;
 	}
 
@@ -199,6 +204,9 @@ acpi_pci_link_check_current(struct acpi_resource *resource, void *context)
 
 
 	switch (resource->type) {
+	case ACPI_RESOURCE_TYPE_START_DEPENDENT:
+	case ACPI_RESOURCE_TYPE_END_TAG:
+		return AE_OK;
 	case ACPI_RESOURCE_TYPE_IRQ:
 		{
 			struct acpi_resource_irq *p = &resource->data.irq;
@@ -208,7 +216,7 @@ acpi_pci_link_check_current(struct acpi_resource *resource, void *context)
 				 * particularly those those w/ _STA disabled
 				 */
 				ACPI_DEBUG_PRINT((ACPI_DB_INFO,
-						  "Blank IRQ resource\n"));
+						  "Blank _CRS IRQ resource\n"));
 				return AE_OK;
 			}
 			*irq = p->interrupts[0];
@@ -224,7 +232,7 @@ acpi_pci_link_check_current(struct acpi_resource *resource, void *context)
 				 * return at least 1 IRQ
 				 */
 				printk(KERN_WARNING PREFIX
-					      "Blank EXT IRQ resource\n");
+					      "Blank _CRS EXT IRQ resource\n");
 				return AE_OK;
 			}
 			*irq = p->interrupts[0];
@@ -232,10 +240,11 @@ acpi_pci_link_check_current(struct acpi_resource *resource, void *context)
 		}
 		break;
 	default:
-		printk(KERN_ERR PREFIX "Resource %d isn't an IRQ\n", resource->type);
-	case ACPI_RESOURCE_TYPE_END_TAG:
+		printk(KERN_ERR PREFIX "_CRS resource type 0x%x isn't an IRQ\n",
+		       resource->type);
 		return AE_OK;
 	}
+
 	return AE_CTRL_TERMINATE;
 }
 

From c2c789057f075022658b38b498755c29c1ba8055 Mon Sep 17 00:00:00 2001
From: Zhao Yakui <yakui.zhao@intel.com>
Date: Thu, 17 Jul 2008 10:46:05 +0800
Subject: [PATCH 0063/1171] ACPI: Ignore _BQC object when registering backlight
 device

According to acpi spec , the objectes of  _BCL and _BCM are required if
integrated LCD is present and supports brightness level and the _BQC is
the optional object. So the _BQC object will be ignored when the backlight
device is registered.
At the same time when there is no _BQC object, the current brightness will be
set to the maximum.

http://bugzilla.kernel.org/show_bug.cgi?id=10206

Signed-off-by: Zhao Yakui <yakui.zhao@intel.com>
Signed-off-by: Zhang Rui  <rui.zhang@intel.com>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
---
 drivers/acpi/video.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/acpi/video.c b/drivers/acpi/video.c
index 64c889331f3b..e32b6c14d928 100644
--- a/drivers/acpi/video.c
+++ b/drivers/acpi/video.c
@@ -741,7 +741,7 @@ static void acpi_video_device_find_cap(struct acpi_video_device *device)
 
 	max_level = acpi_video_init_brightness(device);
 
-	if (device->cap._BCL && device->cap._BCM && device->cap._BQC && max_level > 0){
+	if (device->cap._BCL && device->cap._BCM && max_level > 0) {
 		int result;
 		static int count = 0;
 		char *name;
@@ -753,7 +753,17 @@ static void acpi_video_device_find_cap(struct acpi_video_device *device)
 		device->backlight = backlight_device_register(name,
 			NULL, device, &acpi_backlight_ops);
 		device->backlight->props.max_brightness = device->brightness->count-3;
-		device->backlight->props.brightness = acpi_video_get_brightness(device->backlight);
+		/*
+		 * If there exists the _BQC object, the _BQC object will be
+		 * called to get the current backlight brightness. Otherwise
+		 * the brightness will be set to the maximum.
+		 */
+		if (device->cap._BQC)
+			device->backlight->props.brightness =
+				acpi_video_get_brightness(device->backlight);
+		else
+			device->backlight->props.brightness =
+				device->backlight->props.max_brightness;
 		backlight_update_status(device->backlight);
 		kfree(name);
 

From ea51011a27db48ea0a80a5e20de3969b292d5d4d Mon Sep 17 00:00:00 2001
From: Zhao Yakui <yakui.zhao@intel.com>
Date: Mon, 14 Jul 2008 15:14:03 +0800
Subject: [PATCH 0064/1171] ACPI : Set FAN device to correct state in boot
 phase

Subject:ACPI: Set FAN device to correct state in boot phase
From: Zhao Yakui <yakui.zhao@intel.com>

On some laptops when ACPI FAN driver is loaded, maybe the FAN device will be
turned on. But if the temperature is below the threshold, the corresponding
FAN device should be turned off in the course of loading thermal driver.

So it is necessary to set the FAN device to the correct state in course of loading
the thermal driver.

http://bugzilla.kernel.org/show_bug.cgi?id=8049

Signed-off-by: Zhao Yakui <yakui.zhao@intel.com>
Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
---
 drivers/acpi/thermal.c | 43 ++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 43 insertions(+)

diff --git a/drivers/acpi/thermal.c b/drivers/acpi/thermal.c
index 84c795fb9b1e..9adfd180df6a 100644
--- a/drivers/acpi/thermal.c
+++ b/drivers/acpi/thermal.c
@@ -769,6 +769,47 @@ static void acpi_thermal_run(unsigned long data)
 		acpi_os_execute(OSL_GPE_HANDLER, acpi_thermal_check, (void *)data);
 }
 
+static void acpi_thermal_active_off(void *data)
+{
+	int result = 0;
+	struct acpi_thermal *tz = data;
+	int i = 0;
+	int j = 0;
+	struct acpi_thermal_active *active = NULL;
+
+	if (!tz) {
+		printk(KERN_ERR PREFIX "Invalid (NULL) context\n");
+		return;
+	}
+
+	result = acpi_thermal_get_temperature(tz);
+	if (result)
+		return;
+
+	for (i = 0; i < ACPI_THERMAL_MAX_ACTIVE; i++) {
+		active = &(tz->trips.active[i]);
+		if (!active || !active->flags.valid)
+			break;
+		if (tz->temperature >= active->temperature) {
+			/*
+			 * If the thermal temperature is greater than the
+			 * active threshod, unnecessary to turn off the
+			 * the active cooling device.
+			 */
+			continue;
+		}
+		/*
+		 * Below Threshold?
+		 * ----------------
+		 * Turn OFF all cooling devices associated with this
+		 * threshold.
+		 */
+		for (j = 0; j < active->devices.count; j++)
+			result = acpi_bus_set_power(active->devices.handles[j],
+						    ACPI_STATE_D3);
+	}
+}
+
 static void acpi_thermal_check(void *data)
 {
 	int result = 0;
@@ -1624,6 +1665,8 @@ static int acpi_thermal_add(struct acpi_device *device)
 
 	init_timer(&tz->timer);
 
+	acpi_thermal_active_off(tz);
+
 	acpi_thermal_check(tz);
 
 	status = acpi_install_notify_handler(device->handle,

From b1d77fae0c429d1be84ca0c9e627d9ab0e2a6d0b Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@linux.intel.com>
Date: Fri, 18 Jul 2008 01:42:20 +0200
Subject: [PATCH 0065/1171] Revert "Fix FADT parsing"

This reverts commit 01a5bba576b9364b33f61f0cd9fa70c2cf5535e2.

There seem to be some FADTs around with bogus information
in the v2 fields. Revert this patch for now until
this can be properly resolved.

Signed-off-by: Andi Kleen <ak@linux.intel.com>
---
 drivers/acpi/tables/tbfadt.c | 17 ++++++-----------
 1 file changed, 6 insertions(+), 11 deletions(-)

diff --git a/drivers/acpi/tables/tbfadt.c b/drivers/acpi/tables/tbfadt.c
index ccb5b64bbef3..a4a41ba2484b 100644
--- a/drivers/acpi/tables/tbfadt.c
+++ b/drivers/acpi/tables/tbfadt.c
@@ -124,7 +124,7 @@ static struct acpi_fadt_info fadt_info_table[] = {
 
 static void inline
 acpi_tb_init_generic_address(struct acpi_generic_address *generic_address,
-			     u8 byte_width, u64 address)
+			     u8 bit_width, u64 address)
 {
 
 	/*
@@ -136,7 +136,7 @@ acpi_tb_init_generic_address(struct acpi_generic_address *generic_address,
 	/* All other fields are byte-wide */
 
 	generic_address->space_id = ACPI_ADR_SPACE_SYSTEM_IO;
-	generic_address->bit_width = byte_width << 3;
+	generic_address->bit_width = bit_width;
 	generic_address->bit_offset = 0;
 	generic_address->access_width = 0;
 }
@@ -343,11 +343,9 @@ static void acpi_tb_convert_fadt(void)
 	 *
 	 * The PM event blocks are split into two register blocks, first is the
 	 * PM Status Register block, followed immediately by the PM Enable Register
-	 * block. Each is of length (xpm1x_event_block.bit_width/2)
+	 * block. Each is of length (pm1_event_length/2)
 	 */
-	WARN_ON(ACPI_MOD_16(acpi_gbl_FADT.xpm1a_event_block.bit_width));
-	pm1_register_length = (u8) ACPI_DIV_16(acpi_gbl_FADT
-					       .xpm1a_event_block.bit_width);
+	pm1_register_length = (u8) ACPI_DIV_2(acpi_gbl_FADT.pm1_event_length);
 
 	/* The PM1A register block is required */
 
@@ -362,17 +360,14 @@ static void acpi_tb_convert_fadt(void)
 	/* The PM1B register block is optional, ignore if not present */
 
 	if (acpi_gbl_FADT.xpm1b_event_block.address) {
-		WARN_ON(ACPI_MOD_16(acpi_gbl_FADT.xpm1b_event_block.bit_width));
-		pm1_register_length = (u8) ACPI_DIV_16(acpi_gbl_FADT
-						       .xpm1b_event_block
-						       .bit_width);
 		acpi_tb_init_generic_address(&acpi_gbl_xpm1b_enable,
 					     pm1_register_length,
 					     (acpi_gbl_FADT.xpm1b_event_block.
 					      address + pm1_register_length));
 		/* Don't forget to copy space_id of the GAS */
 		acpi_gbl_xpm1b_enable.space_id =
-		    acpi_gbl_FADT.xpm1b_event_block.space_id;
+		    acpi_gbl_FADT.xpm1a_event_block.space_id;
+
 	}
 }
 

From e1469c34eb623cd1945ef09bfd7de7bc2f9ff6b3 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@linux.intel.com>
Date: Fri, 18 Jul 2008 01:43:08 +0200
Subject: [PATCH 0066/1171] Revert "dock: bay: Don't call acpi_walk_namespace()
 when ACPI is disabled."

Revert double commit by mistake. Noticed by Thomas Gleixner.

This reverts commit cc7e51666d82aedfd6b9a033ca1a10d71c21f1ca.

Signed-off-by: Andi Kleen <ak@linux.intel.com>
---
 drivers/acpi/bay.c  | 3 ---
 drivers/acpi/dock.c | 3 ---
 2 files changed, 6 deletions(-)

diff --git a/drivers/acpi/bay.c b/drivers/acpi/bay.c
index e6caf5d42e0e..61b6c5beb2d3 100644
--- a/drivers/acpi/bay.c
+++ b/drivers/acpi/bay.c
@@ -377,9 +377,6 @@ static int __init bay_init(void)
 
 	INIT_LIST_HEAD(&drive_bays);
 
-	if (acpi_disabled)
-		return -ENODEV;
-
 	if (acpi_disabled)
 		return -ENODEV;
 
diff --git a/drivers/acpi/dock.c b/drivers/acpi/dock.c
index 1e872e79db33..bb7c51f712bd 100644
--- a/drivers/acpi/dock.c
+++ b/drivers/acpi/dock.c
@@ -917,9 +917,6 @@ static int __init dock_init(void)
 
 	dock_station = NULL;
 
-	if (acpi_disabled)
-		return 0;
-
 	if (acpi_disabled)
 		return 0;
 

From 725c3a2d70f958adee807c178178819a50f68a56 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@linux.intel.com>
Date: Fri, 18 Jul 2008 09:12:49 +0200
Subject: [PATCH 0067/1171] Revert "ACPI: don't walk tables if ACPI was
 disabled"

This reverts commit d1857056904d5f313f11184fcfa624652ff9620a.

Double commit, noticed by Thomas Gleixner.

Signed-off-by: Andi Kleen <ak@linux.intel.com>
---
 drivers/acpi/glue.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/drivers/acpi/glue.c b/drivers/acpi/glue.c
index 0f2dd81736bd..2f173e83f8a7 100644
--- a/drivers/acpi/glue.c
+++ b/drivers/acpi/glue.c
@@ -335,9 +335,6 @@ static int __init acpi_rtc_init(void)
 {
 	struct device *dev = get_rtc_dev();
 
-	if (acpi_disabled)
-		return 0;
-
 	if (acpi_disabled)
 		return 0;
 

From e59494f441c834ca7aaa0e6fa6678ddbd3d72743 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <rostedt@goodmis.org>
Date: Wed, 16 Jul 2008 00:13:45 -0400
Subject: [PATCH 0068/1171] ftrace: fix 4d3702b6 (post-v2.6.26): WARNING: at
 kernel/lockdep.c:2731 check_flags (ftrace)

On Wed, 16 Jul 2008, Vegard Nossum wrote:

> When booting 4d3702b6, I got this huge thing:
>
> Testing tracer wakeup: <4>------------[ cut here ]------------
> WARNING: at kernel/lockdep.c:2731 check_flags+0x123/0x160()
> Modules linked in:
> Pid: 1, comm: swapper Not tainted 2.6.26-crashing-02127-g4d3702b6 #30
>  [<c015c349>] warn_on_slowpath+0x59/0xb0
>  [<c01276c6>] ? ftrace_call+0x5/0x8
>  [<c012d800>] ? native_read_tsc+0x0/0x20
>  [<c0158de2>] ? sub_preempt_count+0x12/0xf0
>  [<c01814eb>] ? trace_hardirqs_off+0xb/0x10
>  [<c0182fbc>] ? __lock_acquire+0x2cc/0x1120
>  [<c01814eb>] ? trace_hardirqs_off+0xb/0x10
>  [<c01276af>] ? mcount_call+0x5/0xa
>  [<c017ff53>] check_flags+0x123/0x160
>  [<c0183e61>] lock_acquire+0x51/0xd0
>  [<c01276c6>] ? ftrace_call+0x5/0x8
>  [<c0613d4f>] _spin_lock_irqsave+0x5f/0xa0
>  [<c01a8d45>] ? ftrace_record_ip+0xf5/0x220
>  [<c02d5413>] ? debug_locks_off+0x3/0x50
>  [<c01a8d45>] ftrace_record_ip+0xf5/0x220
>  [<c01276af>] mcount_call+0x5/0xa
>  [<c02d5418>] ? debug_locks_off+0x8/0x50
>  [<c017ff27>] check_flags+0xf7/0x160
>  [<c0183e61>] lock_acquire+0x51/0xd0
>  [<c01276c6>] ? ftrace_call+0x5/0x8
>  [<c0613d4f>] _spin_lock_irqsave+0x5f/0xa0
>  [<c01affcd>] ? wakeup_tracer_call+0x6d/0xf0
>  [<c01625e2>] ? _local_bh_enable+0x62/0xb0
>  [<c0158ddd>] ? sub_preempt_count+0xd/0xf0
>  [<c01affcd>] wakeup_tracer_call+0x6d/0xf0
>  [<c0162724>] ? __do_softirq+0xf4/0x110
>  [<c01afff1>] ? wakeup_tracer_call+0x91/0xf0
>  [<c01276c6>] ftrace_call+0x5/0x8
>  [<c0162724>] ? __do_softirq+0xf4/0x110
>  [<c0158de2>] ? sub_preempt_count+0x12/0xf0
>  [<c01625e2>] _local_bh_enable+0x62/0xb0
>  [<c0162724>] __do_softirq+0xf4/0x110
>  [<c01627ed>] do_softirq+0xad/0xb0
>  [<c0162a15>] irq_exit+0xa5/0xb0
>  [<c013a506>] smp_apic_timer_interrupt+0x66/0xa0
>  [<c02d3fac>] ? trace_hardirqs_off_thunk+0xc/0x10
>  [<c0127449>] apic_timer_interrupt+0x2d/0x34
>  [<c018007b>] ? find_usage_backwards+0xb/0xf0
>  [<c0613a09>] ? _spin_unlock_irqrestore+0x69/0x80
>  [<c014ef32>] tg_shares_up+0x132/0x1d0
>  [<c014d2a2>] walk_tg_tree+0x62/0xa0
>  [<c014ee00>] ? tg_shares_up+0x0/0x1d0
>  [<c014a860>] ? tg_nop+0x0/0x10
>  [<c015499d>] update_shares+0x5d/0x80
>  [<c0154a2f>] try_to_wake_up+0x6f/0x280
>  [<c01a8b90>] ? __ftrace_modify_code+0x0/0xc0
>  [<c01a8b90>] ? __ftrace_modify_code+0x0/0xc0
>  [<c0154c94>] wake_up_process+0x14/0x20
>  [<c01725f6>] kthread_create+0x66/0xb0
>  [<c0195400>] ? do_stop+0x0/0x200
>  [<c0195320>] ? __stop_machine_run+0x30/0xb0
>  [<c0195340>] __stop_machine_run+0x50/0xb0
>  [<c0195400>] ? do_stop+0x0/0x200
>  [<c01a8b90>] ? __ftrace_modify_code+0x0/0xc0
>  [<c061242d>] ? mutex_unlock+0xd/0x10
>  [<c01953cc>] stop_machine_run+0x2c/0x60
>  [<c01a94d3>] unregister_ftrace_function+0x103/0x180
>  [<c01b0517>] stop_wakeup_tracer+0x17/0x60
>  [<c01b056f>] wakeup_tracer_ctrl_update+0xf/0x30
>  [<c01ab8d5>] trace_selftest_startup_wakeup+0xb5/0x130
>  [<c01ab950>] ? trace_wakeup_test_thread+0x0/0x70
>  [<c01aadf5>] register_tracer+0x135/0x1b0
>  [<c0877d02>] init_wakeup_tracer+0xd/0xf
>  [<c085d437>] kernel_init+0x1a9/0x2ce
>  [<c061397b>] ? _spin_unlock_irq+0x3b/0x60
>  [<c02d3f9c>] ? trace_hardirqs_on_thunk+0xc/0x10
>  [<c0877cf5>] ? init_wakeup_tracer+0x0/0xf
>  [<c0182646>] ? trace_hardirqs_on_caller+0x126/0x180
>  [<c02d3f9c>] ? trace_hardirqs_on_thunk+0xc/0x10
>  [<c01269c8>] ? restore_nocheck_notrace+0x0/0xe
>  [<c085d28e>] ? kernel_init+0x0/0x2ce
>  [<c085d28e>] ? kernel_init+0x0/0x2ce
>  [<c01275fb>] kernel_thread_helper+0x7/0x10
>  =======================
> ---[ end trace a7919e7f17c0a725 ]---
> irq event stamp: 579530
> hardirqs last  enabled at (579528): [<c01826ab>] trace_hardirqs_on+0xb/0x10
> hardirqs last disabled at (579529): [<c01814eb>] trace_hardirqs_off+0xb/0x10
> softirqs last  enabled at (579530): [<c0162724>] __do_softirq+0xf4/0x110
> softirqs last disabled at (579517): [<c01627ed>] do_softirq+0xad/0xb0
> irq event stamp: 579530
> hardirqs last  enabled at (579528): [<c01826ab>] trace_hardirqs_on+0xb/0x10
> hardirqs last disabled at (579529): [<c01814eb>] trace_hardirqs_off+0xb/0x10
> softirqs last  enabled at (579530): [<c0162724>] __do_softirq+0xf4/0x110
> softirqs last disabled at (579517): [<c01627ed>] do_softirq+0xad/0xb0
> PASSED
>
> Incidentally, the kernel also hung while I was typing in this report.

Things get weird between lockdep and ftrace because ftrace can be called
within lockdep internal code (via the mcount pointer) and lockdep can be
called with ftrace (via spin_locks).

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Tested-by: Vegard Nossum <vegard.nossum@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/trace/trace_sched_wakeup.c | 27 ++++++++++++++++-----------
 1 file changed, 16 insertions(+), 11 deletions(-)

diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c
index 3c8d61df4474..e303ccb62cdf 100644
--- a/kernel/trace/trace_sched_wakeup.c
+++ b/kernel/trace/trace_sched_wakeup.c
@@ -26,7 +26,8 @@ static struct task_struct	*wakeup_task;
 static int			wakeup_cpu;
 static unsigned			wakeup_prio = -1;
 
-static DEFINE_SPINLOCK(wakeup_lock);
+static raw_spinlock_t wakeup_lock =
+	(raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED;
 
 static void __wakeup_reset(struct trace_array *tr);
 
@@ -56,7 +57,8 @@ wakeup_tracer_call(unsigned long ip, unsigned long parent_ip)
 	if (unlikely(disabled != 1))
 		goto out;
 
-	spin_lock_irqsave(&wakeup_lock, flags);
+	local_irq_save(flags);
+	__raw_spin_lock(&wakeup_lock);
 
 	if (unlikely(!wakeup_task))
 		goto unlock;
@@ -71,7 +73,8 @@ wakeup_tracer_call(unsigned long ip, unsigned long parent_ip)
 	trace_function(tr, data, ip, parent_ip, flags);
 
  unlock:
-	spin_unlock_irqrestore(&wakeup_lock, flags);
+	__raw_spin_unlock(&wakeup_lock);
+	local_irq_restore(flags);
 
  out:
 	atomic_dec(&data->disabled);
@@ -145,7 +148,8 @@ wakeup_sched_switch(void *private, void *rq, struct task_struct *prev,
 	if (likely(disabled != 1))
 		goto out;
 
-	spin_lock_irqsave(&wakeup_lock, flags);
+	local_irq_save(flags);
+	__raw_spin_lock(&wakeup_lock);
 
 	/* We could race with grabbing wakeup_lock */
 	if (unlikely(!tracer_enabled || next != wakeup_task))
@@ -174,7 +178,8 @@ wakeup_sched_switch(void *private, void *rq, struct task_struct *prev,
 
 out_unlock:
 	__wakeup_reset(tr);
-	spin_unlock_irqrestore(&wakeup_lock, flags);
+	__raw_spin_unlock(&wakeup_lock);
+	local_irq_restore(flags);
 out:
 	atomic_dec(&tr->data[cpu]->disabled);
 }
@@ -209,8 +214,6 @@ static void __wakeup_reset(struct trace_array *tr)
 	struct trace_array_cpu *data;
 	int cpu;
 
-	assert_spin_locked(&wakeup_lock);
-
 	for_each_possible_cpu(cpu) {
 		data = tr->data[cpu];
 		tracing_reset(data);
@@ -229,9 +232,11 @@ static void wakeup_reset(struct trace_array *tr)
 {
 	unsigned long flags;
 
-	spin_lock_irqsave(&wakeup_lock, flags);
+	local_irq_save(flags);
+	__raw_spin_lock(&wakeup_lock);
 	__wakeup_reset(tr);
-	spin_unlock_irqrestore(&wakeup_lock, flags);
+	__raw_spin_unlock(&wakeup_lock);
+	local_irq_restore(flags);
 }
 
 static void
@@ -252,7 +257,7 @@ wakeup_check_start(struct trace_array *tr, struct task_struct *p,
 		goto out;
 
 	/* interrupts should be off from try_to_wake_up */
-	spin_lock(&wakeup_lock);
+	__raw_spin_lock(&wakeup_lock);
 
 	/* check for races. */
 	if (!tracer_enabled || p->prio >= wakeup_prio)
@@ -274,7 +279,7 @@ wakeup_check_start(struct trace_array *tr, struct task_struct *p,
 		       CALLER_ADDR1, CALLER_ADDR2, flags);
 
 out_locked:
-	spin_unlock(&wakeup_lock);
+	__raw_spin_unlock(&wakeup_lock);
 out:
 	atomic_dec(&tr->data[cpu]->disabled);
 }

From 8b95d9172be7146c87e7a998310ce2919c851adc Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Mon, 14 Jul 2008 23:32:32 +0200
Subject: [PATCH 0069/1171] fix core/stacktrace changes on avr32, mips, sh

Fixes this type of problem:

  CC      arch/s390/kernel/stacktrace.o
arch/s390/kernel/stacktrace.c:84: warning: data definition has no type or storage class
arch/s390/kernel/stacktrace.c:84: warning: type defaults to 'int' in declaration of 'EXPORT_SYMBOL_GPL'
arch/s390/kernel/stacktrace.c:84: warning: parameter names (without types) in function declaration
arch/s390/kernel/stacktrace.c:97: warning: data definition has no type or storage class
arch/s390/kernel/stacktrace.c:97: warning: type defaults to 'int' in declaration of 'EXPORT_SYMBOL_GPL'
arch/s390/kernel/stacktrace.c:97: warning: parameter names (without types) in function declaration

caused by "stacktrace: export save_stack_trace[_tsk]"

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/avr32/kernel/stacktrace.c | 1 +
 arch/mips/kernel/stacktrace.c  | 1 +
 arch/sh/kernel/stacktrace.c    | 1 +
 3 files changed, 3 insertions(+)

diff --git a/arch/avr32/kernel/stacktrace.c b/arch/avr32/kernel/stacktrace.c
index f4bdb448049c..c09f0d8dd679 100644
--- a/arch/avr32/kernel/stacktrace.c
+++ b/arch/avr32/kernel/stacktrace.c
@@ -10,6 +10,7 @@
 #include <linux/sched.h>
 #include <linux/stacktrace.h>
 #include <linux/thread_info.h>
+#include <linux/module.h>
 
 register unsigned long current_frame_pointer asm("r7");
 
diff --git a/arch/mips/kernel/stacktrace.c b/arch/mips/kernel/stacktrace.c
index 5eb4681a73d2..702e2e92a1ca 100644
--- a/arch/mips/kernel/stacktrace.c
+++ b/arch/mips/kernel/stacktrace.c
@@ -7,6 +7,7 @@
  */
 #include <linux/sched.h>
 #include <linux/stacktrace.h>
+#include <linux/module.h
 #include <asm/stacktrace.h>
 
 /*
diff --git a/arch/sh/kernel/stacktrace.c b/arch/sh/kernel/stacktrace.c
index 1b2ae35c4a76..54d1f61aa007 100644
--- a/arch/sh/kernel/stacktrace.c
+++ b/arch/sh/kernel/stacktrace.c
@@ -12,6 +12,7 @@
 #include <linux/sched.h>
 #include <linux/stacktrace.h>
 #include <linux/thread_info.h>
+#include <linux/module.h>
 #include <asm/ptrace.h>
 
 /*

From b8f8c3cf0a4ac0632ec3f0e15e9dc0c29de917af Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 18 Jul 2008 17:27:28 +0200
Subject: [PATCH 0070/1171] nohz: prevent tick stop outside of the idle loop

Jack Ren and Eric Miao tracked down the following long standing
problem in the NOHZ code:

	scheduler switch to idle task
	enable interrupts

Window starts here

	----> interrupt happens (does not set NEED_RESCHED)
	      	irq_exit() stops the tick

	----> interrupt happens (does set NEED_RESCHED)

	return from schedule()

	cpu_idle(): preempt_disable();

Window ends here

The interrupts can happen at any point inside the race window. The
first interrupt stops the tick, the second one causes the scheduler to
rerun and switch away from idle again and we end up with the tick
disabled.

The fact that it needs two interrupts where the first one does not set
NEED_RESCHED and the second one does made the bug obscure and extremly
hard to reproduce and analyse. Kudos to Jack and Eric.

Solution: Limit the NOHZ functionality to the idle loop to make sure
that we can not run into such a situation ever again.

cpu_idle()
{
	preempt_disable();

	while(1) {
		 tick_nohz_stop_sched_tick(1); <- tell NOHZ code that we
		 			          are in the idle loop

		 while (!need_resched())
		       halt();

		 tick_nohz_restart_sched_tick(); <- disables NOHZ mode
		 preempt_enable_no_resched();
		 schedule();
		 preempt_disable();
	}
}

In hindsight we should have done this forever, but ...

/me grabs a large brown paperbag.

Debugged-by: Jack Ren <jack.ren@marvell.com>,
Debugged-by: eric miao <eric.y.miao@gmail.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/arm/kernel/process.c              |  2 +-
 arch/avr32/kernel/process.c            |  2 +-
 arch/blackfin/kernel/process.c         |  2 +-
 arch/mips/kernel/process.c             |  2 +-
 arch/powerpc/kernel/idle.c             |  2 +-
 arch/powerpc/platforms/iseries/setup.c |  4 ++--
 arch/sh/kernel/process_32.c            |  2 +-
 arch/sparc64/kernel/process.c          |  2 +-
 arch/um/kernel/process.c               |  2 +-
 arch/x86/kernel/process_32.c           |  2 +-
 arch/x86/kernel/process_64.c           |  2 +-
 include/linux/tick.h                   |  5 +++--
 kernel/softirq.c                       |  2 +-
 kernel/time/tick-sched.c               | 12 ++++++++++--
 14 files changed, 26 insertions(+), 17 deletions(-)

diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c
index 46bf2ede6128..84f5a4c778fb 100644
--- a/arch/arm/kernel/process.c
+++ b/arch/arm/kernel/process.c
@@ -164,7 +164,7 @@ void cpu_idle(void)
 		if (!idle)
 			idle = default_idle;
 		leds_event(led_idle_start);
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		while (!need_resched())
 			idle();
 		leds_event(led_idle_end);
diff --git a/arch/avr32/kernel/process.c b/arch/avr32/kernel/process.c
index 6cf9df176274..ff820a9e743a 100644
--- a/arch/avr32/kernel/process.c
+++ b/arch/avr32/kernel/process.c
@@ -31,7 +31,7 @@ void cpu_idle(void)
 {
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		while (!need_resched())
 			cpu_idle_sleep();
 		tick_nohz_restart_sched_tick();
diff --git a/arch/blackfin/kernel/process.c b/arch/blackfin/kernel/process.c
index 53c2cd255441..77800dd83e57 100644
--- a/arch/blackfin/kernel/process.c
+++ b/arch/blackfin/kernel/process.c
@@ -105,7 +105,7 @@ void cpu_idle(void)
 #endif
 		if (!idle)
 			idle = default_idle;
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		while (!need_resched())
 			idle();
 		tick_nohz_restart_sched_tick();
diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c
index 2c09a442e5e5..bdead3aad253 100644
--- a/arch/mips/kernel/process.c
+++ b/arch/mips/kernel/process.c
@@ -53,7 +53,7 @@ void __noreturn cpu_idle(void)
 {
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		while (!need_resched()) {
 #ifdef CONFIG_SMTC_IDLE_HOOK_DEBUG
 			extern void smtc_idle_loop_hook(void);
diff --git a/arch/powerpc/kernel/idle.c b/arch/powerpc/kernel/idle.c
index c3cf0e8f3ac1..d308a9f70f1b 100644
--- a/arch/powerpc/kernel/idle.c
+++ b/arch/powerpc/kernel/idle.c
@@ -60,7 +60,7 @@ void cpu_idle(void)
 
 	set_thread_flag(TIF_POLLING_NRFLAG);
 	while (1) {
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		while (!need_resched() && !cpu_should_die()) {
 			ppc64_runlatch_off();
 
diff --git a/arch/powerpc/platforms/iseries/setup.c b/arch/powerpc/platforms/iseries/setup.c
index b72120751bbe..70b688c1aefb 100644
--- a/arch/powerpc/platforms/iseries/setup.c
+++ b/arch/powerpc/platforms/iseries/setup.c
@@ -561,7 +561,7 @@ static void yield_shared_processor(void)
 static void iseries_shared_idle(void)
 {
 	while (1) {
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		while (!need_resched() && !hvlpevent_is_pending()) {
 			local_irq_disable();
 			ppc64_runlatch_off();
@@ -591,7 +591,7 @@ static void iseries_dedicated_idle(void)
 	set_thread_flag(TIF_POLLING_NRFLAG);
 
 	while (1) {
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		if (!need_resched()) {
 			while (!need_resched()) {
 				ppc64_runlatch_off();
diff --git a/arch/sh/kernel/process_32.c b/arch/sh/kernel/process_32.c
index b98e37a1f54c..921892c351da 100644
--- a/arch/sh/kernel/process_32.c
+++ b/arch/sh/kernel/process_32.c
@@ -86,7 +86,7 @@ void cpu_idle(void)
 		if (!idle)
 			idle = default_idle;
 
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		while (!need_resched())
 			idle();
 		tick_nohz_restart_sched_tick();
diff --git a/arch/sparc64/kernel/process.c b/arch/sparc64/kernel/process.c
index 2084f81a76e1..0798928ba361 100644
--- a/arch/sparc64/kernel/process.c
+++ b/arch/sparc64/kernel/process.c
@@ -97,7 +97,7 @@ void cpu_idle(void)
 	set_thread_flag(TIF_POLLING_NRFLAG);
 
 	while(1) {
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 
 		while (!need_resched() && !cpu_is_offline(cpu))
 			sparc64_yield(cpu);
diff --git a/arch/um/kernel/process.c b/arch/um/kernel/process.c
index 83603cfbde81..a1c6d07cac3e 100644
--- a/arch/um/kernel/process.c
+++ b/arch/um/kernel/process.c
@@ -243,7 +243,7 @@ void default_idle(void)
 		if (need_resched())
 			schedule();
 
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		nsecs = disable_timer();
 		idle_sleep(nsecs);
 		tick_nohz_restart_sched_tick();
diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
index f8476dfbb60d..1f5fa1cf16dd 100644
--- a/arch/x86/kernel/process_32.c
+++ b/arch/x86/kernel/process_32.c
@@ -166,7 +166,7 @@ void cpu_idle(void)
 
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		while (!need_resched()) {
 			void (*idle)(void);
 
diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c
index e2319f39988b..c0a5c2a687e6 100644
--- a/arch/x86/kernel/process_64.c
+++ b/arch/x86/kernel/process_64.c
@@ -148,7 +148,7 @@ void cpu_idle(void)
 	current_thread_info()->status |= TS_POLLING;
 	/* endless idle loop with no priority at all */
 	while (1) {
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		while (!need_resched()) {
 			void (*idle)(void);
 
diff --git a/include/linux/tick.h b/include/linux/tick.h
index a881c652f7e9..d3c02695dc5d 100644
--- a/include/linux/tick.h
+++ b/include/linux/tick.h
@@ -49,6 +49,7 @@ struct tick_sched {
 	unsigned long			check_clocks;
 	enum tick_nohz_mode		nohz_mode;
 	ktime_t				idle_tick;
+	int				inidle;
 	int				tick_stopped;
 	unsigned long			idle_jiffies;
 	unsigned long			idle_calls;
@@ -105,14 +106,14 @@ static inline int tick_check_oneshot_change(int allow_nohz) { return 0; }
 #endif /* !CONFIG_GENERIC_CLOCKEVENTS */
 
 # ifdef CONFIG_NO_HZ
-extern void tick_nohz_stop_sched_tick(void);
+extern void tick_nohz_stop_sched_tick(int inidle);
 extern void tick_nohz_restart_sched_tick(void);
 extern void tick_nohz_update_jiffies(void);
 extern ktime_t tick_nohz_get_sleep_length(void);
 extern void tick_nohz_stop_idle(int cpu);
 extern u64 get_cpu_idle_time_us(int cpu, u64 *last_update_time);
 # else
-static inline void tick_nohz_stop_sched_tick(void) { }
+static inline void tick_nohz_stop_sched_tick(int inidle) { }
 static inline void tick_nohz_restart_sched_tick(void) { }
 static inline void tick_nohz_update_jiffies(void) { }
 static inline ktime_t tick_nohz_get_sleep_length(void)
diff --git a/kernel/softirq.c b/kernel/softirq.c
index 36e061740047..05f248039d77 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -312,7 +312,7 @@ void irq_exit(void)
 #ifdef CONFIG_NO_HZ
 	/* Make sure that timer wheel updates are propagated */
 	if (!in_interrupt() && idle_cpu(smp_processor_id()) && !need_resched())
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(0);
 	rcu_irq_exit();
 #endif
 	preempt_enable_no_resched();
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 86baa4f0dfe4..ee962d11107b 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -195,7 +195,7 @@ u64 get_cpu_idle_time_us(int cpu, u64 *last_update_time)
  * Called either from the idle loop or from irq_exit() when an idle period was
  * just interrupted by an interrupt which did not cause a reschedule.
  */
-void tick_nohz_stop_sched_tick(void)
+void tick_nohz_stop_sched_tick(int inidle)
 {
 	unsigned long seq, last_jiffies, next_jiffies, delta_jiffies, flags;
 	struct tick_sched *ts;
@@ -224,6 +224,11 @@ void tick_nohz_stop_sched_tick(void)
 	if (unlikely(ts->nohz_mode == NOHZ_MODE_INACTIVE))
 		goto end;
 
+	if (!inidle && !ts->inidle)
+		goto end;
+
+	ts->inidle = 1;
+
 	if (need_resched())
 		goto end;
 
@@ -372,11 +377,14 @@ void tick_nohz_restart_sched_tick(void)
 	local_irq_disable();
 	tick_nohz_stop_idle(cpu);
 
-	if (!ts->tick_stopped) {
+	if (!ts->inidle || !ts->tick_stopped) {
+		ts->inidle = 0;
 		local_irq_enable();
 		return;
 	}
 
+	ts->inidle = 0;
+
 	rcu_exit_nohz();
 
 	/* Update jiffies first */

From 8df185a95c9b84fc0c3c02224e64fdc5b83bae34 Mon Sep 17 00:00:00 2001
From: Mike Travis <travis@sgi.com>
Date: Tue, 8 Jul 2008 15:55:48 -0700
Subject: [PATCH 0071/1171] kthread: reduce stack pressure in create_kthread
 and kthreadd

  * Replace:

  	set_cpus_allowed(..., CPU_MASK_ALL)

    with:

  	set_cpus_allowed_ptr(..., CPU_MASK_ALL_PTR)

    to remove excessive stack requirements when NR_CPUS=4096.

Signed-off-by: Mike Travis <travis@sgi.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/kthread.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/kthread.c b/kernel/kthread.c
index ac3fb7326641..6111c27491b1 100644
--- a/kernel/kthread.c
+++ b/kernel/kthread.c
@@ -106,7 +106,7 @@ static void create_kthread(struct kthread_create_info *create)
 		 */
 		sched_setscheduler(create->result, SCHED_NORMAL, &param);
 		set_user_nice(create->result, KTHREAD_NICE_LEVEL);
-		set_cpus_allowed(create->result, CPU_MASK_ALL);
+		set_cpus_allowed_ptr(create->result, CPU_MASK_ALL_PTR);
 	}
 	complete(&create->done);
 }
@@ -233,7 +233,7 @@ int kthreadd(void *unused)
 	set_task_comm(tsk, "kthreadd");
 	ignore_signals(tsk);
 	set_user_nice(tsk, KTHREAD_NICE_LEVEL);
-	set_cpus_allowed(tsk, CPU_MASK_ALL);
+	set_cpus_allowed_ptr(tsk, CPU_MASK_ALL_PTR);
 
 	current->flags |= PF_NOFREEZE | PF_FREEZER_NOSIG;
 

From 1e01cb0c6ff7e9ddb6547551794c6aa82785a7cb Mon Sep 17 00:00:00 2001
From: Steven Rostedt <rostedt@goodmis.org>
Date: Tue, 15 Jul 2008 09:53:37 -0400
Subject: [PATCH 0072/1171] ftrace: only trace preempt off with preempt tracer

When PREEMPT_TRACER and IRQSOFF_TRACER are both configured and irqsoff
tracer is running, the preempt_off sections might also be traced.

Thanks to Andrew Morton for pointing out my mistake of spin_lock disabling
interrupts while he was reviewing ftrace.txt. Seems that my example I used
actually hit this bug.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/trace/trace_irqsoff.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c
index 421d6fe3650e..b1e4a89b08eb 100644
--- a/kernel/trace/trace_irqsoff.c
+++ b/kernel/trace/trace_irqsoff.c
@@ -337,12 +337,14 @@ EXPORT_SYMBOL(trace_hardirqs_off_caller);
 #ifdef CONFIG_PREEMPT_TRACER
 void trace_preempt_on(unsigned long a0, unsigned long a1)
 {
-	stop_critical_timing(a0, a1);
+	if (preempt_trace())
+		stop_critical_timing(a0, a1);
 }
 
 void trace_preempt_off(unsigned long a0, unsigned long a1)
 {
-	start_critical_timing(a0, a1);
+	if (preempt_trace())
+		start_critical_timing(a0, a1);
 }
 #endif /* CONFIG_PREEMPT_TRACER */
 

From 5d481f497559245ecfb1b95cafe39bfbf037fda5 Mon Sep 17 00:00:00 2001
From: Sonic Zhang <sonic.zhang@analog.com>
Date: Sat, 19 Jul 2008 14:51:31 +0800
Subject: [PATCH 0073/1171] Blackfin arch:  change L1 malloc to base on slab
 cache and lists.

Remove the sram piece limitation and improve the performance to
alloc/free sram piece data.

Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/mm/blackfin_sram.c | 405 ++++++++++++++++++-------------
 arch/blackfin/mm/blackfin_sram.h |   4 +-
 arch/blackfin/mm/init.c          |  12 +-
 3 files changed, 253 insertions(+), 168 deletions(-)

diff --git a/arch/blackfin/mm/blackfin_sram.c b/arch/blackfin/mm/blackfin_sram.c
index 8f6fdc245330..b58cf196d7cc 100644
--- a/arch/blackfin/mm/blackfin_sram.c
+++ b/arch/blackfin/mm/blackfin_sram.c
@@ -41,215 +41,276 @@
 #include <asm/blackfin.h>
 #include "blackfin_sram.h"
 
-spinlock_t l1sram_lock, l1_data_sram_lock, l1_inst_sram_lock;
-
-#if CONFIG_L1_MAX_PIECE < 16
-#undef CONFIG_L1_MAX_PIECE
-#define CONFIG_L1_MAX_PIECE        16
-#endif
-
-#if CONFIG_L1_MAX_PIECE > 1024
-#undef CONFIG_L1_MAX_PIECE
-#define CONFIG_L1_MAX_PIECE        1024
-#endif
-
-#define SRAM_SLT_NULL      0
-#define SRAM_SLT_FREE      1
-#define SRAM_SLT_ALLOCATED 2
+static spinlock_t l1sram_lock, l1_data_sram_lock, l1_inst_sram_lock;
 
 /* the data structure for L1 scratchpad and DATA SRAM */
-struct l1_sram_piece {
+struct sram_piece {
 	void *paddr;
 	int size;
-	int flag;
 	pid_t pid;
+	struct sram_piece *next;
 };
 
-static struct l1_sram_piece l1_ssram[CONFIG_L1_MAX_PIECE];
+static struct sram_piece free_l1_ssram_head, used_l1_ssram_head;
 
 #if L1_DATA_A_LENGTH != 0
-static struct l1_sram_piece l1_data_A_sram[CONFIG_L1_MAX_PIECE];
+static struct sram_piece free_l1_data_A_sram_head, used_l1_data_A_sram_head;
 #endif
 
 #if L1_DATA_B_LENGTH != 0
-static struct l1_sram_piece l1_data_B_sram[CONFIG_L1_MAX_PIECE];
+static struct sram_piece free_l1_data_B_sram_head, used_l1_data_B_sram_head;
 #endif
 
 #if L1_CODE_LENGTH != 0
-static struct l1_sram_piece l1_inst_sram[CONFIG_L1_MAX_PIECE];
+static struct sram_piece free_l1_inst_sram_head, used_l1_inst_sram_head;
 #endif
 
-/* L1 Scratchpad SRAM initialization function */
-void __init l1sram_init(void)
-{
-	printk(KERN_INFO "Blackfin Scratchpad data SRAM: %d KB\n",
-	       L1_SCRATCH_LENGTH >> 10);
+static struct kmem_cache *sram_piece_cache;
 
-	memset(&l1_ssram, 0x00, sizeof(l1_ssram));
-	l1_ssram[0].paddr = (void *)L1_SCRATCH_START;
-	l1_ssram[0].size = L1_SCRATCH_LENGTH;
-	l1_ssram[0].flag = SRAM_SLT_FREE;
+/* L1 Scratchpad SRAM initialization function */
+static void __init l1sram_init(void)
+{
+	free_l1_ssram_head.next =
+		kmem_cache_alloc(sram_piece_cache, GFP_KERNEL);
+	if (!free_l1_ssram_head.next) {
+		printk(KERN_INFO"Fail to initialize Scratchpad data SRAM.\n");
+		return;
+	}
+
+	free_l1_ssram_head.next->paddr = (void *)L1_SCRATCH_START;
+	free_l1_ssram_head.next->size = L1_SCRATCH_LENGTH;
+	free_l1_ssram_head.next->pid = 0;
+	free_l1_ssram_head.next->next = NULL;
+
+	used_l1_ssram_head.next = NULL;
 
 	/* mutex initialize */
 	spin_lock_init(&l1sram_lock);
+
+	printk(KERN_INFO "Blackfin Scratchpad data SRAM: %d KB\n",
+	       L1_SCRATCH_LENGTH >> 10);
 }
 
-void __init l1_data_sram_init(void)
+static void __init l1_data_sram_init(void)
 {
 #if L1_DATA_A_LENGTH != 0
-	memset(&l1_data_A_sram, 0x00, sizeof(l1_data_A_sram));
-	l1_data_A_sram[0].paddr = (void *)L1_DATA_A_START +
-					(_ebss_l1 - _sdata_l1);
-	l1_data_A_sram[0].size = L1_DATA_A_LENGTH - (_ebss_l1 - _sdata_l1);
-	l1_data_A_sram[0].flag = SRAM_SLT_FREE;
+	free_l1_data_A_sram_head.next =
+		kmem_cache_alloc(sram_piece_cache, GFP_KERNEL);
+	if (!free_l1_data_A_sram_head.next) {
+		printk(KERN_INFO"Fail to initialize Data A SRAM.\n");
+		return;
+	}
+
+	free_l1_data_A_sram_head.next->paddr =
+		(void *)L1_DATA_A_START + (_ebss_l1 - _sdata_l1);
+	free_l1_data_A_sram_head.next->size =
+		L1_DATA_A_LENGTH - (_ebss_l1 - _sdata_l1);
+	free_l1_data_A_sram_head.next->pid = 0;
+	free_l1_data_A_sram_head.next->next = NULL;
+
+	used_l1_data_A_sram_head.next = NULL;
 
 	printk(KERN_INFO "Blackfin Data A SRAM: %d KB (%d KB free)\n",
-	       L1_DATA_A_LENGTH >> 10, l1_data_A_sram[0].size >> 10);
+		L1_DATA_A_LENGTH >> 10,
+		free_l1_data_A_sram_head.next->size >> 10);
 #endif
 #if L1_DATA_B_LENGTH != 0
-	memset(&l1_data_B_sram, 0x00, sizeof(l1_data_B_sram));
-	l1_data_B_sram[0].paddr = (void *)L1_DATA_B_START +
-				(_ebss_b_l1 - _sdata_b_l1);
-	l1_data_B_sram[0].size = L1_DATA_B_LENGTH - (_ebss_b_l1 - _sdata_b_l1);
-	l1_data_B_sram[0].flag = SRAM_SLT_FREE;
+	free_l1_data_B_sram_head.next =
+		kmem_cache_alloc(sram_piece_cache, GFP_KERNEL);
+	if (!free_l1_data_B_sram_head.next) {
+		printk(KERN_INFO"Fail to initialize Data B SRAM.\n");
+		return;
+	}
+
+	free_l1_data_B_sram_head.next->paddr =
+		(void *)L1_DATA_B_START + (_ebss_b_l1 - _sdata_b_l1);
+	free_l1_data_B_sram_head.next->size =
+		L1_DATA_B_LENGTH - (_ebss_b_l1 - _sdata_b_l1);
+	free_l1_data_B_sram_head.next->pid = 0;
+	free_l1_data_B_sram_head.next->next = NULL;
+
+	used_l1_data_B_sram_head.next = NULL;
 
 	printk(KERN_INFO "Blackfin Data B SRAM: %d KB (%d KB free)\n",
-	       L1_DATA_B_LENGTH >> 10, l1_data_B_sram[0].size >> 10);
+		L1_DATA_B_LENGTH >> 10,
+		free_l1_data_B_sram_head.next->size >> 10);
 #endif
 
 	/* mutex initialize */
 	spin_lock_init(&l1_data_sram_lock);
 }
 
-void __init l1_inst_sram_init(void)
+static void __init l1_inst_sram_init(void)
 {
 #if L1_CODE_LENGTH != 0
-	memset(&l1_inst_sram, 0x00, sizeof(l1_inst_sram));
-	l1_inst_sram[0].paddr = (void *)L1_CODE_START + (_etext_l1 - _stext_l1);
-	l1_inst_sram[0].size = L1_CODE_LENGTH - (_etext_l1 - _stext_l1);
-	l1_inst_sram[0].flag = SRAM_SLT_FREE;
+	free_l1_inst_sram_head.next =
+		kmem_cache_alloc(sram_piece_cache, GFP_KERNEL);
+	if (!free_l1_inst_sram_head.next) {
+		printk(KERN_INFO"Fail to initialize Instruction SRAM.\n");
+		return;
+	}
+
+	free_l1_inst_sram_head.next->paddr =
+		(void *)L1_CODE_START + (_etext_l1 - _stext_l1);
+	free_l1_inst_sram_head.next->size =
+		L1_CODE_LENGTH - (_etext_l1 - _stext_l1);
+	free_l1_inst_sram_head.next->pid = 0;
+	free_l1_inst_sram_head.next->next = NULL;
+
+	used_l1_inst_sram_head.next = NULL;
 
 	printk(KERN_INFO "Blackfin Instruction SRAM: %d KB (%d KB free)\n",
-	       L1_CODE_LENGTH >> 10, l1_inst_sram[0].size >> 10);
+		L1_CODE_LENGTH >> 10,
+		free_l1_inst_sram_head.next->size >> 10);
 #endif
 
 	/* mutex initialize */
 	spin_lock_init(&l1_inst_sram_lock);
 }
 
-/* L1 memory allocate function */
-static void *_l1_sram_alloc(size_t size, struct l1_sram_piece *pfree, int count)
+void __init bfin_sram_init(void)
 {
-	int i, index = 0;
-	void *addr = NULL;
+	sram_piece_cache = kmem_cache_create("sram_piece_cache",
+				sizeof(struct sram_piece),
+				0, SLAB_PANIC, NULL);
 
-	if (size <= 0)
+	l1sram_init();
+	l1_data_sram_init();
+	l1_inst_sram_init();
+}
+
+/* L1 memory allocate function */
+static void *_l1_sram_alloc(size_t size, struct sram_piece *pfree_head,
+		struct sram_piece *pused_head)
+{
+	struct sram_piece *pslot, *plast, *pavail;
+
+	if (size <= 0 || !pfree_head || !pused_head)
 		return NULL;
 
 	/* Align the size */
 	size = (size + 3) & ~3;
 
-	/* not use the good method to match the best slot !!! */
-	/* search an available memory slot */
-	for (i = 0; i < count; i++) {
-		if ((pfree[i].flag == SRAM_SLT_FREE)
-		    && (pfree[i].size >= size)) {
-			addr = pfree[i].paddr;
-			pfree[i].flag = SRAM_SLT_ALLOCATED;
-			pfree[i].pid = current->pid;
-			index = i;
-			break;
-		}
+	pslot = pfree_head->next;
+	plast = pfree_head;
+
+	/* search an available piece slot */
+	while (pslot != NULL && size > pslot->size) {
+		plast = pslot;
+		pslot = pslot->next;
 	}
-	if (i >= count)
+
+	if (!pslot)
 		return NULL;
 
-	/* updated the NULL memory slot !!! */
-	if (pfree[i].size > size) {
-		for (i = 0; i < count; i++) {
-			if (pfree[i].flag == SRAM_SLT_NULL) {
-				pfree[i].pid = 0;
-				pfree[i].flag = SRAM_SLT_FREE;
-				pfree[i].paddr = addr + size;
-				pfree[i].size = pfree[index].size - size;
-				pfree[index].size = size;
-				break;
-			}
-		}
+	if (pslot->size == size) {
+		plast->next = pslot->next;
+		pavail = pslot;
+	} else {
+		pavail = kmem_cache_alloc(sram_piece_cache, GFP_KERNEL);
+
+		if (!pavail)
+			return NULL;
+
+		pavail->paddr = pslot->paddr;
+		pavail->size = size;
+		pslot->paddr += size;
+		pslot->size -= size;
 	}
 
-	return addr;
+	pavail->pid = current->pid;
+
+	pslot = pused_head->next;
+	plast = pused_head;
+
+	/* insert new piece into used piece list !!! */
+	while (pslot != NULL && pavail->paddr < pslot->paddr) {
+		plast = pslot;
+		pslot = pslot->next;
+	}
+
+	pavail->next = pslot;
+	plast->next = pavail;
+
+	return pavail->paddr;
 }
 
 /* Allocate the largest available block.  */
-static void *_l1_sram_alloc_max(struct l1_sram_piece *pfree, int count,
+static void *_l1_sram_alloc_max(struct sram_piece *pfree_head,
+				struct sram_piece *pused_head,
 				unsigned long *psize)
 {
-	unsigned long best = 0;
-	int i, index = -1;
-	void *addr = NULL;
+	struct sram_piece *pslot, *pmax;
 
-	/* search an available memory slot */
-	for (i = 0; i < count; i++) {
-		if (pfree[i].flag == SRAM_SLT_FREE && pfree[i].size > best) {
-			addr = pfree[i].paddr;
-			index = i;
-			best = pfree[i].size;
-		}
-	}
-	if (index < 0)
+	if (!pfree_head || !pused_head)
 		return NULL;
-	*psize = best;
 
-	pfree[index].pid = current->pid;
-	pfree[index].flag = SRAM_SLT_ALLOCATED;
-	return addr;
+	pmax = pslot = pfree_head->next;
+
+	/* search an available piece slot */
+	while (pslot != NULL) {
+		if (pslot->size > pmax->size)
+			pmax = pslot;
+		pslot = pslot->next;
+	}
+
+	if (!pmax)
+		return NULL;
+
+	*psize = pmax->size;
+
+	return _l1_sram_alloc(*psize, pfree_head, pused_head);
 }
 
 /* L1 memory free function */
 static int _l1_sram_free(const void *addr,
-			struct l1_sram_piece *pfree,
-			int count)
+			struct sram_piece *pfree_head,
+			struct sram_piece *pused_head)
 {
-	int i, index = 0;
+	struct sram_piece *pslot, *plast, *pavail;
 
-	/* search the relevant memory slot */
-	for (i = 0; i < count; i++) {
-		if (pfree[i].paddr == addr) {
-			if (pfree[i].flag != SRAM_SLT_ALLOCATED) {
-				/* error log */
-				return -1;
-			}
-			index = i;
-			break;
-		}
-	}
-	if (i >= count)
+	if (!pfree_head || !pused_head)
 		return -1;
 
-	pfree[index].pid = 0;
-	pfree[index].flag = SRAM_SLT_FREE;
+	/* search the relevant memory slot */
+	pslot = pused_head->next;
+	plast = pused_head;
 
-	/* link the next address slot */
-	for (i = 0; i < count; i++) {
-		if (((pfree[index].paddr + pfree[index].size) == pfree[i].paddr)
-		    && (pfree[i].flag == SRAM_SLT_FREE)) {
-			pfree[i].pid = 0;
-			pfree[i].flag = SRAM_SLT_NULL;
-			pfree[index].size += pfree[i].size;
-			pfree[index].flag = SRAM_SLT_FREE;
-			break;
-		}
+	/* search an available piece slot */
+	while (pslot != NULL && pslot->paddr != addr) {
+		plast = pslot;
+		pslot = pslot->next;
 	}
 
-	/* link the last address slot */
-	for (i = 0; i < count; i++) {
-		if (((pfree[i].paddr + pfree[i].size) == pfree[index].paddr) &&
-		    (pfree[i].flag == SRAM_SLT_FREE)) {
-			pfree[index].flag = SRAM_SLT_NULL;
-			pfree[i].size += pfree[index].size;
-			break;
-		}
+	if (!pslot)
+		return -1;
+
+	plast->next = pslot->next;
+	pavail = pslot;
+	pavail->pid = 0;
+
+	/* insert free pieces back to the free list */
+	pslot = pfree_head->next;
+	plast = pfree_head;
+
+	while (pslot != NULL && addr > pslot->paddr) {
+		plast = pslot;
+		pslot = pslot->next;
+	}
+
+	if (plast != pfree_head && plast->paddr + plast->size == pavail->paddr) {
+		plast->size += pavail->size;
+		kmem_cache_free(sram_piece_cache, pavail);
+	} else {
+		pavail->next = plast;
+		plast->next = pavail;
+		plast = pavail;
+	}
+
+	if (pslot && plast->paddr + plast->size == pslot->paddr) {
+		plast->size += pslot->size;
+		plast->next = pslot->next;
+		kmem_cache_free(sram_piece_cache, pslot);
 	}
 
 	return 0;
@@ -287,7 +348,8 @@ void *l1_data_A_sram_alloc(size_t size)
 	spin_lock_irqsave(&l1_data_sram_lock, flags);
 
 #if L1_DATA_A_LENGTH != 0
-	addr = _l1_sram_alloc(size, l1_data_A_sram, ARRAY_SIZE(l1_data_A_sram));
+	addr = _l1_sram_alloc(size, &free_l1_data_A_sram_head,
+			&used_l1_data_A_sram_head);
 #endif
 
 	/* add mutex operation */
@@ -309,8 +371,8 @@ int l1_data_A_sram_free(const void *addr)
 	spin_lock_irqsave(&l1_data_sram_lock, flags);
 
 #if L1_DATA_A_LENGTH != 0
-	ret = _l1_sram_free(addr,
-			   l1_data_A_sram, ARRAY_SIZE(l1_data_A_sram));
+	ret = _l1_sram_free(addr, &free_l1_data_A_sram_head,
+			&used_l1_data_A_sram_head);
 #else
 	ret = -1;
 #endif
@@ -331,7 +393,8 @@ void *l1_data_B_sram_alloc(size_t size)
 	/* add mutex operation */
 	spin_lock_irqsave(&l1_data_sram_lock, flags);
 
-	addr = _l1_sram_alloc(size, l1_data_B_sram, ARRAY_SIZE(l1_data_B_sram));
+	addr = _l1_sram_alloc(size, &free_l1_data_B_sram_head,
+			&used_l1_data_B_sram_head);
 
 	/* add mutex operation */
 	spin_unlock_irqrestore(&l1_data_sram_lock, flags);
@@ -355,7 +418,8 @@ int l1_data_B_sram_free(const void *addr)
 	/* add mutex operation */
 	spin_lock_irqsave(&l1_data_sram_lock, flags);
 
-	ret = _l1_sram_free(addr, l1_data_B_sram, ARRAY_SIZE(l1_data_B_sram));
+	ret = _l1_sram_free(addr, &free_l1_data_B_sram_head,
+			&used_l1_data_B_sram_head);
 
 	/* add mutex operation */
 	spin_unlock_irqrestore(&l1_data_sram_lock, flags);
@@ -408,7 +472,8 @@ void *l1_inst_sram_alloc(size_t size)
 	/* add mutex operation */
 	spin_lock_irqsave(&l1_inst_sram_lock, flags);
 
-	addr = _l1_sram_alloc(size, l1_inst_sram, ARRAY_SIZE(l1_inst_sram));
+	addr = _l1_sram_alloc(size, &free_l1_inst_sram_head,
+			&used_l1_inst_sram_head);
 
 	/* add mutex operation */
 	spin_unlock_irqrestore(&l1_inst_sram_lock, flags);
@@ -432,7 +497,8 @@ int l1_inst_sram_free(const void *addr)
 	/* add mutex operation */
 	spin_lock_irqsave(&l1_inst_sram_lock, flags);
 
-	ret = _l1_sram_free(addr, l1_inst_sram, ARRAY_SIZE(l1_inst_sram));
+	ret = _l1_sram_free(addr, &free_l1_inst_sram_head,
+			&used_l1_inst_sram_head);
 
 	/* add mutex operation */
 	spin_unlock_irqrestore(&l1_inst_sram_lock, flags);
@@ -453,7 +519,8 @@ void *l1sram_alloc(size_t size)
 	/* add mutex operation */
 	spin_lock_irqsave(&l1sram_lock, flags);
 
-	addr = _l1_sram_alloc(size, l1_ssram, ARRAY_SIZE(l1_ssram));
+	addr = _l1_sram_alloc(size, &free_l1_ssram_head,
+			&used_l1_ssram_head);
 
 	/* add mutex operation */
 	spin_unlock_irqrestore(&l1sram_lock, flags);
@@ -470,7 +537,8 @@ void *l1sram_alloc_max(size_t *psize)
 	/* add mutex operation */
 	spin_lock_irqsave(&l1sram_lock, flags);
 
-	addr = _l1_sram_alloc_max(l1_ssram, ARRAY_SIZE(l1_ssram), psize);
+	addr = _l1_sram_alloc_max(&free_l1_ssram_head,
+			&used_l1_ssram_head, psize);
 
 	/* add mutex operation */
 	spin_unlock_irqrestore(&l1sram_lock, flags);
@@ -487,7 +555,8 @@ int l1sram_free(const void *addr)
 	/* add mutex operation */
 	spin_lock_irqsave(&l1sram_lock, flags);
 
-	ret = _l1_sram_free(addr, l1_ssram, ARRAY_SIZE(l1_ssram));
+	ret = _l1_sram_free(addr, &free_l1_ssram_head,
+			&used_l1_ssram_head);
 
 	/* add mutex operation */
 	spin_unlock_irqrestore(&l1sram_lock, flags);
@@ -553,28 +622,38 @@ EXPORT_SYMBOL(sram_alloc_with_lsl);
  * (including newline).
  */
 static int _l1sram_proc_read(char *buf, int *len, int count, const char *desc,
-		struct l1_sram_piece *pfree, const int array_size)
+		struct sram_piece *pfree_head,
+		struct sram_piece *pused_head)
 {
-	int i;
+	struct sram_piece *pslot;
+
+	if (!pfree_head || !pused_head)
+		return -1;
 
 	*len += sprintf(&buf[*len], "--- L1 %-14s Size   PID State     \n", desc);
-	for (i = 0; i < array_size && *len < count; ++i) {
-		const char *alloc_type;
-		switch (pfree[i].flag) {
-		case SRAM_SLT_NULL:      alloc_type = "NULL"; break;
-		case SRAM_SLT_FREE:      alloc_type = "FREE"; break;
-		case SRAM_SLT_ALLOCATED: alloc_type = "ALLOCATED"; break;
-		default:                 alloc_type = "????"; break;
-		}
-		/* if we've got a lot of space to cover, omit things */
-		if ((PAGE_SIZE - 1024) < (CONFIG_L1_MAX_PIECE + 1) * 4 * 44 &&
-		    pfree[i].size == 0)
-			continue;
+
+	/* search the relevant memory slot */
+	pslot = pused_head->next;
+
+	while (pslot != NULL) {
 		*len += sprintf(&buf[*len], "%p-%p %8i %5i %-10s\n",
-			pfree[i].paddr, pfree[i].paddr + pfree[i].size,
-			pfree[i].size, pfree[i].pid, alloc_type);
+			pslot->paddr, pslot->paddr + pslot->size,
+			pslot->size, pslot->pid, "ALLOCATED");
+
+		pslot = pslot->next;
 	}
-	return (i != array_size);
+
+	pslot = pfree_head->next;
+
+	while (pslot != NULL) {
+		*len += sprintf(&buf[*len], "%p-%p %8i %5i %-10s\n",
+			pslot->paddr, pslot->paddr + pslot->size,
+			pslot->size, pslot->pid, "FREE");
+
+		pslot = pslot->next;
+	}
+
+	return 0;
 }
 static int l1sram_proc_read(char *buf, char **start, off_t offset, int count,
 		int *eof, void *data)
@@ -582,21 +661,23 @@ static int l1sram_proc_read(char *buf, char **start, off_t offset, int count,
 	int len = 0;
 
 	if (_l1sram_proc_read(buf, &len, count, "Scratchpad",
-			l1_ssram, ARRAY_SIZE(l1_ssram)))
+			&free_l1_ssram_head, &used_l1_ssram_head))
 		goto not_done;
 #if L1_DATA_A_LENGTH != 0
 	if (_l1sram_proc_read(buf, &len, count, "Data A",
-			l1_data_A_sram, ARRAY_SIZE(l1_data_A_sram)))
+			&free_l1_data_A_sram_head,
+			&used_l1_data_A_sram_head))
 		goto not_done;
 #endif
 #if L1_DATA_B_LENGTH != 0
 	if (_l1sram_proc_read(buf, &len, count, "Data B",
-			l1_data_B_sram, ARRAY_SIZE(l1_data_B_sram)))
+			&free_l1_data_B_sram_head,
+			&used_l1_data_B_sram_head))
 		goto not_done;
 #endif
 #if L1_CODE_LENGTH != 0
 	if (_l1sram_proc_read(buf, &len, count, "Instruction",
-			l1_inst_sram, ARRAY_SIZE(l1_inst_sram)))
+			&free_l1_inst_sram_head, &used_l1_inst_sram_head))
 		goto not_done;
 #endif
 
diff --git a/arch/blackfin/mm/blackfin_sram.h b/arch/blackfin/mm/blackfin_sram.h
index 0fb73b78dd60..8cb0945563f9 100644
--- a/arch/blackfin/mm/blackfin_sram.h
+++ b/arch/blackfin/mm/blackfin_sram.h
@@ -30,9 +30,7 @@
 #ifndef __BLACKFIN_SRAM_H__
 #define __BLACKFIN_SRAM_H__
 
-extern void l1sram_init(void);
-extern void l1_inst_sram_init(void);
-extern void l1_data_sram_init(void);
+extern void bfin_sram_init(void);
 extern void *l1sram_alloc(size_t);
 
 #endif
diff --git a/arch/blackfin/mm/init.c b/arch/blackfin/mm/init.c
index ec3141fefd20..4aab21f44096 100644
--- a/arch/blackfin/mm/init.c
+++ b/arch/blackfin/mm/init.c
@@ -164,11 +164,14 @@ void __init mem_init(void)
 		"(%uk init code, %uk kernel code, %uk data, %uk dma, %uk reserved)\n",
 		(unsigned long) freepages << (PAGE_SHIFT-10), _ramend >> 10,
 		initk, codek, datak, DMA_UNCACHED_REGION >> 10, (reservedpages << (PAGE_SHIFT-10)));
+}
+
+static int __init sram_init(void)
+{
+	unsigned long tmp;
 
 	/* Initialize the blackfin L1 Memory. */
-	l1sram_init();
-	l1_data_sram_init();
-	l1_inst_sram_init();
+	bfin_sram_init();
 
 	/* Allocate this once; never free it.  We assume this gives us a
 	   pointer to the start of L1 scratchpad memory; panic if it
@@ -179,7 +182,10 @@ void __init mem_init(void)
 			tmp, (unsigned long)L1_SCRATCH_TASK_INFO);
 		panic("No L1, time to give up\n");
 	}
+
+	return 0;
 }
+pure_initcall(sram_init);
 
 static void __init free_init_pages(const char *what, unsigned long begin, unsigned long end)
 {

From bafcc1b97323261a42d47960db99947bcc1be133 Mon Sep 17 00:00:00 2001
From: Robin Getz <rgetz@blackfin.uclinux.org>
Date: Sat, 19 Jul 2008 15:11:15 +0800
Subject: [PATCH 0074/1171] Blackfin arch: Remove useless config option.

Signed-off-by: Robin Getz <rgetz@blackfin.uclinux.org>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/Kconfig | 8 --------
 1 file changed, 8 deletions(-)

diff --git a/arch/blackfin/Kconfig b/arch/blackfin/Kconfig
index 8f21b78b2e6f..5a08194e4118 100644
--- a/arch/blackfin/Kconfig
+++ b/arch/blackfin/Kconfig
@@ -729,14 +729,6 @@ config BFIN_WT
 
 endchoice
 
-config L1_MAX_PIECE
-	int "Set the max L1 SRAM pieces"
-	default 16
-	help
-	  Set the max memory pieces for the L1 SRAM allocation algorithm.
-	  Min value is 16. Max value is 1024.
-
-
 config MPU
 	bool "Enable the memory protection unit (EXPERIMENTAL)"
 	default n

From 262c3825a9f3eb0f4f30ebb4b1ee57397bcb3ffc Mon Sep 17 00:00:00 2001
From: Sonic Zhang <sonic.zhang@analog.com>
Date: Sat, 19 Jul 2008 15:42:41 +0800
Subject: [PATCH 0075/1171] Blackfin arch:  Extend sram malloc to handle L2
 SRAM.

Extend system call to alloc L2 SRAM in application.
Automatically move following sections to L2 SRAM:
1. kernel built-in l2 attribute section
2. kernel module l2 attribute section
3. elf-fdpic application l2 attribute section

Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/kernel/module.c      |  74 ++++++++++---
 arch/blackfin/kernel/setup.c       |  10 ++
 arch/blackfin/kernel/vmlinux.lds.S |  40 ++++++-
 arch/blackfin/mm/blackfin_sram.c   | 170 +++++++++++++++++++++++------
 include/asm-blackfin/bfin-global.h |   8 +-
 include/asm-blackfin/elf.h         |   2 +
 include/asm-blackfin/module.h      |   5 +-
 7 files changed, 253 insertions(+), 56 deletions(-)

diff --git a/arch/blackfin/kernel/module.c b/arch/blackfin/kernel/module.c
index 14a42848f37f..e1bebc80a5bf 100644
--- a/arch/blackfin/kernel/module.c
+++ b/arch/blackfin/kernel/module.c
@@ -173,7 +173,7 @@ module_frob_arch_sections(Elf_Ehdr * hdr, Elf_Shdr * sechdrs,
 	for (s = sechdrs; s < sechdrs_end; ++s) {
 		if ((strcmp(".l1.text", secstrings + s->sh_name) == 0) ||
 		    ((strcmp(".text", secstrings + s->sh_name) == 0) &&
-		     (hdr->e_flags & FLG_CODE_IN_L1) && (s->sh_size > 0))) {
+		     (hdr->e_flags & EF_BFIN_CODE_IN_L1) && (s->sh_size > 0))) {
 			dest = l1_inst_sram_alloc(s->sh_size);
 			mod->arch.text_l1 = dest;
 			if (dest == NULL) {
@@ -188,7 +188,7 @@ module_frob_arch_sections(Elf_Ehdr * hdr, Elf_Shdr * sechdrs,
 		}
 		if ((strcmp(".l1.data", secstrings + s->sh_name) == 0) ||
 		    ((strcmp(".data", secstrings + s->sh_name) == 0) &&
-		     (hdr->e_flags & FLG_DATA_IN_L1) && (s->sh_size > 0))) {
+		     (hdr->e_flags & EF_BFIN_DATA_IN_L1) && (s->sh_size > 0))) {
 			dest = l1_data_sram_alloc(s->sh_size);
 			mod->arch.data_a_l1 = dest;
 			if (dest == NULL) {
@@ -203,7 +203,7 @@ module_frob_arch_sections(Elf_Ehdr * hdr, Elf_Shdr * sechdrs,
 		}
 		if (strcmp(".l1.bss", secstrings + s->sh_name) == 0 ||
 		    ((strcmp(".bss", secstrings + s->sh_name) == 0) &&
-		     (hdr->e_flags & FLG_DATA_IN_L1) && (s->sh_size > 0))) {
+		     (hdr->e_flags & EF_BFIN_DATA_IN_L1) && (s->sh_size > 0))) {
 			dest = l1_data_sram_alloc(s->sh_size);
 			mod->arch.bss_a_l1 = dest;
 			if (dest == NULL) {
@@ -242,6 +242,51 @@ module_frob_arch_sections(Elf_Ehdr * hdr, Elf_Shdr * sechdrs,
 			s->sh_flags &= ~SHF_ALLOC;
 			s->sh_addr = (unsigned long)dest;
 		}
+		if ((strcmp(".l2.text", secstrings + s->sh_name) == 0) ||
+		    ((strcmp(".text", secstrings + s->sh_name) == 0) &&
+		     (hdr->e_flags & EF_BFIN_CODE_IN_L2) && (s->sh_size > 0))) {
+			dest = l2_sram_alloc(s->sh_size);
+			mod->arch.text_l2 = dest;
+			if (dest == NULL) {
+				printk(KERN_ERR
+				       "module %s: L2 SRAM allocation failed\n",
+				       mod->name);
+				return -1;
+			}
+			memcpy(dest, (void *)s->sh_addr, s->sh_size);
+			s->sh_flags &= ~SHF_ALLOC;
+			s->sh_addr = (unsigned long)dest;
+		}
+		if ((strcmp(".l2.data", secstrings + s->sh_name) == 0) ||
+		    ((strcmp(".data", secstrings + s->sh_name) == 0) &&
+		     (hdr->e_flags & EF_BFIN_DATA_IN_L2) && (s->sh_size > 0))) {
+			dest = l2_sram_alloc(s->sh_size);
+			mod->arch.data_l2 = dest;
+			if (dest == NULL) {
+				printk(KERN_ERR
+					"module %s: L2 SRAM allocation failed\n",
+					mod->name);
+				return -1;
+			}
+			memcpy(dest, (void *)s->sh_addr, s->sh_size);
+			s->sh_flags &= ~SHF_ALLOC;
+			s->sh_addr = (unsigned long)dest;
+		}
+		if (strcmp(".l2.bss", secstrings + s->sh_name) == 0 ||
+		    ((strcmp(".bss", secstrings + s->sh_name) == 0) &&
+		     (hdr->e_flags & EF_BFIN_DATA_IN_L2) && (s->sh_size > 0))) {
+			dest = l2_sram_alloc(s->sh_size);
+			mod->arch.bss_l2 = dest;
+			if (dest == NULL) {
+				printk(KERN_ERR
+					"module %s: L2 SRAM allocation failed\n",
+					mod->name);
+				return -1;
+			}
+			memset(dest, 0, s->sh_size);
+			s->sh_flags &= ~SHF_ALLOC;
+			s->sh_addr = (unsigned long)dest;
+		}
 	}
 	return 0;
 }
@@ -411,9 +456,10 @@ module_finalize(const Elf_Ehdr * hdr,
 			continue;
 
 		if ((sechdrs[i].sh_type == SHT_RELA) &&
-		    ((strcmp(".rela.l1.text", secstrings + sechdrs[i].sh_name) == 0) ||
+		    ((strcmp(".rela.l2.text", secstrings + sechdrs[i].sh_name) == 0) ||
+		    (strcmp(".rela.l1.text", secstrings + sechdrs[i].sh_name) == 0) ||
 		    ((strcmp(".rela.text", secstrings + sechdrs[i].sh_name) == 0) &&
-			 (hdr->e_flags & FLG_CODE_IN_L1)))) {
+			(hdr->e_flags & (EF_BFIN_CODE_IN_L1|EF_BFIN_CODE_IN_L2))))) {
 			apply_relocate_add((Elf_Shdr *) sechdrs, strtab,
 					   symindex, i, mod);
 		}
@@ -423,14 +469,12 @@ module_finalize(const Elf_Ehdr * hdr,
 
 void module_arch_cleanup(struct module *mod)
 {
-	if (mod->arch.text_l1)
-		l1_inst_sram_free((void *)mod->arch.text_l1);
-	if (mod->arch.data_a_l1)
-		l1_data_sram_free((void *)mod->arch.data_a_l1);
-	if (mod->arch.bss_a_l1)
-		l1_data_sram_free((void *)mod->arch.bss_a_l1);
-	if (mod->arch.data_b_l1)
-		l1_data_B_sram_free((void *)mod->arch.data_b_l1);
-	if (mod->arch.bss_b_l1)
-		l1_data_B_sram_free((void *)mod->arch.bss_b_l1);
+	l1_inst_sram_free(mod->arch.text_l1);
+	l1_data_A_sram_free(mod->arch.data_a_l1);
+	l1_data_A_sram_free(mod->arch.bss_a_l1);
+	l1_data_B_sram_free(mod->arch.data_b_l1);
+	l1_data_B_sram_free(mod->arch.bss_b_l1);
+	l2_sram_free(mod->arch.text_l2);
+	l2_sram_free(mod->arch.data_l2);
+	l2_sram_free(mod->arch.bss_l2);
 }
diff --git a/arch/blackfin/kernel/setup.c b/arch/blackfin/kernel/setup.c
index 861a1db74df8..8671d1db1f99 100644
--- a/arch/blackfin/kernel/setup.c
+++ b/arch/blackfin/kernel/setup.c
@@ -104,6 +104,7 @@ void __init bf53x_relocate_l1_mem(void)
 	unsigned long l1_code_length;
 	unsigned long l1_data_a_length;
 	unsigned long l1_data_b_length;
+	unsigned long l2_length;
 
 	l1_code_length = _etext_l1 - _stext_l1;
 	if (l1_code_length > L1_CODE_LENGTH)
@@ -129,6 +130,15 @@ void __init bf53x_relocate_l1_mem(void)
 	/* Copy _sdata_b_l1 to _ebss_b_l1 to L1 data bank B SRAM */
 	dma_memcpy(_sdata_b_l1, _l1_lma_start + l1_code_length +
 			l1_data_a_length, l1_data_b_length);
+
+#ifdef L2_LENGTH
+	l2_length = _ebss_l2 - _stext_l2;
+	if (l2_length > L2_LENGTH)
+		panic("L2 SRAM Overflow\n");
+
+	/* Copy _stext_l2 to _edata_l2 to L2 SRAM */
+	dma_memcpy(_stext_l2, _l2_lma_start, l2_length);
+#endif
 }
 
 /* add_memory_region to memmap */
diff --git a/arch/blackfin/kernel/vmlinux.lds.S b/arch/blackfin/kernel/vmlinux.lds.S
index 3ecc64cab3be..0896e38d6108 100644
--- a/arch/blackfin/kernel/vmlinux.lds.S
+++ b/arch/blackfin/kernel/vmlinux.lds.S
@@ -101,6 +101,11 @@ SECTIONS
 #if !L1_DATA_B_LENGTH
 		*(.l1.data.B)
 #endif
+#ifndef L2_LENGTH
+		. = ALIGN(32);
+		*(.data_l2.cacheline_aligned)
+		*(.l2.data)
+#endif
 
 		DATA_DATA
 		*(.data.*)
@@ -182,13 +187,12 @@ SECTIONS
 		*(.l1.data)
 		__edata_l1 = .;
 
-		. = ALIGN(4);
-		__sbss_l1 = .;
-		*(.l1.bss)
-
 		. = ALIGN(32);
 		*(.data_l1.cacheline_aligned)
 
+		. = ALIGN(4);
+		__sbss_l1 = .;
+		*(.l1.bss)
 		. = ALIGN(4);
 		__ebss_l1 = .;
 	}
@@ -203,11 +207,37 @@ SECTIONS
 		. = ALIGN(4);
 		__sbss_b_l1 = .;
 		*(.l1.bss.B)
-
 		. = ALIGN(4);
 		__ebss_b_l1 = .;
 	}
 
+#ifdef L2_LENGTH
+	__l2_lma_start = .;
+
+	.text_data_l2 L2_START : AT(LOADADDR(.data_b_l1) + SIZEOF(.data_b_l1))
+	{
+		. = ALIGN(4);
+		__stext_l2 = .;
+		*(.l1.text)
+		. = ALIGN(4);
+		__etext_l2 = .;
+
+		. = ALIGN(4);
+		__sdata_l2 = .;
+		*(.l1.data)
+		__edata_l2 = .;
+
+		. = ALIGN(32);
+		*(.data_l2.cacheline_aligned)
+
+		. = ALIGN(4);
+		__sbss_l2 = .;
+		*(.l1.bss)
+		. = ALIGN(4);
+		__ebss_l2 = .;
+	}
+#endif
+
 	/* Force trailing alignment of our init section so that when we
 	 * free our init memory, we don't leave behind a partial page.
 	 */
diff --git a/arch/blackfin/mm/blackfin_sram.c b/arch/blackfin/mm/blackfin_sram.c
index b58cf196d7cc..5af3c31c9365 100644
--- a/arch/blackfin/mm/blackfin_sram.c
+++ b/arch/blackfin/mm/blackfin_sram.c
@@ -42,6 +42,7 @@
 #include "blackfin_sram.h"
 
 static spinlock_t l1sram_lock, l1_data_sram_lock, l1_inst_sram_lock;
+static spinlock_t l2_sram_lock;
 
 /* the data structure for L1 scratchpad and DATA SRAM */
 struct sram_piece {
@@ -65,6 +66,10 @@ static struct sram_piece free_l1_data_B_sram_head, used_l1_data_B_sram_head;
 static struct sram_piece free_l1_inst_sram_head, used_l1_inst_sram_head;
 #endif
 
+#ifdef L2_LENGTH
+static struct sram_piece free_l2_sram_head, used_l2_sram_head;
+#endif
+
 static struct kmem_cache *sram_piece_cache;
 
 /* L1 Scratchpad SRAM initialization function */
@@ -97,7 +102,7 @@ static void __init l1_data_sram_init(void)
 	free_l1_data_A_sram_head.next =
 		kmem_cache_alloc(sram_piece_cache, GFP_KERNEL);
 	if (!free_l1_data_A_sram_head.next) {
-		printk(KERN_INFO"Fail to initialize Data A SRAM.\n");
+		printk(KERN_INFO"Fail to initialize L1 Data A SRAM.\n");
 		return;
 	}
 
@@ -110,7 +115,7 @@ static void __init l1_data_sram_init(void)
 
 	used_l1_data_A_sram_head.next = NULL;
 
-	printk(KERN_INFO "Blackfin Data A SRAM: %d KB (%d KB free)\n",
+	printk(KERN_INFO "Blackfin L1 Data A SRAM: %d KB (%d KB free)\n",
 		L1_DATA_A_LENGTH >> 10,
 		free_l1_data_A_sram_head.next->size >> 10);
 #endif
@@ -118,7 +123,7 @@ static void __init l1_data_sram_init(void)
 	free_l1_data_B_sram_head.next =
 		kmem_cache_alloc(sram_piece_cache, GFP_KERNEL);
 	if (!free_l1_data_B_sram_head.next) {
-		printk(KERN_INFO"Fail to initialize Data B SRAM.\n");
+		printk(KERN_INFO"Fail to initialize L1 Data B SRAM.\n");
 		return;
 	}
 
@@ -131,7 +136,7 @@ static void __init l1_data_sram_init(void)
 
 	used_l1_data_B_sram_head.next = NULL;
 
-	printk(KERN_INFO "Blackfin Data B SRAM: %d KB (%d KB free)\n",
+	printk(KERN_INFO "Blackfin L1 Data B SRAM: %d KB (%d KB free)\n",
 		L1_DATA_B_LENGTH >> 10,
 		free_l1_data_B_sram_head.next->size >> 10);
 #endif
@@ -146,7 +151,7 @@ static void __init l1_inst_sram_init(void)
 	free_l1_inst_sram_head.next =
 		kmem_cache_alloc(sram_piece_cache, GFP_KERNEL);
 	if (!free_l1_inst_sram_head.next) {
-		printk(KERN_INFO"Fail to initialize Instruction SRAM.\n");
+		printk(KERN_INFO"Fail to initialize L1 Instruction SRAM.\n");
 		return;
 	}
 
@@ -159,7 +164,7 @@ static void __init l1_inst_sram_init(void)
 
 	used_l1_inst_sram_head.next = NULL;
 
-	printk(KERN_INFO "Blackfin Instruction SRAM: %d KB (%d KB free)\n",
+	printk(KERN_INFO "Blackfin L1 Instruction SRAM: %d KB (%d KB free)\n",
 		L1_CODE_LENGTH >> 10,
 		free_l1_inst_sram_head.next->size >> 10);
 #endif
@@ -168,6 +173,33 @@ static void __init l1_inst_sram_init(void)
 	spin_lock_init(&l1_inst_sram_lock);
 }
 
+static void __init l2_sram_init(void)
+{
+#ifdef L2_LENGTH
+	free_l2_sram_head.next =
+		kmem_cache_alloc(sram_piece_cache, GFP_KERNEL);
+	if (!free_l2_sram_head.next) {
+		printk(KERN_INFO"Fail to initialize L2 SRAM.\n");
+		return;
+	}
+
+	free_l2_sram_head.next->paddr = (void *)L2_START +
+		(_etext_l2 - _stext_l2) + (_edata_l2 - _sdata_l2);
+	free_l2_sram_head.next->size = L2_LENGTH -
+		(_etext_l2 - _stext_l2) + (_edata_l2 - _sdata_l2);
+	free_l2_sram_head.next->pid = 0;
+	free_l2_sram_head.next->next = NULL;
+
+	used_l2_sram_head.next = NULL;
+
+	printk(KERN_INFO "Blackfin L2 SRAM: %d KB (%d KB free)\n",
+		L2_LENGTH >> 10,
+		free_l2_sram_head.next->size >> 10);
+#endif
+
+	/* mutex initialize */
+	spin_lock_init(&l2_sram_lock);
+}
 void __init bfin_sram_init(void)
 {
 	sram_piece_cache = kmem_cache_create("sram_piece_cache",
@@ -177,10 +209,11 @@ void __init bfin_sram_init(void)
 	l1sram_init();
 	l1_data_sram_init();
 	l1_inst_sram_init();
+	l2_sram_init();
 }
 
-/* L1 memory allocate function */
-static void *_l1_sram_alloc(size_t size, struct sram_piece *pfree_head,
+/* SRAM allocate function */
+static void *_sram_alloc(size_t size, struct sram_piece *pfree_head,
 		struct sram_piece *pused_head)
 {
 	struct sram_piece *pslot, *plast, *pavail;
@@ -236,7 +269,7 @@ static void *_l1_sram_alloc(size_t size, struct sram_piece *pfree_head,
 }
 
 /* Allocate the largest available block.  */
-static void *_l1_sram_alloc_max(struct sram_piece *pfree_head,
+static void *_sram_alloc_max(struct sram_piece *pfree_head,
 				struct sram_piece *pused_head,
 				unsigned long *psize)
 {
@@ -259,11 +292,11 @@ static void *_l1_sram_alloc_max(struct sram_piece *pfree_head,
 
 	*psize = pmax->size;
 
-	return _l1_sram_alloc(*psize, pfree_head, pused_head);
+	return _sram_alloc(*psize, pfree_head, pused_head);
 }
 
-/* L1 memory free function */
-static int _l1_sram_free(const void *addr,
+/* SRAM free function */
+static int _sram_free(const void *addr,
 			struct sram_piece *pfree_head,
 			struct sram_piece *pused_head)
 {
@@ -333,6 +366,11 @@ int sram_free(const void *addr)
 	else if (addr >= (void *)L1_DATA_B_START
 		 && addr < (void *)(L1_DATA_B_START + L1_DATA_B_LENGTH))
 		return l1_data_B_sram_free(addr);
+#endif
+#ifdef L2_LENGTH
+	else if (addr >= (void *)L2_START
+		 && addr < (void *)(L2_START + L2_LENGTH))
+		return l2_sram_free(addr);
 #endif
 	else
 		return -1;
@@ -348,7 +386,7 @@ void *l1_data_A_sram_alloc(size_t size)
 	spin_lock_irqsave(&l1_data_sram_lock, flags);
 
 #if L1_DATA_A_LENGTH != 0
-	addr = _l1_sram_alloc(size, &free_l1_data_A_sram_head,
+	addr = _sram_alloc(size, &free_l1_data_A_sram_head,
 			&used_l1_data_A_sram_head);
 #endif
 
@@ -371,7 +409,7 @@ int l1_data_A_sram_free(const void *addr)
 	spin_lock_irqsave(&l1_data_sram_lock, flags);
 
 #if L1_DATA_A_LENGTH != 0
-	ret = _l1_sram_free(addr, &free_l1_data_A_sram_head,
+	ret = _sram_free(addr, &free_l1_data_A_sram_head,
 			&used_l1_data_A_sram_head);
 #else
 	ret = -1;
@@ -393,7 +431,7 @@ void *l1_data_B_sram_alloc(size_t size)
 	/* add mutex operation */
 	spin_lock_irqsave(&l1_data_sram_lock, flags);
 
-	addr = _l1_sram_alloc(size, &free_l1_data_B_sram_head,
+	addr = _sram_alloc(size, &free_l1_data_B_sram_head,
 			&used_l1_data_B_sram_head);
 
 	/* add mutex operation */
@@ -418,7 +456,7 @@ int l1_data_B_sram_free(const void *addr)
 	/* add mutex operation */
 	spin_lock_irqsave(&l1_data_sram_lock, flags);
 
-	ret = _l1_sram_free(addr, &free_l1_data_B_sram_head,
+	ret = _sram_free(addr, &free_l1_data_B_sram_head,
 			&used_l1_data_B_sram_head);
 
 	/* add mutex operation */
@@ -472,7 +510,7 @@ void *l1_inst_sram_alloc(size_t size)
 	/* add mutex operation */
 	spin_lock_irqsave(&l1_inst_sram_lock, flags);
 
-	addr = _l1_sram_alloc(size, &free_l1_inst_sram_head,
+	addr = _sram_alloc(size, &free_l1_inst_sram_head,
 			&used_l1_inst_sram_head);
 
 	/* add mutex operation */
@@ -497,7 +535,7 @@ int l1_inst_sram_free(const void *addr)
 	/* add mutex operation */
 	spin_lock_irqsave(&l1_inst_sram_lock, flags);
 
-	ret = _l1_sram_free(addr, &free_l1_inst_sram_head,
+	ret = _sram_free(addr, &free_l1_inst_sram_head,
 			&used_l1_inst_sram_head);
 
 	/* add mutex operation */
@@ -519,7 +557,7 @@ void *l1sram_alloc(size_t size)
 	/* add mutex operation */
 	spin_lock_irqsave(&l1sram_lock, flags);
 
-	addr = _l1_sram_alloc(size, &free_l1_ssram_head,
+	addr = _sram_alloc(size, &free_l1_ssram_head,
 			&used_l1_ssram_head);
 
 	/* add mutex operation */
@@ -537,7 +575,7 @@ void *l1sram_alloc_max(size_t *psize)
 	/* add mutex operation */
 	spin_lock_irqsave(&l1sram_lock, flags);
 
-	addr = _l1_sram_alloc_max(&free_l1_ssram_head,
+	addr = _sram_alloc_max(&free_l1_ssram_head,
 			&used_l1_ssram_head, psize);
 
 	/* add mutex operation */
@@ -555,7 +593,7 @@ int l1sram_free(const void *addr)
 	/* add mutex operation */
 	spin_lock_irqsave(&l1sram_lock, flags);
 
-	ret = _l1_sram_free(addr, &free_l1_ssram_head,
+	ret = _sram_free(addr, &free_l1_ssram_head,
 			&used_l1_ssram_head);
 
 	/* add mutex operation */
@@ -564,6 +602,64 @@ int l1sram_free(const void *addr)
 	return ret;
 }
 
+void *l2_sram_alloc(size_t size)
+{
+#ifdef L2_LENGTH
+	unsigned flags;
+	void *addr;
+
+	/* add mutex operation */
+	spin_lock_irqsave(&l2_sram_lock, flags);
+
+	addr = _sram_alloc(size, &free_l2_sram_head,
+			&used_l2_sram_head);
+
+	/* add mutex operation */
+	spin_unlock_irqrestore(&l2_sram_lock, flags);
+
+	pr_debug("Allocated address in l2_sram_alloc is 0x%lx+0x%lx\n",
+		 (long unsigned int)addr, size);
+
+	return addr;
+#else
+	return NULL;
+#endif
+}
+EXPORT_SYMBOL(l2_sram_alloc);
+
+void *l2_sram_zalloc(size_t size)
+{
+	void *addr = l2_sram_alloc(size);
+
+	if (addr)
+		memset(addr, 0x00, size);
+
+	return addr;
+}
+EXPORT_SYMBOL(l2_sram_zalloc);
+
+int l2_sram_free(const void *addr)
+{
+#ifdef L2_LENGTH
+	unsigned flags;
+	int ret;
+
+	/* add mutex operation */
+	spin_lock_irqsave(&l2_sram_lock, flags);
+
+	ret = _sram_free(addr, &free_l2_sram_head,
+			&used_l2_sram_head);
+
+	/* add mutex operation */
+	spin_unlock_irqrestore(&l2_sram_lock, flags);
+
+	return ret;
+#else
+	return -1;
+#endif
+}
+EXPORT_SYMBOL(l2_sram_free);
+
 int sram_free_with_lsl(const void *addr)
 {
 	struct sram_list_struct *lsl, **tmp;
@@ -602,6 +698,9 @@ void *sram_alloc_with_lsl(size_t size, unsigned long flags)
 	if (addr == NULL && (flags & L1_DATA_B_SRAM))
 		addr = l1_data_B_sram_alloc(size);
 
+	if (addr == NULL && (flags & L2_SRAM))
+		addr = l2_sram_alloc(size);
+
 	if (addr == NULL) {
 		kfree(lsl);
 		return NULL;
@@ -621,7 +720,7 @@ EXPORT_SYMBOL(sram_alloc_with_lsl);
 /* Need to keep line of output the same.  Currently, that is 44 bytes
  * (including newline).
  */
-static int _l1sram_proc_read(char *buf, int *len, int count, const char *desc,
+static int _sram_proc_read(char *buf, int *len, int count, const char *desc,
 		struct sram_piece *pfree_head,
 		struct sram_piece *pused_head)
 {
@@ -630,13 +729,13 @@ static int _l1sram_proc_read(char *buf, int *len, int count, const char *desc,
 	if (!pfree_head || !pused_head)
 		return -1;
 
-	*len += sprintf(&buf[*len], "--- L1 %-14s Size   PID State     \n", desc);
+	*len += sprintf(&buf[*len], "--- SRAM %-14s Size   PID State     \n", desc);
 
 	/* search the relevant memory slot */
 	pslot = pused_head->next;
 
 	while (pslot != NULL) {
-		*len += sprintf(&buf[*len], "%p-%p %8i %5i %-10s\n",
+		*len += sprintf(&buf[*len], "%p-%p %10i %5i %-10s\n",
 			pslot->paddr, pslot->paddr + pslot->size,
 			pslot->size, pslot->pid, "ALLOCATED");
 
@@ -646,7 +745,7 @@ static int _l1sram_proc_read(char *buf, int *len, int count, const char *desc,
 	pslot = pfree_head->next;
 
 	while (pslot != NULL) {
-		*len += sprintf(&buf[*len], "%p-%p %8i %5i %-10s\n",
+		*len += sprintf(&buf[*len], "%p-%p %10i %5i %-10s\n",
 			pslot->paddr, pslot->paddr + pslot->size,
 			pslot->size, pslot->pid, "FREE");
 
@@ -655,38 +754,43 @@ static int _l1sram_proc_read(char *buf, int *len, int count, const char *desc,
 
 	return 0;
 }
-static int l1sram_proc_read(char *buf, char **start, off_t offset, int count,
+static int sram_proc_read(char *buf, char **start, off_t offset, int count,
 		int *eof, void *data)
 {
 	int len = 0;
 
-	if (_l1sram_proc_read(buf, &len, count, "Scratchpad",
+	if (_sram_proc_read(buf, &len, count, "Scratchpad",
 			&free_l1_ssram_head, &used_l1_ssram_head))
 		goto not_done;
 #if L1_DATA_A_LENGTH != 0
-	if (_l1sram_proc_read(buf, &len, count, "Data A",
+	if (_sram_proc_read(buf, &len, count, "L1 Data A",
 			&free_l1_data_A_sram_head,
 			&used_l1_data_A_sram_head))
 		goto not_done;
 #endif
 #if L1_DATA_B_LENGTH != 0
-	if (_l1sram_proc_read(buf, &len, count, "Data B",
+	if (_sram_proc_read(buf, &len, count, "L1 Data B",
 			&free_l1_data_B_sram_head,
 			&used_l1_data_B_sram_head))
 		goto not_done;
 #endif
 #if L1_CODE_LENGTH != 0
-	if (_l1sram_proc_read(buf, &len, count, "Instruction",
+	if (_sram_proc_read(buf, &len, count, "L1 Instruction",
 			&free_l1_inst_sram_head, &used_l1_inst_sram_head))
 		goto not_done;
 #endif
+#ifdef L2_LENGTH
+	if (_sram_proc_read(buf, &len, count, "L2",
+			&free_l2_sram_head, &used_l2_sram_head))
+		goto not_done;
+#endif
 
 	*eof = 1;
  not_done:
 	return len;
 }
 
-static int __init l1sram_proc_init(void)
+static int __init sram_proc_init(void)
 {
 	struct proc_dir_entry *ptr;
 	ptr = create_proc_entry("sram", S_IFREG | S_IRUGO, NULL);
@@ -695,8 +799,8 @@ static int __init l1sram_proc_init(void)
 		return -1;
 	}
 	ptr->owner = THIS_MODULE;
-	ptr->read_proc = l1sram_proc_read;
+	ptr->read_proc = sram_proc_read;
 	return 0;
 }
-late_initcall(l1sram_proc_init);
+late_initcall(sram_proc_init);
 #endif
diff --git a/include/asm-blackfin/bfin-global.h b/include/asm-blackfin/bfin-global.h
index 76033831eb35..320aa5e167e9 100644
--- a/include/asm-blackfin/bfin-global.h
+++ b/include/asm-blackfin/bfin-global.h
@@ -92,16 +92,20 @@ extern void *l1_data_B_sram_alloc(size_t);
 extern void *l1_inst_sram_alloc(size_t);
 extern void *l1_data_sram_alloc(size_t);
 extern void *l1_data_sram_zalloc(size_t);
+extern void *l2_sram_alloc(size_t);
+extern void *l2_sram_zalloc(size_t);
 extern int l1_data_A_sram_free(const void*);
 extern int l1_data_B_sram_free(const void*);
 extern int l1_inst_sram_free(const void*);
 extern int l1_data_sram_free(const void*);
+extern int l2_sram_free(const void *);
 extern int sram_free(const void*);
 
 #define L1_INST_SRAM		0x00000001
 #define L1_DATA_A_SRAM		0x00000002
 #define L1_DATA_B_SRAM		0x00000004
 #define L1_DATA_SRAM		0x00000006
+#define L2_SRAM			0x00000008
 extern void *sram_alloc_with_lsl(size_t, unsigned long);
 extern int sram_free_with_lsl(const void*);
 
@@ -114,7 +118,9 @@ extern struct file_operations dpmc_fops;
 extern unsigned long _ramstart, _ramend, _rambase;
 extern unsigned long memory_start, memory_end, physical_mem_end;
 extern char _stext_l1[], _etext_l1[], _sdata_l1[], _edata_l1[], _sbss_l1[],
-    _ebss_l1[], _l1_lma_start[], _sdata_b_l1[], _ebss_b_l1[];
+	_ebss_l1[], _l1_lma_start[], _sdata_b_l1[], _ebss_b_l1[],
+	_stext_l2[], _etext_l2[], _sdata_l2[], _edata_l2[], _sbss_l2[],
+	_ebss_l2[], _l2_lma_start[];
 
 #ifdef CONFIG_MTD_UCLINUX
 extern unsigned long memory_mtd_start, memory_mtd_end, mtd_size;
diff --git a/include/asm-blackfin/elf.h b/include/asm-blackfin/elf.h
index 30303fc8292c..67a03a8a353e 100644
--- a/include/asm-blackfin/elf.h
+++ b/include/asm-blackfin/elf.h
@@ -15,6 +15,8 @@
 #define EF_BFIN_FDPIC		0x00000002	/* -mfdpic */
 #define EF_BFIN_CODE_IN_L1	0x00000010	/* --code-in-l1 */
 #define EF_BFIN_DATA_IN_L1	0x00000020	/* --data-in-l1 */
+#define EF_BFIN_CODE_IN_L2	0x00000040	/* --code-in-l2 */
+#define EF_BFIN_DATA_IN_L2	0x00000080	/* --data-in-l2 */
 
 typedef unsigned long elf_greg_t;
 
diff --git a/include/asm-blackfin/module.h b/include/asm-blackfin/module.h
index 3c7ce1644280..e3128df139d6 100644
--- a/include/asm-blackfin/module.h
+++ b/include/asm-blackfin/module.h
@@ -6,8 +6,6 @@
 #define Elf_Shdr        Elf32_Shdr
 #define Elf_Sym         Elf32_Sym
 #define Elf_Ehdr        Elf32_Ehdr
-#define FLG_CODE_IN_L1	0x10
-#define FLG_DATA_IN_L1	0x20
 
 struct mod_arch_specific {
 	Elf_Shdr	*text_l1;
@@ -15,5 +13,8 @@ struct mod_arch_specific {
 	Elf_Shdr	*bss_a_l1;
 	Elf_Shdr	*data_b_l1;
 	Elf_Shdr	*bss_b_l1;
+	Elf_Shdr	*text_l2;
+	Elf_Shdr	*data_l2;
+	Elf_Shdr	*bss_l2;
 };
 #endif				/* _ASM_BFIN_MODULE_H */

From e338125b8a886923ba8367207c144764dc352584 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Sat, 19 Jul 2008 09:33:21 +0200
Subject: [PATCH 0076/1171] nohz: adjust tick_nohz_stop_sched_tick() call of
 s390 as well

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/s390/kernel/process.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/s390/kernel/process.c b/arch/s390/kernel/process.c
index 85defd01d293..9839767d0842 100644
--- a/arch/s390/kernel/process.c
+++ b/arch/s390/kernel/process.c
@@ -142,7 +142,7 @@ static void default_idle(void)
 void cpu_idle(void)
 {
 	for (;;) {
-		tick_nohz_stop_sched_tick();
+		tick_nohz_stop_sched_tick(1);
 		while (!need_resched())
 			default_idle();
 		tick_nohz_restart_sched_tick();

From aa3348f461da1df5c583f9916ab80298ddd68eff Mon Sep 17 00:00:00 2001
From: Graf Yang <graf.yang@analog.com>
Date: Sat, 19 Jul 2008 15:54:10 +0800
Subject: [PATCH 0077/1171] Blackfin arch: Add return value check in
 bfin_sir_probe(), remove SSYNC().

Signed-off-by: Graf Yang <graf.yang@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 include/asm-blackfin/mach-bf527/bfin_sir.h | 21 ++++++++----
 include/asm-blackfin/mach-bf533/bfin_sir.h | 13 +++++---
 include/asm-blackfin/mach-bf537/bfin_sir.h | 21 ++++++++----
 include/asm-blackfin/mach-bf548/bfin_sir.h | 37 ++++++++++++++++------
 include/asm-blackfin/mach-bf561/bfin_sir.h | 13 +++++---
 5 files changed, 75 insertions(+), 30 deletions(-)

diff --git a/include/asm-blackfin/mach-bf527/bfin_sir.h b/include/asm-blackfin/mach-bf527/bfin_sir.h
index 0612d0c9501c..cfd8ad4f1f2c 100644
--- a/include/asm-blackfin/mach-bf527/bfin_sir.h
+++ b/include/asm-blackfin/mach-bf527/bfin_sir.h
@@ -118,16 +118,25 @@ static inline void SIR_UART_CLEAR_LSR(struct bfin_sir_port *port)
 
 #define DRIVER_NAME "bfin_sir"
 
-static void bfin_sir_hw_init(void)
+static int bfin_sir_hw_init(void)
 {
+	int ret = -ENODEV;
 #ifdef CONFIG_BFIN_SIR0
-	peripheral_request(P_UART0_TX, DRIVER_NAME);
-	peripheral_request(P_UART0_RX, DRIVER_NAME);
+	ret = peripheral_request(P_UART0_TX, DRIVER_NAME);
+	if (ret)
+		return ret;
+	ret = peripheral_request(P_UART0_RX, DRIVER_NAME);
+	if (ret)
+		return ret;
 #endif
 
 #ifdef CONFIG_BFIN_SIR1
-	peripheral_request(P_UART1_TX, DRIVER_NAME);
-	peripheral_request(P_UART1_RX, DRIVER_NAME);
+	ret = peripheral_request(P_UART1_TX, DRIVER_NAME);
+	if (ret)
+		return ret;
+	ret = peripheral_request(P_UART1_RX, DRIVER_NAME);
+	if (ret)
+		return ret;
 #endif
-	SSYNC();
+	return ret;
 }
diff --git a/include/asm-blackfin/mach-bf533/bfin_sir.h b/include/asm-blackfin/mach-bf533/bfin_sir.h
index cefcf8bb505b..9bb87e9e2e9b 100644
--- a/include/asm-blackfin/mach-bf533/bfin_sir.h
+++ b/include/asm-blackfin/mach-bf533/bfin_sir.h
@@ -110,11 +110,16 @@ static inline void SIR_UART_CLEAR_LSR(struct bfin_sir_port *port)
 
 #define DRIVER_NAME "bfin_sir"
 
-static void bfin_sir_hw_init(void)
+static int bfin_sir_hw_init(void)
 {
+	int ret = -ENODEV;
 #ifdef CONFIG_BFIN_SIR0
-	peripheral_request(P_UART0_TX, DRIVER_NAME);
-	peripheral_request(P_UART0_RX, DRIVER_NAME);
+	ret = peripheral_request(P_UART0_TX, DRIVER_NAME);
+	if (ret)
+		return ret;
+	ret = peripheral_request(P_UART0_RX, DRIVER_NAME);
+	if (ret)
+		return ret;
 #endif
-	SSYNC();
+	return ret;
 }
diff --git a/include/asm-blackfin/mach-bf537/bfin_sir.h b/include/asm-blackfin/mach-bf537/bfin_sir.h
index 0612d0c9501c..cfd8ad4f1f2c 100644
--- a/include/asm-blackfin/mach-bf537/bfin_sir.h
+++ b/include/asm-blackfin/mach-bf537/bfin_sir.h
@@ -118,16 +118,25 @@ static inline void SIR_UART_CLEAR_LSR(struct bfin_sir_port *port)
 
 #define DRIVER_NAME "bfin_sir"
 
-static void bfin_sir_hw_init(void)
+static int bfin_sir_hw_init(void)
 {
+	int ret = -ENODEV;
 #ifdef CONFIG_BFIN_SIR0
-	peripheral_request(P_UART0_TX, DRIVER_NAME);
-	peripheral_request(P_UART0_RX, DRIVER_NAME);
+	ret = peripheral_request(P_UART0_TX, DRIVER_NAME);
+	if (ret)
+		return ret;
+	ret = peripheral_request(P_UART0_RX, DRIVER_NAME);
+	if (ret)
+		return ret;
 #endif
 
 #ifdef CONFIG_BFIN_SIR1
-	peripheral_request(P_UART1_TX, DRIVER_NAME);
-	peripheral_request(P_UART1_RX, DRIVER_NAME);
+	ret = peripheral_request(P_UART1_TX, DRIVER_NAME);
+	if (ret)
+		return ret;
+	ret = peripheral_request(P_UART1_RX, DRIVER_NAME);
+	if (ret)
+		return ret;
 #endif
-	SSYNC();
+	return ret;
 }
diff --git a/include/asm-blackfin/mach-bf548/bfin_sir.h b/include/asm-blackfin/mach-bf548/bfin_sir.h
index 5e94271c7e3b..c41f9cf00268 100644
--- a/include/asm-blackfin/mach-bf548/bfin_sir.h
+++ b/include/asm-blackfin/mach-bf548/bfin_sir.h
@@ -124,26 +124,43 @@ struct bfin_sir_self {
 
 #define DRIVER_NAME "bfin_sir"
 
-static void bfin_sir_hw_init(void)
+static int bfin_sir_hw_init(void)
 {
+	int ret = -ENODEV;
 #ifdef CONFIG_BFIN_SIR0
-	peripheral_request(P_UART0_TX, DRIVER_NAME);
-	peripheral_request(P_UART0_RX, DRIVER_NAME);
+	ret = peripheral_request(P_UART0_TX, DRIVER_NAME);
+	if (ret)
+		return ret;
+	ret = peripheral_request(P_UART0_RX, DRIVER_NAME);
+	if (ret)
+		return ret;
 #endif
 
 #ifdef CONFIG_BFIN_SIR1
-	peripheral_request(P_UART1_TX, DRIVER_NAME);
-	peripheral_request(P_UART1_RX, DRIVER_NAME);
+	ret = peripheral_request(P_UART1_TX, DRIVER_NAME);
+	if (ret)
+		return ret;
+	ret = peripheral_request(P_UART1_RX, DRIVER_NAME);
+	if (ret)
+		return ret;
 #endif
 
 #ifdef CONFIG_BFIN_SIR2
-	peripheral_request(P_UART2_TX, DRIVER_NAME);
-	peripheral_request(P_UART2_RX, DRIVER_NAME);
+	ret = peripheral_request(P_UART2_TX, DRIVER_NAME);
+	if (ret)
+		return ret;
+	ret = peripheral_request(P_UART2_RX, DRIVER_NAME);
+	if (ret)
+		return ret;
 #endif
 
 #ifdef CONFIG_BFIN_SIR3
-	peripheral_request(P_UART3_TX, DRIVER_NAME);
-	peripheral_request(P_UART3_RX, DRIVER_NAME);
+	ret = peripheral_request(P_UART3_TX, DRIVER_NAME);
+	if (ret)
+		return ret;
+	ret = peripheral_request(P_UART3_RX, DRIVER_NAME);
+	if (ret)
+		return ret;
 #endif
-	SSYNC();
+	return ret;
 }
diff --git a/include/asm-blackfin/mach-bf561/bfin_sir.h b/include/asm-blackfin/mach-bf561/bfin_sir.h
index cefcf8bb505b..9bb87e9e2e9b 100644
--- a/include/asm-blackfin/mach-bf561/bfin_sir.h
+++ b/include/asm-blackfin/mach-bf561/bfin_sir.h
@@ -110,11 +110,16 @@ static inline void SIR_UART_CLEAR_LSR(struct bfin_sir_port *port)
 
 #define DRIVER_NAME "bfin_sir"
 
-static void bfin_sir_hw_init(void)
+static int bfin_sir_hw_init(void)
 {
+	int ret = -ENODEV;
 #ifdef CONFIG_BFIN_SIR0
-	peripheral_request(P_UART0_TX, DRIVER_NAME);
-	peripheral_request(P_UART0_RX, DRIVER_NAME);
+	ret = peripheral_request(P_UART0_TX, DRIVER_NAME);
+	if (ret)
+		return ret;
+	ret = peripheral_request(P_UART0_RX, DRIVER_NAME);
+	if (ret)
+		return ret;
 #endif
-	SSYNC();
+	return ret;
 }

From cc2e16bd35f6d78f559467358259e7d7264d4e96 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier.adi@gmail.com>
Date: Sat, 19 Jul 2008 16:43:51 +0800
Subject: [PATCH 0078/1171] Blackfin arch: add missing IORESOURCE_MEM flags to
 UART3

Signed-off-by: Mike Frysinger <vapier.adi@gmail.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/mach-bf548/boards/ezkit.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/blackfin/mach-bf548/boards/ezkit.c b/arch/blackfin/mach-bf548/boards/ezkit.c
index af7c211a580e..547dd253a3f3 100644
--- a/arch/blackfin/mach-bf548/boards/ezkit.c
+++ b/arch/blackfin/mach-bf548/boards/ezkit.c
@@ -177,6 +177,7 @@ static struct resource bfin_uart_resources[] = {
 	{
 		.start = 0xFFC03100,
 		.end = 0xFFC031FF,
+		.flags = IORESOURCE_MEM,
 	},
 #endif
 };

From 0138da6101fa3cdfea7f470d014c2f13cc03e7a9 Mon Sep 17 00:00:00 2001
From: Michael Hennerich <michael.hennerich@analog.com>
Date: Sat, 19 Jul 2008 16:56:53 +0800
Subject: [PATCH 0079/1171] Blackfin arch: fix bug - detect 0.1 silicon
 revision BF527-EZKIT as 0.0 version

Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 include/asm-blackfin/processor.h | 21 ++++++++++++++++++++-
 1 file changed, 20 insertions(+), 1 deletion(-)

diff --git a/include/asm-blackfin/processor.h b/include/asm-blackfin/processor.h
index 1c0040724612..6f3995b119d8 100644
--- a/include/asm-blackfin/processor.h
+++ b/include/asm-blackfin/processor.h
@@ -112,7 +112,26 @@ unsigned long get_wchan(struct task_struct *p);
 static inline uint32_t __pure bfin_revid(void)
 {
 	/* stored in the upper 4 bits */
-	return bfin_read_CHIPID() >> 28;
+	uint32_t revid = bfin_read_CHIPID() >> 28;
+
+#ifdef CONFIG_BF52x
+	/* ANOMALY_05000357
+	 * Incorrect Revision Number in DSPID Register
+	 */
+	if (revid == 0)
+		switch (bfin_read16(_BOOTROM_GET_DXE_ADDRESS_TWI)) {
+		case 0x0010:
+			revid = 0;
+			break;
+		case 0x2796:
+			revid = 1;
+			break;
+		default:
+			revid = 0xFFFF;
+			break;
+		}
+#endif
+	return revid;
 }
 
 static inline uint32_t __pure bfin_compiled_revid(void)

From 0a6304a9517aa3593913ecfbbcad80e798641723 Mon Sep 17 00:00:00 2001
From: Michael Hennerich <michael.hennerich@analog.com>
Date: Sat, 26 Jul 2008 16:14:57 +0800
Subject: [PATCH 0080/1171] Blackfin arch: Add ISP1760 board resources to
 BF548-EZKIT

Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/mach-bf548/boards/ezkit.c | 43 +++++++++++++++++++++++++
 1 file changed, 43 insertions(+)

diff --git a/arch/blackfin/mach-bf548/boards/ezkit.c b/arch/blackfin/mach-bf548/boards/ezkit.c
index 547dd253a3f3..166fa2201ee7 100644
--- a/arch/blackfin/mach-bf548/boards/ezkit.c
+++ b/arch/blackfin/mach-bf548/boards/ezkit.c
@@ -61,6 +61,49 @@ const char bfin_board_name[] = "ADSP-BF548-EZKIT";
  *  Driver needs to know address, irq and flag pin.
  */
 
+#if defined(CONFIG_USB_ISP1760_HCD) || defined(CONFIG_USB_ISP1760_HCD_MODULE)
+static struct resource bfin_isp1761_resources[] = {
+	[0] = {
+		.name	= "isp1761-regs",
+		.start  = 0x2C0C0000,
+		.end    = 0x2C0C0000 + 0xfffff,
+		.flags  = IORESOURCE_MEM,
+	},
+	[1] = {
+		.start  = IRQ_PG7,
+		.end    = IRQ_PG7,
+		.flags  = IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device bfin_isp1761_device = {
+	.name           = "isp1761",
+	.id             = 0,
+	.num_resources  = ARRAY_SIZE(bfin_isp1761_resources),
+	.resource       = bfin_isp1761_resources,
+};
+
+static struct platform_device *bfin_isp1761_devices[] = {
+	&bfin_isp1761_device,
+};
+
+int __init bfin_isp1761_init(void)
+{
+	unsigned int num_devices = ARRAY_SIZE(bfin_isp1761_devices);
+
+	printk(KERN_INFO "%s(): registering device resources\n", __func__);
+	set_irq_type(bfin_isp1761_resources[1].start, IRQF_TRIGGER_FALLING);
+
+	return platform_add_devices(bfin_isp1761_devices, num_devices);
+}
+
+void __exit bfin_isp1761_exit(void)
+{
+	platform_device_unregister(&bfin_isp1761_device);
+}
+arch_initcall(bfin_isp1761_init);
+#endif
+
 #if defined(CONFIG_FB_BF54X_LQ043) || defined(CONFIG_FB_BF54X_LQ043_MODULE)
 
 #include <asm/mach/bf54x-lq043.h>

From f09630bff51daaf427968c61c0f2370c64148e06 Mon Sep 17 00:00:00 2001
From: Robin Getz <rgetz@blackfin.uclinux.org>
Date: Sat, 26 Jul 2008 19:45:46 +0800
Subject: [PATCH 0081/1171] Blackfin arch: Add unwinding for stack info, and a
 little more detail on trace buffer

Signed-off-by: Robin Getz <rgetz@blackfin.uclinux.org>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/kernel/traps.c | 300 ++++++++++++++++++++++++++++-------
 1 file changed, 243 insertions(+), 57 deletions(-)

diff --git a/arch/blackfin/kernel/traps.c b/arch/blackfin/kernel/traps.c
index f061f5181623..ad922ab91543 100644
--- a/arch/blackfin/kernel/traps.c
+++ b/arch/blackfin/kernel/traps.c
@@ -69,8 +69,6 @@ void __init trap_init(void)
 
 unsigned long saved_icplb_fault_addr, saved_dcplb_fault_addr;
 
-int kstack_depth_to_print = 48;
-
 static void decode_address(char *buf, unsigned long address)
 {
 	struct vm_list_struct *vml;
@@ -163,6 +161,9 @@ static void decode_address(char *buf, unsigned long address)
 				if (!in_atomic)
 					mmput(mm);
 
+				if (!strlen(buf))
+					sprintf(buf, "<0x%p> [ %s ] dynamic memory", (void *)address, name);
+
 				goto done;
 			}
 
@@ -173,7 +174,7 @@ static void decode_address(char *buf, unsigned long address)
 	}
 
 	/* we were unable to find this address anywhere */
-	sprintf(buf, "<0x%p> /* unknown address */", (void *)address);
+	sprintf(buf, "<0x%p> /* kernel dynamic memory */", (void *)address);
 
 done:
 	write_unlock_irqrestore(&tasklist_lock, flags);
@@ -494,7 +495,7 @@ asmlinkage void trap_c(struct pt_regs *fp)
 	BUG_ON(sig == 0);
 
 	if (sig != SIGTRAP) {
-		unsigned long stack;
+		unsigned long *stack;
 		dump_bfin_process(fp);
 		dump_bfin_mem(fp);
 		show_regs(fp);
@@ -508,14 +509,23 @@ asmlinkage void trap_c(struct pt_regs *fp)
 		else
 #endif
 			dump_bfin_trace_buffer();
-		show_stack(current, &stack);
+
 		if (oops_in_progress) {
+			/* Dump the current kernel stack */
+			printk(KERN_NOTICE "\n" KERN_NOTICE "Kernel Stack\n");
+			show_stack(current, NULL);
+
 			print_modules();
 #ifndef CONFIG_ACCESS_CHECK
 			printk(KERN_EMERG "Please turn on "
 			       "CONFIG_ACCESS_CHECK\n");
 #endif
 			panic("Kernel exception");
+		} else {
+			/* Dump the user space stack */
+			stack = (unsigned long *)rdusp();
+			printk(KERN_NOTICE "Userspace Stack\n");
+			show_stack(NULL, stack);
 		}
 	}
 
@@ -532,11 +542,71 @@ asmlinkage void trap_c(struct pt_regs *fp)
 
 #define EXPAND_LEN ((1 << CONFIG_DEBUG_BFIN_HWTRACE_EXPAND_LEN) * 256 - 1)
 
+/*
+ * Similar to get_user, do some address checking, then dereference
+ * Return true on sucess, false on bad address
+ */
+bool get_instruction(unsigned short *val, unsigned short *address)
+{
+
+	unsigned long addr;
+
+	addr = (unsigned long)address;
+
+	/* Check for odd addresses */
+	if (addr & 0x1)
+		return false;
+
+	/* Check that things do not wrap around */
+	if (addr > (addr + 2))
+		return false;
+
+	/*
+	 * Since we are in exception context, we need to do a little address checking
+	 * We need to make sure we are only accessing valid memory, and
+	 * we don't read something in the async space that can hang forever
+	 */
+	if ((addr >= FIXED_CODE_START && (addr + 2) <= physical_mem_end) ||
+#ifdef L2_START
+	    (addr >= L2_START && (addr + 2) <= (L2_START + L2_LENGTH)) ||
+#endif
+	    (addr >= BOOT_ROM_START && (addr + 2) <= (BOOT_ROM_START + BOOT_ROM_LENGTH)) ||
+#if L1_DATA_A_LENGTH != 0
+	    (addr >= L1_DATA_A_START && (addr + 2) <= (L1_DATA_A_START + L1_DATA_A_LENGTH)) ||
+#endif
+#if L1_DATA_B_LENGTH != 0
+	    (addr >= L1_DATA_B_START && (addr + 2) <= (L1_DATA_B_START + L1_DATA_B_LENGTH)) ||
+#endif
+	    (addr >= L1_SCRATCH_START && (addr + 2) <= (L1_SCRATCH_START + L1_SCRATCH_LENGTH)) ||
+	    (!(bfin_read_EBIU_AMBCTL0() & B0RDYEN) &&
+	       addr >= ASYNC_BANK0_BASE && (addr + 2) <= (ASYNC_BANK0_BASE + ASYNC_BANK0_SIZE)) ||
+	    (!(bfin_read_EBIU_AMBCTL0() & B1RDYEN) &&
+	       addr >= ASYNC_BANK1_BASE && (addr + 2) <= (ASYNC_BANK1_BASE + ASYNC_BANK1_SIZE)) ||
+	    (!(bfin_read_EBIU_AMBCTL1() & B2RDYEN) &&
+	       addr >= ASYNC_BANK2_BASE && (addr + 2) <= (ASYNC_BANK2_BASE + ASYNC_BANK1_SIZE)) ||
+	    (!(bfin_read_EBIU_AMBCTL1() & B3RDYEN) &&
+	      addr >= ASYNC_BANK3_BASE && (addr + 2) <= (ASYNC_BANK3_BASE + ASYNC_BANK1_SIZE))) {
+		*val = *address;
+		return true;
+	}
+
+#if L1_CODE_LENGTH != 0
+	if (addr >= L1_CODE_START && (addr + 2) <= (L1_CODE_START + L1_CODE_LENGTH)) {
+		dma_memcpy(val, address, 2);
+		return true;
+	}
+#endif
+
+
+	return false;
+}
+
 void dump_bfin_trace_buffer(void)
 {
 #ifdef CONFIG_DEBUG_BFIN_HWTRACE_ON
 	int tflags, i = 0;
 	char buf[150];
+	unsigned short val = 0, *addr;
 #ifdef CONFIG_DEBUG_BFIN_HWTRACE_EXPAND
 	int j, index;
 #endif
@@ -549,8 +619,42 @@ void dump_bfin_trace_buffer(void)
 		for (; bfin_read_TBUFSTAT() & TBUFCNT; i++) {
 			decode_address(buf, (unsigned long)bfin_read_TBUF());
 			printk(KERN_NOTICE "%4i Target : %s\n", i, buf);
-			decode_address(buf, (unsigned long)bfin_read_TBUF());
-			printk(KERN_NOTICE "     Source : %s\n", buf);
+			addr = (unsigned short *)bfin_read_TBUF();
+			decode_address(buf, (unsigned long)addr);
+			printk(KERN_NOTICE "     Source : %s ", buf);
+			if (get_instruction(&val, addr)) {
+				if (val == 0x0010)
+					printk("RTS");
+				else if (val == 0x0011)
+					printk("RTI");
+				else if (val == 0x0012)
+					printk("RTX");
+				else if (val >= 0x0050 && val <= 0x0057)
+					printk("JUMP (P%i)", val & 7);
+				else if (val >= 0x0060 && val <= 0x0067)
+					printk("CALL (P%i)", val & 7);
+				else if (val >= 0x0070 && val <= 0x0077)
+					printk("CALL (PC+P%i)", val & 7);
+				else if (val >= 0x0080 && val <= 0x0087)
+					printk("JUMP (PC+P%i)", val & 7);
+				else if ((val >= 0x1000 && val <= 0x13FF) ||
+				    (val >= 0x1800 && val <= 0x1BFF))
+					printk("IF !CC JUMP");
+				else if ((val >= 0x1400 && val <= 0x17ff) ||
+				    (val >= 0x1c00 && val <= 0x1fff))
+					printk("IF CC JUMP");
+				else if (val >= 0x2000 && val <= 0x2fff)
+					printk("JUMP.S");
+				else if (val >= 0xe080 && val <= 0xe0ff)
+					printk("LSETUP");
+				else if (val >= 0xe200 && val <= 0xe2ff)
+					printk("JUMP.L");
+				else if (val >= 0xe300 && val <= 0xe3ff)
+					printk("CALL pcrel");
+				else
+					printk("0x%04x", val);
+			}
+			printk("\n");
 		}
 	}
 
@@ -582,59 +686,151 @@ void dump_bfin_trace_buffer(void)
 }
 EXPORT_SYMBOL(dump_bfin_trace_buffer);
 
-static void show_trace(struct task_struct *tsk, unsigned long *sp)
+/*
+ * Checks to see if the address pointed to is either a
+ * 16-bit CALL instruction, or a 32-bit CALL instruction
+ */
+bool is_bfin_call(unsigned short *addr)
 {
-	unsigned long addr;
+	unsigned short opcode = 0, *ins_addr;
+	ins_addr = (unsigned short *)addr;
 
-	printk(KERN_NOTICE "\n" KERN_NOTICE "Call Trace:\n");
+	if (!get_instruction(&opcode, ins_addr))
+		return false;
 
-	while (!kstack_end(sp)) {
-		addr = *sp++;
-		/*
-		 * If the address is either in the text segment of the
-		 * kernel, or in the region which contains vmalloc'ed
-		 * memory, it *may* be the address of a calling
-		 * routine; if so, print it so that someone tracing
-		 * down the cause of the crash will be able to figure
-		 * out the call path that was taken.
-		 */
-		if (kernel_text_address(addr))
-			print_ip_sym(addr);
-	}
+	if ((opcode >= 0x0060 && opcode <= 0x0067) ||
+	    (opcode >= 0x0070 && opcode <= 0x0077))
+		return true;
+
+	ins_addr--;
+	if (!get_instruction(&opcode, ins_addr))
+		return false;
+
+	if (opcode >= 0xE300 && opcode <= 0xE3FF)
+		return true;
+
+	return false;
 
-	printk(KERN_NOTICE "\n");
 }
-
 void show_stack(struct task_struct *task, unsigned long *stack)
 {
-	unsigned long *endstack, addr;
-	int i;
+	unsigned int *addr, *endstack, *fp = 0, *frame;
+	unsigned short *ins_addr;
+	char buf[150];
+	unsigned int i, j, ret_addr, frame_no = 0;
 
-	/* Cannot call dump_bfin_trace_buffer() here as show_stack() is
-	 * called externally in some places in the kernel.
+	/*
+	 * If we have been passed a specific stack, use that one otherwise
+	 *    if we have been passed a task structure, use that, otherwise
+	 *    use the stack of where the variable "stack" exists
 	 */
 
-	if (!stack) {
-		if (task)
+	if (stack == NULL) {
+		if (task) {
+			/* We know this is a kernel stack, so this is the start/end */
 			stack = (unsigned long *)task->thread.ksp;
-		else
+			endstack = (unsigned int *)(((unsigned int)(stack) & ~(THREAD_SIZE - 1)) + THREAD_SIZE);
+		} else {
+			/* print out the existing stack info */
 			stack = (unsigned long *)&stack;
+			endstack = (unsigned int *)PAGE_ALIGN((unsigned int)stack);
+		}
+	} else
+		endstack = (unsigned int *)PAGE_ALIGN((unsigned int)stack);
+
+	decode_address(buf, (unsigned int)stack);
+	printk(KERN_NOTICE "Stack info:\n" KERN_NOTICE " SP: [0x%p] %s\n", stack, buf);
+	addr = (unsigned int *)((unsigned int)stack & ~0x3F);
+
+	/* First thing is to look for a frame pointer */
+	for (addr = (unsigned int *)((unsigned int)stack & ~0xF), i = 0;
+		addr < endstack; addr++, i++) {
+		if (*addr & 0x1)
+			continue;
+		ins_addr = (unsigned short *)*addr;
+		ins_addr--;
+		if (is_bfin_call(ins_addr))
+			fp = addr - 1;
+
+		if (fp) {
+			/* Let's check to see if it is a frame pointer */
+			while (fp >= (addr - 1) && fp < endstack && fp)
+				fp = (unsigned int *)*fp;
+			if (fp == 0 || fp == endstack) {
+				fp = addr - 1;
+				break;
+			}
+			fp = 0;
+		}
+	}
+	if (fp) {
+		frame = fp;
+		printk(" FP: (0x%p)\n", fp);
+	} else
+		frame = 0;
+
+	/*
+	 * Now that we think we know where things are, we
+	 * walk the stack again, this time printing things out
+	 * incase there is no frame pointer, we still look for
+	 * valid return addresses
+	 */
+
+	/* First time print out data, next time, print out symbols */
+	for (j = 0; j <= 1; j++) {
+		if (j)
+			printk(KERN_NOTICE "Return addresses in stack:\n");
+		else
+			printk(KERN_NOTICE " Memory from 0x%08lx to %p", ((long unsigned int)stack & ~0xF), endstack);
+
+		fp = frame;
+		frame_no = 0;
+
+		for (addr = (unsigned int *)((unsigned int)stack & ~0xF), i = 0;
+		     addr <= endstack; addr++, i++) {
+
+			ret_addr = 0;
+			if (!j && i % 8 == 0)
+				printk("\n" KERN_NOTICE "%p:",addr);
+
+			/* if it is an odd address, or zero, just skip it */
+			if (*addr & 0x1 || !*addr)
+				goto print;
+
+			ins_addr = (unsigned short *)*addr;
+
+			/* Go back one instruction, and see if it is a CALL */
+			ins_addr--;
+			ret_addr = is_bfin_call(ins_addr);
+ print:
+			if (!j && stack == (unsigned long *)addr)
+				printk("[%08x]", *addr);
+			else if (ret_addr)
+				if (j) {
+					decode_address(buf, (unsigned int)*addr);
+					if (frame == addr) {
+						printk(KERN_NOTICE "   frame %2i : %s\n", frame_no, buf);
+						continue;
+					}
+					printk(KERN_NOTICE "    address : %s\n", buf);
+				} else
+					printk("<%08x>", *addr);
+			else if (fp == addr) {
+				if (j)
+					frame = addr+1;
+				else
+					printk("(%08x)", *addr);
+
+				fp = (unsigned int *)*addr;
+				frame_no++;
+
+			} else if (!j)
+				printk(" %08x ", *addr);
+		}
+		if (!j)
+			printk("\n");
 	}
 
-	addr = (unsigned long)stack;
-	endstack = (unsigned long *)PAGE_ALIGN(addr);
-
-	printk(KERN_NOTICE "Stack from %08lx:", (unsigned long)stack);
-	for (i = 0; i < kstack_depth_to_print; i++) {
-		if (stack + 1 > endstack)
-			break;
-		if (i % 8 == 0)
-			printk("\n" KERN_NOTICE "       ");
-		printk(" %08lx", *stack++);
-	}
-	printk("\n");
-
-	show_trace(task, stack);
 }
 
 void dump_stack(void)
@@ -715,19 +911,9 @@ void dump_bfin_mem(struct pt_regs *fp)
 		if (!((unsigned long)addr & 0xF))
 			printk("\n" KERN_NOTICE "0x%p: ", addr);
 
-		if (get_user(val, addr)) {
-			if (addr >= (unsigned short *)L1_CODE_START &&
-			    addr < (unsigned short *)(L1_CODE_START + L1_CODE_LENGTH)) {
-				dma_memcpy(&val, addr, sizeof(val));
-				sprintf(buf, "%04x", val);
-			} else if (addr >= (unsigned short *)FIXED_CODE_START &&
-				addr <= (unsigned short *)memory_start) {
-				val = bfin_read16(addr);
-				sprintf(buf, "%04x", val);
-			} else {
+		if (get_instruction(&val, addr)) {
 				val = 0;
 				sprintf(buf, "????");
-			}
 		} else
 			sprintf(buf, "%04x", val);
 

From 9db144fe897bb09838294aab2229cb423ab40988 Mon Sep 17 00:00:00 2001
From: Michael Hennerich <michael.hennerich@analog.com>
Date: Sat, 19 Jul 2008 17:16:07 +0800
Subject: [PATCH 0082/1171] Blackfin arch: Apply Bluetechnix CM-BF527 board
 support patch

Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/Kconfig                      |    5 +-
 arch/blackfin/configs/CM-BF527_defconfig   | 1185 ++++++++++++++++++++
 arch/blackfin/mach-bf527/boards/Kconfig    |    5 +
 arch/blackfin/mach-bf527/boards/Makefile   |    1 +
 arch/blackfin/mach-bf527/boards/cm_bf527.c | 1011 +++++++++++++++++
 5 files changed, 2204 insertions(+), 3 deletions(-)
 create mode 100644 arch/blackfin/configs/CM-BF527_defconfig
 create mode 100644 arch/blackfin/mach-bf527/boards/cm_bf527.c

diff --git a/arch/blackfin/Kconfig b/arch/blackfin/Kconfig
index 5a08194e4118..efa77f3a30d8 100644
--- a/arch/blackfin/Kconfig
+++ b/arch/blackfin/Kconfig
@@ -234,7 +234,7 @@ config MEM_MT48LC16M16A2TG_75
 	bool
 	depends on (BFIN533_EZKIT || BFIN561_EZKIT \
 		|| BFIN533_BLUETECHNIX_CM || BFIN537_BLUETECHNIX_CM \
-		|| H8606_HVSISTEMAS)
+		|| H8606_HVSISTEMAS || BFIN527_BLUETECHNIX_CM)
 	default y
 
 config MEM_MT48LC32M8A2_75
@@ -330,8 +330,7 @@ config VCO_MULT
 	default "45" if BFIN533_STAMP
 	default "20" if (BFIN537_STAMP || BFIN527_EZKIT || BFIN548_EZKIT || BFIN548_BLUETECHNIX_CM)
 	default "22" if BFIN533_BLUETECHNIX_CM
-	default "20" if BFIN537_BLUETECHNIX_CM
-	default "20" if BFIN561_BLUETECHNIX_CM
+	default "20" if (BFIN537_BLUETECHNIX_CM || BFIN527_BLUETECHNIX_CM || BFIN561_BLUETECHNIX_CM)
 	default "20" if BFIN561_EZKIT
 	default "16" if H8606_HVSISTEMAS
 	help
diff --git a/arch/blackfin/configs/CM-BF527_defconfig b/arch/blackfin/configs/CM-BF527_defconfig
new file mode 100644
index 000000000000..0799aa9bba9d
--- /dev/null
+++ b/arch/blackfin/configs/CM-BF527_defconfig
@@ -0,0 +1,1185 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.24.7
+# Fri Jul 18 18:00:41 2008
+#
+# CONFIG_MMU is not set
+# CONFIG_FPU is not set
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+# CONFIG_RWSEM_XCHGADD_ALGORITHM is not set
+CONFIG_BLACKFIN=y
+CONFIG_ZONE_DMA=y
+CONFIG_SEMAPHORE_SLEEPERS=y
+CONFIG_GENERIC_FIND_NEXT_BIT=y
+CONFIG_GENERIC_HWEIGHT=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_IRQ_PROBE=y
+CONFIG_GENERIC_GPIO=y
+CONFIG_FORCE_MAX_ZONEORDER=14
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
+
+#
+# General setup
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+CONFIG_LOCALVERSION=""
+CONFIG_LOCALVERSION_AUTO=y
+CONFIG_SYSVIPC=y
+CONFIG_SYSVIPC_SYSCTL=y
+# CONFIG_POSIX_MQUEUE is not set
+# CONFIG_BSD_PROCESS_ACCT is not set
+# CONFIG_TASKSTATS is not set
+# CONFIG_USER_NS is not set
+# CONFIG_PID_NS is not set
+# CONFIG_AUDIT is not set
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
+CONFIG_LOG_BUF_SHIFT=14
+# CONFIG_CGROUPS is not set
+CONFIG_FAIR_GROUP_SCHED=y
+CONFIG_FAIR_USER_SCHED=y
+# CONFIG_FAIR_CGROUP_SCHED is not set
+CONFIG_SYSFS_DEPRECATED=y
+# CONFIG_RELAY is not set
+CONFIG_BLK_DEV_INITRD=y
+CONFIG_INITRAMFS_SOURCE=""
+# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
+CONFIG_SYSCTL=y
+CONFIG_EMBEDDED=y
+CONFIG_UID16=y
+CONFIG_SYSCTL_SYSCALL=y
+CONFIG_KALLSYMS=y
+# CONFIG_KALLSYMS_EXTRA_PASS is not set
+CONFIG_HOTPLUG=y
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_ELF_CORE=y
+CONFIG_BASE_FULL=y
+CONFIG_FUTEX=y
+CONFIG_ANON_INODES=y
+CONFIG_EPOLL=y
+CONFIG_SIGNALFD=y
+CONFIG_EVENTFD=y
+CONFIG_VM_EVENT_COUNTERS=y
+CONFIG_SLAB=y
+# CONFIG_SLUB is not set
+# CONFIG_SLOB is not set
+CONFIG_SLABINFO=y
+CONFIG_RT_MUTEXES=y
+CONFIG_TINY_SHMEM=y
+CONFIG_BASE_SMALL=0
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+# CONFIG_MODULE_FORCE_UNLOAD is not set
+# CONFIG_MODVERSIONS is not set
+# CONFIG_MODULE_SRCVERSION_ALL is not set
+CONFIG_KMOD=y
+CONFIG_BLOCK=y
+# CONFIG_LBD is not set
+# CONFIG_BLK_DEV_IO_TRACE is not set
+# CONFIG_LSF is not set
+# CONFIG_BLK_DEV_BSG is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+# CONFIG_IOSCHED_AS is not set
+# CONFIG_IOSCHED_DEADLINE is not set
+CONFIG_IOSCHED_CFQ=y
+# CONFIG_DEFAULT_AS is not set
+# CONFIG_DEFAULT_DEADLINE is not set
+CONFIG_DEFAULT_CFQ=y
+# CONFIG_DEFAULT_NOOP is not set
+CONFIG_DEFAULT_IOSCHED="cfq"
+# CONFIG_PREEMPT_NONE is not set
+CONFIG_PREEMPT_VOLUNTARY=y
+# CONFIG_PREEMPT is not set
+
+#
+# Blackfin Processor Options
+#
+
+#
+# Processor and Board Settings
+#
+# CONFIG_BF522 is not set
+# CONFIG_BF523 is not set
+# CONFIG_BF524 is not set
+# CONFIG_BF525 is not set
+# CONFIG_BF526 is not set
+CONFIG_BF527=y
+# CONFIG_BF531 is not set
+# CONFIG_BF532 is not set
+# CONFIG_BF533 is not set
+# CONFIG_BF534 is not set
+# CONFIG_BF536 is not set
+# CONFIG_BF537 is not set
+# CONFIG_BF542 is not set
+# CONFIG_BF544 is not set
+# CONFIG_BF547 is not set
+# CONFIG_BF548 is not set
+# CONFIG_BF549 is not set
+# CONFIG_BF561 is not set
+# CONFIG_BF_REV_0_0 is not set
+CONFIG_BF_REV_0_1=y
+# CONFIG_BF_REV_0_2 is not set
+# CONFIG_BF_REV_0_3 is not set
+# CONFIG_BF_REV_0_4 is not set
+# CONFIG_BF_REV_0_5 is not set
+# CONFIG_BF_REV_ANY is not set
+# CONFIG_BF_REV_NONE is not set
+CONFIG_BF52x=y
+CONFIG_MEM_MT48LC16M16A2TG_75=y
+# CONFIG_BFIN527_EZKIT is not set
+CONFIG_BFIN527_BLUETECHNIX_CM=y
+
+#
+# BF527 Specific Configuration
+#
+
+#
+# Alternative Multiplexing Scheme
+#
+# CONFIG_BF527_SPORT0_PORTF is not set
+CONFIG_BF527_SPORT0_PORTG=y
+CONFIG_BF527_SPORT0_TSCLK_PG10=y
+# CONFIG_BF527_SPORT0_TSCLK_PG14 is not set
+CONFIG_BF527_UART1_PORTF=y
+# CONFIG_BF527_UART1_PORTG is not set
+# CONFIG_BF527_NAND_D_PORTF is not set
+CONFIG_BF527_NAND_D_PORTH=y
+
+#
+# Interrupt Priority Assignment
+#
+
+#
+# Priority
+#
+CONFIG_IRQ_PLL_WAKEUP=7
+CONFIG_IRQ_DMA0_ERROR=7
+CONFIG_IRQ_DMAR0_BLK=7
+CONFIG_IRQ_DMAR1_BLK=7
+CONFIG_IRQ_DMAR0_OVR=7
+CONFIG_IRQ_DMAR1_OVR=7
+CONFIG_IRQ_PPI_ERROR=7
+CONFIG_IRQ_MAC_ERROR=7
+CONFIG_IRQ_SPORT0_ERROR=7
+CONFIG_IRQ_SPORT1_ERROR=7
+CONFIG_IRQ_UART0_ERROR=7
+CONFIG_IRQ_UART1_ERROR=7
+CONFIG_IRQ_RTC=8
+CONFIG_IRQ_PPI=8
+CONFIG_IRQ_SPORT0_RX=9
+CONFIG_IRQ_SPORT0_TX=9
+CONFIG_IRQ_SPORT1_RX=9
+CONFIG_IRQ_SPORT1_TX=9
+CONFIG_IRQ_TWI=10
+CONFIG_IRQ_SPI=10
+CONFIG_IRQ_UART0_RX=10
+CONFIG_IRQ_UART0_TX=10
+CONFIG_IRQ_UART1_RX=10
+CONFIG_IRQ_UART1_TX=10
+CONFIG_IRQ_OPTSEC=11
+CONFIG_IRQ_CNT=11
+CONFIG_IRQ_MAC_RX=11
+CONFIG_IRQ_PORTH_INTA=11
+CONFIG_IRQ_MAC_TX=11
+CONFIG_IRQ_PORTH_INTB=11
+CONFIG_IRQ_TMR0=12
+CONFIG_IRQ_TMR1=12
+CONFIG_IRQ_TMR2=12
+CONFIG_IRQ_TMR3=12
+CONFIG_IRQ_TMR4=12
+CONFIG_IRQ_TMR5=12
+CONFIG_IRQ_TMR6=12
+CONFIG_IRQ_TMR7=12
+CONFIG_IRQ_PORTG_INTA=12
+CONFIG_IRQ_PORTG_INTB=12
+CONFIG_IRQ_MEM_DMA0=13
+CONFIG_IRQ_MEM_DMA1=13
+CONFIG_IRQ_WATCH=13
+CONFIG_IRQ_PORTF_INTA=13
+CONFIG_IRQ_PORTF_INTB=13
+CONFIG_IRQ_SPI_ERROR=7
+CONFIG_IRQ_NFC_ERROR=7
+CONFIG_IRQ_HDMA_ERROR=7
+CONFIG_IRQ_HDMA=7
+CONFIG_IRQ_USB_EINT=10
+CONFIG_IRQ_USB_INT0=11
+CONFIG_IRQ_USB_INT1=11
+CONFIG_IRQ_USB_INT2=11
+CONFIG_IRQ_USB_DMA=11
+
+#
+# Board customizations
+#
+# CONFIG_CMDLINE_BOOL is not set
+CONFIG_BOOT_LOAD=0x1000
+
+#
+# Clock/PLL Setup
+#
+CONFIG_CLKIN_HZ=25000000
+# CONFIG_BFIN_KERNEL_CLOCK is not set
+CONFIG_MAX_MEM_SIZE=512
+CONFIG_MAX_VCO_HZ=600000000
+CONFIG_MIN_VCO_HZ=50000000
+CONFIG_MAX_SCLK_HZ=133333333
+CONFIG_MIN_SCLK_HZ=27000000
+
+#
+# Kernel Timer/Scheduler
+#
+# CONFIG_HZ_100 is not set
+CONFIG_HZ_250=y
+# CONFIG_HZ_300 is not set
+# CONFIG_HZ_1000 is not set
+CONFIG_HZ=250
+CONFIG_GENERIC_TIME=y
+CONFIG_GENERIC_CLOCKEVENTS=y
+# CONFIG_CYCLES_CLOCKSOURCE is not set
+# CONFIG_TICK_ONESHOT is not set
+# CONFIG_NO_HZ is not set
+# CONFIG_HIGH_RES_TIMERS is not set
+CONFIG_GENERIC_CLOCKEVENTS_BUILD=y
+
+#
+# Misc
+#
+CONFIG_BFIN_SCRATCH_REG_RETN=y
+# CONFIG_BFIN_SCRATCH_REG_RETE is not set
+# CONFIG_BFIN_SCRATCH_REG_CYCLES is not set
+
+#
+# Blackfin Kernel Optimizations
+#
+
+#
+# Memory Optimizations
+#
+CONFIG_I_ENTRY_L1=y
+CONFIG_EXCPT_IRQ_SYSC_L1=y
+CONFIG_DO_IRQ_L1=y
+CONFIG_CORE_TIMER_IRQ_L1=y
+CONFIG_IDLE_L1=y
+# CONFIG_SCHEDULE_L1 is not set
+CONFIG_ARITHMETIC_OPS_L1=y
+CONFIG_ACCESS_OK_L1=y
+# CONFIG_MEMSET_L1 is not set
+# CONFIG_MEMCPY_L1 is not set
+# CONFIG_SYS_BFIN_SPINLOCK_L1 is not set
+# CONFIG_IP_CHECKSUM_L1 is not set
+CONFIG_CACHELINE_ALIGNED_L1=y
+# CONFIG_SYSCALL_TAB_L1 is not set
+# CONFIG_CPLB_SWITCH_TAB_L1 is not set
+CONFIG_RAMKERNEL=y
+# CONFIG_ROMKERNEL is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+# CONFIG_SPARSEMEM_STATIC is not set
+# CONFIG_SPARSEMEM_VMEMMAP_ENABLE is not set
+CONFIG_SPLIT_PTLOCK_CPUS=4
+# CONFIG_RESOURCES_64BIT is not set
+CONFIG_ZONE_DMA_FLAG=1
+CONFIG_VIRT_TO_BUS=y
+CONFIG_BFIN_GPTIMERS=y
+CONFIG_BFIN_DMA_5XX=y
+# CONFIG_DMA_UNCACHED_4M is not set
+# CONFIG_DMA_UNCACHED_2M is not set
+CONFIG_DMA_UNCACHED_1M=y
+# CONFIG_DMA_UNCACHED_NONE is not set
+
+#
+# Cache Support
+#
+CONFIG_BFIN_ICACHE=y
+CONFIG_BFIN_DCACHE=y
+# CONFIG_BFIN_DCACHE_BANKA is not set
+# CONFIG_BFIN_ICACHE_LOCK is not set
+# CONFIG_BFIN_WB is not set
+CONFIG_BFIN_WT=y
+# CONFIG_MPU is not set
+
+#
+# Asynchonous Memory Configuration
+#
+
+#
+# EBIU_AMGCTL Global Control
+#
+CONFIG_C_AMCKEN=y
+CONFIG_C_CDPRIO=y
+# CONFIG_C_AMBEN is not set
+# CONFIG_C_AMBEN_B0 is not set
+# CONFIG_C_AMBEN_B0_B1 is not set
+# CONFIG_C_AMBEN_B0_B1_B2 is not set
+CONFIG_C_AMBEN_ALL=y
+
+#
+# EBIU_AMBCTL Control
+#
+CONFIG_BANK_0=0x7BB0
+CONFIG_BANK_1=0x5554
+CONFIG_BANK_2=0x7BB0
+CONFIG_BANK_3=0xFFC0
+
+#
+# Bus options (PCI, PCMCIA, EISA, MCA, ISA)
+#
+# CONFIG_PCI is not set
+# CONFIG_ARCH_SUPPORTS_MSI is not set
+# CONFIG_PCCARD is not set
+
+#
+# Executable file formats
+#
+CONFIG_BINFMT_ELF_FDPIC=y
+CONFIG_BINFMT_FLAT=y
+CONFIG_BINFMT_ZFLAT=y
+# CONFIG_BINFMT_SHARED_FLAT is not set
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Power management options
+#
+# CONFIG_PM is not set
+CONFIG_SUSPEND_UP_POSSIBLE=y
+# CONFIG_PM_BFIN_SLEEP_DEEPER is not set
+# CONFIG_PM_BFIN_SLEEP is not set
+# CONFIG_PM_WAKEUP_BY_GPIO is not set
+
+#
+# CPU Frequency scaling
+#
+# CONFIG_CPU_FREQ is not set
+
+#
+# Networking
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+CONFIG_PACKET=y
+# CONFIG_PACKET_MMAP is not set
+CONFIG_UNIX=y
+CONFIG_XFRM=y
+# CONFIG_XFRM_USER is not set
+# CONFIG_XFRM_SUB_POLICY is not set
+# CONFIG_XFRM_MIGRATE is not set
+# CONFIG_NET_KEY is not set
+CONFIG_INET=y
+# CONFIG_IP_MULTICAST is not set
+# CONFIG_IP_ADVANCED_ROUTER is not set
+CONFIG_IP_FIB_HASH=y
+CONFIG_IP_PNP=y
+# CONFIG_IP_PNP_DHCP is not set
+# CONFIG_IP_PNP_BOOTP is not set
+# CONFIG_IP_PNP_RARP is not set
+# CONFIG_NET_IPIP is not set
+# CONFIG_NET_IPGRE is not set
+# CONFIG_ARPD is not set
+CONFIG_SYN_COOKIES=y
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_XFRM_TUNNEL is not set
+# CONFIG_INET_TUNNEL is not set
+CONFIG_INET_XFRM_MODE_TRANSPORT=y
+CONFIG_INET_XFRM_MODE_TUNNEL=y
+CONFIG_INET_XFRM_MODE_BEET=y
+# CONFIG_INET_LRO is not set
+CONFIG_INET_DIAG=y
+CONFIG_INET_TCP_DIAG=y
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_CUBIC=y
+CONFIG_DEFAULT_TCP_CONG="cubic"
+# CONFIG_TCP_MD5SIG is not set
+# CONFIG_IPV6 is not set
+# CONFIG_INET6_XFRM_TUNNEL is not set
+# CONFIG_INET6_TUNNEL is not set
+# CONFIG_NETLABEL is not set
+# CONFIG_NETWORK_SECMARK is not set
+# CONFIG_NETFILTER is not set
+# CONFIG_IP_DCCP is not set
+# CONFIG_IP_SCTP is not set
+# CONFIG_TIPC is not set
+# CONFIG_ATM is not set
+# CONFIG_BRIDGE is not set
+# CONFIG_VLAN_8021Q is not set
+# CONFIG_DECNET is not set
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+# CONFIG_NET_SCHED is not set
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+# CONFIG_AF_RXRPC is not set
+
+#
+# Wireless
+#
+# CONFIG_CFG80211 is not set
+# CONFIG_WIRELESS_EXT is not set
+# CONFIG_MAC80211 is not set
+# CONFIG_IEEE80211 is not set
+# CONFIG_RFKILL is not set
+# CONFIG_NET_9P is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+# CONFIG_FW_LOADER is not set
+# CONFIG_SYS_HYPERVISOR is not set
+# CONFIG_CONNECTOR is not set
+CONFIG_MTD=y
+# CONFIG_MTD_DEBUG is not set
+# CONFIG_MTD_CONCAT is not set
+CONFIG_MTD_PARTITIONS=y
+# CONFIG_MTD_REDBOOT_PARTS is not set
+# CONFIG_MTD_CMDLINE_PARTS is not set
+
+#
+# User Modules And Translation Layers
+#
+CONFIG_MTD_CHAR=m
+CONFIG_MTD_BLKDEVS=y
+CONFIG_MTD_BLOCK=y
+# CONFIG_FTL is not set
+# CONFIG_NFTL is not set
+# CONFIG_INFTL is not set
+# CONFIG_RFD_FTL is not set
+# CONFIG_SSFDC is not set
+# CONFIG_MTD_OOPS is not set
+
+#
+# RAM/ROM/Flash chip drivers
+#
+# CONFIG_MTD_CFI is not set
+CONFIG_MTD_JEDECPROBE=m
+CONFIG_MTD_GEN_PROBE=m
+# CONFIG_MTD_CFI_ADV_OPTIONS is not set
+CONFIG_MTD_MAP_BANK_WIDTH_1=y
+CONFIG_MTD_MAP_BANK_WIDTH_2=y
+CONFIG_MTD_MAP_BANK_WIDTH_4=y
+# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set
+CONFIG_MTD_CFI_I1=y
+CONFIG_MTD_CFI_I2=y
+# CONFIG_MTD_CFI_I4 is not set
+# CONFIG_MTD_CFI_I8 is not set
+# CONFIG_MTD_CFI_INTELEXT is not set
+# CONFIG_MTD_CFI_AMDSTD is not set
+# CONFIG_MTD_CFI_STAA is not set
+CONFIG_MTD_RAM=y
+CONFIG_MTD_ROM=m
+# CONFIG_MTD_ABSENT is not set
+
+#
+# Mapping drivers for chip access
+#
+CONFIG_MTD_COMPLEX_MAPPINGS=y
+# CONFIG_MTD_PHYSMAP is not set
+# CONFIG_MTD_UCLINUX is not set
+# CONFIG_MTD_PLATRAM is not set
+
+#
+# Self-contained MTD device drivers
+#
+# CONFIG_MTD_DATAFLASH is not set
+# CONFIG_MTD_M25P80 is not set
+# CONFIG_MTD_SLRAM is not set
+# CONFIG_MTD_PHRAM is not set
+# CONFIG_MTD_MTDRAM is not set
+# CONFIG_MTD_BLOCK2MTD is not set
+
+#
+# Disk-On-Chip Device Drivers
+#
+# CONFIG_MTD_DOC2000 is not set
+# CONFIG_MTD_DOC2001 is not set
+# CONFIG_MTD_DOC2001PLUS is not set
+# CONFIG_MTD_NAND is not set
+# CONFIG_MTD_ONENAND is not set
+
+#
+# UBI - Unsorted block images
+#
+# CONFIG_MTD_UBI is not set
+# CONFIG_PARPORT is not set
+CONFIG_BLK_DEV=y
+# CONFIG_BLK_DEV_COW_COMMON is not set
+# CONFIG_BLK_DEV_LOOP is not set
+# CONFIG_BLK_DEV_NBD is not set
+# CONFIG_BLK_DEV_UB is not set
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_COUNT=16
+CONFIG_BLK_DEV_RAM_SIZE=4096
+CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024
+# CONFIG_CDROM_PKTCDVD is not set
+# CONFIG_ATA_OVER_ETH is not set
+# CONFIG_MISC_DEVICES is not set
+# CONFIG_IDE is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+# CONFIG_SCSI is not set
+# CONFIG_SCSI_DMA is not set
+# CONFIG_SCSI_NETLINK is not set
+# CONFIG_ATA is not set
+# CONFIG_MD is not set
+CONFIG_NETDEVICES=y
+# CONFIG_NETDEVICES_MULTIQUEUE is not set
+# CONFIG_DUMMY is not set
+# CONFIG_BONDING is not set
+# CONFIG_MACVLAN is not set
+# CONFIG_EQUALIZER is not set
+# CONFIG_TUN is not set
+# CONFIG_VETH is not set
+CONFIG_PHYLIB=y
+
+#
+# MII PHY device drivers
+#
+# CONFIG_MARVELL_PHY is not set
+# CONFIG_DAVICOM_PHY is not set
+# CONFIG_QSEMI_PHY is not set
+# CONFIG_LXT_PHY is not set
+# CONFIG_CICADA_PHY is not set
+# CONFIG_VITESSE_PHY is not set
+# CONFIG_SMSC_PHY is not set
+# CONFIG_BROADCOM_PHY is not set
+# CONFIG_ICPLUS_PHY is not set
+# CONFIG_FIXED_PHY is not set
+# CONFIG_MDIO_BITBANG is not set
+CONFIG_NET_ETHERNET=y
+CONFIG_MII=y
+CONFIG_BFIN_MAC=y
+CONFIG_BFIN_MAC_USE_L1=y
+CONFIG_BFIN_TX_DESC_NUM=10
+CONFIG_BFIN_RX_DESC_NUM=20
+CONFIG_BFIN_MAC_RMII=y
+# CONFIG_SMC91X is not set
+# CONFIG_SMSC911X is not set
+# CONFIG_DM9000 is not set
+# CONFIG_IBM_NEW_EMAC_ZMII is not set
+# CONFIG_IBM_NEW_EMAC_RGMII is not set
+# CONFIG_IBM_NEW_EMAC_TAH is not set
+# CONFIG_IBM_NEW_EMAC_EMAC4 is not set
+# CONFIG_B44 is not set
+# CONFIG_NETDEV_1000 is not set
+# CONFIG_NETDEV_10000 is not set
+
+#
+# Wireless LAN
+#
+# CONFIG_WLAN_PRE80211 is not set
+# CONFIG_WLAN_80211 is not set
+
+#
+# USB Network Adapters
+#
+# CONFIG_USB_CATC is not set
+# CONFIG_USB_KAWETH is not set
+# CONFIG_USB_PEGASUS is not set
+# CONFIG_USB_RTL8150 is not set
+# CONFIG_USB_USBNET is not set
+# CONFIG_WAN is not set
+# CONFIG_PPP is not set
+# CONFIG_SLIP is not set
+# CONFIG_SHAPER is not set
+# CONFIG_NETCONSOLE is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+# CONFIG_ISDN is not set
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+# CONFIG_INPUT is not set
+
+#
+# Hardware I/O ports
+#
+# CONFIG_SERIO is not set
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+# CONFIG_AD9960 is not set
+# CONFIG_SPI_ADC_BF533 is not set
+# CONFIG_BF5xx_PPIFCD is not set
+# CONFIG_BFIN_SIMPLE_TIMER is not set
+# CONFIG_BF5xx_PPI is not set
+CONFIG_BFIN_OTP=y
+# CONFIG_BFIN_OTP_WRITE_ENABLE is not set
+# CONFIG_BFIN_SPORT is not set
+# CONFIG_BFIN_TIMER_LATENCY is not set
+# CONFIG_TWI_LCD is not set
+CONFIG_SIMPLE_GPIO=m
+# CONFIG_VT is not set
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_BFIN=y
+CONFIG_SERIAL_BFIN_CONSOLE=y
+CONFIG_SERIAL_BFIN_DMA=y
+# CONFIG_SERIAL_BFIN_PIO is not set
+CONFIG_SERIAL_BFIN_UART0=y
+# CONFIG_BFIN_UART0_CTSRTS is not set
+CONFIG_SERIAL_BFIN_UART1=y
+# CONFIG_BFIN_UART1_CTSRTS is not set
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_BFIN_SPORT is not set
+CONFIG_UNIX98_PTYS=y
+# CONFIG_LEGACY_PTYS is not set
+
+#
+# CAN, the car bus and industrial fieldbus
+#
+# CONFIG_CAN4LINUX is not set
+# CONFIG_IPMI_HANDLER is not set
+CONFIG_HW_RANDOM=y
+# CONFIG_GEN_RTC is not set
+# CONFIG_R3964 is not set
+# CONFIG_RAW_DRIVER is not set
+# CONFIG_TCG_TPM is not set
+CONFIG_I2C=y
+CONFIG_I2C_BOARDINFO=y
+CONFIG_I2C_CHARDEV=m
+
+#
+# I2C Algorithms
+#
+# CONFIG_I2C_ALGOBIT is not set
+# CONFIG_I2C_ALGOPCF is not set
+# CONFIG_I2C_ALGOPCA is not set
+
+#
+# I2C Hardware Bus support
+#
+CONFIG_I2C_BLACKFIN_TWI=m
+CONFIG_I2C_BLACKFIN_TWI_CLK_KHZ=50
+# CONFIG_I2C_GPIO is not set
+# CONFIG_I2C_OCORES is not set
+# CONFIG_I2C_PARPORT_LIGHT is not set
+# CONFIG_I2C_SIMTEC is not set
+# CONFIG_I2C_TAOS_EVM is not set
+# CONFIG_I2C_STUB is not set
+# CONFIG_I2C_TINY_USB is not set
+
+#
+# Miscellaneous I2C Chip support
+#
+# CONFIG_SENSORS_DS1337 is not set
+# CONFIG_SENSORS_DS1374 is not set
+# CONFIG_DS1682 is not set
+# CONFIG_SENSORS_AD5252 is not set
+# CONFIG_SENSORS_EEPROM is not set
+# CONFIG_SENSORS_PCF8574 is not set
+# CONFIG_SENSORS_PCF8575 is not set
+# CONFIG_SENSORS_PCA9539 is not set
+# CONFIG_SENSORS_PCF8591 is not set
+# CONFIG_SENSORS_MAX6875 is not set
+# CONFIG_SENSORS_TSL2550 is not set
+# CONFIG_I2C_DEBUG_CORE is not set
+# CONFIG_I2C_DEBUG_ALGO is not set
+# CONFIG_I2C_DEBUG_BUS is not set
+# CONFIG_I2C_DEBUG_CHIP is not set
+
+#
+# SPI support
+#
+CONFIG_SPI=y
+CONFIG_SPI_MASTER=y
+
+#
+# SPI Master Controller Drivers
+#
+CONFIG_SPI_BFIN=y
+# CONFIG_SPI_BITBANG is not set
+
+#
+# SPI Protocol Masters
+#
+# CONFIG_SPI_AT25 is not set
+# CONFIG_SPI_SPIDEV is not set
+# CONFIG_SPI_TLE62X0 is not set
+# CONFIG_W1 is not set
+# CONFIG_POWER_SUPPLY is not set
+CONFIG_HWMON=y
+# CONFIG_HWMON_VID is not set
+# CONFIG_SENSORS_AD7418 is not set
+# CONFIG_SENSORS_ADM1021 is not set
+# CONFIG_SENSORS_ADM1025 is not set
+# CONFIG_SENSORS_ADM1026 is not set
+# CONFIG_SENSORS_ADM1029 is not set
+# CONFIG_SENSORS_ADM1031 is not set
+# CONFIG_SENSORS_ADM9240 is not set
+# CONFIG_SENSORS_ADT7470 is not set
+# CONFIG_SENSORS_ATXP1 is not set
+# CONFIG_SENSORS_DS1621 is not set
+# CONFIG_SENSORS_F71805F is not set
+# CONFIG_SENSORS_F71882FG is not set
+# CONFIG_SENSORS_F75375S is not set
+# CONFIG_SENSORS_GL518SM is not set
+# CONFIG_SENSORS_GL520SM is not set
+# CONFIG_SENSORS_IT87 is not set
+# CONFIG_SENSORS_LM63 is not set
+# CONFIG_SENSORS_LM70 is not set
+# CONFIG_SENSORS_LM75 is not set
+# CONFIG_SENSORS_LM77 is not set
+# CONFIG_SENSORS_LM78 is not set
+# CONFIG_SENSORS_LM80 is not set
+# CONFIG_SENSORS_LM83 is not set
+# CONFIG_SENSORS_LM85 is not set
+# CONFIG_SENSORS_LM87 is not set
+# CONFIG_SENSORS_LM90 is not set
+# CONFIG_SENSORS_LM92 is not set
+# CONFIG_SENSORS_LM93 is not set
+# CONFIG_SENSORS_MAX1619 is not set
+# CONFIG_SENSORS_MAX6650 is not set
+# CONFIG_SENSORS_PC87360 is not set
+# CONFIG_SENSORS_PC87427 is not set
+# CONFIG_SENSORS_DME1737 is not set
+# CONFIG_SENSORS_SMSC47M1 is not set
+# CONFIG_SENSORS_SMSC47M192 is not set
+# CONFIG_SENSORS_SMSC47B397 is not set
+# CONFIG_SENSORS_THMC50 is not set
+# CONFIG_SENSORS_VT1211 is not set
+# CONFIG_SENSORS_W83781D is not set
+# CONFIG_SENSORS_W83791D is not set
+# CONFIG_SENSORS_W83792D is not set
+# CONFIG_SENSORS_W83793 is not set
+# CONFIG_SENSORS_W83L785TS is not set
+# CONFIG_SENSORS_W83627HF is not set
+# CONFIG_SENSORS_W83627EHF is not set
+# CONFIG_HWMON_DEBUG_CHIP is not set
+CONFIG_WATCHDOG=y
+# CONFIG_WATCHDOG_NOWAYOUT is not set
+
+#
+# Watchdog Device Drivers
+#
+# CONFIG_SOFT_WATCHDOG is not set
+CONFIG_BFIN_WDT=y
+
+#
+# USB-based Watchdog Cards
+#
+# CONFIG_USBPCWATCHDOG is not set
+
+#
+# Sonics Silicon Backplane
+#
+CONFIG_SSB_POSSIBLE=y
+# CONFIG_SSB is not set
+
+#
+# Multifunction device drivers
+#
+# CONFIG_MFD_SM501 is not set
+
+#
+# Multimedia devices
+#
+# CONFIG_VIDEO_DEV is not set
+# CONFIG_DVB_CORE is not set
+# CONFIG_DAB is not set
+
+#
+# Graphics support
+#
+# CONFIG_VGASTATE is not set
+# CONFIG_VIDEO_OUTPUT_CONTROL is not set
+# CONFIG_FB is not set
+# CONFIG_BACKLIGHT_LCD_SUPPORT is not set
+
+#
+# Display device support
+#
+# CONFIG_DISPLAY_SUPPORT is not set
+
+#
+# Sound
+#
+# CONFIG_SOUND is not set
+CONFIG_USB_SUPPORT=y
+CONFIG_USB_ARCH_HAS_HCD=y
+# CONFIG_USB_ARCH_HAS_OHCI is not set
+# CONFIG_USB_ARCH_HAS_EHCI is not set
+CONFIG_USB=y
+# CONFIG_USB_DEBUG is not set
+
+#
+# Miscellaneous USB options
+#
+# CONFIG_USB_DEVICEFS is not set
+CONFIG_USB_DEVICE_CLASS=y
+# CONFIG_USB_DYNAMIC_MINORS is not set
+# CONFIG_USB_OTG is not set
+# CONFIG_USB_OTG_WHITELIST is not set
+CONFIG_USB_OTG_BLACKLIST_HUB=y
+
+#
+# USB Host Controller Drivers
+#
+# CONFIG_USB_ISP116X_HCD is not set
+# CONFIG_USB_ISP1362_HCD is not set
+# CONFIG_USB_ISP1760_HCD is not set
+# CONFIG_USB_SL811_HCD is not set
+# CONFIG_USB_R8A66597_HCD is not set
+CONFIG_USB_MUSB_HDRC=y
+CONFIG_USB_MUSB_SOC=y
+
+#
+# Blackfin high speed USB support
+#
+CONFIG_USB_MUSB_HOST=y
+# CONFIG_USB_MUSB_PERIPHERAL is not set
+# CONFIG_USB_MUSB_OTG is not set
+CONFIG_USB_MUSB_HDRC_HCD=y
+CONFIG_MUSB_PIO_ONLY=y
+CONFIG_USB_MUSB_LOGLEVEL=0
+
+#
+# USB Device Class drivers
+#
+# CONFIG_USB_ACM is not set
+# CONFIG_USB_PRINTER is not set
+
+#
+# NOTE: USB_STORAGE enables SCSI, and 'SCSI disk support'
+#
+
+#
+# may also be needed; see USB_STORAGE Help for more information
+#
+# CONFIG_USB_LIBUSUAL is not set
+
+#
+# USB Imaging devices
+#
+# CONFIG_USB_MDC800 is not set
+CONFIG_USB_MON=y
+
+#
+# USB port drivers
+#
+
+#
+# USB Serial Converter support
+#
+# CONFIG_USB_SERIAL is not set
+
+#
+# USB Miscellaneous drivers
+#
+# CONFIG_USB_EMI62 is not set
+# CONFIG_USB_EMI26 is not set
+# CONFIG_USB_ADUTUX is not set
+# CONFIG_USB_AUERSWALD is not set
+# CONFIG_USB_RIO500 is not set
+# CONFIG_USB_LEGOTOWER is not set
+# CONFIG_USB_LCD is not set
+# CONFIG_USB_BERRY_CHARGE is not set
+# CONFIG_USB_LED is not set
+# CONFIG_USB_CYPRESS_CY7C63 is not set
+# CONFIG_USB_CYTHERM is not set
+# CONFIG_USB_PHIDGET is not set
+# CONFIG_USB_IDMOUSE is not set
+# CONFIG_USB_FTDI_ELAN is not set
+# CONFIG_USB_APPLEDISPLAY is not set
+# CONFIG_USB_SISUSBVGA is not set
+# CONFIG_USB_LD is not set
+# CONFIG_USB_TRANCEVIBRATOR is not set
+# CONFIG_USB_IOWARRIOR is not set
+
+#
+# USB DSL modem support
+#
+
+#
+# USB Gadget Support
+#
+# CONFIG_USB_GADGET is not set
+# CONFIG_MMC is not set
+# CONFIG_NEW_LEDS is not set
+CONFIG_RTC_LIB=y
+CONFIG_RTC_CLASS=y
+CONFIG_RTC_HCTOSYS=y
+CONFIG_RTC_HCTOSYS_DEVICE="rtc0"
+# CONFIG_RTC_DEBUG is not set
+
+#
+# RTC interfaces
+#
+CONFIG_RTC_INTF_SYSFS=y
+CONFIG_RTC_INTF_PROC=y
+CONFIG_RTC_INTF_DEV=y
+# CONFIG_RTC_INTF_DEV_UIE_EMUL is not set
+# CONFIG_RTC_DRV_TEST is not set
+
+#
+# I2C RTC drivers
+#
+# CONFIG_RTC_DRV_DS1307 is not set
+# CONFIG_RTC_DRV_DS1374 is not set
+# CONFIG_RTC_DRV_DS1672 is not set
+# CONFIG_RTC_DRV_MAX6900 is not set
+# CONFIG_RTC_DRV_RS5C372 is not set
+# CONFIG_RTC_DRV_ISL1208 is not set
+# CONFIG_RTC_DRV_X1205 is not set
+# CONFIG_RTC_DRV_PCF8563 is not set
+# CONFIG_RTC_DRV_PCF8583 is not set
+# CONFIG_RTC_DRV_M41T80 is not set
+
+#
+# SPI RTC drivers
+#
+# CONFIG_RTC_DRV_RS5C348 is not set
+# CONFIG_RTC_DRV_MAX6902 is not set
+
+#
+# Platform RTC drivers
+#
+# CONFIG_RTC_DRV_DS1553 is not set
+# CONFIG_RTC_DRV_STK17TA8 is not set
+# CONFIG_RTC_DRV_DS1742 is not set
+# CONFIG_RTC_DRV_M48T86 is not set
+# CONFIG_RTC_DRV_M48T59 is not set
+# CONFIG_RTC_DRV_V3020 is not set
+
+#
+# on-CPU RTC drivers
+#
+CONFIG_RTC_DRV_BFIN=y
+
+#
+# Userspace I/O
+#
+# CONFIG_UIO is not set
+
+#
+# File systems
+#
+# CONFIG_EXT2_FS is not set
+# CONFIG_EXT3_FS is not set
+# CONFIG_EXT4DEV_FS is not set
+# CONFIG_REISERFS_FS is not set
+# CONFIG_JFS_FS is not set
+# CONFIG_FS_POSIX_ACL is not set
+# CONFIG_XFS_FS is not set
+# CONFIG_GFS2_FS is not set
+# CONFIG_OCFS2_FS is not set
+# CONFIG_MINIX_FS is not set
+# CONFIG_ROMFS_FS is not set
+CONFIG_INOTIFY=y
+CONFIG_INOTIFY_USER=y
+# CONFIG_QUOTA is not set
+# CONFIG_DNOTIFY is not set
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+# CONFIG_FUSE_FS is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+# CONFIG_ISO9660_FS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+# CONFIG_MSDOS_FS is not set
+# CONFIG_VFAT_FS is not set
+# CONFIG_NTFS_FS is not set
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+CONFIG_PROC_SYSCTL=y
+CONFIG_SYSFS=y
+# CONFIG_TMPFS is not set
+# CONFIG_HUGETLB_PAGE is not set
+# CONFIG_CONFIGFS_FS is not set
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+# CONFIG_YAFFS_FS is not set
+# CONFIG_JFFS2_FS is not set
+# CONFIG_CRAMFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+CONFIG_NETWORK_FILESYSTEMS=y
+CONFIG_NFS_FS=m
+CONFIG_NFS_V3=y
+# CONFIG_NFS_V3_ACL is not set
+# CONFIG_NFS_V4 is not set
+# CONFIG_NFS_DIRECTIO is not set
+# CONFIG_NFSD is not set
+CONFIG_LOCKD=m
+CONFIG_LOCKD_V4=y
+CONFIG_NFS_COMMON=y
+CONFIG_SUNRPC=m
+# CONFIG_SUNRPC_BIND34 is not set
+# CONFIG_RPCSEC_GSS_KRB5 is not set
+# CONFIG_RPCSEC_GSS_SPKM3 is not set
+CONFIG_SMB_FS=m
+# CONFIG_SMB_NLS_DEFAULT is not set
+# CONFIG_CIFS is not set
+# CONFIG_NCP_FS is not set
+# CONFIG_CODA_FS is not set
+# CONFIG_AFS_FS is not set
+
+#
+# Partition Types
+#
+# CONFIG_PARTITION_ADVANCED is not set
+CONFIG_MSDOS_PARTITION=y
+CONFIG_NLS=m
+CONFIG_NLS_DEFAULT="iso8859-1"
+# CONFIG_NLS_CODEPAGE_437 is not set
+# CONFIG_NLS_CODEPAGE_737 is not set
+# CONFIG_NLS_CODEPAGE_775 is not set
+# CONFIG_NLS_CODEPAGE_850 is not set
+# CONFIG_NLS_CODEPAGE_852 is not set
+# CONFIG_NLS_CODEPAGE_855 is not set
+# CONFIG_NLS_CODEPAGE_857 is not set
+# CONFIG_NLS_CODEPAGE_860 is not set
+# CONFIG_NLS_CODEPAGE_861 is not set
+# CONFIG_NLS_CODEPAGE_862 is not set
+# CONFIG_NLS_CODEPAGE_863 is not set
+# CONFIG_NLS_CODEPAGE_864 is not set
+# CONFIG_NLS_CODEPAGE_865 is not set
+# CONFIG_NLS_CODEPAGE_866 is not set
+# CONFIG_NLS_CODEPAGE_869 is not set
+# CONFIG_NLS_CODEPAGE_936 is not set
+# CONFIG_NLS_CODEPAGE_950 is not set
+# CONFIG_NLS_CODEPAGE_932 is not set
+# CONFIG_NLS_CODEPAGE_949 is not set
+# CONFIG_NLS_CODEPAGE_874 is not set
+# CONFIG_NLS_ISO8859_8 is not set
+# CONFIG_NLS_CODEPAGE_1250 is not set
+# CONFIG_NLS_CODEPAGE_1251 is not set
+# CONFIG_NLS_ASCII is not set
+# CONFIG_NLS_ISO8859_1 is not set
+# CONFIG_NLS_ISO8859_2 is not set
+# CONFIG_NLS_ISO8859_3 is not set
+# CONFIG_NLS_ISO8859_4 is not set
+# CONFIG_NLS_ISO8859_5 is not set
+# CONFIG_NLS_ISO8859_6 is not set
+# CONFIG_NLS_ISO8859_7 is not set
+# CONFIG_NLS_ISO8859_9 is not set
+# CONFIG_NLS_ISO8859_13 is not set
+# CONFIG_NLS_ISO8859_14 is not set
+# CONFIG_NLS_ISO8859_15 is not set
+# CONFIG_NLS_KOI8_R is not set
+# CONFIG_NLS_KOI8_U is not set
+# CONFIG_NLS_UTF8 is not set
+# CONFIG_DLM is not set
+# CONFIG_INSTRUMENTATION is not set
+
+#
+# Kernel hacking
+#
+# CONFIG_PRINTK_TIME is not set
+CONFIG_ENABLE_WARN_DEPRECATED=y
+CONFIG_ENABLE_MUST_CHECK=y
+# CONFIG_MAGIC_SYSRQ is not set
+# CONFIG_UNUSED_SYMBOLS is not set
+CONFIG_DEBUG_FS=y
+# CONFIG_HEADERS_CHECK is not set
+# CONFIG_DEBUG_KERNEL is not set
+# CONFIG_DEBUG_BUGVERBOSE is not set
+# CONFIG_SAMPLES is not set
+CONFIG_DEBUG_MMRS=y
+CONFIG_DEBUG_HUNT_FOR_ZERO=y
+CONFIG_DEBUG_BFIN_HWTRACE_ON=y
+CONFIG_DEBUG_BFIN_HWTRACE_COMPRESSION_OFF=y
+# CONFIG_DEBUG_BFIN_HWTRACE_COMPRESSION_ONE is not set
+# CONFIG_DEBUG_BFIN_HWTRACE_COMPRESSION_TWO is not set
+CONFIG_DEBUG_BFIN_HWTRACE_COMPRESSION=0
+# CONFIG_DEBUG_BFIN_HWTRACE_EXPAND is not set
+# CONFIG_DEBUG_BFIN_NO_KERN_HWTRACE is not set
+CONFIG_EARLY_PRINTK=y
+# CONFIG_CPLB_INFO is not set
+CONFIG_ACCESS_CHECK=y
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+CONFIG_SECURITY=y
+# CONFIG_SECURITY_NETWORK is not set
+# CONFIG_SECURITY_CAPABILITIES is not set
+# CONFIG_SECURITY_ROOTPLUG is not set
+# CONFIG_CRYPTO is not set
+
+#
+# Library routines
+#
+CONFIG_BITREVERSE=y
+CONFIG_CRC_CCITT=m
+# CONFIG_CRC16 is not set
+# CONFIG_CRC_ITU_T is not set
+CONFIG_CRC32=y
+# CONFIG_CRC7 is not set
+# CONFIG_LIBCRC32C is not set
+CONFIG_ZLIB_INFLATE=y
+CONFIG_PLIST=y
+CONFIG_HAS_IOMEM=y
+CONFIG_HAS_IOPORT=y
+CONFIG_HAS_DMA=y
diff --git a/arch/blackfin/mach-bf527/boards/Kconfig b/arch/blackfin/mach-bf527/boards/Kconfig
index 6a570ad03746..8bf9e58f0148 100644
--- a/arch/blackfin/mach-bf527/boards/Kconfig
+++ b/arch/blackfin/mach-bf527/boards/Kconfig
@@ -9,4 +9,9 @@ config BFIN527_EZKIT
 	help
 	  BF527-EZKIT-LITE board support.
 
+config BFIN527_BLUETECHNIX_CM
+	bool "Bluetechnix CM-BF527"
+	help
+	  CM-BF527 support for EVAL- and DEV-Board.
+
 endchoice
diff --git a/arch/blackfin/mach-bf527/boards/Makefile b/arch/blackfin/mach-bf527/boards/Makefile
index 7277d35ef111..7ba7d256bbb8 100644
--- a/arch/blackfin/mach-bf527/boards/Makefile
+++ b/arch/blackfin/mach-bf527/boards/Makefile
@@ -3,3 +3,4 @@
 #
 
 obj-$(CONFIG_BFIN527_EZKIT)            += ezkit.o
+obj-$(CONFIG_BFIN527_BLUETECHNIX_CM)   += cm_bf527.o
diff --git a/arch/blackfin/mach-bf527/boards/cm_bf527.c b/arch/blackfin/mach-bf527/boards/cm_bf527.c
new file mode 100644
index 000000000000..0b26ae2de5ee
--- /dev/null
+++ b/arch/blackfin/mach-bf527/boards/cm_bf527.c
@@ -0,0 +1,1011 @@
+/*
+ * File:         arch/blackfin/mach-bf527/boards/cm-bf527.c
+ * Based on:     arch/blackfin/mach-bf537/boards/stamp.c
+ * Author:       Aidan Williams <aidan@nicta.com.au>
+ *
+ * Created:
+ * Description:
+ *
+ * Modified:
+ *               Copyright 2005 National ICT Australia (NICTA)
+ *               Copyright 2004-2008 Analog Devices Inc.
+ *
+ * Bugs:         Enter bugs at http://blackfin.uclinux.org/
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, see the file COPYING, or write
+ * to the Free Software Foundation, Inc.,
+ * 51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
+ */
+
+#include <linux/device.h>
+#include <linux/platform_device.h>
+#include <linux/mtd/mtd.h>
+#include <linux/mtd/partitions.h>
+#include <linux/mtd/physmap.h>
+#include <linux/spi/spi.h>
+#include <linux/spi/flash.h>
+#include <linux/etherdevice.h>
+#if defined(CONFIG_USB_ISP1362_HCD) || defined(CONFIG_USB_ISP1362_HCD_MODULE)
+#include <linux/usb/isp1362.h>
+#endif
+#include <linux/pata_platform.h>
+#include <linux/i2c.h>
+#include <linux/irq.h>
+#include <linux/interrupt.h>
+#include <linux/usb/sl811.h>
+#if defined(CONFIG_USB_MUSB_HDRC) || defined(CONFIG_USB_MUSB_HDRC_MODULE)
+#include <linux/usb/musb.h>
+#endif
+#include <asm/cplb.h>
+#include <asm/dma.h>
+#include <asm/bfin5xx_spi.h>
+#include <asm/reboot.h>
+#include <asm/nand.h>
+#include <asm/portmux.h>
+#include <asm/dpmc.h>
+#include <linux/spi/ad7877.h>
+
+/*
+ * Name the Board for the /proc/cpuinfo
+ */
+const char bfin_board_name[] = "Bluetechnix CM-BF527";
+
+/*
+ *  Driver needs to know address, irq and flag pin.
+ */
+
+#define ISP1761_BASE       0x203C0000
+#define ISP1761_IRQ        IRQ_PF7
+
+#if defined(CONFIG_USB_ISP1760_HCD) || defined(CONFIG_USB_ISP1760_HCD_MODULE)
+static struct resource bfin_isp1761_resources[] = {
+	[0] = {
+		.name	= "isp1761-regs",
+		.start  = ISP1761_BASE + 0x00000000,
+		.end    = ISP1761_BASE + 0x000fffff,
+		.flags  = IORESOURCE_MEM,
+	},
+	[1] = {
+		.start  = ISP1761_IRQ,
+		.end    = ISP1761_IRQ,
+		.flags  = IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device bfin_isp1761_device = {
+	.name           = "isp1761",
+	.id             = 0,
+	.num_resources  = ARRAY_SIZE(bfin_isp1761_resources),
+	.resource       = bfin_isp1761_resources,
+};
+
+static struct platform_device *bfin_isp1761_devices[] = {
+	&bfin_isp1761_device,
+};
+
+int __init bfin_isp1761_init(void)
+{
+	unsigned int num_devices = ARRAY_SIZE(bfin_isp1761_devices);
+
+	printk(KERN_INFO "%s(): registering device resources\n", __func__);
+	set_irq_type(ISP1761_IRQ, IRQF_TRIGGER_FALLING);
+
+	return platform_add_devices(bfin_isp1761_devices, num_devices);
+}
+
+void __exit bfin_isp1761_exit(void)
+{
+	platform_device_unregister(&bfin_isp1761_device);
+}
+
+arch_initcall(bfin_isp1761_init);
+#endif
+
+#if defined(CONFIG_USB_MUSB_HDRC) || defined(CONFIG_USB_MUSB_HDRC_MODULE)
+static struct resource musb_resources[] = {
+	[0] = {
+		.start	= 0xffc03800,
+		.end	= 0xffc03cff,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {	/* general IRQ */
+		.start	= IRQ_USB_INT0,
+		.end	= IRQ_USB_INT0,
+		.flags	= IORESOURCE_IRQ | IORESOURCE_IRQ_HIGHLEVEL,
+	},
+	[2] = {	/* DMA IRQ */
+		.start	= IRQ_USB_DMA,
+		.end	= IRQ_USB_DMA,
+		.flags	= IORESOURCE_IRQ | IORESOURCE_IRQ_HIGHLEVEL,
+	},
+};
+
+static struct musb_hdrc_platform_data musb_plat = {
+#if defined(CONFIG_USB_MUSB_OTG)
+	.mode		= MUSB_OTG,
+#elif defined(CONFIG_USB_MUSB_HDRC_HCD)
+	.mode		= MUSB_HOST,
+#elif defined(CONFIG_USB_GADGET_MUSB_HDRC)
+	.mode		= MUSB_PERIPHERAL,
+#endif
+	.multipoint	= 0,
+};
+
+static u64 musb_dmamask = ~(u32)0;
+
+static struct platform_device musb_device = {
+	.name		= "musb_hdrc",
+	.id		= 0,
+	.dev = {
+		.dma_mask		= &musb_dmamask,
+		.coherent_dma_mask	= 0xffffffff,
+		.platform_data		= &musb_plat,
+	},
+	.num_resources	= ARRAY_SIZE(musb_resources),
+	.resource	= musb_resources,
+};
+#endif
+
+#if defined(CONFIG_MTD_PHYSMAP) || defined(CONFIG_MTD_PHYSMAP_MODULE)
+static struct mtd_partition ezkit_partitions[] = {
+	{
+		.name       = "Bootloader",
+		.size       = 0x40000,
+		.offset     = 0,
+	}, {
+		.name       = "Kernel",
+		.size       = 0x1C0000,
+		.offset     = MTDPART_OFS_APPEND,
+	}, {
+		.name       = "RootFS",
+		.size       = MTDPART_SIZ_FULL,
+		.offset     = MTDPART_OFS_APPEND,
+	}
+};
+
+static struct physmap_flash_data ezkit_flash_data = {
+	.width      = 2,
+	.parts      = ezkit_partitions,
+	.nr_parts   = ARRAY_SIZE(ezkit_partitions),
+};
+
+static struct resource ezkit_flash_resource = {
+	.start = 0x20000000,
+	.end   = 0x201fffff,
+	.flags = IORESOURCE_MEM,
+};
+
+static struct platform_device ezkit_flash_device = {
+	.name          = "physmap-flash",
+	.id            = 0,
+	.dev = {
+		.platform_data = &ezkit_flash_data,
+	},
+	.num_resources = 1,
+	.resource      = &ezkit_flash_resource,
+};
+#endif
+
+#if defined(CONFIG_MTD_NAND_BF5XX) || defined(CONFIG_MTD_NAND_BF5XX_MODULE)
+static struct mtd_partition partition_info[] = {
+	{
+		.name = "Linux Kernel",
+		.offset = 0,
+		.size = 4 * SIZE_1M,
+	},
+	{
+		.name = "File System",
+		.offset = MTDPART_OFS_APPEND,
+		.size = MTDPART_SIZ_FULL,
+	},
+};
+
+static struct bf5xx_nand_platform bf5xx_nand_platform = {
+	.page_size = NFC_PG_SIZE_256,
+	.data_width = NFC_NWIDTH_8,
+	.partitions = partition_info,
+	.nr_partitions = ARRAY_SIZE(partition_info),
+	.rd_dly = 3,
+	.wr_dly = 3,
+};
+
+static struct resource bf5xx_nand_resources[] = {
+	{
+		.start = NFC_CTL,
+		.end = NFC_DATA_RD + 2,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = CH_NFC,
+		.end = CH_NFC,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device bf5xx_nand_device = {
+	.name = "bf5xx-nand",
+	.id = 0,
+	.num_resources = ARRAY_SIZE(bf5xx_nand_resources),
+	.resource = bf5xx_nand_resources,
+	.dev = {
+		.platform_data = &bf5xx_nand_platform,
+	},
+};
+#endif
+
+#if defined(CONFIG_BFIN_CFPCMCIA) || defined(CONFIG_BFIN_CFPCMCIA_MODULE)
+static struct resource bfin_pcmcia_cf_resources[] = {
+	{
+		.start = 0x20310000, /* IO PORT */
+		.end = 0x20312000,
+		.flags = IORESOURCE_MEM,
+	}, {
+		.start = 0x20311000, /* Attribute Memory */
+		.end = 0x20311FFF,
+		.flags = IORESOURCE_MEM,
+	}, {
+		.start = IRQ_PF4,
+		.end = IRQ_PF4,
+		.flags = IORESOURCE_IRQ | IORESOURCE_IRQ_LOWLEVEL,
+	}, {
+		.start = 6, /* Card Detect PF6 */
+		.end = 6,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device bfin_pcmcia_cf_device = {
+	.name = "bfin_cf_pcmcia",
+	.id = -1,
+	.num_resources = ARRAY_SIZE(bfin_pcmcia_cf_resources),
+	.resource = bfin_pcmcia_cf_resources,
+};
+#endif
+
+#if defined(CONFIG_RTC_DRV_BFIN) || defined(CONFIG_RTC_DRV_BFIN_MODULE)
+static struct platform_device rtc_device = {
+	.name = "rtc-bfin",
+	.id   = -1,
+};
+#endif
+
+#if defined(CONFIG_SMC91X) || defined(CONFIG_SMC91X_MODULE)
+static struct resource smc91x_resources[] = {
+	{
+		.name = "smc91x-regs",
+		.start = 0x20300300,
+		.end = 0x20300300 + 16,
+		.flags = IORESOURCE_MEM,
+	}, {
+
+		.start = IRQ_PF7,
+		.end = IRQ_PF7,
+		.flags = IORESOURCE_IRQ | IORESOURCE_IRQ_HIGHLEVEL,
+	},
+};
+static struct platform_device smc91x_device = {
+	.name = "smc91x",
+	.id = 0,
+	.num_resources = ARRAY_SIZE(smc91x_resources),
+	.resource = smc91x_resources,
+};
+#endif
+
+#if defined(CONFIG_DM9000) || defined(CONFIG_DM9000_MODULE)
+static struct resource dm9000_resources[] = {
+	[0] = {
+		.start	= 0x203FB800,
+		.end	= 0x203FB800 + 8,
+		.flags	= IORESOURCE_MEM,
+	},
+	[1] = {
+		.start	= IRQ_PF9,
+		.end	= IRQ_PF9,
+		.flags	= (IORESOURCE_IRQ | IORESOURCE_IRQ_HIGHEDGE),
+	},
+};
+
+static struct platform_device dm9000_device = {
+	.name		= "dm9000",
+	.id		= -1,
+	.num_resources	= ARRAY_SIZE(dm9000_resources),
+	.resource	= dm9000_resources,
+};
+#endif
+
+#if defined(CONFIG_USB_SL811_HCD) || defined(CONFIG_USB_SL811_HCD_MODULE)
+static struct resource sl811_hcd_resources[] = {
+	{
+		.start = 0x20340000,
+		.end = 0x20340000,
+		.flags = IORESOURCE_MEM,
+	}, {
+		.start = 0x20340004,
+		.end = 0x20340004,
+		.flags = IORESOURCE_MEM,
+	}, {
+		.start = CONFIG_USB_SL811_BFIN_IRQ,
+		.end = CONFIG_USB_SL811_BFIN_IRQ,
+		.flags = IORESOURCE_IRQ | IORESOURCE_IRQ_HIGHLEVEL,
+	},
+};
+
+#if defined(CONFIG_USB_SL811_BFIN_USE_VBUS)
+void sl811_port_power(struct device *dev, int is_on)
+{
+	gpio_request(CONFIG_USB_SL811_BFIN_GPIO_VBUS, "usb:SL811_VBUS");
+	gpio_direction_output(CONFIG_USB_SL811_BFIN_GPIO_VBUS, is_on);
+}
+#endif
+
+static struct sl811_platform_data sl811_priv = {
+	.potpg = 10,
+	.power = 250,       /* == 500mA */
+#if defined(CONFIG_USB_SL811_BFIN_USE_VBUS)
+	.port_power = &sl811_port_power,
+#endif
+};
+
+static struct platform_device sl811_hcd_device = {
+	.name = "sl811-hcd",
+	.id = 0,
+	.dev = {
+		.platform_data = &sl811_priv,
+	},
+	.num_resources = ARRAY_SIZE(sl811_hcd_resources),
+	.resource = sl811_hcd_resources,
+};
+#endif
+
+#if defined(CONFIG_USB_ISP1362_HCD) || defined(CONFIG_USB_ISP1362_HCD_MODULE)
+static struct resource isp1362_hcd_resources[] = {
+	{
+		.start = 0x20360000,
+		.end = 0x20360000,
+		.flags = IORESOURCE_MEM,
+	}, {
+		.start = 0x20360004,
+		.end = 0x20360004,
+		.flags = IORESOURCE_MEM,
+	}, {
+		.start = CONFIG_USB_ISP1362_BFIN_GPIO_IRQ,
+		.end = CONFIG_USB_ISP1362_BFIN_GPIO_IRQ,
+		.flags = IORESOURCE_IRQ | IORESOURCE_IRQ_HIGHLEVEL,
+	},
+};
+
+static struct isp1362_platform_data isp1362_priv = {
+	.sel15Kres = 1,
+	.clknotstop = 0,
+	.oc_enable = 0,
+	.int_act_high = 0,
+	.int_edge_triggered = 0,
+	.remote_wakeup_connected = 0,
+	.no_power_switching = 1,
+	.power_switching_mode = 0,
+};
+
+static struct platform_device isp1362_hcd_device = {
+	.name = "isp1362-hcd",
+	.id = 0,
+	.dev = {
+		.platform_data = &isp1362_priv,
+	},
+	.num_resources = ARRAY_SIZE(isp1362_hcd_resources),
+	.resource = isp1362_hcd_resources,
+};
+#endif
+
+#if defined(CONFIG_BFIN_MAC) || defined(CONFIG_BFIN_MAC_MODULE)
+static struct platform_device bfin_mac_device = {
+	.name = "bfin_mac",
+};
+#endif
+
+#if defined(CONFIG_USB_NET2272) || defined(CONFIG_USB_NET2272_MODULE)
+static struct resource net2272_bfin_resources[] = {
+	{
+		.start = 0x20300000,
+		.end = 0x20300000 + 0x100,
+		.flags = IORESOURCE_MEM,
+	}, {
+		.start = IRQ_PF7,
+		.end = IRQ_PF7,
+		.flags = IORESOURCE_IRQ | IORESOURCE_IRQ_HIGHLEVEL,
+	},
+};
+
+static struct platform_device net2272_bfin_device = {
+	.name = "net2272",
+	.id = -1,
+	.num_resources = ARRAY_SIZE(net2272_bfin_resources),
+	.resource = net2272_bfin_resources,
+};
+#endif
+
+#if defined(CONFIG_MTD_M25P80) \
+	|| defined(CONFIG_MTD_M25P80_MODULE)
+static struct mtd_partition bfin_spi_flash_partitions[] = {
+	{
+		.name = "bootloader",
+		.size = 0x00040000,
+		.offset = 0,
+		.mask_flags = MTD_CAP_ROM
+	}, {
+		.name = "linux kernel",
+		.size = MTDPART_SIZ_FULL,
+		.offset = MTDPART_OFS_APPEND,
+	}
+};
+
+static struct flash_platform_data bfin_spi_flash_data = {
+	.name = "m25p80",
+	.parts = bfin_spi_flash_partitions,
+	.nr_parts = ARRAY_SIZE(bfin_spi_flash_partitions),
+	.type = "m25p16",
+};
+
+/* SPI flash chip (m25p64) */
+static struct bfin5xx_spi_chip spi_flash_chip_info = {
+	.enable_dma = 0,         /* use dma transfer with this chip*/
+	.bits_per_word = 8,
+};
+#endif
+
+#if defined(CONFIG_SPI_ADC_BF533) \
+	|| defined(CONFIG_SPI_ADC_BF533_MODULE)
+/* SPI ADC chip */
+static struct bfin5xx_spi_chip spi_adc_chip_info = {
+	.enable_dma = 1,         /* use dma transfer with this chip*/
+	.bits_per_word = 16,
+};
+#endif
+
+#if defined(CONFIG_SND_BLACKFIN_AD1836) \
+	|| defined(CONFIG_SND_BLACKFIN_AD1836_MODULE)
+static struct bfin5xx_spi_chip ad1836_spi_chip_info = {
+	.enable_dma = 0,
+	.bits_per_word = 16,
+};
+#endif
+
+#if defined(CONFIG_AD9960) || defined(CONFIG_AD9960_MODULE)
+static struct bfin5xx_spi_chip ad9960_spi_chip_info = {
+	.enable_dma = 0,
+	.bits_per_word = 16,
+};
+#endif
+
+#if defined(CONFIG_SPI_MMC) || defined(CONFIG_SPI_MMC_MODULE)
+static struct bfin5xx_spi_chip spi_mmc_chip_info = {
+	.enable_dma = 1,
+	.bits_per_word = 8,
+};
+#endif
+
+#if defined(CONFIG_PBX)
+static struct bfin5xx_spi_chip spi_si3xxx_chip_info = {
+	.ctl_reg	= 0x4, /* send zero */
+	.enable_dma	= 0,
+	.bits_per_word	= 8,
+	.cs_change_per_word = 1,
+};
+#endif
+
+#if defined(CONFIG_TOUCHSCREEN_AD7877) || defined(CONFIG_TOUCHSCREEN_AD7877_MODULE)
+static struct bfin5xx_spi_chip spi_ad7877_chip_info = {
+	.enable_dma = 0,
+	.bits_per_word = 16,
+};
+
+static const struct ad7877_platform_data bfin_ad7877_ts_info = {
+	.model			= 7877,
+	.vref_delay_usecs	= 50,	/* internal, no capacitor */
+	.x_plate_ohms		= 419,
+	.y_plate_ohms		= 486,
+	.pressure_max		= 1000,
+	.pressure_min		= 0,
+	.stopacq_polarity 	= 1,
+	.first_conversion_delay = 3,
+	.acquisition_time 	= 1,
+	.averaging 		= 1,
+	.pen_down_acc_interval 	= 1,
+};
+#endif
+
+#if defined(CONFIG_SND_SOC_WM8731) || defined(CONFIG_SND_SOC_WM8731_MODULE) \
+	 && defined(CONFIG_SND_SOC_WM8731_SPI)
+static struct bfin5xx_spi_chip spi_wm8731_chip_info = {
+	.enable_dma = 0,
+	.bits_per_word = 16,
+};
+#endif
+
+#if defined(CONFIG_SPI_SPIDEV) || defined(CONFIG_SPI_SPIDEV_MODULE)
+static struct bfin5xx_spi_chip spidev_chip_info = {
+	.enable_dma = 0,
+	.bits_per_word = 8,
+};
+#endif
+
+static struct spi_board_info bfin_spi_board_info[] __initdata = {
+#if defined(CONFIG_MTD_M25P80) \
+	|| defined(CONFIG_MTD_M25P80_MODULE)
+	{
+		/* the modalias must be the same as spi device driver name */
+		.modalias = "m25p80", /* Name of spi_driver for this device */
+		.max_speed_hz = 25000000,     /* max spi clock (SCK) speed in HZ */
+		.bus_num = 0, /* Framework bus number */
+		.chip_select = 1, /* Framework chip select. On STAMP537 it is SPISSEL1*/
+		.platform_data = &bfin_spi_flash_data,
+		.controller_data = &spi_flash_chip_info,
+		.mode = SPI_MODE_3,
+	},
+#endif
+
+#if defined(CONFIG_SPI_ADC_BF533) \
+	|| defined(CONFIG_SPI_ADC_BF533_MODULE)
+	{
+		.modalias = "bfin_spi_adc", /* Name of spi_driver for this device */
+		.max_speed_hz = 6250000,     /* max spi clock (SCK) speed in HZ */
+		.bus_num = 0, /* Framework bus number */
+		.chip_select = 1, /* Framework chip select. */
+		.platform_data = NULL, /* No spi_driver specific config */
+		.controller_data = &spi_adc_chip_info,
+	},
+#endif
+
+#if defined(CONFIG_SND_BLACKFIN_AD1836) \
+	|| defined(CONFIG_SND_BLACKFIN_AD1836_MODULE)
+	{
+		.modalias = "ad1836-spi",
+		.max_speed_hz = 3125000,     /* max spi clock (SCK) speed in HZ */
+		.bus_num = 0,
+		.chip_select = CONFIG_SND_BLACKFIN_SPI_PFBIT,
+		.controller_data = &ad1836_spi_chip_info,
+	},
+#endif
+#if defined(CONFIG_AD9960) || defined(CONFIG_AD9960_MODULE)
+	{
+		.modalias = "ad9960-spi",
+		.max_speed_hz = 10000000,     /* max spi clock (SCK) speed in HZ */
+		.bus_num = 0,
+		.chip_select = 1,
+		.controller_data = &ad9960_spi_chip_info,
+	},
+#endif
+#if defined(CONFIG_SPI_MMC) || defined(CONFIG_SPI_MMC_MODULE)
+	{
+		.modalias = "spi_mmc_dummy",
+		.max_speed_hz = 25000000,     /* max spi clock (SCK) speed in HZ */
+		.bus_num = 0,
+		.chip_select = 0,
+		.platform_data = NULL,
+		.controller_data = &spi_mmc_chip_info,
+		.mode = SPI_MODE_3,
+	},
+	{
+		.modalias = "spi_mmc",
+		.max_speed_hz = 25000000,     /* max spi clock (SCK) speed in HZ */
+		.bus_num = 0,
+		.chip_select = CONFIG_SPI_MMC_CS_CHAN,
+		.platform_data = NULL,
+		.controller_data = &spi_mmc_chip_info,
+		.mode = SPI_MODE_3,
+	},
+#endif
+#if defined(CONFIG_PBX)
+	{
+		.modalias = "fxs-spi",
+		.max_speed_hz = 12500000,     /* max spi clock (SCK) speed in HZ */
+		.bus_num = 0,
+		.chip_select = 8 - CONFIG_J11_JUMPER,
+		.controller_data = &spi_si3xxx_chip_info,
+		.mode = SPI_MODE_3,
+	},
+	{
+		.modalias = "fxo-spi",
+		.max_speed_hz = 12500000,     /* max spi clock (SCK) speed in HZ */
+		.bus_num = 0,
+		.chip_select = 8 - CONFIG_J19_JUMPER,
+		.controller_data = &spi_si3xxx_chip_info,
+		.mode = SPI_MODE_3,
+	},
+#endif
+#if defined(CONFIG_TOUCHSCREEN_AD7877) || defined(CONFIG_TOUCHSCREEN_AD7877_MODULE)
+	{
+		.modalias		= "ad7877",
+		.platform_data		= &bfin_ad7877_ts_info,
+		.irq			= IRQ_PF8,
+		.max_speed_hz	= 12500000,     /* max spi clock (SCK) speed in HZ */
+		.bus_num	= 0,
+		.chip_select  = 2,
+		.controller_data = &spi_ad7877_chip_info,
+	},
+#endif
+#if defined(CONFIG_SND_SOC_WM8731) || defined(CONFIG_SND_SOC_WM8731_MODULE) \
+	 && defined(CONFIG_SND_SOC_WM8731_SPI)
+	{
+		.modalias	= "wm8731",
+		.max_speed_hz	= 3125000,     /* max spi clock (SCK) speed in HZ */
+		.bus_num	= 0,
+		.chip_select    = 5,
+		.controller_data = &spi_wm8731_chip_info,
+		.mode = SPI_MODE_0,
+	},
+#endif
+#if defined(CONFIG_SPI_SPIDEV) || defined(CONFIG_SPI_SPIDEV_MODULE)
+	{
+		.modalias = "spidev",
+		.max_speed_hz = 3125000,     /* max spi clock (SCK) speed in HZ */
+		.bus_num = 0,
+		.chip_select = 1,
+		.controller_data = &spidev_chip_info,
+	},
+#endif
+};
+
+#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE)
+/* SPI controller data */
+static struct bfin5xx_spi_master bfin_spi0_info = {
+	.num_chipselect = 8,
+	.enable_dma = 1,  /* master has the ability to do dma transfer */
+	.pin_req = {P_SPI0_SCK, P_SPI0_MISO, P_SPI0_MOSI, 0},
+};
+
+/* SPI (0) */
+static struct resource bfin_spi0_resource[] = {
+	[0] = {
+		.start = SPI0_REGBASE,
+		.end   = SPI0_REGBASE + 0xFF,
+		.flags = IORESOURCE_MEM,
+		},
+	[1] = {
+		.start = CH_SPI,
+		.end   = CH_SPI,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device bfin_spi0_device = {
+	.name = "bfin-spi",
+	.id = 0, /* Bus number */
+	.num_resources = ARRAY_SIZE(bfin_spi0_resource),
+	.resource = bfin_spi0_resource,
+	.dev = {
+		.platform_data = &bfin_spi0_info, /* Passed to driver */
+	},
+};
+#endif  /* spi master and devices */
+
+#if defined(CONFIG_FB_BFIN_7393) || defined(CONFIG_FB_BFIN_7393_MODULE)
+static struct platform_device bfin_fb_adv7393_device = {
+	.name = "bfin-adv7393",
+};
+#endif
+
+#if defined(CONFIG_SERIAL_BFIN) || defined(CONFIG_SERIAL_BFIN_MODULE)
+static struct resource bfin_uart_resources[] = {
+#ifdef CONFIG_SERIAL_BFIN_UART0
+	{
+		.start = 0xFFC00400,
+		.end = 0xFFC004FF,
+		.flags = IORESOURCE_MEM,
+	},
+#endif
+#ifdef CONFIG_SERIAL_BFIN_UART1
+	{
+		.start = 0xFFC02000,
+		.end = 0xFFC020FF,
+		.flags = IORESOURCE_MEM,
+	},
+#endif
+};
+
+static struct platform_device bfin_uart_device = {
+	.name = "bfin-uart",
+	.id = 1,
+	.num_resources = ARRAY_SIZE(bfin_uart_resources),
+	.resource = bfin_uart_resources,
+};
+#endif
+
+#if defined(CONFIG_BFIN_SIR) || defined(CONFIG_BFIN_SIR_MODULE)
+static struct resource bfin_sir_resources[] = {
+#ifdef CONFIG_BFIN_SIR0
+	{
+		.start = 0xFFC00400,
+		.end = 0xFFC004FF,
+		.flags = IORESOURCE_MEM,
+	},
+#endif
+#ifdef CONFIG_BFIN_SIR1
+	{
+		.start = 0xFFC02000,
+		.end = 0xFFC020FF,
+		.flags = IORESOURCE_MEM,
+	},
+#endif
+};
+
+static struct platform_device bfin_sir_device = {
+	.name = "bfin_sir",
+	.id = 0,
+	.num_resources = ARRAY_SIZE(bfin_sir_resources),
+	.resource = bfin_sir_resources,
+};
+#endif
+
+#if defined(CONFIG_I2C_BLACKFIN_TWI) || defined(CONFIG_I2C_BLACKFIN_TWI_MODULE)
+static struct resource bfin_twi0_resource[] = {
+	[0] = {
+		.start = TWI0_REGBASE,
+		.end   = TWI0_REGBASE,
+		.flags = IORESOURCE_MEM,
+	},
+	[1] = {
+		.start = IRQ_TWI,
+		.end   = IRQ_TWI,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device i2c_bfin_twi_device = {
+	.name = "i2c-bfin-twi",
+	.id = 0,
+	.num_resources = ARRAY_SIZE(bfin_twi0_resource),
+	.resource = bfin_twi0_resource,
+};
+#endif
+
+#ifdef CONFIG_I2C_BOARDINFO
+static struct i2c_board_info __initdata bfin_i2c_board_info[] = {
+#if defined(CONFIG_TWI_LCD) || defined(CONFIG_TWI_LCD_MODULE)
+	{
+		I2C_BOARD_INFO("pcf8574_lcd", 0x22),
+		.type = "pcf8574_lcd",
+	},
+#endif
+#if defined(CONFIG_TWI_KEYPAD) || defined(CONFIG_TWI_KEYPAD_MODULE)
+	{
+		I2C_BOARD_INFO("pcf8574_keypad", 0x27),
+		.type = "pcf8574_keypad",
+		.irq = IRQ_PF8,
+	},
+#endif
+};
+#endif
+
+#if defined(CONFIG_SERIAL_BFIN_SPORT) || defined(CONFIG_SERIAL_BFIN_SPORT_MODULE)
+static struct platform_device bfin_sport0_uart_device = {
+	.name = "bfin-sport-uart",
+	.id = 0,
+};
+
+static struct platform_device bfin_sport1_uart_device = {
+	.name = "bfin-sport-uart",
+	.id = 1,
+};
+#endif
+
+#if defined(CONFIG_PATA_PLATFORM) || defined(CONFIG_PATA_PLATFORM_MODULE)
+#define PATA_INT	55
+
+static struct pata_platform_info bfin_pata_platform_data = {
+	.ioport_shift = 1,
+	.irq_type = IRQF_TRIGGER_HIGH | IRQF_DISABLED,
+};
+
+static struct resource bfin_pata_resources[] = {
+	{
+		.start = 0x20314020,
+		.end = 0x2031403F,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = 0x2031401C,
+		.end = 0x2031401F,
+		.flags = IORESOURCE_MEM,
+	},
+	{
+		.start = PATA_INT,
+		.end = PATA_INT,
+		.flags = IORESOURCE_IRQ,
+	},
+};
+
+static struct platform_device bfin_pata_device = {
+	.name = "pata_platform",
+	.id = -1,
+	.num_resources = ARRAY_SIZE(bfin_pata_resources),
+	.resource = bfin_pata_resources,
+	.dev = {
+		.platform_data = &bfin_pata_platform_data,
+	}
+};
+#endif
+
+#if defined(CONFIG_KEYBOARD_GPIO) || defined(CONFIG_KEYBOARD_GPIO_MODULE)
+#include <linux/input.h>
+#include <linux/gpio_keys.h>
+
+static struct gpio_keys_button bfin_gpio_keys_table[] = {
+	{BTN_0, GPIO_PF14, 1, "gpio-keys: BTN0"},
+};
+
+static struct gpio_keys_platform_data bfin_gpio_keys_data = {
+	.buttons        = bfin_gpio_keys_table,
+	.nbuttons       = ARRAY_SIZE(bfin_gpio_keys_table),
+};
+
+static struct platform_device bfin_device_gpiokeys = {
+	.name      = "gpio-keys",
+	.dev = {
+		.platform_data = &bfin_gpio_keys_data,
+	},
+};
+#endif
+
+static struct resource bfin_gpios_resources = {
+	.start = 0,
+	.end   = MAX_BLACKFIN_GPIOS - 1,
+	.flags = IORESOURCE_IRQ,
+};
+
+static struct platform_device bfin_gpios_device = {
+	.name = "simple-gpio",
+	.id = -1,
+	.num_resources = 1,
+	.resource = &bfin_gpios_resources,
+};
+
+static const unsigned int cclk_vlev_datasheet[] =
+{
+	VRPAIR(VLEV_100, 400000000),
+	VRPAIR(VLEV_105, 426000000),
+	VRPAIR(VLEV_110, 500000000),
+	VRPAIR(VLEV_115, 533000000),
+	VRPAIR(VLEV_120, 600000000),
+};
+
+static struct bfin_dpmc_platform_data bfin_dmpc_vreg_data = {
+	.tuple_tab = cclk_vlev_datasheet,
+	.tabsize = ARRAY_SIZE(cclk_vlev_datasheet),
+	.vr_settling_time = 25 /* us */,
+};
+
+static struct platform_device bfin_dpmc = {
+	.name = "bfin dpmc",
+	.dev = {
+		.platform_data = &bfin_dmpc_vreg_data,
+	},
+};
+
+static struct platform_device *stamp_devices[] __initdata = {
+
+	&bfin_dpmc,
+
+#if defined(CONFIG_MTD_NAND_BF5XX) || defined(CONFIG_MTD_NAND_BF5XX_MODULE)
+	&bf5xx_nand_device,
+#endif
+
+#if defined(CONFIG_BFIN_CFPCMCIA) || defined(CONFIG_BFIN_CFPCMCIA_MODULE)
+	&bfin_pcmcia_cf_device,
+#endif
+
+#if defined(CONFIG_RTC_DRV_BFIN) || defined(CONFIG_RTC_DRV_BFIN_MODULE)
+	&rtc_device,
+#endif
+
+#if defined(CONFIG_USB_SL811_HCD) || defined(CONFIG_USB_SL811_HCD_MODULE)
+	&sl811_hcd_device,
+#endif
+
+#if defined(CONFIG_USB_ISP1362_HCD) || defined(CONFIG_USB_ISP1362_HCD_MODULE)
+	&isp1362_hcd_device,
+#endif
+
+#if defined(CONFIG_USB_MUSB_HDRC) || defined(CONFIG_USB_MUSB_HDRC_MODULE)
+	&musb_device,
+#endif
+
+#if defined(CONFIG_SMC91X) || defined(CONFIG_SMC91X_MODULE)
+	&smc91x_device,
+#endif
+
+#if defined(CONFIG_DM9000) || defined(CONFIG_DM9000_MODULE)
+	&dm9000_device,
+#endif
+
+#if defined(CONFIG_BFIN_MAC) || defined(CONFIG_BFIN_MAC_MODULE)
+	&bfin_mac_device,
+#endif
+
+#if defined(CONFIG_USB_NET2272) || defined(CONFIG_USB_NET2272_MODULE)
+	&net2272_bfin_device,
+#endif
+
+#if defined(CONFIG_SPI_BFIN) || defined(CONFIG_SPI_BFIN_MODULE)
+	&bfin_spi0_device,
+#endif
+
+#if defined(CONFIG_FB_BFIN_7393) || defined(CONFIG_FB_BFIN_7393_MODULE)
+	&bfin_fb_adv7393_device,
+#endif
+
+#if defined(CONFIG_SERIAL_BFIN) || defined(CONFIG_SERIAL_BFIN_MODULE)
+	&bfin_uart_device,
+#endif
+
+#if defined(CONFIG_BFIN_SIR) || defined(CONFIG_BFIN_SIR_MODULE)
+	&bfin_sir_device,
+#endif
+
+#if defined(CONFIG_I2C_BLACKFIN_TWI) || defined(CONFIG_I2C_BLACKFIN_TWI_MODULE)
+	&i2c_bfin_twi_device,
+#endif
+
+#if defined(CONFIG_SERIAL_BFIN_SPORT) || defined(CONFIG_SERIAL_BFIN_SPORT_MODULE)
+	&bfin_sport0_uart_device,
+	&bfin_sport1_uart_device,
+#endif
+
+#if defined(CONFIG_PATA_PLATFORM) || defined(CONFIG_PATA_PLATFORM_MODULE)
+	&bfin_pata_device,
+#endif
+
+#if defined(CONFIG_KEYBOARD_GPIO) || defined(CONFIG_KEYBOARD_GPIO_MODULE)
+	&bfin_device_gpiokeys,
+#endif
+
+#if defined(CONFIG_MTD_PHYSMAP) || defined(CONFIG_MTD_PHYSMAP_MODULE)
+	&ezkit_flash_device,
+#endif
+
+	&bfin_gpios_device,
+};
+
+static int __init stamp_init(void)
+{
+	printk(KERN_INFO "%s(): registering device resources\n", __func__);
+
+#ifdef CONFIG_I2C_BOARDINFO
+	i2c_register_board_info(0, bfin_i2c_board_info,
+				ARRAY_SIZE(bfin_i2c_board_info));
+#endif
+
+	platform_add_devices(stamp_devices, ARRAY_SIZE(stamp_devices));
+	spi_register_board_info(bfin_spi_board_info, ARRAY_SIZE(bfin_spi_board_info));
+
+#if defined(CONFIG_PATA_PLATFORM) || defined(CONFIG_PATA_PLATFORM_MODULE)
+	irq_desc[PATA_INT].status |= IRQ_NOAUTOEN;
+#endif
+	return 0;
+}
+
+arch_initcall(stamp_init);
+
+void native_machine_restart(char *cmd)
+{
+	/* workaround reboot hang when booting from SPI */
+	if ((bfin_read_SYSCR() & 0x7) == 0x3)
+		bfin_gpio_reset_spi0_ssel1();
+}
+
+void bfin_get_ether_addr(char *addr)
+{
+	random_ether_addr(addr);
+	printk(KERN_WARNING "%s:%s: Setting Ethernet MAC to a random one\n", __FILE__, __func__);
+}
+EXPORT_SYMBOL(bfin_get_ether_addr);

From 3a87208028ef59215a88a143c723ac0b83c11df0 Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Mon, 21 Jul 2008 09:15:49 -0300
Subject: [PATCH 0083/1171] ACPI: thinkpad-acpi: minor refactor on radio switch
 init

Change the code of hotkey_init, wan_init and bluetooth_init a bit to make it
much easier to add some Kconfig-selected debugging code later.

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
---
 drivers/misc/thinkpad_acpi.c | 49 +++++++++++++++++++-----------------
 1 file changed, 26 insertions(+), 23 deletions(-)

diff --git a/drivers/misc/thinkpad_acpi.c b/drivers/misc/thinkpad_acpi.c
index b5969298f3d3..c800855be27f 100644
--- a/drivers/misc/thinkpad_acpi.c
+++ b/drivers/misc/thinkpad_acpi.c
@@ -2167,9 +2167,10 @@ static int __init hotkey_init(struct ibm_init_struct *iibm)
 		printk(TPACPI_INFO
 			"radio switch found; radios are %s\n",
 			enabled(status, 0));
+	}
+	if (tp_features.hotkey_wlsw)
 		res = add_to_attr_set(hotkey_dev_attributes,
 				&dev_attr_hotkey_radio_sw.attr);
-	}
 
 	/* For X41t, X60t, X61t Tablets... */
 	if (!res && acpi_evalf(hkey_handle, &status, "MHKG", "qd")) {
@@ -2646,18 +2647,19 @@ static int __init bluetooth_init(struct ibm_init_struct *iibm)
 		str_supported(tp_features.bluetooth),
 		status);
 
+	if (tp_features.bluetooth &&
+	    !(status & TP_ACPI_BLUETOOTH_HWPRESENT)) {
+		/* no bluetooth hardware present in system */
+		tp_features.bluetooth = 0;
+		dbg_printk(TPACPI_DBG_INIT,
+			   "bluetooth hardware not installed\n");
+	}
+
 	if (tp_features.bluetooth) {
-		if (!(status & TP_ACPI_BLUETOOTH_HWPRESENT)) {
-			/* no bluetooth hardware present in system */
-			tp_features.bluetooth = 0;
-			dbg_printk(TPACPI_DBG_INIT,
-				   "bluetooth hardware not installed\n");
-		} else {
-			res = sysfs_create_group(&tpacpi_pdev->dev.kobj,
-					&bluetooth_attr_group);
-			if (res)
-				return res;
-		}
+		res = sysfs_create_group(&tpacpi_pdev->dev.kobj,
+				&bluetooth_attr_group);
+		if (res)
+			return res;
 	}
 
 	return (tp_features.bluetooth)? 0 : 1;
@@ -2818,18 +2820,19 @@ static int __init wan_init(struct ibm_init_struct *iibm)
 		str_supported(tp_features.wan),
 		status);
 
+	if (tp_features.wan &&
+	    !(status & TP_ACPI_WANCARD_HWPRESENT)) {
+		/* no wan hardware present in system */
+		tp_features.wan = 0;
+		dbg_printk(TPACPI_DBG_INIT,
+			   "wan hardware not installed\n");
+	}
+
 	if (tp_features.wan) {
-		if (!(status & TP_ACPI_WANCARD_HWPRESENT)) {
-			/* no wan hardware present in system */
-			tp_features.wan = 0;
-			dbg_printk(TPACPI_DBG_INIT,
-				   "wan hardware not installed\n");
-		} else {
-			res = sysfs_create_group(&tpacpi_pdev->dev.kobj,
-					&wan_attr_group);
-			if (res)
-				return res;
-		}
+		res = sysfs_create_group(&tpacpi_pdev->dev.kobj,
+				&wan_attr_group);
+		if (res)
+			return res;
 	}
 
 	return (tp_features.wan)? 0 : 1;

From 733e27c1cc86afae2d9481838693661b3d839950 Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Mon, 21 Jul 2008 09:15:49 -0300
Subject: [PATCH 0084/1171] ACPI: thinkpad-acpi: consolidate wlsw notification
 function

Rename tpacpi_input_send_radiosw() to tpacpi_send_radiosw_update(), and
make it a central point to issue "radio switch changed state" notifications
by consolidating also the poll() notification in the same function.

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
---
 drivers/misc/thinkpad_acpi.c | 39 ++++++++++++++++++------------------
 1 file changed, 19 insertions(+), 20 deletions(-)

diff --git a/drivers/misc/thinkpad_acpi.c b/drivers/misc/thinkpad_acpi.c
index c800855be27f..9179f2367d42 100644
--- a/drivers/misc/thinkpad_acpi.c
+++ b/drivers/misc/thinkpad_acpi.c
@@ -1285,21 +1285,6 @@ static int hotkey_status_set(int status)
 	return 0;
 }
 
-static void tpacpi_input_send_radiosw(void)
-{
-	int wlsw;
-
-	if (tp_features.hotkey_wlsw && !hotkey_get_wlsw(&wlsw)) {
-		mutex_lock(&tpacpi_inputdev_send_mutex);
-
-		input_report_switch(tpacpi_inputdev,
-				    SW_RFKILL_ALL, !!wlsw);
-		input_sync(tpacpi_inputdev);
-
-		mutex_unlock(&tpacpi_inputdev_send_mutex);
-	}
-}
-
 static void tpacpi_input_send_tabletsw(void)
 {
 	int state;
@@ -1921,6 +1906,22 @@ static struct attribute *hotkey_mask_attributes[] __initdata = {
 	&dev_attr_hotkey_wakeup_hotunplug_complete.attr,
 };
 
+static void tpacpi_send_radiosw_update(void)
+{
+	int wlsw;
+
+	if (tp_features.hotkey_wlsw && !hotkey_get_wlsw(&wlsw)) {
+		mutex_lock(&tpacpi_inputdev_send_mutex);
+
+		input_report_switch(tpacpi_inputdev,
+				    SW_RFKILL_ALL, !!wlsw);
+		input_sync(tpacpi_inputdev);
+
+		mutex_unlock(&tpacpi_inputdev_send_mutex);
+	}
+	hotkey_radio_sw_notify_change();
+}
+
 static void hotkey_exit(void)
 {
 #ifdef CONFIG_THINKPAD_ACPI_HOTKEY_POLL
@@ -2288,7 +2289,7 @@ static int __init hotkey_init(struct ibm_init_struct *iibm)
 	tpacpi_inputdev->close = &hotkey_inputdev_close;
 
 	hotkey_poll_setup_safe(1);
-	tpacpi_input_send_radiosw();
+	tpacpi_send_radiosw_update();
 	tpacpi_input_send_tabletsw();
 
 	return 0;
@@ -2420,8 +2421,7 @@ static void hotkey_notify(struct ibm_struct *ibm, u32 event)
 		case 7:
 			/* 0x7000-0x7FFF: misc */
 			if (tp_features.hotkey_wlsw && hkey == 0x7000) {
-				tpacpi_input_send_radiosw();
-				hotkey_radio_sw_notify_change();
+				tpacpi_send_radiosw_update();
 				send_acpi_ev = 0;
 				break;
 			}
@@ -2464,8 +2464,7 @@ static void hotkey_resume(void)
 		printk(TPACPI_ERR
 		       "error while trying to read hot key mask "
 		       "from firmware\n");
-	tpacpi_input_send_radiosw();
-	hotkey_radio_sw_notify_change();
+	tpacpi_send_radiosw_update();
 	hotkey_tablet_mode_notify_change();
 	hotkey_wakeup_reason_notify_change();
 	hotkey_wakeup_hotunplug_complete_notify_change();

From 07431ec82bf9dc74b470a1d820b41c92c4d86e6f Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Mon, 21 Jul 2008 09:15:50 -0300
Subject: [PATCH 0085/1171] ACPI: thinkpad-acpi: prepare for bluetooth and wwan
 rfkill support

Get rid of some forward definitions by moving code around, this will make
the rfkill conversion of wwan and bluetooth a bit cleaner.

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
---
 drivers/misc/thinkpad_acpi.c | 154 +++++++++++++++++------------------
 1 file changed, 74 insertions(+), 80 deletions(-)

diff --git a/drivers/misc/thinkpad_acpi.c b/drivers/misc/thinkpad_acpi.c
index 9179f2367d42..743a4d6098e8 100644
--- a/drivers/misc/thinkpad_acpi.c
+++ b/drivers/misc/thinkpad_acpi.c
@@ -2581,8 +2581,37 @@ enum {
 	TP_ACPI_BLUETOOTH_UNK		= 0x04,	/* unknown function */
 };
 
-static int bluetooth_get_radiosw(void);
-static int bluetooth_set_radiosw(int radio_on);
+static int bluetooth_get_radiosw(void)
+{
+	int status;
+
+	if (!tp_features.bluetooth)
+		return -ENODEV;
+
+	if (!acpi_evalf(hkey_handle, &status, "GBDC", "d"))
+		return -EIO;
+
+	return (status & TP_ACPI_BLUETOOTH_RADIOSSW) != 0;
+}
+
+static int bluetooth_set_radiosw(int radio_on)
+{
+	int status;
+
+	if (!tp_features.bluetooth)
+		return -ENODEV;
+
+	if (!acpi_evalf(hkey_handle, &status, "GBDC", "d"))
+		return -EIO;
+	if (radio_on)
+		status |= TP_ACPI_BLUETOOTH_RADIOSSW;
+	else
+		status &= ~TP_ACPI_BLUETOOTH_RADIOSSW;
+	if (!acpi_evalf(hkey_handle, NULL, "SBDC", "vd", status))
+		return -EIO;
+
+	return 0;
+}
 
 /* sysfs bluetooth enable ---------------------------------------------- */
 static ssize_t bluetooth_enable_show(struct device *dev,
@@ -2628,6 +2657,12 @@ static const struct attribute_group bluetooth_attr_group = {
 	.attrs = bluetooth_attributes,
 };
 
+static void bluetooth_exit(void)
+{
+	sysfs_remove_group(&tpacpi_pdev->dev.kobj,
+			&bluetooth_attr_group);
+}
+
 static int __init bluetooth_init(struct ibm_init_struct *iibm)
 {
 	int res;
@@ -2664,44 +2699,6 @@ static int __init bluetooth_init(struct ibm_init_struct *iibm)
 	return (tp_features.bluetooth)? 0 : 1;
 }
 
-static void bluetooth_exit(void)
-{
-	sysfs_remove_group(&tpacpi_pdev->dev.kobj,
-			&bluetooth_attr_group);
-}
-
-static int bluetooth_get_radiosw(void)
-{
-	int status;
-
-	if (!tp_features.bluetooth)
-		return -ENODEV;
-
-	if (!acpi_evalf(hkey_handle, &status, "GBDC", "d"))
-		return -EIO;
-
-	return ((status & TP_ACPI_BLUETOOTH_RADIOSSW) != 0);
-}
-
-static int bluetooth_set_radiosw(int radio_on)
-{
-	int status;
-
-	if (!tp_features.bluetooth)
-		return -ENODEV;
-
-	if (!acpi_evalf(hkey_handle, &status, "GBDC", "d"))
-		return -EIO;
-	if (radio_on)
-		status |= TP_ACPI_BLUETOOTH_RADIOSSW;
-	else
-		status &= ~TP_ACPI_BLUETOOTH_RADIOSSW;
-	if (!acpi_evalf(hkey_handle, NULL, "SBDC", "vd", status))
-		return -EIO;
-
-	return 0;
-}
-
 /* procfs -------------------------------------------------------------- */
 static int bluetooth_read(char *p)
 {
@@ -2756,8 +2753,37 @@ enum {
 	TP_ACPI_WANCARD_UNK		= 0x04,	/* unknown function */
 };
 
-static int wan_get_radiosw(void);
-static int wan_set_radiosw(int radio_on);
+static int wan_get_radiosw(void)
+{
+	int status;
+
+	if (!tp_features.wan)
+		return -ENODEV;
+
+	if (!acpi_evalf(hkey_handle, &status, "GWAN", "d"))
+		return -EIO;
+
+	return (status & TP_ACPI_WANCARD_RADIOSSW) != 0;
+}
+
+static int wan_set_radiosw(int radio_on)
+{
+	int status;
+
+	if (!tp_features.wan)
+		return -ENODEV;
+
+	if (!acpi_evalf(hkey_handle, &status, "GWAN", "d"))
+		return -EIO;
+	if (radio_on)
+		status |= TP_ACPI_WANCARD_RADIOSSW;
+	else
+		status &= ~TP_ACPI_WANCARD_RADIOSSW;
+	if (!acpi_evalf(hkey_handle, NULL, "SWAN", "vd", status))
+		return -EIO;
+
+	return 0;
+}
 
 /* sysfs wan enable ---------------------------------------------------- */
 static ssize_t wan_enable_show(struct device *dev,
@@ -2803,6 +2829,12 @@ static const struct attribute_group wan_attr_group = {
 	.attrs = wan_attributes,
 };
 
+static void wan_exit(void)
+{
+	sysfs_remove_group(&tpacpi_pdev->dev.kobj,
+		&wan_attr_group);
+}
+
 static int __init wan_init(struct ibm_init_struct *iibm)
 {
 	int res;
@@ -2837,44 +2869,6 @@ static int __init wan_init(struct ibm_init_struct *iibm)
 	return (tp_features.wan)? 0 : 1;
 }
 
-static void wan_exit(void)
-{
-	sysfs_remove_group(&tpacpi_pdev->dev.kobj,
-		&wan_attr_group);
-}
-
-static int wan_get_radiosw(void)
-{
-	int status;
-
-	if (!tp_features.wan)
-		return -ENODEV;
-
-	if (!acpi_evalf(hkey_handle, &status, "GWAN", "d"))
-		return -EIO;
-
-	return ((status & TP_ACPI_WANCARD_RADIOSSW) != 0);
-}
-
-static int wan_set_radiosw(int radio_on)
-{
-	int status;
-
-	if (!tp_features.wan)
-		return -ENODEV;
-
-	if (!acpi_evalf(hkey_handle, &status, "GWAN", "d"))
-		return -EIO;
-	if (radio_on)
-		status |= TP_ACPI_WANCARD_RADIOSSW;
-	else
-		status &= ~TP_ACPI_WANCARD_RADIOSSW;
-	if (!acpi_evalf(hkey_handle, NULL, "SWAN", "vd", status))
-		return -EIO;
-
-	return 0;
-}
-
 /* procfs -------------------------------------------------------------- */
 static int wan_read(char *p)
 {

From 133ec3bd3ae409895eacdce326cdc8d73c249e8a Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Mon, 21 Jul 2008 09:15:50 -0300
Subject: [PATCH 0086/1171] ACPI: thinkpad-acpi: WLSW overrides other rfkill
 switches

On ThinkPads where the WLSW switch exists, the firmware or the hardware
ANDs the WLSW state with the device-specific switches (WWAN, Bluetooth).
It is downright impossible to enable WWAN or Bluetooth when WLSW is
blocking the radios.

This reality does not necessarily carry over to the WWAN and Bluetooth
firmware interfaces, though... so the state thinkpad-acpi was reporting
could be incorrect.

Tie the three switches in the driver so that we keep their state sane.
When WLSL is off, force the other switches to off as well.

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
---
 drivers/misc/thinkpad_acpi.c | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/drivers/misc/thinkpad_acpi.c b/drivers/misc/thinkpad_acpi.c
index 743a4d6098e8..202d63e1b391 100644
--- a/drivers/misc/thinkpad_acpi.c
+++ b/drivers/misc/thinkpad_acpi.c
@@ -2588,6 +2588,10 @@ static int bluetooth_get_radiosw(void)
 	if (!tp_features.bluetooth)
 		return -ENODEV;
 
+	/* WLSW overrides bluetooth in firmware/hardware, reflect that */
+	if (tp_features.hotkey_wlsw && !hotkey_get_wlsw(&status) && !status)
+		return 0;
+
 	if (!acpi_evalf(hkey_handle, &status, "GBDC", "d"))
 		return -EIO;
 
@@ -2601,6 +2605,12 @@ static int bluetooth_set_radiosw(int radio_on)
 	if (!tp_features.bluetooth)
 		return -ENODEV;
 
+	/* WLSW overrides bluetooth in firmware/hardware, but there is no
+	 * reason to risk weird behaviour. */
+	if (tp_features.hotkey_wlsw && !hotkey_get_wlsw(&status) && !status
+	    && radio_on)
+		return -EPERM;
+
 	if (!acpi_evalf(hkey_handle, &status, "GBDC", "d"))
 		return -EIO;
 	if (radio_on)
@@ -2760,6 +2770,10 @@ static int wan_get_radiosw(void)
 	if (!tp_features.wan)
 		return -ENODEV;
 
+	/* WLSW overrides WWAN in firmware/hardware, reflect that */
+	if (tp_features.hotkey_wlsw && !hotkey_get_wlsw(&status) && !status)
+		return 0;
+
 	if (!acpi_evalf(hkey_handle, &status, "GWAN", "d"))
 		return -EIO;
 
@@ -2773,6 +2787,12 @@ static int wan_set_radiosw(int radio_on)
 	if (!tp_features.wan)
 		return -ENODEV;
 
+	/* WLSW overrides bluetooth in firmware/hardware, but there is no
+	 * reason to risk weird behaviour. */
+	if (tp_features.hotkey_wlsw && !hotkey_get_wlsw(&status) && !status
+	    && radio_on)
+		return -EPERM;
+
 	if (!acpi_evalf(hkey_handle, &status, "GWAN", "d"))
 		return -EIO;
 	if (radio_on)

From 0e74dc2646db04b644faa8ea10ff4f408d55cf90 Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Mon, 21 Jul 2008 09:15:51 -0300
Subject: [PATCH 0087/1171] ACPI: thinkpad-acpi: add bluetooth and WWAN rfkill
 support

Add a read/write rfkill interface to the bluetooth radio switch on the
bluetooth submodule, and one for the wireless wan radio switch to the wan
submodule.

Since rfkill does care for when a switch changes state, use WLSW
notifications to also check if the WWAN or Bluetooth switches did not
change state (due to them being slaves of WLSW in firmware/hardware, but
that reality not being always properly exported by the thinkpad firmware).

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Cc: Ivo van Doorn <IvDoorn@gmail.com>
Cc: John W. Linville <linville@tuxdriver.com>
---
 Documentation/laptops/thinkpad-acpi.txt |  22 ++-
 drivers/misc/Kconfig                    |   2 +
 drivers/misc/thinkpad_acpi.c            | 208 +++++++++++++++++++++---
 3 files changed, 200 insertions(+), 32 deletions(-)

diff --git a/Documentation/laptops/thinkpad-acpi.txt b/Documentation/laptops/thinkpad-acpi.txt
index 64b3f146e4b0..1c1c0217ebd1 100644
--- a/Documentation/laptops/thinkpad-acpi.txt
+++ b/Documentation/laptops/thinkpad-acpi.txt
@@ -621,7 +621,8 @@ Bluetooth
 ---------
 
 procfs: /proc/acpi/ibm/bluetooth
-sysfs device attribute: bluetooth_enable
+sysfs device attribute: bluetooth_enable (deprecated)
+sysfs rfkill class: switch "tpacpi_bluetooth_sw"
 
 This feature shows the presence and current state of a ThinkPad
 Bluetooth device in the internal ThinkPad CDC slot.
@@ -643,8 +644,12 @@ Sysfs notes:
 		0: disables Bluetooth / Bluetooth is disabled
 		1: enables Bluetooth / Bluetooth is enabled.
 
-	Note: this interface will be probably be superseded by the
-	generic rfkill class, so it is NOT to be considered stable yet.
+	Note: this interface has been superseded by the	generic rfkill
+	class.  It has been deprecated, and it will be removed in year
+	2010.
+
+	rfkill controller switch "tpacpi_bluetooth_sw": refer to
+	Documentation/rfkill.txt for details.
 
 Video output control -- /proc/acpi/ibm/video
 --------------------------------------------
@@ -1374,7 +1379,8 @@ EXPERIMENTAL: WAN
 -----------------
 
 procfs: /proc/acpi/ibm/wan
-sysfs device attribute: wwan_enable
+sysfs device attribute: wwan_enable (deprecated)
+sysfs rfkill class: switch "tpacpi_wwan_sw"
 
 This feature is marked EXPERIMENTAL because the implementation
 directly accesses hardware registers and may not work as expected. USE
@@ -1404,8 +1410,12 @@ Sysfs notes:
 		0: disables WWAN card / WWAN card is disabled
 		1: enables WWAN card / WWAN card is enabled.
 
-	Note: this interface will be probably be superseded by the
-	generic rfkill class, so it is NOT to be considered stable yet.
+	Note: this interface has been superseded by the	generic rfkill
+	class.  It has been deprecated, and it will be removed in year
+	2010.
+
+	rfkill controller switch "tpacpi_wwan_sw": refer to
+	Documentation/rfkill.txt for details.
 
 Multiple Commands, Module Parameters
 ------------------------------------
diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig
index 1921b8dbb242..b27ca91fd15e 100644
--- a/drivers/misc/Kconfig
+++ b/drivers/misc/Kconfig
@@ -279,6 +279,8 @@ config THINKPAD_ACPI
 	select INPUT
 	select NEW_LEDS
 	select LEDS_CLASS
+	select NET
+	select RFKILL
 	---help---
 	  This is a driver for the IBM and Lenovo ThinkPad laptops. It adds
 	  support for Fn-Fx key combinations, Bluetooth control, video
diff --git a/drivers/misc/thinkpad_acpi.c b/drivers/misc/thinkpad_acpi.c
index 202d63e1b391..dc8d00a45701 100644
--- a/drivers/misc/thinkpad_acpi.c
+++ b/drivers/misc/thinkpad_acpi.c
@@ -68,6 +68,7 @@
 #include <linux/hwmon-sysfs.h>
 #include <linux/input.h>
 #include <linux/leds.h>
+#include <linux/rfkill.h>
 #include <asm/uaccess.h>
 
 #include <linux/dmi.h>
@@ -144,6 +145,12 @@ enum {
 
 #define TPACPI_MAX_ACPI_ARGS 3
 
+/* rfkill switches */
+enum {
+	TPACPI_RFK_BLUETOOTH_SW_ID = 0,
+	TPACPI_RFK_WWAN_SW_ID,
+};
+
 /* Debugging */
 #define TPACPI_LOG TPACPI_FILE ": "
 #define TPACPI_ERR	   KERN_ERR    TPACPI_LOG
@@ -905,6 +912,43 @@ static int __init tpacpi_check_std_acpi_brightness_support(void)
 	return 0;
 }
 
+static int __init tpacpi_new_rfkill(const unsigned int id,
+			struct rfkill **rfk,
+			const enum rfkill_type rfktype,
+			const char *name,
+			int (*toggle_radio)(void *, enum rfkill_state),
+			int (*get_state)(void *, enum rfkill_state *))
+{
+	int res;
+	enum rfkill_state initial_state;
+
+	*rfk = rfkill_allocate(&tpacpi_pdev->dev, rfktype);
+	if (!*rfk) {
+		printk(TPACPI_ERR
+			"failed to allocate memory for rfkill class\n");
+		return -ENOMEM;
+	}
+
+	(*rfk)->name = name;
+	(*rfk)->get_state = get_state;
+	(*rfk)->toggle_radio = toggle_radio;
+
+	if (!get_state(NULL, &initial_state))
+		(*rfk)->state = initial_state;
+
+	res = rfkill_register(*rfk);
+	if (res < 0) {
+		printk(TPACPI_ERR
+			"failed to register %s rfkill switch: %d\n",
+			name, res);
+		rfkill_free(*rfk);
+		*rfk = NULL;
+		return res;
+	}
+
+	return 0;
+}
+
 /*************************************************************************
  * thinkpad-acpi driver attributes
  */
@@ -1906,10 +1950,18 @@ static struct attribute *hotkey_mask_attributes[] __initdata = {
 	&dev_attr_hotkey_wakeup_hotunplug_complete.attr,
 };
 
+static void bluetooth_update_rfk(void);
+static void wan_update_rfk(void);
 static void tpacpi_send_radiosw_update(void)
 {
 	int wlsw;
 
+	/* Sync these BEFORE sending any rfkill events */
+	if (tp_features.bluetooth)
+		bluetooth_update_rfk();
+	if (tp_features.wan)
+		wan_update_rfk();
+
 	if (tp_features.hotkey_wlsw && !hotkey_get_wlsw(&wlsw)) {
 		mutex_lock(&tpacpi_inputdev_send_mutex);
 
@@ -2581,6 +2633,8 @@ enum {
 	TP_ACPI_BLUETOOTH_UNK		= 0x04,	/* unknown function */
 };
 
+static struct rfkill *tpacpi_bluetooth_rfkill;
+
 static int bluetooth_get_radiosw(void)
 {
 	int status;
@@ -2590,15 +2644,29 @@ static int bluetooth_get_radiosw(void)
 
 	/* WLSW overrides bluetooth in firmware/hardware, reflect that */
 	if (tp_features.hotkey_wlsw && !hotkey_get_wlsw(&status) && !status)
-		return 0;
+		return RFKILL_STATE_HARD_BLOCKED;
 
 	if (!acpi_evalf(hkey_handle, &status, "GBDC", "d"))
 		return -EIO;
 
-	return (status & TP_ACPI_BLUETOOTH_RADIOSSW) != 0;
+	return ((status & TP_ACPI_BLUETOOTH_RADIOSSW) != 0) ?
+		RFKILL_STATE_UNBLOCKED : RFKILL_STATE_SOFT_BLOCKED;
 }
 
-static int bluetooth_set_radiosw(int radio_on)
+static void bluetooth_update_rfk(void)
+{
+	int status;
+
+	if (!tpacpi_bluetooth_rfkill)
+		return;
+
+	status = bluetooth_get_radiosw();
+	if (status < 0)
+		return;
+	rfkill_force_state(tpacpi_bluetooth_rfkill, status);
+}
+
+static int bluetooth_set_radiosw(int radio_on, int update_rfk)
 {
 	int status;
 
@@ -2620,6 +2688,9 @@ static int bluetooth_set_radiosw(int radio_on)
 	if (!acpi_evalf(hkey_handle, NULL, "SBDC", "vd", status))
 		return -EIO;
 
+	if (update_rfk)
+		bluetooth_update_rfk();
+
 	return 0;
 }
 
@@ -2634,7 +2705,8 @@ static ssize_t bluetooth_enable_show(struct device *dev,
 	if (status < 0)
 		return status;
 
-	return snprintf(buf, PAGE_SIZE, "%d\n", status ? 1 : 0);
+	return snprintf(buf, PAGE_SIZE, "%d\n",
+			(status == RFKILL_STATE_UNBLOCKED) ? 1 : 0);
 }
 
 static ssize_t bluetooth_enable_store(struct device *dev,
@@ -2647,7 +2719,7 @@ static ssize_t bluetooth_enable_store(struct device *dev,
 	if (parse_strtoul(buf, 1, &t))
 		return -EINVAL;
 
-	res = bluetooth_set_radiosw(t);
+	res = bluetooth_set_radiosw(t, 1);
 
 	return (res) ? res : count;
 }
@@ -2667,8 +2739,27 @@ static const struct attribute_group bluetooth_attr_group = {
 	.attrs = bluetooth_attributes,
 };
 
+static int tpacpi_bluetooth_rfk_get(void *data, enum rfkill_state *state)
+{
+	int bts = bluetooth_get_radiosw();
+
+	if (bts < 0)
+		return bts;
+
+	*state = bts;
+	return 0;
+}
+
+static int tpacpi_bluetooth_rfk_set(void *data, enum rfkill_state state)
+{
+	return bluetooth_set_radiosw((state == RFKILL_STATE_UNBLOCKED), 0);
+}
+
 static void bluetooth_exit(void)
 {
+	if (tpacpi_bluetooth_rfkill)
+		rfkill_unregister(tpacpi_bluetooth_rfkill);
+
 	sysfs_remove_group(&tpacpi_pdev->dev.kobj,
 			&bluetooth_attr_group);
 }
@@ -2699,14 +2790,26 @@ static int __init bluetooth_init(struct ibm_init_struct *iibm)
 			   "bluetooth hardware not installed\n");
 	}
 
-	if (tp_features.bluetooth) {
-		res = sysfs_create_group(&tpacpi_pdev->dev.kobj,
+	if (!tp_features.bluetooth)
+		return 1;
+
+	res = sysfs_create_group(&tpacpi_pdev->dev.kobj,
 				&bluetooth_attr_group);
-		if (res)
-			return res;
+	if (res)
+		return res;
+
+	res = tpacpi_new_rfkill(TPACPI_RFK_BLUETOOTH_SW_ID,
+				&tpacpi_bluetooth_rfkill,
+				RFKILL_TYPE_BLUETOOTH,
+				"tpacpi_bluetooth_sw",
+				tpacpi_bluetooth_rfk_set,
+				tpacpi_bluetooth_rfk_get);
+	if (res) {
+		bluetooth_exit();
+		return res;
 	}
 
-	return (tp_features.bluetooth)? 0 : 1;
+	return 0;
 }
 
 /* procfs -------------------------------------------------------------- */
@@ -2719,7 +2822,8 @@ static int bluetooth_read(char *p)
 		len += sprintf(p + len, "status:\t\tnot supported\n");
 	else {
 		len += sprintf(p + len, "status:\t\t%s\n",
-				(status)? "enabled" : "disabled");
+				(status == RFKILL_STATE_UNBLOCKED) ?
+					"enabled" : "disabled");
 		len += sprintf(p + len, "commands:\tenable, disable\n");
 	}
 
@@ -2735,9 +2839,9 @@ static int bluetooth_write(char *buf)
 
 	while ((cmd = next_cmd(&buf))) {
 		if (strlencmp(cmd, "enable") == 0) {
-			bluetooth_set_radiosw(1);
+			bluetooth_set_radiosw(1, 1);
 		} else if (strlencmp(cmd, "disable") == 0) {
-			bluetooth_set_radiosw(0);
+			bluetooth_set_radiosw(0, 1);
 		} else
 			return -EINVAL;
 	}
@@ -2763,6 +2867,8 @@ enum {
 	TP_ACPI_WANCARD_UNK		= 0x04,	/* unknown function */
 };
 
+static struct rfkill *tpacpi_wan_rfkill;
+
 static int wan_get_radiosw(void)
 {
 	int status;
@@ -2772,15 +2878,29 @@ static int wan_get_radiosw(void)
 
 	/* WLSW overrides WWAN in firmware/hardware, reflect that */
 	if (tp_features.hotkey_wlsw && !hotkey_get_wlsw(&status) && !status)
-		return 0;
+		return RFKILL_STATE_HARD_BLOCKED;
 
 	if (!acpi_evalf(hkey_handle, &status, "GWAN", "d"))
 		return -EIO;
 
-	return (status & TP_ACPI_WANCARD_RADIOSSW) != 0;
+	return ((status & TP_ACPI_WANCARD_RADIOSSW) != 0) ?
+		RFKILL_STATE_UNBLOCKED : RFKILL_STATE_SOFT_BLOCKED;
 }
 
-static int wan_set_radiosw(int radio_on)
+static void wan_update_rfk(void)
+{
+	int status;
+
+	if (!tpacpi_wan_rfkill)
+		return;
+
+	status = wan_get_radiosw();
+	if (status < 0)
+		return;
+	rfkill_force_state(tpacpi_wan_rfkill, status);
+}
+
+static int wan_set_radiosw(int radio_on, int update_rfk)
 {
 	int status;
 
@@ -2802,6 +2922,9 @@ static int wan_set_radiosw(int radio_on)
 	if (!acpi_evalf(hkey_handle, NULL, "SWAN", "vd", status))
 		return -EIO;
 
+	if (update_rfk)
+		wan_update_rfk();
+
 	return 0;
 }
 
@@ -2816,7 +2939,8 @@ static ssize_t wan_enable_show(struct device *dev,
 	if (status < 0)
 		return status;
 
-	return snprintf(buf, PAGE_SIZE, "%d\n", status ? 1 : 0);
+	return snprintf(buf, PAGE_SIZE, "%d\n",
+			(status == RFKILL_STATE_UNBLOCKED) ? 1 : 0);
 }
 
 static ssize_t wan_enable_store(struct device *dev,
@@ -2829,7 +2953,7 @@ static ssize_t wan_enable_store(struct device *dev,
 	if (parse_strtoul(buf, 1, &t))
 		return -EINVAL;
 
-	res = wan_set_radiosw(t);
+	res = wan_set_radiosw(t, 1);
 
 	return (res) ? res : count;
 }
@@ -2849,8 +2973,27 @@ static const struct attribute_group wan_attr_group = {
 	.attrs = wan_attributes,
 };
 
+static int tpacpi_wan_rfk_get(void *data, enum rfkill_state *state)
+{
+	int wans = wan_get_radiosw();
+
+	if (wans < 0)
+		return wans;
+
+	*state = wans;
+	return 0;
+}
+
+static int tpacpi_wan_rfk_set(void *data, enum rfkill_state state)
+{
+	return wan_set_radiosw((state == RFKILL_STATE_UNBLOCKED), 0);
+}
+
 static void wan_exit(void)
 {
+	if (tpacpi_wan_rfkill)
+		rfkill_unregister(tpacpi_wan_rfkill);
+
 	sysfs_remove_group(&tpacpi_pdev->dev.kobj,
 		&wan_attr_group);
 }
@@ -2879,14 +3022,26 @@ static int __init wan_init(struct ibm_init_struct *iibm)
 			   "wan hardware not installed\n");
 	}
 
-	if (tp_features.wan) {
-		res = sysfs_create_group(&tpacpi_pdev->dev.kobj,
+	if (!tp_features.wan)
+		return 1;
+
+	res = sysfs_create_group(&tpacpi_pdev->dev.kobj,
 				&wan_attr_group);
-		if (res)
-			return res;
+	if (res)
+		return res;
+
+	res = tpacpi_new_rfkill(TPACPI_RFK_WWAN_SW_ID,
+				&tpacpi_wan_rfkill,
+				RFKILL_TYPE_WWAN,
+				"tpacpi_wwan_sw",
+				tpacpi_wan_rfk_set,
+				tpacpi_wan_rfk_get);
+	if (res) {
+		wan_exit();
+		return res;
 	}
 
-	return (tp_features.wan)? 0 : 1;
+	return 0;
 }
 
 /* procfs -------------------------------------------------------------- */
@@ -2899,7 +3054,8 @@ static int wan_read(char *p)
 		len += sprintf(p + len, "status:\t\tnot supported\n");
 	else {
 		len += sprintf(p + len, "status:\t\t%s\n",
-				(status)? "enabled" : "disabled");
+				(status == RFKILL_STATE_UNBLOCKED) ?
+					"enabled" : "disabled");
 		len += sprintf(p + len, "commands:\tenable, disable\n");
 	}
 
@@ -2915,9 +3071,9 @@ static int wan_write(char *buf)
 
 	while ((cmd = next_cmd(&buf))) {
 		if (strlencmp(cmd, "enable") == 0) {
-			wan_set_radiosw(1);
+			wan_set_radiosw(1, 1);
 		} else if (strlencmp(cmd, "disable") == 0) {
-			wan_set_radiosw(0);
+			wan_set_radiosw(0, 1);
 		} else
 			return -EINVAL;
 	}

From 490673dc98adfc7de1703cc88508902bd10f446b Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Mon, 21 Jul 2008 09:15:51 -0300
Subject: [PATCH 0088/1171] ACPI: thinkpad-acpi: bump up version to 0.21

rfkill support deserves a new version checkpoint...

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
---
 Documentation/laptops/thinkpad-acpi.txt | 4 ++--
 drivers/misc/thinkpad_acpi.c            | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/Documentation/laptops/thinkpad-acpi.txt b/Documentation/laptops/thinkpad-acpi.txt
index 1c1c0217ebd1..02dc748b76c4 100644
--- a/Documentation/laptops/thinkpad-acpi.txt
+++ b/Documentation/laptops/thinkpad-acpi.txt
@@ -1,7 +1,7 @@
 		     ThinkPad ACPI Extras Driver
 
-                            Version 0.20
-                          April 09th, 2008
+                            Version 0.21
+                           May 29th, 2008
 
                Borislav Deianov <borislav@users.sf.net>
              Henrique de Moraes Holschuh <hmh@hmh.eng.br>
diff --git a/drivers/misc/thinkpad_acpi.c b/drivers/misc/thinkpad_acpi.c
index dc8d00a45701..3eb01afe4306 100644
--- a/drivers/misc/thinkpad_acpi.c
+++ b/drivers/misc/thinkpad_acpi.c
@@ -21,7 +21,7 @@
  *  02110-1301, USA.
  */
 
-#define TPACPI_VERSION "0.20"
+#define TPACPI_VERSION "0.21"
 #define TPACPI_SYSFS_VERSION 0x020200
 
 /*

From bf20e740a4bcc686de02e2fd1c1810a58872f46e Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Mon, 21 Jul 2008 09:15:51 -0300
Subject: [PATCH 0089/1171] ACPI: thinkpad-acpi: don't misdetect in
 get_thinkpad_model_data() on -ENOMEM

Explicitly check for memory allocation failures, and return status to
indicate that we could not collect data due to errors.

This lets the driver have a far more predictable failure mode on ENOMEM in
that codepath: it will refuse to load.  This is far better than trying to
proceed with missing data which is used to detect quirks, etc.

Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
---
 drivers/misc/thinkpad_acpi.c | 47 +++++++++++++++++++++++++-----------
 1 file changed, 33 insertions(+), 14 deletions(-)

diff --git a/drivers/misc/thinkpad_acpi.c b/drivers/misc/thinkpad_acpi.c
index 3eb01afe4306..d3eb7903c346 100644
--- a/drivers/misc/thinkpad_acpi.c
+++ b/drivers/misc/thinkpad_acpi.c
@@ -6340,13 +6340,18 @@ err_out:
 
 /* Probing */
 
-static void __init get_thinkpad_model_data(struct thinkpad_id_data *tp)
+/* returns 0 - probe ok, or < 0 - probe error.
+ * Probe ok doesn't mean thinkpad found.
+ * On error, kfree() cleanup on tp->* is not performed, caller must do it */
+static int __must_check __init get_thinkpad_model_data(
+						struct thinkpad_id_data *tp)
 {
 	const struct dmi_device *dev = NULL;
 	char ec_fw_string[18];
+	char const *s;
 
 	if (!tp)
-		return;
+		return -EINVAL;
 
 	memset(tp, 0, sizeof(*tp));
 
@@ -6355,12 +6360,14 @@ static void __init get_thinkpad_model_data(struct thinkpad_id_data *tp)
 	else if (dmi_name_in_vendors("LENOVO"))
 		tp->vendor = PCI_VENDOR_ID_LENOVO;
 	else
-		return;
+		return 0;
 
-	tp->bios_version_str = kstrdup(dmi_get_system_info(DMI_BIOS_VERSION),
-					GFP_KERNEL);
+	s = dmi_get_system_info(DMI_BIOS_VERSION);
+	tp->bios_version_str = kstrdup(s, GFP_KERNEL);
+	if (s && !tp->bios_version_str)
+		return -ENOMEM;
 	if (!tp->bios_version_str)
-		return;
+		return 0;
 	tp->bios_model = tp->bios_version_str[0]
 			 | (tp->bios_version_str[1] << 8);
 
@@ -6379,21 +6386,27 @@ static void __init get_thinkpad_model_data(struct thinkpad_id_data *tp)
 			ec_fw_string[strcspn(ec_fw_string, " ]")] = 0;
 
 			tp->ec_version_str = kstrdup(ec_fw_string, GFP_KERNEL);
+			if (!tp->ec_version_str)
+				return -ENOMEM;
 			tp->ec_model = ec_fw_string[0]
 					| (ec_fw_string[1] << 8);
 			break;
 		}
 	}
 
-	tp->model_str = kstrdup(dmi_get_system_info(DMI_PRODUCT_VERSION),
-					GFP_KERNEL);
-	if (tp->model_str && strnicmp(tp->model_str, "ThinkPad", 8) != 0) {
-		kfree(tp->model_str);
-		tp->model_str = NULL;
+	s = dmi_get_system_info(DMI_PRODUCT_VERSION);
+	if (s && !strnicmp(s, "ThinkPad", 8)) {
+		tp->model_str = kstrdup(s, GFP_KERNEL);
+		if (!tp->model_str)
+			return -ENOMEM;
 	}
 
-	tp->nummodel_str = kstrdup(dmi_get_system_info(DMI_PRODUCT_NAME),
-					GFP_KERNEL);
+	s = dmi_get_system_info(DMI_PRODUCT_NAME);
+	tp->nummodel_str = kstrdup(s, GFP_KERNEL);
+	if (s && !tp->nummodel_str)
+		return -ENOMEM;
+
+	return 0;
 }
 
 static int __init probe_for_thinkpad(void)
@@ -6656,7 +6669,13 @@ static int __init thinkpad_acpi_module_init(void)
 
 	/* Driver-level probe */
 
-	get_thinkpad_model_data(&thinkpad_id);
+	ret = get_thinkpad_model_data(&thinkpad_id);
+	if (ret) {
+		printk(TPACPI_ERR
+			"unable to get DMI data: %d\n", ret);
+		thinkpad_acpi_module_exit();
+		return ret;
+	}
 	ret = probe_for_thinkpad();
 	if (ret) {
 		thinkpad_acpi_module_exit();

From f88133d76ea38761b7379d6233b752ed82250a4a Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Mon, 21 Jul 2008 15:57:45 +0200
Subject: [PATCH 0090/1171] acpi: fix crash in core ACPI code, triggered by
 CONFIG_ACPI_PCI_SLOT=y

-tip testing found the following boot crash on 32-bit x86 (Core2Duo
laptop) yesterday:

[    5.606664] scsi4 : ata_piix
[    5.606664] scsi5 : ata_piix
[    5.606664] ACPI Error (psargs-0358): [\_SB_.PCI0.LPC_.EC__.BSTA] Namespace lookup failure, AE_NOT_FOUND
[    5.606664] ACPI Error (psparse-0530): ACPI Error (nsnames-0186): Invalid NS Node (f7c0e960) while traversing path [20080609]
[    5.606664] BUG: unable to handle kernel NULL pointer dereference at 0000000f
[    5.606664] IP: [<80339e2f>] acpi_ns_build_external_path+0x1f/0x80
[    5.609997] *pdpt = 0000000000a03001 *pde = 0000000000000000
[    5.609997] Oops: 0002 [#1] SMP
[    5.609997]
[    5.609997] Pid: 1, comm: swapper Not tainted (2.6.26-tip-03965-gbbfb62e-dirty #3153)
[    5.609997] EIP: 0060:[<80339e2f>] EFLAGS: 00010286 CPU: 0
[    5.609997] EIP is at acpi_ns_build_external_path+0x1f/0x80
[    5.609997] EAX: f7c18c18 EBX: ffffffff ECX: 00000010 EDX: 00000000
[    5.609997] ESI: f7c18c18 EDI: 00000010 EBP: f7c4dc28 ESP: f7c4dc18
[    5.609997]  DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: 0068
[    5.609997] Process swapper (pid: 1, ti=f7c4c000 task=f7c50000 task.ti=f7c4c000)
[    5.609997] Stack: 00000000 00000000 f7c18c18 f7c4dc48 f7c4dc40 80339ed0 00000000 f7c18c18
[    5.609997]        8084c1b6 8084c1b6 f7c4dc58 8033a60a 00000000 00000010 00000000 f7c18c18
[    5.609997]        f7c4dc70 8033a68f f7c18c18 00000000 f6de7600 00000005 f7c4dc98 8033c34d
[    5.609997] Call Trace:
[    5.609997]  [<80339ed0>] ? acpi_ns_handle_to_pathname+0x40/0x72
[    5.609997]  [<8033a60a>] ? acpi_ns_print_node_pathname+0x2c/0x61
[    5.609997]  [<8033a68f>] ? acpi_ns_report_method_error+0x50/0x6d
[    5.609997]  [<8033c34d>] ? acpi_ps_parse_aml+0x149/0x2f9
[    5.609997]  [<8033d6dd>] ? acpi_ps_execute_method+0x132/0x201
[    5.609997]  [<80339d19>] ? acpi_ns_evaluate+0x1ad/0x258
[    5.609997]  [<803406c4>] ? acpi_ut_evaluate_object+0x55/0x18f
[    5.609997]  [<803408b7>] ? acpi_ut_execute_STA+0x22/0x7a
[    5.609997]  [<8033a907>] ? acpi_get_object_info+0x131/0x1be
[    5.609997]  [<80344bb2>] ? do_acpi_find_child+0x22/0x4b
[    5.609997]  [<8033b855>] ? acpi_ns_walk_namespace+0xa5/0x124
[    5.609997]  [<803394f3>] ? acpi_walk_namespace+0x54/0x74
[    5.609997]  [<80344b90>] ? do_acpi_find_child+0x0/0x4b
[    5.609997]  [<80344b85>] ? acpi_get_child+0x38/0x43
[    5.609997]  [<80344b90>] ? do_acpi_find_child+0x0/0x4b
[    5.609997]  [<804d0148>] ? ata_acpi_associate+0xb5/0x1b5
[    5.609997]  [<804c6ecb>] ? ata_scsi_add_hosts+0x8e/0xdc
[    5.609997]  [<804c40c8>] ? ata_host_register+0x9f/0x1d6
[    5.609997]  [<804cbc7f>] ? ata_pci_sff_activate_host+0x179/0x19f
[    5.609997]  [<804cdd45>] ? ata_sff_interrupt+0x0/0x1c7
[    5.609997]  [<8069b033>] ? piix_init_one+0x569/0x5b0
[    5.609997]  [<801bd400>] ? sysfs_ilookup_test+0x0/0x11
[    5.609997]  [<801987d7>] ? ilookup5_nowait+0x29/0x30
[    5.609997]  [<802efc7e>] ? pci_match_device+0x99/0xa3
[    5.609997]  [<802efd3c>] ? pci_device_probe+0x39/0x59
[    5.609997]  [<803bc4af>] ? driver_probe_device+0xa0/0x11b
[    5.609997]  [<803bc564>] ? __driver_attach+0x3a/0x59
[    5.609997]  [<803bbde3>] ? bus_for_each_dev+0x36/0x58
[    5.609997]  [<803bc354>] ? driver_attach+0x14/0x16
[    5.609997]  [<803bc52a>] ? __driver_attach+0x0/0x59
[    5.609997]  [<803bc161>] ? bus_add_driver+0x93/0x196
[    5.609997]  [<803bc773>] ? driver_register+0x71/0xcd
[    5.609997]  [<802eff05>] ? __pci_register_driver+0x3f/0x6e
[    5.609997]  [<809af7ff>] ? piix_init+0x14/0x24
[    5.609997]  [<80984568>] ? kernel_init+0x128/0x269
[    5.609997]  [<809af7eb>] ? piix_init+0x0/0x24
[    5.609997]  [<802e2758>] ? trace_hardirqs_on_thunk+0xc/0x10
[    5.609997]  [<80116aef>] ? restore_nocheck_notrace+0x0/0xe
[    5.609997]  [<80984440>] ? kernel_init+0x0/0x269
[    5.609997]  [<80984440>] ? kernel_init+0x0/0x269
[    5.609997]  [<80117d87>] ? kernel_thread_helper+0x7/0x10
[    5.609997]  =======================
[    5.609997] Code: 75 02 b3 01 8d 43 01 8b 5d fc c9 c3 55 89 e5 57 89 cf 56 53 89 d3 4b 83 ec 04 83 fb 03 89 55 f0 77 09 c6 01 5c c6 41 01 00 eb 59 <c6> 04 19 00 8b 55 f0 8d 34 11 89 c2 eb 19 8b 42 08 83 eb 05 89
[    5.609997] EIP: [<80339e2f>] acpi_ns_build_external_path+0x1f/0x80 SS:ESP 0068:f7c4dc18
[    5.613331] Kernel panic - not syncing: Fatal exception
[    5.613331] Rebooting in 1 seconds..[    4.646664] ata1: SATA link up 1.5 Gbps (SStatus 113 SControl 300)

I have bisected it down to:

 # bad:  [5b664cbe] Merge branch 'upstream-linus' of git://git.kernel.
 # good: [bce7f795] Linux 2.6.26
 # good: [e18425ab] Merge branch 'tracing/for-linus' of git://git.kern
 # good: [cadc7236] Merge branch 'bkl-removal' into next
 # good: [4515889a] Merge branch 'merge' of git://git.kernel.org/pub/s
 # good: [42fdd14e] Merge git://git.kernel.org/pub/scm/linux/kernel/gi
 # good: [8a0ca91f] Merge branch 'for-linus' of git://git.kernel.org/p
 # bad:  [0af4b8cb] ACPI: Introduce new device wakeup flag 'prepared'
 # good: [fe997407] PCI: construct one fakephp slot per PCI slot
 # bad:  [531f254a] PCIE: aer: use dev_printk when possible
 # bad:  [15650a20] x86/PCI: fixup early quirk probing
 # good: [0e6859d9] ACPI PM: Remove obsolete Toshiba workaround
 # bad:  [8344b566] PCI: ACPI PCI slot detection driver
 # good: [f46753c9] PCI: introduce pci_slot

 | 8344b568f5bdc7ee1bba909de3294c6348c36056 is first bad commit
 | commit 8344b568f5bdc7ee1bba909de3294c6348c36056
 | Author: Alex Chiang <achiang@hp.com>
 | Date:   Tue Jun 10 15:30:42 2008 -0600
 |
 |     PCI: ACPI PCI slot detection driver
 |
 |     Detect all physical PCI slots as described by ACPI, and create entries in
 |     /sys/bus/pci/slots/.

I.e. the new CONFIG_ACPI_PCI_SLOT=y option was causing this crash.

But the bug is not mainly in this new PCI code - that code was just
hitting the ACPI code in a new way which made ACPI break.

The crash signature shows that we are crashing on this instruction:

   movb $0x0, (%ecx, %ebx, 1)

ECX and EBX are 0x10 and -1. It's this line in
drivers/acpi/namespace/nsnames.c's acpi_ns_build_external_path():

        name_buffer[index] = 0;

I.e. name_buffer is 0x10 and index is -1.

index -1 corresponds to size 0, and name_buffer 0x10 is slab's
ZERO_SIZE_PTR special-case for zero-sized allocations.

I.e. when we called acpi_ns_handle_to_pathname(), we got required_size
of 0 due to an error condition, but this is passed to the ACPI allocator
unconditionally:

        required_size = acpi_ns_get_pathname_length(node);

        /* Validate/Allocate/Clear caller buffer */

        status = acpi_ut_initialize_buffer(buffer, required_size);
        if (ACPI_FAILURE(status)) {
                return_ACPI_STATUS(status);
        }

Where acpi_ut_initialize_buffer(), through many (unnecessary) layers,
ends up calling kzalloc(0). Which returns 0x10 and that then causes the
crash later on.

So fix both callers of acpi_ns_get_pathname_length(), which can return 0
in case of an invalid node.

Also add a WARN_ON() against zero sized allocations in
acpi_ut_initialize_buffer() to make it easier to find similar instances
of this bug.

I have tested this patch for the past 24 hours and the crash has not
reappeared.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
---
 drivers/acpi/namespace/nsnames.c | 8 ++++++++
 drivers/acpi/utilities/utalloc.c | 4 ++++
 2 files changed, 12 insertions(+)

diff --git a/drivers/acpi/namespace/nsnames.c b/drivers/acpi/namespace/nsnames.c
index cffef1bcbdbc..549db42f16cf 100644
--- a/drivers/acpi/namespace/nsnames.c
+++ b/drivers/acpi/namespace/nsnames.c
@@ -137,6 +137,10 @@ char *acpi_ns_get_external_pathname(struct acpi_namespace_node *node)
 	/* Calculate required buffer size based on depth below root */
 
 	size = acpi_ns_get_pathname_length(node);
+	if (!size) {
+		ACPI_ERROR((AE_INFO, "Invalid node failure"));
+		return_PTR(NULL);
+	}
 
 	/* Allocate a buffer to be returned to caller */
 
@@ -229,6 +233,10 @@ acpi_ns_handle_to_pathname(acpi_handle target_handle,
 	/* Determine size required for the caller buffer */
 
 	required_size = acpi_ns_get_pathname_length(node);
+	if (!required_size) {
+		ACPI_ERROR((AE_INFO, "Invalid node failure"));
+		return_ACPI_STATUS(AE_ERROR);
+	}
 
 	/* Validate/Allocate/Clear caller buffer */
 
diff --git a/drivers/acpi/utilities/utalloc.c b/drivers/acpi/utilities/utalloc.c
index 3dfb8a442b26..e7bf34a7b1d2 100644
--- a/drivers/acpi/utilities/utalloc.c
+++ b/drivers/acpi/utilities/utalloc.c
@@ -242,6 +242,10 @@ acpi_ut_initialize_buffer(struct acpi_buffer * buffer,
 {
 	acpi_status status = AE_OK;
 
+	if (!required_length) {
+		WARN_ON(1);
+		return AE_ERROR;
+	}
 	switch (buffer->length) {
 	case ACPI_NO_BUFFER:
 

From 11d579ee0a19052a5a90ebfe0c39e7ed8ce8a9dc Mon Sep 17 00:00:00 2001
From: Wolfram Sang <w.sang@pengutronix.de>
Date: Sat, 28 Jun 2008 20:31:52 +0200
Subject: [PATCH 0091/1171] powerpc/mpc5200: Fix wrong 'no interrupt' handling
 in of_i2c

If an I2C device node does not specify an interrupt, the .irq member of the
board_info struct was set to -1. This caused crashes on following
irq_dispose_mappings. Leave it NO_IRQ as returned from irq_of_parse_and_map.
(Suggesting -1 as 'i2c-no-irq' used to be a bug in linux/i2c.h.)

Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
Acked-by: Sean MacLennan <smaclennan@pikatech.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
---
 drivers/of/of_i2c.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/of/of_i2c.c b/drivers/of/of_i2c.c
index 5c015d310d4a..344e1b03dd8b 100644
--- a/drivers/of/of_i2c.c
+++ b/drivers/of/of_i2c.c
@@ -91,8 +91,6 @@ void of_register_i2c_devices(struct i2c_adapter *adap,
 		}
 
 		info.irq = irq_of_parse_and_map(node, 0);
-		if (info.irq == NO_IRQ)
-			info.irq = -1;
 
 		if (of_find_i2c_driver(node, &info) < 0) {
 			irq_dispose_mapping(info.irq);

From 6a4a636fad018500c5db7a2b56a00caeb21cbb2c Mon Sep 17 00:00:00 2001
From: Jon Smirl <jonsmirl@gmail.com>
Date: Sun, 20 Jul 2008 11:27:22 -0400
Subject: [PATCH 0092/1171] powerpc/mpc5200: Add AC97 register definitions for
 the MPC52xx PSC

Needed by the PSC AC97 sound driver

Signed-off-by: Jon Smirl <jonsmirl@gmail.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
---
 include/asm-powerpc/mpc52xx_psc.h | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/include/asm-powerpc/mpc52xx_psc.h b/include/asm-powerpc/mpc52xx_psc.h
index 710c5d36efaa..5467c2c0faa7 100644
--- a/include/asm-powerpc/mpc52xx_psc.h
+++ b/include/asm-powerpc/mpc52xx_psc.h
@@ -132,8 +132,12 @@ struct mpc52xx_psc {
 	u8		reserved5[3];
 	u8		ctlr;		/* PSC + 0x1c */
 	u8		reserved6[3];
-	u16		ccr;		/* PSC + 0x20 */
-	u8		reserved7[14];
+	/* BitClkDiv field of CCR is byte swapped in
+	 * the hardware for mpc5200/b compatibility */
+	u32		ccr;		/* PSC + 0x20 */
+	u32		ac97_slots;	/* PSC + 0x24 */
+	u32		ac97_cmd;	/* PSC + 0x28 */
+	u32		ac97_data;	/* PSC + 0x2c */
 	u8		ivr;		/* PSC + 0x30 */
 	u8		reserved8[3];
 	u8		ip;		/* PSC + 0x34 */

From 78f56bd3d2dbe173bf1a946b353bf72ab9c0b94e Mon Sep 17 00:00:00 2001
From: Jon Smirl <jonsmirl@gmail.com>
Date: Sun, 20 Jul 2008 11:30:08 -0400
Subject: [PATCH 0093/1171] powerpc/mpc5200: Remove fsl-soc.c from mpc5200
 build, it is not needed.

Signed-off-by: Jon Smirl <jonsmirl@gmail.com>
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
---
 arch/powerpc/platforms/52xx/Kconfig | 1 -
 1 file changed, 1 deletion(-)

diff --git a/arch/powerpc/platforms/52xx/Kconfig b/arch/powerpc/platforms/52xx/Kconfig
index acd2fc8cf492..981b84b7599a 100644
--- a/arch/powerpc/platforms/52xx/Kconfig
+++ b/arch/powerpc/platforms/52xx/Kconfig
@@ -1,7 +1,6 @@
 config PPC_MPC52xx
 	bool "52xx-based boards"
 	depends on PPC_MULTIPLATFORM && PPC32
-	select FSL_SOC
 	select PPC_CLOCK
 
 config PPC_MPC5200_SIMPLE

From 6d5509babce654fd9ce0ff6689dbdf6ce56c43ae Mon Sep 17 00:00:00 2001
From: Grant Likely <grant.likely@secretlab.ca>
Date: Thu, 15 May 2008 17:04:53 -0600
Subject: [PATCH 0094/1171] powerpc/mpc5200: Make mpc5200 GPIO driver select
 the GENERIC_GPIO config

CONFIG_GENERIC_GPIO is needed for the gpio driver to work.

Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
---
 arch/powerpc/platforms/52xx/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/arch/powerpc/platforms/52xx/Kconfig b/arch/powerpc/platforms/52xx/Kconfig
index 981b84b7599a..c1ca4f3ee5da 100644
--- a/arch/powerpc/platforms/52xx/Kconfig
+++ b/arch/powerpc/platforms/52xx/Kconfig
@@ -46,6 +46,7 @@ config PPC_MPC5200_BUGFIX
 config PPC_MPC5200_GPIO
 	bool "MPC5200 GPIO support"
 	depends on PPC_MPC52xx
+	select GENERIC_GPIO
 	select HAVE_GPIO_LIB
 	help
 	  Enable gpiolib support for mpc5200 based boards

From a19dd1bd7df839c52a668abcf288c2239442c3c9 Mon Sep 17 00:00:00 2001
From: Grant Likely <grant.likely@secretlab.ca>
Date: Tue, 22 Jul 2008 01:13:54 -0600
Subject: [PATCH 0095/1171] powerpc/mpc5200: add PSC SICR bit definitions

Required by the PSC I2S audio driver.

Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
---
 include/asm-powerpc/mpc52xx_psc.h | 32 ++++++++++++++++++++++++++++++-
 1 file changed, 31 insertions(+), 1 deletion(-)

diff --git a/include/asm-powerpc/mpc52xx_psc.h b/include/asm-powerpc/mpc52xx_psc.h
index 5467c2c0faa7..8917ed630565 100644
--- a/include/asm-powerpc/mpc52xx_psc.h
+++ b/include/asm-powerpc/mpc52xx_psc.h
@@ -60,10 +60,12 @@
 #define MPC52xx_PSC_RXTX_FIFO_ALARM	0x0002
 #define MPC52xx_PSC_RXTX_FIFO_EMPTY	0x0001
 
-/* PSC interrupt mask bits */
+/* PSC interrupt status/mask bits */
 #define MPC52xx_PSC_IMR_TXRDY		0x0100
 #define MPC52xx_PSC_IMR_RXRDY		0x0200
 #define MPC52xx_PSC_IMR_DB		0x0400
+#define MPC52xx_PSC_IMR_TXEMP		0x0800
+#define MPC52xx_PSC_IMR_ORERR		0x1000
 #define MPC52xx_PSC_IMR_IPC		0x8000
 
 /* PSC input port change bit */
@@ -92,6 +94,34 @@
 
 #define MPC52xx_PSC_RFNUM_MASK	0x01ff
 
+#define MPC52xx_PSC_SICR_DTS1			(1 << 29)
+#define MPC52xx_PSC_SICR_SHDR			(1 << 28)
+#define MPC52xx_PSC_SICR_SIM_MASK		(0xf << 24)
+#define MPC52xx_PSC_SICR_SIM_UART		(0x0 << 24)
+#define MPC52xx_PSC_SICR_SIM_UART_DCD		(0x8 << 24)
+#define MPC52xx_PSC_SICR_SIM_CODEC_8		(0x1 << 24)
+#define MPC52xx_PSC_SICR_SIM_CODEC_16		(0x2 << 24)
+#define MPC52xx_PSC_SICR_SIM_AC97		(0x3 << 24)
+#define MPC52xx_PSC_SICR_SIM_SIR		(0x8 << 24)
+#define MPC52xx_PSC_SICR_SIM_SIR_DCD		(0xc << 24)
+#define MPC52xx_PSC_SICR_SIM_MIR		(0x5 << 24)
+#define MPC52xx_PSC_SICR_SIM_FIR		(0x6 << 24)
+#define MPC52xx_PSC_SICR_SIM_CODEC_24		(0x7 << 24)
+#define MPC52xx_PSC_SICR_SIM_CODEC_32		(0xf << 24)
+#define MPC52xx_PSC_SICR_GENCLK			(1 << 23)
+#define MPC52xx_PSC_SICR_I2S			(1 << 22)
+#define MPC52xx_PSC_SICR_CLKPOL			(1 << 21)
+#define MPC52xx_PSC_SICR_SYNCPOL		(1 << 20)
+#define MPC52xx_PSC_SICR_CELLSLAVE		(1 << 19)
+#define MPC52xx_PSC_SICR_CELL2XCLK		(1 << 18)
+#define MPC52xx_PSC_SICR_ESAI			(1 << 17)
+#define MPC52xx_PSC_SICR_ENAC97			(1 << 16)
+#define MPC52xx_PSC_SICR_SPI			(1 << 15)
+#define MPC52xx_PSC_SICR_MSTR			(1 << 14)
+#define MPC52xx_PSC_SICR_CPOL			(1 << 13)
+#define MPC52xx_PSC_SICR_CPHA			(1 << 12)
+#define MPC52xx_PSC_SICR_USEEOF			(1 << 11)
+#define MPC52xx_PSC_SICR_DISABLEEOF		(1 << 10)
 
 /* Structure of the hardware registers */
 struct mpc52xx_psc {

From 79c28acb2b7d66ca48d23e1c8b5e9e043aa634f8 Mon Sep 17 00:00:00 2001
From: Grant Likely <grant.likely@secretlab.ca>
Date: Fri, 11 Jul 2008 16:17:57 -0600
Subject: [PATCH 0096/1171] of-bindings: Add binding documentation for SPI
 busses and devices

Add documentation about how to describe SPI busses in the device tree.

Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
Acked-by: Segher Boessenkool <segher@kernel.crashing.org>
---
 Documentation/powerpc/booting-without-of.txt | 57 ++++++++++++++++++++
 1 file changed, 57 insertions(+)

diff --git a/Documentation/powerpc/booting-without-of.txt b/Documentation/powerpc/booting-without-of.txt
index aee243a846a2..ee92fedada1a 100644
--- a/Documentation/powerpc/booting-without-of.txt
+++ b/Documentation/powerpc/booting-without-of.txt
@@ -59,6 +59,7 @@ Table of Contents
       p) Freescale Synchronous Serial Interface
 	  q) USB EHCI controllers
       r) MDIO on GPIOs
+      s) SPI busses
 
   VII - Marvell Discovery mv64[345]6x System Controller chips
     1) The /system-controller node
@@ -1881,6 +1882,62 @@ platforms are moved over to use the flattened-device-tree model.
 			 &qe_pio_c 6>;
 	};
 
+    s) SPI (Serial Peripheral Interface) busses
+
+    SPI busses can be described with a node for the SPI master device
+    and a set of child nodes for each SPI slave on the bus.  For this
+    discussion, it is assumed that the system's SPI controller is in
+    SPI master mode.  This binding does not describe SPI controllers
+    in slave mode.
+
+    The SPI master node requires the following properties:
+    - #address-cells  - number of cells required to define a chip select
+			address on the SPI bus.
+    - #size-cells     - should be zero.
+    - compatible      - name of SPI bus controller following generic names
+			recommended practice.
+    No other properties are required in the SPI bus node.  It is assumed
+    that a driver for an SPI bus device will understand that it is an SPI bus.
+    However, the binding does not attempt to define the specific method for
+    assigning chip select numbers.  Since SPI chip select configuration is
+    flexible and non-standardized, it is left out of this binding with the
+    assumption that board specific platform code will be used to manage
+    chip selects.  Individual drivers can define additional properties to
+    support describing the chip select layout.
+
+    SPI slave nodes must be children of the SPI master node and can
+    contain the following properties.
+    - reg             - (required) chip select address of device.
+    - compatible      - (required) name of SPI device following generic names
+			recommended practice
+    - spi-max-frequency - (required) Maximum SPI clocking speed of device in Hz
+    - spi-cpol        - (optional) Empty property indicating device requires
+			inverse clock polarity (CPOL) mode
+    - spi-cpha        - (optional) Empty property indicating device requires
+			shifted clock phase (CPHA) mode
+
+    SPI example for an MPC5200 SPI bus:
+		spi@f00 {
+			#address-cells = <1>;
+			#size-cells = <0>;
+			compatible = "fsl,mpc5200b-spi","fsl,mpc5200-spi";
+			reg = <0xf00 0x20>;
+			interrupts = <2 13 0 2 14 0>;
+			interrupt-parent = <&mpc5200_pic>;
+
+			ethernet-switch@0 {
+				compatible = "micrel,ks8995m";
+				spi-max-frequency = <1000000>;
+				reg = <0>;
+			};
+
+			codec@1 {
+				compatible = "ti,tlv320aic26";
+				spi-max-frequency = <100000>;
+				reg = <1>;
+			};
+		};
+
 VII - Marvell Discovery mv64[345]6x System Controller chips
 ===========================================================
 

From 04e1e0cccade330ab3715ce59234f7e3b087e246 Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Tue, 22 Jul 2008 13:04:18 +0000
Subject: [PATCH 0097/1171] [CIFS] Fix compiler warning on 64-bit

Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/asn1.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/cifs/asn1.c b/fs/cifs/asn1.c
index f58e41d3ba48..42765465701c 100644
--- a/fs/cifs/asn1.c
+++ b/fs/cifs/asn1.c
@@ -400,7 +400,7 @@ asn1_oid_decode(struct asn1_ctx *ctx,
 	size = eoc - ctx->pointer + 1;
 
 	/* first subid actually encodes first two subids */
-	if (size < 2 || size > ULONG_MAX/sizeof(unsigned long))
+	if (size < 2 || size > UINT_MAX/sizeof(unsigned long))
 		return 0;
 
 	*oid = kmalloc(size * sizeof(unsigned long), GFP_ATOMIC);

From e4268aad42e9f37d01925022830b16bab3d0d5af Mon Sep 17 00:00:00 2001
From: Alex Chiang <achiang@hp.com>
Date: Thu, 17 Jul 2008 11:13:32 -0600
Subject: [PATCH 0098/1171] PCI hotplug: fix error path in pci_slot's
 register_slot

Juha Leppnen noticed that an error path in register_slot() wasn't
returning appropriately, leading to a condition where we might access a
kfree'ed pointer, so let's fix that.

Additionally, fix up the copyright information in the file while
we're in there.

Signed-off-by: Alex Chiang <achiang@hp.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/acpi/pci_slot.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/acpi/pci_slot.c b/drivers/acpi/pci_slot.c
index b9ab030a52d5..dd376f7ad090 100644
--- a/drivers/acpi/pci_slot.c
+++ b/drivers/acpi/pci_slot.c
@@ -6,8 +6,8 @@
  *  Thanks to Kenji Kaneshige <kaneshige.kenji@jp.fujitsu.com> for code
  *  review and fixes.
  *
- *  Copyright (C) 2007 Alex Chiang <achiang@hp.com>
- *  Copyright (C) 2007 Hewlett-Packard Development Company, L.P.
+ *  Copyright (C) 2007-2008 Hewlett-Packard Development Company, L.P.
+ *  	Alex Chiang <achiang@hp.com>
  *
  *  This program is free software; you can redistribute it and/or modify it
  *  under the terms and conditions of the GNU General Public License,
@@ -158,6 +158,7 @@ register_slot(acpi_handle handle, u32 lvl, void *context, void **rv)
 	if (IS_ERR(pci_slot)) {
 		err("pci_create_slot returned %ld\n", PTR_ERR(pci_slot));
 		kfree(slot);
+		return AE_OK;
 	}
 
 	slot->root_handle = parent_context->root_handle;

From f42e86d95fa53d3a62b2795515da18b4f41b0480 Mon Sep 17 00:00:00 2001
From: Yong Wang <yong.y.wang@linux.intel.com>
Date: Tue, 22 Jul 2008 14:14:18 -0700
Subject: [PATCH 0099/1171] PCI/DMAR: don't assume presence of RMRRs

RMRRs do not necessarily have to be present on all VT-d capable platforms.
The printk is just informational and does not need to be followed by an error
return.

Signed-off-by: Yong Y Wang <yong.y.wang@intel.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: mark gross <mgross@linux.intel.com>
Cc: Keshavamurthy, Anil S <anil.s.keshavamurthy@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/dmar.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/pci/dmar.c b/drivers/pci/dmar.c
index f941f609dbf3..8bf86ae2333f 100644
--- a/drivers/pci/dmar.c
+++ b/drivers/pci/dmar.c
@@ -317,10 +317,8 @@ int __init dmar_table_init(void)
 		return -ENODEV;
 	}
 
-	if (list_empty(&dmar_rmrr_units)) {
+	if (list_empty(&dmar_rmrr_units))
 		printk(KERN_INFO PREFIX "No RMRR found\n");
-		return -ENODEV;
-	}
 
 	return 0;
 }

From dd5bdff83b19d9174126e0398b47117c3a80e22d Mon Sep 17 00:00:00 2001
From: Or Gerlitz <ogerlitz@voltaire.com>
Date: Tue, 22 Jul 2008 14:14:22 -0700
Subject: [PATCH 0100/1171] RDMA/cma: Add RDMA_CM_EVENT_ADDR_CHANGE event

Add an RDMA_CM_EVENT_ADDR_CHANGE event can be used by rdma-cm
consumers that wish to have their RDMA sessions always use the same
links (eg <hca/port>) as the IP stack does.  In the current code, this
does not happen when bonding is used and fail-over happened but the IB
link used by an already existing session is operating fine.

Use the netevent notification for sensing that a change has happened
in the IP stack, then scan the rdma-cm ID list to see if there is an
ID that is "misaligned" with respect to the IP stack, and deliver
RDMA_CM_EVENT_ADDR_CHANGE for this ID.  The consumer can act on the
event or just ignore it.

Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/core/cma.c | 92 +++++++++++++++++++++++++++++++++++
 include/rdma/rdma_cm.h        |  3 +-
 2 files changed, 94 insertions(+), 1 deletion(-)

diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c
index ae11d5cc74d0..79792c92e6fb 100644
--- a/drivers/infiniband/core/cma.c
+++ b/drivers/infiniband/core/cma.c
@@ -168,6 +168,12 @@ struct cma_work {
 	struct rdma_cm_event	event;
 };
 
+struct cma_ndev_work {
+	struct work_struct	work;
+	struct rdma_id_private	*id;
+	struct rdma_cm_event	event;
+};
+
 union cma_ip_addr {
 	struct in6_addr ip6;
 	struct {
@@ -1598,6 +1604,30 @@ out:
 	kfree(work);
 }
 
+static void cma_ndev_work_handler(struct work_struct *_work)
+{
+	struct cma_ndev_work *work = container_of(_work, struct cma_ndev_work, work);
+	struct rdma_id_private *id_priv = work->id;
+	int destroy = 0;
+
+	mutex_lock(&id_priv->handler_mutex);
+	if (id_priv->state == CMA_DESTROYING ||
+	    id_priv->state == CMA_DEVICE_REMOVAL)
+		goto out;
+
+	if (id_priv->id.event_handler(&id_priv->id, &work->event)) {
+		cma_exch(id_priv, CMA_DESTROYING);
+		destroy = 1;
+	}
+
+out:
+	mutex_unlock(&id_priv->handler_mutex);
+	cma_deref_id(id_priv);
+	if (destroy)
+		rdma_destroy_id(&id_priv->id);
+	kfree(work);
+}
+
 static int cma_resolve_ib_route(struct rdma_id_private *id_priv, int timeout_ms)
 {
 	struct rdma_route *route = &id_priv->id.route;
@@ -2723,6 +2753,65 @@ void rdma_leave_multicast(struct rdma_cm_id *id, struct sockaddr *addr)
 }
 EXPORT_SYMBOL(rdma_leave_multicast);
 
+static int cma_netdev_change(struct net_device *ndev, struct rdma_id_private *id_priv)
+{
+	struct rdma_dev_addr *dev_addr;
+	struct cma_ndev_work *work;
+
+	dev_addr = &id_priv->id.route.addr.dev_addr;
+
+	if ((dev_addr->src_dev == ndev) &&
+	    memcmp(dev_addr->src_dev_addr, ndev->dev_addr, ndev->addr_len)) {
+		printk(KERN_INFO "RDMA CM addr change for ndev %s used by id %p\n",
+		       ndev->name, &id_priv->id);
+		work = kzalloc(sizeof *work, GFP_KERNEL);
+		if (!work)
+			return -ENOMEM;
+
+		INIT_WORK(&work->work, cma_ndev_work_handler);
+		work->id = id_priv;
+		work->event.event = RDMA_CM_EVENT_ADDR_CHANGE;
+		atomic_inc(&id_priv->refcount);
+		queue_work(cma_wq, &work->work);
+	}
+
+	return 0;
+}
+
+static int cma_netdev_callback(struct notifier_block *self, unsigned long event,
+			       void *ctx)
+{
+	struct net_device *ndev = (struct net_device *)ctx;
+	struct cma_device *cma_dev;
+	struct rdma_id_private *id_priv;
+	int ret = NOTIFY_DONE;
+
+	if (dev_net(ndev) != &init_net)
+		return NOTIFY_DONE;
+
+	if (event != NETDEV_BONDING_FAILOVER)
+		return NOTIFY_DONE;
+
+	if (!(ndev->flags & IFF_MASTER) || !(ndev->priv_flags & IFF_BONDING))
+		return NOTIFY_DONE;
+
+	mutex_lock(&lock);
+	list_for_each_entry(cma_dev, &dev_list, list)
+		list_for_each_entry(id_priv, &cma_dev->id_list, list) {
+			ret = cma_netdev_change(ndev, id_priv);
+			if (ret)
+				goto out;
+		}
+
+out:
+	mutex_unlock(&lock);
+	return ret;
+}
+
+static struct notifier_block cma_nb = {
+	.notifier_call = cma_netdev_callback
+};
+
 static void cma_add_one(struct ib_device *device)
 {
 	struct cma_device *cma_dev;
@@ -2831,6 +2920,7 @@ static int cma_init(void)
 
 	ib_sa_register_client(&sa_client);
 	rdma_addr_register_client(&addr_client);
+	register_netdevice_notifier(&cma_nb);
 
 	ret = ib_register_client(&cma_client);
 	if (ret)
@@ -2838,6 +2928,7 @@ static int cma_init(void)
 	return 0;
 
 err:
+	unregister_netdevice_notifier(&cma_nb);
 	rdma_addr_unregister_client(&addr_client);
 	ib_sa_unregister_client(&sa_client);
 	destroy_workqueue(cma_wq);
@@ -2847,6 +2938,7 @@ err:
 static void cma_cleanup(void)
 {
 	ib_unregister_client(&cma_client);
+	unregister_netdevice_notifier(&cma_nb);
 	rdma_addr_unregister_client(&addr_client);
 	ib_sa_unregister_client(&sa_client);
 	destroy_workqueue(cma_wq);
diff --git a/include/rdma/rdma_cm.h b/include/rdma/rdma_cm.h
index 22bb2e7bab1a..001d606517ff 100644
--- a/include/rdma/rdma_cm.h
+++ b/include/rdma/rdma_cm.h
@@ -57,7 +57,8 @@ enum rdma_cm_event_type {
 	RDMA_CM_EVENT_DISCONNECTED,
 	RDMA_CM_EVENT_DEVICE_REMOVAL,
 	RDMA_CM_EVENT_MULTICAST_JOIN,
-	RDMA_CM_EVENT_MULTICAST_ERROR
+	RDMA_CM_EVENT_MULTICAST_ERROR,
+	RDMA_CM_EVENT_ADDR_CHANGE
 };
 
 enum rdma_port_space {

From 38ca83a588662f0af684ba2567dd910a564268ab Mon Sep 17 00:00:00 2001
From: Amir Vadai <amirv@mellanox.co.il>
Date: Tue, 22 Jul 2008 14:14:23 -0700
Subject: [PATCH 0101/1171] RDMA/cma: Add RDMA_CM_EVENT_TIMEWAIT_EXIT event

Consumers that want to re-use their QPs in new connections need to
know when the QP has exited the timewait state.  Report the timewait
event through the rdma_cm.

Signed-off-by: Amir Vadai <amirv@mellanox.co.il>
Acked-by: Sean Hefty <sean.hefty@intel.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/core/cma.c | 7 ++++++-
 include/rdma/rdma_cm.h        | 3 ++-
 2 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c
index 79792c92e6fb..e980ff3335db 100644
--- a/drivers/infiniband/core/cma.c
+++ b/drivers/infiniband/core/cma.c
@@ -920,7 +920,10 @@ static int cma_ib_handler(struct ib_cm_id *cm_id, struct ib_cm_event *ib_event)
 	struct rdma_cm_event event;
 	int ret = 0;
 
-	if (cma_disable_callback(id_priv, CMA_CONNECT))
+	if ((ib_event->event != IB_CM_TIMEWAIT_EXIT &&
+		cma_disable_callback(id_priv, CMA_CONNECT)) ||
+	    (ib_event->event == IB_CM_TIMEWAIT_EXIT &&
+		cma_disable_callback(id_priv, CMA_DISCONNECT)))
 		return 0;
 
 	memset(&event, 0, sizeof event);
@@ -956,6 +959,8 @@ static int cma_ib_handler(struct ib_cm_id *cm_id, struct ib_cm_event *ib_event)
 		event.event = RDMA_CM_EVENT_DISCONNECTED;
 		break;
 	case IB_CM_TIMEWAIT_EXIT:
+		event.event = RDMA_CM_EVENT_TIMEWAIT_EXIT;
+		break;
 	case IB_CM_MRA_RECEIVED:
 		/* ignore event */
 		goto out;
diff --git a/include/rdma/rdma_cm.h b/include/rdma/rdma_cm.h
index 001d606517ff..df7faf09d66f 100644
--- a/include/rdma/rdma_cm.h
+++ b/include/rdma/rdma_cm.h
@@ -58,7 +58,8 @@ enum rdma_cm_event_type {
 	RDMA_CM_EVENT_DEVICE_REMOVAL,
 	RDMA_CM_EVENT_MULTICAST_JOIN,
 	RDMA_CM_EVENT_MULTICAST_ERROR,
-	RDMA_CM_EVENT_ADDR_CHANGE
+	RDMA_CM_EVENT_ADDR_CHANGE,
+	RDMA_CM_EVENT_TIMEWAIT_EXIT
 };
 
 enum rdma_port_space {

From 2f5de1512884da8c74bec2c76e8f114b972ab4be Mon Sep 17 00:00:00 2001
From: Or Gerlitz <ogerlitz@voltaire.com>
Date: Tue, 22 Jul 2008 14:16:21 -0700
Subject: [PATCH 0102/1171] IB/iser: Add support for RDMA_CM_EVENT_ADDR_CHANGE
 event

Enhance iser to act upon notification on network stack changes that
make its RDMA connection unaligned with the link used by the stack for
the <src,dst> IPs used to establish the connection.

When RDMA_CM_EVENT_ADDR_CHANGE arrives, just disconnect the
connection, assuming that the user space iscsid daemon will reconnect,
and the new connection will be aligned with the IP stack.

Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/ulp/iser/iser_verbs.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/infiniband/ulp/iser/iser_verbs.c b/drivers/infiniband/ulp/iser/iser_verbs.c
index 3a917c1f796f..63462ecca147 100644
--- a/drivers/infiniband/ulp/iser/iser_verbs.c
+++ b/drivers/infiniband/ulp/iser/iser_verbs.c
@@ -483,6 +483,7 @@ static int iser_cma_handler(struct rdma_cm_id *cma_id, struct rdma_cm_event *eve
 		break;
 	case RDMA_CM_EVENT_DISCONNECTED:
 	case RDMA_CM_EVENT_DEVICE_REMOVAL:
+	case RDMA_CM_EVENT_ADDR_CHANGE:
 		iser_disconnected_handler(cma_id);
 		break;
 	default:

From 5b673b71c8ca0fbdb99dc1b1434cfb554212d6ff Mon Sep 17 00:00:00 2001
From: Joachim Fenkes <fenkes@de.ibm.com>
Date: Tue, 22 Jul 2008 14:18:07 -0700
Subject: [PATCH 0103/1171] IB/ehca: Filter PATH_MIG events if QP was never
 armed

Certain firmware versions sometimes cause spurious PATH_MIG events to
occur during QP creation.  Filter these events by making sure PATH_MIG
events are only handed down when they actually make sense (i.e. when
the QP has been armed at least once).

Signed-off-by: Joachim Fenkes <fenkes@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/ehca/ehca_classes.h | 1 +
 drivers/infiniband/hw/ehca/ehca_irq.c     | 4 ++++
 drivers/infiniband/hw/ehca/ehca_qp.c      | 2 ++
 3 files changed, 7 insertions(+)

diff --git a/drivers/infiniband/hw/ehca/ehca_classes.h b/drivers/infiniband/hw/ehca/ehca_classes.h
index 1e9e99a13933..0b0618edd645 100644
--- a/drivers/infiniband/hw/ehca/ehca_classes.h
+++ b/drivers/infiniband/hw/ehca/ehca_classes.h
@@ -194,6 +194,7 @@ struct ehca_qp {
 	u32 packet_count;
 	atomic_t nr_events; /* events seen */
 	wait_queue_head_t wait_completion;
+	int mig_armed;
 };
 
 #define IS_SRQ(qp) (qp->ext_type == EQPT_SRQ)
diff --git a/drivers/infiniband/hw/ehca/ehca_irq.c b/drivers/infiniband/hw/ehca/ehca_irq.c
index 0792d930c481..99642a6e17c4 100644
--- a/drivers/infiniband/hw/ehca/ehca_irq.c
+++ b/drivers/infiniband/hw/ehca/ehca_irq.c
@@ -178,6 +178,10 @@ static void dispatch_qp_event(struct ehca_shca *shca, struct ehca_qp *qp,
 {
 	struct ib_event event;
 
+	/* PATH_MIG without the QP ever having been armed is false alarm */
+	if (event_type == IB_EVENT_PATH_MIG && !qp->mig_armed)
+		return;
+
 	event.device = &shca->ib_device;
 	event.event = event_type;
 
diff --git a/drivers/infiniband/hw/ehca/ehca_qp.c b/drivers/infiniband/hw/ehca/ehca_qp.c
index 3f59587338ea..ea13efddf175 100644
--- a/drivers/infiniband/hw/ehca/ehca_qp.c
+++ b/drivers/infiniband/hw/ehca/ehca_qp.c
@@ -1460,6 +1460,8 @@ static int internal_modify_qp(struct ib_qp *ibqp,
 			goto modify_qp_exit2;
 		}
 		mqpcb->path_migration_state = attr->path_mig_state + 1;
+		if (attr->path_mig_state == IB_MIG_REARM)
+			my_qp->mig_armed = 1;
 		update_mask |=
 			EHCA_BMASK_SET(MQPCB_MASK_PATH_MIGRATION_STATE, 1);
 	}

From 593e4d4a05c8263a6dbd5452c21d47c5bdadd40c Mon Sep 17 00:00:00 2001
From: Joachim Fenkes <fenkes@de.ibm.com>
Date: Tue, 22 Jul 2008 14:18:08 -0700
Subject: [PATCH 0104/1171] IB/ehca: Use default value for Local CA ACK Delay
 if FW returns 0

Some firmware versions report a Local CA ACK Delay of 0.  In that
case, return a more sensible default value of 12 (-> 16 msec) instead.

Signed-off-by: Joachim Fenkes <fenkes@de.ibm.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/ehca/ehca_hca.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/infiniband/hw/ehca/ehca_hca.c b/drivers/infiniband/hw/ehca/ehca_hca.c
index bc3b37d2070f..46288220cfbb 100644
--- a/drivers/infiniband/hw/ehca/ehca_hca.c
+++ b/drivers/infiniband/hw/ehca/ehca_hca.c
@@ -114,7 +114,9 @@ int ehca_query_device(struct ib_device *ibdev, struct ib_device_attr *props)
 	}
 
 	props->max_pkeys           = 16;
-	props->local_ca_ack_delay  = min_t(u8, rblock->local_ca_ack_delay, 255);
+	/* Some FW versions say 0 here; insert sensible value in that case */
+	props->local_ca_ack_delay  = rblock->local_ca_ack_delay ?
+		min_t(u8, rblock->local_ca_ack_delay, 255) : 12;
 	props->max_raw_ipv6_qp     = limit_uint(rblock->max_raw_ipv6_qp);
 	props->max_raw_ethy_qp     = limit_uint(rblock->max_raw_ethy_qp);
 	props->max_mcast_grp       = limit_uint(rblock->max_mcast_grp);

From 1a867c33bb65f2921351a9bdd98548bb96f0ff8c Mon Sep 17 00:00:00 2001
From: Julia Lawall <julia@diku.dk>
Date: Tue, 22 Jul 2008 14:18:10 -0700
Subject: [PATCH 0105/1171] IB/ehca: Release mutex in error path of
 alloc_small_queue_page()

The pd->lock mutex is released on a successful return, so it should be
released on an error return as well.

The semantic patch that makes this change is as follows:
(http://www.emn.fr/x-info/coccinelle/)

// <smpl>
@@
expression l;
@@

mutex_lock(l);
... when != mutex_unlock(l)
    when any
    when strict
(
if (...) { ... when != mutex_unlock(l)
+   mutex_unlock(l);
    return ...;
}
|
mutex_unlock(l);
)
// </smpl>

Signed-off-by: Julia Lawall <julia@diku.dk>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/ehca/ipz_pt_fn.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/infiniband/hw/ehca/ipz_pt_fn.c b/drivers/infiniband/hw/ehca/ipz_pt_fn.c
index 661f8db62706..c3a328465431 100644
--- a/drivers/infiniband/hw/ehca/ipz_pt_fn.c
+++ b/drivers/infiniband/hw/ehca/ipz_pt_fn.c
@@ -163,6 +163,7 @@ static int alloc_small_queue_page(struct ipz_queue *queue, struct ehca_pd *pd)
 
 out:
 	ehca_err(pd->ib_pd.device, "failed to allocate small queue page");
+	mutex_unlock(&pd->lock);
 	return 0;
 }
 

From 64b784b583061ebfe1d484dd1fdc5a26c6d4293f Mon Sep 17 00:00:00 2001
From: Ralph Campbell <ralph.campbell@qlogic.com>
Date: Tue, 22 Jul 2008 14:18:33 -0700
Subject: [PATCH 0106/1171] IB/sa_query: Check if sm_ah is NULL in
 ib_sa_remove_one()

If update_sm_ah() fails, it leaves the port's sm_ah as NULL.  Then if
the device or module is removed, ib_sa_remove_one() will dereference a
NULL pointer when it calls kref_put().  Fix this by testing if sm_ah
is NULL before dropping the reference.

Signed-off-by: Ralph Campbell <ralph.campbell@qlogic.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/core/sa_query.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/infiniband/core/sa_query.c b/drivers/infiniband/core/sa_query.c
index 1341de793e51..7863a50d56f2 100644
--- a/drivers/infiniband/core/sa_query.c
+++ b/drivers/infiniband/core/sa_query.c
@@ -1064,7 +1064,8 @@ static void ib_sa_remove_one(struct ib_device *device)
 
 	for (i = 0; i <= sa_dev->end_port - sa_dev->start_port; ++i) {
 		ib_unregister_mad_agent(sa_dev->port[i].agent);
-		kref_put(&sa_dev->port[i].sm_ah->ref, free_sm_ah);
+		if (sa_dev->port[i].sm_ah)
+			kref_put(&sa_dev->port[i].sm_ah->ref, free_sm_ah);
 	}
 
 	kfree(sa_dev);

From 01b3fc8b15432f7931e40fe099839e1559fb0e09 Mon Sep 17 00:00:00 2001
From: Or Gerlitz <ogerlitz@voltaire.com>
Date: Tue, 22 Jul 2008 14:18:34 -0700
Subject: [PATCH 0107/1171] IPoIB: Include err code in trace message for
 ib_sa_path_rec_get() failures

Print the return code of ib_sa_path_rec_get() if it fails to help
debug errors.

Signed-off-by: Or Gerlitz <ogerlitz@voltaire.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/ulp/ipoib/ipoib_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c
index 8be9ea0436e6..f51201b17bfd 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_main.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c
@@ -548,7 +548,7 @@ static int path_rec_start(struct net_device *dev,
 				   path_rec_completion,
 				   path, &path->query);
 	if (path->query_id < 0) {
-		ipoib_warn(priv, "ib_sa_path_rec_get failed\n");
+		ipoib_warn(priv, "ib_sa_path_rec_get failed: %d\n", path->query_id);
 		path->query = NULL;
 		return path->query_id;
 	}

From 1ca8d15619f725e223c19137350b0336b9196193 Mon Sep 17 00:00:00 2001
From: Dotan Barak <dotanba@gmail.com>
Date: Tue, 22 Jul 2008 14:18:34 -0700
Subject: [PATCH 0108/1171] RDMA/iwcm: Remove IB_ACCESS_LOCAL_WRITE from remote
 QP attributes

Remove IB_ACCESS_LOCAL_WRITE from qp.qp_access_flags because this
attribute is only used to set remote permissions.

Signed-off-by: Dotan Barak <dotanba@gmail.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/core/iwcm.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/infiniband/core/iwcm.c b/drivers/infiniband/core/iwcm.c
index 81c9195b512a..8f9509e1ebf7 100644
--- a/drivers/infiniband/core/iwcm.c
+++ b/drivers/infiniband/core/iwcm.c
@@ -942,8 +942,7 @@ static int iwcm_init_qp_init_attr(struct iwcm_id_private *cm_id_priv,
 	case IW_CM_STATE_CONN_RECV:
 	case IW_CM_STATE_ESTABLISHED:
 		*qp_attr_mask = IB_QP_STATE | IB_QP_ACCESS_FLAGS;
-		qp_attr->qp_access_flags = IB_ACCESS_LOCAL_WRITE |
-					   IB_ACCESS_REMOTE_WRITE|
+		qp_attr->qp_access_flags = IB_ACCESS_REMOTE_WRITE|
 					   IB_ACCESS_REMOTE_READ;
 		ret = 0;
 		break;

From 51f5f0ee22b98980f7816d42647467cd5f4b3b45 Mon Sep 17 00:00:00 2001
From: Jack Morgenstein <jackm@dev.mellanox.co.il>
Date: Tue, 22 Jul 2008 14:19:37 -0700
Subject: [PATCH 0109/1171] mlx4_core: Add module parameter to enable QoS
 support

Add a module parameter "enable_qos" to mlx4_core.  If this param is
set, enable support for QoS in the INIT_HCA command.  By default, the
parameter is set to 0 (disabled).

Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/net/mlx4/fw.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/drivers/net/mlx4/fw.c b/drivers/net/mlx4/fw.c
index 2b5006b9be67..0851ebdddfd4 100644
--- a/drivers/net/mlx4/fw.c
+++ b/drivers/net/mlx4/fw.c
@@ -46,6 +46,10 @@ enum {
 extern void __buggy_use_of_MLX4_GET(void);
 extern void __buggy_use_of_MLX4_PUT(void);
 
+static int enable_qos;
+module_param(enable_qos, bool, 0444);
+MODULE_PARM_DESC(enable_qos, "Enable Quality of Service support in the HCA (default: off)");
+
 #define MLX4_GET(dest, source, offset)				      \
 	do {							      \
 		void *__p = (char *) (source) + (offset);	      \
@@ -737,6 +741,10 @@ int mlx4_INIT_HCA(struct mlx4_dev *dev, struct mlx4_init_hca_param *param)
 	if (dev->caps.flags & MLX4_DEV_CAP_FLAG_IPOIB_CSUM)
 		*(inbox + INIT_HCA_FLAGS_OFFSET / 4) |= cpu_to_be32(1 << 3);
 
+	/* Enable QoS support if module parameter set */
+	if (enable_qos)
+		*(inbox + INIT_HCA_FLAGS_OFFSET / 4) |= cpu_to_be32(1 << 2);
+
 	/* QPC/EEC/CQC/EQC/RDMARC attributes */
 
 	MLX4_PUT(inbox, param->qpc_base,      INIT_HCA_QPC_BASE_OFFSET);

From 47b374752aed1c029f995473c7c463ee3ae5fbaa Mon Sep 17 00:00:00 2001
From: Roland Dreier <rolandd@cisco.com>
Date: Tue, 22 Jul 2008 14:19:39 -0700
Subject: [PATCH 0110/1171] IB/mlx4: Rename struct mlx4_lso_seg to
 mlx4_wqe_lso_seg

Make the struct name consistent with other WQE segment struct types
defined in <linux/mlx4/qp.h>.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/mlx4/qp.c | 2 +-
 include/linux/mlx4/qp.h         | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
index 89eb6cbe592e..bda0859a5ac5 100644
--- a/drivers/infiniband/hw/mlx4/qp.c
+++ b/drivers/infiniband/hw/mlx4/qp.c
@@ -1395,7 +1395,7 @@ static void __set_data_seg(struct mlx4_wqe_data_seg *dseg, struct ib_sge *sg)
 	dseg->addr       = cpu_to_be64(sg->addr);
 }
 
-static int build_lso_seg(struct mlx4_lso_seg *wqe, struct ib_send_wr *wr,
+static int build_lso_seg(struct mlx4_wqe_lso_seg *wqe, struct ib_send_wr *wr,
 			 struct mlx4_ib_qp *qp, unsigned *lso_seg_len)
 {
 	unsigned halign = ALIGN(sizeof *wqe + wr->wr.ud.hlen, 16);
diff --git a/include/linux/mlx4/qp.h b/include/linux/mlx4/qp.h
index 7f128b266faa..f02e9ed36cfa 100644
--- a/include/linux/mlx4/qp.h
+++ b/include/linux/mlx4/qp.h
@@ -219,7 +219,7 @@ struct mlx4_wqe_datagram_seg {
 	__be32			reservd[2];
 };
 
-struct mlx4_lso_seg {
+struct mlx4_wqe_lso_seg {
 	__be32			mss_hdr_size;
 	__be32			header[0];
 };

From 899698dad72340b562478b8b770317f2f0fe0c09 Mon Sep 17 00:00:00 2001
From: Jack Morgenstein <jackm@dev.mellanox.co.il>
Date: Tue, 22 Jul 2008 14:19:39 -0700
Subject: [PATCH 0111/1171] mlx4_code: Add missing FW status return code

Add ICM_ERROR firmware status code.  In mapping to errnos, -ENFILE
seems closest.

This is in preparation for providing more detailed log info using
mlx4_err() in low-level driver when a non-zero status is returned.

Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/net/mlx4/cmd.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/drivers/net/mlx4/cmd.c b/drivers/net/mlx4/cmd.c
index 70dff94a8bc6..04d5bc69a6f8 100644
--- a/drivers/net/mlx4/cmd.c
+++ b/drivers/net/mlx4/cmd.c
@@ -67,6 +67,8 @@ enum {
 	CMD_STAT_BAD_INDEX	= 0x0a,
 	/* FW image corrupted: */
 	CMD_STAT_BAD_NVMEM	= 0x0b,
+	/* Error in ICM mapping (e.g. not enough auxiliary ICM pages to execute command): */
+	CMD_STAT_ICM_ERROR	= 0x0c,
 	/* Attempt to modify a QP/EE which is not in the presumed state: */
 	CMD_STAT_BAD_QP_STATE   = 0x10,
 	/* Bad segment parameters (Address/Size): */
@@ -119,6 +121,7 @@ static int mlx4_status_to_errno(u8 status)
 		[CMD_STAT_BAD_RES_STATE]  = -EBADF,
 		[CMD_STAT_BAD_INDEX]	  = -EBADF,
 		[CMD_STAT_BAD_NVMEM]	  = -EFAULT,
+		[CMD_STAT_ICM_ERROR]	  = -ENFILE,
 		[CMD_STAT_BAD_QP_STATE]   = -EINVAL,
 		[CMD_STAT_BAD_SEG_PARAM]  = -EFAULT,
 		[CMD_STAT_REG_BOUND]	  = -EBUSY,

From e4044cfc493338cd09870bd45dc646336bb66e9f Mon Sep 17 00:00:00 2001
From: Roland Dreier <rolandd@cisco.com>
Date: Tue, 22 Jul 2008 14:19:40 -0700
Subject: [PATCH 0112/1171] mlx4_core: Keep free count for MTT buddy allocator

MTT entries are allocated with a buddy allocator, which just keeps
bitmaps for each level of the buddy table.  However, all free space
starts out at the highest order, and small allocations start scanning
from the lowest order.  When the lowest order tables have no free
space, this can lead to scanning potentially millions of bits before
finding a free entry at a higher order.

We can avoid this by just keeping a count of how many free entries
each order has, and skipping the bitmap scan when an order is
completely empty.  This provides a nice performance boost for a
negligible increase in memory usage.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/net/mlx4/mlx4.h |  1 +
 drivers/net/mlx4/mr.c   | 28 +++++++++++++++++++---------
 2 files changed, 20 insertions(+), 9 deletions(-)

diff --git a/drivers/net/mlx4/mlx4.h b/drivers/net/mlx4/mlx4.h
index a4023c2dd050..78038499cff5 100644
--- a/drivers/net/mlx4/mlx4.h
+++ b/drivers/net/mlx4/mlx4.h
@@ -118,6 +118,7 @@ struct mlx4_bitmap {
 
 struct mlx4_buddy {
 	unsigned long	      **bits;
+	unsigned int	       *num_free;
 	int			max_order;
 	spinlock_t		lock;
 };
diff --git a/drivers/net/mlx4/mr.c b/drivers/net/mlx4/mr.c
index 03a9abcce524..b3ea93b98689 100644
--- a/drivers/net/mlx4/mr.c
+++ b/drivers/net/mlx4/mr.c
@@ -79,23 +79,26 @@ static u32 mlx4_buddy_alloc(struct mlx4_buddy *buddy, int order)
 
 	spin_lock(&buddy->lock);
 
-	for (o = order; o <= buddy->max_order; ++o) {
-		m = 1 << (buddy->max_order - o);
-		seg = find_first_bit(buddy->bits[o], m);
-		if (seg < m)
-			goto found;
-	}
+	for (o = order; o <= buddy->max_order; ++o)
+		if (buddy->num_free[o]) {
+			m = 1 << (buddy->max_order - o);
+			seg = find_first_bit(buddy->bits[o], m);
+			if (seg < m)
+				goto found;
+		}
 
 	spin_unlock(&buddy->lock);
 	return -1;
 
  found:
 	clear_bit(seg, buddy->bits[o]);
+	--buddy->num_free[o];
 
 	while (o > order) {
 		--o;
 		seg <<= 1;
 		set_bit(seg ^ 1, buddy->bits[o]);
+		++buddy->num_free[o];
 	}
 
 	spin_unlock(&buddy->lock);
@@ -113,11 +116,13 @@ static void mlx4_buddy_free(struct mlx4_buddy *buddy, u32 seg, int order)
 
 	while (test_bit(seg ^ 1, buddy->bits[order])) {
 		clear_bit(seg ^ 1, buddy->bits[order]);
+		--buddy->num_free[order];
 		seg >>= 1;
 		++order;
 	}
 
 	set_bit(seg, buddy->bits[order]);
+	++buddy->num_free[order];
 
 	spin_unlock(&buddy->lock);
 }
@@ -131,7 +136,9 @@ static int mlx4_buddy_init(struct mlx4_buddy *buddy, int max_order)
 
 	buddy->bits = kzalloc((buddy->max_order + 1) * sizeof (long *),
 			      GFP_KERNEL);
-	if (!buddy->bits)
+	buddy->num_free = kzalloc((buddy->max_order + 1) * sizeof (int *),
+				  GFP_KERNEL);
+	if (!buddy->bits || !buddy->num_free)
 		goto err_out;
 
 	for (i = 0; i <= buddy->max_order; ++i) {
@@ -143,6 +150,7 @@ static int mlx4_buddy_init(struct mlx4_buddy *buddy, int max_order)
 	}
 
 	set_bit(0, buddy->bits[buddy->max_order]);
+	buddy->num_free[buddy->max_order] = 1;
 
 	return 0;
 
@@ -150,9 +158,10 @@ err_out_free:
 	for (i = 0; i <= buddy->max_order; ++i)
 		kfree(buddy->bits[i]);
 
-	kfree(buddy->bits);
-
 err_out:
+	kfree(buddy->bits);
+	kfree(buddy->num_free);
+
 	return -ENOMEM;
 }
 
@@ -164,6 +173,7 @@ static void mlx4_buddy_cleanup(struct mlx4_buddy *buddy)
 		kfree(buddy->bits[i]);
 
 	kfree(buddy->bits);
+	kfree(buddy->num_free);
 }
 
 static u32 mlx4_alloc_mtt_range(struct mlx4_dev *dev, int order)

From e8bb4beb2b1f90d499134f2849727ed04c3bedc4 Mon Sep 17 00:00:00 2001
From: Roland Dreier <rolandd@cisco.com>
Date: Tue, 22 Jul 2008 14:20:05 -0700
Subject: [PATCH 0113/1171] IB/mthca: Keep free count for MTT buddy allocator

MTT entries are allocated with a buddy allocator, which just keeps
bitmaps for each level of the buddy table.  However, all free space
starts out at the highest order, and small allocations start scanning
from the lowest order.  When the lowest order tables have no free
space, this can lead to scanning potentially millions of bits before
finding a free entry at a higher order.

We can avoid this by just keeping a count of how many free entries
each order has, and skipping the bitmap scan when an order is
completely empty.  This provides a nice performance boost for a
negligible increase in memory usage.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/mthca/mthca_dev.h |  1 +
 drivers/infiniband/hw/mthca/mthca_mr.c  | 28 +++++++++++++++++--------
 2 files changed, 20 insertions(+), 9 deletions(-)

diff --git a/drivers/infiniband/hw/mthca/mthca_dev.h b/drivers/infiniband/hw/mthca/mthca_dev.h
index ee4d073c889f..252590116df5 100644
--- a/drivers/infiniband/hw/mthca/mthca_dev.h
+++ b/drivers/infiniband/hw/mthca/mthca_dev.h
@@ -202,6 +202,7 @@ struct mthca_pd_table {
 
 struct mthca_buddy {
 	unsigned long **bits;
+	int	       *num_free;
 	int             max_order;
 	spinlock_t      lock;
 };
diff --git a/drivers/infiniband/hw/mthca/mthca_mr.c b/drivers/infiniband/hw/mthca/mthca_mr.c
index 8489b1e81c0f..882e6b735915 100644
--- a/drivers/infiniband/hw/mthca/mthca_mr.c
+++ b/drivers/infiniband/hw/mthca/mthca_mr.c
@@ -89,23 +89,26 @@ static u32 mthca_buddy_alloc(struct mthca_buddy *buddy, int order)
 
 	spin_lock(&buddy->lock);
 
-	for (o = order; o <= buddy->max_order; ++o) {
-		m = 1 << (buddy->max_order - o);
-		seg = find_first_bit(buddy->bits[o], m);
-		if (seg < m)
-			goto found;
-	}
+	for (o = order; o <= buddy->max_order; ++o)
+		if (buddy->num_free[o]) {
+			m = 1 << (buddy->max_order - o);
+			seg = find_first_bit(buddy->bits[o], m);
+			if (seg < m)
+				goto found;
+		}
 
 	spin_unlock(&buddy->lock);
 	return -1;
 
  found:
 	clear_bit(seg, buddy->bits[o]);
+	--buddy->num_free[o];
 
 	while (o > order) {
 		--o;
 		seg <<= 1;
 		set_bit(seg ^ 1, buddy->bits[o]);
+		++buddy->num_free[o];
 	}
 
 	spin_unlock(&buddy->lock);
@@ -123,11 +126,13 @@ static void mthca_buddy_free(struct mthca_buddy *buddy, u32 seg, int order)
 
 	while (test_bit(seg ^ 1, buddy->bits[order])) {
 		clear_bit(seg ^ 1, buddy->bits[order]);
+		--buddy->num_free[order];
 		seg >>= 1;
 		++order;
 	}
 
 	set_bit(seg, buddy->bits[order]);
+	++buddy->num_free[order];
 
 	spin_unlock(&buddy->lock);
 }
@@ -141,7 +146,9 @@ static int mthca_buddy_init(struct mthca_buddy *buddy, int max_order)
 
 	buddy->bits = kzalloc((buddy->max_order + 1) * sizeof (long *),
 			      GFP_KERNEL);
-	if (!buddy->bits)
+	buddy->num_free = kzalloc((buddy->max_order + 1) * sizeof (int *),
+				  GFP_KERNEL);
+	if (!buddy->bits || !buddy->num_free)
 		goto err_out;
 
 	for (i = 0; i <= buddy->max_order; ++i) {
@@ -154,6 +161,7 @@ static int mthca_buddy_init(struct mthca_buddy *buddy, int max_order)
 	}
 
 	set_bit(0, buddy->bits[buddy->max_order]);
+	buddy->num_free[buddy->max_order] = 1;
 
 	return 0;
 
@@ -161,9 +169,10 @@ err_out_free:
 	for (i = 0; i <= buddy->max_order; ++i)
 		kfree(buddy->bits[i]);
 
-	kfree(buddy->bits);
-
 err_out:
+	kfree(buddy->bits);
+	kfree(buddy->num_free);
+
 	return -ENOMEM;
 }
 
@@ -175,6 +184,7 @@ static void mthca_buddy_cleanup(struct mthca_buddy *buddy)
 		kfree(buddy->bits[i]);
 
 	kfree(buddy->bits);
+	kfree(buddy->num_free);
 }
 
 static u32 mthca_alloc_mtt_range(struct mthca_dev *dev, int order,

From e5899e1b7d73e67de758a32174a859cc2586c0b9 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Sat, 19 Jul 2008 14:39:24 +0200
Subject: [PATCH 0114/1171] PCI PM: make more PCI PM core functionality
 available to drivers

Make more PCI PM core functionality available to drivers

* Export pci_pme_capable() so that it can be called directly by
  drivers (for example, tg3 needs that).

* Move the state choosing part of pci_prepare_to_sleep() to a
  separate function, pci_target_state(), that can be called directly
  by drivers (for example, tg3 needs that).

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/pci.c   | 34 ++++++++++++++++++++++++----------
 include/linux/pci.h |  2 ++
 2 files changed, 26 insertions(+), 10 deletions(-)

diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index d00f0e0d8453..e9c356236d27 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -1040,7 +1040,7 @@ int pci_set_pcie_reset_state(struct pci_dev *dev, enum pcie_reset_state state)
  * @dev: PCI device to handle.
  * @state: PCI state from which device will issue PME#.
  */
-static bool pci_pme_capable(struct pci_dev *dev, pci_power_t state)
+bool pci_pme_capable(struct pci_dev *dev, pci_power_t state)
 {
 	if (!dev->pm_cap)
 		return false;
@@ -1123,17 +1123,10 @@ int pci_enable_wake(struct pci_dev *dev, pci_power_t state, int enable)
 }
 
 /**
- * pci_prepare_to_sleep - prepare PCI device for system-wide transition into a sleep state
- * @dev: Device to handle.
- *
- * Choose the power state appropriate for the device depending on whether
- * it can wake up the system and/or is power manageable by the platform
- * (PCI_D3hot is the default) and put the device into that state.
  */
-int pci_prepare_to_sleep(struct pci_dev *dev)
+pci_power_t pci_target_state(struct pci_dev *dev)
 {
 	pci_power_t target_state = PCI_D3hot;
-	int error;
 
 	if (platform_pci_power_manageable(dev)) {
 		/*
@@ -1160,7 +1153,7 @@ int pci_prepare_to_sleep(struct pci_dev *dev)
 		 * to generate PME#.
 		 */
 		if (!dev->pm_cap)
-			return -EIO;
+			return PCI_POWER_ERROR;
 
 		if (dev->pme_support) {
 			while (target_state
@@ -1169,6 +1162,25 @@ int pci_prepare_to_sleep(struct pci_dev *dev)
 		}
 	}
 
+	return target_state;
+}
+
+/**
+ * pci_prepare_to_sleep - prepare PCI device for system-wide transition into a sleep state
+ * @dev: Device to handle.
+ *
+ * Choose the power state appropriate for the device depending on whether
+ * it can wake up the system and/or is power manageable by the platform
+ * (PCI_D3hot is the default) and put the device into that state.
+ */
+int pci_prepare_to_sleep(struct pci_dev *dev)
+{
+	pci_power_t target_state = pci_target_state(dev);
+	int error;
+
+	if (target_state == PCI_POWER_ERROR)
+		return -EIO;
+
 	pci_enable_wake(dev, target_state, true);
 
 	error = pci_set_power_state(dev, target_state);
@@ -1918,7 +1930,9 @@ EXPORT_SYMBOL(pci_select_bars);
 EXPORT_SYMBOL(pci_set_power_state);
 EXPORT_SYMBOL(pci_save_state);
 EXPORT_SYMBOL(pci_restore_state);
+EXPORT_SYMBOL(pci_pme_capable);
 EXPORT_SYMBOL(pci_enable_wake);
+EXPORT_SYMBOL(pci_target_state);
 EXPORT_SYMBOL(pci_prepare_to_sleep);
 EXPORT_SYMBOL(pci_back_from_sleep);
 EXPORT_SYMBOL_GPL(pci_set_pcie_reset_state);
diff --git a/include/linux/pci.h b/include/linux/pci.h
index a6a088e1a804..1d296d31abe0 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -638,7 +638,9 @@ int pci_save_state(struct pci_dev *dev);
 int pci_restore_state(struct pci_dev *dev);
 int pci_set_power_state(struct pci_dev *dev, pci_power_t state);
 pci_power_t pci_choose_state(struct pci_dev *dev, pm_message_t state);
+bool pci_pme_capable(struct pci_dev *dev, pci_power_t state);
 int pci_enable_wake(struct pci_dev *dev, pci_power_t state, int enable);
+pci_power_t pci_target_state(struct pci_dev *dev);
 int pci_prepare_to_sleep(struct pci_dev *dev);
 int pci_back_from_sleep(struct pci_dev *dev);
 

From f17a077e61b627e58db5926bc474cf308318dad9 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Tue, 22 Jul 2008 14:40:47 -0700
Subject: [PATCH 0115/1171] PCI: fixup sparse endianness warnings in proc.c

drivers/pci/proc.c:91:3: warning: cast from restricted __le16
drivers/pci/proc.c:100:3: warning: cast from restricted __le32
drivers/pci/proc.c:109:3: warning: cast from restricted __le16
drivers/pci/proc.c:161:40: warning: cast to restricted __le16
drivers/pci/proc.c:170:41: warning: cast to restricted __le32
drivers/pci/proc.c:179:40: warning: cast to restricted __le16

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Jesse Barnes <jbarnes@virtuousgeek.org>
---
 drivers/pci/proc.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/pci/proc.c b/drivers/pci/proc.c
index 4400dffbd93a..e1098c302c45 100644
--- a/drivers/pci/proc.c
+++ b/drivers/pci/proc.c
@@ -88,7 +88,7 @@ proc_bus_pci_read(struct file *file, char __user *buf, size_t nbytes, loff_t *pp
 	if ((pos & 3) && cnt > 2) {
 		unsigned short val;
 		pci_user_read_config_word(dev, pos, &val);
-		__put_user(cpu_to_le16(val), (unsigned short __user *) buf);
+		__put_user(cpu_to_le16(val), (__le16 __user *) buf);
 		buf += 2;
 		pos += 2;
 		cnt -= 2;
@@ -97,7 +97,7 @@ proc_bus_pci_read(struct file *file, char __user *buf, size_t nbytes, loff_t *pp
 	while (cnt >= 4) {
 		unsigned int val;
 		pci_user_read_config_dword(dev, pos, &val);
-		__put_user(cpu_to_le32(val), (unsigned int __user *) buf);
+		__put_user(cpu_to_le32(val), (__le32 __user *) buf);
 		buf += 4;
 		pos += 4;
 		cnt -= 4;
@@ -106,7 +106,7 @@ proc_bus_pci_read(struct file *file, char __user *buf, size_t nbytes, loff_t *pp
 	if (cnt >= 2) {
 		unsigned short val;
 		pci_user_read_config_word(dev, pos, &val);
-		__put_user(cpu_to_le16(val), (unsigned short __user *) buf);
+		__put_user(cpu_to_le16(val), (__le16 __user *) buf);
 		buf += 2;
 		pos += 2;
 		cnt -= 2;
@@ -156,8 +156,8 @@ proc_bus_pci_write(struct file *file, const char __user *buf, size_t nbytes, lof
 	}
 
 	if ((pos & 3) && cnt > 2) {
-		unsigned short val;
-		__get_user(val, (unsigned short __user *) buf);
+		__le16 val;
+		__get_user(val, (__le16 __user *) buf);
 		pci_user_write_config_word(dev, pos, le16_to_cpu(val));
 		buf += 2;
 		pos += 2;
@@ -165,8 +165,8 @@ proc_bus_pci_write(struct file *file, const char __user *buf, size_t nbytes, lof
 	}
 
 	while (cnt >= 4) {
-		unsigned int val;
-		__get_user(val, (unsigned int __user *) buf);
+		__le32 val;
+		__get_user(val, (__le32 __user *) buf);
 		pci_user_write_config_dword(dev, pos, le32_to_cpu(val));
 		buf += 4;
 		pos += 4;
@@ -174,8 +174,8 @@ proc_bus_pci_write(struct file *file, const char __user *buf, size_t nbytes, lof
 	}
 
 	if (cnt >= 2) {
-		unsigned short val;
-		__get_user(val, (unsigned short __user *) buf);
+		__le16 val;
+		__get_user(val, (__le16 __user *) buf);
 		pci_user_write_config_word(dev, pos, le16_to_cpu(val));
 		buf += 2;
 		pos += 2;

From 9bcab8405c98c34849c5795c717b7e6a3e2d3875 Mon Sep 17 00:00:00 2001
From: Milton Miller <miltonm@bga.com>
Date: Fri, 11 Jul 2008 09:03:09 +1000
Subject: [PATCH 0116/1171] powerpc/spufs: correct kcalloc usage

kcalloc is supposed to be called with the count as its first argument and
the element size as the second.

Signed-off-by: Milton Miller <miltonm@bga.com>
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
---
 arch/powerpc/platforms/cell/spufs/sputrace.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/arch/powerpc/platforms/cell/spufs/sputrace.c b/arch/powerpc/platforms/cell/spufs/sputrace.c
index 8c0e95766a62..92d20e993ede 100644
--- a/arch/powerpc/platforms/cell/spufs/sputrace.c
+++ b/arch/powerpc/platforms/cell/spufs/sputrace.c
@@ -196,8 +196,7 @@ static int __init sputrace_init(void)
 	struct proc_dir_entry *entry;
 	int i, error = -ENOMEM;
 
-	sputrace_log = kcalloc(sizeof(struct sputrace),
-				bufsize, GFP_KERNEL);
+	sputrace_log = kcalloc(bufsize, sizeof(struct sputrace), GFP_KERNEL);
 	if (!sputrace_log)
 		goto out;
 

From 8a6d2ea0cd121e3bfff4dbce5bc111874cf9e9d2 Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@vyatta.com>
Date: Tue, 22 Jul 2008 21:53:40 -0700
Subject: [PATCH 0117/1171] sky2: don't stop queue on shutdown

It is unnecessary, to stop queue and turn off carrier in shutdown
routine. With new netdev_queue this causes warnings.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/sky2.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c
index 711e4a8948e0..5257cf464f1a 100644
--- a/drivers/net/sky2.c
+++ b/drivers/net/sky2.c
@@ -1829,9 +1829,6 @@ static int sky2_down(struct net_device *dev)
 	if (netif_msg_ifdown(sky2))
 		printk(KERN_INFO PFX "%s: disabling interface\n", dev->name);
 
-	/* Stop more packets from being queued */
-	netif_stop_queue(dev);
-
 	/* Disable port IRQ */
 	imask = sky2_read32(hw, B0_IMSK);
 	imask &= ~portirq_msk[port];
@@ -1887,8 +1884,6 @@ static int sky2_down(struct net_device *dev)
 
 	sky2_phy_power_down(hw, port);
 
-	netif_carrier_off(dev);
-
 	/* turn off LED's */
 	sky2_write16(hw, B0_Y2LED, LED_STAT_OFF);
 

From deca05c3e81df4fcc38aa891eb8d8add14bce68b Mon Sep 17 00:00:00 2001
From: Greg Ungerer <gerg@goober.(none)>
Date: Wed, 4 Jun 2008 21:20:32 +1000
Subject: [PATCH 0118/1171] m68knommu: change to a configs directory for board
 configurations

Remove the old example m68knommu defconfig. Create a configs directory
for specific board configurations. Make the m5208evb the default.

Signed-off-by: Greg Ungerer <gerg@uclinux.org>
---
 arch/m68knommu/configs/m5208evb_defconfig | 610 ++++++++++++++++++++++
 1 file changed, 610 insertions(+)
 create mode 100644 arch/m68knommu/configs/m5208evb_defconfig

diff --git a/arch/m68knommu/configs/m5208evb_defconfig b/arch/m68knommu/configs/m5208evb_defconfig
new file mode 100644
index 000000000000..6fae33a05e2a
--- /dev/null
+++ b/arch/m68knommu/configs/m5208evb_defconfig
@@ -0,0 +1,610 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.26-rc1
+#
+CONFIG_M68K=y
+# CONFIG_MMU is not set
+# CONFIG_FPU is not set
+CONFIG_ZONE_DMA=y
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+# CONFIG_RWSEM_XCHGADD_ALGORITHM is not set
+# CONFIG_ARCH_HAS_ILOG2_U32 is not set
+# CONFIG_ARCH_HAS_ILOG2_U64 is not set
+CONFIG_GENERIC_FIND_NEXT_BIT=y
+CONFIG_GENERIC_HWEIGHT=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+CONFIG_GENERIC_TIME=y
+CONFIG_TIME_LOW_RES=y
+CONFIG_NO_IOPORT=y
+CONFIG_ARCH_SUPPORTS_AOUT=y
+CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
+
+#
+# General setup
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+CONFIG_LOCALVERSION=""
+CONFIG_LOCALVERSION_AUTO=y
+# CONFIG_SYSVIPC is not set
+# CONFIG_POSIX_MQUEUE is not set
+# CONFIG_BSD_PROCESS_ACCT is not set
+# CONFIG_TASKSTATS is not set
+# CONFIG_AUDIT is not set
+# CONFIG_IKCONFIG is not set
+CONFIG_LOG_BUF_SHIFT=14
+# CONFIG_CGROUPS is not set
+# CONFIG_GROUP_SCHED is not set
+# CONFIG_RELAY is not set
+# CONFIG_NAMESPACES is not set
+# CONFIG_BLK_DEV_INITRD is not set
+# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
+CONFIG_SYSCTL=y
+CONFIG_EMBEDDED=y
+# CONFIG_UID16 is not set
+# CONFIG_SYSCTL_SYSCALL is not set
+# CONFIG_KALLSYMS is not set
+# CONFIG_HOTPLUG is not set
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_ELF_CORE=y
+# CONFIG_COMPAT_BRK is not set
+CONFIG_BASE_FULL=y
+# CONFIG_FUTEX is not set
+# CONFIG_EPOLL is not set
+# CONFIG_SIGNALFD is not set
+# CONFIG_TIMERFD is not set
+# CONFIG_EVENTFD is not set
+# CONFIG_VM_EVENT_COUNTERS is not set
+CONFIG_SLAB=y
+# CONFIG_SLUB is not set
+# CONFIG_SLOB is not set
+# CONFIG_PROFILING is not set
+# CONFIG_MARKERS is not set
+# CONFIG_HAVE_OPROFILE is not set
+# CONFIG_HAVE_KPROBES is not set
+# CONFIG_HAVE_KRETPROBES is not set
+# CONFIG_HAVE_DMA_ATTRS is not set
+CONFIG_SLABINFO=y
+CONFIG_TINY_SHMEM=y
+CONFIG_BASE_SMALL=0
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+# CONFIG_MODULE_FORCE_UNLOAD is not set
+# CONFIG_MODVERSIONS is not set
+# CONFIG_MODULE_SRCVERSION_ALL is not set
+# CONFIG_KMOD is not set
+CONFIG_BLOCK=y
+# CONFIG_LBD is not set
+# CONFIG_LSF is not set
+# CONFIG_BLK_DEV_BSG is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+# CONFIG_IOSCHED_AS is not set
+# CONFIG_IOSCHED_DEADLINE is not set
+# CONFIG_IOSCHED_CFQ is not set
+# CONFIG_DEFAULT_AS is not set
+# CONFIG_DEFAULT_DEADLINE is not set
+# CONFIG_DEFAULT_CFQ is not set
+CONFIG_DEFAULT_NOOP=y
+CONFIG_DEFAULT_IOSCHED="noop"
+CONFIG_CLASSIC_RCU=y
+
+#
+# Processor type and features
+#
+# CONFIG_M68328 is not set
+# CONFIG_M68EZ328 is not set
+# CONFIG_M68VZ328 is not set
+# CONFIG_M68360 is not set
+# CONFIG_M5206 is not set
+# CONFIG_M5206e is not set
+CONFIG_M520x=y
+# CONFIG_M523x is not set
+# CONFIG_M5249 is not set
+# CONFIG_M5271 is not set
+# CONFIG_M5272 is not set
+# CONFIG_M5275 is not set
+# CONFIG_M528x is not set
+# CONFIG_M5307 is not set
+# CONFIG_M532x is not set
+# CONFIG_M5407 is not set
+CONFIG_COLDFIRE=y
+CONFIG_CLOCK_SET=y
+CONFIG_CLOCK_FREQ=166666666
+CONFIG_CLOCK_DIV=2
+
+#
+# Platform
+#
+CONFIG_M5208EVB=y
+CONFIG_FREESCALE=y
+# CONFIG_4KSTACKS is not set
+CONFIG_HZ=100
+
+#
+# RAM configuration
+#
+CONFIG_RAMBASE=0x40000000
+CONFIG_RAMSIZE=0x2000000
+CONFIG_VECTORBASE=0x40000000
+CONFIG_KERNELBASE=0x40020000
+# CONFIG_RAMAUTOBIT is not set
+# CONFIG_RAM8BIT is not set
+CONFIG_RAM16BIT=y
+# CONFIG_RAM32BIT is not set
+
+#
+# ROM configuration
+#
+# CONFIG_ROM is not set
+CONFIG_RAMKERNEL=y
+# CONFIG_ROMKERNEL is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+# CONFIG_SPARSEMEM_STATIC is not set
+# CONFIG_SPARSEMEM_VMEMMAP_ENABLE is not set
+CONFIG_PAGEFLAGS_EXTENDED=y
+CONFIG_SPLIT_PTLOCK_CPUS=4
+# CONFIG_RESOURCES_64BIT is not set
+CONFIG_ZONE_DMA_FLAG=1
+CONFIG_VIRT_TO_BUS=y
+CONFIG_ISA_DMA_API=y
+
+#
+# Bus options (PCI, PCMCIA, EISA, MCA, ISA)
+#
+# CONFIG_PCI is not set
+# CONFIG_ARCH_SUPPORTS_MSI is not set
+
+#
+# Executable file formats
+#
+CONFIG_BINFMT_FLAT=y
+# CONFIG_BINFMT_ZFLAT is not set
+# CONFIG_BINFMT_SHARED_FLAT is not set
+# CONFIG_BINFMT_AOUT is not set
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Power management options
+#
+# CONFIG_PM is not set
+
+#
+# Networking
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+CONFIG_PACKET=y
+# CONFIG_PACKET_MMAP is not set
+CONFIG_UNIX=y
+# CONFIG_NET_KEY is not set
+CONFIG_INET=y
+# CONFIG_IP_MULTICAST is not set
+# CONFIG_IP_ADVANCED_ROUTER is not set
+CONFIG_IP_FIB_HASH=y
+# CONFIG_IP_PNP is not set
+# CONFIG_NET_IPIP is not set
+# CONFIG_NET_IPGRE is not set
+# CONFIG_ARPD is not set
+# CONFIG_SYN_COOKIES is not set
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_XFRM_TUNNEL is not set
+# CONFIG_INET_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
+# CONFIG_INET_LRO is not set
+# CONFIG_INET_DIAG is not set
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_CUBIC=y
+CONFIG_DEFAULT_TCP_CONG="cubic"
+# CONFIG_TCP_MD5SIG is not set
+# CONFIG_IPV6 is not set
+# CONFIG_NETWORK_SECMARK is not set
+# CONFIG_NETFILTER is not set
+# CONFIG_IP_DCCP is not set
+# CONFIG_IP_SCTP is not set
+# CONFIG_TIPC is not set
+# CONFIG_ATM is not set
+# CONFIG_BRIDGE is not set
+# CONFIG_VLAN_8021Q is not set
+# CONFIG_DECNET is not set
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+# CONFIG_NET_SCHED is not set
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_CAN is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+# CONFIG_AF_RXRPC is not set
+
+#
+# Wireless
+#
+# CONFIG_CFG80211 is not set
+# CONFIG_WIRELESS_EXT is not set
+# CONFIG_MAC80211 is not set
+# CONFIG_IEEE80211 is not set
+# CONFIG_RFKILL is not set
+# CONFIG_NET_9P is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+# CONFIG_SYS_HYPERVISOR is not set
+# CONFIG_CONNECTOR is not set
+CONFIG_MTD=y
+# CONFIG_MTD_DEBUG is not set
+# CONFIG_MTD_CONCAT is not set
+CONFIG_MTD_PARTITIONS=y
+# CONFIG_MTD_REDBOOT_PARTS is not set
+# CONFIG_MTD_CMDLINE_PARTS is not set
+# CONFIG_MTD_AR7_PARTS is not set
+
+#
+# User Modules And Translation Layers
+#
+CONFIG_MTD_CHAR=y
+CONFIG_MTD_BLKDEVS=y
+CONFIG_MTD_BLOCK=y
+# CONFIG_FTL is not set
+# CONFIG_NFTL is not set
+# CONFIG_INFTL is not set
+# CONFIG_RFD_FTL is not set
+# CONFIG_SSFDC is not set
+# CONFIG_MTD_OOPS is not set
+
+#
+# RAM/ROM/Flash chip drivers
+#
+CONFIG_MTD_CFI=y
+# CONFIG_MTD_JEDECPROBE is not set
+CONFIG_MTD_GEN_PROBE=y
+# CONFIG_MTD_CFI_ADV_OPTIONS is not set
+CONFIG_MTD_MAP_BANK_WIDTH_1=y
+CONFIG_MTD_MAP_BANK_WIDTH_2=y
+CONFIG_MTD_MAP_BANK_WIDTH_4=y
+# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set
+CONFIG_MTD_CFI_I1=y
+CONFIG_MTD_CFI_I2=y
+# CONFIG_MTD_CFI_I4 is not set
+# CONFIG_MTD_CFI_I8 is not set
+# CONFIG_MTD_CFI_INTELEXT is not set
+CONFIG_MTD_CFI_AMDSTD=y
+# CONFIG_MTD_CFI_STAA is not set
+CONFIG_MTD_CFI_UTIL=y
+CONFIG_MTD_RAM=y
+# CONFIG_MTD_ROM is not set
+# CONFIG_MTD_ABSENT is not set
+
+#
+# Mapping drivers for chip access
+#
+# CONFIG_MTD_COMPLEX_MAPPINGS is not set
+# CONFIG_MTD_PHYSMAP is not set
+CONFIG_MTD_UCLINUX=y
+# CONFIG_MTD_PLATRAM is not set
+
+#
+# Self-contained MTD device drivers
+#
+# CONFIG_MTD_SLRAM is not set
+# CONFIG_MTD_PHRAM is not set
+# CONFIG_MTD_MTDRAM is not set
+# CONFIG_MTD_BLOCK2MTD is not set
+
+#
+# Disk-On-Chip Device Drivers
+#
+# CONFIG_MTD_DOC2000 is not set
+# CONFIG_MTD_DOC2001 is not set
+# CONFIG_MTD_DOC2001PLUS is not set
+# CONFIG_MTD_NAND is not set
+# CONFIG_MTD_ONENAND is not set
+
+#
+# UBI - Unsorted block images
+#
+# CONFIG_MTD_UBI is not set
+# CONFIG_PARPORT is not set
+CONFIG_BLK_DEV=y
+# CONFIG_BLK_DEV_COW_COMMON is not set
+# CONFIG_BLK_DEV_LOOP is not set
+# CONFIG_BLK_DEV_NBD is not set
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_COUNT=16
+CONFIG_BLK_DEV_RAM_SIZE=4096
+# CONFIG_BLK_DEV_XIP is not set
+# CONFIG_CDROM_PKTCDVD is not set
+# CONFIG_ATA_OVER_ETH is not set
+# CONFIG_MISC_DEVICES is not set
+CONFIG_HAVE_IDE=y
+# CONFIG_IDE is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+# CONFIG_SCSI is not set
+# CONFIG_SCSI_DMA is not set
+# CONFIG_SCSI_NETLINK is not set
+# CONFIG_MD is not set
+CONFIG_NETDEVICES=y
+# CONFIG_NETDEVICES_MULTIQUEUE is not set
+# CONFIG_DUMMY is not set
+# CONFIG_BONDING is not set
+# CONFIG_MACVLAN is not set
+# CONFIG_EQUALIZER is not set
+# CONFIG_TUN is not set
+# CONFIG_VETH is not set
+# CONFIG_PHYLIB is not set
+CONFIG_NET_ETHERNET=y
+# CONFIG_MII is not set
+# CONFIG_IBM_NEW_EMAC_ZMII is not set
+# CONFIG_IBM_NEW_EMAC_RGMII is not set
+# CONFIG_IBM_NEW_EMAC_TAH is not set
+# CONFIG_IBM_NEW_EMAC_EMAC4 is not set
+# CONFIG_B44 is not set
+CONFIG_FEC=y
+# CONFIG_FEC2 is not set
+# CONFIG_NETDEV_1000 is not set
+# CONFIG_NETDEV_10000 is not set
+
+#
+# Wireless LAN
+#
+# CONFIG_WLAN_PRE80211 is not set
+# CONFIG_WLAN_80211 is not set
+# CONFIG_IWLWIFI is not set
+# CONFIG_IWLWIFI_LEDS is not set
+# CONFIG_WAN is not set
+# CONFIG_PPP is not set
+# CONFIG_SLIP is not set
+# CONFIG_NETCONSOLE is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+# CONFIG_ISDN is not set
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+# CONFIG_INPUT is not set
+
+#
+# Hardware I/O ports
+#
+# CONFIG_SERIO is not set
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+# CONFIG_VT is not set
+# CONFIG_DEVKMEM is not set
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_COLDFIRE is not set
+CONFIG_SERIAL_MCF=y
+CONFIG_SERIAL_MCF_BAUDRATE=115200
+CONFIG_SERIAL_MCF_CONSOLE=y
+# CONFIG_UNIX98_PTYS is not set
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=256
+# CONFIG_IPMI_HANDLER is not set
+# CONFIG_HW_RANDOM is not set
+# CONFIG_GEN_RTC is not set
+# CONFIG_R3964 is not set
+# CONFIG_RAW_DRIVER is not set
+# CONFIG_TCG_TPM is not set
+# CONFIG_I2C is not set
+# CONFIG_SPI is not set
+# CONFIG_W1 is not set
+# CONFIG_POWER_SUPPLY is not set
+# CONFIG_HWMON is not set
+# CONFIG_THERMAL is not set
+# CONFIG_WATCHDOG is not set
+
+#
+# Sonics Silicon Backplane
+#
+CONFIG_SSB_POSSIBLE=y
+# CONFIG_SSB is not set
+
+#
+# Multifunction device drivers
+#
+# CONFIG_MFD_SM501 is not set
+# CONFIG_HTC_PASIC3 is not set
+
+#
+# Multimedia devices
+#
+
+#
+# Multimedia core support
+#
+# CONFIG_VIDEO_DEV is not set
+# CONFIG_DVB_CORE is not set
+
+#
+# Multimedia drivers
+#
+# CONFIG_DAB is not set
+
+#
+# Graphics support
+#
+# CONFIG_VGASTATE is not set
+# CONFIG_VIDEO_OUTPUT_CONTROL is not set
+# CONFIG_FB is not set
+# CONFIG_BACKLIGHT_LCD_SUPPORT is not set
+
+#
+# Display device support
+#
+# CONFIG_DISPLAY_SUPPORT is not set
+
+#
+# Sound
+#
+# CONFIG_SOUND is not set
+# CONFIG_USB_SUPPORT is not set
+# CONFIG_MMC is not set
+# CONFIG_MEMSTICK is not set
+# CONFIG_NEW_LEDS is not set
+# CONFIG_ACCESSIBILITY is not set
+# CONFIG_RTC_CLASS is not set
+# CONFIG_UIO is not set
+
+#
+# File systems
+#
+CONFIG_EXT2_FS=y
+# CONFIG_EXT2_FS_XATTR is not set
+# CONFIG_EXT3_FS is not set
+# CONFIG_EXT4DEV_FS is not set
+# CONFIG_REISERFS_FS is not set
+# CONFIG_JFS_FS is not set
+# CONFIG_FS_POSIX_ACL is not set
+# CONFIG_XFS_FS is not set
+# CONFIG_DNOTIFY is not set
+# CONFIG_INOTIFY is not set
+# CONFIG_QUOTA is not set
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+# CONFIG_FUSE_FS is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+# CONFIG_ISO9660_FS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+# CONFIG_MSDOS_FS is not set
+# CONFIG_VFAT_FS is not set
+# CONFIG_NTFS_FS is not set
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+CONFIG_PROC_SYSCTL=y
+# CONFIG_SYSFS is not set
+# CONFIG_TMPFS is not set
+# CONFIG_HUGETLB_PAGE is not set
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+# CONFIG_JFFS2_FS is not set
+# CONFIG_CRAMFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_MINIX_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+CONFIG_ROMFS_FS=y
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+# CONFIG_NETWORK_FILESYSTEMS is not set
+
+#
+# Partition Types
+#
+# CONFIG_PARTITION_ADVANCED is not set
+CONFIG_MSDOS_PARTITION=y
+# CONFIG_NLS is not set
+
+#
+# Kernel hacking
+#
+# CONFIG_PRINTK_TIME is not set
+CONFIG_ENABLE_WARN_DEPRECATED=y
+CONFIG_ENABLE_MUST_CHECK=y
+CONFIG_FRAME_WARN=1024
+# CONFIG_MAGIC_SYSRQ is not set
+# CONFIG_UNUSED_SYMBOLS is not set
+# CONFIG_HEADERS_CHECK is not set
+# CONFIG_DEBUG_KERNEL is not set
+# CONFIG_DEBUG_BUGVERBOSE is not set
+# CONFIG_SAMPLES is not set
+CONFIG_FULLDEBUG=y
+# CONFIG_HIGHPROFILE is not set
+# CONFIG_BOOTPARAM is not set
+# CONFIG_NO_KERNEL_MSG is not set
+# CONFIG_BDM_DISABLE is not set
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+# CONFIG_SECURITY_FILE_CAPABILITIES is not set
+# CONFIG_CRYPTO is not set
+
+#
+# Library routines
+#
+CONFIG_BITREVERSE=y
+# CONFIG_GENERIC_FIND_FIRST_BIT is not set
+# CONFIG_CRC_CCITT is not set
+# CONFIG_CRC16 is not set
+# CONFIG_CRC_ITU_T is not set
+CONFIG_CRC32=y
+# CONFIG_CRC7 is not set
+# CONFIG_LIBCRC32C is not set
+CONFIG_HAS_IOMEM=y
+CONFIG_HAS_DMA=y

From 8c81b0574fd8877b5214f2d33816b199c62e3335 Mon Sep 17 00:00:00 2001
From: Greg Ungerer <gerg@goober.(none)>
Date: Wed, 4 Jun 2008 21:22:18 +1000
Subject: [PATCH 0119/1171] m68knommu: defconfig for M5249EVB board

Add a defconfig for the Freescale M5249EVB board.

Signed-off-by: Greg Ungerer <gerg@uclinux.org>
---
 arch/m68knommu/configs/m5249evb_defconfig | 497 ++++++++++++++++++++++
 1 file changed, 497 insertions(+)
 create mode 100644 arch/m68knommu/configs/m5249evb_defconfig

diff --git a/arch/m68knommu/configs/m5249evb_defconfig b/arch/m68knommu/configs/m5249evb_defconfig
new file mode 100644
index 000000000000..cc6458333d67
--- /dev/null
+++ b/arch/m68knommu/configs/m5249evb_defconfig
@@ -0,0 +1,497 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.26-rc1
+#
+CONFIG_M68K=y
+# CONFIG_MMU is not set
+# CONFIG_FPU is not set
+CONFIG_ZONE_DMA=y
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+# CONFIG_RWSEM_XCHGADD_ALGORITHM is not set
+# CONFIG_ARCH_HAS_ILOG2_U32 is not set
+# CONFIG_ARCH_HAS_ILOG2_U64 is not set
+CONFIG_GENERIC_FIND_NEXT_BIT=y
+CONFIG_GENERIC_HWEIGHT=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+CONFIG_GENERIC_TIME=y
+CONFIG_TIME_LOW_RES=y
+CONFIG_NO_IOPORT=y
+CONFIG_ARCH_SUPPORTS_AOUT=y
+CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
+
+#
+# General setup
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+CONFIG_LOCALVERSION=""
+CONFIG_LOCALVERSION_AUTO=y
+# CONFIG_SYSVIPC is not set
+# CONFIG_BSD_PROCESS_ACCT is not set
+# CONFIG_IKCONFIG is not set
+CONFIG_LOG_BUF_SHIFT=14
+# CONFIG_CGROUPS is not set
+# CONFIG_GROUP_SCHED is not set
+# CONFIG_SYSFS_DEPRECATED_V2 is not set
+# CONFIG_RELAY is not set
+# CONFIG_NAMESPACES is not set
+# CONFIG_BLK_DEV_INITRD is not set
+# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
+CONFIG_SYSCTL=y
+CONFIG_EMBEDDED=y
+# CONFIG_UID16 is not set
+# CONFIG_SYSCTL_SYSCALL is not set
+# CONFIG_KALLSYMS is not set
+# CONFIG_HOTPLUG is not set
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_ELF_CORE=y
+# CONFIG_COMPAT_BRK is not set
+CONFIG_BASE_FULL=y
+# CONFIG_FUTEX is not set
+# CONFIG_EPOLL is not set
+# CONFIG_SIGNALFD is not set
+# CONFIG_TIMERFD is not set
+# CONFIG_EVENTFD is not set
+# CONFIG_VM_EVENT_COUNTERS is not set
+CONFIG_SLAB=y
+# CONFIG_SLUB is not set
+# CONFIG_SLOB is not set
+# CONFIG_PROFILING is not set
+# CONFIG_MARKERS is not set
+# CONFIG_HAVE_OPROFILE is not set
+# CONFIG_HAVE_KPROBES is not set
+# CONFIG_HAVE_KRETPROBES is not set
+# CONFIG_HAVE_DMA_ATTRS is not set
+CONFIG_SLABINFO=y
+CONFIG_TINY_SHMEM=y
+CONFIG_BASE_SMALL=0
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+# CONFIG_MODULE_FORCE_UNLOAD is not set
+# CONFIG_MODVERSIONS is not set
+# CONFIG_MODULE_SRCVERSION_ALL is not set
+# CONFIG_KMOD is not set
+CONFIG_BLOCK=y
+# CONFIG_LBD is not set
+# CONFIG_BLK_DEV_IO_TRACE is not set
+# CONFIG_LSF is not set
+# CONFIG_BLK_DEV_BSG is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+# CONFIG_IOSCHED_AS is not set
+# CONFIG_IOSCHED_DEADLINE is not set
+# CONFIG_IOSCHED_CFQ is not set
+# CONFIG_DEFAULT_AS is not set
+# CONFIG_DEFAULT_DEADLINE is not set
+# CONFIG_DEFAULT_CFQ is not set
+CONFIG_DEFAULT_NOOP=y
+CONFIG_DEFAULT_IOSCHED="noop"
+CONFIG_CLASSIC_RCU=y
+
+#
+# Processor type and features
+#
+# CONFIG_M68328 is not set
+# CONFIG_M68EZ328 is not set
+# CONFIG_M68VZ328 is not set
+# CONFIG_M68360 is not set
+# CONFIG_M5206 is not set
+# CONFIG_M5206e is not set
+# CONFIG_M520x is not set
+# CONFIG_M523x is not set
+CONFIG_M5249=y
+# CONFIG_M5271 is not set
+# CONFIG_M5272 is not set
+# CONFIG_M5275 is not set
+# CONFIG_M528x is not set
+# CONFIG_M5307 is not set
+# CONFIG_M532x is not set
+# CONFIG_M5407 is not set
+CONFIG_COLDFIRE=y
+CONFIG_CLOCK_SET=y
+CONFIG_CLOCK_FREQ=140000000
+CONFIG_CLOCK_DIV=2
+
+#
+# Platform
+#
+CONFIG_M5249C3=y
+CONFIG_FREESCALE=y
+CONFIG_4KSTACKS=y
+CONFIG_HZ=100
+
+#
+# RAM configuration
+#
+CONFIG_RAMBASE=0x00000000
+CONFIG_RAMSIZE=0x00800000
+CONFIG_VECTORBASE=0x00000000
+CONFIG_KERNELBASE=0x00020000
+CONFIG_RAMAUTOBIT=y
+# CONFIG_RAM8BIT is not set
+# CONFIG_RAM16BIT is not set
+# CONFIG_RAM32BIT is not set
+
+#
+# ROM configuration
+#
+# CONFIG_ROM is not set
+CONFIG_RAMKERNEL=y
+# CONFIG_ROMKERNEL is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+# CONFIG_SPARSEMEM_STATIC is not set
+# CONFIG_SPARSEMEM_VMEMMAP_ENABLE is not set
+CONFIG_PAGEFLAGS_EXTENDED=y
+CONFIG_SPLIT_PTLOCK_CPUS=4
+# CONFIG_RESOURCES_64BIT is not set
+CONFIG_ZONE_DMA_FLAG=1
+CONFIG_VIRT_TO_BUS=y
+CONFIG_ISA_DMA_API=y
+
+#
+# Bus options (PCI, PCMCIA, EISA, MCA, ISA)
+#
+# CONFIG_PCI is not set
+# CONFIG_ARCH_SUPPORTS_MSI is not set
+
+#
+# Executable file formats
+#
+CONFIG_BINFMT_FLAT=y
+# CONFIG_BINFMT_ZFLAT is not set
+# CONFIG_BINFMT_SHARED_FLAT is not set
+# CONFIG_BINFMT_AOUT is not set
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Power management options
+#
+# CONFIG_PM is not set
+
+#
+# Networking
+#
+# CONFIG_NET is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+# CONFIG_SYS_HYPERVISOR is not set
+CONFIG_MTD=y
+# CONFIG_MTD_DEBUG is not set
+# CONFIG_MTD_CONCAT is not set
+CONFIG_MTD_PARTITIONS=y
+# CONFIG_MTD_REDBOOT_PARTS is not set
+# CONFIG_MTD_CMDLINE_PARTS is not set
+# CONFIG_MTD_AR7_PARTS is not set
+
+#
+# User Modules And Translation Layers
+#
+CONFIG_MTD_CHAR=y
+CONFIG_MTD_BLKDEVS=y
+CONFIG_MTD_BLOCK=y
+# CONFIG_FTL is not set
+# CONFIG_NFTL is not set
+# CONFIG_INFTL is not set
+# CONFIG_RFD_FTL is not set
+# CONFIG_SSFDC is not set
+# CONFIG_MTD_OOPS is not set
+
+#
+# RAM/ROM/Flash chip drivers
+#
+# CONFIG_MTD_CFI is not set
+# CONFIG_MTD_JEDECPROBE is not set
+CONFIG_MTD_MAP_BANK_WIDTH_1=y
+CONFIG_MTD_MAP_BANK_WIDTH_2=y
+CONFIG_MTD_MAP_BANK_WIDTH_4=y
+# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set
+CONFIG_MTD_CFI_I1=y
+CONFIG_MTD_CFI_I2=y
+# CONFIG_MTD_CFI_I4 is not set
+# CONFIG_MTD_CFI_I8 is not set
+CONFIG_MTD_RAM=y
+# CONFIG_MTD_ROM is not set
+# CONFIG_MTD_ABSENT is not set
+
+#
+# Mapping drivers for chip access
+#
+# CONFIG_MTD_COMPLEX_MAPPINGS is not set
+CONFIG_MTD_UCLINUX=y
+# CONFIG_MTD_PLATRAM is not set
+
+#
+# Self-contained MTD device drivers
+#
+# CONFIG_MTD_SLRAM is not set
+# CONFIG_MTD_PHRAM is not set
+# CONFIG_MTD_MTDRAM is not set
+# CONFIG_MTD_BLOCK2MTD is not set
+
+#
+# Disk-On-Chip Device Drivers
+#
+# CONFIG_MTD_DOC2000 is not set
+# CONFIG_MTD_DOC2001 is not set
+# CONFIG_MTD_DOC2001PLUS is not set
+# CONFIG_MTD_NAND is not set
+# CONFIG_MTD_ONENAND is not set
+
+#
+# UBI - Unsorted block images
+#
+# CONFIG_MTD_UBI is not set
+# CONFIG_PARPORT is not set
+CONFIG_BLK_DEV=y
+# CONFIG_BLK_DEV_COW_COMMON is not set
+# CONFIG_BLK_DEV_LOOP is not set
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_COUNT=16
+CONFIG_BLK_DEV_RAM_SIZE=4096
+# CONFIG_BLK_DEV_XIP is not set
+# CONFIG_CDROM_PKTCDVD is not set
+CONFIG_MISC_DEVICES=y
+# CONFIG_EEPROM_93CX6 is not set
+# CONFIG_ENCLOSURE_SERVICES is not set
+CONFIG_HAVE_IDE=y
+# CONFIG_IDE is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+# CONFIG_SCSI is not set
+# CONFIG_SCSI_DMA is not set
+# CONFIG_SCSI_NETLINK is not set
+# CONFIG_MD is not set
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+# CONFIG_INPUT is not set
+
+#
+# Hardware I/O ports
+#
+# CONFIG_SERIO is not set
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+# CONFIG_VT is not set
+# CONFIG_DEVKMEM is not set
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_COLDFIRE is not set
+CONFIG_SERIAL_MCF=y
+CONFIG_SERIAL_MCF_BAUDRATE=19200
+CONFIG_SERIAL_MCF_CONSOLE=y
+# CONFIG_UNIX98_PTYS is not set
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=256
+# CONFIG_IPMI_HANDLER is not set
+# CONFIG_HW_RANDOM is not set
+# CONFIG_GEN_RTC is not set
+# CONFIG_R3964 is not set
+# CONFIG_RAW_DRIVER is not set
+# CONFIG_TCG_TPM is not set
+# CONFIG_I2C is not set
+# CONFIG_SPI is not set
+# CONFIG_W1 is not set
+# CONFIG_POWER_SUPPLY is not set
+# CONFIG_HWMON is not set
+# CONFIG_THERMAL is not set
+# CONFIG_WATCHDOG is not set
+
+#
+# Sonics Silicon Backplane
+#
+CONFIG_SSB_POSSIBLE=y
+# CONFIG_SSB is not set
+
+#
+# Multifunction device drivers
+#
+# CONFIG_MFD_SM501 is not set
+# CONFIG_HTC_PASIC3 is not set
+
+#
+# Multimedia devices
+#
+
+#
+# Multimedia core support
+#
+# CONFIG_VIDEO_DEV is not set
+
+#
+# Multimedia drivers
+#
+# CONFIG_DAB is not set
+
+#
+# Graphics support
+#
+# CONFIG_VGASTATE is not set
+# CONFIG_VIDEO_OUTPUT_CONTROL is not set
+# CONFIG_FB is not set
+# CONFIG_BACKLIGHT_LCD_SUPPORT is not set
+
+#
+# Display device support
+#
+# CONFIG_DISPLAY_SUPPORT is not set
+
+#
+# Sound
+#
+# CONFIG_SOUND is not set
+# CONFIG_USB_SUPPORT is not set
+# CONFIG_MMC is not set
+# CONFIG_MEMSTICK is not set
+# CONFIG_NEW_LEDS is not set
+# CONFIG_ACCESSIBILITY is not set
+# CONFIG_RTC_CLASS is not set
+# CONFIG_UIO is not set
+
+#
+# File systems
+#
+CONFIG_EXT2_FS=y
+# CONFIG_EXT2_FS_XATTR is not set
+# CONFIG_EXT3_FS is not set
+# CONFIG_EXT4DEV_FS is not set
+# CONFIG_REISERFS_FS is not set
+# CONFIG_JFS_FS is not set
+# CONFIG_FS_POSIX_ACL is not set
+# CONFIG_XFS_FS is not set
+# CONFIG_DNOTIFY is not set
+# CONFIG_INOTIFY is not set
+# CONFIG_QUOTA is not set
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+# CONFIG_FUSE_FS is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+# CONFIG_ISO9660_FS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+# CONFIG_MSDOS_FS is not set
+# CONFIG_VFAT_FS is not set
+# CONFIG_NTFS_FS is not set
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+CONFIG_PROC_SYSCTL=y
+CONFIG_SYSFS=y
+# CONFIG_TMPFS is not set
+# CONFIG_HUGETLB_PAGE is not set
+# CONFIG_CONFIGFS_FS is not set
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+# CONFIG_JFFS2_FS is not set
+# CONFIG_CRAMFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_MINIX_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+CONFIG_ROMFS_FS=y
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+
+#
+# Partition Types
+#
+# CONFIG_PARTITION_ADVANCED is not set
+CONFIG_MSDOS_PARTITION=y
+# CONFIG_NLS is not set
+
+#
+# Kernel hacking
+#
+# CONFIG_PRINTK_TIME is not set
+CONFIG_ENABLE_WARN_DEPRECATED=y
+CONFIG_ENABLE_MUST_CHECK=y
+CONFIG_FRAME_WARN=1024
+# CONFIG_MAGIC_SYSRQ is not set
+# CONFIG_UNUSED_SYMBOLS is not set
+# CONFIG_DEBUG_FS is not set
+# CONFIG_HEADERS_CHECK is not set
+# CONFIG_DEBUG_KERNEL is not set
+# CONFIG_DEBUG_BUGVERBOSE is not set
+# CONFIG_SAMPLES is not set
+# CONFIG_FULLDEBUG is not set
+# CONFIG_HIGHPROFILE is not set
+# CONFIG_BOOTPARAM is not set
+# CONFIG_NO_KERNEL_MSG is not set
+# CONFIG_BDM_DISABLE is not set
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+# CONFIG_SECURITY is not set
+# CONFIG_SECURITY_FILE_CAPABILITIES is not set
+# CONFIG_CRYPTO is not set
+
+#
+# Library routines
+#
+# CONFIG_GENERIC_FIND_FIRST_BIT is not set
+# CONFIG_CRC_CCITT is not set
+# CONFIG_CRC16 is not set
+# CONFIG_CRC_ITU_T is not set
+# CONFIG_CRC32 is not set
+# CONFIG_CRC7 is not set
+# CONFIG_LIBCRC32C is not set
+CONFIG_HAS_IOMEM=y
+CONFIG_HAS_DMA=y

From c26def07f97d662ee96362795400cbf6f44d11ce Mon Sep 17 00:00:00 2001
From: Greg Ungerer <gerg@goober.(none)>
Date: Wed, 4 Jun 2008 21:23:43 +1000
Subject: [PATCH 0120/1171] m68knommu: defconfig for M5275EVB board

Add a defconfig for the Freescale M5275EVB board.

Signed-off-by: Greg Ungerer <gerg@uclinux.org>
---
 arch/m68knommu/configs/m5275evb_defconfig | 627 ++++++++++++++++++++++
 1 file changed, 627 insertions(+)
 create mode 100644 arch/m68knommu/configs/m5275evb_defconfig

diff --git a/arch/m68knommu/configs/m5275evb_defconfig b/arch/m68knommu/configs/m5275evb_defconfig
new file mode 100644
index 000000000000..0d1256f5addb
--- /dev/null
+++ b/arch/m68knommu/configs/m5275evb_defconfig
@@ -0,0 +1,627 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.26-rc1
+#
+CONFIG_M68K=y
+# CONFIG_MMU is not set
+# CONFIG_FPU is not set
+CONFIG_ZONE_DMA=y
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+# CONFIG_RWSEM_XCHGADD_ALGORITHM is not set
+# CONFIG_ARCH_HAS_ILOG2_U32 is not set
+# CONFIG_ARCH_HAS_ILOG2_U64 is not set
+CONFIG_GENERIC_FIND_NEXT_BIT=y
+CONFIG_GENERIC_HWEIGHT=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+CONFIG_GENERIC_TIME=y
+CONFIG_TIME_LOW_RES=y
+CONFIG_NO_IOPORT=y
+CONFIG_ARCH_SUPPORTS_AOUT=y
+CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
+
+#
+# General setup
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+CONFIG_LOCALVERSION=""
+CONFIG_LOCALVERSION_AUTO=y
+# CONFIG_SYSVIPC is not set
+# CONFIG_POSIX_MQUEUE is not set
+# CONFIG_BSD_PROCESS_ACCT is not set
+# CONFIG_TASKSTATS is not set
+# CONFIG_AUDIT is not set
+# CONFIG_IKCONFIG is not set
+CONFIG_LOG_BUF_SHIFT=14
+# CONFIG_CGROUPS is not set
+# CONFIG_GROUP_SCHED is not set
+# CONFIG_SYSFS_DEPRECATED_V2 is not set
+# CONFIG_RELAY is not set
+# CONFIG_NAMESPACES is not set
+# CONFIG_BLK_DEV_INITRD is not set
+# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
+CONFIG_SYSCTL=y
+CONFIG_EMBEDDED=y
+# CONFIG_UID16 is not set
+# CONFIG_SYSCTL_SYSCALL is not set
+# CONFIG_KALLSYMS is not set
+# CONFIG_HOTPLUG is not set
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_ELF_CORE=y
+# CONFIG_COMPAT_BRK is not set
+CONFIG_BASE_FULL=y
+# CONFIG_FUTEX is not set
+# CONFIG_EPOLL is not set
+# CONFIG_SIGNALFD is not set
+# CONFIG_TIMERFD is not set
+# CONFIG_EVENTFD is not set
+# CONFIG_VM_EVENT_COUNTERS is not set
+CONFIG_SLAB=y
+# CONFIG_SLUB is not set
+# CONFIG_SLOB is not set
+# CONFIG_PROFILING is not set
+# CONFIG_MARKERS is not set
+# CONFIG_HAVE_OPROFILE is not set
+# CONFIG_HAVE_KPROBES is not set
+# CONFIG_HAVE_KRETPROBES is not set
+# CONFIG_HAVE_DMA_ATTRS is not set
+CONFIG_SLABINFO=y
+CONFIG_TINY_SHMEM=y
+CONFIG_BASE_SMALL=0
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+# CONFIG_MODULE_FORCE_UNLOAD is not set
+# CONFIG_MODVERSIONS is not set
+# CONFIG_MODULE_SRCVERSION_ALL is not set
+# CONFIG_KMOD is not set
+CONFIG_BLOCK=y
+# CONFIG_LBD is not set
+# CONFIG_BLK_DEV_IO_TRACE is not set
+# CONFIG_LSF is not set
+# CONFIG_BLK_DEV_BSG is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+# CONFIG_IOSCHED_AS is not set
+# CONFIG_IOSCHED_DEADLINE is not set
+# CONFIG_IOSCHED_CFQ is not set
+# CONFIG_DEFAULT_AS is not set
+# CONFIG_DEFAULT_DEADLINE is not set
+# CONFIG_DEFAULT_CFQ is not set
+CONFIG_DEFAULT_NOOP=y
+CONFIG_DEFAULT_IOSCHED="noop"
+CONFIG_CLASSIC_RCU=y
+
+#
+# Processor type and features
+#
+# CONFIG_M68328 is not set
+# CONFIG_M68EZ328 is not set
+# CONFIG_M68VZ328 is not set
+# CONFIG_M68360 is not set
+# CONFIG_M5206 is not set
+# CONFIG_M5206e is not set
+# CONFIG_M520x is not set
+# CONFIG_M523x is not set
+# CONFIG_M5249 is not set
+# CONFIG_M5271 is not set
+# CONFIG_M5272 is not set
+CONFIG_M5275=y
+# CONFIG_M528x is not set
+# CONFIG_M5307 is not set
+# CONFIG_M532x is not set
+# CONFIG_M5407 is not set
+CONFIG_M527x=y
+CONFIG_COLDFIRE=y
+CONFIG_CLOCK_SET=y
+CONFIG_CLOCK_FREQ=150000000
+CONFIG_CLOCK_DIV=2
+
+#
+# Platform
+#
+CONFIG_M5275EVB=y
+CONFIG_FREESCALE=y
+# CONFIG_4KSTACKS is not set
+CONFIG_HZ=100
+
+#
+# RAM configuration
+#
+CONFIG_RAMBASE=0x00000000
+CONFIG_RAMSIZE=0x00000000
+CONFIG_VECTORBASE=0x00000000
+CONFIG_KERNELBASE=0x00020000
+CONFIG_RAMAUTOBIT=y
+# CONFIG_RAM8BIT is not set
+# CONFIG_RAM16BIT is not set
+# CONFIG_RAM32BIT is not set
+
+#
+# ROM configuration
+#
+# CONFIG_ROM is not set
+CONFIG_RAMKERNEL=y
+# CONFIG_ROMKERNEL is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+# CONFIG_SPARSEMEM_STATIC is not set
+# CONFIG_SPARSEMEM_VMEMMAP_ENABLE is not set
+CONFIG_PAGEFLAGS_EXTENDED=y
+CONFIG_SPLIT_PTLOCK_CPUS=4
+# CONFIG_RESOURCES_64BIT is not set
+CONFIG_ZONE_DMA_FLAG=1
+CONFIG_VIRT_TO_BUS=y
+CONFIG_ISA_DMA_API=y
+
+#
+# Bus options (PCI, PCMCIA, EISA, MCA, ISA)
+#
+# CONFIG_PCI is not set
+# CONFIG_ARCH_SUPPORTS_MSI is not set
+
+#
+# Executable file formats
+#
+CONFIG_BINFMT_FLAT=y
+# CONFIG_BINFMT_ZFLAT is not set
+# CONFIG_BINFMT_SHARED_FLAT is not set
+# CONFIG_BINFMT_AOUT is not set
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Power management options
+#
+# CONFIG_PM is not set
+
+#
+# Networking
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+CONFIG_PACKET=y
+# CONFIG_PACKET_MMAP is not set
+CONFIG_UNIX=y
+# CONFIG_NET_KEY is not set
+CONFIG_INET=y
+# CONFIG_IP_MULTICAST is not set
+# CONFIG_IP_ADVANCED_ROUTER is not set
+CONFIG_IP_FIB_HASH=y
+# CONFIG_IP_PNP is not set
+# CONFIG_NET_IPIP is not set
+# CONFIG_NET_IPGRE is not set
+# CONFIG_ARPD is not set
+# CONFIG_SYN_COOKIES is not set
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_XFRM_TUNNEL is not set
+# CONFIG_INET_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
+# CONFIG_INET_LRO is not set
+# CONFIG_INET_DIAG is not set
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_CUBIC=y
+CONFIG_DEFAULT_TCP_CONG="cubic"
+# CONFIG_TCP_MD5SIG is not set
+# CONFIG_IPV6 is not set
+# CONFIG_NETWORK_SECMARK is not set
+# CONFIG_NETFILTER is not set
+# CONFIG_IP_DCCP is not set
+# CONFIG_IP_SCTP is not set
+# CONFIG_TIPC is not set
+# CONFIG_ATM is not set
+# CONFIG_BRIDGE is not set
+# CONFIG_VLAN_8021Q is not set
+# CONFIG_DECNET is not set
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+# CONFIG_NET_SCHED is not set
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_CAN is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+# CONFIG_AF_RXRPC is not set
+
+#
+# Wireless
+#
+# CONFIG_CFG80211 is not set
+# CONFIG_WIRELESS_EXT is not set
+# CONFIG_MAC80211 is not set
+# CONFIG_IEEE80211 is not set
+# CONFIG_RFKILL is not set
+# CONFIG_NET_9P is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+# CONFIG_SYS_HYPERVISOR is not set
+# CONFIG_CONNECTOR is not set
+CONFIG_MTD=y
+# CONFIG_MTD_DEBUG is not set
+# CONFIG_MTD_CONCAT is not set
+CONFIG_MTD_PARTITIONS=y
+# CONFIG_MTD_REDBOOT_PARTS is not set
+# CONFIG_MTD_CMDLINE_PARTS is not set
+# CONFIG_MTD_AR7_PARTS is not set
+
+#
+# User Modules And Translation Layers
+#
+CONFIG_MTD_CHAR=y
+CONFIG_MTD_BLKDEVS=y
+CONFIG_MTD_BLOCK=y
+# CONFIG_FTL is not set
+# CONFIG_NFTL is not set
+# CONFIG_INFTL is not set
+# CONFIG_RFD_FTL is not set
+# CONFIG_SSFDC is not set
+# CONFIG_MTD_OOPS is not set
+
+#
+# RAM/ROM/Flash chip drivers
+#
+# CONFIG_MTD_CFI is not set
+# CONFIG_MTD_JEDECPROBE is not set
+CONFIG_MTD_MAP_BANK_WIDTH_1=y
+CONFIG_MTD_MAP_BANK_WIDTH_2=y
+CONFIG_MTD_MAP_BANK_WIDTH_4=y
+# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set
+CONFIG_MTD_CFI_I1=y
+CONFIG_MTD_CFI_I2=y
+# CONFIG_MTD_CFI_I4 is not set
+# CONFIG_MTD_CFI_I8 is not set
+CONFIG_MTD_RAM=y
+# CONFIG_MTD_ROM is not set
+# CONFIG_MTD_ABSENT is not set
+
+#
+# Mapping drivers for chip access
+#
+# CONFIG_MTD_COMPLEX_MAPPINGS is not set
+CONFIG_MTD_UCLINUX=y
+# CONFIG_MTD_PLATRAM is not set
+
+#
+# Self-contained MTD device drivers
+#
+# CONFIG_MTD_SLRAM is not set
+# CONFIG_MTD_PHRAM is not set
+# CONFIG_MTD_MTDRAM is not set
+# CONFIG_MTD_BLOCK2MTD is not set
+
+#
+# Disk-On-Chip Device Drivers
+#
+# CONFIG_MTD_DOC2000 is not set
+# CONFIG_MTD_DOC2001 is not set
+# CONFIG_MTD_DOC2001PLUS is not set
+# CONFIG_MTD_NAND is not set
+# CONFIG_MTD_ONENAND is not set
+
+#
+# UBI - Unsorted block images
+#
+# CONFIG_MTD_UBI is not set
+# CONFIG_PARPORT is not set
+CONFIG_BLK_DEV=y
+# CONFIG_BLK_DEV_COW_COMMON is not set
+# CONFIG_BLK_DEV_LOOP is not set
+# CONFIG_BLK_DEV_NBD is not set
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_COUNT=16
+CONFIG_BLK_DEV_RAM_SIZE=4096
+# CONFIG_BLK_DEV_XIP is not set
+# CONFIG_CDROM_PKTCDVD is not set
+# CONFIG_ATA_OVER_ETH is not set
+# CONFIG_MISC_DEVICES is not set
+CONFIG_HAVE_IDE=y
+# CONFIG_IDE is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+# CONFIG_SCSI is not set
+# CONFIG_SCSI_DMA is not set
+# CONFIG_SCSI_NETLINK is not set
+# CONFIG_MD is not set
+CONFIG_NETDEVICES=y
+# CONFIG_NETDEVICES_MULTIQUEUE is not set
+# CONFIG_DUMMY is not set
+# CONFIG_BONDING is not set
+# CONFIG_MACVLAN is not set
+# CONFIG_EQUALIZER is not set
+# CONFIG_TUN is not set
+# CONFIG_VETH is not set
+# CONFIG_PHYLIB is not set
+CONFIG_NET_ETHERNET=y
+# CONFIG_MII is not set
+# CONFIG_IBM_NEW_EMAC_ZMII is not set
+# CONFIG_IBM_NEW_EMAC_RGMII is not set
+# CONFIG_IBM_NEW_EMAC_TAH is not set
+# CONFIG_IBM_NEW_EMAC_EMAC4 is not set
+# CONFIG_B44 is not set
+CONFIG_FEC=y
+CONFIG_FEC2=y
+# CONFIG_NETDEV_1000 is not set
+# CONFIG_NETDEV_10000 is not set
+
+#
+# Wireless LAN
+#
+# CONFIG_WLAN_PRE80211 is not set
+# CONFIG_WLAN_80211 is not set
+# CONFIG_IWLWIFI is not set
+# CONFIG_IWLWIFI_LEDS is not set
+# CONFIG_WAN is not set
+CONFIG_PPP=y
+# CONFIG_PPP_MULTILINK is not set
+# CONFIG_PPP_FILTER is not set
+# CONFIG_PPP_ASYNC is not set
+# CONFIG_PPP_SYNC_TTY is not set
+# CONFIG_PPP_DEFLATE is not set
+# CONFIG_PPP_BSDCOMP is not set
+# CONFIG_PPP_MPPE is not set
+# CONFIG_PPPOE is not set
+# CONFIG_PPPOL2TP is not set
+# CONFIG_SLIP is not set
+CONFIG_SLHC=y
+# CONFIG_NETCONSOLE is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+# CONFIG_ISDN is not set
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+# CONFIG_INPUT is not set
+
+#
+# Hardware I/O ports
+#
+# CONFIG_SERIO is not set
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+# CONFIG_VT is not set
+# CONFIG_DEVKMEM is not set
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_COLDFIRE is not set
+CONFIG_SERIAL_MCF=y
+CONFIG_SERIAL_MCF_BAUDRATE=19200
+CONFIG_SERIAL_MCF_CONSOLE=y
+# CONFIG_UNIX98_PTYS is not set
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=256
+# CONFIG_IPMI_HANDLER is not set
+# CONFIG_HW_RANDOM is not set
+# CONFIG_GEN_RTC is not set
+# CONFIG_R3964 is not set
+# CONFIG_RAW_DRIVER is not set
+# CONFIG_TCG_TPM is not set
+# CONFIG_I2C is not set
+# CONFIG_SPI is not set
+# CONFIG_W1 is not set
+# CONFIG_POWER_SUPPLY is not set
+# CONFIG_HWMON is not set
+# CONFIG_THERMAL is not set
+# CONFIG_WATCHDOG is not set
+
+#
+# Sonics Silicon Backplane
+#
+CONFIG_SSB_POSSIBLE=y
+# CONFIG_SSB is not set
+
+#
+# Multifunction device drivers
+#
+# CONFIG_MFD_SM501 is not set
+# CONFIG_HTC_PASIC3 is not set
+
+#
+# Multimedia devices
+#
+
+#
+# Multimedia core support
+#
+# CONFIG_VIDEO_DEV is not set
+# CONFIG_DVB_CORE is not set
+
+#
+# Multimedia drivers
+#
+CONFIG_DAB=y
+
+#
+# Graphics support
+#
+# CONFIG_VGASTATE is not set
+# CONFIG_VIDEO_OUTPUT_CONTROL is not set
+# CONFIG_FB is not set
+# CONFIG_BACKLIGHT_LCD_SUPPORT is not set
+
+#
+# Display device support
+#
+# CONFIG_DISPLAY_SUPPORT is not set
+
+#
+# Sound
+#
+# CONFIG_SOUND is not set
+# CONFIG_USB_SUPPORT is not set
+# CONFIG_MMC is not set
+# CONFIG_MEMSTICK is not set
+# CONFIG_NEW_LEDS is not set
+# CONFIG_ACCESSIBILITY is not set
+# CONFIG_RTC_CLASS is not set
+# CONFIG_UIO is not set
+
+#
+# File systems
+#
+CONFIG_EXT2_FS=y
+# CONFIG_EXT2_FS_XATTR is not set
+# CONFIG_EXT3_FS is not set
+# CONFIG_EXT4DEV_FS is not set
+# CONFIG_REISERFS_FS is not set
+# CONFIG_JFS_FS is not set
+# CONFIG_FS_POSIX_ACL is not set
+# CONFIG_XFS_FS is not set
+# CONFIG_OCFS2_FS is not set
+# CONFIG_DNOTIFY is not set
+# CONFIG_INOTIFY is not set
+# CONFIG_QUOTA is not set
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+# CONFIG_FUSE_FS is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+# CONFIG_ISO9660_FS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+# CONFIG_MSDOS_FS is not set
+# CONFIG_VFAT_FS is not set
+# CONFIG_NTFS_FS is not set
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+CONFIG_PROC_SYSCTL=y
+CONFIG_SYSFS=y
+# CONFIG_TMPFS is not set
+# CONFIG_HUGETLB_PAGE is not set
+# CONFIG_CONFIGFS_FS is not set
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+# CONFIG_JFFS2_FS is not set
+# CONFIG_CRAMFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_MINIX_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+CONFIG_ROMFS_FS=y
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+CONFIG_NETWORK_FILESYSTEMS=y
+# CONFIG_NFS_FS is not set
+# CONFIG_NFSD is not set
+# CONFIG_SMB_FS is not set
+# CONFIG_CIFS is not set
+# CONFIG_NCP_FS is not set
+# CONFIG_CODA_FS is not set
+# CONFIG_AFS_FS is not set
+
+#
+# Partition Types
+#
+# CONFIG_PARTITION_ADVANCED is not set
+CONFIG_MSDOS_PARTITION=y
+# CONFIG_NLS is not set
+# CONFIG_DLM is not set
+
+#
+# Kernel hacking
+#
+# CONFIG_PRINTK_TIME is not set
+CONFIG_ENABLE_WARN_DEPRECATED=y
+CONFIG_ENABLE_MUST_CHECK=y
+CONFIG_FRAME_WARN=1024
+# CONFIG_MAGIC_SYSRQ is not set
+# CONFIG_UNUSED_SYMBOLS is not set
+# CONFIG_DEBUG_FS is not set
+# CONFIG_HEADERS_CHECK is not set
+# CONFIG_DEBUG_KERNEL is not set
+# CONFIG_DEBUG_BUGVERBOSE is not set
+# CONFIG_SAMPLES is not set
+# CONFIG_FULLDEBUG is not set
+# CONFIG_HIGHPROFILE is not set
+# CONFIG_BOOTPARAM is not set
+# CONFIG_NO_KERNEL_MSG is not set
+# CONFIG_BDM_DISABLE is not set
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+# CONFIG_SECURITY is not set
+# CONFIG_SECURITY_FILE_CAPABILITIES is not set
+# CONFIG_CRYPTO is not set
+
+#
+# Library routines
+#
+# CONFIG_GENERIC_FIND_FIRST_BIT is not set
+# CONFIG_CRC_CCITT is not set
+# CONFIG_CRC16 is not set
+# CONFIG_CRC_ITU_T is not set
+# CONFIG_CRC32 is not set
+# CONFIG_CRC7 is not set
+# CONFIG_LIBCRC32C is not set
+CONFIG_HAS_IOMEM=y
+CONFIG_HAS_DMA=y

From f3a64eaf9ea87d70487b482a6d733cfd52ae4499 Mon Sep 17 00:00:00 2001
From: Greg Ungerer <gerg@goober.(none)>
Date: Wed, 4 Jun 2008 21:24:54 +1000
Subject: [PATCH 0121/1171] m68knommu: defconfig for M5307C3 board

Add a defconfig for the Freescale M5307C3 board.

Signed-off-by: Greg Ungerer <gerg@uclinux.org>
---
 arch/m68knommu/configs/m5307c3_defconfig | 580 +++++++++++++++++++++++
 1 file changed, 580 insertions(+)
 create mode 100644 arch/m68knommu/configs/m5307c3_defconfig

diff --git a/arch/m68knommu/configs/m5307c3_defconfig b/arch/m68knommu/configs/m5307c3_defconfig
new file mode 100644
index 000000000000..fe2acdfa4d76
--- /dev/null
+++ b/arch/m68knommu/configs/m5307c3_defconfig
@@ -0,0 +1,580 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.26-rc1
+#
+CONFIG_M68K=y
+# CONFIG_MMU is not set
+# CONFIG_FPU is not set
+CONFIG_ZONE_DMA=y
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+# CONFIG_RWSEM_XCHGADD_ALGORITHM is not set
+# CONFIG_ARCH_HAS_ILOG2_U32 is not set
+# CONFIG_ARCH_HAS_ILOG2_U64 is not set
+CONFIG_GENERIC_FIND_NEXT_BIT=y
+CONFIG_GENERIC_HWEIGHT=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+CONFIG_GENERIC_TIME=y
+CONFIG_TIME_LOW_RES=y
+CONFIG_NO_IOPORT=y
+CONFIG_ARCH_SUPPORTS_AOUT=y
+CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
+
+#
+# General setup
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+CONFIG_LOCALVERSION=""
+CONFIG_LOCALVERSION_AUTO=y
+# CONFIG_SYSVIPC is not set
+# CONFIG_POSIX_MQUEUE is not set
+# CONFIG_BSD_PROCESS_ACCT is not set
+# CONFIG_TASKSTATS is not set
+# CONFIG_AUDIT is not set
+# CONFIG_IKCONFIG is not set
+CONFIG_LOG_BUF_SHIFT=14
+# CONFIG_CGROUPS is not set
+# CONFIG_GROUP_SCHED is not set
+# CONFIG_SYSFS_DEPRECATED_V2 is not set
+# CONFIG_RELAY is not set
+# CONFIG_NAMESPACES is not set
+# CONFIG_BLK_DEV_INITRD is not set
+# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
+CONFIG_SYSCTL=y
+CONFIG_EMBEDDED=y
+# CONFIG_UID16 is not set
+# CONFIG_SYSCTL_SYSCALL is not set
+# CONFIG_KALLSYMS is not set
+# CONFIG_HOTPLUG is not set
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_ELF_CORE=y
+# CONFIG_COMPAT_BRK is not set
+CONFIG_BASE_FULL=y
+# CONFIG_FUTEX is not set
+# CONFIG_EPOLL is not set
+# CONFIG_SIGNALFD is not set
+# CONFIG_TIMERFD is not set
+# CONFIG_EVENTFD is not set
+# CONFIG_VM_EVENT_COUNTERS is not set
+CONFIG_SLAB=y
+# CONFIG_SLUB is not set
+# CONFIG_SLOB is not set
+# CONFIG_PROFILING is not set
+# CONFIG_MARKERS is not set
+# CONFIG_HAVE_OPROFILE is not set
+# CONFIG_HAVE_KPROBES is not set
+# CONFIG_HAVE_KRETPROBES is not set
+# CONFIG_HAVE_DMA_ATTRS is not set
+CONFIG_SLABINFO=y
+CONFIG_TINY_SHMEM=y
+CONFIG_BASE_SMALL=0
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+# CONFIG_MODULE_FORCE_UNLOAD is not set
+# CONFIG_MODVERSIONS is not set
+# CONFIG_MODULE_SRCVERSION_ALL is not set
+# CONFIG_KMOD is not set
+CONFIG_BLOCK=y
+# CONFIG_LBD is not set
+# CONFIG_BLK_DEV_IO_TRACE is not set
+# CONFIG_LSF is not set
+# CONFIG_BLK_DEV_BSG is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+# CONFIG_IOSCHED_AS is not set
+# CONFIG_IOSCHED_DEADLINE is not set
+# CONFIG_IOSCHED_CFQ is not set
+# CONFIG_DEFAULT_AS is not set
+# CONFIG_DEFAULT_DEADLINE is not set
+# CONFIG_DEFAULT_CFQ is not set
+CONFIG_DEFAULT_NOOP=y
+CONFIG_DEFAULT_IOSCHED="noop"
+CONFIG_CLASSIC_RCU=y
+
+#
+# Processor type and features
+#
+# CONFIG_M68328 is not set
+# CONFIG_M68EZ328 is not set
+# CONFIG_M68VZ328 is not set
+# CONFIG_M68360 is not set
+# CONFIG_M5206 is not set
+# CONFIG_M5206e is not set
+# CONFIG_M520x is not set
+# CONFIG_M523x is not set
+# CONFIG_M5249 is not set
+# CONFIG_M5271 is not set
+# CONFIG_M5272 is not set
+# CONFIG_M5275 is not set
+# CONFIG_M528x is not set
+CONFIG_M5307=y
+# CONFIG_M532x is not set
+# CONFIG_M5407 is not set
+CONFIG_COLDFIRE=y
+CONFIG_CLOCK_SET=y
+CONFIG_CLOCK_FREQ=90000000
+CONFIG_CLOCK_DIV=2
+# CONFIG_OLDMASK is not set
+
+#
+# Platform
+#
+# CONFIG_ARN5307 is not set
+CONFIG_M5307C3=y
+# CONFIG_eLIA is not set
+# CONFIG_SECUREEDGEMP3 is not set
+# CONFIG_CLEOPATRA is not set
+# CONFIG_NETtel is not set
+CONFIG_FREESCALE=y
+# CONFIG_4KSTACKS is not set
+CONFIG_HZ=100
+
+#
+# RAM configuration
+#
+CONFIG_RAMBASE=0x00000000
+CONFIG_RAMSIZE=0x00800000
+CONFIG_VECTORBASE=0x00000000
+CONFIG_KERNELBASE=0x00020000
+CONFIG_RAMAUTOBIT=y
+# CONFIG_RAM8BIT is not set
+# CONFIG_RAM16BIT is not set
+# CONFIG_RAM32BIT is not set
+
+#
+# ROM configuration
+#
+# CONFIG_ROM is not set
+CONFIG_RAMKERNEL=y
+# CONFIG_ROMKERNEL is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+# CONFIG_SPARSEMEM_STATIC is not set
+# CONFIG_SPARSEMEM_VMEMMAP_ENABLE is not set
+CONFIG_PAGEFLAGS_EXTENDED=y
+CONFIG_SPLIT_PTLOCK_CPUS=4
+# CONFIG_RESOURCES_64BIT is not set
+CONFIG_ZONE_DMA_FLAG=1
+CONFIG_VIRT_TO_BUS=y
+CONFIG_ISA_DMA_API=y
+
+#
+# Bus options (PCI, PCMCIA, EISA, MCA, ISA)
+#
+# CONFIG_PCI is not set
+# CONFIG_COMEMPCI is not set
+# CONFIG_ARCH_SUPPORTS_MSI is not set
+
+#
+# Executable file formats
+#
+CONFIG_BINFMT_FLAT=y
+# CONFIG_BINFMT_ZFLAT is not set
+# CONFIG_BINFMT_SHARED_FLAT is not set
+# CONFIG_BINFMT_AOUT is not set
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Power management options
+#
+# CONFIG_PM is not set
+
+#
+# Networking
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+CONFIG_PACKET=y
+# CONFIG_PACKET_MMAP is not set
+CONFIG_UNIX=y
+# CONFIG_NET_KEY is not set
+CONFIG_INET=y
+# CONFIG_IP_MULTICAST is not set
+# CONFIG_IP_ADVANCED_ROUTER is not set
+CONFIG_IP_FIB_HASH=y
+# CONFIG_IP_PNP is not set
+# CONFIG_NET_IPIP is not set
+# CONFIG_NET_IPGRE is not set
+# CONFIG_ARPD is not set
+# CONFIG_SYN_COOKIES is not set
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_XFRM_TUNNEL is not set
+# CONFIG_INET_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
+# CONFIG_INET_LRO is not set
+# CONFIG_INET_DIAG is not set
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_CUBIC=y
+CONFIG_DEFAULT_TCP_CONG="cubic"
+# CONFIG_TCP_MD5SIG is not set
+# CONFIG_IPV6 is not set
+# CONFIG_NETWORK_SECMARK is not set
+# CONFIG_NETFILTER is not set
+# CONFIG_IP_DCCP is not set
+# CONFIG_IP_SCTP is not set
+# CONFIG_TIPC is not set
+# CONFIG_ATM is not set
+# CONFIG_BRIDGE is not set
+# CONFIG_VLAN_8021Q is not set
+# CONFIG_DECNET is not set
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+# CONFIG_NET_SCHED is not set
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_CAN is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+# CONFIG_AF_RXRPC is not set
+
+#
+# Wireless
+#
+# CONFIG_CFG80211 is not set
+# CONFIG_WIRELESS_EXT is not set
+# CONFIG_MAC80211 is not set
+# CONFIG_IEEE80211 is not set
+# CONFIG_RFKILL is not set
+# CONFIG_NET_9P is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+# CONFIG_SYS_HYPERVISOR is not set
+# CONFIG_CONNECTOR is not set
+# CONFIG_MTD is not set
+# CONFIG_PARPORT is not set
+CONFIG_BLK_DEV=y
+# CONFIG_BLK_DEV_COW_COMMON is not set
+# CONFIG_BLK_DEV_LOOP is not set
+# CONFIG_BLK_DEV_NBD is not set
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_COUNT=16
+CONFIG_BLK_DEV_RAM_SIZE=4096
+# CONFIG_BLK_DEV_XIP is not set
+# CONFIG_CDROM_PKTCDVD is not set
+# CONFIG_ATA_OVER_ETH is not set
+# CONFIG_MISC_DEVICES is not set
+CONFIG_HAVE_IDE=y
+# CONFIG_IDE is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+# CONFIG_SCSI is not set
+# CONFIG_SCSI_DMA is not set
+# CONFIG_SCSI_NETLINK is not set
+# CONFIG_MD is not set
+CONFIG_NETDEVICES=y
+# CONFIG_NETDEVICES_MULTIQUEUE is not set
+# CONFIG_DUMMY is not set
+# CONFIG_BONDING is not set
+# CONFIG_MACVLAN is not set
+# CONFIG_EQUALIZER is not set
+# CONFIG_TUN is not set
+# CONFIG_VETH is not set
+# CONFIG_PHYLIB is not set
+CONFIG_NET_ETHERNET=y
+# CONFIG_MII is not set
+# CONFIG_IBM_NEW_EMAC_ZMII is not set
+# CONFIG_IBM_NEW_EMAC_RGMII is not set
+# CONFIG_IBM_NEW_EMAC_TAH is not set
+# CONFIG_IBM_NEW_EMAC_EMAC4 is not set
+# CONFIG_B44 is not set
+# CONFIG_NETDEV_1000 is not set
+# CONFIG_NETDEV_10000 is not set
+
+#
+# Wireless LAN
+#
+# CONFIG_WLAN_PRE80211 is not set
+# CONFIG_WLAN_80211 is not set
+# CONFIG_IWLWIFI is not set
+# CONFIG_IWLWIFI_LEDS is not set
+# CONFIG_WAN is not set
+CONFIG_PPP=y
+# CONFIG_PPP_MULTILINK is not set
+# CONFIG_PPP_FILTER is not set
+# CONFIG_PPP_ASYNC is not set
+# CONFIG_PPP_SYNC_TTY is not set
+# CONFIG_PPP_DEFLATE is not set
+# CONFIG_PPP_BSDCOMP is not set
+# CONFIG_PPP_MPPE is not set
+# CONFIG_PPPOE is not set
+# CONFIG_PPPOL2TP is not set
+CONFIG_SLIP=y
+CONFIG_SLIP_COMPRESSED=y
+CONFIG_SLHC=y
+# CONFIG_SLIP_SMART is not set
+# CONFIG_SLIP_MODE_SLIP6 is not set
+# CONFIG_NETCONSOLE is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+# CONFIG_ISDN is not set
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+CONFIG_INPUT=y
+# CONFIG_INPUT_FF_MEMLESS is not set
+# CONFIG_INPUT_POLLDEV is not set
+
+#
+# Userland interfaces
+#
+# CONFIG_INPUT_MOUSEDEV is not set
+# CONFIG_INPUT_JOYDEV is not set
+# CONFIG_INPUT_EVDEV is not set
+# CONFIG_INPUT_EVBUG is not set
+
+#
+# Input Device Drivers
+#
+# CONFIG_INPUT_KEYBOARD is not set
+# CONFIG_INPUT_MOUSE is not set
+# CONFIG_INPUT_JOYSTICK is not set
+# CONFIG_INPUT_TABLET is not set
+# CONFIG_INPUT_TOUCHSCREEN is not set
+# CONFIG_INPUT_MISC is not set
+
+#
+# Hardware I/O ports
+#
+# CONFIG_SERIO is not set
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+# CONFIG_VT is not set
+# CONFIG_DEVKMEM is not set
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_COLDFIRE is not set
+CONFIG_SERIAL_MCF=y
+CONFIG_SERIAL_MCF_BAUDRATE=19200
+CONFIG_SERIAL_MCF_CONSOLE=y
+CONFIG_UNIX98_PTYS=y
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=256
+# CONFIG_IPMI_HANDLER is not set
+# CONFIG_HW_RANDOM is not set
+# CONFIG_GEN_RTC is not set
+# CONFIG_R3964 is not set
+# CONFIG_RAW_DRIVER is not set
+# CONFIG_TCG_TPM is not set
+# CONFIG_I2C is not set
+# CONFIG_SPI is not set
+# CONFIG_W1 is not set
+# CONFIG_POWER_SUPPLY is not set
+# CONFIG_HWMON is not set
+# CONFIG_THERMAL is not set
+# CONFIG_WATCHDOG is not set
+
+#
+# Sonics Silicon Backplane
+#
+CONFIG_SSB_POSSIBLE=y
+# CONFIG_SSB is not set
+
+#
+# Multifunction device drivers
+#
+# CONFIG_MFD_SM501 is not set
+# CONFIG_HTC_PASIC3 is not set
+
+#
+# Multimedia devices
+#
+
+#
+# Multimedia core support
+#
+# CONFIG_VIDEO_DEV is not set
+# CONFIG_DVB_CORE is not set
+
+#
+# Multimedia drivers
+#
+CONFIG_DAB=y
+
+#
+# Graphics support
+#
+# CONFIG_VGASTATE is not set
+# CONFIG_VIDEO_OUTPUT_CONTROL is not set
+# CONFIG_FB is not set
+# CONFIG_BACKLIGHT_LCD_SUPPORT is not set
+
+#
+# Display device support
+#
+# CONFIG_DISPLAY_SUPPORT is not set
+
+#
+# Sound
+#
+# CONFIG_SOUND is not set
+# CONFIG_HID_SUPPORT is not set
+# CONFIG_USB_SUPPORT is not set
+# CONFIG_MMC is not set
+# CONFIG_MEMSTICK is not set
+# CONFIG_NEW_LEDS is not set
+# CONFIG_ACCESSIBILITY is not set
+# CONFIG_RTC_CLASS is not set
+# CONFIG_UIO is not set
+
+#
+# File systems
+#
+CONFIG_EXT2_FS=y
+# CONFIG_EXT2_FS_XATTR is not set
+# CONFIG_EXT3_FS is not set
+# CONFIG_EXT4DEV_FS is not set
+# CONFIG_REISERFS_FS is not set
+# CONFIG_JFS_FS is not set
+# CONFIG_FS_POSIX_ACL is not set
+# CONFIG_XFS_FS is not set
+# CONFIG_OCFS2_FS is not set
+# CONFIG_DNOTIFY is not set
+# CONFIG_INOTIFY is not set
+# CONFIG_QUOTA is not set
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+# CONFIG_FUSE_FS is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+# CONFIG_ISO9660_FS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+# CONFIG_MSDOS_FS is not set
+# CONFIG_VFAT_FS is not set
+# CONFIG_NTFS_FS is not set
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+CONFIG_PROC_SYSCTL=y
+CONFIG_SYSFS=y
+# CONFIG_TMPFS is not set
+# CONFIG_HUGETLB_PAGE is not set
+# CONFIG_CONFIGFS_FS is not set
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+# CONFIG_CRAMFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_MINIX_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+CONFIG_ROMFS_FS=y
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+# CONFIG_NETWORK_FILESYSTEMS is not set
+
+#
+# Partition Types
+#
+# CONFIG_PARTITION_ADVANCED is not set
+CONFIG_MSDOS_PARTITION=y
+# CONFIG_NLS is not set
+# CONFIG_DLM is not set
+
+#
+# Kernel hacking
+#
+# CONFIG_PRINTK_TIME is not set
+CONFIG_ENABLE_WARN_DEPRECATED=y
+CONFIG_ENABLE_MUST_CHECK=y
+CONFIG_FRAME_WARN=1024
+# CONFIG_MAGIC_SYSRQ is not set
+# CONFIG_UNUSED_SYMBOLS is not set
+# CONFIG_DEBUG_FS is not set
+# CONFIG_HEADERS_CHECK is not set
+# CONFIG_DEBUG_KERNEL is not set
+# CONFIG_DEBUG_BUGVERBOSE is not set
+# CONFIG_SAMPLES is not set
+CONFIG_FULLDEBUG=y
+# CONFIG_HIGHPROFILE is not set
+# CONFIG_BOOTPARAM is not set
+# CONFIG_NO_KERNEL_MSG is not set
+# CONFIG_BDM_DISABLE is not set
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+# CONFIG_SECURITY is not set
+# CONFIG_SECURITY_FILE_CAPABILITIES is not set
+# CONFIG_CRYPTO is not set
+
+#
+# Library routines
+#
+# CONFIG_GENERIC_FIND_FIRST_BIT is not set
+# CONFIG_CRC_CCITT is not set
+# CONFIG_CRC16 is not set
+# CONFIG_CRC_ITU_T is not set
+# CONFIG_CRC32 is not set
+# CONFIG_CRC7 is not set
+# CONFIG_LIBCRC32C is not set
+CONFIG_HAS_IOMEM=y
+CONFIG_HAS_DMA=y

From ab88e474c8ffa300660f03a8e6b08ea660956bef Mon Sep 17 00:00:00 2001
From: Greg Ungerer <gerg@goober.(none)>
Date: Wed, 4 Jun 2008 21:26:38 +1000
Subject: [PATCH 0122/1171] m68knommu: defconfig for M5407C3 board

Add a defconfig for the Freescale M5407C3 board.

Signed-off-by: Greg Ungerer <gerg@uclinux.org>
---
 arch/m68knommu/configs/m5407c3_defconfig | 641 +++++++++++++++++++++++
 1 file changed, 641 insertions(+)
 create mode 100644 arch/m68knommu/configs/m5407c3_defconfig

diff --git a/arch/m68knommu/configs/m5407c3_defconfig b/arch/m68knommu/configs/m5407c3_defconfig
new file mode 100644
index 000000000000..1118936d20e3
--- /dev/null
+++ b/arch/m68knommu/configs/m5407c3_defconfig
@@ -0,0 +1,641 @@
+#
+# Automatically generated make config: don't edit
+# Linux kernel version: 2.6.26-rc1
+# Wed May  7 10:25:16 2008
+#
+CONFIG_M68K=y
+# CONFIG_MMU is not set
+# CONFIG_FPU is not set
+CONFIG_ZONE_DMA=y
+CONFIG_RWSEM_GENERIC_SPINLOCK=y
+# CONFIG_RWSEM_XCHGADD_ALGORITHM is not set
+# CONFIG_ARCH_HAS_ILOG2_U32 is not set
+# CONFIG_ARCH_HAS_ILOG2_U64 is not set
+CONFIG_GENERIC_FIND_NEXT_BIT=y
+CONFIG_GENERIC_HWEIGHT=y
+CONFIG_GENERIC_HARDIRQS=y
+CONFIG_GENERIC_CALIBRATE_DELAY=y
+CONFIG_GENERIC_TIME=y
+CONFIG_TIME_LOW_RES=y
+CONFIG_NO_IOPORT=y
+CONFIG_ARCH_SUPPORTS_AOUT=y
+CONFIG_DEFCONFIG_LIST="/lib/modules/$UNAME_RELEASE/.config"
+
+#
+# General setup
+#
+CONFIG_EXPERIMENTAL=y
+CONFIG_BROKEN_ON_SMP=y
+CONFIG_INIT_ENV_ARG_LIMIT=32
+CONFIG_LOCALVERSION=""
+CONFIG_LOCALVERSION_AUTO=y
+# CONFIG_SYSVIPC is not set
+# CONFIG_POSIX_MQUEUE is not set
+# CONFIG_BSD_PROCESS_ACCT is not set
+# CONFIG_TASKSTATS is not set
+# CONFIG_AUDIT is not set
+# CONFIG_IKCONFIG is not set
+CONFIG_LOG_BUF_SHIFT=14
+# CONFIG_CGROUPS is not set
+# CONFIG_GROUP_SCHED is not set
+# CONFIG_SYSFS_DEPRECATED_V2 is not set
+# CONFIG_RELAY is not set
+# CONFIG_NAMESPACES is not set
+# CONFIG_BLK_DEV_INITRD is not set
+# CONFIG_CC_OPTIMIZE_FOR_SIZE is not set
+CONFIG_SYSCTL=y
+CONFIG_EMBEDDED=y
+# CONFIG_UID16 is not set
+# CONFIG_SYSCTL_SYSCALL is not set
+# CONFIG_KALLSYMS is not set
+# CONFIG_HOTPLUG is not set
+CONFIG_PRINTK=y
+CONFIG_BUG=y
+CONFIG_ELF_CORE=y
+# CONFIG_COMPAT_BRK is not set
+CONFIG_BASE_FULL=y
+# CONFIG_FUTEX is not set
+# CONFIG_EPOLL is not set
+# CONFIG_SIGNALFD is not set
+# CONFIG_TIMERFD is not set
+# CONFIG_EVENTFD is not set
+# CONFIG_VM_EVENT_COUNTERS is not set
+CONFIG_SLAB=y
+# CONFIG_SLUB is not set
+# CONFIG_SLOB is not set
+# CONFIG_PROFILING is not set
+# CONFIG_MARKERS is not set
+# CONFIG_HAVE_OPROFILE is not set
+# CONFIG_HAVE_KPROBES is not set
+# CONFIG_HAVE_KRETPROBES is not set
+# CONFIG_HAVE_DMA_ATTRS is not set
+CONFIG_SLABINFO=y
+CONFIG_TINY_SHMEM=y
+CONFIG_BASE_SMALL=0
+CONFIG_MODULES=y
+CONFIG_MODULE_UNLOAD=y
+# CONFIG_MODULE_FORCE_UNLOAD is not set
+# CONFIG_MODVERSIONS is not set
+# CONFIG_MODULE_SRCVERSION_ALL is not set
+# CONFIG_KMOD is not set
+CONFIG_BLOCK=y
+# CONFIG_LBD is not set
+# CONFIG_BLK_DEV_IO_TRACE is not set
+# CONFIG_LSF is not set
+# CONFIG_BLK_DEV_BSG is not set
+
+#
+# IO Schedulers
+#
+CONFIG_IOSCHED_NOOP=y
+# CONFIG_IOSCHED_AS is not set
+# CONFIG_IOSCHED_DEADLINE is not set
+# CONFIG_IOSCHED_CFQ is not set
+# CONFIG_DEFAULT_AS is not set
+# CONFIG_DEFAULT_DEADLINE is not set
+# CONFIG_DEFAULT_CFQ is not set
+CONFIG_DEFAULT_NOOP=y
+CONFIG_DEFAULT_IOSCHED="noop"
+CONFIG_CLASSIC_RCU=y
+
+#
+# Processor type and features
+#
+# CONFIG_M68328 is not set
+# CONFIG_M68EZ328 is not set
+# CONFIG_M68VZ328 is not set
+# CONFIG_M68360 is not set
+# CONFIG_M5206 is not set
+# CONFIG_M5206e is not set
+# CONFIG_M520x is not set
+# CONFIG_M523x is not set
+# CONFIG_M5249 is not set
+# CONFIG_M5271 is not set
+# CONFIG_M5272 is not set
+# CONFIG_M5275 is not set
+# CONFIG_M528x is not set
+# CONFIG_M5307 is not set
+# CONFIG_M532x is not set
+CONFIG_M5407=y
+CONFIG_COLDFIRE=y
+CONFIG_CLOCK_SET=y
+CONFIG_CLOCK_FREQ=50000000
+CONFIG_CLOCK_DIV=1
+
+#
+# Platform
+#
+CONFIG_M5407C3=y
+# CONFIG_CLEOPATRA is not set
+CONFIG_FREESCALE=y
+CONFIG_4KSTACKS=y
+CONFIG_HZ=100
+
+#
+# RAM configuration
+#
+CONFIG_RAMBASE=0x00000000
+CONFIG_RAMSIZE=0x00000000
+CONFIG_VECTORBASE=0x00000000
+CONFIG_KERNELBASE=0x00020000
+CONFIG_RAMAUTOBIT=y
+# CONFIG_RAM8BIT is not set
+# CONFIG_RAM16BIT is not set
+# CONFIG_RAM32BIT is not set
+
+#
+# ROM configuration
+#
+# CONFIG_ROM is not set
+CONFIG_RAMKERNEL=y
+# CONFIG_ROMKERNEL is not set
+CONFIG_SELECT_MEMORY_MODEL=y
+CONFIG_FLATMEM_MANUAL=y
+# CONFIG_DISCONTIGMEM_MANUAL is not set
+# CONFIG_SPARSEMEM_MANUAL is not set
+CONFIG_FLATMEM=y
+CONFIG_FLAT_NODE_MEM_MAP=y
+# CONFIG_SPARSEMEM_STATIC is not set
+# CONFIG_SPARSEMEM_VMEMMAP_ENABLE is not set
+CONFIG_PAGEFLAGS_EXTENDED=y
+CONFIG_SPLIT_PTLOCK_CPUS=4
+# CONFIG_RESOURCES_64BIT is not set
+CONFIG_ZONE_DMA_FLAG=1
+CONFIG_VIRT_TO_BUS=y
+CONFIG_ISA_DMA_API=y
+
+#
+# Bus options (PCI, PCMCIA, EISA, MCA, ISA)
+#
+# CONFIG_PCI is not set
+# CONFIG_COMEMPCI is not set
+# CONFIG_ARCH_SUPPORTS_MSI is not set
+
+#
+# Executable file formats
+#
+CONFIG_BINFMT_FLAT=y
+# CONFIG_BINFMT_ZFLAT is not set
+# CONFIG_BINFMT_SHARED_FLAT is not set
+# CONFIG_BINFMT_AOUT is not set
+# CONFIG_BINFMT_MISC is not set
+
+#
+# Power management options
+#
+# CONFIG_PM is not set
+
+#
+# Networking
+#
+CONFIG_NET=y
+
+#
+# Networking options
+#
+CONFIG_PACKET=y
+# CONFIG_PACKET_MMAP is not set
+CONFIG_UNIX=y
+# CONFIG_NET_KEY is not set
+CONFIG_INET=y
+# CONFIG_IP_MULTICAST is not set
+# CONFIG_IP_ADVANCED_ROUTER is not set
+CONFIG_IP_FIB_HASH=y
+# CONFIG_IP_PNP is not set
+# CONFIG_NET_IPIP is not set
+# CONFIG_NET_IPGRE is not set
+# CONFIG_ARPD is not set
+# CONFIG_SYN_COOKIES is not set
+# CONFIG_INET_AH is not set
+# CONFIG_INET_ESP is not set
+# CONFIG_INET_IPCOMP is not set
+# CONFIG_INET_XFRM_TUNNEL is not set
+# CONFIG_INET_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_TRANSPORT is not set
+# CONFIG_INET_XFRM_MODE_TUNNEL is not set
+# CONFIG_INET_XFRM_MODE_BEET is not set
+# CONFIG_INET_LRO is not set
+# CONFIG_INET_DIAG is not set
+# CONFIG_TCP_CONG_ADVANCED is not set
+CONFIG_TCP_CONG_CUBIC=y
+CONFIG_DEFAULT_TCP_CONG="cubic"
+# CONFIG_TCP_MD5SIG is not set
+# CONFIG_IPV6 is not set
+# CONFIG_NETWORK_SECMARK is not set
+# CONFIG_NETFILTER is not set
+# CONFIG_IP_DCCP is not set
+# CONFIG_IP_SCTP is not set
+# CONFIG_TIPC is not set
+# CONFIG_ATM is not set
+# CONFIG_BRIDGE is not set
+# CONFIG_VLAN_8021Q is not set
+# CONFIG_DECNET is not set
+# CONFIG_LLC2 is not set
+# CONFIG_IPX is not set
+# CONFIG_ATALK is not set
+# CONFIG_X25 is not set
+# CONFIG_LAPB is not set
+# CONFIG_ECONET is not set
+# CONFIG_WAN_ROUTER is not set
+# CONFIG_NET_SCHED is not set
+
+#
+# Network testing
+#
+# CONFIG_NET_PKTGEN is not set
+# CONFIG_HAMRADIO is not set
+# CONFIG_CAN is not set
+# CONFIG_IRDA is not set
+# CONFIG_BT is not set
+# CONFIG_AF_RXRPC is not set
+
+#
+# Wireless
+#
+# CONFIG_CFG80211 is not set
+# CONFIG_WIRELESS_EXT is not set
+# CONFIG_MAC80211 is not set
+# CONFIG_IEEE80211 is not set
+# CONFIG_RFKILL is not set
+# CONFIG_NET_9P is not set
+
+#
+# Device Drivers
+#
+
+#
+# Generic Driver Options
+#
+CONFIG_STANDALONE=y
+CONFIG_PREVENT_FIRMWARE_BUILD=y
+# CONFIG_SYS_HYPERVISOR is not set
+# CONFIG_CONNECTOR is not set
+CONFIG_MTD=y
+# CONFIG_MTD_DEBUG is not set
+# CONFIG_MTD_CONCAT is not set
+CONFIG_MTD_PARTITIONS=y
+# CONFIG_MTD_REDBOOT_PARTS is not set
+# CONFIG_MTD_CMDLINE_PARTS is not set
+# CONFIG_MTD_AR7_PARTS is not set
+
+#
+# User Modules And Translation Layers
+#
+CONFIG_MTD_CHAR=y
+CONFIG_MTD_BLKDEVS=y
+CONFIG_MTD_BLOCK=y
+# CONFIG_FTL is not set
+# CONFIG_NFTL is not set
+# CONFIG_INFTL is not set
+# CONFIG_RFD_FTL is not set
+# CONFIG_SSFDC is not set
+# CONFIG_MTD_OOPS is not set
+
+#
+# RAM/ROM/Flash chip drivers
+#
+# CONFIG_MTD_CFI is not set
+# CONFIG_MTD_JEDECPROBE is not set
+CONFIG_MTD_MAP_BANK_WIDTH_1=y
+CONFIG_MTD_MAP_BANK_WIDTH_2=y
+CONFIG_MTD_MAP_BANK_WIDTH_4=y
+# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set
+# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set
+CONFIG_MTD_CFI_I1=y
+CONFIG_MTD_CFI_I2=y
+# CONFIG_MTD_CFI_I4 is not set
+# CONFIG_MTD_CFI_I8 is not set
+CONFIG_MTD_RAM=y
+# CONFIG_MTD_ROM is not set
+# CONFIG_MTD_ABSENT is not set
+
+#
+# Mapping drivers for chip access
+#
+# CONFIG_MTD_COMPLEX_MAPPINGS is not set
+CONFIG_MTD_UCLINUX=y
+# CONFIG_MTD_PLATRAM is not set
+
+#
+# Self-contained MTD device drivers
+#
+# CONFIG_MTD_SLRAM is not set
+# CONFIG_MTD_PHRAM is not set
+# CONFIG_MTD_MTDRAM is not set
+# CONFIG_MTD_BLOCK2MTD is not set
+
+#
+# Disk-On-Chip Device Drivers
+#
+# CONFIG_MTD_DOC2000 is not set
+# CONFIG_MTD_DOC2001 is not set
+# CONFIG_MTD_DOC2001PLUS is not set
+# CONFIG_MTD_NAND is not set
+# CONFIG_MTD_ONENAND is not set
+
+#
+# UBI - Unsorted block images
+#
+# CONFIG_MTD_UBI is not set
+# CONFIG_PARPORT is not set
+CONFIG_BLK_DEV=y
+# CONFIG_BLK_DEV_COW_COMMON is not set
+# CONFIG_BLK_DEV_LOOP is not set
+# CONFIG_BLK_DEV_NBD is not set
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_COUNT=16
+CONFIG_BLK_DEV_RAM_SIZE=4096
+# CONFIG_BLK_DEV_XIP is not set
+# CONFIG_CDROM_PKTCDVD is not set
+# CONFIG_ATA_OVER_ETH is not set
+# CONFIG_MISC_DEVICES is not set
+CONFIG_HAVE_IDE=y
+# CONFIG_IDE is not set
+
+#
+# SCSI device support
+#
+# CONFIG_RAID_ATTRS is not set
+# CONFIG_SCSI is not set
+# CONFIG_SCSI_DMA is not set
+# CONFIG_SCSI_NETLINK is not set
+# CONFIG_MD is not set
+CONFIG_NETDEVICES=y
+# CONFIG_NETDEVICES_MULTIQUEUE is not set
+# CONFIG_DUMMY is not set
+# CONFIG_BONDING is not set
+# CONFIG_MACVLAN is not set
+# CONFIG_EQUALIZER is not set
+# CONFIG_TUN is not set
+# CONFIG_VETH is not set
+# CONFIG_PHYLIB is not set
+CONFIG_NET_ETHERNET=y
+# CONFIG_MII is not set
+# CONFIG_IBM_NEW_EMAC_ZMII is not set
+# CONFIG_IBM_NEW_EMAC_RGMII is not set
+# CONFIG_IBM_NEW_EMAC_TAH is not set
+# CONFIG_IBM_NEW_EMAC_EMAC4 is not set
+# CONFIG_B44 is not set
+# CONFIG_NETDEV_1000 is not set
+# CONFIG_NETDEV_10000 is not set
+
+#
+# Wireless LAN
+#
+# CONFIG_WLAN_PRE80211 is not set
+# CONFIG_WLAN_80211 is not set
+# CONFIG_IWLWIFI is not set
+# CONFIG_IWLWIFI_LEDS is not set
+# CONFIG_WAN is not set
+CONFIG_PPP=y
+# CONFIG_PPP_MULTILINK is not set
+# CONFIG_PPP_FILTER is not set
+# CONFIG_PPP_ASYNC is not set
+# CONFIG_PPP_SYNC_TTY is not set
+# CONFIG_PPP_DEFLATE is not set
+# CONFIG_PPP_BSDCOMP is not set
+# CONFIG_PPP_MPPE is not set
+# CONFIG_PPPOE is not set
+# CONFIG_PPPOL2TP is not set
+# CONFIG_SLIP is not set
+CONFIG_SLHC=y
+# CONFIG_NETCONSOLE is not set
+# CONFIG_NETPOLL is not set
+# CONFIG_NET_POLL_CONTROLLER is not set
+# CONFIG_ISDN is not set
+# CONFIG_PHONE is not set
+
+#
+# Input device support
+#
+CONFIG_INPUT=y
+# CONFIG_INPUT_FF_MEMLESS is not set
+# CONFIG_INPUT_POLLDEV is not set
+
+#
+# Userland interfaces
+#
+# CONFIG_INPUT_MOUSEDEV is not set
+# CONFIG_INPUT_JOYDEV is not set
+# CONFIG_INPUT_EVDEV is not set
+# CONFIG_INPUT_EVBUG is not set
+
+#
+# Input Device Drivers
+#
+# CONFIG_INPUT_KEYBOARD is not set
+# CONFIG_INPUT_MOUSE is not set
+# CONFIG_INPUT_JOYSTICK is not set
+# CONFIG_INPUT_TABLET is not set
+# CONFIG_INPUT_TOUCHSCREEN is not set
+# CONFIG_INPUT_MISC is not set
+
+#
+# Hardware I/O ports
+#
+# CONFIG_SERIO is not set
+# CONFIG_GAMEPORT is not set
+
+#
+# Character devices
+#
+# CONFIG_VT is not set
+# CONFIG_DEVKMEM is not set
+# CONFIG_SERIAL_NONSTANDARD is not set
+
+#
+# Serial drivers
+#
+# CONFIG_SERIAL_8250 is not set
+
+#
+# Non-8250 serial port support
+#
+CONFIG_SERIAL_CORE=y
+CONFIG_SERIAL_CORE_CONSOLE=y
+# CONFIG_SERIAL_COLDFIRE is not set
+CONFIG_SERIAL_MCF=y
+CONFIG_SERIAL_MCF_BAUDRATE=19200
+CONFIG_SERIAL_MCF_CONSOLE=y
+# CONFIG_UNIX98_PTYS is not set
+CONFIG_LEGACY_PTYS=y
+CONFIG_LEGACY_PTY_COUNT=256
+# CONFIG_IPMI_HANDLER is not set
+# CONFIG_HW_RANDOM is not set
+# CONFIG_GEN_RTC is not set
+# CONFIG_R3964 is not set
+# CONFIG_RAW_DRIVER is not set
+# CONFIG_TCG_TPM is not set
+# CONFIG_I2C is not set
+# CONFIG_SPI is not set
+# CONFIG_W1 is not set
+# CONFIG_POWER_SUPPLY is not set
+# CONFIG_HWMON is not set
+# CONFIG_THERMAL is not set
+# CONFIG_WATCHDOG is not set
+
+#
+# Sonics Silicon Backplane
+#
+CONFIG_SSB_POSSIBLE=y
+# CONFIG_SSB is not set
+
+#
+# Multifunction device drivers
+#
+# CONFIG_MFD_SM501 is not set
+# CONFIG_HTC_PASIC3 is not set
+
+#
+# Multimedia devices
+#
+
+#
+# Multimedia core support
+#
+# CONFIG_VIDEO_DEV is not set
+# CONFIG_DVB_CORE is not set
+
+#
+# Multimedia drivers
+#
+CONFIG_DAB=y
+
+#
+# Graphics support
+#
+# CONFIG_VGASTATE is not set
+# CONFIG_VIDEO_OUTPUT_CONTROL is not set
+# CONFIG_FB is not set
+# CONFIG_BACKLIGHT_LCD_SUPPORT is not set
+
+#
+# Display device support
+#
+# CONFIG_DISPLAY_SUPPORT is not set
+
+#
+# Sound
+#
+# CONFIG_SOUND is not set
+# CONFIG_HID_SUPPORT is not set
+# CONFIG_USB_SUPPORT is not set
+# CONFIG_MMC is not set
+# CONFIG_MEMSTICK is not set
+# CONFIG_NEW_LEDS is not set
+# CONFIG_ACCESSIBILITY is not set
+# CONFIG_RTC_CLASS is not set
+# CONFIG_UIO is not set
+
+#
+# File systems
+#
+CONFIG_EXT2_FS=y
+# CONFIG_EXT2_FS_XATTR is not set
+# CONFIG_EXT3_FS is not set
+# CONFIG_EXT4DEV_FS is not set
+# CONFIG_REISERFS_FS is not set
+# CONFIG_JFS_FS is not set
+# CONFIG_FS_POSIX_ACL is not set
+# CONFIG_XFS_FS is not set
+# CONFIG_OCFS2_FS is not set
+# CONFIG_DNOTIFY is not set
+# CONFIG_INOTIFY is not set
+# CONFIG_QUOTA is not set
+# CONFIG_AUTOFS_FS is not set
+# CONFIG_AUTOFS4_FS is not set
+# CONFIG_FUSE_FS is not set
+
+#
+# CD-ROM/DVD Filesystems
+#
+# CONFIG_ISO9660_FS is not set
+# CONFIG_UDF_FS is not set
+
+#
+# DOS/FAT/NT Filesystems
+#
+# CONFIG_MSDOS_FS is not set
+# CONFIG_VFAT_FS is not set
+# CONFIG_NTFS_FS is not set
+
+#
+# Pseudo filesystems
+#
+CONFIG_PROC_FS=y
+CONFIG_PROC_SYSCTL=y
+CONFIG_SYSFS=y
+# CONFIG_TMPFS is not set
+# CONFIG_HUGETLB_PAGE is not set
+# CONFIG_CONFIGFS_FS is not set
+
+#
+# Miscellaneous filesystems
+#
+# CONFIG_ADFS_FS is not set
+# CONFIG_AFFS_FS is not set
+# CONFIG_HFS_FS is not set
+# CONFIG_HFSPLUS_FS is not set
+# CONFIG_BEFS_FS is not set
+# CONFIG_BFS_FS is not set
+# CONFIG_EFS_FS is not set
+# CONFIG_JFFS2_FS is not set
+# CONFIG_CRAMFS is not set
+# CONFIG_VXFS_FS is not set
+# CONFIG_MINIX_FS is not set
+# CONFIG_HPFS_FS is not set
+# CONFIG_QNX4FS_FS is not set
+CONFIG_ROMFS_FS=y
+# CONFIG_SYSV_FS is not set
+# CONFIG_UFS_FS is not set
+# CONFIG_NETWORK_FILESYSTEMS is not set
+
+#
+# Partition Types
+#
+# CONFIG_PARTITION_ADVANCED is not set
+CONFIG_MSDOS_PARTITION=y
+# CONFIG_NLS is not set
+# CONFIG_DLM is not set
+
+#
+# Kernel hacking
+#
+# CONFIG_PRINTK_TIME is not set
+CONFIG_ENABLE_WARN_DEPRECATED=y
+CONFIG_ENABLE_MUST_CHECK=y
+CONFIG_FRAME_WARN=1024
+# CONFIG_MAGIC_SYSRQ is not set
+# CONFIG_UNUSED_SYMBOLS is not set
+# CONFIG_DEBUG_FS is not set
+# CONFIG_HEADERS_CHECK is not set
+# CONFIG_DEBUG_KERNEL is not set
+# CONFIG_DEBUG_BUGVERBOSE is not set
+# CONFIG_SAMPLES is not set
+# CONFIG_FULLDEBUG is not set
+# CONFIG_HIGHPROFILE is not set
+# CONFIG_BOOTPARAM is not set
+# CONFIG_NO_KERNEL_MSG is not set
+# CONFIG_BDM_DISABLE is not set
+
+#
+# Security options
+#
+# CONFIG_KEYS is not set
+# CONFIG_SECURITY is not set
+# CONFIG_SECURITY_FILE_CAPABILITIES is not set
+# CONFIG_CRYPTO is not set
+
+#
+# Library routines
+#
+# CONFIG_GENERIC_FIND_FIRST_BIT is not set
+# CONFIG_CRC_CCITT is not set
+# CONFIG_CRC16 is not set
+# CONFIG_CRC_ITU_T is not set
+# CONFIG_CRC32 is not set
+# CONFIG_CRC7 is not set
+# CONFIG_LIBCRC32C is not set
+CONFIG_HAS_IOMEM=y
+CONFIG_HAS_DMA=y

From 5732b38ddb770b98110ea218232fc072e5626b87 Mon Sep 17 00:00:00 2001
From: Sebastian Siewior <bigeasy@linutronix.de>
Date: Fri, 9 May 2008 16:18:33 +0200
Subject: [PATCH 0123/1171] m68knommu: Add Coldfire DMA Timer support

This one could be used as a hrtimer.

Signed-off-by: Benedikt Spranger <b.spranger@linutronix.de>
Signed-off-by: Sebastian Siewior <bigeasy@linutronix.de>
Signed-off-by: Greg Ungerer <gerg@uclinux.org>
---
 arch/m68knommu/platform/coldfire/Makefile    |  2 +-
 arch/m68knommu/platform/coldfire/dma_timer.c | 68 ++++++++++++++++++++
 2 files changed, 69 insertions(+), 1 deletion(-)
 create mode 100644 arch/m68knommu/platform/coldfire/dma_timer.c

diff --git a/arch/m68knommu/platform/coldfire/Makefile b/arch/m68knommu/platform/coldfire/Makefile
index 40cf20be1b90..4f416a91a829 100644
--- a/arch/m68knommu/platform/coldfire/Makefile
+++ b/arch/m68knommu/platform/coldfire/Makefile
@@ -18,7 +18,7 @@ obj-$(CONFIG_COLDFIRE)	+= dma.o entry.o vectors.o
 obj-$(CONFIG_M5206)	+= timers.o
 obj-$(CONFIG_M5206e)	+= timers.o
 obj-$(CONFIG_M520x)	+= pit.o
-obj-$(CONFIG_M523x)	+= pit.o
+obj-$(CONFIG_M523x)	+= pit.o dma_timer.o
 obj-$(CONFIG_M5249)	+= timers.o
 obj-$(CONFIG_M527x)	+= pit.o
 obj-$(CONFIG_M5272)	+= timers.o
diff --git a/arch/m68knommu/platform/coldfire/dma_timer.c b/arch/m68knommu/platform/coldfire/dma_timer.c
new file mode 100644
index 000000000000..b623c993219e
--- /dev/null
+++ b/arch/m68knommu/platform/coldfire/dma_timer.c
@@ -0,0 +1,68 @@
+/*
+ * dma_timer.c -- Freescale ColdFire DMA Timer.
+ *
+ * Copyright (C) 2007, Benedikt Spranger <b.spranger@linutronix.de>
+ * Copyright (C) 2008. Sebastian Siewior, Linutronix
+ *
+ */
+
+#include <linux/clocksource.h>
+#include <linux/io.h>
+
+#include <asm/machdep.h>
+#include <asm/coldfire.h>
+#include <asm/mcfpit.h>
+#include <asm/mcfsim.h>
+
+#define DMA_TIMER_0	(0x00)
+#define DMA_TIMER_1	(0x40)
+#define DMA_TIMER_2	(0x80)
+#define DMA_TIMER_3	(0xc0)
+
+#define DTMR0	(MCF_IPSBAR + DMA_TIMER_0 + 0x400)
+#define DTXMR0	(MCF_IPSBAR + DMA_TIMER_0 + 0x402)
+#define DTER0	(MCF_IPSBAR + DMA_TIMER_0 + 0x403)
+#define DTRR0	(MCF_IPSBAR + DMA_TIMER_0 + 0x404)
+#define DTCR0	(MCF_IPSBAR + DMA_TIMER_0 + 0x408)
+#define DTCN0	(MCF_IPSBAR + DMA_TIMER_0 + 0x40c)
+
+#define DMA_FREQ    ((MCF_CLK / 2) / 16)
+
+/* DTMR */
+#define DMA_DTMR_RESTART	(1 << 3)
+#define DMA_DTMR_CLK_DIV_1	(1 << 1)
+#define DMA_DTMR_CLK_DIV_16	(2 << 1)
+#define DMA_DTMR_ENABLE		(1 << 0)
+
+static cycle_t cf_dt_get_cycles(void)
+{
+	return __raw_readl(DTCN0);
+}
+
+static struct clocksource clocksource_cf_dt = {
+	.name		= "coldfire_dma_timer",
+	.rating		= 200,
+	.read		= cf_dt_get_cycles,
+	.mask		= CLOCKSOURCE_MASK(32),
+	.shift		= 20,
+	.flags		= CLOCK_SOURCE_IS_CONTINUOUS,
+};
+
+static int __init  init_cf_dt_clocksource(void)
+{
+	/*
+	 * We setup DMA timer 0 in free run mode. This incrementing counter is
+	 * used as a highly precious clock source. With MCF_CLOCK = 150 MHz we
+	 * get a ~213 ns resolution and the 32bit register will overflow almost
+	 * every 15 minutes.
+	 */
+	__raw_writeb(0x00, DTXMR0);
+	__raw_writeb(0x00, DTER0);
+	__raw_writel(0x00000000, DTRR0);
+	__raw_writew(DMA_DTMR_CLK_DIV_16 | DMA_DTMR_ENABLE, DTMR0);
+	clocksource_cf_dt.mult = clocksource_hz2mult(DMA_FREQ,
+						     clocksource_cf_dt.shift);
+	return clocksource_register(&clocksource_cf_dt);
+}
+
+arch_initcall(init_cf_dt_clocksource);

From 1fda83d83c664ad74bfec8ce093a86d4d962f093 Mon Sep 17 00:00:00 2001
From: Sebastian Siewior <bigeasy@linutronix.de>
Date: Fri, 9 May 2008 16:13:36 +0200
Subject: [PATCH 0124/1171] m68knommu: m68knommu: add old stack trace method

The old method is used when frame pointers are not available.
Also fix formating with CONFIG_KALLSYMS=n which eliminates \n.

Signed-off-by: Sebastian Siewior <bigeasy@linutronix.de>
Signed-off-by: Greg Ungerer <gerg@uclinux.org>
---
 arch/m68knommu/kernel/traps.c | 38 ++++++++++++++++++++++++++++++++---
 1 file changed, 35 insertions(+), 3 deletions(-)

diff --git a/arch/m68knommu/kernel/traps.c b/arch/m68knommu/kernel/traps.c
index ec9aea652e79..46f8f9d0c408 100644
--- a/arch/m68knommu/kernel/traps.c
+++ b/arch/m68knommu/kernel/traps.c
@@ -103,12 +103,28 @@ asmlinkage void buserr_c(struct frame *fp)
 	force_sig(SIGSEGV, current);
 }
 
+static void print_this_address(unsigned long addr, int i)
+{
+#ifdef CONFIG_KALLSYMS
+	printk(KERN_EMERG " [%08lx] ", addr);
+	print_symbol(KERN_CONT "%s\n", addr);
+#else
+	if (i % 5)
+		printk(KERN_CONT " [%08lx] ", addr);
+	else
+		printk(KERN_CONT "\n" KERN_EMERG " [%08lx] ", addr);
+	i++;
+#endif
+}
+
 int kstack_depth_to_print = 48;
 
 static void __show_stack(struct task_struct *task, unsigned long *stack)
 {
 	unsigned long *endstack, addr;
+#ifdef CONFIG_FRAME_POINTER
 	unsigned long *last_stack;
+#endif
 	int i;
 
 	if (!stack)
@@ -126,6 +142,7 @@ static void __show_stack(struct task_struct *task, unsigned long *stack)
 		printk(" %08lx", *(stack + i));
 	}
 	printk("\n");
+	i = 0;
 
 #ifdef CONFIG_FRAME_POINTER
 	printk(KERN_EMERG "Call Trace:\n");
@@ -134,15 +151,30 @@ static void __show_stack(struct task_struct *task, unsigned long *stack)
 	while (stack <= endstack && stack > last_stack) {
 
 		addr = *(stack + 1);
-		printk(KERN_EMERG " [%08lx] ", addr);
-		print_symbol(KERN_CONT "%s\n", addr);
+		print_this_address(addr, i);
+		i++;
 
 		last_stack = stack;
 		stack = (unsigned long *)*stack;
 	}
 	printk("\n");
 #else
-	printk(KERN_EMERG "CONFIG_FRAME_POINTER disabled, no symbolic call trace\n");
+	printk(KERN_EMERG "Call Trace with CONFIG_FRAME_POINTER disabled:\n");
+	while (stack <= endstack) {
+		addr = *stack++;
+		/*
+		 * If the address is either in the text segment of the kernel,
+		 * or in a region which is occupied by a module then it *may*
+		 * be the address of a calling routine; if so, print it so that
+		 * someone tracing down the cause of the crash will be able to
+		 * figure out the call path that was taken.
+		 */
+		if (__kernel_text_address(addr)) {
+			print_this_address(addr, i);
+			i++;
+		}
+	}
+	printk(KERN_CONT "\n");
 #endif
 }
 

From 0df185f5a1430ab8b437be402d286ee0728ef9f8 Mon Sep 17 00:00:00 2001
From: Sebastian Siewior <bigeasy@linutronix.de>
Date: Mon, 28 Apr 2008 11:43:00 +0200
Subject: [PATCH 0125/1171] m68knommu: move code within time.c

This patch creates two functions do_set_rtc() and read_rtc_mmss()
based on allready available code.

Signed-off-by: Sebastian Siewior <bigeasy@linutronix.de>
Signed-off-by: Greg Ungerer <gerg@uclinux.org>
---
 arch/m68knommu/kernel/time.c | 40 +++++++++++++++++++++++-------------
 1 file changed, 26 insertions(+), 14 deletions(-)

diff --git a/arch/m68knommu/kernel/time.c b/arch/m68knommu/kernel/time.c
index 0ccfb2ad6380..d33ed9a84cc7 100644
--- a/arch/m68knommu/kernel/time.c
+++ b/arch/m68knommu/kernel/time.c
@@ -33,22 +33,11 @@ static inline int set_rtc_mmss(unsigned long nowtime)
 	return -1;
 }
 
-/*
- * timer_interrupt() needs to keep up the real-time clock,
- * as well as call the "do_timer()" routine every clocktick
- */
-irqreturn_t arch_timer_interrupt(int irq, void *dummy)
+static inline void do_set_rtc(void)
 {
 	/* last time the cmos clock got updated */
 	static long last_rtc_update=0;
 
-	if (current->pid)
-		profile_tick(CPU_PROFILING);
-
-	write_seqlock(&xtime_lock);
-
-	do_timer(1);
-
 	/*
 	 * If we have an externally synchronized Linux clock, then update
 	 * CMOS clock accordingly every ~11 minutes. Set_rtc_mmss() has to be
@@ -63,6 +52,23 @@ irqreturn_t arch_timer_interrupt(int irq, void *dummy)
 	  else
 	    last_rtc_update = xtime.tv_sec - 600; /* do it again in 60 s */
 	}
+}
+
+/*
+ * timer_interrupt() needs to keep up the real-time clock,
+ * as well as call the "do_timer()" routine every clocktick
+ */
+irqreturn_t arch_timer_interrupt(int irq, void *dummy)
+{
+
+	if (current->pid)
+		profile_tick(CPU_PROFILING);
+
+	write_seqlock(&xtime_lock);
+
+	do_timer(1);
+
+	do_set_rtc();
 
 	write_sequnlock(&xtime_lock);
 
@@ -72,7 +78,7 @@ irqreturn_t arch_timer_interrupt(int irq, void *dummy)
 	return(IRQ_HANDLED);
 }
 
-void time_init(void)
+static unsigned long read_rtc_mmss(void)
 {
 	unsigned int year, mon, day, hour, min, sec;
 
@@ -83,7 +89,13 @@ void time_init(void)
 
 	if ((year += 1900) < 1970)
 		year += 100;
-	xtime.tv_sec = mktime(year, mon, day, hour, min, sec);
+
+	return  mktime(year, mon, day, hour, min, sec);;
+}
+
+void time_init(void)
+{
+	xtime.tv_sec = read_rtc_mmss();
 	xtime.tv_nsec = 0;
 	wall_to_monotonic.tv_sec = -xtime.tv_sec;
 

From 95469bd64a7a9ab405b566deb8c81d4aaf67ed9e Mon Sep 17 00:00:00 2001
From: Sebastian Siewior <bigeasy@linutronix.de>
Date: Mon, 28 Apr 2008 11:43:01 +0200
Subject: [PATCH 0126/1171] m68knommu: complete generic time

do_set_rtc() isn't required because the work that is handled is
allready served if read_persistent_clock() & update_persistent_clock()
are implemented and CONFIG_GENERIC_CMOS_UPDATE is. sync_cmos_clock()
looks very familiar :)

Signed-off-by: Sebastian Siewior <bigeasy@linutronix.de>
Signed-off-by: Greg Ungerer <gerg@uclinux.org>
---
 arch/m68knommu/Kconfig       |  4 ++++
 arch/m68knommu/kernel/time.c | 40 ++++++++++--------------------------
 2 files changed, 15 insertions(+), 29 deletions(-)

diff --git a/arch/m68knommu/Kconfig b/arch/m68knommu/Kconfig
index 8e8441587c22..bfd35304d586 100644
--- a/arch/m68knommu/Kconfig
+++ b/arch/m68knommu/Kconfig
@@ -58,6 +58,10 @@ config GENERIC_TIME
 	bool
 	default y
 
+config GENERIC_CMOS_UPDATE
+	bool
+	default y
+
 config TIME_LOW_RES
 	bool
 	default y
diff --git a/arch/m68knommu/kernel/time.c b/arch/m68knommu/kernel/time.c
index d33ed9a84cc7..67944aa27280 100644
--- a/arch/m68knommu/kernel/time.c
+++ b/arch/m68knommu/kernel/time.c
@@ -33,27 +33,6 @@ static inline int set_rtc_mmss(unsigned long nowtime)
 	return -1;
 }
 
-static inline void do_set_rtc(void)
-{
-	/* last time the cmos clock got updated */
-	static long last_rtc_update=0;
-
-	/*
-	 * If we have an externally synchronized Linux clock, then update
-	 * CMOS clock accordingly every ~11 minutes. Set_rtc_mmss() has to be
-	 * called as close as possible to 500 ms before the new second starts.
-	 */
-	if (ntp_synced() &&
-	    xtime.tv_sec > last_rtc_update + 660 &&
-	    (xtime.tv_nsec / 1000) >= 500000 - ((unsigned) TICK_SIZE) / 2 &&
-	    (xtime.tv_nsec  / 1000) <= 500000 + ((unsigned) TICK_SIZE) / 2) {
-	  if (set_rtc_mmss(xtime.tv_sec) == 0)
-	    last_rtc_update = xtime.tv_sec;
-	  else
-	    last_rtc_update = xtime.tv_sec - 600; /* do it again in 60 s */
-	}
-}
-
 /*
  * timer_interrupt() needs to keep up the real-time clock,
  * as well as call the "do_timer()" routine every clocktick
@@ -68,8 +47,6 @@ irqreturn_t arch_timer_interrupt(int irq, void *dummy)
 
 	do_timer(1);
 
-	do_set_rtc();
-
 	write_sequnlock(&xtime_lock);
 
 #ifndef CONFIG_SMP
@@ -93,12 +70,17 @@ static unsigned long read_rtc_mmss(void)
 	return  mktime(year, mon, day, hour, min, sec);;
 }
 
-void time_init(void)
+unsigned long read_persistent_clock(void)
 {
-	xtime.tv_sec = read_rtc_mmss();
-	xtime.tv_nsec = 0;
-	wall_to_monotonic.tv_sec = -xtime.tv_sec;
-
-	hw_timer_init();
+	return read_rtc_mmss();
 }
 
+int update_persistent_clock(struct timespec now)
+{
+	return set_rtc_mmss(now.tv_sec);
+}
+
+void time_init(void)
+{
+	hw_timer_init();
+}

From 5bed10a5ee272fbf18ce0ce764245bbb8f28e2e6 Mon Sep 17 00:00:00 2001
From: Sebastian Siewior <bigeasy@linutronix.de>
Date: Mon, 28 Apr 2008 11:43:03 +0200
Subject: [PATCH 0127/1171] m68knommu: add sched_clock() for the DMA timer

with this printk() and other sched_clock() user use the more precise
timestamps. The highly optimized math is from arch/x86/kernel/tsc_32.c.

Signed-off-by: Sebastian Siewior <bigeasy@linutronix.de>
Signed-off-by: Greg Ungerer <gerg@uclinux.org>
---
 arch/m68knommu/platform/coldfire/dma_timer.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/arch/m68knommu/platform/coldfire/dma_timer.c b/arch/m68knommu/platform/coldfire/dma_timer.c
index b623c993219e..772578b1084f 100644
--- a/arch/m68knommu/platform/coldfire/dma_timer.c
+++ b/arch/m68knommu/platform/coldfire/dma_timer.c
@@ -66,3 +66,19 @@ static int __init  init_cf_dt_clocksource(void)
 }
 
 arch_initcall(init_cf_dt_clocksource);
+
+#define CYC2NS_SCALE_FACTOR 10 /* 2^10, carefully chosen */
+#define CYC2NS_SCALE	((1000000 << CYC2NS_SCALE_FACTOR) / (DMA_FREQ / 1000))
+
+static unsigned long long cycles2ns(unsigned long cycl)
+{
+	return (unsigned long long) ((unsigned long long)cycl *
+			CYC2NS_SCALE) >> CYC2NS_SCALE_FACTOR;
+}
+
+unsigned long long sched_clock(void)
+{
+	unsigned long cycl = __raw_readl(DTCN0);
+
+	return cycles2ns(cycl);
+}

From a6260ef84103fa8a51a67b6a58e5e16c676e08ad Mon Sep 17 00:00:00 2001
From: Sebastian Siewior <bigeasy@linutronix.de>
Date: Fri, 9 May 2008 16:10:37 +0200
Subject: [PATCH 0128/1171] m68knommu: add ffs and __ffs plattform which
 support ISA A+ or ISA C

the ff1 and bitrev opcode appears in ISA C and ISA A+ what isn't
supported by all plattforms. The assembly optimization is automaticly
enabled if the compiler understand the required cpu keyword.
My m5235 seems to boot and run fine so far.

Signed-off-by: Sebastian Siewior <bigeasy@linutronix.de>
Signed-off-by: Greg Ungerer <gerg@uclinux.org>
---
 arch/m68knommu/Makefile        | 11 +++++++----
 include/asm-m68knommu/bitops.h | 30 ++++++++++++++++++++++++++++++
 2 files changed, 37 insertions(+), 4 deletions(-)

diff --git a/arch/m68knommu/Makefile b/arch/m68knommu/Makefile
index e0b5f62e395c..b63bbcf874ff 100644
--- a/arch/m68knommu/Makefile
+++ b/arch/m68knommu/Makefile
@@ -8,6 +8,8 @@
 # (C) Copyright 2002, Greg Ungerer <gerg@snapgear.com>
 #
 
+KBUILD_DEFCONFIG := m5208evb_defconfig
+
 platform-$(CONFIG_M68328)	:= 68328
 platform-$(CONFIG_M68EZ328)	:= 68EZ328
 platform-$(CONFIG_M68VZ328)	:= 68VZ328
@@ -90,13 +92,14 @@ export PLATFORM BOARD MODEL CPUCLASS
 cflags-$(CONFIG_M5206)		:= -m5200
 cflags-$(CONFIG_M5206e)		:= -m5200
 cflags-$(CONFIG_M520x)		:= -m5307
-cflags-$(CONFIG_M523x)		:= -m5307
+cflags-$(CONFIG_M523x)		:= $(call cc-option,-mcpu=523x,-m5307)
 cflags-$(CONFIG_M5249)		:= -m5200
-cflags-$(CONFIG_M527x)		:= -m5307
+cflags-$(CONFIG_M5271)		:= $(call cc-option,-mcpu=5271,-m5307)
 cflags-$(CONFIG_M5272)		:= -m5307
-cflags-$(CONFIG_M528x)		:= -m5307
+cflags-$(CONFIG_M5275)		:= $(call cc-option,-mcpu=5275,-m5307)
+cflags-$(CONFIG_M528x)		:= $(call cc-option,-m528x,-m5307)
 cflags-$(CONFIG_M5307)		:= -m5307
-cflags-$(CONFIG_M532x)		:= -m5307
+cflags-$(CONFIG_M532x)		:= $(call cc-option,-mcpu=532x,-m5307)
 cflags-$(CONFIG_M5407)		:= -m5200
 cflags-$(CONFIG_M68328)		:= -m68000
 cflags-$(CONFIG_M68EZ328)	:= -m68000
diff --git a/include/asm-m68knommu/bitops.h b/include/asm-m68knommu/bitops.h
index c142fbf2f376..6f3685eab44c 100644
--- a/include/asm-m68knommu/bitops.h
+++ b/include/asm-m68knommu/bitops.h
@@ -14,8 +14,38 @@
 #error only <linux/bitops.h> can be included directly
 #endif
 
+#if defined (__mcfisaaplus__) || defined (__mcfisac__)
+static inline int ffs(unsigned int val)
+{
+        if (!val)
+                return 0;
+
+        asm volatile(
+                        "bitrev %0\n\t"
+                        "ff1 %0\n\t"
+                        : "=d" (val)
+                        : "0" (val)
+		    );
+        val++;
+        return val;
+}
+
+static inline int __ffs(unsigned int val)
+{
+        asm volatile(
+                        "bitrev %0\n\t"
+                        "ff1 %0\n\t"
+                        : "=d" (val)
+                        : "0" (val)
+		    );
+        return val;
+}
+
+#else
 #include <asm-generic/bitops/ffs.h>
 #include <asm-generic/bitops/__ffs.h>
+#endif
+
 #include <asm-generic/bitops/sched.h>
 #include <asm-generic/bitops/ffz.h>
 

From e872504b311cec52f7a316a0037fb959080dbea0 Mon Sep 17 00:00:00 2001
From: Sebastian Siewior <bigeasy@linutronix.de>
Date: Sat, 17 May 2008 21:51:15 +0200
Subject: [PATCH 0129/1171] m68knommu: add byteswap assembly opcode for ISA A+

Signed-off-by: Sebastian Siewior <bigeasy@linutronix.de>
Signed-off-by: Greg Ungerer <gerg@uclinux.org>
---
 include/asm-m68knommu/byteorder.h | 16 +++++++++++++++-
 1 file changed, 15 insertions(+), 1 deletion(-)

diff --git a/include/asm-m68knommu/byteorder.h b/include/asm-m68knommu/byteorder.h
index 8fcde907b0f9..20bb4426b610 100644
--- a/include/asm-m68knommu/byteorder.h
+++ b/include/asm-m68knommu/byteorder.h
@@ -1,13 +1,27 @@
 #ifndef _M68KNOMMU_BYTEORDER_H
 #define _M68KNOMMU_BYTEORDER_H
 
-#include <asm/types.h>
+#include <linux/types.h>
 
 #if defined(__GNUC__) && !defined(__STRICT_ANSI__) || defined(__KERNEL__)
 #  define __BYTEORDER_HAS_U64__
 #  define __SWAB_64_THRU_32__
 #endif
 
+#if defined (__mcfisaaplus__) || defined (__mcfisac__)
+static inline __attribute_const__ __u32 ___arch__swab32(__u32 val)
+{
+	asm(
+			"byterev %0"
+			: "=d" (val)
+			: "0" (val)
+	   );
+	return val;
+}
+
+#define __arch__swab32(x) ___arch__swab32(x)
+#endif
+
 #include <linux/byteorder/big_endian.h>
 
 #endif /* _M68KNOMMU_BYTEORDER_H */

From 6dbeb456baaba05d60e7ca8213da26142062408a Mon Sep 17 00:00:00 2001
From: Sebastian Siewior <bigeasy@linutronix.de>
Date: Tue, 13 May 2008 18:52:44 +0200
Subject: [PATCH 0130/1171] m68knommu: add read_barrier_depends() and
 irqs_disabled_flags()

/home/bigeasy/git/linux-2.6-ftrace/kernel/trace/trace.c: In function 'tracing_generic_entry_update':
/home/bigeasy/git/linux-2.6-ftrace/kernel/trace/trace.c:802: error: implicit declaration of function 'irqs_disabled_flags'
make[3]: *** [kernel/trace/trace.o] Error 1
/home/bigeasy/git/linux-2.6-ftrace/kernel/trace/ftrace.c: In function 'ftrace_list_func':
/home/bigeasy/git/linux-2.6-ftrace/kernel/trace/ftrace.c:61: error: implicit declaration of function 'read_barrier_depends'

Signed-off-by: Sebastian Siewior <bigeasy@linutronix.de>
Signed-off-by: Greg Ungerer <gerg@uclinux.org>
---
 include/asm-m68knommu/system.h | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/include/asm-m68knommu/system.h b/include/asm-m68knommu/system.h
index 64c64432bbb8..40f49de69821 100644
--- a/include/asm-m68knommu/system.h
+++ b/include/asm-m68knommu/system.h
@@ -118,6 +118,8 @@ asmlinkage void resume(void);
 #define smp_read_barrier_depends()	do { } while(0)
 #endif
 
+#define read_barrier_depends()  ((void)0)
+
 #define xchg(ptr,x) ((__typeof__(*(ptr)))__xchg((unsigned long)(x),(ptr),sizeof(*(ptr))))
 
 struct __xchg_dummy { unsigned long a[100]; };
@@ -310,4 +312,13 @@ static inline unsigned long __xchg(unsigned long x, volatile void * ptr, int siz
 #endif
 #define arch_align_stack(x) (x)
 
+
+static inline int irqs_disabled_flags(unsigned long flags)
+{
+	if (flags & 0x0700)
+		return 0;
+	else
+		return 1;
+}
+
 #endif /* _M68KNOMMU_SYSTEM_H */

From 2b9a69861c39ae4c232385def833816acc07a0a4 Mon Sep 17 00:00:00 2001
From: Sebastian Siewior <bigeasy@linutronix.de>
Date: Mon, 28 Apr 2008 11:43:04 +0200
Subject: [PATCH 0131/1171] m68knommu: MCF5307 PIT GENERIC_CLOCKEVENTS support

The PIT code has been changed in order to suppport GENERIC_CLOCKEVENTS.
The priority of the PIT clocksource has been decreased in favor of the
DMA timer.

pit_cycles_per_jiffy become a constant (PIT_CYCLES_PER_JIFFY) because it
is known at compile time and does not change afterwards.

Signed-off-by: Benedikt Spranger <b.spranger@linutronix.de>
Signed-off-by: Sebastian Siewior <bigeasy@linutronix.de>
Signed-off-by: Greg Ungerer <gerg@uclinux.org>
---
 arch/m68knommu/Kconfig                 | 11 ++++
 arch/m68knommu/kernel/time.c           |  2 +
 arch/m68knommu/platform/coldfire/pit.c | 91 ++++++++++++++++++++++----
 3 files changed, 91 insertions(+), 13 deletions(-)

diff --git a/arch/m68knommu/Kconfig b/arch/m68knommu/Kconfig
index bfd35304d586..2e7515e8db98 100644
--- a/arch/m68knommu/Kconfig
+++ b/arch/m68knommu/Kconfig
@@ -66,6 +66,10 @@ config TIME_LOW_RES
 	bool
 	default y
 
+config GENERIC_CLOCKEVENTS
+	bool
+	default n
+
 config NO_IOPORT
 	def_bool y
 
@@ -112,11 +116,13 @@ config M5206e
 
 config M520x
 	bool "MCF520x"
+	select GENERIC_CLOCKEVENTS
 	help
 	   Freescale Coldfire 5207/5208 processor support.
 
 config M523x
 	bool "MCF523x"
+	select GENERIC_CLOCKEVENTS
 	help
 	  Freescale Coldfire 5230/1/2/4/5 processor support
 
@@ -142,6 +148,7 @@ config M5275
 
 config M528x
 	bool "MCF528x"
+	select GENERIC_CLOCKEVENTS
 	help
 	  Motorola ColdFire 5280/5282 processor support.
 
@@ -165,6 +172,7 @@ endchoice
 config M527x
 	bool
 	depends on (M5271 || M5275)
+	select GENERIC_CLOCKEVENTS
 	default y
 
 config COLDFIRE
@@ -678,6 +686,9 @@ endchoice
 if COLDFIRE
 source "kernel/Kconfig.preempt"
 endif
+
+source "kernel/time/Kconfig"
+
 source "mm/Kconfig"
 
 endmenu
diff --git a/arch/m68knommu/kernel/time.c b/arch/m68knommu/kernel/time.c
index 67944aa27280..d182b2f72211 100644
--- a/arch/m68knommu/kernel/time.c
+++ b/arch/m68knommu/kernel/time.c
@@ -33,6 +33,7 @@ static inline int set_rtc_mmss(unsigned long nowtime)
 	return -1;
 }
 
+#ifndef CONFIG_GENERIC_CLOCKEVENTS
 /*
  * timer_interrupt() needs to keep up the real-time clock,
  * as well as call the "do_timer()" routine every clocktick
@@ -54,6 +55,7 @@ irqreturn_t arch_timer_interrupt(int irq, void *dummy)
 #endif
 	return(IRQ_HANDLED);
 }
+#endif
 
 static unsigned long read_rtc_mmss(void)
 {
diff --git a/arch/m68knommu/platform/coldfire/pit.c b/arch/m68knommu/platform/coldfire/pit.c
index 4290638012e0..c5b916700b22 100644
--- a/arch/m68knommu/platform/coldfire/pit.c
+++ b/arch/m68knommu/platform/coldfire/pit.c
@@ -18,7 +18,7 @@
 #include <linux/init.h>
 #include <linux/interrupt.h>
 #include <linux/irq.h>
-#include <linux/clocksource.h>
+#include <linux/clockchips.h>
 #include <asm/machdep.h>
 #include <asm/io.h>
 #include <asm/coldfire.h>
@@ -33,22 +33,86 @@
 #define	FREQ	((MCF_CLK / 2) / 64)
 #define	TA(a)	(MCF_IPSBAR + MCFPIT_BASE1 + (a))
 #define	INTC0	(MCF_IPSBAR + MCFICM_INTC0)
+#define PIT_CYCLES_PER_JIFFY (FREQ / HZ)
 
-static u32 pit_cycles_per_jiffy;
 static u32 pit_cnt;
 
+/*
+ * Initialize the PIT timer.
+ *
+ * This is also called after resume to bring the PIT into operation again.
+ */
+
+static void init_cf_pit_timer(enum clock_event_mode mode,
+                             struct clock_event_device *evt)
+{
+	switch (mode) {
+	case CLOCK_EVT_MODE_PERIODIC:
+
+		__raw_writew(MCFPIT_PCSR_DISABLE, TA(MCFPIT_PCSR));
+		__raw_writew(PIT_CYCLES_PER_JIFFY, TA(MCFPIT_PMR));
+		__raw_writew(MCFPIT_PCSR_EN | MCFPIT_PCSR_PIE | \
+				MCFPIT_PCSR_OVW | MCFPIT_PCSR_RLD | \
+				MCFPIT_PCSR_CLK64, TA(MCFPIT_PCSR));
+		break;
+
+	case CLOCK_EVT_MODE_SHUTDOWN:
+	case CLOCK_EVT_MODE_UNUSED:
+
+		__raw_writew(MCFPIT_PCSR_DISABLE, TA(MCFPIT_PCSR));
+		break;
+
+	case CLOCK_EVT_MODE_ONESHOT:
+
+		__raw_writew(MCFPIT_PCSR_DISABLE, TA(MCFPIT_PCSR));
+		__raw_writew(MCFPIT_PCSR_EN | MCFPIT_PCSR_PIE | \
+				MCFPIT_PCSR_OVW | MCFPIT_PCSR_CLK64, \
+				TA(MCFPIT_PCSR));
+		break;
+
+	case CLOCK_EVT_MODE_RESUME:
+		/* Nothing to do here */
+		break;
+	}
+}
+
+/*
+ * Program the next event in oneshot mode
+ *
+ * Delta is given in PIT ticks
+ */
+static int cf_pit_next_event(unsigned long delta,
+		struct clock_event_device *evt)
+{
+	__raw_writew(delta, TA(MCFPIT_PMR));
+	return 0;
+}
+
+struct clock_event_device cf_pit_clockevent = {
+	.name		= "pit",
+	.features	= CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT,
+	.set_mode	= init_cf_pit_timer,
+	.set_next_event	= cf_pit_next_event,
+	.shift		= 32,
+	.irq		= MCFINT_VECBASE + MCFINT_PIT1,
+};
+
+
+
 /***************************************************************************/
 
 static irqreturn_t pit_tick(int irq, void *dummy)
 {
+	struct clock_event_device *evt = &cf_pit_clockevent;
 	u16 pcsr;
 
 	/* Reset the ColdFire timer */
 	pcsr = __raw_readw(TA(MCFPIT_PCSR));
 	__raw_writew(pcsr | MCFPIT_PCSR_PIF, TA(MCFPIT_PCSR));
 
-	pit_cnt += pit_cycles_per_jiffy;
-	return arch_timer_interrupt(irq, dummy);
+	pit_cnt += PIT_CYCLES_PER_JIFFY;
+	evt->event_handler(evt);
+	return IRQ_HANDLED;
 }
 
 /***************************************************************************/
@@ -72,14 +136,14 @@ static cycle_t pit_read_clk(void)
 	cycles = pit_cnt;
 	local_irq_restore(flags);
 
-	return cycles + pit_cycles_per_jiffy - pcntr;
+	return cycles + PIT_CYCLES_PER_JIFFY - pcntr;
 }
 
 /***************************************************************************/
 
 static struct clocksource pit_clk = {
 	.name	= "pit",
-	.rating	= 250,
+	.rating	= 100,
 	.read	= pit_read_clk,
 	.shift	= 20,
 	.mask	= CLOCKSOURCE_MASK(32),
@@ -92,6 +156,14 @@ void hw_timer_init(void)
 {
 	u32 imr;
 
+	cf_pit_clockevent.cpumask = cpumask_of_cpu(smp_processor_id());
+	cf_pit_clockevent.mult = div_sc(FREQ, NSEC_PER_SEC, 32);
+	cf_pit_clockevent.max_delta_ns =
+		clockevent_delta2ns(0xFFFF, &cf_pit_clockevent);
+	cf_pit_clockevent.min_delta_ns =
+		clockevent_delta2ns(0x3f, &cf_pit_clockevent);
+	clockevents_register_device(&cf_pit_clockevent);
+
 	setup_irq(MCFINT_VECBASE + MCFINT_PIT1, &pit_irq);
 
 	__raw_writeb(ICR_INTRCONF, INTC0 + MCFINTC_ICR0 + MCFINT_PIT1);
@@ -99,13 +171,6 @@ void hw_timer_init(void)
 	imr &= ~MCFPIT_IMR_IBIT;
 	__raw_writel(imr, INTC0 + MCFPIT_IMR);
 
-	/* Set up PIT timer 1 as poll clock */
-	pit_cycles_per_jiffy = FREQ / HZ;
-	__raw_writew(MCFPIT_PCSR_DISABLE, TA(MCFPIT_PCSR));
-	__raw_writew(pit_cycles_per_jiffy, TA(MCFPIT_PMR));
-	__raw_writew(MCFPIT_PCSR_EN | MCFPIT_PCSR_PIE | MCFPIT_PCSR_OVW |
-		MCFPIT_PCSR_RLD | MCFPIT_PCSR_CLK64, TA(MCFPIT_PCSR));
-
 	pit_clk.mult = clocksource_hz2mult(FREQ, pit_clk.shift);
 	clocksource_register(&pit_clk);
 }

From 0d176af5b7e658490b75427ccd23ff6a158c472b Mon Sep 17 00:00:00 2001
From: Sebastian Siewior <bigeasy@linutronix.de>
Date: Mon, 28 Apr 2008 11:43:20 +0200
Subject: [PATCH 0132/1171] m68knommu: fec: remove FADS

I found config FADS only in ppc/Kconfig. Bye bye relic.

Signed-off-by: Sebastian Siewior <bigeasy@linutronix.de>
Signed-off-by: Greg Ungerer <gerg@uclinux.org>
---
 drivers/net/fec.c | 10 ----------
 1 file changed, 10 deletions(-)

diff --git a/drivers/net/fec.c b/drivers/net/fec.c
index 32a4f17d35fc..6abbcd5f7c3a 100644
--- a/drivers/net/fec.c
+++ b/drivers/net/fec.c
@@ -2,12 +2,6 @@
  * Fast Ethernet Controller (FEC) driver for Motorola MPC8xx.
  * Copyright (c) 1997 Dan Malek (dmalek@jlc.net)
  *
- * This version of the driver is specific to the FADS implementation,
- * since the board contains control registers external to the processor
- * for the control of the LevelOne LXT970 transceiver.  The MPC860T manual
- * describes connections using the internal parallel port I/O, which
- * is basically all of Port D.
- *
  * Right now, I am very wasteful with the buffers.  I allocate memory
  * pages and then divide them into 2K frame buffers.  This way I know I
  * have buffers large enough to hold one frame within one buffer descriptor.
@@ -1809,10 +1803,6 @@ static void __inline__ fec_request_intrs(struct net_device *dev)
 	*/
 	*((uint *) RPX_CSR_ADDR) &= ~BCSR2_FETHLEDMODE;
 #endif
-#ifdef CONFIG_FADS
-	if (request_8xxirq(SIU_IRQ2, mii_link_interrupt, 0, "mii", dev) != 0)
-		panic("Could not allocate MII IRQ!");
-#endif
 }
 
 static void __inline__ fec_get_mac(struct net_device *dev)

From c1863bed8c88324405dc2a922c153fe5d7df716c Mon Sep 17 00:00:00 2001
From: Sebastian Siewior <bigeasy@linutronix.de>
Date: Mon, 28 Apr 2008 11:43:17 +0200
Subject: [PATCH 0133/1171] m68knommu: remove RPXCLASSIC from the m68k tree

This ifdefs are leftovers from the time as the driver was running
on a ppc.
Signed-off-by: Sebastian Siewior <bigeasy@linutronix.de>
Signed-off-by: Greg Ungerer <gerg@uclinux.org>
---
 drivers/net/fec.c                | 42 --------------------------------
 include/asm-m68knommu/commproc.h | 19 ---------------
 2 files changed, 61 deletions(-)

diff --git a/drivers/net/fec.c b/drivers/net/fec.c
index 6abbcd5f7c3a..0ef7226efd5c 100644
--- a/drivers/net/fec.c
+++ b/drivers/net/fec.c
@@ -43,17 +43,9 @@
 #include <asm/pgtable.h>
 #include <asm/cacheflush.h>
 
-#if defined(CONFIG_M523x) || defined(CONFIG_M527x) || \
-    defined(CONFIG_M5272) || defined(CONFIG_M528x) || \
-    defined(CONFIG_M520x) || defined(CONFIG_M532x)
 #include <asm/coldfire.h>
 #include <asm/mcfsim.h>
 #include "fec.h"
-#else
-#include <asm/8xx_immap.h>
-#include <asm/mpc8xx.h>
-#include "commproc.h"
-#endif
 
 #if defined(CONFIG_FEC2)
 #define	FEC_MAX_PORTS	2
@@ -1229,14 +1221,9 @@ static phy_info_t const * const phy_info[] = {
 
 /* ------------------------------------------------------------------------- */
 #ifdef HAVE_mii_link_interrupt
-#ifdef CONFIG_RPXCLASSIC
-static void
-mii_link_interrupt(void *dev_id);
-#else
 static irqreturn_t
 mii_link_interrupt(int irq, void * dev_id);
 #endif
-#endif
 
 #if defined(CONFIG_M5272)
 /*
@@ -1789,20 +1776,6 @@ static void __inline__ fec_request_intrs(struct net_device *dev)
 
 	if (request_8xxirq(FEC_INTERRUPT, fec_enet_interrupt, 0, "fec", dev) != 0)
 		panic("Could not allocate FEC IRQ!");
-
-#ifdef CONFIG_RPXCLASSIC
-	/* Make Port C, bit 15 an input that causes interrupts.
-	*/
-	immap->im_ioport.iop_pcpar &= ~0x0001;
-	immap->im_ioport.iop_pcdir &= ~0x0001;
-	immap->im_ioport.iop_pcso &= ~0x0001;
-	immap->im_ioport.iop_pcint |= 0x0001;
-	cpm_install_handler(CPMVEC_PIO_PC15, mii_link_interrupt, dev);
-
-	/* Make LEDS reflect Link status.
-	*/
-	*((uint *) RPX_CSR_ADDR) &= ~BCSR2_FETHLEDMODE;
-#endif
 }
 
 static void __inline__ fec_get_mac(struct net_device *dev)
@@ -1811,16 +1784,6 @@ static void __inline__ fec_get_mac(struct net_device *dev)
 
 	bd = (bd_t *)__res;
 	memcpy(dev->dev_addr, bd->bi_enetaddr, ETH_ALEN);
-
-#ifdef CONFIG_RPXCLASSIC
-	/* The Embedded Planet boards have only one MAC address in
-	 * the EEPROM, but can have two Ethernet ports.  For the
-	 * FEC port, we create another address by setting one of
-	 * the address bits above something that would have (up to
-	 * now) been allocated.
-	 */
-	dev->dev_adrd[3] |= 0x80;
-#endif
 }
 
 static void __inline__ fec_set_mii(struct net_device *dev, struct fec_enet_private *fep)
@@ -2099,13 +2062,8 @@ mii_discover_phy(uint mii_reg, struct net_device *dev)
 /* This interrupt occurs when the PHY detects a link change.
 */
 #ifdef HAVE_mii_link_interrupt
-#ifdef CONFIG_RPXCLASSIC
-static void
-mii_link_interrupt(void *dev_id)
-#else
 static irqreturn_t
 mii_link_interrupt(int irq, void * dev_id)
-#endif
 {
 	struct	net_device *dev = dev_id;
 	struct fec_enet_private *fep = netdev_priv(dev);
diff --git a/include/asm-m68knommu/commproc.h b/include/asm-m68knommu/commproc.h
index 36e870b468ef..edf5eb6c08d2 100644
--- a/include/asm-m68knommu/commproc.h
+++ b/include/asm-m68knommu/commproc.h
@@ -519,25 +519,6 @@ typedef struct scc_enet {
 #define SICR_ENET_CLKRT	((uint)0x00002c00)
 #endif
 
-#ifdef CONFIG_RPXCLASSIC
-/* Bits in parallel I/O port registers that have to be set/cleared
- * to configure the pins for SCC1 use.
- */
-#define PA_ENET_RXD	((ushort)0x0001)
-#define PA_ENET_TXD	((ushort)0x0002)
-#define PA_ENET_TCLK	((ushort)0x0200)
-#define PA_ENET_RCLK	((ushort)0x0800)
-#define PB_ENET_TENA	((uint)0x00001000)
-#define PC_ENET_CLSN	((ushort)0x0010)
-#define PC_ENET_RENA	((ushort)0x0020)
-
-/* Control bits in the SICR to route TCLK (CLK2) and RCLK (CLK4) to
- * SCC1.  Also, make sure GR1 (bit 24) and SC1 (bit 25) are zero.
- */
-#define SICR_ENET_MASK	((uint)0x000000ff)
-#define SICR_ENET_CLKRT	((uint)0x0000003d)
-#endif
-
 /* SCC Event register as used by Ethernet.
 */
 #define SCCE_ENET_GRA	((ushort)0x0080)	/* Graceful stop complete */

From 87f4abb45bc640638e6986f0f4d412b2d0ea21e1 Mon Sep 17 00:00:00 2001
From: Greg Ungerer <gerg@uclinux.org>
Date: Fri, 6 Jun 2008 15:55:36 +1000
Subject: [PATCH 0134/1171] m68knommu: remove last use of CONFIG_FADS and
 CONFIG_RPXCLASSIC

They have never been used in this port of the driver. It is has only
ever been used on the ColdFire SoC ethernet core.

Signed-off-by: Greg Ungerer <gerg@uclinux.org>
---
 drivers/net/fec.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/fec.c b/drivers/net/fec.c
index 0ef7226efd5c..ecd5c71a7a8a 100644
--- a/drivers/net/fec.c
+++ b/drivers/net/fec.c
@@ -53,7 +53,7 @@
 #define	FEC_MAX_PORTS	1
 #endif
 
-#if defined(CONFIG_FADS) || defined(CONFIG_RPXCLASSIC) || defined(CONFIG_M5272)
+#if defined(CONFIG_M5272)
 #define HAVE_mii_link_interrupt
 #endif
 

From 9b0e74102494971ca37a425c63031fea68bb5b79 Mon Sep 17 00:00:00 2001
From: Greg Ungerer <gerg@uclinux.org>
Date: Fri, 11 Jul 2008 15:29:36 +1000
Subject: [PATCH 0135/1171] m68knommu: put ColdFire head code into .text.head
 section

Switch the ColdFire head start up code to be in the .text.head segment.
And make sure that segment is at the start of the final linked text
segment. Fixes the linker warnings about section use mis-matches:

  WARNING: vmlinux.o(.text+0xa8): Section mismatch in reference from the variable _clear_bss to the function .init.text:start_kernel()
  The function _clear_bss() references
  the function __init start_kernel().
  This is often because _clear_bss lacks a __init
  annotation or the annotation of start_kernel is wrong.

Signed-off-by: Greg Ungerer <gerg@uclinux.org>
---
 arch/m68knommu/kernel/vmlinux.lds.S     | 1 +
 arch/m68knommu/platform/coldfire/head.S | 3 ++-
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/arch/m68knommu/kernel/vmlinux.lds.S b/arch/m68knommu/kernel/vmlinux.lds.S
index 93e69236ed6f..69ba9b10767a 100644
--- a/arch/m68knommu/kernel/vmlinux.lds.S
+++ b/arch/m68knommu/kernel/vmlinux.lds.S
@@ -62,6 +62,7 @@ SECTIONS {
 	.text : {
 		_text = .;
 		_stext = . ;
+		HEAD_TEXT
 		TEXT_TEXT
 		SCHED_TEXT
 		LOCK_TEXT
diff --git a/arch/m68knommu/platform/coldfire/head.S b/arch/m68knommu/platform/coldfire/head.S
index b9aa0ca29bfb..2b0d73c0cc32 100644
--- a/arch/m68knommu/platform/coldfire/head.S
+++ b/arch/m68knommu/platform/coldfire/head.S
@@ -10,6 +10,7 @@
 
 #include <linux/sys.h>
 #include <linux/linkage.h>
+#include <linux/init.h>
 #include <asm/asm-offsets.h>
 #include <asm/coldfire.h>
 #include <asm/mcfcache.h>
@@ -126,7 +127,7 @@ _ramend:
 
 /*****************************************************************************/
 
-.text
+__HEAD
 
 /*
  *	This is the codes first entry point. This is where it all

From b7c2a75725dee9b5643a0aae3a4cb47f52e00a49 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Tue, 22 Jul 2008 22:34:29 -0700
Subject: [PATCH 0136/1171] sparc64: Fix lockdep issues in LDC protocol layer.

We're calling request_irq() with a IRQs disabled.

No straightforward fix exists because we want to
enable these IRQs and setup state atomically before
getting into the IRQ handler the first time.

What happens now is that we mark the VIRQ to not be
automatically enabled by request_irq().  Then we
make explicit enable_irq() calls when we grab the
LDC channel.

This way we don't need to call request_irq() illegally
under the LDC channel lock any more.

Bump LDC version and release date.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/sparc64/kernel/irq.c | 10 +++++++++-
 arch/sparc64/kernel/ldc.c | 38 +++++++++++++++++++-------------------
 2 files changed, 28 insertions(+), 20 deletions(-)

diff --git a/arch/sparc64/kernel/irq.c b/arch/sparc64/kernel/irq.c
index b441a26b73b0..c481673d249c 100644
--- a/arch/sparc64/kernel/irq.c
+++ b/arch/sparc64/kernel/irq.c
@@ -621,8 +621,9 @@ unsigned int sun4v_build_irq(u32 devhandle, unsigned int devino)
 unsigned int sun4v_build_virq(u32 devhandle, unsigned int devino)
 {
 	struct irq_handler_data *data;
-	struct ino_bucket *bucket;
 	unsigned long hv_err, cookie;
+	struct ino_bucket *bucket;
+	struct irq_desc *desc;
 	unsigned int virt_irq;
 
 	bucket = kzalloc(sizeof(struct ino_bucket), GFP_ATOMIC);
@@ -643,6 +644,13 @@ unsigned int sun4v_build_virq(u32 devhandle, unsigned int devino)
 	if (unlikely(!data))
 		return 0;
 
+	/* In order to make the LDC channel startup sequence easier,
+	 * especially wrt. locking, we do not let request_irq() enable
+	 * the interrupt.
+	 */
+	desc = irq_desc + virt_irq;
+	desc->status |= IRQ_NOAUTOEN;
+
 	set_irq_chip_data(virt_irq, data);
 
 	/* Catch accidental accesses to these things.  IMAP/ICLR handling
diff --git a/arch/sparc64/kernel/ldc.c b/arch/sparc64/kernel/ldc.c
index 63969f610284..d68982330f66 100644
--- a/arch/sparc64/kernel/ldc.c
+++ b/arch/sparc64/kernel/ldc.c
@@ -1,6 +1,6 @@
 /* ldc.c: Logical Domain Channel link-layer protocol driver.
  *
- * Copyright (C) 2007 David S. Miller <davem@davemloft.net>
+ * Copyright (C) 2007, 2008 David S. Miller <davem@davemloft.net>
  */
 
 #include <linux/kernel.h>
@@ -23,8 +23,8 @@
 
 #define DRV_MODULE_NAME		"ldc"
 #define PFX DRV_MODULE_NAME	": "
-#define DRV_MODULE_VERSION	"1.0"
-#define DRV_MODULE_RELDATE	"June 25, 2007"
+#define DRV_MODULE_VERSION	"1.1"
+#define DRV_MODULE_RELDATE	"July 22, 2008"
 
 static char version[] __devinitdata =
 	DRV_MODULE_NAME ".c:v" DRV_MODULE_VERSION " (" DRV_MODULE_RELDATE ")\n";
@@ -1235,13 +1235,9 @@ int ldc_bind(struct ldc_channel *lp, const char *name)
 	unsigned long hv_err, flags;
 	int err = -EINVAL;
 
-	spin_lock_irqsave(&lp->lock, flags);
-
-	if (!name)
-		goto out_err;
-
-	if (lp->state != LDC_STATE_INIT)
-		goto out_err;
+	if (!name ||
+	    (lp->state != LDC_STATE_INIT))
+		return -EINVAL;
 
 	snprintf(lp->rx_irq_name, LDC_IRQ_NAME_MAX, "%s RX", name);
 	snprintf(lp->tx_irq_name, LDC_IRQ_NAME_MAX, "%s TX", name);
@@ -1250,25 +1246,32 @@ int ldc_bind(struct ldc_channel *lp, const char *name)
 			  IRQF_SAMPLE_RANDOM | IRQF_SHARED,
 			  lp->rx_irq_name, lp);
 	if (err)
-		goto out_err;
+		return err;
 
 	err = request_irq(lp->cfg.tx_irq, ldc_tx,
 			  IRQF_SAMPLE_RANDOM | IRQF_SHARED,
 			  lp->tx_irq_name, lp);
-	if (err)
-		goto out_free_rx_irq;
+	if (err) {
+		free_irq(lp->cfg.rx_irq, lp);
+		return err;
+	}
 
 
+	spin_lock_irqsave(&lp->lock, flags);
+
+	enable_irq(lp->cfg.rx_irq);
+	enable_irq(lp->cfg.tx_irq);
+
 	lp->flags |= LDC_FLAG_REGISTERED_IRQS;
 
 	err = -ENODEV;
 	hv_err = sun4v_ldc_tx_qconf(lp->id, 0, 0);
 	if (hv_err)
-		goto out_free_tx_irq;
+		goto out_free_irqs;
 
 	hv_err = sun4v_ldc_tx_qconf(lp->id, lp->tx_ra, lp->tx_num_entries);
 	if (hv_err)
-		goto out_free_tx_irq;
+		goto out_free_irqs;
 
 	hv_err = sun4v_ldc_rx_qconf(lp->id, 0, 0);
 	if (hv_err)
@@ -1304,14 +1307,11 @@ out_unmap_rx:
 out_unmap_tx:
 	sun4v_ldc_tx_qconf(lp->id, 0, 0);
 
-out_free_tx_irq:
+out_free_irqs:
 	lp->flags &= ~LDC_FLAG_REGISTERED_IRQS;
 	free_irq(lp->cfg.tx_irq, lp);
-
-out_free_rx_irq:
 	free_irq(lp->cfg.rx_irq, lp);
 
-out_err:
 	spin_unlock_irqrestore(&lp->lock, flags);
 
 	return err;

From e14fa82439d33cef67eaafc1a48960bbfa610c8e Mon Sep 17 00:00:00 2001
From: Riku Voipio <riku.voipio@iki.fi>
Date: Sat, 31 May 2008 14:43:41 +0100
Subject: [PATCH 0137/1171] leds: Add pca9532 led driver

NXP pca9532 is a LED dimmer/controller attached to i2c bus.  It allows
attaching upto 16 leds which can either be on, off or dimmed and/or blinked
with the two PWM modulators available.

This driver is a "new-style" i2c driver that adheres to the driver model and
implements the led framework api.  Since the leds connected to the driver are
platform specific, it is only useful when platform data is passed to the
driver to define what leds are connected to which pins.

Signed-off-by: Riku Voipio <riku.voipio@iki.fi>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
---
 drivers/leds/Kconfig         |   8 +
 drivers/leds/Makefile        |   1 +
 drivers/leds/leds-pca9532.c  | 337 +++++++++++++++++++++++++++++++++++
 include/linux/leds-pca9532.h |  45 +++++
 4 files changed, 391 insertions(+)
 create mode 100644 drivers/leds/leds-pca9532.c
 create mode 100644 include/linux/leds-pca9532.h

diff --git a/drivers/leds/Kconfig b/drivers/leds/Kconfig
index 86a369bc57d6..1c35dfaef721 100644
--- a/drivers/leds/Kconfig
+++ b/drivers/leds/Kconfig
@@ -103,6 +103,14 @@ config LEDS_HP6XX
 	  This option enables led support for the handheld
 	  HP Jornada 620/660/680/690.
 
+config LEDS_PCA9532
+	tristate "LED driver for PCA9532 dimmer"
+	depends on LEDS_CLASS && I2C && INPUT && EXPERIMENTAL
+	help
+	  This option enables support for NXP pca9532
+	  led controller. It is generally only usefull
+	  as a platform driver
+
 config LEDS_GPIO
 	tristate "LED Support for GPIO connected LEDs"
 	depends on LEDS_CLASS && GENERIC_GPIO
diff --git a/drivers/leds/Makefile b/drivers/leds/Makefile
index 973d626f5f4a..7156f9970fa9 100644
--- a/drivers/leds/Makefile
+++ b/drivers/leds/Makefile
@@ -16,6 +16,7 @@ obj-$(CONFIG_LEDS_WRAP)			+= leds-wrap.o
 obj-$(CONFIG_LEDS_H1940)		+= leds-h1940.o
 obj-$(CONFIG_LEDS_COBALT_QUBE)		+= leds-cobalt-qube.o
 obj-$(CONFIG_LEDS_COBALT_RAQ)		+= leds-cobalt-raq.o
+obj-$(CONFIG_LEDS_PCA9532)		+= leds-pca9532.o
 obj-$(CONFIG_LEDS_GPIO)			+= leds-gpio.o
 obj-$(CONFIG_LEDS_CM_X270)              += leds-cm-x270.o
 obj-$(CONFIG_LEDS_CLEVO_MAIL)		+= leds-clevo-mail.o
diff --git a/drivers/leds/leds-pca9532.c b/drivers/leds/leds-pca9532.c
new file mode 100644
index 000000000000..4064d4f6b33b
--- /dev/null
+++ b/drivers/leds/leds-pca9532.c
@@ -0,0 +1,337 @@
+/*
+ * pca9532.c - 16-bit Led dimmer
+ *
+ * Copyright (C) 2008 Riku Voipio <riku.voipio@movial.fi>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * Datasheet: http://www.nxp.com/acrobat/datasheets/PCA9532_3.pdf
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/i2c.h>
+#include <linux/leds.h>
+#include <linux/input.h>
+#include <linux/mutex.h>
+#include <linux/leds-pca9532.h>
+
+static const unsigned short normal_i2c[] = { /*0x60,*/ I2C_CLIENT_END};
+I2C_CLIENT_INSMOD_1(pca9532);
+
+#define PCA9532_REG_PSC(i) (0x2+(i)*2)
+#define PCA9532_REG_PWM(i) (0x3+(i)*2)
+#define PCA9532_REG_LS0  0x6
+#define LED_REG(led) ((led>>2)+PCA9532_REG_LS0)
+#define LED_NUM(led) (led & 0x3)
+
+#define ldev_to_led(c)       container_of(c, struct pca9532_led, ldev)
+
+struct pca9532_data {
+	struct i2c_client *client;
+	struct pca9532_led leds[16];
+	struct mutex update_lock;
+	struct input_dev    *idev;
+	u8 pwm[2];
+	u8 psc[2];
+};
+
+static int pca9532_probe(struct i2c_client *client,
+	const struct i2c_device_id *id);
+static int pca9532_remove(struct i2c_client *client);
+
+static const struct i2c_device_id pca9532_id[] = {
+	{ "pca9532", 0 },
+	{ }
+};
+
+MODULE_DEVICE_TABLE(i2c, pca9532_id);
+
+static struct i2c_driver pca9532_driver = {
+	.driver = {
+		.name   = "pca9532",
+	},
+	.probe  = pca9532_probe,
+	.remove = pca9532_remove,
+	.id_table = pca9532_id,
+};
+
+/* We have two pwm/blinkers, but 16 possible leds to drive. Additionaly,
+ * the clever Thecus people are using one pwm to drive the beeper. So,
+ * as a compromise we average one pwm to the values requested by all
+ * leds that are not ON/OFF.
+ * */
+static int pca9532_setpwm(struct i2c_client *client, int pwm, int blink,
+	enum led_brightness value)
+{
+	int a = 0, b = 0, i = 0;
+	struct pca9532_data *data = i2c_get_clientdata(client);
+	for (i = 0; i < 16; i++) {
+		if (data->leds[i].type == PCA9532_TYPE_LED &&
+			data->leds[i].state == PCA9532_PWM0+pwm) {
+				a++;
+				b += data->leds[i].ldev.brightness;
+		}
+	}
+	if (a == 0) {
+		dev_err(&client->dev,
+		"fear of division by zero %d/%d, wanted %d\n",
+			b, a, value);
+		return -EINVAL;
+	}
+	b = b/a;
+	if (b > 0xFF)
+		return -EINVAL;
+	mutex_lock(&data->update_lock);
+	data->pwm[pwm] = b;
+	i2c_smbus_write_byte_data(client, PCA9532_REG_PWM(pwm),
+		data->pwm[pwm]);
+	data->psc[pwm] = blink;
+	i2c_smbus_write_byte_data(client, PCA9532_REG_PSC(pwm),
+		data->psc[pwm]);
+	mutex_unlock(&data->update_lock);
+	return 0;
+}
+
+/* Set LED routing */
+static void pca9532_setled(struct pca9532_led *led)
+{
+	struct i2c_client *client = led->client;
+	struct pca9532_data *data = i2c_get_clientdata(client);
+	char reg;
+
+	mutex_lock(&data->update_lock);
+	reg = i2c_smbus_read_byte_data(client, LED_REG(led->id));
+	/* zero led bits */
+	reg = reg & ~(0x3<<LED_NUM(led->id)*2);
+	/* set the new value */
+	reg = reg | (led->state << LED_NUM(led->id)*2);
+	i2c_smbus_write_byte_data(client, LED_REG(led->id), reg);
+	mutex_unlock(&data->update_lock);
+}
+
+static void pca9532_set_brightness(struct led_classdev *led_cdev,
+	enum led_brightness value)
+{
+	int err = 0;
+	struct pca9532_led *led = ldev_to_led(led_cdev);
+
+	if (value == LED_OFF)
+		led->state = PCA9532_OFF;
+	else if (value == LED_FULL)
+		led->state = PCA9532_ON;
+	else {
+		led->state = PCA9532_PWM0; /* Thecus: hardcode one pwm */
+		err = pca9532_setpwm(led->client, 0, 0, value);
+		if (err)
+			return; /* XXX: led api doesn't allow error code? */
+	}
+	pca9532_setled(led);
+}
+
+static int pca9532_set_blink(struct led_classdev *led_cdev,
+	unsigned long *delay_on, unsigned long *delay_off)
+{
+	struct pca9532_led *led = ldev_to_led(led_cdev);
+	struct i2c_client *client = led->client;
+	int psc;
+
+	if (*delay_on == 0 && *delay_off == 0) {
+	/* led subsystem ask us for a blink rate */
+		*delay_on  = 1000;
+		*delay_off = 1000;
+	}
+	if (*delay_on != *delay_off || *delay_on > 1690 || *delay_on < 6)
+		return -EINVAL;
+
+	/* Thecus specific: only use PSC/PWM 0 */
+	psc = (*delay_on * 152-1)/1000;
+	return pca9532_setpwm(client, 0, psc, led_cdev->brightness);
+}
+
+int pca9532_event(struct input_dev *dev, unsigned int type, unsigned int code,
+	int value)
+{
+	struct pca9532_data *data = input_get_drvdata(dev);
+
+	if (type != EV_SND && (code != SND_BELL || code != SND_TONE))
+		return -1;
+
+	/* XXX: allow different kind of beeps with psc/pwm modifications */
+	if (value > 1 && value < 32767)
+		data->pwm[1] = 127;
+	else
+		data->pwm[1] = 0;
+
+	dev_info(&dev->dev, "setting beep to %d \n", data->pwm[1]);
+	mutex_lock(&data->update_lock);
+	i2c_smbus_write_byte_data(data->client, PCA9532_REG_PWM(1),
+		data->pwm[1]);
+	mutex_unlock(&data->update_lock);
+
+	return 0;
+}
+
+static int pca9532_configure(struct i2c_client *client,
+	struct pca9532_data *data, struct pca9532_platform_data *pdata)
+{
+	int i, err = 0;
+
+	for (i = 0; i < 2; i++)	{
+		data->pwm[i] = pdata->pwm[i];
+		data->psc[i] = pdata->psc[i];
+		i2c_smbus_write_byte_data(client, PCA9532_REG_PWM(i),
+			data->pwm[i]);
+		i2c_smbus_write_byte_data(client, PCA9532_REG_PSC(i),
+			data->psc[i]);
+	}
+
+	for (i = 0; i < 16; i++) {
+		struct pca9532_led *led = &data->leds[i];
+		struct pca9532_led *pled = &pdata->leds[i];
+		led->client = client;
+		led->id = i;
+		led->type = pled->type;
+		switch (led->type) {
+		case PCA9532_TYPE_NONE:
+			break;
+		case PCA9532_TYPE_LED:
+			led->state = pled->state;
+			led->name =  pled->name;
+			led->ldev.name = led->name;
+			led->ldev.brightness = LED_OFF;
+			led->ldev.brightness_set = pca9532_set_brightness;
+			led->ldev.blink_set = pca9532_set_blink;
+			if (led_classdev_register(&client->dev,
+				&led->ldev) < 0)	{
+				dev_err(&client->dev,
+					"couldn't register LED %s\n",
+					led->name);
+				goto exit;
+			}
+			pca9532_setled(led);
+			break;
+		case PCA9532_TYPE_N2100_BEEP:
+			BUG_ON(data->idev);
+			led->state = PCA9532_PWM1;
+			pca9532_setled(led);
+			data->idev = input_allocate_device();
+			if (data->idev == NULL) {
+				err = -ENOMEM;
+				goto exit;
+			}
+			data->idev->name = pled->name;
+			data->idev->phys = "i2c/pca9532";
+			data->idev->id.bustype = BUS_HOST;
+			data->idev->id.vendor  = 0x001f;
+			data->idev->id.product = 0x0001;
+			data->idev->id.version = 0x0100;
+			data->idev->evbit[0] = BIT_MASK(EV_SND);
+			data->idev->sndbit[0] = BIT_MASK(SND_BELL) |
+						BIT_MASK(SND_TONE);
+			data->idev->event = pca9532_event;
+			input_set_drvdata(data->idev, data);
+			err = input_register_device(data->idev);
+			if (err) {
+				input_free_device(data->idev);
+				data->idev = NULL;
+				goto exit;
+			}
+			break;
+		}
+	}
+	return 0;
+
+exit:
+	if (i > 0)
+		for (i = i - 1; i >= 0; i--)
+			switch (data->leds[i].type) {
+			case PCA9532_TYPE_NONE:
+				break;
+			case PCA9532_TYPE_LED:
+				led_classdev_unregister(&data->leds[i].ldev);
+				break;
+			case PCA9532_TYPE_N2100_BEEP:
+				if (data->idev != NULL) {
+					input_unregister_device(data->idev);
+					input_free_device(data->idev);
+					data->idev = NULL;
+				}
+				break;
+			}
+
+	return err;
+
+}
+
+static int pca9532_probe(struct i2c_client *client,
+	const struct i2c_device_id *id)
+{
+	struct pca9532_data *data = i2c_get_clientdata(client);
+	struct pca9532_platform_data *pca9532_pdata = client->dev.platform_data;
+
+	if (!i2c_check_functionality(client->adapter,
+		I2C_FUNC_SMBUS_BYTE_DATA))
+		return -EIO;
+
+	data = kzalloc(sizeof(struct pca9532_data), GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
+
+	dev_info(&client->dev, "setting platform data\n");
+	i2c_set_clientdata(client, data);
+	data->client = client;
+	mutex_init(&data->update_lock);
+
+	if (pca9532_pdata == NULL)
+		return -EIO;
+
+	pca9532_configure(client, data, pca9532_pdata);
+	return 0;
+
+}
+
+static int pca9532_remove(struct i2c_client *client)
+{
+	struct pca9532_data *data = i2c_get_clientdata(client);
+	int i;
+	for (i = 0; i < 16; i++)
+		switch (data->leds[i].type) {
+		case PCA9532_TYPE_NONE:
+			break;
+		case PCA9532_TYPE_LED:
+			led_classdev_unregister(&data->leds[i].ldev);
+			break;
+		case PCA9532_TYPE_N2100_BEEP:
+			if (data->idev != NULL) {
+				input_unregister_device(data->idev);
+				input_free_device(data->idev);
+				data->idev = NULL;
+			}
+			break;
+		}
+
+	kfree(data);
+	i2c_set_clientdata(client, NULL);
+	return 0;
+}
+
+static int __init pca9532_init(void)
+{
+	return i2c_add_driver(&pca9532_driver);
+}
+
+static void __exit pca9532_exit(void)
+{
+	i2c_del_driver(&pca9532_driver);
+}
+
+MODULE_AUTHOR("Riku Voipio <riku.voipio@movial.fi>");
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("PCA 9532 LED dimmer");
+
+module_init(pca9532_init);
+module_exit(pca9532_exit);
+
diff --git a/include/linux/leds-pca9532.h b/include/linux/leds-pca9532.h
new file mode 100644
index 000000000000..81b4207deb95
--- /dev/null
+++ b/include/linux/leds-pca9532.h
@@ -0,0 +1,45 @@
+/*
+ * pca9532.h - platform data structure for pca9532 led controller
+ *
+ * Copyright (C) 2008 Riku Voipio <riku.voipio@movial.fi>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * Datasheet: http://www.nxp.com/acrobat/datasheets/PCA9532_3.pdf
+ *
+ */
+
+#ifndef __LINUX_PCA9532_H
+#define __LINUX_PCA9532_H
+
+#include <linux/leds.h>
+
+enum pca9532_state {
+	PCA9532_OFF  = 0x0,
+	PCA9532_ON   = 0x1,
+	PCA9532_PWM0 = 0x2,
+	PCA9532_PWM1 = 0x3
+};
+
+enum pca9532_type { PCA9532_TYPE_NONE, PCA9532_TYPE_LED,
+	PCA9532_TYPE_N2100_BEEP };
+
+struct pca9532_led {
+	u8 id;
+	struct i2c_client *client;
+	char *name;
+	struct led_classdev ldev;
+	enum pca9532_type type;
+	enum pca9532_state state;
+};
+
+struct pca9532_platform_data {
+	struct pca9532_led leds[16];
+	u8 pwm[2];
+	u8 psc[2];
+};
+
+#endif /* __LINUX_PCA9532_H */
+

From 30be0486791fb637e758c771956c8f73bef3467c Mon Sep 17 00:00:00 2001
From: Riku Voipio <riku.voipio@iki.fi>
Date: Sat, 31 May 2008 14:45:16 +0100
Subject: [PATCH 0138/1171] leds: Add pca9532 platform data for Thecus N2100

Thecus N2100 has leds and a buzzer attached to a pca9532 controller.  Attach
the driver to the i2c bus and define the pca9532 pin coniguration for this
platform in n2100_leds.

With this patch, support for N2100 should be complete in mainline Linux.

Signed-off-by: Riku Voipio <riku.voipio@iki.fi>
Acked-by: Lennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
---
 arch/arm/mach-iop32x/n2100.c | 52 ++++++++++++++++++++++++++++++++++++
 1 file changed, 52 insertions(+)

diff --git a/arch/arm/mach-iop32x/n2100.c b/arch/arm/mach-iop32x/n2100.c
index 2741063bf361..28f164ea4726 100644
--- a/arch/arm/mach-iop32x/n2100.c
+++ b/arch/arm/mach-iop32x/n2100.c
@@ -17,6 +17,7 @@
 #include <linux/mm.h>
 #include <linux/init.h>
 #include <linux/f75375s.h>
+#include <linux/leds-pca9532.h>
 #include <linux/delay.h>
 #include <linux/kernel.h>
 #include <linux/pci.h>
@@ -206,6 +207,53 @@ static struct f75375s_platform_data n2100_f75375s = {
 	.pwm_enable = { 0, 0 },
 };
 
+static struct pca9532_platform_data n2100_leds = {
+	.leds = {
+	{	.name = "n2100:red:satafail0",
+		.state = PCA9532_OFF,
+		.type = PCA9532_TYPE_LED,
+	},
+	{	.name = "n2100:red:satafail1",
+		.state = PCA9532_OFF,
+		.type = PCA9532_TYPE_LED,
+	},
+	{	.name = "n2100:blue:usb",
+		.state = PCA9532_OFF,
+		.type = PCA9532_TYPE_LED,
+	},
+	{ 	.type = PCA9532_TYPE_NONE },
+
+	{ 	.type = PCA9532_TYPE_NONE },
+	{ 	.type = PCA9532_TYPE_NONE },
+	{ 	.type = PCA9532_TYPE_NONE },
+	{	.name = "n2100:red:usb",
+		.state = PCA9532_OFF,
+		.type = PCA9532_TYPE_LED,
+	},
+
+	{	.type = PCA9532_TYPE_NONE }, /* power OFF gpio */
+	{	.type = PCA9532_TYPE_NONE }, /* reset gpio */
+	{	.type = PCA9532_TYPE_NONE },
+	{	.type = PCA9532_TYPE_NONE },
+
+	{	.type = PCA9532_TYPE_NONE },
+	{	.name = "n2100:orange:system",
+		.state = PCA9532_OFF,
+		.type = PCA9532_TYPE_LED,
+	},
+	{	.name = "n2100:red:system",
+		.state = PCA9532_OFF,
+		.type = PCA9532_TYPE_LED,
+	},
+	{	.name = "N2100 beeper"  ,
+		.state =  PCA9532_OFF,
+		.type = PCA9532_TYPE_N2100_BEEP,
+	},
+	},
+	.psc = { 0, 0 },
+	.pwm = { 0, 0 },
+};
+
 static struct i2c_board_info __initdata n2100_i2c_devices[] = {
 	{
 		I2C_BOARD_INFO("rs5c372b", 0x32),
@@ -214,6 +262,10 @@ static struct i2c_board_info __initdata n2100_i2c_devices[] = {
 		I2C_BOARD_INFO("f75375", 0x2e),
 		.platform_data = &n2100_f75375s,
 	},
+	{
+		I2C_BOARD_INFO("pca9532", 0x60),
+		.platform_data = &n2100_leds,
+	},
 };
 
 /*

From e49575f46cdb40014e14789a18e637f8fb917317 Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Sat, 31 May 2008 15:18:55 +0100
Subject: [PATCH 0139/1171] leds: fix unsigned value overflow in atmel pwm
 driver

Fix an unsigned value overflow in the error handling code in the
Atmel PWM driver.

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
---
 drivers/leds/leds-atmel-pwm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/leds/leds-atmel-pwm.c b/drivers/leds/leds-atmel-pwm.c
index 28db6c1444ed..52297c3ab246 100644
--- a/drivers/leds/leds-atmel-pwm.c
+++ b/drivers/leds/leds-atmel-pwm.c
@@ -37,7 +37,7 @@ static int __init pwmled_probe(struct platform_device *pdev)
 {
 	const struct gpio_led_platform_data	*pdata;
 	struct pwmled				*leds;
-	unsigned				i;
+	int					i;
 	int					status;
 
 	pdata = pdev->dev.platform_data;

From 781a54e7664cc0089287a90d27086e9656ac68a1 Mon Sep 17 00:00:00 2001
From: Anton Vorontsov <avorontsov@ru.mvista.com>
Date: Sat, 31 May 2008 15:23:19 +0100
Subject: [PATCH 0140/1171] leds: mark led_classdev.default_trigger as const

LED classdev core doesn't modify memory pointed by the default_trigger,
so mark it as const and we'll able to pass const char *s without getting
compiler warnings.

Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
---
 include/linux/leds.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/linux/leds.h b/include/linux/leds.h
index 519df72e939d..e7a5e89932fe 100644
--- a/include/linux/leds.h
+++ b/include/linux/leds.h
@@ -48,7 +48,7 @@ struct led_classdev {
 
 	struct device		*dev;
 	struct list_head	 node;			/* LED Device list */
-	char			*default_trigger;	/* Trigger to use */
+	const char		*default_trigger;	/* Trigger to use */
 
 #ifdef CONFIG_LEDS_TRIGGERS
 	/* Protects the trigger data below */

From dd1160dc1842ae172495a6da274a77e35c593ed8 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Mon, 9 Jun 2008 22:00:49 +0100
Subject: [PATCH 0141/1171] leds: Fix sparse warnings in leds-h1940 driver

Fixes the following sparse errors:
drivers/leds/leds-h1940.c:26:6: warning: symbol 'h1940_greenled_set' was not declared. Should it be static?
drivers/leds/leds-h1940.c:55:6: warning: symbol 'h1940_redled_set' was not declared. Should it be static?
drivers/leds/leds-h1940.c:85:6: warning: symbol 'h1940_blueled_set' was not declared. Should it be static?

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
---
 drivers/leds/leds-h1940.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/leds/leds-h1940.c b/drivers/leds/leds-h1940.c
index bcec42230389..73c705021686 100644
--- a/drivers/leds/leds-h1940.c
+++ b/drivers/leds/leds-h1940.c
@@ -23,7 +23,8 @@
 /*
  * Green led.
  */
-void h1940_greenled_set(struct led_classdev *led_dev, enum led_brightness value)
+static void h1940_greenled_set(struct led_classdev *led_dev,
+			       enum led_brightness value)
 {
 	switch (value) {
 	case LED_HALF:
@@ -52,7 +53,8 @@ static struct led_classdev h1940_greenled = {
 /*
  * Red led.
  */
-void h1940_redled_set(struct led_classdev *led_dev, enum led_brightness value)
+static void h1940_redled_set(struct led_classdev *led_dev,
+			     enum led_brightness value)
 {
 	switch (value) {
 	case LED_HALF:
@@ -82,7 +84,8 @@ static struct led_classdev h1940_redled = {
  * Blue led.
  * (it can only be blue flashing led)
  */
-void h1940_blueled_set(struct led_classdev *led_dev, enum led_brightness value)
+static void h1940_blueled_set(struct led_classdev *led_dev,
+			      enum led_brightness value)
 {
 	if (value) {
 		/* flashing Blue */

From f46e9203d9a100bae216cc06e17f2e77351aa8d8 Mon Sep 17 00:00:00 2001
From: Nate Case <ncase@xes-inc.com>
Date: Wed, 16 Jul 2008 22:49:55 +0100
Subject: [PATCH 0142/1171] leds: Add support for Philips PCA955x I2C LED
 drivers

This driver supports the PCA9550, PCA9551, PCA9552, and PCA9553
LED driver chips.

Signed-off-by: Nate Case <ncase@xes-inc.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
---
 drivers/leds/Kconfig        |   8 +
 drivers/leds/Makefile       |   1 +
 drivers/leds/leds-pca955x.c | 384 ++++++++++++++++++++++++++++++++++++
 include/linux/leds.h        |  14 ++
 4 files changed, 407 insertions(+)
 create mode 100644 drivers/leds/leds-pca955x.c

diff --git a/drivers/leds/Kconfig b/drivers/leds/Kconfig
index 1c35dfaef721..9556262dda5a 100644
--- a/drivers/leds/Kconfig
+++ b/drivers/leds/Kconfig
@@ -155,6 +155,14 @@ config LEDS_CLEVO_MAIL
 	  To compile this driver as a module, choose M here: the
 	  module will be called leds-clevo-mail.
 
+config LEDS_PCA955X
+	tristate "LED Support for PCA955x I2C chips"
+	depends on LEDS_CLASS && I2C
+	help
+	  This option enables support for LEDs connected to PCA955x
+	  LED driver chips accessed via the I2C bus.  Supported
+	  devices include PCA9550, PCA9551, PCA9552, and PCA9553.
+
 comment "LED Triggers"
 
 config LEDS_TRIGGERS
diff --git a/drivers/leds/Makefile b/drivers/leds/Makefile
index 7156f9970fa9..ff7982b44565 100644
--- a/drivers/leds/Makefile
+++ b/drivers/leds/Makefile
@@ -22,6 +22,7 @@ obj-$(CONFIG_LEDS_CM_X270)              += leds-cm-x270.o
 obj-$(CONFIG_LEDS_CLEVO_MAIL)		+= leds-clevo-mail.o
 obj-$(CONFIG_LEDS_HP6XX)		+= leds-hp6xx.o
 obj-$(CONFIG_LEDS_FSG)			+= leds-fsg.o
+obj-$(CONFIG_LEDS_PCA955X)		+= leds-pca955x.o
 
 # LED Triggers
 obj-$(CONFIG_LEDS_TRIGGER_TIMER)	+= ledtrig-timer.o
diff --git a/drivers/leds/leds-pca955x.c b/drivers/leds/leds-pca955x.c
new file mode 100644
index 000000000000..146c06972863
--- /dev/null
+++ b/drivers/leds/leds-pca955x.c
@@ -0,0 +1,384 @@
+/*
+ * Copyright 2007-2008 Extreme Engineering Solutions, Inc.
+ *
+ * Author: Nate Case <ncase@xes-inc.com>
+ *
+ * This file is subject to the terms and conditions of version 2 of
+ * the GNU General Public License.  See the file COPYING in the main
+ * directory of this archive for more details.
+ *
+ * LED driver for various PCA955x I2C LED drivers
+ *
+ * Supported devices:
+ *
+ *	Device		Description		7-bit slave address
+ *	------		-----------		-------------------
+ *	PCA9550		2-bit driver		0x60 .. 0x61
+ *	PCA9551		8-bit driver		0x60 .. 0x67
+ *	PCA9552		16-bit driver		0x60 .. 0x67
+ *	PCA9553/01	4-bit driver		0x62
+ *	PCA9553/02	4-bit driver		0x63
+ *
+ * Philips PCA955x LED driver chips follow a register map as shown below:
+ *
+ *	Control Register		Description
+ *	----------------		-----------
+ *	0x0				Input register 0
+ *					..
+ *	NUM_INPUT_REGS - 1		Last Input register X
+ *
+ *	NUM_INPUT_REGS			Frequency prescaler 0
+ *	NUM_INPUT_REGS + 1		PWM register 0
+ *	NUM_INPUT_REGS + 2		Frequency prescaler 1
+ *	NUM_INPUT_REGS + 3		PWM register 1
+ *
+ *	NUM_INPUT_REGS + 4		LED selector 0
+ *	NUM_INPUT_REGS + 4
+ *	    + NUM_LED_REGS - 1		Last LED selector
+ *
+ *  where NUM_INPUT_REGS and NUM_LED_REGS vary depending on how many
+ *  bits the chip supports.
+ */
+
+#include <linux/module.h>
+#include <linux/delay.h>
+#include <linux/string.h>
+#include <linux/ctype.h>
+#include <linux/leds.h>
+#include <linux/err.h>
+#include <linux/i2c.h>
+#include <linux/workqueue.h>
+
+/* LED select registers determine the source that drives LED outputs */
+#define PCA955X_LS_LED_ON	0x0	/* Output LOW */
+#define PCA955X_LS_LED_OFF	0x1	/* Output HI-Z */
+#define PCA955X_LS_BLINK0	0x2	/* Blink at PWM0 rate */
+#define PCA955X_LS_BLINK1	0x3	/* Blink at PWM1 rate */
+
+enum pca955x_type {
+	pca9550,
+	pca9551,
+	pca9552,
+	pca9553,
+};
+
+struct pca955x_chipdef {
+	int			bits;
+	u8			slv_addr;	/* 7-bit slave address mask */
+	int			slv_addr_shift;	/* Number of bits to ignore */
+};
+
+static struct pca955x_chipdef pca955x_chipdefs[] = {
+	[pca9550] = {
+		.bits		= 2,
+		.slv_addr	= /* 110000x */ 0x60,
+		.slv_addr_shift	= 1,
+	},
+	[pca9551] = {
+		.bits		= 8,
+		.slv_addr	= /* 1100xxx */ 0x60,
+		.slv_addr_shift	= 3,
+	},
+	[pca9552] = {
+		.bits		= 16,
+		.slv_addr	= /* 1100xxx */ 0x60,
+		.slv_addr_shift	= 3,
+	},
+	[pca9553] = {
+		.bits		= 4,
+		.slv_addr	= /* 110001x */ 0x62,
+		.slv_addr_shift	= 1,
+	},
+};
+
+static const struct i2c_device_id pca955x_id[] = {
+	{ "pca9550", pca9550 },
+	{ "pca9551", pca9551 },
+	{ "pca9552", pca9552 },
+	{ "pca9553", pca9553 },
+	{ }
+};
+MODULE_DEVICE_TABLE(i2c, pca955x_id);
+
+struct pca955x_led {
+	struct pca955x_chipdef	*chipdef;
+	struct i2c_client	*client;
+	struct work_struct	work;
+	spinlock_t		lock;
+	enum led_brightness	brightness;
+	struct led_classdev	led_cdev;
+	int			led_num;	/* 0 .. 15 potentially */
+	char			name[32];
+};
+
+/* 8 bits per input register */
+static inline int pca95xx_num_input_regs(int bits)
+{
+	return (bits + 7) / 8;
+}
+
+/* 4 bits per LED selector register */
+static inline int pca95xx_num_led_regs(int bits)
+{
+	return (bits + 3)  / 4;
+}
+
+/*
+ * Return an LED selector register value based on an existing one, with
+ * the appropriate 2-bit state value set for the given LED number (0-3).
+ */
+static inline u8 pca955x_ledsel(u8 oldval, int led_num, int state)
+{
+	return (oldval & (~(0x3 << (led_num << 1)))) |
+		((state & 0x3) << (led_num << 1));
+}
+
+/*
+ * Write to frequency prescaler register, used to program the
+ * period of the PWM output.  period = (PSCx + 1) / 38
+ */
+static void pca955x_write_psc(struct i2c_client *client, int n, u8 val)
+{
+	struct pca955x_led *pca955x = i2c_get_clientdata(client);
+
+	i2c_smbus_write_byte_data(client,
+		pca95xx_num_input_regs(pca955x->chipdef->bits) + 2*n,
+		val);
+}
+
+/*
+ * Write to PWM register, which determines the duty cycle of the
+ * output.  LED is OFF when the count is less than the value of this
+ * register, and ON when it is greater.  If PWMx == 0, LED is always OFF.
+ *
+ * Duty cycle is (256 - PWMx) / 256
+ */
+static void pca955x_write_pwm(struct i2c_client *client, int n, u8 val)
+{
+	struct pca955x_led *pca955x = i2c_get_clientdata(client);
+
+	i2c_smbus_write_byte_data(client,
+		pca95xx_num_input_regs(pca955x->chipdef->bits) + 1 + 2*n,
+		val);
+}
+
+/*
+ * Write to LED selector register, which determines the source that
+ * drives the LED output.
+ */
+static void pca955x_write_ls(struct i2c_client *client, int n, u8 val)
+{
+	struct pca955x_led *pca955x = i2c_get_clientdata(client);
+
+	i2c_smbus_write_byte_data(client,
+		pca95xx_num_input_regs(pca955x->chipdef->bits) + 4 + n,
+		val);
+}
+
+/*
+ * Read the LED selector register, which determines the source that
+ * drives the LED output.
+ */
+static u8 pca955x_read_ls(struct i2c_client *client, int n)
+{
+	struct pca955x_led *pca955x = i2c_get_clientdata(client);
+
+	return (u8) i2c_smbus_read_byte_data(client,
+		pca95xx_num_input_regs(pca955x->chipdef->bits) + 4 + n);
+}
+
+static void pca955x_led_work(struct work_struct *work)
+{
+	struct pca955x_led *pca955x;
+	u8 ls;
+	int chip_ls;	/* which LSx to use (0-3 potentially) */
+	int ls_led;	/* which set of bits within LSx to use (0-3) */
+
+	pca955x = container_of(work, struct pca955x_led, work);
+	chip_ls = pca955x->led_num / 4;
+	ls_led = pca955x->led_num % 4;
+
+	ls = pca955x_read_ls(pca955x->client, chip_ls);
+
+	switch (pca955x->brightness) {
+	case LED_FULL:
+		ls = pca955x_ledsel(ls, ls_led, PCA955X_LS_LED_ON);
+		break;
+	case LED_OFF:
+		ls = pca955x_ledsel(ls, ls_led, PCA955X_LS_LED_OFF);
+		break;
+	case LED_HALF:
+		ls = pca955x_ledsel(ls, ls_led, PCA955X_LS_BLINK0);
+		break;
+	default:
+		/*
+		 * Use PWM1 for all other values.  This has the unwanted
+		 * side effect of making all LEDs on the chip share the
+		 * same brightness level if set to a value other than
+		 * OFF, HALF, or FULL.  But, this is probably better than
+		 * just turning off for all other values.
+		 */
+		pca955x_write_pwm(pca955x->client, 1, 255-pca955x->brightness);
+		ls = pca955x_ledsel(ls, ls_led, PCA955X_LS_BLINK1);
+		break;
+	}
+
+	pca955x_write_ls(pca955x->client, chip_ls, ls);
+}
+
+void pca955x_led_set(struct led_classdev *led_cdev, enum led_brightness value)
+{
+	struct pca955x_led *pca955x;
+
+	pca955x = container_of(led_cdev, struct pca955x_led, led_cdev);
+
+	spin_lock(&pca955x->lock);
+	pca955x->brightness = value;
+
+	/*
+	 * Must use workqueue for the actual I/O since I2C operations
+	 * can sleep.
+	 */
+	schedule_work(&pca955x->work);
+
+	spin_unlock(&pca955x->lock);
+}
+
+static int __devinit pca955x_probe(struct i2c_client *client,
+					const struct i2c_device_id *id)
+{
+	struct pca955x_led *pca955x;
+	int i;
+	int err = -ENODEV;
+	struct pca955x_chipdef *chip;
+	struct i2c_adapter *adapter;
+	struct led_platform_data *pdata;
+
+	chip = &pca955x_chipdefs[id->driver_data];
+	adapter = to_i2c_adapter(client->dev.parent);
+	pdata = client->dev.platform_data;
+
+	/* Make sure the slave address / chip type combo given is possible */
+	if ((client->addr & ~((1 << chip->slv_addr_shift) - 1)) !=
+	    chip->slv_addr) {
+		dev_err(&client->dev, "invalid slave address %02x\n",
+				client->addr);
+		return -ENODEV;
+	}
+
+	printk(KERN_INFO "leds-pca955x: Using %s %d-bit LED driver at "
+			"slave address 0x%02x\n",
+			id->name, chip->bits, client->addr);
+
+	if (!i2c_check_functionality(adapter, I2C_FUNC_I2C))
+		return -EIO;
+
+	if (pdata) {
+		if (pdata->num_leds != chip->bits) {
+			dev_err(&client->dev, "board info claims %d LEDs"
+					" on a %d-bit chip\n",
+					pdata->num_leds, chip->bits);
+			return -ENODEV;
+		}
+	}
+
+	for (i = 0; i < chip->bits; i++) {
+		pca955x = kzalloc(sizeof(struct pca955x_led), GFP_KERNEL);
+		if (!pca955x) {
+			err = -ENOMEM;
+			goto exit;
+		}
+
+		pca955x->chipdef = chip;
+		pca955x->client = client;
+		pca955x->led_num = i;
+		/* Platform data can specify LED names and default triggers */
+		if (pdata) {
+			if (pdata->leds[i].name)
+				snprintf(pca955x->name, 32, "pca955x:%s",
+							pdata->leds[i].name);
+			if (pdata->leds[i].default_trigger)
+				pca955x->led_cdev.default_trigger =
+					pdata->leds[i].default_trigger;
+		} else {
+			snprintf(pca955x->name, 32, "pca955x:%d", i);
+		}
+		spin_lock_init(&pca955x->lock);
+
+		pca955x->led_cdev.name = pca955x->name;
+		pca955x->led_cdev.brightness_set =
+				pca955x_led_set;
+
+		/*
+		 * Client data is a pointer to the _first_ pca955x_led
+		 * struct
+		 */
+		if (i == 0)
+			i2c_set_clientdata(client, pca955x);
+
+		INIT_WORK(&(pca955x->work), pca955x_led_work);
+
+		led_classdev_register(&client->dev, &(pca955x->led_cdev));
+	}
+
+	/* Turn off LEDs */
+	for (i = 0; i < pca95xx_num_led_regs(chip->bits); i++)
+		pca955x_write_ls(client, i, 0x55);
+
+	/* PWM0 is used for half brightness or 50% duty cycle */
+	pca955x_write_pwm(client, 0, 255-LED_HALF);
+
+	/* PWM1 is used for variable brightness, default to OFF */
+	pca955x_write_pwm(client, 1, 0);
+
+	/* Set to fast frequency so we do not see flashing */
+	pca955x_write_psc(client, 0, 0);
+	pca955x_write_psc(client, 1, 0);
+
+	return 0;
+exit:
+	return err;
+}
+
+static int __devexit pca955x_remove(struct i2c_client *client)
+{
+	struct pca955x_led *pca955x = i2c_get_clientdata(client);
+	int leds = pca955x->chipdef->bits;
+	int i;
+
+	for (i = 0; i < leds; i++) {
+		led_classdev_unregister(&(pca955x->led_cdev));
+		cancel_work_sync(&(pca955x->work));
+		kfree(pca955x);
+		pca955x = pca955x + 1;
+	}
+
+	return 0;
+}
+
+static struct i2c_driver pca955x_driver = {
+	.driver = {
+		.name	= "leds-pca955x",
+		.owner	= THIS_MODULE,
+	},
+	.probe	= pca955x_probe,
+	.remove	= __devexit_p(pca955x_remove),
+	.id_table = pca955x_id,
+};
+
+static int __init pca955x_leds_init(void)
+{
+	return i2c_add_driver(&pca955x_driver);
+}
+
+static void __exit pca955x_leds_exit(void)
+{
+	i2c_del_driver(&pca955x_driver);
+}
+
+module_init(pca955x_leds_init);
+module_exit(pca955x_leds_exit);
+
+MODULE_AUTHOR("Nate Case <ncase@xes-inc.com>");
+MODULE_DESCRIPTION("PCA955x LED driver");
+MODULE_LICENSE("GPL v2");
diff --git a/include/linux/leds.h b/include/linux/leds.h
index e7a5e89932fe..d41ccb56146a 100644
--- a/include/linux/leds.h
+++ b/include/linux/leds.h
@@ -118,6 +118,20 @@ extern void ledtrig_ide_activity(void);
 #define ledtrig_ide_activity() do {} while(0)
 #endif
 
+/*
+ * Generic LED platform data for describing LED names and default triggers.
+ */
+struct led_info {
+	const char	*name;
+	char		*default_trigger;
+	int		flags;
+};
+
+struct led_platform_data {
+	int		num_leds;
+	struct led_info	*leds;
+};
+
 /* For the leds-gpio driver */
 struct gpio_led {
 	const char *name;

From fe3025b55c8ed06929afe94e9c9095fc19d15aa0 Mon Sep 17 00:00:00 2001
From: Dmitry Baryshkov <dbaryshkov@gmail.com>
Date: Wed, 16 Jul 2008 22:51:14 +0100
Subject: [PATCH 0143/1171] leds: Ensure led->trigger is set earlier

Make sure led->trigger is valid before calling trigger->activate

Signed-off-by: Dmitry Baryshkov <dbaryshkov@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
---
 drivers/leds/led-triggers.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/leds/led-triggers.c b/drivers/leds/led-triggers.c
index 0f242b3f09b6..f910eaffe3a6 100644
--- a/drivers/leds/led-triggers.c
+++ b/drivers/leds/led-triggers.c
@@ -111,16 +111,17 @@ void led_trigger_set(struct led_classdev *led_cdev, struct led_trigger *trigger)
 			flags);
 		if (led_cdev->trigger->deactivate)
 			led_cdev->trigger->deactivate(led_cdev);
+		led_cdev->trigger = NULL;
 		led_set_brightness(led_cdev, LED_OFF);
 	}
 	if (trigger) {
 		write_lock_irqsave(&trigger->leddev_list_lock, flags);
 		list_add_tail(&led_cdev->trig_list, &trigger->led_cdevs);
 		write_unlock_irqrestore(&trigger->leddev_list_lock, flags);
+		led_cdev->trigger = trigger;
 		if (trigger->activate)
 			trigger->activate(led_cdev);
 	}
-	led_cdev->trigger = trigger;
 }
 EXPORT_SYMBOL_GPL(led_trigger_set);
 

From 7be35c72e6454059a33ad844153349973d22fcb7 Mon Sep 17 00:00:00 2001
From: Matthew Garrett <mjg@redhat.com>
Date: Mon, 9 Jun 2008 21:56:16 +0100
Subject: [PATCH 0144/1171] backlight: Add Nvidia-based Apple Macbook Pro
 backlight driver

Nvidia-based Apple Macbook Pros don't appear to handle backlight control
through the graphics card registers or ACPI, but instead trigger changes
via SMI calls. This driver registers a generic backlight device that
lets existing userspace deal with it. Code derived from Julien Blache's
Pommed application.

Signed-off-by: Julien Blache <jb@jblache.org>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
---
 drivers/video/backlight/Kconfig         |   9 ++
 drivers/video/backlight/Makefile        |   2 +
 drivers/video/backlight/mbp_nvidia_bl.c | 116 ++++++++++++++++++++++++
 3 files changed, 127 insertions(+)
 create mode 100644 drivers/video/backlight/mbp_nvidia_bl.c

diff --git a/drivers/video/backlight/Kconfig b/drivers/video/backlight/Kconfig
index 30bf7f2f1635..62547bd2ea49 100644
--- a/drivers/video/backlight/Kconfig
+++ b/drivers/video/backlight/Kconfig
@@ -119,3 +119,12 @@ config BACKLIGHT_PWM
 	help
 	  If you have a LCD backlight adjustable by PWM, say Y to enable
 	  this driver.
+
+config BACKLIGHT_MBP_NVIDIA
+       tristate "MacBook Pro Nvidia Backlight Driver"
+       depends on BACKLIGHT_CLASS_DEVICE && X86
+       default n
+       help
+         If you have an Apple Macbook Pro with Nvidia graphics hardware say Y
+	 to enable a driver for its backlight
+
diff --git a/drivers/video/backlight/Makefile b/drivers/video/backlight/Makefile
index b51a7cd12500..c7c4d95fdc1c 100644
--- a/drivers/video/backlight/Makefile
+++ b/drivers/video/backlight/Makefile
@@ -11,3 +11,5 @@ obj-$(CONFIG_BACKLIGHT_OMAP1)	+= omap1_bl.o
 obj-$(CONFIG_BACKLIGHT_PROGEAR) += progear_bl.o
 obj-$(CONFIG_BACKLIGHT_CARILLO_RANCH) += cr_bllcd.o
 obj-$(CONFIG_BACKLIGHT_PWM)	+= pwm_bl.o
+obj-$(CONFIG_BACKLIGHT_MBP_NVIDIA) += mbp_nvidia_bl.o
+
diff --git a/drivers/video/backlight/mbp_nvidia_bl.c b/drivers/video/backlight/mbp_nvidia_bl.c
new file mode 100644
index 000000000000..385cba40ea87
--- /dev/null
+++ b/drivers/video/backlight/mbp_nvidia_bl.c
@@ -0,0 +1,116 @@
+/*
+ *  Backlight Driver for Nvidia 8600 in Macbook Pro
+ *
+ *  Copyright (c) Red Hat <mjg@redhat.com>
+ *  Based on code from Pommed:
+ *  Copyright (C) 2006 Nicolas Boichat <nicolas @boichat.ch>
+ *  Copyright (C) 2006 Felipe Alfaro Solana <felipe_alfaro @linuxmail.org>
+ *  Copyright (C) 2007 Julien BLACHE <jb@jblache.org>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License version 2 as
+ *  published by the Free Software Foundation.
+ *
+ *  This driver triggers SMIs which cause the firmware to change the
+ *  backlight brightness. This is icky in many ways, but it's impractical to
+ *  get at the firmware code in order to figure out what it's actually doing.
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/platform_device.h>
+#include <linux/backlight.h>
+#include <linux/err.h>
+#include <linux/dmi.h>
+#include <linux/io.h>
+
+static struct backlight_device *mbp_backlight_device;
+
+static struct dmi_system_id __initdata mbp_device_table[] = {
+	{
+		.ident = "3,1",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "MacBookPro3,1"),
+		},
+	},
+	{
+		.ident = "3,2",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "MacBookPro3,2"),
+		},
+	},
+	{
+		.ident = "4,1",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "MacBookPro4,1"),
+		},
+	},
+	{ }
+};
+
+static int mbp_send_intensity(struct backlight_device *bd)
+{
+	int intensity = bd->props.brightness;
+
+	outb(0x04 | (intensity << 4), 0xb3);
+	outb(0xbf, 0xb2);
+
+	return 0;
+}
+
+static int mbp_get_intensity(struct backlight_device *bd)
+{
+	outb(0x03, 0xb3);
+	outb(0xbf, 0xb2);
+	return inb(0xb3) >> 4;
+}
+
+static struct backlight_ops mbp_ops = {
+	.get_brightness = mbp_get_intensity,
+	.update_status  = mbp_send_intensity,
+};
+
+static int __init mbp_init(void)
+{
+	if (!dmi_check_system(mbp_device_table))
+		return -ENODEV;
+
+	if (!request_region(0xb2, 2, "Macbook Pro backlight"))
+		return -ENXIO;
+
+	mbp_backlight_device = backlight_device_register("mbp_backlight",
+							 NULL, NULL,
+							 &mbp_ops);
+	if (IS_ERR(mbp_backlight_device)) {
+		release_region(0xb2, 2);
+		return PTR_ERR(mbp_backlight_device);
+	}
+
+	mbp_backlight_device->props.max_brightness = 15;
+	mbp_backlight_device->props.brightness =
+		mbp_get_intensity(mbp_backlight_device);
+	backlight_update_status(mbp_backlight_device);
+
+	return 0;
+}
+
+static void __exit mbp_exit(void)
+{
+	backlight_device_unregister(mbp_backlight_device);
+
+	release_region(0xb2, 2);
+}
+
+module_init(mbp_init);
+module_exit(mbp_exit);
+
+MODULE_AUTHOR("Matthew Garrett <mjg@redhat.com>");
+MODULE_DESCRIPTION("Nvidia-based Macbook Pro Backlight Driver");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("svnAppleInc.:pnMacBookPro3,1");
+MODULE_ALIAS("svnAppleInc.:pnMacBookPro3,2");
+MODULE_ALIAS("svnAppleInc.:pnMacBookPro4,1");

From f6ec2d96796d0accda6c325890206f3629130729 Mon Sep 17 00:00:00 2001
From: Sebastian Siewior <bigeasy@linutronix.de>
Date: Wed, 16 Jul 2008 23:05:49 +0100
Subject: [PATCH 0145/1171] backlight: Fix missing kernel doc entry

Signed-off-by: Sebastian Siewior <bigeasy@linutronix.de>
Signed-off-by: Richard Purdie <rpurdie@rpsys.net>
---
 drivers/video/backlight/backlight.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/video/backlight/backlight.c b/drivers/video/backlight/backlight.c
index 39394757679c..fab0bc874b58 100644
--- a/drivers/video/backlight/backlight.c
+++ b/drivers/video/backlight/backlight.c
@@ -191,6 +191,7 @@ static struct device_attribute bl_device_attributes[] = {
  *   backlight_device class.
  * @name: the name of the new object(must be the same as the name of the
  *   respective framebuffer device).
+ * @parent: a pointer to the parent device
  * @devdata: an optional pointer to be stored for private driver use. The
  *   methods may retrieve it by using bl_get_data(bd).
  * @ops: the backlight operations structure.

From 422037bafde8083acc3c539ceba3dfc60a04110c Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 23 Jul 2008 11:16:38 +0200
Subject: [PATCH 0146/1171] sched: fix hrtick & generic-ipi dependency

Andrew Morton reported this s390 allmodconfig build failure:

 kernel/built-in.o: In function `hrtick_start_fair':
 sched.c:(.text+0x69c6): undefined reference to `__smp_call_function_single'

the reason is that s390 is not a generic-ipi SMP platform yet, while
the hrtick code relies on it. Fix the dependency.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/Kconfig.hz | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/Kconfig.hz b/kernel/Kconfig.hz
index 2a202a846757..382dd5a8b2d7 100644
--- a/kernel/Kconfig.hz
+++ b/kernel/Kconfig.hz
@@ -55,4 +55,4 @@ config HZ
 	default 1000 if HZ_1000
 
 config SCHED_HRTICK
-	def_bool HIGH_RES_TIMERS
+	def_bool HIGH_RES_TIMERS && USE_GENERIC_SMP_HELPERS

From 36bd53d07243ae83c1b73bae549086cea2252854 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@linux-foundation.org>
Date: Wed, 23 Jul 2008 00:58:13 -0700
Subject: [PATCH 0147/1171] arch/mips/kernel/stacktrace.c: Heiko can't type

Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/mips/kernel/stacktrace.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/mips/kernel/stacktrace.c b/arch/mips/kernel/stacktrace.c
index 702e2e92a1ca..0632e2a849c0 100644
--- a/arch/mips/kernel/stacktrace.c
+++ b/arch/mips/kernel/stacktrace.c
@@ -7,7 +7,7 @@
  */
 #include <linux/sched.h>
 #include <linux/stacktrace.h>
-#include <linux/module.h
+#include <linux/module.h>
 #include <asm/stacktrace.h>
 
 /*

From 95d04f0735b4fc837bff9aedcc3f3efb20ddc3d1 Mon Sep 17 00:00:00 2001
From: Roland Dreier <rolandd@cisco.com>
Date: Wed, 23 Jul 2008 08:12:26 -0700
Subject: [PATCH 0148/1171] IB/mlx4: Add support for memory management
 extensions and local DMA L_Key

Add support for the following operations to mlx4 when device firmware
supports them:

 - Send with invalidate and local invalidate send queue work requests;
 - Allocate/free fast register MRs;
 - Allocate/free fast register MR page lists;
 - Fast register MR send queue work requests;
 - Local DMA L_Key.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/mlx4/cq.c      | 12 +++++
 drivers/infiniband/hw/mlx4/main.c    | 11 +++++
 drivers/infiniband/hw/mlx4/mlx4_ib.h | 15 ++++++
 drivers/infiniband/hw/mlx4/mr.c      | 70 +++++++++++++++++++++++++++
 drivers/infiniband/hw/mlx4/qp.c      | 72 ++++++++++++++++++++++++++--
 drivers/net/mlx4/fw.c                | 10 ++--
 drivers/net/mlx4/fw.h                |  2 +-
 drivers/net/mlx4/main.c              |  2 +
 drivers/net/mlx4/mr.c                | 23 +++++++--
 include/linux/mlx4/device.h          | 10 ++++
 include/linux/mlx4/qp.h              | 16 +++++--
 11 files changed, 221 insertions(+), 22 deletions(-)

diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c
index 299f20832ab6..0b191a4842ce 100644
--- a/drivers/infiniband/hw/mlx4/cq.c
+++ b/drivers/infiniband/hw/mlx4/cq.c
@@ -637,6 +637,7 @@ repoll:
 		case MLX4_OPCODE_SEND_IMM:
 			wc->wc_flags |= IB_WC_WITH_IMM;
 		case MLX4_OPCODE_SEND:
+		case MLX4_OPCODE_SEND_INVAL:
 			wc->opcode    = IB_WC_SEND;
 			break;
 		case MLX4_OPCODE_RDMA_READ:
@@ -657,6 +658,12 @@ repoll:
 		case MLX4_OPCODE_LSO:
 			wc->opcode    = IB_WC_LSO;
 			break;
+		case MLX4_OPCODE_FMR:
+			wc->opcode    = IB_WC_FAST_REG_MR;
+			break;
+		case MLX4_OPCODE_LOCAL_INVAL:
+			wc->opcode    = IB_WC_LOCAL_INV;
+			break;
 		}
 	} else {
 		wc->byte_len = be32_to_cpu(cqe->byte_cnt);
@@ -667,6 +674,11 @@ repoll:
 			wc->wc_flags	= IB_WC_WITH_IMM;
 			wc->ex.imm_data = cqe->immed_rss_invalid;
 			break;
+		case MLX4_RECV_OPCODE_SEND_INVAL:
+			wc->opcode	= IB_WC_RECV;
+			wc->wc_flags	= IB_WC_WITH_INVALIDATE;
+			wc->ex.invalidate_rkey = be32_to_cpu(cqe->immed_rss_invalid);
+			break;
 		case MLX4_RECV_OPCODE_SEND:
 			wc->opcode   = IB_WC_RECV;
 			wc->wc_flags = 0;
diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index bcf50648fa18..38d6907ab521 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -104,6 +104,12 @@ static int mlx4_ib_query_device(struct ib_device *ibdev,
 		props->device_cap_flags |= IB_DEVICE_UD_IP_CSUM;
 	if (dev->dev->caps.max_gso_sz)
 		props->device_cap_flags |= IB_DEVICE_UD_TSO;
+	if (dev->dev->caps.bmme_flags & MLX4_BMME_FLAG_RESERVED_LKEY)
+		props->device_cap_flags |= IB_DEVICE_LOCAL_DMA_LKEY;
+	if ((dev->dev->caps.bmme_flags & MLX4_BMME_FLAG_LOCAL_INV) &&
+	    (dev->dev->caps.bmme_flags & MLX4_BMME_FLAG_REMOTE_INV) &&
+	    (dev->dev->caps.bmme_flags & MLX4_BMME_FLAG_FAST_REG_WR))
+		props->device_cap_flags |= IB_DEVICE_MEM_MGT_EXTENSIONS;
 
 	props->vendor_id	   = be32_to_cpup((__be32 *) (out_mad->data + 36)) &
 		0xffffff;
@@ -127,6 +133,7 @@ static int mlx4_ib_query_device(struct ib_device *ibdev,
 	props->max_srq		   = dev->dev->caps.num_srqs - dev->dev->caps.reserved_srqs;
 	props->max_srq_wr	   = dev->dev->caps.max_srq_wqes - 1;
 	props->max_srq_sge	   = dev->dev->caps.max_srq_sge;
+	props->max_fast_reg_page_list_len = PAGE_SIZE / sizeof (u64);
 	props->local_ca_ack_delay  = dev->dev->caps.local_ca_ack_delay;
 	props->atomic_cap	   = dev->dev->caps.flags & MLX4_DEV_CAP_FLAG_ATOMIC ?
 		IB_ATOMIC_HCA : IB_ATOMIC_NONE;
@@ -565,6 +572,7 @@ static void *mlx4_ib_add(struct mlx4_dev *dev)
 	strlcpy(ibdev->ib_dev.name, "mlx4_%d", IB_DEVICE_NAME_MAX);
 	ibdev->ib_dev.owner		= THIS_MODULE;
 	ibdev->ib_dev.node_type		= RDMA_NODE_IB_CA;
+	ibdev->ib_dev.local_dma_lkey	= dev->caps.reserved_lkey;
 	ibdev->ib_dev.phys_port_cnt	= dev->caps.num_ports;
 	ibdev->ib_dev.num_comp_vectors	= 1;
 	ibdev->ib_dev.dma_device	= &dev->pdev->dev;
@@ -627,6 +635,9 @@ static void *mlx4_ib_add(struct mlx4_dev *dev)
 	ibdev->ib_dev.get_dma_mr	= mlx4_ib_get_dma_mr;
 	ibdev->ib_dev.reg_user_mr	= mlx4_ib_reg_user_mr;
 	ibdev->ib_dev.dereg_mr		= mlx4_ib_dereg_mr;
+	ibdev->ib_dev.alloc_fast_reg_mr = mlx4_ib_alloc_fast_reg_mr;
+	ibdev->ib_dev.alloc_fast_reg_page_list = mlx4_ib_alloc_fast_reg_page_list;
+	ibdev->ib_dev.free_fast_reg_page_list  = mlx4_ib_free_fast_reg_page_list;
 	ibdev->ib_dev.attach_mcast	= mlx4_ib_mcg_attach;
 	ibdev->ib_dev.detach_mcast	= mlx4_ib_mcg_detach;
 	ibdev->ib_dev.process_mad	= mlx4_ib_process_mad;
diff --git a/drivers/infiniband/hw/mlx4/mlx4_ib.h b/drivers/infiniband/hw/mlx4/mlx4_ib.h
index c4cf5b69eefa..d26a91317d4d 100644
--- a/drivers/infiniband/hw/mlx4/mlx4_ib.h
+++ b/drivers/infiniband/hw/mlx4/mlx4_ib.h
@@ -83,6 +83,11 @@ struct mlx4_ib_mr {
 	struct ib_umem	       *umem;
 };
 
+struct mlx4_ib_fast_reg_page_list {
+	struct ib_fast_reg_page_list	ibfrpl;
+	dma_addr_t			map;
+};
+
 struct mlx4_ib_fmr {
 	struct ib_fmr           ibfmr;
 	struct mlx4_fmr         mfmr;
@@ -199,6 +204,11 @@ static inline struct mlx4_ib_mr *to_mmr(struct ib_mr *ibmr)
 	return container_of(ibmr, struct mlx4_ib_mr, ibmr);
 }
 
+static inline struct mlx4_ib_fast_reg_page_list *to_mfrpl(struct ib_fast_reg_page_list *ibfrpl)
+{
+	return container_of(ibfrpl, struct mlx4_ib_fast_reg_page_list, ibfrpl);
+}
+
 static inline struct mlx4_ib_fmr *to_mfmr(struct ib_fmr *ibfmr)
 {
 	return container_of(ibfmr, struct mlx4_ib_fmr, ibfmr);
@@ -239,6 +249,11 @@ struct ib_mr *mlx4_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 				  u64 virt_addr, int access_flags,
 				  struct ib_udata *udata);
 int mlx4_ib_dereg_mr(struct ib_mr *mr);
+struct ib_mr *mlx4_ib_alloc_fast_reg_mr(struct ib_pd *pd,
+					int max_page_list_len);
+struct ib_fast_reg_page_list *mlx4_ib_alloc_fast_reg_page_list(struct ib_device *ibdev,
+							       int page_list_len);
+void mlx4_ib_free_fast_reg_page_list(struct ib_fast_reg_page_list *page_list);
 
 int mlx4_ib_modify_cq(struct ib_cq *cq, u16 cq_count, u16 cq_period);
 int mlx4_ib_resize_cq(struct ib_cq *ibcq, int entries, struct ib_udata *udata);
diff --git a/drivers/infiniband/hw/mlx4/mr.c b/drivers/infiniband/hw/mlx4/mr.c
index 68e92485fc76..db2086faa4ed 100644
--- a/drivers/infiniband/hw/mlx4/mr.c
+++ b/drivers/infiniband/hw/mlx4/mr.c
@@ -183,6 +183,76 @@ int mlx4_ib_dereg_mr(struct ib_mr *ibmr)
 	return 0;
 }
 
+struct ib_mr *mlx4_ib_alloc_fast_reg_mr(struct ib_pd *pd,
+					int max_page_list_len)
+{
+	struct mlx4_ib_dev *dev = to_mdev(pd->device);
+	struct mlx4_ib_mr *mr;
+	int err;
+
+	mr = kmalloc(sizeof *mr, GFP_KERNEL);
+	if (!mr)
+		return ERR_PTR(-ENOMEM);
+
+	err = mlx4_mr_alloc(dev->dev, to_mpd(pd)->pdn, 0, 0, 0,
+			    max_page_list_len, 0, &mr->mmr);
+	if (err)
+		goto err_free;
+
+	err = mlx4_mr_enable(dev->dev, &mr->mmr);
+	if (err)
+		goto err_mr;
+
+	return &mr->ibmr;
+
+err_mr:
+	mlx4_mr_free(dev->dev, &mr->mmr);
+
+err_free:
+	kfree(mr);
+	return ERR_PTR(err);
+}
+
+struct ib_fast_reg_page_list *mlx4_ib_alloc_fast_reg_page_list(struct ib_device *ibdev,
+							       int page_list_len)
+{
+	struct mlx4_ib_dev *dev = to_mdev(ibdev);
+	struct mlx4_ib_fast_reg_page_list *mfrpl;
+	int size = page_list_len * sizeof (u64);
+
+	if (size > PAGE_SIZE)
+		return ERR_PTR(-EINVAL);
+
+	mfrpl = kmalloc(sizeof *mfrpl, GFP_KERNEL);
+	if (!mfrpl)
+		return ERR_PTR(-ENOMEM);
+
+	mfrpl->ibfrpl.page_list = dma_alloc_coherent(&dev->dev->pdev->dev,
+						     size, &mfrpl->map,
+						     GFP_KERNEL);
+	if (!mfrpl->ibfrpl.page_list)
+		goto err_free;
+
+	WARN_ON(mfrpl->map & 0x3f);
+
+	return &mfrpl->ibfrpl;
+
+err_free:
+	kfree(mfrpl);
+	return ERR_PTR(-ENOMEM);
+}
+
+void mlx4_ib_free_fast_reg_page_list(struct ib_fast_reg_page_list *page_list)
+{
+	struct mlx4_ib_dev *dev = to_mdev(page_list->device);
+	struct mlx4_ib_fast_reg_page_list *mfrpl = to_mfrpl(page_list);
+	int size = page_list->max_page_list_len * sizeof (u64);
+
+	dma_free_coherent(&dev->dev->pdev->dev, size, page_list->page_list,
+			  mfrpl->map);
+	kfree(mfrpl);
+}
+
 struct ib_fmr *mlx4_ib_fmr_alloc(struct ib_pd *pd, int acc,
 				 struct ib_fmr_attr *fmr_attr)
 {
diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
index bda0859a5ac5..02a99bc4442e 100644
--- a/drivers/infiniband/hw/mlx4/qp.c
+++ b/drivers/infiniband/hw/mlx4/qp.c
@@ -78,6 +78,9 @@ static const __be32 mlx4_ib_opcode[] = {
 	[IB_WR_RDMA_READ]		= __constant_cpu_to_be32(MLX4_OPCODE_RDMA_READ),
 	[IB_WR_ATOMIC_CMP_AND_SWP]	= __constant_cpu_to_be32(MLX4_OPCODE_ATOMIC_CS),
 	[IB_WR_ATOMIC_FETCH_AND_ADD]	= __constant_cpu_to_be32(MLX4_OPCODE_ATOMIC_FA),
+	[IB_WR_SEND_WITH_INV]		= __constant_cpu_to_be32(MLX4_OPCODE_SEND_INVAL),
+	[IB_WR_LOCAL_INV]		= __constant_cpu_to_be32(MLX4_OPCODE_LOCAL_INVAL),
+	[IB_WR_FAST_REG_MR]		= __constant_cpu_to_be32(MLX4_OPCODE_FMR),
 };
 
 static struct mlx4_ib_sqp *to_msqp(struct mlx4_ib_qp *mqp)
@@ -976,6 +979,10 @@ static int __mlx4_ib_modify_qp(struct ib_qp *ibqp,
 	context->pd	    = cpu_to_be32(to_mpd(ibqp->pd)->pdn);
 	context->params1    = cpu_to_be32(MLX4_IB_ACK_REQ_FREQ << 28);
 
+	/* Set "fast registration enabled" for all kernel QPs */
+	if (!qp->ibqp.uobject)
+		context->params1 |= cpu_to_be32(1 << 11);
+
 	if (attr_mask & IB_QP_RNR_RETRY) {
 		context->params1 |= cpu_to_be32(attr->rnr_retry << 13);
 		optpar |= MLX4_QP_OPTPAR_RNR_RETRY;
@@ -1322,6 +1329,38 @@ static int mlx4_wq_overflow(struct mlx4_ib_wq *wq, int nreq, struct ib_cq *ib_cq
 	return cur + nreq >= wq->max_post;
 }
 
+static __be32 convert_access(int acc)
+{
+	return (acc & IB_ACCESS_REMOTE_ATOMIC ? cpu_to_be32(MLX4_WQE_FMR_PERM_ATOMIC)       : 0) |
+	       (acc & IB_ACCESS_REMOTE_WRITE  ? cpu_to_be32(MLX4_WQE_FMR_PERM_REMOTE_WRITE) : 0) |
+	       (acc & IB_ACCESS_REMOTE_READ   ? cpu_to_be32(MLX4_WQE_FMR_PERM_REMOTE_READ)  : 0) |
+	       (acc & IB_ACCESS_LOCAL_WRITE   ? cpu_to_be32(MLX4_WQE_FMR_PERM_LOCAL_WRITE)  : 0) |
+		cpu_to_be32(MLX4_WQE_FMR_PERM_LOCAL_READ);
+}
+
+static void set_fmr_seg(struct mlx4_wqe_fmr_seg *fseg, struct ib_send_wr *wr)
+{
+	struct mlx4_ib_fast_reg_page_list *mfrpl = to_mfrpl(wr->wr.fast_reg.page_list);
+
+	fseg->flags		= convert_access(wr->wr.fast_reg.access_flags);
+	fseg->mem_key		= cpu_to_be32(wr->wr.fast_reg.rkey);
+	fseg->buf_list		= cpu_to_be64(mfrpl->map);
+	fseg->start_addr	= cpu_to_be64(wr->wr.fast_reg.iova_start);
+	fseg->reg_len		= cpu_to_be64(wr->wr.fast_reg.length);
+	fseg->offset		= 0; /* XXX -- is this just for ZBVA? */
+	fseg->page_size		= cpu_to_be32(wr->wr.fast_reg.page_shift);
+	fseg->reserved[0]	= 0;
+	fseg->reserved[1]	= 0;
+}
+
+static void set_local_inv_seg(struct mlx4_wqe_local_inval_seg *iseg, u32 rkey)
+{
+	iseg->flags	= 0;
+	iseg->mem_key	= cpu_to_be32(rkey);
+	iseg->guest_id	= 0;
+	iseg->pa	= 0;
+}
+
 static __always_inline void set_raddr_seg(struct mlx4_wqe_raddr_seg *rseg,
 					  u64 remote_addr, u32 rkey)
 {
@@ -1423,6 +1462,21 @@ static int build_lso_seg(struct mlx4_wqe_lso_seg *wqe, struct ib_send_wr *wr,
 	return 0;
 }
 
+static __be32 send_ieth(struct ib_send_wr *wr)
+{
+	switch (wr->opcode) {
+	case IB_WR_SEND_WITH_IMM:
+	case IB_WR_RDMA_WRITE_WITH_IMM:
+		return wr->ex.imm_data;
+
+	case IB_WR_SEND_WITH_INV:
+		return cpu_to_be32(wr->ex.invalidate_rkey);
+
+	default:
+		return 0;
+	}
+}
+
 int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 		      struct ib_send_wr **bad_wr)
 {
@@ -1469,11 +1523,7 @@ int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 				     MLX4_WQE_CTRL_TCP_UDP_CSUM) : 0) |
 			qp->sq_signal_bits;
 
-		if (wr->opcode == IB_WR_SEND_WITH_IMM ||
-		    wr->opcode == IB_WR_RDMA_WRITE_WITH_IMM)
-			ctrl->imm = wr->ex.imm_data;
-		else
-			ctrl->imm = 0;
+		ctrl->imm = send_ieth(wr);
 
 		wqe += sizeof *ctrl;
 		size = sizeof *ctrl / 16;
@@ -1505,6 +1555,18 @@ int mlx4_ib_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
 				size += sizeof (struct mlx4_wqe_raddr_seg) / 16;
 				break;
 
+			case IB_WR_LOCAL_INV:
+				set_local_inv_seg(wqe, wr->ex.invalidate_rkey);
+				wqe  += sizeof (struct mlx4_wqe_local_inval_seg);
+				size += sizeof (struct mlx4_wqe_local_inval_seg) / 16;
+				break;
+
+			case IB_WR_FAST_REG_MR:
+				set_fmr_seg(wqe, wr);
+				wqe  += sizeof (struct mlx4_wqe_fmr_seg);
+				size += sizeof (struct mlx4_wqe_fmr_seg) / 16;
+				break;
+
 			default:
 				/* No extra segments required for sends */
 				break;
diff --git a/drivers/net/mlx4/fw.c b/drivers/net/mlx4/fw.c
index 0851ebdddfd4..57278224ba1e 100644
--- a/drivers/net/mlx4/fw.c
+++ b/drivers/net/mlx4/fw.c
@@ -202,7 +202,7 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 #define QUERY_DEV_CAP_C_MPT_ENTRY_SZ_OFFSET	0x8e
 #define QUERY_DEV_CAP_MTT_ENTRY_SZ_OFFSET	0x90
 #define QUERY_DEV_CAP_D_MPT_ENTRY_SZ_OFFSET	0x92
-#define QUERY_DEV_CAP_BMME_FLAGS_OFFSET		0x97
+#define QUERY_DEV_CAP_BMME_FLAGS_OFFSET		0x94
 #define QUERY_DEV_CAP_RSVD_LKEY_OFFSET		0x98
 #define QUERY_DEV_CAP_MAX_ICM_SZ_OFFSET		0xa0
 
@@ -377,12 +377,8 @@ int mlx4_QUERY_DEV_CAP(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 		}
 	}
 
-	if (dev_cap->bmme_flags & 1)
-		mlx4_dbg(dev, "Base MM extensions: yes "
-			 "(flags %d, rsvd L_Key %08x)\n",
-			 dev_cap->bmme_flags, dev_cap->reserved_lkey);
-	else
-		mlx4_dbg(dev, "Base MM extensions: no\n");
+	mlx4_dbg(dev, "Base MM extensions: flags %08x, rsvd L_Key %08x\n",
+		 dev_cap->bmme_flags, dev_cap->reserved_lkey);
 
 	/*
 	 * Each UAR has 4 EQ doorbells; so if a UAR is reserved, then
diff --git a/drivers/net/mlx4/fw.h b/drivers/net/mlx4/fw.h
index a0e046c149b7..fbf0e22be122 100644
--- a/drivers/net/mlx4/fw.h
+++ b/drivers/net/mlx4/fw.h
@@ -98,7 +98,7 @@ struct mlx4_dev_cap {
 	int cmpt_entry_sz;
 	int mtt_entry_sz;
 	int resize_srq;
-	u8  bmme_flags;
+	u32 bmme_flags;
 	u32 reserved_lkey;
 	u64 max_icm_sz;
 	int max_gso_sz;
diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c
index d3736013fe9b..8e1d24cda1b0 100644
--- a/drivers/net/mlx4/main.c
+++ b/drivers/net/mlx4/main.c
@@ -158,6 +158,8 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	dev->caps.max_msg_sz         = dev_cap->max_msg_sz;
 	dev->caps.page_size_cap	     = ~(u32) (dev_cap->min_page_sz - 1);
 	dev->caps.flags		     = dev_cap->flags;
+	dev->caps.bmme_flags	     = dev_cap->bmme_flags;
+	dev->caps.reserved_lkey	     = dev_cap->reserved_lkey;
 	dev->caps.stat_rate_support  = dev_cap->stat_rate_support;
 	dev->caps.max_gso_sz	     = dev_cap->max_gso_sz;
 
diff --git a/drivers/net/mlx4/mr.c b/drivers/net/mlx4/mr.c
index b3ea93b98689..a3c04c5f12c2 100644
--- a/drivers/net/mlx4/mr.c
+++ b/drivers/net/mlx4/mr.c
@@ -47,7 +47,7 @@ struct mlx4_mpt_entry {
 	__be32 flags;
 	__be32 qpn;
 	__be32 key;
-	__be32 pd;
+	__be32 pd_flags;
 	__be64 start;
 	__be64 length;
 	__be32 lkey;
@@ -61,11 +61,15 @@ struct mlx4_mpt_entry {
 } __attribute__((packed));
 
 #define MLX4_MPT_FLAG_SW_OWNS	    (0xfUL << 28)
+#define MLX4_MPT_FLAG_FREE	    (0x3UL << 28)
 #define MLX4_MPT_FLAG_MIO	    (1 << 17)
 #define MLX4_MPT_FLAG_BIND_ENABLE   (1 << 15)
 #define MLX4_MPT_FLAG_PHYSICAL	    (1 <<  9)
 #define MLX4_MPT_FLAG_REGION	    (1 <<  8)
 
+#define MLX4_MPT_PD_FLAG_FAST_REG   (1 << 26)
+#define MLX4_MPT_PD_FLAG_EN_INV	    (3 << 24)
+
 #define MLX4_MTT_FLAG_PRESENT		1
 
 #define MLX4_MPT_STATUS_SW		0xF0
@@ -324,21 +328,30 @@ int mlx4_mr_enable(struct mlx4_dev *dev, struct mlx4_mr *mr)
 
 	memset(mpt_entry, 0, sizeof *mpt_entry);
 
-	mpt_entry->flags = cpu_to_be32(MLX4_MPT_FLAG_SW_OWNS	 |
-				       MLX4_MPT_FLAG_MIO	 |
+	mpt_entry->flags = cpu_to_be32(MLX4_MPT_FLAG_MIO	 |
 				       MLX4_MPT_FLAG_REGION	 |
 				       mr->access);
 
 	mpt_entry->key	       = cpu_to_be32(key_to_hw_index(mr->key));
-	mpt_entry->pd	       = cpu_to_be32(mr->pd);
+	mpt_entry->pd_flags    = cpu_to_be32(mr->pd | MLX4_MPT_PD_FLAG_EN_INV);
 	mpt_entry->start       = cpu_to_be64(mr->iova);
 	mpt_entry->length      = cpu_to_be64(mr->size);
 	mpt_entry->entity_size = cpu_to_be32(mr->mtt.page_shift);
+
 	if (mr->mtt.order < 0) {
 		mpt_entry->flags |= cpu_to_be32(MLX4_MPT_FLAG_PHYSICAL);
 		mpt_entry->mtt_seg = 0;
-	} else
+	} else {
 		mpt_entry->mtt_seg = cpu_to_be64(mlx4_mtt_addr(dev, &mr->mtt));
+	}
+
+	if (mr->mtt.order >= 0 && mr->mtt.page_shift == 0) {
+		/* fast register MR in free state */
+		mpt_entry->flags    |= cpu_to_be32(MLX4_MPT_FLAG_FREE);
+		mpt_entry->pd_flags |= cpu_to_be32(MLX4_MPT_PD_FLAG_FAST_REG);
+	} else {
+		mpt_entry->flags    |= cpu_to_be32(MLX4_MPT_FLAG_SW_OWNS);
+	}
 
 	err = mlx4_SW2HW_MPT(dev, mailbox,
 			     key_to_hw_index(mr->key) & (dev->caps.num_mpts - 1));
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 81b3dd5206e0..655ea0d1ee14 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -68,6 +68,14 @@ enum {
 	MLX4_DEV_CAP_FLAG_UD_MCAST	= 1 << 21
 };
 
+enum {
+	MLX4_BMME_FLAG_LOCAL_INV	= 1 <<  6,
+	MLX4_BMME_FLAG_REMOTE_INV	= 1 <<  7,
+	MLX4_BMME_FLAG_TYPE_2_WIN	= 1 <<  9,
+	MLX4_BMME_FLAG_RESERVED_LKEY	= 1 << 10,
+	MLX4_BMME_FLAG_FAST_REG_WR	= 1 << 11,
+};
+
 enum mlx4_event {
 	MLX4_EVENT_TYPE_COMP		   = 0x00,
 	MLX4_EVENT_TYPE_PATH_MIG	   = 0x01,
@@ -184,6 +192,8 @@ struct mlx4_caps {
 	u32			max_msg_sz;
 	u32			page_size_cap;
 	u32			flags;
+	u32			bmme_flags;
+	u32			reserved_lkey;
 	u16			stat_rate_support;
 	u8			port_width_cap[MLX4_MAX_PORTS + 1];
 	int			max_gso_sz;
diff --git a/include/linux/mlx4/qp.h b/include/linux/mlx4/qp.h
index f02e9ed36cfa..e27082cd650e 100644
--- a/include/linux/mlx4/qp.h
+++ b/include/linux/mlx4/qp.h
@@ -233,6 +233,14 @@ struct mlx4_wqe_bind_seg {
 	__be64			length;
 };
 
+enum {
+	MLX4_WQE_FMR_PERM_LOCAL_READ	= 1 << 27,
+	MLX4_WQE_FMR_PERM_LOCAL_WRITE	= 1 << 28,
+	MLX4_WQE_FMR_PERM_REMOTE_READ	= 1 << 29,
+	MLX4_WQE_FMR_PERM_REMOTE_WRITE	= 1 << 30,
+	MLX4_WQE_FMR_PERM_ATOMIC	= 1 << 31
+};
+
 struct mlx4_wqe_fmr_seg {
 	__be32			flags;
 	__be32			mem_key;
@@ -255,11 +263,11 @@ struct mlx4_wqe_fmr_ext_seg {
 };
 
 struct mlx4_wqe_local_inval_seg {
-	u8			flags;
-	u8			reserved1[3];
+	__be32			flags;
+	u32			reserved1;
 	__be32			mem_key;
-	u8			reserved2[3];
-	u8			guest_id;
+	u32			reserved2[2];
+	__be32			guest_id;
 	__be64			pa;
 };
 

From 76442640829163d0cdb67c2bf0cb4b81a0fe537b Mon Sep 17 00:00:00 2001
From: Roland Dreier <rolandd@cisco.com>
Date: Wed, 23 Jul 2008 08:12:47 -0700
Subject: [PATCH 0149/1171] mlx4_core: Improve error message when not enough
 UAR pages are available

If an mlx4 device with default FW (which gives a UAR BAR size of 8 MB)
is used in a system with 64 KB pages, then there are only 8192/64==128
UAR pages available.  However, the first 128 UAR pages are reserved
for use with event queue doorbells, so no UAR pages are available to
do anything else with, which means that the driver cannot work.

The current driver fails with a fairly cryptic "Failed to allocate
driver access region, aborting" message in this situation.  Fix the
driver to detect the problem earlier and print out a clearer
description of the problem and a suggestion of how to fix it (use a
new firmware image).

Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/net/mlx4/pd.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/net/mlx4/pd.c b/drivers/net/mlx4/pd.c
index 3a93c5f0f7ab..aa616892d09c 100644
--- a/drivers/net/mlx4/pd.c
+++ b/drivers/net/mlx4/pd.c
@@ -91,6 +91,13 @@ EXPORT_SYMBOL_GPL(mlx4_uar_free);
 
 int mlx4_init_uar_table(struct mlx4_dev *dev)
 {
+	if (dev->caps.num_uars <= 128) {
+		mlx4_err(dev, "Only %d UAR pages (need more than 128)\n",
+			 dev->caps.num_uars);
+		mlx4_err(dev, "Increase firmware log2_uar_bar_megabytes?\n");
+		return -ENODEV;
+	}
+
 	return mlx4_bitmap_init(&mlx4_priv(dev)->uar_table.bitmap,
 				dev->caps.num_uars, dev->caps.num_uars - 1,
 				max(128, dev->caps.reserved_uars));

From 09e50d55a9490e9b7a6fdfbf8fc078924b25ecb5 Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@redhat.com>
Date: Wed, 23 Jul 2008 10:11:19 -0400
Subject: [PATCH 0150/1171] lockdep: annotate cifs in-kernel sockets

Put CIFS sockets in their own class to avoid some lockdep warnings. CIFS
sockets are not exposed to user-space, and so are not subject to the
same deadlock scenarios.

A similar change was made a couple of years ago for RPC sockets in commit
ed07536ed6731775219c1df7fa26a7588753e693.

This patch should prevent lockdep false-positives like this one:

=======================================================
[ INFO: possible circular locking dependency detected ]
2.6.18-98.el5.jtltest.38.bz456320.1debug #1
-------------------------------------------------------
test5/2483 is trying to acquire lock:
 (sk_lock-AF_INET){--..}, at: [<ffffffff800270d2>] tcp_sendmsg+0x1c/0xb2f

but task is already holding lock:
 (&inode->i_alloc_sem){--..}, at: [<ffffffff8002e454>] notify_change+0xf5/0x2e0

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:

-> #3 (&inode->i_alloc_sem){--..}:
       [<ffffffff800a817c>] __lock_acquire+0x9a9/0xadf
       [<ffffffff800a8a72>] lock_acquire+0x55/0x70
       [<ffffffff8002e454>] notify_change+0xf5/0x2e0
       [<ffffffff800a4e36>] down_write+0x3c/0x68
       [<ffffffff8002e454>] notify_change+0xf5/0x2e0
       [<ffffffff800e358d>] do_truncate+0x50/0x6b
       [<ffffffff8005197c>] get_write_access+0x40/0x46
       [<ffffffff80012cf1>] may_open+0x1d3/0x22e
       [<ffffffff8001bc81>] open_namei+0x2c6/0x6dd
       [<ffffffff800289c6>] do_filp_open+0x1c/0x38
       [<ffffffff800683ef>] _spin_unlock+0x17/0x20
       [<ffffffff800167a7>] get_unused_fd+0xf9/0x107
       [<ffffffff8001a704>] do_sys_open+0x44/0xbe
       [<ffffffff80060116>] system_call+0x7e/0x83
       [<ffffffffffffffff>] 0xffffffffffffffff

-> #2 (&sysfs_inode_imutex_key){--..}:
       [<ffffffff800a817c>] __lock_acquire+0x9a9/0xadf
       [<ffffffff8010f6df>] create_dir+0x26/0x1d7
       [<ffffffff800a8a72>] lock_acquire+0x55/0x70
       [<ffffffff8010f6df>] create_dir+0x26/0x1d7
       [<ffffffff800671c0>] mutex_lock_nested+0x104/0x29c
       [<ffffffff800a819d>] __lock_acquire+0x9ca/0xadf
       [<ffffffff8010f6df>] create_dir+0x26/0x1d7
       [<ffffffff8010fc67>] sysfs_create_dir+0x58/0x76
       [<ffffffff8015144c>] kobject_add+0xdb/0x198
       [<ffffffff801be765>] class_device_add+0xb2/0x465
       [<ffffffff8005a6ff>] kobject_get+0x12/0x17
       [<ffffffff80225265>] register_netdevice+0x270/0x33e
       [<ffffffff8022538c>] register_netdev+0x59/0x67
       [<ffffffff80464d40>] net_olddevs_init+0xb/0xac
       [<ffffffff80448a79>] init+0x1f9/0x2fc
       [<ffffffff80068885>] _spin_unlock_irq+0x24/0x27
       [<ffffffff80067f86>] trace_hardirqs_on_thunk+0x35/0x37
       [<ffffffff80061079>] child_rip+0xa/0x11
       [<ffffffff80068885>] _spin_unlock_irq+0x24/0x27
       [<ffffffff800606a8>] restore_args+0x0/0x30
       [<ffffffff80179a59>] acpi_ds_init_one_object+0x0/0x80
       [<ffffffff80448880>] init+0x0/0x2fc
       [<ffffffff8006106f>] child_rip+0x0/0x11
       [<ffffffffffffffff>] 0xffffffffffffffff

-> #1 (rtnl_mutex){--..}:
       [<ffffffff800a817c>] __lock_acquire+0x9a9/0xadf
       [<ffffffff8025acf8>] ip_mc_leave_group+0x23/0xb7
       [<ffffffff800a8a72>] lock_acquire+0x55/0x70
       [<ffffffff8025acf8>] ip_mc_leave_group+0x23/0xb7
       [<ffffffff800671c0>] mutex_lock_nested+0x104/0x29c
       [<ffffffff8025acf8>] ip_mc_leave_group+0x23/0xb7
       [<ffffffff802451b0>] do_ip_setsockopt+0x6d1/0x9bf
       [<ffffffff800a575e>] lock_release_holdtime+0x27/0x48
       [<ffffffff800a575e>] lock_release_holdtime+0x27/0x48
       [<ffffffff8006a85e>] do_page_fault+0x503/0x835
       [<ffffffff8012cbf6>] socket_has_perm+0x5b/0x68
       [<ffffffff80245556>] ip_setsockopt+0x22/0x78
       [<ffffffff8021c973>] sys_setsockopt+0x91/0xb7
       [<ffffffff800602a6>] tracesys+0xd5/0xdf
       [<ffffffffffffffff>] 0xffffffffffffffff

-> #0 (sk_lock-AF_INET){--..}:
       [<ffffffff800a5037>] print_stack_trace+0x59/0x68
       [<ffffffff800a8092>] __lock_acquire+0x8bf/0xadf
       [<ffffffff800a8a72>] lock_acquire+0x55/0x70
       [<ffffffff800270d2>] tcp_sendmsg+0x1c/0xb2f
       [<ffffffff80035466>] lock_sock+0xd4/0xe4
       [<ffffffff80096e91>] _local_bh_enable+0xcb/0xe0
       [<ffffffff800606a8>] restore_args+0x0/0x30
       [<ffffffff800270d2>] tcp_sendmsg+0x1c/0xb2f
       [<ffffffff80057540>] sock_sendmsg+0xf3/0x110
       [<ffffffff800a2bb6>] autoremove_wake_function+0x0/0x2e
       [<ffffffff800a10e4>] kernel_text_address+0x1a/0x26
       [<ffffffff8006f4e2>] dump_trace+0x211/0x23a
       [<ffffffff800a6d3d>] find_usage_backwards+0x5f/0x88
       [<ffffffff8840221a>] MD5Final+0xaf/0xc2 [cifs]
       [<ffffffff884032ec>] cifs_calculate_signature+0x55/0x69 [cifs]
       [<ffffffff8021d891>] kernel_sendmsg+0x35/0x47
       [<ffffffff883ff38e>] smb_send+0xa3/0x151 [cifs]
       [<ffffffff883ff5de>] SendReceive+0x1a2/0x448 [cifs]
       [<ffffffff800a812f>] __lock_acquire+0x95c/0xadf
       [<ffffffff883e758a>] CIFSSMBSetEOF+0x20d/0x25b [cifs]
       [<ffffffff883fa430>] cifs_set_file_size+0x110/0x3b7 [cifs]
       [<ffffffff883faa89>] cifs_setattr+0x3b2/0x6f6 [cifs]
       [<ffffffff8002e454>] notify_change+0xf5/0x2e0
       [<ffffffff8002e4a4>] notify_change+0x145/0x2e0
       [<ffffffff800e358d>] do_truncate+0x50/0x6b
       [<ffffffff8005197c>] get_write_access+0x40/0x46
       [<ffffffff80012cf1>] may_open+0x1d3/0x22e
       [<ffffffff8001bc81>] open_namei+0x2c6/0x6dd
       [<ffffffff800289c6>] do_filp_open+0x1c/0x38
       [<ffffffff800683ef>] _spin_unlock+0x17/0x20
       [<ffffffff800167a7>] get_unused_fd+0xf9/0x107
       [<ffffffff8001a704>] do_sys_open+0x44/0xbe
       [<ffffffff800602a6>] tracesys+0xd5/0xdf
       [<ffffffffffffffff>] 0xffffffffffffffff

other info that might help us debug this:

2 locks held by test5/2483:
 #0:  (&inode->i_mutex){--..}, at: [<ffffffff800e3582>] do_truncate+0x45/0x6b
 #1:  (&inode->i_alloc_sem){--..}, at: [<ffffffff8002e454>] notify_change+0xf5/0x2e0

stack backtrace:

Call Trace:
 [<ffffffff800a6a7b>] print_circular_bug_tail+0x65/0x6e
 [<ffffffff800a5037>] print_stack_trace+0x59/0x68
 [<ffffffff800a8092>] __lock_acquire+0x8bf/0xadf
 [<ffffffff800a8a72>] lock_acquire+0x55/0x70
 [<ffffffff800270d2>] tcp_sendmsg+0x1c/0xb2f
 [<ffffffff80035466>] lock_sock+0xd4/0xe4
 [<ffffffff80096e91>] _local_bh_enable+0xcb/0xe0
 [<ffffffff800606a8>] restore_args+0x0/0x30
 [<ffffffff800270d2>] tcp_sendmsg+0x1c/0xb2f
 [<ffffffff80057540>] sock_sendmsg+0xf3/0x110
 [<ffffffff800a2bb6>] autoremove_wake_function+0x0/0x2e
 [<ffffffff800a10e4>] kernel_text_address+0x1a/0x26
 [<ffffffff8006f4e2>] dump_trace+0x211/0x23a
 [<ffffffff800a6d3d>] find_usage_backwards+0x5f/0x88
 [<ffffffff8840221a>] :cifs:MD5Final+0xaf/0xc2
 [<ffffffff884032ec>] :cifs:cifs_calculate_signature+0x55/0x69
 [<ffffffff8021d891>] kernel_sendmsg+0x35/0x47
 [<ffffffff883ff38e>] :cifs:smb_send+0xa3/0x151
 [<ffffffff883ff5de>] :cifs:SendReceive+0x1a2/0x448
 [<ffffffff800a812f>] __lock_acquire+0x95c/0xadf
 [<ffffffff883e758a>] :cifs:CIFSSMBSetEOF+0x20d/0x25b
 [<ffffffff883fa430>] :cifs:cifs_set_file_size+0x110/0x3b7
 [<ffffffff883faa89>] :cifs:cifs_setattr+0x3b2/0x6f6
 [<ffffffff8002e454>] notify_change+0xf5/0x2e0
 [<ffffffff8002e4a4>] notify_change+0x145/0x2e0
 [<ffffffff800e358d>] do_truncate+0x50/0x6b
 [<ffffffff8005197c>] get_write_access+0x40/0x46
 [<ffffffff80012cf1>] may_open+0x1d3/0x22e
 [<ffffffff8001bc81>] open_namei+0x2c6/0x6dd
 [<ffffffff800289c6>] do_filp_open+0x1c/0x38
 [<ffffffff800683ef>] _spin_unlock+0x17/0x20
 [<ffffffff800167a7>] get_unused_fd+0xf9/0x107
 [<ffffffff8001a704>] do_sys_open+0x44/0xbe
 [<ffffffff800602a6>] tracesys+0xd5/0xdf

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/connect.c | 35 +++++++++++++++++++++++++++++++++++
 1 file changed, 35 insertions(+)

diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index e8fa46c7cff2..644462729387 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -1461,6 +1461,39 @@ get_dfs_path(int xid, struct cifsSesInfo *pSesInfo, const char *old_path,
 	return rc;
 }
 
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+static struct lock_class_key cifs_key[2];
+static struct lock_class_key cifs_slock_key[2];
+
+static inline void
+cifs_reclassify_socket4(struct socket *sock)
+{
+	struct sock *sk = sock->sk;
+	BUG_ON(sock_owned_by_user(sk));
+	sock_lock_init_class_and_name(sk, "slock-AF_INET-CIFS",
+		&cifs_slock_key[0], "sk_lock-AF_INET-CIFS", &cifs_key[0]);
+}
+
+static inline void
+cifs_reclassify_socket6(struct socket *sock)
+{
+	struct sock *sk = sock->sk;
+	BUG_ON(sock_owned_by_user(sk));
+	sock_lock_init_class_and_name(sk, "slock-AF_INET6-CIFS",
+		&cifs_slock_key[1], "sk_lock-AF_INET6-CIFS", &cifs_key[1]);
+}
+#else
+static inline void
+cifs_reclassify_socket4(struct socket *sock)
+{
+}
+
+static inline void
+cifs_reclassify_socket6(struct socket *sock)
+{
+}
+#endif
+
 /* See RFC1001 section 14 on representation of Netbios names */
 static void rfc1002mangle(char *target, char *source, unsigned int length)
 {
@@ -1495,6 +1528,7 @@ ipv4_connect(struct sockaddr_in *psin_server, struct socket **csocket,
 		/* BB other socket options to set KEEPALIVE, NODELAY? */
 			cFYI(1, ("Socket created"));
 			(*csocket)->sk->sk_allocation = GFP_NOFS;
+			cifs_reclassify_socket4(*csocket);
 		}
 	}
 
@@ -1627,6 +1661,7 @@ ipv6_connect(struct sockaddr_in6 *psin_server, struct socket **csocket)
 		/* BB other socket options to set KEEPALIVE, NODELAY? */
 			 cFYI(1, ("ipv6 Socket created"));
 			(*csocket)->sk->sk_allocation = GFP_NOFS;
+			cifs_reclassify_socket6(*csocket);
 		}
 	}
 

From 5b3ab1dbd401b36ba2f9bfee2d2dae252fd62cd8 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Wed, 23 Jul 2008 14:01:29 -0700
Subject: [PATCH 0151/1171] netdev: Remove warning from __netif_schedule().

It isn't helping anything and we aren't going to be able to change all
the drivers that do queue wakeups in strange situations.

Just letting a noop_qdisc get scheduled will work because when
qdisc_run() executes via net_tx_work() it will simply find no packets
pending when it makes the ->dequeue() call in qdisc_restart.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/dev.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index 6bf217da9d8f..ccf97f9f37eb 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1341,9 +1341,6 @@ static void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev)
 
 void __netif_schedule(struct Qdisc *q)
 {
-	if (WARN_ON_ONCE(q == &noop_qdisc))
-		return;
-
 	if (!test_and_set_bit(__QDISC_STATE_SCHED, &q->state)) {
 		struct softnet_data *sd;
 		unsigned long flags;

From b4942af65028c5eb516fdd9053020ccb2ee186ce Mon Sep 17 00:00:00 2001
From: Oliver Hartkopp <oliver@hartkopp.net>
Date: Wed, 23 Jul 2008 14:06:04 -0700
Subject: [PATCH 0152/1171] net: Update entry in af_family_clock_key_strings

In the merge phase of the CAN subsystem the
af_family_clock_key_strings[] have been added to sock.c in commit
443aef0eddfa44c158d1b94ebb431a70638fcab4
(lockdep: fixup sk_callback_lock annotation). This trivial patch adds
the missing name for address family 29 (AF_CAN).

Signed-off-by: Oliver Hartkopp <oliver@hartkopp.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/core/sock.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/core/sock.c b/net/core/sock.c
index 10a64d57078c..91f8bbc93526 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -180,7 +180,7 @@ static const char *af_family_clock_key_strings[AF_MAX+1] = {
   "clock-AF_ASH"   , "clock-AF_ECONET"   , "clock-AF_ATMSVC"   ,
   "clock-21"       , "clock-AF_SNA"      , "clock-AF_IRDA"     ,
   "clock-AF_PPPOX" , "clock-AF_WANPIPE"  , "clock-AF_LLC"      ,
-  "clock-27"       , "clock-28"          , "clock-29"          ,
+  "clock-27"       , "clock-28"          , "clock-AF_CAN"      ,
   "clock-AF_TIPC"  , "clock-AF_BLUETOOTH", "clock-AF_IUCV"     ,
   "clock-AF_RXRPC" , "clock-AF_MAX"
 };

From 1fa6d8181b7bb0361512170c30e436dcc95591ee Mon Sep 17 00:00:00 2001
From: Roland Dreier <rolandd@cisco.com>
Date: Wed, 23 Jul 2008 14:20:12 -0700
Subject: [PATCH 0153/1171] MAINTAINERS: Remove Glenn Streiff from NetEffect
 entry

Glenn is no longer at NetEffect.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 MAINTAINERS | 2 --
 1 file changed, 2 deletions(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 11944b44c2ff..2a73da0cd075 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2915,8 +2915,6 @@ P:	Faisal Latif
 M:	flatif@neteffect.com
 P:	Chien Tung
 M:	ctung@neteffect.com
-P:	Glenn Streiff
-M:	gstreiff@neteffect.com
 L:	general@lists.openfabrics.org
 W:	http://www.neteffect.com
 S:	Supported

From 8efdbde647f542ce0d303273df7ad4157caa03d0 Mon Sep 17 00:00:00 2001
From: Jeff Layton <jlayton@redhat.com>
Date: Wed, 23 Jul 2008 21:28:12 +0000
Subject: [PATCH 0154/1171] [CIFS] break ATTR_SIZE changes out into their own
 function

Move the code that handles ATTR_SIZE changes to its own function. This
makes for a smaller function and reduces the level of indentation.

Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/inode.c | 151 +++++++++++++++++++++++++-----------------------
 1 file changed, 78 insertions(+), 73 deletions(-)

diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c
index 2e904bd111c8..46e54d39461d 100644
--- a/fs/cifs/inode.c
+++ b/fs/cifs/inode.c
@@ -1413,6 +1413,82 @@ out_busy:
 	return -ETXTBSY;
 }
 
+static int
+cifs_set_file_size(struct inode *inode, struct iattr *attrs,
+		   int xid, char *full_path)
+{
+	int rc;
+	struct cifsFileInfo *open_file;
+	struct cifsInodeInfo *cifsInode = CIFS_I(inode);
+	struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb);
+	struct cifsTconInfo *pTcon = cifs_sb->tcon;
+
+	/*
+	 * To avoid spurious oplock breaks from server, in the case of
+	 * inodes that we already have open, avoid doing path based
+	 * setting of file size if we can do it by handle.
+	 * This keeps our caching token (oplock) and avoids timeouts
+	 * when the local oplock break takes longer to flush
+	 * writebehind data than the SMB timeout for the SetPathInfo
+	 * request would allow
+	 */
+	open_file = find_writable_file(cifsInode);
+	if (open_file) {
+		__u16 nfid = open_file->netfid;
+		__u32 npid = open_file->pid;
+		rc = CIFSSMBSetFileSize(xid, pTcon, attrs->ia_size, nfid,
+					npid, false);
+		atomic_dec(&open_file->wrtPending);
+		cFYI(1, ("SetFSize for attrs rc = %d", rc));
+		if ((rc == -EINVAL) || (rc == -EOPNOTSUPP)) {
+			unsigned int bytes_written;
+			rc = CIFSSMBWrite(xid, pTcon, nfid, 0, attrs->ia_size,
+					  &bytes_written, NULL, NULL, 1);
+			cFYI(1, ("Wrt seteof rc %d", rc));
+		}
+	} else
+		rc = -EINVAL;
+
+	if (rc != 0) {
+		/* Set file size by pathname rather than by handle
+		   either because no valid, writeable file handle for
+		   it was found or because there was an error setting
+		   it by handle */
+		rc = CIFSSMBSetEOF(xid, pTcon, full_path, attrs->ia_size,
+				   false, cifs_sb->local_nls,
+				   cifs_sb->mnt_cifs_flags &
+					CIFS_MOUNT_MAP_SPECIAL_CHR);
+		cFYI(1, ("SetEOF by path (setattrs) rc = %d", rc));
+		if ((rc == -EINVAL) || (rc == -EOPNOTSUPP)) {
+			__u16 netfid;
+			int oplock = 0;
+
+			rc = SMBLegacyOpen(xid, pTcon, full_path,
+				FILE_OPEN, GENERIC_WRITE,
+				CREATE_NOT_DIR, &netfid, &oplock, NULL,
+				cifs_sb->local_nls,
+				cifs_sb->mnt_cifs_flags &
+					CIFS_MOUNT_MAP_SPECIAL_CHR);
+			if (rc == 0) {
+				unsigned int bytes_written;
+				rc = CIFSSMBWrite(xid, pTcon, netfid, 0,
+						  attrs->ia_size,
+						  &bytes_written, NULL,
+						  NULL, 1);
+				cFYI(1, ("wrt seteof rc %d", rc));
+				CIFSSMBClose(xid, pTcon, netfid);
+			}
+		}
+	}
+
+	if (rc == 0) {
+		rc = cifs_vmtruncate(inode, attrs->ia_size);
+		cifs_truncate_page(inode->i_mapping, inode->i_size);
+	}
+
+	return rc;
+}
+
 int cifs_setattr(struct dentry *direntry, struct iattr *attrs)
 {
 	int xid;
@@ -1420,7 +1496,6 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs)
 	struct cifsTconInfo *pTcon;
 	char *full_path = NULL;
 	int rc = -EACCES;
-	struct cifsFileInfo *open_file = NULL;
 	FILE_BASIC_INFO time_buf;
 	bool set_time = false;
 	bool set_dosattr = false;
@@ -1472,78 +1547,8 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs)
 	}
 
 	if (attrs->ia_valid & ATTR_SIZE) {
-		/* To avoid spurious oplock breaks from server, in the case of
-		   inodes that we already have open, avoid doing path based
-		   setting of file size if we can do it by handle.
-		   This keeps our caching token (oplock) and avoids timeouts
-		   when the local oplock break takes longer to flush
-		   writebehind data than the SMB timeout for the SetPathInfo
-		   request would allow */
-
-		open_file = find_writable_file(cifsInode);
-		if (open_file) {
-			__u16 nfid = open_file->netfid;
-			__u32 npid = open_file->pid;
-			rc = CIFSSMBSetFileSize(xid, pTcon, attrs->ia_size,
-						nfid, npid, false);
-			atomic_dec(&open_file->wrtPending);
-			cFYI(1, ("SetFSize for attrs rc = %d", rc));
-			if ((rc == -EINVAL) || (rc == -EOPNOTSUPP)) {
-				unsigned int bytes_written;
-				rc = CIFSSMBWrite(xid, pTcon,
-						  nfid, 0, attrs->ia_size,
-						  &bytes_written, NULL, NULL,
-						  1 /* 45 seconds */);
-				cFYI(1, ("Wrt seteof rc %d", rc));
-			}
-		} else
-			rc = -EINVAL;
-
-		if (rc != 0) {
-			/* Set file size by pathname rather than by handle
-			   either because no valid, writeable file handle for
-			   it was found or because there was an error setting
-			   it by handle */
-			rc = CIFSSMBSetEOF(xid, pTcon, full_path,
-					   attrs->ia_size, false,
-					   cifs_sb->local_nls,
-					   cifs_sb->mnt_cifs_flags &
-						CIFS_MOUNT_MAP_SPECIAL_CHR);
-			cFYI(1, ("SetEOF by path (setattrs) rc = %d", rc));
-			if ((rc == -EINVAL) || (rc == -EOPNOTSUPP)) {
-				__u16 netfid;
-				int oplock = 0;
-
-				rc = SMBLegacyOpen(xid, pTcon, full_path,
-					FILE_OPEN, GENERIC_WRITE,
-					CREATE_NOT_DIR, &netfid, &oplock,
-					NULL, cifs_sb->local_nls,
-					cifs_sb->mnt_cifs_flags &
-						CIFS_MOUNT_MAP_SPECIAL_CHR);
-				if (rc == 0) {
-					unsigned int bytes_written;
-					rc = CIFSSMBWrite(xid, pTcon,
-							netfid, 0,
-							attrs->ia_size,
-							&bytes_written, NULL,
-							NULL, 1 /* 45 sec */);
-					cFYI(1, ("wrt seteof rc %d", rc));
-					CIFSSMBClose(xid, pTcon, netfid);
-				}
-
-			}
-		}
-
-		/* Server is ok setting allocation size implicitly - no need
-		   to call:
-		CIFSSMBSetEOF(xid, pTcon, full_path, attrs->ia_size, true,
-			 cifs_sb->local_nls);
-		   */
-
-		if (rc == 0) {
-			rc = cifs_vmtruncate(inode, attrs->ia_size);
-			cifs_truncate_page(inode->i_mapping, inode->i_size);
-		} else
+		rc = cifs_set_file_size(inode, attrs, xid, full_path);
+		if (rc != 0)
 			goto cifs_setattr_exit;
 	}
 

From e8ebe3b893792887317bc24cc4608753f81b81d3 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Wed, 23 Jul 2008 15:30:52 -0700
Subject: [PATCH 0155/1171] e1000e: fix e1000_netpoll(), remove extraneous
 e1000_clean_tx_irq() call

Evgeniy Polyakov noticed that drivers/net/e1000e/netdev.c:e1000_netpoll()
was calling e1000_clean_tx_irq() without taking the TX lock.

David Miller suggested to remove the call altogether: since in this
callpah there's periodic calls to ->poll() anyway which will do
e1000_clean_tx_irq() and will garbage-collect any finished TX ring
descriptors.

This fix solved the e1000e+netconsole crashes i've been seeing:

=============================================================================
BUG skbuff_head_cache: Poison overwritten
-----------------------------------------------------------------------------

INFO: 0xf658ae9c-0xf658ae9c. First byte 0x6a instead of 0x6b
INFO: Allocated in __alloc_skb+0x2c/0x110 age=0 cpu=0 pid=5098
INFO: Freed in __kfree_skb+0x31/0x80 age=0 cpu=1 pid=4440
INFO: Slab 0xc16cc140 objects=16 used=1 fp=0xf658ae00 flags=0x400000c3
INFO: Object 0xf658ae00 @offset=3584 fp=0xf658af00

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/e1000e/netdev.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/net/e1000e/netdev.c b/drivers/net/e1000e/netdev.c
index 869544b8c05c..9c0f56b3c518 100644
--- a/drivers/net/e1000e/netdev.c
+++ b/drivers/net/e1000e/netdev.c
@@ -4067,8 +4067,6 @@ static void e1000_netpoll(struct net_device *netdev)
 	disable_irq(adapter->pdev->irq);
 	e1000_intr(adapter->pdev->irq, netdev);
 
-	e1000_clean_tx_irq(adapter);
-
 	enable_irq(adapter->pdev->irq);
 }
 #endif

From 7ae93f51d7fa8b9130d47e0b7d17979a165c5bc3 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Wed, 23 Jul 2008 16:21:07 -0700
Subject: [PATCH 0156/1171] sparc64: Fix cpufreq notifier registry.

Based upon a report by Daniel Smolik.

We do it too early, which triggers a BUG in
cpufreq_register_notifier().

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/sparc64/kernel/time.c | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)

diff --git a/arch/sparc64/kernel/time.c b/arch/sparc64/kernel/time.c
index bedc4c159b1c..a0c6a97eec6e 100644
--- a/arch/sparc64/kernel/time.c
+++ b/arch/sparc64/kernel/time.c
@@ -884,6 +884,16 @@ static struct notifier_block sparc64_cpufreq_notifier_block = {
 	.notifier_call	= sparc64_cpufreq_notifier
 };
 
+static int __init register_sparc64_cpufreq_notifier(void)
+{
+
+	cpufreq_register_notifier(&sparc64_cpufreq_notifier_block,
+				  CPUFREQ_TRANSITION_NOTIFIER);
+	return 0;
+}
+
+core_initcall(register_sparc64_cpufreq_notifier);
+
 #endif /* CONFIG_CPU_FREQ */
 
 static int sparc64_next_event(unsigned long delta,
@@ -1050,11 +1060,6 @@ void __init time_init(void)
 	       sparc64_clockevent.mult, sparc64_clockevent.shift);
 
 	setup_sparc64_timer();
-
-#ifdef CONFIG_CPU_FREQ
-	cpufreq_register_notifier(&sparc64_cpufreq_notifier_block,
-				  CPUFREQ_TRANSITION_NOTIFIER);
-#endif
 }
 
 unsigned long long sched_clock(void)

From 1685a03e98b7e9b83e0aa692c1cc470b3aa37597 Mon Sep 17 00:00:00 2001
From: Jan Nikitenko <jan.nikitenko@gmail.com>
Date: Thu, 24 Jul 2008 01:27:07 +0200
Subject: [PATCH 0157/1171] mmc_spi: put signals to low power off fix

The original intention was to write a zero byte to mmc to force spi
signals to low when doing power off. Somehow the spi_w8r8 call got there
so a read followed the write of single zero byte. This patch changes
that to simple write of zero byte without the following read.
This way the power off is more reliable and completely sufficient.

Signed-off-by: Jan Nikitenko <jan.nikitenko@gmail.com>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
---
 drivers/mmc/host/mmc_spi.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/mmc/host/mmc_spi.c b/drivers/mmc/host/mmc_spi.c
index 41cc63360e43..7503b81374e0 100644
--- a/drivers/mmc/host/mmc_spi.c
+++ b/drivers/mmc/host/mmc_spi.c
@@ -1076,6 +1076,7 @@ static void mmc_spi_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
 		 */
 		if (canpower && ios->power_mode == MMC_POWER_OFF) {
 			int mres;
+			u8 nullbyte = 0;
 
 			host->spi->mode &= ~(SPI_CPOL|SPI_CPHA);
 			mres = spi_setup(host->spi);
@@ -1083,7 +1084,7 @@ static void mmc_spi_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
 				dev_dbg(&host->spi->dev,
 					"switch to SPI mode 0 failed\n");
 
-			if (spi_w8r8(host->spi, 0x00) < 0)
+			if (spi_write(host->spi, &nullbyte, 1) < 0)
 				dev_dbg(&host->spi->dev,
 					"put spi signals to low failed\n");
 

From 4b53fb67e385b856a991d402096379dab462170a Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Wed, 23 Jul 2008 16:38:45 -0700
Subject: [PATCH 0158/1171] tcp: Clear probes_out more aggressively in
 tcp_ack().

This is based upon an excellent bug report from Eric Dumazet.

tcp_ack() should clear ->icsk_probes_out even if there are packets
outstanding.  Otherwise if we get a sequence of ACKs while we do have
packets outstanding over and over again, we'll never clear the
probes_out value and eventually think the connection is too sick and
we'll reset it.

This appears to be some "optimization" added to tcp_ack() in the 2.4.x
timeframe.  In 2.2.x, probes_out is pretty much always cleared by
tcp_ack().

Here is Eric's original report:

----------------------------------------
Apparently, we can in some situations reset TCP connections in a couple of seconds when some frames are lost.

In order to reproduce the problem, please try the following program on linux-2.6.25.*

Setup some iptables rules to allow two frames per second sent on loopback interface to tcp destination port 12000

iptables -N SLOWLO
iptables -A SLOWLO -m hashlimit --hashlimit 2 --hashlimit-burst 1 --hashlimit-mode dstip --hashlimit-name slow2 -j ACCEPT
iptables -A SLOWLO -j DROP

iptables -A OUTPUT -o lo -p tcp --dport 12000 -j SLOWLO

Then run the attached program and see the output :

# ./loop
State      Recv-Q Send-Q                                  Local Address:Port                                    Peer Address:Port
ESTAB      0      40                                          127.0.0.1:54455                                      127.0.0.1:12000  timer:(persist,200ms,1)
State      Recv-Q Send-Q                                  Local Address:Port                                    Peer Address:Port
ESTAB      0      40                                          127.0.0.1:54455                                      127.0.0.1:12000  timer:(persist,200ms,3)
State      Recv-Q Send-Q                                  Local Address:Port                                    Peer Address:Port
ESTAB      0      40                                          127.0.0.1:54455                                      127.0.0.1:12000  timer:(persist,200ms,5)
State      Recv-Q Send-Q                                  Local Address:Port                                    Peer Address:Port
ESTAB      0      40                                          127.0.0.1:54455                                      127.0.0.1:12000  timer:(persist,200ms,7)
State      Recv-Q Send-Q                                  Local Address:Port                                    Peer Address:Port
ESTAB      0      40                                          127.0.0.1:54455                                      127.0.0.1:12000  timer:(persist,200ms,9)
State      Recv-Q Send-Q                                  Local Address:Port                                    Peer Address:Port
ESTAB      0      40                                          127.0.0.1:54455                                      127.0.0.1:12000  timer:(persist,200ms,11)
State      Recv-Q Send-Q                                  Local Address:Port                                    Peer Address:Port
ESTAB      0      40                                          127.0.0.1:54455                                      127.0.0.1:12000  timer:(persist,201ms,13)
State      Recv-Q Send-Q                                  Local Address:Port                                    Peer Address:Port
ESTAB      0      40                                          127.0.0.1:54455                                      127.0.0.1:12000  timer:(persist,188ms,15)
write(): Connection timed out
wrote 890 bytes but was interrupted after 9 seconds
ESTAB      0      0                 127.0.0.1:12000            127.0.0.1:54455
Exiting read() because no data available (4000 ms timeout).
read 860 bytes

While this tcp session makes progress (sending frames with 50 bytes of payload, every 500ms), linux tcp stack decides to reset it, when tcp_retries 2 is reached (default value : 15)

tcpdump :

15:30:28.856695 IP 127.0.0.1.56554 > 127.0.0.1.12000: S 33788768:33788768(0) win 32792 <mss 16396,nop,nop,sackOK,nop,wscale 7>
15:30:28.856711 IP 127.0.0.1.12000 > 127.0.0.1.56554: S 33899253:33899253(0) ack 33788769 win 32792 <mss 16396,nop,nop,sackOK,nop,wscale 7>
15:30:29.356947 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 1:61(60) ack 1 win 257
15:30:29.356966 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 61 win 257
15:30:29.866415 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 61:111(50) ack 1 win 257
15:30:29.866427 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 111 win 257
15:30:30.366516 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 111:161(50) ack 1 win 257
15:30:30.366527 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 161 win 257
15:30:30.876196 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 161:211(50) ack 1 win 257
15:30:30.876207 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 211 win 257
15:30:31.376282 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 211:261(50) ack 1 win 257
15:30:31.376290 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 261 win 257
15:30:31.885619 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 261:311(50) ack 1 win 257
15:30:31.885631 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 311 win 257
15:30:32.385705 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 311:361(50) ack 1 win 257
15:30:32.385715 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 361 win 257
15:30:32.895249 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 361:411(50) ack 1 win 257
15:30:32.895266 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 411 win 257
15:30:33.395341 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 411:461(50) ack 1 win 257
15:30:33.395351 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 461 win 257
15:30:33.918085 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 461:511(50) ack 1 win 257
15:30:33.918096 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 511 win 257
15:30:34.418163 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 511:561(50) ack 1 win 257
15:30:34.418172 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 561 win 257
15:30:34.927685 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 561:611(50) ack 1 win 257
15:30:34.927698 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 611 win 257
15:30:35.427757 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 611:661(50) ack 1 win 257
15:30:35.427766 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 661 win 257
15:30:35.937359 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 661:711(50) ack 1 win 257
15:30:35.937376 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 711 win 257
15:30:36.437451 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 711:761(50) ack 1 win 257
15:30:36.437464 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 761 win 257
15:30:36.947022 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 761:811(50) ack 1 win 257
15:30:36.947039 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 811 win 257
15:30:37.447135 IP 127.0.0.1.56554 > 127.0.0.1.12000: P 811:861(50) ack 1 win 257
15:30:37.447203 IP 127.0.0.1.12000 > 127.0.0.1.56554: . ack 861 win 257
15:30:41.448171 IP 127.0.0.1.12000 > 127.0.0.1.56554: F 1:1(0) ack 861 win 257
15:30:41.448189 IP 127.0.0.1.56554 > 127.0.0.1.12000: R 33789629:33789629(0) win 0

Source of program :

/*
 * small producer/consumer program.
 * setup a listener on 127.0.0.1:12000
 * Forks a child
 *   child connect to 127.0.0.1, and sends 10 bytes on this tcp socket every 100 ms
 * Father accepts connection, and read all data
 */
#include <sys/types.h>
#include <sys/socket.h>
#include <netinet/in.h>
#include <unistd.h>
#include <stdio.h>
#include <time.h>
#include <sys/poll.h>

int port = 12000;
char buffer[4096];
int main(int argc, char *argv[])
{
        int lfd = socket(AF_INET, SOCK_STREAM, 0);
        struct sockaddr_in socket_address;
        time_t t0, t1;
        int on = 1, sfd, res;
        unsigned long total = 0;
        socklen_t alen = sizeof(socket_address);
        pid_t pid;

        time(&t0);
        socket_address.sin_family = AF_INET;
        socket_address.sin_port = htons(port);
        socket_address.sin_addr.s_addr = htonl(INADDR_LOOPBACK);

        if (lfd == -1) {
                perror("socket()");
                return 1;
        }
        setsockopt(lfd, SOL_SOCKET, SO_REUSEADDR, &on, sizeof(int));
        if (bind(lfd, (struct sockaddr *)&socket_address, sizeof(socket_address)) == -1) {
                perror("bind");
                close(lfd);
                return 1;
        }
        if (listen(lfd, 1) == -1) {
                perror("listen()");
                close(lfd);
                return 1;
        }
        pid = fork();
        if (pid == 0) {
                int i, cfd = socket(AF_INET, SOCK_STREAM, 0);
                close(lfd);
                if (connect(cfd, (struct sockaddr *)&socket_address, sizeof(socket_address)) == -1) {
                        perror("connect()");
                        return 1;
                        }
                for (i = 0 ; ;) {
                        res = write(cfd, "blablabla\n", 10);
                        if (res > 0) total += res;
                        else if (res == -1) {
                                perror("write()");
                                break;
                        } else break;
                        usleep(100000);
                        if (++i == 10) {
                                system("ss -on dst 127.0.0.1:12000");
                                i = 0;
                        }
                }
                time(&t1);
                fprintf(stderr, "wrote %lu bytes but was interrupted after %g seconds\n", total, difftime(t1, t0));
                system("ss -on | grep 127.0.0.1:12000");
                close(cfd);
                return 0;
        }
        sfd = accept(lfd, (struct sockaddr *)&socket_address, &alen);
        if (sfd == -1) {
                perror("accept");
                return 1;
        }
        close(lfd);
        while (1) {
                struct pollfd pfd[1];
                pfd[0].fd = sfd;
                pfd[0].events = POLLIN;
                if (poll(pfd, 1, 4000) == 0) {
                        fprintf(stderr, "Exiting read() because no data available (4000 ms timeout).\n");
                        break;
                }
                res = read(sfd, buffer, sizeof(buffer));
                if (res > 0) total += res;
                else if (res == 0) break;
                else perror("read()");
        }
        fprintf(stderr, "read %lu bytes\n", total);
        close(sfd);
        return 0;
}
----------------------------------------

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/tcp_input.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 1f5e6049883e..75efd244f2af 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -3292,6 +3292,7 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag)
 	 * log. Something worked...
 	 */
 	sk->sk_err_soft = 0;
+	icsk->icsk_probes_out = 0;
 	tp->rcv_tstamp = tcp_time_stamp;
 	prior_packets = tp->packets_out;
 	if (!prior_packets)
@@ -3324,8 +3325,6 @@ static int tcp_ack(struct sock *sk, struct sk_buff *skb, int flag)
 	return 1;
 
 no_queue:
-	icsk->icsk_probes_out = 0;
-
 	/* If this ack opens up a zero window, clear backoff.  It was
 	 * being used to time the probes, and is probably far higher than
 	 * it needs to be for normal retransmission.

From 70eed75d76635ba7350651b9bd96529a306ec67a Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Wed, 23 Jul 2008 16:42:42 -0700
Subject: [PATCH 0159/1171] netfilter: make security table depend on
 NETFILTER_ADVANCED

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/netfilter/Kconfig | 2 +-
 net/ipv6/netfilter/Kconfig | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig
index f23e60c93ef9..90eb7cb47e77 100644
--- a/net/ipv4/netfilter/Kconfig
+++ b/net/ipv4/netfilter/Kconfig
@@ -369,7 +369,7 @@ config IP_NF_SECURITY
 	tristate "Security table"
 	depends on IP_NF_IPTABLES
 	depends on SECURITY
-	default m if NETFILTER_ADVANCED=n
+	depends on NETFILTER_ADVANCED
 	help
 	  This option adds a `security' table to iptables, for use
 	  with Mandatory Access Control (MAC) policy.
diff --git a/net/ipv6/netfilter/Kconfig b/net/ipv6/netfilter/Kconfig
index 689dec899c57..0cfcce7b18d8 100644
--- a/net/ipv6/netfilter/Kconfig
+++ b/net/ipv6/netfilter/Kconfig
@@ -213,7 +213,7 @@ config IP6_NF_SECURITY
        tristate "Security table"
        depends on IP6_NF_IPTABLES
        depends on SECURITY
-       default m if NETFILTER_ADVANCED=n
+       depends on NETFILTER_ADVANCED
        help
          This option adds a `security' table to iptables, for use
          with Mandatory Access Control (MAC) policy.

From 0855b543222e79cbbd9d66dd56cb54740e7d524f Mon Sep 17 00:00:00 2001
From: Andre Detsch <adetsch@br.ibm.com>
Date: Thu, 24 Jul 2008 10:57:26 +1000
Subject: [PATCH 0160/1171] powerpc/spufs: fix aff_mutex and
 cbe_spu_info[n].list_mutex deadlock

Currenlt,, it is possible to lock aff_mutex and
cbe_spu_info[n].list_mutex in different orders, allowing a deadlock to
occur. With this change, aff_mutex is not taken within a list_mutex
critical section anymore.

Signed-off-by: Andre Detsch <adetsch@br.ibm.com>
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
---
 arch/powerpc/platforms/cell/spufs/sched.c | 18 ++++++------------
 1 file changed, 6 insertions(+), 12 deletions(-)

diff --git a/arch/powerpc/platforms/cell/spufs/sched.c b/arch/powerpc/platforms/cell/spufs/sched.c
index 34654743363d..f293963cd855 100644
--- a/arch/powerpc/platforms/cell/spufs/sched.c
+++ b/arch/powerpc/platforms/cell/spufs/sched.c
@@ -389,6 +389,9 @@ static int has_affinity(struct spu_context *ctx)
 	if (list_empty(&ctx->aff_list))
 		return 0;
 
+	if (atomic_read(&ctx->gang->aff_sched_count) == 0)
+		ctx->gang->aff_ref_spu = NULL;
+
 	if (!gang->aff_ref_spu) {
 		if (!(gang->aff_flags & AFF_MERGED))
 			aff_merge_remaining_ctxs(gang);
@@ -416,14 +419,8 @@ static void spu_unbind_context(struct spu *spu, struct spu_context *ctx)
  	if (spu->ctx->flags & SPU_CREATE_NOSCHED)
 		atomic_dec(&cbe_spu_info[spu->node].reserved_spus);
 
-	if (ctx->gang){
-		mutex_lock(&ctx->gang->aff_mutex);
-		if (has_affinity(ctx)) {
-			if (atomic_dec_and_test(&ctx->gang->aff_sched_count))
-				ctx->gang->aff_ref_spu = NULL;
-		}
-		mutex_unlock(&ctx->gang->aff_mutex);
-	}
+	if (ctx->gang)
+		atomic_dec_if_positive(&ctx->gang->aff_sched_count);
 
 	spu_switch_notify(spu, NULL);
 	spu_unmap_mappings(ctx);
@@ -562,10 +559,7 @@ static struct spu *spu_get_idle(struct spu_context *ctx)
 				goto found;
 			mutex_unlock(&cbe_spu_info[node].list_mutex);
 
-			mutex_lock(&ctx->gang->aff_mutex);
-			if (atomic_dec_and_test(&ctx->gang->aff_sched_count))
-				ctx->gang->aff_ref_spu = NULL;
-			mutex_unlock(&ctx->gang->aff_mutex);
+			atomic_dec(&ctx->gang->aff_sched_count);
 			goto not_found;
 		}
 		mutex_unlock(&ctx->gang->aff_mutex);

From ad1ede127760d6ca4903f44dfe1a8a38b3bfb36c Mon Sep 17 00:00:00 2001
From: Andre Detsch <adetsch@br.ibm.com>
Date: Thu, 24 Jul 2008 11:01:54 +1000
Subject: [PATCH 0161/1171] powerpc/spufs: better placement of spu affinity
 reference context

This patch adjusts the placement of a reference context from
a spu affinity chain. The reference context can now be placed
only on nodes that have enough spus not intended to be used by
another gang (already running on the node).

Signed-off-by: Andre Detsch <adetsch@br.ibm.com>
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
---
 arch/powerpc/platforms/cell/spufs/sched.c | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/arch/powerpc/platforms/cell/spufs/sched.c b/arch/powerpc/platforms/cell/spufs/sched.c
index f293963cd855..2deeeba7eccf 100644
--- a/arch/powerpc/platforms/cell/spufs/sched.c
+++ b/arch/powerpc/platforms/cell/spufs/sched.c
@@ -312,10 +312,27 @@ static struct spu *aff_ref_location(struct spu_context *ctx, int mem_aff,
 	 */
 	node = cpu_to_node(raw_smp_processor_id());
 	for (n = 0; n < MAX_NUMNODES; n++, node++) {
+		int available_spus;
+
 		node = (node < MAX_NUMNODES) ? node : 0;
 		if (!node_allowed(ctx, node))
 			continue;
+
+		available_spus = 0;
 		mutex_lock(&cbe_spu_info[node].list_mutex);
+		list_for_each_entry(spu, &cbe_spu_info[node].spus, cbe_list) {
+			if (spu->ctx && spu->ctx->gang
+					&& spu->ctx->aff_offset == 0)
+				available_spus -=
+					(spu->ctx->gang->contexts - 1);
+			else
+				available_spus++;
+		}
+		if (available_spus < ctx->gang->contexts) {
+			mutex_unlock(&cbe_spu_info[node].list_mutex);
+			continue;
+		}
+
 		list_for_each_entry(spu, &cbe_spu_info[node].spus, cbe_list) {
 			if ((!mem_aff || spu->has_mem_affinity) &&
 							sched_spu(spu)) {

From 5ca33c6ac30596fc1403a092f46ea48c406734e4 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 23 Jul 2008 17:45:58 -0700
Subject: [PATCH 0162/1171] cifs: assorted endian annotations

fs/cifs/cifssmb.c:3917:13: warning: incorrect type in assignment (different base types)
fs/cifs/cifssmb.c:3917:13:    expected bool [unsigned] [usertype] is_unicode
fs/cifs/cifssmb.c:3917:13:    got restricted __le16

The comment explains why __force is used here.
fs/cifs/connect.c:458:16: warning: cast to restricted __be32
fs/cifs/connect.c:458:16: warning: cast to restricted __be32
fs/cifs/connect.c:458:16: warning: cast to restricted __be32
fs/cifs/connect.c:458:16: warning: cast to restricted __be32
fs/cifs/connect.c:458:16: warning: cast to restricted __be32
fs/cifs/connect.c:458:16: warning: cast to restricted __be32

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/cifssmb.c | 5 ++++-
 fs/cifs/connect.c | 2 +-
 2 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c
index 4511b708f0f3..7e175e1399d1 100644
--- a/fs/cifs/cifssmb.c
+++ b/fs/cifs/cifssmb.c
@@ -3914,7 +3914,10 @@ parse_DFS_referrals(TRANSACTION2_GET_DFS_REFER_RSP *pSMBr,
 	bool is_unicode;
 	struct dfs_referral_level_3 *ref;
 
-	is_unicode = pSMBr->hdr.Flags2 & SMBFLG2_UNICODE;
+	if (pSMBr->hdr.Flags2 & SMBFLG2_UNICODE)
+		is_unicode = true;
+	else
+		is_unicode = false;
 	*num_of_nodes = le16_to_cpu(pSMBr->NumberOfReferrals);
 
 	if (*num_of_nodes < 1) {
diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index 644462729387..b51d5777cde6 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -455,7 +455,7 @@ incomplete_rcv:
 		/* Note that FC 1001 length is big endian on the wire,
 		but we convert it here so it is always manipulated
 		as host byte order */
-		pdu_length = ntohl(smb_buffer->smb_buf_length);
+		pdu_length = be32_to_cpu((__force __be32)smb_buffer->smb_buf_length);
 		smb_buffer->smb_buf_length = pdu_length;
 
 		cFYI(1, ("rfc1002 length 0x%x", pdu_length+4));

From f984c7b98253c541ed6e702ee521f6a84c8113d4 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Thu, 24 Jul 2008 02:34:24 +0000
Subject: [PATCH 0163/1171] Signed-off-by: Alexey Dobriyan
 <adobriyan@gmail.com> Cc: Steven French <sfrench@us.ibm.com> Signed-off-by:
 Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Steve French
 <sfrench@us.ibm.com>

---
 fs/cifs/cifs_debug.c | 658 +++++++++++++++++--------------------------
 1 file changed, 252 insertions(+), 406 deletions(-)

diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c
index cc950f69e51e..73925a77375e 100644
--- a/fs/cifs/cifs_debug.c
+++ b/fs/cifs/cifs_debug.c
@@ -107,9 +107,7 @@ void cifs_dump_mids(struct TCP_Server_Info *server)
 #endif /* CONFIG_CIFS_DEBUG2 */
 
 #ifdef CONFIG_PROC_FS
-static int
-cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset,
-		     int count, int *eof, void *data)
+static int cifs_debug_data_proc_show(struct seq_file *m, void *v)
 {
 	struct list_head *tmp;
 	struct list_head *tmp1;
@@ -117,23 +115,13 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset,
 	struct cifsSesInfo *ses;
 	struct cifsTconInfo *tcon;
 	int i;
-	int length = 0;
-	char *original_buf = buf;
 
-	*beginBuffer = buf + offset;
-
-	length =
-	    sprintf(buf,
+	seq_puts(m,
 		    "Display Internal CIFS Data Structures for Debugging\n"
 		    "---------------------------------------------------\n");
-	buf += length;
-	length = sprintf(buf, "CIFS Version %s\n", CIFS_VERSION);
-	buf += length;
-	length = sprintf(buf,
-		"Active VFS Requests: %d\n", GlobalTotalActiveXid);
-	buf += length;
-	length = sprintf(buf, "Servers:");
-	buf += length;
+	seq_printf(m, "CIFS Version %s\n", CIFS_VERSION);
+	seq_printf(m, "Active VFS Requests: %d\n", GlobalTotalActiveXid);
+	seq_printf(m, "Servers:");
 
 	i = 0;
 	read_lock(&GlobalSMBSeslock);
@@ -142,11 +130,10 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset,
 		ses = list_entry(tmp, struct cifsSesInfo, cifsSessionList);
 		if ((ses->serverDomain == NULL) || (ses->serverOS == NULL) ||
 		   (ses->serverNOS == NULL)) {
-			buf += sprintf(buf, "\nentry for %s not fully "
+			seq_printf(m, "\nentry for %s not fully "
 					"displayed\n\t", ses->serverName);
 		} else {
-			length =
-			    sprintf(buf,
+			seq_printf(m,
 				    "\n%d) Name: %s  Domain: %s Mounts: %d OS:"
 				    " %s  \n\tNOS: %s\tCapability: 0x%x\n\tSMB"
 				    " session status: %d\t",
@@ -154,10 +141,9 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset,
 				atomic_read(&ses->inUse),
 				ses->serverOS, ses->serverNOS,
 				ses->capabilities, ses->status);
-			buf += length;
 		}
 		if (ses->server) {
-			buf += sprintf(buf, "TCP status: %d\n\tLocal Users To "
+			seq_printf(m, "TCP status: %d\n\tLocal Users To "
 				    "Server: %d SecMode: 0x%x Req On Wire: %d",
 				ses->server->tcpStatus,
 				atomic_read(&ses->server->socketUseCount),
@@ -165,13 +151,12 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset,
 				atomic_read(&ses->server->inFlight));
 
 #ifdef CONFIG_CIFS_STATS2
-			buf += sprintf(buf, " In Send: %d In MaxReq Wait: %d",
+			seq_printf(m, " In Send: %d In MaxReq Wait: %d",
 				atomic_read(&ses->server->inSend),
 				atomic_read(&ses->server->num_waiters));
 #endif
 
-			length = sprintf(buf, "\nMIDs:\n");
-			buf += length;
+			seq_puts(m, "\nMIDs:\n");
 
 			spin_lock(&GlobalMid_Lock);
 			list_for_each(tmp1, &ses->server->pending_mid_q) {
@@ -179,7 +164,7 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset,
 					mid_q_entry,
 					qhead);
 				if (mid_entry) {
-					length = sprintf(buf,
+					seq_printf(m,
 							"State: %d com: %d pid:"
 							" %d tsk: %p mid %d\n",
 							mid_entry->midState,
@@ -187,7 +172,6 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset,
 							mid_entry->pid,
 							mid_entry->tsk,
 							mid_entry->mid);
-					buf += length;
 				}
 			}
 			spin_unlock(&GlobalMid_Lock);
@@ -195,11 +179,9 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset,
 
 	}
 	read_unlock(&GlobalSMBSeslock);
-	sprintf(buf, "\n");
-	buf++;
+	seq_putc(m, '\n');
 
-	length = sprintf(buf, "Shares:");
-	buf += length;
+	seq_puts(m, "Shares:");
 
 	i = 0;
 	read_lock(&GlobalSMBSeslock);
@@ -208,62 +190,52 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset,
 		i++;
 		tcon = list_entry(tmp, struct cifsTconInfo, cifsConnectionList);
 		dev_type = le32_to_cpu(tcon->fsDevInfo.DeviceType);
-		length = sprintf(buf, "\n%d) %s Uses: %d ", i,
+		seq_printf(m, "\n%d) %s Uses: %d ", i,
 				 tcon->treeName, atomic_read(&tcon->useCount));
-		buf += length;
 		if (tcon->nativeFileSystem) {
-			length = sprintf(buf, "Type: %s ",
+			seq_printf(m, "Type: %s ",
 					 tcon->nativeFileSystem);
-			buf += length;
 		}
-		length = sprintf(buf, "DevInfo: 0x%x Attributes: 0x%x"
+		seq_printf(m, "DevInfo: 0x%x Attributes: 0x%x"
 				 "\nPathComponentMax: %d Status: %d",
 			    le32_to_cpu(tcon->fsDevInfo.DeviceCharacteristics),
 			    le32_to_cpu(tcon->fsAttrInfo.Attributes),
 			    le32_to_cpu(tcon->fsAttrInfo.MaxPathNameComponentLength),
 			    tcon->tidStatus);
-		buf += length;
 		if (dev_type == FILE_DEVICE_DISK)
-			length = sprintf(buf, " type: DISK ");
+			seq_puts(m, " type: DISK ");
 		else if (dev_type == FILE_DEVICE_CD_ROM)
-			length = sprintf(buf, " type: CDROM ");
+			seq_puts(m, " type: CDROM ");
 		else
-			length =
-			    sprintf(buf, " type: %d ", dev_type);
-		buf += length;
+			seq_printf(m, " type: %d ", dev_type);
 		if (tcon->tidStatus == CifsNeedReconnect) {
-			buf += sprintf(buf, "\tDISCONNECTED ");
-			length += 14;
+			seq_puts(m, "\tDISCONNECTED ");
 		}
 	}
 	read_unlock(&GlobalSMBSeslock);
 
-	length = sprintf(buf, "\n");
-	buf += length;
+	seq_putc(m, '\n');
 
 	/* BB add code to dump additional info such as TCP session info now */
-	/* Now calculate total size of returned data */
-	length = buf - original_buf;
-
-	if (offset + count >= length)
-		*eof = 1;
-	if (length < offset) {
-		*eof = 1;
-		return 0;
-	} else {
-		length = length - offset;
-	}
-	if (length > count)
-		length = count;
-
-	return length;
+	return 0;
 }
 
-#ifdef CONFIG_CIFS_STATS
+static int cifs_debug_data_proc_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, cifs_debug_data_proc_show, NULL);
+}
 
-static int
-cifs_stats_write(struct file *file, const char __user *buffer,
-		 unsigned long count, void *data)
+static const struct file_operations cifs_debug_data_proc_fops = {
+	.owner		= THIS_MODULE,
+	.open		= cifs_debug_data_proc_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
+#ifdef CONFIG_CIFS_STATS
+static ssize_t cifs_stats_proc_write(struct file *file,
+		const char __user *buffer, size_t count, loff_t *ppos)
 {
 	char c;
 	int rc;
@@ -307,236 +279,127 @@ cifs_stats_write(struct file *file, const char __user *buffer,
 	return count;
 }
 
-static int
-cifs_stats_read(char *buf, char **beginBuffer, off_t offset,
-		  int count, int *eof, void *data)
+static int cifs_stats_proc_show(struct seq_file *m, void *v)
 {
-	int item_length, i, length;
+	int i;
 	struct list_head *tmp;
 	struct cifsTconInfo *tcon;
 
-	*beginBuffer = buf + offset;
-
-	length = sprintf(buf,
+	seq_printf(m,
 			"Resources in use\nCIFS Session: %d\n",
 			sesInfoAllocCount.counter);
-	buf += length;
-	item_length =
-		sprintf(buf, "Share (unique mount targets): %d\n",
+	seq_printf(m, "Share (unique mount targets): %d\n",
 			tconInfoAllocCount.counter);
-	length += item_length;
-	buf += item_length;
-	item_length =
-		sprintf(buf, "SMB Request/Response Buffer: %d Pool size: %d\n",
+	seq_printf(m, "SMB Request/Response Buffer: %d Pool size: %d\n",
 			bufAllocCount.counter,
 			cifs_min_rcv + tcpSesAllocCount.counter);
-	length += item_length;
-	buf += item_length;
-	item_length =
-		sprintf(buf, "SMB Small Req/Resp Buffer: %d Pool size: %d\n",
+	seq_printf(m, "SMB Small Req/Resp Buffer: %d Pool size: %d\n",
 			smBufAllocCount.counter, cifs_min_small);
-	length += item_length;
-	buf += item_length;
 #ifdef CONFIG_CIFS_STATS2
-	item_length = sprintf(buf, "Total Large %d Small %d Allocations\n",
+	seq_printf(m, "Total Large %d Small %d Allocations\n",
 				atomic_read(&totBufAllocCount),
 				atomic_read(&totSmBufAllocCount));
-	length += item_length;
-	buf += item_length;
 #endif /* CONFIG_CIFS_STATS2 */
 
-	item_length =
-		sprintf(buf, "Operations (MIDs): %d\n",
-			midCount.counter);
-	length += item_length;
-	buf += item_length;
-	item_length = sprintf(buf,
+	seq_printf(m, "Operations (MIDs): %d\n", midCount.counter);
+	seq_printf(m,
 		"\n%d session %d share reconnects\n",
 		tcpSesReconnectCount.counter, tconInfoReconnectCount.counter);
-	length += item_length;
-	buf += item_length;
 
-	item_length = sprintf(buf,
+	seq_printf(m,
 		"Total vfs operations: %d maximum at one time: %d\n",
 		GlobalCurrentXid, GlobalMaxActiveXid);
-	length += item_length;
-	buf += item_length;
 
 	i = 0;
 	read_lock(&GlobalSMBSeslock);
 	list_for_each(tmp, &GlobalTreeConnectionList) {
 		i++;
 		tcon = list_entry(tmp, struct cifsTconInfo, cifsConnectionList);
-		item_length = sprintf(buf, "\n%d) %s", i, tcon->treeName);
-		buf += item_length;
-		length += item_length;
-		if (tcon->tidStatus == CifsNeedReconnect) {
-			buf += sprintf(buf, "\tDISCONNECTED ");
-			length += 14;
-		}
-		item_length = sprintf(buf, "\nSMBs: %d Oplock Breaks: %d",
+		seq_printf(m, "\n%d) %s", i, tcon->treeName);
+		if (tcon->tidStatus == CifsNeedReconnect)
+			seq_puts(m, "\tDISCONNECTED ");
+		seq_printf(m, "\nSMBs: %d Oplock Breaks: %d",
 			atomic_read(&tcon->num_smbs_sent),
 			atomic_read(&tcon->num_oplock_brks));
-		buf += item_length;
-		length += item_length;
-		item_length = sprintf(buf, "\nReads:  %d Bytes: %lld",
+		seq_printf(m, "\nReads:  %d Bytes: %lld",
 			atomic_read(&tcon->num_reads),
 			(long long)(tcon->bytes_read));
-		buf += item_length;
-		length += item_length;
-		item_length = sprintf(buf, "\nWrites: %d Bytes: %lld",
+		seq_printf(m, "\nWrites: %d Bytes: %lld",
 			atomic_read(&tcon->num_writes),
 			(long long)(tcon->bytes_written));
-		buf += item_length;
-		length += item_length;
-		item_length = sprintf(buf,
+		seq_printf(m,
 			"\nLocks: %d HardLinks: %d Symlinks: %d",
 			atomic_read(&tcon->num_locks),
 			atomic_read(&tcon->num_hardlinks),
 			atomic_read(&tcon->num_symlinks));
-		buf += item_length;
-		length += item_length;
 
-		item_length = sprintf(buf, "\nOpens: %d Closes: %d Deletes: %d",
+		seq_printf(m, "\nOpens: %d Closes: %d Deletes: %d",
 			atomic_read(&tcon->num_opens),
 			atomic_read(&tcon->num_closes),
 			atomic_read(&tcon->num_deletes));
-		buf += item_length;
-		length += item_length;
-		item_length = sprintf(buf, "\nMkdirs: %d Rmdirs: %d",
+		seq_printf(m, "\nMkdirs: %d Rmdirs: %d",
 			atomic_read(&tcon->num_mkdirs),
 			atomic_read(&tcon->num_rmdirs));
-		buf += item_length;
-		length += item_length;
-		item_length = sprintf(buf, "\nRenames: %d T2 Renames %d",
+		seq_printf(m, "\nRenames: %d T2 Renames %d",
 			atomic_read(&tcon->num_renames),
 			atomic_read(&tcon->num_t2renames));
-		buf += item_length;
-		length += item_length;
-		item_length = sprintf(buf, "\nFindFirst: %d FNext %d FClose %d",
+		seq_printf(m, "\nFindFirst: %d FNext %d FClose %d",
 			atomic_read(&tcon->num_ffirst),
 			atomic_read(&tcon->num_fnext),
 			atomic_read(&tcon->num_fclose));
-		buf += item_length;
-		length += item_length;
 	}
 	read_unlock(&GlobalSMBSeslock);
 
-	buf += sprintf(buf, "\n");
-	length++;
-
-	if (offset + count >= length)
-		*eof = 1;
-	if (length < offset) {
-		*eof = 1;
-		return 0;
-	} else {
-		length = length - offset;
-	}
-	if (length > count)
-		length = count;
-
-	return length;
+	seq_putc(m, '\n');
+	return 0;
 }
+
+static int cifs_stats_proc_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, cifs_stats_proc_show, NULL);
+}
+
+static const struct file_operations cifs_stats_proc_fops = {
+	.owner		= THIS_MODULE,
+	.open		= cifs_stats_proc_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+	.write		= cifs_stats_proc_write,
+};
 #endif /* STATS */
 
 static struct proc_dir_entry *proc_fs_cifs;
-read_proc_t cifs_txanchor_read;
-static read_proc_t cifsFYI_read;
-static write_proc_t cifsFYI_write;
-static read_proc_t oplockEnabled_read;
-static write_proc_t oplockEnabled_write;
-static read_proc_t lookupFlag_read;
-static write_proc_t lookupFlag_write;
-static read_proc_t traceSMB_read;
-static write_proc_t traceSMB_write;
-static read_proc_t multiuser_mount_read;
-static write_proc_t multiuser_mount_write;
-static read_proc_t security_flags_read;
-static write_proc_t security_flags_write;
-/* static read_proc_t ntlmv2_enabled_read;
-static write_proc_t ntlmv2_enabled_write;
-static read_proc_t packet_signing_enabled_read;
-static write_proc_t packet_signing_enabled_write;*/
-static read_proc_t experimEnabled_read;
-static write_proc_t experimEnabled_write;
-static read_proc_t linuxExtensionsEnabled_read;
-static write_proc_t linuxExtensionsEnabled_write;
+static const struct file_operations cifsFYI_proc_fops;
+static const struct file_operations cifs_oplock_proc_fops;
+static const struct file_operations cifs_lookup_cache_proc_fops;
+static const struct file_operations traceSMB_proc_fops;
+static const struct file_operations cifs_multiuser_mount_proc_fops;
+static const struct file_operations cifs_security_flags_proc_fops;
+static const struct file_operations cifs_experimental_proc_fops;
+static const struct file_operations cifs_linux_ext_proc_fops;
 
 void
 cifs_proc_init(void)
 {
-	struct proc_dir_entry *pde;
-
 	proc_fs_cifs = proc_mkdir("fs/cifs", NULL);
 	if (proc_fs_cifs == NULL)
 		return;
 
 	proc_fs_cifs->owner = THIS_MODULE;
-	create_proc_read_entry("DebugData", 0, proc_fs_cifs,
-				cifs_debug_data_read, NULL);
+	proc_create("DebugData", 0, proc_fs_cifs, &cifs_debug_data_proc_fops);
 
 #ifdef CONFIG_CIFS_STATS
-	pde = create_proc_read_entry("Stats", 0, proc_fs_cifs,
-				cifs_stats_read, NULL);
-	if (pde)
-		pde->write_proc = cifs_stats_write;
+	proc_create("Stats", 0, proc_fs_cifs, &cifs_stats_proc_fops);
 #endif /* STATS */
-	pde = create_proc_read_entry("cifsFYI", 0, proc_fs_cifs,
-				cifsFYI_read, NULL);
-	if (pde)
-		pde->write_proc = cifsFYI_write;
-
-	pde =
-	    create_proc_read_entry("traceSMB", 0, proc_fs_cifs,
-				traceSMB_read, NULL);
-	if (pde)
-		pde->write_proc = traceSMB_write;
-
-	pde = create_proc_read_entry("OplockEnabled", 0, proc_fs_cifs,
-				oplockEnabled_read, NULL);
-	if (pde)
-		pde->write_proc = oplockEnabled_write;
-
-	pde = create_proc_read_entry("Experimental", 0, proc_fs_cifs,
-				experimEnabled_read, NULL);
-	if (pde)
-		pde->write_proc = experimEnabled_write;
-
-	pde = create_proc_read_entry("LinuxExtensionsEnabled", 0, proc_fs_cifs,
-				linuxExtensionsEnabled_read, NULL);
-	if (pde)
-		pde->write_proc = linuxExtensionsEnabled_write;
-
-	pde =
-	    create_proc_read_entry("MultiuserMount", 0, proc_fs_cifs,
-				multiuser_mount_read, NULL);
-	if (pde)
-		pde->write_proc = multiuser_mount_write;
-
-	pde =
-	    create_proc_read_entry("SecurityFlags", 0, proc_fs_cifs,
-				security_flags_read, NULL);
-	if (pde)
-		pde->write_proc = security_flags_write;
-
-	pde =
-	create_proc_read_entry("LookupCacheEnabled", 0, proc_fs_cifs,
-				lookupFlag_read, NULL);
-	if (pde)
-		pde->write_proc = lookupFlag_write;
-
-/*	pde =
-	    create_proc_read_entry("NTLMV2Enabled", 0, proc_fs_cifs,
-				ntlmv2_enabled_read, NULL);
-	if (pde)
-		pde->write_proc = ntlmv2_enabled_write;
-
-	pde =
-	    create_proc_read_entry("PacketSigningEnabled", 0, proc_fs_cifs,
-				packet_signing_enabled_read, NULL);
-	if (pde)
-		pde->write_proc = packet_signing_enabled_write;*/
+	proc_create("cifsFYI", 0, proc_fs_cifs, &cifsFYI_proc_fops);
+	proc_create("traceSMB", 0, proc_fs_cifs, &traceSMB_proc_fops);
+	proc_create("OplockEnabled", 0, proc_fs_cifs, &cifs_oplock_proc_fops);
+	proc_create("Experimental", 0, proc_fs_cifs, &cifs_experimental_proc_fops);
+	proc_create("LinuxExtensionsEnabled", 0, proc_fs_cifs, &cifs_linux_ext_proc_fops);
+	proc_create("MultiuserMount", 0, proc_fs_cifs, &cifs_multiuser_mount_proc_fops);
+	proc_create("SecurityFlags", 0, proc_fs_cifs, &cifs_security_flags_proc_fops);
+	proc_create("LookupCacheEnabled", 0, proc_fs_cifs, &cifs_lookup_cache_proc_fops);
 }
 
 void
@@ -553,39 +416,26 @@ cifs_proc_clean(void)
 #endif
 	remove_proc_entry("MultiuserMount", proc_fs_cifs);
 	remove_proc_entry("OplockEnabled", proc_fs_cifs);
-/*	remove_proc_entry("NTLMV2Enabled",proc_fs_cifs); */
 	remove_proc_entry("SecurityFlags", proc_fs_cifs);
-/*	remove_proc_entry("PacketSigningEnabled", proc_fs_cifs); */
 	remove_proc_entry("LinuxExtensionsEnabled", proc_fs_cifs);
 	remove_proc_entry("Experimental", proc_fs_cifs);
 	remove_proc_entry("LookupCacheEnabled", proc_fs_cifs);
 	remove_proc_entry("fs/cifs", NULL);
 }
 
-static int
-cifsFYI_read(char *page, char **start, off_t off, int count,
-	     int *eof, void *data)
+static int cifsFYI_proc_show(struct seq_file *m, void *v)
 {
-	int len;
-
-	len = sprintf(page, "%d\n", cifsFYI);
-
-	len -= off;
-	*start = page + off;
-
-	if (len > count)
-		len = count;
-	else
-		*eof = 1;
-
-	if (len < 0)
-		len = 0;
-
-	return len;
+	seq_printf(m, "%d\n", cifsFYI);
+	return 0;
 }
-static int
-cifsFYI_write(struct file *file, const char __user *buffer,
-	      unsigned long count, void *data)
+
+static int cifsFYI_proc_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, cifsFYI_proc_show, NULL);
+}
+
+static ssize_t cifsFYI_proc_write(struct file *file, const char __user *buffer,
+		size_t count, loff_t *ppos)
 {
 	char c;
 	int rc;
@@ -603,30 +453,28 @@ cifsFYI_write(struct file *file, const char __user *buffer,
 	return count;
 }
 
-static int
-oplockEnabled_read(char *page, char **start, off_t off,
-		   int count, int *eof, void *data)
+static const struct file_operations cifsFYI_proc_fops = {
+	.owner		= THIS_MODULE,
+	.open		= cifsFYI_proc_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+	.write		= cifsFYI_proc_write,
+};
+
+static int cifs_oplock_proc_show(struct seq_file *m, void *v)
 {
-	int len;
-
-	len = sprintf(page, "%d\n", oplockEnabled);
-
-	len -= off;
-	*start = page + off;
-
-	if (len > count)
-		len = count;
-	else
-		*eof = 1;
-
-	if (len < 0)
-		len = 0;
-
-	return len;
+	seq_printf(m, "%d\n", oplockEnabled);
+	return 0;
 }
-static int
-oplockEnabled_write(struct file *file, const char __user *buffer,
-		    unsigned long count, void *data)
+
+static int cifs_oplock_proc_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, cifs_oplock_proc_show, NULL);
+}
+
+static ssize_t cifs_oplock_proc_write(struct file *file,
+		const char __user *buffer, size_t count, loff_t *ppos)
 {
 	char c;
 	int rc;
@@ -642,30 +490,28 @@ oplockEnabled_write(struct file *file, const char __user *buffer,
 	return count;
 }
 
-static int
-experimEnabled_read(char *page, char **start, off_t off,
-		    int count, int *eof, void *data)
+static const struct file_operations cifs_oplock_proc_fops = {
+	.owner		= THIS_MODULE,
+	.open		= cifs_oplock_proc_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+	.write		= cifs_oplock_proc_write,
+};
+
+static int cifs_experimental_proc_show(struct seq_file *m, void *v)
 {
-	int len;
-
-	len = sprintf(page, "%d\n", experimEnabled);
-
-	len -= off;
-	*start = page + off;
-
-	if (len > count)
-		len = count;
-	else
-		*eof = 1;
-
-	if (len < 0)
-		len = 0;
-
-	return len;
+	seq_printf(m, "%d\n", experimEnabled);
+	return 0;
 }
-static int
-experimEnabled_write(struct file *file, const char __user *buffer,
-		     unsigned long count, void *data)
+
+static int cifs_experimental_proc_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, cifs_experimental_proc_show, NULL);
+}
+
+static ssize_t cifs_experimental_proc_write(struct file *file,
+		const char __user *buffer, size_t count, loff_t *ppos)
 {
 	char c;
 	int rc;
@@ -683,29 +529,28 @@ experimEnabled_write(struct file *file, const char __user *buffer,
 	return count;
 }
 
-static int
-linuxExtensionsEnabled_read(char *page, char **start, off_t off,
-			    int count, int *eof, void *data)
+static const struct file_operations cifs_experimental_proc_fops = {
+	.owner		= THIS_MODULE,
+	.open		= cifs_experimental_proc_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+	.write		= cifs_experimental_proc_write,
+};
+
+static int cifs_linux_ext_proc_show(struct seq_file *m, void *v)
 {
-	int len;
-
-	len = sprintf(page, "%d\n", linuxExtEnabled);
-	len -= off;
-	*start = page + off;
-
-	if (len > count)
-		len = count;
-	else
-		*eof = 1;
-
-	if (len < 0)
-		len = 0;
-
-	return len;
+	seq_printf(m, "%d\n", linuxExtEnabled);
+	return 0;
 }
-static int
-linuxExtensionsEnabled_write(struct file *file, const char __user *buffer,
-			     unsigned long count, void *data)
+
+static int cifs_linux_ext_proc_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, cifs_linux_ext_proc_show, NULL);
+}
+
+static ssize_t cifs_linux_ext_proc_write(struct file *file,
+		const char __user *buffer, size_t count, loff_t *ppos)
 {
 	char c;
 	int rc;
@@ -721,31 +566,28 @@ linuxExtensionsEnabled_write(struct file *file, const char __user *buffer,
 	return count;
 }
 
+static const struct file_operations cifs_linux_ext_proc_fops = {
+	.owner		= THIS_MODULE,
+	.open		= cifs_linux_ext_proc_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+	.write		= cifs_linux_ext_proc_write,
+};
 
-static int
-lookupFlag_read(char *page, char **start, off_t off,
-		int count, int *eof, void *data)
+static int cifs_lookup_cache_proc_show(struct seq_file *m, void *v)
 {
-	int len;
-
-	len = sprintf(page, "%d\n", lookupCacheEnabled);
-
-	len -= off;
-	*start = page + off;
-
-	if (len > count)
-		len = count;
-	else
-		*eof = 1;
-
-	if (len < 0)
-		len = 0;
-
-	return len;
+	seq_printf(m, "%d\n", lookupCacheEnabled);
+	return 0;
 }
-static int
-lookupFlag_write(struct file *file, const char __user *buffer,
-		    unsigned long count, void *data)
+
+static int cifs_lookup_cache_proc_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, cifs_lookup_cache_proc_show, NULL);
+}
+
+static ssize_t cifs_lookup_cache_proc_write(struct file *file,
+		const char __user *buffer, size_t count, loff_t *ppos)
 {
 	char c;
 	int rc;
@@ -760,30 +602,29 @@ lookupFlag_write(struct file *file, const char __user *buffer,
 
 	return count;
 }
-static int
-traceSMB_read(char *page, char **start, off_t off, int count,
-	      int *eof, void *data)
+
+static const struct file_operations cifs_lookup_cache_proc_fops = {
+	.owner		= THIS_MODULE,
+	.open		= cifs_lookup_cache_proc_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+	.write		= cifs_lookup_cache_proc_write,
+};
+
+static int traceSMB_proc_show(struct seq_file *m, void *v)
 {
-	int len;
-
-	len = sprintf(page, "%d\n", traceSMB);
-
-	len -= off;
-	*start = page + off;
-
-	if (len > count)
-		len = count;
-	else
-		*eof = 1;
-
-	if (len < 0)
-		len = 0;
-
-	return len;
+	seq_printf(m, "%d\n", traceSMB);
+	return 0;
 }
-static int
-traceSMB_write(struct file *file, const char __user *buffer,
-	       unsigned long count, void *data)
+
+static int traceSMB_proc_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, traceSMB_proc_show, NULL);
+}
+
+static ssize_t traceSMB_proc_write(struct file *file, const char __user *buffer,
+		size_t count, loff_t *ppos)
 {
 	char c;
 	int rc;
@@ -799,30 +640,28 @@ traceSMB_write(struct file *file, const char __user *buffer,
 	return count;
 }
 
-static int
-multiuser_mount_read(char *page, char **start, off_t off,
-		     int count, int *eof, void *data)
+static const struct file_operations traceSMB_proc_fops = {
+	.owner		= THIS_MODULE,
+	.open		= traceSMB_proc_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+	.write		= traceSMB_proc_write,
+};
+
+static int cifs_multiuser_mount_proc_show(struct seq_file *m, void *v)
 {
-	int len;
-
-	len = sprintf(page, "%d\n", multiuser_mount);
-
-	len -= off;
-	*start = page + off;
-
-	if (len > count)
-		len = count;
-	else
-		*eof = 1;
-
-	if (len < 0)
-		len = 0;
-
-	return len;
+	seq_printf(m, "%d\n", multiuser_mount);
+	return 0;
 }
-static int
-multiuser_mount_write(struct file *file, const char __user *buffer,
-		      unsigned long count, void *data)
+
+static int cifs_multiuser_mount_proc_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, cifs_multiuser_mount_proc_show, NULL);
+}
+
+static ssize_t cifs_multiuser_mount_proc_write(struct file *file,
+		const char __user *buffer, size_t count, loff_t *ppos)
 {
 	char c;
 	int rc;
@@ -838,30 +677,28 @@ multiuser_mount_write(struct file *file, const char __user *buffer,
 	return count;
 }
 
-static int
-security_flags_read(char *page, char **start, off_t off,
-		       int count, int *eof, void *data)
+static const struct file_operations cifs_multiuser_mount_proc_fops = {
+	.owner		= THIS_MODULE,
+	.open		= cifs_multiuser_mount_proc_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+	.write		= cifs_multiuser_mount_proc_write,
+};
+
+static int cifs_security_flags_proc_show(struct seq_file *m, void *v)
 {
-	int len;
-
-	len = sprintf(page, "0x%x\n", extended_security);
-
-	len -= off;
-	*start = page + off;
-
-	if (len > count)
-		len = count;
-	else
-		*eof = 1;
-
-	if (len < 0)
-		len = 0;
-
-	return len;
+	seq_printf(m, "0x%x\n", extended_security);
+	return 0;
 }
-static int
-security_flags_write(struct file *file, const char __user *buffer,
-			unsigned long count, void *data)
+
+static int cifs_security_flags_proc_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, cifs_security_flags_proc_show, NULL);
+}
+
+static ssize_t cifs_security_flags_proc_write(struct file *file,
+		const char __user *buffer, size_t count, loff_t *ppos)
 {
 	unsigned int flags;
 	char flags_string[12];
@@ -917,6 +754,15 @@ security_flags_write(struct file *file, const char __user *buffer,
 	/* BB should we turn on MAY flags for other MUST options? */
 	return count;
 }
+
+static const struct file_operations cifs_security_flags_proc_fops = {
+	.owner		= THIS_MODULE,
+	.open		= cifs_security_flags_proc_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+	.write		= cifs_security_flags_proc_write,
+};
 #else
 inline void cifs_proc_init(void)
 {

From 99b1f5b2f6cd2f65cce02c5f63302df5878a5fbc Mon Sep 17 00:00:00 2001
From: Steve French <sfrench@us.ibm.com>
Date: Thu, 24 Jul 2008 02:37:45 +0000
Subject: [PATCH 0164/1171] [CIFS] remove checkpatch warning

Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/cifs_debug.c | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)

diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c
index 73925a77375e..c213aa40064e 100644
--- a/fs/cifs/cifs_debug.c
+++ b/fs/cifs/cifs_debug.c
@@ -395,11 +395,16 @@ cifs_proc_init(void)
 	proc_create("cifsFYI", 0, proc_fs_cifs, &cifsFYI_proc_fops);
 	proc_create("traceSMB", 0, proc_fs_cifs, &traceSMB_proc_fops);
 	proc_create("OplockEnabled", 0, proc_fs_cifs, &cifs_oplock_proc_fops);
-	proc_create("Experimental", 0, proc_fs_cifs, &cifs_experimental_proc_fops);
-	proc_create("LinuxExtensionsEnabled", 0, proc_fs_cifs, &cifs_linux_ext_proc_fops);
-	proc_create("MultiuserMount", 0, proc_fs_cifs, &cifs_multiuser_mount_proc_fops);
-	proc_create("SecurityFlags", 0, proc_fs_cifs, &cifs_security_flags_proc_fops);
-	proc_create("LookupCacheEnabled", 0, proc_fs_cifs, &cifs_lookup_cache_proc_fops);
+	proc_create("Experimental", 0, proc_fs_cifs,
+		    &cifs_experimental_proc_fops);
+	proc_create("LinuxExtensionsEnabled", 0, proc_fs_cifs,
+		    &cifs_linux_ext_proc_fops);
+	proc_create("MultiuserMount", 0, proc_fs_cifs,
+		    &cifs_multiuser_mount_proc_fops);
+	proc_create("SecurityFlags", 0, proc_fs_cifs,
+		    &cifs_security_flags_proc_fops);
+	proc_create("LookupCacheEnabled", 0, proc_fs_cifs,
+		    &cifs_lookup_cache_proc_fops);
 }
 
 void
@@ -655,9 +660,9 @@ static int cifs_multiuser_mount_proc_show(struct seq_file *m, void *v)
 	return 0;
 }
 
-static int cifs_multiuser_mount_proc_open(struct inode *inode, struct file *file)
+static int cifs_multiuser_mount_proc_open(struct inode *inode, struct file *fh)
 {
-	return single_open(file, cifs_multiuser_mount_proc_show, NULL);
+	return single_open(fh, cifs_multiuser_mount_proc_show, NULL);
 }
 
 static ssize_t cifs_multiuser_mount_proc_write(struct file *file,

From 6f75a9b6426e686649ac440c37ec7c249501f9a5 Mon Sep 17 00:00:00 2001
From: Chas Williams <chas@cmf.nrl.navy.mil>
Date: Wed, 23 Jul 2008 20:29:21 -0700
Subject: [PATCH 0165/1171] atm: [fore200e] use MODULE_FIRMWARE() and other
 suggested cleanups

Signed-off-by: Chas Williams <chas@cmf.nrl.navy.mil>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/atm/fore200e.c | 15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)

diff --git a/drivers/atm/fore200e.c b/drivers/atm/fore200e.c
index d5c1bbfbe79d..73338d231db9 100644
--- a/drivers/atm/fore200e.c
+++ b/drivers/atm/fore200e.c
@@ -2562,7 +2562,8 @@ fore200e_load_and_start_fw(struct fore200e* fore200e)
     const struct firmware *firmware;
     struct device *device;
     struct fw_header *fw_header;
-    u32 *fw_data, fw_size;
+    const __le32 *fw_data;
+    u32 fw_size;
     u32 __iomem *load_addr;
     char buf[48];
     int err = -ENODEV;
@@ -2582,7 +2583,7 @@ fore200e_load_and_start_fw(struct fore200e* fore200e)
 	return err;
     }
 
-    fw_data = (u32 *) firmware->data;
+    fw_data = (__le32 *) firmware->data;
     fw_size = firmware->size / sizeof(u32);
     fw_header = (struct fw_header *) firmware->data;
     load_addr = fore200e->virt_base + le32_to_cpu(fw_header->load_offset);
@@ -3199,6 +3200,14 @@ static const struct fore200e_bus fore200e_bus[] = {
     {}
 };
 
-#ifdef MODULE_LICENSE
 MODULE_LICENSE("GPL");
+#ifdef CONFIG_PCI
+#ifdef __LITTLE_ENDIAN__
+MODULE_FIRMWARE("pca200e.bin");
+#else
+MODULE_FIRMWARE("pca200e_ecd.bin2");
+#endif
+#endif /* CONFIG_PCI */
+#ifdef CONFIG_SBUS
+MODULE_FIRMWARE("sba200e_ecd.bin2");
 #endif

From f867e6af94239a04ec23aeec2fcda5aa58e41db7 Mon Sep 17 00:00:00 2001
From: Jarek Poplawski <jarkao2@gmail.com>
Date: Wed, 23 Jul 2008 21:34:27 -0700
Subject: [PATCH 0166/1171] pkt_sched: sch_sfq: dump a real number of flows

Dump the "flows" number according to the number of active flows
instead of repeating the "limit".

Reported-by: Denys Fedoryshchenko <denys@visp.net.lb>
Signed-off-by: Jarek Poplawski <jarkao2@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sched/sch_sfq.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index 8589da666568..73f53844ce97 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -536,7 +536,14 @@ static int sfq_dump(struct Qdisc *sch, struct sk_buff *skb)
 
 	opt.limit = q->limit;
 	opt.divisor = SFQ_HASH_DIVISOR;
-	opt.flows = q->limit;
+	opt.flows = 0;
+	if (q->tail != SFQ_DEPTH) {
+		unsigned int i;
+
+		for (i = 0; i < SFQ_HASH_DIVISOR; i++)
+			if (q->ht[i] != SFQ_DEPTH)
+				opt.flows++;
+	}
 
 	NLA_PUT(skb, TCA_OPTIONS, sizeof(opt), &opt);
 

From 1986b0cb1671ea39178b4e2b00461109728fc935 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Thu, 24 Jul 2008 08:10:02 +0200
Subject: [PATCH 0167/1171] ftrace: remove latency-tracer leftover

remove the :vim=ft=help tag from trace files.

I used them years ago to syntax-highlight traces and forgot about this hack.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/trace/trace.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index 868e121c8e38..fc20e09a6cb1 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -1203,9 +1203,6 @@ static void *s_next(struct seq_file *m, void *v, loff_t *pos)
 
 	iter->pos = *pos;
 
-	if (last_ent && !ent)
-		seq_puts(m, "\n\nvim:ft=help\n");
-
 	return ent;
 }
 

From 9e882c9282512cc622752f29ec7c29ce338fc1eb Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Mon, 21 Jul 2008 16:49:54 -0700
Subject: [PATCH 0168/1171] x86: call early_cpu_init at the same point

Call early_cpu_init() at the same (early) point in setup_arch().
The x86_64 code was calling it relatively late, after when other arch
code need to do cpu-related setup which depends on it.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/setup.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index b4aacb9f52e3..b520dae02bf4 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -597,11 +597,11 @@ void __init setup_arch(char **cmdline_p)
 	memcpy(&boot_cpu_data, &new_cpu_data, sizeof(new_cpu_data));
 	visws_early_detect();
 	pre_setup_arch_hook();
-	early_cpu_init();
 #else
 	printk(KERN_INFO "Command line: %s\n", boot_command_line);
 #endif
 
+	early_cpu_init();
 	early_ioremap_init();
 
 	ROOT_DEV = old_decode_dev(boot_params.hdr.root_dev);
@@ -665,9 +665,6 @@ void __init setup_arch(char **cmdline_p)
 	bss_resource.start = virt_to_phys(&__bss_start);
 	bss_resource.end = virt_to_phys(&__bss_stop)-1;
 
-#ifdef CONFIG_X86_64
-	early_cpu_init();
-#endif
 	strlcpy(command_line, boot_command_line, COMMAND_LINE_SIZE);
 	*cmdline_p = command_line;
 

From 2dc1697eb355c34f9f7bcbbb83f490de248c360a Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Mon, 21 Jul 2008 16:49:58 -0700
Subject: [PATCH 0169/1171] xen: don't use sysret for sysexit32

When implementing sysexit32, don't let Xen use sysret to return to
userspace.  That results in usermode register state being trashed.

Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/xen/xen-asm_64.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/xen/xen-asm_64.S b/arch/x86/xen/xen-asm_64.S
index 4038cbfe3331..7f58304fafb3 100644
--- a/arch/x86/xen/xen-asm_64.S
+++ b/arch/x86/xen/xen-asm_64.S
@@ -173,7 +173,7 @@ ENTRY(xen_sysexit)
 	pushq $__USER32_CS
 	pushq %rdx
 
-	pushq $VGCF_in_syscall
+	pushq $0
 1:	jmp hypercall_iret
 ENDPATCH(xen_sysexit)
 RELOC(xen_sysexit, 1b+1)

From 979c9296bdcfded58ebac41905c3397317df0355 Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Wed, 14 May 2008 16:10:33 +0300
Subject: [PATCH 0170/1171] UBI: print error code

Print error code if checking failed which is very useful
to identify problems.

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/vtbl.c | 39 +++++++++++++++++++++------------------
 1 file changed, 21 insertions(+), 18 deletions(-)

diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c
index af36b12be278..3c4d68f2cfd4 100644
--- a/drivers/mtd/ubi/vtbl.c
+++ b/drivers/mtd/ubi/vtbl.c
@@ -127,7 +127,7 @@ static int vtbl_check(const struct ubi_device *ubi,
 		      const struct ubi_vtbl_record *vtbl)
 {
 	int i, n, reserved_pebs, alignment, data_pad, vol_type, name_len;
-	int upd_marker;
+	int upd_marker, err;
 	uint32_t crc;
 	const char *name;
 
@@ -153,7 +153,7 @@ static int vtbl_check(const struct ubi_device *ubi,
 		if (reserved_pebs == 0) {
 			if (memcmp(&vtbl[i], &empty_vtbl_record,
 						UBI_VTBL_RECORD_SIZE)) {
-				dbg_err("bad empty record");
+				err = 2;
 				goto bad;
 			}
 			continue;
@@ -161,56 +161,57 @@ static int vtbl_check(const struct ubi_device *ubi,
 
 		if (reserved_pebs < 0 || alignment < 0 || data_pad < 0 ||
 		    name_len < 0) {
-			dbg_err("negative values");
+			err = 3;
 			goto bad;
 		}
 
 		if (alignment > ubi->leb_size || alignment == 0) {
-			dbg_err("bad alignment");
+			err = 4;
 			goto bad;
 		}
 
 		n = alignment % ubi->min_io_size;
 		if (alignment != 1 && n) {
-			dbg_err("alignment is not multiple of min I/O unit");
+			err = 5;
 			goto bad;
 		}
 
 		n = ubi->leb_size % alignment;
 		if (data_pad != n) {
 			dbg_err("bad data_pad, has to be %d", n);
+			err = 6;
 			goto bad;
 		}
 
 		if (vol_type != UBI_VID_DYNAMIC && vol_type != UBI_VID_STATIC) {
-			dbg_err("bad vol_type");
+			err = 7;
 			goto bad;
 		}
 
 		if (upd_marker != 0 && upd_marker != 1) {
-			dbg_err("bad upd_marker");
+			err = 8;
 			goto bad;
 		}
 
 		if (reserved_pebs > ubi->good_peb_count) {
 			dbg_err("too large reserved_pebs, good PEBs %d",
 				ubi->good_peb_count);
+			err = 9;
 			goto bad;
 		}
 
 		if (name_len > UBI_VOL_NAME_MAX) {
-			dbg_err("too long volume name, max %d",
-				UBI_VOL_NAME_MAX);
+			err = 10;
 			goto bad;
 		}
 
 		if (name[0] == '\0') {
-			dbg_err("NULL volume name");
+			err = 11;
 			goto bad;
 		}
 
 		if (name_len != strnlen(name, name_len + 1)) {
-			dbg_err("bad name_len");
+			err = 12;
 			goto bad;
 		}
 	}
@@ -235,7 +236,7 @@ static int vtbl_check(const struct ubi_device *ubi,
 	return 0;
 
 bad:
-	ubi_err("volume table check failed, record %d", i);
+	ubi_err("volume table check failed: record %d, error %d", i, err);
 	ubi_dbg_dump_vtbl_record(&vtbl[i], i);
 	return -EINVAL;
 }
@@ -620,30 +621,32 @@ static int init_volumes(struct ubi_device *ubi, const struct ubi_scan_info *si,
 static int check_sv(const struct ubi_volume *vol,
 		    const struct ubi_scan_volume *sv)
 {
+	int err;
+
 	if (sv->highest_lnum >= vol->reserved_pebs) {
-		dbg_err("bad highest_lnum");
+		err = 1;
 		goto bad;
 	}
 	if (sv->leb_count > vol->reserved_pebs) {
-		dbg_err("bad leb_count");
+		err = 2;
 		goto bad;
 	}
 	if (sv->vol_type != vol->vol_type) {
-		dbg_err("bad vol_type");
+		err = 3;
 		goto bad;
 	}
 	if (sv->used_ebs > vol->reserved_pebs) {
-		dbg_err("bad used_ebs");
+		err = 4;
 		goto bad;
 	}
 	if (sv->data_pad != vol->data_pad) {
-		dbg_err("bad data_pad");
+		err = 5;
 		goto bad;
 	}
 	return 0;
 
 bad:
-	ubi_err("bad scanning information");
+	ubi_err("bad scanning information, error %d", err);
 	ubi_dbg_dump_sv(sv);
 	ubi_dbg_dump_vol_info(vol);
 	return -EINVAL;

From beeea636030622f6de67d15c61f5b311a03d188c Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Tue, 20 May 2008 09:54:02 +0300
Subject: [PATCH 0171/1171] UBI: add a comment

It is not clear why we schedule PEB for scrubbing in case of
-EBADMSG. Elaborate.

Requested-by: Kyungmin Park <kmpark@infradead.org>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/vtbl.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c
index 3c4d68f2cfd4..42a7815086b7 100644
--- a/drivers/mtd/ubi/vtbl.c
+++ b/drivers/mtd/ubi/vtbl.c
@@ -385,7 +385,16 @@ static struct ubi_vtbl_record *process_lvol(struct ubi_device *ubi,
 		err = ubi_io_read_data(ubi, leb[seb->lnum], seb->pnum, 0,
 				       ubi->vtbl_size);
 		if (err == UBI_IO_BITFLIPS || err == -EBADMSG)
-			/* Scrub the PEB later */
+			/*
+			 * Scrub the PEB later. Note, -EBADMSG indicates an
+			 * uncorrectable ECC error, but we have our own CRC and
+			 * the data will be checked later. If the data is OK,
+			 * the PEB will be scrubbed (because we set
+			 * seb->scrub). If the data is not OK, the contents of
+			 * the PEB will be recovered from the second copy, and
+			 * seb->scrub will be cleared in
+			 * 'ubi_scan_add_used()'.
+			 */
 			seb->scrub = 1;
 		else if (err)
 			goto out_free;

From a0fd1efd488092951f310fdb777b8a540cf84dcb Mon Sep 17 00:00:00 2001
From: Kyungmin Park <kmpark@infradead.org>
Date: Wed, 21 May 2008 14:34:56 +0300
Subject: [PATCH 0172/1171] UBI: fix buffer padding

Instead of correctly pad the buffer wich we are writing to the
eraseblock during update, we used weird construct:

memset(buf + len, 0xFF, len - len);

Fix this.

Signed-off-by: Kyungmin Park <kmpark@infradead.org>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/upd.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/mtd/ubi/upd.c b/drivers/mtd/ubi/upd.c
index ddaa1a56cc69..6fa1ab3f2a70 100644
--- a/drivers/mtd/ubi/upd.c
+++ b/drivers/mtd/ubi/upd.c
@@ -237,10 +237,10 @@ static int write_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
 	int err;
 
 	if (vol->vol_type == UBI_DYNAMIC_VOLUME) {
-		len = ALIGN(len, ubi->min_io_size);
-		memset(buf + len, 0xFF, len - len);
+		int l = ALIGN(len, ubi->min_io_size);
 
-		len = ubi_calc_data_len(ubi, buf, len);
+		memset(buf + len, 0xFF, l - len);
+		len = ubi_calc_data_len(ubi, buf, l);
 		if (len == 0) {
 			dbg_msg("all %d bytes contain 0xFF - skip", len);
 			return 0;

From cadb40ccc16a26a738f1cbc963e35b21edd93e79 Mon Sep 17 00:00:00 2001
From: Kyungmin Park <kyungmin.park@samsung.com>
Date: Thu, 22 May 2008 10:32:18 +0900
Subject: [PATCH 0173/1171] UBI: avoid unnecessary division operations

UBI already checks that @min io size is the power of 2 at io_init.
It is save to use bit operations then.

Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/build.c | 8 ++++++--
 drivers/mtd/ubi/cdev.c  | 6 +++---
 drivers/mtd/ubi/eba.c   | 2 +-
 drivers/mtd/ubi/kapi.c  | 6 +++---
 drivers/mtd/ubi/misc.c  | 2 +-
 drivers/mtd/ubi/vmt.c   | 2 +-
 drivers/mtd/ubi/vtbl.c  | 5 ++---
 drivers/mtd/ubi/wl.c    | 3 +--
 8 files changed, 18 insertions(+), 16 deletions(-)

diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c
index 961416ac0616..ff4425de1527 100644
--- a/drivers/mtd/ubi/build.c
+++ b/drivers/mtd/ubi/build.c
@@ -530,7 +530,11 @@ static int io_init(struct ubi_device *ubi)
 	ubi->min_io_size = ubi->mtd->writesize;
 	ubi->hdrs_min_io_size = ubi->mtd->writesize >> ubi->mtd->subpage_sft;
 
-	/* Make sure minimal I/O unit is power of 2 */
+	/*
+	 * Make sure minimal I/O unit is power of 2. Note, there is no
+	 * fundamental reason for this assumption. It is just an optimization
+	 * which allows us to avoid costly division operations.
+	 */
 	if (!is_power_of_2(ubi->min_io_size)) {
 		ubi_err("min. I/O unit (%d) is not power of 2",
 			ubi->min_io_size);
@@ -581,7 +585,7 @@ static int io_init(struct ubi_device *ubi)
 	if (ubi->vid_hdr_offset < UBI_EC_HDR_SIZE ||
 	    ubi->leb_start < ubi->vid_hdr_offset + UBI_VID_HDR_SIZE ||
 	    ubi->leb_start > ubi->peb_size - UBI_VID_HDR_SIZE ||
-	    ubi->leb_start % ubi->min_io_size) {
+	    ubi->leb_start & (ubi->min_io_size - 1)) {
 		ubi_err("bad VID header (%d) or data offsets (%d)",
 			ubi->vid_hdr_offset, ubi->leb_start);
 		return -EINVAL;
diff --git a/drivers/mtd/ubi/cdev.c b/drivers/mtd/ubi/cdev.c
index 89193ba9451e..0cdaf9fba7b0 100644
--- a/drivers/mtd/ubi/cdev.c
+++ b/drivers/mtd/ubi/cdev.c
@@ -295,7 +295,7 @@ static ssize_t vol_cdev_direct_write(struct file *file, const char __user *buf,
 	off = do_div(tmp, vol->usable_leb_size);
 	lnum = tmp;
 
-	if (off % ubi->min_io_size) {
+	if (off & (ubi->min_io_size - 1)) {
 		dbg_err("unaligned position");
 		return -EINVAL;
 	}
@@ -304,7 +304,7 @@ static ssize_t vol_cdev_direct_write(struct file *file, const char __user *buf,
 		count_save = count = vol->used_bytes - *offp;
 
 	/* We can write only in fractions of the minimum I/O unit */
-	if (count % ubi->min_io_size) {
+	if (count & (ubi->min_io_size - 1)) {
 		dbg_err("unaligned write length");
 		return -EINVAL;
 	}
@@ -564,7 +564,7 @@ static int verify_mkvol_req(const struct ubi_device *ubi,
 	if (req->alignment > ubi->leb_size)
 		goto bad;
 
-	n = req->alignment % ubi->min_io_size;
+	n = req->alignment & (ubi->min_io_size - 1);
 	if (req->alignment != 1 && n)
 		goto bad;
 
diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index 7ce91ca742b1..37d778447943 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -752,7 +752,7 @@ int ubi_eba_write_leb_st(struct ubi_device *ubi, struct ubi_volume *vol,
 		/* If this is the last LEB @len may be unaligned */
 		len = ALIGN(data_size, ubi->min_io_size);
 	else
-		ubi_assert(len % ubi->min_io_size == 0);
+		ubi_assert(!(len & (ubi->min_io_size - 1)));
 
 	vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_NOFS);
 	if (!vid_hdr)
diff --git a/drivers/mtd/ubi/kapi.c b/drivers/mtd/ubi/kapi.c
index a70d58823f8d..51508832566d 100644
--- a/drivers/mtd/ubi/kapi.c
+++ b/drivers/mtd/ubi/kapi.c
@@ -397,8 +397,8 @@ int ubi_leb_write(struct ubi_volume_desc *desc, int lnum, const void *buf,
 		return -EROFS;
 
 	if (lnum < 0 || lnum >= vol->reserved_pebs || offset < 0 || len < 0 ||
-	    offset + len > vol->usable_leb_size || offset % ubi->min_io_size ||
-	    len % ubi->min_io_size)
+	    offset + len > vol->usable_leb_size ||
+	    offset & (ubi->min_io_size - 1) || len & (ubi->min_io_size - 1))
 		return -EINVAL;
 
 	if (dtype != UBI_LONGTERM && dtype != UBI_SHORTTERM &&
@@ -447,7 +447,7 @@ int ubi_leb_change(struct ubi_volume_desc *desc, int lnum, const void *buf,
 		return -EROFS;
 
 	if (lnum < 0 || lnum >= vol->reserved_pebs || len < 0 ||
-	    len > vol->usable_leb_size || len % ubi->min_io_size)
+	    len > vol->usable_leb_size || len & (ubi->min_io_size - 1))
 		return -EINVAL;
 
 	if (dtype != UBI_LONGTERM && dtype != UBI_SHORTTERM &&
diff --git a/drivers/mtd/ubi/misc.c b/drivers/mtd/ubi/misc.c
index 93e052812012..22ad31402945 100644
--- a/drivers/mtd/ubi/misc.c
+++ b/drivers/mtd/ubi/misc.c
@@ -37,7 +37,7 @@ int ubi_calc_data_len(const struct ubi_device *ubi, const void *buf,
 {
 	int i;
 
-	ubi_assert(length % ubi->min_io_size == 0);
+	ubi_assert(!(length & (ubi->min_io_size - 1)));
 
 	for (i = length - 1; i >= 0; i--)
 		if (((const uint8_t *)buf)[i] != 0xFF)
diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c
index 5be58d85c639..7402025ded94 100644
--- a/drivers/mtd/ubi/vmt.c
+++ b/drivers/mtd/ubi/vmt.c
@@ -727,7 +727,7 @@ static void paranoid_check_volume(struct ubi_device *ubi, int vol_id)
 		goto fail;
 	}
 
-	n = vol->alignment % ubi->min_io_size;
+	n = vol->alignment & (ubi->min_io_size - 1);
 	if (vol->alignment != 1 && n) {
 		ubi_err("alignment is not multiple of min I/O unit");
 		goto fail;
diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c
index 42a7815086b7..d9af11a8682b 100644
--- a/drivers/mtd/ubi/vtbl.c
+++ b/drivers/mtd/ubi/vtbl.c
@@ -170,7 +170,7 @@ static int vtbl_check(const struct ubi_device *ubi,
 			goto bad;
 		}
 
-		n = alignment % ubi->min_io_size;
+		n = alignment & (ubi->min_io_size - 1);
 		if (alignment != 1 && n) {
 			err = 5;
 			goto bad;
@@ -684,14 +684,13 @@ static int check_scanning_info(const struct ubi_device *ubi,
 		return -EINVAL;
 	}
 
-	if (si->highest_vol_id >= ubi->vtbl_slots + UBI_INT_VOL_COUNT&&
+	if (si->highest_vol_id >= ubi->vtbl_slots + UBI_INT_VOL_COUNT &&
 	    si->highest_vol_id < UBI_INTERNAL_VOL_START) {
 		ubi_err("too large volume ID %d found by scanning",
 			si->highest_vol_id);
 		return -EINVAL;
 	}
 
-
 	for (i = 0; i < ubi->vtbl_slots + UBI_INT_VOL_COUNT; i++) {
 		cond_resched();
 
diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c
index a471a491f0ab..cc8fe2934d2b 100644
--- a/drivers/mtd/ubi/wl.c
+++ b/drivers/mtd/ubi/wl.c
@@ -1368,7 +1368,7 @@ int ubi_thread(void *u)
 		int err;
 
 		if (kthread_should_stop())
-			goto out;
+			break;
 
 		if (try_to_freeze())
 			continue;
@@ -1403,7 +1403,6 @@ int ubi_thread(void *u)
 		cond_resched();
 	}
 
-out:
 	dbg_wl("background thread \"%s\" is killed", ubi->bgt_name);
 	return 0;
 }

From abc5e92262d87f9c5c628492bffc55f81c7dcb80 Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Wed, 4 Jun 2008 16:48:12 +0300
Subject: [PATCH 0174/1171] UBI: fix memory leak

ubi_free_volume() function sets ubi->volumes[] to NULL, so
ubi_eba_close() is useless, it does not free what has to be freed.
So zap it and free vol->eba_tbl at the volume release function.

Pointed-out-by: Adrian Hunter <ext-adrian.hunter@nokia.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/build.c |  2 --
 drivers/mtd/ubi/eba.c   | 17 -----------------
 drivers/mtd/ubi/ubi.h   |  1 -
 drivers/mtd/ubi/vmt.c   | 18 +++++++++---------
 4 files changed, 9 insertions(+), 29 deletions(-)

diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c
index ff4425de1527..7b42b4d05b3a 100644
--- a/drivers/mtd/ubi/build.c
+++ b/drivers/mtd/ubi/build.c
@@ -840,7 +840,6 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset)
 out_uif:
 	uif_close(ubi);
 out_detach:
-	ubi_eba_close(ubi);
 	ubi_wl_close(ubi);
 	vfree(ubi->vtbl);
 out_free:
@@ -903,7 +902,6 @@ int ubi_detach_mtd_dev(int ubi_num, int anyway)
 		kthread_stop(ubi->bgt_thread);
 
 	uif_close(ubi);
-	ubi_eba_close(ubi);
 	ubi_wl_close(ubi);
 	vfree(ubi->vtbl);
 	put_mtd_device(ubi->mtd);
diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index 37d778447943..623d25f4855f 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -1233,20 +1233,3 @@ out_free:
 	}
 	return err;
 }
-
-/**
- * ubi_eba_close - close EBA unit.
- * @ubi: UBI device description object
- */
-void ubi_eba_close(const struct ubi_device *ubi)
-{
-	int i, num_volumes = ubi->vtbl_slots + UBI_INT_VOL_COUNT;
-
-	dbg_eba("close EBA unit");
-
-	for (i = 0; i < num_volumes; i++) {
-		if (!ubi->volumes[i])
-			continue;
-		kfree(ubi->volumes[i]->eba_tbl);
-	}
-}
diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h
index 67dcbd11c15c..940f6b7deec3 100644
--- a/drivers/mtd/ubi/ubi.h
+++ b/drivers/mtd/ubi/ubi.h
@@ -477,7 +477,6 @@ int ubi_eba_atomic_leb_change(struct ubi_device *ubi, struct ubi_volume *vol,
 int ubi_eba_copy_leb(struct ubi_device *ubi, int from, int to,
 		     struct ubi_vid_hdr *vid_hdr);
 int ubi_eba_init_scan(struct ubi_device *ubi, struct ubi_scan_info *si);
-void ubi_eba_close(const struct ubi_device *ubi);
 
 /* wl.c */
 int ubi_wl_get_peb(struct ubi_device *ubi, int dtype);
diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c
index 7402025ded94..367b04176e0a 100644
--- a/drivers/mtd/ubi/vmt.c
+++ b/drivers/mtd/ubi/vmt.c
@@ -127,6 +127,7 @@ static void vol_release(struct device *dev)
 {
 	struct ubi_volume *vol = container_of(dev, struct ubi_volume, dev);
 
+	kfree(vol->eba_tbl);
 	kfree(vol);
 }
 
@@ -201,7 +202,7 @@ static void volume_sysfs_close(struct ubi_volume *vol)
  */
 int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req)
 {
-	int i, err, vol_id = req->vol_id, dont_free = 0;
+	int i, err, vol_id = req->vol_id, do_free = 1;
 	struct ubi_volume *vol;
 	struct ubi_vtbl_record vtbl_rec;
 	uint64_t bytes;
@@ -365,14 +366,14 @@ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req)
 
 out_sysfs:
 	/*
-	 * We have registered our device, we should not free the volume*
+	 * We have registered our device, we should not free the volume
 	 * description object in this function in case of an error - it is
 	 * freed by the release function.
 	 *
 	 * Get device reference to prevent the release function from being
 	 * called just after sysfs has been closed.
 	 */
-	dont_free = 1;
+	do_free = 0;
 	get_device(&vol->dev);
 	volume_sysfs_close(vol);
 out_gluebi:
@@ -382,17 +383,18 @@ out_gluebi:
 out_cdev:
 	cdev_del(&vol->cdev);
 out_mapping:
-	kfree(vol->eba_tbl);
+	if (do_free)
+		kfree(vol->eba_tbl);
 out_acc:
 	spin_lock(&ubi->volumes_lock);
 	ubi->rsvd_pebs -= vol->reserved_pebs;
 	ubi->avail_pebs += vol->reserved_pebs;
 out_unlock:
 	spin_unlock(&ubi->volumes_lock);
-	if (dont_free)
-		put_device(&vol->dev);
-	else
+	if (do_free)
 		kfree(vol);
+	else
+		put_device(&vol->dev);
 	ubi_err("cannot create volume %d, error %d", vol_id, err);
 	return err;
 }
@@ -445,8 +447,6 @@ int ubi_remove_volume(struct ubi_volume_desc *desc)
 			goto out_err;
 	}
 
-	kfree(vol->eba_tbl);
-	vol->eba_tbl = NULL;
 	cdev_del(&vol->cdev);
 	volume_sysfs_close(vol);
 

From 505d1caa79cd61a70615e9a7eae2eab85e797a83 Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Wed, 4 Jun 2008 17:00:35 +0300
Subject: [PATCH 0175/1171] UBI: do not forget to free internal volumes

UBI forgets to free internal volumes when detaching MTD device.
Fix this.

Pointed-out-by: Adrian Hunter <ext-adrian.hunter@nokia.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/build.c | 28 +++++++++++++++++++++++++---
 1 file changed, 25 insertions(+), 3 deletions(-)

diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c
index 7b42b4d05b3a..33205e4c1f5b 100644
--- a/drivers/mtd/ubi/build.c
+++ b/drivers/mtd/ubi/build.c
@@ -422,6 +422,10 @@ out_unreg:
 /**
  * uif_close - close user interfaces for an UBI device.
  * @ubi: UBI device description object
+ *
+ * Note, since this function un-registers UBI volume device objects (@vol->dev),
+ * the memory allocated voe the volumes is freed as well (in the release
+ * function).
  */
 static void uif_close(struct ubi_device *ubi)
 {
@@ -431,6 +435,21 @@ static void uif_close(struct ubi_device *ubi)
 	unregister_chrdev_region(ubi->cdev.dev, ubi->vtbl_slots + 1);
 }
 
+/**
+ * free_internal_volumes - free internal volumes.
+ * @ubi: UBI device description object
+ */
+static void free_internal_volumes(struct ubi_device *ubi)
+{
+	int i;
+
+	for (i = ubi->vtbl_slots;
+	     i < ubi->vtbl_slots + UBI_INT_VOL_COUNT; i++) {
+		kfree(ubi->volumes[i]->eba_tbl);
+		kfree(ubi->volumes[i]);
+	}
+}
+
 /**
  * attach_by_scanning - attach an MTD device using scanning method.
  * @ubi: UBI device descriptor
@@ -475,6 +494,7 @@ static int attach_by_scanning(struct ubi_device *ubi)
 out_wl:
 	ubi_wl_close(ubi);
 out_vtbl:
+	free_internal_volumes(ubi);
 	vfree(ubi->vtbl);
 out_si:
 	ubi_scan_destroy_si(si);
@@ -650,7 +670,7 @@ static int autoresize(struct ubi_device *ubi, int vol_id)
 
 	/*
 	 * Clear the auto-resize flag in the volume in-memory copy of the
-	 * volume table, and 'ubi_resize_volume()' will propogate this change
+	 * volume table, and 'ubi_resize_volume()' will propagate this change
 	 * to the flash.
 	 */
 	ubi->vtbl[vol_id].flags &= ~UBI_VTBL_AUTORESIZE_FLG;
@@ -659,7 +679,7 @@ static int autoresize(struct ubi_device *ubi, int vol_id)
 		struct ubi_vtbl_record vtbl_rec;
 
 		/*
-		 * No avalilable PEBs to re-size the volume, clear the flag on
+		 * No available PEBs to re-size the volume, clear the flag on
 		 * flash and exit.
 		 */
 		memcpy(&vtbl_rec, &ubi->vtbl[vol_id],
@@ -692,7 +712,7 @@ static int autoresize(struct ubi_device *ubi, int vol_id)
  *
  * This function attaches MTD device @mtd_dev to UBI and assign @ubi_num number
  * to the newly created UBI device, unless @ubi_num is %UBI_DEV_NUM_AUTO, in
- * which case this function finds a vacant device nubert and assings it
+ * which case this function finds a vacant device number and assigns it
  * automatically. Returns the new UBI device number in case of success and a
  * negative error code in case of failure.
  *
@@ -841,6 +861,7 @@ out_uif:
 	uif_close(ubi);
 out_detach:
 	ubi_wl_close(ubi);
+	free_internal_volumes(ubi);
 	vfree(ubi->vtbl);
 out_free:
 	vfree(ubi->peb_buf1);
@@ -903,6 +924,7 @@ int ubi_detach_mtd_dev(int ubi_num, int anyway)
 
 	uif_close(ubi);
 	ubi_wl_close(ubi);
+	free_internal_volumes(ubi);
 	vfree(ubi->vtbl);
 	put_mtd_device(ubi->mtd);
 	vfree(ubi->peb_buf1);

From 472018f73e7308a7f29b753ee8c742b6f45f103f Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Wed, 4 Jun 2008 17:58:37 +0300
Subject: [PATCH 0176/1171] UBI: fix memory leak on error path

Normally UBI volumes are freed in the release function of
the struct device object. However, on error path they may
have to be freed before the struct device objects have been
initialized.

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/build.c | 34 ++++++++++++++++++++++++++++++----
 1 file changed, 30 insertions(+), 4 deletions(-)

diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c
index 33205e4c1f5b..a5b19944eca8 100644
--- a/drivers/mtd/ubi/build.c
+++ b/drivers/mtd/ubi/build.c
@@ -354,16 +354,35 @@ static void kill_volumes(struct ubi_device *ubi)
 			ubi_free_volume(ubi, ubi->volumes[i]);
 }
 
+/**
+ * free_user_volumes - free all user volumes.
+ * @ubi: UBI device description object
+ *
+ * Normally the volumes are freed at the release function of the volume device
+ * objects. However, on error paths the volumes have to be freed before the
+ * device objects have been initialized.
+ */
+static void free_user_volumes(struct ubi_device *ubi)
+{
+	int i;
+
+	for (i = 0; i < ubi->vtbl_slots; i++)
+		if (ubi->volumes[i]) {
+			kfree(ubi->volumes[i]->eba_tbl);
+			kfree(ubi->volumes[i]);
+		}
+}
+
 /**
  * uif_init - initialize user interfaces for an UBI device.
  * @ubi: UBI device description object
  *
  * This function returns zero in case of success and a negative error code in
- * case of failure.
+ * case of failure. Note, this function destroys all volumes if it failes.
  */
 static int uif_init(struct ubi_device *ubi)
 {
-	int i, err;
+	int i, err, do_free = 0;
 	dev_t dev;
 
 	sprintf(ubi->ubi_name, UBI_NAME_STR "%d", ubi->ubi_num);
@@ -410,10 +429,13 @@ static int uif_init(struct ubi_device *ubi)
 
 out_volumes:
 	kill_volumes(ubi);
+	do_free = 0;
 out_sysfs:
 	ubi_sysfs_close(ubi);
 	cdev_del(&ubi->cdev);
 out_unreg:
+	if (do_free)
+		free_user_volumes(ubi);
 	unregister_chrdev_region(ubi->cdev.dev, ubi->vtbl_slots + 1);
 	ubi_err("cannot initialize UBI %s, error %d", ubi->ubi_name, err);
 	return err;
@@ -722,7 +744,7 @@ static int autoresize(struct ubi_device *ubi, int vol_id)
 int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset)
 {
 	struct ubi_device *ubi;
-	int i, err;
+	int i, err, do_free = 1;
 
 	/*
 	 * Check if we already have the same MTD device attached.
@@ -822,7 +844,7 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset)
 
 	err = uif_init(ubi);
 	if (err)
-		goto out_detach;
+		goto out_nofree;
 
 	ubi->bgt_thread = kthread_create(ubi_thread, ubi, ubi->bgt_name);
 	if (IS_ERR(ubi->bgt_thread)) {
@@ -859,8 +881,12 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset)
 
 out_uif:
 	uif_close(ubi);
+out_nofree:
+	do_free = 0;
 out_detach:
 	ubi_wl_close(ubi);
+	if (do_free)
+		free_user_volumes(ubi);
 	free_internal_volumes(ubi);
 	vfree(ubi->vtbl);
 out_free:

From 23add7455c42eef63f8719bd268328047d4aed69 Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Mon, 16 Jun 2008 13:35:23 +0300
Subject: [PATCH 0177/1171] UBI: fix LEB locking

leb_read_unlock() may be called simultaniously by several tasks.
The would race at the following code:

 up_read(&le->mutex);
 if (free)
         kfree(le);

And it is possible that one task frees 'le' before the other tasks
do 'up_read()'. Fix this by doing up_read and free inside the
'ubi->ltree' lock. Below it the oops we had because of this:

BUG: spinlock bad magic on CPU#0, integck/7504
BUG: unable to handle kernel paging request at 6b6b6c4f
IP: [<c0211221>] spin_bug+0x5c/0xdb
*pde = 00000000 Oops: 0000 [#1] PREEMPT SMP Modules linked in: ubifs ubi nandsim nand nand_ids nand_ecc video output

Pid: 7504, comm: integck Not tainted (2.6.26-rc3ubifs26 #8)
EIP: 0060:[<c0211221>] EFLAGS: 00010002 CPU: 0
EIP is at spin_bug+0x5c/0xdb
EAX: 00000032 EBX: 6b6b6b6b ECX: 6b6b6b6b EDX: f7f7ce30
ESI: f76491dc EDI: c044f51f EBP: e8a736cc ESP: e8a736a8
DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 0068
Process integck (pid: 7504, ti=e8a72000 task=f7f7ce30 task.ti=e8a72000)
Stack: c044f754 c044f51f 00000000 f7f7d024 00001d50 00000001 f76491dc 00000296       f6df50e0 e8a736d8 c02112f0 f76491dc e8a736e8 c039157a f7d9e830 f76491d8       e8a7370c c020b975 f76491dc 00000296 f76491f8 00000000 f76491d8 00000000 Call Trace:
[<c02112f0>] ? _raw_spin_unlock+0x50/0x7c
[<c039157a>] ? _spin_unlock_irqrestore+0x20/0x58
[<c020b975>] ? rwsem_wake+0x4b/0x122
[<c0390e0a>] ? call_rwsem_wake+0xa/0xc
[<c0139ee7>] ? up_read+0x28/0x31
[<f8873b3c>] ? leb_read_unlock+0x73/0x7b [ubi]
[<f88742a3>] ? ubi_eba_read_leb+0x195/0x2b0 [ubi]
[<f8872a04>] ? ubi_leb_read+0xaf/0xf8 [ubi]

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/eba.c | 27 +++++++--------------------
 1 file changed, 7 insertions(+), 20 deletions(-)

diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index 623d25f4855f..8dc488fc0cdf 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -223,22 +223,18 @@ static int leb_read_lock(struct ubi_device *ubi, int vol_id, int lnum)
  */
 static void leb_read_unlock(struct ubi_device *ubi, int vol_id, int lnum)
 {
-	int free = 0;
 	struct ubi_ltree_entry *le;
 
 	spin_lock(&ubi->ltree_lock);
 	le = ltree_lookup(ubi, vol_id, lnum);
 	le->users -= 1;
 	ubi_assert(le->users >= 0);
+	up_read(&le->mutex);
 	if (le->users == 0) {
 		rb_erase(&le->rb, &ubi->ltree);
-		free = 1;
+		kfree(le);
 	}
 	spin_unlock(&ubi->ltree_lock);
-
-	up_read(&le->mutex);
-	if (free)
-		kfree(le);
 }
 
 /**
@@ -274,7 +270,6 @@ static int leb_write_lock(struct ubi_device *ubi, int vol_id, int lnum)
  */
 static int leb_write_trylock(struct ubi_device *ubi, int vol_id, int lnum)
 {
-	int free;
 	struct ubi_ltree_entry *le;
 
 	le = ltree_add_entry(ubi, vol_id, lnum);
@@ -289,12 +284,9 @@ static int leb_write_trylock(struct ubi_device *ubi, int vol_id, int lnum)
 	ubi_assert(le->users >= 0);
 	if (le->users == 0) {
 		rb_erase(&le->rb, &ubi->ltree);
-		free = 1;
-	} else
-		free = 0;
-	spin_unlock(&ubi->ltree_lock);
-	if (free)
 		kfree(le);
+	}
+	spin_unlock(&ubi->ltree_lock);
 
 	return 1;
 }
@@ -307,23 +299,18 @@ static int leb_write_trylock(struct ubi_device *ubi, int vol_id, int lnum)
  */
 static void leb_write_unlock(struct ubi_device *ubi, int vol_id, int lnum)
 {
-	int free;
 	struct ubi_ltree_entry *le;
 
 	spin_lock(&ubi->ltree_lock);
 	le = ltree_lookup(ubi, vol_id, lnum);
 	le->users -= 1;
 	ubi_assert(le->users >= 0);
+	up_write(&le->mutex);
 	if (le->users == 0) {
 		rb_erase(&le->rb, &ubi->ltree);
-		free = 1;
-	} else
-		free = 0;
-	spin_unlock(&ubi->ltree_lock);
-
-	up_write(&le->mutex);
-	if (free)
 		kfree(le);
+	}
+	spin_unlock(&ubi->ltree_lock);
 }
 
 /**

From 73789a3d9fd8e500e121c1d4a5a2b16dd748ab5f Mon Sep 17 00:00:00 2001
From: Bruce Leonard <brucle@selinc.com>
Date: Thu, 3 Jul 2008 10:35:49 +0300
Subject: [PATCH 0178/1171] UBI: fix 64-bit calculations

Signed-off-by: Bruce Leonard <brucle@selinc.com>
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/cdev.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/mtd/ubi/cdev.c b/drivers/mtd/ubi/cdev.c
index 0cdaf9fba7b0..3e3449ec07f0 100644
--- a/drivers/mtd/ubi/cdev.c
+++ b/drivers/mtd/ubi/cdev.c
@@ -437,7 +437,8 @@ static int vol_cdev_ioctl(struct inode *inode, struct file *file,
 			break;
 		}
 
-		rsvd_bytes = vol->reserved_pebs * (ubi->leb_size-vol->data_pad);
+		rsvd_bytes = (long long)vol->reserved_pebs *
+					ubi->leb_size-vol->data_pad;
 		if (bytes < 0 || bytes > rsvd_bytes) {
 			err = -EINVAL;
 			break;

From a5bf6190417cbbf80443a9f71c65b653e13e9982 Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Thu, 10 Jul 2008 18:38:33 +0300
Subject: [PATCH 0179/1171] UBI: add ubi_sync() interface

To flush MTD device caches.

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/kapi.c  | 24 ++++++++++++++++++++++++
 include/linux/mtd/ubi.h |  1 +
 2 files changed, 25 insertions(+)

diff --git a/drivers/mtd/ubi/kapi.c b/drivers/mtd/ubi/kapi.c
index 51508832566d..e65c8e0bcd5d 100644
--- a/drivers/mtd/ubi/kapi.c
+++ b/drivers/mtd/ubi/kapi.c
@@ -632,3 +632,27 @@ int ubi_is_mapped(struct ubi_volume_desc *desc, int lnum)
 	return vol->eba_tbl[lnum] >= 0;
 }
 EXPORT_SYMBOL_GPL(ubi_is_mapped);
+
+/**
+ * ubi_sync - synchronize UBI device buffers.
+ * @ubi_num: UBI device to synchronize
+ *
+ * The underlying MTD device may cache data in hardware or in software. This
+ * function ensures the caches are flushed. Returns zero in case of success and
+ * a negative error code in case of failure.
+ */
+int ubi_sync(int ubi_num)
+{
+	struct ubi_device *ubi;
+
+	ubi = ubi_get_device(ubi_num);
+	if (!ubi)
+		return -ENODEV;
+
+	if (ubi->mtd->sync)
+		ubi->mtd->sync(ubi->mtd);
+
+	ubi_put_device(ubi);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ubi_sync);
diff --git a/include/linux/mtd/ubi.h b/include/linux/mtd/ubi.h
index f71201d0f3e7..83302bbbddb4 100644
--- a/include/linux/mtd/ubi.h
+++ b/include/linux/mtd/ubi.h
@@ -152,6 +152,7 @@ int ubi_leb_erase(struct ubi_volume_desc *desc, int lnum);
 int ubi_leb_unmap(struct ubi_volume_desc *desc, int lnum);
 int ubi_leb_map(struct ubi_volume_desc *desc, int lnum, int dtype);
 int ubi_is_mapped(struct ubi_volume_desc *desc, int lnum);
+int ubi_sync(int ubi_num);
 
 /*
  * This function is the same as the 'ubi_leb_read()' function, but it does not

From a6ea440769e11c46828cddd20f91ab57261701d5 Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Sun, 13 Jul 2008 21:46:24 +0300
Subject: [PATCH 0180/1171] UBI: improve mkvol request validation

Check that volume name is not shorter than 'name_len'.

No need to copy the trailing zero byte because whole array
was zeroed earlier.

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/cdev.c | 7 +++++--
 drivers/mtd/ubi/vmt.c  | 4 ++--
 2 files changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/mtd/ubi/cdev.c b/drivers/mtd/ubi/cdev.c
index 3e3449ec07f0..4fb84e3e6502 100644
--- a/drivers/mtd/ubi/cdev.c
+++ b/drivers/mtd/ubi/cdev.c
@@ -574,6 +574,10 @@ static int verify_mkvol_req(const struct ubi_device *ubi,
 		goto bad;
 	}
 
+	n = strnlen(req->name, req->name_len + 1);
+	if (n != req->name_len)
+		goto bad;
+
 	return 0;
 
 bad:
@@ -629,12 +633,11 @@ static int ubi_cdev_ioctl(struct inode *inode, struct file *file,
 			break;
 		}
 
+		req.name[req.name_len] = '\0';
 		err = verify_mkvol_req(ubi, &req);
 		if (err)
 			break;
 
-		req.name[req.name_len] = '\0';
-
 		mutex_lock(&ubi->volumes_mutex);
 		err = ubi_create_volume(ubi, &req);
 		mutex_unlock(&ubi->volumes_mutex);
diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c
index 367b04176e0a..bfa7c5d2e064 100644
--- a/drivers/mtd/ubi/vmt.c
+++ b/drivers/mtd/ubi/vmt.c
@@ -275,7 +275,7 @@ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req)
 	vol->data_pad  = ubi->leb_size % vol->alignment;
 	vol->vol_type  = req->vol_type;
 	vol->name_len  = req->name_len;
-	memcpy(vol->name, req->name, vol->name_len + 1);
+	memcpy(vol->name, req->name, vol->name_len);
 	vol->ubi = ubi;
 
 	/*
@@ -350,7 +350,7 @@ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req)
 		vtbl_rec.vol_type = UBI_VID_DYNAMIC;
 	else
 		vtbl_rec.vol_type = UBI_VID_STATIC;
-	memcpy(vtbl_rec.name, vol->name, vol->name_len + 1);
+	memcpy(vtbl_rec.name, vol->name, vol->name_len);
 
 	err = ubi_change_vtbl_record(ubi, vol_id, &vtbl_rec);
 	if (err)

From bb84c1a199558962edf4b4aeb4480fb09aa09b91 Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Mon, 14 Jul 2008 12:57:27 +0300
Subject: [PATCH 0181/1171] UBI: fix error message

The ubi_err() macro will add \n.

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/gluebi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/ubi/gluebi.c b/drivers/mtd/ubi/gluebi.c
index e909b390069a..ae76ab638b2f 100644
--- a/drivers/mtd/ubi/gluebi.c
+++ b/drivers/mtd/ubi/gluebi.c
@@ -299,7 +299,7 @@ int ubi_create_gluebi(struct ubi_device *ubi, struct ubi_volume *vol)
 		mtd->size = vol->used_bytes;
 
 	if (add_mtd_device(mtd)) {
-		ubi_err("cannot not add MTD device\n");
+		ubi_err("cannot not add MTD device");
 		kfree(mtd->name);
 		return -ENFILE;
 	}

From 85c6e6e28259e9b58b8984db536c45bc3161f40c Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Wed, 16 Jul 2008 10:25:56 +0300
Subject: [PATCH 0182/1171] UBI: amend commentaries

Hch asked not to use "unit" for sub-systems, let it be so.
Also some other commentaries modifications.

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/build.c     |  2 +-
 drivers/mtd/ubi/debug.h     |  6 +--
 drivers/mtd/ubi/eba.c       | 22 ++++-----
 drivers/mtd/ubi/io.c        | 22 ++++-----
 drivers/mtd/ubi/scan.c      | 28 ++++++-----
 drivers/mtd/ubi/scan.h      | 19 ++++----
 drivers/mtd/ubi/ubi-media.h | 23 ++++-----
 drivers/mtd/ubi/ubi.h       | 37 +++++++--------
 drivers/mtd/ubi/wl.c        | 94 ++++++++++++++++++-------------------
 include/linux/mtd/ubi.h     |  4 +-
 10 files changed, 129 insertions(+), 128 deletions(-)

diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c
index a5b19944eca8..27271fe32e02 100644
--- a/drivers/mtd/ubi/build.c
+++ b/drivers/mtd/ubi/build.c
@@ -524,7 +524,7 @@ out_si:
 }
 
 /**
- * io_init - initialize I/O unit for a given UBI device.
+ * io_init - initialize I/O sub-system for a given UBI device.
  * @ubi: UBI device description object
  *
  * If @ubi->vid_hdr_offset or @ubi->leb_start is zero, default offsets are
diff --git a/drivers/mtd/ubi/debug.h b/drivers/mtd/ubi/debug.h
index 8ea99d8c9e1f..7d8d77c31dfe 100644
--- a/drivers/mtd/ubi/debug.h
+++ b/drivers/mtd/ubi/debug.h
@@ -76,21 +76,21 @@ void ubi_dbg_dump_mkvol_req(const struct ubi_mkvol_req *req);
 #endif /* CONFIG_MTD_UBI_DEBUG_MSG */
 
 #ifdef CONFIG_MTD_UBI_DEBUG_MSG_EBA
-/* Messages from the eraseblock association unit */
+/* Messages from the eraseblock association sub-system */
 #define dbg_eba(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__)
 #else
 #define dbg_eba(fmt, ...) ({})
 #endif
 
 #ifdef CONFIG_MTD_UBI_DEBUG_MSG_WL
-/* Messages from the wear-leveling unit */
+/* Messages from the wear-leveling sub-system */
 #define dbg_wl(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__)
 #else
 #define dbg_wl(fmt, ...) ({})
 #endif
 
 #ifdef CONFIG_MTD_UBI_DEBUG_MSG_IO
-/* Messages from the input/output unit */
+/* Messages from the input/output sub-system */
 #define dbg_io(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__)
 #else
 #define dbg_io(fmt, ...) ({})
diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index 8dc488fc0cdf..613cd1e51648 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -19,20 +19,20 @@
  */
 
 /*
- * The UBI Eraseblock Association (EBA) unit.
+ * The UBI Eraseblock Association (EBA) sub-system.
  *
- * This unit is responsible for I/O to/from logical eraseblock.
+ * This sub-system is responsible for I/O to/from logical eraseblock.
  *
  * Although in this implementation the EBA table is fully kept and managed in
  * RAM, which assumes poor scalability, it might be (partially) maintained on
  * flash in future implementations.
  *
- * The EBA unit implements per-logical eraseblock locking. Before accessing a
- * logical eraseblock it is locked for reading or writing. The per-logical
- * eraseblock locking is implemented by means of the lock tree. The lock tree
- * is an RB-tree which refers all the currently locked logical eraseblocks. The
- * lock tree elements are &struct ubi_ltree_entry objects. They are indexed by
- * (@vol_id, @lnum) pairs.
+ * The EBA sub-system implements per-logical eraseblock locking. Before
+ * accessing a logical eraseblock it is locked for reading or writing. The
+ * per-logical eraseblock locking is implemented by means of the lock tree. The
+ * lock tree is an RB-tree which refers all the currently locked logical
+ * eraseblocks. The lock tree elements are &struct ubi_ltree_entry objects.
+ * They are indexed by (@vol_id, @lnum) pairs.
  *
  * EBA also maintains the global sequence counter which is incremented each
  * time a logical eraseblock is mapped to a physical eraseblock and it is
@@ -1128,7 +1128,7 @@ out_unlock_leb:
 }
 
 /**
- * ubi_eba_init_scan - initialize the EBA unit using scanning information.
+ * ubi_eba_init_scan - initialize the EBA sub-system using scanning information.
  * @ubi: UBI device description object
  * @si: scanning information
  *
@@ -1143,7 +1143,7 @@ int ubi_eba_init_scan(struct ubi_device *ubi, struct ubi_scan_info *si)
 	struct ubi_scan_leb *seb;
 	struct rb_node *rb;
 
-	dbg_eba("initialize EBA unit");
+	dbg_eba("initialize EBA sub-system");
 
 	spin_lock_init(&ubi->ltree_lock);
 	mutex_init(&ubi->alc_mutex);
@@ -1209,7 +1209,7 @@ int ubi_eba_init_scan(struct ubi_device *ubi, struct ubi_scan_info *si)
 		ubi->rsvd_pebs  += ubi->beb_rsvd_pebs;
 	}
 
-	dbg_eba("EBA unit is initialized");
+	dbg_eba("EBA sub-system is initialized");
 	return 0;
 
 out_free:
diff --git a/drivers/mtd/ubi/io.c b/drivers/mtd/ubi/io.c
index 4ac11df7b048..561e7b2f96cb 100644
--- a/drivers/mtd/ubi/io.c
+++ b/drivers/mtd/ubi/io.c
@@ -20,15 +20,15 @@
  */
 
 /*
- * UBI input/output unit.
+ * UBI input/output sub-system.
  *
- * This unit provides a uniform way to work with all kinds of the underlying
- * MTD devices. It also implements handy functions for reading and writing UBI
- * headers.
+ * This sub-system provides a uniform way to work with all kinds of the
+ * underlying MTD devices. It also implements handy functions for reading and
+ * writing UBI headers.
  *
  * We are trying to have a paranoid mindset and not to trust to what we read
- * from the flash media in order to be more secure and robust. So this unit
- * validates every single header it reads from the flash media.
+ * from the flash media in order to be more secure and robust. So this
+ * sub-system validates every single header it reads from the flash media.
  *
  * Some words about how the eraseblock headers are stored.
  *
@@ -79,11 +79,11 @@
  * 512-byte chunks, we have to allocate one more buffer and copy our VID header
  * to offset 448 of this buffer.
  *
- * The I/O unit does the following trick in order to avoid this extra copy.
- * It always allocates a @ubi->vid_hdr_alsize bytes buffer for the VID header
- * and returns a pointer to offset @ubi->vid_hdr_shift of this buffer. When the
- * VID header is being written out, it shifts the VID header pointer back and
- * writes the whole sub-page.
+ * The I/O sub-system does the following trick in order to avoid this extra
+ * copy. It always allocates a @ubi->vid_hdr_alsize bytes buffer for the VID
+ * header and returns a pointer to offset @ubi->vid_hdr_shift of this buffer.
+ * When the VID header is being written out, it shifts the VID header pointer
+ * back and writes the whole sub-page.
  */
 
 #include <linux/crc32.h>
diff --git a/drivers/mtd/ubi/scan.c b/drivers/mtd/ubi/scan.c
index 96d410e106ab..892c2ba49777 100644
--- a/drivers/mtd/ubi/scan.c
+++ b/drivers/mtd/ubi/scan.c
@@ -19,9 +19,9 @@
  */
 
 /*
- * UBI scanning unit.
+ * UBI scanning sub-system.
  *
- * This unit is responsible for scanning the flash media, checking UBI
+ * This sub-system is responsible for scanning the flash media, checking UBI
  * headers and providing complete information about the UBI flash image.
  *
  * The scanning information is represented by a &struct ubi_scan_info' object.
@@ -103,7 +103,7 @@ static int add_to_list(struct ubi_scan_info *si, int pnum, int ec,
  * non-zero if an inconsistency was found and zero if not.
  *
  * Note, UBI does sanity check of everything it reads from the flash media.
- * Most of the checks are done in the I/O unit. Here we check that the
+ * Most of the checks are done in the I/O sub-system. Here we check that the
  * information in the VID header is consistent to the information in other VID
  * headers of the same volume.
  */
@@ -256,8 +256,8 @@ static int compare_lebs(struct ubi_device *ubi, const struct ubi_scan_leb *seb,
 		 * that versions that are close to %0xFFFFFFFF are less then
 		 * versions that are close to %0.
 		 *
-		 * The UBI WL unit guarantees that the number of pending tasks
-		 * is not greater then %0x7FFFFFFF. So, if the difference
+		 * The UBI WL sub-system guarantees that the number of pending
+		 * tasks is not greater then %0x7FFFFFFF. So, if the difference
 		 * between any two versions is greater or equivalent to
 		 * %0x7FFFFFFF, there was an overflow and the logical
 		 * eraseblock with lower version is actually newer then the one
@@ -645,9 +645,9 @@ void ubi_scan_rm_volume(struct ubi_scan_info *si, struct ubi_scan_volume *sv)
  *
  * This function erases physical eraseblock 'pnum', and writes the erase
  * counter header to it. This function should only be used on UBI device
- * initialization stages, when the EBA unit had not been yet initialized. This
- * function returns zero in case of success and a negative error code in case
- * of failure.
+ * initialization stages, when the EBA sub-system had not been yet initialized.
+ * This function returns zero in case of success and a negative error code in
+ * case of failure.
  */
 int ubi_scan_erase_peb(struct ubi_device *ubi, const struct ubi_scan_info *si,
 		       int pnum, int ec)
@@ -687,9 +687,10 @@ out_free:
  * @si: scanning information
  *
  * This function returns a free physical eraseblock. It is supposed to be
- * called on the UBI initialization stages when the wear-leveling unit is not
- * initialized yet. This function picks a physical eraseblocks from one of the
- * lists, writes the EC header if it is needed, and removes it from the list.
+ * called on the UBI initialization stages when the wear-leveling sub-system is
+ * not initialized yet. This function picks a physical eraseblocks from one of
+ * the lists, writes the EC header if it is needed, and removes it from the
+ * list.
  *
  * This function returns scanning physical eraseblock information in case of
  * success and an error code in case of failure.
@@ -764,8 +765,9 @@ static int process_eb(struct ubi_device *ubi, struct ubi_scan_info *si, int pnum
 		return err;
 	else if (err) {
 		/*
-		 * FIXME: this is actually duty of the I/O unit to initialize
-		 * this, but MTD does not provide enough information.
+		 * FIXME: this is actually duty of the I/O sub-system to
+		 * initialize this, but MTD does not provide enough
+		 * information.
 		 */
 		si->bad_peb_count += 1;
 		return 0;
diff --git a/drivers/mtd/ubi/scan.h b/drivers/mtd/ubi/scan.h
index 966b9b682a42..4e2e3cc0becd 100644
--- a/drivers/mtd/ubi/scan.h
+++ b/drivers/mtd/ubi/scan.h
@@ -59,16 +59,16 @@ struct ubi_scan_leb {
  * @leb_count: number of logical eraseblocks in this volume
  * @vol_type: volume type
  * @used_ebs: number of used logical eraseblocks in this volume (only for
- * static volumes)
+ *            static volumes)
  * @last_data_size: amount of data in the last logical eraseblock of this
- * volume (always equivalent to the usable logical eraseblock size in case of
- * dynamic volumes)
+ *                  volume (always equivalent to the usable logical eraseblock
+ *                  size in case of dynamic volumes)
  * @data_pad: how many bytes at the end of logical eraseblocks of this volume
- * are not used (due to volume alignment)
+ *            are not used (due to volume alignment)
  * @compat: compatibility flags of this volume
  * @rb: link in the volume RB-tree
  * @root: root of the RB-tree containing all the eraseblock belonging to this
- * volume (&struct ubi_scan_leb objects)
+ *        volume (&struct ubi_scan_leb objects)
  *
  * One object of this type is allocated for each volume during scanning.
  */
@@ -92,8 +92,8 @@ struct ubi_scan_volume {
  * @free: list of free physical eraseblocks
  * @erase: list of physical eraseblocks which have to be erased
  * @alien: list of physical eraseblocks which should not be used by UBI (e.g.,
+ *         those belonging to "preserve"-compatible internal volumes)
  * @bad_peb_count: count of bad physical eraseblocks
- * those belonging to "preserve"-compatible internal volumes)
  * @vols_found: number of volumes found during scanning
  * @highest_vol_id: highest volume ID
  * @alien_peb_count: count of physical eraseblocks in the @alien list
@@ -106,8 +106,8 @@ struct ubi_scan_volume {
  * @ec_count: a temporary variable used when calculating @mean_ec
  *
  * This data structure contains the result of scanning and may be used by other
- * UBI units to build final UBI data structures, further error-recovery and so
- * on.
+ * UBI sub-systems to build final UBI data structures, further error-recovery
+ * and so on.
  */
 struct ubi_scan_info {
 	struct rb_root volumes;
@@ -132,8 +132,7 @@ struct ubi_device;
 struct ubi_vid_hdr;
 
 /*
- * ubi_scan_move_to_list - move a physical eraseblock from the volume tree to a
- * list.
+ * ubi_scan_move_to_list - move a PEB from the volume tree to a list.
  *
  * @sv: volume scanning information
  * @seb: scanning eraseblock infprmation
diff --git a/drivers/mtd/ubi/ubi-media.h b/drivers/mtd/ubi/ubi-media.h
index c3185d9fd048..26bb7af9787a 100644
--- a/drivers/mtd/ubi/ubi-media.h
+++ b/drivers/mtd/ubi/ubi-media.h
@@ -98,10 +98,11 @@ enum {
  * Compatibility constants used by internal volumes.
  *
  * @UBI_COMPAT_DELETE: delete this internal volume before anything is written
- * to the flash
+ *                     to the flash
  * @UBI_COMPAT_RO: attach this device in read-only mode
  * @UBI_COMPAT_PRESERVE: preserve this internal volume - do not touch its
- * physical eraseblocks, don't allow the wear-leveling unit to move them
+ *                       physical eraseblocks, don't allow the wear-leveling
+ *                       sub-system to move them
  * @UBI_COMPAT_REJECT: reject this UBI image
  */
 enum {
@@ -123,7 +124,7 @@ enum {
  * struct ubi_ec_hdr - UBI erase counter header.
  * @magic: erase counter header magic number (%UBI_EC_HDR_MAGIC)
  * @version: version of UBI implementation which is supposed to accept this
- * UBI image
+ *           UBI image
  * @padding1: reserved for future, zeroes
  * @ec: the erase counter
  * @vid_hdr_offset: where the VID header starts
@@ -159,20 +160,20 @@ struct ubi_ec_hdr {
  * struct ubi_vid_hdr - on-flash UBI volume identifier header.
  * @magic: volume identifier header magic number (%UBI_VID_HDR_MAGIC)
  * @version: UBI implementation version which is supposed to accept this UBI
- * image (%UBI_VERSION)
+ *           image (%UBI_VERSION)
  * @vol_type: volume type (%UBI_VID_DYNAMIC or %UBI_VID_STATIC)
  * @copy_flag: if this logical eraseblock was copied from another physical
- * eraseblock (for wear-leveling reasons)
+ *             eraseblock (for wear-leveling reasons)
  * @compat: compatibility of this volume (%0, %UBI_COMPAT_DELETE,
- * %UBI_COMPAT_IGNORE, %UBI_COMPAT_PRESERVE, or %UBI_COMPAT_REJECT)
+ *          %UBI_COMPAT_IGNORE, %UBI_COMPAT_PRESERVE, or %UBI_COMPAT_REJECT)
  * @vol_id: ID of this volume
  * @lnum: logical eraseblock number
  * @leb_ver: version of this logical eraseblock (IMPORTANT: obsolete, to be
- * removed, kept only for not breaking older UBI users)
+ *           removed, kept only for not breaking older UBI users)
  * @data_size: how many bytes of data this logical eraseblock contains
  * @used_ebs: total number of used logical eraseblocks in this volume
  * @data_pad: how many bytes at the end of this physical eraseblock are not
- * used
+ *            used
  * @data_crc: CRC checksum of the data stored in this logical eraseblock
  * @padding1: reserved for future, zeroes
  * @sqnum: sequence number
@@ -248,9 +249,9 @@ struct ubi_ec_hdr {
  * The @data_crc field contains the CRC checksum of the contents of the logical
  * eraseblock if this is a static volume. In case of dynamic volumes, it does
  * not contain the CRC checksum as a rule. The only exception is when the
- * data of the physical eraseblock was moved by the wear-leveling unit, then
- * the wear-leveling unit calculates the data CRC and stores it in the
- * @data_crc field. And of course, the @copy_flag is %in this case.
+ * data of the physical eraseblock was moved by the wear-leveling sub-system,
+ * then the wear-leveling sub-system calculates the data CRC and stores it in
+ * the @data_crc field. And of course, the @copy_flag is %in this case.
  *
  * The @data_size field is used only for static volumes because UBI has to know
  * how many bytes of data are stored in this eraseblock. For dynamic volumes,
diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h
index 940f6b7deec3..1fc32c863b78 100644
--- a/drivers/mtd/ubi/ubi.h
+++ b/drivers/mtd/ubi/ubi.h
@@ -74,15 +74,15 @@
 #define UBI_IO_RETRIES 3
 
 /*
- * Error codes returned by the I/O unit.
+ * Error codes returned by the I/O sub-system.
  *
  * UBI_IO_PEB_EMPTY: the physical eraseblock is empty, i.e. it contains only
- * 0xFF bytes
+ *                   %0xFF bytes
  * UBI_IO_PEB_FREE: the physical eraseblock is free, i.e. it contains only a
- * valid erase counter header, and the rest are %0xFF bytes
+ *                  valid erase counter header, and the rest are %0xFF bytes
  * UBI_IO_BAD_EC_HDR: the erase counter header is corrupted (bad magic or CRC)
  * UBI_IO_BAD_VID_HDR: the volume identifier header is corrupted (bad magic or
- * CRC)
+ *                     CRC)
  * UBI_IO_BITFLIPS: bit-flips were detected and corrected
  */
 enum {
@@ -99,9 +99,9 @@ enum {
  * @ec: erase counter
  * @pnum: physical eraseblock number
  *
- * This data structure is used in the WL unit. Each physical eraseblock has a
- * corresponding &struct wl_entry object which may be kept in different
- * RB-trees. See WL unit for details.
+ * This data structure is used in the WL sub-system. Each physical eraseblock
+ * has a corresponding &struct wl_entry object which may be kept in different
+ * RB-trees. See WL sub-system for details.
  */
 struct ubi_wl_entry {
 	struct rb_node rb;
@@ -118,10 +118,10 @@ struct ubi_wl_entry {
  * @mutex: read/write mutex to implement read/write access serialization to
  *         the (@vol_id, @lnum) logical eraseblock
  *
- * This data structure is used in the EBA unit to implement per-LEB locking.
- * When a logical eraseblock is being locked - corresponding
+ * This data structure is used in the EBA sub-system to implement per-LEB
+ * locking. When a logical eraseblock is being locked - corresponding
  * &struct ubi_ltree_entry object is inserted to the lock tree (@ubi->ltree).
- * See EBA unit for details.
+ * See EBA sub-system for details.
  */
 struct ubi_ltree_entry {
 	struct rb_node rb;
@@ -225,7 +225,7 @@ struct ubi_volume {
 #ifdef CONFIG_MTD_UBI_GLUEBI
 	/*
 	 * Gluebi-related stuff may be compiled out.
-	 * TODO: this should not be built into UBI but should be a separate
+	 * Note: this should not be built into UBI but should be a separate
 	 * ubimtd driver which works on top of UBI and emulates MTD devices.
 	 */
 	struct ubi_volume_desc *gluebi_desc;
@@ -235,8 +235,7 @@ struct ubi_volume {
 };
 
 /**
- * struct ubi_volume_desc - descriptor of the UBI volume returned when it is
- * opened.
+ * struct ubi_volume_desc - UBI volume descriptor returned when it is opened.
  * @vol: reference to the corresponding volume description object
  * @mode: open mode (%UBI_READONLY, %UBI_READWRITE, or %UBI_EXCLUSIVE)
  */
@@ -316,11 +315,11 @@ struct ubi_wl_entry;
  * @ro_mode: if the UBI device is in read-only mode
  * @leb_size: logical eraseblock size
  * @leb_start: starting offset of logical eraseblocks within physical
- * eraseblocks
+ *             eraseblocks
  * @ec_hdr_alsize: size of the EC header aligned to @hdrs_min_io_size
  * @vid_hdr_alsize: size of the VID header aligned to @hdrs_min_io_size
  * @vid_hdr_offset: starting offset of the volume identifier header (might be
- * unaligned)
+ *                  unaligned)
  * @vid_hdr_aloffset: starting offset of the VID header aligned to
  * @hdrs_min_io_size
  * @vid_hdr_shift: contains @vid_hdr_offset - @vid_hdr_aloffset
@@ -356,16 +355,16 @@ struct ubi_device {
 	struct mutex volumes_mutex;
 
 	int max_ec;
-	/* TODO: mean_ec is not updated run-time, fix */
+	/* Note, mean_ec is not updated run-time - should be fixed */
 	int mean_ec;
 
-	/* EBA unit's stuff */
+	/* EBA sub-system's stuff */
 	unsigned long long global_sqnum;
 	spinlock_t ltree_lock;
 	struct rb_root ltree;
 	struct mutex alc_mutex;
 
-	/* Wear-leveling unit's stuff */
+	/* Wear-leveling sub-system's stuff */
 	struct rb_root used;
 	struct rb_root free;
 	struct rb_root scrub;
@@ -388,7 +387,7 @@ struct ubi_device {
 	int thread_enabled;
 	char bgt_name[sizeof(UBI_BGT_NAME_PATTERN)+2];
 
-	/* I/O unit's stuff */
+	/* I/O sub-system's stuff */
 	long long flash_size;
 	int peb_count;
 	int peb_size;
diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c
index cc8fe2934d2b..761952ba125b 100644
--- a/drivers/mtd/ubi/wl.c
+++ b/drivers/mtd/ubi/wl.c
@@ -19,22 +19,22 @@
  */
 
 /*
- * UBI wear-leveling unit.
+ * UBI wear-leveling sub-system.
  *
- * This unit is responsible for wear-leveling. It works in terms of physical
- * eraseblocks and erase counters and knows nothing about logical eraseblocks,
- * volumes, etc. From this unit's perspective all physical eraseblocks are of
- * two types - used and free. Used physical eraseblocks are those that were
- * "get" by the 'ubi_wl_get_peb()' function, and free physical eraseblocks are
- * those that were put by the 'ubi_wl_put_peb()' function.
+ * This sub-system is responsible for wear-leveling. It works in terms of
+ * physical* eraseblocks and erase counters and knows nothing about logical
+ * eraseblocks, volumes, etc. From this sub-system's perspective all physical
+ * eraseblocks are of two types - used and free. Used physical eraseblocks are
+ * those that were "get" by the 'ubi_wl_get_peb()' function, and free physical
+ * eraseblocks are those that were put by the 'ubi_wl_put_peb()' function.
  *
  * Physical eraseblocks returned by 'ubi_wl_get_peb()' have only erase counter
- * header. The rest of the physical eraseblock contains only 0xFF bytes.
+ * header. The rest of the physical eraseblock contains only %0xFF bytes.
  *
- * When physical eraseblocks are returned to the WL unit by means of the
+ * When physical eraseblocks are returned to the WL sub-system by means of the
  * 'ubi_wl_put_peb()' function, they are scheduled for erasure. The erasure is
  * done asynchronously in context of the per-UBI device background thread,
- * which is also managed by the WL unit.
+ * which is also managed by the WL sub-system.
  *
  * The wear-leveling is ensured by means of moving the contents of used
  * physical eraseblocks with low erase counter to free physical eraseblocks
@@ -43,34 +43,36 @@
  * The 'ubi_wl_get_peb()' function accepts data type hints which help to pick
  * an "optimal" physical eraseblock. For example, when it is known that the
  * physical eraseblock will be "put" soon because it contains short-term data,
- * the WL unit may pick a free physical eraseblock with low erase counter, and
- * so forth.
+ * the WL sub-system may pick a free physical eraseblock with low erase
+ * counter, and so forth.
  *
- * If the WL unit fails to erase a physical eraseblock, it marks it as bad.
+ * If the WL sub-system fails to erase a physical eraseblock, it marks it as
+ * bad.
  *
- * This unit is also responsible for scrubbing. If a bit-flip is detected in a
- * physical eraseblock, it has to be moved. Technically this is the same as
- * moving it for wear-leveling reasons.
+ * This sub-system is also responsible for scrubbing. If a bit-flip is detected
+ * in a physical eraseblock, it has to be moved. Technically this is the same
+ * as moving it for wear-leveling reasons.
  *
- * As it was said, for the UBI unit all physical eraseblocks are either "free"
- * or "used". Free eraseblock are kept in the @wl->free RB-tree, while used
- * eraseblocks are kept in a set of different RB-trees: @wl->used,
+ * As it was said, for the UBI sub-system all physical eraseblocks are either
+ * "free" or "used". Free eraseblock are kept in the @wl->free RB-tree, while
+ * used eraseblocks are kept in a set of different RB-trees: @wl->used,
  * @wl->prot.pnum, @wl->prot.aec, and @wl->scrub.
  *
  * Note, in this implementation, we keep a small in-RAM object for each physical
  * eraseblock. This is surely not a scalable solution. But it appears to be good
  * enough for moderately large flashes and it is simple. In future, one may
- * re-work this unit and make it more scalable.
+ * re-work this sub-system and make it more scalable.
  *
- * At the moment this unit does not utilize the sequence number, which was
- * introduced relatively recently. But it would be wise to do this because the
- * sequence number of a logical eraseblock characterizes how old is it. For
+ * At the moment this sub-system does not utilize the sequence number, which
+ * was introduced relatively recently. But it would be wise to do this because
+ * the sequence number of a logical eraseblock characterizes how old is it. For
  * example, when we move a PEB with low erase counter, and we need to pick the
  * target PEB, we pick a PEB with the highest EC if our PEB is "old" and we
  * pick target PEB with an average EC if our PEB is not very "old". This is a
- * room for future re-works of the WL unit.
+ * room for future re-works of the WL sub-system.
  *
- * FIXME: looks too complex, should be simplified (later).
+ * Note: the stuff with protection trees looks too complex and is difficult to
+ * understand. Should be fixed.
  */
 
 #include <linux/slab.h>
@@ -92,20 +94,21 @@
 
 /*
  * Maximum difference between two erase counters. If this threshold is
- * exceeded, the WL unit starts moving data from used physical eraseblocks with
- * low erase counter to free physical eraseblocks with high erase counter.
+ * exceeded, the WL sub-system starts moving data from used physical
+ * eraseblocks with low erase counter to free physical eraseblocks with high
+ * erase counter.
  */
 #define UBI_WL_THRESHOLD CONFIG_MTD_UBI_WL_THRESHOLD
 
 /*
- * When a physical eraseblock is moved, the WL unit has to pick the target
+ * When a physical eraseblock is moved, the WL sub-system has to pick the target
  * physical eraseblock to move to. The simplest way would be just to pick the
  * one with the highest erase counter. But in certain workloads this could lead
  * to an unlimited wear of one or few physical eraseblock. Indeed, imagine a
  * situation when the picked physical eraseblock is constantly erased after the
  * data is written to it. So, we have a constant which limits the highest erase
- * counter of the free physical eraseblock to pick. Namely, the WL unit does
- * not pick eraseblocks with erase counter greater then the lowest erase
+ * counter of the free physical eraseblock to pick. Namely, the WL sub-system
+ * does not pick eraseblocks with erase counter greater then the lowest erase
  * counter plus %WL_FREE_MAX_DIFF.
  */
 #define WL_FREE_MAX_DIFF (2*UBI_WL_THRESHOLD)
@@ -123,11 +126,11 @@
  * @abs_ec: the absolute erase counter value when the protection ends
  * @e: the wear-leveling entry of the physical eraseblock under protection
  *
- * When the WL unit returns a physical eraseblock, the physical eraseblock is
- * protected from being moved for some "time". For this reason, the physical
- * eraseblock is not directly moved from the @wl->free tree to the @wl->used
- * tree. There is one more tree in between where this physical eraseblock is
- * temporarily stored (@wl->prot).
+ * When the WL sub-system returns a physical eraseblock, the physical
+ * eraseblock is protected from being moved for some "time". For this reason,
+ * the physical eraseblock is not directly moved from the @wl->free tree to the
+ * @wl->used tree. There is one more tree in between where this physical
+ * eraseblock is temporarily stored (@wl->prot).
  *
  * All this protection stuff is needed because:
  *  o we don't want to move physical eraseblocks just after we have given them
@@ -175,7 +178,6 @@ struct ubi_wl_prot_entry {
  * @list: a link in the list of pending works
  * @func: worker function
  * @priv: private data of the worker function
- *
  * @e: physical eraseblock to erase
  * @torture: if the physical eraseblock has to be tortured
  *
@@ -1136,7 +1138,7 @@ out_ro:
 }
 
 /**
- * ubi_wl_put_peb - return a physical eraseblock to the wear-leveling unit.
+ * ubi_wl_put_peb - return a PEB to the wear-leveling sub-system.
  * @ubi: UBI device description object
  * @pnum: physical eraseblock to return
  * @torture: if this physical eraseblock has to be tortured
@@ -1175,11 +1177,11 @@ retry:
 		/*
 		 * User is putting the physical eraseblock which was selected
 		 * as the target the data is moved to. It may happen if the EBA
-		 * unit already re-mapped the LEB in 'ubi_eba_copy_leb()' but
-		 * the WL unit has not put the PEB to the "used" tree yet, but
-		 * it is about to do this. So we just set a flag which will
-		 * tell the WL worker that the PEB is not needed anymore and
-		 * should be scheduled for erasure.
+		 * sub-system already re-mapped the LEB in 'ubi_eba_copy_leb()'
+		 * but the WL sub-system has not put the PEB to the "used" tree
+		 * yet, but it is about to do this. So we just set a flag which
+		 * will tell the WL worker that the PEB is not needed anymore
+		 * and should be scheduled for erasure.
 		 */
 		dbg_wl("PEB %d is the target of data moving", pnum);
 		ubi_assert(!ubi->move_to_put);
@@ -1425,8 +1427,7 @@ static void cancel_pending(struct ubi_device *ubi)
 }
 
 /**
- * ubi_wl_init_scan - initialize the wear-leveling unit using scanning
- * information.
+ * ubi_wl_init_scan - initialize the WL sub-system using scanning information.
  * @ubi: UBI device description object
  * @si: scanning information
  *
@@ -1583,13 +1584,12 @@ static void protection_trees_destroy(struct ubi_device *ubi)
 }
 
 /**
- * ubi_wl_close - close the wear-leveling unit.
+ * ubi_wl_close - close the wear-leveling sub-system.
  * @ubi: UBI device description object
  */
 void ubi_wl_close(struct ubi_device *ubi)
 {
-	dbg_wl("close the UBI wear-leveling unit");
-
+	dbg_wl("close the WL sub-system");
 	cancel_pending(ubi);
 	protection_trees_destroy(ubi);
 	tree_destroy(&ubi->used);
diff --git a/include/linux/mtd/ubi.h b/include/linux/mtd/ubi.h
index 83302bbbddb4..6316fafe5c2a 100644
--- a/include/linux/mtd/ubi.h
+++ b/include/linux/mtd/ubi.h
@@ -45,13 +45,13 @@ enum {
  * @size: how many physical eraseblocks are reserved for this volume
  * @used_bytes: how many bytes of data this volume contains
  * @used_ebs: how many physical eraseblocks of this volume actually contain any
- * data
+ *            data
  * @vol_type: volume type (%UBI_DYNAMIC_VOLUME or %UBI_STATIC_VOLUME)
  * @corrupted: non-zero if the volume is corrupted (static volumes only)
  * @upd_marker: non-zero if the volume has update marker set
  * @alignment: volume alignment
  * @usable_leb_size: how many bytes are available in logical eraseblocks of
- * this volume
+ *                   this volume
  * @name_len: volume name length
  * @name: volume name
  * @cdev: UBI volume character device major and minor numbers

From c8566350a3229ca505b84313c65d1403b4d0cbfc Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Wed, 16 Jul 2008 17:40:22 +0300
Subject: [PATCH 0183/1171] UBI: fix and re-work debugging stuff

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/build.c  |   2 +-
 drivers/mtd/ubi/cdev.c   |  26 +++----
 drivers/mtd/ubi/debug.c  | 160 ++++++++++++++++++++-------------------
 drivers/mtd/ubi/debug.h  |  68 +++++++++++------
 drivers/mtd/ubi/gluebi.c |  10 +--
 drivers/mtd/ubi/io.c     |   4 +-
 drivers/mtd/ubi/kapi.c   |  20 ++---
 drivers/mtd/ubi/scan.c   |   2 +-
 drivers/mtd/ubi/upd.c    |  16 ++--
 drivers/mtd/ubi/vmt.c    |  73 +++++++++---------
 drivers/mtd/ubi/vtbl.c   |   2 +-
 11 files changed, 206 insertions(+), 177 deletions(-)

diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c
index 27271fe32e02..7210e1da1fcb 100644
--- a/drivers/mtd/ubi/build.c
+++ b/drivers/mtd/ubi/build.c
@@ -403,7 +403,7 @@ static int uif_init(struct ubi_device *ubi)
 
 	ubi_assert(MINOR(dev) == 0);
 	cdev_init(&ubi->cdev, &ubi_cdev_operations);
-	dbg_msg("%s major is %u", ubi->ubi_name, MAJOR(dev));
+	dbg_gen("%s major is %u", ubi->ubi_name, MAJOR(dev));
 	ubi->cdev.owner = THIS_MODULE;
 
 	err = cdev_add(&ubi->cdev, dev, 1);
diff --git a/drivers/mtd/ubi/cdev.c b/drivers/mtd/ubi/cdev.c
index 4fb84e3e6502..7c19918cc914 100644
--- a/drivers/mtd/ubi/cdev.c
+++ b/drivers/mtd/ubi/cdev.c
@@ -116,7 +116,7 @@ static int vol_cdev_open(struct inode *inode, struct file *file)
 	else
 		mode = UBI_READONLY;
 
-	dbg_msg("open volume %d, mode %d", vol_id, mode);
+	dbg_gen("open volume %d, mode %d", vol_id, mode);
 
 	desc = ubi_open_volume(ubi_num, vol_id, mode);
 	unlock_kernel();
@@ -132,7 +132,7 @@ static int vol_cdev_release(struct inode *inode, struct file *file)
 	struct ubi_volume_desc *desc = file->private_data;
 	struct ubi_volume *vol = desc->vol;
 
-	dbg_msg("release volume %d, mode %d", vol->vol_id, desc->mode);
+	dbg_gen("release volume %d, mode %d", vol->vol_id, desc->mode);
 
 	if (vol->updating) {
 		ubi_warn("update of volume %d not finished, volume is damaged",
@@ -141,7 +141,7 @@ static int vol_cdev_release(struct inode *inode, struct file *file)
 		vol->updating = 0;
 		vfree(vol->upd_buf);
 	} else if (vol->changing_leb) {
-		dbg_msg("only %lld of %lld bytes received for atomic LEB change"
+		dbg_gen("only %lld of %lld bytes received for atomic LEB change"
 			" for volume %d:%d, cancel", vol->upd_received,
 			vol->upd_bytes, vol->ubi->ubi_num, vol->vol_id);
 		vol->changing_leb = 0;
@@ -183,7 +183,7 @@ static loff_t vol_cdev_llseek(struct file *file, loff_t offset, int origin)
 		return -EINVAL;
 	}
 
-	dbg_msg("seek volume %d, offset %lld, origin %d, new offset %lld",
+	dbg_gen("seek volume %d, offset %lld, origin %d, new offset %lld",
 		vol->vol_id, offset, origin, new_offset);
 
 	file->f_pos = new_offset;
@@ -201,7 +201,7 @@ static ssize_t vol_cdev_read(struct file *file, __user char *buf, size_t count,
 	void *tbuf;
 	uint64_t tmp;
 
-	dbg_msg("read %zd bytes from offset %lld of volume %d",
+	dbg_gen("read %zd bytes from offset %lld of volume %d",
 		count, *offp, vol->vol_id);
 
 	if (vol->updating) {
@@ -216,7 +216,7 @@ static ssize_t vol_cdev_read(struct file *file, __user char *buf, size_t count,
 		return 0;
 
 	if (vol->corrupted)
-		dbg_msg("read from corrupted volume %d", vol->vol_id);
+		dbg_gen("read from corrupted volume %d", vol->vol_id);
 
 	if (*offp + count > vol->used_bytes)
 		count_save = count = vol->used_bytes - *offp;
@@ -285,7 +285,7 @@ static ssize_t vol_cdev_direct_write(struct file *file, const char __user *buf,
 	char *tbuf;
 	uint64_t tmp;
 
-	dbg_msg("requested: write %zd bytes to offset %lld of volume %u",
+	dbg_gen("requested: write %zd bytes to offset %lld of volume %u",
 		count, *offp, vol->vol_id);
 
 	if (vol->vol_type == UBI_STATIC_VOLUME)
@@ -514,7 +514,7 @@ static int vol_cdev_ioctl(struct inode *inode, struct file *file,
 			break;
 		}
 
-		dbg_msg("erase LEB %d:%d", vol->vol_id, lnum);
+		dbg_gen("erase LEB %d:%d", vol->vol_id, lnum);
 		err = ubi_eba_unmap_leb(ubi, vol, lnum);
 		if (err)
 			break;
@@ -626,7 +626,7 @@ static int ubi_cdev_ioctl(struct inode *inode, struct file *file,
 	{
 		struct ubi_mkvol_req req;
 
-		dbg_msg("create volume");
+		dbg_gen("create volume");
 		err = copy_from_user(&req, argp, sizeof(struct ubi_mkvol_req));
 		if (err) {
 			err = -EFAULT;
@@ -656,7 +656,7 @@ static int ubi_cdev_ioctl(struct inode *inode, struct file *file,
 	{
 		int vol_id;
 
-		dbg_msg("remove volume");
+		dbg_gen("remove volume");
 		err = get_user(vol_id, (__user int32_t *)argp);
 		if (err) {
 			err = -EFAULT;
@@ -689,7 +689,7 @@ static int ubi_cdev_ioctl(struct inode *inode, struct file *file,
 		uint64_t tmp;
 		struct ubi_rsvol_req req;
 
-		dbg_msg("re-size volume");
+		dbg_gen("re-size volume");
 		err = copy_from_user(&req, argp, sizeof(struct ubi_rsvol_req));
 		if (err) {
 			err = -EFAULT;
@@ -742,7 +742,7 @@ static int ctrl_cdev_ioctl(struct inode *inode, struct file *file,
 		struct ubi_attach_req req;
 		struct mtd_info *mtd;
 
-		dbg_msg("attach MTD device");
+		dbg_gen("attach MTD device");
 		err = copy_from_user(&req, argp, sizeof(struct ubi_attach_req));
 		if (err) {
 			err = -EFAULT;
@@ -782,7 +782,7 @@ static int ctrl_cdev_ioctl(struct inode *inode, struct file *file,
 	{
 		int ubi_num;
 
-		dbg_msg("dettach MTD device");
+		dbg_gen("dettach MTD device");
 		err = get_user(ubi_num, (__user int32_t *)argp);
 		if (err) {
 			err = -EFAULT;
diff --git a/drivers/mtd/ubi/debug.c b/drivers/mtd/ubi/debug.c
index 56956ec2845f..21e0d7d76a44 100644
--- a/drivers/mtd/ubi/debug.c
+++ b/drivers/mtd/ubi/debug.c
@@ -24,7 +24,7 @@
  * changes.
  */
 
-#ifdef CONFIG_MTD_UBI_DEBUG_MSG
+#ifdef CONFIG_MTD_UBI_DEBUG
 
 #include "ubi.h"
 
@@ -34,14 +34,19 @@
  */
 void ubi_dbg_dump_ec_hdr(const struct ubi_ec_hdr *ec_hdr)
 {
-	dbg_msg("erase counter header dump:");
-	dbg_msg("magic          %#08x", be32_to_cpu(ec_hdr->magic));
-	dbg_msg("version        %d",    (int)ec_hdr->version);
-	dbg_msg("ec             %llu",  (long long)be64_to_cpu(ec_hdr->ec));
-	dbg_msg("vid_hdr_offset %d",    be32_to_cpu(ec_hdr->vid_hdr_offset));
-	dbg_msg("data_offset    %d",    be32_to_cpu(ec_hdr->data_offset));
-	dbg_msg("hdr_crc        %#08x", be32_to_cpu(ec_hdr->hdr_crc));
-	dbg_msg("erase counter header hexdump:");
+	printk(KERN_DEBUG "Erase counter header dump:\n");
+	printk(KERN_DEBUG "\tmagic          %#08x\n",
+	       be32_to_cpu(ec_hdr->magic));
+	printk(KERN_DEBUG "\tversion        %d\n", (int)ec_hdr->version);
+	printk(KERN_DEBUG "\tec             %llu\n",
+	       (long long)be64_to_cpu(ec_hdr->ec));
+	printk(KERN_DEBUG "\tvid_hdr_offset %d\n",
+	       be32_to_cpu(ec_hdr->vid_hdr_offset));
+	printk(KERN_DEBUG "\tdata_offset    %d\n",
+	       be32_to_cpu(ec_hdr->data_offset));
+	printk(KERN_DEBUG "\thdr_crc        %#08x\n",
+	       be32_to_cpu(ec_hdr->hdr_crc));
+	printk(KERN_DEBUG "erase counter header hexdump:\n");
 	print_hex_dump(KERN_DEBUG, "", DUMP_PREFIX_OFFSET, 32, 1,
 		       ec_hdr, UBI_EC_HDR_SIZE, 1);
 }
@@ -52,22 +57,24 @@ void ubi_dbg_dump_ec_hdr(const struct ubi_ec_hdr *ec_hdr)
  */
 void ubi_dbg_dump_vid_hdr(const struct ubi_vid_hdr *vid_hdr)
 {
-	dbg_msg("volume identifier header dump:");
-	dbg_msg("magic     %08x", be32_to_cpu(vid_hdr->magic));
-	dbg_msg("version   %d",   (int)vid_hdr->version);
-	dbg_msg("vol_type  %d",   (int)vid_hdr->vol_type);
-	dbg_msg("copy_flag %d",   (int)vid_hdr->copy_flag);
-	dbg_msg("compat    %d",   (int)vid_hdr->compat);
-	dbg_msg("vol_id    %d",   be32_to_cpu(vid_hdr->vol_id));
-	dbg_msg("lnum      %d",   be32_to_cpu(vid_hdr->lnum));
-	dbg_msg("leb_ver   %u",   be32_to_cpu(vid_hdr->leb_ver));
-	dbg_msg("data_size %d",   be32_to_cpu(vid_hdr->data_size));
-	dbg_msg("used_ebs  %d",   be32_to_cpu(vid_hdr->used_ebs));
-	dbg_msg("data_pad  %d",   be32_to_cpu(vid_hdr->data_pad));
-	dbg_msg("sqnum     %llu",
+	printk(KERN_DEBUG "Volume identifier header dump:\n");
+	printk(KERN_DEBUG "\tmagic     %08x\n", be32_to_cpu(vid_hdr->magic));
+	printk(KERN_DEBUG "\tversion   %d\n",   (int)vid_hdr->version);
+	printk(KERN_DEBUG "\tvol_type  %d\n",   (int)vid_hdr->vol_type);
+	printk(KERN_DEBUG "\tcopy_flag %d\n",   (int)vid_hdr->copy_flag);
+	printk(KERN_DEBUG "\tcompat    %d\n",   (int)vid_hdr->compat);
+	printk(KERN_DEBUG "\tvol_id    %d\n",   be32_to_cpu(vid_hdr->vol_id));
+	printk(KERN_DEBUG "\tlnum      %d\n",   be32_to_cpu(vid_hdr->lnum));
+	printk(KERN_DEBUG "\tleb_ver   %u\n",   be32_to_cpu(vid_hdr->leb_ver));
+	printk(KERN_DEBUG "\tdata_size %d\n",   be32_to_cpu(vid_hdr->data_size));
+	printk(KERN_DEBUG "\tused_ebs  %d\n",   be32_to_cpu(vid_hdr->used_ebs));
+	printk(KERN_DEBUG "\tdata_pad  %d\n",   be32_to_cpu(vid_hdr->data_pad));
+	printk(KERN_DEBUG "\tsqnum     %llu\n",
 		(unsigned long long)be64_to_cpu(vid_hdr->sqnum));
-	dbg_msg("hdr_crc   %08x", be32_to_cpu(vid_hdr->hdr_crc));
-	dbg_msg("volume identifier header hexdump:");
+	printk(KERN_DEBUG "\thdr_crc   %08x\n", be32_to_cpu(vid_hdr->hdr_crc));
+	printk(KERN_DEBUG "Volume identifier header hexdump:\n");
+	print_hex_dump(KERN_DEBUG, "", DUMP_PREFIX_OFFSET, 32, 1,
+		       vid_hdr, UBI_VID_HDR_SIZE, 1);
 }
 
 /**
@@ -76,27 +83,27 @@ void ubi_dbg_dump_vid_hdr(const struct ubi_vid_hdr *vid_hdr)
  */
 void ubi_dbg_dump_vol_info(const struct ubi_volume *vol)
 {
-	dbg_msg("volume information dump:");
-	dbg_msg("vol_id          %d", vol->vol_id);
-	dbg_msg("reserved_pebs   %d", vol->reserved_pebs);
-	dbg_msg("alignment       %d", vol->alignment);
-	dbg_msg("data_pad        %d", vol->data_pad);
-	dbg_msg("vol_type        %d", vol->vol_type);
-	dbg_msg("name_len        %d", vol->name_len);
-	dbg_msg("usable_leb_size %d", vol->usable_leb_size);
-	dbg_msg("used_ebs        %d", vol->used_ebs);
-	dbg_msg("used_bytes      %lld", vol->used_bytes);
-	dbg_msg("last_eb_bytes   %d", vol->last_eb_bytes);
-	dbg_msg("corrupted       %d", vol->corrupted);
-	dbg_msg("upd_marker      %d", vol->upd_marker);
+	printk(KERN_DEBUG "Volume information dump:\n");
+	printk(KERN_DEBUG "\tvol_id          %d\n", vol->vol_id);
+	printk(KERN_DEBUG "\treserved_pebs   %d\n", vol->reserved_pebs);
+	printk(KERN_DEBUG "\talignment       %d\n", vol->alignment);
+	printk(KERN_DEBUG "\tdata_pad        %d\n", vol->data_pad);
+	printk(KERN_DEBUG "\tvol_type        %d\n", vol->vol_type);
+	printk(KERN_DEBUG "\tname_len        %d\n", vol->name_len);
+	printk(KERN_DEBUG "\tusable_leb_size %d\n", vol->usable_leb_size);
+	printk(KERN_DEBUG "\tused_ebs        %d\n", vol->used_ebs);
+	printk(KERN_DEBUG "\tused_bytes      %lld\n", vol->used_bytes);
+	printk(KERN_DEBUG "\tlast_eb_bytes   %d\n", vol->last_eb_bytes);
+	printk(KERN_DEBUG "\tcorrupted       %d\n", vol->corrupted);
+	printk(KERN_DEBUG "\tupd_marker      %d\n", vol->upd_marker);
 
 	if (vol->name_len <= UBI_VOL_NAME_MAX &&
 	    strnlen(vol->name, vol->name_len + 1) == vol->name_len) {
-		dbg_msg("name            %s", vol->name);
+		printk(KERN_DEBUG "\tname            %s\n", vol->name);
 	} else {
-		dbg_msg("the 1st 5 characters of the name: %c%c%c%c%c",
-			vol->name[0], vol->name[1], vol->name[2],
-			vol->name[3], vol->name[4]);
+		printk(KERN_DEBUG "\t1st 5 characters of name: %c%c%c%c%c\n",
+		       vol->name[0], vol->name[1], vol->name[2],
+		       vol->name[3], vol->name[4]);
 	}
 }
 
@@ -109,28 +116,29 @@ void ubi_dbg_dump_vtbl_record(const struct ubi_vtbl_record *r, int idx)
 {
 	int name_len = be16_to_cpu(r->name_len);
 
-	dbg_msg("volume table record %d dump:", idx);
-	dbg_msg("reserved_pebs   %d", be32_to_cpu(r->reserved_pebs));
-	dbg_msg("alignment       %d", be32_to_cpu(r->alignment));
-	dbg_msg("data_pad        %d", be32_to_cpu(r->data_pad));
-	dbg_msg("vol_type        %d", (int)r->vol_type);
-	dbg_msg("upd_marker      %d", (int)r->upd_marker);
-	dbg_msg("name_len        %d", name_len);
+	printk(KERN_DEBUG "Volume table record %d dump:\n", idx);
+	printk(KERN_DEBUG "\treserved_pebs   %d\n",
+	       be32_to_cpu(r->reserved_pebs));
+	printk(KERN_DEBUG "\talignment       %d\n", be32_to_cpu(r->alignment));
+	printk(KERN_DEBUG "\tdata_pad        %d\n", be32_to_cpu(r->data_pad));
+	printk(KERN_DEBUG "\tvol_type        %d\n", (int)r->vol_type);
+	printk(KERN_DEBUG "\tupd_marker      %d\n", (int)r->upd_marker);
+	printk(KERN_DEBUG "\tname_len        %d\n", name_len);
 
 	if (r->name[0] == '\0') {
-		dbg_msg("name            NULL");
+		printk(KERN_DEBUG "\tname            NULL\n");
 		return;
 	}
 
 	if (name_len <= UBI_VOL_NAME_MAX &&
 	    strnlen(&r->name[0], name_len + 1) == name_len) {
-		dbg_msg("name            %s", &r->name[0]);
+		printk(KERN_DEBUG "\tname            %s\n", &r->name[0]);
 	} else {
-		dbg_msg("1st 5 characters of the name: %c%c%c%c%c",
+		printk(KERN_DEBUG "\t1st 5 characters of name: %c%c%c%c%c\n",
 			r->name[0], r->name[1], r->name[2], r->name[3],
 			r->name[4]);
 	}
-	dbg_msg("crc             %#08x", be32_to_cpu(r->crc));
+	printk(KERN_DEBUG "\tcrc             %#08x\n", be32_to_cpu(r->crc));
 }
 
 /**
@@ -139,15 +147,15 @@ void ubi_dbg_dump_vtbl_record(const struct ubi_vtbl_record *r, int idx)
  */
 void ubi_dbg_dump_sv(const struct ubi_scan_volume *sv)
 {
-	dbg_msg("volume scanning information dump:");
-	dbg_msg("vol_id         %d", sv->vol_id);
-	dbg_msg("highest_lnum   %d", sv->highest_lnum);
-	dbg_msg("leb_count      %d", sv->leb_count);
-	dbg_msg("compat         %d", sv->compat);
-	dbg_msg("vol_type       %d", sv->vol_type);
-	dbg_msg("used_ebs       %d", sv->used_ebs);
-	dbg_msg("last_data_size %d", sv->last_data_size);
-	dbg_msg("data_pad       %d", sv->data_pad);
+	printk(KERN_DEBUG "Volume scanning information dump:\n");
+	printk(KERN_DEBUG "\tvol_id         %d\n", sv->vol_id);
+	printk(KERN_DEBUG "\thighest_lnum   %d\n", sv->highest_lnum);
+	printk(KERN_DEBUG "\tleb_count      %d\n", sv->leb_count);
+	printk(KERN_DEBUG "\tcompat         %d\n", sv->compat);
+	printk(KERN_DEBUG "\tvol_type       %d\n", sv->vol_type);
+	printk(KERN_DEBUG "\tused_ebs       %d\n", sv->used_ebs);
+	printk(KERN_DEBUG "\tlast_data_size %d\n", sv->last_data_size);
+	printk(KERN_DEBUG "\tdata_pad       %d\n", sv->data_pad);
 }
 
 /**
@@ -157,14 +165,14 @@ void ubi_dbg_dump_sv(const struct ubi_scan_volume *sv)
  */
 void ubi_dbg_dump_seb(const struct ubi_scan_leb *seb, int type)
 {
-	dbg_msg("eraseblock scanning information dump:");
-	dbg_msg("ec       %d", seb->ec);
-	dbg_msg("pnum     %d", seb->pnum);
+	printk(KERN_DEBUG "eraseblock scanning information dump:\n");
+	printk(KERN_DEBUG "\tec       %d\n", seb->ec);
+	printk(KERN_DEBUG "\tpnum     %d\n", seb->pnum);
 	if (type == 0) {
-		dbg_msg("lnum     %d", seb->lnum);
-		dbg_msg("scrub    %d", seb->scrub);
-		dbg_msg("sqnum    %llu", seb->sqnum);
-		dbg_msg("leb_ver  %u", seb->leb_ver);
+		printk(KERN_DEBUG "\tlnum     %d\n", seb->lnum);
+		printk(KERN_DEBUG "\tscrub    %d\n", seb->scrub);
+		printk(KERN_DEBUG "\tsqnum    %llu\n", seb->sqnum);
+		printk(KERN_DEBUG "\tleb_ver  %u\n", seb->leb_ver);
 	}
 }
 
@@ -176,16 +184,16 @@ void ubi_dbg_dump_mkvol_req(const struct ubi_mkvol_req *req)
 {
 	char nm[17];
 
-	dbg_msg("volume creation request dump:");
-	dbg_msg("vol_id    %d",   req->vol_id);
-	dbg_msg("alignment %d",   req->alignment);
-	dbg_msg("bytes     %lld", (long long)req->bytes);
-	dbg_msg("vol_type  %d",   req->vol_type);
-	dbg_msg("name_len  %d",   req->name_len);
+	printk(KERN_DEBUG "Volume creation request dump:\n");
+	printk(KERN_DEBUG "\tvol_id    %d\n",   req->vol_id);
+	printk(KERN_DEBUG "\talignment %d\n",   req->alignment);
+	printk(KERN_DEBUG "\tbytes     %lld\n", (long long)req->bytes);
+	printk(KERN_DEBUG "\tvol_type  %d\n",   req->vol_type);
+	printk(KERN_DEBUG "\tname_len  %d\n",   req->name_len);
 
 	memcpy(nm, req->name, 16);
 	nm[16] = 0;
-	dbg_msg("the 1st 16 characters of the name: %s", nm);
+	printk(KERN_DEBUG "\t1st 16 characters of name: %s\n", nm);
 }
 
-#endif /* CONFIG_MTD_UBI_DEBUG_MSG */
+#endif /* CONFIG_MTD_UBI_DEBUG */
diff --git a/drivers/mtd/ubi/debug.h b/drivers/mtd/ubi/debug.h
index 7d8d77c31dfe..78e914d23ece 100644
--- a/drivers/mtd/ubi/debug.h
+++ b/drivers/mtd/ubi/debug.h
@@ -24,21 +24,16 @@
 #ifdef CONFIG_MTD_UBI_DEBUG
 #include <linux/random.h>
 
-#define ubi_assert(expr)  BUG_ON(!(expr))
 #define dbg_err(fmt, ...) ubi_err(fmt, ##__VA_ARGS__)
-#else
-#define ubi_assert(expr)  ({})
-#define dbg_err(fmt, ...) ({})
-#endif
 
-#ifdef CONFIG_MTD_UBI_DEBUG_DISABLE_BGT
-#define DBG_DISABLE_BGT 1
-#else
-#define DBG_DISABLE_BGT 0
-#endif
+#define ubi_assert(expr)  do {                                               \
+        if (unlikely(!(expr))) {                                             \
+                printk(KERN_CRIT "UBI assert failed in %s at %u (pid %d)\n", \
+                       __func__, __LINE__, current->pid);                    \
+                ubi_dbg_dump_stack();                                        \
+        }                                                                    \
+} while (0)
 
-#ifdef CONFIG_MTD_UBI_DEBUG_MSG
-/* Generic debugging message */
 #define dbg_msg(fmt, ...)                                    \
 	printk(KERN_DEBUG "UBI DBG (pid %d): %s: " fmt "\n", \
 	       current->pid, __func__, ##__VA_ARGS__)
@@ -61,19 +56,12 @@ void ubi_dbg_dump_sv(const struct ubi_scan_volume *sv);
 void ubi_dbg_dump_seb(const struct ubi_scan_leb *seb, int type);
 void ubi_dbg_dump_mkvol_req(const struct ubi_mkvol_req *req);
 
+#ifdef CONFIG_MTD_UBI_DEBUG_MSG
+/* General debugging messages */
+#define dbg_gen(fmt, ...) dbg_msg(fmt, ##__VA_ARGS__)
 #else
-
-#define dbg_msg(fmt, ...)    ({})
-#define ubi_dbg_dump_stack() ({})
-#define ubi_dbg_dump_ec_hdr(ec_hdr)      ({})
-#define ubi_dbg_dump_vid_hdr(vid_hdr)    ({})
-#define ubi_dbg_dump_vol_info(vol)       ({})
-#define ubi_dbg_dump_vtbl_record(r, idx) ({})
-#define ubi_dbg_dump_sv(sv)              ({})
-#define ubi_dbg_dump_seb(seb, type)      ({})
-#define ubi_dbg_dump_mkvol_req(req)      ({})
-
-#endif /* CONFIG_MTD_UBI_DEBUG_MSG */
+#define dbg_gen(fmt, ...) ({})
+#endif
 
 #ifdef CONFIG_MTD_UBI_DEBUG_MSG_EBA
 /* Messages from the eraseblock association sub-system */
@@ -105,6 +93,12 @@ void ubi_dbg_dump_mkvol_req(const struct ubi_mkvol_req *req);
 #define UBI_IO_DEBUG 0
 #endif
 
+#ifdef CONFIG_MTD_UBI_DEBUG_DISABLE_BGT
+#define DBG_DISABLE_BGT 1
+#else
+#define DBG_DISABLE_BGT 0
+#endif
+
 #ifdef CONFIG_MTD_UBI_DEBUG_EMULATE_BITFLIPS
 /**
  * ubi_dbg_is_bitflip - if it is time to emulate a bit-flip.
@@ -149,4 +143,30 @@ static inline int ubi_dbg_is_erase_failure(void)
 #define ubi_dbg_is_erase_failure() 0
 #endif
 
+#else
+
+#define ubi_assert(expr)                 ({})
+#define dbg_err(fmt, ...)                ({})
+#define dbg_msg(fmt, ...)                ({})
+#define dbg_gen(fmt, ...)                ({})
+#define dbg_eba(fmt, ...)                ({})
+#define dbg_wl(fmt, ...)                 ({})
+#define dbg_io(fmt, ...)                 ({})
+#define dbg_bld(fmt, ...)                ({})
+#define ubi_dbg_dump_stack()             ({})
+#define ubi_dbg_dump_ec_hdr(ec_hdr)      ({})
+#define ubi_dbg_dump_vid_hdr(vid_hdr)    ({})
+#define ubi_dbg_dump_vol_info(vol)       ({})
+#define ubi_dbg_dump_vtbl_record(r, idx) ({})
+#define ubi_dbg_dump_sv(sv)              ({})
+#define ubi_dbg_dump_seb(seb, type)      ({})
+#define ubi_dbg_dump_mkvol_req(req)      ({})
+
+#define UBI_IO_DEBUG               0
+#define DBG_DISABLE_BGT            0
+#define ubi_dbg_is_bitflip()       0
+#define ubi_dbg_is_write_failure() 0
+#define ubi_dbg_is_erase_failure() 0
+
+#endif /* !CONFIG_MTD_UBI_DEBUG */
 #endif /* !__UBI_DEBUG_H__ */
diff --git a/drivers/mtd/ubi/gluebi.c b/drivers/mtd/ubi/gluebi.c
index ae76ab638b2f..49f52dceea93 100644
--- a/drivers/mtd/ubi/gluebi.c
+++ b/drivers/mtd/ubi/gluebi.c
@@ -111,7 +111,7 @@ static int gluebi_read(struct mtd_info *mtd, loff_t from, size_t len,
 	struct ubi_device *ubi;
 	uint64_t tmp = from;
 
-	dbg_msg("read %zd bytes from offset %lld", len, from);
+	dbg_gen("read %zd bytes from offset %lld", len, from);
 
 	if (len < 0 || from < 0 || from + len > mtd->size)
 		return -EINVAL;
@@ -162,7 +162,7 @@ static int gluebi_write(struct mtd_info *mtd, loff_t to, size_t len,
 	struct ubi_device *ubi;
 	uint64_t tmp = to;
 
-	dbg_msg("write %zd bytes to offset %lld", len, to);
+	dbg_gen("write %zd bytes to offset %lld", len, to);
 
 	if (len < 0 || to < 0 || len + to > mtd->size)
 		return -EINVAL;
@@ -215,7 +215,7 @@ static int gluebi_erase(struct mtd_info *mtd, struct erase_info *instr)
 	struct ubi_volume *vol;
 	struct ubi_device *ubi;
 
-	dbg_msg("erase %u bytes at offset %u", instr->len, instr->addr);
+	dbg_gen("erase %u bytes at offset %u", instr->len, instr->addr);
 
 	if (instr->addr < 0 || instr->addr > mtd->size - mtd->erasesize)
 		return -EINVAL;
@@ -304,7 +304,7 @@ int ubi_create_gluebi(struct ubi_device *ubi, struct ubi_volume *vol)
 		return -ENFILE;
 	}
 
-	dbg_msg("added mtd%d (\"%s\"), size %u, EB size %u",
+	dbg_gen("added mtd%d (\"%s\"), size %u, EB size %u",
 		mtd->index, mtd->name, mtd->size, mtd->erasesize);
 	return 0;
 }
@@ -322,7 +322,7 @@ int ubi_destroy_gluebi(struct ubi_volume *vol)
 	int err;
 	struct mtd_info *mtd = &vol->gluebi_mtd;
 
-	dbg_msg("remove mtd%d", mtd->index);
+	dbg_gen("remove mtd%d", mtd->index);
 	err = del_mtd_device(mtd);
 	if (err)
 		return err;
diff --git a/drivers/mtd/ubi/io.c b/drivers/mtd/ubi/io.c
index 561e7b2f96cb..27b9c2c2fc60 100644
--- a/drivers/mtd/ubi/io.c
+++ b/drivers/mtd/ubi/io.c
@@ -187,7 +187,7 @@ retry:
 		ubi_assert(len == read);
 
 		if (ubi_dbg_is_bitflip()) {
-			dbg_msg("bit-flip (emulated)");
+			dbg_gen("bit-flip (emulated)");
 			err = UBI_IO_BITFLIPS;
 		}
 	}
@@ -1256,7 +1256,7 @@ static int paranoid_check_all_ff(struct ubi_device *ubi, int pnum, int offset,
 
 fail:
 	ubi_err("paranoid check failed for PEB %d", pnum);
-	dbg_msg("hex dump of the %d-%d region", offset, offset + len);
+	ubi_msg("hex dump of the %d-%d region", offset, offset + len);
 	print_hex_dump(KERN_DEBUG, "", DUMP_PREFIX_OFFSET, 32, 1,
 		       ubi->dbg_peb_buf, len, 1);
 	err = 1;
diff --git a/drivers/mtd/ubi/kapi.c b/drivers/mtd/ubi/kapi.c
index e65c8e0bcd5d..5d9bcf109c13 100644
--- a/drivers/mtd/ubi/kapi.c
+++ b/drivers/mtd/ubi/kapi.c
@@ -106,7 +106,7 @@ struct ubi_volume_desc *ubi_open_volume(int ubi_num, int vol_id, int mode)
 	struct ubi_device *ubi;
 	struct ubi_volume *vol;
 
-	dbg_msg("open device %d volume %d, mode %d", ubi_num, vol_id, mode);
+	dbg_gen("open device %d volume %d, mode %d", ubi_num, vol_id, mode);
 
 	if (ubi_num < 0 || ubi_num >= UBI_MAX_DEVICES)
 		return ERR_PTR(-EINVAL);
@@ -215,7 +215,7 @@ struct ubi_volume_desc *ubi_open_volume_nm(int ubi_num, const char *name,
 	struct ubi_device *ubi;
 	struct ubi_volume_desc *ret;
 
-	dbg_msg("open volume %s, mode %d", name, mode);
+	dbg_gen("open volume %s, mode %d", name, mode);
 
 	if (!name)
 		return ERR_PTR(-EINVAL);
@@ -266,7 +266,7 @@ void ubi_close_volume(struct ubi_volume_desc *desc)
 	struct ubi_volume *vol = desc->vol;
 	struct ubi_device *ubi = vol->ubi;
 
-	dbg_msg("close volume %d, mode %d", vol->vol_id, desc->mode);
+	dbg_gen("close volume %d, mode %d", vol->vol_id, desc->mode);
 
 	spin_lock(&ubi->volumes_lock);
 	switch (desc->mode) {
@@ -323,7 +323,7 @@ int ubi_leb_read(struct ubi_volume_desc *desc, int lnum, char *buf, int offset,
 	struct ubi_device *ubi = vol->ubi;
 	int err, vol_id = vol->vol_id;
 
-	dbg_msg("read %d bytes from LEB %d:%d:%d", len, vol_id, lnum, offset);
+	dbg_gen("read %d bytes from LEB %d:%d:%d", len, vol_id, lnum, offset);
 
 	if (vol_id < 0 || vol_id >= ubi->vtbl_slots || lnum < 0 ||
 	    lnum >= vol->used_ebs || offset < 0 || len < 0 ||
@@ -388,7 +388,7 @@ int ubi_leb_write(struct ubi_volume_desc *desc, int lnum, const void *buf,
 	struct ubi_device *ubi = vol->ubi;
 	int vol_id = vol->vol_id;
 
-	dbg_msg("write %d bytes to LEB %d:%d:%d", len, vol_id, lnum, offset);
+	dbg_gen("write %d bytes to LEB %d:%d:%d", len, vol_id, lnum, offset);
 
 	if (vol_id < 0 || vol_id >= ubi->vtbl_slots)
 		return -EINVAL;
@@ -438,7 +438,7 @@ int ubi_leb_change(struct ubi_volume_desc *desc, int lnum, const void *buf,
 	struct ubi_device *ubi = vol->ubi;
 	int vol_id = vol->vol_id;
 
-	dbg_msg("atomically write %d bytes to LEB %d:%d", len, vol_id, lnum);
+	dbg_gen("atomically write %d bytes to LEB %d:%d", len, vol_id, lnum);
 
 	if (vol_id < 0 || vol_id >= ubi->vtbl_slots)
 		return -EINVAL;
@@ -482,7 +482,7 @@ int ubi_leb_erase(struct ubi_volume_desc *desc, int lnum)
 	struct ubi_device *ubi = vol->ubi;
 	int err;
 
-	dbg_msg("erase LEB %d:%d", vol->vol_id, lnum);
+	dbg_gen("erase LEB %d:%d", vol->vol_id, lnum);
 
 	if (desc->mode == UBI_READONLY || vol->vol_type == UBI_STATIC_VOLUME)
 		return -EROFS;
@@ -542,7 +542,7 @@ int ubi_leb_unmap(struct ubi_volume_desc *desc, int lnum)
 	struct ubi_volume *vol = desc->vol;
 	struct ubi_device *ubi = vol->ubi;
 
-	dbg_msg("unmap LEB %d:%d", vol->vol_id, lnum);
+	dbg_gen("unmap LEB %d:%d", vol->vol_id, lnum);
 
 	if (desc->mode == UBI_READONLY || vol->vol_type == UBI_STATIC_VOLUME)
 		return -EROFS;
@@ -579,7 +579,7 @@ int ubi_leb_map(struct ubi_volume_desc *desc, int lnum, int dtype)
 	struct ubi_volume *vol = desc->vol;
 	struct ubi_device *ubi = vol->ubi;
 
-	dbg_msg("unmap LEB %d:%d", vol->vol_id, lnum);
+	dbg_gen("unmap LEB %d:%d", vol->vol_id, lnum);
 
 	if (desc->mode == UBI_READONLY || vol->vol_type == UBI_STATIC_VOLUME)
 		return -EROFS;
@@ -621,7 +621,7 @@ int ubi_is_mapped(struct ubi_volume_desc *desc, int lnum)
 {
 	struct ubi_volume *vol = desc->vol;
 
-	dbg_msg("test LEB %d:%d", vol->vol_id, lnum);
+	dbg_gen("test LEB %d:%d", vol->vol_id, lnum);
 
 	if (lnum < 0 || lnum >= vol->reserved_pebs)
 		return -EINVAL;
diff --git a/drivers/mtd/ubi/scan.c b/drivers/mtd/ubi/scan.c
index 892c2ba49777..40eca9ce5fab 100644
--- a/drivers/mtd/ubi/scan.c
+++ b/drivers/mtd/ubi/scan.c
@@ -932,7 +932,7 @@ struct ubi_scan_info *ubi_scan(struct ubi_device *ubi)
 	for (pnum = 0; pnum < ubi->peb_count; pnum++) {
 		cond_resched();
 
-		dbg_msg("process PEB %d", pnum);
+		dbg_gen("process PEB %d", pnum);
 		err = process_eb(ubi, si, pnum);
 		if (err < 0)
 			goto out_vidh;
diff --git a/drivers/mtd/ubi/upd.c b/drivers/mtd/ubi/upd.c
index 6fa1ab3f2a70..1230a5e1b530 100644
--- a/drivers/mtd/ubi/upd.c
+++ b/drivers/mtd/ubi/upd.c
@@ -56,11 +56,11 @@ static int set_update_marker(struct ubi_device *ubi, struct ubi_volume *vol)
 	int err;
 	struct ubi_vtbl_record vtbl_rec;
 
-	dbg_msg("set update marker for volume %d", vol->vol_id);
+	dbg_gen("set update marker for volume %d", vol->vol_id);
 
 	if (vol->upd_marker) {
 		ubi_assert(ubi->vtbl[vol->vol_id].upd_marker);
-		dbg_msg("already set");
+		dbg_gen("already set");
 		return 0;
 	}
 
@@ -92,7 +92,7 @@ static int clear_update_marker(struct ubi_device *ubi, struct ubi_volume *vol,
 	uint64_t tmp;
 	struct ubi_vtbl_record vtbl_rec;
 
-	dbg_msg("clear update marker for volume %d", vol->vol_id);
+	dbg_gen("clear update marker for volume %d", vol->vol_id);
 
 	memcpy(&vtbl_rec, &ubi->vtbl[vol->vol_id],
 	       sizeof(struct ubi_vtbl_record));
@@ -133,7 +133,7 @@ int ubi_start_update(struct ubi_device *ubi, struct ubi_volume *vol,
 	int i, err;
 	uint64_t tmp;
 
-	dbg_msg("start update of volume %d, %llu bytes", vol->vol_id, bytes);
+	dbg_gen("start update of volume %d, %llu bytes", vol->vol_id, bytes);
 	ubi_assert(!vol->updating && !vol->changing_leb);
 	vol->updating = 1;
 
@@ -183,7 +183,7 @@ int ubi_start_leb_change(struct ubi_device *ubi, struct ubi_volume *vol,
 {
 	ubi_assert(!vol->updating && !vol->changing_leb);
 
-	dbg_msg("start changing LEB %d:%d, %u bytes",
+	dbg_gen("start changing LEB %d:%d, %u bytes",
 		vol->vol_id, req->lnum, req->bytes);
 	if (req->bytes == 0)
 		return ubi_eba_atomic_leb_change(ubi, vol, req->lnum, NULL, 0,
@@ -242,7 +242,7 @@ static int write_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
 		memset(buf + len, 0xFF, l - len);
 		len = ubi_calc_data_len(ubi, buf, l);
 		if (len == 0) {
-			dbg_msg("all %d bytes contain 0xFF - skip", len);
+			dbg_gen("all %d bytes contain 0xFF - skip", len);
 			return 0;
 		}
 
@@ -283,7 +283,7 @@ int ubi_more_update_data(struct ubi_device *ubi, struct ubi_volume *vol,
 	uint64_t tmp;
 	int lnum, offs, err = 0, len, to_write = count;
 
-	dbg_msg("write %d of %lld bytes, %lld already passed",
+	dbg_gen("write %d of %lld bytes, %lld already passed",
 		count, vol->upd_bytes, vol->upd_received);
 
 	if (ubi->ro_mode)
@@ -400,7 +400,7 @@ int ubi_more_leb_change_data(struct ubi_device *ubi, struct ubi_volume *vol,
 {
 	int err;
 
-	dbg_msg("write %d of %lld bytes, %lld already passed",
+	dbg_gen("write %d of %lld bytes, %lld already passed",
 		count, vol->upd_bytes, vol->upd_received);
 
 	if (ubi->ro_mode)
diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c
index bfa7c5d2e064..2cd886a5adac 100644
--- a/drivers/mtd/ubi/vmt.c
+++ b/drivers/mtd/ubi/vmt.c
@@ -28,9 +28,9 @@
 #include "ubi.h"
 
 #ifdef CONFIG_MTD_UBI_DEBUG_PARANOID
-static void paranoid_check_volumes(struct ubi_device *ubi);
+static int paranoid_check_volumes(struct ubi_device *ubi);
 #else
-#define paranoid_check_volumes(ubi)
+#define paranoid_check_volumes(ubi) 0
 #endif
 
 static ssize_t vol_attribute_show(struct device *dev,
@@ -218,7 +218,7 @@ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req)
 	spin_lock(&ubi->volumes_lock);
 	if (vol_id == UBI_VOL_NUM_AUTO) {
 		/* Find unused volume ID */
-		dbg_msg("search for vacant volume ID");
+		dbg_gen("search for vacant volume ID");
 		for (i = 0; i < ubi->vtbl_slots; i++)
 			if (!ubi->volumes[i]) {
 				vol_id = i;
@@ -233,7 +233,7 @@ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req)
 		req->vol_id = vol_id;
 	}
 
-	dbg_msg("volume ID %d, %llu bytes, type %d, name %s",
+	dbg_gen("volume ID %d, %llu bytes, type %d, name %s",
 		vol_id, (unsigned long long)req->bytes,
 		(int)req->vol_type, req->name);
 
@@ -361,8 +361,8 @@ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req)
 	ubi->vol_count += 1;
 	spin_unlock(&ubi->volumes_lock);
 
-	paranoid_check_volumes(ubi);
-	return 0;
+	err = paranoid_check_volumes(ubi);
+	return err;
 
 out_sysfs:
 	/*
@@ -414,7 +414,7 @@ int ubi_remove_volume(struct ubi_volume_desc *desc)
 	struct ubi_device *ubi = vol->ubi;
 	int i, err, vol_id = vol->vol_id, reserved_pebs = vol->reserved_pebs;
 
-	dbg_msg("remove UBI volume %d", vol_id);
+	dbg_gen("remove UBI volume %d", vol_id);
 	ubi_assert(desc->mode == UBI_EXCLUSIVE);
 	ubi_assert(vol == ubi->volumes[vol_id]);
 
@@ -465,8 +465,8 @@ int ubi_remove_volume(struct ubi_volume_desc *desc)
 	ubi->vol_count -= 1;
 	spin_unlock(&ubi->volumes_lock);
 
-	paranoid_check_volumes(ubi);
-	return 0;
+	err = paranoid_check_volumes(ubi);
+	return err;
 
 out_err:
 	ubi_err("cannot remove volume %d, error %d", vol_id, err);
@@ -497,7 +497,7 @@ int ubi_resize_volume(struct ubi_volume_desc *desc, int reserved_pebs)
 	if (ubi->ro_mode)
 		return -EROFS;
 
-	dbg_msg("re-size volume %d to from %d to %d PEBs",
+	dbg_gen("re-size volume %d to from %d to %d PEBs",
 		vol_id, vol->reserved_pebs, reserved_pebs);
 
 	if (vol->vol_type == UBI_STATIC_VOLUME &&
@@ -586,8 +586,8 @@ int ubi_resize_volume(struct ubi_volume_desc *desc, int reserved_pebs)
 			(long long)vol->used_ebs * vol->usable_leb_size;
 	}
 
-	paranoid_check_volumes(ubi);
-	return 0;
+	err = paranoid_check_volumes(ubi);
+	return err;
 
 out_acc:
 	if (pebs > 0) {
@@ -615,8 +615,7 @@ int ubi_add_volume(struct ubi_device *ubi, struct ubi_volume *vol)
 	int err, vol_id = vol->vol_id;
 	dev_t dev;
 
-	dbg_msg("add volume %d", vol_id);
-	ubi_dbg_dump_vol_info(vol);
+	dbg_gen("add volume %d", vol_id);
 
 	/* Register character device for the volume */
 	cdev_init(&vol->cdev, &ubi_vol_cdev_operations);
@@ -650,8 +649,8 @@ int ubi_add_volume(struct ubi_device *ubi, struct ubi_volume *vol)
 		return err;
 	}
 
-	paranoid_check_volumes(ubi);
-	return 0;
+	err = paranoid_check_volumes(ubi);
+	return err;
 
 out_gluebi:
 	err = ubi_destroy_gluebi(vol);
@@ -672,7 +671,7 @@ void ubi_free_volume(struct ubi_device *ubi, struct ubi_volume *vol)
 {
 	int err;
 
-	dbg_msg("free volume %d", vol->vol_id);
+	dbg_gen("free volume %d", vol->vol_id);
 
 	ubi->volumes[vol->vol_id] = NULL;
 	err = ubi_destroy_gluebi(vol);
@@ -686,8 +685,10 @@ void ubi_free_volume(struct ubi_device *ubi, struct ubi_volume *vol)
  * paranoid_check_volume - check volume information.
  * @ubi: UBI device description object
  * @vol_id: volume ID
+ *
+ * Returns zero if volume is all right and a a negative error code if not.
  */
-static void paranoid_check_volume(struct ubi_device *ubi, int vol_id)
+static int paranoid_check_volume(struct ubi_device *ubi, int vol_id)
 {
 	int idx = vol_id2idx(ubi, vol_id);
 	int reserved_pebs, alignment, data_pad, vol_type, name_len, upd_marker;
@@ -705,16 +706,7 @@ static void paranoid_check_volume(struct ubi_device *ubi, int vol_id)
 			goto fail;
 		}
 		spin_unlock(&ubi->volumes_lock);
-		return;
-	}
-
-	if (vol->exclusive) {
-		/*
-		 * The volume may be being created at the moment, do not check
-		 * it (e.g., it may be in the middle of ubi_create_volume().
-		 */
-		spin_unlock(&ubi->volumes_lock);
-		return;
+		return 0;
 	}
 
 	if (vol->reserved_pebs < 0 || vol->alignment < 0 || vol->data_pad < 0 ||
@@ -830,25 +822,34 @@ static void paranoid_check_volume(struct ubi_device *ubi, int vol_id)
 	}
 
 	spin_unlock(&ubi->volumes_lock);
-	return;
+	return 0;
 
 fail:
 	ubi_err("paranoid check failed for volume %d", vol_id);
-	ubi_dbg_dump_vol_info(vol);
-	ubi_dbg_dump_vtbl_record(&ubi->vtbl[vol_id], vol_id);
+	if (vol) {
+		ubi_dbg_dump_vol_info(vol);
+		ubi_dbg_dump_vtbl_record(&ubi->vtbl[vol_id], vol_id);
+	}
 	spin_unlock(&ubi->volumes_lock);
-	BUG();
+	return -EINVAL;
 }
 
 /**
  * paranoid_check_volumes - check information about all volumes.
  * @ubi: UBI device description object
+ *
+ * Returns zero if volumes are all right and a a negative error code if not.
  */
-static void paranoid_check_volumes(struct ubi_device *ubi)
+static int paranoid_check_volumes(struct ubi_device *ubi)
 {
-	int i;
+	int i, err = 0;
 
-	for (i = 0; i < ubi->vtbl_slots; i++)
-		paranoid_check_volume(ubi, i);
+	for (i = 0; i < ubi->vtbl_slots; i++) {
+		err = paranoid_check_volume(ubi, i);
+		if (err)
+			break;
+	}
+
+	return err;
 }
 #endif
diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c
index d9af11a8682b..05fb72fd268f 100644
--- a/drivers/mtd/ubi/vtbl.c
+++ b/drivers/mtd/ubi/vtbl.c
@@ -371,7 +371,7 @@ static struct ubi_vtbl_record *process_lvol(struct ubi_device *ubi,
 	 *    to LEB 0.
 	 */
 
-	dbg_msg("check layout volume");
+	dbg_gen("check layout volume");
 
 	/* Read both LEB 0 and LEB 1 into memory */
 	ubi_rb_for_each_entry(rb, seb, &sv->root, u.rb) {

From f40ac9cdf6991287f19bdafe9b0752ee40137908 Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Sun, 13 Jul 2008 21:47:47 +0300
Subject: [PATCH 0184/1171] UBI: implement multiple volumes rename

Quite useful ioctl which allows to make atomic system upgrades.
The idea belongs to Richard Titmuss <richard_titmuss@logitech.com>

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/build.c |   1 +
 drivers/mtd/ubi/cdev.c  | 188 +++++++++++++++++++++++++++++++++++++++-
 drivers/mtd/ubi/ubi.h   |  33 ++++++-
 drivers/mtd/ubi/vmt.c   |  57 ++++++++++--
 drivers/mtd/ubi/vtbl.c  |  51 +++++++++++
 include/mtd/ubi-user.h  |  60 ++++++++++++-
 6 files changed, 375 insertions(+), 15 deletions(-)

diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c
index 7210e1da1fcb..4418a2369b56 100644
--- a/drivers/mtd/ubi/build.c
+++ b/drivers/mtd/ubi/build.c
@@ -806,6 +806,7 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset)
 
 	mutex_init(&ubi->buf_mutex);
 	mutex_init(&ubi->ckvol_mutex);
+	mutex_init(&ubi->mult_mutex);
 	mutex_init(&ubi->volumes_mutex);
 	spin_lock_init(&ubi->volumes_lock);
 
diff --git a/drivers/mtd/ubi/cdev.c b/drivers/mtd/ubi/cdev.c
index 7c19918cc914..bc8199c6a9f5 100644
--- a/drivers/mtd/ubi/cdev.c
+++ b/drivers/mtd/ubi/cdev.c
@@ -605,6 +605,166 @@ static int verify_rsvol_req(const struct ubi_device *ubi,
 	return 0;
 }
 
+/**
+ * rename_volumes - rename UBI volumes.
+ * @ubi: UBI device description object
+ * @req: volumes re-name request
+ *
+ * This is a helper function for the volume re-name IOCTL which validates the
+ * the request, opens the volume and calls corresponding volumes management
+ * function. Returns zero in case of success and a negative error code in case
+ * of failure.
+ */
+static int rename_volumes(struct ubi_device *ubi,
+			  struct ubi_rnvol_req *req)
+{
+	int i, n, err;
+	struct list_head rename_list;
+	struct ubi_rename_entry *re, *re1;
+
+	if (req->count < 0 || req->count > UBI_MAX_RNVOL)
+		return -EINVAL;
+
+	if (req->count == 0)
+		return 0;
+
+	/* Validate volume IDs and names in the request */
+	for (i = 0; i < req->count; i++) {
+		if (req->ents[i].vol_id < 0 ||
+		    req->ents[i].vol_id >= ubi->vtbl_slots)
+			return -EINVAL;
+		if (req->ents[i].name_len < 0)
+			return -EINVAL;
+		if (req->ents[i].name_len > UBI_VOL_NAME_MAX)
+			return -ENAMETOOLONG;
+		req->ents[i].name[req->ents[i].name_len] = '\0';
+		n = strlen(req->ents[i].name);
+		if (n != req->ents[i].name_len)
+			err = -EINVAL;
+	}
+
+	/* Make sure volume IDs and names are unique */
+	for (i = 0; i < req->count - 1; i++) {
+		for (n = i + 1; n < req->count; n++) {
+			if (req->ents[i].vol_id == req->ents[n].vol_id) {
+				dbg_err("duplicated volume id %d",
+					req->ents[i].vol_id);
+				return -EINVAL;
+			}
+			if (!strcmp(req->ents[i].name, req->ents[n].name)) {
+				dbg_err("duplicated volume name \"%s\"",
+					req->ents[i].name);
+				return -EINVAL;
+			}
+		}
+	}
+
+	/* Create the re-name list */
+	INIT_LIST_HEAD(&rename_list);
+	for (i = 0; i < req->count; i++) {
+		int vol_id = req->ents[i].vol_id;
+		int name_len = req->ents[i].name_len;
+		const char *name = req->ents[i].name;
+
+		re = kzalloc(sizeof(struct ubi_rename_entry), GFP_KERNEL);
+		if (!re) {
+			err = -ENOMEM;
+			goto out_free;
+		}
+
+		re->desc = ubi_open_volume(ubi->ubi_num, vol_id, UBI_EXCLUSIVE);
+		if (IS_ERR(re->desc)) {
+			err = PTR_ERR(re->desc);
+			dbg_err("cannot open volume %d, error %d", vol_id, err);
+			kfree(re);
+			goto out_free;
+		}
+
+		/* Skip this re-naming if the name does not really change */
+		if (re->desc->vol->name_len == name_len &&
+		    !memcmp(re->desc->vol->name, name, name_len)) {
+			ubi_close_volume(re->desc);
+			kfree(re);
+			continue;
+		}
+
+		re->new_name_len = name_len;
+		memcpy(re->new_name, name, name_len);
+		list_add_tail(&re->list, &rename_list);
+		dbg_msg("will rename volume %d from \"%s\" to \"%s\"",
+			vol_id, re->desc->vol->name, name);
+	}
+
+	if (list_empty(&rename_list))
+		return 0;
+
+	/* Find out the volumes which have to be removed */
+	list_for_each_entry(re, &rename_list, list) {
+		struct ubi_volume_desc *desc;
+		int no_remove_needed = 0;
+
+		/*
+		 * Volume @re->vol_id is going to be re-named to
+		 * @re->new_name, while its current name is @name. If a volume
+		 * with name @re->new_name currently exists, it has to be
+		 * removed, unless it is also re-named in the request (@req).
+		 */
+		list_for_each_entry(re1, &rename_list, list) {
+			if (re->new_name_len == re1->desc->vol->name_len &&
+			    !memcmp(re->new_name, re1->desc->vol->name,
+				    re1->desc->vol->name_len)) {
+				no_remove_needed = 1;
+				break;
+			}
+		}
+
+		if (no_remove_needed)
+			continue;
+
+		/*
+		 * It seems we need to remove volume with name @re->new_name,
+		 * if it exists.
+		 */
+		desc = ubi_open_volume_nm(ubi->ubi_num, re->new_name, UBI_EXCLUSIVE);
+		if (IS_ERR(desc)) {
+			err = PTR_ERR(desc);
+			if (err == -ENODEV)
+				/* Re-naming into a non-existing volume name */
+				continue;
+
+			/* The volume exists but busy, or an error occurred */
+			dbg_err("cannot open volume \"%s\", error %d",
+				re->new_name, err);
+			goto out_free;
+		}
+
+		re = kzalloc(sizeof(struct ubi_rename_entry), GFP_KERNEL);
+		if (!re) {
+			err = -ENOMEM;
+			ubi_close_volume(desc);
+			goto out_free;
+		}
+
+		re->remove = 1;
+		re->desc = desc;
+		list_add(&re->list, &rename_list);
+		dbg_msg("will remove volume %d, name \"%s\"",
+			re->desc->vol->vol_id, re->desc->vol->name);
+	}
+
+	mutex_lock(&ubi->volumes_mutex);
+	err = ubi_rename_volumes(ubi, &rename_list);
+	mutex_unlock(&ubi->volumes_mutex);
+
+out_free:
+	list_for_each_entry_safe(re, re1, &rename_list, list) {
+		ubi_close_volume(re->desc);
+		list_del(&re->list);
+		kfree(re);
+	}
+	return err;
+}
+
 static int ubi_cdev_ioctl(struct inode *inode, struct file *file,
 			  unsigned int cmd, unsigned long arg)
 {
@@ -670,7 +830,7 @@ static int ubi_cdev_ioctl(struct inode *inode, struct file *file,
 		}
 
 		mutex_lock(&ubi->volumes_mutex);
-		err = ubi_remove_volume(desc);
+		err = ubi_remove_volume(desc, 0);
 		mutex_unlock(&ubi->volumes_mutex);
 
 		/*
@@ -717,6 +877,32 @@ static int ubi_cdev_ioctl(struct inode *inode, struct file *file,
 		break;
 	}
 
+	/* Re-name volumes command */
+	case UBI_IOCRNVOL:
+	{
+		struct ubi_rnvol_req *req;
+
+		dbg_msg("re-name volumes");
+		req = kmalloc(sizeof(struct ubi_rnvol_req), GFP_KERNEL);
+		if (!req) {
+			err = -ENOMEM;
+			break;
+		};
+
+		err = copy_from_user(req, argp, sizeof(struct ubi_rnvol_req));
+		if (err) {
+			err = -EFAULT;
+			kfree(req);
+			break;
+		}
+
+		mutex_lock(&ubi->mult_mutex);
+		err = rename_volumes(ubi, req);
+		mutex_unlock(&ubi->mult_mutex);
+		kfree(req);
+		break;
+	}
+
 	default:
 		err = -ENOTTY;
 		break;
diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h
index 1fc32c863b78..274c67916b34 100644
--- a/drivers/mtd/ubi/ubi.h
+++ b/drivers/mtd/ubi/ubi.h
@@ -131,6 +131,27 @@ struct ubi_ltree_entry {
 	struct rw_semaphore mutex;
 };
 
+/**
+ * struct ubi_rename_entry - volume re-name description data structure.
+ * @new_name_len: new volume name length
+ * @new_name: new volume name
+ * @remove: if not zero, this volume should be removed, not re-named
+ * @desc: descriptor of the volume
+ * @list: links re-name entries into a list
+ *
+ * This data structure is utilized in the multiple volume re-name code. Namely,
+ * UBI first creates a list of &struct ubi_rename_entry objects from the
+ * &struct ubi_rnvol_req request object, and then utilizes this list to do all
+ * the job.
+ */
+struct ubi_rename_entry {
+	int new_name_len;
+	char new_name[UBI_VOL_NAME_MAX + 1];
+	int remove;
+	struct ubi_volume_desc *desc;
+	struct list_head list;
+};
+
 struct ubi_volume_desc;
 
 /**
@@ -206,7 +227,7 @@ struct ubi_volume {
 	int alignment;
 	int data_pad;
 	int name_len;
-	char name[UBI_VOL_NAME_MAX+1];
+	char name[UBI_VOL_NAME_MAX + 1];
 
 	int upd_ebs;
 	int ch_lnum;
@@ -272,7 +293,7 @@ struct ubi_wl_entry;
  * @vtbl_size: size of the volume table in bytes
  * @vtbl: in-RAM volume table copy
  * @volumes_mutex: protects on-flash volume table and serializes volume
- *                 changes, like creation, deletion, update, resize
+ *                 changes, like creation, deletion, update, re-size and re-name
  *
  * @max_ec: current highest erase counter value
  * @mean_ec: current mean erase counter value
@@ -330,6 +351,8 @@ struct ubi_wl_entry;
  * @peb_buf1: a buffer of PEB size used for different purposes
  * @peb_buf2: another buffer of PEB size used for different purposes
  * @buf_mutex: proptects @peb_buf1 and @peb_buf2
+ * @ckvol_mutex: serializes static volume checking when opening
+ * @mult_mutex: serializes operations on multiple volumes, like re-nameing
  * @dbg_peb_buf: buffer of PEB size used for debugging
  * @dbg_buf_mutex: proptects @dbg_peb_buf
  */
@@ -410,6 +433,7 @@ struct ubi_device {
 	void *peb_buf2;
 	struct mutex buf_mutex;
 	struct mutex ckvol_mutex;
+	struct mutex mult_mutex;
 #ifdef CONFIG_MTD_UBI_DEBUG
 	void *dbg_peb_buf;
 	struct mutex dbg_buf_mutex;
@@ -426,12 +450,15 @@ extern struct mutex ubi_devices_mutex;
 /* vtbl.c */
 int ubi_change_vtbl_record(struct ubi_device *ubi, int idx,
 			   struct ubi_vtbl_record *vtbl_rec);
+int ubi_vtbl_rename_volumes(struct ubi_device *ubi,
+			    struct list_head *rename_list);
 int ubi_read_volume_table(struct ubi_device *ubi, struct ubi_scan_info *si);
 
 /* vmt.c */
 int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req);
-int ubi_remove_volume(struct ubi_volume_desc *desc);
+int ubi_remove_volume(struct ubi_volume_desc *desc, int no_vtbl);
 int ubi_resize_volume(struct ubi_volume_desc *desc, int reserved_pebs);
+int ubi_rename_volumes(struct ubi_device *ubi, struct list_head *rename_list);
 int ubi_add_volume(struct ubi_device *ubi, struct ubi_volume *vol);
 void ubi_free_volume(struct ubi_device *ubi, struct ubi_volume *vol);
 
diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c
index 2cd886a5adac..4be4014c70df 100644
--- a/drivers/mtd/ubi/vmt.c
+++ b/drivers/mtd/ubi/vmt.c
@@ -402,13 +402,14 @@ out_unlock:
 /**
  * ubi_remove_volume - remove volume.
  * @desc: volume descriptor
+ * @no_vtbl: do not change volume table if not zero
  *
  * This function removes volume described by @desc. The volume has to be opened
  * in "exclusive" mode. Returns zero in case of success and a negative error
  * code in case of failure. The caller has to have the @ubi->volumes_mutex
  * locked.
  */
-int ubi_remove_volume(struct ubi_volume_desc *desc)
+int ubi_remove_volume(struct ubi_volume_desc *desc, int no_vtbl)
 {
 	struct ubi_volume *vol = desc->vol;
 	struct ubi_device *ubi = vol->ubi;
@@ -437,9 +438,11 @@ int ubi_remove_volume(struct ubi_volume_desc *desc)
 	if (err)
 		goto out_err;
 
-	err = ubi_change_vtbl_record(ubi, vol_id, NULL);
-	if (err)
-		goto out_err;
+	if (!no_vtbl) {
+		err = ubi_change_vtbl_record(ubi, vol_id, NULL);
+		if (err)
+			goto out_err;
+	}
 
 	for (i = 0; i < vol->reserved_pebs; i++) {
 		err = ubi_eba_unmap_leb(ubi, vol, i);
@@ -465,7 +468,8 @@ int ubi_remove_volume(struct ubi_volume_desc *desc)
 	ubi->vol_count -= 1;
 	spin_unlock(&ubi->volumes_lock);
 
-	err = paranoid_check_volumes(ubi);
+	if (!no_vtbl)
+		err = paranoid_check_volumes(ubi);
 	return err;
 
 out_err:
@@ -601,6 +605,44 @@ out_free:
 	return err;
 }
 
+/**
+ * ubi_rename_volumes - re-name UBI volumes.
+ * @ubi: UBI device description object
+ * @renam_list: list of &struct ubi_rename_entry objects
+ *
+ * This function re-names or removes volumes specified in the re-name list.
+ * Returns zero in case of success and a negative error code in case of
+ * failure.
+ */
+int ubi_rename_volumes(struct ubi_device *ubi, struct list_head *rename_list)
+{
+	int err;
+	struct ubi_rename_entry *re;
+
+	err = ubi_vtbl_rename_volumes(ubi, rename_list);
+	if (err)
+		return err;
+
+	list_for_each_entry(re, rename_list, list) {
+		if (re->remove) {
+			err = ubi_remove_volume(re->desc, 1);
+			if (err)
+				break;
+		} else {
+			struct ubi_volume *vol = re->desc->vol;
+
+			spin_lock(&ubi->volumes_lock);
+			vol->name_len = re->new_name_len;
+			memcpy(vol->name, re->new_name, re->new_name_len + 1);
+			spin_unlock(&ubi->volumes_lock);
+		}
+	}
+
+	if (!err)
+		paranoid_check_volumes(ubi);
+	return err;
+}
+
 /**
  * ubi_add_volume - add volume.
  * @ubi: UBI device description object
@@ -826,10 +868,9 @@ static int paranoid_check_volume(struct ubi_device *ubi, int vol_id)
 
 fail:
 	ubi_err("paranoid check failed for volume %d", vol_id);
-	if (vol) {
+	if (vol)
 		ubi_dbg_dump_vol_info(vol);
-		ubi_dbg_dump_vtbl_record(&ubi->vtbl[vol_id], vol_id);
-	}
+	ubi_dbg_dump_vtbl_record(&ubi->vtbl[vol_id], vol_id);
 	spin_unlock(&ubi->volumes_lock);
 	return -EINVAL;
 }
diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c
index 05fb72fd268f..23c5376234b2 100644
--- a/drivers/mtd/ubi/vtbl.c
+++ b/drivers/mtd/ubi/vtbl.c
@@ -114,6 +114,57 @@ int ubi_change_vtbl_record(struct ubi_device *ubi, int idx,
 	return 0;
 }
 
+/**
+ * ubi_vtbl_rename_volumes - rename UBI volumes in the volume table.
+ * @ubi: UBI device description object
+ * @renam_list: list of &struct ubi_rename_entry objects
+ *
+ * This function re-names multiple volumes specified in @req in the volume
+ * table. Returns zero in case of success and a negative error code in case of
+ * failure.
+ */
+int ubi_vtbl_rename_volumes(struct ubi_device *ubi,
+			    struct list_head *rename_list)
+{
+	int i, err;
+	struct ubi_rename_entry *re;
+	struct ubi_volume *layout_vol;
+
+	list_for_each_entry(re, rename_list, list) {
+		uint32_t crc;
+		struct ubi_volume *vol = re->desc->vol;
+		struct ubi_vtbl_record *vtbl_rec = &ubi->vtbl[vol->vol_id];
+
+		if (re->remove) {
+			memcpy(vtbl_rec, &empty_vtbl_record,
+			       sizeof(struct ubi_vtbl_record));
+			continue;
+		}
+
+		vtbl_rec->name_len = cpu_to_be16(re->new_name_len);
+		memcpy(vtbl_rec->name, re->new_name, re->new_name_len);
+		memset(vtbl_rec->name + re->new_name_len, 0,
+		       UBI_VOL_NAME_MAX + 1 - re->new_name_len);
+		crc = crc32(UBI_CRC32_INIT, vtbl_rec,
+			    UBI_VTBL_RECORD_SIZE_CRC);
+		vtbl_rec->crc = cpu_to_be32(crc);
+	}
+
+	layout_vol = ubi->volumes[vol_id2idx(ubi, UBI_LAYOUT_VOLUME_ID)];
+	for (i = 0; i < UBI_LAYOUT_VOLUME_EBS; i++) {
+		err = ubi_eba_unmap_leb(ubi, layout_vol, i);
+		if (err)
+			return err;
+
+		err = ubi_eba_write_leb(ubi, layout_vol, i, ubi->vtbl, 0,
+					ubi->vtbl_size, UBI_LONGTERM);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
 /**
  * vtbl_check - check if volume table is not corrupted and contains sensible
  *              data.
diff --git a/include/mtd/ubi-user.h b/include/mtd/ubi-user.h
index a7421f130cc0..e8e57c3dfcdb 100644
--- a/include/mtd/ubi-user.h
+++ b/include/mtd/ubi-user.h
@@ -58,6 +58,13 @@
  * device should be used. A &struct ubi_rsvol_req object has to be properly
  * filled and a pointer to it has to be passed to the IOCTL.
  *
+ * UBI volumes re-name
+ * ~~~~~~~~~~~~~~~~~~~
+ *
+ * To re-name several volumes atomically at one go, the %UBI_IOCRNVOL command
+ * of the UBI character device should be used. A &struct ubi_rnvol_req object
+ * has to be properly filled and a pointer to it has to be passed to the IOCTL.
+ *
  * UBI volume update
  * ~~~~~~~~~~~~~~~~~
  *
@@ -104,6 +111,8 @@
 #define UBI_IOCRMVOL _IOW(UBI_IOC_MAGIC, 1, int32_t)
 /* Re-size an UBI volume */
 #define UBI_IOCRSVOL _IOW(UBI_IOC_MAGIC, 2, struct ubi_rsvol_req)
+/* Re-name volumes */
+#define UBI_IOCRNVOL _IOW(UBI_IOC_MAGIC, 3, struct ubi_rnvol_req)
 
 /* IOCTL commands of the UBI control character device */
 
@@ -128,6 +137,9 @@
 /* Maximum MTD device name length supported by UBI */
 #define MAX_UBI_MTD_NAME_LEN 127
 
+/* Maximum amount of UBI volumes that can be re-named at one go */
+#define UBI_MAX_RNVOL 32
+
 /*
  * UBI data type hint constants.
  *
@@ -189,7 +201,7 @@ struct ubi_attach_req {
 	int32_t ubi_num;
 	int32_t mtd_num;
 	int32_t vid_hdr_offset;
-	uint8_t padding[12];
+	int8_t padding[12];
 };
 
 /**
@@ -250,6 +262,48 @@ struct ubi_rsvol_req {
 	int32_t vol_id;
 } __attribute__ ((packed));
 
+/**
+ * struct ubi_rnvol_req - volumes re-name request.
+ * @count: count of volumes to re-name
+ * @padding1:  reserved for future, not used, has to be zeroed
+ * @vol_id: ID of the volume to re-name
+ * @name_len: name length
+ * @padding2:  reserved for future, not used, has to be zeroed
+ * @name: new volume name
+ *
+ * UBI allows to re-name up to %32 volumes at one go. The count of volumes to
+ * re-name is specified in the @count field. The ID of the volumes to re-name
+ * and the new names are specified in the @vol_id and @name fields.
+ *
+ * The UBI volume re-name operation is atomic, which means that should power cut
+ * happen, the volumes will have either old name or new name. So the possible
+ * use-cases of this command is atomic upgrade. Indeed, to upgrade, say, volumes
+ * A and B one may create temporary volumes %A1 and %B1 with the new contents,
+ * then atomically re-name A1->A and B1->B, in which case old %A and %B will
+ * be removed.
+ *
+ * If it is not desirable to remove old A and B, the re-name request has to
+ * contain 4 entries: A1->A, A->A1, B1->B, B->B1, in which case old A1 and B1
+ * become A and B, and old A and B will become A1 and B1.
+ *
+ * It is also OK to request: A1->A, A1->X, B1->B, B->Y, in which case old A1
+ * and B1 become A and B, and old A and B become X and Y.
+ *
+ * In other words, in case of re-naming into an existing volume name, the
+ * existing volume is removed, unless it is re-named as well at the same
+ * re-name request.
+ */
+struct ubi_rnvol_req {
+	int32_t count;
+	int8_t padding1[12];
+	struct {
+		int32_t vol_id;
+		int16_t name_len;
+		int8_t  padding2[2];
+		char    name[UBI_MAX_VOLUME_NAME + 1];
+	} ents[UBI_MAX_RNVOL];
+} __attribute__ ((packed));
+
 /**
  * struct ubi_leb_change_req - a data structure used in atomic logical
  *                             eraseblock change requests.
@@ -261,8 +315,8 @@ struct ubi_rsvol_req {
 struct ubi_leb_change_req {
 	int32_t lnum;
 	int32_t bytes;
-	uint8_t dtype;
-	uint8_t padding[7];
+	int8_t  dtype;
+	int8_t  padding[7];
 } __attribute__ ((packed));
 
 #endif /* __UBI_USER_H__ */

From 8c1e6ee10bd87d70faada065a8d1f70732c17382 Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Fri, 18 Jul 2008 12:20:23 +0300
Subject: [PATCH 0185/1171] UBI: rework scrubbing messages

If bit-flips happen often, UBI prints to many messages. Lessen
the amount by only printing the messages when the PEB has been
scrubbed. Also, print torturing messages.

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/io.c | 8 +++++++-
 drivers/mtd/ubi/wl.c | 6 +++++-
 2 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/ubi/io.c b/drivers/mtd/ubi/io.c
index 27b9c2c2fc60..2bebb39d19b3 100644
--- a/drivers/mtd/ubi/io.c
+++ b/drivers/mtd/ubi/io.c
@@ -156,8 +156,12 @@ retry:
 			/*
 			 * -EUCLEAN is reported if there was a bit-flip which
 			 * was corrected, so this is harmless.
+			 *
+			 * We do not report about it here unless debugging is
+			 * enabled. A corresponding message will be printed
+			 * later, when it is has been scrubbed.
 			 */
-			ubi_msg("fixable bit-flip detected at PEB %d", pnum);
+			dbg_msg("fixable bit-flip detected at PEB %d", pnum);
 			ubi_assert(len == read);
 			return UBI_IO_BITFLIPS;
 		}
@@ -391,6 +395,7 @@ static int torture_peb(struct ubi_device *ubi, int pnum)
 {
 	int err, i, patt_count;
 
+	ubi_msg("run torture test for PEB %d", pnum);
 	patt_count = ARRAY_SIZE(patterns);
 	ubi_assert(patt_count > 0);
 
@@ -434,6 +439,7 @@ static int torture_peb(struct ubi_device *ubi, int pnum)
 	}
 
 	err = patt_count;
+	ubi_msg("PEB %d passed torture test, do not mark it a bad", pnum);
 
 out:
 	mutex_unlock(&ubi->buf_mutex);
diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c
index 761952ba125b..6821952bcdb8 100644
--- a/drivers/mtd/ubi/wl.c
+++ b/drivers/mtd/ubi/wl.c
@@ -873,6 +873,10 @@ static int wear_leveling_worker(struct ubi_device *ubi, struct ubi_work *wrk,
 	}
 
 	ubi_free_vid_hdr(ubi, vid_hdr);
+	if (scrubbing && !protect)
+		ubi_msg("scrubbed PEB %d, data moved to PEB %d",
+			e1->pnum, e2->pnum);
+
 	spin_lock(&ubi->wl_lock);
 	if (protect)
 		prot_tree_add(ubi, e1, pe, protect);
@@ -1231,7 +1235,7 @@ int ubi_wl_scrub_peb(struct ubi_device *ubi, int pnum)
 {
 	struct ubi_wl_entry *e;
 
-	ubi_msg("schedule PEB %d for scrubbing", pnum);
+	dbg_msg("schedule PEB %d for scrubbing", pnum);
 
 retry:
 	spin_lock(&ubi->wl_lock);

From 4d88de4beb6f327dfc7c2221eab532dad5b2bb3e Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Fri, 18 Jul 2008 12:42:14 +0300
Subject: [PATCH 0186/1171] UBI: bugfix - do not torture PEB needlessly

This is probably a copy-paste bug - we torture the old PEB
in the atomic LEB change function, but we should not do this.

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/eba.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index 613cd1e51648..e14208152c36 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -906,7 +906,7 @@ retry:
 	}
 
 	if (vol->eba_tbl[lnum] >= 0) {
-		err = ubi_wl_put_peb(ubi, vol->eba_tbl[lnum], 1);
+		err = ubi_wl_put_peb(ubi, vol->eba_tbl[lnum], 0);
 		if (err)
 			goto out_leb_unlock;
 	}

From 9c9ec147709e63e4e8ac6a037c6bb50688ff8e9c Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Fri, 18 Jul 2008 13:19:52 +0300
Subject: [PATCH 0187/1171] UBI: fix checkpatch.pl errors and warnings

Just out or curiousity ran checkpatch.pl for whole UBI,
and discovered there are quite a few of stylistic issues.
Fix them.

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/build.c  |  8 ++--
 drivers/mtd/ubi/cdev.c   |  4 +-
 drivers/mtd/ubi/eba.c    |  7 +--
 drivers/mtd/ubi/gluebi.c |  4 +-
 drivers/mtd/ubi/io.c     |  8 ++--
 drivers/mtd/ubi/scan.c   |  9 ++--
 drivers/mtd/ubi/ubi.h    |  3 +-
 drivers/mtd/ubi/upd.c    |  8 ++--
 drivers/mtd/ubi/vmt.c    |  4 +-
 drivers/mtd/ubi/vtbl.c   | 12 +++---
 drivers/mtd/ubi/wl.c     | 92 +++++++++++++++++++---------------------
 include/mtd/ubi-user.h   | 16 +++----
 12 files changed, 86 insertions(+), 89 deletions(-)

diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c
index 4418a2369b56..535d9a8a6ba8 100644
--- a/drivers/mtd/ubi/build.c
+++ b/drivers/mtd/ubi/build.c
@@ -51,14 +51,13 @@
  * @name: MTD device name or number string
  * @vid_hdr_offs: VID header offset
  */
-struct mtd_dev_param
-{
+struct mtd_dev_param {
 	char name[MTD_PARAM_LEN_MAX];
 	int vid_hdr_offs;
 };
 
 /* Numbers of elements set in the @mtd_dev_param array */
-static int mtd_devs = 0;
+static int mtd_devs;
 
 /* MTD devices specification parameters */
 static struct mtd_dev_param mtd_dev_param[UBI_MAX_DEVICES];
@@ -781,7 +780,8 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset)
 			if (!ubi_devices[ubi_num])
 				break;
 		if (ubi_num == UBI_MAX_DEVICES) {
-			dbg_err("only %d UBI devices may be created", UBI_MAX_DEVICES);
+			dbg_err("only %d UBI devices may be created",
+				UBI_MAX_DEVICES);
 			return -ENFILE;
 		}
 	} else {
diff --git a/drivers/mtd/ubi/cdev.c b/drivers/mtd/ubi/cdev.c
index bc8199c6a9f5..03c759b4eeb5 100644
--- a/drivers/mtd/ubi/cdev.c
+++ b/drivers/mtd/ubi/cdev.c
@@ -39,9 +39,9 @@
 #include <linux/stat.h>
 #include <linux/ioctl.h>
 #include <linux/capability.h>
+#include <linux/uaccess.h>
 #include <linux/smp_lock.h>
 #include <mtd/ubi-user.h>
-#include <asm/uaccess.h>
 #include <asm/div64.h>
 #include "ubi.h"
 
@@ -352,7 +352,7 @@ static ssize_t vol_cdev_direct_write(struct file *file, const char __user *buf,
 }
 
 #else
-#define vol_cdev_direct_write(file, buf, count, offp) -EPERM
+#define vol_cdev_direct_write(file, buf, count, offp) (-EPERM)
 #endif /* CONFIG_MTD_UBI_DEBUG_USERSPACE_IO */
 
 static ssize_t vol_cdev_write(struct file *file, const char __user *buf,
diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c
index e14208152c36..e04bcf1dff87 100644
--- a/drivers/mtd/ubi/eba.c
+++ b/drivers/mtd/ubi/eba.c
@@ -189,9 +189,7 @@ static struct ubi_ltree_entry *ltree_add_entry(struct ubi_device *ubi,
 	le->users += 1;
 	spin_unlock(&ubi->ltree_lock);
 
-	if (le_free)
-		kfree(le_free);
-
+	kfree(le_free);
 	return le;
 }
 
@@ -503,9 +501,8 @@ static int recover_peb(struct ubi_device *ubi, int pnum, int vol_id, int lnum,
 	struct ubi_vid_hdr *vid_hdr;
 
 	vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_NOFS);
-	if (!vid_hdr) {
+	if (!vid_hdr)
 		return -ENOMEM;
-	}
 
 	mutex_lock(&ubi->buf_mutex);
 
diff --git a/drivers/mtd/ubi/gluebi.c b/drivers/mtd/ubi/gluebi.c
index 49f52dceea93..605812bb0b1a 100644
--- a/drivers/mtd/ubi/gluebi.c
+++ b/drivers/mtd/ubi/gluebi.c
@@ -249,8 +249,8 @@ static int gluebi_erase(struct mtd_info *mtd, struct erase_info *instr)
 	if (err)
 		goto out_err;
 
-        instr->state = MTD_ERASE_DONE;
-        mtd_erase_callback(instr);
+	instr->state = MTD_ERASE_DONE;
+	mtd_erase_callback(instr);
 	return 0;
 
 out_err:
diff --git a/drivers/mtd/ubi/io.c b/drivers/mtd/ubi/io.c
index 2bebb39d19b3..a84f0db0a03a 100644
--- a/drivers/mtd/ubi/io.c
+++ b/drivers/mtd/ubi/io.c
@@ -167,8 +167,8 @@ retry:
 		}
 
 		if (read != len && retries++ < UBI_IO_RETRIES) {
-			dbg_io("error %d while reading %d bytes from PEB %d:%d, "
-			       "read only %zd bytes, retry",
+			dbg_io("error %d while reading %d bytes from PEB %d:%d,"
+			       " read only %zd bytes, retry",
 			       err, len, pnum, offset, read);
 			yield();
 			goto retry;
@@ -705,8 +705,8 @@ int ubi_io_read_ec_hdr(struct ubi_device *ubi, int pnum,
 
 	if (hdr_crc != crc) {
 		if (verbose) {
-			ubi_warn("bad EC header CRC at PEB %d, calculated %#08x,"
-				 " read %#08x", pnum, crc, hdr_crc);
+			ubi_warn("bad EC header CRC at PEB %d, calculated "
+				 "%#08x, read %#08x", pnum, crc, hdr_crc);
 			ubi_dbg_dump_ec_hdr(ec_hdr);
 		}
 		return UBI_IO_BAD_EC_HDR;
diff --git a/drivers/mtd/ubi/scan.c b/drivers/mtd/ubi/scan.c
index 40eca9ce5fab..0bb7488862d2 100644
--- a/drivers/mtd/ubi/scan.c
+++ b/drivers/mtd/ubi/scan.c
@@ -248,7 +248,8 @@ static int compare_lebs(struct ubi_device *ubi, const struct ubi_scan_leb *seb,
 	unsigned long long sqnum2 = be64_to_cpu(vid_hdr->sqnum);
 
 	if (seb->sqnum == 0 && sqnum2 == 0) {
-		long long abs, v1 = seb->leb_ver, v2 = be32_to_cpu(vid_hdr->leb_ver);
+		long long abs;
+		long long v1 = seb->leb_ver, v2 = be32_to_cpu(vid_hdr->leb_ver);
 
 		/*
 		 * UBI constantly increases the logical eraseblock version
@@ -752,7 +753,8 @@ struct ubi_scan_leb *ubi_scan_get_free_peb(struct ubi_device *ubi,
  * This function returns a zero if the physical eraseblock was successfully
  * handled and a negative error code in case of failure.
  */
-static int process_eb(struct ubi_device *ubi, struct ubi_scan_info *si, int pnum)
+static int process_eb(struct ubi_device *ubi, struct ubi_scan_info *si,
+		      int pnum)
 {
 	long long uninitialized_var(ec);
 	int err, bitflips = 0, vol_id, ec_corr = 0;
@@ -1301,8 +1303,7 @@ static int paranoid_check_si(struct ubi_device *ubi, struct ubi_scan_info *si)
 		if (err < 0) {
 			kfree(buf);
 			return err;
-		}
-		else if (err)
+		} else if (err)
 			buf[pnum] = 1;
 	}
 
diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h
index 274c67916b34..14a5596d2d9e 100644
--- a/drivers/mtd/ubi/ubi.h
+++ b/drivers/mtd/ubi/ubi.h
@@ -473,7 +473,8 @@ int ubi_more_leb_change_data(struct ubi_device *ubi, struct ubi_volume *vol,
 			     const void __user *buf, int count);
 
 /* misc.c */
-int ubi_calc_data_len(const struct ubi_device *ubi, const void *buf, int length);
+int ubi_calc_data_len(const struct ubi_device *ubi, const void *buf,
+		      int length);
 int ubi_check_volume(struct ubi_device *ubi, int vol_id);
 void ubi_calculate_reserved(struct ubi_device *ubi);
 
diff --git a/drivers/mtd/ubi/upd.c b/drivers/mtd/ubi/upd.c
index 1230a5e1b530..3b8beb8545c4 100644
--- a/drivers/mtd/ubi/upd.c
+++ b/drivers/mtd/ubi/upd.c
@@ -39,7 +39,7 @@
  */
 
 #include <linux/err.h>
-#include <asm/uaccess.h>
+#include <linux/uaccess.h>
 #include <asm/div64.h>
 #include "ubi.h"
 
@@ -246,7 +246,8 @@ static int write_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
 			return 0;
 		}
 
-		err = ubi_eba_write_leb(ubi, vol, lnum, buf, 0, len, UBI_UNKNOWN);
+		err = ubi_eba_write_leb(ubi, vol, lnum, buf, 0, len,
+					UBI_UNKNOWN);
 	} else {
 		/*
 		 * When writing static volume, and this is the last logical
@@ -418,7 +419,8 @@ int ubi_more_leb_change_data(struct ubi_device *ubi, struct ubi_volume *vol,
 	if (vol->upd_received == vol->upd_bytes) {
 		int len = ALIGN((int)vol->upd_bytes, ubi->min_io_size);
 
-		memset(vol->upd_buf + vol->upd_bytes, 0xFF, len - vol->upd_bytes);
+		memset(vol->upd_buf + vol->upd_bytes, 0xFF,
+		       len - vol->upd_bytes);
 		len = ubi_calc_data_len(ubi, vol->upd_buf, len);
 		err = ubi_eba_atomic_leb_change(ubi, vol, vol->ch_lnum,
 						vol->upd_buf, len, UBI_UNKNOWN);
diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c
index 4be4014c70df..852482d8b18f 100644
--- a/drivers/mtd/ubi/vmt.c
+++ b/drivers/mtd/ubi/vmt.c
@@ -253,7 +253,7 @@ int ubi_create_volume(struct ubi_device *ubi, struct ubi_mkvol_req *req)
 			goto out_unlock;
 		}
 
-        /* Calculate how many eraseblocks are requested */
+	/* Calculate how many eraseblocks are requested */
 	vol->usable_leb_size = ubi->leb_size - ubi->leb_size % req->alignment;
 	bytes = req->bytes;
 	if (do_div(bytes, vol->usable_leb_size))
@@ -858,7 +858,7 @@ static int paranoid_check_volume(struct ubi_device *ubi, int vol_id)
 
 	if (alignment != vol->alignment || data_pad != vol->data_pad ||
 	    upd_marker != vol->upd_marker || vol_type != vol->vol_type ||
-	    name_len!= vol->name_len || strncmp(name, vol->name, name_len)) {
+	    name_len != vol->name_len || strncmp(name, vol->name, name_len)) {
 		ubi_err("volume info is different");
 		goto fail;
 	}
diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c
index 23c5376234b2..10c22257f601 100644
--- a/drivers/mtd/ubi/vtbl.c
+++ b/drivers/mtd/ubi/vtbl.c
@@ -461,7 +461,8 @@ static struct ubi_vtbl_record *process_lvol(struct ubi_device *ubi,
 	if (!leb_corrupted[0]) {
 		/* LEB 0 is OK */
 		if (leb[1])
-			leb_corrupted[1] = memcmp(leb[0], leb[1], ubi->vtbl_size);
+			leb_corrupted[1] = memcmp(leb[0], leb[1],
+						  ubi->vtbl_size);
 		if (leb_corrupted[1]) {
 			ubi_warn("volume table copy #2 is corrupted");
 			err = create_vtbl(ubi, si, 1, leb[0]);
@@ -859,11 +860,10 @@ int ubi_read_volume_table(struct ubi_device *ubi, struct ubi_scan_info *si)
 
 out_free:
 	vfree(ubi->vtbl);
-	for (i = 0; i < ubi->vtbl_slots + UBI_INT_VOL_COUNT; i++)
-		if (ubi->volumes[i]) {
-			kfree(ubi->volumes[i]);
-			ubi->volumes[i] = NULL;
-		}
+	for (i = 0; i < ubi->vtbl_slots + UBI_INT_VOL_COUNT; i++) {
+		kfree(ubi->volumes[i]);
+		ubi->volumes[i] = NULL;
+	}
 	return err;
 }
 
diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c
index 6821952bcdb8..2a5d2a0e14ad 100644
--- a/drivers/mtd/ubi/wl.c
+++ b/drivers/mtd/ubi/wl.c
@@ -475,52 +475,47 @@ retry:
 	}
 
 	switch (dtype) {
-		case UBI_LONGTERM:
-			/*
-			 * For long term data we pick a physical eraseblock
-			 * with high erase counter. But the highest erase
-			 * counter we can pick is bounded by the the lowest
-			 * erase counter plus %WL_FREE_MAX_DIFF.
-			 */
-			e = find_wl_entry(&ubi->free, WL_FREE_MAX_DIFF);
-			protect = LT_PROTECTION;
-			break;
-		case UBI_UNKNOWN:
-			/*
-			 * For unknown data we pick a physical eraseblock with
-			 * medium erase counter. But we by no means can pick a
-			 * physical eraseblock with erase counter greater or
-			 * equivalent than the lowest erase counter plus
-			 * %WL_FREE_MAX_DIFF.
-			 */
-			first = rb_entry(rb_first(&ubi->free),
-					 struct ubi_wl_entry, rb);
-			last = rb_entry(rb_last(&ubi->free),
-					struct ubi_wl_entry, rb);
+	case UBI_LONGTERM:
+		/*
+		 * For long term data we pick a physical eraseblock with high
+		 * erase counter. But the highest erase counter we can pick is
+		 * bounded by the the lowest erase counter plus
+		 * %WL_FREE_MAX_DIFF.
+		 */
+		e = find_wl_entry(&ubi->free, WL_FREE_MAX_DIFF);
+		protect = LT_PROTECTION;
+		break;
+	case UBI_UNKNOWN:
+		/*
+		 * For unknown data we pick a physical eraseblock with medium
+		 * erase counter. But we by no means can pick a physical
+		 * eraseblock with erase counter greater or equivalent than the
+		 * lowest erase counter plus %WL_FREE_MAX_DIFF.
+		 */
+		first = rb_entry(rb_first(&ubi->free), struct ubi_wl_entry, rb);
+		last = rb_entry(rb_last(&ubi->free), struct ubi_wl_entry, rb);
 
-			if (last->ec - first->ec < WL_FREE_MAX_DIFF)
-				e = rb_entry(ubi->free.rb_node,
-						struct ubi_wl_entry, rb);
-			else {
-				medium_ec = (first->ec + WL_FREE_MAX_DIFF)/2;
-				e = find_wl_entry(&ubi->free, medium_ec);
-			}
-			protect = U_PROTECTION;
-			break;
-		case UBI_SHORTTERM:
-			/*
-			 * For short term data we pick a physical eraseblock
-			 * with the lowest erase counter as we expect it will
-			 * be erased soon.
-			 */
-			e = rb_entry(rb_first(&ubi->free),
-				     struct ubi_wl_entry, rb);
-			protect = ST_PROTECTION;
-			break;
-		default:
-			protect = 0;
-			e = NULL;
-			BUG();
+		if (last->ec - first->ec < WL_FREE_MAX_DIFF)
+			e = rb_entry(ubi->free.rb_node,
+					struct ubi_wl_entry, rb);
+		else {
+			medium_ec = (first->ec + WL_FREE_MAX_DIFF)/2;
+			e = find_wl_entry(&ubi->free, medium_ec);
+		}
+		protect = U_PROTECTION;
+		break;
+	case UBI_SHORTTERM:
+		/*
+		 * For short term data we pick a physical eraseblock with the
+		 * lowest erase counter as we expect it will be erased soon.
+		 */
+		e = rb_entry(rb_first(&ubi->free), struct ubi_wl_entry, rb);
+		protect = ST_PROTECTION;
+		break;
+	default:
+		protect = 0;
+		e = NULL;
+		BUG();
 	}
 
 	/*
@@ -584,7 +579,8 @@ found:
  * This function returns zero in case of success and a negative error code in
  * case of failure.
  */
-static int sync_erase(struct ubi_device *ubi, struct ubi_wl_entry *e, int torture)
+static int sync_erase(struct ubi_device *ubi, struct ubi_wl_entry *e,
+		      int torture)
 {
 	int err;
 	struct ubi_ec_hdr *ec_hdr;
@@ -1060,8 +1056,8 @@ static int erase_worker(struct ubi_device *ubi, struct ubi_work *wl_wrk,
 		spin_unlock(&ubi->wl_lock);
 
 		/*
-		 * One more erase operation has happened, take care about protected
-		 * physical eraseblocks.
+		 * One more erase operation has happened, take care about
+		 * protected physical eraseblocks.
 		 */
 		check_protection_over(ubi);
 
diff --git a/include/mtd/ubi-user.h b/include/mtd/ubi-user.h
index e8e57c3dfcdb..ccdc562e444e 100644
--- a/include/mtd/ubi-user.h
+++ b/include/mtd/ubi-user.h
@@ -188,14 +188,14 @@ enum {
  * it will be 512 in case of a 2KiB page NAND flash with 4 512-byte sub-pages.
  *
  * But in rare cases, if this optimizes things, the VID header may be placed to
- * a different offset. For example, the boot-loader might do things faster if the
- * VID header sits at the end of the first 2KiB NAND page with 4 sub-pages. As
- * the boot-loader would not normally need to read EC headers (unless it needs
- * UBI in RW mode), it might be faster to calculate ECC. This is weird example,
- * but it real-life example. So, in this example, @vid_hdr_offer would be
- * 2KiB-64 bytes = 1984. Note, that this position is not even 512-bytes
- * aligned, which is OK, as UBI is clever enough to realize this is 4th sub-page
- * of the first page and add needed padding.
+ * a different offset. For example, the boot-loader might do things faster if
+ * the VID header sits at the end of the first 2KiB NAND page with 4 sub-pages.
+ * As the boot-loader would not normally need to read EC headers (unless it
+ * needs UBI in RW mode), it might be faster to calculate ECC. This is weird
+ * example, but it real-life example. So, in this example, @vid_hdr_offer would
+ * be 2KiB-64 bytes = 1984. Note, that this position is not even 512-bytes
+ * aligned, which is OK, as UBI is clever enough to realize this is 4th
+ * sub-page of the first page and add needed padding.
  */
 struct ubi_attach_req {
 	int32_t ubi_num;

From ebaaf1af3e9ef05c4fb7c61e4530c15e1ad10e3b Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Fri, 18 Jul 2008 13:34:32 +0300
Subject: [PATCH 0188/1171] UBI: fix kernel-doc errors and warnings

No functional changes, just tweak comments to make kernel-doc
work fine and stop complaining.

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/build.c |  8 +++-----
 drivers/mtd/ubi/io.c    |  6 ++----
 drivers/mtd/ubi/scan.c  | 18 ++++++------------
 drivers/mtd/ubi/ubi.h   |  1 +
 drivers/mtd/ubi/upd.c   |  2 ++
 drivers/mtd/ubi/vmt.c   |  2 +-
 drivers/mtd/ubi/vtbl.c  |  8 +++-----
 drivers/mtd/ubi/wl.c    | 13 +++++--------
 8 files changed, 23 insertions(+), 35 deletions(-)

diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c
index 535d9a8a6ba8..eba760b3b8c7 100644
--- a/drivers/mtd/ubi/build.c
+++ b/drivers/mtd/ubi/build.c
@@ -159,8 +159,7 @@ void ubi_put_device(struct ubi_device *ubi)
 }
 
 /**
- * ubi_get_by_major - get UBI device description object by character device
- *                    major number.
+ * ubi_get_by_major - get UBI device by character device major number.
  * @major: major number
  *
  * This function is similar to 'ubi_get_device()', but it searches the device
@@ -727,7 +726,7 @@ static int autoresize(struct ubi_device *ubi, int vol_id)
 
 /**
  * ubi_attach_mtd_dev - attach an MTD device.
- * @mtd_dev: MTD device description object
+ * @mtd: MTD device description object
  * @ubi_num: number to assign to the new UBI device
  * @vid_hdr_offset: VID header offset
  *
@@ -1095,8 +1094,7 @@ static void __exit ubi_exit(void)
 module_exit(ubi_exit);
 
 /**
- * bytes_str_to_int - convert a string representing number of bytes to an
- * integer.
+ * bytes_str_to_int - convert a number of bytes string into an integer.
  * @str: the string to convert
  *
  * This function returns positive resulting integer in case of success and a
diff --git a/drivers/mtd/ubi/io.c b/drivers/mtd/ubi/io.c
index a84f0db0a03a..2fb64be44f1b 100644
--- a/drivers/mtd/ubi/io.c
+++ b/drivers/mtd/ubi/io.c
@@ -1101,8 +1101,7 @@ fail:
 }
 
 /**
- * paranoid_check_peb_ec_hdr - check that the erase counter header of a
- * physical eraseblock is in-place and is all right.
+ * paranoid_check_peb_ec_hdr - check erase counter header.
  * @ubi: UBI device description object
  * @pnum: the physical eraseblock number to check
  *
@@ -1180,8 +1179,7 @@ fail:
 }
 
 /**
- * paranoid_check_peb_vid_hdr - check that the volume identifier header of a
- * physical eraseblock is in-place and is all right.
+ * paranoid_check_peb_vid_hdr - check volume identifier header.
  * @ubi: UBI device description object
  * @pnum: the physical eraseblock number to check
  *
diff --git a/drivers/mtd/ubi/scan.c b/drivers/mtd/ubi/scan.c
index 0bb7488862d2..4dfbf27b0656 100644
--- a/drivers/mtd/ubi/scan.c
+++ b/drivers/mtd/ubi/scan.c
@@ -93,8 +93,7 @@ static int add_to_list(struct ubi_scan_info *si, int pnum, int ec,
 }
 
 /**
- * validate_vid_hdr - check that volume identifier header is correct and
- * consistent.
+ * validate_vid_hdr - check volume identifier header.
  * @vid_hdr: the volume identifier header to check
  * @sv: information about the volume this logical eraseblock belongs to
  * @pnum: physical eraseblock number the VID header came from
@@ -380,8 +379,7 @@ out_free_vidh:
 }
 
 /**
- * ubi_scan_add_used - add information about a physical eraseblock to the
- * scanning information.
+ * ubi_scan_add_used - add physical eraseblock to the scanning information.
  * @ubi: UBI device description object
  * @si: scanning information
  * @pnum: the physical eraseblock number
@@ -555,8 +553,7 @@ int ubi_scan_add_used(struct ubi_device *ubi, struct ubi_scan_info *si,
 }
 
 /**
- * ubi_scan_find_sv - find information about a particular volume in the
- * scanning information.
+ * ubi_scan_find_sv - find volume in the scanning information.
  * @si: scanning information
  * @vol_id: the requested volume ID
  *
@@ -585,8 +582,7 @@ struct ubi_scan_volume *ubi_scan_find_sv(const struct ubi_scan_info *si,
 }
 
 /**
- * ubi_scan_find_seb - find information about a particular logical
- * eraseblock in the volume scanning information.
+ * ubi_scan_find_seb - find LEB in the volume scanning information.
  * @sv: a pointer to the volume scanning information
  * @lnum: the requested logical eraseblock
  *
@@ -744,8 +740,7 @@ struct ubi_scan_leb *ubi_scan_get_free_peb(struct ubi_device *ubi,
 }
 
 /**
- * process_eb - read UBI headers, check them and add corresponding data
- * to the scanning information.
+ * process_eb - read, check UBI headers, and add them to scanning information.
  * @ubi: UBI device description object
  * @si: scanning information
  * @pnum: the physical eraseblock number
@@ -1083,8 +1078,7 @@ void ubi_scan_destroy_si(struct ubi_scan_info *si)
 #ifdef CONFIG_MTD_UBI_DEBUG_PARANOID
 
 /**
- * paranoid_check_si - check if the scanning information is correct and
- * consistent.
+ * paranoid_check_si - check the scanning information.
  * @ubi: UBI device description object
  * @si: scanning information
  *
diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h
index 14a5596d2d9e..1c3fa18c26a7 100644
--- a/drivers/mtd/ubi/ubi.h
+++ b/drivers/mtd/ubi/ubi.h
@@ -313,6 +313,7 @@ struct ubi_wl_entry;
  *           @move_to, @move_to_put @erase_pending, @wl_scheduled, and @works
  *           fields
  * @move_mutex: serializes eraseblock moves
+ * @work_sem: sycnhronizes the WL worker with use tasks
  * @wl_scheduled: non-zero if the wear-leveling was scheduled
  * @lookuptbl: a table to quickly find a &struct ubi_wl_entry object for any
  *             physical eraseblock
diff --git a/drivers/mtd/ubi/upd.c b/drivers/mtd/ubi/upd.c
index 3b8beb8545c4..8b89cc18ff0b 100644
--- a/drivers/mtd/ubi/upd.c
+++ b/drivers/mtd/ubi/upd.c
@@ -268,6 +268,7 @@ static int write_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
 
 /**
  * ubi_more_update_data - write more update data.
+ * @ubi: UBI device description object
  * @vol: volume description object
  * @buf: write data (user-space memory buffer)
  * @count: how much bytes to write
@@ -385,6 +386,7 @@ int ubi_more_update_data(struct ubi_device *ubi, struct ubi_volume *vol,
 
 /**
  * ubi_more_leb_change_data - accept more data for atomic LEB change.
+ * @ubi: UBI device description object
  * @vol: volume description object
  * @buf: write data (user-space memory buffer)
  * @count: how much bytes to write
diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c
index 852482d8b18f..d40066833abb 100644
--- a/drivers/mtd/ubi/vmt.c
+++ b/drivers/mtd/ubi/vmt.c
@@ -608,7 +608,7 @@ out_free:
 /**
  * ubi_rename_volumes - re-name UBI volumes.
  * @ubi: UBI device description object
- * @renam_list: list of &struct ubi_rename_entry objects
+ * @rename_list: list of &struct ubi_rename_entry objects
  *
  * This function re-names or removes volumes specified in the re-name list.
  * Returns zero in case of success and a negative error code in case of
diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c
index 10c22257f601..4e1c489a3bae 100644
--- a/drivers/mtd/ubi/vtbl.c
+++ b/drivers/mtd/ubi/vtbl.c
@@ -117,7 +117,7 @@ int ubi_change_vtbl_record(struct ubi_device *ubi, int idx,
 /**
  * ubi_vtbl_rename_volumes - rename UBI volumes in the volume table.
  * @ubi: UBI device description object
- * @renam_list: list of &struct ubi_rename_entry objects
+ * @rename_list: list of &struct ubi_rename_entry objects
  *
  * This function re-names multiple volumes specified in @req in the volume
  * table. Returns zero in case of success and a negative error code in case of
@@ -166,8 +166,7 @@ int ubi_vtbl_rename_volumes(struct ubi_device *ubi,
 }
 
 /**
- * vtbl_check - check if volume table is not corrupted and contains sensible
- *              data.
+ * vtbl_check - check if volume table is not corrupted and sensible.
  * @ubi: UBI device description object
  * @vtbl: volume table
  *
@@ -780,8 +779,7 @@ static int check_scanning_info(const struct ubi_device *ubi,
 }
 
 /**
- * ubi_read_volume_table - read volume table.
- * information.
+ * ubi_read_volume_table - read the volume table.
  * @ubi: UBI device description object
  * @si: scanning information
  *
diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c
index 2a5d2a0e14ad..05d70937b543 100644
--- a/drivers/mtd/ubi/wl.c
+++ b/drivers/mtd/ubi/wl.c
@@ -632,8 +632,7 @@ out_free:
 }
 
 /**
- * check_protection_over - check if it is time to stop protecting some
- * physical eraseblocks.
+ * check_protection_over - check if it is time to stop protecting some PEBs.
  * @ubi: UBI device description object
  *
  * This function is called after each erase operation, when the absolute erase
@@ -1601,8 +1600,7 @@ void ubi_wl_close(struct ubi_device *ubi)
 #ifdef CONFIG_MTD_UBI_DEBUG_PARANOID
 
 /**
- * paranoid_check_ec - make sure that the erase counter of a physical eraseblock
- * is correct.
+ * paranoid_check_ec - make sure that the erase counter of a PEB is correct.
  * @ubi: UBI device description object
  * @pnum: the physical eraseblock number to check
  * @ec: the erase counter to check
@@ -1643,13 +1641,12 @@ out_free:
 }
 
 /**
- * paranoid_check_in_wl_tree - make sure that a wear-leveling entry is present
- * in a WL RB-tree.
+ * paranoid_check_in_wl_tree - check that wear-leveling entry is in WL RB-tree.
  * @e: the wear-leveling entry to check
  * @root: the root of the tree
  *
- * This function returns zero if @e is in the @root RB-tree and %1 if it
- * is not.
+ * This function returns zero if @e is in the @root RB-tree and %1 if it is
+ * not.
  */
 static int paranoid_check_in_wl_tree(struct ubi_wl_entry *e,
 				     struct rb_root *root)

From 9869cd801c107bbae91663c3f4edbb6b5715919f Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Fri, 18 Jul 2008 13:53:39 +0300
Subject: [PATCH 0189/1171] UBI: remove pre-sqnum images support

Before UBI got into mainline, there was a slight flash format
change - we did not have sequence number support, then added it.

We have carried full support of those ancient images till this
moment. Now the support is removed, well, not fully removed.

Now UBI will support only _clean_ old images, which were cleanly
detached last time (just before kernel upgrade). This is most
likely the case.

But we will not support unclean ancient images. Surprisingly,
this allows us to remove a big chunk of legacy code.

And the same should be true for downgrading: clean images should
downgrade fine, but unclean ones will not.

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/debug.c     |  2 -
 drivers/mtd/ubi/scan.c      | 87 ++++++++++---------------------------
 drivers/mtd/ubi/scan.h      |  2 -
 drivers/mtd/ubi/ubi-media.h | 17 +++-----
 drivers/mtd/ubi/vtbl.c      |  1 -
 5 files changed, 30 insertions(+), 79 deletions(-)

diff --git a/drivers/mtd/ubi/debug.c b/drivers/mtd/ubi/debug.c
index 21e0d7d76a44..c0ed60e8ade9 100644
--- a/drivers/mtd/ubi/debug.c
+++ b/drivers/mtd/ubi/debug.c
@@ -65,7 +65,6 @@ void ubi_dbg_dump_vid_hdr(const struct ubi_vid_hdr *vid_hdr)
 	printk(KERN_DEBUG "\tcompat    %d\n",   (int)vid_hdr->compat);
 	printk(KERN_DEBUG "\tvol_id    %d\n",   be32_to_cpu(vid_hdr->vol_id));
 	printk(KERN_DEBUG "\tlnum      %d\n",   be32_to_cpu(vid_hdr->lnum));
-	printk(KERN_DEBUG "\tleb_ver   %u\n",   be32_to_cpu(vid_hdr->leb_ver));
 	printk(KERN_DEBUG "\tdata_size %d\n",   be32_to_cpu(vid_hdr->data_size));
 	printk(KERN_DEBUG "\tused_ebs  %d\n",   be32_to_cpu(vid_hdr->used_ebs));
 	printk(KERN_DEBUG "\tdata_pad  %d\n",   be32_to_cpu(vid_hdr->data_pad));
@@ -172,7 +171,6 @@ void ubi_dbg_dump_seb(const struct ubi_scan_leb *seb, int type)
 		printk(KERN_DEBUG "\tlnum     %d\n", seb->lnum);
 		printk(KERN_DEBUG "\tscrub    %d\n", seb->scrub);
 		printk(KERN_DEBUG "\tsqnum    %llu\n", seb->sqnum);
-		printk(KERN_DEBUG "\tleb_ver  %u\n", seb->leb_ver);
 	}
 }
 
diff --git a/drivers/mtd/ubi/scan.c b/drivers/mtd/ubi/scan.c
index 4dfbf27b0656..967bb4406df9 100644
--- a/drivers/mtd/ubi/scan.c
+++ b/drivers/mtd/ubi/scan.c
@@ -246,46 +246,21 @@ static int compare_lebs(struct ubi_device *ubi, const struct ubi_scan_leb *seb,
 	struct ubi_vid_hdr *vh = NULL;
 	unsigned long long sqnum2 = be64_to_cpu(vid_hdr->sqnum);
 
-	if (seb->sqnum == 0 && sqnum2 == 0) {
-		long long abs;
-		long long v1 = seb->leb_ver, v2 = be32_to_cpu(vid_hdr->leb_ver);
-
+	if (sqnum2 == seb->sqnum) {
 		/*
-		 * UBI constantly increases the logical eraseblock version
-		 * number and it can overflow. Thus, we have to bear in mind
-		 * that versions that are close to %0xFFFFFFFF are less then
-		 * versions that are close to %0.
-		 *
-		 * The UBI WL sub-system guarantees that the number of pending
-		 * tasks is not greater then %0x7FFFFFFF. So, if the difference
-		 * between any two versions is greater or equivalent to
-		 * %0x7FFFFFFF, there was an overflow and the logical
-		 * eraseblock with lower version is actually newer then the one
-		 * with higher version.
-		 *
-		 * FIXME: but this is anyway obsolete and will be removed at
-		 * some point.
+		 * This must be a really ancient UBI image which has been
+		 * created before sequence numbers support has been added. At
+		 * that times we used 32-bit LEB versions stored in logical
+		 * eraseblocks. That was before UBI got into mainline. We do not
+		 * support these images anymore. Well, those images will work
+		 * still work, but only if no unclean reboots happened.
 		 */
-		dbg_bld("using old crappy leb_ver stuff");
+		ubi_err("unsupported on-flash UBI format\n");
+		return -EINVAL;
+	}
 
-		if (v1 == v2) {
-			ubi_err("PEB %d and PEB %d have the same version %lld",
-				seb->pnum, pnum, v1);
-			return -EINVAL;
-		}
-
-		abs = v1 - v2;
-		if (abs < 0)
-			abs = -abs;
-
-		if (abs < 0x7FFFFFFF)
-			/* Non-overflow situation */
-			second_is_newer = (v2 > v1);
-		else
-			second_is_newer = (v2 < v1);
-	} else
-		/* Obviously the LEB with lower sequence counter is older */
-		second_is_newer = sqnum2 > seb->sqnum;
+	/* Obviously the LEB with lower sequence counter is older */
+	second_is_newer = !!(sqnum2 > seb->sqnum);
 
 	/*
 	 * Now we know which copy is newer. If the copy flag of the PEB with
@@ -293,7 +268,7 @@ static int compare_lebs(struct ubi_device *ubi, const struct ubi_scan_leb *seb,
 	 * check data CRC. For the second PEB we already have the VID header,
 	 * for the first one - we'll need to re-read it from flash.
 	 *
-	 * FIXME: this may be optimized so that we wouldn't read twice.
+	 * Note: this may be optimized so that we wouldn't read twice.
 	 */
 
 	if (second_is_newer) {
@@ -399,7 +374,6 @@ int ubi_scan_add_used(struct ubi_device *ubi, struct ubi_scan_info *si,
 		      int bitflips)
 {
 	int err, vol_id, lnum;
-	uint32_t leb_ver;
 	unsigned long long sqnum;
 	struct ubi_scan_volume *sv;
 	struct ubi_scan_leb *seb;
@@ -408,10 +382,9 @@ int ubi_scan_add_used(struct ubi_device *ubi, struct ubi_scan_info *si,
 	vol_id = be32_to_cpu(vid_hdr->vol_id);
 	lnum = be32_to_cpu(vid_hdr->lnum);
 	sqnum = be64_to_cpu(vid_hdr->sqnum);
-	leb_ver = be32_to_cpu(vid_hdr->leb_ver);
 
-	dbg_bld("PEB %d, LEB %d:%d, EC %d, sqnum %llu, ver %u, bitflips %d",
-		pnum, vol_id, lnum, ec, sqnum, leb_ver, bitflips);
+	dbg_bld("PEB %d, LEB %d:%d, EC %d, sqnum %llu, bitflips %d",
+		pnum, vol_id, lnum, ec, sqnum, bitflips);
 
 	sv = add_volume(si, vol_id, pnum, vid_hdr);
 	if (IS_ERR(sv) < 0)
@@ -444,25 +417,20 @@ int ubi_scan_add_used(struct ubi_device *ubi, struct ubi_scan_info *si,
 		 */
 
 		dbg_bld("this LEB already exists: PEB %d, sqnum %llu, "
-			"LEB ver %u, EC %d", seb->pnum, seb->sqnum,
-			seb->leb_ver, seb->ec);
-
-		/*
-		 * Make sure that the logical eraseblocks have different
-		 * versions. Otherwise the image is bad.
-		 */
-		if (seb->leb_ver == leb_ver && leb_ver != 0) {
-			ubi_err("two LEBs with same version %u", leb_ver);
-			ubi_dbg_dump_seb(seb, 0);
-			ubi_dbg_dump_vid_hdr(vid_hdr);
-			return -EINVAL;
-		}
+			"EC %d", seb->pnum, seb->sqnum, seb->ec);
 
 		/*
 		 * Make sure that the logical eraseblocks have different
 		 * sequence numbers. Otherwise the image is bad.
 		 *
-		 * FIXME: remove 'sqnum != 0' check when leb_ver is removed.
+		 * However, if the sequence number is zero, we assume it must
+		 * be an ancient UBI image from the era when UBI did not have
+		 * sequence numbers. We still can attach these images, unless
+		 * there is a need to distinguish between old and new
+		 * eraseblocks, in which case we'll refuse the image in
+		 * 'compare_lebs()'. In other words, we attach old clean
+		 * images, but refuse attaching old images with duplicated
+		 * logical eraseblocks because there was an unclean reboot.
 		 */
 		if (seb->sqnum == sqnum && sqnum != 0) {
 			ubi_err("two LEBs with same sequence number %llu",
@@ -502,7 +470,6 @@ int ubi_scan_add_used(struct ubi_device *ubi, struct ubi_scan_info *si,
 			seb->pnum = pnum;
 			seb->scrub = ((cmp_res & 2) || bitflips);
 			seb->sqnum = sqnum;
-			seb->leb_ver = leb_ver;
 
 			if (sv->highest_lnum == lnum)
 				sv->last_data_size =
@@ -539,7 +506,6 @@ int ubi_scan_add_used(struct ubi_device *ubi, struct ubi_scan_info *si,
 	seb->lnum = lnum;
 	seb->sqnum = sqnum;
 	seb->scrub = bitflips;
-	seb->leb_ver = leb_ver;
 
 	if (sv->highest_lnum <= lnum) {
 		sv->highest_lnum = lnum;
@@ -1263,11 +1229,6 @@ static int paranoid_check_si(struct ubi_device *ubi, struct ubi_scan_info *si)
 				ubi_err("bad data_pad %d", sv->data_pad);
 				goto bad_vid_hdr;
 			}
-
-			if (seb->leb_ver != be32_to_cpu(vidh->leb_ver)) {
-				ubi_err("bad leb_ver %u", seb->leb_ver);
-				goto bad_vid_hdr;
-			}
 		}
 
 		if (!last_seb)
diff --git a/drivers/mtd/ubi/scan.h b/drivers/mtd/ubi/scan.h
index 4e2e3cc0becd..61df208e2f20 100644
--- a/drivers/mtd/ubi/scan.h
+++ b/drivers/mtd/ubi/scan.h
@@ -34,7 +34,6 @@
  * @u: unions RB-tree or @list links
  * @u.rb: link in the per-volume RB-tree of &struct ubi_scan_leb objects
  * @u.list: link in one of the eraseblock lists
- * @leb_ver: logical eraseblock version (obsolete)
  *
  * One object of this type is allocated for each physical eraseblock during
  * scanning.
@@ -49,7 +48,6 @@ struct ubi_scan_leb {
 		struct rb_node rb;
 		struct list_head list;
 	} u;
-	uint32_t leb_ver;
 };
 
 /**
diff --git a/drivers/mtd/ubi/ubi-media.h b/drivers/mtd/ubi/ubi-media.h
index 26bb7af9787a..2ad940409053 100644
--- a/drivers/mtd/ubi/ubi-media.h
+++ b/drivers/mtd/ubi/ubi-media.h
@@ -168,16 +168,15 @@ struct ubi_ec_hdr {
  *          %UBI_COMPAT_IGNORE, %UBI_COMPAT_PRESERVE, or %UBI_COMPAT_REJECT)
  * @vol_id: ID of this volume
  * @lnum: logical eraseblock number
- * @leb_ver: version of this logical eraseblock (IMPORTANT: obsolete, to be
- *           removed, kept only for not breaking older UBI users)
+ * @padding1: reserved for future, zeroes
  * @data_size: how many bytes of data this logical eraseblock contains
  * @used_ebs: total number of used logical eraseblocks in this volume
  * @data_pad: how many bytes at the end of this physical eraseblock are not
  *            used
  * @data_crc: CRC checksum of the data stored in this logical eraseblock
- * @padding1: reserved for future, zeroes
- * @sqnum: sequence number
  * @padding2: reserved for future, zeroes
+ * @sqnum: sequence number
+ * @padding3: reserved for future, zeroes
  * @hdr_crc: volume identifier header CRC checksum
  *
  * The @sqnum is the value of the global sequence counter at the time when this
@@ -225,10 +224,6 @@ struct ubi_ec_hdr {
  * checksum is correct, this physical eraseblock is selected (P1). Otherwise
  * the older one (P) is selected.
  *
- * Note, there is an obsolete @leb_ver field which was used instead of @sqnum
- * in the past. But it is not used anymore and we keep it in order to be able
- * to deal with old UBI images. It will be removed at some point.
- *
  * There are 2 sorts of volumes in UBI: user volumes and internal volumes.
  * Internal volumes are not seen from outside and are used for various internal
  * UBI purposes. In this implementation there is only one internal volume - the
@@ -278,14 +273,14 @@ struct ubi_vid_hdr {
 	__u8    compat;
 	__be32  vol_id;
 	__be32  lnum;
-	__be32  leb_ver; /* obsolete, to be removed, don't use */
+	__u8    padding1[4];
 	__be32  data_size;
 	__be32  used_ebs;
 	__be32  data_pad;
 	__be32  data_crc;
-	__u8    padding1[4];
+	__u8    padding2[4];
 	__be64  sqnum;
-	__u8    padding2[12];
+	__u8    padding3[12];
 	__be32  hdr_crc;
 } __attribute__ ((packed));
 
diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c
index 4e1c489a3bae..217d0e111b2a 100644
--- a/drivers/mtd/ubi/vtbl.c
+++ b/drivers/mtd/ubi/vtbl.c
@@ -338,7 +338,6 @@ retry:
 			     vid_hdr->data_pad = cpu_to_be32(0);
 	vid_hdr->lnum = cpu_to_be32(copy);
 	vid_hdr->sqnum = cpu_to_be64(++si->max_sqnum);
-	vid_hdr->leb_ver = cpu_to_be32(old_seb ? old_seb->leb_ver + 1: 0);
 
 	/* The EC header is already there, write the VID header */
 	err = ubi_io_write_vid_hdr(ubi, new_seb->pnum, vid_hdr);

From eeb16e87b6747c9a4f5769f33467c9d173e9f5ee Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Wed, 23 Jul 2008 15:51:46 +0300
Subject: [PATCH 0190/1171] UBI: fix gcc warning

Fix the following warning:

drivers/mtd/ubi/vmt.c: In function 'ubi_rename_volumes':
drivers/mtd/ubi/vmt.c:642: warning: statement with no effect

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/vmt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mtd/ubi/vmt.c b/drivers/mtd/ubi/vmt.c
index d40066833abb..3531ca9a1e24 100644
--- a/drivers/mtd/ubi/vmt.c
+++ b/drivers/mtd/ubi/vmt.c
@@ -639,7 +639,7 @@ int ubi_rename_volumes(struct ubi_device *ubi, struct list_head *rename_list)
 	}
 
 	if (!err)
-		paranoid_check_volumes(ubi);
+		err = paranoid_check_volumes(ubi);
 	return err;
 }
 

From 9d25d4db81833029d30b7b03cc1000cbbe09e192 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Mon, 21 Jul 2008 18:41:26 +0100
Subject: [PATCH 0191/1171] x86: BUILD_IRQ say .text to avoid .data.percpu

When I edit the x86_64 Makefile to -fno-unit-at-a-time, bootup panics
on 0xCCs in IRQ0x3e_interrupt(): IRQ0x20_interrupt etc. have got linked
into .data.percpu.  Perhaps there are other ways of triggering that:
specify ".text" in the BUILD_IRQ() macro for safety.

I've been using -fno-unit-at-a-time (to lessen inlining, for easier
debugging) for a long time.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: Mike Travis <travis@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/irqinit_64.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/kernel/irqinit_64.c b/arch/x86/kernel/irqinit_64.c
index 0373e88de95a..9414125f19ce 100644
--- a/arch/x86/kernel/irqinit_64.c
+++ b/arch/x86/kernel/irqinit_64.c
@@ -43,7 +43,7 @@
 
 #define BUILD_IRQ(nr)				\
 	asmlinkage void IRQ_NAME(nr);		\
-	asm("\n.p2align\n"			\
+	asm("\n.text\n.p2align\n"		\
 	    "IRQ" #nr "_interrupt:\n\t"		\
 	    "push $~(" #nr ") ; "		\
 	    "jmp common_interrupt");

From 04bbe430f73c6c31bbd067349c029e907e153a8d Mon Sep 17 00:00:00 2001
From: Vegard Nossum <vegard.nossum@gmail.com>
Date: Mon, 21 Jul 2008 15:06:35 +0200
Subject: [PATCH 0192/1171] x86: fix header export, asm-x86/processor-flags.h,
 CONFIG_* leaks

Apparently,

commit 6330a30a76c1e62d4b4ec238368957f8febf9113
Author: Vegard Nossum <vegard.nossum@gmail.com>
Date:   Wed May 28 09:46:19 2008 +0200

    x86: break mutual header inclusion

introduced some CONFIG names to processor-flags.h, which was exported in

commit 6093015db2bd9e70cf20cdd23be1a50733baafdd
Author: Ingo Molnar <mingo@elte.hu>
Date:   Sun Mar 30 11:45:23 2008 +0200

    x86: cleanup replace most vm86 flags with flags from processor-flags.h, fix

Fix it by wrapping the CONFIG parts in __KERNEL__.

Reported-by: Olaf Hering <olaf@aepfle.de>
Signed-off-by: Vegard Nossum <vegard.nossum@gmail.com>
Cc: Adrian Bunk <bunk@kernel.org>
Cc: Olaf Hering <olaf@aepfle.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/processor-flags.h | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/include/asm-x86/processor-flags.h b/include/asm-x86/processor-flags.h
index 092b39b3a7e6..eff2ecd7fff0 100644
--- a/include/asm-x86/processor-flags.h
+++ b/include/asm-x86/processor-flags.h
@@ -88,10 +88,12 @@
 #define CX86_ARR_BASE	0xc4
 #define CX86_RCR_BASE	0xdc
 
+#ifdef __KERNEL__
 #ifdef CONFIG_VM86
 #define X86_VM_MASK	X86_EFLAGS_VM
 #else
 #define X86_VM_MASK	0 /* No VM86 support */
 #endif
+#endif
 
 #endif	/* __ASM_I386_PROCESSOR_FLAGS_H */

From 58838cf3ca3337d76141c33d6c68376490263468 Mon Sep 17 00:00:00 2001
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Thu, 24 Jul 2008 12:43:13 +0200
Subject: [PATCH 0193/1171] sched: clean up compiler warning

Reported-by: Daniel Walker <dwalker@mvista.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sched_rt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c
index 147004c651c0..93ac8ee08271 100644
--- a/kernel/sched_rt.c
+++ b/kernel/sched_rt.c
@@ -253,7 +253,7 @@ static int do_balance_runtime(struct rt_rq *rt_rq)
 
 		diff = iter->rt_runtime - iter->rt_time;
 		if (diff > 0) {
-			do_div(diff, weight);
+			diff = div_u64((u64)diff, weight);
 			if (rt_rq->rt_runtime + diff > rt_period)
 				diff = rt_period - rt_rq->rt_runtime;
 			iter->rt_runtime -= diff;

From 78305de2f99e9f43ab860dd95bb430b20e26c695 Mon Sep 17 00:00:00 2001
From: Matthew Wilcox <matthew@wil.cx>
Date: Wed, 23 Apr 2008 07:20:41 -0400
Subject: [PATCH 0194/1171] Remove mention of semaphores from kernel-locking

Since the consensus seems to be to eliminate semaphores where possible,
we shouldn't be educating people about how to use them as locks.  Use
mutexes instead.  Semaphores should be described in a separate document
if we end up keeping them.

Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
---
 Documentation/DocBook/kernel-locking.tmpl | 57 ++++++++++-------------
 1 file changed, 24 insertions(+), 33 deletions(-)

diff --git a/Documentation/DocBook/kernel-locking.tmpl b/Documentation/DocBook/kernel-locking.tmpl
index 2510763295d0..084f6ad7b7a0 100644
--- a/Documentation/DocBook/kernel-locking.tmpl
+++ b/Documentation/DocBook/kernel-locking.tmpl
@@ -219,10 +219,10 @@
    </para>
 
    <sect1 id="lock-intro">
-   <title>Three Main Types of Kernel Locks: Spinlocks, Mutexes and Semaphores</title>
+   <title>Two Main Types of Kernel Locks: Spinlocks and Mutexes</title>
 
    <para>
-     There are three main types of kernel locks.  The fundamental type
+     There are two main types of kernel locks.  The fundamental type
      is the spinlock 
      (<filename class="headerfile">include/asm/spinlock.h</filename>),
      which is a very simple single-holder lock: if you can't get the 
@@ -239,14 +239,6 @@
      can't sleep (see <xref linkend="sleeping-things"/>), and so have to
      use a spinlock instead.
    </para>
-   <para>
-     The third type is a semaphore
-     (<filename class="headerfile">include/linux/semaphore.h</filename>): it
-     can have more than one holder at any time (the number decided at
-     initialization time), although it is most commonly used as a
-     single-holder lock (a mutex).  If you can't get a semaphore, your
-     task will be suspended and later on woken up - just like for mutexes.
-   </para>
    <para>
      Neither type of lock is recursive: see
      <xref linkend="deadlock"/>.
@@ -278,7 +270,7 @@
     </para>
 
     <para>
-      Semaphores still exist, because they are required for
+      Mutexes still exist, because they are required for
       synchronization between <firstterm linkend="gloss-usercontext">user 
       contexts</firstterm>, as we will see below.
     </para>
@@ -289,18 +281,17 @@
 
      <para>
        If you have a data structure which is only ever accessed from
-       user context, then you can use a simple semaphore
-       (<filename>linux/linux/semaphore.h</filename>) to protect it.  This
-       is the most trivial case: you initialize the semaphore to the number 
-       of resources available (usually 1), and call
-       <function>down_interruptible()</function> to grab the semaphore, and 
-       <function>up()</function> to release it.  There is also a 
-       <function>down()</function>, which should be avoided, because it 
+       user context, then you can use a simple mutex
+       (<filename>include/linux/mutex.h</filename>) to protect it.  This
+       is the most trivial case: you initialize the mutex.  Then you can
+       call <function>mutex_lock_interruptible()</function> to grab the mutex,
+       and <function>mutex_unlock()</function> to release it.  There is also a 
+       <function>mutex_lock()</function>, which should be avoided, because it 
        will not return if a signal is received.
      </para>
 
      <para>
-       Example: <filename>linux/net/core/netfilter.c</filename> allows 
+       Example: <filename>net/netfilter/nf_sockopt.c</filename> allows 
        registration of new <function>setsockopt()</function> and 
        <function>getsockopt()</function> calls, with
        <function>nf_register_sockopt()</function>.  Registration and 
@@ -515,7 +506,7 @@
       <listitem>
 	<para>
           If you are in a process context (any syscall) and want to
-	lock other process out, use a semaphore.  You can take a semaphore
+	lock other process out, use a mutex.  You can take a mutex
 	and sleep (<function>copy_from_user*(</function> or
 	<function>kmalloc(x,GFP_KERNEL)</function>).
       </para>
@@ -662,7 +653,7 @@
 <entry>SLBH</entry>
 <entry>SLBH</entry>
 <entry>SLBH</entry>
-<entry>DI</entry>
+<entry>MLI</entry>
 <entry>None</entry>
 </row>
 
@@ -692,8 +683,8 @@
 <entry>spin_lock_bh</entry>
 </row>
 <row>
-<entry>DI</entry>
-<entry>down_interruptible</entry>
+<entry>MLI</entry>
+<entry>mutex_lock_interruptible</entry>
 </row>
 
 </tbody>
@@ -1310,7 +1301,7 @@ as Alan Cox says, <quote>Lock data, not code</quote>.
     <para>
       There is a coding bug where a piece of code tries to grab a
       spinlock twice: it will spin forever, waiting for the lock to
-      be released (spinlocks, rwlocks and semaphores are not
+      be released (spinlocks, rwlocks and mutexes are not
       recursive in Linux).  This is trivial to diagnose: not a
       stay-up-five-nights-talk-to-fluffy-code-bunnies kind of
       problem.
@@ -1335,7 +1326,7 @@ as Alan Cox says, <quote>Lock data, not code</quote>.
 
     <para>
       This complete lockup is easy to diagnose: on SMP boxes the
-      watchdog timer or compiling with <symbol>DEBUG_SPINLOCKS</symbol> set
+      watchdog timer or compiling with <symbol>DEBUG_SPINLOCK</symbol> set
       (<filename>include/linux/spinlock.h</filename>) will show this up 
       immediately when it happens.
     </para>
@@ -1558,7 +1549,7 @@ the amount of locking which needs to be done.
    <title>Read/Write Lock Variants</title>
 
    <para>
-      Both spinlocks and semaphores have read/write variants:
+      Both spinlocks and mutexes have read/write variants:
       <type>rwlock_t</type> and <structname>struct rw_semaphore</structname>.
       These divide users into two classes: the readers and the writers.  If
       you are only reading the data, you can get a read lock, but to write to
@@ -1681,7 +1672,7 @@ the amount of locking which needs to be done.
  #include &lt;linux/slab.h&gt;
  #include &lt;linux/string.h&gt;
 +#include &lt;linux/rcupdate.h&gt;
- #include &lt;linux/semaphore.h&gt;
+ #include &lt;linux/mutex.h&gt;
  #include &lt;asm/errno.h&gt;
 
  struct object
@@ -1913,7 +1904,7 @@ machines due to caching.
        </listitem>
        <listitem>
         <para>
-          <function> put_user()</function>
+          <function>put_user()</function>
         </para>
        </listitem>
       </itemizedlist>
@@ -1927,13 +1918,13 @@ machines due to caching.
 
      <listitem>
       <para>
-      <function>down_interruptible()</function> and
-      <function>down()</function>
+      <function>mutex_lock_interruptible()</function> and
+      <function>mutex_lock()</function>
       </para>
       <para>
-       There is a <function>down_trylock()</function> which can be
+       There is a <function>mutex_trylock()</function> which can be
        used inside interrupt context, as it will not sleep.
-       <function>up()</function> will also never sleep.
+       <function>mutex_unlock()</function> will also never sleep.
       </para>
      </listitem>
     </itemizedlist>
@@ -2023,7 +2014,7 @@ machines due to caching.
       <para>
         Prior to 2.5, or when <symbol>CONFIG_PREEMPT</symbol> is
         unset, processes in user context inside the kernel would not
-        preempt each other (ie. you had that CPU until you have it up,
+        preempt each other (ie. you had that CPU until you gave it up,
         except for interrupts).  With the addition of
         <symbol>CONFIG_PREEMPT</symbol> in 2.5.4, this changed: when
         in user context, higher priority tasks can "cut in": spinlocks

From 0f17e4c796e89d1f69f13b653aba60e6ccfb8ae0 Mon Sep 17 00:00:00 2001
From: Matthew Wilcox <matthew@wil.cx>
Date: Thu, 24 Jul 2008 08:30:48 -0400
Subject: [PATCH 0195/1171] Add missing semaphore.h includes

These files use semaphores but don't include semaphore.h

Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
---
 drivers/input/keyboard/hil_kbd.c | 1 +
 drivers/input/misc/hp_sdc_rtc.c  | 1 +
 drivers/input/serio/hp_sdc.c     | 1 +
 3 files changed, 3 insertions(+)

diff --git a/drivers/input/keyboard/hil_kbd.c b/drivers/input/keyboard/hil_kbd.c
index adbf29f0169d..71c1971abf80 100644
--- a/drivers/input/keyboard/hil_kbd.c
+++ b/drivers/input/keyboard/hil_kbd.c
@@ -37,6 +37,7 @@
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/init.h>
+#include <linux/semaphore.h>
 #include <linux/slab.h>
 #include <linux/pci_ids.h>
 
diff --git a/drivers/input/misc/hp_sdc_rtc.c b/drivers/input/misc/hp_sdc_rtc.c
index 49d8abfe38fe..daa9d4220331 100644
--- a/drivers/input/misc/hp_sdc_rtc.c
+++ b/drivers/input/misc/hp_sdc_rtc.c
@@ -44,6 +44,7 @@
 #include <linux/proc_fs.h>
 #include <linux/poll.h>
 #include <linux/rtc.h>
+#include <linux/semaphore.h>
 
 MODULE_AUTHOR("Brian S. Julin <bri@calyx.com>");
 MODULE_DESCRIPTION("HP i8042 SDC + MSM-58321 RTC Driver");
diff --git a/drivers/input/serio/hp_sdc.c b/drivers/input/serio/hp_sdc.c
index 7b233a492ad5..aad664d5259f 100644
--- a/drivers/input/serio/hp_sdc.c
+++ b/drivers/input/serio/hp_sdc.c
@@ -67,6 +67,7 @@
 #include <linux/module.h>
 #include <linux/ioport.h>
 #include <linux/time.h>
+#include <linux/semaphore.h>
 #include <linux/slab.h>
 #include <linux/hil.h>
 #include <linux/semaphore.h>

From 6310e472717ed736c9bff9840febb71f7bb400ed Mon Sep 17 00:00:00 2001
From: Matthew Wilcox <matthew@wil.cx>
Date: Thu, 24 Jul 2008 08:08:09 -0400
Subject: [PATCH 0196/1171] Remove use of asm/semaphore.h

Change to use linux/semaphore.h

Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
---
 arch/arm/mach-ns9xxx/clock.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm/mach-ns9xxx/clock.c b/arch/arm/mach-ns9xxx/clock.c
index f8639161068f..44ed20d4a388 100644
--- a/arch/arm/mach-ns9xxx/clock.c
+++ b/arch/arm/mach-ns9xxx/clock.c
@@ -14,8 +14,8 @@
 #include <linux/clk.h>
 #include <linux/string.h>
 #include <linux/platform_device.h>
+#include <linux/semaphore.h>
 
-#include <asm/semaphore.h>
 #include "clock.h"
 
 static LIST_HEAD(clocks);

From 2351ec533ed0dd56052ab96988d2161d5ecc8ed9 Mon Sep 17 00:00:00 2001
From: Matthew Wilcox <matthew@wil.cx>
Date: Thu, 24 Jul 2008 08:09:32 -0400
Subject: [PATCH 0197/1171] Remove asm/semaphore.h

All users have now been converted to linux/semaphore.h and we don't need
to keep these files around any longer.

Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
---
 Documentation/feature-removal-schedule.txt | 8 --------
 include/asm-alpha/semaphore.h              | 1 -
 include/asm-arm/semaphore.h                | 1 -
 include/asm-avr32/semaphore.h              | 1 -
 include/asm-blackfin/semaphore.h           | 1 -
 include/asm-cris/semaphore.h               | 1 -
 include/asm-frv/semaphore.h                | 1 -
 include/asm-h8300/semaphore.h              | 1 -
 include/asm-ia64/semaphore.h               | 1 -
 include/asm-m32r/semaphore.h               | 1 -
 include/asm-m68k/semaphore.h               | 1 -
 include/asm-m68knommu/semaphore.h          | 1 -
 include/asm-mips/semaphore.h               | 1 -
 include/asm-mn10300/semaphore.h            | 1 -
 include/asm-parisc/semaphore.h             | 1 -
 include/asm-powerpc/semaphore.h            | 1 -
 include/asm-s390/semaphore.h               | 1 -
 include/asm-sh/semaphore.h                 | 1 -
 include/asm-sparc/semaphore.h              | 1 -
 include/asm-sparc64/semaphore.h            | 1 -
 include/asm-um/semaphore.h                 | 1 -
 include/asm-v850/semaphore.h               | 1 -
 include/asm-x86/semaphore.h                | 1 -
 include/asm-xtensa/semaphore.h             | 1 -
 24 files changed, 31 deletions(-)
 delete mode 100644 include/asm-alpha/semaphore.h
 delete mode 100644 include/asm-arm/semaphore.h
 delete mode 100644 include/asm-avr32/semaphore.h
 delete mode 100644 include/asm-blackfin/semaphore.h
 delete mode 100644 include/asm-cris/semaphore.h
 delete mode 100644 include/asm-frv/semaphore.h
 delete mode 100644 include/asm-h8300/semaphore.h
 delete mode 100644 include/asm-ia64/semaphore.h
 delete mode 100644 include/asm-m32r/semaphore.h
 delete mode 100644 include/asm-m68k/semaphore.h
 delete mode 100644 include/asm-m68knommu/semaphore.h
 delete mode 100644 include/asm-mips/semaphore.h
 delete mode 100644 include/asm-mn10300/semaphore.h
 delete mode 100644 include/asm-parisc/semaphore.h
 delete mode 100644 include/asm-powerpc/semaphore.h
 delete mode 100644 include/asm-s390/semaphore.h
 delete mode 100644 include/asm-sh/semaphore.h
 delete mode 100644 include/asm-sparc/semaphore.h
 delete mode 100644 include/asm-sparc64/semaphore.h
 delete mode 100644 include/asm-um/semaphore.h
 delete mode 100644 include/asm-v850/semaphore.h
 delete mode 100644 include/asm-x86/semaphore.h
 delete mode 100644 include/asm-xtensa/semaphore.h

diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
index 9f73587219e8..09c4a1efb8e3 100644
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -300,14 +300,6 @@ Who:	ocfs2-devel@oss.oracle.com
 
 ---------------------------
 
-What:	asm/semaphore.h
-When:	2.6.26
-Why:	Implementation became generic; users should now include
-	linux/semaphore.h instead.
-Who:	Matthew Wilcox <willy@linux.intel.com>
-
----------------------------
-
 What:	SCTP_GET_PEER_ADDRS_NUM_OLD, SCTP_GET_PEER_ADDRS_OLD,
 	SCTP_GET_LOCAL_ADDRS_NUM_OLD, SCTP_GET_LOCAL_ADDRS_OLD
 When: 	June 2009
diff --git a/include/asm-alpha/semaphore.h b/include/asm-alpha/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-alpha/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>
diff --git a/include/asm-arm/semaphore.h b/include/asm-arm/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-arm/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>
diff --git a/include/asm-avr32/semaphore.h b/include/asm-avr32/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-avr32/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>
diff --git a/include/asm-blackfin/semaphore.h b/include/asm-blackfin/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-blackfin/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>
diff --git a/include/asm-cris/semaphore.h b/include/asm-cris/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-cris/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>
diff --git a/include/asm-frv/semaphore.h b/include/asm-frv/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-frv/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>
diff --git a/include/asm-h8300/semaphore.h b/include/asm-h8300/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-h8300/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>
diff --git a/include/asm-ia64/semaphore.h b/include/asm-ia64/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-ia64/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>
diff --git a/include/asm-m32r/semaphore.h b/include/asm-m32r/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-m32r/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>
diff --git a/include/asm-m68k/semaphore.h b/include/asm-m68k/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-m68k/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>
diff --git a/include/asm-m68knommu/semaphore.h b/include/asm-m68knommu/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-m68knommu/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>
diff --git a/include/asm-mips/semaphore.h b/include/asm-mips/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-mips/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>
diff --git a/include/asm-mn10300/semaphore.h b/include/asm-mn10300/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-mn10300/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>
diff --git a/include/asm-parisc/semaphore.h b/include/asm-parisc/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-parisc/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>
diff --git a/include/asm-powerpc/semaphore.h b/include/asm-powerpc/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-powerpc/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>
diff --git a/include/asm-s390/semaphore.h b/include/asm-s390/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-s390/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>
diff --git a/include/asm-sh/semaphore.h b/include/asm-sh/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-sh/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>
diff --git a/include/asm-sparc/semaphore.h b/include/asm-sparc/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-sparc/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>
diff --git a/include/asm-sparc64/semaphore.h b/include/asm-sparc64/semaphore.h
deleted file mode 100644
index 39362afde5fe..000000000000
--- a/include/asm-sparc64/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-sparc/semaphore.h>
diff --git a/include/asm-um/semaphore.h b/include/asm-um/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-um/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>
diff --git a/include/asm-v850/semaphore.h b/include/asm-v850/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-v850/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>
diff --git a/include/asm-x86/semaphore.h b/include/asm-x86/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-x86/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>
diff --git a/include/asm-xtensa/semaphore.h b/include/asm-xtensa/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-xtensa/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>

From b552068999b0b05087c454e525b30b785c79dc9b Mon Sep 17 00:00:00 2001
From: Matthew Wilcox <matthew@wil.cx>
Date: Wed, 23 Apr 2008 10:07:27 -0400
Subject: [PATCH 0198/1171] Remove __DECLARE_SEMAPHORE_GENERIC

There are no users of __DECLARE_SEMAPHORE_GENERIC in the kernel

Signed-off-by: Matthew Wilcox <willy@linux.intel.com>
---
 include/linux/semaphore.h | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/include/linux/semaphore.h b/include/linux/semaphore.h
index 9cae64b00d6b..7415839ac890 100644
--- a/include/linux/semaphore.h
+++ b/include/linux/semaphore.h
@@ -26,10 +26,8 @@ struct semaphore {
 	.wait_list	= LIST_HEAD_INIT((name).wait_list),		\
 }
 
-#define __DECLARE_SEMAPHORE_GENERIC(name, count) \
-	struct semaphore name = __SEMAPHORE_INITIALIZER(name, count)
-
-#define DECLARE_MUTEX(name)	__DECLARE_SEMAPHORE_GENERIC(name, 1)
+#define DECLARE_MUTEX(name)	\
+	struct semaphore name = __SEMAPHORE_INITIALIZER(name, 1)
 
 static inline void sema_init(struct semaphore *sem, int val)
 {

From 76c510ad2e7d56cfe8f2cc7b23783e5c687cf704 Mon Sep 17 00:00:00 2001
From: Shirish Pargaonkar <shirishp@gmail.com>
Date: Thu, 24 Jul 2008 14:48:33 +0000
Subject: [PATCH 0199/1171] [CIFS] Fix possible double free if search
 immediately after search rewind fails

Signed-off-by: Shirish Pargaonkar <shirishp@us.ibm.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/readdir.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c
index 83f306954883..5f40ed3473f5 100644
--- a/fs/cifs/readdir.c
+++ b/fs/cifs/readdir.c
@@ -690,6 +690,7 @@ static int find_cifs_entry(const int xid, struct cifsTconInfo *pTcon,
 			else
 				cifs_buf_release(cifsFile->srch_inf.
 						ntwrk_buf_start);
+			cifsFile->srch_inf.ntwrk_buf_start = NULL;
 		}
 		rc = initiate_cifs_search(xid, file);
 		if (rc) {

From b1910ad6222a705650dc991c003af43b94cdb3e1 Mon Sep 17 00:00:00 2001
From: Shirish Pargaonkar <shirishp@gmail.com>
Date: Thu, 24 Jul 2008 14:53:20 +0000
Subject: [PATCH 0200/1171] [CIFS] Fix improper endian conversion of ACL
 subauth field

In mode_to_acl when converting a Unix mode to a Windows ACL
the subauth fields of the SID in the ACL were translated
incorrectly on bigendian architectures

Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/cifsacl.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/fs/cifs/cifsacl.c b/fs/cifs/cifsacl.c
index 0e9fc2ba90ee..18faf65b9114 100644
--- a/fs/cifs/cifsacl.c
+++ b/fs/cifs/cifsacl.c
@@ -169,8 +169,7 @@ static void copy_sec_desc(const struct cifs_ntsd *pntsd,
 	for (i = 0; i < 6; i++)
 		ngroup_sid_ptr->authority[i] = group_sid_ptr->authority[i];
 	for (i = 0; i < 5; i++)
-		ngroup_sid_ptr->sub_auth[i] =
-				cpu_to_le32(group_sid_ptr->sub_auth[i]);
+		ngroup_sid_ptr->sub_auth[i] = group_sid_ptr->sub_auth[i];
 
 	return;
 }

From ef571cadd516e7ffcdeac6bb8054e5908fcccfcf Mon Sep 17 00:00:00 2001
From: Shirish Pargaonkar <shirishp@gmail.com>
Date: Thu, 24 Jul 2008 15:56:05 +0000
Subject: [PATCH 0201/1171] [CIFS] Fix warnings from checkpatch

Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/cifs_debug.c  |  4 ++--
 fs/cifs/cifsacl.c     | 38 +++++++++++++++++++-------------------
 fs/cifs/cifsencrypt.c |  3 +--
 fs/cifs/cifsglob.h    |  6 +++---
 fs/cifs/cifspdu.h     |  8 ++++----
 fs/cifs/cifssmb.c     |  5 ++---
 6 files changed, 31 insertions(+), 33 deletions(-)

diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c
index c213aa40064e..688a2d42153f 100644
--- a/fs/cifs/cifs_debug.c
+++ b/fs/cifs/cifs_debug.c
@@ -208,9 +208,9 @@ static int cifs_debug_data_proc_show(struct seq_file *m, void *v)
 			seq_puts(m, " type: CDROM ");
 		else
 			seq_printf(m, " type: %d ", dev_type);
-		if (tcon->tidStatus == CifsNeedReconnect) {
+
+		if (tcon->tidStatus == CifsNeedReconnect)
 			seq_puts(m, "\tDISCONNECTED ");
-		}
 	}
 	read_unlock(&GlobalSMBSeslock);
 
diff --git a/fs/cifs/cifsacl.c b/fs/cifs/cifsacl.c
index 18faf65b9114..57ecdc83c26f 100644
--- a/fs/cifs/cifsacl.c
+++ b/fs/cifs/cifsacl.c
@@ -56,7 +56,7 @@ int match_sid(struct cifs_sid *ctsid)
 	struct cifs_sid *cwsid;
 
 	if (!ctsid)
-		return (-1);
+		return -1;
 
 	for (i = 0; i < NUM_WK_SIDS; ++i) {
 		cwsid = &(wksidarr[i].cifssid);
@@ -87,11 +87,11 @@ int match_sid(struct cifs_sid *ctsid)
 		}
 
 		cFYI(1, ("matching sid: %s\n", wksidarr[i].sidname));
-		return (0); /* sids compare/match */
+		return 0; /* sids compare/match */
 	}
 
 	cFYI(1, ("No matching sid"));
-	return (-1);
+	return -1;
 }
 
 /* if the two SIDs (roughly equivalent to a UUID for a user or group) are
@@ -102,16 +102,16 @@ int compare_sids(const struct cifs_sid *ctsid, const struct cifs_sid *cwsid)
 	int num_subauth, num_sat, num_saw;
 
 	if ((!ctsid) || (!cwsid))
-		return (0);
+		return 0;
 
 	/* compare the revision */
 	if (ctsid->revision != cwsid->revision)
-		return (0);
+		return 0;
 
 	/* compare all of the six auth values */
 	for (i = 0; i < 6; ++i) {
 		if (ctsid->authority[i] != cwsid->authority[i])
-			return (0);
+			return 0;
 	}
 
 	/* compare all of the subauth values if any */
@@ -121,11 +121,11 @@ int compare_sids(const struct cifs_sid *ctsid, const struct cifs_sid *cwsid)
 	if (num_subauth) {
 		for (i = 0; i < num_subauth; ++i) {
 			if (ctsid->sub_auth[i] != cwsid->sub_auth[i])
-				return (0);
+				return 0;
 		}
 	}
 
-	return (1); /* sids compare/match */
+	return 1; /* sids compare/match */
 }
 
 
@@ -284,7 +284,7 @@ static __u16 fill_ace_for_sid(struct cifs_ace *pntace,
 	size = 1 + 1 + 2 + 4 + 1 + 1 + 6 + (psid->num_subauth * 4);
 	pntace->size = cpu_to_le16(size);
 
-	return (size);
+	return size;
 }
 
 
@@ -425,7 +425,7 @@ static int set_chmod_dacl(struct cifs_acl *pndacl, struct cifs_sid *pownersid,
 	pndacl->size = cpu_to_le16(size + sizeof(struct cifs_acl));
 	pndacl->num_aces = cpu_to_le32(3);
 
-	return (0);
+	return 0;
 }
 
 
@@ -509,7 +509,7 @@ static int parse_sec_desc(struct cifs_ntsd *pntsd, int acl_len,
 			sizeof(struct cifs_sid)); */
 
 
-	return (0);
+	return 0;
 }
 
 
@@ -526,7 +526,7 @@ static int build_sec_desc(struct cifs_ntsd *pntsd, struct cifs_ntsd *pnntsd,
 	struct cifs_acl *ndacl_ptr = NULL; /* no need for SACL ptr */
 
 	if ((inode == NULL) || (pntsd == NULL) || (pnntsd == NULL))
-		return (-EIO);
+		return -EIO;
 
 	owner_sid_ptr = (struct cifs_sid *)((char *)pntsd +
 				le32_to_cpu(pntsd->osidoffset));
@@ -549,7 +549,7 @@ static int build_sec_desc(struct cifs_ntsd *pntsd, struct cifs_ntsd *pnntsd,
 	/* copy security descriptor control portion and owner and group sid */
 	copy_sec_desc(pntsd, pnntsd, sidsoffset);
 
-	return (rc);
+	return rc;
 }
 
 
@@ -628,11 +628,11 @@ static int set_cifs_acl(struct cifs_ntsd *pnntsd, __u32 acllen,
 	cFYI(DBG2, ("set ACL for %s from mode 0x%x", path, inode->i_mode));
 
 	if (!inode)
-		return (rc);
+		return rc;
 
 	sb = inode->i_sb;
 	if (sb == NULL)
-		return (rc);
+		return rc;
 
 	cifs_sb = CIFS_SB(sb);
 	xid = GetXid();
@@ -651,7 +651,7 @@ static int set_cifs_acl(struct cifs_ntsd *pnntsd, __u32 acllen,
 		if (rc != 0) {
 			cERROR(1, ("Unable to open file to set ACL"));
 			FreeXid(xid);
-			return (rc);
+			return rc;
 		}
 	}
 
@@ -664,7 +664,7 @@ static int set_cifs_acl(struct cifs_ntsd *pnntsd, __u32 acllen,
 
 	FreeXid(xid);
 
-	return (rc);
+	return rc;
 }
 
 /* Translate the CIFS ACL (simlar to NTFS ACL) for a file into mode bits */
@@ -714,7 +714,7 @@ int mode_to_acl(struct inode *inode, const char *path, __u64 nmode)
 		if (!pnntsd) {
 			cERROR(1, ("Unable to allocate security descriptor"));
 			kfree(pntsd);
-			return (-ENOMEM);
+			return -ENOMEM;
 		}
 
 		rc = build_sec_desc(pntsd, pnntsd, inode, nmode);
@@ -731,6 +731,6 @@ int mode_to_acl(struct inode *inode, const char *path, __u64 nmode)
 		kfree(pntsd);
 	}
 
-	return (rc);
+	return rc;
 }
 #endif /* CONFIG_CIFS_EXPERIMENTAL */
diff --git a/fs/cifs/cifsencrypt.c b/fs/cifs/cifsencrypt.c
index 4ff8939c6cc7..83fd40dc1ef0 100644
--- a/fs/cifs/cifsencrypt.c
+++ b/fs/cifs/cifsencrypt.c
@@ -310,9 +310,8 @@ void calc_lanman_hash(struct cifsSesInfo *ses, char *lnm_session_key)
 	utf8 and other multibyte codepages each need their own strupper
 	function since a byte at a time will ont work. */
 
-	for (i = 0; i < CIFS_ENCPWD_SIZE; i++) {
+	for (i = 0; i < CIFS_ENCPWD_SIZE; i++)
 		password_with_pad[i] = toupper(password_with_pad[i]);
-	}
 
 	SMBencrypt(password_with_pad, ses->server->cryptKey, lnm_session_key);
 	/* clear password before we return/free memory */
diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h
index 9cfcf326ead3..7e1cf262effe 100644
--- a/fs/cifs/cifsglob.h
+++ b/fs/cifs/cifsglob.h
@@ -27,7 +27,7 @@
 #define MAX_SES_INFO 2
 #define MAX_TCON_INFO 4
 
-#define MAX_TREE_SIZE 2 + MAX_SERVER_SIZE + 1 + MAX_SHARE_SIZE + 1
+#define MAX_TREE_SIZE (2 + MAX_SERVER_SIZE + 1 + MAX_SHARE_SIZE + 1)
 #define MAX_SERVER_SIZE 15
 #define MAX_SHARE_SIZE  64	/* used to be 20, this should still be enough */
 #define MAX_USERNAME_SIZE 32	/* 32 is to allow for 15 char names + null
@@ -537,8 +537,8 @@ require use of the stronger protocol */
 #endif /* WEAK_PW_HASH */
 #define   CIFSSEC_MUST_SEAL	0x40040 /* not supported yet */
 
-#define   CIFSSEC_DEF  CIFSSEC_MAY_SIGN | CIFSSEC_MAY_NTLM | CIFSSEC_MAY_NTLMV2
-#define   CIFSSEC_MAX  CIFSSEC_MUST_SIGN | CIFSSEC_MUST_NTLMV2
+#define   CIFSSEC_DEF (CIFSSEC_MAY_SIGN | CIFSSEC_MAY_NTLM | CIFSSEC_MAY_NTLMV2)
+#define   CIFSSEC_MAX (CIFSSEC_MUST_SIGN | CIFSSEC_MUST_NTLMV2)
 #define   CIFSSEC_AUTH_MASK (CIFSSEC_MAY_NTLM | CIFSSEC_MAY_NTLMV2 | CIFSSEC_MAY_LANMAN | CIFSSEC_MAY_PLNTXT | CIFSSEC_MAY_KRB5)
 /*
  *****************************************************************
diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h
index 0f327c224da3..409abce12732 100644
--- a/fs/cifs/cifspdu.h
+++ b/fs/cifs/cifspdu.h
@@ -31,7 +31,7 @@
 #else
 #define CIFS_PROT   0
 #endif
-#define POSIX_PROT  CIFS_PROT+1
+#define POSIX_PROT  (CIFS_PROT+1)
 #define BAD_PROT 0xFFFF
 
 /* SMB command codes */
@@ -341,7 +341,7 @@
 #define CREATE_COMPLETE_IF_OPLK 0x00000100	/* should be zero */
 #define CREATE_NO_EA_KNOWLEDGE  0x00000200
 #define CREATE_EIGHT_DOT_THREE  0x00000400	/* doc says this is obsolete
-						 "open for recovery" flag - should
+						 "open for recovery" flag should
 						 be zero in any case */
 #define CREATE_OPEN_FOR_RECOVERY 0x00000400
 #define CREATE_RANDOM_ACCESS	0x00000800
@@ -414,8 +414,8 @@ struct smb_hdr {
 	__u8 WordCount;
 } __attribute__((packed));
 /* given a pointer to an smb_hdr retrieve the value of byte count */
-#define BCC(smb_var) ( *(__u16 *)((char *)smb_var + sizeof(struct smb_hdr) + (2 * smb_var->WordCount)))
-#define BCC_LE(smb_var) ( *(__le16 *)((char *)smb_var + sizeof(struct smb_hdr) + (2 * smb_var->WordCount)))
+#define BCC(smb_var) (*(__u16 *)((char *)smb_var + sizeof(struct smb_hdr) + (2 * smb_var->WordCount)))
+#define BCC_LE(smb_var) (*(__le16 *)((char *)smb_var + sizeof(struct smb_hdr) + (2 * smb_var->WordCount)))
 /* given a pointer to an smb_hdr retrieve the pointer to the byte area */
 #define pByteArea(smb_var) ((unsigned char *)smb_var + sizeof(struct smb_hdr) + (2 * smb_var->WordCount) + 2)
 
diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c
index 7e175e1399d1..c621ffa2ca90 100644
--- a/fs/cifs/cifssmb.c
+++ b/fs/cifs/cifssmb.c
@@ -686,11 +686,10 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses)
 						 SecurityBlob,
 						 count - 16,
 						 &server->secType);
-			if (rc == 1) {
+			if (rc == 1)
 				rc = 0;
-			} else {
+			else
 				rc = -EINVAL;
-			}
 		}
 	} else
 		server->capabilities &= ~CAP_EXTENDED_SECURITY;

From e108526e77aa41c89b3be96f75d97615db2b751c Mon Sep 17 00:00:00 2001
From: Akinobu Mita <akinobu.mita@gmail.com>
Date: Wed, 23 Jul 2008 21:26:44 -0700
Subject: [PATCH 0202/1171] move memory_read_from_buffer() from fs.h to
 string.h

James Bottomley warns that inclusion of linux/fs.h in a low level
driver was always a danger signal.  This patch moves
memory_read_from_buffer() from fs.h to string.h and fixes includes in
existing memory_read_from_buffer() users.

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: James Bottomley <James.Bottomley@hansenpartnership.com>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Zhang Rui <rui.zhang@intel.com>
Cc: Bob Moore <robert.moore@intel.com>
Cc: Thomas Renninger <trenn@suse.de>
Cc: Len Brown <lenb@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/acpi/system.c       | 1 +
 drivers/zorro/zorro-sysfs.c | 1 -
 include/linux/fs.h          | 2 --
 include/linux/string.h      | 3 +++
 4 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/acpi/system.c b/drivers/acpi/system.c
index d8e3f153b295..91dec448b3ed 100644
--- a/drivers/acpi/system.c
+++ b/drivers/acpi/system.c
@@ -26,6 +26,7 @@
 #include <linux/proc_fs.h>
 #include <linux/seq_file.h>
 #include <linux/init.h>
+#include <linux/string.h>
 #include <asm/uaccess.h>
 
 #include <acpi/acpi_drivers.h>
diff --git a/drivers/zorro/zorro-sysfs.c b/drivers/zorro/zorro-sysfs.c
index 3da712cc7708..5290552d2ef7 100644
--- a/drivers/zorro/zorro-sysfs.c
+++ b/drivers/zorro/zorro-sysfs.c
@@ -15,7 +15,6 @@
 #include <linux/zorro.h>
 #include <linux/stat.h>
 #include <linux/string.h>
-#include <linux/fs.h>
 
 #include "zorro.h"
 
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 9c2ac5c0ef5c..ff54ae4933f3 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -2006,8 +2006,6 @@ extern void simple_release_fs(struct vfsmount **mount, int *count);
 
 extern ssize_t simple_read_from_buffer(void __user *to, size_t count,
 			loff_t *ppos, const void *from, size_t available);
-extern ssize_t memory_read_from_buffer(void *to, size_t count, loff_t *ppos,
-			const void *from, size_t available);
 
 #ifdef CONFIG_MIGRATION
 extern int buffer_migrate_page(struct address_space *,
diff --git a/include/linux/string.h b/include/linux/string.h
index efdc44593b52..810d80df0a1d 100644
--- a/include/linux/string.h
+++ b/include/linux/string.h
@@ -111,5 +111,8 @@ extern void argv_free(char **argv);
 
 extern bool sysfs_streq(const char *s1, const char *s2);
 
+extern ssize_t memory_read_from_buffer(void *to, size_t count, loff_t *ppos,
+			const void *from, size_t available);
+
 #endif
 #endif /* _LINUX_STRING_H_ */

From d36e74c4392b5f26a5c4d94d7881a156ddc8e593 Mon Sep 17 00:00:00 2001
From: Clemens Ladisch <clemens@ladisch.de>
Date: Wed, 23 Jul 2008 21:26:46 -0700
Subject: [PATCH 0203/1171] hpet: clarify maintainer entry

The existing HPET maintainer entries are somewhat unclear about which one
applies to what part of the kernel.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 MAINTAINERS | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index 5d8971c76a7f..7ffd78c4e277 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -1984,7 +1984,7 @@ P:	Carlos Corbacho
 M:	carlos@strangeworlds.co.uk
 S:	Odd Fixes
 
-HPET:	High Precision Event Timers driver (hpet.c)
+HPET:	High Precision Event Timers driver (drivers/char/hpet.c)
 P:	Clemens Ladisch
 M:	clemens@ladisch.de
 S:	Maintained

From d7ce20b2024d318b9ba88859226af1441270d99f Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Wed, 23 Jul 2008 21:26:47 -0700
Subject: [PATCH 0204/1171] remove is_tty()

This patch removes the no longer used is_tty().

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Acked-by: Alan Cox <alan@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/tty_io.c | 13 -------------
 1 file changed, 13 deletions(-)

diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index fa48dba5ba5e..6f4d856df987 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -1119,19 +1119,6 @@ int tty_hung_up_p(struct file *filp)
 
 EXPORT_SYMBOL(tty_hung_up_p);
 
-/**
- *	is_tty	-	checker whether file is a TTY
- *	@filp:		file handle that may be a tty
- *
- *	Check if the file handle is a tty handle.
- */
-
-int is_tty(struct file *filp)
-{
-	return filp->f_op->read == tty_read
-		|| filp->f_op->read == hung_up_tty_read;
-}
-
 static void session_clear_tty(struct pid *session)
 {
 	struct task_struct *p;

From 9483a578df27fe7603605d565eefe039c1ba5845 Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Wed, 23 Jul 2008 21:26:48 -0700
Subject: [PATCH 0205/1171] add HAVE_CLK to Kconfig, for driver dependencies

Flag platforms as HAVE_CLK (or not) in Kconfig, based on whether they
support <linux/clk.h> calls, so that otherwise portable drivers which need
those calls can list that dependency.

Something like this is a prerequisite for merging the musb_hdrc driver,
currently used on platforms including Davinci, OMAP2430, OMAP3xx ...  and
the discrete TUSB6010 chip, which doesn't have a natural platform
dependency.  (Used with OMAP 2420 in current Nokia N8x0 tablets.)

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Russell King <rmk@arm.linux.org.uk>
Acked-by: Haavard Skinnemoen <hskinnemoen@atmel.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/Kconfig         |  7 +++++++
 arch/arm/Kconfig     | 13 +++++++++++++
 arch/avr32/Kconfig   |  1 +
 arch/powerpc/Kconfig |  1 +
 arch/sh/Kconfig      |  1 +
 5 files changed, 23 insertions(+)

diff --git a/arch/Kconfig b/arch/Kconfig
index ad89a33d8c6e..4d5ebbc1e72b 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -42,3 +42,10 @@ config HAVE_DMA_ATTRS
 
 config USE_GENERIC_SMP_HELPERS
 	def_bool n
+
+config HAVE_CLK
+	def_bool n
+	help
+	  The <linux/clk.h> calls support software clock gating and
+	  thus are a key power management tool on many systems.
+
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index d048f6887d0b..6fb4f03369f2 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -198,12 +198,14 @@ choice
 config ARCH_AAEC2000
 	bool "Agilent AAEC-2000 based"
 	select ARM_AMBA
+	select HAVE_CLK
 	help
 	  This enables support for systems based on the Agilent AAEC-2000
 
 config ARCH_INTEGRATOR
 	bool "ARM Ltd. Integrator family"
 	select ARM_AMBA
+	select HAVE_CLK
 	select ICST525
 	help
 	  Support for ARM's Integrator platform.
@@ -211,6 +213,7 @@ config ARCH_INTEGRATOR
 config ARCH_REALVIEW
 	bool "ARM Ltd. RealView family"
 	select ARM_AMBA
+	select HAVE_CLK
 	select ICST307
 	select GENERIC_TIME
 	select GENERIC_CLOCKEVENTS
@@ -221,6 +224,7 @@ config ARCH_VERSATILE
 	bool "ARM Ltd. Versatile family"
 	select ARM_AMBA
 	select ARM_VIC
+	select HAVE_CLK
 	select ICST307
 	select GENERIC_TIME
 	select GENERIC_CLOCKEVENTS
@@ -262,6 +266,8 @@ config ARCH_EP93XX
 	select ARM_AMBA
 	select ARM_VIC
 	select GENERIC_GPIO
+	select HAVE_CLK
+	select HAVE_CLK
 	select HAVE_GPIO_LIB
 	help
 	  This enables support for the Cirrus EP93xx series of CPUs.
@@ -381,6 +387,7 @@ config ARCH_NS9XXX
 	select GENERIC_GPIO
 	select GENERIC_TIME
 	select GENERIC_CLOCKEVENTS
+	select HAVE_CLK
 	help
 	  Say Y here if you intend to run this kernel on a NetSilicon NS9xxx
 	  System.
@@ -430,6 +437,7 @@ config ARCH_ORION5X
 
 config ARCH_PNX4008
 	bool "Philips Nexperia PNX4008 Mobile"
+	select HAVE_CLK
 	help
 	  This enables support for Philips PNX4008 mobile platform.
 
@@ -438,6 +446,7 @@ config ARCH_PXA
 	depends on MMU
 	select ARCH_MTD_XIP
 	select GENERIC_GPIO
+	select HAVE_CLK
 	select HAVE_GPIO_LIB
 	select GENERIC_TIME
 	select GENERIC_CLOCKEVENTS
@@ -468,6 +477,7 @@ config ARCH_SA1100
 	select GENERIC_GPIO
 	select GENERIC_TIME
 	select GENERIC_CLOCKEVENTS
+	select HAVE_CLK
 	select TICK_ONESHOT
 	select HAVE_GPIO_LIB
 	help
@@ -476,6 +486,7 @@ config ARCH_SA1100
 config ARCH_S3C2410
 	bool "Samsung S3C2410, S3C2412, S3C2413, S3C2440, S3C2442, S3C2443"
 	select GENERIC_GPIO
+	select HAVE_CLK
 	help
 	  Samsung S3C2410X CPU based systems, such as the Simtec Electronics
 	  BAST (<http://www.simtec.co.uk/products/EB110ITX/>), the IPAQ 1940 or
@@ -503,12 +514,14 @@ config ARCH_DAVINCI
 	select GENERIC_TIME
 	select GENERIC_CLOCKEVENTS
 	select GENERIC_GPIO
+	select HAVE_CLK
 	help
 	  Support for TI's DaVinci platform.
 
 config ARCH_OMAP
 	bool "TI OMAP"
 	select GENERIC_GPIO
+	select HAVE_CLK
 	select HAVE_GPIO_LIB
 	select GENERIC_TIME
 	select GENERIC_CLOCKEVENTS
diff --git a/arch/avr32/Kconfig b/arch/avr32/Kconfig
index 45d63c986015..df4adefedb42 100644
--- a/arch/avr32/Kconfig
+++ b/arch/avr32/Kconfig
@@ -10,6 +10,7 @@ config AVR32
 	# With EMBEDDED=n, we get lots of stuff automatically selected
 	# that we usually don't need on AVR32.
 	select EMBEDDED
+	select HAVE_CLK
 	select HAVE_OPROFILE
 	select HAVE_KPROBES
 	help
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 737ebf9d12bb..4d7e2ba10bae 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -843,6 +843,7 @@ source "crypto/Kconfig"
 config PPC_CLOCK
 	bool
 	default n
+	select HAVE_CLK
 
 config PPC_LIB_RHEAP
 	bool
diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig
index 3e7384f4619c..8879938f3356 100644
--- a/arch/sh/Kconfig
+++ b/arch/sh/Kconfig
@@ -8,6 +8,7 @@ mainmenu "Linux/SuperH Kernel Configuration"
 config SUPERH
 	def_bool y
 	select EMBEDDED
+	select HAVE_CLK
 	select HAVE_IDE
 	select HAVE_OPROFILE
 	help

From 6b74ab97bc12ce74acec900f1d89a4aee2e4d70d Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Wed, 23 Jul 2008 21:26:49 -0700
Subject: [PATCH 0206/1171] mm: add a basic debugging framework for memory
 initialisation

Boot initialisation is very complex, with significant numbers of
architecture-specific routines, hooks and code ordering.  While significant
amounts of the initialisation is architecture-independent, it trusts the data
received from the architecture layer.  This is a mistake, and has resulted in
a number of difficult-to-diagnose bugs.

This patchset adds some validation and tracing to memory initialisation.  It
also introduces a few basic defensive measures.  The validation code can be
explicitly disabled for embedded systems.

This patch:

Add additional debugging and verification code for memory initialisation.

Once enabled, the verification checks are always run and when required
additional debugging information may be outputted via a mminit_loglevel=
command-line parameter.

The verification code is placed in a new file mm/mm_init.c.  Ideally other mm
initialisation code will be moved here over time.

Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/kernel-parameters.txt |  8 ++++++++
 lib/Kconfig.debug                   | 12 ++++++++++++
 mm/Makefile                         |  1 +
 mm/internal.h                       | 27 +++++++++++++++++++++++++++
 mm/mm_init.c                        | 18 ++++++++++++++++++
 mm/page_alloc.c                     | 22 +++++++++++++---------
 6 files changed, 79 insertions(+), 9 deletions(-)
 create mode 100644 mm/mm_init.c

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 47e7d8794fc6..5e20ccb5a736 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1225,6 +1225,14 @@ and is between 256 and 4096 characters. It is defined in the file
 
 	mga=		[HW,DRM]
 
+	mminit_loglevel=
+			[KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this
+			parameter allows control of the logging verbosity for
+			the additional memory initialisation checks. A value
+			of 0 disables mminit logging and a level of 4 will
+			log everything. Information is printed at KERN_DEBUG
+			so loglevel=8 may also need to be specified.
+
 	mousedev.tap_time=
 			[MOUSE] Maximum time between finger touching and
 			leaving touchpad surface for touch to be considered
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 882c51048993..e1d4764435ed 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -505,6 +505,18 @@ config DEBUG_WRITECOUNT
 
 	  If unsure, say N.
 
+config DEBUG_MEMORY_INIT
+	bool "Debug memory initialisation" if EMBEDDED
+	default !EMBEDDED
+	help
+	  Enable this for additional checks during memory initialisation.
+	  The sanity checks verify aspects of the VM such as the memory model
+	  and other information provided by the architecture. Verbose
+	  information will be printed at KERN_DEBUG loglevel depending
+	  on the mminit_loglevel= command-line option.
+
+	  If unsure, say Y
+
 config DEBUG_LIST
 	bool "Debug linked list manipulation"
 	depends on DEBUG_KERNEL
diff --git a/mm/Makefile b/mm/Makefile
index 18c143b3c46c..4bbc8f094ff0 100644
--- a/mm/Makefile
+++ b/mm/Makefile
@@ -26,6 +26,7 @@ obj-$(CONFIG_TMPFS_POSIX_ACL) += shmem_acl.o
 obj-$(CONFIG_TINY_SHMEM) += tiny-shmem.o
 obj-$(CONFIG_SLOB) += slob.o
 obj-$(CONFIG_SLAB) += slab.o
+obj-$(CONFIG_DEBUG_MEMORY_INIT) += mm_init.o
 obj-$(CONFIG_SLUB) += slub.o
 obj-$(CONFIG_MEMORY_HOTPLUG) += memory_hotplug.o
 obj-$(CONFIG_FS_XIP) += filemap_xip.o
diff --git a/mm/internal.h b/mm/internal.h
index 0034e947e4bc..a7ee05253294 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -59,4 +59,31 @@ static inline unsigned long page_order(struct page *page)
 #define __paginginit __init
 #endif
 
+/* Memory initialisation debug and verification */
+enum mminit_level {
+	MMINIT_WARNING,
+	MMINIT_VERIFY,
+	MMINIT_TRACE
+};
+
+#ifdef CONFIG_DEBUG_MEMORY_INIT
+
+extern int mminit_loglevel;
+
+#define mminit_dprintk(level, prefix, fmt, arg...) \
+do { \
+	if (level < mminit_loglevel) { \
+		printk(level <= MMINIT_WARNING ? KERN_WARNING : KERN_DEBUG); \
+		printk(KERN_CONT "mminit::" prefix " " fmt, ##arg); \
+	} \
+} while (0)
+
+#else
+
+static inline void mminit_dprintk(enum mminit_level level,
+				const char *prefix, const char *fmt, ...)
+{
+}
+
+#endif /* CONFIG_DEBUG_MEMORY_INIT */
 #endif
diff --git a/mm/mm_init.c b/mm/mm_init.c
new file mode 100644
index 000000000000..c01d8dfec817
--- /dev/null
+++ b/mm/mm_init.c
@@ -0,0 +1,18 @@
+/*
+ * mm_init.c - Memory initialisation verification and debugging
+ *
+ * Copyright 2008 IBM Corporation, 2008
+ * Author Mel Gorman <mel@csn.ul.ie>
+ *
+ */
+#include <linux/kernel.h>
+#include <linux/init.h>
+
+int __meminitdata mminit_loglevel;
+
+static __init int set_mminit_loglevel(char *str)
+{
+	get_option(&str, &mminit_loglevel);
+	return 0;
+}
+early_param("mminit_loglevel", set_mminit_loglevel);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 79ac4afc908c..0908352ba727 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -2975,7 +2975,8 @@ void __init sparse_memory_present_with_active_regions(int nid)
 void __init push_node_boundaries(unsigned int nid,
 		unsigned long start_pfn, unsigned long end_pfn)
 {
-	printk(KERN_DEBUG "Entering push_node_boundaries(%u, %lu, %lu)\n",
+	mminit_dprintk(MMINIT_TRACE, "zoneboundary",
+			"Entering push_node_boundaries(%u, %lu, %lu)\n",
 			nid, start_pfn, end_pfn);
 
 	/* Initialise the boundary for this node if necessary */
@@ -2993,7 +2994,8 @@ void __init push_node_boundaries(unsigned int nid,
 static void __meminit account_node_boundary(unsigned int nid,
 		unsigned long *start_pfn, unsigned long *end_pfn)
 {
-	printk(KERN_DEBUG "Entering account_node_boundary(%u, %lu, %lu)\n",
+	mminit_dprintk(MMINIT_TRACE, "zoneboundary",
+			"Entering account_node_boundary(%u, %lu, %lu)\n",
 			nid, *start_pfn, *end_pfn);
 
 	/* Return if boundary information has not been provided */
@@ -3368,8 +3370,8 @@ static void __paginginit free_area_init_core(struct pglist_data *pgdat,
 			PAGE_ALIGN(size * sizeof(struct page)) >> PAGE_SHIFT;
 		if (realsize >= memmap_pages) {
 			realsize -= memmap_pages;
-			printk(KERN_DEBUG
-				"  %s zone: %lu pages used for memmap\n",
+			mminit_dprintk(MMINIT_TRACE, "memmap_init",
+				"%s zone: %lu pages used for memmap\n",
 				zone_names[j], memmap_pages);
 		} else
 			printk(KERN_WARNING
@@ -3379,7 +3381,8 @@ static void __paginginit free_area_init_core(struct pglist_data *pgdat,
 		/* Account for reserved pages */
 		if (j == 0 && realsize > dma_reserve) {
 			realsize -= dma_reserve;
-			printk(KERN_DEBUG "  %s zone: %lu pages reserved\n",
+			mminit_dprintk(MMINIT_TRACE, "memmap_init",
+					"%s zone: %lu pages reserved\n",
 					zone_names[0], dma_reserve);
 		}
 
@@ -3520,10 +3523,11 @@ void __init add_active_range(unsigned int nid, unsigned long start_pfn,
 {
 	int i;
 
-	printk(KERN_DEBUG "Entering add_active_range(%d, %#lx, %#lx) "
-			  "%d entries of %d used\n",
-			  nid, start_pfn, end_pfn,
-			  nr_nodemap_entries, MAX_ACTIVE_REGIONS);
+	mminit_dprintk(MMINIT_TRACE, "memory_register",
+			"Entering add_active_range(%d, %#lx, %#lx) "
+			"%d entries of %d used\n",
+			nid, start_pfn, end_pfn,
+			nr_nodemap_entries, MAX_ACTIVE_REGIONS);
 
 	/* Merge with existing active regions if possible */
 	for (i = 0; i < nr_nodemap_entries; i++) {

From 708614e6180f398cd307ea0048d48ba6fa274610 Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Wed, 23 Jul 2008 21:26:51 -0700
Subject: [PATCH 0207/1171] mm: verify the page links and memory model

Print out information on how the page flags are being used if mminit_loglevel
is MMINIT_VERIFY or higher and unconditionally performs sanity checks on the
flags regardless of loglevel.

When the page flags are updated with section, node and zone information, a
check are made to ensure the values can be retrieved correctly.  Finally we
confirm that pfn_to_page and page_to_pfn are the correct inverse functions.

[akpm@linux-foundation.org: fix printk warnings]
Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/internal.h   | 12 +++++++++
 mm/mm_init.c    | 71 +++++++++++++++++++++++++++++++++++++++++++++++++
 mm/page_alloc.c |  8 ++++++
 3 files changed, 91 insertions(+)

diff --git a/mm/internal.h b/mm/internal.h
index a7ee05253294..7a4a2885dc8e 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -78,6 +78,10 @@ do { \
 	} \
 } while (0)
 
+extern void mminit_verify_pageflags_layout(void);
+extern void mminit_verify_page_links(struct page *page,
+		enum zone_type zone, unsigned long nid, unsigned long pfn);
+
 #else
 
 static inline void mminit_dprintk(enum mminit_level level,
@@ -85,5 +89,13 @@ static inline void mminit_dprintk(enum mminit_level level,
 {
 }
 
+static inline void mminit_verify_pageflags_layout(void)
+{
+}
+
+static inline void mminit_verify_page_links(struct page *page,
+		enum zone_type zone, unsigned long nid, unsigned long pfn)
+{
+}
 #endif /* CONFIG_DEBUG_MEMORY_INIT */
 #endif
diff --git a/mm/mm_init.c b/mm/mm_init.c
index c01d8dfec817..e16990d629e6 100644
--- a/mm/mm_init.c
+++ b/mm/mm_init.c
@@ -7,9 +7,80 @@
  */
 #include <linux/kernel.h>
 #include <linux/init.h>
+#include "internal.h"
 
 int __meminitdata mminit_loglevel;
 
+void __init mminit_verify_pageflags_layout(void)
+{
+	int shift, width;
+	unsigned long or_mask, add_mask;
+
+	shift = 8 * sizeof(unsigned long);
+	width = shift - SECTIONS_WIDTH - NODES_WIDTH - ZONES_WIDTH;
+	mminit_dprintk(MMINIT_TRACE, "pageflags_layout_widths",
+		"Section %d Node %d Zone %d Flags %d\n",
+		SECTIONS_WIDTH,
+		NODES_WIDTH,
+		ZONES_WIDTH,
+		NR_PAGEFLAGS);
+	mminit_dprintk(MMINIT_TRACE, "pageflags_layout_shifts",
+		"Section %d Node %d Zone %d\n",
+#ifdef SECTIONS_SHIFT
+		SECTIONS_SHIFT,
+#else
+		0,
+#endif
+		NODES_SHIFT,
+		ZONES_SHIFT);
+	mminit_dprintk(MMINIT_TRACE, "pageflags_layout_offsets",
+		"Section %lu Node %lu Zone %lu\n",
+		(unsigned long)SECTIONS_PGSHIFT,
+		(unsigned long)NODES_PGSHIFT,
+		(unsigned long)ZONES_PGSHIFT);
+	mminit_dprintk(MMINIT_TRACE, "pageflags_layout_zoneid",
+		"Zone ID: %lu -> %lu\n",
+		(unsigned long)ZONEID_PGOFF,
+		(unsigned long)(ZONEID_PGOFF + ZONEID_SHIFT));
+	mminit_dprintk(MMINIT_TRACE, "pageflags_layout_usage",
+		"location: %d -> %d unused %d -> %d flags %d -> %d\n",
+		shift, width, width, NR_PAGEFLAGS, NR_PAGEFLAGS, 0);
+#ifdef NODE_NOT_IN_PAGE_FLAGS
+	mminit_dprintk(MMINIT_TRACE, "pageflags_layout_nodeflags",
+		"Node not in page flags");
+#endif
+
+	if (SECTIONS_WIDTH) {
+		shift -= SECTIONS_WIDTH;
+		BUG_ON(shift != SECTIONS_PGSHIFT);
+	}
+	if (NODES_WIDTH) {
+		shift -= NODES_WIDTH;
+		BUG_ON(shift != NODES_PGSHIFT);
+	}
+	if (ZONES_WIDTH) {
+		shift -= ZONES_WIDTH;
+		BUG_ON(shift != ZONES_PGSHIFT);
+	}
+
+	/* Check for bitmask overlaps */
+	or_mask = (ZONES_MASK << ZONES_PGSHIFT) |
+			(NODES_MASK << NODES_PGSHIFT) |
+			(SECTIONS_MASK << SECTIONS_PGSHIFT);
+	add_mask = (ZONES_MASK << ZONES_PGSHIFT) +
+			(NODES_MASK << NODES_PGSHIFT) +
+			(SECTIONS_MASK << SECTIONS_PGSHIFT);
+	BUG_ON(or_mask != add_mask);
+}
+
+void __meminit mminit_verify_page_links(struct page *page, enum zone_type zone,
+			unsigned long nid, unsigned long pfn)
+{
+	BUG_ON(page_to_nid(page) != nid);
+	BUG_ON(page_zonenum(page) != zone);
+	BUG_ON(page_to_pfn(page) != pfn);
+}
+
 static __init int set_mminit_loglevel(char *str)
 {
 	get_option(&str, &mminit_loglevel);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 0908352ba727..acab6ad326df 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -2534,6 +2534,7 @@ void __meminit memmap_init_zone(unsigned long size, int nid, unsigned long zone,
 		}
 		page = pfn_to_page(pfn);
 		set_page_links(page, zone, nid, pfn);
+		mminit_verify_page_links(page, zone, nid, pfn);
 		init_page_count(page);
 		reset_page_mapcount(page);
 		SetPageReserved(page);
@@ -2836,6 +2837,12 @@ __meminit int init_currently_empty_zone(struct zone *zone,
 
 	zone->zone_start_pfn = zone_start_pfn;
 
+	mminit_dprintk(MMINIT_TRACE, "memmap_init",
+			"Initialising map node %d zone %lu pfns %lu -> %lu\n",
+			pgdat->node_id,
+			(unsigned long)zone_idx(zone),
+			zone_start_pfn, (zone_start_pfn + size));
+
 	zone_init_free_lists(zone);
 
 	return 0;
@@ -3961,6 +3968,7 @@ void __init free_area_init_nodes(unsigned long *max_zone_pfn)
 						early_node_map[i].end_pfn);
 
 	/* Initialise every node */
+	mminit_verify_pageflags_layout();
 	setup_nr_node_ids();
 	for_each_online_node(nid) {
 		pg_data_t *pgdat = NODE_DATA(nid);

From 2dbb51c49f4fecb8330e43247a0edfbc4b2b8974 Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Wed, 23 Jul 2008 21:26:52 -0700
Subject: [PATCH 0208/1171] mm: make defensive checks around PFN values
 registered for memory usage

There are a number of different views to how much memory is currently active.
There is the arch-independent zone-sizing view, the bootmem allocator and
memory models view.

Architectures register this information at different times and is not
necessarily in sync particularly with respect to some SPARSEMEM limitations.

This patch introduces mminit_validate_memmodel_limits() which is able to
validate and correct PFN ranges with respect to the memory model.  It is only
SPARSEMEM that currently validates itself.

Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/bootmem.c    |  1 +
 mm/internal.h   | 12 ++++++++++++
 mm/page_alloc.c |  2 ++
 mm/sparse.c     | 37 +++++++++++++++++++++++++++++--------
 4 files changed, 44 insertions(+), 8 deletions(-)

diff --git a/mm/bootmem.c b/mm/bootmem.c
index 8d9f60e06f62..9f4bbc5da73f 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -91,6 +91,7 @@ static unsigned long __init init_bootmem_core(pg_data_t *pgdat,
 	bootmem_data_t *bdata = pgdat->bdata;
 	unsigned long mapsize;
 
+	mminit_validate_memmodel_limits(&start, &end);
 	bdata->node_bootmem_map = phys_to_virt(PFN_PHYS(mapstart));
 	bdata->node_boot_start = PFN_PHYS(start);
 	bdata->node_low_pfn = end;
diff --git a/mm/internal.h b/mm/internal.h
index 7a4a2885dc8e..5d17f3efac41 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -98,4 +98,16 @@ static inline void mminit_verify_page_links(struct page *page,
 {
 }
 #endif /* CONFIG_DEBUG_MEMORY_INIT */
+
+/* mminit_validate_memmodel_limits is independent of CONFIG_DEBUG_MEMORY_INIT */
+#if defined(CONFIG_SPARSEMEM)
+extern void mminit_validate_memmodel_limits(unsigned long *start_pfn,
+				unsigned long *end_pfn);
+#else
+static inline void mminit_validate_memmodel_limits(unsigned long *start_pfn,
+				unsigned long *end_pfn)
+{
+}
+#endif /* CONFIG_SPARSEMEM */
+
 #endif
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index acab6ad326df..0adb66e711e6 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -3536,6 +3536,8 @@ void __init add_active_range(unsigned int nid, unsigned long start_pfn,
 			nid, start_pfn, end_pfn,
 			nr_nodemap_entries, MAX_ACTIVE_REGIONS);
 
+	mminit_validate_memmodel_limits(&start_pfn, &end_pfn);
+
 	/* Merge with existing active regions if possible */
 	for (i = 0; i < nr_nodemap_entries; i++) {
 		if (early_node_map[i].nid != nid)
diff --git a/mm/sparse.c b/mm/sparse.c
index 36511c7b5e2c..7a3650923d9a 100644
--- a/mm/sparse.c
+++ b/mm/sparse.c
@@ -12,6 +12,7 @@
 #include <asm/dma.h>
 #include <asm/pgalloc.h>
 #include <asm/pgtable.h>
+#include "internal.h"
 
 /*
  * Permanent SPARSEMEM data:
@@ -147,22 +148,41 @@ static inline int sparse_early_nid(struct mem_section *section)
 	return (section->section_mem_map >> SECTION_NID_SHIFT);
 }
 
-/* Record a memory area against a node. */
-void __init memory_present(int nid, unsigned long start, unsigned long end)
+/* Validate the physical addressing limitations of the model */
+void __meminit mminit_validate_memmodel_limits(unsigned long *start_pfn,
+						unsigned long *end_pfn)
 {
-	unsigned long max_arch_pfn = 1UL << (MAX_PHYSMEM_BITS-PAGE_SHIFT);
-	unsigned long pfn;
+	unsigned long max_sparsemem_pfn = 1UL << (MAX_PHYSMEM_BITS-PAGE_SHIFT);
 
 	/*
 	 * Sanity checks - do not allow an architecture to pass
 	 * in larger pfns than the maximum scope of sparsemem:
 	 */
-	if (start >= max_arch_pfn)
-		return;
-	if (end >= max_arch_pfn)
-		end = max_arch_pfn;
+	if (*start_pfn > max_sparsemem_pfn) {
+		mminit_dprintk(MMINIT_WARNING, "pfnvalidation",
+			"Start of range %lu -> %lu exceeds SPARSEMEM max %lu\n",
+			*start_pfn, *end_pfn, max_sparsemem_pfn);
+		WARN_ON_ONCE(1);
+		*start_pfn = max_sparsemem_pfn;
+		*end_pfn = max_sparsemem_pfn;
+	}
+
+	if (*end_pfn > max_sparsemem_pfn) {
+		mminit_dprintk(MMINIT_WARNING, "pfnvalidation",
+			"End of range %lu -> %lu exceeds SPARSEMEM max %lu\n",
+			*start_pfn, *end_pfn, max_sparsemem_pfn);
+		WARN_ON_ONCE(1);
+		*end_pfn = max_sparsemem_pfn;
+	}
+}
+
+/* Record a memory area against a node. */
+void __init memory_present(int nid, unsigned long start, unsigned long end)
+{
+	unsigned long pfn;
 
 	start &= PAGE_SECTION_MASK;
+	mminit_validate_memmodel_limits(&start, &end);
 	for (pfn = start; pfn < end; pfn += PAGES_PER_SECTION) {
 		unsigned long section = pfn_to_section_nr(pfn);
 		struct mem_section *ms;
@@ -187,6 +207,7 @@ unsigned long __init node_memmap_size_bytes(int nid, unsigned long start_pfn,
 	unsigned long pfn;
 	unsigned long nr_pages = 0;
 
+	mminit_validate_memmodel_limits(&start_pfn, &end_pfn);
 	for (pfn = start_pfn; pfn < end_pfn; pfn += PAGES_PER_SECTION) {
 		if (nid != early_pfn_to_nid(pfn))
 			continue;

From 68ad8df42e12037c3894c9706ab428bf5cd6426b Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Wed, 23 Jul 2008 21:26:52 -0700
Subject: [PATCH 0209/1171] mm: print out the zonelists on request for manual
 verification

This patch prints out the zonelists during boot for manual verification by the
user if the mminit_loglevel is MMINIT_VERIFY or higher.

Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/internal.h   |  5 +++++
 mm/mm_init.c    | 45 +++++++++++++++++++++++++++++++++++++++++++++
 mm/page_alloc.c |  1 +
 3 files changed, 51 insertions(+)

diff --git a/mm/internal.h b/mm/internal.h
index 5d17f3efac41..50807e12490e 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -81,6 +81,7 @@ do { \
 extern void mminit_verify_pageflags_layout(void);
 extern void mminit_verify_page_links(struct page *page,
 		enum zone_type zone, unsigned long nid, unsigned long pfn);
+extern void mminit_verify_zonelist(void);
 
 #else
 
@@ -97,6 +98,10 @@ static inline void mminit_verify_page_links(struct page *page,
 		enum zone_type zone, unsigned long nid, unsigned long pfn)
 {
 }
+
+static inline void mminit_verify_zonelist(void)
+{
+}
 #endif /* CONFIG_DEBUG_MEMORY_INIT */
 
 /* mminit_validate_memmodel_limits is independent of CONFIG_DEBUG_MEMORY_INIT */
diff --git a/mm/mm_init.c b/mm/mm_init.c
index e16990d629e6..ce445ca097e7 100644
--- a/mm/mm_init.c
+++ b/mm/mm_init.c
@@ -11,6 +11,51 @@
 
 int __meminitdata mminit_loglevel;
 
+/* The zonelists are simply reported, validation is manual. */
+void mminit_verify_zonelist(void)
+{
+	int nid;
+
+	if (mminit_loglevel < MMINIT_VERIFY)
+		return;
+
+	for_each_online_node(nid) {
+		pg_data_t *pgdat = NODE_DATA(nid);
+		struct zone *zone;
+		struct zoneref *z;
+		struct zonelist *zonelist;
+		int i, listid, zoneid;
+
+		BUG_ON(MAX_ZONELISTS > 2);
+		for (i = 0; i < MAX_ZONELISTS * MAX_NR_ZONES; i++) {
+
+			/* Identify the zone and nodelist */
+			zoneid = i % MAX_NR_ZONES;
+			listid = i / MAX_NR_ZONES;
+			zonelist = &pgdat->node_zonelists[listid];
+			zone = &pgdat->node_zones[zoneid];
+			if (!populated_zone(zone))
+				continue;
+
+			/* Print information about the zonelist */
+			printk(KERN_DEBUG "mminit::zonelist %s %d:%s = ",
+				listid > 0 ? "thisnode" : "general", nid,
+				zone->name);
+
+			/* Iterate the zonelist */
+			for_each_zone_zonelist(zone, z, zonelist, zoneid) {
+#ifdef CONFIG_NUMA
+				printk(KERN_CONT "%d:%s ",
+					zone->node, zone->name);
+#else
+				printk(KERN_CONT "0:%s ", zone->name);
+#endif /* CONFIG_NUMA */
+			}
+			printk(KERN_CONT "\n");
+		}
+	}
+}
+
 void __init mminit_verify_pageflags_layout(void)
 {
 	int shift, width;
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 0adb66e711e6..9ece07ce65b0 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -2352,6 +2352,7 @@ void build_all_zonelists(void)
 
 	if (system_state == SYSTEM_BOOTING) {
 		__build_all_zonelists(NULL);
+		mminit_verify_zonelist();
 		cpuset_init_current_mems_allowed();
 	} else {
 		/* we have to stop all cpus to guarantee there is no user

From 8b05c7e6e159d2f33c9275281b8b909a89eb7c5d Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Wed, 23 Jul 2008 21:26:53 -0700
Subject: [PATCH 0210/1171] add a helper function to test if an object is on
 the stack

lib/debugobjects.c has a function to test if an object is on the stack.
The block layer and ide needs it (they need to avoid DMA from/to stack
buffers).  This patch moves the function to include/linux/sched.h so that
everyone can use it.

lib/debugobjects.c uses current->stack but this patch uses a
task_stack_page() accessor, which is a preferable way to access the stack.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/sched.h | 7 +++++++
 lib/debugobjects.c    | 4 +---
 2 files changed, 8 insertions(+), 3 deletions(-)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index dc7e592c473a..6aca4a16e377 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1983,6 +1983,13 @@ static inline unsigned long *end_of_stack(struct task_struct *p)
 
 #endif
 
+static inline int object_is_on_stack(void *obj)
+{
+	void *stack = task_stack_page(current);
+
+	return (obj >= stack) && (obj < (stack + THREAD_SIZE));
+}
+
 extern void thread_info_cache_init(void);
 
 /* set thread flags in other task's structures
diff --git a/lib/debugobjects.c b/lib/debugobjects.c
index 85b18d79be89..f86196390cfd 100644
--- a/lib/debugobjects.c
+++ b/lib/debugobjects.c
@@ -226,15 +226,13 @@ debug_object_fixup(int (*fixup)(void *addr, enum debug_obj_state state),
 
 static void debug_object_is_on_stack(void *addr, int onstack)
 {
-	void *stack = current->stack;
 	int is_on_stack;
 	static int limit;
 
 	if (limit > 4)
 		return;
 
-	is_on_stack = (addr >= stack && addr < (stack + THREAD_SIZE));
-
+	is_on_stack = object_is_on_stack(addr);
 	if (is_on_stack == onstack)
 		return;
 

From b61bfa3c462671c48a51fb5c31af337c5a996a04 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Wed, 23 Jul 2008 21:26:55 -0700
Subject: [PATCH 0211/1171] mm: move bootmem descriptors definition to a single
 place

There are a lot of places that define either a single bootmem descriptor or an
array of them.  Use only one central array with MAX_NUMNODES items instead.

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Acked-by: Ralf Baechle <ralf@linux-mips.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Hirokazu Takata <takata@linux-m32r.org>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Kyle McMartin <kyle@parisc-linux.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: David S. Miller <davem@davemloft.net>
Cc: Yinghai Lu <yhlu.kernel@gmail.com>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/alpha/mm/numa.c             |  8 ++++----
 arch/arm/mm/discontig.c          | 34 +++++++++++++++-----------------
 arch/ia64/mm/discontig.c         | 11 +++++------
 arch/m32r/mm/discontig.c         |  4 +---
 arch/m68k/mm/init.c              |  4 +---
 arch/mips/sgi-ip27/ip27-memory.c |  4 +---
 arch/parisc/mm/init.c            |  3 +--
 arch/powerpc/mm/numa.c           |  3 +--
 arch/sh/mm/numa.c                |  5 ++---
 arch/sparc64/mm/init.c           |  3 +--
 arch/x86/mm/discontig_32.c       |  3 +--
 arch/x86/mm/numa_64.c            |  4 +---
 include/linux/bootmem.h          |  2 ++
 mm/bootmem.c                     |  2 ++
 mm/page_alloc.c                  |  4 +---
 15 files changed, 40 insertions(+), 54 deletions(-)

diff --git a/arch/alpha/mm/numa.c b/arch/alpha/mm/numa.c
index 10ab7833e83c..a53fda0481ca 100644
--- a/arch/alpha/mm/numa.c
+++ b/arch/alpha/mm/numa.c
@@ -19,7 +19,6 @@
 #include <asm/pgalloc.h>
 
 pg_data_t node_data[MAX_NUMNODES];
-bootmem_data_t node_bdata[MAX_NUMNODES];
 EXPORT_SYMBOL(node_data);
 
 #undef DEBUG_DISCONTIG
@@ -141,7 +140,7 @@ setup_memory_node(int nid, void *kernel_end)
 		printk(" not enough mem to reserve NODE_DATA");
 		return;
 	}
-	NODE_DATA(nid)->bdata = &node_bdata[nid];
+	NODE_DATA(nid)->bdata = &bootmem_node_data[nid];
 
 	printk(" Detected node memory:   start %8lu, end %8lu\n",
 	       node_min_pfn, node_max_pfn);
@@ -304,8 +303,9 @@ void __init paging_init(void)
 	dma_local_pfn = virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT;
 
 	for_each_online_node(nid) {
-		unsigned long start_pfn = node_bdata[nid].node_boot_start >> PAGE_SHIFT;
-		unsigned long end_pfn = node_bdata[nid].node_low_pfn;
+		bootmem_data_t *bdata = &bootmem_node_data[nid];
+		unsigned long start_pfn = bdata->node_boot_start >> PAGE_SHIFT;
+		unsigned long end_pfn = bdata->node_low_pfn;
 
 		if (dma_local_pfn >= end_pfn - start_pfn)
 			zones_size[ZONE_DMA] = end_pfn - start_pfn;
diff --git a/arch/arm/mm/discontig.c b/arch/arm/mm/discontig.c
index 1e5602189507..c8c0c4b0f0a3 100644
--- a/arch/arm/mm/discontig.c
+++ b/arch/arm/mm/discontig.c
@@ -21,26 +21,24 @@
  * Our node_data structure for discontiguous memory.
  */
 
-static bootmem_data_t node_bootmem_data[MAX_NUMNODES];
-
 pg_data_t discontig_node_data[MAX_NUMNODES] = {
-  { .bdata = &node_bootmem_data[0] },
-  { .bdata = &node_bootmem_data[1] },
-  { .bdata = &node_bootmem_data[2] },
-  { .bdata = &node_bootmem_data[3] },
+  { .bdata = &bootmem_node_data[0] },
+  { .bdata = &bootmem_node_data[1] },
+  { .bdata = &bootmem_node_data[2] },
+  { .bdata = &bootmem_node_data[3] },
 #if MAX_NUMNODES == 16
-  { .bdata = &node_bootmem_data[4] },
-  { .bdata = &node_bootmem_data[5] },
-  { .bdata = &node_bootmem_data[6] },
-  { .bdata = &node_bootmem_data[7] },
-  { .bdata = &node_bootmem_data[8] },
-  { .bdata = &node_bootmem_data[9] },
-  { .bdata = &node_bootmem_data[10] },
-  { .bdata = &node_bootmem_data[11] },
-  { .bdata = &node_bootmem_data[12] },
-  { .bdata = &node_bootmem_data[13] },
-  { .bdata = &node_bootmem_data[14] },
-  { .bdata = &node_bootmem_data[15] },
+  { .bdata = &bootmem_node_data[4] },
+  { .bdata = &bootmem_node_data[5] },
+  { .bdata = &bootmem_node_data[6] },
+  { .bdata = &bootmem_node_data[7] },
+  { .bdata = &bootmem_node_data[8] },
+  { .bdata = &bootmem_node_data[9] },
+  { .bdata = &bootmem_node_data[10] },
+  { .bdata = &bootmem_node_data[11] },
+  { .bdata = &bootmem_node_data[12] },
+  { .bdata = &bootmem_node_data[13] },
+  { .bdata = &bootmem_node_data[14] },
+  { .bdata = &bootmem_node_data[15] },
 #endif
 };
 
diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c
index 544dc420c65e..2fcf8464331e 100644
--- a/arch/ia64/mm/discontig.c
+++ b/arch/ia64/mm/discontig.c
@@ -36,7 +36,6 @@ struct early_node_data {
 	struct ia64_node_data *node_data;
 	unsigned long pernode_addr;
 	unsigned long pernode_size;
-	struct bootmem_data bootmem_data;
 	unsigned long num_physpages;
 #ifdef CONFIG_ZONE_DMA
 	unsigned long num_dma_physpages;
@@ -76,7 +75,7 @@ static int __init build_node_maps(unsigned long start, unsigned long len,
 				  int node)
 {
 	unsigned long cstart, epfn, end = start + len;
-	struct bootmem_data *bdp = &mem_data[node].bootmem_data;
+	struct bootmem_data *bdp = &bootmem_node_data[node];
 
 	epfn = GRANULEROUNDUP(end) >> PAGE_SHIFT;
 	cstart = GRANULEROUNDDOWN(start);
@@ -167,7 +166,7 @@ static void __init fill_pernode(int node, unsigned long pernode,
 {
 	void *cpu_data;
 	int cpus = early_nr_cpus_node(node);
-	struct bootmem_data *bdp = &mem_data[node].bootmem_data;
+	struct bootmem_data *bdp = &bootmem_node_data[node];
 
 	mem_data[node].pernode_addr = pernode;
 	mem_data[node].pernode_size = pernodesize;
@@ -224,7 +223,7 @@ static int __init find_pernode_space(unsigned long start, unsigned long len,
 {
 	unsigned long epfn;
 	unsigned long pernodesize = 0, pernode, pages, mapsize;
-	struct bootmem_data *bdp = &mem_data[node].bootmem_data;
+	struct bootmem_data *bdp = &bootmem_node_data[node];
 
 	epfn = (start + len) >> PAGE_SHIFT;
 
@@ -440,7 +439,7 @@ void __init find_memory(void)
 	efi_memmap_walk(find_max_min_low_pfn, NULL);
 
 	for_each_online_node(node)
-		if (mem_data[node].bootmem_data.node_low_pfn) {
+		if (bootmem_node_data[node].node_low_pfn) {
 			node_clear(node, memory_less_mask);
 			mem_data[node].min_pfn = ~0UL;
 		}
@@ -460,7 +459,7 @@ void __init find_memory(void)
 		else if (node_isset(node, memory_less_mask))
 			continue;
 
-		bdp = &mem_data[node].bootmem_data;
+		bdp = &bootmem_node_data[node];
 		pernode = mem_data[node].pernode_addr;
 		pernodesize = mem_data[node].pernode_size;
 		map = pernode + pernodesize;
diff --git a/arch/m32r/mm/discontig.c b/arch/m32r/mm/discontig.c
index 07c1af7dc0e2..aa9145ef6cca 100644
--- a/arch/m32r/mm/discontig.c
+++ b/arch/m32r/mm/discontig.c
@@ -20,7 +20,6 @@ extern char _end[];
 
 struct pglist_data *node_data[MAX_NUMNODES];
 EXPORT_SYMBOL(node_data);
-static bootmem_data_t node_bdata[MAX_NUMNODES] __initdata;
 
 pg_data_t m32r_node_data[MAX_NUMNODES];
 
@@ -81,7 +80,7 @@ unsigned long __init setup_memory(void)
 	for_each_online_node(nid) {
 		mp = &mem_prof[nid];
 		NODE_DATA(nid)=(pg_data_t *)&m32r_node_data[nid];
-		NODE_DATA(nid)->bdata = &node_bdata[nid];
+		NODE_DATA(nid)->bdata = &bootmem_node_data[nid];
 		min_pfn = mp->start_pfn;
 		max_pfn = mp->start_pfn + mp->pages;
 		bootmap_size = init_bootmem_node(NODE_DATA(nid), mp->free_pfn,
@@ -163,4 +162,3 @@ unsigned long __init zone_sizes_init(void)
 
 	return holes;
 }
-
diff --git a/arch/m68k/mm/init.c b/arch/m68k/mm/init.c
index d8fb9c5303cc..79f5f94d4800 100644
--- a/arch/m68k/mm/init.c
+++ b/arch/m68k/mm/init.c
@@ -32,8 +32,6 @@
 
 DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
 
-static bootmem_data_t __initdata bootmem_data[MAX_NUMNODES];
-
 pg_data_t pg_data_map[MAX_NUMNODES];
 EXPORT_SYMBOL(pg_data_map);
 
@@ -58,7 +56,7 @@ void __init m68k_setup_node(int node)
 		pg_data_table[i] = pg_data_map + node;
 	}
 #endif
-	pg_data_map[node].bdata = bootmem_data + node;
+	pg_data_map[node].bdata = bootmem_node_data + node;
 	node_set_online(node);
 }
 
diff --git a/arch/mips/sgi-ip27/ip27-memory.c b/arch/mips/sgi-ip27/ip27-memory.c
index 42cd10956306..060d853d7b35 100644
--- a/arch/mips/sgi-ip27/ip27-memory.c
+++ b/arch/mips/sgi-ip27/ip27-memory.c
@@ -33,8 +33,6 @@
 #define SLOT_PFNSHIFT           (SLOT_SHIFT - PAGE_SHIFT)
 #define PFN_NASIDSHFT           (NASID_SHFT - PAGE_SHIFT)
 
-static struct bootmem_data __initdata plat_node_bdata[MAX_COMPACT_NODES];
-
 struct node_data *__node_data[MAX_COMPACT_NODES];
 
 EXPORT_SYMBOL(__node_data);
@@ -403,7 +401,7 @@ static void __init node_mem_init(cnodeid_t node)
 	 */
 	__node_data[node] = __va(slot_freepfn << PAGE_SHIFT);
 
-	NODE_DATA(node)->bdata = &plat_node_bdata[node];
+	NODE_DATA(node)->bdata = &bootmem_node_data[node];
 	NODE_DATA(node)->node_start_pfn = start_pfn;
 	NODE_DATA(node)->node_spanned_pages = end_pfn - start_pfn;
 
diff --git a/arch/parisc/mm/init.c b/arch/parisc/mm/init.c
index b4d6c8777ed0..0ddf4904640a 100644
--- a/arch/parisc/mm/init.c
+++ b/arch/parisc/mm/init.c
@@ -36,7 +36,6 @@ extern int  data_start;
 
 #ifdef CONFIG_DISCONTIGMEM
 struct node_map_data node_data[MAX_NUMNODES] __read_mostly;
-bootmem_data_t bmem_data[MAX_NUMNODES] __read_mostly;
 unsigned char pfnnid_map[PFNNID_MAP_MAX] __read_mostly;
 #endif
 
@@ -262,7 +261,7 @@ static void __init setup_bootmem(void)
 #ifdef CONFIG_DISCONTIGMEM
 	for (i = 0; i < MAX_PHYSMEM_RANGES; i++) {
 		memset(NODE_DATA(i), 0, sizeof(pg_data_t));
-		NODE_DATA(i)->bdata = &bmem_data[i];
+		NODE_DATA(i)->bdata = &bootmem_node_data[i];
 	}
 	memset(pfnnid_map, 0xff, sizeof(pfnnid_map));
 
diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c
index cf4bffba6f7c..d9a181351332 100644
--- a/arch/powerpc/mm/numa.c
+++ b/arch/powerpc/mm/numa.c
@@ -39,7 +39,6 @@ EXPORT_SYMBOL(numa_cpu_lookup_table);
 EXPORT_SYMBOL(numa_cpumask_lookup_table);
 EXPORT_SYMBOL(node_data);
 
-static bootmem_data_t __initdata plat_node_bdata[MAX_NUMNODES];
 static int min_common_depth;
 static int n_mem_addr_cells, n_mem_size_cells;
 
@@ -816,7 +815,7 @@ void __init do_init_bootmem(void)
   		dbg("node %d\n", nid);
 		dbg("NODE_DATA() = %p\n", NODE_DATA(nid));
 
-		NODE_DATA(nid)->bdata = &plat_node_bdata[nid];
+		NODE_DATA(nid)->bdata = &bootmem_node_data[nid];
 		NODE_DATA(nid)->node_start_pfn = start_pfn;
 		NODE_DATA(nid)->node_spanned_pages = end_pfn - start_pfn;
 
diff --git a/arch/sh/mm/numa.c b/arch/sh/mm/numa.c
index 1663199ce888..095d93bec7cd 100644
--- a/arch/sh/mm/numa.c
+++ b/arch/sh/mm/numa.c
@@ -14,7 +14,6 @@
 #include <linux/pfn.h>
 #include <asm/sections.h>
 
-static bootmem_data_t plat_node_bdata[MAX_NUMNODES];
 struct pglist_data *node_data[MAX_NUMNODES] __read_mostly;
 EXPORT_SYMBOL_GPL(node_data);
 
@@ -35,7 +34,7 @@ void __init setup_memory(void)
 	NODE_DATA(0) = pfn_to_kaddr(free_pfn);
 	memset(NODE_DATA(0), 0, sizeof(struct pglist_data));
 	free_pfn += PFN_UP(sizeof(struct pglist_data));
-	NODE_DATA(0)->bdata = &plat_node_bdata[0];
+	NODE_DATA(0)->bdata = &bootmem_node_data[0];
 
 	/* Set up node 0 */
 	setup_bootmem_allocator(free_pfn);
@@ -66,7 +65,7 @@ void __init setup_bootmem_node(int nid, unsigned long start, unsigned long end)
 	free_pfn += PFN_UP(sizeof(struct pglist_data));
 	memset(NODE_DATA(nid), 0, sizeof(struct pglist_data));
 
-	NODE_DATA(nid)->bdata = &plat_node_bdata[nid];
+	NODE_DATA(nid)->bdata = &bootmem_node_data[nid];
 	NODE_DATA(nid)->node_start_pfn = start_pfn;
 	NODE_DATA(nid)->node_spanned_pages = end_pfn - start_pfn;
 
diff --git a/arch/sparc64/mm/init.c b/arch/sparc64/mm/init.c
index 84898c44dd4d..713297473951 100644
--- a/arch/sparc64/mm/init.c
+++ b/arch/sparc64/mm/init.c
@@ -788,7 +788,6 @@ int numa_cpu_lookup_table[NR_CPUS];
 cpumask_t numa_cpumask_lookup_table[MAX_NUMNODES];
 
 #ifdef CONFIG_NEED_MULTIPLE_NODES
-static bootmem_data_t plat_node_bdata[MAX_NUMNODES];
 
 struct mdesc_mblock {
 	u64	base;
@@ -871,7 +870,7 @@ static void __init allocate_node_data(int nid)
 	NODE_DATA(nid) = __va(paddr);
 	memset(NODE_DATA(nid), 0, sizeof(struct pglist_data));
 
-	NODE_DATA(nid)->bdata = &plat_node_bdata[nid];
+	NODE_DATA(nid)->bdata = &bootmem_node_data[nid];
 #endif
 
 	p = NODE_DATA(nid);
diff --git a/arch/x86/mm/discontig_32.c b/arch/x86/mm/discontig_32.c
index 5dfef9fa061a..62fa440678d8 100644
--- a/arch/x86/mm/discontig_32.c
+++ b/arch/x86/mm/discontig_32.c
@@ -42,7 +42,6 @@
 
 struct pglist_data *node_data[MAX_NUMNODES] __read_mostly;
 EXPORT_SYMBOL(node_data);
-static bootmem_data_t node0_bdata;
 
 /*
  * numa interface - we expect the numa architecture specific code to have
@@ -385,7 +384,7 @@ void __init initmem_init(unsigned long start_pfn,
 	for_each_online_node(nid)
 		memset(NODE_DATA(nid), 0, sizeof(struct pglist_data));
 
-	NODE_DATA(0)->bdata = &node0_bdata;
+	NODE_DATA(0)->bdata = &bootmem_node_data[0];
 	setup_bootmem_allocator();
 }
 
diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c
index 9782f42dd319..a4dd793d6003 100644
--- a/arch/x86/mm/numa_64.c
+++ b/arch/x86/mm/numa_64.c
@@ -23,8 +23,6 @@
 struct pglist_data *node_data[MAX_NUMNODES] __read_mostly;
 EXPORT_SYMBOL(node_data);
 
-static bootmem_data_t plat_node_bdata[MAX_NUMNODES];
-
 struct memnode memnode;
 
 s16 apicid_to_node[MAX_LOCAL_APIC] __cpuinitdata = {
@@ -198,7 +196,7 @@ void __init setup_node_bootmem(int nodeid, unsigned long start,
 		nodedata_phys + pgdat_size - 1);
 
 	memset(NODE_DATA(nodeid), 0, sizeof(pg_data_t));
-	NODE_DATA(nodeid)->bdata = &plat_node_bdata[nodeid];
+	NODE_DATA(nodeid)->bdata = &bootmem_node_data[nodeid];
 	NODE_DATA(nodeid)->node_start_pfn = start_pfn;
 	NODE_DATA(nodeid)->node_spanned_pages = last_pfn - start_pfn;
 
diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index a1d9b79078ea..2599c741405e 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -38,6 +38,8 @@ typedef struct bootmem_data {
 	struct list_head list;
 } bootmem_data_t;
 
+extern bootmem_data_t bootmem_node_data[];
+
 extern unsigned long bootmem_bootmap_pages(unsigned long);
 extern unsigned long init_bootmem(unsigned long addr, unsigned long memend);
 extern void free_bootmem(unsigned long addr, unsigned long size);
diff --git a/mm/bootmem.c b/mm/bootmem.c
index 9f4bbc5da73f..35b3cb667036 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -36,6 +36,8 @@ static LIST_HEAD(bdata_list);
 unsigned long saved_max_pfn;
 #endif
 
+bootmem_data_t bootmem_node_data[MAX_NUMNODES] __initdata;
+
 /* return the number of _pages_ that will be allocated for the boot bitmap */
 unsigned long __init bootmem_bootmap_pages(unsigned long pages)
 {
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 9ece07ce65b0..e089b92cdfff 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -4040,9 +4040,7 @@ void __init set_dma_reserve(unsigned long new_dma_reserve)
 }
 
 #ifndef CONFIG_NEED_MULTIPLE_NODES
-static bootmem_data_t contig_bootmem_data;
-struct pglist_data contig_page_data = { .bdata = &contig_bootmem_data };
-
+struct pglist_data contig_page_data = { .bdata = &bootmem_node_data[0] };
 EXPORT_SYMBOL(contig_page_data);
 #endif
 

From 6b312c0e6e2f44b020e12953d1dd37eed60e3609 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Wed, 23 Jul 2008 21:26:58 -0700
Subject: [PATCH 0212/1171] mm: fix free_all_bootmem_core alignment check

The check for node_boot_start is bogus because we start freeing at the
corresponding pfn.  So check if the pfn is properly aligned instead in a more
readable way and adjust the documentation.

Also remove an unneeded accounting variable.

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Yinghai Lu <yhlu.kernel@gmail.com>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/bootmem.c | 21 ++++++++++-----------
 1 file changed, 10 insertions(+), 11 deletions(-)

diff --git a/mm/bootmem.c b/mm/bootmem.c
index 35b3cb667036..319a79bce7cd 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -377,7 +377,7 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat)
 	struct page *page;
 	unsigned long pfn;
 	bootmem_data_t *bdata = pgdat->bdata;
-	unsigned long i, count, total = 0;
+	unsigned long i, count;
 	unsigned long idx;
 	unsigned long *map; 
 	int gofast = 0;
@@ -389,10 +389,13 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat)
 	pfn = PFN_DOWN(bdata->node_boot_start);
 	idx = bdata->node_low_pfn - pfn;
 	map = bdata->node_bootmem_map;
-	/* Check physaddr is O(LOG2(BITS_PER_LONG)) page aligned */
-	if (bdata->node_boot_start == 0 ||
-	    ffs(bdata->node_boot_start) - PAGE_SHIFT > ffs(BITS_PER_LONG))
+	/*
+	 * Check if we are aligned to BITS_PER_LONG pages.  If so, we might
+	 * be able to free page orders of that size at once.
+	 */
+	if (!(pfn & (BITS_PER_LONG-1)))
 		gofast = 1;
+
 	for (i = 0; i < idx; ) {
 		unsigned long v = ~map[i / BITS_PER_LONG];
 
@@ -420,23 +423,19 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat)
 		}
 		pfn += BITS_PER_LONG;
 	}
-	total += count;
 
 	/*
 	 * Now free the allocator bitmap itself, it's not
 	 * needed anymore:
 	 */
 	page = virt_to_page(bdata->node_bootmem_map);
-	count = 0;
 	idx = (get_mapsize(bdata) + PAGE_SIZE-1) >> PAGE_SHIFT;
-	for (i = 0; i < idx; i++, page++) {
+	for (i = 0; i < idx; i++, page++)
 		__free_pages_bootmem(page, 0);
-		count++;
-	}
-	total += count;
+	count += i;
 	bdata->node_bootmem_map = NULL;
 
-	return total;
+	return count;
 }
 
 unsigned long __init init_bootmem_node(pg_data_t *pgdat, unsigned long freepfn,

From 8ae04463077324ed9f6b04ab3a5b17ae1ee4dd35 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Wed, 23 Jul 2008 21:26:59 -0700
Subject: [PATCH 0213/1171] mm: normalize internal argument passing of bootmem
 data

All _core functions only need the bootmem data, not the whole node descriptor.
Adjust the two functions that take the node descriptor unneededly.

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Yinghai Lu <yhlu.kernel@gmail.com>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/bootmem.c | 14 ++++++--------
 1 file changed, 6 insertions(+), 8 deletions(-)

diff --git a/mm/bootmem.c b/mm/bootmem.c
index 319a79bce7cd..251c66c5d96a 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -87,10 +87,9 @@ static unsigned long __init get_mapsize(bootmem_data_t *bdata)
 /*
  * Called once to set up the allocator itself.
  */
-static unsigned long __init init_bootmem_core(pg_data_t *pgdat,
+static unsigned long __init init_bootmem_core(bootmem_data_t *bdata,
 	unsigned long mapstart, unsigned long start, unsigned long end)
 {
-	bootmem_data_t *bdata = pgdat->bdata;
 	unsigned long mapsize;
 
 	mminit_validate_memmodel_limits(&start, &end);
@@ -372,11 +371,10 @@ found:
 	return ret;
 }
 
-static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat)
+static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata)
 {
 	struct page *page;
 	unsigned long pfn;
-	bootmem_data_t *bdata = pgdat->bdata;
 	unsigned long i, count;
 	unsigned long idx;
 	unsigned long *map; 
@@ -441,7 +439,7 @@ static unsigned long __init free_all_bootmem_core(pg_data_t *pgdat)
 unsigned long __init init_bootmem_node(pg_data_t *pgdat, unsigned long freepfn,
 				unsigned long startpfn, unsigned long endpfn)
 {
-	return init_bootmem_core(pgdat, freepfn, startpfn, endpfn);
+	return init_bootmem_core(pgdat->bdata, freepfn, startpfn, endpfn);
 }
 
 int __init reserve_bootmem_node(pg_data_t *pgdat, unsigned long physaddr,
@@ -466,14 +464,14 @@ void __init free_bootmem_node(pg_data_t *pgdat, unsigned long physaddr,
 unsigned long __init free_all_bootmem_node(pg_data_t *pgdat)
 {
 	register_page_bootmem_info_node(pgdat);
-	return free_all_bootmem_core(pgdat);
+	return free_all_bootmem_core(pgdat->bdata);
 }
 
 unsigned long __init init_bootmem(unsigned long start, unsigned long pages)
 {
 	max_low_pfn = pages;
 	min_low_pfn = start;
-	return init_bootmem_core(NODE_DATA(0), start, 0, pages);
+	return init_bootmem_core(NODE_DATA(0)->bdata, start, 0, pages);
 }
 
 #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
@@ -504,7 +502,7 @@ void __init free_bootmem(unsigned long addr, unsigned long size)
 
 unsigned long __init free_all_bootmem(void)
 {
-	return free_all_bootmem_core(NODE_DATA(0));
+	return free_all_bootmem_core(NODE_DATA(0)->bdata);
 }
 
 void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align,

From ffc6421f0720f433b5b35b89ff56e998eabff93b Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Wed, 23 Jul 2008 21:26:59 -0700
Subject: [PATCH 0214/1171] mm: unexport __alloc_bootmem_core()

This function has no external callers, so unexport it.  Also fix its naming
inconsistency.

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Yinghai Lu <yhlu.kernel@gmail.com>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/bootmem.h |  5 -----
 mm/bootmem.c            | 24 ++++++++++++------------
 2 files changed, 12 insertions(+), 17 deletions(-)

diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index 2599c741405e..dd8fee6c46d9 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -56,11 +56,6 @@ extern void *__alloc_bootmem_low_node(pg_data_t *pgdat,
 				      unsigned long size,
 				      unsigned long align,
 				      unsigned long goal);
-extern void *__alloc_bootmem_core(struct bootmem_data *bdata,
-				  unsigned long size,
-				  unsigned long align,
-				  unsigned long goal,
-				  unsigned long limit);
 
 /*
  * flags for reserve_bootmem (also if CONFIG_HAVE_ARCH_BOOTMEM_NODE,
diff --git a/mm/bootmem.c b/mm/bootmem.c
index 251c66c5d96a..4bc6ae2fbaa3 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -234,9 +234,9 @@ static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr,
  *
  * NOTE:  This function is _not_ reentrant.
  */
-void * __init
-__alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size,
-	      unsigned long align, unsigned long goal, unsigned long limit)
+static void * __init
+alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size,
+		unsigned long align, unsigned long goal, unsigned long limit)
 {
 	unsigned long areasize, preferred;
 	unsigned long i, start = 0, incr, eidx, end_pfn;
@@ -245,7 +245,7 @@ __alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size,
 	void *node_bootmem_map;
 
 	if (!size) {
-		printk("__alloc_bootmem_core(): zero-sized request\n");
+		printk("alloc_bootmem_core(): zero-sized request\n");
 		BUG();
 	}
 	BUG_ON(align & (align-1));
@@ -512,7 +512,7 @@ void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align,
 	void *ptr;
 
 	list_for_each_entry(bdata, &bdata_list, list) {
-		ptr = __alloc_bootmem_core(bdata, size, align, goal, 0);
+		ptr = alloc_bootmem_core(bdata, size, align, goal, 0);
 		if (ptr)
 			return ptr;
 	}
@@ -540,7 +540,7 @@ void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size,
 {
 	void *ptr;
 
-	ptr = __alloc_bootmem_core(pgdat->bdata, size, align, goal, 0);
+	ptr = alloc_bootmem_core(pgdat->bdata, size, align, goal, 0);
 	if (ptr)
 		return ptr;
 
@@ -559,8 +559,8 @@ void * __init alloc_bootmem_section(unsigned long size,
 	goal = PFN_PHYS(pfn);
 	limit = PFN_PHYS(section_nr_to_pfn(section_nr + 1)) - 1;
 	pgdat = NODE_DATA(early_pfn_to_nid(pfn));
-	ptr = __alloc_bootmem_core(pgdat->bdata, size, SMP_CACHE_BYTES, goal,
-				   limit);
+	ptr = alloc_bootmem_core(pgdat->bdata, size, SMP_CACHE_BYTES, goal,
+				limit);
 
 	if (!ptr)
 		return NULL;
@@ -589,8 +589,8 @@ void * __init __alloc_bootmem_low(unsigned long size, unsigned long align,
 	void *ptr;
 
 	list_for_each_entry(bdata, &bdata_list, list) {
-		ptr = __alloc_bootmem_core(bdata, size, align, goal,
-						ARCH_LOW_ADDRESS_LIMIT);
+		ptr = alloc_bootmem_core(bdata, size, align, goal,
+					ARCH_LOW_ADDRESS_LIMIT);
 		if (ptr)
 			return ptr;
 	}
@@ -606,6 +606,6 @@ void * __init __alloc_bootmem_low(unsigned long size, unsigned long align,
 void * __init __alloc_bootmem_low_node(pg_data_t *pgdat, unsigned long size,
 				       unsigned long align, unsigned long goal)
 {
-	return __alloc_bootmem_core(pgdat->bdata, size, align, goal,
-				    ARCH_LOW_ADDRESS_LIMIT);
+	return alloc_bootmem_core(pgdat->bdata, size, align, goal,
+				ARCH_LOW_ADDRESS_LIMIT);
 }

From efe9e77997f6e0306fedc6efa98df491dcf5ecb0 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Wed, 23 Jul 2008 21:27:00 -0700
Subject: [PATCH 0215/1171] mspec: convert nopfn to fault

[akpm@linux-foundation.org: remove unused variable]
Signed-off-by: Nick Piggin <npiggin@suse.de>
Acked-by: Jes Sorensen <jes@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/mspec.c | 23 ++++++++++++++---------
 1 file changed, 14 insertions(+), 9 deletions(-)

diff --git a/drivers/char/mspec.c b/drivers/char/mspec.c
index fe2a95b5d3c0..30f095a8c2d4 100644
--- a/drivers/char/mspec.c
+++ b/drivers/char/mspec.c
@@ -193,25 +193,23 @@ mspec_close(struct vm_area_struct *vma)
 }
 
 /*
- * mspec_nopfn
+ * mspec_fault
  *
  * Creates a mspec page and maps it to user space.
  */
-static unsigned long
-mspec_nopfn(struct vm_area_struct *vma, unsigned long address)
+static int
+mspec_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 {
 	unsigned long paddr, maddr;
 	unsigned long pfn;
-	int index;
+	pgoff_t index = vmf->pgoff;
 	struct vma_data *vdata = vma->vm_private_data;
 
-	BUG_ON(address < vdata->vm_start || address >= vdata->vm_end);
-	index = (address - vdata->vm_start) >> PAGE_SHIFT;
 	maddr = (volatile unsigned long) vdata->maddr[index];
 	if (maddr == 0) {
 		maddr = uncached_alloc_page(numa_node_id(), 1);
 		if (maddr == 0)
-			return NOPFN_OOM;
+			return VM_FAULT_OOM;
 
 		spin_lock(&vdata->lock);
 		if (vdata->maddr[index] == 0) {
@@ -231,13 +229,20 @@ mspec_nopfn(struct vm_area_struct *vma, unsigned long address)
 
 	pfn = paddr >> PAGE_SHIFT;
 
-	return pfn;
+	/*
+	 * vm_insert_pfn can fail with -EBUSY, but in that case it will
+	 * be because another thread has installed the pte first, so it
+	 * is no problem.
+	 */
+	vm_insert_pfn(vma, (unsigned long)vmf->virtual_address, pfn);
+
+	return VM_FAULT_NOPAGE;
 }
 
 static struct vm_operations_struct mspec_vm_ops = {
 	.open = mspec_open,
 	.close = mspec_close,
-	.nopfn = mspec_nopfn
+	.fault = mspec_fault,
 };
 
 /*

From e4048e5dc4aecec670f48ed007a28779f09cebd6 Mon Sep 17 00:00:00 2001
From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Date: Wed, 23 Jul 2008 21:27:01 -0700
Subject: [PATCH 0216/1171] page allocator: inline some __alloc_pages()
 wrappers

Two zonelist patch series rewrote __page_alloc() largely.  Now, it is just
a wrapper function.  Inlining them will save a function call.

[akpm@linux-foundation.org: export __alloc_pages_internal]
Cc: Lee Schermerhorn <lee.schermerhorn@hp.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/gfp.h | 21 +++++++++++++++++----
 mm/page_alloc.c     | 19 ++-----------------
 2 files changed, 19 insertions(+), 21 deletions(-)

diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index b414be387180..f640ed241422 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -173,11 +173,24 @@ static inline void arch_free_page(struct page *page, int order) { }
 static inline void arch_alloc_page(struct page *page, int order) { }
 #endif
 
-extern struct page *__alloc_pages(gfp_t, unsigned int, struct zonelist *);
+struct page *
+__alloc_pages_internal(gfp_t gfp_mask, unsigned int order,
+		       struct zonelist *zonelist, nodemask_t *nodemask);
+
+static inline struct page *
+__alloc_pages(gfp_t gfp_mask, unsigned int order,
+		struct zonelist *zonelist)
+{
+	return __alloc_pages_internal(gfp_mask, order, zonelist, NULL);
+}
+
+static inline struct page *
+__alloc_pages_nodemask(gfp_t gfp_mask, unsigned int order,
+		struct zonelist *zonelist, nodemask_t *nodemask)
+{
+	return __alloc_pages_internal(gfp_mask, order, zonelist, nodemask);
+}
 
-extern struct page *
-__alloc_pages_nodemask(gfp_t, unsigned int,
-				struct zonelist *, nodemask_t *nodemask);
 
 static inline struct page *alloc_pages_node(int nid, gfp_t gfp_mask,
 						unsigned int order)
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index e089b92cdfff..35b1347d81bb 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1429,7 +1429,7 @@ try_next_zone:
 /*
  * This is the 'heart' of the zoned buddy allocator.
  */
-static struct page *
+struct page *
 __alloc_pages_internal(gfp_t gfp_mask, unsigned int order,
 			struct zonelist *zonelist, nodemask_t *nodemask)
 {
@@ -1632,22 +1632,7 @@ nopage:
 got_pg:
 	return page;
 }
-
-struct page *
-__alloc_pages(gfp_t gfp_mask, unsigned int order,
-		struct zonelist *zonelist)
-{
-	return __alloc_pages_internal(gfp_mask, order, zonelist, NULL);
-}
-
-struct page *
-__alloc_pages_nodemask(gfp_t gfp_mask, unsigned int order,
-		struct zonelist *zonelist, nodemask_t *nodemask)
-{
-	return __alloc_pages_internal(gfp_mask, order, zonelist, nodemask);
-}
-
-EXPORT_SYMBOL(__alloc_pages);
+EXPORT_SYMBOL(__alloc_pages_internal);
 
 /*
  * Common helper functions.

From 4f5ca265788973e3f5a1129a96ee4a9cbf587f2b Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Wed, 23 Jul 2008 21:27:02 -0700
Subject: [PATCH 0217/1171] mm/migrate.c should #include <linux/syscalls.h>

Every file should include the headers containing the externs for its
global functions (in this case for sys_move_pages()).

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Acked-by: Christoph Lameter <cl@linux-foundation.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/migrate.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/mm/migrate.c b/mm/migrate.c
index 55bd355d170d..e7d13a708da0 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -30,6 +30,7 @@
 #include <linux/vmalloc.h>
 #include <linux/security.h>
 #include <linux/memcontrol.h>
+#include <linux/syscalls.h>
 
 #include "internal.h"
 

From c748e1340e0de3fa7fed86f8bdf499be9242afff Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Wed, 23 Jul 2008 21:27:03 -0700
Subject: [PATCH 0218/1171] mm/vmstat.c: proper externs

This patch adds proper extern declarations for five variables in
include/linux/vmstat.h

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/proc/proc_misc.c    | 4 ----
 include/linux/vmstat.h | 6 ++++++
 kernel/sysctl.c        | 2 +-
 mm/vmstat.c            | 1 +
 4 files changed, 8 insertions(+), 5 deletions(-)

diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c
index c652d469dc08..b14f43d25e9e 100644
--- a/fs/proc/proc_misc.c
+++ b/fs/proc/proc_misc.c
@@ -232,7 +232,6 @@ static int meminfo_read_proc(char *page, char **start, off_t off,
 #undef K
 }
 
-extern const struct seq_operations fragmentation_op;
 static int fragmentation_open(struct inode *inode, struct file *file)
 {
 	(void)inode;
@@ -246,7 +245,6 @@ static const struct file_operations fragmentation_file_operations = {
 	.release	= seq_release,
 };
 
-extern const struct seq_operations pagetypeinfo_op;
 static int pagetypeinfo_open(struct inode *inode, struct file *file)
 {
 	return seq_open(file, &pagetypeinfo_op);
@@ -259,7 +257,6 @@ static const struct file_operations pagetypeinfo_file_ops = {
 	.release	= seq_release,
 };
 
-extern const struct seq_operations zoneinfo_op;
 static int zoneinfo_open(struct inode *inode, struct file *file)
 {
 	return seq_open(file, &zoneinfo_op);
@@ -356,7 +353,6 @@ static const struct file_operations proc_devinfo_operations = {
 	.release	= seq_release,
 };
 
-extern const struct seq_operations vmstat_op;
 static int vmstat_open(struct inode *inode, struct file *file)
 {
 	return seq_open(file, &vmstat_op);
diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h
index e83b69346d23..58334d439516 100644
--- a/include/linux/vmstat.h
+++ b/include/linux/vmstat.h
@@ -44,6 +44,12 @@ enum vm_event_item { PGPGIN, PGPGOUT, PSWPIN, PSWPOUT,
 		NR_VM_EVENT_ITEMS
 };
 
+extern const struct seq_operations fragmentation_op;
+extern const struct seq_operations pagetypeinfo_op;
+extern const struct seq_operations zoneinfo_op;
+extern const struct seq_operations vmstat_op;
+extern int sysctl_stat_interval;
+
 #ifdef CONFIG_VM_EVENT_COUNTERS
 /*
  * Light weight per cpu counter implementation.
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 2a7b9d88706b..1f7b3b76a166 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -43,6 +43,7 @@
 #include <linux/limits.h>
 #include <linux/dcache.h>
 #include <linux/syscalls.h>
+#include <linux/vmstat.h>
 #include <linux/nfs_fs.h>
 #include <linux/acpi.h>
 #include <linux/reboot.h>
@@ -80,7 +81,6 @@ extern int sysctl_drop_caches;
 extern int percpu_pagelist_fraction;
 extern int compat_log;
 extern int maps_protect;
-extern int sysctl_stat_interval;
 extern int latencytop_enabled;
 extern int sysctl_nr_open_min, sysctl_nr_open_max;
 #ifdef CONFIG_RCU_TORTURE_TEST
diff --git a/mm/vmstat.c b/mm/vmstat.c
index c3d4a781802f..b0d08e667ece 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -13,6 +13,7 @@
 #include <linux/err.h>
 #include <linux/module.h>
 #include <linux/cpu.h>
+#include <linux/vmstat.h>
 #include <linux/sched.h>
 
 #ifdef CONFIG_VM_EVENT_COUNTERS

From 75353bed36cfbbfb55bbde0896bbf5a02d9ba355 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Wed, 23 Jul 2008 21:27:03 -0700
Subject: [PATCH 0219/1171] mm/hugetlb.c: fix duplicate variable

It's confusing that set_max_huge_pages() contained two different
variables named "ret", and although the code works correctly this should
be fixed.

The inner of the two variables can simply be removed.

Spotted by sparse.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Cc:  "KOSAKI Motohiro" <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/hugetlb.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index ab171274ef21..2c5c9ee4220d 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -603,7 +603,6 @@ static unsigned long set_max_huge_pages(unsigned long count)
 	}
 
 	while (count > persistent_huge_pages) {
-		int ret;
 		/*
 		 * If this allocation races such that we no longer need the
 		 * page, free_huge_page will handle it by freeing the page

From a969e903a944f69309ee5cc9e7c7b08310d1151e Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Wed, 23 Jul 2008 21:27:04 -0700
Subject: [PATCH 0220/1171] kill generic_file_direct_IO()

generic_file_direct_IO is a common helper around the invocation of
->direct_IO.  But there's almost nothing shared between the read and write
side, so we're better off without this helper.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/filemap.c | 117 ++++++++++++++++++++++-----------------------------
 1 file changed, 51 insertions(+), 66 deletions(-)

diff --git a/mm/filemap.c b/mm/filemap.c
index 65d9d9e2b755..6343f3c841b7 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -42,9 +42,6 @@
 
 #include <asm/mman.h>
 
-static ssize_t
-generic_file_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov,
-	loff_t offset, unsigned long nr_segs);
 
 /*
  * Shared mappings implemented 30.11.1994. It's not fully working yet,
@@ -1205,8 +1202,11 @@ generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov,
 			goto out; /* skip atime */
 		size = i_size_read(inode);
 		if (pos < size) {
-			retval = generic_file_direct_IO(READ, iocb,
-						iov, pos, nr_segs);
+			retval = filemap_write_and_wait(mapping);
+			if (!retval) {
+				retval = mapping->a_ops->direct_IO(READ, iocb,
+							iov, pos, nr_segs);
+			}
 			if (retval > 0)
 				*ppos = pos + retval;
 		}
@@ -2004,11 +2004,55 @@ generic_file_direct_write(struct kiocb *iocb, const struct iovec *iov,
 	struct address_space *mapping = file->f_mapping;
 	struct inode	*inode = mapping->host;
 	ssize_t		written;
+	size_t		write_len;
+	pgoff_t		end;
 
 	if (count != ocount)
 		*nr_segs = iov_shorten((struct iovec *)iov, *nr_segs, count);
 
-	written = generic_file_direct_IO(WRITE, iocb, iov, pos, *nr_segs);
+	/*
+	 * Unmap all mmappings of the file up-front.
+	 *
+	 * This will cause any pte dirty bits to be propagated into the
+	 * pageframes for the subsequent filemap_write_and_wait().
+	 */
+	write_len = iov_length(iov, *nr_segs);
+	end = (pos + write_len - 1) >> PAGE_CACHE_SHIFT;
+	if (mapping_mapped(mapping))
+		unmap_mapping_range(mapping, pos, write_len, 0);
+
+	written = filemap_write_and_wait(mapping);
+	if (written)
+		goto out;
+
+	/*
+	 * After a write we want buffered reads to be sure to go to disk to get
+	 * the new data.  We invalidate clean cached page from the region we're
+	 * about to write.  We do this *before* the write so that we can return
+	 * -EIO without clobbering -EIOCBQUEUED from ->direct_IO().
+	 */
+	if (mapping->nrpages) {
+		written = invalidate_inode_pages2_range(mapping,
+					pos >> PAGE_CACHE_SHIFT, end);
+		if (written)
+			goto out;
+	}
+
+	written = mapping->a_ops->direct_IO(WRITE, iocb, iov, pos, *nr_segs);
+
+	/*
+	 * Finally, try again to invalidate clean pages which might have been
+	 * cached by non-direct readahead, or faulted in by get_user_pages()
+	 * if the source of the write was an mmap'ed region of the file
+	 * we're writing.  Either one is a pretty crazy thing to do,
+	 * so we don't support it 100%.  If this invalidation
+	 * fails, tough, the write still worked...
+	 */
+	if (mapping->nrpages) {
+		invalidate_inode_pages2_range(mapping,
+					      pos >> PAGE_CACHE_SHIFT, end);
+	}
+
 	if (written > 0) {
 		loff_t end = pos + written;
 		if (end > i_size_read(inode) && !S_ISBLK(inode->i_mode)) {
@@ -2024,6 +2068,7 @@ generic_file_direct_write(struct kiocb *iocb, const struct iovec *iov,
 	 * i_mutex is held, which protects generic_osync_inode() from
 	 * livelocking.  AIO O_DIRECT ops attempt to sync metadata here.
 	 */
+out:
 	if ((written >= 0 || written == -EIOCBQUEUED) &&
 	    ((file->f_flags & O_SYNC) || IS_SYNC(inode))) {
 		int err = generic_osync_inode(inode, mapping, OSYNC_METADATA);
@@ -2511,66 +2556,6 @@ ssize_t generic_file_aio_write(struct kiocb *iocb, const struct iovec *iov,
 }
 EXPORT_SYMBOL(generic_file_aio_write);
 
-/*
- * Called under i_mutex for writes to S_ISREG files.   Returns -EIO if something
- * went wrong during pagecache shootdown.
- */
-static ssize_t
-generic_file_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov,
-	loff_t offset, unsigned long nr_segs)
-{
-	struct file *file = iocb->ki_filp;
-	struct address_space *mapping = file->f_mapping;
-	ssize_t retval;
-	size_t write_len;
-	pgoff_t end = 0; /* silence gcc */
-
-	/*
-	 * If it's a write, unmap all mmappings of the file up-front.  This
-	 * will cause any pte dirty bits to be propagated into the pageframes
-	 * for the subsequent filemap_write_and_wait().
-	 */
-	if (rw == WRITE) {
-		write_len = iov_length(iov, nr_segs);
-		end = (offset + write_len - 1) >> PAGE_CACHE_SHIFT;
-	       	if (mapping_mapped(mapping))
-			unmap_mapping_range(mapping, offset, write_len, 0);
-	}
-
-	retval = filemap_write_and_wait(mapping);
-	if (retval)
-		goto out;
-
-	/*
-	 * After a write we want buffered reads to be sure to go to disk to get
-	 * the new data.  We invalidate clean cached page from the region we're
-	 * about to write.  We do this *before* the write so that we can return
-	 * -EIO without clobbering -EIOCBQUEUED from ->direct_IO().
-	 */
-	if (rw == WRITE && mapping->nrpages) {
-		retval = invalidate_inode_pages2_range(mapping,
-					offset >> PAGE_CACHE_SHIFT, end);
-		if (retval)
-			goto out;
-	}
-
-	retval = mapping->a_ops->direct_IO(rw, iocb, iov, offset, nr_segs);
-
-	/*
-	 * Finally, try again to invalidate clean pages which might have been
-	 * cached by non-direct readahead, or faulted in by get_user_pages()
-	 * if the source of the write was an mmap'ed region of the file
-	 * we're writing.  Either one is a pretty crazy thing to do,
-	 * so we don't support it 100%.  If this invalidation
-	 * fails, tough, the write still worked...
-	 */
-	if (rw == WRITE && mapping->nrpages) {
-		invalidate_inode_pages2_range(mapping, offset >> PAGE_CACHE_SHIFT, end);
-	}
-out:
-	return retval;
-}
-
 /**
  * try_to_release_page() - release old fs-specific metadata on a page
  *

From 0d71d10a4252a3938e6b70189bc776171c02e076 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Wed, 23 Jul 2008 21:27:05 -0700
Subject: [PATCH 0221/1171] mm: remove nopfn

There are no users of nopfn in the tree. Remove it.

[hugh@veritas.com: fix build error]
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/mm.h |  9 -------
 mm/memory.c        | 67 +++++-----------------------------------------
 2 files changed, 7 insertions(+), 69 deletions(-)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 2128ef7780c6..eb815cfc1b35 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -166,8 +166,6 @@ struct vm_operations_struct {
 	void (*open)(struct vm_area_struct * area);
 	void (*close)(struct vm_area_struct * area);
 	int (*fault)(struct vm_area_struct *vma, struct vm_fault *vmf);
-	unsigned long (*nopfn)(struct vm_area_struct *area,
-			unsigned long address);
 
 	/* notification that a previously read-only page is about to become
 	 * writable, if an error is returned it will cause a SIGBUS */
@@ -674,13 +672,6 @@ static inline int page_mapped(struct page *page)
 	return atomic_read(&(page)->_mapcount) >= 0;
 }
 
-/*
- * Error return values for the *_nopfn functions
- */
-#define NOPFN_SIGBUS	((unsigned long) -1)
-#define NOPFN_OOM	((unsigned long) -2)
-#define NOPFN_REFAULT	((unsigned long) -3)
-
 /*
  * Different kinds of faults, as returned by handle_mm_fault().
  * Used to decide whether a process gets delivered SIGBUS or
diff --git a/mm/memory.c b/mm/memory.c
index 2302d228fe04..46dbed4b7446 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1058,11 +1058,9 @@ static inline int use_zero_page(struct vm_area_struct *vma)
 	if (vma->vm_flags & (VM_LOCKED | VM_SHARED))
 		return 0;
 	/*
-	 * And if we have a fault or a nopfn routine, it's not an
-	 * anonymous region.
+	 * And if we have a fault routine, it's not an anonymous region.
 	 */
-	return !vma->vm_ops ||
-		(!vma->vm_ops->fault && !vma->vm_ops->nopfn);
+	return !vma->vm_ops || !vma->vm_ops->fault;
 }
 
 int get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
@@ -1338,6 +1336,11 @@ out:
  *
  * This function should only be called from a vm_ops->fault handler, and
  * in that case the handler should return NULL.
+ *
+ * vma cannot be a COW mapping.
+ *
+ * As this is called only for pages that do not currently exist, we
+ * do not need to flush old virtual caches or the TLB.
  */
 int vm_insert_pfn(struct vm_area_struct *vma, unsigned long addr,
 			unsigned long pfn)
@@ -2501,59 +2504,6 @@ static int do_linear_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 	return __do_fault(mm, vma, address, pmd, pgoff, flags, orig_pte);
 }
 
-
-/*
- * do_no_pfn() tries to create a new page mapping for a page without
- * a struct_page backing it
- *
- * As this is called only for pages that do not currently exist, we
- * do not need to flush old virtual caches or the TLB.
- *
- * We enter with non-exclusive mmap_sem (to exclude vma changes,
- * but allow concurrent faults), and pte mapped but not yet locked.
- * We return with mmap_sem still held, but pte unmapped and unlocked.
- *
- * It is expected that the ->nopfn handler always returns the same pfn
- * for a given virtual mapping.
- *
- * Mark this `noinline' to prevent it from bloating the main pagefault code.
- */
-static noinline int do_no_pfn(struct mm_struct *mm, struct vm_area_struct *vma,
-		     unsigned long address, pte_t *page_table, pmd_t *pmd,
-		     int write_access)
-{
-	spinlock_t *ptl;
-	pte_t entry;
-	unsigned long pfn;
-
-	pte_unmap(page_table);
-	BUG_ON(!(vma->vm_flags & (VM_PFNMAP|VM_MIXEDMAP)));
-	BUG_ON((vma->vm_flags & VM_PFNMAP) && is_cow_mapping(vma->vm_flags));
-
-	pfn = vma->vm_ops->nopfn(vma, address & PAGE_MASK);
-
-	BUG_ON((vma->vm_flags & VM_MIXEDMAP) && pfn_valid(pfn));
-
-	if (unlikely(pfn == NOPFN_OOM))
-		return VM_FAULT_OOM;
-	else if (unlikely(pfn == NOPFN_SIGBUS))
-		return VM_FAULT_SIGBUS;
-	else if (unlikely(pfn == NOPFN_REFAULT))
-		return 0;
-
-	page_table = pte_offset_map_lock(mm, pmd, address, &ptl);
-
-	/* Only go through if we didn't race with anybody else... */
-	if (pte_none(*page_table)) {
-		entry = pfn_pte(pfn, vma->vm_page_prot);
-		if (write_access)
-			entry = maybe_mkwrite(pte_mkdirty(entry), vma);
-		set_pte_at(mm, address, page_table, entry);
-	}
-	pte_unmap_unlock(page_table, ptl);
-	return 0;
-}
-
 /*
  * Fault of a previously existing named mapping. Repopulate the pte
  * from the encoded file_pte if possible. This enables swappable
@@ -2614,9 +2564,6 @@ static inline int handle_pte_fault(struct mm_struct *mm,
 				if (likely(vma->vm_ops->fault))
 					return do_linear_fault(mm, vma, address,
 						pte, pmd, write_access, entry);
-				if (unlikely(vma->vm_ops->nopfn))
-					return do_no_pfn(mm, vma, address, pte,
-							 pmd, write_access);
 			}
 			return do_anonymous_page(mm, vma, address,
 						 pte, pmd, write_access);

From 28b2ee20c7cba812b6f2ccf6d722cf86d00a84dc Mon Sep 17 00:00:00 2001
From: Rik van Riel <riel@redhat.com>
Date: Wed, 23 Jul 2008 21:27:05 -0700
Subject: [PATCH 0222/1171] access_process_vm device memory infrastructure

In order to be able to debug things like the X server and programs using
the PPC Cell SPUs, the debugger needs to be able to access device memory
through ptrace and /proc/pid/mem.

This patch:

Add the generic_access_phys access function and put the hooks in place
to allow access_process_vm to access device or PPC Cell SPU memory.

[riel@redhat.com: Add documentation for the vm_ops->access function]
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Benjamin Herrensmidt <benh@kernel.crashing.org>
Cc: Dave Airlie <airlied@linux.ie>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/filesystems/Locking |   7 ++
 arch/Kconfig                      |   3 +
 arch/x86/Kconfig                  |   1 +
 arch/x86/mm/ioremap.c             |   8 ++
 include/asm-x86/io_32.h           |   2 +
 include/asm-x86/io_64.h           |   2 +
 include/linux/mm.h                |   8 ++
 mm/memory.c                       | 131 ++++++++++++++++++++++++++----
 8 files changed, 144 insertions(+), 18 deletions(-)

diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking
index 8b22d7d8b991..680fb566b928 100644
--- a/Documentation/filesystems/Locking
+++ b/Documentation/filesystems/Locking
@@ -510,6 +510,7 @@ prototypes:
 	void (*close)(struct vm_area_struct*);
 	int (*fault)(struct vm_area_struct*, struct vm_fault *);
 	int (*page_mkwrite)(struct vm_area_struct *, struct page *);
+	int (*access)(struct vm_area_struct *, unsigned long, void*, int, int);
 
 locking rules:
 		BKL	mmap_sem	PageLocked(page)
@@ -517,6 +518,7 @@ open:		no	yes
 close:		no	yes
 fault:		no	yes
 page_mkwrite:	no	yes		no
+access:		no	yes
 
 	->page_mkwrite() is called when a previously read-only page is
 about to become writeable. The file system is responsible for
@@ -525,6 +527,11 @@ taking to lock out truncate, the page range should be verified to be
 within i_size. The page mapping should also be checked that it is not
 NULL.
 
+	->access() is called when get_user_pages() fails in
+acces_process_vm(), typically used to debug a process through
+/proc/pid/mem or ptrace.  This function is needed only for
+VM_IO | VM_PFNMAP VMAs.
+
 ================================================================================
 			Dubious stuff
 
diff --git a/arch/Kconfig b/arch/Kconfig
index 4d5ebbc1e72b..6093c0be58b0 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -31,6 +31,9 @@ config KRETPROBES
 	def_bool y
 	depends on KPROBES && HAVE_KRETPROBES
 
+config HAVE_IOREMAP_PROT
+	def_bool n
+
 config HAVE_KPROBES
 	def_bool n
 
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 03980cb04291..b2ddfcf01728 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -21,6 +21,7 @@ config X86
 	select HAVE_UNSTABLE_SCHED_CLOCK
 	select HAVE_IDE
 	select HAVE_OPROFILE
+	select HAVE_IOREMAP_PROT
 	select HAVE_KPROBES
 	select HAVE_KRETPROBES
 	select HAVE_DYNAMIC_FTRACE
diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c
index 24c1d3c30186..016f335bbeea 100644
--- a/arch/x86/mm/ioremap.c
+++ b/arch/x86/mm/ioremap.c
@@ -330,6 +330,14 @@ static void __iomem *ioremap_default(resource_size_t phys_addr,
 	return (void __iomem *)ret;
 }
 
+void __iomem *ioremap_prot(resource_size_t phys_addr, unsigned long size,
+				unsigned long prot_val)
+{
+	return __ioremap_caller(phys_addr, size, (prot_val & _PAGE_CACHE_MASK),
+				__builtin_return_address(0));
+}
+EXPORT_SYMBOL(ioremap_prot);
+
 /**
  * iounmap - Free a IO remapping
  * @addr: virtual address from ioremap_*
diff --git a/include/asm-x86/io_32.h b/include/asm-x86/io_32.h
index 4df44ed54077..e876d89ac156 100644
--- a/include/asm-x86/io_32.h
+++ b/include/asm-x86/io_32.h
@@ -110,6 +110,8 @@ static inline void *phys_to_virt(unsigned long address)
  */
 extern void __iomem *ioremap_nocache(resource_size_t offset, unsigned long size);
 extern void __iomem *ioremap_cache(resource_size_t offset, unsigned long size);
+extern void __iomem *ioremap_prot(resource_size_t offset, unsigned long size,
+				unsigned long prot_val);
 
 /*
  * The default ioremap() behavior is non-cached:
diff --git a/include/asm-x86/io_64.h b/include/asm-x86/io_64.h
index ddd8058a5026..22995c5c5adc 100644
--- a/include/asm-x86/io_64.h
+++ b/include/asm-x86/io_64.h
@@ -175,6 +175,8 @@ extern void early_iounmap(void *addr, unsigned long size);
  */
 extern void __iomem *ioremap_nocache(resource_size_t offset, unsigned long size);
 extern void __iomem *ioremap_cache(resource_size_t offset, unsigned long size);
+extern void __iomem *ioremap_prot(resource_size_t offset, unsigned long size,
+				unsigned long prot_val);
 
 /*
  * The default ioremap() behavior is non-cached:
diff --git a/include/linux/mm.h b/include/linux/mm.h
index eb815cfc1b35..5c7f8f64f70e 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -170,6 +170,12 @@ struct vm_operations_struct {
 	/* notification that a previously read-only page is about to become
 	 * writable, if an error is returned it will cause a SIGBUS */
 	int (*page_mkwrite)(struct vm_area_struct *vma, struct page *page);
+
+	/* called by access_process_vm when get_user_pages() fails, typically
+	 * for use by special VMAs that can switch between memory and hardware
+	 */
+	int (*access)(struct vm_area_struct *vma, unsigned long addr,
+		      void *buf, int len, int write);
 #ifdef CONFIG_NUMA
 	/*
 	 * set_policy() op must add a reference to any non-NULL @new mempolicy
@@ -771,6 +777,8 @@ int copy_page_range(struct mm_struct *dst, struct mm_struct *src,
 			struct vm_area_struct *vma);
 void unmap_mapping_range(struct address_space *mapping,
 		loff_t const holebegin, loff_t const holelen, int even_cows);
+int generic_access_phys(struct vm_area_struct *vma, unsigned long addr,
+			void *buf, int len, int write);
 
 static inline void unmap_shared_mapping_range(struct address_space *mapping,
 		loff_t const holebegin, loff_t const holelen)
diff --git a/mm/memory.c b/mm/memory.c
index 46dbed4b7446..87350321e66f 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -2751,6 +2751,86 @@ int in_gate_area_no_task(unsigned long addr)
 
 #endif	/* __HAVE_ARCH_GATE_AREA */
 
+#ifdef CONFIG_HAVE_IOREMAP_PROT
+static resource_size_t follow_phys(struct vm_area_struct *vma,
+			unsigned long address, unsigned int flags,
+			unsigned long *prot)
+{
+	pgd_t *pgd;
+	pud_t *pud;
+	pmd_t *pmd;
+	pte_t *ptep, pte;
+	spinlock_t *ptl;
+	resource_size_t phys_addr = 0;
+	struct mm_struct *mm = vma->vm_mm;
+
+	VM_BUG_ON(!(vma->vm_flags & (VM_IO | VM_PFNMAP)));
+
+	pgd = pgd_offset(mm, address);
+	if (pgd_none(*pgd) || unlikely(pgd_bad(*pgd)))
+		goto no_page_table;
+
+	pud = pud_offset(pgd, address);
+	if (pud_none(*pud) || unlikely(pud_bad(*pud)))
+		goto no_page_table;
+
+	pmd = pmd_offset(pud, address);
+	if (pmd_none(*pmd) || unlikely(pmd_bad(*pmd)))
+		goto no_page_table;
+
+	/* We cannot handle huge page PFN maps. Luckily they don't exist. */
+	if (pmd_huge(*pmd))
+		goto no_page_table;
+
+	ptep = pte_offset_map_lock(mm, pmd, address, &ptl);
+	if (!ptep)
+		goto out;
+
+	pte = *ptep;
+	if (!pte_present(pte))
+		goto unlock;
+	if ((flags & FOLL_WRITE) && !pte_write(pte))
+		goto unlock;
+	phys_addr = pte_pfn(pte);
+	phys_addr <<= PAGE_SHIFT; /* Shift here to avoid overflow on PAE */
+
+	*prot = pgprot_val(pte_pgprot(pte));
+
+unlock:
+	pte_unmap_unlock(ptep, ptl);
+out:
+	return phys_addr;
+no_page_table:
+	return 0;
+}
+
+int generic_access_phys(struct vm_area_struct *vma, unsigned long addr,
+			void *buf, int len, int write)
+{
+	resource_size_t phys_addr;
+	unsigned long prot = 0;
+	void *maddr;
+	int offset = addr & (PAGE_SIZE-1);
+
+	if (!(vma->vm_flags & (VM_IO | VM_PFNMAP)))
+		return -EINVAL;
+
+	phys_addr = follow_phys(vma, addr, write, &prot);
+
+	if (!phys_addr)
+		return -EINVAL;
+
+	maddr = ioremap_prot(phys_addr, PAGE_SIZE, prot);
+	if (write)
+		memcpy_toio(maddr + offset, buf, len);
+	else
+		memcpy_fromio(buf, maddr + offset, len);
+	iounmap(maddr);
+
+	return len;
+}
+#endif
+
 /*
  * Access another process' address space.
  * Source/target buffer must be kernel space,
@@ -2760,7 +2840,6 @@ int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, in
 {
 	struct mm_struct *mm;
 	struct vm_area_struct *vma;
-	struct page *page;
 	void *old_buf = buf;
 
 	mm = get_task_mm(tsk);
@@ -2772,28 +2851,44 @@ int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, in
 	while (len) {
 		int bytes, ret, offset;
 		void *maddr;
+		struct page *page = NULL;
 
 		ret = get_user_pages(tsk, mm, addr, 1,
 				write, 1, &page, &vma);
-		if (ret <= 0)
-			break;
-
-		bytes = len;
-		offset = addr & (PAGE_SIZE-1);
-		if (bytes > PAGE_SIZE-offset)
-			bytes = PAGE_SIZE-offset;
-
-		maddr = kmap(page);
-		if (write) {
-			copy_to_user_page(vma, page, addr,
-					  maddr + offset, buf, bytes);
-			set_page_dirty_lock(page);
+		if (ret <= 0) {
+			/*
+			 * Check if this is a VM_IO | VM_PFNMAP VMA, which
+			 * we can access using slightly different code.
+			 */
+#ifdef CONFIG_HAVE_IOREMAP_PROT
+			vma = find_vma(mm, addr);
+			if (!vma)
+				break;
+			if (vma->vm_ops && vma->vm_ops->access)
+				ret = vma->vm_ops->access(vma, addr, buf,
+							  len, write);
+			if (ret <= 0)
+#endif
+				break;
+			bytes = ret;
 		} else {
-			copy_from_user_page(vma, page, addr,
-					    buf, maddr + offset, bytes);
+			bytes = len;
+			offset = addr & (PAGE_SIZE-1);
+			if (bytes > PAGE_SIZE-offset)
+				bytes = PAGE_SIZE-offset;
+
+			maddr = kmap(page);
+			if (write) {
+				copy_to_user_page(vma, page, addr,
+						  maddr + offset, buf, bytes);
+				set_page_dirty_lock(page);
+			} else {
+				copy_from_user_page(vma, page, addr,
+						    buf, maddr + offset, bytes);
+			}
+			kunmap(page);
+			page_cache_release(page);
 		}
-		kunmap(page);
-		page_cache_release(page);
 		len -= bytes;
 		buf += bytes;
 		addr += bytes;

From 7ae8ed5053a39082d224a3f48409e016baca9c16 Mon Sep 17 00:00:00 2001
From: Rik van Riel <riel@redhat.com>
Date: Wed, 23 Jul 2008 21:27:07 -0700
Subject: [PATCH 0223/1171] use generic_access_phys for /dev/mem mappings

Use generic_access_phys as the access_process_vm access function for
/dev/mem mappings.  This makes it possible to debug the X server.

[akpm@linux-foundation.org: repair all the architectures which broke]
Signed-off-by: Rik van Riel <riel@redhat.com>
Cc: Benjamin Herrensmidt <benh@kernel.crashing.org>
Cc: Dave Airlie <airlied@linux.ie>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/pci/i386.c | 1 +
 drivers/char/mem.c  | 5 ++++-
 2 files changed, 5 insertions(+), 1 deletion(-)

diff --git a/arch/x86/pci/i386.c b/arch/x86/pci/i386.c
index 2aafb67dc5f1..a09505806b82 100644
--- a/arch/x86/pci/i386.c
+++ b/arch/x86/pci/i386.c
@@ -280,6 +280,7 @@ static void pci_track_mmap_page_range(struct vm_area_struct *vma)
 static struct vm_operations_struct pci_mmap_ops = {
 	.open  = pci_track_mmap_page_range,
 	.close = pci_unmap_page_range,
+	.access = generic_access_phys,
 };
 
 int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma,
diff --git a/drivers/char/mem.c b/drivers/char/mem.c
index c2dba82eb5f7..672b08e694d0 100644
--- a/drivers/char/mem.c
+++ b/drivers/char/mem.c
@@ -327,7 +327,10 @@ static void mmap_mem_close(struct vm_area_struct *vma)
 
 static struct vm_operations_struct mmap_mem_ops = {
 	.open  = mmap_mem_open,
-	.close = mmap_mem_close
+	.close = mmap_mem_close,
+#ifdef CONFIG_HAVE_IOREMAP_PROT
+	.access = generic_access_phys
+#endif
 };
 
 static int mmap_mem(struct file * file, struct vm_area_struct * vma)

From a1f242ff460e4b50a045fa237c3c56cce9eabf83 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Wed, 23 Jul 2008 21:27:08 -0700
Subject: [PATCH 0224/1171] powerpc ioremap_prot

This adds ioremap_prot and pte_pgprot() so that one can extract protection
bits from a PTE and use them to ioremap_prot() (in order to support ptrace
of VM_IO | VM_PFNMAP as per Rik's patch).

This moves a couple of flag checks around in the ioremap implementations
of arch/powerpc.  There's a side effect of allowing non-cacheable and
non-guarded mappings on ppc32 which before would always have _PAGE_GUARDED
set whenever _PAGE_NO_CACHE is.

(standard ioremap will still set _PAGE_GUARDED, but ioremap_prot will be
capable of setting such a non guarded mapping).

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Rik van Riel <riel@redhat.com>
Cc: Dave Airlie <airlied@linux.ie>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/powerpc/Kconfig                |  1 +
 arch/powerpc/mm/pgtable_32.c        | 22 ++++++++++++++++------
 arch/powerpc/mm/pgtable_64.c        | 16 ++++++++++++++++
 include/asm-powerpc/io.h            |  5 ++++-
 include/asm-powerpc/pgtable-4k.h    |  3 +++
 include/asm-powerpc/pgtable-ppc32.h | 16 ++++++++++++++++
 include/asm-powerpc/pgtable-ppc64.h |  8 ++++++++
 7 files changed, 64 insertions(+), 7 deletions(-)

diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index 4d7e2ba10bae..a487671c282f 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -111,6 +111,7 @@ config PPC
 	select HAVE_DYNAMIC_FTRACE
 	select HAVE_FTRACE
 	select HAVE_IDE
+	select HAVE_IOREMAP_PROT
 	select HAVE_KPROBES
 	select HAVE_ARCH_KGDB
 	select HAVE_KRETPROBES
diff --git a/arch/powerpc/mm/pgtable_32.c b/arch/powerpc/mm/pgtable_32.c
index c7584072dfcc..2001abdb1912 100644
--- a/arch/powerpc/mm/pgtable_32.c
+++ b/arch/powerpc/mm/pgtable_32.c
@@ -145,13 +145,20 @@ void pte_free(struct mm_struct *mm, pgtable_t ptepage)
 void __iomem *
 ioremap(phys_addr_t addr, unsigned long size)
 {
-	return __ioremap(addr, size, _PAGE_NO_CACHE);
+	return __ioremap(addr, size, _PAGE_NO_CACHE | _PAGE_GUARDED);
 }
 EXPORT_SYMBOL(ioremap);
 
 void __iomem *
 ioremap_flags(phys_addr_t addr, unsigned long size, unsigned long flags)
 {
+	/* writeable implies dirty for kernel addresses */
+	if (flags & _PAGE_RW)
+		flags |= _PAGE_DIRTY | _PAGE_HWWRITE;
+
+	/* we don't want to let _PAGE_USER and _PAGE_EXEC leak out */
+	flags &= ~(_PAGE_USER | _PAGE_EXEC | _PAGE_HWEXEC);
+
 	return __ioremap(addr, size, flags);
 }
 EXPORT_SYMBOL(ioremap_flags);
@@ -163,6 +170,14 @@ __ioremap(phys_addr_t addr, unsigned long size, unsigned long flags)
 	phys_addr_t p;
 	int err;
 
+	/* Make sure we have the base flags */
+	if ((flags & _PAGE_PRESENT) == 0)
+		flags |= _PAGE_KERNEL;
+
+	/* Non-cacheable page cannot be coherent */
+	if (flags & _PAGE_NO_CACHE)
+		flags &= ~_PAGE_COHERENT;
+
 	/*
 	 * Choose an address to map it to.
 	 * Once the vmalloc system is running, we use it.
@@ -219,11 +234,6 @@ __ioremap(phys_addr_t addr, unsigned long size, unsigned long flags)
 		v = (ioremap_bot -= size);
 	}
 
-	if ((flags & _PAGE_PRESENT) == 0)
-		flags |= _PAGE_KERNEL;
-	if (flags & _PAGE_NO_CACHE)
-		flags |= _PAGE_GUARDED;
-
 	/*
 	 * Should check if it is a candidate for a BAT mapping
 	 */
diff --git a/arch/powerpc/mm/pgtable_64.c b/arch/powerpc/mm/pgtable_64.c
index 3ef0ad2f9ca0..365e61ae5dbc 100644
--- a/arch/powerpc/mm/pgtable_64.c
+++ b/arch/powerpc/mm/pgtable_64.c
@@ -107,9 +107,18 @@ void __iomem * __ioremap_at(phys_addr_t pa, void *ea, unsigned long size,
 {
 	unsigned long i;
 
+	/* Make sure we have the base flags */
 	if ((flags & _PAGE_PRESENT) == 0)
 		flags |= pgprot_val(PAGE_KERNEL);
 
+	/* Non-cacheable page cannot be coherent */
+	if (flags & _PAGE_NO_CACHE)
+		flags &= ~_PAGE_COHERENT;
+
+	/* We don't support the 4K PFN hack with ioremap */
+	if (flags & _PAGE_4K_PFN)
+		return NULL;
+
 	WARN_ON(pa & ~PAGE_MASK);
 	WARN_ON(((unsigned long)ea) & ~PAGE_MASK);
 	WARN_ON(size & ~PAGE_MASK);
@@ -190,6 +199,13 @@ void __iomem * ioremap(phys_addr_t addr, unsigned long size)
 void __iomem * ioremap_flags(phys_addr_t addr, unsigned long size,
 			     unsigned long flags)
 {
+	/* writeable implies dirty for kernel addresses */
+	if (flags & _PAGE_RW)
+		flags |= _PAGE_DIRTY;
+
+	/* we don't want to let _PAGE_USER and _PAGE_EXEC leak out */
+	flags &= ~(_PAGE_USER | _PAGE_EXEC);
+
 	if (ppc_md.ioremap)
 		return ppc_md.ioremap(addr, size, flags);
 	return __ioremap(addr, size, flags);
diff --git a/include/asm-powerpc/io.h b/include/asm-powerpc/io.h
index 8b627823f5f9..77c7fa025e65 100644
--- a/include/asm-powerpc/io.h
+++ b/include/asm-powerpc/io.h
@@ -617,7 +617,8 @@ static inline void iosync(void)
  *   and can be hooked by the platform via ppc_md
  *
  * * ioremap_flags allows to specify the page flags as an argument and can
- *   also be hooked by the platform via ppc_md
+ *   also be hooked by the platform via ppc_md. ioremap_prot is the exact
+ *   same thing as ioremap_flags.
  *
  * * ioremap_nocache is identical to ioremap
  *
@@ -639,6 +640,8 @@ extern void __iomem *ioremap(phys_addr_t address, unsigned long size);
 extern void __iomem *ioremap_flags(phys_addr_t address, unsigned long size,
 				   unsigned long flags);
 #define ioremap_nocache(addr, size)	ioremap((addr), (size))
+#define ioremap_prot(addr, size, prot)	ioremap_flags((addr), (size), (prot))
+
 extern void iounmap(volatile void __iomem *addr);
 
 extern void __iomem *__ioremap(phys_addr_t, unsigned long size,
diff --git a/include/asm-powerpc/pgtable-4k.h b/include/asm-powerpc/pgtable-4k.h
index fd2090dc1dce..c9601dfb4a1e 100644
--- a/include/asm-powerpc/pgtable-4k.h
+++ b/include/asm-powerpc/pgtable-4k.h
@@ -51,6 +51,9 @@
 #define _PAGE_HPTEFLAGS (_PAGE_BUSY | _PAGE_HASHPTE | \
 			 _PAGE_SECONDARY | _PAGE_GROUP_IX)
 
+/* There is no 4K PFN hack on 4K pages */
+#define _PAGE_4K_PFN	0
+
 /* PAGE_MASK gives the right answer below, but only by accident */
 /* It should be preserving the high 48 bits and then specifically */
 /* preserving _PAGE_SECONDARY | _PAGE_GROUP_IX */
diff --git a/include/asm-powerpc/pgtable-ppc32.h b/include/asm-powerpc/pgtable-ppc32.h
index 3a96d001cb75..bdbab72f3ebc 100644
--- a/include/asm-powerpc/pgtable-ppc32.h
+++ b/include/asm-powerpc/pgtable-ppc32.h
@@ -395,6 +395,12 @@ extern int icache_44x_need_flush;
 #ifndef _PAGE_EXEC
 #define _PAGE_EXEC	0
 #endif
+#ifndef _PAGE_ENDIAN
+#define _PAGE_ENDIAN	0
+#endif
+#ifndef _PAGE_COHERENT
+#define _PAGE_COHERENT	0
+#endif
 #ifndef _PMD_PRESENT_MASK
 #define _PMD_PRESENT_MASK	_PMD_PRESENT
 #endif
@@ -405,6 +411,12 @@ extern int icache_44x_need_flush;
 
 #define _PAGE_CHG_MASK	(PAGE_MASK | _PAGE_ACCESSED | _PAGE_DIRTY)
 
+
+#define PAGE_PROT_BITS	__pgprot(_PAGE_GUARDED | _PAGE_COHERENT | _PAGE_NO_CACHE | \
+				 _PAGE_WRITETHRU | _PAGE_ENDIAN | \
+				 _PAGE_USER | _PAGE_ACCESSED | \
+				 _PAGE_RW | _PAGE_HWWRITE | _PAGE_DIRTY | \
+				 _PAGE_EXEC | _PAGE_HWEXEC)
 /*
  * Note: the _PAGE_COHERENT bit automatically gets set in the hardware
  * PTE if CONFIG_SMP is defined (hash_page does this); there is no need
@@ -538,6 +550,10 @@ static inline pte_t pte_mkyoung(pte_t pte) {
 	pte_val(pte) |= _PAGE_ACCESSED; return pte; }
 static inline pte_t pte_mkspecial(pte_t pte) {
 	return pte; }
+static inline unsigned long pte_pgprot(pte_t pte)
+{
+	return __pgprot(pte_val(pte)) & PAGE_PROT_BITS;
+}
 
 static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
 {
diff --git a/include/asm-powerpc/pgtable-ppc64.h b/include/asm-powerpc/pgtable-ppc64.h
index ab98a9c80b28..ba8000352b9a 100644
--- a/include/asm-powerpc/pgtable-ppc64.h
+++ b/include/asm-powerpc/pgtable-ppc64.h
@@ -117,6 +117,10 @@
 #define PAGE_AGP	__pgprot(_PAGE_BASE | _PAGE_WRENABLE | _PAGE_NO_CACHE)
 #define HAVE_PAGE_AGP
 
+#define PAGE_PROT_BITS	__pgprot(_PAGE_GUARDED | _PAGE_COHERENT | \
+				 _PAGE_NO_CACHE | _PAGE_WRITETHRU | \
+				 _PAGE_4K_PFN | _PAGE_RW | _PAGE_USER | \
+ 				 _PAGE_ACCESSED | _PAGE_DIRTY | _PAGE_EXEC)
 /* PTEIDX nibble */
 #define _PTEIDX_SECONDARY	0x8
 #define _PTEIDX_GROUP_IX	0x7
@@ -262,6 +266,10 @@ static inline pte_t pte_mkhuge(pte_t pte) {
 	return pte; }
 static inline pte_t pte_mkspecial(pte_t pte) {
 	return pte; }
+static inline unsigned long pte_pgprot(pte_t pte)
+{
+	return __pgprot(pte_val(pte)) & PAGE_PROT_BITS;
+}
 
 /* Atomic PTE updates */
 static inline unsigned long pte_update(struct mm_struct *mm,

From a352894d07059649398c4769dc8b645e1a1dad88 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Wed, 23 Jul 2008 21:27:09 -0700
Subject: [PATCH 0225/1171] spufs: use new vm_ops->access to allow local state
 access from gdb

This uses the new vm_ops->access to allow gdb to access the SPU local
store.  We currently prevent access to problem state registers, this can
be done later if really needed but it's safer not to.

[akpm@linux-foundation.org: fix typo]
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Rik van Riel <riel@redhat.com>
Cc: Dave Airlie <airlied@linux.ie>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/powerpc/platforms/cell/spufs/file.c | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/arch/powerpc/platforms/cell/spufs/file.c b/arch/powerpc/platforms/cell/spufs/file.c
index 99c73066b82f..010a51f59796 100644
--- a/arch/powerpc/platforms/cell/spufs/file.c
+++ b/arch/powerpc/platforms/cell/spufs/file.c
@@ -288,9 +288,32 @@ spufs_mem_mmap_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 	return VM_FAULT_NOPAGE;
 }
 
+static int spufs_mem_mmap_access(struct vm_area_struct *vma,
+				unsigned long address,
+				void *buf, int len, int write)
+{
+	struct spu_context *ctx = vma->vm_file->private_data;
+	unsigned long offset = address - vma->vm_start;
+	char *local_store;
+
+	if (write && !(vma->vm_flags & VM_WRITE))
+		return -EACCES;
+	if (spu_acquire(ctx))
+		return -EINTR;
+	if ((offset + len) > vma->vm_end)
+		len = vma->vm_end - offset;
+	local_store = ctx->ops->get_ls(ctx);
+	if (write)
+		memcpy_toio(local_store + offset, buf, len);
+	else
+		memcpy_fromio(buf, local_store + offset, len);
+	spu_release(ctx);
+	return len;
+}
 
 static struct vm_operations_struct spufs_mem_mmap_vmops = {
 	.fault = spufs_mem_mmap_fault,
+	.access = spufs_mem_mmap_access,
 };
 
 static int spufs_mem_mmap(struct file *file, struct vm_area_struct *vma)

From 42b7772812d15b86543a23b82bd6070eef9a08b1 Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Wed, 23 Jul 2008 21:27:10 -0700
Subject: [PATCH 0226/1171] mm: remove double indirection on tlb parameter to
 free_pgd_range() & Co

The double indirection here is not needed anywhere and hence (at least)
confusing.

Signed-off-by: Jan Beulich <jbeulich@novell.com>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Nick Piggin <npiggin@suse.de>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: "Luck, Tony" <tony.luck@intel.com>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: "David S. Miller" <davem@davemloft.net>
Acked-by: Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/ia64/mm/hugetlbpage.c    |  2 +-
 arch/powerpc/mm/hugetlbpage.c |  8 ++++----
 fs/exec.c                     |  4 ++--
 include/asm-ia64/hugetlb.h    |  2 +-
 include/asm-powerpc/hugetlb.h |  2 +-
 include/asm-sh/hugetlb.h      |  2 +-
 include/asm-sparc/hugetlb.h   |  2 +-
 include/asm-x86/hugetlb.h     |  2 +-
 include/linux/mm.h            |  4 +---
 mm/internal.h                 |  3 +++
 mm/memory.c                   | 10 ++++++----
 mm/mmap.c                     |  6 ++++--
 12 files changed, 26 insertions(+), 21 deletions(-)

diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c
index d3ce8f3bcaa6..cd49e2860eef 100644
--- a/arch/ia64/mm/hugetlbpage.c
+++ b/arch/ia64/mm/hugetlbpage.c
@@ -112,7 +112,7 @@ follow_huge_pmd(struct mm_struct *mm, unsigned long address, pmd_t *pmd, int wri
 	return NULL;
 }
 
-void hugetlb_free_pgd_range(struct mmu_gather **tlb,
+void hugetlb_free_pgd_range(struct mmu_gather *tlb,
 			unsigned long addr, unsigned long end,
 			unsigned long floor, unsigned long ceiling)
 {
diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c
index 0d12fba31bc5..1a96cc891cf5 100644
--- a/arch/powerpc/mm/hugetlbpage.c
+++ b/arch/powerpc/mm/hugetlbpage.c
@@ -255,7 +255,7 @@ static void hugetlb_free_pud_range(struct mmu_gather *tlb, pgd_t *pgd,
  *
  * Must be called with pagetable lock held.
  */
-void hugetlb_free_pgd_range(struct mmu_gather **tlb,
+void hugetlb_free_pgd_range(struct mmu_gather *tlb,
 			    unsigned long addr, unsigned long end,
 			    unsigned long floor, unsigned long ceiling)
 {
@@ -315,13 +315,13 @@ void hugetlb_free_pgd_range(struct mmu_gather **tlb,
 		return;
 
 	start = addr;
-	pgd = pgd_offset((*tlb)->mm, addr);
+	pgd = pgd_offset(tlb->mm, addr);
 	do {
-		BUG_ON(get_slice_psize((*tlb)->mm, addr) != mmu_huge_psize);
+		BUG_ON(get_slice_psize(tlb->mm, addr) != mmu_huge_psize);
 		next = pgd_addr_end(addr, end);
 		if (pgd_none_or_clear_bad(pgd))
 			continue;
-		hugetlb_free_pud_range(*tlb, pgd, addr, next, floor, ceiling);
+		hugetlb_free_pud_range(tlb, pgd, addr, next, floor, ceiling);
 	} while (pgd++, addr = next, addr != end);
 }
 
diff --git a/fs/exec.c b/fs/exec.c
index fd9234379e8d..190ed1f92774 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -541,7 +541,7 @@ static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift)
 		/*
 		 * when the old and new regions overlap clear from new_end.
 		 */
-		free_pgd_range(&tlb, new_end, old_end, new_end,
+		free_pgd_range(tlb, new_end, old_end, new_end,
 			vma->vm_next ? vma->vm_next->vm_start : 0);
 	} else {
 		/*
@@ -550,7 +550,7 @@ static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift)
 		 * have constraints on va-space that make this illegal (IA64) -
 		 * for the others its just a little faster.
 		 */
-		free_pgd_range(&tlb, old_start, old_end, new_end,
+		free_pgd_range(tlb, old_start, old_end, new_end,
 			vma->vm_next ? vma->vm_next->vm_start : 0);
 	}
 	tlb_finish_mmu(tlb, new_end, old_end);
diff --git a/include/asm-ia64/hugetlb.h b/include/asm-ia64/hugetlb.h
index f28a9701f1cf..e9d1e5e2382d 100644
--- a/include/asm-ia64/hugetlb.h
+++ b/include/asm-ia64/hugetlb.h
@@ -4,7 +4,7 @@
 #include <asm/page.h>
 
 
-void hugetlb_free_pgd_range(struct mmu_gather **tlb, unsigned long addr,
+void hugetlb_free_pgd_range(struct mmu_gather *tlb, unsigned long addr,
 			    unsigned long end, unsigned long floor,
 			    unsigned long ceiling);
 
diff --git a/include/asm-powerpc/hugetlb.h b/include/asm-powerpc/hugetlb.h
index be32ff02f4a0..0a37aa5ecaa5 100644
--- a/include/asm-powerpc/hugetlb.h
+++ b/include/asm-powerpc/hugetlb.h
@@ -7,7 +7,7 @@
 int is_hugepage_only_range(struct mm_struct *mm, unsigned long addr,
 			   unsigned long len);
 
-void hugetlb_free_pgd_range(struct mmu_gather **tlb, unsigned long addr,
+void hugetlb_free_pgd_range(struct mmu_gather *tlb, unsigned long addr,
 			    unsigned long end, unsigned long floor,
 			    unsigned long ceiling);
 
diff --git a/include/asm-sh/hugetlb.h b/include/asm-sh/hugetlb.h
index 02402303d89b..fb30018938c7 100644
--- a/include/asm-sh/hugetlb.h
+++ b/include/asm-sh/hugetlb.h
@@ -26,7 +26,7 @@ static inline int prepare_hugepage_range(unsigned long addr, unsigned long len)
 static inline void hugetlb_prefault_arch_hook(struct mm_struct *mm) {
 }
 
-static inline void hugetlb_free_pgd_range(struct mmu_gather **tlb,
+static inline void hugetlb_free_pgd_range(struct mmu_gather *tlb,
 					  unsigned long addr, unsigned long end,
 					  unsigned long floor,
 					  unsigned long ceiling)
diff --git a/include/asm-sparc/hugetlb.h b/include/asm-sparc/hugetlb.h
index 412af58926a0..aeb92374ca3d 100644
--- a/include/asm-sparc/hugetlb.h
+++ b/include/asm-sparc/hugetlb.h
@@ -31,7 +31,7 @@ static inline int prepare_hugepage_range(unsigned long addr, unsigned long len)
 	return 0;
 }
 
-static inline void hugetlb_free_pgd_range(struct mmu_gather **tlb,
+static inline void hugetlb_free_pgd_range(struct mmu_gather *tlb,
 					  unsigned long addr, unsigned long end,
 					  unsigned long floor,
 					  unsigned long ceiling)
diff --git a/include/asm-x86/hugetlb.h b/include/asm-x86/hugetlb.h
index 14171a4924f6..7eed6e0883bf 100644
--- a/include/asm-x86/hugetlb.h
+++ b/include/asm-x86/hugetlb.h
@@ -26,7 +26,7 @@ static inline int prepare_hugepage_range(unsigned long addr, unsigned long len)
 static inline void hugetlb_prefault_arch_hook(struct mm_struct *mm) {
 }
 
-static inline void hugetlb_free_pgd_range(struct mmu_gather **tlb,
+static inline void hugetlb_free_pgd_range(struct mmu_gather *tlb,
 					  unsigned long addr, unsigned long end,
 					  unsigned long floor,
 					  unsigned long ceiling)
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 5c7f8f64f70e..f8071097302a 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -769,10 +769,8 @@ struct mm_walk {
 
 int walk_page_range(unsigned long addr, unsigned long end,
 		struct mm_walk *walk);
-void free_pgd_range(struct mmu_gather **tlb, unsigned long addr,
+void free_pgd_range(struct mmu_gather *tlb, unsigned long addr,
 		unsigned long end, unsigned long floor, unsigned long ceiling);
-void free_pgtables(struct mmu_gather **tlb, struct vm_area_struct *start_vma,
-		unsigned long floor, unsigned long ceiling);
 int copy_page_range(struct mm_struct *dst, struct mm_struct *src,
 			struct vm_area_struct *vma);
 void unmap_mapping_range(struct address_space *mapping,
diff --git a/mm/internal.h b/mm/internal.h
index 50807e12490e..858ad01864dc 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -13,6 +13,9 @@
 
 #include <linux/mm.h>
 
+void free_pgtables(struct mmu_gather *tlb, struct vm_area_struct *start_vma,
+		unsigned long floor, unsigned long ceiling);
+
 static inline void set_page_count(struct page *page, int v)
 {
 	atomic_set(&page->_count, v);
diff --git a/mm/memory.c b/mm/memory.c
index 87350321e66f..82f3f1c5cf17 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -61,6 +61,8 @@
 #include <linux/swapops.h>
 #include <linux/elf.h>
 
+#include "internal.h"
+
 #ifndef CONFIG_NEED_MULTIPLE_NODES
 /* use the per-pgdat data instead for discontigmem - mbligh */
 unsigned long max_mapnr;
@@ -211,7 +213,7 @@ static inline void free_pud_range(struct mmu_gather *tlb, pgd_t *pgd,
  *
  * Must be called with pagetable lock held.
  */
-void free_pgd_range(struct mmu_gather **tlb,
+void free_pgd_range(struct mmu_gather *tlb,
 			unsigned long addr, unsigned long end,
 			unsigned long floor, unsigned long ceiling)
 {
@@ -262,16 +264,16 @@ void free_pgd_range(struct mmu_gather **tlb,
 		return;
 
 	start = addr;
-	pgd = pgd_offset((*tlb)->mm, addr);
+	pgd = pgd_offset(tlb->mm, addr);
 	do {
 		next = pgd_addr_end(addr, end);
 		if (pgd_none_or_clear_bad(pgd))
 			continue;
-		free_pud_range(*tlb, pgd, addr, next, floor, ceiling);
+		free_pud_range(tlb, pgd, addr, next, floor, ceiling);
 	} while (pgd++, addr = next, addr != end);
 }
 
-void free_pgtables(struct mmu_gather **tlb, struct vm_area_struct *vma,
+void free_pgtables(struct mmu_gather *tlb, struct vm_area_struct *vma,
 		unsigned long floor, unsigned long ceiling)
 {
 	while (vma) {
diff --git a/mm/mmap.c b/mm/mmap.c
index 1d102b956fd8..75e0d0673d78 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -32,6 +32,8 @@
 #include <asm/tlb.h>
 #include <asm/mmu_context.h>
 
+#include "internal.h"
+
 #ifndef arch_mmap_check
 #define arch_mmap_check(addr, len, flags)	(0)
 #endif
@@ -1763,7 +1765,7 @@ static void unmap_region(struct mm_struct *mm,
 	update_hiwater_rss(mm);
 	unmap_vmas(&tlb, vma, start, end, &nr_accounted, NULL);
 	vm_unacct_memory(nr_accounted);
-	free_pgtables(&tlb, vma, prev? prev->vm_end: FIRST_USER_ADDRESS,
+	free_pgtables(tlb, vma, prev? prev->vm_end: FIRST_USER_ADDRESS,
 				 next? next->vm_start: 0);
 	tlb_finish_mmu(tlb, start, end);
 }
@@ -2063,7 +2065,7 @@ void exit_mmap(struct mm_struct *mm)
 	/* Use -1 here to ensure all VMAs in the mm are unmapped */
 	end = unmap_vmas(&tlb, vma, 0, -1, &nr_accounted, NULL);
 	vm_unacct_memory(nr_accounted);
-	free_pgtables(&tlb, vma, FIRST_USER_ADDRESS, 0);
+	free_pgtables(tlb, vma, FIRST_USER_ADDRESS, 0);
 	tlb_finish_mmu(tlb, 0, end);
 
 	/*

From 3c82d0ce2c4f642b2f24ef98707a030543b06b90 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:27:11 -0700
Subject: [PATCH 0227/1171] buddy: clarify comments describing buddy merge

In __free_one_page(), the comment "Move the buddy up one level" appears
attached to the break and by implication when the break is taken we are
moving it up one level:

	if (!page_is_buddy(page, buddy, order))
		break;          /* Move the buddy up one level. */

In reality the inverse is true, we break out when we can no longer merge
this page with its buddy.  Looking back into pre-history (into the full
git history) it appears that these two lines accidentally got joined as
part of another change.

Move the comment down where it belongs below the if and clarify its
language.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/page_alloc.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 35b1347d81bb..24aa3d1b9d96 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -432,8 +432,9 @@ static inline void __free_one_page(struct page *page,
 
 		buddy = __page_find_buddy(page, page_idx, order);
 		if (!page_is_buddy(page, buddy, order))
-			break;		/* Move the buddy up one level. */
+			break;
 
+		/* Our buddy is free, merge with it and move up one order. */
 		list_del(&buddy->lru);
 		zone->free_area[order].nr_free--;
 		rmv_page_order(buddy);

From da3bbdd4632c0171406b2677e31494afa5bde2f8 Mon Sep 17 00:00:00 2001
From: Kentaro Makita <k-makita@np.css.fujitsu.com>
Date: Wed, 23 Jul 2008 21:27:13 -0700
Subject: [PATCH 0228/1171] fix soft lock up at NFS mount via per-SB LRU-list
 of unused dentries

[Summary]

 Split LRU-list of unused dentries to one per superblock to avoid soft
 lock up during NFS mounts and remounting of any filesystem.

 Previously I posted here:
 http://lkml.org/lkml/2008/3/5/590

[Descriptions]

- background

  dentry_unused is a list of dentries which are not referenced.
  dentry_unused grows up when references on directories or files are
  released.  This list can be very long if there is huge free memory.

- the problem

  When shrink_dcache_sb() is called, it scans all dentry_unused linearly
  under spin_lock(), and if dentry->d_sb is differnt from given
  superblock, scan next dentry.  This scan costs very much if there are
  many entries, and very ineffective if there are many superblocks.

  IOW, When we need to shrink unused dentries on one dentry, but scans
  unused dentries on all superblocks in the system.  For example, we scan
  500 dentries to unmount a filesystem, but scans 1,000,000 or more unused
  dentries on other superblocks.

  In our case , At mounting NFS*, shrink_dcache_sb() is called to shrink
  unused dentries on NFS, but scans 100,000,000 unused dentries on
  superblocks in the system such as local ext3 filesystems.  I hear NFS
  mounting took 1 min on some system in use.

* : NFS uses virtual filesystem in rpc layer, so NFS is affected by
  this problem.

  100,000,000 is possible number on large systems.

  Per-superblock LRU of unused dentried can reduce the cost in
  reasonable manner.

- How to fix

  I found this problem is solved by David Chinner's "Per-superblock
  unused dentry LRU lists V3"(1), so I rebase it and add some fix to
  reclaim with fairness, which is in Andrew Morton's comments(2).

  1) http://lkml.org/lkml/2006/5/25/318
  2) http://lkml.org/lkml/2006/5/25/320

  Split LRU-list of unused dentries to each superblocks.  Then, NFS
  mounting will check dentries under a superblock instead of all.  But
  this spliting will break LRU of dentry-unused.  So, I've attempted to
  make reclaim unused dentrins with fairness by calculate number of
  dentries to scan on this sb based on following way

  number of dentries to scan on this sb =
  count * (number of dentries on this sb / number of dentries in the machine)

- ToDo
 - I have to measuring performance number and do stress tests.

 - When unmount occurs during prune_dcache(), scanning on same
  superblock, It is unable to reach next superblock because it is gone
  away.  We restart scannig superblock from first one, it causes
  unfairness of reclaim unused dentries on first superblock.  But I think
  this happens very rarely.

- Test Results

  Result on 6GB boxes with excessive unused dentries.

Without patch:

$ cat /proc/sys/fs/dentry-state
10181835        10180203        45      0       0       0
# mount -t nfs 10.124.60.70:/work/kernel-src nfs
real    0m1.830s
user    0m0.001s
sys     0m1.653s

 With this patch:
$ cat /proc/sys/fs/dentry-state
10236610        10234751        45      0       0       0
# mount -t nfs 10.124.60.70:/work/kernel-src nfs
real    0m0.106s
user    0m0.002s
sys     0m0.032s

[akpm@linux-foundation.org: fix comments]
Signed-off-by: Kentaro Makita <k-makita@np.css.fujitsu.com>
Cc: Neil Brown <neilb@suse.de>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: David Chinner <dgc@sgi.com>
Cc: "J. Bruce Fields" <bfields@fieldses.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/dcache.c        | 343 ++++++++++++++++++++++++---------------------
 fs/super.c         |   1 +
 include/linux/fs.h |   4 +
 3 files changed, 189 insertions(+), 159 deletions(-)

diff --git a/fs/dcache.c b/fs/dcache.c
index 6068c25b393c..3818d6ab76ca 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -61,7 +61,6 @@ static struct kmem_cache *dentry_cache __read_mostly;
 static unsigned int d_hash_mask __read_mostly;
 static unsigned int d_hash_shift __read_mostly;
 static struct hlist_head *dentry_hashtable __read_mostly;
-static LIST_HEAD(dentry_unused);
 
 /* Statistics gathering. */
 struct dentry_stat_t dentry_stat = {
@@ -96,14 +95,6 @@ static void d_free(struct dentry *dentry)
 		call_rcu(&dentry->d_u.d_rcu, d_callback);
 }
 
-static void dentry_lru_remove(struct dentry *dentry)
-{
-	if (!list_empty(&dentry->d_lru)) {
-		list_del_init(&dentry->d_lru);
-		dentry_stat.nr_unused--;
-	}
-}
-
 /*
  * Release the dentry's inode, using the filesystem
  * d_iput() operation if defined.
@@ -130,6 +121,41 @@ static void dentry_iput(struct dentry * dentry)
 	}
 }
 
+/*
+ * dentry_lru_(add|add_tail|del|del_init) must be called with dcache_lock held.
+ */
+static void dentry_lru_add(struct dentry *dentry)
+{
+	list_add(&dentry->d_lru, &dentry->d_sb->s_dentry_lru);
+	dentry->d_sb->s_nr_dentry_unused++;
+	dentry_stat.nr_unused++;
+}
+
+static void dentry_lru_add_tail(struct dentry *dentry)
+{
+	list_add_tail(&dentry->d_lru, &dentry->d_sb->s_dentry_lru);
+	dentry->d_sb->s_nr_dentry_unused++;
+	dentry_stat.nr_unused++;
+}
+
+static void dentry_lru_del(struct dentry *dentry)
+{
+	if (!list_empty(&dentry->d_lru)) {
+		list_del(&dentry->d_lru);
+		dentry->d_sb->s_nr_dentry_unused--;
+		dentry_stat.nr_unused--;
+	}
+}
+
+static void dentry_lru_del_init(struct dentry *dentry)
+{
+	if (likely(!list_empty(&dentry->d_lru))) {
+		list_del_init(&dentry->d_lru);
+		dentry->d_sb->s_nr_dentry_unused--;
+		dentry_stat.nr_unused--;
+	}
+}
+
 /**
  * d_kill - kill dentry and return parent
  * @dentry: dentry to kill
@@ -212,8 +238,7 @@ repeat:
 		goto kill_it;
   	if (list_empty(&dentry->d_lru)) {
   		dentry->d_flags |= DCACHE_REFERENCED;
-  		list_add(&dentry->d_lru, &dentry_unused);
-  		dentry_stat.nr_unused++;
+		dentry_lru_add(dentry);
   	}
  	spin_unlock(&dentry->d_lock);
 	spin_unlock(&dcache_lock);
@@ -222,7 +247,8 @@ repeat:
 unhash_it:
 	__d_drop(dentry);
 kill_it:
-	dentry_lru_remove(dentry);
+	/* if dentry was on the d_lru list delete it from there */
+	dentry_lru_del(dentry);
 	dentry = d_kill(dentry);
 	if (dentry)
 		goto repeat;
@@ -290,7 +316,7 @@ int d_invalidate(struct dentry * dentry)
 static inline struct dentry * __dget_locked(struct dentry *dentry)
 {
 	atomic_inc(&dentry->d_count);
-	dentry_lru_remove(dentry);
+	dentry_lru_del_init(dentry);
 	return dentry;
 }
 
@@ -406,132 +432,166 @@ static void prune_one_dentry(struct dentry * dentry)
 
 		if (dentry->d_op && dentry->d_op->d_delete)
 			dentry->d_op->d_delete(dentry);
-		dentry_lru_remove(dentry);
+		dentry_lru_del_init(dentry);
 		__d_drop(dentry);
 		dentry = d_kill(dentry);
 		spin_lock(&dcache_lock);
 	}
 }
 
-/**
- * prune_dcache - shrink the dcache
- * @count: number of entries to try and free
- * @sb: if given, ignore dentries for other superblocks
- *         which are being unmounted.
- *
- * Shrink the dcache. This is done when we need
- * more memory, or simply when we need to unmount
- * something (at which point we need to unuse
- * all dentries).
- *
- * This function may fail to free any resources if
- * all the dentries are in use.
+/*
+ * Shrink the dentry LRU on a given superblock.
+ * @sb   : superblock to shrink dentry LRU.
+ * @count: If count is NULL, we prune all dentries on superblock.
+ * @flags: If flags is non-zero, we need to do special processing based on
+ * which flags are set. This means we don't need to maintain multiple
+ * similar copies of this loop.
  */
- 
-static void prune_dcache(int count, struct super_block *sb)
+static void __shrink_dcache_sb(struct super_block *sb, int *count, int flags)
 {
+	LIST_HEAD(referenced);
+	LIST_HEAD(tmp);
+	struct dentry *dentry;
+	int cnt = 0;
+
+	BUG_ON(!sb);
+	BUG_ON((flags & DCACHE_REFERENCED) && count == NULL);
 	spin_lock(&dcache_lock);
-	for (; count ; count--) {
-		struct dentry *dentry;
-		struct list_head *tmp;
-		struct rw_semaphore *s_umount;
+	if (count != NULL)
+		/* called from prune_dcache() and shrink_dcache_parent() */
+		cnt = *count;
+restart:
+	if (count == NULL)
+		list_splice_init(&sb->s_dentry_lru, &tmp);
+	else {
+		while (!list_empty(&sb->s_dentry_lru)) {
+			dentry = list_entry(sb->s_dentry_lru.prev,
+					struct dentry, d_lru);
+			BUG_ON(dentry->d_sb != sb);
 
-		cond_resched_lock(&dcache_lock);
-
-		tmp = dentry_unused.prev;
-		if (sb) {
-			/* Try to find a dentry for this sb, but don't try
-			 * too hard, if they aren't near the tail they will
-			 * be moved down again soon
+			spin_lock(&dentry->d_lock);
+			/*
+			 * If we are honouring the DCACHE_REFERENCED flag and
+			 * the dentry has this flag set, don't free it. Clear
+			 * the flag and put it back on the LRU.
 			 */
-			int skip = count;
-			while (skip && tmp != &dentry_unused &&
-			    list_entry(tmp, struct dentry, d_lru)->d_sb != sb) {
-				skip--;
-				tmp = tmp->prev;
+			if ((flags & DCACHE_REFERENCED)
+				&& (dentry->d_flags & DCACHE_REFERENCED)) {
+				dentry->d_flags &= ~DCACHE_REFERENCED;
+				list_move_tail(&dentry->d_lru, &referenced);
+				spin_unlock(&dentry->d_lock);
+			} else {
+				list_move_tail(&dentry->d_lru, &tmp);
+				spin_unlock(&dentry->d_lock);
+				cnt--;
+				if (!cnt)
+					break;
 			}
 		}
-		if (tmp == &dentry_unused)
-			break;
-		list_del_init(tmp);
-		prefetch(dentry_unused.prev);
- 		dentry_stat.nr_unused--;
-		dentry = list_entry(tmp, struct dentry, d_lru);
-
- 		spin_lock(&dentry->d_lock);
+	}
+	while (!list_empty(&tmp)) {
+		dentry = list_entry(tmp.prev, struct dentry, d_lru);
+		dentry_lru_del_init(dentry);
+		spin_lock(&dentry->d_lock);
 		/*
 		 * We found an inuse dentry which was not removed from
-		 * dentry_unused because of laziness during lookup.  Do not free
-		 * it - just keep it off the dentry_unused list.
+		 * the LRU because of laziness during lookup.  Do not free
+		 * it - just keep it off the LRU list.
 		 */
- 		if (atomic_read(&dentry->d_count)) {
- 			spin_unlock(&dentry->d_lock);
+		if (atomic_read(&dentry->d_count)) {
+			spin_unlock(&dentry->d_lock);
 			continue;
 		}
-		/* If the dentry was recently referenced, don't free it. */
-		if (dentry->d_flags & DCACHE_REFERENCED) {
-			dentry->d_flags &= ~DCACHE_REFERENCED;
- 			list_add(&dentry->d_lru, &dentry_unused);
- 			dentry_stat.nr_unused++;
- 			spin_unlock(&dentry->d_lock);
-			continue;
-		}
-		/*
-		 * If the dentry is not DCACHED_REFERENCED, it is time
-		 * to remove it from the dcache, provided the super block is
-		 * NULL (which means we are trying to reclaim memory)
-		 * or this dentry belongs to the same super block that
-		 * we want to shrink.
-		 */
-		/*
-		 * If this dentry is for "my" filesystem, then I can prune it
-		 * without taking the s_umount lock (I already hold it).
-		 */
-		if (sb && dentry->d_sb == sb) {
-			prune_one_dentry(dentry);
-			continue;
-		}
-		/*
-		 * ...otherwise we need to be sure this filesystem isn't being
-		 * unmounted, otherwise we could race with
-		 * generic_shutdown_super(), and end up holding a reference to
-		 * an inode while the filesystem is unmounted.
-		 * So we try to get s_umount, and make sure s_root isn't NULL.
-		 * (Take a local copy of s_umount to avoid a use-after-free of
-		 * `dentry').
-		 */
-		s_umount = &dentry->d_sb->s_umount;
-		if (down_read_trylock(s_umount)) {
-			if (dentry->d_sb->s_root != NULL) {
-				prune_one_dentry(dentry);
-				up_read(s_umount);
-				continue;
-			}
-			up_read(s_umount);
-		}
-		spin_unlock(&dentry->d_lock);
-		/*
-		 * Insert dentry at the head of the list as inserting at the
-		 * tail leads to a cycle.
-		 */
- 		list_add(&dentry->d_lru, &dentry_unused);
-		dentry_stat.nr_unused++;
+		prune_one_dentry(dentry);
+		/* dentry->d_lock was dropped in prune_one_dentry() */
+		cond_resched_lock(&dcache_lock);
 	}
+	if (count == NULL && !list_empty(&sb->s_dentry_lru))
+		goto restart;
+	if (count != NULL)
+		*count = cnt;
+	if (!list_empty(&referenced))
+		list_splice(&referenced, &sb->s_dentry_lru);
 	spin_unlock(&dcache_lock);
 }
 
-/*
- * Shrink the dcache for the specified super block.
- * This allows us to unmount a device without disturbing
- * the dcache for the other devices.
+/**
+ * prune_dcache - shrink the dcache
+ * @count: number of entries to try to free
  *
- * This implementation makes just two traversals of the
- * unused list.  On the first pass we move the selected
- * dentries to the most recent end, and on the second
- * pass we free them.  The second pass must restart after
- * each dput(), but since the target dentries are all at
- * the end, it's really just a single traversal.
+ * Shrink the dcache. This is done when we need more memory, or simply when we
+ * need to unmount something (at which point we need to unuse all dentries).
+ *
+ * This function may fail to free any resources if all the dentries are in use.
  */
+static void prune_dcache(int count)
+{
+	struct super_block *sb;
+	int w_count;
+	int unused = dentry_stat.nr_unused;
+	int prune_ratio;
+	int pruned;
+
+	if (unused == 0 || count == 0)
+		return;
+	spin_lock(&dcache_lock);
+restart:
+	if (count >= unused)
+		prune_ratio = 1;
+	else
+		prune_ratio = unused / count;
+	spin_lock(&sb_lock);
+	list_for_each_entry(sb, &super_blocks, s_list) {
+		if (sb->s_nr_dentry_unused == 0)
+			continue;
+		sb->s_count++;
+		/* Now, we reclaim unused dentrins with fairness.
+		 * We reclaim them same percentage from each superblock.
+		 * We calculate number of dentries to scan on this sb
+		 * as follows, but the implementation is arranged to avoid
+		 * overflows:
+		 * number of dentries to scan on this sb =
+		 * count * (number of dentries on this sb /
+		 * number of dentries in the machine)
+		 */
+		spin_unlock(&sb_lock);
+		if (prune_ratio != 1)
+			w_count = (sb->s_nr_dentry_unused / prune_ratio) + 1;
+		else
+			w_count = sb->s_nr_dentry_unused;
+		pruned = w_count;
+		/*
+		 * We need to be sure this filesystem isn't being unmounted,
+		 * otherwise we could race with generic_shutdown_super(), and
+		 * end up holding a reference to an inode while the filesystem
+		 * is unmounted.  So we try to get s_umount, and make sure
+		 * s_root isn't NULL.
+		 */
+		if (down_read_trylock(&sb->s_umount)) {
+			if ((sb->s_root != NULL) &&
+			    (!list_empty(&sb->s_dentry_lru))) {
+				spin_unlock(&dcache_lock);
+				__shrink_dcache_sb(sb, &w_count,
+						DCACHE_REFERENCED);
+				pruned -= w_count;
+				spin_lock(&dcache_lock);
+			}
+			up_read(&sb->s_umount);
+		}
+		spin_lock(&sb_lock);
+		count -= pruned;
+		/*
+		 * restart only when sb is no longer on the list and
+		 * we have more work to do.
+		 */
+		if (__put_super_and_need_restart(sb) && count > 0) {
+			spin_unlock(&sb_lock);
+			goto restart;
+		}
+	}
+	spin_unlock(&sb_lock);
+	spin_unlock(&dcache_lock);
+}
 
 /**
  * shrink_dcache_sb - shrink dcache for a superblock
@@ -541,44 +601,9 @@ static void prune_dcache(int count, struct super_block *sb)
  * is used to free the dcache before unmounting a file
  * system
  */
-
 void shrink_dcache_sb(struct super_block * sb)
 {
-	struct list_head *tmp, *next;
-	struct dentry *dentry;
-
-	/*
-	 * Pass one ... move the dentries for the specified
-	 * superblock to the most recent end of the unused list.
-	 */
-	spin_lock(&dcache_lock);
-	list_for_each_prev_safe(tmp, next, &dentry_unused) {
-		dentry = list_entry(tmp, struct dentry, d_lru);
-		if (dentry->d_sb != sb)
-			continue;
-		list_move_tail(tmp, &dentry_unused);
-	}
-
-	/*
-	 * Pass two ... free the dentries for this superblock.
-	 */
-repeat:
-	list_for_each_prev_safe(tmp, next, &dentry_unused) {
-		dentry = list_entry(tmp, struct dentry, d_lru);
-		if (dentry->d_sb != sb)
-			continue;
-		dentry_stat.nr_unused--;
-		list_del_init(tmp);
-		spin_lock(&dentry->d_lock);
-		if (atomic_read(&dentry->d_count)) {
-			spin_unlock(&dentry->d_lock);
-			continue;
-		}
-		prune_one_dentry(dentry);
-		cond_resched_lock(&dcache_lock);
-		goto repeat;
-	}
-	spin_unlock(&dcache_lock);
+	__shrink_dcache_sb(sb, NULL, 0);
 }
 
 /*
@@ -595,7 +620,7 @@ static void shrink_dcache_for_umount_subtree(struct dentry *dentry)
 
 	/* detach this root from the system */
 	spin_lock(&dcache_lock);
-	dentry_lru_remove(dentry);
+	dentry_lru_del_init(dentry);
 	__d_drop(dentry);
 	spin_unlock(&dcache_lock);
 
@@ -609,7 +634,7 @@ static void shrink_dcache_for_umount_subtree(struct dentry *dentry)
 			spin_lock(&dcache_lock);
 			list_for_each_entry(loop, &dentry->d_subdirs,
 					    d_u.d_child) {
-				dentry_lru_remove(loop);
+				dentry_lru_del_init(loop);
 				__d_drop(loop);
 				cond_resched_lock(&dcache_lock);
 			}
@@ -791,14 +816,13 @@ resume:
 		struct dentry *dentry = list_entry(tmp, struct dentry, d_u.d_child);
 		next = tmp->next;
 
-		dentry_lru_remove(dentry);
+		dentry_lru_del_init(dentry);
 		/* 
 		 * move only zero ref count dentries to the end 
 		 * of the unused list for prune_dcache
 		 */
 		if (!atomic_read(&dentry->d_count)) {
-			list_add_tail(&dentry->d_lru, &dentry_unused);
-			dentry_stat.nr_unused++;
+			dentry_lru_add_tail(dentry);
 			found++;
 		}
 
@@ -840,10 +864,11 @@ out:
  
 void shrink_dcache_parent(struct dentry * parent)
 {
+	struct super_block *sb = parent->d_sb;
 	int found;
 
 	while ((found = select_parent(parent)) != 0)
-		prune_dcache(found, parent->d_sb);
+		__shrink_dcache_sb(sb, &found, 0);
 }
 
 /*
@@ -863,7 +888,7 @@ static int shrink_dcache_memory(int nr, gfp_t gfp_mask)
 	if (nr) {
 		if (!(gfp_mask & __GFP_FS))
 			return -1;
-		prune_dcache(nr, NULL);
+		prune_dcache(nr);
 	}
 	return (dentry_stat.nr_unused / 100) * sysctl_vfs_cache_pressure;
 }
@@ -1215,7 +1240,7 @@ struct dentry *d_splice_alias(struct inode *inode, struct dentry *dentry)
  * rcu_read_lock() and rcu_read_unlock() are used to disable preemption while
  * lookup is going on.
  *
- * dentry_unused list is not updated even if lookup finds the required dentry
+ * The dentry unused LRU is not updated even if lookup finds the required dentry
  * in there. It is updated in places such as prune_dcache, shrink_dcache_sb,
  * select_parent and __dget_locked. This laziness saves lookup from dcache_lock
  * acquisition.
diff --git a/fs/super.c b/fs/super.c
index 453877c5697b..e931ae9511fe 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -70,6 +70,7 @@ static struct super_block *alloc_super(struct file_system_type *type)
 		INIT_LIST_HEAD(&s->s_instances);
 		INIT_HLIST_HEAD(&s->s_anon);
 		INIT_LIST_HEAD(&s->s_inodes);
+		INIT_LIST_HEAD(&s->s_dentry_lru);
 		init_rwsem(&s->s_umount);
 		mutex_init(&s->s_lock);
 		lockdep_set_class(&s->s_umount, &type->s_umount_key);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index ff54ae4933f3..e5e6a244096c 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1025,6 +1025,7 @@ extern int send_sigurg(struct fown_struct *fown);
 extern struct list_head super_blocks;
 extern spinlock_t sb_lock;
 
+#define sb_entry(list)  list_entry((list), struct super_block, s_list)
 #define S_BIAS (1<<30)
 struct super_block {
 	struct list_head	s_list;		/* Keep this first */
@@ -1058,6 +1059,9 @@ struct super_block {
 	struct list_head	s_more_io;	/* parked for more writeback */
 	struct hlist_head	s_anon;		/* anonymous dentries for (nfs) exporting */
 	struct list_head	s_files;
+	/* s_dentry_lru and s_nr_dentry_unused are protected by dcache_lock */
+	struct list_head	s_dentry_lru;	/* unused dentry lru */
+	int			s_nr_dentry_unused;	/* # of dentry on lru */
 
 	struct block_device	*s_bdev;
 	struct mtd_info		*s_mtd;

From 0cad47cf13bc2e9142d3a11d9f50523797d0d4ea Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:27:16 -0700
Subject: [PATCH 0229/1171] page-flags: record page flag overlays explicitly

With the recent page flag reorganisation we have a single enum which
defines the valid page flags and their values, nice and clear.  However
there are a number of bits which are overloaded by different subsystems.
Firstly there is PG_owner_priv_1 which is used by filesystems and by XEN.
Secondly both SLOB and SLUB use a couple of extra page bits to manage
internal state for pages they own; both overlay other bits.  All of these
"aliases" are scattered about the source making it very hard for a reader
to know if the bits are safe to rely on in all contexts; confusion here is
bad.

As we now have a single place where the bits are clearly assigned it makes
sense to clarify the reuse of bits by making the aliases explicit and
visible with the original bit assignments.  This patch creates explicit
aliases within the enum itself for the overloaded bits, creates standard
bit accessors PageFoo etc.  and uses those throughout.

This version pulls the bit manipulation out to standard named page bit
accessors as suggested by Christoph, it retains the explicit mapping to
the overlayed bits.  A fusion of both ideas.  This has been SLUB and SLOB
have been compile tested on x86_64 only, and SLUB boot tested.  If people
feel this is worth doing then I can run a fuller set of testing.

This patch:

Some page flags are used for more than one purpose, for example
PG_owner_priv_1.  Currently there are individual accessors for each user,
each built using the common flag name far away from the bit definitions.
This makes it hard to see all possible uses of these bits.

Now that we have a single enum to generate the bit orders it makes sense
to express overlays in the same place.  So create per use aliases for this
bit in the main page-flags enum and use those in the accessors.

[akpm@linux-foundation.org: fix xen]
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Matt Mackall <mpm@selenic.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/page-flags.h | 15 +++++++++++----
 1 file changed, 11 insertions(+), 4 deletions(-)

diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h
index 0d2a4e7012aa..7d8db1233e44 100644
--- a/include/linux/page-flags.h
+++ b/include/linux/page-flags.h
@@ -96,7 +96,14 @@ enum pageflags {
 #ifdef CONFIG_IA64_UNCACHED_ALLOCATOR
 	PG_uncached,		/* Page has been mapped as uncached */
 #endif
-	__NR_PAGEFLAGS
+	__NR_PAGEFLAGS,
+
+	/* Filesystems */
+	PG_checked = PG_owner_priv_1,
+
+	/* XEN */
+	PG_pinned = PG_owner_priv_1,
+	PG_savepinned = PG_dirty,
 };
 
 #ifndef __GENERATING_BOUNDS_H
@@ -155,9 +162,9 @@ PAGEFLAG(Dirty, dirty) TESTSCFLAG(Dirty, dirty) __CLEARPAGEFLAG(Dirty, dirty)
 PAGEFLAG(LRU, lru) __CLEARPAGEFLAG(LRU, lru)
 PAGEFLAG(Active, active) __CLEARPAGEFLAG(Active, active)
 __PAGEFLAG(Slab, slab)
-PAGEFLAG(Checked, owner_priv_1)		/* Used by some filesystems */
-PAGEFLAG(Pinned, owner_priv_1) TESTSCFLAG(Pinned, owner_priv_1) /* Xen */
-PAGEFLAG(SavePinned, dirty);					/* Xen */
+PAGEFLAG(Checked, checked)		/* Used by some filesystems */
+PAGEFLAG(Pinned, pinned) TESTSCFLAG(Pinned, pinned)	/* Xen */
+PAGEFLAG(SavePinned, savepinned);			/* Xen */
 PAGEFLAG(Reserved, reserved) __CLEARPAGEFLAG(Reserved, reserved)
 PAGEFLAG(Private, private) __CLEARPAGEFLAG(Private, private)
 	__SETPAGEFLAG(Private, private)

From 8a38082d21cbc5ec961da7dda195e98a9a064dcf Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:27:18 -0700
Subject: [PATCH 0230/1171] slub: record page flag overlays explicitly

SLUB reuses two page bits for internal purposes, it overlays PG_active and
PG_error.  This is hidden away in slub.c.  Document these overlays
explicitly in the main page-flags enum along with all the others.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Matt Mackall <mpm@selenic.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Tested-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/page-flags.h |  7 ++++
 mm/slub.c                  | 65 ++++++++++----------------------------
 2 files changed, 24 insertions(+), 48 deletions(-)

diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h
index 7d8db1233e44..3fc586b7b90b 100644
--- a/include/linux/page-flags.h
+++ b/include/linux/page-flags.h
@@ -104,6 +104,10 @@ enum pageflags {
 	/* XEN */
 	PG_pinned = PG_owner_priv_1,
 	PG_savepinned = PG_dirty,
+
+	/* SLUB */
+	PG_slub_frozen = PG_active,
+	PG_slub_debug = PG_error,
 };
 
 #ifndef __GENERATING_BOUNDS_H
@@ -169,6 +173,9 @@ PAGEFLAG(Reserved, reserved) __CLEARPAGEFLAG(Reserved, reserved)
 PAGEFLAG(Private, private) __CLEARPAGEFLAG(Private, private)
 	__SETPAGEFLAG(Private, private)
 
+__PAGEFLAG(SlubFrozen, slub_frozen)
+__PAGEFLAG(SlubDebug, slub_debug)
+
 /*
  * Only test-and-set exist for PG_writeback.  The unconditional operators are
  * risky: they bypass page accounting.
diff --git a/mm/slub.c b/mm/slub.c
index 6d4a49c1ff2f..77c21cf53ff9 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -102,44 +102,12 @@
  * 			the fast path and disables lockless freelists.
  */
 
-#define FROZEN (1 << PG_active)
-
 #ifdef CONFIG_SLUB_DEBUG
-#define SLABDEBUG (1 << PG_error)
+#define SLABDEBUG 1
 #else
 #define SLABDEBUG 0
 #endif
 
-static inline int SlabFrozen(struct page *page)
-{
-	return page->flags & FROZEN;
-}
-
-static inline void SetSlabFrozen(struct page *page)
-{
-	page->flags |= FROZEN;
-}
-
-static inline void ClearSlabFrozen(struct page *page)
-{
-	page->flags &= ~FROZEN;
-}
-
-static inline int SlabDebug(struct page *page)
-{
-	return page->flags & SLABDEBUG;
-}
-
-static inline void SetSlabDebug(struct page *page)
-{
-	page->flags |= SLABDEBUG;
-}
-
-static inline void ClearSlabDebug(struct page *page)
-{
-	page->flags &= ~SLABDEBUG;
-}
-
 /*
  * Issues still to be resolved:
  *
@@ -971,7 +939,7 @@ static int free_debug_processing(struct kmem_cache *s, struct page *page,
 	}
 
 	/* Special debug activities for freeing objects */
-	if (!SlabFrozen(page) && !page->freelist)
+	if (!PageSlubFrozen(page) && !page->freelist)
 		remove_full(s, page);
 	if (s->flags & SLAB_STORE_USER)
 		set_track(s, object, TRACK_FREE, addr);
@@ -1157,7 +1125,7 @@ static struct page *new_slab(struct kmem_cache *s, gfp_t flags, int node)
 	page->flags |= 1 << PG_slab;
 	if (s->flags & (SLAB_DEBUG_FREE | SLAB_RED_ZONE | SLAB_POISON |
 			SLAB_STORE_USER | SLAB_TRACE))
-		SetSlabDebug(page);
+		__SetPageSlubDebug(page);
 
 	start = page_address(page);
 
@@ -1184,14 +1152,14 @@ static void __free_slab(struct kmem_cache *s, struct page *page)
 	int order = compound_order(page);
 	int pages = 1 << order;
 
-	if (unlikely(SlabDebug(page))) {
+	if (unlikely(SLABDEBUG && PageSlubDebug(page))) {
 		void *p;
 
 		slab_pad_check(s, page);
 		for_each_object(p, s, page_address(page),
 						page->objects)
 			check_object(s, page, p, 0);
-		ClearSlabDebug(page);
+		__ClearPageSlubDebug(page);
 	}
 
 	mod_zone_page_state(page_zone(page),
@@ -1288,7 +1256,7 @@ static inline int lock_and_freeze_slab(struct kmem_cache_node *n,
 	if (slab_trylock(page)) {
 		list_del(&page->lru);
 		n->nr_partial--;
-		SetSlabFrozen(page);
+		__SetPageSlubFrozen(page);
 		return 1;
 	}
 	return 0;
@@ -1398,7 +1366,7 @@ static void unfreeze_slab(struct kmem_cache *s, struct page *page, int tail)
 	struct kmem_cache_node *n = get_node(s, page_to_nid(page));
 	struct kmem_cache_cpu *c = get_cpu_slab(s, smp_processor_id());
 
-	ClearSlabFrozen(page);
+	__ClearPageSlubFrozen(page);
 	if (page->inuse) {
 
 		if (page->freelist) {
@@ -1406,7 +1374,8 @@ static void unfreeze_slab(struct kmem_cache *s, struct page *page, int tail)
 			stat(c, tail ? DEACTIVATE_TO_TAIL : DEACTIVATE_TO_HEAD);
 		} else {
 			stat(c, DEACTIVATE_FULL);
-			if (SlabDebug(page) && (s->flags & SLAB_STORE_USER))
+			if (SLABDEBUG && PageSlubDebug(page) &&
+						(s->flags & SLAB_STORE_USER))
 				add_full(n, page);
 		}
 		slab_unlock(page);
@@ -1551,7 +1520,7 @@ load_freelist:
 	object = c->page->freelist;
 	if (unlikely(!object))
 		goto another_slab;
-	if (unlikely(SlabDebug(c->page)))
+	if (unlikely(SLABDEBUG && PageSlubDebug(c->page)))
 		goto debug;
 
 	c->freelist = object[c->offset];
@@ -1588,7 +1557,7 @@ new_slab:
 		if (c->page)
 			flush_slab(s, c);
 		slab_lock(new);
-		SetSlabFrozen(new);
+		__SetPageSlubFrozen(new);
 		c->page = new;
 		goto load_freelist;
 	}
@@ -1674,7 +1643,7 @@ static void __slab_free(struct kmem_cache *s, struct page *page,
 	stat(c, FREE_SLOWPATH);
 	slab_lock(page);
 
-	if (unlikely(SlabDebug(page)))
+	if (unlikely(SLABDEBUG && PageSlubDebug(page)))
 		goto debug;
 
 checks_ok:
@@ -1682,7 +1651,7 @@ checks_ok:
 	page->freelist = object;
 	page->inuse--;
 
-	if (unlikely(SlabFrozen(page))) {
+	if (unlikely(PageSlubFrozen(page))) {
 		stat(c, FREE_FROZEN);
 		goto out_unlock;
 	}
@@ -3317,12 +3286,12 @@ static void validate_slab_slab(struct kmem_cache *s, struct page *page,
 			s->name, page);
 
 	if (s->flags & DEBUG_DEFAULT_FLAGS) {
-		if (!SlabDebug(page))
-			printk(KERN_ERR "SLUB %s: SlabDebug not set "
+		if (!PageSlubDebug(page))
+			printk(KERN_ERR "SLUB %s: SlubDebug not set "
 				"on slab 0x%p\n", s->name, page);
 	} else {
-		if (SlabDebug(page))
-			printk(KERN_ERR "SLUB %s: SlabDebug set on "
+		if (PageSlubDebug(page))
+			printk(KERN_ERR "SLUB %s: SlubDebug set on "
 				"slab 0x%p\n", s->name, page);
 	}
 }

From 9023cb7e8564d95a1893f8cb6895a293be9a71fe Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:27:19 -0700
Subject: [PATCH 0231/1171] slob: record page flag overlays explicitly

SLOB reuses two page bits for internal purposes, it overlays PG_active and
PG_private.  This is hidden away in slob.c.  Document these overlays
explicitly in the main page-flags enum along with all the others.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Matt Mackall <mpm@selenic.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Jeremy Fitzhardinge <jeremy@goop.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/page-flags.h |  7 +++++++
 mm/slob.c                  | 12 ++++++------
 2 files changed, 13 insertions(+), 6 deletions(-)

diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h
index 3fc586b7b90b..54590a9a103e 100644
--- a/include/linux/page-flags.h
+++ b/include/linux/page-flags.h
@@ -105,6 +105,10 @@ enum pageflags {
 	PG_pinned = PG_owner_priv_1,
 	PG_savepinned = PG_dirty,
 
+	/* SLOB */
+	PG_slob_page = PG_active,
+	PG_slob_free = PG_private,
+
 	/* SLUB */
 	PG_slub_frozen = PG_active,
 	PG_slub_debug = PG_error,
@@ -173,6 +177,9 @@ PAGEFLAG(Reserved, reserved) __CLEARPAGEFLAG(Reserved, reserved)
 PAGEFLAG(Private, private) __CLEARPAGEFLAG(Private, private)
 	__SETPAGEFLAG(Private, private)
 
+__PAGEFLAG(SlobPage, slob_page)
+__PAGEFLAG(SlobFree, slob_free)
+
 __PAGEFLAG(SlubFrozen, slub_frozen)
 __PAGEFLAG(SlubDebug, slub_debug)
 
diff --git a/mm/slob.c b/mm/slob.c
index a3ad6671adf1..de268eb7ac70 100644
--- a/mm/slob.c
+++ b/mm/slob.c
@@ -130,17 +130,17 @@ static LIST_HEAD(free_slob_large);
  */
 static inline int slob_page(struct slob_page *sp)
 {
-	return test_bit(PG_active, &sp->flags);
+	return PageSlobPage((struct page *)sp);
 }
 
 static inline void set_slob_page(struct slob_page *sp)
 {
-	__set_bit(PG_active, &sp->flags);
+	__SetPageSlobPage((struct page *)sp);
 }
 
 static inline void clear_slob_page(struct slob_page *sp)
 {
-	__clear_bit(PG_active, &sp->flags);
+	__ClearPageSlobPage((struct page *)sp);
 }
 
 /*
@@ -148,19 +148,19 @@ static inline void clear_slob_page(struct slob_page *sp)
  */
 static inline int slob_page_free(struct slob_page *sp)
 {
-	return test_bit(PG_private, &sp->flags);
+	return PageSlobFree((struct page *)sp);
 }
 
 static void set_slob_page_free(struct slob_page *sp, struct list_head *list)
 {
 	list_add(&sp->list, list);
-	__set_bit(PG_private, &sp->flags);
+	__SetPageSlobFree((struct page *)sp);
 }
 
 static inline void clear_slob_page_free(struct slob_page *sp)
 {
 	list_del(&sp->list);
-	__clear_bit(PG_private, &sp->flags);
+	__ClearPageSlobFree((struct page *)sp);
 }
 
 #define SLOB_UNIT sizeof(slob_t)

From 2185e69f680ae8c8496b6fc15e20c889d5b39b67 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@linux-foundation.org>
Date: Wed, 23 Jul 2008 21:27:19 -0700
Subject: [PATCH 0232/1171] mapping_set_error: add unlikely()

This is called on a per-page basis and in the vast majority of cases
`error' is zero.

Cc: Guillaume Chazarain <guichaz@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/pagemap.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h
index d2fca802f809..ee1ec2c7723c 100644
--- a/include/linux/pagemap.h
+++ b/include/linux/pagemap.h
@@ -22,7 +22,7 @@
 
 static inline void mapping_set_error(struct address_space *mapping, int error)
 {
-	if (error) {
+	if (unlikely(error)) {
 		if (error == -ENOSPC)
 			set_bit(AS_ENOSPC, &mapping->flags);
 		else

From 9109fb7b3520de187ebc3646c209d66a233f7169 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Wed, 23 Jul 2008 21:27:20 -0700
Subject: [PATCH 0233/1171] mm: drop unneeded pgdat argument from
 free_area_init_node()

free_area_init_node() gets passed in the node id as well as the node
descriptor.  This is redundant as the function can trivially get the node
descriptor itself by means of NODE_DATA() and the node's id.

I checked all the users and NODE_DATA() seems to be usable everywhere
from where this function is called.

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/alpha/mm/numa.c         |  2 +-
 arch/arm/mm/init.c           |  2 +-
 arch/avr32/mm/init.c         |  2 +-
 arch/cris/arch-v10/mm/init.c |  2 +-
 arch/cris/arch-v32/mm/init.c |  2 +-
 arch/m32r/mm/discontig.c     |  3 +--
 arch/m32r/mm/init.c          |  2 +-
 arch/m68k/mm/motorola.c      |  2 +-
 arch/m68k/mm/sun3mmu.c       |  2 +-
 arch/parisc/mm/init.c        |  2 +-
 arch/sparc/mm/srmmu.c        |  3 +--
 arch/sparc/mm/sun4c.c        |  3 +--
 arch/v850/kernel/setup.c     |  3 +--
 include/linux/mm.h           |  5 ++---
 mm/memory_hotplug.c          |  2 +-
 mm/page_alloc.c              | 11 ++++++-----
 16 files changed, 22 insertions(+), 26 deletions(-)

diff --git a/arch/alpha/mm/numa.c b/arch/alpha/mm/numa.c
index a53fda0481ca..def0c74a78a8 100644
--- a/arch/alpha/mm/numa.c
+++ b/arch/alpha/mm/numa.c
@@ -313,7 +313,7 @@ void __init paging_init(void)
 			zones_size[ZONE_DMA] = dma_local_pfn;
 			zones_size[ZONE_NORMAL] = (end_pfn - start_pfn) - dma_local_pfn;
 		}
-		free_area_init_node(nid, NODE_DATA(nid), zones_size, start_pfn, NULL);
+		free_area_init_node(nid, zones_size, start_pfn, NULL);
 	}
 
 	/* Initialize the kernel's ZERO_PGE. */
diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c
index b657f1719af0..e6352946dde0 100644
--- a/arch/arm/mm/init.c
+++ b/arch/arm/mm/init.c
@@ -284,7 +284,7 @@ bootmem_init_node(int node, int initrd_node, struct meminfo *mi)
 	 */
 	arch_adjust_zones(node, zone_size, zhole_size);
 
-	free_area_init_node(node, pgdat, zone_size, start_pfn, zhole_size);
+	free_area_init_node(node, zone_size, start_pfn, zhole_size);
 
 	return end_pfn;
 }
diff --git a/arch/avr32/mm/init.c b/arch/avr32/mm/init.c
index 3f90a87527bb..786de88a82a7 100644
--- a/arch/avr32/mm/init.c
+++ b/arch/avr32/mm/init.c
@@ -129,7 +129,7 @@ void __init paging_init(void)
 		printk("Node %u: start_pfn = 0x%lx, low = 0x%lx\n",
 		       nid, start_pfn, low);
 
-		free_area_init_node(nid, pgdat, zones_size, start_pfn, NULL);
+		free_area_init_node(nid, zones_size, start_pfn, NULL);
 
 		printk("Node %u: mem_map starts at %p\n",
 		       pgdat->node_id, pgdat->node_mem_map);
diff --git a/arch/cris/arch-v10/mm/init.c b/arch/cris/arch-v10/mm/init.c
index e0fcd1a9bfd5..742fd1974c2e 100644
--- a/arch/cris/arch-v10/mm/init.c
+++ b/arch/cris/arch-v10/mm/init.c
@@ -182,7 +182,7 @@ paging_init(void)
 	 * mem_map page array.
 	 */
 
-	free_area_init_node(0, &contig_page_data, zones_size, PAGE_OFFSET >> PAGE_SHIFT, 0);
+	free_area_init_node(0, zones_size, PAGE_OFFSET >> PAGE_SHIFT, 0);
 }
 
 /* Initialize remaps of some I/O-ports. It is important that this
diff --git a/arch/cris/arch-v32/mm/init.c b/arch/cris/arch-v32/mm/init.c
index 5a9ac5834647..8a34b8b74293 100644
--- a/arch/cris/arch-v32/mm/init.c
+++ b/arch/cris/arch-v32/mm/init.c
@@ -162,7 +162,7 @@ paging_init(void)
 	 * substantially higher than 0, like us (we start at PAGE_OFFSET). This
 	 * saves space in the mem_map page array.
 	 */
-	free_area_init_node(0, &contig_page_data, zones_size, PAGE_OFFSET >> PAGE_SHIFT, 0);
+	free_area_init_node(0, zones_size, PAGE_OFFSET >> PAGE_SHIFT, 0);
 
 	mem_map = contig_page_data.node_mem_map;
 }
diff --git a/arch/m32r/mm/discontig.c b/arch/m32r/mm/discontig.c
index aa9145ef6cca..cc23934bc41e 100644
--- a/arch/m32r/mm/discontig.c
+++ b/arch/m32r/mm/discontig.c
@@ -147,8 +147,7 @@ unsigned long __init zone_sizes_init(void)
 		zholes_size[ZONE_DMA] = mp->holes;
 		holes += zholes_size[ZONE_DMA];
 
-		free_area_init_node(nid, NODE_DATA(nid), zones_size,
-			start_pfn, zholes_size);
+		free_area_init_node(nid, zones_size, start_pfn, zholes_size);
 	}
 
 	/*
diff --git a/arch/m32r/mm/init.c b/arch/m32r/mm/init.c
index bbd97c85bc5d..28799af15e95 100644
--- a/arch/m32r/mm/init.c
+++ b/arch/m32r/mm/init.c
@@ -123,7 +123,7 @@ unsigned long __init zone_sizes_init(void)
 	start_pfn = __MEMORY_START >> PAGE_SHIFT;
 #endif /* CONFIG_MMU */
 
-	free_area_init_node(0, NODE_DATA(0), zones_size, start_pfn, 0);
+	free_area_init_node(0, zones_size, start_pfn, 0);
 
 	return 0;
 }
diff --git a/arch/m68k/mm/motorola.c b/arch/m68k/mm/motorola.c
index 226795bdf355..c5dbb9bdb322 100644
--- a/arch/m68k/mm/motorola.c
+++ b/arch/m68k/mm/motorola.c
@@ -296,7 +296,7 @@ void __init paging_init(void)
 #endif
 	for (i = 0; i < m68k_num_memory; i++) {
 		zones_size[ZONE_DMA] = m68k_memory[i].size >> PAGE_SHIFT;
-		free_area_init_node(i, pg_data_map + i, zones_size,
+		free_area_init_node(i, zones_size,
 				    m68k_memory[i].addr >> PAGE_SHIFT, NULL);
 	}
 }
diff --git a/arch/m68k/mm/sun3mmu.c b/arch/m68k/mm/sun3mmu.c
index edceefc18870..1b902dbd4376 100644
--- a/arch/m68k/mm/sun3mmu.c
+++ b/arch/m68k/mm/sun3mmu.c
@@ -94,7 +94,7 @@ void __init paging_init(void)
 
 	/* I really wish I knew why the following change made things better...  -- Sam */
 /*	free_area_init(zones_size); */
-	free_area_init_node(0, NODE_DATA(0), zones_size,
+	free_area_init_node(0, zones_size,
 			    (__pa(PAGE_OFFSET) >> PAGE_SHIFT) + 1, NULL);
 
 
diff --git a/arch/parisc/mm/init.c b/arch/parisc/mm/init.c
index 0ddf4904640a..7c155c254e72 100644
--- a/arch/parisc/mm/init.c
+++ b/arch/parisc/mm/init.c
@@ -887,7 +887,7 @@ void __init paging_init(void)
 		}
 #endif
 
-		free_area_init_node(i, NODE_DATA(i), zones_size,
+		free_area_init_node(i, zones_size,
 				pmem_ranges[i].start_pfn, NULL);
 	}
 }
diff --git a/arch/sparc/mm/srmmu.c b/arch/sparc/mm/srmmu.c
index c624e04ff03e..ee30462598fc 100644
--- a/arch/sparc/mm/srmmu.c
+++ b/arch/sparc/mm/srmmu.c
@@ -1352,8 +1352,7 @@ void __init srmmu_paging_init(void)
 		zones_size[ZONE_HIGHMEM] = npages;
 		zholes_size[ZONE_HIGHMEM] = npages - calc_highpages();
 
-		free_area_init_node(0, &contig_page_data, zones_size,
-				    pfn_base, zholes_size);
+		free_area_init_node(0, zones_size, pfn_base, zholes_size);
 	}
 }
 
diff --git a/arch/sparc/mm/sun4c.c b/arch/sparc/mm/sun4c.c
index 2375fe9dc312..d1782f6368be 100644
--- a/arch/sparc/mm/sun4c.c
+++ b/arch/sparc/mm/sun4c.c
@@ -2123,8 +2123,7 @@ void __init sun4c_paging_init(void)
 		zones_size[ZONE_HIGHMEM] = npages;
 		zholes_size[ZONE_HIGHMEM] = npages - calc_highpages();
 
-		free_area_init_node(0, &contig_page_data, zones_size,
-				    pfn_base, zholes_size);
+		free_area_init_node(0, zones_size, pfn_base, zholes_size);
 	}
 
 	cnt = 0;
diff --git a/arch/v850/kernel/setup.c b/arch/v850/kernel/setup.c
index a0a8456a8430..10335cecf7bd 100644
--- a/arch/v850/kernel/setup.c
+++ b/arch/v850/kernel/setup.c
@@ -295,8 +295,7 @@ init_mem_alloc (unsigned long ram_start, unsigned long ram_len)
 #error MAX_ORDER is too large for given PAGE_OFFSET (use CONFIG_FORCE_MAX_ZONEORDER to change it)
 #endif
 	NODE_DATA(0)->node_mem_map = NULL;
-	free_area_init_node (0, NODE_DATA(0), zones_size,
-			     ADDR_TO_PAGE (PAGE_OFFSET), 0);
+	free_area_init_node(0, zones_size, ADDR_TO_PAGE (PAGE_OFFSET), 0);
 }
 
 
diff --git a/include/linux/mm.h b/include/linux/mm.h
index f8071097302a..196924b657bc 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -962,9 +962,8 @@ static inline void pgtable_page_dtor(struct page *page)
 		NULL: pte_offset_kernel(pmd, address))
 
 extern void free_area_init(unsigned long * zones_size);
-extern void free_area_init_node(int nid, pg_data_t *pgdat,
-	unsigned long * zones_size, unsigned long zone_start_pfn, 
-	unsigned long *zholes_size);
+extern void free_area_init_node(int nid, unsigned long * zones_size,
+		unsigned long zone_start_pfn, unsigned long *zholes_size);
 #ifdef CONFIG_ARCH_POPULATES_NODE_MAP
 /*
  * With CONFIG_ARCH_POPULATES_NODE_MAP set, an architecture may initialise its
diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index 833f854eabe5..6e26adc08f14 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -455,7 +455,7 @@ static pg_data_t *hotadd_new_pgdat(int nid, u64 start)
 	/* we can use NODE_DATA(nid) from here */
 
 	/* init node's zones as empty zones, we don't have any present pages.*/
-	free_area_init_node(nid, pgdat, zones_size, start_pfn, zholes_size);
+	free_area_init_node(nid, zones_size, start_pfn, zholes_size);
 
 	return pgdat;
 }
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 24aa3d1b9d96..e43aae135b38 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -3461,10 +3461,11 @@ static void __init_refok alloc_node_mem_map(struct pglist_data *pgdat)
 #endif /* CONFIG_FLAT_NODE_MEM_MAP */
 }
 
-void __paginginit free_area_init_node(int nid, struct pglist_data *pgdat,
-		unsigned long *zones_size, unsigned long node_start_pfn,
-		unsigned long *zholes_size)
+void __paginginit free_area_init_node(int nid, unsigned long *zones_size,
+		unsigned long node_start_pfn, unsigned long *zholes_size)
 {
+	pg_data_t *pgdat = NODE_DATA(nid);
+
 	pgdat->node_id = nid;
 	pgdat->node_start_pfn = node_start_pfn;
 	calculate_node_totalpages(pgdat, zones_size, zholes_size);
@@ -3961,7 +3962,7 @@ void __init free_area_init_nodes(unsigned long *max_zone_pfn)
 	setup_nr_node_ids();
 	for_each_online_node(nid) {
 		pg_data_t *pgdat = NODE_DATA(nid);
-		free_area_init_node(nid, pgdat, NULL,
+		free_area_init_node(nid, NULL,
 				find_min_pfn_for_node(nid), NULL);
 
 		/* Any memory on that node */
@@ -4032,7 +4033,7 @@ EXPORT_SYMBOL(contig_page_data);
 
 void __init free_area_init(unsigned long *zones_size)
 {
-	free_area_init_node(0, NODE_DATA(0), zones_size,
+	free_area_init_node(0, zones_size,
 			__pa(PAGE_OFFSET) >> PAGE_SHIFT, NULL);
 }
 

From fc1b8a73dd71226902a11928dd5500326e101df9 Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Wed, 23 Jul 2008 21:27:22 -0700
Subject: [PATCH 0234/1171] hugetlb: move hugetlb_acct_memory()

This is a patchset to give reliable behaviour to a process that
successfully calls mmap(MAP_PRIVATE) on a hugetlbfs file.  Currently, it
is possible for the process to be killed due to a small hugepage pool size
even if it calls mlock().

MAP_SHARED mappings on hugetlbfs reserve huge pages at mmap() time.  This
guarantees all future faults against the mapping will succeed.  This
allows local allocations at first use improving NUMA locality whilst
retaining reliability.

MAP_PRIVATE mappings do not reserve pages.  This can result in an
application being SIGKILLed later if a huge page is not available at fault
time.  This makes huge pages usage very ill-advised in some cases as the
unexpected application failure cannot be detected and handled as it is
immediately fatal.  Although an application may force instantiation of the
pages using mlock(), this may lead to poor memory placement and the
process may still be killed when performing COW.

This patchset introduces a reliability guarantee for the process which
creates a private mapping, i.e.  the process that calls mmap() on a
hugetlbfs file successfully.  The first patch of the set is purely
mechanical code move to make later diffs easier to read.  The second patch
will guarantee faults up until the process calls fork().  After patch two,
as long as the child keeps the mappings, the parent is no longer
guaranteed to be reliable.  Patch 3 guarantees that the parent will always
successfully COW by unmapping the pages from the child in the event there
are insufficient pages in the hugepage pool in allocate a new page, be it
via a static or dynamic pool.

Existing hugepage-aware applications are unlikely to be affected by this
change.  For much of hugetlbfs's history, pages were pre-faulted at mmap()
time or mmap() failed which acts in a reserve-like manner.  If the pool is
sized correctly already so that parent and child can fault reliably, the
application will not even notice the reserves.  It's only when the pool is
too small for the application to function perfectly reliably that the
reserves come into play.

Credit goes to Andy Whitcroft for cleaning up a number of mistakes during
review before the patches were released.

This patch:

A later patch in this set needs to call hugetlb_acct_memory() before it is
defined.  This patch moves the function without modification.  This makes
later diffs easier to read.

Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Acked-by: Adam Litke <agl@us.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/hugetlb.c | 82 ++++++++++++++++++++++++++--------------------------
 1 file changed, 41 insertions(+), 41 deletions(-)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 2c5c9ee4220d..a4dbba8965f3 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -716,6 +716,47 @@ unsigned long hugetlb_total_pages(void)
 	return nr_huge_pages * (HPAGE_SIZE / PAGE_SIZE);
 }
 
+static int hugetlb_acct_memory(long delta)
+{
+	int ret = -ENOMEM;
+
+	spin_lock(&hugetlb_lock);
+	/*
+	 * When cpuset is configured, it breaks the strict hugetlb page
+	 * reservation as the accounting is done on a global variable. Such
+	 * reservation is completely rubbish in the presence of cpuset because
+	 * the reservation is not checked against page availability for the
+	 * current cpuset. Application can still potentially OOM'ed by kernel
+	 * with lack of free htlb page in cpuset that the task is in.
+	 * Attempt to enforce strict accounting with cpuset is almost
+	 * impossible (or too ugly) because cpuset is too fluid that
+	 * task or memory node can be dynamically moved between cpusets.
+	 *
+	 * The change of semantics for shared hugetlb mapping with cpuset is
+	 * undesirable. However, in order to preserve some of the semantics,
+	 * we fall back to check against current free page availability as
+	 * a best attempt and hopefully to minimize the impact of changing
+	 * semantics that cpuset has.
+	 */
+	if (delta > 0) {
+		if (gather_surplus_pages(delta) < 0)
+			goto out;
+
+		if (delta > cpuset_mems_nr(free_huge_pages_node)) {
+			return_unused_surplus_pages(delta);
+			goto out;
+		}
+	}
+
+	ret = 0;
+	if (delta < 0)
+		return_unused_surplus_pages((unsigned long) -delta);
+
+out:
+	spin_unlock(&hugetlb_lock);
+	return ret;
+}
+
 /*
  * We cannot handle pagefaults against hugetlb pages at all.  They cause
  * handle_mm_fault() to try to instantiate regular-sized pages in the
@@ -1248,47 +1289,6 @@ static long region_truncate(struct list_head *head, long end)
 	return chg;
 }
 
-static int hugetlb_acct_memory(long delta)
-{
-	int ret = -ENOMEM;
-
-	spin_lock(&hugetlb_lock);
-	/*
-	 * When cpuset is configured, it breaks the strict hugetlb page
-	 * reservation as the accounting is done on a global variable. Such
-	 * reservation is completely rubbish in the presence of cpuset because
-	 * the reservation is not checked against page availability for the
-	 * current cpuset. Application can still potentially OOM'ed by kernel
-	 * with lack of free htlb page in cpuset that the task is in.
-	 * Attempt to enforce strict accounting with cpuset is almost
-	 * impossible (or too ugly) because cpuset is too fluid that
-	 * task or memory node can be dynamically moved between cpusets.
-	 *
-	 * The change of semantics for shared hugetlb mapping with cpuset is
-	 * undesirable. However, in order to preserve some of the semantics,
-	 * we fall back to check against current free page availability as
-	 * a best attempt and hopefully to minimize the impact of changing
-	 * semantics that cpuset has.
-	 */
-	if (delta > 0) {
-		if (gather_surplus_pages(delta) < 0)
-			goto out;
-
-		if (delta > cpuset_mems_nr(free_huge_pages_node)) {
-			return_unused_surplus_pages(delta);
-			goto out;
-		}
-	}
-
-	ret = 0;
-	if (delta < 0)
-		return_unused_surplus_pages((unsigned long) -delta);
-
-out:
-	spin_unlock(&hugetlb_lock);
-	return ret;
-}
-
 int hugetlb_reserve_pages(struct inode *inode, long from, long to)
 {
 	long ret, chg;

From a1e78772d72b2616ed20e54896e68e0e7044854e Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Wed, 23 Jul 2008 21:27:23 -0700
Subject: [PATCH 0235/1171] hugetlb: reserve huge pages for reliable
 MAP_PRIVATE hugetlbfs mappings until fork()

This patch reserves huge pages at mmap() time for MAP_PRIVATE mappings in
a similar manner to the reservations taken for MAP_SHARED mappings.  The
reserve count is accounted both globally and on a per-VMA basis for
private mappings.  This guarantees that a process that successfully calls
mmap() will successfully fault all pages in the future unless fork() is
called.

The characteristics of private mappings of hugetlbfs files behaviour after
this patch are;

1. The process calling mmap() is guaranteed to succeed all future faults until
   it forks().
2. On fork(), the parent may die due to SIGKILL on writes to the private
   mapping if enough pages are not available for the COW. For reasonably
   reliable behaviour in the face of a small huge page pool, children of
   hugepage-aware processes should not reference the mappings; such as
   might occur when fork()ing to exec().
3. On fork(), the child VMAs inherit no reserves. Reads on pages already
   faulted by the parent will succeed. Successful writes will depend on enough
   huge pages being free in the pool.
4. Quotas of the hugetlbfs mount are checked at reserve time for the mapper
   and at fault time otherwise.

Before this patch, all reads or writes in the child potentially needs page
allocations that can later lead to the death of the parent.  This applies
to reads and writes of uninstantiated pages as well as COW.  After the
patch it is only a write to an instantiated page that causes problems.

Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Acked-by: Adam Litke <agl@us.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/hugetlbfs/inode.c    |   8 +-
 include/linux/hugetlb.h |   9 ++-
 kernel/fork.c           |   9 +++
 mm/hugetlb.c            | 174 +++++++++++++++++++++++++++++-----------
 4 files changed, 148 insertions(+), 52 deletions(-)

diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index aeabf80f81a5..1576bbecd084 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -103,9 +103,9 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma)
 	ret = -ENOMEM;
 	len = vma_len + ((loff_t)vma->vm_pgoff << PAGE_SHIFT);
 
-	if (vma->vm_flags & VM_MAYSHARE &&
-	    hugetlb_reserve_pages(inode, vma->vm_pgoff >> (HPAGE_SHIFT-PAGE_SHIFT),
-				  len >> HPAGE_SHIFT))
+	if (hugetlb_reserve_pages(inode,
+				vma->vm_pgoff >> (HPAGE_SHIFT-PAGE_SHIFT),
+				len >> HPAGE_SHIFT, vma))
 		goto out;
 
 	ret = 0;
@@ -942,7 +942,7 @@ struct file *hugetlb_file_setup(const char *name, size_t size)
 		goto out_dentry;
 
 	error = -ENOMEM;
-	if (hugetlb_reserve_pages(inode, 0, size >> HPAGE_SHIFT))
+	if (hugetlb_reserve_pages(inode, 0, size >> HPAGE_SHIFT, NULL))
 		goto out_inode;
 
 	d_instantiate(dentry, inode);
diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index a79e80b689d8..185b14c9f021 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -17,6 +17,7 @@ static inline int is_vm_hugetlb_page(struct vm_area_struct *vma)
 	return vma->vm_flags & VM_HUGETLB;
 }
 
+void reset_vma_resv_huge_pages(struct vm_area_struct *vma);
 int hugetlb_sysctl_handler(struct ctl_table *, int, struct file *, void __user *, size_t *, loff_t *);
 int hugetlb_overcommit_handler(struct ctl_table *, int, struct file *, void __user *, size_t *, loff_t *);
 int hugetlb_treat_movable_handler(struct ctl_table *, int, struct file *, void __user *, size_t *, loff_t *);
@@ -30,7 +31,8 @@ int hugetlb_report_node_meminfo(int, char *);
 unsigned long hugetlb_total_pages(void);
 int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 			unsigned long address, int write_access);
-int hugetlb_reserve_pages(struct inode *inode, long from, long to);
+int hugetlb_reserve_pages(struct inode *inode, long from, long to,
+						struct vm_area_struct *vma);
 void hugetlb_unreserve_pages(struct inode *inode, long offset, long freed);
 
 extern unsigned long max_huge_pages;
@@ -58,6 +60,11 @@ static inline int is_vm_hugetlb_page(struct vm_area_struct *vma)
 {
 	return 0;
 }
+
+static inline void reset_vma_resv_huge_pages(struct vm_area_struct *vma)
+{
+}
+
 static inline unsigned long hugetlb_total_pages(void)
 {
 	return 0;
diff --git a/kernel/fork.c b/kernel/fork.c
index adefc1131f27..552c8d8e77ad 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -33,6 +33,7 @@
 #include <linux/cpu.h>
 #include <linux/cgroup.h>
 #include <linux/security.h>
+#include <linux/hugetlb.h>
 #include <linux/swap.h>
 #include <linux/syscalls.h>
 #include <linux/jiffies.h>
@@ -306,6 +307,14 @@ static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm)
 			spin_unlock(&file->f_mapping->i_mmap_lock);
 		}
 
+		/*
+		 * Clear hugetlb-related page reserves for children. This only
+		 * affects MAP_PRIVATE mappings. Faults generated by the child
+		 * are not guaranteed to succeed, even if read-only
+		 */
+		if (is_vm_hugetlb_page(tmp))
+			reset_vma_resv_huge_pages(tmp);
+
 		/*
 		 * Link in the new vma and copy the page table entries.
 		 */
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index a4dbba8965f3..0af500db3632 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -40,6 +40,69 @@ static int hugetlb_next_nid;
  */
 static DEFINE_SPINLOCK(hugetlb_lock);
 
+/*
+ * These helpers are used to track how many pages are reserved for
+ * faults in a MAP_PRIVATE mapping. Only the process that called mmap()
+ * is guaranteed to have their future faults succeed.
+ *
+ * With the exception of reset_vma_resv_huge_pages() which is called at fork(),
+ * the reserve counters are updated with the hugetlb_lock held. It is safe
+ * to reset the VMA at fork() time as it is not in use yet and there is no
+ * chance of the global counters getting corrupted as a result of the values.
+ */
+static unsigned long vma_resv_huge_pages(struct vm_area_struct *vma)
+{
+	VM_BUG_ON(!is_vm_hugetlb_page(vma));
+	if (!(vma->vm_flags & VM_SHARED))
+		return (unsigned long)vma->vm_private_data;
+	return 0;
+}
+
+static void set_vma_resv_huge_pages(struct vm_area_struct *vma,
+							unsigned long reserve)
+{
+	VM_BUG_ON(!is_vm_hugetlb_page(vma));
+	VM_BUG_ON(vma->vm_flags & VM_SHARED);
+
+	vma->vm_private_data = (void *)reserve;
+}
+
+/* Decrement the reserved pages in the hugepage pool by one */
+static void decrement_hugepage_resv_vma(struct vm_area_struct *vma)
+{
+	if (vma->vm_flags & VM_SHARED) {
+		/* Shared mappings always use reserves */
+		resv_huge_pages--;
+	} else {
+		/*
+		 * Only the process that called mmap() has reserves for
+		 * private mappings.
+		 */
+		if (vma_resv_huge_pages(vma)) {
+			resv_huge_pages--;
+			reserve = (unsigned long)vma->vm_private_data - 1;
+			vma->vm_private_data = (void *)reserve;
+		}
+	}
+}
+
+void reset_vma_resv_huge_pages(struct vm_area_struct *vma)
+{
+	VM_BUG_ON(!is_vm_hugetlb_page(vma));
+	if (!(vma->vm_flags & VM_SHARED))
+		vma->vm_private_data = (void *)0;
+}
+
+/* Returns true if the VMA has associated reserve pages */
+static int vma_has_private_reserves(struct vm_area_struct *vma)
+{
+	if (vma->vm_flags & VM_SHARED)
+		return 0;
+	if (!vma_resv_huge_pages(vma))
+		return 0;
+	return 1;
+}
+
 static void clear_huge_page(struct page *page, unsigned long addr)
 {
 	int i;
@@ -101,6 +164,15 @@ static struct page *dequeue_huge_page_vma(struct vm_area_struct *vma,
 	struct zone *zone;
 	struct zoneref *z;
 
+	/*
+	 * A child process with MAP_PRIVATE mappings created by their parent
+	 * have no page reserves. This check ensures that reservations are
+	 * not "stolen". The child may still get SIGKILLed
+	 */
+	if (!vma_has_private_reserves(vma) &&
+			free_huge_pages - resv_huge_pages == 0)
+		return NULL;
+
 	for_each_zone_zonelist_nodemask(zone, z, zonelist,
 						MAX_NR_ZONES - 1, nodemask) {
 		nid = zone_to_nid(zone);
@@ -111,8 +183,8 @@ static struct page *dequeue_huge_page_vma(struct vm_area_struct *vma,
 			list_del(&page->lru);
 			free_huge_pages--;
 			free_huge_pages_node[nid]--;
-			if (vma && vma->vm_flags & VM_MAYSHARE)
-				resv_huge_pages--;
+			decrement_hugepage_resv_vma(vma);
+
 			break;
 		}
 	}
@@ -461,55 +533,40 @@ static void return_unused_surplus_pages(unsigned long unused_resv_pages)
 	}
 }
 
-
-static struct page *alloc_huge_page_shared(struct vm_area_struct *vma,
-						unsigned long addr)
-{
-	struct page *page;
-
-	spin_lock(&hugetlb_lock);
-	page = dequeue_huge_page_vma(vma, addr);
-	spin_unlock(&hugetlb_lock);
-	return page ? page : ERR_PTR(-VM_FAULT_OOM);
-}
-
-static struct page *alloc_huge_page_private(struct vm_area_struct *vma,
-						unsigned long addr)
-{
-	struct page *page = NULL;
-
-	if (hugetlb_get_quota(vma->vm_file->f_mapping, 1))
-		return ERR_PTR(-VM_FAULT_SIGBUS);
-
-	spin_lock(&hugetlb_lock);
-	if (free_huge_pages > resv_huge_pages)
-		page = dequeue_huge_page_vma(vma, addr);
-	spin_unlock(&hugetlb_lock);
-	if (!page) {
-		page = alloc_buddy_huge_page(vma, addr);
-		if (!page) {
-			hugetlb_put_quota(vma->vm_file->f_mapping, 1);
-			return ERR_PTR(-VM_FAULT_OOM);
-		}
-	}
-	return page;
-}
-
 static struct page *alloc_huge_page(struct vm_area_struct *vma,
 				    unsigned long addr)
 {
 	struct page *page;
 	struct address_space *mapping = vma->vm_file->f_mapping;
+	struct inode *inode = mapping->host;
+	unsigned int chg = 0;
 
-	if (vma->vm_flags & VM_MAYSHARE)
-		page = alloc_huge_page_shared(vma, addr);
-	else
-		page = alloc_huge_page_private(vma, addr);
-
-	if (!IS_ERR(page)) {
-		set_page_refcounted(page);
-		set_page_private(page, (unsigned long) mapping);
+	/*
+	 * Processes that did not create the mapping will have no reserves and
+	 * will not have accounted against quota. Check that the quota can be
+	 * made before satisfying the allocation
+	 */
+	if (!vma_has_private_reserves(vma)) {
+		chg = 1;
+		if (hugetlb_get_quota(inode->i_mapping, chg))
+			return ERR_PTR(-ENOSPC);
 	}
+
+	spin_lock(&hugetlb_lock);
+	page = dequeue_huge_page_vma(vma, addr);
+	spin_unlock(&hugetlb_lock);
+
+	if (!page) {
+		page = alloc_buddy_huge_page(vma, addr);
+		if (!page) {
+			hugetlb_put_quota(inode->i_mapping, chg);
+			return ERR_PTR(-VM_FAULT_OOM);
+		}
+	}
+
+	set_page_refcounted(page);
+	set_page_private(page, (unsigned long) mapping);
+
 	return page;
 }
 
@@ -757,6 +814,13 @@ out:
 	return ret;
 }
 
+static void hugetlb_vm_op_close(struct vm_area_struct *vma)
+{
+	unsigned long reserve = vma_resv_huge_pages(vma);
+	if (reserve)
+		hugetlb_acct_memory(-reserve);
+}
+
 /*
  * We cannot handle pagefaults against hugetlb pages at all.  They cause
  * handle_mm_fault() to try to instantiate regular-sized pages in the
@@ -771,6 +835,7 @@ static int hugetlb_vm_op_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 
 struct vm_operations_struct hugetlb_vm_ops = {
 	.fault = hugetlb_vm_op_fault,
+	.close = hugetlb_vm_op_close,
 };
 
 static pte_t make_huge_pte(struct vm_area_struct *vma, struct page *page,
@@ -1289,11 +1354,25 @@ static long region_truncate(struct list_head *head, long end)
 	return chg;
 }
 
-int hugetlb_reserve_pages(struct inode *inode, long from, long to)
+int hugetlb_reserve_pages(struct inode *inode,
+					long from, long to,
+					struct vm_area_struct *vma)
 {
 	long ret, chg;
 
-	chg = region_chg(&inode->i_mapping->private_list, from, to);
+	/*
+	 * Shared mappings base their reservation on the number of pages that
+	 * are already allocated on behalf of the file. Private mappings need
+	 * to reserve the full area even if read-only as mprotect() may be
+	 * called to make the mapping read-write. Assume !vma is a shm mapping
+	 */
+	if (!vma || vma->vm_flags & VM_SHARED)
+		chg = region_chg(&inode->i_mapping->private_list, from, to);
+	else {
+		chg = to - from;
+		set_vma_resv_huge_pages(vma, chg);
+	}
+
 	if (chg < 0)
 		return chg;
 
@@ -1304,7 +1383,8 @@ int hugetlb_reserve_pages(struct inode *inode, long from, long to)
 		hugetlb_put_quota(inode->i_mapping, chg);
 		return ret;
 	}
-	region_add(&inode->i_mapping->private_list, from, to);
+	if (!vma || vma->vm_flags & VM_SHARED)
+		region_add(&inode->i_mapping->private_list, from, to);
 	return 0;
 }
 

From 04f2cbe35699d22dbf428373682ead85ca1240f5 Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Wed, 23 Jul 2008 21:27:25 -0700
Subject: [PATCH 0236/1171] hugetlb: guarantee that COW faults for a process
 that called mmap(MAP_PRIVATE) on hugetlbfs will succeed

After patch 2 in this series, a process that successfully calls mmap() for
a MAP_PRIVATE mapping will be guaranteed to successfully fault until a
process calls fork().  At that point, the next write fault from the parent
could fail due to COW if the child still has a reference.

We only reserve pages for the parent but a copy must be made to avoid
leaking data from the parent to the child after fork().  Reserves could be
taken for both parent and child at fork time to guarantee faults but if
the mapping is large it is highly likely we will not have sufficient pages
for the reservation, and it is common to fork only to exec() immediatly
after.  A failure here would be very undesirable.

Note that the current behaviour of mainline with MAP_PRIVATE pages is
pretty bad.  The following situation is allowed to occur today.

1. Process calls mmap(MAP_PRIVATE)
2. Process calls mlock() to fault all pages and makes sure it succeeds
3. Process forks()
4. Process writes to MAP_PRIVATE mapping while child still exists
5. If the COW fails at this point, the process gets SIGKILLed even though it
   had taken care to ensure the pages existed

This patch improves the situation by guaranteeing the reliability of the
process that successfully calls mmap().  When the parent performs COW, it
will try to satisfy the allocation without using reserves.  If that fails
the parent will steal the page leaving any children without a page.
Faults from the child after that point will result in failure.  If the
child COW happens first, an attempt will be made to allocate the page
without reserves and the child will get SIGKILLed on failure.

To summarise the new behaviour:

1. If the original mapper performs COW on a private mapping with multiple
   references, it will attempt to allocate a hugepage from the pool or
   the buddy allocator without using the existing reserves. On fail, VMAs
   mapping the same area are traversed and the page being COW'd is unmapped
   where found. It will then steal the original page as the last mapper in
   the normal way.

2. The VMAs the pages were unmapped from are flagged to note that pages
   with data no longer exist. Future no-page faults on those VMAs will
   terminate the process as otherwise it would appear that data was corrupted.
   A warning is printed to the console that this situation occured.

2. If the child performs COW first, it will attempt to satisfy the COW
   from the pool if there are enough pages or via the buddy allocator if
   overcommit is allowed and the buddy allocator can satisfy the request. If
   it fails, the child will be killed.

If the pool is large enough, existing applications will not notice that
the reserves were a factor.  Existing applications depending on the
no-reserves been set are unlikely to exist as for much of the history of
hugetlbfs, pages were prefaulted at mmap(), allocating the pages at that
point or failing the mmap().

[npiggin@suse.de: fix CONFIG_HUGETLB=n build]
Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Acked-by: Adam Litke <agl@us.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/hugetlbfs/inode.c    |   2 +-
 include/linux/hugetlb.h |   8 +-
 mm/hugetlb.c            | 201 ++++++++++++++++++++++++++++++++++++----
 mm/memory.c             |   2 +-
 4 files changed, 190 insertions(+), 23 deletions(-)

diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index 1576bbecd084..428eff5b73f3 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -441,7 +441,7 @@ hugetlb_vmtruncate_list(struct prio_tree_root *root, pgoff_t pgoff)
 			v_offset = 0;
 
 		__unmap_hugepage_range(vma,
-				vma->vm_start + v_offset, vma->vm_end);
+				vma->vm_start + v_offset, vma->vm_end, NULL);
 	}
 }
 
diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index 185b14c9f021..abbc187193a1 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -23,8 +23,10 @@ int hugetlb_overcommit_handler(struct ctl_table *, int, struct file *, void __us
 int hugetlb_treat_movable_handler(struct ctl_table *, int, struct file *, void __user *, size_t *, loff_t *);
 int copy_hugetlb_page_range(struct mm_struct *, struct mm_struct *, struct vm_area_struct *);
 int follow_hugetlb_page(struct mm_struct *, struct vm_area_struct *, struct page **, struct vm_area_struct **, unsigned long *, int *, int, int);
-void unmap_hugepage_range(struct vm_area_struct *, unsigned long, unsigned long);
-void __unmap_hugepage_range(struct vm_area_struct *, unsigned long, unsigned long);
+void unmap_hugepage_range(struct vm_area_struct *,
+			unsigned long, unsigned long, struct page *);
+void __unmap_hugepage_range(struct vm_area_struct *,
+			unsigned long, unsigned long, struct page *);
 int hugetlb_prefault(struct address_space *, struct vm_area_struct *);
 int hugetlb_report_meminfo(char *);
 int hugetlb_report_node_meminfo(int, char *);
@@ -74,7 +76,7 @@ static inline unsigned long hugetlb_total_pages(void)
 #define follow_huge_addr(mm, addr, write)	ERR_PTR(-EINVAL)
 #define copy_hugetlb_page_range(src, dst, vma)	({ BUG(); 0; })
 #define hugetlb_prefault(mapping, vma)		({ BUG(); 0; })
-#define unmap_hugepage_range(vma, start, end)	BUG()
+#define unmap_hugepage_range(vma, start, end, page)	BUG()
 #define hugetlb_report_meminfo(buf)		0
 #define hugetlb_report_node_meminfo(n, buf)	0
 #define follow_huge_pmd(mm, addr, pmd, write)	NULL
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 0af500db3632..a2d29b84501f 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -40,6 +40,9 @@ static int hugetlb_next_nid;
  */
 static DEFINE_SPINLOCK(hugetlb_lock);
 
+#define HPAGE_RESV_OWNER    (1UL << (BITS_PER_LONG - 1))
+#define HPAGE_RESV_UNMAPPED (1UL << (BITS_PER_LONG - 2))
+#define HPAGE_RESV_MASK (HPAGE_RESV_OWNER | HPAGE_RESV_UNMAPPED)
 /*
  * These helpers are used to track how many pages are reserved for
  * faults in a MAP_PRIVATE mapping. Only the process that called mmap()
@@ -54,17 +57,32 @@ static unsigned long vma_resv_huge_pages(struct vm_area_struct *vma)
 {
 	VM_BUG_ON(!is_vm_hugetlb_page(vma));
 	if (!(vma->vm_flags & VM_SHARED))
-		return (unsigned long)vma->vm_private_data;
+		return (unsigned long)vma->vm_private_data & ~HPAGE_RESV_MASK;
 	return 0;
 }
 
 static void set_vma_resv_huge_pages(struct vm_area_struct *vma,
 							unsigned long reserve)
 {
+	unsigned long flags;
 	VM_BUG_ON(!is_vm_hugetlb_page(vma));
 	VM_BUG_ON(vma->vm_flags & VM_SHARED);
 
-	vma->vm_private_data = (void *)reserve;
+	flags = (unsigned long)vma->vm_private_data & HPAGE_RESV_MASK;
+	vma->vm_private_data = (void *)(reserve | flags);
+}
+
+static void set_vma_resv_flags(struct vm_area_struct *vma, unsigned long flags)
+{
+	unsigned long reserveflags = (unsigned long)vma->vm_private_data;
+	VM_BUG_ON(!is_vm_hugetlb_page(vma));
+	vma->vm_private_data = (void *)(reserveflags | flags);
+}
+
+static int is_vma_resv_set(struct vm_area_struct *vma, unsigned long flag)
+{
+	VM_BUG_ON(!is_vm_hugetlb_page(vma));
+	return ((unsigned long)vma->vm_private_data & flag) != 0;
 }
 
 /* Decrement the reserved pages in the hugepage pool by one */
@@ -78,14 +96,18 @@ static void decrement_hugepage_resv_vma(struct vm_area_struct *vma)
 		 * Only the process that called mmap() has reserves for
 		 * private mappings.
 		 */
-		if (vma_resv_huge_pages(vma)) {
+		if (is_vma_resv_set(vma, HPAGE_RESV_OWNER)) {
+			unsigned long flags, reserve;
 			resv_huge_pages--;
+			flags = (unsigned long)vma->vm_private_data &
+							HPAGE_RESV_MASK;
 			reserve = (unsigned long)vma->vm_private_data - 1;
-			vma->vm_private_data = (void *)reserve;
+			vma->vm_private_data = (void *)(reserve | flags);
 		}
 	}
 }
 
+/* Reset counters to 0 and clear all HPAGE_RESV_* flags */
 void reset_vma_resv_huge_pages(struct vm_area_struct *vma)
 {
 	VM_BUG_ON(!is_vm_hugetlb_page(vma));
@@ -153,7 +175,7 @@ static struct page *dequeue_huge_page(void)
 }
 
 static struct page *dequeue_huge_page_vma(struct vm_area_struct *vma,
-				unsigned long address)
+				unsigned long address, int avoid_reserve)
 {
 	int nid;
 	struct page *page = NULL;
@@ -173,6 +195,10 @@ static struct page *dequeue_huge_page_vma(struct vm_area_struct *vma,
 			free_huge_pages - resv_huge_pages == 0)
 		return NULL;
 
+	/* If reserves cannot be used, ensure enough pages are in the pool */
+	if (avoid_reserve && free_huge_pages - resv_huge_pages == 0)
+		return NULL;
+
 	for_each_zone_zonelist_nodemask(zone, z, zonelist,
 						MAX_NR_ZONES - 1, nodemask) {
 		nid = zone_to_nid(zone);
@@ -183,7 +209,9 @@ static struct page *dequeue_huge_page_vma(struct vm_area_struct *vma,
 			list_del(&page->lru);
 			free_huge_pages--;
 			free_huge_pages_node[nid]--;
-			decrement_hugepage_resv_vma(vma);
+
+			if (!avoid_reserve)
+				decrement_hugepage_resv_vma(vma);
 
 			break;
 		}
@@ -534,7 +562,7 @@ static void return_unused_surplus_pages(unsigned long unused_resv_pages)
 }
 
 static struct page *alloc_huge_page(struct vm_area_struct *vma,
-				    unsigned long addr)
+				    unsigned long addr, int avoid_reserve)
 {
 	struct page *page;
 	struct address_space *mapping = vma->vm_file->f_mapping;
@@ -546,14 +574,15 @@ static struct page *alloc_huge_page(struct vm_area_struct *vma,
 	 * will not have accounted against quota. Check that the quota can be
 	 * made before satisfying the allocation
 	 */
-	if (!vma_has_private_reserves(vma)) {
+	if (!(vma->vm_flags & VM_SHARED) &&
+			!is_vma_resv_set(vma, HPAGE_RESV_OWNER)) {
 		chg = 1;
 		if (hugetlb_get_quota(inode->i_mapping, chg))
 			return ERR_PTR(-ENOSPC);
 	}
 
 	spin_lock(&hugetlb_lock);
-	page = dequeue_huge_page_vma(vma, addr);
+	page = dequeue_huge_page_vma(vma, addr, avoid_reserve);
 	spin_unlock(&hugetlb_lock);
 
 	if (!page) {
@@ -909,7 +938,7 @@ nomem:
 }
 
 void __unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start,
-			    unsigned long end)
+			    unsigned long end, struct page *ref_page)
 {
 	struct mm_struct *mm = vma->vm_mm;
 	unsigned long address;
@@ -937,6 +966,27 @@ void __unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start,
 		if (huge_pmd_unshare(mm, &address, ptep))
 			continue;
 
+		/*
+		 * If a reference page is supplied, it is because a specific
+		 * page is being unmapped, not a range. Ensure the page we
+		 * are about to unmap is the actual page of interest.
+		 */
+		if (ref_page) {
+			pte = huge_ptep_get(ptep);
+			if (huge_pte_none(pte))
+				continue;
+			page = pte_page(pte);
+			if (page != ref_page)
+				continue;
+
+			/*
+			 * Mark the VMA as having unmapped its page so that
+			 * future faults in this VMA will fail rather than
+			 * looking like data was lost
+			 */
+			set_vma_resv_flags(vma, HPAGE_RESV_UNMAPPED);
+		}
+
 		pte = huge_ptep_get_and_clear(mm, address, ptep);
 		if (huge_pte_none(pte))
 			continue;
@@ -955,7 +1005,7 @@ void __unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start,
 }
 
 void unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start,
-			  unsigned long end)
+			  unsigned long end, struct page *ref_page)
 {
 	/*
 	 * It is undesirable to test vma->vm_file as it should be non-null
@@ -967,19 +1017,68 @@ void unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start,
 	 */
 	if (vma->vm_file) {
 		spin_lock(&vma->vm_file->f_mapping->i_mmap_lock);
-		__unmap_hugepage_range(vma, start, end);
+		__unmap_hugepage_range(vma, start, end, ref_page);
 		spin_unlock(&vma->vm_file->f_mapping->i_mmap_lock);
 	}
 }
 
+/*
+ * This is called when the original mapper is failing to COW a MAP_PRIVATE
+ * mappping it owns the reserve page for. The intention is to unmap the page
+ * from other VMAs and let the children be SIGKILLed if they are faulting the
+ * same region.
+ */
+int unmap_ref_private(struct mm_struct *mm,
+					struct vm_area_struct *vma,
+					struct page *page,
+					unsigned long address)
+{
+	struct vm_area_struct *iter_vma;
+	struct address_space *mapping;
+	struct prio_tree_iter iter;
+	pgoff_t pgoff;
+
+	/*
+	 * vm_pgoff is in PAGE_SIZE units, hence the different calculation
+	 * from page cache lookup which is in HPAGE_SIZE units.
+	 */
+	address = address & huge_page_mask(hstate_vma(vma));
+	pgoff = ((address - vma->vm_start) >> PAGE_SHIFT)
+		+ (vma->vm_pgoff >> PAGE_SHIFT);
+	mapping = (struct address_space *)page_private(page);
+
+	vma_prio_tree_foreach(iter_vma, &iter, &mapping->i_mmap, pgoff, pgoff) {
+		/* Do not unmap the current VMA */
+		if (iter_vma == vma)
+			continue;
+
+		/*
+		 * Unmap the page from other VMAs without their own reserves.
+		 * They get marked to be SIGKILLed if they fault in these
+		 * areas. This is because a future no-page fault on this VMA
+		 * could insert a zeroed page instead of the data existing
+		 * from the time of fork. This would look like data corruption
+		 */
+		if (!is_vma_resv_set(iter_vma, HPAGE_RESV_OWNER))
+			unmap_hugepage_range(iter_vma,
+				address, address + HPAGE_SIZE,
+				page);
+	}
+
+	return 1;
+}
+
 static int hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma,
-			unsigned long address, pte_t *ptep, pte_t pte)
+			unsigned long address, pte_t *ptep, pte_t pte,
+			struct page *pagecache_page)
 {
 	struct page *old_page, *new_page;
 	int avoidcopy;
+	int outside_reserve = 0;
 
 	old_page = pte_page(pte);
 
+retry_avoidcopy:
 	/* If no-one else is actually using this page, avoid the copy
 	 * and just make the page writable */
 	avoidcopy = (page_count(old_page) == 1);
@@ -988,11 +1087,43 @@ static int hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma,
 		return 0;
 	}
 
+	/*
+	 * If the process that created a MAP_PRIVATE mapping is about to
+	 * perform a COW due to a shared page count, attempt to satisfy
+	 * the allocation without using the existing reserves. The pagecache
+	 * page is used to determine if the reserve at this address was
+	 * consumed or not. If reserves were used, a partial faulted mapping
+	 * at the time of fork() could consume its reserves on COW instead
+	 * of the full address range.
+	 */
+	if (!(vma->vm_flags & VM_SHARED) &&
+			is_vma_resv_set(vma, HPAGE_RESV_OWNER) &&
+			old_page != pagecache_page)
+		outside_reserve = 1;
+
 	page_cache_get(old_page);
-	new_page = alloc_huge_page(vma, address);
+	new_page = alloc_huge_page(vma, address, outside_reserve);
 
 	if (IS_ERR(new_page)) {
 		page_cache_release(old_page);
+
+		/*
+		 * If a process owning a MAP_PRIVATE mapping fails to COW,
+		 * it is due to references held by a child and an insufficient
+		 * huge page pool. To guarantee the original mappers
+		 * reliability, unmap the page from child processes. The child
+		 * may get SIGKILLed if it later faults.
+		 */
+		if (outside_reserve) {
+			BUG_ON(huge_pte_none(pte));
+			if (unmap_ref_private(mm, vma, old_page, address)) {
+				BUG_ON(page_count(old_page) != 1);
+				BUG_ON(huge_pte_none(pte));
+				goto retry_avoidcopy;
+			}
+			WARN_ON_ONCE(1);
+		}
+
 		return -PTR_ERR(new_page);
 	}
 
@@ -1015,6 +1146,20 @@ static int hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma,
 	return 0;
 }
 
+/* Return the pagecache page at a given address within a VMA */
+static struct page *hugetlbfs_pagecache_page(struct vm_area_struct *vma,
+			unsigned long address)
+{
+	struct address_space *mapping;
+	unsigned long idx;
+
+	mapping = vma->vm_file->f_mapping;
+	idx = ((address - vma->vm_start) >> HPAGE_SHIFT)
+		+ (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT));
+
+	return find_lock_page(mapping, idx);
+}
+
 static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma,
 			unsigned long address, pte_t *ptep, int write_access)
 {
@@ -1025,6 +1170,18 @@ static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	struct address_space *mapping;
 	pte_t new_pte;
 
+	/*
+	 * Currently, we are forced to kill the process in the event the
+	 * original mapper has unmapped pages from the child due to a failed
+	 * COW. Warn that such a situation has occured as it may not be obvious
+	 */
+	if (is_vma_resv_set(vma, HPAGE_RESV_UNMAPPED)) {
+		printk(KERN_WARNING
+			"PID %d killed due to inadequate hugepage pool\n",
+			current->pid);
+		return ret;
+	}
+
 	mapping = vma->vm_file->f_mapping;
 	idx = ((address - vma->vm_start) >> HPAGE_SHIFT)
 		+ (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT));
@@ -1039,7 +1196,7 @@ retry:
 		size = i_size_read(mapping->host) >> HPAGE_SHIFT;
 		if (idx >= size)
 			goto out;
-		page = alloc_huge_page(vma, address);
+		page = alloc_huge_page(vma, address, 0);
 		if (IS_ERR(page)) {
 			ret = -PTR_ERR(page);
 			goto out;
@@ -1081,7 +1238,7 @@ retry:
 
 	if (write_access && !(vma->vm_flags & VM_SHARED)) {
 		/* Optimization, do the COW without a second fault */
-		ret = hugetlb_cow(mm, vma, address, ptep, new_pte);
+		ret = hugetlb_cow(mm, vma, address, ptep, new_pte, page);
 	}
 
 	spin_unlock(&mm->page_table_lock);
@@ -1126,8 +1283,15 @@ int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 	spin_lock(&mm->page_table_lock);
 	/* Check for a racing update before calling hugetlb_cow */
 	if (likely(pte_same(entry, huge_ptep_get(ptep))))
-		if (write_access && !pte_write(entry))
-			ret = hugetlb_cow(mm, vma, address, ptep, entry);
+		if (write_access && !pte_write(entry)) {
+			struct page *page;
+			page = hugetlbfs_pagecache_page(vma, address);
+			ret = hugetlb_cow(mm, vma, address, ptep, entry, page);
+			if (page) {
+				unlock_page(page);
+				put_page(page);
+			}
+		}
 	spin_unlock(&mm->page_table_lock);
 	mutex_unlock(&hugetlb_instantiation_mutex);
 
@@ -1371,6 +1535,7 @@ int hugetlb_reserve_pages(struct inode *inode,
 	else {
 		chg = to - from;
 		set_vma_resv_huge_pages(vma, chg);
+		set_vma_resv_flags(vma, HPAGE_RESV_OWNER);
 	}
 
 	if (chg < 0)
diff --git a/mm/memory.c b/mm/memory.c
index 82f3f1c5cf17..72932489a082 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -901,7 +901,7 @@ unsigned long unmap_vmas(struct mmu_gather **tlbp,
 			}
 
 			if (unlikely(is_vm_hugetlb_page(vma))) {
-				unmap_hugepage_range(vma, start, end);
+				unmap_hugepage_range(vma, start, end, NULL);
 				zap_work -= (end - start) /
 						(HPAGE_SIZE / PAGE_SIZE);
 				start = end;

From e7c4b0bfd025f71cf7624b7c1be174f63caade33 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:27:26 -0700
Subject: [PATCH 0237/1171] huge page private reservation review cleanups

Create some new accessors for vma private data to cut down on and contain
the casts.  Encapsulates the huge and small page offset calculations.
Also adds a couple of VM_BUG_ONs for consistency.

[akpm@linux-foundation.org: Make things static]
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Cc: Adam Litke <agl@us.ibm.com>
Cc: Johannes Weiner <hannes@saeurebad.de>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/hugetlb.c | 58 ++++++++++++++++++++++++++++++++++++++++------------
 1 file changed, 45 insertions(+), 13 deletions(-)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index a2d29b84501f..3e873f0101fb 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -40,6 +40,28 @@ static int hugetlb_next_nid;
  */
 static DEFINE_SPINLOCK(hugetlb_lock);
 
+/*
+ * Convert the address within this vma to the page offset within
+ * the mapping, in base page units.
+ */
+static pgoff_t vma_page_offset(struct vm_area_struct *vma,
+				unsigned long address)
+{
+	return ((address - vma->vm_start) >> PAGE_SHIFT) +
+					(vma->vm_pgoff >> PAGE_SHIFT);
+}
+
+/*
+ * Convert the address within this vma to the page offset within
+ * the mapping, in pagecache page units; huge pages here.
+ */
+static pgoff_t vma_pagecache_offset(struct vm_area_struct *vma,
+					unsigned long address)
+{
+	return ((address - vma->vm_start) >> HPAGE_SHIFT) +
+			(vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT));
+}
+
 #define HPAGE_RESV_OWNER    (1UL << (BITS_PER_LONG - 1))
 #define HPAGE_RESV_UNMAPPED (1UL << (BITS_PER_LONG - 2))
 #define HPAGE_RESV_MASK (HPAGE_RESV_OWNER | HPAGE_RESV_UNMAPPED)
@@ -53,36 +75,48 @@ static DEFINE_SPINLOCK(hugetlb_lock);
  * to reset the VMA at fork() time as it is not in use yet and there is no
  * chance of the global counters getting corrupted as a result of the values.
  */
+static unsigned long get_vma_private_data(struct vm_area_struct *vma)
+{
+	return (unsigned long)vma->vm_private_data;
+}
+
+static void set_vma_private_data(struct vm_area_struct *vma,
+							unsigned long value)
+{
+	vma->vm_private_data = (void *)value;
+}
+
 static unsigned long vma_resv_huge_pages(struct vm_area_struct *vma)
 {
 	VM_BUG_ON(!is_vm_hugetlb_page(vma));
 	if (!(vma->vm_flags & VM_SHARED))
-		return (unsigned long)vma->vm_private_data & ~HPAGE_RESV_MASK;
+		return get_vma_private_data(vma) & ~HPAGE_RESV_MASK;
 	return 0;
 }
 
 static void set_vma_resv_huge_pages(struct vm_area_struct *vma,
 							unsigned long reserve)
 {
-	unsigned long flags;
 	VM_BUG_ON(!is_vm_hugetlb_page(vma));
 	VM_BUG_ON(vma->vm_flags & VM_SHARED);
 
-	flags = (unsigned long)vma->vm_private_data & HPAGE_RESV_MASK;
-	vma->vm_private_data = (void *)(reserve | flags);
+	set_vma_private_data(vma,
+		(get_vma_private_data(vma) & HPAGE_RESV_MASK) | reserve);
 }
 
 static void set_vma_resv_flags(struct vm_area_struct *vma, unsigned long flags)
 {
-	unsigned long reserveflags = (unsigned long)vma->vm_private_data;
 	VM_BUG_ON(!is_vm_hugetlb_page(vma));
-	vma->vm_private_data = (void *)(reserveflags | flags);
+	VM_BUG_ON(vma->vm_flags & VM_SHARED);
+
+	set_vma_private_data(vma, get_vma_private_data(vma) | flags);
 }
 
 static int is_vma_resv_set(struct vm_area_struct *vma, unsigned long flag)
 {
 	VM_BUG_ON(!is_vm_hugetlb_page(vma));
-	return ((unsigned long)vma->vm_private_data & flag) != 0;
+
+	return (get_vma_private_data(vma) & flag) != 0;
 }
 
 /* Decrement the reserved pages in the hugepage pool by one */
@@ -1151,11 +1185,10 @@ static struct page *hugetlbfs_pagecache_page(struct vm_area_struct *vma,
 			unsigned long address)
 {
 	struct address_space *mapping;
-	unsigned long idx;
+	pgoff_t idx;
 
 	mapping = vma->vm_file->f_mapping;
-	idx = ((address - vma->vm_start) >> HPAGE_SHIFT)
-		+ (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT));
+	idx = vma_pagecache_offset(vma, address);
 
 	return find_lock_page(mapping, idx);
 }
@@ -1164,7 +1197,7 @@ static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma,
 			unsigned long address, pte_t *ptep, int write_access)
 {
 	int ret = VM_FAULT_SIGBUS;
-	unsigned long idx;
+	pgoff_t idx;
 	unsigned long size;
 	struct page *page;
 	struct address_space *mapping;
@@ -1183,8 +1216,7 @@ static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	}
 
 	mapping = vma->vm_file->f_mapping;
-	idx = ((address - vma->vm_start) >> HPAGE_SHIFT)
-		+ (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT));
+	idx = vma_pagecache_offset(vma, address);
 
 	/*
 	 * Use page lock to guard against racing truncation

From cdfd4325c0d878679bd6a3ba8285b71d9980e3c0 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:27:28 -0700
Subject: [PATCH 0238/1171] mm: record MAP_NORESERVE status on vmas and fix
 small page mprotect reservations

With Mel's hugetlb private reservation support patches applied, strict
overcommit semantics are applied to both shared and private huge page
mappings.  This can be a problem if an application relied on unlimited
overcommit semantics for private mappings.  An example of this would be an
application which maps a huge area with the intention of using it very
sparsely.  These application would benefit from being able to opt-out of
the strict overcommit.  It should be noted that prior to hugetlb
supporting demand faulting all mappings were fully populated and so
applications of this type should be rare.

This patch stack implements the MAP_NORESERVE mmap() flag for huge page
mappings.  This flag has the same meaning as for small page mappings,
suppressing reservations for that mapping.

Thanks to Mel Gorman for reviewing a number of early versions of these
patches.

This patch:

When a small page mapping is created with mmap() reservations are created
by default for any memory pages required.  When the region is read/write
the reservation is increased for every page, no reservation is needed for
read-only regions (as they implicitly share the zero page).  Reservations
are tracked via the VM_ACCOUNT vma flag which is present when the region
has reservation backing it.  When we convert a region from read-only to
read-write new reservations are aquired and VM_ACCOUNT is set.  However,
when a read-only map is created with MAP_NORESERVE it is indistinguishable
from a normal mapping.  When we then convert that to read/write we are
forced to incorrectly create reservations for it as we have no record of
the original MAP_NORESERVE.

This patch introduces a new vma flag VM_NORESERVE which records the
presence of the original MAP_NORESERVE flag.  This allows us to
distinguish these two circumstances and correctly account the reserve.

As well as fixing this FIXME in the code, this makes it much easier to
introduce MAP_NORESERVE support for huge pages as this flag is available
consistantly for the life of the mapping.  VM_ACCOUNT on the other hand is
heavily used at the generic level in association with small pages.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Adam Litke <agl@us.ibm.com>
Cc: Johannes Weiner <hannes@saeurebad.de>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/mm.h | 1 +
 mm/mmap.c          | 3 +++
 mm/mprotect.c      | 6 ++----
 3 files changed, 6 insertions(+), 4 deletions(-)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index 196924b657bc..df322fb4df31 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -100,6 +100,7 @@ extern unsigned int kobjsize(const void *objp);
 #define VM_DONTEXPAND	0x00040000	/* Cannot expand with mremap() */
 #define VM_RESERVED	0x00080000	/* Count as reserved_vm like IO */
 #define VM_ACCOUNT	0x00100000	/* Is a VM accounted object */
+#define VM_NORESERVE	0x00200000	/* should the VM suppress accounting */
 #define VM_HUGETLB	0x00400000	/* Huge TLB Page VM */
 #define VM_NONLINEAR	0x00800000	/* Is non-linear (remap_file_pages) */
 #define VM_MAPPED_COPY	0x01000000	/* T if mapped copy of data (nommu mmap) */
diff --git a/mm/mmap.c b/mm/mmap.c
index 75e0d0673d78..57d3b6097deb 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -1110,6 +1110,9 @@ munmap_back:
 	if (!may_expand_vm(mm, len >> PAGE_SHIFT))
 		return -ENOMEM;
 
+	if (flags & MAP_NORESERVE)
+		vm_flags |= VM_NORESERVE;
+
 	if (accountable && (!(flags & MAP_NORESERVE) ||
 			    sysctl_overcommit_memory == OVERCOMMIT_NEVER)) {
 		if (vm_flags & VM_SHARED) {
diff --git a/mm/mprotect.c b/mm/mprotect.c
index 360d9cc8b38c..abd645a3b0a0 100644
--- a/mm/mprotect.c
+++ b/mm/mprotect.c
@@ -153,12 +153,10 @@ mprotect_fixup(struct vm_area_struct *vma, struct vm_area_struct **pprev,
 	 * If we make a private mapping writable we increase our commit;
 	 * but (without finer accounting) cannot reduce our commit if we
 	 * make it unwritable again.
-	 *
-	 * FIXME? We haven't defined a VM_NORESERVE flag, so mprotecting
-	 * a MAP_NORESERVE private mapping to writable will now reserve.
 	 */
 	if (newflags & VM_WRITE) {
-		if (!(oldflags & (VM_ACCOUNT|VM_WRITE|VM_SHARED))) {
+		if (!(oldflags & (VM_ACCOUNT|VM_WRITE|
+						VM_SHARED|VM_NORESERVE))) {
 			charged = nrpages;
 			if (security_vm_enough_memory(charged))
 				return -ENOMEM;

From 9682290484370ce68ba23cd2ec2838e301934199 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:27:29 -0700
Subject: [PATCH 0239/1171] hugetlb: move reservation region support earlier

The following patch will require use of the reservation regions support.
Move this earlier in the file.  No changes have been made to this code.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Acked-by: Adam Litke <agl@us.ibm.com>
Cc: Johannes Weiner <hannes@saeurebad.de>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/hugetlb.c | 246 ++++++++++++++++++++++++++-------------------------
 1 file changed, 125 insertions(+), 121 deletions(-)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 3e873f0101fb..05bc9af4fca9 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -40,6 +40,131 @@ static int hugetlb_next_nid;
  */
 static DEFINE_SPINLOCK(hugetlb_lock);
 
+/*
+ * Region tracking -- allows tracking of reservations and instantiated pages
+ *                    across the pages in a mapping.
+ */
+struct file_region {
+	struct list_head link;
+	long from;
+	long to;
+};
+
+static long region_add(struct list_head *head, long f, long t)
+{
+	struct file_region *rg, *nrg, *trg;
+
+	/* Locate the region we are either in or before. */
+	list_for_each_entry(rg, head, link)
+		if (f <= rg->to)
+			break;
+
+	/* Round our left edge to the current segment if it encloses us. */
+	if (f > rg->from)
+		f = rg->from;
+
+	/* Check for and consume any regions we now overlap with. */
+	nrg = rg;
+	list_for_each_entry_safe(rg, trg, rg->link.prev, link) {
+		if (&rg->link == head)
+			break;
+		if (rg->from > t)
+			break;
+
+		/* If this area reaches higher then extend our area to
+		 * include it completely.  If this is not the first area
+		 * which we intend to reuse, free it. */
+		if (rg->to > t)
+			t = rg->to;
+		if (rg != nrg) {
+			list_del(&rg->link);
+			kfree(rg);
+		}
+	}
+	nrg->from = f;
+	nrg->to = t;
+	return 0;
+}
+
+static long region_chg(struct list_head *head, long f, long t)
+{
+	struct file_region *rg, *nrg;
+	long chg = 0;
+
+	/* Locate the region we are before or in. */
+	list_for_each_entry(rg, head, link)
+		if (f <= rg->to)
+			break;
+
+	/* If we are below the current region then a new region is required.
+	 * Subtle, allocate a new region at the position but make it zero
+	 * size such that we can guarantee to record the reservation. */
+	if (&rg->link == head || t < rg->from) {
+		nrg = kmalloc(sizeof(*nrg), GFP_KERNEL);
+		if (!nrg)
+			return -ENOMEM;
+		nrg->from = f;
+		nrg->to   = f;
+		INIT_LIST_HEAD(&nrg->link);
+		list_add(&nrg->link, rg->link.prev);
+
+		return t - f;
+	}
+
+	/* Round our left edge to the current segment if it encloses us. */
+	if (f > rg->from)
+		f = rg->from;
+	chg = t - f;
+
+	/* Check for and consume any regions we now overlap with. */
+	list_for_each_entry(rg, rg->link.prev, link) {
+		if (&rg->link == head)
+			break;
+		if (rg->from > t)
+			return chg;
+
+		/* We overlap with this area, if it extends futher than
+		 * us then we must extend ourselves.  Account for its
+		 * existing reservation. */
+		if (rg->to > t) {
+			chg += rg->to - t;
+			t = rg->to;
+		}
+		chg -= rg->to - rg->from;
+	}
+	return chg;
+}
+
+static long region_truncate(struct list_head *head, long end)
+{
+	struct file_region *rg, *trg;
+	long chg = 0;
+
+	/* Locate the region we are either in or before. */
+	list_for_each_entry(rg, head, link)
+		if (end <= rg->to)
+			break;
+	if (&rg->link == head)
+		return 0;
+
+	/* If we are in the middle of a region then adjust it. */
+	if (end > rg->from) {
+		chg = rg->to - end;
+		rg->to = end;
+		rg = list_entry(rg->link.next, typeof(*rg), link);
+	}
+
+	/* Drop any remaining regions. */
+	list_for_each_entry_safe(rg, trg, rg->link.prev, link) {
+		if (&rg->link == head)
+			break;
+		chg += rg->to - rg->from;
+		list_del(&rg->link);
+		kfree(rg);
+	}
+	return chg;
+}
+
 /*
  * Convert the address within this vma to the page offset within
  * the mapping, in base page units.
@@ -1429,127 +1554,6 @@ void hugetlb_change_protection(struct vm_area_struct *vma,
 	flush_tlb_range(vma, start, end);
 }
 
-struct file_region {
-	struct list_head link;
-	long from;
-	long to;
-};
-
-static long region_add(struct list_head *head, long f, long t)
-{
-	struct file_region *rg, *nrg, *trg;
-
-	/* Locate the region we are either in or before. */
-	list_for_each_entry(rg, head, link)
-		if (f <= rg->to)
-			break;
-
-	/* Round our left edge to the current segment if it encloses us. */
-	if (f > rg->from)
-		f = rg->from;
-
-	/* Check for and consume any regions we now overlap with. */
-	nrg = rg;
-	list_for_each_entry_safe(rg, trg, rg->link.prev, link) {
-		if (&rg->link == head)
-			break;
-		if (rg->from > t)
-			break;
-
-		/* If this area reaches higher then extend our area to
-		 * include it completely.  If this is not the first area
-		 * which we intend to reuse, free it. */
-		if (rg->to > t)
-			t = rg->to;
-		if (rg != nrg) {
-			list_del(&rg->link);
-			kfree(rg);
-		}
-	}
-	nrg->from = f;
-	nrg->to = t;
-	return 0;
-}
-
-static long region_chg(struct list_head *head, long f, long t)
-{
-	struct file_region *rg, *nrg;
-	long chg = 0;
-
-	/* Locate the region we are before or in. */
-	list_for_each_entry(rg, head, link)
-		if (f <= rg->to)
-			break;
-
-	/* If we are below the current region then a new region is required.
-	 * Subtle, allocate a new region at the position but make it zero
-	 * size such that we can guarantee to record the reservation. */
-	if (&rg->link == head || t < rg->from) {
-		nrg = kmalloc(sizeof(*nrg), GFP_KERNEL);
-		if (!nrg)
-			return -ENOMEM;
-		nrg->from = f;
-		nrg->to   = f;
-		INIT_LIST_HEAD(&nrg->link);
-		list_add(&nrg->link, rg->link.prev);
-
-		return t - f;
-	}
-
-	/* Round our left edge to the current segment if it encloses us. */
-	if (f > rg->from)
-		f = rg->from;
-	chg = t - f;
-
-	/* Check for and consume any regions we now overlap with. */
-	list_for_each_entry(rg, rg->link.prev, link) {
-		if (&rg->link == head)
-			break;
-		if (rg->from > t)
-			return chg;
-
-		/* We overlap with this area, if it extends futher than
-		 * us then we must extend ourselves.  Account for its
-		 * existing reservation. */
-		if (rg->to > t) {
-			chg += rg->to - t;
-			t = rg->to;
-		}
-		chg -= rg->to - rg->from;
-	}
-	return chg;
-}
-
-static long region_truncate(struct list_head *head, long end)
-{
-	struct file_region *rg, *trg;
-	long chg = 0;
-
-	/* Locate the region we are either in or before. */
-	list_for_each_entry(rg, head, link)
-		if (end <= rg->to)
-			break;
-	if (&rg->link == head)
-		return 0;
-
-	/* If we are in the middle of a region then adjust it. */
-	if (end > rg->from) {
-		chg = rg->to - end;
-		rg->to = end;
-		rg = list_entry(rg->link.next, typeof(*rg), link);
-	}
-
-	/* Drop any remaining regions. */
-	list_for_each_entry_safe(rg, trg, rg->link.prev, link) {
-		if (&rg->link == head)
-			break;
-		chg += rg->to - rg->from;
-		list_del(&rg->link);
-		kfree(rg);
-	}
-	return chg;
-}
-
 int hugetlb_reserve_pages(struct inode *inode,
 					long from, long to,
 					struct vm_area_struct *vma)

From c37f9fb11c976ffc08200d631dada6dcbfd07ea4 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:27:30 -0700
Subject: [PATCH 0240/1171] hugetlb: allow huge page mappings to be created
 without reservations

By default all shared mappings and most private mappings now have
reservations associated with them.  This improves semantics by providing
allocation guarentees to the mapper.  However a small number of
applications may attempt to make very large sparse mappings, with these
strict reservations the system will never be able to honour the mapping.

This patch set brings MAP_NORESERVE support to hugetlb files.  This allows
new mappings to be made to hugetlbfs files without an associated
reservation, for both shared and private mappings.  This allows
applications which want to create very sparse mappings to opt-out of the
reservation system.  Obviously as there is no reservation they are liable
to fault at runtime if the huge page pool becomes exhausted; buyer beware.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Adam Litke <agl@us.ibm.com>
Cc: Johannes Weiner <hannes@saeurebad.de>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/hugetlb.c | 58 +++++++++++++++++++++++++++++++++++++++++++++++-----
 1 file changed, 53 insertions(+), 5 deletions(-)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 05bc9af4fca9..72acbb29d2cc 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -247,6 +247,9 @@ static int is_vma_resv_set(struct vm_area_struct *vma, unsigned long flag)
 /* Decrement the reserved pages in the hugepage pool by one */
 static void decrement_hugepage_resv_vma(struct vm_area_struct *vma)
 {
+	if (vma->vm_flags & VM_NORESERVE)
+		return;
+
 	if (vma->vm_flags & VM_SHARED) {
 		/* Shared mappings always use reserves */
 		resv_huge_pages--;
@@ -720,25 +723,65 @@ static void return_unused_surplus_pages(unsigned long unused_resv_pages)
 	}
 }
 
+/*
+ * Determine if the huge page at addr within the vma has an associated
+ * reservation.  Where it does not we will need to logically increase
+ * reservation and actually increase quota before an allocation can occur.
+ * Where any new reservation would be required the reservation change is
+ * prepared, but not committed.  Once the page has been quota'd allocated
+ * an instantiated the change should be committed via vma_commit_reservation.
+ * No action is required on failure.
+ */
+static int vma_needs_reservation(struct vm_area_struct *vma, unsigned long addr)
+{
+	struct address_space *mapping = vma->vm_file->f_mapping;
+	struct inode *inode = mapping->host;
+
+	if (vma->vm_flags & VM_SHARED) {
+		pgoff_t idx = vma_pagecache_offset(vma, addr);
+		return region_chg(&inode->i_mapping->private_list,
+							idx, idx + 1);
+
+	} else {
+		if (!is_vma_resv_set(vma, HPAGE_RESV_OWNER))
+			return 1;
+	}
+
+	return 0;
+}
+static void vma_commit_reservation(struct vm_area_struct *vma,
+							unsigned long addr)
+{
+	struct address_space *mapping = vma->vm_file->f_mapping;
+	struct inode *inode = mapping->host;
+
+	if (vma->vm_flags & VM_SHARED) {
+		pgoff_t idx = vma_pagecache_offset(vma, addr);
+		region_add(&inode->i_mapping->private_list, idx, idx + 1);
+	}
+}
+
 static struct page *alloc_huge_page(struct vm_area_struct *vma,
 				    unsigned long addr, int avoid_reserve)
 {
 	struct page *page;
 	struct address_space *mapping = vma->vm_file->f_mapping;
 	struct inode *inode = mapping->host;
-	unsigned int chg = 0;
+	unsigned int chg;
 
 	/*
 	 * Processes that did not create the mapping will have no reserves and
 	 * will not have accounted against quota. Check that the quota can be
 	 * made before satisfying the allocation
+	 * MAP_NORESERVE mappings may also need pages and quota allocated
+	 * if no reserve mapping overlaps.
 	 */
-	if (!(vma->vm_flags & VM_SHARED) &&
-			!is_vma_resv_set(vma, HPAGE_RESV_OWNER)) {
-		chg = 1;
+	chg = vma_needs_reservation(vma, addr);
+	if (chg < 0)
+		return ERR_PTR(chg);
+	if (chg)
 		if (hugetlb_get_quota(inode->i_mapping, chg))
 			return ERR_PTR(-ENOSPC);
-	}
 
 	spin_lock(&hugetlb_lock);
 	page = dequeue_huge_page_vma(vma, addr, avoid_reserve);
@@ -755,6 +798,8 @@ static struct page *alloc_huge_page(struct vm_area_struct *vma,
 	set_page_refcounted(page);
 	set_page_private(page, (unsigned long) mapping);
 
+	vma_commit_reservation(vma, addr);
+
 	return page;
 }
 
@@ -1560,6 +1605,9 @@ int hugetlb_reserve_pages(struct inode *inode,
 {
 	long ret, chg;
 
+	if (vma && vma->vm_flags & VM_NORESERVE)
+		return 0;
+
 	/*
 	 * Shared mappings base their reservation on the number of pages that
 	 * are already allocated on behalf of the file. Private mappings need

From 84afd99b8398c9d73af8238aa3cd835858e3097a Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:27:32 -0700
Subject: [PATCH 0241/1171] hugetlb reservations: fix hugetlb MAP_PRIVATE
 reservations across vma splits

When a hugetlb mapping with a reservation is split, a new VMA is cloned
from the original.  This new VMA is a direct copy of the original
including the reservation count.  When this pair of VMAs are unmapped we
will incorrect double account the unused reservation and the overall
reservation count will be incorrect, in extreme cases it will wrap.

The problem occurs when we split an existing VMA say to unmap a page in
the middle.  split_vma() will create a new VMA copying all fields from the
original.  As we are storing our reservation count in vm_private_data this
is also copies, endowing the new VMA with a duplicate of the original
VMA's reservation.  Neither of the new VMAs can exhaust these reservations
as they are too small, but when we unmap and close these VMAs we will
incorrect credit the remainder twice and resv_huge_pages will become out
of sync.  This can lead to allocation failures on mappings with
reservations and even to resv_huge_pages wrapping which prevents all
subsequent hugepage allocations.

The simple fix would be to correctly apportion the remaining reservation
count when the split is made.  However the only hook we have vm_ops->open
only has the new VMA we do not know the identity of the preceeding VMA.
Also even if we did have that VMA to hand we do not know how much of the
reservation was consumed each side of the split.

This patch therefore takes a different tack.  We know that the whole of
any private mapping (which has a reservation) has a reservation over its
whole size.  Any present pages represent consumed reservation.  Therefore
if we track the instantiated pages we can calculate the remaining
reservation.

This patch reuses the existing regions code to track the regions for which
we have consumed reservation (ie.  the instantiated pages), as each page
is faulted in we record the consumption of reservation for the new page.
When we need to return unused reservations at unmap time we simply count
the consumed reservation region subtracting that from the whole of the
map.  During a VMA split the newly opened VMA will point to the same
region map, as this map is offset oriented it remains valid for both of
the split VMAs.  This map is referenced counted so that it is removed when
all VMAs which are part of the mmap are gone.

Thanks to Adam Litke and Mel Gorman for their review feedback.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Cc: Adam Litke <agl@us.ibm.com>
Cc: Johannes Weiner <hannes@saeurebad.de>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Cc: Jon Tollefson <kniht@linux.vnet.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/hugetlb.c | 172 +++++++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 145 insertions(+), 27 deletions(-)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 72acbb29d2cc..65616941a383 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -43,6 +43,16 @@ static DEFINE_SPINLOCK(hugetlb_lock);
 /*
  * Region tracking -- allows tracking of reservations and instantiated pages
  *                    across the pages in a mapping.
+ *
+ * The region data structures are protected by a combination of the mmap_sem
+ * and the hugetlb_instantion_mutex.  To access or modify a region the caller
+ * must either hold the mmap_sem for write, or the mmap_sem for read and
+ * the hugetlb_instantiation mutex:
+ *
+ * 	down_write(&mm->mmap_sem);
+ * or
+ * 	down_read(&mm->mmap_sem);
+ * 	mutex_lock(&hugetlb_instantiation_mutex);
  */
 struct file_region {
 	struct list_head link;
@@ -165,6 +175,30 @@ static long region_truncate(struct list_head *head, long end)
 	return chg;
 }
 
+static long region_count(struct list_head *head, long f, long t)
+{
+	struct file_region *rg;
+	long chg = 0;
+
+	/* Locate each segment we overlap with, and count that overlap. */
+	list_for_each_entry(rg, head, link) {
+		int seg_from;
+		int seg_to;
+
+		if (rg->to <= f)
+			continue;
+		if (rg->from >= t)
+			break;
+
+		seg_from = max(rg->from, f);
+		seg_to = min(rg->to, t);
+
+		chg += seg_to - seg_from;
+	}
+
+	return chg;
+}
+
 /*
  * Convert the address within this vma to the page offset within
  * the mapping, in base page units.
@@ -187,9 +221,15 @@ static pgoff_t vma_pagecache_offset(struct vm_area_struct *vma,
 			(vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT));
 }
 
-#define HPAGE_RESV_OWNER    (1UL << (BITS_PER_LONG - 1))
-#define HPAGE_RESV_UNMAPPED (1UL << (BITS_PER_LONG - 2))
+/*
+ * Flags for MAP_PRIVATE reservations.  These are stored in the bottom
+ * bits of the reservation map pointer, which are always clear due to
+ * alignment.
+ */
+#define HPAGE_RESV_OWNER    (1UL << 0)
+#define HPAGE_RESV_UNMAPPED (1UL << 1)
 #define HPAGE_RESV_MASK (HPAGE_RESV_OWNER | HPAGE_RESV_UNMAPPED)
+
 /*
  * These helpers are used to track how many pages are reserved for
  * faults in a MAP_PRIVATE mapping. Only the process that called mmap()
@@ -199,6 +239,15 @@ static pgoff_t vma_pagecache_offset(struct vm_area_struct *vma,
  * the reserve counters are updated with the hugetlb_lock held. It is safe
  * to reset the VMA at fork() time as it is not in use yet and there is no
  * chance of the global counters getting corrupted as a result of the values.
+ *
+ * The private mapping reservation is represented in a subtly different
+ * manner to a shared mapping.  A shared mapping has a region map associated
+ * with the underlying file, this region map represents the backing file
+ * pages which have ever had a reservation assigned which this persists even
+ * after the page is instantiated.  A private mapping has a region map
+ * associated with the original mmap which is attached to all VMAs which
+ * reference it, this region map represents those offsets which have consumed
+ * reservation ie. where pages have been instantiated.
  */
 static unsigned long get_vma_private_data(struct vm_area_struct *vma)
 {
@@ -211,22 +260,48 @@ static void set_vma_private_data(struct vm_area_struct *vma,
 	vma->vm_private_data = (void *)value;
 }
 
-static unsigned long vma_resv_huge_pages(struct vm_area_struct *vma)
+struct resv_map {
+	struct kref refs;
+	struct list_head regions;
+};
+
+struct resv_map *resv_map_alloc(void)
+{
+	struct resv_map *resv_map = kmalloc(sizeof(*resv_map), GFP_KERNEL);
+	if (!resv_map)
+		return NULL;
+
+	kref_init(&resv_map->refs);
+	INIT_LIST_HEAD(&resv_map->regions);
+
+	return resv_map;
+}
+
+void resv_map_release(struct kref *ref)
+{
+	struct resv_map *resv_map = container_of(ref, struct resv_map, refs);
+
+	/* Clear out any active regions before we release the map. */
+	region_truncate(&resv_map->regions, 0);
+	kfree(resv_map);
+}
+
+static struct resv_map *vma_resv_map(struct vm_area_struct *vma)
 {
 	VM_BUG_ON(!is_vm_hugetlb_page(vma));
 	if (!(vma->vm_flags & VM_SHARED))
-		return get_vma_private_data(vma) & ~HPAGE_RESV_MASK;
+		return (struct resv_map *)(get_vma_private_data(vma) &
+							~HPAGE_RESV_MASK);
 	return 0;
 }
 
-static void set_vma_resv_huge_pages(struct vm_area_struct *vma,
-							unsigned long reserve)
+static void set_vma_resv_map(struct vm_area_struct *vma, struct resv_map *map)
 {
 	VM_BUG_ON(!is_vm_hugetlb_page(vma));
 	VM_BUG_ON(vma->vm_flags & VM_SHARED);
 
-	set_vma_private_data(vma,
-		(get_vma_private_data(vma) & HPAGE_RESV_MASK) | reserve);
+	set_vma_private_data(vma, (get_vma_private_data(vma) &
+				HPAGE_RESV_MASK) | (unsigned long)map);
 }
 
 static void set_vma_resv_flags(struct vm_area_struct *vma, unsigned long flags)
@@ -253,19 +328,12 @@ static void decrement_hugepage_resv_vma(struct vm_area_struct *vma)
 	if (vma->vm_flags & VM_SHARED) {
 		/* Shared mappings always use reserves */
 		resv_huge_pages--;
-	} else {
+	} else if (is_vma_resv_set(vma, HPAGE_RESV_OWNER)) {
 		/*
 		 * Only the process that called mmap() has reserves for
 		 * private mappings.
 		 */
-		if (is_vma_resv_set(vma, HPAGE_RESV_OWNER)) {
-			unsigned long flags, reserve;
-			resv_huge_pages--;
-			flags = (unsigned long)vma->vm_private_data &
-							HPAGE_RESV_MASK;
-			reserve = (unsigned long)vma->vm_private_data - 1;
-			vma->vm_private_data = (void *)(reserve | flags);
-		}
+		resv_huge_pages--;
 	}
 }
 
@@ -282,7 +350,7 @@ static int vma_has_private_reserves(struct vm_area_struct *vma)
 {
 	if (vma->vm_flags & VM_SHARED)
 		return 0;
-	if (!vma_resv_huge_pages(vma))
+	if (!is_vma_resv_set(vma, HPAGE_RESV_OWNER))
 		return 0;
 	return 1;
 }
@@ -742,12 +810,19 @@ static int vma_needs_reservation(struct vm_area_struct *vma, unsigned long addr)
 		return region_chg(&inode->i_mapping->private_list,
 							idx, idx + 1);
 
-	} else {
-		if (!is_vma_resv_set(vma, HPAGE_RESV_OWNER))
-			return 1;
-	}
+	} else if (!is_vma_resv_set(vma, HPAGE_RESV_OWNER)) {
+		return 1;
 
-	return 0;
+	} else  {
+		int err;
+		pgoff_t idx = vma_pagecache_offset(vma, addr);
+		struct resv_map *reservations = vma_resv_map(vma);
+
+		err = region_chg(&reservations->regions, idx, idx + 1);
+		if (err < 0)
+			return err;
+		return 0;
+	}
 }
 static void vma_commit_reservation(struct vm_area_struct *vma,
 							unsigned long addr)
@@ -758,6 +833,13 @@ static void vma_commit_reservation(struct vm_area_struct *vma,
 	if (vma->vm_flags & VM_SHARED) {
 		pgoff_t idx = vma_pagecache_offset(vma, addr);
 		region_add(&inode->i_mapping->private_list, idx, idx + 1);
+
+	} else if (is_vma_resv_set(vma, HPAGE_RESV_OWNER)) {
+		pgoff_t idx = vma_pagecache_offset(vma, addr);
+		struct resv_map *reservations = vma_resv_map(vma);
+
+		/* Mark this page used in the map. */
+		region_add(&reservations->regions, idx, idx + 1);
 	}
 }
 
@@ -1047,11 +1129,41 @@ out:
 	return ret;
 }
 
+static void hugetlb_vm_op_open(struct vm_area_struct *vma)
+{
+	struct resv_map *reservations = vma_resv_map(vma);
+
+	/*
+	 * This new VMA should share its siblings reservation map if present.
+	 * The VMA will only ever have a valid reservation map pointer where
+	 * it is being copied for another still existing VMA.  As that VMA
+	 * has a reference to the reservation map it cannot dissappear until
+	 * after this open call completes.  It is therefore safe to take a
+	 * new reference here without additional locking.
+	 */
+	if (reservations)
+		kref_get(&reservations->refs);
+}
+
 static void hugetlb_vm_op_close(struct vm_area_struct *vma)
 {
-	unsigned long reserve = vma_resv_huge_pages(vma);
-	if (reserve)
-		hugetlb_acct_memory(-reserve);
+	struct resv_map *reservations = vma_resv_map(vma);
+	unsigned long reserve;
+	unsigned long start;
+	unsigned long end;
+
+	if (reservations) {
+		start = vma_pagecache_offset(vma, vma->vm_start);
+		end = vma_pagecache_offset(vma, vma->vm_end);
+
+		reserve = (end - start) -
+			region_count(&reservations->regions, start, end);
+
+		kref_put(&reservations->refs, resv_map_release);
+
+		if (reserve)
+			hugetlb_acct_memory(-reserve);
+	}
 }
 
 /*
@@ -1068,6 +1180,7 @@ static int hugetlb_vm_op_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 
 struct vm_operations_struct hugetlb_vm_ops = {
 	.fault = hugetlb_vm_op_fault,
+	.open = hugetlb_vm_op_open,
 	.close = hugetlb_vm_op_close,
 };
 
@@ -1617,8 +1730,13 @@ int hugetlb_reserve_pages(struct inode *inode,
 	if (!vma || vma->vm_flags & VM_SHARED)
 		chg = region_chg(&inode->i_mapping->private_list, from, to);
 	else {
+		struct resv_map *resv_map = resv_map_alloc();
+		if (!resv_map)
+			return -ENOMEM;
+
 		chg = to - from;
-		set_vma_resv_huge_pages(vma, chg);
+
+		set_vma_resv_map(vma, resv_map);
 		set_vma_resv_flags(vma, HPAGE_RESV_OWNER);
 	}
 

From a858f7b2e9bb4eb665176dde5cf32eeaaf90f153 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Wed, 23 Jul 2008 21:27:33 -0700
Subject: [PATCH 0242/1171] vma_page_offset() has no callees: drop it

Hugh adds: vma_pagecache_offset() has a dangerously misleading name, since
it's using hugepage units: rename it to vma_hugecache_offset().

[apw@shadowen.org: restack onto fixed MAP_PRIVATE reservations]
[akpm@linux-foundation.org: vma_split conversion]
Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: Adam Litke <agl@us.ibm.com>
Cc: Nishanth Aravamudan <nacc@us.ibm.com>
Cc: Andi Kleen <ak@suse.de>
Cc: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/hugetlb.c | 29 +++++++++--------------------
 1 file changed, 9 insertions(+), 20 deletions(-)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 65616941a383..eda9642254a0 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -199,22 +199,11 @@ static long region_count(struct list_head *head, long f, long t)
 	return chg;
 }
 
-/*
- * Convert the address within this vma to the page offset within
- * the mapping, in base page units.
- */
-static pgoff_t vma_page_offset(struct vm_area_struct *vma,
-				unsigned long address)
-{
-	return ((address - vma->vm_start) >> PAGE_SHIFT) +
-					(vma->vm_pgoff >> PAGE_SHIFT);
-}
-
 /*
  * Convert the address within this vma to the page offset within
  * the mapping, in pagecache page units; huge pages here.
  */
-static pgoff_t vma_pagecache_offset(struct vm_area_struct *vma,
+static pgoff_t vma_hugecache_offset(struct vm_area_struct *vma,
 					unsigned long address)
 {
 	return ((address - vma->vm_start) >> HPAGE_SHIFT) +
@@ -806,7 +795,7 @@ static int vma_needs_reservation(struct vm_area_struct *vma, unsigned long addr)
 	struct inode *inode = mapping->host;
 
 	if (vma->vm_flags & VM_SHARED) {
-		pgoff_t idx = vma_pagecache_offset(vma, addr);
+		pgoff_t idx = vma_hugecache_offset(vma, addr);
 		return region_chg(&inode->i_mapping->private_list,
 							idx, idx + 1);
 
@@ -815,7 +804,7 @@ static int vma_needs_reservation(struct vm_area_struct *vma, unsigned long addr)
 
 	} else  {
 		int err;
-		pgoff_t idx = vma_pagecache_offset(vma, addr);
+		pgoff_t idx = vma_hugecache_offset(vma, addr);
 		struct resv_map *reservations = vma_resv_map(vma);
 
 		err = region_chg(&reservations->regions, idx, idx + 1);
@@ -831,11 +820,11 @@ static void vma_commit_reservation(struct vm_area_struct *vma,
 	struct inode *inode = mapping->host;
 
 	if (vma->vm_flags & VM_SHARED) {
-		pgoff_t idx = vma_pagecache_offset(vma, addr);
+		pgoff_t idx = vma_hugecache_offset(vma, addr);
 		region_add(&inode->i_mapping->private_list, idx, idx + 1);
 
 	} else if (is_vma_resv_set(vma, HPAGE_RESV_OWNER)) {
-		pgoff_t idx = vma_pagecache_offset(vma, addr);
+		pgoff_t idx = vma_hugecache_offset(vma, addr);
 		struct resv_map *reservations = vma_resv_map(vma);
 
 		/* Mark this page used in the map. */
@@ -1153,8 +1142,8 @@ static void hugetlb_vm_op_close(struct vm_area_struct *vma)
 	unsigned long end;
 
 	if (reservations) {
-		start = vma_pagecache_offset(vma, vma->vm_start);
-		end = vma_pagecache_offset(vma, vma->vm_end);
+		start = vma_hugecache_offset(vma, vma->vm_start);
+		end = vma_hugecache_offset(vma, vma->vm_end);
 
 		reserve = (end - start) -
 			region_count(&reservations->regions, start, end);
@@ -1471,7 +1460,7 @@ static struct page *hugetlbfs_pagecache_page(struct vm_area_struct *vma,
 	pgoff_t idx;
 
 	mapping = vma->vm_file->f_mapping;
-	idx = vma_pagecache_offset(vma, address);
+	idx = vma_hugecache_offset(vma, address);
 
 	return find_lock_page(mapping, idx);
 }
@@ -1499,7 +1488,7 @@ static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	}
 
 	mapping = vma->vm_file->f_mapping;
-	idx = vma_pagecache_offset(vma, address);
+	idx = vma_hugecache_offset(vma, address);
 
 	/*
 	 * Use page lock to guard against racing truncation

From 11fa977ecde652ab324dd79c179deb52e82a8df1 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Wed, 23 Jul 2008 21:27:34 -0700
Subject: [PATCH 0243/1171] generic_file_aio_read() cleanups

As akpm points out, there's really no need for generic_file_aio_read to
make a special case of count 0: just loop through nr_segs doing nothing.
And as Harvey Harrison points out, there's no need to reset retval to 0
where it's already 0.

Setting count (or ocount) to 0 before calling generic_segment_checks is
unnecessary too; but reluctantly I'll leave that removal to someone with a
wider range of gcc versions to hand - 4.1.2 and 4.2.1 don't warn about it,
but perhaps others do - I forget which are the warniest versions.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Tested-by: Lawrence Greenfield <leg@google.com>
Cc: Christoph Rohland <hans-christoph.rohland@sap.com>
Cc: Badari Pulavarty <pbadari@us.ibm.com>
Cc: Zach Brown <zach.brown@oracle.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/filemap.c | 42 +++++++++++++++++++-----------------------
 1 file changed, 19 insertions(+), 23 deletions(-)

diff --git a/mm/filemap.c b/mm/filemap.c
index 6343f3c841b7..7675b91f4f63 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -1197,7 +1197,6 @@ generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov,
 
 		mapping = filp->f_mapping;
 		inode = mapping->host;
-		retval = 0;
 		if (!count)
 			goto out; /* skip atime */
 		size = i_size_read(inode);
@@ -1209,33 +1208,30 @@ generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov,
 			}
 			if (retval > 0)
 				*ppos = pos + retval;
-		}
-		if (likely(retval != 0)) {
-			file_accessed(filp);
-			goto out;
+			if (retval) {
+				file_accessed(filp);
+				goto out;
+			}
 		}
 	}
 
-	retval = 0;
-	if (count) {
-		for (seg = 0; seg < nr_segs; seg++) {
-			read_descriptor_t desc;
+	for (seg = 0; seg < nr_segs; seg++) {
+		read_descriptor_t desc;
 
-			desc.written = 0;
-			desc.arg.buf = iov[seg].iov_base;
-			desc.count = iov[seg].iov_len;
-			if (desc.count == 0)
-				continue;
-			desc.error = 0;
-			do_generic_file_read(filp,ppos,&desc,file_read_actor);
-			retval += desc.written;
-			if (desc.error) {
-				retval = retval ?: desc.error;
-				break;
-			}
-			if (desc.count > 0)
-				break;
+		desc.written = 0;
+		desc.arg.buf = iov[seg].iov_base;
+		desc.count = iov[seg].iov_len;
+		if (desc.count == 0)
+			continue;
+		desc.error = 0;
+		do_generic_file_read(filp, ppos, &desc, file_read_actor);
+		retval += desc.written;
+		if (desc.error) {
+			retval = retval ?: desc.error;
+			break;
 		}
+		if (desc.count > 0)
+			break;
 	}
 out:
 	return retval;

From bcd78e49613c41b5bed96fa288e983876f286a59 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Wed, 23 Jul 2008 21:27:35 -0700
Subject: [PATCH 0244/1171] tmpfs: support aio

We have a request for tmpfs to support the AIO interface: easily done, no
more than replacing the old shmem_file_read by shmem_file_aio_read,
cribbed from generic_file_aio_read.  (In 2.6.25 its write side was already
changed to use generic_file_aio_write.)

Incorporate cleanups from Andrew Morton and Harvey Harrison.

Tests out fine with LTP's ltp-aiodio.sh, given hacks (not included) to
support O_DIRECT.  tmpfs cannot honestly support O_DIRECT: its
cache-avoiding-IO nature is at odds with direct IO-avoiding-cache.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Tested-by: Lawrence Greenfield <leg@google.com>
Cc: Christoph Rohland <hans-christoph.rohland@sap.com>
Cc: Badari Pulavarty <pbadari@us.ibm.com>
Cc: Zach Brown <zach.brown@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/shmem.c | 47 ++++++++++++++++++++++++++++++-----------------
 1 file changed, 30 insertions(+), 17 deletions(-)

diff --git a/mm/shmem.c b/mm/shmem.c
index e2a6ae1a44e9..9ffbea9b79e1 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -1690,26 +1690,38 @@ static void do_shmem_file_read(struct file *filp, loff_t *ppos, read_descriptor_
 	file_accessed(filp);
 }
 
-static ssize_t shmem_file_read(struct file *filp, char __user *buf, size_t count, loff_t *ppos)
+static ssize_t shmem_file_aio_read(struct kiocb *iocb,
+		const struct iovec *iov, unsigned long nr_segs, loff_t pos)
 {
-	read_descriptor_t desc;
+	struct file *filp = iocb->ki_filp;
+	ssize_t retval;
+	unsigned long seg;
+	size_t count;
+	loff_t *ppos = &iocb->ki_pos;
 
-	if ((ssize_t) count < 0)
-		return -EINVAL;
-	if (!access_ok(VERIFY_WRITE, buf, count))
-		return -EFAULT;
-	if (!count)
-		return 0;
+	retval = generic_segment_checks(iov, &nr_segs, &count, VERIFY_WRITE);
+	if (retval)
+		return retval;
 
-	desc.written = 0;
-	desc.count = count;
-	desc.arg.buf = buf;
-	desc.error = 0;
+	for (seg = 0; seg < nr_segs; seg++) {
+		read_descriptor_t desc;
 
-	do_shmem_file_read(filp, ppos, &desc, file_read_actor);
-	if (desc.written)
-		return desc.written;
-	return desc.error;
+		desc.written = 0;
+		desc.arg.buf = iov[seg].iov_base;
+		desc.count = iov[seg].iov_len;
+		if (desc.count == 0)
+			continue;
+		desc.error = 0;
+		do_shmem_file_read(filp, ppos, &desc, file_read_actor);
+		retval += desc.written;
+		if (desc.error) {
+			retval = retval ?: desc.error;
+			break;
+		}
+		if (desc.count > 0)
+			break;
+	}
+	return retval;
 }
 
 static int shmem_statfs(struct dentry *dentry, struct kstatfs *buf)
@@ -2369,8 +2381,9 @@ static const struct file_operations shmem_file_operations = {
 	.mmap		= shmem_mmap,
 #ifdef CONFIG_TMPFS
 	.llseek		= generic_file_llseek,
-	.read		= shmem_file_read,
+	.read		= do_sync_read,
 	.write		= do_sync_write,
+	.aio_read	= shmem_file_aio_read,
 	.aio_write	= generic_file_aio_write,
 	.fsync		= simple_sync_file,
 	.splice_read	= generic_file_splice_read,

From cce770815869e9209171d819dfde89bcc738ab62 Mon Sep 17 00:00:00 2001
From: Pavel Machek <pavel@suse.cz>
Date: Wed, 23 Jul 2008 21:27:36 -0700
Subject: [PATCH 0245/1171] SYNC_FILE_RANGE_WRITE may and will block. Document
 that.

[akpm@linux-foundation.org: fix comment text]
Signed-off-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/sync.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/fs/sync.c b/fs/sync.c
index 228e17b5e9ee..2967562d416f 100644
--- a/fs/sync.c
+++ b/fs/sync.c
@@ -139,7 +139,8 @@ asmlinkage long sys_fdatasync(unsigned int fd)
  * before performing the write.
  *
  * SYNC_FILE_RANGE_WRITE: initiate writeout of all those dirty pages in the
- * range which are not presently under writeback.
+ * range which are not presently under writeback. Note that this may block for
+ * significant periods due to exhaustion of disk request structures.
  *
  * SYNC_FILE_RANGE_WAIT_AFTER: wait upon writeout of all pages in the range
  * after performing the write.

From a47a126ad5ea072aca3e611ed8f8dc6adad24bab Mon Sep 17 00:00:00 2001
From: Eric Dumazet <dada1@cosmosbay.com>
Date: Wed, 23 Jul 2008 21:27:38 -0700
Subject: [PATCH 0246/1171] vmallocinfo: add NUMA information

Christoph recently added /proc/vmallocinfo file to get information about
vmalloc allocations.

This patch adds NUMA specific information, giving number of pages
allocated on each memory node.

This should help to check that vmalloc() is able to respect NUMA policies.

Example of output on a four nodes machine (one cpu per node)

1) network hash tables are evenly spreaded on four nodes (OK) (Same
   point for inodes and dentries hash tables)

2) iptables tables (x_tables) are correctly allocated on each cpu node
   (OK).

3) sys_swapon() allocates its memory from one node only.

4) each loaded module is using memory on one node.

Sysadmins could tune their setup to change points 3) and 4) if necessary.

grep "pages="  /proc/vmallocinfo
0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204/0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128
0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204/0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64
0xffffc2000031a000-0xffffc2000031d000   12288 alloc_large_system_hash+0x204/0x2c0 pages=2 vmalloc N1=1 N2=1
0xffffc2000031f000-0xffffc2000032b000   49152 cramfs_uncompress_init+0x2e/0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3
0xffffc2000033e000-0xffffc20000341000   12288 sys_swapon+0x640/0xac0 pages=2 vmalloc N0=2
0xffffc20000341000-0xffffc20000344000   12288 xt_alloc_table_info+0xfe/0x130 [x_tables] pages=2 vmalloc N0=2
0xffffc20000344000-0xffffc20000347000   12288 xt_alloc_table_info+0xfe/0x130 [x_tables] pages=2 vmalloc N1=2
0xffffc20000347000-0xffffc2000034a000   12288 xt_alloc_table_info+0xfe/0x130 [x_tables] pages=2 vmalloc N2=2
0xffffc2000034a000-0xffffc2000034d000   12288 xt_alloc_table_info+0xfe/0x130 [x_tables] pages=2 vmalloc N3=2
0xffffc20004381000-0xffffc20004402000  528384 alloc_large_system_hash+0x204/0x2c0 pages=128 vmalloc N0=32 N1=32 N2=32 N3=32
0xffffc20004402000-0xffffc20004803000 4198400 alloc_large_system_hash+0x204/0x2c0 pages=1024 vmalloc vpages N0=256 N1=256 N2=256 N3=256
0xffffc20004803000-0xffffc20004904000 1052672 alloc_large_system_hash+0x204/0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64
0xffffc20004904000-0xffffc20004bec000 3047424 sys_swapon+0x640/0xac0 pages=743 vmalloc vpages N0=743
0xffffffffa0000000-0xffffffffa000f000   61440 sys_init_module+0xc27/0x1d00 pages=14 vmalloc N1=14
0xffffffffa000f000-0xffffffffa0014000   20480 sys_init_module+0xc27/0x1d00 pages=4 vmalloc N0=4
0xffffffffa0014000-0xffffffffa0017000   12288 sys_init_module+0xc27/0x1d00 pages=2 vmalloc N0=2
0xffffffffa0017000-0xffffffffa0022000   45056 sys_init_module+0xc27/0x1d00 pages=10 vmalloc N1=10
0xffffffffa0022000-0xffffffffa0028000   24576 sys_init_module+0xc27/0x1d00 pages=5 vmalloc N3=5
0xffffffffa0028000-0xffffffffa0050000  163840 sys_init_module+0xc27/0x1d00 pages=39 vmalloc N1=39
0xffffffffa0050000-0xffffffffa0052000    8192 sys_init_module+0xc27/0x1d00 pages=1 vmalloc N1=1
0xffffffffa0052000-0xffffffffa0056000   16384 sys_init_module+0xc27/0x1d00 pages=3 vmalloc N1=3
0xffffffffa0056000-0xffffffffa0081000  176128 sys_init_module+0xc27/0x1d00 pages=42 vmalloc N3=42
0xffffffffa0081000-0xffffffffa00ae000  184320 sys_init_module+0xc27/0x1d00 pages=44 vmalloc N3=44
0xffffffffa00ae000-0xffffffffa00b1000   12288 sys_init_module+0xc27/0x1d00 pages=2 vmalloc N3=2
0xffffffffa00b1000-0xffffffffa00b9000   32768 sys_init_module+0xc27/0x1d00 pages=7 vmalloc N0=7
0xffffffffa00b9000-0xffffffffa00c4000   45056 sys_init_module+0xc27/0x1d00 pages=10 vmalloc N3=10
0xffffffffa00c6000-0xffffffffa00e0000  106496 sys_init_module+0xc27/0x1d00 pages=25 vmalloc N2=25
0xffffffffa00e0000-0xffffffffa00f1000   69632 sys_init_module+0xc27/0x1d00 pages=16 vmalloc N2=16
0xffffffffa00f1000-0xffffffffa00f4000   12288 sys_init_module+0xc27/0x1d00 pages=2 vmalloc N3=2
0xffffffffa00f4000-0xffffffffa00f7000   12288 sys_init_module+0xc27/0x1d00 pages=2 vmalloc N3=2

[akpm@linux-foundation.org: fix comment]
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/filesystems/proc.txt | 44 ++++++++++++++++++++++++++++++
 fs/proc/proc_misc.c                | 15 ++++++++--
 mm/vmalloc.c                       | 20 ++++++++++++++
 3 files changed, 77 insertions(+), 2 deletions(-)

diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt
index 7f268f327d75..8c6384bdfed4 100644
--- a/Documentation/filesystems/proc.txt
+++ b/Documentation/filesystems/proc.txt
@@ -296,6 +296,7 @@ Table 1-4: Kernel info in /proc
  uptime      System uptime                                     
  version     Kernel version                                    
  video	     bttv info of video resources			(2.4)
+ vmallocinfo Show vmalloced areas
 ..............................................................................
 
 You can,  for  example,  check  which interrupts are currently in use and what
@@ -557,6 +558,49 @@ VmallocTotal: total size of vmalloc memory area
  VmallocUsed: amount of vmalloc area which is used
 VmallocChunk: largest contigious block of vmalloc area which is free
 
+..............................................................................
+
+vmallocinfo:
+
+Provides information about vmalloced/vmaped areas. One line per area,
+containing the virtual address range of the area, size in bytes,
+caller information of the creator, and optional information depending
+on the kind of area :
+
+ pages=nr    number of pages
+ phys=addr   if a physical address was specified
+ ioremap     I/O mapping (ioremap() and friends)
+ vmalloc     vmalloc() area
+ vmap        vmap()ed pages
+ user        VM_USERMAP area
+ vpages      buffer for pages pointers was vmalloced (huge area)
+ N<node>=nr  (Only on NUMA kernels)
+             Number of pages allocated on memory node <node>
+
+> cat /proc/vmallocinfo
+0xffffc20000000000-0xffffc20000201000 2101248 alloc_large_system_hash+0x204 ...
+  /0x2c0 pages=512 vmalloc N0=128 N1=128 N2=128 N3=128
+0xffffc20000201000-0xffffc20000302000 1052672 alloc_large_system_hash+0x204 ...
+  /0x2c0 pages=256 vmalloc N0=64 N1=64 N2=64 N3=64
+0xffffc20000302000-0xffffc20000304000    8192 acpi_tb_verify_table+0x21/0x4f...
+  phys=7fee8000 ioremap
+0xffffc20000304000-0xffffc20000307000   12288 acpi_tb_verify_table+0x21/0x4f...
+  phys=7fee7000 ioremap
+0xffffc2000031d000-0xffffc2000031f000    8192 init_vdso_vars+0x112/0x210
+0xffffc2000031f000-0xffffc2000032b000   49152 cramfs_uncompress_init+0x2e ...
+  /0x80 pages=11 vmalloc N0=3 N1=3 N2=2 N3=3
+0xffffc2000033a000-0xffffc2000033d000   12288 sys_swapon+0x640/0xac0      ...
+  pages=2 vmalloc N1=2
+0xffffc20000347000-0xffffc2000034c000   20480 xt_alloc_table_info+0xfe ...
+  /0x130 [x_tables] pages=4 vmalloc N0=4
+0xffffffffa0000000-0xffffffffa000f000   61440 sys_init_module+0xc27/0x1d00 ...
+   pages=14 vmalloc N2=14
+0xffffffffa000f000-0xffffffffa0014000   20480 sys_init_module+0xc27/0x1d00 ...
+   pages=4 vmalloc N1=4
+0xffffffffa0014000-0xffffffffa0017000   12288 sys_init_module+0xc27/0x1d00 ...
+   pages=2 vmalloc N1=2
+0xffffffffa0017000-0xffffffffa0022000   45056 sys_init_module+0xc27/0x1d00 ...
+   pages=10 vmalloc N0=10
 
 1.3 IDE devices in /proc/ide
 ----------------------------
diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c
index b14f43d25e9e..ded969862960 100644
--- a/fs/proc/proc_misc.c
+++ b/fs/proc/proc_misc.c
@@ -464,14 +464,25 @@ static const struct file_operations proc_slabstats_operations = {
 #ifdef CONFIG_MMU
 static int vmalloc_open(struct inode *inode, struct file *file)
 {
-	return seq_open(file, &vmalloc_op);
+	unsigned int *ptr = NULL;
+	int ret;
+
+	if (NUMA_BUILD)
+		ptr = kmalloc(nr_node_ids * sizeof(unsigned int), GFP_KERNEL);
+	ret = seq_open(file, &vmalloc_op);
+	if (!ret) {
+		struct seq_file *m = file->private_data;
+		m->private = ptr;
+	} else
+		kfree(ptr);
+	return ret;
 }
 
 static const struct file_operations proc_vmalloc_operations = {
 	.open		= vmalloc_open,
 	.read		= seq_read,
 	.llseek		= seq_lseek,
-	.release	= seq_release,
+	.release	= seq_release_private,
 };
 #endif
 
diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index 6e45b0f3d125..35f293816294 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -931,6 +931,25 @@ static void s_stop(struct seq_file *m, void *p)
 	read_unlock(&vmlist_lock);
 }
 
+static void show_numa_info(struct seq_file *m, struct vm_struct *v)
+{
+	if (NUMA_BUILD) {
+		unsigned int nr, *counters = m->private;
+
+		if (!counters)
+			return;
+
+		memset(counters, 0, nr_node_ids * sizeof(unsigned int));
+
+		for (nr = 0; nr < v->nr_pages; nr++)
+			counters[page_to_nid(v->pages[nr])]++;
+
+		for_each_node_state(nr, N_HIGH_MEMORY)
+			if (counters[nr])
+				seq_printf(m, " N%u=%u", nr, counters[nr]);
+	}
+}
+
 static int s_show(struct seq_file *m, void *p)
 {
 	struct vm_struct *v = p;
@@ -967,6 +986,7 @@ static int s_show(struct seq_file *m, void *p)
 	if (v->flags & VM_VPAGES)
 		seq_printf(m, " vpages");
 
+	show_numa_info(m, v);
 	seq_putc(m, '\n');
 	return 0;
 }

From 5e9426abe209cf134adbbd62c5e73ef534eb73e9 Mon Sep 17 00:00:00 2001
From: Nishanth Aravamudan <nacc@us.ibm.com>
Date: Wed, 23 Jul 2008 21:27:39 -0700
Subject: [PATCH 0247/1171] mm: remove mm_init compilation dependency on
 CONFIG_DEBUG_MEMORY_INIT

Towards the end of putting all core mm initialization in mm_init.c, I
plan on putting the creation of a mm kobject in a function in that file.
However, the file is currently only compiled if CONFIG_DEBUG_MEMORY_INIT
is set. Remove this dependency, but put the code under an #ifdef on the
same config option. This should result in no functional changes.

Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/Makefile  | 3 +--
 mm/mm_init.c | 2 ++
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/mm/Makefile b/mm/Makefile
index 4bbc8f094ff0..06ca2381fef1 100644
--- a/mm/Makefile
+++ b/mm/Makefile
@@ -11,7 +11,7 @@ obj-y			:= bootmem.o filemap.o mempool.o oom_kill.o fadvise.o \
 			   maccess.o page_alloc.o page-writeback.o pdflush.o \
 			   readahead.o swap.o truncate.o vmscan.o \
 			   prio_tree.o util.o mmzone.o vmstat.o backing-dev.o \
-			   page_isolation.o $(mmu-y)
+			   page_isolation.o mm_init.o $(mmu-y)
 
 obj-$(CONFIG_PROC_PAGE_MONITOR) += pagewalk.o
 obj-$(CONFIG_BOUNCE)	+= bounce.o
@@ -26,7 +26,6 @@ obj-$(CONFIG_TMPFS_POSIX_ACL) += shmem_acl.o
 obj-$(CONFIG_TINY_SHMEM) += tiny-shmem.o
 obj-$(CONFIG_SLOB) += slob.o
 obj-$(CONFIG_SLAB) += slab.o
-obj-$(CONFIG_DEBUG_MEMORY_INIT) += mm_init.o
 obj-$(CONFIG_SLUB) += slub.o
 obj-$(CONFIG_MEMORY_HOTPLUG) += memory_hotplug.o
 obj-$(CONFIG_FS_XIP) += filemap_xip.o
diff --git a/mm/mm_init.c b/mm/mm_init.c
index ce445ca097e7..eaf0d3b47099 100644
--- a/mm/mm_init.c
+++ b/mm/mm_init.c
@@ -9,6 +9,7 @@
 #include <linux/init.h>
 #include "internal.h"
 
+#ifdef CONFIG_DEBUG_MEMORY_INIT
 int __meminitdata mminit_loglevel;
 
 /* The zonelists are simply reported, validation is manual. */
@@ -132,3 +133,4 @@ static __init int set_mminit_loglevel(char *str)
 	return 0;
 }
 early_param("mminit_loglevel", set_mminit_loglevel);
+#endif /* CONFIG_DEBUG_MEMORY_INIT */

From ff7ea79cf7c3a481851bd4b2185fdeb6ce4afa29 Mon Sep 17 00:00:00 2001
From: Nishanth Aravamudan <nacc@us.ibm.com>
Date: Wed, 23 Jul 2008 21:27:39 -0700
Subject: [PATCH 0248/1171] mm: create /sys/kernel/mm

Add a kobject to create /sys/kernel/mm when sysfs is mounted.  The kobject
will exist regardless.  This will allow for the hugepage related sysfs
directories to exist under the mm "subsystem" directory.  Add an ABI file
appropriately.

[kosaki.motohiro@jp.fujitsu.com: fix build]
Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/ABI/testing/sysfs-kernel-mm |  6 ++++++
 include/linux/kobject.h                   |  2 ++
 mm/mm_init.c                              | 16 ++++++++++++++++
 3 files changed, 24 insertions(+)
 create mode 100644 Documentation/ABI/testing/sysfs-kernel-mm

diff --git a/Documentation/ABI/testing/sysfs-kernel-mm b/Documentation/ABI/testing/sysfs-kernel-mm
new file mode 100644
index 000000000000..190d523ac159
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-kernel-mm
@@ -0,0 +1,6 @@
+What:		/sys/kernel/mm
+Date:		July 2008
+Contact:	Nishanth Aravamudan <nacc@us.ibm.com>, VM maintainers
+Description:
+		/sys/kernel/mm/ should contain any and all VM
+		related information in /sys/kernel/.
diff --git a/include/linux/kobject.h b/include/linux/kobject.h
index 60f0d418ae32..5437ac0276e2 100644
--- a/include/linux/kobject.h
+++ b/include/linux/kobject.h
@@ -186,6 +186,8 @@ extern struct kobject *kset_find_obj(struct kset *, const char *);
 
 /* The global /sys/kernel/ kobject for people to chain off of */
 extern struct kobject *kernel_kobj;
+/* The global /sys/kernel/mm/ kobject for people to chain off of */
+extern struct kobject *mm_kobj;
 /* The global /sys/hypervisor/ kobject for people to chain off of */
 extern struct kobject *hypervisor_kobj;
 /* The global /sys/power/ kobject for people to chain off of */
diff --git a/mm/mm_init.c b/mm/mm_init.c
index eaf0d3b47099..c6af41ea9994 100644
--- a/mm/mm_init.c
+++ b/mm/mm_init.c
@@ -7,6 +7,8 @@
  */
 #include <linux/kernel.h>
 #include <linux/init.h>
+#include <linux/kobject.h>
+#include <linux/module.h>
 #include "internal.h"
 
 #ifdef CONFIG_DEBUG_MEMORY_INIT
@@ -134,3 +136,17 @@ static __init int set_mminit_loglevel(char *str)
 }
 early_param("mminit_loglevel", set_mminit_loglevel);
 #endif /* CONFIG_DEBUG_MEMORY_INIT */
+
+struct kobject *mm_kobj;
+EXPORT_SYMBOL_GPL(mm_kobj);
+
+static int __init mm_sysfs_init(void)
+{
+	mm_kobj = kobject_create_and_add("mm", kernel_kobj);
+	if (!mm_kobj)
+		return -ENOMEM;
+
+	return 0;
+}
+
+__initcall(mm_sysfs_init);

From b7ba30c679ed1eb7ed3ed8f281f6493282042bd4 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 23 Jul 2008 21:27:40 -0700
Subject: [PATCH 0249/1171] hugetlb: factor out prep_new_huge_page

Needed to avoid code duplication in follow up patches.

Acked-by: Adam Litke <agl@us.ibm.com>
Acked-by: Nishanth Aravamudan <nacc@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/hugetlb.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index eda9642254a0..32dff4290c66 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -513,6 +513,16 @@ static int adjust_pool_surplus(int delta)
 	return ret;
 }
 
+static void prep_new_huge_page(struct page *page, int nid)
+{
+	set_compound_page_dtor(page, free_huge_page);
+	spin_lock(&hugetlb_lock);
+	nr_huge_pages++;
+	nr_huge_pages_node[nid]++;
+	spin_unlock(&hugetlb_lock);
+	put_page(page); /* free it into the hugepage allocator */
+}
+
 static struct page *alloc_fresh_huge_page_node(int nid)
 {
 	struct page *page;
@@ -526,12 +536,7 @@ static struct page *alloc_fresh_huge_page_node(int nid)
 			__free_pages(page, HUGETLB_PAGE_ORDER);
 			return NULL;
 		}
-		set_compound_page_dtor(page, free_huge_page);
-		spin_lock(&hugetlb_lock);
-		nr_huge_pages++;
-		nr_huge_pages_node[nid]++;
-		spin_unlock(&hugetlb_lock);
-		put_page(page); /* free it into the hugepage allocator */
+		prep_new_huge_page(page, nid);
 	}
 
 	return page;

From a5516438959d90b071ff0a484ce4f3f523dc3152 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 23 Jul 2008 21:27:41 -0700
Subject: [PATCH 0250/1171] hugetlb: modular state for hugetlb page size

The goal of this patchset is to support multiple hugetlb page sizes.  This
is achieved by introducing a new struct hstate structure, which
encapsulates the important hugetlb state and constants (eg.  huge page
size, number of huge pages currently allocated, etc).

The hstate structure is then passed around the code which requires these
fields, they will do the right thing regardless of the exact hstate they
are operating on.

This patch adds the hstate structure, with a single global instance of it
(default_hstate), and does the basic work of converting hugetlb to use the
hstate.

Future patches will add more hstate structures to allow for different
hugetlbfs mounts to have different page sizes.

[akpm@linux-foundation.org: coding-style fixes]
Acked-by: Adam Litke <agl@us.ibm.com>
Acked-by: Nishanth Aravamudan <nacc@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/ia64/mm/hugetlbpage.c    |   7 +-
 arch/powerpc/mm/hugetlbpage.c |   3 +-
 arch/s390/mm/hugetlbpage.c    |   3 +-
 arch/sh/mm/hugetlbpage.c      |   3 +-
 arch/sparc64/mm/hugetlbpage.c |   5 +-
 arch/x86/mm/hugetlbpage.c     |   5 +-
 fs/hugetlbfs/inode.c          |  52 +++--
 include/asm-ia64/hugetlb.h    |   3 +-
 include/asm-powerpc/hugetlb.h |   3 +-
 include/asm-s390/hugetlb.h    |   3 +-
 include/asm-sh/hugetlb.h      |   3 +-
 include/asm-sparc/hugetlb.h   |   3 +-
 include/asm-x86/hugetlb.h     |   8 +-
 include/linux/hugetlb.h       |  88 +++++++-
 ipc/shm.c                     |   3 +-
 mm/hugetlb.c                  | 370 +++++++++++++++++++---------------
 mm/memory.c                   |   2 +-
 mm/mempolicy.c                |   9 +-
 mm/mmap.c                     |   3 +-
 19 files changed, 357 insertions(+), 219 deletions(-)

diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c
index cd49e2860eef..6170f097d255 100644
--- a/arch/ia64/mm/hugetlbpage.c
+++ b/arch/ia64/mm/hugetlbpage.c
@@ -24,7 +24,7 @@
 unsigned int hpage_shift=HPAGE_SHIFT_DEFAULT;
 
 pte_t *
-huge_pte_alloc (struct mm_struct *mm, unsigned long addr)
+huge_pte_alloc(struct mm_struct *mm, unsigned long addr, unsigned long sz)
 {
 	unsigned long taddr = htlbpage_to_page(addr);
 	pgd_t *pgd;
@@ -75,7 +75,8 @@ int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep)
  * Don't actually need to do any preparation, but need to make sure
  * the address is in the right region.
  */
-int prepare_hugepage_range(unsigned long addr, unsigned long len)
+int prepare_hugepage_range(struct file *file,
+			unsigned long addr, unsigned long len)
 {
 	if (len & ~HPAGE_MASK)
 		return -EINVAL;
@@ -149,7 +150,7 @@ unsigned long hugetlb_get_unmapped_area(struct file *file, unsigned long addr, u
 
 	/* Handle MAP_FIXED */
 	if (flags & MAP_FIXED) {
-		if (prepare_hugepage_range(addr, len))
+		if (prepare_hugepage_range(file, addr, len))
 			return -EINVAL;
 		return addr;
 	}
diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c
index 1a96cc891cf5..c94dc71af989 100644
--- a/arch/powerpc/mm/hugetlbpage.c
+++ b/arch/powerpc/mm/hugetlbpage.c
@@ -128,7 +128,8 @@ pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
 	return NULL;
 }
 
-pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
+pte_t *huge_pte_alloc(struct mm_struct *mm,
+			unsigned long addr, unsigned long sz)
 {
 	pgd_t *pg;
 	pud_t *pu;
diff --git a/arch/s390/mm/hugetlbpage.c b/arch/s390/mm/hugetlbpage.c
index f4b6124fdb75..9162dc84f77f 100644
--- a/arch/s390/mm/hugetlbpage.c
+++ b/arch/s390/mm/hugetlbpage.c
@@ -72,7 +72,8 @@ void arch_release_hugepage(struct page *page)
 	page[1].index = 0;
 }
 
-pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
+pte_t *huge_pte_alloc(struct mm_struct *mm,
+			unsigned long addr, unsigned long sz)
 {
 	pgd_t *pgdp;
 	pud_t *pudp;
diff --git a/arch/sh/mm/hugetlbpage.c b/arch/sh/mm/hugetlbpage.c
index ae8c321d6e2a..2f9dbe0ef4ac 100644
--- a/arch/sh/mm/hugetlbpage.c
+++ b/arch/sh/mm/hugetlbpage.c
@@ -22,7 +22,8 @@
 #include <asm/tlbflush.h>
 #include <asm/cacheflush.h>
 
-pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
+pte_t *huge_pte_alloc(struct mm_struct *mm,
+			unsigned long addr, unsigned long sz)
 {
 	pgd_t *pgd;
 	pud_t *pud;
diff --git a/arch/sparc64/mm/hugetlbpage.c b/arch/sparc64/mm/hugetlbpage.c
index ebefd2a14375..1307b23f6a76 100644
--- a/arch/sparc64/mm/hugetlbpage.c
+++ b/arch/sparc64/mm/hugetlbpage.c
@@ -175,7 +175,7 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
 		return -ENOMEM;
 
 	if (flags & MAP_FIXED) {
-		if (prepare_hugepage_range(addr, len))
+		if (prepare_hugepage_range(file, addr, len))
 			return -EINVAL;
 		return addr;
 	}
@@ -195,7 +195,8 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
 				pgoff, flags);
 }
 
-pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
+pte_t *huge_pte_alloc(struct mm_struct *mm,
+			unsigned long addr, unsigned long sz)
 {
 	pgd_t *pgd;
 	pud_t *pud;
diff --git a/arch/x86/mm/hugetlbpage.c b/arch/x86/mm/hugetlbpage.c
index 0b3d567e686d..52476fde8996 100644
--- a/arch/x86/mm/hugetlbpage.c
+++ b/arch/x86/mm/hugetlbpage.c
@@ -124,7 +124,8 @@ int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep)
 	return 1;
 }
 
-pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
+pte_t *huge_pte_alloc(struct mm_struct *mm,
+			unsigned long addr, unsigned long sz)
 {
 	pgd_t *pgd;
 	pud_t *pud;
@@ -368,7 +369,7 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
 		return -ENOMEM;
 
 	if (flags & MAP_FIXED) {
-		if (prepare_hugepage_range(addr, len))
+		if (prepare_hugepage_range(file, addr, len))
 			return -EINVAL;
 		return addr;
 	}
diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index 428eff5b73f3..516c581b5371 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -80,6 +80,7 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma)
 	struct inode *inode = file->f_path.dentry->d_inode;
 	loff_t len, vma_len;
 	int ret;
+	struct hstate *h = hstate_file(file);
 
 	/*
 	 * vma address alignment (but not the pgoff alignment) has
@@ -92,7 +93,7 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma)
 	vma->vm_flags |= VM_HUGETLB | VM_RESERVED;
 	vma->vm_ops = &hugetlb_vm_ops;
 
-	if (vma->vm_pgoff & ~(HPAGE_MASK >> PAGE_SHIFT))
+	if (vma->vm_pgoff & ~(huge_page_mask(h) >> PAGE_SHIFT))
 		return -EINVAL;
 
 	vma_len = (loff_t)(vma->vm_end - vma->vm_start);
@@ -104,8 +105,8 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma)
 	len = vma_len + ((loff_t)vma->vm_pgoff << PAGE_SHIFT);
 
 	if (hugetlb_reserve_pages(inode,
-				vma->vm_pgoff >> (HPAGE_SHIFT-PAGE_SHIFT),
-				len >> HPAGE_SHIFT, vma))
+				vma->vm_pgoff >> huge_page_order(h),
+				len >> huge_page_shift(h), vma))
 		goto out;
 
 	ret = 0;
@@ -130,20 +131,21 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
 	struct mm_struct *mm = current->mm;
 	struct vm_area_struct *vma;
 	unsigned long start_addr;
+	struct hstate *h = hstate_file(file);
 
-	if (len & ~HPAGE_MASK)
+	if (len & ~huge_page_mask(h))
 		return -EINVAL;
 	if (len > TASK_SIZE)
 		return -ENOMEM;
 
 	if (flags & MAP_FIXED) {
-		if (prepare_hugepage_range(addr, len))
+		if (prepare_hugepage_range(file, addr, len))
 			return -EINVAL;
 		return addr;
 	}
 
 	if (addr) {
-		addr = ALIGN(addr, HPAGE_SIZE);
+		addr = ALIGN(addr, huge_page_size(h));
 		vma = find_vma(mm, addr);
 		if (TASK_SIZE - len >= addr &&
 		    (!vma || addr + len <= vma->vm_start))
@@ -156,7 +158,7 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
 		start_addr = TASK_UNMAPPED_BASE;
 
 full_search:
-	addr = ALIGN(start_addr, HPAGE_SIZE);
+	addr = ALIGN(start_addr, huge_page_size(h));
 
 	for (vma = find_vma(mm, addr); ; vma = vma->vm_next) {
 		/* At this point:  (!vma || addr < vma->vm_end). */
@@ -174,7 +176,7 @@ full_search:
 
 		if (!vma || addr + len <= vma->vm_start)
 			return addr;
-		addr = ALIGN(vma->vm_end, HPAGE_SIZE);
+		addr = ALIGN(vma->vm_end, huge_page_size(h));
 	}
 }
 #endif
@@ -225,10 +227,11 @@ hugetlbfs_read_actor(struct page *page, unsigned long offset,
 static ssize_t hugetlbfs_read(struct file *filp, char __user *buf,
 			      size_t len, loff_t *ppos)
 {
+	struct hstate *h = hstate_file(filp);
 	struct address_space *mapping = filp->f_mapping;
 	struct inode *inode = mapping->host;
-	unsigned long index = *ppos >> HPAGE_SHIFT;
-	unsigned long offset = *ppos & ~HPAGE_MASK;
+	unsigned long index = *ppos >> huge_page_shift(h);
+	unsigned long offset = *ppos & ~huge_page_mask(h);
 	unsigned long end_index;
 	loff_t isize;
 	ssize_t retval = 0;
@@ -243,17 +246,17 @@ static ssize_t hugetlbfs_read(struct file *filp, char __user *buf,
 	if (!isize)
 		goto out;
 
-	end_index = (isize - 1) >> HPAGE_SHIFT;
+	end_index = (isize - 1) >> huge_page_shift(h);
 	for (;;) {
 		struct page *page;
-		int nr, ret;
+		unsigned long nr, ret;
 
 		/* nr is the maximum number of bytes to copy from this page */
-		nr = HPAGE_SIZE;
+		nr = huge_page_size(h);
 		if (index >= end_index) {
 			if (index > end_index)
 				goto out;
-			nr = ((isize - 1) & ~HPAGE_MASK) + 1;
+			nr = ((isize - 1) & ~huge_page_mask(h)) + 1;
 			if (nr <= offset) {
 				goto out;
 			}
@@ -287,8 +290,8 @@ static ssize_t hugetlbfs_read(struct file *filp, char __user *buf,
 		offset += ret;
 		retval += ret;
 		len -= ret;
-		index += offset >> HPAGE_SHIFT;
-		offset &= ~HPAGE_MASK;
+		index += offset >> huge_page_shift(h);
+		offset &= ~huge_page_mask(h);
 
 		if (page)
 			page_cache_release(page);
@@ -298,7 +301,7 @@ static ssize_t hugetlbfs_read(struct file *filp, char __user *buf,
 			break;
 	}
 out:
-	*ppos = ((loff_t)index << HPAGE_SHIFT) + offset;
+	*ppos = ((loff_t)index << huge_page_shift(h)) + offset;
 	mutex_unlock(&inode->i_mutex);
 	return retval;
 }
@@ -339,8 +342,9 @@ static void truncate_huge_page(struct page *page)
 
 static void truncate_hugepages(struct inode *inode, loff_t lstart)
 {
+	struct hstate *h = hstate_inode(inode);
 	struct address_space *mapping = &inode->i_data;
-	const pgoff_t start = lstart >> HPAGE_SHIFT;
+	const pgoff_t start = lstart >> huge_page_shift(h);
 	struct pagevec pvec;
 	pgoff_t next;
 	int i, freed = 0;
@@ -449,8 +453,9 @@ static int hugetlb_vmtruncate(struct inode *inode, loff_t offset)
 {
 	pgoff_t pgoff;
 	struct address_space *mapping = inode->i_mapping;
+	struct hstate *h = hstate_inode(inode);
 
-	BUG_ON(offset & ~HPAGE_MASK);
+	BUG_ON(offset & ~huge_page_mask(h));
 	pgoff = offset >> PAGE_SHIFT;
 
 	i_size_write(inode, offset);
@@ -465,6 +470,7 @@ static int hugetlb_vmtruncate(struct inode *inode, loff_t offset)
 static int hugetlbfs_setattr(struct dentry *dentry, struct iattr *attr)
 {
 	struct inode *inode = dentry->d_inode;
+	struct hstate *h = hstate_inode(inode);
 	int error;
 	unsigned int ia_valid = attr->ia_valid;
 
@@ -476,7 +482,7 @@ static int hugetlbfs_setattr(struct dentry *dentry, struct iattr *attr)
 
 	if (ia_valid & ATTR_SIZE) {
 		error = -EINVAL;
-		if (!(attr->ia_size & ~HPAGE_MASK))
+		if (!(attr->ia_size & ~huge_page_mask(h)))
 			error = hugetlb_vmtruncate(inode, attr->ia_size);
 		if (error)
 			goto out;
@@ -610,9 +616,10 @@ static int hugetlbfs_set_page_dirty(struct page *page)
 static int hugetlbfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 {
 	struct hugetlbfs_sb_info *sbinfo = HUGETLBFS_SB(dentry->d_sb);
+	struct hstate *h = hstate_inode(dentry->d_inode);
 
 	buf->f_type = HUGETLBFS_MAGIC;
-	buf->f_bsize = HPAGE_SIZE;
+	buf->f_bsize = huge_page_size(h);
 	if (sbinfo) {
 		spin_lock(&sbinfo->stat_lock);
 		/* If no limits set, just report 0 for max/free/used
@@ -942,7 +949,8 @@ struct file *hugetlb_file_setup(const char *name, size_t size)
 		goto out_dentry;
 
 	error = -ENOMEM;
-	if (hugetlb_reserve_pages(inode, 0, size >> HPAGE_SHIFT, NULL))
+	if (hugetlb_reserve_pages(inode, 0,
+			size >> huge_page_shift(hstate_inode(inode)), NULL))
 		goto out_inode;
 
 	d_instantiate(dentry, inode);
diff --git a/include/asm-ia64/hugetlb.h b/include/asm-ia64/hugetlb.h
index e9d1e5e2382d..da55c63728e0 100644
--- a/include/asm-ia64/hugetlb.h
+++ b/include/asm-ia64/hugetlb.h
@@ -8,7 +8,8 @@ void hugetlb_free_pgd_range(struct mmu_gather *tlb, unsigned long addr,
 			    unsigned long end, unsigned long floor,
 			    unsigned long ceiling);
 
-int prepare_hugepage_range(unsigned long addr, unsigned long len);
+int prepare_hugepage_range(struct file *file,
+			unsigned long addr, unsigned long len);
 
 static inline int is_hugepage_only_range(struct mm_struct *mm,
 					 unsigned long addr,
diff --git a/include/asm-powerpc/hugetlb.h b/include/asm-powerpc/hugetlb.h
index 0a37aa5ecaa5..ca37c4af27b1 100644
--- a/include/asm-powerpc/hugetlb.h
+++ b/include/asm-powerpc/hugetlb.h
@@ -21,7 +21,8 @@ pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr,
  * If the arch doesn't supply something else, assume that hugepage
  * size aligned regions are ok without further preparation.
  */
-static inline int prepare_hugepage_range(unsigned long addr, unsigned long len)
+static inline int prepare_hugepage_range(struct file *file,
+			unsigned long addr, unsigned long len)
 {
 	if (len & ~HPAGE_MASK)
 		return -EINVAL;
diff --git a/include/asm-s390/hugetlb.h b/include/asm-s390/hugetlb.h
index 600a776f8f75..670a1d1745d2 100644
--- a/include/asm-s390/hugetlb.h
+++ b/include/asm-s390/hugetlb.h
@@ -22,7 +22,8 @@ void set_huge_pte_at(struct mm_struct *mm, unsigned long addr,
  * If the arch doesn't supply something else, assume that hugepage
  * size aligned regions are ok without further preparation.
  */
-static inline int prepare_hugepage_range(unsigned long addr, unsigned long len)
+static inline int prepare_hugepage_range(struct file *file,
+			unsigned long addr, unsigned long len)
 {
 	if (len & ~HPAGE_MASK)
 		return -EINVAL;
diff --git a/include/asm-sh/hugetlb.h b/include/asm-sh/hugetlb.h
index fb30018938c7..967068fb79ac 100644
--- a/include/asm-sh/hugetlb.h
+++ b/include/asm-sh/hugetlb.h
@@ -14,7 +14,8 @@ static inline int is_hugepage_only_range(struct mm_struct *mm,
  * If the arch doesn't supply something else, assume that hugepage
  * size aligned regions are ok without further preparation.
  */
-static inline int prepare_hugepage_range(unsigned long addr, unsigned long len)
+static inline int prepare_hugepage_range(struct file *file,
+			unsigned long addr, unsigned long len)
 {
 	if (len & ~HPAGE_MASK)
 		return -EINVAL;
diff --git a/include/asm-sparc/hugetlb.h b/include/asm-sparc/hugetlb.h
index aeb92374ca3d..177061064ee6 100644
--- a/include/asm-sparc/hugetlb.h
+++ b/include/asm-sparc/hugetlb.h
@@ -22,7 +22,8 @@ static inline int is_hugepage_only_range(struct mm_struct *mm,
  * If the arch doesn't supply something else, assume that hugepage
  * size aligned regions are ok without further preparation.
  */
-static inline int prepare_hugepage_range(unsigned long addr, unsigned long len)
+static inline int prepare_hugepage_range(struct file *file,
+			unsigned long addr, unsigned long len)
 {
 	if (len & ~HPAGE_MASK)
 		return -EINVAL;
diff --git a/include/asm-x86/hugetlb.h b/include/asm-x86/hugetlb.h
index 7eed6e0883bf..439a9acc132d 100644
--- a/include/asm-x86/hugetlb.h
+++ b/include/asm-x86/hugetlb.h
@@ -14,11 +14,13 @@ static inline int is_hugepage_only_range(struct mm_struct *mm,
  * If the arch doesn't supply something else, assume that hugepage
  * size aligned regions are ok without further preparation.
  */
-static inline int prepare_hugepage_range(unsigned long addr, unsigned long len)
+static inline int prepare_hugepage_range(struct file *file,
+			unsigned long addr, unsigned long len)
 {
-	if (len & ~HPAGE_MASK)
+	struct hstate *h = hstate_file(file);
+	if (len & ~huge_page_mask(h))
 		return -EINVAL;
-	if (addr & ~HPAGE_MASK)
+	if (addr & ~huge_page_mask(h))
 		return -EINVAL;
 	return 0;
 }
diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index abbc187193a1..ad2271e11f9b 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -8,7 +8,6 @@
 #include <linux/mempolicy.h>
 #include <linux/shm.h>
 #include <asm/tlbflush.h>
-#include <asm/hugetlb.h>
 
 struct ctl_table;
 
@@ -45,7 +44,8 @@ extern int sysctl_hugetlb_shm_group;
 
 /* arch callbacks */
 
-pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr);
+pte_t *huge_pte_alloc(struct mm_struct *mm,
+			unsigned long addr, unsigned long sz);
 pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr);
 int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep);
 struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address,
@@ -80,7 +80,7 @@ static inline unsigned long hugetlb_total_pages(void)
 #define hugetlb_report_meminfo(buf)		0
 #define hugetlb_report_node_meminfo(n, buf)	0
 #define follow_huge_pmd(mm, addr, pmd, write)	NULL
-#define prepare_hugepage_range(addr,len)	(-EINVAL)
+#define prepare_hugepage_range(file, addr, len)	(-EINVAL)
 #define pmd_huge(x)	0
 #define is_hugepage_only_range(mm, addr, len)	0
 #define hugetlb_free_pgd_range(tlb, addr, end, floor, ceiling) ({BUG(); 0; })
@@ -134,8 +134,6 @@ struct file *hugetlb_file_setup(const char *name, size_t);
 int hugetlb_get_quota(struct address_space *mapping, long delta);
 void hugetlb_put_quota(struct address_space *mapping, long delta);
 
-#define BLOCKS_PER_HUGEPAGE	(HPAGE_SIZE / 512)
-
 static inline int is_file_hugepages(struct file *file)
 {
 	if (file->f_op == &hugetlbfs_file_operations)
@@ -164,4 +162,84 @@ unsigned long hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
 					unsigned long flags);
 #endif /* HAVE_ARCH_HUGETLB_UNMAPPED_AREA */
 
+#ifdef CONFIG_HUGETLB_PAGE
+
+/* Defines one hugetlb page size */
+struct hstate {
+	int hugetlb_next_nid;
+	unsigned int order;
+	unsigned long mask;
+	unsigned long max_huge_pages;
+	unsigned long nr_huge_pages;
+	unsigned long free_huge_pages;
+	unsigned long resv_huge_pages;
+	unsigned long surplus_huge_pages;
+	unsigned long nr_overcommit_huge_pages;
+	struct list_head hugepage_freelists[MAX_NUMNODES];
+	unsigned int nr_huge_pages_node[MAX_NUMNODES];
+	unsigned int free_huge_pages_node[MAX_NUMNODES];
+	unsigned int surplus_huge_pages_node[MAX_NUMNODES];
+};
+
+extern struct hstate default_hstate;
+
+static inline struct hstate *hstate_vma(struct vm_area_struct *vma)
+{
+	return &default_hstate;
+}
+
+static inline struct hstate *hstate_file(struct file *f)
+{
+	return &default_hstate;
+}
+
+static inline struct hstate *hstate_inode(struct inode *i)
+{
+	return &default_hstate;
+}
+
+static inline unsigned long huge_page_size(struct hstate *h)
+{
+	return (unsigned long)PAGE_SIZE << h->order;
+}
+
+static inline unsigned long huge_page_mask(struct hstate *h)
+{
+	return h->mask;
+}
+
+static inline unsigned int huge_page_order(struct hstate *h)
+{
+	return h->order;
+}
+
+static inline unsigned huge_page_shift(struct hstate *h)
+{
+	return h->order + PAGE_SHIFT;
+}
+
+static inline unsigned int pages_per_huge_page(struct hstate *h)
+{
+	return 1 << h->order;
+}
+
+static inline unsigned int blocks_per_huge_page(struct hstate *h)
+{
+	return huge_page_size(h) / 512;
+}
+
+#include <asm/hugetlb.h>
+
+#else
+struct hstate {};
+#define hstate_file(f) NULL
+#define hstate_vma(v) NULL
+#define hstate_inode(i) NULL
+#define huge_page_size(h) PAGE_SIZE
+#define huge_page_mask(h) PAGE_MASK
+#define huge_page_order(h) 0
+#define huge_page_shift(h) PAGE_SHIFT
+#define pages_per_huge_page(h) 1
+#endif
+
 #endif /* _LINUX_HUGETLB_H */
diff --git a/ipc/shm.c b/ipc/shm.c
index 790240cd067f..a726aebce7d7 100644
--- a/ipc/shm.c
+++ b/ipc/shm.c
@@ -577,7 +577,8 @@ static void shm_get_stat(struct ipc_namespace *ns, unsigned long *rss,
 
 		if (is_file_hugepages(shp->shm_file)) {
 			struct address_space *mapping = inode->i_mapping;
-			*rss += (HPAGE_SIZE/PAGE_SIZE)*mapping->nrpages;
+			struct hstate *h = hstate_file(shp->shm_file);
+			*rss += pages_per_huge_page(h) * mapping->nrpages;
 		} else {
 			struct shmem_inode_info *info = SHMEM_I(inode);
 			spin_lock(&info->lock);
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 32dff4290c66..0d8153e25f09 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -22,18 +22,12 @@
 #include "internal.h"
 
 const unsigned long hugetlb_zero = 0, hugetlb_infinity = ~0UL;
-static unsigned long nr_huge_pages, free_huge_pages, resv_huge_pages;
-static unsigned long surplus_huge_pages;
-static unsigned long nr_overcommit_huge_pages;
 unsigned long max_huge_pages;
 unsigned long sysctl_overcommit_huge_pages;
-static struct list_head hugepage_freelists[MAX_NUMNODES];
-static unsigned int nr_huge_pages_node[MAX_NUMNODES];
-static unsigned int free_huge_pages_node[MAX_NUMNODES];
-static unsigned int surplus_huge_pages_node[MAX_NUMNODES];
 static gfp_t htlb_alloc_mask = GFP_HIGHUSER;
 unsigned long hugepages_treat_as_movable;
-static int hugetlb_next_nid;
+
+struct hstate default_hstate;
 
 /*
  * Protects updates to hugepage_freelists, nr_huge_pages, and free_huge_pages
@@ -203,11 +197,11 @@ static long region_count(struct list_head *head, long f, long t)
  * Convert the address within this vma to the page offset within
  * the mapping, in pagecache page units; huge pages here.
  */
-static pgoff_t vma_hugecache_offset(struct vm_area_struct *vma,
-					unsigned long address)
+static pgoff_t vma_hugecache_offset(struct hstate *h,
+			struct vm_area_struct *vma, unsigned long address)
 {
-	return ((address - vma->vm_start) >> HPAGE_SHIFT) +
-			(vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT));
+	return ((address - vma->vm_start) >> huge_page_shift(h)) +
+			(vma->vm_pgoff >> huge_page_order(h));
 }
 
 /*
@@ -309,20 +303,21 @@ static int is_vma_resv_set(struct vm_area_struct *vma, unsigned long flag)
 }
 
 /* Decrement the reserved pages in the hugepage pool by one */
-static void decrement_hugepage_resv_vma(struct vm_area_struct *vma)
+static void decrement_hugepage_resv_vma(struct hstate *h,
+			struct vm_area_struct *vma)
 {
 	if (vma->vm_flags & VM_NORESERVE)
 		return;
 
 	if (vma->vm_flags & VM_SHARED) {
 		/* Shared mappings always use reserves */
-		resv_huge_pages--;
+		h->resv_huge_pages--;
 	} else if (is_vma_resv_set(vma, HPAGE_RESV_OWNER)) {
 		/*
 		 * Only the process that called mmap() has reserves for
 		 * private mappings.
 		 */
-		resv_huge_pages--;
+		h->resv_huge_pages--;
 	}
 }
 
@@ -344,12 +339,13 @@ static int vma_has_private_reserves(struct vm_area_struct *vma)
 	return 1;
 }
 
-static void clear_huge_page(struct page *page, unsigned long addr)
+static void clear_huge_page(struct page *page,
+			unsigned long addr, unsigned long sz)
 {
 	int i;
 
 	might_sleep();
-	for (i = 0; i < (HPAGE_SIZE/PAGE_SIZE); i++) {
+	for (i = 0; i < sz/PAGE_SIZE; i++) {
 		cond_resched();
 		clear_user_highpage(page + i, addr + i * PAGE_SIZE);
 	}
@@ -359,41 +355,43 @@ static void copy_huge_page(struct page *dst, struct page *src,
 			   unsigned long addr, struct vm_area_struct *vma)
 {
 	int i;
+	struct hstate *h = hstate_vma(vma);
 
 	might_sleep();
-	for (i = 0; i < HPAGE_SIZE/PAGE_SIZE; i++) {
+	for (i = 0; i < pages_per_huge_page(h); i++) {
 		cond_resched();
 		copy_user_highpage(dst + i, src + i, addr + i*PAGE_SIZE, vma);
 	}
 }
 
-static void enqueue_huge_page(struct page *page)
+static void enqueue_huge_page(struct hstate *h, struct page *page)
 {
 	int nid = page_to_nid(page);
-	list_add(&page->lru, &hugepage_freelists[nid]);
-	free_huge_pages++;
-	free_huge_pages_node[nid]++;
+	list_add(&page->lru, &h->hugepage_freelists[nid]);
+	h->free_huge_pages++;
+	h->free_huge_pages_node[nid]++;
 }
 
-static struct page *dequeue_huge_page(void)
+static struct page *dequeue_huge_page(struct hstate *h)
 {
 	int nid;
 	struct page *page = NULL;
 
 	for (nid = 0; nid < MAX_NUMNODES; ++nid) {
-		if (!list_empty(&hugepage_freelists[nid])) {
-			page = list_entry(hugepage_freelists[nid].next,
+		if (!list_empty(&h->hugepage_freelists[nid])) {
+			page = list_entry(h->hugepage_freelists[nid].next,
 					  struct page, lru);
 			list_del(&page->lru);
-			free_huge_pages--;
-			free_huge_pages_node[nid]--;
+			h->free_huge_pages--;
+			h->free_huge_pages_node[nid]--;
 			break;
 		}
 	}
 	return page;
 }
 
-static struct page *dequeue_huge_page_vma(struct vm_area_struct *vma,
+static struct page *dequeue_huge_page_vma(struct hstate *h,
+				struct vm_area_struct *vma,
 				unsigned long address, int avoid_reserve)
 {
 	int nid;
@@ -411,26 +409,26 @@ static struct page *dequeue_huge_page_vma(struct vm_area_struct *vma,
 	 * not "stolen". The child may still get SIGKILLed
 	 */
 	if (!vma_has_private_reserves(vma) &&
-			free_huge_pages - resv_huge_pages == 0)
+			h->free_huge_pages - h->resv_huge_pages == 0)
 		return NULL;
 
 	/* If reserves cannot be used, ensure enough pages are in the pool */
-	if (avoid_reserve && free_huge_pages - resv_huge_pages == 0)
+	if (avoid_reserve && h->free_huge_pages - h->resv_huge_pages == 0)
 		return NULL;
 
 	for_each_zone_zonelist_nodemask(zone, z, zonelist,
 						MAX_NR_ZONES - 1, nodemask) {
 		nid = zone_to_nid(zone);
 		if (cpuset_zone_allowed_softwall(zone, htlb_alloc_mask) &&
-		    !list_empty(&hugepage_freelists[nid])) {
-			page = list_entry(hugepage_freelists[nid].next,
+		    !list_empty(&h->hugepage_freelists[nid])) {
+			page = list_entry(h->hugepage_freelists[nid].next,
 					  struct page, lru);
 			list_del(&page->lru);
-			free_huge_pages--;
-			free_huge_pages_node[nid]--;
+			h->free_huge_pages--;
+			h->free_huge_pages_node[nid]--;
 
 			if (!avoid_reserve)
-				decrement_hugepage_resv_vma(vma);
+				decrement_hugepage_resv_vma(h, vma);
 
 			break;
 		}
@@ -439,12 +437,13 @@ static struct page *dequeue_huge_page_vma(struct vm_area_struct *vma,
 	return page;
 }
 
-static void update_and_free_page(struct page *page)
+static void update_and_free_page(struct hstate *h, struct page *page)
 {
 	int i;
-	nr_huge_pages--;
-	nr_huge_pages_node[page_to_nid(page)]--;
-	for (i = 0; i < (HPAGE_SIZE / PAGE_SIZE); i++) {
+
+	h->nr_huge_pages--;
+	h->nr_huge_pages_node[page_to_nid(page)]--;
+	for (i = 0; i < pages_per_huge_page(h); i++) {
 		page[i].flags &= ~(1 << PG_locked | 1 << PG_error | 1 << PG_referenced |
 				1 << PG_dirty | 1 << PG_active | 1 << PG_reserved |
 				1 << PG_private | 1<< PG_writeback);
@@ -452,11 +451,16 @@ static void update_and_free_page(struct page *page)
 	set_compound_page_dtor(page, NULL);
 	set_page_refcounted(page);
 	arch_release_hugepage(page);
-	__free_pages(page, HUGETLB_PAGE_ORDER);
+	__free_pages(page, huge_page_order(h));
 }
 
 static void free_huge_page(struct page *page)
 {
+	/*
+	 * Can't pass hstate in here because it is called from the
+	 * compound page destructor.
+	 */
+	struct hstate *h = &default_hstate;
 	int nid = page_to_nid(page);
 	struct address_space *mapping;
 
@@ -466,12 +470,12 @@ static void free_huge_page(struct page *page)
 	INIT_LIST_HEAD(&page->lru);
 
 	spin_lock(&hugetlb_lock);
-	if (surplus_huge_pages_node[nid]) {
-		update_and_free_page(page);
-		surplus_huge_pages--;
-		surplus_huge_pages_node[nid]--;
+	if (h->surplus_huge_pages_node[nid]) {
+		update_and_free_page(h, page);
+		h->surplus_huge_pages--;
+		h->surplus_huge_pages_node[nid]--;
 	} else {
-		enqueue_huge_page(page);
+		enqueue_huge_page(h, page);
 	}
 	spin_unlock(&hugetlb_lock);
 	if (mapping)
@@ -483,7 +487,7 @@ static void free_huge_page(struct page *page)
  * balanced by operating on them in a round-robin fashion.
  * Returns 1 if an adjustment was made.
  */
-static int adjust_pool_surplus(int delta)
+static int adjust_pool_surplus(struct hstate *h, int delta)
 {
 	static int prev_nid;
 	int nid = prev_nid;
@@ -496,15 +500,15 @@ static int adjust_pool_surplus(int delta)
 			nid = first_node(node_online_map);
 
 		/* To shrink on this node, there must be a surplus page */
-		if (delta < 0 && !surplus_huge_pages_node[nid])
+		if (delta < 0 && !h->surplus_huge_pages_node[nid])
 			continue;
 		/* Surplus cannot exceed the total number of pages */
-		if (delta > 0 && surplus_huge_pages_node[nid] >=
-						nr_huge_pages_node[nid])
+		if (delta > 0 && h->surplus_huge_pages_node[nid] >=
+						h->nr_huge_pages_node[nid])
 			continue;
 
-		surplus_huge_pages += delta;
-		surplus_huge_pages_node[nid] += delta;
+		h->surplus_huge_pages += delta;
+		h->surplus_huge_pages_node[nid] += delta;
 		ret = 1;
 		break;
 	} while (nid != prev_nid);
@@ -513,46 +517,46 @@ static int adjust_pool_surplus(int delta)
 	return ret;
 }
 
-static void prep_new_huge_page(struct page *page, int nid)
+static void prep_new_huge_page(struct hstate *h, struct page *page, int nid)
 {
 	set_compound_page_dtor(page, free_huge_page);
 	spin_lock(&hugetlb_lock);
-	nr_huge_pages++;
-	nr_huge_pages_node[nid]++;
+	h->nr_huge_pages++;
+	h->nr_huge_pages_node[nid]++;
 	spin_unlock(&hugetlb_lock);
 	put_page(page); /* free it into the hugepage allocator */
 }
 
-static struct page *alloc_fresh_huge_page_node(int nid)
+static struct page *alloc_fresh_huge_page_node(struct hstate *h, int nid)
 {
 	struct page *page;
 
 	page = alloc_pages_node(nid,
 		htlb_alloc_mask|__GFP_COMP|__GFP_THISNODE|
 						__GFP_REPEAT|__GFP_NOWARN,
-		HUGETLB_PAGE_ORDER);
+		huge_page_order(h));
 	if (page) {
 		if (arch_prepare_hugepage(page)) {
 			__free_pages(page, HUGETLB_PAGE_ORDER);
 			return NULL;
 		}
-		prep_new_huge_page(page, nid);
+		prep_new_huge_page(h, page, nid);
 	}
 
 	return page;
 }
 
-static int alloc_fresh_huge_page(void)
+static int alloc_fresh_huge_page(struct hstate *h)
 {
 	struct page *page;
 	int start_nid;
 	int next_nid;
 	int ret = 0;
 
-	start_nid = hugetlb_next_nid;
+	start_nid = h->hugetlb_next_nid;
 
 	do {
-		page = alloc_fresh_huge_page_node(hugetlb_next_nid);
+		page = alloc_fresh_huge_page_node(h, h->hugetlb_next_nid);
 		if (page)
 			ret = 1;
 		/*
@@ -566,11 +570,11 @@ static int alloc_fresh_huge_page(void)
 		 * if we just successfully allocated a hugepage so that
 		 * the next caller gets hugepages on the next node.
 		 */
-		next_nid = next_node(hugetlb_next_nid, node_online_map);
+		next_nid = next_node(h->hugetlb_next_nid, node_online_map);
 		if (next_nid == MAX_NUMNODES)
 			next_nid = first_node(node_online_map);
-		hugetlb_next_nid = next_nid;
-	} while (!page && hugetlb_next_nid != start_nid);
+		h->hugetlb_next_nid = next_nid;
+	} while (!page && h->hugetlb_next_nid != start_nid);
 
 	if (ret)
 		count_vm_event(HTLB_BUDDY_PGALLOC);
@@ -580,8 +584,8 @@ static int alloc_fresh_huge_page(void)
 	return ret;
 }
 
-static struct page *alloc_buddy_huge_page(struct vm_area_struct *vma,
-						unsigned long address)
+static struct page *alloc_buddy_huge_page(struct hstate *h,
+			struct vm_area_struct *vma, unsigned long address)
 {
 	struct page *page;
 	unsigned int nid;
@@ -610,18 +614,18 @@ static struct page *alloc_buddy_huge_page(struct vm_area_struct *vma,
 	 * per-node value is checked there.
 	 */
 	spin_lock(&hugetlb_lock);
-	if (surplus_huge_pages >= nr_overcommit_huge_pages) {
+	if (h->surplus_huge_pages >= h->nr_overcommit_huge_pages) {
 		spin_unlock(&hugetlb_lock);
 		return NULL;
 	} else {
-		nr_huge_pages++;
-		surplus_huge_pages++;
+		h->nr_huge_pages++;
+		h->surplus_huge_pages++;
 	}
 	spin_unlock(&hugetlb_lock);
 
 	page = alloc_pages(htlb_alloc_mask|__GFP_COMP|
 					__GFP_REPEAT|__GFP_NOWARN,
-					HUGETLB_PAGE_ORDER);
+					huge_page_order(h));
 
 	spin_lock(&hugetlb_lock);
 	if (page) {
@@ -636,12 +640,12 @@ static struct page *alloc_buddy_huge_page(struct vm_area_struct *vma,
 		/*
 		 * We incremented the global counters already
 		 */
-		nr_huge_pages_node[nid]++;
-		surplus_huge_pages_node[nid]++;
+		h->nr_huge_pages_node[nid]++;
+		h->surplus_huge_pages_node[nid]++;
 		__count_vm_event(HTLB_BUDDY_PGALLOC);
 	} else {
-		nr_huge_pages--;
-		surplus_huge_pages--;
+		h->nr_huge_pages--;
+		h->surplus_huge_pages--;
 		__count_vm_event(HTLB_BUDDY_PGALLOC_FAIL);
 	}
 	spin_unlock(&hugetlb_lock);
@@ -653,16 +657,16 @@ static struct page *alloc_buddy_huge_page(struct vm_area_struct *vma,
  * Increase the hugetlb pool such that it can accomodate a reservation
  * of size 'delta'.
  */
-static int gather_surplus_pages(int delta)
+static int gather_surplus_pages(struct hstate *h, int delta)
 {
 	struct list_head surplus_list;
 	struct page *page, *tmp;
 	int ret, i;
 	int needed, allocated;
 
-	needed = (resv_huge_pages + delta) - free_huge_pages;
+	needed = (h->resv_huge_pages + delta) - h->free_huge_pages;
 	if (needed <= 0) {
-		resv_huge_pages += delta;
+		h->resv_huge_pages += delta;
 		return 0;
 	}
 
@@ -673,7 +677,7 @@ static int gather_surplus_pages(int delta)
 retry:
 	spin_unlock(&hugetlb_lock);
 	for (i = 0; i < needed; i++) {
-		page = alloc_buddy_huge_page(NULL, 0);
+		page = alloc_buddy_huge_page(h, NULL, 0);
 		if (!page) {
 			/*
 			 * We were not able to allocate enough pages to
@@ -694,7 +698,8 @@ retry:
 	 * because either resv_huge_pages or free_huge_pages may have changed.
 	 */
 	spin_lock(&hugetlb_lock);
-	needed = (resv_huge_pages + delta) - (free_huge_pages + allocated);
+	needed = (h->resv_huge_pages + delta) -
+			(h->free_huge_pages + allocated);
 	if (needed > 0)
 		goto retry;
 
@@ -707,7 +712,7 @@ retry:
 	 * before they are reserved.
 	 */
 	needed += allocated;
-	resv_huge_pages += delta;
+	h->resv_huge_pages += delta;
 	ret = 0;
 free:
 	/* Free the needed pages to the hugetlb pool */
@@ -715,7 +720,7 @@ free:
 		if ((--needed) < 0)
 			break;
 		list_del(&page->lru);
-		enqueue_huge_page(page);
+		enqueue_huge_page(h, page);
 	}
 
 	/* Free unnecessary surplus pages to the buddy allocator */
@@ -743,7 +748,8 @@ free:
  * allocated to satisfy the reservation must be explicitly freed if they were
  * never used.
  */
-static void return_unused_surplus_pages(unsigned long unused_resv_pages)
+static void return_unused_surplus_pages(struct hstate *h,
+					unsigned long unused_resv_pages)
 {
 	static int nid = -1;
 	struct page *page;
@@ -758,27 +764,27 @@ static void return_unused_surplus_pages(unsigned long unused_resv_pages)
 	unsigned long remaining_iterations = num_online_nodes();
 
 	/* Uncommit the reservation */
-	resv_huge_pages -= unused_resv_pages;
+	h->resv_huge_pages -= unused_resv_pages;
 
-	nr_pages = min(unused_resv_pages, surplus_huge_pages);
+	nr_pages = min(unused_resv_pages, h->surplus_huge_pages);
 
 	while (remaining_iterations-- && nr_pages) {
 		nid = next_node(nid, node_online_map);
 		if (nid == MAX_NUMNODES)
 			nid = first_node(node_online_map);
 
-		if (!surplus_huge_pages_node[nid])
+		if (!h->surplus_huge_pages_node[nid])
 			continue;
 
-		if (!list_empty(&hugepage_freelists[nid])) {
-			page = list_entry(hugepage_freelists[nid].next,
+		if (!list_empty(&h->hugepage_freelists[nid])) {
+			page = list_entry(h->hugepage_freelists[nid].next,
 					  struct page, lru);
 			list_del(&page->lru);
-			update_and_free_page(page);
-			free_huge_pages--;
-			free_huge_pages_node[nid]--;
-			surplus_huge_pages--;
-			surplus_huge_pages_node[nid]--;
+			update_and_free_page(h, page);
+			h->free_huge_pages--;
+			h->free_huge_pages_node[nid]--;
+			h->surplus_huge_pages--;
+			h->surplus_huge_pages_node[nid]--;
 			nr_pages--;
 			remaining_iterations = num_online_nodes();
 		}
@@ -794,13 +800,14 @@ static void return_unused_surplus_pages(unsigned long unused_resv_pages)
  * an instantiated the change should be committed via vma_commit_reservation.
  * No action is required on failure.
  */
-static int vma_needs_reservation(struct vm_area_struct *vma, unsigned long addr)
+static int vma_needs_reservation(struct hstate *h,
+			struct vm_area_struct *vma, unsigned long addr)
 {
 	struct address_space *mapping = vma->vm_file->f_mapping;
 	struct inode *inode = mapping->host;
 
 	if (vma->vm_flags & VM_SHARED) {
-		pgoff_t idx = vma_hugecache_offset(vma, addr);
+		pgoff_t idx = vma_hugecache_offset(h, vma, addr);
 		return region_chg(&inode->i_mapping->private_list,
 							idx, idx + 1);
 
@@ -809,7 +816,7 @@ static int vma_needs_reservation(struct vm_area_struct *vma, unsigned long addr)
 
 	} else  {
 		int err;
-		pgoff_t idx = vma_hugecache_offset(vma, addr);
+		pgoff_t idx = vma_hugecache_offset(h, vma, addr);
 		struct resv_map *reservations = vma_resv_map(vma);
 
 		err = region_chg(&reservations->regions, idx, idx + 1);
@@ -818,18 +825,18 @@ static int vma_needs_reservation(struct vm_area_struct *vma, unsigned long addr)
 		return 0;
 	}
 }
-static void vma_commit_reservation(struct vm_area_struct *vma,
-							unsigned long addr)
+static void vma_commit_reservation(struct hstate *h,
+			struct vm_area_struct *vma, unsigned long addr)
 {
 	struct address_space *mapping = vma->vm_file->f_mapping;
 	struct inode *inode = mapping->host;
 
 	if (vma->vm_flags & VM_SHARED) {
-		pgoff_t idx = vma_hugecache_offset(vma, addr);
+		pgoff_t idx = vma_hugecache_offset(h, vma, addr);
 		region_add(&inode->i_mapping->private_list, idx, idx + 1);
 
 	} else if (is_vma_resv_set(vma, HPAGE_RESV_OWNER)) {
-		pgoff_t idx = vma_hugecache_offset(vma, addr);
+		pgoff_t idx = vma_hugecache_offset(h, vma, addr);
 		struct resv_map *reservations = vma_resv_map(vma);
 
 		/* Mark this page used in the map. */
@@ -840,6 +847,7 @@ static void vma_commit_reservation(struct vm_area_struct *vma,
 static struct page *alloc_huge_page(struct vm_area_struct *vma,
 				    unsigned long addr, int avoid_reserve)
 {
+	struct hstate *h = hstate_vma(vma);
 	struct page *page;
 	struct address_space *mapping = vma->vm_file->f_mapping;
 	struct inode *inode = mapping->host;
@@ -852,7 +860,7 @@ static struct page *alloc_huge_page(struct vm_area_struct *vma,
 	 * MAP_NORESERVE mappings may also need pages and quota allocated
 	 * if no reserve mapping overlaps.
 	 */
-	chg = vma_needs_reservation(vma, addr);
+	chg = vma_needs_reservation(h, vma, addr);
 	if (chg < 0)
 		return ERR_PTR(chg);
 	if (chg)
@@ -860,11 +868,11 @@ static struct page *alloc_huge_page(struct vm_area_struct *vma,
 			return ERR_PTR(-ENOSPC);
 
 	spin_lock(&hugetlb_lock);
-	page = dequeue_huge_page_vma(vma, addr, avoid_reserve);
+	page = dequeue_huge_page_vma(h, vma, addr, avoid_reserve);
 	spin_unlock(&hugetlb_lock);
 
 	if (!page) {
-		page = alloc_buddy_huge_page(vma, addr);
+		page = alloc_buddy_huge_page(h, vma, addr);
 		if (!page) {
 			hugetlb_put_quota(inode->i_mapping, chg);
 			return ERR_PTR(-VM_FAULT_OOM);
@@ -874,7 +882,7 @@ static struct page *alloc_huge_page(struct vm_area_struct *vma,
 	set_page_refcounted(page);
 	set_page_private(page, (unsigned long) mapping);
 
-	vma_commit_reservation(vma, addr);
+	vma_commit_reservation(h, vma, addr);
 
 	return page;
 }
@@ -882,21 +890,28 @@ static struct page *alloc_huge_page(struct vm_area_struct *vma,
 static int __init hugetlb_init(void)
 {
 	unsigned long i;
+	struct hstate *h = &default_hstate;
 
 	if (HPAGE_SHIFT == 0)
 		return 0;
 
-	for (i = 0; i < MAX_NUMNODES; ++i)
-		INIT_LIST_HEAD(&hugepage_freelists[i]);
+	if (!h->order) {
+		h->order = HPAGE_SHIFT - PAGE_SHIFT;
+		h->mask = HPAGE_MASK;
+	}
 
-	hugetlb_next_nid = first_node(node_online_map);
+	for (i = 0; i < MAX_NUMNODES; ++i)
+		INIT_LIST_HEAD(&h->hugepage_freelists[i]);
+
+	h->hugetlb_next_nid = first_node(node_online_map);
 
 	for (i = 0; i < max_huge_pages; ++i) {
-		if (!alloc_fresh_huge_page())
+		if (!alloc_fresh_huge_page(h))
 			break;
 	}
-	max_huge_pages = free_huge_pages = nr_huge_pages = i;
-	printk("Total HugeTLB memory allocated, %ld\n", free_huge_pages);
+	max_huge_pages = h->free_huge_pages = h->nr_huge_pages = i;
+	printk(KERN_INFO "Total HugeTLB memory allocated, %ld\n",
+			h->free_huge_pages);
 	return 0;
 }
 module_init(hugetlb_init);
@@ -922,34 +937,36 @@ static unsigned int cpuset_mems_nr(unsigned int *array)
 
 #ifdef CONFIG_SYSCTL
 #ifdef CONFIG_HIGHMEM
-static void try_to_free_low(unsigned long count)
+static void try_to_free_low(struct hstate *h, unsigned long count)
 {
 	int i;
 
 	for (i = 0; i < MAX_NUMNODES; ++i) {
 		struct page *page, *next;
-		list_for_each_entry_safe(page, next, &hugepage_freelists[i], lru) {
-			if (count >= nr_huge_pages)
+		struct list_head *freel = &h->hugepage_freelists[i];
+		list_for_each_entry_safe(page, next, freel, lru) {
+			if (count >= h->nr_huge_pages)
 				return;
 			if (PageHighMem(page))
 				continue;
 			list_del(&page->lru);
 			update_and_free_page(page);
-			free_huge_pages--;
-			free_huge_pages_node[page_to_nid(page)]--;
+			h->free_huge_pages--;
+			h->free_huge_pages_node[page_to_nid(page)]--;
 		}
 	}
 }
 #else
-static inline void try_to_free_low(unsigned long count)
+static inline void try_to_free_low(struct hstate *h, unsigned long count)
 {
 }
 #endif
 
-#define persistent_huge_pages (nr_huge_pages - surplus_huge_pages)
+#define persistent_huge_pages(h) (h->nr_huge_pages - h->surplus_huge_pages)
 static unsigned long set_max_huge_pages(unsigned long count)
 {
 	unsigned long min_count, ret;
+	struct hstate *h = &default_hstate;
 
 	/*
 	 * Increase the pool size
@@ -963,19 +980,19 @@ static unsigned long set_max_huge_pages(unsigned long count)
 	 * within all the constraints specified by the sysctls.
 	 */
 	spin_lock(&hugetlb_lock);
-	while (surplus_huge_pages && count > persistent_huge_pages) {
-		if (!adjust_pool_surplus(-1))
+	while (h->surplus_huge_pages && count > persistent_huge_pages(h)) {
+		if (!adjust_pool_surplus(h, -1))
 			break;
 	}
 
-	while (count > persistent_huge_pages) {
+	while (count > persistent_huge_pages(h)) {
 		/*
 		 * If this allocation races such that we no longer need the
 		 * page, free_huge_page will handle it by freeing the page
 		 * and reducing the surplus.
 		 */
 		spin_unlock(&hugetlb_lock);
-		ret = alloc_fresh_huge_page();
+		ret = alloc_fresh_huge_page(h);
 		spin_lock(&hugetlb_lock);
 		if (!ret)
 			goto out;
@@ -997,21 +1014,21 @@ static unsigned long set_max_huge_pages(unsigned long count)
 	 * and won't grow the pool anywhere else. Not until one of the
 	 * sysctls are changed, or the surplus pages go out of use.
 	 */
-	min_count = resv_huge_pages + nr_huge_pages - free_huge_pages;
+	min_count = h->resv_huge_pages + h->nr_huge_pages - h->free_huge_pages;
 	min_count = max(count, min_count);
-	try_to_free_low(min_count);
-	while (min_count < persistent_huge_pages) {
-		struct page *page = dequeue_huge_page();
+	try_to_free_low(h, min_count);
+	while (min_count < persistent_huge_pages(h)) {
+		struct page *page = dequeue_huge_page(h);
 		if (!page)
 			break;
-		update_and_free_page(page);
+		update_and_free_page(h, page);
 	}
-	while (count < persistent_huge_pages) {
-		if (!adjust_pool_surplus(1))
+	while (count < persistent_huge_pages(h)) {
+		if (!adjust_pool_surplus(h, 1))
 			break;
 	}
 out:
-	ret = persistent_huge_pages;
+	ret = persistent_huge_pages(h);
 	spin_unlock(&hugetlb_lock);
 	return ret;
 }
@@ -1041,9 +1058,10 @@ int hugetlb_overcommit_handler(struct ctl_table *table, int write,
 			struct file *file, void __user *buffer,
 			size_t *length, loff_t *ppos)
 {
+	struct hstate *h = &default_hstate;
 	proc_doulongvec_minmax(table, write, file, buffer, length, ppos);
 	spin_lock(&hugetlb_lock);
-	nr_overcommit_huge_pages = sysctl_overcommit_huge_pages;
+	h->nr_overcommit_huge_pages = sysctl_overcommit_huge_pages;
 	spin_unlock(&hugetlb_lock);
 	return 0;
 }
@@ -1052,37 +1070,40 @@ int hugetlb_overcommit_handler(struct ctl_table *table, int write,
 
 int hugetlb_report_meminfo(char *buf)
 {
+	struct hstate *h = &default_hstate;
 	return sprintf(buf,
 			"HugePages_Total: %5lu\n"
 			"HugePages_Free:  %5lu\n"
 			"HugePages_Rsvd:  %5lu\n"
 			"HugePages_Surp:  %5lu\n"
 			"Hugepagesize:    %5lu kB\n",
-			nr_huge_pages,
-			free_huge_pages,
-			resv_huge_pages,
-			surplus_huge_pages,
-			HPAGE_SIZE/1024);
+			h->nr_huge_pages,
+			h->free_huge_pages,
+			h->resv_huge_pages,
+			h->surplus_huge_pages,
+			1UL << (huge_page_order(h) + PAGE_SHIFT - 10));
 }
 
 int hugetlb_report_node_meminfo(int nid, char *buf)
 {
+	struct hstate *h = &default_hstate;
 	return sprintf(buf,
 		"Node %d HugePages_Total: %5u\n"
 		"Node %d HugePages_Free:  %5u\n"
 		"Node %d HugePages_Surp:  %5u\n",
-		nid, nr_huge_pages_node[nid],
-		nid, free_huge_pages_node[nid],
-		nid, surplus_huge_pages_node[nid]);
+		nid, h->nr_huge_pages_node[nid],
+		nid, h->free_huge_pages_node[nid],
+		nid, h->surplus_huge_pages_node[nid]);
 }
 
 /* Return the number pages of memory we physically have, in PAGE_SIZE units. */
 unsigned long hugetlb_total_pages(void)
 {
-	return nr_huge_pages * (HPAGE_SIZE / PAGE_SIZE);
+	struct hstate *h = &default_hstate;
+	return h->nr_huge_pages * pages_per_huge_page(h);
 }
 
-static int hugetlb_acct_memory(long delta)
+static int hugetlb_acct_memory(struct hstate *h, long delta)
 {
 	int ret = -ENOMEM;
 
@@ -1105,18 +1126,18 @@ static int hugetlb_acct_memory(long delta)
 	 * semantics that cpuset has.
 	 */
 	if (delta > 0) {
-		if (gather_surplus_pages(delta) < 0)
+		if (gather_surplus_pages(h, delta) < 0)
 			goto out;
 
-		if (delta > cpuset_mems_nr(free_huge_pages_node)) {
-			return_unused_surplus_pages(delta);
+		if (delta > cpuset_mems_nr(h->free_huge_pages_node)) {
+			return_unused_surplus_pages(h, delta);
 			goto out;
 		}
 	}
 
 	ret = 0;
 	if (delta < 0)
-		return_unused_surplus_pages((unsigned long) -delta);
+		return_unused_surplus_pages(h, (unsigned long) -delta);
 
 out:
 	spin_unlock(&hugetlb_lock);
@@ -1141,14 +1162,15 @@ static void hugetlb_vm_op_open(struct vm_area_struct *vma)
 
 static void hugetlb_vm_op_close(struct vm_area_struct *vma)
 {
+	struct hstate *h = hstate_vma(vma);
 	struct resv_map *reservations = vma_resv_map(vma);
 	unsigned long reserve;
 	unsigned long start;
 	unsigned long end;
 
 	if (reservations) {
-		start = vma_hugecache_offset(vma, vma->vm_start);
-		end = vma_hugecache_offset(vma, vma->vm_end);
+		start = vma_hugecache_offset(h, vma, vma->vm_start);
+		end = vma_hugecache_offset(h, vma, vma->vm_end);
 
 		reserve = (end - start) -
 			region_count(&reservations->regions, start, end);
@@ -1156,7 +1178,7 @@ static void hugetlb_vm_op_close(struct vm_area_struct *vma)
 		kref_put(&reservations->refs, resv_map_release);
 
 		if (reserve)
-			hugetlb_acct_memory(-reserve);
+			hugetlb_acct_memory(h, -reserve);
 	}
 }
 
@@ -1214,14 +1236,16 @@ int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
 	struct page *ptepage;
 	unsigned long addr;
 	int cow;
+	struct hstate *h = hstate_vma(vma);
+	unsigned long sz = huge_page_size(h);
 
 	cow = (vma->vm_flags & (VM_SHARED | VM_MAYWRITE)) == VM_MAYWRITE;
 
-	for (addr = vma->vm_start; addr < vma->vm_end; addr += HPAGE_SIZE) {
+	for (addr = vma->vm_start; addr < vma->vm_end; addr += sz) {
 		src_pte = huge_pte_offset(src, addr);
 		if (!src_pte)
 			continue;
-		dst_pte = huge_pte_alloc(dst, addr);
+		dst_pte = huge_pte_alloc(dst, addr, sz);
 		if (!dst_pte)
 			goto nomem;
 
@@ -1257,6 +1281,9 @@ void __unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start,
 	pte_t pte;
 	struct page *page;
 	struct page *tmp;
+	struct hstate *h = hstate_vma(vma);
+	unsigned long sz = huge_page_size(h);
+
 	/*
 	 * A page gathering list, protected by per file i_mmap_lock. The
 	 * lock is used to avoid list corruption from multiple unmapping
@@ -1265,11 +1292,11 @@ void __unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start,
 	LIST_HEAD(page_list);
 
 	WARN_ON(!is_vm_hugetlb_page(vma));
-	BUG_ON(start & ~HPAGE_MASK);
-	BUG_ON(end & ~HPAGE_MASK);
+	BUG_ON(start & ~huge_page_mask(h));
+	BUG_ON(end & ~huge_page_mask(h));
 
 	spin_lock(&mm->page_table_lock);
-	for (address = start; address < end; address += HPAGE_SIZE) {
+	for (address = start; address < end; address += sz) {
 		ptep = huge_pte_offset(mm, address);
 		if (!ptep)
 			continue;
@@ -1383,6 +1410,7 @@ static int hugetlb_cow(struct mm_struct *mm, struct vm_area_struct *vma,
 			unsigned long address, pte_t *ptep, pte_t pte,
 			struct page *pagecache_page)
 {
+	struct hstate *h = hstate_vma(vma);
 	struct page *old_page, *new_page;
 	int avoidcopy;
 	int outside_reserve = 0;
@@ -1443,7 +1471,7 @@ retry_avoidcopy:
 	__SetPageUptodate(new_page);
 	spin_lock(&mm->page_table_lock);
 
-	ptep = huge_pte_offset(mm, address & HPAGE_MASK);
+	ptep = huge_pte_offset(mm, address & huge_page_mask(h));
 	if (likely(pte_same(huge_ptep_get(ptep), pte))) {
 		/* Break COW */
 		huge_ptep_clear_flush(vma, address, ptep);
@@ -1458,14 +1486,14 @@ retry_avoidcopy:
 }
 
 /* Return the pagecache page at a given address within a VMA */
-static struct page *hugetlbfs_pagecache_page(struct vm_area_struct *vma,
-			unsigned long address)
+static struct page *hugetlbfs_pagecache_page(struct hstate *h,
+			struct vm_area_struct *vma, unsigned long address)
 {
 	struct address_space *mapping;
 	pgoff_t idx;
 
 	mapping = vma->vm_file->f_mapping;
-	idx = vma_hugecache_offset(vma, address);
+	idx = vma_hugecache_offset(h, vma, address);
 
 	return find_lock_page(mapping, idx);
 }
@@ -1473,6 +1501,7 @@ static struct page *hugetlbfs_pagecache_page(struct vm_area_struct *vma,
 static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma,
 			unsigned long address, pte_t *ptep, int write_access)
 {
+	struct hstate *h = hstate_vma(vma);
 	int ret = VM_FAULT_SIGBUS;
 	pgoff_t idx;
 	unsigned long size;
@@ -1493,7 +1522,7 @@ static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	}
 
 	mapping = vma->vm_file->f_mapping;
-	idx = vma_hugecache_offset(vma, address);
+	idx = vma_hugecache_offset(h, vma, address);
 
 	/*
 	 * Use page lock to guard against racing truncation
@@ -1502,7 +1531,7 @@ static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma,
 retry:
 	page = find_lock_page(mapping, idx);
 	if (!page) {
-		size = i_size_read(mapping->host) >> HPAGE_SHIFT;
+		size = i_size_read(mapping->host) >> huge_page_shift(h);
 		if (idx >= size)
 			goto out;
 		page = alloc_huge_page(vma, address, 0);
@@ -1510,7 +1539,7 @@ retry:
 			ret = -PTR_ERR(page);
 			goto out;
 		}
-		clear_huge_page(page, address);
+		clear_huge_page(page, address, huge_page_size(h));
 		__SetPageUptodate(page);
 
 		if (vma->vm_flags & VM_SHARED) {
@@ -1526,14 +1555,14 @@ retry:
 			}
 
 			spin_lock(&inode->i_lock);
-			inode->i_blocks += BLOCKS_PER_HUGEPAGE;
+			inode->i_blocks += blocks_per_huge_page(h);
 			spin_unlock(&inode->i_lock);
 		} else
 			lock_page(page);
 	}
 
 	spin_lock(&mm->page_table_lock);
-	size = i_size_read(mapping->host) >> HPAGE_SHIFT;
+	size = i_size_read(mapping->host) >> huge_page_shift(h);
 	if (idx >= size)
 		goto backout;
 
@@ -1569,8 +1598,9 @@ int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 	pte_t entry;
 	int ret;
 	static DEFINE_MUTEX(hugetlb_instantiation_mutex);
+	struct hstate *h = hstate_vma(vma);
 
-	ptep = huge_pte_alloc(mm, address);
+	ptep = huge_pte_alloc(mm, address, huge_page_size(h));
 	if (!ptep)
 		return VM_FAULT_OOM;
 
@@ -1594,7 +1624,7 @@ int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 	if (likely(pte_same(entry, huge_ptep_get(ptep))))
 		if (write_access && !pte_write(entry)) {
 			struct page *page;
-			page = hugetlbfs_pagecache_page(vma, address);
+			page = hugetlbfs_pagecache_page(h, vma, address);
 			ret = hugetlb_cow(mm, vma, address, ptep, entry, page);
 			if (page) {
 				unlock_page(page);
@@ -1615,6 +1645,7 @@ int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	unsigned long pfn_offset;
 	unsigned long vaddr = *position;
 	int remainder = *length;
+	struct hstate *h = hstate_vma(vma);
 
 	spin_lock(&mm->page_table_lock);
 	while (vaddr < vma->vm_end && remainder) {
@@ -1626,7 +1657,7 @@ int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
 		 * each hugepage.  We have to make * sure we get the
 		 * first, for the page indexing below to work.
 		 */
-		pte = huge_pte_offset(mm, vaddr & HPAGE_MASK);
+		pte = huge_pte_offset(mm, vaddr & huge_page_mask(h));
 
 		if (!pte || huge_pte_none(huge_ptep_get(pte)) ||
 		    (write && !pte_write(huge_ptep_get(pte)))) {
@@ -1644,7 +1675,7 @@ int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
 			break;
 		}
 
-		pfn_offset = (vaddr & ~HPAGE_MASK) >> PAGE_SHIFT;
+		pfn_offset = (vaddr & ~huge_page_mask(h)) >> PAGE_SHIFT;
 		page = pte_page(huge_ptep_get(pte));
 same_page:
 		if (pages) {
@@ -1660,7 +1691,7 @@ same_page:
 		--remainder;
 		++i;
 		if (vaddr < vma->vm_end && remainder &&
-				pfn_offset < HPAGE_SIZE/PAGE_SIZE) {
+				pfn_offset < pages_per_huge_page(h)) {
 			/*
 			 * We use pfn_offset to avoid touching the pageframes
 			 * of this compound page.
@@ -1682,13 +1713,14 @@ void hugetlb_change_protection(struct vm_area_struct *vma,
 	unsigned long start = address;
 	pte_t *ptep;
 	pte_t pte;
+	struct hstate *h = hstate_vma(vma);
 
 	BUG_ON(address >= end);
 	flush_cache_range(vma, address, end);
 
 	spin_lock(&vma->vm_file->f_mapping->i_mmap_lock);
 	spin_lock(&mm->page_table_lock);
-	for (; address < end; address += HPAGE_SIZE) {
+	for (; address < end; address += huge_page_size(h)) {
 		ptep = huge_pte_offset(mm, address);
 		if (!ptep)
 			continue;
@@ -1711,6 +1743,7 @@ int hugetlb_reserve_pages(struct inode *inode,
 					struct vm_area_struct *vma)
 {
 	long ret, chg;
+	struct hstate *h = hstate_inode(inode);
 
 	if (vma && vma->vm_flags & VM_NORESERVE)
 		return 0;
@@ -1739,7 +1772,7 @@ int hugetlb_reserve_pages(struct inode *inode,
 
 	if (hugetlb_get_quota(inode->i_mapping, chg))
 		return -ENOSPC;
-	ret = hugetlb_acct_memory(chg);
+	ret = hugetlb_acct_memory(h, chg);
 	if (ret < 0) {
 		hugetlb_put_quota(inode->i_mapping, chg);
 		return ret;
@@ -1751,12 +1784,13 @@ int hugetlb_reserve_pages(struct inode *inode,
 
 void hugetlb_unreserve_pages(struct inode *inode, long offset, long freed)
 {
+	struct hstate *h = hstate_inode(inode);
 	long chg = region_truncate(&inode->i_mapping->private_list, offset);
 
 	spin_lock(&inode->i_lock);
-	inode->i_blocks -= BLOCKS_PER_HUGEPAGE * freed;
+	inode->i_blocks -= blocks_per_huge_page(h);
 	spin_unlock(&inode->i_lock);
 
 	hugetlb_put_quota(inode->i_mapping, (chg - freed));
-	hugetlb_acct_memory(-(chg - freed));
+	hugetlb_acct_memory(h, -(chg - freed));
 }
diff --git a/mm/memory.c b/mm/memory.c
index 72932489a082..c1c1d6d8c22b 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -903,7 +903,7 @@ unsigned long unmap_vmas(struct mmu_gather **tlbp,
 			if (unlikely(is_vm_hugetlb_page(vma))) {
 				unmap_hugepage_range(vma, start, end, NULL);
 				zap_work -= (end - start) /
-						(HPAGE_SIZE / PAGE_SIZE);
+					pages_per_huge_page(hstate_vma(vma));
 				start = end;
 			} else
 				start = unmap_page_range(*tlbp, vma,
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index c94e58b192c3..e550bec20582 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -1481,7 +1481,7 @@ struct zonelist *huge_zonelist(struct vm_area_struct *vma, unsigned long addr,
 
 	if (unlikely((*mpol)->mode == MPOL_INTERLEAVE)) {
 		zl = node_zonelist(interleave_nid(*mpol, vma, addr,
-						HPAGE_SHIFT), gfp_flags);
+				huge_page_shift(hstate_vma(vma))), gfp_flags);
 	} else {
 		zl = policy_zonelist(gfp_flags, *mpol);
 		if ((*mpol)->mode == MPOL_BIND)
@@ -2220,9 +2220,12 @@ static void check_huge_range(struct vm_area_struct *vma,
 {
 	unsigned long addr;
 	struct page *page;
+	struct hstate *h = hstate_vma(vma);
+	unsigned long sz = huge_page_size(h);
 
-	for (addr = start; addr < end; addr += HPAGE_SIZE) {
-		pte_t *ptep = huge_pte_offset(vma->vm_mm, addr & HPAGE_MASK);
+	for (addr = start; addr < end; addr += sz) {
+		pte_t *ptep = huge_pte_offset(vma->vm_mm,
+						addr & huge_page_mask(h));
 		pte_t pte;
 
 		if (!ptep)
diff --git a/mm/mmap.c b/mm/mmap.c
index 57d3b6097deb..5e0cc99e9cd5 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -1812,7 +1812,8 @@ int split_vma(struct mm_struct * mm, struct vm_area_struct * vma,
 	struct mempolicy *pol;
 	struct vm_area_struct *new;
 
-	if (is_vm_hugetlb_page(vma) && (addr & ~HPAGE_MASK))
+	if (is_vm_hugetlb_page(vma) && (addr &
+					~(huge_page_mask(hstate_vma(vma)))))
 		return -EINVAL;
 
 	if (mm->map_count >= sysctl_max_map_count)

From e5ff215941d59f8ae6bf58f6428dc5c26745a612 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 23 Jul 2008 21:27:42 -0700
Subject: [PATCH 0251/1171] hugetlb: multiple hstates for multiple page sizes

Add basic support for more than one hstate in hugetlbfs.  This is the key
to supporting multiple hugetlbfs page sizes at once.

- Rather than a single hstate, we now have an array, with an iterator
- default_hstate continues to be the struct hstate which we use by default
- Add functions for architectures to register new hstates

[akpm@linux-foundation.org: coding-style fixes]
Acked-by: Adam Litke <agl@us.ibm.com>
Acked-by: Nishanth Aravamudan <nacc@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/hugetlb.h |  19 +++++-
 kernel/sysctl.c         |   8 ++-
 mm/hugetlb.c            | 148 ++++++++++++++++++++++++++++++++--------
 3 files changed, 142 insertions(+), 33 deletions(-)

diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index ad2271e11f9b..b75bdb4deba3 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -36,8 +36,6 @@ int hugetlb_reserve_pages(struct inode *inode, long from, long to,
 						struct vm_area_struct *vma);
 void hugetlb_unreserve_pages(struct inode *inode, long offset, long freed);
 
-extern unsigned long max_huge_pages;
-extern unsigned long sysctl_overcommit_huge_pages;
 extern unsigned long hugepages_treat_as_movable;
 extern const unsigned long hugetlb_zero, hugetlb_infinity;
 extern int sysctl_hugetlb_shm_group;
@@ -181,7 +179,17 @@ struct hstate {
 	unsigned int surplus_huge_pages_node[MAX_NUMNODES];
 };
 
-extern struct hstate default_hstate;
+void __init hugetlb_add_hstate(unsigned order);
+struct hstate *size_to_hstate(unsigned long size);
+
+#ifndef HUGE_MAX_HSTATE
+#define HUGE_MAX_HSTATE 1
+#endif
+
+extern struct hstate hstates[HUGE_MAX_HSTATE];
+extern unsigned int default_hstate_idx;
+
+#define default_hstate (hstates[default_hstate_idx])
 
 static inline struct hstate *hstate_vma(struct vm_area_struct *vma)
 {
@@ -230,6 +238,11 @@ static inline unsigned int blocks_per_huge_page(struct hstate *h)
 
 #include <asm/hugetlb.h>
 
+static inline struct hstate *page_hstate(struct page *page)
+{
+	return size_to_hstate(PAGE_SIZE << compound_order(page));
+}
+
 #else
 struct hstate {};
 #define hstate_file(f) NULL
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 1f7b3b76a166..1a8299d1fe59 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -959,7 +959,7 @@ static struct ctl_table vm_table[] = {
 #ifdef CONFIG_HUGETLB_PAGE
 	 {
 		.procname	= "nr_hugepages",
-		.data		= &max_huge_pages,
+		.data		= NULL,
 		.maxlen		= sizeof(unsigned long),
 		.mode		= 0644,
 		.proc_handler	= &hugetlb_sysctl_handler,
@@ -985,10 +985,12 @@ static struct ctl_table vm_table[] = {
 	{
 		.ctl_name	= CTL_UNNUMBERED,
 		.procname	= "nr_overcommit_hugepages",
-		.data		= &sysctl_overcommit_huge_pages,
-		.maxlen		= sizeof(sysctl_overcommit_huge_pages),
+		.data		= NULL,
+		.maxlen		= sizeof(unsigned long),
 		.mode		= 0644,
 		.proc_handler	= &hugetlb_overcommit_handler,
+		.extra1		= (void *)&hugetlb_zero,
+		.extra2		= (void *)&hugetlb_infinity,
 	},
 #endif
 	{
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 0d8153e25f09..82378d44a0c5 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -22,12 +22,19 @@
 #include "internal.h"
 
 const unsigned long hugetlb_zero = 0, hugetlb_infinity = ~0UL;
-unsigned long max_huge_pages;
-unsigned long sysctl_overcommit_huge_pages;
 static gfp_t htlb_alloc_mask = GFP_HIGHUSER;
 unsigned long hugepages_treat_as_movable;
 
-struct hstate default_hstate;
+static int max_hstate;
+unsigned int default_hstate_idx;
+struct hstate hstates[HUGE_MAX_HSTATE];
+
+/* for command line parsing */
+static struct hstate * __initdata parsed_hstate;
+static unsigned long __initdata default_hstate_max_huge_pages;
+
+#define for_each_hstate(h) \
+	for ((h) = hstates; (h) < &hstates[max_hstate]; (h)++)
 
 /*
  * Protects updates to hugepage_freelists, nr_huge_pages, and free_huge_pages
@@ -454,13 +461,24 @@ static void update_and_free_page(struct hstate *h, struct page *page)
 	__free_pages(page, huge_page_order(h));
 }
 
+struct hstate *size_to_hstate(unsigned long size)
+{
+	struct hstate *h;
+
+	for_each_hstate(h) {
+		if (huge_page_size(h) == size)
+			return h;
+	}
+	return NULL;
+}
+
 static void free_huge_page(struct page *page)
 {
 	/*
 	 * Can't pass hstate in here because it is called from the
 	 * compound page destructor.
 	 */
-	struct hstate *h = &default_hstate;
+	struct hstate *h = page_hstate(page);
 	int nid = page_to_nid(page);
 	struct address_space *mapping;
 
@@ -887,39 +905,94 @@ static struct page *alloc_huge_page(struct vm_area_struct *vma,
 	return page;
 }
 
-static int __init hugetlb_init(void)
+static void __init hugetlb_init_one_hstate(struct hstate *h)
 {
 	unsigned long i;
-	struct hstate *h = &default_hstate;
-
-	if (HPAGE_SHIFT == 0)
-		return 0;
-
-	if (!h->order) {
-		h->order = HPAGE_SHIFT - PAGE_SHIFT;
-		h->mask = HPAGE_MASK;
-	}
 
 	for (i = 0; i < MAX_NUMNODES; ++i)
 		INIT_LIST_HEAD(&h->hugepage_freelists[i]);
 
 	h->hugetlb_next_nid = first_node(node_online_map);
 
-	for (i = 0; i < max_huge_pages; ++i) {
+	for (i = 0; i < h->max_huge_pages; ++i) {
 		if (!alloc_fresh_huge_page(h))
 			break;
 	}
-	max_huge_pages = h->free_huge_pages = h->nr_huge_pages = i;
-	printk(KERN_INFO "Total HugeTLB memory allocated, %ld\n",
-			h->free_huge_pages);
+	h->max_huge_pages = h->free_huge_pages = h->nr_huge_pages = i;
+}
+
+static void __init hugetlb_init_hstates(void)
+{
+	struct hstate *h;
+
+	for_each_hstate(h) {
+		hugetlb_init_one_hstate(h);
+	}
+}
+
+static void __init report_hugepages(void)
+{
+	struct hstate *h;
+
+	for_each_hstate(h) {
+		printk(KERN_INFO "Total HugeTLB memory allocated, "
+				"%ld %dMB pages\n",
+				h->free_huge_pages,
+				1 << (h->order + PAGE_SHIFT - 20));
+	}
+}
+
+static int __init hugetlb_init(void)
+{
+	BUILD_BUG_ON(HPAGE_SHIFT == 0);
+
+	if (!size_to_hstate(HPAGE_SIZE)) {
+		hugetlb_add_hstate(HUGETLB_PAGE_ORDER);
+		parsed_hstate->max_huge_pages = default_hstate_max_huge_pages;
+	}
+	default_hstate_idx = size_to_hstate(HPAGE_SIZE) - hstates;
+
+	hugetlb_init_hstates();
+
+	report_hugepages();
+
 	return 0;
 }
 module_init(hugetlb_init);
 
+/* Should be called on processing a hugepagesz=... option */
+void __init hugetlb_add_hstate(unsigned order)
+{
+	struct hstate *h;
+	if (size_to_hstate(PAGE_SIZE << order)) {
+		printk(KERN_WARNING "hugepagesz= specified twice, ignoring\n");
+		return;
+	}
+	BUG_ON(max_hstate >= HUGE_MAX_HSTATE);
+	BUG_ON(order == 0);
+	h = &hstates[max_hstate++];
+	h->order = order;
+	h->mask = ~((1ULL << (order + PAGE_SHIFT)) - 1);
+	hugetlb_init_one_hstate(h);
+	parsed_hstate = h;
+}
+
 static int __init hugetlb_setup(char *s)
 {
-	if (sscanf(s, "%lu", &max_huge_pages) <= 0)
-		max_huge_pages = 0;
+	unsigned long *mhp;
+
+	/*
+	 * !max_hstate means we haven't parsed a hugepagesz= parameter yet,
+	 * so this hugepages= parameter goes to the "default hstate".
+	 */
+	if (!max_hstate)
+		mhp = &default_hstate_max_huge_pages;
+	else
+		mhp = &parsed_hstate->max_huge_pages;
+
+	if (sscanf(s, "%lu", mhp) <= 0)
+		*mhp = 0;
+
 	return 1;
 }
 __setup("hugepages=", hugetlb_setup);
@@ -950,7 +1023,7 @@ static void try_to_free_low(struct hstate *h, unsigned long count)
 			if (PageHighMem(page))
 				continue;
 			list_del(&page->lru);
-			update_and_free_page(page);
+			update_and_free_page(h, page);
 			h->free_huge_pages--;
 			h->free_huge_pages_node[page_to_nid(page)]--;
 		}
@@ -963,10 +1036,9 @@ static inline void try_to_free_low(struct hstate *h, unsigned long count)
 #endif
 
 #define persistent_huge_pages(h) (h->nr_huge_pages - h->surplus_huge_pages)
-static unsigned long set_max_huge_pages(unsigned long count)
+static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count)
 {
 	unsigned long min_count, ret;
-	struct hstate *h = &default_hstate;
 
 	/*
 	 * Increase the pool size
@@ -1037,8 +1109,19 @@ int hugetlb_sysctl_handler(struct ctl_table *table, int write,
 			   struct file *file, void __user *buffer,
 			   size_t *length, loff_t *ppos)
 {
+	struct hstate *h = &default_hstate;
+	unsigned long tmp;
+
+	if (!write)
+		tmp = h->max_huge_pages;
+
+	table->data = &tmp;
+	table->maxlen = sizeof(unsigned long);
 	proc_doulongvec_minmax(table, write, file, buffer, length, ppos);
-	max_huge_pages = set_max_huge_pages(max_huge_pages);
+
+	if (write)
+		h->max_huge_pages = set_max_huge_pages(h, tmp);
+
 	return 0;
 }
 
@@ -1059,10 +1142,21 @@ int hugetlb_overcommit_handler(struct ctl_table *table, int write,
 			size_t *length, loff_t *ppos)
 {
 	struct hstate *h = &default_hstate;
+	unsigned long tmp;
+
+	if (!write)
+		tmp = h->nr_overcommit_huge_pages;
+
+	table->data = &tmp;
+	table->maxlen = sizeof(unsigned long);
 	proc_doulongvec_minmax(table, write, file, buffer, length, ppos);
-	spin_lock(&hugetlb_lock);
-	h->nr_overcommit_huge_pages = sysctl_overcommit_huge_pages;
-	spin_unlock(&hugetlb_lock);
+
+	if (write) {
+		spin_lock(&hugetlb_lock);
+		h->nr_overcommit_huge_pages = tmp;
+		spin_unlock(&hugetlb_lock);
+	}
+
 	return 0;
 }
 

From a137e1cc6d6e7d315fef03962a2a5a113348b13b Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 23 Jul 2008 21:27:43 -0700
Subject: [PATCH 0252/1171] hugetlbfs: per mount huge page sizes

Add the ability to configure the hugetlb hstate used on a per mount basis.

- Add a new pagesize= option to the hugetlbfs mount that allows setting
  the page size
- This option causes the mount code to find the hstate corresponding to the
  specified size, and sets up a pointer to the hstate in the mount's
  superblock.
- Change the hstate accessors to use this information rather than the
  global_hstate they were using (requires a slight change in mm/memory.c
  so we don't NULL deref in the error-unmap path -- see comments).

[np: take hstate out of hugetlbfs inode and vma->vm_private_data]

Acked-by: Adam Litke <agl@us.ibm.com>
Acked-by: Nishanth Aravamudan <nacc@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/hugetlbfs/inode.c    | 45 ++++++++++++++++++++++++++++++++---------
 include/linux/hugetlb.h | 14 ++++++++-----
 mm/hugetlb.c            | 16 +++------------
 mm/memory.c             | 18 +++++++++++++++--
 4 files changed, 64 insertions(+), 29 deletions(-)

diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index 516c581b5371..dbd01d262ca4 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -53,6 +53,7 @@ int sysctl_hugetlb_shm_group;
 enum {
 	Opt_size, Opt_nr_inodes,
 	Opt_mode, Opt_uid, Opt_gid,
+	Opt_pagesize,
 	Opt_err,
 };
 
@@ -62,6 +63,7 @@ static match_table_t tokens = {
 	{Opt_mode,	"mode=%o"},
 	{Opt_uid,	"uid=%u"},
 	{Opt_gid,	"gid=%u"},
+	{Opt_pagesize,	"pagesize=%s"},
 	{Opt_err,	NULL},
 };
 
@@ -750,6 +752,8 @@ hugetlbfs_parse_options(char *options, struct hugetlbfs_config *pconfig)
 	char *p, *rest;
 	substring_t args[MAX_OPT_ARGS];
 	int option;
+	unsigned long long size = 0;
+	enum { NO_SIZE, SIZE_STD, SIZE_PERCENT } setsize = NO_SIZE;
 
 	if (!options)
 		return 0;
@@ -780,17 +784,13 @@ hugetlbfs_parse_options(char *options, struct hugetlbfs_config *pconfig)
 			break;
 
 		case Opt_size: {
- 			unsigned long long size;
 			/* memparse() will accept a K/M/G without a digit */
 			if (!isdigit(*args[0].from))
 				goto bad_val;
 			size = memparse(args[0].from, &rest);
-			if (*rest == '%') {
-				size <<= HPAGE_SHIFT;
-				size *= max_huge_pages;
-				do_div(size, 100);
-			}
-			pconfig->nr_blocks = (size >> HPAGE_SHIFT);
+			setsize = SIZE_STD;
+			if (*rest == '%')
+				setsize = SIZE_PERCENT;
 			break;
 		}
 
@@ -801,6 +801,19 @@ hugetlbfs_parse_options(char *options, struct hugetlbfs_config *pconfig)
 			pconfig->nr_inodes = memparse(args[0].from, &rest);
 			break;
 
+		case Opt_pagesize: {
+			unsigned long ps;
+			ps = memparse(args[0].from, &rest);
+			pconfig->hstate = size_to_hstate(ps);
+			if (!pconfig->hstate) {
+				printk(KERN_ERR
+				"hugetlbfs: Unsupported page size %lu MB\n",
+					ps >> 20);
+				return -EINVAL;
+			}
+			break;
+		}
+
 		default:
 			printk(KERN_ERR "hugetlbfs: Bad mount option: \"%s\"\n",
 				 p);
@@ -808,6 +821,18 @@ hugetlbfs_parse_options(char *options, struct hugetlbfs_config *pconfig)
 			break;
 		}
 	}
+
+	/* Do size after hstate is set up */
+	if (setsize > NO_SIZE) {
+		struct hstate *h = pconfig->hstate;
+		if (setsize == SIZE_PERCENT) {
+			size <<= huge_page_shift(h);
+			size *= h->max_huge_pages;
+			do_div(size, 100);
+		}
+		pconfig->nr_blocks = (size >> huge_page_shift(h));
+	}
+
 	return 0;
 
 bad_val:
@@ -832,6 +857,7 @@ hugetlbfs_fill_super(struct super_block *sb, void *data, int silent)
 	config.uid = current->fsuid;
 	config.gid = current->fsgid;
 	config.mode = 0755;
+	config.hstate = &default_hstate;
 	ret = hugetlbfs_parse_options(data, &config);
 	if (ret)
 		return ret;
@@ -840,14 +866,15 @@ hugetlbfs_fill_super(struct super_block *sb, void *data, int silent)
 	if (!sbinfo)
 		return -ENOMEM;
 	sb->s_fs_info = sbinfo;
+	sbinfo->hstate = config.hstate;
 	spin_lock_init(&sbinfo->stat_lock);
 	sbinfo->max_blocks = config.nr_blocks;
 	sbinfo->free_blocks = config.nr_blocks;
 	sbinfo->max_inodes = config.nr_inodes;
 	sbinfo->free_inodes = config.nr_inodes;
 	sb->s_maxbytes = MAX_LFS_FILESIZE;
-	sb->s_blocksize = HPAGE_SIZE;
-	sb->s_blocksize_bits = HPAGE_SHIFT;
+	sb->s_blocksize = huge_page_size(config.hstate);
+	sb->s_blocksize_bits = huge_page_shift(config.hstate);
 	sb->s_magic = HUGETLBFS_MAGIC;
 	sb->s_op = &hugetlbfs_ops;
 	sb->s_time_gran = 1;
diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index b75bdb4deba3..ba9263e631b9 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -100,6 +100,7 @@ struct hugetlbfs_config {
 	umode_t mode;
 	long	nr_blocks;
 	long	nr_inodes;
+	struct hstate *hstate;
 };
 
 struct hugetlbfs_sb_info {
@@ -108,6 +109,7 @@ struct hugetlbfs_sb_info {
 	long	max_inodes;   /* inodes allowed */
 	long	free_inodes;  /* inodes free */
 	spinlock_t	stat_lock;
+	struct hstate *hstate;
 };
 
 
@@ -191,19 +193,21 @@ extern unsigned int default_hstate_idx;
 
 #define default_hstate (hstates[default_hstate_idx])
 
-static inline struct hstate *hstate_vma(struct vm_area_struct *vma)
+static inline struct hstate *hstate_inode(struct inode *i)
 {
-	return &default_hstate;
+	struct hugetlbfs_sb_info *hsb;
+	hsb = HUGETLBFS_SB(i->i_sb);
+	return hsb->hstate;
 }
 
 static inline struct hstate *hstate_file(struct file *f)
 {
-	return &default_hstate;
+	return hstate_inode(f->f_dentry->d_inode);
 }
 
-static inline struct hstate *hstate_inode(struct inode *i)
+static inline struct hstate *hstate_vma(struct vm_area_struct *vma)
 {
-	return &default_hstate;
+	return hstate_file(vma->vm_file);
 }
 
 static inline unsigned long huge_page_size(struct hstate *h)
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 82378d44a0c5..4cf7a90e9140 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -1439,19 +1439,9 @@ void __unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start,
 void unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start,
 			  unsigned long end, struct page *ref_page)
 {
-	/*
-	 * It is undesirable to test vma->vm_file as it should be non-null
-	 * for valid hugetlb area. However, vm_file will be NULL in the error
-	 * cleanup path of do_mmap_pgoff. When hugetlbfs ->mmap method fails,
-	 * do_mmap_pgoff() nullifies vma->vm_file before calling this function
-	 * to clean up. Since no pte has actually been setup, it is safe to
-	 * do nothing in this case.
-	 */
-	if (vma->vm_file) {
-		spin_lock(&vma->vm_file->f_mapping->i_mmap_lock);
-		__unmap_hugepage_range(vma, start, end, ref_page);
-		spin_unlock(&vma->vm_file->f_mapping->i_mmap_lock);
-	}
+	spin_lock(&vma->vm_file->f_mapping->i_mmap_lock);
+	__unmap_hugepage_range(vma, start, end, ref_page);
+	spin_unlock(&vma->vm_file->f_mapping->i_mmap_lock);
 }
 
 /*
diff --git a/mm/memory.c b/mm/memory.c
index c1c1d6d8c22b..02fc6b1047b0 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -901,9 +901,23 @@ unsigned long unmap_vmas(struct mmu_gather **tlbp,
 			}
 
 			if (unlikely(is_vm_hugetlb_page(vma))) {
-				unmap_hugepage_range(vma, start, end, NULL);
-				zap_work -= (end - start) /
+				/*
+				 * It is undesirable to test vma->vm_file as it
+				 * should be non-null for valid hugetlb area.
+				 * However, vm_file will be NULL in the error
+				 * cleanup path of do_mmap_pgoff. When
+				 * hugetlbfs ->mmap method fails,
+				 * do_mmap_pgoff() nullifies vma->vm_file
+				 * before calling this function to clean up.
+				 * Since no pte has actually been setup, it is
+				 * safe to do nothing in this case.
+				 */
+				if (vma->vm_file) {
+					unmap_hugepage_range(vma, start, end, NULL);
+					zap_work -= (end - start) /
 					pages_per_huge_page(hstate_vma(vma));
+				}
+
 				start = end;
 			} else
 				start = unmap_page_range(*tlbp, vma,

From a3437870160cf2caaac6bdd76c7377a5a4145a8c Mon Sep 17 00:00:00 2001
From: Nishanth Aravamudan <nacc@us.ibm.com>
Date: Wed, 23 Jul 2008 21:27:44 -0700
Subject: [PATCH 0253/1171] hugetlb: new sysfs interface

Provide new hugepages user APIs that are more suited to multiple hstates
in sysfs.  There is a new directory, /sys/kernel/hugepages.  Underneath
that directory there will be a directory per-supported hugepage size,
e.g.:

/sys/kernel/hugepages/hugepages-64kB
/sys/kernel/hugepages/hugepages-16384kB
/sys/kernel/hugepages/hugepages-16777216kB

corresponding to 64k, 16m and 16g respectively.  Within each
hugepages-size directory there are a number of files, corresponding to the
tracked counters in the hstate, e.g.:

/sys/kernel/hugepages/hugepages-64/nr_hugepages
/sys/kernel/hugepages/hugepages-64/nr_overcommit_hugepages
/sys/kernel/hugepages/hugepages-64/free_hugepages
/sys/kernel/hugepages/hugepages-64/resv_hugepages
/sys/kernel/hugepages/hugepages-64/surplus_hugepages

Of these files, the first two are read-write and the latter three are
read-only.  The size of the hugepage being manipulated is trivially
deducible from the enclosing directory and is always expressed in kB (to
match meminfo).

[dave@linux.vnet.ibm.com: fix build]
[nacc@us.ibm.com: hugetlb: hang off of /sys/kernel/mm rather than /sys/kernel]
[nacc@us.ibm.com: hugetlb: remove CONFIG_SYSFS dependency]
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 .../ABI/testing/sysfs-kernel-mm-hugepages     |  15 +
 Documentation/vm/hugetlbpage.txt              |  23 ++
 include/linux/hugetlb.h                       |   2 +
 mm/hugetlb.c                                  | 288 ++++++++++++++----
 4 files changed, 262 insertions(+), 66 deletions(-)
 create mode 100644 Documentation/ABI/testing/sysfs-kernel-mm-hugepages

diff --git a/Documentation/ABI/testing/sysfs-kernel-mm-hugepages b/Documentation/ABI/testing/sysfs-kernel-mm-hugepages
new file mode 100644
index 000000000000..e21c00571cf4
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-kernel-mm-hugepages
@@ -0,0 +1,15 @@
+What:		/sys/kernel/mm/hugepages/
+Date:		June 2008
+Contact:	Nishanth Aravamudan <nacc@us.ibm.com>, hugetlb maintainers
+Description:
+		/sys/kernel/mm/hugepages/ contains a number of subdirectories
+		of the form hugepages-<size>kB, where <size> is the page size
+		of the hugepages supported by the kernel/CPU combination.
+
+		Under these directories are a number of files:
+			nr_hugepages
+			nr_overcommit_hugepages
+			free_hugepages
+			surplus_hugepages
+			resv_hugepages
+		See Documentation/vm/hugetlbpage.txt for details.
diff --git a/Documentation/vm/hugetlbpage.txt b/Documentation/vm/hugetlbpage.txt
index 3102b81bef88..8a5b5763f0fe 100644
--- a/Documentation/vm/hugetlbpage.txt
+++ b/Documentation/vm/hugetlbpage.txt
@@ -95,6 +95,29 @@ this condition holds, however, no more surplus huge pages will be
 allowed on the system until one of the two sysctls are increased
 sufficiently, or the surplus huge pages go out of use and are freed.
 
+With support for multiple hugepage pools at run-time available, much of
+the hugepage userspace interface has been duplicated in sysfs. The above
+information applies to the default hugepage size (which will be
+controlled by the proc interfaces for backwards compatibility). The root
+hugepage control directory is
+
+	/sys/kernel/mm/hugepages
+
+For each hugepage size supported by the running kernel, a subdirectory
+will exist, of the form
+
+	hugepages-${size}kB
+
+Inside each of these directories, the same set of files will exist:
+
+	nr_hugepages
+	nr_overcommit_hugepages
+	free_hugepages
+	resv_hugepages
+	surplus_hugepages
+
+which function as described above for the default hugepage-sized case.
+
 If the user applications are going to request hugepages using mmap system
 call, then it is required that system administrator mount a file system of
 type hugetlbfs:
diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index ba9263e631b9..58c0de32e7f0 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -164,6 +164,7 @@ unsigned long hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
 
 #ifdef CONFIG_HUGETLB_PAGE
 
+#define HSTATE_NAME_LEN 32
 /* Defines one hugetlb page size */
 struct hstate {
 	int hugetlb_next_nid;
@@ -179,6 +180,7 @@ struct hstate {
 	unsigned int nr_huge_pages_node[MAX_NUMNODES];
 	unsigned int free_huge_pages_node[MAX_NUMNODES];
 	unsigned int surplus_huge_pages_node[MAX_NUMNODES];
+	char name[HSTATE_NAME_LEN];
 };
 
 void __init hugetlb_add_hstate(unsigned order);
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 4cf7a90e9140..bb49ce5d0067 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -14,6 +14,7 @@
 #include <linux/mempolicy.h>
 #include <linux/cpuset.h>
 #include <linux/mutex.h>
+#include <linux/sysfs.h>
 
 #include <asm/page.h>
 #include <asm/pgtable.h>
@@ -942,72 +943,6 @@ static void __init report_hugepages(void)
 	}
 }
 
-static int __init hugetlb_init(void)
-{
-	BUILD_BUG_ON(HPAGE_SHIFT == 0);
-
-	if (!size_to_hstate(HPAGE_SIZE)) {
-		hugetlb_add_hstate(HUGETLB_PAGE_ORDER);
-		parsed_hstate->max_huge_pages = default_hstate_max_huge_pages;
-	}
-	default_hstate_idx = size_to_hstate(HPAGE_SIZE) - hstates;
-
-	hugetlb_init_hstates();
-
-	report_hugepages();
-
-	return 0;
-}
-module_init(hugetlb_init);
-
-/* Should be called on processing a hugepagesz=... option */
-void __init hugetlb_add_hstate(unsigned order)
-{
-	struct hstate *h;
-	if (size_to_hstate(PAGE_SIZE << order)) {
-		printk(KERN_WARNING "hugepagesz= specified twice, ignoring\n");
-		return;
-	}
-	BUG_ON(max_hstate >= HUGE_MAX_HSTATE);
-	BUG_ON(order == 0);
-	h = &hstates[max_hstate++];
-	h->order = order;
-	h->mask = ~((1ULL << (order + PAGE_SHIFT)) - 1);
-	hugetlb_init_one_hstate(h);
-	parsed_hstate = h;
-}
-
-static int __init hugetlb_setup(char *s)
-{
-	unsigned long *mhp;
-
-	/*
-	 * !max_hstate means we haven't parsed a hugepagesz= parameter yet,
-	 * so this hugepages= parameter goes to the "default hstate".
-	 */
-	if (!max_hstate)
-		mhp = &default_hstate_max_huge_pages;
-	else
-		mhp = &parsed_hstate->max_huge_pages;
-
-	if (sscanf(s, "%lu", mhp) <= 0)
-		*mhp = 0;
-
-	return 1;
-}
-__setup("hugepages=", hugetlb_setup);
-
-static unsigned int cpuset_mems_nr(unsigned int *array)
-{
-	int node;
-	unsigned int nr = 0;
-
-	for_each_node_mask(node, cpuset_current_mems_allowed)
-		nr += array[node];
-
-	return nr;
-}
-
 #ifdef CONFIG_SYSCTL
 #ifdef CONFIG_HIGHMEM
 static void try_to_free_low(struct hstate *h, unsigned long count)
@@ -1105,6 +1040,227 @@ out:
 	return ret;
 }
 
+#define HSTATE_ATTR_RO(_name) \
+	static struct kobj_attribute _name##_attr = __ATTR_RO(_name)
+
+#define HSTATE_ATTR(_name) \
+	static struct kobj_attribute _name##_attr = \
+		__ATTR(_name, 0644, _name##_show, _name##_store)
+
+static struct kobject *hugepages_kobj;
+static struct kobject *hstate_kobjs[HUGE_MAX_HSTATE];
+
+static struct hstate *kobj_to_hstate(struct kobject *kobj)
+{
+	int i;
+	for (i = 0; i < HUGE_MAX_HSTATE; i++)
+		if (hstate_kobjs[i] == kobj)
+			return &hstates[i];
+	BUG();
+	return NULL;
+}
+
+static ssize_t nr_hugepages_show(struct kobject *kobj,
+					struct kobj_attribute *attr, char *buf)
+{
+	struct hstate *h = kobj_to_hstate(kobj);
+	return sprintf(buf, "%lu\n", h->nr_huge_pages);
+}
+static ssize_t nr_hugepages_store(struct kobject *kobj,
+		struct kobj_attribute *attr, const char *buf, size_t count)
+{
+	int err;
+	unsigned long input;
+	struct hstate *h = kobj_to_hstate(kobj);
+
+	err = strict_strtoul(buf, 10, &input);
+	if (err)
+		return 0;
+
+	h->max_huge_pages = set_max_huge_pages(h, input);
+
+	return count;
+}
+HSTATE_ATTR(nr_hugepages);
+
+static ssize_t nr_overcommit_hugepages_show(struct kobject *kobj,
+					struct kobj_attribute *attr, char *buf)
+{
+	struct hstate *h = kobj_to_hstate(kobj);
+	return sprintf(buf, "%lu\n", h->nr_overcommit_huge_pages);
+}
+static ssize_t nr_overcommit_hugepages_store(struct kobject *kobj,
+		struct kobj_attribute *attr, const char *buf, size_t count)
+{
+	int err;
+	unsigned long input;
+	struct hstate *h = kobj_to_hstate(kobj);
+
+	err = strict_strtoul(buf, 10, &input);
+	if (err)
+		return 0;
+
+	spin_lock(&hugetlb_lock);
+	h->nr_overcommit_huge_pages = input;
+	spin_unlock(&hugetlb_lock);
+
+	return count;
+}
+HSTATE_ATTR(nr_overcommit_hugepages);
+
+static ssize_t free_hugepages_show(struct kobject *kobj,
+					struct kobj_attribute *attr, char *buf)
+{
+	struct hstate *h = kobj_to_hstate(kobj);
+	return sprintf(buf, "%lu\n", h->free_huge_pages);
+}
+HSTATE_ATTR_RO(free_hugepages);
+
+static ssize_t resv_hugepages_show(struct kobject *kobj,
+					struct kobj_attribute *attr, char *buf)
+{
+	struct hstate *h = kobj_to_hstate(kobj);
+	return sprintf(buf, "%lu\n", h->resv_huge_pages);
+}
+HSTATE_ATTR_RO(resv_hugepages);
+
+static ssize_t surplus_hugepages_show(struct kobject *kobj,
+					struct kobj_attribute *attr, char *buf)
+{
+	struct hstate *h = kobj_to_hstate(kobj);
+	return sprintf(buf, "%lu\n", h->surplus_huge_pages);
+}
+HSTATE_ATTR_RO(surplus_hugepages);
+
+static struct attribute *hstate_attrs[] = {
+	&nr_hugepages_attr.attr,
+	&nr_overcommit_hugepages_attr.attr,
+	&free_hugepages_attr.attr,
+	&resv_hugepages_attr.attr,
+	&surplus_hugepages_attr.attr,
+	NULL,
+};
+
+static struct attribute_group hstate_attr_group = {
+	.attrs = hstate_attrs,
+};
+
+static int __init hugetlb_sysfs_add_hstate(struct hstate *h)
+{
+	int retval;
+
+	hstate_kobjs[h - hstates] = kobject_create_and_add(h->name,
+							hugepages_kobj);
+	if (!hstate_kobjs[h - hstates])
+		return -ENOMEM;
+
+	retval = sysfs_create_group(hstate_kobjs[h - hstates],
+							&hstate_attr_group);
+	if (retval)
+		kobject_put(hstate_kobjs[h - hstates]);
+
+	return retval;
+}
+
+static void __init hugetlb_sysfs_init(void)
+{
+	struct hstate *h;
+	int err;
+
+	hugepages_kobj = kobject_create_and_add("hugepages", mm_kobj);
+	if (!hugepages_kobj)
+		return;
+
+	for_each_hstate(h) {
+		err = hugetlb_sysfs_add_hstate(h);
+		if (err)
+			printk(KERN_ERR "Hugetlb: Unable to add hstate %s",
+								h->name);
+	}
+}
+
+static void __exit hugetlb_exit(void)
+{
+	struct hstate *h;
+
+	for_each_hstate(h) {
+		kobject_put(hstate_kobjs[h - hstates]);
+	}
+
+	kobject_put(hugepages_kobj);
+}
+module_exit(hugetlb_exit);
+
+static int __init hugetlb_init(void)
+{
+	BUILD_BUG_ON(HPAGE_SHIFT == 0);
+
+	if (!size_to_hstate(HPAGE_SIZE)) {
+		hugetlb_add_hstate(HUGETLB_PAGE_ORDER);
+		parsed_hstate->max_huge_pages = default_hstate_max_huge_pages;
+	}
+	default_hstate_idx = size_to_hstate(HPAGE_SIZE) - hstates;
+
+	hugetlb_init_hstates();
+
+	report_hugepages();
+
+	hugetlb_sysfs_init();
+
+	return 0;
+}
+module_init(hugetlb_init);
+
+/* Should be called on processing a hugepagesz=... option */
+void __init hugetlb_add_hstate(unsigned order)
+{
+	struct hstate *h;
+	if (size_to_hstate(PAGE_SIZE << order)) {
+		printk(KERN_WARNING "hugepagesz= specified twice, ignoring\n");
+		return;
+	}
+	BUG_ON(max_hstate >= HUGE_MAX_HSTATE);
+	BUG_ON(order == 0);
+	h = &hstates[max_hstate++];
+	h->order = order;
+	h->mask = ~((1ULL << (order + PAGE_SHIFT)) - 1);
+	snprintf(h->name, HSTATE_NAME_LEN, "hugepages-%lukB",
+					huge_page_size(h)/1024);
+	hugetlb_init_one_hstate(h);
+	parsed_hstate = h;
+}
+
+static int __init hugetlb_setup(char *s)
+{
+	unsigned long *mhp;
+
+	/*
+	 * !max_hstate means we haven't parsed a hugepagesz= parameter yet,
+	 * so this hugepages= parameter goes to the "default hstate".
+	 */
+	if (!max_hstate)
+		mhp = &default_hstate_max_huge_pages;
+	else
+		mhp = &parsed_hstate->max_huge_pages;
+
+	if (sscanf(s, "%lu", mhp) <= 0)
+		*mhp = 0;
+
+	return 1;
+}
+__setup("hugepages=", hugetlb_setup);
+
+static unsigned int cpuset_mems_nr(unsigned int *array)
+{
+	int node;
+	unsigned int nr = 0;
+
+	for_each_node_mask(node, cpuset_current_mems_allowed)
+		nr += array[node];
+
+	return nr;
+}
+
 int hugetlb_sysctl_handler(struct ctl_table *table, int write,
 			   struct file *file, void __user *buffer,
 			   size_t *length, loff_t *ppos)

From 5ced66c901f1cf0b684feb15c2cd8b126e263d07 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 23 Jul 2008 21:27:45 -0700
Subject: [PATCH 0254/1171] hugetlb: abstract numa round robin selection

Need this as a separate function for a future patch.

No behaviour change.

Acked-by: Adam Litke <agl@us.ibm.com>
Acked-by: Nishanth Aravamudan <nacc@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/hugetlb.c | 37 ++++++++++++++++++++++---------------
 1 file changed, 22 insertions(+), 15 deletions(-)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index bb49ce5d0067..5e620e25cf08 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -565,6 +565,27 @@ static struct page *alloc_fresh_huge_page_node(struct hstate *h, int nid)
 	return page;
 }
 
+/*
+ * Use a helper variable to find the next node and then
+ * copy it back to hugetlb_next_nid afterwards:
+ * otherwise there's a window in which a racer might
+ * pass invalid nid MAX_NUMNODES to alloc_pages_node.
+ * But we don't need to use a spin_lock here: it really
+ * doesn't matter if occasionally a racer chooses the
+ * same nid as we do.  Move nid forward in the mask even
+ * if we just successfully allocated a hugepage so that
+ * the next caller gets hugepages on the next node.
+ */
+static int hstate_next_node(struct hstate *h)
+{
+	int next_nid;
+	next_nid = next_node(h->hugetlb_next_nid, node_online_map);
+	if (next_nid == MAX_NUMNODES)
+		next_nid = first_node(node_online_map);
+	h->hugetlb_next_nid = next_nid;
+	return next_nid;
+}
+
 static int alloc_fresh_huge_page(struct hstate *h)
 {
 	struct page *page;
@@ -578,21 +599,7 @@ static int alloc_fresh_huge_page(struct hstate *h)
 		page = alloc_fresh_huge_page_node(h, h->hugetlb_next_nid);
 		if (page)
 			ret = 1;
-		/*
-		 * Use a helper variable to find the next node and then
-		 * copy it back to hugetlb_next_nid afterwards:
-		 * otherwise there's a window in which a racer might
-		 * pass invalid nid MAX_NUMNODES to alloc_pages_node.
-		 * But we don't need to use a spin_lock here: it really
-		 * doesn't matter if occasionally a racer chooses the
-		 * same nid as we do.  Move nid forward in the mask even
-		 * if we just successfully allocated a hugepage so that
-		 * the next caller gets hugepages on the next node.
-		 */
-		next_nid = next_node(h->hugetlb_next_nid, node_online_map);
-		if (next_nid == MAX_NUMNODES)
-			next_nid = first_node(node_online_map);
-		h->hugetlb_next_nid = next_nid;
+		next_nid = hstate_next_node(h);
 	} while (!page && h->hugetlb_next_nid != start_nid);
 
 	if (ret)

From b54bbf7b81170f03597c17dd0b559e3006bc9868 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 23 Jul 2008 21:27:45 -0700
Subject: [PATCH 0255/1171] mm: introduce non panic alloc_bootmem

Straight forward variant of the existing __alloc_bootmem_node, only
subsequent patch when allocating giant hugepages at boot -- don't want to
panic if we can't allocate as many as the user asked for.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/bootmem.h |  4 ++++
 mm/bootmem.c            | 12 ++++++++++++
 2 files changed, 16 insertions(+)

diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index dd8fee6c46d9..f352c5f125b4 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -89,6 +89,10 @@ extern void *__alloc_bootmem_node(pg_data_t *pgdat,
 				  unsigned long size,
 				  unsigned long align,
 				  unsigned long goal);
+extern void *__alloc_bootmem_node_nopanic(pg_data_t *pgdat,
+				  unsigned long size,
+				  unsigned long align,
+				  unsigned long goal);
 extern unsigned long init_bootmem_node(pg_data_t *pgdat,
 				       unsigned long freepfn,
 				       unsigned long startpfn,
diff --git a/mm/bootmem.c b/mm/bootmem.c
index 4bc6ae2fbaa3..9ac972535fff 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -578,6 +578,18 @@ void * __init alloc_bootmem_section(unsigned long size,
 }
 #endif
 
+void * __init __alloc_bootmem_node_nopanic(pg_data_t *pgdat, unsigned long size,
+				   unsigned long align, unsigned long goal)
+{
+	void *ptr;
+
+	ptr = alloc_bootmem_core(pgdat->bdata, size, align, goal, 0);
+	if (ptr)
+		return ptr;
+
+	return __alloc_bootmem_nopanic(size, align, goal);
+}
+
 #ifndef ARCH_LOW_ADDRESS_LIMIT
 #define ARCH_LOW_ADDRESS_LIMIT	0xffffffffUL
 #endif

From 01ad1c0827db5b3695c53e296dbb2c1da16a0911 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 23 Jul 2008 21:27:46 -0700
Subject: [PATCH 0256/1171] mm: export prep_compound_page to mm

hugetlb will need to get compound pages from bootmem to handle the case of
them being greater than or equal to MAX_ORDER.  Export the constructor
function needed for this.

Acked-by: Adam Litke <agl@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/internal.h   | 2 ++
 mm/page_alloc.c | 2 +-
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/mm/internal.h b/mm/internal.h
index 858ad01864dc..1f43f7416972 100644
--- a/mm/internal.h
+++ b/mm/internal.h
@@ -16,6 +16,8 @@
 void free_pgtables(struct mmu_gather *tlb, struct vm_area_struct *start_vma,
 		unsigned long floor, unsigned long ceiling);
 
+extern void prep_compound_page(struct page *page, unsigned long order);
+
 static inline void set_page_count(struct page *page, int v)
 {
 	atomic_set(&page->_count, v);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index e43aae135b38..eaa86671ebbd 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -264,7 +264,7 @@ static void free_compound_page(struct page *page)
 	__free_pages_ok(page, compound_order(page));
 }
 
-static void prep_compound_page(struct page *page, unsigned long order)
+void prep_compound_page(struct page *page, unsigned long order)
 {
 	int i;
 	int nr_pages = 1 << order;

From aa888a74977a8f2120ae9332376e179c39a6b07d Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 23 Jul 2008 21:27:47 -0700
Subject: [PATCH 0257/1171] hugetlb: support larger than MAX_ORDER

This is needed on x86-64 to handle GB pages in hugetlbfs, because it is
not practical to enlarge MAX_ORDER to 1GB.

Instead the 1GB pages are only allocated at boot using the bootmem
allocator using the hugepages=...  option.

These 1G bootmem pages are never freed.  In theory it would be possible to
implement that with some complications, but since it would be a one-way
street (>= MAX_ORDER pages cannot be allocated later) I decided not to
currently.

The >= MAX_ORDER code is not ifdef'ed per architecture.  It is not very
big and the ifdef uglyness seemed not be worth it.

Known problems: /proc/meminfo and "free" do not display the memory
allocated for gb pages in "Total".  This is a little confusing for the
user.

Acked-by: Andrew Hastings <abh@cray.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/hugetlb.c | 83 ++++++++++++++++++++++++++++++++++++++++++++++++++--
 1 file changed, 81 insertions(+), 2 deletions(-)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 5e620e25cf08..1a6fe87555b2 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -14,6 +14,7 @@
 #include <linux/mempolicy.h>
 #include <linux/cpuset.h>
 #include <linux/mutex.h>
+#include <linux/bootmem.h>
 #include <linux/sysfs.h>
 
 #include <asm/page.h>
@@ -489,7 +490,7 @@ static void free_huge_page(struct page *page)
 	INIT_LIST_HEAD(&page->lru);
 
 	spin_lock(&hugetlb_lock);
-	if (h->surplus_huge_pages_node[nid]) {
+	if (h->surplus_huge_pages_node[nid] && huge_page_order(h) < MAX_ORDER) {
 		update_and_free_page(h, page);
 		h->surplus_huge_pages--;
 		h->surplus_huge_pages_node[nid]--;
@@ -550,6 +551,9 @@ static struct page *alloc_fresh_huge_page_node(struct hstate *h, int nid)
 {
 	struct page *page;
 
+	if (h->order >= MAX_ORDER)
+		return NULL;
+
 	page = alloc_pages_node(nid,
 		htlb_alloc_mask|__GFP_COMP|__GFP_THISNODE|
 						__GFP_REPEAT|__GFP_NOWARN,
@@ -616,6 +620,9 @@ static struct page *alloc_buddy_huge_page(struct hstate *h,
 	struct page *page;
 	unsigned int nid;
 
+	if (h->order >= MAX_ORDER)
+		return NULL;
+
 	/*
 	 * Assume we will successfully allocate the surplus page to
 	 * prevent racing processes from causing the surplus to exceed
@@ -792,6 +799,10 @@ static void return_unused_surplus_pages(struct hstate *h,
 	/* Uncommit the reservation */
 	h->resv_huge_pages -= unused_resv_pages;
 
+	/* Cannot return gigantic pages currently */
+	if (h->order >= MAX_ORDER)
+		return;
+
 	nr_pages = min(unused_resv_pages, h->surplus_huge_pages);
 
 	while (remaining_iterations-- && nr_pages) {
@@ -913,6 +924,63 @@ static struct page *alloc_huge_page(struct vm_area_struct *vma,
 	return page;
 }
 
+static __initdata LIST_HEAD(huge_boot_pages);
+
+struct huge_bootmem_page {
+	struct list_head list;
+	struct hstate *hstate;
+};
+
+static int __init alloc_bootmem_huge_page(struct hstate *h)
+{
+	struct huge_bootmem_page *m;
+	int nr_nodes = nodes_weight(node_online_map);
+
+	while (nr_nodes) {
+		void *addr;
+
+		addr = __alloc_bootmem_node_nopanic(
+				NODE_DATA(h->hugetlb_next_nid),
+				huge_page_size(h), huge_page_size(h), 0);
+
+		if (addr) {
+			/*
+			 * Use the beginning of the huge page to store the
+			 * huge_bootmem_page struct (until gather_bootmem
+			 * puts them into the mem_map).
+			 */
+			m = addr;
+			if (m)
+				goto found;
+		}
+		hstate_next_node(h);
+		nr_nodes--;
+	}
+	return 0;
+
+found:
+	BUG_ON((unsigned long)virt_to_phys(m) & (huge_page_size(h) - 1));
+	/* Put them into a private list first because mem_map is not up yet */
+	list_add(&m->list, &huge_boot_pages);
+	m->hstate = h;
+	return 1;
+}
+
+/* Put bootmem huge pages into the standard lists after mem_map is up */
+static void __init gather_bootmem_prealloc(void)
+{
+	struct huge_bootmem_page *m;
+
+	list_for_each_entry(m, &huge_boot_pages, list) {
+		struct page *page = virt_to_page(m);
+		struct hstate *h = m->hstate;
+		__ClearPageReserved(page);
+		WARN_ON(page_count(page) != 1);
+		prep_compound_page(page, h->order);
+		prep_new_huge_page(h, page, page_to_nid(page));
+	}
+}
+
 static void __init hugetlb_init_one_hstate(struct hstate *h)
 {
 	unsigned long i;
@@ -923,7 +991,10 @@ static void __init hugetlb_init_one_hstate(struct hstate *h)
 	h->hugetlb_next_nid = first_node(node_online_map);
 
 	for (i = 0; i < h->max_huge_pages; ++i) {
-		if (!alloc_fresh_huge_page(h))
+		if (h->order >= MAX_ORDER) {
+			if (!alloc_bootmem_huge_page(h))
+				break;
+		} else if (!alloc_fresh_huge_page(h))
 			break;
 	}
 	h->max_huge_pages = h->free_huge_pages = h->nr_huge_pages = i;
@@ -956,6 +1027,9 @@ static void try_to_free_low(struct hstate *h, unsigned long count)
 {
 	int i;
 
+	if (h->order >= MAX_ORDER)
+		return;
+
 	for (i = 0; i < MAX_NUMNODES; ++i) {
 		struct page *page, *next;
 		struct list_head *freel = &h->hugepage_freelists[i];
@@ -982,6 +1056,9 @@ static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count)
 {
 	unsigned long min_count, ret;
 
+	if (h->order >= MAX_ORDER)
+		return h->max_huge_pages;
+
 	/*
 	 * Increase the pool size
 	 * First take pages out of surplus state.  Then make up the
@@ -1210,6 +1287,8 @@ static int __init hugetlb_init(void)
 
 	hugetlb_init_hstates();
 
+	gather_bootmem_prealloc();
+
 	report_hugepages();
 
 	hugetlb_sysfs_init();

From 8faa8b077b2cdc4e4646842fe50b07840955a013 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 23 Jul 2008 21:27:48 -0700
Subject: [PATCH 0258/1171] hugetlb: support boot allocate different sizes

Make some infrastructure changes to allow boot-time allocation of
different hugepage page sizes.

- move all basic hstate initialisation into hugetlb_add_hstate
- create a new function hugetlb_hstate_alloc_pages() to do the
  actual initial page allocations. Call this function early in
  order to allocate giant pages from bootmem.
- Check for multiple hugepages= parameters

Acked-by: Adam Litke <agl@us.ibm.com>
Acked-by: Nishanth Aravamudan <nacc@us.ibm.com>
Acked-by: Andrew Hastings <abh@cray.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/hugetlb.c | 39 ++++++++++++++++++++++++++++++---------
 1 file changed, 30 insertions(+), 9 deletions(-)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 1a6fe87555b2..243a8684d180 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -981,15 +981,10 @@ static void __init gather_bootmem_prealloc(void)
 	}
 }
 
-static void __init hugetlb_init_one_hstate(struct hstate *h)
+static void __init hugetlb_hstate_alloc_pages(struct hstate *h)
 {
 	unsigned long i;
 
-	for (i = 0; i < MAX_NUMNODES; ++i)
-		INIT_LIST_HEAD(&h->hugepage_freelists[i]);
-
-	h->hugetlb_next_nid = first_node(node_online_map);
-
 	for (i = 0; i < h->max_huge_pages; ++i) {
 		if (h->order >= MAX_ORDER) {
 			if (!alloc_bootmem_huge_page(h))
@@ -997,7 +992,7 @@ static void __init hugetlb_init_one_hstate(struct hstate *h)
 		} else if (!alloc_fresh_huge_page(h))
 			break;
 	}
-	h->max_huge_pages = h->free_huge_pages = h->nr_huge_pages = i;
+	h->max_huge_pages = i;
 }
 
 static void __init hugetlb_init_hstates(void)
@@ -1005,7 +1000,9 @@ static void __init hugetlb_init_hstates(void)
 	struct hstate *h;
 
 	for_each_hstate(h) {
-		hugetlb_init_one_hstate(h);
+		/* oversize hugepages were init'ed in early boot */
+		if (h->order < MAX_ORDER)
+			hugetlb_hstate_alloc_pages(h);
 	}
 }
 
@@ -1301,6 +1298,8 @@ module_init(hugetlb_init);
 void __init hugetlb_add_hstate(unsigned order)
 {
 	struct hstate *h;
+	unsigned long i;
+
 	if (size_to_hstate(PAGE_SIZE << order)) {
 		printk(KERN_WARNING "hugepagesz= specified twice, ignoring\n");
 		return;
@@ -1310,15 +1309,21 @@ void __init hugetlb_add_hstate(unsigned order)
 	h = &hstates[max_hstate++];
 	h->order = order;
 	h->mask = ~((1ULL << (order + PAGE_SHIFT)) - 1);
+	h->nr_huge_pages = 0;
+	h->free_huge_pages = 0;
+	for (i = 0; i < MAX_NUMNODES; ++i)
+		INIT_LIST_HEAD(&h->hugepage_freelists[i]);
+	h->hugetlb_next_nid = first_node(node_online_map);
 	snprintf(h->name, HSTATE_NAME_LEN, "hugepages-%lukB",
 					huge_page_size(h)/1024);
-	hugetlb_init_one_hstate(h);
+
 	parsed_hstate = h;
 }
 
 static int __init hugetlb_setup(char *s)
 {
 	unsigned long *mhp;
+	static unsigned long *last_mhp;
 
 	/*
 	 * !max_hstate means we haven't parsed a hugepagesz= parameter yet,
@@ -1329,9 +1334,25 @@ static int __init hugetlb_setup(char *s)
 	else
 		mhp = &parsed_hstate->max_huge_pages;
 
+	if (mhp == last_mhp) {
+		printk(KERN_WARNING "hugepages= specified twice without "
+			"interleaving hugepagesz=, ignoring\n");
+		return 1;
+	}
+
 	if (sscanf(s, "%lu", mhp) <= 0)
 		*mhp = 0;
 
+	/*
+	 * Global state is always initialized later in hugetlb_init.
+	 * But we need to allocate >= MAX_ORDER hstates here early to still
+	 * use the bootmem allocator.
+	 */
+	if (max_hstate && parsed_hstate->order >= MAX_ORDER)
+		hugetlb_hstate_alloc_pages(parsed_hstate);
+
+	last_mhp = mhp;
+
 	return 1;
 }
 __setup("hugepages=", hugetlb_setup);

From 4abd32dbab201c3ced0b0af12accea77cd9eeffc Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 23 Jul 2008 21:27:49 -0700
Subject: [PATCH 0259/1171] hugetlb: printk cleanup

- Reword sentence to clarify meaning with multiple options
- Add support for using GB prefixes for the page size
- Add extra printk to delayed > MAX_ORDER allocation code

Acked-by: Adam Litke <agl@us.ibm.com>
Acked-by: Nishanth Aravamudan <nacc@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/hugetlb.c | 20 ++++++++++++++++----
 1 file changed, 16 insertions(+), 4 deletions(-)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 243a8684d180..0c74c14dd2f7 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -1006,15 +1006,27 @@ static void __init hugetlb_init_hstates(void)
 	}
 }
 
+static char * __init memfmt(char *buf, unsigned long n)
+{
+	if (n >= (1UL << 30))
+		sprintf(buf, "%lu GB", n >> 30);
+	else if (n >= (1UL << 20))
+		sprintf(buf, "%lu MB", n >> 20);
+	else
+		sprintf(buf, "%lu KB", n >> 10);
+	return buf;
+}
+
 static void __init report_hugepages(void)
 {
 	struct hstate *h;
 
 	for_each_hstate(h) {
-		printk(KERN_INFO "Total HugeTLB memory allocated, "
-				"%ld %dMB pages\n",
-				h->free_huge_pages,
-				1 << (h->order + PAGE_SHIFT - 20));
+		char buf[32];
+		printk(KERN_INFO "HugeTLB registered %s page size, "
+				 "pre-allocated %ld pages\n",
+			memfmt(buf, huge_page_size(h)),
+			h->free_huge_pages);
 	}
 }
 

From ceb868796181dc95ea01a110e123afd391639873 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 23 Jul 2008 21:27:50 -0700
Subject: [PATCH 0260/1171] hugetlb: introduce pud_huge

Straight forward extensions for huge pages located in the PUD instead of
PMDs.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/ia64/mm/hugetlbpage.c    |  6 ++++++
 arch/powerpc/mm/hugetlbpage.c |  5 +++++
 arch/s390/mm/hugetlbpage.c    |  5 +++++
 arch/sh/mm/hugetlbpage.c      |  5 +++++
 arch/sparc64/mm/hugetlbpage.c |  5 +++++
 arch/x86/mm/hugetlbpage.c     | 25 ++++++++++++++++++++++++-
 include/linux/hugetlb.h       |  5 +++++
 mm/hugetlb.c                  |  9 +++++++++
 mm/memory.c                   | 15 +++++++++++----
 9 files changed, 75 insertions(+), 5 deletions(-)

diff --git a/arch/ia64/mm/hugetlbpage.c b/arch/ia64/mm/hugetlbpage.c
index 6170f097d255..c45fc7f5a979 100644
--- a/arch/ia64/mm/hugetlbpage.c
+++ b/arch/ia64/mm/hugetlbpage.c
@@ -107,6 +107,12 @@ int pmd_huge(pmd_t pmd)
 {
 	return 0;
 }
+
+int pud_huge(pud_t pud)
+{
+	return 0;
+}
+
 struct page *
 follow_huge_pmd(struct mm_struct *mm, unsigned long address, pmd_t *pmd, int write)
 {
diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c
index c94dc71af989..63db7adce717 100644
--- a/arch/powerpc/mm/hugetlbpage.c
+++ b/arch/powerpc/mm/hugetlbpage.c
@@ -369,6 +369,11 @@ int pmd_huge(pmd_t pmd)
 	return 0;
 }
 
+int pud_huge(pud_t pud)
+{
+	return 0;
+}
+
 struct page *
 follow_huge_pmd(struct mm_struct *mm, unsigned long address,
 		pmd_t *pmd, int write)
diff --git a/arch/s390/mm/hugetlbpage.c b/arch/s390/mm/hugetlbpage.c
index 9162dc84f77f..f28c43d2f61d 100644
--- a/arch/s390/mm/hugetlbpage.c
+++ b/arch/s390/mm/hugetlbpage.c
@@ -120,6 +120,11 @@ int pmd_huge(pmd_t pmd)
 	return !!(pmd_val(pmd) & _SEGMENT_ENTRY_LARGE);
 }
 
+int pud_huge(pud_t pud)
+{
+	return 0;
+}
+
 struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address,
 			     pmd_t *pmdp, int write)
 {
diff --git a/arch/sh/mm/hugetlbpage.c b/arch/sh/mm/hugetlbpage.c
index 2f9dbe0ef4ac..9304117039c4 100644
--- a/arch/sh/mm/hugetlbpage.c
+++ b/arch/sh/mm/hugetlbpage.c
@@ -79,6 +79,11 @@ int pmd_huge(pmd_t pmd)
 	return 0;
 }
 
+int pud_huge(pud_t pud)
+{
+	return 0;
+}
+
 struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address,
 			     pmd_t *pmd, int write)
 {
diff --git a/arch/sparc64/mm/hugetlbpage.c b/arch/sparc64/mm/hugetlbpage.c
index 1307b23f6a76..f27d10369e0c 100644
--- a/arch/sparc64/mm/hugetlbpage.c
+++ b/arch/sparc64/mm/hugetlbpage.c
@@ -295,6 +295,11 @@ int pmd_huge(pmd_t pmd)
 	return 0;
 }
 
+int pud_huge(pud_t pud)
+{
+	return 0;
+}
+
 struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address,
 			     pmd_t *pmd, int write)
 {
diff --git a/arch/x86/mm/hugetlbpage.c b/arch/x86/mm/hugetlbpage.c
index 52476fde8996..a4789e87a315 100644
--- a/arch/x86/mm/hugetlbpage.c
+++ b/arch/x86/mm/hugetlbpage.c
@@ -189,6 +189,11 @@ int pmd_huge(pmd_t pmd)
 	return 0;
 }
 
+int pud_huge(pud_t pud)
+{
+	return 0;
+}
+
 struct page *
 follow_huge_pmd(struct mm_struct *mm, unsigned long address,
 		pmd_t *pmd, int write)
@@ -209,6 +214,11 @@ int pmd_huge(pmd_t pmd)
 	return !!(pmd_val(pmd) & _PAGE_PSE);
 }
 
+int pud_huge(pud_t pud)
+{
+	return 0;
+}
+
 struct page *
 follow_huge_pmd(struct mm_struct *mm, unsigned long address,
 		pmd_t *pmd, int write)
@@ -217,9 +227,22 @@ follow_huge_pmd(struct mm_struct *mm, unsigned long address,
 
 	page = pte_page(*(pte_t *)pmd);
 	if (page)
-		page += ((address & ~HPAGE_MASK) >> PAGE_SHIFT);
+		page += ((address & ~PMD_MASK) >> PAGE_SHIFT);
 	return page;
 }
+
+struct page *
+follow_huge_pud(struct mm_struct *mm, unsigned long address,
+		pud_t *pud, int write)
+{
+	struct page *page;
+
+	page = pte_page(*(pte_t *)pud);
+	if (page)
+		page += ((address & ~PUD_MASK) >> PAGE_SHIFT);
+	return page;
+}
+
 #endif
 
 /* x86_64 also uses this file */
diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index 58c0de32e7f0..b2c17f62cacb 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -50,7 +50,10 @@ struct page *follow_huge_addr(struct mm_struct *mm, unsigned long address,
 			      int write);
 struct page *follow_huge_pmd(struct mm_struct *mm, unsigned long address,
 				pmd_t *pmd, int write);
+struct page *follow_huge_pud(struct mm_struct *mm, unsigned long address,
+				pud_t *pud, int write);
 int pmd_huge(pmd_t pmd);
+int pud_huge(pud_t pmd);
 void hugetlb_change_protection(struct vm_area_struct *vma,
 		unsigned long address, unsigned long end, pgprot_t newprot);
 
@@ -78,8 +81,10 @@ static inline unsigned long hugetlb_total_pages(void)
 #define hugetlb_report_meminfo(buf)		0
 #define hugetlb_report_node_meminfo(n, buf)	0
 #define follow_huge_pmd(mm, addr, pmd, write)	NULL
+#define follow_huge_pud(mm, addr, pud, write)	NULL
 #define prepare_hugepage_range(file, addr, len)	(-EINVAL)
 #define pmd_huge(x)	0
+#define pud_huge(x)	0
 #define is_hugepage_only_range(mm, addr, len)	0
 #define hugetlb_free_pgd_range(tlb, addr, end, floor, ceiling) ({BUG(); 0; })
 #define hugetlb_fault(mm, vma, addr, write)	({ BUG(); 0; })
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 0c74c14dd2f7..107c1ce223cb 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -1996,6 +1996,15 @@ int hugetlb_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 	return ret;
 }
 
+/* Can be overriden by architectures */
+__attribute__((weak)) struct page *
+follow_huge_pud(struct mm_struct *mm, unsigned long address,
+	       pud_t *pud, int write)
+{
+	BUG();
+	return NULL;
+}
+
 int follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
 			struct page **pages, struct vm_area_struct **vmas,
 			unsigned long *position, int *length, int i,
diff --git a/mm/memory.c b/mm/memory.c
index 02fc6b1047b0..262e3eb6601a 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -998,19 +998,24 @@ struct page *follow_page(struct vm_area_struct *vma, unsigned long address,
 		goto no_page_table;
 
 	pud = pud_offset(pgd, address);
-	if (pud_none(*pud) || unlikely(pud_bad(*pud)))
+	if (pud_none(*pud))
 		goto no_page_table;
-	
+	if (pud_huge(*pud)) {
+		BUG_ON(flags & FOLL_GET);
+		page = follow_huge_pud(mm, address, pud, flags & FOLL_WRITE);
+		goto out;
+	}
+	if (unlikely(pud_bad(*pud)))
+		goto no_page_table;
+
 	pmd = pmd_offset(pud, address);
 	if (pmd_none(*pmd))
 		goto no_page_table;
-
 	if (pmd_huge(*pmd)) {
 		BUG_ON(flags & FOLL_GET);
 		page = follow_huge_pmd(mm, address, pmd, flags & FOLL_WRITE);
 		goto out;
 	}
-
 	if (unlikely(pmd_bad(*pmd)))
 		goto no_page_table;
 
@@ -1567,6 +1572,8 @@ static int apply_to_pmd_range(struct mm_struct *mm, pud_t *pud,
 	unsigned long next;
 	int err;
 
+	BUG_ON(pud_huge(*pud));
+
 	pmd = pmd_alloc(mm, pud, addr);
 	if (!pmd)
 		return -ENOMEM;

From 39c11e6c05b7fedbf7ed4df3908b25f622d56204 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 23 Jul 2008 21:27:50 -0700
Subject: [PATCH 0261/1171] x86: support GB hugepages on 64-bit

Acked-by: Adam Litke <agl@us.ibm.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/mm/hugetlbpage.c | 33 ++++++++++++++++++++++-----------
 1 file changed, 22 insertions(+), 11 deletions(-)

diff --git a/arch/x86/mm/hugetlbpage.c b/arch/x86/mm/hugetlbpage.c
index a4789e87a315..b7a65a07af03 100644
--- a/arch/x86/mm/hugetlbpage.c
+++ b/arch/x86/mm/hugetlbpage.c
@@ -134,9 +134,14 @@ pte_t *huge_pte_alloc(struct mm_struct *mm,
 	pgd = pgd_offset(mm, addr);
 	pud = pud_alloc(mm, pgd, addr);
 	if (pud) {
-		if (pud_none(*pud))
-			huge_pmd_share(mm, addr, pud);
-		pte = (pte_t *) pmd_alloc(mm, pud, addr);
+		if (sz == PUD_SIZE) {
+			pte = (pte_t *)pud;
+		} else {
+			BUG_ON(sz != PMD_SIZE);
+			if (pud_none(*pud))
+				huge_pmd_share(mm, addr, pud);
+			pte = (pte_t *) pmd_alloc(mm, pud, addr);
+		}
 	}
 	BUG_ON(pte && !pte_none(*pte) && !pte_huge(*pte));
 
@@ -152,8 +157,11 @@ pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
 	pgd = pgd_offset(mm, addr);
 	if (pgd_present(*pgd)) {
 		pud = pud_offset(pgd, addr);
-		if (pud_present(*pud))
+		if (pud_present(*pud)) {
+			if (pud_large(*pud))
+				return (pte_t *)pud;
 			pmd = pmd_offset(pud, addr);
+		}
 	}
 	return (pte_t *) pmd;
 }
@@ -216,7 +224,7 @@ int pmd_huge(pmd_t pmd)
 
 int pud_huge(pud_t pud)
 {
-	return 0;
+	return !!(pud_val(pud) & _PAGE_PSE);
 }
 
 struct page *
@@ -252,6 +260,7 @@ static unsigned long hugetlb_get_unmapped_area_bottomup(struct file *file,
 		unsigned long addr, unsigned long len,
 		unsigned long pgoff, unsigned long flags)
 {
+	struct hstate *h = hstate_file(file);
 	struct mm_struct *mm = current->mm;
 	struct vm_area_struct *vma;
 	unsigned long start_addr;
@@ -264,7 +273,7 @@ static unsigned long hugetlb_get_unmapped_area_bottomup(struct file *file,
 	}
 
 full_search:
-	addr = ALIGN(start_addr, HPAGE_SIZE);
+	addr = ALIGN(start_addr, huge_page_size(h));
 
 	for (vma = find_vma(mm, addr); ; vma = vma->vm_next) {
 		/* At this point:  (!vma || addr < vma->vm_end). */
@@ -286,7 +295,7 @@ full_search:
 		}
 		if (addr + mm->cached_hole_size < vma->vm_start)
 		        mm->cached_hole_size = vma->vm_start - addr;
-		addr = ALIGN(vma->vm_end, HPAGE_SIZE);
+		addr = ALIGN(vma->vm_end, huge_page_size(h));
 	}
 }
 
@@ -294,6 +303,7 @@ static unsigned long hugetlb_get_unmapped_area_topdown(struct file *file,
 		unsigned long addr0, unsigned long len,
 		unsigned long pgoff, unsigned long flags)
 {
+	struct hstate *h = hstate_file(file);
 	struct mm_struct *mm = current->mm;
 	struct vm_area_struct *vma, *prev_vma;
 	unsigned long base = mm->mmap_base, addr = addr0;
@@ -314,7 +324,7 @@ try_again:
 		goto fail;
 
 	/* either no address requested or cant fit in requested address hole */
-	addr = (mm->free_area_cache - len) & HPAGE_MASK;
+	addr = (mm->free_area_cache - len) & huge_page_mask(h);
 	do {
 		/*
 		 * Lookup failure means no vma is above this address,
@@ -345,7 +355,7 @@ try_again:
 		        largest_hole = vma->vm_start - addr;
 
 		/* try just below the current vma->vm_start */
-		addr = (vma->vm_start - len) & HPAGE_MASK;
+		addr = (vma->vm_start - len) & huge_page_mask(h);
 	} while (len <= vma->vm_start);
 
 fail:
@@ -383,10 +393,11 @@ unsigned long
 hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
 		unsigned long len, unsigned long pgoff, unsigned long flags)
 {
+	struct hstate *h = hstate_file(file);
 	struct mm_struct *mm = current->mm;
 	struct vm_area_struct *vma;
 
-	if (len & ~HPAGE_MASK)
+	if (len & ~huge_page_mask(h))
 		return -EINVAL;
 	if (len > TASK_SIZE)
 		return -ENOMEM;
@@ -398,7 +409,7 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
 	}
 
 	if (addr) {
-		addr = ALIGN(addr, HPAGE_SIZE);
+		addr = ALIGN(addr, huge_page_size(h));
 		vma = find_vma(mm, addr);
 		if (TASK_SIZE - len >= addr &&
 		    (!vma || addr + len <= vma->vm_start))

From b4718e628dbf68a2dee23b5709e2aa3190409c56 Mon Sep 17 00:00:00 2001
From: Andi Kleen <ak@suse.de>
Date: Wed, 23 Jul 2008 21:27:51 -0700
Subject: [PATCH 0262/1171] x86: add hugepagesz option on 64-bit

Add an hugepagesz=...  option similar to IA64, PPC etc.  to x86-64.

This finally allows to select GB pages for hugetlbfs in x86 now that all
the infrastructure is in place.

Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/kernel-parameters.txt | 11 +++++++++--
 arch/x86/mm/hugetlbpage.c           | 17 +++++++++++++++++
 include/asm-x86/page.h              |  2 ++
 3 files changed, 28 insertions(+), 2 deletions(-)

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 5e20ccb5a736..d55fd88fd0a9 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -774,8 +774,15 @@ and is between 256 and 4096 characters. It is defined in the file
 	hisax=		[HW,ISDN]
 			See Documentation/isdn/README.HiSax.
 
-	hugepages=	[HW,X86-32,IA-64] Maximal number of HugeTLB pages.
-	hugepagesz=	[HW,IA-64,PPC] The size of the HugeTLB pages.
+	hugepages=	[HW,X86-32,IA-64] HugeTLB pages to allocate at boot.
+	hugepagesz=	[HW,IA-64,PPC,X86-64] The size of the HugeTLB pages.
+			On x86 this option can be specified multiple times
+			interleaved with hugepages= to reserve huge pages
+			of different sizes. Valid pages sizes on x86-64
+			are 2M (when the CPU supports "pse") and 1G (when the
+			CPU supports the "pdpe1gb" cpuinfo flag)
+			Note that 1GB pages can only be allocated at boot time
+			using hugepages= and not freed afterwards.
 
 	i8042.direct	[HW] Put keyboard port into non-translated mode
 	i8042.dumbkbd	[HW] Pretend that controller can only read data from
diff --git a/arch/x86/mm/hugetlbpage.c b/arch/x86/mm/hugetlbpage.c
index b7a65a07af03..8f307d914c2e 100644
--- a/arch/x86/mm/hugetlbpage.c
+++ b/arch/x86/mm/hugetlbpage.c
@@ -425,3 +425,20 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
 
 #endif /*HAVE_ARCH_HUGETLB_UNMAPPED_AREA*/
 
+#ifdef CONFIG_X86_64
+static __init int setup_hugepagesz(char *opt)
+{
+	unsigned long ps = memparse(opt, &opt);
+	if (ps == PMD_SIZE) {
+		hugetlb_add_hstate(PMD_SHIFT - PAGE_SHIFT);
+	} else if (ps == PUD_SIZE && cpu_has_gbpages) {
+		hugetlb_add_hstate(PUD_SHIFT - PAGE_SHIFT);
+	} else {
+		printk(KERN_ERR "hugepagesz: Unsupported page size %lu M\n",
+			ps >> 20);
+		return 0;
+	}
+	return 1;
+}
+__setup("hugepagesz=", setup_hugepagesz);
+#endif
diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h
index 6c846228948d..6e02098b1605 100644
--- a/include/asm-x86/page.h
+++ b/include/asm-x86/page.h
@@ -32,6 +32,8 @@
 #define HPAGE_MASK		(~(HPAGE_SIZE - 1))
 #define HUGETLB_PAGE_ORDER	(HPAGE_SHIFT - PAGE_SHIFT)
 
+#define HUGE_MAX_HSTATE 2
+
 /* to align the pointer to the (next) page boundary */
 #define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
 

From e11bfbfcb08ef4223b863799897c19cdf7c5bc00 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Wed, 23 Jul 2008 21:27:52 -0700
Subject: [PATCH 0263/1171] hugetlb: override default huge page size

Allow configurations with the default huge page size which is different to
the traditional HPAGE_SIZE size.  The default huge page size is the one
represented in the legacy /proc ABIs, SHM, and which is defaulted to when
mounting hugetlbfs filesystems.

This is implemented with a new kernel option default_hugepagesz=, which
defaults to HPAGE_SIZE if not specified.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/kernel-parameters.txt |  7 +++++++
 mm/hugetlb.c                        | 23 +++++++++++++++++------
 2 files changed, 24 insertions(+), 6 deletions(-)

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index d55fd88fd0a9..30278e9e5211 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -783,6 +783,13 @@ and is between 256 and 4096 characters. It is defined in the file
 			CPU supports the "pdpe1gb" cpuinfo flag)
 			Note that 1GB pages can only be allocated at boot time
 			using hugepages= and not freed afterwards.
+	default_hugepagesz=
+			[same as hugepagesz=] The size of the default
+			HugeTLB page size. This is the size represented by
+			the legacy /proc/ hugepages APIs, used for SHM, and
+			default size when mounting hugetlbfs filesystems.
+			Defaults to the default architecture's huge page size
+			if not specified.
 
 	i8042.direct	[HW] Put keyboard port into non-translated mode
 	i8042.dumbkbd	[HW] Pretend that controller can only read data from
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 107c1ce223cb..2a2f6e869401 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -34,6 +34,7 @@ struct hstate hstates[HUGE_MAX_HSTATE];
 /* for command line parsing */
 static struct hstate * __initdata parsed_hstate;
 static unsigned long __initdata default_hstate_max_huge_pages;
+static unsigned long __initdata default_hstate_size;
 
 #define for_each_hstate(h) \
 	for ((h) = hstates; (h) < &hstates[max_hstate]; (h)++)
@@ -1288,11 +1289,14 @@ static int __init hugetlb_init(void)
 {
 	BUILD_BUG_ON(HPAGE_SHIFT == 0);
 
-	if (!size_to_hstate(HPAGE_SIZE)) {
-		hugetlb_add_hstate(HUGETLB_PAGE_ORDER);
-		parsed_hstate->max_huge_pages = default_hstate_max_huge_pages;
+	if (!size_to_hstate(default_hstate_size)) {
+		default_hstate_size = HPAGE_SIZE;
+		if (!size_to_hstate(default_hstate_size))
+			hugetlb_add_hstate(HUGETLB_PAGE_ORDER);
 	}
-	default_hstate_idx = size_to_hstate(HPAGE_SIZE) - hstates;
+	default_hstate_idx = size_to_hstate(default_hstate_size) - hstates;
+	if (default_hstate_max_huge_pages)
+		default_hstate.max_huge_pages = default_hstate_max_huge_pages;
 
 	hugetlb_init_hstates();
 
@@ -1332,7 +1336,7 @@ void __init hugetlb_add_hstate(unsigned order)
 	parsed_hstate = h;
 }
 
-static int __init hugetlb_setup(char *s)
+static int __init hugetlb_nrpages_setup(char *s)
 {
 	unsigned long *mhp;
 	static unsigned long *last_mhp;
@@ -1367,7 +1371,14 @@ static int __init hugetlb_setup(char *s)
 
 	return 1;
 }
-__setup("hugepages=", hugetlb_setup);
+__setup("hugepages=", hugetlb_nrpages_setup);
+
+static int __init hugetlb_default_setup(char *s)
+{
+	default_hstate_size = memparse(s, &s);
+	return 1;
+}
+__setup("default_hugepagesz=", hugetlb_default_setup);
 
 static unsigned int cpuset_mems_nr(unsigned int *array)
 {

From 53ba51d21d6e048424ab8aadfebdb1f25ae07b60 Mon Sep 17 00:00:00 2001
From: Jon Tollefson <kniht@linux.vnet.ibm.com>
Date: Wed, 23 Jul 2008 21:27:52 -0700
Subject: [PATCH 0264/1171] hugetlb: allow arch overridden hugepage allocation

Allow alloc_bootmem_huge_page() to be overridden by architectures that
can't always use bootmem.  This requires huge_boot_pages to be available
for use by this function.

This is required for powerpc 16G pages, which have to be reserved prior to
boot-time.  The location of these pages are indicated in the device tree.

Acked-by: Adam Litke <agl@us.ibm.com>
Signed-off-by: Jon Tollefson <kniht@linux.vnet.ibm.com>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/hugetlb.h | 10 ++++++++++
 mm/hugetlb.c            | 11 +++--------
 2 files changed, 13 insertions(+), 8 deletions(-)

diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index b2c17f62cacb..9a71d4cc88c8 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -39,6 +39,7 @@ void hugetlb_unreserve_pages(struct inode *inode, long offset, long freed);
 extern unsigned long hugepages_treat_as_movable;
 extern const unsigned long hugetlb_zero, hugetlb_infinity;
 extern int sysctl_hugetlb_shm_group;
+extern struct list_head huge_boot_pages;
 
 /* arch callbacks */
 
@@ -188,6 +189,14 @@ struct hstate {
 	char name[HSTATE_NAME_LEN];
 };
 
+struct huge_bootmem_page {
+	struct list_head list;
+	struct hstate *hstate;
+};
+
+/* arch callback */
+int __init alloc_bootmem_huge_page(struct hstate *h);
+
 void __init hugetlb_add_hstate(unsigned order);
 struct hstate *size_to_hstate(unsigned long size);
 
@@ -256,6 +265,7 @@ static inline struct hstate *page_hstate(struct page *page)
 
 #else
 struct hstate {};
+#define alloc_bootmem_huge_page(h) NULL
 #define hstate_file(f) NULL
 #define hstate_vma(v) NULL
 #define hstate_inode(i) NULL
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 2a2f6e869401..3e1506b808a3 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -31,6 +31,8 @@ static int max_hstate;
 unsigned int default_hstate_idx;
 struct hstate hstates[HUGE_MAX_HSTATE];
 
+__initdata LIST_HEAD(huge_boot_pages);
+
 /* for command line parsing */
 static struct hstate * __initdata parsed_hstate;
 static unsigned long __initdata default_hstate_max_huge_pages;
@@ -925,14 +927,7 @@ static struct page *alloc_huge_page(struct vm_area_struct *vma,
 	return page;
 }
 
-static __initdata LIST_HEAD(huge_boot_pages);
-
-struct huge_bootmem_page {
-	struct list_head list;
-	struct hstate *hstate;
-};
-
-static int __init alloc_bootmem_huge_page(struct hstate *h)
+__attribute__((weak)) int alloc_bootmem_huge_page(struct hstate *h)
 {
 	struct huge_bootmem_page *m;
 	int nr_nodes = nodes_weight(node_online_map);

From ec4b2c0c8312d1118c2acd00c89988ecf955d5cc Mon Sep 17 00:00:00 2001
From: Jon Tollefson <kniht@linux.vnet.ibm.com>
Date: Wed, 23 Jul 2008 21:27:53 -0700
Subject: [PATCH 0265/1171] powerpc: function to allocate gigantic hugepages

The 16G page locations have been saved during early boot in an array.  The
alloc_bootmem_huge_page() function adds a page from here to the
huge_boot_pages list.

Acked-by: Adam Litke <agl@us.ibm.com>
Signed-off-by: Jon Tollefson <kniht@linux.vnet.ibm.com>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/powerpc/mm/hugetlbpage.c | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)

diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c
index 63db7adce717..5df82186fc93 100644
--- a/arch/powerpc/mm/hugetlbpage.c
+++ b/arch/powerpc/mm/hugetlbpage.c
@@ -29,6 +29,12 @@
 
 #define NUM_LOW_AREAS	(0x100000000UL >> SID_SHIFT)
 #define NUM_HIGH_AREAS	(PGTABLE_RANGE >> HTLB_AREA_SHIFT)
+#define MAX_NUMBER_GPAGES	1024
+
+/* Tracks the 16G pages after the device tree is scanned and before the
+ * huge_boot_pages list is ready.  */
+static unsigned long gpage_freearray[MAX_NUMBER_GPAGES];
+static unsigned nr_gpages;
 
 unsigned int hugepte_shift;
 #define PTRS_PER_HUGEPTE	(1 << hugepte_shift)
@@ -104,6 +110,21 @@ pmd_t *hpmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long addr)
 }
 #endif
 
+/* Moves the gigantic page addresses from the temporary list to the
+ * huge_boot_pages list.  */
+int alloc_bootmem_huge_page(struct hstate *h)
+{
+	struct huge_bootmem_page *m;
+	if (nr_gpages == 0)
+		return 0;
+	m = phys_to_virt(gpage_freearray[--nr_gpages]);
+	gpage_freearray[nr_gpages] = 0;
+	list_add(&m->list, &huge_boot_pages);
+	m->hstate = h;
+	return 1;
+}
+
+
 /* Modelled after find_linux_pte() */
 pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
 {

From 658013e93eb70494f7300bc90457b09a807232a4 Mon Sep 17 00:00:00 2001
From: Jon Tollefson <kniht@linux.vnet.ibm.com>
Date: Wed, 23 Jul 2008 21:27:54 -0700
Subject: [PATCH 0266/1171] powerpc: scan device tree for gigantic pages

The 16G huge pages have to be reserved in the HMC prior to boot.  The
location of the pages are placed in the device tree.  This patch adds code
to scan the device tree during very early boot and save these page
locations until hugetlbfs is ready for them.

Acked-by: Adam Litke <agl@us.ibm.com>
Signed-off-by: Jon Tollefson <kniht@linux.vnet.ibm.com>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/powerpc/mm/hash_utils_64.c  | 44 +++++++++++++++++++++++++++++++-
 arch/powerpc/mm/hugetlbpage.c    | 16 ++++++++++++
 include/asm-powerpc/mmu-hash64.h |  2 ++
 3 files changed, 61 insertions(+), 1 deletion(-)

diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c
index 8d3b58ebd38e..ae4c717243a5 100644
--- a/arch/powerpc/mm/hash_utils_64.c
+++ b/arch/powerpc/mm/hash_utils_64.c
@@ -68,6 +68,7 @@
 
 #define KB (1024)
 #define MB (1024*KB)
+#define GB (1024L*MB)
 
 /*
  * Note:  pte   --> Linux PTE
@@ -329,6 +330,44 @@ static int __init htab_dt_scan_page_sizes(unsigned long node,
 	return 0;
 }
 
+/* Scan for 16G memory blocks that have been set aside for huge pages
+ * and reserve those blocks for 16G huge pages.
+ */
+static int __init htab_dt_scan_hugepage_blocks(unsigned long node,
+					const char *uname, int depth,
+					void *data) {
+	char *type = of_get_flat_dt_prop(node, "device_type", NULL);
+	unsigned long *addr_prop;
+	u32 *page_count_prop;
+	unsigned int expected_pages;
+	long unsigned int phys_addr;
+	long unsigned int block_size;
+
+	/* We are scanning "memory" nodes only */
+	if (type == NULL || strcmp(type, "memory") != 0)
+		return 0;
+
+	/* This property is the log base 2 of the number of virtual pages that
+	 * will represent this memory block. */
+	page_count_prop = of_get_flat_dt_prop(node, "ibm,expected#pages", NULL);
+	if (page_count_prop == NULL)
+		return 0;
+	expected_pages = (1 << page_count_prop[0]);
+	addr_prop = of_get_flat_dt_prop(node, "reg", NULL);
+	if (addr_prop == NULL)
+		return 0;
+	phys_addr = addr_prop[0];
+	block_size = addr_prop[1];
+	if (block_size != (16 * GB))
+		return 0;
+	printk(KERN_INFO "Huge page(16GB) memory: "
+			"addr = 0x%lX size = 0x%lX pages = %d\n",
+			phys_addr, block_size, expected_pages);
+	lmb_reserve(phys_addr, block_size * expected_pages);
+	add_gpage(phys_addr, block_size, expected_pages);
+	return 0;
+}
+
 static void __init htab_init_page_sizes(void)
 {
 	int rc;
@@ -418,7 +457,10 @@ static void __init htab_init_page_sizes(void)
 	       );
 
 #ifdef CONFIG_HUGETLB_PAGE
-	/* Init large page size. Currently, we pick 16M or 1M depending
+	/* Reserve 16G huge page memory sections for huge pages */
+	of_scan_flat_dt(htab_dt_scan_hugepage_blocks, NULL);
+
+/* Init large page size. Currently, we pick 16M or 1M depending
 	 * on what is available
 	 */
 	if (mmu_psize_defs[MMU_PAGE_16M].shift)
diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c
index 5df82186fc93..e2a650a9e533 100644
--- a/arch/powerpc/mm/hugetlbpage.c
+++ b/arch/powerpc/mm/hugetlbpage.c
@@ -110,6 +110,22 @@ pmd_t *hpmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long addr)
 }
 #endif
 
+/* Build list of addresses of gigantic pages.  This function is used in early
+ * boot before the buddy or bootmem allocator is setup.
+ */
+void add_gpage(unsigned long addr, unsigned long page_size,
+	unsigned long number_of_pages)
+{
+	if (!addr)
+		return;
+	while (number_of_pages > 0) {
+		gpage_freearray[nr_gpages] = addr;
+		nr_gpages++;
+		number_of_pages--;
+		addr += page_size;
+	}
+}
+
 /* Moves the gigantic page addresses from the temporary list to the
  * huge_boot_pages list.  */
 int alloc_bootmem_huge_page(struct hstate *h)
diff --git a/include/asm-powerpc/mmu-hash64.h b/include/asm-powerpc/mmu-hash64.h
index d1dc16afb118..b61181aa7746 100644
--- a/include/asm-powerpc/mmu-hash64.h
+++ b/include/asm-powerpc/mmu-hash64.h
@@ -281,6 +281,8 @@ extern int htab_bolt_mapping(unsigned long vstart, unsigned long vend,
 			     unsigned long pstart, unsigned long mode,
 			     int psize, int ssize);
 extern void set_huge_psize(int psize);
+extern void add_gpage(unsigned long addr, unsigned long page_size,
+			  unsigned long number_of_pages);
 extern void demote_segment_4k(struct mm_struct *mm, unsigned long addr);
 
 extern void htab_initialize(void);

From 91224346aa8c1cdaa660300a98e0b074a3a95030 Mon Sep 17 00:00:00 2001
From: Jon Tollefson <kniht@linux.vnet.ibm.com>
Date: Wed, 23 Jul 2008 21:27:55 -0700
Subject: [PATCH 0267/1171] powerpc: define support for 16G hugepages

The huge page size is defined for 16G pages.  If a hugepagesz of 16G is
specified at boot-time then it becomes the huge page size instead of the
default 16M.

The change in pgtable-64K.h is to the macro pte_iterate_hashed_subpages to
make the increment to va (the 1 being shifted) be a long so that it is not
shifted to 0.  Otherwise it would create an infinite loop when the shift
value is for a 16G page (when base page size is 64K).

Signed-off-by: Jon Tollefson <kniht@linux.vnet.ibm.com>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/powerpc/mm/hugetlbpage.c     | 62 ++++++++++++++++++++++---------
 include/asm-powerpc/pgtable-64k.h |  2 +-
 2 files changed, 45 insertions(+), 19 deletions(-)

diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c
index e2a650a9e533..19b1a9cec6d5 100644
--- a/arch/powerpc/mm/hugetlbpage.c
+++ b/arch/powerpc/mm/hugetlbpage.c
@@ -24,8 +24,9 @@
 #include <asm/cputable.h>
 #include <asm/spu.h>
 
-#define HPAGE_SHIFT_64K	16
-#define HPAGE_SHIFT_16M	24
+#define PAGE_SHIFT_64K	16
+#define PAGE_SHIFT_16M	24
+#define PAGE_SHIFT_16G	34
 
 #define NUM_LOW_AREAS	(0x100000000UL >> SID_SHIFT)
 #define NUM_HIGH_AREAS	(PGTABLE_RANGE >> HTLB_AREA_SHIFT)
@@ -95,7 +96,7 @@ static int __hugepte_alloc(struct mm_struct *mm, hugepd_t *hpdp,
 static inline
 pmd_t *hpmd_offset(pud_t *pud, unsigned long addr)
 {
-	if (HPAGE_SHIFT == HPAGE_SHIFT_64K)
+	if (HPAGE_SHIFT == PAGE_SHIFT_64K)
 		return pmd_offset(pud, addr);
 	else
 		return (pmd_t *) pud;
@@ -103,7 +104,7 @@ pmd_t *hpmd_offset(pud_t *pud, unsigned long addr)
 static inline
 pmd_t *hpmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long addr)
 {
-	if (HPAGE_SHIFT == HPAGE_SHIFT_64K)
+	if (HPAGE_SHIFT == PAGE_SHIFT_64K)
 		return pmd_alloc(mm, pud, addr);
 	else
 		return (pmd_t *) pud;
@@ -260,7 +261,7 @@ static void hugetlb_free_pud_range(struct mmu_gather *tlb, pgd_t *pgd,
 			continue;
 		hugetlb_free_pmd_range(tlb, pud, addr, next, floor, ceiling);
 #else
-		if (HPAGE_SHIFT == HPAGE_SHIFT_64K) {
+		if (HPAGE_SHIFT == PAGE_SHIFT_64K) {
 			if (pud_none_or_clear_bad(pud))
 				continue;
 			hugetlb_free_pmd_range(tlb, pud, addr, next, floor, ceiling);
@@ -592,20 +593,40 @@ void set_huge_psize(int psize)
 {
 	/* Check that it is a page size supported by the hardware and
 	 * that it fits within pagetable limits. */
-	if (mmu_psize_defs[psize].shift && mmu_psize_defs[psize].shift < SID_SHIFT &&
+	if (mmu_psize_defs[psize].shift &&
+		mmu_psize_defs[psize].shift < SID_SHIFT_1T &&
 		(mmu_psize_defs[psize].shift > MIN_HUGEPTE_SHIFT ||
-			mmu_psize_defs[psize].shift == HPAGE_SHIFT_64K)) {
+		 mmu_psize_defs[psize].shift == PAGE_SHIFT_64K ||
+		 mmu_psize_defs[psize].shift == PAGE_SHIFT_16G)) {
+		/* Return if huge page size is the same as the
+		 * base page size. */
+		if (mmu_psize_defs[psize].shift == PAGE_SHIFT)
+			return;
+
 		HPAGE_SHIFT = mmu_psize_defs[psize].shift;
 		mmu_huge_psize = psize;
-#ifdef CONFIG_PPC_64K_PAGES
-		hugepte_shift = (PMD_SHIFT-HPAGE_SHIFT);
-#else
-		if (HPAGE_SHIFT == HPAGE_SHIFT_64K)
-			hugepte_shift = (PMD_SHIFT-HPAGE_SHIFT);
-		else
-			hugepte_shift = (PUD_SHIFT-HPAGE_SHIFT);
-#endif
 
+		switch (HPAGE_SHIFT) {
+		case PAGE_SHIFT_64K:
+		    /* We only allow 64k hpages with 4k base page,
+		     * which was checked above, and always put them
+		     * at the PMD */
+		    hugepte_shift = PMD_SHIFT;
+		    break;
+		case PAGE_SHIFT_16M:
+		    /* 16M pages can be at two different levels
+		     * of pagestables based on base page size */
+		    if (PAGE_SHIFT == PAGE_SHIFT_64K)
+			    hugepte_shift = PMD_SHIFT;
+		    else /* 4k base page */
+			    hugepte_shift = PUD_SHIFT;
+		    break;
+		case PAGE_SHIFT_16G:
+		    /* 16G pages are always at PGD level */
+		    hugepte_shift = PGDIR_SHIFT;
+		    break;
+		}
+		hugepte_shift -= HPAGE_SHIFT;
 	} else
 		HPAGE_SHIFT = 0;
 }
@@ -621,17 +642,22 @@ static int __init hugepage_setup_sz(char *str)
 	shift = __ffs(size);
 	switch (shift) {
 #ifndef CONFIG_PPC_64K_PAGES
-	case HPAGE_SHIFT_64K:
+	case PAGE_SHIFT_64K:
 		mmu_psize = MMU_PAGE_64K;
 		break;
 #endif
-	case HPAGE_SHIFT_16M:
+	case PAGE_SHIFT_16M:
 		mmu_psize = MMU_PAGE_16M;
 		break;
+	case PAGE_SHIFT_16G:
+		mmu_psize = MMU_PAGE_16G;
+		break;
 	}
 
-	if (mmu_psize >=0 && mmu_psize_defs[mmu_psize].shift)
+	if (mmu_psize >= 0 && mmu_psize_defs[mmu_psize].shift) {
 		set_huge_psize(mmu_psize);
+		hugetlb_add_hstate(shift - PAGE_SHIFT);
+	}
 	else
 		printk(KERN_WARNING "Invalid huge page size specified(%llu)\n", size);
 
diff --git a/include/asm-powerpc/pgtable-64k.h b/include/asm-powerpc/pgtable-64k.h
index c5007712473f..7e54adb35596 100644
--- a/include/asm-powerpc/pgtable-64k.h
+++ b/include/asm-powerpc/pgtable-64k.h
@@ -138,7 +138,7 @@ static inline struct subpage_prot_table *pgd_subpage_prot(pgd_t *pgd)
                 unsigned __split = (psize == MMU_PAGE_4K ||                 \
 				    psize == MMU_PAGE_64K_AP);              \
                 shift = mmu_psize_defs[psize].shift;                        \
-	        for (index = 0; va < __end; index++, va += (1 << shift)) {  \
+		for (index = 0; va < __end; index++, va += (1L << shift)) { \
 		        if (!__split || __rpte_sub_valid(rpte, index)) do { \
 
 #define pte_iterate_hashed_end() } while(0); } } while(0)

From f4a67cceee4a6f5ed38011a698c9e34747270ae5 Mon Sep 17 00:00:00 2001
From: Jon Tollefson <kniht@linux.vnet.ibm.com>
Date: Wed, 23 Jul 2008 21:27:55 -0700
Subject: [PATCH 0268/1171] fs: check for statfs overflow

Adds a check for an overflow in the filesystem size so if someone is
checking with statfs() on a 16G blocksize hugetlbfs in a 32bit binary that
it will report back EOVERFLOW instead of a size of 0.

Acked-by: Nishanth Aravamudan <nacc@us.ibm.com>
Signed-off-by: Jon Tollefson <kniht@linux.vnet.ibm.com>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/compat.c | 8 ++++----
 fs/open.c   | 3 ++-
 2 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/fs/compat.c b/fs/compat.c
index ed43e17a5dc6..b46604281766 100644
--- a/fs/compat.c
+++ b/fs/compat.c
@@ -197,8 +197,8 @@ static int put_compat_statfs(struct compat_statfs __user *ubuf, struct kstatfs *
 {
 	
 	if (sizeof ubuf->f_blocks == 4) {
-		if ((kbuf->f_blocks | kbuf->f_bfree | kbuf->f_bavail) &
-		    0xffffffff00000000ULL)
+		if ((kbuf->f_blocks | kbuf->f_bfree | kbuf->f_bavail |
+		     kbuf->f_bsize | kbuf->f_frsize) & 0xffffffff00000000ULL)
 			return -EOVERFLOW;
 		/* f_files and f_ffree may be -1; it's okay
 		 * to stuff that into 32 bits */
@@ -271,8 +271,8 @@ out:
 static int put_compat_statfs64(struct compat_statfs64 __user *ubuf, struct kstatfs *kbuf)
 {
 	if (sizeof ubuf->f_blocks == 4) {
-		if ((kbuf->f_blocks | kbuf->f_bfree | kbuf->f_bavail) &
-		    0xffffffff00000000ULL)
+		if ((kbuf->f_blocks | kbuf->f_bfree | kbuf->f_bavail |
+		     kbuf->f_bsize | kbuf->f_frsize) & 0xffffffff00000000ULL)
 			return -EOVERFLOW;
 		/* f_files and f_ffree may be -1; it's okay
 		 * to stuff that into 32 bits */
diff --git a/fs/open.c b/fs/open.c
index a99ad09c3197..bb98d2fe809f 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -64,7 +64,8 @@ static int vfs_statfs_native(struct dentry *dentry, struct statfs *buf)
 		memcpy(buf, &st, sizeof(st));
 	else {
 		if (sizeof buf->f_blocks == 4) {
-			if ((st.f_blocks | st.f_bfree | st.f_bavail) &
+			if ((st.f_blocks | st.f_bfree | st.f_bavail |
+			     st.f_bsize | st.f_frsize) &
 			    0xffffffff00000000ULL)
 				return -EOVERFLOW;
 			/*

From 0d9ea75443dc7e37843e656b8ebc947a6d16d618 Mon Sep 17 00:00:00 2001
From: Jon Tollefson <kniht@linux.vnet.ibm.com>
Date: Wed, 23 Jul 2008 21:27:56 -0700
Subject: [PATCH 0269/1171] powerpc: support multiple hugepage sizes

Instead of using the variable mmu_huge_psize to keep track of the huge
page size we use an array of MMU_PAGE_* values.  For each supported huge
page size we need to know the hugepte_shift value and have a
pgtable_cache.  The hstate or an mmu_huge_psizes index is passed to
functions so that they know which huge page size they should use.

The hugepage sizes 16M and 64K are setup(if available on the hardware) so
that they don't have to be set on the boot cmd line in order to use them.
The number of 16G pages have to be specified at boot-time though (e.g.
hugepagesz=16G hugepages=5).

Signed-off-by: Jon Tollefson <kniht@linux.vnet.ibm.com>
Signed-off-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/kernel-parameters.txt |  10 +-
 arch/powerpc/mm/hash_utils_64.c     |   9 +-
 arch/powerpc/mm/hugetlbpage.c       | 272 ++++++++++++++++++----------
 arch/powerpc/mm/init_64.c           |   8 +-
 arch/powerpc/mm/tlb_64.c            |   2 +-
 include/asm-powerpc/hugetlb.h       |   5 +-
 include/asm-powerpc/mmu-hash64.h    |   4 +-
 include/asm-powerpc/page_64.h       |   1 +
 include/asm-powerpc/pgalloc-64.h    |   4 +-
 9 files changed, 198 insertions(+), 117 deletions(-)

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 30278e9e5211..01a2992b5754 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -776,11 +776,11 @@ and is between 256 and 4096 characters. It is defined in the file
 
 	hugepages=	[HW,X86-32,IA-64] HugeTLB pages to allocate at boot.
 	hugepagesz=	[HW,IA-64,PPC,X86-64] The size of the HugeTLB pages.
-			On x86 this option can be specified multiple times
-			interleaved with hugepages= to reserve huge pages
-			of different sizes. Valid pages sizes on x86-64
-			are 2M (when the CPU supports "pse") and 1G (when the
-			CPU supports the "pdpe1gb" cpuinfo flag)
+			On x86-64 and powerpc, this option can be specified
+			multiple times interleaved with hugepages= to reserve
+			huge pages of different sizes. Valid pages sizes on
+			x86-64 are 2M (when the CPU supports "pse") and 1G
+			(when the CPU supports the "pdpe1gb" cpuinfo flag)
 			Note that 1GB pages can only be allocated at boot time
 			using hugepages= and not freed afterwards.
 	default_hugepagesz=
diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c
index ae4c717243a5..5ce5a4dcd008 100644
--- a/arch/powerpc/mm/hash_utils_64.c
+++ b/arch/powerpc/mm/hash_utils_64.c
@@ -103,7 +103,6 @@ int mmu_kernel_ssize = MMU_SEGSIZE_256M;
 int mmu_highuser_ssize = MMU_SEGSIZE_256M;
 u16 mmu_slb_size = 64;
 #ifdef CONFIG_HUGETLB_PAGE
-int mmu_huge_psize = MMU_PAGE_16M;
 unsigned int HPAGE_SHIFT;
 #endif
 #ifdef CONFIG_PPC_64K_PAGES
@@ -460,15 +459,15 @@ static void __init htab_init_page_sizes(void)
 	/* Reserve 16G huge page memory sections for huge pages */
 	of_scan_flat_dt(htab_dt_scan_hugepage_blocks, NULL);
 
-/* Init large page size. Currently, we pick 16M or 1M depending
+/* Set default large page size. Currently, we pick 16M or 1M depending
 	 * on what is available
 	 */
 	if (mmu_psize_defs[MMU_PAGE_16M].shift)
-		set_huge_psize(MMU_PAGE_16M);
+		HPAGE_SHIFT = mmu_psize_defs[MMU_PAGE_16M].shift;
 	/* With 4k/4level pagetables, we can't (for now) cope with a
 	 * huge page size < PMD_SIZE */
 	else if (mmu_psize_defs[MMU_PAGE_1M].shift)
-		set_huge_psize(MMU_PAGE_1M);
+		HPAGE_SHIFT = mmu_psize_defs[MMU_PAGE_1M].shift;
 #endif /* CONFIG_HUGETLB_PAGE */
 }
 
@@ -889,7 +888,7 @@ int hash_page(unsigned long ea, unsigned long access, unsigned long trap)
 
 #ifdef CONFIG_HUGETLB_PAGE
 	/* Handle hugepage regions */
-	if (HPAGE_SHIFT && psize == mmu_huge_psize) {
+	if (HPAGE_SHIFT && mmu_huge_psizes[psize]) {
 		DBG_LOW(" -> huge page !\n");
 		return hash_huge_page(mm, access, ea, vsid, local, trap);
 	}
diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c
index 19b1a9cec6d5..fb42c4dd3217 100644
--- a/arch/powerpc/mm/hugetlbpage.c
+++ b/arch/powerpc/mm/hugetlbpage.c
@@ -37,15 +37,30 @@
 static unsigned long gpage_freearray[MAX_NUMBER_GPAGES];
 static unsigned nr_gpages;
 
-unsigned int hugepte_shift;
-#define PTRS_PER_HUGEPTE	(1 << hugepte_shift)
-#define HUGEPTE_TABLE_SIZE	(sizeof(pte_t) << hugepte_shift)
+/* Array of valid huge page sizes - non-zero value(hugepte_shift) is
+ * stored for the huge page sizes that are valid.
+ */
+unsigned int mmu_huge_psizes[MMU_PAGE_COUNT] = { }; /* initialize all to 0 */
 
-#define HUGEPD_SHIFT		(HPAGE_SHIFT + hugepte_shift)
-#define HUGEPD_SIZE		(1UL << HUGEPD_SHIFT)
-#define HUGEPD_MASK		(~(HUGEPD_SIZE-1))
+#define hugepte_shift			mmu_huge_psizes
+#define PTRS_PER_HUGEPTE(psize)		(1 << hugepte_shift[psize])
+#define HUGEPTE_TABLE_SIZE(psize)	(sizeof(pte_t) << hugepte_shift[psize])
 
-#define huge_pgtable_cache	(pgtable_cache[HUGEPTE_CACHE_NUM])
+#define HUGEPD_SHIFT(psize)		(mmu_psize_to_shift(psize) \
+						+ hugepte_shift[psize])
+#define HUGEPD_SIZE(psize)		(1UL << HUGEPD_SHIFT(psize))
+#define HUGEPD_MASK(psize)		(~(HUGEPD_SIZE(psize)-1))
+
+/* Subtract one from array size because we don't need a cache for 4K since
+ * is not a huge page size */
+#define huge_pgtable_cache(psize)	(pgtable_cache[HUGEPTE_CACHE_NUM \
+							+ psize-1])
+#define HUGEPTE_CACHE_NAME(psize)	(huge_pgtable_cache_name[psize])
+
+static const char *huge_pgtable_cache_name[MMU_PAGE_COUNT] = {
+	"unused_4K", "hugepte_cache_64K", "unused_64K_AP",
+	"hugepte_cache_1M", "hugepte_cache_16M", "hugepte_cache_16G"
+};
 
 /* Flag to mark huge PD pointers.  This means pmd_bad() and pud_bad()
  * will choke on pointers to hugepte tables, which is handy for
@@ -56,24 +71,49 @@ typedef struct { unsigned long pd; } hugepd_t;
 
 #define hugepd_none(hpd)	((hpd).pd == 0)
 
+static inline int shift_to_mmu_psize(unsigned int shift)
+{
+	switch (shift) {
+#ifndef CONFIG_PPC_64K_PAGES
+	case PAGE_SHIFT_64K:
+	    return MMU_PAGE_64K;
+#endif
+	case PAGE_SHIFT_16M:
+	    return MMU_PAGE_16M;
+	case PAGE_SHIFT_16G:
+	    return MMU_PAGE_16G;
+	}
+	return -1;
+}
+
+static inline unsigned int mmu_psize_to_shift(unsigned int mmu_psize)
+{
+	if (mmu_psize_defs[mmu_psize].shift)
+		return mmu_psize_defs[mmu_psize].shift;
+	BUG();
+}
+
 static inline pte_t *hugepd_page(hugepd_t hpd)
 {
 	BUG_ON(!(hpd.pd & HUGEPD_OK));
 	return (pte_t *)(hpd.pd & ~HUGEPD_OK);
 }
 
-static inline pte_t *hugepte_offset(hugepd_t *hpdp, unsigned long addr)
+static inline pte_t *hugepte_offset(hugepd_t *hpdp, unsigned long addr,
+				    struct hstate *hstate)
 {
-	unsigned long idx = ((addr >> HPAGE_SHIFT) & (PTRS_PER_HUGEPTE-1));
+	unsigned int shift = huge_page_shift(hstate);
+	int psize = shift_to_mmu_psize(shift);
+	unsigned long idx = ((addr >> shift) & (PTRS_PER_HUGEPTE(psize)-1));
 	pte_t *dir = hugepd_page(*hpdp);
 
 	return dir + idx;
 }
 
 static int __hugepte_alloc(struct mm_struct *mm, hugepd_t *hpdp,
-			   unsigned long address)
+			   unsigned long address, unsigned int psize)
 {
-	pte_t *new = kmem_cache_alloc(huge_pgtable_cache,
+	pte_t *new = kmem_cache_alloc(huge_pgtable_cache(psize),
 				      GFP_KERNEL|__GFP_REPEAT);
 
 	if (! new)
@@ -81,7 +121,7 @@ static int __hugepte_alloc(struct mm_struct *mm, hugepd_t *hpdp,
 
 	spin_lock(&mm->page_table_lock);
 	if (!hugepd_none(*hpdp))
-		kmem_cache_free(huge_pgtable_cache, new);
+		kmem_cache_free(huge_pgtable_cache(psize), new);
 	else
 		hpdp->pd = (unsigned long)new | HUGEPD_OK;
 	spin_unlock(&mm->page_table_lock);
@@ -90,21 +130,22 @@ static int __hugepte_alloc(struct mm_struct *mm, hugepd_t *hpdp,
 
 /* Base page size affects how we walk hugetlb page tables */
 #ifdef CONFIG_PPC_64K_PAGES
-#define hpmd_offset(pud, addr)		pmd_offset(pud, addr)
-#define hpmd_alloc(mm, pud, addr)	pmd_alloc(mm, pud, addr)
+#define hpmd_offset(pud, addr, h)	pmd_offset(pud, addr)
+#define hpmd_alloc(mm, pud, addr, h)	pmd_alloc(mm, pud, addr)
 #else
 static inline
-pmd_t *hpmd_offset(pud_t *pud, unsigned long addr)
+pmd_t *hpmd_offset(pud_t *pud, unsigned long addr, struct hstate *hstate)
 {
-	if (HPAGE_SHIFT == PAGE_SHIFT_64K)
+	if (huge_page_shift(hstate) == PAGE_SHIFT_64K)
 		return pmd_offset(pud, addr);
 	else
 		return (pmd_t *) pud;
 }
 static inline
-pmd_t *hpmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long addr)
+pmd_t *hpmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long addr,
+		  struct hstate *hstate)
 {
-	if (HPAGE_SHIFT == PAGE_SHIFT_64K)
+	if (huge_page_shift(hstate) == PAGE_SHIFT_64K)
 		return pmd_alloc(mm, pud, addr);
 	else
 		return (pmd_t *) pud;
@@ -128,8 +169,9 @@ void add_gpage(unsigned long addr, unsigned long page_size,
 }
 
 /* Moves the gigantic page addresses from the temporary list to the
- * huge_boot_pages list.  */
-int alloc_bootmem_huge_page(struct hstate *h)
+ * huge_boot_pages list.
+ */
+int alloc_bootmem_huge_page(struct hstate *hstate)
 {
 	struct huge_bootmem_page *m;
 	if (nr_gpages == 0)
@@ -137,7 +179,7 @@ int alloc_bootmem_huge_page(struct hstate *h)
 	m = phys_to_virt(gpage_freearray[--nr_gpages]);
 	gpage_freearray[nr_gpages] = 0;
 	list_add(&m->list, &huge_boot_pages);
-	m->hstate = h;
+	m->hstate = hstate;
 	return 1;
 }
 
@@ -149,17 +191,25 @@ pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
 	pud_t *pu;
 	pmd_t *pm;
 
-	BUG_ON(get_slice_psize(mm, addr) != mmu_huge_psize);
+	unsigned int psize;
+	unsigned int shift;
+	unsigned long sz;
+	struct hstate *hstate;
+	psize = get_slice_psize(mm, addr);
+	shift = mmu_psize_to_shift(psize);
+	sz = ((1UL) << shift);
+	hstate = size_to_hstate(sz);
 
-	addr &= HPAGE_MASK;
+	addr &= hstate->mask;
 
 	pg = pgd_offset(mm, addr);
 	if (!pgd_none(*pg)) {
 		pu = pud_offset(pg, addr);
 		if (!pud_none(*pu)) {
-			pm = hpmd_offset(pu, addr);
+			pm = hpmd_offset(pu, addr, hstate);
 			if (!pmd_none(*pm))
-				return hugepte_offset((hugepd_t *)pm, addr);
+				return hugepte_offset((hugepd_t *)pm, addr,
+						      hstate);
 		}
 	}
 
@@ -173,16 +223,20 @@ pte_t *huge_pte_alloc(struct mm_struct *mm,
 	pud_t *pu;
 	pmd_t *pm;
 	hugepd_t *hpdp = NULL;
+	struct hstate *hstate;
+	unsigned int psize;
+	hstate = size_to_hstate(sz);
 
-	BUG_ON(get_slice_psize(mm, addr) != mmu_huge_psize);
+	psize = get_slice_psize(mm, addr);
+	BUG_ON(!mmu_huge_psizes[psize]);
 
-	addr &= HPAGE_MASK;
+	addr &= hstate->mask;
 
 	pg = pgd_offset(mm, addr);
 	pu = pud_alloc(mm, pg, addr);
 
 	if (pu) {
-		pm = hpmd_alloc(mm, pu, addr);
+		pm = hpmd_alloc(mm, pu, addr, hstate);
 		if (pm)
 			hpdp = (hugepd_t *)pm;
 	}
@@ -190,10 +244,10 @@ pte_t *huge_pte_alloc(struct mm_struct *mm,
 	if (! hpdp)
 		return NULL;
 
-	if (hugepd_none(*hpdp) && __hugepte_alloc(mm, hpdp, addr))
+	if (hugepd_none(*hpdp) && __hugepte_alloc(mm, hpdp, addr, psize))
 		return NULL;
 
-	return hugepte_offset(hpdp, addr);
+	return hugepte_offset(hpdp, addr, hstate);
 }
 
 int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep)
@@ -201,19 +255,22 @@ int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep)
 	return 0;
 }
 
-static void free_hugepte_range(struct mmu_gather *tlb, hugepd_t *hpdp)
+static void free_hugepte_range(struct mmu_gather *tlb, hugepd_t *hpdp,
+			       unsigned int psize)
 {
 	pte_t *hugepte = hugepd_page(*hpdp);
 
 	hpdp->pd = 0;
 	tlb->need_flush = 1;
-	pgtable_free_tlb(tlb, pgtable_free_cache(hugepte, HUGEPTE_CACHE_NUM,
+	pgtable_free_tlb(tlb, pgtable_free_cache(hugepte,
+						 HUGEPTE_CACHE_NUM+psize-1,
 						 PGF_CACHENUM_MASK));
 }
 
 static void hugetlb_free_pmd_range(struct mmu_gather *tlb, pud_t *pud,
 				   unsigned long addr, unsigned long end,
-				   unsigned long floor, unsigned long ceiling)
+				   unsigned long floor, unsigned long ceiling,
+				   unsigned int psize)
 {
 	pmd_t *pmd;
 	unsigned long next;
@@ -225,7 +282,7 @@ static void hugetlb_free_pmd_range(struct mmu_gather *tlb, pud_t *pud,
 		next = pmd_addr_end(addr, end);
 		if (pmd_none(*pmd))
 			continue;
-		free_hugepte_range(tlb, (hugepd_t *)pmd);
+		free_hugepte_range(tlb, (hugepd_t *)pmd, psize);
 	} while (pmd++, addr = next, addr != end);
 
 	start &= PUD_MASK;
@@ -251,6 +308,9 @@ static void hugetlb_free_pud_range(struct mmu_gather *tlb, pgd_t *pgd,
 	pud_t *pud;
 	unsigned long next;
 	unsigned long start;
+	unsigned int shift;
+	unsigned int psize = get_slice_psize(tlb->mm, addr);
+	shift = mmu_psize_to_shift(psize);
 
 	start = addr;
 	pud = pud_offset(pgd, addr);
@@ -259,16 +319,18 @@ static void hugetlb_free_pud_range(struct mmu_gather *tlb, pgd_t *pgd,
 #ifdef CONFIG_PPC_64K_PAGES
 		if (pud_none_or_clear_bad(pud))
 			continue;
-		hugetlb_free_pmd_range(tlb, pud, addr, next, floor, ceiling);
+		hugetlb_free_pmd_range(tlb, pud, addr, next, floor, ceiling,
+				       psize);
 #else
-		if (HPAGE_SHIFT == PAGE_SHIFT_64K) {
+		if (shift == PAGE_SHIFT_64K) {
 			if (pud_none_or_clear_bad(pud))
 				continue;
-			hugetlb_free_pmd_range(tlb, pud, addr, next, floor, ceiling);
+			hugetlb_free_pmd_range(tlb, pud, addr, next, floor,
+					       ceiling, psize);
 		} else {
 			if (pud_none(*pud))
 				continue;
-			free_hugepte_range(tlb, (hugepd_t *)pud);
+			free_hugepte_range(tlb, (hugepd_t *)pud, psize);
 		}
 #endif
 	} while (pud++, addr = next, addr != end);
@@ -336,27 +398,29 @@ void hugetlb_free_pgd_range(struct mmu_gather *tlb,
 	 * now has no other vmas using it, so can be freed, we don't
 	 * bother to round floor or end up - the tests don't need that.
 	 */
+	unsigned int psize = get_slice_psize(tlb->mm, addr);
 
-	addr &= HUGEPD_MASK;
+	addr &= HUGEPD_MASK(psize);
 	if (addr < floor) {
-		addr += HUGEPD_SIZE;
+		addr += HUGEPD_SIZE(psize);
 		if (!addr)
 			return;
 	}
 	if (ceiling) {
-		ceiling &= HUGEPD_MASK;
+		ceiling &= HUGEPD_MASK(psize);
 		if (!ceiling)
 			return;
 	}
 	if (end - 1 > ceiling - 1)
-		end -= HUGEPD_SIZE;
+		end -= HUGEPD_SIZE(psize);
 	if (addr > end - 1)
 		return;
 
 	start = addr;
 	pgd = pgd_offset(tlb->mm, addr);
 	do {
-		BUG_ON(get_slice_psize(tlb->mm, addr) != mmu_huge_psize);
+		psize = get_slice_psize(tlb->mm, addr);
+		BUG_ON(!mmu_huge_psizes[psize]);
 		next = pgd_addr_end(addr, end);
 		if (pgd_none_or_clear_bad(pgd))
 			continue;
@@ -373,7 +437,11 @@ void set_huge_pte_at(struct mm_struct *mm, unsigned long addr,
 		 * necessary anymore if we make hpte_need_flush() get the
 		 * page size from the slices
 		 */
-		pte_update(mm, addr & HPAGE_MASK, ptep, ~0UL, 1);
+		unsigned int psize = get_slice_psize(mm, addr);
+		unsigned int shift = mmu_psize_to_shift(psize);
+		unsigned long sz = ((1UL) << shift);
+		struct hstate *hstate = size_to_hstate(sz);
+		pte_update(mm, addr & hstate->mask, ptep, ~0UL, 1);
 	}
 	*ptep = __pte(pte_val(pte) & ~_PAGE_HPTEFLAGS);
 }
@@ -390,14 +458,19 @@ follow_huge_addr(struct mm_struct *mm, unsigned long address, int write)
 {
 	pte_t *ptep;
 	struct page *page;
+	unsigned int mmu_psize = get_slice_psize(mm, address);
 
-	if (get_slice_psize(mm, address) != mmu_huge_psize)
+	/* Verify it is a huge page else bail. */
+	if (!mmu_huge_psizes[mmu_psize])
 		return ERR_PTR(-EINVAL);
 
 	ptep = huge_pte_offset(mm, address);
 	page = pte_page(*ptep);
-	if (page)
-		page += (address % HPAGE_SIZE) / PAGE_SIZE;
+	if (page) {
+		unsigned int shift = mmu_psize_to_shift(mmu_psize);
+		unsigned long sz = ((1UL) << shift);
+		page += (address % sz) / PAGE_SIZE;
+	}
 
 	return page;
 }
@@ -425,15 +498,16 @@ unsigned long hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
 					unsigned long len, unsigned long pgoff,
 					unsigned long flags)
 {
-	return slice_get_unmapped_area(addr, len, flags,
-				       mmu_huge_psize, 1, 0);
+	struct hstate *hstate = hstate_file(file);
+	int mmu_psize = shift_to_mmu_psize(huge_page_shift(hstate));
+	return slice_get_unmapped_area(addr, len, flags, mmu_psize, 1, 0);
 }
 
 /*
  * Called by asm hashtable.S for doing lazy icache flush
  */
 static unsigned int hash_huge_page_do_lazy_icache(unsigned long rflags,
-						  pte_t pte, int trap)
+					pte_t pte, int trap, unsigned long sz)
 {
 	struct page *page;
 	int i;
@@ -446,7 +520,7 @@ static unsigned int hash_huge_page_do_lazy_icache(unsigned long rflags,
 	/* page is dirty */
 	if (!test_bit(PG_arch_1, &page->flags) && !PageReserved(page)) {
 		if (trap == 0x400) {
-			for (i = 0; i < (HPAGE_SIZE / PAGE_SIZE); i++)
+			for (i = 0; i < (sz / PAGE_SIZE); i++)
 				__flush_dcache_icache(page_address(page+i));
 			set_bit(PG_arch_1, &page->flags);
 		} else {
@@ -462,11 +536,16 @@ int hash_huge_page(struct mm_struct *mm, unsigned long access,
 {
 	pte_t *ptep;
 	unsigned long old_pte, new_pte;
-	unsigned long va, rflags, pa;
+	unsigned long va, rflags, pa, sz;
 	long slot;
 	int err = 1;
 	int ssize = user_segment_size(ea);
+	unsigned int mmu_psize;
+	int shift;
+	mmu_psize = get_slice_psize(mm, ea);
 
+	if (!mmu_huge_psizes[mmu_psize])
+		goto out;
 	ptep = huge_pte_offset(mm, ea);
 
 	/* Search the Linux page table for a match with va */
@@ -509,30 +588,32 @@ int hash_huge_page(struct mm_struct *mm, unsigned long access,
 	rflags = 0x2 | (!(new_pte & _PAGE_RW));
  	/* _PAGE_EXEC -> HW_NO_EXEC since it's inverted */
 	rflags |= ((new_pte & _PAGE_EXEC) ? 0 : HPTE_R_N);
+	shift = mmu_psize_to_shift(mmu_psize);
+	sz = ((1UL) << shift);
 	if (!cpu_has_feature(CPU_FTR_COHERENT_ICACHE))
 		/* No CPU has hugepages but lacks no execute, so we
 		 * don't need to worry about that case */
 		rflags = hash_huge_page_do_lazy_icache(rflags, __pte(old_pte),
-						       trap);
+						       trap, sz);
 
 	/* Check if pte already has an hpte (case 2) */
 	if (unlikely(old_pte & _PAGE_HASHPTE)) {
 		/* There MIGHT be an HPTE for this pte */
 		unsigned long hash, slot;
 
-		hash = hpt_hash(va, HPAGE_SHIFT, ssize);
+		hash = hpt_hash(va, shift, ssize);
 		if (old_pte & _PAGE_F_SECOND)
 			hash = ~hash;
 		slot = (hash & htab_hash_mask) * HPTES_PER_GROUP;
 		slot += (old_pte & _PAGE_F_GIX) >> 12;
 
-		if (ppc_md.hpte_updatepp(slot, rflags, va, mmu_huge_psize,
+		if (ppc_md.hpte_updatepp(slot, rflags, va, mmu_psize,
 					 ssize, local) == -1)
 			old_pte &= ~_PAGE_HPTEFLAGS;
 	}
 
 	if (likely(!(old_pte & _PAGE_HASHPTE))) {
-		unsigned long hash = hpt_hash(va, HPAGE_SHIFT, ssize);
+		unsigned long hash = hpt_hash(va, shift, ssize);
 		unsigned long hpte_group;
 
 		pa = pte_pfn(__pte(old_pte)) << PAGE_SHIFT;
@@ -553,7 +634,7 @@ repeat:
 
 		/* Insert into the hash table, primary slot */
 		slot = ppc_md.hpte_insert(hpte_group, va, pa, rflags, 0,
-					  mmu_huge_psize, ssize);
+					  mmu_psize, ssize);
 
 		/* Primary is full, try the secondary */
 		if (unlikely(slot == -1)) {
@@ -561,7 +642,7 @@ repeat:
 				      HPTES_PER_GROUP) & ~0x7UL; 
 			slot = ppc_md.hpte_insert(hpte_group, va, pa, rflags,
 						  HPTE_V_SECONDARY,
-						  mmu_huge_psize, ssize);
+						  mmu_psize, ssize);
 			if (slot == -1) {
 				if (mftb() & 0x1)
 					hpte_group = ((hash & htab_hash_mask) *
@@ -598,66 +679,50 @@ void set_huge_psize(int psize)
 		(mmu_psize_defs[psize].shift > MIN_HUGEPTE_SHIFT ||
 		 mmu_psize_defs[psize].shift == PAGE_SHIFT_64K ||
 		 mmu_psize_defs[psize].shift == PAGE_SHIFT_16G)) {
-		/* Return if huge page size is the same as the
-		 * base page size. */
-		if (mmu_psize_defs[psize].shift == PAGE_SHIFT)
+		/* Return if huge page size has already been setup or is the
+		 * same as the base page size. */
+		if (mmu_huge_psizes[psize] ||
+		   mmu_psize_defs[psize].shift == PAGE_SHIFT)
 			return;
+		hugetlb_add_hstate(mmu_psize_defs[psize].shift - PAGE_SHIFT);
 
-		HPAGE_SHIFT = mmu_psize_defs[psize].shift;
-		mmu_huge_psize = psize;
-
-		switch (HPAGE_SHIFT) {
+		switch (mmu_psize_defs[psize].shift) {
 		case PAGE_SHIFT_64K:
 		    /* We only allow 64k hpages with 4k base page,
 		     * which was checked above, and always put them
 		     * at the PMD */
-		    hugepte_shift = PMD_SHIFT;
+		    hugepte_shift[psize] = PMD_SHIFT;
 		    break;
 		case PAGE_SHIFT_16M:
 		    /* 16M pages can be at two different levels
 		     * of pagestables based on base page size */
 		    if (PAGE_SHIFT == PAGE_SHIFT_64K)
-			    hugepte_shift = PMD_SHIFT;
+			    hugepte_shift[psize] = PMD_SHIFT;
 		    else /* 4k base page */
-			    hugepte_shift = PUD_SHIFT;
+			    hugepte_shift[psize] = PUD_SHIFT;
 		    break;
 		case PAGE_SHIFT_16G:
 		    /* 16G pages are always at PGD level */
-		    hugepte_shift = PGDIR_SHIFT;
+		    hugepte_shift[psize] = PGDIR_SHIFT;
 		    break;
 		}
-		hugepte_shift -= HPAGE_SHIFT;
+		hugepte_shift[psize] -= mmu_psize_defs[psize].shift;
 	} else
-		HPAGE_SHIFT = 0;
+		hugepte_shift[psize] = 0;
 }
 
 static int __init hugepage_setup_sz(char *str)
 {
 	unsigned long long size;
-	int mmu_psize = -1;
+	int mmu_psize;
 	int shift;
 
 	size = memparse(str, &str);
 
 	shift = __ffs(size);
-	switch (shift) {
-#ifndef CONFIG_PPC_64K_PAGES
-	case PAGE_SHIFT_64K:
-		mmu_psize = MMU_PAGE_64K;
-		break;
-#endif
-	case PAGE_SHIFT_16M:
-		mmu_psize = MMU_PAGE_16M;
-		break;
-	case PAGE_SHIFT_16G:
-		mmu_psize = MMU_PAGE_16G;
-		break;
-	}
-
-	if (mmu_psize >= 0 && mmu_psize_defs[mmu_psize].shift) {
+	mmu_psize = shift_to_mmu_psize(shift);
+	if (mmu_psize >= 0 && mmu_psize_defs[mmu_psize].shift)
 		set_huge_psize(mmu_psize);
-		hugetlb_add_hstate(shift - PAGE_SHIFT);
-	}
 	else
 		printk(KERN_WARNING "Invalid huge page size specified(%llu)\n", size);
 
@@ -672,16 +737,31 @@ static void zero_ctor(struct kmem_cache *cache, void *addr)
 
 static int __init hugetlbpage_init(void)
 {
+	unsigned int psize;
+
 	if (!cpu_has_feature(CPU_FTR_16M_PAGE))
 		return -ENODEV;
+	/* Add supported huge page sizes.  Need to change HUGE_MAX_HSTATE
+	 * and adjust PTE_NONCACHE_NUM if the number of supported huge page
+	 * sizes changes.
+	 */
+	set_huge_psize(MMU_PAGE_16M);
+	set_huge_psize(MMU_PAGE_64K);
+	set_huge_psize(MMU_PAGE_16G);
 
-	huge_pgtable_cache = kmem_cache_create("hugepte_cache",
-					       HUGEPTE_TABLE_SIZE,
-					       HUGEPTE_TABLE_SIZE,
-					       0,
-					       zero_ctor);
-	if (! huge_pgtable_cache)
-		panic("hugetlbpage_init(): could not create hugepte cache\n");
+	for (psize = 0; psize < MMU_PAGE_COUNT; ++psize) {
+		if (mmu_huge_psizes[psize]) {
+			huge_pgtable_cache(psize) = kmem_cache_create(
+						HUGEPTE_CACHE_NAME(psize),
+						HUGEPTE_TABLE_SIZE(psize),
+						HUGEPTE_TABLE_SIZE(psize),
+						0,
+						zero_ctor);
+			if (!huge_pgtable_cache(psize))
+				panic("hugetlbpage_init(): could not create %s"\
+				      "\n", HUGEPTE_CACHE_NAME(psize));
+		}
+	}
 
 	return 0;
 }
diff --git a/arch/powerpc/mm/init_64.c b/arch/powerpc/mm/init_64.c
index 6ef63caca682..a41bc5aa2043 100644
--- a/arch/powerpc/mm/init_64.c
+++ b/arch/powerpc/mm/init_64.c
@@ -153,10 +153,10 @@ static const char *pgtable_cache_name[ARRAY_SIZE(pgtable_cache_size)] = {
 };
 
 #ifdef CONFIG_HUGETLB_PAGE
-/* Hugepages need one extra cache, initialized in hugetlbpage.c.  We
- * can't put into the tables above, because HPAGE_SHIFT is not compile
- * time constant. */
-struct kmem_cache *pgtable_cache[ARRAY_SIZE(pgtable_cache_size)+1];
+/* Hugepages need an extra cache per hugepagesize, initialized in
+ * hugetlbpage.c.  We can't put into the tables above, because HPAGE_SHIFT
+ * is not compile time constant. */
+struct kmem_cache *pgtable_cache[ARRAY_SIZE(pgtable_cache_size)+MMU_PAGE_COUNT];
 #else
 struct kmem_cache *pgtable_cache[ARRAY_SIZE(pgtable_cache_size)];
 #endif
diff --git a/arch/powerpc/mm/tlb_64.c b/arch/powerpc/mm/tlb_64.c
index a01b5c608ff9..409fcc7b63ce 100644
--- a/arch/powerpc/mm/tlb_64.c
+++ b/arch/powerpc/mm/tlb_64.c
@@ -147,7 +147,7 @@ void hpte_need_flush(struct mm_struct *mm, unsigned long addr,
 	 */
 	if (huge) {
 #ifdef CONFIG_HUGETLB_PAGE
-		psize = mmu_huge_psize;
+		psize = get_slice_psize(mm, addr);;
 #else
 		BUG();
 		psize = pte_pagesize_index(mm, addr, pte); /* shutup gcc */
diff --git a/include/asm-powerpc/hugetlb.h b/include/asm-powerpc/hugetlb.h
index ca37c4af27b1..26f0d0ab27a5 100644
--- a/include/asm-powerpc/hugetlb.h
+++ b/include/asm-powerpc/hugetlb.h
@@ -24,9 +24,10 @@ pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr,
 static inline int prepare_hugepage_range(struct file *file,
 			unsigned long addr, unsigned long len)
 {
-	if (len & ~HPAGE_MASK)
+	struct hstate *h = hstate_file(file);
+	if (len & ~huge_page_mask(h))
 		return -EINVAL;
-	if (addr & ~HPAGE_MASK)
+	if (addr & ~huge_page_mask(h))
 		return -EINVAL;
 	return 0;
 }
diff --git a/include/asm-powerpc/mmu-hash64.h b/include/asm-powerpc/mmu-hash64.h
index b61181aa7746..19c7a9403490 100644
--- a/include/asm-powerpc/mmu-hash64.h
+++ b/include/asm-powerpc/mmu-hash64.h
@@ -194,9 +194,9 @@ extern int mmu_ci_restrictions;
 
 #ifdef CONFIG_HUGETLB_PAGE
 /*
- * The page size index of the huge pages for use by hugetlbfs
+ * The page size indexes of the huge pages for use by hugetlbfs
  */
-extern int mmu_huge_psize;
+extern unsigned int mmu_huge_psizes[MMU_PAGE_COUNT];
 
 #endif /* CONFIG_HUGETLB_PAGE */
 
diff --git a/include/asm-powerpc/page_64.h b/include/asm-powerpc/page_64.h
index 02fd80710e9d..043bfdfe4f73 100644
--- a/include/asm-powerpc/page_64.h
+++ b/include/asm-powerpc/page_64.h
@@ -90,6 +90,7 @@ extern unsigned int HPAGE_SHIFT;
 #define HPAGE_SIZE		((1UL) << HPAGE_SHIFT)
 #define HPAGE_MASK		(~(HPAGE_SIZE - 1))
 #define HUGETLB_PAGE_ORDER	(HPAGE_SHIFT - PAGE_SHIFT)
+#define HUGE_MAX_HSTATE		3
 
 #endif /* __ASSEMBLY__ */
 
diff --git a/include/asm-powerpc/pgalloc-64.h b/include/asm-powerpc/pgalloc-64.h
index 68980990f62a..812a1d8f35cb 100644
--- a/include/asm-powerpc/pgalloc-64.h
+++ b/include/asm-powerpc/pgalloc-64.h
@@ -22,7 +22,7 @@ extern struct kmem_cache *pgtable_cache[];
 #define PUD_CACHE_NUM		1
 #define PMD_CACHE_NUM		1
 #define HUGEPTE_CACHE_NUM	2
-#define PTE_NONCACHE_NUM	3  /* from GFP rather than kmem_cache */
+#define PTE_NONCACHE_NUM	7  /* from GFP rather than kmem_cache */
 
 static inline pgd_t *pgd_alloc(struct mm_struct *mm)
 {
@@ -119,7 +119,7 @@ static inline void pte_free(struct mm_struct *mm, pgtable_t ptepage)
 	__free_page(ptepage);
 }
 
-#define PGF_CACHENUM_MASK	0x3
+#define PGF_CACHENUM_MASK	0x7
 
 typedef struct pgtable_free {
 	unsigned long val;

From 7f09ca51e925ba62e9ebfd4979f093e97e38adeb Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Wed, 23 Jul 2008 21:27:58 -0700
Subject: [PATCH 0270/1171] hugetlb: fix a hugepage reservation check for
 MAP_SHARED

When removing a huge page from the hugepage pool for a fault the system checks
to see if the mapping requires additional pages to be reserved, and if it does
whether there are any unreserved pages remaining.  If not, the allocation
fails without even attempting to get a page.  In order to determine whether to
apply this check we call vma_has_private_reserves() which tells us if this vma
is MAP_PRIVATE and is the owner.  This incorrectly triggers the remaining
reservation test for MAP_SHARED mappings which prevents allocation of the
final page in the pool even though it is reserved for this mapping.

In reality we only want to check this for MAP_PRIVATE mappings where the
process is not the original mapper.  Replace vma_has_private_reserves() with
vma_has_reserves() which indicates whether further reserves are required, and
update the caller.

Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Acked-by: Adam Litke <agl@us.ibm.com>
Acked-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/hugetlb.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 3e1506b808a3..8c20aed62b9c 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -342,13 +342,13 @@ void reset_vma_resv_huge_pages(struct vm_area_struct *vma)
 }
 
 /* Returns true if the VMA has associated reserve pages */
-static int vma_has_private_reserves(struct vm_area_struct *vma)
+static int vma_has_reserves(struct vm_area_struct *vma)
 {
 	if (vma->vm_flags & VM_SHARED)
-		return 0;
-	if (!is_vma_resv_set(vma, HPAGE_RESV_OWNER))
-		return 0;
-	return 1;
+		return 1;
+	if (is_vma_resv_set(vma, HPAGE_RESV_OWNER))
+		return 1;
+	return 0;
 }
 
 static void clear_huge_page(struct page *page,
@@ -420,7 +420,7 @@ static struct page *dequeue_huge_page_vma(struct hstate *h,
 	 * have no page reserves. This check ensures that reservations are
 	 * not "stolen". The child may still get SIGKILLed
 	 */
-	if (!vma_has_private_reserves(vma) &&
+	if (!vma_has_reserves(vma) &&
 			h->free_huge_pages - h->resv_huge_pages == 0)
 		return NULL;
 

From 7251ff78b94c2a68d267623d09b32672b20662c1 Mon Sep 17 00:00:00 2001
From: Adam Litke <agl@us.ibm.com>
Date: Wed, 23 Jul 2008 21:27:59 -0700
Subject: [PATCH 0271/1171] hugetlb: quota is not freed for unused reserved
 private huge pages

With shared reservations (and now also with private reservations), we reserve
huge pages at mmap time.  We also account for the mapping against fs quota to
prevent a reservation from being preempted by quota exhaustion.

When testing with the libhugetlbfs test suite, I found a problem with quota
accounting.  FS quota for allocated pages is handled correctly but we are not
releasing quota for private pages that were reserved but never allocated.  Do
this in hugetlb_vm_op_close() at the same time as unused page reservations are
released.

Signed-off-by: Adam Litke <agl@us.ibm.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Johannes Weiner <hannes@saeurebad.de>
Cc: William Lee Irwin III <wli@holomorphy.com>
Cc: Hugh Dickins <hugh@veritas.com>
Acked-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/hugetlb.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 8c20aed62b9c..41341c414194 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -1552,8 +1552,10 @@ static void hugetlb_vm_op_close(struct vm_area_struct *vma)
 
 		kref_put(&reservations->refs, resv_map_release);
 
-		if (reserve)
+		if (reserve) {
 			hugetlb_acct_memory(h, -reserve);
+			hugetlb_put_quota(vma->vm_file->f_mapping, reserve);
+		}
 	}
 }
 

From 223e8dc9249c9e15f6c8b638d73fcad78ccb0a88 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Wed, 23 Jul 2008 21:28:00 -0700
Subject: [PATCH 0272/1171] bootmem: reorder code to match new bootmem
 structure

This only reorders functions so that further patches will be easier to
read.  No code changed.

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/bootmem.h |  86 +++++-----
 mm/bootmem.c            | 362 ++++++++++++++++++++--------------------
 2 files changed, 225 insertions(+), 223 deletions(-)

diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index f352c5f125b4..5000fd70b04f 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -41,36 +41,62 @@ typedef struct bootmem_data {
 extern bootmem_data_t bootmem_node_data[];
 
 extern unsigned long bootmem_bootmap_pages(unsigned long);
+
+extern unsigned long init_bootmem_node(pg_data_t *pgdat,
+				       unsigned long freepfn,
+				       unsigned long startpfn,
+				       unsigned long endpfn);
 extern unsigned long init_bootmem(unsigned long addr, unsigned long memend);
+
+extern unsigned long free_all_bootmem_node(pg_data_t *pgdat);
+extern unsigned long free_all_bootmem(void);
+
+extern void free_bootmem_node(pg_data_t *pgdat,
+			      unsigned long addr,
+			      unsigned long size);
 extern void free_bootmem(unsigned long addr, unsigned long size);
-extern void *__alloc_bootmem(unsigned long size,
+
+/*
+ * Flags for reserve_bootmem (also if CONFIG_HAVE_ARCH_BOOTMEM_NODE,
+ * the architecture-specific code should honor this).
+ *
+ * If flags is 0, then the return value is always 0 (success). If
+ * flags contains BOOTMEM_EXCLUSIVE, then -EBUSY is returned if the
+ * memory already was reserved.
+ */
+#define BOOTMEM_DEFAULT		0
+#define BOOTMEM_EXCLUSIVE	(1<<0)
+
+extern int reserve_bootmem_node(pg_data_t *pgdat,
+				 unsigned long physaddr,
+				 unsigned long size,
+				 int flags);
+#ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
+extern int reserve_bootmem(unsigned long addr, unsigned long size, int flags);
+#endif
+
+extern void *__alloc_bootmem_nopanic(unsigned long size,
 			     unsigned long align,
 			     unsigned long goal);
-extern void *__alloc_bootmem_nopanic(unsigned long size,
+extern void *__alloc_bootmem(unsigned long size,
 				     unsigned long align,
 				     unsigned long goal);
 extern void *__alloc_bootmem_low(unsigned long size,
 				 unsigned long align,
 				 unsigned long goal);
+extern void *__alloc_bootmem_node(pg_data_t *pgdat,
+				  unsigned long size,
+				  unsigned long align,
+				  unsigned long goal);
+extern void *__alloc_bootmem_node_nopanic(pg_data_t *pgdat,
+				  unsigned long size,
+				  unsigned long align,
+				  unsigned long goal);
 extern void *__alloc_bootmem_low_node(pg_data_t *pgdat,
 				      unsigned long size,
 				      unsigned long align,
 				      unsigned long goal);
-
-/*
- * flags for reserve_bootmem (also if CONFIG_HAVE_ARCH_BOOTMEM_NODE,
- * the architecture-specific code should honor this)
- */
-#define BOOTMEM_DEFAULT		0
-#define BOOTMEM_EXCLUSIVE	(1<<0)
-
 #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
-/*
- * If flags is 0, then the return value is always 0 (success). If
- * flags contains BOOTMEM_EXCLUSIVE, then -EBUSY is returned if the
- * memory already was reserved.
- */
-extern int reserve_bootmem(unsigned long addr, unsigned long size, int flags);
 #define alloc_bootmem(x) \
 	__alloc_bootmem(x, SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
 #define alloc_bootmem_low(x) \
@@ -83,38 +109,16 @@ extern int reserve_bootmem(unsigned long addr, unsigned long size, int flags);
 
 extern int reserve_bootmem_generic(unsigned long addr, unsigned long size,
 				   int flags);
-extern unsigned long free_all_bootmem(void);
-extern unsigned long free_all_bootmem_node(pg_data_t *pgdat);
-extern void *__alloc_bootmem_node(pg_data_t *pgdat,
-				  unsigned long size,
-				  unsigned long align,
-				  unsigned long goal);
-extern void *__alloc_bootmem_node_nopanic(pg_data_t *pgdat,
-				  unsigned long size,
-				  unsigned long align,
-				  unsigned long goal);
-extern unsigned long init_bootmem_node(pg_data_t *pgdat,
-				       unsigned long freepfn,
-				       unsigned long startpfn,
-				       unsigned long endpfn);
-extern int reserve_bootmem_node(pg_data_t *pgdat,
-				 unsigned long physaddr,
-				 unsigned long size,
-				 int flags);
-extern void free_bootmem_node(pg_data_t *pgdat,
-			      unsigned long addr,
-			      unsigned long size);
-extern void *alloc_bootmem_section(unsigned long size,
-				   unsigned long section_nr);
 
-#ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
 #define alloc_bootmem_node(pgdat, x) \
 	__alloc_bootmem_node(pgdat, x, SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
 #define alloc_bootmem_pages_node(pgdat, x) \
 	__alloc_bootmem_node(pgdat, x, PAGE_SIZE, __pa(MAX_DMA_ADDRESS))
 #define alloc_bootmem_low_pages_node(pgdat, x) \
 	__alloc_bootmem_low_node(pgdat, x, PAGE_SIZE, 0)
-#endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
+
+extern void *alloc_bootmem_section(unsigned long size,
+				   unsigned long section_nr);
 
 #ifdef CONFIG_HAVE_ARCH_ALLOC_REMAP
 extern void *alloc_remap(int nid, unsigned long size);
diff --git a/mm/bootmem.c b/mm/bootmem.c
index 9ac972535fff..24eacf52c50e 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -38,6 +38,19 @@ unsigned long saved_max_pfn;
 
 bootmem_data_t bootmem_node_data[MAX_NUMNODES] __initdata;
 
+/*
+ * Given an initialised bdata, it returns the size of the boot bitmap
+ */
+static unsigned long __init get_mapsize(bootmem_data_t *bdata)
+{
+	unsigned long mapsize;
+	unsigned long start = PFN_DOWN(bdata->node_boot_start);
+	unsigned long end = bdata->node_low_pfn;
+
+	mapsize = ((end - start) + 7) / 8;
+	return ALIGN(mapsize, sizeof(long));
+}
+
 /* return the number of _pages_ that will be allocated for the boot bitmap */
 unsigned long __init bootmem_bootmap_pages(unsigned long pages)
 {
@@ -71,19 +84,6 @@ static void __init link_bootmem(bootmem_data_t *bdata)
 	list_add_tail(&bdata->list, &bdata_list);
 }
 
-/*
- * Given an initialised bdata, it returns the size of the boot bitmap
- */
-static unsigned long __init get_mapsize(bootmem_data_t *bdata)
-{
-	unsigned long mapsize;
-	unsigned long start = PFN_DOWN(bdata->node_boot_start);
-	unsigned long end = bdata->node_low_pfn;
-
-	mapsize = ((end - start) + 7) / 8;
-	return ALIGN(mapsize, sizeof(long));
-}
-
 /*
  * Called once to set up the allocator itself.
  */
@@ -108,6 +108,146 @@ static unsigned long __init init_bootmem_core(bootmem_data_t *bdata,
 	return mapsize;
 }
 
+unsigned long __init init_bootmem_node(pg_data_t *pgdat, unsigned long freepfn,
+				unsigned long startpfn, unsigned long endpfn)
+{
+	return init_bootmem_core(pgdat->bdata, freepfn, startpfn, endpfn);
+}
+
+unsigned long __init init_bootmem(unsigned long start, unsigned long pages)
+{
+	max_low_pfn = pages;
+	min_low_pfn = start;
+	return init_bootmem_core(NODE_DATA(0)->bdata, start, 0, pages);
+}
+
+static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata)
+{
+	struct page *page;
+	unsigned long pfn;
+	unsigned long i, count;
+	unsigned long idx;
+	unsigned long *map;
+	int gofast = 0;
+
+	BUG_ON(!bdata->node_bootmem_map);
+
+	count = 0;
+	/* first extant page of the node */
+	pfn = PFN_DOWN(bdata->node_boot_start);
+	idx = bdata->node_low_pfn - pfn;
+	map = bdata->node_bootmem_map;
+	/*
+	 * Check if we are aligned to BITS_PER_LONG pages.  If so, we might
+	 * be able to free page orders of that size at once.
+	 */
+	if (!(pfn & (BITS_PER_LONG-1)))
+		gofast = 1;
+
+	for (i = 0; i < idx; ) {
+		unsigned long v = ~map[i / BITS_PER_LONG];
+
+		if (gofast && v == ~0UL) {
+			int order;
+
+			page = pfn_to_page(pfn);
+			count += BITS_PER_LONG;
+			order = ffs(BITS_PER_LONG) - 1;
+			__free_pages_bootmem(page, order);
+			i += BITS_PER_LONG;
+			page += BITS_PER_LONG;
+		} else if (v) {
+			unsigned long m;
+
+			page = pfn_to_page(pfn);
+			for (m = 1; m && i < idx; m<<=1, page++, i++) {
+				if (v & m) {
+					count++;
+					__free_pages_bootmem(page, 0);
+				}
+			}
+		} else {
+			i += BITS_PER_LONG;
+		}
+		pfn += BITS_PER_LONG;
+	}
+
+	/*
+	 * Now free the allocator bitmap itself, it's not
+	 * needed anymore:
+	 */
+	page = virt_to_page(bdata->node_bootmem_map);
+	idx = (get_mapsize(bdata) + PAGE_SIZE-1) >> PAGE_SHIFT;
+	for (i = 0; i < idx; i++, page++)
+		__free_pages_bootmem(page, 0);
+	count += i;
+	bdata->node_bootmem_map = NULL;
+
+	return count;
+}
+
+unsigned long __init free_all_bootmem_node(pg_data_t *pgdat)
+{
+	register_page_bootmem_info_node(pgdat);
+	return free_all_bootmem_core(pgdat->bdata);
+}
+
+unsigned long __init free_all_bootmem(void)
+{
+	return free_all_bootmem_core(NODE_DATA(0)->bdata);
+}
+
+static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr,
+				     unsigned long size)
+{
+	unsigned long sidx, eidx;
+	unsigned long i;
+
+	BUG_ON(!size);
+
+	/* out range */
+	if (addr + size < bdata->node_boot_start ||
+		PFN_DOWN(addr) > bdata->node_low_pfn)
+		return;
+	/*
+	 * round down end of usable mem, partially free pages are
+	 * considered reserved.
+	 */
+
+	if (addr >= bdata->node_boot_start && addr < bdata->last_success)
+		bdata->last_success = addr;
+
+	/*
+	 * Round up to index to the range.
+	 */
+	if (PFN_UP(addr) > PFN_DOWN(bdata->node_boot_start))
+		sidx = PFN_UP(addr) - PFN_DOWN(bdata->node_boot_start);
+	else
+		sidx = 0;
+
+	eidx = PFN_DOWN(addr + size - bdata->node_boot_start);
+	if (eidx > bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start))
+		eidx = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start);
+
+	for (i = sidx; i < eidx; i++) {
+		if (unlikely(!test_and_clear_bit(i, bdata->node_bootmem_map)))
+			BUG();
+	}
+}
+
+void __init free_bootmem_node(pg_data_t *pgdat, unsigned long physaddr,
+			      unsigned long size)
+{
+	free_bootmem_core(pgdat->bdata, physaddr, size);
+}
+
+void __init free_bootmem(unsigned long addr, unsigned long size)
+{
+	bootmem_data_t *bdata;
+	list_for_each_entry(bdata, &bdata_list, list)
+		free_bootmem_core(bdata, addr, size);
+}
+
 /*
  * Marks a particular physical memory range as unallocatable. Usable RAM
  * might be used for boot-time allocations - or it might get added
@@ -183,44 +323,37 @@ static void __init reserve_bootmem_core(bootmem_data_t *bdata,
 	}
 }
 
-static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr,
-				     unsigned long size)
+int __init reserve_bootmem_node(pg_data_t *pgdat, unsigned long physaddr,
+				 unsigned long size, int flags)
 {
-	unsigned long sidx, eidx;
-	unsigned long i;
+	int ret;
 
-	BUG_ON(!size);
-
-	/* out range */
-	if (addr + size < bdata->node_boot_start ||
-		PFN_DOWN(addr) > bdata->node_low_pfn)
-		return;
-	/*
-	 * round down end of usable mem, partially free pages are
-	 * considered reserved.
-	 */
-
-	if (addr >= bdata->node_boot_start && addr < bdata->last_success)
-		bdata->last_success = addr;
-
-	/*
-	 * Round up to index to the range.
-	 */
-	if (PFN_UP(addr) > PFN_DOWN(bdata->node_boot_start))
-		sidx = PFN_UP(addr) - PFN_DOWN(bdata->node_boot_start);
-	else
-		sidx = 0;
-
-	eidx = PFN_DOWN(addr + size - bdata->node_boot_start);
-	if (eidx > bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start))
-		eidx = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start);
-
-	for (i = sidx; i < eidx; i++) {
-		if (unlikely(!test_and_clear_bit(i, bdata->node_bootmem_map)))
-			BUG();
-	}
+	ret = can_reserve_bootmem_core(pgdat->bdata, physaddr, size, flags);
+	if (ret < 0)
+		return -ENOMEM;
+	reserve_bootmem_core(pgdat->bdata, physaddr, size, flags);
+	return 0;
 }
 
+#ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
+int __init reserve_bootmem(unsigned long addr, unsigned long size,
+			    int flags)
+{
+	bootmem_data_t *bdata;
+	int ret;
+
+	list_for_each_entry(bdata, &bdata_list, list) {
+		ret = can_reserve_bootmem_core(bdata, addr, size, flags);
+		if (ret < 0)
+			return ret;
+	}
+	list_for_each_entry(bdata, &bdata_list, list)
+		reserve_bootmem_core(bdata, addr, size, flags);
+
+	return 0;
+}
+#endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
+
 /*
  * We 'merge' subsequent allocations to save space. We might 'lose'
  * some fraction of a page if allocations cannot be satisfied due to
@@ -371,140 +504,6 @@ found:
 	return ret;
 }
 
-static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata)
-{
-	struct page *page;
-	unsigned long pfn;
-	unsigned long i, count;
-	unsigned long idx;
-	unsigned long *map; 
-	int gofast = 0;
-
-	BUG_ON(!bdata->node_bootmem_map);
-
-	count = 0;
-	/* first extant page of the node */
-	pfn = PFN_DOWN(bdata->node_boot_start);
-	idx = bdata->node_low_pfn - pfn;
-	map = bdata->node_bootmem_map;
-	/*
-	 * Check if we are aligned to BITS_PER_LONG pages.  If so, we might
-	 * be able to free page orders of that size at once.
-	 */
-	if (!(pfn & (BITS_PER_LONG-1)))
-		gofast = 1;
-
-	for (i = 0; i < idx; ) {
-		unsigned long v = ~map[i / BITS_PER_LONG];
-
-		if (gofast && v == ~0UL) {
-			int order;
-
-			page = pfn_to_page(pfn);
-			count += BITS_PER_LONG;
-			order = ffs(BITS_PER_LONG) - 1;
-			__free_pages_bootmem(page, order);
-			i += BITS_PER_LONG;
-			page += BITS_PER_LONG;
-		} else if (v) {
-			unsigned long m;
-
-			page = pfn_to_page(pfn);
-			for (m = 1; m && i < idx; m<<=1, page++, i++) {
-				if (v & m) {
-					count++;
-					__free_pages_bootmem(page, 0);
-				}
-			}
-		} else {
-			i += BITS_PER_LONG;
-		}
-		pfn += BITS_PER_LONG;
-	}
-
-	/*
-	 * Now free the allocator bitmap itself, it's not
-	 * needed anymore:
-	 */
-	page = virt_to_page(bdata->node_bootmem_map);
-	idx = (get_mapsize(bdata) + PAGE_SIZE-1) >> PAGE_SHIFT;
-	for (i = 0; i < idx; i++, page++)
-		__free_pages_bootmem(page, 0);
-	count += i;
-	bdata->node_bootmem_map = NULL;
-
-	return count;
-}
-
-unsigned long __init init_bootmem_node(pg_data_t *pgdat, unsigned long freepfn,
-				unsigned long startpfn, unsigned long endpfn)
-{
-	return init_bootmem_core(pgdat->bdata, freepfn, startpfn, endpfn);
-}
-
-int __init reserve_bootmem_node(pg_data_t *pgdat, unsigned long physaddr,
-				 unsigned long size, int flags)
-{
-	int ret;
-
-	ret = can_reserve_bootmem_core(pgdat->bdata, physaddr, size, flags);
-	if (ret < 0)
-		return -ENOMEM;
-	reserve_bootmem_core(pgdat->bdata, physaddr, size, flags);
-
-	return 0;
-}
-
-void __init free_bootmem_node(pg_data_t *pgdat, unsigned long physaddr,
-			      unsigned long size)
-{
-	free_bootmem_core(pgdat->bdata, physaddr, size);
-}
-
-unsigned long __init free_all_bootmem_node(pg_data_t *pgdat)
-{
-	register_page_bootmem_info_node(pgdat);
-	return free_all_bootmem_core(pgdat->bdata);
-}
-
-unsigned long __init init_bootmem(unsigned long start, unsigned long pages)
-{
-	max_low_pfn = pages;
-	min_low_pfn = start;
-	return init_bootmem_core(NODE_DATA(0)->bdata, start, 0, pages);
-}
-
-#ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
-int __init reserve_bootmem(unsigned long addr, unsigned long size,
-			    int flags)
-{
-	bootmem_data_t *bdata;
-	int ret;
-
-	list_for_each_entry(bdata, &bdata_list, list) {
-		ret = can_reserve_bootmem_core(bdata, addr, size, flags);
-		if (ret < 0)
-			return ret;
-	}
-	list_for_each_entry(bdata, &bdata_list, list)
-		reserve_bootmem_core(bdata, addr, size, flags);
-
-	return 0;
-}
-#endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
-
-void __init free_bootmem(unsigned long addr, unsigned long size)
-{
-	bootmem_data_t *bdata;
-	list_for_each_entry(bdata, &bdata_list, list)
-		free_bootmem_core(bdata, addr, size);
-}
-
-unsigned long __init free_all_bootmem(void)
-{
-	return free_all_bootmem_core(NODE_DATA(0)->bdata);
-}
-
 void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align,
 				      unsigned long goal)
 {
@@ -534,7 +533,6 @@ void * __init __alloc_bootmem(unsigned long size, unsigned long align,
 	return NULL;
 }
 
-
 void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size,
 				   unsigned long align, unsigned long goal)
 {

From 57cfc29efac6670355ee0e107c8dbae8237d406b Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Wed, 23 Jul 2008 21:28:00 -0700
Subject: [PATCH 0273/1171] bootmem: clean up bootmem.c file header

Change the description, move a misplaced comment about the allocator
itself and add me to the list of copyright holders.

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/bootmem.c | 14 +++++---------
 1 file changed, 5 insertions(+), 9 deletions(-)

diff --git a/mm/bootmem.c b/mm/bootmem.c
index 24eacf52c50e..286e12c536ae 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -1,12 +1,12 @@
 /*
- *  linux/mm/bootmem.c
+ *  bootmem - A boot-time physical memory allocator and configurator
  *
  *  Copyright (C) 1999 Ingo Molnar
- *  Discontiguous memory support, Kanoj Sarcar, SGI, Nov 1999
+ *                1999 Kanoj Sarcar, SGI
+ *                2008 Johannes Weiner
  *
- *  simple boot-time physical memory area allocator and
- *  free memory collector. It's used to deal with reserved
- *  system memory and memory holes as well.
+ * Access to this subsystem has to be serialized externally (which is true
+ * for the boot process anyway).
  */
 #include <linux/init.h>
 #include <linux/pfn.h>
@@ -19,10 +19,6 @@
 
 #include "internal.h"
 
-/*
- * Access to this subsystem has to be serialized externally. (this is
- * true for the boot process anyway)
- */
 unsigned long max_low_pfn;
 unsigned long min_low_pfn;
 unsigned long max_pfn;

From a66fd7daec1f40c1f0eac466f0da9206b615fe2a Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Wed, 23 Jul 2008 21:28:01 -0700
Subject: [PATCH 0274/1171] bootmem: add documentation to API functions

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/bootmem.c | 150 ++++++++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 149 insertions(+), 1 deletion(-)

diff --git a/mm/bootmem.c b/mm/bootmem.c
index 286e12c536ae..105ad4cff2e1 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -47,7 +47,10 @@ static unsigned long __init get_mapsize(bootmem_data_t *bdata)
 	return ALIGN(mapsize, sizeof(long));
 }
 
-/* return the number of _pages_ that will be allocated for the boot bitmap */
+/**
+ * bootmem_bootmap_pages - calculate bitmap size in pages
+ * @pages: number of pages the bitmap has to represent
+ */
 unsigned long __init bootmem_bootmap_pages(unsigned long pages)
 {
 	unsigned long mapsize;
@@ -104,12 +107,28 @@ static unsigned long __init init_bootmem_core(bootmem_data_t *bdata,
 	return mapsize;
 }
 
+/**
+ * init_bootmem_node - register a node as boot memory
+ * @pgdat: node to register
+ * @freepfn: pfn where the bitmap for this node is to be placed
+ * @startpfn: first pfn on the node
+ * @endpfn: first pfn after the node
+ *
+ * Returns the number of bytes needed to hold the bitmap for this node.
+ */
 unsigned long __init init_bootmem_node(pg_data_t *pgdat, unsigned long freepfn,
 				unsigned long startpfn, unsigned long endpfn)
 {
 	return init_bootmem_core(pgdat->bdata, freepfn, startpfn, endpfn);
 }
 
+/**
+ * init_bootmem - register boot memory
+ * @start: pfn where the bitmap is to be placed
+ * @pages: number of available physical pages
+ *
+ * Returns the number of bytes needed to hold the bitmap.
+ */
 unsigned long __init init_bootmem(unsigned long start, unsigned long pages)
 {
 	max_low_pfn = pages;
@@ -182,12 +201,23 @@ static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata)
 	return count;
 }
 
+/**
+ * free_all_bootmem_node - release a node's free pages to the buddy allocator
+ * @pgdat: node to be released
+ *
+ * Returns the number of pages actually released.
+ */
 unsigned long __init free_all_bootmem_node(pg_data_t *pgdat)
 {
 	register_page_bootmem_info_node(pgdat);
 	return free_all_bootmem_core(pgdat->bdata);
 }
 
+/**
+ * free_all_bootmem - release free pages to the buddy allocator
+ *
+ * Returns the number of pages actually released.
+ */
 unsigned long __init free_all_bootmem(void)
 {
 	return free_all_bootmem_core(NODE_DATA(0)->bdata);
@@ -231,12 +261,32 @@ static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr,
 	}
 }
 
+/**
+ * free_bootmem_node - mark a page range as usable
+ * @pgdat: node the range resides on
+ * @physaddr: starting address of the range
+ * @size: size of the range in bytes
+ *
+ * Partial pages will be considered reserved and left as they are.
+ *
+ * Only physical pages that actually reside on @pgdat are marked.
+ */
 void __init free_bootmem_node(pg_data_t *pgdat, unsigned long physaddr,
 			      unsigned long size)
 {
 	free_bootmem_core(pgdat->bdata, physaddr, size);
 }
 
+/**
+ * free_bootmem - mark a page range as usable
+ * @addr: starting address of the range
+ * @size: size of the range in bytes
+ *
+ * Partial pages will be considered reserved and left as they are.
+ *
+ * All physical pages within the range are marked, no matter what
+ * node they reside on.
+ */
 void __init free_bootmem(unsigned long addr, unsigned long size)
 {
 	bootmem_data_t *bdata;
@@ -319,6 +369,17 @@ static void __init reserve_bootmem_core(bootmem_data_t *bdata,
 	}
 }
 
+/**
+ * reserve_bootmem_node - mark a page range as reserved
+ * @pgdat: node the range resides on
+ * @physaddr: starting address of the range
+ * @size: size of the range in bytes
+ * @flags: reservation flags (see linux/bootmem.h)
+ *
+ * Partial pages will be reserved.
+ *
+ * Only physical pages that actually reside on @pgdat are marked.
+ */
 int __init reserve_bootmem_node(pg_data_t *pgdat, unsigned long physaddr,
 				 unsigned long size, int flags)
 {
@@ -332,6 +393,17 @@ int __init reserve_bootmem_node(pg_data_t *pgdat, unsigned long physaddr,
 }
 
 #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
+/**
+ * reserve_bootmem - mark a page range as usable
+ * @addr: starting address of the range
+ * @size: size of the range in bytes
+ * @flags: reservation flags (see linux/bootmem.h)
+ *
+ * Partial pages will be reserved.
+ *
+ * All physical pages within the range are marked, no matter what
+ * node they reside on.
+ */
 int __init reserve_bootmem(unsigned long addr, unsigned long size,
 			    int flags)
 {
@@ -500,6 +572,19 @@ found:
 	return ret;
 }
 
+/**
+ * __alloc_bootmem_nopanic - allocate boot memory without panicking
+ * @size: size of the request in bytes
+ * @align: alignment of the region
+ * @goal: preferred starting address of the region
+ *
+ * The goal is dropped if it can not be satisfied and the allocation will
+ * fall back to memory below @goal.
+ *
+ * Allocation may happen on any node in the system.
+ *
+ * Returns NULL on failure.
+ */
 void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align,
 				      unsigned long goal)
 {
@@ -514,6 +599,19 @@ void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align,
 	return NULL;
 }
 
+/**
+ * __alloc_bootmem - allocate boot memory
+ * @size: size of the request in bytes
+ * @align: alignment of the region
+ * @goal: preferred starting address of the region
+ *
+ * The goal is dropped if it can not be satisfied and the allocation will
+ * fall back to memory below @goal.
+ *
+ * Allocation may happen on any node in the system.
+ *
+ * The function panics if the request can not be satisfied.
+ */
 void * __init __alloc_bootmem(unsigned long size, unsigned long align,
 			      unsigned long goal)
 {
@@ -529,6 +627,21 @@ void * __init __alloc_bootmem(unsigned long size, unsigned long align,
 	return NULL;
 }
 
+/**
+ * __alloc_bootmem_node - allocate boot memory from a specific node
+ * @pgdat: node to allocate from
+ * @size: size of the request in bytes
+ * @align: alignment of the region
+ * @goal: preferred starting address of the region
+ *
+ * The goal is dropped if it can not be satisfied and the allocation will
+ * fall back to memory below @goal.
+ *
+ * Allocation may fall back to any node in the system if the specified node
+ * can not hold the requested memory.
+ *
+ * The function panics if the request can not be satisfied.
+ */
 void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size,
 				   unsigned long align, unsigned long goal)
 {
@@ -542,6 +655,13 @@ void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size,
 }
 
 #ifdef CONFIG_SPARSEMEM
+/**
+ * alloc_bootmem_section - allocate boot memory from a specific section
+ * @size: size of the request in bytes
+ * @section_nr: sparse map section to allocate from
+ *
+ * Return NULL on failure.
+ */
 void * __init alloc_bootmem_section(unsigned long size,
 				    unsigned long section_nr)
 {
@@ -588,6 +708,19 @@ void * __init __alloc_bootmem_node_nopanic(pg_data_t *pgdat, unsigned long size,
 #define ARCH_LOW_ADDRESS_LIMIT	0xffffffffUL
 #endif
 
+/**
+ * __alloc_bootmem_low - allocate low boot memory
+ * @size: size of the request in bytes
+ * @align: alignment of the region
+ * @goal: preferred starting address of the region
+ *
+ * The goal is dropped if it can not be satisfied and the allocation will
+ * fall back to memory below @goal.
+ *
+ * Allocation may happen on any node in the system.
+ *
+ * The function panics if the request can not be satisfied.
+ */
 void * __init __alloc_bootmem_low(unsigned long size, unsigned long align,
 				  unsigned long goal)
 {
@@ -609,6 +742,21 @@ void * __init __alloc_bootmem_low(unsigned long size, unsigned long align,
 	return NULL;
 }
 
+/**
+ * __alloc_bootmem_low_node - allocate low boot memory from a specific node
+ * @pgdat: node to allocate from
+ * @size: size of the request in bytes
+ * @align: alignment of the region
+ * @goal: preferred starting address of the region
+ *
+ * The goal is dropped if it can not be satisfied and the allocation will
+ * fall back to memory below @goal.
+ *
+ * Allocation may fall back to any node in the system if the specified node
+ * can not hold the requested memory.
+ *
+ * The function panics if the request can not be satisfied.
+ */
 void * __init __alloc_bootmem_low_node(pg_data_t *pgdat, unsigned long size,
 				       unsigned long align, unsigned long goal)
 {

From 2e5237daf0cc3c8d87762f53f704dc54fa91dcf6 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Wed, 23 Jul 2008 21:28:02 -0700
Subject: [PATCH 0275/1171] bootmem: add debugging framework

Introduce the bootmem_debug kernel parameter that enables very verbose
diagnostics regarding all range operations of bootmem as well as the
initialization and release of nodes.

[akpm@linux-foundation.org: fix printk warnings]
Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/bootmem.c | 51 ++++++++++++++++++++++++++++++++++++++++++++-------
 1 file changed, 44 insertions(+), 7 deletions(-)

diff --git a/mm/bootmem.c b/mm/bootmem.c
index 105ad4cff2e1..4e085ee1d98e 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -34,6 +34,22 @@ unsigned long saved_max_pfn;
 
 bootmem_data_t bootmem_node_data[MAX_NUMNODES] __initdata;
 
+static int bootmem_debug;
+
+static int __init bootmem_debug_setup(char *buf)
+{
+	bootmem_debug = 1;
+	return 0;
+}
+early_param("bootmem_debug", bootmem_debug_setup);
+
+#define bdebug(fmt, args...) ({				\
+	if (unlikely(bootmem_debug))			\
+		printk(KERN_INFO			\
+			"bootmem::%s " fmt,		\
+			__FUNCTION__, ## args);		\
+})
+
 /*
  * Given an initialised bdata, it returns the size of the boot bitmap
  */
@@ -104,6 +120,9 @@ static unsigned long __init init_bootmem_core(bootmem_data_t *bdata,
 	mapsize = get_mapsize(bdata);
 	memset(bdata->node_bootmem_map, 0xff, mapsize);
 
+	bdebug("nid=%td start=%lx map=%lx end=%lx mapsize=%lx\n",
+		bdata - bootmem_node_data, start, mapstart, end, mapsize);
+
 	return mapsize;
 }
 
@@ -198,6 +217,8 @@ static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata)
 	count += i;
 	bdata->node_bootmem_map = NULL;
 
+	bdebug("nid=%td released=%lx\n", bdata - bootmem_node_data, count);
+
 	return count;
 }
 
@@ -255,6 +276,10 @@ static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr,
 	if (eidx > bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start))
 		eidx = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start);
 
+	bdebug("nid=%td start=%lx end=%lx\n", bdata - bootmem_node_data,
+		sidx + PFN_DOWN(bdata->node_boot_start),
+		eidx + PFN_DOWN(bdata->node_boot_start));
+
 	for (i = sidx; i < eidx; i++) {
 		if (unlikely(!test_and_clear_bit(i, bdata->node_bootmem_map)))
 			BUG();
@@ -360,13 +385,16 @@ static void __init reserve_bootmem_core(bootmem_data_t *bdata,
 	if (eidx > bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start))
 		eidx = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start);
 
-	for (i = sidx; i < eidx; i++) {
-		if (test_and_set_bit(i, bdata->node_bootmem_map)) {
-#ifdef CONFIG_DEBUG_BOOTMEM
-			printk("hm, page %08lx reserved twice.\n", i*PAGE_SIZE);
-#endif
-		}
-	}
+	bdebug("nid=%td start=%lx end=%lx flags=%x\n",
+		bdata - bootmem_node_data,
+		sidx + PFN_DOWN(bdata->node_boot_start),
+		eidx + PFN_DOWN(bdata->node_boot_start),
+		flags);
+
+	for (i = sidx; i < eidx; i++)
+		if (test_and_set_bit(i, bdata->node_bootmem_map))
+			bdebug("hm, page %lx reserved twice.\n",
+				PFN_DOWN(bdata->node_boot_start) + i);
 }
 
 /**
@@ -455,6 +483,10 @@ alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size,
 	if (!bdata->node_bootmem_map)
 		return NULL;
 
+	bdebug("nid=%td size=%lx [%lu pages] align=%lx goal=%lx limit=%lx\n",
+		bdata - bootmem_node_data, size, PAGE_ALIGN(size) >> PAGE_SHIFT,
+		align, goal, limit);
+
 	/* bdata->node_boot_start is supposed to be (12+6)bits alignment on x86_64 ? */
 	node_boot_start = bdata->node_boot_start;
 	node_bootmem_map = bdata->node_bootmem_map;
@@ -562,6 +594,11 @@ found:
 		ret = phys_to_virt(start * PAGE_SIZE + node_boot_start);
 	}
 
+	bdebug("nid=%td start=%lx end=%lx\n",
+		bdata - bootmem_node_data,
+		start + PFN_DOWN(bdata->node_boot_start),
+		start + areasize + PFN_DOWN(bdata->node_boot_start));
+
 	/*
 	 * Reserve the area now:
 	 */

From df049a5f41a3b2eee2131221959e3b558ba7c705 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Wed, 23 Jul 2008 21:28:02 -0700
Subject: [PATCH 0276/1171] bootmem: revisit bitmap size calculations

Reincarnate get_mapsize as bootmap_bytes and implement
bootmem_bootmap_pages on top of it.

Adjust users of these helpers and make free_all_bootmem_core use
bootmem_bootmap_pages instead of open-coding it.

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/bootmem.c | 27 +++++++++------------------
 1 file changed, 9 insertions(+), 18 deletions(-)

diff --git a/mm/bootmem.c b/mm/bootmem.c
index 4e085ee1d98e..484849bfc8c4 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -50,17 +50,11 @@ early_param("bootmem_debug", bootmem_debug_setup);
 			__FUNCTION__, ## args);		\
 })
 
-/*
- * Given an initialised bdata, it returns the size of the boot bitmap
- */
-static unsigned long __init get_mapsize(bootmem_data_t *bdata)
+static unsigned long __init bootmap_bytes(unsigned long pages)
 {
-	unsigned long mapsize;
-	unsigned long start = PFN_DOWN(bdata->node_boot_start);
-	unsigned long end = bdata->node_low_pfn;
+	unsigned long bytes = (pages + 7) / 8;
 
-	mapsize = ((end - start) + 7) / 8;
-	return ALIGN(mapsize, sizeof(long));
+	return ALIGN(bytes, sizeof(long));
 }
 
 /**
@@ -69,13 +63,9 @@ static unsigned long __init get_mapsize(bootmem_data_t *bdata)
  */
 unsigned long __init bootmem_bootmap_pages(unsigned long pages)
 {
-	unsigned long mapsize;
+	unsigned long bytes = bootmap_bytes(pages);
 
-	mapsize = (pages+7)/8;
-	mapsize = (mapsize + ~PAGE_MASK) & PAGE_MASK;
-	mapsize >>= PAGE_SHIFT;
-
-	return mapsize;
+	return PAGE_ALIGN(bytes) >> PAGE_SHIFT;
 }
 
 /*
@@ -117,7 +107,7 @@ static unsigned long __init init_bootmem_core(bootmem_data_t *bdata,
 	 * Initially all pages are reserved - setup_arch() has to
 	 * register free RAM areas explicitly.
 	 */
-	mapsize = get_mapsize(bdata);
+	mapsize = bootmap_bytes(end - start);
 	memset(bdata->node_bootmem_map, 0xff, mapsize);
 
 	bdebug("nid=%td start=%lx map=%lx end=%lx mapsize=%lx\n",
@@ -160,7 +150,7 @@ static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata)
 	struct page *page;
 	unsigned long pfn;
 	unsigned long i, count;
-	unsigned long idx;
+	unsigned long idx, pages;
 	unsigned long *map;
 	int gofast = 0;
 
@@ -211,7 +201,8 @@ static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata)
 	 * needed anymore:
 	 */
 	page = virt_to_page(bdata->node_bootmem_map);
-	idx = (get_mapsize(bdata) + PAGE_SIZE-1) >> PAGE_SHIFT;
+	pages = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start);
+	idx = bootmem_bootmap_pages(pages);
 	for (i = 0; i < idx; i++, page++)
 		__free_pages_bootmem(page, 0);
 	count += i;

From 636cc40cb79f511d9caa27ef098a83e4fa4971fb Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Wed, 23 Jul 2008 21:28:03 -0700
Subject: [PATCH 0277/1171] bootmem: revisit bootmem descriptor list handling

link_bootmem handles an insertion of a new descriptor into the sorted list
in more or less three explicit branches; empty list, insert in between and
append.  These cases can be expressed implicite.

Also mark the sorted list as initdata as it can be thrown away after boot
as well.

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/bootmem.c | 23 ++++++++++-------------
 1 file changed, 10 insertions(+), 13 deletions(-)

diff --git a/mm/bootmem.c b/mm/bootmem.c
index 484849bfc8c4..9da7d4097810 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -23,7 +23,6 @@ unsigned long max_low_pfn;
 unsigned long min_low_pfn;
 unsigned long max_pfn;
 
-static LIST_HEAD(bdata_list);
 #ifdef CONFIG_CRASH_DUMP
 /*
  * If we have booted due to a crash, max_pfn will be a very low value. We need
@@ -34,6 +33,8 @@ unsigned long saved_max_pfn;
 
 bootmem_data_t bootmem_node_data[MAX_NUMNODES] __initdata;
 
+static struct list_head bdata_list __initdata = LIST_HEAD_INIT(bdata_list);
+
 static int bootmem_debug;
 
 static int __init bootmem_debug_setup(char *buf)
@@ -73,20 +74,16 @@ unsigned long __init bootmem_bootmap_pages(unsigned long pages)
  */
 static void __init link_bootmem(bootmem_data_t *bdata)
 {
-	bootmem_data_t *ent;
+	struct list_head *iter;
 
-	if (list_empty(&bdata_list)) {
-		list_add(&bdata->list, &bdata_list);
-		return;
+	list_for_each(iter, &bdata_list) {
+		bootmem_data_t *ent;
+
+		ent = list_entry(iter, bootmem_data_t, list);
+		if (bdata->node_boot_start < ent->node_boot_start)
+			break;
 	}
-	/* insert in order */
-	list_for_each_entry(ent, &bdata_list, list) {
-		if (bdata->node_boot_start < ent->node_boot_start) {
-			list_add_tail(&bdata->list, &ent->list);
-			return;
-		}
-	}
-	list_add_tail(&bdata->list, &bdata_list);
+	list_add_tail(&bdata->list, iter);
 }
 
 /*

From 41546c17418fba08ece978bad72a33072715b8f3 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Wed, 23 Jul 2008 21:28:03 -0700
Subject: [PATCH 0278/1171] bootmem: clean up free_all_bootmem_core

Rewrite the code in a more concise way using less variables.

[akpm@linux-foundation.org: fix printk warnings]
Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Yinghai Lu <yhlu.kernel@gmail.com>
Cc: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/bootmem.c | 83 ++++++++++++++++++++++++----------------------------
 1 file changed, 38 insertions(+), 45 deletions(-)

diff --git a/mm/bootmem.c b/mm/bootmem.c
index 9da7d4097810..300d126ec533 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -144,66 +144,59 @@ unsigned long __init init_bootmem(unsigned long start, unsigned long pages)
 
 static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata)
 {
+	int aligned;
 	struct page *page;
-	unsigned long pfn;
-	unsigned long i, count;
-	unsigned long idx, pages;
-	unsigned long *map;
-	int gofast = 0;
+	unsigned long start, end, pages, count = 0;
 
-	BUG_ON(!bdata->node_bootmem_map);
+	if (!bdata->node_bootmem_map)
+		return 0;
+
+	start = PFN_DOWN(bdata->node_boot_start);
+	end = bdata->node_low_pfn;
 
-	count = 0;
-	/* first extant page of the node */
-	pfn = PFN_DOWN(bdata->node_boot_start);
-	idx = bdata->node_low_pfn - pfn;
-	map = bdata->node_bootmem_map;
 	/*
-	 * Check if we are aligned to BITS_PER_LONG pages.  If so, we might
-	 * be able to free page orders of that size at once.
+	 * If the start is aligned to the machines wordsize, we might
+	 * be able to free pages in bulks of that order.
 	 */
-	if (!(pfn & (BITS_PER_LONG-1)))
-		gofast = 1;
+	aligned = !(start & (BITS_PER_LONG - 1));
 
-	for (i = 0; i < idx; ) {
-		unsigned long v = ~map[i / BITS_PER_LONG];
+	bdebug("nid=%td start=%lx end=%lx aligned=%d\n",
+		bdata - bootmem_node_data, start, end, aligned);
 
-		if (gofast && v == ~0UL) {
-			int order;
+	while (start < end) {
+		unsigned long *map, idx, vec;
 
-			page = pfn_to_page(pfn);
+		map = bdata->node_bootmem_map;
+		idx = start - PFN_DOWN(bdata->node_boot_start);
+		vec = ~map[idx / BITS_PER_LONG];
+
+		if (aligned && vec == ~0UL && start + BITS_PER_LONG < end) {
+			int order = ilog2(BITS_PER_LONG);
+
+			__free_pages_bootmem(pfn_to_page(start), order);
 			count += BITS_PER_LONG;
-			order = ffs(BITS_PER_LONG) - 1;
-			__free_pages_bootmem(page, order);
-			i += BITS_PER_LONG;
-			page += BITS_PER_LONG;
-		} else if (v) {
-			unsigned long m;
-
-			page = pfn_to_page(pfn);
-			for (m = 1; m && i < idx; m<<=1, page++, i++) {
-				if (v & m) {
-					count++;
-					__free_pages_bootmem(page, 0);
-				}
-			}
 		} else {
-			i += BITS_PER_LONG;
+			unsigned long off = 0;
+
+			while (vec && off < BITS_PER_LONG) {
+				if (vec & 1) {
+					page = pfn_to_page(start + off);
+					__free_pages_bootmem(page, 0);
+					count++;
+				}
+				vec >>= 1;
+				off++;
+			}
 		}
-		pfn += BITS_PER_LONG;
+		start += BITS_PER_LONG;
 	}
 
-	/*
-	 * Now free the allocator bitmap itself, it's not
-	 * needed anymore:
-	 */
 	page = virt_to_page(bdata->node_bootmem_map);
 	pages = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start);
-	idx = bootmem_bootmap_pages(pages);
-	for (i = 0; i < idx; i++, page++)
-		__free_pages_bootmem(page, 0);
-	count += i;
-	bdata->node_bootmem_map = NULL;
+	pages = bootmem_bootmap_pages(pages);
+	count += pages;
+	while (pages--)
+		__free_pages_bootmem(page++, 0);
 
 	bdebug("nid=%td released=%lx\n", bdata - bootmem_node_data, count);
 

From 5f2809e69c7128f86316048221cf45146f69a4a0 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Wed, 23 Jul 2008 21:28:05 -0700
Subject: [PATCH 0279/1171] bootmem: clean up alloc_bootmem_core

alloc_bootmem_core has become quite nasty to read over time.  This is a
clean rewrite that keeps the semantics.

bdata->last_pos has been dropped.

bdata->last_success has been renamed to hint_idx and it is now an index
relative to the node's range.  Since further block searching might start
at this index, it is now set to the end of a succeeded allocation rather
than its beginning.

bdata->last_offset has been renamed to last_end_off to be more clear that
it represents the ending address of the last allocation relative to the
node.

[y-goto@jp.fujitsu.com: fix new alloc_bootmem_core()]
Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Signed-off-by: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/bootmem.h |   6 +-
 mm/bootmem.c            | 210 ++++++++++++++--------------------------
 2 files changed, 77 insertions(+), 139 deletions(-)

diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index 5000fd70b04f..90921d10ffa2 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -31,10 +31,8 @@ typedef struct bootmem_data {
 	unsigned long node_boot_start;
 	unsigned long node_low_pfn;
 	void *node_bootmem_map;
-	unsigned long last_offset;
-	unsigned long last_pos;
-	unsigned long last_success;	/* Previous allocation point.  To speed
-					 * up searching */
+	unsigned long last_end_off;
+	unsigned long hint_idx;
 	struct list_head list;
 } bootmem_data_t;
 
diff --git a/mm/bootmem.c b/mm/bootmem.c
index 300d126ec533..94ea612deccf 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -242,8 +242,9 @@ static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr,
 	 * considered reserved.
 	 */
 
-	if (addr >= bdata->node_boot_start && addr < bdata->last_success)
-		bdata->last_success = addr;
+	if (addr >= bdata->node_boot_start &&
+			PFN_DOWN(addr - bdata->node_boot_start) < bdata->hint_idx)
+		bdata->hint_idx = PFN_DOWN(addr - bdata->node_boot_start);
 
 	/*
 	 * Round up to index to the range.
@@ -431,36 +432,16 @@ int __init reserve_bootmem(unsigned long addr, unsigned long size,
 }
 #endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
 
-/*
- * We 'merge' subsequent allocations to save space. We might 'lose'
- * some fraction of a page if allocations cannot be satisfied due to
- * size constraints on boxes where there is physical RAM space
- * fragmentation - in these cases (mostly large memory boxes) this
- * is not a problem.
- *
- * On low memory boxes we get it right in 100% of the cases.
- *
- * alignment has to be a power of 2 value.
- *
- * NOTE:  This function is _not_ reentrant.
- */
-static void * __init
-alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size,
-		unsigned long align, unsigned long goal, unsigned long limit)
+static void * __init alloc_bootmem_core(struct bootmem_data *bdata,
+				unsigned long size, unsigned long align,
+				unsigned long goal, unsigned long limit)
 {
-	unsigned long areasize, preferred;
-	unsigned long i, start = 0, incr, eidx, end_pfn;
-	void *ret;
-	unsigned long node_boot_start;
-	void *node_bootmem_map;
+	unsigned long min, max, start, sidx, midx, step;
 
-	if (!size) {
-		printk("alloc_bootmem_core(): zero-sized request\n");
-		BUG();
-	}
-	BUG_ON(align & (align-1));
+	BUG_ON(!size);
+	BUG_ON(align & (align - 1));
+	BUG_ON(limit && goal + size > limit);
 
-	/* on nodes without memory - bootmem_map is NULL */
 	if (!bdata->node_bootmem_map)
 		return NULL;
 
@@ -468,126 +449,85 @@ alloc_bootmem_core(struct bootmem_data *bdata, unsigned long size,
 		bdata - bootmem_node_data, size, PAGE_ALIGN(size) >> PAGE_SHIFT,
 		align, goal, limit);
 
-	/* bdata->node_boot_start is supposed to be (12+6)bits alignment on x86_64 ? */
-	node_boot_start = bdata->node_boot_start;
-	node_bootmem_map = bdata->node_bootmem_map;
-	if (align) {
-		node_boot_start = ALIGN(bdata->node_boot_start, align);
-		if (node_boot_start > bdata->node_boot_start)
-			node_bootmem_map = (unsigned long *)bdata->node_bootmem_map +
-			    PFN_DOWN(node_boot_start - bdata->node_boot_start)/BITS_PER_LONG;
-	}
+	min = PFN_DOWN(bdata->node_boot_start);
+	max = bdata->node_low_pfn;
 
-	if (limit && node_boot_start >= limit)
+	goal >>= PAGE_SHIFT;
+	limit >>= PAGE_SHIFT;
+
+	if (limit && max > limit)
+		max = limit;
+	if (max <= min)
 		return NULL;
 
-	end_pfn = bdata->node_low_pfn;
-	limit = PFN_DOWN(limit);
-	if (limit && end_pfn > limit)
-		end_pfn = limit;
+	step = max(align >> PAGE_SHIFT, 1UL);
 
-	eidx = end_pfn - PFN_DOWN(node_boot_start);
+	if (goal && min < goal && goal < max)
+		start = ALIGN(goal, step);
+	else
+		start = ALIGN(min, step);
 
-	/*
-	 * We try to allocate bootmem pages above 'goal'
-	 * first, then we try to allocate lower pages.
-	 */
-	preferred = 0;
-	if (goal && PFN_DOWN(goal) < end_pfn) {
-		if (goal > node_boot_start)
-			preferred = goal - node_boot_start;
+	sidx = start - PFN_DOWN(bdata->node_boot_start);
+	midx = max - PFN_DOWN(bdata->node_boot_start);
 
-		if (bdata->last_success > node_boot_start &&
-			bdata->last_success - node_boot_start >= preferred)
-			if (!limit || (limit && limit > bdata->last_success))
-				preferred = bdata->last_success - node_boot_start;
+	if (bdata->hint_idx > sidx) {
+		/* Make sure we retry on failure */
+		goal = 1;
+		sidx = ALIGN(bdata->hint_idx, step);
 	}
 
-	preferred = PFN_DOWN(ALIGN(preferred, align));
-	areasize = (size + PAGE_SIZE-1) / PAGE_SIZE;
-	incr = align >> PAGE_SHIFT ? : 1;
+	while (1) {
+		int merge;
+		void *region;
+		unsigned long eidx, i, start_off, end_off;
+find_block:
+		sidx = find_next_zero_bit(bdata->node_bootmem_map, midx, sidx);
+		sidx = ALIGN(sidx, step);
+		eidx = sidx + PFN_UP(size);
 
-restart_scan:
-	for (i = preferred; i < eidx;) {
-		unsigned long j;
-
-		i = find_next_zero_bit(node_bootmem_map, eidx, i);
-		i = ALIGN(i, incr);
-		if (i >= eidx)
+		if (sidx >= midx || eidx > midx)
 			break;
-		if (test_bit(i, node_bootmem_map)) {
-			i += incr;
-			continue;
-		}
-		for (j = i + 1; j < i + areasize; ++j) {
-			if (j >= eidx)
-				goto fail_block;
-			if (test_bit(j, node_bootmem_map))
-				goto fail_block;
-		}
-		start = i;
-		goto found;
-	fail_block:
-		i = ALIGN(j, incr);
-		if (i == j)
-			i += incr;
+
+		for (i = sidx; i < eidx; i++)
+			if (test_bit(i, bdata->node_bootmem_map)) {
+				sidx = ALIGN(i, step);
+				if (sidx == i)
+					sidx += step;
+				goto find_block;
+			}
+
+		if (bdata->last_end_off &&
+				PFN_DOWN(bdata->last_end_off) + 1 == sidx)
+			start_off = ALIGN(bdata->last_end_off, align);
+		else
+			start_off = PFN_PHYS(sidx);
+
+		merge = PFN_DOWN(start_off) < sidx;
+		end_off = start_off + size;
+
+		bdata->last_end_off = end_off;
+		bdata->hint_idx = PFN_UP(end_off);
+
+		/*
+		 * Reserve the area now:
+		 */
+		for (i = PFN_DOWN(start_off) + merge;
+				i < PFN_UP(end_off); i++)
+			if (test_and_set_bit(i, bdata->node_bootmem_map))
+				BUG();
+
+		region = phys_to_virt(bdata->node_boot_start + start_off);
+		memset(region, 0, size);
+		return region;
 	}
 
-	if (preferred > 0) {
-		preferred = 0;
-		goto restart_scan;
+	if (goal) {
+		goal = 0;
+		sidx = 0;
+		goto find_block;
 	}
+
 	return NULL;
-
-found:
-	bdata->last_success = PFN_PHYS(start) + node_boot_start;
-	BUG_ON(start >= eidx);
-
-	/*
-	 * Is the next page of the previous allocation-end the start
-	 * of this allocation's buffer? If yes then we can 'merge'
-	 * the previous partial page with this allocation.
-	 */
-	if (align < PAGE_SIZE &&
-	    bdata->last_offset && bdata->last_pos+1 == start) {
-		unsigned long offset, remaining_size;
-		offset = ALIGN(bdata->last_offset, align);
-		BUG_ON(offset > PAGE_SIZE);
-		remaining_size = PAGE_SIZE - offset;
-		if (size < remaining_size) {
-			areasize = 0;
-			/* last_pos unchanged */
-			bdata->last_offset = offset + size;
-			ret = phys_to_virt(bdata->last_pos * PAGE_SIZE +
-					   offset + node_boot_start);
-		} else {
-			remaining_size = size - remaining_size;
-			areasize = (remaining_size + PAGE_SIZE-1) / PAGE_SIZE;
-			ret = phys_to_virt(bdata->last_pos * PAGE_SIZE +
-					   offset + node_boot_start);
-			bdata->last_pos = start + areasize - 1;
-			bdata->last_offset = remaining_size;
-		}
-		bdata->last_offset &= ~PAGE_MASK;
-	} else {
-		bdata->last_pos = start + areasize - 1;
-		bdata->last_offset = size & ~PAGE_MASK;
-		ret = phys_to_virt(start * PAGE_SIZE + node_boot_start);
-	}
-
-	bdebug("nid=%td start=%lx end=%lx\n",
-		bdata - bootmem_node_data,
-		start + PFN_DOWN(bdata->node_boot_start),
-		start + areasize + PFN_DOWN(bdata->node_boot_start));
-
-	/*
-	 * Reserve the area now:
-	 */
-	for (i = start; i < start + areasize; i++)
-		if (unlikely(test_and_set_bit(i, node_bootmem_map)))
-			BUG();
-	memset(ret, 0, size);
-	return ret;
 }
 
 /**

From d747fa4bcebcf3696607b86a6b0dafa644be0676 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Wed, 23 Jul 2008 21:28:05 -0700
Subject: [PATCH 0280/1171] bootmem: free/reserve helpers

Factor out the common operation of marking a range on the bitmap.

[akpm@linux-foundation.org: fix various warnings]
Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Yinghai Lu <yhlu.kernel@gmail.com>
Cc: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/bootmem.c | 65 ++++++++++++++++++++++++++++++++++------------------
 1 file changed, 43 insertions(+), 22 deletions(-)

diff --git a/mm/bootmem.c b/mm/bootmem.c
index 94ea612deccf..9d03ff651359 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -225,6 +225,44 @@ unsigned long __init free_all_bootmem(void)
 	return free_all_bootmem_core(NODE_DATA(0)->bdata);
 }
 
+static void __init __free(bootmem_data_t *bdata,
+			unsigned long sidx, unsigned long eidx)
+{
+	unsigned long idx;
+
+	bdebug("nid=%td start=%lx end=%lx\n", bdata - bootmem_node_data,
+		sidx + PFN_DOWN(bdata->node_boot_start),
+		eidx + PFN_DOWN(bdata->node_boot_start));
+
+	for (idx = sidx; idx < eidx; idx++)
+		if (!test_and_clear_bit(idx, bdata->node_bootmem_map))
+			BUG();
+}
+
+static int __init __reserve(bootmem_data_t *bdata, unsigned long sidx,
+			unsigned long eidx, int flags)
+{
+	unsigned long idx;
+	int exclusive = flags & BOOTMEM_EXCLUSIVE;
+
+	bdebug("nid=%td start=%lx end=%lx flags=%x\n",
+		bdata - bootmem_node_data,
+		sidx + PFN_DOWN(bdata->node_boot_start),
+		eidx + PFN_DOWN(bdata->node_boot_start),
+		flags);
+
+	for (idx = sidx; idx < eidx; idx++)
+		if (test_and_set_bit(idx, bdata->node_bootmem_map)) {
+			if (exclusive) {
+				__free(bdata, sidx, idx);
+				return -EBUSY;
+			}
+			bdebug("silent double reserve of PFN %lx\n",
+				idx + PFN_DOWN(bdata->node_boot_start));
+		}
+	return 0;
+}
+
 static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr,
 				     unsigned long size)
 {
@@ -258,14 +296,7 @@ static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr,
 	if (eidx > bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start))
 		eidx = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start);
 
-	bdebug("nid=%td start=%lx end=%lx\n", bdata - bootmem_node_data,
-		sidx + PFN_DOWN(bdata->node_boot_start),
-		eidx + PFN_DOWN(bdata->node_boot_start));
-
-	for (i = sidx; i < eidx; i++) {
-		if (unlikely(!test_and_clear_bit(i, bdata->node_bootmem_map)))
-			BUG();
-	}
+	__free(bdata, sidx, eidx);
 }
 
 /**
@@ -367,16 +398,7 @@ static void __init reserve_bootmem_core(bootmem_data_t *bdata,
 	if (eidx > bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start))
 		eidx = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start);
 
-	bdebug("nid=%td start=%lx end=%lx flags=%x\n",
-		bdata - bootmem_node_data,
-		sidx + PFN_DOWN(bdata->node_boot_start),
-		eidx + PFN_DOWN(bdata->node_boot_start),
-		flags);
-
-	for (i = sidx; i < eidx; i++)
-		if (test_and_set_bit(i, bdata->node_bootmem_map))
-			bdebug("hm, page %lx reserved twice.\n",
-				PFN_DOWN(bdata->node_boot_start) + i);
+	return __reserve(bdata, sidx, eidx, flags);
 }
 
 /**
@@ -511,10 +533,9 @@ find_block:
 		/*
 		 * Reserve the area now:
 		 */
-		for (i = PFN_DOWN(start_off) + merge;
-				i < PFN_UP(end_off); i++)
-			if (test_and_set_bit(i, bdata->node_bootmem_map))
-				BUG();
+		if (__reserve(bdata, PFN_DOWN(start_off) + merge,
+				PFN_UP(end_off), BOOTMEM_EXCLUSIVE))
+			BUG();
 
 		region = phys_to_virt(bdata->node_boot_start + start_off);
 		memset(region, 0, size);

From e2bf3cae515090fefe28329e71230dfe7ab873b1 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Wed, 23 Jul 2008 21:28:06 -0700
Subject: [PATCH 0281/1171] bootmem: factor out the marking of a PFN range

Introduce new helpers that mark a range that resides completely on a node
or node-agnostic ranges that might also span node boundaries.

The free/reserve API functions will then directly use these helpers.

Note that the free/reserve semantics become more strict: while the prior
code took basically arbitrary range arguments and marked the PFNs that
happen to fall into that range, the new code requires node-specific ranges
to be completely on the node.  The node-agnostic requests might span node
boundaries as long as the nodes are contiguous.

Passing ranges that do not satisfy these criteria is a bug.

[akpm@linux-foundation.org: fix printk warnings]
Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Yinghai Lu <yhlu.kernel@gmail.com>
Cc: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/bootmem.c | 188 +++++++++++++++++++--------------------------------
 1 file changed, 69 insertions(+), 119 deletions(-)

diff --git a/mm/bootmem.c b/mm/bootmem.c
index 9d03ff651359..e5415a5414a5 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -234,6 +234,9 @@ static void __init __free(bootmem_data_t *bdata,
 		sidx + PFN_DOWN(bdata->node_boot_start),
 		eidx + PFN_DOWN(bdata->node_boot_start));
 
+	if (bdata->hint_idx > sidx)
+		bdata->hint_idx = sidx;
+
 	for (idx = sidx; idx < eidx; idx++)
 		if (!test_and_clear_bit(idx, bdata->node_bootmem_map))
 			BUG();
@@ -263,40 +266,57 @@ static int __init __reserve(bootmem_data_t *bdata, unsigned long sidx,
 	return 0;
 }
 
-static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr,
-				     unsigned long size)
+static int __init mark_bootmem_node(bootmem_data_t *bdata,
+				unsigned long start, unsigned long end,
+				int reserve, int flags)
 {
 	unsigned long sidx, eidx;
-	unsigned long i;
 
-	BUG_ON(!size);
+	bdebug("nid=%td start=%lx end=%lx reserve=%d flags=%x\n",
+		bdata - bootmem_node_data, start, end, reserve, flags);
 
-	/* out range */
-	if (addr + size < bdata->node_boot_start ||
-		PFN_DOWN(addr) > bdata->node_low_pfn)
-		return;
-	/*
-	 * round down end of usable mem, partially free pages are
-	 * considered reserved.
-	 */
+	BUG_ON(start < PFN_DOWN(bdata->node_boot_start));
+	BUG_ON(end > bdata->node_low_pfn);
 
-	if (addr >= bdata->node_boot_start &&
-			PFN_DOWN(addr - bdata->node_boot_start) < bdata->hint_idx)
-		bdata->hint_idx = PFN_DOWN(addr - bdata->node_boot_start);
+	sidx = start - PFN_DOWN(bdata->node_boot_start);
+	eidx = end - PFN_DOWN(bdata->node_boot_start);
 
-	/*
-	 * Round up to index to the range.
-	 */
-	if (PFN_UP(addr) > PFN_DOWN(bdata->node_boot_start))
-		sidx = PFN_UP(addr) - PFN_DOWN(bdata->node_boot_start);
+	if (reserve)
+		return __reserve(bdata, sidx, eidx, flags);
 	else
-		sidx = 0;
+		__free(bdata, sidx, eidx);
+	return 0;
+}
 
-	eidx = PFN_DOWN(addr + size - bdata->node_boot_start);
-	if (eidx > bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start))
-		eidx = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start);
+static int __init mark_bootmem(unsigned long start, unsigned long end,
+				int reserve, int flags)
+{
+	unsigned long pos;
+	bootmem_data_t *bdata;
 
-	__free(bdata, sidx, eidx);
+	pos = start;
+	list_for_each_entry(bdata, &bdata_list, list) {
+		int err;
+		unsigned long max;
+
+		if (pos < PFN_DOWN(bdata->node_boot_start)) {
+			BUG_ON(pos != start);
+			continue;
+		}
+
+		max = min(bdata->node_low_pfn, end);
+
+		err = mark_bootmem_node(bdata, pos, max, reserve, flags);
+		if (reserve && err) {
+			mark_bootmem(start, pos, 0, 0);
+			return err;
+		}
+
+		if (max == end)
+			return 0;
+		pos = bdata->node_low_pfn;
+	}
+	BUG();
 }
 
 /**
@@ -307,12 +327,17 @@ static void __init free_bootmem_core(bootmem_data_t *bdata, unsigned long addr,
  *
  * Partial pages will be considered reserved and left as they are.
  *
- * Only physical pages that actually reside on @pgdat are marked.
+ * The range must reside completely on the specified node.
  */
 void __init free_bootmem_node(pg_data_t *pgdat, unsigned long physaddr,
 			      unsigned long size)
 {
-	free_bootmem_core(pgdat->bdata, physaddr, size);
+	unsigned long start, end;
+
+	start = PFN_UP(physaddr);
+	end = PFN_DOWN(physaddr + size);
+
+	mark_bootmem_node(pgdat->bdata, start, end, 0, 0);
 }
 
 /**
@@ -322,83 +347,16 @@ void __init free_bootmem_node(pg_data_t *pgdat, unsigned long physaddr,
  *
  * Partial pages will be considered reserved and left as they are.
  *
- * All physical pages within the range are marked, no matter what
- * node they reside on.
+ * The range must be contiguous but may span node boundaries.
  */
 void __init free_bootmem(unsigned long addr, unsigned long size)
 {
-	bootmem_data_t *bdata;
-	list_for_each_entry(bdata, &bdata_list, list)
-		free_bootmem_core(bdata, addr, size);
-}
+	unsigned long start, end;
 
-/*
- * Marks a particular physical memory range as unallocatable. Usable RAM
- * might be used for boot-time allocations - or it might get added
- * to the free page pool later on.
- */
-static int __init can_reserve_bootmem_core(bootmem_data_t *bdata,
-			unsigned long addr, unsigned long size, int flags)
-{
-	unsigned long sidx, eidx;
-	unsigned long i;
+	start = PFN_UP(addr);
+	end = PFN_DOWN(addr + size);
 
-	BUG_ON(!size);
-
-	/* out of range, don't hold other */
-	if (addr + size < bdata->node_boot_start ||
-		PFN_DOWN(addr) > bdata->node_low_pfn)
-		return 0;
-
-	/*
-	 * Round up to index to the range.
-	 */
-	if (addr > bdata->node_boot_start)
-		sidx= PFN_DOWN(addr - bdata->node_boot_start);
-	else
-		sidx = 0;
-
-	eidx = PFN_UP(addr + size - bdata->node_boot_start);
-	if (eidx > bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start))
-		eidx = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start);
-
-	for (i = sidx; i < eidx; i++) {
-		if (test_bit(i, bdata->node_bootmem_map)) {
-			if (flags & BOOTMEM_EXCLUSIVE)
-				return -EBUSY;
-		}
-	}
-
-	return 0;
-
-}
-
-static void __init reserve_bootmem_core(bootmem_data_t *bdata,
-			unsigned long addr, unsigned long size, int flags)
-{
-	unsigned long sidx, eidx;
-	unsigned long i;
-
-	BUG_ON(!size);
-
-	/* out of range */
-	if (addr + size < bdata->node_boot_start ||
-		PFN_DOWN(addr) > bdata->node_low_pfn)
-		return;
-
-	/*
-	 * Round up to index to the range.
-	 */
-	if (addr > bdata->node_boot_start)
-		sidx= PFN_DOWN(addr - bdata->node_boot_start);
-	else
-		sidx = 0;
-
-	eidx = PFN_UP(addr + size - bdata->node_boot_start);
-	if (eidx > bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start))
-		eidx = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start);
-
-	return __reserve(bdata, sidx, eidx, flags);
+	mark_bootmem(start, end, 0, 0);
 }
 
 /**
@@ -410,18 +368,17 @@ static void __init reserve_bootmem_core(bootmem_data_t *bdata,
  *
  * Partial pages will be reserved.
  *
- * Only physical pages that actually reside on @pgdat are marked.
+ * The range must reside completely on the specified node.
  */
 int __init reserve_bootmem_node(pg_data_t *pgdat, unsigned long physaddr,
 				 unsigned long size, int flags)
 {
-	int ret;
+	unsigned long start, end;
 
-	ret = can_reserve_bootmem_core(pgdat->bdata, physaddr, size, flags);
-	if (ret < 0)
-		return -ENOMEM;
-	reserve_bootmem_core(pgdat->bdata, physaddr, size, flags);
-	return 0;
+	start = PFN_DOWN(physaddr);
+	end = PFN_UP(physaddr + size);
+
+	return mark_bootmem_node(pgdat->bdata, start, end, 1, flags);
 }
 
 #ifndef CONFIG_HAVE_ARCH_BOOTMEM_NODE
@@ -433,24 +390,17 @@ int __init reserve_bootmem_node(pg_data_t *pgdat, unsigned long physaddr,
  *
  * Partial pages will be reserved.
  *
- * All physical pages within the range are marked, no matter what
- * node they reside on.
+ * The range must be contiguous but may span node boundaries.
  */
 int __init reserve_bootmem(unsigned long addr, unsigned long size,
 			    int flags)
 {
-	bootmem_data_t *bdata;
-	int ret;
+	unsigned long start, end;
 
-	list_for_each_entry(bdata, &bdata_list, list) {
-		ret = can_reserve_bootmem_core(bdata, addr, size, flags);
-		if (ret < 0)
-			return ret;
-	}
-	list_for_each_entry(bdata, &bdata_list, list)
-		reserve_bootmem_core(bdata, addr, size, flags);
+	start = PFN_DOWN(addr);
+	end = PFN_UP(addr + size);
 
-	return 0;
+	return mark_bootmem(start, end, 1, flags);
 }
 #endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
 
@@ -663,7 +613,7 @@ void * __init alloc_bootmem_section(unsigned long size,
 	if (start_nr != section_nr || end_nr != section_nr) {
 		printk(KERN_WARNING "alloc_bootmem failed on section %ld.\n",
 		       section_nr);
-		free_bootmem_core(pgdat->bdata, __pa(ptr), size);
+		free_bootmem_node(pgdat, __pa(ptr), size);
 		ptr = NULL;
 	}
 

From 0f3caba211babef6e3fbde1ba76ddc79321bc92f Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Wed, 23 Jul 2008 21:28:07 -0700
Subject: [PATCH 0282/1171] bootmem: respect goal more likely

The old node-agnostic code tried allocating on all nodes starting from the
one with the lowest range.  alloc_bootmem_core retried without the goal if
it could not satisfy it and so the goal was only respected at all when it
happened to be on the first (lowest page numbers) node (or theoretically
if allocations failed on all nodes before to the one holding the goal).

Introduce a non-panicking helper that starts allocating from the node
holding the goal and falls back only after all thes tries failed, thus
moving the goal fallback code out of alloc_bootmem_core.

Make all other allocation functions benefit from this new helper.

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Yinghai Lu <yhlu.kernel@gmail.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/bootmem.c | 92 ++++++++++++++++++++++++++++++----------------------
 1 file changed, 54 insertions(+), 38 deletions(-)

diff --git a/mm/bootmem.c b/mm/bootmem.c
index e5415a5414a5..89646f77b427 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -408,6 +408,7 @@ static void * __init alloc_bootmem_core(struct bootmem_data *bdata,
 				unsigned long size, unsigned long align,
 				unsigned long goal, unsigned long limit)
 {
+	unsigned long fallback = 0;
 	unsigned long min, max, start, sidx, midx, step;
 
 	BUG_ON(!size);
@@ -443,8 +444,11 @@ static void * __init alloc_bootmem_core(struct bootmem_data *bdata,
 	midx = max - PFN_DOWN(bdata->node_boot_start);
 
 	if (bdata->hint_idx > sidx) {
-		/* Make sure we retry on failure */
-		goal = 1;
+		/*
+		 * Handle the valid case of sidx being zero and still
+		 * catch the fallback below.
+		 */
+		fallback = sidx + 1;
 		sidx = ALIGN(bdata->hint_idx, step);
 	}
 
@@ -492,10 +496,39 @@ find_block:
 		return region;
 	}
 
+	if (fallback) {
+		sidx = ALIGN(fallback - 1, step);
+		fallback = 0;
+		goto find_block;
+	}
+
+	return NULL;
+}
+
+static void * __init ___alloc_bootmem_nopanic(unsigned long size,
+					unsigned long align,
+					unsigned long goal,
+					unsigned long limit)
+{
+	bootmem_data_t *bdata;
+
+restart:
+	list_for_each_entry(bdata, &bdata_list, list) {
+		void *region;
+
+		if (goal && bdata->node_low_pfn <= PFN_DOWN(goal))
+			continue;
+		if (limit && bdata->node_boot_start >= limit)
+			break;
+
+		region = alloc_bootmem_core(bdata, size, align, goal, limit);
+		if (region)
+			return region;
+	}
+
 	if (goal) {
 		goal = 0;
-		sidx = 0;
-		goto find_block;
+		goto restart;
 	}
 
 	return NULL;
@@ -515,16 +548,23 @@ find_block:
  * Returns NULL on failure.
  */
 void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align,
-				      unsigned long goal)
+					unsigned long goal)
 {
-	bootmem_data_t *bdata;
-	void *ptr;
+	return ___alloc_bootmem_nopanic(size, align, goal, 0);
+}
 
-	list_for_each_entry(bdata, &bdata_list, list) {
-		ptr = alloc_bootmem_core(bdata, size, align, goal, 0);
-		if (ptr)
-			return ptr;
-	}
+static void * __init ___alloc_bootmem(unsigned long size, unsigned long align,
+					unsigned long goal, unsigned long limit)
+{
+	void *mem = ___alloc_bootmem_nopanic(size, align, goal, limit);
+
+	if (mem)
+		return mem;
+	/*
+	 * Whoops, we cannot satisfy the allocation request.
+	 */
+	printk(KERN_ALERT "bootmem alloc of %lu bytes failed!\n", size);
+	panic("Out of memory");
 	return NULL;
 }
 
@@ -544,16 +584,7 @@ void * __init __alloc_bootmem_nopanic(unsigned long size, unsigned long align,
 void * __init __alloc_bootmem(unsigned long size, unsigned long align,
 			      unsigned long goal)
 {
-	void *mem = __alloc_bootmem_nopanic(size,align,goal);
-
-	if (mem)
-		return mem;
-	/*
-	 * Whoops, we cannot satisfy the allocation request.
-	 */
-	printk(KERN_ALERT "bootmem alloc of %lu bytes failed!\n", size);
-	panic("Out of memory");
-	return NULL;
+	return ___alloc_bootmem(size, align, goal, 0);
 }
 
 /**
@@ -653,22 +684,7 @@ void * __init __alloc_bootmem_node_nopanic(pg_data_t *pgdat, unsigned long size,
 void * __init __alloc_bootmem_low(unsigned long size, unsigned long align,
 				  unsigned long goal)
 {
-	bootmem_data_t *bdata;
-	void *ptr;
-
-	list_for_each_entry(bdata, &bdata_list, list) {
-		ptr = alloc_bootmem_core(bdata, size, align, goal,
-					ARCH_LOW_ADDRESS_LIMIT);
-		if (ptr)
-			return ptr;
-	}
-
-	/*
-	 * Whoops, we cannot satisfy the allocation request.
-	 */
-	printk(KERN_ALERT "low bootmem alloc of %lu bytes failed!\n", size);
-	panic("Out of low memory");
-	return NULL;
+	return ___alloc_bootmem(size, align, goal, ARCH_LOW_ADDRESS_LIMIT);
 }
 
 /**

From 4cc278b721d5bf3569dfc5f1100253042e097bc3 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Wed, 23 Jul 2008 21:28:08 -0700
Subject: [PATCH 0283/1171] bootmem: Make __alloc_bootmem_low_node fall back to
 other nodes

__alloc_bootmem_node already does this, make the interface consistent.

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Yinghai Lu <yhlu.kernel@gmail.com>
Cc: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/bootmem.c | 25 ++++++++++++++++---------
 1 file changed, 16 insertions(+), 9 deletions(-)

diff --git a/mm/bootmem.c b/mm/bootmem.c
index 89646f77b427..459da4710b8f 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -587,6 +587,19 @@ void * __init __alloc_bootmem(unsigned long size, unsigned long align,
 	return ___alloc_bootmem(size, align, goal, 0);
 }
 
+static void * __init ___alloc_bootmem_node(bootmem_data_t *bdata,
+				unsigned long size, unsigned long align,
+				unsigned long goal, unsigned long limit)
+{
+	void *ptr;
+
+	ptr = alloc_bootmem_core(bdata, size, align, goal, limit);
+	if (ptr)
+		return ptr;
+
+	return ___alloc_bootmem(size, align, goal, limit);
+}
+
 /**
  * __alloc_bootmem_node - allocate boot memory from a specific node
  * @pgdat: node to allocate from
@@ -605,13 +618,7 @@ void * __init __alloc_bootmem(unsigned long size, unsigned long align,
 void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size,
 				   unsigned long align, unsigned long goal)
 {
-	void *ptr;
-
-	ptr = alloc_bootmem_core(pgdat->bdata, size, align, goal, 0);
-	if (ptr)
-		return ptr;
-
-	return __alloc_bootmem(size, align, goal);
+	return ___alloc_bootmem_node(pgdat->bdata, size, align, goal, 0);
 }
 
 #ifdef CONFIG_SPARSEMEM
@@ -705,6 +712,6 @@ void * __init __alloc_bootmem_low(unsigned long size, unsigned long align,
 void * __init __alloc_bootmem_low_node(pg_data_t *pgdat, unsigned long size,
 				       unsigned long align, unsigned long goal)
 {
-	return alloc_bootmem_core(pgdat->bdata, size, align, goal,
-				ARCH_LOW_ADDRESS_LIMIT);
+	return ___alloc_bootmem_node(pgdat->bdata, size, align,
+				goal, ARCH_LOW_ADDRESS_LIMIT);
 }

From 75a56cfe9fdb064d1db1cfbc564315fddb756fb1 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Wed, 23 Jul 2008 21:28:09 -0700
Subject: [PATCH 0284/1171] bootmem: revisit alloc_bootmem_section

Since alloc_bootmem_core does no goal-fallback anymore and just returns
NULL if the allocation fails, we might now use it in alloc_bootmem_section
without all the fixup code for a misplaced allocation.

Also, the limit can be the first PFN of the next section as the semantics
is that the limit is _above_ the allocated region, not within.

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/bootmem.c | 27 ++++++---------------------
 1 file changed, 6 insertions(+), 21 deletions(-)

diff --git a/mm/bootmem.c b/mm/bootmem.c
index 459da4710b8f..282b786c2b15 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -632,30 +632,15 @@ void * __init __alloc_bootmem_node(pg_data_t *pgdat, unsigned long size,
 void * __init alloc_bootmem_section(unsigned long size,
 				    unsigned long section_nr)
 {
-	void *ptr;
-	unsigned long limit, goal, start_nr, end_nr, pfn;
-	struct pglist_data *pgdat;
+	bootmem_data_t *bdata;
+	unsigned long pfn, goal, limit;
 
 	pfn = section_nr_to_pfn(section_nr);
-	goal = PFN_PHYS(pfn);
-	limit = PFN_PHYS(section_nr_to_pfn(section_nr + 1)) - 1;
-	pgdat = NODE_DATA(early_pfn_to_nid(pfn));
-	ptr = alloc_bootmem_core(pgdat->bdata, size, SMP_CACHE_BYTES, goal,
-				limit);
+	goal = pfn << PAGE_SHIFT;
+	limit = section_nr_to_pfn(section_nr + 1) << PAGE_SHIFT;
+	bdata = &bootmem_node_data[early_pfn_to_nid(pfn)];
 
-	if (!ptr)
-		return NULL;
-
-	start_nr = pfn_to_section_nr(PFN_DOWN(__pa(ptr)));
-	end_nr = pfn_to_section_nr(PFN_DOWN(__pa(ptr) + size));
-	if (start_nr != section_nr || end_nr != section_nr) {
-		printk(KERN_WARNING "alloc_bootmem failed on section %ld.\n",
-		       section_nr);
-		free_bootmem_node(pgdat, __pa(ptr), size);
-		ptr = NULL;
-	}
-
-	return ptr;
+	return alloc_bootmem_core(bdata, size, SMP_CACHE_BYTES, goal, limit);
 }
 #endif
 

From 3560e249abda6bee41a07a7bf0383a6e193e2839 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Wed, 23 Jul 2008 21:28:09 -0700
Subject: [PATCH 0285/1171] bootmem: replace node_boot_start in struct
 bootmem_data

Almost all users of this field need a PFN instead of a physical address,
so replace node_boot_start with node_min_pfn.

[Lee.Schermerhorn@hp.com: fix spurious BUG_ON() in mark_bootmem()]
Signed-off-by: Johannes Weiner <hannes@saeureba.de>
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/alpha/mm/numa.c     |  2 +-
 arch/arm/plat-omap/fb.c  |  4 +---
 arch/avr32/mm/init.c     |  3 +--
 arch/ia64/mm/discontig.c | 19 ++++++++++---------
 arch/m32r/mm/discontig.c |  3 +--
 arch/m32r/mm/init.c      |  4 +---
 arch/mn10300/mm/init.c   |  6 +++---
 arch/sh/mm/init.c        |  2 +-
 include/linux/bootmem.h  |  2 +-
 mm/bootmem.c             | 40 +++++++++++++++++++++-------------------
 10 files changed, 41 insertions(+), 44 deletions(-)

diff --git a/arch/alpha/mm/numa.c b/arch/alpha/mm/numa.c
index def0c74a78a8..d8c4ceaf00b9 100644
--- a/arch/alpha/mm/numa.c
+++ b/arch/alpha/mm/numa.c
@@ -304,7 +304,7 @@ void __init paging_init(void)
 
 	for_each_online_node(nid) {
 		bootmem_data_t *bdata = &bootmem_node_data[nid];
-		unsigned long start_pfn = bdata->node_boot_start >> PAGE_SHIFT;
+		unsigned long start_pfn = bdata->node_min_pfn;
 		unsigned long end_pfn = bdata->node_low_pfn;
 
 		if (dma_local_pfn >= end_pfn - start_pfn)
diff --git a/arch/arm/plat-omap/fb.c b/arch/arm/plat-omap/fb.c
index 7854f19b77cf..96d6f0619733 100644
--- a/arch/arm/plat-omap/fb.c
+++ b/arch/arm/plat-omap/fb.c
@@ -182,7 +182,7 @@ void __init omapfb_reserve_sdram(void)
 		return;
 
 	bdata = NODE_DATA(0)->bdata;
-	sdram_start = bdata->node_boot_start;
+	sdram_start = bdata->node_min_pfn << PAGE_SHIFT;
 	sdram_size = (bdata->node_low_pfn << PAGE_SHIFT) - sdram_start;
 	reserved = 0;
 	for (i = 0; ; i++) {
@@ -340,5 +340,3 @@ unsigned long omapfb_reserve_sram(unsigned long sram_pstart,
 
 
 #endif
-
-
diff --git a/arch/avr32/mm/init.c b/arch/avr32/mm/init.c
index 786de88a82a7..3c85fdaa9487 100644
--- a/arch/avr32/mm/init.c
+++ b/arch/avr32/mm/init.c
@@ -119,8 +119,7 @@ void __init paging_init(void)
 		unsigned long zones_size[MAX_NR_ZONES];
 		unsigned long low, start_pfn;
 
-		start_pfn = pgdat->bdata->node_boot_start;
-		start_pfn >>= PAGE_SHIFT;
+		start_pfn = pgdat->bdata->node_min_pfn;
 		low = pgdat->bdata->node_low_pfn;
 
 		memset(zones_size, 0, sizeof(zones_size));
diff --git a/arch/ia64/mm/discontig.c b/arch/ia64/mm/discontig.c
index 2fcf8464331e..d83125e1ed27 100644
--- a/arch/ia64/mm/discontig.c
+++ b/arch/ia64/mm/discontig.c
@@ -74,17 +74,17 @@ pg_data_t *pgdat_list[MAX_NUMNODES];
 static int __init build_node_maps(unsigned long start, unsigned long len,
 				  int node)
 {
-	unsigned long cstart, epfn, end = start + len;
+	unsigned long spfn, epfn, end = start + len;
 	struct bootmem_data *bdp = &bootmem_node_data[node];
 
 	epfn = GRANULEROUNDUP(end) >> PAGE_SHIFT;
-	cstart = GRANULEROUNDDOWN(start);
+	spfn = GRANULEROUNDDOWN(start) >> PAGE_SHIFT;
 
 	if (!bdp->node_low_pfn) {
-		bdp->node_boot_start = cstart;
+		bdp->node_min_pfn = spfn;
 		bdp->node_low_pfn = epfn;
 	} else {
-		bdp->node_boot_start = min(cstart, bdp->node_boot_start);
+		bdp->node_min_pfn = min(spfn, bdp->node_min_pfn);
 		bdp->node_low_pfn = max(epfn, bdp->node_low_pfn);
 	}
 
@@ -221,20 +221,21 @@ static void __init fill_pernode(int node, unsigned long pernode,
 static int __init find_pernode_space(unsigned long start, unsigned long len,
 				     int node)
 {
-	unsigned long epfn;
+	unsigned long spfn, epfn;
 	unsigned long pernodesize = 0, pernode, pages, mapsize;
 	struct bootmem_data *bdp = &bootmem_node_data[node];
 
+	spfn = start >> PAGE_SHIFT;
 	epfn = (start + len) >> PAGE_SHIFT;
 
-	pages = bdp->node_low_pfn - (bdp->node_boot_start >> PAGE_SHIFT);
+	pages = bdp->node_low_pfn - bdp->node_min_pfn;
 	mapsize = bootmem_bootmap_pages(pages) << PAGE_SHIFT;
 
 	/*
 	 * Make sure this memory falls within this node's usable memory
 	 * since we may have thrown some away in build_maps().
 	 */
-	if (start < bdp->node_boot_start || epfn > bdp->node_low_pfn)
+	if (spfn < bdp->node_min_pfn || epfn > bdp->node_low_pfn)
 		return 0;
 
 	/* Don't setup this node's local space twice... */
@@ -296,7 +297,7 @@ static void __init reserve_pernode_space(void)
 		bdp = pdp->bdata;
 
 		/* First the bootmem_map itself */
-		pages = bdp->node_low_pfn - (bdp->node_boot_start>>PAGE_SHIFT);
+		pages = bdp->node_low_pfn - bdp->node_min_pfn;
 		size = bootmem_bootmap_pages(pages) << PAGE_SHIFT;
 		base = __pa(bdp->node_bootmem_map);
 		reserve_bootmem_node(pdp, base, size, BOOTMEM_DEFAULT);
@@ -466,7 +467,7 @@ void __init find_memory(void)
 
 		init_bootmem_node(pgdat_list[node],
 				  map>>PAGE_SHIFT,
-				  bdp->node_boot_start>>PAGE_SHIFT,
+				  bdp->node_min_pfn,
 				  bdp->node_low_pfn);
 	}
 
diff --git a/arch/m32r/mm/discontig.c b/arch/m32r/mm/discontig.c
index cc23934bc41e..cbc3c4c54566 100644
--- a/arch/m32r/mm/discontig.c
+++ b/arch/m32r/mm/discontig.c
@@ -123,8 +123,7 @@ unsigned long __init setup_memory(void)
 	return max_low_pfn;
 }
 
-#define START_PFN(nid)	\
-	(NODE_DATA(nid)->bdata->node_boot_start >> PAGE_SHIFT)
+#define START_PFN(nid)		(NODE_DATA(nid)->bdata->node_min_pfn)
 #define MAX_LOW_PFN(nid)	(NODE_DATA(nid)->bdata->node_low_pfn)
 
 unsigned long __init zone_sizes_init(void)
diff --git a/arch/m32r/mm/init.c b/arch/m32r/mm/init.c
index 28799af15e95..2554eb59cfef 100644
--- a/arch/m32r/mm/init.c
+++ b/arch/m32r/mm/init.c
@@ -93,8 +93,7 @@ void free_initrd_mem(unsigned long, unsigned long);
 #endif
 
 /* It'd be good if these lines were in the standard header file. */
-#define START_PFN(nid)	\
-	(NODE_DATA(nid)->bdata->node_boot_start >> PAGE_SHIFT)
+#define START_PFN(nid)		(NODE_DATA(nid)->bdata->node_min_pfn)
 #define MAX_LOW_PFN(nid)	(NODE_DATA(nid)->bdata->node_low_pfn)
 
 #ifndef CONFIG_DISCONTIGMEM
@@ -252,4 +251,3 @@ void free_initrd_mem(unsigned long start, unsigned long end)
 	printk (KERN_INFO "Freeing initrd memory: %ldk freed\n", (end - start) >> 10);
 }
 #endif
-
diff --git a/arch/mn10300/mm/init.c b/arch/mn10300/mm/init.c
index 8c5d88c7b90a..8cee387a24fd 100644
--- a/arch/mn10300/mm/init.c
+++ b/arch/mn10300/mm/init.c
@@ -67,8 +67,8 @@ void __init paging_init(void)
 
 	/* declare the sizes of the RAM zones (only use the normal zone) */
 	zones_size[ZONE_NORMAL] =
-		(contig_page_data.bdata->node_low_pfn) -
-		(contig_page_data.bdata->node_boot_start >> PAGE_SHIFT);
+		contig_page_data.bdata->node_low_pfn -
+		contig_page_data.bdata->node_min_pfn;
 
 	/* pass the memory from the bootmem allocator to the main allocator */
 	free_area_init(zones_size);
@@ -87,7 +87,7 @@ void __init mem_init(void)
 	if (!mem_map)
 		BUG();
 
-#define START_PFN	(contig_page_data.bdata->node_boot_start >> PAGE_SHIFT)
+#define START_PFN	(contig_page_data.bdata->node_min_pfn)
 #define MAX_LOW_PFN	(contig_page_data.bdata->node_low_pfn)
 
 	max_mapnr = num_physpages = MAX_LOW_PFN - START_PFN;
diff --git a/arch/sh/mm/init.c b/arch/sh/mm/init.c
index d7df26bd1e54..d652d375eb1e 100644
--- a/arch/sh/mm/init.c
+++ b/arch/sh/mm/init.c
@@ -191,7 +191,7 @@ void __init paging_init(void)
 		pg_data_t *pgdat = NODE_DATA(nid);
 		unsigned long low, start_pfn;
 
-		start_pfn = pgdat->bdata->node_boot_start >> PAGE_SHIFT;
+		start_pfn = pgdat->bdata->node_min_pfn;
 		low = pgdat->bdata->node_low_pfn;
 
 		if (max_zone_pfns[ZONE_NORMAL] < low)
diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index 90921d10ffa2..4ddf2922fc8d 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -28,7 +28,7 @@ extern unsigned long saved_max_pfn;
  * memory pages (including holes) on the node.
  */
 typedef struct bootmem_data {
-	unsigned long node_boot_start;
+	unsigned long node_min_pfn;
 	unsigned long node_low_pfn;
 	void *node_bootmem_map;
 	unsigned long last_end_off;
diff --git a/mm/bootmem.c b/mm/bootmem.c
index 282b786c2b15..4af15d0340ad 100644
--- a/mm/bootmem.c
+++ b/mm/bootmem.c
@@ -80,7 +80,7 @@ static void __init link_bootmem(bootmem_data_t *bdata)
 		bootmem_data_t *ent;
 
 		ent = list_entry(iter, bootmem_data_t, list);
-		if (bdata->node_boot_start < ent->node_boot_start)
+		if (bdata->node_min_pfn < ent->node_min_pfn)
 			break;
 	}
 	list_add_tail(&bdata->list, iter);
@@ -96,7 +96,7 @@ static unsigned long __init init_bootmem_core(bootmem_data_t *bdata,
 
 	mminit_validate_memmodel_limits(&start, &end);
 	bdata->node_bootmem_map = phys_to_virt(PFN_PHYS(mapstart));
-	bdata->node_boot_start = PFN_PHYS(start);
+	bdata->node_min_pfn = start;
 	bdata->node_low_pfn = end;
 	link_bootmem(bdata);
 
@@ -151,7 +151,7 @@ static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata)
 	if (!bdata->node_bootmem_map)
 		return 0;
 
-	start = PFN_DOWN(bdata->node_boot_start);
+	start = bdata->node_min_pfn;
 	end = bdata->node_low_pfn;
 
 	/*
@@ -167,7 +167,7 @@ static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata)
 		unsigned long *map, idx, vec;
 
 		map = bdata->node_bootmem_map;
-		idx = start - PFN_DOWN(bdata->node_boot_start);
+		idx = start - bdata->node_min_pfn;
 		vec = ~map[idx / BITS_PER_LONG];
 
 		if (aligned && vec == ~0UL && start + BITS_PER_LONG < end) {
@@ -192,7 +192,7 @@ static unsigned long __init free_all_bootmem_core(bootmem_data_t *bdata)
 	}
 
 	page = virt_to_page(bdata->node_bootmem_map);
-	pages = bdata->node_low_pfn - PFN_DOWN(bdata->node_boot_start);
+	pages = bdata->node_low_pfn - bdata->node_min_pfn;
 	pages = bootmem_bootmap_pages(pages);
 	count += pages;
 	while (pages--)
@@ -231,8 +231,8 @@ static void __init __free(bootmem_data_t *bdata,
 	unsigned long idx;
 
 	bdebug("nid=%td start=%lx end=%lx\n", bdata - bootmem_node_data,
-		sidx + PFN_DOWN(bdata->node_boot_start),
-		eidx + PFN_DOWN(bdata->node_boot_start));
+		sidx + bdata->node_min_pfn,
+		eidx + bdata->node_min_pfn);
 
 	if (bdata->hint_idx > sidx)
 		bdata->hint_idx = sidx;
@@ -250,8 +250,8 @@ static int __init __reserve(bootmem_data_t *bdata, unsigned long sidx,
 
 	bdebug("nid=%td start=%lx end=%lx flags=%x\n",
 		bdata - bootmem_node_data,
-		sidx + PFN_DOWN(bdata->node_boot_start),
-		eidx + PFN_DOWN(bdata->node_boot_start),
+		sidx + bdata->node_min_pfn,
+		eidx + bdata->node_min_pfn,
 		flags);
 
 	for (idx = sidx; idx < eidx; idx++)
@@ -261,7 +261,7 @@ static int __init __reserve(bootmem_data_t *bdata, unsigned long sidx,
 				return -EBUSY;
 			}
 			bdebug("silent double reserve of PFN %lx\n",
-				idx + PFN_DOWN(bdata->node_boot_start));
+				idx + bdata->node_min_pfn);
 		}
 	return 0;
 }
@@ -275,11 +275,11 @@ static int __init mark_bootmem_node(bootmem_data_t *bdata,
 	bdebug("nid=%td start=%lx end=%lx reserve=%d flags=%x\n",
 		bdata - bootmem_node_data, start, end, reserve, flags);
 
-	BUG_ON(start < PFN_DOWN(bdata->node_boot_start));
+	BUG_ON(start < bdata->node_min_pfn);
 	BUG_ON(end > bdata->node_low_pfn);
 
-	sidx = start - PFN_DOWN(bdata->node_boot_start);
-	eidx = end - PFN_DOWN(bdata->node_boot_start);
+	sidx = start - bdata->node_min_pfn;
+	eidx = end - bdata->node_min_pfn;
 
 	if (reserve)
 		return __reserve(bdata, sidx, eidx, flags);
@@ -299,7 +299,8 @@ static int __init mark_bootmem(unsigned long start, unsigned long end,
 		int err;
 		unsigned long max;
 
-		if (pos < PFN_DOWN(bdata->node_boot_start)) {
+		if (pos < bdata->node_min_pfn ||
+		    pos >= bdata->node_low_pfn) {
 			BUG_ON(pos != start);
 			continue;
 		}
@@ -422,7 +423,7 @@ static void * __init alloc_bootmem_core(struct bootmem_data *bdata,
 		bdata - bootmem_node_data, size, PAGE_ALIGN(size) >> PAGE_SHIFT,
 		align, goal, limit);
 
-	min = PFN_DOWN(bdata->node_boot_start);
+	min = bdata->node_min_pfn;
 	max = bdata->node_low_pfn;
 
 	goal >>= PAGE_SHIFT;
@@ -440,8 +441,8 @@ static void * __init alloc_bootmem_core(struct bootmem_data *bdata,
 	else
 		start = ALIGN(min, step);
 
-	sidx = start - PFN_DOWN(bdata->node_boot_start);
-	midx = max - PFN_DOWN(bdata->node_boot_start);
+	sidx = start - bdata->node_min_pfn;;
+	midx = max - bdata->node_min_pfn;
 
 	if (bdata->hint_idx > sidx) {
 		/*
@@ -491,7 +492,8 @@ find_block:
 				PFN_UP(end_off), BOOTMEM_EXCLUSIVE))
 			BUG();
 
-		region = phys_to_virt(bdata->node_boot_start + start_off);
+		region = phys_to_virt(PFN_PHYS(bdata->node_min_pfn) +
+				start_off);
 		memset(region, 0, size);
 		return region;
 	}
@@ -518,7 +520,7 @@ restart:
 
 		if (goal && bdata->node_low_pfn <= PFN_DOWN(goal))
 			continue;
-		if (limit && bdata->node_boot_start >= limit)
+		if (limit && bdata->node_min_pfn >= PFN_DOWN(limit))
 			break;
 
 		region = alloc_bootmem_core(bdata, size, align, goal, limit);

From 2be0ffe2b29bd31d3debd0877797892ff2d91f4c Mon Sep 17 00:00:00 2001
From: Timur Tabi <timur@freescale.com>
Date: Wed, 23 Jul 2008 21:28:11 -0700
Subject: [PATCH 0286/1171] mm: add alloc_pages_exact() and free_pages_exact()

alloc_pages_exact() is similar to alloc_pages(), except that it allocates
the minimum number of pages to fulfill the request.  This is useful if you
want to allocate a very large buffer that is slightly larger than an even
power-of-two number of pages.  In that case, alloc_pages() will waste a
lot of memory.

I have a video driver that wants to allocate a 5MB buffer.  alloc_pages()
wiill waste 3MB of physically-contiguous memory.

Signed-off-by: Timur Tabi <timur@freescale.com>
Cc: Andi Kleen <andi@firstfloor.org>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/gfp.h |  3 +++
 mm/page_alloc.c     | 53 +++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 56 insertions(+)

diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index f640ed241422..e8003afeffba 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -228,6 +228,9 @@ extern struct page *alloc_page_vma(gfp_t gfp_mask,
 extern unsigned long __get_free_pages(gfp_t gfp_mask, unsigned int order);
 extern unsigned long get_zeroed_page(gfp_t gfp_mask);
 
+void *alloc_pages_exact(size_t size, gfp_t gfp_mask);
+void free_pages_exact(void *virt, size_t size);
+
 #define __get_free_page(gfp_mask) \
 		__get_free_pages((gfp_mask),0)
 
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index eaa86671ebbd..8d528d57b403 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1697,6 +1697,59 @@ void free_pages(unsigned long addr, unsigned int order)
 
 EXPORT_SYMBOL(free_pages);
 
+/**
+ * alloc_pages_exact - allocate an exact number physically-contiguous pages.
+ * @size: the number of bytes to allocate
+ * @gfp_mask: GFP flags for the allocation
+ *
+ * This function is similar to alloc_pages(), except that it allocates the
+ * minimum number of pages to satisfy the request.  alloc_pages() can only
+ * allocate memory in power-of-two pages.
+ *
+ * This function is also limited by MAX_ORDER.
+ *
+ * Memory allocated by this function must be released by free_pages_exact().
+ */
+void *alloc_pages_exact(size_t size, gfp_t gfp_mask)
+{
+	unsigned int order = get_order(size);
+	unsigned long addr;
+
+	addr = __get_free_pages(gfp_mask, order);
+	if (addr) {
+		unsigned long alloc_end = addr + (PAGE_SIZE << order);
+		unsigned long used = addr + PAGE_ALIGN(size);
+
+		split_page(virt_to_page(addr), order);
+		while (used < alloc_end) {
+			free_page(used);
+			used += PAGE_SIZE;
+		}
+	}
+
+	return (void *)addr;
+}
+EXPORT_SYMBOL(alloc_pages_exact);
+
+/**
+ * free_pages_exact - release memory allocated via alloc_pages_exact()
+ * @virt: the value returned by alloc_pages_exact.
+ * @size: size of allocation, same value as passed to alloc_pages_exact().
+ *
+ * Release the memory allocated by a previous call to alloc_pages_exact.
+ */
+void free_pages_exact(void *virt, size_t size)
+{
+	unsigned long addr = (unsigned long)virt;
+	unsigned long end = addr + PAGE_ALIGN(size);
+
+	while (addr < end) {
+		free_page(addr);
+		addr += PAGE_SIZE;
+	}
+}
+EXPORT_SYMBOL(free_pages_exact);
+
 static unsigned int nr_free_zone_pages(int offset)
 {
 	struct zoneref *z;

From b69a7288ea7bf171328f313f0edae629f50e3bdb Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Wed, 23 Jul 2008 21:28:12 -0700
Subject: [PATCH 0287/1171] mm/page_alloc.c: cleanups

This patch contains the following cleanups:
- make the following needlessly global variables static:
  - required_kernelcore
  - zone_movable_pfn[]
- make the following needlessly global functions static:
  - move_freepages()
  - move_freepages_block()
  - setup_pageset()
  - find_usable_zone_for_movable()
  - adjust_zone_range_for_zone_movable()
  - __absent_pages_in_range()
  - find_min_pfn_for_node()
  - find_zone_movable_pfns_for_nodes()

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/page_alloc.c | 25 +++++++++++++------------
 1 file changed, 13 insertions(+), 12 deletions(-)

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 8d528d57b403..cd4c41432ef6 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -153,9 +153,9 @@ static unsigned long __meminitdata dma_reserve;
   static unsigned long __meminitdata node_boundary_start_pfn[MAX_NUMNODES];
   static unsigned long __meminitdata node_boundary_end_pfn[MAX_NUMNODES];
 #endif /* CONFIG_MEMORY_HOTPLUG_RESERVE */
-  unsigned long __initdata required_kernelcore;
+  static unsigned long __initdata required_kernelcore;
   static unsigned long __initdata required_movablecore;
-  unsigned long __meminitdata zone_movable_pfn[MAX_NUMNODES];
+  static unsigned long __meminitdata zone_movable_pfn[MAX_NUMNODES];
 
   /* movable_zone is the "real" zone pages in ZONE_MOVABLE are taken from */
   int movable_zone;
@@ -674,9 +674,9 @@ static int fallbacks[MIGRATE_TYPES][MIGRATE_TYPES-1] = {
  * Note that start_page and end_pages are not aligned on a pageblock
  * boundary. If alignment is required, use move_freepages_block()
  */
-int move_freepages(struct zone *zone,
-			struct page *start_page, struct page *end_page,
-			int migratetype)
+static int move_freepages(struct zone *zone,
+			  struct page *start_page, struct page *end_page,
+			  int migratetype)
 {
 	struct page *page;
 	unsigned long order;
@@ -715,7 +715,8 @@ int move_freepages(struct zone *zone,
 	return pages_moved;
 }
 
-int move_freepages_block(struct zone *zone, struct page *page, int migratetype)
+static int move_freepages_block(struct zone *zone, struct page *page,
+				int migratetype)
 {
 	unsigned long start_pfn, end_pfn;
 	struct page *start_page, *end_page;
@@ -2652,7 +2653,7 @@ static int zone_batchsize(struct zone *zone)
 	return batch;
 }
 
-inline void setup_pageset(struct per_cpu_pageset *p, unsigned long batch)
+static void setup_pageset(struct per_cpu_pageset *p, unsigned long batch)
 {
 	struct per_cpu_pages *pcp;
 
@@ -3099,7 +3100,7 @@ void __meminit get_pfn_range_for_nid(unsigned int nid,
  * assumption is made that zones within a node are ordered in monotonic
  * increasing memory addresses so that the "highest" populated zone is used
  */
-void __init find_usable_zone_for_movable(void)
+static void __init find_usable_zone_for_movable(void)
 {
 	int zone_index;
 	for (zone_index = MAX_NR_ZONES - 1; zone_index >= 0; zone_index--) {
@@ -3125,7 +3126,7 @@ void __init find_usable_zone_for_movable(void)
  * highest usable zone for ZONE_MOVABLE. This preserves the assumption that
  * zones within a node are in order of monotonic increases memory addresses
  */
-void __meminit adjust_zone_range_for_zone_movable(int nid,
+static void __meminit adjust_zone_range_for_zone_movable(int nid,
 					unsigned long zone_type,
 					unsigned long node_start_pfn,
 					unsigned long node_end_pfn,
@@ -3186,7 +3187,7 @@ static unsigned long __meminit zone_spanned_pages_in_node(int nid,
  * Return the number of holes in a range on a node. If nid is MAX_NUMNODES,
  * then all holes in the requested range will be accounted for.
  */
-unsigned long __meminit __absent_pages_in_range(int nid,
+static unsigned long __meminit __absent_pages_in_range(int nid,
 				unsigned long range_start_pfn,
 				unsigned long range_end_pfn)
 {
@@ -3723,7 +3724,7 @@ static void __init sort_node_map(void)
 }
 
 /* Find the lowest pfn for a node */
-unsigned long __init find_min_pfn_for_node(int nid)
+static unsigned long __init find_min_pfn_for_node(int nid)
 {
 	int i;
 	unsigned long min_pfn = ULONG_MAX;
@@ -3795,7 +3796,7 @@ static unsigned long __init early_calculate_totalpages(void)
  * memory. When they don't, some nodes will have more kernelcore than
  * others
  */
-void __init find_zone_movable_pfns_for_nodes(unsigned long *movable_pfn)
+static void __init find_zone_movable_pfns_for_nodes(unsigned long *movable_pfn)
 {
 	int i, nid;
 	unsigned long usable_startpfn;

From d92bc318547507a944a22e7ef936793dc0fe167f Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Wed, 23 Jul 2008 21:28:12 -0700
Subject: [PATCH 0288/1171] mm: make register_page_bootmem_info_section()
 static

Make the needlessly global register_page_bootmem_info_section() static.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Acked-by: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/memory_hotplug.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index 6e26adc08f14..ec85c37dcfb9 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -86,7 +86,7 @@ void put_page_bootmem(struct page *page)
 
 }
 
-void register_page_bootmem_info_section(unsigned long start_pfn)
+static void register_page_bootmem_info_section(unsigned long start_pfn)
 {
 	unsigned long *usemap, mapsize, section_nr, i;
 	struct mem_section *ms;

From 27ac792ca0b0a1e7e65f20342260650516c95864 Mon Sep 17 00:00:00 2001
From: Andrea Righi <righi.andrea@gmail.com>
Date: Wed, 23 Jul 2008 21:28:13 -0700
Subject: [PATCH 0289/1171] PAGE_ALIGN(): correctly handle 64-bit values on
 32-bit architectures

On 32-bit architectures PAGE_ALIGN() truncates 64-bit values to the 32-bit
boundary. For example:

	u64 val = PAGE_ALIGN(size);

always returns a value < 4GB even if size is greater than 4GB.

The problem resides in PAGE_MASK definition (from include/asm-x86/page.h for
example):

#define PAGE_SHIFT      12
#define PAGE_SIZE       (_AC(1,UL) << PAGE_SHIFT)
#define PAGE_MASK       (~(PAGE_SIZE-1))
...
#define PAGE_ALIGN(addr)       (((addr)+PAGE_SIZE-1)&PAGE_MASK)

The "~" is performed on a 32-bit value, so everything in "and" with
PAGE_MASK greater than 4GB will be truncated to the 32-bit boundary.
Using the ALIGN() macro seems to be the right way, because it uses
typeof(addr) for the mask.

Also move the PAGE_ALIGN() definitions out of include/asm-*/page.h in
include/linux/mm.h.

See also lkml discussion: http://lkml.org/lkml/2008/6/11/237

[akpm@linux-foundation.org: fix drivers/media/video/uvc/uvc_queue.c]
[akpm@linux-foundation.org: fix v850]
[akpm@linux-foundation.org: fix powerpc]
[akpm@linux-foundation.org: fix arm]
[akpm@linux-foundation.org: fix mips]
[akpm@linux-foundation.org: fix drivers/media/video/pvrusb2/pvrusb2-dvb.c]
[akpm@linux-foundation.org: fix drivers/mtd/maps/uclinux.c]
[akpm@linux-foundation.org: fix powerpc]
Signed-off-by: Andrea Righi <righi.andrea@gmail.com>
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/arm/kernel/module.c                     | 1 +
 arch/arm/plat-omap/fb.c                      | 1 +
 arch/avr32/mm/ioremap.c                      | 1 +
 arch/h8300/kernel/setup.c                    | 1 +
 arch/m68k/amiga/chipram.c                    | 1 +
 arch/m68knommu/kernel/setup.c                | 1 +
 arch/mips/kernel/module.c                    | 1 +
 arch/mips/sgi-ip27/ip27-klnuma.c             | 1 +
 arch/powerpc/kernel/suspend.c                | 1 +
 arch/powerpc/lib/code-patching.c             | 1 +
 arch/sparc64/kernel/iommu_common.h           | 2 +-
 arch/x86/kernel/module_64.c                  | 1 +
 arch/xtensa/kernel/setup.c                   | 1 +
 drivers/char/random.c                        | 1 +
 drivers/ieee1394/iso.c                       | 1 +
 drivers/media/video/pvrusb2/pvrusb2-dvb.c    | 1 +
 drivers/media/video/pvrusb2/pvrusb2-ioread.c | 1 +
 drivers/media/video/uvc/uvc_queue.c          | 1 +
 drivers/media/video/videobuf-core.c          | 1 +
 drivers/mtd/maps/uclinux.c                   | 1 +
 drivers/net/mlx4/eq.c                        | 1 +
 drivers/pcmcia/electra_cf.c                  | 1 +
 drivers/scsi/sun_esp.c                       | 1 +
 drivers/video/acornfb.c                      | 1 +
 drivers/video/imxfb.c                        | 1 +
 drivers/video/omap/dispc.c                   | 1 +
 drivers/video/omap/omapfb_main.c             | 1 +
 drivers/video/pxafb.c                        | 1 +
 drivers/video/sa1100fb.c                     | 1 +
 include/asm-alpha/page.h                     | 3 ---
 include/asm-arm/page-nommu.h                 | 4 +---
 include/asm-arm/page.h                       | 3 ---
 include/asm-avr32/page.h                     | 3 ---
 include/asm-blackfin/page.h                  | 3 ---
 include/asm-cris/page.h                      | 3 ---
 include/asm-frv/page.h                       | 3 ---
 include/asm-h8300/page.h                     | 3 ---
 include/asm-ia64/page.h                      | 1 -
 include/asm-m32r/page.h                      | 3 ---
 include/asm-m68k/dvma.h                      | 2 +-
 include/asm-m68k/page.h                      | 3 ---
 include/asm-m68knommu/page.h                 | 3 ---
 include/asm-mips/page.h                      | 3 ---
 include/asm-mips/processor.h                 | 2 +-
 include/asm-mn10300/page.h                   | 3 ---
 include/asm-parisc/page.h                    | 4 ----
 include/asm-powerpc/page.h                   | 3 ---
 include/asm-s390/page.h                      | 3 ---
 include/asm-sh/page.h                        | 3 ---
 include/asm-sparc/page_32.h                  | 3 ---
 include/asm-sparc/page_64.h                  | 3 ---
 include/asm-um/page.h                        | 3 ---
 include/asm-v850/page.h                      | 4 ----
 include/asm-x86/page.h                       | 3 ---
 include/asm-xtensa/page.h                    | 2 --
 include/linux/mm.h                           | 3 +++
 sound/core/info.c                            | 1 +
 57 files changed, 36 insertions(+), 74 deletions(-)

diff --git a/arch/arm/kernel/module.c b/arch/arm/kernel/module.c
index 79b7e5cf5416..a68259a0cccd 100644
--- a/arch/arm/kernel/module.c
+++ b/arch/arm/kernel/module.c
@@ -13,6 +13,7 @@
 #include <linux/module.h>
 #include <linux/moduleloader.h>
 #include <linux/kernel.h>
+#include <linux/mm.h>
 #include <linux/elf.h>
 #include <linux/vmalloc.h>
 #include <linux/slab.h>
diff --git a/arch/arm/plat-omap/fb.c b/arch/arm/plat-omap/fb.c
index 96d6f0619733..5d107520e6b9 100644
--- a/arch/arm/plat-omap/fb.c
+++ b/arch/arm/plat-omap/fb.c
@@ -23,6 +23,7 @@
 
 #include <linux/module.h>
 #include <linux/kernel.h>
+#include <linux/mm.h>
 #include <linux/init.h>
 #include <linux/platform_device.h>
 #include <linux/bootmem.h>
diff --git a/arch/avr32/mm/ioremap.c b/arch/avr32/mm/ioremap.c
index 3437c82434ac..f03b79f0e0ab 100644
--- a/arch/avr32/mm/ioremap.c
+++ b/arch/avr32/mm/ioremap.c
@@ -6,6 +6,7 @@
  * published by the Free Software Foundation.
  */
 #include <linux/vmalloc.h>
+#include <linux/mm.h>
 #include <linux/module.h>
 #include <linux/io.h>
 
diff --git a/arch/h8300/kernel/setup.c b/arch/h8300/kernel/setup.c
index b1f25c20a5db..7fda657110eb 100644
--- a/arch/h8300/kernel/setup.c
+++ b/arch/h8300/kernel/setup.c
@@ -20,6 +20,7 @@
 #include <linux/sched.h>
 #include <linux/delay.h>
 #include <linux/interrupt.h>
+#include <linux/mm.h>
 #include <linux/fs.h>
 #include <linux/fb.h>
 #include <linux/console.h>
diff --git a/arch/m68k/amiga/chipram.c b/arch/m68k/amiga/chipram.c
index cbe36538af47..61df1d33c050 100644
--- a/arch/m68k/amiga/chipram.c
+++ b/arch/m68k/amiga/chipram.c
@@ -9,6 +9,7 @@
 
 #include <linux/types.h>
 #include <linux/kernel.h>
+#include <linux/mm.h>
 #include <linux/init.h>
 #include <linux/ioport.h>
 #include <linux/slab.h>
diff --git a/arch/m68knommu/kernel/setup.c b/arch/m68knommu/kernel/setup.c
index 03f4fe6a2fc0..5985f1989021 100644
--- a/arch/m68knommu/kernel/setup.c
+++ b/arch/m68knommu/kernel/setup.c
@@ -22,6 +22,7 @@
 #include <linux/interrupt.h>
 #include <linux/fb.h>
 #include <linux/module.h>
+#include <linux/mm.h>
 #include <linux/console.h>
 #include <linux/errno.h>
 #include <linux/string.h>
diff --git a/arch/mips/kernel/module.c b/arch/mips/kernel/module.c
index e7ed0ac48537..1f60e27523d9 100644
--- a/arch/mips/kernel/module.c
+++ b/arch/mips/kernel/module.c
@@ -22,6 +22,7 @@
 
 #include <linux/moduleloader.h>
 #include <linux/elf.h>
+#include <linux/mm.h>
 #include <linux/vmalloc.h>
 #include <linux/slab.h>
 #include <linux/fs.h>
diff --git a/arch/mips/sgi-ip27/ip27-klnuma.c b/arch/mips/sgi-ip27/ip27-klnuma.c
index 48932ce1d730..d9c79d8be81d 100644
--- a/arch/mips/sgi-ip27/ip27-klnuma.c
+++ b/arch/mips/sgi-ip27/ip27-klnuma.c
@@ -4,6 +4,7 @@
  * Copyright 2000 - 2001 Kanoj Sarcar (kanoj@sgi.com)
  */
 #include <linux/init.h>
+#include <linux/mm.h>
 #include <linux/mmzone.h>
 #include <linux/kernel.h>
 #include <linux/nodemask.h>
diff --git a/arch/powerpc/kernel/suspend.c b/arch/powerpc/kernel/suspend.c
index 8cee57107541..6fc6328dc626 100644
--- a/arch/powerpc/kernel/suspend.c
+++ b/arch/powerpc/kernel/suspend.c
@@ -7,6 +7,7 @@
  * Copyright (c) 2001 Patrick Mochel <mochel@osdl.org>
  */
 
+#include <linux/mm.h>
 #include <asm/page.h>
 
 /* References to section boundaries */
diff --git a/arch/powerpc/lib/code-patching.c b/arch/powerpc/lib/code-patching.c
index 0559fe086eb4..7c975d43e3f3 100644
--- a/arch/powerpc/lib/code-patching.c
+++ b/arch/powerpc/lib/code-patching.c
@@ -10,6 +10,7 @@
 #include <linux/kernel.h>
 #include <linux/vmalloc.h>
 #include <linux/init.h>
+#include <linux/mm.h>
 #include <asm/page.h>
 #include <asm/code-patching.h>
 
diff --git a/arch/sparc64/kernel/iommu_common.h b/arch/sparc64/kernel/iommu_common.h
index f3575a614fa2..53b19c8231a9 100644
--- a/arch/sparc64/kernel/iommu_common.h
+++ b/arch/sparc64/kernel/iommu_common.h
@@ -23,7 +23,7 @@
 #define IO_PAGE_SHIFT			13
 #define IO_PAGE_SIZE			(1UL << IO_PAGE_SHIFT)
 #define IO_PAGE_MASK			(~(IO_PAGE_SIZE-1))
-#define IO_PAGE_ALIGN(addr)		(((addr)+IO_PAGE_SIZE-1)&IO_PAGE_MASK)
+#define IO_PAGE_ALIGN(addr)		ALIGN(addr, IO_PAGE_SIZE)
 
 #define IO_TSB_ENTRIES			(128*1024)
 #define IO_TSB_SIZE			(IO_TSB_ENTRIES * 8)
diff --git a/arch/x86/kernel/module_64.c b/arch/x86/kernel/module_64.c
index 0e867676b5a5..6ba87830d4b1 100644
--- a/arch/x86/kernel/module_64.c
+++ b/arch/x86/kernel/module_64.c
@@ -22,6 +22,7 @@
 #include <linux/fs.h>
 #include <linux/string.h>
 #include <linux/kernel.h>
+#include <linux/mm.h>
 #include <linux/slab.h>
 #include <linux/bug.h>
 
diff --git a/arch/xtensa/kernel/setup.c b/arch/xtensa/kernel/setup.c
index 5e6d75c9f92b..a00359e8f7a8 100644
--- a/arch/xtensa/kernel/setup.c
+++ b/arch/xtensa/kernel/setup.c
@@ -16,6 +16,7 @@
 
 #include <linux/errno.h>
 #include <linux/init.h>
+#include <linux/mm.h>
 #include <linux/proc_fs.h>
 #include <linux/screen_info.h>
 #include <linux/bootmem.h>
diff --git a/drivers/char/random.c b/drivers/char/random.c
index 0cf98bd4f2d2..e0d0e371909c 100644
--- a/drivers/char/random.c
+++ b/drivers/char/random.c
@@ -236,6 +236,7 @@
 #include <linux/fs.h>
 #include <linux/genhd.h>
 #include <linux/interrupt.h>
+#include <linux/mm.h>
 #include <linux/spinlock.h>
 #include <linux/percpu.h>
 #include <linux/cryptohash.h>
diff --git a/drivers/ieee1394/iso.c b/drivers/ieee1394/iso.c
index 07ca35c98f96..1cf6487b65ba 100644
--- a/drivers/ieee1394/iso.c
+++ b/drivers/ieee1394/iso.c
@@ -11,6 +11,7 @@
 
 #include <linux/pci.h>
 #include <linux/sched.h>
+#include <linux/mm.h>
 #include <linux/slab.h>
 
 #include "hosts.h"
diff --git a/drivers/media/video/pvrusb2/pvrusb2-dvb.c b/drivers/media/video/pvrusb2/pvrusb2-dvb.c
index 6ec4bf81fc7f..77b3c3385066 100644
--- a/drivers/media/video/pvrusb2/pvrusb2-dvb.c
+++ b/drivers/media/video/pvrusb2/pvrusb2-dvb.c
@@ -20,6 +20,7 @@
 
 #include <linux/kthread.h>
 #include <linux/freezer.h>
+#include <linux/mm.h>
 #include "dvbdev.h"
 #include "pvrusb2-debug.h"
 #include "pvrusb2-hdw-internal.h"
diff --git a/drivers/media/video/pvrusb2/pvrusb2-ioread.c b/drivers/media/video/pvrusb2/pvrusb2-ioread.c
index 05a1376405e7..b4824782d858 100644
--- a/drivers/media/video/pvrusb2/pvrusb2-ioread.c
+++ b/drivers/media/video/pvrusb2/pvrusb2-ioread.c
@@ -22,6 +22,7 @@
 #include "pvrusb2-debug.h"
 #include <linux/errno.h>
 #include <linux/string.h>
+#include <linux/mm.h>
 #include <linux/slab.h>
 #include <linux/mutex.h>
 #include <asm/uaccess.h>
diff --git a/drivers/media/video/uvc/uvc_queue.c b/drivers/media/video/uvc/uvc_queue.c
index 7388d0cee3d4..5646a6a32939 100644
--- a/drivers/media/video/uvc/uvc_queue.c
+++ b/drivers/media/video/uvc/uvc_queue.c
@@ -13,6 +13,7 @@
 
 #include <linux/kernel.h>
 #include <linux/version.h>
+#include <linux/mm.h>
 #include <linux/list.h>
 #include <linux/module.h>
 #include <linux/usb.h>
diff --git a/drivers/media/video/videobuf-core.c b/drivers/media/video/videobuf-core.c
index 0a88c44ace00..b7b05842cf28 100644
--- a/drivers/media/video/videobuf-core.c
+++ b/drivers/media/video/videobuf-core.c
@@ -16,6 +16,7 @@
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/moduleparam.h>
+#include <linux/mm.h>
 #include <linux/slab.h>
 #include <linux/interrupt.h>
 
diff --git a/drivers/mtd/maps/uclinux.c b/drivers/mtd/maps/uclinux.c
index c42f4b83f686..3fcf92130aa4 100644
--- a/drivers/mtd/maps/uclinux.c
+++ b/drivers/mtd/maps/uclinux.c
@@ -15,6 +15,7 @@
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/fs.h>
+#include <linux/mm.h>
 #include <linux/major.h>
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/map.h>
diff --git a/drivers/net/mlx4/eq.c b/drivers/net/mlx4/eq.c
index e141a1513f07..ea3a09aaa844 100644
--- a/drivers/net/mlx4/eq.c
+++ b/drivers/net/mlx4/eq.c
@@ -33,6 +33,7 @@
 
 #include <linux/init.h>
 #include <linux/interrupt.h>
+#include <linux/mm.h>
 #include <linux/dma-mapping.h>
 
 #include <linux/mlx4/cmd.h>
diff --git a/drivers/pcmcia/electra_cf.c b/drivers/pcmcia/electra_cf.c
index c21f9a9c3e3f..a34284b1482a 100644
--- a/drivers/pcmcia/electra_cf.c
+++ b/drivers/pcmcia/electra_cf.c
@@ -28,6 +28,7 @@
 #include <linux/init.h>
 #include <linux/delay.h>
 #include <linux/interrupt.h>
+#include <linux/mm.h>
 #include <linux/vmalloc.h>
 #include <linux/of_platform.h>
 
diff --git a/drivers/scsi/sun_esp.c b/drivers/scsi/sun_esp.c
index 2c87db98cdfb..f9cf70151366 100644
--- a/drivers/scsi/sun_esp.c
+++ b/drivers/scsi/sun_esp.c
@@ -7,6 +7,7 @@
 #include <linux/types.h>
 #include <linux/delay.h>
 #include <linux/module.h>
+#include <linux/mm.h>
 #include <linux/init.h>
 
 #include <asm/irq.h>
diff --git a/drivers/video/acornfb.c b/drivers/video/acornfb.c
index eedb8285e32f..017233d0c481 100644
--- a/drivers/video/acornfb.c
+++ b/drivers/video/acornfb.c
@@ -23,6 +23,7 @@
 #include <linux/string.h>
 #include <linux/ctype.h>
 #include <linux/slab.h>
+#include <linux/mm.h>
 #include <linux/init.h>
 #include <linux/fb.h>
 #include <linux/platform_device.h>
diff --git a/drivers/video/imxfb.c b/drivers/video/imxfb.c
index 94e4d3ac1a05..0c5a475c1cae 100644
--- a/drivers/video/imxfb.c
+++ b/drivers/video/imxfb.c
@@ -24,6 +24,7 @@
 #include <linux/string.h>
 #include <linux/interrupt.h>
 #include <linux/slab.h>
+#include <linux/mm.h>
 #include <linux/fb.h>
 #include <linux/delay.h>
 #include <linux/init.h>
diff --git a/drivers/video/omap/dispc.c b/drivers/video/omap/dispc.c
index ab32ceb06178..ab77c51fe9d6 100644
--- a/drivers/video/omap/dispc.c
+++ b/drivers/video/omap/dispc.c
@@ -20,6 +20,7 @@
  */
 #include <linux/kernel.h>
 #include <linux/dma-mapping.h>
+#include <linux/mm.h>
 #include <linux/vmalloc.h>
 #include <linux/clk.h>
 #include <linux/io.h>
diff --git a/drivers/video/omap/omapfb_main.c b/drivers/video/omap/omapfb_main.c
index 14d0f7a11145..f85af5c4fa68 100644
--- a/drivers/video/omap/omapfb_main.c
+++ b/drivers/video/omap/omapfb_main.c
@@ -25,6 +25,7 @@
  * 59 Temple Place - Suite 330, Boston, MA  02111-1307, USA.
  */
 #include <linux/platform_device.h>
+#include <linux/mm.h>
 #include <linux/uaccess.h>
 
 #include <asm/mach-types.h>
diff --git a/drivers/video/pxafb.c b/drivers/video/pxafb.c
index bb2514369507..5e8a140399fc 100644
--- a/drivers/video/pxafb.c
+++ b/drivers/video/pxafb.c
@@ -30,6 +30,7 @@
 #include <linux/string.h>
 #include <linux/interrupt.h>
 #include <linux/slab.h>
+#include <linux/mm.h>
 #include <linux/fb.h>
 #include <linux/delay.h>
 #include <linux/init.h>
diff --git a/drivers/video/sa1100fb.c b/drivers/video/sa1100fb.c
index ab2b2110478b..4a9f7e121807 100644
--- a/drivers/video/sa1100fb.c
+++ b/drivers/video/sa1100fb.c
@@ -167,6 +167,7 @@
 #include <linux/string.h>
 #include <linux/interrupt.h>
 #include <linux/slab.h>
+#include <linux/mm.h>
 #include <linux/fb.h>
 #include <linux/delay.h>
 #include <linux/init.h>
diff --git a/include/asm-alpha/page.h b/include/asm-alpha/page.h
index 22ff9762d17b..0995f9d13417 100644
--- a/include/asm-alpha/page.h
+++ b/include/asm-alpha/page.h
@@ -80,9 +80,6 @@ typedef struct page *pgtable_t;
 
 #endif /* !__ASSEMBLY__ */
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
-
 #define __pa(x)			((unsigned long) (x) - PAGE_OFFSET)
 #define __va(x)			((void *)((unsigned long) (x) + PAGE_OFFSET))
 #ifndef CONFIG_DISCONTIGMEM
diff --git a/include/asm-arm/page-nommu.h b/include/asm-arm/page-nommu.h
index a1bcad060480..ea1cde84f500 100644
--- a/include/asm-arm/page-nommu.h
+++ b/include/asm-arm/page-nommu.h
@@ -7,6 +7,7 @@
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
  */
+
 #ifndef _ASMARM_PAGE_NOMMU_H
 #define _ASMARM_PAGE_NOMMU_H
 
@@ -42,9 +43,6 @@ typedef unsigned long pgprot_t;
 #define __pmd(x)        (x)
 #define __pgprot(x)     (x)
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
-
 extern unsigned long memory_start;
 extern unsigned long memory_end;
 
diff --git a/include/asm-arm/page.h b/include/asm-arm/page.h
index 8e05bdb5f12f..7c5fc5582e5d 100644
--- a/include/asm-arm/page.h
+++ b/include/asm-arm/page.h
@@ -15,9 +15,6 @@
 #define PAGE_SIZE		(1UL << PAGE_SHIFT)
 #define PAGE_MASK		(~(PAGE_SIZE-1))
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
-
 #ifndef __ASSEMBLY__
 
 #ifndef CONFIG_MMU
diff --git a/include/asm-avr32/page.h b/include/asm-avr32/page.h
index cbbc5ca9728b..f805d1cb11bc 100644
--- a/include/asm-avr32/page.h
+++ b/include/asm-avr32/page.h
@@ -57,9 +57,6 @@ static inline int get_order(unsigned long size)
 
 #endif /* !__ASSEMBLY__ */
 
-/* Align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr) + PAGE_SIZE - 1) & PAGE_MASK)
-
 /*
  * The hardware maps the virtual addresses 0x80000000 -> 0x9fffffff
  * permanently to the physical addresses 0x00000000 -> 0x1fffffff when
diff --git a/include/asm-blackfin/page.h b/include/asm-blackfin/page.h
index c7db0220fbd6..344f6a8c1f22 100644
--- a/include/asm-blackfin/page.h
+++ b/include/asm-blackfin/page.h
@@ -51,9 +51,6 @@ typedef struct page *pgtable_t;
 #define __pgd(x)	((pgd_t) { (x) } )
 #define __pgprot(x)	((pgprot_t) { (x) } )
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
-
 extern unsigned long memory_start;
 extern unsigned long memory_end;
 
diff --git a/include/asm-cris/page.h b/include/asm-cris/page.h
index c45bb1ef397c..d19272ba6b69 100644
--- a/include/asm-cris/page.h
+++ b/include/asm-cris/page.h
@@ -60,9 +60,6 @@ typedef struct page *pgtable_t;
 
 #define page_to_phys(page)     __pa((((page) - mem_map) << PAGE_SHIFT) + PAGE_OFFSET)
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
-
 #ifndef __ASSEMBLY__
 
 #endif /* __ASSEMBLY__ */
diff --git a/include/asm-frv/page.h b/include/asm-frv/page.h
index c2c1e89e747d..bd9c220094c7 100644
--- a/include/asm-frv/page.h
+++ b/include/asm-frv/page.h
@@ -40,9 +40,6 @@ typedef struct page *pgtable_t;
 #define __pgprot(x)	((pgprot_t) { (x) } )
 #define PTE_MASK	PAGE_MASK
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr) + PAGE_SIZE - 1) & PAGE_MASK)
-
 #define devmem_is_allowed(pfn)	1
 
 #define __pa(vaddr)		virt_to_phys((void *) (unsigned long) (vaddr))
diff --git a/include/asm-h8300/page.h b/include/asm-h8300/page.h
index d6a3eaf3b27e..0b6acf0b03aa 100644
--- a/include/asm-h8300/page.h
+++ b/include/asm-h8300/page.h
@@ -43,9 +43,6 @@ typedef struct page *pgtable_t;
 #define __pgd(x)	((pgd_t) { (x) } )
 #define __pgprot(x)	((pgprot_t) { (x) } )
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
-
 extern unsigned long memory_start;
 extern unsigned long memory_end;
 
diff --git a/include/asm-ia64/page.h b/include/asm-ia64/page.h
index 36f39321b768..5f271bc712ee 100644
--- a/include/asm-ia64/page.h
+++ b/include/asm-ia64/page.h
@@ -40,7 +40,6 @@
 
 #define PAGE_SIZE		(__IA64_UL_CONST(1) << PAGE_SHIFT)
 #define PAGE_MASK		(~(PAGE_SIZE - 1))
-#define PAGE_ALIGN(addr)	(((addr) + PAGE_SIZE - 1) & PAGE_MASK)
 
 #define PERCPU_PAGE_SHIFT	16	/* log2() of max. size of per-CPU area */
 #define PERCPU_PAGE_SIZE	(__IA64_UL_CONST(1) << PERCPU_PAGE_SHIFT)
diff --git a/include/asm-m32r/page.h b/include/asm-m32r/page.h
index 8a677f3fca68..c9333089fe11 100644
--- a/include/asm-m32r/page.h
+++ b/include/asm-m32r/page.h
@@ -41,9 +41,6 @@ typedef struct page *pgtable_t;
 
 #endif /* !__ASSEMBLY__ */
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr) + PAGE_SIZE - 1) & PAGE_MASK)
-
 /*
  * This handles the memory map.. We could make this a config
  * option, but too many people screw it up, and too few need
diff --git a/include/asm-m68k/dvma.h b/include/asm-m68k/dvma.h
index 4fff408d0150..890bbf7e7758 100644
--- a/include/asm-m68k/dvma.h
+++ b/include/asm-m68k/dvma.h
@@ -13,7 +13,7 @@
 #define DVMA_PAGE_SHIFT	13
 #define DVMA_PAGE_SIZE	(1UL << DVMA_PAGE_SHIFT)
 #define DVMA_PAGE_MASK	(~(DVMA_PAGE_SIZE-1))
-#define DVMA_PAGE_ALIGN(addr)	(((addr)+DVMA_PAGE_SIZE-1)&DVMA_PAGE_MASK)
+#define DVMA_PAGE_ALIGN(addr)	ALIGN(addr, DVMA_PAGE_SIZE)
 
 extern void dvma_init(void);
 extern int dvma_map_iommu(unsigned long kaddr, unsigned long baddr,
diff --git a/include/asm-m68k/page.h b/include/asm-m68k/page.h
index 880c2cbff8a6..a34b8bad7847 100644
--- a/include/asm-m68k/page.h
+++ b/include/asm-m68k/page.h
@@ -103,9 +103,6 @@ typedef struct page *pgtable_t;
 #define __pgd(x)	((pgd_t) { (x) } )
 #define __pgprot(x)	((pgprot_t) { (x) } )
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
-
 #endif /* !__ASSEMBLY__ */
 
 #include <asm/page_offset.h>
diff --git a/include/asm-m68knommu/page.h b/include/asm-m68knommu/page.h
index 1e82ebb7d644..3a1ede4544cb 100644
--- a/include/asm-m68knommu/page.h
+++ b/include/asm-m68knommu/page.h
@@ -43,9 +43,6 @@ typedef struct page *pgtable_t;
 #define __pgd(x)	((pgd_t) { (x) } )
 #define __pgprot(x)	((pgprot_t) { (x) } )
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
-
 extern unsigned long memory_start;
 extern unsigned long memory_end;
 
diff --git a/include/asm-mips/page.h b/include/asm-mips/page.h
index 494f00ba9541..fe7a88ea066e 100644
--- a/include/asm-mips/page.h
+++ b/include/asm-mips/page.h
@@ -137,9 +137,6 @@ typedef struct { unsigned long pgprot; } pgprot_t;
 
 #endif /* !__ASSEMBLY__ */
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr) + PAGE_SIZE - 1) & PAGE_MASK)
-
 /*
  * __pa()/__va() should be used only during mem init.
  */
diff --git a/include/asm-mips/processor.h b/include/asm-mips/processor.h
index 58cbac5a64e4..a1e4453469f9 100644
--- a/include/asm-mips/processor.h
+++ b/include/asm-mips/processor.h
@@ -45,7 +45,7 @@ extern unsigned int vced_count, vcei_count;
  * This decides where the kernel will search for a free chunk of vm
  * space during mmap's.
  */
-#define TASK_UNMAPPED_BASE	(PAGE_ALIGN(TASK_SIZE / 3))
+#define TASK_UNMAPPED_BASE	((TASK_SIZE / 3) & ~(PAGE_SIZE))
 #endif
 
 #ifdef CONFIG_64BIT
diff --git a/include/asm-mn10300/page.h b/include/asm-mn10300/page.h
index 124971b9fb9b..8288e124165b 100644
--- a/include/asm-mn10300/page.h
+++ b/include/asm-mn10300/page.h
@@ -61,9 +61,6 @@ typedef struct page *pgtable_t;
 
 #endif /* !__ASSEMBLY__ */
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr) + PAGE_SIZE - 1) & PAGE_MASK)
-
 /*
  * This handles the memory map.. We could make this a config
  * option, but too many people screw it up, and too few need
diff --git a/include/asm-parisc/page.h b/include/asm-parisc/page.h
index 27d50b859541..c3941f09a878 100644
--- a/include/asm-parisc/page.h
+++ b/include/asm-parisc/page.h
@@ -119,10 +119,6 @@ extern int npmem_ranges;
 #define PMD_ENTRY_SIZE	(1UL << BITS_PER_PMD_ENTRY)
 #define PTE_ENTRY_SIZE	(1UL << BITS_PER_PTE_ENTRY)
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
-
-
 #define LINUX_GATEWAY_SPACE     0
 
 /* This governs the relationship between virtual and physical addresses.
diff --git a/include/asm-powerpc/page.h b/include/asm-powerpc/page.h
index cffdf0eb0df6..e088545cb3f5 100644
--- a/include/asm-powerpc/page.h
+++ b/include/asm-powerpc/page.h
@@ -119,9 +119,6 @@ extern phys_addr_t kernstart_addr;
 /* align addr on a size boundary - adjust address up if needed */
 #define _ALIGN(addr,size)     _ALIGN_UP(addr,size)
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	_ALIGN(addr, PAGE_SIZE)
-
 /*
  * Don't compare things with KERNELBASE or PAGE_OFFSET to test for
  * "kernelness", use is_kernel_addr() - it should do what you want.
diff --git a/include/asm-s390/page.h b/include/asm-s390/page.h
index 12fd9c4f0f15..991ba939408c 100644
--- a/include/asm-s390/page.h
+++ b/include/asm-s390/page.h
@@ -138,9 +138,6 @@ void arch_alloc_page(struct page *page, int order);
 
 #endif /* !__ASSEMBLY__ */
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)        (((addr)+PAGE_SIZE-1)&PAGE_MASK)
-
 #define __PAGE_OFFSET           0x0UL
 #define PAGE_OFFSET             0x0UL
 #define __pa(x)                 (unsigned long)(x)
diff --git a/include/asm-sh/page.h b/include/asm-sh/page.h
index 304c30b5d947..5dc01d2fcc4c 100644
--- a/include/asm-sh/page.h
+++ b/include/asm-sh/page.h
@@ -22,9 +22,6 @@
 #define PAGE_MASK	(~(PAGE_SIZE-1))
 #define PTE_MASK	PAGE_MASK
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
-
 #if defined(CONFIG_HUGETLB_PAGE_SIZE_64K)
 #define HPAGE_SHIFT	16
 #elif defined(CONFIG_HUGETLB_PAGE_SIZE_256K)
diff --git a/include/asm-sparc/page_32.h b/include/asm-sparc/page_32.h
index 14de518cc38f..cf5fb70ca1c1 100644
--- a/include/asm-sparc/page_32.h
+++ b/include/asm-sparc/page_32.h
@@ -134,9 +134,6 @@ BTFIXUPDEF_SETHI(sparc_unmapped_base)
 
 #endif /* !(__ASSEMBLY__) */
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)  (((addr)+PAGE_SIZE-1)&PAGE_MASK)
-
 #define PAGE_OFFSET	0xf0000000
 #ifndef __ASSEMBLY__
 extern unsigned long phys_base;
diff --git a/include/asm-sparc/page_64.h b/include/asm-sparc/page_64.h
index a8a2bba032c1..b579b910ef51 100644
--- a/include/asm-sparc/page_64.h
+++ b/include/asm-sparc/page_64.h
@@ -106,9 +106,6 @@ typedef struct page *pgtable_t;
 
 #endif /* !(__ASSEMBLY__) */
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
-
 /* We used to stick this into a hard-coded global register (%g4)
  * but that does not make sense anymore.
  */
diff --git a/include/asm-um/page.h b/include/asm-um/page.h
index 916e1a61999f..335c57383c02 100644
--- a/include/asm-um/page.h
+++ b/include/asm-um/page.h
@@ -92,9 +92,6 @@ typedef struct page *pgtable_t;
 #define __pgd(x) ((pgd_t) { (x) } )
 #define __pgprot(x)	((pgprot_t) { (x) } )
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
-
 extern unsigned long uml_physmem;
 
 #define PAGE_OFFSET (uml_physmem)
diff --git a/include/asm-v850/page.h b/include/asm-v850/page.h
index 74a539a9bd59..f9de35d873fa 100644
--- a/include/asm-v850/page.h
+++ b/include/asm-v850/page.h
@@ -94,10 +94,6 @@ typedef unsigned long pgprot_t;
 #endif /* !__ASSEMBLY__ */
 
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr) + PAGE_SIZE - 1) & PAGE_MASK)
-
-
 /* No current v850 processor has virtual memory.  */
 #define __virt_to_phys(addr)	(addr)
 #define __phys_to_virt(addr)	(addr)
diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h
index 6e02098b1605..49982110e4d9 100644
--- a/include/asm-x86/page.h
+++ b/include/asm-x86/page.h
@@ -34,9 +34,6 @@
 
 #define HUGE_MAX_HSTATE 2
 
-/* to align the pointer to the (next) page boundary */
-#define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE-1)&PAGE_MASK)
-
 #ifndef __ASSEMBLY__
 #include <linux/types.h>
 #endif
diff --git a/include/asm-xtensa/page.h b/include/asm-xtensa/page.h
index 80a6ae0dd259..11f7dc2dbec7 100644
--- a/include/asm-xtensa/page.h
+++ b/include/asm-xtensa/page.h
@@ -26,13 +26,11 @@
 
 /*
  * PAGE_SHIFT determines the page size
- * PAGE_ALIGN(x) aligns the pointer to the (next) page boundary
  */
 
 #define PAGE_SHIFT		12
 #define PAGE_SIZE		(__XTENSA_UL_CONST(1) << PAGE_SHIFT)
 #define PAGE_MASK		(~(PAGE_SIZE-1))
-#define PAGE_ALIGN(addr)	(((addr)+PAGE_SIZE - 1) & PAGE_MASK)
 
 #define PAGE_OFFSET		XCHAL_KSEG_CACHED_VADDR
 #define MAX_MEM_PFN		XCHAL_KSEG_SIZE
diff --git a/include/linux/mm.h b/include/linux/mm.h
index df322fb4df31..d87a5a5fe87d 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -41,6 +41,9 @@ extern unsigned long mmap_min_addr;
 
 #define nth_page(page,n) pfn_to_page(page_to_pfn((page)) + (n))
 
+/* to align the pointer to the (next) page boundary */
+#define PAGE_ALIGN(addr) ALIGN(addr, PAGE_SIZE)
+
 /*
  * Linux kernel virtual memory manager primitives.
  * The idea being to have a "virtual" mm in the same way
diff --git a/sound/core/info.c b/sound/core/info.c
index cb5ead3e202d..c67773ad9298 100644
--- a/sound/core/info.c
+++ b/sound/core/info.c
@@ -21,6 +21,7 @@
 
 #include <linux/init.h>
 #include <linux/time.h>
+#include <linux/mm.h>
 #include <linux/smp_lock.h>
 #include <linux/string.h>
 #include <sound/core.h>

From f84f9504bddeec33a72d64ebe95143d3aaeb3f9b Mon Sep 17 00:00:00 2001
From: Vegard Nossum <vegard.nossum@gmail.com>
Date: Wed, 23 Jul 2008 21:28:14 -0700
Subject: [PATCH 0290/1171] mm: remove initialization of static per-cpu
 variables

This was required by some old, no-longer-used gcc on sparc.

Signed-off-by: Vegard Nossum <vegard.nossum@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/swap.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/mm/swap.c b/mm/swap.c
index 45c9f25a8a3b..dd89234ee51f 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -34,9 +34,9 @@
 /* How many pages do we try to swap or page in/out together? */
 int page_cluster;
 
-static DEFINE_PER_CPU(struct pagevec, lru_add_pvecs) = { 0, };
-static DEFINE_PER_CPU(struct pagevec, lru_add_active_pvecs) = { 0, };
-static DEFINE_PER_CPU(struct pagevec, lru_rotate_pvecs) = { 0, };
+static DEFINE_PER_CPU(struct pagevec, lru_add_pvecs);
+static DEFINE_PER_CPU(struct pagevec, lru_add_active_pvecs);
+static DEFINE_PER_CPU(struct pagevec, lru_rotate_pvecs);
 
 /*
  * This path almost never happens for VM activity - pages are normally
@@ -493,7 +493,7 @@ EXPORT_SYMBOL(pagevec_lookup_tag);
  */
 #define ACCT_THRESHOLD	max(16, NR_CPUS * 2)
 
-static DEFINE_PER_CPU(long, committed_space) = 0;
+static DEFINE_PER_CPU(long, committed_space);
 
 void vm_acct_memory(long pages)
 {

From 48c906823f3927b981db9f0b03c2e2499977ee93 Mon Sep 17 00:00:00 2001
From: Yasunori Goto <y-goto@jp.fujitsu.com>
Date: Wed, 23 Jul 2008 21:28:15 -0700
Subject: [PATCH 0291/1171] memory hotplug: allocate usemap on the section with
 pgdat

Usemaps are allocated on the section which has pgdat by this.

Because usemap size is very small, many other sections usemaps are
allocated on only one page.  If a section has usemap, it can't be removed
until removing other sections.  This dependency is not desirable for
memory removing.

Pgdat has similar feature.  When a section has pgdat area, it must be the
last section for removing on the node.  So, if section A has pgdat and
section B has usemap for section A, Both sections can't be removed due to
dependency each other.

To solve this issue, this patch collects usemap on same section with pgdat
as much as possible.  If other sections doesn't have any dependency, this
section will be able to be removed finally.

Signed-off-by: Yasunori Goto <y-goto@jp.fujitsu.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: David Miller <davem@davemloft.net>
Cc: Badari Pulavarty <pbadari@us.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Hiroyuki KAMEZAWA <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Tony Breeds <tony@bakeyournoodle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/sparse.c | 78 ++++++++++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 77 insertions(+), 1 deletion(-)

diff --git a/mm/sparse.c b/mm/sparse.c
index 7a3650923d9a..8ffc08990008 100644
--- a/mm/sparse.c
+++ b/mm/sparse.c
@@ -269,16 +269,92 @@ static unsigned long *__kmalloc_section_usemap(void)
 }
 #endif /* CONFIG_MEMORY_HOTPLUG */
 
+#ifdef CONFIG_MEMORY_HOTREMOVE
+static unsigned long * __init
+sparse_early_usemap_alloc_pgdat_section(struct pglist_data *pgdat)
+{
+	unsigned long section_nr;
+
+	/*
+	 * A page may contain usemaps for other sections preventing the
+	 * page being freed and making a section unremovable while
+	 * other sections referencing the usemap retmain active. Similarly,
+	 * a pgdat can prevent a section being removed. If section A
+	 * contains a pgdat and section B contains the usemap, both
+	 * sections become inter-dependent. This allocates usemaps
+	 * from the same section as the pgdat where possible to avoid
+	 * this problem.
+	 */
+	section_nr = pfn_to_section_nr(__pa(pgdat) >> PAGE_SHIFT);
+	return alloc_bootmem_section(usemap_size(), section_nr);
+}
+
+static void __init check_usemap_section_nr(int nid, unsigned long *usemap)
+{
+	unsigned long usemap_snr, pgdat_snr;
+	static unsigned long old_usemap_snr = NR_MEM_SECTIONS;
+	static unsigned long old_pgdat_snr = NR_MEM_SECTIONS;
+	struct pglist_data *pgdat = NODE_DATA(nid);
+	int usemap_nid;
+
+	usemap_snr = pfn_to_section_nr(__pa(usemap) >> PAGE_SHIFT);
+	pgdat_snr = pfn_to_section_nr(__pa(pgdat) >> PAGE_SHIFT);
+	if (usemap_snr == pgdat_snr)
+		return;
+
+	if (old_usemap_snr == usemap_snr && old_pgdat_snr == pgdat_snr)
+		/* skip redundant message */
+		return;
+
+	old_usemap_snr = usemap_snr;
+	old_pgdat_snr = pgdat_snr;
+
+	usemap_nid = sparse_early_nid(__nr_to_section(usemap_snr));
+	if (usemap_nid != nid) {
+		printk(KERN_INFO
+		       "node %d must be removed before remove section %ld\n",
+		       nid, usemap_snr);
+		return;
+	}
+	/*
+	 * There is a circular dependency.
+	 * Some platforms allow un-removable section because they will just
+	 * gather other removable sections for dynamic partitioning.
+	 * Just notify un-removable section's number here.
+	 */
+	printk(KERN_INFO "Section %ld and %ld (node %d)", usemap_snr,
+	       pgdat_snr, nid);
+	printk(KERN_CONT
+	       " have a circular dependency on usemap and pgdat allocations\n");
+}
+#else
+static unsigned long * __init
+sparse_early_usemap_alloc_pgdat_section(struct pglist_data *pgdat)
+{
+	return NULL;
+}
+
+static void __init check_usemap_section_nr(int nid, unsigned long *usemap)
+{
+}
+#endif /* CONFIG_MEMORY_HOTREMOVE */
+
 static unsigned long *__init sparse_early_usemap_alloc(unsigned long pnum)
 {
 	unsigned long *usemap;
 	struct mem_section *ms = __nr_to_section(pnum);
 	int nid = sparse_early_nid(ms);
 
-	usemap = alloc_bootmem_node(NODE_DATA(nid), usemap_size());
+	usemap = sparse_early_usemap_alloc_pgdat_section(NODE_DATA(nid));
 	if (usemap)
 		return usemap;
 
+	usemap = alloc_bootmem_node(NODE_DATA(nid), usemap_size());
+	if (usemap) {
+		check_usemap_section_nr(nid, usemap);
+		return usemap;
+	}
+
 	/* Stupid: suppress gcc warning for SPARSEMEM && !NUMA */
 	nid = 0;
 

From af370fb8cb3031f20438f246798d5f0d98089f29 Mon Sep 17 00:00:00 2001
From: Yasunori Goto <y-goto@jp.fujitsu.com>
Date: Wed, 23 Jul 2008 21:28:17 -0700
Subject: [PATCH 0292/1171] memory hotplug: small fixes to bootmem freeing for
 memory hotremove

- Change some naming
  * Magic -> types
  * MIX_INFO -> MIX_SECTION_INFO
  * Change definition of bootmem type from direct hex value

- __free_pages_bootmem() becomes __meminit.

Signed-off-by: Yasunori Goto <y-goto@jp.fujitsu.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Badari Pulavarty <pbadari@us.ibm.com>
Cc: Yinghai Lu <yhlu.kernel@gmail.com>
Cc: Johannes Weiner <hannes@saeurebad.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/memory_hotplug.h |  8 ++++----
 mm/memory_hotplug.c            | 12 ++++++------
 mm/page_alloc.c                |  2 +-
 3 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/include/linux/memory_hotplug.h b/include/linux/memory_hotplug.h
index ea9f5ad9ec8e..3628e5088f64 100644
--- a/include/linux/memory_hotplug.h
+++ b/include/linux/memory_hotplug.h
@@ -13,12 +13,12 @@ struct mem_section;
 #ifdef CONFIG_MEMORY_HOTPLUG
 
 /*
- * Magic number for free bootmem.
+ * Types for free bootmem.
  * The normal smallest mapcount is -1. Here is smaller value than it.
  */
-#define SECTION_INFO		0xfffffffe
-#define MIX_INFO		0xfffffffd
-#define NODE_INFO		0xfffffffc
+#define SECTION_INFO		(-1 - 1)
+#define MIX_SECTION_INFO	(-1 - 2)
+#define NODE_INFO		(-1 - 3)
 
 /*
  * pgdat resizing functions
diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index ec85c37dcfb9..0fb05b258f0c 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -62,9 +62,9 @@ static void release_memory_resource(struct resource *res)
 
 #ifdef CONFIG_MEMORY_HOTPLUG_SPARSE
 #ifndef CONFIG_SPARSEMEM_VMEMMAP
-static void get_page_bootmem(unsigned long info,  struct page *page, int magic)
+static void get_page_bootmem(unsigned long info,  struct page *page, int type)
 {
-	atomic_set(&page->_mapcount, magic);
+	atomic_set(&page->_mapcount, type);
 	SetPagePrivate(page);
 	set_page_private(page, info);
 	atomic_inc(&page->_count);
@@ -72,10 +72,10 @@ static void get_page_bootmem(unsigned long info,  struct page *page, int magic)
 
 void put_page_bootmem(struct page *page)
 {
-	int magic;
+	int type;
 
-	magic = atomic_read(&page->_mapcount);
-	BUG_ON(magic >= -1);
+	type = atomic_read(&page->_mapcount);
+	BUG_ON(type >= -1);
 
 	if (atomic_dec_return(&page->_count) == 1) {
 		ClearPagePrivate(page);
@@ -119,7 +119,7 @@ static void register_page_bootmem_info_section(unsigned long start_pfn)
 	mapsize = PAGE_ALIGN(usemap_size()) >> PAGE_SHIFT;
 
 	for (i = 0; i < mapsize; i++, page++)
-		get_page_bootmem(section_nr, page, MIX_INFO);
+		get_page_bootmem(section_nr, page, MIX_SECTION_INFO);
 
 }
 
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index cd4c41432ef6..6da667274df5 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -533,7 +533,7 @@ static void __free_pages_ok(struct page *page, unsigned int order)
 /*
  * permit the bootmem allocator to evade page validation on high-order frees
  */
-void __free_pages_bootmem(struct page *page, unsigned int order)
+void __meminit __free_pages_bootmem(struct page *page, unsigned int order)
 {
 	if (order == 0) {
 		__ClearPageReserved(page);

From 2f7f24eca31c4fc2fdb134b2ef743ccd67cfb9a9 Mon Sep 17 00:00:00 2001
From: Kent Liu <kent.liu@linux.intel.com>
Date: Wed, 23 Jul 2008 21:28:18 -0700
Subject: [PATCH 0293/1171] memory-hotplug: don't calculate vm_total_pages
 twice when rebuilding zonelists in online_pages()

If zonelist is required to be rebuilt in online_pages(), there is no need
to recalculate vm_total_pages in that function, as it has been updated in
the call build_all_zonelists().

Signed-off-by: Kent Liu <kent.liu@linux.intel.com>
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/memory_hotplug.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index 0fb05b258f0c..93aba78dc8b6 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -429,7 +429,9 @@ int online_pages(unsigned long pfn, unsigned long nr_pages)
 
 	if (need_zonelists_rebuild)
 		build_all_zonelists();
-	vm_total_pages = nr_free_pagecache_pages();
+	else
+		vm_total_pages = nr_free_pagecache_pages();
+
 	writeback_set_ratelimit();
 
 	if (onlined_pages)

From 5c755e9fd813810680abd56ec09a5f90143e815b Mon Sep 17 00:00:00 2001
From: Badari Pulavarty <pbadari@us.ibm.com>
Date: Wed, 23 Jul 2008 21:28:19 -0700
Subject: [PATCH 0294/1171] memory-hotplug: add sysfs removable attribute for
 hotplug memory remove

Memory may be hot-removed on a per-memory-block basis, particularly on
POWER where the SPARSEMEM section size often matches the memory-block
size.  A user-level agent must be able to identify which sections of
memory are likely to be removable before attempting the potentially
expensive operation.  This patch adds a file called "removable" to the
memory directory in sysfs to help such an agent.  In this patch, a memory
block is considered removable if;

o It contains only MOVABLE pageblocks
o It contains only pageblocks with free pages regardless of pageblock type

On the other hand, a memory block starting with a PageReserved() page will
never be considered removable.  Without this patch, the user-agent is
forced to choose a memory block to remove randomly.

Sample output of the sysfs files:

./memory/memory0/removable: 0
./memory/memory1/removable: 0
./memory/memory2/removable: 0
./memory/memory3/removable: 0
./memory/memory4/removable: 0
./memory/memory5/removable: 0
./memory/memory6/removable: 0
./memory/memory7/removable: 1
./memory/memory8/removable: 0
./memory/memory9/removable: 0
./memory/memory10/removable: 0
./memory/memory11/removable: 0
./memory/memory12/removable: 0
./memory/memory13/removable: 0
./memory/memory14/removable: 0
./memory/memory15/removable: 0
./memory/memory16/removable: 0
./memory/memory17/removable: 1
./memory/memory18/removable: 1
./memory/memory19/removable: 1
./memory/memory20/removable: 1
./memory/memory21/removable: 1
./memory/memory22/removable: 1

Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com>
Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 .../ABI/testing/sysfs-devices-memory          | 24 ++++++++
 drivers/base/memory.c                         | 19 ++++++
 include/linux/memory_hotplug.h                | 12 ++++
 mm/memory_hotplug.c                           | 60 +++++++++++++++++++
 4 files changed, 115 insertions(+)
 create mode 100644 Documentation/ABI/testing/sysfs-devices-memory

diff --git a/Documentation/ABI/testing/sysfs-devices-memory b/Documentation/ABI/testing/sysfs-devices-memory
new file mode 100644
index 000000000000..7a16fe1e2270
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-devices-memory
@@ -0,0 +1,24 @@
+What:		/sys/devices/system/memory
+Date:		June 2008
+Contact:	Badari Pulavarty <pbadari@us.ibm.com>
+Description:
+		The /sys/devices/system/memory contains a snapshot of the
+		internal state of the kernel memory blocks. Files could be
+		added or removed dynamically to represent hot-add/remove
+		operations.
+
+Users:		hotplug memory add/remove tools
+		https://w3.opensource.ibm.com/projects/powerpc-utils/
+
+What:		/sys/devices/system/memory/memoryX/removable
+Date:		June 2008
+Contact:	Badari Pulavarty <pbadari@us.ibm.com>
+Description:
+		The file /sys/devices/system/memory/memoryX/removable
+		indicates whether this memory block is removable or not.
+		This is useful for a user-level agent to determine
+		identify removable sections of the memory before attempting
+		potentially expensive hot-remove memory operation
+
+Users:		hotplug memory remove tools
+		https://w3.opensource.ibm.com/projects/powerpc-utils/
diff --git a/drivers/base/memory.c b/drivers/base/memory.c
index 4d4e0e7b6e92..855ed1a9f97b 100644
--- a/drivers/base/memory.c
+++ b/drivers/base/memory.c
@@ -100,6 +100,21 @@ static ssize_t show_mem_phys_index(struct sys_device *dev,
 	return sprintf(buf, "%08lx\n", mem->phys_index);
 }
 
+/*
+ * Show whether the section of memory is likely to be hot-removable
+ */
+static ssize_t show_mem_removable(struct sys_device *dev, char *buf)
+{
+	unsigned long start_pfn;
+	int ret;
+	struct memory_block *mem =
+		container_of(dev, struct memory_block, sysdev);
+
+	start_pfn = section_nr_to_pfn(mem->phys_index);
+	ret = is_mem_section_removable(start_pfn, PAGES_PER_SECTION);
+	return sprintf(buf, "%d\n", ret);
+}
+
 /*
  * online, offline, going offline, etc.
  */
@@ -262,6 +277,7 @@ static ssize_t show_phys_device(struct sys_device *dev,
 static SYSDEV_ATTR(phys_index, 0444, show_mem_phys_index, NULL);
 static SYSDEV_ATTR(state, 0644, show_mem_state, store_mem_state);
 static SYSDEV_ATTR(phys_device, 0444, show_phys_device, NULL);
+static SYSDEV_ATTR(removable, 0444, show_mem_removable, NULL);
 
 #define mem_create_simple_file(mem, attr_name)	\
 	sysdev_create_file(&mem->sysdev, &attr_##attr_name)
@@ -350,6 +366,8 @@ static int add_memory_block(unsigned long node_id, struct mem_section *section,
 		ret = mem_create_simple_file(mem, state);
 	if (!ret)
 		ret = mem_create_simple_file(mem, phys_device);
+	if (!ret)
+		ret = mem_create_simple_file(mem, removable);
 
 	return ret;
 }
@@ -394,6 +412,7 @@ int remove_memory_block(unsigned long node_id, struct mem_section *section,
 	mem_remove_simple_file(mem, phys_index);
 	mem_remove_simple_file(mem, state);
 	mem_remove_simple_file(mem, phys_device);
+	mem_remove_simple_file(mem, removable);
 	unregister_memory(mem, section);
 
 	return 0;
diff --git a/include/linux/memory_hotplug.h b/include/linux/memory_hotplug.h
index 3628e5088f64..763ba81fc0f0 100644
--- a/include/linux/memory_hotplug.h
+++ b/include/linux/memory_hotplug.h
@@ -199,6 +199,18 @@ extern int walk_memory_resource(unsigned long start_pfn,
 			unsigned long nr_pages, void *arg,
 			int (*func)(unsigned long, unsigned long, void *));
 
+#ifdef CONFIG_MEMORY_HOTREMOVE
+
+extern int is_mem_section_removable(unsigned long pfn, unsigned long nr_pages);
+
+#else
+static inline int is_mem_section_removable(unsigned long pfn,
+					unsigned long nr_pages)
+{
+	return 0;
+}
+#endif /* CONFIG_MEMORY_HOTREMOVE */
+
 extern int add_memory(int nid, u64 start, u64 size);
 extern int arch_add_memory(int nid, u64 start, u64 size);
 extern int remove_memory(u64 start, u64 size);
diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index 93aba78dc8b6..89fee2dcb039 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -522,6 +522,66 @@ error:
 EXPORT_SYMBOL_GPL(add_memory);
 
 #ifdef CONFIG_MEMORY_HOTREMOVE
+/*
+ * A free page on the buddy free lists (not the per-cpu lists) has PageBuddy
+ * set and the size of the free page is given by page_order(). Using this,
+ * the function determines if the pageblock contains only free pages.
+ * Due to buddy contraints, a free page at least the size of a pageblock will
+ * be located at the start of the pageblock
+ */
+static inline int pageblock_free(struct page *page)
+{
+	return PageBuddy(page) && page_order(page) >= pageblock_order;
+}
+
+/* Return the start of the next active pageblock after a given page */
+static struct page *next_active_pageblock(struct page *page)
+{
+	int pageblocks_stride;
+
+	/* Ensure the starting page is pageblock-aligned */
+	BUG_ON(page_to_pfn(page) & (pageblock_nr_pages - 1));
+
+	/* Move forward by at least 1 * pageblock_nr_pages */
+	pageblocks_stride = 1;
+
+	/* If the entire pageblock is free, move to the end of free page */
+	if (pageblock_free(page))
+		pageblocks_stride += page_order(page) - pageblock_order;
+
+	return page + (pageblocks_stride * pageblock_nr_pages);
+}
+
+/* Checks if this range of memory is likely to be hot-removable. */
+int is_mem_section_removable(unsigned long start_pfn, unsigned long nr_pages)
+{
+	int type;
+	struct page *page = pfn_to_page(start_pfn);
+	struct page *end_page = page + nr_pages;
+
+	/* Check the starting page of each pageblock within the range */
+	for (; page < end_page; page = next_active_pageblock(page)) {
+		type = get_pageblock_migratetype(page);
+
+		/*
+		 * A pageblock containing MOVABLE or free pages is considered
+		 * removable
+		 */
+		if (type != MIGRATE_MOVABLE && !pageblock_free(page))
+			return 0;
+
+		/*
+		 * A pageblock starting with a PageReserved page is not
+		 * considered removable.
+		 */
+		if (PageReserved(page))
+			return 0;
+	}
+
+	/* All pageblocks in the memory block are likely to be hot-removable */
+	return 1;
+}
+
 /*
  * Confirm all pages in a range [start, end) is belongs to the same zone.
  */

From 9ca908f47bc784c90e17a553ce33e756c73feac4 Mon Sep 17 00:00:00 2001
From: Milton Miller <miltonm@bga.com>
Date: Wed, 23 Jul 2008 21:28:20 -0700
Subject: [PATCH 0295/1171] kcalloc: remove runtime division

While in all cases in the kernel we know the size of the elements to be
created, we don't always know the count of elements.  By commuting the size
and count in the overflow check, the compiler can reduce the runtime division
of size_t with a compare to a (unique) constant in these cases.

Signed-off-by: Milton Miller <miltonm@bga.com>
Cc: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/slab.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/linux/slab.h b/include/linux/slab.h
index 9aa90a6f20e0..41103910f8a2 100644
--- a/include/linux/slab.h
+++ b/include/linux/slab.h
@@ -180,7 +180,7 @@ size_t ksize(const void *);
  */
 static inline void *kcalloc(size_t n, size_t size, gfp_t flags)
 {
-	if (n != 0 && size > ULONG_MAX / n)
+	if (size != 0 && n > ULONG_MAX / size)
 		return NULL;
 	return __kmalloc(n * size, flags | __GFP_ZERO);
 }

From 83d1674a946141c3c59d430e96c224f7937e6158 Mon Sep 17 00:00:00 2001
From: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Date: Wed, 23 Jul 2008 21:28:22 -0700
Subject: [PATCH 0296/1171] mm: make CONFIG_MIGRATION available w/o CONFIG_NUMA

We'd like to support CONFIG_MEMORY_HOTREMOVE on s390, which depends on
CONFIG_MIGRATION.  So far, CONFIG_MIGRATION is only available with NUMA
support.

This patch makes CONFIG_MIGRATION selectable for architectures that define
ARCH_ENABLE_MEMORY_HOTREMOVE.  When MIGRATION is enabled w/o NUMA, the
kernel won't compile because migrate_vmas() does not know about
vm_ops->migrate() and vma_migratable() does not know about policy_zone.
To fix this, those two functions can be restricted to '#ifdef CONFIG_NUMA'
because they are not being used w/o NUMA.  vma_migratable() is moved over
from migrate.h to mempolicy.h.

[kosaki.motohiro@jp.fujitsu.com: build fix]
Acked-by: Christoph Lameter <cl@linux-foundation.org>
Signed-off-by: Gerald Schaefer <gerald.schaefer@de.ibm.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: KOSAKI Motorhiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/mempolicy.h | 19 +++++++++++++++++++
 include/linux/migrate.h   | 21 ---------------------
 mm/Kconfig                |  2 +-
 mm/migrate.c              |  2 +-
 4 files changed, 21 insertions(+), 23 deletions(-)

diff --git a/include/linux/mempolicy.h b/include/linux/mempolicy.h
index 3a39570b81b8..085c903fe0f1 100644
--- a/include/linux/mempolicy.h
+++ b/include/linux/mempolicy.h
@@ -59,6 +59,7 @@ enum {
 #include <linux/rbtree.h>
 #include <linux/spinlock.h>
 #include <linux/nodemask.h>
+#include <linux/pagemap.h>
 
 struct mm_struct;
 
@@ -220,6 +221,24 @@ extern int mpol_parse_str(char *str, struct mempolicy **mpol, int no_context);
 extern int mpol_to_str(char *buffer, int maxlen, struct mempolicy *pol,
 			int no_context);
 #endif
+
+/* Check if a vma is migratable */
+static inline int vma_migratable(struct vm_area_struct *vma)
+{
+	if (vma->vm_flags & (VM_IO|VM_HUGETLB|VM_PFNMAP|VM_RESERVED))
+		return 0;
+	/*
+	 * Migration allocates pages in the highest zone. If we cannot
+	 * do so then migration (at least from node to node) is not
+	 * possible.
+	 */
+	if (vma->vm_file &&
+		gfp_zone(mapping_gfp_mask(vma->vm_file->f_mapping))
+								< policy_zone)
+			return 0;
+	return 1;
+}
+
 #else
 
 struct mempolicy {};
diff --git a/include/linux/migrate.h b/include/linux/migrate.h
index e10a90a93b5d..03aea612d284 100644
--- a/include/linux/migrate.h
+++ b/include/linux/migrate.h
@@ -3,28 +3,10 @@
 
 #include <linux/mm.h>
 #include <linux/mempolicy.h>
-#include <linux/pagemap.h>
 
 typedef struct page *new_page_t(struct page *, unsigned long private, int **);
 
 #ifdef CONFIG_MIGRATION
-/* Check if a vma is migratable */
-static inline int vma_migratable(struct vm_area_struct *vma)
-{
-	if (vma->vm_flags & (VM_IO|VM_HUGETLB|VM_PFNMAP|VM_RESERVED))
-		return 0;
-	/*
-	 * Migration allocates pages in the highest zone. If we cannot
-	 * do so then migration (at least from node to node) is not
-	 * possible.
-	 */
-	if (vma->vm_file &&
-		gfp_zone(mapping_gfp_mask(vma->vm_file->f_mapping))
-								< policy_zone)
-			return 0;
-	return 1;
-}
-
 extern int isolate_lru_page(struct page *p, struct list_head *pagelist);
 extern int putback_lru_pages(struct list_head *l);
 extern int migrate_page(struct address_space *,
@@ -39,9 +21,6 @@ extern int migrate_vmas(struct mm_struct *mm,
 		const nodemask_t *from, const nodemask_t *to,
 		unsigned long flags);
 #else
-static inline int vma_migratable(struct vm_area_struct *vma)
-					{ return 0; }
-
 static inline int isolate_lru_page(struct page *p, struct list_head *list)
 					{ return -ENOSYS; }
 static inline int putback_lru_pages(struct list_head *l) { return 0; }
diff --git a/mm/Kconfig b/mm/Kconfig
index c4de85285bb4..aa799007a11b 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -174,7 +174,7 @@ config SPLIT_PTLOCK_CPUS
 config MIGRATION
 	bool "Page migration"
 	def_bool y
-	depends on NUMA
+	depends on NUMA || ARCH_ENABLE_MEMORY_HOTREMOVE
 	help
 	  Allows the migration of the physical location of pages of processes
 	  while the virtual addresses are not changed. This is useful for
diff --git a/mm/migrate.c b/mm/migrate.c
index e7d13a708da0..376cceba82f9 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -1071,7 +1071,6 @@ out2:
 	mmput(mm);
 	return err;
 }
-#endif
 
 /*
  * Call migration functions in the vma_ops that may prepare
@@ -1093,3 +1092,4 @@ int migrate_vmas(struct mm_struct *mm, const nodemask_t *to,
  	}
  	return err;
 }
+#endif

From 78ecba081224a2db5876b6b81cfed0b78f58adc7 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Wed, 23 Jul 2008 21:28:23 -0700
Subject: [PATCH 0297/1171] mm: fix ever-decreasing swap priority

Vegard Nossum has noticed the ever-decreasing negative priority in a
swapon /swapoff loop, which eventually would misprioritize when int wraps
positive.  Not worth spending much code on, but probably better fixed.

It's easy to handle the swapping on and off of just one area, but there's
not much point if a pair or more still misbehave.  To handle the general
case, swapoff should compact negative priorities, keeping them always from
-1 to -MAX_SWAPFILES.  That's a change, but should cause no regression,
since these negative (unspecified) priorities are disjoint from the the
positive specified priorities 0 to 32767.

One small functional difference, which seems appropriate: when swapoff
fails to free all swap from a negative priority area, that area is now
reinserted at lowest priority, rather than at its original priority.

In moving down swapon's setting of priority, I notice that an area is
visible to /proc/swaps when it has swap_map set, yet that was being set
before all the visible fields were properly filled in: corrected.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Reported-by: Vegard Nossum <vegard.nossum@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/swapfile.c | 49 +++++++++++++++++++++++++------------------------
 1 file changed, 25 insertions(+), 24 deletions(-)

diff --git a/mm/swapfile.c b/mm/swapfile.c
index bd1bb5920306..2f33edb8bee9 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -37,6 +37,7 @@ DEFINE_SPINLOCK(swap_lock);
 unsigned int nr_swapfiles;
 long total_swap_pages;
 static int swap_overflow;
+static int least_priority;
 
 static const char Bad_file[] = "Bad swap file entry ";
 static const char Unused_file[] = "Unused swap file entry ";
@@ -1260,6 +1261,11 @@ asmlinkage long sys_swapoff(const char __user * specialfile)
 		/* just pick something that's safe... */
 		swap_list.next = swap_list.head;
 	}
+	if (p->prio < 0) {
+		for (i = p->next; i >= 0; i = swap_info[i].next)
+			swap_info[i].prio = p->prio--;
+		least_priority++;
+	}
 	nr_swap_pages -= p->pages;
 	total_swap_pages -= p->pages;
 	p->flags &= ~SWP_WRITEOK;
@@ -1272,9 +1278,14 @@ asmlinkage long sys_swapoff(const char __user * specialfile)
 	if (err) {
 		/* re-insert swap space back into swap_list */
 		spin_lock(&swap_lock);
-		for (prev = -1, i = swap_list.head; i >= 0; prev = i, i = swap_info[i].next)
+		if (p->prio < 0)
+			p->prio = --least_priority;
+		prev = -1;
+		for (i = swap_list.head; i >= 0; i = swap_info[i].next) {
 			if (p->prio >= swap_info[i].prio)
 				break;
+			prev = i;
+		}
 		p->next = i;
 		if (prev < 0)
 			swap_list.head = swap_list.next = p - swap_info;
@@ -1447,7 +1458,6 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags)
 	unsigned int type;
 	int i, prev;
 	int error;
-	static int least_priority;
 	union swap_header *swap_header = NULL;
 	int swap_header_version;
 	unsigned int nr_good_pages = 0;
@@ -1455,7 +1465,7 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags)
 	sector_t span;
 	unsigned long maxpages = 1;
 	int swapfilesize;
-	unsigned short *swap_map;
+	unsigned short *swap_map = NULL;
 	struct page *page = NULL;
 	struct inode *inode = NULL;
 	int did_down = 0;
@@ -1474,22 +1484,10 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags)
 	}
 	if (type >= nr_swapfiles)
 		nr_swapfiles = type+1;
+	memset(p, 0, sizeof(*p));
 	INIT_LIST_HEAD(&p->extent_list);
 	p->flags = SWP_USED;
-	p->swap_file = NULL;
-	p->old_block_size = 0;
-	p->swap_map = NULL;
-	p->lowest_bit = 0;
-	p->highest_bit = 0;
-	p->cluster_nr = 0;
-	p->inuse_pages = 0;
 	p->next = -1;
-	if (swap_flags & SWAP_FLAG_PREFER) {
-		p->prio =
-		  (swap_flags & SWAP_FLAG_PRIO_MASK)>>SWAP_FLAG_PRIO_SHIFT;
-	} else {
-		p->prio = --least_priority;
-	}
 	spin_unlock(&swap_lock);
 	name = getname(specialfile);
 	error = PTR_ERR(name);
@@ -1632,19 +1630,20 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags)
 			goto bad_swap;
 
 		/* OK, set up the swap map and apply the bad block list */
-		if (!(p->swap_map = vmalloc(maxpages * sizeof(short)))) {
+		swap_map = vmalloc(maxpages * sizeof(short));
+		if (!swap_map) {
 			error = -ENOMEM;
 			goto bad_swap;
 		}
 
 		error = 0;
-		memset(p->swap_map, 0, maxpages * sizeof(short));
+		memset(swap_map, 0, maxpages * sizeof(short));
 		for (i = 0; i < swap_header->info.nr_badpages; i++) {
 			int page_nr = swap_header->info.badpages[i];
 			if (page_nr <= 0 || page_nr >= swap_header->info.last_page)
 				error = -EINVAL;
 			else
-				p->swap_map[page_nr] = SWAP_MAP_BAD;
+				swap_map[page_nr] = SWAP_MAP_BAD;
 		}
 		nr_good_pages = swap_header->info.last_page -
 				swap_header->info.nr_badpages -
@@ -1654,7 +1653,7 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags)
 	}
 
 	if (nr_good_pages) {
-		p->swap_map[0] = SWAP_MAP_BAD;
+		swap_map[0] = SWAP_MAP_BAD;
 		p->max = maxpages;
 		p->pages = nr_good_pages;
 		nr_extents = setup_swap_extents(p, &span);
@@ -1672,6 +1671,12 @@ asmlinkage long sys_swapon(const char __user * specialfile, int swap_flags)
 
 	mutex_lock(&swapon_mutex);
 	spin_lock(&swap_lock);
+	if (swap_flags & SWAP_FLAG_PREFER)
+		p->prio =
+		  (swap_flags & SWAP_FLAG_PRIO_MASK) >> SWAP_FLAG_PRIO_SHIFT;
+	else
+		p->prio = --least_priority;
+	p->swap_map = swap_map;
 	p->flags = SWP_ACTIVE;
 	nr_swap_pages += nr_good_pages;
 	total_swap_pages += nr_good_pages;
@@ -1707,12 +1712,8 @@ bad_swap:
 	destroy_swap_extents(p);
 bad_swap_2:
 	spin_lock(&swap_lock);
-	swap_map = p->swap_map;
 	p->swap_file = NULL;
-	p->swap_map = NULL;
 	p->flags = 0;
-	if (!(swap_flags & SWAP_FLAG_PREFER))
-		++least_priority;
 	spin_unlock(&swap_lock);
 	vfree(swap_map);
 	if (swap_file)

From 5459c164f0591ee75ed0203bb8f3817f25948e2f Mon Sep 17 00:00:00 2001
From: "Andrew G. Morgan" <morgan@kernel.org>
Date: Wed, 23 Jul 2008 21:28:24 -0700
Subject: [PATCH 0298/1171] security: protect legacy applications from
 executing with insufficient privilege

When cap_bset suppresses some of the forced (fP) capabilities of a file,
it is generally only safe to execute the program if it understands how to
recognize it doesn't have enough privilege to work correctly.  For legacy
applications (fE!=0), which have no non-destructive way to determine that
they are missing privilege, we fail to execute (EPERM) any executable that
requires fP capabilities, but would otherwise get pP' < fP.  This is a
fail-safe permission check.

For some discussion of why it is problematic for (legacy) privileged
applications to run with less than the set of capabilities requested for
them, see:

 http://userweb.kernel.org/~morgan/sendmail-capabilities-war-story.html

With this iteration of this support, we do not include setuid-0 based
privilege protection from the bounding set.  That is, the admin can still
(ab)use the bounding set to suppress the privileges of a setuid-0 program.

[akpm@linux-foundation.org: coding-style fixes]
[akpm@linux-foundation.org: cleanup]
Signed-off-by: Andrew G. Morgan <morgan@kernel.org>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/binfmts.h |   2 +-
 security/commoncap.c    | 108 ++++++++++++++++++++++------------------
 2 files changed, 60 insertions(+), 50 deletions(-)

diff --git a/include/linux/binfmts.h b/include/linux/binfmts.h
index ee0ed48e8348..826f62350805 100644
--- a/include/linux/binfmts.h
+++ b/include/linux/binfmts.h
@@ -38,7 +38,7 @@ struct linux_binprm{
 		     misc_bang:1;
 	struct file * file;
 	int e_uid, e_gid;
-	kernel_cap_t cap_inheritable, cap_permitted;
+	kernel_cap_t cap_post_exec_permitted;
 	bool cap_effective;
 	void *security;
 	int argc, envc;
diff --git a/security/commoncap.c b/security/commoncap.c
index 0b6537a3672d..4afbece37a08 100644
--- a/security/commoncap.c
+++ b/security/commoncap.c
@@ -162,8 +162,7 @@ void cap_capset_set (struct task_struct *target, kernel_cap_t *effective,
 
 static inline void bprm_clear_caps(struct linux_binprm *bprm)
 {
-	cap_clear(bprm->cap_inheritable);
-	cap_clear(bprm->cap_permitted);
+	cap_clear(bprm->cap_post_exec_permitted);
 	bprm->cap_effective = false;
 }
 
@@ -198,6 +197,7 @@ static inline int cap_from_disk(struct vfs_cap_data *caps,
 {
 	__u32 magic_etc;
 	unsigned tocopy, i;
+	int ret;
 
 	if (size < sizeof(magic_etc))
 		return -EINVAL;
@@ -225,19 +225,40 @@ static inline int cap_from_disk(struct vfs_cap_data *caps,
 		bprm->cap_effective = false;
 	}
 
-	for (i = 0; i < tocopy; ++i) {
-		bprm->cap_permitted.cap[i] =
-			le32_to_cpu(caps->data[i].permitted);
-		bprm->cap_inheritable.cap[i] =
-			le32_to_cpu(caps->data[i].inheritable);
-	}
-	while (i < VFS_CAP_U32) {
-		bprm->cap_permitted.cap[i] = 0;
-		bprm->cap_inheritable.cap[i] = 0;
-		i++;
+	ret = 0;
+
+	CAP_FOR_EACH_U32(i) {
+		__u32 value_cpu;
+
+		if (i >= tocopy) {
+			/*
+			 * Legacy capability sets have no upper bits
+			 */
+			bprm->cap_post_exec_permitted.cap[i] = 0;
+			continue;
+		}
+		/*
+		 * pP' = (X & fP) | (pI & fI)
+		 */
+		value_cpu = le32_to_cpu(caps->data[i].permitted);
+		bprm->cap_post_exec_permitted.cap[i] =
+			(current->cap_bset.cap[i] & value_cpu) |
+			(current->cap_inheritable.cap[i] &
+				le32_to_cpu(caps->data[i].inheritable));
+		if (value_cpu & ~bprm->cap_post_exec_permitted.cap[i]) {
+			/*
+			 * insufficient to execute correctly
+			 */
+			ret = -EPERM;
+		}
 	}
 
-	return 0;
+	/*
+	 * For legacy apps, with no internal support for recognizing they
+	 * do not have enough capabilities, we return an error if they are
+	 * missing some "forced" (aka file-permitted) capabilities.
+	 */
+	return bprm->cap_effective ? ret : 0;
 }
 
 /* Locate any VFS capabilities: */
@@ -269,9 +290,9 @@ static int get_file_caps(struct linux_binprm *bprm)
 		goto out;
 
 	rc = cap_from_disk(&vcaps, bprm, rc);
-	if (rc)
+	if (rc == -EINVAL)
 		printk(KERN_NOTICE "%s: cap_from_disk returned %d for %s\n",
-			__func__, rc, bprm->filename);
+		       __func__, rc, bprm->filename);
 
 out:
 	dput(dentry);
@@ -304,25 +325,24 @@ int cap_bprm_set_security (struct linux_binprm *bprm)
 	int ret;
 
 	ret = get_file_caps(bprm);
-	if (ret)
-		printk(KERN_NOTICE "%s: get_file_caps returned %d for %s\n",
-			__func__, ret, bprm->filename);
 
-	/*  To support inheritance of root-permissions and suid-root
-	 *  executables under compatibility mode, we raise all three
-	 *  capability sets for the file.
-	 *
-	 *  If only the real uid is 0, we only raise the inheritable
-	 *  and permitted sets of the executable file.
-	 */
-
-	if (!issecure (SECURE_NOROOT)) {
+	if (!issecure(SECURE_NOROOT)) {
+		/*
+		 * To support inheritance of root-permissions and suid-root
+		 * executables under compatibility mode, we override the
+		 * capability sets for the file.
+		 *
+		 * If only the real uid is 0, we do not set the effective
+		 * bit.
+		 */
 		if (bprm->e_uid == 0 || current->uid == 0) {
-			cap_set_full (bprm->cap_inheritable);
-			cap_set_full (bprm->cap_permitted);
+			/* pP' = (cap_bset & ~0) | (pI & ~0) */
+			bprm->cap_post_exec_permitted = cap_combine(
+				current->cap_bset, current->cap_inheritable
+				);
+			bprm->cap_effective = (bprm->e_uid == 0);
+			ret = 0;
 		}
-		if (bprm->e_uid == 0)
-			bprm->cap_effective = true;
 	}
 
 	return ret;
@@ -330,17 +350,9 @@ int cap_bprm_set_security (struct linux_binprm *bprm)
 
 void cap_bprm_apply_creds (struct linux_binprm *bprm, int unsafe)
 {
-	/* Derived from fs/exec.c:compute_creds. */
-	kernel_cap_t new_permitted, working;
-
-	new_permitted = cap_intersect(bprm->cap_permitted,
-				 current->cap_bset);
-	working = cap_intersect(bprm->cap_inheritable,
-				 current->cap_inheritable);
-	new_permitted = cap_combine(new_permitted, working);
-
 	if (bprm->e_uid != current->uid || bprm->e_gid != current->gid ||
-	    !cap_issubset (new_permitted, current->cap_permitted)) {
+	    !cap_issubset(bprm->cap_post_exec_permitted,
+			  current->cap_permitted)) {
 		set_dumpable(current->mm, suid_dumpable);
 		current->pdeath_signal = 0;
 
@@ -350,9 +362,9 @@ void cap_bprm_apply_creds (struct linux_binprm *bprm, int unsafe)
 				bprm->e_gid = current->gid;
 			}
 			if (cap_limit_ptraced_target()) {
-				new_permitted =
-					cap_intersect(new_permitted,
-						      current->cap_permitted);
+				bprm->cap_post_exec_permitted = cap_intersect(
+					bprm->cap_post_exec_permitted,
+					current->cap_permitted);
 			}
 		}
 	}
@@ -364,9 +376,9 @@ void cap_bprm_apply_creds (struct linux_binprm *bprm, int unsafe)
 	 * in the init_task struct. Thus we skip the usual
 	 * capability rules */
 	if (!is_global_init(current)) {
-		current->cap_permitted = new_permitted;
+		current->cap_permitted = bprm->cap_post_exec_permitted;
 		if (bprm->cap_effective)
-			current->cap_effective = new_permitted;
+			current->cap_effective = bprm->cap_post_exec_permitted;
 		else
 			cap_clear(current->cap_effective);
 	}
@@ -381,9 +393,7 @@ int cap_bprm_secureexec (struct linux_binprm *bprm)
 	if (current->uid != 0) {
 		if (bprm->cap_effective)
 			return 1;
-		if (!cap_isclear(bprm->cap_permitted))
-			return 1;
-		if (!cap_isclear(bprm->cap_inheritable))
+		if (!cap_isclear(bprm->cap_post_exec_permitted))
 			return 1;
 	}
 

From ab763c7112ce0e2559c73f921617c81dc7287ca6 Mon Sep 17 00:00:00 2001
From: "Andrew G. Morgan" <morgan@kernel.org>
Date: Wed, 23 Jul 2008 21:28:25 -0700
Subject: [PATCH 0299/1171] security: filesystem capabilities refactor kernel
 code

To date, we've tried hard to confine filesystem support for capabilities
to the security modules.  This has left a lot of the code in
kernel/capability.c in a state where it looks like it supports something
that filesystem support for capabilities actually suppresses when the LSM
security/commmoncap.c code runs.  What is left is a lot of code that uses
sub-optimal locking in the main kernel

With this change we refactor the main kernel code and make it explicit
which locks are needed and that the only remaining kernel races in this
area are associated with non-filesystem capability code.

Signed-off-by: Andrew G. Morgan <morgan@kernel.org>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/capability.c | 338 +++++++++++++++++++++++++++++---------------
 1 file changed, 221 insertions(+), 117 deletions(-)

diff --git a/kernel/capability.c b/kernel/capability.c
index 901e0fdc3fff..0101e847603e 100644
--- a/kernel/capability.c
+++ b/kernel/capability.c
@@ -115,11 +115,208 @@ static int cap_validate_magic(cap_user_header_t header, unsigned *tocopy)
 	return 0;
 }
 
+#ifndef CONFIG_SECURITY_FILE_CAPABILITIES
+
 /*
- * For sys_getproccap() and sys_setproccap(), any of the three
- * capability set pointers may be NULL -- indicating that that set is
- * uninteresting and/or not to be changed.
+ * Without filesystem capability support, we nominally support one process
+ * setting the capabilities of another
  */
+static inline int cap_get_target_pid(pid_t pid, kernel_cap_t *pEp,
+				     kernel_cap_t *pIp, kernel_cap_t *pPp)
+{
+	struct task_struct *target;
+	int ret;
+
+	spin_lock(&task_capability_lock);
+	read_lock(&tasklist_lock);
+
+	if (pid && pid != task_pid_vnr(current)) {
+		target = find_task_by_vpid(pid);
+		if (!target) {
+			ret = -ESRCH;
+			goto out;
+		}
+	} else
+		target = current;
+
+	ret = security_capget(target, pEp, pIp, pPp);
+
+out:
+	read_unlock(&tasklist_lock);
+	spin_unlock(&task_capability_lock);
+
+	return ret;
+}
+
+/*
+ * cap_set_pg - set capabilities for all processes in a given process
+ * group.  We call this holding task_capability_lock and tasklist_lock.
+ */
+static inline int cap_set_pg(int pgrp_nr, kernel_cap_t *effective,
+			     kernel_cap_t *inheritable,
+			     kernel_cap_t *permitted)
+{
+	struct task_struct *g, *target;
+	int ret = -EPERM;
+	int found = 0;
+	struct pid *pgrp;
+
+	spin_lock(&task_capability_lock);
+	read_lock(&tasklist_lock);
+
+	pgrp = find_vpid(pgrp_nr);
+	do_each_pid_task(pgrp, PIDTYPE_PGID, g) {
+		target = g;
+		while_each_thread(g, target) {
+			if (!security_capset_check(target, effective,
+						   inheritable, permitted)) {
+				security_capset_set(target, effective,
+						    inheritable, permitted);
+				ret = 0;
+			}
+			found = 1;
+		}
+	} while_each_pid_task(pgrp, PIDTYPE_PGID, g);
+
+	read_unlock(&tasklist_lock);
+	spin_unlock(&task_capability_lock);
+
+	if (!found)
+		ret = 0;
+	return ret;
+}
+
+/*
+ * cap_set_all - set capabilities for all processes other than init
+ * and self.  We call this holding task_capability_lock and tasklist_lock.
+ */
+static inline int cap_set_all(kernel_cap_t *effective,
+			      kernel_cap_t *inheritable,
+			      kernel_cap_t *permitted)
+{
+	struct task_struct *g, *target;
+	int ret = -EPERM;
+	int found = 0;
+
+	spin_lock(&task_capability_lock);
+	read_lock(&tasklist_lock);
+
+	do_each_thread(g, target) {
+		if (target == current
+		    || is_container_init(target->group_leader))
+			continue;
+		found = 1;
+		if (security_capset_check(target, effective, inheritable,
+					  permitted))
+			continue;
+		ret = 0;
+		security_capset_set(target, effective, inheritable, permitted);
+	} while_each_thread(g, target);
+
+	read_unlock(&tasklist_lock);
+	spin_unlock(&task_capability_lock);
+
+	if (!found)
+		ret = 0;
+
+	return ret;
+}
+
+/*
+ * Given the target pid does not refer to the current process we
+ * need more elaborate support... (This support is not present when
+ * filesystem capabilities are configured.)
+ */
+static inline int do_sys_capset_other_tasks(pid_t pid, kernel_cap_t *effective,
+					    kernel_cap_t *inheritable,
+					    kernel_cap_t *permitted)
+{
+	struct task_struct *target;
+	int ret;
+
+	if (!capable(CAP_SETPCAP))
+		return -EPERM;
+
+	if (pid == -1)	          /* all procs other than current and init */
+		return cap_set_all(effective, inheritable, permitted);
+
+	else if (pid < 0)                    /* all procs in process group */
+		return cap_set_pg(-pid, effective, inheritable, permitted);
+
+	/* target != current */
+	spin_lock(&task_capability_lock);
+	read_lock(&tasklist_lock);
+
+	target = find_task_by_vpid(pid);
+	if (!target)
+		ret = -ESRCH;
+	else {
+		ret = security_capset_check(target, effective, inheritable,
+					    permitted);
+
+		/* having verified that the proposed changes are legal,
+		   we now put them into effect. */
+		if (!ret)
+			security_capset_set(target, effective, inheritable,
+					    permitted);
+	}
+
+	read_unlock(&tasklist_lock);
+	spin_unlock(&task_capability_lock);
+
+	return ret;
+}
+
+#else /* ie., def CONFIG_SECURITY_FILE_CAPABILITIES */
+
+/*
+ * If we have configured with filesystem capability support, then the
+ * only thing that can change the capabilities of the current process
+ * is the current process. As such, we can't be in this code at the
+ * same time as we are in the process of setting capabilities in this
+ * process. The net result is that we can limit our use of locks to
+ * when we are reading the caps of another process.
+ */
+static inline int cap_get_target_pid(pid_t pid, kernel_cap_t *pEp,
+				     kernel_cap_t *pIp, kernel_cap_t *pPp)
+{
+	int ret;
+
+	if (pid && (pid != task_pid_vnr(current))) {
+		struct task_struct *target;
+
+		spin_lock(&task_capability_lock);
+		read_lock(&tasklist_lock);
+
+		target = find_task_by_vpid(pid);
+		if (!target)
+			ret = -ESRCH;
+		else
+			ret = security_capget(target, pEp, pIp, pPp);
+
+		read_unlock(&tasklist_lock);
+		spin_unlock(&task_capability_lock);
+	} else
+		ret = security_capget(current, pEp, pIp, pPp);
+
+	return ret;
+}
+
+/*
+ * With filesystem capability support configured, the kernel does not
+ * permit the changing of capabilities in one process by another
+ * process. (CAP_SETPCAP has much less broad semantics when configured
+ * this way.)
+ */
+static inline int do_sys_capset_other_tasks(pid_t pid,
+					    kernel_cap_t *effective,
+					    kernel_cap_t *inheritable,
+					    kernel_cap_t *permitted)
+{
+	return -EPERM;
+}
+
+#endif /* ie., ndef CONFIG_SECURITY_FILE_CAPABILITIES */
 
 /*
  * Atomically modify the effective capabilities returning the original
@@ -155,7 +352,6 @@ asmlinkage long sys_capget(cap_user_header_t header, cap_user_data_t dataptr)
 {
 	int ret = 0;
 	pid_t pid;
-	struct task_struct *target;
 	unsigned tocopy;
 	kernel_cap_t pE, pI, pP;
 
@@ -169,23 +365,7 @@ asmlinkage long sys_capget(cap_user_header_t header, cap_user_data_t dataptr)
 	if (pid < 0)
 		return -EINVAL;
 
-	spin_lock(&task_capability_lock);
-	read_lock(&tasklist_lock);
-
-	if (pid && pid != task_pid_vnr(current)) {
-		target = find_task_by_vpid(pid);
-		if (!target) {
-			ret = -ESRCH;
-			goto out;
-		}
-	} else
-		target = current;
-
-	ret = security_capget(target, &pE, &pI, &pP);
-
-out:
-	read_unlock(&tasklist_lock);
-	spin_unlock(&task_capability_lock);
+	ret = cap_get_target_pid(pid, &pE, &pI, &pP);
 
 	if (!ret) {
 		struct __user_cap_data_struct kdata[_KERNEL_CAPABILITY_U32S];
@@ -216,7 +396,6 @@ out:
 		 * before modification is attempted and the application
 		 * fails.
 		 */
-
 		if (copy_to_user(dataptr, kdata, tocopy
 				 * sizeof(struct __user_cap_data_struct))) {
 			return -EFAULT;
@@ -226,70 +405,8 @@ out:
 	return ret;
 }
 
-/*
- * cap_set_pg - set capabilities for all processes in a given process
- * group.  We call this holding task_capability_lock and tasklist_lock.
- */
-static inline int cap_set_pg(int pgrp_nr, kernel_cap_t *effective,
-			      kernel_cap_t *inheritable,
-			      kernel_cap_t *permitted)
-{
-	struct task_struct *g, *target;
-	int ret = -EPERM;
-	int found = 0;
-	struct pid *pgrp;
-
-	pgrp = find_vpid(pgrp_nr);
-	do_each_pid_task(pgrp, PIDTYPE_PGID, g) {
-		target = g;
-		while_each_thread(g, target) {
-			if (!security_capset_check(target, effective,
-							inheritable,
-							permitted)) {
-				security_capset_set(target, effective,
-							inheritable,
-							permitted);
-				ret = 0;
-			}
-			found = 1;
-		}
-	} while_each_pid_task(pgrp, PIDTYPE_PGID, g);
-
-	if (!found)
-		ret = 0;
-	return ret;
-}
-
-/*
- * cap_set_all - set capabilities for all processes other than init
- * and self.  We call this holding task_capability_lock and tasklist_lock.
- */
-static inline int cap_set_all(kernel_cap_t *effective,
-			       kernel_cap_t *inheritable,
-			       kernel_cap_t *permitted)
-{
-     struct task_struct *g, *target;
-     int ret = -EPERM;
-     int found = 0;
-
-     do_each_thread(g, target) {
-             if (target == current || is_container_init(target->group_leader))
-                     continue;
-             found = 1;
-	     if (security_capset_check(target, effective, inheritable,
-						permitted))
-		     continue;
-	     ret = 0;
-	     security_capset_set(target, effective, inheritable, permitted);
-     } while_each_thread(g, target);
-
-     if (!found)
-	     ret = 0;
-     return ret;
-}
-
 /**
- * sys_capset - set capabilities for a process or a group of processes
+ * sys_capset - set capabilities for a process or (*) a group of processes
  * @header: pointer to struct that contains capability version and
  *	target pid data
  * @data: pointer to struct that contains the effective, permitted,
@@ -313,7 +430,6 @@ asmlinkage long sys_capset(cap_user_header_t header, const cap_user_data_t data)
 	struct __user_cap_data_struct kdata[_KERNEL_CAPABILITY_U32S];
 	unsigned i, tocopy;
 	kernel_cap_t inheritable, permitted, effective;
-	struct task_struct *target;
 	int ret;
 	pid_t pid;
 
@@ -324,9 +440,6 @@ asmlinkage long sys_capset(cap_user_header_t header, const cap_user_data_t data)
 	if (get_user(pid, &header->pid))
 		return -EFAULT;
 
-	if (pid && pid != task_pid_vnr(current) && !capable(CAP_SETPCAP))
-		return -EPERM;
-
 	if (copy_from_user(&kdata, data, tocopy
 			   * sizeof(struct __user_cap_data_struct))) {
 		return -EFAULT;
@@ -344,40 +457,31 @@ asmlinkage long sys_capset(cap_user_header_t header, const cap_user_data_t data)
 		i++;
 	}
 
-	spin_lock(&task_capability_lock);
-	read_lock(&tasklist_lock);
+	if (pid && (pid != task_pid_vnr(current)))
+		ret = do_sys_capset_other_tasks(pid, &effective, &inheritable,
+						&permitted);
+	else {
+		/*
+		 * This lock is required even when filesystem
+		 * capability support is configured - it protects the
+		 * sys_capget() call from returning incorrect data in
+		 * the case that the targeted process is not the
+		 * current one.
+		 */
+		spin_lock(&task_capability_lock);
 
-	if (pid > 0 && pid != task_pid_vnr(current)) {
-		target = find_task_by_vpid(pid);
-		if (!target) {
-			ret = -ESRCH;
-			goto out;
-		}
-	} else
-		target = current;
-
-	ret = 0;
-
-	/* having verified that the proposed changes are legal,
-	   we now put them into effect. */
-	if (pid < 0) {
-		if (pid == -1)	/* all procs other than current and init */
-			ret = cap_set_all(&effective, &inheritable, &permitted);
-
-		else		/* all procs in process group */
-			ret = cap_set_pg(-pid, &effective, &inheritable,
-					 &permitted);
-	} else {
-		ret = security_capset_check(target, &effective, &inheritable,
+		ret = security_capset_check(current, &effective, &inheritable,
 					    &permitted);
+		/*
+		 * Having verified that the proposed changes are
+		 * legal, we now put them into effect.
+		 */
 		if (!ret)
-			security_capset_set(target, &effective, &inheritable,
+			security_capset_set(current, &effective, &inheritable,
 					    &permitted);
+		spin_unlock(&task_capability_lock);
 	}
 
-out:
-	read_unlock(&tasklist_lock);
-	spin_unlock(&task_capability_lock);
 
 	return ret;
 }

From 84aaa7ab4c40b66d6dd9aa393901551ad50ec640 Mon Sep 17 00:00:00 2001
From: "Andrew G. Morgan" <morgan@kernel.org>
Date: Wed, 23 Jul 2008 21:28:25 -0700
Subject: [PATCH 0300/1171] security: filesystem capabilities no longer
 experimental

Filesystem capabilities have come of age.  Remove the experimental tag for
configuring filesystem capabilities.

Signed-off-by: Andrew G. Morgan <morgan@kernel.org>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 security/Kconfig | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/security/Kconfig b/security/Kconfig
index 62ed4717d334..559293922a47 100644
--- a/security/Kconfig
+++ b/security/Kconfig
@@ -74,8 +74,7 @@ config SECURITY_NETWORK_XFRM
 	  If you are unsure how to answer this question, answer N.
 
 config SECURITY_FILE_CAPABILITIES
-	bool "File POSIX Capabilities (EXPERIMENTAL)"
-	depends on EXPERIMENTAL
+	bool "File POSIX Capabilities"
 	default n
 	help
 	  This enables filesystem capabilities, allowing you to give

From 9b3e43a747c74029b0acf6acf4666601f132f471 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Wed, 23 Jul 2008 21:28:26 -0700
Subject: [PATCH 0301/1171] security: remove unused forwards

Why would linux/security.h need forward declarations for nfsctl_arg and
swap_info_struct?  It's hard to imagine: remove them.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/security.h | 2 --
 1 file changed, 2 deletions(-)

diff --git a/include/linux/security.h b/include/linux/security.h
index 31c8851ec5d0..f0e9adb22ac2 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -102,9 +102,7 @@ extern unsigned long mmap_min_addr;
 #define LSM_SETID_FS	8
 
 /* forward declares to avoid warnings */
-struct nfsctl_arg;
 struct sched_param;
-struct swap_info_struct;
 struct request_sock;
 
 /* bprm_apply_creds unsafe reasons */

From 5002779d37b261271da9883e06c14b097d4781c4 Mon Sep 17 00:00:00 2001
From: Tilman Schmidt <tilman@imap.cc>
Date: Wed, 23 Jul 2008 21:28:27 -0700
Subject: [PATCH 0302/1171] gigaset: use dev_ macros for messages

The info() / warn() / err() macros from usb.h for generating kernel
messages are considered inferior to dev_info() / dev_warn() / dev_err()
from device.h.  Replace them where possible.  Also correct the severity
level and improve the text of one message.

Signed-off-by: Tilman Schmidt <tilman@imap.cc>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/isdn/gigaset/asyncdata.c   |  3 +-
 drivers/isdn/gigaset/common.c      |  2 +-
 drivers/isdn/gigaset/gigaset.h     |  3 --
 drivers/isdn/gigaset/i4l.c         | 56 ++++++++++++++++++------------
 drivers/isdn/gigaset/interface.c   | 25 ++++++-------
 drivers/isdn/gigaset/usb-gigaset.c |  7 ++--
 6 files changed, 53 insertions(+), 43 deletions(-)

diff --git a/drivers/isdn/gigaset/asyncdata.c b/drivers/isdn/gigaset/asyncdata.c
index 091deb9d1c47..c2bd97d29273 100644
--- a/drivers/isdn/gigaset/asyncdata.c
+++ b/drivers/isdn/gigaset/asyncdata.c
@@ -575,7 +575,8 @@ int gigaset_m10x_send_skb(struct bc_state *bcs, struct sk_buff *skb)
 	else
 		skb = iraw_encode(skb, HW_HDR_LEN, 0);
 	if (!skb) {
-		err("unable to allocate memory for encoding!\n");
+		dev_err(bcs->cs->dev,
+			"unable to allocate memory for encoding!\n");
 		return -ENOMEM;
 	}
 
diff --git a/drivers/isdn/gigaset/common.c b/drivers/isdn/gigaset/common.c
index 827c32c16795..9d3ce7718e58 100644
--- a/drivers/isdn/gigaset/common.c
+++ b/drivers/isdn/gigaset/common.c
@@ -287,7 +287,7 @@ struct event_t *gigaset_add_event(struct cardstate *cs,
 	tail = cs->ev_tail;
 	next = (tail + 1) % MAX_EVENTS;
 	if (unlikely(next == cs->ev_head))
-		err("event queue full");
+		dev_err(cs->dev, "event queue full\n");
 	else {
 		event = cs->events + tail;
 		event->type = type;
diff --git a/drivers/isdn/gigaset/gigaset.h b/drivers/isdn/gigaset/gigaset.h
index f365993161fc..003752954993 100644
--- a/drivers/isdn/gigaset/gigaset.h
+++ b/drivers/isdn/gigaset/gigaset.h
@@ -106,7 +106,6 @@ enum debuglevel {
 #undef err
 #undef info
 #undef warn
-#undef notice
 
 #define err(format, arg...) printk(KERN_ERR KBUILD_MODNAME ": " \
 	format "\n" , ## arg)
@@ -114,8 +113,6 @@ enum debuglevel {
 	format "\n" , ## arg)
 #define warn(format, arg...) printk(KERN_WARNING KBUILD_MODNAME ": " \
 	format "\n" , ## arg)
-#define notice(format, arg...) printk(KERN_NOTICE KBUILD_MODNAME ": " \
-	format "\n" , ## arg)
 
 #ifdef CONFIG_GIGASET_DEBUG
 
diff --git a/drivers/isdn/gigaset/i4l.c b/drivers/isdn/gigaset/i4l.c
index 9e089f06a942..3c127a8cbaf2 100644
--- a/drivers/isdn/gigaset/i4l.c
+++ b/drivers/isdn/gigaset/i4l.c
@@ -46,7 +46,8 @@ static int writebuf_from_LL(int driverID, int channel, int ack,
 		return -ENODEV;
 	}
 	if (channel < 0 || channel >= cs->channels) {
-		err("%s: invalid channel ID (%d)", __func__, channel);
+		dev_err(cs->dev, "%s: invalid channel ID (%d)\n",
+			__func__, channel);
 		return -ENODEV;
 	}
 	bcs = &cs->bcs[channel];
@@ -58,11 +59,13 @@ static int writebuf_from_LL(int driverID, int channel, int ack,
 
 	if (!len) {
 		if (ack)
-			notice("%s: not ACKing empty packet", __func__);
+			dev_notice(cs->dev, "%s: not ACKing empty packet\n",
+				   __func__);
 		return 0;
 	}
 	if (len > MAX_BUF_SIZE) {
-		err("%s: packet too large (%d bytes)", __func__, len);
+		dev_err(cs->dev, "%s: packet too large (%d bytes)\n",
+			__func__, len);
 		return -EINVAL;
 	}
 
@@ -116,8 +119,7 @@ static int command_from_LL(isdn_ctrl *cntrl)
 	gigaset_debugdrivers();
 
 	if (!cs) {
-		warn("LL tried to access unknown device with nr. %d",
-		     cntrl->driver);
+		err("%s: invalid driver ID (%d)", __func__, cntrl->driver);
 		return -ENODEV;
 	}
 
@@ -126,7 +128,7 @@ static int command_from_LL(isdn_ctrl *cntrl)
 		gig_dbg(DEBUG_ANY, "ISDN_CMD_IOCTL (driver: %d, arg: %ld)",
 			cntrl->driver, cntrl->arg);
 
-		warn("ISDN_CMD_IOCTL is not supported.");
+		dev_warn(cs->dev, "ISDN_CMD_IOCTL not supported\n");
 		return -EINVAL;
 
 	case ISDN_CMD_DIAL:
@@ -138,22 +140,23 @@ static int command_from_LL(isdn_ctrl *cntrl)
 			cntrl->parm.setup.si1, cntrl->parm.setup.si2);
 
 		if (cntrl->arg >= cs->channels) {
-			err("ISDN_CMD_DIAL: invalid channel (%d)",
-			    (int) cntrl->arg);
+			dev_err(cs->dev,
+				"ISDN_CMD_DIAL: invalid channel (%d)\n",
+				(int) cntrl->arg);
 			return -EINVAL;
 		}
 
 		bcs = cs->bcs + cntrl->arg;
 
 		if (!gigaset_get_channel(bcs)) {
-			err("ISDN_CMD_DIAL: channel not free");
+			dev_err(cs->dev, "ISDN_CMD_DIAL: channel not free\n");
 			return -EBUSY;
 		}
 
 		sp = kmalloc(sizeof *sp, GFP_ATOMIC);
 		if (!sp) {
 			gigaset_free_channel(bcs);
-			err("ISDN_CMD_DIAL: out of memory");
+			dev_err(cs->dev, "ISDN_CMD_DIAL: out of memory\n");
 			return -ENOMEM;
 		}
 		*sp = cntrl->parm.setup;
@@ -173,8 +176,9 @@ static int command_from_LL(isdn_ctrl *cntrl)
 		gig_dbg(DEBUG_ANY, "ISDN_CMD_ACCEPTD");
 
 		if (cntrl->arg >= cs->channels) {
-			err("ISDN_CMD_ACCEPTD: invalid channel (%d)",
-			    (int) cntrl->arg);
+			dev_err(cs->dev,
+				"ISDN_CMD_ACCEPTD: invalid channel (%d)\n",
+				(int) cntrl->arg);
 			return -EINVAL;
 		}
 
@@ -196,8 +200,9 @@ static int command_from_LL(isdn_ctrl *cntrl)
 			(int) cntrl->arg);
 
 		if (cntrl->arg >= cs->channels) {
-			err("ISDN_CMD_HANGUP: invalid channel (%u)",
-			    (unsigned) cntrl->arg);
+			dev_err(cs->dev,
+				"ISDN_CMD_HANGUP: invalid channel (%d)\n",
+				(int) cntrl->arg);
 			return -EINVAL;
 		}
 
@@ -224,8 +229,9 @@ static int command_from_LL(isdn_ctrl *cntrl)
 			cntrl->arg & 0xff, (cntrl->arg >> 8));
 
 		if ((cntrl->arg & 0xff) >= cs->channels) {
-			err("ISDN_CMD_SETL2: invalid channel (%u)",
-			    (unsigned) cntrl->arg & 0xff);
+			dev_err(cs->dev,
+				"ISDN_CMD_SETL2: invalid channel (%d)\n",
+				(int) cntrl->arg & 0xff);
 			return -EINVAL;
 		}
 
@@ -244,14 +250,16 @@ static int command_from_LL(isdn_ctrl *cntrl)
 			cntrl->arg & 0xff, (cntrl->arg >> 8));
 
 		if ((cntrl->arg & 0xff) >= cs->channels) {
-			err("ISDN_CMD_SETL3: invalid channel (%u)",
-			    (unsigned) cntrl->arg & 0xff);
+			dev_err(cs->dev,
+				"ISDN_CMD_SETL3: invalid channel (%d)\n",
+				(int) cntrl->arg & 0xff);
 			return -EINVAL;
 		}
 
 		if (cntrl->arg >> 8 != ISDN_PROTO_L3_TRANS) {
-			err("ISDN_CMD_SETL3: invalid protocol %lu",
-			    cntrl->arg >> 8);
+			dev_err(cs->dev,
+				"ISDN_CMD_SETL3: invalid protocol %lu\n",
+				cntrl->arg >> 8);
 			return -EINVAL;
 		}
 
@@ -262,8 +270,9 @@ static int command_from_LL(isdn_ctrl *cntrl)
 	case ISDN_CMD_ALERT:
 		gig_dbg(DEBUG_ANY, "ISDN_CMD_ALERT"); //FIXME
 		if (cntrl->arg >= cs->channels) {
-			err("ISDN_CMD_ALERT: invalid channel (%d)",
-			    (int) cntrl->arg);
+			dev_err(cs->dev,
+				"ISDN_CMD_ALERT: invalid channel (%d)\n",
+				(int) cntrl->arg);
 			return -EINVAL;
 		}
 		//bcs = cs->bcs + cntrl->arg;
@@ -295,7 +304,8 @@ static int command_from_LL(isdn_ctrl *cntrl)
 		gig_dbg(DEBUG_ANY, "ISDN_CMD_GETSIL");
 		break;
 	default:
-		err("unknown command %d from LL", cntrl->command);
+		dev_err(cs->dev, "unknown command %d from LL\n",
+			cntrl->command);
 		return -EINVAL;
 	}
 
diff --git a/drivers/isdn/gigaset/interface.c b/drivers/isdn/gigaset/interface.c
index af195b07c191..521951a898ec 100644
--- a/drivers/isdn/gigaset/interface.c
+++ b/drivers/isdn/gigaset/interface.c
@@ -197,7 +197,7 @@ static void if_close(struct tty_struct *tty, struct file *filp)
 	mutex_lock(&cs->mutex);
 
 	if (!cs->open_count)
-		warn("%s: device not opened", __func__);
+		dev_warn(cs->dev, "%s: device not opened\n", __func__);
 	else {
 		if (!--cs->open_count) {
 			spin_lock_irqsave(&cs->lock, flags);
@@ -232,7 +232,7 @@ static int if_ioctl(struct tty_struct *tty, struct file *file,
 		return -ERESTARTSYS; // FIXME -EINTR?
 
 	if (!cs->open_count)
-		warn("%s: device not opened", __func__);
+		dev_warn(cs->dev, "%s: device not opened\n", __func__);
 	else {
 		retval = 0;
 		switch (cmd) {
@@ -364,9 +364,9 @@ static int if_write(struct tty_struct *tty, const unsigned char *buf, int count)
 		return -ERESTARTSYS; // FIXME -EINTR?
 
 	if (!cs->open_count)
-		warn("%s: device not opened", __func__);
+		dev_warn(cs->dev, "%s: device not opened\n", __func__);
 	else if (cs->mstate != MS_LOCKED) {
-		warn("can't write to unlocked device");
+		dev_warn(cs->dev, "can't write to unlocked device\n");
 		retval = -EBUSY;
 	} else if (!cs->connected) {
 		gig_dbg(DEBUG_ANY, "can't write to unplugged device");
@@ -398,9 +398,9 @@ static int if_write_room(struct tty_struct *tty)
 		return -ERESTARTSYS; // FIXME -EINTR?
 
 	if (!cs->open_count)
-		warn("%s: device not opened", __func__);
+		dev_warn(cs->dev, "%s: device not opened\n", __func__);
 	else if (cs->mstate != MS_LOCKED) {
-		warn("can't write to unlocked device");
+		dev_warn(cs->dev, "can't write to unlocked device\n");
 		retval = -EBUSY;
 	} else if (!cs->connected) {
 		gig_dbg(DEBUG_ANY, "can't write to unplugged device");
@@ -430,9 +430,9 @@ static int if_chars_in_buffer(struct tty_struct *tty)
 		return -ERESTARTSYS; // FIXME -EINTR?
 
 	if (!cs->open_count)
-		warn("%s: device not opened", __func__);
+		dev_warn(cs->dev, "%s: device not opened\n", __func__);
 	else if (cs->mstate != MS_LOCKED) {
-		warn("can't write to unlocked device");
+		dev_warn(cs->dev, "can't write to unlocked device\n");
 		retval = -EBUSY;
 	} else if (!cs->connected) {
 		gig_dbg(DEBUG_ANY, "can't write to unplugged device");
@@ -460,7 +460,7 @@ static void if_throttle(struct tty_struct *tty)
 	mutex_lock(&cs->mutex);
 
 	if (!cs->open_count)
-		warn("%s: device not opened", __func__);
+		dev_warn(cs->dev, "%s: device not opened\n", __func__);
 	else {
 		//FIXME
 	}
@@ -483,7 +483,7 @@ static void if_unthrottle(struct tty_struct *tty)
 	mutex_lock(&cs->mutex);
 
 	if (!cs->open_count)
-		warn("%s: device not opened", __func__);
+		dev_warn(cs->dev, "%s: device not opened\n", __func__);
 	else {
 		//FIXME
 	}
@@ -510,7 +510,7 @@ static void if_set_termios(struct tty_struct *tty, struct ktermios *old)
 	mutex_lock(&cs->mutex);
 
 	if (!cs->open_count) {
-		warn("%s: device not opened", __func__);
+		dev_warn(cs->dev, "%s: device not opened\n", __func__);
 		goto out;
 	}
 
@@ -623,7 +623,8 @@ void gigaset_if_init(struct cardstate *cs)
 	if (!IS_ERR(cs->tty_dev))
 		dev_set_drvdata(cs->tty_dev, cs);
 	else {
-		warn("could not register device to the tty subsystem");
+		dev_warn(cs->dev,
+			 "could not register device to the tty subsystem\n");
 		cs->tty_dev = NULL;
 	}
 	mutex_unlock(&cs->mutex);
diff --git a/drivers/isdn/gigaset/usb-gigaset.c b/drivers/isdn/gigaset/usb-gigaset.c
index 77d20ab0cd4d..4661830a49db 100644
--- a/drivers/isdn/gigaset/usb-gigaset.c
+++ b/drivers/isdn/gigaset/usb-gigaset.c
@@ -498,8 +498,9 @@ static int send_cb(struct cardstate *cs, struct cmdbuf_t *cb)
 
 			if (status) {
 				ucs->busy = 0;
-				err("could not submit urb (error %d)\n",
-				    -status);
+				dev_err(cs->dev,
+					"could not submit urb (error %d)\n",
+					-status);
 				cb->len = 0; /* skip urb => remove cb+wakeup
 						in next loop cycle */
 			}
@@ -670,7 +671,7 @@ static int write_modem(struct cardstate *cs)
 	spin_unlock_irqrestore(&cs->lock, flags);
 
 	if (ret) {
-		err("could not submit urb (error %d)\n", -ret);
+		dev_err(cs->dev, "could not submit urb (error %d)\n", -ret);
 		ucs->busy = 0;
 	}
 

From 5f09c4c797d00bef5700e1ca085b4efcedaf34b8 Mon Sep 17 00:00:00 2001
From: Tilman Schmidt <tilman@imap.cc>
Date: Wed, 23 Jul 2008 21:28:27 -0700
Subject: [PATCH 0303/1171] gigaset: gigaset_isowbuf_getbytes() may return
 signed unnoticed

ifd->offset is unsigned.  gigaset_isowbuf_getbytes() may return signed
unnoticed.  Revised version of patch originally submitted by Roel Kluin
<12o3l@tiscali.nl>.

Signed-off-by: Tilman Schmidt <tilman@imap.cc>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/isdn/gigaset/bas-gigaset.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/isdn/gigaset/bas-gigaset.c b/drivers/isdn/gigaset/bas-gigaset.c
index 5255b5e20e13..3f11910c7ccd 100644
--- a/drivers/isdn/gigaset/bas-gigaset.c
+++ b/drivers/isdn/gigaset/bas-gigaset.c
@@ -1050,10 +1050,9 @@ static int submit_iso_write_urb(struct isow_urbctx_t *ucx)
 		}
 
 		/* retrieve block of data to send */
-		ifd->offset = gigaset_isowbuf_getbytes(ubc->isooutbuf,
-						       ifd->length);
-		if (ifd->offset < 0) {
-			if (ifd->offset == -EBUSY) {
+		rc = gigaset_isowbuf_getbytes(ubc->isooutbuf, ifd->length);
+		if (rc < 0) {
+			if (rc == -EBUSY) {
 				gig_dbg(DEBUG_ISO,
 					"%s: buffer busy at frame %d",
 					__func__, nframe);
@@ -1062,11 +1061,12 @@ static int submit_iso_write_urb(struct isow_urbctx_t *ucx)
 			} else {
 				dev_err(ucx->bcs->cs->dev,
 					"%s: buffer error %d at frame %d\n",
-					__func__, ifd->offset, nframe);
-				return ifd->offset;
+					__func__, rc, nframe);
+				return rc;
 			}
 			break;
 		}
+		ifd->offset = rc;
 		ucx->limit = ubc->isooutbuf->nextread;
 		ifd->status = 0;
 		ifd->actual_length = 0;

From e53f12cc6c43b69f54937f15c5706d83f67c2fdd Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Wed, 23 Jul 2008 21:28:28 -0700
Subject: [PATCH 0304/1171] remove include/asm-h8300/keyboard.h

This patch removes the unused include/asm-h8300/keyboard.h

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Acked-by: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/asm-h8300/keyboard.h | 24 ------------------------
 1 file changed, 24 deletions(-)
 delete mode 100644 include/asm-h8300/keyboard.h

diff --git a/include/asm-h8300/keyboard.h b/include/asm-h8300/keyboard.h
deleted file mode 100644
index 90efbd655390..000000000000
--- a/include/asm-h8300/keyboard.h
+++ /dev/null
@@ -1,24 +0,0 @@
-/*
- *  linux/include/asm-h8300/keyboard.h
- *  Created 04 Dec 2001 by Khaled Hassounah <khassounah@mediumware.net>
- *  This file contains the Dragonball architecture specific keyboard definitions
- */
-
-#ifndef _H8300_KEYBOARD_H
-#define _H8300_KEYBOARD_H
-
-
-/* dummy i.e. no real keyboard */
-#define kbd_setkeycode(x...)	(-ENOSYS)
-#define kbd_getkeycode(x...)	(-ENOSYS)
-#define kbd_translate(x...)	(0)
-#define kbd_unexpected_up(x...)	(1)
-#define kbd_leds(x...)		do {;} while (0)
-#define kbd_init_hw(x...)	do {;} while (0)
-#define kbd_enable_irq(x...)	do {;} while (0)
-#define kbd_disable_irq(x...)	do {;} while (0)
-
-#endif  /* _H8300_KEYBOARD_H */
-
-
-

From fb9ba4e95921f71d874beee2d0964fc2322b47a2 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Wed, 23 Jul 2008 21:28:29 -0700
Subject: [PATCH 0305/1171] alpha: remove the unused ALPHA_CORE_AGP option

The real option is named AGP_ALPHA_CORE.

Reviewed-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/alpha/Kconfig | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/arch/alpha/Kconfig b/arch/alpha/Kconfig
index dbe8c280fea9..1bec55d63ef6 100644
--- a/arch/alpha/Kconfig
+++ b/arch/alpha/Kconfig
@@ -333,11 +333,6 @@ config PCI_SYSCALL
 config IOMMU_HELPER
 	def_bool PCI
 
-config ALPHA_CORE_AGP
-	bool
-	depends on ALPHA_GENERIC || ALPHA_TITAN || ALPHA_MARVEL
-	default y
-
 config ALPHA_NONAME
 	bool
 	depends on ALPHA_BOOK1 || ALPHA_NONAME_CH

From 0d63081d418c73cc187c893069e0f24c4c6eecd3 Mon Sep 17 00:00:00 2001
From: Pavel Machek <pavel@suse.cz>
Date: Wed, 23 Jul 2008 21:28:32 -0700
Subject: [PATCH 0306/1171] swsusp: provide users with a hint about the
 no_console_suspend option

Tell the user about the no_console_suspend option, so that we don't have to
tell each bug reporter personally.

[akpm@linux-foundation.org: clarify the text a little]
Signed-off-by: Pavel Machek <pavel@suse.cz>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/printk.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/printk.c b/kernel/printk.c
index 07ad9e7f7a66..3f7a2a94583b 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -933,7 +933,7 @@ void suspend_console(void)
 {
 	if (!console_suspend_enabled)
 		return;
-	printk("Suspending console(s)\n");
+	printk("Suspending console(s) (use no_console_suspend to debug)\n");
 	acquire_console_sem();
 	console_suspended = 1;
 }

From 77437fd4e61f87cc94d9314baa5cbf50e3ccdf54 Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Wed, 23 Jul 2008 21:28:33 -0700
Subject: [PATCH 0307/1171] pm: boot time suspend selftest

Boot-time test for system suspend states (STR or standby).  The generic
RTC framework triggers wakeup alarms, which are used to exit those states.

  - Measures some aspects of suspend time ... this uses "jiffies" until
    someone converts it to use a timebase that works properly even while
    timer IRQs are disabled.

  - Triggered by a command line parameter.  By default nothing even
    vaguely troublesome will happen, but "test_suspend=mem" will give
    you a brief STR test during system boot.  (Or you may need to use
    "test_suspend=standby" instead, if your hardware needs that.)

This isn't without problems.  It fires early enough during boot that for
example both PCMCIA and MMC stacks have misbehaved.  The workaround in
those cases was to boot without such media cards inserted.

[matthltc@us.ibm.com: fix compile failure in boot time suspend selftest]
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Pavel Machek <pavel@suse.cz>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: Matt Helsley <matthltc@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/kernel-parameters.txt |   9 +-
 kernel/power/Kconfig                |  11 ++
 kernel/power/main.c                 | 194 +++++++++++++++++++++++++++-
 3 files changed, 212 insertions(+), 2 deletions(-)

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 01a2992b5754..4d705713cabc 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -87,7 +87,8 @@ parameter is applicable:
 	SH	SuperH architecture is enabled.
 	SMP	The kernel is an SMP kernel.
 	SPARC	Sparc architecture is enabled.
-	SWSUSP	Software suspend is enabled.
+	SWSUSP	Software suspend (hibernation) is enabled.
+	SUSPEND	System suspend states are enabled.
 	TS	Appropriate touchscreen support is enabled.
 	USB	USB support is enabled.
 	USBHID	USB Human Interface Device support is enabled.
@@ -2123,6 +2124,12 @@ and is between 256 and 4096 characters. It is defined in the file
 
 	tdfx=		[HW,DRM]
 
+	test_suspend=	[SUSPEND]
+			Specify "mem" (for Suspend-to-RAM) or "standby" (for
+			standby suspend) as the system sleep state to briefly
+			enter during system startup.  The system is woken from
+			this state using a wakeup-capable RTC alarm.
+
 	thash_entries=	[KNL,NET]
 			Set number of hash buckets for TCP connection
 
diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig
index 59dfdf1e1d20..dcd165f92a88 100644
--- a/kernel/power/Kconfig
+++ b/kernel/power/Kconfig
@@ -94,6 +94,17 @@ config SUSPEND
 	  powered and thus its contents are preserved, such as the
 	  suspend-to-RAM state (e.g. the ACPI S3 state).
 
+config PM_TEST_SUSPEND
+	bool "Test suspend/resume and wakealarm during bootup"
+	depends on SUSPEND && PM_DEBUG && RTC_LIB=y
+	---help---
+	This option will let you suspend your machine during bootup, and
+	make it wake up a few seconds later using an RTC wakeup alarm.
+	Enable this with a kernel parameter like "test_suspend=mem".
+
+	You probably want to have your system's RTC driver statically
+	linked, ensuring that it's available when this test runs.
+
 config SUSPEND_FREEZER
 	bool "Enable freezer for suspend to RAM/standby" \
 		if ARCH_WANTS_FREEZER_CONTROL || BROKEN
diff --git a/kernel/power/main.c b/kernel/power/main.c
index 3398f4651aa1..95bff23ecdaa 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -132,6 +132,61 @@ static inline int suspend_test(int level) { return 0; }
 
 #ifdef CONFIG_SUSPEND
 
+#ifdef CONFIG_PM_TEST_SUSPEND
+
+/*
+ * We test the system suspend code by setting an RTC wakealarm a short
+ * time in the future, then suspending.  Suspending the devices won't
+ * normally take long ... some systems only need a few milliseconds.
+ *
+ * The time it takes is system-specific though, so when we test this
+ * during system bootup we allow a LOT of time.
+ */
+#define TEST_SUSPEND_SECONDS	5
+
+static unsigned long suspend_test_start_time;
+
+static void suspend_test_start(void)
+{
+	/* FIXME Use better timebase than "jiffies", ideally a clocksource.
+	 * What we want is a hardware counter that will work correctly even
+	 * during the irqs-are-off stages of the suspend/resume cycle...
+	 */
+	suspend_test_start_time = jiffies;
+}
+
+static void suspend_test_finish(const char *label)
+{
+	long nj = jiffies - suspend_test_start_time;
+	unsigned msec;
+
+	msec = jiffies_to_msecs(abs(nj));
+	pr_info("PM: %s took %d.%03d seconds\n", label,
+			msec / 1000, msec % 1000);
+
+	/* Warning on suspend means the RTC alarm period needs to be
+	 * larger -- the system was sooo slooowwww to suspend that the
+	 * alarm (should have) fired before the system went to sleep!
+	 *
+	 * Warning on either suspend or resume also means the system
+	 * has some performance issues.  The stack dump of a WARN_ON
+	 * is more likely to get the right attention than a printk...
+	 */
+	WARN_ON(msec > (TEST_SUSPEND_SECONDS * 1000));
+}
+
+#else
+
+static void suspend_test_start(void)
+{
+}
+
+static void suspend_test_finish(const char *label)
+{
+}
+
+#endif
+
 /* This is just an arbitrary number */
 #define FREE_PAGE_NUMBER (100)
 
@@ -266,12 +321,13 @@ int suspend_devices_and_enter(suspend_state_t state)
 			goto Close;
 	}
 	suspend_console();
+	suspend_test_start();
 	error = device_suspend(PMSG_SUSPEND);
 	if (error) {
 		printk(KERN_ERR "PM: Some devices failed to suspend\n");
 		goto Recover_platform;
 	}
-
+	suspend_test_finish("suspend devices");
 	if (suspend_test(TEST_DEVICES))
 		goto Recover_platform;
 
@@ -293,7 +349,9 @@ int suspend_devices_and_enter(suspend_state_t state)
 	if (suspend_ops->finish)
 		suspend_ops->finish();
  Resume_devices:
+	suspend_test_start();
 	device_resume(PMSG_RESUME);
+	suspend_test_finish("resume devices");
 	resume_console();
  Close:
 	if (suspend_ops->end)
@@ -521,3 +579,137 @@ static int __init pm_init(void)
 }
 
 core_initcall(pm_init);
+
+
+#ifdef CONFIG_PM_TEST_SUSPEND
+
+#include <linux/rtc.h>
+
+/*
+ * To test system suspend, we need a hands-off mechanism to resume the
+ * system.  RTCs wake alarms are a common self-contained mechanism.
+ */
+
+static void __init test_wakealarm(struct rtc_device *rtc, suspend_state_t state)
+{
+	static char err_readtime[] __initdata =
+		KERN_ERR "PM: can't read %s time, err %d\n";
+	static char err_wakealarm [] __initdata =
+		KERN_ERR "PM: can't set %s wakealarm, err %d\n";
+	static char err_suspend[] __initdata =
+		KERN_ERR "PM: suspend test failed, error %d\n";
+	static char info_test[] __initdata =
+		KERN_INFO "PM: test RTC wakeup from '%s' suspend\n";
+
+	unsigned long		now;
+	struct rtc_wkalrm	alm;
+	int			status;
+
+	/* this may fail if the RTC hasn't been initialized */
+	status = rtc_read_time(rtc, &alm.time);
+	if (status < 0) {
+		printk(err_readtime, rtc->dev.bus_id, status);
+		return;
+	}
+	rtc_tm_to_time(&alm.time, &now);
+
+	memset(&alm, 0, sizeof alm);
+	rtc_time_to_tm(now + TEST_SUSPEND_SECONDS, &alm.time);
+	alm.enabled = true;
+
+	status = rtc_set_alarm(rtc, &alm);
+	if (status < 0) {
+		printk(err_wakealarm, rtc->dev.bus_id, status);
+		return;
+	}
+
+	if (state == PM_SUSPEND_MEM) {
+		printk(info_test, pm_states[state]);
+		status = pm_suspend(state);
+		if (status == -ENODEV)
+			state = PM_SUSPEND_STANDBY;
+	}
+	if (state == PM_SUSPEND_STANDBY) {
+		printk(info_test, pm_states[state]);
+		status = pm_suspend(state);
+	}
+	if (status < 0)
+		printk(err_suspend, status);
+}
+
+static int __init has_wakealarm(struct device *dev, void *name_ptr)
+{
+	struct rtc_device *candidate = to_rtc_device(dev);
+
+	if (!candidate->ops->set_alarm)
+		return 0;
+	if (!device_may_wakeup(candidate->dev.parent))
+		return 0;
+
+	*(char **)name_ptr = dev->bus_id;
+	return 1;
+}
+
+/*
+ * Kernel options like "test_suspend=mem" force suspend/resume sanity tests
+ * at startup time.  They're normally disabled, for faster boot and because
+ * we can't know which states really work on this particular system.
+ */
+static suspend_state_t test_state __initdata = PM_SUSPEND_ON;
+
+static char warn_bad_state[] __initdata =
+	KERN_WARNING "PM: can't test '%s' suspend state\n";
+
+static int __init setup_test_suspend(char *value)
+{
+	unsigned i;
+
+	/* "=mem" ==> "mem" */
+	value++;
+	for (i = 0; i < PM_SUSPEND_MAX; i++) {
+		if (!pm_states[i])
+			continue;
+		if (strcmp(pm_states[i], value) != 0)
+			continue;
+		test_state = (__force suspend_state_t) i;
+		return 0;
+	}
+	printk(warn_bad_state, value);
+	return 0;
+}
+__setup("test_suspend", setup_test_suspend);
+
+static int __init test_suspend(void)
+{
+	static char		warn_no_rtc[] __initdata =
+		KERN_WARNING "PM: no wakealarm-capable RTC driver is ready\n";
+
+	char			*pony = NULL;
+	struct rtc_device	*rtc = NULL;
+
+	/* PM is initialized by now; is that state testable? */
+	if (test_state == PM_SUSPEND_ON)
+		goto done;
+	if (!valid_state(test_state)) {
+		printk(warn_bad_state, pm_states[test_state]);
+		goto done;
+	}
+
+	/* RTCs have initialized by now too ... can we use one? */
+	class_find_device(rtc_class, NULL, &pony, has_wakealarm);
+	if (pony)
+		rtc = rtc_class_open(pony);
+	if (!rtc) {
+		printk(warn_no_rtc);
+		goto done;
+	}
+
+	/* go for it */
+	test_wakealarm(rtc, test_state);
+	rtc_class_close(rtc);
+done:
+	return 0;
+}
+late_initcall(test_suspend);
+
+#endif /* CONFIG_PM_TEST_SUSPEND */

From d75f65fd247fe85d90a3880d143b1bb22fe13a48 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Wed, 23 Jul 2008 21:28:34 -0700
Subject: [PATCH 0308/1171] remove include/linux/pm_legacy.h

Remove the obsolete and no longer used include/linux/pm_legacy.h

Reviewed-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Cc: Pavel Machek <pavel@suse.cz>
Acked-by: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/frv/kernel/pm.c            |  1 -
 arch/mips/au1000/common/power.c |  1 -
 arch/x86/kernel/apm_32.c        |  1 -
 include/linux/pm_legacy.h       | 35 ---------------------------------
 4 files changed, 38 deletions(-)
 delete mode 100644 include/linux/pm_legacy.h

diff --git a/arch/frv/kernel/pm.c b/arch/frv/kernel/pm.c
index 73f3aeefd203..d1113c5031f5 100644
--- a/arch/frv/kernel/pm.c
+++ b/arch/frv/kernel/pm.c
@@ -14,7 +14,6 @@
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/pm.h>
-#include <linux/pm_legacy.h>
 #include <linux/sched.h>
 #include <linux/interrupt.h>
 #include <linux/sysctl.h>
diff --git a/arch/mips/au1000/common/power.c b/arch/mips/au1000/common/power.c
index 2166b9e1e80c..bd854a6d1d89 100644
--- a/arch/mips/au1000/common/power.c
+++ b/arch/mips/au1000/common/power.c
@@ -31,7 +31,6 @@
 
 #include <linux/init.h>
 #include <linux/pm.h>
-#include <linux/pm_legacy.h>
 #include <linux/sysctl.h>
 #include <linux/jiffies.h>
 
diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c
index bf9b441331e9..9ee24e6bc4b0 100644
--- a/arch/x86/kernel/apm_32.c
+++ b/arch/x86/kernel/apm_32.c
@@ -219,7 +219,6 @@
 #include <linux/time.h>
 #include <linux/sched.h>
 #include <linux/pm.h>
-#include <linux/pm_legacy.h>
 #include <linux/capability.h>
 #include <linux/device.h>
 #include <linux/kernel.h>
diff --git a/include/linux/pm_legacy.h b/include/linux/pm_legacy.h
deleted file mode 100644
index 446f4f42b952..000000000000
--- a/include/linux/pm_legacy.h
+++ /dev/null
@@ -1,35 +0,0 @@
-#ifndef __LINUX_PM_LEGACY_H__
-#define __LINUX_PM_LEGACY_H__
-
-
-#ifdef CONFIG_PM_LEGACY
-
-/*
- * Register a device with power management
- */
-struct pm_dev __deprecated *
-pm_register(pm_dev_t type, unsigned long id, pm_callback callback);
-
-/*
- * Send a request to all devices
- */
-int __deprecated pm_send_all(pm_request_t rqst, void *data);
-
-#else /* CONFIG_PM_LEGACY */
-
-static inline struct pm_dev *pm_register(pm_dev_t type,
-					 unsigned long id,
-					 pm_callback callback)
-{
-	return NULL;
-}
-
-static inline int pm_send_all(pm_request_t rqst, void *data)
-{
-	return 0;
-}
-
-#endif /* CONFIG_PM_LEGACY */
-
-#endif /* __LINUX_PM_LEGACY_H__ */
-

From 558481f038e587b22d02167af58914c814ce9de5 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Wed, 23 Jul 2008 21:28:35 -0700
Subject: [PATCH 0309/1171] pm: remove definition of struct pm_dev

Remove the definition of 'struct pm_dev', which is not used any more,
along with some related stuff from include/linux/pm.h .

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/pm.h | 24 ------------------------
 1 file changed, 24 deletions(-)

diff --git a/include/linux/pm.h b/include/linux/pm.h
index 4ad9de94449a..5bf1ce89cfbb 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -68,30 +68,6 @@ enum
  */
 #define PM_PCI_ID(dev) ((dev)->bus->number << 16 | (dev)->devfn)
 
-/*
- * Request handler callback
- */
-struct pm_dev;
-
-typedef int (*pm_callback)(struct pm_dev *dev, pm_request_t rqst, void *data);
-
-/*
- * Dynamic device information
- */
-struct pm_dev
-{
-	pm_dev_t	 type;
-	unsigned long	 id;
-	pm_callback	 callback;
-	void		*data;
-
-	unsigned long	 flags;
-	unsigned long	 state;
-	unsigned long	 prev_state;
-
-	struct list_head entry;
-};
-
 /* Functions above this comment are list-based old-style power
  * management. Please avoid using them.  */
 

From e7ecb331e11d1f7aa66aeef9170fc20781c9bb55 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Wed, 23 Jul 2008 21:28:35 -0700
Subject: [PATCH 0310/1171] pm: remove remaining obsolete definitions from pm.h

Remove the remaining obsolete definitions from include/linux/pm.h and move
the definitions of PM_SUSPEND and PM_RESUME to the header of h3600 which
is the only user of them.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/asm-arm/arch-sa1100/h3600.h |  5 ++++
 include/linux/pm.h                  | 46 -----------------------------
 2 files changed, 5 insertions(+), 46 deletions(-)

diff --git a/include/asm-arm/arch-sa1100/h3600.h b/include/asm-arm/arch-sa1100/h3600.h
index 1b6355971574..3ca0ecf095e6 100644
--- a/include/asm-arm/arch-sa1100/h3600.h
+++ b/include/asm-arm/arch-sa1100/h3600.h
@@ -23,6 +23,11 @@
 #ifndef _INCLUDE_H3600_H_
 #define _INCLUDE_H3600_H_
 
+typedef int __bitwise pm_request_t;
+
+#define PM_SUSPEND	((__force pm_request_t) 1)	/* enter D1-D3 */
+#define PM_RESUME	((__force pm_request_t) 2)	/* enter D0 */
+
 /* generalized support for H3xxx series Compaq Pocket PC's */
 #define machine_is_h3xxx() (machine_is_h3100() || machine_is_h3600() || machine_is_h3800())
 
diff --git a/include/linux/pm.h b/include/linux/pm.h
index 5bf1ce89cfbb..390dd95a375e 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -25,52 +25,6 @@
 #include <asm/atomic.h>
 #include <asm/errno.h>
 
-/*
- * Power management requests... these are passed to pm_send_all() and friends.
- *
- * these functions are old and deprecated, see below.
- */
-typedef int __bitwise pm_request_t;
-
-#define PM_SUSPEND	((__force pm_request_t) 1)	/* enter D1-D3 */
-#define PM_RESUME	((__force pm_request_t) 2)	/* enter D0 */
-
-
-/*
- * Device types... these are passed to pm_register
- */
-typedef int __bitwise pm_dev_t;
-
-#define PM_UNKNOWN_DEV	((__force pm_dev_t) 0)	/* generic */
-#define PM_SYS_DEV	((__force pm_dev_t) 1)	/* system device (fan, KB controller, ...) */
-#define PM_PCI_DEV	((__force pm_dev_t) 2)	/* PCI device */
-#define PM_USB_DEV	((__force pm_dev_t) 3)	/* USB device */
-#define PM_SCSI_DEV	((__force pm_dev_t) 4)	/* SCSI device */
-#define PM_ISA_DEV	((__force pm_dev_t) 5)	/* ISA device */
-#define	PM_MTD_DEV	((__force pm_dev_t) 6)	/* Memory Technology Device */
-
-/*
- * System device hardware ID (PnP) values
- */
-enum
-{
-	PM_SYS_UNKNOWN = 0x00000000, /* generic */
-	PM_SYS_KBC =	 0x41d00303, /* keyboard controller */
-	PM_SYS_COM =	 0x41d00500, /* serial port */
-	PM_SYS_IRDA =	 0x41d00510, /* IRDA controller */
-	PM_SYS_FDC =	 0x41d00700, /* floppy controller */
-	PM_SYS_VGA =	 0x41d00900, /* VGA controller */
-	PM_SYS_PCMCIA =	 0x41d00e00, /* PCMCIA controller */
-};
-
-/*
- * Device identifier
- */
-#define PM_PCI_ID(dev) ((dev)->bus->number << 16 | (dev)->devfn)
-
-/* Functions above this comment are list-based old-style power
- * management. Please avoid using them.  */
-
 /*
  * Callbacks for platform drivers to implement.
  */

From 40b4ac33b4d1bdd5cbeb2241be2399c550fa3696 Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Wed, 23 Jul 2008 21:28:36 -0700
Subject: [PATCH 0311/1171] pm: remove obsolete piece of PM documentation

Remove some obsolete PM documentation.

The majority of contents of Documentation/power/pm.txt are
outdated.  Remove the outdated parts of this file and move the rest
to Documentation/power/apm-acpi.txt .  Update the index in
Documentation/power/ as appropriate.

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Acked-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/power/00-INDEX     |   4 +-
 Documentation/power/apm-acpi.txt |  32 ++++
 Documentation/power/pm.txt       | 257 -------------------------------
 3 files changed, 34 insertions(+), 259 deletions(-)
 create mode 100644 Documentation/power/apm-acpi.txt
 delete mode 100644 Documentation/power/pm.txt

diff --git a/Documentation/power/00-INDEX b/Documentation/power/00-INDEX
index a55d7f1c836d..fb742c213c9e 100644
--- a/Documentation/power/00-INDEX
+++ b/Documentation/power/00-INDEX
@@ -1,5 +1,7 @@
 00-INDEX
 	- This file
+apm-acpi.txt
+	- basic info about the APM and ACPI support.
 basic-pm-debugging.txt
 	- Debugging suspend and resume
 devices.txt
@@ -14,8 +16,6 @@ notifiers.txt
 	- Registering suspend notifiers in device drivers
 pci.txt
 	- How the PCI Subsystem Does Power Management
-pm.txt
-	- info on Linux power management support.
 pm_qos_interface.txt
 	- info on Linux PM Quality of Service interface
 power_supply_class.txt
diff --git a/Documentation/power/apm-acpi.txt b/Documentation/power/apm-acpi.txt
new file mode 100644
index 000000000000..1bd799dc17e8
--- /dev/null
+++ b/Documentation/power/apm-acpi.txt
@@ -0,0 +1,32 @@
+APM or ACPI?
+------------
+If you have a relatively recent x86 mobile, desktop, or server system,
+odds are it supports either Advanced Power Management (APM) or
+Advanced Configuration and Power Interface (ACPI).  ACPI is the newer
+of the two technologies and puts power management in the hands of the
+operating system, allowing for more intelligent power management than
+is possible with BIOS controlled APM.
+
+The best way to determine which, if either, your system supports is to
+build a kernel with both ACPI and APM enabled (as of 2.3.x ACPI is
+enabled by default).  If a working ACPI implementation is found, the
+ACPI driver will override and disable APM, otherwise the APM driver
+will be used.
+
+No, sorry, you cannot have both ACPI and APM enabled and running at
+once.  Some people with broken ACPI or broken APM implementations
+would like to use both to get a full set of working features, but you
+simply cannot mix and match the two.  Only one power management
+interface can be in control of the machine at once.  Think about it..
+
+User-space Daemons
+------------------
+Both APM and ACPI rely on user-space daemons, apmd and acpid
+respectively, to be completely functional.  Obtain both of these
+daemons from your Linux distribution or from the Internet (see below)
+and be sure that they are started sometime in the system boot process.
+Go ahead and start both.  If ACPI or APM is not available on your
+system the associated daemon will exit gracefully.
+
+  apmd:   http://worldvisions.ca/~apenwarr/apmd/
+  acpid:  http://acpid.sf.net/
diff --git a/Documentation/power/pm.txt b/Documentation/power/pm.txt
deleted file mode 100644
index be841507e43f..000000000000
--- a/Documentation/power/pm.txt
+++ /dev/null
@@ -1,257 +0,0 @@
-               Linux Power Management Support
-
-This document briefly describes how to use power management with your
-Linux system and how to add power management support to Linux drivers.
-
-APM or ACPI?
-------------
-If you have a relatively recent x86 mobile, desktop, or server system,
-odds are it supports either Advanced Power Management (APM) or
-Advanced Configuration and Power Interface (ACPI).  ACPI is the newer
-of the two technologies and puts power management in the hands of the
-operating system, allowing for more intelligent power management than
-is possible with BIOS controlled APM.
-
-The best way to determine which, if either, your system supports is to
-build a kernel with both ACPI and APM enabled (as of 2.3.x ACPI is
-enabled by default).  If a working ACPI implementation is found, the
-ACPI driver will override and disable APM, otherwise the APM driver
-will be used.
-
-No, sorry, you cannot have both ACPI and APM enabled and running at
-once.  Some people with broken ACPI or broken APM implementations
-would like to use both to get a full set of working features, but you
-simply cannot mix and match the two.  Only one power management
-interface can be in control of the machine at once.  Think about it..
-
-User-space Daemons
-------------------
-Both APM and ACPI rely on user-space daemons, apmd and acpid
-respectively, to be completely functional.  Obtain both of these
-daemons from your Linux distribution or from the Internet (see below)
-and be sure that they are started sometime in the system boot process.
-Go ahead and start both.  If ACPI or APM is not available on your
-system the associated daemon will exit gracefully.
-
-  apmd:   http://worldvisions.ca/~apenwarr/apmd/
-  acpid:  http://acpid.sf.net/
-
-Driver Interface -- OBSOLETE, DO NOT USE!
-----------------*************************
-
-Note: pm_register(), pm_access(), pm_dev_idle() and friends are
-obsolete. Please do not use them. Instead you should properly hook
-your driver into the driver model, and use its suspend()/resume()
-callbacks to do this kind of stuff.
-
-If you are writing a new driver or maintaining an old driver, it
-should include power management support.  Without power management
-support, a single driver may prevent a system with power management
-capabilities from ever being able to suspend (safely).
-
-Overview:
-1) Register each instance of a device with "pm_register"
-2) Call "pm_access" before accessing the hardware.
-   (this will ensure that the hardware is awake and ready)
-3) Your "pm_callback" is called before going into a
-   suspend state (ACPI D1-D3) or after resuming (ACPI D0)
-   from a suspend.
-4) Call "pm_dev_idle" when the device is not being used
-   (optional but will improve device idle detection)
-5) When unloaded, unregister the device with "pm_unregister"
-
-/*
- * Description: Register a device with the power-management subsystem
- *
- * Parameters:
- *   type - device type (PCI device, system device, ...)
- *   id - instance number or unique identifier
- *   cback - request handler callback (suspend, resume, ...)
- *
- * Returns: Registered PM device or NULL on error
- *
- * Examples:
- *   dev = pm_register(PM_SYS_DEV, PM_SYS_VGA, vga_callback);
- *
- *   struct pci_dev *pci_dev = pci_find_dev(...);
- *   dev = pm_register(PM_PCI_DEV, PM_PCI_ID(pci_dev), callback);
- */
-struct pm_dev *pm_register(pm_dev_t type, unsigned long id, pm_callback cback);
-
-/*
- * Description: Unregister a device with the power management subsystem
- *
- * Parameters:
- *   dev - PM device previously returned from pm_register
- */
-void pm_unregister(struct pm_dev *dev);
-
-/*
- * Description: Unregister all devices with a matching callback function
- *
- * Parameters:
- *   cback - previously registered request callback
- *
- * Notes: Provided for easier porting from old APM interface
- */
-void pm_unregister_all(pm_callback cback);
-
-/*
- * Power management request callback
- *
- * Parameters:
- *   dev - PM device previously returned from pm_register
- *   rqst - request type
- *   data - data, if any, associated with the request
- *
- * Returns: 0 if the request is successful
- *          EINVAL if the request is not supported
- *          EBUSY if the device is now busy and cannot handle the request
- *          ENOMEM if the device was unable to handle the request due to memory
- *
- * Details: The device request callback will be called before the
- *          device/system enters a suspend state (ACPI D1-D3) or
- *          or after the device/system resumes from suspend (ACPI D0).
- *          For PM_SUSPEND, the ACPI D-state being entered is passed
- *          as the "data" argument to the callback.  The device
- *          driver should save (PM_SUSPEND) or restore (PM_RESUME)
- *          device context when the request callback is called.
- *
- *          Once a driver returns 0 (success) from a suspend
- *          request, it should not process any further requests or
- *          access the device hardware until a call to "pm_access" is made.
- */
-typedef int (*pm_callback)(struct pm_dev *dev, pm_request_t rqst, void *data);
-
-Driver Details
---------------
-This is just a quick Q&A as a stopgap until a real driver writers'
-power management guide is available.
-
-Q: When is a device suspended?
-
-Devices can be suspended based on direct user request (eg. laptop lid
-closes), system power policy (eg.  sleep after 30 minutes of console
-inactivity), or device power policy (eg. power down device after 5
-minutes of inactivity)
-
-Q: Must a driver honor a suspend request?
-
-No, a driver can return -EBUSY from a suspend request and this
-will stop the system from suspending.  When a suspend request
-fails, all suspended devices are resumed and the system continues
-to run.  Suspend can be retried at a later time.
-
-Q: Can the driver block suspend/resume requests?
-
-Yes, a driver can delay its return from a suspend or resume
-request until the device is ready to handle requests.  It
-is advantageous to return as quickly as possible from a
-request as suspend/resume are done serially.
-
-Q: What context is a suspend/resume initiated from?
-
-A suspend or resume is initiated from a kernel thread context.
-It is safe to block, allocate memory, initiate requests
-or anything else you can do within the kernel.
-
-Q: Will requests continue to arrive after a suspend?
-
-Possibly.  It is the driver's responsibility to queue(*),
-fail, or drop any requests that arrive after returning
-success to a suspend request.  It is important that the
-driver not access its device until after it receives
-a resume request as the device's bus may no longer
-be active.
-
-(*) If a driver queues requests for processing after
-    resume be aware that the device, network, etc.
-    might be in a different state than at suspend time.
-    It's probably better to drop requests unless
-    the driver is a storage device.
-
-Q: Do I have to manage bus-specific power management registers
-
-No.  It is the responsibility of the bus driver to manage
-PCI, USB, etc. power management registers.  The bus driver
-or the power management subsystem will also enable any
-wake-on functionality that the device has.
-
-Q: So, really, what do I need to do to support suspend/resume?
-
-You need to save any device context that would
-be lost if the device was powered off and then restore
-it at resume time.  When ACPI is active, there are
-three levels of device suspend states; D1, D2, and D3.
-(The suspend state is passed as the "data" argument
-to the device callback.)  With D3, the device is powered
-off and loses all context, D1 and D2 are shallower power
-states and require less device context to be saved.  To
-play it safe, just save everything at suspend and restore
-everything at resume.
-
-Q: Where do I store device context for suspend?
-
-Anywhere in memory, kmalloc a buffer or store it
-in the device descriptor.  You are guaranteed that the
-contents of memory will be restored and accessible
-before resume, even when the system suspends to disk.
-
-Q: What do I need to do for ACPI vs. APM vs. etc?
-
-Drivers need not be aware of the specific power management
-technology that is active.  They just need to be aware
-of when the overlying power management system requests
-that they suspend or resume.
-
-Q: What about device dependencies?
-
-When a driver registers a device, the power management
-subsystem uses the information provided to build a
-tree of device dependencies (eg. USB device X is on
-USB controller Y which is on PCI bus Z)  When power
-management wants to suspend a device, it first sends
-a suspend request to its driver, then the bus driver,
-and so on up to the system bus.  Device resumes
-proceed in the opposite direction.
-
-Q: Who do I contact for additional information about
-   enabling power management for my specific driver/device?
-
-ACPI Development mailing list: linux-acpi@vger.kernel.org
-
-System Interface -- OBSOLETE, DO NOT USE!
-----------------*************************
-If you are providing new power management support to Linux (ie.
-adding support for something like APM or ACPI), you should
-communicate with drivers through the existing generic power
-management interface.
-
-/*
- * Send a request to all devices
- *
- * Parameters:
- *   rqst - request type
- *   data - data, if any, associated with the request
- *
- * Returns: 0 if the request is successful
- *          See "pm_callback" return for errors
- *
- * Details: Walk list of registered devices and call pm_send
- *          for each until complete or an error is encountered.
- *          If an error is encountered for a suspend request,
- *          return all devices to the state they were in before
- *          the suspend request.
- */
-int pm_send_all(pm_request_t rqst, void *data);
-
-/*
- * Find a matching device
- *
- * Parameters:
- *   type - device type (PCI device, system device, or 0 to match all devices)
- *   from - previous match or NULL to start from the beginning
- *
- * Returns: Matching device or NULL if none found
- */
-struct pm_dev *pm_find(pm_dev_t type, struct pm_dev *from);

From 8c363265d57d755e62053e9f69a1f2164e83f7ea Mon Sep 17 00:00:00 2001
From: "Rafael J. Wysocki" <rjw@sisk.pl>
Date: Wed, 23 Jul 2008 21:28:37 -0700
Subject: [PATCH 0312/1171] pm: drop unnecessary includes from pm.h

Drop unnecessary includes from include/linux/pm.h .

Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/pm.h | 2 --
 1 file changed, 2 deletions(-)

diff --git a/include/linux/pm.h b/include/linux/pm.h
index 390dd95a375e..ed98d967f9fb 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -22,8 +22,6 @@
 #define _LINUX_PM_H
 
 #include <linux/list.h>
-#include <asm/atomic.h>
-#include <asm/errno.h>
 
 /*
  * Callbacks for platform drivers to implement.

From 8111d1b552349921aae1acf73e4e8cea98e80970 Mon Sep 17 00:00:00 2001
From: Alan Stern <stern@rowland.harvard.edu>
Date: Wed, 23 Jul 2008 21:28:37 -0700
Subject: [PATCH 0313/1171] pm: add new PM_EVENT codes for runtime power
 transitions

This patch (as1112) adds some new PM_EVENT_* codes for use by kernel
subsystems.  They describe runtime power-state transitions of the sort already
implemented by the USB subsystem.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/pm.h | 37 +++++++++++++++++++++++++++++++++++--
 1 file changed, 35 insertions(+), 2 deletions(-)

diff --git a/include/linux/pm.h b/include/linux/pm.h
index ed98d967f9fb..4dcce54b6d76 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -245,6 +245,21 @@ struct pm_ext_ops {
  * RECOVER	Creation of a hibernation image or restoration of the main
  *		memory contents from a hibernation image has failed, call
  *		->thaw() and ->complete() for all devices.
+ *
+ * The following PM_EVENT_ messages are defined for internal use by
+ * kernel subsystems.  They are never issued by the PM core.
+ *
+ * USER_SUSPEND		Manual selective suspend was issued by userspace.
+ *
+ * USER_RESUME		Manual selective resume was issued by userspace.
+ *
+ * REMOTE_WAKEUP	Remote-wakeup request was received from the device.
+ *
+ * AUTO_SUSPEND		Automatic (device idle) runtime suspend was
+ *			initiated by the subsystem.
+ *
+ * AUTO_RESUME		Automatic (device needed) runtime resume was
+ *			requested by a driver.
  */
 
 #define PM_EVENT_ON		0x0000
@@ -256,9 +271,18 @@ struct pm_ext_ops {
 #define PM_EVENT_THAW		0x0020
 #define PM_EVENT_RESTORE	0x0040
 #define PM_EVENT_RECOVER	0x0080
+#define PM_EVENT_USER		0x0100
+#define PM_EVENT_REMOTE		0x0200
+#define PM_EVENT_AUTO		0x0400
 
-#define PM_EVENT_SLEEP	(PM_EVENT_SUSPEND | PM_EVENT_HIBERNATE)
+#define PM_EVENT_SLEEP		(PM_EVENT_SUSPEND | PM_EVENT_HIBERNATE)
+#define PM_EVENT_USER_SUSPEND	(PM_EVENT_USER | PM_EVENT_SUSPEND)
+#define PM_EVENT_USER_RESUME	(PM_EVENT_USER | PM_EVENT_RESUME)
+#define PM_EVENT_REMOTE_WAKEUP	(PM_EVENT_REMOTE | PM_EVENT_RESUME)
+#define PM_EVENT_AUTO_SUSPEND	(PM_EVENT_AUTO | PM_EVENT_SUSPEND)
+#define PM_EVENT_AUTO_RESUME	(PM_EVENT_AUTO | PM_EVENT_RESUME)
 
+#define PMSG_ON		((struct pm_message){ .event = PM_EVENT_ON, })
 #define PMSG_FREEZE	((struct pm_message){ .event = PM_EVENT_FREEZE, })
 #define PMSG_QUIESCE	((struct pm_message){ .event = PM_EVENT_QUIESCE, })
 #define PMSG_SUSPEND	((struct pm_message){ .event = PM_EVENT_SUSPEND, })
@@ -267,7 +291,16 @@ struct pm_ext_ops {
 #define PMSG_THAW	((struct pm_message){ .event = PM_EVENT_THAW, })
 #define PMSG_RESTORE	((struct pm_message){ .event = PM_EVENT_RESTORE, })
 #define PMSG_RECOVER	((struct pm_message){ .event = PM_EVENT_RECOVER, })
-#define PMSG_ON		((struct pm_message){ .event = PM_EVENT_ON, })
+#define PMSG_USER_SUSPEND	((struct pm_messge) \
+					{ .event = PM_EVENT_USER_SUSPEND, })
+#define PMSG_USER_RESUME	((struct pm_messge) \
+					{ .event = PM_EVENT_USER_RESUME, })
+#define PMSG_REMOTE_RESUME	((struct pm_messge) \
+					{ .event = PM_EVENT_REMOTE_RESUME, })
+#define PMSG_AUTO_SUSPEND	((struct pm_messge) \
+					{ .event = PM_EVENT_AUTO_SUSPEND, })
+#define PMSG_AUTO_RESUME		((struct pm_messge) \
+					{ .event = PM_EVENT_AUTO_RESUME, })
 
 /**
  * Device power management states

From 0d83304c7e7bd3b05be90281b3a47841bc8f057a Mon Sep 17 00:00:00 2001
From: Akinobu Mita <akinobu.mita@gmail.com>
Date: Wed, 23 Jul 2008 21:28:38 -0700
Subject: [PATCH 0314/1171] pm: hibernation: simplify memory bitmap

This patch simplifies the memory bitmap manipulations.

- remove the member size in struct bm_block

It is not necessary for struct bm_block to have the number of bit chunks that
can be calculated by using end_pfn and start_pfn.

- use find_next_bit() for memory_bm_next_pfn

No need to invent the bitmap library only for the memory bitmap.

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/power/snapshot.c | 86 ++++++++++-------------------------------
 1 file changed, 20 insertions(+), 66 deletions(-)

diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c
index 5f91a07c4eac..5d2ab836e998 100644
--- a/kernel/power/snapshot.c
+++ b/kernel/power/snapshot.c
@@ -205,8 +205,7 @@ static void chain_free(struct chain_allocator *ca, int clear_page_nosave)
  *	objects.  The main list's elements are of type struct zone_bitmap
  *	and each of them corresonds to one zone.  For each zone bitmap
  *	object there is a list of objects of type struct bm_block that
- *	represent each blocks of bit chunks in which information is
- *	stored.
+ *	represent each blocks of bitmap in which information is stored.
  *
  *	struct memory_bitmap contains a pointer to the main list of zone
  *	bitmap objects, a struct bm_position used for browsing the bitmap,
@@ -224,26 +223,27 @@ static void chain_free(struct chain_allocator *ca, int clear_page_nosave)
  *	pfns that correspond to the start and end of the represented zone.
  *
  *	struct bm_block contains a pointer to the memory page in which
- *	information is stored (in the form of a block of bit chunks
- *	of type unsigned long each).  It also contains the pfns that
- *	correspond to the start and end of the represented memory area and
- *	the number of bit chunks in the block.
+ *	information is stored (in the form of a block of bitmap)
+ *	It also contains the pfns that correspond to the start and end of
+ *	the represented memory area.
  */
 
 #define BM_END_OF_MAP	(~0UL)
 
-#define BM_CHUNKS_PER_BLOCK	(PAGE_SIZE / sizeof(long))
-#define BM_BITS_PER_CHUNK	(sizeof(long) << 3)
 #define BM_BITS_PER_BLOCK	(PAGE_SIZE << 3)
 
 struct bm_block {
 	struct bm_block *next;		/* next element of the list */
 	unsigned long start_pfn;	/* pfn represented by the first bit */
 	unsigned long end_pfn;	/* pfn represented by the last bit plus 1 */
-	unsigned int size;	/* number of bit chunks */
-	unsigned long *data;	/* chunks of bits representing pages */
+	unsigned long *data;	/* bitmap representing pages */
 };
 
+static inline unsigned long bm_block_bits(struct bm_block *bb)
+{
+	return bb->end_pfn - bb->start_pfn;
+}
+
 struct zone_bitmap {
 	struct zone_bitmap *next;	/* next element of the list */
 	unsigned long start_pfn;	/* minimal pfn in this zone */
@@ -257,7 +257,6 @@ struct zone_bitmap {
 struct bm_position {
 	struct zone_bitmap *zone_bm;
 	struct bm_block *block;
-	int chunk;
 	int bit;
 };
 
@@ -272,12 +271,6 @@ struct memory_bitmap {
 
 /* Functions that operate on memory bitmaps */
 
-static inline void memory_bm_reset_chunk(struct memory_bitmap *bm)
-{
-	bm->cur.chunk = 0;
-	bm->cur.bit = -1;
-}
-
 static void memory_bm_position_reset(struct memory_bitmap *bm)
 {
 	struct zone_bitmap *zone_bm;
@@ -285,7 +278,7 @@ static void memory_bm_position_reset(struct memory_bitmap *bm)
 	zone_bm = bm->zone_bm_list;
 	bm->cur.zone_bm = zone_bm;
 	bm->cur.block = zone_bm->bm_blocks;
-	memory_bm_reset_chunk(bm);
+	bm->cur.bit = 0;
 }
 
 static void memory_bm_free(struct memory_bitmap *bm, int clear_nosave_free);
@@ -394,12 +387,10 @@ memory_bm_create(struct memory_bitmap *bm, gfp_t gfp_mask, int safe_needed)
 			bb->start_pfn = pfn;
 			if (nr >= BM_BITS_PER_BLOCK) {
 				pfn += BM_BITS_PER_BLOCK;
-				bb->size = BM_CHUNKS_PER_BLOCK;
 				nr -= BM_BITS_PER_BLOCK;
 			} else {
 				/* This is executed only once in the loop */
 				pfn += nr;
-				bb->size = DIV_ROUND_UP(nr, BM_BITS_PER_CHUNK);
 			}
 			bb->end_pfn = pfn;
 			bb = bb->next;
@@ -478,8 +469,8 @@ static int memory_bm_find_bit(struct memory_bitmap *bm, unsigned long pfn,
 	}
 	zone_bm->cur_block = bb;
 	pfn -= bb->start_pfn;
-	*bit_nr = pfn % BM_BITS_PER_CHUNK;
-	*addr = bb->data + pfn / BM_BITS_PER_CHUNK;
+	*bit_nr = pfn;
+	*addr = bb->data;
 	return 0;
 }
 
@@ -528,36 +519,6 @@ static int memory_bm_test_bit(struct memory_bitmap *bm, unsigned long pfn)
 	return test_bit(bit, addr);
 }
 
-/* Two auxiliary functions for memory_bm_next_pfn */
-
-/* Find the first set bit in the given chunk, if there is one */
-
-static inline int next_bit_in_chunk(int bit, unsigned long *chunk_p)
-{
-	bit++;
-	while (bit < BM_BITS_PER_CHUNK) {
-		if (test_bit(bit, chunk_p))
-			return bit;
-
-		bit++;
-	}
-	return -1;
-}
-
-/* Find a chunk containing some bits set in given block of bits */
-
-static inline int next_chunk_in_block(int n, struct bm_block *bb)
-{
-	n++;
-	while (n < bb->size) {
-		if (bb->data[n])
-			return n;
-
-		n++;
-	}
-	return -1;
-}
-
 /**
  *	memory_bm_next_pfn - find the pfn that corresponds to the next set bit
  *	in the bitmap @bm.  If the pfn cannot be found, BM_END_OF_MAP is
@@ -571,40 +532,33 @@ static unsigned long memory_bm_next_pfn(struct memory_bitmap *bm)
 {
 	struct zone_bitmap *zone_bm;
 	struct bm_block *bb;
-	int chunk;
 	int bit;
 
 	do {
 		bb = bm->cur.block;
 		do {
-			chunk = bm->cur.chunk;
 			bit = bm->cur.bit;
-			do {
-				bit = next_bit_in_chunk(bit, bb->data + chunk);
-				if (bit >= 0)
-					goto Return_pfn;
+			bit = find_next_bit(bb->data, bm_block_bits(bb), bit);
+			if (bit < bm_block_bits(bb))
+				goto Return_pfn;
 
-				chunk = next_chunk_in_block(chunk, bb);
-				bit = -1;
-			} while (chunk >= 0);
 			bb = bb->next;
 			bm->cur.block = bb;
-			memory_bm_reset_chunk(bm);
+			bm->cur.bit = 0;
 		} while (bb);
 		zone_bm = bm->cur.zone_bm->next;
 		if (zone_bm) {
 			bm->cur.zone_bm = zone_bm;
 			bm->cur.block = zone_bm->bm_blocks;
-			memory_bm_reset_chunk(bm);
+			bm->cur.bit = 0;
 		}
 	} while (zone_bm);
 	memory_bm_position_reset(bm);
 	return BM_END_OF_MAP;
 
  Return_pfn:
-	bm->cur.chunk = chunk;
-	bm->cur.bit = bit;
-	return bb->start_pfn + chunk * BM_BITS_PER_CHUNK + bit;
+	bm->cur.bit = bit + 1;
+	return bb->start_pfn + bit;
 }
 
 /**

From c1a220e7acf8ad2c03504891f4a70cd9c32c904b Mon Sep 17 00:00:00 2001
From: Zhang Rui <rui.zhang@intel.com>
Date: Wed, 23 Jul 2008 21:28:39 -0700
Subject: [PATCH 0315/1171] pm: introduce new interfaces schedule_work_on() and
 queue_work_on()

This interface allows adding a job on a specific cpu.

Although a work struct on a cpu will be scheduled to other cpu if the cpu
dies, there is a recursion if a work task tries to offline the cpu it's
running on.  we need to schedule the task to a specific cpu in this case.
http://bugzilla.kernel.org/show_bug.cgi?id=10897

[oleg@tv-sign.ru: cleanups]
Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Tested-by: Rus <harbour@sfinx.od.ua>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/workqueue.h |  3 +++
 kernel/workqueue.c        | 39 ++++++++++++++++++++++++++++++++++++++-
 2 files changed, 41 insertions(+), 1 deletion(-)

diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h
index 542526c6e8ef..14d47120682b 100644
--- a/include/linux/workqueue.h
+++ b/include/linux/workqueue.h
@@ -179,6 +179,8 @@ __create_workqueue_key(const char *name, int singlethread,
 extern void destroy_workqueue(struct workqueue_struct *wq);
 
 extern int queue_work(struct workqueue_struct *wq, struct work_struct *work);
+extern int queue_work_on(int cpu, struct workqueue_struct *wq,
+			struct work_struct *work);
 extern int queue_delayed_work(struct workqueue_struct *wq,
 			struct delayed_work *work, unsigned long delay);
 extern int queue_delayed_work_on(int cpu, struct workqueue_struct *wq,
@@ -188,6 +190,7 @@ extern void flush_workqueue(struct workqueue_struct *wq);
 extern void flush_scheduled_work(void);
 
 extern int schedule_work(struct work_struct *work);
+extern int schedule_work_on(int cpu, struct work_struct *work);
 extern int schedule_delayed_work(struct delayed_work *work, unsigned long delay);
 extern int schedule_delayed_work_on(int cpu, struct delayed_work *work,
 					unsigned long delay);
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index a6d36346d10a..6fd158b21026 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -140,7 +140,6 @@ static void insert_work(struct cpu_workqueue_struct *cwq,
 	wake_up(&cwq->more_work);
 }
 
-/* Preempt must be disabled. */
 static void __queue_work(struct cpu_workqueue_struct *cwq,
 			 struct work_struct *work)
 {
@@ -175,6 +174,31 @@ int queue_work(struct workqueue_struct *wq, struct work_struct *work)
 }
 EXPORT_SYMBOL_GPL(queue_work);
 
+/**
+ * queue_work_on - queue work on specific cpu
+ * @cpu: CPU number to execute work on
+ * @wq: workqueue to use
+ * @work: work to queue
+ *
+ * Returns 0 if @work was already on a queue, non-zero otherwise.
+ *
+ * We queue the work to a specific CPU, the caller must ensure it
+ * can't go away.
+ */
+int
+queue_work_on(int cpu, struct workqueue_struct *wq, struct work_struct *work)
+{
+	int ret = 0;
+
+	if (!test_and_set_bit(WORK_STRUCT_PENDING, work_data_bits(work))) {
+		BUG_ON(!list_empty(&work->entry));
+		__queue_work(wq_per_cpu(wq, cpu), work);
+		ret = 1;
+	}
+	return ret;
+}
+EXPORT_SYMBOL_GPL(queue_work_on);
+
 static void delayed_work_timer_fn(unsigned long __data)
 {
 	struct delayed_work *dwork = (struct delayed_work *)__data;
@@ -553,6 +577,19 @@ int schedule_work(struct work_struct *work)
 }
 EXPORT_SYMBOL(schedule_work);
 
+/*
+ * schedule_work_on - put work task on a specific cpu
+ * @cpu: cpu to put the work task on
+ * @work: job to be done
+ *
+ * This puts a job on a specific cpu
+ */
+int schedule_work_on(int cpu, struct work_struct *work)
+{
+	return queue_work_on(cpu, keventd_wq, work);
+}
+EXPORT_SYMBOL(schedule_work_on);
+
 /**
  * schedule_delayed_work - put work task in global workqueue after delay
  * @dwork: job to be done

From 2f15fc4bdf91eb399da3f47a09c55831d9f22826 Mon Sep 17 00:00:00 2001
From: Zhang Rui <rui.zhang@intel.com>
Date: Wed, 23 Jul 2008 21:28:40 -0700
Subject: [PATCH 0316/1171] pm: schedule sysrq poweroff on boot cpu

schedule sysrq poweroff on boot cpu.

sysrq poweroff needs to disable nonboot cpus, and we need to run this on boot
cpu to avoid any recursion.  http://bugzilla.kernel.org/show_bug.cgi?id=10897

[kosaki.motohiro@jp.fujitsu.com: build fix]
Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Tested-by: Rus <harbour@sfinx.od.ua>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/power/poweroff.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/kernel/power/poweroff.c b/kernel/power/poweroff.c
index 678ec736076b..72016f051477 100644
--- a/kernel/power/poweroff.c
+++ b/kernel/power/poweroff.c
@@ -10,6 +10,7 @@
 #include <linux/pm.h>
 #include <linux/workqueue.h>
 #include <linux/reboot.h>
+#include <linux/cpumask.h>
 
 /*
  * When the user hits Sys-Rq o to power down the machine this is the
@@ -25,7 +26,8 @@ static DECLARE_WORK(poweroff_work, do_poweroff);
 
 static void handle_poweroff(int key, struct tty_struct *tty)
 {
-	schedule_work(&poweroff_work);
+	/* run sysrq poweroff on boot cpu */
+	schedule_work_on(first_cpu(cpu_online_map), &poweroff_work);
 }
 
 static struct sysrq_key_op	sysrq_poweroff_op = {

From bdfe6b7c681669148dae4db27eb24ee5408ba371 Mon Sep 17 00:00:00 2001
From: Shaohua Li <shaohua.li@intel.com>
Date: Wed, 23 Jul 2008 21:28:41 -0700
Subject: [PATCH 0317/1171] pm: acpi hibernation: utilize hardware signature

ACPI defines a hardware signature.  BIOS calculates the signature according to
hardware configure and if hardware changes while hibernated, the signature
will change.  In that case, S4 resume should fail.

Still, there may be systems on which this mechanism does not work correctly,
so it is better to provide a workaround for them.  For this reason, add a new
switch to the acpi_sleep= command line argument allowing one to disable
hardware signature checking.

[shaohua.li@intel.com: build fix]
Signed-off-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Len Brown <lenb@kernel.org>
Acked-by: Pavel Machek <pavel@ucw.cz>
Cc: <Valdis.Kletnieks@vt.edu>
Cc: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/kernel-parameters.txt |  4 +++-
 arch/x86/kernel/acpi/sleep.c        |  4 ++++
 drivers/acpi/sleep/main.c           | 22 ++++++++++++++++++++++
 include/linux/acpi.h                |  1 +
 4 files changed, 30 insertions(+), 1 deletion(-)

diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 4d705713cabc..497a98dafdaa 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -148,10 +148,12 @@ and is between 256 and 4096 characters. It is defined in the file
 			default: 0
 
 	acpi_sleep=	[HW,ACPI] Sleep options
-			Format: { s3_bios, s3_mode, s3_beep, old_ordering }
+			Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig, old_ordering }
 			See Documentation/power/video.txt for s3_bios and s3_mode.
 			s3_beep is for debugging; it makes the PC's speaker beep
 			as soon as the kernel's real-mode entry point is called.
+			s4_nohwsig prevents ACPI hardware signature from being
+			used during resume from hibernation.
 			old_ordering causes the ACPI 1.0 ordering of the _PTS
 			control method, wrt putting devices into low power
 			states, to be enforced (the ACPI 2.0 ordering of _PTS is
diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c
index a3ddad18aaa3..fa2161d5003b 100644
--- a/arch/x86/kernel/acpi/sleep.c
+++ b/arch/x86/kernel/acpi/sleep.c
@@ -150,6 +150,10 @@ static int __init acpi_sleep_setup(char *str)
 			acpi_realmode_flags |= 2;
 		if (strncmp(str, "s3_beep", 7) == 0)
 			acpi_realmode_flags |= 4;
+#ifdef CONFIG_HIBERNATION
+		if (strncmp(str, "s4_nohwsig", 10) == 0)
+			acpi_no_s4_hw_signature();
+#endif
 		if (strncmp(str, "old_ordering", 12) == 0)
 			acpi_old_suspend_ordering();
 		str = strchr(str, ',');
diff --git a/drivers/acpi/sleep/main.c b/drivers/acpi/sleep/main.c
index 0489a7d1d42c..313507accf18 100644
--- a/drivers/acpi/sleep/main.c
+++ b/drivers/acpi/sleep/main.c
@@ -283,6 +283,15 @@ static struct platform_suspend_ops acpi_suspend_ops_old = {
 #endif /* CONFIG_SUSPEND */
 
 #ifdef CONFIG_HIBERNATION
+static unsigned long s4_hardware_signature;
+static struct acpi_table_facs *facs;
+static bool nosigcheck;
+
+void __init acpi_no_s4_hw_signature(void)
+{
+	nosigcheck = true;
+}
+
 static int acpi_hibernation_begin(void)
 {
 	acpi_target_sleep_state = ACPI_STATE_S4;
@@ -316,6 +325,12 @@ static void acpi_hibernation_leave(void)
 	acpi_enable();
 	/* Reprogram control registers and execute _BFS */
 	acpi_leave_sleep_state_prep(ACPI_STATE_S4);
+	/* Check the hardware signature */
+	if (facs && s4_hardware_signature != facs->hardware_signature) {
+		printk(KERN_EMERG "ACPI: Hardware changed while hibernated, "
+			"cannot resume!\n");
+		panic("ACPI S4 hardware signature mismatch");
+	}
 }
 
 static void acpi_pm_enable_gpes(void)
@@ -544,6 +559,13 @@ int __init acpi_sleep_init(void)
 			&acpi_hibernation_ops_old : &acpi_hibernation_ops);
 		sleep_states[ACPI_STATE_S4] = 1;
 		printk(" S4");
+		if (!nosigcheck) {
+			acpi_get_table_by_index(ACPI_TABLE_INDEX_FACS,
+				(struct acpi_table_header **)&facs);
+			if (facs)
+				s4_hardware_signature =
+					facs->hardware_signature;
+		}
 	}
 #endif
 	status = acpi_get_sleep_type_data(ACPI_STATE_S5, &type_a, &type_b);
diff --git a/include/linux/acpi.h b/include/linux/acpi.h
index a17177639376..702f79dad16a 100644
--- a/include/linux/acpi.h
+++ b/include/linux/acpi.h
@@ -236,6 +236,7 @@ int acpi_check_mem_region(resource_size_t start, resource_size_t n,
 		      const char *name);
 
 #ifdef CONFIG_PM_SLEEP
+void __init acpi_no_s4_hw_signature(void);
 void __init acpi_old_suspend_ordering(void);
 #endif /* CONFIG_PM_SLEEP */
 #else	/* CONFIG_ACPI */

From e41fb7c58e3ca18ec5c9c9bb7bb68e8e653c9e8e Mon Sep 17 00:00:00 2001
From: Carlos Corbacho <carlos@strangeworlds.co.uk>
Date: Wed, 23 Jul 2008 21:28:43 -0700
Subject: [PATCH 0318/1171] pm: acpi pm: add DMI quirk list for ACPI 1.0
 suspend ordering

There are a few BIOSes that we know of already that need to use the ACPI 1.0
suspend order.  This appears to be only be a small minority of mostly nVidia
based systems.

Based on observation of Windows behaviour, it's clear that Windows is also
doing maintaining its own list of broken hardware that needs this workaround.

Signed-off-by: Carlos Corbacho <carlos@strangeworlds.co.uk>
Signed-off-by: Rafael J. Wysocki <rjw@sisk.pl>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Len Brown <lenb@kernel.org>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/acpi/sleep/main.c | 20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

diff --git a/drivers/acpi/sleep/main.c b/drivers/acpi/sleep/main.c
index 313507accf18..d13194a031bf 100644
--- a/drivers/acpi/sleep/main.c
+++ b/drivers/acpi/sleep/main.c
@@ -280,6 +280,24 @@ static struct platform_suspend_ops acpi_suspend_ops_old = {
 	.end = acpi_pm_end,
 	.recover = acpi_pm_finish,
 };
+
+static int __init init_old_suspend_ordering(const struct dmi_system_id *d)
+{
+	old_suspend_ordering = true;
+	return 0;
+}
+
+static struct dmi_system_id __initdata acpisleep_dmi_table[] = {
+	{
+	.callback = init_old_suspend_ordering,
+	.ident = "Abit KN9 (nForce4 variant)",
+	.matches = {
+		DMI_MATCH(DMI_BOARD_VENDOR, "http://www.abit.com.tw/"),
+		DMI_MATCH(DMI_BOARD_NAME, "KN9 Series(NF-CK804)"),
+		},
+	},
+	{},
+};
 #endif /* CONFIG_SUSPEND */
 
 #ifdef CONFIG_HIBERNATION
@@ -531,6 +549,8 @@ int __init acpi_sleep_init(void)
 	u8 type_a, type_b;
 #ifdef CONFIG_SUSPEND
 	int i = 0;
+
+	dmi_check_system(acpisleep_dmi_table);
 #endif
 
 	if (acpi_disabled)

From f0af566da6e9a4a2f5a83c5a70f3d0a772050e21 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Wed, 23 Jul 2008 21:28:44 -0700
Subject: [PATCH 0319/1171] pm: fix try_to_freeze_tasks()'s use of do_div()

Fix try_to_freeze_tasks()'s use of do_div() on an s64 by making
elapsed_csecs64 a u64 instead and dividing that.

Possibly this should be guarded lest the interval calculation turn up
negative, but the possible negativity of the result of the division is
cast away anyway.

This was introduced by patch 438e2ce68dfd4af4cfcec2f873564fb921db4bb5.

Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: "Rafael J. Wysocki" <rjw@sisk.pl>
Acked-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/power/process.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/power/process.c b/kernel/power/process.c
index 5fb87652f214..278946aecaf0 100644
--- a/kernel/power/process.c
+++ b/kernel/power/process.c
@@ -149,7 +149,7 @@ static int try_to_freeze_tasks(bool sig_only)
 	unsigned long end_time;
 	unsigned int todo;
 	struct timeval start, end;
-	s64 elapsed_csecs64;
+	u64 elapsed_csecs64;
 	unsigned int elapsed_csecs;
 
 	do_gettimeofday(&start);

From 912019572180f287e85b5534fbb1c1e3ca6df6c9 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Wed, 23 Jul 2008 21:28:45 -0700
Subject: [PATCH 0320/1171] mn10300: move sg_dma_{address,len}() to
 asm/scatterlist.h

mn10300 was the only architecture where sg_dma_{address,len}() were not
in asm/scatterlist.h, and it's not a big surprise that this caused a
compile error somewhere:

/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/media/video/videobuf-dma-sg.c: In function `videobuf_dma_map':
/home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/media/video/videobuf-dma-sg.c:238: error: implicit declaration of function 'sg_dma_address'

Acked-by: David Howells <dhowells@redhat.com>
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/asm-mn10300/pci.h         | 9 ---------
 include/asm-mn10300/scatterlist.h | 9 +++++++++
 2 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/include/asm-mn10300/pci.h b/include/asm-mn10300/pci.h
index 205192c52bb5..cd9cc5c89cea 100644
--- a/include/asm-mn10300/pci.h
+++ b/include/asm-mn10300/pci.h
@@ -74,15 +74,6 @@ struct pci_dev;
 /* This is always fine. */
 #define pci_dac_dma_supported(pci_dev, mask)	(0)
 
-/*
- * These macros should be used after a pci_map_sg call has been done
- * to get bus addresses of each of the SG entries and their lengths.
- * You should only work with the number of sg entries pci_map_sg
- * returns.
- */
-#define sg_dma_address(sg)	((sg)->dma_address)
-#define sg_dma_len(sg)		((sg)->length)
-
 /* Return the index of the PCI controller for device. */
 static inline int pci_controller_num(struct pci_dev *dev)
 {
diff --git a/include/asm-mn10300/scatterlist.h b/include/asm-mn10300/scatterlist.h
index e29d91dbcf2b..67535901b9ff 100644
--- a/include/asm-mn10300/scatterlist.h
+++ b/include/asm-mn10300/scatterlist.h
@@ -43,4 +43,13 @@ struct scatterlist {
 
 #define ISA_DMA_THRESHOLD (0x00ffffff)
 
+/*
+ * These macros should be used after a pci_map_sg call has been done
+ * to get bus addresses of each of the SG entries and their lengths.
+ * You should only work with the number of sg entries pci_map_sg
+ * returns.
+ */
+#define sg_dma_address(sg)	((sg)->dma_address)
+#define sg_dma_len(sg)		((sg)->length)
+
 #endif /* _ASM_SCATTERLIST_H */

From d50004b0867a59f8a81116f000edb352595343d9 Mon Sep 17 00:00:00 2001
From: Fernando Luis Vazquez Cao <fernando@intellilink.co.jp>
Date: Wed, 23 Jul 2008 21:28:45 -0700
Subject: [PATCH 0321/1171] cris: remove unused global_flush_tlb

global_flush_tlb is declared but never used.

Signed-off-by: Fernando Luis Vazquez Cao <fernando@oss.ntt.co.jp>
Cc: Mikael Starvik <starvik@axis.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/asm-cris/cacheflush.h | 1 -
 1 file changed, 1 deletion(-)

diff --git a/include/asm-cris/cacheflush.h b/include/asm-cris/cacheflush.h
index 01af2de27c5b..cf60e3f69f8d 100644
--- a/include/asm-cris/cacheflush.h
+++ b/include/asm-cris/cacheflush.h
@@ -26,7 +26,6 @@
 #define copy_from_user_page(vma, page, vaddr, dst, src, len) \
 	memcpy(dst, src, len)
 
-void global_flush_tlb(void); 
 int change_page_attr(struct page *page, int numpages, pgprot_t prot);
 
 #endif /* _CRIS_CACHEFLUSH_H */

From ed62f77bb631bc4a2d8acb0521b720cb55e58183 Mon Sep 17 00:00:00 2001
From: Akinobu Mita <akinobu.mita@gmail.com>
Date: Wed, 23 Jul 2008 21:28:46 -0700
Subject: [PATCH 0322/1171] cris: use simple_read_from_buffer()

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Mikael Starvik <starvik@axis.com>
Cc: Jesper Nilsson <jesper.nilsson@axis.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/cris/kernel/profile.c | 17 +++++++----------
 1 file changed, 7 insertions(+), 10 deletions(-)

diff --git a/arch/cris/kernel/profile.c b/arch/cris/kernel/profile.c
index 44f7b4f79476..9aa571169bcc 100644
--- a/arch/cris/kernel/profile.c
+++ b/arch/cris/kernel/profile.c
@@ -35,19 +35,16 @@ read_cris_profile(struct file *file, char __user *buf,
 		  size_t count, loff_t *ppos)
 {
 	unsigned long p = *ppos;
+	ssize_t ret;
 
-	if (p > SAMPLE_BUFFER_SIZE)
-		return 0;
+	ret = simple_read_from_buffer(buf, count, ppos, sample_buffer,
+						SAMPLE_BUFFER_SIZE);
+	if (ret < 0)
+		return ret;
 
-	if (p + count > SAMPLE_BUFFER_SIZE)
-		count = SAMPLE_BUFFER_SIZE - p;
-	if (copy_to_user(buf, sample_buffer + p,count))
-		return -EFAULT;
+	memset(sample_buffer + p, 0, ret);
 
-	memset(sample_buffer + p, 0, count);
-	*ppos += count;
-
-	return count;
+	return ret;
 }
 
 static ssize_t

From 4c182ae7810f3fe444e666f3f78c209a7c116fdf Mon Sep 17 00:00:00 2001
From: WANG Cong <xiyou.wangcong@gmail.com>
Date: Wed, 23 Jul 2008 21:28:47 -0700
Subject: [PATCH 0323/1171] arch/um/kernel/irq.c: clean up some functions

Make activate_fd() and free_irq_by_irq_and_dev() static.  Remove
init_aio_irq() since it has no users.

Cc: Jeff Dike <jdike@addtoit.com>
Signed-off-by: WANG Cong <wangcong@zeuux.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/um/include/irq_kern.h |  2 --
 arch/um/include/irq_user.h |  2 --
 arch/um/kernel/irq.c       | 35 ++---------------------------------
 3 files changed, 2 insertions(+), 37 deletions(-)

diff --git a/arch/um/include/irq_kern.h b/arch/um/include/irq_kern.h
index 4f775597fd5f..fba3895274f9 100644
--- a/arch/um/include/irq_kern.h
+++ b/arch/um/include/irq_kern.h
@@ -13,8 +13,6 @@ extern int um_request_irq(unsigned int irq, int fd, int type,
 			  irq_handler_t handler,
 			  unsigned long irqflags,  const char * devname,
 			  void *dev_id);
-extern int init_aio_irq(int irq, char *name,
-			irq_handler_t handler);
 
 #endif
 
diff --git a/arch/um/include/irq_user.h b/arch/um/include/irq_user.h
index e60b31873de1..c6c784df2673 100644
--- a/arch/um/include/irq_user.h
+++ b/arch/um/include/irq_user.h
@@ -21,8 +21,6 @@ struct irq_fd {
 enum { IRQ_READ, IRQ_WRITE };
 
 extern void sigio_handler(int sig, struct uml_pt_regs *regs);
-extern int activate_fd(int irq, int fd, int type, void *dev_id);
-extern void free_irq_by_irq_and_dev(unsigned int irq, void *dev_id);
 extern void free_irq_by_fd(int fd);
 extern void reactivate_fd(int fd, int irqnum);
 extern void deactivate_fd(int fd, int irqnum);
diff --git a/arch/um/kernel/irq.c b/arch/um/kernel/irq.c
index 91587f8db340..3d7aad09b171 100644
--- a/arch/um/kernel/irq.c
+++ b/arch/um/kernel/irq.c
@@ -102,7 +102,7 @@ void sigio_handler(int sig, struct uml_pt_regs *regs)
 
 static DEFINE_SPINLOCK(irq_lock);
 
-int activate_fd(int irq, int fd, int type, void *dev_id)
+static int activate_fd(int irq, int fd, int type, void *dev_id)
 {
 	struct pollfd *tmp_pfd;
 	struct irq_fd *new_fd, *irq_fd;
@@ -216,7 +216,7 @@ static int same_irq_and_dev(struct irq_fd *irq, void *d)
 	return ((irq->irq == data->irq) && (irq->id == data->dev));
 }
 
-void free_irq_by_irq_and_dev(unsigned int irq, void *dev)
+static void free_irq_by_irq_and_dev(unsigned int irq, void *dev)
 {
 	struct irq_and_dev data = ((struct irq_and_dev) { .irq  = irq,
 							  .dev  = dev });
@@ -403,37 +403,6 @@ void __init init_IRQ(void)
 	}
 }
 
-int init_aio_irq(int irq, char *name, irq_handler_t handler)
-{
-	int fds[2], err;
-
-	err = os_pipe(fds, 1, 1);
-	if (err) {
-		printk(KERN_ERR "init_aio_irq - os_pipe failed, err = %d\n",
-		       -err);
-		goto out;
-	}
-
-	err = um_request_irq(irq, fds[0], IRQ_READ, handler,
-			     IRQF_DISABLED | IRQF_SAMPLE_RANDOM, name,
-			     (void *) (long) fds[0]);
-	if (err) {
-		printk(KERN_ERR "init_aio_irq - : um_request_irq failed, "
-		       "err = %d\n",
-		       err);
-		goto out_close;
-	}
-
-	err = fds[1];
-	goto out;
-
- out_close:
-	os_close_file(fds[0]);
-	os_close_file(fds[1]);
- out:
-	return err;
-}
-
 /*
  * IRQ stack entry and exit:
  *

From 4a5675820436e4ad738dd442c1cc8a165101509b Mon Sep 17 00:00:00 2001
From: WANG Cong <xiyou.wangcong@gmail.com>
Date: Wed, 23 Jul 2008 21:28:49 -0700
Subject: [PATCH 0324/1171] arch/um/kernel/mem.c: remove arch_validate()

- Remove arch_validate(), because no one uses it.

- Remove useless macro HAVE_ARCH_VALIDATE.

- Make the variable 'empty_bad_page' static.

Cc: Jeff Dike <jdike@addtoit.com>
Signed-off-by: WANG Cong <wangcong@zeuux.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/um/kernel/ksyms.c |  1 -
 arch/um/kernel/mem.c   | 33 +--------------------------------
 include/asm-um/page.h  |  3 ---
 3 files changed, 1 insertion(+), 36 deletions(-)

diff --git a/arch/um/kernel/ksyms.c b/arch/um/kernel/ksyms.c
index ccc02a616c22..836fc9b94707 100644
--- a/arch/um/kernel/ksyms.c
+++ b/arch/um/kernel/ksyms.c
@@ -18,7 +18,6 @@ EXPORT_SYMBOL(get_signals);
 EXPORT_SYMBOL(kernel_thread);
 EXPORT_SYMBOL(sys_waitpid);
 EXPORT_SYMBOL(flush_tlb_range);
-EXPORT_SYMBOL(arch_validate);
 
 EXPORT_SYMBOL(high_physmem);
 EXPORT_SYMBOL(empty_zero_page);
diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c
index b0ee64622ff7..e2274ef3155d 100644
--- a/arch/um/kernel/mem.c
+++ b/arch/um/kernel/mem.c
@@ -21,7 +21,7 @@
 /* allocated in paging_init, zeroed in mem_init, and unchanged thereafter */
 unsigned long *empty_zero_page = NULL;
 /* allocated in paging_init and unchanged thereafter */
-unsigned long *empty_bad_page = NULL;
+static unsigned long *empty_bad_page = NULL;
 
 /*
  * Initialized during boot, and readonly for initializing page tables
@@ -240,37 +240,6 @@ void __init paging_init(void)
 #endif
 }
 
-struct page *arch_validate(struct page *page, gfp_t mask, int order)
-{
-	unsigned long addr, zero = 0;
-	int i;
-
- again:
-	if (page == NULL)
-		return page;
-	if (PageHighMem(page))
-		return page;
-
-	addr = (unsigned long) page_address(page);
-	for (i = 0; i < (1 << order); i++) {
-		current->thread.fault_addr = (void *) addr;
-		if (__do_copy_to_user((void __user *) addr, &zero,
-				     sizeof(zero),
-				     &current->thread.fault_addr,
-				     &current->thread.fault_catcher)) {
-			if (!(mask & __GFP_WAIT))
-				return NULL;
-			else break;
-		}
-		addr += PAGE_SIZE;
-	}
-
-	if (i == (1 << order))
-		return page;
-	page = alloc_pages(mask, order);
-	goto again;
-}
-
 /*
  * This can't do anything because nothing in the kernel image can be freed
  * since it's not in kernel physical memory.
diff --git a/include/asm-um/page.h b/include/asm-um/page.h
index 335c57383c02..a6df1f13d732 100644
--- a/include/asm-um/page.h
+++ b/include/asm-um/page.h
@@ -115,9 +115,6 @@ extern unsigned long uml_physmem;
 #define pfn_valid(pfn) ((pfn) < max_mapnr)
 #define virt_addr_valid(v) pfn_valid(phys_to_pfn(__pa(v)))
 
-extern struct page *arch_validate(struct page *page, gfp_t mask, int order);
-#define HAVE_ARCH_VALIDATE
-
 #include <asm-generic/memory_model.h>
 #include <asm-generic/page.h>
 

From 99764fa4ceeecba8b9e0a8a5565b418a2e94f83b Mon Sep 17 00:00:00 2001
From: WANG Cong <xiyou.wangcong@gmail.com>
Date: Wed, 23 Jul 2008 21:28:49 -0700
Subject: [PATCH 0325/1171] UML: make several more things static

- Make some variables and functions static, since they don't need to be
  global.

- Remove an unused function - arch/um/kernel/time.c::sched_clock().

- Clean the style a bit as complained by checkpatch.pl.

Cc: Jeff Dike <jdike@addtoit.com>
Signed-off-by: WANG Cong <wangcong@zeuux.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/um/include/skas/skas.h     | 1 -
 arch/um/include/um_uaccess.h    | 1 -
 arch/um/kernel/physmem.c        | 2 +-
 arch/um/kernel/ptrace.c         | 2 +-
 arch/um/kernel/time.c           | 8 --------
 arch/um/kernel/uaccess.c        | 2 +-
 arch/um/os-Linux/sigio.c        | 2 +-
 arch/um/os-Linux/signal.c       | 2 +-
 arch/um/os-Linux/skas/process.c | 2 +-
 arch/um/os-Linux/umid.c         | 2 +-
 arch/um/sys-i386/bugs.c         | 2 +-
 arch/um/sys-i386/checksum.S     | 5 ++---
 arch/um/sys-i386/ldt.c          | 4 ++--
 include/asm-um/ptrace-generic.h | 3 ---
 14 files changed, 12 insertions(+), 26 deletions(-)

diff --git a/arch/um/include/skas/skas.h b/arch/um/include/skas/skas.h
index b073f8a86bd3..64d2c7443306 100644
--- a/arch/um/include/skas/skas.h
+++ b/arch/um/include/skas/skas.h
@@ -16,7 +16,6 @@ extern int user_thread(unsigned long stack, int flags);
 extern void new_thread_handler(void);
 extern void handle_syscall(struct uml_pt_regs *regs);
 extern int new_mm(unsigned long stack);
-extern void get_skas_faultinfo(int pid, struct faultinfo * fi);
 extern long execute_syscall_skas(void *r);
 extern unsigned long current_stub_stack(void);
 
diff --git a/arch/um/include/um_uaccess.h b/arch/um/include/um_uaccess.h
index 2b6fc8e0f071..45c04999d670 100644
--- a/arch/um/include/um_uaccess.h
+++ b/arch/um/include/um_uaccess.h
@@ -34,7 +34,6 @@ extern int copy_to_user(void __user *to, const void *from, int n);
 
 extern int __do_copy_to_user(void *to, const void *from, int n,
 			     void **fault_addr, jmp_buf **fault_catcher);
-extern void __do_copy(void *to, const void *from, int n);
 
 /*
  * strncpy_from_user: - Copy a NUL terminated string from userspace.
diff --git a/arch/um/kernel/physmem.c b/arch/um/kernel/physmem.c
index 9757085a0220..a1a9090254c2 100644
--- a/arch/um/kernel/physmem.c
+++ b/arch/um/kernel/physmem.c
@@ -185,7 +185,7 @@ unsigned long find_iomem(char *driver, unsigned long *len_out)
 	return 0;
 }
 
-int setup_iomem(void)
+static int setup_iomem(void)
 {
 	struct iomem_region *region = iomem_regions;
 	unsigned long iomem_start = high_physmem + PAGE_SIZE;
diff --git a/arch/um/kernel/ptrace.c b/arch/um/kernel/ptrace.c
index 47b57b497d55..15e8b7c4de13 100644
--- a/arch/um/kernel/ptrace.c
+++ b/arch/um/kernel/ptrace.c
@@ -225,7 +225,7 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data)
 	return ret;
 }
 
-void send_sigtrap(struct task_struct *tsk, struct uml_pt_regs *regs,
+static void send_sigtrap(struct task_struct *tsk, struct uml_pt_regs *regs,
 		  int error_code)
 {
 	struct siginfo info;
diff --git a/arch/um/kernel/time.c b/arch/um/kernel/time.c
index c3e2f369c33c..47f04f4a3464 100644
--- a/arch/um/kernel/time.c
+++ b/arch/um/kernel/time.c
@@ -13,14 +13,6 @@
 #include "kern_util.h"
 #include "os.h"
 
-/*
- * Scheduler clock - returns current time in nanosec units.
- */
-unsigned long long sched_clock(void)
-{
-	return (unsigned long long)jiffies_64 * (NSEC_PER_SEC / HZ);
-}
-
 void timer_handler(int sig, struct uml_pt_regs *regs)
 {
 	unsigned long flags;
diff --git a/arch/um/kernel/uaccess.c b/arch/um/kernel/uaccess.c
index f0f4b040d7c5..dd33f040c526 100644
--- a/arch/um/kernel/uaccess.c
+++ b/arch/um/kernel/uaccess.c
@@ -12,7 +12,7 @@
 #include <linux/string.h>
 #include "os.h"
 
-void __do_copy(void *to, const void *from, int n)
+static void __do_copy(void *to, const void *from, int n)
 {
 	memcpy(to, from, n);
 }
diff --git a/arch/um/os-Linux/sigio.c b/arch/um/os-Linux/sigio.c
index eb8f2e4be192..63d299df152b 100644
--- a/arch/um/os-Linux/sigio.c
+++ b/arch/um/os-Linux/sigio.c
@@ -530,7 +530,7 @@ static void tty_close(int master, int slave)
 		printk(UM_KERN_CONT "No, enabling workaround\n");
 }
 
-void __init check_sigio(void)
+static void __init check_sigio(void)
 {
 	if ((access("/dev/ptmx", R_OK) < 0) &&
 	    (access("/dev/ptyp0", R_OK) < 0)) {
diff --git a/arch/um/os-Linux/signal.c b/arch/um/os-Linux/signal.c
index 5aade6027e40..6ae180703a63 100644
--- a/arch/um/os-Linux/signal.c
+++ b/arch/um/os-Linux/signal.c
@@ -126,7 +126,7 @@ void set_sigstack(void *sig_stack, int size)
 		panic("enabling signal stack failed, errno = %d\n", errno);
 }
 
-void (*handlers[_NSIG])(int sig, struct sigcontext *sc);
+static void (*handlers[_NSIG])(int sig, struct sigcontext *sc);
 
 void handle_signal(int sig, struct sigcontext *sc)
 {
diff --git a/arch/um/os-Linux/skas/process.c b/arch/um/os-Linux/skas/process.c
index 172ad8f72e12..d6e0a2234b86 100644
--- a/arch/um/os-Linux/skas/process.c
+++ b/arch/um/os-Linux/skas/process.c
@@ -96,7 +96,7 @@ bad_wait:
 
 extern unsigned long current_stub_stack(void);
 
-void get_skas_faultinfo(int pid, struct faultinfo * fi)
+static void get_skas_faultinfo(int pid, struct faultinfo *fi)
 {
 	int err;
 
diff --git a/arch/um/os-Linux/umid.c b/arch/um/os-Linux/umid.c
index 106fa8641553..a27defb81884 100644
--- a/arch/um/os-Linux/umid.c
+++ b/arch/um/os-Linux/umid.c
@@ -245,7 +245,7 @@ int __init set_umid(char *name)
 /* Changed in make_umid, which is called during early boot */
 static int umid_setup = 0;
 
-int __init make_umid(void)
+static int __init make_umid(void)
 {
 	int fd, err;
 	char tmp[256];
diff --git a/arch/um/sys-i386/bugs.c b/arch/um/sys-i386/bugs.c
index a74442d13762..2c6d0d731c12 100644
--- a/arch/um/sys-i386/bugs.c
+++ b/arch/um/sys-i386/bugs.c
@@ -12,7 +12,7 @@
 #include "sysdep/ptrace.h"
 
 /* Set during early boot */
-int host_has_cmov = 1;
+static int host_has_cmov = 1;
 static jmp_buf cmov_test_return;
 
 static void cmov_sigill_test_handler(int sig)
diff --git a/arch/um/sys-i386/checksum.S b/arch/um/sys-i386/checksum.S
index 62c7e564f22e..f058d2f82e18 100644
--- a/arch/um/sys-i386/checksum.S
+++ b/arch/um/sys-i386/checksum.S
@@ -243,13 +243,12 @@ unsigned int csum_partial_copy_generic (const char *src, char *dst,
 	.previous
 
 .align 4
-.globl csum_partial_copy_generic_i386
-				
+
 #ifndef CONFIG_X86_USE_PPRO_CHECKSUM
 
 #define ARGBASE 16		
 #define FP		12
-		
+
 csum_partial_copy_generic_i386:
 	subl  $4,%esp	
 	pushl %edi
diff --git a/arch/um/sys-i386/ldt.c b/arch/um/sys-i386/ldt.c
index a34263e6b08d..a4846a84a7be 100644
--- a/arch/um/sys-i386/ldt.c
+++ b/arch/um/sys-i386/ldt.c
@@ -14,8 +14,8 @@
 
 extern int modify_ldt(int func, void *ptr, unsigned long bytecount);
 
-long write_ldt_entry(struct mm_id * mm_idp, int func, struct user_desc * desc,
-		     void **addr, int done)
+static long write_ldt_entry(struct mm_id *mm_idp, int func,
+		     struct user_desc *desc, void **addr, int done)
 {
 	long res;
 
diff --git a/include/asm-um/ptrace-generic.h b/include/asm-um/ptrace-generic.h
index 6aefcd32fc61..315749705ea1 100644
--- a/include/asm-um/ptrace-generic.h
+++ b/include/asm-um/ptrace-generic.h
@@ -47,9 +47,6 @@ extern int set_fpregs(struct user_i387_struct __user *buf,
 
 extern void show_regs(struct pt_regs *regs);
 
-extern void send_sigtrap(struct task_struct *tsk, struct uml_pt_regs *regs,
-			 int error_code);
-
 extern int arch_copy_tls(struct task_struct *new);
 extern void clear_flushed_tls(struct task_struct *task);
 

From f606ddf42fd4edc558eeb48bfee66d2c591571d2 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Wed, 23 Jul 2008 21:28:50 -0700
Subject: [PATCH 0326/1171] remove the v850 port

Trying to compile the v850 port brings many compile errors, one of them exists
since at least kernel 2.6.19.

There also seems to be noone willing to bring this port back into a usable
state.

This patch therefore removes the v850 port.

If anyone ever decides to revive the v850 port the code will still be
available from older kernels, and it wouldn't be impossible for the port to
reenter the kernel if it would become actively maintained again.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Acked-by: Greg Ungerer <gerg@uclinux.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 MAINTAINERS                            |    3 -
 arch/v850/Kconfig                      |  353 --------
 arch/v850/Kconfig.debug                |   10 -
 arch/v850/Makefile                     |   54 --
 arch/v850/README                       |   44 -
 arch/v850/configs/rte-ma1-cb_defconfig |  617 -------------
 arch/v850/configs/rte-me2-cb_defconfig |  462 ----------
 arch/v850/configs/sim_defconfig        |  451 ----------
 arch/v850/kernel/Makefile              |   40 -
 arch/v850/kernel/anna-rom.ld           |   16 -
 arch/v850/kernel/anna.c                |  202 -----
 arch/v850/kernel/anna.ld               |   20 -
 arch/v850/kernel/as85ep1-rom.ld        |   21 -
 arch/v850/kernel/as85ep1.c             |  234 -----
 arch/v850/kernel/as85ep1.ld            |   49 --
 arch/v850/kernel/asm-offsets.c         |   58 --
 arch/v850/kernel/bug.c                 |  142 ---
 arch/v850/kernel/entry.S               | 1121 ------------------------
 arch/v850/kernel/fpga85e2c.c           |  167 ----
 arch/v850/kernel/fpga85e2c.ld          |   62 --
 arch/v850/kernel/gbus_int.c            |  271 ------
 arch/v850/kernel/head.S                |  128 ---
 arch/v850/kernel/highres_timer.c       |  132 ---
 arch/v850/kernel/init_task.c           |   48 -
 arch/v850/kernel/intv.S                |   87 --
 arch/v850/kernel/irq.c                 |  123 ---
 arch/v850/kernel/ma.c                  |   69 --
 arch/v850/kernel/mach.c                |   17 -
 arch/v850/kernel/mach.h                |   56 --
 arch/v850/kernel/me2.c                 |   73 --
 arch/v850/kernel/memcons.c             |  135 ---
 arch/v850/kernel/module.c              |  237 -----
 arch/v850/kernel/process.c             |  217 -----
 arch/v850/kernel/procfs.c              |   67 --
 arch/v850/kernel/ptrace.c              |  235 -----
 arch/v850/kernel/rte_cb.c              |  193 ----
 arch/v850/kernel/rte_cb_leds.c         |  137 ---
 arch/v850/kernel/rte_cb_multi.c        |  121 ---
 arch/v850/kernel/rte_ma1_cb-rom.ld     |   14 -
 arch/v850/kernel/rte_ma1_cb.c          |  107 ---
 arch/v850/kernel/rte_ma1_cb.ld         |   57 --
 arch/v850/kernel/rte_mb_a_pci.c        |  819 -----------------
 arch/v850/kernel/rte_me2_cb.c          |  298 -------
 arch/v850/kernel/rte_me2_cb.ld         |   30 -
 arch/v850/kernel/rte_nb85e_cb-multi.ld |   57 --
 arch/v850/kernel/rte_nb85e_cb.c        |   81 --
 arch/v850/kernel/rte_nb85e_cb.ld       |   22 -
 arch/v850/kernel/setup.c               |  329 -------
 arch/v850/kernel/signal.c              |  523 -----------
 arch/v850/kernel/sim.c                 |  172 ----
 arch/v850/kernel/sim.ld                |   13 -
 arch/v850/kernel/sim85e2.c             |  195 -----
 arch/v850/kernel/sim85e2.ld            |   36 -
 arch/v850/kernel/simcons.c             |  161 ----
 arch/v850/kernel/syscalls.c            |  196 -----
 arch/v850/kernel/teg.c                 |   62 --
 arch/v850/kernel/time.c                |  106 ---
 arch/v850/kernel/v850_ksyms.c          |   51 --
 arch/v850/kernel/v850e2_cache.c        |  127 ---
 arch/v850/kernel/v850e_cache.c         |  174 ----
 arch/v850/kernel/v850e_intc.c          |  104 ---
 arch/v850/kernel/v850e_timer_d.c       |   54 --
 arch/v850/kernel/v850e_utils.c         |   62 --
 arch/v850/kernel/vmlinux.lds.S         |  306 -------
 arch/v850/lib/Makefile                 |    6 -
 arch/v850/lib/ashldi3.c                |   62 --
 arch/v850/lib/ashrdi3.c                |   63 --
 arch/v850/lib/checksum.c               |  155 ----
 arch/v850/lib/lshrdi3.c                |   62 --
 arch/v850/lib/memcpy.c                 |   92 --
 arch/v850/lib/memset.c                 |   68 --
 arch/v850/lib/muldi3.c                 |   61 --
 arch/v850/lib/negdi2.c                 |   25 -
 drivers/serial/Kconfig                 |   16 -
 drivers/watchdog/Kconfig               |    2 -
 drivers/watchdog/Makefile              |    2 -
 include/asm-v850/Kbuild                |    1 -
 include/asm-v850/a.out.h               |   21 -
 include/asm-v850/anna.h                |  137 ---
 include/asm-v850/as85ep1.h             |  152 ----
 include/asm-v850/asm.h                 |   32 -
 include/asm-v850/atomic.h              |  131 ---
 include/asm-v850/auxvec.h              |    4 -
 include/asm-v850/bitops.h              |  161 ----
 include/asm-v850/bug.h                 |   25 -
 include/asm-v850/bugs.h                |   16 -
 include/asm-v850/byteorder.h           |   48 -
 include/asm-v850/cache.h               |   26 -
 include/asm-v850/cacheflush.h          |   70 --
 include/asm-v850/checksum.h            |  112 ---
 include/asm-v850/clinkage.h            |   26 -
 include/asm-v850/cputime.h             |    6 -
 include/asm-v850/current.h             |   47 -
 include/asm-v850/delay.h               |   47 -
 include/asm-v850/device.h              |    7 -
 include/asm-v850/div64.h               |    1 -
 include/asm-v850/dma-mapping.h         |   11 -
 include/asm-v850/dma.h                 |   18 -
 include/asm-v850/elf.h                 |   99 ---
 include/asm-v850/emergency-restart.h   |    6 -
 include/asm-v850/entry.h               |  113 ---
 include/asm-v850/errno.h               |    6 -
 include/asm-v850/fb.h                  |   12 -
 include/asm-v850/fcntl.h               |   11 -
 include/asm-v850/flat.h                |  133 ---
 include/asm-v850/fpga85e2c.h           |   82 --
 include/asm-v850/futex.h               |    6 -
 include/asm-v850/gbus_int.h            |   97 --
 include/asm-v850/hardirq.h             |   28 -
 include/asm-v850/highres_timer.h       |   44 -
 include/asm-v850/hw_irq.h              |    4 -
 include/asm-v850/io.h                  |  142 ---
 include/asm-v850/ioctl.h               |    1 -
 include/asm-v850/ioctls.h              |   84 --
 include/asm-v850/ipcbuf.h              |   29 -
 include/asm-v850/irq.h                 |   55 --
 include/asm-v850/irq_regs.h            |    1 -
 include/asm-v850/kdebug.h              |    1 -
 include/asm-v850/kmap_types.h          |   19 -
 include/asm-v850/kvm.h                 |    6 -
 include/asm-v850/linkage.h             |    8 -
 include/asm-v850/local.h               |    6 -
 include/asm-v850/ma.h                  |  101 ---
 include/asm-v850/ma1.h                 |   50 --
 include/asm-v850/machdep.h             |   60 --
 include/asm-v850/macrology.h           |   17 -
 include/asm-v850/me2.h                 |  182 ----
 include/asm-v850/mman.h                |   15 -
 include/asm-v850/mmu.h                 |   11 -
 include/asm-v850/mmu_context.h         |   13 -
 include/asm-v850/module.h              |   62 --
 include/asm-v850/msgbuf.h              |   31 -
 include/asm-v850/mutex.h               |    9 -
 include/asm-v850/page.h                |  124 ---
 include/asm-v850/param.h               |   33 -
 include/asm-v850/pci.h                 |  119 ---
 include/asm-v850/percpu.h              |   14 -
 include/asm-v850/pgalloc.h             |   22 -
 include/asm-v850/pgtable.h             |   59 --
 include/asm-v850/poll.h                |    9 -
 include/asm-v850/posix_types.h         |   72 --
 include/asm-v850/processor.h           |  120 ---
 include/asm-v850/ptrace.h              |  121 ---
 include/asm-v850/resource.h            |    6 -
 include/asm-v850/rte_cb.h              |   78 --
 include/asm-v850/rte_ma1_cb.h          |  128 ---
 include/asm-v850/rte_mb_a_pci.h        |   56 --
 include/asm-v850/rte_me2_cb.h          |  202 -----
 include/asm-v850/rte_nb85e_cb.h        |  111 ---
 include/asm-v850/scatterlist.h         |   31 -
 include/asm-v850/sections.h            |    6 -
 include/asm-v850/segment.h             |   36 -
 include/asm-v850/semaphore.h           |    1 -
 include/asm-v850/sembuf.h              |   25 -
 include/asm-v850/serial.h              |   56 --
 include/asm-v850/setup.h               |    6 -
 include/asm-v850/shmbuf.h              |   42 -
 include/asm-v850/shmparam.h            |    6 -
 include/asm-v850/sigcontext.h          |   25 -
 include/asm-v850/siginfo.h             |    6 -
 include/asm-v850/signal.h              |  168 ----
 include/asm-v850/sim.h                 |   47 -
 include/asm-v850/sim85e2.h             |   69 --
 include/asm-v850/sim85e2c.h            |   26 -
 include/asm-v850/sim85e2s.h            |   28 -
 include/asm-v850/simsyscall.h          |   99 ---
 include/asm-v850/socket.h              |   57 --
 include/asm-v850/sockios.h             |   13 -
 include/asm-v850/stat.h                |   73 --
 include/asm-v850/statfs.h              |    6 -
 include/asm-v850/string.h              |   25 -
 include/asm-v850/system.h              |  123 ---
 include/asm-v850/teg.h                 |  101 ---
 include/asm-v850/termbits.h            |  200 -----
 include/asm-v850/termios.h             |   90 --
 include/asm-v850/thread_info.h         |  129 ---
 include/asm-v850/timex.h               |   18 -
 include/asm-v850/tlb.h                 |   21 -
 include/asm-v850/tlbflush.h            |   64 --
 include/asm-v850/topology.h            |    6 -
 include/asm-v850/types.h               |   36 -
 include/asm-v850/uaccess.h             |  159 ----
 include/asm-v850/ucontext.h            |   14 -
 include/asm-v850/unaligned.h           |   22 -
 include/asm-v850/unistd.h              |  244 ------
 include/asm-v850/user.h                |   52 --
 include/asm-v850/v850e.h               |   21 -
 include/asm-v850/v850e2.h              |   69 --
 include/asm-v850/v850e2_cache.h        |   75 --
 include/asm-v850/v850e_cache.h         |   48 -
 include/asm-v850/v850e_intc.h          |  133 ---
 include/asm-v850/v850e_timer_c.h       |   48 -
 include/asm-v850/v850e_timer_d.h       |   62 --
 include/asm-v850/v850e_uart.h          |   76 --
 include/asm-v850/v850e_uarta.h         |  278 ------
 include/asm-v850/v850e_uartb.h         |  262 ------
 include/asm-v850/v850e_utils.h         |   35 -
 include/linux/audit.h                  |    1 -
 include/linux/module.h                 |    2 +-
 include/linux/serial_core.h            |    3 -
 include/linux/syscalls.h               |    2 +-
 scripts/genksyms/genksyms.c            |    3 +-
 scripts/mod/file2alias.c               |    2 +-
 scripts/mod/mk_elfconfig.c             |    2 +-
 204 files changed, 5 insertions(+), 18406 deletions(-)
 delete mode 100644 arch/v850/Kconfig
 delete mode 100644 arch/v850/Kconfig.debug
 delete mode 100644 arch/v850/Makefile
 delete mode 100644 arch/v850/README
 delete mode 100644 arch/v850/configs/rte-ma1-cb_defconfig
 delete mode 100644 arch/v850/configs/rte-me2-cb_defconfig
 delete mode 100644 arch/v850/configs/sim_defconfig
 delete mode 100644 arch/v850/kernel/Makefile
 delete mode 100644 arch/v850/kernel/anna-rom.ld
 delete mode 100644 arch/v850/kernel/anna.c
 delete mode 100644 arch/v850/kernel/anna.ld
 delete mode 100644 arch/v850/kernel/as85ep1-rom.ld
 delete mode 100644 arch/v850/kernel/as85ep1.c
 delete mode 100644 arch/v850/kernel/as85ep1.ld
 delete mode 100644 arch/v850/kernel/asm-offsets.c
 delete mode 100644 arch/v850/kernel/bug.c
 delete mode 100644 arch/v850/kernel/entry.S
 delete mode 100644 arch/v850/kernel/fpga85e2c.c
 delete mode 100644 arch/v850/kernel/fpga85e2c.ld
 delete mode 100644 arch/v850/kernel/gbus_int.c
 delete mode 100644 arch/v850/kernel/head.S
 delete mode 100644 arch/v850/kernel/highres_timer.c
 delete mode 100644 arch/v850/kernel/init_task.c
 delete mode 100644 arch/v850/kernel/intv.S
 delete mode 100644 arch/v850/kernel/irq.c
 delete mode 100644 arch/v850/kernel/ma.c
 delete mode 100644 arch/v850/kernel/mach.c
 delete mode 100644 arch/v850/kernel/mach.h
 delete mode 100644 arch/v850/kernel/me2.c
 delete mode 100644 arch/v850/kernel/memcons.c
 delete mode 100644 arch/v850/kernel/module.c
 delete mode 100644 arch/v850/kernel/process.c
 delete mode 100644 arch/v850/kernel/procfs.c
 delete mode 100644 arch/v850/kernel/ptrace.c
 delete mode 100644 arch/v850/kernel/rte_cb.c
 delete mode 100644 arch/v850/kernel/rte_cb_leds.c
 delete mode 100644 arch/v850/kernel/rte_cb_multi.c
 delete mode 100644 arch/v850/kernel/rte_ma1_cb-rom.ld
 delete mode 100644 arch/v850/kernel/rte_ma1_cb.c
 delete mode 100644 arch/v850/kernel/rte_ma1_cb.ld
 delete mode 100644 arch/v850/kernel/rte_mb_a_pci.c
 delete mode 100644 arch/v850/kernel/rte_me2_cb.c
 delete mode 100644 arch/v850/kernel/rte_me2_cb.ld
 delete mode 100644 arch/v850/kernel/rte_nb85e_cb-multi.ld
 delete mode 100644 arch/v850/kernel/rte_nb85e_cb.c
 delete mode 100644 arch/v850/kernel/rte_nb85e_cb.ld
 delete mode 100644 arch/v850/kernel/setup.c
 delete mode 100644 arch/v850/kernel/signal.c
 delete mode 100644 arch/v850/kernel/sim.c
 delete mode 100644 arch/v850/kernel/sim.ld
 delete mode 100644 arch/v850/kernel/sim85e2.c
 delete mode 100644 arch/v850/kernel/sim85e2.ld
 delete mode 100644 arch/v850/kernel/simcons.c
 delete mode 100644 arch/v850/kernel/syscalls.c
 delete mode 100644 arch/v850/kernel/teg.c
 delete mode 100644 arch/v850/kernel/time.c
 delete mode 100644 arch/v850/kernel/v850_ksyms.c
 delete mode 100644 arch/v850/kernel/v850e2_cache.c
 delete mode 100644 arch/v850/kernel/v850e_cache.c
 delete mode 100644 arch/v850/kernel/v850e_intc.c
 delete mode 100644 arch/v850/kernel/v850e_timer_d.c
 delete mode 100644 arch/v850/kernel/v850e_utils.c
 delete mode 100644 arch/v850/kernel/vmlinux.lds.S
 delete mode 100644 arch/v850/lib/Makefile
 delete mode 100644 arch/v850/lib/ashldi3.c
 delete mode 100644 arch/v850/lib/ashrdi3.c
 delete mode 100644 arch/v850/lib/checksum.c
 delete mode 100644 arch/v850/lib/lshrdi3.c
 delete mode 100644 arch/v850/lib/memcpy.c
 delete mode 100644 arch/v850/lib/memset.c
 delete mode 100644 arch/v850/lib/muldi3.c
 delete mode 100644 arch/v850/lib/negdi2.c
 delete mode 100644 include/asm-v850/Kbuild
 delete mode 100644 include/asm-v850/a.out.h
 delete mode 100644 include/asm-v850/anna.h
 delete mode 100644 include/asm-v850/as85ep1.h
 delete mode 100644 include/asm-v850/asm.h
 delete mode 100644 include/asm-v850/atomic.h
 delete mode 100644 include/asm-v850/auxvec.h
 delete mode 100644 include/asm-v850/bitops.h
 delete mode 100644 include/asm-v850/bug.h
 delete mode 100644 include/asm-v850/bugs.h
 delete mode 100644 include/asm-v850/byteorder.h
 delete mode 100644 include/asm-v850/cache.h
 delete mode 100644 include/asm-v850/cacheflush.h
 delete mode 100644 include/asm-v850/checksum.h
 delete mode 100644 include/asm-v850/clinkage.h
 delete mode 100644 include/asm-v850/cputime.h
 delete mode 100644 include/asm-v850/current.h
 delete mode 100644 include/asm-v850/delay.h
 delete mode 100644 include/asm-v850/device.h
 delete mode 100644 include/asm-v850/div64.h
 delete mode 100644 include/asm-v850/dma-mapping.h
 delete mode 100644 include/asm-v850/dma.h
 delete mode 100644 include/asm-v850/elf.h
 delete mode 100644 include/asm-v850/emergency-restart.h
 delete mode 100644 include/asm-v850/entry.h
 delete mode 100644 include/asm-v850/errno.h
 delete mode 100644 include/asm-v850/fb.h
 delete mode 100644 include/asm-v850/fcntl.h
 delete mode 100644 include/asm-v850/flat.h
 delete mode 100644 include/asm-v850/fpga85e2c.h
 delete mode 100644 include/asm-v850/futex.h
 delete mode 100644 include/asm-v850/gbus_int.h
 delete mode 100644 include/asm-v850/hardirq.h
 delete mode 100644 include/asm-v850/highres_timer.h
 delete mode 100644 include/asm-v850/hw_irq.h
 delete mode 100644 include/asm-v850/io.h
 delete mode 100644 include/asm-v850/ioctl.h
 delete mode 100644 include/asm-v850/ioctls.h
 delete mode 100644 include/asm-v850/ipcbuf.h
 delete mode 100644 include/asm-v850/irq.h
 delete mode 100644 include/asm-v850/irq_regs.h
 delete mode 100644 include/asm-v850/kdebug.h
 delete mode 100644 include/asm-v850/kmap_types.h
 delete mode 100644 include/asm-v850/kvm.h
 delete mode 100644 include/asm-v850/linkage.h
 delete mode 100644 include/asm-v850/local.h
 delete mode 100644 include/asm-v850/ma.h
 delete mode 100644 include/asm-v850/ma1.h
 delete mode 100644 include/asm-v850/machdep.h
 delete mode 100644 include/asm-v850/macrology.h
 delete mode 100644 include/asm-v850/me2.h
 delete mode 100644 include/asm-v850/mman.h
 delete mode 100644 include/asm-v850/mmu.h
 delete mode 100644 include/asm-v850/mmu_context.h
 delete mode 100644 include/asm-v850/module.h
 delete mode 100644 include/asm-v850/msgbuf.h
 delete mode 100644 include/asm-v850/mutex.h
 delete mode 100644 include/asm-v850/page.h
 delete mode 100644 include/asm-v850/param.h
 delete mode 100644 include/asm-v850/pci.h
 delete mode 100644 include/asm-v850/percpu.h
 delete mode 100644 include/asm-v850/pgalloc.h
 delete mode 100644 include/asm-v850/pgtable.h
 delete mode 100644 include/asm-v850/poll.h
 delete mode 100644 include/asm-v850/posix_types.h
 delete mode 100644 include/asm-v850/processor.h
 delete mode 100644 include/asm-v850/ptrace.h
 delete mode 100644 include/asm-v850/resource.h
 delete mode 100644 include/asm-v850/rte_cb.h
 delete mode 100644 include/asm-v850/rte_ma1_cb.h
 delete mode 100644 include/asm-v850/rte_mb_a_pci.h
 delete mode 100644 include/asm-v850/rte_me2_cb.h
 delete mode 100644 include/asm-v850/rte_nb85e_cb.h
 delete mode 100644 include/asm-v850/scatterlist.h
 delete mode 100644 include/asm-v850/sections.h
 delete mode 100644 include/asm-v850/segment.h
 delete mode 100644 include/asm-v850/semaphore.h
 delete mode 100644 include/asm-v850/sembuf.h
 delete mode 100644 include/asm-v850/serial.h
 delete mode 100644 include/asm-v850/setup.h
 delete mode 100644 include/asm-v850/shmbuf.h
 delete mode 100644 include/asm-v850/shmparam.h
 delete mode 100644 include/asm-v850/sigcontext.h
 delete mode 100644 include/asm-v850/siginfo.h
 delete mode 100644 include/asm-v850/signal.h
 delete mode 100644 include/asm-v850/sim.h
 delete mode 100644 include/asm-v850/sim85e2.h
 delete mode 100644 include/asm-v850/sim85e2c.h
 delete mode 100644 include/asm-v850/sim85e2s.h
 delete mode 100644 include/asm-v850/simsyscall.h
 delete mode 100644 include/asm-v850/socket.h
 delete mode 100644 include/asm-v850/sockios.h
 delete mode 100644 include/asm-v850/stat.h
 delete mode 100644 include/asm-v850/statfs.h
 delete mode 100644 include/asm-v850/string.h
 delete mode 100644 include/asm-v850/system.h
 delete mode 100644 include/asm-v850/teg.h
 delete mode 100644 include/asm-v850/termbits.h
 delete mode 100644 include/asm-v850/termios.h
 delete mode 100644 include/asm-v850/thread_info.h
 delete mode 100644 include/asm-v850/timex.h
 delete mode 100644 include/asm-v850/tlb.h
 delete mode 100644 include/asm-v850/tlbflush.h
 delete mode 100644 include/asm-v850/topology.h
 delete mode 100644 include/asm-v850/types.h
 delete mode 100644 include/asm-v850/uaccess.h
 delete mode 100644 include/asm-v850/ucontext.h
 delete mode 100644 include/asm-v850/unaligned.h
 delete mode 100644 include/asm-v850/unistd.h
 delete mode 100644 include/asm-v850/user.h
 delete mode 100644 include/asm-v850/v850e.h
 delete mode 100644 include/asm-v850/v850e2.h
 delete mode 100644 include/asm-v850/v850e2_cache.h
 delete mode 100644 include/asm-v850/v850e_cache.h
 delete mode 100644 include/asm-v850/v850e_intc.h
 delete mode 100644 include/asm-v850/v850e_timer_c.h
 delete mode 100644 include/asm-v850/v850e_timer_d.h
 delete mode 100644 include/asm-v850/v850e_uart.h
 delete mode 100644 include/asm-v850/v850e_uarta.h
 delete mode 100644 include/asm-v850/v850e_uartb.h
 delete mode 100644 include/asm-v850/v850e_utils.h

diff --git a/MAINTAINERS b/MAINTAINERS
index 7ffd78c4e277..7e5c7b0290bb 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -4131,9 +4131,6 @@ W:	http://www.uclinux.org/
 L:	uclinux-dev@uclinux.org  (subscribers-only)
 S:	Maintained
 
-UCLINUX FOR NEC V850
-P:	Miles Bader
-
 UCLINUX FOR RENESAS H8/300
 P:	Yoshinori Sato
 M:	ysato@users.sourceforge.jp
diff --git a/arch/v850/Kconfig b/arch/v850/Kconfig
deleted file mode 100644
index 4379f43505ef..000000000000
--- a/arch/v850/Kconfig
+++ /dev/null
@@ -1,353 +0,0 @@
-#############################################################################
-#
-# For a description of the syntax of this configuration file,
-# see Documentation/kbuild/kconfig-language.txt.
-#
-#############################################################################
-
-mainmenu "uClinux/v850 (w/o MMU) Kernel Configuration"
-
-config MMU
-       	bool
-	default n
-config ZONE_DMA
-	bool
-	default y
-config RWSEM_GENERIC_SPINLOCK
-	bool
-	default y
-config RWSEM_XCHGADD_ALGORITHM
-	bool
-	default n
-config GENERIC_FIND_NEXT_BIT
-	bool
-	default y
-config GENERIC_HWEIGHT
-	bool
-	default y
-config GENERIC_CALIBRATE_DELAY
-	bool
-	default y
-
-config GENERIC_HARDIRQS
-	bool
-	default y
-
-config GENERIC_IRQ_PROBE
-	bool
-	default y
-
-config GENERIC_TIME
-	bool
-	default y
-
-config TIME_LOW_RES
-	bool
-	default y
-
-config ARCH_HAS_ILOG2_U32
-	bool
-	default n
-
-config ARCH_HAS_ILOG2_U64
-	bool
-	default n
-
-config ARCH_SUPPORTS_AOUT
-	def_bool y
-
-# Turn off some random 386 crap that can affect device config
-config ISA
-	bool
-	default n
-config ISAPNP
-	bool
-	default n
-config EISA
-	bool
-	default n
-config MCA
-	bool
-	default n
-
-
-#############################################################################
-#### v850-specific config
-
-# Define the architecture
-config V850
-	bool
-	default y
-	select HAVE_IDE
-
-menu "Processor type and features"
-
-   choice
-	  prompt "Platform"
-	  default GDB
-      config V850E_SIM
-      	     bool "GDB"
-      config RTE_CB_MA1
-      	     bool "RTE-V850E/MA1-CB"
-      config RTE_CB_NB85E
-      	     bool "RTE-V850E/NB85E-CB"
-      config RTE_CB_ME2
-      	     bool "RTE-V850E/ME2-CB"
-      config V850E_AS85EP1
-      	     bool "AS85EP1"
-      config V850E2_SIM85E2C
-      	     bool "sim85e2c"
-      config V850E2_SIM85E2S
-      	     bool "sim85e2s"
-      config V850E2_FPGA85E2C
-      	     bool "NA85E2C-FPGA"
-      config V850E2_ANNA
-      	     bool "Anna"
-   endchoice
-
-   #### V850E processor-specific config
-
-   # All CPUs currently supported use the v850e architecture
-   config V850E
-   	  bool
-	  default y
-
-   # The RTE-V850E/MA1-CB is the only type of V850E/MA1 platform we
-   # currently support
-   config V850E_MA1
-   	  bool
-	  depends on RTE_CB_MA1
-	  default y
-   # Similarly for the RTE-V850E/NB85E-CB - V850E/TEG
-   config V850E_TEG
-   	  bool
-	  depends on RTE_CB_NB85E
-	  default y
-   # ... and the RTE-V850E/ME2-CB - V850E/ME2
-   config V850E_ME2
-   	  bool
-	  depends on RTE_CB_ME2
-	  default y
-
-
-   #### sim85e2-specific config
-
-   config V850E2_SIM85E2
-   	  bool
-	  depends on V850E2_SIM85E2C || V850E2_SIM85E2S
-	  default y
-
-
-   #### V850E2 processor-specific config
-
-   # V850E2 processors
-   config V850E2
-   	  bool
-	  depends on V850E2_SIM85E2 || V850E2_FPGA85E2C || V850E2_ANNA
-	  default y
-
-
-   #### RTE-CB platform-specific config
-
-   # Boards in the RTE-x-CB series
-   config RTE_CB
-   	  bool
-	  depends on RTE_CB_MA1 || RTE_CB_NB85E || RTE_CB_ME2
-	  default y
-
-   config RTE_CB_MULTI
-   	  bool
-	  # RTE_CB_NB85E can either have multi ROM support or not, but
-	  # other platforms (currently only RTE_CB_MA1) require it.
-	  prompt "Multi monitor ROM support" if RTE_CB_NB85E
-	  depends on RTE_CB_MA1 || RTE_CB_NB85E
-	  default y
-
-   config RTE_CB_MULTI_DBTRAP
-   	  bool "Pass illegal insn trap / dbtrap to kernel"
-	  depends on RTE_CB_MULTI
-	  default n
-
-   config RTE_CB_MA1_KSRAM
-   	  bool "Kernel in SRAM (limits size of kernel)"
-	  depends on RTE_CB_MA1 && RTE_CB_MULTI
-	  default n
-
-   config RTE_MB_A_PCI
-   	  bool "Mother-A PCI support"
-	  depends on RTE_CB
-	  default y
-
-   # The GBUS is used to talk to the RTE-MOTHER-A board
-   config RTE_GBUS_INT
-   	  bool
-	  depends on RTE_MB_A_PCI
-	  default y
-
-   # The only PCI bus we support is on the RTE-MOTHER-A board
-   config PCI
-   	  bool
-	  default RTE_MB_A_PCI
-
-   #### Some feature-specific configs
-
-   # Everything except for the GDB simulator uses the same interrupt controller
-   config V850E_INTC
-   	  bool
-	  default !V850E_SIM
-
-   # Everything except for the various simulators uses the "Timer D" unit
-   config V850E_TIMER_D
-   	  bool
-	  default !V850E_SIM && !V850E2_SIM85E2
-
-   # Cache control used on some v850e1 processors
-   config V850E_CACHE
-          bool
-	  default V850E_TEG || V850E_ME2
-
-   # Cache control used on v850e2 processors; I think this should
-   # actually apply to more, but currently only the SIM85E2S uses it
-   config V850E2_CACHE
-   	  bool
-	  default V850E2_SIM85E2S
-
-   config NO_CACHE
-   	  bool
-	  default !V850E_CACHE && !V850E2_CACHE
-
-   # HZ depends on the platform
-   config HZ
-	  int
-	  default 24  if V850E_SIM || V850E2_SIM85E2
-	  default 122 if V850E2_FPGA85E2C
-	  default 100
-
-   #### Misc config
-
-   config ROM_KERNEL
-   	  bool "Kernel in ROM"
-	  depends on V850E2_ANNA || V850E_AS85EP1 || RTE_CB_ME2
-
-   # Some platforms pre-zero memory, in which case the kernel doesn't need to
-   config ZERO_BSS
-   	  bool
-	  depends on !V850E2_SIM85E2C
-	  default y
-
-   # The crappy-ass zone allocator requires that the start of allocatable
-   # memory be aligned to the largest possible allocation.
-   config FORCE_MAX_ZONEORDER
-   	  int
-	  default 8 if V850E2_SIM85E2C || V850E2_FPGA85E2C
-
-   config V850E_HIGHRES_TIMER
-   	  bool "High resolution timer support"
-	  depends on V850E_TIMER_D
-   config TIME_BOOTUP
-   	  bool "Time bootup"
-	  depends on V850E_HIGHRES_TIMER
-
-   config RESET_GUARD
-   	  bool "Reset Guard"
-
-source "mm/Kconfig"
-
-endmenu
-
-
-#############################################################################
-
-source init/Kconfig
-
-#############################################################################
-
-menu "Bus options (PCI, PCMCIA, EISA, MCA, ISA)"
-
-#    config PCI
-# 	   bool "PCI support"
-# 	   help
-# 	     Support for PCI bus.
-
-source "drivers/pci/Kconfig"
-
-source "drivers/pcmcia/Kconfig"
-
-source "drivers/pci/hotplug/Kconfig"
-
-endmenu
-
-menu "Executable file formats"
-
-source "fs/Kconfig.binfmt"
-
-endmenu
-
-source "net/Kconfig"
-
-#############################################################################
-
-source "drivers/base/Kconfig"
-
-source drivers/mtd/Kconfig
-
-source drivers/parport/Kconfig
-
-#source drivers/pnp/Kconfig
-
-source drivers/block/Kconfig
-
-#############################################################################
-
-menu "Disk device support"
-
-source "drivers/ide/Kconfig"
-
-source "drivers/scsi/Kconfig"
-
-endmenu
-
-#############################################################################
-
-
-source "drivers/md/Kconfig"
-
-source "drivers/message/fusion/Kconfig"
-
-source "drivers/ieee1394/Kconfig"
-
-source "drivers/message/i2o/Kconfig"
-
-source "drivers/net/Kconfig"
-
-source "drivers/isdn/Kconfig"
-
-#source "drivers/telephony/Kconfig"
-
-#
-# input before char - char/joystick depends on it. As does USB.
-#
-source "drivers/input/Kconfig"
-
-source "drivers/char/Kconfig"
-
-#source drivers/misc/Config.in
-source "drivers/media/Kconfig"
-
-source "fs/Kconfig"
-
-source "drivers/video/Kconfig"
-
-source "sound/Kconfig"
-
-source "drivers/usb/Kconfig"
-
-source "arch/v850/Kconfig.debug"
-
-source "security/Kconfig"
-
-source "crypto/Kconfig"
-
-source "lib/Kconfig"
-
-#############################################################################
diff --git a/arch/v850/Kconfig.debug b/arch/v850/Kconfig.debug
deleted file mode 100644
index 4acfb9cca1ca..000000000000
--- a/arch/v850/Kconfig.debug
+++ /dev/null
@@ -1,10 +0,0 @@
-menu "Kernel hacking"
-
-source "lib/Kconfig.debug"
-
-config NO_KERNEL_MSG
-	bool "Suppress Kernel BUG Messages"
-	help
-	  Do not output any debug BUG messages within the kernel.
-
-endmenu
diff --git a/arch/v850/Makefile b/arch/v850/Makefile
deleted file mode 100644
index 8b629df0029a..000000000000
--- a/arch/v850/Makefile
+++ /dev/null
@@ -1,54 +0,0 @@
-#
-# arch/v850/Makefile
-#
-#  Copyright (C) 2001,02,03,05  NEC Corporation
-#  Copyright (C) 2001,02,03,05  Miles Bader <miles@gnu.org>
-#
-# This file is included by the global makefile so that you can add your own
-# architecture-specific flags and dependencies. Remember to do have actions
-# for "archclean" and "archdep" for cleaning up and making dependencies for
-# this architecture
-#
-# This file is subject to the terms and conditions of the GNU General Public
-# License.  See the file "COPYING" in the main directory of this archive
-# for more details.
-#
-
-arch_dir = arch/v850
-
-KBUILD_CFLAGS += -mv850e
-# r16 is a fixed pointer to the current task
-KBUILD_CFLAGS += -ffixed-r16 -mno-prolog-function
-KBUILD_CFLAGS += -fno-builtin
-KBUILD_CFLAGS += -D__linux__ -DUTS_SYSNAME=\"uClinux\"
-
-# By default, build a kernel that runs on the gdb v850 simulator.
-KBUILD_DEFCONFIG := sim_defconfig
-
-# This prevents the linker from consolidating the .gnu.linkonce.this_module
-# section into .text (which the v850 default linker script for -r does for
-# some reason)
-LDFLAGS_MODULE += --unique=.gnu.linkonce.this_module
-
-OBJCOPY_FLAGS_BLOB := -I binary -O elf32-little -B v850e
-
-
-head-y := $(arch_dir)/kernel/head.o $(arch_dir)/kernel/init_task.o
-core-y += $(arch_dir)/kernel/
-libs-y += $(arch_dir)/lib/
-
-
-# Deal with the initial contents of the root device
-ifdef ROOT_FS_IMAGE
-core-y += root_fs_image.o
-
-# Because the kernel build-system erases all explicit .o build rules, we
-# have to use an intermediate target to fool it into building for us.
-# This results in it being built anew each time, but that's alright.
-root_fs_image.o: root_fs_image_force
-
-root_fs_image_force: $(ROOT_FS_IMAGE)
-	$(OBJCOPY) $(OBJCOPY_FLAGS_BLOB) --rename-section .data=.root,alloc,load,readonly,data,contents $< root_fs_image.o
-endif
-
-CLEAN_FILES += root_fs_image.o
diff --git a/arch/v850/README b/arch/v850/README
deleted file mode 100644
index 12f7f7a665e0..000000000000
--- a/arch/v850/README
+++ /dev/null
@@ -1,44 +0,0 @@
-This port to the NEC V850E processor supports the following platforms:
-
-   "sim"
-	The gdb v850e simulator (CONFIG_V850E_SIM).
-
-   "rte-ma1-cb"
-	The Midas labs RTE-V850E/MA1-CB and RTE-V850E/NB85E-CB evaluation
-	boards (CONFIG_RTE_CB_MA1 and CONFIG_RTE_CB_NB85E).  This support
-	has only been tested when running with the Multi-debugger monitor
-	ROM (for the Green Hills Multi debugger).  The optional NEC
-	Solution Gear RTE-MOTHER-A motherboard is also supported, which
-	allows PCI boards to be used (CONFIG_RTE_MB_A_PCI).
-
-   "rte-me2-cb"
-	The Midas labs RTE-V850E/ME2-CB evaluation board (CONFIG_RTE_CB_ME2).
-     	This has only been tested using a kernel downloaded via an ICE
-     	connection using the Multi debugger.  Support for the RTE-MOTHER-A is
-     	present, but hasn't been tested (unlike the other Midas labs cpu
-     	boards, the RTE-V850E/ME2-CB includes an ethernet adaptor).
-
-   "as85ep1"
-	The NEC AS85EP1 V850E evaluation chip/board (CONFIG_V850E_AS85EP1).
-
-   "anna"
-	The NEC `Anna' (board/chip) implementation of the V850E2 processor
-	(CONFIG_V850E2_ANNA).
-
-   "sim85e2c", "sim85e2s"
-   	The sim85e2c and sim85e2s simulators, which are verilog simulations
-	of the V850E2 NA85E2C/NA85E2S cpu cores (CONFIG_V850E2_SIM85E2C and
-	CONFIG_V850E2_SIM85E2S).
-
-   "fpga85e2c"
-	A FPGA implementation of the V850E2 NA85E2C cpu core
-	(CONFIG_V850E2_FPGA85E2C).
-
-To get a default kernel configuration for a particular platform, you can
-use a <platform>_defconfig make target (e.g., "make rte-me2-cb_defconfig");
-to see which default configurations are possible, look in the directory
-"arch/v850/configs".
-
-Porting to anything with a V850E/MA1 or MA2 processor should be simple.
-See the file <asm-v850/machdep.h> and the files it includes for an example of
-how to add platform/chip-specific support.
diff --git a/arch/v850/configs/rte-ma1-cb_defconfig b/arch/v850/configs/rte-ma1-cb_defconfig
deleted file mode 100644
index 1a5beda36e29..000000000000
--- a/arch/v850/configs/rte-ma1-cb_defconfig
+++ /dev/null
@@ -1,617 +0,0 @@
-#
-# Automatically generated make config: don't edit
-# Linux kernel version: 2.6.13-uc0
-# Fri Sep  2 13:54:27 2005
-#
-# CONFIG_MMU is not set
-# CONFIG_UID16 is not set
-CONFIG_RWSEM_GENERIC_SPINLOCK=y
-# CONFIG_RWSEM_XCHGADD_ALGORITHM is not set
-CONFIG_GENERIC_CALIBRATE_DELAY=y
-# CONFIG_ISA is not set
-# CONFIG_ISAPNP is not set
-# CONFIG_EISA is not set
-# CONFIG_MCA is not set
-CONFIG_V850=y
-
-#
-# Processor type and features
-#
-# CONFIG_V850E_SIM is not set
-CONFIG_RTE_CB_MA1=y
-# CONFIG_RTE_CB_NB85E is not set
-# CONFIG_RTE_CB_ME2 is not set
-# CONFIG_V850E_AS85EP1 is not set
-# CONFIG_V850E2_SIM85E2C is not set
-# CONFIG_V850E2_SIM85E2S is not set
-# CONFIG_V850E2_FPGA85E2C is not set
-# CONFIG_V850E2_ANNA is not set
-CONFIG_V850E=y
-CONFIG_V850E_MA1=y
-CONFIG_RTE_CB=y
-CONFIG_RTE_CB_MULTI=y
-CONFIG_RTE_CB_MULTI_DBTRAP=y
-# CONFIG_RTE_CB_MA1_KSRAM is not set
-CONFIG_RTE_MB_A_PCI=y
-CONFIG_RTE_GBUS_INT=y
-CONFIG_PCI=y
-CONFIG_V850E_INTC=y
-CONFIG_V850E_TIMER_D=y
-# CONFIG_V850E_CACHE is not set
-# CONFIG_V850E2_CACHE is not set
-CONFIG_NO_CACHE=y
-CONFIG_ZERO_BSS=y
-# CONFIG_V850E_HIGHRES_TIMER is not set
-# CONFIG_RESET_GUARD is not set
-CONFIG_LARGE_ALLOCS=y
-CONFIG_FLATMEM=y
-CONFIG_FLAT_NODE_MEM_MAP=y
-
-#
-# Code maturity level options
-#
-# CONFIG_EXPERIMENTAL is not set
-CONFIG_CLEAN_COMPILE=y
-CONFIG_BROKEN_ON_SMP=y
-CONFIG_INIT_ENV_ARG_LIMIT=32
-
-#
-# General setup
-#
-CONFIG_LOCALVERSION=""
-# CONFIG_BSD_PROCESS_ACCT is not set
-# CONFIG_SYSCTL is not set
-# CONFIG_AUDIT is not set
-# CONFIG_HOTPLUG is not set
-CONFIG_KOBJECT_UEVENT=y
-# CONFIG_IKCONFIG is not set
-CONFIG_EMBEDDED=y
-# CONFIG_KALLSYMS is not set
-CONFIG_PRINTK=y
-CONFIG_BUG=y
-# CONFIG_BASE_FULL is not set
-# CONFIG_FUTEX is not set
-# CONFIG_EPOLL is not set
-CONFIG_CC_OPTIMIZE_FOR_SIZE=y
-CONFIG_CC_ALIGN_FUNCTIONS=0
-CONFIG_CC_ALIGN_LABELS=0
-CONFIG_CC_ALIGN_LOOPS=0
-CONFIG_CC_ALIGN_JUMPS=0
-CONFIG_BASE_SMALL=1
-
-#
-# Loadable module support
-#
-CONFIG_MODULES=y
-CONFIG_MODULE_UNLOAD=y
-CONFIG_OBSOLETE_MODPARM=y
-# CONFIG_MODULE_SRCVERSION_ALL is not set
-CONFIG_KMOD=y
-
-#
-# Bus options (PCI, PCMCIA, EISA, MCA, ISA)
-#
-# CONFIG_PCI_LEGACY_PROC is not set
-# CONFIG_PCI_NAMES is not set
-# CONFIG_PCI_DEBUG is not set
-
-#
-# PCCARD (PCMCIA/CardBus) support
-#
-# CONFIG_PCCARD is not set
-
-#
-# PCI Hotplug Support
-#
-
-#
-# Executable file formats
-#
-CONFIG_BINFMT_FLAT=y
-# CONFIG_BINFMT_ZFLAT is not set
-# CONFIG_BINFMT_SHARED_FLAT is not set
-# CONFIG_BINFMT_MISC is not set
-
-#
-# Networking
-#
-CONFIG_NET=y
-
-#
-# Networking options
-#
-# CONFIG_PACKET is not set
-# CONFIG_UNIX is not set
-# CONFIG_NET_KEY is not set
-CONFIG_INET=y
-# CONFIG_IP_MULTICAST is not set
-# CONFIG_IP_ADVANCED_ROUTER is not set
-CONFIG_IP_FIB_HASH=y
-# CONFIG_IP_PNP is not set
-# CONFIG_NET_IPIP is not set
-# CONFIG_NET_IPGRE is not set
-# CONFIG_SYN_COOKIES is not set
-# CONFIG_INET_AH is not set
-# CONFIG_INET_ESP is not set
-# CONFIG_INET_IPCOMP is not set
-# CONFIG_INET_TUNNEL is not set
-# CONFIG_IP_TCPDIAG is not set
-# CONFIG_IP_TCPDIAG_IPV6 is not set
-# CONFIG_TCP_CONG_ADVANCED is not set
-CONFIG_TCP_CONG_BIC=y
-# CONFIG_IPV6 is not set
-# CONFIG_NETFILTER is not set
-# CONFIG_BRIDGE is not set
-# CONFIG_VLAN_8021Q is not set
-# CONFIG_DECNET is not set
-# CONFIG_LLC2 is not set
-# CONFIG_IPX is not set
-# CONFIG_ATALK is not set
-# CONFIG_NET_SCHED is not set
-# CONFIG_NET_CLS_ROUTE is not set
-
-#
-# Network testing
-#
-# CONFIG_NET_PKTGEN is not set
-# CONFIG_HAMRADIO is not set
-# CONFIG_IRDA is not set
-# CONFIG_BT is not set
-
-#
-# Generic Driver Options
-#
-CONFIG_STANDALONE=y
-CONFIG_PREVENT_FIRMWARE_BUILD=y
-# CONFIG_FW_LOADER is not set
-# CONFIG_DEBUG_DRIVER is not set
-
-#
-# Memory Technology Devices (MTD)
-#
-CONFIG_MTD=y
-# CONFIG_MTD_DEBUG is not set
-# CONFIG_MTD_CONCAT is not set
-# CONFIG_MTD_PARTITIONS is not set
-
-#
-# User Modules And Translation Layers
-#
-# CONFIG_MTD_CHAR is not set
-CONFIG_MTD_BLOCK=y
-# CONFIG_FTL is not set
-# CONFIG_NFTL is not set
-# CONFIG_INFTL is not set
-
-#
-# RAM/ROM/Flash chip drivers
-#
-# CONFIG_MTD_CFI is not set
-# CONFIG_MTD_JEDECPROBE is not set
-CONFIG_MTD_MAP_BANK_WIDTH_1=y
-CONFIG_MTD_MAP_BANK_WIDTH_2=y
-CONFIG_MTD_MAP_BANK_WIDTH_4=y
-# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set
-# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set
-# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set
-CONFIG_MTD_CFI_I1=y
-CONFIG_MTD_CFI_I2=y
-# CONFIG_MTD_CFI_I4 is not set
-# CONFIG_MTD_CFI_I8 is not set
-# CONFIG_MTD_RAM is not set
-# CONFIG_MTD_ROM is not set
-# CONFIG_MTD_ABSENT is not set
-
-#
-# Mapping drivers for chip access
-#
-# CONFIG_MTD_COMPLEX_MAPPINGS is not set
-# CONFIG_MTD_PLATRAM is not set
-
-#
-# Self-contained MTD device drivers
-#
-# CONFIG_MTD_PMC551 is not set
-CONFIG_MTD_SLRAM=y
-# CONFIG_MTD_PHRAM is not set
-# CONFIG_MTD_MTDRAM is not set
-# CONFIG_MTD_BLKMTD is not set
-
-#
-# Disk-On-Chip Device Drivers
-#
-# CONFIG_MTD_DOC2000 is not set
-# CONFIG_MTD_DOC2001 is not set
-# CONFIG_MTD_DOC2001PLUS is not set
-
-#
-# NAND Flash Device Drivers
-#
-# CONFIG_MTD_NAND is not set
-
-#
-# Parallel port support
-#
-# CONFIG_PARPORT is not set
-
-#
-# Block devices
-#
-# CONFIG_BLK_DEV_FD is not set
-# CONFIG_BLK_CPQ_DA is not set
-# CONFIG_BLK_CPQ_CISS_DA is not set
-# CONFIG_BLK_DEV_DAC960 is not set
-# CONFIG_BLK_DEV_COW_COMMON is not set
-# CONFIG_BLK_DEV_LOOP is not set
-# CONFIG_BLK_DEV_NBD is not set
-# CONFIG_BLK_DEV_SX8 is not set
-# CONFIG_BLK_DEV_RAM is not set
-CONFIG_BLK_DEV_RAM_COUNT=16
-CONFIG_INITRAMFS_SOURCE=""
-# CONFIG_CDROM_PKTCDVD is not set
-
-#
-# IO Schedulers
-#
-CONFIG_IOSCHED_NOOP=y
-# CONFIG_IOSCHED_AS is not set
-# CONFIG_IOSCHED_DEADLINE is not set
-# CONFIG_IOSCHED_CFQ is not set
-# CONFIG_ATA_OVER_ETH is not set
-
-#
-# Disk device support
-#
-
-#
-# ATA/ATAPI/MFM/RLL support
-#
-# CONFIG_IDE is not set
-
-#
-# SCSI device support
-#
-# CONFIG_SCSI is not set
-
-#
-# Multi-device support (RAID and LVM)
-#
-# CONFIG_MD is not set
-
-#
-# Fusion MPT device support
-#
-# CONFIG_FUSION is not set
-
-#
-# IEEE 1394 (FireWire) support
-#
-# CONFIG_IEEE1394 is not set
-
-#
-# I2O device support
-#
-# CONFIG_I2O is not set
-
-#
-# Network device support
-#
-CONFIG_NETDEVICES=y
-# CONFIG_DUMMY is not set
-# CONFIG_BONDING is not set
-# CONFIG_EQUALIZER is not set
-# CONFIG_TUN is not set
-
-#
-# ARCnet devices
-#
-# CONFIG_ARCNET is not set
-
-#
-# Ethernet (10 or 100Mbit)
-#
-CONFIG_NET_ETHERNET=y
-CONFIG_MII=y
-# CONFIG_HAPPYMEAL is not set
-# CONFIG_SUNGEM is not set
-# CONFIG_NET_VENDOR_3COM is not set
-# CONFIG_NET_VENDOR_SMC is not set
-
-#
-# Tulip family network device support
-#
-# CONFIG_NET_TULIP is not set
-# CONFIG_HP100 is not set
-# CONFIG_NE2000 is not set
-CONFIG_NET_PCI=y
-# CONFIG_PCNET32 is not set
-# CONFIG_AMD8111_ETH is not set
-# CONFIG_ADAPTEC_STARFIRE is not set
-# CONFIG_DGRS is not set
-CONFIG_EEPRO100=y
-# CONFIG_E100 is not set
-# CONFIG_FEALNX is not set
-# CONFIG_NATSEMI is not set
-# CONFIG_NE2K_PCI is not set
-# CONFIG_8139TOO is not set
-# CONFIG_SIS900 is not set
-# CONFIG_EPIC100 is not set
-# CONFIG_SUNDANCE is not set
-# CONFIG_TLAN is not set
-# CONFIG_VIA_RHINE is not set
-
-#
-# Ethernet (1000 Mbit)
-#
-# CONFIG_ACENIC is not set
-# CONFIG_DL2K is not set
-# CONFIG_E1000 is not set
-# CONFIG_NS83820 is not set
-# CONFIG_HAMACHI is not set
-# CONFIG_R8169 is not set
-# CONFIG_SK98LIN is not set
-# CONFIG_VIA_VELOCITY is not set
-# CONFIG_TIGON3 is not set
-# CONFIG_BNX2 is not set
-
-#
-# Ethernet (10000 Mbit)
-#
-# CONFIG_IXGB is not set
-# CONFIG_S2IO is not set
-
-#
-# Token Ring devices
-#
-# CONFIG_TR is not set
-
-#
-# Wireless LAN (non-hamradio)
-#
-# CONFIG_NET_RADIO is not set
-
-#
-# Wan interfaces
-#
-# CONFIG_WAN is not set
-# CONFIG_FDDI is not set
-# CONFIG_PPP is not set
-# CONFIG_SLIP is not set
-# CONFIG_NETPOLL is not set
-# CONFIG_NET_POLL_CONTROLLER is not set
-
-#
-# ISDN subsystem
-#
-# CONFIG_ISDN is not set
-
-#
-# Input device support
-#
-CONFIG_INPUT=y
-
-#
-# Userland interfaces
-#
-# CONFIG_INPUT_MOUSEDEV is not set
-# CONFIG_INPUT_JOYDEV is not set
-# CONFIG_INPUT_TSDEV is not set
-# CONFIG_INPUT_EVDEV is not set
-# CONFIG_INPUT_EVBUG is not set
-
-#
-# Input Device Drivers
-#
-# CONFIG_INPUT_KEYBOARD is not set
-# CONFIG_INPUT_MOUSE is not set
-# CONFIG_INPUT_JOYSTICK is not set
-# CONFIG_INPUT_TOUCHSCREEN is not set
-# CONFIG_INPUT_MISC is not set
-
-#
-# Hardware I/O ports
-#
-# CONFIG_SERIO is not set
-# CONFIG_GAMEPORT is not set
-
-#
-# Character devices
-#
-# CONFIG_VT is not set
-# CONFIG_SERIAL_NONSTANDARD is not set
-
-#
-# Serial drivers
-#
-# CONFIG_SERIAL_8250 is not set
-
-#
-# Non-8250 serial port support
-#
-CONFIG_V850E_UART=y
-CONFIG_V850E_UART_CONSOLE=y
-CONFIG_SERIAL_CORE=y
-CONFIG_SERIAL_CORE_CONSOLE=y
-# CONFIG_SERIAL_JSM is not set
-# CONFIG_UNIX98_PTYS is not set
-# CONFIG_LEGACY_PTYS is not set
-
-#
-# IPMI
-#
-# CONFIG_IPMI_HANDLER is not set
-
-#
-# Watchdog Cards
-#
-# CONFIG_WATCHDOG is not set
-# CONFIG_RTC is not set
-# CONFIG_GEN_RTC is not set
-# CONFIG_DTLK is not set
-# CONFIG_R3964 is not set
-# CONFIG_APPLICOM is not set
-
-#
-# Ftape, the floppy tape device driver
-#
-# CONFIG_DRM is not set
-# CONFIG_RAW_DRIVER is not set
-
-#
-# TPM devices
-#
-
-#
-# Multimedia devices
-#
-# CONFIG_VIDEO_DEV is not set
-
-#
-# Digital Video Broadcasting Devices
-#
-# CONFIG_DVB is not set
-
-#
-# File systems
-#
-# CONFIG_EXT2_FS is not set
-# CONFIG_EXT3_FS is not set
-# CONFIG_JBD is not set
-# CONFIG_REISERFS_FS is not set
-# CONFIG_JFS_FS is not set
-# CONFIG_FS_POSIX_ACL is not set
-
-#
-# XFS support
-#
-# CONFIG_XFS_FS is not set
-# CONFIG_MINIX_FS is not set
-CONFIG_ROMFS_FS=y
-# CONFIG_MAGIC_ROM_PTR is not set
-CONFIG_INOTIFY=y
-# CONFIG_QUOTA is not set
-CONFIG_DNOTIFY=y
-# CONFIG_AUTOFS_FS is not set
-# CONFIG_AUTOFS4_FS is not set
-
-#
-# CD-ROM/DVD Filesystems
-#
-# CONFIG_ISO9660_FS is not set
-# CONFIG_UDF_FS is not set
-
-#
-# DOS/FAT/NT Filesystems
-#
-# CONFIG_MSDOS_FS is not set
-# CONFIG_VFAT_FS is not set
-# CONFIG_NTFS_FS is not set
-
-#
-# Pseudo filesystems
-#
-CONFIG_PROC_FS=y
-CONFIG_SYSFS=y
-# CONFIG_TMPFS is not set
-# CONFIG_HUGETLB_PAGE is not set
-CONFIG_RAMFS=y
-
-#
-# Miscellaneous filesystems
-#
-# CONFIG_HFSPLUS_FS is not set
-# CONFIG_JFFS_FS is not set
-# CONFIG_JFFS2_FS is not set
-# CONFIG_CRAMFS is not set
-# CONFIG_VXFS_FS is not set
-# CONFIG_HPFS_FS is not set
-# CONFIG_QNX4FS_FS is not set
-# CONFIG_SYSV_FS is not set
-# CONFIG_UFS_FS is not set
-
-#
-# Network File Systems
-#
-CONFIG_NFS_FS=y
-CONFIG_NFS_V3=y
-# CONFIG_NFS_V3_ACL is not set
-# CONFIG_NFSD is not set
-CONFIG_LOCKD=y
-CONFIG_LOCKD_V4=y
-CONFIG_NFS_COMMON=y
-CONFIG_SUNRPC=y
-# CONFIG_SMB_FS is not set
-# CONFIG_CIFS is not set
-# CONFIG_NCP_FS is not set
-# CONFIG_CODA_FS is not set
-
-#
-# Partition Types
-#
-# CONFIG_PARTITION_ADVANCED is not set
-CONFIG_MSDOS_PARTITION=y
-
-#
-# Native Language Support
-#
-# CONFIG_NLS is not set
-
-#
-# Graphics support
-#
-# CONFIG_FB is not set
-
-#
-# Sound
-#
-# CONFIG_SOUND is not set
-
-#
-# USB support
-#
-CONFIG_USB_ARCH_HAS_HCD=y
-CONFIG_USB_ARCH_HAS_OHCI=y
-# CONFIG_USB is not set
-
-#
-# USB Gadget Support
-#
-# CONFIG_USB_GADGET is not set
-
-#
-# Kernel hacking
-#
-# CONFIG_PRINTK_TIME is not set
-CONFIG_DEBUG_KERNEL=y
-# CONFIG_MAGIC_SYSRQ is not set
-CONFIG_LOG_BUF_SHIFT=14
-# CONFIG_SCHEDSTATS is not set
-# CONFIG_DEBUG_SLAB is not set
-# CONFIG_DEBUG_SPINLOCK is not set
-# CONFIG_DEBUG_SPINLOCK_SLEEP is not set
-# CONFIG_DEBUG_KOBJECT is not set
-CONFIG_DEBUG_INFO=y
-# CONFIG_DEBUG_FS is not set
-# CONFIG_NO_KERNEL_MSG is not set
-
-#
-# Security options
-#
-# CONFIG_KEYS is not set
-# CONFIG_SECURITY is not set
-
-#
-# Cryptographic options
-#
-# CONFIG_CRYPTO is not set
-
-#
-# Hardware crypto devices
-#
-
-#
-# Library routines
-#
-# CONFIG_CRC_CCITT is not set
-# CONFIG_CRC32 is not set
-# CONFIG_LIBCRC32C is not set
diff --git a/arch/v850/configs/rte-me2-cb_defconfig b/arch/v850/configs/rte-me2-cb_defconfig
deleted file mode 100644
index 15e666478061..000000000000
--- a/arch/v850/configs/rte-me2-cb_defconfig
+++ /dev/null
@@ -1,462 +0,0 @@
-#
-# Automatically generated make config: don't edit
-# Linux kernel version: 2.6.13-uc0
-# Fri Sep  2 13:47:50 2005
-#
-# CONFIG_MMU is not set
-# CONFIG_UID16 is not set
-CONFIG_RWSEM_GENERIC_SPINLOCK=y
-# CONFIG_RWSEM_XCHGADD_ALGORITHM is not set
-CONFIG_GENERIC_CALIBRATE_DELAY=y
-# CONFIG_ISA is not set
-# CONFIG_ISAPNP is not set
-# CONFIG_EISA is not set
-# CONFIG_MCA is not set
-CONFIG_V850=y
-
-#
-# Processor type and features
-#
-# CONFIG_V850E_SIM is not set
-# CONFIG_RTE_CB_MA1 is not set
-# CONFIG_RTE_CB_NB85E is not set
-CONFIG_RTE_CB_ME2=y
-# CONFIG_V850E_AS85EP1 is not set
-# CONFIG_V850E2_SIM85E2C is not set
-# CONFIG_V850E2_SIM85E2S is not set
-# CONFIG_V850E2_FPGA85E2C is not set
-# CONFIG_V850E2_ANNA is not set
-CONFIG_V850E=y
-CONFIG_V850E_ME2=y
-CONFIG_RTE_CB=y
-# CONFIG_RTE_MB_A_PCI is not set
-# CONFIG_PCI is not set
-CONFIG_V850E_INTC=y
-CONFIG_V850E_TIMER_D=y
-CONFIG_V850E_CACHE=y
-# CONFIG_V850E2_CACHE is not set
-# CONFIG_NO_CACHE is not set
-# CONFIG_ROM_KERNEL is not set
-CONFIG_ZERO_BSS=y
-# CONFIG_V850E_HIGHRES_TIMER is not set
-# CONFIG_RESET_GUARD is not set
-CONFIG_LARGE_ALLOCS=y
-CONFIG_FLATMEM=y
-CONFIG_FLAT_NODE_MEM_MAP=y
-
-#
-# Code maturity level options
-#
-# CONFIG_EXPERIMENTAL is not set
-CONFIG_CLEAN_COMPILE=y
-CONFIG_BROKEN_ON_SMP=y
-CONFIG_INIT_ENV_ARG_LIMIT=32
-
-#
-# General setup
-#
-CONFIG_LOCALVERSION=""
-# CONFIG_BSD_PROCESS_ACCT is not set
-# CONFIG_SYSCTL is not set
-# CONFIG_HOTPLUG is not set
-# CONFIG_IKCONFIG is not set
-CONFIG_EMBEDDED=y
-# CONFIG_KALLSYMS is not set
-CONFIG_PRINTK=y
-CONFIG_BUG=y
-# CONFIG_BASE_FULL is not set
-# CONFIG_FUTEX is not set
-# CONFIG_EPOLL is not set
-CONFIG_CC_OPTIMIZE_FOR_SIZE=y
-CONFIG_CC_ALIGN_FUNCTIONS=0
-CONFIG_CC_ALIGN_LABELS=0
-CONFIG_CC_ALIGN_LOOPS=0
-CONFIG_CC_ALIGN_JUMPS=0
-CONFIG_BASE_SMALL=1
-
-#
-# Loadable module support
-#
-CONFIG_MODULES=y
-CONFIG_MODULE_UNLOAD=y
-CONFIG_OBSOLETE_MODPARM=y
-# CONFIG_MODULE_SRCVERSION_ALL is not set
-CONFIG_KMOD=y
-
-#
-# Bus options (PCI, PCMCIA, EISA, MCA, ISA)
-#
-
-#
-# PCCARD (PCMCIA/CardBus) support
-#
-# CONFIG_PCCARD is not set
-
-#
-# PCI Hotplug Support
-#
-
-#
-# Executable file formats
-#
-CONFIG_BINFMT_FLAT=y
-# CONFIG_BINFMT_ZFLAT is not set
-# CONFIG_BINFMT_SHARED_FLAT is not set
-# CONFIG_BINFMT_MISC is not set
-
-#
-# Networking
-#
-# CONFIG_NET is not set
-
-#
-# Generic Driver Options
-#
-CONFIG_STANDALONE=y
-CONFIG_PREVENT_FIRMWARE_BUILD=y
-# CONFIG_FW_LOADER is not set
-# CONFIG_DEBUG_DRIVER is not set
-
-#
-# Memory Technology Devices (MTD)
-#
-CONFIG_MTD=y
-# CONFIG_MTD_DEBUG is not set
-# CONFIG_MTD_CONCAT is not set
-# CONFIG_MTD_PARTITIONS is not set
-
-#
-# User Modules And Translation Layers
-#
-# CONFIG_MTD_CHAR is not set
-CONFIG_MTD_BLOCK=y
-# CONFIG_FTL is not set
-# CONFIG_NFTL is not set
-# CONFIG_INFTL is not set
-
-#
-# RAM/ROM/Flash chip drivers
-#
-# CONFIG_MTD_CFI is not set
-# CONFIG_MTD_JEDECPROBE is not set
-CONFIG_MTD_MAP_BANK_WIDTH_1=y
-CONFIG_MTD_MAP_BANK_WIDTH_2=y
-CONFIG_MTD_MAP_BANK_WIDTH_4=y
-# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set
-# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set
-# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set
-CONFIG_MTD_CFI_I1=y
-CONFIG_MTD_CFI_I2=y
-# CONFIG_MTD_CFI_I4 is not set
-# CONFIG_MTD_CFI_I8 is not set
-# CONFIG_MTD_RAM is not set
-# CONFIG_MTD_ROM is not set
-# CONFIG_MTD_ABSENT is not set
-
-#
-# Mapping drivers for chip access
-#
-# CONFIG_MTD_COMPLEX_MAPPINGS is not set
-# CONFIG_MTD_PLATRAM is not set
-
-#
-# Self-contained MTD device drivers
-#
-CONFIG_MTD_SLRAM=y
-# CONFIG_MTD_PHRAM is not set
-# CONFIG_MTD_MTDRAM is not set
-# CONFIG_MTD_BLKMTD is not set
-
-#
-# Disk-On-Chip Device Drivers
-#
-# CONFIG_MTD_DOC2000 is not set
-# CONFIG_MTD_DOC2001 is not set
-# CONFIG_MTD_DOC2001PLUS is not set
-
-#
-# NAND Flash Device Drivers
-#
-# CONFIG_MTD_NAND is not set
-
-#
-# Parallel port support
-#
-# CONFIG_PARPORT is not set
-
-#
-# Block devices
-#
-# CONFIG_BLK_DEV_FD is not set
-# CONFIG_BLK_DEV_COW_COMMON is not set
-# CONFIG_BLK_DEV_LOOP is not set
-# CONFIG_BLK_DEV_RAM is not set
-CONFIG_BLK_DEV_RAM_COUNT=16
-CONFIG_INITRAMFS_SOURCE=""
-# CONFIG_CDROM_PKTCDVD is not set
-
-#
-# IO Schedulers
-#
-CONFIG_IOSCHED_NOOP=y
-# CONFIG_IOSCHED_AS is not set
-# CONFIG_IOSCHED_DEADLINE is not set
-# CONFIG_IOSCHED_CFQ is not set
-
-#
-# Disk device support
-#
-
-#
-# ATA/ATAPI/MFM/RLL support
-#
-# CONFIG_IDE is not set
-
-#
-# SCSI device support
-#
-# CONFIG_SCSI is not set
-
-#
-# Multi-device support (RAID and LVM)
-#
-# CONFIG_MD is not set
-
-#
-# Fusion MPT device support
-#
-# CONFIG_FUSION is not set
-
-#
-# IEEE 1394 (FireWire) support
-#
-
-#
-# I2O device support
-#
-
-#
-# Network device support
-#
-# CONFIG_NETPOLL is not set
-# CONFIG_NET_POLL_CONTROLLER is not set
-
-#
-# ISDN subsystem
-#
-
-#
-# Input device support
-#
-CONFIG_INPUT=y
-
-#
-# Userland interfaces
-#
-# CONFIG_INPUT_MOUSEDEV is not set
-# CONFIG_INPUT_JOYDEV is not set
-# CONFIG_INPUT_TSDEV is not set
-# CONFIG_INPUT_EVDEV is not set
-# CONFIG_INPUT_EVBUG is not set
-
-#
-# Input Device Drivers
-#
-# CONFIG_INPUT_KEYBOARD is not set
-# CONFIG_INPUT_MOUSE is not set
-# CONFIG_INPUT_JOYSTICK is not set
-# CONFIG_INPUT_TOUCHSCREEN is not set
-# CONFIG_INPUT_MISC is not set
-
-#
-# Hardware I/O ports
-#
-CONFIG_SERIO=y
-# CONFIG_SERIO_I8042 is not set
-# CONFIG_SERIO_SERPORT is not set
-# CONFIG_SERIO_LIBPS2 is not set
-# CONFIG_SERIO_RAW is not set
-# CONFIG_GAMEPORT is not set
-
-#
-# Character devices
-#
-# CONFIG_VT is not set
-# CONFIG_SERIAL_NONSTANDARD is not set
-
-#
-# Serial drivers
-#
-CONFIG_SERIAL_8250=y
-CONFIG_SERIAL_8250_CONSOLE=y
-CONFIG_SERIAL_8250_NR_UARTS=1
-# CONFIG_SERIAL_8250_EXTENDED is not set
-
-#
-# Non-8250 serial port support
-#
-# CONFIG_V850E_UART is not set
-CONFIG_SERIAL_CORE=y
-CONFIG_SERIAL_CORE_CONSOLE=y
-# CONFIG_UNIX98_PTYS is not set
-# CONFIG_LEGACY_PTYS is not set
-
-#
-# IPMI
-#
-# CONFIG_IPMI_HANDLER is not set
-
-#
-# Watchdog Cards
-#
-# CONFIG_WATCHDOG is not set
-# CONFIG_RTC is not set
-# CONFIG_GEN_RTC is not set
-# CONFIG_DTLK is not set
-# CONFIG_R3964 is not set
-
-#
-# Ftape, the floppy tape device driver
-#
-# CONFIG_RAW_DRIVER is not set
-
-#
-# TPM devices
-#
-
-#
-# Multimedia devices
-#
-# CONFIG_VIDEO_DEV is not set
-
-#
-# Digital Video Broadcasting Devices
-#
-
-#
-# File systems
-#
-# CONFIG_EXT2_FS is not set
-# CONFIG_EXT3_FS is not set
-# CONFIG_JBD is not set
-# CONFIG_REISERFS_FS is not set
-# CONFIG_JFS_FS is not set
-# CONFIG_FS_POSIX_ACL is not set
-
-#
-# XFS support
-#
-# CONFIG_XFS_FS is not set
-# CONFIG_MINIX_FS is not set
-CONFIG_ROMFS_FS=y
-# CONFIG_MAGIC_ROM_PTR is not set
-CONFIG_INOTIFY=y
-# CONFIG_QUOTA is not set
-CONFIG_DNOTIFY=y
-# CONFIG_AUTOFS_FS is not set
-# CONFIG_AUTOFS4_FS is not set
-
-#
-# CD-ROM/DVD Filesystems
-#
-# CONFIG_ISO9660_FS is not set
-# CONFIG_UDF_FS is not set
-
-#
-# DOS/FAT/NT Filesystems
-#
-# CONFIG_MSDOS_FS is not set
-# CONFIG_VFAT_FS is not set
-# CONFIG_NTFS_FS is not set
-
-#
-# Pseudo filesystems
-#
-CONFIG_PROC_FS=y
-CONFIG_SYSFS=y
-# CONFIG_TMPFS is not set
-# CONFIG_HUGETLB_PAGE is not set
-CONFIG_RAMFS=y
-
-#
-# Miscellaneous filesystems
-#
-# CONFIG_HFSPLUS_FS is not set
-# CONFIG_JFFS_FS is not set
-# CONFIG_JFFS2_FS is not set
-# CONFIG_CRAMFS is not set
-# CONFIG_VXFS_FS is not set
-# CONFIG_HPFS_FS is not set
-# CONFIG_QNX4FS_FS is not set
-# CONFIG_SYSV_FS is not set
-# CONFIG_UFS_FS is not set
-
-#
-# Partition Types
-#
-# CONFIG_PARTITION_ADVANCED is not set
-CONFIG_MSDOS_PARTITION=y
-
-#
-# Native Language Support
-#
-# CONFIG_NLS is not set
-
-#
-# Graphics support
-#
-# CONFIG_FB is not set
-
-#
-# Sound
-#
-# CONFIG_SOUND is not set
-
-#
-# USB support
-#
-# CONFIG_USB_ARCH_HAS_HCD is not set
-# CONFIG_USB_ARCH_HAS_OHCI is not set
-
-#
-# USB Gadget Support
-#
-# CONFIG_USB_GADGET is not set
-
-#
-# Kernel hacking
-#
-# CONFIG_PRINTK_TIME is not set
-CONFIG_DEBUG_KERNEL=y
-# CONFIG_MAGIC_SYSRQ is not set
-CONFIG_LOG_BUF_SHIFT=14
-# CONFIG_SCHEDSTATS is not set
-# CONFIG_DEBUG_SLAB is not set
-# CONFIG_DEBUG_SPINLOCK is not set
-# CONFIG_DEBUG_SPINLOCK_SLEEP is not set
-# CONFIG_DEBUG_KOBJECT is not set
-CONFIG_DEBUG_INFO=y
-# CONFIG_DEBUG_FS is not set
-# CONFIG_NO_KERNEL_MSG is not set
-
-#
-# Security options
-#
-# CONFIG_KEYS is not set
-# CONFIG_SECURITY is not set
-
-#
-# Cryptographic options
-#
-# CONFIG_CRYPTO is not set
-
-#
-# Hardware crypto devices
-#
-
-#
-# Library routines
-#
-# CONFIG_CRC_CCITT is not set
-# CONFIG_CRC32 is not set
-# CONFIG_LIBCRC32C is not set
diff --git a/arch/v850/configs/sim_defconfig b/arch/v850/configs/sim_defconfig
deleted file mode 100644
index f31ba7398ad0..000000000000
--- a/arch/v850/configs/sim_defconfig
+++ /dev/null
@@ -1,451 +0,0 @@
-#
-# Automatically generated make config: don't edit
-# Linux kernel version: 2.6.13-uc0
-# Fri Sep  2 13:36:43 2005
-#
-# CONFIG_MMU is not set
-# CONFIG_UID16 is not set
-CONFIG_RWSEM_GENERIC_SPINLOCK=y
-# CONFIG_RWSEM_XCHGADD_ALGORITHM is not set
-CONFIG_GENERIC_CALIBRATE_DELAY=y
-# CONFIG_ISA is not set
-# CONFIG_ISAPNP is not set
-# CONFIG_EISA is not set
-# CONFIG_MCA is not set
-CONFIG_V850=y
-
-#
-# Processor type and features
-#
-CONFIG_V850E_SIM=y
-# CONFIG_RTE_CB_MA1 is not set
-# CONFIG_RTE_CB_NB85E is not set
-# CONFIG_RTE_CB_ME2 is not set
-# CONFIG_V850E_AS85EP1 is not set
-# CONFIG_V850E2_SIM85E2C is not set
-# CONFIG_V850E2_SIM85E2S is not set
-# CONFIG_V850E2_FPGA85E2C is not set
-# CONFIG_V850E2_ANNA is not set
-CONFIG_V850E=y
-# CONFIG_PCI is not set
-# CONFIG_V850E_INTC is not set
-# CONFIG_V850E_TIMER_D is not set
-# CONFIG_V850E_CACHE is not set
-# CONFIG_V850E2_CACHE is not set
-CONFIG_NO_CACHE=y
-CONFIG_ZERO_BSS=y
-# CONFIG_RESET_GUARD is not set
-CONFIG_LARGE_ALLOCS=y
-CONFIG_FLATMEM=y
-CONFIG_FLAT_NODE_MEM_MAP=y
-
-#
-# Code maturity level options
-#
-# CONFIG_EXPERIMENTAL is not set
-CONFIG_CLEAN_COMPILE=y
-CONFIG_BROKEN_ON_SMP=y
-CONFIG_INIT_ENV_ARG_LIMIT=32
-
-#
-# General setup
-#
-CONFIG_LOCALVERSION=""
-# CONFIG_BSD_PROCESS_ACCT is not set
-# CONFIG_SYSCTL is not set
-# CONFIG_HOTPLUG is not set
-# CONFIG_IKCONFIG is not set
-CONFIG_EMBEDDED=y
-# CONFIG_KALLSYMS is not set
-CONFIG_PRINTK=y
-CONFIG_BUG=y
-# CONFIG_BASE_FULL is not set
-# CONFIG_FUTEX is not set
-# CONFIG_EPOLL is not set
-CONFIG_CC_OPTIMIZE_FOR_SIZE=y
-CONFIG_CC_ALIGN_FUNCTIONS=0
-CONFIG_CC_ALIGN_LABELS=0
-CONFIG_CC_ALIGN_LOOPS=0
-CONFIG_CC_ALIGN_JUMPS=0
-CONFIG_BASE_SMALL=1
-
-#
-# Loadable module support
-#
-CONFIG_MODULES=y
-CONFIG_MODULE_UNLOAD=y
-CONFIG_OBSOLETE_MODPARM=y
-# CONFIG_MODULE_SRCVERSION_ALL is not set
-CONFIG_KMOD=y
-
-#
-# Bus options (PCI, PCMCIA, EISA, MCA, ISA)
-#
-
-#
-# PCCARD (PCMCIA/CardBus) support
-#
-# CONFIG_PCCARD is not set
-
-#
-# PCI Hotplug Support
-#
-
-#
-# Executable file formats
-#
-CONFIG_BINFMT_FLAT=y
-# CONFIG_BINFMT_ZFLAT is not set
-# CONFIG_BINFMT_SHARED_FLAT is not set
-# CONFIG_BINFMT_MISC is not set
-
-#
-# Networking
-#
-# CONFIG_NET is not set
-
-#
-# Generic Driver Options
-#
-CONFIG_STANDALONE=y
-CONFIG_PREVENT_FIRMWARE_BUILD=y
-# CONFIG_FW_LOADER is not set
-# CONFIG_DEBUG_DRIVER is not set
-
-#
-# Memory Technology Devices (MTD)
-#
-CONFIG_MTD=y
-# CONFIG_MTD_DEBUG is not set
-# CONFIG_MTD_CONCAT is not set
-# CONFIG_MTD_PARTITIONS is not set
-
-#
-# User Modules And Translation Layers
-#
-# CONFIG_MTD_CHAR is not set
-CONFIG_MTD_BLOCK=y
-# CONFIG_FTL is not set
-# CONFIG_NFTL is not set
-# CONFIG_INFTL is not set
-
-#
-# RAM/ROM/Flash chip drivers
-#
-# CONFIG_MTD_CFI is not set
-# CONFIG_MTD_JEDECPROBE is not set
-CONFIG_MTD_MAP_BANK_WIDTH_1=y
-CONFIG_MTD_MAP_BANK_WIDTH_2=y
-CONFIG_MTD_MAP_BANK_WIDTH_4=y
-# CONFIG_MTD_MAP_BANK_WIDTH_8 is not set
-# CONFIG_MTD_MAP_BANK_WIDTH_16 is not set
-# CONFIG_MTD_MAP_BANK_WIDTH_32 is not set
-CONFIG_MTD_CFI_I1=y
-CONFIG_MTD_CFI_I2=y
-# CONFIG_MTD_CFI_I4 is not set
-# CONFIG_MTD_CFI_I8 is not set
-# CONFIG_MTD_RAM is not set
-# CONFIG_MTD_ROM is not set
-# CONFIG_MTD_ABSENT is not set
-
-#
-# Mapping drivers for chip access
-#
-# CONFIG_MTD_COMPLEX_MAPPINGS is not set
-# CONFIG_MTD_PLATRAM is not set
-
-#
-# Self-contained MTD device drivers
-#
-CONFIG_MTD_SLRAM=y
-# CONFIG_MTD_PHRAM is not set
-# CONFIG_MTD_MTDRAM is not set
-# CONFIG_MTD_BLKMTD is not set
-
-#
-# Disk-On-Chip Device Drivers
-#
-# CONFIG_MTD_DOC2000 is not set
-# CONFIG_MTD_DOC2001 is not set
-# CONFIG_MTD_DOC2001PLUS is not set
-
-#
-# NAND Flash Device Drivers
-#
-# CONFIG_MTD_NAND is not set
-
-#
-# Parallel port support
-#
-# CONFIG_PARPORT is not set
-
-#
-# Block devices
-#
-# CONFIG_BLK_DEV_FD is not set
-# CONFIG_BLK_DEV_COW_COMMON is not set
-# CONFIG_BLK_DEV_LOOP is not set
-# CONFIG_BLK_DEV_RAM is not set
-CONFIG_BLK_DEV_RAM_COUNT=16
-CONFIG_INITRAMFS_SOURCE=""
-# CONFIG_CDROM_PKTCDVD is not set
-
-#
-# IO Schedulers
-#
-CONFIG_IOSCHED_NOOP=y
-# CONFIG_IOSCHED_AS is not set
-# CONFIG_IOSCHED_DEADLINE is not set
-# CONFIG_IOSCHED_CFQ is not set
-
-#
-# Disk device support
-#
-
-#
-# ATA/ATAPI/MFM/RLL support
-#
-# CONFIG_IDE is not set
-
-#
-# SCSI device support
-#
-# CONFIG_SCSI is not set
-
-#
-# Multi-device support (RAID and LVM)
-#
-# CONFIG_MD is not set
-
-#
-# Fusion MPT device support
-#
-# CONFIG_FUSION is not set
-
-#
-# IEEE 1394 (FireWire) support
-#
-
-#
-# I2O device support
-#
-
-#
-# Network device support
-#
-# CONFIG_NETPOLL is not set
-# CONFIG_NET_POLL_CONTROLLER is not set
-
-#
-# ISDN subsystem
-#
-
-#
-# Input device support
-#
-CONFIG_INPUT=y
-
-#
-# Userland interfaces
-#
-# CONFIG_INPUT_MOUSEDEV is not set
-# CONFIG_INPUT_JOYDEV is not set
-# CONFIG_INPUT_TSDEV is not set
-# CONFIG_INPUT_EVDEV is not set
-# CONFIG_INPUT_EVBUG is not set
-
-#
-# Input Device Drivers
-#
-# CONFIG_INPUT_KEYBOARD is not set
-# CONFIG_INPUT_MOUSE is not set
-# CONFIG_INPUT_JOYSTICK is not set
-# CONFIG_INPUT_TOUCHSCREEN is not set
-# CONFIG_INPUT_MISC is not set
-
-#
-# Hardware I/O ports
-#
-CONFIG_SERIO=y
-# CONFIG_SERIO_I8042 is not set
-# CONFIG_SERIO_SERPORT is not set
-# CONFIG_SERIO_LIBPS2 is not set
-# CONFIG_SERIO_RAW is not set
-# CONFIG_GAMEPORT is not set
-
-#
-# Character devices
-#
-# CONFIG_VT is not set
-# CONFIG_SERIAL_NONSTANDARD is not set
-
-#
-# Serial drivers
-#
-# CONFIG_SERIAL_8250 is not set
-
-#
-# Non-8250 serial port support
-#
-# CONFIG_UNIX98_PTYS is not set
-# CONFIG_LEGACY_PTYS is not set
-
-#
-# IPMI
-#
-# CONFIG_IPMI_HANDLER is not set
-
-#
-# Watchdog Cards
-#
-# CONFIG_WATCHDOG is not set
-# CONFIG_RTC is not set
-# CONFIG_GEN_RTC is not set
-# CONFIG_DTLK is not set
-# CONFIG_R3964 is not set
-
-#
-# Ftape, the floppy tape device driver
-#
-# CONFIG_RAW_DRIVER is not set
-
-#
-# TPM devices
-#
-
-#
-# Multimedia devices
-#
-# CONFIG_VIDEO_DEV is not set
-
-#
-# Digital Video Broadcasting Devices
-#
-
-#
-# File systems
-#
-# CONFIG_EXT2_FS is not set
-# CONFIG_EXT3_FS is not set
-# CONFIG_JBD is not set
-# CONFIG_REISERFS_FS is not set
-# CONFIG_JFS_FS is not set
-# CONFIG_FS_POSIX_ACL is not set
-
-#
-# XFS support
-#
-# CONFIG_XFS_FS is not set
-# CONFIG_MINIX_FS is not set
-CONFIG_ROMFS_FS=y
-# CONFIG_MAGIC_ROM_PTR is not set
-CONFIG_INOTIFY=y
-# CONFIG_QUOTA is not set
-CONFIG_DNOTIFY=y
-# CONFIG_AUTOFS_FS is not set
-# CONFIG_AUTOFS4_FS is not set
-
-#
-# CD-ROM/DVD Filesystems
-#
-# CONFIG_ISO9660_FS is not set
-# CONFIG_UDF_FS is not set
-
-#
-# DOS/FAT/NT Filesystems
-#
-# CONFIG_MSDOS_FS is not set
-# CONFIG_VFAT_FS is not set
-# CONFIG_NTFS_FS is not set
-
-#
-# Pseudo filesystems
-#
-CONFIG_PROC_FS=y
-CONFIG_SYSFS=y
-# CONFIG_TMPFS is not set
-# CONFIG_HUGETLB_PAGE is not set
-CONFIG_RAMFS=y
-
-#
-# Miscellaneous filesystems
-#
-# CONFIG_HFSPLUS_FS is not set
-# CONFIG_JFFS_FS is not set
-# CONFIG_JFFS2_FS is not set
-# CONFIG_CRAMFS is not set
-# CONFIG_VXFS_FS is not set
-# CONFIG_HPFS_FS is not set
-# CONFIG_QNX4FS_FS is not set
-# CONFIG_SYSV_FS is not set
-# CONFIG_UFS_FS is not set
-
-#
-# Partition Types
-#
-# CONFIG_PARTITION_ADVANCED is not set
-CONFIG_MSDOS_PARTITION=y
-
-#
-# Native Language Support
-#
-# CONFIG_NLS is not set
-
-#
-# Graphics support
-#
-# CONFIG_FB is not set
-
-#
-# Sound
-#
-# CONFIG_SOUND is not set
-
-#
-# USB support
-#
-# CONFIG_USB_ARCH_HAS_HCD is not set
-# CONFIG_USB_ARCH_HAS_OHCI is not set
-
-#
-# USB Gadget Support
-#
-# CONFIG_USB_GADGET is not set
-
-#
-# Kernel hacking
-#
-# CONFIG_PRINTK_TIME is not set
-CONFIG_DEBUG_KERNEL=y
-# CONFIG_MAGIC_SYSRQ is not set
-CONFIG_LOG_BUF_SHIFT=14
-# CONFIG_SCHEDSTATS is not set
-# CONFIG_DEBUG_SLAB is not set
-# CONFIG_DEBUG_SPINLOCK is not set
-# CONFIG_DEBUG_SPINLOCK_SLEEP is not set
-# CONFIG_DEBUG_KOBJECT is not set
-CONFIG_DEBUG_INFO=y
-# CONFIG_DEBUG_FS is not set
-# CONFIG_NO_KERNEL_MSG is not set
-
-#
-# Security options
-#
-# CONFIG_KEYS is not set
-# CONFIG_SECURITY is not set
-
-#
-# Cryptographic options
-#
-# CONFIG_CRYPTO is not set
-
-#
-# Hardware crypto devices
-#
-
-#
-# Library routines
-#
-# CONFIG_CRC_CCITT is not set
-# CONFIG_CRC32 is not set
-# CONFIG_LIBCRC32C is not set
diff --git a/arch/v850/kernel/Makefile b/arch/v850/kernel/Makefile
deleted file mode 100644
index da5889c53576..000000000000
--- a/arch/v850/kernel/Makefile
+++ /dev/null
@@ -1,40 +0,0 @@
-#
-# arch/v850/kernel/Makefile
-#
-#  Copyright (C) 2001,02,03  NEC Electronics Corporation
-#  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
-#
-# This file is subject to the terms and conditions of the GNU General Public
-# License.  See the file "COPYING" in the main directory of this archive
-# for more details.
-#
-
-extra-y := head.o init_task.o vmlinux.lds
-
-obj-y += intv.o entry.o process.o syscalls.o time.o setup.o \
-	 signal.o irq.o mach.o ptrace.o bug.o
-obj-$(CONFIG_MODULES)		+= module.o v850_ksyms.o
-# chip-specific code
-obj-$(CONFIG_V850E_MA1)		+= ma.o
-obj-$(CONFIG_V850E_ME2)		+= me2.o
-obj-$(CONFIG_V850E_TEG)		+= teg.o
-obj-$(CONFIG_V850E_AS85EP1)	+= as85ep1.o
-obj-$(CONFIG_V850E2_ANNA)	+= anna.o
-# platform-specific code
-obj-$(CONFIG_V850E_SIM)		+= sim.o simcons.o
-obj-$(CONFIG_V850E2_SIM85E2)	+= sim85e2.o memcons.o
-obj-$(CONFIG_V850E2_FPGA85E2C)	+= fpga85e2c.o memcons.o
-obj-$(CONFIG_RTE_CB)		+= rte_cb.o rte_cb_leds.o
-obj-$(CONFIG_RTE_CB_MA1)	+= rte_ma1_cb.o
-obj-$(CONFIG_RTE_CB_ME2)	+= rte_me2_cb.o
-obj-$(CONFIG_RTE_CB_NB85E)	+= rte_nb85e_cb.o
-obj-$(CONFIG_RTE_CB_MULTI)	+= rte_cb_multi.o
-obj-$(CONFIG_RTE_MB_A_PCI)	+= rte_mb_a_pci.o
-obj-$(CONFIG_RTE_GBUS_INT)	+= gbus_int.o
-# feature-specific code
-obj-$(CONFIG_V850E_INTC)	+= v850e_intc.o
-obj-$(CONFIG_V850E_TIMER_D)	+= v850e_timer_d.o v850e_utils.o
-obj-$(CONFIG_V850E_CACHE)	+= v850e_cache.o
-obj-$(CONFIG_V850E2_CACHE)	+= v850e2_cache.o
-obj-$(CONFIG_V850E_HIGHRES_TIMER) += highres_timer.o
-obj-$(CONFIG_PROC_FS)		+= procfs.o
diff --git a/arch/v850/kernel/anna-rom.ld b/arch/v850/kernel/anna-rom.ld
deleted file mode 100644
index 7c54e7e3f1b1..000000000000
--- a/arch/v850/kernel/anna-rom.ld
+++ /dev/null
@@ -1,16 +0,0 @@
-/* Linker script for the Midas labs Anna V850E2 evaluation board
-   (CONFIG_V850E2_ANNA), with kernel in ROM (CONFIG_ROM_KERNEL).  */
-
-MEMORY {
-	/* 8MB of flash ROM.  */
-	ROM   : ORIGIN = 0,          LENGTH = 0x00800000
-
-	/* 1MB of static RAM.  This memory is mirrored 64 times.  */
-	SRAM  : ORIGIN = SRAM_ADDR,  LENGTH = SRAM_SIZE
-	/* 64MB of DRAM.  */
-	SDRAM : ORIGIN = SDRAM_ADDR, LENGTH = SDRAM_SIZE
-}
-
-SECTIONS {
-	ROMK_SECTIONS(ROM, SRAM)
-}
diff --git a/arch/v850/kernel/anna.c b/arch/v850/kernel/anna.c
deleted file mode 100644
index 5978a25170fb..000000000000
--- a/arch/v850/kernel/anna.c
+++ /dev/null
@@ -1,202 +0,0 @@
-/*
- * arch/v850/kernel/anna.c -- Anna V850E2 evaluation chip/board
- *
- *  Copyright (C) 2002,03  NEC Electronics Corporation
- *  Copyright (C) 2002,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/init.h>
-#include <linux/bootmem.h>
-#include <linux/major.h>
-#include <linux/irq.h>
-
-#include <asm/machdep.h>
-#include <asm/atomic.h>
-#include <asm/page.h>
-#include <asm/v850e_timer_d.h>
-#include <asm/v850e_uart.h>
-
-#include "mach.h"
-
-
-/* SRAM and SDRAM are vaguely contiguous (with a big hole in between; see
-   mach_reserve_bootmem for details); use both as one big area.  */
-#define RAM_START 	SRAM_ADDR
-#define RAM_END		(SDRAM_ADDR + SDRAM_SIZE)
-
-/* The bits of this port are connected to an 8-LED bar-graph.  */
-#define LEDS_PORT	0
-
-
-static void anna_led_tick (void);
-
-
-void __init mach_early_init (void)
-{
-	ANNA_ILBEN    = 0;
-
-	V850E2_CSC(0) = 0x402F;
-	V850E2_CSC(1) = 0x4000;
-	V850E2_BPC    = 0;
-	V850E2_BSC    = 0xAAAA;
-	V850E2_BEC    = 0;
-
-#if 0
-	V850E2_BHC    = 0xFFFF;	/* icache all memory, dcache all */
-#else
-	V850E2_BHC    = 0;	/* cache no memory */
-#endif
-	V850E2_BCT(0) = 0xB088;
-	V850E2_BCT(1) = 0x0008;
-	V850E2_DWC(0) = 0x0027;
-	V850E2_DWC(1) = 0;
-	V850E2_BCC    = 0x0006;
-	V850E2_ASC    = 0;
-	V850E2_LBS    = 0x0089;
-	V850E2_SCR(3) = 0x21A9;
-	V850E2_RFS(3) = 0x8121;
-
-	v850e_intc_disable_irqs ();
-}
-
-void __init mach_setup (char **cmdline)
-{
-	ANNA_PORT_PM (LEDS_PORT) = 0;	/* Make all LED pins output pins.  */
-	mach_tick = anna_led_tick;
-}
-
-void __init mach_get_physical_ram (unsigned long *ram_start,
-				   unsigned long *ram_len)
-{
-	*ram_start = RAM_START;
-	*ram_len = RAM_END - RAM_START;
-}
-
-void __init mach_reserve_bootmem ()
-{
-	/* The space between SRAM and SDRAM is filled with duplicate
-	   images of SRAM.  Prevent the kernel from using them.  */
-	reserve_bootmem (SRAM_ADDR + SRAM_SIZE,
-			 SDRAM_ADDR - (SRAM_ADDR + SRAM_SIZE),
-			 BOOTMEM_DEFAULT);
-}
-
-void mach_gettimeofday (struct timespec *tv)
-{
-	tv->tv_sec = 0;
-	tv->tv_nsec = 0;
-}
-
-void __init mach_sched_init (struct irqaction *timer_action)
-{
-	/* Start hardware timer.  */
-	v850e_timer_d_configure (0, HZ);
-	/* Install timer interrupt handler.  */
-	setup_irq (IRQ_INTCMD(0), timer_action);
-}
-
-static struct v850e_intc_irq_init irq_inits[] = {
-	{ "IRQ", 0, 		NUM_MACH_IRQS,	1, 7 },
-	{ "PIN", IRQ_INTP(0),   IRQ_INTP_NUM,   1, 4 },
-	{ "CCC", IRQ_INTCCC(0),	IRQ_INTCCC_NUM, 1, 5 },
-	{ "CMD", IRQ_INTCMD(0), IRQ_INTCMD_NUM,	1, 5 },
-	{ "DMA", IRQ_INTDMA(0), IRQ_INTDMA_NUM,	1, 2 },
-	{ "DMXER", IRQ_INTDMXER,1,		1, 2 },
-	{ "SRE", IRQ_INTSRE(0), IRQ_INTSRE_NUM,	3, 3 },
-	{ "SR",	 IRQ_INTSR(0),	IRQ_INTSR_NUM, 	3, 4 },
-	{ "ST",  IRQ_INTST(0), 	IRQ_INTST_NUM, 	3, 5 },
-	{ 0 }
-};
-#define NUM_IRQ_INITS (ARRAY_SIZE(irq_inits) - 1)
-
-static struct hw_interrupt_type hw_itypes[NUM_IRQ_INITS];
-
-void __init mach_init_irqs (void)
-{
-	v850e_intc_init_irq_types (irq_inits, hw_itypes);
-}
-
-void machine_restart (char *__unused)
-{
-#ifdef CONFIG_RESET_GUARD
-	disable_reset_guard ();
-#endif
-	asm ("jmp r0"); /* Jump to the reset vector.  */
-}
-
-void machine_halt (void)
-{
-#ifdef CONFIG_RESET_GUARD
-	disable_reset_guard ();
-#endif
-	local_irq_disable ();	/* Ignore all interrupts.  */
-	ANNA_PORT_IO(LEDS_PORT) = 0xAA;	/* Note that we halted.  */
-	for (;;)
-		asm ("halt; nop; nop; nop; nop; nop");
-}
-
-void machine_power_off (void)
-{
-	machine_halt ();
-}
-
-/* Called before configuring an on-chip UART.  */
-void anna_uart_pre_configure (unsigned chan, unsigned cflags, unsigned baud)
-{
-	/* The Anna connects some general-purpose I/O pins on the CPU to
-	   the RTS/CTS lines of UART 1's serial connection.  I/O pins P07
-	   and P37 are RTS and CTS respectively.  */
-	if (chan == 1) {
-		ANNA_PORT_PM(0) &= ~0x80; /* P07 in output mode */
-		ANNA_PORT_PM(3) |=  0x80; /* P37 in input mode */
-	}
-}
-
-/* Minimum and maximum bounds for the moving upper LED boundary in the
-   clock tick display.  We can't use the last bit because it's used for
-   UART0's CTS output.  */
-#define MIN_MAX_POS 0
-#define MAX_MAX_POS 6
-
-/* There are MAX_MAX_POS^2 - MIN_MAX_POS^2 cycles in the animation, so if
-   we pick 6 and 0 as above, we get 49 cycles, which is when divided into
-   the standard 100 value for HZ, gives us an almost 1s total time.  */
-#define TICKS_PER_FRAME \
-	(HZ / (MAX_MAX_POS * MAX_MAX_POS - MIN_MAX_POS * MIN_MAX_POS))
-
-static void anna_led_tick ()
-{
-	static unsigned counter = 0;
-	
-	if (++counter == TICKS_PER_FRAME) {
-		static int pos = 0, max_pos = MAX_MAX_POS, dir = 1;
-
-		if (dir > 0 && pos == max_pos) {
-			dir = -1;
-			if (max_pos == MIN_MAX_POS)
-				max_pos = MAX_MAX_POS;
-			else
-				max_pos--;
-		} else {
-			if (dir < 0 && pos == 0)
-				dir = 1;
-
-			if (pos + dir <= max_pos) {
-				/* Each bit of port 0 has a LED. */
-				clear_bit (pos, &ANNA_PORT_IO(LEDS_PORT));
-				pos += dir;
-				set_bit (pos, &ANNA_PORT_IO(LEDS_PORT));
-			}
-		}
-
-		counter = 0;
-	}
-}
diff --git a/arch/v850/kernel/anna.ld b/arch/v850/kernel/anna.ld
deleted file mode 100644
index df7f80f2833d..000000000000
--- a/arch/v850/kernel/anna.ld
+++ /dev/null
@@ -1,20 +0,0 @@
-/* Linker script for the Midas labs Anna V850E2 evaluation board
-   (CONFIG_V850E2_ANNA).  */
-
-MEMORY {
-	/* 256KB of internal memory (followed by one mirror).  */
-	iMEM0 : ORIGIN = 0,	     LENGTH = 0x00040000
-	/* 256KB of internal memory (followed by one mirror).  */
-	iMEM1 : ORIGIN = 0x00040000, LENGTH = 0x00040000
-
-	/* 1MB of static RAM.  This memory is mirrored 64 times.  */
-	SRAM  : ORIGIN = SRAM_ADDR,  LENGTH = SRAM_SIZE
-	/* 64MB of DRAM.  */
-	SDRAM : ORIGIN = SDRAM_ADDR, LENGTH = SDRAM_SIZE
-}
-
-SECTIONS {
-	.intv : { INTV_CONTENTS } > iMEM0
-	.sram : { RAMK_KRAM_CONTENTS } > SRAM
-	.root : { ROOT_FS_CONTENTS } > SDRAM
-}
diff --git a/arch/v850/kernel/as85ep1-rom.ld b/arch/v850/kernel/as85ep1-rom.ld
deleted file mode 100644
index fe2a9a3ab525..000000000000
--- a/arch/v850/kernel/as85ep1-rom.ld
+++ /dev/null
@@ -1,21 +0,0 @@
-/* Linker script for the NEC AS85EP1 V850E evaluation board
-   (CONFIG_V850E_AS85EP1), with kernel in ROM (CONFIG_ROM_KERNEL).  */
-
-MEMORY {
-	/* 4MB of flash ROM.  */
-	ROM   : ORIGIN = 0,          LENGTH = 0x00400000
-
-	/* 1MB of static RAM.  */
-	SRAM  : ORIGIN = SRAM_ADDR,  LENGTH = SRAM_SIZE
-
-	/* About 58MB of DRAM.  This can actually be at one of two
-	   positions, determined by jumper JP3; we have to use the first
-	   position because the second is partially out of processor
-	   instruction addressing range (though in the second position
-	   there's actually 64MB available).  */
-	SDRAM : ORIGIN = SDRAM_ADDR, LENGTH = SDRAM_SIZE
-}
-
-SECTIONS {
-	ROMK_SECTIONS(ROM, SRAM)
-}
diff --git a/arch/v850/kernel/as85ep1.c b/arch/v850/kernel/as85ep1.c
deleted file mode 100644
index b525ecf3aea4..000000000000
--- a/arch/v850/kernel/as85ep1.c
+++ /dev/null
@@ -1,234 +0,0 @@
-/*
- * arch/v850/kernel/as85ep1.c -- AS85EP1 V850E evaluation chip/board
- *
- *  Copyright (C) 2002,03  NEC Electronics Corporation
- *  Copyright (C) 2002,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/init.h>
-#include <linux/bootmem.h>
-#include <linux/major.h>
-#include <linux/irq.h>
-
-#include <asm/machdep.h>
-#include <asm/atomic.h>
-#include <asm/page.h>
-#include <asm/v850e_timer_d.h>
-#include <asm/v850e_uart.h>
-
-#include "mach.h"
-
-
-/* SRAM and SDRAM are vaguely contiguous (with a big hole in between; see
-   mach_reserve_bootmem for details); use both as one big area.  */
-#define RAM_START 	SRAM_ADDR
-#define RAM_END		(SDRAM_ADDR + SDRAM_SIZE)
-
-/* The bits of this port are connected to an 8-LED bar-graph.  */
-#define LEDS_PORT	4
-
-
-static void as85ep1_led_tick (void);
-
-extern char _intv_copy_src_start, _intv_copy_src_end;
-extern char _intv_copy_dst_start;
-
-
-void __init mach_early_init (void)
-{
-#ifndef CONFIG_ROM_KERNEL
-	const u32 *src;
-	register u32 *dst asm ("ep");
-#endif
-
-	AS85EP1_CSC(0) = 0x0403;
-	AS85EP1_BCT(0) = 0xB8B8;
-	AS85EP1_DWC(0) = 0x0104;
-	AS85EP1_BCC    = 0x0012;
-	AS85EP1_ASC    = 0;
-	AS85EP1_LBS    = 0x00A9;
-
-	AS85EP1_PORT_PMC(6)  = 0xFF; /* valid A0,A1,A20-A25 */
-	AS85EP1_PORT_PMC(7)  = 0x0E; /* valid CS1-CS3       */
-	AS85EP1_PORT_PMC(9)  = 0xFF; /* valid D16-D23       */
-	AS85EP1_PORT_PMC(10) = 0xFF; /* valid D24-D31       */
-
-	AS85EP1_RFS(1) = 0x800c;
-	AS85EP1_RFS(3) = 0x800c;
-	AS85EP1_SCR(1) = 0x20A9;
-	AS85EP1_SCR(3) = 0x20A9;
-
-#ifndef CONFIG_ROM_KERNEL
-	/* The early chip we have is buggy, and writing the interrupt
-	   vectors into low RAM may screw up, so for non-ROM kernels, we
-	   only rely on the reset vector being downloaded, and copy the
-	   rest of the interrupt vectors into place here.  The specific bug
-	   is that writing address N, where (N & 0x10) == 0x10, will _also_
-	   write to address (N - 0x10).  We avoid this (effectively) by
-	   writing in 16-byte chunks backwards from the end.  */
-
-	AS85EP1_IRAMM = 0x3;	/* "write-mode" for the internal instruction memory */
-
-	src = (u32 *)(((u32)&_intv_copy_src_end - 1) & ~0xF);
-	dst = (u32 *)&_intv_copy_dst_start
-		+ (src - (u32 *)&_intv_copy_src_start);
-	do {
-		u32 t0 = src[0], t1 = src[1], t2 = src[2], t3 = src[3];
-		dst[0] = t0; dst[1] = t1; dst[2] = t2; dst[3] = t3;
-		dst -= 4;
-		src -= 4;
-	} while (src > (u32 *)&_intv_copy_src_start);
-
-	AS85EP1_IRAMM = 0x0;	/* "read-mode" for the internal instruction memory */
-#endif /* !CONFIG_ROM_KERNEL */
-
-	v850e_intc_disable_irqs ();
-}
-
-void __init mach_setup (char **cmdline)
-{
-	AS85EP1_PORT_PMC (LEDS_PORT) = 0; /* Make the LEDs port an I/O port. */
-	AS85EP1_PORT_PM (LEDS_PORT) = 0; /* Make all the bits output pins.  */
-	mach_tick = as85ep1_led_tick;
-}
-
-void __init mach_get_physical_ram (unsigned long *ram_start,
-				   unsigned long *ram_len)
-{
-	*ram_start = RAM_START;
-	*ram_len = RAM_END - RAM_START;
-}
-
-/* Convenience macros.  */
-#define SRAM_END	(SRAM_ADDR + SRAM_SIZE)
-#define SDRAM_END	(SDRAM_ADDR + SDRAM_SIZE)
-
-void __init mach_reserve_bootmem ()
-{
-	if (SDRAM_ADDR < RAM_END && SDRAM_ADDR > RAM_START)
-		/* We can't use the space between SRAM and SDRAM, so
-		   prevent the kernel from trying.  */
-		reserve_bootmem(SRAM_END, SDRAM_ADDR - SRAM_END,
-				BOOTMEM_DEFAULT);
-}
-
-void mach_gettimeofday (struct timespec *tv)
-{
-	tv->tv_sec = 0;
-	tv->tv_nsec = 0;
-}
-
-void __init mach_sched_init (struct irqaction *timer_action)
-{
-	/* Start hardware timer.  */
-	v850e_timer_d_configure (0, HZ);
-	/* Install timer interrupt handler.  */
-	setup_irq (IRQ_INTCMD(0), timer_action);
-}
-
-static struct v850e_intc_irq_init irq_inits[] = {
-	{ "IRQ", 0, 		NUM_MACH_IRQS,	1, 7 },
-	{ "CCC", IRQ_INTCCC(0),	IRQ_INTCCC_NUM, 1, 5 },
-	{ "CMD", IRQ_INTCMD(0), IRQ_INTCMD_NUM,	1, 5 },
-	{ "SRE", IRQ_INTSRE(0), IRQ_INTSRE_NUM,	3, 3 },
-	{ "SR",	 IRQ_INTSR(0),	IRQ_INTSR_NUM, 	3, 4 },
-	{ "ST",  IRQ_INTST(0), 	IRQ_INTST_NUM, 	3, 5 },
-	{ 0 }
-};
-#define NUM_IRQ_INITS (ARRAY_SIZE(irq_inits) - 1)
-
-static struct hw_interrupt_type hw_itypes[NUM_IRQ_INITS];
-
-void __init mach_init_irqs (void)
-{
-	v850e_intc_init_irq_types (irq_inits, hw_itypes);
-}
-
-void machine_restart (char *__unused)
-{
-#ifdef CONFIG_RESET_GUARD
-	disable_reset_guard ();
-#endif
-	asm ("jmp r0"); /* Jump to the reset vector.  */
-}
-
-void machine_halt (void)
-{
-#ifdef CONFIG_RESET_GUARD
-	disable_reset_guard ();
-#endif
-	local_irq_disable ();	/* Ignore all interrupts.  */
-	AS85EP1_PORT_IO (LEDS_PORT) = 0xAA;	/* Note that we halted.  */
-	for (;;)
-		asm ("halt; nop; nop; nop; nop; nop");
-}
-
-void machine_power_off (void)
-{
-	machine_halt ();
-}
-
-/* Called before configuring an on-chip UART.  */
-void as85ep1_uart_pre_configure (unsigned chan, unsigned cflags, unsigned baud)
-{
-	/* Make the shared uart/port pins be uart pins.  */
-	AS85EP1_PORT_PMC(3) |= (0x5 << chan);
-
-	/* The AS85EP1 connects some general-purpose I/O pins on the CPU to
-	   the RTS/CTS lines of UART 1's serial connection.  I/O pins P53
-	   and P54 are RTS and CTS respectively.  */
-	if (chan == 1) {
-		/* Put P53 & P54 in I/O port mode.  */
-		AS85EP1_PORT_PMC(5) &= ~0x18;
-		/* Make P53 an output, and P54 an input.  */
-		AS85EP1_PORT_PM(5) |=  0x10;
-	}
-}
-
-/* Minimum and maximum bounds for the moving upper LED boundary in the
-   clock tick display.  */
-#define MIN_MAX_POS 0
-#define MAX_MAX_POS 7
-
-/* There are MAX_MAX_POS^2 - MIN_MAX_POS^2 cycles in the animation, so if
-   we pick 6 and 0 as above, we get 49 cycles, which is when divided into
-   the standard 100 value for HZ, gives us an almost 1s total time.  */
-#define TICKS_PER_FRAME \
-	(HZ / (MAX_MAX_POS * MAX_MAX_POS - MIN_MAX_POS * MIN_MAX_POS))
-
-static void as85ep1_led_tick ()
-{
-	static unsigned counter = 0;
-	
-	if (++counter == TICKS_PER_FRAME) {
-		static int pos = 0, max_pos = MAX_MAX_POS, dir = 1;
-
-		if (dir > 0 && pos == max_pos) {
-			dir = -1;
-			if (max_pos == MIN_MAX_POS)
-				max_pos = MAX_MAX_POS;
-			else
-				max_pos--;
-		} else {
-			if (dir < 0 && pos == 0)
-				dir = 1;
-
-			if (pos + dir <= max_pos) {
-				/* Each bit of port 0 has a LED. */
-				set_bit (pos, &AS85EP1_PORT_IO(LEDS_PORT));
-				pos += dir;
-				clear_bit (pos, &AS85EP1_PORT_IO(LEDS_PORT));
-			}
-		}
-
-		counter = 0;
-	}
-}
diff --git a/arch/v850/kernel/as85ep1.ld b/arch/v850/kernel/as85ep1.ld
deleted file mode 100644
index ef2c4399063e..000000000000
--- a/arch/v850/kernel/as85ep1.ld
+++ /dev/null
@@ -1,49 +0,0 @@
-/* Linker script for the NEC AS85EP1 V850E evaluation board
-   (CONFIG_V850E_AS85EP1).  */
-
-MEMORY {
-	/* 1MB of internal instruction memory. */
-	iMEM0 : ORIGIN = 0,	     LENGTH = 0x00100000
-
-	/* 1MB of static RAM.  */
-	SRAM  : ORIGIN = SRAM_ADDR,  LENGTH = SRAM_SIZE
-
-	/* About 58MB of DRAM.  This can actually be at one of two
-	   positions, determined by jump JP3; we have to use the first
-	   position because the second is partially out of processor
-	   instruction addressing range (though in the second position
-	   there's actually 64MB available).  */
-	SDRAM : ORIGIN = SDRAM_ADDR, LENGTH = SDRAM_SIZE
-}
-
-SECTIONS {
-	.resetv : {
-		__intv_start = . ;
-			*(.intv.reset)	/* Reset vector */
-	} > iMEM0
-
-	.sram : {
-		RAMK_KRAM_CONTENTS
-
-		/* We stick most of the interrupt vectors here; they'll be
-		   copied into the proper location by the early init code (we
-		   can't put them directly in the right place because of
-		   hardware bugs).  The vectors shouldn't need to be
-		   relocated, so we don't have to use `> ...  AT> ...' to
-		   split the load/vm addresses (and we can't because of
-		   problems with the loader).  */
-		. = ALIGN (0x10) ;
-		__intv_copy_src_start = . ;
-			*(.intv.common)	/* Vectors common to all v850e proc. */
-			*(.intv.mach)	/* Machine-specific int. vectors.  */
-		. = ALIGN (0x10) ;
-		__intv_copy_src_end = . ;
-	} > SRAM
-
-	/* Where we end up putting the vectors.  */
-	__intv_copy_dst_start = 0x10 ;
-	__intv_copy_dst_end = __intv_copy_dst_start + (__intv_copy_src_end - __intv_copy_src_start) ;
-	__intv_end = __intv_copy_dst_end ;
-
-	.root : { ROOT_FS_CONTENTS } > SDRAM
-}
diff --git a/arch/v850/kernel/asm-offsets.c b/arch/v850/kernel/asm-offsets.c
deleted file mode 100644
index 581e6986a776..000000000000
--- a/arch/v850/kernel/asm-offsets.c
+++ /dev/null
@@ -1,58 +0,0 @@
-/*
- * This program is used to generate definitions needed by
- * assembly language modules.
- *
- * We use the technique used in the OSF Mach kernel code:
- * generate asm statements containing #defines,
- * compile this file to assembler, and then extract the
- * #defines from the assembly-language output.
- */
-
-#include <linux/stddef.h>
-#include <linux/sched.h>
-#include <linux/kernel_stat.h>
-#include <linux/ptrace.h>
-#include <linux/hardirq.h>
-#include <linux/kbuild.h>
-
-#include <asm/irq.h>
-#include <asm/errno.h>
-
-int main (void)
-{
-	/* offsets into the task struct */
-	DEFINE (TASK_STATE, offsetof (struct task_struct, state));
-	DEFINE (TASK_FLAGS, offsetof (struct task_struct, flags));
-	DEFINE (TASK_PTRACE, offsetof (struct task_struct, ptrace));
-	DEFINE (TASK_BLOCKED, offsetof (struct task_struct, blocked));
-	DEFINE (TASK_THREAD, offsetof (struct task_struct, thread));
-	DEFINE (TASK_THREAD_INFO, offsetof (struct task_struct, stack));
-	DEFINE (TASK_MM, offsetof (struct task_struct, mm));
-	DEFINE (TASK_ACTIVE_MM, offsetof (struct task_struct, active_mm));
-	DEFINE (TASK_PID, offsetof (struct task_struct, pid));
-
-	/* offsets into the kernel_stat struct */
-	DEFINE (STAT_IRQ, offsetof (struct kernel_stat, irqs));
-
-
-	/* signal defines */
-	DEFINE (SIGSEGV, SIGSEGV);
-	DEFINE (SEGV_MAPERR, SEGV_MAPERR);
-	DEFINE (SIGTRAP, SIGTRAP);
-	DEFINE (SIGCHLD, SIGCHLD);
-	DEFINE (SIGILL, SIGILL);
-	DEFINE (TRAP_TRACE, TRAP_TRACE);
-
-	/* ptrace flag bits */
-	DEFINE (PT_PTRACED, PT_PTRACED);
-	DEFINE (PT_DTRACE, PT_DTRACE);
-
-	/* error values */
-	DEFINE (ENOSYS, ENOSYS);
-
-	/* clone flag bits */
-	DEFINE (CLONE_VFORK, CLONE_VFORK);
-	DEFINE (CLONE_VM, CLONE_VM);
-
-	return 0;
-}
diff --git a/arch/v850/kernel/bug.c b/arch/v850/kernel/bug.c
deleted file mode 100644
index c78cf750915a..000000000000
--- a/arch/v850/kernel/bug.c
+++ /dev/null
@@ -1,142 +0,0 @@
-/*
- * arch/v850/kernel/bug.c -- Bug reporting functions
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/kernel.h>
-#include <linux/reboot.h>
-#include <linux/sched.h>
-#include <linux/module.h>
-
-#include <asm/errno.h>
-#include <asm/ptrace.h>
-#include <asm/processor.h>
-#include <asm/current.h>
-
-/* We should use __builtin_return_address, but it doesn't work in gcc-2.90
-   (which is currently our standard compiler on the v850).  */
-#define ret_addr() ({ register u32 lp asm ("lp"); lp; })
-#define stack_addr() ({ register u32 sp asm ("sp"); sp; })
-
-void __bug ()
-{
-	printk (KERN_CRIT "kernel BUG at PC 0x%x (SP ~0x%x)!\n",
-		ret_addr() - 4, /* - 4 for `jarl' */
-		stack_addr());
-	machine_halt ();
-}
-
-int bad_trap (int trap_num, struct pt_regs *regs)
-{
-	printk (KERN_CRIT
-		"unimplemented trap %d called at 0x%08lx, pid %d!\n",
-		trap_num, regs->pc, current->pid);
-	return -ENOSYS;
-}
-
-#ifdef CONFIG_RESET_GUARD
-void unexpected_reset (unsigned long ret_addr, unsigned long kmode,
-		       struct task_struct *task, unsigned long sp)
-{
-	printk (KERN_CRIT
-		"unexpected reset in %s mode, pid %d"
-		" (ret_addr = 0x%lx, sp = 0x%lx)\n",
-		kmode ? "kernel" : "user",
-		task ? task->pid : -1,
-		ret_addr, sp);
-
-	machine_halt ();
-}
-#endif /* CONFIG_RESET_GUARD */
-
-
-
-struct spec_reg_name {
-	const char *name;
-	int gpr;
-};
-
-struct spec_reg_name spec_reg_names[] = {
-	{ "sp", GPR_SP },
-	{ "gp", GPR_GP },
-	{ "tp", GPR_TP },
-	{ "ep", GPR_EP },
-	{ "lp", GPR_LP },
-	{ 0, 0 }
-};
-
-void show_regs (struct pt_regs *regs)
-{
-	int gpr_base, gpr_offs;
-
-	printk ("     pc 0x%08lx    psw 0x%08lx                       kernel_mode %d\n",
-		regs->pc, regs->psw, regs->kernel_mode);
-	printk ("   ctpc 0x%08lx  ctpsw 0x%08lx   ctbp 0x%08lx\n",
-		regs->ctpc, regs->ctpsw, regs->ctbp);
-
-	for (gpr_base = 0; gpr_base < NUM_GPRS; gpr_base += 4) {
-		for (gpr_offs = 0; gpr_offs < 4; gpr_offs++) {
-			int gpr = gpr_base + gpr_offs;
-			long val = regs->gpr[gpr];
-			struct spec_reg_name *srn;
-
-			for (srn = spec_reg_names; srn->name; srn++)
-				if (srn->gpr == gpr)
-					break;
-
-			if (srn->name)
-				printk ("%7s 0x%08lx", srn->name, val);
-			else
-				printk ("    r%02d 0x%08lx", gpr, val);
-		}
-
-		printk ("\n");
-	}
-}
-
-/*
- * TASK is a pointer to the task whose backtrace we want to see (or NULL
- * for current task), SP is the stack pointer of the first frame that
- * should be shown in the back trace (or NULL if the entire call-chain of
- * the task should be shown).
- */
-void show_stack (struct task_struct *task, unsigned long *sp)
-{
-	unsigned long addr, end;
-
-	if (sp)
-		addr = (unsigned long)sp;
-	else if (task)
-		addr = task_sp (task);
-	else
-		addr = stack_addr ();
-
-	addr = addr & ~3;
-	end = (addr + THREAD_SIZE - 1) & THREAD_MASK;
-
-	while (addr < end) {
-		printk ("%8lX: ", addr);
-		while (addr < end) {
-			printk (" %8lX", *(unsigned long *)addr);
-			addr += sizeof (unsigned long);
-			if (! (addr & 0xF))
-				break;
-		}
-		printk ("\n");
-	}
-}
-
-void dump_stack ()
-{
-	show_stack (0, 0);
-}
-
-EXPORT_SYMBOL(dump_stack);
diff --git a/arch/v850/kernel/entry.S b/arch/v850/kernel/entry.S
deleted file mode 100644
index e4327a8d6bcd..000000000000
--- a/arch/v850/kernel/entry.S
+++ /dev/null
@@ -1,1121 +0,0 @@
-/*
- * arch/v850/kernel/entry.S -- Low-level system-call handling, trap handlers,
- *	and context-switching
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/sys.h>
-
-#include <asm/entry.h>
-#include <asm/current.h>
-#include <asm/thread_info.h>
-#include <asm/clinkage.h>
-#include <asm/processor.h>
-#include <asm/irq.h>
-#include <asm/errno.h>
-
-#include <asm/asm-offsets.h>
-
-
-/* Make a slightly more convenient alias for C_SYMBOL_NAME.  */
-#define CSYM	C_SYMBOL_NAME
-
-
-/* The offset of the struct pt_regs in a state-save-frame on the stack.  */
-#define PTO	STATE_SAVE_PT_OFFSET
-
-
-/* Save argument registers to the state-save-frame pointed to by EP.  */
-#define SAVE_ARG_REGS							      \
-	sst.w	r6, PTO+PT_GPR(6)[ep];					      \
-	sst.w	r7, PTO+PT_GPR(7)[ep];					      \
-	sst.w	r8, PTO+PT_GPR(8)[ep];					      \
-	sst.w	r9, PTO+PT_GPR(9)[ep]
-/* Restore argument registers from the state-save-frame pointed to by EP.  */
-#define RESTORE_ARG_REGS						      \
-	sld.w	PTO+PT_GPR(6)[ep], r6;					      \
-	sld.w	PTO+PT_GPR(7)[ep], r7;					      \
-	sld.w	PTO+PT_GPR(8)[ep], r8;					      \
-	sld.w	PTO+PT_GPR(9)[ep], r9
-
-/* Save value return registers to the state-save-frame pointed to by EP.  */
-#define SAVE_RVAL_REGS							      \
-	sst.w	r10, PTO+PT_GPR(10)[ep];				      \
-	sst.w	r11, PTO+PT_GPR(11)[ep]
-/* Restore value return registers from the state-save-frame pointed to by EP.  */
-#define RESTORE_RVAL_REGS						      \
-	sld.w	PTO+PT_GPR(10)[ep], r10;				      \
-	sld.w	PTO+PT_GPR(11)[ep], r11
-
-
-#define SAVE_CALL_CLOBBERED_REGS_BEFORE_ARGS				      \
-	sst.w	r1, PTO+PT_GPR(1)[ep];					      \
-	sst.w	r5, PTO+PT_GPR(5)[ep]
-#define SAVE_CALL_CLOBBERED_REGS_AFTER_RVAL				      \
-	sst.w	r12, PTO+PT_GPR(12)[ep];				      \
-	sst.w	r13, PTO+PT_GPR(13)[ep];				      \
-	sst.w	r14, PTO+PT_GPR(14)[ep];				      \
-	sst.w	r15, PTO+PT_GPR(15)[ep];				      \
-	sst.w	r16, PTO+PT_GPR(16)[ep];				      \
-	sst.w	r17, PTO+PT_GPR(17)[ep];				      \
-	sst.w	r18, PTO+PT_GPR(18)[ep];				      \
-	sst.w	r19, PTO+PT_GPR(19)[ep]
-#define RESTORE_CALL_CLOBBERED_REGS_BEFORE_ARGS				      \
-	sld.w	PTO+PT_GPR(1)[ep], r1;					      \
-	sld.w	PTO+PT_GPR(5)[ep], r5
-#define RESTORE_CALL_CLOBBERED_REGS_AFTER_RVAL				      \
-	sld.w	PTO+PT_GPR(12)[ep], r12;				      \
-	sld.w	PTO+PT_GPR(13)[ep], r13;				      \
-	sld.w	PTO+PT_GPR(14)[ep], r14;				      \
-	sld.w	PTO+PT_GPR(15)[ep], r15;				      \
-	sld.w	PTO+PT_GPR(16)[ep], r16;				      \
-	sld.w	PTO+PT_GPR(17)[ep], r17;				      \
-	sld.w	PTO+PT_GPR(18)[ep], r18;				      \
-	sld.w	PTO+PT_GPR(19)[ep], r19
-
-/* Save `call clobbered' registers to the state-save-frame pointed to by EP.  */
-#define SAVE_CALL_CLOBBERED_REGS					      \
-	SAVE_CALL_CLOBBERED_REGS_BEFORE_ARGS;				      \
-	SAVE_ARG_REGS;							      \
-	SAVE_RVAL_REGS;							      \
-	SAVE_CALL_CLOBBERED_REGS_AFTER_RVAL
-/* Restore `call clobbered' registers from the state-save-frame pointed to
-   by EP.  */
-#define RESTORE_CALL_CLOBBERED_REGS					      \
-	RESTORE_CALL_CLOBBERED_REGS_BEFORE_ARGS;			      \
-	RESTORE_ARG_REGS;						      \
-	RESTORE_RVAL_REGS;						      \
-	RESTORE_CALL_CLOBBERED_REGS_AFTER_RVAL
-
-/* Save `call clobbered' registers except for the return-value registers
-   to the state-save-frame pointed to by EP.  */
-#define SAVE_CALL_CLOBBERED_REGS_NO_RVAL				      \
-	SAVE_CALL_CLOBBERED_REGS_BEFORE_ARGS;				      \
-	SAVE_ARG_REGS;							      \
-	SAVE_CALL_CLOBBERED_REGS_AFTER_RVAL
-/* Restore `call clobbered' registers except for the return-value registers
-   from the state-save-frame pointed to by EP.  */
-#define RESTORE_CALL_CLOBBERED_REGS_NO_RVAL				      \
-	RESTORE_CALL_CLOBBERED_REGS_BEFORE_ARGS;			      \
-	RESTORE_ARG_REGS;						      \
-	RESTORE_CALL_CLOBBERED_REGS_AFTER_RVAL
-
-/* Save `call saved' registers to the state-save-frame pointed to by EP.  */
-#define SAVE_CALL_SAVED_REGS						      \
-	sst.w	r2, PTO+PT_GPR(2)[ep];					      \
-	sst.w	r20, PTO+PT_GPR(20)[ep];				      \
-	sst.w	r21, PTO+PT_GPR(21)[ep];				      \
-	sst.w	r22, PTO+PT_GPR(22)[ep];				      \
-	sst.w	r23, PTO+PT_GPR(23)[ep];				      \
-	sst.w	r24, PTO+PT_GPR(24)[ep];				      \
-	sst.w	r25, PTO+PT_GPR(25)[ep];				      \
-	sst.w	r26, PTO+PT_GPR(26)[ep];				      \
-	sst.w	r27, PTO+PT_GPR(27)[ep];				      \
-	sst.w	r28, PTO+PT_GPR(28)[ep];				      \
-	sst.w	r29, PTO+PT_GPR(29)[ep]
-/* Restore `call saved' registers from the state-save-frame pointed to by EP.  */
-#define RESTORE_CALL_SAVED_REGS						      \
-	sld.w	PTO+PT_GPR(2)[ep], r2;					      \
-	sld.w	PTO+PT_GPR(20)[ep], r20;				      \
-	sld.w	PTO+PT_GPR(21)[ep], r21;				      \
-	sld.w	PTO+PT_GPR(22)[ep], r22;				      \
-	sld.w	PTO+PT_GPR(23)[ep], r23;				      \
-	sld.w	PTO+PT_GPR(24)[ep], r24;				      \
-	sld.w	PTO+PT_GPR(25)[ep], r25;				      \
-	sld.w	PTO+PT_GPR(26)[ep], r26;				      \
-	sld.w	PTO+PT_GPR(27)[ep], r27;				      \
-	sld.w	PTO+PT_GPR(28)[ep], r28;				      \
-	sld.w	PTO+PT_GPR(29)[ep], r29
-
-
-/* Save the PC stored in the special register SAVEREG to the state-save-frame
-   pointed to by EP.  r19 is clobbered.  */
-#define SAVE_PC(savereg)						      \
-	stsr	SR_ ## savereg, r19;					      \
-	sst.w	r19, PTO+PT_PC[ep]
-/* Restore the PC from the state-save-frame pointed to by EP, to the special
-   register SAVEREG.  LP is clobbered (it is used as a scratch register
-   because the POP_STATE macro restores it, and this macro is usually used
-   inside POP_STATE).  */
-#define RESTORE_PC(savereg)						      \
-	sld.w	PTO+PT_PC[ep], lp;					      \
-	ldsr	lp, SR_ ## savereg
-/* Save the PSW register stored in the special register SAVREG to the
-   state-save-frame pointed to by EP.  r19 is clobbered.  */
-#define SAVE_PSW(savereg)						      \
-	stsr	SR_ ## savereg, r19;					      \
-	sst.w	r19, PTO+PT_PSW[ep]
-/* Restore the PSW register from the state-save-frame pointed to by EP, to
-   the special register SAVEREG.  LP is clobbered (it is used as a scratch
-   register because the POP_STATE macro restores it, and this macro is
-   usually used inside POP_STATE).  */
-#define RESTORE_PSW(savereg)						      \
-	sld.w	PTO+PT_PSW[ep], lp;					      \
-	ldsr	lp, SR_ ## savereg
-
-/* Save CTPC/CTPSW/CTBP registers to the state-save-frame pointed to by REG.
-   r19 is clobbered.  */
-#define SAVE_CT_REGS							      \
-	stsr	SR_CTPC, r19;						      \
-	sst.w	r19, PTO+PT_CTPC[ep];					      \
-	stsr	SR_CTPSW, r19;						      \
-	sst.w	r19, PTO+PT_CTPSW[ep];					      \
-	stsr	SR_CTBP, r19;						      \
-	sst.w	r19, PTO+PT_CTBP[ep]
-/* Restore CTPC/CTPSW/CTBP registers from the state-save-frame pointed to by EP.
-   LP is clobbered (it is used as a scratch register because the POP_STATE
-   macro restores it, and this macro is usually used inside POP_STATE).  */
-#define RESTORE_CT_REGS							      \
-	sld.w	PTO+PT_CTPC[ep], lp;					      \
-	ldsr	lp, SR_CTPC;						      \
-	sld.w	PTO+PT_CTPSW[ep], lp;					      \
-	ldsr	lp, SR_CTPSW;						      \
-	sld.w	PTO+PT_CTBP[ep], lp;					      \
-	ldsr	lp, SR_CTBP
-
-
-/* Push register state, except for the stack pointer, on the stack in the
-   form of a state-save-frame (plus some extra padding), in preparation for
-   a system call.  This macro makes sure that the EP, GP, and LP
-   registers are saved, and TYPE identifies the set of extra registers to
-   be saved as well.  Also copies (the new value of) SP to EP.  */
-#define PUSH_STATE(type)						      \
-	addi	-STATE_SAVE_SIZE, sp, sp; /* Make room on the stack.  */      \
-	st.w	ep, PTO+PT_GPR(GPR_EP)[sp];				      \
-	mov	sp, ep;							      \
-	sst.w	gp, PTO+PT_GPR(GPR_GP)[ep];				      \
-	sst.w	lp, PTO+PT_GPR(GPR_LP)[ep];				      \
-	type ## _STATE_SAVER
-/* Pop a register state pushed by PUSH_STATE, except for the stack pointer,
-   from the stack.  */
-#define POP_STATE(type)							      \
-	mov	sp, ep;							      \
-	type ## _STATE_RESTORER;					      \
-	sld.w	PTO+PT_GPR(GPR_GP)[ep], gp;				      \
-	sld.w	PTO+PT_GPR(GPR_LP)[ep], lp;				      \
-	sld.w	PTO+PT_GPR(GPR_EP)[ep], ep;				      \
-	addi	STATE_SAVE_SIZE, sp, sp /* Clean up our stack space.  */
-
-
-/* Switch to the kernel stack if necessary, and push register state on the
-   stack in the form of a state-save-frame.  Also load the current task
-   pointer if switching from user mode.  The stack-pointer (r3) should have
-   already been saved to the memory location SP_SAVE_LOC (the reason for
-   this is that the interrupt vectors may be beyond a 22-bit signed offset
-   jump from the actual interrupt handler, and this allows them to save the
-   stack-pointer and use that register to do an indirect jump).  This macro
-   makes sure that `special' registers, system registers, and the stack
-   pointer are saved; TYPE identifies the set of extra registers to be
-   saved as well.  SYSCALL_NUM is the register in which the system-call
-   number this state is for is stored (r0 if this isn't a system call).
-   Interrupts should already be disabled when calling this.  */
-#define SAVE_STATE(type, syscall_num, sp_save_loc)			      \
-	tst1	0, KM;			/* See if already in kernel mode.  */ \
-	bz	1f;							      \
-	ld.w	sp_save_loc, sp;	/* ... yes, use saved SP.  */	      \
-	br	2f;							      \
-1:	ld.w	KSP, sp;		/* ... no, switch to kernel stack. */ \
-2:	PUSH_STATE(type);						      \
-	ld.b	KM, r19;		/* Remember old kernel-mode.  */      \
-	sst.w	r19, PTO+PT_KERNEL_MODE[ep];				      \
-	ld.w	sp_save_loc, r19;	/* Remember old SP.  */		      \
-	sst.w	r19, PTO+PT_GPR(GPR_SP)[ep];				      \
-	mov	1, r19;			/* Now definitely in kernel-mode. */  \
-	st.b	r19, KM;						      \
-	GET_CURRENT_TASK(CURRENT_TASK);	/* Fetch the current task pointer. */ \
-	/* Save away the syscall number.  */				      \
-	sst.w	syscall_num, PTO+PT_CUR_SYSCALL[ep]
-
-
-/* Save register state not normally saved by PUSH_STATE for TYPE, to the
-   state-save-frame on the stack; also copies SP to EP.  r19 may be trashed. */
-#define SAVE_EXTRA_STATE(type)						      \
-	mov	sp, ep;							      \
-	type ## _EXTRA_STATE_SAVER
-/* Restore register state not normally restored by POP_STATE for TYPE,
-   from the state-save-frame on the stack; also copies SP to EP.
-   r19 may be trashed.  */
-#define RESTORE_EXTRA_STATE(type)					      \
-	mov	sp, ep;							      \
-	type ## _EXTRA_STATE_RESTORER
-
-/* Save any call-clobbered registers not normally saved by PUSH_STATE for
-   TYPE, to the state-save-frame on the stack.
-   EP may be trashed, but is not guaranteed to contain a copy of SP
-   (unlike after most SAVE_... macros).  r19 may be trashed.  */
-#define SAVE_EXTRA_STATE_FOR_SCHEDULE(type)				      \
-	type ## _SCHEDULE_EXTRA_STATE_SAVER
-/* Restore any call-clobbered registers not normally restored by
-   POP_STATE for TYPE, to the state-save-frame on the stack.
-   EP may be trashed, but is not guaranteed to contain a copy of SP
-   (unlike after most RESTORE_... macros).  r19 may be trashed.  */
-#define RESTORE_EXTRA_STATE_FOR_SCHEDULE(type)				      \
-	type ## _SCHEDULE_EXTRA_STATE_RESTORER
-
-
-/* These are extra_state_saver/restorer values for a user trap.  Note
-   that we save the argument registers so that restarted syscalls will
-   function properly (otherwise it wouldn't be necessary), and we must
-   _not_ restore the return-value registers (so that traps can return a
-   value!), but call-clobbered registers are not saved at all, as the
-   caller of the syscall function should have saved them.  */
-
-#define TRAP_RET reti
-/* Traps don't save call-clobbered registers (but do still save arg regs).
-   We preserve PSw to keep long-term state, namely interrupt status (for traps
-   from kernel-mode), and the single-step flag (for user traps).  */
-#define TRAP_STATE_SAVER						      \
-	SAVE_ARG_REGS;							      \
-	SAVE_PC(EIPC);							      \
-	SAVE_PSW(EIPSW)
-/* When traps return, they just leave call-clobbered registers (except for arg
-   regs) with whatever value they have from the kernel.  Traps don't preserve
-   the PSW, but we zero EIPSW to ensure it doesn't contain anything dangerous
-   (in particular, the single-step flag).  */
-#define TRAP_STATE_RESTORER						      \
-	RESTORE_ARG_REGS;						      \
-	RESTORE_PC(EIPC);						      \
-	RESTORE_PSW(EIPSW)
-/* Save registers not normally saved by traps.  We need to save r12, even
-   though it's nominally call-clobbered, because it's used when restarting
-   a system call (the signal-handling path uses SAVE_EXTRA_STATE, and
-   expects r12 to be restored when the trap returns).  */
-#define TRAP_EXTRA_STATE_SAVER						      \
-	SAVE_RVAL_REGS;							      \
-	sst.w	r12, PTO+PT_GPR(12)[ep];				      \
-	SAVE_CALL_SAVED_REGS;						      \
-	SAVE_CT_REGS
-#define TRAP_EXTRA_STATE_RESTORER					      \
-	RESTORE_RVAL_REGS;						      \
-	sld.w	PTO+PT_GPR(12)[ep], r12;				      \
-	RESTORE_CALL_SAVED_REGS;					      \
-	RESTORE_CT_REGS
-/* Save registers prior to calling scheduler (just before trap returns).
-   We have to save the return-value registers to preserve the trap's return
-   value.  Note that ..._SCHEDULE_EXTRA_STATE_SAVER, unlike most ..._SAVER
-   macros, is required to setup EP itself if EP is needed (this is because
-   in many cases, the macro is empty).  */
-#define TRAP_SCHEDULE_EXTRA_STATE_SAVER					      \
-	mov sp, ep;							      \
-	SAVE_RVAL_REGS
-/* Note that ..._SCHEDULE_EXTRA_STATE_RESTORER, unlike most ..._RESTORER
-   macros, is required to setup EP itself if EP is needed (this is because
-   in many cases, the macro is empty).  */
-#define TRAP_SCHEDULE_EXTRA_STATE_RESTORER				      \
-	mov sp, ep;							      \
-	RESTORE_RVAL_REGS
-
-/* Register saving/restoring for maskable interrupts.  */
-#define IRQ_RET reti
-#define IRQ_STATE_SAVER							      \
-	SAVE_CALL_CLOBBERED_REGS;					      \
-	SAVE_PC(EIPC);							      \
-	SAVE_PSW(EIPSW)
-#define IRQ_STATE_RESTORER						      \
-	RESTORE_CALL_CLOBBERED_REGS;					      \
-	RESTORE_PC(EIPC);						      \
-	RESTORE_PSW(EIPSW)
-#define IRQ_EXTRA_STATE_SAVER						      \
-	SAVE_CALL_SAVED_REGS;						      \
-	SAVE_CT_REGS
-#define IRQ_EXTRA_STATE_RESTORER					      \
-	RESTORE_CALL_SAVED_REGS;					      \
-	RESTORE_CT_REGS
-#define IRQ_SCHEDULE_EXTRA_STATE_SAVER	     /* nothing */
-#define IRQ_SCHEDULE_EXTRA_STATE_RESTORER    /* nothing */
-
-/* Register saving/restoring for non-maskable interrupts.  */
-#define NMI_RET reti
-#define NMI_STATE_SAVER							      \
-	SAVE_CALL_CLOBBERED_REGS;					      \
-	SAVE_PC(FEPC);							      \
-	SAVE_PSW(FEPSW);
-#define NMI_STATE_RESTORER						      \
-	RESTORE_CALL_CLOBBERED_REGS;					      \
-	RESTORE_PC(FEPC);						      \
-	RESTORE_PSW(FEPSW);
-#define NMI_EXTRA_STATE_SAVER						      \
-	SAVE_CALL_SAVED_REGS;						      \
-	SAVE_CT_REGS
-#define NMI_EXTRA_STATE_RESTORER					      \
-	RESTORE_CALL_SAVED_REGS;					      \
-	RESTORE_CT_REGS
-#define NMI_SCHEDULE_EXTRA_STATE_SAVER	     /* nothing */
-#define NMI_SCHEDULE_EXTRA_STATE_RESTORER    /* nothing */
-
-/* Register saving/restoring for debug traps.  */
-#define DBTRAP_RET .long 0x014607E0 /* `dbret', but gas doesn't support it. */
-#define DBTRAP_STATE_SAVER						      \
-	SAVE_CALL_CLOBBERED_REGS;					      \
-	SAVE_PC(DBPC);							      \
-	SAVE_PSW(DBPSW)
-#define DBTRAP_STATE_RESTORER						      \
-	RESTORE_CALL_CLOBBERED_REGS;					      \
-	RESTORE_PC(DBPC);						      \
-	RESTORE_PSW(DBPSW)
-#define DBTRAP_EXTRA_STATE_SAVER					      \
-	SAVE_CALL_SAVED_REGS;						      \
-	SAVE_CT_REGS
-#define DBTRAP_EXTRA_STATE_RESTORER					      \
-	RESTORE_CALL_SAVED_REGS;					      \
-	RESTORE_CT_REGS
-#define DBTRAP_SCHEDULE_EXTRA_STATE_SAVER	/* nothing */
-#define DBTRAP_SCHEDULE_EXTRA_STATE_RESTORER	/* nothing */
-
-/* Register saving/restoring for a context switch.  We don't need to save
-   too many registers, because context-switching looks like a function call
-   (via the function `switch_thread'), so callers will save any
-   call-clobbered registers themselves.  We do need to save the CT regs, as
-   they're normally not saved during kernel entry (the kernel doesn't use
-   them).  We save PSW so that interrupt-status state will correctly follow
-   each thread (mostly NMI vs. normal-IRQ/trap), though for the most part
-   it doesn't matter since threads are always in almost exactly the same
-   processor state during a context switch.  The stack pointer and return
-   value are handled by switch_thread itself.  */
-#define SWITCH_STATE_SAVER						      \
-	SAVE_CALL_SAVED_REGS;						      \
-	SAVE_PSW(PSW);							      \
-	SAVE_CT_REGS
-#define SWITCH_STATE_RESTORER						      \
-	RESTORE_CALL_SAVED_REGS;					      \
-	RESTORE_PSW(PSW);						      \
-	RESTORE_CT_REGS
-
-
-/* Restore register state from the state-save-frame on the stack, switch back
-   to the user stack if necessary, and return from the trap/interrupt.
-   EXTRA_STATE_RESTORER is a sequence of assembly language statements to
-   restore anything not restored by this macro.  Only registers not saved by
-   the C compiler are restored (that is, R3(sp), R4(gp), R31(lp), and
-   anything restored by EXTRA_STATE_RESTORER).  */
-#define RETURN(type)							      \
-	ld.b	PTO+PT_KERNEL_MODE[sp], r19;				      \
-	di;				/* Disable interrupts */	      \
-	cmp	r19, r0;		/* See if returning to kernel mode, */\
-	bne	2f;			/* ... if so, skip resched &c.  */    \
-									      \
-	/* We're returning to user mode, so check for various conditions that \
-	   trigger rescheduling. */					      \
-	GET_CURRENT_THREAD(r18);					      \
-	ld.w	TI_FLAGS[r18], r19;					      \
-	andi	_TIF_NEED_RESCHED, r19, r0;				      \
-	bnz	3f;			/* Call the scheduler.  */	      \
-5:	andi	_TIF_SIGPENDING, r19, r18;				      \
-	ld.w	TASK_PTRACE[CURRENT_TASK], r19; /* ptrace flags */	      \
-	or	r18, r19;		/* see if either is non-zero */	      \
-	bnz	4f;			/* if so, handle them */	      \
-									      \
-/* Return to user state.  */						      \
-1:	st.b	r0, KM;			/* Now officially in user state. */   \
-									      \
-/* Final return.  The stack-pointer fiddling is not needed when returning     \
-   to kernel-mode, but they don't hurt, and this way we can share the	      \
-   (sometimes rather lengthy) POP_STATE macro.  */			      \
-2:	POP_STATE(type);						      \
-	st.w	sp, KSP;		/* Save the kernel stack pointer. */  \
-	ld.w	PT_GPR(GPR_SP)-PT_SIZE[sp], sp; /* Restore stack pointer. */  \
-	type ## _RET;			/* Return from the trap/interrupt. */ \
-									      \
-/* Call the scheduler before returning from a syscall/trap. */		      \
-3:	SAVE_EXTRA_STATE_FOR_SCHEDULE(type); /* Prepare to call scheduler. */ \
-	jarl	call_scheduler, lp;	/* Call scheduler */		      \
-	di;				/* The scheduler enables interrupts */\
-	RESTORE_EXTRA_STATE_FOR_SCHEDULE(type);				      \
-	GET_CURRENT_THREAD(r18);					      \
-	ld.w	TI_FLAGS[r18], r19;					      \
-	br	5b;			/* Continue with return path. */      \
-									      \
-/* Handle a signal or ptraced process return.				      \
-   r18 should be non-zero if there are pending signals.  */		      \
-4:	/* Not all registers are saved by the normal trap/interrupt entry     \
-	   points (for instance, call-saved registers (because the normal     \
-	   C-compiler calling sequence in the kernel makes sure they're	      \
-	   preserved), and call-clobbered registers in the case of	      \
-	   traps), but signal handlers may want to examine or change the      \
-	   complete register state.  Here we save anything not saved by	      \
-	   the normal entry sequence, so that it may be safely restored	      \
-	   (in a possibly modified form) after do_signal returns.  */	      \
-	SAVE_EXTRA_STATE(type);		/* Save state not saved by entry. */  \
-	jarl	handle_signal_or_ptrace_return, lp;			      \
-	RESTORE_EXTRA_STATE(type);	/* Restore extra regs.  */	      \
-	br	1b
-
-
-/* Jump to the appropriate function for the system call number in r12
-   (r12 is not preserved), or return an error if r12 is not valid.  The
-   LP register should point to the location where the called function
-   should return.  [note that MAKE_SYS_CALL uses label 1]  */
-#define MAKE_SYS_CALL							      \
-	/* Figure out which function to use for this system call.  */	      \
-	shl	2, r12;							      \
-	/* See if the system call number is valid.  */			      \
-	addi	lo(CSYM(sys_call_table) - sys_call_table_end), r12, r0;	      \
-	bnh	1f;							      \
-	mov	hilo(CSYM(sys_call_table)), r19;			      \
-	add	r19, r12;						      \
-	ld.w	0[r12], r12;						      \
-	/* Make the system call.  */					      \
-	jmp	[r12];							      \
-	/* The syscall number is invalid, return an error.  */		      \
-1:	addi	-ENOSYS, r0, r10;					      \
-	jmp	[lp]
-
-
-	.text
-
-/*
- * User trap.
- *
- * Trap 0 system calls are also handled here.
- *
- * The stack-pointer (r3) should have already been saved to the memory
- * location ENTRY_SP (the reason for this is that the interrupt vectors may be
- * beyond a 22-bit signed offset jump from the actual interrupt handler, and
- * this allows them to save the stack-pointer and use that register to do an
- * indirect jump).
- *
- * Syscall protocol:
- *   Syscall number in r12, args in r6-r9
- *   Return value in r10
- */
-G_ENTRY(trap):
-	SAVE_STATE (TRAP, r12, ENTRY_SP) // Save registers.
-	stsr	SR_ECR, r19		// Find out which trap it was.
-	ei				// Enable interrupts.
-	mov	hilo(ret_from_trap), lp	// where the trap should return
-
-	// The following two shifts (1) clear out extraneous NMI data in the
-	// upper 16-bits, (2) convert the 0x40 - 0x5f range of trap ECR
-	// numbers into the (0-31) << 2 range we want, (3) set the flags.
-	shl	27, r19			// chop off all high bits
-	shr	25, r19			// scale back down and then << 2
-	bnz	2f			// See if not trap 0.
-
-	// Trap 0 is a `short' system call, skip general trap table.
-	MAKE_SYS_CALL			// Jump to the syscall function.
-
-2:	// For other traps, use a table lookup.
-	mov	hilo(CSYM(trap_table)), r18
-	add	r19, r18
-	ld.w	0[r18], r18
-	jmp	[r18]			// Jump to the trap handler.
-END(trap)
-
-/* This is just like ret_from_trap, but first restores extra registers
-   saved by some wrappers.  */
-L_ENTRY(restore_extra_regs_and_ret_from_trap):
-	RESTORE_EXTRA_STATE(TRAP)
-	// fall through
-END(restore_extra_regs_and_ret_from_trap)
-
-/* Entry point used to return from a syscall/trap.  */
-L_ENTRY(ret_from_trap):
-	RETURN(TRAP)
-END(ret_from_trap)
-
-
-/* This the initial entry point for a new child thread, with an appropriate
-   stack in place that makes it look that the child is in the middle of an
-   syscall.  This function is actually `returned to' from switch_thread
-   (copy_thread makes ret_from_fork the return address in each new thread's
-   saved context).  */
-C_ENTRY(ret_from_fork):
-	mov	r10, r6			// switch_thread returns the prev task.
-	jarl	CSYM(schedule_tail), lp	// ...which is schedule_tail's arg
-	mov	r0, r10			// Child's fork call should return 0.
-	br	ret_from_trap		// Do normal trap return.
-C_END(ret_from_fork)
-
-
-/*
- * Trap 1: `long' system calls
- * `Long' syscall protocol:
- *   Syscall number in r12, args in r6-r9, r13-r14
- *   Return value in r10
- */
-L_ENTRY(syscall_long):
-	// Push extra arguments on the stack.  Note that by default, the trap
-	// handler reserves enough stack space for 6 arguments, so we don't
-	// have to make any additional room.
-	st.w	r13, 16[sp]		// arg 5
-	st.w	r14, 20[sp]		// arg 6
-
-	// Make sure r13 and r14 are preserved, in case we have to restart a
-	// system call because of a signal (ep has already been set by caller).
-	st.w	r13, PTO+PT_GPR(13)[sp]
-	st.w	r14, PTO+PT_GPR(13)[sp]
-	mov	hilo(ret_from_long_syscall), lp
-
-	MAKE_SYS_CALL			// Jump to the syscall function.
-END(syscall_long)
-
-/* Entry point used to return from a long syscall.  Only needed to restore
-   r13/r14 if the general trap mechanism doesnt' do so.  */
-L_ENTRY(ret_from_long_syscall):
-	ld.w	PTO+PT_GPR(13)[sp], r13 // Restore the extra registers
-	ld.w	PTO+PT_GPR(13)[sp], r14
-	br	ret_from_trap		// The rest is the same as other traps
-END(ret_from_long_syscall)
-
-
-/* These syscalls need access to the struct pt_regs on the stack, so we
-   implement them in assembly (they're basically all wrappers anyway).  */
-
-L_ENTRY(sys_fork_wrapper):
-#ifdef CONFIG_MMU
-	addi	SIGCHLD, r0, r6		   // Arg 0: flags
-	ld.w	PTO+PT_GPR(GPR_SP)[sp], r7 // Arg 1: child SP (use parent's)
-	movea	PTO, sp, r8		   // Arg 2: parent context
-	mov	r0, r9			   // Arg 3/4/5: 0
-	st.w	r0, 16[sp]
-	st.w	r0, 20[sp]
-	mov	hilo(CSYM(do_fork)), r18   // Where the real work gets done
-	br	save_extra_state_tramp	   // Save state and go there
-#else
-	// fork almost works, enough to trick you into looking elsewhere :-(
-	addi	-EINVAL, r0, r10
-	jmp	[lp]
-#endif
-END(sys_fork_wrapper)
-
-L_ENTRY(sys_vfork_wrapper):
-	addi	CLONE_VFORK | CLONE_VM | SIGCHLD, r0, r6 // Arg 0: flags
-	ld.w	PTO+PT_GPR(GPR_SP)[sp], r7 // Arg 1: child SP (use parent's)
-	movea	PTO, sp, r8		   // Arg 2: parent context
-	mov	r0, r9			   // Arg 3/4/5: 0
-	st.w	r0, 16[sp]
-	st.w	r0, 20[sp]
-	mov	hilo(CSYM(do_fork)), r18   // Where the real work gets done
-	br	save_extra_state_tramp	   // Save state and go there
-END(sys_vfork_wrapper)
-
-L_ENTRY(sys_clone_wrapper):
-	ld.w	PTO+PT_GPR(GPR_SP)[sp], r19// parent's stack pointer
-	cmp	r7, r0			   // See if child SP arg (arg 1) is 0.
-	cmov	z, r19, r7, r7		   // ... and use the parent's if so.
-	movea	PTO, sp, r8		   // Arg 2: parent context
-	mov	r0, r9			   // Arg 3/4/5: 0
-	st.w	r0, 16[sp]
-	st.w	r0, 20[sp]
-	mov	hilo(CSYM(do_fork)), r18   // Where the real work gets done
-	br	save_extra_state_tramp	   // Save state and go there
-END(sys_clone_wrapper)
-
-
-L_ENTRY(sys_execve_wrapper):
-	movea	PTO, sp, r9		// add user context as 4th arg
-	jr	CSYM(sys_execve)	// Do real work (tail-call).
-END(sys_execve_wrapper)
-
-
-L_ENTRY(sys_sigsuspend_wrapper):
-	movea	PTO, sp, r7		// add user context as 2nd arg
-	mov	hilo(CSYM(sys_sigsuspend)), r18	// syscall function
-	jarl	save_extra_state_tramp, lp	// Save state and do it
-	br	restore_extra_regs_and_ret_from_trap
-END(sys_sigsuspend_wrapper)
-L_ENTRY(sys_rt_sigsuspend_wrapper):
-	movea	PTO, sp, r8		// add user context as 3rd arg
-	mov	hilo(CSYM(sys_rt_sigsuspend)), r18 // syscall function
-	jarl	save_extra_state_tramp, lp	   // Save state and do it
-	br	restore_extra_regs_and_ret_from_trap
-END(sys_rt_sigsuspend_wrapper)
-
-L_ENTRY(sys_sigreturn_wrapper):
-	movea	PTO, sp, r6		// add user context as 1st arg
-	mov	hilo(CSYM(sys_sigreturn)), r18	// syscall function
-	jarl	save_extra_state_tramp, lp	// Save state and do it
-	br	restore_extra_regs_and_ret_from_trap
-END(sys_sigreturn_wrapper)
-L_ENTRY(sys_rt_sigreturn_wrapper):
-	movea	PTO, sp, r6		// add user context as 1st arg
-	mov	hilo(CSYM(sys_rt_sigreturn)), r18// syscall function
-	jarl	save_extra_state_tramp, lp	 // Save state and do it
-	br	restore_extra_regs_and_ret_from_trap
-END(sys_rt_sigreturn_wrapper)
-
-
-/* Save any state not saved by SAVE_STATE(TRAP), and jump to r18.
-   It's main purpose is to share the rather lengthy code sequence that
-   SAVE_STATE expands into among the above wrapper functions.  */
-L_ENTRY(save_extra_state_tramp):
-	SAVE_EXTRA_STATE(TRAP)		// Save state not saved by entry.
-	jmp	[r18]			// Do the work the caller wants
-END(save_extra_state_tramp)
-
-
-/*
- * Hardware maskable interrupts.
- *
- * The stack-pointer (r3) should have already been saved to the memory
- * location ENTRY_SP (the reason for this is that the interrupt vectors may be
- * beyond a 22-bit signed offset jump from the actual interrupt handler, and
- * this allows them to save the stack-pointer and use that register to do an
- * indirect jump).
- */
-G_ENTRY(irq):
-	SAVE_STATE (IRQ, r0, ENTRY_SP)	// Save registers.
-
-	stsr	SR_ECR, r6		// Find out which interrupt it was.
-	movea	PTO, sp, r7		// User regs are arg2
-
-	// All v850 implementations I know about encode their interrupts as
-	// multiples of 0x10, starting at 0x80 (after NMIs and software
-	// interrupts).  Convert this number into a simple IRQ index for the
-	// rest of the kernel.  We also clear the upper 16 bits, which hold
-	// NMI info, and don't appear to be cleared when a NMI returns.
-	shl	16, r6			// clear upper 16 bits
-	shr	20, r6			// shift back, and remove lower nibble
-	add	-8, r6			// remove bias for irqs
-
-	// Call the high-level interrupt handling code.
-	jarl	CSYM(handle_irq), lp
-
-	RETURN(IRQ)
-END(irq)
-
-
-/*
- * Debug trap / illegal-instruction exception
- *
- * The stack-pointer (r3) should have already been saved to the memory
- * location ENTRY_SP (the reason for this is that the interrupt vectors may be
- * beyond a 22-bit signed offset jump from the actual interrupt handler, and
- * this allows them to save the stack-pointer and use that register to do an
- * indirect jump).
- */
-G_ENTRY(dbtrap):
-	SAVE_STATE (DBTRAP, r0, ENTRY_SP)// Save registers.
-
-	/* First see if we came from kernel mode; if so, the dbtrap
-	   instruction has a special meaning, to set the DIR (`debug
-	   information register') register.  This is because the DIR register
-	   can _only_ be manipulated/read while in `debug mode,' and debug
-	   mode is only active while we're inside the dbtrap handler.  The
-	   exact functionality is:  { DIR = (DIR | r6) & ~r7; return DIR; }. */
-	ld.b	PTO+PT_KERNEL_MODE[sp], r19
-	cmp	r19, r0
-	bz	1f
-
-	stsr	SR_DIR, r10
-	or	r6, r10
-	not	r7, r7
-	and	r7, r10
-	ldsr	r10, SR_DIR
-	stsr	SR_DIR, r10		// Confirm the value we set
-	st.w	r10, PTO+PT_GPR(10)[sp]	// return it
-	br	3f
-
-1:	ei				// Enable interrupts.
-
-	/* The default signal type we raise.  */
-	mov	SIGTRAP, r6
-
-	/* See if it's a single-step trap.  */
-	stsr	SR_DBPSW, r19
-	andi	0x0800, r19, r19
-	bnz	2f
-
-	/* Look to see if the preceding instruction was is a dbtrap or not,
-	   to decide which signal we should use.  */
-	stsr	SR_DBPC, r19		// PC following trapping insn
-	ld.hu	-2[r19], r19
-	ori	0xf840, r0, r20		// DBTRAP insn
-	cmp	r19, r20		// Was this trap caused by DBTRAP?
-	cmov	ne, SIGILL, r6, r6	// Choose signal appropriately
-
-	/* Raise the desired signal.  */
-2:	mov	CURRENT_TASK, r7	// Arg 1: task
-	jarl	CSYM(send_sig), lp	// tail call
-
-3:	RETURN(DBTRAP)
-END(dbtrap)
-
-
-/*
- * Hardware non-maskable interrupts.
- *
- * The stack-pointer (r3) should have already been saved to the memory
- * location ENTRY_SP (the reason for this is that the interrupt vectors may be
- * beyond a 22-bit signed offset jump from the actual interrupt handler, and
- * this allows them to save the stack-pointer and use that register to do an
- * indirect jump).
- */
-G_ENTRY(nmi):
-	SAVE_STATE (NMI, r0, NMI_ENTRY_SP); /* Save registers.  */
-
-	stsr	SR_ECR, r6;		/* Find out which nmi it was.  */
-	shr	20, r6;			/* Extract NMI code in bits 20-24. */
-	movea	PTO, sp, r7;		/* User regs are arg2.  */
-
-	/* Non-maskable interrupts always lie right after maskable interrupts.
-	   Call the generic IRQ handler, with two arguments, the IRQ number,
-	   and a pointer to the user registers, to handle the specifics.
-	   (we subtract one because the first NMI has code 1).  */
-	addi	FIRST_NMI - 1, r6, r6
-	jarl	CSYM(handle_irq), lp
-
-	RETURN(NMI)
-END(nmi)
-
-
-/*
- * Trap with no handler
- */
-L_ENTRY(bad_trap_wrapper):
-	mov	r19, r6			// Arg 0: trap number
-	movea	PTO, sp, r7		// Arg 1: user regs
-	jr	CSYM(bad_trap)		// tail call handler
-END(bad_trap_wrapper)
-
-
-/*
- * Invoke the scheduler, called from the trap/irq kernel exit path.
- *
- * This basically just calls `schedule', but also arranges for extra
- * registers to be saved for ptrace'd processes, so ptrace can modify them.
- */
-L_ENTRY(call_scheduler):
-	ld.w	TASK_PTRACE[CURRENT_TASK], r19	// See if task is ptrace'd
-	cmp	r19, r0
-	bnz	1f			// ... yes, do special stuff
-	jr	CSYM(schedule)		// ... no, just tail-call scheduler
-
-	// Save extra regs for ptrace'd task.  We want to save anything
-	// that would otherwise only be `implicitly' saved by the normal
-	// compiler calling-convention.
-1:	mov	sp, ep			// Setup EP for SAVE_CALL_SAVED_REGS
-	SAVE_CALL_SAVED_REGS		// Save call-saved registers to stack
-	mov	lp, r20			// Save LP in a callee-saved register
-
-	jarl	CSYM(schedule), lp	// Call scheduler
-
-	mov	r20, lp
-	mov	sp, ep			// We can't rely on EP after return
-	RESTORE_CALL_SAVED_REGS		// Restore (possibly modified) regs
-	jmp	[lp]			// Return to the return path
-END(call_scheduler)
-
-
-/*
- * This is an out-of-line handler for two special cases during the kernel
- * trap/irq exit sequence:
- *
- *  (1) If r18 is non-zero then a signal needs to be handled, which is
- *	done, and then the caller returned to.
- *
- *  (2) If r18 is non-zero then we're returning to a ptraced process, which
- *	has several special cases -- single-stepping and trap tracing, both
- *	of which require using the `dbret' instruction to exit the kernel
- *	instead of the normal `reti' (this is because the CPU not correctly
- *	single-step after a reti).  In this case, of course, this handler
- *	never returns to the caller.
- *
- * In either case, all registers should have been saved to the current
- * state-save-frame on the stack, except for callee-saved registers.
- *
- * [These two different cases are combined merely to avoid bloating the
- * macro-inlined code, not because they really make much sense together!]
- */
-L_ENTRY(handle_signal_or_ptrace_return):
-	cmp	r18, r0			// See if handling a signal
-	bz	1f			// ... nope, go do ptrace return
-
-	// Handle a signal
-	mov	lp, r20			// Save link-pointer
-	mov	r10, r21		// Save return-values (for trap)
-	mov	r11, r22
-
-	movea	PTO, sp, r6		// Arg 1: struct pt_regs *regs
-	mov	r0, r7			// Arg 2: sigset_t *oldset
-	jarl	CSYM(do_signal), lp	// Handle the signal
-	di				// sig handling enables interrupts
-
-	mov	r20, lp			// Restore link-pointer
-	mov	r21, r10		// Restore return-values (for trap)
-	mov	r22, r11
-	ld.w	TASK_PTRACE[CURRENT_TASK], r19  // check ptrace flags too
-	cmp	r19, r0
-	bnz	1f			// ... some set, so look more
-2:	jmp	[lp]			// ... none set, so return normally
-
-	// ptrace return
-1:	ld.w	PTO+PT_PSW[sp], r19	// Look at user-processes's flags
-	andi	0x0800, r19, r19	// See if single-step flag is set
-	bz	2b			// ... nope, return normally
-
-	// Return as if from a dbtrap insn
-	st.b	r0, KM			// Now officially in user state.
-	POP_STATE(DBTRAP)		// Restore regs
-	st.w	sp, KSP			// Save the kernel stack pointer.
-	ld.w	PT_GPR(GPR_SP)-PT_SIZE[sp], sp // Restore user stack pointer.
-	DBTRAP_RET			// Return from the trap/interrupt.
-END(handle_signal_or_ptrace_return)
-
-
-/*
- * This is where we switch between two threads.  The arguments are:
- *   r6 -- pointer to the struct thread for the `current' process
- *   r7 -- pointer to the struct thread for the `new' process.
- * when this function returns, it will return to the new thread.
- */
-C_ENTRY(switch_thread):
-	// Return the previous task (r10 is not clobbered by restore below)
-	mov	CURRENT_TASK, r10
-	// First, push the current processor state on the stack
-	PUSH_STATE(SWITCH)
-	// Now save the location of the kernel stack pointer for this thread;
-	// since we've pushed all other state on the stack, this is enough to
-	// restore it all later.
-	st.w	sp, THREAD_KSP[r6]
-	// Now restore the stack pointer from the new process
-	ld.w	THREAD_KSP[r7], sp
-	// ... and restore all state from that
-	POP_STATE(SWITCH)
-	// Update the current task pointer
-	GET_CURRENT_TASK(CURRENT_TASK)
-	// Now return into the new thread
-	jmp	[lp]
-C_END(switch_thread)
-
-
-	.data
-
-	.align 4
-C_DATA(trap_table):
-	.long bad_trap_wrapper		// trap 0, doesn't use trap table.
-	.long syscall_long		// trap 1, `long' syscall.
-	.long bad_trap_wrapper
-	.long bad_trap_wrapper
-	.long bad_trap_wrapper
-	.long bad_trap_wrapper
-	.long bad_trap_wrapper
-	.long bad_trap_wrapper
-	.long bad_trap_wrapper
-	.long bad_trap_wrapper
-	.long bad_trap_wrapper
-	.long bad_trap_wrapper
-	.long bad_trap_wrapper
-	.long bad_trap_wrapper
-	.long bad_trap_wrapper
-	.long bad_trap_wrapper
-C_END(trap_table)
-
-
-	.section .rodata
-
-	.align 4
-C_DATA(sys_call_table):
-	.long CSYM(sys_restart_syscall)	// 0
-	.long CSYM(sys_exit)
-	.long sys_fork_wrapper
-	.long CSYM(sys_read)
-	.long CSYM(sys_write)
-	.long CSYM(sys_open)		// 5
-	.long CSYM(sys_close)
-	.long CSYM(sys_waitpid)
-	.long CSYM(sys_creat)
-	.long CSYM(sys_link)
-	.long CSYM(sys_unlink)		// 10
-	.long sys_execve_wrapper
-	.long CSYM(sys_chdir)
-	.long CSYM(sys_time)
-	.long CSYM(sys_mknod)
-	.long CSYM(sys_chmod)		// 15
-	.long CSYM(sys_chown)
-	.long CSYM(sys_ni_syscall)	// was: break
-	.long CSYM(sys_ni_syscall)	// was: oldstat (aka stat)
-	.long CSYM(sys_lseek)
-	.long CSYM(sys_getpid)		// 20
-	.long CSYM(sys_mount)
-	.long CSYM(sys_oldumount)
-	.long CSYM(sys_setuid)
-	.long CSYM(sys_getuid)
-	.long CSYM(sys_stime)		// 25
-	.long CSYM(sys_ptrace)
-	.long CSYM(sys_alarm)
-	.long CSYM(sys_ni_syscall)	// was: oldfstat (aka fstat)
-	.long CSYM(sys_pause)
-	.long CSYM(sys_utime)		// 30
-	.long CSYM(sys_ni_syscall)	// was: stty
-	.long CSYM(sys_ni_syscall)	// was: gtty
-	.long CSYM(sys_access)
-	.long CSYM(sys_nice)
-	.long CSYM(sys_ni_syscall)	// 35, was: ftime
-	.long CSYM(sys_sync)
-	.long CSYM(sys_kill)
-	.long CSYM(sys_rename)
-	.long CSYM(sys_mkdir)
-	.long CSYM(sys_rmdir)		// 40
-	.long CSYM(sys_dup)
-	.long CSYM(sys_pipe)
-	.long CSYM(sys_times)
-	.long CSYM(sys_ni_syscall)	// was: prof
-	.long CSYM(sys_brk)		// 45
-	.long CSYM(sys_setgid)
-	.long CSYM(sys_getgid)
-	.long CSYM(sys_signal)
-	.long CSYM(sys_geteuid)
-	.long CSYM(sys_getegid)		// 50
-	.long CSYM(sys_acct)
-	.long CSYM(sys_umount)		// recycled never used phys()
-	.long CSYM(sys_ni_syscall)	// was: lock
-	.long CSYM(sys_ioctl)
-	.long CSYM(sys_fcntl)		// 55
-	.long CSYM(sys_ni_syscall)	// was: mpx
-	.long CSYM(sys_setpgid)
-	.long CSYM(sys_ni_syscall)	// was: ulimit
-	.long CSYM(sys_ni_syscall)
-	.long CSYM(sys_umask)		// 60
-	.long CSYM(sys_chroot)
-	.long CSYM(sys_ustat)
-	.long CSYM(sys_dup2)
-	.long CSYM(sys_getppid)
-	.long CSYM(sys_getpgrp)		// 65
-	.long CSYM(sys_setsid)
-	.long CSYM(sys_sigaction)
-	.long CSYM(sys_sgetmask)
-	.long CSYM(sys_ssetmask)
-	.long CSYM(sys_setreuid)	// 70
-	.long CSYM(sys_setregid)
-	.long sys_sigsuspend_wrapper
-	.long CSYM(sys_sigpending)
-	.long CSYM(sys_sethostname)
-	.long CSYM(sys_setrlimit)	// 75
-	.long CSYM(sys_getrlimit)
-	.long CSYM(sys_getrusage)
-	.long CSYM(sys_gettimeofday)
-	.long CSYM(sys_settimeofday)
-	.long CSYM(sys_getgroups)	// 80
-	.long CSYM(sys_setgroups)
-	.long CSYM(sys_select)
-	.long CSYM(sys_symlink)
-	.long CSYM(sys_ni_syscall)	// was: oldlstat (aka lstat)
-	.long CSYM(sys_readlink)	// 85
-	.long CSYM(sys_uselib)
-	.long CSYM(sys_swapon)
-	.long CSYM(sys_reboot)
-	.long CSYM(old_readdir)
-	.long CSYM(sys_mmap)		// 90
-	.long CSYM(sys_munmap)
-	.long CSYM(sys_truncate)
-	.long CSYM(sys_ftruncate)
-	.long CSYM(sys_fchmod)
-	.long CSYM(sys_fchown)		// 95
-	.long CSYM(sys_getpriority)
-	.long CSYM(sys_setpriority)
-	.long CSYM(sys_ni_syscall)	// was: profil
-	.long CSYM(sys_statfs)
-	.long CSYM(sys_fstatfs)		// 100
-	.long CSYM(sys_ni_syscall)	// i386: ioperm
-	.long CSYM(sys_socketcall)
-	.long CSYM(sys_syslog)
-	.long CSYM(sys_setitimer)
-	.long CSYM(sys_getitimer)	// 105
-	.long CSYM(sys_newstat)
-	.long CSYM(sys_newlstat)
-	.long CSYM(sys_newfstat)
-	.long CSYM(sys_ni_syscall)	// was: olduname (aka uname)
-	.long CSYM(sys_ni_syscall)	// 110, i386: iopl
-	.long CSYM(sys_vhangup)
-	.long CSYM(sys_ni_syscall)	// was: idle
-	.long CSYM(sys_ni_syscall)	// i386: vm86old
-	.long CSYM(sys_wait4)
-	.long CSYM(sys_swapoff)		// 115
-	.long CSYM(sys_sysinfo)
-	.long CSYM(sys_ipc)
-	.long CSYM(sys_fsync)
-	.long sys_sigreturn_wrapper
-	.long sys_clone_wrapper		// 120
-	.long CSYM(sys_setdomainname)
-	.long CSYM(sys_newuname)
-	.long CSYM(sys_ni_syscall)	// i386: modify_ldt, m68k: cacheflush
-	.long CSYM(sys_adjtimex)
-	.long CSYM(sys_ni_syscall)	// 125 - sys_mprotect
-	.long CSYM(sys_sigprocmask)
-	.long CSYM(sys_ni_syscall)	// sys_create_module
-	.long CSYM(sys_init_module)
-	.long CSYM(sys_delete_module)
-	.long CSYM(sys_ni_syscall)	// 130 - sys_get_kernel_syms
-	.long CSYM(sys_quotactl)
-	.long CSYM(sys_getpgid)
-	.long CSYM(sys_fchdir)
-	.long CSYM(sys_bdflush)
-	.long CSYM(sys_sysfs)		// 135
-	.long CSYM(sys_personality)
-	.long CSYM(sys_ni_syscall)	// for afs_syscall
-	.long CSYM(sys_setfsuid)
-	.long CSYM(sys_setfsgid)
-	.long CSYM(sys_llseek)		// 140
-	.long CSYM(sys_getdents)
-	.long CSYM(sys_select)		// for backward compat; remove someday
-	.long CSYM(sys_flock)
-	.long CSYM(sys_ni_syscall)	// sys_msync
-	.long CSYM(sys_readv)		// 145
-	.long CSYM(sys_writev)
-	.long CSYM(sys_getsid)
-	.long CSYM(sys_fdatasync)
-	.long CSYM(sys_sysctl)
-	.long CSYM(sys_ni_syscall)	// 150 - sys_mlock
-	.long CSYM(sys_ni_syscall)	// sys_munlock
-	.long CSYM(sys_ni_syscall)	// sys_mlockall
-	.long CSYM(sys_ni_syscall)	// sys_munlockall
-	.long CSYM(sys_sched_setparam)
-	.long CSYM(sys_sched_getparam)	// 155
-	.long CSYM(sys_sched_setscheduler)
-	.long CSYM(sys_sched_getscheduler)
-	.long CSYM(sys_sched_yield)
-	.long CSYM(sys_sched_get_priority_max)
-	.long CSYM(sys_sched_get_priority_min)	// 160
-	.long CSYM(sys_sched_rr_get_interval)
-	.long CSYM(sys_nanosleep)
-	.long CSYM(sys_ni_syscall)	// sys_mremap
-	.long CSYM(sys_setresuid)
-	.long CSYM(sys_getresuid)	// 165
-	.long CSYM(sys_ni_syscall)	// for vm86
-	.long CSYM(sys_ni_syscall)	// sys_query_module
-	.long CSYM(sys_poll)
-	.long CSYM(sys_nfsservctl)
-	.long CSYM(sys_setresgid)	// 170
-	.long CSYM(sys_getresgid)
-	.long CSYM(sys_prctl)
-	.long sys_rt_sigreturn_wrapper
-	.long CSYM(sys_rt_sigaction)
-	.long CSYM(sys_rt_sigprocmask)	// 175
-	.long CSYM(sys_rt_sigpending)
-	.long CSYM(sys_rt_sigtimedwait)
-	.long CSYM(sys_rt_sigqueueinfo)
-	.long sys_rt_sigsuspend_wrapper
-	.long CSYM(sys_pread64)		// 180
-	.long CSYM(sys_pwrite64)
-	.long CSYM(sys_lchown)
-	.long CSYM(sys_getcwd)
-	.long CSYM(sys_capget)
-	.long CSYM(sys_capset)		// 185
-	.long CSYM(sys_sigaltstack)
-	.long CSYM(sys_sendfile)
-	.long CSYM(sys_ni_syscall)	// streams1
-	.long CSYM(sys_ni_syscall)	// streams2
-	.long sys_vfork_wrapper		// 190
-	.long CSYM(sys_ni_syscall)
-	.long CSYM(sys_mmap2)
-	.long CSYM(sys_truncate64)
-	.long CSYM(sys_ftruncate64)
-	.long CSYM(sys_stat64)		// 195
-	.long CSYM(sys_lstat64)
-	.long CSYM(sys_fstat64)
-	.long CSYM(sys_fcntl64)
-	.long CSYM(sys_getdents64)
-	.long CSYM(sys_pivot_root)	// 200
-	.long CSYM(sys_gettid)
-	.long CSYM(sys_tkill)
-sys_call_table_end:
-C_END(sys_call_table)
diff --git a/arch/v850/kernel/fpga85e2c.c b/arch/v850/kernel/fpga85e2c.c
deleted file mode 100644
index ab9cf16a85c8..000000000000
--- a/arch/v850/kernel/fpga85e2c.c
+++ /dev/null
@@ -1,167 +0,0 @@
-/*
- * arch/v850/kernel/fpga85e2c.h -- Machine-dependent defs for
- *	FPGA implementation of V850E2/NA85E2C
- *
- *  Copyright (C) 2002,03  NEC Electronics Corporation
- *  Copyright (C) 2002,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/init.h>
-#include <linux/mm.h>
-#include <linux/swap.h>
-#include <linux/bootmem.h>
-#include <linux/irq.h>
-#include <linux/bitops.h>
-
-#include <asm/atomic.h>
-#include <asm/page.h>
-#include <asm/machdep.h>
-
-#include "mach.h"
-
-extern void memcons_setup (void);
-
-
-#define REG_DUMP_ADDR		0x220000
-
-
-extern struct irqaction reg_snap_action; /* fwd decl */
-
-
-void __init mach_early_init (void)
-{
-	int i;
-	const u32 *src;
-	register u32 *dst asm ("ep");
-	extern u32 _intv_end, _intv_load_start;
-
-	/* Set bus sizes: CS0 32-bit, CS1 16-bit, CS7 8-bit,
-	   everything else 32-bit.  */
-	V850E2_BSC = 0x2AA6;
-	for (i = 2; i <= 6; i++)
-		CSDEV(i) = 0;	/* 32 bit */
-
-	/* Ensure that the simulator halts on a panic, instead of going
-	   into an infinite loop inside the panic function.  */
-	panic_timeout = -1;
-
-	/* Move the interrupt vectors into their real location.  Note that
-	   any relocations there are relative to the real location, so we
-	   don't have to fix anything up.  We use a loop instead of calling
-	   memcpy to keep this a leaf function (to avoid a function
-	   prologue being generated).  */
-	dst = 0x10;		/* &_intv_start + 0x10.  */
-	src = &_intv_load_start;
-	do {
-		u32 t0 = src[0], t1 = src[1], t2 = src[2], t3 = src[3];
-		u32 t4 = src[4], t5 = src[5], t6 = src[6], t7 = src[7];
-		dst[0] = t0; dst[1] = t1; dst[2] = t2; dst[3] = t3;
-		dst[4] = t4; dst[5] = t5; dst[6] = t6; dst[7] = t7;
-		dst += 8;
-		src += 8;
-	} while (dst < &_intv_end);
-}
-
-void __init mach_setup (char **cmdline)
-{
-	memcons_setup ();
-
-	/* Setup up NMI0 to copy the registers to a known memory location.
-	   The FGPA board has a button that produces NMI0 when pressed, so
-	   this allows us to push the button, and then look at memory to see
-	   what's in the registers (there's no other way to easily do so).
-	   We have to use `setup_irq' instead of `request_irq' because it's
-	   still too early to do memory allocation.  */
-	setup_irq (IRQ_NMI (0), &reg_snap_action);
-}
-
-void mach_get_physical_ram (unsigned long *ram_start, unsigned long *ram_len)
-{
-	*ram_start = ERAM_ADDR;
-	*ram_len = ERAM_SIZE;
-}
-
-void __init mach_sched_init (struct irqaction *timer_action)
-{
-	/* Setup up the timer interrupt.  The FPGA peripheral control
-	   registers _only_ work with single-bit writes (set1/clr1)!  */
-	__clear_bit (RPU_GTMC_CE_BIT, &RPU_GTMC);
-	__clear_bit (RPU_GTMC_CLK_BIT, &RPU_GTMC);
-	__set_bit (RPU_GTMC_CE_BIT, &RPU_GTMC);
-
-	/* We use the first RPU interrupt, which occurs every 8.192ms.  */
-	setup_irq (IRQ_RPU (0), timer_action);
-}
-
-
-void mach_gettimeofday (struct timespec *tv)
-{
-	tv->tv_sec = 0;
-	tv->tv_nsec = 0;
-}
-
-void machine_halt (void) __attribute__ ((noreturn));
-void machine_halt (void)
-{
-	for (;;) {
-		DWC(0) = 0x7777;
-		DWC(1) = 0x7777;
-		ASC = 0xffff;
-		FLGREG(0) = 1;	/* Halt immediately.  */
-		asm ("di; halt; nop; nop; nop; nop; nop");
-	}
-}
-
-void machine_restart (char *__unused)
-{
-	machine_halt ();
-}
-
-void machine_power_off (void)
-{
-	machine_halt ();
-}
-
-
-/* Interrupts */
-
-struct v850e_intc_irq_init irq_inits[] = {
-	{ "IRQ", 0, 		NUM_MACH_IRQS,	1, 7 },
-	{ "RPU", IRQ_RPU(0),	IRQ_RPU_NUM,	1, 6 },
-	{ 0 }
-};
-#define NUM_IRQ_INITS (ARRAY_SIZE(irq_inits) - 1)
-
-struct hw_interrupt_type hw_itypes[NUM_IRQ_INITS];
-
-/* Initialize interrupts.  */
-void __init mach_init_irqs (void)
-{
-	v850e_intc_init_irq_types (irq_inits, hw_itypes);
-}
-
-
-/* An interrupt handler that copies the registers to a known memory location,
-   for debugging purposes.  */
-
-static void make_reg_snap (int irq, void *dummy, struct pt_regs *regs)
-{
-	(*(unsigned *)REG_DUMP_ADDR)++;
-	(*(struct pt_regs *)(REG_DUMP_ADDR + sizeof (unsigned))) = *regs;
-}
-
-static int reg_snap_dev_id;
-static struct irqaction reg_snap_action = {
-	.handler = make_reg_snap,
-	.mask = CPU_MASK_NONE,
-	.name = "reg_snap",
-	.dev_id = &reg_snap_dev_id,
-};
diff --git a/arch/v850/kernel/fpga85e2c.ld b/arch/v850/kernel/fpga85e2c.ld
deleted file mode 100644
index b5d4578ae411..000000000000
--- a/arch/v850/kernel/fpga85e2c.ld
+++ /dev/null
@@ -1,62 +0,0 @@
-/* Linker script for the FPGA implementation of the V850E2 NA85E2C cpu core
-   (CONFIG_V850E2_FPGA85E2C).  */
-
-MEMORY {
-	/* Reset vector.  */
-	RESET	 : ORIGIN = 0, LENGTH = 0x10
-	/* Interrupt vectors.  */
-	INTV      : ORIGIN = 0x10, LENGTH = 0x470
-	/* The `window' in RAM were we're allowed to load stuff.  */
-	RAM_LOW   : ORIGIN = 0x480, LENGTH = 0x0005FB80
-	/* Some more ram above the window were we can put bss &c.  */
-	RAM_HIGH  : ORIGIN = 0x00060000, LENGTH = 0x000A0000
-	/* This is the area visible from the outside world (we can use
-	   this only for uninitialized data).  */
-	VISIBLE   : ORIGIN = 0x00200000, LENGTH = 0x00060000
-}
-
-SECTIONS {
-	.reset : {
-		__kram_start = . ;
-		__intv_start = . ;
-	        	*(.intv.reset)	/* Reset vector */
-	} > RESET
-
-	.ram_low : {
-		__r0_ram = . ;		/* Must be near address 0.  */
-		. = . + 32 ;
-
-		TEXT_CONTENTS
-		DATA_CONTENTS
-		ROOT_FS_CONTENTS
-		RAMK_INIT_CONTENTS_NO_END
-		INITRAMFS_CONTENTS
-	} > RAM_LOW
-
-        /* Where the interrupt vectors are initially loaded.  */
-	__intv_load_start = . ;
-
-	.intv : {
-			*(.intv.common)	/* Vectors common to all v850e proc. */
-			*(.intv.mach)	/* Machine-specific int. vectors.  */
-		__intv_end = . ;
-	} > INTV  AT> RAM_LOW
-
-	.ram_high : {
-		/* This is here so that when we free init memory the
-		   load-time copy of the interrupt vectors and any empty
-		   space at the end of the `RAM_LOW' area is freed too.  */
-		. = ALIGN (4096);
-		__init_end = . ;
-
-		BSS_CONTENTS
-		__kram_end = . ;
-		BOOTMAP_CONTENTS
-	} > RAM_HIGH
-
-	.visible : {
-		_memcons_output = . ;
-		. = . + 0x8000 ;
-		_memcons_output_end = . ;
-	} > VISIBLE
-}
diff --git a/arch/v850/kernel/gbus_int.c b/arch/v850/kernel/gbus_int.c
deleted file mode 100644
index b2bcc251f65b..000000000000
--- a/arch/v850/kernel/gbus_int.c
+++ /dev/null
@@ -1,271 +0,0 @@
-/*
- * arch/v850/kernel/gbus_int.c -- Midas labs GBUS interrupt support
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/types.h>
-#include <linux/init.h>
-#include <linux/irq.h>
-#include <linux/interrupt.h>
-#include <linux/signal.h>
-#include <linux/kernel.h>
-
-#include <asm/machdep.h>
-
-
-/* The number of shared GINT interrupts. */
-#define NUM_GINTS   	4
-
-/* For each GINT interrupt, how many GBUS interrupts are using it.  */
-static unsigned gint_num_active_irqs[NUM_GINTS] = { 0 };
-
-/* A table of GINTn interrupts we actually use.
-   Note that we don't use GINT0 because all the boards we support treat it
-   specially.  */
-struct used_gint {
-	unsigned gint;
-	unsigned priority;
-} used_gint[] = {
-	{ 1, GBUS_INT_PRIORITY_HIGH },
-	{ 3, GBUS_INT_PRIORITY_LOW }
-};
-#define NUM_USED_GINTS ARRAY_SIZE(used_gint)
-
-/* A table of which GINT is used by each GBUS interrupts (they are
-   assigned based on priority).  */
-static unsigned char gbus_int_gint[IRQ_GBUS_INT_NUM];
-
-
-/* Interrupt enabling/disabling.  */
-
-/* Enable interrupt handling for interrupt IRQ.  */
-void gbus_int_enable_irq (unsigned irq)
-{
-	unsigned gint = gbus_int_gint[irq - GBUS_INT_BASE_IRQ];
-	GBUS_INT_ENABLE (GBUS_INT_IRQ_WORD(irq), gint)
-		|= GBUS_INT_IRQ_MASK (irq);
-}
-
-/* Disable interrupt handling for interrupt IRQ.  Note that any
-   interrupts received while disabled will be delivered once the
-   interrupt is enabled again, unless they are explicitly cleared using
-   `gbus_int_clear_pending_irq'.  */
-void gbus_int_disable_irq (unsigned irq)
-{
-	unsigned gint = gbus_int_gint[irq - GBUS_INT_BASE_IRQ];
-	GBUS_INT_ENABLE (GBUS_INT_IRQ_WORD(irq), gint)
-		&= ~GBUS_INT_IRQ_MASK (irq);
-}
-
-/* Return true if interrupt handling for interrupt IRQ is enabled.  */
-int gbus_int_irq_enabled (unsigned irq)
-{
-	unsigned gint = gbus_int_gint[irq - GBUS_INT_BASE_IRQ];
-	return (GBUS_INT_ENABLE (GBUS_INT_IRQ_WORD(irq), gint)
-		& GBUS_INT_IRQ_MASK(irq));
-}
-
-/* Disable all GBUS irqs.  */
-void gbus_int_disable_irqs ()
-{
-	unsigned w, n;
-	for (w = 0; w < GBUS_INT_NUM_WORDS; w++)
-		for (n = 0; n < IRQ_GINT_NUM; n++)
-			GBUS_INT_ENABLE (w, n) = 0;
-}
-
-/* Clear any pending interrupts for IRQ.  */
-void gbus_int_clear_pending_irq (unsigned irq)
-{
-	GBUS_INT_CLEAR (GBUS_INT_IRQ_WORD(irq)) = GBUS_INT_IRQ_MASK (irq);
-}
-
-/* Return true if interrupt IRQ is pending (but disabled).  */
-int gbus_int_irq_pending (unsigned irq)
-{
-	return (GBUS_INT_STATUS (GBUS_INT_IRQ_WORD(irq))
-		& GBUS_INT_IRQ_MASK(irq));
-}
-
-
-/* Delegating interrupts.  */
-
-/* Handle a shared GINT interrupt by passing to the appropriate GBUS
-   interrupt handler.  */
-static irqreturn_t gbus_int_handle_irq (int irq, void *dev_id,
-					struct pt_regs *regs)
-{
-	unsigned w;
-	irqreturn_t rval = IRQ_NONE;
-	unsigned gint = irq - IRQ_GINT (0);
-
-	for (w = 0; w < GBUS_INT_NUM_WORDS; w++) {
-		unsigned status = GBUS_INT_STATUS (w);
-		unsigned enable = GBUS_INT_ENABLE (w, gint);
-
-		/* Only pay attention to enabled interrupts.  */
-		status &= enable;
-		if (status) {
-			irq = IRQ_GBUS_INT (w * GBUS_INT_BITS_PER_WORD);
-			do {
-				/* There's an active interrupt in word
-				   W, find out which one, and call its
-				   handler.  */
-
-				while (! (status & 0x1)) {
-					irq++;
-					status >>= 1;
-				}
-				status &= ~0x1;
-
-				/* Recursively call handle_irq to handle it. */
-				handle_irq (irq, regs);
-				rval = IRQ_HANDLED;
-			} while (status);
-		}
-	}
-
-	/* Toggle the `all enable' bit back and forth, which should cause
-	   another edge transition if there are any other interrupts
-	   still pending, and so result in another CPU interrupt.  */
-	GBUS_INT_ENABLE (0, gint) &= ~0x1;
-	GBUS_INT_ENABLE (0, gint) |=  0x1;
-
-	return rval;
-}
-
-
-/* Initialize GBUS interrupt sources.  */
-
-static void irq_nop (unsigned irq) { }
-
-static unsigned gbus_int_startup_irq (unsigned irq)
-{
-	unsigned gint = gbus_int_gint[irq - GBUS_INT_BASE_IRQ];
-
-	if (gint_num_active_irqs[gint] == 0) {
-		/* First enable the CPU interrupt.  */
-		int rval =
-			request_irq (IRQ_GINT(gint), gbus_int_handle_irq,
-				     IRQF_DISABLED,
-				     "gbus_int_handler",
-				     &gint_num_active_irqs[gint]);
-		if (rval != 0)
-			return rval;
-	}
-
-	gint_num_active_irqs[gint]++;
-
-	gbus_int_clear_pending_irq (irq);
-	gbus_int_enable_irq (irq);
-
-	return 0;
-}
-
-static void gbus_int_shutdown_irq (unsigned irq)
-{
-	unsigned gint = gbus_int_gint[irq - GBUS_INT_BASE_IRQ];
-
-	gbus_int_disable_irq (irq);
-
-	if (--gint_num_active_irqs[gint] == 0)
-		/* Disable the CPU interrupt.  */
-		free_irq (IRQ_GINT(gint), &gint_num_active_irqs[gint]);
-}
-
-/* Initialize HW_IRQ_TYPES for INTC-controlled irqs described in array
-   INITS (which is terminated by an entry with the name field == 0).  */
-void __init gbus_int_init_irq_types (struct gbus_int_irq_init *inits,
-				     struct hw_interrupt_type *hw_irq_types)
-{
-	struct gbus_int_irq_init *init;
-	for (init = inits; init->name; init++) {
-		unsigned i;
-		struct hw_interrupt_type *hwit = hw_irq_types++;
-
-		hwit->typename = init->name;
-
-		hwit->startup  = gbus_int_startup_irq;
-		hwit->shutdown = gbus_int_shutdown_irq;
-		hwit->enable   = gbus_int_enable_irq;
-		hwit->disable  = gbus_int_disable_irq;
-		hwit->ack      = irq_nop;
-		hwit->end      = irq_nop;
-		
-		/* Initialize kernel IRQ infrastructure for this interrupt.  */
-		init_irq_handlers(init->base, init->num, init->interval, hwit);
-
-		/* Set the interrupt priorities.  */
-		for (i = 0; i < init->num; i++) {
-			unsigned j;
-			for (j = 0; j < NUM_USED_GINTS; j++)
-				if (used_gint[j].priority > init->priority)
-					break;
-			/* Wherever we stopped looking is one past the
-			   GINT we want. */
-			gbus_int_gint[init->base + i * init->interval
-				      - GBUS_INT_BASE_IRQ]
-				= used_gint[j > 0 ? j - 1 : 0].gint;
-		}
-	}
-}
-
-
-/* Initialize IRQS.  */
-
-/* Chip interrupts (GINTn) shared among GBUS interrupts.  */
-static struct hw_interrupt_type gint_hw_itypes[NUM_USED_GINTS];
-
-
-/* GBUS interrupts themselves.  */
-
-struct gbus_int_irq_init gbus_irq_inits[] __initdata = {
-	/* First set defaults.  */
-	{ "GBUS_INT", IRQ_GBUS_INT(0), IRQ_GBUS_INT_NUM, 1, 6},
-	{ 0 }
-};
-#define NUM_GBUS_IRQ_INITS (ARRAY_SIZE(gbus_irq_inits) - 1)
-
-static struct hw_interrupt_type gbus_hw_itypes[NUM_GBUS_IRQ_INITS];
-
-
-/* Initialize GBUS interrupts.  */
-void __init gbus_int_init_irqs (void)
-{
-	unsigned i;
-
-	/* First initialize the shared gint interrupts.  */
-	for (i = 0; i < NUM_USED_GINTS; i++) {
-		unsigned gint = used_gint[i].gint;
-		struct v850e_intc_irq_init gint_irq_init[2];
-
-		/* We initialize one GINT interrupt at a time.  */
-		gint_irq_init[0].name = "GINT";
-		gint_irq_init[0].base = IRQ_GINT (gint);
-		gint_irq_init[0].num = 1;
-		gint_irq_init[0].interval = 1;
-		gint_irq_init[0].priority = used_gint[i].priority;
-
-		gint_irq_init[1].name = 0; /* Terminate the vector.  */
-
-		v850e_intc_init_irq_types (gint_irq_init, gint_hw_itypes);
-	}
-
-	/* Then the GBUS interrupts.  */
-	gbus_int_disable_irqs ();
-	gbus_int_init_irq_types (gbus_irq_inits, gbus_hw_itypes);
-	/* Turn on the `all enable' bits, which are ANDed with
-	   individual interrupt enable bits; we only want to bother with
-	   the latter.  They are the first bit in the first word of each
-	   interrupt-enable area.  */
-	for (i = 0; i < NUM_USED_GINTS; i++)
-		GBUS_INT_ENABLE (0, used_gint[i].gint) = 0x1;
-}
diff --git a/arch/v850/kernel/head.S b/arch/v850/kernel/head.S
deleted file mode 100644
index c490b937ef14..000000000000
--- a/arch/v850/kernel/head.S
+++ /dev/null
@@ -1,128 +0,0 @@
-/*
- * arch/v850/kernel/head.S -- Lowest-level startup code
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <asm/clinkage.h>
-#include <asm/current.h>
-#include <asm/entry.h>
-#include <asm/thread_info.h>
-#include <asm/irq.h>
-
-
-/* Make a slightly more convenient alias for C_SYMBOL_NAME.  */
-#define CSYM	C_SYMBOL_NAME
-
-
-	.text
-
-	// Define `mach_early_init' as a weak symbol
-	.global	CSYM(mach_early_init)
-	.weak	CSYM(mach_early_init)
-
-C_ENTRY(start):
-	// Make sure interrupts are turned off, just in case
-	di
-
-#ifdef CONFIG_RESET_GUARD
-	// See if we got here via an unexpected reset
-	ld.w	RESET_GUARD, r19	// Check current value of reset guard
-	mov	RESET_GUARD_ACTIVE, r20
-	cmp	r19, r20
-	bne	1f			// Guard was not active
-
-	// If we get here, the reset guard was active.  Load up some
-	// interesting values as arguments, and jump to the handler.
-	st.w	r0, RESET_GUARD		// Allow further resets to succeed
-	mov	lp, r6			// Arg 0: return address
-	ld.b	KM, r7			// Arg 1: kernel mode
-	mov	sp, r9			// Arg 3: stack pointer
-	ld.w	KSP, r19		// maybe switch to kernel stack
-	cmp	r7, r0			// see if already in kernel mode
-	cmov	z, r19, sp, sp		//  and switch to kernel stack if not
-	GET_CURRENT_TASK(r8)		// Arg 2: task pointer
-	jr	CSYM(unexpected_reset)
-
-1:	st.w	r20, RESET_GUARD	// Turn on reset guard
-#endif /* CONFIG_RESET_GUARD */
-
-	// Setup a temporary stack for doing pre-initialization function calls.
-	// 
-	// We can't use the initial kernel stack, because (1) it may be
-	// located in memory we're not allowed to touch, and (2) since
-	// it's in the data segment, calling memcpy to initialize that
-	// area from ROM will overwrite memcpy's return address.
-	mov	hilo(CSYM(_init_stack_end) - 4), sp
-
-	// See if there's a platform-specific early-initialization routine
-	// defined; it's a weak symbol, so it will have an address of zero if
-	// there's not.
-	mov	hilo(CSYM(mach_early_init)), r6
-	cmp	r6, r0
-	bz	3f
-
-	// There is one, so call it.  If this function is written in C, it
-	// should be very careful -- the stack pointer is valid, but very
-	// little else is (e.g., bss is not zeroed yet, and initialized data
-	// hasn't been).
-	jarl	2f, lp			// first figure out return address
-2:	add	3f - ., lp
-	jmp	[r6]			// do call
-3:
-
-#ifdef CONFIG_ROM_KERNEL
-	// Copy the data area from ROM to RAM
-	mov	hilo(CSYM(_rom_copy_dst_start)), r6
-	mov	hilo(CSYM(_rom_copy_src_start)), r7
-	mov	hilo(CSYM(_rom_copy_dst_end)), r8
-	sub	r6, r8
-	jarl	CSYM(memcpy), lp
-#endif
-
-	// Load the initial thread's stack, and current task pointer (in r16)
-	mov	hilo(CSYM(init_thread_union)), r19
-	movea	THREAD_SIZE, r19, sp
-	ld.w	TI_TASK[r19], CURRENT_TASK
-
-#ifdef CONFIG_TIME_BOOTUP
-	/* This stuff must come after mach_early_init, because interrupts may
-	   not work until after its been called.  */
-	jarl	CSYM(highres_timer_reset), lp
-	jarl	CSYM(highres_timer_start), lp
-#endif
-
-	// Kernel stack pointer save location
-	st.w	sp, KSP
-
-	// Assert that we're in `kernel mode'
-	mov	1, r19
-	st.w	r19, KM
-
-#ifdef CONFIG_ZERO_BSS
-	// Zero bss area, since we can't rely upon any loader to do so
-	mov	hilo(CSYM(_sbss)), r6
-	mov	r0, r7
-	mov	hilo(CSYM(_ebss)), r8
-	sub	r6, r8
-	jarl	CSYM(memset), lp
-#endif
-
-	// What happens if the main kernel function returns (it shouldn't)
-	mov	hilo(CSYM(machine_halt)), lp
-
-	// Start the linux kernel.  We use an indirect jump to get extra
-	// range, because on some platforms this initial startup code
-	// (and the associated platform-specific code in mach_early_init)
-	// are located far away from the main kernel, e.g. so that they
-	// can initialize RAM first and copy the kernel or something.
-	mov	hilo(CSYM(start_kernel)), r12
-	jmp	[r12]
-C_END(start)
diff --git a/arch/v850/kernel/highres_timer.c b/arch/v850/kernel/highres_timer.c
deleted file mode 100644
index b16ad1eaf966..000000000000
--- a/arch/v850/kernel/highres_timer.c
+++ /dev/null
@@ -1,132 +0,0 @@
-/*
- * arch/v850/kernel/highres_timer.c -- High resolution timing routines
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <asm/system.h>
-#include <asm/v850e_timer_d.h>
-#include <asm/highres_timer.h>
-
-#define HIGHRES_TIMER_USEC_SHIFT   12
-
-/* Pre-calculated constant used for converting ticks to real time
-   units.  We initialize it to prevent it being put into BSS.  */
-static u32 highres_timer_usec_prescale = 1;
-
-void highres_timer_slow_tick_irq (void) __attribute__ ((noreturn));
-void highres_timer_slow_tick_irq (void)
-{
-	/* This is an interrupt handler, so it must be very careful to
-	   not to trash any registers.  At this point, the stack-pointer
-	   (r3) has been saved in the chip ram location ENTRY_SP by the
-	   interrupt vector, so we can use it as a scratch register; we
-	   must also restore it before returning.  */
-	asm ("ld.w	%0[r0], sp;"
-	     "add	1, sp;"
-	     "st.w	sp, %0[r0];"
-	     "ld.w	%1[r0], sp;" /* restore pre-irq stack-pointer */
-	     "reti"
-	     ::
-	      "i" (HIGHRES_TIMER_SLOW_TICKS_ADDR),
-	      "i" (ENTRY_SP_ADDR)
-	     : "memory");
-}
-
-void highres_timer_reset (void)
-{
-	V850E_TIMER_D_TMD (HIGHRES_TIMER_TIMER_D_UNIT) = 0;
-	HIGHRES_TIMER_SLOW_TICKS = 0;
-}
-
-void highres_timer_start (void)
-{
-	u32 fast_tick_rate;
-
-	/* Start hardware timer.  */
-	v850e_timer_d_configure (HIGHRES_TIMER_TIMER_D_UNIT,
-				 HIGHRES_TIMER_SLOW_TICK_RATE);
-
-	fast_tick_rate =
-		(V850E_TIMER_D_BASE_FREQ
-		 >> V850E_TIMER_D_DIVLOG2 (HIGHRES_TIMER_TIMER_D_UNIT));
-
-	/* The obvious way of calculating microseconds from fast ticks
-	   is to do:
-
-	     usec = fast_ticks * 10^6 / fast_tick_rate
-
-	   However, divisions are much slower than multiplications, and
-	   the above calculation can overflow, so we do this instead:
-
-	     usec = fast_ticks * (10^6 * 2^12 / fast_tick_rate) / 2^12
-
-           since we can pre-calculate (10^6 * (2^12 / fast_tick_rate))
-	   and use a shift for dividing by 2^12, this avoids division,
-	   and is almost as accurate (it differs by about 2 microseconds
-	   at the extreme value of the fast-tick counter's ranger).  */
-	highres_timer_usec_prescale = ((1000000 << HIGHRES_TIMER_USEC_SHIFT)
-				       / fast_tick_rate);
-
-	/* Enable the interrupt (which is hardwired to this use), and
-	   give it the highest priority.  */
-	V850E_INTC_IC (IRQ_INTCMD (HIGHRES_TIMER_TIMER_D_UNIT)) = 0;
-}
-
-void highres_timer_stop (void)
-{
-	/* Stop the timer.  */
-	V850E_TIMER_D_TMCD (HIGHRES_TIMER_TIMER_D_UNIT) =
-		V850E_TIMER_D_TMCD_CAE;
-	/* Disable its interrupt, just in case.  */
-	v850e_intc_disable_irq (IRQ_INTCMD (HIGHRES_TIMER_TIMER_D_UNIT));
-}
-
-inline void highres_timer_read_ticks (u32 *slow_ticks, u32 *fast_ticks)
-{
-	int flags;
-	u32 fast_ticks_1, fast_ticks_2, _slow_ticks;
-
-	local_irq_save (flags);
-	fast_ticks_1 = V850E_TIMER_D_TMD (HIGHRES_TIMER_TIMER_D_UNIT);
-	_slow_ticks = HIGHRES_TIMER_SLOW_TICKS;
-	fast_ticks_2 = V850E_TIMER_D_TMD (HIGHRES_TIMER_TIMER_D_UNIT);
-	local_irq_restore (flags);
-
-	if (fast_ticks_2 < fast_ticks_1)
-		_slow_ticks++;
-
-	*slow_ticks = _slow_ticks;
-	*fast_ticks = fast_ticks_2;
-}
-
-inline void highres_timer_ticks_to_timeval (u32 slow_ticks, u32 fast_ticks,
-					    struct timeval *tv)
-{
-	unsigned long sec, sec_rem, usec;
-
-	usec = ((fast_ticks * highres_timer_usec_prescale)
-		>> HIGHRES_TIMER_USEC_SHIFT);
-
-	sec = slow_ticks / HIGHRES_TIMER_SLOW_TICK_RATE;
-	sec_rem = slow_ticks % HIGHRES_TIMER_SLOW_TICK_RATE;
-
-	usec += sec_rem * (1000000 / HIGHRES_TIMER_SLOW_TICK_RATE);
-
-	tv->tv_sec = sec;
-	tv->tv_usec = usec;
-}
-
-void highres_timer_read (struct timeval *tv)
-{
-	u32 fast_ticks, slow_ticks;
-	highres_timer_read_ticks (&slow_ticks, &fast_ticks);
-	highres_timer_ticks_to_timeval (slow_ticks, fast_ticks, tv);
-}
diff --git a/arch/v850/kernel/init_task.c b/arch/v850/kernel/init_task.c
deleted file mode 100644
index 44b274dff33f..000000000000
--- a/arch/v850/kernel/init_task.c
+++ /dev/null
@@ -1,48 +0,0 @@
-/*
- * arch/v850/kernel/init_task.c -- Initial task/thread structures
- *
- *  Copyright (C) 2002,03  NEC Electronics Corporation
- *  Copyright (C) 2002,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- */
-
-#include <linux/mm.h>
-#include <linux/module.h>
-#include <linux/sched.h>
-#include <linux/init.h>
-#include <linux/init_task.h>
-#include <linux/fs.h>
-#include <linux/mqueue.h>
-
-#include <asm/uaccess.h>
-#include <asm/pgtable.h>
-
-static struct fs_struct init_fs = INIT_FS;
-static struct signal_struct init_signals = INIT_SIGNALS (init_signals);
-static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
-struct mm_struct init_mm = INIT_MM (init_mm);
-
-EXPORT_SYMBOL(init_mm);
-
-/*
- * Initial task structure.
- *
- * All other task structs will be allocated on slabs in fork.c
- */
-struct task_struct init_task = INIT_TASK (init_task);
-
-EXPORT_SYMBOL(init_task);
-
-/*
- * Initial thread structure.
- *
- * We need to make sure that this is 8192-byte aligned due to the
- * way process stacks are handled. This is done by having a special
- * "init_task" linker map entry.
- */
-union thread_union init_thread_union 
-	__attribute__((__section__(".data.init_task"))) =
-		{ INIT_THREAD_INFO(init_task) };
diff --git a/arch/v850/kernel/intv.S b/arch/v850/kernel/intv.S
deleted file mode 100644
index 671e4c6150dd..000000000000
--- a/arch/v850/kernel/intv.S
+++ /dev/null
@@ -1,87 +0,0 @@
-/*
- * arch/v850/kernel/intv.S -- Interrupt vectors
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <asm/clinkage.h>
-#include <asm/irq.h>
-#include <asm/machdep.h>
-#include <asm/entry.h>
-
-#ifdef CONFIG_V850E_HIGHRES_TIMER
-#include <asm/highres_timer.h>
-#endif
-
-/* Jump to an interrupt/trap handler.  These handlers (defined in entry.S)
-   expect the stack-pointer to be saved in ENTRY_SP, so we use sp to do an
-   indirect jump (which avoids problems when the handler is more than a signed
-   22-bit offset away).  */
-#define JUMP_TO_HANDLER(name, sp_save_loc)				      \
-	st.w	sp, sp_save_loc;					      \
-	mov	hilo(name), sp;						      \
-	jmp	[sp]
-
-
-	/* Reset vector.  */
-	.section	.intv.reset, "ax"
-	.org	0x0
-	mov	hilo(C_SYMBOL_NAME(start)), r1;
-	jmp	[r1]
-
-
-	/* Generic interrupt vectors.  */
-	.section	.intv.common, "ax"
-	.balign	0x10
-	JUMP_TO_HANDLER (nmi, NMI_ENTRY_SP)	// 0x10 - NMI0
-	.balign	0x10
-	JUMP_TO_HANDLER (nmi, NMI_ENTRY_SP)	// 0x20 - NMI1
-	.balign	0x10
-	JUMP_TO_HANDLER (nmi, NMI_ENTRY_SP)	// 0x30 - NMI2
-	
-	.balign	0x10
-	JUMP_TO_HANDLER (trap, ENTRY_SP)	// 0x40 - TRAP0n
-	.balign	0x10
-	JUMP_TO_HANDLER (trap, ENTRY_SP)	// 0x50 - TRAP1n
-
-	.balign	0x10
-	JUMP_TO_HANDLER (dbtrap, ENTRY_SP)	// 0x60 - Illegal op / DBTRAP insn
-
-
-	/* Hardware interrupt vectors.  */
-	.section	.intv.mach, "ax"
-	.org	0x0
-
-#if defined (CONFIG_V850E_HIGHRES_TIMER) && defined (IRQ_INTCMD)
-
-	/* Interrupts before the highres timer interrupt.  */
-	.rept	IRQ_INTCMD (HIGHRES_TIMER_TIMER_D_UNIT)
-	.balign	0x10
-	JUMP_TO_HANDLER (irq, ENTRY_SP)
-	.endr
-
-	/* The highres timer interrupt.  */
-	.balign	0x10
-	JUMP_TO_HANDLER (C_SYMBOL_NAME (highres_timer_slow_tick_irq), ENTRY_SP)
-
-	/* Interrupts after the highres timer interrupt.  */
-	.rept	NUM_CPU_IRQS - IRQ_INTCMD (HIGHRES_TIMER_TIMER_D_UNIT) - 1
-	.balign	0x10
-	JUMP_TO_HANDLER (irq, ENTRY_SP)
-	.endr
-
-#else /* No highres timer */
-
-	.rept	NUM_CPU_IRQS
-	.balign	0x10
-	JUMP_TO_HANDLER (irq, ENTRY_SP)
-	.endr
-
-#endif /* Highres timer */
diff --git a/arch/v850/kernel/irq.c b/arch/v850/kernel/irq.c
deleted file mode 100644
index 858c45819aab..000000000000
--- a/arch/v850/kernel/irq.c
+++ /dev/null
@@ -1,123 +0,0 @@
-/*
- * arch/v850/kernel/irq.c -- High-level interrupt handling
- *
- *  Copyright (C) 2001,02,03,04,05  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03,04,05  Miles Bader <miles@gnu.org>
- *  Copyright (C) 1994-2000  Ralf Baechle
- *  Copyright (C) 1992  Linus Torvalds
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * This file was was derived from the mips version, arch/mips/kernel/irq.c
- */
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/irq.h>
-#include <linux/init.h>
-#include <linux/interrupt.h>
-#include <linux/kernel_stat.h>
-#include <linux/slab.h>
-#include <linux/mm.h>
-#include <linux/random.h>
-#include <linux/seq_file.h>
-
-#include <asm/system.h>
-
-/*
- * 'what should we do if we get a hw irq event on an illegal vector'.
- * each architecture has to answer this themselves, it doesn't deserve
- * a generic callback i think.
- */
-void ack_bad_irq(unsigned int irq)
-{
-	printk("received IRQ %d with unknown interrupt type\n", irq);
-}
-
-volatile unsigned long irq_err_count, spurious_count;
-
-/*
- * Generic, controller-independent functions:
- */
-
-int show_interrupts(struct seq_file *p, void *v)
-{
-	int irq = *(loff_t *) v;
-
-	if (irq == 0) {
-		int cpu;
-		seq_puts(p, "           ");
-		for (cpu=0; cpu < 1 /*smp_num_cpus*/; cpu++)
-			seq_printf(p, "CPU%d       ", cpu);
-		seq_putc(p, '\n');
-	}
-
-	if (irq < NR_IRQS) {
-		unsigned long flags;
-		struct irqaction *action;
-
-		spin_lock_irqsave(&irq_desc[irq].lock, flags);
-
-		action = irq_desc[irq].action;
-		if (action) {
-			int j;
-			int count = 0;
-			int num = -1;
-			const char *type_name = irq_desc[irq].chip->typename;
-
-			for (j = 0; j < NR_IRQS; j++)
-				if (irq_desc[j].chip->typename == type_name){
-					if (irq == j)
-						num = count;
-					count++;
-				}
-
-			seq_printf(p, "%3d: ",irq);
-			seq_printf(p, "%10u ", kstat_irqs(irq));
-			if (count > 1) {
-				int prec = (num >= 100 ? 3 : num >= 10 ? 2 : 1);
-				seq_printf(p, " %*s%d", 14 - prec,
-					   type_name, num);
-			} else
-				seq_printf(p, " %14s", type_name);
-		
-			seq_printf(p, "  %s", action->name);
-			for (action=action->next; action; action = action->next)
-				seq_printf(p, ", %s", action->name);
-			seq_putc(p, '\n');
-		}
-
-		spin_unlock_irqrestore(&irq_desc[irq].lock, flags);
-	} else if (irq == NR_IRQS)
-		seq_printf(p, "ERR: %10lu\n", irq_err_count);
-
-	return 0;
-}
-
-/* Handle interrupt IRQ.  REGS are the registers at the time of ther
-   interrupt.  */
-unsigned int handle_irq (int irq, struct pt_regs *regs)
-{
-	irq_enter();
-	__do_IRQ(irq, regs);
-	irq_exit();
-	return 1;
-}
-
-/* Initialize irq handling for IRQs.
-   BASE_IRQ, BASE_IRQ+INTERVAL, ..., BASE_IRQ+NUM*INTERVAL
-   to IRQ_TYPE.  An IRQ_TYPE of 0 means to use a generic interrupt type.  */
-void __init
-init_irq_handlers (int base_irq, int num, int interval,
-		   struct hw_interrupt_type *irq_type)
-{
-	while (num-- > 0) {
-		irq_desc[base_irq].status  = IRQ_DISABLED;
-		irq_desc[base_irq].action  = NULL;
-		irq_desc[base_irq].depth   = 1;
-		irq_desc[base_irq].chip = irq_type;
-		base_irq += interval;
-	}
-}
diff --git a/arch/v850/kernel/ma.c b/arch/v850/kernel/ma.c
deleted file mode 100644
index 143774de75e1..000000000000
--- a/arch/v850/kernel/ma.c
+++ /dev/null
@@ -1,69 +0,0 @@
-/*
- * arch/v850/kernel/ma.c -- V850E/MA series of cpu chips
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/kernel.h>
-#include <linux/init.h>
-#include <linux/mm.h>
-#include <linux/swap.h>
-#include <linux/bootmem.h>
-#include <linux/irq.h>
-
-#include <asm/atomic.h>
-#include <asm/page.h>
-#include <asm/machdep.h>
-#include <asm/v850e_timer_d.h>
-
-#include "mach.h"
-
-void __init mach_sched_init (struct irqaction *timer_action)
-{
-	/* Start hardware timer.  */
-	v850e_timer_d_configure (0, HZ);
-	/* Install timer interrupt handler.  */
-	setup_irq (IRQ_INTCMD(0), timer_action);
-}
-
-static struct v850e_intc_irq_init irq_inits[] = {
-	{ "IRQ", 0, 		NUM_MACH_IRQS,	1, 7 },
-	{ "CMD", IRQ_INTCMD(0), IRQ_INTCMD_NUM,	1, 5 },
-	{ "DMA", IRQ_INTDMA(0), IRQ_INTDMA_NUM,	1, 2 },
-	{ "CSI", IRQ_INTCSI(0), IRQ_INTCSI_NUM,	4, 4 },
-	{ "SER", IRQ_INTSER(0), IRQ_INTSER_NUM,	4, 3 },
-	{ "SR",	 IRQ_INTSR(0),	IRQ_INTSR_NUM, 	4, 4 },
-	{ "ST",  IRQ_INTST(0), 	IRQ_INTST_NUM, 	4, 5 },
-	{ 0 }
-};
-#define NUM_IRQ_INITS (ARRAY_SIZE(irq_inits) - 1)
-
-static struct hw_interrupt_type hw_itypes[NUM_IRQ_INITS];
-
-/* Initialize MA chip interrupts.  */
-void __init ma_init_irqs (void)
-{
-	v850e_intc_init_irq_types (irq_inits, hw_itypes);
-}
-
-/* Called before configuring an on-chip UART.  */
-void ma_uart_pre_configure (unsigned chan, unsigned cflags, unsigned baud)
-{
-	/* We only know about the first two UART channels (though
-	   specific chips may have more).  */
-	if (chan < 2) {
-		unsigned bits = 0x3 << (chan * 3);
-		/* Specify that the relevant pins on the chip should do
-		   serial I/O, not direct I/O.  */
-		MA_PORT4_PMC |= bits;
-		/* Specify that we're using the UART, not the CSI device.  */
-		MA_PORT4_PFC |= bits;
-	}
-}
diff --git a/arch/v850/kernel/mach.c b/arch/v850/kernel/mach.c
deleted file mode 100644
index b9db278d2b71..000000000000
--- a/arch/v850/kernel/mach.c
+++ /dev/null
@@ -1,17 +0,0 @@
-/*
- * arch/v850/kernel/mach.c -- Defaults for some things defined by "mach.h"
- *
- *  Copyright (C) 2001  NEC Corporation
- *  Copyright (C) 2001  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include "mach.h"
-
-/* Called with each timer tick, if non-zero.  */
-void (*mach_tick)(void) = 0;
diff --git a/arch/v850/kernel/mach.h b/arch/v850/kernel/mach.h
deleted file mode 100644
index 9e0e4816ec56..000000000000
--- a/arch/v850/kernel/mach.h
+++ /dev/null
@@ -1,56 +0,0 @@
-/*
- * arch/v850/kernel/mach.h -- Machine-dependent functions used by v850 port
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_MACH_H__
-#define __V850_MACH_H__
-
-#include <linux/kernel.h>
-#include <linux/time.h>
-#include <linux/spinlock.h>
-#include <linux/interrupt.h>
-#include <linux/fs.h>
-#include <linux/seq_file.h>
-
-#include <asm/ptrace.h>
-#include <asm/entry.h>
-#include <asm/clinkage.h>
-
-void mach_setup (char **cmdline);
-void mach_gettimeofday (struct timespec *tv);
-void mach_sched_init (struct irqaction *timer_action);
-void mach_get_physical_ram (unsigned long *ram_start, unsigned long *ram_len);
-void mach_init_irqs (void);
-
-/* If defined, is called very early in the kernel initialization.  The
-   stack pointer is valid, but very little has been initialized (e.g.,
-   bss is not zeroed yet) when this is called, so care must taken.  */
-void mach_early_init (void);
-
-/* If defined, called after the bootmem allocator has been initialized,
-   to allow the platform-dependent code to reserve any areas of RAM that
-   the kernel shouldn't touch.  */
-void mach_reserve_bootmem (void) __attribute__ ((__weak__));
-
-/* Called with each timer tick, if non-zero.  */
-extern void (*mach_tick) (void);
-
-/* The following establishes aliases for various mach_ functions to the
-   name by which the rest of the kernel calls them.  These statements
-   should only have an effect in the file that defines the actual functions. */
-#define MACH_ALIAS(to, from)						      \
-   asm (".global " macrology_stringify (C_SYMBOL_NAME (to)) ";"		      \
-	macrology_stringify (C_SYMBOL_NAME (to))			      \
-	" = " macrology_stringify (C_SYMBOL_NAME (from)))
-/* e.g.: MACH_ALIAS (kernel_name,	arch_spec_name);  */
-
-#endif /* __V850_MACH_H__ */
diff --git a/arch/v850/kernel/me2.c b/arch/v850/kernel/me2.c
deleted file mode 100644
index 007115dc9ce0..000000000000
--- a/arch/v850/kernel/me2.c
+++ /dev/null
@@ -1,73 +0,0 @@
-/*
- * arch/v850/kernel/me2.c -- V850E/ME2 chip-specific support
- *
- *  Copyright (C) 2003  NEC Corporation
- *  Copyright (C) 2003  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/kernel.h>
-#include <linux/init.h>
-#include <linux/mm.h>
-#include <linux/swap.h>
-#include <linux/bootmem.h>
-#include <linux/irq.h>
-
-#include <asm/atomic.h>
-#include <asm/page.h>
-#include <asm/machdep.h>
-#include <asm/v850e_timer_d.h>
-
-#include "mach.h"
-
-void __init mach_sched_init (struct irqaction *timer_action)
-{
-	/* Start hardware timer.  */
-	v850e_timer_d_configure (0, HZ);
-	/* Install timer interrupt handler.  */
-	setup_irq (IRQ_INTCMD(0), timer_action);
-}
-
-static struct v850e_intc_irq_init irq_inits[] = {
-	{ "IRQ",    0,                NUM_CPU_IRQS,      1, 7 },
-	{ "INTP",   IRQ_INTP(0),      IRQ_INTP_NUM,      1, 5 },
-	{ "CMD",    IRQ_INTCMD(0),    IRQ_INTCMD_NUM,    1, 3 },
-	{ "UBTIRE", IRQ_INTUBTIRE(0), IRQ_INTUBTIRE_NUM, 5, 4 },
-	{ "UBTIR",  IRQ_INTUBTIR(0),  IRQ_INTUBTIR_NUM,  5, 4 },
-	{ "UBTIT",  IRQ_INTUBTIT(0),  IRQ_INTUBTIT_NUM,  5, 4 },
-	{ "UBTIF",  IRQ_INTUBTIF(0),  IRQ_INTUBTIF_NUM,  5, 4 },
-	{ "UBTITO", IRQ_INTUBTITO(0), IRQ_INTUBTITO_NUM, 5, 4 },
-	{ 0 }
-};
-#define NUM_IRQ_INITS (ARRAY_SIZE(irq_inits) - 1)
-
-static struct hw_interrupt_type hw_itypes[NUM_IRQ_INITS];
-
-/* Initialize V850E/ME2 chip interrupts.  */
-void __init me2_init_irqs (void)
-{
-	v850e_intc_init_irq_types (irq_inits, hw_itypes);
-}
-
-/* Called before configuring an on-chip UART.  */
-void me2_uart_pre_configure (unsigned chan, unsigned cflags, unsigned baud)
-{
-	if (chan == 0) {
-		/* Specify that the relevant pins on the chip should do
-		   serial I/O, not direct I/O.  */
-		ME2_PORT1_PMC |= 0xC;
-		/* Specify that we're using the UART, not the CSI device. */
-		ME2_PORT1_PFC |= 0xC;
-	} else if (chan == 1) {
-		/* Specify that the relevant pins on the chip should do
-		   serial I/O, not direct I/O.  */
-		ME2_PORT2_PMC |= 0x6;
-		/* Specify that we're using the UART, not the CSI device. */
-		ME2_PORT2_PFC |= 0x6;
-	}
-}
diff --git a/arch/v850/kernel/memcons.c b/arch/v850/kernel/memcons.c
deleted file mode 100644
index 92f514fdcc79..000000000000
--- a/arch/v850/kernel/memcons.c
+++ /dev/null
@@ -1,135 +0,0 @@
-/*
- * arch/v850/kernel/memcons.c -- Console I/O to a memory buffer
- *
- *  Copyright (C) 2001,02  NEC Corporation
- *  Copyright (C) 2001,02  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/kernel.h>
-#include <linux/console.h>
-#include <linux/tty.h>
-#include <linux/tty_driver.h>
-#include <linux/init.h>
-
-/* If this device is enabled, the linker map should define start and
-   end points for its buffer. */
-extern char memcons_output[], memcons_output_end;
-
-/* Current offset into the buffer.  */
-static unsigned long memcons_offs = 0;
-
-/* Spinlock protecting memcons_offs.  */
-static DEFINE_SPINLOCK(memcons_lock);
-
-
-static size_t write (const char *buf, size_t len)
-{
-	unsigned long flags;
-	char *point;
-
-	spin_lock_irqsave (memcons_lock, flags);
-
-	point = memcons_output + memcons_offs;
-	if (point + len >= &memcons_output_end) {
-		len = &memcons_output_end - point;
-		memcons_offs = 0;
-	} else
-		memcons_offs += len;
-
-	spin_unlock_irqrestore (memcons_lock, flags);
-
-	memcpy (point, buf, len);
-
-	return len;
-}
-
-
-/*  Low-level console. */
-
-static void memcons_write (struct console *co, const char *buf, unsigned len)
-{
-	while (len > 0)
-		len -= write (buf, len);
-}
-
-static struct tty_driver *tty_driver;
-
-static struct tty_driver *memcons_device (struct console *co, int *index)
-{
-	*index = co->index;
-	return tty_driver;
-}
-
-static struct console memcons =
-{
-    .name	= "memcons",
-    .write	= memcons_write,
-    .device	= memcons_device,
-    .flags	= CON_PRINTBUFFER,
-    .index	= -1,
-};
-
-void memcons_setup (void)
-{
-	register_console (&memcons);
-	printk (KERN_INFO "Console: static memory buffer (memcons)\n");
-}
-
-/* Higher level TTY interface.  */
-
-int memcons_tty_open (struct tty_struct *tty, struct file *filp)
-{
-	return 0;
-}
-
-int memcons_tty_write (struct tty_struct *tty, const unsigned char *buf, int len)
-{
-	return write (buf, len);
-}
-
-int memcons_tty_write_room (struct tty_struct *tty)
-{
-	return &memcons_output_end - (memcons_output + memcons_offs);
-}
-
-int memcons_tty_chars_in_buffer (struct tty_struct *tty)
-{
-	/* We have no buffer.  */
-	return 0;
-}
-
-static const struct tty_operations ops = {
-	.open = memcons_tty_open,
-	.write = memcons_tty_write,
-	.write_room = memcons_tty_write_room,
-	.chars_in_buffer = memcons_tty_chars_in_buffer,
-};
-
-int __init memcons_tty_init (void)
-{
-	int err;
-	struct tty_driver *driver = alloc_tty_driver(1);
-	if (!driver)
-		return -ENOMEM;
-
-	driver->name = "memcons";
-	driver->major = TTY_MAJOR;
-	driver->minor_start = 64;
-	driver->type = TTY_DRIVER_TYPE_SYSCONS;
-	driver->init_termios = tty_std_termios;
-	tty_set_operations(driver, &ops);
-	err = tty_register_driver(driver);
-	if (err) {
-		put_tty_driver(driver);
-		return err;
-	}
-	tty_driver = driver;
-	return 0;
-}
-__initcall (memcons_tty_init);
diff --git a/arch/v850/kernel/module.c b/arch/v850/kernel/module.c
deleted file mode 100644
index 64aeb3e37c52..000000000000
--- a/arch/v850/kernel/module.c
+++ /dev/null
@@ -1,237 +0,0 @@
-/*
- * arch/v850/kernel/module.c -- Architecture-specific module functions
- *
- *  Copyright (C) 2002,03  NEC Electronics Corporation
- *  Copyright (C) 2002,03  Miles Bader <miles@gnu.org>
- *  Copyright (C) 2001,03  Rusty Russell
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- *
- * Derived in part from arch/ppc/kernel/module.c
- */
-
-#include <linux/kernel.h>
-#include <linux/vmalloc.h>
-#include <linux/moduleloader.h>
-#include <linux/elf.h>
-
-#if 0
-#define DEBUGP printk
-#else
-#define DEBUGP(fmt , ...)
-#endif
-
-void *module_alloc (unsigned long size)
-{
-	return size == 0 ? 0 : vmalloc (size);
-}
-
-void module_free (struct module *mod, void *module_region)
-{
-	vfree (module_region);
-	/* FIXME: If module_region == mod->init_region, trim exception
-           table entries. */
-}
-
-int module_finalize (const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs,
-		     struct module *mod)
-{
-	return 0;
-}
-
-/* Count how many different relocations (different symbol, different
-   addend) */
-static unsigned int count_relocs(const Elf32_Rela *rela, unsigned int num)
-{
-	unsigned int i, j, ret = 0;
-
-	/* Sure, this is order(n^2), but it's usually short, and not
-           time critical */
-	for (i = 0; i < num; i++) {
-		for (j = 0; j < i; j++) {
-			/* If this addend appeared before, it's
-                           already been counted */
-			if (ELF32_R_SYM(rela[i].r_info)
-			    == ELF32_R_SYM(rela[j].r_info)
-			    && rela[i].r_addend == rela[j].r_addend)
-				break;
-		}
-		if (j == i) ret++;
-	}
-	return ret;
-}
-
-/* Get the potential trampolines size required of the init and
-   non-init sections */
-static unsigned long get_plt_size(const Elf32_Ehdr *hdr,
-				  const Elf32_Shdr *sechdrs,
-				  const char *secstrings,
-				  int is_init)
-{
-	unsigned long ret = 0;
-	unsigned i;
-
-	/* Everything marked ALLOC (this includes the exported
-           symbols) */
-	for (i = 1; i < hdr->e_shnum; i++) {
-		/* If it's called *.init*, and we're not init, we're
-                   not interested */
-		if ((strstr(secstrings + sechdrs[i].sh_name, ".init") != 0)
-		    != is_init)
-			continue;
-
-		if (sechdrs[i].sh_type == SHT_RELA) {
-			DEBUGP("Found relocations in section %u\n", i);
-			DEBUGP("Ptr: %p.  Number: %u\n",
-			       (void *)hdr + sechdrs[i].sh_offset,
-			       sechdrs[i].sh_size / sizeof(Elf32_Rela));
-			ret += count_relocs((void *)hdr
-					     + sechdrs[i].sh_offset,
-					     sechdrs[i].sh_size
-					     / sizeof(Elf32_Rela))
-				* sizeof(struct v850_plt_entry);
-		}
-	}
-
-	return ret;
-}
-
-int module_frob_arch_sections(Elf32_Ehdr *hdr,
-			      Elf32_Shdr *sechdrs,
-			      char *secstrings,
-			      struct module *me)
-{
-	unsigned int i;
-
-	/* Find .plt and .pltinit sections */
-	for (i = 0; i < hdr->e_shnum; i++) {
-		if (strcmp(secstrings + sechdrs[i].sh_name, ".init.plt") == 0)
-			me->arch.init_plt_section = i;
-		else if (strcmp(secstrings + sechdrs[i].sh_name, ".plt") == 0)
-			me->arch.core_plt_section = i;
-	}
-	if (!me->arch.core_plt_section || !me->arch.init_plt_section) {
-		printk("Module doesn't contain .plt or .plt.init sections.\n");
-		return -ENOEXEC;
-	}
-
-	/* Override their sizes */
-	sechdrs[me->arch.core_plt_section].sh_size
-		= get_plt_size(hdr, sechdrs, secstrings, 0);
-	sechdrs[me->arch.init_plt_section].sh_size
-		= get_plt_size(hdr, sechdrs, secstrings, 1);
-	return 0;
-}
-
-int apply_relocate (Elf32_Shdr *sechdrs, const char *strtab,
-		    unsigned int symindex, unsigned int relsec,
-		    struct module *mod)
-{
-	printk ("Barf\n");
-	return -ENOEXEC;
-}
-
-/* Set up a trampoline in the PLT to bounce us to the distant function */
-static uint32_t do_plt_call (void *location, Elf32_Addr val,
-			     Elf32_Shdr *sechdrs, struct module *mod)
-{
-	struct v850_plt_entry *entry;
-	/* Instructions used to do the indirect jump.  */
-	uint32_t tramp[2];
-
-	/* We have to trash a register, so we assume that any control
-	   transfer more than 21-bits away must be a function call
-	   (so we can use a call-clobbered register).  */
-	tramp[0] = 0x0621 + ((val & 0xffff) << 16);   /* mov sym, r1 ... */
-	tramp[1] = ((val >> 16) & 0xffff) + 0x610000; /* ...; jmp r1 */
-
-	/* Init, or core PLT? */
-	if (location >= mod->module_core
-	    && location < mod->module_core + mod->core_size)
-		entry = (void *)sechdrs[mod->arch.core_plt_section].sh_addr;
-	else
-		entry = (void *)sechdrs[mod->arch.init_plt_section].sh_addr;
-
-	/* Find this entry, or if that fails, the next avail. entry */
-	while (entry->tramp[0])
-		if (entry->tramp[0] == tramp[0] && entry->tramp[1] == tramp[1])
-			return (uint32_t)entry;
-		else
-			entry++;
-
-	entry->tramp[0] = tramp[0];
-	entry->tramp[1] = tramp[1];
-
-	return (uint32_t)entry;
-}
-
-int apply_relocate_add (Elf32_Shdr *sechdrs, const char *strtab,
-			unsigned int symindex, unsigned int relsec,
-			struct module *mod)
-{
-	unsigned int i;
-	Elf32_Rela *rela = (void *)sechdrs[relsec].sh_addr;
-
-	DEBUGP ("Applying relocate section %u to %u\n", relsec,
-		sechdrs[relsec].sh_info);
-
-	for (i = 0; i < sechdrs[relsec].sh_size / sizeof (*rela); i++) {
-		/* This is where to make the change */
-		uint32_t *loc
-			= ((void *)sechdrs[sechdrs[relsec].sh_info].sh_addr
-			   + rela[i].r_offset);
-		/* This is the symbol it is referring to.  Note that all
-		   undefined symbols have been resolved.  */
-		Elf32_Sym *sym
-			= ((Elf32_Sym *)sechdrs[symindex].sh_addr
-			   + ELF32_R_SYM (rela[i].r_info));
-		uint32_t val = sym->st_value + rela[i].r_addend;
-
-		switch (ELF32_R_TYPE (rela[i].r_info)) {
-		case R_V850_32:
-			/* We write two shorts instead of a long because even
-			   32-bit insns only need half-word alignment, but
-			   32-bit data writes need to be long-word aligned.  */
-			val += ((uint16_t *)loc)[0];
-			val += ((uint16_t *)loc)[1] << 16;
-			((uint16_t *)loc)[0] = val & 0xffff;
-			((uint16_t *)loc)[1] = (val >> 16) & 0xffff;
-			break;
-
-		case R_V850_22_PCREL:
-			/* Maybe jump indirectly via a PLT table entry.  */
-			if ((int32_t)(val - (uint32_t)loc) > 0x1fffff
-			    || (int32_t)(val - (uint32_t)loc) < -0x200000)
-				val = do_plt_call (loc, val, sechdrs, mod);
-
-			val -= (uint32_t)loc;
-
-			/* We write two shorts instead of a long because
-			   even 32-bit insns only need half-word alignment,
-			   but 32-bit data writes need to be long-word
-			   aligned.  */
-			((uint16_t *)loc)[0] =
-				(*(uint16_t *)loc & 0xffc0) /* opcode + reg */
-				| ((val >> 16) & 0xffc03f); /* offs high */
-			((uint16_t *)loc)[1] =
-				(val & 0xffff);		    /* offs low */
-			break;
-
-		default:
-			printk (KERN_ERR "module %s: Unknown reloc: %u\n",
-				mod->name, ELF32_R_TYPE (rela[i].r_info));
-			return -ENOEXEC;
-		}
-	}
-
-	return 0;
-}
-
-void
-module_arch_cleanup(struct module *mod)
-{
-}
diff --git a/arch/v850/kernel/process.c b/arch/v850/kernel/process.c
deleted file mode 100644
index e4a4b8e7d5a3..000000000000
--- a/arch/v850/kernel/process.c
+++ /dev/null
@@ -1,217 +0,0 @@
-/*
- * arch/v850/kernel/process.c -- Arch-dependent process handling
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/errno.h>
-#include <linux/sched.h>
-#include <linux/kernel.h>
-#include <linux/mm.h>
-#include <linux/smp.h>
-#include <linux/stddef.h>
-#include <linux/unistd.h>
-#include <linux/ptrace.h>
-#include <linux/slab.h>
-#include <linux/user.h>
-#include <linux/a.out.h>
-#include <linux/reboot.h>
-
-#include <asm/uaccess.h>
-#include <asm/system.h>
-#include <asm/pgtable.h>
-
-void (*pm_power_off)(void) = NULL;
-EXPORT_SYMBOL(pm_power_off);
-
-extern void ret_from_fork (void);
-
-
-/* The idle loop.  */
-static void default_idle (void)
-{
-	while (! need_resched ())
-		asm ("halt; nop; nop; nop; nop; nop" ::: "cc");
-}
-
-void (*idle)(void) = default_idle;
-
-/*
- * The idle thread. There's no useful work to be
- * done, so just try to conserve power and have a
- * low exit latency (ie sit in a loop waiting for
- * somebody to say that they'd like to reschedule)
- */
-void cpu_idle (void)
-{
-	/* endless idle loop with no priority at all */
-	while (1) {
-		while (!need_resched())
-			(*idle) ();
-
-		preempt_enable_no_resched();
-		schedule();
-		preempt_disable();
-	}
-}
-
-/*
- * This is the mechanism for creating a new kernel thread.
- *
- * NOTE! Only a kernel-only process (ie the swapper or direct descendants who
- * haven't done an "execve()") should use this: it will work within a system
- * call from a "real" process, but the process memory space will not be free'd
- * until both the parent and the child have exited.
- */
-int kernel_thread (int (*fn)(void *), void *arg, unsigned long flags)
-{
-	register mm_segment_t fs = get_fs ();
-	register unsigned long syscall asm (SYSCALL_NUM);
-	register unsigned long arg0 asm (SYSCALL_ARG0);
-	register unsigned long ret asm (SYSCALL_RET);
-
-	set_fs (KERNEL_DS);
-
-	/* Clone this thread.  Note that we don't pass the clone syscall's
-	   second argument -- it's ignored for calls from kernel mode (the
-	   child's SP is always set to the top of the kernel stack).  */
-	arg0 = flags | CLONE_VM;
-	syscall = __NR_clone;
-	asm volatile ("trap " SYSCALL_SHORT_TRAP
-		      : "=r" (ret), "=r" (syscall)
-		      : "1" (syscall), "r" (arg0)
-		      : SYSCALL_SHORT_CLOBBERS);
-
-	if (ret == 0) {
-		/* In child thread, call FN and exit.  */
-		arg0 = (*fn) (arg);
-		syscall = __NR_exit;
-		asm volatile ("trap " SYSCALL_SHORT_TRAP
-			      : "=r" (ret), "=r" (syscall)
-			      : "1" (syscall), "r" (arg0)
-			      : SYSCALL_SHORT_CLOBBERS);
-	}
-
-	/* In parent.  */
-	set_fs (fs);
-
-	return ret;
-}
-
-void flush_thread (void)
-{
-	set_fs (USER_DS);
-}
-
-int copy_thread (int nr, unsigned long clone_flags,
-		 unsigned long stack_start, unsigned long stack_size,
-		 struct task_struct *p, struct pt_regs *regs)
-{
-	/* Start pushing stuff from the top of the child's kernel stack.  */
-	unsigned long orig_ksp = task_tos(p);
-	unsigned long ksp = orig_ksp;
-	/* We push two `state save' stack fames (see entry.S) on the new
-	   kernel stack:
-	     1) The innermost one is what switch_thread would have
-	        pushed, and is used when we context switch to the child
-		thread for the first time.  It's set up to return to
-		ret_from_fork in entry.S.
-	     2) The outermost one (nearest the top) is what a syscall
-	        trap would have pushed, and is set up to return to the
-		same location as the parent thread, but with a return
-		value of 0. */
-	struct pt_regs *child_switch_regs, *child_trap_regs;
-
-	/* Trap frame.  */
-	ksp -= STATE_SAVE_SIZE;
-	child_trap_regs = (struct pt_regs *)(ksp + STATE_SAVE_PT_OFFSET);
-	/* Switch frame.  */
-	ksp -= STATE_SAVE_SIZE;
-	child_switch_regs = (struct pt_regs *)(ksp + STATE_SAVE_PT_OFFSET);
-
-	/* First copy parent's register state to child.  */
-	*child_switch_regs = *regs;
-	*child_trap_regs = *regs;
-
-	/* switch_thread returns to the restored value of the lp
-	   register (r31), so we make that the place where we want to
-	   jump when the child thread begins running.  */
-	child_switch_regs->gpr[GPR_LP] = (v850_reg_t)ret_from_fork;
-
-	if (regs->kernel_mode)
-		/* Since we're returning to kernel-mode, make sure the child's
-		   stored kernel stack pointer agrees with what the actual
-		   stack pointer will be at that point (the trap return code
-		   always restores the SP, even when returning to
-		   kernel-mode).  */
-		child_trap_regs->gpr[GPR_SP] = orig_ksp;
-	else
-		/* Set the child's user-mode stack-pointer (the name
-		   `stack_start' is a misnomer, it's just the initial SP
-		   value).  */
-		child_trap_regs->gpr[GPR_SP] = stack_start;
-
-	/* Thread state for the child (everything else is on the stack).  */
-	p->thread.ksp = ksp;
-
-	return 0;
-}
-
-/*
- * sys_execve() executes a new program.
- */
-int sys_execve (char *name, char **argv, char **envp, struct pt_regs *regs)
-{
-	char *filename = getname (name);
-	int error = PTR_ERR (filename);
-
-	if (! IS_ERR (filename)) {
-		error = do_execve (filename, argv, envp, regs);
-		putname (filename);
-	}
-
-	return error;
-}
-
-
-/*
- * These bracket the sleeping functions..
- */
-#define first_sched	((unsigned long)__sched_text_start)
-#define last_sched	((unsigned long)__sched_text_end)
-
-unsigned long get_wchan (struct task_struct *p)
-{
-#if 0  /* Barf.  Figure out the stack-layout later.  XXX  */
-	unsigned long fp, pc;
-	int count = 0;
-
-	if (!p || p == current || p->state == TASK_RUNNING)
-		return 0;
-
-	pc = thread_saved_pc (p);
-
-	/* This quite disgusting function walks up the stack, following
-	   saved return address, until it something that's out of bounds
-	   (as defined by `first_sched' and `last_sched').  It then
-	   returns the last PC that was in-bounds.  */
-	do {
-		if (fp < stack_page + sizeof (struct task_struct) ||
-		    fp >= 8184+stack_page)
-			return 0;
-		pc = ((unsigned long *)fp)[1];
-		if (pc < first_sched || pc >= last_sched)
-			return pc;
-		fp = *(unsigned long *) fp;
-	} while (count++ < 16);
-#endif
-
-	return 0;
-}
diff --git a/arch/v850/kernel/procfs.c b/arch/v850/kernel/procfs.c
deleted file mode 100644
index e433cde789b4..000000000000
--- a/arch/v850/kernel/procfs.c
+++ /dev/null
@@ -1,67 +0,0 @@
-/*
- * arch/v850/kernel/procfs.c -- Introspection functions for /proc filesystem
- *
- *  Copyright (C) 2001,02  NEC Corporation
- *  Copyright (C) 2001,02  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include "mach.h"
-
-static int cpuinfo_print (struct seq_file *m, void *v)
-{
-	extern unsigned long loops_per_jiffy;
-	
-	seq_printf (m, "CPU-Family:	v850\nCPU-Arch:	%s\n", CPU_ARCH);
-
-#ifdef CPU_MODEL_LONG
-	seq_printf (m, "CPU-Model:	%s (%s)\n", CPU_MODEL, CPU_MODEL_LONG);
-#else
-	seq_printf (m, "CPU-Model:	%s\n", CPU_MODEL);
-#endif
-
-#ifdef CPU_CLOCK_FREQ
-	seq_printf (m, "CPU-Clock:	%ld (%ld MHz)\n",
-		    (long)CPU_CLOCK_FREQ,
-		    (long)CPU_CLOCK_FREQ / 1000000);
-#endif
-
-	seq_printf (m, "BogoMips:	%lu.%02lu\n",
-		    loops_per_jiffy/(500000/HZ),
-		    (loops_per_jiffy/(5000/HZ)) % 100);
-
-#ifdef PLATFORM_LONG
-	seq_printf (m, "Platform:	%s (%s)\n", PLATFORM, PLATFORM_LONG);
-#elif defined (PLATFORM)
-	seq_printf (m, "Platform:	%s\n", PLATFORM);
-#endif
-
-	return 0;
-}
-
-static void *cpuinfo_start (struct seq_file *m, loff_t *pos)
-{
-	return *pos < NR_CPUS ? ((void *) 0x12345678) : NULL;
-}
-
-static void *cpuinfo_next (struct seq_file *m, void *v, loff_t *pos)
-{
-	++*pos;
-	return cpuinfo_start (m, pos);
-}
-
-static void cpuinfo_stop (struct seq_file *m, void *v)
-{
-}
-
-const struct seq_operations cpuinfo_op = {
-	.start	= cpuinfo_start,
-	.next	= cpuinfo_next,
-	.stop	= cpuinfo_stop,
-	.show	= cpuinfo_print
-};
diff --git a/arch/v850/kernel/ptrace.c b/arch/v850/kernel/ptrace.c
deleted file mode 100644
index a458ac941b25..000000000000
--- a/arch/v850/kernel/ptrace.c
+++ /dev/null
@@ -1,235 +0,0 @@
-/*
- * arch/v850/kernel/ptrace.c -- `ptrace' system call
- *
- *  Copyright (C) 2002,03,04  NEC Electronics Corporation
- *  Copyright (C) 2002,03,04  Miles Bader <miles@gnu.org>
- *
- * Derived from arch/mips/kernel/ptrace.c:
- *
- *  Copyright (C) 1992 Ross Biro
- *  Copyright (C) Linus Torvalds
- *  Copyright (C) 1994, 95, 96, 97, 98, 2000 Ralf Baechle
- *  Copyright (C) 1996 David S. Miller
- *  Kevin D. Kissell, kevink@mips.com and Carsten Langgaard, carstenl@mips.com
- *  Copyright (C) 1999 MIPS Technologies, Inc.
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- */
-
-#include <linux/kernel.h>
-#include <linux/mm.h>
-#include <linux/sched.h>
-#include <linux/ptrace.h>
-#include <linux/signal.h>
-
-#include <asm/errno.h>
-#include <asm/ptrace.h>
-#include <asm/processor.h>
-#include <asm/uaccess.h>
-
-/* Returns the address where the register at REG_OFFS in P is stashed away.  */
-static v850_reg_t *reg_save_addr (unsigned reg_offs, struct task_struct *t)
-{
-	struct pt_regs *regs;
-
-	/* Three basic cases:
-
-	   (1) A register normally saved before calling the scheduler, is
-	       available in the kernel entry pt_regs structure at the top
-	       of the kernel stack.  The kernel trap/irq exit path takes
-	       care to save/restore almost all registers for ptrace'd
-	       processes.
-
-	   (2) A call-clobbered register, where the process P entered the
-	       kernel via [syscall] trap, is not stored anywhere; that's
-	       OK, because such registers are not expected to be preserved
-	       when the trap returns anyway (so we don't actually bother to
-	       test for this case).
-
-	   (3) A few registers not used at all by the kernel, and so
-	       normally never saved except by context-switches, are in the
-	       context switch state.  */
-
-	if (reg_offs == PT_CTPC || reg_offs == PT_CTPSW || reg_offs == PT_CTBP)
-		/* Register saved during context switch.  */
-		regs = thread_saved_regs (t);
-	else
-		/* Register saved during kernel entry (or not available).  */
-		regs = task_pt_regs (t);
-
-	return (v850_reg_t *)((char *)regs + reg_offs);
-}
-
-/* Set the bits SET and clear the bits CLEAR in the v850e DIR
-   (`debug information register').  Returns the new value of DIR.  */
-static inline v850_reg_t set_dir (v850_reg_t set, v850_reg_t clear)
-{
-	register v850_reg_t rval asm ("r10");
-	register v850_reg_t arg0 asm ("r6") = set;
-	register v850_reg_t arg1 asm ("r7") = clear;
-
-	/* The dbtrap handler has exactly this functionality when called
-	   from kernel mode.  0xf840 is a `dbtrap' insn.  */
-	asm (".short 0xf840" : "=r" (rval) : "r" (arg0), "r" (arg1));
-
-	return rval;
-}
-
-/* Makes sure hardware single-stepping is (globally) enabled.
-   Returns true if successful.  */
-static inline int enable_single_stepping (void)
-{
-	static int enabled = 0;	/* Remember whether we already did it.  */
-	if (! enabled) {
-		/* Turn on the SE (`single-step enable') bit, 0x100, in the
-		   DIR (`debug information register').  This may fail if a
-		   processor doesn't support it or something.  We also try
-		   to clear bit 0x40 (`INI'), which is necessary to use the
-		   debug stuff on the v850e2; on the v850e, clearing 0x40
-		   shouldn't cause any problem.  */
-		v850_reg_t dir = set_dir (0x100, 0x40);
-		/* Make sure it really got set.  */
-		if (dir & 0x100)
-			enabled = 1;
-	}
-	return enabled;
-}
-
-/* Try to set CHILD's single-step flag to VAL.  Returns true if successful.  */
-static int set_single_step (struct task_struct *t, int val)
-{
-	v850_reg_t *psw_addr = reg_save_addr(PT_PSW, t);
-	if (val) {
-		/* Make sure single-stepping is enabled.  */
-		if (! enable_single_stepping ())
-			return 0;
-		/* Set T's single-step flag.  */
-		*psw_addr |= 0x800;
-	} else
-		*psw_addr &= ~0x800;
-	return 1;
-}
-
-long arch_ptrace(struct task_struct *child, long request, long addr, long data)
-{
-	int rval;
-
-	switch (request) {
-		unsigned long val;
-
-	case PTRACE_PEEKTEXT: /* read word at location addr. */
-	case PTRACE_PEEKDATA:
-		rval = generic_ptrace_peekdata(child, addr, data);
-		goto out;
-
-	case PTRACE_POKETEXT: /* write the word at location addr. */
-	case PTRACE_POKEDATA:
-		rval = generic_ptrace_pokedata(child, addr, data);
-		goto out;
-
-	/* Read/write the word at location ADDR in the registers.  */
-	case PTRACE_PEEKUSR:
-	case PTRACE_POKEUSR:
-		rval = 0;
-		if (addr >= PT_SIZE && request == PTRACE_PEEKUSR) {
-			/* Special requests that don't actually correspond
-			   to offsets in struct pt_regs.  */
-			if (addr == PT_TEXT_ADDR)
-				val = child->mm->start_code;
-			else if (addr == PT_DATA_ADDR)
-				val = child->mm->start_data;
-			else if (addr == PT_TEXT_LEN)
-				val = child->mm->end_code
-					- child->mm->start_code;
-			else
-				rval = -EIO;
-		} else if (addr >= 0 && addr < PT_SIZE && (addr & 0x3) == 0) {
-			v850_reg_t *reg_addr = reg_save_addr(addr, child);
-			if (request == PTRACE_PEEKUSR)
-				val = *reg_addr;
-			else
-				*reg_addr = data;
-		} else
-			rval = -EIO;
-
-		if (rval == 0 && request == PTRACE_PEEKUSR)
-			rval = put_user (val, (unsigned long *)data);
-		goto out;
-
-	/* Continue and stop at next (return from) syscall */
-	case PTRACE_SYSCALL:
-	/* Restart after a signal.  */
-	case PTRACE_CONT:
-	/* Execute a single instruction. */
-	case PTRACE_SINGLESTEP:
-		rval = -EIO;
-		if (!valid_signal(data))
-			break;
-
-		/* Turn CHILD's single-step flag on or off.  */
-		if (! set_single_step (child, request == PTRACE_SINGLESTEP))
-			break;
-
-		if (request == PTRACE_SYSCALL)
-			set_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
-		else
-			clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
-
-		child->exit_code = data;
-		wake_up_process(child);
-		rval = 0;
-		break;
-
-	/*
-	 * make the child exit.  Best I can do is send it a sigkill.
-	 * perhaps it should be put in the status that it wants to
-	 * exit.
-	 */
-	case PTRACE_KILL:
-		rval = 0;
-		if (child->exit_state == EXIT_ZOMBIE)	/* already dead */
-			break;
-		child->exit_code = SIGKILL;
-		wake_up_process(child);
-		break;
-
-	case PTRACE_DETACH: /* detach a process that was attached. */
-		set_single_step (child, 0);  /* Clear single-step flag */
-		rval = ptrace_detach(child, data);
-		break;
-
-	default:
-		rval = -EIO;
-		goto out;
-	}
- out:
-	return rval;
-}
-
-asmlinkage void syscall_trace(void)
-{
-	if (!test_thread_flag(TIF_SYSCALL_TRACE))
-		return;
-	if (!(current->ptrace & PT_PTRACED))
-		return;
-	/* The 0x80 provides a way for the tracing parent to distinguish
-	   between a syscall stop and SIGTRAP delivery */
-	ptrace_notify(SIGTRAP | ((current->ptrace & PT_TRACESYSGOOD)
-				 ? 0x80 : 0));
-	/*
-	 * this isn't the same as continuing with a signal, but it will do
-	 * for normal use.  strace only continues with a signal if the
-	 * stopping signal is not SIGTRAP.  -brl
-	 */
-	if (current->exit_code) {
-		send_sig(current->exit_code, current, 1);
-		current->exit_code = 0;
-	}
-}
-
-void ptrace_disable (struct task_struct *child)
-{
-	/* nothing to do */
-}
diff --git a/arch/v850/kernel/rte_cb.c b/arch/v850/kernel/rte_cb.c
deleted file mode 100644
index 43018e1edebd..000000000000
--- a/arch/v850/kernel/rte_cb.c
+++ /dev/null
@@ -1,193 +0,0 @@
-/*
- * include/asm-v850/rte_cb.c -- Midas lab RTE-CB series of evaluation boards
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/init.h>
-#include <linux/irq.h>
-#include <linux/fs.h>
-#include <linux/module.h>
-#include <linux/kernel.h>
-
-#include <asm/machdep.h>
-#include <asm/v850e_uart.h>
-
-#include "mach.h"
-
-static void led_tick (void);
-
-/* LED access routines.  */
-extern unsigned read_leds (int pos, char *buf, int len);
-extern unsigned write_leds (int pos, const char *buf, int len);
-
-#ifdef CONFIG_RTE_CB_MULTI
-extern void multi_init (void);
-#endif
-
-
-void __init rte_cb_early_init (void)
-{
-	v850e_intc_disable_irqs ();
-
-#ifdef CONFIG_RTE_CB_MULTI
-	multi_init ();
-#endif
-}
-
-void __init mach_setup (char **cmdline)
-{
-#ifdef CONFIG_RTE_MB_A_PCI
-	/* Probe for Mother-A, and print a message if we find it.  */
-	*(volatile unsigned long *)MB_A_SRAM_ADDR = 0xDEADBEEF;
-	if (*(volatile unsigned long *)MB_A_SRAM_ADDR == 0xDEADBEEF) {
-		*(volatile unsigned long *)MB_A_SRAM_ADDR = 0x12345678;
-		if (*(volatile unsigned long *)MB_A_SRAM_ADDR == 0x12345678)
-			printk (KERN_INFO
-				"          NEC SolutionGear/Midas lab"
-				" RTE-MOTHER-A motherboard\n");
-	}
-#endif /* CONFIG_RTE_MB_A_PCI */
-
-	mach_tick = led_tick;
-}
-
-void machine_restart (char *__unused)
-{
-#ifdef CONFIG_RESET_GUARD
-	disable_reset_guard ();
-#endif
-	asm ("jmp r0"); /* Jump to the reset vector.  */
-}
-
-/* This says `HALt.' in LEDese.  */
-static unsigned char halt_leds_msg[] = { 0x76, 0x77, 0x38, 0xF8 };
-
-void machine_halt (void)
-{
-#ifdef CONFIG_RESET_GUARD
-	disable_reset_guard ();
-#endif
-
-	/* Ignore all interrupts.  */
-	local_irq_disable ();
-
-	/* Write a little message.  */
-	write_leds (0, halt_leds_msg, sizeof halt_leds_msg);
-
-	/* Really halt.  */
-	for (;;)
-		asm ("halt; nop; nop; nop; nop; nop");
-}
-
-void machine_power_off (void)
-{
-	machine_halt ();
-}
-
-
-/* Animated LED display for timer tick.  */
-
-#define TICK_UPD_FREQ	6
-static int tick_frames[][10] = {
-	{ 0x01, 0x02, 0x04, 0x08, 0x10, 0x20, -1 },
-	{ 0x63, 0x5c, -1 },
-	{ 0x5c, 0x00, -1 },
-	{ 0x63, 0x00, -1 },
-	{ -1 }
-};
-
-static void led_tick ()
-{
-	static unsigned counter = 0;
-	
-	if (++counter == (HZ / TICK_UPD_FREQ)) {
-		/* Which frame we're currently displaying for each digit.  */
-		static unsigned frame_nums[LED_NUM_DIGITS] = { 0 };
-		/* Display image.  */
-		static unsigned char image[LED_NUM_DIGITS] = { 0 };
-		unsigned char prev_image[LED_NUM_DIGITS];
-		int write_to_leds = 1; /* true if we should actually display */
-		int digit;
-
-		/* We check to see if the physical LEDs contains what we last
-		   wrote to them; if not, we suppress display (this is so that
-		   users can write to the LEDs, and not have their output
-		   overwritten).  As a special case, we start writing again if
-		   all the LEDs are blank, or our display image is all zeros
-		   (indicating that this is the initial update, when the actual
-		   LEDs might contain random data).  */
-		read_leds (0, prev_image, LED_NUM_DIGITS);
-		for (digit = 0; digit < LED_NUM_DIGITS; digit++)
-			if (image[digit] != prev_image[digit]
-			    && image[digit] && prev_image[digit])
-			{
-				write_to_leds = 0;
-				break;
-			}
-
-		/* Update display image.  */
-		for (digit = 0;
-		     digit < LED_NUM_DIGITS && tick_frames[digit][0] >= 0;
-		     digit++)
-		{
-			int frame = tick_frames[digit][frame_nums[digit]];
-			if (frame < 0) {
-				image[digit] = tick_frames[digit][0];
-				frame_nums[digit] = 1;
-			} else {
-				image[digit] = frame;
-				frame_nums[digit]++;
-				break;
-			}
-		}
-
-		if (write_to_leds)
-			/* Write the display image to the physical LEDs.  */
-			write_leds (0, image, LED_NUM_DIGITS);
-
-		counter = 0;
-	}
-}
-
-
-/* Mother-A interrupts.  */
-
-#ifdef CONFIG_RTE_GBUS_INT
-
-#define L GBUS_INT_PRIORITY_LOW
-#define M GBUS_INT_PRIORITY_MEDIUM
-#define H GBUS_INT_PRIORITY_HIGH
-
-static struct gbus_int_irq_init gbus_irq_inits[] = {
-#ifdef CONFIG_RTE_MB_A_PCI
-	{ "MB_A_LAN",	IRQ_MB_A_LAN,		1,		     1, L },
-	{ "MB_A_PCI1",	IRQ_MB_A_PCI1(0),	IRQ_MB_A_PCI1_NUM,   1, L },
-	{ "MB_A_PCI2",	IRQ_MB_A_PCI2(0),	IRQ_MB_A_PCI2_NUM,   1, L },
-	{ "MB_A_EXT",	IRQ_MB_A_EXT(0),	IRQ_MB_A_EXT_NUM,    1, L },
-	{ "MB_A_USB_OC",IRQ_MB_A_USB_OC(0),	IRQ_MB_A_USB_OC_NUM, 1, L },
-	{ "MB_A_PCMCIA_OC",IRQ_MB_A_PCMCIA_OC,	1,		     1, L },
-#endif
-	{ 0 }
-};
-#define NUM_GBUS_IRQ_INITS (ARRAY_SIZE(gbus_irq_inits) - 1)
-
-static struct hw_interrupt_type gbus_hw_itypes[NUM_GBUS_IRQ_INITS];
-
-#endif /* CONFIG_RTE_GBUS_INT */
-
-
-void __init rte_cb_init_irqs (void)
-{
-#ifdef CONFIG_RTE_GBUS_INT
-	gbus_int_init_irqs ();
-	gbus_int_init_irq_types (gbus_irq_inits, gbus_hw_itypes);
-#endif /* CONFIG_RTE_GBUS_INT */
-}
diff --git a/arch/v850/kernel/rte_cb_leds.c b/arch/v850/kernel/rte_cb_leds.c
deleted file mode 100644
index aa47ab1dcd87..000000000000
--- a/arch/v850/kernel/rte_cb_leds.c
+++ /dev/null
@@ -1,137 +0,0 @@
-/*
- * include/asm-v850/rte_cb_leds.c -- Midas lab RTE-CB board LED device support
- *
- *  Copyright (C) 2002,03  NEC Electronics Corporation
- *  Copyright (C) 2002,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/init.h>
-#include <linux/spinlock.h>
-#include <linux/fs.h>
-#include <linux/miscdevice.h>
-
-#include <asm/uaccess.h>
-
-#define LEDS_MINOR	169	/* Minor device number, using misc major.  */
-
-/* The actual LED hardware is write-only, so we hold the contents here too.  */
-static unsigned char leds_image[LED_NUM_DIGITS] = { 0 };
-
-/* Spinlock protecting the above leds.  */
-static DEFINE_SPINLOCK(leds_lock);
-
-/* Common body of LED read/write functions, checks POS and LEN for
-   correctness, declares a variable using IMG_DECL, initialized pointing at
-   the POS position in the LED image buffer, and and iterates COPY_EXPR
-   until BUF is equal to the last buffer position; finally, sets LEN to be
-   the amount actually copied.  IMG should be a variable declaration
-   (without an initializer or a terminating semicolon); POS, BUF, and LEN
-   should all be simple variables.  */
-#define DO_LED_COPY(img_decl, pos, buf, len, copy_expr)			\
-do {									\
-	if (pos > LED_NUM_DIGITS)					\
-		len = 0;						\
-	else {								\
-		if (pos + len > LED_NUM_DIGITS)				\
-			len = LED_NUM_DIGITS - pos;			\
-									\
-		if (len > 0) {						\
-			unsigned long _flags;				\
-			const char *_end = buf + len;			\
-			img_decl = &leds_image[pos];			\
-									\
-			spin_lock_irqsave (leds_lock, _flags);		\
-			do						\
-				(copy_expr);				\
-			while (buf != _end);				\
-			spin_unlock_irqrestore (leds_lock, _flags);	\
-		}							\
-	}								\
-} while (0)
-
-/* Read LEN bytes from LEDs at position POS, into BUF.
-   Returns actual amount read.  */
-unsigned read_leds (unsigned pos, char *buf, unsigned len)
-{
-	DO_LED_COPY (const char *img, pos, buf, len, *buf++ = *img++);
-	return len;
-}
-
-/* Write LEN bytes to LEDs at position POS, from BUF.
-   Returns actual amount written.  */
-unsigned write_leds (unsigned pos, const char *buf, unsigned len)
-{
-	/* We write the actual LED values backwards, because
-	   increasing memory addresses reflect LEDs right-to-left. */
-	volatile char *led = &LED (LED_NUM_DIGITS - pos - 1);
-	/* We invert the value written to the hardware, because 1 = off,
-	   and 0 = on.  */
-	DO_LED_COPY (char *img, pos, buf, len,
-		     *led-- = 0xFF ^ (*img++ = *buf++));
-	return len;
-}
-
-
-/* Device functions.  */
-
-static ssize_t leds_dev_read (struct file *file, char *buf, size_t len,
-			      loff_t *pos)
-{
-	char temp_buf[LED_NUM_DIGITS];
-	len = read_leds (*pos, temp_buf, len);
-	if (copy_to_user (buf, temp_buf, len))
-		return -EFAULT;
-	*pos += len;
-	return len;
-}
-
-static ssize_t leds_dev_write (struct file *file, const char *buf, size_t len,
-			       loff_t *pos)
-{
-	char temp_buf[LED_NUM_DIGITS];
-	if (copy_from_user (temp_buf, buf, min_t(size_t, len, LED_NUM_DIGITS)))
-		return -EFAULT;
-	len = write_leds (*pos, temp_buf, len);
-	*pos += len;
-	return len;
-}
-
-static loff_t leds_dev_lseek (struct file *file, loff_t offs, int whence)
-{
-	if (whence == 1)
-		offs += file->f_pos; /* relative */
-	else if (whence == 2)
-		offs += LED_NUM_DIGITS; /* end-relative */
-
-	if (offs < 0 || offs > LED_NUM_DIGITS)
-		return -EINVAL;
-
-	file->f_pos = offs;
-
-	return 0;
-}
-
-static const struct file_operations leds_fops = {
-	.read		= leds_dev_read,
-	.write		= leds_dev_write,
-	.llseek		= leds_dev_lseek
-};
-
-static struct miscdevice leds_miscdev = {
-	.name		= "leds",
-	.minor		= LEDS_MINOR,
-	.fops		= &leds_fops
-};
-
-int __init leds_dev_init (void)
-{
-	return misc_register (&leds_miscdev);
-}
-
-__initcall (leds_dev_init);
diff --git a/arch/v850/kernel/rte_cb_multi.c b/arch/v850/kernel/rte_cb_multi.c
deleted file mode 100644
index 963d55ab34cc..000000000000
--- a/arch/v850/kernel/rte_cb_multi.c
+++ /dev/null
@@ -1,121 +0,0 @@
-/*
- * include/asm-v850/rte_multi.c -- Support for Multi debugger monitor ROM
- * 	on Midas lab RTE-CB series of evaluation boards
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/init.h>
-
-#include <asm/machdep.h>
-
-#define IRQ_ADDR(irq) (0x80 + (irq) * 0x10)
-
-/* A table of which interrupt vectors to install, since blindly
-   installing all of them makes the debugger stop working.  This is a
-   list of offsets in the interrupt vector area; each entry means to
-   copy that particular 16-byte vector.  An entry less than zero ends
-   the table.  */
-static long multi_intv_install_table[] = {
-	/* Trap vectors */
-	0x40, 0x50,		
-
-#ifdef CONFIG_RTE_CB_MULTI_DBTRAP
-	/* Illegal insn / dbtrap.  These are used by multi, so only handle
-	   them if configured to do so.  */
-	0x60,
-#endif
-
-	/* GINT1 - GINT3 (note, not GINT0!) */
-	IRQ_ADDR (IRQ_GINT(1)),
-	IRQ_ADDR (IRQ_GINT(2)),
-	IRQ_ADDR (IRQ_GINT(3)),
-
-	/* Timer D interrupts (up to 4 timers) */
-	IRQ_ADDR (IRQ_INTCMD(0)),
-#if IRQ_INTCMD_NUM > 1
-	IRQ_ADDR (IRQ_INTCMD(1)),
-#if IRQ_INTCMD_NUM > 2
-	IRQ_ADDR (IRQ_INTCMD(2)),
-#if IRQ_INTCMD_NUM > 3
-	IRQ_ADDR (IRQ_INTCMD(3)),
-#endif
-#endif
-#endif
-	
-	/* UART interrupts (up to 3 channels) */
-	IRQ_ADDR (IRQ_INTSER (0)), /* err */
-	IRQ_ADDR (IRQ_INTSR  (0)), /* rx */
-	IRQ_ADDR (IRQ_INTST  (0)), /* tx */
-#if IRQ_INTSR_NUM > 1
-	IRQ_ADDR (IRQ_INTSER (1)), /* err */
-	IRQ_ADDR (IRQ_INTSR  (1)), /* rx */
-	IRQ_ADDR (IRQ_INTST  (1)), /* tx */
-#if IRQ_INTSR_NUM > 2
-	IRQ_ADDR (IRQ_INTSER (2)), /* err */
-	IRQ_ADDR (IRQ_INTSR  (2)), /* rx */
-	IRQ_ADDR (IRQ_INTST  (2)), /* tx */
-#endif
-#endif
-
-	-1
-};
-
-/* Early initialization for kernel using Multi debugger ROM monitor.  */
-void __init multi_init (void)
-{
-	/* We're using the Multi debugger monitor, so we have to install
-	   the interrupt vectors.  The monitor doesn't allow them to be
-	   initially downloaded into their final destination because
-	   it's in the monitor's scratch-RAM area.  Unfortunately, Multi
-	   also doesn't deal correctly with ELF sections where the LMA
-	   and VMA differ -- it just ignores the LMA -- so we can't use
-	   that feature to work around the problem.  What we do instead
-	   is just put the interrupt vectors into a normal section, and
-	   do the necessary copying and relocation here.  Since the
-	   interrupt vector basically only contains `jr' instructions
-	   and no-ops, it's not that hard.  */
-	extern unsigned long _intv_load_start, _intv_start;
-	register unsigned long *src = &_intv_load_start;
-	register unsigned long *dst = (unsigned long *)INTV_BASE;
-	register unsigned long jr_fixup = (char *)&_intv_start - (char *)dst;
-	register long *ii;
-
-	/* Copy interrupt vectors as instructed by multi_intv_install_table. */
-	for (ii = multi_intv_install_table; *ii >= 0; ii++) {
-		/* Copy 16-byte interrupt vector at offset *ii.  */
-		int boffs;
-		for (boffs = 0; boffs < 0x10; boffs += sizeof *src) {
-			/* Copy a single word, fixing up the jump offs
-			   if it's a `jr' instruction.  */
-			int woffs = (*ii + boffs) / sizeof *src;
-			unsigned long word = src[woffs];
-
-			if ((word & 0xFC0) == 0x780) {
-				/* A `jr' insn, fix up its offset (and yes, the
-				   weird half-word swapping is intentional). */
-				unsigned short hi = word & 0xFFFF;
-				unsigned short lo = word >> 16;
-				unsigned long udisp22
-					= lo + ((hi & 0x3F) << 16);
-				long disp22 = (long)(udisp22 << 10) >> 10;
-
-				disp22 += jr_fixup;
-
-				hi = ((disp22 >> 16) & 0x3F) | 0x780;
-				lo = disp22 & 0xFFFF;
-
-				word = hi + (lo << 16);
-			}
-
-			dst[woffs] = word;
-		}
-	}
-}
diff --git a/arch/v850/kernel/rte_ma1_cb-rom.ld b/arch/v850/kernel/rte_ma1_cb-rom.ld
deleted file mode 100644
index 87b618f8253b..000000000000
--- a/arch/v850/kernel/rte_ma1_cb-rom.ld
+++ /dev/null
@@ -1,14 +0,0 @@
-/* Linker script for the Midas labs RTE-V850E/MA1-CB evaluation board
-   (CONFIG_RTE_CB_MA1), with kernel in ROM.  */
-
-MEMORY {
-	ROM   : ORIGIN = 0x00000000, LENGTH = 0x00100000
-	/* 1MB of SRAM.  This memory is mirrored 4 times.  */
-	SRAM  : ORIGIN = SRAM_ADDR,  LENGTH = SRAM_SIZE
-	/* 32MB of SDRAM.  */
-	SDRAM : ORIGIN = SDRAM_ADDR, LENGTH = SDRAM_SIZE
-}
-
-SECTIONS {
-	ROMK_SECTIONS(ROM, SRAM)
-}
diff --git a/arch/v850/kernel/rte_ma1_cb.c b/arch/v850/kernel/rte_ma1_cb.c
deleted file mode 100644
index 08abf3d5f8df..000000000000
--- a/arch/v850/kernel/rte_ma1_cb.c
+++ /dev/null
@@ -1,107 +0,0 @@
-/*
- * arch/v850/kernel/rte_ma1_cb.c -- Midas labs RTE-V850E/MA1-CB board
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/kernel.h>
-#include <linux/init.h>
-#include <linux/bootmem.h>
-
-#include <asm/atomic.h>
-#include <asm/page.h>
-#include <asm/ma1.h>
-#include <asm/rte_ma1_cb.h>
-#include <asm/v850e_timer_c.h>
-
-#include "mach.h"
-
-
-/* SRAM and SDRAM are almost contiguous (with a small hole in between;
-   see mach_reserve_bootmem for details), so just use both as one big area.  */
-#define RAM_START 	SRAM_ADDR
-#define RAM_END		(SDRAM_ADDR + SDRAM_SIZE)
-
-
-void __init mach_early_init (void)
-{
-	rte_cb_early_init ();
-}
-
-void __init mach_get_physical_ram (unsigned long *ram_start,
-				   unsigned long *ram_len)
-{
-	*ram_start = RAM_START;
-	*ram_len = RAM_END - RAM_START;
-}
-
-void __init mach_reserve_bootmem ()
-{
-#ifdef CONFIG_RTE_CB_MULTI
-	/* Prevent the kernel from touching the monitor's scratch RAM.  */
-	reserve_bootmem(MON_SCRATCH_ADDR, MON_SCRATCH_SIZE,
-			BOOTMEM_DEFAULT);
-#endif
-
-	/* The space between SRAM and SDRAM is filled with duplicate
-	   images of SRAM.  Prevent the kernel from using them.  */
-	reserve_bootmem (SRAM_ADDR + SRAM_SIZE,
-			 SDRAM_ADDR - (SRAM_ADDR + SRAM_SIZE),
-			 BOOTMEM_DEFAULT);
-}
-
-void mach_gettimeofday (struct timespec *tv)
-{
-	tv->tv_sec = 0;
-	tv->tv_nsec = 0;
-}
-
-/* Called before configuring an on-chip UART.  */
-void rte_ma1_cb_uart_pre_configure (unsigned chan,
-				    unsigned cflags, unsigned baud)
-{
-	/* The RTE-MA1-CB connects some general-purpose I/O pins on the
-	   CPU to the RTS/CTS lines of UART 0's serial connection.
-	   I/O pins P42 and P43 are RTS and CTS respectively.  */
-	if (chan == 0) {
-		/* Put P42 & P43 in I/O port mode.  */
-		MA_PORT4_PMC &= ~0xC;
-		/* Make P42 an output, and P43 an input.  */
-		MA_PORT4_PM = (MA_PORT4_PM & ~0xC) | 0x8;
-	}
-
-	/* Do pre-configuration for the actual UART.  */
-	ma_uart_pre_configure (chan, cflags, baud);
-}
-
-void __init mach_init_irqs (void)
-{
-	unsigned tc;
-
-	/* Initialize interrupts.  */
-	ma_init_irqs ();
-	rte_cb_init_irqs ();
-
-	/* Use falling-edge-sensitivity for interrupts .  */
-	V850E_TIMER_C_SESC (0) &= ~0xC;
-	V850E_TIMER_C_SESC (1) &= ~0xF;
-
-	/* INTP000-INTP011 are shared with `Timer C', so we have to set
-	   up Timer C to pass them through as raw interrupts.  */
-	for (tc = 0; tc < 2; tc++)
-		/* Turn on the timer.  */
-		V850E_TIMER_C_TMCC0 (tc) |= V850E_TIMER_C_TMCC0_CAE;
-
-	/* Make sure the relevant port0/port1 pins are assigned
-	   interrupt duty.  We used INTP001-INTP011 (don't screw with
-	   INTP000 because the monitor uses it).  */
-	MA_PORT0_PMC |= 0x4;	/* P02 (INTP001) in IRQ mode.  */
-	MA_PORT1_PMC |= 0x6;	/* P11 (INTP010) & P12 (INTP011) in IRQ mode.*/
-}
diff --git a/arch/v850/kernel/rte_ma1_cb.ld b/arch/v850/kernel/rte_ma1_cb.ld
deleted file mode 100644
index c8e16d16be41..000000000000
--- a/arch/v850/kernel/rte_ma1_cb.ld
+++ /dev/null
@@ -1,57 +0,0 @@
-/* Linker script for the Midas labs RTE-V850E/MA1-CB evaluation board
-   (CONFIG_RTE_CB_MA1), with kernel in SDRAM, under Multi debugger.  */
-
-MEMORY {
-	/* 1MB of SRAM; we can't use the last 32KB, because it's used by
-	   the monitor scratch-RAM.  This memory is mirrored 4 times.  */
-	SRAM  : ORIGIN = SRAM_ADDR,  LENGTH = (SRAM_SIZE - MON_SCRATCH_SIZE)
-	/* Monitor scratch RAM; only the interrupt vectors should go here.  */
-	MRAM  : ORIGIN = MON_SCRATCH_ADDR,  LENGTH = MON_SCRATCH_SIZE
-	/* 32MB of SDRAM.  */
-	SDRAM : ORIGIN = SDRAM_ADDR, LENGTH = SDRAM_SIZE
-}
-
-#ifdef CONFIG_RTE_CB_MA1_KSRAM
-# define KRAM SRAM
-#else
-# define KRAM SDRAM
-#endif
-
-SECTIONS {
-	/* We can't use RAMK_KRAM_CONTENTS because that puts the whole
-	   kernel in a single ELF segment, and the Multi debugger (which
-	   we use to load the kernel) appears to have bizarre problems
-	   dealing with it.  */
-
-	.text : {
-		__kram_start = . ;
-		TEXT_CONTENTS
-	} > KRAM
-
-	.data : {
-		DATA_CONTENTS
-		BSS_CONTENTS
-		RAMK_INIT_CONTENTS
-		__kram_end = . ;
-		BOOTMAP_CONTENTS
-
-		/* The address at which the interrupt vectors are initially
-		   loaded by the loader.  We can't load the interrupt vectors
-		   directly into their target location, because the monitor
-		   ROM for the GHS Multi debugger barfs if we try.
-		   Unfortunately, Multi also doesn't deal correctly with ELF
-		   sections where the LMA and VMA differ (it just ignores the
-		   LMA), so we can't use that feature to work around the
-		   problem!  What we do instead is just put the interrupt
-		   vectors into a normal section, and have the
-		   `mach_early_init' function for Midas boards do the
-		   necessary copying and relocation at runtime (this section
-		   basically only contains `jr' instructions, so it's not
-		   that hard).  */
-		. = ALIGN (0x10) ;
-		__intv_load_start = . ;
-		INTV_CONTENTS
-	} > KRAM
-
-	.root ALIGN (4096) : { ROOT_FS_CONTENTS } > SDRAM
-}
diff --git a/arch/v850/kernel/rte_mb_a_pci.c b/arch/v850/kernel/rte_mb_a_pci.c
deleted file mode 100644
index 687e367d8b64..000000000000
--- a/arch/v850/kernel/rte_mb_a_pci.c
+++ /dev/null
@@ -1,819 +0,0 @@
-/*
- * arch/v850/kernel/mb_a_pci.c -- PCI support for Midas lab RTE-MOTHER-A board
- *
- *  Copyright (C) 2001,02,03,05  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03,05  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/init.h>
-#include <linux/slab.h>
-#include <linux/spinlock.h>
-#include <linux/pci.h>
-
-#include <asm/machdep.h>
-
-/* __nomods_init is like __devinit, but is a no-op when modules are enabled.
-   This is used by some routines that can be called either during boot
-   or by a module.  */
-#ifdef CONFIG_MODULES
-#define __nomods_init /*nothing*/
-#else
-#define __nomods_init __devinit
-#endif
-
-/* PCI devices on the Mother-A board can only do DMA to/from the MB SRAM
-   (the RTE-V850E/MA1-CB cpu board doesn't support PCI access to
-   CPU-board memory), and since linux DMA buffers are allocated in
-   normal kernel memory, we basically have to copy DMA blocks around
-   (this is like a `bounce buffer').  When a DMA block is `mapped', we
-   allocate an identically sized block in MB SRAM, and if we're doing
-   output to the device, copy the CPU-memory block to the MB-SRAM block.
-   When an active block is `unmapped', we will copy the block back to
-   CPU memory if necessary, and then deallocate the MB SRAM block.
-   Ack.  */
-
-/* Where the motherboard SRAM is in the PCI-bus address space (the
-   first 512K of it is also mapped at PCI address 0).  */
-#define PCI_MB_SRAM_ADDR 0x800000
-
-/* Convert CPU-view MB SRAM address to/from PCI-view addresses of the
-   same memory.  */
-#define MB_SRAM_TO_PCI(mb_sram_addr) \
-   ((dma_addr_t)mb_sram_addr - MB_A_SRAM_ADDR + PCI_MB_SRAM_ADDR)
-#define PCI_TO_MB_SRAM(pci_addr)     \
-   (void *)(pci_addr - PCI_MB_SRAM_ADDR + MB_A_SRAM_ADDR)
-
-static void pcibios_assign_resources (void);
-
-struct mb_pci_dev_irq {
-	unsigned dev;		/* PCI device number */
-	unsigned irq_base;	/* First IRQ  */
-	unsigned query_pin;	/* True if we should read the device's
-				   Interrupt Pin info, and allocate
-				   interrupt IRQ_BASE + PIN.  */
-};
-
-/* PCI interrupts are mapped statically to GBUS interrupts.  */
-static struct mb_pci_dev_irq mb_pci_dev_irqs[] = {
-	/* Motherboard SB82558 ethernet controller */
-	{ 10,	IRQ_MB_A_LAN,		0 },
-	/* PCI slot 1 */
-	{ 8, 	IRQ_MB_A_PCI1(0),	1 },
-	/* PCI slot 2 */
-	{ 9, 	IRQ_MB_A_PCI2(0),	1 }
-};
-#define NUM_MB_PCI_DEV_IRQS ARRAY_SIZE(mb_pci_dev_irqs)
-
-
-/* PCI configuration primitives.  */
-
-#define CONFIG_DMCFGA(bus, devfn, offs)					\
-   (0x80000000								\
-    | ((offs) & ~0x3)							\
-    | ((devfn) << 8)							\
-    | ((bus)->number << 16))
-
-static int
-mb_pci_read (struct pci_bus *bus, unsigned devfn, int offs, int size, u32 *rval)
-{
-	u32 addr;
-	int flags;
-
-	local_irq_save (flags);
-
-	MB_A_PCI_PCICR = 0x7;
-	MB_A_PCI_DMCFGA = CONFIG_DMCFGA (bus, devfn, offs);
-
-	addr = MB_A_PCI_IO_ADDR + (offs & 0x3);
-
-	switch (size) {
-	case 1:	*rval = *(volatile  u8 *)addr; break;
-	case 2:	*rval = *(volatile u16 *)addr; break;
-	case 4:	*rval = *(volatile u32 *)addr; break;
-	}
-
-        if (MB_A_PCI_PCISR & 0x2000) {
-		MB_A_PCI_PCISR = 0x2000;
-		*rval = ~0;
-        }
-
-	MB_A_PCI_DMCFGA = 0;
-
-	local_irq_restore (flags);
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static int
-mb_pci_write (struct pci_bus *bus, unsigned devfn, int offs, int size, u32 val)
-{
-	u32 addr;
-	int flags;
-
-	local_irq_save (flags);
-
-	MB_A_PCI_PCICR = 0x7;
-	MB_A_PCI_DMCFGA = CONFIG_DMCFGA (bus, devfn, offs);
-
-	addr = MB_A_PCI_IO_ADDR + (offs & 0x3);
-
-	switch (size) {
-	case 1: *(volatile  u8 *)addr = val; break;
-	case 2: *(volatile u16 *)addr = val; break;
-	case 4: *(volatile u32 *)addr = val; break;
-	}
-
-        if (MB_A_PCI_PCISR & 0x2000)
-		MB_A_PCI_PCISR = 0x2000;
-
-	MB_A_PCI_DMCFGA = 0;
-
-	local_irq_restore (flags);
-
-	return PCIBIOS_SUCCESSFUL;
-}
-
-static struct pci_ops mb_pci_config_ops = {
-	.read	= mb_pci_read,
-	.write	= mb_pci_write,
-};
-
-
-/* PCI Initialization.  */
-
-static struct pci_bus *mb_pci_bus = 0;
-
-/* Do initial PCI setup.  */
-static int __devinit pcibios_init (void)
-{
-	u32 id = MB_A_PCI_PCIHIDR;
-	u16 vendor = id & 0xFFFF;
-	u16 device = (id >> 16) & 0xFFFF;
-
-	if (vendor == PCI_VENDOR_ID_PLX && device == PCI_DEVICE_ID_PLX_9080) {
-		printk (KERN_INFO
-			"PCI: PLX Technology PCI9080 HOST/PCI bridge\n");
-
-		MB_A_PCI_PCICR = 0x147;
-
-		MB_A_PCI_PCIBAR0 = 0x007FFF00;
-		MB_A_PCI_PCIBAR1 = 0x0000FF00;
-		MB_A_PCI_PCIBAR2 = 0x00800000;
-
-		MB_A_PCI_PCILTR = 0x20;
-
-		MB_A_PCI_PCIPBAM |= 0x3;
-
-		MB_A_PCI_PCISR =  ~0; /* Clear errors.  */
-
-		/* Reprogram the motherboard's IO/config address space,
-		   as we don't support the GCS7 address space that the
-		   default uses.  */
-
-		/* Significant address bits used for decoding PCI GCS5 space
-		   accesses.  */
-		MB_A_PCI_DMRR = ~(MB_A_PCI_MEM_SIZE - 1);
-
-		/* I don't understand this, but the SolutionGear example code
-		   uses such an offset, and it doesn't work without it.  XXX */
-#if GCS5_SIZE == 0x00800000
-#define GCS5_CFG_OFFS 0x00800000
-#else
-#define GCS5_CFG_OFFS 0
-#endif
-
-		/* Address bit values for matching.  Note that we have to give
-		   the address from the motherboard's point of view, which is
-		   different than the CPU's.  */
-		/* PCI memory space.  */
-		MB_A_PCI_DMLBAM = GCS5_CFG_OFFS + 0x0;
-		/* PCI I/O space.  */
-		MB_A_PCI_DMLBAI =
-			GCS5_CFG_OFFS + (MB_A_PCI_IO_ADDR - GCS5_ADDR);
-
-		mb_pci_bus = pci_scan_bus (0, &mb_pci_config_ops, 0);
-
-		pcibios_assign_resources ();
-	} else
-		printk (KERN_ERR "PCI: HOST/PCI bridge not found\n");
-
-	return 0;
-}
-
-subsys_initcall (pcibios_init);
-
-char __devinit *pcibios_setup (char *option)
-{
-	/* Don't handle any options. */
-	return option;
-}
-
-
-int __nomods_init pcibios_enable_device (struct pci_dev *dev, int mask)
-{
-	u16 cmd, old_cmd;
-	int idx;
-	struct resource *r;
-
-	pci_read_config_word(dev, PCI_COMMAND, &cmd);
-	old_cmd = cmd;
-	for (idx = 0; idx < 6; idx++) {
-		r = &dev->resource[idx];
-		if (!r->start && r->end) {
-			printk(KERN_ERR "PCI: Device %s not available because "
-			       "of resource collisions\n", pci_name(dev));
-			return -EINVAL;
-		}
-		if (r->flags & IORESOURCE_IO)
-			cmd |= PCI_COMMAND_IO;
-		if (r->flags & IORESOURCE_MEM)
-			cmd |= PCI_COMMAND_MEMORY;
-	}
-	if (cmd != old_cmd) {
-		printk("PCI: Enabling device %s (%04x -> %04x)\n",
-		       pci_name(dev), old_cmd, cmd);
-		pci_write_config_word(dev, PCI_COMMAND, cmd);
-	}
-	return 0;
-}
-
-
-/* Resource allocation.  */
-static void __devinit pcibios_assign_resources (void)
-{
-	struct pci_dev *dev = NULL;
-	struct resource *r;
-
-	for_each_pci_dev(dev) {
-		unsigned di_num;
-		unsigned class = dev->class >> 8;
-
-		if (class && class != PCI_CLASS_BRIDGE_HOST) {
-			unsigned r_num;
-			for(r_num = 0; r_num < 6; r_num++) {
-				r = &dev->resource[r_num];
-				if (!r->start && r->end)
-					pci_assign_resource (dev, r_num);
-			}
-		}
-
-		/* Assign interrupts.  */
-		for (di_num = 0; di_num < NUM_MB_PCI_DEV_IRQS; di_num++) {
-			struct mb_pci_dev_irq *di = &mb_pci_dev_irqs[di_num];
-
-			if (di->dev == PCI_SLOT (dev->devfn)) {
-				unsigned irq = di->irq_base;
-
-				if (di->query_pin) {
-					/* Find out which interrupt pin
-					   this device uses (each PCI
-					   slot has 4).  */
-					u8 irq_pin;
-
-					pci_read_config_byte (dev,
-							     PCI_INTERRUPT_PIN,
-							      &irq_pin);
-
-					if (irq_pin == 0)
-						/* Doesn't use interrupts.  */ 
-						continue;
-					else
-						irq += irq_pin - 1;
-				}
-
-				pcibios_update_irq (dev, irq);
-			}
-		}
-	}
-}
-
-void __devinit pcibios_update_irq (struct pci_dev *dev, int irq)
-{
-	dev->irq = irq;
-	pci_write_config_byte (dev, PCI_INTERRUPT_LINE, irq);
-}
-
-void __devinit
-pcibios_resource_to_bus(struct pci_dev *dev, struct pci_bus_region *region,
-			struct resource *res)
-{
-	unsigned long offset = 0;
-
-	if (res->flags & IORESOURCE_IO) {
-		offset = MB_A_PCI_IO_ADDR;
-	} else if (res->flags & IORESOURCE_MEM) {
-		offset = MB_A_PCI_MEM_ADDR;
-	}
-
-	region->start = res->start - offset;
-	region->end = res->end - offset;
-}
-
-
-/* Stubs for things we don't use.  */
-
-/* Called after each bus is probed, but before its children are examined. */
-void pcibios_fixup_bus(struct pci_bus *b)
-{
-}
-
-void
-pcibios_align_resource (void *data, struct resource *res,
-			resource_size_t size, resource_size_t align)
-{
-}
-
-void pcibios_set_master (struct pci_dev *dev)
-{
-}
-
-
-/* Mother-A SRAM memory allocation.  This is a simple first-fit allocator.  */
-
-/* A memory free-list node.  */
-struct mb_sram_free_area {
-	void *mem;
-	unsigned long size;
-	struct mb_sram_free_area *next;
-};
-
-/* The tail of the free-list, which starts out containing all the SRAM.  */
-static struct mb_sram_free_area mb_sram_free_tail = {
-	(void *)MB_A_SRAM_ADDR, MB_A_SRAM_SIZE, 0
-};
-
-/* The free-list.  */
-static struct mb_sram_free_area *mb_sram_free_areas = &mb_sram_free_tail;
-
-/* The free-list of free free-list nodes. (:-)  */
-static struct mb_sram_free_area *mb_sram_free_free_areas = 0;
-
-/* Spinlock protecting the above globals.  */
-static DEFINE_SPINLOCK(mb_sram_lock);
-
-/* Allocate a memory block at least SIZE bytes long in the Mother-A SRAM
-   space.  */
-static void *alloc_mb_sram (size_t size)
-{
-	struct mb_sram_free_area *prev, *fa;
-	unsigned long flags;
-	void *mem = 0;
-
-	spin_lock_irqsave (mb_sram_lock, flags);
-
-	/* Look for a free area that can contain SIZE bytes.  */
-	for (prev = 0, fa = mb_sram_free_areas; fa; prev = fa, fa = fa->next)
-		if (fa->size >= size) {
-			/* Found one!  */
-			mem = fa->mem;
-
-			if (fa->size == size) {
-				/* In fact, it fits exactly, so remove
-				   this node from the free-list.  */
-				if (prev)
-					prev->next = fa->next;
-				else
-					mb_sram_free_areas = fa->next;
-				/* Put it on the free-list-entry-free-list. */
-				fa->next = mb_sram_free_free_areas;
-				mb_sram_free_free_areas = fa;
-			} else {
-				/* FA is bigger than SIZE, so just
-				   reduce its size to account for this
-				   allocation.  */
-				fa->mem += size;
-				fa->size -= size;
-			}
-
-			break;
-		}
-
-	spin_unlock_irqrestore (mb_sram_lock, flags);
-
-	return mem;
-}
-
-/* Return the memory area MEM of size SIZE to the MB SRAM free pool.  */
-static void free_mb_sram (void *mem, size_t size)
-{
-	struct mb_sram_free_area *prev, *fa, *new_fa;
-	unsigned long flags;
-	void *end = mem + size;
-
-	spin_lock_irqsave (mb_sram_lock, flags);
-
- retry:
-	/* Find an adjacent free-list entry.  */
-	for (prev = 0, fa = mb_sram_free_areas; fa; prev = fa, fa = fa->next)
-		if (fa->mem == end) {
-			/* FA is just after MEM, grow down to encompass it. */
-			fa->mem = mem;
-			fa->size += size;
-			goto done;
-		} else if (fa->mem + fa->size == mem) {
-			struct mb_sram_free_area *next_fa = fa->next;
-
-			/* FA is just before MEM, expand to encompass it. */
-			fa->size += size;
-
-			/* See if FA can now be merged with its successor. */
-			if (next_fa && fa->mem + fa->size == next_fa->mem) {
-				/* Yup; merge NEXT_FA's info into FA.  */
-				fa->size += next_fa->size;
-				fa->next = next_fa->next;
-				/* Free NEXT_FA.  */
-				next_fa->next = mb_sram_free_free_areas;
-				mb_sram_free_free_areas = next_fa;
-			}
-			goto done;
-		} else if (fa->mem > mem)
-			/* We've reached the right spot in the free-list
-			   without finding an adjacent free-area, so add
-			   a new free area to hold mem. */
-			break;
-
-	/* Make a new free-list entry.  */
-
-	/* First, get a free-list entry.  */
-	if (! mb_sram_free_free_areas) {
-		/* There are none, so make some.  */
-		void *block;
-		size_t block_size = sizeof (struct mb_sram_free_area) * 8;
-
-		/* Don't hold the lock while calling kmalloc (I'm not
-		   sure whether it would be a problem, since we use
-		   GFP_ATOMIC, but it makes me nervous).  */
-		spin_unlock_irqrestore (mb_sram_lock, flags);
-
-		block = kmalloc (block_size, GFP_ATOMIC);
-		if (! block)
-			panic ("free_mb_sram: can't allocate free-list entry");
-
-		/* Now get the lock back.  */
-		spin_lock_irqsave (mb_sram_lock, flags);
-
-		/* Add the new free free-list entries.  */
-		while (block_size > 0) {
-			struct mb_sram_free_area *nfa = block;
-			nfa->next = mb_sram_free_free_areas;
-			mb_sram_free_free_areas = nfa;
-			block += sizeof *nfa;
-			block_size -= sizeof *nfa;
-		}
-
-		/* Since we dropped the lock to call kmalloc, the
-		   free-list could have changed, so retry from the
-		   beginning.  */
-		goto retry;
-	}
-
-	/* Remove NEW_FA from the free-list of free-list entries.  */
-	new_fa = mb_sram_free_free_areas;
-	mb_sram_free_free_areas = new_fa->next;
-
-	/* NEW_FA initially holds only MEM.  */
-	new_fa->mem = mem;
-	new_fa->size = size;
-
-	/* Insert NEW_FA in the free-list between PREV and FA. */
-	new_fa->next = fa;
-	if (prev)
-		prev->next = new_fa;
-	else
-		mb_sram_free_areas = new_fa;
-
- done:
-	spin_unlock_irqrestore (mb_sram_lock, flags);
-}
-
-
-/* Maintainence of CPU -> Mother-A DMA mappings.  */
-
-struct dma_mapping {
-	void *cpu_addr;
-	void *mb_sram_addr;
-	size_t size;
-	struct dma_mapping *next;
-};
-
-/* A list of mappings from CPU addresses to MB SRAM addresses for active
-   DMA blocks (that have been `granted' to the PCI device).  */
-static struct dma_mapping *active_dma_mappings = 0;
-
-/* A list of free mapping objects.  */
-static struct dma_mapping *free_dma_mappings = 0;
-
-/* Spinlock protecting the above globals.  */
-static DEFINE_SPINLOCK(dma_mappings_lock);
-
-static struct dma_mapping *new_dma_mapping (size_t size)
-{
-	unsigned long flags;
-	struct dma_mapping *mapping;
-	void *mb_sram_block = alloc_mb_sram (size);
-
-	if (! mb_sram_block)
-		return 0;
-
-	spin_lock_irqsave (dma_mappings_lock, flags);
-
-	if (! free_dma_mappings) {
-		/* We're out of mapping structures, make more.  */
-		void *mblock;
-		size_t mblock_size = sizeof (struct dma_mapping) * 8;
-
-		/* Don't hold the lock while calling kmalloc (I'm not
-		   sure whether it would be a problem, since we use
-		   GFP_ATOMIC, but it makes me nervous).  */
-		spin_unlock_irqrestore (dma_mappings_lock, flags);
-
-		mblock = kmalloc (mblock_size, GFP_ATOMIC);
-		if (! mblock) {
-			free_mb_sram (mb_sram_block, size);
-			return 0;
-		}
-
-		/* Get the lock back.  */
-		spin_lock_irqsave (dma_mappings_lock, flags);
-
-		/* Add the new mapping structures to the free-list.  */
-		while (mblock_size > 0) {
-			struct dma_mapping *fm = mblock;
-			fm->next = free_dma_mappings;
-			free_dma_mappings = fm;
-			mblock += sizeof *fm;
-			mblock_size -= sizeof *fm;
-		}
-	}
-
-	/* Get a mapping struct from the freelist.  */
-	mapping = free_dma_mappings;
-	free_dma_mappings = mapping->next;
-
-	/* Initialize the mapping.  Other fields should be filled in by
-	   caller.  */
-	mapping->mb_sram_addr = mb_sram_block;
-	mapping->size = size;
-
-	/* Add it to the list of active mappings.  */
-	mapping->next = active_dma_mappings;
-	active_dma_mappings = mapping;
-
-	spin_unlock_irqrestore (dma_mappings_lock, flags);
-
-	return mapping;
-}
-
-static struct dma_mapping *find_dma_mapping (void *mb_sram_addr)
-{
-	unsigned long flags;
-	struct dma_mapping *mapping;
-
-	spin_lock_irqsave (dma_mappings_lock, flags);
-
-	for (mapping = active_dma_mappings; mapping; mapping = mapping->next)
-		if (mapping->mb_sram_addr == mb_sram_addr) {
-			spin_unlock_irqrestore (dma_mappings_lock, flags);
-			return mapping;
-		}
-
-	panic ("find_dma_mapping: unmapped PCI DMA addr 0x%x",
-	       MB_SRAM_TO_PCI (mb_sram_addr));
-}
-
-static struct dma_mapping *deactivate_dma_mapping (void *mb_sram_addr)
-{
-	unsigned long flags;
-	struct dma_mapping *mapping, *prev;
-
-	spin_lock_irqsave (dma_mappings_lock, flags);
-
-	for (prev = 0, mapping = active_dma_mappings;
-	     mapping;
-	     prev = mapping, mapping = mapping->next)
-	{
-		if (mapping->mb_sram_addr == mb_sram_addr) {
-			/* This is the MAPPING; deactivate it.  */
-			if (prev)
-				prev->next = mapping->next;
-			else
-				active_dma_mappings = mapping->next;
-
-			spin_unlock_irqrestore (dma_mappings_lock, flags);
-
-			return mapping;
-		}
-	}
-
-	panic ("deactivate_dma_mapping: unmapped PCI DMA addr 0x%x",
-	       MB_SRAM_TO_PCI (mb_sram_addr));
-}
-
-/* Return MAPPING to the freelist.  */
-static inline void
-free_dma_mapping (struct dma_mapping *mapping)
-{
-	unsigned long flags;
-
-	free_mb_sram (mapping->mb_sram_addr, mapping->size);
-
-	spin_lock_irqsave (dma_mappings_lock, flags);
-
-	mapping->next = free_dma_mappings;
-	free_dma_mappings = mapping;
-
-	spin_unlock_irqrestore (dma_mappings_lock, flags);
-}
-
-
-/* Single PCI DMA mappings.  */
-
-/* `Grant' to PDEV the memory block at CPU_ADDR, for doing DMA.  The
-   32-bit PCI bus mastering address to use is returned.  the device owns
-   this memory until either pci_unmap_single or pci_dma_sync_single is
-   performed.  */
-dma_addr_t
-pci_map_single (struct pci_dev *pdev, void *cpu_addr, size_t size, int dir)
-{
-	struct dma_mapping *mapping = new_dma_mapping (size);
-
-	if (! mapping)
-		return 0;
-
-	mapping->cpu_addr = cpu_addr;
-
-	if (dir == PCI_DMA_BIDIRECTIONAL || dir == PCI_DMA_TODEVICE)
-		memcpy (mapping->mb_sram_addr, cpu_addr, size);
-
-	return MB_SRAM_TO_PCI (mapping->mb_sram_addr);
-}
-
-/* Return to the CPU the PCI DMA memory block previously `granted' to
-   PDEV, at DMA_ADDR.  */
-void pci_unmap_single (struct pci_dev *pdev, dma_addr_t dma_addr, size_t size,
-		       int dir)
-{
-	void *mb_sram_addr = PCI_TO_MB_SRAM (dma_addr);
-	struct dma_mapping *mapping = deactivate_dma_mapping (mb_sram_addr);
-
-	if (size != mapping->size)
-		panic ("pci_unmap_single: size (%d) doesn't match"
-		       " size of mapping at PCI DMA addr 0x%x (%d)\n",
-		       size, dma_addr, mapping->size);
-
-	/* Copy back the DMA'd contents if necessary.  */
-	if (dir == PCI_DMA_BIDIRECTIONAL || dir == PCI_DMA_FROMDEVICE)
-		memcpy (mapping->cpu_addr, mb_sram_addr, size);
-
-	/* Return mapping to the freelist.  */
-	free_dma_mapping (mapping);
-}
-
-/* Make physical memory consistent for a single streaming mode DMA
-   translation after a transfer.
-
-   If you perform a pci_map_single() but wish to interrogate the
-   buffer using the cpu, yet do not wish to teardown the PCI dma
-   mapping, you must call this function before doing so.  At the next
-   point you give the PCI dma address back to the card, you must first
-   perform a pci_dma_sync_for_device, and then the device again owns
-   the buffer.  */
-void
-pci_dma_sync_single_for_cpu (struct pci_dev *pdev, dma_addr_t dma_addr, size_t size,
-		     int dir)
-{
-	void *mb_sram_addr = PCI_TO_MB_SRAM (dma_addr);
-	struct dma_mapping *mapping = find_dma_mapping (mb_sram_addr);
-
-	/* Synchronize the DMA buffer with the CPU buffer if necessary.  */
-	if (dir == PCI_DMA_FROMDEVICE)
-		memcpy (mapping->cpu_addr, mb_sram_addr, size);
-	else if (dir == PCI_DMA_TODEVICE)
-		; /* nothing to do */
-	else
-		panic("pci_dma_sync_single: unsupported sync dir: %d", dir);
-}
-
-void
-pci_dma_sync_single_for_device (struct pci_dev *pdev, dma_addr_t dma_addr, size_t size,
-				int dir)
-{
-	void *mb_sram_addr = PCI_TO_MB_SRAM (dma_addr);
-	struct dma_mapping *mapping = find_dma_mapping (mb_sram_addr);
-
-	/* Synchronize the DMA buffer with the CPU buffer if necessary.  */
-	if (dir == PCI_DMA_FROMDEVICE)
-		; /* nothing to do */
-	else if (dir == PCI_DMA_TODEVICE)
-		memcpy (mb_sram_addr, mapping->cpu_addr, size);
-	else
-		panic("pci_dma_sync_single: unsupported sync dir: %d", dir);
-}
-
-
-/* Scatter-gather PCI DMA mappings.  */
-
-/* Do multiple DMA mappings at once.  */
-int
-pci_map_sg (struct pci_dev *pdev, struct scatterlist *sg, int sg_len, int dir)
-{
-	BUG ();
-	return 0;
-}
-
-/* Unmap multiple DMA mappings at once.  */
-void
-pci_unmap_sg (struct pci_dev *pdev, struct scatterlist *sg, int sg_len,int dir)
-{
-	BUG ();
-}
-
-/* Make physical memory consistent for a set of streaming mode DMA
-   translations after a transfer.  The same as pci_dma_sync_single_* but
-   for a scatter-gather list, same rules and usage.  */
-
-void
-pci_dma_sync_sg_for_cpu (struct pci_dev *dev,
-			 struct scatterlist *sg, int sg_len,
-			 int dir)
-{
-	BUG ();
-}
-
-void
-pci_dma_sync_sg_for_device (struct pci_dev *dev,
-			    struct scatterlist *sg, int sg_len,
-			    int dir)
-{
-	BUG ();
-}
-
-
-/* PCI mem mapping.  */
-
-/* Allocate and map kernel buffer using consistent mode DMA for PCI
-   device.  Returns non-NULL cpu-view pointer to the buffer if
-   successful and sets *DMA_ADDR to the pci side dma address as well,
-   else DMA_ADDR is undefined.  */
-void *
-pci_alloc_consistent (struct pci_dev *pdev, size_t size, dma_addr_t *dma_addr)
-{
-	void *mb_sram_mem = alloc_mb_sram (size);
-	if (mb_sram_mem)
-		*dma_addr = MB_SRAM_TO_PCI (mb_sram_mem);
-	return mb_sram_mem;
-}
-
-/* Free and unmap a consistent DMA buffer.  CPU_ADDR and DMA_ADDR must
-   be values that were returned from pci_alloc_consistent.  SIZE must be
-   the same as what as passed into pci_alloc_consistent.  References to
-   the memory and mappings associated with CPU_ADDR or DMA_ADDR past
-   this call are illegal.  */
-void
-pci_free_consistent (struct pci_dev *pdev, size_t size, void *cpu_addr,
-		     dma_addr_t dma_addr)
-{
-	void *mb_sram_mem = PCI_TO_MB_SRAM (dma_addr);
-	free_mb_sram (mb_sram_mem, size);
-}
-
-
-/* iomap/iomap */
-
-void __iomem *pci_iomap (struct pci_dev *dev, int bar, unsigned long max)
-{
-	resource_size_t start = pci_resource_start (dev, bar);
-	resource_size_t len = pci_resource_len (dev, bar);
-
-	if (!start || len == 0)
-		return 0;
-
-	/* None of the ioremap functions actually do anything, other than
-	   re-casting their argument, so don't bother differentiating them.  */
-	return ioremap (start, len);
-}
-
-void pci_iounmap (struct pci_dev *dev, void __iomem *addr)
-{
-	/* nothing */
-}
-
-
-/* symbol exports (for modules) */
-
-EXPORT_SYMBOL (pci_map_single);
-EXPORT_SYMBOL (pci_unmap_single);
-EXPORT_SYMBOL (pci_alloc_consistent);
-EXPORT_SYMBOL (pci_free_consistent);
-EXPORT_SYMBOL (pci_dma_sync_single_for_cpu);
-EXPORT_SYMBOL (pci_dma_sync_single_for_device);
-EXPORT_SYMBOL (pci_iomap);
-EXPORT_SYMBOL (pci_iounmap);
diff --git a/arch/v850/kernel/rte_me2_cb.c b/arch/v850/kernel/rte_me2_cb.c
deleted file mode 100644
index 46803d48dffe..000000000000
--- a/arch/v850/kernel/rte_me2_cb.c
+++ /dev/null
@@ -1,298 +0,0 @@
-/*
- * arch/v850/kernel/rte_me2_cb.c -- Midas labs RTE-V850E/ME2-CB board
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/kernel.h>
-#include <linux/init.h>
-#include <linux/bootmem.h>
-#include <linux/irq.h>
-#include <linux/fs.h>
-#include <linux/major.h>
-#include <linux/sched.h>
-#include <linux/delay.h>
-
-#include <asm/atomic.h>
-#include <asm/page.h>
-#include <asm/me2.h>
-#include <asm/rte_me2_cb.h>
-#include <asm/machdep.h>
-#include <asm/v850e_intc.h>
-#include <asm/v850e_cache.h>
-#include <asm/irq.h>
-
-#include "mach.h"
-
-extern unsigned long *_intv_start;
-extern unsigned long *_intv_end;
-
-/* LED access routines.  */
-extern unsigned read_leds (int pos, char *buf, int len);
-extern unsigned write_leds (int pos, const char *buf, int len);
-
-
-/* SDRAM are almost contiguous (with a small hole in between;
-   see mach_reserve_bootmem for details), so just use both as one big area.  */
-#define RAM_START 	SDRAM_ADDR
-#define RAM_END		(SDRAM_ADDR + SDRAM_SIZE)
-
-
-void __init mach_get_physical_ram (unsigned long *ram_start,
-				   unsigned long *ram_len)
-{
-	*ram_start = RAM_START;
-	*ram_len = RAM_END - RAM_START;
-}
-
-void mach_gettimeofday (struct timespec *tv)
-{
-	tv->tv_sec = 0;
-	tv->tv_nsec = 0;
-}
-
-/* Called before configuring an on-chip UART.  */
-void rte_me2_cb_uart_pre_configure (unsigned chan,
-				    unsigned cflags, unsigned baud)
-{
-	/* The RTE-V850E/ME2-CB connects some general-purpose I/O
-	   pins on the CPU to the RTS/CTS lines of UARTB channel 0's
-	   serial connection.
-	   I/O pins P21 and P22 are RTS and CTS respectively.  */
-	if (chan == 0) {
-		/* Put P21 & P22 in I/O port mode.  */
-		ME2_PORT2_PMC &= ~0x6;
-		/* Make P21 and output, and P22 an input.  */
-		ME2_PORT2_PM = (ME2_PORT2_PM & ~0xC) | 0x4;
-	}
-
-	me2_uart_pre_configure (chan, cflags, baud);
-}
-
-void __init mach_init_irqs (void)
-{
-	/* Initialize interrupts.  */
-	me2_init_irqs ();
-	rte_me2_cb_init_irqs ();
-}
-
-#ifdef CONFIG_ROM_KERNEL
-/* Initialization for kernel in ROM.  */
-static inline rom_kernel_init (void)
-{
-	/* If the kernel is in ROM, we have to copy any initialized data
-	   from ROM into RAM.  */
-	extern unsigned long _data_load_start, _sdata, _edata;
-	register unsigned long *src = &_data_load_start;
-	register unsigned long *dst = &_sdata, *end = &_edata;
-
-	while (dst != end)
-		*dst++ = *src++;
-}
-#endif /* CONFIG_ROM_KERNEL */
-
-static void install_interrupt_vectors (void)
-{
-	unsigned long *p1, *p2;
-
-	ME2_IRAMM = 0x03; /* V850E/ME2 iRAM write mode */
-
-	/* vector copy to iRAM */
-	p1 = (unsigned long *)0; /* v85x vector start */
-	p2 = (unsigned long *)&_intv_start;
-	while (p2 < (unsigned long *)&_intv_end)
-		*p1++ = *p2++;
-
-	ME2_IRAMM = 0x00; /* V850E/ME2 iRAM read mode */
-}
-
-/* CompactFlash */
-
-static void cf_power_on (void)
-{
-	/* CF card detected? */
-	if (CB_CF_STS0 & 0x0030)
-		return;
-
-	CB_CF_REG0 = 0x0002; /* reest on */
-	mdelay (10);
-	CB_CF_REG0 = 0x0003; /* power on */
-	mdelay (10);
-	CB_CF_REG0 = 0x0001; /* reset off */
-	mdelay (10);
-}
-
-static void cf_power_off (void)
-{
-	CB_CF_REG0 = 0x0003; /* power on */
-	mdelay (10);
-	CB_CF_REG0 = 0x0002; /* reest on */
-	mdelay (10);
-}
-
-void __init mach_early_init (void)
-{
-	install_interrupt_vectors ();
-
-	/* CS1 SDRAM instruction cache enable */
-	v850e_cache_enable (0x04, 0x03, 0);
-
-	rte_cb_early_init ();
-
-	/* CompactFlash power on */
-	cf_power_on ();
-
-#if defined (CONFIG_ROM_KERNEL)
-	rom_kernel_init ();
-#endif
-}
-
-
-/* RTE-V850E/ME2-CB Programmable Interrupt Controller.  */
-
-static struct cb_pic_irq_init cb_pic_irq_inits[] = {
-	{ "CB_EXTTM0",       IRQ_CB_EXTTM0,       1, 1, 6 },
-	{ "CB_EXTSIO",       IRQ_CB_EXTSIO,       1, 1, 6 },
-	{ "CB_TOVER",        IRQ_CB_TOVER,        1, 1, 6 },
-	{ "CB_GINT0",        IRQ_CB_GINT0,        1, 1, 6 },
-	{ "CB_USB",          IRQ_CB_USB,          1, 1, 6 },
-	{ "CB_LANC",         IRQ_CB_LANC,         1, 1, 6 },
-	{ "CB_USB_VBUS_ON",  IRQ_CB_USB_VBUS_ON,  1, 1, 6 },
-	{ "CB_USB_VBUS_OFF", IRQ_CB_USB_VBUS_OFF, 1, 1, 6 },
-	{ "CB_EXTTM1",       IRQ_CB_EXTTM1,       1, 1, 6 },
-	{ "CB_EXTTM2",       IRQ_CB_EXTTM2,       1, 1, 6 },
-	{ 0 }
-};
-#define NUM_CB_PIC_IRQ_INITS (ARRAY_SIZE(cb_pic_irq_inits) - 1)
-
-static struct hw_interrupt_type cb_pic_hw_itypes[NUM_CB_PIC_IRQ_INITS];
-static unsigned char cb_pic_active_irqs = 0;
-
-void __init rte_me2_cb_init_irqs (void)
-{
-	cb_pic_init_irq_types (cb_pic_irq_inits, cb_pic_hw_itypes);
-
-	/* Initalize on board PIC1 (not PIC0) enable */
-	CB_PIC_INT0M  = 0x0000;
-	CB_PIC_INT1M  = 0x0000;
-	CB_PIC_INTR   = 0x0000;
-	CB_PIC_INTEN |= CB_PIC_INT1EN;
-
-	ME2_PORT2_PMC 	 |= 0x08;	/* INTP23/SCK1 mode */
-	ME2_PORT2_PFC 	 &= ~0x08;	/* INTP23 mode */
-	ME2_INTR(2) 	 &= ~0x08;	/* INTP23 falling-edge detect */
-	ME2_INTF(2) 	 &= ~0x08;	/*   " */
-
-	rte_cb_init_irqs ();	/* gbus &c */
-}
-
-
-/* Enable interrupt handling for interrupt IRQ.  */
-void cb_pic_enable_irq (unsigned irq)
-{
-	CB_PIC_INT1M |= 1 << (irq - CB_PIC_BASE_IRQ);
-}
-
-void cb_pic_disable_irq (unsigned irq)
-{
-	CB_PIC_INT1M &= ~(1 << (irq - CB_PIC_BASE_IRQ));
-}
-
-void cb_pic_shutdown_irq (unsigned irq)
-{
-	cb_pic_disable_irq (irq);
-
-	if (--cb_pic_active_irqs == 0)
-		free_irq (IRQ_CB_PIC, 0);
-
-	CB_PIC_INT1M &= ~(1 << (irq - CB_PIC_BASE_IRQ));
-}
-
-static irqreturn_t cb_pic_handle_irq (int irq, void *dev_id,
-				      struct pt_regs *regs)
-{
-	irqreturn_t rval = IRQ_NONE;
-	unsigned status = CB_PIC_INTR;
-	unsigned enable = CB_PIC_INT1M;
-
-	/* Only pay attention to enabled interrupts.  */
-	status &= enable;
-
-	CB_PIC_INTEN &= ~CB_PIC_INT1EN;
-
-	if (status) {
-		unsigned mask = 1;
-
-		irq = CB_PIC_BASE_IRQ;
-		do {
-			/* There's an active interrupt, find out which one,
-			   and call its handler.  */
-			while (! (status & mask)) {
-				irq++;
-				mask <<= 1;
-			}
-			status &= ~mask;
-
-			CB_PIC_INTR = mask;
-
-			/* Recursively call handle_irq to handle it. */
-			handle_irq (irq, regs);
-			rval = IRQ_HANDLED;
-		} while (status);
-	}
-
-	CB_PIC_INTEN |= CB_PIC_INT1EN;
-
-	return rval;
-}
-
-
-static void irq_nop (unsigned irq) { }
-
-static unsigned cb_pic_startup_irq (unsigned irq)
-{
-	int rval;
-
-	if (cb_pic_active_irqs == 0) {
-		rval = request_irq (IRQ_CB_PIC, cb_pic_handle_irq,
-				    IRQF_DISABLED, "cb_pic_handler", 0);
-		if (rval != 0)
-			return rval;
-	}
-
-	cb_pic_active_irqs++;
-
-	cb_pic_enable_irq (irq);
-
-	return 0;
-}
-
-/* Initialize HW_IRQ_TYPES for INTC-controlled irqs described in array
-   INITS (which is terminated by an entry with the name field == 0).  */
-void __init cb_pic_init_irq_types (struct cb_pic_irq_init *inits,
-				   struct hw_interrupt_type *hw_irq_types)
-{
-	struct cb_pic_irq_init *init;
-	for (init = inits; init->name; init++) {
-		struct hw_interrupt_type *hwit = hw_irq_types++;
-
-		hwit->typename = init->name;
-
-		hwit->startup  = cb_pic_startup_irq;
-		hwit->shutdown = cb_pic_shutdown_irq;
-		hwit->enable   = cb_pic_enable_irq;
-		hwit->disable  = cb_pic_disable_irq;
-		hwit->ack      = irq_nop;
-		hwit->end      = irq_nop;
-
-		/* Initialize kernel IRQ infrastructure for this interrupt.  */
-		init_irq_handlers(init->base, init->num, init->interval, hwit);
-	}
-}
diff --git a/arch/v850/kernel/rte_me2_cb.ld b/arch/v850/kernel/rte_me2_cb.ld
deleted file mode 100644
index cf0766065ec6..000000000000
--- a/arch/v850/kernel/rte_me2_cb.ld
+++ /dev/null
@@ -1,30 +0,0 @@
-/* Linker script for the Midas labs RTE-V850E/ME2-CB evaluation board
-   (CONFIG_RTE_CB_ME2), with kernel in SDRAM.  */
-
-MEMORY {
-	/* 128Kbyte of IRAM */
-	IRAM : ORIGIN = 0x00000000, LENGTH = 0x00020000
-
-	/* 32MB of SDRAM.  */
-	SDRAM : ORIGIN = SDRAM_ADDR, LENGTH = SDRAM_SIZE
-}
-
-#define KRAM SDRAM
-
-SECTIONS {
-	.text : {
-		__kram_start = . ;
-		TEXT_CONTENTS
-		INTV_CONTENTS	/* copy to iRAM (0x0-0x620) */
-	} > KRAM
-
-	.data : {
-		DATA_CONTENTS
-		BSS_CONTENTS
-		RAMK_INIT_CONTENTS
-		__kram_end = . ;
-		BOOTMAP_CONTENTS
-	} > KRAM
-	
-	.root ALIGN (4096) : { ROOT_FS_CONTENTS } > SDRAM
-}
diff --git a/arch/v850/kernel/rte_nb85e_cb-multi.ld b/arch/v850/kernel/rte_nb85e_cb-multi.ld
deleted file mode 100644
index de347b4fffac..000000000000
--- a/arch/v850/kernel/rte_nb85e_cb-multi.ld
+++ /dev/null
@@ -1,57 +0,0 @@
-/* Linker script for the Midas labs RTE-NB85E-CB evaluation board
-   (CONFIG_RTE_CB_NB85E), with the Multi debugger ROM monitor .  */
-
-MEMORY {
-	/* 1MB of SRAM; we can't use the last 96KB, because it's used by
-	   the monitor scratch-RAM.  This memory is mirrored 4 times.  */
-	SRAM  : ORIGIN = SRAM_ADDR,  LENGTH = (SRAM_SIZE - MON_SCRATCH_SIZE)
-	/* Monitor scratch RAM; only the interrupt vectors should go here.  */
-	MRAM  : ORIGIN = MON_SCRATCH_ADDR,  LENGTH = MON_SCRATCH_SIZE
-	/* 16MB of SDRAM.  */
-	SDRAM : ORIGIN = SDRAM_ADDR, LENGTH = SDRAM_SIZE
-}
-
-#ifdef CONFIG_RTE_CB_NB85E_KSRAM
-# define KRAM SRAM
-#else
-# define KRAM SDRAM
-#endif
-
-SECTIONS {
-	/* We can't use RAMK_KRAM_CONTENTS because that puts the whole
-	   kernel in a single ELF segment, and the Multi debugger (which
-	   we use to load the kernel) appears to have bizarre problems
-	   dealing with it.  */
-
-	.text : {
-		__kram_start = . ;
-		TEXT_CONTENTS
-	} > KRAM
-
-	.data : {
-		DATA_CONTENTS
-		BSS_CONTENTS
-		RAMK_INIT_CONTENTS
-		__kram_end = . ;
-		BOOTMAP_CONTENTS
-
-		/* The address at which the interrupt vectors are initially
-		   loaded by the loader.  We can't load the interrupt vectors
-		   directly into their target location, because the monitor
-		   ROM for the GHS Multi debugger barfs if we try.
-		   Unfortunately, Multi also doesn't deal correctly with ELF
-		   sections where the LMA and VMA differ (it just ignores the
-		   LMA), so we can't use that feature to work around the
-		   problem!  What we do instead is just put the interrupt
-		   vectors into a normal section, and have the
-		   `mach_early_init' function for Midas boards do the
-		   necessary copying and relocation at runtime (this section
-		   basically only contains `jr' instructions, so it's not
-		   that hard).  */
-		. = ALIGN (0x10) ;
-		__intv_load_start = . ;
-		INTV_CONTENTS
-	} > KRAM
-
-	.root ALIGN (4096) : { ROOT_FS_CONTENTS } > SDRAM
-}
diff --git a/arch/v850/kernel/rte_nb85e_cb.c b/arch/v850/kernel/rte_nb85e_cb.c
deleted file mode 100644
index b4a045da5d70..000000000000
--- a/arch/v850/kernel/rte_nb85e_cb.c
+++ /dev/null
@@ -1,81 +0,0 @@
-/*
- * arch/v850/kernel/rte_nb85e_cb.c -- Midas labs RTE-V850E/NB85E-CB board
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/kernel.h>
-#include <linux/init.h>
-#include <linux/mm.h>
-#include <linux/swap.h>
-#include <linux/bootmem.h>
-#include <linux/irq.h>
-
-#include <asm/atomic.h>
-#include <asm/page.h>
-#include <asm/v850e.h>
-#include <asm/rte_nb85e_cb.h>
-
-#include "mach.h"
-
-void __init mach_early_init (void)
-{
-	/* Configure caching; some possible settings:
-
-	     BHC = 0x0000, DCC = 0x0000	 -- all caching disabled
-	     BHC = 0x0040, DCC = 0x0000	 -- SDRAM: icache only
-	     BHC = 0x0080, DCC = 0x0C00	 -- SDRAM: write-back dcache only
-	     BHC = 0x00C0, DCC = 0x0C00	 -- SDRAM: icache + write-back dcache
-	     BHC = 0x00C0, DCC = 0x0800	 -- SDRAM: icache + write-thru dcache
-
-	   We can only cache SDRAM (we can't use cache SRAM because it's in
-	   the same memory region as the on-chip RAM and I/O space).
-
-	   Unfortunately, the dcache seems to be buggy, so we only use the
-	   icache for now.  */
-	v850e_cache_enable (0x0040 /*BHC*/, 0x0003 /*ICC*/, 0x0000 /*DCC*/);
-
-	rte_cb_early_init ();
-}
-
-void __init mach_get_physical_ram (unsigned long *ram_start,
-				   unsigned long *ram_len)
-{
-	/* We just use SDRAM here.  */
-	*ram_start = SDRAM_ADDR;
-	*ram_len = SDRAM_SIZE;
-}
-
-void mach_gettimeofday (struct timespec *tv)
-{
-	tv->tv_sec = 0;
-	tv->tv_nsec = 0;
-}
-
-/* Called before configuring an on-chip UART.  */
-void rte_nb85e_cb_uart_pre_configure (unsigned chan,
-				    unsigned cflags, unsigned baud)
-{
-	/* The RTE-NB85E-CB connects some general-purpose I/O pins on the
-	   CPU to the RTS/CTS lines the UART's serial connection, as follows:
-	   P00 = CTS (in), P01 = DSR (in), P02 = RTS (out), P03 = DTR (out). */
-
-	TEG_PORT0_PM = 0x03;	/* P00 and P01 inputs, P02 and P03 outputs */
-	TEG_PORT0_IO = 0x03;	/* Accept input */
-
-	/* Do pre-configuration for the actual UART.  */
-	teg_uart_pre_configure (chan, cflags, baud);
-}
-
-void __init mach_init_irqs (void)
-{
-	teg_init_irqs ();
-	rte_cb_init_irqs ();
-}
diff --git a/arch/v850/kernel/rte_nb85e_cb.ld b/arch/v850/kernel/rte_nb85e_cb.ld
deleted file mode 100644
index b672f484f085..000000000000
--- a/arch/v850/kernel/rte_nb85e_cb.ld
+++ /dev/null
@@ -1,22 +0,0 @@
-/* Linker script for the Midas labs RTE-NB85E-CB evaluation board
-   (CONFIG_RTE_CB_NB85E).  */
-
-MEMORY {
-	LOW   : ORIGIN = 0x0,	     LENGTH = 0x00100000
-	/* 1MB of SRAM  This memory is mirrored 4 times.  */
-	SRAM  : ORIGIN = SRAM_ADDR,  LENGTH = SRAM_SIZE
-	/* 16MB of SDRAM.  */
-	SDRAM : ORIGIN = SDRAM_ADDR, LENGTH = SDRAM_SIZE
-}
-
-#ifdef CONFIG_RTE_CB_NB85E_KSRAM
-# define KRAM SRAM
-#else
-# define KRAM SDRAM
-#endif
-
-SECTIONS {
-	.intv : { INTV_CONTENTS } > LOW
-	.sram : { RAMK_KRAM_CONTENTS } > KRAM
-	.root : { ROOT_FS_CONTENTS } > SDRAM
-}
diff --git a/arch/v850/kernel/setup.c b/arch/v850/kernel/setup.c
deleted file mode 100644
index 10335cecf7bd..000000000000
--- a/arch/v850/kernel/setup.c
+++ /dev/null
@@ -1,329 +0,0 @@
-/*
- * arch/v850/kernel/setup.c -- Arch-dependent initialization functions
- *
- *  Copyright (C) 2001,02,03,05,06  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03,05,06  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/mm.h>
-#include <linux/bootmem.h>
-#include <linux/swap.h>		/* we don't have swap, but for nr_free_pages */
-#include <linux/irq.h>
-#include <linux/reboot.h>
-#include <linux/personality.h>
-#include <linux/major.h>
-#include <linux/root_dev.h>
-#include <linux/mtd/mtd.h>
-#include <linux/init.h>
-
-#include <asm/irq.h>
-#include <asm/setup.h>
-
-#include "mach.h"
-
-/* These symbols are all defined in the linker map to delineate various
-   statically allocated regions of memory.  */
-
-extern char _intv_start, _intv_end;
-/* `kram' is only used if the kernel uses part of normal user RAM.  */
-extern char _kram_start __attribute__ ((__weak__));
-extern char _kram_end __attribute__ ((__weak__));
-extern char _init_start, _init_end;
-extern char _bootmap;
-extern char _stext, _etext, _sdata, _edata, _sbss, _ebss;
-/* Many platforms use an embedded root image.  */
-extern char _root_fs_image_start __attribute__ ((__weak__));
-extern char _root_fs_image_end __attribute__ ((__weak__));
-
-
-char __initdata command_line[COMMAND_LINE_SIZE];
-
-/* Memory not used by the kernel.  */
-static unsigned long total_ram_pages;
-
-/* System RAM.  */
-static unsigned long ram_start = 0, ram_len = 0;
-
-
-#define ADDR_TO_PAGE_UP(x)   ((((unsigned long)x) + PAGE_SIZE-1) >> PAGE_SHIFT)
-#define ADDR_TO_PAGE(x)	     (((unsigned long)x) >> PAGE_SHIFT)
-#define PAGE_TO_ADDR(x)	     (((unsigned long)x) << PAGE_SHIFT)
-
-static void init_mem_alloc (unsigned long ram_start, unsigned long ram_len);
-
-void set_mem_root (void *addr, size_t len, char *cmd_line);
-
-
-void __init setup_arch (char **cmdline)
-{
-	/* Keep a copy of command line */
-	*cmdline = command_line;
-	memcpy (boot_command_line, command_line, COMMAND_LINE_SIZE);
-	boot_command_line[COMMAND_LINE_SIZE - 1] = '\0';
-
-	console_verbose ();
-
-	init_mm.start_code = (unsigned long) &_stext;
-	init_mm.end_code = (unsigned long) &_etext;
-	init_mm.end_data = (unsigned long) &_edata;
-	init_mm.brk = (unsigned long) &_kram_end;
-
-	/* Find out what mem this machine has.  */
-	mach_get_physical_ram (&ram_start, &ram_len);
-	/* ... and tell the kernel about it.  */
-	init_mem_alloc (ram_start, ram_len);
-
-	printk (KERN_INFO "CPU: %s\nPlatform: %s\n",
-		CPU_MODEL_LONG, PLATFORM_LONG);
-
-	/* do machine-specific setups.  */
-	mach_setup (cmdline);
-
-#ifdef CONFIG_MTD
-	if (!ROOT_DEV && &_root_fs_image_end > &_root_fs_image_start)
-		set_mem_root (&_root_fs_image_start,
-			      &_root_fs_image_end - &_root_fs_image_start,
-			      *cmdline);
-#endif
-}
-
-void __init trap_init (void)
-{
-}
-
-#ifdef CONFIG_MTD
-
-/* From drivers/mtd/devices/slram.c */
-#define SLRAM_BLK_SZ 0x4000
-
-/* Set the root filesystem to be the given memory region.
-   Some parameter may be appended to CMD_LINE.  */
-void set_mem_root (void *addr, size_t len, char *cmd_line)
-{
-	/* Some sort of idiocy in MTD means we must supply a length that's
-	   a multiple of SLRAM_BLK_SZ.  We just round up the real length,
-	   as the file system shouldn't attempt to access anything beyond
-	   the end of the image anyway.  */
-	len = (((len - 1) + SLRAM_BLK_SZ) / SLRAM_BLK_SZ) * SLRAM_BLK_SZ;
-
-	/* The only way to pass info to the MTD slram driver is via
-	   the command line.  */
-	if (*cmd_line) {
-		cmd_line += strlen (cmd_line);
-		*cmd_line++ = ' ';
-	}
-	sprintf (cmd_line, "slram=root,0x%x,+0x%x", (u32)addr, (u32)len);
-
-	ROOT_DEV = MKDEV (MTD_BLOCK_MAJOR, 0);
-}
-#endif
-
-
-static void irq_nop (unsigned irq) { }
-static unsigned irq_zero (unsigned irq) { return 0; }
-
-static void nmi_end (unsigned irq)
-{
-	if (irq != IRQ_NMI (0)) {
-		printk (KERN_CRIT "NMI %d is unrecoverable; restarting...",
-			irq - IRQ_NMI (0));
-		machine_restart (0);
-	}
-}
-
-static struct hw_interrupt_type nmi_irq_type = {
-	.typename = "NMI",
-	.startup = irq_zero,		/* startup */
-	.shutdown = irq_nop,		/* shutdown */
-	.enable = irq_nop,		/* enable */
-	.disable = irq_nop,		/* disable */
-	.ack = irq_nop,		/* ack */
-	.end = nmi_end,		/* end */
-};
-
-void __init init_IRQ (void)
-{
-	init_irq_handlers (0, NUM_MACH_IRQS, 1, 0);
-	init_irq_handlers (IRQ_NMI (0), NUM_NMIS, 1, &nmi_irq_type);
-	mach_init_irqs ();
-}
-
-
-void __init mem_init (void)
-{
-	max_mapnr = MAP_NR (ram_start + ram_len);
-
-	num_physpages = ADDR_TO_PAGE (ram_len);
-
-	total_ram_pages = free_all_bootmem ();
-
-	printk (KERN_INFO
-		"Memory: %luK/%luK available"
-		" (%luK kernel code, %luK data)\n",
-		PAGE_TO_ADDR (nr_free_pages()) / 1024,
-		ram_len / 1024,
-		((unsigned long)&_etext - (unsigned long)&_stext) / 1024,
-		((unsigned long)&_ebss - (unsigned long)&_sdata) / 1024);
-}
-
-void free_initmem (void)
-{
-	unsigned long ram_end = ram_start + ram_len;
-	unsigned long start = PAGE_ALIGN ((unsigned long)(&_init_start));
-
-	if (start >= ram_start && start < ram_end) {
-		unsigned long addr;
-		unsigned long end = PAGE_ALIGN ((unsigned long)(&_init_end));
-
-		if (end > ram_end)
-			end = ram_end;
-
-		printk("Freeing unused kernel memory: %ldK freed\n",
-		       (end - start) / 1024);
-
-		for (addr = start; addr < end; addr += PAGE_SIZE) {
-			struct page *page = virt_to_page (addr);
-			ClearPageReserved (page);
-			init_page_count (page);
-			__free_page (page);
-			total_ram_pages++;
-		}
-	}
-}
-
-
-/* Initialize the `bootmem allocator'.  RAM_START and RAM_LEN identify
-   what RAM may be used.  */
-static void __init
-init_bootmem_alloc (unsigned long ram_start, unsigned long ram_len)
-{
-	/* The part of the kernel that's in the same managed RAM space
-	   used for general allocation.  */
-	unsigned long kram_start = (unsigned long)&_kram_start;
-	unsigned long kram_end = (unsigned long)&_kram_end;
-	/* End of the managed RAM space.  */
-	unsigned long ram_end = ram_start + ram_len;
-	/* Address range of the interrupt vector table.  */
-	unsigned long intv_start = (unsigned long)&_intv_start;
-	unsigned long intv_end = (unsigned long)&_intv_end;
-	/* True if the interrupt vectors are in the managed RAM area.  */
-	int intv_in_ram = (intv_end > ram_start && intv_start < ram_end);
-	/* True if the interrupt vectors are inside the kernel's RAM.  */
-	int intv_in_kram = (intv_end > kram_start && intv_start < kram_end);
-	/* A pointer to an optional function that reserves platform-specific
-	   memory regions.  We declare the pointer `volatile' to avoid gcc
-	   turning the call into a static call (the problem is that since
-	   it's a weak symbol, a static call may end up trying to reference
-	   the location 0x0, which is not always reachable).  */
-	void (*volatile mrb) (void) = mach_reserve_bootmem;
-	/* The bootmem allocator's allocation bitmap.  */
-	unsigned long bootmap = (unsigned long)&_bootmap;
-	unsigned long bootmap_len;
-
-	/* Round bootmap location up to next page.  */
-	bootmap = PAGE_TO_ADDR (ADDR_TO_PAGE_UP (bootmap));
-
-	/* Initialize bootmem allocator.  */
-	bootmap_len = init_bootmem_node (NODE_DATA (0),
-					 ADDR_TO_PAGE (bootmap),
-					 ADDR_TO_PAGE (PAGE_OFFSET),
-					 ADDR_TO_PAGE (ram_end));
-
-	/* Now make the RAM actually allocatable (it starts out `reserved'). */
-	free_bootmem (ram_start, ram_len);
-
-	if (kram_end > kram_start)
-		/* Reserve the RAM part of the kernel's address space, so it
-		   doesn't get allocated.  */
-		reserve_bootmem(kram_start, kram_end - kram_start,
-				BOOTMEM_DEFAULT);
-	
-	if (intv_in_ram && !intv_in_kram)
-		/* Reserve the interrupt vector space.  */
-		reserve_bootmem(intv_start, intv_end - intv_start,
-				BOOTMEM_DEFAULT);
-
-	if (bootmap >= ram_start && bootmap < ram_end)
-		/* Reserve the bootmap space.  */
-		reserve_bootmem(bootmap, bootmap_len,
-				BOOTMEM_DEFAULT);
-
-	/* Reserve the memory used by the root filesystem image if it's
-	   in RAM.  */
-	if (&_root_fs_image_end > &_root_fs_image_start
-	    && (unsigned long)&_root_fs_image_start >= ram_start
-	    && (unsigned long)&_root_fs_image_start < ram_end)
-		reserve_bootmem ((unsigned long)&_root_fs_image_start,
-				 &_root_fs_image_end - &_root_fs_image_start,
-				 BOOTMEM_DEFAULT);
-
-	/* Let the platform-dependent code reserve some too.  */
-	if (mrb)
-		(*mrb) ();
-}
-
-/* Tell the kernel about what RAM it may use for memory allocation.  */
-static void __init
-init_mem_alloc (unsigned long ram_start, unsigned long ram_len)
-{
-	unsigned i;
-	unsigned long zones_size[MAX_NR_ZONES];
-
-	init_bootmem_alloc (ram_start, ram_len);
-
-	for (i = 0; i < MAX_NR_ZONES; i++)
-		zones_size[i] = 0;
-
-	/* We stuff all the memory into one area, which includes the
-	   initial gap from PAGE_OFFSET to ram_start.  */
-	zones_size[ZONE_DMA]
-		= ADDR_TO_PAGE (ram_len + (ram_start - PAGE_OFFSET));
-
-	/* The allocator is very picky about the address of the first
-	   allocatable page -- it must be at least as aligned as the
-	   maximum allocation -- so try to detect cases where it will get
-	   confused and signal them at compile time (this is a common
-	   problem when porting to a new platform with ).  There is a
-	   similar runtime check in free_area_init_core.  */
-#if ((PAGE_OFFSET >> PAGE_SHIFT) & ((1UL << (MAX_ORDER - 1)) - 1))
-#error MAX_ORDER is too large for given PAGE_OFFSET (use CONFIG_FORCE_MAX_ZONEORDER to change it)
-#endif
-	NODE_DATA(0)->node_mem_map = NULL;
-	free_area_init_node(0, zones_size, ADDR_TO_PAGE (PAGE_OFFSET), 0);
-}
-
-
-
-/* Taken from m68knommu */
-void show_mem(void)
-{
-    unsigned long i;
-    int free = 0, total = 0, reserved = 0, shared = 0;
-    int cached = 0;
-
-    printk(KERN_INFO "\nMem-info:\n");
-    show_free_areas();
-    i = max_mapnr;
-    while (i-- > 0) {
-	total++;
-	if (PageReserved(mem_map+i))
-	    reserved++;
-	else if (PageSwapCache(mem_map+i))
-	    cached++;
-	else if (!page_count(mem_map+i))
-	    free++;
-	else
-	    shared += page_count(mem_map+i) - 1;
-    }
-    printk(KERN_INFO "%d pages of RAM\n",total);
-    printk(KERN_INFO "%d free pages\n",free);
-    printk(KERN_INFO "%d reserved pages\n",reserved);
-    printk(KERN_INFO "%d pages shared\n",shared);
-    printk(KERN_INFO "%d pages swap cached\n",cached);
-}
diff --git a/arch/v850/kernel/signal.c b/arch/v850/kernel/signal.c
deleted file mode 100644
index bf166e7e762c..000000000000
--- a/arch/v850/kernel/signal.c
+++ /dev/null
@@ -1,523 +0,0 @@
-/*
- * arch/v850/kernel/signal.c -- Signal handling
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *  Copyright (C) 1999,2000,2002  Niibe Yutaka & Kaz Kojima
- *  Copyright (C) 1991,1992  Linus Torvalds
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * 1997-11-28  Modified for POSIX.1b signals by Richard Henderson
- *
- * This file was derived from the sh version, arch/sh/kernel/signal.c
- */
-
-#include <linux/mm.h>
-#include <linux/smp.h>
-#include <linux/kernel.h>
-#include <linux/signal.h>
-#include <linux/errno.h>
-#include <linux/wait.h>
-#include <linux/ptrace.h>
-#include <linux/unistd.h>
-#include <linux/stddef.h>
-#include <linux/personality.h>
-#include <linux/tty.h>
-
-#include <asm/ucontext.h>
-#include <asm/uaccess.h>
-#include <asm/pgtable.h>
-#include <asm/pgalloc.h>
-#include <asm/thread_info.h>
-#include <asm/cacheflush.h>
-
-#define DEBUG_SIG 0
-
-#define _BLOCKABLE (~(sigmask(SIGKILL) | sigmask(SIGSTOP)))
-
-asmlinkage int do_signal(struct pt_regs *regs, sigset_t *oldset);
-
-/*
- * Atomically swap in the new signal mask, and wait for a signal.
- */
-asmlinkage int
-sys_sigsuspend(old_sigset_t mask, struct pt_regs *regs)
-{
-	sigset_t saveset;
-
-	mask &= _BLOCKABLE;
-	spin_lock_irq(&current->sighand->siglock);
-	saveset = current->blocked;
-	siginitset(&current->blocked, mask);
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-
-	regs->gpr[GPR_RVAL] = -EINTR;
-	while (1) {
-		current->state = TASK_INTERRUPTIBLE;
-		schedule();
-		if (do_signal(regs, &saveset))
-			return -EINTR;
-	}
-}
-
-asmlinkage int
-sys_rt_sigsuspend(sigset_t *unewset, size_t sigsetsize,
-		  struct pt_regs *regs)
-{
-	sigset_t saveset, newset;
-
-	/* XXX: Don't preclude handling different sized sigset_t's.  */
-	if (sigsetsize != sizeof(sigset_t))
-		return -EINVAL;
-
-	if (copy_from_user(&newset, unewset, sizeof(newset)))
-		return -EFAULT;
-	sigdelsetmask(&newset, ~_BLOCKABLE);
-	spin_lock_irq(&current->sighand->siglock);
-	saveset = current->blocked;
-	current->blocked = newset;
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-
-	regs->gpr[GPR_RVAL] = -EINTR;
-	while (1) {
-		current->state = TASK_INTERRUPTIBLE;
-		schedule();
-		if (do_signal(regs, &saveset))
-			return -EINTR;
-	}
-}
-
-asmlinkage int 
-sys_sigaction(int sig, const struct old_sigaction *act,
-	      struct old_sigaction *oact)
-{
-	struct k_sigaction new_ka, old_ka;
-	int ret;
-
-	if (act) {
-		old_sigset_t mask;
-		if (!access_ok(VERIFY_READ, act, sizeof(*act)) ||
-		    __get_user(new_ka.sa.sa_handler, &act->sa_handler) ||
-		    __get_user(new_ka.sa.sa_restorer, &act->sa_restorer))
-			return -EFAULT;
-		__get_user(new_ka.sa.sa_flags, &act->sa_flags);
-		__get_user(mask, &act->sa_mask);
-		siginitset(&new_ka.sa.sa_mask, mask);
-	}
-
-	ret = do_sigaction(sig, act ? &new_ka : NULL, oact ? &old_ka : NULL);
-
-	if (!ret && oact) {
-		if (!access_ok(VERIFY_WRITE, oact, sizeof(*oact)) ||
-		    __put_user(old_ka.sa.sa_handler, &oact->sa_handler) ||
-		    __put_user(old_ka.sa.sa_restorer, &oact->sa_restorer))
-			return -EFAULT;
-		__put_user(old_ka.sa.sa_flags, &oact->sa_flags);
-		__put_user(old_ka.sa.sa_mask.sig[0], &oact->sa_mask);
-	}
-
-	return ret;
-}
-
-asmlinkage int
-sys_sigaltstack(const stack_t *uss, stack_t *uoss,
-		struct pt_regs *regs)
-{
-	return do_sigaltstack(uss, uoss, regs->gpr[GPR_SP]);
-}
-
-
-/*
- * Do a signal return; undo the signal stack.
- */
-
-struct sigframe
-{
-	struct sigcontext sc;
-	unsigned long extramask[_NSIG_WORDS-1];
-	unsigned long tramp[2];	/* signal trampoline */
-};
-
-struct rt_sigframe
-{
-	struct siginfo info;
-	struct ucontext uc;
-	unsigned long tramp[2];	/* signal trampoline */
-};
-
-static int
-restore_sigcontext(struct pt_regs *regs, struct sigcontext *sc, int *rval_p)
-{
-	unsigned int err = 0;
-
-#define COPY(x)		err |= __get_user(regs->x, &sc->regs.x)
-	COPY(gpr[0]);	COPY(gpr[1]);	COPY(gpr[2]);	COPY(gpr[3]);
-	COPY(gpr[4]);	COPY(gpr[5]);	COPY(gpr[6]);	COPY(gpr[7]);
-	COPY(gpr[8]);	COPY(gpr[9]);	COPY(gpr[10]);	COPY(gpr[11]);
-	COPY(gpr[12]);	COPY(gpr[13]);	COPY(gpr[14]);	COPY(gpr[15]);
-	COPY(gpr[16]);	COPY(gpr[17]);	COPY(gpr[18]);	COPY(gpr[19]);
-	COPY(gpr[20]);	COPY(gpr[21]);	COPY(gpr[22]);	COPY(gpr[23]);
-	COPY(gpr[24]);	COPY(gpr[25]);	COPY(gpr[26]);	COPY(gpr[27]);
-	COPY(gpr[28]);	COPY(gpr[29]);	COPY(gpr[30]);	COPY(gpr[31]);
-	COPY(pc);	COPY(psw);
-	COPY(ctpc);	COPY(ctpsw);	COPY(ctbp);
-#undef COPY
-
-	return err;
-}
-
-asmlinkage int sys_sigreturn(struct pt_regs *regs)
-{
-	struct sigframe *frame = (struct sigframe *)regs->gpr[GPR_SP];
-	sigset_t set;
-	int rval;
-
-	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
-		goto badframe;
-
-	if (__get_user(set.sig[0], &frame->sc.oldmask)
-	    || (_NSIG_WORDS > 1
-		&& __copy_from_user(&set.sig[1], &frame->extramask,
-				    sizeof(frame->extramask))))
-		goto badframe;
-
-	sigdelsetmask(&set, ~_BLOCKABLE);
-	spin_lock_irq(&current->sighand->siglock);
-	current->blocked = set;
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-
-	if (restore_sigcontext(regs, &frame->sc, &rval))
-		goto badframe;
-	return rval;
-
-badframe:
-	force_sig(SIGSEGV, current);
-	return 0;
-}
-
-asmlinkage int sys_rt_sigreturn(struct pt_regs *regs)
-{
-	struct rt_sigframe *frame = (struct rt_sigframe *)regs->gpr[GPR_SP];
-	sigset_t set;
-	stack_t st;
-	int rval;
-
-	if (!access_ok(VERIFY_READ, frame, sizeof(*frame)))
-		goto badframe;
-
-	if (__copy_from_user(&set, &frame->uc.uc_sigmask, sizeof(set)))
-		goto badframe;
-
-	sigdelsetmask(&set, ~_BLOCKABLE);
-	spin_lock_irq(&current->sighand->siglock);
-	current->blocked = set;
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-
-	if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &rval))
-		goto badframe;
-
-	if (__copy_from_user(&st, &frame->uc.uc_stack, sizeof(st)))
-		goto badframe;
-	/* It is more difficult to avoid calling this function than to
-	   call it and ignore errors.  */
-	do_sigaltstack(&st, NULL, regs->gpr[GPR_SP]);
-
-	return rval;
-
-badframe:
-	force_sig(SIGSEGV, current);
-	return 0;
-}	
-
-/*
- * Set up a signal frame.
- */
-
-static int
-setup_sigcontext(struct sigcontext *sc, struct pt_regs *regs,
-		 unsigned long mask)
-{
-	int err = 0;
-
-#define COPY(x)		err |= __put_user(regs->x, &sc->regs.x)
-	COPY(gpr[0]);	COPY(gpr[1]);	COPY(gpr[2]);	COPY(gpr[3]);
-	COPY(gpr[4]);	COPY(gpr[5]);	COPY(gpr[6]);	COPY(gpr[7]);
-	COPY(gpr[8]);	COPY(gpr[9]);	COPY(gpr[10]);	COPY(gpr[11]);
-	COPY(gpr[12]);	COPY(gpr[13]);	COPY(gpr[14]);	COPY(gpr[15]);
-	COPY(gpr[16]);	COPY(gpr[17]);	COPY(gpr[18]);	COPY(gpr[19]);
-	COPY(gpr[20]);	COPY(gpr[21]);	COPY(gpr[22]);	COPY(gpr[23]);
-	COPY(gpr[24]);	COPY(gpr[25]);	COPY(gpr[26]);	COPY(gpr[27]);
-	COPY(gpr[28]);	COPY(gpr[29]);	COPY(gpr[30]);	COPY(gpr[31]);
-	COPY(pc);	COPY(psw);
-	COPY(ctpc);	COPY(ctpsw);	COPY(ctbp);
-#undef COPY
-
-	err |= __put_user(mask, &sc->oldmask);
-
-	return err;
-}
-
-/*
- * Determine which stack to use..
- */
-static inline void *
-get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size)
-{
-	/* Default to using normal stack */
-	unsigned long sp = regs->gpr[GPR_SP];
-
-	if ((ka->sa.sa_flags & SA_ONSTACK) != 0 && ! sas_ss_flags(sp))
-		sp = current->sas_ss_sp + current->sas_ss_size;
-
-	return (void *)((sp - frame_size) & -8UL);
-}
-
-static void setup_frame(int sig, struct k_sigaction *ka,
-			sigset_t *set, struct pt_regs *regs)
-{
-	struct sigframe *frame;
-	int err = 0;
-	int signal;
-
-	frame = get_sigframe(ka, regs, sizeof(*frame));
-
-	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
-		goto give_sigsegv;
-
-	signal = current_thread_info()->exec_domain
-		&& current_thread_info()->exec_domain->signal_invmap
-		&& sig < 32
-		? current_thread_info()->exec_domain->signal_invmap[sig]
-		: sig;
-
-	err |= setup_sigcontext(&frame->sc, regs, set->sig[0]);
-
-	if (_NSIG_WORDS > 1) {
-		err |= __copy_to_user(frame->extramask, &set->sig[1],
-				      sizeof(frame->extramask));
-	}
-
-	/* Set up to return from userspace.  If provided, use a stub
-	   already in userspace.  */
-	if (ka->sa.sa_flags & SA_RESTORER) {
-		regs->gpr[GPR_LP] = (unsigned long) ka->sa.sa_restorer;
-	} else {
-		/* Note, these encodings are _little endian_!  */
-
-		/* addi  __NR_sigreturn, r0, r12  */
-		err |= __put_user(0x6600 | (__NR_sigreturn << 16),
-				  frame->tramp + 0);
-		/* trap 0 */
-		err |= __put_user(0x010007e0,
-				  frame->tramp + 1);
-
-		regs->gpr[GPR_LP] = (unsigned long)frame->tramp;
-
-		flush_cache_sigtramp (regs->gpr[GPR_LP]);
-	}
-
-	if (err)
-		goto give_sigsegv;
-
-	/* Set up registers for signal handler.  */
-	regs->pc = (v850_reg_t) ka->sa.sa_handler;
-	regs->gpr[GPR_SP] = (v850_reg_t)frame;
-	/* Signal handler args:  */
-	regs->gpr[GPR_ARG0] = signal; /* arg 0: signum */
-	regs->gpr[GPR_ARG1] = (v850_reg_t)&frame->sc;/* arg 1: sigcontext */
-
-	set_fs(USER_DS);
-
-#if DEBUG_SIG
-	printk("SIG deliver (%s:%d): sp=%p pc=%08lx ra=%08lx\n",
-		current->comm, current->pid, frame, regs->pc, );
-#endif
-
-	return;
-
-give_sigsegv:
-	force_sigsegv(sig, current);
-}
-
-static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
-			   sigset_t *set, struct pt_regs *regs)
-{
-	struct rt_sigframe *frame;
-	int err = 0;
-	int signal;
-
-	frame = get_sigframe(ka, regs, sizeof(*frame));
-
-	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
-		goto give_sigsegv;
-
-	signal = current_thread_info()->exec_domain
-		&& current_thread_info()->exec_domain->signal_invmap
-		&& sig < 32
-		? current_thread_info()->exec_domain->signal_invmap[sig]
-		: sig;
-
-	err |= copy_siginfo_to_user(&frame->info, info);
-
-	/* Create the ucontext.  */
-	err |= __put_user(0, &frame->uc.uc_flags);
-	err |= __put_user(0, &frame->uc.uc_link);
-	err |= __put_user((void *)current->sas_ss_sp,
-			  &frame->uc.uc_stack.ss_sp);
-	err |= __put_user(sas_ss_flags(regs->gpr[GPR_SP]),
-			  &frame->uc.uc_stack.ss_flags);
-	err |= __put_user(current->sas_ss_size, &frame->uc.uc_stack.ss_size);
-	err |= setup_sigcontext(&frame->uc.uc_mcontext,
-			        regs, set->sig[0]);
-	err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set));
-
-	/* Set up to return from userspace.  If provided, use a stub
-	   already in userspace.  */
-	if (ka->sa.sa_flags & SA_RESTORER) {
-		regs->gpr[GPR_LP] = (unsigned long) ka->sa.sa_restorer;
-	} else {
-		/* Note, these encodings are _little endian_!  */
-
-		/* addi  __NR_sigreturn, r0, r12  */
-		err |= __put_user(0x6600 | (__NR_sigreturn << 16),
-				  frame->tramp + 0);
-		/* trap 0 */
-		err |= __put_user(0x010007e0,
-				  frame->tramp + 1);
-
-		regs->gpr[GPR_LP] = (unsigned long)frame->tramp;
-
-		flush_cache_sigtramp (regs->gpr[GPR_LP]);
-	}
-
-	if (err)
-		goto give_sigsegv;
-
-	/* Set up registers for signal handler.  */
-	regs->pc = (v850_reg_t) ka->sa.sa_handler;
-	regs->gpr[GPR_SP] = (v850_reg_t)frame;
-	/* Signal handler args:  */
-	regs->gpr[GPR_ARG0] = signal; /* arg 0: signum */
-	regs->gpr[GPR_ARG1] = (v850_reg_t)&frame->info; /* arg 1: siginfo */
-	regs->gpr[GPR_ARG2] = (v850_reg_t)&frame->uc; /* arg 2: ucontext */
-
-	set_fs(USER_DS);
-
-#if DEBUG_SIG
-	printk("SIG deliver (%s:%d): sp=%p pc=%08lx pr=%08lx\n",
-		current->comm, current->pid, frame, regs->pc, regs->pr);
-#endif
-
-	return;
-
-give_sigsegv:
-	force_sigsegv(sig, current);
-}
-
-/*
- * OK, we're invoking a handler
- */	
-
-static void
-handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka,
-	      sigset_t *oldset,	struct pt_regs * regs)
-{
-	/* Are we from a system call? */
-	if (PT_REGS_SYSCALL (regs)) {
-		/* If so, check system call restarting.. */
-		switch (regs->gpr[GPR_RVAL]) {
-		case -ERESTART_RESTARTBLOCK:
-			current_thread_info()->restart_block.fn =
-				do_no_restart_syscall;
-			/* fall through */
-		case -ERESTARTNOHAND:
-			regs->gpr[GPR_RVAL] = -EINTR;
-			break;
-
-		case -ERESTARTSYS:
-			if (!(ka->sa.sa_flags & SA_RESTART)) {
-				regs->gpr[GPR_RVAL] = -EINTR;
-				break;
-			}
-			/* fallthrough */
-		case -ERESTARTNOINTR:
-			regs->gpr[12] = PT_REGS_SYSCALL (regs);
-			regs->pc -= 4; /* Size of `trap 0' insn.  */
-		}
-
-		PT_REGS_SET_SYSCALL (regs, 0);
-	}
-
-	/* Set up the stack frame */
-	if (ka->sa.sa_flags & SA_SIGINFO)
-		setup_rt_frame(sig, ka, info, oldset, regs);
-	else
-		setup_frame(sig, ka, oldset, regs);
-
-	spin_lock_irq(&current->sighand->siglock);
-	sigorsets(&current->blocked,&current->blocked,&ka->sa.sa_mask);
-	if (!(ka->sa.sa_flags & SA_NODEFER))
-		sigaddset(&current->blocked,sig);
-	recalc_sigpending();
-	spin_unlock_irq(&current->sighand->siglock);
-}
-
-/*
- * Note that 'init' is a special process: it doesn't get signals it doesn't
- * want to handle. Thus you cannot kill init even with a SIGKILL even by
- * mistake.
- *
- * Note that we go through the signals twice: once to check the signals that
- * the kernel can handle, and then we build all the user-level signal handling
- * stack-frames in one go after that.
- */
-int do_signal(struct pt_regs *regs, sigset_t *oldset)
-{
-	siginfo_t info;
-	int signr;
-	struct k_sigaction ka;
-
-	/*
-	 * We want the common case to go fast, which
-	 * is why we may in certain cases get here from
-	 * kernel mode. Just return without doing anything
-	 * if so.
-	 */
-	if (!user_mode(regs))
-		return 1;
-
-	if (!oldset)
-		oldset = &current->blocked;
-
-	signr = get_signal_to_deliver(&info, &ka, regs, NULL);
-	if (signr > 0) {
-		/* Whee!  Actually deliver the signal.  */
-		handle_signal(signr, &info, &ka, oldset, regs);
-		return 1;
-	}
-
-	/* Did we come from a system call? */
-	if (PT_REGS_SYSCALL (regs)) {
-		int rval = (int)regs->gpr[GPR_RVAL];
-		/* Restart the system call - no handlers present */
-		if (rval == -ERESTARTNOHAND
-		    || rval == -ERESTARTSYS
-		    || rval == -ERESTARTNOINTR)
-		{
-			regs->gpr[12] = PT_REGS_SYSCALL (regs);
-			regs->pc -= 4; /* Size of `trap 0' insn.  */
-		}
-		else if (rval == -ERESTART_RESTARTBLOCK) {
-			regs->gpr[12] = __NR_restart_syscall;
-			regs->pc -= 4; /* Size of `trap 0' insn.  */
-		}
-	}
-	return 0;
-}
diff --git a/arch/v850/kernel/sim.c b/arch/v850/kernel/sim.c
deleted file mode 100644
index 467b4aa0acdd..000000000000
--- a/arch/v850/kernel/sim.c
+++ /dev/null
@@ -1,172 +0,0 @@
-/*
- * arch/v850/kernel/sim.c -- Machine-specific stuff for GDB v850e simulator
- *
- *  Copyright (C) 2001,02  NEC Corporation
- *  Copyright (C) 2001,02  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/init.h>
-#include <linux/mm.h>
-#include <linux/swap.h>
-#include <linux/bootmem.h>
-#include <linux/irq.h>
-
-#include <asm/atomic.h>
-#include <asm/page.h>
-#include <asm/machdep.h>
-#include <asm/simsyscall.h>
-
-#include "mach.h"
-
-/* The name of a file containing the root filesystem.  */
-#define ROOT_FS "rootfs.image"
-
-extern void simcons_setup (void);
-extern void simcons_poll_ttys (void);
-extern void set_mem_root (void *addr, size_t len, char *cmd_line);
-
-static int read_file (const char *name,
-		      unsigned long *addr, unsigned long *len,
-		      const char **err);
-
-void __init mach_setup (char **cmdline)
-{
-	const char *err;
-	unsigned long root_dev_addr, root_dev_len;
-
-	simcons_setup ();
-
-	printk (KERN_INFO "Reading root filesystem: %s", ROOT_FS);
-
-	if (read_file (ROOT_FS, &root_dev_addr, &root_dev_len, &err)) {
-		printk (" (size %luK)\n", root_dev_len / 1024);
-		set_mem_root ((void *)root_dev_addr, (size_t)root_dev_len,
-			      *cmdline);
-	} else
-		printk ("...%s failed!\n", err);
-}
-
-void mach_get_physical_ram (unsigned long *ram_start, unsigned long *ram_len)
-{
-	*ram_start = RAM_ADDR;
-	*ram_len = RAM_SIZE;
-}
-
-void __init mach_sched_init (struct irqaction *timer_action)
-{
-	/* ...do magic timer initialization?...  */
-	mach_tick = simcons_poll_ttys;
-	setup_irq (0, timer_action);
-}
-
-
-static void irq_nop (unsigned irq) { }
-static unsigned irq_zero (unsigned irq) { return 0; }
-
-static struct hw_interrupt_type sim_irq_type = {
-	.typename = "IRQ",
-	.startup = irq_zero,		/* startup */
-	.shutdown = irq_nop,		/* shutdown */
-	.enable = irq_nop,		/* enable */
-	.disable = irq_nop,		/* disable */
-	.ack = irq_nop,		/* ack */
-	.end = irq_nop,		/* end */
-};
-
-void __init mach_init_irqs (void)
-{
-	init_irq_handlers (0, NUM_MACH_IRQS, 1, &sim_irq_type);
-}
-
-
-void mach_gettimeofday (struct timespec *tv)
-{
-	long timeval[2], timezone[2];
-	int rval = V850_SIM_SYSCALL (gettimeofday, timeval, timezone);
-	if (rval == 0) {
-		tv->tv_sec = timeval[0];
-		tv->tv_nsec = timeval[1] * 1000;
-	}
-}
-
-void machine_restart (char *__unused)
-{
-	V850_SIM_SYSCALL (write, 1, "RESTART\n", 8);
-	V850_SIM_SYSCALL (exit, 0);
-}
-
-void machine_halt (void)
-{
-	V850_SIM_SYSCALL (write, 1, "HALT\n", 5);
-	V850_SIM_SYSCALL (exit, 0);
-}
-
-void machine_power_off (void)
-{
-	V850_SIM_SYSCALL (write, 1, "POWER OFF\n", 10);
-	V850_SIM_SYSCALL (exit, 0);
-}
-
-
-/* Load data from a file called NAME into ram.  The address and length
-   of the data image are returned in ADDR and LEN.  */
-static int __init
-read_file (const char *name,
-	   unsigned long *addr, unsigned long *len,
-	   const char **err)
-{
-	int rval, fd;
-	unsigned long cur, left;
-	/* Note this is not a normal stat buffer, it's an ad-hoc
-	   structure defined by the simulator.  */
-	unsigned long stat_buf[10];
-
-	/* Stat the file to find out the length.  */
-	rval = V850_SIM_SYSCALL (stat, name, stat_buf);
-	if (rval < 0) {
-		if (err) *err = "stat";
-		return 0;
-	}
-	*len = stat_buf[4];
-
-	/* Open the file; `0' is O_RDONLY.  */
-	fd = V850_SIM_SYSCALL (open, name, 0);
-	if (fd < 0) {
-		if (err) *err = "open";
-		return 0;
-	}
-
-	*addr = (unsigned long)alloc_bootmem(*len);
-	if (! *addr) {
-		V850_SIM_SYSCALL (close, fd);
-		if (err) *err = "alloc_bootmem";
-		return 0;
-	}
-
-	cur = *addr;
-	left = *len;
-	while (left > 0) {
-		int chunk = V850_SIM_SYSCALL (read, fd, cur, left);
-		if (chunk <= 0)
-			break;
-		cur += chunk;
-		left -= chunk;
-	}
-	V850_SIM_SYSCALL (close, fd);
-	if (left > 0) {
-		/* Some read failed.  */
-		free_bootmem (*addr, *len);
-		if (err) *err = "read";
-		return 0;
-	}
-
-	return 1;
-}
diff --git a/arch/v850/kernel/sim.ld b/arch/v850/kernel/sim.ld
deleted file mode 100644
index 101885f3c9f0..000000000000
--- a/arch/v850/kernel/sim.ld
+++ /dev/null
@@ -1,13 +0,0 @@
-/* Linker script for the gdb v850e simulator (CONFIG_V850E_SIM).  */
-
-MEMORY {
-	/* Interrupt vectors.  */
-	INTV  : ORIGIN = 0x0, LENGTH = 0xe0
-	/* Main RAM.  */
-	RAM   : ORIGIN = RAM_ADDR, LENGTH = RAM_SIZE
-}
-
-SECTIONS {
-	.intv : { INTV_CONTENTS } > INTV
-	.ram : { RAMK_KRAM_CONTENTS } > RAM
-}
diff --git a/arch/v850/kernel/sim85e2.c b/arch/v850/kernel/sim85e2.c
deleted file mode 100644
index 566dde5e6070..000000000000
--- a/arch/v850/kernel/sim85e2.c
+++ /dev/null
@@ -1,195 +0,0 @@
-/*
- * arch/v850/kernel/sim85e2.c -- Machine-specific stuff for
- *	V850E2 RTL simulator
- *
- *  Copyright (C) 2002,03  NEC Electronics Corporation
- *  Copyright (C) 2002,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/init.h>
-#include <linux/mm.h>
-#include <linux/swap.h>
-#include <linux/bootmem.h>
-#include <linux/irq.h>
-
-#include <asm/atomic.h>
-#include <asm/page.h>
-#include <asm/machdep.h>
-
-#include "mach.h"
-
-
-/* There are 4 possible areas we can use:
-
-     IRAM (1MB) is fast for instruction fetches, but slow for data
-     DRAM (1020KB) is fast for data, but slow for instructions
-     ERAM is cached, so should be fast for both insns and data
-     SDRAM is external DRAM, similar to ERAM
-*/
-
-#define INIT_MEMC_FOR_SDRAM
-#define USE_SDRAM_AREA
-#define KERNEL_IN_SDRAM_AREA
-
-#define DCACHE_MODE	V850E2_CACHE_BTSC_DCM_WT
-/*#define DCACHE_MODE	V850E2_CACHE_BTSC_DCM_WB_ALLOC*/
-
-#ifdef USE_SDRAM_AREA
-#define RAM_START 	SDRAM_ADDR
-#define RAM_END		(SDRAM_ADDR + SDRAM_SIZE)
-#else
-/* When we use DRAM, we need to account for the fact that the end of it is
-   used for R0_RAM.  */
-#define RAM_START	DRAM_ADDR
-#define RAM_END		R0_RAM_ADDR
-#endif
-
-
-extern void memcons_setup (void);
-
-
-#ifdef KERNEL_IN_SDRAM_AREA
-#define EARLY_INIT_SECTION_ATTR __attribute__ ((section (".early.text")))
-#else
-#define EARLY_INIT_SECTION_ATTR __init
-#endif
-
-void EARLY_INIT_SECTION_ATTR mach_early_init (void)
-{
-	/* The sim85e2 simulator tracks `undefined' values, so to make
-	   debugging easier, we begin by zeroing out all otherwise
-	   undefined registers.  This is not strictly necessary.
-
-	   The registers we zero are:
-	       Every GPR except:
-	           stack-pointer (r3)
-		   task-pointer (r16)
-		   our return addr (r31)
-	       Every system register (SPR) that we know about except for
-	       the PSW (SPR 5), which we zero except for the
-	       disable-interrupts bit.
-	*/
-
-	/* GPRs */
-	asm volatile ("             mov r0, r1 ; mov r0, r2              ");
-	asm volatile ("mov r0, r4 ; mov r0, r5 ; mov r0, r6 ; mov r0, r7 ");
-	asm volatile ("mov r0, r8 ; mov r0, r9 ; mov r0, r10; mov r0, r11");
-	asm volatile ("mov r0, r12; mov r0, r13; mov r0, r14; mov r0, r15");
-	asm volatile ("             mov r0, r17; mov r0, r18; mov r0, r19");
-	asm volatile ("mov r0, r20; mov r0, r21; mov r0, r22; mov r0, r23");
-	asm volatile ("mov r0, r24; mov r0, r25; mov r0, r26; mov r0, r27");
-	asm volatile ("mov r0, r28; mov r0, r29; mov r0, r30");
-
-	/* SPRs */
-	asm volatile ("ldsr r0, 0;  ldsr r0, 1;  ldsr r0, 2;  ldsr r0, 3");
-	asm volatile ("ldsr r0, 4");
-	asm volatile ("addi 0x20, r0, r1; ldsr r1, 5"); /* PSW */
-	asm volatile ("ldsr r0, 16; ldsr r0, 17; ldsr r0, 18; ldsr r0, 19");
-	asm volatile ("ldsr r0, 20");
-
-
-#ifdef INIT_MEMC_FOR_SDRAM
-	/* Settings for SDRAM controller.  */
-	V850E2_VSWC   = 0x0042;
-	V850E2_BSC    = 0x9286;
-	V850E2_BCT(0) = 0xb000;	/* was: 0 */
-	V850E2_BCT(1) = 0x000b;
-	V850E2_ASC    = 0;
-	V850E2_LBS    = 0xa9aa;	/* was: 0xaaaa */
-	V850E2_LBC(0) = 0;
-	V850E2_LBC(1) = 0;	/* was: 0x3 */
-	V850E2_BCC    = 0;
-	V850E2_RFS(4) = 0x800a;	/* was: 0xf109 */
-	V850E2_SCR(4) = 0x2091;	/* was: 0x20a1 */
-	V850E2_RFS(3) = 0x800c;
-	V850E2_SCR(3) = 0x20a1;
-	V850E2_DWC(0) = 0;
-	V850E2_DWC(1) = 0;
-#endif
-
-#if 0
-#ifdef CONFIG_V850E2_SIM85E2S
-	/* Turn on the caches.  */
-	V850E2_CACHE_BTSC = V850E2_CACHE_BTSC_ICM | DCACHE_MODE;
-	V850E2_BHC  = 0x1010;
-#elif CONFIG_V850E2_SIM85E2C
-	V850E2_CACHE_BTSC |= (V850E2_CACHE_BTSC_ICM | V850E2_CACHE_BTSC_DCM0);
-	V850E2_BUSM_BHC = 0xFFFF;
-#endif
-#else
-	V850E2_BHC  = 0;
-#endif
-
-	/* Don't stop the simulator at `halt' instructions.  */
-	SIM85E2_NOTHAL = 1;
-
-	/* Ensure that the simulator halts on a panic, instead of going
-	   into an infinite loop inside the panic function.  */
-	panic_timeout = -1;
-}
-
-void __init mach_setup (char **cmdline)
-{
-	memcons_setup ();
-}
-
-void mach_get_physical_ram (unsigned long *ram_start, unsigned long *ram_len)
-{
-	*ram_start = RAM_START;
-	*ram_len = RAM_END - RAM_START;
-}
-
-void __init mach_sched_init (struct irqaction *timer_action)
-{
-	/* The simulator actually cycles through all interrupts
-	   periodically.  We just pay attention to IRQ0, which gives us
-	   1/64 the rate of the periodic interrupts.  */
-	setup_irq (0, timer_action);
-}
-
-void mach_gettimeofday (struct timespec *tv)
-{
-	tv->tv_sec = 0;
-	tv->tv_nsec = 0;
-}
-
-/* Interrupts */
-
-struct v850e_intc_irq_init irq_inits[] = {
-	{ "IRQ", 0, NUM_MACH_IRQS, 1, 7 },
-	{ 0 }
-};
-struct hw_interrupt_type hw_itypes[1];
-
-/* Initialize interrupts.  */
-void __init mach_init_irqs (void)
-{
-	v850e_intc_init_irq_types (irq_inits, hw_itypes);
-}
-
-
-void machine_halt (void) __attribute__ ((noreturn));
-void machine_halt (void)
-{
-	SIM85E2_SIMFIN = 0;	/* Halt immediately.  */
-	for (;;) {}
-}
-
-void machine_restart (char *__unused)
-{
-	machine_halt ();
-}
-
-void machine_power_off (void)
-{
-	machine_halt ();
-}
-
diff --git a/arch/v850/kernel/sim85e2.ld b/arch/v850/kernel/sim85e2.ld
deleted file mode 100644
index 7470fd2ffb5b..000000000000
--- a/arch/v850/kernel/sim85e2.ld
+++ /dev/null
@@ -1,36 +0,0 @@
-/* Linker script for the sim85e2c simulator, which is a verilog simulation of
-   the V850E2 NA85E2C cpu core (CONFIG_V850E2_SIM85E2C).  */
-
-MEMORY {
-	/* 1MB of `instruction RAM', starting at 0.
-	   Instruction fetches are much faster from IRAM than from DRAM.  */
-	IRAM : ORIGIN = IRAM_ADDR, LENGTH = IRAM_SIZE
-
-	/* 1MB of `data RAM', below and contiguous with the I/O space.
-	   Data fetches are much faster from DRAM than from IRAM.  */
-	DRAM : ORIGIN = DRAM_ADDR, LENGTH = DRAM_SIZE
-
-	/* `external ram' (CS1 area), comes after IRAM.  */
-	ERAM : ORIGIN = ERAM_ADDR, LENGTH = ERAM_SIZE
-
-	/* Dynamic RAM; uses memory controller.  */
-	SDRAM : ORIGIN = SDRAM_ADDR, LENGTH = SDRAM_SIZE
-}
-
-SECTIONS {
-	.iram : {
-		INTV_CONTENTS
-		*arch/v850/kernel/head.o
-		*(.early.text)
-	} > IRAM
-	.dram : {
-		_memcons_output = . ;
-		. = . + 0x8000 ;
-		_memcons_output_end = . ;
-	} > DRAM
-	.sdram : {
-		/* We stick console output into a buffer here.  */
-		RAMK_KRAM_CONTENTS
-		ROOT_FS_CONTENTS
-	} > SDRAM
-}
diff --git a/arch/v850/kernel/simcons.c b/arch/v850/kernel/simcons.c
deleted file mode 100644
index 9973596ae304..000000000000
--- a/arch/v850/kernel/simcons.c
+++ /dev/null
@@ -1,161 +0,0 @@
-/*
- * arch/v850/kernel/simcons.c -- Console I/O for GDB v850e simulator
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/kernel.h>
-#include <linux/console.h>
-#include <linux/tty.h>
-#include <linux/tty_flip.h>
-#include <linux/tty_driver.h>
-#include <linux/init.h>
-
-#include <asm/poll.h>
-#include <asm/string.h>
-#include <asm/simsyscall.h>
-
-
-/*  Low-level console. */
-
-static void simcons_write (struct console *co, const char *buf, unsigned len)
-{
-	V850_SIM_SYSCALL (write, 1, buf, len);
-}
-
-static int simcons_read (struct console *co, char *buf, unsigned len)
-{
-	return V850_SIM_SYSCALL (read, 0, buf, len);
-}
-
-static struct tty_driver *tty_driver;
-static struct tty_driver *simcons_device (struct console *c, int *index)
-{
-	*index = c->index;
-	return tty_driver;
-}
-
-static struct console simcons =
-{
-    .name	= "simcons",
-    .write	= simcons_write,
-    .read	= simcons_read,
-    .device	= simcons_device,
-    .flags	= CON_PRINTBUFFER,
-    .index	= -1,
-};
-
-/* Higher level TTY interface.  */
-
-int simcons_tty_open (struct tty_struct *tty, struct file *filp)
-{
-	return 0;
-}
-
-int simcons_tty_write (struct tty_struct *tty,
-		       const unsigned char *buf, int count)
-{
-	return V850_SIM_SYSCALL (write, 1, buf, count);
-}
-
-int simcons_tty_write_room (struct tty_struct *tty)
-{
-	/* Completely arbitrary.  */
-	return 0x100000;
-}
-
-int simcons_tty_chars_in_buffer (struct tty_struct *tty)
-{
-	/* We have no buffer.  */
-	return 0;
-}
-
-static const struct tty_operations ops = {
-	.open = simcons_tty_open,
-	.write = simcons_tty_write,
-	.write_room = simcons_tty_write_room,
-	.chars_in_buffer = simcons_tty_chars_in_buffer,
-};
-
-int __init simcons_tty_init (void)
-{
-	struct tty_driver *driver = alloc_tty_driver(1);
-	int err;
-	if (!driver)
-		return -ENOMEM;
-	driver->name = "simcons";
-	driver->major = TTY_MAJOR;
-	driver->minor_start = 64;
-	driver->type = TTY_DRIVER_TYPE_SYSCONS;
-	driver->init_termios = tty_std_termios;
-	tty_set_operations(driver, &ops);
-	err = tty_register_driver(driver);
-	if (err) {
-		put_tty_driver(driver);
-		return err;
-	}
-	tty_driver = driver;
-	return 0;
-}
-/* We use `late_initcall' instead of just `__initcall' as a workaround for
-   the fact that (1) simcons_tty_init can't be called before tty_init,
-   (2) tty_init is called via `module_init', (3) if statically linked,
-   module_init == device_init, and (4) there's no ordering of init lists.
-   We can do this easily because simcons is always statically linked, but
-   other tty drivers that depend on tty_init and which must use
-   `module_init' to declare their init routines are likely to be broken.  */
-late_initcall(simcons_tty_init);
-
-/* Poll for input on the console, and if there's any, deliver it to the
-   tty driver.  */
-void simcons_poll_tty (struct tty_struct *tty)
-{
-	char buf[32];	/* Not the nicest way to do it but I need it correct first */
-	int flip = 0, send_break = 0;
-	struct pollfd pfd;
-	pfd.fd = 0;
-	pfd.events = POLLIN;
-
-	if (V850_SIM_SYSCALL (poll, &pfd, 1, 0) > 0) {
-		if (pfd.revents & POLLIN) {
-			/* Real block hardware knows the transfer size before
-			   transfer so the new tty buffering doesn't try to handle
-			   this rather weird simulator specific case well */
-			int rd = V850_SIM_SYSCALL (read, 0, buf, 32);
-			if (rd > 0) {
-				tty_insert_flip_string(tty, buf, rd);
-				flip = 1;
-			} else
-				send_break = 1;
-		} else if (pfd.revents & POLLERR)
-			send_break = 1;
-	}
-
-	if (send_break) {
-		tty_insert_flip_char (tty, 0, TTY_BREAK);		
-		flip = 1;
-	}
-
-	if (flip)
-		tty_schedule_flip (tty);
-}
-
-void simcons_poll_ttys (void)
-{
-	if (tty_driver && tty_driver->ttys[0])
-		simcons_poll_tty (tty_driver->ttys[0]);
-}
-
-void simcons_setup (void)
-{
-	V850_SIM_SYSCALL (make_raw, 0);
-	register_console (&simcons);
-	printk (KERN_INFO "Console: GDB V850E simulator stdio\n");
-}
diff --git a/arch/v850/kernel/syscalls.c b/arch/v850/kernel/syscalls.c
deleted file mode 100644
index 1a83daf8e24f..000000000000
--- a/arch/v850/kernel/syscalls.c
+++ /dev/null
@@ -1,196 +0,0 @@
-/*
- * arch/v850/kernel/syscalls.c -- Various system-call definitions not
- * 	defined in machine-independent code
- *
- *  Copyright (C) 2001,02  NEC Corporation
- *  Copyright (C) 2001,02  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * This file was derived the ppc version, arch/ppc/kernel/syscalls.c
- * ... which was derived from "arch/i386/kernel/sys_i386.c" by Gary Thomas;
- *     modified by Cort Dougan (cort@cs.nmt.edu)
- *     and Paul Mackerras (paulus@cs.anu.edu.au).
- */
-
-#include <linux/errno.h>
-#include <linux/mm.h>
-#include <linux/smp.h>
-#include <linux/syscalls.h>
-#include <linux/sem.h>
-#include <linux/msg.h>
-#include <linux/shm.h>
-#include <linux/stat.h>
-#include <linux/mman.h>
-#include <linux/sys.h>
-#include <linux/ipc.h>
-#include <linux/utsname.h>
-#include <linux/file.h>
-
-#include <asm/uaccess.h>
-#include <asm/unistd.h>
-
-/*
- * sys_ipc() is the de-multiplexer for the SysV IPC calls..
- *
- * This is really horribly ugly.
- */
-int
-sys_ipc (uint call, int first, int second, int third, void *ptr, long fifth)
-{
-	int version, ret;
-
-	version = call >> 16; /* hack for backward compatibility */
-	call &= 0xffff;
-
-	ret = -EINVAL;
-	switch (call) {
-	case SEMOP:
-		ret = sys_semop (first, (struct sembuf *)ptr, second);
-		break;
-	case SEMGET:
-		ret = sys_semget (first, second, third);
-		break;
-	case SEMCTL:
-	{
-		union semun fourth;
-
-		if (!ptr)
-			break;
-		if ((ret = access_ok(VERIFY_READ, ptr, sizeof(long)) ? 0 : -EFAULT)
-		    || (ret = get_user(fourth.__pad, (void **)ptr)))
-			break;
-		ret = sys_semctl (first, second, third, fourth);
-		break;
-	}
-	case MSGSND:
-		ret = sys_msgsnd (first, (struct msgbuf *) ptr, second, third);
-		break;
-	case MSGRCV:
-		switch (version) {
-		case 0: {
-			struct ipc_kludge tmp;
-
-			if (!ptr)
-				break;
-			if ((ret = access_ok(VERIFY_READ, ptr, sizeof(tmp)) ? 0 : -EFAULT)
-			    || (ret = copy_from_user(&tmp,
-						(struct ipc_kludge *) ptr,
-						sizeof (tmp))))
-				break;
-			ret = sys_msgrcv (first, tmp.msgp, second, tmp.msgtyp,
-					  third);
-			break;
-			}
-		default:
-			ret = sys_msgrcv (first, (struct msgbuf *) ptr,
-					  second, fifth, third);
-			break;
-		}
-		break;
-	case MSGGET:
-		ret = sys_msgget ((key_t) first, second);
-		break;
-	case MSGCTL:
-		ret = sys_msgctl (first, second, (struct msqid_ds *) ptr);
-		break;
-	case SHMAT:
-		switch (version) {
-		default: {
-			ulong raddr;
-
-			if ((ret = access_ok(VERIFY_WRITE, (ulong*) third,
-					       sizeof(ulong)) ? 0 : -EFAULT))
-				break;
-			ret = do_shmat (first, (char *) ptr, second, &raddr);
-			if (ret)
-				break;
-			ret = put_user (raddr, (ulong *) third);
-			break;
-			}
-		case 1:	/* iBCS2 emulator entry point */
-			if (!segment_eq(get_fs(), get_ds()))
-				break;
-			ret = do_shmat (first, (char *) ptr, second,
-					 (ulong *) third);
-			break;
-		}
-		break;
-	case SHMDT: 
-		ret = sys_shmdt ((char *)ptr);
-		break;
-	case SHMGET:
-		ret = sys_shmget (first, second, third);
-		break;
-	case SHMCTL:
-		ret = sys_shmctl (first, second, (struct shmid_ds *) ptr);
-		break;
-	}
-
-	return ret;
-}
-
-static inline unsigned long
-do_mmap2 (unsigned long addr, size_t len,
-	 unsigned long prot, unsigned long flags,
-	 unsigned long fd, unsigned long pgoff)
-{
-	struct file * file = NULL;
-	int ret = -EBADF;
-
-	flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE);
-	if (! (flags & MAP_ANONYMOUS)) {
-		if (!(file = fget (fd)))
-			goto out;
-	}
-	
-	down_write (&current->mm->mmap_sem);
-	ret = do_mmap_pgoff (file, addr, len, prot, flags, pgoff);
-	up_write (&current->mm->mmap_sem);
-	if (file)
-		fput (file);
-out:
-	return ret;
-}
-
-unsigned long sys_mmap2 (unsigned long addr, size_t len,
-			unsigned long prot, unsigned long flags,
-			unsigned long fd, unsigned long pgoff)
-{
-	return do_mmap2 (addr, len, prot, flags, fd, pgoff);
-}
-
-unsigned long sys_mmap (unsigned long addr, size_t len,
-		       unsigned long prot, unsigned long flags,
-		       unsigned long fd, off_t offset)
-{
-	int err = -EINVAL;
-
-	if (offset & ~PAGE_MASK)
-		goto out;
-
-	err = do_mmap2 (addr, len, prot, flags, fd, offset >> PAGE_SHIFT);
-out:
-	return err;
-}
-
-/*
- * Do a system call from kernel instead of calling sys_execve so we
- * end up with proper pt_regs.
- */
-int kernel_execve(const char *filename, char *const argv[], char *const envp[])
-{
-	register char *__a __asm__ ("r6") = filename;
-	register void *__b __asm__ ("r7") = argv;
-	register void *__c __asm__ ("r8") = envp;
-	register unsigned long __syscall __asm__ ("r12") = __NR_execve;
-	register unsigned long __ret __asm__ ("r10");
-	__asm__ __volatile__ ("trap 0"
-			: "=r" (__ret), "=r" (__syscall)
-			: "1" (__syscall), "r" (__a), "r" (__b), "r" (__c)
-			: "r1", "r5", "r11", "r13", "r14",
-			  "r15", "r16", "r17", "r18", "r19");
-	return __ret;
-}
diff --git a/arch/v850/kernel/teg.c b/arch/v850/kernel/teg.c
deleted file mode 100644
index 699248f92aae..000000000000
--- a/arch/v850/kernel/teg.c
+++ /dev/null
@@ -1,62 +0,0 @@
-/*
- * arch/v850/kernel/teg.c -- NB85E-TEG cpu chip
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/kernel.h>
-#include <linux/init.h>
-#include <linux/mm.h>
-#include <linux/swap.h>
-#include <linux/bootmem.h>
-#include <linux/irq.h>
-
-#include <asm/atomic.h>
-#include <asm/page.h>
-#include <asm/machdep.h>
-#include <asm/v850e_timer_d.h>
-
-#include "mach.h"
-
-void __init mach_sched_init (struct irqaction *timer_action)
-{
-	/* Select timer interrupt instead of external pin.  */
-	TEG_ISS |= 0x1;
-	/* Start hardware timer.  */
-	v850e_timer_d_configure (0, HZ);
-	/* Install timer interrupt handler.  */
-	setup_irq (IRQ_INTCMD(0), timer_action);
-}
-
-static struct v850e_intc_irq_init irq_inits[] = {
-	{ "IRQ", 0,		NUM_CPU_IRQS,	1, 7 },
-	{ "CMD", IRQ_INTCMD(0),	IRQ_INTCMD_NUM,	1, 5 },
-	{ "SER", IRQ_INTSER(0),	IRQ_INTSER_NUM,	1, 3 },
-	{ "SR",	 IRQ_INTSR(0),	IRQ_INTSR_NUM,	1, 4 },
-	{ "ST",	 IRQ_INTST(0),	IRQ_INTST_NUM,	1, 5 },
-	{ 0 }
-};
-#define NUM_IRQ_INITS (ARRAY_SIZE(irq_inits) - 1)
-
-static struct hw_interrupt_type hw_itypes[NUM_IRQ_INITS];
-
-/* Initialize MA chip interrupts.  */
-void __init teg_init_irqs (void)
-{
-	v850e_intc_init_irq_types (irq_inits, hw_itypes);
-}
-
-/* Called before configuring an on-chip UART.  */
-void teg_uart_pre_configure (unsigned chan, unsigned cflags, unsigned baud)
-{
-	/* Enable UART I/O pins instead of external interrupt pins, and
-	   UART interrupts instead of external pin interrupts.  */
-	TEG_ISS |= 0x4E;
-}
diff --git a/arch/v850/kernel/time.c b/arch/v850/kernel/time.c
deleted file mode 100644
index d810c93fe665..000000000000
--- a/arch/v850/kernel/time.c
+++ /dev/null
@@ -1,106 +0,0 @@
-/*
- * linux/arch/v850/kernel/time.c -- Arch-dependent timer functions
- *
- *  Copyright (C) 1991, 1992, 1995, 2001, 2002  Linus Torvalds
- *
- * This file contains the v850-specific time handling details.
- * Most of the stuff is located in the machine specific files.
- *
- * 1997-09-10	Updated NTP code according to technical memorandum Jan '96
- *		"A Kernel Model for Precision Timekeeping" by Dave Mills
- */
-
-#include <linux/errno.h>
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/param.h>
-#include <linux/string.h>
-#include <linux/mm.h>
-#include <linux/interrupt.h>
-#include <linux/time.h>
-#include <linux/timex.h>
-#include <linux/profile.h>
-
-#include <asm/io.h>
-
-#include "mach.h"
-
-#define TICK_SIZE	(tick_nsec / 1000)
-
-/*
- * timer_interrupt() needs to keep up the real-time clock,
- * as well as call the "do_timer()" routine every clocktick
- */
-static irqreturn_t timer_interrupt (int irq, void *dummy, struct pt_regs *regs)
-{
-#if 0
-	/* last time the cmos clock got updated */
-	static long last_rtc_update=0;
-#endif
-
-	/* may need to kick the hardware timer */
-	if (mach_tick)
-	  mach_tick ();
-
-	do_timer (1);
-#ifndef CONFIG_SMP
-	update_process_times(user_mode(regs));
-#endif
-	profile_tick(CPU_PROFILING, regs);
-#if 0
-	/*
-	 * If we have an externally synchronized Linux clock, then update
-	 * CMOS clock accordingly every ~11 minutes. Set_rtc_mmss() has to be
-	 * called as close as possible to 500 ms before the new second starts.
-	 */
-	if (ntp_synced() &&
-	    xtime.tv_sec > last_rtc_update + 660 &&
-	    (xtime.tv_nsec / 1000) >= 500000 - ((unsigned) TICK_SIZE) / 2 &&
-	    (xtime.tv_nsec / 1000) <= 500000 + ((unsigned) TICK_SIZE) / 2) {
-	  if (set_rtc_mmss (xtime.tv_sec) == 0)
-	    last_rtc_update = xtime.tv_sec;
-	  else
-	    last_rtc_update = xtime.tv_sec - 600; /* do it again in 60 s */
-	}
-#ifdef CONFIG_HEARTBEAT
-	/* use power LED as a heartbeat instead -- much more useful
-	   for debugging -- based on the version for PReP by Cort */
-	/* acts like an actual heart beat -- ie thump-thump-pause... */
-	if (mach_heartbeat) {
-	    static unsigned cnt = 0, period = 0, dist = 0;
-
-	    if (cnt == 0 || cnt == dist)
-		mach_heartbeat ( 1 );
-	    else if (cnt == 7 || cnt == dist+7)
-		mach_heartbeat ( 0 );
-
-	    if (++cnt > period) {
-		cnt = 0;
-		/* The hyperbolic function below modifies the heartbeat period
-		 * length in dependency of the current (5min) load. It goes
-		 * through the points f(0)=126, f(1)=86, f(5)=51,
-		 * f(inf)->30. */
-		period = ((672<<FSHIFT)/(5*avenrun[0]+(7<<FSHIFT))) + 30;
-		dist = period / 4;
-	    }
-	}
-#endif /* CONFIG_HEARTBEAT */
-#endif /* 0 */
-
-	return IRQ_HANDLED;
-}
-
-static int timer_dev_id;
-static struct irqaction timer_irqaction = {
-	.handler = timer_interrupt,
-	.flags = IRQF_DISABLED,
-	.mask = CPU_MASK_NONE,
-	.name = "timer",
-	.dev_id = &timer_dev_id,
-};
-
-void time_init (void)
-{
-	mach_gettimeofday (&xtime);
-	mach_sched_init (&timer_irqaction);
-}
diff --git a/arch/v850/kernel/v850_ksyms.c b/arch/v850/kernel/v850_ksyms.c
deleted file mode 100644
index 8d386a5dbc4a..000000000000
--- a/arch/v850/kernel/v850_ksyms.c
+++ /dev/null
@@ -1,51 +0,0 @@
-#include <linux/module.h>
-#include <linux/linkage.h>
-#include <linux/sched.h>
-#include <linux/string.h>
-#include <linux/mm.h>
-#include <linux/user.h>
-#include <linux/elfcore.h>
-#include <linux/in6.h>
-#include <linux/interrupt.h>
-
-#include <asm/pgalloc.h>
-#include <asm/irq.h>
-#include <asm/io.h>
-#include <asm/checksum.h>
-#include <asm/current.h>
-
-
-extern void *trap_table;
-EXPORT_SYMBOL (trap_table);
-
-/* platform dependent support */
-EXPORT_SYMBOL (kernel_thread);
-EXPORT_SYMBOL (__bug);
-
-/* Networking helper routines. */
-EXPORT_SYMBOL (csum_partial_copy_nocheck);
-EXPORT_SYMBOL (csum_partial_copy_from_user);
-EXPORT_SYMBOL (ip_compute_csum);
-EXPORT_SYMBOL (ip_fast_csum);
-
-/* string / mem functions */
-EXPORT_SYMBOL (memset);
-EXPORT_SYMBOL (memcpy);
-EXPORT_SYMBOL (memmove);
-
-/*
- * libgcc functions - functions that are used internally by the
- * compiler...  (prototypes are not correct though, but that
- * doesn't really matter since they're not versioned).
- */
-extern void __ashldi3 (void);
-extern void __ashrdi3 (void);
-extern void __lshrdi3 (void);
-extern void __muldi3 (void);
-extern void __negdi2 (void);
-
-EXPORT_SYMBOL (__ashldi3);
-EXPORT_SYMBOL (__ashrdi3);
-EXPORT_SYMBOL (__lshrdi3);
-EXPORT_SYMBOL (__muldi3);
-EXPORT_SYMBOL (__negdi2);
diff --git a/arch/v850/kernel/v850e2_cache.c b/arch/v850/kernel/v850e2_cache.c
deleted file mode 100644
index 4570312c689c..000000000000
--- a/arch/v850/kernel/v850e2_cache.c
+++ /dev/null
@@ -1,127 +0,0 @@
-/*
- * arch/v850/kernel/v850e2_cache.c -- Cache control for V850E2 cache
- * 	memories
- *
- *  Copyright (C) 2003  NEC Electronics Corporation
- *  Copyright (C) 2003  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/mm.h>
-
-#include <asm/v850e2_cache.h>
-
-/* Cache operations we can do.  The encoding corresponds directly to the
-   value we need to write into the COPR register.  */
-enum cache_op {
-	OP_SYNC_IF_DIRTY 	   = V850E2_CACHE_COPR_CFC(0), /* 000 */
-	OP_SYNC_IF_VALID 	   = V850E2_CACHE_COPR_CFC(1), /* 001 */
-	OP_SYNC_IF_VALID_AND_CLEAR = V850E2_CACHE_COPR_CFC(3), /* 011 */
-	OP_WAY_CLEAR 		   = V850E2_CACHE_COPR_CFC(4), /* 100 */
-	OP_FILL 		   = V850E2_CACHE_COPR_CFC(5), /* 101 */
-	OP_CLEAR 		   = V850E2_CACHE_COPR_CFC(6), /* 110 */
-	OP_CREATE_DIRTY 	   = V850E2_CACHE_COPR_CFC(7)  /* 111 */
-};
-
-/* Which cache to use.  This encoding also corresponds directly to the
-   value we need to write into the COPR register. */
-enum cache {
-	ICACHE = 0,
-	DCACHE = V850E2_CACHE_COPR_LBSL
-};
-
-/* Returns ADDR rounded down to the beginning of its cache-line.  */
-#define CACHE_LINE_ADDR(addr)  \
-   ((addr) & ~(V850E2_CACHE_LINE_SIZE - 1))
-/* Returns END_ADDR rounded up to the `limit' of its cache-line.  */
-#define CACHE_LINE_END_ADDR(end_addr)  \
-   CACHE_LINE_ADDR(end_addr + (V850E2_CACHE_LINE_SIZE - 1))
-
-
-/* Low-level cache ops.  */
-
-/* Apply cache-op OP to all entries in CACHE.  */
-static inline void cache_op_all (enum cache_op op, enum cache cache)
-{
-	int cmd = op | cache | V850E2_CACHE_COPR_WSLE | V850E2_CACHE_COPR_STRT;
-
-	if (op != OP_WAY_CLEAR) {
-		/* The WAY_CLEAR operation does the whole way, but other
-		   ops take begin-index and count params; we just indicate
-		   the entire cache.  */
-		V850E2_CACHE_CADL = 0;
-		V850E2_CACHE_CADH = 0;
-		V850E2_CACHE_CCNT = V850E2_CACHE_WAY_SIZE - 1;
-	}
-
-	V850E2_CACHE_COPR = cmd | V850E2_CACHE_COPR_WSL(0); /* way 0 */
-	V850E2_CACHE_COPR = cmd | V850E2_CACHE_COPR_WSL(1); /* way 1 */
-	V850E2_CACHE_COPR = cmd | V850E2_CACHE_COPR_WSL(2); /* way 2 */
-	V850E2_CACHE_COPR = cmd | V850E2_CACHE_COPR_WSL(3); /* way 3 */
-}
-
-/* Apply cache-op OP to all entries in CACHE covering addresses ADDR
-   through ADDR+LEN.  */
-static inline void cache_op_range (enum cache_op op, u32 addr, u32 len,
-				   enum cache cache)
-{
-	u32 start = CACHE_LINE_ADDR (addr);
-	u32 end = CACHE_LINE_END_ADDR (addr + len);
-	u32 num_lines = (end - start) >> V850E2_CACHE_LINE_SIZE_BITS;
-
-	V850E2_CACHE_CADL = start & 0xFFFF;
-	V850E2_CACHE_CADH = start >> 16;
-	V850E2_CACHE_CCNT = num_lines - 1;
-
-	V850E2_CACHE_COPR = op | cache | V850E2_CACHE_COPR_STRT;
-}
-
-
-/* High-level ops.  */
-
-static void cache_exec_after_store_all (void)
-{
-	cache_op_all (OP_SYNC_IF_DIRTY, DCACHE);
-	cache_op_all (OP_WAY_CLEAR, ICACHE);
-}
-
-static void cache_exec_after_store_range (u32 start, u32 len)
-{
-	cache_op_range (OP_SYNC_IF_DIRTY, start, len, DCACHE);
-	cache_op_range (OP_CLEAR, start, len, ICACHE);
-}
-
-
-/* Exported functions.  */
-
-void flush_icache (void)
-{
-	cache_exec_after_store_all ();
-}
-
-void flush_icache_range (unsigned long start, unsigned long end)
-{
-	cache_exec_after_store_range (start, end - start);
-}
-
-void flush_icache_page (struct vm_area_struct *vma, struct page *page)
-{
-	cache_exec_after_store_range (page_to_virt (page), PAGE_SIZE);
-}
-
-void flush_icache_user_range (struct vm_area_struct *vma, struct page *page,
-			      unsigned long addr, int len)
-{
-	cache_exec_after_store_range (addr, len);
-}
-
-void flush_cache_sigtramp (unsigned long addr)
-{
-	/* For the exact size, see signal.c, but 16 bytes should be enough.  */
-	cache_exec_after_store_range (addr, 16);
-}
diff --git a/arch/v850/kernel/v850e_cache.c b/arch/v850/kernel/v850e_cache.c
deleted file mode 100644
index ea3e51cfb259..000000000000
--- a/arch/v850/kernel/v850e_cache.c
+++ /dev/null
@@ -1,174 +0,0 @@
-/*
- * arch/v850/kernel/v850e_cache.c -- Cache control for V850E cache memories
- *
- *  Copyright (C) 2003  NEC Electronics Corporation
- *  Copyright (C) 2003  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-/* This file implements cache control for the rather simple cache used on
-   some V850E CPUs, specifically the NB85E/TEG CPU-core and the V850E/ME2
-   CPU.  V850E2 processors have their own (better) cache
-   implementation.  */
-
-#include <asm/entry.h>
-#include <asm/cacheflush.h>
-#include <asm/v850e_cache.h>
-
-#define WAIT_UNTIL_CLEAR(value) while (value) {}
-
-/* Set caching params via the BHC and DCC registers.  */
-void v850e_cache_enable (u16 bhc, u16 icc, u16 dcc)
-{
-	unsigned long *r0_ram = (unsigned long *)R0_RAM_ADDR;
-	register u16 bhc_val asm ("r6") = bhc;
-
-	/* Read the instruction cache control register (ICC) and confirm
-	   that bits 0 and 1 (TCLR0, TCLR1) are all cleared.  */
-	WAIT_UNTIL_CLEAR (V850E_CACHE_ICC & 0x3);
-	V850E_CACHE_ICC = icc;
-
-#ifdef V850E_CACHE_DCC
-	/* Configure data-cache.  */
-	V850E_CACHE_DCC = dcc;
-#endif /* V850E_CACHE_DCC */
-
-	/* Configure caching for various memory regions by writing the BHC
-	   register.  The documentation says that an instruction _cannot_
-	   enable/disable caching for the memory region in which the
-	   instruction itself exists; to work around this, we store
-	   appropriate instructions into the on-chip RAM area (which is never
-	   cached), and briefly jump there to do the work.  */
-#ifdef V850E_CACHE_WRITE_IBS
-	*r0_ram++ 	= 0xf0720760;	/* st.h r0, 0xfffff072[r0] */
-#endif
-	*r0_ram++ 	= 0xf06a3760;	/* st.h r6, 0xfffff06a[r0] */
-	*r0_ram 	= 0x5640006b;	/* jmp [r11] */
-
-	asm ("mov hilo(1f), r11; jmp [%1]; 1:;"
-	     :: "r" (bhc_val), "r" (R0_RAM_ADDR) : "r11");
-}
-
-static void clear_icache (void)
-{
-	/* 1. Read the instruction cache control register (ICC) and confirm
-	      that bits 0 and 1 (TCLR0, TCLR1) are all cleared.  */
-	WAIT_UNTIL_CLEAR (V850E_CACHE_ICC & 0x3);
-
-	/* 2. Read the ICC register and confirm that bit 12 (LOCK0) is
-  	      cleared.  Bit 13 of the ICC register is always cleared.  */
-	WAIT_UNTIL_CLEAR (V850E_CACHE_ICC & 0x1000);
-
-	/* 3. Set the TCLR0 and TCLR1 bits of the ICC register as follows,
-	      when clearing way 0 and way 1 at the same time:
-	        (a) Set the TCLR0 and TCLR1 bits.
-		(b) Read the TCLR0 and TCLR1 bits to confirm that these bits
-		    are cleared.
-		(c) Perform (a) and (b) above again.  */
-	V850E_CACHE_ICC |= 0x3;
-	WAIT_UNTIL_CLEAR (V850E_CACHE_ICC & 0x3);
-
-#ifdef V850E_CACHE_REPEAT_ICC_WRITE
-	/* Do it again.  */
-	V850E_CACHE_ICC |= 0x3;
-	WAIT_UNTIL_CLEAR (V850E_CACHE_ICC & 0x3);
-#endif
-}
-
-#ifdef V850E_CACHE_DCC
-/* Flush or clear (or both) the data cache, depending on the value of FLAGS;
-   the procedure is the same for both, just the control bits used differ (and
-   both may be performed simultaneously).  */
-static void dcache_op (unsigned short flags)
-{
-	/* 1. Read the data cache control register (DCC) and confirm that bits
-	      0, 1, 4, and 5 (DC00, DC01, DC04, DC05) are all cleared.  */
-	WAIT_UNTIL_CLEAR (V850E_CACHE_DCC & 0x33);
-
-	/* 2. Clear DCC register bit 12 (DC12), bit 13 (DC13), or both
-	      depending on the way for which tags are to be cleared.  */
-	V850E_CACHE_DCC &= ~0xC000;
-
-	/* 3. Set DCC register bit 0 (DC00), bit 1 (DC01) or both depending on
-	      the way for which tags are to be cleared.
-	      ...
-	      Set DCC register bit 4 (DC04), bit 5 (DC05), or both depending
-	      on the way to be data flushed.  */
-	V850E_CACHE_DCC |= flags;
-
-	/* 4. Read DCC register bit DC00, DC01 [DC04, DC05], or both depending
-	      on the way for which tags were cleared [flushed] and confirm
-	      that that bit is cleared.  */
-	WAIT_UNTIL_CLEAR (V850E_CACHE_DCC & flags);
-}
-#endif /* V850E_CACHE_DCC */
-
-/* Flushes the contents of the dcache to memory.  */
-static inline void flush_dcache (void)
-{
-#ifdef V850E_CACHE_DCC
-	/* We only need to do something if in write-back mode.  */
-	if (V850E_CACHE_DCC & 0x0400)
-		dcache_op (0x30);
-#endif /* V850E_CACHE_DCC */
-}
-
-/* Flushes the contents of the dcache to memory, and then clears it.  */
-static inline void clear_dcache (void)
-{
-#ifdef V850E_CACHE_DCC
-	/* We only need to do something if the dcache is enabled.  */
-	if (V850E_CACHE_DCC & 0x0C00)
-		dcache_op (0x33);
-#endif /* V850E_CACHE_DCC */
-}
-
-/* Clears the dcache without flushing to memory first.  */
-static inline void clear_dcache_no_flush (void)
-{
-#ifdef V850E_CACHE_DCC
-	/* We only need to do something if the dcache is enabled.  */
-	if (V850E_CACHE_DCC & 0x0C00)
-		dcache_op (0x3);
-#endif /* V850E_CACHE_DCC */
-}
-
-static inline void cache_exec_after_store (void)
-{
-	flush_dcache ();
-	clear_icache ();
-}
-
-
-/* Exported functions.  */
-
-void flush_icache (void)
-{
-	cache_exec_after_store ();
-}
-
-void flush_icache_range (unsigned long start, unsigned long end)
-{
-	cache_exec_after_store ();
-}
-
-void flush_icache_page (struct vm_area_struct *vma, struct page *page)
-{
-	cache_exec_after_store ();
-}
-
-void flush_icache_user_range (struct vm_area_struct *vma, struct page *page,
-			      unsigned long adr, int len)
-{
-	cache_exec_after_store ();
-}
-
-void flush_cache_sigtramp (unsigned long addr)
-{
-	cache_exec_after_store ();
-}
diff --git a/arch/v850/kernel/v850e_intc.c b/arch/v850/kernel/v850e_intc.c
deleted file mode 100644
index 8d39a52ee6d1..000000000000
--- a/arch/v850/kernel/v850e_intc.c
+++ /dev/null
@@ -1,104 +0,0 @@
-/*
- * arch/v850/kernel/v850e_intc.c -- V850E interrupt controller (INTC)
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/kernel.h>
-#include <linux/init.h>
-#include <linux/irq.h>
-
-#include <asm/v850e_intc.h>
-
-static void irq_nop (unsigned irq) { }
-
-static unsigned v850e_intc_irq_startup (unsigned irq)
-{
-	v850e_intc_clear_pending_irq (irq);
-	v850e_intc_enable_irq (irq);
-	return 0;
-}
-
-static void v850e_intc_end_irq (unsigned irq)
-{
-	unsigned long psw, temp;
-
-	/* Clear the highest-level bit in the In-service priority register
-	   (ISPR), to allow this interrupt (or another of the same or
-	   lesser priority) to happen again.
-
-	   The `reti' instruction normally does this automatically when the
-	   PSW bits EP and NP are zero, but we can't always rely on reti
-	   being used consistently to return after an interrupt (another
-	   process can be scheduled, for instance, which can delay the
-	   associated reti for a long time, or this process may be being
-	   single-stepped, which uses the `dbret' instruction to return
-	   from the kernel).
-
-	   We also set the PSW EP bit, which prevents reti from also
-	   trying to modify the ISPR itself.  */
-
-	/* Get PSW and disable interrupts.  */
-	asm volatile ("stsr psw, %0; di" : "=r" (psw));
-	/* We don't want to do anything for NMIs (they don't use the ISPR).  */
-	if (! (psw & 0xC0)) {
-		/* Transition to `trap' state, so that an eventual real
-		   reti instruction won't modify the ISPR.  */
-		psw |= 0x40;
-		/* Fake an interrupt return, which automatically clears the
-		   appropriate bit in the ISPR.  */
-		asm volatile ("mov hilo(1f), %0;"
-			      "ldsr %0, eipc; ldsr %1, eipsw;"
-			      "reti;"
-			      "1:"
-			      : "=&r" (temp) : "r" (psw));
-	}
-}
-
-/* Initialize HW_IRQ_TYPES for INTC-controlled irqs described in array
-   INITS (which is terminated by an entry with the name field == 0).  */
-void __init v850e_intc_init_irq_types (struct v850e_intc_irq_init *inits,
-				       struct hw_interrupt_type *hw_irq_types)
-{
-	struct v850e_intc_irq_init *init;
-	for (init = inits; init->name; init++) {
-		unsigned i;
-		struct hw_interrupt_type *hwit = hw_irq_types++;
-
-		hwit->typename = init->name;
-
-		hwit->startup  = v850e_intc_irq_startup;
-		hwit->shutdown = v850e_intc_disable_irq;
-		hwit->enable   = v850e_intc_enable_irq;
-		hwit->disable  = v850e_intc_disable_irq;
-		hwit->ack      = irq_nop;
-		hwit->end      = v850e_intc_end_irq;
-		
-		/* Initialize kernel IRQ infrastructure for this interrupt.  */
-		init_irq_handlers(init->base, init->num, init->interval, hwit);
-
-		/* Set the interrupt priorities.  */
-		for (i = 0; i < init->num; i++) {
-			unsigned irq = init->base + i * init->interval;
-
-			/* If the interrupt is currently enabled (all
-			   interrupts are initially disabled), then
-			   assume whoever enabled it has set things up
-			   properly, and avoid messing with it.  */
-			if (! v850e_intc_irq_enabled (irq))
-				/* This write also (1) disables the
-				   interrupt, and (2) clears any pending
-				   interrupts.  */
-				V850E_INTC_IC (irq)
-					= (V850E_INTC_IC_PR (init->priority)
-					   | V850E_INTC_IC_MK);
-		}
-	}
-}
diff --git a/arch/v850/kernel/v850e_timer_d.c b/arch/v850/kernel/v850e_timer_d.c
deleted file mode 100644
index d2a4ece2574c..000000000000
--- a/arch/v850/kernel/v850e_timer_d.c
+++ /dev/null
@@ -1,54 +0,0 @@
-/*
- * include/asm-v850/v850e_timer_d.c -- `Timer D' component often used
- *	with V850E CPUs
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/kernel.h>
-
-#include <asm/v850e_utils.h>
-#include <asm/v850e_timer_d.h>
-
-/* Start interval timer TIMER (0-3).  The timer will issue the
-   corresponding INTCMD interrupt RATE times per second.
-   This function does not enable the interrupt.  */
-void v850e_timer_d_configure (unsigned timer, unsigned rate)
-{
-	unsigned divlog2, count;
-
-	/* Calculate params for timer.  */
-	if (! calc_counter_params (
-		    V850E_TIMER_D_BASE_FREQ, rate,
-		    V850E_TIMER_D_TMCD_CS_MIN, V850E_TIMER_D_TMCD_CS_MAX, 16,
-		    &divlog2, &count))
-		printk (KERN_WARNING
-			"Cannot find interval timer %d setting suitable"
-			" for rate of %dHz.\n"
-			"Using rate of %dHz instead.\n",
-			timer, rate,
-			(V850E_TIMER_D_BASE_FREQ >> divlog2) >> 16);
-
-	/* Do the actual hardware timer initialization:  */
-
-	/* Enable timer.  */
-	V850E_TIMER_D_TMCD(timer) = V850E_TIMER_D_TMCD_CAE;
-	/* Set clock divider.  */
-	V850E_TIMER_D_TMCD(timer)
-		= V850E_TIMER_D_TMCD_CAE
-		| V850E_TIMER_D_TMCD_CS(divlog2);
-	/* Set timer compare register.  */
-	V850E_TIMER_D_CMD(timer) = count;
-	/* Start counting.  */
-	V850E_TIMER_D_TMCD(timer)
-		= V850E_TIMER_D_TMCD_CAE
-		| V850E_TIMER_D_TMCD_CS(divlog2)
-		| V850E_TIMER_D_TMCD_CE;
-}
diff --git a/arch/v850/kernel/v850e_utils.c b/arch/v850/kernel/v850e_utils.c
deleted file mode 100644
index e6807ef8dee6..000000000000
--- a/arch/v850/kernel/v850e_utils.c
+++ /dev/null
@@ -1,62 +0,0 @@
-/*
- * include/asm-v850/v850e_utils.h -- Utility functions associated with
- *	V850E CPUs
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <asm/v850e_utils.h>
-
-/* Calculate counter clock-divider and count values to attain the
-   desired frequency RATE from the base frequency BASE_FREQ.  The
-   counter is expected to have a clock-divider, which can divide the
-   system cpu clock by a power of two value from MIN_DIVLOG2 to
-   MAX_DIV_LOG2, and a word-size of COUNTER_SIZE bits (the counter
-   counts up and resets whenever it's equal to the compare register,
-   generating an interrupt or whatever when it does so).  The returned
-   values are: *DIVLOG2 -- log2 of the desired clock divider and *COUNT
-   -- the counter compare value to use.  Returns true if it was possible
-   to find a reasonable value, otherwise false (and the other return
-   values will be set to be as good as possible).  */
-int calc_counter_params (unsigned long base_freq,
-			 unsigned long rate,
-			 unsigned min_divlog2, unsigned max_divlog2,
-			 unsigned counter_size,
-			 unsigned *divlog2, unsigned *count)
-{
-	unsigned _divlog2;
-	int ok = 0;
-
-	/* Find the lowest clock divider setting that can represent RATE.  */
-	for (_divlog2 = min_divlog2; _divlog2 <= max_divlog2; _divlog2++) {
-		/* Minimum interrupt rate possible using this divider.  */
-		unsigned min_int_rate
-			= (base_freq >> _divlog2) >> counter_size;
-
-		if (min_int_rate <= rate) {
-			/* This setting is the highest resolution
-			   setting that's slow enough enough to attain
-			   RATE interrupts per second, so use it.  */
-			ok = 1;
-			break;
-		}
-	}
-
-	if (_divlog2 > max_divlog2)
-		/* Can't find correct setting.  */
-		_divlog2 = max_divlog2;
-
-	if (divlog2)
-		*divlog2 = _divlog2;
-	if (count)
-		*count = ((base_freq >> _divlog2) + rate/2) / rate;
-
-	return ok;
-}
diff --git a/arch/v850/kernel/vmlinux.lds.S b/arch/v850/kernel/vmlinux.lds.S
deleted file mode 100644
index d08cd1d27f27..000000000000
--- a/arch/v850/kernel/vmlinux.lds.S
+++ /dev/null
@@ -1,306 +0,0 @@
-/*
- * arch/v850/vmlinux.lds.S -- kernel linker script for v850 platforms
- *
- *  Copyright (C) 2002,03,04,05  NEC Electronics Corporation
- *  Copyright (C) 2002,03,04,05  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-
-#define VMLINUX_SYMBOL(_sym_) _##_sym_
-#include <asm-generic/vmlinux.lds.h>
-
-/* For most platforms, this will define useful things like RAM addr/size.  */
-#include <asm/machdep.h>
-
-
-/* The following macros contain the usual definitions for various data areas.
-   The prefix `RAMK_' is used to indicate macros suitable for kernels loaded
-   into RAM, and similarly `ROMK_' for ROM-resident kernels.  Note that all
-   symbols are prefixed with an extra `_' for compatibility with the v850
-   toolchain.  */
-
-	
-/* Interrupt vectors.  */
-#define INTV_CONTENTS							      \
-		. = ALIGN (0x10) ;					      \
-		__intv_start = . ;					      \
-			*(.intv.reset)	/* Reset vector */		      \
-		. = __intv_start + 0x10 ;				      \
-			*(.intv.common)	/* Vectors common to all v850e proc */\
-		. = __intv_start + 0x80 ;				      \
-			*(.intv.mach)	/* Machine-specific int. vectors.  */ \
-		__intv_end = . ;
-
-#define RODATA_CONTENTS							      \
-		. = ALIGN (16) ;					      \
-			*(.rodata) *(.rodata.*)				      \
-			*(__vermagic)		/* Kernel version magic */    \
-			*(.rodata1)					      \
-		/* PCI quirks */					      \
-		___start_pci_fixups_early = . ;				      \
-			*(.pci_fixup_early)				      \
-		___end_pci_fixups_early = . ;				      \
-		___start_pci_fixups_header = . ;			      \
-			*(.pci_fixup_header)				      \
-		___end_pci_fixups_header = . ;				      \
-		___start_pci_fixups_final = . ;				      \
-			*(.pci_fixup_final)				      \
-		___end_pci_fixups_final = . ;				      \
-		___start_pci_fixups_enable = . ;			      \
-			*(.pci_fixup_enable)				      \
-		___end_pci_fixups_enable = . ;				      \
-		/* Kernel symbol table: Normal symbols */		      \
-		___start___ksymtab = .;					      \
-			*(__ksymtab)					      \
-		___stop___ksymtab = .;					      \
-		/* Kernel symbol table: GPL-only symbols */		      \
-		___start___ksymtab_gpl = .;				      \
-			*(__ksymtab_gpl)				      \
-		___stop___ksymtab_gpl = .;				      \
-		/* Kernel symbol table: GPL-future symbols */		      \
-		___start___ksymtab_gpl_future = .;			      \
-			*(__ksymtab_gpl_future)				      \
-		___stop___ksymtab_gpl_future = .;			      \
-		/* Kernel symbol table: strings */			      \
-			*(__ksymtab_strings)				      \
-		/* Kernel symbol table: Normal symbols */		      \
-		___start___kcrctab = .;					      \
-			*(__kcrctab)					      \
-		___stop___kcrctab = .;					      \
-		/* Kernel symbol table: GPL-only symbols */		      \
-		___start___kcrctab_gpl = .;				      \
-			*(__kcrctab_gpl)				      \
-		___stop___kcrctab_gpl = .;				      \
-		/* Kernel symbol table: GPL-future symbols */		      \
-		___start___kcrctab_gpl_future = .;			      \
-			*(__kcrctab_gpl_future)				      \
-		___stop___kcrctab_gpl_future = .;			      \
-		/* Built-in module parameters */			      \
-		. = ALIGN (4) ;						      \
-		___start___param = .;					      \
-		*(__param)						      \
-		___stop___param = .;
-
-
-/* Kernel text segment, and some constant data areas.  */
-#define TEXT_CONTENTS							      \
-		_text = .;						      \
-		__stext = . ;						      \
-		TEXT_TEXT						      \
-		SCHED_TEXT						      \
-			*(.exit.text)	/* 2.5 convention */		      \
-			*(.text.exit)	/* 2.4 convention */		      \
-			*(.text.lock)					      \
-			*(.exitcall.exit)				      \
-		__real_etext = . ;	/* There may be data after here.  */  \
-		RODATA_CONTENTS						      \
-		. = ALIGN (4) ;						      \
-		    	*(.call_table_data)				      \
-			*(.call_table_text)				      \
-		. = ALIGN (16) ;	/* Exception table.  */		      \
-		___start___ex_table = . ;				      \
-			*(__ex_table)					      \
-		___stop___ex_table = . ;				      \
-		. = ALIGN (4) ;						      \
-		__etext = . ;
-
-/* Kernel data segment.  */
-#define DATA_CONTENTS							      \
-		__sdata = . ;						      \
-		DATA_DATA						      \
-			EXIT_DATA	/* 2.5 convention */		      \
-			*(.data.exit)	/* 2.4 convention */		      \
-		. = ALIGN (16) ;					      \
-		*(.data.cacheline_aligned)				      \
-		. = ALIGN (0x2000) ;					      \
-        	*(.data.init_task)					      \
-		. = ALIGN (0x2000) ;					      \
-		__edata = . ;
-
-/* Kernel BSS segment.  */
-#define BSS_CONTENTS							      \
-		__sbss = . ;						      \
-			*(.bss)						      \
-			*(COMMON)					      \
-		. = ALIGN (4) ;						      \
-		__init_stack_end = . ;					      \
-		__ebss = . ;
-
-/* `initcall' tables.  */
-#define INITCALL_CONTENTS						      \
-		. = ALIGN (16) ;					      \
-		___setup_start = . ;					      \
-			*(.init.setup)	/* 2.5 convention */		      \
-			*(.setup.init)	/* 2.4 convention */		      \
-		___setup_end = . ;					      \
-		___initcall_start = . ;					      \
-			*(.initcall.init)				      \
-			INITCALLS					      \
-		. = ALIGN (4) ;						      \
-		___initcall_end = . ;					      \
-		___con_initcall_start = .;				      \
-			*(.con_initcall.init)				      \
-		___con_initcall_end = .;
-
-/* Contents of `init' section for a kernel that's loaded into RAM.  */
-#define RAMK_INIT_CONTENTS						      \
-		RAMK_INIT_CONTENTS_NO_END				      \
-		__init_end = . ;
-/* Same as RAMK_INIT_CONTENTS, but doesn't define the `__init_end' symbol.  */
-#define RAMK_INIT_CONTENTS_NO_END					      \
-		. = ALIGN (4096) ;					      \
-		__init_start = . ;					      \
-			__sinittext = .;				      \
-			INIT_TEXT	/* 2.5 convention */		      \
-			__einittext = .;				      \
-			INIT_DATA					      \
-			*(.text.init)	/* 2.4 convention */		      \
-			*(.data.init)					      \
-		INITCALL_CONTENTS					      \
-		INITRAMFS_CONTENTS
-
-/* The contents of `init' section for a ROM-resident kernel which
-   should go into RAM.  */	
-#define ROMK_INIT_RAM_CONTENTS						      \
-		. = ALIGN (4096) ;					      \
-		__init_start = . ;					      \
-			INIT_DATA	/* 2.5 convention */		      \
-			*(.data.init)	/* 2.4 convention */		      \
-		__init_end = . ;					      \
-		. = ALIGN (4096) ;
-
-/* The contents of `init' section for a ROM-resident kernel which
-   should go into ROM.  */	
-#define ROMK_INIT_ROM_CONTENTS						      \
-			_sinittext = .;					      \
-			INIT_TEXT	/* 2.5 convention */		      \
-			_einittext = .;					      \
-			*(.text.init)	/* 2.4 convention */		      \
-		INITCALL_CONTENTS					      \
-		INITRAMFS_CONTENTS
-
-/* A root filesystem image, for kernels with an embedded root filesystem.  */
-#define ROOT_FS_CONTENTS						      \
-		__root_fs_image_start = . ;				      \
-		*(.root)						      \
-		__root_fs_image_end = . ;
-
-#ifdef CONFIG_BLK_DEV_INITRD
-/* The initramfs archive.  */
-#define INITRAMFS_CONTENTS						      \
-		. = ALIGN (4) ;						      \
-		___initramfs_start = . ;				      \
-			*(.init.ramfs)					      \
-		___initramfs_end = . ;
-#endif
-
-/* Where the initial bootmap (bitmap for the boot-time memory allocator) 
-   should be place.  */
-#define BOOTMAP_CONTENTS						      \
-		. = ALIGN (4096) ;					      \
-		__bootmap = . ;						      \
-		. = . + 4096 ;		/* enough for 128MB.   */
-
-/* The contents of a `typical' kram area for a kernel in RAM.  */
-#define RAMK_KRAM_CONTENTS						      \
-		__kram_start = . ;					      \
-		TEXT_CONTENTS						      \
-		DATA_CONTENTS						      \
-		BSS_CONTENTS						      \
-		RAMK_INIT_CONTENTS					      \
-		__kram_end = . ;					      \
-		BOOTMAP_CONTENTS
-
-
-/* Define output sections normally used for a ROM-resident kernel.  
-   ROM and RAM should be appropriate memory areas to use for kernel
-   ROM and RAM data.  This assumes that ROM starts at 0 (and thus can
-   hold the interrupt vectors).  */
-#define ROMK_SECTIONS(ROM, RAM)						      \
-	.rom : {							      \
-		INTV_CONTENTS						      \
-		TEXT_CONTENTS						      \
-		ROMK_INIT_ROM_CONTENTS					      \
-		ROOT_FS_CONTENTS					      \
-	} > ROM								      \
-									      \
-	__rom_copy_src_start = . ;					      \
-									      \
-	.data : {							      \
-		__kram_start = . ;					      \
-		__rom_copy_dst_start = . ;				      \
-		DATA_CONTENTS						      \
-		ROMK_INIT_RAM_CONTENTS					      \
-		__rom_copy_dst_end = . ;				      \
-	} > RAM  AT> ROM						      \
-									      \
-	.bss ALIGN (4) : {						      \
-		BSS_CONTENTS						      \
-		__kram_end = . ;					      \
-		BOOTMAP_CONTENTS					      \
-	} > RAM
-
-
-/* The 32-bit variable `jiffies' is just the lower 32-bits of `jiffies_64'.  */
-_jiffies = _jiffies_64 ;
-
-
-/* Include an appropriate platform-dependent linker-script (which
-   usually should use the above macros to do most of the work).  */
-
-#ifdef CONFIG_V850E_SIM
-# include "sim.ld"
-#endif
-
-#ifdef CONFIG_V850E2_SIM85E2
-# include "sim85e2.ld"
-#endif
-
-#ifdef CONFIG_V850E2_FPGA85E2C
-# include "fpga85e2c.ld"
-#endif
-
-#ifdef CONFIG_V850E2_ANNA
-# ifdef CONFIG_ROM_KERNEL
-#  include "anna-rom.ld"
-# else
-#  include "anna.ld"
-# endif
-#endif
-
-#ifdef CONFIG_V850E_AS85EP1
-# ifdef CONFIG_ROM_KERNEL
-#  include "as85ep1-rom.ld"
-# else
-#  include "as85ep1.ld"
-# endif
-#endif
-
-#ifdef CONFIG_RTE_CB_MA1
-# ifdef CONFIG_ROM_KERNEL
-#  include "rte_ma1_cb-rom.ld"
-# else
-#  include "rte_ma1_cb.ld"
-# endif
-#endif
-
-#ifdef CONFIG_RTE_CB_NB85E
-# ifdef CONFIG_ROM_KERNEL
-#  include "rte_nb85e_cb-rom.ld"
-# elif defined(CONFIG_RTE_CB_MULTI)
-#  include "rte_nb85e_cb-multi.ld"
-# else
-#  include "rte_nb85e_cb.ld"
-# endif
-#endif
-
-#ifdef CONFIG_RTE_CB_ME2
-#  include "rte_me2_cb.ld"
-#endif
-
diff --git a/arch/v850/lib/Makefile b/arch/v850/lib/Makefile
deleted file mode 100644
index 1c78b728a117..000000000000
--- a/arch/v850/lib/Makefile
+++ /dev/null
@@ -1,6 +0,0 @@
-#
-# arch/v850/lib/Makefile
-#
-
-lib-y  = ashrdi3.o ashldi3.o lshrdi3.o muldi3.o negdi2.o \
-	 checksum.o memcpy.o memset.o
diff --git a/arch/v850/lib/ashldi3.c b/arch/v850/lib/ashldi3.c
deleted file mode 100644
index 9e792d53f0e4..000000000000
--- a/arch/v850/lib/ashldi3.c
+++ /dev/null
@@ -1,62 +0,0 @@
-/* ashldi3.c extracted from gcc-2.95.2/libgcc2.c which is: */
-/* Copyright (C) 1989, 92-98, 1999 Free Software Foundation, Inc.
-
-This file is part of GNU CC.
-
-GNU CC is free software; you can redistribute it and/or modify
-it under the terms of the GNU General Public License as published by
-the Free Software Foundation; either version 2, or (at your option)
-any later version.
-
-GNU CC is distributed in the hope that it will be useful,
-but WITHOUT ANY WARRANTY; without even the implied warranty of
-MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-GNU General Public License for more details.
-
-You should have received a copy of the GNU General Public License
-along with GNU CC; see the file COPYING.  If not, write to
-the Free Software Foundation, 59 Temple Place - Suite 330,
-Boston, MA 02111-1307, USA.  */
-
-#define BITS_PER_UNIT 8
-
-typedef 	 int SItype	__attribute__ ((mode (SI)));
-typedef unsigned int USItype	__attribute__ ((mode (SI)));
-typedef		 int DItype	__attribute__ ((mode (DI)));
-typedef int word_type __attribute__ ((mode (__word__)));
-
-struct DIstruct {SItype high, low;};
-
-typedef union
-{
-  struct DIstruct s;
-  DItype ll;
-} DIunion;
-
-DItype
-__ashldi3 (DItype u, word_type b)
-{
-  DIunion w;
-  word_type bm;
-  DIunion uu;
-
-  if (b == 0)
-    return u;
-
-  uu.ll = u;
-
-  bm = (sizeof (SItype) * BITS_PER_UNIT) - b;
-  if (bm <= 0)
-    {
-      w.s.low = 0;
-      w.s.high = (USItype)uu.s.low << -bm;
-    }
-  else
-    {
-      USItype carries = (USItype)uu.s.low >> bm;
-      w.s.low = (USItype)uu.s.low << b;
-      w.s.high = ((USItype)uu.s.high << b) | carries;
-    }
-
-  return w.ll;
-}
diff --git a/arch/v850/lib/ashrdi3.c b/arch/v850/lib/ashrdi3.c
deleted file mode 100644
index 78efb65e315a..000000000000
--- a/arch/v850/lib/ashrdi3.c
+++ /dev/null
@@ -1,63 +0,0 @@
-/* ashrdi3.c extracted from gcc-2.7.2/libgcc2.c which is: */
-/* Copyright (C) 1989, 1992, 1993, 1994, 1995 Free Software Foundation, Inc.
-
-This file is part of GNU CC.
-
-GNU CC is free software; you can redistribute it and/or modify
-it under the terms of the GNU General Public License as published by
-the Free Software Foundation; either version 2, or (at your option)
-any later version.
-
-GNU CC is distributed in the hope that it will be useful,
-but WITHOUT ANY WARRANTY; without even the implied warranty of
-MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-GNU General Public License for more details.
-
-You should have received a copy of the GNU General Public License
-along with GNU CC; see the file COPYING.  If not, write to
-the Free Software Foundation, 59 Temple Place - Suite 330,
-Boston, MA 02111-1307, USA.  */
-
-#define BITS_PER_UNIT 8
-
-typedef 	 int SItype	__attribute__ ((mode (SI)));
-typedef unsigned int USItype	__attribute__ ((mode (SI)));
-typedef		 int DItype	__attribute__ ((mode (DI)));
-typedef int word_type __attribute__ ((mode (__word__)));
-
-struct DIstruct {SItype high, low;};
-
-typedef union
-{
-  struct DIstruct s;
-  DItype ll;
-} DIunion;
-
-DItype
-__ashrdi3 (DItype u, word_type b)
-{
-  DIunion w;
-  word_type bm;
-  DIunion uu;
-
-  if (b == 0)
-    return u;
-
-  uu.ll = u;
-
-  bm = (sizeof (SItype) * BITS_PER_UNIT) - b;
-  if (bm <= 0)
-    {
-      /* w.s.high = 1..1 or 0..0 */
-      w.s.high = uu.s.high >> (sizeof (SItype) * BITS_PER_UNIT - 1);
-      w.s.low = uu.s.high >> -bm;
-    }
-  else
-    {
-      USItype carries = (USItype)uu.s.high << bm;
-      w.s.high = uu.s.high >> b;
-      w.s.low = ((USItype)uu.s.low >> b) | carries;
-    }
-
-  return w.ll;
-}
diff --git a/arch/v850/lib/checksum.c b/arch/v850/lib/checksum.c
deleted file mode 100644
index 042158dfe17a..000000000000
--- a/arch/v850/lib/checksum.c
+++ /dev/null
@@ -1,155 +0,0 @@
-/*
- * INET		An implementation of the TCP/IP protocol suite for the LINUX
- *		operating system.  INET is implemented using the  BSD Socket
- *		interface as the means of communication with the user level.
- *
- *		MIPS specific IP/TCP/UDP checksumming routines
- *
- * Authors:	Ralf Baechle, <ralf@waldorf-gmbh.de>
- *		Lots of code moved from tcp.c and ip.c; see those files
- *		for more names.
- *
- *		This program is free software; you can redistribute it and/or
- *		modify it under the terms of the GNU General Public License
- *		as published by the Free Software Foundation; either version
- *		2 of the License, or (at your option) any later version.
- *
- * $Id: checksum.c,v 1.1 2002/09/28 14:58:40 gerg Exp $
- */
-#include <net/checksum.h>
-#include <linux/module.h>
-#include <linux/types.h>
-#include <asm/byteorder.h>
-#include <asm/string.h>
-#include <asm/uaccess.h>
-
-static inline unsigned short from32to16 (unsigned long sum)
-{
-	unsigned int result;
-	/*
-			        %0		%1
-	      hsw %1, %0	H     L		L     H
-	      add %1, %0	H     L		H+L+C H+L
-	*/
-	asm ("hsw %1, %0; add %1, %0" : "=&r" (result) : "r" (sum));
-	return result >> 16;
-}
-
-static inline unsigned int do_csum(const unsigned char * buff, int len)
-{
-	int odd, count;
-	unsigned int result = 0;
-
-	if (len <= 0)
-		goto out;
-	odd = 1 & (unsigned long) buff;
-	if (odd) {
-		result = be16_to_cpu(*buff);
-		len--;
-		buff++;
-	}
-	count = len >> 1;		/* nr of 16-bit words.. */
-	if (count) {
-		if (2 & (unsigned long) buff) {
-			result += *(unsigned short *) buff;
-			count--;
-			len -= 2;
-			buff += 2;
-		}
-		count >>= 1;		/* nr of 32-bit words.. */
-		if (count) {
-			unsigned int carry = 0;
-			do {
-				unsigned int w = *(unsigned int *) buff;
-				count--;
-				buff += 4;
-				result += carry;
-				result += w;
-				carry = (w > result);
-			} while (count);
-			result += carry;
-			result = (result & 0xffff) + (result >> 16);
-		}
-		if (len & 2) {
-			result += *(unsigned short *) buff;
-			buff += 2;
-		}
-	}
-	if (len & 1)
-		result += le16_to_cpu(*buff);
-	result = from32to16(result);
-	if (odd)
-		result = ((result >> 8) & 0xff) | ((result & 0xff) << 8);
-out:
-	return result;
-}
-
-/*
- *	This is a version of ip_compute_csum() optimized for IP headers,
- *	which always checksum on 4 octet boundaries.
- */
-__sum16 ip_fast_csum(const void *iph, unsigned int ihl)
-{
-	return (__force __sum16)~do_csum(iph,ihl*4);
-}
-
-/*
- * this routine is used for miscellaneous IP-like checksums, mainly
- * in icmp.c
- */
-__sum16 ip_compute_csum(const void *buff, int len)
-{
-	return (__force __sum16)~do_csum(buff,len);
-}
-
-/*
- * computes a partial checksum, e.g. for TCP/UDP fragments
- */
-__wsum csum_partial(const void *buff, int len, __wsum sum)
-{
-	unsigned int result = do_csum(buff, len);
-
-	/* add in old sum, and carry.. */
-	result += (__force u32)sum;
-	if ((__force u32)sum > result)
-		result += 1;
-	return (__force __wsum)result;
-}
-
-EXPORT_SYMBOL(csum_partial);
-
-/*
- * copy while checksumming, otherwise like csum_partial
- */
-__wsum csum_partial_copy_nocheck(const void *src, void *dst,
-                               int len, __wsum sum)
-{
-	/*
-	 * It's 2:30 am and I don't feel like doing it real ...
-	 * This is lots slower than the real thing (tm)
-	 */
-	sum = csum_partial(src, len, sum);
-	memcpy(dst, src, len);
-
-	return sum;
-}
-
-/*
- * Copy from userspace and compute checksum.  If we catch an exception
- * then zero the rest of the buffer.
- */
-__wsum csum_partial_copy_from_user (const void *src,
-					  void *dst,
-                                          int len, __wsum sum,
-                                          int *err_ptr)
-{
-	int missing;
-
-	missing = copy_from_user(dst, src, len);
-	if (missing) {
-		memset(dst + len - missing, 0, missing);
-		*err_ptr = -EFAULT;
-	}
-		
-	return csum_partial(dst, len, sum);
-}
diff --git a/arch/v850/lib/lshrdi3.c b/arch/v850/lib/lshrdi3.c
deleted file mode 100644
index 93b1cb6fdee8..000000000000
--- a/arch/v850/lib/lshrdi3.c
+++ /dev/null
@@ -1,62 +0,0 @@
-/* lshrdi3.c extracted from gcc-2.7.2/libgcc2.c which is: */
-/* Copyright (C) 1989, 1992, 1993, 1994, 1995 Free Software Foundation, Inc.
-
-This file is part of GNU CC.
-
-GNU CC is free software; you can redistribute it and/or modify
-it under the terms of the GNU General Public License as published by
-the Free Software Foundation; either version 2, or (at your option)
-any later version.
-
-GNU CC is distributed in the hope that it will be useful,
-but WITHOUT ANY WARRANTY; without even the implied warranty of
-MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-GNU General Public License for more details.
-
-You should have received a copy of the GNU General Public License
-along with GNU CC; see the file COPYING.  If not, write to
-the Free Software Foundation, 59 Temple Place - Suite 330,
-Boston, MA 02111-1307, USA.  */
-
-#define BITS_PER_UNIT 8
-
-typedef 	 int SItype	__attribute__ ((mode (SI)));
-typedef unsigned int USItype	__attribute__ ((mode (SI)));
-typedef		 int DItype	__attribute__ ((mode (DI)));
-typedef int word_type __attribute__ ((mode (__word__)));
-
-struct DIstruct {SItype high, low;};
-
-typedef union
-{
-  struct DIstruct s;
-  DItype ll;
-} DIunion;
-
-DItype
-__lshrdi3 (DItype u, word_type b)
-{
-  DIunion w;
-  word_type bm;
-  DIunion uu;
-
-  if (b == 0)
-    return u;
-
-  uu.ll = u;
-
-  bm = (sizeof (SItype) * BITS_PER_UNIT) - b;
-  if (bm <= 0)
-    {
-      w.s.high = 0;
-      w.s.low = (USItype)uu.s.high >> -bm;
-    }
-  else
-    {
-      USItype carries = (USItype)uu.s.high << bm;
-      w.s.high = (USItype)uu.s.high >> b;
-      w.s.low = ((USItype)uu.s.low >> b) | carries;
-    }
-
-  return w.ll;
-}
diff --git a/arch/v850/lib/memcpy.c b/arch/v850/lib/memcpy.c
deleted file mode 100644
index 492847b3e612..000000000000
--- a/arch/v850/lib/memcpy.c
+++ /dev/null
@@ -1,92 +0,0 @@
-/*
- * arch/v850/lib/memcpy.c -- Memory copying
- *
- *  Copyright (C) 2001,02  NEC Corporation
- *  Copyright (C) 2001,02  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/types.h>
-#include <asm/string.h>
-
-#define CHUNK_SIZE		32 /* bytes */
-#define CHUNK_ALIGNED(addr)	(((unsigned long)addr & 0x3) == 0)
-
-/* Note that this macro uses 8 call-clobbered registers (not including
-   R1), which are few enough so that the following functions don't need
-   to spill anything to memory.  It also uses R1, which is nominally
-   reserved for the assembler, but here it should be OK.  */
-#define COPY_CHUNK(src, dst)			\
-   asm ("mov %0, ep;"				\
-	"sld.w 0[ep], r1; sld.w 4[ep], r12;"	\
-	"sld.w 8[ep], r13; sld.w 12[ep], r14;"	\
-	"sld.w 16[ep], r15; sld.w 20[ep], r17;"	\
-	"sld.w 24[ep], r18; sld.w 28[ep], r19;"	\
-	"mov %1, ep;"				\
-	"sst.w r1, 0[ep]; sst.w r12, 4[ep];"	\
-	"sst.w r13, 8[ep]; sst.w r14, 12[ep];"	\
-	"sst.w r15, 16[ep]; sst.w r17, 20[ep];"	\
-	"sst.w r18, 24[ep]; sst.w r19, 28[ep]"	\
-	:: "r" (src), "r" (dst)			\
-	: "r1", "r12", "r13", "r14", "r15",	\
-	  "r17", "r18", "r19", "ep", "memory");
-
-void *memcpy (void *dst, const void *src, __kernel_size_t size)
-{
-	char *_dst = dst;
-	const char *_src = src;
-
-	if (size >= CHUNK_SIZE && CHUNK_ALIGNED(_src) && CHUNK_ALIGNED(_dst)) {
-		/* Copy large blocks efficiently.  */
-		unsigned count;
-		for (count = size / CHUNK_SIZE; count; count--) {
-			COPY_CHUNK (_src, _dst);
-			_src += CHUNK_SIZE;
-			_dst += CHUNK_SIZE;
-		}
-		size %= CHUNK_SIZE;
-	}
-
-	if (size > 0)
-		do
-			*_dst++ = *_src++;
-		while (--size);
-
-	return dst;
-}
-
-void *memmove (void *dst, const void *src, __kernel_size_t size)
-{
-	if ((unsigned long)dst < (unsigned long)src
-	    || (unsigned long)src + size < (unsigned long)dst)
-		return memcpy (dst, src, size);
-	else {
-		char *_dst = dst + size;
-		const char *_src = src + size;
-
-		if (size >= CHUNK_SIZE
-		    && CHUNK_ALIGNED (_src) && CHUNK_ALIGNED (_dst))
-		{
-			/* Copy large blocks efficiently.  */
-			unsigned count;
-			for (count = size / CHUNK_SIZE; count; count--) {
-				_src -= CHUNK_SIZE;
-				_dst -= CHUNK_SIZE;
-				COPY_CHUNK (_src, _dst);
-			}
-			size %= CHUNK_SIZE;
-		}
-
-		if (size > 0)
-			do
-				*--_dst = *--_src;
-			while (--size);
-
-		return _dst;
-	}
-}
diff --git a/arch/v850/lib/memset.c b/arch/v850/lib/memset.c
deleted file mode 100644
index d1b2ad821b15..000000000000
--- a/arch/v850/lib/memset.c
+++ /dev/null
@@ -1,68 +0,0 @@
-/*
- * arch/v850/lib/memset.c -- Memory initialization
- *
- *  Copyright (C) 2001,02,04  NEC Corporation
- *  Copyright (C) 2001,02,04  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#include <linux/types.h>
-
-void *memset (void *dst, int val, __kernel_size_t count)
-{
-	if (count) {
-		register unsigned loop;
-		register void *ptr asm ("ep") = dst;
-
-		/* replicate VAL into a long.  */
-		val &= 0xff;
-		val |= val << 8;
-		val |= val << 16;
-
-		/* copy initial unaligned bytes.  */
-		if ((long)ptr & 1) {
-			*(char *)ptr = val;
-			ptr = (void *)((char *)ptr + 1);
-			count--;
-		}
-		if (count > 2 && ((long)ptr & 2)) {
-			*(short *)ptr = val;
-			ptr = (void *)((short *)ptr + 1);
-			count -= 2;
-		}
-
-		/* 32-byte copying loop.  */
-		for (loop = count / 32; loop; loop--) {
-			asm ("sst.w %0, 0[ep]; sst.w %0, 4[ep];"
-			     "sst.w %0, 8[ep]; sst.w %0, 12[ep];"
-			     "sst.w %0, 16[ep]; sst.w %0, 20[ep];"
-			     "sst.w %0, 24[ep]; sst.w %0, 28[ep]"
-			     :: "r" (val) : "memory");
-			ptr += 32;
-		}
-		count %= 32;
-
-		/* long copying loop.  */
-		for (loop = count / 4; loop; loop--) {
-			*(long *)ptr = val;
-			ptr = (void *)((long *)ptr + 1);
-		}
-		count %= 4;
-
-		/* finish up with any trailing bytes.  */
-		if (count & 2) {
-			*(short *)ptr = val;
-			ptr = (void *)((short *)ptr + 1);
-		}
-		if (count & 1) {
-			*(char *)ptr = val;
-		}
-	}
-
-	return dst;
-}
diff --git a/arch/v850/lib/muldi3.c b/arch/v850/lib/muldi3.c
deleted file mode 100644
index 277ca25c82c8..000000000000
--- a/arch/v850/lib/muldi3.c
+++ /dev/null
@@ -1,61 +0,0 @@
-/* muldi3.c extracted from gcc-2.7.2.3/libgcc2.c and 
-			   gcc-2.7.2.3/longlong.h which is: */
-/* Copyright (C) 1989, 1992, 1993, 1994, 1995, 2001 Free Software Foundation, Inc.
-
-This file is part of GNU CC.
-
-GNU CC is free software; you can redistribute it and/or modify
-it under the terms of the GNU General Public License as published by
-the Free Software Foundation; either version 2, or (at your option)
-any later version.
-
-GNU CC is distributed in the hope that it will be useful,
-but WITHOUT ANY WARRANTY; without even the implied warranty of
-MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-GNU General Public License for more details.
-
-You should have received a copy of the GNU General Public License
-along with GNU CC; see the file COPYING.  If not, write to
-the Free Software Foundation, 59 Temple Place - Suite 330,
-Boston, MA 02111-1307, USA.  */
-
-#define umul_ppmm(w1, w0, u, v) \
-  __asm__ ("mulu %3, %0, %1"						\
-           : "=r" ((USItype)(w0)),					\
-             "=r" ((USItype)(w1))					\
-           : "%0" ((USItype)(u)),					\
-             "r" ((USItype)(v)))
-
-#define __umulsidi3(u, v) \
-  ({DIunion __w;							\
-    umul_ppmm (__w.s.high, __w.s.low, u, v);				\
-    __w.ll; })
-
-typedef 	 int SItype	__attribute__ ((mode (SI)));
-typedef unsigned int USItype	__attribute__ ((mode (SI)));
-typedef		 int DItype	__attribute__ ((mode (DI)));
-typedef int word_type __attribute__ ((mode (__word__)));
-
-struct DIstruct {SItype high, low;};
-
-typedef union
-{
-  struct DIstruct s;
-  DItype ll;
-} DIunion;
-
-DItype
-__muldi3 (DItype u, DItype v)
-{
-  DIunion w;
-  DIunion uu, vv;
-
-  uu.ll = u,
-  vv.ll = v;
-
-  w.ll = __umulsidi3 (uu.s.low, vv.s.low);
-  w.s.high += ((USItype) uu.s.low * (USItype) vv.s.high
-	       + (USItype) uu.s.high * (USItype) vv.s.low);
-
-  return w.ll;
-}
diff --git a/arch/v850/lib/negdi2.c b/arch/v850/lib/negdi2.c
deleted file mode 100644
index 571e04fc619a..000000000000
--- a/arch/v850/lib/negdi2.c
+++ /dev/null
@@ -1,25 +0,0 @@
-/*
- * arch/v850/lib/negdi2.c -- 64-bit negation
- *
- *  Copyright (C) 2001  NEC Corporation
- *  Copyright (C) 2001  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-typedef		 int DItype	__attribute__ ((mode (DI)));
-
-DItype __negdi2 (DItype x)
-{
-	__asm__ __volatile__
-		("not	r6, r10;"
-		 "add	1, r10;"
-		 "setf	c, r6;"
-		 "not	r7, r11;"
-		 "add	r6, r11"
-		 ::: "r6", "r7", "r10", "r11");
-}
diff --git a/drivers/serial/Kconfig b/drivers/serial/Kconfig
index 8fc7451c0049..3b4a14e355c1 100644
--- a/drivers/serial/Kconfig
+++ b/drivers/serial/Kconfig
@@ -942,22 +942,6 @@ config SERIAL_IP22_ZILOG_CONSOLE
 	depends on SERIAL_IP22_ZILOG=y
 	select SERIAL_CORE_CONSOLE
 
-config V850E_UART
-	bool "NEC V850E on-chip UART support"
-	depends on V850E_MA1 || V850E_ME2 || V850E_TEG || V850E2_ANNA || V850E_AS85EP1
-	select SERIAL_CORE
-	default y
-
-config V850E_UARTB
-        bool
-	depends on V850E_UART && V850E_ME2
-	default y
-
-config V850E_UART_CONSOLE
-	bool "Use NEC V850E on-chip UART for console"
-	depends on V850E_UART
-	select SERIAL_CORE_CONSOLE
-
 config SERIAL_SH_SCI
 	tristate "SuperH SCI(F) serial port support"
 	depends on SUPERH || H8300
diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig
index ccb78f66c2b6..48399e134c0d 100644
--- a/drivers/watchdog/Kconfig
+++ b/drivers/watchdog/Kconfig
@@ -788,8 +788,6 @@ config WATCHDOG_RIO
 	  machines.  The watchdog timeout period is normally one minute but
 	  can be changed with a boot-time parameter.
 
-# V850 Architecture
-
 # XTENSA Architecture
 
 #
diff --git a/drivers/watchdog/Makefile b/drivers/watchdog/Makefile
index 25b352b664d9..edd305a64e63 100644
--- a/drivers/watchdog/Makefile
+++ b/drivers/watchdog/Makefile
@@ -119,8 +119,6 @@ obj-$(CONFIG_SH_WDT) += shwdt.o
 
 # SPARC64 Architecture
 
-# V850 Architecture
-
 # XTENSA Architecture
 
 # Architecture Independant
diff --git a/include/asm-v850/Kbuild b/include/asm-v850/Kbuild
deleted file mode 100644
index c68e1680da01..000000000000
--- a/include/asm-v850/Kbuild
+++ /dev/null
@@ -1 +0,0 @@
-include include/asm-generic/Kbuild.asm
diff --git a/include/asm-v850/a.out.h b/include/asm-v850/a.out.h
deleted file mode 100644
index e9439a0708f6..000000000000
--- a/include/asm-v850/a.out.h
+++ /dev/null
@@ -1,21 +0,0 @@
-#ifndef __V850_A_OUT_H__
-#define __V850_A_OUT_H__
-
-struct exec
-{
-  unsigned long a_info;		/* Use macros N_MAGIC, etc for access */
-  unsigned a_text;		/* length of text, in bytes */
-  unsigned a_data;		/* length of data, in bytes */
-  unsigned a_bss;		/* length of uninitialized data area for file, in bytes */
-  unsigned a_syms;		/* length of symbol table data in file, in bytes */
-  unsigned a_entry;		/* start address */
-  unsigned a_trsize;		/* length of relocation info for text, in bytes */
-  unsigned a_drsize;		/* length of relocation info for data, in bytes */
-};
-
-#define N_TRSIZE(a)	((a).a_trsize)
-#define N_DRSIZE(a)	((a).a_drsize)
-#define N_SYMSIZE(a)	((a).a_syms)
-
-
-#endif /* __V850_A_OUT_H__ */
diff --git a/include/asm-v850/anna.h b/include/asm-v850/anna.h
deleted file mode 100644
index cd5eaee103b0..000000000000
--- a/include/asm-v850/anna.h
+++ /dev/null
@@ -1,137 +0,0 @@
-/*
- * include/asm-v850/anna.h -- Anna V850E2 evaluation cpu chip/board
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_ANNA_H__
-#define __V850_ANNA_H__
-
-#include <asm/v850e2.h>		/* Based on V850E2 core.  */
-
-
-#define CPU_MODEL	"v850e2/anna"
-#define CPU_MODEL_LONG	"NEC V850E2/Anna"
-#define PLATFORM	"anna"
-#define PLATFORM_LONG	"NEC/Midas lab V850E2/Anna evaluation board"
-
-#define CPU_CLOCK_FREQ	200000000 /*  200MHz */
-#define SYS_CLOCK_FREQ	 33300000 /* 33.3MHz */
-
-
-/* 1MB of static RAM.  This memory is mirrored 64 times.  */
-#define SRAM_ADDR	0x04000000
-#define SRAM_SIZE	0x00100000 /* 1MB */
-/* 64MB of DRAM.  */
-#define SDRAM_ADDR	0x08000000	
-#define SDRAM_SIZE	0x04000000 /* 64MB */
-
-
-/* For <asm/page.h> */
-#define PAGE_OFFSET 	SRAM_ADDR
-
-/* We use on-chip RAM, for a few miscellaneous variables that must be
-   accessible using a load instruction relative to R0.  The Anna chip has
-   128K of `dLB' ram nominally located at 0xFFF00000, but it's mirrored
-   every 128K, so we can use the `last mirror' (except for the portion at
-   the top which is overridden by I/O space).  In addition, the early
-   sample chip we're using has lots of memory errors in the dLB ram, so we
-   use a specially chosen location that has at least 20 bytes of contiguous
-   valid memory (xxxF0020 - xxxF003F).  */
-#define R0_RAM_ADDR			0xFFFF8020
-
-
-/* Anna specific control registers.  */
-#define ANNA_ILBEN_ADDR			0xFFFFF7F2
-#define ANNA_ILBEN			(*(volatile u16 *)ANNA_ILBEN_ADDR)
-
-
-/* I/O port P0-P3. */
-/* Direct I/O.  Bits 0-7 are pins Pn0-Pn7.  */
-#define ANNA_PORT_IO_ADDR(n)		(0xFFFFF400 + (n) * 2)
-#define ANNA_PORT_IO(n)			(*(volatile u8 *)ANNA_PORT_IO_ADDR(n))
-/* Port mode (for direct I/O, 0 = output, 1 = input).  */
-#define ANNA_PORT_PM_ADDR(n)		(0xFFFFF410 + (n) * 2)
-#define ANNA_PORT_PM(n)			(*(volatile u8 *)ANNA_PORT_PM_ADDR(n))
-
-
-/* Hardware-specific interrupt numbers (in the kernel IRQ namespace).  */
-#define IRQ_INTP(n)	(n)	/* Pnnn (pin) interrupts 0-15 */
-#define IRQ_INTP_NUM	16
-#define IRQ_INTOV(n)	(0x10 + (n)) /* 0-2 */
-#define IRQ_INTOV_NUM	2
-#define IRQ_INTCCC(n)	(0x12 + (n))
-#define IRQ_INTCCC_NUM	4
-#define IRQ_INTCMD(n)	(0x16 + (n)) /* interval timer interrupts 0-5 */
-#define IRQ_INTCMD_NUM	6
-#define IRQ_INTDMA(n)	(0x1C + (n)) /* DMA interrupts 0-3 */
-#define IRQ_INTDMA_NUM	4
-#define IRQ_INTDMXER	0x20
-#define IRQ_INTSRE(n)	(0x21 + (n)*3) /* UART 0-1 reception error */
-#define IRQ_INTSRE_NUM	2
-#define IRQ_INTSR(n)	(0x22 + (n)*3) /* UART 0-1 reception completion */
-#define IRQ_INTSR_NUM	2
-#define IRQ_INTST(n)	(0x23 + (n)*3) /* UART 0-1 transmission completion */
-#define IRQ_INTST_NUM	2
-
-#define NUM_CPU_IRQS	64
-
-#ifndef __ASSEMBLY__
-/* Initialize chip interrupts.  */
-extern void anna_init_irqs (void);
-#endif
-
-
-/* Anna UART details (basically the same as the V850E/MA1, but 2 channels).  */
-#define V850E_UART_NUM_CHANNELS		2
-#define V850E_UART_BASE_FREQ		(SYS_CLOCK_FREQ / 2)
-#define V850E_UART_CHIP_NAME 		"V850E2/NA85E2A"
-
-/* This is the UART channel that's actually connected on the board.  */
-#define V850E_UART_CONSOLE_CHANNEL	1
-
-/* This is a function that gets called before configuring the UART.  */
-#define V850E_UART_PRE_CONFIGURE	anna_uart_pre_configure
-#ifndef __ASSEMBLY__
-extern void anna_uart_pre_configure (unsigned chan,
-				     unsigned cflags, unsigned baud);
-#endif
-
-/* This board supports RTS/CTS for the on-chip UART, but only for channel 1. */
-
-/* CTS for UART channel 1 is pin P37 (bit 7 of port 3).  */
-#define V850E_UART_CTS(chan)	((chan) == 1 ? !(ANNA_PORT_IO(3) & 0x80) : 1)
-/* RTS for UART channel 1 is pin P07 (bit 7 of port 0).  */
-#define V850E_UART_SET_RTS(chan, val)					      \
-   do {									      \
-	   if (chan == 1) {						      \
-		   unsigned old = ANNA_PORT_IO(0); 			      \
-		   if (val)						      \
-			   ANNA_PORT_IO(0) = old & ~0x80;		      \
-		   else							      \
-			   ANNA_PORT_IO(0) = old | 0x80;		      \
-	   }								      \
-   } while (0)
-
-
-/* Timer C details.  */
-#define V850E_TIMER_C_BASE_ADDR		0xFFFFF600
-
-/* Timer D details (the Anna actually has 5 of these; should change later). */
-#define V850E_TIMER_D_BASE_ADDR		0xFFFFF540
-#define V850E_TIMER_D_TMD_BASE_ADDR 	(V850E_TIMER_D_BASE_ADDR + 0x0)
-#define V850E_TIMER_D_CMD_BASE_ADDR 	(V850E_TIMER_D_BASE_ADDR + 0x2)
-#define V850E_TIMER_D_TMCD_BASE_ADDR 	(V850E_TIMER_D_BASE_ADDR + 0x4)
-
-#define V850E_TIMER_D_BASE_FREQ		SYS_CLOCK_FREQ
-#define V850E_TIMER_D_TMCD_CS_MIN	1 /* min 2^1 divider */
-
-
-#endif /* __V850_ANNA_H__ */
diff --git a/include/asm-v850/as85ep1.h b/include/asm-v850/as85ep1.h
deleted file mode 100644
index 5a5ca9073d09..000000000000
--- a/include/asm-v850/as85ep1.h
+++ /dev/null
@@ -1,152 +0,0 @@
-/*
- * include/asm-v850/as85ep1.h -- AS85EP1 evaluation CPU chip/board
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_AS85EP1_H__
-#define __V850_AS85EP1_H__
-
-#include <asm/v850e.h>
-
-
-#define CPU_MODEL	"as85ep1"
-#define CPU_MODEL_LONG	"NEC V850E/AS85EP1"
-#define PLATFORM	"AS85EP1"
-#define PLATFORM_LONG	"NEC V850E/AS85EP1 evaluation board"
-
-#define CPU_CLOCK_FREQ	96000000 /*  96MHz */
-#define SYS_CLOCK_FREQ	CPU_CLOCK_FREQ
-
-
-/* 1MB of static RAM.  */
-#define SRAM_ADDR	0x00400000
-#define SRAM_SIZE	0x00100000 /* 1MB */
-/* About 58MB of DRAM.  This can actually be at one of two positions,
-   determined by jump JP3; we have to use the first position because the
-   second is partially out of processor instruction addressing range
-   (though in the second position there's actually 64MB available).  */
-#define SDRAM_ADDR	0x00600000
-#define SDRAM_SIZE	0x039F8000 /* approx 58MB */
-
-/* For <asm/page.h> */
-#define PAGE_OFFSET 	SRAM_ADDR
-
-/* We use on-chip RAM, for a few miscellaneous variables that must be
-   accessible using a load instruction relative to R0.  The AS85EP1 chip
-   16K of internal RAM located slightly before I/O space.  */
-#define R0_RAM_ADDR	0xFFFF8000
-
-
-/* AS85EP1 specific control registers.  */
-#define AS85EP1_CSC_ADDR(n)	(0xFFFFF060 + (n) * 2)
-#define AS85EP1_CSC(n)		(*(volatile u16 *)AS85EP1_CSC_ADDR(n))
-#define AS85EP1_BSC_ADDR	0xFFFFF066
-#define AS85EP1_BSC		(*(volatile u16 *)AS85EP1_BSC_ADDR)
-#define AS85EP1_BCT_ADDR(n)	(0xFFFFF480 + (n) * 2)
-#define AS85EP1_BCT(n)		(*(volatile u16 *)AS85EP1_BCT_ADDR(n))
-#define AS85EP1_DWC_ADDR(n)	(0xFFFFF484 + (n) * 2)
-#define AS85EP1_DWC(n)		(*(volatile u16 *)AS85EP1_DWC_ADDR(n))
-#define AS85EP1_BCC_ADDR	0xFFFFF488
-#define AS85EP1_BCC		(*(volatile u16 *)AS85EP1_BCC_ADDR)
-#define AS85EP1_ASC_ADDR	0xFFFFF48A
-#define AS85EP1_ASC		(*(volatile u16 *)AS85EP1_ASC_ADDR)
-#define AS85EP1_BCP_ADDR	0xFFFFF48C
-#define AS85EP1_BCP		(*(volatile u16 *)AS85EP1_BCP_ADDR)
-#define AS85EP1_LBS_ADDR	0xFFFFF48E
-#define AS85EP1_LBS		(*(volatile u16 *)AS85EP1_LBS_ADDR)
-#define AS85EP1_BMC_ADDR	0xFFFFF498
-#define AS85EP1_BMC		(*(volatile u16 *)AS85EP1_BMC_ADDR)
-#define AS85EP1_PRC_ADDR	0xFFFFF49A
-#define AS85EP1_PRC		(*(volatile u16 *)AS85EP1_PRC_ADDR)
-#define AS85EP1_SCR_ADDR(n)	(0xFFFFF4A0 + (n) * 4)
-#define AS85EP1_SCR(n)		(*(volatile u16 *)AS85EP1_SCR_ADDR(n))
-#define AS85EP1_RFS_ADDR(n)	(0xFFFFF4A2 + (n) * 4)
-#define AS85EP1_RFS(n)		(*(volatile u16 *)AS85EP1_RFS_ADDR(n))
-#define AS85EP1_IRAMM_ADDR	0xFFFFF80A
-#define AS85EP1_IRAMM		(*(volatile u8 *)AS85EP1_IRAMM_ADDR)
-
-
-
-/* I/O port P0-P13. */
-/* Direct I/O.  Bits 0-7 are pins Pn0-Pn7.  */
-#define AS85EP1_PORT_IO_ADDR(n)	(0xFFFFF400 + (n) * 2)
-#define AS85EP1_PORT_IO(n)	(*(volatile u8 *)AS85EP1_PORT_IO_ADDR(n))
-/* Port mode (for direct I/O, 0 = output, 1 = input).  */
-#define AS85EP1_PORT_PM_ADDR(n)	(0xFFFFF420 + (n) * 2)
-#define AS85EP1_PORT_PM(n)	(*(volatile u8 *)AS85EP1_PORT_PM_ADDR(n))
-/* Port mode control (0 = direct I/O mode, 1 = alternative I/O mode).  */
-#define AS85EP1_PORT_PMC_ADDR(n) (0xFFFFF440 + (n) * 2)
-#define AS85EP1_PORT_PMC(n)	(*(volatile u8 *)AS85EP1_PORT_PMC_ADDR(n))
-
-
-/* Hardware-specific interrupt numbers (in the kernel IRQ namespace).  */
-#define IRQ_INTCCC(n)	(0x0C + (n))
-#define IRQ_INTCCC_NUM	8
-#define IRQ_INTCMD(n)	(0x14 + (n)) /* interval timer interrupts 0-5 */
-#define IRQ_INTCMD_NUM	6
-#define IRQ_INTSRE(n)	(0x1E + (n)*3) /* UART 0-1 reception error */
-#define IRQ_INTSRE_NUM	2
-#define IRQ_INTSR(n)	(0x1F + (n)*3) /* UART 0-1 reception completion */
-#define IRQ_INTSR_NUM	2
-#define IRQ_INTST(n)	(0x20 + (n)*3) /* UART 0-1 transmission completion */
-#define IRQ_INTST_NUM	2
-
-#define NUM_CPU_IRQS	64
-
-#ifndef __ASSEMBLY__
-/* Initialize chip interrupts.  */
-extern void as85ep1_init_irqs (void);
-#endif
-
-
-/* AS85EP1 UART details (basically the same as the V850E/MA1, but 2 channels).  */
-#define V850E_UART_NUM_CHANNELS		2
-#define V850E_UART_BASE_FREQ		(SYS_CLOCK_FREQ / 4)
-#define V850E_UART_CHIP_NAME 		"V850E/NA85E"
-
-/* This is a function that gets called before configuring the UART.  */
-#define V850E_UART_PRE_CONFIGURE	as85ep1_uart_pre_configure
-#ifndef __ASSEMBLY__
-extern void as85ep1_uart_pre_configure (unsigned chan,
-					unsigned cflags, unsigned baud);
-#endif
-
-/* This board supports RTS/CTS for the on-chip UART, but only for channel 1. */
-
-/* CTS for UART channel 1 is pin P54 (bit 4 of port 5).  */
-#define V850E_UART_CTS(chan)   ((chan) == 1 ? !(AS85EP1_PORT_IO(5) & 0x10) : 1)
-/* RTS for UART channel 1 is pin P53 (bit 3 of port 5).  */
-#define V850E_UART_SET_RTS(chan, val)					      \
-   do {									      \
-	   if (chan == 1) {						      \
-		   unsigned old = AS85EP1_PORT_IO(5); 			      \
-		   if (val)						      \
-			   AS85EP1_PORT_IO(5) = old & ~0x8;		      \
-		   else							      \
-			   AS85EP1_PORT_IO(5) = old | 0x8;		      \
-	   }								      \
-   } while (0)
-
-
-/* Timer C details.  */
-#define V850E_TIMER_C_BASE_ADDR		0xFFFFF600
-
-/* Timer D details (the AS85EP1 actually has 5 of these; should change later). */
-#define V850E_TIMER_D_BASE_ADDR		0xFFFFF540
-#define V850E_TIMER_D_TMD_BASE_ADDR 	(V850E_TIMER_D_BASE_ADDR + 0x0)
-#define V850E_TIMER_D_CMD_BASE_ADDR 	(V850E_TIMER_D_BASE_ADDR + 0x2)
-#define V850E_TIMER_D_TMCD_BASE_ADDR 	(V850E_TIMER_D_BASE_ADDR + 0x4)
-
-#define V850E_TIMER_D_BASE_FREQ		SYS_CLOCK_FREQ
-#define V850E_TIMER_D_TMCD_CS_MIN	2 /* min 2^2 divider */
-
-
-#endif /* __V850_AS85EP1_H__ */
diff --git a/include/asm-v850/asm.h b/include/asm-v850/asm.h
deleted file mode 100644
index bf1e785a5dde..000000000000
--- a/include/asm-v850/asm.h
+++ /dev/null
@@ -1,32 +0,0 @@
-/*
- * include/asm-v850/asm.h -- Macros for writing assembly code
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#define G_ENTRY(name)							      \
-   .balign 4;								      \
-   .globl name;								      \
-   .type  name,@function;						      \
-   name
-#define G_DATA(name)							      \
-   .globl name;								      \
-   .type  name,@object;							      \
-   name
-#define END(name)							      \
-   .size  name,.-name
-
-#define L_ENTRY(name)							      \
-   .balign 4;								      \
-   .type  name,@function;						      \
-   name
-#define L_DATA(name)							      \
-   .type  name,@object;							      \
-   name
diff --git a/include/asm-v850/atomic.h b/include/asm-v850/atomic.h
deleted file mode 100644
index e4e57de08f73..000000000000
--- a/include/asm-v850/atomic.h
+++ /dev/null
@@ -1,131 +0,0 @@
-/*
- * include/asm-v850/atomic.h -- Atomic operations
- *
- *  Copyright (C) 2001,02  NEC Corporation
- *  Copyright (C) 2001,02  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_ATOMIC_H__
-#define __V850_ATOMIC_H__
-
-
-#include <asm/system.h>
-
-#ifdef CONFIG_SMP
-#error SMP not supported
-#endif
-
-typedef struct { int counter; } atomic_t;
-
-#define ATOMIC_INIT(i)	{ (i) }
-
-#ifdef __KERNEL__
-
-#define atomic_read(v)		((v)->counter)
-#define atomic_set(v,i)		(((v)->counter) = (i))
-
-static inline int atomic_add_return (int i, volatile atomic_t *v)
-{
-	unsigned long flags;
-	int res;
-
-	local_irq_save (flags);
-	res = v->counter + i;
-	v->counter = res;
-	local_irq_restore (flags);
-
-	return res;
-}
-
-static __inline__ int atomic_sub_return (int i, volatile atomic_t *v)
-{
-	unsigned long flags;
-	int res;
-
-	local_irq_save (flags);
-	res = v->counter - i;
-	v->counter = res;
-	local_irq_restore (flags);
-
-	return res;
-}
-
-static __inline__ void atomic_clear_mask (unsigned long mask, unsigned long *addr)
-{
-	unsigned long flags;
-
-	local_irq_save (flags);
-	*addr &= ~mask;
-	local_irq_restore (flags);
-}
-
-#endif
-
-#define atomic_add(i, v)	atomic_add_return ((i), (v))
-#define atomic_sub(i, v)	atomic_sub_return ((i), (v))
-
-#define atomic_dec_return(v)	atomic_sub_return (1, (v))
-#define atomic_inc_return(v)	atomic_add_return (1, (v))
-#define atomic_inc(v) 		atomic_inc_return (v)
-#define atomic_dec(v) 		atomic_dec_return (v)
-
-/*
- * atomic_inc_and_test - increment and test
- * @v: pointer of type atomic_t
- *
- * Atomically increments @v by 1
- * and returns true if the result is zero, or false for all
- * other cases.
- */
-#define atomic_inc_and_test(v) (atomic_inc_return(v) == 0)
-
-#define atomic_sub_and_test(i,v)	(atomic_sub_return ((i), (v)) == 0)
-#define atomic_dec_and_test(v)		(atomic_sub_return (1, (v)) == 0)
-#define atomic_add_negative(i,v)	(atomic_add_return ((i), (v)) < 0)
-
-static inline int atomic_cmpxchg(atomic_t *v, int old, int new)
-{
-	int ret;
-	unsigned long flags;
-
-	local_irq_save(flags);
-	ret = v->counter;
-	if (likely(ret == old))
-		v->counter = new;
-	local_irq_restore(flags);
-
-	return ret;
-}
-
-#define atomic_xchg(v, new) (xchg(&((v)->counter), new))
-
-static inline int atomic_add_unless(atomic_t *v, int a, int u)
-{
-	int ret;
-	unsigned long flags;
-
-	local_irq_save(flags);
-	ret = v->counter;
-	if (ret != u)
-		v->counter += a;
-	local_irq_restore(flags);
-
-	return ret != u;
-}
-
-#define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
-
-/* Atomic operations are already serializing on ARM */
-#define smp_mb__before_atomic_dec()	barrier()
-#define smp_mb__after_atomic_dec()	barrier()
-#define smp_mb__before_atomic_inc()	barrier()
-#define smp_mb__after_atomic_inc()	barrier()
-
-#include <asm-generic/atomic.h>
-#endif /* __V850_ATOMIC_H__ */
diff --git a/include/asm-v850/auxvec.h b/include/asm-v850/auxvec.h
deleted file mode 100644
index f493232d0224..000000000000
--- a/include/asm-v850/auxvec.h
+++ /dev/null
@@ -1,4 +0,0 @@
-#ifndef __V850_AUXVEC_H__
-#define __V850_AUXVEC_H__
-
-#endif /* __V850_AUXVEC_H__ */
diff --git a/include/asm-v850/bitops.h b/include/asm-v850/bitops.h
deleted file mode 100644
index f82f5b4a56e0..000000000000
--- a/include/asm-v850/bitops.h
+++ /dev/null
@@ -1,161 +0,0 @@
-/*
- * include/asm-v850/bitops.h -- Bit operations
- *
- *  Copyright (C) 2001,02,03,04,05  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03,04,05  Miles Bader <miles@gnu.org>
- *  Copyright (C) 1992  Linus Torvalds.
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- */
-
-#ifndef __V850_BITOPS_H__
-#define __V850_BITOPS_H__
-
-#ifndef _LINUX_BITOPS_H
-#error only <linux/bitops.h> can be included directly
-#endif
-
-#include <linux/compiler.h>	/* unlikely  */
-#include <asm/byteorder.h>	/* swab32 */
-#include <asm/system.h>		/* interrupt enable/disable */
-
-
-#ifdef __KERNEL__
-
-#include <asm-generic/bitops/ffz.h>
-
-/*
- * The __ functions are not atomic
- */
-
-/* In the following constant-bit-op macros, a "g" constraint is used when
-   we really need an integer ("i" constraint).  This is to avoid
-   warnings/errors from the compiler in the case where the associated
-   operand _isn't_ an integer, and shouldn't produce bogus assembly because
-   use of that form is protected by a guard statement that checks for
-   constants, and should otherwise be removed by the optimizer.  This
-   _usually_ works -- however, __builtin_constant_p returns true for a
-   variable with a known constant value too, and unfortunately gcc will
-   happily put the variable in a register and use the register for the "g"
-   constraint'd asm operand.  To avoid the latter problem, we add a
-   constant offset to the operand and subtract it back in the asm code;
-   forcing gcc to do arithmetic on the value is usually enough to get it
-   to use a real constant value.  This is horrible, and ultimately
-   unreliable too, but it seems to work for now (hopefully gcc will offer
-   us more control in the future, so we can do a better job).  */
-
-#define __const_bit_op(op, nr, addr)					\
-  ({ __asm__ (op " (%0 - 0x123), %1"					\
-	      :: "g" (((nr) & 0x7) + 0x123),				\
-		 "m" (*((char *)(addr) + ((nr) >> 3)))			\
-	      : "memory"); })
-#define __var_bit_op(op, nr, addr)					\
-  ({ int __nr = (nr);							\
-     __asm__ (op " %0, [%1]"						\
-	      :: "r" (__nr & 0x7),					\
-		 "r" ((char *)(addr) + (__nr >> 3))			\
-	      : "memory"); })
-#define __bit_op(op, nr, addr)						\
-  ((__builtin_constant_p (nr) && (unsigned)(nr) <= 0x7FFFF)		\
-   ? __const_bit_op (op, nr, addr)					\
-   : __var_bit_op (op, nr, addr))
-
-#define __set_bit(nr, addr)		__bit_op ("set1", nr, addr)
-#define __clear_bit(nr, addr)		__bit_op ("clr1", nr, addr)
-#define __change_bit(nr, addr)		__bit_op ("not1", nr, addr)
-
-/* The bit instructions used by `non-atomic' variants are actually atomic.  */
-#define set_bit __set_bit
-#define clear_bit __clear_bit
-#define change_bit __change_bit
-
-
-#define __const_tns_bit_op(op, nr, addr)				      \
-  ({ int __tns_res;							      \
-     __asm__ __volatile__ (						      \
-	     "tst1 (%1 - 0x123), %2; setf nz, %0; " op " (%1 - 0x123), %2"    \
-	     : "=&r" (__tns_res)					      \
-	     : "g" (((nr) & 0x7) + 0x123),				      \
-	       "m" (*((char *)(addr) + ((nr) >> 3)))			      \
-	     : "memory");						      \
-     __tns_res;								      \
-  })
-#define __var_tns_bit_op(op, nr, addr)					      \
-  ({ int __nr = (nr);							      \
-     int __tns_res;							      \
-     __asm__ __volatile__ (						      \
-	     "tst1 %1, [%2]; setf nz, %0; " op " %1, [%2]"		      \
-	      : "=&r" (__tns_res)					      \
-	      : "r" (__nr & 0x7),					      \
-		"r" ((char *)(addr) + (__nr >> 3))			      \
-	      : "memory");						      \
-     __tns_res;								      \
-  })
-#define __tns_bit_op(op, nr, addr)					\
-  ((__builtin_constant_p (nr) && (unsigned)(nr) <= 0x7FFFF)		\
-   ? __const_tns_bit_op (op, nr, addr)					\
-   : __var_tns_bit_op (op, nr, addr))
-#define __tns_atomic_bit_op(op, nr, addr)				\
-  ({ int __tns_atomic_res, __tns_atomic_flags;				\
-     local_irq_save (__tns_atomic_flags);				\
-     __tns_atomic_res = __tns_bit_op (op, nr, addr);			\
-     local_irq_restore (__tns_atomic_flags);				\
-     __tns_atomic_res;							\
-  })
-
-#define __test_and_set_bit(nr, addr)	__tns_bit_op ("set1", nr, addr)
-#define test_and_set_bit(nr, addr)	__tns_atomic_bit_op ("set1", nr, addr)
-
-#define __test_and_clear_bit(nr, addr)	__tns_bit_op ("clr1", nr, addr)
-#define test_and_clear_bit(nr, addr)	__tns_atomic_bit_op ("clr1", nr, addr)
-
-#define __test_and_change_bit(nr, addr)	__tns_bit_op ("not1", nr, addr)
-#define test_and_change_bit(nr, addr)	__tns_atomic_bit_op ("not1", nr, addr)
-
-
-#define __const_test_bit(nr, addr)					      \
-  ({ int __test_bit_res;						      \
-     __asm__ __volatile__ ("tst1 (%1 - 0x123), %2; setf nz, %0"		      \
-			   : "=r" (__test_bit_res)			      \
-			   : "g" (((nr) & 0x7) + 0x123),		      \
-			     "m" (*((const char *)(addr) + ((nr) >> 3))));    \
-     __test_bit_res;							      \
-  })
-static inline int __test_bit (int nr, const void *addr)
-{
-	int res;
-	__asm__ __volatile__ ("tst1 %1, [%2]; setf nz, %0"
-			      : "=r" (res)
-			      : "r" (nr & 0x7), "r" (addr + (nr >> 3)));
-	return res;
-}
-#define test_bit(nr,addr)						\
-  ((__builtin_constant_p (nr) && (unsigned)(nr) <= 0x7FFFF)		\
-   ? __const_test_bit ((nr), (addr))					\
-   : __test_bit ((nr), (addr)))
-
-
-/* clear_bit doesn't provide any barrier for the compiler.  */
-#define smp_mb__before_clear_bit()	barrier ()
-#define smp_mb__after_clear_bit()	barrier ()
-
-#include <asm-generic/bitops/ffs.h>
-#include <asm-generic/bitops/fls.h>
-#include <asm-generic/bitops/fls64.h>
-#include <asm-generic/bitops/__ffs.h>
-#include <asm-generic/bitops/find.h>
-#include <asm-generic/bitops/sched.h>
-#include <asm-generic/bitops/hweight.h>
-#include <asm-generic/bitops/lock.h>
-
-#include <asm-generic/bitops/ext2-non-atomic.h>
-#define ext2_set_bit_atomic(l,n,a)      test_and_set_bit(n,a)
-#define ext2_clear_bit_atomic(l,n,a)    test_and_clear_bit(n,a)
-
-#include <asm-generic/bitops/minix.h>
-
-#endif /* __KERNEL__ */
-
-#endif /* __V850_BITOPS_H__ */
diff --git a/include/asm-v850/bug.h b/include/asm-v850/bug.h
deleted file mode 100644
index b0ed2d35f3e8..000000000000
--- a/include/asm-v850/bug.h
+++ /dev/null
@@ -1,25 +0,0 @@
-/*
- * include/asm-v850/bug.h -- Bug reporting
- *
- *  Copyright (C) 2003  NEC Electronics Corporation
- *  Copyright (C) 2003  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_BUG_H__
-#define __V850_BUG_H__
-
-#ifdef CONFIG_BUG
-extern void __bug (void) __attribute__ ((noreturn));
-#define BUG()		__bug()
-#define HAVE_ARCH_BUG
-#endif
-
-#include <asm-generic/bug.h>
-
-#endif /* __V850_BUG_H__ */
diff --git a/include/asm-v850/bugs.h b/include/asm-v850/bugs.h
deleted file mode 100644
index 71110a65c1d7..000000000000
--- a/include/asm-v850/bugs.h
+++ /dev/null
@@ -1,16 +0,0 @@
-/*
- *  include/asm-v850e/bugs.h
- *
- *  Copyright (C) 1994  Linus Torvalds
- */
-
-/*
- * This is included by init/main.c to check for architecture-dependent bugs.
- *
- * Needs:
- *	void check_bugs(void);
- */
-
-static void check_bugs(void)
-{
-}
diff --git a/include/asm-v850/byteorder.h b/include/asm-v850/byteorder.h
deleted file mode 100644
index a6f07530050e..000000000000
--- a/include/asm-v850/byteorder.h
+++ /dev/null
@@ -1,48 +0,0 @@
-/*
- * include/asm-v850/byteorder.h -- Endian id and conversion ops
- *
- *  Copyright (C) 2001  NEC Corporation
- *  Copyright (C) 2001  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_BYTEORDER_H__
-#define __V850_BYTEORDER_H__
-
-#include <asm/types.h>
-#include <linux/compiler.h>
-
-#ifdef __GNUC__
-
-static __inline__ __attribute_const__ __u32 ___arch__swab32 (__u32 word)
-{
-	__u32 res;
-	__asm__ ("bsw %1, %0" : "=r" (res) : "r" (word));
-	return res;
-}
-
-static __inline__ __attribute_const__ __u16 ___arch__swab16 (__u16 half_word)
-{
-	__u16 res;
-	__asm__ ("bsh %1, %0" : "=r" (res) : "r" (half_word));
-	return res;
-}
-
-#define __arch__swab32(x) ___arch__swab32(x)
-#define __arch__swab16(x) ___arch__swab16(x)
-
-#if !defined(__STRICT_ANSI__) || defined(__KERNEL__)
-#  define __BYTEORDER_HAS_U64__
-#  define __SWAB_64_THRU_32__
-#endif
-
-#endif /* __GNUC__ */
-
-#include <linux/byteorder/little_endian.h>
-
-#endif /* __V850_BYTEORDER_H__ */
diff --git a/include/asm-v850/cache.h b/include/asm-v850/cache.h
deleted file mode 100644
index 8832c7ea3242..000000000000
--- a/include/asm-v850/cache.h
+++ /dev/null
@@ -1,26 +0,0 @@
-/*
- * include/asm-v850/cache.h -- Cache operations
- *
- *  Copyright (C) 2001,05  NEC Corporation
- *  Copyright (C) 2001,05  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_CACHE_H__
-#define __V850_CACHE_H__
-
-/* All cache operations are machine-dependent.  */
-#include <asm/machdep.h>
-
-#ifndef L1_CACHE_BYTES
-/* This processor has no cache, so just choose an arbitrary value.  */
-#define L1_CACHE_BYTES		16
-#define L1_CACHE_SHIFT		4
-#endif
-
-#endif /* __V850_CACHE_H__ */
diff --git a/include/asm-v850/cacheflush.h b/include/asm-v850/cacheflush.h
deleted file mode 100644
index 9ece05a202ef..000000000000
--- a/include/asm-v850/cacheflush.h
+++ /dev/null
@@ -1,70 +0,0 @@
-/*
- * include/asm-v850/cacheflush.h
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_CACHEFLUSH_H__
-#define __V850_CACHEFLUSH_H__
-
-/* Somebody depends on this; sigh...  */
-#include <linux/mm.h>
-
-#include <asm/machdep.h>
-
-
-/* The following are all used by the kernel in ways that only affect
-   systems with MMUs, so we don't need them.  */
-#define flush_cache_all()			((void)0)
-#define flush_cache_mm(mm)			((void)0)
-#define flush_cache_dup_mm(mm)			((void)0)
-#define flush_cache_range(vma, start, end)	((void)0)
-#define flush_cache_page(vma, vmaddr, pfn)	((void)0)
-#define flush_dcache_page(page)			((void)0)
-#define flush_dcache_mmap_lock(mapping)		((void)0)
-#define flush_dcache_mmap_unlock(mapping)	((void)0)
-#define flush_cache_vmap(start, end)		((void)0)
-#define flush_cache_vunmap(start, end)		((void)0)
-
-#ifdef CONFIG_NO_CACHE
-
-/* Some systems have no cache at all, in which case we don't need these
-   either.  */
-#define flush_icache()				((void)0)
-#define flush_icache_range(start, end)		((void)0)
-#define flush_icache_page(vma,pg)		((void)0)
-#define flush_icache_user_range(vma,pg,adr,len)	((void)0)
-#define flush_cache_sigtramp(vaddr)		((void)0)
-
-#else /* !CONFIG_NO_CACHE */
-
-struct page;
-struct mm_struct;
-struct vm_area_struct;
-
-/* Otherwise, somebody had better define them.  */
-extern void flush_icache (void);
-extern void flush_icache_range (unsigned long start, unsigned long end);
-extern void flush_icache_page (struct vm_area_struct *vma, struct page *page);
-extern void flush_icache_user_range (struct vm_area_struct *vma,
-				     struct page *page,
-				     unsigned long adr, int len);
-extern void flush_cache_sigtramp (unsigned long addr);
-
-#endif /* CONFIG_NO_CACHE */
-
-#define copy_to_user_page(vma, page, vaddr, dst, src, len) \
-do { memcpy(dst, src, len); \
-     flush_icache_user_range(vma, page, vaddr, len); \
-} while (0)
-#define copy_from_user_page(vma, page, vaddr, dst, src, len) \
-	memcpy(dst, src, len)
-
-#endif /* __V850_CACHEFLUSH_H__ */
diff --git a/include/asm-v850/checksum.h b/include/asm-v850/checksum.h
deleted file mode 100644
index d1dddd938262..000000000000
--- a/include/asm-v850/checksum.h
+++ /dev/null
@@ -1,112 +0,0 @@
-/*
- * include/asm-v850/checksum.h -- Checksum ops
- *
- *  Copyright (C) 2001,2005  NEC Corporation
- *  Copyright (C) 2001,2005  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_CHECKSUM_H__
-#define __V850_CHECKSUM_H__
-
-/*
- * computes the checksum of a memory block at buff, length len,
- * and adds in "sum" (32-bit)
- *
- * returns a 32-bit number suitable for feeding into itself
- * or csum_tcpudp_magic
- *
- * this function must be called with even lengths, except
- * for the last fragment, which may be odd
- *
- * it's best to have buff aligned on a 32-bit boundary
- */
-extern __wsum csum_partial(const void *buff, int len, __wsum sum);
-
-/*
- * the same as csum_partial, but copies from src while it
- * checksums
- *
- * here even more important to align src and dst on a 32-bit (or even
- * better 64-bit) boundary
- */
-extern __wsum csum_partial_copy_nocheck(const void *src,
-				   void *dst, int len, __wsum sum);
-
-
-/*
- * the same as csum_partial_copy, but copies from user space.
- *
- * here even more important to align src and dst on a 32-bit (or even
- * better 64-bit) boundary
- */
-extern __wsum csum_partial_copy_from_user (const void *src,
-					     void *dst,
-					     int len, __wsum sum,
-					     int *csum_err);
-
-__sum16 ip_fast_csum(const void *iph, unsigned int ihl);
-
-/*
- *	Fold a partial checksum
- */
-static inline __sum16 csum_fold (__wsum sum)
-{
-	unsigned int result;
-	/*
-			        %0		%1
-	      hsw %1, %0	H     L		L     H
-	      add %1, %0	H     L		H+L+C H+L
-	*/
-	asm ("hsw %1, %0; add %1, %0" : "=&r" (result) : "r" (sum));
-	return (__force __sum16)(~result >> 16);
-}
-
-
-/*
- * computes the checksum of the TCP/UDP pseudo-header
- * returns a 16-bit checksum, already complemented
- */
-static inline __wsum
-csum_tcpudp_nofold (__be32 saddr, __be32 daddr,
-		    unsigned short len,
-		    unsigned short proto, __wsum sum)
-{
-	int __carry;
-	__asm__ ("add %2, %0;"
-		 "setf c, %1;"
-		 "add %1, %0;"
-		 "add %3, %0;"
-		 "setf c, %1;"
-		 "add %1, %0;"
-		 "add %4, %0;"
-		 "setf c, %1;"
-		 "add %1, %0"
-		 : "=&r" (sum), "=&r" (__carry)
-		 : "r" (daddr), "r" (saddr),
-		 "r" ((len + proto) << 8),
-		 "0" (sum));
-	return sum;
-}
-
-static inline __sum16
-csum_tcpudp_magic (__be32 saddr, __be32 daddr,
-		   unsigned short len,
-		   unsigned short proto, __wsum sum)
-{
-	return csum_fold (csum_tcpudp_nofold (saddr, daddr, len, proto, sum));
-}
-
-/*
- * this routine is used for miscellaneous IP-like checksums, mainly
- * in icmp.c
- */
-extern __sum16 ip_compute_csum(const void *buff, int len);
-
-
-#endif /* __V850_CHECKSUM_H__ */
diff --git a/include/asm-v850/clinkage.h b/include/asm-v850/clinkage.h
deleted file mode 100644
index c389691d6f86..000000000000
--- a/include/asm-v850/clinkage.h
+++ /dev/null
@@ -1,26 +0,0 @@
-/*
- * include/asm-v850/clinkage.h -- Macros to reflect C symbol-naming conventions
- *
- *  Copyright (C) 2001,02  NEC Corporatione
- *  Copyright (C) 2001,02  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_CLINKAGE_H__
-#define __V850_CLINKAGE_H__
-
-#include <asm/macrology.h>
-#include <asm/asm.h>
-
-#define C_SYMBOL_NAME(name) 	macrology_paste(_, name)
-#define C_SYMBOL_STRING(name)	macrology_stringify(C_SYMBOL_NAME(name))
-#define C_ENTRY(name)		G_ENTRY(C_SYMBOL_NAME(name))
-#define C_DATA(name)		G_DATA(C_SYMBOL_NAME(name))
-#define C_END(name)		END(C_SYMBOL_NAME(name))
-
-#endif /* __V850_CLINKAGE_H__ */
diff --git a/include/asm-v850/cputime.h b/include/asm-v850/cputime.h
deleted file mode 100644
index 7c799c33b8a9..000000000000
--- a/include/asm-v850/cputime.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __V850_CPUTIME_H
-#define __V850_CPUTIME_H
-
-#include <asm-generic/cputime.h>
-
-#endif /* __V850_CPUTIME_H */
diff --git a/include/asm-v850/current.h b/include/asm-v850/current.h
deleted file mode 100644
index 30aae5673770..000000000000
--- a/include/asm-v850/current.h
+++ /dev/null
@@ -1,47 +0,0 @@
-/*
- * include/asm-v850/current.h -- Current task
- *
- *  Copyright (C) 2001,02  NEC Corporation
- *  Copyright (C) 2001,02  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_CURRENT_H__
-#define __V850_CURRENT_H__
-
-#ifndef __ASSEMBLY__ /* <linux/thread_info.h> is not asm-safe.  */
-#include <linux/thread_info.h>
-#endif
-
-#include <asm/macrology.h>
-
-
-/* Register used to hold the current task pointer while in the kernel.
-   Any `call clobbered' register without a special meaning should be OK,
-   but check asm/v850/kernel/entry.S to be sure.  */
-#define CURRENT_TASK_REGNUM	16
-#define CURRENT_TASK 		macrology_paste (r, CURRENT_TASK_REGNUM)
-
-
-#ifdef __ASSEMBLY__
-
-/* Put a pointer to the current task structure into REG.  */
-#define GET_CURRENT_TASK(reg)						\
-	GET_CURRENT_THREAD(reg);					\
-	ld.w	TI_TASK[reg], reg
-
-#else /* !__ASSEMBLY__ */
-
-/* A pointer to the current task.  */
-register struct task_struct *current					\
-   __asm__ (macrology_stringify (CURRENT_TASK));
-
-#endif /* __ASSEMBLY__ */
-
-
-#endif /* _V850_CURRENT_H */
diff --git a/include/asm-v850/delay.h b/include/asm-v850/delay.h
deleted file mode 100644
index 6d028e6b2354..000000000000
--- a/include/asm-v850/delay.h
+++ /dev/null
@@ -1,47 +0,0 @@
-/*
- * include/asm-v850/delay.h -- Delay routines, using a pre-computed
- * 	"loops_per_second" value
- *
- *  Copyright (C) 2001,03  NEC Corporation
- *  Copyright (C) 2001,03  Miles Bader <miles@gnu.org>
- *  Copyright (C) 1994 Hamish Macdonald
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- */
-
-#ifndef __V850_DELAY_H__
-#define __V850_DELAY_H__
-
-#include <asm/param.h>
-
-static inline void __delay(unsigned long loops)
-{
-	if (loops)
-		__asm__ __volatile__ ("1: add -1, %0; bnz 1b"
-				      : "=r" (loops) : "0" (loops));
-}
-
-/*
- * Use only for very small delays ( < 1 msec).  Should probably use a
- * lookup table, really, as the multiplications take much too long with
- * short delays.  This is a "reasonable" implementation, though (and the
- * first constant multiplications gets optimized away if the delay is
- * a constant)  
- */
-
-extern unsigned long loops_per_jiffy;
-
-static inline void udelay(unsigned long usecs)
-{
-	register unsigned long full_loops, part_loops;
-
-	full_loops = ((usecs * HZ) / 1000000) * loops_per_jiffy;
-	usecs %= (1000000 / HZ);
-	part_loops = (usecs * HZ * loops_per_jiffy) / 1000000;
-
-	__delay(full_loops + part_loops);
-}
-
-#endif /* __V850_DELAY_H__ */
diff --git a/include/asm-v850/device.h b/include/asm-v850/device.h
deleted file mode 100644
index d8f9872b0e2d..000000000000
--- a/include/asm-v850/device.h
+++ /dev/null
@@ -1,7 +0,0 @@
-/*
- * Arch specific extensions to struct device
- *
- * This file is released under the GPLv2
- */
-#include <asm-generic/device.h>
-
diff --git a/include/asm-v850/div64.h b/include/asm-v850/div64.h
deleted file mode 100644
index 6cd978cefb28..000000000000
--- a/include/asm-v850/div64.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/div64.h>
diff --git a/include/asm-v850/dma-mapping.h b/include/asm-v850/dma-mapping.h
deleted file mode 100644
index 1cc42c603a1b..000000000000
--- a/include/asm-v850/dma-mapping.h
+++ /dev/null
@@ -1,11 +0,0 @@
-#ifndef __V850_DMA_MAPPING_H__
-#define __V850_DMA_MAPPING_H__
-
-
-#ifdef CONFIG_PCI
-#include <asm-generic/dma-mapping.h>
-#else
-#include <asm-generic/dma-mapping-broken.h>
-#endif
-
-#endif /* __V850_DMA_MAPPING_H__ */
diff --git a/include/asm-v850/dma.h b/include/asm-v850/dma.h
deleted file mode 100644
index 2369849e2d0a..000000000000
--- a/include/asm-v850/dma.h
+++ /dev/null
@@ -1,18 +0,0 @@
-#ifndef __V850_DMA_H__
-#define __V850_DMA_H__
-
-/* What should this be?  */
-#define MAX_DMA_ADDRESS	0xFFFFFFFF
-
-/* reserve a DMA channel */
-extern int request_dma (unsigned int dmanr, const char * device_id);
-/* release it again */
-extern void free_dma (unsigned int dmanr);
-
-#ifdef CONFIG_PCI
-extern int isa_dma_bridge_buggy;
-#else
-#define isa_dma_bridge_buggy    (0)
-#endif
-
-#endif /* __V850_DMA_H__ */
diff --git a/include/asm-v850/elf.h b/include/asm-v850/elf.h
deleted file mode 100644
index 28f5b176ff1a..000000000000
--- a/include/asm-v850/elf.h
+++ /dev/null
@@ -1,99 +0,0 @@
-#ifndef __V850_ELF_H__
-#define __V850_ELF_H__
-
-/*
- * ELF register definitions..
- */
-
-#include <asm/ptrace.h>
-#include <asm/user.h>
-#include <asm/byteorder.h>
-
-typedef unsigned long elf_greg_t;
-
-#define ELF_NGREG (sizeof (struct pt_regs) / sizeof(elf_greg_t))
-typedef elf_greg_t elf_gregset_t[ELF_NGREG];
-
-typedef struct user_fpu_struct elf_fpregset_t;
-
-/*
- * This is used to ensure we don't load something for the wrong architecture.
- */
-#define elf_check_arch(x)  \
-  ((x)->e_machine == EM_V850 || (x)->e_machine == EM_CYGNUS_V850)
-
-
-/* v850 relocation types.  */
-#define R_V850_NONE		0
-#define R_V850_9_PCREL		1
-#define R_V850_22_PCREL		2
-#define R_V850_HI16_S		3
-#define R_V850_HI16		4
-#define R_V850_LO16		5
-#define R_V850_32		6
-#define R_V850_16		7
-#define R_V850_8		8
-#define R_V850_SDA_16_16_OFFSET	9	/* For ld.b, st.b, set1, clr1,
-					   not1, tst1, movea, movhi */
-#define R_V850_SDA_15_16_OFFSET	10	/* For ld.w, ld.h, ld.hu, st.w, st.h */
-#define R_V850_ZDA_16_16_OFFSET	11	/* For ld.b, st.b, set1, clr1,
-					   not1, tst1, movea, movhi */
-#define R_V850_ZDA_15_16_OFFSET	12	/* For ld.w, ld.h, ld.hu, st.w, st.h */
-#define R_V850_TDA_6_8_OFFSET	13	/* For sst.w, sld.w */
-#define R_V850_TDA_7_8_OFFSET	14	/* For sst.h, sld.h */
-#define R_V850_TDA_7_7_OFFSET	15	/* For sst.b, sld.b */
-#define R_V850_TDA_16_16_OFFSET	16	/* For set1, clr1, not1, tst1,
-					   movea, movhi */
-#define R_V850_NUM		17
-
-
-/*
- * These are used to set parameters in the core dumps.
- */
-#define ELF_CLASS	ELFCLASS32
-#ifdef __LITTLE_ENDIAN__
-#define ELF_DATA	ELFDATA2LSB
-#else
-#define ELF_DATA	ELFDATA2MSB
-#endif
-#define ELF_ARCH	EM_V850
-
-#define USE_ELF_CORE_DUMP
-#define ELF_EXEC_PAGESIZE	4096
-
-
-#define ELF_CORE_COPY_REGS(_dest,_regs)				\
-	memcpy((char *) &_dest, (char *) _regs,			\
-	       sizeof(struct pt_regs));
-
-/* This yields a mask that user programs can use to figure out what
-   instruction set this CPU supports.  This could be done in user space,
-   but it's not easy, and we've already done it here.  */
-
-#define ELF_HWCAP	(0)
-
-/* This yields a string that ld.so will use to load implementation
-   specific libraries for optimization.  This is more specific in
-   intent than poking at uname or /proc/cpuinfo.
-
-   For the moment, we have only optimizations for the Intel generations,
-   but that could change... */
-
-#define ELF_PLATFORM  (NULL)
-
-#define ELF_PLAT_INIT(_r, load_addr)					      \
-  do {									      \
-	 _r->gpr[0] =  _r->gpr[1] =  _r->gpr[2] =  _r->gpr[3] =		      \
-	 _r->gpr[4] =  _r->gpr[5] =  _r->gpr[6] =  _r->gpr[7] =		      \
-	 _r->gpr[8] =  _r->gpr[9] = _r->gpr[10] = _r->gpr[11] =		      \
-	_r->gpr[12] = _r->gpr[13] = _r->gpr[14] = _r->gpr[15] =		      \
-	_r->gpr[16] = _r->gpr[17] = _r->gpr[18] = _r->gpr[19] =		      \
-	_r->gpr[20] = _r->gpr[21] = _r->gpr[22] = _r->gpr[23] =		      \
-	_r->gpr[24] = _r->gpr[25] = _r->gpr[26] = _r->gpr[27] =		      \
-	_r->gpr[28] = _r->gpr[29] = _r->gpr[30] = _r->gpr[31] =		      \
-	0;								      \
-  } while (0)
-
-#define SET_PERSONALITY(ex, ibcs2) set_personality(PER_LINUX_32BIT)
-
-#endif /* __V850_ELF_H__ */
diff --git a/include/asm-v850/emergency-restart.h b/include/asm-v850/emergency-restart.h
deleted file mode 100644
index 108d8c48e42e..000000000000
--- a/include/asm-v850/emergency-restart.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef _ASM_EMERGENCY_RESTART_H
-#define _ASM_EMERGENCY_RESTART_H
-
-#include <asm-generic/emergency-restart.h>
-
-#endif /* _ASM_EMERGENCY_RESTART_H */
diff --git a/include/asm-v850/entry.h b/include/asm-v850/entry.h
deleted file mode 100644
index d9df8ac48584..000000000000
--- a/include/asm-v850/entry.h
+++ /dev/null
@@ -1,113 +0,0 @@
-/*
- * include/asm-v850/entry.h -- Definitions used by low-level trap handlers
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_ENTRY_H__
-#define __V850_ENTRY_H__
-
-
-#include <asm/ptrace.h>
-#include <asm/machdep.h>
-
-
-/* These are special variables using by the kernel trap/interrupt code
-   to save registers in, at a time when there are no spare registers we
-   can use to do so, and we can't depend on the value of the stack
-   pointer.  This means that they must be within a signed 16-bit
-   displacement of 0x00000000.  */
-
-#define KERNEL_VAR_SPACE_ADDR	R0_RAM_ADDR
-
-#ifdef __ASSEMBLY__
-#define KERNEL_VAR(addr)	addr[r0]
-#else
-#define KERNEL_VAR(addr)	(*(volatile unsigned long *)(addr))
-#endif
-
-/* Kernel stack pointer, 4 bytes.  */
-#define KSP_ADDR		(KERNEL_VAR_SPACE_ADDR +  0)
-#define KSP			KERNEL_VAR (KSP_ADDR)
-/* 1 if in kernel-mode, 0 if in user mode, 1 byte.  */
-#define KM_ADDR 		(KERNEL_VAR_SPACE_ADDR +  4)
-#define KM			KERNEL_VAR (KM_ADDR)
-/* Temporary storage for interrupt handlers, 4 bytes.  */
-#define INT_SCRATCH_ADDR	(KERNEL_VAR_SPACE_ADDR +  8)
-#define INT_SCRATCH		KERNEL_VAR (INT_SCRATCH_ADDR)
-/* Where the stack-pointer is saved when jumping to various sorts of
-   interrupt handlers.  ENTRY_SP is used by everything except NMIs,
-   which have their own location.  Higher-priority NMIs can clobber the
-   value written by a lower priority NMI, since they can't be disabled,
-   but that's OK, because only NMI0 (the lowest-priority one) is allowed
-   to return.  */
-#define ENTRY_SP_ADDR		(KERNEL_VAR_SPACE_ADDR + 12)
-#define ENTRY_SP		KERNEL_VAR (ENTRY_SP_ADDR)
-#define NMI_ENTRY_SP_ADDR	(KERNEL_VAR_SPACE_ADDR + 16)
-#define NMI_ENTRY_SP		KERNEL_VAR (NMI_ENTRY_SP_ADDR)
-
-#ifdef CONFIG_RESET_GUARD
-/* Used to detect unexpected resets (since the v850 has no MMU, any call
-   through a null pointer will jump to the reset vector).  We detect
-   such resets by checking for a magic value, RESET_GUARD_ACTIVE, in
-   this location.  Properly resetting the machine stores zero there, so
-   it shouldn't trigger the guard; the power-on value is uncertain, but
-   it's unlikely to be RESET_GUARD_ACTIVE.  */
-#define RESET_GUARD_ADDR	(KERNEL_VAR_SPACE_ADDR + 28)
-#define RESET_GUARD		KERNEL_VAR (RESET_GUARD_ADDR)
-#define RESET_GUARD_ACTIVE	0xFAB4BEEF
-#endif /* CONFIG_RESET_GUARD */
-
-#ifdef CONFIG_V850E_HIGHRES_TIMER
-#define HIGHRES_TIMER_SLOW_TICKS_ADDR (KERNEL_VAR_SPACE_ADDR + 32)
-#define HIGHRES_TIMER_SLOW_TICKS     KERNEL_VAR (HIGHRES_TIMER_SLOW_TICKS_ADDR)
-#endif /* CONFIG_V850E_HIGHRES_TIMER */
-
-#ifndef __ASSEMBLY__
-
-#ifdef CONFIG_RESET_GUARD
-/* Turn off reset guard, so that resetting the machine works normally.
-   This should be called in the various machine_halt, etc., functions.  */
-static inline void disable_reset_guard (void)
-{
-	RESET_GUARD = 0;
-}
-#endif /* CONFIG_RESET_GUARD */
-
-#endif /* !__ASSEMBLY__ */
-
-
-/* A `state save frame' is a struct pt_regs preceded by some extra space
-   suitable for a function call stack frame.  */
-
-/* Amount of room on the stack reserved for arguments and to satisfy the
-   C calling conventions, in addition to the space used by the struct
-   pt_regs that actually holds saved values.  */
-#define STATE_SAVE_ARG_SPACE	(6*4) /* Up to six arguments.  */
-
-
-#ifdef __ASSEMBLY__
-
-/* The size of a state save frame.  */
-#define STATE_SAVE_SIZE		(PT_SIZE + STATE_SAVE_ARG_SPACE)
-
-#else /* !__ASSEMBLY__ */
-
-/* The size of a state save frame.  */
-#define STATE_SAVE_SIZE	       (sizeof (struct pt_regs) + STATE_SAVE_ARG_SPACE)
-
-#endif /* __ASSEMBLY__ */
-
-
-/* Offset of the struct pt_regs in a state save frame.  */
-#define STATE_SAVE_PT_OFFSET	STATE_SAVE_ARG_SPACE
-
-
-#endif /* __V850_ENTRY_H__ */
diff --git a/include/asm-v850/errno.h b/include/asm-v850/errno.h
deleted file mode 100644
index 31c91df01205..000000000000
--- a/include/asm-v850/errno.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __V850_ERRNO_H__
-#define __V850_ERRNO_H__
-
-#include <asm-generic/errno.h>
-
-#endif /* __V850_ERRNO_H__ */
diff --git a/include/asm-v850/fb.h b/include/asm-v850/fb.h
deleted file mode 100644
index c7df38030992..000000000000
--- a/include/asm-v850/fb.h
+++ /dev/null
@@ -1,12 +0,0 @@
-#ifndef _ASM_FB_H_
-#define _ASM_FB_H_
-#include <linux/fb.h>
-
-#define fb_pgprotect(...) do {} while (0)
-
-static inline int fb_is_primary_device(struct fb_info *info)
-{
-	return 0;
-}
-
-#endif /* _ASM_FB_H_ */
diff --git a/include/asm-v850/fcntl.h b/include/asm-v850/fcntl.h
deleted file mode 100644
index 3af4d56776dd..000000000000
--- a/include/asm-v850/fcntl.h
+++ /dev/null
@@ -1,11 +0,0 @@
-#ifndef __V850_FCNTL_H__
-#define __V850_FCNTL_H__
-
-#define O_DIRECTORY	040000	/* must be a directory */
-#define O_NOFOLLOW     0100000	/* don't follow links */
-#define O_DIRECT       0200000	/* direct disk access hint - currently ignored */
-#define O_LARGEFILE    0400000
-
-#include <asm-generic/fcntl.h>
-
-#endif /* __V850_FCNTL_H__ */
diff --git a/include/asm-v850/flat.h b/include/asm-v850/flat.h
deleted file mode 100644
index 17f0ea566611..000000000000
--- a/include/asm-v850/flat.h
+++ /dev/null
@@ -1,133 +0,0 @@
-/*
- * include/asm-v850/flat.h -- uClinux flat-format executables
- *
- *  Copyright (C) 2002,03  NEC Electronics Corporation
- *  Copyright (C) 2002,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_FLAT_H__
-#define __V850_FLAT_H__
-
-/* The amount by which a relocation can exceed the program image limits
-   without being regarded as an error.  On the v850, the relocations of
-   some base-pointers can be offset by 0x8000 (to allow better usage of the
-   space offered by 16-bit signed offsets -- in most cases the offsets used
-   with such a base-pointer will be negative).  */
-
-#define	flat_reloc_valid(reloc, size)	((reloc) <= (size + 0x8000))
-
-#define	flat_stack_align(sp)		/* nothing needed */
-#define	flat_argvp_envp_on_stack()	0
-#define	flat_old_ram_flag(flags)	(flags)
-#define	flat_set_persistent(relval, p)	0
-
-/* We store the type of relocation in the top 4 bits of the `relval.' */
-
-/* Convert a relocation entry into an address.  */
-static inline unsigned long
-flat_get_relocate_addr (unsigned long relval)
-{
-	return relval & 0x0fffffff; /* Mask out top 4-bits */
-}
-
-#define flat_v850_get_reloc_type(relval) ((relval) >> 28)
-
-#define FLAT_V850_R_32		0 /* Normal 32-bit reloc */
-#define FLAT_V850_R_HI16S_LO15	1 /* High 16-bits + signed 15-bit low field */
-#define FLAT_V850_R_HI16S_LO16	2 /* High 16-bits + signed 16-bit low field */
-
-/* Extract the address to be relocated from the symbol reference at RP;
-   RELVAL is the raw relocation-table entry from which RP is derived.
-   For the v850, RP should always be half-word aligned.  */
-static inline unsigned long flat_get_addr_from_rp (unsigned long *rp,
-						   unsigned long relval,
-						   unsigned long flags,
-						   unsigned long *persistent)
-{
-	short *srp = (short *)rp;
-
-	switch (flat_v850_get_reloc_type (relval))
-	{
-	case FLAT_V850_R_32:
-		/* Simple 32-bit address.  */
-		return srp[0] | (srp[1] << 16);
-
-	case FLAT_V850_R_HI16S_LO16:
-		/* The high and low halves of the address are in the 16
-		   bits at RP, and the 2nd word of the 32-bit instruction
-		   following that, respectively.  The low half is _signed_
-		   so we have to sign-extend it and add it to the upper
-		   half instead of simply or-ing them together.
-
-		   Unlike most relocated address, this one is stored in
-		   native (little-endian) byte-order to avoid problems with
-		   trashing the low-order bit, so we have to convert to
-		   network-byte-order before returning, as that's what the
-		   caller expects.  */
-		return htonl ((srp[0] << 16) + srp[2]);
-
-	case FLAT_V850_R_HI16S_LO15:
-		/* The high and low halves of the address are in the 16
-		   bits at RP, and the upper 15 bits of the 2nd word of the
-		   32-bit instruction following that, respectively.  The
-		   low half is _signed_ so we have to sign-extend it and
-		   add it to the upper half instead of simply or-ing them
-		   together.  The lowest bit is always zero.
-
-		   Unlike most relocated address, this one is stored in
-		   native (little-endian) byte-order to avoid problems with
-		   trashing the low-order bit, so we have to convert to
-		   network-byte-order before returning, as that's what the
-		   caller expects.  */
-		return htonl ((srp[0] << 16) + (srp[2] & ~0x1));
-
-	default:
-		return ~0;	/* bogus value */
-	}
-}
-
-/* Insert the address ADDR into the symbol reference at RP;
-   RELVAL is the raw relocation-table entry from which RP is derived.
-   For the v850, RP should always be half-word aligned.  */
-static inline void flat_put_addr_at_rp (unsigned long *rp, unsigned long addr,
-					unsigned long relval)
-{
-	short *srp = (short *)rp;
-
-	switch (flat_v850_get_reloc_type (relval)) {
-	case FLAT_V850_R_32:
-		/* Simple 32-bit address.  */
-		srp[0] = addr & 0xFFFF;
-		srp[1] = (addr >> 16);
-		break;
-
-	case FLAT_V850_R_HI16S_LO16:
-		/* The high and low halves of the address are in the 16
-		   bits at RP, and the 2nd word of the 32-bit instruction
-		   following that, respectively.  The low half is _signed_
-		   so we must carry its sign bit to the upper half before
-		   writing the upper half.  */
-		srp[0] = (addr >> 16) + ((addr >> 15) & 0x1);
-		srp[2] = addr & 0xFFFF;
-		break;
-
-	case FLAT_V850_R_HI16S_LO15:
-		/* The high and low halves of the address are in the 16
-		   bits at RP, and the upper 15 bits of the 2nd word of the
-		   32-bit instruction following that, respectively.  The
-		   low half is _signed_ so we must carry its sign bit to
-		   the upper half before writing the upper half.  The
-		   lowest bit we preserve from the existing instruction.  */
-		srp[0] = (addr >> 16) + ((addr >> 15) & 0x1);
-		srp[2] = (addr & 0xFFFE) | (srp[2] & 0x1);
-		break;
-	}
-}
-
-#endif /* __V850_FLAT_H__ */
diff --git a/include/asm-v850/fpga85e2c.h b/include/asm-v850/fpga85e2c.h
deleted file mode 100644
index 23aae666c718..000000000000
--- a/include/asm-v850/fpga85e2c.h
+++ /dev/null
@@ -1,82 +0,0 @@
-/*
- * include/asm-v850/fpga85e2c.h -- Machine-dependent defs for
- *	FPGA implementation of V850E2/NA85E2C
- *
- *  Copyright (C) 2002,03  NEC Electronics Corporation
- *  Copyright (C) 2002,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_FPGA85E2C_H__
-#define __V850_FPGA85E2C_H__
-
-#include <asm/v850e2.h>
-#include <asm/clinkage.h>
-
-
-#define CPU_MODEL	"v850e2/fpga85e2c"
-#define CPU_MODEL_LONG	"NEC V850E2/NA85E2C"
-#define PLATFORM	"fpga85e2c"
-#define PLATFORM_LONG	"NA85E2C FPGA implementation"
-
-
-/* `external ram'.  */
-#define ERAM_ADDR		0
-#define ERAM_SIZE		0x00100000 /* 1MB */
-
-
-/* FPGA specific control registers.  */
-
-/* Writing a non-zero value to FLGREG(0) will signal the controlling CPU
-   to stop execution.  */
-#define FLGREG_ADDR(n)		(0xFFE80100 + 2*(n))
-#define FLGREG(n)		(*(volatile unsigned char *)FLGREG_ADDR (n))
-#define FLGREG_NUM		2
-
-#define CSDEV_ADDR(n)		(0xFFE80110 + 2*(n))
-#define CSDEV(n)		(*(volatile unsigned char *)CSDEV_ADDR (n))
-
-
-/* Timer interrupts 0-3, interrupt at intervals from CLK/4096 to CLK/16384.  */
-#define IRQ_RPU(n)		(60 + (n))
-#define IRQ_RPU_NUM		4
-
-/* For <asm/irq.h> */
-#define NUM_CPU_IRQS		64
-
-
-/* General-purpose timer.  */
-/* control/status register (can only be read/written via bit insns) */
-#define RPU_GTMC_ADDR		0xFFFFFB00
-#define RPU_GTMC		(*(volatile unsigned char *)RPU_GTMC_ADDR)
-#define RPU_GTMC_CE_BIT		7 /* clock enable (control) */
-#define RPU_GTMC_OV_BIT		6 /* overflow (status) */
-#define RPU_GTMC_CLK_BIT	1 /* 0 = .5 MHz CLK, 1 = 1 Mhz (control) */
-/* 32-bit count (8 least-significant bits are always zero).  */
-#define RPU_GTM_ADDR		0xFFFFFB28
-#define RPU_GTM			(*(volatile unsigned long *)RPU_GTMC_ADDR)
-
-
-/* For <asm/page.h> */
-#define PAGE_OFFSET		ERAM_ADDR /* minimum allocatable address */
-
-
-/* For <asm/entry.h> */
-/* `R0 RAM', used for a few miscellaneous variables that must be accessible
-   using a load instruction relative to R0.  The FPGA implementation
-   actually has no on-chip RAM, so we use part of main ram just after the
-   interrupt vectors.  */
-#ifdef __ASSEMBLY__
-#define R0_RAM_ADDR		lo(C_SYMBOL_NAME(_r0_ram))
-#else
-extern char _r0_ram;
-#define R0_RAM_ADDR		((unsigned long)&_r0_ram);
-#endif
-
-
-#endif /* __V850_FPGA85E2C_H__ */
diff --git a/include/asm-v850/futex.h b/include/asm-v850/futex.h
deleted file mode 100644
index 6a332a9f099c..000000000000
--- a/include/asm-v850/futex.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef _ASM_FUTEX_H
-#define _ASM_FUTEX_H
-
-#include <asm-generic/futex.h>
-
-#endif
diff --git a/include/asm-v850/gbus_int.h b/include/asm-v850/gbus_int.h
deleted file mode 100644
index 0c4bce753c7e..000000000000
--- a/include/asm-v850/gbus_int.h
+++ /dev/null
@@ -1,97 +0,0 @@
-/*
- * include/asm-v850/gbus_int.h -- Midas labs GBUS interrupt support
- *
- *  Copyright (C) 2001,02  NEC Corporation
- *  Copyright (C) 2001,02  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_GBUS_INT_H__
-#define __V850_GBUS_INT_H__
-
-
-/* The GBUS interrupt interface has 32 interrupts shared among 4
-   processor interrupts.  The 32 GBUS interrupts are divided into two
-   sets of 16 each, for allocating among control registers, etc (there
-   are two of each control register, with bits 0-15 controlling an
-   interrupt each).  */
-
-/* The GBUS interrupts themselves.  */
-#define IRQ_GBUS_INT(n)		(GBUS_INT_BASE_IRQ + (n))
-#define IRQ_GBUS_INT_NUM	32
-
-/* Control registers.  */
-#define GBUS_INT_STATUS_ADDR(w)	(GBUS_INT_BASE_ADDR + (w)*0x40)
-#define GBUS_INT_STATUS(w)	(*(volatile u16 *)GBUS_INT_STATUS_ADDR(w))
-#define GBUS_INT_CLEAR_ADDR(w)	(GBUS_INT_BASE_ADDR + 0x10 + (w)*0x40)
-#define GBUS_INT_CLEAR(w)	(*(volatile u16 *)GBUS_INT_CLEAR_ADDR(w))
-#define GBUS_INT_EDGE_ADDR(w)	(GBUS_INT_BASE_ADDR + 0x20 + (w)*0x40)
-#define GBUS_INT_EDGE(w)	(*(volatile u16 *)GBUS_INT_EDGE_ADDR(w))
-#define GBUS_INT_POLARITY_ADDR(w)	(GBUS_INT_BASE_ADDR + 0x30 + (w)*0x40)
-#define GBUS_INT_POLARITY(w)	(*(volatile u16 *)GBUS_INT_POLARITY_ADDR(w))
-/* This allows enabling interrupt bits in word W for interrupt GINTn.  */
-#define GBUS_INT_ENABLE_ADDR(w, n) \
-   (GBUS_INT_BASE_ADDR + 0x100 + (w)*0x10 + (n)*0x20)
-#define GBUS_INT_ENABLE(w, n)	(*(volatile u16 *)GBUS_INT_ENABLE_ADDR(w, n))
-
-/* Mapping between kernel interrupt numbers and hardware control regs/bits.  */
-#define GBUS_INT_BITS_PER_WORD	16
-#define GBUS_INT_NUM_WORDS	(IRQ_GBUS_INT_NUM / GBUS_INT_BITS_PER_WORD)
-#define GBUS_INT_IRQ_WORD(irq)	(((irq) - GBUS_INT_BASE_IRQ) >> 4)
-#define GBUS_INT_IRQ_BIT(irq)	(((irq) - GBUS_INT_BASE_IRQ) & 0xF)
-#define GBUS_INT_IRQ_MASK(irq)	(1 << GBUS_INT_IRQ_BIT(irq))
-
-
-/* Possible priorities for GBUS interrupts.  */
-#define GBUS_INT_PRIORITY_HIGH		2
-#define GBUS_INT_PRIORITY_MEDIUM	4
-#define GBUS_INT_PRIORITY_LOW		6
-
-
-#ifndef __ASSEMBLY__
-
-/* Enable interrupt handling for interrupt IRQ.  */
-extern void gbus_int_enable_irq (unsigned irq);
-/* Disable interrupt handling for interrupt IRQ.  Note that any
-   interrupts received while disabled will be delivered once the
-   interrupt is enabled again, unless they are explicitly cleared using
-   `gbus_int_clear_pending_irq'.  */
-extern void gbus_int_disable_irq (unsigned irq);
-/* Return true if interrupt handling for interrupt IRQ is enabled.  */
-extern int gbus_int_irq_enabled (unsigned irq);
-/* Disable all GBUS irqs.  */
-extern void gbus_int_disable_irqs (void);
-/* Clear any pending interrupts for IRQ.  */
-extern void gbus_int_clear_pending_irq (unsigned irq);
-/* Return true if interrupt IRQ is pending (but disabled).  */
-extern int gbus_int_irq_pending (unsigned irq);
-
-
-struct gbus_int_irq_init {
-	const char *name;	/* name of interrupt type */
-
-	/* Range of kernel irq numbers for this type:
-	   BASE, BASE+INTERVAL, ..., BASE+INTERVAL*NUM  */
-	unsigned base, num, interval;
-
-	unsigned priority;	/* interrupt priority to assign */
-};
-struct hw_interrupt_type;	/* fwd decl */
-
-/* Initialize HW_IRQ_TYPES for GBUS irqs described in array
-   INITS (which is terminated by an entry with the name field == 0).  */
-extern void gbus_int_init_irq_types (struct gbus_int_irq_init *inits,
-				     struct hw_interrupt_type *hw_irq_types);
-
-/* Initialize GBUS interrupts.  */
-extern void gbus_int_init_irqs (void);
-
-#endif /* !__ASSEMBLY__ */
-
-
-#endif /* __V850_GBUS_INT_H__ */
diff --git a/include/asm-v850/hardirq.h b/include/asm-v850/hardirq.h
deleted file mode 100644
index 04e20127c5af..000000000000
--- a/include/asm-v850/hardirq.h
+++ /dev/null
@@ -1,28 +0,0 @@
-#ifndef __V850_HARDIRQ_H__
-#define __V850_HARDIRQ_H__
-
-#include <linux/threads.h>
-#include <linux/cache.h>
-
-#include <asm/irq.h>
-
-typedef struct {
-	unsigned int __softirq_pending;
-} ____cacheline_aligned irq_cpustat_t;
-
-#include <linux/irq_cpustat.h>	/* Standard mappings for irq_cpustat_t above */
-
-#define HARDIRQ_BITS	8
-
-/*
- * The hardirq mask has to be large enough to have
- * space for potentially all IRQ sources in the system
- * nesting on a single CPU:
- */
-#if (1 << HARDIRQ_BITS) < NR_IRQS
-# error HARDIRQ_BITS is too low!
-#endif
-
-void ack_bad_irq(unsigned int irq);
-
-#endif /* __V850_HARDIRQ_H__ */
diff --git a/include/asm-v850/highres_timer.h b/include/asm-v850/highres_timer.h
deleted file mode 100644
index 486fb49ceab6..000000000000
--- a/include/asm-v850/highres_timer.h
+++ /dev/null
@@ -1,44 +0,0 @@
-/*
- * include/asm-v850/highres_timer.h -- High resolution timing routines
- *
- *  Copyright (C) 2001,03  NEC Electronics Corporation
- *  Copyright (C) 2001,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_HIGHRES_TIMER_H__
-#define __V850_HIGHRES_TIMER_H__
-
-#ifndef __ASSEMBLY__
-#include <linux/time.h>
-#endif
-
-#include <asm/entry.h>
-
-
-/* Frequency of the `slow ticks' (one tick each time the fast-tick
-   counter overflows).  */
-#define HIGHRES_TIMER_SLOW_TICK_RATE	25
-
-/* Which timer in the V850E `Timer D' we use.  */
-#define HIGHRES_TIMER_TIMER_D_UNIT	3
-
-
-#ifndef __ASSEMBLY__
-
-extern void highres_timer_start (void), highres_timer_stop (void);
-extern void highres_timer_reset (void);
-extern void highres_timer_read_ticks (u32 *slow_ticks, u32 *fast_ticks);
-extern void highres_timer_ticks_to_timeval (u32 slow_ticks, u32 fast_ticks,
-					    struct timeval *tv);
-extern void highres_timer_read (struct timeval *tv);
-
-#endif /* !__ASSEMBLY__ */
-
-
-#endif /* __V850_HIGHRES_TIMER_H__ */
diff --git a/include/asm-v850/hw_irq.h b/include/asm-v850/hw_irq.h
deleted file mode 100644
index 043e94bb6bd8..000000000000
--- a/include/asm-v850/hw_irq.h
+++ /dev/null
@@ -1,4 +0,0 @@
-#ifndef __V850_HW_IRQ_H__
-#define __V850_HW_IRQ_H__
-
-#endif /* __V850_HW_IRQ_H__ */
diff --git a/include/asm-v850/io.h b/include/asm-v850/io.h
deleted file mode 100644
index cdad251fba9f..000000000000
--- a/include/asm-v850/io.h
+++ /dev/null
@@ -1,142 +0,0 @@
-/*
- * include/asm-v850/io.h -- Misc I/O operations
- *
- *  Copyright (C) 2001,02,03,04,05  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03,04,05  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_IO_H__
-#define __V850_IO_H__
-
-#define IO_SPACE_LIMIT 0xFFFFFFFF
-
-#define readb(addr) \
-  ({ unsigned char __v = (*(volatile unsigned char *) (addr)); __v; })
-#define readw(addr) \
-  ({ unsigned short __v = (*(volatile unsigned short *) (addr)); __v; })
-#define readl(addr) \
-  ({ unsigned long __v = (*(volatile unsigned long *) (addr)); __v; })
-
-#define readb_relaxed(a) readb(a)
-#define readw_relaxed(a) readw(a)
-#define readl_relaxed(a) readl(a)
-
-#define writeb(val, addr) \
-  (void)((*(volatile unsigned char *) (addr)) = (val))
-#define writew(val, addr) \
-  (void)((*(volatile unsigned short *) (addr)) = (val))
-#define writel(val, addr) \
-  (void)((*(volatile unsigned int *) (addr)) = (val))
-
-#define __raw_readb readb
-#define __raw_readw readw
-#define __raw_readl readl
-#define __raw_writeb writeb
-#define __raw_writew writew
-#define __raw_writel writel
-
-#define inb(addr)	readb (addr)
-#define inw(addr)	readw (addr)
-#define inl(addr)	readl (addr)
-#define outb(x, addr)	((void) writeb (x, addr))
-#define outw(x, addr)	((void) writew (x, addr))
-#define outl(x, addr)	((void) writel (x, addr))
-
-#define inb_p(port)		inb((port))
-#define outb_p(val, port)	outb((val), (port))
-#define inw_p(port)		inw((port))
-#define outw_p(val, port)	outw((val), (port))
-#define inl_p(port)		inl((port))
-#define outl_p(val, port)	outl((val), (port))
-
-static inline void insb (unsigned long port, void *dst, unsigned long count)
-{
-	unsigned char *p = dst;
-	while (count--)
-		*p++ = inb (port);
-}
-static inline void insw (unsigned long port, void *dst, unsigned long count)
-{
-	unsigned short *p = dst;
-	while (count--)
-		*p++ = inw (port);
-}
-static inline void insl (unsigned long port, void *dst, unsigned long count)
-{
-	unsigned long *p = dst;
-	while (count--)
-		*p++ = inl (port);
-}
-
-static inline void
-outsb (unsigned long port, const void *src, unsigned long count)
-{
-	const unsigned char *p = src;
-	while (count--)
-		outb (*p++, port);
-}
-static inline void
-outsw (unsigned long port, const void *src, unsigned long count)
-{
-	const unsigned short *p = src;
-	while (count--)
-		outw (*p++, port);
-}
-static inline void
-outsl (unsigned long port, const void *src, unsigned long count)
-{
-	const unsigned long *p = src;
-	while (count--)
-		outl (*p++, port);
-}
-
-
-/* Some places try to pass in an loff_t for PHYSADDR (?!), so we cast it to
-   long before casting it to a pointer to avoid compiler warnings.  */
-#define ioremap(physaddr, size)	((void __iomem *)(unsigned long)(physaddr))
-#define iounmap(addr)		((void)0)
-
-#define ioremap_nocache(physaddr, size)		ioremap (physaddr, size)
-#define ioremap_writethrough(physaddr, size)	ioremap (physaddr, size)
-#define ioremap_fullcache(physaddr, size)	ioremap (physaddr, size)
-
-#define ioread8(addr)		readb (addr)
-#define ioread16(addr)		readw (addr)
-#define ioread32(addr)		readl (addr)
-#define iowrite8(val, addr)	writeb (val, addr)
-#define iowrite16(val, addr)	writew (val, addr)
-#define iowrite32(val, addr)	writel (val, addr)
-
-#define mmiowb()
-
-#define page_to_phys(page)      ((page - mem_map) << PAGE_SHIFT)
-#if 0
-/* This is really stupid; don't define it.  */
-#define page_to_bus(page)       page_to_phys (page)
-#endif
-
-/* Conversion between virtual and physical mappings.  */
-#define phys_to_virt(addr)	((void *)__phys_to_virt (addr))
-#define virt_to_phys(addr)	((unsigned long)__virt_to_phys (addr))
-
-#define memcpy_fromio(dst, src, len) memcpy (dst, (void *)src, len)
-#define memcpy_toio(dst, src, len) memcpy ((void *)dst, src, len)
-
-/*
- * Convert a physical pointer to a virtual kernel pointer for /dev/mem
- * access
- */
-#define xlate_dev_mem_ptr(p)	__va(p)
-
-/*
- * Convert a virtual cached pointer to an uncached pointer
- */
-#define xlate_dev_kmem_ptr(p)	p
-
-#endif /* __V850_IO_H__ */
diff --git a/include/asm-v850/ioctl.h b/include/asm-v850/ioctl.h
deleted file mode 100644
index b279fe06dfe5..000000000000
--- a/include/asm-v850/ioctl.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/ioctl.h>
diff --git a/include/asm-v850/ioctls.h b/include/asm-v850/ioctls.h
deleted file mode 100644
index 5313abd5f388..000000000000
--- a/include/asm-v850/ioctls.h
+++ /dev/null
@@ -1,84 +0,0 @@
-#ifndef __V850_IOCTLS_H__
-#define __V850_IOCTLS_H__
-
-#include <asm/ioctl.h>
-
-/* 0x54 is just a magic number to make these relatively unique ('T') */
-
-#define TCGETS		0x5401
-#define TCSETS		0x5402
-#define TCSETSW		0x5403
-#define TCSETSF		0x5404
-#define TCGETA		0x5405
-#define TCSETA		0x5406
-#define TCSETAW		0x5407
-#define TCSETAF		0x5408
-#define TCSBRK		0x5409
-#define TCXONC		0x540A
-#define TCFLSH		0x540B
-#define TIOCEXCL	0x540C
-#define TIOCNXCL	0x540D
-#define TIOCSCTTY	0x540E
-#define TIOCGPGRP	0x540F
-#define TIOCSPGRP	0x5410
-#define TIOCOUTQ	0x5411
-#define TIOCSTI		0x5412
-#define TIOCGWINSZ	0x5413
-#define TIOCSWINSZ	0x5414
-#define TIOCMGET	0x5415
-#define TIOCMBIS	0x5416
-#define TIOCMBIC	0x5417
-#define TIOCMSET	0x5418
-#define TIOCGSOFTCAR	0x5419
-#define TIOCSSOFTCAR	0x541A
-#define FIONREAD	0x541B
-#define TIOCINQ		FIONREAD
-#define TIOCLINUX	0x541C
-#define TIOCCONS	0x541D
-#define TIOCGSERIAL	0x541E
-#define TIOCSSERIAL	0x541F
-#define TIOCPKT		0x5420
-#define FIONBIO		0x5421
-#define TIOCNOTTY	0x5422
-#define TIOCSETD	0x5423
-#define TIOCGETD	0x5424
-#define TCSBRKP		0x5425	/* Needed for POSIX tcsendbreak() */
-#define TIOCSBRK	0x5427  /* BSD compatibility */
-#define TIOCCBRK	0x5428  /* BSD compatibility */
-#define TIOCGSID	0x5429  /* Return the session ID of FD */
-#define TCGETS2		_IOR('T',0x2A, struct termios2)
-#define TCSETS2		_IOW('T',0x2B, struct termios2)
-#define TCSETSW2	_IOW('T',0x2C, struct termios2)
-#define TCSETSF2	_IOW('T',0x2D, struct termios2)
-#define TIOCGPTN	_IOR('T',0x30, unsigned int) /* Get Pty Number (of pty-mux device) */
-#define TIOCSPTLCK	_IOW('T',0x31, int)  /* Lock/unlock Pty */
-
-#define FIONCLEX	0x5450  /* these numbers need to be adjusted. */
-#define FIOCLEX		0x5451
-#define FIOASYNC	0x5452
-#define TIOCSERCONFIG	0x5453
-#define TIOCSERGWILD	0x5454
-#define TIOCSERSWILD	0x5455
-#define TIOCGLCKTRMIOS	0x5456
-#define TIOCSLCKTRMIOS	0x5457
-#define TIOCSERGSTRUCT	0x5458 /* For debugging only */
-#define TIOCSERGETLSR   0x5459 /* Get line status register */
-#define TIOCSERGETMULTI 0x545A /* Get multiport config  */
-#define TIOCSERSETMULTI 0x545B /* Set multiport config */
-
-#define TIOCMIWAIT	0x545C	/* wait for a change on serial input line(s) */
-#define TIOCGICOUNT	0x545D	/* read serial port inline interrupt counts */
-#define FIOQSIZE	0x545E
-
-/* Used for packet mode */
-#define TIOCPKT_DATA		 0
-#define TIOCPKT_FLUSHREAD	 1
-#define TIOCPKT_FLUSHWRITE	 2
-#define TIOCPKT_STOP		 4
-#define TIOCPKT_START		 8
-#define TIOCPKT_NOSTOP		16
-#define TIOCPKT_DOSTOP		32
-
-#define TIOCSER_TEMT    0x01	/* Transmitter physically empty */
-
-#endif /* __V850_IOCTLS_H__ */
diff --git a/include/asm-v850/ipcbuf.h b/include/asm-v850/ipcbuf.h
deleted file mode 100644
index d8cbe9886d95..000000000000
--- a/include/asm-v850/ipcbuf.h
+++ /dev/null
@@ -1,29 +0,0 @@
-#ifndef __V850E_IPCBUF_H__
-#define __V850E_IPCBUF_H__
-
-/*
- * The user_ipc_perm structure for v850e architecture.
- * Note extra padding because this structure is passed back and forth
- * between kernel and user space.
- *
- * Pad space is left for:
- * - 32-bit mode_t and seq
- * - 2 miscellaneous 32-bit values
- */
-
-struct ipc64_perm
-{
-	__kernel_key_t		key;
-	__kernel_uid32_t	uid;
-	__kernel_gid32_t	gid;
-	__kernel_uid32_t	cuid;
-	__kernel_gid32_t	cgid;
-	__kernel_mode_t		mode;
-	unsigned short		__pad1;
-	unsigned short		seq;
-	unsigned short		__pad2;
-	unsigned long		__unused1;
-	unsigned long		__unused2;
-};
-
-#endif /* __V850E_IPCBUF_H__ */
diff --git a/include/asm-v850/irq.h b/include/asm-v850/irq.h
deleted file mode 100644
index 7d0d4cd1ce54..000000000000
--- a/include/asm-v850/irq.h
+++ /dev/null
@@ -1,55 +0,0 @@
-/*
- * include/asm-v850/irq.h -- Machine interrupt handling
- *
- *  Copyright (C) 2001,02,04  NEC Electronics Corporation
- *  Copyright (C) 2001,02,04  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_IRQ_H__
-#define __V850_IRQ_H__
-
-#include <asm/machdep.h>
-
-/* Default NUM_MACH_IRQS.  */
-#ifndef NUM_MACH_IRQS
-#define NUM_MACH_IRQS	NUM_CPU_IRQS
-#endif
-
-/* NMIs have IRQ numbers from FIRST_NMI to FIRST_NMI+NUM_NMIS-1.  */
-#define FIRST_NMI	NUM_MACH_IRQS
-#define IRQ_NMI(n)	(FIRST_NMI + (n))
-/* v850 processors have 3 non-maskable interrupts.  */
-#define NUM_NMIS	3
-
-/* Includes both maskable and non-maskable irqs.  */
-#define NR_IRQS		(NUM_MACH_IRQS + NUM_NMIS)
-
-
-#ifndef __ASSEMBLY__
-
-struct pt_regs;
-struct hw_interrupt_type;
-struct irqaction;
-
-#define irq_canonicalize(irq)	(irq)
-
-/* Initialize irq handling for IRQs.
-   BASE_IRQ, BASE_IRQ+INTERVAL, ..., BASE_IRQ+NUM*INTERVAL
-   to IRQ_TYPE.  An IRQ_TYPE of 0 means to use a generic interrupt type.  */
-extern void
-init_irq_handlers (int base_irq, int num, int interval,
-		   struct hw_interrupt_type *irq_type);
-
-/* Handle interrupt IRQ.  REGS are the registers at the time of ther
-   interrupt.  */
-extern unsigned int handle_irq (int irq, struct pt_regs *regs);
-
-#endif /* !__ASSEMBLY__ */
-
-#endif /* __V850_IRQ_H__ */
diff --git a/include/asm-v850/irq_regs.h b/include/asm-v850/irq_regs.h
deleted file mode 100644
index 3dd9c0b70270..000000000000
--- a/include/asm-v850/irq_regs.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/irq_regs.h>
diff --git a/include/asm-v850/kdebug.h b/include/asm-v850/kdebug.h
deleted file mode 100644
index 6ece1b037665..000000000000
--- a/include/asm-v850/kdebug.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-generic/kdebug.h>
diff --git a/include/asm-v850/kmap_types.h b/include/asm-v850/kmap_types.h
deleted file mode 100644
index 3288976b161f..000000000000
--- a/include/asm-v850/kmap_types.h
+++ /dev/null
@@ -1,19 +0,0 @@
-#ifndef __V850_KMAP_TYPES_H__
-#define __V850_KMAP_TYPES_H__
-
-enum km_type {
-	KM_BOUNCE_READ,
-	KM_SKB_SUNRPC_DATA,
-	KM_SKB_DATA_SOFTIRQ,
-	KM_USER0,
-	KM_USER1,
-	KM_BIO_SRC_IRQ,
-	KM_BIO_DST_IRQ,
-	KM_PTE0,
-	KM_PTE1,
-	KM_IRQ0,
-	KM_IRQ1,
-	KM_TYPE_NR
-};
-
-#endif /* __V850_KMAP_TYPES_H__ */
diff --git a/include/asm-v850/kvm.h b/include/asm-v850/kvm.h
deleted file mode 100644
index 3f729b79febc..000000000000
--- a/include/asm-v850/kvm.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __LINUX_KVM_V850_H
-#define __LINUX_KVM_V850_H
-
-/* v850 does not support KVM */
-
-#endif
diff --git a/include/asm-v850/linkage.h b/include/asm-v850/linkage.h
deleted file mode 100644
index b6185d3cfe68..000000000000
--- a/include/asm-v850/linkage.h
+++ /dev/null
@@ -1,8 +0,0 @@
-#ifndef __ASM_LINKAGE_H
-#define __ASM_LINKAGE_H
-
-#ifdef __ASSEMBLY__
-#include <asm/asm.h>
-#endif
-
-#endif
diff --git a/include/asm-v850/local.h b/include/asm-v850/local.h
deleted file mode 100644
index 705148abe276..000000000000
--- a/include/asm-v850/local.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __V850_LOCAL_H__
-#define __V850_LOCAL_H__
-
-#include <asm-generic/local.h>
-
-#endif /* __V850_LOCAL_H__ */
diff --git a/include/asm-v850/ma.h b/include/asm-v850/ma.h
deleted file mode 100644
index 89e66473a176..000000000000
--- a/include/asm-v850/ma.h
+++ /dev/null
@@ -1,101 +0,0 @@
-/*
- * include/asm-v850/ma.h -- V850E/MA series of cpu chips
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_MA_H__
-#define __V850_MA_H__
-
-/* The MA series uses the V850E cpu core.  */
-#include <asm/v850e.h>
-
-
-/* For <asm/entry.h> */
-/* We use on-chip RAM, for a few miscellaneous variables that must be
-   accessible using a load instruction relative to R0.  The amount
-   varies between chip models, but there's always at least 4K, and it
-   should always start at FFFFC000.  */
-#define R0_RAM_ADDR			0xFFFFC000
-
-
-/* MA series UART details.  */
-#define V850E_UART_BASE_FREQ		CPU_CLOCK_FREQ
-
-/* This is a function that gets called before configuring the UART.  */
-#define V850E_UART_PRE_CONFIGURE	ma_uart_pre_configure
-#ifndef __ASSEMBLY__
-extern void ma_uart_pre_configure (unsigned chan,
-				   unsigned cflags, unsigned baud);
-#endif
-
-
-/* MA series timer C details.  */
-#define V850E_TIMER_C_BASE_ADDR		0xFFFFF600
-
-
-/* MA series timer D details.  */
-#define V850E_TIMER_D_BASE_ADDR		0xFFFFF540
-#define V850E_TIMER_D_TMD_BASE_ADDR 	(V850E_TIMER_D_BASE_ADDR + 0x0)
-#define V850E_TIMER_D_CMD_BASE_ADDR 	(V850E_TIMER_D_BASE_ADDR + 0x2)
-#define V850E_TIMER_D_TMCD_BASE_ADDR 	(V850E_TIMER_D_BASE_ADDR + 0x4)
-
-#define V850E_TIMER_D_BASE_FREQ		CPU_CLOCK_FREQ
-
-
-/* Port 0 */
-/* Direct I/O.  Bits 0-7 are pins P00-P07.  */
-#define MA_PORT0_IO_ADDR		0xFFFFF400
-#define MA_PORT0_IO			(*(volatile u8 *)MA_PORT0_IO_ADDR)
-/* Port mode (for direct I/O, 0 = output, 1 = input).  */
-#define MA_PORT0_PM_ADDR		0xFFFFF420
-#define MA_PORT0_PM			(*(volatile u8 *)MA_PORT0_PM_ADDR)
-/* Port mode control (0 = direct I/O mode, 1 = alternative I/O mode).  */
-#define MA_PORT0_PMC_ADDR		0xFFFFF440
-#define MA_PORT0_PMC			(*(volatile u8 *)MA_PORT0_PMC_ADDR)
-/* Port function control (for P04-P07, 0 = IRQ, 1 = DMARQ).  */
-#define MA_PORT0_PFC_ADDR		0xFFFFF460
-#define MA_PORT0_PFC			(*(volatile u8 *)MA_PORT0_PFC_ADDR)
-
-/* Port 1 */
-/* Direct I/O.  Bits 0-3 are pins P10-P13.  */
-#define MA_PORT1_IO_ADDR		0xFFFFF402
-#define MA_PORT1_IO			(*(volatile u8 *)MA_PORT1_IO_ADDR)
-/* Port mode (for direct I/O, 0 = output, 1 = input).  */
-#define MA_PORT1_PM_ADDR		0xFFFFF420
-#define MA_PORT1_PM			(*(volatile u8 *)MA_PORT1_PM_ADDR)
-/* Port mode control (0 = direct I/O mode, 1 = alternative I/O mode).  */
-#define MA_PORT1_PMC_ADDR		0xFFFFF442
-#define MA_PORT1_PMC			(*(volatile u8 *)MA_PORT1_PMC_ADDR)
-
-/* Port 4 */
-/* Direct I/O.  Bits 0-5 are pins P40-P45.  */
-#define MA_PORT4_IO_ADDR		0xFFFFF408
-#define MA_PORT4_IO			(*(volatile u8 *)MA_PORT4_IO_ADDR)
-/* Port mode (for direct I/O, 0 = output, 1 = input).  */
-#define MA_PORT4_PM_ADDR		0xFFFFF428
-#define MA_PORT4_PM			(*(volatile u8 *)MA_PORT4_PM_ADDR)
-/* Port mode control (0 = direct I/O mode, 1 = alternative I/O mode).  */
-#define MA_PORT4_PMC_ADDR		0xFFFFF448
-#define MA_PORT4_PMC			(*(volatile u8 *)MA_PORT4_PMC_ADDR)
-/* Port function control (for serial interfaces, 0 = CSI, 1 = UART).  */
-#define MA_PORT4_PFC_ADDR		0xFFFFF468
-#define MA_PORT4_PFC			(*(volatile u8 *)MA_PORT4_PFC_ADDR)
-
-
-#ifndef __ASSEMBLY__
-
-/* Initialize MA chip interrupts.  */
-extern void ma_init_irqs (void);
-
-#endif /* !__ASSEMBLY__ */
-
-
-#endif /* __V850_MA_H__ */
diff --git a/include/asm-v850/ma1.h b/include/asm-v850/ma1.h
deleted file mode 100644
index ede1f1de2b7a..000000000000
--- a/include/asm-v850/ma1.h
+++ /dev/null
@@ -1,50 +0,0 @@
-/*
- * include/asm-v850/ma1.h -- V850E/MA1 cpu chip
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_MA1_H__
-#define __V850_MA1_H__
-
-/* Inherit more generic details from MA series.  */
-#include <asm/ma.h>
-
-
-#define CPU_MODEL	"v850e/ma1"
-#define CPU_MODEL_LONG	"NEC V850E/MA1"
-
-
-/* Hardware-specific interrupt numbers (in the kernel IRQ namespace).  */
-#define IRQ_INTOV(n)	(n)	/* 0-3 */
-#define IRQ_INTOV_NUM	4
-#define IRQ_INTP(n)	(0x4  + (n)) /* Pnnn (pin) interrupts */
-#define IRQ_INTP_NUM	24
-#define IRQ_INTCMD(n)	(0x1c + (n)) /* interval timer interrupts 0-3 */
-#define IRQ_INTCMD_NUM	4
-#define IRQ_INTDMA(n)	(0x20 + (n)) /* DMA interrupts 0-3 */
-#define IRQ_INTDMA_NUM	4
-#define IRQ_INTCSI(n)	(0x24 + (n)*4)/* CSI 0-2 transmit/receive completion */
-#define IRQ_INTCSI_NUM	3
-#define IRQ_INTSER(n)	(0x25 + (n)*4) /* UART 0-2 reception error */
-#define IRQ_INTSER_NUM	3
-#define IRQ_INTSR(n)	(0x26 + (n)*4) /* UART 0-2 reception completion */
-#define IRQ_INTSR_NUM	3
-#define IRQ_INTST(n)	(0x27 + (n)*4) /* UART 0-2 transmission completion */
-#define IRQ_INTST_NUM	3
-
-#define NUM_CPU_IRQS	0x30
-
-
-/* The MA1 has a UART with 3 channels.  */
-#define V850E_UART_NUM_CHANNELS	3
-
-
-#endif /* __V850_MA1_H__ */
diff --git a/include/asm-v850/machdep.h b/include/asm-v850/machdep.h
deleted file mode 100644
index f1e3b8b91508..000000000000
--- a/include/asm-v850/machdep.h
+++ /dev/null
@@ -1,60 +0,0 @@
-/*
- * include/asm-v850/machdep.h -- Machine-dependent definitions
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_MACHDEP_H__
-#define __V850_MACHDEP_H__
-
-
-/* chips */
-#ifdef CONFIG_V850E_MA1
-#include <asm/ma1.h>
-#endif
-#ifdef CONFIG_V850E_ME2
-#include <asm/me2.h>
-#endif
-#ifdef CONFIG_V850E_TEG
-#include <asm/teg.h>
-#endif
-
-/* These are both chips _and_ platforms, so put them in the middle... */
-#ifdef CONFIG_V850E2_ANNA
-#include <asm/anna.h>
-#endif
-#ifdef CONFIG_V850E_AS85EP1
-#include <asm/as85ep1.h>
-#endif
-
-/* platforms */
-#ifdef CONFIG_RTE_CB_MA1
-#include <asm/rte_ma1_cb.h>
-#endif
-#ifdef CONFIG_RTE_CB_ME2
-#include <asm/rte_me2_cb.h>
-#endif
-#ifdef CONFIG_RTE_CB_NB85E
-#include <asm/rte_nb85e_cb.h>
-#endif
-#ifdef CONFIG_V850E_SIM
-#include <asm/sim.h>
-#endif
-#ifdef CONFIG_V850E2_SIM85E2C
-#include <asm/sim85e2c.h>
-#endif
-#ifdef CONFIG_V850E2_SIM85E2S
-#include <asm/sim85e2s.h>
-#endif
-#ifdef CONFIG_V850E2_FPGA85E2C
-#include <asm/fpga85e2c.h>
-#endif
-
-#endif /* __V850_MACHDEP_H__ */
diff --git a/include/asm-v850/macrology.h b/include/asm-v850/macrology.h
deleted file mode 100644
index 37abf874832c..000000000000
--- a/include/asm-v850/macrology.h
+++ /dev/null
@@ -1,17 +0,0 @@
-/*
- * include/asm-v850/macrology.h -- Various useful CPP macros
- *
- *  Copyright (C) 2001  NEC Corporation
- *  Copyright (C) 2001  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#define macrology_paste(arg1, arg2)	macrology_paste_1(arg1, arg2)
-#define macrology_paste_1(arg1, arg2)	arg1 ## arg2
-#define macrology_stringify(sym)	macrology_stringify_1(sym)
-#define macrology_stringify_1(sym)	#sym
diff --git a/include/asm-v850/me2.h b/include/asm-v850/me2.h
deleted file mode 100644
index ac7c9ce0bdc1..000000000000
--- a/include/asm-v850/me2.h
+++ /dev/null
@@ -1,182 +0,0 @@
-/*
- * include/asm-v850/me2.h -- V850E/ME2 cpu chip
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_ME2_H__
-#define __V850_ME2_H__
-
-#include <asm/v850e.h>
-#include <asm/v850e_cache.h>
-
-
-#define CPU_MODEL	"v850e/me2"
-#define CPU_MODEL_LONG	"NEC V850E/ME2"
-
-
-/* Hardware-specific interrupt numbers (in the kernel IRQ namespace).  */
-#define IRQ_INTP(n)       (n) /* Pnnn (pin) interrupts */
-#define IRQ_INTP_NUM      31
-#define IRQ_INTCMD(n)     (0x31 + (n)) /* interval timer interrupts 0-3 */
-#define IRQ_INTCMD_NUM    4
-#define IRQ_INTDMA(n)     (0x41 + (n)) /* DMA interrupts 0-3 */
-#define IRQ_INTDMA_NUM    4
-#define IRQ_INTUBTIRE(n)  (0x49 + (n)*5)/* UARTB 0-1 reception error */
-#define IRQ_INTUBTIRE_NUM 2
-#define IRQ_INTUBTIR(n)   (0x4a + (n)*5) /* UARTB 0-1 reception complete */
-#define IRQ_INTUBTIR_NUM  2
-#define IRQ_INTUBTIT(n)   (0x4b + (n)*5) /* UARTB 0-1 transmission complete */
-#define IRQ_INTUBTIT_NUM  2
-#define IRQ_INTUBTIF(n)   (0x4c + (n)*5) /* UARTB 0-1 FIFO trans. complete */
-#define IRQ_INTUBTIF_NUM  2
-#define IRQ_INTUBTITO(n)  (0x4d + (n)*5) /* UARTB 0-1 reception timeout */
-#define IRQ_INTUBTITO_NUM 2
-
-/* For <asm/irq.h> */
-#define NUM_CPU_IRQS		0x59 /* V850E/ME2 */
-
-
-/* For <asm/entry.h> */
-/* We use on-chip RAM, for a few miscellaneous variables that must be
-   accessible using a load instruction relative to R0.  */
-#define R0_RAM_ADDR			0xFFFFB000 /* V850E/ME2 */
-
-
-/* V850E/ME2 UARTB details.*/
-#define V850E_UART_NUM_CHANNELS		2
-#define V850E_UARTB_BASE_FREQ		(CPU_CLOCK_FREQ / 4)
-
-/* This is a function that gets called before configuring the UART.  */
-#define V850E_UART_PRE_CONFIGURE	me2_uart_pre_configure
-#ifndef __ASSEMBLY__
-extern void me2_uart_pre_configure (unsigned chan,
-				    unsigned cflags, unsigned baud);
-#endif /* __ASSEMBLY__ */
-
-
-/* V850E/ME2 timer C details.  */
-#define V850E_TIMER_C_BASE_ADDR		0xFFFFF600
-
-
-/* V850E/ME2 timer D details.  */
-#define V850E_TIMER_D_BASE_ADDR		0xFFFFF540
-#define V850E_TIMER_D_TMD_BASE_ADDR	(V850E_TIMER_D_BASE_ADDR + 0x0)
-#define V850E_TIMER_D_CMD_BASE_ADDR 	(V850E_TIMER_D_BASE_ADDR + 0x2)
-#define V850E_TIMER_D_TMCD_BASE_ADDR	(V850E_TIMER_D_BASE_ADDR + 0x4)
-
-#define V850E_TIMER_D_BASE_FREQ		(CPU_CLOCK_FREQ / 2)
-
-
-/* Select iRAM mode.  */
-#define ME2_IRAMM_ADDR			0xFFFFF80A
-#define ME2_IRAMM			(*(volatile u8*)ME2_IRAMM_ADDR)
-
-
-/* Interrupt edge-detection configuration.  INTF(n) and INTR(n) are only
-   valid for n == 1, 2, or 5.  */
-#define ME2_INTF_ADDR(n)		(0xFFFFFC00 + (n) * 0x2)
-#define ME2_INTF(n)			(*(volatile u8*)ME2_INTF_ADDR(n))
-#define ME2_INTR_ADDR(n)		(0xFFFFFC20 + (n) * 0x2)
-#define ME2_INTR(n)			(*(volatile u8*)ME2_INTR_ADDR(n))
-#define ME2_INTFAL_ADDR			0xFFFFFC10
-#define ME2_INTFAL			(*(volatile u8*)ME2_INTFAL_ADDR)
-#define ME2_INTRAL_ADDR			0xFFFFFC30
-#define ME2_INTRAL			(*(volatile u8*)ME2_INTRAL_ADDR)
-#define ME2_INTFDH_ADDR			0xFFFFFC16
-#define ME2_INTFDH			(*(volatile u16*)ME2_INTFDH_ADDR)
-#define ME2_INTRDH_ADDR			0xFFFFFC36
-#define ME2_INTRDH			(*(volatile u16*)ME2_INTRDH_ADDR)
-#define ME2_SESC_ADDR(n)		(0xFFFFF609 + (n) * 0x10)
-#define ME2_SESC(n)			(*(volatile u8*)ME2_SESC_ADDR(n))
-#define ME2_SESA10_ADDR			0xFFFFF5AD
-#define ME2_SESA10			(*(volatile u8*)ME2_SESA10_ADDR)
-#define ME2_SESA11_ADDR			0xFFFFF5DD
-#define ME2_SESA11			(*(volatile u8*)ME2_SESA11_ADDR)
-
-
-/* Port 1 */
-/* Direct I/O.  Bits 0-3 are pins P10-P13.  */
-#define ME2_PORT1_IO_ADDR		0xFFFFF402
-#define ME2_PORT1_IO			(*(volatile u8 *)ME2_PORT1_IO_ADDR)
-/* Port mode (for direct I/O, 0 = output, 1 = input).  */
-#define ME2_PORT1_PM_ADDR		0xFFFFF422
-#define ME2_PORT1_PM			(*(volatile u8 *)ME2_PORT1_PM_ADDR)
-/* Port mode control (0 = direct I/O mode, 1 = alternative I/O mode).  */
-#define ME2_PORT1_PMC_ADDR		0xFFFFF442
-#define ME2_PORT1_PMC			(*(volatile u8 *)ME2_PORT1_PMC_ADDR)
-/* Port function control (for serial interfaces, 0 = CSI30, 1 = UARTB0 ).  */
-#define ME2_PORT1_PFC_ADDR		0xFFFFF462
-#define ME2_PORT1_PFC			(*(volatile u8 *)ME2_PORT1_PFC_ADDR)
-
-/* Port 2 */
-/* Direct I/O.  Bits 0-3 are pins P20-P25.  */
-#define ME2_PORT2_IO_ADDR		0xFFFFF404
-#define ME2_PORT2_IO			(*(volatile u8 *)ME2_PORT2_IO_ADDR)
-/* Port mode (for direct I/O, 0 = output, 1 = input).  */
-#define ME2_PORT2_PM_ADDR		0xFFFFF424
-#define ME2_PORT2_PM			(*(volatile u8 *)ME2_PORT2_PM_ADDR)
-/* Port mode control (0 = direct I/O mode, 1 = alternative I/O mode).  */
-#define ME2_PORT2_PMC_ADDR		0xFFFFF444
-#define ME2_PORT2_PMC			(*(volatile u8 *)ME2_PORT2_PMC_ADDR)
-/* Port function control (for serial interfaces, 0 = INTP2x, 1 = UARTB1 ).  */
-#define ME2_PORT2_PFC_ADDR		0xFFFFF464
-#define ME2_PORT2_PFC			(*(volatile u8 *)ME2_PORT2_PFC_ADDR)
-
-/* Port 5 */
-/* Direct I/O.  Bits 0-5 are pins P50-P55.  */
-#define ME2_PORT5_IO_ADDR		0xFFFFF40A
-#define ME2_PORT5_IO			(*(volatile u8 *)ME2_PORT5_IO_ADDR)
-/* Port mode (for direct I/O, 0 = output, 1 = input).  */
-#define ME2_PORT5_PM_ADDR		0xFFFFF42A
-#define ME2_PORT5_PM			(*(volatile u8 *)ME2_PORT5_PM_ADDR)
-/* Port mode control (0 = direct I/O mode, 1 = alternative I/O mode).  */
-#define ME2_PORT5_PMC_ADDR		0xFFFFF44A
-#define ME2_PORT5_PMC			(*(volatile u8 *)ME2_PORT5_PMC_ADDR)
-/* Port function control ().  */
-#define ME2_PORT5_PFC_ADDR		0xFFFFF46A
-#define ME2_PORT5_PFC			(*(volatile u8 *)ME2_PORT5_PFC_ADDR)
-
-/* Port 6 */
-/* Direct I/O.  Bits 5-7 are pins P65-P67.  */
-#define ME2_PORT6_IO_ADDR		0xFFFFF40C
-#define ME2_PORT6_IO			(*(volatile u8 *)ME2_PORT6_IO_ADDR)
-/* Port mode (for direct I/O, 0 = output, 1 = input).  */
-#define ME2_PORT6_PM_ADDR		0xFFFFF42C
-#define ME2_PORT6_PM			(*(volatile u8 *)ME2_PORT6_PM_ADDR)
-/* Port mode control (0 = direct I/O mode, 1 = alternative I/O mode).  */
-#define ME2_PORT6_PMC_ADDR		0xFFFFF44C
-#define ME2_PORT6_PMC			(*(volatile u8 *)ME2_PORT6_PMC_ADDR)
-/* Port function control ().  */
-#define ME2_PORT6_PFC_ADDR		0xFFFFF46C
-#define ME2_PORT6_PFC			(*(volatile u8 *)ME2_PORT6_PFC_ADDR)
-
-/* Port 7 */
-/* Direct I/O.  Bits 2-7 are pins P72-P77.  */
-#define ME2_PORT7_IO_ADDR		0xFFFFF40E
-#define ME2_PORT7_IO			(*(volatile u8 *)ME2_PORT7_IO_ADDR)
-/* Port mode (for direct I/O, 0 = output, 1 = input).  */
-#define ME2_PORT7_PM_ADDR		0xFFFFF42E
-#define ME2_PORT7_PM			(*(volatile u8 *)ME2_PORT7_PM_ADDR)
-/* Port mode control (0 = direct I/O mode, 1 = alternative I/O mode).  */
-#define ME2_PORT7_PMC_ADDR		0xFFFFF44E
-#define ME2_PORT7_PMC			(*(volatile u8 *)ME2_PORT7_PMC_ADDR)
-/* Port function control ().  */
-#define ME2_PORT7_PFC_ADDR		0xFFFFF46E
-#define ME2_PORT7_PFC			(*(volatile u8 *)ME2_PORT7_PFC_ADDR)
-
-
-#ifndef __ASSEMBLY__
-/* Initialize V850E/ME2 chip interrupts.  */
-extern void me2_init_irqs (void);
-#endif /* !__ASSEMBLY__ */
-
-
-#endif /* __V850_ME2_H__ */
diff --git a/include/asm-v850/mman.h b/include/asm-v850/mman.h
deleted file mode 100644
index edbf6edbfb37..000000000000
--- a/include/asm-v850/mman.h
+++ /dev/null
@@ -1,15 +0,0 @@
-#ifndef __V850_MMAN_H__
-#define __V850_MMAN_H__
-
-#include <asm-generic/mman.h>
-
-#define MAP_GROWSDOWN	0x0100		/* stack-like segment */
-#define MAP_DENYWRITE	0x0800		/* ETXTBSY */
-#define MAP_EXECUTABLE	0x1000		/* mark it as an executable */
-#define MAP_LOCKED	0x2000		/* pages are locked */
-#define MAP_NORESERVE	0x4000		/* don't check for reservations */
-
-#define MCL_CURRENT	1		/* lock all current mappings */
-#define MCL_FUTURE	2		/* lock all future mappings */
-
-#endif /* __V850_MMAN_H__ */
diff --git a/include/asm-v850/mmu.h b/include/asm-v850/mmu.h
deleted file mode 100644
index 267768c66ef6..000000000000
--- a/include/asm-v850/mmu.h
+++ /dev/null
@@ -1,11 +0,0 @@
-/* Copyright (C) 2002, 2005, David McCullough <davidm@snapgear.com> */
-
-#ifndef __V850_MMU_H__
-#define __V850_MMU_H__
-
-typedef struct {
-	struct vm_list_struct	*vmlist;
-	unsigned long		end_brk;
-} mm_context_t;
-
-#endif /* __V850_MMU_H__ */
diff --git a/include/asm-v850/mmu_context.h b/include/asm-v850/mmu_context.h
deleted file mode 100644
index 01daacd5474e..000000000000
--- a/include/asm-v850/mmu_context.h
+++ /dev/null
@@ -1,13 +0,0 @@
-#ifndef __V850_MMU_CONTEXT_H__
-#define __V850_MMU_CONTEXT_H__
-
-#include <asm-generic/mm_hooks.h>
-
-#define destroy_context(mm)		((void)0)
-#define init_new_context(tsk,mm)	0
-#define switch_mm(prev,next,tsk)	((void)0)
-#define deactivate_mm(tsk,mm)		do { } while (0)
-#define activate_mm(prev,next)		((void)0)
-#define enter_lazy_tlb(mm,tsk)		((void)0)
-
-#endif /* __V850_MMU_CONTEXT_H__ */
diff --git a/include/asm-v850/module.h b/include/asm-v850/module.h
deleted file mode 100644
index 2c2f4944f09f..000000000000
--- a/include/asm-v850/module.h
+++ /dev/null
@@ -1,62 +0,0 @@
-/*
- * include/asm-v850/module.h -- Architecture-specific module hooks
- *
- *  Copyright (C) 2001,02,03,04  NEC Corporation
- *  Copyright (C) 2001,02,03,04  Miles Bader <miles@gnu.org>
- *  Copyright (C) 2001,03  Rusty Russell
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- *
- * Derived in part from include/asm-ppc/module.h
- */
-
-#ifndef __V850_MODULE_H__
-#define __V850_MODULE_H__
-
-#define MODULE_SYMBOL_PREFIX "_"
-
-struct v850_plt_entry
-{
-	/* Indirect jump instruction sequence (6-byte mov + 2-byte jr).  */
-	unsigned long tramp[2];
-};
-
-struct mod_arch_specific
-{
-	/* Indices of PLT sections within module. */
-	unsigned int core_plt_section, init_plt_section;
-};
-
-#define Elf_Shdr Elf32_Shdr
-#define Elf_Sym Elf32_Sym
-#define Elf_Ehdr Elf32_Ehdr
-
-/* Make empty sections for module_frob_arch_sections to expand. */
-#ifdef MODULE
-asm(".section .plt,\"ax\",@nobits; .align 3; .previous");
-asm(".section .init.plt,\"ax\",@nobits; .align 3; .previous");
-#endif
-
-/* We don't do exception tables.  */
-struct exception_table_entry;
-static inline const struct exception_table_entry *
-search_extable(const struct exception_table_entry *first,
-	       const struct exception_table_entry *last,
-	       unsigned long value)
-{
-	return 0;
-}
-#define ARCH_HAS_SEARCH_EXTABLE
-static inline void
-sort_extable(struct exception_table_entry *start,
-	     struct exception_table_entry *finish)
-{
-	/* nada */
-}
-#define ARCH_HAS_SORT_EXTABLE
-
-#endif /* __V850_MODULE_H__ */
diff --git a/include/asm-v850/msgbuf.h b/include/asm-v850/msgbuf.h
deleted file mode 100644
index ed07dbd01637..000000000000
--- a/include/asm-v850/msgbuf.h
+++ /dev/null
@@ -1,31 +0,0 @@
-#ifndef __V850_MSGBUF_H__
-#define __V850_MSGBUF_H__
-
-/* 
- * The msqid64_ds structure for v850 architecture.
- * Note extra padding because this structure is passed back and forth
- * between kernel and user space.
- *
- * Pad space is left for:
- * - 64-bit time_t to solve y2038 problem
- * - 2 miscellaneous 32-bit values
- */
-
-struct msqid64_ds {
-	struct ipc64_perm msg_perm;
-	__kernel_time_t msg_stime;	/* last msgsnd time */
-	unsigned long	__unused1;
-	__kernel_time_t msg_rtime;	/* last msgrcv time */
-	unsigned long	__unused2;
-	__kernel_time_t msg_ctime;	/* last change time */
-	unsigned long	__unused3;
-	unsigned long  msg_cbytes;	/* current number of bytes on queue */
-	unsigned long  msg_qnum;	/* number of messages in queue */
-	unsigned long  msg_qbytes;	/* max number of bytes on queue */
-	__kernel_pid_t msg_lspid;	/* pid of last msgsnd */
-	__kernel_pid_t msg_lrpid;	/* last receive pid */
-	unsigned long  __unused4;
-	unsigned long  __unused5;
-};
-
-#endif /* __V850_MSGBUF_H__ */
diff --git a/include/asm-v850/mutex.h b/include/asm-v850/mutex.h
deleted file mode 100644
index 458c1f7fbc18..000000000000
--- a/include/asm-v850/mutex.h
+++ /dev/null
@@ -1,9 +0,0 @@
-/*
- * Pull in the generic implementation for the mutex fastpath.
- *
- * TODO: implement optimized primitives instead, or leave the generic
- * implementation in place, or pick the atomic_xchg() based generic
- * implementation. (see asm-generic/mutex-xchg.h for details)
- */
-
-#include <asm-generic/mutex-dec.h>
diff --git a/include/asm-v850/page.h b/include/asm-v850/page.h
deleted file mode 100644
index f9de35d873fa..000000000000
--- a/include/asm-v850/page.h
+++ /dev/null
@@ -1,124 +0,0 @@
-/*
- * include/asm-v850/page.h -- VM ops
- *
- *  Copyright (C) 2001,02,03,05  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03,05  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_PAGE_H__
-#define __V850_PAGE_H__
-
-#include <asm/machdep.h>
-
-
-#define PAGE_SHIFT	12
-#define PAGE_SIZE       (1UL << PAGE_SHIFT)
-#define PAGE_MASK       (~(PAGE_SIZE-1))
-
-
-/*
- * PAGE_OFFSET -- the first address of the first page of memory. For archs with
- * no MMU this corresponds to the first free page in physical memory (aligned
- * on a page boundary).
- */
-#ifndef PAGE_OFFSET
-#define PAGE_OFFSET  0x0000000
-#endif
-
-
-#ifndef __ASSEMBLY__
-
-#define STRICT_MM_TYPECHECKS
-
-#define clear_page(page)	memset ((void *)(page), 0, PAGE_SIZE)
-#define copy_page(to, from)	memcpy ((void *)(to), (void *)from, PAGE_SIZE)
-
-#define clear_user_page(addr, vaddr, page)	\
-	do { 	clear_page(addr);		\
-		flush_dcache_page(page);	\
-	} while (0)
-#define copy_user_page(to, from, vaddr, page)	\
-	do {	copy_page(to, from);		\
-		flush_dcache_page(page);	\
-	} while (0)
-
-#ifdef STRICT_MM_TYPECHECKS
-/*
- * These are used to make use of C type-checking..
- */
-
-typedef struct { unsigned long pte; } pte_t;
-typedef struct { unsigned long pmd; } pmd_t;
-typedef struct { unsigned long pgd; } pgd_t;
-typedef struct { unsigned long pgprot; } pgprot_t;
-typedef struct page *pgtable_t;
-
-#define pte_val(x)      ((x).pte)
-#define pmd_val(x)      ((x).pmd)
-#define pgd_val(x)      ((x).pgd)
-#define pgprot_val(x)   ((x).pgprot)
-
-#define __pte(x)        ((pte_t) { (x) } )
-#define __pmd(x)        ((pmd_t) { (x) } )
-#define __pgd(x)        ((pgd_t) { (x) } )
-#define __pgprot(x)     ((pgprot_t) { (x) } )
-
-#else /* !STRICT_MM_TYPECHECKS */
-/*
- * .. while these make it easier on the compiler
- */
-
-typedef unsigned long pte_t;
-typedef unsigned long pmd_t;
-typedef unsigned long pgd_t;
-typedef unsigned long pgprot_t;
-
-#define pte_val(x)      (x)
-#define pmd_val(x)      (x)
-#define pgd_val(x)      (x)
-#define pgprot_val(x)   (x)
-
-#define __pte(x)        (x)
-#define __pmd(x)        (x)
-#define __pgd(x)        (x)
-#define __pgprot(x)     (x)
-
-#endif /* STRICT_MM_TYPECHECKS */
-
-#endif /* !__ASSEMBLY__ */
-
-
-/* No current v850 processor has virtual memory.  */
-#define __virt_to_phys(addr)	(addr)
-#define __phys_to_virt(addr)	(addr)
-
-#define virt_to_pfn(kaddr)	(__virt_to_phys (kaddr) >> PAGE_SHIFT)
-#define pfn_to_virt(pfn)	__phys_to_virt ((pfn) << PAGE_SHIFT)
-
-#define MAP_NR(kaddr) \
-  (((unsigned long)(kaddr) - PAGE_OFFSET) >> PAGE_SHIFT)
-#define virt_to_page(kaddr)	(mem_map + MAP_NR (kaddr))
-#define page_to_virt(page) \
-  ((((page) - mem_map) << PAGE_SHIFT) + PAGE_OFFSET)
-
-#define ARCH_PFN_OFFSET		(PAGE_OFFSET >> PAGE_SHIFT)
-#define pfn_valid(pfn)	        ((pfn) < max_mapnr)
-
-#define	virt_addr_valid(kaddr)						\
-  (((void *)(kaddr) >= (void *)PAGE_OFFSET) && MAP_NR (kaddr) < max_mapnr)
-
-
-#define __pa(x)		     __virt_to_phys ((unsigned long)(x))
-#define __va(x)		     ((void *)__phys_to_virt ((unsigned long)(x)))
-
-
-#include <asm-generic/memory_model.h>
-#include <asm-generic/page.h>
-
-#endif /* __V850_PAGE_H__ */
diff --git a/include/asm-v850/param.h b/include/asm-v850/param.h
deleted file mode 100644
index 4391f5fe0204..000000000000
--- a/include/asm-v850/param.h
+++ /dev/null
@@ -1,33 +0,0 @@
-/*
- * include/asm-v850/param.h -- Varions kernel parameters
- *
- *  Copyright (C) 2001,02  NEC Corporation
- *  Copyright (C) 2001,02  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_PARAM_H__
-#define __V850_PARAM_H__
-
-#define EXEC_PAGESIZE	4096
-
-#ifndef NOGROUP
-#define NOGROUP		(-1)
-#endif
-
-#define MAXHOSTNAMELEN	64	/* max length of hostname */
-
-#ifdef __KERNEL__
-# define HZ		CONFIG_HZ
-# define USER_HZ	100
-# define CLOCKS_PER_SEC	USER_HZ
-#else
-# define HZ		100
-#endif
-
-#endif /* __V850_PARAM_H__ */
diff --git a/include/asm-v850/pci.h b/include/asm-v850/pci.h
deleted file mode 100644
index de2a7d0a81cc..000000000000
--- a/include/asm-v850/pci.h
+++ /dev/null
@@ -1,119 +0,0 @@
-/*
- * include/asm-v850/pci.h -- PCI support
- *
- *  Copyright (C) 2001,02,05  NEC Corporation
- *  Copyright (C) 2001,02,05  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_PCI_H__
-#define __V850_PCI_H__
-
-/* Get any platform-dependent definitions.  */
-#include <asm/machdep.h>
-
-#define pcibios_scan_all_fns(a, b)	0
-
-/* Generic declarations.  */
-
-struct scatterlist;
-
-extern void pcibios_set_master (struct pci_dev *dev);
-
-/* `Grant' to PDEV the memory block at CPU_ADDR, for doing DMA.  The
-   32-bit PCI bus mastering address to use is returned.  the device owns
-   this memory until either pci_unmap_single or pci_dma_sync_single_for_cpu is
-   performed.  */
-extern dma_addr_t
-pci_map_single (struct pci_dev *pdev, void *cpu_addr, size_t size, int dir);
-
-/* Return to the CPU the PCI DMA memory block previously `granted' to
-   PDEV, at DMA_ADDR.  */
-extern void
-pci_unmap_single (struct pci_dev *pdev, dma_addr_t dma_addr, size_t size,
-		  int dir);
-
-/* Make physical memory consistent for a single streaming mode DMA
-   translation after a transfer.
-
-   If you perform a pci_map_single() but wish to interrogate the
-   buffer using the cpu, yet do not wish to teardown the PCI dma
-   mapping, you must call this function before doing so.  At the next
-   point you give the PCI dma address back to the card, you must first
-   perform a pci_dma_sync_for_device, and then the device again owns
-   the buffer.  */
-extern void
-pci_dma_sync_single_for_cpu (struct pci_dev *dev, dma_addr_t dma_addr,
-			     size_t size, int dir);
-
-extern void
-pci_dma_sync_single_for_device (struct pci_dev *dev, dma_addr_t dma_addr,
-				size_t size, int dir);
-
-
-/* Do multiple DMA mappings at once.  */
-extern int
-pci_map_sg (struct pci_dev *pdev, struct scatterlist *sg, int sg_len, int dir);
-
-/* Unmap multiple DMA mappings at once.  */
-extern void
-pci_unmap_sg (struct pci_dev *pdev, struct scatterlist *sg, int sg_len,
-	      int dir);
-
-/* SG-list versions of pci_dma_sync functions.  */
-extern void
-pci_dma_sync_sg_for_cpu (struct pci_dev *dev,
-			 struct scatterlist *sg, int sg_len,
-			 int dir);
-extern void
-pci_dma_sync_sg_for_device (struct pci_dev *dev,
-			    struct scatterlist *sg, int sg_len,
-			    int dir);
-
-#define pci_map_page(dev, page, offs, size, dir) \
-  pci_map_single(dev, (page_address(page) + (offs)), size, dir)
-#define pci_unmap_page(dev,addr,sz,dir) \
-  pci_unmap_single(dev, addr, sz, dir)
-
-/* Test for pci_map_single or pci_map_page having generated an error.  */
-static inline int
-pci_dma_mapping_error (dma_addr_t dma_addr)
-{
-	return dma_addr == 0;
-}
-
-/* Allocate and map kernel buffer using consistent mode DMA for PCI
-   device.  Returns non-NULL cpu-view pointer to the buffer if
-   successful and sets *DMA_ADDR to the pci side dma address as well,
-   else DMA_ADDR is undefined.  */
-extern void *
-pci_alloc_consistent (struct pci_dev *pdev, size_t size, dma_addr_t *dma_addr);
-
-/* Free and unmap a consistent DMA buffer.  CPU_ADDR and DMA_ADDR must
-   be values that were returned from pci_alloc_consistent.  SIZE must be
-   the same as what as passed into pci_alloc_consistent.  References to
-   the memory and mappings assosciated with CPU_ADDR or DMA_ADDR past
-   this call are illegal.  */
-extern void
-pci_free_consistent (struct pci_dev *pdev, size_t size, void *cpu_addr,
-		     dma_addr_t dma_addr);
-
-#ifdef CONFIG_PCI
-static inline void pci_dma_burst_advice(struct pci_dev *pdev,
-					enum pci_dma_burst_strategy *strat,
-					unsigned long *strategy_parameter)
-{
-	*strat = PCI_DMA_BURST_INFINITY;
-	*strategy_parameter = ~0UL;
-}
-#endif
-
-extern void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long max);
-extern void pci_iounmap (struct pci_dev *dev, void __iomem *addr);
-
-#endif /* __V850_PCI_H__ */
diff --git a/include/asm-v850/percpu.h b/include/asm-v850/percpu.h
deleted file mode 100644
index 755ac6522b63..000000000000
--- a/include/asm-v850/percpu.h
+++ /dev/null
@@ -1,14 +0,0 @@
-#ifndef __V850_PERCPU_H__
-#define __V850_PERCPU_H__
-
-#include <asm-generic/percpu.h>
-
-/* This is a stupid hack to satisfy some grotty implicit include-file
-   dependency; basically, <linux/smp.h> uses BUG_ON, which calls BUG, but
-   doesn't include the necessary headers to define it.  In the twisted
-   festering mess of includes this must all be resolved somehow on other
-   platforms, but I haven't the faintest idea how, and don't care; here will
-   do, even though doesn't actually make any sense.  */
-#include <asm/page.h>
-
-#endif /* __V850_PERCPU_H__ */
diff --git a/include/asm-v850/pgalloc.h b/include/asm-v850/pgalloc.h
deleted file mode 100644
index b91eb2d02bfd..000000000000
--- a/include/asm-v850/pgalloc.h
+++ /dev/null
@@ -1,22 +0,0 @@
-/*
- * include/asm-v850/pgalloc.h
- *
- *  Copyright (C) 2001,02  NEC Corporation
- *  Copyright (C) 2001,02  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_PGALLOC_H__
-#define __V850_PGALLOC_H__
-
-#include <linux/mm.h>  /* some crap code expects this */
-
-/* ... and then, there was one.  */
-#define check_pgt_cache()	((void)0)
-
-#endif /* __V850_PGALLOC_H__ */
diff --git a/include/asm-v850/pgtable.h b/include/asm-v850/pgtable.h
deleted file mode 100644
index 1ea2a900f0f8..000000000000
--- a/include/asm-v850/pgtable.h
+++ /dev/null
@@ -1,59 +0,0 @@
-#ifndef __V850_PGTABLE_H__
-#define __V850_PGTABLE_H__
-
-#include <asm-generic/4level-fixup.h>
-
-#include <asm/page.h>
-
-
-#define pgd_present(pgd)	(1) /* pages are always present on NO_MM */
-#define pgd_none(pgd)		(0)
-#define pgd_bad(pgd)		(0)
-#define pgd_clear(pgdp)		((void)0)
-
-#define	pmd_offset(a, b)	((void *)0)
-
-#define kern_addr_valid(addr)	(1)
-
-
-#define __swp_type(x)		(0)
-#define __swp_offset(x)		(0)
-#define __swp_entry(typ,off)	((swp_entry_t) { ((typ) | ((off) << 7)) })
-#define __pte_to_swp_entry(pte)	((swp_entry_t) { pte_val(pte) })
-#define __swp_entry_to_pte(x)	((pte_t) { (x).val })
-
-static inline int pte_file (pte_t pte) { return 0; }
-
-
-/* These mean nothing to !CONFIG_MMU.  */
-#define PAGE_NONE		__pgprot(0)
-#define PAGE_SHARED		__pgprot(0)
-#define PAGE_COPY		__pgprot(0)
-#define PAGE_READONLY		__pgprot(0)
-#define PAGE_KERNEL		__pgprot(0)
-
-
-/*
- * ZERO_PAGE is a global shared page that is always zero: used
- * for zero-mapped memory areas etc.  When CONFIG_MMU is not defined, this
- * should never actually be used, so just define it to something that's
- * will hopefully cause a bus error if it is.
- */
-#define ZERO_PAGE(vaddr)	((void *)0x87654321)
-
-
-/* Some bogus code in procfs uses these; whatever.  */
-#define VMALLOC_START	0
-#define VMALLOC_END	(~0)
-
-
-extern void paging_init (void);
-#define swapper_pg_dir ((pgd_t *) 0)
-
-#define pgtable_cache_init()   ((void)0)
-
-
-extern unsigned int kobjsize(const void *objp);
-
-
-#endif /* __V850_PGTABLE_H__ */
diff --git a/include/asm-v850/poll.h b/include/asm-v850/poll.h
deleted file mode 100644
index 803cad0b9b59..000000000000
--- a/include/asm-v850/poll.h
+++ /dev/null
@@ -1,9 +0,0 @@
-#ifndef __V850_POLL_H__
-#define __V850_POLL_H__
-
-#define POLLWRNORM	POLLOUT
-#define POLLWRBAND	0x0100
-
-#include <asm-generic/poll.h>
-
-#endif /* __V850_POLL_H__ */
diff --git a/include/asm-v850/posix_types.h b/include/asm-v850/posix_types.h
deleted file mode 100644
index 7f403b765390..000000000000
--- a/include/asm-v850/posix_types.h
+++ /dev/null
@@ -1,72 +0,0 @@
-/*
- * include/asm-v850/posix_types.h -- Kernel versions of standard types
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_POSIX_TYPES_H__
-#define __V850_POSIX_TYPES_H__
-
-typedef unsigned long	__kernel_ino_t;
-typedef unsigned long long __kernel_ino64_t;
-typedef unsigned int	__kernel_mode_t;
-typedef unsigned int	__kernel_nlink_t;
-typedef long		__kernel_off_t;
-typedef long long	__kernel_loff_t;
-typedef int		__kernel_pid_t;
-typedef unsigned short	__kernel_ipc_pid_t;
-typedef unsigned int	__kernel_uid_t;
-typedef unsigned int	__kernel_gid_t;
-typedef unsigned int	__kernel_size_t;
-typedef int		__kernel_ssize_t;
-typedef int		__kernel_ptrdiff_t;
-typedef long		__kernel_time_t;
-typedef long		__kernel_suseconds_t;
-typedef long		__kernel_clock_t;
-typedef int		__kernel_timer_t;
-typedef int		__kernel_clockid_t;
-typedef int		__kernel_daddr_t;
-typedef char *		__kernel_caddr_t;
-typedef unsigned short	__kernel_uid16_t;
-typedef unsigned short	__kernel_gid16_t;
-typedef unsigned int	__kernel_uid32_t;
-typedef unsigned int	__kernel_gid32_t;
-
-/* Some bogus code depends on this; we don't care.  */
-typedef __kernel_uid_t __kernel_old_uid_t;
-typedef unsigned int	__kernel_old_dev_t;
-
-typedef struct {
-	int	val[2];
-} __kernel_fsid_t;
-
-
-#if defined(__KERNEL__)
-
-/* We used to include <asm/bitops.h> here, which seems the right thing, but
-   it caused nasty include-file definition order problems.  Removing the
-   include seems to work, so fingers crossed...  */
-
-#undef	__FD_SET
-#define __FD_SET(fd, fd_set) \
-  __set_bit (fd, (void *)&((__kernel_fd_set *)fd_set)->fds_bits)
-#undef __FD_CLR
-#define __FD_CLR(fd, fd_set) \
-  __clear_bit (fd, (void *)&((__kernel_fd_set *)fd_set)->fds_bits)
-#undef	__FD_ISSET
-#define __FD_ISSET(fd, fd_set) \
-  __test_bit (fd, (void *)&((__kernel_fd_set *)fd_set)->fds_bits)
-#undef	__FD_ZERO
-#define __FD_ZERO(fd_set) \
-  memset (fd_set, 0, sizeof (*(fd_set *)fd_set))
-
-#endif /* defined(__KERNEL__) */
-
-#endif /* __V850_POSIX_TYPES_H__ */
diff --git a/include/asm-v850/processor.h b/include/asm-v850/processor.h
deleted file mode 100644
index 979e3467f9af..000000000000
--- a/include/asm-v850/processor.h
+++ /dev/null
@@ -1,120 +0,0 @@
-/*
- * include/asm-v850/processor.h
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_PROCESSOR_H__
-#define __V850_PROCESSOR_H__
-
-#ifndef __ASSEMBLY__ /* <linux/thread_info.h> is not asm-safe.  */
-#include <linux/thread_info.h>
-#endif
-
-#include <linux/compiler.h>
-#include <asm/ptrace.h>
-#include <asm/entry.h>
-
-/* Some code expects `segment' stuff to be defined here.  */
-#include <asm/segment.h>
-
-
-/*
- * The only places this is used seem to be horrible bletcherous kludges,
- * so we just define it to be as large as possible.
- */
-#define TASK_SIZE	(0xFFFFFFFF)
-
-/*
- * This decides where the kernel will search for a free chunk of vm
- * space during mmap's.  We won't be using it.
- */
-#define TASK_UNMAPPED_BASE	0
-
-
-#ifndef __ASSEMBLY__
-
-
-/*
- * Default implementation of macro that returns current
- * instruction pointer ("program counter").
- */
-#define current_text_addr()	({ __label__ _l; _l: &&_l;})
-
-/* If you change this, you must change the associated assembly-languages
-   constants defined below, THREAD_*.  */
-struct thread_struct {
-	/* kernel stack pointer (must be first field in structure) */
-	unsigned long  ksp;
-};
-
-#define INIT_THREAD { sizeof init_stack + (unsigned long)init_stack }
-
-
-/* Do necessary setup to start up a newly executed thread.  */
-static inline void start_thread (struct pt_regs *regs,
-				 unsigned long pc, unsigned long usp)
-{
-	regs->pc = pc;
-	regs->gpr[GPR_SP] = usp;
-	regs->kernel_mode = 0;
-}
-
-/* Free all resources held by a thread. */
-static inline void release_thread (struct task_struct *dead_task)
-{
-}
-
-/* Prepare to copy thread state - unlazy all lazy status */
-#define prepare_to_copy(tsk)	do { } while (0)
-
-extern int kernel_thread (int (*fn)(void *), void * arg, unsigned long flags);
-
-/* Free current thread data structures etc.  */
-static inline void exit_thread (void)
-{
-}
-
-
-/* Return the registers saved during context-switch by the currently
-   not-running thread T.  Note that this only includes some registers!
-   See entry.S for details.  */
-#define thread_saved_regs(t) \
-   ((struct pt_regs*)((t)->thread.ksp + STATE_SAVE_PT_OFFSET))
-/* Return saved (kernel) PC of a blocked thread.  Actually, we return the
-   LP register, because the thread is actually blocked in switch_thread,
-   and we're interested in the PC it will _return_ to.  */
-#define thread_saved_pc(t)   (thread_saved_regs(t)->gpr[GPR_LP])
-
-
-unsigned long get_wchan (struct task_struct *p);
-
-
-/* Return some info about the user process TASK.  */
-#define task_tos(task)	((unsigned long)task_stack_page(task) + THREAD_SIZE)
-#define task_pt_regs(task) ((struct pt_regs *)task_tos (task) - 1)
-#define task_sp(task)	(task_pt_regs (task)->gpr[GPR_SP])
-#define task_pc(task)	(task_pt_regs (task)->pc)
-/* Grotty old names for some.  */
-#define KSTK_EIP(task)	task_pc (task)
-#define KSTK_ESP(task)	task_sp (task)
-
-
-#define cpu_relax()    barrier()
-
-
-#else /* __ASSEMBLY__ */
-
-#define THREAD_KSP	0
-
-#endif /* !__ASSEMBLY__ */
-
-
-#endif /* __V850_PROCESSOR_H__ */
diff --git a/include/asm-v850/ptrace.h b/include/asm-v850/ptrace.h
deleted file mode 100644
index 4f35cf2cd641..000000000000
--- a/include/asm-v850/ptrace.h
+++ /dev/null
@@ -1,121 +0,0 @@
-/*
- * include/asm-v850/ptrace.h -- Access to CPU registers
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_PTRACE_H__
-#define __V850_PTRACE_H__
-
-
-/* v850 general purpose registers with special meanings.  */
-#define GPR_ZERO	0	/* constant zero */
-#define GPR_ASM		1	/* reserved for assembler */
-#define GPR_SP		3	/* stack pointer */
-#define GPR_GP		4	/* global data pointer */
-#define GPR_TP		5	/* `text pointer' */
-#define GPR_EP		30	/* `element pointer' */
-#define GPR_LP		31	/* link pointer (current return address) */
-
-/* These aren't official names, but they make some code more descriptive.  */
-#define GPR_ARG0	6
-#define GPR_ARG1	7
-#define GPR_ARG2	8
-#define GPR_ARG3	9
-#define GPR_RVAL0	10
-#define GPR_RVAL1	11
-#define GPR_RVAL	GPR_RVAL0
-
-#define NUM_GPRS	32
-
-/* v850 `system' registers.  */
-#define SR_EIPC		0
-#define SR_EIPSW	1
-#define SR_FEPC		2
-#define SR_FEPSW	3
-#define SR_ECR		4
-#define SR_PSW		5
-#define SR_CTPC		16
-#define SR_CTPSW	17
-#define SR_DBPC		18
-#define SR_DBPSW	19
-#define SR_CTBP		20
-#define SR_DIR		21
-#define SR_ASID		23
-
-
-#ifndef __ASSEMBLY__
-
-typedef unsigned long v850_reg_t;
-
-/* How processor state is stored on the stack during a syscall/signal.
-   If you change this structure, change the associated assembly-language
-   macros below too (PT_*)!  */
-struct pt_regs
-{
-	/* General purpose registers.  */
-	v850_reg_t gpr[NUM_GPRS];
-
-	v850_reg_t pc;		/* program counter */
-	v850_reg_t psw;		/* program status word */
-
-	/* Registers used by `callt' instruction:  */
-	v850_reg_t ctpc;	/* saved program counter */
-	v850_reg_t ctpsw;	/* saved psw */
-	v850_reg_t ctbp;	/* base pointer for callt table */
-
-	char kernel_mode;	/* 1 if in `kernel mode', 0 if user mode */
-};
-
-
-#define instruction_pointer(regs)	((regs)->pc)
-#define profile_pc(regs) instruction_pointer(regs)
-#define user_mode(regs)			(!(regs)->kernel_mode)
-
-/* When a struct pt_regs is used to save user state for a system call in
-   the kernel, the system call is stored in the space for R0 (since it's
-   never used otherwise, R0 being a constant 0).  Non-system-calls
-   simply store 0 there.  */
-#define PT_REGS_SYSCALL(regs)		(regs)->gpr[0]
-#define PT_REGS_SET_SYSCALL(regs, val)	((regs)->gpr[0] = (val))
-
-#endif /* !__ASSEMBLY__ */
-
-
-/* The number of bytes used to store each register.  */
-#define _PT_REG_SIZE	4
-
-/* Offset of a general purpose register in a struct pt_regs.  */
-#define PT_GPR(num)	((num) * _PT_REG_SIZE)
-
-/* Offsets of various special registers & fields in a struct pt_regs.  */
-#define PT_PC		((NUM_GPRS + 0) * _PT_REG_SIZE)
-#define PT_PSW		((NUM_GPRS + 1) * _PT_REG_SIZE)
-#define PT_CTPC		((NUM_GPRS + 2) * _PT_REG_SIZE)
-#define PT_CTPSW	((NUM_GPRS + 3) * _PT_REG_SIZE)
-#define PT_CTBP		((NUM_GPRS + 4) * _PT_REG_SIZE)
-#define PT_KERNEL_MODE	((NUM_GPRS + 5) * _PT_REG_SIZE)
-
-/* Where the current syscall number is stashed; obviously only valid in
-   the kernel!  */
-#define PT_CUR_SYSCALL	PT_GPR(0)
-
-/* Size of struct pt_regs, including alignment.  */
-#define PT_SIZE		((NUM_GPRS + 6) * _PT_REG_SIZE)
-
-
-/* These are `magic' values for PTRACE_PEEKUSR that return info about where
-   a process is located in memory.  */
-#define PT_TEXT_ADDR	(PT_SIZE + 1)
-#define PT_TEXT_LEN	(PT_SIZE + 2)
-#define PT_DATA_ADDR	(PT_SIZE + 3)
-
-
-#endif /* __V850_PTRACE_H__ */
diff --git a/include/asm-v850/resource.h b/include/asm-v850/resource.h
deleted file mode 100644
index 4b9dcd44f8d1..000000000000
--- a/include/asm-v850/resource.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __V850_RESOURCE_H__
-#define __V850_RESOURCE_H__
-
-#include <asm-generic/resource.h>
-
-#endif /* __V850_RESOURCE_H__ */
diff --git a/include/asm-v850/rte_cb.h b/include/asm-v850/rte_cb.h
deleted file mode 100644
index db9879f00aa7..000000000000
--- a/include/asm-v850/rte_cb.h
+++ /dev/null
@@ -1,78 +0,0 @@
-/*
- * include/asm-v850/rte_cb.h -- Midas labs RTE-CB series of evaluation boards
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_RTE_CB_H__
-#define __V850_RTE_CB_H__
-
-
-/* The SRAM on the Mother-A motherboard.  */
-#define MB_A_SRAM_ADDR		GCS0_ADDR
-#define MB_A_SRAM_SIZE		0x00200000 /* 2MB */
-
-
-#ifdef CONFIG_RTE_GBUS_INT
-/* GBUS interrupt support.  */
-
-# include <asm/gbus_int.h>
-
-# define GBUS_INT_BASE_IRQ	NUM_RTE_CB_IRQS
-# define GBUS_INT_BASE_ADDR	(GCS2_ADDR + 0x00006000)
-
-/* Some specific interrupts.  */
-# define IRQ_MB_A_LAN		IRQ_GBUS_INT(10)
-# define IRQ_MB_A_PCI1(n)	(IRQ_GBUS_INT(16) + (n))
-# define IRQ_MB_A_PCI1_NUM	4
-# define IRQ_MB_A_PCI2(n)	(IRQ_GBUS_INT(20) + (n))
-# define IRQ_MB_A_PCI2_NUM	4
-# define IRQ_MB_A_EXT(n)	(IRQ_GBUS_INT(24) + (n))
-# define IRQ_MB_A_EXT_NUM	4
-# define IRQ_MB_A_USB_OC(n)	(IRQ_GBUS_INT(28) + (n))
-# define IRQ_MB_A_USB_OC_NUM	2
-# define IRQ_MB_A_PCMCIA_OC	IRQ_GBUS_INT(30)
-
-/* We define NUM_MACH_IRQS to include extra interrupts from the GBUS.  */
-# define NUM_MACH_IRQS		(NUM_RTE_CB_IRQS + IRQ_GBUS_INT_NUM)
-
-#else /* !CONFIG_RTE_GBUS_INT */
-
-# define NUM_MACH_IRQS		NUM_RTE_CB_IRQS
-
-#endif /* CONFIG_RTE_GBUS_INT */
-
-
-#ifdef CONFIG_RTE_MB_A_PCI
-/* Mother-A PCI bus support.  */
-
-# include <asm/rte_mb_a_pci.h>
-
-/* These are the base addresses used for allocating device address
-   space.  512K of the motherboard SRAM is in the same space, so we have
-   to be careful not to let it be allocated.  */
-# define PCIBIOS_MIN_MEM	(MB_A_PCI_MEM_ADDR + 0x80000)
-# define PCIBIOS_MIN_IO		MB_A_PCI_IO_ADDR
-
-/* As we don't really support PCI DMA to cpu memory, and use bounce-buffers
-   instead, perversely enough, this becomes always true! */
-# define pci_dma_supported(dev, mask)		1
-# define pcibios_assign_all_busses()		1
-
-#endif /* CONFIG_RTE_MB_A_PCI */
-
-
-#ifndef __ASSEMBLY__
-extern void rte_cb_early_init (void);
-extern void rte_cb_init_irqs (void);
-#endif /* !__ASSEMBLY__ */
-
-
-#endif /* __V850_RTE_CB_H__ */
diff --git a/include/asm-v850/rte_ma1_cb.h b/include/asm-v850/rte_ma1_cb.h
deleted file mode 100644
index bd3162ab9844..000000000000
--- a/include/asm-v850/rte_ma1_cb.h
+++ /dev/null
@@ -1,128 +0,0 @@
-/*
- * include/asm-v850/rte_ma1_cb.h -- Midas labs RTE-V850/MA1-CB board
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_RTE_MA1_CB_H__
-#define __V850_RTE_MA1_CB_H__
-
-#include <asm/rte_cb.h>		/* Common defs for Midas RTE-CB boards.  */
-
-
-#define PLATFORM		"rte-v850e/ma1-cb"
-#define PLATFORM_LONG		"Midas lab RTE-V850E/MA1-CB"
-
-#define CPU_CLOCK_FREQ		50000000 /* 50MHz */
-
-/* 1MB of onboard SRAM.  Note that the monitor ROM uses parts of this
-   for its own purposes, so care must be taken.  Some address lines are
-   not decoded, so the SRAM area is mirrored every 1MB from 0x400000 to
-   0x800000 (exclusive).  */
-#define SRAM_ADDR		0x00400000
-#define SRAM_SIZE		0x00100000 /* 1MB */
-
-/* 32MB of onbard SDRAM.  */
-#define SDRAM_ADDR		0x00800000
-#define SDRAM_SIZE		0x02000000 /* 32MB */
-
-
-/* CPU addresses of GBUS memory spaces.  */
-#define GCS0_ADDR		0x05000000 /* GCS0 - Common SRAM (2MB) */
-#define GCS0_SIZE		0x00200000 /*   2MB */
-#define GCS1_ADDR		0x06000000 /* GCS1 - Flash ROM (8MB) */
-#define GCS1_SIZE		0x00800000 /*   8MB */
-#define GCS2_ADDR		0x07900000 /* GCS2 - I/O registers */
-#define GCS2_SIZE		0x00400000 /*   4MB */
-#define GCS5_ADDR		0x04000000 /* GCS5 - PCI bus space */
-#define GCS5_SIZE		0x01000000 /*   16MB */
-#define GCS6_ADDR		0x07980000 /* GCS6 - PCI control registers */
-#define GCS6_SIZE		0x00000200 /*   512B */
-
-
-/* For <asm/page.h> */
-#define PAGE_OFFSET 		SRAM_ADDR
-
-
-/* The GBUS GINT0 - GINT3 interrupts are connected to the INTP000 - INTP011
-   pins on the CPU.  These are shared among the GBUS interrupts.  */
-#define IRQ_GINT(n)		IRQ_INTP(n)
-#define IRQ_GINT_NUM		4
-
-/* Used by <asm/rte_cb.h> to derive NUM_MACH_IRQS.  */
-#define NUM_RTE_CB_IRQS		NUM_CPU_IRQS
-
-
-#ifdef CONFIG_ROM_KERNEL
-/* Kernel is in ROM, starting at address 0.  */
-
-#define INTV_BASE		0
-
-#else /* !CONFIG_ROM_KERNEL */
-
-#ifdef CONFIG_RTE_CB_MULTI
-/* Using RAM kernel with ROM monitor for Multi debugger.  */
-
-/* The chip's real interrupt vectors are in ROM, but they jump to a
-   secondary interrupt vector table in RAM.  */
-#define INTV_BASE		0x004F8000
-
-/* Scratch memory used by the ROM monitor, which shouldn't be used by
-   linux (except for the alternate interrupt vector area, defined
-   above).  */
-#define MON_SCRATCH_ADDR	0x004F8000
-#define MON_SCRATCH_SIZE	0x00008000 /* 32KB */
-
-#else /* !CONFIG_RTE_CB_MULTI */
-/* Using RAM-kernel.  Assume some sort of boot-loader got us loaded at
-   address 0.  */
-
-#define INTV_BASE		0
-
-#endif /* CONFIG_RTE_CB_MULTI */
-
-#endif /* CONFIG_ROM_KERNEL */
-
-
-/* Some misc. on-board devices.  */
-
-/* Seven-segment LED display (two digits).  Write-only.  */
-#define LED_ADDR(n)		(0x07802000 + (n))
-#define LED(n)			(*(volatile unsigned char *)LED_ADDR(n))
-#define LED_NUM_DIGITS		2
-
-
-/* Override the basic MA uart pre-initialization so that we can
-   initialize extra stuff.  */
-#undef V850E_UART_PRE_CONFIGURE	/* should be defined by <asm/ma.h> */
-#define V850E_UART_PRE_CONFIGURE	rte_ma1_cb_uart_pre_configure
-#ifndef __ASSEMBLY__
-extern void rte_ma1_cb_uart_pre_configure (unsigned chan,
-					   unsigned cflags, unsigned baud);
-#endif
-
-/* This board supports RTS/CTS for the on-chip UART, but only for channel 0. */
-
-/* CTS for UART channel 0 is pin P43 (bit 3 of port 4).  */
-#define V850E_UART_CTS(chan)	((chan) == 0 ? !(MA_PORT4_IO & 0x8) : 1)
-/* RTS for UART channel 0 is pin P42 (bit 2 of port 4).  */
-#define V850E_UART_SET_RTS(chan, val)					      \
-   do {									      \
-	   if (chan == 0) {						      \
-		   unsigned old = MA_PORT4_IO; 				      \
-		   if (val)						      \
-			   MA_PORT4_IO = old & ~0x4;			      \
-		   else							      \
-			   MA_PORT4_IO = old | 0x4;			      \
-	   }								      \
-   } while (0)
-
-
-#endif /* __V850_RTE_MA1_CB_H__ */
diff --git a/include/asm-v850/rte_mb_a_pci.h b/include/asm-v850/rte_mb_a_pci.h
deleted file mode 100644
index 41ac185ca9cd..000000000000
--- a/include/asm-v850/rte_mb_a_pci.h
+++ /dev/null
@@ -1,56 +0,0 @@
-/*
- * include/asm-v850/mb_a_pci.h -- PCI support for Midas lab RTE-MOTHER-A board
- *
- *  Copyright (C) 2001  NEC Corporation
- *  Copyright (C) 2001  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_MB_A_PCI_H__
-#define __V850_MB_A_PCI_H__
-
-
-#define MB_A_PCI_MEM_ADDR	GCS5_ADDR
-#define MB_A_PCI_MEM_SIZE	(GCS5_SIZE / 2)
-#define MB_A_PCI_IO_ADDR	(GCS5_ADDR + MB_A_PCI_MEM_SIZE)
-#define MB_A_PCI_IO_SIZE	(GCS5_SIZE / 2)
-#define MB_A_PCI_REG_BASE_ADDR	GCS6_ADDR
-
-#define MB_A_PCI_PCICR_ADDR	(MB_A_PCI_REG_BASE_ADDR + 0x4)
-#define MB_A_PCI_PCICR		(*(volatile u16 *)MB_A_PCI_PCICR_ADDR)
-#define MB_A_PCI_PCISR_ADDR	(MB_A_PCI_REG_BASE_ADDR + 0x6)
-#define MB_A_PCI_PCISR		(*(volatile u16 *)MB_A_PCI_PCISR_ADDR)
-#define MB_A_PCI_PCILTR_ADDR	(MB_A_PCI_REG_BASE_ADDR + 0xD)
-#define MB_A_PCI_PCILTR		(*(volatile u8 *)MB_A_PCI_PCILTR_ADDR)
-#define MB_A_PCI_PCIBAR0_ADDR	(MB_A_PCI_REG_BASE_ADDR + 0x10)
-#define MB_A_PCI_PCIBAR0	(*(volatile u32 *)MB_A_PCI_PCIBAR0_ADDR)
-#define MB_A_PCI_PCIBAR1_ADDR	(MB_A_PCI_REG_BASE_ADDR + 0x14)
-#define MB_A_PCI_PCIBAR1	(*(volatile u32 *)MB_A_PCI_PCIBAR1_ADDR)
-#define MB_A_PCI_PCIBAR2_ADDR	(MB_A_PCI_REG_BASE_ADDR + 0x18)
-#define MB_A_PCI_PCIBAR2	(*(volatile u32 *)MB_A_PCI_PCIBAR2_ADDR)
-#define MB_A_PCI_VENDOR_ID_ADDR	(MB_A_PCI_REG_BASE_ADDR + 0x2C)
-#define MB_A_PCI_VENDOR_ID	(*(volatile u16 *)MB_A_PCI_VENDOR_ID_ADDR)
-#define MB_A_PCI_DEVICE_ID_ADDR	(MB_A_PCI_REG_BASE_ADDR + 0x2E)
-#define MB_A_PCI_DEVICE_ID	(*(volatile u16 *)MB_A_PCI_DEVICE_ID_ADDR)
-#define MB_A_PCI_DMRR_ADDR	(MB_A_PCI_REG_BASE_ADDR + 0x9C)
-#define MB_A_PCI_DMRR		(*(volatile u32 *)MB_A_PCI_DMRR_ADDR)
-#define MB_A_PCI_DMLBAM_ADDR	(MB_A_PCI_REG_BASE_ADDR + 0xA0)
-#define MB_A_PCI_DMLBAM		(*(volatile u32 *)MB_A_PCI_DMLBAM_ADDR)
-#define MB_A_PCI_DMLBAI_ADDR	(MB_A_PCI_REG_BASE_ADDR + 0xA4)
-#define MB_A_PCI_DMLBAI		(*(volatile u32 *)MB_A_PCI_DMLBAI_ADDR)
-#define MB_A_PCI_PCIPBAM_ADDR	(MB_A_PCI_REG_BASE_ADDR + 0xA8)
-#define MB_A_PCI_PCIPBAM	(*(volatile u32 *)MB_A_PCI_PCIPBAM_ADDR)
-/* `PCI Configuration Address Register for Direct Master to PCI IO/CFG'  */
-#define MB_A_PCI_DMCFGA_ADDR	(MB_A_PCI_REG_BASE_ADDR + 0xAC)
-#define MB_A_PCI_DMCFGA		(*(volatile u32 *)MB_A_PCI_DMCFGA_ADDR)
-/* `PCI Permanent Configuration ID Register'  */
-#define MB_A_PCI_PCIHIDR_ADDR	(MB_A_PCI_REG_BASE_ADDR + 0xF0)
-#define MB_A_PCI_PCIHIDR	(*(volatile u32 *)MB_A_PCI_PCIHIDR_ADDR)
-
-
-#endif /* __V850_MB_A_PCI_H__ */
diff --git a/include/asm-v850/rte_me2_cb.h b/include/asm-v850/rte_me2_cb.h
deleted file mode 100644
index 9922c85c85a8..000000000000
--- a/include/asm-v850/rte_me2_cb.h
+++ /dev/null
@@ -1,202 +0,0 @@
-/*
- * include/asm-v850/rte_me2_cb.h -- Midas labs RTE-V850E/ME2-CB board
- *
- *  Copyright (C) 2001,02,03  NEC Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_RTE_ME2_CB_H__
-#define __V850_RTE_ME2_CB_H__
-
-#include <asm/rte_cb.h>		/* Common defs for Midas RTE-CB boards.  */
-
-
-#define PLATFORM		"rte-v850e/me2-cb"
-#define PLATFORM_LONG		"Midas lab RTE-V850E/ME2-CB"
-
-#define CPU_CLOCK_FREQ		150000000 /* 150MHz */
-#define FIXED_BOGOMIPS		50
-
-/* 32MB of onbard SDRAM.  */
-#define SDRAM_ADDR		0x00800000
-#define SDRAM_SIZE		0x02000000 /* 32MB */
-
-
-/* CPU addresses of GBUS memory spaces.  */
-#define GCS0_ADDR		0x04000000 /* GCS0 - Common SRAM (2MB) */
-#define GCS0_SIZE		0x00800000 /*   8MB */
-#define GCS1_ADDR		0x04800000 /* GCS1 - Flash ROM (8MB) */
-#define GCS1_SIZE		0x00800000 /*   8MB */
-#define GCS2_ADDR		0x07000000 /* GCS2 - I/O registers */
-#define GCS2_SIZE		0x00800000 /*   8MB */
-#define GCS5_ADDR		0x08000000 /* GCS5 - PCI bus space */
-#define GCS5_SIZE		0x02000000 /*   32MB */
-#define GCS6_ADDR		0x07800000 /* GCS6 - PCI control registers */
-#define GCS6_SIZE		0x00800000 /*   8MB */
-
-
-/* For <asm/page.h> */
-#define PAGE_OFFSET 		SDRAM_ADDR
-
-
-#ifdef CONFIG_ROM_KERNEL
-/* Kernel is in ROM, starting at address 0.  */
-
-#define INTV_BASE		0
-#define ROOT_FS_IMAGE_RW	0
-
-#else /* !CONFIG_ROM_KERNEL */
-/* Using RAM-kernel.  Assume some sort of boot-loader got us loaded at
-   address 0.  */
-
-#define INTV_BASE		0
-#define ROOT_FS_IMAGE_RW	1
-
-#endif /* CONFIG_ROM_KERNEL */
-
-
-/* Some misc. on-board devices.  */
-
-/* Seven-segment LED display (four digits).  */
-#define LED_ADDR(n)		(0x0FE02000 + (n))
-#define LED(n)			(*(volatile unsigned char *)LED_ADDR(n))
-#define LED_NUM_DIGITS		4
-
-
-/* On-board PIC.  */
-
-#define CB_PIC_BASE_ADDR 	0x0FE04000
-
-#define CB_PIC_INT0M_ADDR 	(CB_PIC_BASE_ADDR + 0x00)
-#define CB_PIC_INT0M      	(*(volatile u16 *)CB_PIC_INT0M_ADDR)
-#define CB_PIC_INT1M_ADDR 	(CB_PIC_BASE_ADDR + 0x10)
-#define CB_PIC_INT1M      	(*(volatile u16 *)CB_PIC_INT1M_ADDR)
-#define CB_PIC_INTR_ADDR  	(CB_PIC_BASE_ADDR + 0x20)
-#define CB_PIC_INTR       	(*(volatile u16 *)CB_PIC_INTR_ADDR)
-#define CB_PIC_INTEN_ADDR 	(CB_PIC_BASE_ADDR + 0x30)
-#define CB_PIC_INTEN      	(*(volatile u16 *)CB_PIC_INTEN_ADDR)
-
-#define CB_PIC_INT0EN        	0x0001
-#define CB_PIC_INT1EN        	0x0002
-#define CB_PIC_INT0SEL       	0x0080
-
-/* The PIC interrupts themselves.  */
-#define CB_PIC_BASE_IRQ		NUM_CPU_IRQS
-#define IRQ_CB_PIC_NUM		10
-
-/* Some specific CB_PIC interrupts. */
-#define IRQ_CB_EXTTM0		(CB_PIC_BASE_IRQ + 0)
-#define IRQ_CB_EXTSIO		(CB_PIC_BASE_IRQ + 1)
-#define IRQ_CB_TOVER		(CB_PIC_BASE_IRQ + 2)
-#define IRQ_CB_GINT0		(CB_PIC_BASE_IRQ + 3)
-#define IRQ_CB_USB		(CB_PIC_BASE_IRQ + 4)
-#define IRQ_CB_LANC		(CB_PIC_BASE_IRQ + 5)
-#define IRQ_CB_USB_VBUS_ON	(CB_PIC_BASE_IRQ + 6)
-#define IRQ_CB_USB_VBUS_OFF	(CB_PIC_BASE_IRQ + 7)
-#define IRQ_CB_EXTTM1		(CB_PIC_BASE_IRQ + 8)
-#define IRQ_CB_EXTTM2		(CB_PIC_BASE_IRQ + 9)
-
-/* The GBUS GINT1 - GINT3 (note, not GINT0!) interrupts are connected to
-   the INTP65 - INTP67 pins on the CPU.  These are shared among the GBUS
-   interrupts.  */
-#define IRQ_GINT(n)		IRQ_INTP((n) + 9)  /* 0 is unused! */
-#define IRQ_GINT_NUM		4		   /* 0 is unused! */
-
-/* The shared interrupt line from the PIC is connected to CPU pin INTP23.  */
-#define IRQ_CB_PIC		IRQ_INTP(4) /* P23 */
-
-/* Used by <asm/rte_cb.h> to derive NUM_MACH_IRQS.  */
-#define NUM_RTE_CB_IRQS		(NUM_CPU_IRQS + IRQ_CB_PIC_NUM)
-
-
-#ifndef __ASSEMBLY__
-struct cb_pic_irq_init {
-	const char *name;	/* name of interrupt type */
-
-	/* Range of kernel irq numbers for this type:
-	   BASE, BASE+INTERVAL, ..., BASE+INTERVAL*NUM  */
-	unsigned base, num, interval;
-
-	unsigned priority;	/* interrupt priority to assign */
-};
-struct hw_interrupt_type;	/* fwd decl */
-
-/* Enable interrupt handling for interrupt IRQ.  */
-extern void cb_pic_enable_irq (unsigned irq);
-/* Disable interrupt handling for interrupt IRQ.  Note that any interrupts
-   received while disabled will be delivered once the interrupt is enabled
-   again, unless they are explicitly cleared using `cb_pic_clear_pending_irq'.  */
-extern void cb_pic_disable_irq (unsigned irq);
-/* Initialize HW_IRQ_TYPES for PIC irqs described in array INITS (which is
-   terminated by an entry with the name field == 0).  */
-extern void cb_pic_init_irq_types (struct cb_pic_irq_init *inits,
-				   struct hw_interrupt_type *hw_irq_types);
-/* Initialize PIC interrupts.  */
-extern void cb_pic_init_irqs (void);
-#endif /* __ASSEMBLY__ */
-
-
-/* TL16C550C on board UART see also asm/serial.h */
-#define CB_UART_BASE    	0x0FE08000
-#define CB_UART_REG_GAP 	0x10
-#define CB_UART_CLOCK   	0x16000000
-
-/* CompactFlash setting */
-#define CB_CF_BASE     		0x0FE0C000
-#define CB_CF_CCR_ADDR 		(CB_CF_BASE+0x200)
-#define CB_CF_CCR      		(*(volatile u8 *)CB_CF_CCR_ADDR)
-#define CB_CF_REG0_ADDR		(CB_CF_BASE+0x1000)
-#define CB_CF_REG0     		(*(volatile u16 *)CB_CF_REG0_ADDR)
-#define CB_CF_STS0_ADDR		(CB_CF_BASE+0x1004)
-#define CB_CF_STS0     		(*(volatile u16 *)CB_CF_STS0_ADDR)
-#define CB_PCATA_BASE  		(CB_CF_BASE+0x800)
-#define CB_IDE_BASE    		(CB_CF_BASE+0x9F0)
-#define CB_IDE_CTRL    		(CB_CF_BASE+0xBF6)
-#define CB_IDE_REG_OFFS		0x1
-
-
-/* SMSC LAN91C111 setting */
-#if defined(CONFIG_SMC91111)
-#define CB_LANC_BASE 		0x0FE10300
-#define CONFIG_SMC16BITONLY
-#define ETH0_ADDR 		CB_LANC_BASE
-#define ETH0_IRQ 		IRQ_CB_LANC
-#endif /* CONFIG_SMC16BITONLY */
-
-
-#undef V850E_UART_PRE_CONFIGURE
-#define V850E_UART_PRE_CONFIGURE	rte_me2_cb_uart_pre_configure
-#ifndef __ASSEMBLY__
-extern void rte_me2_cb_uart_pre_configure (unsigned chan,
-					   unsigned cflags, unsigned baud);
-#endif /* __ASSEMBLY__ */
-
-/* This board supports RTS/CTS for the on-chip UART, but only for channel 0. */
-
-/* CTS for UART channel 0 is pin P22 (bit 2 of port 2).  */
-#define V850E_UART_CTS(chan)	((chan) == 0 ? !(ME2_PORT2_IO & 0x4) : 1)
-/* RTS for UART channel 0 is pin P21 (bit 1 of port 2).  */
-#define V850E_UART_SET_RTS(chan, val)					      \
-   do {									      \
-	   if (chan == 0) {						      \
-		   unsigned old = ME2_PORT2_IO; 			      \
-		   if (val)						      \
-			   ME2_PORT2_IO = old & ~0x2;			      \
-		   else							      \
-			   ME2_PORT2_IO = old | 0x2;			      \
-	   }								      \
-   } while (0)
-
-
-#ifndef __ASSEMBLY__
-extern void rte_me2_cb_init_irqs (void);
-#endif /* !__ASSEMBLY__ */
-
-
-#endif /* __V850_RTE_ME2_CB_H__ */
diff --git a/include/asm-v850/rte_nb85e_cb.h b/include/asm-v850/rte_nb85e_cb.h
deleted file mode 100644
index f56591cad90a..000000000000
--- a/include/asm-v850/rte_nb85e_cb.h
+++ /dev/null
@@ -1,111 +0,0 @@
-/*
- * include/asm-v850/rte_nb85e_cb.h -- Midas labs RTE-V850/NB85E-CB board
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_RTE_NB85E_CB_H__
-#define __V850_RTE_NB85E_CB_H__
-
-#include <asm/rte_cb.h>		/* Common defs for Midas RTE-CB boards.  */
-
-
-#define PLATFORM		"rte-v850e/nb85e-cb"
-#define PLATFORM_LONG		"Midas lab RTE-V850E/NB85E-CB"
-
-#define CPU_CLOCK_FREQ		50000000 /* 50MHz */
-
-/* 1MB of onboard SRAM.  Note that the monitor ROM uses parts of this
-   for its own purposes, so care must be taken.  */
-#define SRAM_ADDR		0x03C00000
-#define SRAM_SIZE		0x00100000 /* 1MB */
-
-/* 16MB of onbard SDRAM.  */
-#define SDRAM_ADDR		0x01000000
-#define SDRAM_SIZE		0x01000000 /* 16MB */
-
-
-/* CPU addresses of GBUS memory spaces.  */
-#define GCS0_ADDR		0x00400000 /* GCS0 - Common SRAM (2MB) */
-#define GCS0_SIZE		0x00400000 /*   4MB */
-#define GCS1_ADDR		0x02000000 /* GCS1 - Flash ROM (8MB) */
-#define GCS1_SIZE		0x00800000 /*   8MB */
-#define GCS2_ADDR		0x03900000 /* GCS2 - I/O registers */
-#define GCS2_SIZE		0x00080000 /*   512KB */
-#define GCS3_ADDR		0x02800000 /* GCS3 - EXT-bus: memory space */
-#define GCS3_SIZE		0x00800000 /*   8MB */
-#define GCS4_ADDR		0x03A00000 /* GCS4 - EXT-bus: I/O space */
-#define GCS4_SIZE		0x00200000 /*   2MB */
-#define GCS5_ADDR		0x00800000 /* GCS5 - PCI bus space */
-#define GCS5_SIZE		0x00800000 /*   8MB */
-#define GCS6_ADDR		0x03980000 /* GCS6 - PCI control registers */
-#define GCS6_SIZE		0x00010000 /*   64KB */
-
-
-/* The GBUS GINT0 - GINT3 interrupts are connected to CPU interrupts 10-12.
-   These are shared among the GBUS interrupts.  */
-#define IRQ_GINT(n)		(10 + (n))
-#define IRQ_GINT_NUM		3
-
-/* Used by <asm/rte_cb.h> to derive NUM_MACH_IRQS.  */
-#define NUM_RTE_CB_IRQS		NUM_CPU_IRQS
-
-
-#ifdef CONFIG_ROM_KERNEL
-/* Kernel is in ROM, starting at address 0.  */
-
-#define INTV_BASE	0
-
-#else /* !CONFIG_ROM_KERNEL */
-/* We're using the ROM monitor.  */
-
-/* The chip's real interrupt vectors are in ROM, but they jump to a
-   secondary interrupt vector table in RAM.  */
-#define INTV_BASE		0x03CF8000
-
-/* Scratch memory used by the ROM monitor, which shouldn't be used by
-   linux (except for the alternate interrupt vector area, defined
-   above).  */
-#define MON_SCRATCH_ADDR	0x03CE8000
-#define MON_SCRATCH_SIZE	0x00018000 /* 96KB */
-
-#endif /* CONFIG_ROM_KERNEL */
-
-
-/* Some misc. on-board devices.  */
-
-/* Seven-segment LED display (two digits).  Write-only.  */
-#define LED_ADDR(n)	(0x03802000 + (n))
-#define LED(n)		(*(volatile unsigned char *)LED_ADDR(n))
-#define LED_NUM_DIGITS	4
-
-
-/* Override the basic TEG UART pre-initialization so that we can
-   initialize extra stuff.  */
-#undef V850E_UART_PRE_CONFIGURE	/* should be defined by <asm/teg.h> */
-#define V850E_UART_PRE_CONFIGURE	rte_nb85e_cb_uart_pre_configure
-#ifndef __ASSEMBLY__
-extern void rte_nb85e_cb_uart_pre_configure (unsigned chan,
-					     unsigned cflags, unsigned baud);
-#endif
-
-/* This board supports RTS/CTS for the on-chip UART. */
-
-/* CTS is pin P00.  */
-#define V850E_UART_CTS(chan)	(! (TEG_PORT0_IO & 0x1))
-/* RTS is pin P02.  */
-#define V850E_UART_SET_RTS(chan, val)					      \
-   do {									      \
-	   unsigned old = TEG_PORT0_IO;					      \
-	   TEG_PORT0_IO = val ? (old & ~0x4) : (old | 0x4);		      \
-   } while (0)
-
-
-#endif /* __V850_RTE_NB85E_CB_H__ */
diff --git a/include/asm-v850/scatterlist.h b/include/asm-v850/scatterlist.h
deleted file mode 100644
index 02d27b3fb061..000000000000
--- a/include/asm-v850/scatterlist.h
+++ /dev/null
@@ -1,31 +0,0 @@
-/*
- * include/asm-v850/scatterlist.h
- *
- *  Copyright (C) 2001,02  NEC Corporation
- *  Copyright (C) 2001,02  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_SCATTERLIST_H__
-#define __V850_SCATTERLIST_H__
-
-#include <asm/types.h>
-
-struct scatterlist {
-#ifdef CONFIG_DEBUG_SG
-	unsigned long	sg_magic;
-#endif
-	unsigned long	page_link;
-	unsigned	offset;
-	dma_addr_t	dma_address;
-	unsigned	length;
-};
-
-#define ISA_DMA_THRESHOLD	(~0UL)
-
-#endif /* __V850_SCATTERLIST_H__ */
diff --git a/include/asm-v850/sections.h b/include/asm-v850/sections.h
deleted file mode 100644
index e0238253a0d0..000000000000
--- a/include/asm-v850/sections.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __V850_SECTIONS_H__
-#define __V850_SECTIONS_H__
-
-#include <asm-generic/sections.h>
-
-#endif /* __V850_SECTIONS_H__ */
diff --git a/include/asm-v850/segment.h b/include/asm-v850/segment.h
deleted file mode 100644
index 5e2b15dcf3d9..000000000000
--- a/include/asm-v850/segment.h
+++ /dev/null
@@ -1,36 +0,0 @@
-#ifndef __V850_SEGMENT_H__
-#define __V850_SEGMENT_H__
-
-
-#ifndef __ASSEMBLY__
-
-typedef unsigned long mm_segment_t;	/* domain register */
-
-#endif /* !__ASSEMBLY__ */
-
-
-#define __KERNEL_CS	0x0
-#define __KERNEL_DS	0x0
-
-#define __USER_CS	0x1
-#define __USER_DS	0x1
-
-#define KERNEL_DS	__KERNEL_DS
-#define KERNEL_CS	__KERNEL_CS
-#define USER_DS		__USER_DS
-#define USER_CS		__USER_CS
-
-#define segment_eq(a,b)	((a) == (b))
-
-#define get_ds()	(KERNEL_DS)
-#define get_fs()	(USER_DS)
-
-#define set_fs(seg)	((void)(seg))
-
-
-#define copy_segments(task, mm)	((void)((void)(task), (mm)))
-#define release_segments(mm)	((void)(mm))
-#define forget_segments()	((void)0)
-
-
-#endif /* __V850_SEGMENT_H__ */
diff --git a/include/asm-v850/semaphore.h b/include/asm-v850/semaphore.h
deleted file mode 100644
index d9b2034ed1d2..000000000000
--- a/include/asm-v850/semaphore.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <linux/semaphore.h>
diff --git a/include/asm-v850/sembuf.h b/include/asm-v850/sembuf.h
deleted file mode 100644
index 1622231a8b85..000000000000
--- a/include/asm-v850/sembuf.h
+++ /dev/null
@@ -1,25 +0,0 @@
-#ifndef __V850_SEMBUF_H__
-#define __V850_SEMBUF_H__
-
-/* 
- * The semid64_ds structure for v850 architecture.
- * Note extra padding because this structure is passed back and forth
- * between kernel and user space.
- *
- * Pad space is left for:
- * - 64-bit time_t to solve y2038 problem
- * - 2 miscellaneous 32-bit values
- */
-
-struct semid64_ds {
-	struct ipc64_perm sem_perm;		/* permissions .. see ipc.h */
-	__kernel_time_t	sem_otime;		/* last semop time */
-	unsigned long	__unused1;
-	__kernel_time_t	sem_ctime;		/* last change time */
-	unsigned long	__unused2;
-	unsigned long	sem_nsems;		/* no. of semaphores in array */
-	unsigned long	__unused3;
-	unsigned long	__unused4;
-};
-
-#endif /* __V850_SEMBUF_H__ */
diff --git a/include/asm-v850/serial.h b/include/asm-v850/serial.h
deleted file mode 100644
index 36d8f4cbbf39..000000000000
--- a/include/asm-v850/serial.h
+++ /dev/null
@@ -1,56 +0,0 @@
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 1999 by Ralf Baechle
- * Copyright (C) 1999, 2000 Silicon Graphics, Inc.
- */ 
-
-#ifdef CONFIG_RTE_CB_ME2
-
-#include <asm/rte_me2_cb.h>
-
-#define STD_COM_FLAGS (ASYNC_BOOT_AUTOCONF | ASYNC_SKIP_TEST)
-
-#define irq_cannonicalize(x) (x)
-#define BASE_BAUD	250000	/* (16MHz / (16 * 38400)) * 9600 */
-#define SERIAL_PORT_DFNS \
-   { 0, BASE_BAUD, CB_UART_BASE, IRQ_CB_EXTSIO, STD_COM_FLAGS },
-
-/* Redefine UART register offsets.  */
-#undef UART_RX
-#undef UART_TX
-#undef UART_DLL
-#undef UART_TRG
-#undef UART_DLM
-#undef UART_IER
-#undef UART_FCTR
-#undef UART_IIR
-#undef UART_FCR
-#undef UART_EFR
-#undef UART_LCR
-#undef UART_MCR
-#undef UART_LSR
-#undef UART_MSR
-#undef UART_SCR
-#undef UART_EMSR
-
-#define UART_RX		(0 * CB_UART_REG_GAP)
-#define UART_TX		(0 * CB_UART_REG_GAP)
-#define UART_DLL	(0 * CB_UART_REG_GAP)
-#define UART_TRG	(0 * CB_UART_REG_GAP)
-#define UART_DLM	(1 * CB_UART_REG_GAP)
-#define UART_IER	(1 * CB_UART_REG_GAP)
-#define UART_FCTR	(1 * CB_UART_REG_GAP)
-#define UART_IIR	(2 * CB_UART_REG_GAP)
-#define UART_FCR	(2 * CB_UART_REG_GAP)
-#define UART_EFR	(2 * CB_UART_REG_GAP)
-#define UART_LCR	(3 * CB_UART_REG_GAP)
-#define UART_MCR	(4 * CB_UART_REG_GAP)
-#define UART_LSR	(5 * CB_UART_REG_GAP)
-#define UART_MSR	(6 * CB_UART_REG_GAP)
-#define UART_SCR	(7 * CB_UART_REG_GAP)
-#define UART_EMSR	(7 * CB_UART_REG_GAP)
-
-#endif /* CONFIG_RTE_CB_ME2 */
diff --git a/include/asm-v850/setup.h b/include/asm-v850/setup.h
deleted file mode 100644
index c48a9b97d05b..000000000000
--- a/include/asm-v850/setup.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef _V850_SETUP_H
-#define _V850_SETUP_H
-
-#define COMMAND_LINE_SIZE	512
-
-#endif /* __SETUP_H */
diff --git a/include/asm-v850/shmbuf.h b/include/asm-v850/shmbuf.h
deleted file mode 100644
index 3d085c9c418e..000000000000
--- a/include/asm-v850/shmbuf.h
+++ /dev/null
@@ -1,42 +0,0 @@
-#ifndef __V850_SHMBUF_H__
-#define __V850_SHMBUF_H__
-
-/* 
- * The shmid64_ds structure for v850 architecture.
- * Note extra padding because this structure is passed back and forth
- * between kernel and user space.
- *
- * Pad space is left for:
- * - 64-bit time_t to solve y2038 problem
- * - 2 miscellaneous 32-bit values
- */
-
-struct shmid64_ds {
-	struct ipc64_perm	shm_perm;	/* operation perms */
-	size_t			shm_segsz;	/* size of segment (bytes) */
-	__kernel_time_t		shm_atime;	/* last attach time */
-	unsigned long		__unused1;
-	__kernel_time_t		shm_dtime;	/* last detach time */
-	unsigned long		__unused2;
-	__kernel_time_t		shm_ctime;	/* last change time */
-	unsigned long		__unused3;
-	__kernel_pid_t		shm_cpid;	/* pid of creator */
-	__kernel_pid_t		shm_lpid;	/* pid of last operator */
-	unsigned long		shm_nattch;	/* no. of current attaches */
-	unsigned long		__unused4;
-	unsigned long		__unused5;
-};
-
-struct shminfo64 {
-	unsigned long	shmmax;
-	unsigned long	shmmin;
-	unsigned long	shmmni;
-	unsigned long	shmseg;
-	unsigned long	shmall;
-	unsigned long	__unused1;
-	unsigned long	__unused2;
-	unsigned long	__unused3;
-	unsigned long	__unused4;
-};
-
-#endif /* __V850_SHMBUF_H__ */
diff --git a/include/asm-v850/shmparam.h b/include/asm-v850/shmparam.h
deleted file mode 100644
index 7dcb6739073e..000000000000
--- a/include/asm-v850/shmparam.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __V850_SHMPARAM_H__
-#define __V850_SHMPARAM_H__
-
-#define	SHMLBA		PAGE_SIZE	/* attach addr a multiple of this */
-
-#endif /* __V850_SHMPARAM_H__ */
diff --git a/include/asm-v850/sigcontext.h b/include/asm-v850/sigcontext.h
deleted file mode 100644
index e0890f6f4bc9..000000000000
--- a/include/asm-v850/sigcontext.h
+++ /dev/null
@@ -1,25 +0,0 @@
-/*
- * include/asm-v850/sigcontext.h -- Signal contexts
- *
- *  Copyright (C) 2001  NEC Corporation
- *  Copyright (C) 2001  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_SIGCONTEXT_H__
-#define __V850_SIGCONTEXT_H__
-
-#include <asm/ptrace.h>
-
-struct sigcontext
-{
-	struct pt_regs 	regs;
-	unsigned long	oldmask;
-};
-
-#endif /* __V850_SIGCONTEXT_H__ */
diff --git a/include/asm-v850/siginfo.h b/include/asm-v850/siginfo.h
deleted file mode 100644
index 7eb94703dce0..000000000000
--- a/include/asm-v850/siginfo.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __V850_SIGINFO_H__
-#define __V850_SIGINFO_H__
-
-#include <asm-generic/siginfo.h>
-
-#endif /* __V850_SIGINFO_H__ */
diff --git a/include/asm-v850/signal.h b/include/asm-v850/signal.h
deleted file mode 100644
index a38df0834bbf..000000000000
--- a/include/asm-v850/signal.h
+++ /dev/null
@@ -1,168 +0,0 @@
-#ifndef __V850_SIGNAL_H__
-#define __V850_SIGNAL_H__
-
-#include <linux/types.h>
-
-/* Avoid too many header ordering problems.  */
-struct siginfo;
-
-
-#ifdef __KERNEL__
-
-/* Most things should be clean enough to redefine this at will, if care
-   is taken to make libc match.  */
-#define _NSIG		64
-#define _NSIG_BPW	32
-#define _NSIG_WORDS	(_NSIG / _NSIG_BPW)
-
-typedef unsigned long old_sigset_t;		/* at least 32 bits */
-
-typedef struct {
-	unsigned long sig[_NSIG_WORDS];
-} sigset_t;
-
-#else /* !__KERNEL__ */
-
-/* Here we must cater to libcs that poke about in kernel headers.  */
-
-#define NSIG		32
-typedef unsigned long sigset_t;
-
-#endif /* __KERNEL__ */
-
-
-#define SIGHUP		 1
-#define SIGINT		 2
-#define SIGQUIT		 3
-#define SIGILL		 4
-#define SIGTRAP		 5
-#define SIGABRT		 6
-#define SIGIOT		 6
-#define SIGBUS		 7
-#define SIGFPE		 8
-#define SIGKILL		 9
-#define SIGUSR1		10
-#define SIGSEGV		11
-#define SIGUSR2		12
-#define SIGPIPE		13
-#define SIGALRM		14
-#define SIGTERM		15
-#define SIGSTKFLT	16
-#define SIGCHLD		17
-#define SIGCONT		18
-#define SIGSTOP		19
-#define SIGTSTP		20
-#define SIGTTIN		21
-#define SIGTTOU		22
-#define SIGURG		23
-#define SIGXCPU		24
-#define SIGXFSZ		25
-#define SIGVTALRM	26
-#define SIGPROF		27
-#define SIGWINCH	28
-#define SIGIO		29
-#define SIGPOLL		SIGIO
-/*
-#define SIGLOST		29
-*/
-#define SIGPWR		30
-#define SIGSYS		31
-#define	SIGUNUSED	31
-
-/* These should not be considered constants from userland.  */
-#define SIGRTMIN	32
-#define SIGRTMAX	_NSIG
-
-/*
- * SA_FLAGS values:
- *
- * SA_ONSTACK indicates that a registered stack_t will be used.
- * SA_RESTART flag to get restarting signals (which were the default long ago)
- * SA_NOCLDSTOP flag to turn off SIGCHLD when children stop.
- * SA_RESETHAND clears the handler when the signal is delivered.
- * SA_NOCLDWAIT flag on SIGCHLD to inhibit zombies.
- * SA_NODEFER prevents the current signal from being masked in the handler.
- *
- * SA_ONESHOT and SA_NOMASK are the historical Linux names for the Single
- * Unix names RESETHAND and NODEFER respectively.
- */
-#define SA_NOCLDSTOP	0x00000001
-#define SA_NOCLDWAIT	0x00000002
-#define SA_SIGINFO	0x00000004
-#define SA_ONSTACK	0x08000000
-#define SA_RESTART	0x10000000
-#define SA_NODEFER	0x40000000
-#define SA_RESETHAND	0x80000000
-
-#define SA_NOMASK	SA_NODEFER
-#define SA_ONESHOT	SA_RESETHAND
-
-#define SA_RESTORER	0x04000000
-
-/* 
- * sigaltstack controls
- */
-#define SS_ONSTACK	1
-#define SS_DISABLE	2
-
-#define MINSIGSTKSZ	2048
-#define SIGSTKSZ	8192
-
-#include <asm-generic/signal.h>
-
-#ifdef __KERNEL__
-
-struct old_sigaction {
-	__sighandler_t sa_handler;
-	old_sigset_t sa_mask;
-	unsigned long sa_flags;
-	void (*sa_restorer)(void);
-};
-
-struct sigaction {
-	__sighandler_t sa_handler;
-	unsigned long sa_flags;
-	void (*sa_restorer)(void);
-	sigset_t sa_mask;		/* mask last for extensibility */
-};
-
-struct k_sigaction {
-	struct sigaction sa;
-};
-
-#else /* !__KERNEL__ */
-
-/* Here we must cater to libcs that poke about in kernel headers.  */
-
-struct sigaction {
-	union {
-	  __sighandler_t _sa_handler;
-	  void (*_sa_sigaction)(int, struct siginfo *, void *);
-	} _u;
-	sigset_t sa_mask;
-	unsigned long sa_flags;
-	void (*sa_restorer)(void);
-};
-
-#define sa_handler	_u._sa_handler
-#define sa_sigaction	_u._sa_sigaction
-
-#endif /* __KERNEL__ */
-
-
-typedef struct sigaltstack {
-	void *ss_sp;
-	int ss_flags;
-	size_t ss_size;
-} stack_t;
-
-#ifdef __KERNEL__
-
-#include <asm/sigcontext.h>
-#undef __HAVE_ARCH_SIG_BITOPS
-
-#define ptrace_signal_deliver(regs, cookie) do { } while (0)
-
-#endif /* __KERNEL__ */
-
-#endif /* __V850_SIGNAL_H__ */
diff --git a/include/asm-v850/sim.h b/include/asm-v850/sim.h
deleted file mode 100644
index 026932d476cd..000000000000
--- a/include/asm-v850/sim.h
+++ /dev/null
@@ -1,47 +0,0 @@
-/*
- * include/asm-v850/sim.h -- Machine-dependent defs for GDB v850e simulator
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_SIM_H__
-#define __V850_SIM_H__
-
-
-#define CPU_ARCH		"v850e"
-#define CPU_MODEL		"v850e"
-#define CPU_MODEL_LONG		"NEC V850E"
-#define PLATFORM		"gdb/v850e"
-#define PLATFORM_LONG		"GDB V850E simulator"
-
-
-/* We use a weird value for RAM, not just 0, for testing purposes.
-   These must match the values used in the linker script.  */
-#define RAM_ADDR		0x8F000000
-#define RAM_SIZE		0x03000000
-
-
-/* For <asm/page.h> */
-#define PAGE_OFFSET 		RAM_ADDR
-
-
-/* For <asm/entry.h> */
-/* `R0 RAM', used for a few miscellaneous variables that must be
-   accessible using a load instruction relative to R0.  On real
-   processors, this usually is on-chip RAM, but here we just
-   choose an arbitrary address that meets the above constraint.  */
-#define R0_RAM_ADDR		0xFFFFF000
-
-
-/* For <asm/irq.h> */
-#define NUM_CPU_IRQS		6
-
-
-#endif /* __V850_SIM_H__ */
diff --git a/include/asm-v850/sim85e2.h b/include/asm-v850/sim85e2.h
deleted file mode 100644
index 8b4d6974066c..000000000000
--- a/include/asm-v850/sim85e2.h
+++ /dev/null
@@ -1,69 +0,0 @@
-/*
- * include/asm-v850/sim85e2.h -- Machine-dependent defs for
- *	V850E2 RTL simulator
- *
- *  Copyright (C) 2002,03  NEC Electronics Corporation
- *  Copyright (C) 2002,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_SIM85E2_H__
-#define __V850_SIM85E2_H__
-
-
-#include <asm/v850e2.h>		/* Based on V850E2 core.  */
-
-
-/* Various memory areas supported by the simulator.
-   These should match the corresponding definitions in the linker script.  */
-
-/* `instruction RAM'; instruction fetches are much faster from IRAM than
-   from DRAM.  */
-#define IRAM_ADDR		0
-#define IRAM_SIZE		0x00100000 /* 1MB */
-/* `data RAM', below and contiguous with the I/O space.
-   Data fetches are much faster from DRAM than from IRAM.  */
-#define DRAM_ADDR		0xfff00000
-#define DRAM_SIZE		0x000ff000 /* 1020KB */
-/* `external ram'.  Unlike the above RAM areas, this memory is cached,
-   so both instruction and data fetches should be (mostly) fast --
-   however, currently only write-through caching is supported, so writes
-   to ERAM will be slow.  */
-#define ERAM_ADDR		0x00100000
-#define ERAM_SIZE		0x07f00000 /* 127MB (max) */
-/* Dynamic RAM; uses memory controller.  */
-#define SDRAM_ADDR		0x10000000
-#define SDRAM_SIZE		0x01000000 /* 16MB */
-
-
-/* Simulator specific control registers.  */
-/* NOTHAL controls whether the simulator will stop at a `halt' insn.  */
-#define SIM85E2_NOTHAL_ADDR	0xffffff22
-#define SIM85E2_NOTHAL		(*(volatile u8 *)SIM85E2_NOTHAL_ADDR)
-/* The simulator will stop N cycles after N is written to SIMFIN.  */
-#define SIM85E2_SIMFIN_ADDR	0xffffff24
-#define SIM85E2_SIMFIN		(*(volatile u16 *)SIM85E2_SIMFIN_ADDR)
-
-
-/* For <asm/irq.h> */
-#define NUM_CPU_IRQS		64
-
-
-/* For <asm/page.h> */
-#define PAGE_OFFSET		SDRAM_ADDR
-
-
-/* For <asm/entry.h> */
-/* `R0 RAM', used for a few miscellaneous variables that must be accessible
-   using a load instruction relative to R0.  The sim85e2 simulator
-   actually puts 1020K of RAM from FFF00000 to FFFFF000, so we arbitarily
-   choose a small portion at the end of that.  */
-#define R0_RAM_ADDR		0xFFFFE000
-
-
-#endif /* __V850_SIM85E2_H__ */
diff --git a/include/asm-v850/sim85e2c.h b/include/asm-v850/sim85e2c.h
deleted file mode 100644
index eee543ff3af8..000000000000
--- a/include/asm-v850/sim85e2c.h
+++ /dev/null
@@ -1,26 +0,0 @@
-/*
- * include/asm-v850/sim85e2c.h -- Machine-dependent defs for
- *	V850E2 RTL simulator
- *
- *  Copyright (C) 2002  NEC Corporation
- *  Copyright (C) 2002  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_SIM85E2C_H__
-#define __V850_SIM85E2C_H__
-
-/* Use generic sim85e2 settings, other than the various names.  */
-#include <asm/sim85e2.h>
-
-#define CPU_MODEL	"v850e2"
-#define CPU_MODEL_LONG	"NEC V850E2"
-#define PLATFORM	"sim85e2c"
-#define PLATFORM_LONG	"SIM85E2C V850E2 simulator"
-
-#endif /* __V850_SIM85E2C_H__ */
diff --git a/include/asm-v850/sim85e2s.h b/include/asm-v850/sim85e2s.h
deleted file mode 100644
index ee066d5d3c51..000000000000
--- a/include/asm-v850/sim85e2s.h
+++ /dev/null
@@ -1,28 +0,0 @@
-/*
- * include/asm-v850/sim85e2s.h -- Machine-dependent defs for
- *	V850E2 RTL simulator
- *
- *  Copyright (C) 2003  NEC Electronics Corporation
- *  Copyright (C) 2003  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_SIM85E2S_H__
-#define __V850_SIM85E2S_H__
-
-#include <asm/sim85e2.h>	/* Use generic sim85e2 settings.  */
-#if 0
-#include <asm/v850e2_cache.h>	/* + cache */
-#endif
-
-#define CPU_MODEL	"v850e2"
-#define CPU_MODEL_LONG	"NEC V850E2"
-#define PLATFORM	"sim85e2s"
-#define PLATFORM_LONG	"SIM85E2S V850E2 simulator"
-
-#endif /* __V850_SIM85E2S_H__ */
diff --git a/include/asm-v850/simsyscall.h b/include/asm-v850/simsyscall.h
deleted file mode 100644
index 4a19d5ae9d17..000000000000
--- a/include/asm-v850/simsyscall.h
+++ /dev/null
@@ -1,99 +0,0 @@
-/*
- * include/asm-v850/simsyscall.h -- `System calls' under the v850e emulator
- *
- *  Copyright (C) 2001  NEC Corporation
- *  Copyright (C) 2001  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_SIMSYSCALL_H__
-#define __V850_SIMSYSCALL_H__
-
-#define V850_SIM_SYS_exit(a...)		V850_SIM_SYSCALL_1 (1	, ##a)
-#define V850_SIM_SYS_fork(a...)		V850_SIM_SYSCALL_0 (2	, ##a)
-#define V850_SIM_SYS_read(a...)		V850_SIM_SYSCALL_3 (3	, ##a)
-#define V850_SIM_SYS_write(a...)	V850_SIM_SYSCALL_3 (4	, ##a)
-#define V850_SIM_SYS_open(a...)		V850_SIM_SYSCALL_2 (5	, ##a)
-#define V850_SIM_SYS_close(a...)	V850_SIM_SYSCALL_1 (6	, ##a)
-#define V850_SIM_SYS_wait4(a...)	V850_SIM_SYSCALL_4 (7	, ##a)
-/* #define V850_SIM_SYS_creat(a...)	V850_SIM_SYSCALL_1 (8	, ##a) */
-/* #define V850_SIM_SYS_link(a...)	V850_SIM_SYSCALL_1 (9	, ##a) */
-/* #define V850_SIM_SYS_unlink(a...)	V850_SIM_SYSCALL_1 (10	, ##a) */
-#define V850_SIM_SYS_execv(a...)	V850_SIM_SYSCALL_2 (11	, ##a)
-/* #define V850_SIM_SYS_chdir(a...)	V850_SIM_SYSCALL_1 (12	, ##a) */
-/* #define V850_SIM_SYS_mknod(a...)	V850_SIM_SYSCALL_1 (14	, ##a) */
-#define V850_SIM_SYS_chmod(a...)	V850_SIM_SYSCALL_2 (15	, ##a)
-#define V850_SIM_SYS_chown(a...)	V850_SIM_SYSCALL_2 (16	, ##a)
-#define V850_SIM_SYS_lseek(a...)	V850_SIM_SYSCALL_3 (19	, ##a)
-/* #define V850_SIM_SYS_getpid(a...)	V850_SIM_SYSCALL_1 (20	, ##a) */
-/* #define V850_SIM_SYS_isatty(a...)	V850_SIM_SYSCALL_1 (21	, ##a) */
-/* #define V850_SIM_SYS_fstat(a...)	V850_SIM_SYSCALL_1 (22	, ##a) */
-#define V850_SIM_SYS_time(a...)		V850_SIM_SYSCALL_1 (23	, ##a)
-#define V850_SIM_SYS_poll(a...)		V850_SIM_SYSCALL_3 (24	, ##a)
-#define V850_SIM_SYS_stat(a...)		V850_SIM_SYSCALL_2 (38	, ##a)
-#define V850_SIM_SYS_pipe(a...)		V850_SIM_SYSCALL_1 (42	, ##a)
-#define V850_SIM_SYS_times(a...)	V850_SIM_SYSCALL_1 (43	, ##a)
-#define V850_SIM_SYS_execve(a...)	V850_SIM_SYSCALL_3 (59	, ##a)
-#define V850_SIM_SYS_gettimeofday(a...)	V850_SIM_SYSCALL_2 (116	, ##a)
-/* #define V850_SIM_SYS_utime(a...)	V850_SIM_SYSCALL_2 (201	, ##a) */
-/* #define V850_SIM_SYS_wait(a...)	V850_SIM_SYSCALL_1 (202	, ##a) */
-
-#define V850_SIM_SYS_make_raw(a...)	V850_SIM_SYSCALL_1 (1024 , ##a)
-
-
-#define V850_SIM_SYSCALL_0(_call)					      \
-({									      \
-	register int call __asm__ ("r6") = _call;			      \
-	register int rval __asm__ ("r10");				      \
-	__asm__ __volatile__ ("trap 31"					      \
-			      : "=r" (rval)				      \
-			      : "r" (call)				      \
-			      : "r11", "memory");			      \
-	rval;								      \
-})
-#define V850_SIM_SYSCALL_1(_call, _arg0)				      \
-({									      \
-	register int call __asm__ ("r6") = _call;			      \
-	register long arg0 __asm__ ("r7") = (long)_arg0;		      \
-	register int rval __asm__ ("r10");				      \
-	__asm__ __volatile__ ("trap 31"					      \
-			      : "=r" (rval)				      \
-			      : "r" (call), "r" (arg0)			      \
-			      : "r11", "memory");			      \
-	rval;								      \
-})
-#define V850_SIM_SYSCALL_2(_call, _arg0, _arg1)				      \
-({									      \
-	register int call __asm__ ("r6") = _call;			      \
-	register long arg0 __asm__ ("r7") = (long)_arg0;		      \
-	register long arg1 __asm__ ("r8") = (long)_arg1;		      \
-	register int rval __asm__ ("r10");				      \
-	__asm__ __volatile__ ("trap 31"					      \
-			      : "=r" (rval)				      \
-			      : "r" (call), "r" (arg0), "r" (arg1)	      \
-			      : "r11", "memory");			      \
-	rval;								      \
-})
-#define V850_SIM_SYSCALL_3(_call, _arg0, _arg1, _arg2)			      \
-({									      \
-	register int call __asm__ ("r6") = _call;			      \
-	register long arg0 __asm__ ("r7") = (long)_arg0;		      \
-	register long arg1 __asm__ ("r8") = (long)_arg1;		      \
-	register long arg2 __asm__ ("r9") = (long)_arg2;		      \
-	register int rval __asm__ ("r10");				      \
-	__asm__ __volatile__ ("trap 31"					      \
-			      : "=r" (rval)				      \
-			      : "r" (call), "r" (arg0), "r" (arg1), "r" (arg2)\
-			      : "r11", "memory");			      \
-	rval;								      \
-})
-
-#define V850_SIM_SYSCALL(call, args...) \
-   V850_SIM_SYS_##call (args)
-
-#endif /* __V850_SIMSYSCALL_H__ */
diff --git a/include/asm-v850/socket.h b/include/asm-v850/socket.h
deleted file mode 100644
index e199a2bf12aa..000000000000
--- a/include/asm-v850/socket.h
+++ /dev/null
@@ -1,57 +0,0 @@
-#ifndef __V850_SOCKET_H__
-#define __V850_SOCKET_H__
-
-#include <asm/sockios.h>
-
-/* For setsockoptions(2) */
-#define SOL_SOCKET	1
-
-#define SO_DEBUG	1
-#define SO_REUSEADDR	2
-#define SO_TYPE		3
-#define SO_ERROR	4
-#define SO_DONTROUTE	5
-#define SO_BROADCAST	6
-#define SO_SNDBUF	7
-#define SO_RCVBUF	8
-#define SO_SNDBUFFORCE	32
-#define SO_RCVBUFFORCE	33
-#define SO_KEEPALIVE	9
-#define SO_OOBINLINE	10
-#define SO_NO_CHECK	11
-#define SO_PRIORITY	12
-#define SO_LINGER	13
-#define SO_BSDCOMPAT	14
-/* To add :#define SO_REUSEPORT 15 */
-#define SO_PASSCRED	16
-#define SO_PEERCRED	17
-#define SO_RCVLOWAT	18
-#define SO_SNDLOWAT	19
-#define SO_RCVTIMEO	20
-#define SO_SNDTIMEO	21
-
-/* Security levels - as per NRL IPv6 - don't actually do anything */
-#define SO_SECURITY_AUTHENTICATION		22
-#define SO_SECURITY_ENCRYPTION_TRANSPORT	23
-#define SO_SECURITY_ENCRYPTION_NETWORK		24
-
-#define SO_BINDTODEVICE	25
-
-/* Socket filtering */
-#define SO_ATTACH_FILTER        26
-#define SO_DETACH_FILTER        27
-
-#define SO_PEERNAME             28
-#define SO_TIMESTAMP		29
-#define SCM_TIMESTAMP		SO_TIMESTAMP
-
-#define SO_ACCEPTCONN		30
-
-#define SO_PEERSEC		31
-#define SO_PASSSEC		34
-#define SO_TIMESTAMPNS		35
-#define SCM_TIMESTAMPNS		SO_TIMESTAMPNS
-
-#define SO_MARK			36
-
-#endif /* __V850_SOCKET_H__ */
diff --git a/include/asm-v850/sockios.h b/include/asm-v850/sockios.h
deleted file mode 100644
index 823e106e6cd0..000000000000
--- a/include/asm-v850/sockios.h
+++ /dev/null
@@ -1,13 +0,0 @@
-#ifndef __V850_SOCKIOS_H__
-#define __V850_SOCKIOS_H__
-
-/* Socket-level I/O control calls. */
-#define FIOSETOWN 	0x8901
-#define SIOCSPGRP	0x8902
-#define FIOGETOWN	0x8903
-#define SIOCGPGRP	0x8904
-#define SIOCATMARK	0x8905
-#define SIOCGSTAMP	0x8906		/* Get stamp (timeval) */
-#define SIOCGSTAMPNS	0x8907		/* Get stamp (timespec) */
-
-#endif /* __V850_SOCKIOS_H__ */
diff --git a/include/asm-v850/stat.h b/include/asm-v850/stat.h
deleted file mode 100644
index c68c60d06e2f..000000000000
--- a/include/asm-v850/stat.h
+++ /dev/null
@@ -1,73 +0,0 @@
-/*
- * include/asm-v850/stat.h -- v850 stat structure
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_STAT_H__
-#define __V850_STAT_H__
-
-#include <asm/posix_types.h>
-
-struct stat {
-	unsigned int	st_dev;
-	unsigned long	st_ino;
-	unsigned int	st_mode;
-	unsigned int	st_nlink;
-	unsigned int 	st_uid;
-	unsigned int 	st_gid;
-	unsigned int	st_rdev;
-	long		st_size;
-	unsigned long	st_blksize;
-	unsigned long	st_blocks;
-	unsigned long	st_atime;
-	unsigned long	__unused1;
-	unsigned long	st_mtime;
-	unsigned long	__unused2;
-	unsigned long	st_ctime;
-	unsigned long	__unused3;
-	unsigned long	__unused4;
-	unsigned long	__unused5;
-};
-
-struct stat64 {
-	unsigned long long	st_dev;
-	unsigned long	__unused1;
-
-	unsigned long long	st_ino;
-
-	unsigned int	st_mode;
-	unsigned int	st_nlink;
-
-	unsigned int	st_uid;
-	unsigned int	st_gid;
-
-	unsigned long long	st_rdev;
-	unsigned long	__unused3;
-
-	long long	st_size;
-	unsigned long	st_blksize;
-
-	unsigned long	st_blocks; /* No. of 512-byte blocks allocated */
-	unsigned long	__unused4; /* future possible st_blocks high bits */
-
-	unsigned long	st_atime;
-	unsigned long	st_atime_nsec;
-
-	unsigned long	st_mtime;
-	unsigned long	st_mtime_nsec;
-
-	unsigned long	st_ctime;
-	unsigned long	st_ctime_nsec;
-
-	unsigned long	__unused8;
-};
-
-#endif /* __V850_STAT_H__ */
diff --git a/include/asm-v850/statfs.h b/include/asm-v850/statfs.h
deleted file mode 100644
index ea1596607f26..000000000000
--- a/include/asm-v850/statfs.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __V850_STATFS_H__
-#define __V850_STATFS_H__
-
-#include <asm-generic/statfs.h>
-
-#endif /* __V850_STATFS_H__ */
diff --git a/include/asm-v850/string.h b/include/asm-v850/string.h
deleted file mode 100644
index 478e234789d6..000000000000
--- a/include/asm-v850/string.h
+++ /dev/null
@@ -1,25 +0,0 @@
-/*
- * include/asm-v850/string.h -- Architecture specific string routines
- *
- *  Copyright (C) 2001,02  NEC Corporation
- *  Copyright (C) 2001,02  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_STRING_H__
-#define __V850_STRING_H__
-
-#define __HAVE_ARCH_MEMCPY
-#define __HAVE_ARCH_MEMSET
-#define __HAVE_ARCH_MEMMOVE
-
-extern void *memcpy (void *, const void *, __kernel_size_t);
-extern void *memset (void *, int, __kernel_size_t);
-extern void *memmove (void *, const void *, __kernel_size_t);
-
-#endif /* __V850_STRING_H__ */
diff --git a/include/asm-v850/system.h b/include/asm-v850/system.h
deleted file mode 100644
index 7daf1fdee119..000000000000
--- a/include/asm-v850/system.h
+++ /dev/null
@@ -1,123 +0,0 @@
-/*
- * include/asm-v850/system.h -- Low-level interrupt/thread ops
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_SYSTEM_H__
-#define __V850_SYSTEM_H__
-
-#include <linux/linkage.h>
-#include <asm/ptrace.h>
-
-
-/*
- * switch_to(n) should switch tasks to task ptr, first checking that
- * ptr isn't the current task, in which case it does nothing.
- */
-struct thread_struct;
-extern void *switch_thread (struct thread_struct *last,
-			    struct thread_struct *next);
-#define switch_to(prev,next,last)					      \
-  do {									      \
-        if (prev != next) {						      \
- 		(last) = switch_thread (&prev->thread, &next->thread);	      \
-	}								      \
-  } while (0)
-
-
-/* Enable/disable interrupts.  */
-#define local_irq_enable()	__asm__ __volatile__ ("ei")
-#define local_irq_disable()	__asm__ __volatile__ ("di")
-
-#define local_save_flags(flags) \
-  __asm__ __volatile__ ("stsr %1, %0" : "=r" (flags) : "i" (SR_PSW))
-#define local_restore_flags(flags) \
-  __asm__ __volatile__ ("ldsr %0, %1" :: "r" (flags), "i" (SR_PSW))
-
-/* For spinlocks etc */
-#define	local_irq_save(flags) \
-  do { local_save_flags (flags); local_irq_disable (); } while (0) 
-#define local_irq_restore(flags) \
-  local_restore_flags (flags);
-
-
-static inline int irqs_disabled (void)
-{
-	unsigned flags;
-	local_save_flags (flags);
-	return !!(flags & 0x20);
-}
-
-
-/*
- * Force strict CPU ordering.
- * Not really required on v850...
- */
-#define nop()			__asm__ __volatile__ ("nop")
-#define mb()			__asm__ __volatile__ ("" ::: "memory")
-#define rmb()			mb ()
-#define wmb()			mb ()
-#define read_barrier_depends()	((void)0)
-#define set_mb(var, value)	do { xchg (&var, value); } while (0)
-
-#define smp_mb()	mb ()
-#define smp_rmb()	rmb ()
-#define smp_wmb()	wmb ()
-#define smp_read_barrier_depends()	read_barrier_depends()
-
-#define xchg(ptr, with) \
-  ((__typeof__ (*(ptr)))__xchg ((unsigned long)(with), (ptr), sizeof (*(ptr))))
-
-static inline unsigned long __xchg (unsigned long with,
-				    __volatile__ void *ptr, int size)
-{
-	unsigned long tmp, flags;
-
-	local_irq_save (flags);
-
-	switch (size) {
-	case 1:
-		tmp = *(unsigned char *)ptr;
-		*(unsigned char *)ptr = with;
-		break;
-	case 2:
-		tmp = *(unsigned short *)ptr;
-		*(unsigned short *)ptr = with;
-		break;
-	case 4:
-		tmp = *(unsigned long *)ptr;
-		*(unsigned long *)ptr = with;
-		break;
-	}
-
-	local_irq_restore (flags);
-
-	return tmp;
-}
-
-#include <asm-generic/cmpxchg-local.h>
-
-/*
- * cmpxchg_local and cmpxchg64_local are atomic wrt current CPU. Always make
- * them available.
- */
-#define cmpxchg_local(ptr, o, n)				  	       \
-	((__typeof__(*(ptr)))__cmpxchg_local_generic((ptr), (unsigned long)(o),\
-			(unsigned long)(n), sizeof(*(ptr))))
-#define cmpxchg64_local(ptr, o, n) __cmpxchg64_local_generic((ptr), (o), (n))
-
-#ifndef CONFIG_SMP
-#include <asm-generic/cmpxchg.h>
-#endif
-
-#define arch_align_stack(x) (x)
-
-#endif /* __V850_SYSTEM_H__ */
diff --git a/include/asm-v850/teg.h b/include/asm-v850/teg.h
deleted file mode 100644
index acc8c7d95329..000000000000
--- a/include/asm-v850/teg.h
+++ /dev/null
@@ -1,101 +0,0 @@
-/*
- * include/asm-v850/teg.h -- NB85E-TEG cpu chip
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_TEG_H__
-#define __V850_TEG_H__
-
-
-/* The TEG uses the V850E cpu core.  */
-#include <asm/v850e.h>
-#include <asm/v850e_cache.h>
-
-
-#define CPU_MODEL	"v850e/nb85e-teg"
-#define CPU_MODEL_LONG	"NEC V850E/NB85E TEG"
-
-
-/* For <asm/entry.h> */
-/* We use on-chip RAM, for a few miscellaneous variables that must be
-   accessible using a load instruction relative to R0.  On the NB85E/TEG,
-   There's 60KB of iRAM starting at 0xFFFF0000, however we need the base
-   address to be addressable by a 16-bit signed offset, so we only use the
-   second half of it starting from 0xFFFF8000.  */
-#define R0_RAM_ADDR			0xFFFF8000
-
-
-/* Hardware-specific interrupt numbers (in the kernel IRQ namespace).
-   Some of these are parameterized even though there's only a single
-   interrupt, for compatibility with some generic code that works on other
-   processor models.  */
-#define IRQ_INTCMD(n)	6	/* interval timer interrupt */
-#define IRQ_INTCMD_NUM	1
-#define IRQ_INTSER(n)	16	/* UART reception error */
-#define IRQ_INTSER_NUM	1
-#define IRQ_INTSR(n)	17	/* UART reception completion */
-#define IRQ_INTSR_NUM	1
-#define IRQ_INTST(n)	18	/* UART transmission completion */
-#define IRQ_INTST_NUM	1
-
-/* For <asm/irq.h> */
-#define NUM_CPU_IRQS	64
-
-
-/* TEG UART details.  */
-#define V850E_UART_BASE_ADDR(n)		(0xFFFFF600 + 0x10 * (n))
-#define V850E_UART_ASIM_ADDR(n)		(V850E_UART_BASE_ADDR(n) + 0x0)
-#define V850E_UART_ASIS_ADDR(n)		(V850E_UART_BASE_ADDR(n) + 0x2)
-#define V850E_UART_ASIF_ADDR(n)		(V850E_UART_BASE_ADDR(n) + 0x4)
-#define V850E_UART_CKSR_ADDR(n)		(V850E_UART_BASE_ADDR(n) + 0x6)
-#define V850E_UART_BRGC_ADDR(n)		(V850E_UART_BASE_ADDR(n) + 0x8)
-#define V850E_UART_TXB_ADDR(n)		(V850E_UART_BASE_ADDR(n) + 0xA)
-#define V850E_UART_RXB_ADDR(n)		(V850E_UART_BASE_ADDR(n) + 0xC)
-#define V850E_UART_NUM_CHANNELS		1
-#define V850E_UART_BASE_FREQ		CPU_CLOCK_FREQ
-/* This is a function that gets called before configuring the UART.  */
-#define V850E_UART_PRE_CONFIGURE	teg_uart_pre_configure
-#ifndef __ASSEMBLY__
-extern void teg_uart_pre_configure (unsigned chan,
-				    unsigned cflags, unsigned baud);
-#endif
-
-
-/* The TEG RTPU.  */
-#define V850E_RTPU_BASE_ADDR		0xFFFFF210
-
-
-/* TEG series timer D details.  */
-#define V850E_TIMER_D_BASE_ADDR		0xFFFFF210
-#define V850E_TIMER_D_TMCD_BASE_ADDR 	(V850E_TIMER_D_BASE_ADDR + 0x0)
-#define V850E_TIMER_D_TMD_BASE_ADDR 	(V850E_TIMER_D_BASE_ADDR + 0x4)
-#define V850E_TIMER_D_CMD_BASE_ADDR 	(V850E_TIMER_D_BASE_ADDR + 0x8)
-#define V850E_TIMER_D_BASE_FREQ		CPU_CLOCK_FREQ
-
-
-/* `Interrupt Source Select' control register.  */
-#define TEG_ISS_ADDR			0xFFFFF7FA
-#define TEG_ISS				(*(volatile u8 *)TEG_ISS_ADDR)
-
-/* Port 0 I/O register (bits 0-3 used).  */
-#define TEG_PORT0_IO_ADDR		0xFFFFF7F2
-#define TEG_PORT0_IO			(*(volatile u8 *)TEG_PORT0_IO_ADDR)
-/* Port 0 control register (bits 0-3 control mode, 0 = output, 1 = input).  */
-#define TEG_PORT0_PM_ADDR		0xFFFFF7F4
-#define TEG_PORT0_PM			(*(volatile u8 *)TEG_PORT0_PM_ADDR)
-
-
-#ifndef __ASSEMBLY__
-extern void teg_init_irqs (void);
-#endif
-
-
-#endif /* __V850_TEG_H__ */
diff --git a/include/asm-v850/termbits.h b/include/asm-v850/termbits.h
deleted file mode 100644
index 295d7bf69451..000000000000
--- a/include/asm-v850/termbits.h
+++ /dev/null
@@ -1,200 +0,0 @@
-#ifndef __V850_TERMBITS_H__
-#define __V850_TERMBITS_H__
-
-#include <linux/posix_types.h>
-
-typedef unsigned char	cc_t;
-typedef unsigned int	speed_t;
-typedef unsigned int	tcflag_t;
-
-#define NCCS 19
-struct termios {
-	tcflag_t c_iflag;		/* input mode flags */
-	tcflag_t c_oflag;		/* output mode flags */
-	tcflag_t c_cflag;		/* control mode flags */
-	tcflag_t c_lflag;		/* local mode flags */
-	cc_t c_line;			/* line discipline */
-	cc_t c_cc[NCCS];		/* control characters */
-};
-
-struct termios2 {
-	tcflag_t c_iflag;		/* input mode flags */
-	tcflag_t c_oflag;		/* output mode flags */
-	tcflag_t c_cflag;		/* control mode flags */
-	tcflag_t c_lflag;		/* local mode flags */
-	cc_t c_line;			/* line discipline */
-	cc_t c_cc[NCCS];		/* control characters */
-	speed_t c_ispeed;		/* input speed */
-	speed_t c_ospeed;		/* output speed */
-};
-
-struct ktermios {
-	tcflag_t c_iflag;		/* input mode flags */
-	tcflag_t c_oflag;		/* output mode flags */
-	tcflag_t c_cflag;		/* control mode flags */
-	tcflag_t c_lflag;		/* local mode flags */
-	cc_t c_line;			/* line discipline */
-	cc_t c_cc[NCCS];		/* control characters */
-	speed_t c_ispeed;		/* input speed */
-	speed_t c_ospeed;		/* output speed */
-};
-
-/* c_cc characters */
-#define VINTR 0
-#define VQUIT 1
-#define VERASE 2
-#define VKILL 3
-#define VEOF 4
-#define VTIME 5
-#define VMIN 6
-#define VSWTC 7
-#define VSTART 8
-#define VSTOP 9
-#define VSUSP 10
-#define VEOL 11
-#define VREPRINT 12
-#define VDISCARD 13
-#define VWERASE 14
-#define VLNEXT 15
-#define VEOL2 16
-
-
-/* c_iflag bits */
-#define IGNBRK	0000001
-#define BRKINT	0000002
-#define IGNPAR	0000004
-#define PARMRK	0000010
-#define INPCK	0000020
-#define ISTRIP	0000040
-#define INLCR	0000100
-#define IGNCR	0000200
-#define ICRNL	0000400
-#define IUCLC	0001000
-#define IXON	0002000
-#define IXANY	0004000
-#define IXOFF	0010000
-#define IMAXBEL	0020000
-#define IUTF8	0040000
-
-/* c_oflag bits */
-#define OPOST	0000001
-#define OLCUC	0000002
-#define ONLCR	0000004
-#define OCRNL	0000010
-#define ONOCR	0000020
-#define ONLRET	0000040
-#define OFILL	0000100
-#define OFDEL	0000200
-#define NLDLY	0000400
-#define   NL0	0000000
-#define   NL1	0000400
-#define CRDLY	0003000
-#define   CR0	0000000
-#define   CR1	0001000
-#define   CR2	0002000
-#define   CR3	0003000
-#define TABDLY	0014000
-#define   TAB0	0000000
-#define   TAB1	0004000
-#define   TAB2	0010000
-#define   TAB3	0014000
-#define   XTABS	0014000
-#define BSDLY	0020000
-#define   BS0	0000000
-#define   BS1	0020000
-#define VTDLY	0040000
-#define   VT0	0000000
-#define   VT1	0040000
-#define FFDLY	0100000
-#define   FF0	0000000
-#define   FF1	0100000
-
-/* c_cflag bit meaning */
-#define CBAUD	0010017
-#define  B0	0000000		/* hang up */
-#define  B50	0000001
-#define  B75	0000002
-#define  B110	0000003
-#define  B134	0000004
-#define  B150	0000005
-#define  B200	0000006
-#define  B300	0000007
-#define  B600	0000010
-#define  B1200	0000011
-#define  B1800	0000012
-#define  B2400	0000013
-#define  B4800	0000014
-#define  B9600	0000015
-#define  B19200	0000016
-#define  B38400	0000017
-#define EXTA B19200
-#define EXTB B38400
-#define CSIZE	0000060
-#define   CS5	0000000
-#define   CS6	0000020
-#define   CS7	0000040
-#define   CS8	0000060
-#define CSTOPB	0000100
-#define CREAD	0000200
-#define PARENB	0000400
-#define PARODD	0001000
-#define HUPCL	0002000
-#define CLOCAL	0004000
-#define CBAUDEX 0010000
-#define    BOTHER 0010000
-#define    B57600 0010001
-#define   B115200 0010002
-#define   B230400 0010003
-#define   B460800 0010004
-#define   B500000 0010005
-#define   B576000 0010006
-#define   B921600 0010007
-#define  B1000000 0010010
-#define  B1152000 0010011
-#define  B1500000 0010012
-#define  B2000000 0010013
-#define  B2500000 0010014
-#define  B3000000 0010015
-#define  B3500000 0010016
-#define  B4000000 0010017
-#define CIBAUD	  002003600000		/* input baud rate */
-#define CMSPAR	  010000000000		/* mark or space (stick) parity */
-#define CRTSCTS	  020000000000		/* flow control */
-
-#define	IBSHIFT	16		/* Shifr from CBAUD to CIBAUD */
-
-/* c_lflag bits */
-#define ISIG	0000001
-#define ICANON	0000002
-#define XCASE	0000004
-#define ECHO	0000010
-#define ECHOE	0000020
-#define ECHOK	0000040
-#define ECHONL	0000100
-#define NOFLSH	0000200
-#define TOSTOP	0000400
-#define ECHOCTL	0001000
-#define ECHOPRT	0002000
-#define ECHOKE	0004000
-#define FLUSHO	0010000
-#define PENDIN	0040000
-#define IEXTEN	0100000
-
-
-/* tcflow() and TCXONC use these */
-#define	TCOOFF		0
-#define	TCOON		1
-#define	TCIOFF		2
-#define	TCION		3
-
-/* tcflush() and TCFLSH use these */
-#define	TCIFLUSH	0
-#define	TCOFLUSH	1
-#define	TCIOFLUSH	2
-
-/* tcsetattr uses these */
-#define	TCSANOW		0
-#define	TCSADRAIN	1
-#define	TCSAFLUSH	2
-
-#endif /* __V850_TERMBITS_H__ */
diff --git a/include/asm-v850/termios.h b/include/asm-v850/termios.h
deleted file mode 100644
index fcd171838d9c..000000000000
--- a/include/asm-v850/termios.h
+++ /dev/null
@@ -1,90 +0,0 @@
-#ifndef __V850_TERMIOS_H__
-#define __V850_TERMIOS_H__
-
-#include <asm/termbits.h>
-#include <asm/ioctls.h>
-
-struct winsize {
-	unsigned short ws_row;
-	unsigned short ws_col;
-	unsigned short ws_xpixel;
-	unsigned short ws_ypixel;
-};
-
-#define NCC 8
-struct termio {
-	unsigned short c_iflag;		/* input mode flags */
-	unsigned short c_oflag;		/* output mode flags */
-	unsigned short c_cflag;		/* control mode flags */
-	unsigned short c_lflag;		/* local mode flags */
-	unsigned char c_line;		/* line discipline */
-	unsigned char c_cc[NCC];	/* control characters */
-};
-
-/* modem lines */
-#define TIOCM_LE	0x001
-#define TIOCM_DTR	0x002
-#define TIOCM_RTS	0x004
-#define TIOCM_ST	0x008
-#define TIOCM_SR	0x010
-#define TIOCM_CTS	0x020
-#define TIOCM_CAR	0x040
-#define TIOCM_RNG	0x080
-#define TIOCM_DSR	0x100
-#define TIOCM_CD	TIOCM_CAR
-#define TIOCM_RI	TIOCM_RNG
-#define TIOCM_OUT1	0x2000
-#define TIOCM_OUT2	0x4000
-#define TIOCM_LOOP	0x8000
-
-/* ioctl (fd, TIOCSERGETLSR, &result) where result may be as below */
-
-#ifdef __KERNEL__
-
-/*	intr=^C		quit=^\		erase=del	kill=^U
-	eof=^D		vtime=\0	vmin=\1		sxtc=\0
-	start=^Q	stop=^S		susp=^Z		eol=\0
-	reprint=^R	discard=^U	werase=^W	lnext=^V
-	eol2=\0
-*/
-#define INIT_C_CC "\003\034\177\025\004\0\1\0\021\023\032\0\022\017\027\026\0"
-
-/*
- * Translate a "termio" structure into a "termios". Ugh.
- */
-#define SET_LOW_TERMIOS_BITS(termios, termio, x) { \
-	unsigned short __tmp; \
-	get_user(__tmp,&(termio)->x); \
-	*(unsigned short *) &(termios)->x = __tmp; \
-}
-
-#define user_termio_to_kernel_termios(termios, termio) \
-({ \
-	SET_LOW_TERMIOS_BITS(termios, termio, c_iflag); \
-	SET_LOW_TERMIOS_BITS(termios, termio, c_oflag); \
-	SET_LOW_TERMIOS_BITS(termios, termio, c_cflag); \
-	SET_LOW_TERMIOS_BITS(termios, termio, c_lflag); \
-	copy_from_user((termios)->c_cc, (termio)->c_cc, NCC); \
-})
-
-/*
- * Translate a "termios" structure into a "termio". Ugh.
- */
-#define kernel_termios_to_user_termio(termio, termios) \
-({ \
-	put_user((termios)->c_iflag, &(termio)->c_iflag); \
-	put_user((termios)->c_oflag, &(termio)->c_oflag); \
-	put_user((termios)->c_cflag, &(termio)->c_cflag); \
-	put_user((termios)->c_lflag, &(termio)->c_lflag); \
-	put_user((termios)->c_line,  &(termio)->c_line); \
-	copy_to_user((termio)->c_cc, (termios)->c_cc, NCC); \
-})
-
-#define user_termios_to_kernel_termios(k, u) copy_from_user(k, u, sizeof(struct termios2))
-#define kernel_termios_to_user_termios(u, k) copy_to_user(u, k, sizeof(struct termios2))
-#define user_termios_to_kernel_termios_1(k, u) copy_from_user(k, u, sizeof(struct termios))
-#define kernel_termios_to_user_termios_1(u, k) copy_to_user(u, k, sizeof(struct termios))
-
-#endif	/* __KERNEL__ */
-
-#endif	/* __V850_TERMIOS_H__ */
diff --git a/include/asm-v850/thread_info.h b/include/asm-v850/thread_info.h
deleted file mode 100644
index 1a9e6ae0c5fd..000000000000
--- a/include/asm-v850/thread_info.h
+++ /dev/null
@@ -1,129 +0,0 @@
-/*
- * include/asm-v850/thread_info.h -- v850 low-level thread information
- *
- *  Copyright (C) 2002  NEC Corporation
- *  Copyright (C) 2002  Miles Bader <miles@gnu.org>
- *  Copyright (C) 2002  David Howells (dhowells@redhat.com)
- *    - Incorporating suggestions made by Linus Torvalds and Dave Miller
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * This file was derived from the PPC version, include/asm-ppc/thread_info.h
- * which was adapted from the i386 version by Paul Mackerras
- */
-
-#ifndef __V850_THREAD_INFO_H__
-#define __V850_THREAD_INFO_H__
-
-#ifdef __KERNEL__
-
-#ifndef __ASSEMBLY__
-
-/*
- * low level task data.
- * If you change this, change the TI_* offsets below to match.
- */
-struct thread_info {
-	struct task_struct	*task;		/* main task structure */
-	struct exec_domain	*exec_domain;	/* execution domain */
-	unsigned long		flags;		/* low level flags */
-	int			cpu;		/* cpu we're on */
-	int			preempt_count;	/* 0 => preemptable,
-						   <0 => BUG */
-	struct restart_block	restart_block;
-};
-
-#define INIT_THREAD_INFO(tsk)						      \
-{									      \
-	.task =		&tsk,						      \
-	.exec_domain =	&default_exec_domain,				      \
-	.flags =	0,						      \
-	.cpu =		0,						      \
-	.preempt_count = 1,						      \
-	.restart_block = {						      \
-		.fn = do_no_restart_syscall,				      \
-	},								      \
-}
-
-#define init_thread_info	(init_thread_union.thread_info)
-#define init_stack		(init_thread_union.stack)
-
-/*
- * macros/functions for gaining access to the thread information structure
- */
-
-/* thread information allocation */
-#define alloc_thread_info(tsk) ((struct thread_info *) \
-				__get_free_pages(GFP_KERNEL, 1))
-#define free_thread_info(ti)	free_pages((unsigned long) (ti), 1)
-
-#endif /* __ASSEMBLY__ */
-
-
-/*
- * Offsets in thread_info structure, used in assembly code
- */
-#define TI_TASK		0
-#define TI_EXECDOMAIN	4
-#define TI_FLAGS	8
-#define TI_CPU		12
-#define TI_PREEMPT	16
-
-#define PREEMPT_ACTIVE		0x4000000
-
-/*
- * thread information flag bit numbers
- */
-#define TIF_SYSCALL_TRACE	0	/* syscall trace active */
-#define TIF_SIGPENDING		1	/* signal pending */
-#define TIF_NEED_RESCHED	2	/* rescheduling necessary */
-#define TIF_POLLING_NRFLAG	3	/* true if poll_idle() is polling
-					   TIF_NEED_RESCHED */
-#define TIF_MEMDIE		4
-
-/* as above, but as bit values */
-#define _TIF_SYSCALL_TRACE	(1<<TIF_SYSCALL_TRACE)
-#define _TIF_SIGPENDING		(1<<TIF_SIGPENDING)
-#define _TIF_NEED_RESCHED	(1<<TIF_NEED_RESCHED)
-#define _TIF_POLLING_NRFLAG	(1<<TIF_POLLING_NRFLAG)
-
-
-/* Size of kernel stack for each process.  */
-#define THREAD_SIZE		0x2000
-
-/* The alignment of kernel threads, with thread_info structures at their
-   base.  Thus, a pointer for a task's task structure can be derived from
-   its kernel stack pointer.  */
-#define THREAD_ALIGNMENT	THREAD_SIZE
-#define THREAD_MASK		(-THREAD_ALIGNMENT)
-
-
-#ifdef __ASSEMBLY__
-
-/* Put a pointer to the current thread_info structure into REG.  Note that
-   this definition requires THREAD_MASK to be representable as a signed
-   16-bit value.  */
-#define GET_CURRENT_THREAD(reg)						\
-        /* Use `addi' and then `and' instead of just `andi', because	\
-	   `addi' sign-extends the immediate value, whereas `andi'	\
-	   zero-extends it.  */						\
-	addi	THREAD_MASK, r0, reg;					\
-	and	sp, reg
-
-#else
-
-/* Return a pointer to the current thread_info structure.  */
-static inline struct thread_info *current_thread_info (void)
-{
-	register unsigned long sp __asm__ ("sp");
-	return (struct thread_info *)(sp & THREAD_MASK);
-}
-
-#endif /* __ASSEMBLY__ */
-
-
-#endif /* __KERNEL__ */
-
-#endif /* __V850_THREAD_INFO_H__ */
diff --git a/include/asm-v850/timex.h b/include/asm-v850/timex.h
deleted file mode 100644
index 6279e5a0ee8e..000000000000
--- a/include/asm-v850/timex.h
+++ /dev/null
@@ -1,18 +0,0 @@
-/*
- * linux/include/asm-v850/timex.h
- *
- * v850 architecture timex specifications
- */
-#ifndef __V850_TIMEX_H__
-#define __V850_TIMEX_H__
-
-#define CLOCK_TICK_RATE	1193180 /* Underlying HZ */
-
-typedef unsigned long cycles_t;
-
-static inline cycles_t get_cycles(void)
-{
-	return 0;
-}
-
-#endif /* __V850_TIMEX_H__ */
diff --git a/include/asm-v850/tlb.h b/include/asm-v850/tlb.h
deleted file mode 100644
index 73bc9ead40dd..000000000000
--- a/include/asm-v850/tlb.h
+++ /dev/null
@@ -1,21 +0,0 @@
-/*
- * include/asm-v850/tlb.h
- *
- *  Copyright (C) 2002  NEC Corporation
- *  Copyright (C) 2002  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_TLB_H__
-#define __V850_TLB_H__
-
-#define tlb_flush(tlb)	((void)0)
-
-#include <asm-generic/tlb.h>
-
-#endif /* __V850_TLB_H__ */
diff --git a/include/asm-v850/tlbflush.h b/include/asm-v850/tlbflush.h
deleted file mode 100644
index c44aa64449c8..000000000000
--- a/include/asm-v850/tlbflush.h
+++ /dev/null
@@ -1,64 +0,0 @@
-/*
- * include/asm-v850/tlbflush.h
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_TLBFLUSH_H__
-#define __V850_TLBFLUSH_H__
-
-#include <asm/machdep.h>
-
-
-/*
- * flush all user-space atc entries.
- */
-static inline void __flush_tlb(void)
-{
-	BUG ();
-}
-
-static inline void __flush_tlb_one(unsigned long addr)
-{
-	BUG ();
-}
-
-#define flush_tlb() __flush_tlb()
-
-/*
- * flush all atc entries (both kernel and user-space entries).
- */
-static inline void flush_tlb_all(void)
-{
-	BUG ();
-}
-
-static inline void flush_tlb_mm(struct mm_struct *mm)
-{
-	BUG ();
-}
-
-static inline void flush_tlb_page(struct vm_area_struct *vma, unsigned long addr)
-{
-	BUG ();
-}
-
-static inline void flush_tlb_range(struct vm_area_struct *vma,
-				   unsigned long start, unsigned long end)
-{
-	BUG ();
-}
-
-static inline void flush_tlb_kernel_page(unsigned long addr)
-{
-	BUG ();
-}
-
-#endif /* __V850_TLBFLUSH_H__ */
diff --git a/include/asm-v850/topology.h b/include/asm-v850/topology.h
deleted file mode 100644
index 6040e41d7945..000000000000
--- a/include/asm-v850/topology.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __V850_TOPOLOGY_H__
-#define __V850_TOPOLOGY_H__
-
-#include <asm-generic/topology.h>
-
-#endif /* __V850_TOPOLOGY_H__ */
diff --git a/include/asm-v850/types.h b/include/asm-v850/types.h
deleted file mode 100644
index 89f735ee41dd..000000000000
--- a/include/asm-v850/types.h
+++ /dev/null
@@ -1,36 +0,0 @@
-#ifndef __V850_TYPES_H__
-#define __V850_TYPES_H__
-
-#ifndef __ASSEMBLY__
-
-/*
- * This file is never included by application software unless
- * explicitly requested (e.g., via linux/types.h) in which case the
- * application is Linux specific so (user-) name space pollution is
- * not a major issue.  However, for interoperability, libraries still
- * need to be careful to avoid a name clashes.
- */
-#include <asm-generic/int-ll64.h>
-
-typedef unsigned short umode_t;
-
-#endif /* !__ASSEMBLY__ */
-
-/*
- * These aren't exported outside the kernel to avoid name space clashes
- */
-#ifdef __KERNEL__
-
-#define BITS_PER_LONG 32
-
-#ifndef __ASSEMBLY__
-
-/* Dma addresses are 32-bits wide.  */
-
-typedef u32 dma_addr_t;
-
-#endif /* !__ASSEMBLY__ */
-
-#endif /* __KERNEL__ */
-
-#endif /* __V850_TYPES_H__ */
diff --git a/include/asm-v850/uaccess.h b/include/asm-v850/uaccess.h
deleted file mode 100644
index 64563c409bb2..000000000000
--- a/include/asm-v850/uaccess.h
+++ /dev/null
@@ -1,159 +0,0 @@
-#ifndef __V850_UACCESS_H__
-#define __V850_UACCESS_H__
-
-/*
- * User space memory access functions
- */
-
-#include <linux/errno.h>
-#include <linux/string.h>
-
-#include <asm/segment.h>
-#include <asm/machdep.h>
-
-#define VERIFY_READ	0
-#define VERIFY_WRITE	1
-
-static inline int access_ok (int type, const void *addr, unsigned long size)
-{
-	/* XXX I guess we should check against real ram bounds at least, and
-	   possibly make sure ADDR is not within the kernel.
-	   For now we just check to make sure it's not a small positive
-	   or negative value, as that will at least catch some kinds of
-	   error.  In particular, we make sure that ADDR's not within the
-	   interrupt vector area, which we know starts at zero, or within the
-	   peripheral-I/O area, which is located just _before_ zero.  */
-	unsigned long val = (unsigned long)addr;
-	return val >= (0x80 + NUM_CPU_IRQS*16) && val < 0xFFFFF000;
-}
-
-/*
- * The exception table consists of pairs of addresses: the first is the
- * address of an instruction that is allowed to fault, and the second is
- * the address at which the program should continue.  No registers are
- * modified, so it is entirely up to the continuation code to figure out
- * what to do.
- *
- * All the routines below use bits of fixup code that are out of line
- * with the main instruction path.  This means when everything is well,
- * we don't even have to jump over them.  Further, they do not intrude
- * on our cache or tlb entries.
- */
-
-struct exception_table_entry
-{
-	unsigned long insn, fixup;
-};
-
-/* Returns 0 if exception not found and fixup otherwise.  */
-extern unsigned long search_exception_table (unsigned long);
-
-
-/*
- * These are the main single-value transfer routines.  They automatically
- * use the right size if we just have the right pointer type.
- */
-
-extern int bad_user_access_length (void);
-
-#define __get_user(var, ptr)						      \
-  ({									      \
-	  int __gu_err = 0;						      \
-	  typeof(*(ptr)) __gu_val = 0;					      \
-	  switch (sizeof (*(ptr))) {					      \
-	  case 1:							      \
-	  case 2:							      \
-	  case 4:							      \
-		  __gu_val = *(ptr);					      \
-		  break;						      \
-	  case 8:							      \
-		  memcpy(&__gu_val, ptr, sizeof(__gu_val));		      \
-		  break;						      \
-	  default:							      \
-		  __gu_val = 0;						      \
-		  __gu_err = __get_user_bad ();				      \
-		  break;						      \
-	  }								      \
-	  (var) = __gu_val;						      \
-	  __gu_err;							      \
-  })
-#define __get_user_bad()	(bad_user_access_length (), (-EFAULT))
-
-#define __put_user(var, ptr)						      \
-  ({									      \
-	  int __pu_err = 0;						      \
-	  switch (sizeof (*(ptr))) {					      \
-	  case 1:							      \
-	  case 2:							      \
-	  case 4:							      \
-		  *(ptr) = (var);					      \
-		  break;						      \
-	  case 8: {							      \
-	  	  typeof(*(ptr)) __pu_val = 0;				      \
-		  memcpy(ptr, &__pu_val, sizeof(__pu_val));		      \
-		  }							      \
-		  break;						      \
-	  default:							      \
-		  __pu_err = __put_user_bad ();				      \
-		  break;						      \
-	  }								      \
-	  __pu_err;							      \
-  })
-#define __put_user_bad()	(bad_user_access_length (), (-EFAULT))
-
-#define put_user(x, ptr)	__put_user(x, ptr)
-#define get_user(x, ptr)	__get_user(x, ptr)
-
-#define __copy_from_user(to, from, n)	(memcpy (to, from, n), 0)
-#define __copy_to_user(to, from, n)	(memcpy(to, from, n), 0)
-
-#define __copy_to_user_inatomic __copy_to_user
-#define __copy_from_user_inatomic __copy_from_user
-
-#define copy_from_user(to, from, n)	__copy_from_user (to, from, n)
-#define copy_to_user(to, from, n) 	__copy_to_user(to, from, n)
-
-#define copy_to_user_ret(to,from,n,retval) \
-  ({ if (copy_to_user (to,from,n)) return retval; })
-
-#define copy_from_user_ret(to,from,n,retval) \
-  ({ if (copy_from_user (to,from,n)) return retval; })
-
-/*
- * Copy a null terminated string from userspace.
- */
-
-static inline long
-strncpy_from_user (char *dst, const char *src, long count)
-{
-	char *tmp;
-	strncpy (dst, src, count);
-	for (tmp = dst; *tmp && count > 0; tmp++, count--)
-		;
-	return tmp - dst;
-}
-
-/*
- * Return the size of a string (including the ending 0)
- *
- * Return 0 on exception, a value greater than N if too long
- */
-static inline long strnlen_user (const char *src, long n)
-{
-	return strlen (src) + 1;
-}
-
-#define strlen_user(str)	strnlen_user (str, 32767)
-
-/*
- * Zero Userspace
- */
-
-static inline unsigned long
-clear_user (void *to, unsigned long n)
-{
-	memset (to, 0, n);
-	return 0;
-}
-
-#endif /* __V850_UACCESS_H__ */
diff --git a/include/asm-v850/ucontext.h b/include/asm-v850/ucontext.h
deleted file mode 100644
index 303c21590cff..000000000000
--- a/include/asm-v850/ucontext.h
+++ /dev/null
@@ -1,14 +0,0 @@
-#ifndef __V850_UCONTEXT_H__
-#define __V850_UCONTEXT_H__
-
-#include <asm/sigcontext.h>
-
-struct ucontext {
-	unsigned long	  uc_flags;
-	struct ucontext  *uc_link;
-	stack_t		  uc_stack;
-	struct sigcontext uc_mcontext;
-	sigset_t	  uc_sigmask;	/* mask last for extensibility */
-};
-
-#endif /* __V850_UCONTEXT_H__ */
diff --git a/include/asm-v850/unaligned.h b/include/asm-v850/unaligned.h
deleted file mode 100644
index 53122b28491e..000000000000
--- a/include/asm-v850/unaligned.h
+++ /dev/null
@@ -1,22 +0,0 @@
-/*
- *  Copyright (C) 2001  NEC Corporation
- *  Copyright (C) 2001  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Note that some v850 chips support unaligned access, but it seems too
- * annoying to use.
- */
-#ifndef _ASM_V850_UNALIGNED_H
-#define _ASM_V850_UNALIGNED_H
-
-#include <linux/unaligned/be_byteshift.h>
-#include <linux/unaligned/le_byteshift.h>
-#include <linux/unaligned/generic.h>
-
-#define get_unaligned	__get_unaligned_le
-#define put_unaligned	__put_unaligned_le
-
-#endif /* _ASM_V850_UNALIGNED_H */
diff --git a/include/asm-v850/unistd.h b/include/asm-v850/unistd.h
deleted file mode 100644
index 2241ed45ecfe..000000000000
--- a/include/asm-v850/unistd.h
+++ /dev/null
@@ -1,244 +0,0 @@
-/*
- * include/asm-v850/unistd.h -- System call numbers and invocation mechanism
- *
- *  Copyright (C) 2001,02,03,04  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03,04  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_UNISTD_H__
-#define __V850_UNISTD_H__
-
-#define __NR_restart_syscall	  0
-#define __NR_exit		  1
-#define __NR_fork		  2
-#define __NR_read		  3
-#define __NR_write		  4
-#define __NR_open		  5
-#define __NR_close		  6
-#define __NR_waitpid		  7
-#define __NR_creat		  8
-#define __NR_link		  9
-#define __NR_unlink		 10
-#define __NR_execve		 11
-#define __NR_chdir		 12
-#define __NR_time		 13
-#define __NR_mknod		 14
-#define __NR_chmod		 15
-#define __NR_chown		 16
-#define __NR_break		 17
-#define __NR_lseek		 19
-#define __NR_getpid		 20
-#define __NR_mount		 21
-#define __NR_umount		 22
-#define __NR_setuid		 23
-#define __NR_getuid		 24
-#define __NR_stime		 25
-#define __NR_ptrace		 26
-#define __NR_alarm		 27
-#define __NR_pause		 29
-#define __NR_utime		 30
-#define __NR_stty		 31
-#define __NR_gtty		 32
-#define __NR_access		 33
-#define __NR_nice		 34
-#define __NR_ftime		 35
-#define __NR_sync		 36
-#define __NR_kill		 37
-#define __NR_rename		 38
-#define __NR_mkdir		 39
-#define __NR_rmdir		 40
-#define __NR_dup		 41
-#define __NR_pipe		 42
-#define __NR_times		 43
-#define __NR_prof		 44
-#define __NR_brk		 45
-#define __NR_setgid		 46
-#define __NR_getgid		 47
-#define __NR_signal		 48
-#define __NR_geteuid		 49
-#define __NR_getegid		 50
-#define __NR_acct		 51
-#define __NR_umount2		 52
-#define __NR_lock		 53
-#define __NR_ioctl		 54
-#define __NR_fcntl		 55
-#define __NR_setpgid		 57
-#define __NR_umask		 60
-#define __NR_chroot		 61
-#define __NR_ustat		 62
-#define __NR_dup2		 63
-#define __NR_getppid		 64
-#define __NR_getpgrp		 65
-#define __NR_setsid		 66
-#define __NR_sigaction		 67
-#define __NR_sgetmask		 68
-#define __NR_ssetmask		 69
-#define __NR_setreuid		 70
-#define __NR_setregid		 71
-#define __NR_sigsuspend		 72
-#define __NR_sigpending		 73
-#define __NR_sethostname	 74
-#define __NR_setrlimit		 75
-#define __NR_ugetrlimit	 	 76
-#define __NR_getrusage		 77
-#define __NR_gettimeofday	 78
-#define __NR_settimeofday	 79
-#define __NR_getgroups		 80
-#define __NR_setgroups		 81
-#define __NR_select		 82
-#define __NR_symlink		 83
-#define __NR_readlink		 85
-#define __NR_uselib		 86
-#define __NR_swapon		 87
-#define __NR_reboot		 88
-#define __NR_readdir		 89
-#define __NR_mmap		 90
-#define __NR_munmap		 91
-#define __NR_truncate		 92
-#define __NR_ftruncate		 93
-#define __NR_fchmod		 94
-#define __NR_fchown		 95
-#define __NR_getpriority	 96
-#define __NR_setpriority	 97
-#define __NR_profil		 98
-#define __NR_statfs		 99
-#define __NR_fstatfs		100
-#define __NR_socketcall		102
-#define __NR_syslog		103
-#define __NR_setitimer		104
-#define __NR_getitimer		105
-#define __NR_stat		106
-#define __NR_lstat		107
-#define __NR_fstat		108
-#define __NR_vhangup		111
-#define __NR_wait4		114
-#define __NR_swapoff		115
-#define __NR_sysinfo		116
-#define __NR_ipc		117
-#define __NR_fsync		118
-#define __NR_sigreturn		119
-#define __NR_clone		120
-#define __NR_setdomainname	121
-#define __NR_uname		122
-#define __NR_cacheflush		123
-#define __NR_adjtimex		124
-#define __NR_mprotect		125
-#define __NR_sigprocmask	126
-#define __NR_create_module	127
-#define __NR_init_module	128
-#define __NR_delete_module	129
-#define __NR_get_kernel_syms	130
-#define __NR_quotactl		131
-#define __NR_getpgid		132
-#define __NR_fchdir		133
-#define __NR_bdflush		134
-#define __NR_sysfs		135
-#define __NR_personality	136
-#define __NR_afs_syscall	137 /* Syscall for Andrew File System */
-#define __NR_setfsuid		138
-#define __NR_setfsgid		139
-#define __NR__llseek		140
-#define __NR_getdents		141
-#define __NR_flock		143
-#define __NR_msync		144
-#define __NR_readv		145
-#define __NR_writev		146
-#define __NR_getsid		147
-#define __NR_fdatasync		148
-#define __NR__sysctl		149
-#define __NR_mlock		150
-#define __NR_munlock		151
-#define __NR_mlockall		152
-#define __NR_munlockall		153
-#define __NR_sched_setparam		154
-#define __NR_sched_getparam		155
-#define __NR_sched_setscheduler		156
-#define __NR_sched_getscheduler		157
-#define __NR_sched_yield		158
-#define __NR_sched_get_priority_max	159
-#define __NR_sched_get_priority_min	160
-#define __NR_sched_rr_get_interval	161
-#define __NR_nanosleep		162
-#define __NR_mremap		163
-#define __NR_setresuid		164
-#define __NR_getresuid		165
-#define __NR_query_module	167
-#define __NR_poll		168
-#define __NR_nfsservctl		169
-#define __NR_setresgid		170
-#define __NR_getresgid		171
-#define __NR_prctl		172
-#define __NR_rt_sigreturn	173
-#define __NR_rt_sigaction	174
-#define __NR_rt_sigprocmask	175
-#define __NR_rt_sigpending	176
-#define __NR_rt_sigtimedwait	177
-#define __NR_rt_sigqueueinfo	178
-#define __NR_rt_sigsuspend	179
-#define __NR_pread		180
-#define __NR_pwrite		181
-#define __NR_lchown		182
-#define __NR_getcwd		183
-#define __NR_capget		184
-#define __NR_capset		185
-#define __NR_sigaltstack	186
-#define __NR_sendfile		187
-#define __NR_getpmsg		188	/* some people actually want streams */
-#define __NR_putpmsg		189	/* some people actually want streams */
-#define __NR_vfork		190
-#define __NR_mmap2		192
-#define __NR_truncate64		193
-#define __NR_ftruncate64	194
-#define __NR_stat64		195
-#define __NR_lstat64		196
-#define __NR_fstat64		197
-#define __NR_fcntl64		198
-#define __NR_getdents64		199
-#define __NR_pivot_root		200
-#define __NR_gettid		201
-#define __NR_tkill		202
-
-#ifdef __KERNEL__
-
-#define __ARCH_WANT_IPC_PARSE_VERSION
-#define __ARCH_WANT_OLD_READDIR
-#define __ARCH_WANT_STAT64
-#define __ARCH_WANT_SYS_ALARM
-#define __ARCH_WANT_SYS_GETHOSTNAME
-#define __ARCH_WANT_SYS_PAUSE
-#define __ARCH_WANT_SYS_SGETMASK
-#define __ARCH_WANT_SYS_SIGNAL
-#define __ARCH_WANT_SYS_TIME
-#define __ARCH_WANT_SYS_UTIME
-#define __ARCH_WANT_SYS_WAITPID
-#define __ARCH_WANT_SYS_SOCKETCALL
-#define __ARCH_WANT_SYS_FADVISE64
-#define __ARCH_WANT_SYS_GETPGRP
-#define __ARCH_WANT_SYS_LLSEEK
-#define __ARCH_WANT_SYS_NICE
-#define __ARCH_WANT_SYS_OLDUMOUNT
-#define __ARCH_WANT_SYS_SIGPENDING
-#define __ARCH_WANT_SYS_SIGPROCMASK
-#define __ARCH_WANT_SYS_RT_SIGACTION
-
-/*
- * "Conditional" syscalls
- */
-#define cond_syscall(name)						      \
-  asm (".weak\t" C_SYMBOL_STRING(name) ";"				      \
-       ".set\t" C_SYMBOL_STRING(name) "," C_SYMBOL_STRING(sys_ni_syscall))
-#if 0
-/* This doesn't work if there's a function prototype for NAME visible,
-   because the argument types probably won't match.  */
-#define cond_syscall(name)  \
-  void name (void) __attribute__ ((weak, alias ("sys_ni_syscall")));
-#endif
-
-#endif /* __KERNEL__ */
-#endif /* __V850_UNISTD_H__ */
diff --git a/include/asm-v850/user.h b/include/asm-v850/user.h
deleted file mode 100644
index 63cdc567d272..000000000000
--- a/include/asm-v850/user.h
+++ /dev/null
@@ -1,52 +0,0 @@
-#ifndef __V850_USER_H__
-#define __V850_USER_H__
-
-/* Adapted from <asm-ppc/user.h>.  */
-
-#include <linux/ptrace.h>
-#include <asm/page.h>
-
-/*
- * Core file format: The core file is written in such a way that gdb
- * can understand it and provide useful information to the user (under
- * linux we use the `trad-core' bfd, NOT the osf-core).  The file contents
- * are as follows:
- *
- *  upage: 1 page consisting of a user struct that tells gdb
- *	what is present in the file.  Directly after this is a
- *	copy of the task_struct, which is currently not used by gdb,
- *	but it may come in handy at some point.  All of the registers
- *	are stored as part of the upage.  The upage should always be
- *	only one page long.
- *  data: The data segment follows next.  We use current->end_text to
- *	current->brk to pick up all of the user variables, plus any memory
- *	that may have been sbrk'ed.  No attempt is made to determine if a
- *	page is demand-zero or if a page is totally unused, we just cover
- *	the entire range.  All of the addresses are rounded in such a way
- *	that an integral number of pages is written.
- *  stack: We need the stack information in order to get a meaningful
- *	backtrace.  We need to write the data from usp to
- *	current->start_stack, so we round each of these in order to be able
- *	to write an integer number of pages.
- */
-struct user {
-	struct pt_regs	regs;			/* entire machine state */
-	size_t		u_tsize;		/* text size (pages) */
-	size_t		u_dsize;		/* data size (pages) */
-	size_t		u_ssize;		/* stack size (pages) */
-	unsigned long	start_code;		/* text starting address */
-	unsigned long	start_data;		/* data starting address */
-	unsigned long	start_stack;		/* stack starting address */
-	long int	signal;			/* signal causing core dump */
-	unsigned long	u_ar0;			/* help gdb find registers */
-	unsigned long	magic;			/* identifies a core file */
-	char		u_comm[32];		/* user command name */
-};
-
-#define NBPG			PAGE_SIZE
-#define UPAGES			1
-#define HOST_TEXT_START_ADDR	(u.start_code)
-#define HOST_DATA_START_ADDR	(u.start_data)
-#define HOST_STACK_END_ADDR	(u.start_stack + u.u_ssize * NBPG)
-
-#endif /* __V850_USER_H__ */
diff --git a/include/asm-v850/v850e.h b/include/asm-v850/v850e.h
deleted file mode 100644
index 5a222eb5117f..000000000000
--- a/include/asm-v850/v850e.h
+++ /dev/null
@@ -1,21 +0,0 @@
-/*
- * include/asm-v850/v850e.h -- V850E CPU
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_V850E_H__
-#define __V850_V850E_H__
-
-#include <asm/v850e_intc.h>
-
-#define CPU_ARCH "v850e"
-
-#endif /* __V850_V850E_H__ */
diff --git a/include/asm-v850/v850e2.h b/include/asm-v850/v850e2.h
deleted file mode 100644
index 48680408ab7e..000000000000
--- a/include/asm-v850/v850e2.h
+++ /dev/null
@@ -1,69 +0,0 @@
-/*
- * include/asm-v850/v850e2.h -- Machine-dependent defs for V850E2 CPUs
- *
- *  Copyright (C) 2002,03  NEC Electronics Corporation
- *  Copyright (C) 2002,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_V850E2_H__
-#define __V850_V850E2_H__
-
-#include <asm/v850e_intc.h>	/* v850e-style interrupt system.  */
-
-
-#define CPU_ARCH "v850e2"
-
-
-/* Control registers.  */
-
-/* Chip area select control */ 
-#define V850E2_CSC_ADDR(n)	(0xFFFFF060 + (n) * 2)
-#define V850E2_CSC(n)		(*(volatile u16 *)V850E2_CSC_ADDR(n))
-/* I/O area select control */
-#define V850E2_BPC_ADDR		0xFFFFF064
-#define V850E2_BPC		(*(volatile u16 *)V850E2_BPC_ADDR)
-/* Bus size configuration */
-#define V850E2_BSC_ADDR		0xFFFFF066
-#define V850E2_BSC		(*(volatile u16 *)V850E2_BSC_ADDR)
-/* Endian configuration */
-#define V850E2_BEC_ADDR		0xFFFFF068
-#define V850E2_BEC		(*(volatile u16 *)V850E2_BEC_ADDR)
-/* Cache configuration */
-#define V850E2_BHC_ADDR		0xFFFFF06A
-#define V850E2_BHC		(*(volatile u16 *)V850E2_BHC_ADDR)
-/* NPB strobe-wait configuration */
-#define V850E2_VSWC_ADDR	0xFFFFF06E
-#define V850E2_VSWC		(*(volatile u16 *)V850E2_VSWC_ADDR)
-/* Bus cycle type */
-#define V850E2_BCT_ADDR(n)	(0xFFFFF480 + (n) * 2)
-#define V850E2_BCT(n)		(*(volatile u16 *)V850E2_BCT_ADDR(n))
-/* Data wait control */
-#define V850E2_DWC_ADDR(n)	(0xFFFFF484 + (n) * 2)
-#define V850E2_DWC(n)		(*(volatile u16 *)V850E2_DWC_ADDR(n))
-/* Bus cycle control */
-#define V850E2_BCC_ADDR		0xFFFFF488
-#define V850E2_BCC		(*(volatile u16 *)V850E2_BCC_ADDR)
-/* Address wait control */
-#define V850E2_ASC_ADDR		0xFFFFF48A
-#define V850E2_ASC		(*(volatile u16 *)V850E2_ASC_ADDR)
-/* Local bus sizing control */
-#define V850E2_LBS_ADDR		0xFFFFF48E
-#define V850E2_LBS		(*(volatile u16 *)V850E2_LBS_ADDR)
-/* Line buffer control */
-#define V850E2_LBC_ADDR(n)	(0xFFFFF490 + (n) * 2)
-#define V850E2_LBC(n)		(*(volatile u16 *)V850E2_LBC_ADDR(n))
-/* SDRAM configuration */
-#define V850E2_SCR_ADDR(n)	(0xFFFFF4A0 + (n) * 4)
-#define V850E2_SCR(n)		(*(volatile u16 *)V850E2_SCR_ADDR(n))
-/* SDRAM refresh cycle control */
-#define V850E2_RFS_ADDR(n)	(0xFFFFF4A2 + (n) * 4)
-#define V850E2_RFS(n)		(*(volatile u16 *)V850E2_RFS_ADDR(n))
-
-
-#endif /* __V850_V850E2_H__ */
diff --git a/include/asm-v850/v850e2_cache.h b/include/asm-v850/v850e2_cache.h
deleted file mode 100644
index 87edf0d311d5..000000000000
--- a/include/asm-v850/v850e2_cache.h
+++ /dev/null
@@ -1,75 +0,0 @@
-/*
- * include/asm-v850/v850e2_cache_cache.h -- Cache control for V850E2
- * 	cache memories
- *
- *  Copyright (C) 2003,05  NEC Electronics Corporation
- *  Copyright (C) 2003,05  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_V850E2_CACHE_H__
-#define __V850_V850E2_CACHE_H__
-
-#include <asm/types.h>
-
-
-/* Cache control registers.  */
-
-/* Bus Transaction Control */
-#define V850E2_CACHE_BTSC_ADDR	0xFFFFF070
-#define V850E2_CACHE_BTSC 	(*(volatile u16 *)V850E2_CACHE_BTSC_ADDR)
-#define V850E2_CACHE_BTSC_ICM	0x0001 /* icache enable */
-#define V850E2_CACHE_BTSC_DCM0	0x0004 /* dcache enable, bit 0 */
-#define V850E2_CACHE_BTSC_DCM1	0x0008 /* dcache enable, bit 1 */
-#define V850E2_CACHE_BTSC_DCM_WT		      /* write-through */ \
-			V850E2_CACHE_BTSC_DCM0
-#ifdef CONFIG_V850E2_V850E2S
-# define V850E2_CACHE_BTSC_DCM_WB_NO_ALLOC    /* write-back, non-alloc */ \
-			V850E2_CACHE_BTSC_DCM1	
-# define V850E2_CACHE_BTSC_DCM_WB_ALLOC	      /* write-back, non-alloc */ \
-			(V850E2_CACHE_BTSC_DCM1 | V850E2_CACHE_BTSC_DCM0)
-# define V850E2_CACHE_BTSC_ISEQ	0x0010 /* icache `address sequence mode' */
-# define V850E2_CACHE_BTSC_DSEQ	0x0020 /* dcache `address sequence mode' */
-# define V850E2_CACHE_BTSC_IRFC	0x0030
-# define V850E2_CACHE_BTSC_ILCD	0x4000
-# define V850E2_CACHE_BTSC_VABE	0x8000
-#endif /* CONFIG_V850E2_V850E2S */
-
-/* Cache operation start address register (low-bits).  */
-#define V850E2_CACHE_CADL_ADDR	0xFFFFF074
-#define V850E2_CACHE_CADL 	(*(volatile u16 *)V850E2_CACHE_CADL_ADDR)
-/* Cache operation start address register (high-bits).  */
-#define V850E2_CACHE_CADH_ADDR	0xFFFFF076
-#define V850E2_CACHE_CADH 	(*(volatile u16 *)V850E2_CACHE_CADH_ADDR)
-/* Cache operation count register.  */
-#define V850E2_CACHE_CCNT_ADDR	0xFFFFF078
-#define V850E2_CACHE_CCNT 	(*(volatile u16 *)V850E2_CACHE_CCNT_ADDR)
-/* Cache operation specification register.  */
-#define V850E2_CACHE_COPR_ADDR	0xFFFFF07A
-#define V850E2_CACHE_COPR 	(*(volatile u16 *)V850E2_CACHE_COPR_ADDR)
-#define V850E2_CACHE_COPR_STRT	0x0001 /* start cache operation */
-#define V850E2_CACHE_COPR_LBSL	0x0100 /* 0 = icache, 1 = dcache */
-#define V850E2_CACHE_COPR_WSLE	0x0200 /* operate on cache way */
-#define V850E2_CACHE_COPR_WSL(way) ((way) * 0x0400) /* way select */
-#define V850E2_CACHE_COPR_CFC(op)  ((op)  * 0x1000) /* cache function code */
-
-
-/* Size of a cache line in bytes.  */
-#define V850E2_CACHE_LINE_SIZE_BITS	4
-#define V850E2_CACHE_LINE_SIZE		(1 << V850E2_CACHE_LINE_SIZE_BITS)
-
-/* The size of each cache `way' in lines.  */
-#define V850E2_CACHE_WAY_SIZE		256
-
-
-/* For <asm/cache.h> */
-#define L1_CACHE_BYTES			V850E2_CACHE_LINE_SIZE
-#define L1_CACHE_SHIFT			V850E2_CACHE_LINE_SIZE_BITS
-
-
-#endif /* __V850_V850E2_CACHE_H__ */
diff --git a/include/asm-v850/v850e_cache.h b/include/asm-v850/v850e_cache.h
deleted file mode 100644
index aa7d7eb9da50..000000000000
--- a/include/asm-v850/v850e_cache.h
+++ /dev/null
@@ -1,48 +0,0 @@
-/*
- * include/asm-v850/v850e_cache.h -- Cache control for V850E cache memories
- *
- *  Copyright (C) 2001,03  NEC Electronics Corporation
- *  Copyright (C) 2001,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-/* This file implements cache control for the rather simple cache used on
-   some V850E CPUs, specifically the NB85E/TEG CPU-core and the V850E/ME2
-   CPU.  V850E2 processors have their own (better) cache
-   implementation.  */
-
-#ifndef __V850_V850E_CACHE_H__
-#define __V850_V850E_CACHE_H__
-
-#include <asm/types.h>
-
-
-/* Cache control registers.  */
-#define V850E_CACHE_BHC_ADDR	0xFFFFF06A
-#define V850E_CACHE_BHC		(*(volatile u16 *)V850E_CACHE_BHC_ADDR)
-#define V850E_CACHE_ICC_ADDR	0xFFFFF070
-#define V850E_CACHE_ICC		(*(volatile u16 *)V850E_CACHE_ICC_ADDR)
-#define V850E_CACHE_ISI_ADDR	0xFFFFF072
-#define V850E_CACHE_ISI		(*(volatile u16 *)V850E_CACHE_ISI_ADDR)
-#define V850E_CACHE_DCC_ADDR	0xFFFFF078
-#define V850E_CACHE_DCC		(*(volatile u16 *)V850E_CACHE_DCC_ADDR)
-
-/* Size of a cache line in bytes.  */
-#define V850E_CACHE_LINE_SIZE	16
-
-/* For <asm/cache.h> */
-#define L1_CACHE_BYTES		V850E_CACHE_LINE_SIZE
-
-
-#if defined(__KERNEL__) && !defined(__ASSEMBLY__)
-/* Set caching params via the BHC, ICC, and DCC registers.  */
-void v850e_cache_enable (u16 bhc, u16 icc, u16 dcc);
-#endif /* __KERNEL__ && !__ASSEMBLY__ */
-
-
-#endif /* __V850_V850E_CACHE_H__ */
diff --git a/include/asm-v850/v850e_intc.h b/include/asm-v850/v850e_intc.h
deleted file mode 100644
index 6fdf95708317..000000000000
--- a/include/asm-v850/v850e_intc.h
+++ /dev/null
@@ -1,133 +0,0 @@
-/*
- * include/asm-v850/v850e_intc.h -- V850E CPU interrupt controller (INTC)
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_V850E_INTC_H__
-#define __V850_V850E_INTC_H__
-
-
-/* There are 4 16-bit `Interrupt Mask Registers' located contiguously
-   starting from this base.  Each interrupt uses a single bit to
-   indicated enabled/disabled status.  */
-#define V850E_INTC_IMR_BASE_ADDR  0xFFFFF100
-#define V850E_INTC_IMR_ADDR(irq)  (V850E_INTC_IMR_BASE_ADDR + ((irq) >> 3))
-#define V850E_INTC_IMR_BIT(irq)	  ((irq) & 0x7)
-
-/* Each maskable interrupt has a single-byte control register at this
-   address.  */
-#define V850E_INTC_IC_BASE_ADDR	  0xFFFFF110
-#define V850E_INTC_IC_ADDR(irq)	  (V850E_INTC_IC_BASE_ADDR + ((irq) << 1))
-#define V850E_INTC_IC(irq)	  (*(volatile u8 *)V850E_INTC_IC_ADDR(irq))
-/* Encode priority PR for storing in an interrupt control register.  */
-#define V850E_INTC_IC_PR(pr)	  (pr)
-/* Interrupt disable bit in an interrupt control register.  */
-#define V850E_INTC_IC_MK_BIT	  6
-#define V850E_INTC_IC_MK	  (1 << V850E_INTC_IC_MK_BIT)
-/* Interrupt pending flag in an interrupt control register.  */
-#define V850E_INTC_IC_IF_BIT	  7
-#define V850E_INTC_IC_IF	  (1 << V850E_INTC_IC_IF_BIT)
-
-/* The ISPR (In-service priority register) contains one bit for each interrupt
-   priority level, which is set to one when that level is currently being
-   serviced (and thus blocking any interrupts of equal or lesser level).  */
-#define V850E_INTC_ISPR_ADDR	  0xFFFFF1FA
-#define V850E_INTC_ISPR		  (*(volatile u8 *)V850E_INTC_ISPR_ADDR)
-
-
-#ifndef __ASSEMBLY__
-
-/* Enable interrupt handling for interrupt IRQ.  */
-static inline void v850e_intc_enable_irq (unsigned irq)
-{
-	__asm__ __volatile__ ("clr1 %0, [%1]"
-			      :: "r" (V850E_INTC_IMR_BIT (irq)),
-			         "r" (V850E_INTC_IMR_ADDR (irq))
-			      : "memory");
-}
-
-/* Disable interrupt handling for interrupt IRQ.  Note that any
-   interrupts received while disabled will be delivered once the
-   interrupt is enabled again, unless they are explicitly cleared using
-   `v850e_intc_clear_pending_irq'.  */
-static inline void v850e_intc_disable_irq (unsigned irq)
-{
-	__asm__ __volatile__ ("set1 %0, [%1]"
-			      :: "r" (V850E_INTC_IMR_BIT (irq)),
-			         "r" (V850E_INTC_IMR_ADDR (irq))
-			      : "memory");
-}
-
-/* Return true if interrupt handling for interrupt IRQ is enabled.  */
-static inline int v850e_intc_irq_enabled (unsigned irq)
-{
-	int rval;
-	__asm__ __volatile__ ("tst1 %1, [%2]; setf z, %0"
-			      : "=r" (rval)
-			      : "r" (V850E_INTC_IMR_BIT (irq)),
-			        "r" (V850E_INTC_IMR_ADDR (irq)));
-	return rval;
-}
-
-/* Disable irqs from 0 until LIMIT.  LIMIT must be a multiple of 8.  */
-static inline void _v850e_intc_disable_irqs (unsigned limit)
-{
-	unsigned long addr;
-	for (addr = V850E_INTC_IMR_BASE_ADDR; limit >= 8; addr++, limit -= 8)
-		*(char *)addr = 0xFF;
-}
-
-/* Disable all irqs.  This is purposely a macro, because NUM_MACH_IRQS
-   will be only be defined later.  */
-#define v850e_intc_disable_irqs()   _v850e_intc_disable_irqs (NUM_MACH_IRQS)
-
-/* Clear any pending interrupts for IRQ.  */
-static inline void v850e_intc_clear_pending_irq (unsigned irq)
-{
-	__asm__ __volatile__ ("clr1 %0, 0[%1]"
-			      :: "i" (V850E_INTC_IC_IF_BIT),
-			         "r" (V850E_INTC_IC_ADDR (irq))
-			      : "memory");
-}
-
-/* Return true if interrupt IRQ is pending (but disabled).  */
-static inline int v850e_intc_irq_pending (unsigned irq)
-{
-	int rval;
-	__asm__ __volatile__ ("tst1 %1, 0[%2]; setf nz, %0"
-			      : "=r" (rval)
-			      : "i" (V850E_INTC_IC_IF_BIT),
-			        "r" (V850E_INTC_IC_ADDR (irq)));
-	return rval;
-}
-
-
-struct v850e_intc_irq_init {
-	const char *name;	/* name of interrupt type */
-
-	/* Range of kernel irq numbers for this type:
-	   BASE, BASE+INTERVAL, ..., BASE+INTERVAL*NUM  */
-	unsigned base, num, interval;
-
-	unsigned priority;	/* interrupt priority to assign */
-};
-struct hw_interrupt_type;	/* fwd decl */
-
-/* Initialize HW_IRQ_TYPES for INTC-controlled irqs described in array
-   INITS (which is terminated by an entry with the name field == 0).  */
-extern void v850e_intc_init_irq_types (struct v850e_intc_irq_init *inits,
-				       struct hw_interrupt_type *hw_irq_types);
-
-
-#endif /* !__ASSEMBLY__ */
-
-
-#endif /* __V850_V850E_INTC_H__ */
diff --git a/include/asm-v850/v850e_timer_c.h b/include/asm-v850/v850e_timer_c.h
deleted file mode 100644
index f70575df6ea9..000000000000
--- a/include/asm-v850/v850e_timer_c.h
+++ /dev/null
@@ -1,48 +0,0 @@
-/*
- * include/asm-v850/v850e_timer_c.h -- `Timer C' component often used
- *	with the V850E cpu core
- *
- *  Copyright (C) 2001,03  NEC Electronics Corporation
- *  Copyright (C) 2001,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-/* NOTE: this include file currently contains only enough to allow us to
-   use timer C as an interrupt pass-through.  */
-
-#ifndef __V850_V850E_TIMER_C_H__
-#define __V850_V850E_TIMER_C_H__
-
-#include <asm/types.h>
-#include <asm/machdep.h>	/* Pick up chip-specific defs.  */
-
-
-/* Timer C (16-bit interval timers).  */
-
-/* Control register 0 for timer C.  */
-#define V850E_TIMER_C_TMCC0_ADDR(n) (V850E_TIMER_C_BASE_ADDR + 0x6 + 0x10 *(n))
-#define V850E_TIMER_C_TMCC0(n)	  (*(volatile u8 *)V850E_TIMER_C_TMCC0_ADDR(n))
-#define V850E_TIMER_C_TMCC0_CAE	  0x01 /* clock action enable */
-#define V850E_TIMER_C_TMCC0_CE	  0x02 /* count enable */
-/* ... */
-
-/* Control register 1 for timer C.  */
-#define V850E_TIMER_C_TMCC1_ADDR(n) (V850E_TIMER_C_BASE_ADDR + 0x8 + 0x10 *(n))
-#define V850E_TIMER_C_TMCC1(n)	  (*(volatile u8 *)V850E_TIMER_C_TMCC1_ADDR(n))
-#define V850E_TIMER_C_TMCC1_CMS0  0x01 /* capture/compare mode select (ccc0) */
-#define V850E_TIMER_C_TMCC1_CMS1  0x02 /* capture/compare mode select (ccc1) */
-/* ... */
-
-/* Interrupt edge-sensitivity control for timer C.  */
-#define V850E_TIMER_C_SESC_ADDR(n) (V850E_TIMER_C_BASE_ADDR + 0x9 + 0x10 *(n))
-#define V850E_TIMER_C_SESC(n)	  (*(volatile u8 *)V850E_TIMER_C_SESC_ADDR(n))
-
-/* ...etc... */
-
-
-#endif /* __V850_V850E_TIMER_C_H__  */
diff --git a/include/asm-v850/v850e_timer_d.h b/include/asm-v850/v850e_timer_d.h
deleted file mode 100644
index 417612c5b22f..000000000000
--- a/include/asm-v850/v850e_timer_d.h
+++ /dev/null
@@ -1,62 +0,0 @@
-/*
- * include/asm-v850/v850e_timer_d.h -- `Timer D' component often used
- *	with the V850E cpu core
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_V850E_TIMER_D_H__
-#define __V850_V850E_TIMER_D_H__
-
-#include <asm/types.h>
-#include <asm/machdep.h>	/* Pick up chip-specific defs.  */
-
-
-/* Timer D (16-bit interval timers).  */
-
-/* Count registers for timer D.  */
-#define V850E_TIMER_D_TMD_ADDR(n) (V850E_TIMER_D_TMD_BASE_ADDR + 0x10 * (n))
-#define V850E_TIMER_D_TMD(n)	  (*(volatile u16 *)V850E_TIMER_D_TMD_ADDR(n))
-
-/* Count compare registers for timer D.  */
-#define V850E_TIMER_D_CMD_ADDR(n) (V850E_TIMER_D_CMD_BASE_ADDR + 0x10 * (n))
-#define V850E_TIMER_D_CMD(n)	  (*(volatile u16 *)V850E_TIMER_D_CMD_ADDR(n))
-
-/* Control registers for timer D.  */
-#define V850E_TIMER_D_TMCD_ADDR(n) (V850E_TIMER_D_TMCD_BASE_ADDR + 0x10 * (n))
-#define V850E_TIMER_D_TMCD(n)	   (*(volatile u8 *)V850E_TIMER_D_TMCD_ADDR(n))
-/* Control bits for timer D.  */
-#define V850E_TIMER_D_TMCD_CE  	   0x2 /* count enable */
-#define V850E_TIMER_D_TMCD_CAE	   0x1 /* clock action enable */
-/* Clock divider setting (log2).  */
-#define V850E_TIMER_D_TMCD_CS(divlog2) (((divlog2) - V850E_TIMER_D_TMCD_CS_MIN) << 4)
-/* Minimum clock divider setting (log2).  */
-#ifndef V850E_TIMER_D_TMCD_CS_MIN /* Can be overridden by mach-specific hdrs */
-#define V850E_TIMER_D_TMCD_CS_MIN  2 /* Default is correct for the v850e/ma1 */
-#endif
-/* Maximum clock divider setting (log2).  */
-#define V850E_TIMER_D_TMCD_CS_MAX  (V850E_TIMER_D_TMCD_CS_MIN + 7)
-
-/* Return the clock-divider (log2) of timer D unit N.  */
-#define V850E_TIMER_D_DIVLOG2(n) \
-  (((V850E_TIMER_D_TMCD(n) >> 4) & 0x7) + V850E_TIMER_D_TMCD_CS_MIN)
-
-
-#ifndef __ASSEMBLY__
-
-/* Start interval timer TIMER (0-3).  The timer will issue the
-   corresponding INTCMD interrupt RATE times per second.  This function
-   does not enable the interrupt.  */
-extern void v850e_timer_d_configure (unsigned timer, unsigned rate);
-
-#endif /* !__ASSEMBLY__ */
-
-
-#endif /* __V850_V850E_TIMER_D_H__  */
diff --git a/include/asm-v850/v850e_uart.h b/include/asm-v850/v850e_uart.h
deleted file mode 100644
index 5182fb4cc989..000000000000
--- a/include/asm-v850/v850e_uart.h
+++ /dev/null
@@ -1,76 +0,0 @@
-/*
- * include/asm-v850/v850e_uart.h -- common V850E on-chip UART driver
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-/* There's not actually a single UART implementation used by V850E CPUs,
-   but rather a series of implementations that are all `close' to one
-   another.  This file corresponds to the single driver which handles all
-   of them.  */
-
-#ifndef __V850_V850E_UART_H__
-#define __V850_V850E_UART_H__
-
-#include <linux/termios.h>
-
-#include <asm/v850e_utils.h>
-#include <asm/types.h>
-#include <asm/machdep.h>	/* Pick up chip-specific defs.  */
-
-
-/* Include model-specific definitions.  */
-#ifdef CONFIG_V850E_UART
-# ifdef CONFIG_V850E_UARTB
-#  include <asm-v850/v850e_uartb.h>
-# else
-#  include <asm-v850/v850e_uarta.h> /* original V850E UART */
-# endif
-#endif
-
-
-/* Optional capabilities some hardware provides.  */
-
-/* This UART doesn't implement RTS/CTS by default, but some platforms
-   implement them externally, so check to see if <asm/machdep.h> defined
-   anything.  */
-#ifdef V850E_UART_CTS
-#define v850e_uart_cts(n)		V850E_UART_CTS(n)
-#else
-#define v850e_uart_cts(n)		(1)
-#endif
-
-/* Do the same for RTS.  */
-#ifdef V850E_UART_SET_RTS
-#define v850e_uart_set_rts(n,v)		V850E_UART_SET_RTS(n,v)
-#else
-#define v850e_uart_set_rts(n,v)		((void)0)
-#endif
-
-
-/* This is the serial channel to use for the boot console (if desired).  */
-#ifndef V850E_UART_CONSOLE_CHANNEL
-# define V850E_UART_CONSOLE_CHANNEL 0
-#endif
-
-
-#ifndef __ASSEMBLY__
-
-/* Setup a console using channel 0 of the builtin uart.  */
-extern void v850e_uart_cons_init (unsigned chan);
-
-/* Configure and turn on uart channel CHAN, using the termios `control
-   modes' bits in CFLAGS, and a baud-rate of BAUD.  */
-void v850e_uart_configure (unsigned chan, unsigned cflags, unsigned baud);
-
-#endif /* !__ASSEMBLY__ */
-
-
-#endif /* __V850_V850E_UART_H__ */
diff --git a/include/asm-v850/v850e_uarta.h b/include/asm-v850/v850e_uarta.h
deleted file mode 100644
index e483e0950725..000000000000
--- a/include/asm-v850/v850e_uarta.h
+++ /dev/null
@@ -1,278 +0,0 @@
-/*
- * include/asm-v850/v850e_uarta.h -- original V850E on-chip UART
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-/* This is the original V850E UART implementation is called just `UART' in
-   the docs, but we name this header file <asm/v850e_uarta.h> because the
-   name <asm/v850e_uart.h> is used for the common driver that handles both
-   `UART' and `UARTB' implementations.  */
-
-#ifndef __V850_V850E_UARTA_H__
-#define __V850_V850E_UARTA_H__
-
-
-/* Raw hardware interface.  */
-
-/* The base address of the UART control registers for channel N.
-   The default is the address used on the V850E/MA1.  */
-#ifndef V850E_UART_BASE_ADDR
-#define V850E_UART_BASE_ADDR(n)		(0xFFFFFA00 + 0x10 * (n))
-#endif 
-
-/* Addresses of specific UART control registers for channel N.
-   The defaults are the addresses used on the V850E/MA1; if a platform
-   wants to redefine any of these, it must redefine them all.  */
-#ifndef V850E_UART_ASIM_ADDR
-#define V850E_UART_ASIM_ADDR(n)		(V850E_UART_BASE_ADDR(n) + 0x0)
-#define V850E_UART_RXB_ADDR(n)		(V850E_UART_BASE_ADDR(n) + 0x2)
-#define V850E_UART_ASIS_ADDR(n)		(V850E_UART_BASE_ADDR(n) + 0x3)
-#define V850E_UART_TXB_ADDR(n)		(V850E_UART_BASE_ADDR(n) + 0x4)
-#define V850E_UART_ASIF_ADDR(n)		(V850E_UART_BASE_ADDR(n) + 0x5)
-#define V850E_UART_CKSR_ADDR(n)		(V850E_UART_BASE_ADDR(n) + 0x6)
-#define V850E_UART_BRGC_ADDR(n)		(V850E_UART_BASE_ADDR(n) + 0x7)
-#endif
-
-/* UART config registers.  */
-#define V850E_UART_ASIM(n)	(*(volatile u8 *)V850E_UART_ASIM_ADDR(n))
-/* Control bits for config registers.  */
-#define V850E_UART_ASIM_CAE	0x80 /* clock enable */
-#define V850E_UART_ASIM_TXE	0x40 /* transmit enable */
-#define V850E_UART_ASIM_RXE	0x20 /* receive enable */
-#define V850E_UART_ASIM_PS_MASK	0x18 /* mask covering parity-select bits */
-#define V850E_UART_ASIM_PS_NONE	0x00 /* no parity */
-#define V850E_UART_ASIM_PS_ZERO	0x08 /* zero parity */
-#define V850E_UART_ASIM_PS_ODD	0x10 /* odd parity */
-#define V850E_UART_ASIM_PS_EVEN	0x18 /* even parity */
-#define V850E_UART_ASIM_CL_8	0x04 /* char len is 8 bits (otherwise, 7) */
-#define V850E_UART_ASIM_SL_2	0x02 /* 2 stop bits (otherwise, 1) */
-#define V850E_UART_ASIM_ISRM	0x01 /* generate INTSR interrupt on errors
-					(otherwise, generate INTSER) */
-
-/* UART serial interface status registers.  */
-#define V850E_UART_ASIS(n)	(*(volatile u8 *)V850E_UART_ASIS_ADDR(n))
-/* Control bits for status registers.  */
-#define V850E_UART_ASIS_PE	0x04 /* parity error */
-#define V850E_UART_ASIS_FE	0x02 /* framing error */
-#define V850E_UART_ASIS_OVE	0x01 /* overrun error */
-
-/* UART serial interface transmission status registers.  */
-#define V850E_UART_ASIF(n)	(*(volatile u8 *)V850E_UART_ASIF_ADDR(n))
-#define V850E_UART_ASIF_TXBF	0x02 /* transmit buffer flag (data in TXB) */
-#define V850E_UART_ASIF_TXSF	0x01 /* transmit shift flag (sending data) */
-
-/* UART receive buffer register.  */
-#define V850E_UART_RXB(n)	(*(volatile u8 *)V850E_UART_RXB_ADDR(n))
-
-/* UART transmit buffer register.  */
-#define V850E_UART_TXB(n)	(*(volatile u8 *)V850E_UART_TXB_ADDR(n))
-
-/* UART baud-rate generator control registers.  */
-#define V850E_UART_CKSR(n)	(*(volatile u8 *)V850E_UART_CKSR_ADDR(n))
-#define V850E_UART_CKSR_MAX	11
-#define V850E_UART_BRGC(n)	(*(volatile u8 *)V850E_UART_BRGC_ADDR(n))
-#define V850E_UART_BRGC_MIN	8
-
-
-#ifndef V850E_UART_CKSR_MAX_FREQ
-#define V850E_UART_CKSR_MAX_FREQ (25*1000*1000)
-#endif
-
-/* Calculate the minimum value for CKSR on this processor.  */
-static inline unsigned v850e_uart_cksr_min (void)
-{
-	int min = 0;
-	unsigned freq = V850E_UART_BASE_FREQ;
-	while (freq > V850E_UART_CKSR_MAX_FREQ) {
-		freq >>= 1;
-		min++;
-	}
-	return min;
-}
-
-
-/* Slightly abstract interface used by driver.  */
-
-
-/* Interrupts used by the UART.  */
-
-/* Received when the most recently transmitted character has been sent.  */
-#define V850E_UART_TX_IRQ(chan)		IRQ_INTST (chan)
-/* Received when a new character has been received.  */
-#define V850E_UART_RX_IRQ(chan)		IRQ_INTSR (chan)
-
-
-/* UART clock generator interface.  */
-
-/* This type encapsulates a particular uart frequency.  */
-typedef struct {
-	unsigned clk_divlog2;
-	unsigned brgen_count;
-} v850e_uart_speed_t;
-
-/* Calculate a uart speed from BAUD for this uart.  */
-static inline v850e_uart_speed_t v850e_uart_calc_speed (unsigned baud)
-{
-	v850e_uart_speed_t speed;
-
-	/* Calculate the log2 clock divider and baud-rate counter values
-	   (note that the UART divides the resulting clock by 2, so
-	   multiply BAUD by 2 here to compensate).  */
-	calc_counter_params (V850E_UART_BASE_FREQ, baud * 2,
-			     v850e_uart_cksr_min(),
-			     V850E_UART_CKSR_MAX, 8/*bits*/,
-			     &speed.clk_divlog2, &speed.brgen_count);
-
-	return speed;
-}
-
-/* Return the current speed of uart channel CHAN.  */
-static inline v850e_uart_speed_t v850e_uart_speed (unsigned chan)
-{
-	v850e_uart_speed_t speed;
-	speed.clk_divlog2 = V850E_UART_CKSR (chan);
-	speed.brgen_count = V850E_UART_BRGC (chan);
-	return speed;
-}
-
-/* Set the current speed of uart channel CHAN.  */
-static inline void v850e_uart_set_speed(unsigned chan,v850e_uart_speed_t speed)
-{
-	V850E_UART_CKSR (chan) = speed.clk_divlog2;
-	V850E_UART_BRGC (chan) = speed.brgen_count;
-}
-
-static inline int
-v850e_uart_speed_eq (v850e_uart_speed_t speed1, v850e_uart_speed_t speed2)
-{
-	return speed1.clk_divlog2 == speed2.clk_divlog2
-		&& speed1.brgen_count == speed2.brgen_count;
-}
-
-/* Minimum baud rate possible.  */
-#define v850e_uart_min_baud() \
-   ((V850E_UART_BASE_FREQ >> V850E_UART_CKSR_MAX) / (2 * 255) + 1)
-
-/* Maximum baud rate possible.  The error is quite high at max, though.  */
-#define v850e_uart_max_baud() \
-   ((V850E_UART_BASE_FREQ >> v850e_uart_cksr_min()) / (2 *V850E_UART_BRGC_MIN))
-
-/* The `maximum' clock rate the uart can used, which is wanted (though not
-   really used in any useful way) by the serial framework.  */
-#define v850e_uart_max_clock() \
-   ((V850E_UART_BASE_FREQ >> v850e_uart_cksr_min()) / 2)
-
-
-/* UART configuration interface.  */
-
-/* Type of the uart config register; must be a scalar.  */
-typedef u16 v850e_uart_config_t;
-
-/* The uart hardware config register for channel CHAN.  */
-#define V850E_UART_CONFIG(chan)		V850E_UART_ASIM (chan)
-
-/* This config bit set if the uart is enabled.  */
-#define V850E_UART_CONFIG_ENABLED	V850E_UART_ASIM_CAE
-/* If the uart _isn't_ enabled, store this value to it to do so.  */
-#define V850E_UART_CONFIG_INIT		V850E_UART_ASIM_CAE
-/* Store this config value to disable the uart channel completely.  */
-#define V850E_UART_CONFIG_FINI		0
-
-/* Setting/clearing these bits enable/disable TX/RX, respectively (but
-   otherwise generally leave things running).  */
-#define V850E_UART_CONFIG_RX_ENABLE	V850E_UART_ASIM_RXE
-#define V850E_UART_CONFIG_TX_ENABLE	V850E_UART_ASIM_TXE
-
-/* These masks define which config bits affect TX/RX modes, respectively.  */
-#define V850E_UART_CONFIG_RX_BITS \
-  (V850E_UART_ASIM_PS_MASK | V850E_UART_ASIM_CL_8 | V850E_UART_ASIM_ISRM)
-#define V850E_UART_CONFIG_TX_BITS \
-  (V850E_UART_ASIM_PS_MASK | V850E_UART_ASIM_CL_8 | V850E_UART_ASIM_SL_2)
-
-static inline v850e_uart_config_t v850e_uart_calc_config (unsigned cflags)
-{
-	v850e_uart_config_t config = 0;
-
-	/* Figure out new configuration of control register.  */
-	if (cflags & CSTOPB)
-		/* Number of stop bits, 1 or 2.  */
-		config |= V850E_UART_ASIM_SL_2;
-	if ((cflags & CSIZE) == CS8)
-		/* Number of data bits, 7 or 8.  */
-		config |= V850E_UART_ASIM_CL_8;
-	if (! (cflags & PARENB))
-		/* No parity check/generation.  */
-		config |= V850E_UART_ASIM_PS_NONE;
-	else if (cflags & PARODD)
-		/* Odd parity check/generation.  */
-		config |= V850E_UART_ASIM_PS_ODD;
-	else
-		/* Even parity check/generation.  */
-		config |= V850E_UART_ASIM_PS_EVEN;
-	if (cflags & CREAD)
-		/* Reading enabled.  */
-		config |= V850E_UART_ASIM_RXE;
-
-	config |= V850E_UART_ASIM_CAE;
-	config |= V850E_UART_ASIM_TXE; /* Writing is always enabled.  */
-	config |= V850E_UART_ASIM_ISRM; /* Errors generate a read-irq.  */
-
-	return config;
-}
-
-/* This should delay as long as necessary for a recently written config
-   setting to settle, before we turn the uart back on.  */
-static inline void
-v850e_uart_config_delay (v850e_uart_config_t config, v850e_uart_speed_t speed)
-{
-	/* The UART may not be reset properly unless we wait at least 2
-	   `basic-clocks' until turning on the TXE/RXE bits again.
-	   A `basic clock' is the clock used by the baud-rate generator,
-	   i.e., the cpu clock divided by the 2^new_clk_divlog2.
-	   The loop takes 2 insns, so loop CYCLES / 2 times.  */
-	register unsigned count = 1 << speed.clk_divlog2;
-	while (--count != 0)
-		/* nothing */;
-}
-
-
-/* RX/TX interface.  */
-
-/* Return true if all characters awaiting transmission on uart channel N
-   have been transmitted.  */
-#define v850e_uart_xmit_done(n)						      \
-   (! (V850E_UART_ASIF(n) & V850E_UART_ASIF_TXBF))
-/* Wait for this to be true.  */
-#define v850e_uart_wait_for_xmit_done(n)				      \
-   do { } while (! v850e_uart_xmit_done (n))
-
-/* Return true if uart channel N is ready to transmit a character.  */
-#define v850e_uart_xmit_ok(n)						      \
-   (v850e_uart_xmit_done(n) && v850e_uart_cts(n))
-/* Wait for this to be true.  */
-#define v850e_uart_wait_for_xmit_ok(n)					      \
-   do { } while (! v850e_uart_xmit_ok (n))
-
-/* Write character CH to uart channel CHAN.  */
-#define v850e_uart_putc(chan, ch)	(V850E_UART_TXB(chan) = (ch))
-
-/* Return latest character read on channel CHAN.  */
-#define v850e_uart_getc(chan)		V850E_UART_RXB (chan)
-
-/* Return bit-mask of uart error status.  */
-#define v850e_uart_err(chan)		V850E_UART_ASIS (chan)
-/* Various error bits set in the error result.  */
-#define V850E_UART_ERR_OVERRUN		V850E_UART_ASIS_OVE
-#define V850E_UART_ERR_FRAME		V850E_UART_ASIS_FE
-#define V850E_UART_ERR_PARITY		V850E_UART_ASIS_PE
-
-
-#endif /* __V850_V850E_UARTA_H__ */
diff --git a/include/asm-v850/v850e_uartb.h b/include/asm-v850/v850e_uartb.h
deleted file mode 100644
index 6d4767d5a835..000000000000
--- a/include/asm-v850/v850e_uartb.h
+++ /dev/null
@@ -1,262 +0,0 @@
-/*
- * include/asm-v850/v850e_uartb.h -- V850E on-chip `UARTB' UART
- *
- *  Copyright (C) 2001,02,03  NEC Electronics Corporation
- *  Copyright (C) 2001,02,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-/* The V850E UARTB is basically a superset of the original V850E UART, but
-   even where it's the same, the names and details have changed a bit.
-   It's similar enough to use the same driver (v850e_uart.c), but the
-   details have been abstracted slightly to do so.  */
-
-#ifndef __V850_V850E_UARTB_H__
-#define __V850_V850E_UARTB_H__
-
-
-/* Raw hardware interface.  */
-
-#define V850E_UARTB_BASE_ADDR(n)	(0xFFFFFA00 + 0x10 * (n))
-
-/* Addresses of specific UART control registers for channel N.  */
-#define V850E_UARTB_CTL0_ADDR(n)	(V850E_UARTB_BASE_ADDR(n) + 0x0)
-#define V850E_UARTB_CTL2_ADDR(n)	(V850E_UARTB_BASE_ADDR(n) + 0x2)
-#define V850E_UARTB_STR_ADDR(n)		(V850E_UARTB_BASE_ADDR(n) + 0x4)
-#define V850E_UARTB_RX_ADDR(n)		(V850E_UARTB_BASE_ADDR(n) + 0x6)
-#define V850E_UARTB_RXAP_ADDR(n)	(V850E_UARTB_BASE_ADDR(n) + 0x6)
-#define V850E_UARTB_TX_ADDR(n)		(V850E_UARTB_BASE_ADDR(n) + 0x8)
-#define V850E_UARTB_FIC0_ADDR(n)	(V850E_UARTB_BASE_ADDR(n) + 0xA)
-#define V850E_UARTB_FIC1_ADDR(n)	(V850E_UARTB_BASE_ADDR(n) + 0xB)
-#define V850E_UARTB_FIC2_ADDR(n)	(V850E_UARTB_BASE_ADDR(n) + 0xC)
-#define V850E_UARTB_FIS0_ADDR(n)	(V850E_UARTB_BASE_ADDR(n) + 0xE)
-#define V850E_UARTB_FIS1_ADDR(n)	(V850E_UARTB_BASE_ADDR(n) + 0xF)
-
-/* UARTB control register 0 (general config).  */
-#define V850E_UARTB_CTL0(n)	(*(volatile u8 *)V850E_UARTB_CTL0_ADDR(n))
-/* Control bits for config registers.  */
-#define V850E_UARTB_CTL0_PWR		0x80	/* clock enable */
-#define V850E_UARTB_CTL0_TXE		0x40	/* transmit enable */
-#define V850E_UARTB_CTL0_RXE		0x20	/* receive enable */
-#define V850E_UARTB_CTL0_DIR		0x10	/*  */
-#define V850E_UARTB_CTL0_PS1		0x08	/* parity */
-#define V850E_UARTB_CTL0_PS0		0x04	/* parity */
-#define V850E_UARTB_CTL0_CL		0x02	/* char len 1:8bit, 0:7bit */
-#define V850E_UARTB_CTL0_SL		0x01	/* stop bit 1:2bit, 0:1bit */
-#define V850E_UARTB_CTL0_PS_MASK	0x0C	/* mask covering parity bits */
-#define V850E_UARTB_CTL0_PS_NONE	0x00	/* no parity */
-#define V850E_UARTB_CTL0_PS_ZERO	0x04	/* zero parity */
-#define V850E_UARTB_CTL0_PS_ODD		0x08	/* odd parity */
-#define V850E_UARTB_CTL0_PS_EVEN	0x0C	/* even parity */
-#define V850E_UARTB_CTL0_CL_8		0x02	/* char len 1:8bit, 0:7bit */
-#define V850E_UARTB_CTL0_SL_2		0x01	/* stop bit 1:2bit, 0:1bit */
-
-/* UARTB control register 2 (clock divider).  */
-#define V850E_UARTB_CTL2(n)	(*(volatile u16 *)V850E_UARTB_CTL2_ADDR(n))
-#define V850E_UARTB_CTL2_MIN	4
-#define V850E_UARTB_CTL2_MAX	0xFFFF
-
-/* UARTB serial interface status register.  */
-#define V850E_UARTB_STR(n)	(*(volatile u8 *)V850E_UARTB_STR_ADDR(n))
-/* Control bits for status registers.  */
-#define V850E_UARTB_STR_TSF	0x80	/* UBTX or FIFO exist data  */
-#define V850E_UARTB_STR_OVF	0x08	/* overflow error */
-#define V850E_UARTB_STR_PE	0x04	/* parity error */
-#define V850E_UARTB_STR_FE	0x02	/* framing error */
-#define V850E_UARTB_STR_OVE	0x01	/* overrun error */
-
-/* UARTB receive data register.  */
-#define V850E_UARTB_RX(n)	(*(volatile u8 *)V850E_UARTB_RX_ADDR(n))
-#define V850E_UARTB_RXAP(n)	(*(volatile u16 *)V850E_UARTB_RXAP_ADDR(n))
-/* Control bits for status registers.  */
-#define V850E_UARTB_RXAP_PEF	0x0200 /* parity error */
-#define V850E_UARTB_RXAP_FEF	0x0100 /* framing error */
-
-/* UARTB transmit data register.  */
-#define V850E_UARTB_TX(n)	(*(volatile u8 *)V850E_UARTB_TX_ADDR(n))
-
-/* UARTB FIFO control register 0.  */
-#define V850E_UARTB_FIC0(n)	(*(volatile u8 *)V850E_UARTB_FIC0_ADDR(n))
-
-/* UARTB FIFO control register 1.  */
-#define V850E_UARTB_FIC1(n)	(*(volatile u8 *)V850E_UARTB_FIC1_ADDR(n))
-
-/* UARTB FIFO control register 2.  */
-#define V850E_UARTB_FIC2(n)	(*(volatile u16 *)V850E_UARTB_FIC2_ADDR(n))
-
-/* UARTB FIFO status register 0.  */
-#define V850E_UARTB_FIS0(n)	(*(volatile u8 *)V850E_UARTB_FIS0_ADDR(n))
-
-/* UARTB FIFO status register 1.  */
-#define V850E_UARTB_FIS1(n)	(*(volatile u8 *)V850E_UARTB_FIS1_ADDR(n))
-
-
-/* Slightly abstract interface used by driver.  */
-
-
-/* Interrupts used by the UART.  */
-
-/* Received when the most recently transmitted character has been sent.  */
-#define V850E_UART_TX_IRQ(chan)		IRQ_INTUBTIT (chan)
-/* Received when a new character has been received.  */
-#define V850E_UART_RX_IRQ(chan)		IRQ_INTUBTIR (chan)
-
-/* Use by serial driver for information purposes.  */
-#define V850E_UART_BASE_ADDR(chan)	V850E_UARTB_BASE_ADDR(chan)
-
-
-/* UART clock generator interface.  */
-
-/* This type encapsulates a particular uart frequency.  */
-typedef u16 v850e_uart_speed_t;
-
-/* Calculate a uart speed from BAUD for this uart.  */
-static inline v850e_uart_speed_t v850e_uart_calc_speed (unsigned baud)
-{
-	v850e_uart_speed_t speed;
-
-	/*
-	 * V850E/ME2 UARTB baud rate is determined by the value of UBCTL2
-	 * fx = V850E_UARTB_BASE_FREQ = CPU_CLOCK_FREQ/4
-	 * baud = fx / 2*speed   [ speed >= 4 ]
-	 */
-	speed = V850E_UARTB_CTL2_MIN;
-	while (((V850E_UARTB_BASE_FREQ / 2) / speed ) > baud)
-		speed++;
-
-	return speed;
-}
-
-/* Return the current speed of uart channel CHAN.  */
-#define v850e_uart_speed(chan)		    V850E_UARTB_CTL2 (chan)
-
-/* Set the current speed of uart channel CHAN.  */
-#define v850e_uart_set_speed(chan, speed)   (V850E_UARTB_CTL2 (chan) = (speed))
-
-/* Return true if SPEED1 and SPEED2 are the same.  */
-#define v850e_uart_speed_eq(speed1, speed2) ((speed1) == (speed2))
-
-/* Minimum baud rate possible.  */
-#define v850e_uart_min_baud() \
-   ((V850E_UARTB_BASE_FREQ / 2) / V850E_UARTB_CTL2_MAX)
-
-/* Maximum baud rate possible.  The error is quite high at max, though.  */
-#define v850e_uart_max_baud() \
-   ((V850E_UARTB_BASE_FREQ / 2) / V850E_UARTB_CTL2_MIN)
-
-/* The `maximum' clock rate the uart can used, which is wanted (though not
-   really used in any useful way) by the serial framework.  */
-#define v850e_uart_max_clock() \
-   (V850E_UARTB_BASE_FREQ / 2)
-
-
-/* UART configuration interface.  */
-
-/* Type of the uart config register; must be a scalar.  */
-typedef u16 v850e_uart_config_t;
-
-/* The uart hardware config register for channel CHAN.  */
-#define V850E_UART_CONFIG(chan)		V850E_UARTB_CTL0 (chan)
-
-/* This config bit set if the uart is enabled.  */
-#define V850E_UART_CONFIG_ENABLED	V850E_UARTB_CTL0_PWR
-/* If the uart _isn't_ enabled, store this value to it to do so.  */
-#define V850E_UART_CONFIG_INIT		V850E_UARTB_CTL0_PWR
-/* Store this config value to disable the uart channel completely.  */
-#define V850E_UART_CONFIG_FINI		0
-
-/* Setting/clearing these bits enable/disable TX/RX, respectively (but
-   otherwise generally leave things running).  */
-#define V850E_UART_CONFIG_RX_ENABLE	V850E_UARTB_CTL0_RXE
-#define V850E_UART_CONFIG_TX_ENABLE	V850E_UARTB_CTL0_TXE
-
-/* These masks define which config bits affect TX/RX modes, respectively.  */
-#define V850E_UART_CONFIG_RX_BITS \
-  (V850E_UARTB_CTL0_PS_MASK | V850E_UARTB_CTL0_CL_8)
-#define V850E_UART_CONFIG_TX_BITS \
-  (V850E_UARTB_CTL0_PS_MASK | V850E_UARTB_CTL0_CL_8 | V850E_UARTB_CTL0_SL_2)
-
-static inline v850e_uart_config_t v850e_uart_calc_config (unsigned cflags)
-{
-	v850e_uart_config_t config = 0;
-
-	/* Figure out new configuration of control register.  */
-	if (cflags & CSTOPB)
-		/* Number of stop bits, 1 or 2.  */
-		config |= V850E_UARTB_CTL0_SL_2;
-	if ((cflags & CSIZE) == CS8)
-		/* Number of data bits, 7 or 8.  */
-		config |= V850E_UARTB_CTL0_CL_8;
-	if (! (cflags & PARENB))
-		/* No parity check/generation.  */
-		config |= V850E_UARTB_CTL0_PS_NONE;
-	else if (cflags & PARODD)
-		/* Odd parity check/generation.  */
-		config |= V850E_UARTB_CTL0_PS_ODD;
-	else
-		/* Even parity check/generation.  */
-		config |= V850E_UARTB_CTL0_PS_EVEN;
-	if (cflags & CREAD)
-		/* Reading enabled.  */
-		config |= V850E_UARTB_CTL0_RXE;
-
-	config |= V850E_UARTB_CTL0_PWR;
-	config |= V850E_UARTB_CTL0_TXE; /* Writing is always enabled.  */
-	config |= V850E_UARTB_CTL0_DIR; /* LSB first.  */
-
-	return config;
-}
-
-/* This should delay as long as necessary for a recently written config
-   setting to settle, before we turn the uart back on.  */
-static inline void
-v850e_uart_config_delay (v850e_uart_config_t config, v850e_uart_speed_t speed)
-{
-	/* The UART may not be reset properly unless we wait at least 2
-	   `basic-clocks' until turning on the TXE/RXE bits again.
-	   A `basic clock' is the clock used by the baud-rate generator,
-	   i.e., the cpu clock divided by the 2^new_clk_divlog2.
-	   The loop takes 2 insns, so loop CYCLES / 2 times.  */
-	register unsigned count = 1 << speed;
-	while (--count != 0)
-		/* nothing */;
-}
-
-
-/* RX/TX interface.  */
-
-/* Return true if all characters awaiting transmission on uart channel N
-   have been transmitted.  */
-#define v850e_uart_xmit_done(n)						      \
-   (! (V850E_UARTB_STR(n) & V850E_UARTB_STR_TSF))
-/* Wait for this to be true.  */
-#define v850e_uart_wait_for_xmit_done(n)				      \
-   do { } while (! v850e_uart_xmit_done (n))
-
-/* Return true if uart channel N is ready to transmit a character.  */
-#define v850e_uart_xmit_ok(n)						      \
-   (v850e_uart_xmit_done(n) && v850e_uart_cts(n))
-/* Wait for this to be true.  */
-#define v850e_uart_wait_for_xmit_ok(n)					      \
-   do { } while (! v850e_uart_xmit_ok (n))
-
-/* Write character CH to uart channel CHAN.  */
-#define v850e_uart_putc(chan, ch)	(V850E_UARTB_TX(chan) = (ch))
-
-/* Return latest character read on channel CHAN.  */
-#define v850e_uart_getc(chan)		V850E_UARTB_RX (chan)
-
-/* Return bit-mask of uart error status.  */
-#define v850e_uart_err(chan)		V850E_UARTB_STR (chan)
-/* Various error bits set in the error result.  */
-#define V850E_UART_ERR_OVERRUN		V850E_UARTB_STR_OVE
-#define V850E_UART_ERR_FRAME		V850E_UARTB_STR_FE
-#define V850E_UART_ERR_PARITY		V850E_UARTB_STR_PE
-
-
-#endif /* __V850_V850E_UARTB_H__ */
diff --git a/include/asm-v850/v850e_utils.h b/include/asm-v850/v850e_utils.h
deleted file mode 100644
index 52eb72822d3d..000000000000
--- a/include/asm-v850/v850e_utils.h
+++ /dev/null
@@ -1,35 +0,0 @@
-/*
- * include/asm-v850/v850e_utils.h -- Utility functions associated with
- *	V850E CPUs
- *
- *  Copyright (C) 2001,03  NEC Electronics Corporation
- *  Copyright (C) 2001,03  Miles Bader <miles@gnu.org>
- *
- * This file is subject to the terms and conditions of the GNU General
- * Public License.  See the file COPYING in the main directory of this
- * archive for more details.
- *
- * Written by Miles Bader <miles@gnu.org>
- */
-
-#ifndef __V850_V850E_UTILS_H__
-#define __V850_V850E_UTILS_H__
-
-/* Calculate counter clock-divider and count values to attain the
-   desired frequency RATE from the base frequency BASE_FREQ.  The
-   counter is expected to have a clock-divider, which can divide the
-   system cpu clock by a power of two value from MIN_DIVLOG2 to
-   MAX_DIV_LOG2, and a word-size of COUNTER_SIZE bits (the counter
-   counts up and resets whenever it's equal to the compare register,
-   generating an interrupt or whatever when it does so).  The returned
-   values are: *DIVLOG2 -- log2 of the desired clock divider and *COUNT
-   -- the counter compare value to use.  Returns true if it was possible
-   to find a reasonable value, otherwise false (and the other return
-   values will be set to be as good as possible).  */
-extern int calc_counter_params (unsigned long base_freq,
-				unsigned long rate,
-				unsigned min_divlog2, unsigned max_divlog2,
-				unsigned counter_size,
-				unsigned *divlog2, unsigned *count);
-
-#endif /* __V850_V850E_UTILS_H__ */
diff --git a/include/linux/audit.h b/include/linux/audit.h
index 8b82974bdc12..6272a395d43c 100644
--- a/include/linux/audit.h
+++ b/include/linux/audit.h
@@ -286,7 +286,6 @@
 #define AUDIT_ARCH_SHEL64	(EM_SH|__AUDIT_ARCH_64BIT|__AUDIT_ARCH_LE)
 #define AUDIT_ARCH_SPARC	(EM_SPARC)
 #define AUDIT_ARCH_SPARC64	(EM_SPARCV9|__AUDIT_ARCH_64BIT)
-#define AUDIT_ARCH_V850		(EM_V850|__AUDIT_ARCH_LE)
 #define AUDIT_ARCH_X86_64	(EM_X86_64|__AUDIT_ARCH_64BIT|__AUDIT_ARCH_LE)
 
 #define AUDIT_PERM_EXEC		1
diff --git a/include/linux/module.h b/include/linux/module.h
index fce15ebd0e1c..68e09557c951 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -23,7 +23,7 @@
 /* Not Yet Implemented */
 #define MODULE_SUPPORTED_DEVICE(name)
 
-/* v850 toolchain uses a `_' prefix for all user symbols */
+/* some toolchains uses a `_' prefix for all user symbols */
 #ifndef MODULE_SYMBOL_PREFIX
 #define MODULE_SYMBOL_PREFIX ""
 #endif
diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h
index f3a1c0e45021..3b2f6c04855e 100644
--- a/include/linux/serial_core.h
+++ b/include/linux/serial_core.h
@@ -59,9 +59,6 @@
 #define PORT_SUNZILOG	38
 #define PORT_SUNSAB	39
 
-/* NEC v850.  */
-#define PORT_V850E_UART	40
-
 /* DEC */
 #define PORT_DZ		46
 #define PORT_ZS		47
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 0522f368f9d7..4394dadff813 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -443,7 +443,7 @@ asmlinkage long sys_newuname(struct new_utsname __user *name);
 
 asmlinkage long sys_getrlimit(unsigned int resource,
 				struct rlimit __user *rlim);
-#if defined(COMPAT_RLIM_OLD_INFINITY) || !(defined(CONFIG_IA64) || defined(CONFIG_V850))
+#if defined(COMPAT_RLIM_OLD_INFINITY) || !(defined(CONFIG_IA64))
 asmlinkage long sys_old_getrlimit(unsigned int resource, struct rlimit __user *rlim);
 #endif
 asmlinkage long sys_setrlimit(unsigned int resource,
diff --git a/scripts/genksyms/genksyms.c b/scripts/genksyms/genksyms.c
index dca5e0dd09bf..4f8a3007e457 100644
--- a/scripts/genksyms/genksyms.c
+++ b/scripts/genksyms/genksyms.c
@@ -520,8 +520,7 @@ int main(int argc, char **argv)
 			genksyms_usage();
 			return 1;
 		}
-	if ((strcmp(arch, "v850") == 0) || (strcmp(arch, "h8300") == 0)
-	    || (strcmp(arch, "blackfin") == 0))
+	if ((strcmp(arch, "h8300") == 0) || (strcmp(arch, "blackfin") == 0))
 		mod_prefix = "_";
 	{
 		extern int yydebug;
diff --git a/scripts/mod/file2alias.c b/scripts/mod/file2alias.c
index 1fcaf3284a6a..4fa1f3ad2513 100644
--- a/scripts/mod/file2alias.c
+++ b/scripts/mod/file2alias.c
@@ -623,7 +623,7 @@ static int do_i2c_entry(const char *filename, struct i2c_device_id *id,
 	return 1;
 }
 
-/* Ignore any prefix, eg. v850 prepends _ */
+/* Ignore any prefix, eg. some architectures prepend _ */
 static inline int sym_is(const char *symbol, const char *name)
 {
 	const char *match;
diff --git a/scripts/mod/mk_elfconfig.c b/scripts/mod/mk_elfconfig.c
index db3881f14c2d..6a96d47bd1e6 100644
--- a/scripts/mod/mk_elfconfig.c
+++ b/scripts/mod/mk_elfconfig.c
@@ -55,7 +55,7 @@ main(int argc, char **argv)
 	else
 		exit(1);
 
-	if ((strcmp(argv[1], "v850") == 0) || (strcmp(argv[1], "h8300") == 0)
+	if ((strcmp(argv[1], "h8300") == 0)
 	    || (strcmp(argv[1], "blackfin") == 0))
 		printf("#define MODULE_SYMBOL_PREFIX \"_\"\n");
 	else

From 82736f4d1d2b7063b829cc93171a6e5aea8a9c49 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= <Uwe.Kleine-Koenig@digi.com>
Date: Wed, 23 Jul 2008 21:28:54 -0700
Subject: [PATCH 0327/1171] generic irqs: handle failure of irqchip->set_type
 in setup_irq
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

set_type returns an int indicating success or failure, but up to now
setup_irq ignores that.

In my case this resulted in a machine hang:

gpio-keys requested IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING, but
arm/ns9xxx can only trigger on one direction so set_type didn't touch
the configuration which happens do default to a level sensitiveness and
returned -EINVAL.  setup_irq ignored that and unmasked the irq.  This
resulted in an endless triggering of the gpio-key interrupt service
routine which effectively killed the machine.

With this patch applied setup_irq propagates the error to the caller.

Note that before in the case

	chip && !chip->set_type && !chip->name

a NULL pointer was feed to printk.  This is fixed, too.

Signed-off-by: Uwe Kleine-König <Uwe.Kleine-Koenig@digi.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/irq/manage.c | 64 +++++++++++++++++++++++++++++----------------
 1 file changed, 42 insertions(+), 22 deletions(-)

diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index 3cfc0fefb5ee..5bc6e5ecc493 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -308,6 +308,30 @@ void compat_irq_chip_set_default_handler(struct irq_desc *desc)
 		desc->handle_irq = NULL;
 }
 
+static int __irq_set_trigger(struct irq_chip *chip, unsigned int irq,
+		unsigned long flags)
+{
+	int ret;
+
+	if (!chip || !chip->set_type) {
+		/*
+		 * IRQF_TRIGGER_* but the PIC does not support multiple
+		 * flow-types?
+		 */
+		pr_warning("No set_type function for IRQ %d (%s)\n", irq,
+				chip ? (chip->name ? : "unknown") : "unknown");
+		return 0;
+	}
+
+	ret = chip->set_type(irq, flags & IRQF_TRIGGER_MASK);
+
+	if (ret)
+		pr_err("setting flow type for irq %u failed (%pF)\n",
+				irq, chip->set_type);
+
+	return ret;
+}
+
 /*
  * Internal function to register an irqaction - typically used to
  * allocate special interrupts that are part of the architecture.
@@ -319,6 +343,7 @@ int setup_irq(unsigned int irq, struct irqaction *new)
 	const char *old_name = NULL;
 	unsigned long flags;
 	int shared = 0;
+	int ret;
 
 	if (irq >= NR_IRQS)
 		return -EINVAL;
@@ -376,36 +401,24 @@ int setup_irq(unsigned int irq, struct irqaction *new)
 		shared = 1;
 	}
 
-	*p = new;
-
-	/* Exclude IRQ from balancing */
-	if (new->flags & IRQF_NOBALANCING)
-		desc->status |= IRQ_NO_BALANCING;
-
 	if (!shared) {
 		irq_chip_set_defaults(desc->chip);
 
+		/* Setup the type (level, edge polarity) if configured: */
+		if (new->flags & IRQF_TRIGGER_MASK) {
+			ret = __irq_set_trigger(desc->chip, irq, new->flags);
+
+			if (ret) {
+				spin_unlock_irqrestore(&desc->lock, flags);
+				return ret;
+			}
+		} else
+			compat_irq_chip_set_default_handler(desc);
 #if defined(CONFIG_IRQ_PER_CPU)
 		if (new->flags & IRQF_PERCPU)
 			desc->status |= IRQ_PER_CPU;
 #endif
 
-		/* Setup the type (level, edge polarity) if configured: */
-		if (new->flags & IRQF_TRIGGER_MASK) {
-			if (desc->chip->set_type)
-				desc->chip->set_type(irq,
-						new->flags & IRQF_TRIGGER_MASK);
-			else
-				/*
-				 * IRQF_TRIGGER_* but the PIC does not support
-				 * multiple flow-types?
-				 */
-				printk(KERN_WARNING "No IRQF_TRIGGER set_type "
-				       "function for IRQ %d (%s)\n", irq,
-				       desc->chip->name);
-		} else
-			compat_irq_chip_set_default_handler(desc);
-
 		desc->status &= ~(IRQ_AUTODETECT | IRQ_WAITING |
 				  IRQ_INPROGRESS | IRQ_SPURIOUS_DISABLED);
 
@@ -423,6 +436,13 @@ int setup_irq(unsigned int irq, struct irqaction *new)
 		/* Set default affinity mask once everything is setup */
 		irq_select_affinity(irq);
 	}
+
+	*p = new;
+
+	/* Exclude IRQ from balancing */
+	if (new->flags & IRQF_NOBALANCING)
+		desc->status |= IRQ_NO_BALANCING;
+
 	/* Reset broken irq detection when installing new handler */
 	desc->irq_count = 0;
 	desc->irqs_unhandled = 0;

From 5aa0769d089125e63f8dc23e0283e559e1790493 Mon Sep 17 00:00:00 2001
From: Hans-Christian Egtvedt <hans-christian.egtvedt@atmel.com>
Date: Wed, 23 Jul 2008 21:28:55 -0700
Subject: [PATCH 0328/1171] atmel_pwm: set up only one PWM clock when
 allocating a clock

This patch will only setup one clock, if free, and return this clock to the
caller.  The previous solution would setup both clocks with the same prescaler
and divider and return PWM_CPR_CLKB, thus taking both clocks in the same call
without the caller knowing.

Signed-off-by: Hans-Christian Egtvedt <hans-christian.egtvedt@atmel.com>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/misc/atmel_pwm.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/misc/atmel_pwm.c b/drivers/misc/atmel_pwm.c
index 5b5a14dab3d3..6aa5294dfec4 100644
--- a/drivers/misc/atmel_pwm.c
+++ b/drivers/misc/atmel_pwm.c
@@ -211,8 +211,7 @@ int pwm_clk_alloc(unsigned prescale, unsigned div)
 	if ((mr & 0xffff) == 0) {
 		mr |= val;
 		ret = PWM_CPR_CLKA;
-	}
-	if ((mr & (0xffff << 16)) == 0) {
+	} else if ((mr & (0xffff << 16)) == 0) {
 		mr |= val << 16;
 		ret = PWM_CPR_CLKB;
 	}

From 6cbb2e711128b505209f7c910018aac77335c887 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:28:55 -0700
Subject: [PATCH 0329/1171] checkpatch: Version: 0.20

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 6971bf078d13..66f060ecb81f 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -9,7 +9,7 @@ use strict;
 my $P = $0;
 $P =~ s@.*/@@g;
 
-my $V = '0.19';
+my $V = '0.20';
 
 use Getopt::Long qw(:config no_auto_abbrev);
 

From fee61c47d15270bdea699a8a3dd867f0825c3541 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:28:56 -0700
Subject: [PATCH 0330/1171] checkpatch: return is not a function -- parentheses
 for casts are ok too

Casts require parentheses so it is possible to have something like this:

	return (int)(*a);

This miss trips the complexity function.  Ensure that the two separate
parenthesised sections are not coelesced.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 1 +
 1 file changed, 1 insertion(+)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 66f060ecb81f..83ae37b38621 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -1670,6 +1670,7 @@ sub process {
 			my $value = $2;
 
 			# Flatten any parentheses and braces
+			$value =~ s/\)\(/\) \(/g;
 			while ($value =~ s/\([^\(\)]*\)/1/) {
 			}
 

From c8cb2ca37ed51aa1f3b20e3eff1e72df1c400f70 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:28:57 -0700
Subject: [PATCH 0331/1171] checkpatch: types: some types may also be
 identifiers

Some types such as typedefs may overlap real identifiers.  Be more
targetted about when a type can really exist.  Where it cannot let it be
an identifier.  This prevents false reporting of the minus '-' in unary
context in the following:

	foo[bar->bool - 1];

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 83ae37b38621..5420db6502fc 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -171,6 +171,7 @@ our @modifierList = (
 sub build_types {
 	my $mods = "(?:  \n" . join("|\n  ", @modifierList) . "\n)";
 	my $all = "(?:  \n" . join("|\n  ", @typeList) . "\n)";
+	$Modifier	= qr{(?:$Attribute|$Sparse|$mods)};
 	$NonptrType	= qr{
 			(?:const\s+)?
 			(?:$mods\s+)?
@@ -178,15 +179,14 @@ sub build_types {
 				(?:typeof|__typeof__)\s*\(\s*\**\s*$Ident\s*\)|
 				(?:${all}\b)
 			)
-			(?:\s+$Sparse|\s+const)*
+			(?:\s+$Modifier|\s+const)*
 		  }x;
 	$Type	= qr{
 			$NonptrType
 			(?:\s*\*+\s*const|\s*\*+|(?:\s*\[\s*\])+)?
-			(?:\s+$Inline|\s+$Sparse|\s+$Attribute|\s+$mods)*
+			(?:\s+$Inline|\s+$Modifier)*
 		  }x;
 	$Declare	= qr{(?:$Storage\s+)?$Type};
-	$Modifier	= qr{(?:$Attribute|$Sparse|$mods)};
 }
 build_types();
 
@@ -715,7 +715,7 @@ sub annotate_values {
 				$av_preprocessor = 0;
 			}
 
-		} elsif ($cur =~ /^($Type)/) {
+		} elsif ($cur =~ /^($Type)\s*(?:$Ident|,|\))/) {
 			print "DECLARE($1)\n" if ($dbg_values > 1);
 			$type = 'T';
 
@@ -800,8 +800,9 @@ sub annotate_values {
 				print "PAREN('$1')\n" if ($dbg_values > 1);
 			}
 
-		} elsif ($cur =~ /^($Ident)\(/o) {
+		} elsif ($cur =~ /^($Ident)\s*\(/o) {
 			print "FUNC($1)\n" if ($dbg_values > 1);
+			$type = 'V';
 			$av_pending = 'V';
 
 		} elsif ($cur =~ /^($Ident|$Constant)/o) {

From f3db6639fee577f6ed92c0a1fc881e748c47ec48 Mon Sep 17 00:00:00 2001
From: Michael Ellerman <michael@ellerman.id.au>
Date: Wed, 23 Jul 2008 21:28:57 -0700
Subject: [PATCH 0332/1171] checkpatch: add a checkpatch warning for new uses
 of __initcall().

[apw@shadowen.org: generalise pattern and add tests]
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 5420db6502fc..cf70f123f574 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -2108,6 +2108,10 @@ sub process {
 		if ($line =~ /\bsimple_(strto.*?)\s*\(/) {
 			WARN("consider using strict_$1 in preference to simple_$1\n" . $herecurr);
 		}
+# check for __initcall(), use device_initcall() explicitly please
+		if ($line =~ /^.\s*__initcall\s*\(/) {
+			WARN("please use device_initcall() instead of __initcall()\n" . $herecurr);
+		}
 
 # use of NR_CPUS is usually wrong
 # ignore definitions of NR_CPUS and usage to define arrays as likely right

From d3ddcf471ea90d7ff711dbaa371ef379ed625ec0 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:28:58 -0700
Subject: [PATCH 0333/1171] checkpatch: possible types: __asm__ is never a type

We are false matching __asm__ as a type, and then tripping the external
function checks.  Squash.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index cf70f123f574..fd597a4b5dad 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -846,7 +846,7 @@ sub possible {
 	if ($possible !~ /^(?:$Storage|$Type|DEFINE_\S+)$/ &&
 	    $possible ne 'goto' && $possible ne 'return' &&
 	    $possible ne 'case' && $possible ne 'else' &&
-	    $possible ne 'asm' &&
+	    $possible ne 'asm' && $possible ne '__asm__' &&
 	    $possible !~ /^(typedef|struct|enum)\b/) {
 		# Check for modifiers.
 		$possible =~ s/\s*$Storage\s*//g;

From beae6332493a40116dba24928154621f2e88b9a9 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:28:59 -0700
Subject: [PATCH 0334/1171] checkpatch: comment detection: ignore macro
 continuation when detecting associated comments

When looking for an associated comment they may be suffixed by a macro
continuation.  Ignore this.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index fd597a4b5dad..94250d1a3a47 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -631,7 +631,7 @@ sub ctx_locate_comment {
 	my ($first_line, $end_line) = @_;
 
 	# Catch a comment on the end of the line itself.
-	my ($current_comment) = ($rawlines[$end_line - 1] =~ m@.*(/\*.*\*/)\s*$@);
+	my ($current_comment) = ($rawlines[$end_line - 1] =~ m@.*(/\*.*\*/)\s*(?:\\\s*)?$@);
 	return $current_comment if (defined $current_comment);
 
 	# Look through the context and try and figure out if there is a

From 6ef9b297f6e8850da3be9c9ff5f00385c0977004 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:28:59 -0700
Subject: [PATCH 0335/1171] checkpatch: types: unary -- goto introduces unary
 context

When we see a goto we enter unary context.  For example:

	goto *h;

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 94250d1a3a47..b2b0648ee14a 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -780,7 +780,7 @@ sub annotate_values {
 			$av_pending = 'N';
 			$type = 'N';
 
-		} elsif ($cur =~/^(return|case|else)/o) {
+		} elsif ($cur =~/^(return|case|else|goto)/o) {
 			print "KEYWORD($1)\n" if ($dbg_values > 1);
 			$type = 'N';
 

From a3bb97a7aba36055d476896ed6393ab35a119d5b Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:29:00 -0700
Subject: [PATCH 0336/1171] checkpatch: macros: fix statement counting block
 end detection

We are incorrectly counting the lines in a block while accumulating
the trailing lines in a macro statement, leading to false positives.
Fix end of block handling and general counting for negative context lines.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 23 +++++++++++------------
 1 file changed, 11 insertions(+), 12 deletions(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index b2b0648ee14a..add86862325b 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -470,7 +470,9 @@ sub ctx_statement_block {
 		}
 		$off++;
 	}
+	# We are truly at the end, so shuffle to the next line.
 	if ($off == $len) {
+		$loff = $len + 1;
 		$line++;
 		$remain--;
 	}
@@ -1793,30 +1795,26 @@ sub process {
 				$lines[$ln - 1] =~ /^(?:-|..*\\$)/)
 			{
 				$ctx .= $rawlines[$ln - 1] . "\n";
+				$cnt-- if ($lines[$ln - 1] !~ /^-/);
 				$ln++;
-				$cnt--;
 			}
 			$ctx .= $rawlines[$ln - 1];
 
 			($dstat, $dcond, $ln, $cnt, $off) =
 				ctx_statement_block($linenr, $ln - $linenr + 1, 0);
 			#print "dstat<$dstat> dcond<$dcond> cnt<$cnt> off<$off>\n";
-			#print "LINE<$lines[$ln]> len<" . length($lines[$ln]) . "\n";
+			#print "LINE<$lines[$ln-1]> len<" . length($lines[$ln-1]) . "\n";
 
 			# Extract the remainder of the define (if any) and
 			# rip off surrounding spaces, and trailing \'s.
 			$rest = '';
-			if (defined $lines[$ln - 1] &&
-			    $off > length($lines[$ln - 1]))
-			{
+			while ($off != 0 || ($cnt > 0 && $rest =~ /(?:^|\\)\s*$/)) {
+				#print "ADDING $off <" . substr($lines[$ln - 1], $off) . ">\n";
+				if ($off != 0 || $lines[$ln - 1] !~ /^-/) {
+					$rest .= substr($lines[$ln - 1], $off) . "\n";
+					$cnt--;
+				}
 				$ln++;
-				$cnt--;
-				$off = 0;
-			}
-			while ($cnt > 0) {
-				$rest .= substr($lines[$ln - 1], $off) . "\n";
-				$ln++;
-				$cnt--;
 				$off = 0;
 			}
 			$rest =~ s/\\\n.//g;
@@ -1847,6 +1845,7 @@ sub process {
 				DEFINE_PER_CPU|
 				__typeof__\(
 			}x;
+			#print "REST<$rest>\n";
 			if ($rest ne '') {
 				if ($rest !~ /while\s*\(/ &&
 				    $dstat !~ /$exceptions/)

From 548596d523d83dff5a670beb84be0daf4c3bcd16 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:29:01 -0700
Subject: [PATCH 0337/1171] checkpatch: trailing statement indent: fix end of
 statement location

Fix end of statement location.  Where the last line of the statement is
replaced we are miss reporting the newly added replacement an incorrectly
indented trailing statement for the negative context.  We are also
incorrectly reporting negative statements generally.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index add86862325b..89177c349f93 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -1249,17 +1249,22 @@ sub process {
 			my $pre_ctx = "$1$2";
 
 			my ($level, @ctx) = ctx_statement_level($linenr, $realcnt, 0);
-			my $ctx_ln = $linenr + $#ctx + 1;
 			my $ctx_cnt = $realcnt - $#ctx - 1;
 			my $ctx = join("\n", @ctx);
 
-			##warn "realcnt<$realcnt> ctx_cnt<$ctx_cnt>\n";
+			my $ctx_ln = $linenr;
+			my $ctx_skip = $realcnt;
 
-			# Skip over any removed lines in the context following statement.
-			while (defined($lines[$ctx_ln - 1]) && $lines[$ctx_ln - 1] =~ /^-/) {
+			while ($ctx_skip > $ctx_cnt || ($ctx_skip == $ctx_cnt &&
+					defined $lines[$ctx_ln - 1] &&
+					$lines[$ctx_ln - 1] =~ /^-/)) {
+				##print "SKIP<$ctx_skip> CNT<$ctx_cnt>\n";
+				$ctx_skip-- if (!defined $lines[$ctx_ln - 1] || $lines[$ctx_ln - 1] !~ /^-/);
 				$ctx_ln++;
 			}
-			##warn "pre<$pre_ctx>\nline<$line>\nctx<$ctx>\nnext<$lines[$ctx_ln - 1]>\n";
+
+			##print "realcnt<$realcnt> ctx_cnt<$ctx_cnt>\n";
+			##print "pre<$pre_ctx>\nline<$line>\nctx<$ctx>\nnext<$lines[$ctx_ln - 1]>\n";
 
 			if ($ctx !~ /{\s*/ && defined($lines[$ctx_ln -1]) && $lines[$ctx_ln - 1] =~ /^\+\s*{/) {
 				ERROR("that open brace { should be on the previous line\n" .

From f4c014c0dede10cc0a8463e748892e738e190699 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:29:01 -0700
Subject: [PATCH 0338/1171] checkpatch: allow printk strings to exceed 80
 characters to maintain their searchability

Allow printk strings to break the 80 character width limits, thus keeping
them complete and searchable.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 89177c349f93..614999f29aa0 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -1138,7 +1138,9 @@ sub process {
 		}
 #80 column limit
 		if ($line =~ /^\+/ && $prevrawline !~ /\/\*\*/ &&
-		    $rawline !~ /^.\s*\*\s*\@$Ident\s/ && $length > 80)
+		    $rawline !~ /^.\s*\*\s*\@$Ident\s/ &&
+		    $line !~ /^\+\s*printk\s*\(\s*(?:KERN_\S+\s*)?"[X\t]*"\s*(?:,|\)\s*;)\s*$/ &&
+		    $length > 80)
 		{
 			WARN("line over 80 characters\n" . $herecurr);
 		}

From e2a763c20b89890d2153551b1af6962b135de4c0 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:29:02 -0700
Subject: [PATCH 0339/1171] checkpatch: switch -- report trailing statements on
 case and default

Report trailing statements on case and default lines.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 614999f29aa0..5f71b3050253 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -1244,6 +1244,10 @@ sub process {
 				ERROR("switch and case should be at the same indent\n$hereline$err");
 			}
 		}
+		if ($line =~ /^.\s*(?:case\s*.*|default\s*):/g &&
+		    $line !~ /\G(?:\s*{)?(?:\s*$;*)(?:\s*\\)?\s*$/g) {
+			ERROR("trailing statements should be on next line\n" . $herecurr);
+		}
 
 # if/while/etc brace do not go on next line, unless defining a do while loop,
 # or if that brace on the next line is for something else

From 8d31cfcecf67563d70cd68616cb8fb4384f24b51 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:29:02 -0700
Subject: [PATCH 0340/1171] checkpatch: check spacing for square brackets

Check on the spacing before square brackets.  We should only allow spaces
there if this is part of a type definition or an initialialiser.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 5f71b3050253..e7c8ab1b54bd 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -1435,6 +1435,17 @@ sub process {
 			ERROR("open brace '{' following $1 go on the same line\n" . $hereprev);
 		}
 
+# check for spacing round square brackets; allowed:
+#  1. with a type on the left -- int [] a;
+#  2. at the beginning of a line for slice initialisers -- [0..10] = 5,
+		while ($line =~ /(.*?\s)\[/g) {
+			my ($where, $prefix) = ($-[1], $1);
+			if ($prefix !~ /$Type\s+$/ &&
+			    ($where != 0 || $prefix !~ /^.\s+$/)) {
+				ERROR("space prohibited before open square bracket '['\n" . $herecurr);
+			}
+		}
+
 # check for spaces between functions and their parentheses.
 		while ($line =~ /($Ident)\s+\(/g) {
 			my $name = $1;

From 53210168feeff9a3c780bd42f69936d4c12381d5 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:29:03 -0700
Subject: [PATCH 0341/1171] checkpatch: toughen trailing if statement checks
 and extend them to while and for

Extend the trailing statement checks to report a trailing semi-colon ';'
as we really want it on the next line and indented so it is really really
obvious.  Also extend the tests to include while and for.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index e7c8ab1b54bd..8616baee0aeb 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -1269,8 +1269,8 @@ sub process {
 				$ctx_ln++;
 			}
 
-			##print "realcnt<$realcnt> ctx_cnt<$ctx_cnt>\n";
-			##print "pre<$pre_ctx>\nline<$line>\nctx<$ctx>\nnext<$lines[$ctx_ln - 1]>\n";
+			#print "realcnt<$realcnt> ctx_cnt<$ctx_cnt>\n";
+			#print "pre<$pre_ctx>\nline<$line>\nctx<$ctx>\nnext<$lines[$ctx_ln - 1]>\n";
 
 			if ($ctx !~ /{\s*/ && defined($lines[$ctx_ln -1]) && $lines[$ctx_ln - 1] =~ /^\+\s*{/) {
 				ERROR("that open brace { should be on the previous line\n" .
@@ -1713,7 +1713,7 @@ sub process {
 		}
 
 # Check for illegal assignment in if conditional.
-		if ($line =~ /\bif\s*\(/) {
+		if ($line =~ /\b(?:if|while|for)\s*\(/ && $line !~ /^.\s*#/) {
 			my ($s, $c) = ($stat, $cond);
 
 			if ($c =~ /\bif\s*\(.*[^<>!=]=[^=].*/) {
@@ -1725,8 +1725,8 @@ sub process {
 			substr($s, 0, length($c), '');
 			$s =~ s/\n.*//g;
 			$s =~ s/$;//g; 	# Remove any comments
-			if (length($c) && $s !~ /^\s*({|;|)\s*\\*\s*$/ &&
-			    $c !~ /^.\s*\#\s*if/)
+			if (length($c) && $s !~ /^\s*{?\s*\\*\s*$/ &&
+			    $c !~ /}\s*while\s*/)
 			{
 				ERROR("trailing statements should be on next line\n" . $herecurr);
 			}

From f5fe35dd95549b1b419cdeb2ec3fe61fda94fa93 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:29:03 -0700
Subject: [PATCH 0342/1171] checkpatch: condition/loop indent checks

Check to see if the block/statement which a condition or loop introduces
is indented correctly.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 59 ++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 55 insertions(+), 4 deletions(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 8616baee0aeb..13d7a330b5dd 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -1167,10 +1167,10 @@ sub process {
 		}
 
 # Check for potential 'bare' types
-		my ($stat, $cond);
+		my ($stat, $cond, $line_nr_next, $remain_next);
 		if ($realcnt && $line =~ /.\s*\S/) {
-			($stat, $cond) = ctx_statement_block($linenr,
-								$realcnt, 0);
+			($stat, $cond, $line_nr_next, $remain_next) =
+				ctx_statement_block($linenr, $realcnt, 0);
 			$stat =~ s/\n./\n /g;
 			$cond =~ s/\n./\n /g;
 
@@ -1712,7 +1712,8 @@ sub process {
 			ERROR("space required before the open parenthesis '('\n" . $herecurr);
 		}
 
-# Check for illegal assignment in if conditional.
+# Check for illegal assignment in if conditional -- and check for trailing
+# statements after the conditional.
 		if ($line =~ /\b(?:if|while|for)\s*\(/ && $line !~ /^.\s*#/) {
 			my ($s, $c) = ($stat, $cond);
 
@@ -1732,6 +1733,56 @@ sub process {
 			}
 		}
 
+# Check relative indent for conditionals and blocks.
+		if ($line =~ /\b(?:(?:if|while|for)\s*\(|do\b)/ && $line !~ /^.\s*#/ && $line !~ /\}\s*while\s*/) {
+			my ($s, $c) = ($stat, $cond);
+
+			substr($s, 0, length($c), '');
+
+			# Make sure we remove the line prefixes as we have
+			# none on the first line, and are going to readd them
+			# where necessary.
+			$s =~ s/\n./\n/gs;
+
+			# We want to check the first line inside the block
+			# starting at the end of the conditional, so remove:
+			#  1) any blank line termination
+			#  2) any opening brace { on end of the line
+			#  3) any do (...) {
+			my $continuation = 0;
+			my $check = 0;
+			$s =~ s/^.*\bdo\b//;
+			$s =~ s/^\s*{//;
+			if ($s =~ s/^\s*\\//) {
+				$continuation = 1;
+			}
+			if ($s =~ s/^\s*\n//) {
+				$check = 1;
+			}
+
+			# Also ignore a loop construct at the end of a
+			# preprocessor statement.
+			if (($prevline =~ /^.\s*#\s*define\s/ ||
+			    $prevline =~ /\\\s*$/) && $continuation == 0) {
+				$check = 0;
+			}
+
+			# Ignore the current line if its is a preprocessor
+			# line.
+			if ($s =~ /^\s*#\s*/) {
+				$check = 0;
+			}
+
+			my (undef, $sindent) = line_stats("+" . $s);
+
+			##print "line<$line> prevline<$prevline> indent<$indent> sindent<$sindent> check<$check> continuation<$continuation> s<$s>\n";
+
+			if ($check && (($sindent % 8) != 0 ||
+			    ($sindent <= $indent && $s ne ''))) {
+				WARN("suspect code indent for conditional statements\n" . $herecurr);
+			}
+		}
+
 # Check for bitwise tests written as boolean
 		if ($line =~ /
 			(?:

From 4c432a8f0134504814aa8dcce6cc57c89d175604 Mon Sep 17 00:00:00 2001
From: Greg Kroah-Hartman <gregkh@suse.de>
Date: Wed, 23 Jul 2008 21:29:04 -0700
Subject: [PATCH 0343/1171] checkpatch: usb_free_urb() can take NULL

usb_free_urb() can take a NULL, so let's check and warn about that.

Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 13d7a330b5dd..a4e8087a0cae 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -2078,6 +2078,13 @@ sub process {
 				WARN("kfree(NULL) is safe this check is probabally not required\n" . $hereprev);
 			}
 		}
+# check for needless usb_free_urb() checks
+		if ($prevline =~ /\bif\s*\(([^\)]*)\)/) {
+			my $expr = $1;
+			if ($line =~ /\busb_free_urb\(\Q$expr\E\);/) {
+				WARN("usb_free_urb(NULL) is safe this check is probably not required\n" . $hereprev);
+			}
+		}
 
 # warn about #ifdefs in C files
 #		if ($line =~ /^.\s*\#\s*if(|n)def/ && ($realfile =~ /\.c$/)) {

From 3c232147a7d5b0418b0a0bae0e5b9a62fb81f4f2 Mon Sep 17 00:00:00 2001
From: Wolfram Sang <w.sang@pengutronix.de>
Date: Wed, 23 Jul 2008 21:29:05 -0700
Subject: [PATCH 0344/1171] checkpatch: correct spelling in kfree checks

Correct spelling in the kfree reports.

Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index a4e8087a0cae..3961e759a256 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -2075,7 +2075,7 @@ sub process {
 		if ($prevline =~ /\bif\s*\(([^\)]*)\)/) {
 			my $expr = $1;
 			if ($line =~ /\bkfree\(\Q$expr\E\);/) {
-				WARN("kfree(NULL) is safe this check is probabally not required\n" . $hereprev);
+				WARN("kfree(NULL) is safe this check is probably not required\n" . $hereprev);
 			}
 		}
 # check for needless usb_free_urb() checks

From 389a2fe57ffc59a649bea39db4d7e6d2eff2b562 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:29:05 -0700
Subject: [PATCH 0345/1171] checkpatch: allow for type modifiers on multiple
 declarations

Allow for type modifiers mid declaration on multiple declarations:

	struct mxser_mstatus ms, __user *msu = argp;

Reported by Jiri Slaby.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 3961e759a256..bcfb8ef00fee 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -721,6 +721,10 @@ sub annotate_values {
 			print "DECLARE($1)\n" if ($dbg_values > 1);
 			$type = 'T';
 
+		} elsif ($cur =~ /^($Modifier)\s*/) {
+			print "MODIFIER($1)\n" if ($dbg_values > 1);
+			$type = 'T';
+
 		} elsif ($cur =~ /^(\#\s*define\s*$Ident)(\(?)/o) {
 			print "DEFINE($1,$2)\n" if ($dbg_values > 1);
 			$av_preprocessor = 1;

From 7429c6903e3628fc2cfea65ec7e13bac030c7bfe Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:29:06 -0700
Subject: [PATCH 0346/1171] checkpatch: improve type matcher debug

Improve type matcher debug so we can see what it does match.  As part
of this move us to to using the common debug framework.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index bcfb8ef00fee..077a2ca33043 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -17,7 +17,6 @@ my $quiet = 0;
 my $tree = 1;
 my $chk_signoff = 1;
 my $chk_patch = 1;
-my $tst_type = 0;
 my $tst_only;
 my $emacs = 0;
 my $terse = 0;
@@ -44,7 +43,6 @@ GetOptions(
 	'summary-file!'	=> \$summary_file,
 
 	'debug=s'	=> \%debug,
-	'test-type!'	=> \$tst_type,
 	'test-only=s'	=> \$tst_only,
 ) or exit;
 
@@ -67,6 +65,7 @@ if ($#ARGV < 0) {
 
 my $dbg_values = 0;
 my $dbg_possible = 0;
+my $dbg_type = 0;
 for my $key (keys %debug) {
 	eval "\${dbg_$key} = '$debug{$key}';"
 }
@@ -1307,8 +1306,12 @@ sub process {
 		if ($line=~/^[^\+]/) {next;}
 
 # TEST: allow direct testing of the type matcher.
-		if ($tst_type && $line =~ /^.$Declare$/) {
-			ERROR("TEST: is type $Declare\n" . $herecurr);
+		if ($dbg_type) {
+			if ($line =~ /^.\s*$Declare\s*$/) {
+				ERROR("TEST: is type\n" . $herecurr);
+			} elsif ($dbg_type > 1 && $line =~ /^.+($Declare)/) {
+				ERROR("TEST: is not type ($1 is)\n". $herecurr);
+			}
 			next;
 		}
 

From d2172eb5bd4b7d06577113ec40635083619ca54a Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:29:07 -0700
Subject: [PATCH 0347/1171] checkpatch: possible modifiers are not being
 correctly matched

Although we are finding the added modifier in the declaration below
we are not correctly matching it as a type.  Fix the declaration.

    static void __ref *vmem_alloc_pages(unsigned int order)
    {
    }

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 077a2ca33043..53ec3946670d 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -168,12 +168,11 @@ our @modifierList = (
 );
 
 sub build_types {
-	my $mods = "(?:  \n" . join("|\n  ", @modifierList) . "\n)";
-	my $all = "(?:  \n" . join("|\n  ", @typeList) . "\n)";
+	my $mods = "(?x:  \n" . join("|\n  ", @modifierList) . "\n)";
+	my $all = "(?x:  \n" . join("|\n  ", @typeList) . "\n)";
 	$Modifier	= qr{(?:$Attribute|$Sparse|$mods)};
 	$NonptrType	= qr{
-			(?:const\s+)?
-			(?:$mods\s+)?
+			(?:$Modifier\s+|const\s+)*
 			(?:
 				(?:typeof|__typeof__)\s*\(\s*\**\s*$Ident\s*\)|
 				(?:${all}\b)

From b8f96a31f38c8e9fc75f0a89c6815e7cbc402858 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:29:07 -0700
Subject: [PATCH 0348/1171] checkpatch: macro complexity checks are meaningless
 in linker scripts

Exclude vmlinux.lds.h from the macro complexity checks.  They will never
apply sanely here.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 53ec3946670d..775f2b146aa1 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -1860,7 +1860,8 @@ sub process {
 # multi-statement macros should be enclosed in a do while loop, grab the
 # first statement and ensure its the whole macro if its not enclosed
 # in a known good container
-		if ($line =~ /^.\s*\#\s*define\s*$Ident(\()?/) {
+		if ($realfile !~ m@/vmlinux.lds.h$@ &&
+		    $line =~ /^.\s*\#\s*define\s*$Ident(\()?/) {
 			my $ln = $linenr;
 			my $cnt = $realcnt;
 			my ($off, $dstat, $dcond, $rest);

From 8ea3eb9a20f39d5afa52900a34092b4b5f6b55cb Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:29:08 -0700
Subject: [PATCH 0349/1171] checkpatch: handle return types of pointers to
 functions

Make sure we correctly mark the return type of the pointer to a function
declaration.

    const void *(*sb_tag)(struct sysfs_tag_info *info);

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 775f2b146aa1..6d07b6778c91 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -715,7 +715,7 @@ sub annotate_values {
 				$av_preprocessor = 0;
 			}
 
-		} elsif ($cur =~ /^($Type)\s*(?:$Ident|,|\))/) {
+		} elsif ($cur =~ /^($Type)\s*(?:$Ident|,|\)|\()/) {
 			print "DECLARE($1)\n" if ($dbg_values > 1);
 			$type = 'T';
 

From 0221f55c142b0ac8baf6f0b6c4e1ec89f0c98e96 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:29:08 -0700
Subject: [PATCH 0350/1171] checkpatch: possible types -- known modifiers
 cannot be types

Ensure we do not inadvertantly load known modifiers up as possible types.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 6d07b6778c91..9c209165f252 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -847,7 +847,7 @@ sub possible {
 	my ($possible, $line) = @_;
 
 	print "CHECK<$possible> ($line)\n" if ($dbg_possible > 1);
-	if ($possible !~ /^(?:$Storage|$Type|DEFINE_\S+)$/ &&
+	if ($possible !~ /^(?:$Modifier|$Storage|$Type|DEFINE_\S+)$/ &&
 	    $possible ne 'goto' && $possible ne 'return' &&
 	    $possible ne 'case' && $possible ne 'else' &&
 	    $possible ne 'asm' && $possible ne '__asm__' &&

From d2506586586c59f5db0e2ce00d5d31ccec6260b8 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:29:09 -0700
Subject: [PATCH 0351/1171] checkpatch: possible modifiers -- handle multiple
 modifiers and trailing

Add support for multiple modifiers such as:

	int __one __two foo;

Also handle trailing known modifiers when defecting modifiers:

	int __one foo __read_mostly;

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 9c209165f252..8a3b0fd67ad7 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -859,8 +859,10 @@ sub possible {
 
 		} elsif ($possible =~ /\s/) {
 			$possible =~ s/\s*$Type\s*//g;
-			warn "MODIFIER: $possible ($line)\n" if ($dbg_possible);
-			push(@modifierList, $possible);
+			for my $modifier (split(' ', $possible)) {
+				warn "MODIFIER: $modifier ($possible) ($line)\n" if ($dbg_possible);
+				push(@modifierList, $modifier);
+			}
 
 		} else {
 			warn "POSSIBLE: $possible ($line)\n" if ($dbg_possible);
@@ -1186,7 +1188,7 @@ sub process {
 			} elsif ($s =~ /^.\s*$Ident\s*\(/s) {
 
 			# declarations always start with types
-			} elsif ($prev_values eq 'E' && $s =~ /^.\s*(?:$Storage\s+)?(?:$Inline\s+)?(?:const\s+)?((?:\s*$Ident)+)\b(?:\s+$Sparse)?\s*\**\s*(?:$Ident|\(\*[^\)]*\))\s*(?:;|=|,|\()/s) {
+			} elsif ($prev_values eq 'E' && $s =~ /^.\s*(?:$Storage\s+)?(?:$Inline\s+)?(?:const\s+)?((?:\s*$Ident)+?)\b(?:\s+$Sparse)?\s*\**\s*(?:$Ident|\(\*[^\)]*\))(?:\s*$Modifier)?\s*(?:;|=|,|\()/s) {
 				my $type = $1;
 				$type =~ s/\s+/ /g;
 				possible($type, "A:" . $s);

From 1f65f947a6a875e1fe7867dc08e981c4101d435d Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:29:10 -0700
Subject: [PATCH 0352/1171] checkpatch: add checks for question mark and colon
 spacing

Add checks for the question mark colon operator spacing, and also check
the other uses of colon.  Colon means a number of things:

 - it introduces the else part of the ?: operator,
 - it terminates a goto label,
 - it terminates the case value,
 - it separates the identifier from the bit size on bit fields, and
 - it is used to introduce option types in asm().

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 81 ++++++++++++++++++++++++++++++++++++++-----
 1 file changed, 72 insertions(+), 9 deletions(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 8a3b0fd67ad7..88027f237cdf 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -689,17 +689,20 @@ sub cat_vet {
 my $av_preprocessor = 0;
 my $av_pending;
 my @av_paren_type;
+my $av_pend_colon;
 
 sub annotate_reset {
 	$av_preprocessor = 0;
 	$av_pending = '_';
 	@av_paren_type = ('E');
+	$av_pend_colon = 'O';
 }
 
 sub annotate_values {
 	my ($stream, $type) = @_;
 
 	my $res;
+	my $var = '_' x length($stream);
 	my $cur = $stream;
 
 	print "$stream\n" if ($dbg_values > 1);
@@ -784,7 +787,12 @@ sub annotate_values {
 			$av_pending = 'N';
 			$type = 'N';
 
-		} elsif ($cur =~/^(return|case|else|goto)/o) {
+		} elsif ($cur =~/^(case)/o) {
+			print "CASE($1)\n" if ($dbg_values > 1);
+			$av_pend_colon = 'C';
+			$type = 'N';
+
+		} elsif ($cur =~/^(return|else|goto)/o) {
 			print "KEYWORD($1)\n" if ($dbg_values > 1);
 			$type = 'N';
 
@@ -809,6 +817,15 @@ sub annotate_values {
 			$type = 'V';
 			$av_pending = 'V';
 
+		} elsif ($cur =~ /^($Ident\s*):/) {
+			if ($type eq 'E') {
+				$av_pend_colon = 'L';
+			} elsif ($type eq 'T') {
+				$av_pend_colon = 'B';
+			}
+			print "IDENT_COLON($1,$type>$av_pend_colon)\n" if ($dbg_values > 1);
+			$type = 'V';
+
 		} elsif ($cur =~ /^($Ident|$Constant)/o) {
 			print "IDENT($1)\n" if ($dbg_values > 1);
 			$type = 'V';
@@ -820,8 +837,24 @@ sub annotate_values {
 		} elsif ($cur =~/^(;|{|})/) {
 			print "END($1)\n" if ($dbg_values > 1);
 			$type = 'E';
+			$av_pend_colon = 'O';
 
-		} elsif ($cur =~ /^(;|\?|:|\[)/o) {
+		} elsif ($cur =~ /^(\?)/o) {
+			print "QUESTION($1)\n" if ($dbg_values > 1);
+			$type = 'N';
+
+		} elsif ($cur =~ /^(:)/o) {
+			print "COLON($1,$av_pend_colon)\n" if ($dbg_values > 1);
+
+			substr($var, length($res), 1, $av_pend_colon);
+			if ($av_pend_colon eq 'C' || $av_pend_colon eq 'L') {
+				$type = 'E';
+			} else {
+				$type = 'N';
+			}
+			$av_pend_colon = 'O';
+
+		} elsif ($cur =~ /^(;|\[)/o) {
 			print "CLOSE($1)\n" if ($dbg_values > 1);
 			$type = 'N';
 
@@ -840,7 +873,7 @@ sub annotate_values {
 		}
 	}
 
-	return $res;
+	return ($res, $var);
 }
 
 sub possible {
@@ -1294,12 +1327,14 @@ sub process {
 
 		# Track the 'values' across context and added lines.
 		my $opline = $line; $opline =~ s/^./ /;
-		my $curr_values = annotate_values($opline . "\n", $prev_values);
+		my ($curr_values, $curr_vars) =
+				annotate_values($opline . "\n", $prev_values);
 		$curr_values = $prev_values . $curr_values;
 		if ($dbg_values) {
 			my $outline = $opline; $outline =~ s/\t/ /g;
 			print "$linenr > .$outline\n";
 			print "$linenr > $curr_values\n";
+			print "$linenr >  $curr_vars\n";
 		}
 		$prev_values = substr($curr_values, -1);
 
@@ -1490,7 +1525,8 @@ sub process {
 				<<=|>>=|<=|>=|==|!=|
 				\+=|-=|\*=|\/=|%=|\^=|\|=|&=|
 				=>|->|<<|>>|<|>|=|!|~|
-				&&|\|\||,|\^|\+\+|--|&|\||\+|-|\*|\/|%
+				&&|\|\||,|\^|\+\+|--|&|\||\+|-|\*|\/|%|
+				\?|:
 			}x;
 			my @elements = split(/($ops|;)/, $opline);
 			my $off = 0;
@@ -1554,6 +1590,9 @@ sub process {
 				#	print "UNARY: <$op_left$op_type $is_unary $a:$op:$c> <$ca:$op:$cc> <$unary_ctx>\n";
 				#}
 
+				# Get the full operator variant.
+				my $opv = $op . substr($curr_vars, $off, 1);
+
 				# Ignore operators passed as parameters.
 				if ($op_type ne 'V' &&
 				    $ca =~ /\s$/ && $cc =~ /^\s*,/) {
@@ -1571,8 +1610,10 @@ sub process {
 				# // is a comment
 				} elsif ($op eq '//') {
 
-				# -> should have no spaces
-				} elsif ($op eq '->') {
+				# No spaces for:
+				#   ->
+				#   :   when part of a bitfield
+				} elsif ($op eq '->' || $opv eq ':B') {
 					if ($ctx =~ /Wx.|.xW/) {
 						ERROR("spaces prohibited around that '$op' $at\n" . $hereptr);
 					}
@@ -1628,11 +1669,33 @@ sub process {
 							$hereptr);
 					}
 
+				# A colon needs no spaces before when it is
+				# terminating a case value or a label.
+				} elsif ($opv eq ':C' || $opv eq ':L') {
+					if ($ctx =~ /Wx./) {
+						ERROR("space prohibited before that '$op' $at\n" . $hereptr);
+					}
+
 				# All the others need spaces both sides.
 				} elsif ($ctx !~ /[EWC]x[CWE]/) {
+					my $ok = 0;
+
 					# Ignore email addresses <foo@bar>
-					if (!($op eq '<' && $cb =~ /$;\S+\@\S+>/) &&
-					    !($op eq '>' && $cb =~ /<\S+\@\S+$;/)) {
+					if (($op eq '<' &&
+					     $cc =~ /^\S+\@\S+>/) ||
+					    ($op eq '>' &&
+					     $ca =~ /<\S+\@\S+$/))
+					{
+					    	$ok = 1;
+					}
+
+					# Ignore ?:
+					if (($opv eq ':O' && $ca =~ /\?$/) ||
+					    ($op eq '?' && $cc =~ /^:/)) {
+					    	$ok = 1;
+					}
+
+					if ($ok == 0) {
 						ERROR("spaces required around that '$op' $at\n" . $hereptr);
 					}
 				}

From 74048ed811152a995a88945ba9e0dded34adfff4 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:29:10 -0700
Subject: [PATCH 0353/1171] checkpatch: variants -- move the main unary/binary
 operators to use variants

Now that we have a variants system, move to using that to carry the
unary/binary designation for +, -, &, and *.

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 36 ++++++++++++++++++------------------
 1 file changed, 18 insertions(+), 18 deletions(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 88027f237cdf..8afa88aaed96 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -858,6 +858,19 @@ sub annotate_values {
 			print "CLOSE($1)\n" if ($dbg_values > 1);
 			$type = 'N';
 
+		} elsif ($cur =~ /^(-(?![->])|\+(?!\+)|\*|\&(?!\&))/o) {
+			my $variant;
+
+			print "OPV($1)\n" if ($dbg_values > 1);
+			if ($type eq 'V') {
+				$variant = 'B';
+			} else {
+				$variant = 'U';
+			}
+
+			substr($var, length($res), 1, $variant);
+			$type = 'N';
+
 		} elsif ($cur =~ /^($Operators)/o) {
 			print "OP($1)\n" if ($dbg_values > 1);
 			if ($1 ne '++' && $1 ne '--') {
@@ -1573,22 +1586,8 @@ sub process {
 				my $ptr = substr($blank, 0, $off) . "^";
 				my $hereptr = "$hereline$ptr\n";
 
-				# Classify operators into binary, unary, or
-				# definitions (* only) where they have more
-				# than one mode.
+				# Pull out the value of this operator.
 				my $op_type = substr($curr_values, $off + 1, 1);
-				my $op_left = substr($curr_values, $off, 1);
-				my $is_unary;
-				if ($op_type eq 'T') {
-					$is_unary = 2;
-				} elsif ($op_left eq 'V') {
-					$is_unary = 0;
-				} else {
-					$is_unary = 1;
-				}
-				#if ($op eq '-' || $op eq '&' || $op eq '*') {
-				#	print "UNARY: <$op_left$op_type $is_unary $a:$op:$c> <$ca:$op:$cc> <$unary_ctx>\n";
-				#}
 
 				# Get the full operator variant.
 				my $opv = $op . substr($curr_vars, $off, 1);
@@ -1625,18 +1624,19 @@ sub process {
 					}
 
 				# '*' as part of a type definition -- reported already.
-				} elsif ($op eq '*' && $is_unary == 2) {
+				} elsif ($opv eq '*_') {
 					#warn "'*' is part of type\n";
 
 				# unary operators should have a space before and
 				# none after.  May be left adjacent to another
 				# unary operator, or a cast
 				} elsif ($op eq '!' || $op eq '~' ||
-				         ($is_unary && ($op eq '*' || $op eq '-' || $op eq '&'))) {
+					 $opv eq '*U' || $opv eq '-U' ||
+					 $opv eq '&U') {
 					if ($ctx !~ /[WEBC]x./ && $ca !~ /(?:\)|!|~|\*|-|\&|\||\+\+|\-\-|\{)$/) {
 						ERROR("space required before that '$op' $at\n" . $hereptr);
 					}
-					if ($op  eq '*' && $cc =~/\s*const\b/) {
+					if ($op eq '*' && $cc =~/\s*const\b/) {
 						# A unary '*' may be const
 
 					} elsif ($ctx =~ /.xW/) {

From 292f1a9b342d763f94ea3915726a48905be4acd1 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:29:11 -0700
Subject: [PATCH 0354/1171] checkpatch: complex macros need to ignore comments

Ensure we ignore comments in complex macro detection else we incorrectly
report this:

	#define PFM_GROUP_PERM_ANY     -1      /* any user/group */

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 1 +
 1 file changed, 1 insertion(+)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 8afa88aaed96..96a762be5748 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -1972,6 +1972,7 @@ sub process {
 			} else {
 				$dstat =~ s/^.\s*\#\s*define\s+$Ident\s*//;
 			}
+			$dstat =~ s/$;//g;
 			$dstat =~ s/\\\n.//g;
 			$dstat =~ s/^\s*//s;
 			$dstat =~ s/\s*$//s;

From 234fff6515a11cf3e67c793146689da426787fea Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:29:12 -0700
Subject: [PATCH 0355/1171] checkpatch: types cannot start mid word for pointer
 tests

When checking spacing for pointer checks the type cannot start in the
middle of a word, ie. this is not 'int * bar':

	x = fooint * bar;

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 96a762be5748..022ee557b681 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -1435,11 +1435,11 @@ sub process {
 			ERROR("\"(foo $1 )\" should be \"(foo $1)\"\n" .
 				$herecurr);
 
-		} elsif ($line =~ m{$NonptrType(\*+)(?:\s+(?:$Attribute|$Sparse))?\s+[A-Za-z\d_]+}) {
+		} elsif ($line =~ m{\b$NonptrType(\*+)(?:\s+(?:$Attribute|$Sparse))?\s+[A-Za-z\d_]+}) {
 			ERROR("\"foo$1 bar\" should be \"foo $1bar\"\n" .
 				$herecurr);
 
-		} elsif ($line =~ m{$NonptrType\s+(\*+)(?!\s+(?:$Attribute|$Sparse))\s+[A-Za-z\d_]+}) {
+		} elsif ($line =~ m{\b$NonptrType\s+(\*+)(?!\s+(?:$Attribute|$Sparse))\s+[A-Za-z\d_]+}) {
 			ERROR("\"foo $1 bar\" should be \"foo $1bar\"\n" .
 				$herecurr);
 		}

From 33cba0657393a75e18e1781e3e13613303f18124 Mon Sep 17 00:00:00 2001
From: Andy Whitcroft <apw@shadowen.org>
Date: Wed, 23 Jul 2008 21:29:12 -0700
Subject: [PATCH 0356/1171] checkpatch: version 0.21

Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkpatch.pl | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 022ee557b681..bc6779398229 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -9,7 +9,7 @@ use strict;
 my $P = $0;
 $P =~ s@.*/@@g;
 
-my $V = '0.20';
+my $V = '0.21';
 
 use Getopt::Long qw(:config no_auto_abbrev);
 

From 7102ed519a08b70eadc8fea9d8765d2d990241d1 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Wed, 23 Jul 2008 21:29:13 -0700
Subject: [PATCH 0357/1171] remove the OSS trident driver

SOUND_TRIDENT was the last PCI OSS driver, and since there's already an
ALSA driver for the same hardware we can remove it.

[muli@il.ibm.com: update CREDITS]
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Muli Ben-Yehuda <muli@il.ibm.com>
Signed-off-by: Muli Ben-Yehuda <muli@il.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 CREDITS             |    8 +
 MAINTAINERS         |    6 -
 sound/oss/Kconfig   |   41 -
 sound/oss/Makefile  |    1 -
 sound/oss/trident.c | 4654 -------------------------------------------
 sound/oss/trident.h |  358 ----
 6 files changed, 8 insertions(+), 5060 deletions(-)
 delete mode 100644 sound/oss/trident.c
 delete mode 100644 sound/oss/trident.h

diff --git a/CREDITS b/CREDITS
index 077b147388bd..c62dcb3b7e26 100644
--- a/CREDITS
+++ b/CREDITS
@@ -317,6 +317,14 @@ S: 2322 37th Ave SW
 S: Seattle, Washington 98126-2010
 S: USA
 
+N: Muli Ben-Yehuda
+E: mulix@mulix.org
+E: muli@il.ibm.com
+W: http://www.mulix.org
+D: trident OSS sound driver, x86-64 dma-ops and Calgary IOMMU,
+D: KVM and Xen bits and other misc. hackery.
+S: Haifa, Israel
+
 N: Johannes Berg
 E: johannes@sipsolutions.net
 W: http://johannes.sipsolutions.net/
diff --git a/MAINTAINERS b/MAINTAINERS
index 7e5c7b0290bb..5ecb97e13e59 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -4080,12 +4080,6 @@ W:	http://www.prosec.rub.de/tpm/
 L:	tpmdd-devel@lists.sourceforge.net
 S:	Maintained
 
-TRIDENT 4DWAVE/SIS 7018 PCI AUDIO CORE
-P:	Muli Ben-Yehuda
-M:	mulix@mulix.org
-L:	linux-kernel@vger.kernel.org
-S:	Maintained
-
 TRIVIAL PATCHES
 P:	Jesper Juhl
 M:	trivial@kernel.org
diff --git a/sound/oss/Kconfig b/sound/oss/Kconfig
index 33940139844b..d4fafb6eec6c 100644
--- a/sound/oss/Kconfig
+++ b/sound/oss/Kconfig
@@ -35,47 +35,6 @@ config SOUND_AU1550_AC97
 	tristate "Au1550/Au1200 AC97 Sound"
 	depends on SOC_AU1550 || SOC_AU1200
 
-config SOUND_TRIDENT
-	tristate "Trident 4DWave DX/NX, SiS 7018 or ALi 5451 PCI Audio Core"
-	depends on PCI
-	---help---
-	  Say Y or M if you have a PCI sound card utilizing the Trident
-	  4DWave-DX/NX chipset or your mother board chipset has SiS 7018
-	  or ALi 5451 built-in. The SiS 7018 PCI Audio Core is embedded
-	  in SiS960 Super South Bridge and SiS540/630 Single Chipset.
-	  The ALi 5451 PCI Audio Core is embedded in ALi M1535, M1535D,
-	  M1535+ or M1535D+ South Bridge.
-
-	  Use lspci -n to find out if your sound card or chipset uses
-	  Trident 4DWave or SiS 7018. PCI ID 1023:2000 or 1023:2001 stands
-	  for Trident 4Dwave. PCI ID 1039:7018 stands for SiS7018. PCI ID
-	  10B9:5451 stands for ALi5451.
-
-	  This driver supports S/PDIF in/out (record/playback) for ALi 5451
-	  embedded in ALi M1535+ and M1535D+. Note that they aren't all
-	  enabled by default; you can enable them by saying Y to "/proc file
-	  system support" and "Sysctl support", and after the /proc file
-	  system has been mounted, executing the command
-
-		command			what is enabled
-
-	  echo 0>/proc/ALi5451	pcm out is also set to S/PDIF out. (Default).
-
-	  echo 1>/proc/ALi5451	use S/PDIF out to output pcm data.
-
-	  echo 2>/proc/ALi5451	use S/PDIF out to output non-pcm data.
-	  (AC3...).
-
-	  echo 3>/proc/ALi5451	record from Ac97 in(MIC, Line in...).
-	  (Default).
-
-	  echo 4>/proc/ALi5451	no matter Ac97 settings, record from S/PDIF
-	  in.
-
-
-	  This driver differs slightly from OSS/Free, so PLEASE READ the
-	  comments at the top of <file:sound/oss/trident.c>.
-
 config SOUND_MSNDCLAS
 	tristate "Support for Turtle Beach MultiSound Classic, Tahiti, Monterey"
 	depends on (m || !STANDALONE) && ISA
diff --git a/sound/oss/Makefile b/sound/oss/Makefile
index 1f86299fae40..3a141474fb73 100644
--- a/sound/oss/Makefile
+++ b/sound/oss/Makefile
@@ -29,7 +29,6 @@ obj-$(CONFIG_SOUND_MSNDCLAS)	+= msnd.o msnd_classic.o
 obj-$(CONFIG_SOUND_MSNDPIN)	+= msnd.o msnd_pinnacle.o
 obj-$(CONFIG_SOUND_VWSND)	+= vwsnd.o
 obj-$(CONFIG_SOUND_AU1550_AC97)	+= au1550_ac97.o ac97_codec.o
-obj-$(CONFIG_SOUND_TRIDENT)	+= trident.o ac97_codec.o
 obj-$(CONFIG_SOUND_BCM_CS4297A)	+= swarm_cs4297a.o
 
 obj-$(CONFIG_SOUND_WM97XX)	+= ac97_plugin_wm97xx.o
diff --git a/sound/oss/trident.c b/sound/oss/trident.c
deleted file mode 100644
index f43f91ef86c7..000000000000
--- a/sound/oss/trident.c
+++ /dev/null
@@ -1,4654 +0,0 @@
-/*
- *	OSS driver for Linux 2.[46].x for
- *
- *	Trident 4D-Wave
- *	SiS 7018
- *	ALi 5451
- *	Tvia/IGST CyberPro 5050
- *
- *	Driver: Alan Cox <alan@redhat.com>
- *
- *  Built from:
- *	Low level code: <audio@tridentmicro.com> from ALSA
- *	Framework: Thomas Sailer <sailer@ife.ee.ethz.ch>
- *	Extended by: Zach Brown <zab@redhat.com>
- *
- *  Hacked up by:
- *	Aaron Holtzman <aholtzma@ess.engr.uvic.ca>
- *	Ollie Lho <ollie@sis.com.tw> SiS 7018 Audio Core Support
- *	Ching-Ling Lee <cling-li@ali.com.tw> ALi 5451 Audio Core Support
- *	Matt Wu <mattwu@acersoftech.com.cn> ALi 5451 Audio Core Support
- *	Peter Wächtler <pwaechtler@loewe-komp.de> CyberPro5050 support
- *      Muli Ben-Yehuda <mulix@mulix.org>
- *
- *
- *	This program is free software; you can redistribute it and/or modify
- *	it under the terms of the GNU General Public License as published by
- *	the Free Software Foundation; either version 2 of the License, or
- *	(at your option) any later version.
- *
- *	This program is distributed in the hope that it will be useful,
- *	but WITHOUT ANY WARRANTY; without even the implied warranty of
- *	MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- *	GNU General Public License for more details.
- *
- *	You should have received a copy of the GNU General Public License
- *	along with this program; if not, write to the Free Software
- *	Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
- *
- *  History
- *  v0.14.10j
- *  	January 3 2004 Eugene Teo <eugeneteo@eugeneteo.net>
- *  	minor cleanup to use pr_debug instead of TRDBG since it is already
- *  	defined in linux/kernel.h.
- *  v0.14.10i
- *      December 29 2003 Muli Ben-Yehuda <mulix@mulix.org>
- *      major cleanup for 2.6, fix a few error patch buglets
- *      with returning without properly cleaning up first,
- *      get rid of lock_kernel().
- *  v0.14.10h
- *	Sept 10 2002 Pascal Schmidt <der.eremit@email.de>
- *	added support for ALi 5451 joystick port
- *  v0.14.10g
- *	Sept 05 2002 Alan Cox <alan@redhat.com>
- *	adapt to new pci joystick attachment interface
- *  v0.14.10f
- *      July 24 2002 Muli Ben-Yehuda <mulix@actcom.co.il>
- *      patch from Eric Lemar (via Ian Soboroff): in suspend and resume,
- *      fix wrong cast from pci_dev* to struct trident_card*.
- *  v0.14.10e
- *      July 19 2002 Muli Ben-Yehuda <mulix@actcom.co.il>
- *      rewrite the DMA buffer allocation/deallcoation functions, to make it
- *      modular and fix a bug where we would call free_pages on memory
- *      obtained with pci_alloc_consistent. Also remove unnecessary #ifdef
- *      CONFIG_PROC_FS and various other cleanups.
- *  v0.14.10d
- *      July 19 2002 Muli Ben-Yehuda <mulix@actcom.co.il>
- *      made several printk(KERN_NOTICE...) into TRDBG(...), to avoid spamming
- *      my syslog with hundreds of messages.
- *  v0.14.10c
- *      July 16 2002 Muli Ben-Yehuda <mulix@actcom.co.il>
- *      Cleaned up Lei Hu's 0.4.10 driver to conform to Documentation/CodingStyle
- *      and the coding style used in the rest of the file.
- *  v0.14.10b
- *      June 23 2002 Muli Ben-Yehuda <mulix@actcom.co.il>
- *      add a missing unlock_set_fmt, remove a superflous lock/unlock pair
- *      with nothing in between.
- *  v0.14.10a
- *      June 21 2002 Muli Ben-Yehuda <mulix@actcom.co.il>
- *      use a debug macro instead of #ifdef CONFIG_DEBUG, trim to 80 columns
- *      per line, use 'do {} while (0)' in statement macros.
- *  v0.14.10
- *      June 6 2002 Lei Hu <Lei_hu@ali.com.tw>
- *      rewrite the part to read/write registers of audio codec for Ali5451
- *  v0.14.9e
- *      January 2 2002 Vojtech Pavlik <vojtech@ucw.cz> added gameport
- *      support to avoid resource conflict with pcigame.c
- *  v0.14.9d
- *  	October 8 2001 Arnaldo Carvalho de Melo <acme@conectiva.com.br>
- *	use set_current_state, properly release resources on failure in
- *	trident_probe, get rid of check_region
- *  v0.14.9c
- *	August 10 2001 Peter Wächtler <pwaechtler@loewe-komp.de>
- *	added support for Tvia (formerly Integraphics/IGST) CyberPro5050
- *	this chip is often found in settop boxes (combined video+audio)
- *  v0.14.9b
- *	Switch to static inline not extern inline (gcc 3)
- *  v0.14.9a
- *	Aug 6 2001 Alan Cox
- *	0.14.9 crashed on rmmod due to a timer/bh left running. Simplified
- *	the existing logic (the BH doesn't help as ac97 is lock_irqsave)
- *	and used del_timer_sync to clean up
- *	Fixed a problem where the ALi change broke my generic card
- *  v0.14.9
- *	Jul 10 2001 Matt Wu
- *	Add H/W Volume Control
- *  v0.14.8a
- *	July 7 2001 Alan Cox
- *	Moved Matt Wu's ac97 register cache into the card structure
- *  v0.14.8
- *	Apr 30 2001 Matt Wu
- *	Set EBUF1 and EBUF2 to still mode
- *	Add dc97/ac97 reset function
- *	Fix power management: ali_restore_regs
- *  unreleased
- *	Mar 09 2001 Matt Wu
- *	Add cache for ac97 access
- *  v0.14.7
- *	Feb 06 2001 Matt Wu
- *	Fix ac97 initialization
- *	Fix bug: an extra tail will be played when playing
- *	Jan 05 2001 Matt Wu
- *	Implement multi-channels and S/PDIF in support for ALi 1535+
- *  v0.14.6
- *	Nov 1 2000 Ching-Ling Lee
- *	Fix the bug of memory leak when switching 5.1-channels to 2 channels.
- *	Add lock protection into dynamic changing format of data.
- *	Oct 18 2000 Ching-Ling Lee
- *	5.1-channels support for ALi
- *	June 28 2000 Ching-Ling Lee
- *	S/PDIF out/in(playback/record) support for ALi 1535+, using /proc to be selected by user
- *	Simple Power Management support for ALi
- *  v0.14.5 May 23 2000 Ollie Lho
- *  	Misc bug fix from the Net
- *  v0.14.4 May 20 2000 Aaron Holtzman
- *  	Fix kfree'd memory access in release
- *  	Fix race in open while looking for a free virtual channel slot
- *  	remove open_wait wq (which appears to be unused)
- *  v0.14.3 May 10 2000 Ollie Lho
- *	fixed a small bug in trident_update_ptr, xmms 1.0.1 no longer uses 100% CPU
- *  v0.14.2 Mar 29 2000 Ching-Ling Lee
- *	Add clear to silence advance in trident_update_ptr
- *	fix invalid data of the end of the sound
- *  v0.14.1 Mar 24 2000 Ching-Ling Lee
- *	ALi 5451 support added, playback and recording O.K.
- *	ALi 5451 originally developed and structured based on sonicvibes, and
- *	suggested to merge into this file by Alan Cox.
- *  v0.14 Mar 15 2000 Ollie Lho
- *	5.1 channel output support with channel binding. What's the Matrix ?
- *  v0.13.1 Mar 10 2000 Ollie Lho
- *	few minor bugs on dual codec support, needs more testing
- *  v0.13 Mar 03 2000 Ollie Lho
- *	new pci_* for 2.4 kernel, back ported to 2.2
- *  v0.12 Feb 23 2000 Ollie Lho
- *	Preliminary Recording support
- *  v0.11.2 Feb 19 2000 Ollie Lho
- *	removed incomplete full-dulplex support
- *  v0.11.1 Jan 28 2000 Ollie Lho
- *	small bug in setting sample rate for 4d-nx (reported by Aaron)
- *  v0.11 Jan 27 2000 Ollie Lho
- *	DMA bug, scheduler latency, second try
- *  v0.10 Jan 24 2000 Ollie Lho
- *	DMA bug fixed, found kernel scheduling problem
- *  v0.09 Jan 20 2000 Ollie Lho
- *	Clean up of channel register access routine (prepare for channel binding)
- *  v0.08 Jan 14 2000 Ollie Lho
- *	Isolation of AC97 codec code
- *  v0.07 Jan 13 2000 Ollie Lho
- *	Get rid of ugly old low level access routines (e.g. CHRegs.lp****)
- *  v0.06 Jan 11 2000 Ollie Lho
- *	Preliminary support for dual (more ?) AC97 codecs
- *  v0.05 Jan 08 2000 Luca Montecchiani <m.luca@iname.com>
- *	adapt to 2.3.x new __setup/__init call
- *  v0.04 Dec 31 1999 Ollie Lho
- *	Multiple Open, using Middle Loop Interrupt to smooth playback
- *  v0.03 Dec 24 1999 Ollie Lho
- *	mem leak in prog_dmabuf and dealloc_dmabuf removed
- *  v0.02 Dec 15 1999 Ollie Lho
- *	SiS 7018 support added, playback O.K.
- *  v0.01 Alan Cox et. al.
- *	Initial Release in kernel 2.3.30, does not work
- *
- *  ToDo
- *	Clean up of low level channel register access code. (done)
- *	Fix the bug on dma buffer management in update_ptr, read/write, drain_dac (done)
- *	Dual AC97 codecs support (done)
- *	Recording support (done)
- *	Mmap support
- *	"Channel Binding" ioctl extension (done)
- *	new pci device driver interface for 2.4 kernel (done)
- *
- *	Lock order (high->low)
- *		lock	-	hardware lock
- *		open_mutex - 	guard opens
- *		sem	-	guard dmabuf, write re-entry etc
- */
-
-#include <linux/module.h>
-#include <linux/string.h>
-#include <linux/ctype.h>
-#include <linux/ioport.h>
-#include <linux/sched.h>
-#include <linux/delay.h>
-#include <linux/sound.h>
-#include <linux/slab.h>
-#include <linux/soundcard.h>
-#include <linux/pci.h>
-#include <linux/init.h>
-#include <linux/poll.h>
-#include <linux/spinlock.h>
-#include <linux/ac97_codec.h>
-#include <linux/bitops.h>
-#include <linux/proc_fs.h>
-#include <linux/interrupt.h>
-#include <linux/pm.h>
-#include <linux/gameport.h>
-#include <linux/kernel.h>
-#include <linux/mutex.h>
-#include <linux/mm.h>
-
-#include <asm/uaccess.h>
-#include <asm/io.h>
-#include <asm/dma.h>
-
-#if defined(CONFIG_ALPHA_NAUTILUS) || defined(CONFIG_ALPHA_GENERIC)
-#include <asm/hwrpb.h>
-#endif
-
-#include "trident.h"
-
-#define DRIVER_VERSION "0.14.10j-2.6"
-
-#if defined(CONFIG_GAMEPORT) || (defined(MODULE) && defined(CONFIG_GAMEPORT_MODULE))
-#define SUPPORT_JOYSTICK 1
-#endif
-
-/* magic numbers to protect our data structures */
-#define TRIDENT_CARD_MAGIC	0x5072696E	/* "Prin" */
-#define TRIDENT_STATE_MAGIC	0x63657373	/* "cess" */
-
-#define TRIDENT_DMA_MASK	0x3fffffff	/* DMA buffer mask for pci_alloc_consist */
-#define ALI_DMA_MASK		0x7fffffff	/* ALI Tridents have 31-bit DMA. Wow. */
-
-#define NR_HW_CH		32
-
-/* maximum number of AC97 codecs connected, AC97 2.0 defined 4, but 7018 and 4D-NX only
-   have 2 SDATA_IN lines (currently) */
-#define NR_AC97		2
-
-/* minor number of /dev/swmodem (temporary, experimental) */
-#define SND_DEV_SWMODEM	7
-
-static const unsigned ali_multi_channels_5_1[] = {
-	/*ALI_SURR_LEFT_CHANNEL, ALI_SURR_RIGHT_CHANNEL, */
-	ALI_CENTER_CHANNEL,
-	ALI_LEF_CHANNEL,
-	ALI_SURR_LEFT_CHANNEL,
-	ALI_SURR_RIGHT_CHANNEL
-};
-
-static const unsigned sample_size[] = { 1, 2, 2, 4 };
-static const unsigned sample_shift[] = { 0, 1, 1, 2 };
-
-static const char invalid_magic[] = KERN_CRIT "trident: invalid magic value in %s\n";
-
-enum {
-	TRIDENT_4D_DX = 0,
-	TRIDENT_4D_NX,
-	SIS_7018,
-	ALI_5451,
-	CYBER5050
-};
-
-static char *card_names[] = {
-	"Trident 4DWave DX",
-	"Trident 4DWave NX",
-	"SiS 7018 PCI Audio",
-	"ALi Audio Accelerator",
-	"Tvia/IGST CyberPro 5050"
-};
-
-static struct pci_device_id trident_pci_tbl[] = {
-	{PCI_DEVICE(PCI_VENDOR_ID_TRIDENT, PCI_DEVICE_ID_TRIDENT_4DWAVE_DX),
-		PCI_CLASS_MULTIMEDIA_AUDIO << 8, 0xffff00, TRIDENT_4D_DX},
-	{PCI_DEVICE(PCI_VENDOR_ID_TRIDENT, PCI_DEVICE_ID_TRIDENT_4DWAVE_NX),
-		0, 0, TRIDENT_4D_NX},
-	{PCI_DEVICE(PCI_VENDOR_ID_SI, PCI_DEVICE_ID_SI_7018), 0, 0, SIS_7018},
-	{PCI_DEVICE(PCI_VENDOR_ID_ALI, PCI_DEVICE_ID_ALI_5451), 0, 0, ALI_5451},
-	{PCI_DEVICE(PCI_VENDOR_ID_INTERG, PCI_DEVICE_ID_INTERG_5050),
-		0, 0, CYBER5050},
-	{0,}
-};
-
-MODULE_DEVICE_TABLE(pci, trident_pci_tbl);
-
-/* "software" or virtual channel, an instance of opened /dev/dsp */
-struct trident_state {
-	unsigned int magic;
-	struct trident_card *card;	/* Card info */
-
-	/* file mode */
-	mode_t open_mode;
-
-	/* virtual channel number */
-	int virt;
-
-	struct dmabuf {
-		/* wave sample stuff */
-		unsigned int rate;
-		unsigned char fmt, enable;
-
-		/* hardware channel */
-		struct trident_channel *channel;
-
-		/* OSS buffer management stuff */
-		void *rawbuf;
-		dma_addr_t dma_handle;
-		unsigned buforder;
-		unsigned numfrag;
-		unsigned fragshift;
-
-		/* our buffer acts like a circular ring */
-		unsigned hwptr;	/* where dma last started, updated by update_ptr */
-		unsigned swptr;	/* where driver last clear/filled, updated by read/write */
-		int count;	/* bytes to be comsumed or been generated by dma machine */
-		unsigned total_bytes;	/* total bytes dmaed by hardware */
-
-		unsigned error;	/* number of over/underruns */
-                /* put process on wait queue when no more space in buffer */
-		wait_queue_head_t wait;
-
-		/* redundant, but makes calculations easier */
-		unsigned fragsize;
-		unsigned dmasize;
-		unsigned fragsamples;
-
-		/* OSS stuff */
-		unsigned mapped:1;
-		unsigned ready:1;
-		unsigned endcleared:1;
-		unsigned update_flag;
-		unsigned ossfragshift;
-		int ossmaxfrags;
-		unsigned subdivision;
-
-	} dmabuf;
-
-	/* 5.1 channels */
-	struct trident_state *other_states[4];
-	int multi_channels_adjust_count;
-	unsigned chans_num;
-	unsigned long fmt_flag;
-	/* Guard against mmap/write/read races */
-	struct mutex sem;
-
-};
-
-/* hardware channels */
-struct trident_channel {
-	int num; /* channel number */
-	u32 lba; /* Loop Begine Address, where dma buffer starts */
-	u32 eso; /* End Sample Offset, wehre dma buffer ends */
-	         /* (in the unit of samples) */
-	u32 delta; /* delta value, sample rate / 48k for playback, */
-	           /* 48k/sample rate for recording */
-	u16 attribute; /* control where PCM data go and come  */
-	u16 fm_vol;
-	u32 control; /* signed/unsigned, 8/16 bits, mono/stereo */
-};
-
-struct trident_pcm_bank_address {
-	u32 start;
-	u32 stop;
-	u32 aint;
-	u32 aint_en;
-};
-
-static struct trident_pcm_bank_address bank_a_addrs = {
-	T4D_START_A,
-	T4D_STOP_A,
-	T4D_AINT_A,
-	T4D_AINTEN_A
-};
-
-static struct trident_pcm_bank_address bank_b_addrs = {
-	T4D_START_B,
-	T4D_STOP_B,
-	T4D_AINT_B,
-	T4D_AINTEN_B
-};
-
-struct trident_pcm_bank {
-	/* register addresses to control bank operations */
-	struct trident_pcm_bank_address *addresses;
-	/* each bank has 32 channels */
-	u32 bitmap;		/* channel allocation bitmap */
-	struct trident_channel channels[32];
-};
-
-struct trident_card {
-	unsigned int magic;
-
-	/* We keep trident cards in a linked list */
-	struct trident_card *next;
-
-	/* single open lock mechanism, only used for recording */
-	struct mutex open_mutex;
-
-	/* The trident has a certain amount of cross channel interaction
-	   so we use a single per card lock */
-	spinlock_t lock;
-
-	/* PCI device stuff */
-	struct pci_dev *pci_dev;
-	u16 pci_id;
-	u8 revision;
-
-	/* soundcore stuff */
-	int dev_audio;
-
-	/* structures for abstraction of hardware facilities, codecs, */
-	/* banks and channels */
-	struct ac97_codec *ac97_codec[NR_AC97];
-	struct trident_pcm_bank banks[NR_BANKS];
-	struct trident_state *states[NR_HW_CH];
-
-	/* hardware resources */
-	unsigned long iobase;
-	u32 irq;
-
-	/* Function support */
-	struct trident_channel *(*alloc_pcm_channel) (struct trident_card *);
-	struct trident_channel *(*alloc_rec_pcm_channel) (struct trident_card *);
-	void (*free_pcm_channel) (struct trident_card *, unsigned int chan);
-	void (*address_interrupt) (struct trident_card *);
-
-	/* Added by Matt Wu 01-05-2001 for spdif in */
-	int multi_channel_use_count;
-	int rec_channel_use_count;
-	u16 mixer_regs[64][NR_AC97];	/* Made card local by Alan */
-	int mixer_regs_ready;
-
-	/* Added for hardware volume control */
-	int hwvolctl;
-	struct timer_list timer;
-
-	/* Game port support */
-	struct gameport *gameport;
-};
-
-enum dmabuf_mode {
-	DM_PLAYBACK = 0,
-	DM_RECORD
-};
-
-/* table to map from CHANNELMASK to channel attribute for SiS 7018 */
-static u16 mask2attr[] = {
-	PCM_LR, PCM_LR, SURR_LR, CENTER_LFE,
-	HSET, MIC, MODEM_LINE1, MODEM_LINE2,
-	I2S_LR, SPDIF_LR
-};
-
-/* table to map from channel attribute to CHANNELMASK for SiS 7018 */
-static int attr2mask[] = {
-	DSP_BIND_MODEM1, DSP_BIND_MODEM2, DSP_BIND_FRONT, DSP_BIND_HANDSET,
-	DSP_BIND_I2S, DSP_BIND_CENTER_LFE, DSP_BIND_SURR, DSP_BIND_SPDIF
-};
-
-/* Added by Matt Wu 01-05-2001 for spdif in */
-static int ali_close_multi_channels(void);
-static void ali_delay(struct trident_card *card, int interval);
-static void ali_detect_spdif_rate(struct trident_card *card);
-
-static void ali_ac97_write(struct ac97_codec *codec, u8 reg, u16 val);
-static u16 ali_ac97_read(struct ac97_codec *codec, u8 reg);
-
-static struct trident_card *devs;
-
-static void trident_ac97_set(struct ac97_codec *codec, u8 reg, u16 val);
-static u16 trident_ac97_get(struct ac97_codec *codec, u8 reg);
-
-static int trident_open_mixdev(struct inode *inode, struct file *file);
-static int trident_ioctl_mixdev(struct inode *inode, struct file *file,
-				unsigned int cmd, unsigned long arg);
-
-static void ali_ac97_set(struct trident_card *card, int secondary, u8 reg, u16 val);
-static u16 ali_ac97_get(struct trident_card *card, int secondary, u8 reg);
-static void ali_set_spdif_out_rate(struct trident_card *card, unsigned int rate);
-static void ali_enable_special_channel(struct trident_state *stat);
-static struct trident_channel *ali_alloc_rec_pcm_channel(struct trident_card *card);
-static struct trident_channel *ali_alloc_pcm_channel(struct trident_card *card);
-static void ali_free_pcm_channel(struct trident_card *card, unsigned int channel);
-static int ali_setup_multi_channels(struct trident_card *card, int chan_nums);
-static unsigned int ali_get_spdif_in_rate(struct trident_card *card);
-static void ali_setup_spdif_in(struct trident_card *card);
-static void ali_disable_spdif_in(struct trident_card *card);
-static void ali_disable_special_channel(struct trident_card *card, int ch);
-static void ali_setup_spdif_out(struct trident_card *card, int flag);
-static int ali_write_5_1(struct trident_state *state,
-			 const char __user *buffer,
-			 int cnt_for_multi_channel, unsigned int *copy_count,
-			 unsigned int *state_cnt);
-static int ali_allocate_other_states_resources(struct trident_state *state,
-					       int chan_nums);
-static void ali_free_other_states_resources(struct trident_state *state);
-
-#define seek_offset(dma_ptr, buffer, cnt, offset, copy_count)	do { \
-        (dma_ptr) += (offset);	  \
-	(buffer) += (offset);	  \
-        (cnt) -= (offset);	  \
-	(copy_count) += (offset); \
-} while (0)
-
-static inline int lock_set_fmt(struct trident_state* state)
-{
-	if (test_and_set_bit(0, &state->fmt_flag))
-		return -EFAULT;
-
-	return 0;
-}
-
-static inline void unlock_set_fmt(struct trident_state* state)
-{
-	clear_bit(0, &state->fmt_flag);
-}
-
-static int
-trident_enable_loop_interrupts(struct trident_card *card)
-{
-	u32 global_control;
-
-	global_control = inl(TRID_REG(card, T4D_LFO_GC_CIR));
-
-	switch (card->pci_id) {
-	case PCI_DEVICE_ID_SI_7018:
-		global_control |= (ENDLP_IE | MIDLP_IE | BANK_B_EN);
-		break;
-	case PCI_DEVICE_ID_ALI_5451:
-	case PCI_DEVICE_ID_TRIDENT_4DWAVE_DX:
-	case PCI_DEVICE_ID_TRIDENT_4DWAVE_NX:
-	case PCI_DEVICE_ID_INTERG_5050:
-		global_control |= (ENDLP_IE | MIDLP_IE);
-		break;
-	default:
-		return 0;
-	}
-
-	outl(global_control, TRID_REG(card, T4D_LFO_GC_CIR));
-
-	pr_debug("trident: Enable Loop Interrupts, globctl = 0x%08X\n",
-		 inl(TRID_REG(card, T4D_LFO_GC_CIR)));
-
-	return 1;
-}
-
-static int
-trident_disable_loop_interrupts(struct trident_card *card)
-{
-	u32 global_control;
-
-	global_control = inl(TRID_REG(card, T4D_LFO_GC_CIR));
-	global_control &= ~(ENDLP_IE | MIDLP_IE);
-	outl(global_control, TRID_REG(card, T4D_LFO_GC_CIR));
-
-	pr_debug("trident: Disabled Loop Interrupts, globctl = 0x%08X\n",
-		 global_control);
-
-	return 1;
-}
-
-static void
-trident_enable_voice_irq(struct trident_card *card, unsigned int channel)
-{
-	unsigned int mask = 1 << (channel & 0x1f);
-	struct trident_pcm_bank *bank = &card->banks[channel >> 5];
-	u32 reg, addr = bank->addresses->aint_en;
-
-	reg = inl(TRID_REG(card, addr));
-	reg |= mask;
-	outl(reg, TRID_REG(card, addr));
-
-#ifdef DEBUG
-	reg = inl(TRID_REG(card, addr));
-	pr_debug("trident: enabled IRQ on channel %d, %s = 0x%08x(addr:%X)\n",
-		 channel, addr == T4D_AINTEN_B ? "AINTEN_B" : "AINTEN_A",
-		 reg, addr);
-#endif /* DEBUG */
-}
-
-static void
-trident_disable_voice_irq(struct trident_card *card, unsigned int channel)
-{
-	unsigned int mask = 1 << (channel & 0x1f);
-	struct trident_pcm_bank *bank = &card->banks[channel >> 5];
-	u32 reg, addr = bank->addresses->aint_en;
-
-	reg = inl(TRID_REG(card, addr));
-	reg &= ~mask;
-	outl(reg, TRID_REG(card, addr));
-
-	/* Ack the channel in case the interrupt was set before we disable it. */
-	outl(mask, TRID_REG(card, bank->addresses->aint));
-
-#ifdef DEBUG
-	reg = inl(TRID_REG(card, addr));
-	pr_debug("trident: disabled IRQ on channel %d, %s = 0x%08x(addr:%X)\n",
-		 channel, addr == T4D_AINTEN_B ? "AINTEN_B" : "AINTEN_A",
-		 reg, addr);
-#endif /* DEBUG */
-}
-
-static void
-trident_start_voice(struct trident_card *card, unsigned int channel)
-{
-	unsigned int mask = 1 << (channel & 0x1f);
-	struct trident_pcm_bank *bank = &card->banks[channel >> 5];
-	u32 addr = bank->addresses->start;
-
-#ifdef DEBUG
-	u32 reg;
-#endif /* DEBUG */
-
-	outl(mask, TRID_REG(card, addr));
-
-#ifdef DEBUG
-	reg = inl(TRID_REG(card, addr));
-	pr_debug("trident: start voice on channel %d, %s = 0x%08x(addr:%X)\n",
-		 channel, addr == T4D_START_B ? "START_B" : "START_A",
-		 reg, addr);
-#endif /* DEBUG */
-}
-
-static void
-trident_stop_voice(struct trident_card *card, unsigned int channel)
-{
-	unsigned int mask = 1 << (channel & 0x1f);
-	struct trident_pcm_bank *bank = &card->banks[channel >> 5];
-	u32 addr = bank->addresses->stop;
-
-#ifdef DEBUG
-	u32 reg;
-#endif /* DEBUG */
-
-	outl(mask, TRID_REG(card, addr));
-
-#ifdef DEBUG
-	reg = inl(TRID_REG(card, addr));
-	pr_debug("trident: stop voice on channel %d, %s = 0x%08x(addr:%X)\n",
-		 channel, addr == T4D_STOP_B ? "STOP_B" : "STOP_A",
-		 reg, addr);
-#endif /* DEBUG */
-}
-
-static u32
-trident_get_interrupt_mask(struct trident_card *card, unsigned int channel)
-{
-	struct trident_pcm_bank *bank = &card->banks[channel];
-	u32 addr = bank->addresses->aint;
-	return inl(TRID_REG(card, addr));
-}
-
-static int
-trident_check_channel_interrupt(struct trident_card *card, unsigned int channel)
-{
-	unsigned int mask = 1 << (channel & 0x1f);
-	u32 reg = trident_get_interrupt_mask(card, channel >> 5);
-
-#ifdef DEBUG
-	if (reg & mask)
-		pr_debug("trident: channel %d has interrupt, %s = 0x%08x\n",
-			 channel, reg == T4D_AINT_B ? "AINT_B" : "AINT_A",
-			 reg);
-#endif /* DEBUG */
-	return (reg & mask) ? 1 : 0;
-}
-
-static void
-trident_ack_channel_interrupt(struct trident_card *card, unsigned int channel)
-{
-	unsigned int mask = 1 << (channel & 0x1f);
-	struct trident_pcm_bank *bank = &card->banks[channel >> 5];
-	u32 reg, addr = bank->addresses->aint;
-
-	reg = inl(TRID_REG(card, addr));
-	reg &= mask;
-	outl(reg, TRID_REG(card, addr));
-
-#ifdef DEBUG
-	reg = inl(TRID_REG(card, T4D_AINT_B));
-	pr_debug("trident: Ack channel %d interrupt, AINT_B = 0x%08x\n",
-		 channel, reg);
-#endif /* DEBUG */
-}
-
-static struct trident_channel *
-trident_alloc_pcm_channel(struct trident_card *card)
-{
-	struct trident_pcm_bank *bank;
-	int idx;
-
-	bank = &card->banks[BANK_B];
-
-	for (idx = 31; idx >= 0; idx--) {
-		if (!(bank->bitmap & (1 << idx))) {
-			struct trident_channel *channel = &bank->channels[idx];
-			bank->bitmap |= 1 << idx;
-			channel->num = idx + 32;
-			return channel;
-		}
-	}
-
-	/* no more free channels available */
-	printk(KERN_ERR "trident: no more channels available on Bank B.\n");
-	return NULL;
-}
-
-static void
-trident_free_pcm_channel(struct trident_card *card, unsigned int channel)
-{
-	int bank;
-	unsigned char b;
-
-	if (channel < 31 || channel > 63)
-		return;
-
-	if (card->pci_id == PCI_DEVICE_ID_TRIDENT_4DWAVE_DX ||
-	    card->pci_id == PCI_DEVICE_ID_TRIDENT_4DWAVE_NX) {
-		b = inb(TRID_REG(card, T4D_REC_CH));
-		if ((b & ~0x80) == channel)
-			outb(0x0, TRID_REG(card, T4D_REC_CH));
-	}
-
-	bank = channel >> 5;
-	channel = channel & 0x1f;
-
-	card->banks[bank].bitmap &= ~(1 << (channel));
-}
-
-static struct trident_channel *
-cyber_alloc_pcm_channel(struct trident_card *card)
-{
-	struct trident_pcm_bank *bank;
-	int idx;
-
-	/* The cyberpro 5050 has only 32 voices and one bank */
-	/* .. at least they are not documented (if you want to call that
-	 * crap documentation), perhaps broken ? */
-
-	bank = &card->banks[BANK_A];
-
-	for (idx = 31; idx >= 0; idx--) {
-		if (!(bank->bitmap & (1 << idx))) {
-			struct trident_channel *channel = &bank->channels[idx];
-			bank->bitmap |= 1 << idx;
-			channel->num = idx;
-			return channel;
-		}
-	}
-
-	/* no more free channels available */
-	printk(KERN_ERR "cyberpro5050: no more channels available on Bank A.\n");
-	return NULL;
-}
-
-static void
-cyber_free_pcm_channel(struct trident_card *card, unsigned int channel)
-{
-	if (channel > 31)
-		return;
-	card->banks[BANK_A].bitmap &= ~(1 << (channel));
-}
-
-static inline void
-cyber_outidx(int port, int idx, int data)
-{
-	outb(idx, port);
-	outb(data, port + 1);
-}
-
-static inline int
-cyber_inidx(int port, int idx)
-{
-	outb(idx, port);
-	return inb(port + 1);
-}
-
-static int
-cyber_init_ritual(struct trident_card *card)
-{
-	/* some black magic, taken from SDK samples */
-	/* remove this and nothing will work */
-	int portDat;
-	int ret = 0;
-	unsigned long flags;
-
-	/*
-	 *      Keep interrupts off for the configure - we don't want to
-	 *      clash with another cyberpro config event
-	 */
-
-	spin_lock_irqsave(&card->lock, flags);
-	portDat = cyber_inidx(CYBER_PORT_AUDIO, CYBER_IDX_AUDIO_ENABLE);
-	/* enable, if it was disabled */
-	if ((portDat & CYBER_BMSK_AUENZ) != CYBER_BMSK_AUENZ_ENABLE) {
-		printk(KERN_INFO "cyberpro5050: enabling audio controller\n");
-		cyber_outidx(CYBER_PORT_AUDIO, CYBER_IDX_AUDIO_ENABLE,
-			     portDat | CYBER_BMSK_AUENZ_ENABLE);
-		/* check again if hardware is enabled now */
-		portDat = cyber_inidx(CYBER_PORT_AUDIO, CYBER_IDX_AUDIO_ENABLE);
-	}
-	if ((portDat & CYBER_BMSK_AUENZ) != CYBER_BMSK_AUENZ_ENABLE) {
-		printk(KERN_ERR "cyberpro5050: initAudioAccess: no success\n");
-		ret = -1;
-	} else {
-		cyber_outidx(CYBER_PORT_AUDIO, CYBER_IDX_IRQ_ENABLE,
-			     CYBER_BMSK_AUDIO_INT_ENABLE);
-		cyber_outidx(CYBER_PORT_AUDIO, 0xbf, 0x01);
-		cyber_outidx(CYBER_PORT_AUDIO, 0xba, 0x20);
-		cyber_outidx(CYBER_PORT_AUDIO, 0xbb, 0x08);
-		cyber_outidx(CYBER_PORT_AUDIO, 0xbf, 0x02);
-		cyber_outidx(CYBER_PORT_AUDIO, 0xb3, 0x06);
-		cyber_outidx(CYBER_PORT_AUDIO, 0xbf, 0x00);
-	}
-	spin_unlock_irqrestore(&card->lock, flags);
-	return ret;
-}
-
-/*  called with spin lock held */
-
-static int
-trident_load_channel_registers(struct trident_card *card, u32 * data,
-			       unsigned int channel)
-{
-	int i;
-
-	if (channel > 63)
-		return 0;
-
-	/* select hardware channel to write */
-	outb(channel, TRID_REG(card, T4D_LFO_GC_CIR));
-
-	/* Output the channel registers, but don't write register
-	   three to an ALI chip. */
-	for (i = 0; i < CHANNEL_REGS; i++) {
-		if (i == 3 && card->pci_id == PCI_DEVICE_ID_ALI_5451)
-			continue;
-		outl(data[i], TRID_REG(card, CHANNEL_START + 4 * i));
-	}
-	if (card->pci_id == PCI_DEVICE_ID_ALI_5451 ||
-	    card->pci_id == PCI_DEVICE_ID_INTERG_5050) {
-		outl(ALI_EMOD_Still, TRID_REG(card, ALI_EBUF1));
-		outl(ALI_EMOD_Still, TRID_REG(card, ALI_EBUF2));
-	}
-	return 1;
-}
-
-/* called with spin lock held */
-static int
-trident_write_voice_regs(struct trident_state *state)
-{
-	unsigned int data[CHANNEL_REGS + 1];
-	struct trident_channel *channel;
-
-	channel = state->dmabuf.channel;
-
-	data[1] = channel->lba;
-	data[4] = channel->control;
-
-	switch (state->card->pci_id) {
-	case PCI_DEVICE_ID_ALI_5451:
-		data[0] = 0;	/* Current Sample Offset */
-		data[2] = (channel->eso << 16) | (channel->delta & 0xffff);
-		data[3] = 0;
-		break;
-	case PCI_DEVICE_ID_SI_7018:
-	case PCI_DEVICE_ID_INTERG_5050:
-		data[0] = 0;	/* Current Sample Offset */
-		data[2] = (channel->eso << 16) | (channel->delta & 0xffff);
-		data[3] = (channel->attribute << 16) | (channel->fm_vol & 0xffff);
-		break;
-	case PCI_DEVICE_ID_TRIDENT_4DWAVE_DX:
-		data[0] = 0;	/* Current Sample Offset */
-		data[2] = (channel->eso << 16) | (channel->delta & 0xffff);
-		data[3] = channel->fm_vol & 0xffff;
-		break;
-	case PCI_DEVICE_ID_TRIDENT_4DWAVE_NX:
-		data[0] = (channel->delta << 24);
-		data[2] = ((channel->delta << 16) & 0xff000000) |
-			(channel->eso & 0x00ffffff);
-		data[3] = channel->fm_vol & 0xffff;
-		break;
-	default:
-		return 0;
-	}
-
-	return trident_load_channel_registers(state->card, data, channel->num);
-}
-
-static int
-compute_rate_play(u32 rate)
-{
-	int delta;
-	/* We special case 44100 and 8000 since rounding with the equation
-	   does not give us an accurate enough value. For 11025 and 22050
-	   the equation gives us the best answer. All other frequencies will
-	   also use the equation. JDW */
-	if (rate == 44100)
-		delta = 0xeb3;
-	else if (rate == 8000)
-		delta = 0x2ab;
-	else if (rate == 48000)
-		delta = 0x1000;
-	else
-		delta = (((rate << 12) + rate) / 48000) & 0x0000ffff;
-	return delta;
-}
-
-static int
-compute_rate_rec(u32 rate)
-{
-	int delta;
-
-	if (rate == 44100)
-		delta = 0x116a;
-	else if (rate == 8000)
-		delta = 0x6000;
-	else if (rate == 48000)
-		delta = 0x1000;
-	else
-		delta = ((48000 << 12) / rate) & 0x0000ffff;
-
-	return delta;
-}
-
-/* set playback sample rate */
-static unsigned int
-trident_set_dac_rate(struct trident_state *state, unsigned int rate)
-{
-	struct dmabuf *dmabuf = &state->dmabuf;
-
-	if (rate > 48000)
-		rate = 48000;
-	if (rate < 4000)
-		rate = 4000;
-
-	dmabuf->rate = rate;
-	dmabuf->channel->delta = compute_rate_play(rate);
-
-	trident_write_voice_regs(state);
-
-	pr_debug("trident: called trident_set_dac_rate : rate = %d\n", rate);
-
-	return rate;
-}
-
-/* set recording sample rate */
-static unsigned int
-trident_set_adc_rate(struct trident_state *state, unsigned int rate)
-{
-	struct dmabuf *dmabuf = &state->dmabuf;
-
-	if (rate > 48000)
-		rate = 48000;
-	if (rate < 4000)
-		rate = 4000;
-
-	dmabuf->rate = rate;
-	dmabuf->channel->delta = compute_rate_rec(rate);
-
-	trident_write_voice_regs(state);
-
-	pr_debug("trident: called trident_set_adc_rate : rate = %d\n", rate);
-
-	return rate;
-}
-
-/* prepare channel attributes for playback */
-static void
-trident_play_setup(struct trident_state *state)
-{
-	struct dmabuf *dmabuf = &state->dmabuf;
-	struct trident_channel *channel = dmabuf->channel;
-
-	channel->lba = dmabuf->dma_handle;
-	channel->delta = compute_rate_play(dmabuf->rate);
-
-	channel->eso = dmabuf->dmasize >> sample_shift[dmabuf->fmt];
-	channel->eso -= 1;
-
-	if (state->card->pci_id != PCI_DEVICE_ID_SI_7018) {
-		channel->attribute = 0;
-		if (state->card->pci_id == PCI_DEVICE_ID_ALI_5451) {
-			if ((channel->num == ALI_SPDIF_IN_CHANNEL) ||
-			    (channel->num == ALI_PCM_IN_CHANNEL))
-				ali_disable_special_channel(state->card, channel->num);
-			else if ((inl(TRID_REG(state->card, ALI_GLOBAL_CONTROL))
-				  & ALI_SPDIF_OUT_CH_ENABLE)
-				 && (channel->num == ALI_SPDIF_OUT_CHANNEL)) {
-				ali_set_spdif_out_rate(state->card,
-						       state->dmabuf.rate);
-				state->dmabuf.channel->delta = 0x1000;
-			}
-		}
-	}
-
-	channel->fm_vol = 0x0;
-
-	channel->control = CHANNEL_LOOP;
-	if (dmabuf->fmt & TRIDENT_FMT_16BIT) {
-		/* 16-bits */
-		channel->control |= CHANNEL_16BITS;
-		/* signed */
-		channel->control |= CHANNEL_SIGNED;
-	}
-	if (dmabuf->fmt & TRIDENT_FMT_STEREO)
-		/* stereo */
-		channel->control |= CHANNEL_STEREO;
-
-	pr_debug("trident: trident_play_setup, LBA = 0x%08x, Delta = 0x%08x, "
-		 "ESO = 0x%08x, Control = 0x%08x\n", channel->lba,
-		 channel->delta, channel->eso, channel->control);
-
-	trident_write_voice_regs(state);
-}
-
-/* prepare channel attributes for recording */
-static void
-trident_rec_setup(struct trident_state *state)
-{
-	u16 w;
-	u8 bval;
-
-	struct trident_card *card = state->card;
-	struct dmabuf *dmabuf = &state->dmabuf;
-	struct trident_channel *channel = dmabuf->channel;
-	unsigned int rate;
-
-	/* Enable AC-97 ADC (capture) */
-	switch (card->pci_id) {
-	case PCI_DEVICE_ID_ALI_5451:
-		ali_enable_special_channel(state);
-		break;
-	case PCI_DEVICE_ID_SI_7018:
-		/* for 7018, the ac97 is always in playback/record (duplex) mode */
-		break;
-	case PCI_DEVICE_ID_TRIDENT_4DWAVE_DX:
-		w = inb(TRID_REG(card, DX_ACR2_AC97_COM_STAT));
-		outb(w | 0x48, TRID_REG(card, DX_ACR2_AC97_COM_STAT));
-		/* enable and set record channel */
-		outb(0x80 | channel->num, TRID_REG(card, T4D_REC_CH));
-		break;
-	case PCI_DEVICE_ID_TRIDENT_4DWAVE_NX:
-		w = inw(TRID_REG(card, T4D_MISCINT));
-		outw(w | 0x1000, TRID_REG(card, T4D_MISCINT));
-		/* enable and set record channel */
-		outb(0x80 | channel->num, TRID_REG(card, T4D_REC_CH));
-		break;
-	case PCI_DEVICE_ID_INTERG_5050:
-		/* don't know yet, using special channel 22 in GC1(0xd4)? */
-		break;
-	default:
-		return;
-	}
-
-	channel->lba = dmabuf->dma_handle;
-	channel->delta = compute_rate_rec(dmabuf->rate);
-	if ((card->pci_id == PCI_DEVICE_ID_ALI_5451) &&
-	    (channel->num == ALI_SPDIF_IN_CHANNEL)) {
-		rate = ali_get_spdif_in_rate(card);
-		if (rate == 0) {
-			printk(KERN_WARNING "trident: ALi 5451 "
-			       "S/PDIF input setup error!\n");
-			rate = 48000;
-		}
-		bval = inb(TRID_REG(card, ALI_SPDIF_CTRL));
-		if (bval & 0x10) {
-			outb(bval, TRID_REG(card, ALI_SPDIF_CTRL));
-			printk(KERN_WARNING "trident: cleared ALi "
-			       "5451 S/PDIF parity error flag.\n");
-		}
-
-		if (rate != 48000)
-			channel->delta = ((rate << 12) / dmabuf->rate) & 0x0000ffff;
-	}
-
-	channel->eso = dmabuf->dmasize >> sample_shift[dmabuf->fmt];
-	channel->eso -= 1;
-
-	if (state->card->pci_id != PCI_DEVICE_ID_SI_7018) {
-		channel->attribute = 0;
-	}
-
-	channel->fm_vol = 0x0;
-
-	channel->control = CHANNEL_LOOP;
-	if (dmabuf->fmt & TRIDENT_FMT_16BIT) {
-		/* 16-bits */
-		channel->control |= CHANNEL_16BITS;
-		/* signed */
-		channel->control |= CHANNEL_SIGNED;
-	}
-	if (dmabuf->fmt & TRIDENT_FMT_STEREO)
-		/* stereo */
-		channel->control |= CHANNEL_STEREO;
-
-	pr_debug("trident: trident_rec_setup, LBA = 0x%08x, Delat = 0x%08x, "
-		 "ESO = 0x%08x, Control = 0x%08x\n", channel->lba,
-		 channel->delta, channel->eso, channel->control);
-
-	trident_write_voice_regs(state);
-}
-
-/* get current playback/recording dma buffer pointer (byte offset from LBA),
-   called with spinlock held! */
-static inline unsigned
-trident_get_dma_addr(struct trident_state *state)
-{
-	struct dmabuf *dmabuf = &state->dmabuf;
-	u32 cso;
-
-	if (!dmabuf->enable)
-		return 0;
-
-	outb(dmabuf->channel->num, TRID_REG(state->card, T4D_LFO_GC_CIR));
-
-	switch (state->card->pci_id) {
-	case PCI_DEVICE_ID_ALI_5451:
-	case PCI_DEVICE_ID_SI_7018:
-	case PCI_DEVICE_ID_TRIDENT_4DWAVE_DX:
-	case PCI_DEVICE_ID_INTERG_5050:
-		/* 16 bits ESO, CSO for 7018 and DX */
-		cso = inw(TRID_REG(state->card, CH_DX_CSO_ALPHA_FMS + 2));
-		break;
-	case PCI_DEVICE_ID_TRIDENT_4DWAVE_NX:
-		/* 24 bits ESO, CSO for NX */
-		cso = inl(TRID_REG(state->card, CH_NX_DELTA_CSO)) & 0x00ffffff;
-		break;
-	default:
-		return 0;
-	}
-
-	pr_debug("trident: trident_get_dma_addr: chip reported channel: %d, "
-		 "cso = 0x%04x\n", dmabuf->channel->num, cso);
-
-	/* ESO and CSO are in units of Samples, convert to byte offset */
-	cso <<= sample_shift[dmabuf->fmt];
-
-	return (cso % dmabuf->dmasize);
-}
-
-/* Stop recording (lock held) */
-static inline void
-__stop_adc(struct trident_state *state)
-{
-	struct dmabuf *dmabuf = &state->dmabuf;
-	unsigned int chan_num = dmabuf->channel->num;
-	struct trident_card *card = state->card;
-
-	dmabuf->enable &= ~ADC_RUNNING;
-	trident_stop_voice(card, chan_num);
-	trident_disable_voice_irq(card, chan_num);
-}
-
-static void
-stop_adc(struct trident_state *state)
-{
-	struct trident_card *card = state->card;
-	unsigned long flags;
-
-	spin_lock_irqsave(&card->lock, flags);
-	__stop_adc(state);
-	spin_unlock_irqrestore(&card->lock, flags);
-}
-
-static void
-start_adc(struct trident_state *state)
-{
-	struct dmabuf *dmabuf = &state->dmabuf;
-	unsigned int chan_num = dmabuf->channel->num;
-	struct trident_card *card = state->card;
-	unsigned long flags;
-
-	spin_lock_irqsave(&card->lock, flags);
-	if ((dmabuf->mapped ||
-	     dmabuf->count < (signed) dmabuf->dmasize) &&
-	    dmabuf->ready) {
-		dmabuf->enable |= ADC_RUNNING;
-		trident_enable_voice_irq(card, chan_num);
-		trident_start_voice(card, chan_num);
-	}
-	spin_unlock_irqrestore(&card->lock, flags);
-}
-
-/* stop playback (lock held) */
-static inline void
-__stop_dac(struct trident_state *state)
-{
-	struct dmabuf *dmabuf = &state->dmabuf;
-	unsigned int chan_num = dmabuf->channel->num;
-	struct trident_card *card = state->card;
-
-	dmabuf->enable &= ~DAC_RUNNING;
-	trident_stop_voice(card, chan_num);
-	if (state->chans_num == 6) {
-		trident_stop_voice(card, state->other_states[0]->
-				   dmabuf.channel->num);
-		trident_stop_voice(card, state->other_states[1]->
-				   dmabuf.channel->num);
-		trident_stop_voice(card, state->other_states[2]->
-				   dmabuf.channel->num);
-		trident_stop_voice(card, state->other_states[3]->
-				   dmabuf.channel->num);
-	}
-	trident_disable_voice_irq(card, chan_num);
-}
-
-static void
-stop_dac(struct trident_state *state)
-{
-	struct trident_card *card = state->card;
-	unsigned long flags;
-
-	spin_lock_irqsave(&card->lock, flags);
-	__stop_dac(state);
-	spin_unlock_irqrestore(&card->lock, flags);
-}
-
-static void
-start_dac(struct trident_state *state)
-{
-	struct dmabuf *dmabuf = &state->dmabuf;
-	unsigned int chan_num = dmabuf->channel->num;
-	struct trident_card *card = state->card;
-	unsigned long flags;
-
-	spin_lock_irqsave(&card->lock, flags);
-	if ((dmabuf->mapped || dmabuf->count > 0) && dmabuf->ready) {
-		dmabuf->enable |= DAC_RUNNING;
-		trident_enable_voice_irq(card, chan_num);
-		trident_start_voice(card, chan_num);
-		if (state->chans_num == 6) {
-			trident_start_voice(card, state->other_states[0]->
-					    dmabuf.channel->num);
-			trident_start_voice(card, state->other_states[1]->
-					    dmabuf.channel->num);
-			trident_start_voice(card, state->other_states[2]->
-					    dmabuf.channel->num);
-			trident_start_voice(card, state->other_states[3]->
-					    dmabuf.channel->num);
-		}
-	}
-	spin_unlock_irqrestore(&card->lock, flags);
-}
-
-#define DMABUF_DEFAULTORDER (15-PAGE_SHIFT)
-#define DMABUF_MINORDER 1
-
-/* alloc a DMA buffer of with a buffer of this order */
-static int
-alloc_dmabuf(struct dmabuf *dmabuf, struct pci_dev *pci_dev, int order)
-{
-	void *rawbuf = NULL;
-	struct page *page, *pend;
-
-	if (!(rawbuf = pci_alloc_consistent(pci_dev, PAGE_SIZE << order,
-					    &dmabuf->dma_handle)))
-		return -ENOMEM;
-
-	pr_debug("trident: allocated %ld (order = %d) bytes at %p\n",
-		 PAGE_SIZE << order, order, rawbuf);
-
-	dmabuf->ready = dmabuf->mapped = 0;
-	dmabuf->rawbuf = rawbuf;
-	dmabuf->buforder = order;
-
-	/* now mark the pages as reserved; otherwise */
-	/* remap_pfn_range doesn't do what we want */
-	pend = virt_to_page(rawbuf + (PAGE_SIZE << order) - 1);
-	for (page = virt_to_page(rawbuf); page <= pend; page++)
-		SetPageReserved(page);
-
-	return 0;
-}
-
-/* allocate the main DMA buffer, playback and recording buffer should be */
-/* allocated separately */
-static int
-alloc_main_dmabuf(struct trident_state *state)
-{
-	struct dmabuf *dmabuf = &state->dmabuf;
-	int order;
-	int ret = -ENOMEM;
-
-	/* alloc as big a chunk as we can, FIXME: is this necessary ?? */
-	for (order = DMABUF_DEFAULTORDER; order >= DMABUF_MINORDER; order--) {
-		if (!(ret = alloc_dmabuf(dmabuf, state->card->pci_dev, order)))
-			return 0;
-		/* else try again */
-	}
-	return ret;
-}
-
-/* deallocate a DMA buffer */
-static void
-dealloc_dmabuf(struct dmabuf *dmabuf, struct pci_dev *pci_dev)
-{
-	struct page *page, *pend;
-
-	if (dmabuf->rawbuf) {
-		/* undo marking the pages as reserved */
-		pend = virt_to_page(dmabuf->rawbuf + (PAGE_SIZE << dmabuf->buforder) - 1);
-		for (page = virt_to_page(dmabuf->rawbuf); page <= pend; page++)
-			ClearPageReserved(page);
-		pci_free_consistent(pci_dev, PAGE_SIZE << dmabuf->buforder,
-				    dmabuf->rawbuf, dmabuf->dma_handle);
-		dmabuf->rawbuf = NULL;
-	}
-	dmabuf->mapped = dmabuf->ready = 0;
-}
-
-static int
-prog_dmabuf(struct trident_state *state, enum dmabuf_mode rec)
-{
-	struct dmabuf *dmabuf = &state->dmabuf;
-	unsigned bytepersec;
-	struct trident_state *s = state;
-	unsigned bufsize, dma_nums;
-	unsigned long flags;
-	int ret, i, order;
-
-	if ((ret = lock_set_fmt(state)) < 0)
-		return ret;
-
-	if (state->chans_num == 6)
-		dma_nums = 5;
-	else
-		dma_nums = 1;
-
-	for (i = 0; i < dma_nums; i++) {
-		if (i > 0) {
-			s = state->other_states[i - 1];
-			dmabuf = &s->dmabuf;
-			dmabuf->fmt = state->dmabuf.fmt;
-			dmabuf->rate = state->dmabuf.rate;
-		}
-
-		spin_lock_irqsave(&s->card->lock, flags);
-		dmabuf->hwptr = dmabuf->swptr = dmabuf->total_bytes = 0;
-		dmabuf->count = dmabuf->error = 0;
-		spin_unlock_irqrestore(&s->card->lock, flags);
-
-		/* allocate DMA buffer if not allocated yet */
-		if (!dmabuf->rawbuf) {
-			if (i == 0) {
-				if ((ret = alloc_main_dmabuf(state))) {
-					unlock_set_fmt(state);
-					return ret;
-				}
-			} else {
-				ret = -ENOMEM;
-				order = state->dmabuf.buforder - 1;
-				if (order >= DMABUF_MINORDER) {
-					ret = alloc_dmabuf(dmabuf,
-							   state->card->pci_dev,
-							   order);
-				}
-				if (ret) {
-					/* release the main DMA buffer */
-					dealloc_dmabuf(&state->dmabuf, state->card->pci_dev);
-					/* release the auxiliary DMA buffers */
-					for (i -= 2; i >= 0; i--)
-						dealloc_dmabuf(&state->other_states[i]->dmabuf,
-							       state->card->pci_dev);
-					unlock_set_fmt(state);
-					return ret;
-				}
-			}
-		}
-		/* FIXME: figure out all this OSS fragment stuff */
-		bytepersec = dmabuf->rate << sample_shift[dmabuf->fmt];
-		bufsize = PAGE_SIZE << dmabuf->buforder;
-		if (dmabuf->ossfragshift) {
-			if ((1000 << dmabuf->ossfragshift) < bytepersec)
-				dmabuf->fragshift = ld2(bytepersec / 1000);
-			else
-				dmabuf->fragshift = dmabuf->ossfragshift;
-		} else {
-			/* lets hand out reasonable big ass buffers by default */
-			dmabuf->fragshift = (dmabuf->buforder + PAGE_SHIFT - 2);
-		}
-		dmabuf->numfrag = bufsize >> dmabuf->fragshift;
-		while (dmabuf->numfrag < 4 && dmabuf->fragshift > 3) {
-			dmabuf->fragshift--;
-			dmabuf->numfrag = bufsize >> dmabuf->fragshift;
-		}
-		dmabuf->fragsize = 1 << dmabuf->fragshift;
-		if (dmabuf->ossmaxfrags >= 4 && dmabuf->ossmaxfrags < dmabuf->numfrag)
-			dmabuf->numfrag = dmabuf->ossmaxfrags;
-		dmabuf->fragsamples = dmabuf->fragsize >> sample_shift[dmabuf->fmt];
-		dmabuf->dmasize = dmabuf->numfrag << dmabuf->fragshift;
-
-		memset(dmabuf->rawbuf, (dmabuf->fmt & TRIDENT_FMT_16BIT) ? 0 : 0x80,
-		       dmabuf->dmasize);
-
-		spin_lock_irqsave(&s->card->lock, flags);
-		if (rec == DM_RECORD)
-			trident_rec_setup(s);
-		else /* DM_PLAYBACK */
-			trident_play_setup(s);
-
-		spin_unlock_irqrestore(&s->card->lock, flags);
-
-		/* set the ready flag for the dma buffer */
-		dmabuf->ready = 1;
-
-		pr_debug("trident: prog_dmabuf(%d), sample rate = %d, "
-			 "format = %d, numfrag = %d, fragsize = %d "
-			 "dmasize = %d\n", dmabuf->channel->num,
-			 dmabuf->rate, dmabuf->fmt, dmabuf->numfrag,
-			 dmabuf->fragsize, dmabuf->dmasize);
-	}
-	unlock_set_fmt(state);
-	return 0;
-}
-
-
-static inline int prog_dmabuf_record(struct trident_state* state)
-{
-	return prog_dmabuf(state, DM_RECORD);
-}
-
-static inline int prog_dmabuf_playback(struct trident_state* state)
-{
-	return prog_dmabuf(state, DM_PLAYBACK);
-}
-
-/* we are doing quantum mechanics here, the buffer can only be empty, half or full filled i.e.
-   |------------|------------|   or   |xxxxxxxxxxxx|------------|   or   |xxxxxxxxxxxx|xxxxxxxxxxxx|
-   but we almost always get this
-   |xxxxxx------|------------|   or   |xxxxxxxxxxxx|xxxxx-------|
-   so we have to clear the tail space to "silence"
-   |xxxxxx000000|------------|   or   |xxxxxxxxxxxx|xxxxxx000000|
-*/
-static void
-trident_clear_tail(struct trident_state *state)
-{
-	struct dmabuf *dmabuf = &state->dmabuf;
-	unsigned swptr;
-	unsigned char silence = (dmabuf->fmt & TRIDENT_FMT_16BIT) ? 0 : 0x80;
-	unsigned int len;
-	unsigned long flags;
-
-	spin_lock_irqsave(&state->card->lock, flags);
-	swptr = dmabuf->swptr;
-	spin_unlock_irqrestore(&state->card->lock, flags);
-
-	if (swptr == 0 || swptr == dmabuf->dmasize / 2 ||
-	    swptr == dmabuf->dmasize)
-		return;
-
-	if (swptr < dmabuf->dmasize / 2)
-		len = dmabuf->dmasize / 2 - swptr;
-	else
-		len = dmabuf->dmasize - swptr;
-
-	memset(dmabuf->rawbuf + swptr, silence, len);
-	if (state->card->pci_id != PCI_DEVICE_ID_ALI_5451) {
-		spin_lock_irqsave(&state->card->lock, flags);
-		dmabuf->swptr += len;
-		dmabuf->count += len;
-		spin_unlock_irqrestore(&state->card->lock, flags);
-	}
-
-	/* restart the dma machine in case it is halted */
-	start_dac(state);
-}
-
-static int
-drain_dac(struct trident_state *state, int nonblock)
-{
-	DECLARE_WAITQUEUE(wait, current);
-	struct dmabuf *dmabuf = &state->dmabuf;
-	unsigned long flags;
-	unsigned long tmo;
-	int count;
-	unsigned long diff = 0;
-
-	if (dmabuf->mapped || !dmabuf->ready)
-		return 0;
-
-	add_wait_queue(&dmabuf->wait, &wait);
-	for (;;) {
-		/* It seems that we have to set the current state to TASK_INTERRUPTIBLE
-		   every time to make the process really go to sleep */
-		set_current_state(TASK_INTERRUPTIBLE);
-
-		spin_lock_irqsave(&state->card->lock, flags);
-		count = dmabuf->count;
-		spin_unlock_irqrestore(&state->card->lock, flags);
-
-		if (count <= 0)
-			break;
-
-		if (signal_pending(current))
-			break;
-
-		if (nonblock) {
-			remove_wait_queue(&dmabuf->wait, &wait);
-			set_current_state(TASK_RUNNING);
-			return -EBUSY;
-		}
-
-		/* No matter how much data is left in the buffer, we have to wait until
-		   CSO == ESO/2 or CSO == ESO when address engine interrupts */
-		if (state->card->pci_id == PCI_DEVICE_ID_ALI_5451 ||
-		    state->card->pci_id == PCI_DEVICE_ID_INTERG_5050) {
-			diff = dmabuf->swptr - trident_get_dma_addr(state) + dmabuf->dmasize;
-			diff = diff % (dmabuf->dmasize);
-			tmo = (diff * HZ) / dmabuf->rate;
-		} else {
-			tmo = (dmabuf->dmasize * HZ) / dmabuf->rate;
-		}
-		tmo >>= sample_shift[dmabuf->fmt];
-		if (!schedule_timeout(tmo ? tmo : 1) && tmo) {
-			break;
-		}
-	}
-	remove_wait_queue(&dmabuf->wait, &wait);
-	set_current_state(TASK_RUNNING);
-	if (signal_pending(current))
-		return -ERESTARTSYS;
-
-	return 0;
-}
-
-/* update buffer manangement pointers, especially, */
-/* dmabuf->count and dmabuf->hwptr */
-static void
-trident_update_ptr(struct trident_state *state)
-{
-	struct dmabuf *dmabuf = &state->dmabuf;
-	unsigned hwptr, swptr;
-	int clear_cnt = 0;
-	int diff;
-	unsigned char silence;
-	unsigned half_dmasize;
-
-	/* update hardware pointer */
-	hwptr = trident_get_dma_addr(state);
-	diff = (dmabuf->dmasize + hwptr - dmabuf->hwptr) % dmabuf->dmasize;
-	dmabuf->hwptr = hwptr;
-	dmabuf->total_bytes += diff;
-
-	/* error handling and process wake up for ADC */
-	if (dmabuf->enable == ADC_RUNNING) {
-		if (dmabuf->mapped) {
-			dmabuf->count -= diff;
-			if (dmabuf->count >= (signed) dmabuf->fragsize)
-				wake_up(&dmabuf->wait);
-		} else {
-			dmabuf->count += diff;
-
-			if (dmabuf->count < 0 ||
-			    dmabuf->count > dmabuf->dmasize) {
-				/* buffer underrun or buffer overrun, */
-				/* we have no way to recover it here, just */
-				/* stop the machine and let the process */
-				/* force hwptr and swptr to sync */
-				__stop_adc(state);
-				dmabuf->error++;
-			}
-			if (dmabuf->count < (signed) dmabuf->dmasize / 2)
-				wake_up(&dmabuf->wait);
-		}
-	}
-
-	/* error handling and process wake up for DAC */
-	if (dmabuf->enable == DAC_RUNNING) {
-		if (dmabuf->mapped) {
-			dmabuf->count += diff;
-			if (dmabuf->count >= (signed) dmabuf->fragsize)
-				wake_up(&dmabuf->wait);
-		} else {
-			dmabuf->count -= diff;
-
-			if (dmabuf->count < 0 ||
-			    dmabuf->count > dmabuf->dmasize) {
-				/* buffer underrun or buffer overrun, we have no way to recover
-				   it here, just stop the machine and let the process force hwptr
-				   and swptr to sync */
-				__stop_dac(state);
-				dmabuf->error++;
-			} else if (!dmabuf->endcleared) {
-				swptr = dmabuf->swptr;
-				silence = (dmabuf->fmt & TRIDENT_FMT_16BIT ? 0 : 0x80);
-				if (dmabuf->update_flag & ALI_ADDRESS_INT_UPDATE) {
-					/* We must clear end data of 1/2 dmabuf if needed.
-					   According to 1/2 algorithm of Address Engine Interrupt,
-					   check the validation of the data of half dmasize. */
-					half_dmasize = dmabuf->dmasize / 2;
-					if ((diff = hwptr - half_dmasize) < 0)
-						diff = hwptr;
-					if ((dmabuf->count + diff) < half_dmasize) {
-						//there is invalid data in the end of half buffer
-						if ((clear_cnt = half_dmasize - swptr) < 0)
-							clear_cnt += half_dmasize;
-						//clear the invalid data
-						memset(dmabuf->rawbuf + swptr, silence, clear_cnt);
-						if (state->chans_num == 6) {
-							clear_cnt = clear_cnt / 2;
-							swptr = swptr / 2;
-							memset(state->other_states[0]->dmabuf.rawbuf + swptr,
-							       silence, clear_cnt);
-							memset(state->other_states[1]->dmabuf.rawbuf + swptr,
-							       silence, clear_cnt);
-							memset(state->other_states[2]->dmabuf.rawbuf + swptr,
-							       silence, clear_cnt);
-							memset(state->other_states[3]->dmabuf.rawbuf + swptr,
-							       silence, clear_cnt);
-						}
-						dmabuf->endcleared = 1;
-					}
-				} else if (dmabuf->count < (signed) dmabuf->fragsize) {
-					clear_cnt = dmabuf->fragsize;
-					if ((swptr + clear_cnt) > dmabuf->dmasize)
-						clear_cnt = dmabuf->dmasize - swptr;
-					memset(dmabuf->rawbuf + swptr, silence, clear_cnt);
-					if (state->chans_num == 6) {
-						clear_cnt = clear_cnt / 2;
-						swptr = swptr / 2;
-						memset(state->other_states[0]->dmabuf.rawbuf + swptr,
-						       silence, clear_cnt);
-						memset(state->other_states[1]->dmabuf.rawbuf + swptr,
-						       silence, clear_cnt);
-						memset(state->other_states[2]->dmabuf.rawbuf + swptr,
-						       silence, clear_cnt);
-						memset(state->other_states[3]->dmabuf.rawbuf + swptr,
-						       silence, clear_cnt);
-					}
-					dmabuf->endcleared = 1;
-				}
-			}
-			/* trident_update_ptr is called by interrupt handler or by process via
-			   ioctl/poll, we only wake up the waiting process when we have more
-			   than 1/2 buffer free (always true for interrupt handler) */
-			if (dmabuf->count < (signed) dmabuf->dmasize / 2)
-				wake_up(&dmabuf->wait);
-		}
-	}
-	dmabuf->update_flag &= ~ALI_ADDRESS_INT_UPDATE;
-}
-
-static void
-trident_address_interrupt(struct trident_card *card)
-{
-	int i;
-	struct trident_state *state;
-	unsigned int channel;
-
-	/* Update the pointers for all channels we are running. */
-	/* FIXME: should read interrupt status only once */
-	for (i = 0; i < NR_HW_CH; i++) {
-		channel = 63 - i;
-		if (trident_check_channel_interrupt(card, channel)) {
-			trident_ack_channel_interrupt(card, channel);
-			if ((state = card->states[i]) != NULL) {
-				trident_update_ptr(state);
-			} else {
-				printk(KERN_WARNING "trident: spurious channel "
-				       "irq %d.\n", channel);
-				trident_stop_voice(card, channel);
-				trident_disable_voice_irq(card, channel);
-			}
-		}
-	}
-}
-
-static void
-ali_hwvol_control(struct trident_card *card, int opt)
-{
-	u16 dwTemp, volume[2], mute, diff, *pVol[2];
-
-	dwTemp = ali_ac97_read(card->ac97_codec[0], 0x02);
-	mute = dwTemp & 0x8000;
-	volume[0] = dwTemp & 0x001f;
-	volume[1] = (dwTemp & 0x1f00) >> 8;
-	if (volume[0] < volume[1]) {
-		pVol[0] = &volume[0];
-		pVol[1] = &volume[1];
-	} else {
-		pVol[1] = &volume[0];
-		pVol[0] = &volume[1];
-	}
-	diff = *(pVol[1]) - *(pVol[0]);
-
-	if (opt == 1) {		// MUTE
-		dwTemp ^= 0x8000;
-		ali_ac97_write(card->ac97_codec[0],
-			       0x02, dwTemp);
-	} else if (opt == 2) {	// Down
-		if (mute)
-			return;
-		if (*(pVol[1]) < 0x001f) {
-			(*pVol[1])++;
-			*(pVol[0]) = *(pVol[1]) - diff;
-		}
-		dwTemp &= 0xe0e0;
-		dwTemp |= (volume[0]) | (volume[1] << 8);
-		ali_ac97_write(card->ac97_codec[0], 0x02, dwTemp);
-		card->ac97_codec[0]->mixer_state[0] = ((32 - volume[0]) * 25 / 8) |
-			(((32 - volume[1]) * 25 / 8) << 8);
-	} else if (opt == 4) {	// Up
-		if (mute)
-			return;
-		if (*(pVol[0]) > 0) {
-			(*pVol[0])--;
-			*(pVol[1]) = *(pVol[0]) + diff;
-		}
-		dwTemp &= 0xe0e0;
-		dwTemp |= (volume[0]) | (volume[1] << 8);
-		ali_ac97_write(card->ac97_codec[0], 0x02, dwTemp);
-		card->ac97_codec[0]->mixer_state[0] = ((32 - volume[0]) * 25 / 8) |
-			(((32 - volume[1]) * 25 / 8) << 8);
-	} else {
-		/* Nothing needs doing */
-	}
-}
-
-/*
- *	Re-enable reporting of vol change after 0.1 seconds
- */
-
-static void
-ali_timeout(unsigned long ptr)
-{
-	struct trident_card *card = (struct trident_card *) ptr;
-	u16 temp = 0;
-
-	/* Enable GPIO IRQ (MISCINT bit 18h) */
-	temp = inw(TRID_REG(card, T4D_MISCINT + 2));
-	temp |= 0x0004;
-	outw(temp, TRID_REG(card, T4D_MISCINT + 2));
-}
-
-/*
- *	Set up the timer to clear the vol change notification
- */
-
-static void
-ali_set_timer(struct trident_card *card)
-{
-	/* Add Timer Routine to Enable GPIO IRQ */
-	del_timer(&card->timer);	/* Never queue twice */
-	card->timer.function = ali_timeout;
-	card->timer.data = (unsigned long) card;
-	card->timer.expires = jiffies + HZ / 10;
-	add_timer(&card->timer);
-}
-
-/*
- *	Process a GPIO event
- */
-
-static void
-ali_queue_task(struct trident_card *card, int opt)
-{
-	u16 temp;
-
-	/* Disable GPIO IRQ (MISCINT bit 18h) */
-	temp = inw(TRID_REG(card, T4D_MISCINT + 2));
-	temp &= (u16) (~0x0004);
-	outw(temp, TRID_REG(card, T4D_MISCINT + 2));
-
-	/* Adjust the volume */
-	ali_hwvol_control(card, opt);
-
-	/* Set the timer for 1/10th sec */
-	ali_set_timer(card);
-}
-
-static void
-cyber_address_interrupt(struct trident_card *card)
-{
-	int i, irq_status;
-	struct trident_state *state;
-	unsigned int channel;
-
-	/* Update the pointers for all channels we are running. */
-	/* FIXED: read interrupt status only once */
-	irq_status = inl(TRID_REG(card, T4D_AINT_A));
-
-	pr_debug("cyber_address_interrupt: irq_status 0x%X\n", irq_status);
-
-	for (i = 0; i < NR_HW_CH; i++) {
-		channel = 31 - i;
-		if (irq_status & (1 << channel)) {
-			/* clear bit by writing a 1, zeroes are ignored */
-			outl((1 << channel), TRID_REG(card, T4D_AINT_A));
-
-			pr_debug("cyber_interrupt: channel %d\n", channel);
-
-			if ((state = card->states[i]) != NULL) {
-				trident_update_ptr(state);
-			} else {
-				printk(KERN_WARNING "cyber5050: spurious "
-				       "channel irq %d.\n", channel);
-				trident_stop_voice(card, channel);
-				trident_disable_voice_irq(card, channel);
-			}
-		}
-	}
-}
-
-static irqreturn_t
-trident_interrupt(int irq, void *dev_id)
-{
-	struct trident_card *card = (struct trident_card *) dev_id;
-	u32 event;
-	u32 gpio;
-
-	spin_lock(&card->lock);
-	event = inl(TRID_REG(card, T4D_MISCINT));
-
-	pr_debug("trident: trident_interrupt called, MISCINT = 0x%08x\n",
-		 event);
-
-	if (event & ADDRESS_IRQ) {
-		card->address_interrupt(card);
-	}
-
-	if (card->pci_id == PCI_DEVICE_ID_ALI_5451) {
-		/* GPIO IRQ (H/W Volume Control) */
-		event = inl(TRID_REG(card, T4D_MISCINT));
-		if (event & (1 << 25)) {
-			gpio = inl(TRID_REG(card, ALI_GPIO));
-			if (!timer_pending(&card->timer))
-				ali_queue_task(card, gpio & 0x07);
-		}
-		event = inl(TRID_REG(card, T4D_MISCINT));
-		outl(event | (ST_TARGET_REACHED | MIXER_OVERFLOW | MIXER_UNDERFLOW),
-		     TRID_REG(card, T4D_MISCINT));
-		spin_unlock(&card->lock);
-		return IRQ_HANDLED;
-	}
-
-	/* manually clear interrupt status, bad hardware design, blame T^2 */
-	outl((ST_TARGET_REACHED | MIXER_OVERFLOW | MIXER_UNDERFLOW),
-	     TRID_REG(card, T4D_MISCINT));
-	spin_unlock(&card->lock);
-	return IRQ_HANDLED;
-}
-
-/* in this loop, dmabuf.count signifies the amount of data that is waiting */
-/* to be copied to the user's buffer.  it is filled by the dma machine and */
-/* drained by this loop. */
-static ssize_t
-trident_read(struct file *file, char __user *buffer, size_t count, loff_t * ppos)
-{
-	struct trident_state *state = (struct trident_state *)file->private_data;
-	struct dmabuf *dmabuf = &state->dmabuf;
-	ssize_t ret = 0;
-	unsigned long flags;
-	unsigned swptr;
-	int cnt;
-
-	pr_debug("trident: trident_read called, count = %zd\n", count);
-
-	VALIDATE_STATE(state);
-
-	if (dmabuf->mapped)
-		return -ENXIO;
-	if (!access_ok(VERIFY_WRITE, buffer, count))
-		return -EFAULT;
-
-	mutex_lock(&state->sem);
-	if (!dmabuf->ready && (ret = prog_dmabuf_record(state)))
-		goto out;
-
-	while (count > 0) {
-		spin_lock_irqsave(&state->card->lock, flags);
-		if (dmabuf->count > (signed) dmabuf->dmasize) {
-			/* buffer overrun, we are recovering from */
-			/* sleep_on_timeout, resync hwptr and swptr, */
-			/* make process flush the buffer */
-			dmabuf->count = dmabuf->dmasize;
-			dmabuf->swptr = dmabuf->hwptr;
-		}
-		swptr = dmabuf->swptr;
-		cnt = dmabuf->dmasize - swptr;
-		if (dmabuf->count < cnt)
-			cnt = dmabuf->count;
-		spin_unlock_irqrestore(&state->card->lock, flags);
-
-		if (cnt > count)
-			cnt = count;
-		if (cnt <= 0) {
-			unsigned long tmo;
-			/* buffer is empty, start the dma machine and */
-			/* wait for data to be recorded */
-			start_adc(state);
-			if (file->f_flags & O_NONBLOCK) {
-				if (!ret)
-					ret = -EAGAIN;
-				goto out;
-			}
-
-			mutex_unlock(&state->sem);
-			/* No matter how much space left in the buffer, */
-			/* we have to wait until CSO == ESO/2 or CSO == ESO */
-			/* when address engine interrupts */
-			tmo = (dmabuf->dmasize * HZ) / (dmabuf->rate * 2);
-			tmo >>= sample_shift[dmabuf->fmt];
-			/* There are two situations when sleep_on_timeout returns, one is when
-			   the interrupt is serviced correctly and the process is waked up by
-			   ISR ON TIME. Another is when timeout is expired, which means that
-			   either interrupt is NOT serviced correctly (pending interrupt) or it
-			   is TOO LATE for the process to be scheduled to run (scheduler latency)
-			   which results in a (potential) buffer overrun. And worse, there is
-			   NOTHING we can do to prevent it. */
-			if (!interruptible_sleep_on_timeout(&dmabuf->wait, tmo)) {
-				pr_debug(KERN_ERR "trident: recording schedule timeout, "
-					 "dmasz %u fragsz %u count %i hwptr %u swptr %u\n",
-					 dmabuf->dmasize, dmabuf->fragsize, dmabuf->count,
-					 dmabuf->hwptr, dmabuf->swptr);
-
-				/* a buffer overrun, we delay the recovery until next time the
-				   while loop begin and we REALLY have space to record */
-			}
-			if (signal_pending(current)) {
-				if (!ret)
-					ret = -ERESTARTSYS;
-				goto out;
-			}
-			mutex_lock(&state->sem);
-			if (dmabuf->mapped) {
-				if (!ret)
-					ret = -ENXIO;
-				goto out;
-			}
-			continue;
-		}
-
-		if (copy_to_user(buffer, dmabuf->rawbuf + swptr, cnt)) {
-			if (!ret)
-				ret = -EFAULT;
-			goto out;
-		}
-
-		swptr = (swptr + cnt) % dmabuf->dmasize;
-
-		spin_lock_irqsave(&state->card->lock, flags);
-		dmabuf->swptr = swptr;
-		dmabuf->count -= cnt;
-		spin_unlock_irqrestore(&state->card->lock, flags);
-
-		count -= cnt;
-		buffer += cnt;
-		ret += cnt;
-		start_adc(state);
-	}
-out:
-	mutex_unlock(&state->sem);
-	return ret;
-}
-
-/* in this loop, dmabuf.count signifies the amount of data that is waiting to be dma to
-   the soundcard.  it is drained by the dma machine and filled by this loop. */
-
-static ssize_t
-trident_write(struct file *file, const char __user *buffer, size_t count, loff_t * ppos)
-{
-	struct trident_state *state = (struct trident_state *)file->private_data;
-	struct dmabuf *dmabuf = &state->dmabuf;
-	ssize_t ret;
-	unsigned long flags;
-	unsigned swptr;
-	int cnt;
-	unsigned int state_cnt;
-	unsigned int copy_count;
-	int lret; /* for lock_set_fmt */
-
-	pr_debug("trident: trident_write called, count = %zd\n", count);
-
-	VALIDATE_STATE(state);
-
-	/*
-	 *      Guard against an mmap or ioctl while writing
-	 */
-
-	mutex_lock(&state->sem);
-
-	if (dmabuf->mapped) {
-		ret = -ENXIO;
-		goto out;
-	}
-	if (!dmabuf->ready && (ret = prog_dmabuf_playback(state)))
-		goto out;
-
-	if (!access_ok(VERIFY_READ, buffer, count)) {
-		ret = -EFAULT;
-		goto out;
-	}
-
-	ret = 0;
-
-	while (count > 0) {
-		spin_lock_irqsave(&state->card->lock, flags);
-		if (dmabuf->count < 0) {
-			/* buffer underrun, we are recovering from */
-			/* sleep_on_timeout, resync hwptr and swptr */
-			dmabuf->count = 0;
-			dmabuf->swptr = dmabuf->hwptr;
-		}
-		swptr = dmabuf->swptr;
-		cnt = dmabuf->dmasize - swptr;
-		if (dmabuf->count + cnt > dmabuf->dmasize)
-			cnt = dmabuf->dmasize - dmabuf->count;
-		spin_unlock_irqrestore(&state->card->lock, flags);
-
-		if (cnt > count)
-			cnt = count;
-		if (cnt <= 0) {
-			unsigned long tmo;
-			/* buffer is full, start the dma machine and */
-			/* wait for data to be played */
-			start_dac(state);
-			if (file->f_flags & O_NONBLOCK) {
-				if (!ret)
-					ret = -EAGAIN;
-				goto out;
-			}
-			/* No matter how much data left in the buffer, */
-			/* we have to wait until CSO == ESO/2 or CSO == ESO */
-			/* when address engine interrupts */
-			lock_set_fmt(state);
-			tmo = (dmabuf->dmasize * HZ) / (dmabuf->rate * 2);
-			tmo >>= sample_shift[dmabuf->fmt];
-			unlock_set_fmt(state);
-			mutex_unlock(&state->sem);
-
-			/* There are two situations when sleep_on_timeout */
-			/* returns, one is when the interrupt is serviced */
-			/* correctly and the process is waked up by ISR */
-			/* ON TIME. Another is when timeout is expired, which */
-			/* means that either interrupt is NOT serviced */
-			/* correctly (pending interrupt) or it is TOO LATE */
-			/* for the process to be scheduled to run */
-			/* (scheduler latency) which results in a (potential) */
-			/* buffer underrun. And worse, there is NOTHING we */
-			/* can do to prevent it. */
-			if (!interruptible_sleep_on_timeout(&dmabuf->wait, tmo)) {
-				pr_debug(KERN_ERR "trident: playback schedule "
-					 "timeout, dmasz %u fragsz %u count %i "
-					 "hwptr %u swptr %u\n", dmabuf->dmasize,
-					 dmabuf->fragsize, dmabuf->count,
-					 dmabuf->hwptr, dmabuf->swptr);
-
-				/* a buffer underrun, we delay the recovery */
-				/* until next time the while loop begin and */
-				/* we REALLY have data to play */
-			}
-			if (signal_pending(current)) {
-				if (!ret)
-					ret = -ERESTARTSYS;
-				goto out_nolock;
-			}
-			mutex_lock(&state->sem);
-			if (dmabuf->mapped) {
-				if (!ret)
-					ret = -ENXIO;
-				goto out;
-			}
-			continue;
-		}
-		if ((lret = lock_set_fmt(state)) < 0) {
-			ret = lret;
-			goto out;
-		}
-
-		if (state->chans_num == 6) {
-			copy_count = 0;
-			state_cnt = 0;
-			if (ali_write_5_1(state, buffer, cnt, &copy_count,
-					  &state_cnt) == -EFAULT) {
-				if (state_cnt) {
-					swptr = (swptr + state_cnt) % dmabuf->dmasize;
-					spin_lock_irqsave(&state->card->lock, flags);
-					dmabuf->swptr = swptr;
-					dmabuf->count += state_cnt;
-					dmabuf->endcleared = 0;
-					spin_unlock_irqrestore(&state->card->lock, flags);
-				}
-				ret += copy_count;
-				if (!ret)
-					ret = -EFAULT;
-				unlock_set_fmt(state);
-				goto out;
-			}
-		} else {
-			if (copy_from_user(dmabuf->rawbuf + swptr,
-					   buffer, cnt)) {
-				if (!ret)
-					ret = -EFAULT;
-				unlock_set_fmt(state);
-				goto out;
-			}
-			state_cnt = cnt;
-		}
-		unlock_set_fmt(state);
-
-		swptr = (swptr + state_cnt) % dmabuf->dmasize;
-
-		spin_lock_irqsave(&state->card->lock, flags);
-		dmabuf->swptr = swptr;
-		dmabuf->count += state_cnt;
-		dmabuf->endcleared = 0;
-		spin_unlock_irqrestore(&state->card->lock, flags);
-
-		count -= cnt;
-		buffer += cnt;
-		ret += cnt;
-		start_dac(state);
-	}
-out:
-	mutex_unlock(&state->sem);
-out_nolock:
-	return ret;
-}
-
-/* No kernel lock - we have our own spinlock */
-static unsigned int
-trident_poll(struct file *file, struct poll_table_struct *wait)
-{
-	struct trident_state *state = (struct trident_state *)file->private_data;
-	struct dmabuf *dmabuf = &state->dmabuf;
-	unsigned long flags;
-	unsigned int mask = 0;
-
-	VALIDATE_STATE(state);
-
-	/*
-	 *      Guard against a parallel poll and write causing multiple
-	 *      prog_dmabuf events
-	 */
-
-	mutex_lock(&state->sem);
-
-	if (file->f_mode & FMODE_WRITE) {
-		if (!dmabuf->ready && prog_dmabuf_playback(state)) {
-			mutex_unlock(&state->sem);
-			return 0;
-		}
-		poll_wait(file, &dmabuf->wait, wait);
-	}
-	if (file->f_mode & FMODE_READ) {
-		if (!dmabuf->ready && prog_dmabuf_record(state)) {
-			mutex_unlock(&state->sem);
-			return 0;
-		}
-		poll_wait(file, &dmabuf->wait, wait);
-	}
-
-	mutex_unlock(&state->sem);
-
-	spin_lock_irqsave(&state->card->lock, flags);
-	trident_update_ptr(state);
-	if (file->f_mode & FMODE_READ) {
-		if (dmabuf->count >= (signed) dmabuf->fragsize)
-			mask |= POLLIN | POLLRDNORM;
-	}
-	if (file->f_mode & FMODE_WRITE) {
-		if (dmabuf->mapped) {
-			if (dmabuf->count >= (signed) dmabuf->fragsize)
-				mask |= POLLOUT | POLLWRNORM;
-		} else {
-			if ((signed) dmabuf->dmasize >= dmabuf->count +
-			    (signed) dmabuf->fragsize)
-				mask |= POLLOUT | POLLWRNORM;
-		}
-	}
-	spin_unlock_irqrestore(&state->card->lock, flags);
-
-	return mask;
-}
-
-static int
-trident_mmap(struct file *file, struct vm_area_struct *vma)
-{
-	struct trident_state *state = (struct trident_state *)file->private_data;
-	struct dmabuf *dmabuf = &state->dmabuf;
-	int ret = -EINVAL;
-	unsigned long size;
-
-	VALIDATE_STATE(state);
-
-	/*
-	 *      Lock against poll read write or mmap creating buffers. Also lock
-	 *      a read or write against an mmap.
-	 */
-
-	mutex_lock(&state->sem);
-
-	if (vma->vm_flags & VM_WRITE) {
-		if ((ret = prog_dmabuf_playback(state)) != 0)
-			goto out;
-	} else if (vma->vm_flags & VM_READ) {
-		if ((ret = prog_dmabuf_record(state)) != 0)
-			goto out;
-	} else
-		goto out;
-
-	ret = -EINVAL;
-	if (vma->vm_pgoff != 0)
-		goto out;
-	size = vma->vm_end - vma->vm_start;
-	if (size > (PAGE_SIZE << dmabuf->buforder))
-		goto out;
-	ret = -EAGAIN;
-	if (remap_pfn_range(vma, vma->vm_start,
-			     virt_to_phys(dmabuf->rawbuf) >> PAGE_SHIFT,
-			     size, vma->vm_page_prot))
-		goto out;
-	dmabuf->mapped = 1;
-	ret = 0;
-out:
-	mutex_unlock(&state->sem);
-	return ret;
-}
-
-static int
-trident_ioctl(struct inode *inode, struct file *file,
-	      unsigned int cmd, unsigned long arg)
-{
-	struct trident_state *state = (struct trident_state *)file->private_data;
-	struct dmabuf *dmabuf = &state->dmabuf;
-	unsigned long flags;
-	audio_buf_info abinfo;
-	count_info cinfo;
-	int val, mapped, ret = 0;
-	struct trident_card *card = state->card;
-	void __user *argp = (void __user *)arg;
-	int __user *p = argp;
-
-	VALIDATE_STATE(state);
-
-
-	mapped = ((file->f_mode & (FMODE_WRITE | FMODE_READ)) && dmabuf->mapped);
-
-	pr_debug("trident: trident_ioctl, command = %2d, arg = 0x%08x\n",
-		 _IOC_NR(cmd), arg ? *p : 0);
-
-	switch (cmd) {
-	case OSS_GETVERSION:
-		ret = put_user(SOUND_VERSION, p);
-		break;
-
-	case SNDCTL_DSP_RESET:
-		/* FIXME: spin_lock ? */
-		if (file->f_mode & FMODE_WRITE) {
-			stop_dac(state);
-			synchronize_irq(card->irq);
-			dmabuf->ready = 0;
-			dmabuf->swptr = dmabuf->hwptr = 0;
-			dmabuf->count = dmabuf->total_bytes = 0;
-		}
-		if (file->f_mode & FMODE_READ) {
-			stop_adc(state);
-			synchronize_irq(card->irq);
-			dmabuf->ready = 0;
-			dmabuf->swptr = dmabuf->hwptr = 0;
-			dmabuf->count = dmabuf->total_bytes = 0;
-		}
-		break;
-
-	case SNDCTL_DSP_SYNC:
-		if (file->f_mode & FMODE_WRITE)
-			ret = drain_dac(state, file->f_flags & O_NONBLOCK);
-		break;
-
-	case SNDCTL_DSP_SPEED:	/* set smaple rate */
-		if (get_user(val, p)) {
-			ret = -EFAULT;
-			break;
-		}
-		if (val >= 0) {
-			if (file->f_mode & FMODE_WRITE) {
-				stop_dac(state);
-				dmabuf->ready = 0;
-				spin_lock_irqsave(&state->card->lock, flags);
-				trident_set_dac_rate(state, val);
-				spin_unlock_irqrestore(&state->card->lock, flags);
-			}
-			if (file->f_mode & FMODE_READ) {
-				stop_adc(state);
-				dmabuf->ready = 0;
-				spin_lock_irqsave(&state->card->lock, flags);
-				trident_set_adc_rate(state, val);
-				spin_unlock_irqrestore(&state->card->lock, flags);
-			}
-		}
-		ret = put_user(dmabuf->rate, p);
-		break;
-
-	case SNDCTL_DSP_STEREO:	/* set stereo or mono channel */
-		if (get_user(val, p)) {
-			ret = -EFAULT;
-			break;
-		}
-		if ((ret = lock_set_fmt(state)) < 0)
-			return ret;
-
-		if (file->f_mode & FMODE_WRITE) {
-			stop_dac(state);
-			dmabuf->ready = 0;
-			if (val)
-				dmabuf->fmt |= TRIDENT_FMT_STEREO;
-			else
-				dmabuf->fmt &= ~TRIDENT_FMT_STEREO;
-		}
-		if (file->f_mode & FMODE_READ) {
-			stop_adc(state);
-			dmabuf->ready = 0;
-			if (val)
-				dmabuf->fmt |= TRIDENT_FMT_STEREO;
-			else
-				dmabuf->fmt &= ~TRIDENT_FMT_STEREO;
-		}
-		unlock_set_fmt(state);
-		break;
-
-	case SNDCTL_DSP_GETBLKSIZE:
-		if (file->f_mode & FMODE_WRITE) {
-			if ((val = prog_dmabuf_playback(state)))
-				ret = val;
-			else
-				ret = put_user(dmabuf->fragsize, p);
-			break;
-		}
-		if (file->f_mode & FMODE_READ) {
-			if ((val = prog_dmabuf_record(state)))
-				ret = val;
-			else
-				ret = put_user(dmabuf->fragsize, p);
-			break;
-		}
-		/* neither READ nor WRITE? is this even possible? */
-		ret = -EINVAL;
-		break;
-
-
-	case SNDCTL_DSP_GETFMTS: /* Returns a mask of supported sample format */
-		ret = put_user(AFMT_S16_LE | AFMT_U16_LE | AFMT_S8 |
-			       AFMT_U8, p);
-		break;
-
-	case SNDCTL_DSP_SETFMT:	/* Select sample format */
-		if (get_user(val, p)) {
-			ret = -EFAULT;
-			break;
-		}
-		if ((ret = lock_set_fmt(state)) < 0)
-			return ret;
-
-		if (val != AFMT_QUERY) {
-			if (file->f_mode & FMODE_WRITE) {
-				stop_dac(state);
-				dmabuf->ready = 0;
-				if (val == AFMT_S16_LE)
-					dmabuf->fmt |= TRIDENT_FMT_16BIT;
-				else
-					dmabuf->fmt &= ~TRIDENT_FMT_16BIT;
-			}
-			if (file->f_mode & FMODE_READ) {
-				stop_adc(state);
-				dmabuf->ready = 0;
-				if (val == AFMT_S16_LE)
-					dmabuf->fmt |= TRIDENT_FMT_16BIT;
-				else
-					dmabuf->fmt &= ~TRIDENT_FMT_16BIT;
-			}
-		}
-		unlock_set_fmt(state);
-		ret = put_user((dmabuf->fmt & TRIDENT_FMT_16BIT) ? AFMT_S16_LE :
-			       AFMT_U8, p);
-		break;
-
-	case SNDCTL_DSP_CHANNELS:
-		if (get_user(val, p)) {
-			ret = -EFAULT;
-			break;
-		}
-		if (val != 0) {
-			if ((ret = lock_set_fmt(state)) < 0)
-				return ret;
-
-			if (file->f_mode & FMODE_WRITE) {
-				stop_dac(state);
-				dmabuf->ready = 0;
-
-				//prevent from memory leak
-				if ((state->chans_num > 2) && (state->chans_num != val)) {
-					ali_free_other_states_resources(state);
-					state->chans_num = 1;
-				}
-
-				if (val >= 2) {
-
-					dmabuf->fmt |= TRIDENT_FMT_STEREO;
-					if ((val == 6) && (state->card->pci_id == PCI_DEVICE_ID_ALI_5451)) {
-						if (card->rec_channel_use_count > 0) {
-							printk(KERN_ERR "trident: Record is "
-							       "working on the card!\n");
-							ret = -EBUSY;
-							unlock_set_fmt(state);
-							break;
-						}
-
-						ret = ali_setup_multi_channels(state->card, 6);
-						if (ret < 0) {
-							unlock_set_fmt(state);
-							break;
-						}
-						mutex_lock(&state->card->open_mutex);
-						ret = ali_allocate_other_states_resources(state, 6);
-						if (ret < 0) {
-							mutex_unlock(&state->card->open_mutex);
-							unlock_set_fmt(state);
-							break;
-						}
-						state->card->multi_channel_use_count++;
-						mutex_unlock(&state->card->open_mutex);
-					} else
-						val = 2;	/*yield to 2-channels */
-				} else
-					dmabuf->fmt &= ~TRIDENT_FMT_STEREO;
-				state->chans_num = val;
-			}
-			if (file->f_mode & FMODE_READ) {
-				stop_adc(state);
-				dmabuf->ready = 0;
-				if (val >= 2) {
-					if (!((file->f_mode & FMODE_WRITE) &&
-					      (val == 6)))
-						val = 2;
-					dmabuf->fmt |= TRIDENT_FMT_STEREO;
-				} else
-					dmabuf->fmt &= ~TRIDENT_FMT_STEREO;
-				state->chans_num = val;
-			}
-			unlock_set_fmt(state);
-		}
-		ret = put_user(val, p);
-		break;
-
-	case SNDCTL_DSP_POST:
-		/* Cause the working fragment to be output */
-		break;
-
-	case SNDCTL_DSP_SUBDIVIDE:
-		if (dmabuf->subdivision) {
-			ret = -EINVAL;
-			break;
-		}
-		if (get_user(val, p)) {
-			ret = -EFAULT;
-			break;
-		}
-		if (val != 1 && val != 2 && val != 4) {
-			ret = -EINVAL;
-			break;
-		}
-		dmabuf->subdivision = val;
-		break;
-
-	case SNDCTL_DSP_SETFRAGMENT:
-		if (get_user(val, p)) {
-			ret = -EFAULT;
-			break;
-		}
-
-		dmabuf->ossfragshift = val & 0xffff;
-		dmabuf->ossmaxfrags = (val >> 16) & 0xffff;
-		if (dmabuf->ossfragshift < 4)
-			dmabuf->ossfragshift = 4;
-		if (dmabuf->ossfragshift > 15)
-			dmabuf->ossfragshift = 15;
-		if (dmabuf->ossmaxfrags < 4)
-			dmabuf->ossmaxfrags = 4;
-
-		break;
-
-	case SNDCTL_DSP_GETOSPACE:
-		if (!(file->f_mode & FMODE_WRITE)) {
-			ret = -EINVAL;
-			break;
-		}
-		if (!dmabuf->ready && (val = prog_dmabuf_playback(state)) != 0) {
-			ret = val;
-			break;
-		}
-		spin_lock_irqsave(&state->card->lock, flags);
-		trident_update_ptr(state);
-		abinfo.fragsize = dmabuf->fragsize;
-		abinfo.bytes = dmabuf->dmasize - dmabuf->count;
-		abinfo.fragstotal = dmabuf->numfrag;
-		abinfo.fragments = abinfo.bytes >> dmabuf->fragshift;
-		spin_unlock_irqrestore(&state->card->lock, flags);
-		ret = copy_to_user(argp, &abinfo, sizeof (abinfo)) ?
-			-EFAULT : 0;
-		break;
-
-	case SNDCTL_DSP_GETISPACE:
-		if (!(file->f_mode & FMODE_READ)) {
-			ret = -EINVAL;
-			break;
-		}
-		if (!dmabuf->ready && (val = prog_dmabuf_record(state)) != 0) {
-			ret = val;
-			break;
-		}
-		spin_lock_irqsave(&state->card->lock, flags);
-		trident_update_ptr(state);
-		abinfo.fragsize = dmabuf->fragsize;
-		abinfo.bytes = dmabuf->count;
-		abinfo.fragstotal = dmabuf->numfrag;
-		abinfo.fragments = abinfo.bytes >> dmabuf->fragshift;
-		spin_unlock_irqrestore(&state->card->lock, flags);
-		ret = copy_to_user(argp, &abinfo, sizeof (abinfo)) ?
-			-EFAULT : 0;
-		break;
-
-	case SNDCTL_DSP_NONBLOCK:
-		file->f_flags |= O_NONBLOCK;
-		break;
-
-	case SNDCTL_DSP_GETCAPS:
-		ret = put_user(DSP_CAP_REALTIME | DSP_CAP_TRIGGER |
-			       DSP_CAP_MMAP | DSP_CAP_BIND, p);
-		break;
-
-	case SNDCTL_DSP_GETTRIGGER:
-		val = 0;
-		if ((file->f_mode & FMODE_READ) && dmabuf->enable)
-			val |= PCM_ENABLE_INPUT;
-		if ((file->f_mode & FMODE_WRITE) && dmabuf->enable)
-			val |= PCM_ENABLE_OUTPUT;
-		ret = put_user(val, p);
-		break;
-
-	case SNDCTL_DSP_SETTRIGGER:
-		if (get_user(val, p)) {
-			ret = -EFAULT;
-			break;
-		}
-		if (file->f_mode & FMODE_READ) {
-			if (val & PCM_ENABLE_INPUT) {
-				if (!dmabuf->ready &&
-				    (ret = prog_dmabuf_record(state)))
-					break;
-				start_adc(state);
-			} else
-				stop_adc(state);
-		}
-		if (file->f_mode & FMODE_WRITE) {
-			if (val & PCM_ENABLE_OUTPUT) {
-				if (!dmabuf->ready &&
-				    (ret = prog_dmabuf_playback(state)))
-					break;
-				start_dac(state);
-			} else
-				stop_dac(state);
-		}
-		break;
-
-	case SNDCTL_DSP_GETIPTR:
-		if (!(file->f_mode & FMODE_READ)) {
-			ret = -EINVAL;
-			break;
-		}
-		if (!dmabuf->ready && (val = prog_dmabuf_record(state))
-		    != 0) {
-			ret = val;
-			break;
-		}
-		spin_lock_irqsave(&state->card->lock, flags);
-		trident_update_ptr(state);
-		cinfo.bytes = dmabuf->total_bytes;
-		cinfo.blocks = dmabuf->count >> dmabuf->fragshift;
-		cinfo.ptr = dmabuf->hwptr;
-		if (dmabuf->mapped)
-			dmabuf->count &= dmabuf->fragsize - 1;
-		spin_unlock_irqrestore(&state->card->lock, flags);
-		ret = copy_to_user(argp, &cinfo, sizeof (cinfo)) ?
-			-EFAULT : 0;
-		break;
-
-	case SNDCTL_DSP_GETOPTR:
-		if (!(file->f_mode & FMODE_WRITE)) {
-			ret = -EINVAL;
-			break;
-		}
-		if (!dmabuf->ready && (val = prog_dmabuf_playback(state))
-		    != 0) {
-			ret = val;
-			break;
-		}
-
-		spin_lock_irqsave(&state->card->lock, flags);
-		trident_update_ptr(state);
-		cinfo.bytes = dmabuf->total_bytes;
-		cinfo.blocks = dmabuf->count >> dmabuf->fragshift;
-		cinfo.ptr = dmabuf->hwptr;
-		if (dmabuf->mapped)
-			dmabuf->count &= dmabuf->fragsize - 1;
-		spin_unlock_irqrestore(&state->card->lock, flags);
-		ret = copy_to_user(argp, &cinfo, sizeof (cinfo)) ?
-			-EFAULT : 0;
-		break;
-
-	case SNDCTL_DSP_SETDUPLEX:
-		ret = -EINVAL;
-		break;
-
-	case SNDCTL_DSP_GETODELAY:
-		if (!(file->f_mode & FMODE_WRITE)) {
-			ret = -EINVAL;
-			break;
-		}
-		if (!dmabuf->ready && (val = prog_dmabuf_playback(state)) != 0) {
-			ret = val;
-			break;
-		}
-		spin_lock_irqsave(&state->card->lock, flags);
-		trident_update_ptr(state);
-		val = dmabuf->count;
-		spin_unlock_irqrestore(&state->card->lock, flags);
-		ret = put_user(val, p);
-		break;
-
-	case SOUND_PCM_READ_RATE:
-		ret = put_user(dmabuf->rate, p);
-		break;
-
-	case SOUND_PCM_READ_CHANNELS:
-		ret = put_user((dmabuf->fmt & TRIDENT_FMT_STEREO) ? 2 : 1,
-			       p);
-		break;
-
-	case SOUND_PCM_READ_BITS:
-		ret = put_user((dmabuf->fmt & TRIDENT_FMT_16BIT) ? AFMT_S16_LE :
-			       AFMT_U8, p);
-		break;
-
-	case SNDCTL_DSP_GETCHANNELMASK:
-		ret = put_user(DSP_BIND_FRONT | DSP_BIND_SURR |
-			       DSP_BIND_CENTER_LFE,  p);
-		break;
-
-	case SNDCTL_DSP_BIND_CHANNEL:
-		if (state->card->pci_id != PCI_DEVICE_ID_SI_7018) {
-			ret = -EINVAL;
-			break;
-		}
-
-		if (get_user(val, p)) {
-			ret = -EFAULT;
-			break;
-		}
-		if (val == DSP_BIND_QUERY) {
-			val = dmabuf->channel->attribute | 0x3c00;
-			val = attr2mask[val >> 8];
-		} else {
-			dmabuf->ready = 0;
-			if (file->f_mode & FMODE_READ)
-				dmabuf->channel->attribute = (CHANNEL_REC |
-							      SRC_ENABLE);
-			if (file->f_mode & FMODE_WRITE)
-				dmabuf->channel->attribute = (CHANNEL_SPC_PB |
-							      SRC_ENABLE);
-			dmabuf->channel->attribute |= mask2attr[ffs(val)];
-		}
-		ret = put_user(val, p);
-		break;
-
-	case SNDCTL_DSP_MAPINBUF:
-	case SNDCTL_DSP_MAPOUTBUF:
-	case SNDCTL_DSP_SETSYNCRO:
-	case SOUND_PCM_WRITE_FILTER:
-	case SOUND_PCM_READ_FILTER:
-	default:
-		ret = -EINVAL;
-		break;
-
-	}
-	return ret;
-}
-
-static int
-trident_open(struct inode *inode, struct file *file)
-{
-	int i = 0;
-	int minor = iminor(inode);
-	struct trident_card *card = devs;
-	struct trident_state *state = NULL;
-	struct dmabuf *dmabuf = NULL;
-	unsigned long flags;
-
-	/* Added by Matt Wu 01-05-2001 */
-	/* TODO: there's some redundacy here wrt the check below */
-	/* for multi_use_count > 0. Should we return -EBUSY or find */
-	/* a different card? for now, don't break current behaviour */
-	/* -- mulix */
-	if (file->f_mode & FMODE_READ) {
-		if (card->pci_id == PCI_DEVICE_ID_ALI_5451) {
-			if (card->multi_channel_use_count > 0)
-				return -EBUSY;
-		}
-	}
-
-	/* find an available virtual channel (instance of /dev/dsp) */
-	while (card != NULL) {
-		mutex_lock(&card->open_mutex);
-		if (file->f_mode & FMODE_READ) {
-			/* Skip opens on cards that are in 6 channel mode */
-			if (card->multi_channel_use_count > 0) {
-				mutex_unlock(&card->open_mutex);
-				card = card->next;
-				continue;
-			}
-		}
-		for (i = 0; i < NR_HW_CH; i++) {
-			if (card->states[i] == NULL) {
-				state = card->states[i] = kzalloc(sizeof(*state), GFP_KERNEL);
-				if (state == NULL) {
-					mutex_unlock(&card->open_mutex);
-					return -ENOMEM;
-				}
-				mutex_init(&state->sem);
-				dmabuf = &state->dmabuf;
-				goto found_virt;
-			}
-		}
-		mutex_unlock(&card->open_mutex);
-		card = card->next;
-	}
-	/* no more virtual channel avaiable */
-	if (!state) {
-		return -ENODEV;
-	}
-      found_virt:
-	/* found a free virtual channel, allocate hardware channels */
-	if (file->f_mode & FMODE_READ)
-		dmabuf->channel = card->alloc_rec_pcm_channel(card);
-	else
-		dmabuf->channel = card->alloc_pcm_channel(card);
-
-	if (dmabuf->channel == NULL) {
-		kfree(card->states[i]);
-		card->states[i] = NULL;
-		return -ENODEV;
-	}
-
-	/* initialize the virtual channel */
-	state->virt = i;
-	state->card = card;
-	state->magic = TRIDENT_STATE_MAGIC;
-	init_waitqueue_head(&dmabuf->wait);
-	file->private_data = state;
-
-	/* set default sample format. According to OSS Programmer's */
-	/* Guide  /dev/dsp should be default to unsigned 8-bits, mono, */
-	/* with sample rate 8kHz and /dev/dspW will accept 16-bits sample */
-	if (file->f_mode & FMODE_WRITE) {
-		dmabuf->fmt &= ~TRIDENT_FMT_MASK;
-		if ((minor & 0x0f) == SND_DEV_DSP16)
-			dmabuf->fmt |= TRIDENT_FMT_16BIT;
-		dmabuf->ossfragshift = 0;
-		dmabuf->ossmaxfrags = 0;
-		dmabuf->subdivision = 0;
-		if (card->pci_id == PCI_DEVICE_ID_SI_7018) {
-			/* set default channel attribute to normal playback */
-			dmabuf->channel->attribute = CHANNEL_PB;
-		}
-		spin_lock_irqsave(&card->lock, flags);
-		trident_set_dac_rate(state, 8000);
-		spin_unlock_irqrestore(&card->lock, flags);
-	}
-
-	if (file->f_mode & FMODE_READ) {
-		/* FIXME: Trident 4d can only record in signed 16-bits stereo, */
-		/* 48kHz sample, to be dealed with in trident_set_adc_rate() ?? */
-		dmabuf->fmt &= ~TRIDENT_FMT_MASK;
-		if ((minor & 0x0f) == SND_DEV_DSP16)
-			dmabuf->fmt |= TRIDENT_FMT_16BIT;
-		dmabuf->ossfragshift = 0;
-		dmabuf->ossmaxfrags = 0;
-		dmabuf->subdivision = 0;
-		if (card->pci_id == PCI_DEVICE_ID_SI_7018) {
-			/* set default channel attribute to 0x8a80, record from
-			   PCM L/R FIFO and mono = (left + right + 1)/2 */
-			dmabuf->channel->attribute = (CHANNEL_REC | PCM_LR |
-						      MONO_MIX);
-		}
-		spin_lock_irqsave(&card->lock, flags);
-		trident_set_adc_rate(state, 8000);
-		spin_unlock_irqrestore(&card->lock, flags);
-
-		/* Added by Matt Wu 01-05-2001 */
-		if (card->pci_id == PCI_DEVICE_ID_ALI_5451)
-			card->rec_channel_use_count++;
-	}
-
-	state->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
-	mutex_unlock(&card->open_mutex);
-
-	pr_debug("trident: open virtual channel %d, hard channel %d\n",
-		 state->virt, dmabuf->channel->num);
-
-	return nonseekable_open(inode, file);
-}
-
-static int
-trident_release(struct inode *inode, struct file *file)
-{
-	struct trident_state *state = (struct trident_state *)file->private_data;
-	struct trident_card *card;
-	struct dmabuf *dmabuf;
-
-	VALIDATE_STATE(state);
-
-	card = state->card;
-	dmabuf = &state->dmabuf;
-
-	if (file->f_mode & FMODE_WRITE) {
-		trident_clear_tail(state);
-		drain_dac(state, file->f_flags & O_NONBLOCK);
-	}
-
-	pr_debug("trident: closing virtual channel %d, hard channel %d\n",
-		 state->virt, dmabuf->channel->num);
-
-	/* stop DMA state machine and free DMA buffers/channels */
-	mutex_lock(&card->open_mutex);
-
-	if (file->f_mode & FMODE_WRITE) {
-		stop_dac(state);
-		dealloc_dmabuf(&state->dmabuf, state->card->pci_dev);
-		state->card->free_pcm_channel(state->card, dmabuf->channel->num);
-
-		/* Added by Matt Wu */
-		if (card->pci_id == PCI_DEVICE_ID_ALI_5451) {
-			if (state->chans_num > 2) {
-				if (card->multi_channel_use_count-- < 0)
-					card->multi_channel_use_count = 0;
-				if (card->multi_channel_use_count == 0)
-					ali_close_multi_channels();
-				ali_free_other_states_resources(state);
-			}
-		}
-	}
-	if (file->f_mode & FMODE_READ) {
-		stop_adc(state);
-		dealloc_dmabuf(&state->dmabuf, state->card->pci_dev);
-		state->card->free_pcm_channel(state->card, dmabuf->channel->num);
-
-		/* Added by Matt Wu */
-		if (card->pci_id == PCI_DEVICE_ID_ALI_5451) {
-			if (card->rec_channel_use_count-- < 0)
-				card->rec_channel_use_count = 0;
-		}
-	}
-
-	card->states[state->virt] = NULL;
-	kfree(state);
-
-	/* we're covered by the open_mutex */
-	mutex_unlock(&card->open_mutex);
-
-	return 0;
-}
-
-static const struct file_operations trident_audio_fops = {
-	.owner = THIS_MODULE,
-	.llseek = no_llseek,
-	.read = trident_read,
-	.write = trident_write,
-	.poll = trident_poll,
-	.ioctl = trident_ioctl,
-	.mmap = trident_mmap,
-	.open = trident_open,
-	.release = trident_release,
-};
-
-/* trident specific AC97 functions */
-/* Write AC97 codec registers */
-static void
-trident_ac97_set(struct ac97_codec *codec, u8 reg, u16 val)
-{
-	struct trident_card *card = (struct trident_card *)codec->private_data;
-	unsigned int address, mask, busy;
-	unsigned short count = 0xffff;
-	unsigned long flags;
-	u32 data;
-
-	data = ((u32) val) << 16;
-
-	switch (card->pci_id) {
-	default:
-	case PCI_DEVICE_ID_SI_7018:
-		address = SI_AC97_WRITE;
-		mask = SI_AC97_BUSY_WRITE | SI_AC97_AUDIO_BUSY;
-		if (codec->id)
-			mask |= SI_AC97_SECONDARY;
-		busy = SI_AC97_BUSY_WRITE;
-		break;
-	case PCI_DEVICE_ID_TRIDENT_4DWAVE_DX:
-		address = DX_ACR0_AC97_W;
-		mask = busy = DX_AC97_BUSY_WRITE;
-		break;
-	case PCI_DEVICE_ID_TRIDENT_4DWAVE_NX:
-		address = NX_ACR1_AC97_W;
-		mask = NX_AC97_BUSY_WRITE;
-		if (codec->id)
-			mask |= NX_AC97_WRITE_SECONDARY;
-		busy = NX_AC97_BUSY_WRITE;
-		break;
-	case PCI_DEVICE_ID_INTERG_5050:
-		address = SI_AC97_WRITE;
-		mask = busy = SI_AC97_BUSY_WRITE;
-		if (codec->id)
-			mask |= SI_AC97_SECONDARY;
-		break;
-	}
-
-	spin_lock_irqsave(&card->lock, flags);
-	do {
-		if ((inw(TRID_REG(card, address)) & busy) == 0)
-			break;
-	} while (--count);
-
-	data |= (mask | (reg & AC97_REG_ADDR));
-
-	if (count == 0) {
-		printk(KERN_ERR "trident: AC97 CODEC write timed out.\n");
-		spin_unlock_irqrestore(&card->lock, flags);
-		return;
-	}
-
-	outl(data, TRID_REG(card, address));
-	spin_unlock_irqrestore(&card->lock, flags);
-}
-
-/* Read AC97 codec registers */
-static u16
-trident_ac97_get(struct ac97_codec *codec, u8 reg)
-{
-	struct trident_card *card = (struct trident_card *)codec->private_data;
-	unsigned int address, mask, busy;
-	unsigned short count = 0xffff;
-	unsigned long flags;
-	u32 data;
-
-	switch (card->pci_id) {
-	default:
-	case PCI_DEVICE_ID_SI_7018:
-		address = SI_AC97_READ;
-		mask = SI_AC97_BUSY_READ | SI_AC97_AUDIO_BUSY;
-		if (codec->id)
-			mask |= SI_AC97_SECONDARY;
-		busy = SI_AC97_BUSY_READ;
-		break;
-	case PCI_DEVICE_ID_TRIDENT_4DWAVE_DX:
-		address = DX_ACR1_AC97_R;
-		mask = busy = DX_AC97_BUSY_READ;
-		break;
-	case PCI_DEVICE_ID_TRIDENT_4DWAVE_NX:
-		if (codec->id)
-			address = NX_ACR3_AC97_R_SECONDARY;
-		else
-			address = NX_ACR2_AC97_R_PRIMARY;
-		mask = NX_AC97_BUSY_READ;
-		busy = NX_AC97_BUSY_READ | NX_AC97_BUSY_DATA;
-		break;
-	case PCI_DEVICE_ID_INTERG_5050:
-		address = SI_AC97_READ;
-		mask = busy = SI_AC97_BUSY_READ;
-		if (codec->id)
-			mask |= SI_AC97_SECONDARY;
-		break;
-	}
-
-	data = (mask | (reg & AC97_REG_ADDR));
-
-	spin_lock_irqsave(&card->lock, flags);
-	outl(data, TRID_REG(card, address));
-	do {
-		data = inl(TRID_REG(card, address));
-		if ((data & busy) == 0)
-			break;
-	} while (--count);
-	spin_unlock_irqrestore(&card->lock, flags);
-
-	if (count == 0) {
-		printk(KERN_ERR "trident: AC97 CODEC read timed out.\n");
-		data = 0;
-	}
-	return ((u16) (data >> 16));
-}
-
-/* rewrite ac97 read and write mixer register by hulei for ALI*/
-static int
-acquirecodecaccess(struct trident_card *card)
-{
-	u16 wsemamask = 0x6000;	/* bit 14..13 */
-	u16 wsemabits;
-	u16 wcontrol;
-	int block = 0;
-	int ncount = 25;
-	while (1) {
-		wcontrol = inw(TRID_REG(card, ALI_AC97_WRITE));
-		wsemabits = wcontrol & wsemamask;
-
-		if (wsemabits == 0x4000)
-			return 1;	/* 0x4000 is audio ,then success */
-		if (ncount-- < 0)
-			break;
-		if (wsemabits == 0) {
-		      unlock:
-			outl(((u32) (wcontrol & 0x1eff) | 0x00004000),
-			     TRID_REG(card, ALI_AC97_WRITE));
-			continue;
-		}
-		udelay(20);
-	}
-	if (!block) {
-		pr_debug("accesscodecsemaphore: try unlock\n");
-		block = 1;
-		goto unlock;
-	}
-	return 0;
-}
-
-static void
-releasecodecaccess(struct trident_card *card)
-{
-	unsigned long wcontrol;
-	wcontrol = inl(TRID_REG(card, ALI_AC97_WRITE));
-	outl((wcontrol & 0xffff1eff), TRID_REG(card, ALI_AC97_WRITE));
-}
-
-static int
-waitforstimertick(struct trident_card *card)
-{
-	unsigned long chk1, chk2;
-	unsigned int wcount = 0xffff;
-	chk1 = inl(TRID_REG(card, ALI_STIMER));
-
-	while (1) {
-		chk2 = inl(TRID_REG(card, ALI_STIMER));
-		if ((wcount > 0) && chk1 != chk2)
-			return 1;
-		if (wcount <= 0)
-			break;
-		udelay(50);
-	}
-	return 0;
-}
-
-/* Read AC97 codec registers for ALi*/
-static u16
-ali_ac97_get(struct trident_card *card, int secondary, u8 reg)
-{
-	unsigned int address, mask;
-	unsigned int ncount;
-	unsigned long aud_reg;
-	u32 data;
-	u16 wcontrol;
-	unsigned long flags;
-
-	BUG_ON(!card);
-
-	address = ALI_AC97_READ;
-	if (card->revision == ALI_5451_V02) {
-		address = ALI_AC97_WRITE;
-	}
-	mask = ALI_AC97_READ_ACTION | ALI_AC97_AUDIO_BUSY;
-	if (secondary)
-		mask |= ALI_AC97_SECONDARY;
-
-	spin_lock_irqsave(&card->lock, flags);
-
-	if (!acquirecodecaccess(card))
-		printk(KERN_ERR "access codec fail\n");
-
-	wcontrol = inw(TRID_REG(card, ALI_AC97_WRITE));
-	wcontrol &= 0xfe00;
-	wcontrol |= (0x8000 | reg);
-	outw(wcontrol, TRID_REG(card, ALI_AC97_WRITE));
-
-	data = (mask | (reg & AC97_REG_ADDR));
-
-	if (!waitforstimertick(card)) {
-		printk(KERN_ERR "ali_ac97_read: BIT_CLOCK is dead\n");
-		goto releasecodec;
-	}
-
-	udelay(20);
-
-	ncount = 10;
-
-	while (1) {
-		if ((inw(TRID_REG(card, ALI_AC97_WRITE)) & ALI_AC97_BUSY_READ)
-		    != 0)
-			break;
-		if (ncount <= 0)
-			break;
-		if (ncount-- == 1) {
-			pr_debug("ali_ac97_read :try clear busy flag\n");
-			aud_reg = inl(TRID_REG(card, ALI_AC97_WRITE));
-			outl((aud_reg & 0xffff7fff),
-			     TRID_REG(card, ALI_AC97_WRITE));
-		}
-		udelay(10);
-	}
-
-	data = inl(TRID_REG(card, address));
-
-	spin_unlock_irqrestore(&card->lock, flags);
-
-	return ((u16) (data >> 16));
-
-      releasecodec:
-	releasecodecaccess(card);
-	spin_unlock_irqrestore(&card->lock, flags);
-	printk(KERN_ERR "ali_ac97_read: AC97 CODEC read timed out.\n");
-	return 0;
-}
-
-/* Write AC97 codec registers for hulei*/
-static void
-ali_ac97_set(struct trident_card *card, int secondary, u8 reg, u16 val)
-{
-	unsigned int address, mask;
-	unsigned int ncount;
-	u32 data;
-	u16 wcontrol;
-	unsigned long flags;
-
-	data = ((u32) val) << 16;
-
-	BUG_ON(!card);
-
-	address = ALI_AC97_WRITE;
-	mask = ALI_AC97_WRITE_ACTION | ALI_AC97_AUDIO_BUSY;
-	if (secondary)
-		mask |= ALI_AC97_SECONDARY;
-	if (card->revision == ALI_5451_V02)
-		mask |= ALI_AC97_WRITE_MIXER_REGISTER;
-
-	spin_lock_irqsave(&card->lock, flags);
-	if (!acquirecodecaccess(card))
-		printk(KERN_ERR "ali_ac97_write: access codec fail\n");
-
-	wcontrol = inw(TRID_REG(card, ALI_AC97_WRITE));
-	wcontrol &= 0xff00;
-	wcontrol |= (0x8100 | reg); /* bit 8=1: (ali1535 )reserved/ */
-	                            /* ali1535+ write */
-	outl((data | wcontrol), TRID_REG(card, ALI_AC97_WRITE));
-
-	if (!waitforstimertick(card)) {
-		printk(KERN_ERR "BIT_CLOCK is dead\n");
-		goto releasecodec;
-	}
-
-	ncount = 10;
-	while (1) {
-		wcontrol = inw(TRID_REG(card, ALI_AC97_WRITE));
-		if (!(wcontrol & 0x8000))
-			break;
-		if (ncount <= 0)
-			break;
-		if (ncount-- == 1) {
-			pr_debug("ali_ac97_set :try clear busy flag!!\n");
-			outw(wcontrol & 0x7fff,
-			     TRID_REG(card, ALI_AC97_WRITE));
-		}
-		udelay(10);
-	}
-
-      releasecodec:
-	releasecodecaccess(card);
-	spin_unlock_irqrestore(&card->lock, flags);
-	return;
-}
-
-static void
-ali_enable_special_channel(struct trident_state *stat)
-{
-	struct trident_card *card = stat->card;
-	unsigned long s_channels;
-
-	s_channels = inl(TRID_REG(card, ALI_GLOBAL_CONTROL));
-	s_channels |= (1 << stat->dmabuf.channel->num);
-	outl(s_channels, TRID_REG(card, ALI_GLOBAL_CONTROL));
-}
-
-static u16
-ali_ac97_read(struct ac97_codec *codec, u8 reg)
-{
-	int id;
-	u16 data;
-	struct trident_card *card = NULL;
-
-	/* Added by Matt Wu */
-	BUG_ON(!codec);
-
-	card = (struct trident_card *) codec->private_data;
-
-	if (!card->mixer_regs_ready)
-		return ali_ac97_get(card, codec->id, reg);
-
-	/*
-	 *      FIXME: need to stop this caching some registers
-	 */
-	if (codec->id)
-		id = 1;
-	else
-		id = 0;
-
-	data = card->mixer_regs[reg / 2][id];
-	return data;
-}
-
-static void
-ali_ac97_write(struct ac97_codec *codec, u8 reg, u16 val)
-{
-	int id;
-	struct trident_card *card;
-
-	/*  Added by Matt Wu */
-	BUG_ON(!codec);
-
-	card = (struct trident_card *) codec->private_data;
-
-	if (!card->mixer_regs_ready) {
-		ali_ac97_set(card, codec->id, reg, val);
-		return;
-	}
-
-	if (codec->id)
-		id = 1;
-	else
-		id = 0;
-
-	card->mixer_regs[reg / 2][id] = val;
-	ali_ac97_set(card, codec->id, reg, val);
-}
-
-/*
-flag:	ALI_SPDIF_OUT_TO_SPDIF_OUT
-	ALI_PCM_TO_SPDIF_OUT
-*/
-
-static void
-ali_setup_spdif_out(struct trident_card *card, int flag)
-{
-	unsigned long spdif;
-	unsigned char ch;
-
-	char temp;
-	struct pci_dev *pci_dev = NULL;
-
-	pci_dev = pci_get_device(PCI_VENDOR_ID_AL, PCI_DEVICE_ID_AL_M1533,
-				 pci_dev);
-	if (pci_dev == NULL)
-		return;
-	pci_read_config_byte(pci_dev, 0x61, &temp);
-	temp |= 0x40;
-	pci_write_config_byte(pci_dev, 0x61, temp);
-	pci_read_config_byte(pci_dev, 0x7d, &temp);
-	temp |= 0x01;
-	pci_write_config_byte(pci_dev, 0x7d, temp);
-	pci_read_config_byte(pci_dev, 0x7e, &temp);
-	temp &= (~0x20);
-	temp |= 0x10;
-	pci_write_config_byte(pci_dev, 0x7e, temp);
-
-	pci_dev_put(pci_dev);
-
-	ch = inb(TRID_REG(card, ALI_SCTRL));
-	outb(ch | ALI_SPDIF_OUT_ENABLE, TRID_REG(card, ALI_SCTRL));
-	ch = inb(TRID_REG(card, ALI_SPDIF_CTRL));
-	outb(ch & ALI_SPDIF_OUT_CH_STATUS, TRID_REG(card, ALI_SPDIF_CTRL));
-
-	if (flag & ALI_SPDIF_OUT_TO_SPDIF_OUT) {
-		spdif = inw(TRID_REG(card, ALI_GLOBAL_CONTROL));
-		spdif |= ALI_SPDIF_OUT_CH_ENABLE;
-		spdif &= ALI_SPDIF_OUT_SEL_SPDIF;
-		outw(spdif, TRID_REG(card, ALI_GLOBAL_CONTROL));
-		spdif = inw(TRID_REG(card, ALI_SPDIF_CS));
-		if (flag & ALI_SPDIF_OUT_NON_PCM)
-			spdif |= 0x0002;
-		else
-			spdif &= (~0x0002);
-		outw(spdif, TRID_REG(card, ALI_SPDIF_CS));
-	} else {
-		spdif = inw(TRID_REG(card, ALI_GLOBAL_CONTROL));
-		spdif |= ALI_SPDIF_OUT_SEL_PCM;
-		outw(spdif, TRID_REG(card, ALI_GLOBAL_CONTROL));
-	}
-}
-
-static void
-ali_disable_special_channel(struct trident_card *card, int ch)
-{
-	unsigned long sc;
-
-	sc = inl(TRID_REG(card, ALI_GLOBAL_CONTROL));
-	sc &= ~(1 << ch);
-	outl(sc, TRID_REG(card, ALI_GLOBAL_CONTROL));
-}
-
-static void
-ali_disable_spdif_in(struct trident_card *card)
-{
-	unsigned long spdif;
-
-	spdif = inl(TRID_REG(card, ALI_GLOBAL_CONTROL));
-	spdif &= (~ALI_SPDIF_IN_SUPPORT);
-	outl(spdif, TRID_REG(card, ALI_GLOBAL_CONTROL));
-
-	ali_disable_special_channel(card, ALI_SPDIF_IN_CHANNEL);
-}
-
-static void
-ali_setup_spdif_in(struct trident_card *card)
-{
-	unsigned long spdif;
-
-	//Set SPDIF IN Supported
-	spdif = inl(TRID_REG(card, ALI_GLOBAL_CONTROL));
-	spdif |= ALI_SPDIF_IN_SUPPORT;
-	outl(spdif, TRID_REG(card, ALI_GLOBAL_CONTROL));
-
-	//Set SPDIF IN Rec
-	spdif = inl(TRID_REG(card, ALI_GLOBAL_CONTROL));
-	spdif |= ALI_SPDIF_IN_CH_ENABLE;
-	outl(spdif, TRID_REG(card, ALI_GLOBAL_CONTROL));
-
-	spdif = inb(TRID_REG(card, ALI_SPDIF_CTRL));
-	spdif |= ALI_SPDIF_IN_CH_STATUS;
-	outb(spdif, TRID_REG(card, ALI_SPDIF_CTRL));
-/*
-	spdif = inb(TRID_REG(card, ALI_SPDIF_CTRL));
-	spdif |= ALI_SPDIF_IN_FUNC_ENABLE;
-	outb(spdif, TRID_REG(card, ALI_SPDIF_CTRL));
-*/
-}
-
-static void
-ali_delay(struct trident_card *card, int interval)
-{
-	unsigned long begintimer, currenttimer;
-
-	begintimer = inl(TRID_REG(card, ALI_STIMER));
-	currenttimer = inl(TRID_REG(card, ALI_STIMER));
-
-	while (currenttimer < begintimer + interval)
-		currenttimer = inl(TRID_REG(card, ALI_STIMER));
-}
-
-static void
-ali_detect_spdif_rate(struct trident_card *card)
-{
-	u16 wval = 0;
-	u16 count = 0;
-	u8 bval = 0, R1 = 0, R2 = 0;
-
-	bval = inb(TRID_REG(card, ALI_SPDIF_CTRL));
-	bval |= 0x02;
-	outb(bval, TRID_REG(card, ALI_SPDIF_CTRL));
-
-	bval = inb(TRID_REG(card, ALI_SPDIF_CTRL + 1));
-	bval |= 0x1F;
-	outb(bval, TRID_REG(card, ALI_SPDIF_CTRL + 1));
-
-	while (((R1 < 0x0B) || (R1 > 0x0E)) && (R1 != 0x12) &&
-	       count <= 50000) {
-		count++;
-
-		ali_delay(card, 6);
-
-		bval = inb(TRID_REG(card, ALI_SPDIF_CTRL + 1));
-		R1 = bval & 0x1F;
-	}
-
-	if (count > 50000) {
-		printk(KERN_WARNING "trident: Error in "
-		       "ali_detect_spdif_rate!\n");
-		return;
-	}
-
-	count = 0;
-
-	while (count <= 50000) {
-		count++;
-
-		ali_delay(card, 6);
-
-		bval = inb(TRID_REG(card, ALI_SPDIF_CTRL + 1));
-		R2 = bval & 0x1F;
-
-		if (R2 != R1)
-			R1 = R2;
-		else
-			break;
-	}
-
-	if (count > 50000) {
-		printk(KERN_WARNING "trident: Error in "
-		       "ali_detect_spdif_rate!\n");
-		return;
-	}
-
-	switch (R2) {
-	case 0x0b:
-	case 0x0c:
-	case 0x0d:
-	case 0x0e:
-		wval = inw(TRID_REG(card, ALI_SPDIF_CTRL + 2));
-		wval &= 0xE0F0;
-		wval |= (u16) 0x09 << 8 | (u16) 0x05;
-		outw(wval, TRID_REG(card, ALI_SPDIF_CTRL + 2));
-
-		bval = inb(TRID_REG(card, ALI_SPDIF_CS + 3)) & 0xF0;
-		outb(bval | 0x02, TRID_REG(card, ALI_SPDIF_CS + 3));
-		break;
-
-	case 0x12:
-		wval = inw(TRID_REG(card, ALI_SPDIF_CTRL + 2));
-		wval &= 0xE0F0;
-		wval |= (u16) 0x0E << 8 | (u16) 0x08;
-		outw(wval, TRID_REG(card, ALI_SPDIF_CTRL + 2));
-
-		bval = inb(TRID_REG(card, ALI_SPDIF_CS + 3)) & 0xF0;
-		outb(bval | 0x03, TRID_REG(card, ALI_SPDIF_CS + 3));
-		break;
-
-	default:
-		break;
-	}
-
-}
-
-static unsigned int
-ali_get_spdif_in_rate(struct trident_card *card)
-{
-	u32 dwRate = 0;
-	u8 bval = 0;
-
-	ali_detect_spdif_rate(card);
-
-	bval = inb(TRID_REG(card, ALI_SPDIF_CTRL));
-	bval &= 0x7F;
-	bval |= 0x40;
-	outb(bval, TRID_REG(card, ALI_SPDIF_CTRL));
-
-	bval = inb(TRID_REG(card, ALI_SPDIF_CS + 3));
-	bval &= 0x0F;
-
-	switch (bval) {
-	case 0:
-		dwRate = 44100;
-		break;
-	case 1:
-		dwRate = 48000;
-		break;
-	case 2:
-		dwRate = 32000;
-		break;
-	default:
-		// Error occurs
-		break;
-	}
-
-	return dwRate;
-
-}
-
-static int
-ali_close_multi_channels(void)
-{
-	char temp = 0;
-	struct pci_dev *pci_dev = NULL;
-
-	pci_dev = pci_get_device(PCI_VENDOR_ID_AL, PCI_DEVICE_ID_AL_M1533,
-				 pci_dev);
-	if (pci_dev == NULL)
-		return -1;
-
-	pci_read_config_byte(pci_dev, 0x59, &temp);
-	temp &= ~0x80;
-	pci_write_config_byte(pci_dev, 0x59, temp);
-
-	pci_dev_put(pci_dev);
-
-	pci_dev = pci_get_device(PCI_VENDOR_ID_AL, PCI_DEVICE_ID_AL_M7101,
-				 NULL);
-	if (pci_dev == NULL)
-		return -1;
-
-	pci_read_config_byte(pci_dev, 0xB8, &temp);
-	temp &= ~0x20;
-	pci_write_config_byte(pci_dev, 0xB8, temp);
-
-	pci_dev_put(pci_dev);
-
-	return 0;
-}
-
-static int
-ali_setup_multi_channels(struct trident_card *card, int chan_nums)
-{
-	unsigned long dwValue;
-	char temp = 0;
-	struct pci_dev *pci_dev = NULL;
-
-	pci_dev = pci_get_device(PCI_VENDOR_ID_AL, PCI_DEVICE_ID_AL_M1533,
-				 pci_dev);
-	if (pci_dev == NULL)
-		return -1;
-	pci_read_config_byte(pci_dev, 0x59, &temp);
-	temp |= 0x80;
-	pci_write_config_byte(pci_dev, 0x59, temp);
-
-	pci_dev_put(pci_dev);
-
-	pci_dev = pci_get_device(PCI_VENDOR_ID_AL, PCI_DEVICE_ID_AL_M7101,
-				 NULL);
-	if (pci_dev == NULL)
-		return -1;
-	pci_read_config_byte(pci_dev, (int) 0xB8, &temp);
-	temp |= 0x20;
-	pci_write_config_byte(pci_dev, (int) 0xB8, (u8) temp);
-
-	pci_dev_put(pci_dev);
-
-	if (chan_nums == 6) {
-		dwValue = inl(TRID_REG(card, ALI_SCTRL)) | 0x000f0000;
-		outl(dwValue, TRID_REG(card, ALI_SCTRL));
-		mdelay(4);
-		dwValue = inl(TRID_REG(card, ALI_SCTRL));
-		if (dwValue & 0x2000000) {
-			ali_ac97_write(card->ac97_codec[0], 0x02, 8080);
-			ali_ac97_write(card->ac97_codec[0], 0x36, 0);
-			ali_ac97_write(card->ac97_codec[0], 0x38, 0);
-			/*
-			 *      On a board with a single codec you won't get the
-			 *      surround. On other boards configure it.
-			 */
-			if (card->ac97_codec[1] != NULL) {
-				ali_ac97_write(card->ac97_codec[1], 0x36, 0);
-				ali_ac97_write(card->ac97_codec[1], 0x38, 0);
-				ali_ac97_write(card->ac97_codec[1], 0x02, 0x0606);
-				ali_ac97_write(card->ac97_codec[1], 0x18, 0x0303);
-				ali_ac97_write(card->ac97_codec[1], 0x74, 0x3);
-			}
-			return 1;
-		}
-	}
-	return -EINVAL;
-}
-
-static void
-ali_free_pcm_channel(struct trident_card *card, unsigned int channel)
-{
-	int bank;
-
-	if (channel > 31)
-		return;
-
-	bank = channel >> 5;
-	channel = channel & 0x1f;
-
-	card->banks[bank].bitmap &= ~(1 << (channel));
-}
-
-static int
-ali_allocate_other_states_resources(struct trident_state *state, int chan_nums)
-{
-	struct trident_card *card = state->card;
-	struct trident_state *s;
-	int i, state_count = 0;
-	struct trident_pcm_bank *bank;
-	struct trident_channel *channel;
-	unsigned long num;
-
-	bank = &card->banks[BANK_A];
-
-	if (chan_nums != 6)
-		return 0;
-
-	for (i = 0; (i < ALI_CHANNELS) && (state_count != 4); i++) {
-		if (card->states[i])
-			continue;
-
-		num = ali_multi_channels_5_1[state_count];
-		if (!(bank->bitmap & (1 << num))) {
-			bank->bitmap |= 1 << num;
-			channel = &bank->channels[num];
-			channel->num = num;
-		} else {
-			state_count--;
-			for (; state_count >= 0; state_count--) {
-				kfree(state->other_states[state_count]);
-				num = ali_multi_channels_5_1[state_count];
-					ali_free_pcm_channel(card, num);
-			}
-			return -EBUSY;
-		}
-		s = card->states[i] = kzalloc(sizeof(*state), GFP_KERNEL);
-		if (!s) {
-			num = ali_multi_channels_5_1[state_count];
-			ali_free_pcm_channel(card, num);
-			state_count--;
-			for (; state_count >= 0; state_count--) {
-				num = ali_multi_channels_5_1[state_count];
-				ali_free_pcm_channel(card, num);
-				kfree(state->other_states[state_count]);
-			}
-			return -ENOMEM;
-		}
-
-		s->dmabuf.channel = channel;
-		s->dmabuf.ossfragshift = s->dmabuf.ossmaxfrags =
-			s->dmabuf.subdivision = 0;
-		init_waitqueue_head(&s->dmabuf.wait);
-		s->magic = card->magic;
-		s->card = card;
-		s->virt = i;
-		ali_enable_special_channel(s);
-		state->other_states[state_count++] = s;
-	}
-
-	if (state_count != 4) {
-		state_count--;
-		for (; state_count >= 0; state_count--) {
-			kfree(state->other_states[state_count]);
-			num = ali_multi_channels_5_1[state_count];
-			ali_free_pcm_channel(card, num);
-		}
-		return -EBUSY;
-	}
-	return 0;
-}
-
-#ifdef CONFIG_PM
-/* save registers for ALi Power Management */
-static struct ali_saved_registers {
-	unsigned long global_regs[ALI_GLOBAL_REGS];
-	unsigned long channel_regs[ALI_CHANNELS][ALI_CHANNEL_REGS];
-	unsigned mixer_regs[ALI_MIXER_REGS];
-} ali_registers;
-
-static void
-ali_save_regs(struct trident_card *card)
-{
-	unsigned long flags;
-	int i, j;
-
-	spin_lock_irqsave(&card->lock, flags);
-
-	ali_registers.global_regs[0x2c] = inl(TRID_REG(card, T4D_MISCINT));
-	//ali_registers.global_regs[0x20] = inl(TRID_REG(card,T4D_START_A));
-	ali_registers.global_regs[0x21] = inl(TRID_REG(card, T4D_STOP_A));
-
-	//disable all IRQ bits
-	outl(ALI_DISABLE_ALL_IRQ, TRID_REG(card, T4D_MISCINT));
-
-	for (i = 1; i < ALI_MIXER_REGS; i++)
-		ali_registers.mixer_regs[i] = ali_ac97_read(card->ac97_codec[0],
-							    i * 2);
-
-	for (i = 0; i < ALI_GLOBAL_REGS; i++) {
-		if ((i * 4 == T4D_MISCINT) || (i * 4 == T4D_STOP_A))
-			continue;
-		ali_registers.global_regs[i] = inl(TRID_REG(card, i * 4));
-	}
-
-	for (i = 0; i < ALI_CHANNELS; i++) {
-		outb(i, TRID_REG(card, T4D_LFO_GC_CIR));
-		for (j = 0; j < ALI_CHANNEL_REGS; j++)
-			ali_registers.channel_regs[i][j] = inl(TRID_REG(card,
-									j * 4 + 0xe0));
-	}
-
-	//Stop all HW channel
-	outl(ALI_STOP_ALL_CHANNELS, TRID_REG(card, T4D_STOP_A));
-
-	spin_unlock_irqrestore(&card->lock, flags);
-}
-
-static void
-ali_restore_regs(struct trident_card *card)
-{
-	unsigned long flags;
-	int i, j;
-
-	spin_lock_irqsave(&card->lock, flags);
-
-	for (i = 1; i < ALI_MIXER_REGS; i++)
-		ali_ac97_write(card->ac97_codec[0], i * 2,
-			       ali_registers.mixer_regs[i]);
-
-	for (i = 0; i < ALI_CHANNELS; i++) {
-		outb(i, TRID_REG(card, T4D_LFO_GC_CIR));
-		for (j = 0; j < ALI_CHANNEL_REGS; j++)
-			outl(ali_registers.channel_regs[i][j],
-			     TRID_REG(card, j * 4 + 0xe0));
-	}
-
-	for (i = 0; i < ALI_GLOBAL_REGS; i++) {
-		if ((i * 4 == T4D_MISCINT) || (i * 4 == T4D_STOP_A) ||
-		    (i * 4 == T4D_START_A))
-			continue;
-		outl(ali_registers.global_regs[i], TRID_REG(card, i * 4));
-	}
-
-	//start HW channel
-	outl(ali_registers.global_regs[0x20], TRID_REG(card, T4D_START_A));
-	//restore IRQ enable bits
-	outl(ali_registers.global_regs[0x2c], TRID_REG(card, T4D_MISCINT));
-
-	spin_unlock_irqrestore(&card->lock, flags);
-}
-
-static int
-trident_suspend(struct pci_dev *dev, pm_message_t unused)
-{
-	struct trident_card *card = pci_get_drvdata(dev);
-
-	if (card->pci_id == PCI_DEVICE_ID_ALI_5451) {
-		ali_save_regs(card);
-	}
-	return 0;
-}
-
-static int
-trident_resume(struct pci_dev *dev)
-{
-	struct trident_card *card = pci_get_drvdata(dev);
-
-	if (card->pci_id == PCI_DEVICE_ID_ALI_5451) {
-		ali_restore_regs(card);
-	}
-	return 0;
-}
-#endif
-
-static struct trident_channel *
-ali_alloc_pcm_channel(struct trident_card *card)
-{
-	struct trident_pcm_bank *bank;
-	int idx;
-
-	bank = &card->banks[BANK_A];
-
-	if (inl(TRID_REG(card, ALI_GLOBAL_CONTROL)) &
-	    (ALI_SPDIF_OUT_CH_ENABLE)) {
-		idx = ALI_SPDIF_OUT_CHANNEL;
-		if (!(bank->bitmap & (1 << idx))) {
-			struct trident_channel *channel = &bank->channels[idx];
-			bank->bitmap |= 1 << idx;
-			channel->num = idx;
-			return channel;
-		}
-	}
-
-	for (idx = ALI_PCM_OUT_CHANNEL_FIRST; idx <= ALI_PCM_OUT_CHANNEL_LAST;
-	     idx++) {
-		if (!(bank->bitmap & (1 << idx))) {
-			struct trident_channel *channel = &bank->channels[idx];
-			bank->bitmap |= 1 << idx;
-			channel->num = idx;
-			return channel;
-		}
-	}
-
-	/* no more free channels avaliable */
-#if 0
-	printk(KERN_ERR "ali: no more channels available on Bank A.\n");
-#endif /* 0 */
-	return NULL;
-}
-
-static struct trident_channel *
-ali_alloc_rec_pcm_channel(struct trident_card *card)
-{
-	struct trident_pcm_bank *bank;
-	int idx;
-
-	if (inl(TRID_REG(card, ALI_GLOBAL_CONTROL)) & ALI_SPDIF_IN_SUPPORT)
-		idx = ALI_SPDIF_IN_CHANNEL;
-	else
-		idx = ALI_PCM_IN_CHANNEL;
-
-	bank = &card->banks[BANK_A];
-
-	if (!(bank->bitmap & (1 << idx))) {
-		struct trident_channel *channel = &bank->channels[idx];
-		bank->bitmap |= 1 << idx;
-		channel->num = idx;
-		return channel;
-	}
-
-	/* no free recordable channels avaliable */
-#if 0
-	printk(KERN_ERR "ali: no recordable channels available on Bank A.\n");
-#endif /* 0 */
-	return NULL;
-}
-
-static void
-ali_set_spdif_out_rate(struct trident_card *card, unsigned int rate)
-{
-	unsigned char ch_st_sel;
-	unsigned short status_rate;
-
-	switch (rate) {
-	case 44100:
-		status_rate = 0;
-		break;
-	case 32000:
-		status_rate = 0x300;
-		break;
-	case 48000:
-	default:
-		status_rate = 0x200;
-		break;
-	}
-
-	/* select spdif_out */
-	ch_st_sel = inb(TRID_REG(card, ALI_SPDIF_CTRL)) & ALI_SPDIF_OUT_CH_STATUS;
-
-	ch_st_sel |= 0x80;	/* select right */
-	outb(ch_st_sel, TRID_REG(card, ALI_SPDIF_CTRL));
-	outb(status_rate | 0x20, TRID_REG(card, ALI_SPDIF_CS + 2));
-
-	ch_st_sel &= (~0x80);	/* select left */
-	outb(ch_st_sel, TRID_REG(card, ALI_SPDIF_CTRL));
-	outw(status_rate | 0x10, TRID_REG(card, ALI_SPDIF_CS + 2));
-}
-
-static void
-ali_address_interrupt(struct trident_card *card)
-{
-	int i, channel;
-	struct trident_state *state;
-	u32 mask, channel_mask;
-
-	mask = trident_get_interrupt_mask(card, 0);
-	for (i = 0; i < NR_HW_CH; i++) {
-		if ((state = card->states[i]) == NULL)
-			continue;
-		channel = state->dmabuf.channel->num;
-		if ((channel_mask = 1 << channel) & mask) {
-			mask &= ~channel_mask;
-			trident_ack_channel_interrupt(card, channel);
-			udelay(100);
-			state->dmabuf.update_flag |= ALI_ADDRESS_INT_UPDATE;
-			trident_update_ptr(state);
-		}
-	}
-	if (mask) {
-		for (i = 0; i < NR_HW_CH; i++) {
-			if (mask & (1 << i)) {
-				printk("ali: spurious channel irq %d.\n", i);
-				trident_ack_channel_interrupt(card, i);
-				trident_stop_voice(card, i);
-				trident_disable_voice_irq(card, i);
-			}
-		}
-	}
-}
-
-/* Updating the values of counters of other_states' DMAs without lock
-protection is no harm because all DMAs of multi-channels and interrupt
-depend on a master state's DMA, and changing the counters of the master
-state DMA is protected by a spinlock.
-*/
-static int
-ali_write_5_1(struct trident_state *state, const char __user *buf,
-	      int cnt_for_multi_channel, unsigned int *copy_count,
-	      unsigned int *state_cnt)
-{
-
-	struct dmabuf *dmabuf = &state->dmabuf;
-	struct dmabuf *dmabuf_temp;
-	const char __user *buffer = buf;
-	unsigned swptr, other_dma_nums, sample_s;
-	unsigned int i, loop;
-
-	other_dma_nums = 4;
-	sample_s = sample_size[dmabuf->fmt] >> 1;
-	swptr = dmabuf->swptr;
-
-	if ((i = state->multi_channels_adjust_count) > 0) {
-		if (i == 1) {
-			if (copy_from_user(dmabuf->rawbuf + swptr,
-					   buffer, sample_s))
-				return -EFAULT;
-			seek_offset(swptr, buffer, cnt_for_multi_channel,
-				    sample_s, *copy_count);
-			i--;
-			(*state_cnt) += sample_s;
-			state->multi_channels_adjust_count++;
-		} else
-			i = i - (state->chans_num - other_dma_nums);
-		for (; (i < other_dma_nums) && (cnt_for_multi_channel > 0); i++) {
-			dmabuf_temp = &state->other_states[i]->dmabuf;
-			if (copy_from_user(dmabuf_temp->rawbuf + dmabuf_temp->swptr,
-					   buffer, sample_s))
-				return -EFAULT;
-			seek_offset(dmabuf_temp->swptr, buffer, cnt_for_multi_channel,
-				    sample_s, *copy_count);
-		}
-		if (cnt_for_multi_channel == 0)
-			state->multi_channels_adjust_count += i;
-	}
-	if (cnt_for_multi_channel > 0) {
-		loop = cnt_for_multi_channel / (state->chans_num * sample_s);
-		for (i = 0; i < loop; i++) {
-			if (copy_from_user(dmabuf->rawbuf + swptr, buffer,
-					   sample_s * 2))
-				return -EFAULT;
-			seek_offset(swptr, buffer, cnt_for_multi_channel,
-				    sample_s * 2, *copy_count);
-			(*state_cnt) += (sample_s * 2);
-
-			dmabuf_temp = &state->other_states[0]->dmabuf;
-			if (copy_from_user(dmabuf_temp->rawbuf + dmabuf_temp->swptr,
-					   buffer, sample_s))
-				return -EFAULT;
-			seek_offset(dmabuf_temp->swptr, buffer, cnt_for_multi_channel,
-				    sample_s, *copy_count);
-
-			dmabuf_temp = &state->other_states[1]->dmabuf;
-			if (copy_from_user(dmabuf_temp->rawbuf + dmabuf_temp->swptr,
-					   buffer, sample_s))
-				return -EFAULT;
-			seek_offset(dmabuf_temp->swptr, buffer, cnt_for_multi_channel,
-				    sample_s, *copy_count);
-
-			dmabuf_temp = &state->other_states[2]->dmabuf;
-			if (copy_from_user(dmabuf_temp->rawbuf + dmabuf_temp->swptr,
-					   buffer, sample_s))
-				return -EFAULT;
-			seek_offset(dmabuf_temp->swptr, buffer, cnt_for_multi_channel,
-				    sample_s, *copy_count);
-
-			dmabuf_temp = &state->other_states[3]->dmabuf;
-			if (copy_from_user(dmabuf_temp->rawbuf + dmabuf_temp->swptr,
-					   buffer, sample_s))
-				return -EFAULT;
-			seek_offset(dmabuf_temp->swptr, buffer, cnt_for_multi_channel,
-				    sample_s, *copy_count);
-		}
-
-		if (cnt_for_multi_channel > 0) {
-			state->multi_channels_adjust_count = cnt_for_multi_channel / sample_s;
-
-			if (copy_from_user(dmabuf->rawbuf + swptr, buffer, sample_s))
-				return -EFAULT;
-			seek_offset(swptr, buffer, cnt_for_multi_channel,
-				    sample_s, *copy_count);
-			(*state_cnt) += sample_s;
-
-			if (cnt_for_multi_channel > 0) {
-				if (copy_from_user(dmabuf->rawbuf + swptr,
-						   buffer, sample_s))
-					return -EFAULT;
-				seek_offset(swptr, buffer, cnt_for_multi_channel,
-					    sample_s, *copy_count);
-				(*state_cnt) += sample_s;
-
-				if (cnt_for_multi_channel > 0) {
-					int diff = state->chans_num - other_dma_nums;
-					loop = state->multi_channels_adjust_count - diff;
-					for (i = 0; i < loop; i++) {
-						dmabuf_temp = &state->other_states[i]->dmabuf;
-						if (copy_from_user(dmabuf_temp->rawbuf +
-								   dmabuf_temp->swptr,
-								   buffer, sample_s))
-							return -EFAULT;
-						seek_offset(dmabuf_temp->swptr, buffer,
-							    cnt_for_multi_channel,
-							    sample_s, *copy_count);
-					}
-				}
-			}
-		} else
-			state->multi_channels_adjust_count = 0;
-	}
-	for (i = 0; i < other_dma_nums; i++) {
-		dmabuf_temp = &state->other_states[i]->dmabuf;
-		dmabuf_temp->swptr = dmabuf_temp->swptr % dmabuf_temp->dmasize;
-	}
-	return *state_cnt;
-}
-
-static void
-ali_free_other_states_resources(struct trident_state *state)
-{
-	int i;
-	struct trident_card *card = state->card;
-	struct trident_state *s;
-	unsigned other_states_count;
-
-	other_states_count = state->chans_num - 2; /* except PCM L/R channels */
-	for (i = 0; i < other_states_count; i++) {
-		s = state->other_states[i];
-		dealloc_dmabuf(&s->dmabuf, card->pci_dev);
-		ali_disable_special_channel(s->card, s->dmabuf.channel->num);
-		state->card->free_pcm_channel(s->card, s->dmabuf.channel->num);
-		card->states[s->virt] = NULL;
-		kfree(s);
-	}
-}
-
-static struct proc_dir_entry *res;
-
-static int
-ali_write_proc(struct file *file, const char __user *buffer, unsigned long count, void *data)
-{
-	struct trident_card *card = (struct trident_card *) data;
-	unsigned long flags;
-	char c;
-
-	if (count < 0)
-		return -EINVAL;
-	if (count == 0)
-		return 0;
-	if (get_user(c, buffer))
-		return -EFAULT;
-
-	spin_lock_irqsave(&card->lock, flags);
-	switch (c) {
-	case '0':
-		ali_setup_spdif_out(card, ALI_PCM_TO_SPDIF_OUT);
-		ali_disable_special_channel(card, ALI_SPDIF_OUT_CHANNEL);
-		break;
-	case '1':
-		ali_setup_spdif_out(card, ALI_SPDIF_OUT_TO_SPDIF_OUT |
-				    ALI_SPDIF_OUT_PCM);
-		break;
-	case '2':
-		ali_setup_spdif_out(card, ALI_SPDIF_OUT_TO_SPDIF_OUT |
-				    ALI_SPDIF_OUT_NON_PCM);
-		break;
-	case '3':
-		ali_disable_spdif_in(card);	//default
-		break;
-	case '4':
-		ali_setup_spdif_in(card);
-		break;
-	}
-	spin_unlock_irqrestore(&card->lock, flags);
-
-	return count;
-}
-
-/* OSS /dev/mixer file operation methods */
-static int
-trident_open_mixdev(struct inode *inode, struct file *file)
-{
-	int i = 0;
-	int minor = iminor(inode);
-	struct trident_card *card = devs;
-
-	for (card = devs; card != NULL; card = card->next)
-		for (i = 0; i < NR_AC97; i++)
-			if (card->ac97_codec[i] != NULL &&
-			    card->ac97_codec[i]->dev_mixer == minor)
-				goto match;
-
-	if (!card) {
-		return -ENODEV;
-	}
-      match:
-	file->private_data = card->ac97_codec[i];
-
-	return nonseekable_open(inode, file);
-}
-
-static int
-trident_ioctl_mixdev(struct inode *inode, struct file *file, unsigned int cmd,
-		     unsigned long arg)
-{
-	struct ac97_codec *codec = (struct ac97_codec *) file->private_data;
-
-	return codec->mixer_ioctl(codec, cmd, arg);
-}
-
-static const struct file_operations trident_mixer_fops = {
-	.owner = THIS_MODULE,
-	.llseek = no_llseek,
-	.ioctl = trident_ioctl_mixdev,
-	.open = trident_open_mixdev,
-};
-
-static int
-ali_reset_5451(struct trident_card *card)
-{
-	struct pci_dev *pci_dev = NULL;
-	unsigned int dwVal;
-	unsigned short wCount, wReg;
-
-	pci_dev = pci_get_device(PCI_VENDOR_ID_AL, PCI_DEVICE_ID_AL_M1533,
-				 pci_dev);
-	if (pci_dev == NULL)
-		return -1;
-
-	pci_read_config_dword(pci_dev, 0x7c, &dwVal);
-	pci_write_config_dword(pci_dev, 0x7c, dwVal | 0x08000000);
-	udelay(5000);
-	pci_read_config_dword(pci_dev, 0x7c, &dwVal);
-	pci_write_config_dword(pci_dev, 0x7c, dwVal & 0xf7ffffff);
-	udelay(5000);
-	pci_dev_put(pci_dev);
-
-	pci_dev = card->pci_dev;
-	if (pci_dev == NULL)
-		return -1;
-
-	pci_read_config_dword(pci_dev, 0x44, &dwVal);
-	pci_write_config_dword(pci_dev, 0x44, dwVal | 0x000c0000);
-	udelay(500);
-	pci_read_config_dword(pci_dev, 0x44, &dwVal);
-	pci_write_config_dword(pci_dev, 0x44, dwVal & 0xfffbffff);
-	udelay(5000);
-
-	/* TODO: recognize if we have a PM capable codec and only do this */
-	/* if the codec is PM capable */
-	wCount = 2000;
-	while (wCount--) {
-		wReg = ali_ac97_get(card, 0, AC97_POWER_CONTROL);
-		if ((wReg & 0x000f) == 0x000f)
-			return 0;
-		udelay(5000);
-	}
-	/* This is non fatal if you have a non PM capable codec.. */
-	return 0;
-}
-
-/* AC97 codec initialisation. */
-static int __devinit
-trident_ac97_init(struct trident_card *card)
-{
-	int num_ac97 = 0;
-	unsigned long ready_2nd = 0;
-	struct ac97_codec *codec;
-	int i = 0;
-
-	/* initialize controller side of AC link, and find out if secondary codes
-	   really exist */
-	switch (card->pci_id) {
-	case PCI_DEVICE_ID_ALI_5451:
-		if (ali_reset_5451(card)) {
-			printk(KERN_ERR "trident_ac97_init: error "
-			       "resetting 5451.\n");
-			return -1;
-		}
-		outl(0x80000001, TRID_REG(card, ALI_GLOBAL_CONTROL));
-		outl(0x00000000, TRID_REG(card, T4D_AINTEN_A));
-		outl(0xffffffff, TRID_REG(card, T4D_AINT_A));
-		outl(0x00000000, TRID_REG(card, T4D_MUSICVOL_WAVEVOL));
-		outb(0x10, TRID_REG(card, ALI_MPUR2));
-		ready_2nd = inl(TRID_REG(card, ALI_SCTRL));
-		ready_2nd &= 0x3fff;
-		outl(ready_2nd | PCMOUT | 0x8000, TRID_REG(card, ALI_SCTRL));
-		ready_2nd = inl(TRID_REG(card, ALI_SCTRL));
-		ready_2nd &= SI_AC97_SECONDARY_READY;
-		if (card->revision < ALI_5451_V02)
-			ready_2nd = 0;
-		break;
-	case PCI_DEVICE_ID_SI_7018:
-		/* disable AC97 GPIO interrupt */
-		outl(0x00, TRID_REG(card, SI_AC97_GPIO));
-		/* when power up the AC link is in cold reset mode so stop it */
-		outl(PCMOUT | SURROUT | CENTEROUT | LFEOUT | SECONDARY_ID,
-		     TRID_REG(card, SI_SERIAL_INTF_CTRL));
-		/* it take a long time to recover from a cold reset */
-		/* (especially when you have more than one codec) */
-		udelay(2000);
-		ready_2nd = inl(TRID_REG(card, SI_SERIAL_INTF_CTRL));
-		ready_2nd &= SI_AC97_SECONDARY_READY;
-		break;
-	case PCI_DEVICE_ID_TRIDENT_4DWAVE_DX:
-		/* playback on */
-		outl(DX_AC97_PLAYBACK, TRID_REG(card, DX_ACR2_AC97_COM_STAT));
-		break;
-	case PCI_DEVICE_ID_TRIDENT_4DWAVE_NX:
-		/* enable AC97 Output Slot 3,4 (PCM Left/Right Playback) */
-		outl(NX_AC97_PCM_OUTPUT, TRID_REG(card, NX_ACR0_AC97_COM_STAT));
-		ready_2nd = inl(TRID_REG(card, NX_ACR0_AC97_COM_STAT));
-		ready_2nd &= NX_AC97_SECONDARY_READY;
-		break;
-	case PCI_DEVICE_ID_INTERG_5050:
-		/* disable AC97 GPIO interrupt */
-		outl(0x00, TRID_REG(card, SI_AC97_GPIO));
-		/* when power up, the AC link is in cold reset mode, so stop it */
-		outl(PCMOUT | SURROUT | CENTEROUT | LFEOUT,
-		     TRID_REG(card, SI_SERIAL_INTF_CTRL));
-		/* it take a long time to recover from a cold reset (especially */
-		/* when you have more than one codec) */
-		udelay(2000);
-		ready_2nd = inl(TRID_REG(card, SI_SERIAL_INTF_CTRL));
-		ready_2nd &= SI_AC97_SECONDARY_READY;
-		break;
-	}
-
-	for (num_ac97 = 0; num_ac97 < NR_AC97; num_ac97++) {
-		if ((codec = ac97_alloc_codec()) == NULL)
-			return -ENOMEM;
-
-		/* initialize some basic codec information, other fields */
-		/* will be filled in ac97_probe_codec */
-		codec->private_data = card;
-		codec->id = num_ac97;
-
-		if (card->pci_id == PCI_DEVICE_ID_ALI_5451) {
-			codec->codec_read = ali_ac97_read;
-			codec->codec_write = ali_ac97_write;
-		} else {
-			codec->codec_read = trident_ac97_get;
-			codec->codec_write = trident_ac97_set;
-		}
-
-		if (ac97_probe_codec(codec) == 0)
-			break;
-
-		codec->dev_mixer = register_sound_mixer(&trident_mixer_fops, -1);
-		if (codec->dev_mixer < 0) {
-			printk(KERN_ERR "trident: couldn't register mixer!\n");
-			ac97_release_codec(codec);
-			break;
-		}
-
-		card->ac97_codec[num_ac97] = codec;
-
-		/* if there is no secondary codec at all, don't probe any more */
-		if (!ready_2nd)
-			break;
-	}
-
-	if (card->pci_id == PCI_DEVICE_ID_ALI_5451) {
-		for (num_ac97 = 0; num_ac97 < NR_AC97; num_ac97++) {
-			if (card->ac97_codec[num_ac97] == NULL)
-				break;
-			for (i = 0; i < 64; i++) {
-				u16 reg = ali_ac97_get(card, num_ac97, i * 2);
-				card->mixer_regs[i][num_ac97] = reg;
-			}
-		}
-	}
-	return num_ac97 + 1;
-}
-
-#ifdef SUPPORT_JOYSTICK
-/* Gameport functions for the cards ADC gameport */
-
-static unsigned char trident_game_read(struct gameport *gameport)
-{
-	struct trident_card *card = gameport->port_data;
-
-	return inb(TRID_REG(card, T4D_GAME_LEG));
-}
-
-static void trident_game_trigger(struct gameport *gameport)
-{
-	struct trident_card *card = gameport->port_data;
-
-	outb(0xff, TRID_REG(card, T4D_GAME_LEG));
-}
-
-static int trident_game_cooked_read(struct gameport *gameport,
-				    int *axes, int *buttons)
-{
-	struct trident_card *card = gameport->port_data;
-	int i;
-
-	*buttons = (~inb(TRID_REG(card, T4D_GAME_LEG)) >> 4) & 0xf;
-
-	for (i = 0; i < 4; i++) {
-		axes[i] = inw(TRID_REG(card, T4D_GAME_AXD) + i * sizeof (u16));
-		if (axes[i] == 0xffff)
-			axes[i] = -1;
-	}
-
-	return 0;
-}
-
-static int trident_game_open(struct gameport *gameport, int mode)
-{
-	struct trident_card *card = gameport->port_data;
-
-	switch (mode) {
-	case GAMEPORT_MODE_COOKED:
-		outb(0x80, TRID_REG(card, T4D_GAME_CR));
-		msleep(20);
-		return 0;
-	case GAMEPORT_MODE_RAW:
-		outb(0x00, TRID_REG(card, T4D_GAME_CR));
-		return 0;
-	default:
-		return -1;
-	}
-
-	return 0;
-}
-
-static int __devinit trident_register_gameport(struct trident_card *card)
-{
-	struct gameport *gp;
-
-	card->gameport = gp = gameport_allocate_port();
-	if (!gp) {
-		printk(KERN_ERR "trident: can not allocate memory for gameport\n");
-		return -ENOMEM;
-	}
-
-	gameport_set_name(gp, "Trident 4DWave");
-	gameport_set_phys(gp, "pci%s/gameport0", pci_name(card->pci_dev));
-	gp->read = trident_game_read;
-	gp->trigger = trident_game_trigger;
-	gp->cooked_read = trident_game_cooked_read;
-	gp->open = trident_game_open;
-	gp->fuzz = 64;
-	gp->port_data = card;
-
-	gameport_register_port(gp);
-
-	return 0;
-}
-
-static inline void trident_unregister_gameport(struct trident_card *card)
-{
-	if (card->gameport)
-		gameport_unregister_port(card->gameport);
-}
-
-#else
-static inline int trident_register_gameport(struct trident_card *card) { return -ENOSYS; }
-static inline void trident_unregister_gameport(struct trident_card *card) { }
-#endif /* SUPPORT_JOYSTICK */
-
-/* install the driver, we do not allocate hardware channel nor DMA buffer */
-/* now, they are defered until "ACCESS" time (in prog_dmabuf called by */
-/* open/read/write/ioctl/mmap) */
-static int __devinit
-trident_probe(struct pci_dev *pci_dev, const struct pci_device_id *pci_id)
-{
-	unsigned long iobase;
-	struct trident_card *card;
-	u8 bits;
-	u8 revision;
-	int i = 0;
-	u16 temp;
-	struct pci_dev *pci_dev_m1533 = NULL;
-	int rc = -ENODEV;
-	u64 dma_mask;
-
-	if (pci_enable_device(pci_dev))
-		goto out;
-
-	if (pci_dev->device == PCI_DEVICE_ID_ALI_5451)
-		dma_mask = ALI_DMA_MASK;
-	else
-		dma_mask = TRIDENT_DMA_MASK;
-	if (pci_set_dma_mask(pci_dev, dma_mask)) {
-		printk(KERN_ERR "trident: architecture does not support"
-		       " %s PCI busmaster DMA\n",
-		       pci_dev->device == PCI_DEVICE_ID_ALI_5451 ?
-		       "32-bit" : "30-bit");
-		goto out;
-	}
-	pci_read_config_byte(pci_dev, PCI_CLASS_REVISION, &revision);
-
-	if (pci_id->device == PCI_DEVICE_ID_INTERG_5050)
-		iobase = pci_resource_start(pci_dev, 1);
-	else
-		iobase = pci_resource_start(pci_dev, 0);
-
-	if (!request_region(iobase, 256, card_names[pci_id->driver_data])) {
-		printk(KERN_ERR "trident: can't allocate I/O space at "
-		       "0x%4.4lx\n", iobase);
-		goto out;
-	}
-
-	rc = -ENOMEM;
-	if ((card = kzalloc(sizeof(*card), GFP_KERNEL)) == NULL) {
-		printk(KERN_ERR "trident: out of memory\n");
-		goto out_release_region;
-	}
-
-	init_timer(&card->timer);
-	card->iobase = iobase;
-	card->pci_dev = pci_dev_get(pci_dev);
-	card->pci_id = pci_id->device;
-	card->revision = revision;
-	card->irq = pci_dev->irq;
-	card->next = devs;
-	card->magic = TRIDENT_CARD_MAGIC;
-	card->banks[BANK_A].addresses = &bank_a_addrs;
-	card->banks[BANK_A].bitmap = 0UL;
-	card->banks[BANK_B].addresses = &bank_b_addrs;
-	card->banks[BANK_B].bitmap = 0UL;
-
-	mutex_init(&card->open_mutex);
-	spin_lock_init(&card->lock);
-	init_timer(&card->timer);
-
-	devs = card;
-
-	pci_set_master(pci_dev);
-
-	printk(KERN_INFO "trident: %s found at IO 0x%04lx, IRQ %d\n",
-	       card_names[pci_id->driver_data], card->iobase, card->irq);
-
-	if (card->pci_id == PCI_DEVICE_ID_ALI_5451) {
-		/* ALi channel Management */
-		card->alloc_pcm_channel = ali_alloc_pcm_channel;
-		card->alloc_rec_pcm_channel = ali_alloc_rec_pcm_channel;
-		card->free_pcm_channel = ali_free_pcm_channel;
-
-		card->address_interrupt = ali_address_interrupt;
-
-		/* Added by Matt Wu 01-05-2001 for spdif in */
-		card->multi_channel_use_count = 0;
-		card->rec_channel_use_count = 0;
-
-		/* ALi SPDIF OUT function */
-		if (card->revision == ALI_5451_V02) {
-			ali_setup_spdif_out(card, ALI_PCM_TO_SPDIF_OUT);
-			res = create_proc_entry("ALi5451", 0, NULL);
-			if (res) {
-				res->write_proc = ali_write_proc;
-				res->data = card;
-			}
-		}
-
-		/* Add H/W Volume Control By Matt Wu Jul. 06, 2001 */
-		card->hwvolctl = 0;
-		pci_dev_m1533 = pci_get_device(PCI_VENDOR_ID_AL,
-					       PCI_DEVICE_ID_AL_M1533,
-					       pci_dev_m1533);
-		rc = -ENODEV;
-		if (pci_dev_m1533 == NULL)
-			goto out_proc_fs;
-		pci_read_config_byte(pci_dev_m1533, 0x63, &bits);
-		if (bits & (1 << 5))
-			card->hwvolctl = 1;
-		if (card->hwvolctl) {
-			/* Clear m1533 pci cfg 78h bit 30 to zero, which makes
-			   GPIO11/12/13 work as ACGP_UP/DOWN/MUTE. */
-			pci_read_config_byte(pci_dev_m1533, 0x7b, &bits);
-			bits &= 0xbf;	/*clear bit 6 */
-			pci_write_config_byte(pci_dev_m1533, 0x7b, bits);
-		}
-		pci_dev_put(pci_dev_m1533);
-
-	} else if (card->pci_id == PCI_DEVICE_ID_INTERG_5050) {
-		card->alloc_pcm_channel = cyber_alloc_pcm_channel;
-		card->alloc_rec_pcm_channel = cyber_alloc_pcm_channel;
-		card->free_pcm_channel = cyber_free_pcm_channel;
-		card->address_interrupt = cyber_address_interrupt;
-		cyber_init_ritual(card);
-	} else {
-		card->alloc_pcm_channel = trident_alloc_pcm_channel;
-		card->alloc_rec_pcm_channel = trident_alloc_pcm_channel;
-		card->free_pcm_channel = trident_free_pcm_channel;
-		card->address_interrupt = trident_address_interrupt;
-	}
-
-	/* claim our irq */
-	rc = -ENODEV;
-	if (request_irq(card->irq, &trident_interrupt, IRQF_SHARED,
-			card_names[pci_id->driver_data], card)) {
-		printk(KERN_ERR "trident: unable to allocate irq %d\n",
-		       card->irq);
-		goto out_proc_fs;
-	}
-	/* register /dev/dsp */
-	if ((card->dev_audio = register_sound_dsp(&trident_audio_fops, -1)) < 0) {
-		printk(KERN_ERR "trident: couldn't register DSP device!\n");
-		goto out_free_irq;
-	}
-	card->mixer_regs_ready = 0;
-	/* initialize AC97 codec and register /dev/mixer */
-	if (trident_ac97_init(card) <= 0) {
-		/* unregister audio devices */
-		for (i = 0; i < NR_AC97; i++) {
-			if (card->ac97_codec[i] != NULL) {
-				struct ac97_codec* codec = card->ac97_codec[i];
-				unregister_sound_mixer(codec->dev_mixer);
-				ac97_release_codec(codec);
-			}
-		}
-		goto out_unregister_sound_dsp;
-	}
-	card->mixer_regs_ready = 1;
-	outl(0x00, TRID_REG(card, T4D_MUSICVOL_WAVEVOL));
-
-	if (card->pci_id == PCI_DEVICE_ID_ALI_5451) {
-		/* Add H/W Volume Control By Matt Wu Jul. 06, 2001 */
-		if (card->hwvolctl) {
-			/* Enable GPIO IRQ (MISCINT bit 18h) */
-			temp = inw(TRID_REG(card, T4D_MISCINT + 2));
-			temp |= 0x0004;
-			outw(temp, TRID_REG(card, T4D_MISCINT + 2));
-
-			/* Enable H/W Volume Control GLOVAL CONTROL bit 0 */
-			temp = inw(TRID_REG(card, ALI_GLOBAL_CONTROL));
-			temp |= 0x0001;
-			outw(temp, TRID_REG(card, ALI_GLOBAL_CONTROL));
-
-		}
-		if (card->revision == ALI_5451_V02)
-			ali_close_multi_channels();
-		/* edited by HMSEO for GT sound */
-#if defined(CONFIG_ALPHA_NAUTILUS) || defined(CONFIG_ALPHA_GENERIC)
-		{
-			u16 ac97_data;
-			extern struct hwrpb_struct *hwrpb;
-
-			if ((hwrpb->sys_type) == 201) {
-				printk(KERN_INFO "trident: Running on Alpha system "
-				       "type Nautilus\n");
-				ac97_data = ali_ac97_get(card, 0, AC97_POWER_CONTROL);
-				ali_ac97_set(card, 0, AC97_POWER_CONTROL,
-					     ac97_data | ALI_EAPD_POWER_DOWN);
-			}
-		}
-#endif				/* CONFIG_ALPHA_NAUTILUS || CONFIG_ALPHA_GENERIC */
-		/* edited by HMSEO for GT sound */
-	}
-	rc = 0;
-	pci_set_drvdata(pci_dev, card);
-
-	/* Enable Address Engine Interrupts */
-	trident_enable_loop_interrupts(card);
-
-	/* Register gameport */
-	trident_register_gameport(card);
-
-out:
-	return rc;
-
-out_unregister_sound_dsp:
-	unregister_sound_dsp(card->dev_audio);
-out_free_irq:
-	free_irq(card->irq, card);
-out_proc_fs:
-	pci_dev_put(card->pci_dev);
-	if (res) {
-		remove_proc_entry("ALi5451", NULL);
-		res = NULL;
-	}
-	kfree(card);
-	devs = NULL;
-out_release_region:
-	release_region(iobase, 256);
-	return rc;
-}
-
-static void __devexit
-trident_remove(struct pci_dev *pci_dev)
-{
-	int i;
-	struct trident_card *card = pci_get_drvdata(pci_dev);
-
-	/*
-	 *      Kill running timers before unload. We can't have them
-	 *      going off after rmmod!
-	 */
-	if (card->hwvolctl)
-		del_timer_sync(&card->timer);
-
-	/* ALi S/PDIF and Power Management */
-	if (card->pci_id == PCI_DEVICE_ID_ALI_5451) {
-		ali_setup_spdif_out(card, ALI_PCM_TO_SPDIF_OUT);
-		ali_disable_special_channel(card, ALI_SPDIF_OUT_CHANNEL);
-		ali_disable_spdif_in(card);
-		remove_proc_entry("ALi5451", NULL);
-	}
-
-	/* Unregister gameport */
-	trident_unregister_gameport(card);
-
-	/* Kill interrupts, and SP/DIF */
-	trident_disable_loop_interrupts(card);
-
-	/* free hardware resources */
-	free_irq(card->irq, card);
-	release_region(card->iobase, 256);
-
-	/* unregister audio devices */
-	for (i = 0; i < NR_AC97; i++)
-		if (card->ac97_codec[i] != NULL) {
-			unregister_sound_mixer(card->ac97_codec[i]->dev_mixer);
-			ac97_release_codec(card->ac97_codec[i]);
-		}
-	unregister_sound_dsp(card->dev_audio);
-
-	pci_set_drvdata(pci_dev, NULL);
-	pci_dev_put(card->pci_dev);
-	kfree(card);
-}
-
-MODULE_AUTHOR("Alan Cox, Aaron Holtzman, Ollie Lho, Ching Ling Lee, Muli Ben-Yehuda");
-MODULE_DESCRIPTION("Trident 4DWave/SiS 7018/ALi 5451 and Tvia/IGST CyberPro5050 PCI "
-		   "Audio Driver");
-MODULE_LICENSE("GPL");
-
-#define TRIDENT_MODULE_NAME "trident"
-
-static struct pci_driver trident_pci_driver = {
-	.name = TRIDENT_MODULE_NAME,
-	.id_table = trident_pci_tbl,
-	.probe = trident_probe,
-	.remove = __devexit_p(trident_remove),
-#ifdef CONFIG_PM
-	.suspend = trident_suspend,
-	.resume = trident_resume
-#endif
-};
-
-static int __init
-trident_init_module(void)
-{
-	printk(KERN_INFO "Trident 4DWave/SiS 7018/ALi 5451,Tvia CyberPro "
-	       "5050 PCI Audio, version " DRIVER_VERSION ", " __TIME__ " "
-	       __DATE__ "\n");
-
-	return pci_register_driver(&trident_pci_driver);
-}
-
-static void __exit
-trident_cleanup_module(void)
-{
-	pci_unregister_driver(&trident_pci_driver);
-}
-
-module_init(trident_init_module);
-module_exit(trident_cleanup_module);
diff --git a/sound/oss/trident.h b/sound/oss/trident.h
deleted file mode 100644
index ff30a1d7c2f1..000000000000
--- a/sound/oss/trident.h
+++ /dev/null
@@ -1,358 +0,0 @@
-#ifndef __TRID4DWAVE_H
-#define __TRID4DWAVE_H
-
-/*
- *  audio@tridentmicro.com
- *  Fri Feb 19 15:55:28 MST 1999
- *  Definitions for Trident 4DWave DX/NX chips
- *
- *
- *   This program is free software; you can redistribute it and/or modify
- *   it under the terms of the GNU General Public License as published by
- *   the Free Software Foundation; either version 2 of the License, or
- *   (at your option) any later version.
- *
- *   This program is distributed in the hope that it will be useful,
- *   but WITHOUT ANY WARRANTY; without even the implied warranty of
- *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- *   GNU General Public License for more details.
- *
- *   You should have received a copy of the GNU General Public License
- *   along with this program; if not, write to the Free Software
- *   Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
- *
- */
-
-/* PCI vendor and device ID */ 
-#ifndef PCI_VENDOR_ID_TRIDENT
-#define PCI_VENDOR_ID_TRIDENT		0x1023
-#endif
-
-#ifndef PCI_VENDOR_ID_SI
-#define PCI_VENDOR_ID_SI			0x1039
-#endif
-
-#ifndef PCI_VENDOR_ID_ALI
-#define PCI_VENDOR_ID_ALI			0x10b9
-#endif
-
-#ifndef PCI_DEVICE_ID_TRIDENT_4DWAVE_DX
-#define PCI_DEVICE_ID_TRIDENT_4DWAVE_DX	0x2000
-#endif
-
-#ifndef PCI_DEVICE_ID_TRIDENT_4DWAVE_NX
-#define PCI_DEVICE_ID_TRIDENT_4DWAVE_NX	0x2001
-#endif
-
-#ifndef PCI_DEVICE_ID_SI_7018
-#define PCI_DEVICE_ID_SI_7018		0x7018
-#endif
-
-#ifndef PCI_DEVICE_ID_ALI_5451
-#define PCI_DEVICE_ID_ALI_5451		0x5451
-#endif
-
-#ifndef PCI_DEVICE_ID_ALI_1533
-#define PCI_DEVICE_ID_ALI_1533		0x1533
-#endif
-
-#define CHANNEL_REGS	5
-#define CHANNEL_START	0xe0   // The first bytes of the contiguous register space.
-
-#define BANK_A 		0
-#define BANK_B 		1
-#define NR_BANKS		2
-
-#define TRIDENT_FMT_STEREO     0x01
-#define TRIDENT_FMT_16BIT      0x02
-#define TRIDENT_FMT_MASK       0x03
-
-#define DAC_RUNNING	0x01
-#define ADC_RUNNING	0x02
-
-/* Register Addresses */
-
-/* operational registers common to DX, NX, 7018 */
-enum trident_op_registers {
-	T4D_GAME_CR	= 0x30, T4D_GAME_LEG	= 0x31,
-	T4D_GAME_AXD	= 0x34,
-	T4D_REC_CH	= 0x70,
-	T4D_START_A     = 0x80, T4D_STOP_A      = 0x84,
-	T4D_DLY_A       = 0x88, T4D_SIGN_CSO_A  = 0x8c,
-	T4D_CSPF_A      = 0x90, T4D_CEBC_A      = 0x94,
-	T4D_AINT_A      = 0x98, T4D_EINT_A	= 0x9c,
-	T4D_LFO_GC_CIR	= 0xa0, T4D_AINTEN_A    = 0xa4,
-	T4D_MUSICVOL_WAVEVOL = 0xa8, T4D_SBDELTA_DELTA_R = 0xac,
-	T4D_MISCINT	= 0xb0, T4D_START_B     = 0xb4,
-	T4D_STOP_B      = 0xb8, T4D_CSPF_B	= 0xbc,
-	T4D_SBBL_SBCL	= 0xc0, T4D_SBCTRL_SBE2R_SBDD    = 0xc4,
-	T4D_STIMER	= 0xc8, T4D_LFO_B_I2S_DELTA      = 0xcc,
-	T4D_AINT_B	= 0xd8, T4D_AINTEN_B	= 0xdc,
-	ALI_MPUR2	= 0x22,	ALI_GPIO	= 0x7c,
-	ALI_EBUF1 = 0xf4,
-	ALI_EBUF2 = 0xf8
-};
-
-enum ali_op_registers {
-	ALI_SCTRL		= 0x48,
-	ALI_GLOBAL_CONTROL	= 0xd4,
-	ALI_STIMER		= 0xc8,
-	ALI_SPDIF_CS		= 0x70,
-	ALI_SPDIF_CTRL		= 0x74
-};
-
-enum ali_registers_number {
-	ALI_GLOBAL_REGS		= 56,
-	ALI_CHANNEL_REGS	= 8,
-	ALI_MIXER_REGS		= 20
-};
-
-enum ali_sctrl_control_bit {
-	ALI_SPDIF_OUT_ENABLE	= 0x20
-};
-
-enum ali_global_control_bit {
-	ALI_SPDIF_OUT_SEL_PCM	= 0x00000400,
-	ALI_SPDIF_IN_SUPPORT	= 0x00000800,
-	ALI_SPDIF_OUT_CH_ENABLE	= 0x00008000,
-	ALI_SPDIF_IN_CH_ENABLE	= 0x00080000,
-	ALI_PCM_IN_DISABLE	= 0x7fffffff,
-	ALI_PCM_IN_ENABLE	= 0x80000000,
-	ALI_SPDIF_IN_CH_DISABLE	= 0xfff7ffff,
-	ALI_SPDIF_OUT_CH_DISABLE = 0xffff7fff,
-	ALI_SPDIF_OUT_SEL_SPDIF	= 0xfffffbff
-	
-};
-
-enum ali_spdif_control_bit {
-	ALI_SPDIF_IN_FUNC_ENABLE	= 0x02,
-	ALI_SPDIF_IN_CH_STATUS		= 0x40,
-	ALI_SPDIF_OUT_CH_STATUS		= 0xbf
-	
-};
-
-enum ali_control_all {
-	ALI_DISABLE_ALL_IRQ	= 0,
-	ALI_CHANNELS		= 32,
-	ALI_STOP_ALL_CHANNELS	= 0xffffffff,
-	ALI_MULTI_CHANNELS_START_STOP	= 0x07800000
-};
-
-enum ali_EMOD_control_bit {
-	ALI_EMOD_DEC	= 0x00000000,
-	ALI_EMOD_INC	= 0x10000000,
-	ALI_EMOD_Delay	= 0x20000000,
-	ALI_EMOD_Still	= 0x30000000
-};
-
-enum ali_pcm_in_channel_num {
-	ALI_NORMAL_CHANNEL	= 0,
-	ALI_SPDIF_OUT_CHANNEL	= 15,
-	ALI_SPDIF_IN_CHANNEL    = 19,
-	ALI_LEF_CHANNEL		= 23,
-	ALI_CENTER_CHANNEL	= 24,
-	ALI_SURR_RIGHT_CHANNEL	= 25,
-	ALI_SURR_LEFT_CHANNEL	= 26,
-	ALI_PCM_IN_CHANNEL	= 31
-};
-
-enum ali_pcm_out_channel_num {
-	ALI_PCM_OUT_CHANNEL_FIRST = 0,
-	ALI_PCM_OUT_CHANNEL_LAST = 31
-};
-
-enum ali_ac97_power_control_bit {
-	ALI_EAPD_POWER_DOWN	= 0x8000
-};
-
-enum ali_update_ptr_flags {
-	ALI_ADDRESS_INT_UPDATE	= 0x01
-};
-
-enum ali_revision {
-	ALI_5451_V02	= 0x02
-};
-
-enum ali_spdif_out_control {
-	ALI_PCM_TO_SPDIF_OUT		= 0,
-	ALI_SPDIF_OUT_TO_SPDIF_OUT	= 1,
-	ALI_SPDIF_OUT_PCM		= 0,
-	ALI_SPDIF_OUT_NON_PCM		= 2
-};
-
-/* S/PDIF Operational Registers for 4D-NX */
-enum nx_spdif_registers {
-	NX_SPCTRL_SPCSO	= 0x24, NX_SPLBA = 0x28,
-	NX_SPESO	= 0x2c, NX_SPCSTATUS = 0x64
-};
-
-/* OP registers to access each hardware channel */
-enum channel_registers {
-	CH_DX_CSO_ALPHA_FMS = 0xe0, CH_DX_ESO_DELTA = 0xe8,
-	CH_DX_FMC_RVOL_CVOL = 0xec,
-	CH_NX_DELTA_CSO     = 0xe0, CH_NX_DELTA_ESO = 0xe8,
-	CH_NX_ALPHA_FMS_FMC_RVOL_CVOL = 0xec,
-	CH_LBA              = 0xe4,
-	CH_GVSEL_PAN_VOL_CTRL_EC      = 0xf0
-};
-
-/* registers to read/write/control AC97 codec */
-enum dx_ac97_registers {
-	DX_ACR0_AC97_W        = 0x40, DX_ACR1_AC97_R = 0x44,
-	DX_ACR2_AC97_COM_STAT = 0x48
-};
-
-enum nx_ac97_registers {
-	NX_ACR0_AC97_COM_STAT  = 0x40, NX_ACR1_AC97_W           = 0x44,
-	NX_ACR2_AC97_R_PRIMARY = 0x48, NX_ACR3_AC97_R_SECONDARY	= 0x4c
-};
-
-enum si_ac97_registers {
-	SI_AC97_WRITE       = 0x40, SI_AC97_READ = 0x44,
-	SI_SERIAL_INTF_CTRL = 0x48, SI_AC97_GPIO = 0x4c
-};
-
-enum ali_ac97_registers {
-	ALI_AC97_WRITE       = 0x40, ALI_AC97_READ = 0x44
-};
-
-/* Bit mask for operational registers */
-#define AC97_REG_ADDR      0x000000ff
-
-enum ali_ac97_bits {
-	ALI_AC97_BUSY_WRITE = 0x8000, ALI_AC97_BUSY_READ = 0x8000,
-	ALI_AC97_WRITE_ACTION = 0x8000, ALI_AC97_READ_ACTION = 0x8000,
-	ALI_AC97_AUDIO_BUSY = 0x4000, ALI_AC97_SECONDARY  = 0x0080,
-	ALI_AC97_READ_MIXER_REGISTER = 0xfeff,
-	ALI_AC97_WRITE_MIXER_REGISTER = 0x0100
-};
-
-enum sis7018_ac97_bits {
-	SI_AC97_BUSY_WRITE = 0x8000, SI_AC97_BUSY_READ = 0x8000,
-	SI_AC97_AUDIO_BUSY = 0x4000, SI_AC97_MODEM_BUSY = 0x2000,
-	SI_AC97_SECONDARY  = 0x0080
-};
-
-enum trident_dx_ac97_bits {
-	DX_AC97_BUSY_WRITE = 0x8000, DX_AC97_BUSY_READ = 0x8000,
-	DX_AC97_READY      = 0x0010, DX_AC97_RECORD    = 0x0008,
-	DX_AC97_PLAYBACK   = 0x0002
-};
-
-enum trident_nx_ac97_bits {
-	/* ACR1-3 */
-	NX_AC97_BUSY_WRITE = 0x0800, NX_AC97_BUSY_READ = 0x0800,
-	NX_AC97_BUSY_DATA  = 0x0400, NX_AC97_WRITE_SECONDARY = 0x0100,
-	/* ACR0 */
-	NX_AC97_SECONDARY_READY = 0x0040, NX_AC97_SECONDARY_RECORD = 0x0020,
-	NX_AC97_SURROUND_OUTPUT = 0x0010,
-	NX_AC97_PRIMARY_READY   = 0x0008, NX_AC97_PRIMARY_RECORD   = 0x0004,
-	NX_AC97_PCM_OUTPUT      = 0x0002,
-	NX_AC97_WARM_RESET      = 0x0001
-};
-
-enum serial_intf_ctrl_bits {
-	WARM_REST   = 0x00000001, COLD_RESET  = 0x00000002,
-	I2S_CLOCK   = 0x00000004, PCM_SEC_AC97= 0x00000008,
-	AC97_DBL_RATE = 0x00000010, SPDIF_EN  = 0x00000020,
-	I2S_OUTPUT_EN = 0x00000040, I2S_INPUT_EN = 0x00000080,
-	PCMIN       = 0x00000100, LINE1IN     = 0x00000200,
-	MICIN       = 0x00000400, LINE2IN     = 0x00000800,
-	HEAD_SET_IN = 0x00001000, GPIOIN      = 0x00002000,
-	/* 7018 spec says id = 01 but the demo board routed to 10 
-	   SECONDARY_ID= 0x00004000, */
-	SECONDARY_ID= 0x00004000,
-	PCMOUT      = 0x00010000, SURROUT     = 0x00020000,
-	CENTEROUT   = 0x00040000, LFEOUT      = 0x00080000,
-	LINE1OUT    = 0x00100000, LINE2OUT    = 0x00200000,
-	GPIOOUT     = 0x00400000,
-	SI_AC97_PRIMARY_READY   = 0x01000000,
-	SI_AC97_SECONDARY_READY = 0x02000000,
-};
-
-enum global_control_bits {
-	CHANNLE_IDX = 0x0000003f, PB_RESET    = 0x00000100,
-	PAUSE_ENG   = 0x00000200,
-	OVERRUN_IE  = 0x00000400, UNDERRUN_IE = 0x00000800,
-	ENDLP_IE    = 0x00001000, MIDLP_IE    = 0x00002000,
-	ETOG_IE     = 0x00004000,
-	EDROP_IE    = 0x00008000, BANK_B_EN   = 0x00010000
-};
-
-enum channel_control_bits {
-	CHANNEL_LOOP   = 0x00001000, CHANNEL_SIGNED = 0x00002000,
-	CHANNEL_STEREO = 0x00004000, CHANNEL_16BITS = 0x00008000,
-};
-
-enum channel_attribute {
-	/* playback/record select */
-	CHANNEL_PB     = 0x0000, CHANNEL_SPC_PB = 0x4000,
-	CHANNEL_REC    = 0x8000, CHANNEL_REC_PB = 0xc000,
-	/* playback destination/record source select */
-	MODEM_LINE1    = 0x0000, MODEM_LINE2    = 0x0400,
-	PCM_LR         = 0x0800, HSET           = 0x0c00,
-	I2S_LR         = 0x1000, CENTER_LFE     = 0x1400,
-	SURR_LR        = 0x1800, SPDIF_LR       = 0x1c00,
-	MIC            = 0x1400,
-	/* mist stuff */
-	MONO_LEFT      = 0x0000, MONO_RIGHT     = 0x0100,
-	MONO_MIX       = 0x0200, SRC_ENABLE     = 0x0080,
-};
-
-enum miscint_bits {
-	PB_UNDERRUN_IRO = 0x00000001, REC_OVERRUN_IRQ = 0x00000002,
-	SB_IRQ          = 0x00000004, MPU401_IRQ      = 0x00000008,
-	OPL3_IRQ        = 0x00000010, ADDRESS_IRQ     = 0x00000020,
-	ENVELOPE_IRQ    = 0x00000040, ST_IRQ          = 0x00000080,
-	PB_UNDERRUN     = 0x00000100, REC_OVERRUN     = 0x00000200,
-	MIXER_UNDERFLOW = 0x00000400, MIXER_OVERFLOW  = 0x00000800,
-	ST_TARGET_REACHED = 0x00008000, PB_24K_MODE   = 0x00010000, 
-	ST_IRQ_EN       = 0x00800000, ACGPIO_IRQ      = 0x01000000
-};
-
-#define TRID_REG( trident, x ) ( (trident) -> iobase + (x) )
-
-#define		CYBER_PORT_AUDIO		0x3CE
-#define		CYBER_IDX_AUDIO_ENABLE          0x7B
-#define		CYBER_BMSK_AUDIO_INT_ENABLE	0x09
-#define		CYBER_BMSK_AUENZ		0x01
-#define		CYBER_BMSK_AUENZ_ENABLE		0x00
-#define		CYBER_IDX_IRQ_ENABLE		0x12
-      
-#define VALIDATE_MAGIC(FOO,MAG)				\
-({						  	\
-	if (!(FOO) || (FOO)->magic != MAG) { 		\
-		printk(invalid_magic,__func__);	\
-		return -ENXIO;			  	\
-	}					  	\
-})
-
-#define VALIDATE_STATE(a) VALIDATE_MAGIC(a,TRIDENT_STATE_MAGIC)
-#define VALIDATE_CARD(a) VALIDATE_MAGIC(a,TRIDENT_CARD_MAGIC)
-
-static inline unsigned ld2(unsigned int x)
-{
-	unsigned r = 0;
-	
-	if (x >= 0x10000) {
-		x >>= 16;
-		r += 16;
-	}
-	if (x >= 0x100) {
-		x >>= 8;
-		r += 8;
-	}
-	if (x >= 0x10) {
-		x >>= 4;
-		r += 4;
-	}
-	if (x >= 4) {
-		x >>= 2;
-		r += 2;
-	}
-	if (x >= 2)
-		r++;
-	return r;
-}
-
-#endif /* __TRID4DWAVE_H */

From 76a6f3dc9a7108785c145a298f82c72f9208fe17 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Wed, 23 Jul 2008 21:29:15 -0700
Subject: [PATCH 0358/1171] CONFIG_SOUND_WM97XX: remove stale makefile line

The driver is gone for a long time.

Reported-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 sound/oss/Makefile | 2 --
 1 file changed, 2 deletions(-)

diff --git a/sound/oss/Makefile b/sound/oss/Makefile
index 3a141474fb73..c611514f7ff1 100644
--- a/sound/oss/Makefile
+++ b/sound/oss/Makefile
@@ -31,8 +31,6 @@ obj-$(CONFIG_SOUND_VWSND)	+= vwsnd.o
 obj-$(CONFIG_SOUND_AU1550_AC97)	+= au1550_ac97.o ac97_codec.o
 obj-$(CONFIG_SOUND_BCM_CS4297A)	+= swarm_cs4297a.o
 
-obj-$(CONFIG_SOUND_WM97XX)	+= ac97_plugin_wm97xx.o
-
 obj-$(CONFIG_DMASOUND)		+= dmasound/
 
 # Declare multi-part drivers.

From 6e2c10a12a2170856f5582d62d583cbcd1cb5eaf Mon Sep 17 00:00:00 2001
From: Akinobu Mita <akinobu.mita@gmail.com>
Date: Wed, 23 Jul 2008 21:29:15 -0700
Subject: [PATCH 0359/1171] binfmt_misc: use simple_read_from_buffer()

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/binfmt_misc.c | 20 +++-----------------
 1 file changed, 3 insertions(+), 17 deletions(-)

diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c
index 7191306367c5..756205314c24 100644
--- a/fs/binfmt_misc.c
+++ b/fs/binfmt_misc.c
@@ -27,6 +27,7 @@
 #include <linux/namei.h>
 #include <linux/mount.h>
 #include <linux/syscalls.h>
+#include <linux/fs.h>
 
 #include <asm/uaccess.h>
 
@@ -535,31 +536,16 @@ static ssize_t
 bm_entry_read(struct file * file, char __user * buf, size_t nbytes, loff_t *ppos)
 {
 	Node *e = file->f_path.dentry->d_inode->i_private;
-	loff_t pos = *ppos;
 	ssize_t res;
 	char *page;
-	int len;
 
 	if (!(page = (char*) __get_free_page(GFP_KERNEL)))
 		return -ENOMEM;
 
 	entry_status(e, page);
-	len = strlen(page);
 
-	res = -EINVAL;
-	if (pos < 0)
-		goto out;
-	res = 0;
-	if (pos >= len)
-		goto out;
-	if (len < pos + nbytes)
-		nbytes = len - pos;
-	res = -EFAULT;
-	if (copy_to_user(buf, page + pos, nbytes))
-		goto out;
-	*ppos = pos + nbytes;
-	res = nbytes;
-out:
+	res = simple_read_from_buffer(buf, nbytes, ppos, page, strlen(page));
+
 	free_page((unsigned long) page);
 	return res;
 }

From a677a039be7243357d93502bff2b40850c942e2d Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Wed, 23 Jul 2008 21:29:17 -0700
Subject: [PATCH 0360/1171] flag parameters: socket and socketpair

This patch adds support for flag values which are ORed to the type passwd
to socket and socketpair.  The additional code is minimal.  The flag
values in this implementation can and must match the O_* flags.  This
avoids overhead in the conversion.

The internal functions sock_alloc_fd and sock_map_fd get a new parameters
and all callers are changed.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#include <fcntl.h>
#include <stdio.h>
#include <unistd.h>
#include <netinet/in.h>
#include <sys/socket.h>

#define PORT 57392

/* For Linux these must be the same.  */
#define SOCK_CLOEXEC O_CLOEXEC

int
main (void)
{
  int fd;
  fd = socket (PF_INET, SOCK_STREAM, 0);
  if (fd == -1)
    {
      puts ("socket(0) failed");
      return 1;
    }
  int coe = fcntl (fd, F_GETFD);
  if (coe == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if (coe & FD_CLOEXEC)
    {
      puts ("socket(0) set close-on-exec flag");
      return 1;
    }
  close (fd);

  fd = socket (PF_INET, SOCK_STREAM|SOCK_CLOEXEC, 0);
  if (fd == -1)
    {
      puts ("socket(SOCK_CLOEXEC) failed");
      return 1;
    }
  coe = fcntl (fd, F_GETFD);
  if (coe == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if ((coe & FD_CLOEXEC) == 0)
    {
      puts ("socket(SOCK_CLOEXEC) does not set close-on-exec flag");
      return 1;
    }
  close (fd);

  int fds[2];
  if (socketpair (PF_UNIX, SOCK_STREAM, 0, fds) == -1)
    {
      puts ("socketpair(0) failed");
      return 1;
    }
  for (int i = 0; i < 2; ++i)
    {
      coe = fcntl (fds[i], F_GETFD);
      if (coe == -1)
        {
          puts ("fcntl failed");
          return 1;
        }
      if (coe & FD_CLOEXEC)
        {
          printf ("socketpair(0) set close-on-exec flag for fds[%d]\n", i);
          return 1;
        }
      close (fds[i]);
    }

  if (socketpair (PF_UNIX, SOCK_STREAM|SOCK_CLOEXEC, 0, fds) == -1)
    {
      puts ("socketpair(SOCK_CLOEXEC) failed");
      return 1;
    }
  for (int i = 0; i < 2; ++i)
    {
      coe = fcntl (fds[i], F_GETFD);
      if (coe == -1)
        {
          puts ("fcntl failed");
          return 1;
        }
      if ((coe & FD_CLOEXEC) == 0)
        {
          printf ("socketpair(SOCK_CLOEXEC) does not set close-on-exec flag for fds[%d]\n", i);
          return 1;
        }
      close (fds[i]);
    }

  puts ("OK");

  return 0;
}
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/asm-mips/socket.h |  7 +++++++
 include/linux/net.h       |  9 ++++++++-
 net/9p/trans_fd.c         |  2 +-
 net/sctp/socket.c         |  2 +-
 net/socket.c              | 28 ++++++++++++++++++++--------
 5 files changed, 37 insertions(+), 11 deletions(-)

diff --git a/include/asm-mips/socket.h b/include/asm-mips/socket.h
index 63f60254d308..facc2d7a87ca 100644
--- a/include/asm-mips/socket.h
+++ b/include/asm-mips/socket.h
@@ -102,6 +102,13 @@ enum sock_type {
 };
 
 #define SOCK_MAX (SOCK_PACKET + 1)
+/* Mask which covers at least up to SOCK_MASK-1.  The
+ *  * remaining bits are used as flags. */
+#define SOCK_TYPE_MASK 0xf
+
+/* Flags for socket, socketpair, paccept */
+#define SOCK_CLOEXEC	O_CLOEXEC
+#define SOCK_NONBLOCK	O_NONBLOCK
 
 #define ARCH_HAS_SOCKET_TYPES 1
 
diff --git a/include/linux/net.h b/include/linux/net.h
index 150a48c68d52..8b5383c45b45 100644
--- a/include/linux/net.h
+++ b/include/linux/net.h
@@ -20,6 +20,7 @@
 
 #include <linux/wait.h>
 #include <linux/socket.h>
+#include <linux/fcntl.h>	/* For O_CLOEXEC */
 #include <asm/socket.h>
 
 struct poll_table_struct;
@@ -94,6 +95,12 @@ enum sock_type {
 };
 
 #define SOCK_MAX (SOCK_PACKET + 1)
+/* Mask which covers at least up to SOCK_MASK-1.  The
+ * remaining bits are used as flags. */
+#define SOCK_TYPE_MASK 0xf
+
+/* Flags for socket, socketpair, paccept */
+#define SOCK_CLOEXEC	O_CLOEXEC
 
 #endif /* ARCH_HAS_SOCKET_TYPES */
 
@@ -208,7 +215,7 @@ extern int   	     sock_sendmsg(struct socket *sock, struct msghdr *msg,
 				  size_t len);
 extern int	     sock_recvmsg(struct socket *sock, struct msghdr *msg,
 				  size_t size, int flags);
-extern int 	     sock_map_fd(struct socket *sock);
+extern int 	     sock_map_fd(struct socket *sock, int flags);
 extern struct socket *sockfd_lookup(int fd, int *err);
 #define		     sockfd_put(sock) fput(sock->file)
 extern int	     net_ratelimit(void);
diff --git a/net/9p/trans_fd.c b/net/9p/trans_fd.c
index 4507f744f44e..cdf137af7adc 100644
--- a/net/9p/trans_fd.c
+++ b/net/9p/trans_fd.c
@@ -1285,7 +1285,7 @@ static int p9_socket_open(struct p9_trans *trans, struct socket *csocket)
 	int fd, ret;
 
 	csocket->sk->sk_allocation = GFP_NOIO;
-	fd = sock_map_fd(csocket);
+	fd = sock_map_fd(csocket, 0);
 	if (fd < 0) {
 		P9_EPRINTK(KERN_ERR, "p9_socket_open: failed to map fd\n");
 		return fd;
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index 79bece16aede..dbb79adf8f3c 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -3910,7 +3910,7 @@ static int sctp_getsockopt_peeloff(struct sock *sk, int len, char __user *optval
 		goto out;
 
 	/* Map the socket to an unused fd that can be returned to the user.  */
-	retval = sock_map_fd(newsock);
+	retval = sock_map_fd(newsock, 0);
 	if (retval < 0) {
 		sock_release(newsock);
 		goto out;
diff --git a/net/socket.c b/net/socket.c
index 1ba57d888981..64601f900352 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -349,11 +349,11 @@ static struct dentry_operations sockfs_dentry_operations = {
  *	but we take care of internal coherence yet.
  */
 
-static int sock_alloc_fd(struct file **filep)
+static int sock_alloc_fd(struct file **filep, int flags)
 {
 	int fd;
 
-	fd = get_unused_fd();
+	fd = get_unused_fd_flags(flags);
 	if (likely(fd >= 0)) {
 		struct file *file = get_empty_filp();
 
@@ -396,10 +396,10 @@ static int sock_attach_fd(struct socket *sock, struct file *file)
 	return 0;
 }
 
-int sock_map_fd(struct socket *sock)
+int sock_map_fd(struct socket *sock, int flags)
 {
 	struct file *newfile;
-	int fd = sock_alloc_fd(&newfile);
+	int fd = sock_alloc_fd(&newfile, flags);
 
 	if (likely(fd >= 0)) {
 		int err = sock_attach_fd(sock, newfile);
@@ -1218,12 +1218,18 @@ asmlinkage long sys_socket(int family, int type, int protocol)
 {
 	int retval;
 	struct socket *sock;
+	int flags;
+
+	flags = type & ~SOCK_TYPE_MASK;
+	if (flags & ~SOCK_CLOEXEC)
+		return -EINVAL;
+	type &= SOCK_TYPE_MASK;
 
 	retval = sock_create(family, type, protocol, &sock);
 	if (retval < 0)
 		goto out;
 
-	retval = sock_map_fd(sock);
+	retval = sock_map_fd(sock, flags & O_CLOEXEC);
 	if (retval < 0)
 		goto out_release;
 
@@ -1246,6 +1252,12 @@ asmlinkage long sys_socketpair(int family, int type, int protocol,
 	struct socket *sock1, *sock2;
 	int fd1, fd2, err;
 	struct file *newfile1, *newfile2;
+	int flags;
+
+	flags = type & ~SOCK_TYPE_MASK;
+	if (flags & ~SOCK_CLOEXEC)
+		return -EINVAL;
+	type &= SOCK_TYPE_MASK;
 
 	/*
 	 * Obtain the first socket and check if the underlying protocol
@@ -1264,13 +1276,13 @@ asmlinkage long sys_socketpair(int family, int type, int protocol,
 	if (err < 0)
 		goto out_release_both;
 
-	fd1 = sock_alloc_fd(&newfile1);
+	fd1 = sock_alloc_fd(&newfile1, flags & O_CLOEXEC);
 	if (unlikely(fd1 < 0)) {
 		err = fd1;
 		goto out_release_both;
 	}
 
-	fd2 = sock_alloc_fd(&newfile2);
+	fd2 = sock_alloc_fd(&newfile2, flags & O_CLOEXEC);
 	if (unlikely(fd2 < 0)) {
 		err = fd2;
 		put_filp(newfile1);
@@ -1426,7 +1438,7 @@ asmlinkage long sys_accept(int fd, struct sockaddr __user *upeer_sockaddr,
 	 */
 	__module_get(newsock->ops->owner);
 
-	newfd = sock_alloc_fd(&newfile);
+	newfd = sock_alloc_fd(&newfile, 0);
 	if (unlikely(newfd < 0)) {
 		err = newfd;
 		sock_release(newsock);

From aaca0bdca573f3f51ea03139f9c7289541e7bca3 Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Wed, 23 Jul 2008 21:29:20 -0700
Subject: [PATCH 0361/1171] flag parameters: paccept

This patch is by far the most complex in the series.  It adds a new syscall
paccept.  This syscall differs from accept in that it adds (at the userlevel)
two additional parameters:

- a signal mask
- a flags value

The flags parameter can be used to set flag like SOCK_CLOEXEC.  This is
imlpemented here as well.  Some people argued that this is a property which
should be inherited from the file desriptor for the server but this is against
POSIX.  Additionally, we really want the signal mask parameter as well
(similar to pselect, ppoll, etc).  So an interface change in inevitable.

The flag value is the same as for socket and socketpair.  I think diverging
here will only create confusion.  Similar to the filesystem interfaces where
the use of the O_* constants differs, it is acceptable here.

The signal mask is handled as for pselect etc.  The mask is temporarily
installed for the thread and removed before the call returns.  I modeled the
code after pselect.  If there is a problem it's likely also in pselect.

For architectures which use socketcall I maintained this interface instead of
adding a system call.  The symmetry shouldn't be broken.

The following test must be adjusted for architectures other than x86 and
x86-64 and in case the syscall numbers changed.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#include <errno.h>
#include <fcntl.h>
#include <pthread.h>
#include <signal.h>
#include <stdio.h>
#include <unistd.h>
#include <netinet/in.h>
#include <sys/socket.h>
#include <sys/syscall.h>

#ifndef __NR_paccept
# ifdef __x86_64__
#  define __NR_paccept 288
# elif defined __i386__
#  define SYS_PACCEPT 18
#  define USE_SOCKETCALL 1
# else
#  error "need __NR_paccept"
# endif
#endif

#ifdef USE_SOCKETCALL
# define paccept(fd, addr, addrlen, mask, flags) \
  ({ long args[6] = { \
       (long) fd, (long) addr, (long) addrlen, (long) mask, 8, (long) flags }; \
     syscall (__NR_socketcall, SYS_PACCEPT, args); })
#else
# define paccept(fd, addr, addrlen, mask, flags) \
  syscall (__NR_paccept, fd, addr, addrlen, mask, 8, flags)
#endif

#define PORT 57392

#define SOCK_CLOEXEC O_CLOEXEC

static pthread_barrier_t b;

static void *
tf (void *arg)
{
  pthread_barrier_wait (&b);
  int s = socket (AF_INET, SOCK_STREAM, 0);
  struct sockaddr_in sin;
  sin.sin_family = AF_INET;
  sin.sin_addr.s_addr = htonl (INADDR_LOOPBACK);
  sin.sin_port = htons (PORT);
  connect (s, (const struct sockaddr *) &sin, sizeof (sin));
  close (s);

  pthread_barrier_wait (&b);
  s = socket (AF_INET, SOCK_STREAM, 0);
  sin.sin_port = htons (PORT);
  connect (s, (const struct sockaddr *) &sin, sizeof (sin));
  close (s);
  pthread_barrier_wait (&b);

  pthread_barrier_wait (&b);
  sleep (2);
  pthread_kill ((pthread_t) arg, SIGUSR1);

  return NULL;
}

static void
handler (int s)
{
}

int
main (void)
{
  pthread_barrier_init (&b, NULL, 2);

  struct sockaddr_in sin;
  pthread_t th;
  if (pthread_create (&th, NULL, tf, (void *) pthread_self ()) != 0)
    {
      puts ("pthread_create failed");
      return 1;
    }

  int s = socket (AF_INET, SOCK_STREAM, 0);
  int reuse = 1;
  setsockopt (s, SOL_SOCKET, SO_REUSEADDR, &reuse, sizeof (reuse));
  sin.sin_family = AF_INET;
  sin.sin_addr.s_addr = htonl (INADDR_LOOPBACK);
  sin.sin_port = htons (PORT);
  bind (s, (struct sockaddr *) &sin, sizeof (sin));
  listen (s, SOMAXCONN);

  pthread_barrier_wait (&b);

  int s2 = paccept (s, NULL, 0, NULL, 0);
  if (s2 < 0)
    {
      puts ("paccept(0) failed");
      return 1;
    }

  int coe = fcntl (s2, F_GETFD);
  if (coe & FD_CLOEXEC)
    {
      puts ("paccept(0) set close-on-exec-flag");
      return 1;
    }
  close (s2);

  pthread_barrier_wait (&b);

  s2 = paccept (s, NULL, 0, NULL, SOCK_CLOEXEC);
  if (s2 < 0)
    {
      puts ("paccept(SOCK_CLOEXEC) failed");
      return 1;
    }

  coe = fcntl (s2, F_GETFD);
  if ((coe & FD_CLOEXEC) == 0)
    {
      puts ("paccept(SOCK_CLOEXEC) does not set close-on-exec flag");
      return 1;
    }
  close (s2);

  pthread_barrier_wait (&b);

  struct sigaction sa;
  sa.sa_handler = handler;
  sa.sa_flags = 0;
  sigemptyset (&sa.sa_mask);
  sigaction (SIGUSR1, &sa, NULL);

  sigset_t ss;
  pthread_sigmask (SIG_SETMASK, NULL, &ss);
  sigaddset (&ss, SIGUSR1);
  pthread_sigmask (SIG_SETMASK, &ss, NULL);

  sigdelset (&ss, SIGUSR1);
  alarm (4);
  pthread_barrier_wait (&b);

  errno = 0 ;
  s2 = paccept (s, NULL, 0, &ss, 0);
  if (s2 != -1 || errno != EINTR)
    {
      puts ("paccept did not fail with EINTR");
      return 1;
    }

  close (s);

  puts ("OK");

  return 0;
}
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

[akpm@linux-foundation.org: make it compile]
[akpm@linux-foundation.org: add sys_ni stub]
Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Cc: <linux-arch@vger.kernel.org>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Roland McGrath <roland@redhat.com>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/asm-alpha/socket.h  |  5 +++
 include/asm-parisc/socket.h |  5 +++
 include/asm-x86/unistd_64.h |  2 +
 include/linux/net.h         |  3 ++
 include/linux/syscalls.h    |  2 +
 kernel/sys_ni.c             |  1 +
 net/compat.c                | 52 ++++++++++++++++++++++--
 net/socket.c                | 81 +++++++++++++++++++++++++++++++++----
 8 files changed, 139 insertions(+), 12 deletions(-)

diff --git a/include/asm-alpha/socket.h b/include/asm-alpha/socket.h
index 08c979319929..a1057c2d95e7 100644
--- a/include/asm-alpha/socket.h
+++ b/include/asm-alpha/socket.h
@@ -62,4 +62,9 @@
 
 #define SO_MARK			36
 
+/* O_NONBLOCK clashes with the bits used for socket types.  Therefore we
+ * have to define SOCK_NONBLOCK to a different value here.
+ */
+#define SOCK_NONBLOCK	0x40000000
+
 #endif /* _ASM_SOCKET_H */
diff --git a/include/asm-parisc/socket.h b/include/asm-parisc/socket.h
index 69a7a0d30b02..fba402c95ac2 100644
--- a/include/asm-parisc/socket.h
+++ b/include/asm-parisc/socket.h
@@ -54,4 +54,9 @@
 
 #define SO_MARK			0x401f
 
+/* O_NONBLOCK clashes with the bits used for socket types.  Therefore we
+ * have to define SOCK_NONBLOCK to a different value here.
+ */
+#define SOCK_NONBLOCK   0x40000000
+
 #endif /* _ASM_SOCKET_H */
diff --git a/include/asm-x86/unistd_64.h b/include/asm-x86/unistd_64.h
index 9c1a4a3470d9..e323994a370f 100644
--- a/include/asm-x86/unistd_64.h
+++ b/include/asm-x86/unistd_64.h
@@ -639,6 +639,8 @@ __SYSCALL(__NR_fallocate, sys_fallocate)
 __SYSCALL(__NR_timerfd_settime, sys_timerfd_settime)
 #define __NR_timerfd_gettime			287
 __SYSCALL(__NR_timerfd_gettime, sys_timerfd_gettime)
+#define __NR_paccept				288
+__SYSCALL(__NR_paccept, sys_paccept)
 
 
 #ifndef __NO_STUBS
diff --git a/include/linux/net.h b/include/linux/net.h
index 8b5383c45b45..3a9b06d4d0fe 100644
--- a/include/linux/net.h
+++ b/include/linux/net.h
@@ -47,6 +47,7 @@ struct net;
 #define SYS_GETSOCKOPT	15		/* sys_getsockopt(2)		*/
 #define SYS_SENDMSG	16		/* sys_sendmsg(2)		*/
 #define SYS_RECVMSG	17		/* sys_recvmsg(2)		*/
+#define SYS_PACCEPT	18		/* sys_paccept(2)		*/
 
 typedef enum {
 	SS_FREE = 0,			/* not allocated		*/
@@ -219,6 +220,8 @@ extern int 	     sock_map_fd(struct socket *sock, int flags);
 extern struct socket *sockfd_lookup(int fd, int *err);
 #define		     sockfd_put(sock) fput(sock->file)
 extern int	     net_ratelimit(void);
+extern long	     do_accept(int fd, struct sockaddr __user *upeer_sockaddr,
+			       int __user *upeer_addrlen, int flags);
 
 #define net_random()		random32()
 #define net_srandom(seed)	srandom32((__force u32)seed)
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 4394dadff813..2a2a40af6b2c 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -409,6 +409,8 @@ asmlinkage long sys_getsockopt(int fd, int level, int optname,
 asmlinkage long sys_bind(int, struct sockaddr __user *, int);
 asmlinkage long sys_connect(int, struct sockaddr __user *, int);
 asmlinkage long sys_accept(int, struct sockaddr __user *, int __user *);
+asmlinkage long sys_paccept(int, struct sockaddr __user *, int __user *,
+			    const sigset_t *, size_t, int);
 asmlinkage long sys_getsockname(int, struct sockaddr __user *, int __user *);
 asmlinkage long sys_getpeername(int, struct sockaddr __user *, int __user *);
 asmlinkage long sys_send(int, void __user *, size_t, unsigned);
diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c
index 0fea0ee12da9..2f0b8a2e600f 100644
--- a/kernel/sys_ni.c
+++ b/kernel/sys_ni.c
@@ -31,6 +31,7 @@ cond_syscall(sys_socketpair);
 cond_syscall(sys_bind);
 cond_syscall(sys_listen);
 cond_syscall(sys_accept);
+cond_syscall(sys_paccept);
 cond_syscall(sys_connect);
 cond_syscall(sys_getsockname);
 cond_syscall(sys_getpeername);
diff --git a/net/compat.c b/net/compat.c
index 6e1b03b51933..67fb6a3834a3 100644
--- a/net/compat.c
+++ b/net/compat.c
@@ -722,9 +722,10 @@ EXPORT_SYMBOL(compat_mc_getsockopt);
 
 /* Argument list sizes for compat_sys_socketcall */
 #define AL(x) ((x) * sizeof(u32))
-static unsigned char nas[18]={AL(0),AL(3),AL(3),AL(3),AL(2),AL(3),
+static unsigned char nas[19]={AL(0),AL(3),AL(3),AL(3),AL(2),AL(3),
 				AL(3),AL(3),AL(4),AL(4),AL(4),AL(6),
-				AL(6),AL(2),AL(5),AL(5),AL(3),AL(3)};
+				AL(6),AL(2),AL(5),AL(5),AL(3),AL(3),
+				AL(6)};
 #undef AL
 
 asmlinkage long compat_sys_sendmsg(int fd, struct compat_msghdr __user *msg, unsigned flags)
@@ -737,13 +738,52 @@ asmlinkage long compat_sys_recvmsg(int fd, struct compat_msghdr __user *msg, uns
 	return sys_recvmsg(fd, (struct msghdr __user *)msg, flags | MSG_CMSG_COMPAT);
 }
 
+asmlinkage long compat_sys_paccept(int fd, struct sockaddr __user *upeer_sockaddr,
+				   int __user *upeer_addrlen,
+				   const compat_sigset_t __user *sigmask,
+				   compat_size_t sigsetsize, int flags)
+{
+	compat_sigset_t ss32;
+	sigset_t ksigmask, sigsaved;
+	int ret;
+
+	if (sigmask) {
+		if (sigsetsize != sizeof(compat_sigset_t))
+			return -EINVAL;
+		if (copy_from_user(&ss32, sigmask, sizeof(ss32)))
+			return -EFAULT;
+		sigset_from_compat(&ksigmask, &ss32);
+
+		sigdelsetmask(&ksigmask, sigmask(SIGKILL)|sigmask(SIGSTOP));
+		sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved);
+	}
+
+	ret = do_accept(fd, upeer_sockaddr, upeer_addrlen, flags);
+
+	if (ret == -ERESTARTNOHAND) {
+		/*
+		 * Don't restore the signal mask yet. Let do_signal() deliver
+		 * the signal on the way back to userspace, before the signal
+		 * mask is restored.
+		 */
+		if (sigmask) {
+			memcpy(&current->saved_sigmask, &sigsaved,
+			       sizeof(sigsaved));
+			set_restore_sigmask();
+		}
+	} else if (sigmask)
+		sigprocmask(SIG_SETMASK, &sigsaved, NULL);
+
+	return ret;
+}
+
 asmlinkage long compat_sys_socketcall(int call, u32 __user *args)
 {
 	int ret;
 	u32 a[6];
 	u32 a0, a1;
 
-	if (call < SYS_SOCKET || call > SYS_RECVMSG)
+	if (call < SYS_SOCKET || call > SYS_PACCEPT)
 		return -EINVAL;
 	if (copy_from_user(a, args, nas[call]))
 		return -EFAULT;
@@ -764,7 +804,7 @@ asmlinkage long compat_sys_socketcall(int call, u32 __user *args)
 		ret = sys_listen(a0, a1);
 		break;
 	case SYS_ACCEPT:
-		ret = sys_accept(a0, compat_ptr(a1), compat_ptr(a[2]));
+		ret = do_accept(a0, compat_ptr(a1), compat_ptr(a[2]), 0);
 		break;
 	case SYS_GETSOCKNAME:
 		ret = sys_getsockname(a0, compat_ptr(a1), compat_ptr(a[2]));
@@ -804,6 +844,10 @@ asmlinkage long compat_sys_socketcall(int call, u32 __user *args)
 	case SYS_RECVMSG:
 		ret = compat_sys_recvmsg(a0, compat_ptr(a1), a[2]);
 		break;
+	case SYS_PACCEPT:
+		ret = compat_sys_paccept(a0, compat_ptr(a1), compat_ptr(a[2]),
+					 compat_ptr(a[3]), a[4], a[5]);
+		break;
 	default:
 		ret = -EINVAL;
 		break;
diff --git a/net/socket.c b/net/socket.c
index 64601f900352..a0ce8ad72252 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -63,6 +63,7 @@
 #include <linux/file.h>
 #include <linux/net.h>
 #include <linux/interrupt.h>
+#include <linux/thread_info.h>
 #include <linux/rcupdate.h>
 #include <linux/netdevice.h>
 #include <linux/proc_fs.h>
@@ -1225,6 +1226,9 @@ asmlinkage long sys_socket(int family, int type, int protocol)
 		return -EINVAL;
 	type &= SOCK_TYPE_MASK;
 
+	if (SOCK_NONBLOCK != O_NONBLOCK && (flags & SOCK_NONBLOCK))
+		flags = (flags & ~SOCK_NONBLOCK) | O_NONBLOCK;
+
 	retval = sock_create(family, type, protocol, &sock);
 	if (retval < 0)
 		goto out;
@@ -1259,6 +1263,9 @@ asmlinkage long sys_socketpair(int family, int type, int protocol,
 		return -EINVAL;
 	type &= SOCK_TYPE_MASK;
 
+	if (SOCK_NONBLOCK != O_NONBLOCK && (flags & SOCK_NONBLOCK))
+		flags = (flags & ~SOCK_NONBLOCK) | O_NONBLOCK;
+
 	/*
 	 * Obtain the first socket and check if the underlying protocol
 	 * supports the socketpair call.
@@ -1413,14 +1420,20 @@ asmlinkage long sys_listen(int fd, int backlog)
  *	clean when we restucture accept also.
  */
 
-asmlinkage long sys_accept(int fd, struct sockaddr __user *upeer_sockaddr,
-			   int __user *upeer_addrlen)
+long do_accept(int fd, struct sockaddr __user *upeer_sockaddr,
+	       int __user *upeer_addrlen, int flags)
 {
 	struct socket *sock, *newsock;
 	struct file *newfile;
 	int err, len, newfd, fput_needed;
 	struct sockaddr_storage address;
 
+	if (flags & ~SOCK_CLOEXEC)
+		return -EINVAL;
+
+	if (SOCK_NONBLOCK != O_NONBLOCK && (flags & SOCK_NONBLOCK))
+		flags = (flags & ~SOCK_NONBLOCK) | O_NONBLOCK;
+
 	sock = sockfd_lookup_light(fd, &err, &fput_needed);
 	if (!sock)
 		goto out;
@@ -1438,7 +1451,7 @@ asmlinkage long sys_accept(int fd, struct sockaddr __user *upeer_sockaddr,
 	 */
 	__module_get(newsock->ops->owner);
 
-	newfd = sock_alloc_fd(&newfile, 0);
+	newfd = sock_alloc_fd(&newfile, flags & O_CLOEXEC);
 	if (unlikely(newfd < 0)) {
 		err = newfd;
 		sock_release(newsock);
@@ -1491,6 +1504,50 @@ out_fd:
 	goto out_put;
 }
 
+asmlinkage long sys_paccept(int fd, struct sockaddr __user *upeer_sockaddr,
+			    int __user *upeer_addrlen,
+			    const sigset_t __user *sigmask,
+			    size_t sigsetsize, int flags)
+{
+	sigset_t ksigmask, sigsaved;
+	int ret;
+
+	if (sigmask) {
+		/* XXX: Don't preclude handling different sized sigset_t's.  */
+		if (sigsetsize != sizeof(sigset_t))
+			return -EINVAL;
+		if (copy_from_user(&ksigmask, sigmask, sizeof(ksigmask)))
+			return -EFAULT;
+
+		sigdelsetmask(&ksigmask, sigmask(SIGKILL)|sigmask(SIGSTOP));
+		sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved);
+        }
+
+	ret = do_accept(fd, upeer_sockaddr, upeer_addrlen, flags);
+
+	if (ret < 0 && signal_pending(current)) {
+		/*
+		 * Don't restore the signal mask yet. Let do_signal() deliver
+		 * the signal on the way back to userspace, before the signal
+		 * mask is restored.
+		 */
+		if (sigmask) {
+			memcpy(&current->saved_sigmask, &sigsaved,
+			       sizeof(sigsaved));
+			set_restore_sigmask();
+		}
+	} else if (sigmask)
+		sigprocmask(SIG_SETMASK, &sigsaved, NULL);
+
+	return ret;
+}
+
+asmlinkage long sys_accept(int fd, struct sockaddr __user *upeer_sockaddr,
+			   int __user *upeer_addrlen)
+{
+	return do_accept(fd, upeer_sockaddr, upeer_addrlen, 0);
+}
+
 /*
  *	Attempt to connect to a socket with the server address.  The address
  *	is in user space so we verify it is OK and move it to kernel space.
@@ -2011,10 +2068,11 @@ out:
 
 /* Argument list sizes for sys_socketcall */
 #define AL(x) ((x) * sizeof(unsigned long))
-static const unsigned char nargs[18]={
+static const unsigned char nargs[19]={
 	AL(0),AL(3),AL(3),AL(3),AL(2),AL(3),
 	AL(3),AL(3),AL(4),AL(4),AL(4),AL(6),
-	AL(6),AL(2),AL(5),AL(5),AL(3),AL(3)
+	AL(6),AL(2),AL(5),AL(5),AL(3),AL(3),
+	AL(6)
 };
 
 #undef AL
@@ -2033,7 +2091,7 @@ asmlinkage long sys_socketcall(int call, unsigned long __user *args)
 	unsigned long a0, a1;
 	int err;
 
-	if (call < 1 || call > SYS_RECVMSG)
+	if (call < 1 || call > SYS_PACCEPT)
 		return -EINVAL;
 
 	/* copy_from_user should be SMP safe. */
@@ -2062,8 +2120,8 @@ asmlinkage long sys_socketcall(int call, unsigned long __user *args)
 		break;
 	case SYS_ACCEPT:
 		err =
-		    sys_accept(a0, (struct sockaddr __user *)a1,
-			       (int __user *)a[2]);
+		    do_accept(a0, (struct sockaddr __user *)a1,
+			      (int __user *)a[2], 0);
 		break;
 	case SYS_GETSOCKNAME:
 		err =
@@ -2110,6 +2168,13 @@ asmlinkage long sys_socketcall(int call, unsigned long __user *args)
 	case SYS_RECVMSG:
 		err = sys_recvmsg(a0, (struct msghdr __user *)a1, a[2]);
 		break;
+	case SYS_PACCEPT:
+		err =
+		    sys_paccept(a0, (struct sockaddr __user *)a1,
+			        (int __user *)a[2],
+				(const sigset_t __user *) a[3],
+				a[4], a[5]);
+		break;
 	default:
 		err = -EINVAL;
 		break;

From c019bbc612f6633ede7ed67725cbf68de45ae8a4 Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Wed, 23 Jul 2008 21:29:21 -0700
Subject: [PATCH 0362/1171] flag parameters: paccept w/out set_restore_sigmask

Some platforms do not have support to restore the signal mask in the
return path from a syscall.  For those platforms syscalls like pselect are
not defined at all.  This is, I think, not a good choice for paccept()
since paccept() adds more value on top of accept() than just the signal
mask handling.

Therefore this patch defines a scaled down version of the sys_paccept
function for those platforms.  It returns -EINVAL in case the signal mask
is non-NULL but behaves the same otherwise.

Note that I explicitly included <linux/thread_info.h>.  I saw that it is
currently included but indirectly two levels down.  There is too much risk
in relying on this.  The header might change and then suddenly the
function definition would change without anyone immediately noticing.

Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Cc: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/net.h |  3 +++
 net/socket.c        | 17 +++++++++++++++++
 2 files changed, 20 insertions(+)

diff --git a/include/linux/net.h b/include/linux/net.h
index 3a9b06d4d0fe..39a23af059b4 100644
--- a/include/linux/net.h
+++ b/include/linux/net.h
@@ -102,6 +102,9 @@ enum sock_type {
 
 /* Flags for socket, socketpair, paccept */
 #define SOCK_CLOEXEC	O_CLOEXEC
+#ifndef SOCK_NONBLOCK
+#define SOCK_NONBLOCK	O_NONBLOCK
+#endif
 
 #endif /* ARCH_HAS_SOCKET_TYPES */
 
diff --git a/net/socket.c b/net/socket.c
index a0ce8ad72252..d163adff95bf 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -69,6 +69,7 @@
 #include <linux/proc_fs.h>
 #include <linux/seq_file.h>
 #include <linux/mutex.h>
+#include <linux/thread_info.h>
 #include <linux/wanrouter.h>
 #include <linux/if_bridge.h>
 #include <linux/if_frad.h>
@@ -1504,6 +1505,7 @@ out_fd:
 	goto out_put;
 }
 
+#ifdef HAVE_SET_RESTORE_SIGMASK
 asmlinkage long sys_paccept(int fd, struct sockaddr __user *upeer_sockaddr,
 			    int __user *upeer_addrlen,
 			    const sigset_t __user *sigmask,
@@ -1541,6 +1543,21 @@ asmlinkage long sys_paccept(int fd, struct sockaddr __user *upeer_sockaddr,
 
 	return ret;
 }
+#else
+asmlinkage long sys_paccept(int fd, struct sockaddr __user *upeer_sockaddr,
+			    int __user *upeer_addrlen,
+			    const sigset_t __user *sigmask,
+			    size_t sigsetsize, int flags)
+{
+	/* The platform does not support restoring the signal mask in the
+	 * return path.  So we do not allow using paccept() with a signal
+	 * mask.  */
+	if (sigmask)
+		return -EINVAL;
+
+	return do_accept(fd, upeer_sockaddr, upeer_addrlen, flags);
+}
+#endif
 
 asmlinkage long sys_accept(int fd, struct sockaddr __user *upeer_sockaddr,
 			   int __user *upeer_addrlen)

From 7d9dbca34240ebb6ff88d8a29c6c7bffd098f0c1 Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Wed, 23 Jul 2008 21:29:22 -0700
Subject: [PATCH 0363/1171] flag parameters: anon_inode_getfd extension

This patch just extends the anon_inode_getfd interface to take an additional
parameter with a flag value.  The flag value is passed on to
get_unused_fd_flags in anticipation for a use with the O_CLOEXEC flag.

No actual semantic changes here, the changed callers all pass 0 for now.

[akpm@linux-foundation.org: KVM fix]
Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/anon_inodes.c            | 9 +++++----
 fs/eventfd.c                | 2 +-
 fs/eventpoll.c              | 2 +-
 fs/signalfd.c               | 3 ++-
 fs/timerfd.c                | 2 +-
 include/linux/anon_inodes.h | 2 +-
 virt/kvm/kvm_main.c         | 4 ++--
 7 files changed, 13 insertions(+), 11 deletions(-)

diff --git a/fs/anon_inodes.c b/fs/anon_inodes.c
index 977ef208c051..1a4eee620b0d 100644
--- a/fs/anon_inodes.c
+++ b/fs/anon_inodes.c
@@ -58,8 +58,9 @@ static struct dentry_operations anon_inodefs_dentry_operations = {
  *                    of the file
  *
  * @name:    [in]    name of the "class" of the new file
- * @fops     [in]    file operations for the new file
- * @priv     [in]    private data for the new file (will be file's private_data)
+ * @fops:    [in]    file operations for the new file
+ * @priv:    [in]    private data for the new file (will be file's private_data)
+ * @flags:   [in]    flags
  *
  * Creates a new file by hooking it on a single inode. This is useful for files
  * that do not need to have a full-fledged inode in order to operate correctly.
@@ -68,7 +69,7 @@ static struct dentry_operations anon_inodefs_dentry_operations = {
  * setup.  Returns new descriptor or -error.
  */
 int anon_inode_getfd(const char *name, const struct file_operations *fops,
-		     void *priv)
+		     void *priv, int flags)
 {
 	struct qstr this;
 	struct dentry *dentry;
@@ -78,7 +79,7 @@ int anon_inode_getfd(const char *name, const struct file_operations *fops,
 	if (IS_ERR(anon_inode_inode))
 		return -ENODEV;
 
-	error = get_unused_fd();
+	error = get_unused_fd_flags(flags);
 	if (error < 0)
 		return error;
 	fd = error;
diff --git a/fs/eventfd.c b/fs/eventfd.c
index 343942deeec1..6094265ca409 100644
--- a/fs/eventfd.c
+++ b/fs/eventfd.c
@@ -214,7 +214,7 @@ asmlinkage long sys_eventfd(unsigned int count)
 	 * When we call this, the initialization must be complete, since
 	 * anon_inode_getfd() will install the fd.
 	 */
-	fd = anon_inode_getfd("[eventfd]", &eventfd_fops, ctx);
+	fd = anon_inode_getfd("[eventfd]", &eventfd_fops, ctx, 0);
 	if (fd < 0)
 		kfree(ctx);
 	return fd;
diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index 990c01d2d66b..9392dd968125 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -1068,7 +1068,7 @@ asmlinkage long sys_epoll_create(int size)
 	 * Creates all the items needed to setup an eventpoll file. That is,
 	 * a file structure and a free file descriptor.
 	 */
-	fd = anon_inode_getfd("[eventpoll]", &eventpoll_fops, ep);
+	fd = anon_inode_getfd("[eventpoll]", &eventpoll_fops, ep, 0);
 	if (fd < 0)
 		ep_free(ep);
 
diff --git a/fs/signalfd.c b/fs/signalfd.c
index 619725644c75..ddb328b74bde 100644
--- a/fs/signalfd.c
+++ b/fs/signalfd.c
@@ -227,7 +227,8 @@ asmlinkage long sys_signalfd(int ufd, sigset_t __user *user_mask, size_t sizemas
 		 * When we call this, the initialization must be complete, since
 		 * anon_inode_getfd() will install the fd.
 		 */
-		ufd = anon_inode_getfd("[signalfd]", &signalfd_fops, ctx);
+		ufd = anon_inode_getfd("[signalfd]", &signalfd_fops, ctx,
+				       0);
 		if (ufd < 0)
 			kfree(ctx);
 	} else {
diff --git a/fs/timerfd.c b/fs/timerfd.c
index d87d354ec424..77c2bc92cbee 100644
--- a/fs/timerfd.c
+++ b/fs/timerfd.c
@@ -198,7 +198,7 @@ asmlinkage long sys_timerfd_create(int clockid, int flags)
 	ctx->clockid = clockid;
 	hrtimer_init(&ctx->tmr, clockid, HRTIMER_MODE_ABS);
 
-	ufd = anon_inode_getfd("[timerfd]", &timerfd_fops, ctx);
+	ufd = anon_inode_getfd("[timerfd]", &timerfd_fops, ctx, 0);
 	if (ufd < 0)
 		kfree(ctx);
 
diff --git a/include/linux/anon_inodes.h b/include/linux/anon_inodes.h
index 6129e58ca7c9..e0a0cdc2da43 100644
--- a/include/linux/anon_inodes.h
+++ b/include/linux/anon_inodes.h
@@ -9,7 +9,7 @@
 #define _LINUX_ANON_INODES_H
 
 int anon_inode_getfd(const char *name, const struct file_operations *fops,
-		     void *priv);
+		     void *priv, int flags);
 
 #endif /* _LINUX_ANON_INODES_H */
 
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 904d7b7bd780..a845890b6800 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -902,7 +902,7 @@ static const struct file_operations kvm_vcpu_fops = {
  */
 static int create_vcpu_fd(struct kvm_vcpu *vcpu)
 {
-	int fd = anon_inode_getfd("kvm-vcpu", &kvm_vcpu_fops, vcpu);
+	int fd = anon_inode_getfd("kvm-vcpu", &kvm_vcpu_fops, vcpu, 0);
 	if (fd < 0)
 		kvm_put_kvm(vcpu->kvm);
 	return fd;
@@ -1261,7 +1261,7 @@ static int kvm_dev_ioctl_create_vm(void)
 	kvm = kvm_create_vm();
 	if (IS_ERR(kvm))
 		return PTR_ERR(kvm);
-	fd = anon_inode_getfd("kvm-vm", &kvm_vm_fops, kvm);
+	fd = anon_inode_getfd("kvm-vm", &kvm_vm_fops, kvm, 0);
 	if (fd < 0)
 		kvm_put_kvm(kvm);
 

From 9deb27baedb79759c3ab9435a7d8b841842d56e9 Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Wed, 23 Jul 2008 21:29:24 -0700
Subject: [PATCH 0364/1171] flag parameters: signalfd

This patch adds the new signalfd4 syscall.  It extends the old signalfd
syscall by one parameter which is meant to hold a flag value.  In this
patch the only flag support is SFD_CLOEXEC which causes the close-on-exec
flag for the returned file descriptor to be set.

A new name SFD_CLOEXEC is introduced which in this implementation must
have the same value as O_CLOEXEC.

The following test must be adjusted for architectures other than x86 and
x86-64 and in case the syscall numbers changed.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#include <fcntl.h>
#include <signal.h>
#include <stdio.h>
#include <unistd.h>
#include <sys/syscall.h>

#ifndef __NR_signalfd4
# ifdef __x86_64__
#  define __NR_signalfd4 289
# elif defined __i386__
#  define __NR_signalfd4 327
# else
#  error "need __NR_signalfd4"
# endif
#endif

#define SFD_CLOEXEC O_CLOEXEC

int
main (void)
{
  sigset_t ss;
  sigemptyset (&ss);
  sigaddset (&ss, SIGUSR1);
  int fd = syscall (__NR_signalfd4, -1, &ss, 8, 0);
  if (fd == -1)
    {
      puts ("signalfd4(0) failed");
      return 1;
    }
  int coe = fcntl (fd, F_GETFD);
  if (coe == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if (coe & FD_CLOEXEC)
    {
      puts ("signalfd4(0) set close-on-exec flag");
      return 1;
    }
  close (fd);

  fd = syscall (__NR_signalfd4, -1, &ss, 8, SFD_CLOEXEC);
  if (fd == -1)
    {
      puts ("signalfd4(SFD_CLOEXEC) failed");
      return 1;
    }
  coe = fcntl (fd, F_GETFD);
  if (coe == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if ((coe & FD_CLOEXEC) == 0)
    {
      puts ("signalfd4(SFD_CLOEXEC) does not set close-on-exec flag");
      return 1;
    }
  close (fd);

  puts ("OK");

  return 0;
}
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

[akpm@linux-foundation.org: add sys_ni stub]
Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/ia32/ia32entry.S          |  1 +
 arch/x86/kernel/syscall_table_32.S |  1 +
 fs/compat.c                        | 14 ++++++++++----
 fs/signalfd.c                      | 14 ++++++++++++--
 include/asm-x86/unistd_32.h        |  1 +
 include/asm-x86/unistd_64.h        |  2 ++
 include/linux/signalfd.h           |  5 +++++
 include/linux/syscalls.h           |  1 +
 kernel/sys_ni.c                    |  1 +
 9 files changed, 34 insertions(+), 6 deletions(-)

diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S
index 021d71bc69b5..c308128b9251 100644
--- a/arch/x86/ia32/ia32entry.S
+++ b/arch/x86/ia32/ia32entry.S
@@ -826,4 +826,5 @@ ia32_sys_call_table:
 	.quad sys32_fallocate
 	.quad compat_sys_timerfd_settime	/* 325 */
 	.quad compat_sys_timerfd_gettime
+	.quad compat_sys_signalfd4
 ia32_syscall_end:
diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S
index adff5562f5fd..c12a36c9fd51 100644
--- a/arch/x86/kernel/syscall_table_32.S
+++ b/arch/x86/kernel/syscall_table_32.S
@@ -326,3 +326,4 @@ ENTRY(sys_call_table)
 	.long sys_fallocate
 	.long sys_timerfd_settime	/* 325 */
 	.long sys_timerfd_gettime
+	.long sys_signalfd4
diff --git a/fs/compat.c b/fs/compat.c
index b46604281766..106eba28ec5a 100644
--- a/fs/compat.c
+++ b/fs/compat.c
@@ -2131,9 +2131,9 @@ asmlinkage long compat_sys_epoll_pwait(int epfd,
 
 #ifdef CONFIG_SIGNALFD
 
-asmlinkage long compat_sys_signalfd(int ufd,
-				    const compat_sigset_t __user *sigmask,
-				    compat_size_t sigsetsize)
+asmlinkage long compat_sys_signalfd4(int ufd,
+				     const compat_sigset_t __user *sigmask,
+				     compat_size_t sigsetsize, int flags)
 {
 	compat_sigset_t ss32;
 	sigset_t tmp;
@@ -2148,9 +2148,15 @@ asmlinkage long compat_sys_signalfd(int ufd,
 	if (copy_to_user(ksigmask, &tmp, sizeof(sigset_t)))
 		return -EFAULT;
 
-	return sys_signalfd(ufd, ksigmask, sizeof(sigset_t));
+	return sys_signalfd4(ufd, ksigmask, sizeof(sigset_t), flags);
 }
 
+asmlinkage long compat_sys_signalfd(int ufd,
+				    const compat_sigset_t __user *sigmask,
+				    compat_size_t sigsetsize)
+{
+	return compat_sys_signalfd4(ufd, sigmask, sigsetsize, 0);
+}
 #endif /* CONFIG_SIGNALFD */
 
 #ifdef CONFIG_TIMERFD
diff --git a/fs/signalfd.c b/fs/signalfd.c
index ddb328b74bde..c8609fa51a13 100644
--- a/fs/signalfd.c
+++ b/fs/signalfd.c
@@ -205,11 +205,15 @@ static const struct file_operations signalfd_fops = {
 	.read		= signalfd_read,
 };
 
-asmlinkage long sys_signalfd(int ufd, sigset_t __user *user_mask, size_t sizemask)
+asmlinkage long sys_signalfd4(int ufd, sigset_t __user *user_mask,
+			      size_t sizemask, int flags)
 {
 	sigset_t sigmask;
 	struct signalfd_ctx *ctx;
 
+	if (flags & ~SFD_CLOEXEC)
+		return -EINVAL;
+
 	if (sizemask != sizeof(sigset_t) ||
 	    copy_from_user(&sigmask, user_mask, sizeof(sigmask)))
 		return -EINVAL;
@@ -228,7 +232,7 @@ asmlinkage long sys_signalfd(int ufd, sigset_t __user *user_mask, size_t sizemas
 		 * anon_inode_getfd() will install the fd.
 		 */
 		ufd = anon_inode_getfd("[signalfd]", &signalfd_fops, ctx,
-				       0);
+				       flags & O_CLOEXEC);
 		if (ufd < 0)
 			kfree(ctx);
 	} else {
@@ -250,3 +254,9 @@ asmlinkage long sys_signalfd(int ufd, sigset_t __user *user_mask, size_t sizemas
 
 	return ufd;
 }
+
+asmlinkage long sys_signalfd(int ufd, sigset_t __user *user_mask,
+			     size_t sizemask)
+{
+	return sys_signalfd4(ufd, user_mask, sizemask, 0);
+}
diff --git a/include/asm-x86/unistd_32.h b/include/asm-x86/unistd_32.h
index 8317d94771d3..c310371f5613 100644
--- a/include/asm-x86/unistd_32.h
+++ b/include/asm-x86/unistd_32.h
@@ -332,6 +332,7 @@
 #define __NR_fallocate		324
 #define __NR_timerfd_settime	325
 #define __NR_timerfd_gettime	326
+#define __NR_signalfd4		327
 
 #ifdef __KERNEL__
 
diff --git a/include/asm-x86/unistd_64.h b/include/asm-x86/unistd_64.h
index e323994a370f..e0a9b45b2346 100644
--- a/include/asm-x86/unistd_64.h
+++ b/include/asm-x86/unistd_64.h
@@ -641,6 +641,8 @@ __SYSCALL(__NR_timerfd_settime, sys_timerfd_settime)
 __SYSCALL(__NR_timerfd_gettime, sys_timerfd_gettime)
 #define __NR_paccept				288
 __SYSCALL(__NR_paccept, sys_paccept)
+#define __NR_signalfd4				289
+__SYSCALL(__NR_signalfd4, sys_signalfd4)
 
 
 #ifndef __NO_STUBS
diff --git a/include/linux/signalfd.h b/include/linux/signalfd.h
index ea037f28df91..8b3f7b7420a1 100644
--- a/include/linux/signalfd.h
+++ b/include/linux/signalfd.h
@@ -8,6 +8,11 @@
 #ifndef _LINUX_SIGNALFD_H
 #define _LINUX_SIGNALFD_H
 
+/* For O_CLOEXEC */
+#include <linux/fcntl.h>
+
+/* Flags for signalfd4.  */
+#define SFD_CLOEXEC O_CLOEXEC
 
 struct signalfd_siginfo {
 	__u32 ssi_signo;
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 2a2a40af6b2c..1c2707797845 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -610,6 +610,7 @@ asmlinkage long sys_set_robust_list(struct robust_list_head __user *head,
 				    size_t len);
 asmlinkage long sys_getcpu(unsigned __user *cpu, unsigned __user *node, struct getcpu_cache __user *cache);
 asmlinkage long sys_signalfd(int ufd, sigset_t __user *user_mask, size_t sizemask);
+asmlinkage long sys_signalfd4(int ufd, sigset_t __user *user_mask, size_t sizemask, int flags);
 asmlinkage long sys_timerfd_create(int clockid, int flags);
 asmlinkage long sys_timerfd_settime(int ufd, int flags,
 				    const struct itimerspec __user *utmr,
diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c
index 2f0b8a2e600f..8627c89ae9e8 100644
--- a/kernel/sys_ni.c
+++ b/kernel/sys_ni.c
@@ -156,6 +156,7 @@ cond_syscall(sys_ioprio_get);
 
 /* New file descriptors */
 cond_syscall(sys_signalfd);
+cond_syscall(sys_signalfd4);
 cond_syscall(compat_sys_signalfd);
 cond_syscall(sys_timerfd_create);
 cond_syscall(sys_timerfd_settime);

From b087498eb5605673b0f260a7620d91818cd72304 Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Wed, 23 Jul 2008 21:29:25 -0700
Subject: [PATCH 0365/1171] flag parameters: eventfd

This patch adds the new eventfd2 syscall.  It extends the old eventfd
syscall by one parameter which is meant to hold a flag value.  In this
patch the only flag support is EFD_CLOEXEC which causes the close-on-exec
flag for the returned file descriptor to be set.

A new name EFD_CLOEXEC is introduced which in this implementation must
have the same value as O_CLOEXEC.

The following test must be adjusted for architectures other than x86 and
x86-64 and in case the syscall numbers changed.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#include <fcntl.h>
#include <stdio.h>
#include <unistd.h>
#include <sys/syscall.h>

#ifndef __NR_eventfd2
# ifdef __x86_64__
#  define __NR_eventfd2 290
# elif defined __i386__
#  define __NR_eventfd2 328
# else
#  error "need __NR_eventfd2"
# endif
#endif

#define EFD_CLOEXEC O_CLOEXEC

int
main (void)
{
  int fd = syscall (__NR_eventfd2, 1, 0);
  if (fd == -1)
    {
      puts ("eventfd2(0) failed");
      return 1;
    }
  int coe = fcntl (fd, F_GETFD);
  if (coe == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if (coe & FD_CLOEXEC)
    {
      puts ("eventfd2(0) sets close-on-exec flag");
      return 1;
    }
  close (fd);

  fd = syscall (__NR_eventfd2, 1, EFD_CLOEXEC);
  if (fd == -1)
    {
      puts ("eventfd2(EFD_CLOEXEC) failed");
      return 1;
    }
  coe = fcntl (fd, F_GETFD);
  if (coe == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if ((coe & FD_CLOEXEC) == 0)
    {
      puts ("eventfd2(EFD_CLOEXEC) does not set close-on-exec flag");
      return 1;
    }
  close (fd);

  puts ("OK");

  return 0;
}
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

[akpm@linux-foundation.org: add sys_ni stub]
Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/ia32/ia32entry.S          |  1 +
 arch/x86/kernel/syscall_table_32.S |  1 +
 fs/eventfd.c                       | 13 +++++++++++--
 include/asm-x86/unistd_32.h        |  1 +
 include/asm-x86/unistd_64.h        |  2 ++
 include/linux/eventfd.h            |  6 ++++++
 include/linux/syscalls.h           |  1 +
 kernel/sys_ni.c                    |  1 +
 8 files changed, 24 insertions(+), 2 deletions(-)

diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S
index c308128b9251..cf0eb31745ca 100644
--- a/arch/x86/ia32/ia32entry.S
+++ b/arch/x86/ia32/ia32entry.S
@@ -827,4 +827,5 @@ ia32_sys_call_table:
 	.quad compat_sys_timerfd_settime	/* 325 */
 	.quad compat_sys_timerfd_gettime
 	.quad compat_sys_signalfd4
+	.quad sys_eventfd2
 ia32_syscall_end:
diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S
index c12a36c9fd51..cf112cb11c37 100644
--- a/arch/x86/kernel/syscall_table_32.S
+++ b/arch/x86/kernel/syscall_table_32.S
@@ -327,3 +327,4 @@ ENTRY(sys_call_table)
 	.long sys_timerfd_settime	/* 325 */
 	.long sys_timerfd_gettime
 	.long sys_signalfd4
+	.long sys_eventfd2
diff --git a/fs/eventfd.c b/fs/eventfd.c
index 6094265ca409..bd420e6478ad 100644
--- a/fs/eventfd.c
+++ b/fs/eventfd.c
@@ -198,11 +198,14 @@ struct file *eventfd_fget(int fd)
 	return file;
 }
 
-asmlinkage long sys_eventfd(unsigned int count)
+asmlinkage long sys_eventfd2(unsigned int count, int flags)
 {
 	int fd;
 	struct eventfd_ctx *ctx;
 
+	if (flags & ~EFD_CLOEXEC)
+		return -EINVAL;
+
 	ctx = kmalloc(sizeof(*ctx), GFP_KERNEL);
 	if (!ctx)
 		return -ENOMEM;
@@ -214,9 +217,15 @@ asmlinkage long sys_eventfd(unsigned int count)
 	 * When we call this, the initialization must be complete, since
 	 * anon_inode_getfd() will install the fd.
 	 */
-	fd = anon_inode_getfd("[eventfd]", &eventfd_fops, ctx, 0);
+	fd = anon_inode_getfd("[eventfd]", &eventfd_fops, ctx,
+			      flags & O_CLOEXEC);
 	if (fd < 0)
 		kfree(ctx);
 	return fd;
 }
 
+asmlinkage long sys_eventfd(unsigned int count)
+{
+	return sys_eventfd2(count, 0);
+}
+
diff --git a/include/asm-x86/unistd_32.h b/include/asm-x86/unistd_32.h
index c310371f5613..edbd8723c939 100644
--- a/include/asm-x86/unistd_32.h
+++ b/include/asm-x86/unistd_32.h
@@ -333,6 +333,7 @@
 #define __NR_timerfd_settime	325
 #define __NR_timerfd_gettime	326
 #define __NR_signalfd4		327
+#define __NR_eventfd2		328
 
 #ifdef __KERNEL__
 
diff --git a/include/asm-x86/unistd_64.h b/include/asm-x86/unistd_64.h
index e0a9b45b2346..fb059a6feeb1 100644
--- a/include/asm-x86/unistd_64.h
+++ b/include/asm-x86/unistd_64.h
@@ -643,6 +643,8 @@ __SYSCALL(__NR_timerfd_gettime, sys_timerfd_gettime)
 __SYSCALL(__NR_paccept, sys_paccept)
 #define __NR_signalfd4				289
 __SYSCALL(__NR_signalfd4, sys_signalfd4)
+#define __NR_eventfd2				290
+__SYSCALL(__NR_eventfd2, sys_eventfd2)
 
 
 #ifndef __NO_STUBS
diff --git a/include/linux/eventfd.h b/include/linux/eventfd.h
index a701399b7fed..a6c0eaedb1b0 100644
--- a/include/linux/eventfd.h
+++ b/include/linux/eventfd.h
@@ -10,6 +10,12 @@
 
 #ifdef CONFIG_EVENTFD
 
+/* For O_CLOEXEC */
+#include <linux/fcntl.h>
+
+/* Flags for eventfd2.  */
+#define EFD_CLOEXEC O_CLOEXEC
+
 struct file *eventfd_fget(int fd);
 int eventfd_signal(struct file *file, int n);
 
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 1c2707797845..9ab09926a7f2 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -617,6 +617,7 @@ asmlinkage long sys_timerfd_settime(int ufd, int flags,
 				    struct itimerspec __user *otmr);
 asmlinkage long sys_timerfd_gettime(int ufd, struct itimerspec __user *otmr);
 asmlinkage long sys_eventfd(unsigned int count);
+asmlinkage long sys_eventfd2(unsigned int count, int flags);
 asmlinkage long sys_fallocate(int fd, int mode, loff_t offset, loff_t len);
 
 int kernel_execve(const char *filename, char *const argv[], char *const envp[]);
diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c
index 8627c89ae9e8..2a361ccdc7ca 100644
--- a/kernel/sys_ni.c
+++ b/kernel/sys_ni.c
@@ -164,3 +164,4 @@ cond_syscall(sys_timerfd_gettime);
 cond_syscall(compat_sys_timerfd_settime);
 cond_syscall(compat_sys_timerfd_gettime);
 cond_syscall(sys_eventfd);
+cond_syscall(sys_eventfd2);

From 11fcb6c14676023d0bd437841f5dcd670e7990a0 Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Wed, 23 Jul 2008 21:29:26 -0700
Subject: [PATCH 0366/1171] flag parameters: timerfd_create

The timerfd_create syscall already has a flags parameter.  It just is
unused so far.  This patch changes this by introducing the TFD_CLOEXEC
flag to set the close-on-exec flag for the returned file descriptor.

A new name TFD_CLOEXEC is introduced which in this implementation must
have the same value as O_CLOEXEC.

The following test must be adjusted for architectures other than x86 and
x86-64 and in case the syscall numbers changed.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#include <fcntl.h>
#include <stdio.h>
#include <time.h>
#include <unistd.h>
#include <sys/syscall.h>

#ifndef __NR_timerfd_create
# ifdef __x86_64__
#  define __NR_timerfd_create 283
# elif defined __i386__
#  define __NR_timerfd_create 322
# else
#  error "need __NR_timerfd_create"
# endif
#endif

#define TFD_CLOEXEC O_CLOEXEC

int
main (void)
{
  int fd = syscall (__NR_timerfd_create, CLOCK_REALTIME, 0);
  if (fd == -1)
    {
      puts ("timerfd_create(0) failed");
      return 1;
    }
  int coe = fcntl (fd, F_GETFD);
  if (coe == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if (coe & FD_CLOEXEC)
    {
      puts ("timerfd_create(0) set close-on-exec flag");
      return 1;
    }
  close (fd);

  fd = syscall (__NR_timerfd_create, CLOCK_REALTIME, TFD_CLOEXEC);
  if (fd == -1)
    {
      puts ("timerfd_create(TFD_CLOEXEC) failed");
      return 1;
    }
  coe = fcntl (fd, F_GETFD);
  if (coe == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if ((coe & FD_CLOEXEC) == 0)
    {
      puts ("timerfd_create(TFD_CLOEXEC) set close-on-exec flag");
      return 1;
    }
  close (fd);

  puts ("OK");

  return 0;
}
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/timerfd.c            | 5 +++--
 include/linux/timerfd.h | 5 +++++
 2 files changed, 8 insertions(+), 2 deletions(-)

diff --git a/fs/timerfd.c b/fs/timerfd.c
index 77c2bc92cbee..c6ef5e33cb34 100644
--- a/fs/timerfd.c
+++ b/fs/timerfd.c
@@ -184,7 +184,7 @@ asmlinkage long sys_timerfd_create(int clockid, int flags)
 	int ufd;
 	struct timerfd_ctx *ctx;
 
-	if (flags)
+	if (flags & ~TFD_CLOEXEC)
 		return -EINVAL;
 	if (clockid != CLOCK_MONOTONIC &&
 	    clockid != CLOCK_REALTIME)
@@ -198,7 +198,8 @@ asmlinkage long sys_timerfd_create(int clockid, int flags)
 	ctx->clockid = clockid;
 	hrtimer_init(&ctx->tmr, clockid, HRTIMER_MODE_ABS);
 
-	ufd = anon_inode_getfd("[timerfd]", &timerfd_fops, ctx, 0);
+	ufd = anon_inode_getfd("[timerfd]", &timerfd_fops, ctx,
+			       flags & O_CLOEXEC);
 	if (ufd < 0)
 		kfree(ctx);
 
diff --git a/include/linux/timerfd.h b/include/linux/timerfd.h
index cf2b10d75731..96ed97dff00f 100644
--- a/include/linux/timerfd.h
+++ b/include/linux/timerfd.h
@@ -8,9 +8,14 @@
 #ifndef _LINUX_TIMERFD_H
 #define _LINUX_TIMERFD_H
 
+/* For O_CLOEXEC */
+#include <linux/fcntl.h>
 
+/* Flags for timerfd_settime.  */
 #define TFD_TIMER_ABSTIME (1 << 0)
 
+/* Flags for timerfd_create.  */
+#define TFD_CLOEXEC O_CLOEXEC
 
 
 #endif /* _LINUX_TIMERFD_H */

From a0998b50c3f0b8fdd265c63e0032f86ebe377dbf Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Wed, 23 Jul 2008 21:29:27 -0700
Subject: [PATCH 0367/1171] flag parameters: epoll_create

This patch adds the new epoll_create2 syscall.  It extends the old epoll_create
syscall by one parameter which is meant to hold a flag value.  In this
patch the only flag support is EPOLL_CLOEXEC which causes the close-on-exec
flag for the returned file descriptor to be set.

A new name EPOLL_CLOEXEC is introduced which in this implementation must
have the same value as O_CLOEXEC.

The following test must be adjusted for architectures other than x86 and
x86-64 and in case the syscall numbers changed.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#include <fcntl.h>
#include <stdio.h>
#include <time.h>
#include <unistd.h>
#include <sys/syscall.h>

#ifndef __NR_epoll_create2
# ifdef __x86_64__
#  define __NR_epoll_create2 291
# elif defined __i386__
#  define __NR_epoll_create2 329
# else
#  error "need __NR_epoll_create2"
# endif
#endif

#define EPOLL_CLOEXEC O_CLOEXEC

int
main (void)
{
  int fd = syscall (__NR_epoll_create2, 1, 0);
  if (fd == -1)
    {
      puts ("epoll_create2(0) failed");
      return 1;
    }
  int coe = fcntl (fd, F_GETFD);
  if (coe == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if (coe & FD_CLOEXEC)
    {
      puts ("epoll_create2(0) set close-on-exec flag");
      return 1;
    }
  close (fd);

  fd = syscall (__NR_epoll_create2, 1, EPOLL_CLOEXEC);
  if (fd == -1)
    {
      puts ("epoll_create2(EPOLL_CLOEXEC) failed");
      return 1;
    }
  coe = fcntl (fd, F_GETFD);
  if (coe == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if ((coe & FD_CLOEXEC) == 0)
    {
      puts ("epoll_create2(EPOLL_CLOEXEC) set close-on-exec flag");
      return 1;
    }
  close (fd);

  puts ("OK");

  return 0;
}
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/ia32/ia32entry.S          |  1 +
 arch/x86/kernel/syscall_table_32.S |  1 +
 fs/eventpoll.c                     | 13 +++++++++++--
 include/asm-x86/unistd_32.h        |  1 +
 include/asm-x86/unistd_64.h        |  2 ++
 include/linux/eventpoll.h          |  4 ++++
 include/linux/syscalls.h           |  1 +
 7 files changed, 21 insertions(+), 2 deletions(-)

diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S
index cf0eb31745ca..04366f08f424 100644
--- a/arch/x86/ia32/ia32entry.S
+++ b/arch/x86/ia32/ia32entry.S
@@ -828,4 +828,5 @@ ia32_sys_call_table:
 	.quad compat_sys_timerfd_gettime
 	.quad compat_sys_signalfd4
 	.quad sys_eventfd2
+	.quad sys_epoll_create2
 ia32_syscall_end:
diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S
index cf112cb11c37..4d7007ca263d 100644
--- a/arch/x86/kernel/syscall_table_32.S
+++ b/arch/x86/kernel/syscall_table_32.S
@@ -328,3 +328,4 @@ ENTRY(sys_call_table)
 	.long sys_timerfd_gettime
 	.long sys_signalfd4
 	.long sys_eventfd2
+	.long sys_epoll_create2
diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index 9392dd968125..3fd4014f3c5a 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -1046,11 +1046,14 @@ retry:
  * RB tree. With the current implementation, the "size" parameter is ignored
  * (besides sanity checks).
  */
-asmlinkage long sys_epoll_create(int size)
+asmlinkage long sys_epoll_create2(int size, int flags)
 {
 	int error, fd = -1;
 	struct eventpoll *ep;
 
+	if (flags & ~EPOLL_CLOEXEC)
+		return -EINVAL;
+
 	DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_create(%d)\n",
 		     current, size));
 
@@ -1068,7 +1071,8 @@ asmlinkage long sys_epoll_create(int size)
 	 * Creates all the items needed to setup an eventpoll file. That is,
 	 * a file structure and a free file descriptor.
 	 */
-	fd = anon_inode_getfd("[eventpoll]", &eventpoll_fops, ep, 0);
+	fd = anon_inode_getfd("[eventpoll]", &eventpoll_fops, ep,
+			      flags & O_CLOEXEC);
 	if (fd < 0)
 		ep_free(ep);
 
@@ -1079,6 +1083,11 @@ error_return:
 	return fd;
 }
 
+asmlinkage long sys_epoll_create(int size)
+{
+	return sys_epoll_create2(size, 0);
+}
+
 /*
  * The following function implements the controller interface for
  * the eventpoll file that enables the insertion/removal/change of
diff --git a/include/asm-x86/unistd_32.h b/include/asm-x86/unistd_32.h
index edbd8723c939..a37d6b0c4e1e 100644
--- a/include/asm-x86/unistd_32.h
+++ b/include/asm-x86/unistd_32.h
@@ -334,6 +334,7 @@
 #define __NR_timerfd_gettime	326
 #define __NR_signalfd4		327
 #define __NR_eventfd2		328
+#define __NR_epoll_create2	329
 
 #ifdef __KERNEL__
 
diff --git a/include/asm-x86/unistd_64.h b/include/asm-x86/unistd_64.h
index fb059a6feeb1..a1a4a5b6e5ee 100644
--- a/include/asm-x86/unistd_64.h
+++ b/include/asm-x86/unistd_64.h
@@ -645,6 +645,8 @@ __SYSCALL(__NR_paccept, sys_paccept)
 __SYSCALL(__NR_signalfd4, sys_signalfd4)
 #define __NR_eventfd2				290
 __SYSCALL(__NR_eventfd2, sys_eventfd2)
+#define __NR_epoll_create2			291
+__SYSCALL(__NR_epoll_create2, sys_epoll_create2)
 
 
 #ifndef __NO_STUBS
diff --git a/include/linux/eventpoll.h b/include/linux/eventpoll.h
index cf79853967ff..1cfaa40059c8 100644
--- a/include/linux/eventpoll.h
+++ b/include/linux/eventpoll.h
@@ -14,8 +14,12 @@
 #ifndef _LINUX_EVENTPOLL_H
 #define _LINUX_EVENTPOLL_H
 
+/* For O_CLOEXEC */
+#include <linux/fcntl.h>
 #include <linux/types.h>
 
+/* Flags for epoll_create2.  */
+#define EPOLL_CLOEXEC O_CLOEXEC
 
 /* Valid opcodes to issue to sys_epoll_ctl() */
 #define EPOLL_CTL_ADD 1
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 9ab09926a7f2..85953240f28c 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -430,6 +430,7 @@ asmlinkage long sys_poll(struct pollfd __user *ufds, unsigned int nfds,
 asmlinkage long sys_select(int n, fd_set __user *inp, fd_set __user *outp,
 			fd_set __user *exp, struct timeval __user *tvp);
 asmlinkage long sys_epoll_create(int size);
+asmlinkage long sys_epoll_create2(int size, int flags);
 asmlinkage long sys_epoll_ctl(int epfd, int op, int fd,
 				struct epoll_event __user *event);
 asmlinkage long sys_epoll_wait(int epfd, struct epoll_event __user *events,

From 336dd1f70ff62d7dd8655228caed4c5bfc818c56 Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Wed, 23 Jul 2008 21:29:29 -0700
Subject: [PATCH 0368/1171] flag parameters: dup2

This patch adds the new dup3 syscall.  It extends the old dup2 syscall by one
parameter which is meant to hold a flag value.  Support for the O_CLOEXEC flag
is added in this patch.

The following test must be adjusted for architectures other than x86 and
x86-64 and in case the syscall numbers changed.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#include <fcntl.h>
#include <stdio.h>
#include <time.h>
#include <unistd.h>
#include <sys/syscall.h>

#ifndef __NR_dup3
# ifdef __x86_64__
#  define __NR_dup3 292
# elif defined __i386__
#  define __NR_dup3 330
# else
#  error "need __NR_dup3"
# endif
#endif

int
main (void)
{
  int fd = syscall (__NR_dup3, 1, 4, 0);
  if (fd == -1)
    {
      puts ("dup3(0) failed");
      return 1;
    }
  int coe = fcntl (fd, F_GETFD);
  if (coe == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if (coe & FD_CLOEXEC)
    {
      puts ("dup3(0) set close-on-exec flag");
      return 1;
    }
  close (fd);

  fd = syscall (__NR_dup3, 1, 4, O_CLOEXEC);
  if (fd == -1)
    {
      puts ("dup3(O_CLOEXEC) failed");
      return 1;
    }
  coe = fcntl (fd, F_GETFD);
  if (coe == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if ((coe & FD_CLOEXEC) == 0)
    {
      puts ("dup3(O_CLOEXEC) set close-on-exec flag");
      return 1;
    }
  close (fd);

  puts ("OK");

  return 0;
}
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/ia32/ia32entry.S          |  1 +
 arch/x86/kernel/syscall_table_32.S |  1 +
 fs/fcntl.c                         | 15 +++++++++++++--
 include/asm-x86/unistd_32.h        |  1 +
 include/asm-x86/unistd_64.h        |  2 ++
 include/linux/syscalls.h           |  1 +
 6 files changed, 19 insertions(+), 2 deletions(-)

diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S
index 04366f08f424..5614a8f7bed4 100644
--- a/arch/x86/ia32/ia32entry.S
+++ b/arch/x86/ia32/ia32entry.S
@@ -829,4 +829,5 @@ ia32_sys_call_table:
 	.quad compat_sys_signalfd4
 	.quad sys_eventfd2
 	.quad sys_epoll_create2
+	.quad sys_dup3			/* 330 */
 ia32_syscall_end:
diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S
index 4d7007ca263d..24a3f1ea6a0e 100644
--- a/arch/x86/kernel/syscall_table_32.S
+++ b/arch/x86/kernel/syscall_table_32.S
@@ -329,3 +329,4 @@ ENTRY(sys_call_table)
 	.long sys_signalfd4
 	.long sys_eventfd2
 	.long sys_epoll_create2
+	.long sys_dup3			/* 330 */
diff --git a/fs/fcntl.c b/fs/fcntl.c
index 330a7d782591..9679fcbdeaa0 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
@@ -125,13 +125,16 @@ static int dupfd(struct file *file, unsigned int start, int cloexec)
 	return fd;
 }
 
-asmlinkage long sys_dup2(unsigned int oldfd, unsigned int newfd)
+asmlinkage long sys_dup3(unsigned int oldfd, unsigned int newfd, int flags)
 {
 	int err = -EBADF;
 	struct file * file, *tofree;
 	struct files_struct * files = current->files;
 	struct fdtable *fdt;
 
+	if ((flags & ~O_CLOEXEC) != 0)
+		return -EINVAL;
+
 	spin_lock(&files->file_lock);
 	if (!(file = fcheck(oldfd)))
 		goto out_unlock;
@@ -163,7 +166,10 @@ asmlinkage long sys_dup2(unsigned int oldfd, unsigned int newfd)
 
 	rcu_assign_pointer(fdt->fd[newfd], file);
 	FD_SET(newfd, fdt->open_fds);
-	FD_CLR(newfd, fdt->close_on_exec);
+	if (flags & O_CLOEXEC)
+		FD_SET(newfd, fdt->close_on_exec);
+	else
+		FD_CLR(newfd, fdt->close_on_exec);
 	spin_unlock(&files->file_lock);
 
 	if (tofree)
@@ -181,6 +187,11 @@ out_fput:
 	goto out;
 }
 
+asmlinkage long sys_dup2(unsigned int oldfd, unsigned int newfd)
+{
+	return sys_dup3(oldfd, newfd, 0);
+}
+
 asmlinkage long sys_dup(unsigned int fildes)
 {
 	int ret = -EBADF;
diff --git a/include/asm-x86/unistd_32.h b/include/asm-x86/unistd_32.h
index a37d6b0c4e1e..a1f6383bf695 100644
--- a/include/asm-x86/unistd_32.h
+++ b/include/asm-x86/unistd_32.h
@@ -335,6 +335,7 @@
 #define __NR_signalfd4		327
 #define __NR_eventfd2		328
 #define __NR_epoll_create2	329
+#define __NR_dup3		330
 
 #ifdef __KERNEL__
 
diff --git a/include/asm-x86/unistd_64.h b/include/asm-x86/unistd_64.h
index a1a4a5b6e5ee..f0fb2bd40cdb 100644
--- a/include/asm-x86/unistd_64.h
+++ b/include/asm-x86/unistd_64.h
@@ -647,6 +647,8 @@ __SYSCALL(__NR_signalfd4, sys_signalfd4)
 __SYSCALL(__NR_eventfd2, sys_eventfd2)
 #define __NR_epoll_create2			291
 __SYSCALL(__NR_epoll_create2, sys_epoll_create2)
+#define __NR_dup3				292
+__SYSCALL(__NR_dup3, sys_dup3)
 
 
 #ifndef __NO_STUBS
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 85953240f28c..034d3358549e 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -305,6 +305,7 @@ asmlinkage long sys_fcntl64(unsigned int fd,
 #endif
 asmlinkage long sys_dup(unsigned int fildes);
 asmlinkage long sys_dup2(unsigned int oldfd, unsigned int newfd);
+asmlinkage long sys_dup3(unsigned int oldfd, unsigned int newfd, int flags);
 asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int on);
 asmlinkage long sys_ioctl(unsigned int fd, unsigned int cmd,
 				unsigned long arg);

From ed8cae8ba01348bfd83333f4648dd807b04d7f08 Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Wed, 23 Jul 2008 21:29:30 -0700
Subject: [PATCH 0369/1171] flag parameters: pipe

This patch introduces the new syscall pipe2 which is like pipe but it also
takes an additional parameter which takes a flag value.  This patch implements
the handling of O_CLOEXEC for the flag.  I did not add support for the new
syscall for the architectures which have a special sys_pipe implementation.  I
think the maintainers of those archs have the chance to go with the unified
implementation but that's up to them.

The implementation introduces do_pipe_flags.  I did that instead of changing
all callers of do_pipe because some of the callers are written in assembler.
I would probably screw up changing the assembly code.  To avoid breaking code
do_pipe is now a small wrapper around do_pipe_flags.  Once all callers are
changed over to do_pipe_flags the old do_pipe function can be removed.

The following test must be adjusted for architectures other than x86 and
x86-64 and in case the syscall numbers changed.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#include <fcntl.h>
#include <stdio.h>
#include <unistd.h>
#include <sys/syscall.h>

#ifndef __NR_pipe2
# ifdef __x86_64__
#  define __NR_pipe2 293
# elif defined __i386__
#  define __NR_pipe2 331
# else
#  error "need __NR_pipe2"
# endif
#endif

int
main (void)
{
  int fd[2];
  if (syscall (__NR_pipe2, fd, 0) != 0)
    {
      puts ("pipe2(0) failed");
      return 1;
    }
  for (int i = 0; i < 2; ++i)
    {
      int coe = fcntl (fd[i], F_GETFD);
      if (coe == -1)
        {
          puts ("fcntl failed");
          return 1;
        }
      if (coe & FD_CLOEXEC)
        {
          printf ("pipe2(0) set close-on-exit for fd[%d]\n", i);
          return 1;
        }
    }
  close (fd[0]);
  close (fd[1]);

  if (syscall (__NR_pipe2, fd, O_CLOEXEC) != 0)
    {
      puts ("pipe2(O_CLOEXEC) failed");
      return 1;
    }
  for (int i = 0; i < 2; ++i)
    {
      int coe = fcntl (fd[i], F_GETFD);
      if (coe == -1)
        {
          puts ("fcntl failed");
          return 1;
        }
      if ((coe & FD_CLOEXEC) == 0)
        {
          printf ("pipe2(O_CLOEXEC) does not set close-on-exit for fd[%d]\n", i);
          return 1;
        }
    }
  close (fd[0]);
  close (fd[1]);

  puts ("OK");

  return 0;
}
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/ia64/ia32/sys_ia32.c          |  2 +-
 arch/ia64/kernel/sys_ia64.c        |  2 +-
 arch/mips/kernel/syscall.c         |  2 +-
 arch/parisc/hpux/sys_hpux.c        |  2 +-
 arch/sh/kernel/sys_sh32.c          |  2 +-
 arch/sparc/kernel/sys_sparc.c      |  2 +-
 arch/sparc64/kernel/sys_sparc.c    |  2 +-
 arch/x86/ia32/ia32entry.S          |  1 +
 arch/x86/ia32/sys_ia32.c           |  2 +-
 arch/x86/kernel/syscall_table_32.S |  1 +
 arch/xtensa/kernel/syscall.c       |  2 +-
 fs/pipe.c                          | 23 ++++++++++++++++++-----
 include/asm-x86/unistd_32.h        |  1 +
 include/asm-x86/unistd_64.h        |  2 ++
 include/linux/fs.h                 |  1 +
 15 files changed, 33 insertions(+), 14 deletions(-)

diff --git a/arch/ia64/ia32/sys_ia32.c b/arch/ia64/ia32/sys_ia32.c
index 7e028ceb93ba..465116aecb85 100644
--- a/arch/ia64/ia32/sys_ia32.c
+++ b/arch/ia64/ia32/sys_ia32.c
@@ -1139,7 +1139,7 @@ sys32_pipe (int __user *fd)
 	int retval;
 	int fds[2];
 
-	retval = do_pipe(fds);
+	retval = do_pipe_flags(fds, 0);
 	if (retval)
 		goto out;
 	if (copy_to_user(fd, fds, sizeof(fds)))
diff --git a/arch/ia64/kernel/sys_ia64.c b/arch/ia64/kernel/sys_ia64.c
index 1eda194b9559..bcbb6d8792d3 100644
--- a/arch/ia64/kernel/sys_ia64.c
+++ b/arch/ia64/kernel/sys_ia64.c
@@ -160,7 +160,7 @@ sys_pipe (void)
 	int fd[2];
 	int retval;
 
-	retval = do_pipe(fd);
+	retval = do_pipe_flags(fd, 0);
 	if (retval)
 		goto out;
 	retval = fd[0];
diff --git a/arch/mips/kernel/syscall.c b/arch/mips/kernel/syscall.c
index 3523c8d12eda..343015a2f418 100644
--- a/arch/mips/kernel/syscall.c
+++ b/arch/mips/kernel/syscall.c
@@ -52,7 +52,7 @@ asmlinkage int sysm_pipe(nabi_no_regargs volatile struct pt_regs regs)
 	int fd[2];
 	int error, res;
 
-	error = do_pipe(fd);
+	error = do_pipe_flags(fd, 0);
 	if (error) {
 		res = error;
 		goto out;
diff --git a/arch/parisc/hpux/sys_hpux.c b/arch/parisc/hpux/sys_hpux.c
index 0c5b9dabb475..be255ebb609c 100644
--- a/arch/parisc/hpux/sys_hpux.c
+++ b/arch/parisc/hpux/sys_hpux.c
@@ -448,7 +448,7 @@ int hpux_pipe(int *kstack_fildes)
 	int error;
 
 	lock_kernel();
-	error = do_pipe(kstack_fildes);
+	error = do_pipe_flags(kstack_fildes, 0);
 	unlock_kernel();
 	return error;
 }
diff --git a/arch/sh/kernel/sys_sh32.c b/arch/sh/kernel/sys_sh32.c
index 125e493ead82..f0aa5c398656 100644
--- a/arch/sh/kernel/sys_sh32.c
+++ b/arch/sh/kernel/sys_sh32.c
@@ -29,7 +29,7 @@ asmlinkage int sys_pipe(unsigned long r4, unsigned long r5,
 	int fd[2];
 	int error;
 
-	error = do_pipe(fd);
+	error = do_pipe_flags(fd, 0);
 	if (!error) {
 		regs->regs[1] = fd[1];
 		return fd[0];
diff --git a/arch/sparc/kernel/sys_sparc.c b/arch/sparc/kernel/sys_sparc.c
index 3c6b49a53ae8..4d73421559c3 100644
--- a/arch/sparc/kernel/sys_sparc.c
+++ b/arch/sparc/kernel/sys_sparc.c
@@ -97,7 +97,7 @@ asmlinkage int sparc_pipe(struct pt_regs *regs)
 	int fd[2];
 	int error;
 
-	error = do_pipe(fd);
+	error = do_pipe_flags(fd, 0);
 	if (error)
 		goto out;
 	regs->u_regs[UREG_I1] = fd[1];
diff --git a/arch/sparc64/kernel/sys_sparc.c b/arch/sparc64/kernel/sys_sparc.c
index e1f4eba2e576..39749e32dc7e 100644
--- a/arch/sparc64/kernel/sys_sparc.c
+++ b/arch/sparc64/kernel/sys_sparc.c
@@ -418,7 +418,7 @@ asmlinkage long sparc_pipe(struct pt_regs *regs)
 	int fd[2];
 	int error;
 
-	error = do_pipe(fd);
+	error = do_pipe_flags(fd, 0);
 	if (error)
 		goto out;
 	regs->u_regs[UREG_I1] = fd[1];
diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S
index 5614a8f7bed4..18808b164570 100644
--- a/arch/x86/ia32/ia32entry.S
+++ b/arch/x86/ia32/ia32entry.S
@@ -830,4 +830,5 @@ ia32_sys_call_table:
 	.quad sys_eventfd2
 	.quad sys_epoll_create2
 	.quad sys_dup3			/* 330 */
+	.quad sys_pipe2
 ia32_syscall_end:
diff --git a/arch/x86/ia32/sys_ia32.c b/arch/x86/ia32/sys_ia32.c
index f00afdf61e67..d3c64088b981 100644
--- a/arch/x86/ia32/sys_ia32.c
+++ b/arch/x86/ia32/sys_ia32.c
@@ -238,7 +238,7 @@ asmlinkage long sys32_pipe(int __user *fd)
 	int retval;
 	int fds[2];
 
-	retval = do_pipe(fds);
+	retval = do_pipe_flags(fds, 0);
 	if (retval)
 		goto out;
 	if (copy_to_user(fd, fds, sizeof(fds)))
diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S
index 24a3f1ea6a0e..66154769d52f 100644
--- a/arch/x86/kernel/syscall_table_32.S
+++ b/arch/x86/kernel/syscall_table_32.S
@@ -330,3 +330,4 @@ ENTRY(sys_call_table)
 	.long sys_eventfd2
 	.long sys_epoll_create2
 	.long sys_dup3			/* 330 */
+	.long sys_pipe2
diff --git a/arch/xtensa/kernel/syscall.c b/arch/xtensa/kernel/syscall.c
index f3e16efcd47a..ac15ecbdf919 100644
--- a/arch/xtensa/kernel/syscall.c
+++ b/arch/xtensa/kernel/syscall.c
@@ -49,7 +49,7 @@ asmlinkage long xtensa_pipe(int __user *userfds)
 	int fd[2];
 	int error;
 
-	error = do_pipe(fd);
+	error = do_pipe_flags(fd, 0);
 	if (!error) {
 		if (copy_to_user(userfds, fd, 2 * sizeof(int)))
 			error = -EFAULT;
diff --git a/fs/pipe.c b/fs/pipe.c
index 700f4e0d9572..68e82061070c 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -1027,12 +1027,15 @@ struct file *create_read_pipe(struct file *wrf)
 	return f;
 }
 
-int do_pipe(int *fd)
+int do_pipe_flags(int *fd, int flags)
 {
 	struct file *fw, *fr;
 	int error;
 	int fdw, fdr;
 
+	if (flags & ~O_CLOEXEC)
+		return -EINVAL;
+
 	fw = create_write_pipe();
 	if (IS_ERR(fw))
 		return PTR_ERR(fw);
@@ -1041,12 +1044,12 @@ int do_pipe(int *fd)
 	if (IS_ERR(fr))
 		goto err_write_pipe;
 
-	error = get_unused_fd();
+	error = get_unused_fd_flags(flags);
 	if (error < 0)
 		goto err_read_pipe;
 	fdr = error;
 
-	error = get_unused_fd();
+	error = get_unused_fd_flags(flags);
 	if (error < 0)
 		goto err_fdr;
 	fdw = error;
@@ -1074,16 +1077,21 @@ int do_pipe(int *fd)
 	return error;
 }
 
+int do_pipe(int *fd)
+{
+	return do_pipe_flags(fd, 0);
+}
+
 /*
  * sys_pipe() is the normal C calling standard for creating
  * a pipe. It's not the way Unix traditionally does this, though.
  */
-asmlinkage long __weak sys_pipe(int __user *fildes)
+asmlinkage long __weak sys_pipe2(int __user *fildes, int flags)
 {
 	int fd[2];
 	int error;
 
-	error = do_pipe(fd);
+	error = do_pipe_flags(fd, flags);
 	if (!error) {
 		if (copy_to_user(fildes, fd, sizeof(fd))) {
 			sys_close(fd[0]);
@@ -1094,6 +1102,11 @@ asmlinkage long __weak sys_pipe(int __user *fildes)
 	return error;
 }
 
+asmlinkage long __weak sys_pipe(int __user *fildes)
+{
+	return sys_pipe2(fildes, 0);
+}
+
 /*
  * pipefs should _never_ be mounted by userland - too much of security hassle,
  * no real gain from having the whole whorehouse mounted. So we don't need
diff --git a/include/asm-x86/unistd_32.h b/include/asm-x86/unistd_32.h
index a1f6383bf695..748a05c77da4 100644
--- a/include/asm-x86/unistd_32.h
+++ b/include/asm-x86/unistd_32.h
@@ -336,6 +336,7 @@
 #define __NR_eventfd2		328
 #define __NR_epoll_create2	329
 #define __NR_dup3		330
+#define __NR_pipe2		331
 
 #ifdef __KERNEL__
 
diff --git a/include/asm-x86/unistd_64.h b/include/asm-x86/unistd_64.h
index f0fb2bd40cdb..d2284b43ad58 100644
--- a/include/asm-x86/unistd_64.h
+++ b/include/asm-x86/unistd_64.h
@@ -649,6 +649,8 @@ __SYSCALL(__NR_eventfd2, sys_eventfd2)
 __SYSCALL(__NR_epoll_create2, sys_epoll_create2)
 #define __NR_dup3				292
 __SYSCALL(__NR_dup3, sys_dup3)
+#define __NR_pipe2				293
+__SYSCALL(__NR_pipe2, sys_pipe2)
 
 
 #ifndef __NO_STUBS
diff --git a/include/linux/fs.h b/include/linux/fs.h
index e5e6a244096c..0e80cd717d32 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1777,6 +1777,7 @@ static inline void allow_write_access(struct file *file)
 		atomic_inc(&file->f_path.dentry->d_inode->i_writecount);
 }
 extern int do_pipe(int *);
+extern int do_pipe_flags(int *, int);
 extern struct file *create_read_pipe(struct file *f);
 extern struct file *create_write_pipe(void);
 extern void free_write_pipe(struct file *);

From 4006553b06306b34054529477b06b68a1c66249b Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Wed, 23 Jul 2008 21:29:32 -0700
Subject: [PATCH 0370/1171] flag parameters: inotify_init

This patch introduces the new syscall inotify_init1 (note: the 1 stands for
the one parameter the syscall takes, as opposed to no parameter before).  The
values accepted for this parameter are function-specific and defined in the
inotify.h header.  Here the values must match the O_* flags, though.  In this
patch CLOEXEC support is introduced.

The following test must be adjusted for architectures other than x86 and
x86-64 and in case the syscall numbers changed.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#include <fcntl.h>
#include <stdio.h>
#include <unistd.h>
#include <sys/syscall.h>

#ifndef __NR_inotify_init1
# ifdef __x86_64__
#  define __NR_inotify_init1 294
# elif defined __i386__
#  define __NR_inotify_init1 332
# else
#  error "need __NR_inotify_init1"
# endif
#endif

#define IN_CLOEXEC O_CLOEXEC

int
main (void)
{
  int fd;
  fd = syscall (__NR_inotify_init1, 0);
  if (fd == -1)
    {
      puts ("inotify_init1(0) failed");
      return 1;
    }
  int coe = fcntl (fd, F_GETFD);
  if (coe == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if (coe & FD_CLOEXEC)
    {
      puts ("inotify_init1(0) set close-on-exit");
      return 1;
    }
  close (fd);

  fd = syscall (__NR_inotify_init1, IN_CLOEXEC);
  if (fd == -1)
    {
      puts ("inotify_init1(IN_CLOEXEC) failed");
      return 1;
    }
  coe = fcntl (fd, F_GETFD);
  if (coe == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if ((coe & FD_CLOEXEC) == 0)
    {
      puts ("inotify_init1(O_CLOEXEC) does not set close-on-exit");
      return 1;
    }
  close (fd);

  puts ("OK");

  return 0;
}
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

[akpm@linux-foundation.org: add sys_ni stub]
Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/ia32/ia32entry.S          |  1 +
 arch/x86/kernel/syscall_table_32.S |  1 +
 fs/inotify_user.c                  | 12 ++++++++++--
 include/asm-x86/unistd_32.h        |  1 +
 include/asm-x86/unistd_64.h        |  2 ++
 include/linux/inotify.h            |  5 +++++
 include/linux/syscalls.h           |  1 +
 kernel/sys_ni.c                    |  1 +
 8 files changed, 22 insertions(+), 2 deletions(-)

diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S
index 18808b164570..4541073dd837 100644
--- a/arch/x86/ia32/ia32entry.S
+++ b/arch/x86/ia32/ia32entry.S
@@ -831,4 +831,5 @@ ia32_sys_call_table:
 	.quad sys_epoll_create2
 	.quad sys_dup3			/* 330 */
 	.quad sys_pipe2
+	.quad sys_inotify_init1
 ia32_syscall_end:
diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S
index 66154769d52f..f59aba5ff0f0 100644
--- a/arch/x86/kernel/syscall_table_32.S
+++ b/arch/x86/kernel/syscall_table_32.S
@@ -331,3 +331,4 @@ ENTRY(sys_call_table)
 	.long sys_epoll_create2
 	.long sys_dup3			/* 330 */
 	.long sys_pipe2
+	.long sys_inotify_init1
diff --git a/fs/inotify_user.c b/fs/inotify_user.c
index 6676c06bb7c1..851005998cd4 100644
--- a/fs/inotify_user.c
+++ b/fs/inotify_user.c
@@ -566,7 +566,7 @@ static const struct inotify_operations inotify_user_ops = {
 	.destroy_watch	= free_inotify_user_watch,
 };
 
-asmlinkage long sys_inotify_init(void)
+asmlinkage long sys_inotify_init1(int flags)
 {
 	struct inotify_device *dev;
 	struct inotify_handle *ih;
@@ -574,7 +574,10 @@ asmlinkage long sys_inotify_init(void)
 	struct file *filp;
 	int fd, ret;
 
-	fd = get_unused_fd();
+	if (flags & ~IN_CLOEXEC)
+		return -EINVAL;
+
+	fd = get_unused_fd_flags(flags & O_CLOEXEC);
 	if (fd < 0)
 		return fd;
 
@@ -638,6 +641,11 @@ out_put_fd:
 	return ret;
 }
 
+asmlinkage long sys_inotify_init(void)
+{
+	return sys_inotify_init1(0);
+}
+
 asmlinkage long sys_inotify_add_watch(int fd, const char __user *path, u32 mask)
 {
 	struct inode *inode;
diff --git a/include/asm-x86/unistd_32.h b/include/asm-x86/unistd_32.h
index 748a05c77da4..b3daf503ab93 100644
--- a/include/asm-x86/unistd_32.h
+++ b/include/asm-x86/unistd_32.h
@@ -337,6 +337,7 @@
 #define __NR_epoll_create2	329
 #define __NR_dup3		330
 #define __NR_pipe2		331
+#define __NR_inotify_init1	332
 
 #ifdef __KERNEL__
 
diff --git a/include/asm-x86/unistd_64.h b/include/asm-x86/unistd_64.h
index d2284b43ad58..c8cb88d70c6b 100644
--- a/include/asm-x86/unistd_64.h
+++ b/include/asm-x86/unistd_64.h
@@ -651,6 +651,8 @@ __SYSCALL(__NR_epoll_create2, sys_epoll_create2)
 __SYSCALL(__NR_dup3, sys_dup3)
 #define __NR_pipe2				293
 __SYSCALL(__NR_pipe2, sys_pipe2)
+#define __NR_inotify_init1			294
+__SYSCALL(__NR_inotify_init1, sys_inotify_init1)
 
 
 #ifndef __NO_STUBS
diff --git a/include/linux/inotify.h b/include/linux/inotify.h
index 742b917e7d1b..72ef82120512 100644
--- a/include/linux/inotify.h
+++ b/include/linux/inotify.h
@@ -7,6 +7,8 @@
 #ifndef _LINUX_INOTIFY_H
 #define _LINUX_INOTIFY_H
 
+/* For O_CLOEXEC */
+#include <linux/fcntl.h>
 #include <linux/types.h>
 
 /*
@@ -63,6 +65,9 @@ struct inotify_event {
 			 IN_MOVED_TO | IN_DELETE | IN_CREATE | IN_DELETE_SELF | \
 			 IN_MOVE_SELF)
 
+/* Flags for sys_inotify_init1.  */
+#define IN_CLOEXEC O_CLOEXEC
+
 #ifdef __KERNEL__
 
 #include <linux/dcache.h>
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 034d3358549e..93a7e7f017a6 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -547,6 +547,7 @@ asmlinkage long sys_get_mempolicy(int __user *policy,
 				unsigned long addr, unsigned long flags);
 
 asmlinkage long sys_inotify_init(void);
+asmlinkage long sys_inotify_init1(int flags);
 asmlinkage long sys_inotify_add_watch(int fd, const char __user *path,
 					u32 mask);
 asmlinkage long sys_inotify_rm_watch(int fd, u32 wd);
diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c
index 2a361ccdc7ca..bd66ac5406f3 100644
--- a/kernel/sys_ni.c
+++ b/kernel/sys_ni.c
@@ -96,6 +96,7 @@ cond_syscall(sys_keyctl);
 cond_syscall(compat_sys_keyctl);
 cond_syscall(compat_sys_socketcall);
 cond_syscall(sys_inotify_init);
+cond_syscall(sys_inotify_init1);
 cond_syscall(sys_inotify_add_watch);
 cond_syscall(sys_inotify_rm_watch);
 cond_syscall(sys_migrate_pages);

From 99829b832997d907c30669bfd17da32151e18f04 Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Wed, 23 Jul 2008 21:29:33 -0700
Subject: [PATCH 0371/1171] flag parameters: NONBLOCK in anon_inode_getfd

Building on the previous change to anon_inode_getfd, this patch introduces
support for handling of O_NONBLOCK in addition to the already supported
O_CLOEXEC.  Following patches will take advantage of this support.  As can be
seen, the additional support for supporting this functionality is minimal.

Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/anon_inodes.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/anon_inodes.c b/fs/anon_inodes.c
index 1a4eee620b0d..3662dd44896b 100644
--- a/fs/anon_inodes.c
+++ b/fs/anon_inodes.c
@@ -116,7 +116,7 @@ int anon_inode_getfd(const char *name, const struct file_operations *fops,
 	file->f_mapping = anon_inode_inode->i_mapping;
 
 	file->f_pos = 0;
-	file->f_flags = O_RDWR;
+	file->f_flags = O_RDWR | (flags & O_NONBLOCK);
 	file->f_version = 0;
 	file->private_data = priv;
 

From 77d2720059618b9b6e827a8b73831eb6c6fad63c Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Wed, 23 Jul 2008 21:29:35 -0700
Subject: [PATCH 0372/1171] flag parameters: NONBLOCK in socket and socketpair

This patch introduces support for the SOCK_NONBLOCK flag in socket,
socketpair, and  paccept.  To do this the internal function sock_attach_fd
gets an additional parameter which it uses to set the appropriate flag for
the file descriptor.

Given that in modern, scalable programs almost all socket connections are
non-blocking and the minimal additional cost for the new functionality
I see no reason not to add this code.

The following test must be adjusted for architectures other than x86 and
x86-64 and in case the syscall numbers changed.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#include <fcntl.h>
#include <pthread.h>
#include <stdio.h>
#include <unistd.h>
#include <netinet/in.h>
#include <sys/socket.h>
#include <sys/syscall.h>

#ifndef __NR_paccept
# ifdef __x86_64__
#  define __NR_paccept 288
# elif defined __i386__
#  define SYS_PACCEPT 18
#  define USE_SOCKETCALL 1
# else
#  error "need __NR_paccept"
# endif
#endif

#ifdef USE_SOCKETCALL
# define paccept(fd, addr, addrlen, mask, flags) \
  ({ long args[6] = { \
       (long) fd, (long) addr, (long) addrlen, (long) mask, 8, (long) flags }; \
     syscall (__NR_socketcall, SYS_PACCEPT, args); })
#else
# define paccept(fd, addr, addrlen, mask, flags) \
  syscall (__NR_paccept, fd, addr, addrlen, mask, 8, flags)
#endif

#define PORT 57392

#define SOCK_NONBLOCK O_NONBLOCK

static pthread_barrier_t b;

static void *
tf (void *arg)
{
  pthread_barrier_wait (&b);
  int s = socket (AF_INET, SOCK_STREAM, 0);
  struct sockaddr_in sin;
  sin.sin_family = AF_INET;
  sin.sin_addr.s_addr = htonl (INADDR_LOOPBACK);
  sin.sin_port = htons (PORT);
  connect (s, (const struct sockaddr *) &sin, sizeof (sin));
  close (s);
  pthread_barrier_wait (&b);

  pthread_barrier_wait (&b);
  s = socket (AF_INET, SOCK_STREAM, 0);
  sin.sin_port = htons (PORT);
  connect (s, (const struct sockaddr *) &sin, sizeof (sin));
  close (s);
  pthread_barrier_wait (&b);

  return NULL;
}

int
main (void)
{
  int fd;
  fd = socket (PF_INET, SOCK_STREAM, 0);
  if (fd == -1)
    {
      puts ("socket(0) failed");
      return 1;
    }
  int fl = fcntl (fd, F_GETFL);
  if (fl == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if (fl & O_NONBLOCK)
    {
      puts ("socket(0) set non-blocking mode");
      return 1;
    }
  close (fd);

  fd = socket (PF_INET, SOCK_STREAM|SOCK_NONBLOCK, 0);
  if (fd == -1)
    {
      puts ("socket(SOCK_NONBLOCK) failed");
      return 1;
    }
  fl = fcntl (fd, F_GETFL);
  if (fl == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if ((fl & O_NONBLOCK) == 0)
    {
      puts ("socket(SOCK_NONBLOCK) does not set non-blocking mode");
      return 1;
    }
  close (fd);

  int fds[2];
  if (socketpair (PF_UNIX, SOCK_STREAM, 0, fds) == -1)
    {
      puts ("socketpair(0) failed");
      return 1;
    }
  for (int i = 0; i < 2; ++i)
    {
      fl = fcntl (fds[i], F_GETFL);
      if (fl == -1)
        {
          puts ("fcntl failed");
          return 1;
        }
      if (fl & O_NONBLOCK)
        {
          printf ("socketpair(0) set non-blocking mode for fds[%d]\n", i);
          return 1;
        }
      close (fds[i]);
    }

  if (socketpair (PF_UNIX, SOCK_STREAM|SOCK_NONBLOCK, 0, fds) == -1)
    {
      puts ("socketpair(SOCK_NONBLOCK) failed");
      return 1;
    }
  for (int i = 0; i < 2; ++i)
    {
      fl = fcntl (fds[i], F_GETFL);
      if (fl == -1)
        {
          puts ("fcntl failed");
          return 1;
        }
      if ((fl & O_NONBLOCK) == 0)
        {
          printf ("socketpair(SOCK_NONBLOCK) does not set non-blocking mode for fds[%d]\n", i);
          return 1;
        }
      close (fds[i]);
    }

  pthread_barrier_init (&b, NULL, 2);

  struct sockaddr_in sin;
  pthread_t th;
  if (pthread_create (&th, NULL, tf, NULL) != 0)
    {
      puts ("pthread_create failed");
      return 1;
    }

  int s = socket (AF_INET, SOCK_STREAM, 0);
  int reuse = 1;
  setsockopt (s, SOL_SOCKET, SO_REUSEADDR, &reuse, sizeof (reuse));
  sin.sin_family = AF_INET;
  sin.sin_addr.s_addr = htonl (INADDR_LOOPBACK);
  sin.sin_port = htons (PORT);
  bind (s, (struct sockaddr *) &sin, sizeof (sin));
  listen (s, SOMAXCONN);

  pthread_barrier_wait (&b);

  int s2 = paccept (s, NULL, 0, NULL, 0);
  if (s2 < 0)
    {
      puts ("paccept(0) failed");
      return 1;
    }

  fl = fcntl (s2, F_GETFL);
  if (fl & O_NONBLOCK)
    {
      puts ("paccept(0) set non-blocking mode");
      return 1;
    }
  close (s2);
  close (s);

  pthread_barrier_wait (&b);

  s = socket (AF_INET, SOCK_STREAM, 0);
  sin.sin_port = htons (PORT);
  setsockopt (s, SOL_SOCKET, SO_REUSEADDR, &reuse, sizeof (reuse));
  bind (s, (struct sockaddr *) &sin, sizeof (sin));
  listen (s, SOMAXCONN);

  pthread_barrier_wait (&b);

  s2 = paccept (s, NULL, 0, NULL, SOCK_NONBLOCK);
  if (s2 < 0)
    {
      puts ("paccept(SOCK_NONBLOCK) failed");
      return 1;
    }

  fl = fcntl (s2, F_GETFL);
  if ((fl & O_NONBLOCK) == 0)
    {
      puts ("paccept(SOCK_NONBLOCK) does not set non-blocking mode");
      return 1;
    }
  close (s2);
  close (s);

  pthread_barrier_wait (&b);
  puts ("OK");

  return 0;
}
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/net.h |  2 +-
 net/socket.c        | 20 ++++++++++----------
 2 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/include/linux/net.h b/include/linux/net.h
index 39a23af059b4..2f999fbb188d 100644
--- a/include/linux/net.h
+++ b/include/linux/net.h
@@ -20,7 +20,7 @@
 
 #include <linux/wait.h>
 #include <linux/socket.h>
-#include <linux/fcntl.h>	/* For O_CLOEXEC */
+#include <linux/fcntl.h>	/* For O_CLOEXEC and O_NONBLOCK */
 #include <asm/socket.h>
 
 struct poll_table_struct;
diff --git a/net/socket.c b/net/socket.c
index d163adff95bf..31105f9048a8 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -369,7 +369,7 @@ static int sock_alloc_fd(struct file **filep, int flags)
 	return fd;
 }
 
-static int sock_attach_fd(struct socket *sock, struct file *file)
+static int sock_attach_fd(struct socket *sock, struct file *file, int flags)
 {
 	struct dentry *dentry;
 	struct qstr name = { .name = "" };
@@ -391,7 +391,7 @@ static int sock_attach_fd(struct socket *sock, struct file *file)
 	init_file(file, sock_mnt, dentry, FMODE_READ | FMODE_WRITE,
 		  &socket_file_ops);
 	SOCK_INODE(sock)->i_fop = &socket_file_ops;
-	file->f_flags = O_RDWR;
+	file->f_flags = O_RDWR | (flags & O_NONBLOCK);
 	file->f_pos = 0;
 	file->private_data = sock;
 
@@ -404,7 +404,7 @@ int sock_map_fd(struct socket *sock, int flags)
 	int fd = sock_alloc_fd(&newfile, flags);
 
 	if (likely(fd >= 0)) {
-		int err = sock_attach_fd(sock, newfile);
+		int err = sock_attach_fd(sock, newfile, flags);
 
 		if (unlikely(err < 0)) {
 			put_filp(newfile);
@@ -1223,7 +1223,7 @@ asmlinkage long sys_socket(int family, int type, int protocol)
 	int flags;
 
 	flags = type & ~SOCK_TYPE_MASK;
-	if (flags & ~SOCK_CLOEXEC)
+	if (flags & ~(SOCK_CLOEXEC | SOCK_NONBLOCK))
 		return -EINVAL;
 	type &= SOCK_TYPE_MASK;
 
@@ -1234,7 +1234,7 @@ asmlinkage long sys_socket(int family, int type, int protocol)
 	if (retval < 0)
 		goto out;
 
-	retval = sock_map_fd(sock, flags & O_CLOEXEC);
+	retval = sock_map_fd(sock, flags & (O_CLOEXEC | O_NONBLOCK));
 	if (retval < 0)
 		goto out_release;
 
@@ -1260,7 +1260,7 @@ asmlinkage long sys_socketpair(int family, int type, int protocol,
 	int flags;
 
 	flags = type & ~SOCK_TYPE_MASK;
-	if (flags & ~SOCK_CLOEXEC)
+	if (flags & ~(SOCK_CLOEXEC | SOCK_NONBLOCK))
 		return -EINVAL;
 	type &= SOCK_TYPE_MASK;
 
@@ -1298,12 +1298,12 @@ asmlinkage long sys_socketpair(int family, int type, int protocol,
 		goto out_release_both;
 	}
 
-	err = sock_attach_fd(sock1, newfile1);
+	err = sock_attach_fd(sock1, newfile1, flags & O_NONBLOCK);
 	if (unlikely(err < 0)) {
 		goto out_fd2;
 	}
 
-	err = sock_attach_fd(sock2, newfile2);
+	err = sock_attach_fd(sock2, newfile2, flags & O_NONBLOCK);
 	if (unlikely(err < 0)) {
 		fput(newfile1);
 		goto out_fd1;
@@ -1429,7 +1429,7 @@ long do_accept(int fd, struct sockaddr __user *upeer_sockaddr,
 	int err, len, newfd, fput_needed;
 	struct sockaddr_storage address;
 
-	if (flags & ~SOCK_CLOEXEC)
+	if (flags & ~(SOCK_CLOEXEC | SOCK_NONBLOCK))
 		return -EINVAL;
 
 	if (SOCK_NONBLOCK != O_NONBLOCK && (flags & SOCK_NONBLOCK))
@@ -1459,7 +1459,7 @@ long do_accept(int fd, struct sockaddr __user *upeer_sockaddr,
 		goto out_put;
 	}
 
-	err = sock_attach_fd(newsock, newfile);
+	err = sock_attach_fd(newsock, newfile, flags & O_NONBLOCK);
 	if (err < 0)
 		goto out_fd_simple;
 

From 5fb5e04926a54bc1c22bba7ca166840f4476196f Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Wed, 23 Jul 2008 21:29:37 -0700
Subject: [PATCH 0373/1171] flag parameters: NONBLOCK in signalfd

This patch adds support for the SFD_NONBLOCK flag to signalfd4.  The
additional changes needed are minimal.

The following test must be adjusted for architectures other than x86 and
x86-64 and in case the syscall numbers changed.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#include <fcntl.h>
#include <signal.h>
#include <stdio.h>
#include <unistd.h>
#include <sys/syscall.h>

#ifndef __NR_signalfd4
# ifdef __x86_64__
#  define __NR_signalfd4 289
# elif defined __i386__
#  define __NR_signalfd4 327
# else
#  error "need __NR_signalfd4"
# endif
#endif

#define SFD_NONBLOCK O_NONBLOCK

int
main (void)
{
  sigset_t ss;
  sigemptyset (&ss);
  sigaddset (&ss, SIGUSR1);
  int fd = syscall (__NR_signalfd4, -1, &ss, 8, 0);
  if (fd == -1)
    {
      puts ("signalfd4(0) failed");
      return 1;
    }
  int fl = fcntl (fd, F_GETFL);
  if (fl == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if (fl & O_NONBLOCK)
    {
      puts ("signalfd4(0) set non-blocking mode");
      return 1;
    }
  close (fd);

  fd = syscall (__NR_signalfd4, -1, &ss, 8, SFD_NONBLOCK);
  if (fd == -1)
    {
      puts ("signalfd4(SFD_NONBLOCK) failed");
      return 1;
    }
  fl = fcntl (fd, F_GETFL);
  if (fl == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if ((fl & O_NONBLOCK) == 0)
    {
      puts ("signalfd4(SFD_NONBLOCK) does not set non-blocking mode");
      return 1;
    }
  close (fd);

  puts ("OK");

  return 0;
}
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/signalfd.c            | 4 ++--
 include/linux/signalfd.h | 3 ++-
 2 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/fs/signalfd.c b/fs/signalfd.c
index c8609fa51a13..5441a4bca772 100644
--- a/fs/signalfd.c
+++ b/fs/signalfd.c
@@ -211,7 +211,7 @@ asmlinkage long sys_signalfd4(int ufd, sigset_t __user *user_mask,
 	sigset_t sigmask;
 	struct signalfd_ctx *ctx;
 
-	if (flags & ~SFD_CLOEXEC)
+	if (flags & ~(SFD_CLOEXEC | SFD_NONBLOCK))
 		return -EINVAL;
 
 	if (sizemask != sizeof(sigset_t) ||
@@ -232,7 +232,7 @@ asmlinkage long sys_signalfd4(int ufd, sigset_t __user *user_mask,
 		 * anon_inode_getfd() will install the fd.
 		 */
 		ufd = anon_inode_getfd("[signalfd]", &signalfd_fops, ctx,
-				       flags & O_CLOEXEC);
+				       flags & (O_CLOEXEC | O_NONBLOCK));
 		if (ufd < 0)
 			kfree(ctx);
 	} else {
diff --git a/include/linux/signalfd.h b/include/linux/signalfd.h
index 8b3f7b7420a1..bef0c46d4713 100644
--- a/include/linux/signalfd.h
+++ b/include/linux/signalfd.h
@@ -8,11 +8,12 @@
 #ifndef _LINUX_SIGNALFD_H
 #define _LINUX_SIGNALFD_H
 
-/* For O_CLOEXEC */
+/* For O_CLOEXEC and O_NONBLOCK */
 #include <linux/fcntl.h>
 
 /* Flags for signalfd4.  */
 #define SFD_CLOEXEC O_CLOEXEC
+#define SFD_NONBLOCK O_NONBLOCK
 
 struct signalfd_siginfo {
 	__u32 ssi_signo;

From e7d476dfdf0bcfed478a207aecfdc84f81efecaf Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Wed, 23 Jul 2008 21:29:38 -0700
Subject: [PATCH 0374/1171] flag parameters: NONBLOCK in eventfd

This patch adds support for the EFD_NONBLOCK flag to eventfd2.  The
additional changes needed are minimal.

The following test must be adjusted for architectures other than x86 and
x86-64 and in case the syscall numbers changed.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#include <fcntl.h>
#include <stdio.h>
#include <unistd.h>
#include <sys/syscall.h>

#ifndef __NR_eventfd2
# ifdef __x86_64__
#  define __NR_eventfd2 290
# elif defined __i386__
#  define __NR_eventfd2 328
# else
#  error "need __NR_eventfd2"
# endif
#endif

#define EFD_NONBLOCK O_NONBLOCK

int
main (void)
{
  int fd = syscall (__NR_eventfd2, 1, 0);
  if (fd == -1)
    {
      puts ("eventfd2(0) failed");
      return 1;
    }
  int fl = fcntl (fd, F_GETFL);
  if (fl == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if (fl & O_NONBLOCK)
    {
      puts ("eventfd2(0) sets non-blocking mode");
      return 1;
    }
  close (fd);

  fd = syscall (__NR_eventfd2, 1, EFD_NONBLOCK);
  if (fd == -1)
    {
      puts ("eventfd2(EFD_NONBLOCK) failed");
      return 1;
    }
  fl = fcntl (fd, F_GETFL);
  if (fl == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if ((fl & O_NONBLOCK) == 0)
    {
      puts ("eventfd2(EFD_NONBLOCK) does not set non-blocking mode");
      return 1;
    }
  close (fd);

  puts ("OK");

  return 0;
}
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/eventfd.c            | 4 ++--
 include/linux/eventfd.h | 3 ++-
 2 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/fs/eventfd.c b/fs/eventfd.c
index bd420e6478ad..3ed4466177a7 100644
--- a/fs/eventfd.c
+++ b/fs/eventfd.c
@@ -203,7 +203,7 @@ asmlinkage long sys_eventfd2(unsigned int count, int flags)
 	int fd;
 	struct eventfd_ctx *ctx;
 
-	if (flags & ~EFD_CLOEXEC)
+	if (flags & ~(EFD_CLOEXEC | EFD_NONBLOCK))
 		return -EINVAL;
 
 	ctx = kmalloc(sizeof(*ctx), GFP_KERNEL);
@@ -218,7 +218,7 @@ asmlinkage long sys_eventfd2(unsigned int count, int flags)
 	 * anon_inode_getfd() will install the fd.
 	 */
 	fd = anon_inode_getfd("[eventfd]", &eventfd_fops, ctx,
-			      flags & O_CLOEXEC);
+			      flags & (O_CLOEXEC | O_NONBLOCK));
 	if (fd < 0)
 		kfree(ctx);
 	return fd;
diff --git a/include/linux/eventfd.h b/include/linux/eventfd.h
index a6c0eaedb1b0..a667637b54e3 100644
--- a/include/linux/eventfd.h
+++ b/include/linux/eventfd.h
@@ -10,11 +10,12 @@
 
 #ifdef CONFIG_EVENTFD
 
-/* For O_CLOEXEC */
+/* For O_CLOEXEC and O_NONBLOCK */
 #include <linux/fcntl.h>
 
 /* Flags for eventfd2.  */
 #define EFD_CLOEXEC O_CLOEXEC
+#define EFD_NONBLOCK O_NONBLOCK
 
 struct file *eventfd_fget(int fd);
 int eventfd_signal(struct file *file, int n);

From 6b1ef0e60d42f2fdaec26baee8327eb156347b4f Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Wed, 23 Jul 2008 21:29:39 -0700
Subject: [PATCH 0375/1171] flag parameters: NONBLOCK in timerfd_create

This patch adds support for the TFD_NONBLOCK flag to timerfd_create.  The
additional changes needed are minimal.

The following test must be adjusted for architectures other than x86 and
x86-64 and in case the syscall numbers changed.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#include <fcntl.h>
#include <stdio.h>
#include <time.h>
#include <unistd.h>
#include <sys/syscall.h>

#ifndef __NR_timerfd_create
# ifdef __x86_64__
#  define __NR_timerfd_create 283
# elif defined __i386__
#  define __NR_timerfd_create 322
# else
#  error "need __NR_timerfd_create"
# endif
#endif

#define TFD_NONBLOCK O_NONBLOCK

int
main (void)
{
  int fd = syscall (__NR_timerfd_create, CLOCK_REALTIME, 0);
  if (fd == -1)
    {
      puts ("timerfd_create(0) failed");
      return 1;
    }
  int fl = fcntl (fd, F_GETFL);
  if (fl == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if (fl & O_NONBLOCK)
    {
      puts ("timerfd_create(0) set non-blocking mode");
      return 1;
    }
  close (fd);

  fd = syscall (__NR_timerfd_create, CLOCK_REALTIME, TFD_NONBLOCK);
  if (fd == -1)
    {
      puts ("timerfd_create(TFD_NONBLOCK) failed");
      return 1;
    }
  fl = fcntl (fd, F_GETFL);
  if (fl == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if ((fl & O_NONBLOCK) == 0)
    {
      puts ("timerfd_create(TFD_NONBLOCK) set non-blocking mode");
      return 1;
    }
  close (fd);

  puts ("OK");

  return 0;
}
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/timerfd.c            | 4 ++--
 include/linux/timerfd.h | 3 ++-
 2 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/fs/timerfd.c b/fs/timerfd.c
index c6ef5e33cb34..75d44efe346c 100644
--- a/fs/timerfd.c
+++ b/fs/timerfd.c
@@ -184,7 +184,7 @@ asmlinkage long sys_timerfd_create(int clockid, int flags)
 	int ufd;
 	struct timerfd_ctx *ctx;
 
-	if (flags & ~TFD_CLOEXEC)
+	if (flags & ~(TFD_CLOEXEC | TFD_NONBLOCK))
 		return -EINVAL;
 	if (clockid != CLOCK_MONOTONIC &&
 	    clockid != CLOCK_REALTIME)
@@ -199,7 +199,7 @@ asmlinkage long sys_timerfd_create(int clockid, int flags)
 	hrtimer_init(&ctx->tmr, clockid, HRTIMER_MODE_ABS);
 
 	ufd = anon_inode_getfd("[timerfd]", &timerfd_fops, ctx,
-			       flags & O_CLOEXEC);
+			       flags & (O_CLOEXEC | O_NONBLOCK));
 	if (ufd < 0)
 		kfree(ctx);
 
diff --git a/include/linux/timerfd.h b/include/linux/timerfd.h
index 96ed97dff00f..86cb0501d3e2 100644
--- a/include/linux/timerfd.h
+++ b/include/linux/timerfd.h
@@ -8,7 +8,7 @@
 #ifndef _LINUX_TIMERFD_H
 #define _LINUX_TIMERFD_H
 
-/* For O_CLOEXEC */
+/* For O_CLOEXEC and O_NONBLOCK */
 #include <linux/fcntl.h>
 
 /* Flags for timerfd_settime.  */
@@ -16,6 +16,7 @@
 
 /* Flags for timerfd_create.  */
 #define TFD_CLOEXEC O_CLOEXEC
+#define TFD_NONBLOCK O_NONBLOCK
 
 
 #endif /* _LINUX_TIMERFD_H */

From be61a86d7237dd80510615f38ae21d6e1e98660c Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Wed, 23 Jul 2008 21:29:40 -0700
Subject: [PATCH 0376/1171] flag parameters: NONBLOCK in pipe

This patch adds O_NONBLOCK support to pipe2.  It is minimally more involved
than the patches for eventfd et.al but still trivial.  The interfaces of the
create_write_pipe and create_read_pipe helper functions were changed and the
one other caller as well.

The following test must be adjusted for architectures other than x86 and
x86-64 and in case the syscall numbers changed.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#include <fcntl.h>
#include <stdio.h>
#include <unistd.h>
#include <sys/syscall.h>

#ifndef __NR_pipe2
# ifdef __x86_64__
#  define __NR_pipe2 293
# elif defined __i386__
#  define __NR_pipe2 331
# else
#  error "need __NR_pipe2"
# endif
#endif

int
main (void)
{
  int fds[2];
  if (syscall (__NR_pipe2, fds, 0) == -1)
    {
      puts ("pipe2(0) failed");
      return 1;
    }
  for (int i = 0; i < 2; ++i)
    {
      int fl = fcntl (fds[i], F_GETFL);
      if (fl == -1)
        {
          puts ("fcntl failed");
          return 1;
        }
      if (fl & O_NONBLOCK)
        {
          printf ("pipe2(0) set non-blocking mode for fds[%d]\n", i);
          return 1;
        }
      close (fds[i]);
    }

  if (syscall (__NR_pipe2, fds, O_NONBLOCK) == -1)
    {
      puts ("pipe2(O_NONBLOCK) failed");
      return 1;
    }
  for (int i = 0; i < 2; ++i)
    {
      int fl = fcntl (fds[i], F_GETFL);
      if (fl == -1)
        {
          puts ("fcntl failed");
          return 1;
        }
      if ((fl & O_NONBLOCK) == 0)
        {
          printf ("pipe2(O_NONBLOCK) does not set non-blocking mode for fds[%d]\n", i);
          return 1;
        }
      close (fds[i]);
    }

  puts ("OK");

  return 0;
}
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/pipe.c          | 14 +++++++-------
 include/linux/fs.h |  4 ++--
 kernel/kmod.c      |  4 ++--
 3 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/fs/pipe.c b/fs/pipe.c
index 68e82061070c..10c4e9aa5c49 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -950,7 +950,7 @@ fail_inode:
 	return NULL;
 }
 
-struct file *create_write_pipe(void)
+struct file *create_write_pipe(int flags)
 {
 	int err;
 	struct inode *inode;
@@ -983,7 +983,7 @@ struct file *create_write_pipe(void)
 		goto err_dentry;
 	f->f_mapping = inode->i_mapping;
 
-	f->f_flags = O_WRONLY;
+	f->f_flags = O_WRONLY | (flags & O_NONBLOCK);
 	f->f_version = 0;
 
 	return f;
@@ -1007,7 +1007,7 @@ void free_write_pipe(struct file *f)
 	put_filp(f);
 }
 
-struct file *create_read_pipe(struct file *wrf)
+struct file *create_read_pipe(struct file *wrf, int flags)
 {
 	struct file *f = get_empty_filp();
 	if (!f)
@@ -1019,7 +1019,7 @@ struct file *create_read_pipe(struct file *wrf)
 	f->f_mapping = wrf->f_path.dentry->d_inode->i_mapping;
 
 	f->f_pos = 0;
-	f->f_flags = O_RDONLY;
+	f->f_flags = O_RDONLY | (flags & O_NONBLOCK);
 	f->f_op = &read_pipe_fops;
 	f->f_mode = FMODE_READ;
 	f->f_version = 0;
@@ -1033,13 +1033,13 @@ int do_pipe_flags(int *fd, int flags)
 	int error;
 	int fdw, fdr;
 
-	if (flags & ~O_CLOEXEC)
+	if (flags & ~(O_CLOEXEC | O_NONBLOCK))
 		return -EINVAL;
 
-	fw = create_write_pipe();
+	fw = create_write_pipe(flags);
 	if (IS_ERR(fw))
 		return PTR_ERR(fw);
-	fr = create_read_pipe(fw);
+	fr = create_read_pipe(fw, flags);
 	error = PTR_ERR(fr);
 	if (IS_ERR(fr))
 		goto err_write_pipe;
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 0e80cd717d32..4b86f806014c 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1778,8 +1778,8 @@ static inline void allow_write_access(struct file *file)
 }
 extern int do_pipe(int *);
 extern int do_pipe_flags(int *, int);
-extern struct file *create_read_pipe(struct file *f);
-extern struct file *create_write_pipe(void);
+extern struct file *create_read_pipe(struct file *f, int flags);
+extern struct file *create_write_pipe(int flags);
 extern void free_write_pipe(struct file *);
 
 extern struct file *do_filp_open(int dfd, const char *pathname,
diff --git a/kernel/kmod.c b/kernel/kmod.c
index 90d7af1c1655..2989f67c4446 100644
--- a/kernel/kmod.c
+++ b/kernel/kmod.c
@@ -417,12 +417,12 @@ int call_usermodehelper_stdinpipe(struct subprocess_info *sub_info,
 {
 	struct file *f;
 
-	f = create_write_pipe();
+	f = create_write_pipe(0);
 	if (IS_ERR(f))
 		return PTR_ERR(f);
 	*filp = f;
 
-	f = create_read_pipe(f);
+	f = create_read_pipe(f, 0);
 	if (IS_ERR(f)) {
 		free_write_pipe(*filp);
 		return PTR_ERR(f);

From 510df2dd482496083e1c3b1a8c9b6afd5fa4c7d7 Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Wed, 23 Jul 2008 21:29:41 -0700
Subject: [PATCH 0377/1171] flag parameters: NONBLOCK in inotify_init

This patch adds non-blocking support for inotify_init1.  The
additional changes needed are minimal.

The following test must be adjusted for architectures other than x86 and
x86-64 and in case the syscall numbers changed.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#include <fcntl.h>
#include <stdio.h>
#include <unistd.h>
#include <sys/syscall.h>

#ifndef __NR_inotify_init1
# ifdef __x86_64__
#  define __NR_inotify_init1 294
# elif defined __i386__
#  define __NR_inotify_init1 332
# else
#  error "need __NR_inotify_init1"
# endif
#endif

#define IN_NONBLOCK O_NONBLOCK

int
main (void)
{
  int fd = syscall (__NR_inotify_init1, 0);
  if (fd == -1)
    {
      puts ("inotify_init1(0) failed");
      return 1;
    }
  int fl = fcntl (fd, F_GETFL);
  if (fl == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if (fl & O_NONBLOCK)
    {
      puts ("inotify_init1(0) set non-blocking mode");
      return 1;
    }
  close (fd);

  fd = syscall (__NR_inotify_init1, IN_NONBLOCK);
  if (fd == -1)
    {
      puts ("inotify_init1(IN_NONBLOCK) failed");
      return 1;
    }
  fl = fcntl (fd, F_GETFL);
  if (fl == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if ((fl & O_NONBLOCK) == 0)
    {
      puts ("inotify_init1(IN_NONBLOCK) set non-blocking mode");
      return 1;
    }
  close (fd);

  puts ("OK");

  return 0;
}
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/inotify_user.c       | 4 ++--
 include/linux/inotify.h | 3 ++-
 2 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/fs/inotify_user.c b/fs/inotify_user.c
index 851005998cd4..dc7e1f619748 100644
--- a/fs/inotify_user.c
+++ b/fs/inotify_user.c
@@ -574,7 +574,7 @@ asmlinkage long sys_inotify_init1(int flags)
 	struct file *filp;
 	int fd, ret;
 
-	if (flags & ~IN_CLOEXEC)
+	if (flags & ~(IN_CLOEXEC | IN_NONBLOCK))
 		return -EINVAL;
 
 	fd = get_unused_fd_flags(flags & O_CLOEXEC);
@@ -613,7 +613,7 @@ asmlinkage long sys_inotify_init1(int flags)
 	filp->f_path.dentry = dget(inotify_mnt->mnt_root);
 	filp->f_mapping = filp->f_path.dentry->d_inode->i_mapping;
 	filp->f_mode = FMODE_READ;
-	filp->f_flags = O_RDONLY;
+	filp->f_flags = O_RDONLY | (flags & O_NONBLOCK);
 	filp->private_data = dev;
 
 	INIT_LIST_HEAD(&dev->events);
diff --git a/include/linux/inotify.h b/include/linux/inotify.h
index 72ef82120512..bd578578a8b9 100644
--- a/include/linux/inotify.h
+++ b/include/linux/inotify.h
@@ -7,7 +7,7 @@
 #ifndef _LINUX_INOTIFY_H
 #define _LINUX_INOTIFY_H
 
-/* For O_CLOEXEC */
+/* For O_CLOEXEC and O_NONBLOCK */
 #include <linux/fcntl.h>
 #include <linux/types.h>
 
@@ -67,6 +67,7 @@ struct inotify_event {
 
 /* Flags for sys_inotify_init1.  */
 #define IN_CLOEXEC O_CLOEXEC
+#define IN_NONBLOCK O_NONBLOCK
 
 #ifdef __KERNEL__
 

From e38b36f325153eaadd1c2a7abc5762079233e540 Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Wed, 23 Jul 2008 21:29:42 -0700
Subject: [PATCH 0378/1171] flag parameters: check magic constants

This patch adds test that ensure the boundary conditions for the various
constants introduced in the previous patches is met.  No code is generated.

[akpm@linux-foundation.org: fix alpha]
Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/eventfd.c      | 4 ++++
 fs/eventpoll.c    | 3 +++
 fs/inotify_user.c | 4 ++++
 fs/signalfd.c     | 4 ++++
 fs/timerfd.c      | 4 ++++
 net/socket.c      | 6 ++++++
 6 files changed, 25 insertions(+)

diff --git a/fs/eventfd.c b/fs/eventfd.c
index 3ed4466177a7..08bf558d0408 100644
--- a/fs/eventfd.c
+++ b/fs/eventfd.c
@@ -203,6 +203,10 @@ asmlinkage long sys_eventfd2(unsigned int count, int flags)
 	int fd;
 	struct eventfd_ctx *ctx;
 
+	/* Check the EFD_* constants for consistency.  */
+	BUILD_BUG_ON(EFD_CLOEXEC != O_CLOEXEC);
+	BUILD_BUG_ON(EFD_NONBLOCK != O_NONBLOCK);
+
 	if (flags & ~(EFD_CLOEXEC | EFD_NONBLOCK))
 		return -EINVAL;
 
diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index 3fd4014f3c5a..2fdad4204044 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -1051,6 +1051,9 @@ asmlinkage long sys_epoll_create2(int size, int flags)
 	int error, fd = -1;
 	struct eventpoll *ep;
 
+	/* Check the EPOLL_* constant for consistency.  */
+	BUILD_BUG_ON(EPOLL_CLOEXEC != O_CLOEXEC);
+
 	if (flags & ~EPOLL_CLOEXEC)
 		return -EINVAL;
 
diff --git a/fs/inotify_user.c b/fs/inotify_user.c
index dc7e1f619748..fe79c25d95dc 100644
--- a/fs/inotify_user.c
+++ b/fs/inotify_user.c
@@ -574,6 +574,10 @@ asmlinkage long sys_inotify_init1(int flags)
 	struct file *filp;
 	int fd, ret;
 
+	/* Check the IN_* constants for consistency.  */
+	BUILD_BUG_ON(IN_CLOEXEC != O_CLOEXEC);
+	BUILD_BUG_ON(IN_NONBLOCK != O_NONBLOCK);
+
 	if (flags & ~(IN_CLOEXEC | IN_NONBLOCK))
 		return -EINVAL;
 
diff --git a/fs/signalfd.c b/fs/signalfd.c
index 5441a4bca772..9c39bc7f8431 100644
--- a/fs/signalfd.c
+++ b/fs/signalfd.c
@@ -211,6 +211,10 @@ asmlinkage long sys_signalfd4(int ufd, sigset_t __user *user_mask,
 	sigset_t sigmask;
 	struct signalfd_ctx *ctx;
 
+	/* Check the SFD_* constants for consistency.  */
+	BUILD_BUG_ON(SFD_CLOEXEC != O_CLOEXEC);
+	BUILD_BUG_ON(SFD_NONBLOCK != O_NONBLOCK);
+
 	if (flags & ~(SFD_CLOEXEC | SFD_NONBLOCK))
 		return -EINVAL;
 
diff --git a/fs/timerfd.c b/fs/timerfd.c
index 75d44efe346c..c502c60e4f54 100644
--- a/fs/timerfd.c
+++ b/fs/timerfd.c
@@ -184,6 +184,10 @@ asmlinkage long sys_timerfd_create(int clockid, int flags)
 	int ufd;
 	struct timerfd_ctx *ctx;
 
+	/* Check the TFD_* constants for consistency.  */
+	BUILD_BUG_ON(TFD_CLOEXEC != O_CLOEXEC);
+	BUILD_BUG_ON(TFD_NONBLOCK != O_NONBLOCK);
+
 	if (flags & ~(TFD_CLOEXEC | TFD_NONBLOCK))
 		return -EINVAL;
 	if (clockid != CLOCK_MONOTONIC &&
diff --git a/net/socket.c b/net/socket.c
index 31105f9048a8..1310a82cbba7 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -1222,6 +1222,12 @@ asmlinkage long sys_socket(int family, int type, int protocol)
 	struct socket *sock;
 	int flags;
 
+	/* Check the SOCK_* constants for consistency.  */
+	BUILD_BUG_ON(SOCK_CLOEXEC != O_CLOEXEC);
+	BUILD_BUG_ON((SOCK_MAX | SOCK_TYPE_MASK) != SOCK_TYPE_MASK);
+	BUILD_BUG_ON(SOCK_CLOEXEC & SOCK_TYPE_MASK);
+	BUILD_BUG_ON(SOCK_NONBLOCK & SOCK_TYPE_MASK);
+
 	flags = type & ~SOCK_TYPE_MASK;
 	if (flags & ~(SOCK_CLOEXEC | SOCK_NONBLOCK))
 		return -EINVAL;

From 9fe5ad9c8cef9ad5873d8ee55d1cf00d9b607df0 Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Wed, 23 Jul 2008 21:29:43 -0700
Subject: [PATCH 0379/1171] flag parameters add-on: remove epoll_create size
 param

Remove the size parameter from the new epoll_create syscall and renames the
syscall itself.  The updated test program follows.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
#include <fcntl.h>
#include <stdio.h>
#include <time.h>
#include <unistd.h>
#include <sys/syscall.h>

#ifndef __NR_epoll_create2
# ifdef __x86_64__
#  define __NR_epoll_create2 291
# elif defined __i386__
#  define __NR_epoll_create2 329
# else
#  error "need __NR_epoll_create2"
# endif
#endif

#define EPOLL_CLOEXEC O_CLOEXEC

int
main (void)
{
  int fd = syscall (__NR_epoll_create2, 0);
  if (fd == -1)
    {
      puts ("epoll_create2(0) failed");
      return 1;
    }
  int coe = fcntl (fd, F_GETFD);
  if (coe == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if (coe & FD_CLOEXEC)
    {
      puts ("epoll_create2(0) set close-on-exec flag");
      return 1;
    }
  close (fd);

  fd = syscall (__NR_epoll_create2, EPOLL_CLOEXEC);
  if (fd == -1)
    {
      puts ("epoll_create2(EPOLL_CLOEXEC) failed");
      return 1;
    }
  coe = fcntl (fd, F_GETFD);
  if (coe == -1)
    {
      puts ("fcntl failed");
      return 1;
    }
  if ((coe & FD_CLOEXEC) == 0)
    {
      puts ("epoll_create2(EPOLL_CLOEXEC) set close-on-exec flag");
      return 1;
    }
  close (fd);

  puts ("OK");

  return 0;
}
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Michael Kerrisk <mtk.manpages@googlemail.com>
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/ia32/ia32entry.S          |  2 +-
 arch/x86/kernel/syscall_table_32.S |  2 +-
 fs/eventpoll.c                     | 18 ++++++++++--------
 include/asm-x86/unistd_32.h        |  2 +-
 include/asm-x86/unistd_64.h        |  4 ++--
 include/linux/eventpoll.h          |  2 +-
 include/linux/syscalls.h           |  2 +-
 7 files changed, 17 insertions(+), 15 deletions(-)

diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S
index 4541073dd837..e4bd1793a5e4 100644
--- a/arch/x86/ia32/ia32entry.S
+++ b/arch/x86/ia32/ia32entry.S
@@ -828,7 +828,7 @@ ia32_sys_call_table:
 	.quad compat_sys_timerfd_gettime
 	.quad compat_sys_signalfd4
 	.quad sys_eventfd2
-	.quad sys_epoll_create2
+	.quad sys_epoll_create1
 	.quad sys_dup3			/* 330 */
 	.quad sys_pipe2
 	.quad sys_inotify_init1
diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S
index f59aba5ff0f0..d44395ff34c3 100644
--- a/arch/x86/kernel/syscall_table_32.S
+++ b/arch/x86/kernel/syscall_table_32.S
@@ -328,7 +328,7 @@ ENTRY(sys_call_table)
 	.long sys_timerfd_gettime
 	.long sys_signalfd4
 	.long sys_eventfd2
-	.long sys_epoll_create2
+	.long sys_epoll_create1
 	.long sys_dup3			/* 330 */
 	.long sys_pipe2
 	.long sys_inotify_init1
diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index 2fdad4204044..0c87474f7917 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -1046,7 +1046,7 @@ retry:
  * RB tree. With the current implementation, the "size" parameter is ignored
  * (besides sanity checks).
  */
-asmlinkage long sys_epoll_create2(int size, int flags)
+asmlinkage long sys_epoll_create1(int flags)
 {
 	int error, fd = -1;
 	struct eventpoll *ep;
@@ -1058,14 +1058,13 @@ asmlinkage long sys_epoll_create2(int size, int flags)
 		return -EINVAL;
 
 	DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_create(%d)\n",
-		     current, size));
+		     current, flags));
 
 	/*
-	 * Sanity check on the size parameter, and create the internal data
-	 * structure ( "struct eventpoll" ).
+	 * Create the internal data structure ( "struct eventpoll" ).
 	 */
-	error = -EINVAL;
-	if (size <= 0 || (error = ep_alloc(&ep)) < 0) {
+	error = ep_alloc(&ep);
+	if (error < 0) {
 		fd = error;
 		goto error_return;
 	}
@@ -1081,14 +1080,17 @@ asmlinkage long sys_epoll_create2(int size, int flags)
 
 error_return:
 	DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_create(%d) = %d\n",
-		     current, size, fd));
+		     current, flags, fd));
 
 	return fd;
 }
 
 asmlinkage long sys_epoll_create(int size)
 {
-	return sys_epoll_create2(size, 0);
+	if (size < 0)
+		return -EINVAL;
+
+	return sys_epoll_create1(0);
 }
 
 /*
diff --git a/include/asm-x86/unistd_32.h b/include/asm-x86/unistd_32.h
index b3daf503ab93..d7394673b772 100644
--- a/include/asm-x86/unistd_32.h
+++ b/include/asm-x86/unistd_32.h
@@ -334,7 +334,7 @@
 #define __NR_timerfd_gettime	326
 #define __NR_signalfd4		327
 #define __NR_eventfd2		328
-#define __NR_epoll_create2	329
+#define __NR_epoll_create1	329
 #define __NR_dup3		330
 #define __NR_pipe2		331
 #define __NR_inotify_init1	332
diff --git a/include/asm-x86/unistd_64.h b/include/asm-x86/unistd_64.h
index c8cb88d70c6b..3a341d791792 100644
--- a/include/asm-x86/unistd_64.h
+++ b/include/asm-x86/unistd_64.h
@@ -645,8 +645,8 @@ __SYSCALL(__NR_paccept, sys_paccept)
 __SYSCALL(__NR_signalfd4, sys_signalfd4)
 #define __NR_eventfd2				290
 __SYSCALL(__NR_eventfd2, sys_eventfd2)
-#define __NR_epoll_create2			291
-__SYSCALL(__NR_epoll_create2, sys_epoll_create2)
+#define __NR_epoll_create1			291
+__SYSCALL(__NR_epoll_create1, sys_epoll_create1)
 #define __NR_dup3				292
 __SYSCALL(__NR_dup3, sys_dup3)
 #define __NR_pipe2				293
diff --git a/include/linux/eventpoll.h b/include/linux/eventpoll.h
index 1cfaa40059c8..f1e1d3c47125 100644
--- a/include/linux/eventpoll.h
+++ b/include/linux/eventpoll.h
@@ -18,7 +18,7 @@
 #include <linux/fcntl.h>
 #include <linux/types.h>
 
-/* Flags for epoll_create2.  */
+/* Flags for epoll_create1.  */
 #define EPOLL_CLOEXEC O_CLOEXEC
 
 /* Valid opcodes to issue to sys_epoll_ctl() */
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 93a7e7f017a6..06f2bf76c030 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -431,7 +431,7 @@ asmlinkage long sys_poll(struct pollfd __user *ufds, unsigned int nfds,
 asmlinkage long sys_select(int n, fd_set __user *inp, fd_set __user *outp,
 			fd_set __user *exp, struct timeval __user *tvp);
 asmlinkage long sys_epoll_create(int size);
-asmlinkage long sys_epoll_create2(int size, int flags);
+asmlinkage long sys_epoll_create1(int flags);
 asmlinkage long sys_epoll_ctl(int epfd, int op, int fd,
 				struct epoll_event __user *event);
 asmlinkage long sys_epoll_wait(int epfd, struct epoll_event __user *events,

From 920519c1c31ca46ef6caab1a4be102ed0dfb5fbc Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Wed, 23 Jul 2008 21:29:44 -0700
Subject: [PATCH 0380/1171] serial/8250_gsc.c: add MODULE_LICENSE

This patch adds the missing MODULE_LICENSE("GPL").

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Acked-by: Alan Cox <alan@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/8250_gsc.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/serial/8250_gsc.c b/drivers/serial/8250_gsc.c
index 4eb7437a404a..0416ad3bc127 100644
--- a/drivers/serial/8250_gsc.c
+++ b/drivers/serial/8250_gsc.c
@@ -119,3 +119,5 @@ int __init probe_serial_gsc(void)
 }
 
 module_init(probe_serial_gsc);
+
+MODULE_LICENSE("GPL");

From 7500b1f602aad75901774a67a687ee985d85893f Mon Sep 17 00:00:00 2001
From: Aristeu Rozanski <arozansk@redhat.com>
Date: Wed, 23 Jul 2008 21:29:45 -0700
Subject: [PATCH 0381/1171] 8250: fix break handling for Intel 82571

Intel 82571 has a "Serial Over LAN" feature that doesn't properly
implements the receiving of break characters.  When a break is received,
it doesn't set UART_LSR_DR and unless another character is received, the
break won't be received by the application.

Signed-off-by: Aristeu Rozanski <arozansk@redhat.com>
Acked-by: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/8250.c | 17 ++++++++++++++---
 1 file changed, 14 insertions(+), 3 deletions(-)

diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c
index 27f34a9f9cb7..a97f1ae11f78 100644
--- a/drivers/serial/8250.c
+++ b/drivers/serial/8250.c
@@ -1293,7 +1293,18 @@ receive_chars(struct uart_8250_port *up, unsigned int *status)
 	char flag;
 
 	do {
-		ch = serial_inp(up, UART_RX);
+		if (likely(lsr & UART_LSR_DR))
+			ch = serial_inp(up, UART_RX);
+		else
+			/*
+			 * Intel 82571 has a Serial Over Lan device that will
+			 * set UART_LSR_BI without setting UART_LSR_DR when
+			 * it receives a break. To avoid reading from the
+			 * receive buffer without UART_LSR_DR bit set, we
+			 * just force the read character to be 0
+			 */
+			ch = 0;
+
 		flag = TTY_NORMAL;
 		up->port.icount.rx++;
 
@@ -1342,7 +1353,7 @@ receive_chars(struct uart_8250_port *up, unsigned int *status)
 
 ignore_char:
 		lsr = serial_inp(up, UART_LSR);
-	} while ((lsr & UART_LSR_DR) && (max_count-- > 0));
+	} while ((lsr & (UART_LSR_DR | UART_LSR_BI)) && (max_count-- > 0));
 	spin_unlock(&up->port.lock);
 	tty_flip_buffer_push(tty);
 	spin_lock(&up->port.lock);
@@ -1425,7 +1436,7 @@ serial8250_handle_port(struct uart_8250_port *up)
 
 	DEBUG_INTR("status = %x...", status);
 
-	if (status & UART_LSR_DR)
+	if (status & (UART_LSR_DR | UART_LSR_BI))
 		receive_chars(up, &status);
 	check_modem_status(up);
 	if (status & UART_LSR_THRE)

From b76c5a0717094f0a900d9afd8e36f7ad8dbba587 Mon Sep 17 00:00:00 2001
From: "Catalin(ux) M BOIE" <catab@embedromix.ro>
Date: Wed, 23 Jul 2008 21:29:46 -0700
Subject: [PATCH 0382/1171] serial: add support for a no-name 4 ports
 multiserial card

It is a no-name PCI card.  I found no reference to a producer so I used
"UNKNOWN_0x1584" as the name.

Full lspci:
01:07.0 0780: 10b5:9050 (rev 01)
        Subsystem: 10b5:1584
        Control: I/O+ Mem+ BusMaster- SpecCycle- MemWINV- VGASnoop- \
                ParErr- Stepping- SERR+ FastB2B-
        Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- \
                DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR-
        Interrupt: pin A routed to IRQ 10
        Region 1: I/O ports at ec00 [size=128]
        Region 2: I/O ports at e480 [size=32]
        Region 3: I/O ports at e400 [size=8]
        Capabilities: [40] Power Management version 1
                Flags: PMEClk- DSI- D1- D2- AuxCurrent=0mA \
                        PME(D0+,D1-,D2-,D3hot+,D3cold-)
                Status: D0 PME-Enable- DSel=0 DScale=0 PME-
        Capabilities: [48] #06 [0080]
        Capabilities: [4c] Vital Product Data

After:
0000:01:07.0: ttyS4 at I/O 0xe480 (irq = 10) is a 16550A
0000:01:07.0: ttyS5 at I/O 0xe488 (irq = 10) is a 16550A
0000:01:07.0: ttyS6 at I/O 0xe490 (irq = 10) is a 16550A
0000:01:07.0: ttyS7 at I/O 0xe498 (irq = 10) is a 16550A

Signed-off-by: Catalin(ux) M BOIE <catab@embedromix.ro>
Acked-by: Alan Cox <alan@redhat.com>
Acked-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/8250_pci.c | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

diff --git a/drivers/serial/8250_pci.c b/drivers/serial/8250_pci.c
index 1b36087665a2..c2f23933155b 100644
--- a/drivers/serial/8250_pci.c
+++ b/drivers/serial/8250_pci.c
@@ -767,6 +767,9 @@ pci_default_setup(struct serial_private *priv, struct pciserial_board *board,
 #define PCI_SUBDEVICE_ID_POCTAL232	0x0308
 #define PCI_SUBDEVICE_ID_POCTAL422	0x0408
 
+/* Unknown vendors/cards - this should not be in linux/pci_ids.h */
+#define PCI_SUBDEVICE_ID_UNKNOWN_0x1584	0x1584
+
 /*
  * Master list of serial port init/setup/exit quirks.
  * This does not describe the general nature of the port.
@@ -880,6 +883,15 @@ static struct pci_serial_quirk pci_serial_quirks[] __refdata = {
 		.setup		= pci_default_setup,
 		.exit		= __devexit_p(pci_plx9050_exit),
 	},
+	{
+		.vendor		= PCI_VENDOR_ID_PLX,
+		.device		= PCI_DEVICE_ID_PLX_9050,
+		.subvendor	= PCI_VENDOR_ID_PLX,
+		.subdevice	= PCI_SUBDEVICE_ID_UNKNOWN_0x1584,
+		.init		= pci_plx9050_init,
+		.setup		= pci_default_setup,
+		.exit		= __devexit_p(pci_plx9050_exit),
+	},
 	{
 		.vendor		= PCI_VENDOR_ID_PLX,
 		.device		= PCI_DEVICE_ID_PLX_ROMULUS,
@@ -2197,6 +2209,11 @@ static struct pci_device_id serial_pci_tbl[] = {
 	{	PCI_VENDOR_ID_PLX, PCI_DEVICE_ID_PLX_1077,
 		PCI_ANY_ID, PCI_ANY_ID, 0, 0,
 		pbn_b2_4_921600 },
+	/* Unknown card - subdevice 0x1584 */
+	{	PCI_VENDOR_ID_PLX, PCI_DEVICE_ID_PLX_9050,
+		PCI_VENDOR_ID_PLX,
+		PCI_SUBDEVICE_ID_UNKNOWN_0x1584, 0, 0,
+		pbn_b0_4_115200 },
 	{	PCI_VENDOR_ID_PLX, PCI_DEVICE_ID_PLX_9050,
 		PCI_SUBVENDOR_ID_KEYSPAN,
 		PCI_SUBDEVICE_ID_KEYSPAN_SX2, 0, 0,

From 377135912806ddc87d56d64fafa685f4063c45f1 Mon Sep 17 00:00:00 2001
From: "Maciej W. Rozycki" <macro@linux-mips.org>
Date: Wed, 23 Jul 2008 21:29:48 -0700
Subject: [PATCH 0383/1171] serial: Z85C30: avoid a hang at console switch-over

Changes to the generic console support code that happened a while ago
introduced a scenario where the initial console is used in parallel with
the final console during a brief period when switching between the two is
in progress.  During that time a message about the switch-over is printed.

With some combinations of chips, firmware and drivers, such as the Zilog
Z85C30 SCC used with the DECstation, a hang may happen because the
firmware used for the initial console may not expect the state of the chip
after it has been initialised by the driver.  This is not a bug in the
firmware, as some registers it would have to examine are write-only.

This is a workaround for the Z85C30 which reuses the power-management
callback to keep the transmitter of the line associated with the console
enabled.  It reflects the consensus reached in a discussion a while ago.

Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org>
Cc: Jiri Slaby <jirislaby@gmail.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/zs.c | 21 +++++++++++++++++++--
 1 file changed, 19 insertions(+), 2 deletions(-)

diff --git a/drivers/serial/zs.c b/drivers/serial/zs.c
index bd45b6230fd8..9e6a873f8203 100644
--- a/drivers/serial/zs.c
+++ b/drivers/serial/zs.c
@@ -787,7 +787,6 @@ static int zs_startup(struct uart_port *uport)
 	zport->regs[1] &= ~RxINT_MASK;
 	zport->regs[1] |= RxINT_ALL | TxINT_ENAB | EXT_INT_ENAB;
 	zport->regs[3] |= RxENABLE;
-	zport->regs[5] |= TxENAB;
 	zport->regs[15] |= BRKIE;
 	write_zsreg(zport, R1, zport->regs[1]);
 	write_zsreg(zport, R3, zport->regs[3]);
@@ -814,7 +813,6 @@ static void zs_shutdown(struct uart_port *uport)
 
 	spin_lock_irqsave(&scc->zlock, flags);
 
-	zport->regs[5] &= ~TxENAB;
 	zport->regs[3] &= ~RxENABLE;
 	write_zsreg(zport, R5, zport->regs[5]);
 	write_zsreg(zport, R3, zport->regs[3]);
@@ -959,6 +957,23 @@ static void zs_set_termios(struct uart_port *uport, struct ktermios *termios,
 	spin_unlock_irqrestore(&scc->zlock, flags);
 }
 
+/*
+ * Hack alert!
+ * Required solely so that the initial PROM-based console
+ * works undisturbed in parallel with this one.
+ */
+static void zs_pm(struct uart_port *uport, unsigned int state,
+		  unsigned int oldstate)
+{
+	struct zs_port *zport = to_zport(uport);
+
+	if (state < 3)
+		zport->regs[5] |= TxENAB;
+	else
+		zport->regs[5] &= ~TxENAB;
+	write_zsreg(zport, R5, zport->regs[5]);
+}
+
 
 static const char *zs_type(struct uart_port *uport)
 {
@@ -1041,6 +1056,7 @@ static struct uart_ops zs_ops = {
 	.startup	= zs_startup,
 	.shutdown	= zs_shutdown,
 	.set_termios	= zs_set_termios,
+	.pm		= zs_pm,
 	.type		= zs_type,
 	.release_port	= zs_release_port,
 	.request_port	= zs_request_port,
@@ -1190,6 +1206,7 @@ static int __init zs_console_setup(struct console *co, char *options)
 		return ret;
 
 	zs_reset(zport);
+	zs_pm(uport, 0, -1);
 
 	if (options)
 		uart_parse_options(options, &baud, &parity, &bits, &flow);

From e9a8f4d1de12633bfb71b5fee47745b32877b7b5 Mon Sep 17 00:00:00 2001
From: "Maciej W. Rozycki" <macro@linux-mips.org>
Date: Wed, 23 Jul 2008 21:29:49 -0700
Subject: [PATCH 0384/1171] serial: DZ11: avoid a hang at console switch-over

Changes to the generic console support code that happened a while ago
introduced a scenario where the initial console is used in parallel with
the final console during a brief period when switching between the two is
in progress.  During that time a message about the switch-over is printed.

With some combinations of chips, firmware and drivers, such as the DEC
DZ11 clone used with the DECstation, a hang may happen because the
firmware used for the initial console may not expect the state of the chip
after it has been initialised by the driver.

This is a workaround for the DZ11 which reuses the power-management
callback to keep the transmitter of the line associated with the console
enabled.  It reflects the consensus reached in a discussion a while ago.

Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org>
Cc: Jiri Slaby <jirislaby@gmail.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/dz.c | 24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)

diff --git a/drivers/serial/dz.c b/drivers/serial/dz.c
index a81d2c2ff8a2..6042b87797a1 100644
--- a/drivers/serial/dz.c
+++ b/drivers/serial/dz.c
@@ -642,6 +642,26 @@ static void dz_set_termios(struct uart_port *uport, struct ktermios *termios,
 	spin_unlock_irqrestore(&dport->port.lock, flags);
 }
 
+/*
+ * Hack alert!
+ * Required solely so that the initial PROM-based console
+ * works undisturbed in parallel with this one.
+ */
+static void dz_pm(struct uart_port *uport, unsigned int state,
+		  unsigned int oldstate)
+{
+	struct dz_port *dport = to_dport(uport);
+	unsigned long flags;
+
+	spin_lock_irqsave(&dport->port.lock, flags);
+	if (state < 3)
+		dz_start_tx(&dport->port);
+	else
+		dz_stop_tx(&dport->port);
+	spin_unlock_irqrestore(&dport->port.lock, flags);
+}
+
+
 static const char *dz_type(struct uart_port *uport)
 {
 	return "DZ";
@@ -738,6 +758,7 @@ static struct uart_ops dz_ops = {
 	.startup	= dz_startup,
 	.shutdown	= dz_shutdown,
 	.set_termios	= dz_set_termios,
+	.pm		= dz_pm,
 	.type		= dz_type,
 	.release_port	= dz_release_port,
 	.request_port	= dz_request_port,
@@ -861,7 +882,10 @@ static int __init dz_console_setup(struct console *co, char *options)
 	if (ret)
 		return ret;
 
+	spin_lock_init(&dport->port.lock);	/* For dz_pm().  */
+
 	dz_reset(dport);
+	dz_pm(uport, 0, -1);
 
 	if (options)
 		uart_parse_options(options, &baud, &parity, &bits, &flow);

From ae2d4c396e19f45918ed6e0900b031538d009823 Mon Sep 17 00:00:00 2001
From: Nye Liu <nyet@mrv.com>
Date: Wed, 23 Jul 2008 21:29:50 -0700
Subject: [PATCH 0385/1171] cpm1: don't send break on TX_STOP, don't interrupt
 RX/TX when adjusting termios parameters

Before setting STOP_TX, set _brkcr to 0 so the SMC does not send a break
character.  The driver appears to properly re-initialize _brkcr when the
SMC is restarted.

Do not interrupt RX/TX when the termios is being adjusted; it results in
corrupted characters appearing on the line.

Cc: Vitaly Bordug <vbordug@ru.mvista.com>
Cc: Scott Wood <scottwood@freescale.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Kumar Gala <galak@kernel.crashing.org>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/serial/cpm_uart/cpm_uart_core.c | 21 ++++++++++++++-------
 1 file changed, 14 insertions(+), 7 deletions(-)

diff --git a/drivers/serial/cpm_uart/cpm_uart_core.c b/drivers/serial/cpm_uart/cpm_uart_core.c
index 1ff80de177db..a4f86927a74b 100644
--- a/drivers/serial/cpm_uart/cpm_uart_core.c
+++ b/drivers/serial/cpm_uart/cpm_uart_core.c
@@ -435,10 +435,13 @@ static void cpm_uart_shutdown(struct uart_port *port)
 		}
 
 		/* Shut them really down and reinit buffer descriptors */
-		if (IS_SMC(pinfo))
+		if (IS_SMC(pinfo)) {
+			out_be16(&pinfo->smcup->smc_brkcr, 0);
 			cpm_line_cr_cmd(pinfo, CPM_CR_STOP_TX);
-		else
+		} else {
+			out_be16(&pinfo->sccup->scc_brkcr, 0);
 			cpm_line_cr_cmd(pinfo, CPM_CR_GRA_STOP_TX);
+		}
 
 		cpm_uart_initbd(pinfo);
 	}
@@ -554,9 +557,11 @@ static void cpm_uart_set_termios(struct uart_port *port,
 		 * enables, because we want to put them back if they were
 		 * present.
 		 */
-		prev_mode = in_be16(&smcp->smc_smcmr);
-		out_be16(&smcp->smc_smcmr, smcr_mk_clen(bits) | cval | SMCMR_SM_UART);
-		setbits16(&smcp->smc_smcmr, (prev_mode & (SMCMR_REN | SMCMR_TEN)));
+		prev_mode = in_be16(&smcp->smc_smcmr) & (SMCMR_REN | SMCMR_TEN);
+		/* Output in *one* operation, so we don't interrupt RX/TX if they
+		 * were already enabled. */
+		out_be16(&smcp->smc_smcmr, smcr_mk_clen(bits) | cval |
+		    SMCMR_SM_UART | prev_mode);
 	} else {
 		out_be16(&sccp->scc_psmr, (sbits << 12) | scval);
 	}
@@ -1198,12 +1203,14 @@ static int __init cpm_uart_console_setup(struct console *co, char *options)
 	udbg_putc = NULL;
 #endif
 
-	cpm_line_cr_cmd(pinfo, CPM_CR_STOP_TX);
-
 	if (IS_SMC(pinfo)) {
+		out_be16(&pinfo->smcup->smc_brkcr, 0);
+		cpm_line_cr_cmd(pinfo, CPM_CR_STOP_TX);
 		clrbits8(&pinfo->smcp->smc_smcm, SMCM_RX | SMCM_TX);
 		clrbits16(&pinfo->smcp->smc_smcmr, SMCMR_REN | SMCMR_TEN);
 	} else {
+		out_be16(&pinfo->sccup->scc_brkcr, 0);
+		cpm_line_cr_cmd(pinfo, CPM_CR_GRA_STOP_TX);
 		clrbits16(&pinfo->sccp->scc_sccm, UART_SCCM_TX | UART_SCCM_RX);
 		clrbits32(&pinfo->sccp->scc_gsmrl, SCC_GSMRL_ENR | SCC_GSMRL_ENT);
 	}

From 708d8cefd0f6d8dc13027f899e865ccfa5f63871 Mon Sep 17 00:00:00 2001
From: Andre Haupt <andre@bitwigglers.org>
Date: Wed, 23 Jul 2008 21:29:51 -0700
Subject: [PATCH 0386/1171] stallion: removed unused variable

Signed-off-by: Andre Haupt <andre@bitwigglers.org>
Acked-by: Alan Cox <alan@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/stallion.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/char/stallion.c b/drivers/char/stallion.c
index b976248e1072..19db1eb87c26 100644
--- a/drivers/char/stallion.c
+++ b/drivers/char/stallion.c
@@ -1256,7 +1256,6 @@ static int stl_tiocmset(struct tty_struct *tty, struct file *file,
 static int stl_ioctl(struct tty_struct *tty, struct file *file, unsigned int cmd, unsigned long arg)
 {
 	struct stlport	*portp;
-	unsigned int	ival;
 	int		rc;
 	void __user *argp = (void __user *)arg;
 

From a61f5345eba34772a71523227de890a28410f320 Mon Sep 17 00:00:00 2001
From: Chen Gong <g.chen@freescale.com>
Date: Wed, 23 Jul 2008 21:29:52 -0700
Subject: [PATCH 0387/1171] spi: spi_mpc83xx clockrate fixes

This updates the SPI clock rate calculations for the spi_mpc83xx driver.
Some boundary conditions were wrong, and in several cases divide-by-16
wasn't always needed

Signed-off-by: Chen Gong <g.chen@freescale.com>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/spi/spi_mpc83xx.c | 29 ++++++++++++++++-------------
 1 file changed, 16 insertions(+), 13 deletions(-)

diff --git a/drivers/spi/spi_mpc83xx.c b/drivers/spi/spi_mpc83xx.c
index 6832da6f7109..070c6219e2d6 100644
--- a/drivers/spi/spi_mpc83xx.c
+++ b/drivers/spi/spi_mpc83xx.c
@@ -266,21 +266,24 @@ int mpc83xx_spi_setup_transfer(struct spi_device *spi, struct spi_transfer *t)
 
 	cs->hw_mode |= SPMODE_LEN(bits_per_word);
 
-	if ((mpc83xx_spi->spibrg / hz) >= 64) {
-		pm = mpc83xx_spi->spibrg / (hz * 64) - 1;
-		if (pm > 0x0f) {
-			dev_err(&spi->dev, "Requested speed is too "
-				"low: %d Hz. Will use %d Hz instead.\n",
-				hz, mpc83xx_spi->spibrg / 1024);
-			pm = 0x0f;
+	if ((mpc83xx_spi->spibrg / hz) > 64) {
+		pm = mpc83xx_spi->spibrg / (hz * 64);
+		if (pm > 16) {
+			cs->hw_mode |= SPMODE_DIV16;
+			pm /= 16;
+			if (pm > 16) {
+				dev_err(&spi->dev, "Requested speed is too "
+					"low: %d Hz. Will use %d Hz instead.\n",
+					hz, mpc83xx_spi->spibrg / 1024);
+				pm = 16;
+			}
 		}
-		cs->hw_mode |= SPMODE_PM(pm) | SPMODE_DIV16;
-	} else {
+	} else
 		pm = mpc83xx_spi->spibrg / (hz * 4);
-		if (pm)
-			pm--;
-		cs->hw_mode |= SPMODE_PM(pm);
-	}
+	if (pm)
+		pm--;
+
+	cs->hw_mode |= SPMODE_PM(pm);
 	regval =  mpc83xx_spi_read_reg(&mpc83xx_spi->base->mode);
 	if (cs->hw_mode != regval) {
 		unsigned long flags;

From 166a375b657b7af494f4ce3f72c4d2002180da44 Mon Sep 17 00:00:00 2001
From: Roel Kluin <12o3l@tiscali.nl>
Date: Wed, 23 Jul 2008 21:29:53 -0700
Subject: [PATCH 0388/1171] xilinx_spi: test below 0 on unsigned irq in
 xilinx_spi_probe()

xilinx_spi->irq is unsigned, so the test fails

Signed-off-by: Roel Kluin <12o3l@tiscali.nl>
Cc: David Brownell <dbrownell@users.sourceforge.net>
Cc: Andrei Konovalov <akonovalov@ru.mvista.com>
Cc: Yuri Frolov <yfrolov@ru.mvista.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/spi/xilinx_spi.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/spi/xilinx_spi.c b/drivers/spi/xilinx_spi.c
index 113a0468ffcb..68d6f4988fb5 100644
--- a/drivers/spi/xilinx_spi.c
+++ b/drivers/spi/xilinx_spi.c
@@ -353,11 +353,12 @@ static int __init xilinx_spi_probe(struct platform_device *dev)
 		goto put_master;
 	}
 
-	xspi->irq = platform_get_irq(dev, 0);
-	if (xspi->irq < 0) {
+	ret = platform_get_irq(dev, 0);
+	if (ret < 0) {
 		ret = -ENXIO;
 		goto unmap_io;
 	}
+	xspi->irq = ret;
 
 	master->bus_num = pdata->bus_num;
 	master->num_chipselect = pdata->num_chipselect;

From 6291fe2abce4689d6ee7cbaea16692c79bf0d01b Mon Sep 17 00:00:00 2001
From: "Robert P. J. Day" <rpjday@crashcourse.ca>
Date: Wed, 23 Jul 2008 21:29:53 -0700
Subject: [PATCH 0389/1171] SPI Kconfig simplifications

Use "if SPI_MASTER" to remove numerous dependencies.

[dbrownell@users.sourceforge.net: remove a couple now-needless EXPERIMENTAL dependencies too]
Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/spi/Kconfig | 45 +++++++++++++++++++++++----------------------
 1 file changed, 23 insertions(+), 22 deletions(-)

diff --git a/drivers/spi/Kconfig b/drivers/spi/Kconfig
index 66ec5d8808de..2303521b4f09 100644
--- a/drivers/spi/Kconfig
+++ b/drivers/spi/Kconfig
@@ -49,25 +49,26 @@ config SPI_MASTER
 	  controller and the protocol drivers for the SPI slave chips
 	  that are connected.
 
+if SPI_MASTER
+
 comment "SPI Master Controller Drivers"
-	depends on SPI_MASTER
 
 config SPI_ATMEL
 	tristate "Atmel SPI Controller"
-	depends on (ARCH_AT91 || AVR32) && SPI_MASTER
+	depends on (ARCH_AT91 || AVR32)
 	help
 	  This selects a driver for the Atmel SPI Controller, present on
 	  many AT32 (AVR32) and AT91 (ARM) chips.
 
 config SPI_BFIN
 	tristate "SPI controller driver for ADI Blackfin5xx"
-	depends on SPI_MASTER && BLACKFIN
+	depends on BLACKFIN
 	help
 	  This is the SPI controller master driver for Blackfin 5xx processor.
 
 config SPI_AU1550
 	tristate "Au1550/Au12x0 SPI Controller"
-	depends on SPI_MASTER && (SOC_AU1550 || SOC_AU1200) && EXPERIMENTAL
+	depends on (SOC_AU1550 || SOC_AU1200) && EXPERIMENTAL
 	select SPI_BITBANG
 	help
 	  If you say yes to this option, support will be included for the
@@ -78,7 +79,6 @@ config SPI_AU1550
 
 config SPI_BITBANG
 	tristate "Bitbanging SPI master"
-	depends on SPI_MASTER && EXPERIMENTAL
 	help
 	  With a few GPIO pins, your system can bitbang the SPI protocol.
 	  Select this to get SPI support through I/O pins (GPIO, parallel
@@ -92,7 +92,7 @@ config SPI_BITBANG
 
 config SPI_BUTTERFLY
 	tristate "Parallel port adapter for AVR Butterfly (DEVELOPMENT)"
-	depends on SPI_MASTER && PARPORT && EXPERIMENTAL
+	depends on PARPORT
 	select SPI_BITBANG
 	help
 	  This uses a custom parallel port cable to connect to an AVR
@@ -102,14 +102,14 @@ config SPI_BUTTERFLY
 
 config SPI_IMX
 	tristate "Freescale iMX SPI controller"
-	depends on SPI_MASTER && ARCH_IMX && EXPERIMENTAL
+	depends on ARCH_IMX && EXPERIMENTAL
 	help
 	  This enables using the Freescale iMX SPI controller in master
 	  mode.
 
 config SPI_LM70_LLP
 	tristate "Parallel port adapter for LM70 eval board (DEVELOPMENT)"
-	depends on SPI_MASTER && PARPORT && EXPERIMENTAL
+	depends on PARPORT && EXPERIMENTAL
 	select SPI_BITBANG
 	help
 	  This driver supports the NS LM70 LLP Evaluation Board,
@@ -118,14 +118,14 @@ config SPI_LM70_LLP
 
 config SPI_MPC52xx_PSC
 	tristate "Freescale MPC52xx PSC SPI controller"
-	depends on SPI_MASTER && PPC_MPC52xx && EXPERIMENTAL
+	depends on PPC_MPC52xx && EXPERIMENTAL
 	help
 	  This enables using the Freescale MPC52xx Programmable Serial
 	  Controller in master SPI mode.
 
 config SPI_MPC83xx
 	tristate "Freescale MPC83xx/QUICC Engine SPI controller"
-	depends on SPI_MASTER && (PPC_83xx || QUICC_ENGINE) && EXPERIMENTAL
+	depends on (PPC_83xx || QUICC_ENGINE) && EXPERIMENTAL
 	help
 	  This enables using the Freescale MPC83xx and QUICC Engine SPI
 	  controllers in master mode.
@@ -137,21 +137,21 @@ config SPI_MPC83xx
 
 config SPI_OMAP_UWIRE
 	tristate "OMAP1 MicroWire"
-	depends on SPI_MASTER && ARCH_OMAP1
+	depends on ARCH_OMAP1
 	select SPI_BITBANG
 	help
 	  This hooks up to the MicroWire controller on OMAP1 chips.
 
 config SPI_OMAP24XX
 	tristate "McSPI driver for OMAP24xx/OMAP34xx"
-	depends on SPI_MASTER && (ARCH_OMAP24XX || ARCH_OMAP34XX)
+	depends on ARCH_OMAP24XX || ARCH_OMAP34XX
 	help
 	  SPI master controller for OMAP24xx/OMAP34xx Multichannel SPI
 	  (McSPI) modules.
 
 config SPI_PXA2XX
 	tristate "PXA2xx SSP SPI master"
-	depends on SPI_MASTER && ARCH_PXA && EXPERIMENTAL
+	depends on ARCH_PXA && EXPERIMENTAL
 	select PXA_SSP
 	help
 	  This enables using a PXA2xx SSP port as a SPI master controller.
@@ -160,14 +160,14 @@ config SPI_PXA2XX
 
 config SPI_S3C24XX
 	tristate "Samsung S3C24XX series SPI"
-	depends on SPI_MASTER && ARCH_S3C2410 && EXPERIMENTAL
+	depends on ARCH_S3C2410 && EXPERIMENTAL
 	select SPI_BITBANG
 	help
 	  SPI driver for Samsung S3C24XX series ARM SoCs
 
 config SPI_S3C24XX_GPIO
 	tristate "Samsung S3C24XX series SPI by GPIO"
-	depends on SPI_MASTER && ARCH_S3C2410 && EXPERIMENTAL
+	depends on ARCH_S3C2410 && EXPERIMENTAL
 	select SPI_BITBANG
 	help
 	  SPI driver for Samsung S3C24XX series ARM SoCs using
@@ -177,20 +177,20 @@ config SPI_S3C24XX_GPIO
 
 config SPI_SH_SCI
 	tristate "SuperH SCI SPI controller"
-	depends on SPI_MASTER && SUPERH
+	depends on SUPERH
 	select SPI_BITBANG
 	help
 	  SPI driver for SuperH SCI blocks.
 
 config SPI_TXX9
 	tristate "Toshiba TXx9 SPI controller"
-	depends on SPI_MASTER && GENERIC_GPIO && CPU_TX49XX
+	depends on GENERIC_GPIO && CPU_TX49XX
 	help
 	  SPI driver for Toshiba TXx9 MIPS SoCs
 
 config SPI_XILINX
 	tristate "Xilinx SPI controller"
-	depends on SPI_MASTER && XILINX_VIRTEX && EXPERIMENTAL
+	depends on XILINX_VIRTEX && EXPERIMENTAL
 	select SPI_BITBANG
 	help
 	  This exposes the SPI controller IP from the Xilinx EDK.
@@ -207,11 +207,10 @@ config SPI_XILINX
 # being probably the most widely used ones.
 #
 comment "SPI Protocol Masters"
-	depends on SPI_MASTER
 
 config SPI_AT25
 	tristate "SPI EEPROMs from most vendors"
-	depends on SPI_MASTER && SYSFS
+	depends on SYSFS
 	help
 	  Enable this driver to get read/write support to most SPI EEPROMs,
 	  after you configure the board init code to know about each eeprom
@@ -222,7 +221,7 @@ config SPI_AT25
 
 config SPI_SPIDEV
 	tristate "User mode SPI device driver support"
-	depends on SPI_MASTER && EXPERIMENTAL
+	depends on EXPERIMENTAL
 	help
 	  This supports user mode SPI protocol drivers.
 
@@ -231,7 +230,7 @@ config SPI_SPIDEV
 
 config SPI_TLE62X0
 	tristate "Infineon TLE62X0 (for power switching)"
-	depends on SPI_MASTER && SYSFS
+	depends on SYSFS
 	help
 	  SPI driver for Infineon TLE62X0 series line driver chips,
 	  such as the TLE6220, TLE6230 and TLE6240.  This provides a
@@ -242,6 +241,8 @@ config SPI_TLE62X0
 # Add new SPI protocol masters in alphabetical order above this line
 #
 
+endif # SPI_MASTER
+
 # (slave support would go here)
 
 endif # SPI

From 102eb97564c73ea73645b38599c5cbe6f54b030c Mon Sep 17 00:00:00 2001
From: Grant Likely <grant.likely@secretlab.ca>
Date: Wed, 23 Jul 2008 21:29:55 -0700
Subject: [PATCH 0390/1171] spi: make spi_board_info.modalias a char array

Currently, 'modalias' in the spi_device structure is a 'const char *'.
The spi_new_device() function fills in the modalias value from a passed in
spi_board_info data block.  Since it is a pointer copy, the new spi_device
remains dependent on the spi_board_info structure after the new spi_device
is registered (no other fields in spi_device directly depend on the
spi_board_info structure; all of the other data is copied).

This causes a problem when dynamically propulating the list of attached
SPI devices.  For example, in arch/powerpc, the list of SPI devices can be
populated from data in the device tree.  With the current code, the device
tree adapter must kmalloc() a new spi_board_info structure for each new
SPI device it finds in the device tree, and there is no simple mechanism
in place for keeping track of these allocations.

This patch changes modalias from a 'const char *' to a fixed char array.
By copying the modalias string instead of referencing it, the dependency
on the spi_board_info structure is eliminated and an outside caller does
not need to maintain a separate spi_board_info allocation for each device.

If searched through the code to the best of my ability for any references
to modalias which may be affected by this change and haven't found
anything.  It has been tested with the lite5200b platform in arch/powerpc.

[dbrownell@users.sourceforge.net: cope with linux-next changes: KOBJ_NAME_LEN obliterated, etc]
Signed-off-by: Grant Likely <grant.likely@secretlab.ca>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/spi/spi.c       | 4 +++-
 include/linux/spi/spi.h | 2 +-
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c
index 1771b2456bfa..ecca4a6a6f94 100644
--- a/drivers/spi/spi.c
+++ b/drivers/spi/spi.c
@@ -218,6 +218,8 @@ struct spi_device *spi_new_device(struct spi_master *master,
 	if (!spi_master_get(master))
 		return NULL;
 
+	WARN_ON(strlen(chip->modalias) >= sizeof(proxy->modalias));
+
 	proxy = kzalloc(sizeof *proxy, GFP_KERNEL);
 	if (!proxy) {
 		dev_err(dev, "can't alloc dev for cs%d\n",
@@ -229,7 +231,7 @@ struct spi_device *spi_new_device(struct spi_master *master,
 	proxy->max_speed_hz = chip->max_speed_hz;
 	proxy->mode = chip->mode;
 	proxy->irq = chip->irq;
-	proxy->modalias = chip->modalias;
+	strlcpy(proxy->modalias, chip->modalias, sizeof(proxy->modalias));
 
 	snprintf(proxy->dev.bus_id, sizeof proxy->dev.bus_id,
 			"%s.%u", master->dev.bus_id,
diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h
index b9a76c972084..a9cc29d46653 100644
--- a/include/linux/spi/spi.h
+++ b/include/linux/spi/spi.h
@@ -82,7 +82,7 @@ struct spi_device {
 	int			irq;
 	void			*controller_state;
 	void			*controller_data;
-	const char		*modalias;
+	char			modalias[32];
 
 	/*
 	 * likely need more hooks for more protocol options affecting how

From 4ef754b7d7971a704d5b1b4608839da1bae37e5e Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Wed, 23 Jul 2008 21:29:55 -0700
Subject: [PATCH 0391/1171] spidev: BKL removal

Another step to removing ->ioctl and to removing the BKL

[dbrownell@users.sourceforge.net: take final step; BKL not needed]
Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/spi/spidev.c | 19 +++++++++++++------
 1 file changed, 13 insertions(+), 6 deletions(-)

diff --git a/drivers/spi/spidev.c b/drivers/spi/spidev.c
index 2833fd772a24..e5e0cfed5e3b 100644
--- a/drivers/spi/spidev.c
+++ b/drivers/spi/spidev.c
@@ -228,7 +228,6 @@ static int spidev_message(struct spidev_data *spidev,
 	 * We walk the array of user-provided transfers, using each one
 	 * to initialize a kernel version of the same transfer.
 	 */
-	mutex_lock(&spidev->buf_lock);
 	buf = spidev->buffer;
 	total = 0;
 	for (n = n_xfers, k_tmp = k_xfers, u_tmp = u_xfers;
@@ -296,14 +295,12 @@ static int spidev_message(struct spidev_data *spidev,
 	status = total;
 
 done:
-	mutex_unlock(&spidev->buf_lock);
 	kfree(k_xfers);
 	return status;
 }
 
-static int
-spidev_ioctl(struct inode *inode, struct file *filp,
-		unsigned int cmd, unsigned long arg)
+static long
+spidev_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 {
 	int			err = 0;
 	int			retval = 0;
@@ -341,6 +338,14 @@ spidev_ioctl(struct inode *inode, struct file *filp,
 	if (spi == NULL)
 		return -ESHUTDOWN;
 
+	/* use the buffer lock here for triple duty:
+	 *  - prevent I/O (from us) so calling spi_setup() is safe;
+	 *  - prevent concurrent SPI_IOC_WR_* from morphing
+	 *    data fields while SPI_IOC_RD_* reads them;
+	 *  - SPI_IOC_MESSAGE needs the buffer locked "normally".
+	 */
+	mutex_lock(&spidev->buf_lock);
+
 	switch (cmd) {
 	/* read requests */
 	case SPI_IOC_RD_MODE:
@@ -456,6 +461,8 @@ spidev_ioctl(struct inode *inode, struct file *filp,
 		kfree(ioc);
 		break;
 	}
+
+	mutex_unlock(&spidev->buf_lock);
 	spi_dev_put(spi);
 	return retval;
 }
@@ -533,7 +540,7 @@ static struct file_operations spidev_fops = {
 	 */
 	.write =	spidev_write,
 	.read =		spidev_read,
-	.ioctl =	spidev_ioctl,
+	.unlocked_ioctl = spidev_ioctl,
 	.open =		spidev_open,
 	.release =	spidev_release,
 };

From 3a93a159c61e38a12f7ecbb3a25cf3f012abcf7a Mon Sep 17 00:00:00 2001
From: Manuel Lauss <mano@roarinelk.homelinux.net>
Date: Wed, 23 Jul 2008 21:29:56 -0700
Subject: [PATCH 0392/1171] spi: au1550_spi: proper platform device

Remove the Au1550 resource table and instead extract MMIO/IRQ/DMA
resources from platform resource information like any well-behaved
platform driver.

Signed-off-by: Manuel Lauss <mano@roarinelk.homelinux.net>
Signed-off-by: Jan Nikitenko <jan.nikitenko@gmail.com>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/spi/au1550_spi.c                  | 138 ++++++++++++----------
 include/asm-mips/mach-au1x00/au1550_spi.h |   1 -
 2 files changed, 74 insertions(+), 65 deletions(-)

diff --git a/drivers/spi/au1550_spi.c b/drivers/spi/au1550_spi.c
index 072c4a595334..3860dd2fa5d9 100644
--- a/drivers/spi/au1550_spi.c
+++ b/drivers/spi/au1550_spi.c
@@ -26,6 +26,7 @@
 #include <linux/errno.h>
 #include <linux/device.h>
 #include <linux/platform_device.h>
+#include <linux/resource.h>
 #include <linux/spi/spi.h>
 #include <linux/spi/spi_bitbang.h>
 #include <linux/dma-mapping.h>
@@ -81,6 +82,7 @@ struct au1550_spi {
 	struct spi_master *master;
 	struct device *dev;
 	struct au1550_spi_info *pdata;
+	struct resource *ioarea;
 };
 
 
@@ -96,6 +98,8 @@ static dbdev_tab_t au1550_spi_mem_dbdev =
 	.dev_intpolarity	= 0
 };
 
+static int ddma_memid;	/* id to above mem dma device */
+
 static void au1550_spi_bits_handlers_set(struct au1550_spi *hw, int bpw);
 
 
@@ -732,6 +736,7 @@ static int __init au1550_spi_probe(struct platform_device *pdev)
 {
 	struct au1550_spi *hw;
 	struct spi_master *master;
+	struct resource *r;
 	int err = 0;
 
 	master = spi_alloc_master(&pdev->dev, sizeof(struct au1550_spi));
@@ -753,6 +758,52 @@ static int __init au1550_spi_probe(struct platform_device *pdev)
 		goto err_no_pdata;
 	}
 
+	r = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
+	if (!r) {
+		dev_err(&pdev->dev, "no IRQ\n");
+		err = -ENODEV;
+		goto err_no_iores;
+	}
+	hw->irq = r->start;
+
+	hw->usedma = 0;
+	r = platform_get_resource(pdev, IORESOURCE_DMA, 0);
+	if (r) {
+		hw->dma_tx_id = r->start;
+		r = platform_get_resource(pdev, IORESOURCE_DMA, 1);
+		if (r) {
+			hw->dma_rx_id = r->start;
+			if (usedma && ddma_memid) {
+				if (pdev->dev.dma_mask == NULL)
+					dev_warn(&pdev->dev, "no dma mask\n");
+				else
+					hw->usedma = 1;
+			}
+		}
+	}
+
+	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (!r) {
+		dev_err(&pdev->dev, "no mmio resource\n");
+		err = -ENODEV;
+		goto err_no_iores;
+	}
+
+	hw->ioarea = request_mem_region(r->start, sizeof(psc_spi_t),
+					pdev->name);
+	if (!hw->ioarea) {
+		dev_err(&pdev->dev, "Cannot reserve iomem region\n");
+		err = -ENXIO;
+		goto err_no_iores;
+	}
+
+	hw->regs = (psc_spi_t __iomem *)ioremap(r->start, sizeof(psc_spi_t));
+	if (!hw->regs) {
+		dev_err(&pdev->dev, "cannot ioremap\n");
+		err = -ENXIO;
+		goto err_ioremap;
+	}
+
 	platform_set_drvdata(pdev, hw);
 
 	init_completion(&hw->master_done);
@@ -763,66 +814,8 @@ static int __init au1550_spi_probe(struct platform_device *pdev)
 	hw->bitbang.master->setup = au1550_spi_setup;
 	hw->bitbang.txrx_bufs = au1550_spi_txrx_bufs;
 
-	switch (hw->pdata->bus_num) {
-	case 0:
-		hw->irq = AU1550_PSC0_INT;
-		hw->regs = (volatile psc_spi_t *)PSC0_BASE_ADDR;
-		hw->dma_rx_id = DSCR_CMD0_PSC0_RX;
-		hw->dma_tx_id = DSCR_CMD0_PSC0_TX;
-		break;
-	case 1:
-		hw->irq = AU1550_PSC1_INT;
-		hw->regs = (volatile psc_spi_t *)PSC1_BASE_ADDR;
-		hw->dma_rx_id = DSCR_CMD0_PSC1_RX;
-		hw->dma_tx_id = DSCR_CMD0_PSC1_TX;
-		break;
-	case 2:
-		hw->irq = AU1550_PSC2_INT;
-		hw->regs = (volatile psc_spi_t *)PSC2_BASE_ADDR;
-		hw->dma_rx_id = DSCR_CMD0_PSC2_RX;
-		hw->dma_tx_id = DSCR_CMD0_PSC2_TX;
-		break;
-	case 3:
-		hw->irq = AU1550_PSC3_INT;
-		hw->regs = (volatile psc_spi_t *)PSC3_BASE_ADDR;
-		hw->dma_rx_id = DSCR_CMD0_PSC3_RX;
-		hw->dma_tx_id = DSCR_CMD0_PSC3_TX;
-		break;
-	default:
-		dev_err(&pdev->dev, "Wrong bus_num of SPI\n");
-		err = -ENOENT;
-		goto err_no_pdata;
-	}
-
-	if (request_mem_region((unsigned long)hw->regs, sizeof(psc_spi_t),
-			pdev->name) == NULL) {
-		dev_err(&pdev->dev, "Cannot reserve iomem region\n");
-		err = -ENXIO;
-		goto err_no_iores;
-	}
-
-
-	if (usedma) {
-		if (pdev->dev.dma_mask == NULL)
-			dev_warn(&pdev->dev, "no dma mask\n");
-		else
-			hw->usedma = 1;
-	}
-
 	if (hw->usedma) {
-		/*
-		 * create memory device with 8 bits dev_devwidth
-		 * needed for proper byte ordering to spi fifo
-		 */
-		int memid = au1xxx_ddma_add_device(&au1550_spi_mem_dbdev);
-		if (!memid) {
-			dev_err(&pdev->dev,
-				"Cannot create dma 8 bit mem device\n");
-			err = -ENXIO;
-			goto err_dma_add_dev;
-		}
-
-		hw->dma_tx_ch = au1xxx_dbdma_chan_alloc(memid,
+		hw->dma_tx_ch = au1xxx_dbdma_chan_alloc(ddma_memid,
 			hw->dma_tx_id, NULL, (void *)hw);
 		if (hw->dma_tx_ch == 0) {
 			dev_err(&pdev->dev,
@@ -841,7 +834,7 @@ static int __init au1550_spi_probe(struct platform_device *pdev)
 
 
 		hw->dma_rx_ch = au1xxx_dbdma_chan_alloc(hw->dma_rx_id,
-			memid, NULL, (void *)hw);
+			ddma_memid, NULL, (void *)hw);
 		if (hw->dma_rx_ch == 0) {
 			dev_err(&pdev->dev,
 				"Cannot allocate rx dma channel\n");
@@ -874,7 +867,7 @@ static int __init au1550_spi_probe(struct platform_device *pdev)
 		goto err_no_irq;
 	}
 
-	master->bus_num = hw->pdata->bus_num;
+	master->bus_num = pdev->id;
 	master->num_chipselect = hw->pdata->num_chipselect;
 
 	/*
@@ -924,8 +917,11 @@ err_no_txdma_descr:
 		au1xxx_dbdma_chan_free(hw->dma_tx_ch);
 
 err_no_txdma:
-err_dma_add_dev:
-	release_mem_region((unsigned long)hw->regs, sizeof(psc_spi_t));
+	iounmap((void __iomem *)hw->regs);
+
+err_ioremap:
+	release_resource(hw->ioarea);
+	kfree(hw->ioarea);
 
 err_no_iores:
 err_no_pdata:
@@ -944,7 +940,9 @@ static int __exit au1550_spi_remove(struct platform_device *pdev)
 
 	spi_bitbang_stop(&hw->bitbang);
 	free_irq(hw->irq, hw);
-	release_mem_region((unsigned long)hw->regs, sizeof(psc_spi_t));
+	iounmap((void __iomem *)hw->regs);
+	release_resource(hw->ioarea);
+	kfree(hw->ioarea);
 
 	if (hw->usedma) {
 		au1550_spi_dma_rxtmp_free(hw);
@@ -971,12 +969,24 @@ static struct platform_driver au1550_spi_drv = {
 
 static int __init au1550_spi_init(void)
 {
+	/*
+	 * create memory device with 8 bits dev_devwidth
+	 * needed for proper byte ordering to spi fifo
+	 */
+	if (usedma) {
+		ddma_memid = au1xxx_ddma_add_device(&au1550_spi_mem_dbdev);
+		if (!ddma_memid)
+			printk(KERN_ERR "au1550-spi: cannot add memory"
+					"dbdma device\n");
+	}
 	return platform_driver_probe(&au1550_spi_drv, au1550_spi_probe);
 }
 module_init(au1550_spi_init);
 
 static void __exit au1550_spi_exit(void)
 {
+	if (usedma && ddma_memid)
+		au1xxx_ddma_del_device(ddma_memid);
 	platform_driver_unregister(&au1550_spi_drv);
 }
 module_exit(au1550_spi_exit);
diff --git a/include/asm-mips/mach-au1x00/au1550_spi.h b/include/asm-mips/mach-au1x00/au1550_spi.h
index 40e6c489833a..08e1958e9410 100644
--- a/include/asm-mips/mach-au1x00/au1550_spi.h
+++ b/include/asm-mips/mach-au1x00/au1550_spi.h
@@ -6,7 +6,6 @@
 #define _AU1550_SPI_H_
 
 struct au1550_spi_info {
-	s16 bus_num;		/* defines which PSC and IRQ to use */
 	u32 mainclk_hz;		/* main input clock frequency of PSC */
 	u16 num_chipselect;	/* number of chipselects supported */
 	void (*activate_cs)(struct au1550_spi_info *spi, int cs, int polarity);

From bbe48ecc7f6559318cfc6c023da225a0b0e14ab3 Mon Sep 17 00:00:00 2001
From: Jan Nikitenko <jan.nikitenko@gmail.com>
Date: Wed, 23 Jul 2008 21:29:57 -0700
Subject: [PATCH 0393/1171] spi: au1550_spi: improve pio transfer mode

Improve PIO transfer mode of au1550 spi controller by continuing of spi
transfer, instead of aborting transfer when transmit underflow interrupt
occurrs.

Verified by oscilloscope that the spi clock pauses on trasmit underflow,
so transfer continuation is perfectly valid even though au1550 datasheet
says that on tx underflow zeroes will be transfered.

Also make some error messages more specific.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Jan Nikitenko <jan.nikitenko@gmail.com>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/spi/au1550_spi.c | 69 ++++++++++++++++++++++++++++------------
 1 file changed, 49 insertions(+), 20 deletions(-)

diff --git a/drivers/spi/au1550_spi.c b/drivers/spi/au1550_spi.c
index 3860dd2fa5d9..9149689c79d9 100644
--- a/drivers/spi/au1550_spi.c
+++ b/drivers/spi/au1550_spi.c
@@ -484,9 +484,13 @@ static irqreturn_t au1550_spi_dma_irq_callback(struct au1550_spi *hw)
 		au1xxx_dbdma_reset(hw->dma_tx_ch);
 		au1550_spi_reset_fifos(hw);
 
-		dev_err(hw->dev,
-			"Unexpected SPI error: event=0x%x stat=0x%x!\n",
-			evnt, stat);
+		if (evnt == PSC_SPIEVNT_RO)
+			dev_err(hw->dev,
+				"dma transfer: receive FIFO overflow!\n");
+		else
+			dev_err(hw->dev,
+				"dma transfer: unexpected SPI error "
+				"(event=0x%x stat=0x%x)!\n", evnt, stat);
 
 		complete(&hw->master_done);
 		return IRQ_HANDLED;
@@ -596,17 +600,17 @@ static irqreturn_t au1550_spi_pio_irq_callback(struct au1550_spi *hw)
 
 	if ((evnt & (PSC_SPIEVNT_MM | PSC_SPIEVNT_RO
 				| PSC_SPIEVNT_RU | PSC_SPIEVNT_TO
-				| PSC_SPIEVNT_TU | PSC_SPIEVNT_SD))
+				| PSC_SPIEVNT_SD))
 			!= 0) {
-		dev_err(hw->dev,
-			"Unexpected SPI error: event=0x%x stat=0x%x!\n",
-			evnt, stat);
 		/*
 		 * due to an error we consider transfer as done,
 		 * so mask all events until before next transfer start
 		 */
 		au1550_spi_mask_ack_all(hw);
 		au1550_spi_reset_fifos(hw);
+		dev_err(hw->dev,
+			"pio transfer: unexpected SPI error "
+			"(event=0x%x stat=0x%x)!\n", evnt, stat);
 		complete(&hw->master_done);
 		return IRQ_HANDLED;
 	}
@@ -620,27 +624,50 @@ static irqreturn_t au1550_spi_pio_irq_callback(struct au1550_spi *hw)
 		stat = hw->regs->psc_spistat;
 		au_sync();
 
-		if ((stat & PSC_SPISTAT_RE) == 0 && hw->rx_count < hw->len) {
+		/*
+		 * Take care to not let the Rx FIFO overflow.
+		 *
+		 * We only write a byte if we have read one at least. Initially,
+		 * the write fifo is full, so we should read from the read fifo
+		 * first.
+		 * In case we miss a word from the read fifo, we should get a
+		 * RO event and should back out.
+		 */
+		if (!(stat & PSC_SPISTAT_RE) && hw->rx_count < hw->len) {
 			hw->rx_word(hw);
-			/* ack the receive request event */
-			hw->regs->psc_spievent = PSC_SPIEVNT_RR;
-			au_sync();
 			busy = 1;
-		}
 
-		if ((stat & PSC_SPISTAT_TF) == 0 && hw->tx_count < hw->len) {
-			hw->tx_word(hw);
-			/* ack the transmit request event */
-			hw->regs->psc_spievent = PSC_SPIEVNT_TR;
-			au_sync();
-			busy = 1;
+			if (!(stat & PSC_SPISTAT_TF) && hw->tx_count < hw->len)
+				hw->tx_word(hw);
 		}
 	} while (busy);
 
-	evnt = hw->regs->psc_spievent;
+	hw->regs->psc_spievent = PSC_SPIEVNT_RR | PSC_SPIEVNT_TR;
 	au_sync();
 
-	if (hw->rx_count >= hw->len || (evnt & PSC_SPIEVNT_MD) != 0) {
+	/*
+	 * Restart the SPI transmission in case of a transmit underflow.
+	 * This seems to work despite the notes in the Au1550 data book
+	 * of Figure 8-4 with flowchart for SPI master operation:
+	 *
+	 * """Note 1: An XFR Error Interrupt occurs, unless masked,
+	 * for any of the following events: Tx FIFO Underflow,
+	 * Rx FIFO Overflow, or Multiple-master Error
+	 *    Note 2: In case of a Tx Underflow Error, all zeroes are
+	 * transmitted."""
+	 *
+	 * By simply restarting the spi transfer on Tx Underflow Error,
+	 * we assume that spi transfer was paused instead of zeroes
+	 * transmittion mentioned in the Note 2 of Au1550 data book.
+	 */
+	if (evnt & PSC_SPIEVNT_TU) {
+		hw->regs->psc_spievent = PSC_SPIEVNT_TU | PSC_SPIEVNT_MD;
+		au_sync();
+		hw->regs->psc_spipcr = PSC_SPIPCR_MS;
+		au_sync();
+	}
+
+	if (hw->rx_count >= hw->len) {
 		/* transfer completed successfully */
 		au1550_spi_mask_ack_all(hw);
 		complete(&hw->master_done);
@@ -729,6 +756,8 @@ static void __init au1550_spi_setup_psc_as_spi(struct au1550_spi *hw)
 		stat = hw->regs->psc_spistat;
 		au_sync();
 	} while ((stat & PSC_SPISTAT_DR) == 0);
+
+	au1550_spi_reset_fifos(hw);
 }
 
 

From e0426e6a09954d205da2d674a3d368d2715e3afd Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Wed, 23 Jul 2008 21:29:58 -0700
Subject: [PATCH 0394/1171] vt: hold console_sem across sysfs operations

Hold console sem while creating/destroying sysfs files.  Serialisation is
so far done by BKL held in tty release_dev and chrdev_open, but no other
locks are held in open path.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Aristeu Rozanski <aris@ruivo.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/vt.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/char/vt.c b/drivers/char/vt.c
index e32a076d5f1f..ab53a1d4d895 100644
--- a/drivers/char/vt.c
+++ b/drivers/char/vt.c
@@ -2749,8 +2749,8 @@ static int con_open(struct tty_struct *tty, struct file *filp)
 				tty->termios->c_iflag |= IUTF8;
 			else
 				tty->termios->c_iflag &= ~IUTF8;
-			release_console_sem();
 			vcs_make_sysfs(tty);
+			release_console_sem();
 			return ret;
 		}
 	}
@@ -2775,8 +2775,8 @@ static void con_close(struct tty_struct *tty, struct file *filp)
 		if (vc)
 			vc->vc_tty = NULL;
 		tty->driver_data = NULL;
-		release_console_sem();
 		vcs_remove_sysfs(tty);
+		release_console_sem();
 		mutex_unlock(&tty_mutex);
 		/*
 		 * tty_mutex is released, but we still hold BKL, so there is

From f700d6e5e5549cb9349d22043f4bd153792c621f Mon Sep 17 00:00:00 2001
From: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Date: Wed, 23 Jul 2008 21:29:59 -0700
Subject: [PATCH 0395/1171] vt: do not update when the console is blanked

vt.c DO_UPDATE macro checks if the console is visible but doesn't check if
the console is blanked.

In fact updating fbcon while the console is blanked is not only
unnecessary but can even cause screen corruption.

Therefore I am adding a simple check on console_blanked in DO_UPDATE.

Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/vt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/char/vt.c b/drivers/char/vt.c
index ab53a1d4d895..cb8c90da3934 100644
--- a/drivers/char/vt.c
+++ b/drivers/char/vt.c
@@ -261,7 +261,7 @@ static void notify_update(struct vc_data *vc)
 #ifdef VT_BUF_VRAM_ONLY
 #define DO_UPDATE(vc)	0
 #else
-#define DO_UPDATE(vc)	CON_IS_VISIBLE(vc)
+#define DO_UPDATE(vc)	(CON_IS_VISIBLE(vc) && !console_blanked)
 #endif
 
 static inline unsigned short *screenpos(struct vc_data *vc, int offset, int viewed)

From 0293902a4d66fab27d0ddcc0766e05dae68f004e Mon Sep 17 00:00:00 2001
From: Wang Chen <wangchen@cn.fujitsu.com>
Date: Wed, 23 Jul 2008 21:30:01 -0700
Subject: [PATCH 0396/1171] I2O: handle sysfs_create_link() failures

Compile warning:
ignoring return value of `sysfs_create_link', declared with attribute warn_unused_result.

If sysfs_create_link failed, take care of the return value and do some
error handle after the failure.

Since sysfs_remove_link() will check whether a link exists, when removing the
link in error path, we don't need to care whether a link was created.

Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Cc: Markus Lidel <Markus.Lidel@shadowconnect.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/message/i2o/device.c | 54 ++++++++++++++++++++++++++++--------
 1 file changed, 42 insertions(+), 12 deletions(-)

diff --git a/drivers/message/i2o/device.c b/drivers/message/i2o/device.c
index 489d7c5c4965..8774c670e668 100644
--- a/drivers/message/i2o/device.c
+++ b/drivers/message/i2o/device.c
@@ -243,29 +243,41 @@ static int i2o_device_add(struct i2o_controller *c, i2o_lct_entry *entry)
 
 	/* create user entries for this device */
 	tmp = i2o_iop_find_device(i2o_dev->iop, i2o_dev->lct_data.user_tid);
-	if (tmp && (tmp != i2o_dev))
-		sysfs_create_link(&i2o_dev->device.kobj, &tmp->device.kobj,
-				  "user");
+	if (tmp && (tmp != i2o_dev)) {
+		rc = sysfs_create_link(&i2o_dev->device.kobj,
+				       &tmp->device.kobj, "user");
+		if (rc)
+			goto unreg_dev;
+	}
 
 	/* create user entries refering to this device */
 	list_for_each_entry(tmp, &c->devices, list)
 	    if ((tmp->lct_data.user_tid == i2o_dev->lct_data.tid)
-		&& (tmp != i2o_dev))
-		sysfs_create_link(&tmp->device.kobj,
-				  &i2o_dev->device.kobj, "user");
+		&& (tmp != i2o_dev)) {
+		rc = sysfs_create_link(&tmp->device.kobj,
+				       &i2o_dev->device.kobj, "user");
+		if (rc)
+			goto rmlink1;
+	}
 
 	/* create parent entries for this device */
 	tmp = i2o_iop_find_device(i2o_dev->iop, i2o_dev->lct_data.parent_tid);
-	if (tmp && (tmp != i2o_dev))
-		sysfs_create_link(&i2o_dev->device.kobj, &tmp->device.kobj,
-				  "parent");
+	if (tmp && (tmp != i2o_dev)) {
+		rc = sysfs_create_link(&i2o_dev->device.kobj,
+				       &tmp->device.kobj, "parent");
+		if (rc)
+			goto rmlink1;
+	}
 
 	/* create parent entries refering to this device */
 	list_for_each_entry(tmp, &c->devices, list)
 	    if ((tmp->lct_data.parent_tid == i2o_dev->lct_data.tid)
-		&& (tmp != i2o_dev))
-		sysfs_create_link(&tmp->device.kobj,
-				  &i2o_dev->device.kobj, "parent");
+		&& (tmp != i2o_dev)) {
+		rc = sysfs_create_link(&tmp->device.kobj,
+				       &i2o_dev->device.kobj, "parent");
+		if (rc)
+			goto rmlink2;
+	}
 
 	i2o_driver_notify_device_add_all(i2o_dev);
 
@@ -273,6 +285,24 @@ static int i2o_device_add(struct i2o_controller *c, i2o_lct_entry *entry)
 
 	return 0;
 
+rmlink2:
+	/* If link creating failed halfway, we loop whole list to cleanup.
+	 * And we don't care wrong removing of link, because sysfs_remove_link
+	 * will take care of it.
+	 */
+	list_for_each_entry(tmp, &c->devices, list) {
+		if (tmp->lct_data.parent_tid == i2o_dev->lct_data.tid)
+			sysfs_remove_link(&tmp->device.kobj, "parent");
+	}
+	sysfs_remove_link(&i2o_dev->device.kobj, "parent");
+rmlink1:
+	list_for_each_entry(tmp, &c->devices, list)
+		if (tmp->lct_data.user_tid == i2o_dev->lct_data.tid)
+			sysfs_remove_link(&tmp->device.kobj, "user");
+	sysfs_remove_link(&i2o_dev->device.kobj, "user");
+unreg_dev:
+	list_del(&i2o_dev->list);
+	device_unregister(&i2o_dev->device);
 err:
 	kfree(i2o_dev);
 	return rc;

From 746f1e558bc52b9693c1a1ecdab60f8392e5ff18 Mon Sep 17 00:00:00 2001
From: Michael Halcrow <mhalcrow@us.ibm.com>
Date: Wed, 23 Jul 2008 21:30:02 -0700
Subject: [PATCH 0397/1171] eCryptfs: Privileged kthread for lower file opens

eCryptfs would really like to have read-write access to all files in the
lower filesystem.  Right now, the persistent lower file may be opened
read-only if the attempt to open it read-write fails.  One way to keep
from having to do that is to have a privileged kthread that can open the
lower persistent file on behalf of the user opening the eCryptfs file;
this patch implements this functionality.

This patch will properly allow a less-privileged user to open the eCryptfs
file, followed by a more-privileged user opening the eCryptfs file, with
the first user only being able to read and the second user being able to
both read and write.  eCryptfs currently does this wrong; it will wind up
calling vfs_write() on a file that was opened read-only.  This is fixed in
this patch.

Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com>
Cc: Dave Kleikamp <shaggy@austin.ibm.com>
Cc: Serge Hallyn <serue@us.ibm.com>
Cc: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ecryptfs/Makefile          |   2 +-
 fs/ecryptfs/ecryptfs_kernel.h |  19 ++++
 fs/ecryptfs/file.c            |   7 ++
 fs/ecryptfs/kthread.c         | 203 ++++++++++++++++++++++++++++++++++
 fs/ecryptfs/main.c            |  42 +++----
 5 files changed, 251 insertions(+), 22 deletions(-)
 create mode 100644 fs/ecryptfs/kthread.c

diff --git a/fs/ecryptfs/Makefile b/fs/ecryptfs/Makefile
index 1e34a7fd4884..b4755a85996e 100644
--- a/fs/ecryptfs/Makefile
+++ b/fs/ecryptfs/Makefile
@@ -4,4 +4,4 @@
 
 obj-$(CONFIG_ECRYPT_FS) += ecryptfs.o
 
-ecryptfs-objs := dentry.o file.o inode.o main.o super.o mmap.o read_write.o crypto.o keystore.o messaging.o netlink.o miscdev.o debug.o
+ecryptfs-objs := dentry.o file.o inode.o main.o super.o mmap.o read_write.o crypto.o keystore.o messaging.o netlink.o miscdev.o kthread.o debug.o
diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h
index c15c25745e05..b4a0cccfdd7c 100644
--- a/fs/ecryptfs/ecryptfs_kernel.h
+++ b/fs/ecryptfs/ecryptfs_kernel.h
@@ -559,6 +559,20 @@ extern struct kmem_cache *ecryptfs_key_record_cache;
 extern struct kmem_cache *ecryptfs_key_sig_cache;
 extern struct kmem_cache *ecryptfs_global_auth_tok_cache;
 extern struct kmem_cache *ecryptfs_key_tfm_cache;
+extern struct kmem_cache *ecryptfs_open_req_cache;
+
+struct ecryptfs_open_req {
+#define ECRYPTFS_REQ_PROCESSED 0x00000001
+#define ECRYPTFS_REQ_DROPPED   0x00000002
+#define ECRYPTFS_REQ_ZOMBIE    0x00000004
+	u32 flags;
+	struct file **lower_file;
+	struct dentry *lower_dentry;
+	struct vfsmount *lower_mnt;
+	wait_queue_head_t wait;
+	struct mutex mux;
+	struct list_head kthread_ctl_list;
+};
 
 int ecryptfs_interpose(struct dentry *hidden_dentry,
 		       struct dentry *this_dentry, struct super_block *sb,
@@ -690,5 +704,10 @@ void ecryptfs_msg_ctx_alloc_to_free(struct ecryptfs_msg_ctx *msg_ctx);
 int
 ecryptfs_spawn_daemon(struct ecryptfs_daemon **daemon, uid_t euid,
 		      struct user_namespace *user_ns, struct pid *pid);
+int ecryptfs_init_kthread(void);
+void ecryptfs_destroy_kthread(void);
+int ecryptfs_privileged_open(struct file **lower_file,
+			     struct dentry *lower_dentry,
+			     struct vfsmount *lower_mnt);
 
 #endif /* #ifndef ECRYPTFS_KERNEL_H */
diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c
index 24749bf0668f..f0be29051528 100644
--- a/fs/ecryptfs/file.c
+++ b/fs/ecryptfs/file.c
@@ -192,6 +192,13 @@ static int ecryptfs_open(struct inode *inode, struct file *file)
 				      | ECRYPTFS_ENCRYPTED);
 	}
 	mutex_unlock(&crypt_stat->cs_mutex);
+	if ((ecryptfs_inode_to_private(inode)->lower_file->f_flags & O_RDONLY)
+	    && !(file->f_flags & O_RDONLY)) {
+		rc = -EPERM;
+		printk(KERN_WARNING "%s: Lower persistent file is RO; eCryptfs "
+		       "file must hence be opened RO\n", __func__);
+		goto out;
+	}
 	ecryptfs_set_file_lower(
 		file, ecryptfs_inode_to_private(inode)->lower_file);
 	if (S_ISDIR(ecryptfs_dentry->d_inode->i_mode)) {
diff --git a/fs/ecryptfs/kthread.c b/fs/ecryptfs/kthread.c
new file mode 100644
index 000000000000..c440c6b58b2d
--- /dev/null
+++ b/fs/ecryptfs/kthread.c
@@ -0,0 +1,203 @@
+/**
+ * eCryptfs: Linux filesystem encryption layer
+ *
+ * Copyright (C) 2008 International Business Machines Corp.
+ *   Author(s): Michael A. Halcrow <mahalcro@us.ibm.com>
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation; either version 2 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA
+ * 02111-1307, USA.
+ */
+
+#include <linux/kthread.h>
+#include <linux/freezer.h>
+#include <linux/wait.h>
+#include <linux/mount.h>
+#include "ecryptfs_kernel.h"
+
+struct kmem_cache *ecryptfs_open_req_cache;
+
+static struct ecryptfs_kthread_ctl {
+#define ECRYPTFS_KTHREAD_ZOMBIE 0x00000001
+	u32 flags;
+	struct mutex mux;
+	struct list_head req_list;
+	wait_queue_head_t wait;
+} ecryptfs_kthread_ctl;
+
+static struct task_struct *ecryptfs_kthread;
+
+/**
+ * ecryptfs_threadfn
+ * @ignored: ignored
+ *
+ * The eCryptfs kernel thread that has the responsibility of getting
+ * the lower persistent file with RW permissions.
+ *
+ * Returns zero on success; non-zero otherwise
+ */
+static int ecryptfs_threadfn(void *ignored)
+{
+	set_freezable();
+	while (1)  {
+		struct ecryptfs_open_req *req;
+
+		wait_event_freezable(
+			ecryptfs_kthread_ctl.wait,
+			(!list_empty(&ecryptfs_kthread_ctl.req_list)
+			 || kthread_should_stop()));
+		mutex_lock(&ecryptfs_kthread_ctl.mux);
+		if (ecryptfs_kthread_ctl.flags & ECRYPTFS_KTHREAD_ZOMBIE) {
+			mutex_unlock(&ecryptfs_kthread_ctl.mux);
+			goto out;
+		}
+		while (!list_empty(&ecryptfs_kthread_ctl.req_list)) {
+			req = list_first_entry(&ecryptfs_kthread_ctl.req_list,
+					       struct ecryptfs_open_req,
+					       kthread_ctl_list);
+			mutex_lock(&req->mux);
+			list_del(&req->kthread_ctl_list);
+			if (!(req->flags & ECRYPTFS_REQ_ZOMBIE)) {
+				dget(req->lower_dentry);
+				mntget(req->lower_mnt);
+				(*req->lower_file) = dentry_open(
+					req->lower_dentry, req->lower_mnt,
+					(O_RDWR | O_LARGEFILE));
+				req->flags |= ECRYPTFS_REQ_PROCESSED;
+			}
+			wake_up(&req->wait);
+			mutex_unlock(&req->mux);
+		}
+		mutex_unlock(&ecryptfs_kthread_ctl.mux);
+	}
+out:
+	return 0;
+}
+
+int ecryptfs_init_kthread(void)
+{
+	int rc = 0;
+
+	mutex_init(&ecryptfs_kthread_ctl.mux);
+	init_waitqueue_head(&ecryptfs_kthread_ctl.wait);
+	INIT_LIST_HEAD(&ecryptfs_kthread_ctl.req_list);
+	ecryptfs_kthread = kthread_run(&ecryptfs_threadfn, NULL,
+				       "ecryptfs-kthread");
+	if (IS_ERR(ecryptfs_kthread)) {
+		rc = PTR_ERR(ecryptfs_kthread);
+		printk(KERN_ERR "%s: Failed to create kernel thread; rc = [%d]"
+		       "\n", __func__, rc);
+	}
+	return rc;
+}
+
+void ecryptfs_destroy_kthread(void)
+{
+	struct ecryptfs_open_req *req;
+
+	mutex_lock(&ecryptfs_kthread_ctl.mux);
+	ecryptfs_kthread_ctl.flags |= ECRYPTFS_KTHREAD_ZOMBIE;
+	list_for_each_entry(req, &ecryptfs_kthread_ctl.req_list,
+			    kthread_ctl_list) {
+		mutex_lock(&req->mux);
+		req->flags |= ECRYPTFS_REQ_ZOMBIE;
+		wake_up(&req->wait);
+		mutex_unlock(&req->mux);
+	}
+	mutex_unlock(&ecryptfs_kthread_ctl.mux);
+	kthread_stop(ecryptfs_kthread);
+	wake_up(&ecryptfs_kthread_ctl.wait);
+}
+
+/**
+ * ecryptfs_privileged_open
+ * @lower_file: Result of dentry_open by root on lower dentry
+ * @lower_dentry: Lower dentry for file to open
+ * @lower_mnt: Lower vfsmount for file to open
+ *
+ * This function gets a r/w file opened againt the lower dentry.
+ *
+ * Returns zero on success; non-zero otherwise
+ */
+int ecryptfs_privileged_open(struct file **lower_file,
+			     struct dentry *lower_dentry,
+			     struct vfsmount *lower_mnt)
+{
+	struct ecryptfs_open_req *req;
+	int rc = 0;
+
+	/* Corresponding dput() and mntput() are done when the
+	 * persistent file is fput() when the eCryptfs inode is
+	 * destroyed. */
+	dget(lower_dentry);
+	mntget(lower_mnt);
+	(*lower_file) = dentry_open(lower_dentry, lower_mnt,
+				    (O_RDWR | O_LARGEFILE));
+	if (!IS_ERR(*lower_file))
+		goto out;
+	req = kmem_cache_alloc(ecryptfs_open_req_cache, GFP_KERNEL);
+	if (!req) {
+		rc = -ENOMEM;
+		goto out;
+	}
+	mutex_init(&req->mux);
+	req->lower_file = lower_file;
+	req->lower_dentry = lower_dentry;
+	req->lower_mnt = lower_mnt;
+	init_waitqueue_head(&req->wait);
+	req->flags = 0;
+	mutex_lock(&ecryptfs_kthread_ctl.mux);
+	if (ecryptfs_kthread_ctl.flags & ECRYPTFS_KTHREAD_ZOMBIE) {
+		rc = -EIO;
+		mutex_unlock(&ecryptfs_kthread_ctl.mux);
+		printk(KERN_ERR "%s: We are in the middle of shutting down; "
+		       "aborting privileged request to open lower file\n",
+			__func__);
+		goto out_free;
+	}
+	list_add_tail(&req->kthread_ctl_list, &ecryptfs_kthread_ctl.req_list);
+	mutex_unlock(&ecryptfs_kthread_ctl.mux);
+	wake_up(&ecryptfs_kthread_ctl.wait);
+	wait_event(req->wait, (req->flags != 0));
+	mutex_lock(&req->mux);
+	BUG_ON(req->flags == 0);
+	if (req->flags & ECRYPTFS_REQ_DROPPED
+	    || req->flags & ECRYPTFS_REQ_ZOMBIE) {
+		rc = -EIO;
+		printk(KERN_WARNING "%s: Privileged open request dropped\n",
+		       __func__);
+		goto out_unlock;
+	}
+	if (IS_ERR(*req->lower_file)) {
+		rc = PTR_ERR(*req->lower_file);
+		dget(lower_dentry);
+		mntget(lower_mnt);
+		(*lower_file) = dentry_open(lower_dentry, lower_mnt,
+					    (O_RDONLY | O_LARGEFILE));
+		if (IS_ERR(*lower_file)) {
+			rc = PTR_ERR(*req->lower_file);
+			(*lower_file) = NULL;
+			printk(KERN_WARNING "%s: Error attempting privileged "
+			       "open of lower file with either RW or RO "
+			       "perms; rc = [%d]. Giving up.\n",
+			       __func__, rc);
+		}
+	}
+out_unlock:
+	mutex_unlock(&req->mux);
+out_free:
+	kmem_cache_free(ecryptfs_open_req_cache, req);
+out:
+	return rc;
+}
diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c
index d603631601eb..f36ab2feea28 100644
--- a/fs/ecryptfs/main.c
+++ b/fs/ecryptfs/main.c
@@ -130,26 +130,12 @@ static int ecryptfs_init_persistent_file(struct dentry *ecryptfs_dentry)
 			ecryptfs_dentry_to_lower_mnt(ecryptfs_dentry);
 
 		lower_dentry = ecryptfs_dentry_to_lower(ecryptfs_dentry);
-		/* Corresponding dput() and mntput() are done when the
-		 * persistent file is fput() when the eCryptfs inode
-		 * is destroyed. */
-		dget(lower_dentry);
-		mntget(lower_mnt);
-		inode_info->lower_file = dentry_open(lower_dentry,
-						     lower_mnt,
-						     (O_RDWR | O_LARGEFILE));
-		if (IS_ERR(inode_info->lower_file)) {
-			dget(lower_dentry);
-			mntget(lower_mnt);
-			inode_info->lower_file = dentry_open(lower_dentry,
-							     lower_mnt,
-							     (O_RDONLY
-							      | O_LARGEFILE));
-		}
-		if (IS_ERR(inode_info->lower_file)) {
+		rc = ecryptfs_privileged_open(&inode_info->lower_file,
+						     lower_dentry, lower_mnt);
+		if (rc || IS_ERR(inode_info->lower_file)) {
 			printk(KERN_ERR "Error opening lower persistent file "
-			       "for lower_dentry [0x%p] and lower_mnt [0x%p]\n",
-			       lower_dentry, lower_mnt);
+			       "for lower_dentry [0x%p] and lower_mnt [0x%p]; "
+			       "rc = [%d]\n", lower_dentry, lower_mnt, rc);
 			rc = PTR_ERR(inode_info->lower_file);
 			inode_info->lower_file = NULL;
 		}
@@ -679,6 +665,11 @@ static struct ecryptfs_cache_info {
 		.name = "ecryptfs_key_tfm_cache",
 		.size = sizeof(struct ecryptfs_key_tfm),
 	},
+	{
+		.cache = &ecryptfs_open_req_cache,
+		.name = "ecryptfs_open_req_cache",
+		.size = sizeof(struct ecryptfs_open_req),
+	},
 };
 
 static void ecryptfs_free_kmem_caches(void)
@@ -795,11 +786,17 @@ static int __init ecryptfs_init(void)
 		printk(KERN_ERR "sysfs registration failed\n");
 		goto out_unregister_filesystem;
 	}
+	rc = ecryptfs_init_kthread();
+	if (rc) {
+		printk(KERN_ERR "%s: kthread initialization failed; "
+		       "rc = [%d]\n", __func__, rc);
+		goto out_do_sysfs_unregistration;
+	}
 	rc = ecryptfs_init_messaging(ecryptfs_transport);
 	if (rc) {
-		ecryptfs_printk(KERN_ERR, "Failure occured while attempting to "
+		printk(KERN_ERR "Failure occured while attempting to "
 				"initialize the eCryptfs netlink socket\n");
-		goto out_do_sysfs_unregistration;
+		goto out_destroy_kthread;
 	}
 	rc = ecryptfs_init_crypto();
 	if (rc) {
@@ -814,6 +811,8 @@ static int __init ecryptfs_init(void)
 	goto out;
 out_release_messaging:
 	ecryptfs_release_messaging(ecryptfs_transport);
+out_destroy_kthread:
+	ecryptfs_destroy_kthread();
 out_do_sysfs_unregistration:
 	do_sysfs_unregistration();
 out_unregister_filesystem:
@@ -833,6 +832,7 @@ static void __exit ecryptfs_exit(void)
 		printk(KERN_ERR "Failure whilst attempting to destroy crypto; "
 		       "rc = [%d]\n", rc);
 	ecryptfs_release_messaging(ecryptfs_transport);
+	ecryptfs_destroy_kthread();
 	do_sysfs_unregistration();
 	unregister_filesystem(&ecryptfs_fs_type);
 	ecryptfs_free_kmem_caches();

From 6c4c17b073cd4a5a61bc04329561632870bb21fc Mon Sep 17 00:00:00 2001
From: Tyler Hicks <tyhicks@linux.vnet.ibm.com>
Date: Wed, 23 Jul 2008 21:30:04 -0700
Subject: [PATCH 0398/1171] ecryptfs: discard ecryptfsd registration messages
 in miscdev

The userspace eCryptfs daemon sends HELO and QUIT messages to the kernel
for per-user daemon (un)registration.  These messages are required when
netlink is used as the transport, but (un)registration is handled by
opening and closing the device file when miscdev is the transport.  These
messages should be discarded in the miscdev transport so that a daemon
isn't registered twice.

Signed-off-by: Tyler Hicks <tyhicks@linux.vnet.ibm.com>
Cc: Michael Halcrow <mhalcrow@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ecryptfs/miscdev.c | 59 -------------------------------------------
 1 file changed, 59 deletions(-)

diff --git a/fs/ecryptfs/miscdev.c b/fs/ecryptfs/miscdev.c
index 09a4522f65e6..b484792a0996 100644
--- a/fs/ecryptfs/miscdev.c
+++ b/fs/ecryptfs/miscdev.c
@@ -357,46 +357,6 @@ out_unlock_daemon:
 	return rc;
 }
 
-/**
- * ecryptfs_miscdev_helo
- * @euid: effective user id of miscdevess sending helo packet
- * @user_ns: The namespace in which @euid applies
- * @pid: miscdevess id of miscdevess sending helo packet
- *
- * Returns zero on success; non-zero otherwise
- */
-static int ecryptfs_miscdev_helo(uid_t euid, struct user_namespace *user_ns,
-				 struct pid *pid)
-{
-	int rc;
-
-	rc = ecryptfs_process_helo(ECRYPTFS_TRANSPORT_MISCDEV, euid, user_ns,
-				   pid);
-	if (rc)
-		printk(KERN_WARNING "Error processing HELO; rc = [%d]\n", rc);
-	return rc;
-}
-
-/**
- * ecryptfs_miscdev_quit
- * @euid: effective user id of miscdevess sending quit packet
- * @user_ns: The namespace in which @euid applies
- * @pid: miscdevess id of miscdevess sending quit packet
- *
- * Returns zero on success; non-zero otherwise
- */
-static int ecryptfs_miscdev_quit(uid_t euid, struct user_namespace *user_ns,
-				 struct pid *pid)
-{
-	int rc;
-
-	rc = ecryptfs_process_quit(euid, user_ns, pid);
-	if (rc)
-		printk(KERN_WARNING
-		       "Error processing QUIT message; rc = [%d]\n", rc);
-	return rc;
-}
-
 /**
  * ecryptfs_miscdev_response - miscdevess response to message previously sent to daemon
  * @data: Bytes comprising struct ecryptfs_message
@@ -512,26 +472,7 @@ ecryptfs_miscdev_write(struct file *file, const char __user *buf,
 			       __func__, rc);
 		break;
 	case ECRYPTFS_MSG_HELO:
-		rc = ecryptfs_miscdev_helo(current->euid,
-					   current->nsproxy->user_ns,
-					   task_pid(current));
-		if (rc) {
-			printk(KERN_ERR "%s: Error attempting to process "
-			       "helo from pid [0x%p]; rc = [%d]\n", __func__,
-			       task_pid(current), rc);
-			goto out_free;
-		}
-		break;
 	case ECRYPTFS_MSG_QUIT:
-		rc = ecryptfs_miscdev_quit(current->euid,
-					   current->nsproxy->user_ns,
-					   task_pid(current));
-		if (rc) {
-			printk(KERN_ERR "%s: Error attempting to process "
-			       "quit from pid [0x%p]; rc = [%d]\n", __func__,
-			       task_pid(current), rc);
-			goto out_free;
-		}
 		break;
 	default:
 		ecryptfs_printk(KERN_WARNING, "Dropping miscdev "

From 982363c97f8cad7aea4c3d2cfebffc1cc2d2f166 Mon Sep 17 00:00:00 2001
From: Eric Sandeen <sandeen@redhat.com>
Date: Wed, 23 Jul 2008 21:30:04 -0700
Subject: [PATCH 0399/1171] ecryptfs: propagate key errors up at mount time

Mounting with invalid key signatures should probably fail, if they were
specifically requested but not available.

Also fix case checks in process_request_key_err() for the right sign of
the errnos, as spotted by Jan Tluka.

Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Reviewed-by: Jan Tluka <jtluka@redhat.com>
Acked-by: Michael Halcrow <mhalcrow@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ecryptfs/keystore.c | 9 ++++-----
 fs/ecryptfs/main.c     | 4 ++--
 2 files changed, 6 insertions(+), 7 deletions(-)

diff --git a/fs/ecryptfs/keystore.c b/fs/ecryptfs/keystore.c
index e82b457180be..f5b76a331b9c 100644
--- a/fs/ecryptfs/keystore.c
+++ b/fs/ecryptfs/keystore.c
@@ -44,15 +44,15 @@ static int process_request_key_err(long err_code)
 	int rc = 0;
 
 	switch (err_code) {
-	case ENOKEY:
+	case -ENOKEY:
 		ecryptfs_printk(KERN_WARNING, "No key\n");
 		rc = -ENOENT;
 		break;
-	case EKEYEXPIRED:
+	case -EKEYEXPIRED:
 		ecryptfs_printk(KERN_WARNING, "Key expired\n");
 		rc = -ETIME;
 		break;
-	case EKEYREVOKED:
+	case -EKEYREVOKED:
 		ecryptfs_printk(KERN_WARNING, "Key revoked\n");
 		rc = -EINVAL;
 		break;
@@ -963,8 +963,7 @@ int ecryptfs_keyring_auth_tok_for_sig(struct key **auth_tok_key,
 	if (!(*auth_tok_key) || IS_ERR(*auth_tok_key)) {
 		printk(KERN_ERR "Could not find key with description: [%s]\n",
 		       sig);
-		process_request_key_err(PTR_ERR(*auth_tok_key));
-		rc = -EINVAL;
+		rc = process_request_key_err(PTR_ERR(*auth_tok_key));
 		goto out;
 	}
 	(*auth_tok) = ecryptfs_get_key_payload_data(*auth_tok_key);
diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c
index f36ab2feea28..8876fe7c76e2 100644
--- a/fs/ecryptfs/main.c
+++ b/fs/ecryptfs/main.c
@@ -248,10 +248,11 @@ static int ecryptfs_init_global_auth_toks(
 			       "session keyring for sig specified in mount "
 			       "option: [%s]\n", global_auth_tok->sig);
 			global_auth_tok->flags |= ECRYPTFS_AUTH_TOK_INVALID;
-			rc = 0;
+			goto out;
 		} else
 			global_auth_tok->flags &= ~ECRYPTFS_AUTH_TOK_INVALID;
 	}
+out:
 	return rc;
 }
 
@@ -416,7 +417,6 @@ static int ecryptfs_parse_options(struct super_block *sb, char *options)
 		printk(KERN_WARNING "One or more global auth toks could not "
 		       "properly register; rc = [%d]\n", rc);
 	}
-	rc = 0;
 out:
 	return rc;
 }

From 8f2368095e25018838e1bf145041f58270ccd32e Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Wed, 23 Jul 2008 21:30:05 -0700
Subject: [PATCH 0400/1171] ecryptfs: string copy cleanup

Clean up overcomplicated string copy, which also gets rid of this
bogus warning:

fs/ecryptfs/main.c: In function 'ecryptfs_parse_options':
include/asm/arch/string_32.h:75: warning: array subscript is above array bounds

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Cc: Michael Halcrow <mhalcrow@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ecryptfs/main.c | 18 ++++++------------
 1 file changed, 6 insertions(+), 12 deletions(-)

diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c
index 8876fe7c76e2..10475d93ff53 100644
--- a/fs/ecryptfs/main.c
+++ b/fs/ecryptfs/main.c
@@ -301,7 +301,6 @@ static int ecryptfs_parse_options(struct super_block *sb, char *options)
 	char *cipher_name_dst;
 	char *cipher_name_src;
 	char *cipher_key_bytes_src;
-	int cipher_name_len;
 
 	if (!options) {
 		rc = -EINVAL;
@@ -382,17 +381,12 @@ static int ecryptfs_parse_options(struct super_block *sb, char *options)
 		goto out;
 	}
 	if (!cipher_name_set) {
-		cipher_name_len = strlen(ECRYPTFS_DEFAULT_CIPHER);
-		if (unlikely(cipher_name_len
-			     >= ECRYPTFS_MAX_CIPHER_NAME_SIZE)) {
-			rc = -EINVAL;
-			BUG();
-			goto out;
-		}
-		memcpy(mount_crypt_stat->global_default_cipher_name,
-		       ECRYPTFS_DEFAULT_CIPHER, cipher_name_len);
-		mount_crypt_stat->global_default_cipher_name[cipher_name_len]
-		    = '\0';
+		int cipher_name_len = strlen(ECRYPTFS_DEFAULT_CIPHER);
+
+		BUG_ON(cipher_name_len >= ECRYPTFS_MAX_CIPHER_NAME_SIZE);
+
+		strcpy(mount_crypt_stat->global_default_cipher_name,
+		       ECRYPTFS_DEFAULT_CIPHER);
 	}
 	if (!cipher_key_bytes_set) {
 		mount_crypt_stat->global_default_cipher_key_size = 0;

From 29335c6a41568d4708d4ec3b9187f9b6d302e5ea Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 23 Jul 2008 21:30:06 -0700
Subject: [PATCH 0401/1171] ecryptfs: crypto.c use unaligned byteorder helpers

Fixes the following sparse warnings:
fs/ecryptfs/crypto.c:1036:8: warning: cast to restricted __be32
fs/ecryptfs/crypto.c:1038:8: warning: cast to restricted __be32
fs/ecryptfs/crypto.c:1077:10: warning: cast to restricted __be32
fs/ecryptfs/crypto.c:1103:6: warning: incorrect type in assignment (different base types)
fs/ecryptfs/crypto.c:1105:6: warning: incorrect type in assignment (different base types)
fs/ecryptfs/crypto.c:1124:8: warning: incorrect type in assignment (different base types)
fs/ecryptfs/crypto.c:1241:21: warning: incorrect type in assignment (different base types)
fs/ecryptfs/crypto.c:1244:30: warning: incorrect type in assignment (different base types)
fs/ecryptfs/crypto.c:1414:23: warning: cast to restricted __be32
fs/ecryptfs/crypto.c:1417:32: warning: cast to restricted __be16

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Cc: Michael Halcrow <mhalcrow@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ecryptfs/crypto.c | 37 ++++++++++++++-----------------------
 1 file changed, 14 insertions(+), 23 deletions(-)

diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c
index e2832bc7869a..7b99917ffadc 100644
--- a/fs/ecryptfs/crypto.c
+++ b/fs/ecryptfs/crypto.c
@@ -33,6 +33,7 @@
 #include <linux/crypto.h>
 #include <linux/file.h>
 #include <linux/scatterlist.h>
+#include <asm/unaligned.h>
 #include "ecryptfs_kernel.h"
 
 static int
@@ -1032,10 +1033,8 @@ static int contains_ecryptfs_marker(char *data)
 {
 	u32 m_1, m_2;
 
-	memcpy(&m_1, data, 4);
-	m_1 = be32_to_cpu(m_1);
-	memcpy(&m_2, (data + 4), 4);
-	m_2 = be32_to_cpu(m_2);
+	m_1 = get_unaligned_be32(data);
+	m_2 = get_unaligned_be32(data + 4);
 	if ((m_1 ^ MAGIC_ECRYPTFS_MARKER) == m_2)
 		return 1;
 	ecryptfs_printk(KERN_DEBUG, "m_1 = [0x%.8x]; m_2 = [0x%.8x]; "
@@ -1073,8 +1072,7 @@ static int ecryptfs_process_flags(struct ecryptfs_crypt_stat *crypt_stat,
 	int i;
 	u32 flags;
 
-	memcpy(&flags, page_virt, 4);
-	flags = be32_to_cpu(flags);
+	flags = get_unaligned_be32(page_virt);
 	for (i = 0; i < ((sizeof(ecryptfs_flag_map)
 			  / sizeof(struct ecryptfs_flag_map_elem))); i++)
 		if (flags & ecryptfs_flag_map[i].file_flag) {
@@ -1100,11 +1098,9 @@ static void write_ecryptfs_marker(char *page_virt, size_t *written)
 
 	get_random_bytes(&m_1, (MAGIC_ECRYPTFS_MARKER_SIZE_BYTES / 2));
 	m_2 = (m_1 ^ MAGIC_ECRYPTFS_MARKER);
-	m_1 = cpu_to_be32(m_1);
-	memcpy(page_virt, &m_1, (MAGIC_ECRYPTFS_MARKER_SIZE_BYTES / 2));
-	m_2 = cpu_to_be32(m_2);
-	memcpy(page_virt + (MAGIC_ECRYPTFS_MARKER_SIZE_BYTES / 2), &m_2,
-	       (MAGIC_ECRYPTFS_MARKER_SIZE_BYTES / 2));
+	put_unaligned_be32(m_1, page_virt);
+	page_virt += (MAGIC_ECRYPTFS_MARKER_SIZE_BYTES / 2);
+	put_unaligned_be32(m_2, page_virt);
 	(*written) = MAGIC_ECRYPTFS_MARKER_SIZE_BYTES;
 }
 
@@ -1121,8 +1117,7 @@ write_ecryptfs_flags(char *page_virt, struct ecryptfs_crypt_stat *crypt_stat,
 			flags |= ecryptfs_flag_map[i].file_flag;
 	/* Version is in top 8 bits of the 32-bit flag vector */
 	flags |= ((((u8)crypt_stat->file_version) << 24) & 0xFF000000);
-	flags = cpu_to_be32(flags);
-	memcpy(page_virt, &flags, 4);
+	put_unaligned_be32(flags, page_virt);
 	(*written) = 4;
 }
 
@@ -1238,11 +1233,9 @@ ecryptfs_write_header_metadata(char *virt,
 	num_header_extents_at_front =
 		(u16)(crypt_stat->num_header_bytes_at_front
 		      / crypt_stat->extent_size);
-	header_extent_size = cpu_to_be32(header_extent_size);
-	memcpy(virt, &header_extent_size, 4);
+	put_unaligned_be32(header_extent_size, virt);
 	virt += 4;
-	num_header_extents_at_front = cpu_to_be16(num_header_extents_at_front);
-	memcpy(virt, &num_header_extents_at_front, 2);
+	put_unaligned_be16(num_header_extents_at_front, virt);
 	(*written) = 6;
 }
 
@@ -1410,15 +1403,13 @@ static int parse_header_metadata(struct ecryptfs_crypt_stat *crypt_stat,
 	u32 header_extent_size;
 	u16 num_header_extents_at_front;
 
-	memcpy(&header_extent_size, virt, sizeof(u32));
-	header_extent_size = be32_to_cpu(header_extent_size);
-	virt += sizeof(u32);
-	memcpy(&num_header_extents_at_front, virt, sizeof(u16));
-	num_header_extents_at_front = be16_to_cpu(num_header_extents_at_front);
+	header_extent_size = get_unaligned_be32(virt);
+	virt += sizeof(__be32);
+	num_header_extents_at_front = get_unaligned_be16(virt);
 	crypt_stat->num_header_bytes_at_front =
 		(((size_t)num_header_extents_at_front
 		  * (size_t)header_extent_size));
-	(*bytes_read) = (sizeof(u32) + sizeof(u16));
+	(*bytes_read) = (sizeof(__be32) + sizeof(__be16));
 	if ((validate_header_size == ECRYPTFS_VALIDATE_HEADER_SIZE)
 	    && (crypt_stat->num_header_bytes_at_front
 		< ECRYPTFS_MINIMUM_HEADER_EXTENT_SIZE)) {

From 0a688ad713949643e201431d3f4a4ceddfeb70ca Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Wed, 23 Jul 2008 21:30:07 -0700
Subject: [PATCH 0402/1171] ecryptfs: inode.c mmap.c use unaligned byteorder
 helpers

Fixe sparse warnings:
fs/ecryptfs/inode.c:368:15: warning: cast to restricted __be64
fs/ecryptfs/mmap.c:385:12: warning: incorrect type in assignment (different base types)
fs/ecryptfs/mmap.c:385:12:    expected unsigned long long [unsigned] [assigned] [usertype] file_size
fs/ecryptfs/mmap.c:385:12:    got restricted __be64 [usertype] <noident>
fs/ecryptfs/mmap.c:428:12: warning: incorrect type in assignment (different base types)
fs/ecryptfs/mmap.c:428:12:    expected unsigned long long [unsigned] [assigned] [usertype] file_size
fs/ecryptfs/mmap.c:428:12:    got restricted __be64 [usertype] <noident>

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Cc: Michael Halcrow <mhalcrow@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ecryptfs/inode.c |  4 ++--
 fs/ecryptfs/mmap.c  | 11 +++--------
 2 files changed, 5 insertions(+), 10 deletions(-)

diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c
index c92cc1c00aae..7315547193ea 100644
--- a/fs/ecryptfs/inode.c
+++ b/fs/ecryptfs/inode.c
@@ -31,6 +31,7 @@
 #include <linux/mount.h>
 #include <linux/crypto.h>
 #include <linux/fs_stack.h>
+#include <asm/unaligned.h>
 #include "ecryptfs_kernel.h"
 
 static struct dentry *lock_parent(struct dentry *dentry)
@@ -364,8 +365,7 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry,
 		else
 			file_size = i_size_read(lower_dentry->d_inode);
 	} else {
-		memcpy(&file_size, page_virt, sizeof(file_size));
-		file_size = be64_to_cpu(file_size);
+		file_size = get_unaligned_be64(page_virt);
 	}
 	i_size_write(dentry->d_inode, (loff_t)file_size);
 	kmem_cache_free(ecryptfs_header_cache_2, page_virt);
diff --git a/fs/ecryptfs/mmap.c b/fs/ecryptfs/mmap.c
index 2b6fe1e6e8ba..245c2dc02d5c 100644
--- a/fs/ecryptfs/mmap.c
+++ b/fs/ecryptfs/mmap.c
@@ -32,6 +32,7 @@
 #include <linux/file.h>
 #include <linux/crypto.h>
 #include <linux/scatterlist.h>
+#include <asm/unaligned.h>
 #include "ecryptfs_kernel.h"
 
 /**
@@ -372,7 +373,6 @@ out:
  */
 static int ecryptfs_write_inode_size_to_header(struct inode *ecryptfs_inode)
 {
-	u64 file_size;
 	char *file_size_virt;
 	int rc;
 
@@ -381,9 +381,7 @@ static int ecryptfs_write_inode_size_to_header(struct inode *ecryptfs_inode)
 		rc = -ENOMEM;
 		goto out;
 	}
-	file_size = (u64)i_size_read(ecryptfs_inode);
-	file_size = cpu_to_be64(file_size);
-	memcpy(file_size_virt, &file_size, sizeof(u64));
+	put_unaligned_be64(i_size_read(ecryptfs_inode), file_size_virt);
 	rc = ecryptfs_write_lower(ecryptfs_inode, file_size_virt, 0,
 				  sizeof(u64));
 	kfree(file_size_virt);
@@ -403,7 +401,6 @@ static int ecryptfs_write_inode_size_to_xattr(struct inode *ecryptfs_inode)
 	struct dentry *lower_dentry =
 		ecryptfs_inode_to_private(ecryptfs_inode)->lower_file->f_dentry;
 	struct inode *lower_inode = lower_dentry->d_inode;
-	u64 file_size;
 	int rc;
 
 	if (!lower_inode->i_op->getxattr || !lower_inode->i_op->setxattr) {
@@ -424,9 +421,7 @@ static int ecryptfs_write_inode_size_to_xattr(struct inode *ecryptfs_inode)
 					   xattr_virt, PAGE_CACHE_SIZE);
 	if (size < 0)
 		size = 8;
-	file_size = (u64)i_size_read(ecryptfs_inode);
-	file_size = cpu_to_be64(file_size);
-	memcpy(xattr_virt, &file_size, sizeof(u64));
+	put_unaligned_be64(i_size_read(ecryptfs_inode), xattr_virt);
 	rc = lower_inode->i_op->setxattr(lower_dentry, ECRYPTFS_XATTR_NAME,
 					 xattr_virt, size, 0);
 	mutex_unlock(&lower_inode->i_mutex);

From 72b55fffd631a89e5be6fe1b4f2565bc4cd90deb Mon Sep 17 00:00:00 2001
From: Michael Halcrow <mhalcrow@us.ibm.com>
Date: Wed, 23 Jul 2008 21:30:07 -0700
Subject: [PATCH 0403/1171] eCryptfs: do not try to open device files on mknod

When creating device nodes, eCryptfs needs to delay actually opening the lower
persistent file until an application tries to open.  Device handles may not be
backed by anything when they first come into existence.

[Valdis.Kletnieks@vt.edu: build fix]
Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com>
Cc: <Valdis.Kletnieks@vt.edu}
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ecryptfs/ecryptfs_kernel.h |  6 +++++-
 fs/ecryptfs/file.c            | 14 ++++++++++++++
 fs/ecryptfs/inode.c           |  6 ++++--
 fs/ecryptfs/main.c            | 29 +++++++++++++++++++----------
 4 files changed, 42 insertions(+), 13 deletions(-)

diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h
index b4a0cccfdd7c..b0727f91454e 100644
--- a/fs/ecryptfs/ecryptfs_kernel.h
+++ b/fs/ecryptfs/ecryptfs_kernel.h
@@ -235,6 +235,7 @@ struct ecryptfs_crypt_stat {
 #define ECRYPTFS_METADATA_IN_XATTR  0x00000100
 #define ECRYPTFS_VIEW_AS_ENCRYPTED  0x00000200
 #define ECRYPTFS_KEY_SET            0x00000400
+#define ECRYPTFS_DELAY_PERSISTENT   0x00000800
 	u32 flags;
 	unsigned int file_version;
 	size_t iv_bytes;
@@ -574,9 +575,11 @@ struct ecryptfs_open_req {
 	struct list_head kthread_ctl_list;
 };
 
+#define ECRYPTFS_INTERPOSE_FLAG_D_ADD                 0x00000001
+#define ECRYPTFS_INTERPOSE_FLAG_DELAY_PERSISTENT_FILE 0x00000002
 int ecryptfs_interpose(struct dentry *hidden_dentry,
 		       struct dentry *this_dentry, struct super_block *sb,
-		       int flag);
+		       u32 flags);
 int ecryptfs_fill_zeros(struct file *file, loff_t new_length);
 int ecryptfs_decode_filename(struct ecryptfs_crypt_stat *crypt_stat,
 			     const char *name, int length,
@@ -709,5 +712,6 @@ void ecryptfs_destroy_kthread(void);
 int ecryptfs_privileged_open(struct file **lower_file,
 			     struct dentry *lower_dentry,
 			     struct vfsmount *lower_mnt);
+int ecryptfs_init_persistent_file(struct dentry *ecryptfs_dentry);
 
 #endif /* #ifndef ECRYPTFS_KERNEL_H */
diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c
index f0be29051528..2c2d60df3f60 100644
--- a/fs/ecryptfs/file.c
+++ b/fs/ecryptfs/file.c
@@ -199,6 +199,20 @@ static int ecryptfs_open(struct inode *inode, struct file *file)
 		       "file must hence be opened RO\n", __func__);
 		goto out;
 	}
+	if (!ecryptfs_inode_to_private(inode)->lower_file) {
+		BUG_ON(!(crypt_stat->flags & ECRYPTFS_DELAY_PERSISTENT));
+		mutex_lock(&crypt_stat->cs_mutex);
+		crypt_stat->flags &= ~(ECRYPTFS_ENCRYPTED);
+		mutex_unlock(&crypt_stat->cs_mutex);
+		rc = ecryptfs_init_persistent_file(ecryptfs_dentry);
+		if (rc) {
+			printk(KERN_ERR "%s: Error attempting to initialize "
+			       "the persistent file for the dentry with name "
+			       "[%s]; rc = [%d]\n", __func__,
+			       ecryptfs_dentry->d_name.name, rc);
+			goto out;
+		}
+	}
 	ecryptfs_set_file_lower(
 		file, ecryptfs_inode_to_private(inode)->lower_file);
 	if (S_ISDIR(ecryptfs_dentry->d_inode->i_mode)) {
diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c
index 7315547193ea..26090878c930 100644
--- a/fs/ecryptfs/inode.c
+++ b/fs/ecryptfs/inode.c
@@ -308,7 +308,8 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry,
 		d_add(dentry, NULL);
 		goto out;
 	}
-	rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb, 1);
+	rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb,
+				ECRYPTFS_INTERPOSE_FLAG_D_ADD);
 	if (rc) {
 		ecryptfs_printk(KERN_ERR, "Error interposing\n");
 		goto out_dput;
@@ -537,7 +538,8 @@ ecryptfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev)
 	rc = vfs_mknod(lower_dir_dentry->d_inode, lower_dentry, mode, dev);
 	if (rc || !lower_dentry->d_inode)
 		goto out;
-	rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb, 0);
+	rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb,
+				ECRYPTFS_INTERPOSE_FLAG_DELAY_PERSISTENT_FILE);
 	if (rc)
 		goto out;
 	fsstack_copy_attr_times(dir, lower_dir_dentry->d_inode);
diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c
index 10475d93ff53..ee4f84b20410 100644
--- a/fs/ecryptfs/main.c
+++ b/fs/ecryptfs/main.c
@@ -117,7 +117,7 @@ void __ecryptfs_printk(const char *fmt, ...)
  *
  * Returns zero on success; non-zero otherwise
  */
-static int ecryptfs_init_persistent_file(struct dentry *ecryptfs_dentry)
+int ecryptfs_init_persistent_file(struct dentry *ecryptfs_dentry)
 {
 	struct ecryptfs_inode_info *inode_info =
 		ecryptfs_inode_to_private(ecryptfs_dentry->d_inode);
@@ -149,14 +149,14 @@ static int ecryptfs_init_persistent_file(struct dentry *ecryptfs_dentry)
  * @lower_dentry: Existing dentry in the lower filesystem
  * @dentry: ecryptfs' dentry
  * @sb: ecryptfs's super_block
- * @flag: If set to true, then d_add is called, else d_instantiate is called
+ * @flags: flags to govern behavior of interpose procedure
  *
  * Interposes upper and lower dentries.
  *
  * Returns zero on success; non-zero otherwise
  */
 int ecryptfs_interpose(struct dentry *lower_dentry, struct dentry *dentry,
-		       struct super_block *sb, int flag)
+		       struct super_block *sb, u32 flags)
 {
 	struct inode *lower_inode;
 	struct inode *inode;
@@ -193,7 +193,7 @@ int ecryptfs_interpose(struct dentry *lower_dentry, struct dentry *dentry,
 		init_special_inode(inode, lower_inode->i_mode,
 				   lower_inode->i_rdev);
 	dentry->d_op = &ecryptfs_dops;
-	if (flag)
+	if (flags & ECRYPTFS_INTERPOSE_FLAG_D_ADD)
 		d_add(dentry, inode);
 	else
 		d_instantiate(dentry, inode);
@@ -201,12 +201,21 @@ int ecryptfs_interpose(struct dentry *lower_dentry, struct dentry *dentry,
 	/* This size will be overwritten for real files w/ headers and
 	 * other metadata */
 	fsstack_copy_inode_size(inode, lower_inode);
-	rc = ecryptfs_init_persistent_file(dentry);
-	if (rc) {
-		printk(KERN_ERR "%s: Error attempting to initialize the "
-		       "persistent file for the dentry with name [%s]; "
-		       "rc = [%d]\n", __func__, dentry->d_name.name, rc);
-		goto out;
+	if (!(flags & ECRYPTFS_INTERPOSE_FLAG_DELAY_PERSISTENT_FILE)) {
+		rc = ecryptfs_init_persistent_file(dentry);
+		if (rc) {
+			printk(KERN_ERR "%s: Error attempting to initialize "
+			       "the persistent file for the dentry with name "
+			       "[%s]; rc = [%d]\n", __func__,
+			       dentry->d_name.name, rc);
+			goto out;
+		}
+	} else {
+		struct ecryptfs_inode_info *inode_info =
+			ecryptfs_inode_to_private(dentry->d_inode);
+
+		inode_info->lower_file = NULL;
+		inode_info->crypt_stat.flags |= ECRYPTFS_DELAY_PERSISTENT;
 	}
 out:
 	return rc;

From 391b52f98cf2e9bff227dad8bf9ea206fec43fa4 Mon Sep 17 00:00:00 2001
From: Michael Halcrow <mhalcrow@us.ibm.com>
Date: Wed, 23 Jul 2008 21:30:08 -0700
Subject: [PATCH 0404/1171] eCryptfs: Make all persistent file opens delayed

There is no good reason to immediately open the lower file, and that can
cause problems with files that the user does not intend to immediately
open, such as device nodes.

This patch removes the persistent file open from the interpose step and
pushes that to the locations where eCryptfs really does need the lower
persistent file, such as just before reading or writing the metadata
stored in the lower file header.

Two functions are jumping to out_dput when they should just be jumping to
out on error paths.  This patch also fixes these.

Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ecryptfs/ecryptfs_kernel.h |  2 --
 fs/ecryptfs/file.c            |  4 ----
 fs/ecryptfs/inode.c           | 27 +++++++++++++++++++++++----
 fs/ecryptfs/main.c            | 16 ----------------
 4 files changed, 23 insertions(+), 26 deletions(-)

diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h
index b0727f91454e..b73fb752c5f8 100644
--- a/fs/ecryptfs/ecryptfs_kernel.h
+++ b/fs/ecryptfs/ecryptfs_kernel.h
@@ -235,7 +235,6 @@ struct ecryptfs_crypt_stat {
 #define ECRYPTFS_METADATA_IN_XATTR  0x00000100
 #define ECRYPTFS_VIEW_AS_ENCRYPTED  0x00000200
 #define ECRYPTFS_KEY_SET            0x00000400
-#define ECRYPTFS_DELAY_PERSISTENT   0x00000800
 	u32 flags;
 	unsigned int file_version;
 	size_t iv_bytes;
@@ -576,7 +575,6 @@ struct ecryptfs_open_req {
 };
 
 #define ECRYPTFS_INTERPOSE_FLAG_D_ADD                 0x00000001
-#define ECRYPTFS_INTERPOSE_FLAG_DELAY_PERSISTENT_FILE 0x00000002
 int ecryptfs_interpose(struct dentry *hidden_dentry,
 		       struct dentry *this_dentry, struct super_block *sb,
 		       u32 flags);
diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c
index 2c2d60df3f60..9244d653743e 100644
--- a/fs/ecryptfs/file.c
+++ b/fs/ecryptfs/file.c
@@ -200,10 +200,6 @@ static int ecryptfs_open(struct inode *inode, struct file *file)
 		goto out;
 	}
 	if (!ecryptfs_inode_to_private(inode)->lower_file) {
-		BUG_ON(!(crypt_stat->flags & ECRYPTFS_DELAY_PERSISTENT));
-		mutex_lock(&crypt_stat->cs_mutex);
-		crypt_stat->flags &= ~(ECRYPTFS_ENCRYPTED);
-		mutex_unlock(&crypt_stat->cs_mutex);
 		rc = ecryptfs_init_persistent_file(ecryptfs_dentry);
 		if (rc) {
 			printk(KERN_ERR "%s: Error attempting to initialize "
diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c
index 26090878c930..d755455e3bff 100644
--- a/fs/ecryptfs/inode.c
+++ b/fs/ecryptfs/inode.c
@@ -189,6 +189,16 @@ static int ecryptfs_initialize_file(struct dentry *ecryptfs_dentry)
 				"context; rc = [%d]\n", rc);
 		goto out;
 	}
+	if (!ecryptfs_inode_to_private(ecryptfs_dentry->d_inode)->lower_file) {
+		rc = ecryptfs_init_persistent_file(ecryptfs_dentry);
+		if (rc) {
+			printk(KERN_ERR "%s: Error attempting to initialize "
+			       "the persistent file for the dentry with name "
+			       "[%s]; rc = [%d]\n", __func__,
+			       ecryptfs_dentry->d_name.name, rc);
+			goto out;
+		}
+	}
 	rc = ecryptfs_write_metadata(ecryptfs_dentry);
 	if (rc) {
 		printk(KERN_ERR "Error writing headers; rc = [%d]\n", rc);
@@ -312,7 +322,7 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry,
 				ECRYPTFS_INTERPOSE_FLAG_D_ADD);
 	if (rc) {
 		ecryptfs_printk(KERN_ERR, "Error interposing\n");
-		goto out_dput;
+		goto out;
 	}
 	if (S_ISDIR(lower_inode->i_mode)) {
 		ecryptfs_printk(KERN_DEBUG, "Is a directory; returning\n");
@@ -338,11 +348,21 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry,
 		rc = -ENOMEM;
 		ecryptfs_printk(KERN_ERR,
 				"Cannot ecryptfs_kmalloc a page\n");
-		goto out_dput;
+		goto out;
 	}
 	crypt_stat = &ecryptfs_inode_to_private(dentry->d_inode)->crypt_stat;
 	if (!(crypt_stat->flags & ECRYPTFS_POLICY_APPLIED))
 		ecryptfs_set_default_sizes(crypt_stat);
+	if (!ecryptfs_inode_to_private(dentry->d_inode)->lower_file) {
+		rc = ecryptfs_init_persistent_file(dentry);
+		if (rc) {
+			printk(KERN_ERR "%s: Error attempting to initialize "
+			       "the persistent file for the dentry with name "
+			       "[%s]; rc = [%d]\n", __func__,
+			       dentry->d_name.name, rc);
+			goto out;
+		}
+	}
 	rc = ecryptfs_read_and_validate_header_region(page_virt,
 						      dentry->d_inode);
 	if (rc) {
@@ -538,8 +558,7 @@ ecryptfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev)
 	rc = vfs_mknod(lower_dir_dentry->d_inode, lower_dentry, mode, dev);
 	if (rc || !lower_dentry->d_inode)
 		goto out;
-	rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb,
-				ECRYPTFS_INTERPOSE_FLAG_DELAY_PERSISTENT_FILE);
+	rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb, 0);
 	if (rc)
 		goto out;
 	fsstack_copy_attr_times(dir, lower_dir_dentry->d_inode);
diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c
index ee4f84b20410..6f403cfba14f 100644
--- a/fs/ecryptfs/main.c
+++ b/fs/ecryptfs/main.c
@@ -201,22 +201,6 @@ int ecryptfs_interpose(struct dentry *lower_dentry, struct dentry *dentry,
 	/* This size will be overwritten for real files w/ headers and
 	 * other metadata */
 	fsstack_copy_inode_size(inode, lower_inode);
-	if (!(flags & ECRYPTFS_INTERPOSE_FLAG_DELAY_PERSISTENT_FILE)) {
-		rc = ecryptfs_init_persistent_file(dentry);
-		if (rc) {
-			printk(KERN_ERR "%s: Error attempting to initialize "
-			       "the persistent file for the dentry with name "
-			       "[%s]; rc = [%d]\n", __func__,
-			       dentry->d_name.name, rc);
-			goto out;
-		}
-	} else {
-		struct ecryptfs_inode_info *inode_info =
-			ecryptfs_inode_to_private(dentry->d_inode);
-
-		inode_info->lower_file = NULL;
-		inode_info->crypt_stat.flags |= ECRYPTFS_DELAY_PERSISTENT;
-	}
 out:
 	return rc;
 }

From 5f6f4f28b6ba543beef8bad91aa6f69c7ffeee51 Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Wed, 23 Jul 2008 21:30:09 -0700
Subject: [PATCH 0405/1171] autofs4: don't make expiring dentry negative

Correct the error of making a positive dentry negative after it has been
instantiated.

The code that makes this error attempts to re-use the dentry from a
concurrent expire and mount to resolve a race and the dentry used for the
lookup must be negative for mounts to trigger in the required cases.  The
fact is that the dentry doesn't need to be re-used because all that is
needed is to preserve the flag that indicates an expire is still
incomplete at the time of the mount request.

This change uses the the dentry to check the flag and wait for the expire
to complete then discards it instead of attempting to re-use it.

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/autofs_i.h |   6 +--
 fs/autofs4/inode.c    |   6 +--
 fs/autofs4/root.c     | 118 ++++++++++++++++--------------------------
 3 files changed, 52 insertions(+), 78 deletions(-)

diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h
index c3d352d7fa93..69b1497b0029 100644
--- a/fs/autofs4/autofs_i.h
+++ b/fs/autofs4/autofs_i.h
@@ -52,7 +52,7 @@ struct autofs_info {
 
 	int		flags;
 
-	struct list_head rehash;
+	struct list_head expiring;
 
 	struct autofs_sb_info *sbi;
 	unsigned long last_used;
@@ -112,8 +112,8 @@ struct autofs_sb_info {
 	struct mutex wq_mutex;
 	spinlock_t fs_lock;
 	struct autofs_wait_queue *queues; /* Wait queue pointer */
-	spinlock_t rehash_lock;
-	struct list_head rehash_list;
+	spinlock_t lookup_lock;
+	struct list_head expiring_list;
 };
 
 static inline struct autofs_sb_info *autofs4_sbi(struct super_block *sb)
diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c
index 2fdcf5e1d236..94bfc154d7a2 100644
--- a/fs/autofs4/inode.c
+++ b/fs/autofs4/inode.c
@@ -47,7 +47,7 @@ struct autofs_info *autofs4_init_ino(struct autofs_info *ino,
 	ino->dentry = NULL;
 	ino->size = 0;
 
-	INIT_LIST_HEAD(&ino->rehash);
+	INIT_LIST_HEAD(&ino->expiring);
 
 	ino->last_used = jiffies;
 	atomic_set(&ino->count, 0);
@@ -338,8 +338,8 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent)
 	mutex_init(&sbi->wq_mutex);
 	spin_lock_init(&sbi->fs_lock);
 	sbi->queues = NULL;
-	spin_lock_init(&sbi->rehash_lock);
-	INIT_LIST_HEAD(&sbi->rehash_list);
+	spin_lock_init(&sbi->lookup_lock);
+	INIT_LIST_HEAD(&sbi->expiring_list);
 	s->s_blocksize = 1024;
 	s->s_blocksize_bits = 10;
 	s->s_magic = AUTOFS_SUPER_MAGIC;
diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c
index edf5b6bddb52..9ead2279df4f 100644
--- a/fs/autofs4/root.c
+++ b/fs/autofs4/root.c
@@ -493,10 +493,10 @@ void autofs4_dentry_release(struct dentry *de)
 		struct autofs_sb_info *sbi = autofs4_sbi(de->d_sb);
 
 		if (sbi) {
-			spin_lock(&sbi->rehash_lock);
-			if (!list_empty(&inf->rehash))
-				list_del(&inf->rehash);
-			spin_unlock(&sbi->rehash_lock);
+			spin_lock(&sbi->lookup_lock);
+			if (!list_empty(&inf->expiring))
+				list_del(&inf->expiring);
+			spin_unlock(&sbi->lookup_lock);
 		}
 
 		inf->dentry = NULL;
@@ -518,7 +518,7 @@ static struct dentry_operations autofs4_dentry_operations = {
 	.d_release	= autofs4_dentry_release,
 };
 
-static struct dentry *autofs4_lookup_unhashed(struct autofs_sb_info *sbi, struct dentry *parent, struct qstr *name)
+static struct dentry *autofs4_lookup_expiring(struct autofs_sb_info *sbi, struct dentry *parent, struct qstr *name)
 {
 	unsigned int len = name->len;
 	unsigned int hash = name->hash;
@@ -526,14 +526,14 @@ static struct dentry *autofs4_lookup_unhashed(struct autofs_sb_info *sbi, struct
 	struct list_head *p, *head;
 
 	spin_lock(&dcache_lock);
-	spin_lock(&sbi->rehash_lock);
-	head = &sbi->rehash_list;
+	spin_lock(&sbi->lookup_lock);
+	head = &sbi->expiring_list;
 	list_for_each(p, head) {
 		struct autofs_info *ino;
 		struct dentry *dentry;
 		struct qstr *qstr;
 
-		ino = list_entry(p, struct autofs_info, rehash);
+		ino = list_entry(p, struct autofs_info, expiring);
 		dentry = ino->dentry;
 
 		spin_lock(&dentry->d_lock);
@@ -555,33 +555,16 @@ static struct dentry *autofs4_lookup_unhashed(struct autofs_sb_info *sbi, struct
 			goto next;
 
 		if (d_unhashed(dentry)) {
-			struct inode *inode = dentry->d_inode;
-
-			ino = autofs4_dentry_ino(dentry);
-			list_del_init(&ino->rehash);
 			dget(dentry);
-			/*
-			 * Make the rehashed dentry negative so the VFS
-			 * behaves as it should.
-			 */
-			if (inode) {
-				dentry->d_inode = NULL;
-				list_del_init(&dentry->d_alias);
-				spin_unlock(&dentry->d_lock);
-				spin_unlock(&sbi->rehash_lock);
-				spin_unlock(&dcache_lock);
-				iput(inode);
-				return dentry;
-			}
 			spin_unlock(&dentry->d_lock);
-			spin_unlock(&sbi->rehash_lock);
+			spin_unlock(&sbi->lookup_lock);
 			spin_unlock(&dcache_lock);
 			return dentry;
 		}
 next:
 		spin_unlock(&dentry->d_lock);
 	}
-	spin_unlock(&sbi->rehash_lock);
+	spin_unlock(&sbi->lookup_lock);
 	spin_unlock(&dcache_lock);
 
 	return NULL;
@@ -591,7 +574,7 @@ next:
 static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
 {
 	struct autofs_sb_info *sbi;
-	struct dentry *unhashed;
+	struct dentry *expiring;
 	int oz_mode;
 
 	DPRINTK("name = %.*s",
@@ -607,44 +590,44 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s
 	DPRINTK("pid = %u, pgrp = %u, catatonic = %d, oz_mode = %d",
 		 current->pid, task_pgrp_nr(current), sbi->catatonic, oz_mode);
 
-	unhashed = autofs4_lookup_unhashed(sbi, dentry->d_parent, &dentry->d_name);
-	if (!unhashed) {
-		/*
-		 * Mark the dentry incomplete but don't hash it. We do this
-		 * to serialize our inode creation operations (symlink and
-		 * mkdir) which prevents deadlock during the callback to
-		 * the daemon. Subsequent user space lookups for the same
-		 * dentry are placed on the wait queue while the daemon
-		 * itself is allowed passage unresticted so the create
-		 * operation itself can then hash the dentry. Finally,
-		 * we check for the hashed dentry and return the newly
-		 * hashed dentry.
-		 */
-		dentry->d_op = &autofs4_root_dentry_operations;
-
-		dentry->d_fsdata = NULL;
-		d_instantiate(dentry, NULL);
-	} else {
-		struct autofs_info *ino = autofs4_dentry_ino(unhashed);
-		DPRINTK("rehash %p with %p", dentry, unhashed);
+	expiring = autofs4_lookup_expiring(sbi, dentry->d_parent, &dentry->d_name);
+	if (expiring) {
+		struct autofs_info *ino = autofs4_dentry_ino(expiring);
 		/*
 		 * If we are racing with expire the request might not
 		 * be quite complete but the directory has been removed
 		 * so it must have been successful, so just wait for it.
-		 * We need to ensure the AUTOFS_INF_EXPIRING flag is clear
-		 * before continuing as revalidate may fail when calling
-		 * try_to_fill_dentry (returning EAGAIN) if we don't.
 		 */
 		while (ino && (ino->flags & AUTOFS_INF_EXPIRING)) {
 			DPRINTK("wait for incomplete expire %p name=%.*s",
-				unhashed, unhashed->d_name.len,
-				unhashed->d_name.name);
-			autofs4_wait(sbi, unhashed, NFY_NONE);
+				expiring, expiring->d_name.len,
+				expiring->d_name.name);
+			autofs4_wait(sbi, expiring, NFY_NONE);
 			DPRINTK("request completed");
 		}
-		dentry = unhashed;
+		spin_lock(&sbi->lookup_lock);
+		if (!list_empty(&ino->expiring))
+			list_del_init(&ino->expiring);
+		spin_unlock(&sbi->lookup_lock);
+		dput(expiring);
 	}
 
+	/*
+	 * Mark the dentry incomplete but don't hash it. We do this
+	 * to serialize our inode creation operations (symlink and
+	 * mkdir) which prevents deadlock during the callback to
+	 * the daemon. Subsequent user space lookups for the same
+	 * dentry are placed on the wait queue while the daemon
+	 * itself is allowed passage unresticted so the create
+	 * operation itself can then hash the dentry. Finally,
+	 * we check for the hashed dentry and return the newly
+	 * hashed dentry.
+	 */
+	dentry->d_op = &autofs4_root_dentry_operations;
+
+	dentry->d_fsdata = NULL;
+	d_instantiate(dentry, NULL);
+
 	if (!oz_mode) {
 		spin_lock(&dentry->d_lock);
 		dentry->d_flags |= DCACHE_AUTOFS_PENDING;
@@ -668,8 +651,6 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s
 			if (sigismember (sigset, SIGKILL) ||
 			    sigismember (sigset, SIGQUIT) ||
 			    sigismember (sigset, SIGINT)) {
-			    if (unhashed)
-				dput(unhashed);
 			    return ERR_PTR(-ERESTARTNOINTR);
 			}
 		}
@@ -699,15 +680,9 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s
 		else
 			dentry = ERR_PTR(-ENOENT);
 
-		if (unhashed)
-			dput(unhashed);
-
 		return dentry;
 	}
 
-	if (unhashed)
-		return dentry;
-
 	return NULL;
 }
 
@@ -769,9 +744,8 @@ static int autofs4_dir_symlink(struct inode *dir,
  * that the file no longer exists. However, doing that means that the
  * VFS layer can turn the dentry into a negative dentry.  We don't want
  * this, because the unlink is probably the result of an expire.
- * We simply d_drop it and add it to a rehash candidates list in the
- * super block, which allows the dentry lookup to reuse it retaining
- * the flags, such as expire in progress, in case we're racing with expire.
+ * We simply d_drop it and add it to a expiring list in the super block,
+ * which allows the dentry lookup to check for an incomplete expire.
  *
  * If a process is blocked on the dentry waiting for the expire to finish,
  * it will invalidate the dentry and try to mount with a new one.
@@ -801,9 +775,9 @@ static int autofs4_dir_unlink(struct inode *dir, struct dentry *dentry)
 	dir->i_mtime = CURRENT_TIME;
 
 	spin_lock(&dcache_lock);
-	spin_lock(&sbi->rehash_lock);
-	list_add(&ino->rehash, &sbi->rehash_list);
-	spin_unlock(&sbi->rehash_lock);
+	spin_lock(&sbi->lookup_lock);
+	list_add(&ino->expiring, &sbi->expiring_list);
+	spin_unlock(&sbi->lookup_lock);
 	spin_lock(&dentry->d_lock);
 	__d_drop(dentry);
 	spin_unlock(&dentry->d_lock);
@@ -829,9 +803,9 @@ static int autofs4_dir_rmdir(struct inode *dir, struct dentry *dentry)
 		spin_unlock(&dcache_lock);
 		return -ENOTEMPTY;
 	}
-	spin_lock(&sbi->rehash_lock);
-	list_add(&ino->rehash, &sbi->rehash_list);
-	spin_unlock(&sbi->rehash_lock);
+	spin_lock(&sbi->lookup_lock);
+	list_add(&ino->expiring, &sbi->expiring_list);
+	spin_unlock(&sbi->lookup_lock);
 	spin_lock(&dentry->d_lock);
 	__d_drop(dentry);
 	spin_unlock(&dentry->d_lock);

From caf7da3d5d4d9dd873eb52d025d8cc63b89f1fdb Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Wed, 23 Jul 2008 21:30:11 -0700
Subject: [PATCH 0406/1171] autofs4: revert - redo lookup in ttfd

This patch series enables the use of a single dentry for lookups prior to
the dentry being hashed and so we no longer need to redo the lookup.  This
patch reverts the patch of commit
033790449ba9c4dcf8478a87693d33df625c23b5.

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/root.c | 21 ---------------------
 1 file changed, 21 deletions(-)

diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c
index 9ead2279df4f..53dabe8d5b8b 100644
--- a/fs/autofs4/root.c
+++ b/fs/autofs4/root.c
@@ -242,7 +242,6 @@ static int try_to_fill_dentry(struct dentry *dentry, int flags)
 {
 	struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
 	struct autofs_info *ino = autofs4_dentry_ino(dentry);
-	struct dentry *new;
 	int status;
 
 	/* Block on any pending expiry here; invalidate the dentry
@@ -320,26 +319,6 @@ static int try_to_fill_dentry(struct dentry *dentry, int flags)
 	dentry->d_flags &= ~DCACHE_AUTOFS_PENDING;
 	spin_unlock(&dentry->d_lock);
 
-	/*
-	 * The dentry that is passed in from lookup may not be the one
-	 * we end up using, as mkdir can create a new one.  If this
-	 * happens, and another process tries the lookup at the same time,
-	 * it will set the PENDING flag on this new dentry, but add itself
-	 * to our waitq.  Then, if after the lookup succeeds, the first
-	 * process that requested the mount performs another lookup of the
-	 * same directory, it will show up as still pending!  So, we need
-	 * to redo the lookup here and clear pending on that dentry.
-	 */
-	if (d_unhashed(dentry)) {
-		new = d_lookup(dentry->d_parent, &dentry->d_name);
-		if (new) {
-			spin_lock(&new->d_lock);
-			new->d_flags &= ~DCACHE_AUTOFS_PENDING;
-			spin_unlock(&new->d_lock);
-			dput(new);
-		}
-	}
-
 	return 0;
 }
 

From 2576737873dc1d9ea461a5955a5f6779b569a350 Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Wed, 23 Jul 2008 21:30:12 -0700
Subject: [PATCH 0407/1171] autofs4: use look aside list for lookups

A while ago a patch to resolve a deadlock during directory creation was
merged.  This delayed the hashing of lookup dentrys until the ->mkdir()
(or ->symlink()) operation completed to ensure we always went through
->lookup() instead of also having processes go through ->revalidate() so
our VFS locking remained consistent.

Now we are seeing a couple of side affects of that change in situations
with heavy mount activity.

Two cases have been identified:

1) When a mount request is triggered, due to the delayed hashing, the
   directory created by user space for the mount point doesn't have the
   DCACHE_AUTOFS_PENDING flag set.  In the case of an autofs multi-mount
   where a tree of mount point directories are created this can lead to
   the path walk continuing rather than the dentry being sent to the wait
   queue to wait for request completion.  This is because, if the pending
   flag isn't set, the criteria for deciding this is a mount in progress
   fails to hold, namely that the dentry is not a mount point and has no
   subdirectories.

2) A mount request dentry is initially created negative and unhashed.
   It remains this way until the ->mkdir() callback completes.  Since it
   is unhashed a fresh dentry is used when the user space mount request
   creates the mount point directory.  This leaves the original dentry
   negative and unhashed.  But revalidate has no way to tell the VFS that
   the dentry has changed, other than to force another ->lookup() by
   returning false, which is at best wastefull and at worst not possible.
   This results in an -ENOENT return from the original path walk when in
   fact the mount succeeded.

To resolve this we need to ensure that the same dentry is used in all
calls to ->lookup() during the course of a mount request.  This patch
achieves that by adding the initial dentry to a look aside list and
removes it at ->mkdir() or ->symlink() completion (or when the dentry is
released), since these are the only create operations autofs4 supports.

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/autofs_i.h |   2 +
 fs/autofs4/inode.c    |  25 ++++---
 fs/autofs4/root.c     | 169 ++++++++++++++++++++++++++++++++++--------
 3 files changed, 156 insertions(+), 40 deletions(-)

diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h
index 69b1497b0029..2dce2334737d 100644
--- a/fs/autofs4/autofs_i.h
+++ b/fs/autofs4/autofs_i.h
@@ -52,6 +52,7 @@ struct autofs_info {
 
 	int		flags;
 
+	struct list_head active;
 	struct list_head expiring;
 
 	struct autofs_sb_info *sbi;
@@ -113,6 +114,7 @@ struct autofs_sb_info {
 	spinlock_t fs_lock;
 	struct autofs_wait_queue *queues; /* Wait queue pointer */
 	spinlock_t lookup_lock;
+	struct list_head active_list;
 	struct list_head expiring_list;
 };
 
diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c
index 94bfc154d7a2..e3e70994ab46 100644
--- a/fs/autofs4/inode.c
+++ b/fs/autofs4/inode.c
@@ -24,8 +24,10 @@
 
 static void ino_lnkfree(struct autofs_info *ino)
 {
-	kfree(ino->u.symlink);
-	ino->u.symlink = NULL;
+	if (ino->u.symlink) {
+		kfree(ino->u.symlink);
+		ino->u.symlink = NULL;
+	}
 }
 
 struct autofs_info *autofs4_init_ino(struct autofs_info *ino,
@@ -41,16 +43,18 @@ struct autofs_info *autofs4_init_ino(struct autofs_info *ino,
 	if (ino == NULL)
 		return NULL;
 
-	ino->flags = 0;
+	if (!reinit) {
+		ino->flags = 0;
+		ino->inode = NULL;
+		ino->dentry = NULL;
+		ino->size = 0;
+		INIT_LIST_HEAD(&ino->active);
+		INIT_LIST_HEAD(&ino->expiring);
+		atomic_set(&ino->count, 0);
+	}
+
 	ino->mode = mode;
-	ino->inode = NULL;
-	ino->dentry = NULL;
-	ino->size = 0;
-
-	INIT_LIST_HEAD(&ino->expiring);
-
 	ino->last_used = jiffies;
-	atomic_set(&ino->count, 0);
 
 	ino->sbi = sbi;
 
@@ -339,6 +343,7 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent)
 	spin_lock_init(&sbi->fs_lock);
 	sbi->queues = NULL;
 	spin_lock_init(&sbi->lookup_lock);
+	INIT_LIST_HEAD(&sbi->active_list);
 	INIT_LIST_HEAD(&sbi->expiring_list);
 	s->s_blocksize = 1024;
 	s->s_blocksize_bits = 10;
diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c
index 53dabe8d5b8b..dbb70d5a4882 100644
--- a/fs/autofs4/root.c
+++ b/fs/autofs4/root.c
@@ -473,6 +473,8 @@ void autofs4_dentry_release(struct dentry *de)
 
 		if (sbi) {
 			spin_lock(&sbi->lookup_lock);
+			if (!list_empty(&inf->active))
+				list_del(&inf->active);
 			if (!list_empty(&inf->expiring))
 				list_del(&inf->expiring);
 			spin_unlock(&sbi->lookup_lock);
@@ -497,6 +499,58 @@ static struct dentry_operations autofs4_dentry_operations = {
 	.d_release	= autofs4_dentry_release,
 };
 
+static struct dentry *autofs4_lookup_active(struct autofs_sb_info *sbi, struct dentry *parent, struct qstr *name)
+{
+	unsigned int len = name->len;
+	unsigned int hash = name->hash;
+	const unsigned char *str = name->name;
+	struct list_head *p, *head;
+
+	spin_lock(&dcache_lock);
+	spin_lock(&sbi->lookup_lock);
+	head = &sbi->active_list;
+	list_for_each(p, head) {
+		struct autofs_info *ino;
+		struct dentry *dentry;
+		struct qstr *qstr;
+
+		ino = list_entry(p, struct autofs_info, active);
+		dentry = ino->dentry;
+
+		spin_lock(&dentry->d_lock);
+
+		/* Already gone? */
+		if (atomic_read(&dentry->d_count) == 0)
+			goto next;
+
+		qstr = &dentry->d_name;
+
+		if (dentry->d_name.hash != hash)
+			goto next;
+		if (dentry->d_parent != parent)
+			goto next;
+
+		if (qstr->len != len)
+			goto next;
+		if (memcmp(qstr->name, str, len))
+			goto next;
+
+		if (d_unhashed(dentry)) {
+			dget(dentry);
+			spin_unlock(&dentry->d_lock);
+			spin_unlock(&sbi->lookup_lock);
+			spin_unlock(&dcache_lock);
+			return dentry;
+		}
+next:
+		spin_unlock(&dentry->d_lock);
+	}
+	spin_unlock(&sbi->lookup_lock);
+	spin_unlock(&dcache_lock);
+
+	return NULL;
+}
+
 static struct dentry *autofs4_lookup_expiring(struct autofs_sb_info *sbi, struct dentry *parent, struct qstr *name)
 {
 	unsigned int len = name->len;
@@ -553,7 +607,8 @@ next:
 static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
 {
 	struct autofs_sb_info *sbi;
-	struct dentry *expiring;
+	struct autofs_info *ino;
+	struct dentry *expiring, *unhashed;
 	int oz_mode;
 
 	DPRINTK("name = %.*s",
@@ -571,12 +626,12 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s
 
 	expiring = autofs4_lookup_expiring(sbi, dentry->d_parent, &dentry->d_name);
 	if (expiring) {
-		struct autofs_info *ino = autofs4_dentry_ino(expiring);
 		/*
 		 * If we are racing with expire the request might not
 		 * be quite complete but the directory has been removed
 		 * so it must have been successful, so just wait for it.
 		 */
+		ino = autofs4_dentry_ino(expiring);
 		while (ino && (ino->flags & AUTOFS_INF_EXPIRING)) {
 			DPRINTK("wait for incomplete expire %p name=%.*s",
 				expiring, expiring->d_name.len,
@@ -591,21 +646,41 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s
 		dput(expiring);
 	}
 
-	/*
-	 * Mark the dentry incomplete but don't hash it. We do this
-	 * to serialize our inode creation operations (symlink and
-	 * mkdir) which prevents deadlock during the callback to
-	 * the daemon. Subsequent user space lookups for the same
-	 * dentry are placed on the wait queue while the daemon
-	 * itself is allowed passage unresticted so the create
-	 * operation itself can then hash the dentry. Finally,
-	 * we check for the hashed dentry and return the newly
-	 * hashed dentry.
-	 */
-	dentry->d_op = &autofs4_root_dentry_operations;
+	unhashed = autofs4_lookup_active(sbi, dentry->d_parent, &dentry->d_name);
+	if (unhashed)
+		dentry = unhashed;
+	else {
+		/*
+		 * Mark the dentry incomplete but don't hash it. We do this
+		 * to serialize our inode creation operations (symlink and
+		 * mkdir) which prevents deadlock during the callback to
+		 * the daemon. Subsequent user space lookups for the same
+		 * dentry are placed on the wait queue while the daemon
+		 * itself is allowed passage unresticted so the create
+		 * operation itself can then hash the dentry. Finally,
+		 * we check for the hashed dentry and return the newly
+		 * hashed dentry.
+		 */
+		dentry->d_op = &autofs4_root_dentry_operations;
 
-	dentry->d_fsdata = NULL;
-	d_instantiate(dentry, NULL);
+		/*
+		 * And we need to ensure that the same dentry is used for
+		 * all following lookup calls until it is hashed so that
+		 * the dentry flags are persistent throughout the request.
+		 */
+		ino = autofs4_init_ino(NULL, sbi, 0555);
+		if (!ino)
+			return ERR_PTR(-ENOMEM);
+
+		dentry->d_fsdata = ino;
+		ino->dentry = dentry;
+
+		spin_lock(&sbi->lookup_lock);
+		list_add(&ino->active, &sbi->active_list);
+		spin_unlock(&sbi->lookup_lock);
+
+		d_instantiate(dentry, NULL);
+	}
 
 	if (!oz_mode) {
 		spin_lock(&dentry->d_lock);
@@ -630,12 +705,16 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s
 			if (sigismember (sigset, SIGKILL) ||
 			    sigismember (sigset, SIGQUIT) ||
 			    sigismember (sigset, SIGINT)) {
+			    if (unhashed)
+				dput(unhashed);
 			    return ERR_PTR(-ERESTARTNOINTR);
 			}
 		}
-		spin_lock(&dentry->d_lock);
-		dentry->d_flags &= ~DCACHE_AUTOFS_PENDING;
-		spin_unlock(&dentry->d_lock);
+		if (!oz_mode) {
+			spin_lock(&dentry->d_lock);
+			dentry->d_flags &= ~DCACHE_AUTOFS_PENDING;
+			spin_unlock(&dentry->d_lock);
+		}
 	}
 
 	/*
@@ -659,9 +738,15 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s
 		else
 			dentry = ERR_PTR(-ENOENT);
 
+		if (unhashed)
+			dput(unhashed);
+
 		return dentry;
 	}
 
+	if (unhashed)
+		return unhashed;
+
 	return NULL;
 }
 
@@ -682,20 +767,30 @@ static int autofs4_dir_symlink(struct inode *dir,
 		return -EACCES;
 
 	ino = autofs4_init_ino(ino, sbi, S_IFLNK | 0555);
-	if (ino == NULL)
-		return -ENOSPC;
+	if (!ino)
+		return -ENOMEM;
 
-	ino->size = strlen(symname);
-	ino->u.symlink = cp = kmalloc(ino->size + 1, GFP_KERNEL);
+	spin_lock(&sbi->lookup_lock);
+	if (!list_empty(&ino->active))
+		list_del_init(&ino->active);
+	spin_unlock(&sbi->lookup_lock);
 
-	if (cp == NULL) {
-		kfree(ino);
-		return -ENOSPC;
+	cp = kmalloc(ino->size + 1, GFP_KERNEL);
+	if (!cp) {
+		if (!dentry->d_fsdata)
+			kfree(ino);
+		return -ENOMEM;
 	}
 
 	strcpy(cp, symname);
 
 	inode = autofs4_get_inode(dir->i_sb, ino);
+	if (!inode) {
+		kfree(cp);
+		if (!dentry->d_fsdata)
+			kfree(ino);
+		return -ENOMEM;
+	}
 	d_add(dentry, inode);
 
 	if (dir == dir->i_sb->s_root->d_inode)
@@ -711,6 +806,8 @@ static int autofs4_dir_symlink(struct inode *dir,
 		atomic_inc(&p_ino->count);
 	ino->inode = inode;
 
+	ino->size = strlen(symname);
+	ino->u.symlink = cp;
 	dir->i_mtime = CURRENT_TIME;
 
 	return 0;
@@ -755,7 +852,8 @@ static int autofs4_dir_unlink(struct inode *dir, struct dentry *dentry)
 
 	spin_lock(&dcache_lock);
 	spin_lock(&sbi->lookup_lock);
-	list_add(&ino->expiring, &sbi->expiring_list);
+	if (list_empty(&ino->expiring))
+		list_add(&ino->expiring, &sbi->expiring_list);
 	spin_unlock(&sbi->lookup_lock);
 	spin_lock(&dentry->d_lock);
 	__d_drop(dentry);
@@ -783,7 +881,8 @@ static int autofs4_dir_rmdir(struct inode *dir, struct dentry *dentry)
 		return -ENOTEMPTY;
 	}
 	spin_lock(&sbi->lookup_lock);
-	list_add(&ino->expiring, &sbi->expiring_list);
+	if (list_empty(&ino->expiring))
+		list_add(&ino->expiring, &sbi->expiring_list);
 	spin_unlock(&sbi->lookup_lock);
 	spin_lock(&dentry->d_lock);
 	__d_drop(dentry);
@@ -819,10 +918,20 @@ static int autofs4_dir_mkdir(struct inode *dir, struct dentry *dentry, int mode)
 		dentry, dentry->d_name.len, dentry->d_name.name);
 
 	ino = autofs4_init_ino(ino, sbi, S_IFDIR | 0555);
-	if (ino == NULL)
-		return -ENOSPC;
+	if (!ino)
+		return -ENOMEM;
+
+	spin_lock(&sbi->lookup_lock);
+	if (!list_empty(&ino->active))
+		list_del_init(&ino->active);
+	spin_unlock(&sbi->lookup_lock);
 
 	inode = autofs4_get_inode(dir->i_sb, ino);
+	if (!inode) {
+		if (!dentry->d_fsdata)
+			kfree(ino);
+		return -ENOMEM;
+	}
 	d_add(dentry, inode);
 
 	if (dir == dir->i_sb->s_root->d_inode)

From ef581a742874ebc4c28d24b374c78b762144ebdc Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Wed, 23 Jul 2008 21:30:13 -0700
Subject: [PATCH 0408/1171] autofs4: fix symlink name allocation

The length of the symlink name has been moved but it needs to be set
before allocating space for it in the dentry info struct.  This corrects a
mistake in a recent patch.

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/root.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c
index dbb70d5a4882..324290c68275 100644
--- a/fs/autofs4/root.c
+++ b/fs/autofs4/root.c
@@ -775,6 +775,7 @@ static int autofs4_dir_symlink(struct inode *dir,
 		list_del_init(&ino->active);
 	spin_unlock(&sbi->lookup_lock);
 
+	ino->size = strlen(symname);
 	cp = kmalloc(ino->size + 1, GFP_KERNEL);
 	if (!cp) {
 		if (!dentry->d_fsdata)
@@ -806,7 +807,6 @@ static int autofs4_dir_symlink(struct inode *dir,
 		atomic_inc(&p_ino->count);
 	ino->inode = inode;
 
-	ino->size = strlen(symname);
 	ino->u.symlink = cp;
 	dir->i_mtime = CURRENT_TIME;
 

From c432c2586a0811c7d0030d78f0993568bc889a6f Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Wed, 23 Jul 2008 21:30:14 -0700
Subject: [PATCH 0409/1171] autofs4: don't release directory mutex if called in
 oz_mode

Since we now delay hashing of dentrys until the ->mkdir() call, droping
and re-taking the directory mutex within the ->lookup() function when we
are being called by user space is not needed.  This can lead to a race
when other processes are attempting to access the same directory during
mount point directory creation.

In this case we need to hang onto the mutex to ensure we don't get user
processes trying to create a mount request for a newly created dentry
after the mount point entry has already been created.  This ensures that
when we need to check a dentry passed to autofs4_wait(), if it is hashed,
it is always the mount point dentry and not a new dentry created by
another lookup during directory creation.

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/root.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)

diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c
index 324290c68275..1e901e5ea013 100644
--- a/fs/autofs4/root.c
+++ b/fs/autofs4/root.c
@@ -686,12 +686,11 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s
 		spin_lock(&dentry->d_lock);
 		dentry->d_flags |= DCACHE_AUTOFS_PENDING;
 		spin_unlock(&dentry->d_lock);
-	}
-
-	if (dentry->d_op && dentry->d_op->d_revalidate) {
-		mutex_unlock(&dir->i_mutex);
-		(dentry->d_op->d_revalidate)(dentry, nd);
-		mutex_lock(&dir->i_mutex);
+		if (dentry->d_op && dentry->d_op->d_revalidate) {
+			mutex_unlock(&dir->i_mutex);
+			(dentry->d_op->d_revalidate)(dentry, nd);
+			mutex_lock(&dir->i_mutex);
+		}
 	}
 
 	/*

From 6d5cb926fa0162b1e62f37c117cc7ce763cfcbb9 Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Wed, 23 Jul 2008 21:30:15 -0700
Subject: [PATCH 0410/1171] autofs4: use lookup intent flags to trigger mounts

When an open(2) call is made on an autofs mount point directory that
already exists and the O_DIRECTORY flag is not used the needed mount
callback to the daemon is not done. This leads to the path walk
continuing resulting in a callback to the daemon with an incorrect
key. open(2) is called without O_DIRECTORY by the "find" utility but
this should be handled properly anyway.

This happens because autofs needs to use the lookup flags to decide
when to callback to the daemon to perform a mount to prevent mount
storms. For example, an autofs indirect mount map that has the "browse"
option will have the mount point directories are pre-created and the
stat(2) call made by a color ls against each directory will cause all
these directories to be mounted. It is unfortunate we need to resort
to this but mount maps can be quite large. Additionally, if a user
manually umounts an autofs indirect mount the directory isn't removed
which also leads to this situation.

To resolve this autofs needs to use the lookup intent flags to enable
it to make this decision. This patch adds this check and triggers a
call back if any of the lookup intent flags are set as all these calls
warrant a mount attempt be requested.

I know that external VFS code which uses the lookup flags is something
that the VFS would like to eliminate but I have no choice as I can't
see any other way to do this. A VFS dentry or inode operation callback
which returns the lookup "type" (requires a definition) would be
sufficient. But this change is needed now and I'm not aware of the form
that coming VFS changes will take so I'm not willing to propose anything
along these lines.

If anyone can provide an alternate method I would be happy to use it.

[akpm@linux-foundation.org: fix build for concurrent VFS changes]
Signed-off-by: Ian Kent <raven@themaw.net>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/root.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c
index 1e901e5ea013..87352654ff4e 100644
--- a/fs/autofs4/root.c
+++ b/fs/autofs4/root.c
@@ -31,6 +31,9 @@ static int autofs4_root_readdir(struct file * filp, void * dirent, filldir_t fil
 static struct dentry *autofs4_lookup(struct inode *,struct dentry *, struct nameidata *);
 static void *autofs4_follow_link(struct dentry *, struct nameidata *);
 
+#define TRIGGER_FLAGS   (LOOKUP_CONTINUE | LOOKUP_DIRECTORY)
+#define TRIGGER_INTENTS (LOOKUP_OPEN | LOOKUP_CREATE)
+
 const struct file_operations autofs4_root_operations = {
 	.open		= dcache_dir_open,
 	.release	= dcache_dir_close,
@@ -291,7 +294,7 @@ static int try_to_fill_dentry(struct dentry *dentry, int flags)
 			return status;
 		}
 	/* Trigger mount for path component or follow link */
-	} else if (flags & (LOOKUP_CONTINUE | LOOKUP_DIRECTORY) ||
+	} else if (flags & (TRIGGER_FLAGS | TRIGGER_INTENTS) ||
 			current->link_count) {
 		DPRINTK("waiting for mount name=%.*s",
 			dentry->d_name.len, dentry->d_name.name);
@@ -336,7 +339,7 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd)
 		nd->flags);
 
 	/* If it's our master or we shouldn't trigger a mount we're done */
-	lookup_type = nd->flags & (LOOKUP_CONTINUE | LOOKUP_DIRECTORY);
+	lookup_type = nd->flags & (TRIGGER_FLAGS | TRIGGER_INTENTS);
 	if (oz_mode || !lookup_type)
 		goto done;
 

From 70b52a0a5005ce6a0ceec56e97222437a0ba7506 Mon Sep 17 00:00:00 2001
From: Jeff Moyer <jmoyer@redhat.com>
Date: Wed, 23 Jul 2008 21:30:16 -0700
Subject: [PATCH 0411/1171] autofs4: use struct qstr in waitq.c

The autofs_wait_queue already contains all of the fields of the
struct qstr, so change it into a qstr.

This patch, from Jeff Moyer, has been modified a liitle by myself.

Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/autofs_i.h |  4 +-
 fs/autofs4/waitq.c    | 86 ++++++++++++++++++++++---------------------
 2 files changed, 46 insertions(+), 44 deletions(-)

diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h
index 2dce2334737d..da8882ff31e6 100644
--- a/fs/autofs4/autofs_i.h
+++ b/fs/autofs4/autofs_i.h
@@ -75,9 +75,7 @@ struct autofs_wait_queue {
 	struct autofs_wait_queue *next;
 	autofs_wqt_t wait_queue_token;
 	/* We use the following to see what we are waiting for */
-	unsigned int hash;
-	unsigned int len;
-	char *name;
+	struct qstr name;
 	u32 dev;
 	u64 ino;
 	uid_t uid;
diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c
index 75e5955c3f6d..5208cfb1df4e 100644
--- a/fs/autofs4/waitq.c
+++ b/fs/autofs4/waitq.c
@@ -36,8 +36,10 @@ void autofs4_catatonic_mode(struct autofs_sb_info *sbi)
 	while (wq) {
 		nwq = wq->next;
 		wq->status = -ENOENT; /* Magic is gone - report failure */
-		kfree(wq->name);
-		wq->name = NULL;
+		if (wq->name.name) {
+			kfree(wq->name.name);
+			wq->name.name = NULL;
+		}
 		wake_up_interruptible(&wq->queue);
 		wq = nwq;
 	}
@@ -92,7 +94,7 @@ static void autofs4_notify_daemon(struct autofs_sb_info *sbi,
 	size_t pktsz;
 
 	DPRINTK("wait id = 0x%08lx, name = %.*s, type=%d",
-		wq->wait_queue_token, wq->len, wq->name, type);
+		wq->wait_queue_token, wq->name.len, wq->name.name, type);
 
 	memset(&pkt,0,sizeof pkt); /* For security reasons */
 
@@ -107,9 +109,9 @@ static void autofs4_notify_daemon(struct autofs_sb_info *sbi,
 		pktsz = sizeof(*mp);
 
 		mp->wait_queue_token = wq->wait_queue_token;
-		mp->len = wq->len;
-		memcpy(mp->name, wq->name, wq->len);
-		mp->name[wq->len] = '\0';
+		mp->len = wq->name.len;
+		memcpy(mp->name, wq->name.name, wq->name.len);
+		mp->name[wq->name.len] = '\0';
 		break;
 	}
 	case autofs_ptype_expire_multi:
@@ -119,9 +121,9 @@ static void autofs4_notify_daemon(struct autofs_sb_info *sbi,
 		pktsz = sizeof(*ep);
 
 		ep->wait_queue_token = wq->wait_queue_token;
-		ep->len = wq->len;
-		memcpy(ep->name, wq->name, wq->len);
-		ep->name[wq->len] = '\0';
+		ep->len = wq->name.len;
+		memcpy(ep->name, wq->name.name, wq->name.len);
+		ep->name[wq->name.len] = '\0';
 		break;
 	}
 	/*
@@ -138,9 +140,9 @@ static void autofs4_notify_daemon(struct autofs_sb_info *sbi,
 		pktsz = sizeof(*packet);
 
 		packet->wait_queue_token = wq->wait_queue_token;
-		packet->len = wq->len;
-		memcpy(packet->name, wq->name, wq->len);
-		packet->name[wq->len] = '\0';
+		packet->len = wq->name.len;
+		memcpy(packet->name, wq->name.name, wq->name.len);
+		packet->name[wq->name.len] = '\0';
 		packet->dev = wq->dev;
 		packet->ino = wq->ino;
 		packet->uid = wq->uid;
@@ -191,15 +193,15 @@ static int autofs4_getpath(struct autofs_sb_info *sbi,
 }
 
 static struct autofs_wait_queue *
-autofs4_find_wait(struct autofs_sb_info *sbi,
-		  char *name, unsigned int hash, unsigned int len)
+autofs4_find_wait(struct autofs_sb_info *sbi, struct qstr *qstr)
 {
 	struct autofs_wait_queue *wq;
 
 	for (wq = sbi->queues; wq; wq = wq->next) {
-		if (wq->hash == hash &&
-		    wq->len == len &&
-		    wq->name && !memcmp(wq->name, name, len))
+		if (wq->name.hash == qstr->hash &&
+		    wq->name.len == qstr->len &&
+		    wq->name.name &&
+			 !memcmp(wq->name.name, qstr->name, qstr->len))
 			break;
 	}
 	return wq;
@@ -210,9 +212,8 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 {
 	struct autofs_info *ino;
 	struct autofs_wait_queue *wq;
+	struct qstr qstr;
 	char *name;
-	unsigned int len = 0;
-	unsigned int hash = 0;
 	int status, type;
 
 	/* In catatonic mode, we don't wait for nobody */
@@ -225,22 +226,23 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 
 	/* If this is a direct mount request create a dummy name */
 	if (IS_ROOT(dentry) && (sbi->type & AUTOFS_TYPE_DIRECT))
-		len = sprintf(name, "%p", dentry);
+		qstr.len = sprintf(name, "%p", dentry);
 	else {
-		len = autofs4_getpath(sbi, dentry, &name);
-		if (!len) {
+		qstr.len = autofs4_getpath(sbi, dentry, &name);
+		if (!qstr.len) {
 			kfree(name);
 			return -ENOENT;
 		}
 	}
-	hash = full_name_hash(name, len);
+	qstr.name = name;
+	qstr.hash = full_name_hash(name, qstr.len);
 
 	if (mutex_lock_interruptible(&sbi->wq_mutex)) {
-		kfree(name);
+		kfree(qstr.name);
 		return -EINTR;
 	}
 
-	wq = autofs4_find_wait(sbi, name, hash, len);
+	wq = autofs4_find_wait(sbi, &qstr);
 	ino = autofs4_dentry_ino(dentry);
 	if (!wq && ino && notify == NFY_NONE) {
 		/*
@@ -254,10 +256,10 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 			mutex_unlock(&sbi->wq_mutex);
 			schedule_timeout_interruptible(HZ/10);
 			if (mutex_lock_interruptible(&sbi->wq_mutex)) {
-				kfree(name);
+				kfree(qstr.name);
 				return -EINTR;
 			}
-			wq = autofs4_find_wait(sbi, name, hash, len);
+			wq = autofs4_find_wait(sbi, &qstr);
 			if (wq)
 				break;
 		}
@@ -268,7 +270,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 		 * return status of the wait.
 		 */
 		if (!wq) {
-			kfree(name);
+			kfree(qstr.name);
 			mutex_unlock(&sbi->wq_mutex);
 			return 0;
 		}
@@ -278,7 +280,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 		/* Create a new wait queue */
 		wq = kmalloc(sizeof(struct autofs_wait_queue),GFP_KERNEL);
 		if (!wq) {
-			kfree(name);
+			kfree(qstr.name);
 			mutex_unlock(&sbi->wq_mutex);
 			return -ENOMEM;
 		}
@@ -289,9 +291,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 		wq->next = sbi->queues;
 		sbi->queues = wq;
 		init_waitqueue_head(&wq->queue);
-		wq->hash = hash;
-		wq->name = name;
-		wq->len = len;
+		memcpy(&wq->name, &qstr, sizeof(struct qstr));
 		wq->dev = autofs4_get_dev(sbi);
 		wq->ino = autofs4_get_ino(sbi);
 		wq->uid = current->uid;
@@ -319,16 +319,18 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 		}
 
 		DPRINTK("new wait id = 0x%08lx, name = %.*s, nfy=%d\n",
-			(unsigned long) wq->wait_queue_token, wq->len, wq->name, notify);
+			(unsigned long) wq->wait_queue_token, wq->name.len,
+			wq->name.name, notify);
 
 		/* autofs4_notify_daemon() may block */
 		autofs4_notify_daemon(sbi, wq, type);
 	} else {
 		atomic_inc(&wq->wait_ctr);
 		mutex_unlock(&sbi->wq_mutex);
-		kfree(name);
+		kfree(qstr.name);
 		DPRINTK("existing wait id = 0x%08lx, name = %.*s, nfy=%d",
-			(unsigned long) wq->wait_queue_token, wq->len, wq->name, notify);
+			(unsigned long) wq->wait_queue_token, wq->name.len,
+			wq->name.name, notify);
 	}
 
 	/* wq->name is NULL if and only if the lock is already released */
@@ -336,11 +338,13 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 	if (sbi->catatonic) {
 		/* We might have slept, so check again for catatonic mode */
 		wq->status = -ENOENT;
-		kfree(wq->name);
-		wq->name = NULL;
+		if (wq->name.name) {
+			kfree(wq->name.name);
+			wq->name.name = NULL;
+		}
 	}
 
-	if (wq->name) {
+	if (wq->name.name) {
 		/* Block all but "shutdown" signals while waiting */
 		sigset_t oldset;
 		unsigned long irqflags;
@@ -351,7 +355,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 		recalc_sigpending();
 		spin_unlock_irqrestore(&current->sighand->siglock, irqflags);
 
-		wait_event_interruptible(wq->queue, wq->name == NULL);
+		wait_event_interruptible(wq->queue, wq->name.name == NULL);
 
 		spin_lock_irqsave(&current->sighand->siglock, irqflags);
 		current->blocked = oldset;
@@ -388,8 +392,8 @@ int autofs4_wait_release(struct autofs_sb_info *sbi, autofs_wqt_t wait_queue_tok
 
 	*wql = wq->next;	/* Unlink from chain */
 	mutex_unlock(&sbi->wq_mutex);
-	kfree(wq->name);
-	wq->name = NULL;	/* Do not wait on this queue */
+	kfree(wq->name.name);
+	wq->name.name = NULL;	/* Do not wait on this queue */
 
 	wq->status = status;
 

From 5a11d4d0ee1ff284271f7265929d07ea4a1168a6 Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Wed, 23 Jul 2008 21:30:17 -0700
Subject: [PATCH 0412/1171] autofs4: fix waitq locking

The autofs4_catatonic_mode() function accesses the wait queue without any
locking but can be called at any time.  This could lead to a possible
double free of the name field of the wait and a double fput of the daemon
communication pipe or an fput of a NULL file pointer.

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/inode.c |  4 ++--
 fs/autofs4/waitq.c | 23 ++++++++++++-----------
 2 files changed, 14 insertions(+), 13 deletions(-)

diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c
index e3e70994ab46..7bb3e5ba0537 100644
--- a/fs/autofs4/inode.c
+++ b/fs/autofs4/inode.c
@@ -163,8 +163,8 @@ void autofs4_kill_sb(struct super_block *sb)
 	if (!sbi)
 		goto out_kill_sb;
 
-	if (!sbi->catatonic)
-		autofs4_catatonic_mode(sbi); /* Free wait queues, close pipe */
+	/* Free wait queues, close pipe */
+	autofs4_catatonic_mode(sbi);
 
 	/* Clean up and release dangling references */
 	autofs4_force_release(sbi);
diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c
index 5208cfb1df4e..55aac10cf328 100644
--- a/fs/autofs4/waitq.c
+++ b/fs/autofs4/waitq.c
@@ -28,6 +28,12 @@ void autofs4_catatonic_mode(struct autofs_sb_info *sbi)
 {
 	struct autofs_wait_queue *wq, *nwq;
 
+	mutex_lock(&sbi->wq_mutex);
+	if (sbi->catatonic) {
+		mutex_unlock(&sbi->wq_mutex);
+		return;
+	}
+
 	DPRINTK("entering catatonic mode");
 
 	sbi->catatonic = 1;
@@ -45,6 +51,8 @@ void autofs4_catatonic_mode(struct autofs_sb_info *sbi)
 	}
 	fput(sbi->pipe);	/* Close the pipe */
 	sbi->pipe = NULL;
+	sbi->pipefd = -1;
+	mutex_unlock(&sbi->wq_mutex);
 }
 
 static int autofs4_write(struct file *file, const void *addr, int bytes)
@@ -333,17 +341,10 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 			wq->name.name, notify);
 	}
 
-	/* wq->name is NULL if and only if the lock is already released */
-
-	if (sbi->catatonic) {
-		/* We might have slept, so check again for catatonic mode */
-		wq->status = -ENOENT;
-		if (wq->name.name) {
-			kfree(wq->name.name);
-			wq->name.name = NULL;
-		}
-	}
-
+	/*
+	 * wq->name.name is NULL iff the lock is already released
+	 * or the mount has been made catatonic.
+	 */
 	if (wq->name.name) {
 		/* Block all but "shutdown" signals while waiting */
 		sigset_t oldset;

From a1362fe92f1bde687b3a9e93d6b8d105d0a84f74 Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Wed, 23 Jul 2008 21:30:19 -0700
Subject: [PATCH 0413/1171] autofs4: fix pending mount race

Close a race between a pending mount that is about to finish and a new
lookup for the same directory.

Process P1 triggers a mount of directory foo.  It sets
DCACHE_AUTOFS_PENDING in the ->lookup routine, creates a waitq entry for
'foo', and calls out to the daemon to perform the mount.  The autofs
daemon will then create the directory 'foo', using a new dentry that will
be hashed in the dcache.

Before the mount completes, another process, P2, tries to walk into the
'foo' directory.  The vfs path walking code finds an entry for 'foo' and
calls the revalidate method.  Revalidate finds that the entry is not
PENDING (because PENDING was never set on the dentry created by the
mkdir), but it does find the directory is empty.  Revalidate calls
try_to_fill_dentry, which sets the PENDING flag and then calls into the
autofs4 wait code to trigger or wait for a mount of 'foo'.  The wait code
finds the entry for 'foo' and goes to sleep waiting for the completion of
the mount.

Yet another process, P3, tries to walk into the 'foo' directory.  This
process again finds a dentry in the dcache for 'foo', and calls into the
autofs revalidate code.

The revalidate code finds that the PENDING flag is set, and so calls
try_to_fill_dentry.

a) try_to_fill_dentry sets the PENDING flag redundantly for this
   dentry, then calls into the autofs4 wait code.

b) the autofs4 wait code takes the waitq mutex and searches for an
   entry for 'foo'

Between a and b, P1 is woken up because the mount completed.  P1 takes the
wait queue mutex, clears the PENDING flag from the dentry, and removes the
waitqueue entry for 'foo' from the list.

When it releases the waitq mutex, P3 (eventually) acquires it.  At this
time, it looks for an existing waitq for 'foo', finds none, and so creates
a new one and calls out to the daemon to mount the 'foo' directory.

Now, the reason that three processes are required to trigger this race is
that, because the PENDING flag is not set on the dentry created by mkdir,
the window for the race would be way to slim for it to ever occur.
Basically, between the testing of d_mountpoint(dentry) and the taking of
the waitq mutex, the mount would have to complete and the daemon would
have to be woken up, and that in turn would have to wake up P1.  This is
simply impossible.  Add the third process, though, and it becomes slightly
more likely.

Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/waitq.c | 135 ++++++++++++++++++++++++++++++++-------------
 1 file changed, 97 insertions(+), 38 deletions(-)

diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c
index 55aac10cf328..cd3b2a671696 100644
--- a/fs/autofs4/waitq.c
+++ b/fs/autofs4/waitq.c
@@ -215,19 +215,106 @@ autofs4_find_wait(struct autofs_sb_info *sbi, struct qstr *qstr)
 	return wq;
 }
 
+/*
+ * Check if we have a valid request.
+ * Returns
+ * 1 if the request should continue.
+ *   In this case we can return an autofs_wait_queue entry if one is
+ *   found or NULL to idicate a new wait needs to be created.
+ * 0 or a negative errno if the request shouldn't continue.
+ */
+static int validate_request(struct autofs_wait_queue **wait,
+			    struct autofs_sb_info *sbi,
+			    struct qstr *qstr,
+			    struct dentry*dentry, enum autofs_notify notify)
+{
+	struct autofs_wait_queue *wq;
+	struct autofs_info *ino;
+
+	/* Wait in progress, continue; */
+	wq = autofs4_find_wait(sbi, qstr);
+	if (wq) {
+		*wait = wq;
+		return 1;
+	}
+
+	*wait = NULL;
+
+	/* If we don't yet have any info this is a new request */
+	ino = autofs4_dentry_ino(dentry);
+	if (!ino)
+		return 1;
+
+	/*
+	 * If we've been asked to wait on an existing expire (NFY_NONE)
+	 * but there is no wait in the queue ...
+	 */
+	if (notify == NFY_NONE) {
+		/*
+		 * Either we've betean the pending expire to post it's
+		 * wait or it finished while we waited on the mutex.
+		 * So we need to wait till either, the wait appears
+		 * or the expire finishes.
+		 */
+
+		while (ino->flags & AUTOFS_INF_EXPIRING) {
+			mutex_unlock(&sbi->wq_mutex);
+			schedule_timeout_interruptible(HZ/10);
+			if (mutex_lock_interruptible(&sbi->wq_mutex))
+				return -EINTR;
+
+			wq = autofs4_find_wait(sbi, qstr);
+			if (wq) {
+				*wait = wq;
+				return 1;
+			}
+		}
+
+		/*
+		 * Not ideal but the status has already gone. Of the two
+		 * cases where we wait on NFY_NONE neither depend on the
+		 * return status of the wait.
+		 */
+		return 0;
+	}
+
+	/*
+	 * If we've been asked to trigger a mount and the request
+	 * completed while we waited on the mutex ...
+	 */
+	if (notify == NFY_MOUNT) {
+		/*
+		 * If the dentry isn't hashed just go ahead and try the
+		 * mount again with a new wait (not much else we can do).
+		*/
+		if (!d_unhashed(dentry)) {
+			/*
+			 * But if the dentry is hashed, that means that we
+			 * got here through the revalidate path.  Thus, we
+			 * need to check if the dentry has been mounted
+			 * while we waited on the wq_mutex. If it has,
+			 * simply return success.
+			 */
+			if (d_mountpoint(dentry))
+				return 0;
+		}
+	}
+
+	return 1;
+}
+
 int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 		enum autofs_notify notify)
 {
-	struct autofs_info *ino;
 	struct autofs_wait_queue *wq;
 	struct qstr qstr;
 	char *name;
-	int status, type;
+	int status, ret, type;
 
 	/* In catatonic mode, we don't wait for nobody */
 	if (sbi->catatonic)
 		return -ENOENT;
-	
+
 	name = kmalloc(NAME_MAX + 1, GFP_KERNEL);
 	if (!name)
 		return -ENOMEM;
@@ -245,43 +332,15 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 	qstr.name = name;
 	qstr.hash = full_name_hash(name, qstr.len);
 
-	if (mutex_lock_interruptible(&sbi->wq_mutex)) {
-		kfree(qstr.name);
+	if (mutex_lock_interruptible(&sbi->wq_mutex))
 		return -EINTR;
-	}
 
-	wq = autofs4_find_wait(sbi, &qstr);
-	ino = autofs4_dentry_ino(dentry);
-	if (!wq && ino && notify == NFY_NONE) {
-		/*
-		 * Either we've betean the pending expire to post it's
-		 * wait or it finished while we waited on the mutex.
-		 * So we need to wait till either, the wait appears
-		 * or the expire finishes.
-		 */
-
-		while (ino->flags & AUTOFS_INF_EXPIRING) {
+	ret = validate_request(&wq, sbi, &qstr, dentry, notify);
+	if (ret <= 0) {
+		if (ret == 0)
 			mutex_unlock(&sbi->wq_mutex);
-			schedule_timeout_interruptible(HZ/10);
-			if (mutex_lock_interruptible(&sbi->wq_mutex)) {
-				kfree(qstr.name);
-				return -EINTR;
-			}
-			wq = autofs4_find_wait(sbi, &qstr);
-			if (wq)
-				break;
-		}
-
-		/*
-		 * Not ideal but the status has already gone. Of the two
-		 * cases where we wait on NFY_NONE neither depend on the
-		 * return status of the wait.
-		 */
-		if (!wq) {
-			kfree(qstr.name);
-			mutex_unlock(&sbi->wq_mutex);
-			return 0;
-		}
+		kfree(qstr.name);
+		return ret;
 	}
 
 	if (!wq) {
@@ -392,9 +451,9 @@ int autofs4_wait_release(struct autofs_sb_info *sbi, autofs_wqt_t wait_queue_tok
 	}
 
 	*wql = wq->next;	/* Unlink from chain */
-	mutex_unlock(&sbi->wq_mutex);
 	kfree(wq->name.name);
 	wq->name.name = NULL;	/* Do not wait on this queue */
+	mutex_unlock(&sbi->wq_mutex);
 
 	wq->status = status;
 

From f4c7da02615bebcaf89f15a8d055922f515160b8 Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Wed, 23 Jul 2008 21:30:19 -0700
Subject: [PATCH 0414/1171] autofs4: add missing kfree

It see that the patch tittled "autofs4 - fix pending mount race" is
missing a change that I had recently made.

It's missing a kfree for the case mutex_lock_interruptible() fails
to aquire the wait queue mutex.

Signed-off-by: Ian Kent <raven@themaw.net>
Cc: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/waitq.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c
index cd3b2a671696..1132cc2a0310 100644
--- a/fs/autofs4/waitq.c
+++ b/fs/autofs4/waitq.c
@@ -332,8 +332,10 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 	qstr.name = name;
 	qstr.hash = full_name_hash(name, qstr.len);
 
-	if (mutex_lock_interruptible(&sbi->wq_mutex))
+	if (mutex_lock_interruptible(&sbi->wq_mutex)) {
+		kfree(qstr.name);
 		return -EINTR;
+	}
 
 	ret = validate_request(&wq, sbi, &qstr, dentry, notify);
 	if (ret <= 0) {

From e64be33ccaceaca67c84237dff8805b861398eab Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Wed, 23 Jul 2008 21:30:20 -0700
Subject: [PATCH 0415/1171] autofs4: check kernel communication pipe is valid
 for write

It is possible for an autofs mount to become catatonic (and for the daemon
communication pipe to become NULL) after a wait has been initiallized but
before the request has been sent to the daemon.  We need to check for this
before sending the request packet.

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/waitq.c | 16 ++++++++++++++--
 1 file changed, 14 insertions(+), 2 deletions(-)

diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c
index 1132cc2a0310..dd2914d7ad7f 100644
--- a/fs/autofs4/waitq.c
+++ b/fs/autofs4/waitq.c
@@ -99,6 +99,7 @@ static void autofs4_notify_daemon(struct autofs_sb_info *sbi,
 		union autofs_packet_union v4_pkt;
 		union autofs_v5_packet_union v5_pkt;
 	} pkt;
+	struct file *pipe = NULL;
 	size_t pktsz;
 
 	DPRINTK("wait id = 0x%08lx, name = %.*s, type=%d",
@@ -164,8 +165,19 @@ static void autofs4_notify_daemon(struct autofs_sb_info *sbi,
 		return;
 	}
 
-	if (autofs4_write(sbi->pipe, &pkt, pktsz))
-		autofs4_catatonic_mode(sbi);
+	/* Check if we have become catatonic */
+	mutex_lock(&sbi->wq_mutex);
+	if (!sbi->catatonic) {
+		pipe = sbi->pipe;
+		get_file(pipe);
+	}
+	mutex_unlock(&sbi->wq_mutex);
+
+	if (pipe) {
+		if (autofs4_write(pipe, &pkt, pktsz))
+			autofs4_catatonic_mode(sbi);
+		fput(pipe);
+	}
 }
 
 static int autofs4_getpath(struct autofs_sb_info *sbi,

From 296f7bf78bc5c7a4d772aea580ce800d14040d1a Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Wed, 23 Jul 2008 21:30:21 -0700
Subject: [PATCH 0416/1171] autofs4: fix waitq memory leak

If an autofs mount becomes catatonic before autofs4_wait_release() is
called the wait queue counter will not be decremented down to zero and the
entry will never be freed.  There are also races decrementing the wait
counter in the wait release function.  To deal with this the counter needs
to be updated while holding the wait queue mutex and waiters need to be
woken up unconditionally when the wait is removed from the queue to ensure
we eventually free the wait.

Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/autofs_i.h |  2 +-
 fs/autofs4/waitq.c    | 18 +++++++++---------
 2 files changed, 10 insertions(+), 10 deletions(-)

diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h
index da8882ff31e6..058e1800caec 100644
--- a/fs/autofs4/autofs_i.h
+++ b/fs/autofs4/autofs_i.h
@@ -84,7 +84,7 @@ struct autofs_wait_queue {
 	pid_t tgid;
 	/* This is for status reporting upon return */
 	int status;
-	atomic_t wait_ctr;
+	unsigned int wait_ctr;
 };
 
 #define AUTOFS_SBI_MAGIC 0x6d4a556d
diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c
index dd2914d7ad7f..3458dbc8fff0 100644
--- a/fs/autofs4/waitq.c
+++ b/fs/autofs4/waitq.c
@@ -46,6 +46,7 @@ void autofs4_catatonic_mode(struct autofs_sb_info *sbi)
 			kfree(wq->name.name);
 			wq->name.name = NULL;
 		}
+		wq->wait_ctr--;
 		wake_up_interruptible(&wq->queue);
 		wq = nwq;
 	}
@@ -380,7 +381,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 		wq->pid = current->pid;
 		wq->tgid = current->tgid;
 		wq->status = -EINTR; /* Status return if interrupted */
-		atomic_set(&wq->wait_ctr, 2);
+		wq->wait_ctr = 2;
 		mutex_unlock(&sbi->wq_mutex);
 
 		if (sbi->version < 5) {
@@ -406,7 +407,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 		/* autofs4_notify_daemon() may block */
 		autofs4_notify_daemon(sbi, wq, type);
 	} else {
-		atomic_inc(&wq->wait_ctr);
+		wq->wait_ctr++;
 		mutex_unlock(&sbi->wq_mutex);
 		kfree(qstr.name);
 		DPRINTK("existing wait id = 0x%08lx, name = %.*s, nfy=%d",
@@ -442,8 +443,10 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 	status = wq->status;
 
 	/* Are we the last process to need status? */
-	if (atomic_dec_and_test(&wq->wait_ctr))
+	mutex_lock(&sbi->wq_mutex);
+	if (!--wq->wait_ctr)
 		kfree(wq);
+	mutex_unlock(&sbi->wq_mutex);
 
 	return status;
 }
@@ -467,14 +470,11 @@ int autofs4_wait_release(struct autofs_sb_info *sbi, autofs_wqt_t wait_queue_tok
 	*wql = wq->next;	/* Unlink from chain */
 	kfree(wq->name.name);
 	wq->name.name = NULL;	/* Do not wait on this queue */
-	mutex_unlock(&sbi->wq_mutex);
-
 	wq->status = status;
-
-	if (atomic_dec_and_test(&wq->wait_ctr))	/* Is anyone still waiting for this guy? */
+	wake_up_interruptible(&wq->queue);
+	if (!--wq->wait_ctr)
 		kfree(wq);
-	else
-		wake_up_interruptible(&wq->queue);
+	mutex_unlock(&sbi->wq_mutex);
 
 	return 0;
 }

From eb3b176796b0e53fd26fce86847231542eb0d198 Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Wed, 23 Jul 2008 21:30:22 -0700
Subject: [PATCH 0417/1171] autofs4: detect invalid direct mount requests

autofs v5 direct and offset mounts within an autofs filesystem are
triggered by existing autofs triger mounts so the mount point dentry must
be positive.  If the mount point dentry is negative then the trigger
doesn't exist so we can return fail immediately.

Signed-off-by: Ian Kent <raven@themaw.net>
Cc: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/waitq.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c
index 3458dbc8fff0..bcb6c5265467 100644
--- a/fs/autofs4/waitq.c
+++ b/fs/autofs4/waitq.c
@@ -328,6 +328,10 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 	if (sbi->catatonic)
 		return -ENOENT;
 
+	if (!dentry->d_inode &&
+	    (sbi->type & (AUTOFS_TYPE_DIRECT | AUTOFS_TYPE_OFFSET)))
+		return -ENOENT;
+
 	name = kmalloc(NAME_MAX + 1, GFP_KERNEL);
 	if (!name)
 		return -ENOMEM;

From c72305b5472522299bb6f45b736080128eb1c822 Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Wed, 23 Jul 2008 21:30:23 -0700
Subject: [PATCH 0418/1171] autofs4: indirect dentry must almost always be
 positive

We have been seeing mount requests comming to the automount daemon for
keys of the form "<map key>/<non key directory>" which are lookups for
invalid map keys.  But we can check for this in the kernel module and
return a fail immediately, without having to send a request to the daemon.

It is possible to recognise these requests are invalid based on whether
the request dentry is negative and its relation to the autofs file system
root.

For example, given the indirect multi-mount map entry:

idm1  \
    /mm1  <server>:/<path1>
    /mm2  <server>:/<path2>

For a request to mount idm1, IS_ROOT((idm1)->d_parent) will be always be
true and the dentry may be negative.  But directories idm1/mm1 and
idm1/mm2 will always be created as part of the mount request for idm1.  So
any mount request within idm1 itself must have a positive dentry otherwise
the map key is invalid.

In version 4 these multi-mount entries are all mounted and umounted as a
single request and in version 5 the directories idm1/mm1 and idm1/mm2 are
created and an autofs fs mounted on them to act as a mount trigger so the
above is also true.

This also holds true for the autofs version 4 pseudo direct mount feature.
 When this feature is used without the "--ghost" option automount(8) will
create internal submounts as we go down the map key paths which are
essentially normal indirect mounts for which the above holds.  If the
"--ghost" option is given the directories for map keys are created at
daemon startup so valid map entries correspond to postive dentries in the
autofs fs.

autofs version 5 direct mount maps are similar except that the IS_ROOT
check is not needed.  This has been addressed in a previous patch tittled
"autofs4 - detect invalid direct mount requests".

For example, given the direct multi-mount map entry:

/test/dm1  \
    /mm1  <server>:/<path1>
    /mm2  <server>:/<path2>

An autofs fs is mounted on /test/dm1 as a trigger mount and when a mount
is triggered for /test/dm1, the multi-mount offset directories
/test/dm1/mm1 and /test/dm1/mm2 are created and an autofs fs is mounted on
them to act as mount triggers.  So valid direct mount requests must always
have a positive dentry if they correspond to a valid map entry.

Signed-off-by: Ian Kent <raven@themaw.net>
Acked-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/waitq.c | 17 ++++++++++++++---
 1 file changed, 14 insertions(+), 3 deletions(-)

diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c
index bcb6c5265467..35216d18d8b5 100644
--- a/fs/autofs4/waitq.c
+++ b/fs/autofs4/waitq.c
@@ -328,9 +328,20 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
 	if (sbi->catatonic)
 		return -ENOENT;
 
-	if (!dentry->d_inode &&
-	    (sbi->type & (AUTOFS_TYPE_DIRECT | AUTOFS_TYPE_OFFSET)))
-		return -ENOENT;
+	if (!dentry->d_inode) {
+		/*
+		 * A wait for a negative dentry is invalid for certain
+		 * cases. A direct or offset mount "always" has its mount
+		 * point directory created and so the request dentry must
+		 * be positive or the map key doesn't exist. The situation
+		 * is very similar for indirect mounts except only dentrys
+		 * in the root of the autofs file system may be negative.
+		 */
+		if (sbi->type & (AUTOFS_TYPE_DIRECT|AUTOFS_TYPE_OFFSET))
+			return -ENOENT;
+		else if (!IS_ROOT(dentry->d_parent))
+			return -ENOENT;
+	}
 
 	name = kmalloc(NAME_MAX + 1, GFP_KERNEL);
 	if (!name)

From ff9cd499d6258952385cb2f12e9a3c0908fd5786 Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Wed, 23 Jul 2008 21:30:24 -0700
Subject: [PATCH 0419/1171] autofs4: cleanup redundant readir code

The mount triggering functionality of readdir and related functions is no
longer used (and is quite broken as well).  The unused portions have been
removed.

Signed-off-by: Ian Kent <raven@themaw.net>
Reviewed-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/root.c | 149 +++++-----------------------------------------
 1 file changed, 16 insertions(+), 133 deletions(-)

diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c
index 87352654ff4e..51c873ca8e8d 100644
--- a/fs/autofs4/root.c
+++ b/fs/autofs4/root.c
@@ -25,8 +25,6 @@ static int autofs4_dir_rmdir(struct inode *,struct dentry *);
 static int autofs4_dir_mkdir(struct inode *,struct dentry *,int);
 static int autofs4_root_ioctl(struct inode *, struct file *,unsigned int,unsigned long);
 static int autofs4_dir_open(struct inode *inode, struct file *file);
-static int autofs4_dir_close(struct inode *inode, struct file *file);
-static int autofs4_dir_readdir(struct file * filp, void * dirent, filldir_t filldir);
 static int autofs4_root_readdir(struct file * filp, void * dirent, filldir_t filldir);
 static struct dentry *autofs4_lookup(struct inode *,struct dentry *, struct nameidata *);
 static void *autofs4_follow_link(struct dentry *, struct nameidata *);
@@ -44,9 +42,9 @@ const struct file_operations autofs4_root_operations = {
 
 const struct file_operations autofs4_dir_operations = {
 	.open		= autofs4_dir_open,
-	.release	= autofs4_dir_close,
+	.release	= dcache_dir_close,
 	.read		= generic_read_dir,
-	.readdir	= autofs4_dir_readdir,
+	.readdir	= dcache_readdir,
 };
 
 const struct inode_operations autofs4_indirect_root_inode_operations = {
@@ -98,17 +96,7 @@ static int autofs4_root_readdir(struct file *file, void *dirent,
 static int autofs4_dir_open(struct inode *inode, struct file *file)
 {
 	struct dentry *dentry = file->f_path.dentry;
-	struct vfsmount *mnt = file->f_path.mnt;
 	struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
-	struct dentry *cursor;
-	int status;
-
-	status = dcache_dir_open(inode, file);
-	if (status)
-		goto out;
-
-	cursor = file->private_data;
-	cursor->d_fsdata = NULL;
 
 	DPRINTK("file=%p dentry=%p %.*s",
 		file, dentry, dentry->d_name.len, dentry->d_name.name);
@@ -116,129 +104,24 @@ static int autofs4_dir_open(struct inode *inode, struct file *file)
 	if (autofs4_oz_mode(sbi))
 		goto out;
 
-	if (autofs4_ispending(dentry)) {
-		DPRINTK("dentry busy");
-		dcache_dir_close(inode, file);
-		status = -EBUSY;
-		goto out;
-	}
-
-	status = -ENOENT;
-	if (!d_mountpoint(dentry) && dentry->d_op && dentry->d_op->d_revalidate) {
-		struct nameidata nd;
-		int empty, ret;
-
-		/* In case there are stale directory dentrys from a failed mount */
-		spin_lock(&dcache_lock);
-		empty = list_empty(&dentry->d_subdirs);
+	/*
+	 * An empty directory in an autofs file system is always a
+	 * mount point. The daemon must have failed to mount this
+	 * during lookup so it doesn't exist. This can happen, for
+	 * example, if user space returns an incorrect status for a
+	 * mount request. Otherwise we're doing a readdir on the
+	 * autofs file system so just let the libfs routines handle
+	 * it.
+	 */
+	spin_lock(&dcache_lock);
+	if (!d_mountpoint(dentry) && __simple_empty(dentry)) {
 		spin_unlock(&dcache_lock);
-
-		if (!empty)
-			d_invalidate(dentry);
-
-		nd.flags = LOOKUP_DIRECTORY;
-		ret = (dentry->d_op->d_revalidate)(dentry, &nd);
-
-		if (ret <= 0) {
-			if (ret < 0)
-				status = ret;
-			dcache_dir_close(inode, file);
-			goto out;
-		}
+		return -ENOENT;
 	}
+	spin_unlock(&dcache_lock);
 
-	if (d_mountpoint(dentry)) {
-		struct file *fp = NULL;
-		struct path fp_path = { .dentry = dentry, .mnt = mnt };
-
-		path_get(&fp_path);
-
-		if (!autofs4_follow_mount(&fp_path.mnt, &fp_path.dentry)) {
-			path_put(&fp_path);
-			dcache_dir_close(inode, file);
-			goto out;
-		}
-
-		fp = dentry_open(fp_path.dentry, fp_path.mnt, file->f_flags);
-		status = PTR_ERR(fp);
-		if (IS_ERR(fp)) {
-			dcache_dir_close(inode, file);
-			goto out;
-		}
-		cursor->d_fsdata = fp;
-	}
-	return 0;
 out:
-	return status;
-}
-
-static int autofs4_dir_close(struct inode *inode, struct file *file)
-{
-	struct dentry *dentry = file->f_path.dentry;
-	struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
-	struct dentry *cursor = file->private_data;
-	int status = 0;
-
-	DPRINTK("file=%p dentry=%p %.*s",
-		file, dentry, dentry->d_name.len, dentry->d_name.name);
-
-	if (autofs4_oz_mode(sbi))
-		goto out;
-
-	if (autofs4_ispending(dentry)) {
-		DPRINTK("dentry busy");
-		status = -EBUSY;
-		goto out;
-	}
-
-	if (d_mountpoint(dentry)) {
-		struct file *fp = cursor->d_fsdata;
-		if (!fp) {
-			status = -ENOENT;
-			goto out;
-		}
-		filp_close(fp, current->files);
-	}
-out:
-	dcache_dir_close(inode, file);
-	return status;
-}
-
-static int autofs4_dir_readdir(struct file *file, void *dirent, filldir_t filldir)
-{
-	struct dentry *dentry = file->f_path.dentry;
-	struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
-	struct dentry *cursor = file->private_data;
-	int status;
-
-	DPRINTK("file=%p dentry=%p %.*s",
-		file, dentry, dentry->d_name.len, dentry->d_name.name);
-
-	if (autofs4_oz_mode(sbi))
-		goto out;
-
-	if (autofs4_ispending(dentry)) {
-		DPRINTK("dentry busy");
-		return -EBUSY;
-	}
-
-	if (d_mountpoint(dentry)) {
-		struct file *fp = cursor->d_fsdata;
-
-		if (!fp)
-			return -ENOENT;
-
-		if (!fp->f_op || !fp->f_op->readdir)
-			goto out;
-
-		status = vfs_readdir(fp, filldir, dirent);
-		file->f_pos = fp->f_pos;
-		if (status)
-			autofs4_copy_atime(file, fp);
-		return status;
-	}
-out:
-	return dcache_readdir(file, dirent, filldir);
+	return dcache_dir_open(inode, file);
 }
 
 static int try_to_fill_dentry(struct dentry *dentry, int flags)

From 26e81b3142f1ba497d4cd0365c13661684b784ce Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Wed, 23 Jul 2008 21:30:25 -0700
Subject: [PATCH 0420/1171] autofs4: fix pending checks

There are two cases for which a dentry that has a pending mount request
does not wait for completion.  One is via autofs4_revalidate() and the
other via autofs4_follow_link().

In revalidate, after the mount point directory is created, but before the
mount is done, the check in try_to_fill_dentry() can can fail to send the
dentry to the wait queue since the dentry is positive and the lookup flags
may contain only LOOKUP_FOLLOW.  Although we don't trigger a mount for the
LOOKUP_FOLLOW flag, if ther's one pending we might as well wait and use
the mounted dentry for the lookup.

In autofs4_follow_link() the dentry is not checked to see if it is pending
so it may fail to call try_to_fill_dentry() and not wait for mount
completion.

A dentry that is pending must always be sent to the wait queue.

Signed-off-by: Ian Kent <raven@themaw.net>
Reviewed-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/root.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c
index 51c873ca8e8d..61d1dca16884 100644
--- a/fs/autofs4/root.c
+++ b/fs/autofs4/root.c
@@ -177,7 +177,8 @@ static int try_to_fill_dentry(struct dentry *dentry, int flags)
 			return status;
 		}
 	/* Trigger mount for path component or follow link */
-	} else if (flags & (TRIGGER_FLAGS | TRIGGER_INTENTS) ||
+	} else if (dentry->d_flags & DCACHE_AUTOFS_PENDING ||
+			flags & (TRIGGER_FLAGS | TRIGGER_INTENTS) ||
 			current->link_count) {
 		DPRINTK("waiting for mount name=%.*s",
 			dentry->d_name.len, dentry->d_name.name);
@@ -223,7 +224,8 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd)
 
 	/* If it's our master or we shouldn't trigger a mount we're done */
 	lookup_type = nd->flags & (TRIGGER_FLAGS | TRIGGER_INTENTS);
-	if (oz_mode || !lookup_type)
+	if (oz_mode ||
+	    !(lookup_type || dentry->d_flags & DCACHE_AUTOFS_PENDING))
 		goto done;
 
 	/* If an expire request is pending wait for it. */
@@ -242,7 +244,8 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd)
 	 * don't try to mount it again.
 	 */
 	spin_lock(&dcache_lock);
-	if (!d_mountpoint(dentry) && __simple_empty(dentry)) {
+	if (dentry->d_flags & DCACHE_AUTOFS_PENDING ||
+	    (!d_mountpoint(dentry) && __simple_empty(dentry))) {
 		spin_unlock(&dcache_lock);
 
 		status = try_to_fill_dentry(dentry, 0);

From 97e7449a7ad883bf9f516fc970778d75999c7843 Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Wed, 23 Jul 2008 21:30:26 -0700
Subject: [PATCH 0421/1171] autofs4: fix indirect mount pending expire race

The selection of a dentry for expiration and the setting of the
AUTOFS_INF_EXPIRING flag isn't done atomically which can lead to lookups
walking into an expiring mount.

What happens is that an expire is initiated by the daemon and a dentry is
selected for expire but, since there is no lock held between the selection
and setting of the expiring flag, a process may find the flag clear and
continue walking into the mount tree at the same time the daemon attempts
the expire it.

Signed-off-by: Ian Kent <raven@themaw.net>
Reviewed-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/autofs_i.h | 10 +++-------
 fs/autofs4/expire.c   | 46 ++++++++++++++++++++++++++++++++-----------
 fs/autofs4/root.c     | 32 +++++++++++++++++++++++++-----
 3 files changed, 65 insertions(+), 23 deletions(-)

diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h
index 058e1800caec..5d90ed3b4b43 100644
--- a/fs/autofs4/autofs_i.h
+++ b/fs/autofs4/autofs_i.h
@@ -138,18 +138,14 @@ static inline int autofs4_oz_mode(struct autofs_sb_info *sbi) {
 static inline int autofs4_ispending(struct dentry *dentry)
 {
 	struct autofs_info *inf = autofs4_dentry_ino(dentry);
-	int pending = 0;
 
 	if (dentry->d_flags & DCACHE_AUTOFS_PENDING)
 		return 1;
 
-	if (inf) {
-		spin_lock(&inf->sbi->fs_lock);
-		pending = inf->flags & AUTOFS_INF_EXPIRING;
-		spin_unlock(&inf->sbi->fs_lock);
-	}
+	if (inf->flags & AUTOFS_INF_EXPIRING)
+		return 1;
 
-	return pending;
+	return 0;
 }
 
 static inline void autofs4_copy_atime(struct file *src, struct file *dst)
diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c
index 894fee54d4d8..19f5bea2704f 100644
--- a/fs/autofs4/expire.c
+++ b/fs/autofs4/expire.c
@@ -292,6 +292,8 @@ static struct dentry *autofs4_expire_indirect(struct super_block *sb,
 	struct list_head *next;
 	int do_now = how & AUTOFS_EXP_IMMEDIATE;
 	int exp_leaves = how & AUTOFS_EXP_LEAVES;
+	struct autofs_info *ino;
+	unsigned int ino_count;
 
 	if (!root)
 		return NULL;
@@ -316,6 +318,9 @@ static struct dentry *autofs4_expire_indirect(struct super_block *sb,
 		dentry = dget(dentry);
 		spin_unlock(&dcache_lock);
 
+		spin_lock(&sbi->fs_lock);
+		ino = autofs4_dentry_ino(dentry);
+
 		/*
 		 * Case 1: (i) indirect mount or top level pseudo direct mount
 		 *	   (autofs-4.1).
@@ -326,6 +331,11 @@ static struct dentry *autofs4_expire_indirect(struct super_block *sb,
 			DPRINTK("checking mountpoint %p %.*s",
 				dentry, (int)dentry->d_name.len, dentry->d_name.name);
 
+			/* Path walk currently on this dentry? */
+			ino_count = atomic_read(&ino->count) + 2;
+			if (atomic_read(&dentry->d_count) > ino_count)
+				goto next;
+
 			/* Can we umount this guy */
 			if (autofs4_mount_busy(mnt, dentry))
 				goto next;
@@ -343,23 +353,25 @@ static struct dentry *autofs4_expire_indirect(struct super_block *sb,
 
 		/* Case 2: tree mount, expire iff entire tree is not busy */
 		if (!exp_leaves) {
-			/* Lock the tree as we must expire as a whole */
-			spin_lock(&sbi->fs_lock);
-			if (!autofs4_tree_busy(mnt, dentry, timeout, do_now)) {
-				struct autofs_info *inf = autofs4_dentry_ino(dentry);
+			/* Path walk currently on this dentry? */
+			ino_count = atomic_read(&ino->count) + 1;
+			if (atomic_read(&dentry->d_count) > ino_count)
+				goto next;
 
-				/* Set this flag early to catch sys_chdir and the like */
-				inf->flags |= AUTOFS_INF_EXPIRING;
-				spin_unlock(&sbi->fs_lock);
+			if (!autofs4_tree_busy(mnt, dentry, timeout, do_now)) {
 				expired = dentry;
 				goto found;
 			}
-			spin_unlock(&sbi->fs_lock);
 		/*
 		 * Case 3: pseudo direct mount, expire individual leaves
 		 *	   (autofs-4.1).
 		 */
 		} else {
+			/* Path walk currently on this dentry? */
+			ino_count = atomic_read(&ino->count) + 1;
+			if (atomic_read(&dentry->d_count) > ino_count)
+				goto next;
+
 			expired = autofs4_check_leaves(mnt, dentry, timeout, do_now);
 			if (expired) {
 				dput(dentry);
@@ -367,6 +379,7 @@ static struct dentry *autofs4_expire_indirect(struct super_block *sb,
 			}
 		}
 next:
+		spin_unlock(&sbi->fs_lock);
 		dput(dentry);
 		spin_lock(&dcache_lock);
 		next = next->next;
@@ -377,6 +390,9 @@ next:
 found:
 	DPRINTK("returning %p %.*s",
 		expired, (int)expired->d_name.len, expired->d_name.name);
+	ino = autofs4_dentry_ino(expired);
+	ino->flags |= AUTOFS_INF_EXPIRING;
+	spin_unlock(&sbi->fs_lock);
 	spin_lock(&dcache_lock);
 	list_move(&expired->d_parent->d_subdirs, &expired->d_u.d_child);
 	spin_unlock(&dcache_lock);
@@ -390,7 +406,9 @@ int autofs4_expire_run(struct super_block *sb,
 		      struct autofs_packet_expire __user *pkt_p)
 {
 	struct autofs_packet_expire pkt;
+	struct autofs_info *ino;
 	struct dentry *dentry;
+	int ret = 0;
 
 	memset(&pkt,0,sizeof pkt);
 
@@ -406,9 +424,14 @@ int autofs4_expire_run(struct super_block *sb,
 	dput(dentry);
 
 	if ( copy_to_user(pkt_p, &pkt, sizeof(struct autofs_packet_expire)) )
-		return -EFAULT;
+		ret = -EFAULT;
 
-	return 0;
+	spin_lock(&sbi->fs_lock);
+	ino = autofs4_dentry_ino(dentry);
+	ino->flags &= ~AUTOFS_INF_EXPIRING;
+	spin_unlock(&sbi->fs_lock);
+
+	return ret;
 }
 
 /* Call repeatedly until it returns -EAGAIN, meaning there's nothing
@@ -433,9 +456,10 @@ int autofs4_expire_multi(struct super_block *sb, struct vfsmount *mnt,
 
 		/* This is synchronous because it makes the daemon a
                    little easier */
-		ino->flags |= AUTOFS_INF_EXPIRING;
 		ret = autofs4_wait(sbi, dentry, NFY_EXPIRE);
+		spin_lock(&sbi->fs_lock);
 		ino->flags &= ~AUTOFS_INF_EXPIRING;
+		spin_unlock(&sbi->fs_lock);
 		dput(dentry);
 	}
 
diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c
index 61d1dca16884..1c2579de1f2e 100644
--- a/fs/autofs4/root.c
+++ b/fs/autofs4/root.c
@@ -133,7 +133,10 @@ static int try_to_fill_dentry(struct dentry *dentry, int flags)
 	/* Block on any pending expiry here; invalidate the dentry
            when expiration is done to trigger mount request with a new
            dentry */
-	if (ino && (ino->flags & AUTOFS_INF_EXPIRING)) {
+	spin_lock(&sbi->fs_lock);
+	if (ino->flags & AUTOFS_INF_EXPIRING) {
+		spin_unlock(&sbi->fs_lock);
+
 		DPRINTK("waiting for expire %p name=%.*s",
 			 dentry, dentry->d_name.len, dentry->d_name.name);
 
@@ -149,8 +152,11 @@ static int try_to_fill_dentry(struct dentry *dentry, int flags)
 		status = d_invalidate(dentry);
 		if (status != -EBUSY)
 			return -EAGAIN;
-	}
 
+		goto cont;
+	}
+	spin_unlock(&sbi->fs_lock);
+cont:
 	DPRINTK("dentry=%p %.*s ino=%p",
 		 dentry, dentry->d_name.len, dentry->d_name.name, dentry->d_inode);
 
@@ -229,15 +235,21 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd)
 		goto done;
 
 	/* If an expire request is pending wait for it. */
-	if (ino && (ino->flags & AUTOFS_INF_EXPIRING)) {
+	spin_lock(&sbi->fs_lock);
+	if (ino->flags & AUTOFS_INF_EXPIRING) {
+		spin_unlock(&sbi->fs_lock);
+
 		DPRINTK("waiting for active request %p name=%.*s",
 			dentry, dentry->d_name.len, dentry->d_name.name);
 
 		status = autofs4_wait(sbi, dentry, NFY_NONE);
 
 		DPRINTK("request done status=%d", status);
-	}
 
+		goto cont;
+	}
+	spin_unlock(&sbi->fs_lock);
+cont:
 	/*
 	 * If the dentry contains directories then it is an
 	 * autofs multi-mount with no root mount offset. So
@@ -292,8 +304,11 @@ static int autofs4_revalidate(struct dentry *dentry, struct nameidata *nd)
 	int status = 1;
 
 	/* Pending dentry */
+	spin_lock(&sbi->fs_lock);
 	if (autofs4_ispending(dentry)) {
 		/* The daemon never causes a mount to trigger */
+		spin_unlock(&sbi->fs_lock);
+
 		if (oz_mode)
 			return 1;
 
@@ -316,6 +331,7 @@ static int autofs4_revalidate(struct dentry *dentry, struct nameidata *nd)
 
 		return status;
 	}
+	spin_unlock(&sbi->fs_lock);
 
 	/* Negative dentry.. invalidate if "old" */
 	if (dentry->d_inode == NULL)
@@ -329,6 +345,7 @@ static int autofs4_revalidate(struct dentry *dentry, struct nameidata *nd)
 		DPRINTK("dentry=%p %.*s, emptydir",
 			 dentry, dentry->d_name.len, dentry->d_name.name);
 		spin_unlock(&dcache_lock);
+
 		/* The daemon never causes a mount to trigger */
 		if (oz_mode)
 			return 1;
@@ -521,13 +538,18 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s
 		 * so it must have been successful, so just wait for it.
 		 */
 		ino = autofs4_dentry_ino(expiring);
-		while (ino && (ino->flags & AUTOFS_INF_EXPIRING)) {
+		spin_lock(&sbi->fs_lock);
+		if (ino->flags & AUTOFS_INF_EXPIRING) {
+			spin_unlock(&sbi->fs_lock);
 			DPRINTK("wait for incomplete expire %p name=%.*s",
 				expiring, expiring->d_name.len,
 				expiring->d_name.name);
 			autofs4_wait(sbi, expiring, NFY_NONE);
 			DPRINTK("request completed");
+			goto cont;
 		}
+		spin_unlock(&sbi->fs_lock);
+cont:
 		spin_lock(&sbi->lookup_lock);
 		if (!list_empty(&ino->expiring))
 			list_del_init(&ino->expiring);

From 6e60a9ab5f5d314735467752f623072f5b75157a Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Wed, 23 Jul 2008 21:30:27 -0700
Subject: [PATCH 0422/1171] autofs4: fix direct mount pending expire race

For direct and offset type mounts that are covered by another mount we
cannot check the AUTOFS_INF_EXPIRING flag during a path walk which leads
to lookups walking into an expiring mount while it is being expired.

For example, for the direct multi-mount map entry with a couple of
offsets:

/race/mm1  /      <server1>:/<path1>
           /om1   <server2>:/<path2>
           /om2   <server1>:/<path3>

an autofs trigger mount is mounted on /race/mm1 and when accessed it is
over mounted and trigger mounts made for /race/mm1/om1 and /race/mm1/om2.
So it isn't possible for path walks to see the expiring flag at all and
they happily walk into the file system while it is expiring.

When expiring these mounts follow_down() must stop at the autofs mount and
all processes must block in the ->follow_link() method (except the daemon)
until the expire is complete.  This is done by decrementing the d_mounted
field of the autofs trigger mount root dentry until the expire is
completed.  In ->follow_link() all processes wait on the expire and the
mount following is completed for the daemon until the expire is complete.

Signed-off-by: Ian Kent <raven@themaw.net>
Cc: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/autofs_i.h |  3 ++
 fs/autofs4/expire.c   | 16 ++++++++--
 fs/autofs4/root.c     | 72 +++++++++++++++++++++++++++++--------------
 3 files changed, 65 insertions(+), 26 deletions(-)

diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h
index 5d90ed3b4b43..4b40cbc71e9b 100644
--- a/fs/autofs4/autofs_i.h
+++ b/fs/autofs4/autofs_i.h
@@ -52,6 +52,8 @@ struct autofs_info {
 
 	int		flags;
 
+	struct completion expire_complete;
+
 	struct list_head active;
 	struct list_head expiring;
 
@@ -69,6 +71,7 @@ struct autofs_info {
 };
 
 #define AUTOFS_INF_EXPIRING	(1<<0) /* dentry is in the process of expiring */
+#define AUTOFS_INF_MOUNTPOINT	(1<<1) /* mountpoint status for direct expire */
 
 struct autofs_wait_queue {
 	wait_queue_head_t queue;
diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c
index 19f5bea2704f..705b9f057fb3 100644
--- a/fs/autofs4/expire.c
+++ b/fs/autofs4/expire.c
@@ -259,13 +259,15 @@ static struct dentry *autofs4_expire_direct(struct super_block *sb,
 	now = jiffies;
 	timeout = sbi->exp_timeout;
 
-	/* Lock the tree as we must expire as a whole */
 	spin_lock(&sbi->fs_lock);
 	if (!autofs4_direct_busy(mnt, root, timeout, do_now)) {
 		struct autofs_info *ino = autofs4_dentry_ino(root);
-
-		/* Set this flag early to catch sys_chdir and the like */
+		if (d_mountpoint(root)) {
+			ino->flags |= AUTOFS_INF_MOUNTPOINT;
+			root->d_mounted--;
+		}
 		ino->flags |= AUTOFS_INF_EXPIRING;
+		init_completion(&ino->expire_complete);
 		spin_unlock(&sbi->fs_lock);
 		return root;
 	}
@@ -392,6 +394,7 @@ found:
 		expired, (int)expired->d_name.len, expired->d_name.name);
 	ino = autofs4_dentry_ino(expired);
 	ino->flags |= AUTOFS_INF_EXPIRING;
+	init_completion(&ino->expire_complete);
 	spin_unlock(&sbi->fs_lock);
 	spin_lock(&dcache_lock);
 	list_move(&expired->d_parent->d_subdirs, &expired->d_u.d_child);
@@ -429,6 +432,7 @@ int autofs4_expire_run(struct super_block *sb,
 	spin_lock(&sbi->fs_lock);
 	ino = autofs4_dentry_ino(dentry);
 	ino->flags &= ~AUTOFS_INF_EXPIRING;
+	complete_all(&ino->expire_complete);
 	spin_unlock(&sbi->fs_lock);
 
 	return ret;
@@ -457,8 +461,14 @@ int autofs4_expire_multi(struct super_block *sb, struct vfsmount *mnt,
 		/* This is synchronous because it makes the daemon a
                    little easier */
 		ret = autofs4_wait(sbi, dentry, NFY_EXPIRE);
+
 		spin_lock(&sbi->fs_lock);
+		if (ino->flags & AUTOFS_INF_MOUNTPOINT) {
+			sb->s_root->d_mounted++;
+			ino->flags &= ~AUTOFS_INF_MOUNTPOINT;
+		}
 		ino->flags &= ~AUTOFS_INF_EXPIRING;
+		complete_all(&ino->expire_complete);
 		spin_unlock(&sbi->fs_lock);
 		dput(dentry);
 	}
diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c
index 1c2579de1f2e..adbd8559e870 100644
--- a/fs/autofs4/root.c
+++ b/fs/autofs4/root.c
@@ -141,6 +141,7 @@ static int try_to_fill_dentry(struct dentry *dentry, int flags)
 			 dentry, dentry->d_name.len, dentry->d_name.name);
 
 		status = autofs4_wait(sbi, dentry, NFY_NONE);
+		wait_for_completion(&ino->expire_complete);
 
 		DPRINTK("expire done status=%d", status);
 
@@ -227,14 +228,32 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd)
 	DPRINTK("dentry=%p %.*s oz_mode=%d nd->flags=%d",
 		dentry, dentry->d_name.len, dentry->d_name.name, oz_mode,
 		nd->flags);
-
-	/* If it's our master or we shouldn't trigger a mount we're done */
-	lookup_type = nd->flags & (TRIGGER_FLAGS | TRIGGER_INTENTS);
-	if (oz_mode ||
-	    !(lookup_type || dentry->d_flags & DCACHE_AUTOFS_PENDING))
+	/*
+	 * For an expire of a covered direct or offset mount we need
+	 * to beeak out of follow_down() at the autofs mount trigger
+	 * (d_mounted--), so we can see the expiring flag, and manage
+	 * the blocking and following here until the expire is completed.
+	 */
+	if (oz_mode) {
+		spin_lock(&sbi->fs_lock);
+		if (ino->flags & AUTOFS_INF_EXPIRING) {
+			spin_unlock(&sbi->fs_lock);
+			/* Follow down to our covering mount. */
+			if (!follow_down(&nd->path.mnt, &nd->path.dentry))
+				goto done;
+			/*
+			 * We shouldn't need to do this but we have no way
+			 * of knowing what may have been done so try a follow
+			 * just in case.
+			 */
+			autofs4_follow_mount(&nd->path.mnt, &nd->path.dentry);
+			goto done;
+		}
+		spin_unlock(&sbi->fs_lock);
 		goto done;
+	}
 
-	/* If an expire request is pending wait for it. */
+	/* If an expire request is pending everyone must wait. */
 	spin_lock(&sbi->fs_lock);
 	if (ino->flags & AUTOFS_INF_EXPIRING) {
 		spin_unlock(&sbi->fs_lock);
@@ -243,6 +262,7 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd)
 			dentry, dentry->d_name.len, dentry->d_name.name);
 
 		status = autofs4_wait(sbi, dentry, NFY_NONE);
+		wait_for_completion(&ino->expire_complete);
 
 		DPRINTK("request done status=%d", status);
 
@@ -250,10 +270,15 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd)
 	}
 	spin_unlock(&sbi->fs_lock);
 cont:
+	/* We trigger a mount for almost all flags */
+	lookup_type = nd->flags & (TRIGGER_FLAGS | TRIGGER_INTENTS);
+	if (!(lookup_type || dentry->d_flags & DCACHE_AUTOFS_PENDING))
+		goto done;
+
 	/*
-	 * If the dentry contains directories then it is an
-	 * autofs multi-mount with no root mount offset. So
-	 * don't try to mount it again.
+	 * If the dentry contains directories then it is an autofs
+	 * multi-mount with no root mount offset. So don't try to
+	 * mount it again.
 	 */
 	spin_lock(&dcache_lock);
 	if (dentry->d_flags & DCACHE_AUTOFS_PENDING ||
@@ -264,22 +289,22 @@ cont:
 		if (status)
 			goto out_error;
 
-		/*
-		 * The mount succeeded but if there is no root mount
-		 * it must be an autofs multi-mount with no root offset
-		 * so we don't need to follow the mount.
-		 */
-		if (d_mountpoint(dentry)) {
-			if (!autofs4_follow_mount(&nd->path.mnt,
-						  &nd->path.dentry)) {
-				status = -ENOENT;
-				goto out_error;
-			}
-		}
-
-		goto done;
+		goto follow;
 	}
 	spin_unlock(&dcache_lock);
+follow:
+	/*
+	 * If there is no root mount it must be an autofs
+	 * multi-mount with no root offset so we don't need
+	 * to follow it.
+	 */
+	if (d_mountpoint(dentry)) {
+		if (!autofs4_follow_mount(&nd->path.mnt,
+					  &nd->path.dentry)) {
+			status = -ENOENT;
+			goto out_error;
+		}
+	}
 
 done:
 	return NULL;
@@ -545,6 +570,7 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s
 				expiring, expiring->d_name.len,
 				expiring->d_name.name);
 			autofs4_wait(sbi, expiring, NFY_NONE);
+			wait_for_completion(&ino->expire_complete);
 			DPRINTK("request completed");
 			goto cont;
 		}

From ec6e8c7d3f9073336ec7b2eed3fcda6f922087c3 Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Wed, 23 Jul 2008 21:30:28 -0700
Subject: [PATCH 0423/1171] autofs4: fix direct mount pending expire race -
 correction

Appologies, somehow I seem to have sent an out dated version of this
patch. Here is an additional patch that brings the patch up to date.

Signed-off-by: Ian Kent <raven@themaw.net>
Cc: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/root.c | 10 ++--------
 1 file changed, 2 insertions(+), 8 deletions(-)

diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c
index adbd8559e870..e062ee5a3ed5 100644
--- a/fs/autofs4/root.c
+++ b/fs/autofs4/root.c
@@ -241,13 +241,7 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd)
 			/* Follow down to our covering mount. */
 			if (!follow_down(&nd->path.mnt, &nd->path.dentry))
 				goto done;
-			/*
-			 * We shouldn't need to do this but we have no way
-			 * of knowing what may have been done so try a follow
-			 * just in case.
-			 */
-			autofs4_follow_mount(&nd->path.mnt, &nd->path.dentry);
-			goto done;
+			goto follow;
 		}
 		spin_unlock(&sbi->fs_lock);
 		goto done;
@@ -273,7 +267,7 @@ cont:
 	/* We trigger a mount for almost all flags */
 	lookup_type = nd->flags & (TRIGGER_FLAGS | TRIGGER_INTENTS);
 	if (!(lookup_type || dentry->d_flags & DCACHE_AUTOFS_PENDING))
-		goto done;
+		goto follow;
 
 	/*
 	 * If the dentry contains directories then it is an autofs

From 06a3598552dc3b2b30eb18bd53bbac2a901489d7 Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Wed, 23 Jul 2008 21:30:28 -0700
Subject: [PATCH 0424/1171] autofs4: reorganize expire pending wait function
 calls

This patch re-orgnirzes the checking for and waiting on active expires and
elininates redundant checks.

Signed-off-by: Ian Kent <raven@themaw.net>
Cc: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/autofs_i.h |  1 +
 fs/autofs4/expire.c   | 29 +++++++++++++++++
 fs/autofs4/root.c     | 75 ++++++-------------------------------------
 3 files changed, 40 insertions(+), 65 deletions(-)

diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h
index 4b40cbc71e9b..69a2f5c92319 100644
--- a/fs/autofs4/autofs_i.h
+++ b/fs/autofs4/autofs_i.h
@@ -163,6 +163,7 @@ void autofs4_free_ino(struct autofs_info *);
 
 /* Expiration */
 int is_autofs4_dentry(struct dentry *);
+int autofs4_expire_wait(struct dentry *dentry);
 int autofs4_expire_run(struct super_block *, struct vfsmount *,
 			struct autofs_sb_info *,
 			struct autofs_packet_expire __user *);
diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c
index 705b9f057fb3..cdabb796ff01 100644
--- a/fs/autofs4/expire.c
+++ b/fs/autofs4/expire.c
@@ -402,6 +402,35 @@ found:
 	return expired;
 }
 
+int autofs4_expire_wait(struct dentry *dentry)
+{
+	struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb);
+	struct autofs_info *ino = autofs4_dentry_ino(dentry);
+	int status;
+
+	/* Block on any pending expire */
+	spin_lock(&sbi->fs_lock);
+	if (ino->flags & AUTOFS_INF_EXPIRING) {
+		spin_unlock(&sbi->fs_lock);
+
+		DPRINTK("waiting for expire %p name=%.*s",
+			 dentry, dentry->d_name.len, dentry->d_name.name);
+
+		status = autofs4_wait(sbi, dentry, NFY_NONE);
+		wait_for_completion(&ino->expire_complete);
+
+		DPRINTK("expire done status=%d", status);
+
+		if (d_unhashed(dentry))
+			return -EAGAIN;
+
+		return status;
+	}
+	spin_unlock(&sbi->fs_lock);
+
+	return 0;
+}
+
 /* Perform an expiry operation */
 int autofs4_expire_run(struct super_block *sb,
 		      struct vfsmount *mnt,
diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c
index e062ee5a3ed5..ae22bde0bbd7 100644
--- a/fs/autofs4/root.c
+++ b/fs/autofs4/root.c
@@ -130,34 +130,6 @@ static int try_to_fill_dentry(struct dentry *dentry, int flags)
 	struct autofs_info *ino = autofs4_dentry_ino(dentry);
 	int status;
 
-	/* Block on any pending expiry here; invalidate the dentry
-           when expiration is done to trigger mount request with a new
-           dentry */
-	spin_lock(&sbi->fs_lock);
-	if (ino->flags & AUTOFS_INF_EXPIRING) {
-		spin_unlock(&sbi->fs_lock);
-
-		DPRINTK("waiting for expire %p name=%.*s",
-			 dentry, dentry->d_name.len, dentry->d_name.name);
-
-		status = autofs4_wait(sbi, dentry, NFY_NONE);
-		wait_for_completion(&ino->expire_complete);
-
-		DPRINTK("expire done status=%d", status);
-
-		/*
-		 * If the directory still exists the mount request must
-		 * continue otherwise it can't be followed at the right
-		 * time during the walk.
-		 */
-		status = d_invalidate(dentry);
-		if (status != -EBUSY)
-			return -EAGAIN;
-
-		goto cont;
-	}
-	spin_unlock(&sbi->fs_lock);
-cont:
 	DPRINTK("dentry=%p %.*s ino=%p",
 		 dentry, dentry->d_name.len, dentry->d_name.name, dentry->d_inode);
 
@@ -248,22 +220,8 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd)
 	}
 
 	/* If an expire request is pending everyone must wait. */
-	spin_lock(&sbi->fs_lock);
-	if (ino->flags & AUTOFS_INF_EXPIRING) {
-		spin_unlock(&sbi->fs_lock);
+	autofs4_expire_wait(dentry);
 
-		DPRINTK("waiting for active request %p name=%.*s",
-			dentry, dentry->d_name.len, dentry->d_name.name);
-
-		status = autofs4_wait(sbi, dentry, NFY_NONE);
-		wait_for_completion(&ino->expire_complete);
-
-		DPRINTK("request done status=%d", status);
-
-		goto cont;
-	}
-	spin_unlock(&sbi->fs_lock);
-cont:
 	/* We trigger a mount for almost all flags */
 	lookup_type = nd->flags & (TRIGGER_FLAGS | TRIGGER_INTENTS);
 	if (!(lookup_type || dentry->d_flags & DCACHE_AUTOFS_PENDING))
@@ -331,6 +289,14 @@ static int autofs4_revalidate(struct dentry *dentry, struct nameidata *nd)
 		if (oz_mode)
 			return 1;
 
+		/*
+		 * If the directory has gone away due to an expire
+		 * we have been called as ->d_revalidate() and so
+		 * we need to return false and proceed to ->lookup().
+		 */
+		if (autofs4_expire_wait(dentry) == -EAGAIN)
+			return 0;
+
 		/*
 		 * A zero status is success otherwise we have a
 		 * negative error code.
@@ -339,15 +305,6 @@ static int autofs4_revalidate(struct dentry *dentry, struct nameidata *nd)
 		if (status == 0)
 			return 1;
 
-		/*
-		 * A status of EAGAIN here means that the dentry has gone
-		 * away while waiting for an expire to complete. If we are
-		 * racing with expire lookup will wait for it so this must
-		 * be a revalidate and we need to send it to lookup.
-		 */
-		if (status == -EAGAIN)
-			return 0;
-
 		return status;
 	}
 	spin_unlock(&sbi->fs_lock);
@@ -557,19 +514,7 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s
 		 * so it must have been successful, so just wait for it.
 		 */
 		ino = autofs4_dentry_ino(expiring);
-		spin_lock(&sbi->fs_lock);
-		if (ino->flags & AUTOFS_INF_EXPIRING) {
-			spin_unlock(&sbi->fs_lock);
-			DPRINTK("wait for incomplete expire %p name=%.*s",
-				expiring, expiring->d_name.len,
-				expiring->d_name.name);
-			autofs4_wait(sbi, expiring, NFY_NONE);
-			wait_for_completion(&ino->expire_complete);
-			DPRINTK("request completed");
-			goto cont;
-		}
-		spin_unlock(&sbi->fs_lock);
-cont:
+		autofs4_expire_wait(expiring);
 		spin_lock(&sbi->lookup_lock);
 		if (!list_empty(&ino->expiring))
 			list_del_init(&ino->expiring);

From aa55ddf340c9fa3f303ee16bbf35887e42c50304 Mon Sep 17 00:00:00 2001
From: Ian Kent <raven@themaw.net>
Date: Wed, 23 Jul 2008 21:30:29 -0700
Subject: [PATCH 0425/1171] autofs4: remove unused ioctls

The ioctls AUTOFS_IOC_TOGGLEREGHOST and AUTOFS_IOC_ASKREGHOST were added
several years ago but what they were intended for has never been
implemented (as far as I'm aware noone uses them) so remove them.

Signed-off-by: Ian Kent <raven@themaw.net>
Reviewed-by: Jeff Moyer <jmoyer@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/autofs4/root.c        | 68 +---------------------------------------
 fs/compat_ioctl.c        |  2 --
 include/linux/auto_fs4.h |  2 --
 3 files changed, 1 insertion(+), 71 deletions(-)

diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c
index ae22bde0bbd7..bcfb2dc0a61b 100644
--- a/fs/autofs4/root.c
+++ b/fs/autofs4/root.c
@@ -25,7 +25,6 @@ static int autofs4_dir_rmdir(struct inode *,struct dentry *);
 static int autofs4_dir_mkdir(struct inode *,struct dentry *,int);
 static int autofs4_root_ioctl(struct inode *, struct file *,unsigned int,unsigned long);
 static int autofs4_dir_open(struct inode *inode, struct file *file);
-static int autofs4_root_readdir(struct file * filp, void * dirent, filldir_t filldir);
 static struct dentry *autofs4_lookup(struct inode *,struct dentry *, struct nameidata *);
 static void *autofs4_follow_link(struct dentry *, struct nameidata *);
 
@@ -36,7 +35,7 @@ const struct file_operations autofs4_root_operations = {
 	.open		= dcache_dir_open,
 	.release	= dcache_dir_close,
 	.read		= generic_read_dir,
-	.readdir	= autofs4_root_readdir,
+	.readdir	= dcache_readdir,
 	.ioctl		= autofs4_root_ioctl,
 };
 
@@ -71,28 +70,6 @@ const struct inode_operations autofs4_dir_inode_operations = {
 	.rmdir		= autofs4_dir_rmdir,
 };
 
-static int autofs4_root_readdir(struct file *file, void *dirent,
-				filldir_t filldir)
-{
-	struct autofs_sb_info *sbi = autofs4_sbi(file->f_path.dentry->d_sb);
-	int oz_mode = autofs4_oz_mode(sbi);
-
-	DPRINTK("called, filp->f_pos = %lld", file->f_pos);
-
-	/*
-	 * Don't set reghost flag if:
-	 * 1) f_pos is larger than zero -- we've already been here.
-	 * 2) we haven't even enabled reghosting in the 1st place.
-	 * 3) this is the daemon doing a readdir
-	 */
-	if (oz_mode && file->f_pos == 0 && sbi->reghost_enabled)
-		sbi->needs_reghost = 1;
-
-	DPRINTK("needs_reghost = %d", sbi->needs_reghost);
-
-	return dcache_readdir(file, dirent, filldir);
-}
-
 static int autofs4_dir_open(struct inode *inode, struct file *file)
 {
 	struct dentry *dentry = file->f_path.dentry;
@@ -858,44 +835,6 @@ static inline int autofs4_get_protosubver(struct autofs_sb_info *sbi, int __user
 	return put_user(sbi->sub_version, p);
 }
 
-/*
- * Tells the daemon whether we need to reghost or not. Also, clears
- * the reghost_needed flag.
- */
-static inline int autofs4_ask_reghost(struct autofs_sb_info *sbi, int __user *p)
-{
-	int status;
-
-	DPRINTK("returning %d", sbi->needs_reghost);
-
-	status = put_user(sbi->needs_reghost, p);
-	if (status)
-		return status;
-
-	sbi->needs_reghost = 0;
-	return 0;
-}
-
-/*
- * Enable / Disable reghosting ioctl() operation
- */
-static inline int autofs4_toggle_reghost(struct autofs_sb_info *sbi, int __user *p)
-{
-	int status;
-	int val;
-
-	status = get_user(val, p);
-
-	DPRINTK("reghost = %d", val);
-
-	if (status)
-		return status;
-
-	/* turn on/off reghosting, with the val */
-	sbi->reghost_enabled = val;
-	return 0;
-}
-
 /*
 * Tells the daemon whether it can umount the autofs mount.
 */
@@ -960,11 +899,6 @@ static int autofs4_root_ioctl(struct inode *inode, struct file *filp,
 	case AUTOFS_IOC_SETTIMEOUT:
 		return autofs4_get_set_timeout(sbi, p);
 
-	case AUTOFS_IOC_TOGGLEREGHOST:
-		return autofs4_toggle_reghost(sbi, p);
-	case AUTOFS_IOC_ASKREGHOST:
-		return autofs4_ask_reghost(sbi, p);
-
 	case AUTOFS_IOC_ASKUMOUNT:
 		return autofs4_ask_umount(filp->f_path.mnt, p);
 
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index 7b3a03c7c6a9..18e2c548161d 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -2297,8 +2297,6 @@ COMPATIBLE_IOCTL(AUTOFS_IOC_PROTOVER)
 COMPATIBLE_IOCTL(AUTOFS_IOC_EXPIRE)
 COMPATIBLE_IOCTL(AUTOFS_IOC_EXPIRE_MULTI)
 COMPATIBLE_IOCTL(AUTOFS_IOC_PROTOSUBVER)
-COMPATIBLE_IOCTL(AUTOFS_IOC_ASKREGHOST)
-COMPATIBLE_IOCTL(AUTOFS_IOC_TOGGLEREGHOST)
 COMPATIBLE_IOCTL(AUTOFS_IOC_ASKUMOUNT)
 /* Raw devices */
 COMPATIBLE_IOCTL(RAW_SETBIND)
diff --git a/include/linux/auto_fs4.h b/include/linux/auto_fs4.h
index 31a29541b504..b785c6f8644d 100644
--- a/include/linux/auto_fs4.h
+++ b/include/linux/auto_fs4.h
@@ -98,8 +98,6 @@ union autofs_v5_packet_union {
 #define AUTOFS_IOC_EXPIRE_INDIRECT	AUTOFS_IOC_EXPIRE_MULTI
 #define AUTOFS_IOC_EXPIRE_DIRECT	AUTOFS_IOC_EXPIRE_MULTI
 #define AUTOFS_IOC_PROTOSUBVER		_IOR(0x93,0x67,int)
-#define AUTOFS_IOC_ASKREGHOST           _IOR(0x93,0x68,int)
-#define AUTOFS_IOC_TOGGLEREGHOST        _IOR(0x93,0x69,int)
 #define AUTOFS_IOC_ASKUMOUNT		_IOR(0x93,0x70,int)
 
 

From 35aa64f3a117a16c466f688f52ac3847b3b572e8 Mon Sep 17 00:00:00 2001
From: "Maciej W. Rozycki" <macro@linux-mips.org>
Date: Wed, 23 Jul 2008 21:30:29 -0700
Subject: [PATCH 0426/1171] rtc: m41t80: sort header inclusions for readability

Sort the header inclusions for readability.  No functional changes.

Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org>
Cc: Alessandro Zummo <a.zummo@towertech.it>
Cc: Alexander Bigga <ab@mycable.de>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/rtc/rtc-m41t80.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/rtc/rtc-m41t80.c b/drivers/rtc/rtc-m41t80.c
index 0a19c06019be..4b260664547d 100644
--- a/drivers/rtc/rtc-m41t80.c
+++ b/drivers/rtc/rtc-m41t80.c
@@ -13,21 +13,21 @@
  *
  */
 
-#include <linux/module.h>
+#include <linux/bcd.h>
+#include <linux/i2c.h>
 #include <linux/init.h>
 #include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/rtc.h>
 #include <linux/slab.h>
 #include <linux/smp_lock.h>
 #include <linux/string.h>
-#include <linux/i2c.h>
-#include <linux/rtc.h>
-#include <linux/bcd.h>
 #ifdef CONFIG_RTC_DRV_M41T80_WDT
-#include <linux/miscdevice.h>
-#include <linux/watchdog.h>
-#include <linux/reboot.h>
 #include <linux/fs.h>
 #include <linux/ioctl.h>
+#include <linux/miscdevice.h>
+#include <linux/reboot.h>
+#include <linux/watchdog.h>
 #endif
 
 #define M41T80_REG_SSEC	0

From 4c228db0b30fa12d65ae7461ce29ed1f4da12c5b Mon Sep 17 00:00:00 2001
From: "Maciej W. Rozycki" <macro@linux-mips.org>
Date: Wed, 23 Jul 2008 21:30:32 -0700
Subject: [PATCH 0427/1171] rtc: m41t80: use pr_info() as appropriate

Replace printk(KERN_INFO ...) calls with appropriate pr_info(...)
equivalents.

Signed-off-by: Maciej W. Rozycki <macro@linux-mips.org>
Cc: Alessandro Zummo <a.zummo@towertech.it>
Cc: Alexander Bigga <ab@mycable.de>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/rtc/rtc-m41t80.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/rtc/rtc-m41t80.c b/drivers/rtc/rtc-m41t80.c
index 4b260664547d..24bc1689fc74 100644
--- a/drivers/rtc/rtc-m41t80.c
+++ b/drivers/rtc/rtc-m41t80.c
@@ -631,14 +631,12 @@ static int wdt_ioctl(struct inode *inode, struct file *file, unsigned int cmd,
 			return -EFAULT;
 
 		if (rv & WDIOS_DISABLECARD) {
-			printk(KERN_INFO
-			       "rtc-m41t80: disable watchdog\n");
+			pr_info("rtc-m41t80: disable watchdog\n");
 			wdt_disable();
 		}
 
 		if (rv & WDIOS_ENABLECARD) {
-			printk(KERN_INFO
-			       "rtc-m41t80: enable watchdog\n");
+			pr_info("rtc-m41t80: enable watchdog\n");
 			wdt_ping();
 		}
 

From 53f1b1433da7eac2607a4a0898a221a4485fd732 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Wed, 23 Jul 2008 21:30:32 -0700
Subject: [PATCH 0428/1171] rtc: push the BKL down into the driver ioctl method

For now just wrap the main logic, but this driver is a prime candidate for
someone wanting to eliminate the lock entirely

[lizf@cn.fujitsu.com: fix build failure]
Signed-off-by: Alan Cox <alan@redhat.com>
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/rtc.c | 17 ++++++++++-------
 1 file changed, 10 insertions(+), 7 deletions(-)

diff --git a/drivers/char/rtc.c b/drivers/char/rtc.c
index fa92a8af5a5a..d1569a0d0506 100644
--- a/drivers/char/rtc.c
+++ b/drivers/char/rtc.c
@@ -78,9 +78,10 @@
 #include <linux/wait.h>
 #include <linux/bcd.h>
 #include <linux/delay.h>
+#include <linux/smp_lock.h>
+#include <linux/uaccess.h>
 
 #include <asm/current.h>
-#include <asm/uaccess.h>
 #include <asm/system.h>
 
 #ifdef CONFIG_X86
@@ -144,8 +145,7 @@ static DEFINE_TIMER(rtc_irq_timer, rtc_dropped_irq, 0, 0);
 static ssize_t rtc_read(struct file *file, char __user *buf,
 			size_t count, loff_t *ppos);
 
-static int rtc_ioctl(struct inode *inode, struct file *file,
-		     unsigned int cmd, unsigned long arg);
+static long rtc_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
 
 #ifdef RTC_IRQ
 static unsigned int rtc_poll(struct file *file, poll_table *wait);
@@ -719,10 +719,13 @@ static int rtc_do_ioctl(unsigned int cmd, unsigned long arg, int kernel)
 			    &wtime, sizeof wtime) ? -EFAULT : 0;
 }
 
-static int rtc_ioctl(struct inode *inode, struct file *file, unsigned int cmd,
-		     unsigned long arg)
+static long rtc_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 {
-	return rtc_do_ioctl(cmd, arg, 0);
+	long ret;
+	lock_kernel();
+	ret = rtc_do_ioctl(cmd, arg, 0);
+	unlock_kernel();
+	return ret;
 }
 
 /*
@@ -915,7 +918,7 @@ static const struct file_operations rtc_fops = {
 #ifdef RTC_IRQ
 	.poll		= rtc_poll,
 #endif
-	.ioctl		= rtc_ioctl,
+	.unlocked_ioctl	= rtc_ioctl,
 	.open		= rtc_open,
 	.release	= rtc_release,
 	.fasync		= rtc_fasync,

From 5ad31a575157147b43fa84ef1e21471661653878 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Wed, 23 Jul 2008 21:30:33 -0700
Subject: [PATCH 0429/1171] rtc: remove BKL for ioctl()

Remove implicit use of BKL in ioctl() from the RTC framework.

Instead, the rtc->ops_lock is used.  That's the same lock that already
protects the RTC operations when they're issued through the exported
rtc_*() calls in drivers/rtc/interface.c ...  making this a bugfix, not
just a cleanup, since both ioctl calls and set_alarm() need to update IRQ
enable flags and that implies a common lock (which RTC drivers as a rule
do not provide on their own).

A new comment at the declaration of "struct rtc_class_ops" summarizes
current locking rules.  It's not clear to me that the exceptions listed
there should exist ...  if not, those are pre-existing problems which can
be fixed in a patch that doesn't relate to BKL removal.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Jonathan Corbet <corbet@lwn.net>
Acked-by: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/rtc/rtc-dev.c | 58 ++++++++++++++++++++++++++++---------------
 include/linux/rtc.h   | 17 +++++++++++++
 2 files changed, 55 insertions(+), 20 deletions(-)

diff --git a/drivers/rtc/rtc-dev.c b/drivers/rtc/rtc-dev.c
index 0114a78b7cbb..0a870b7e5c32 100644
--- a/drivers/rtc/rtc-dev.c
+++ b/drivers/rtc/rtc-dev.c
@@ -209,7 +209,7 @@ static unsigned int rtc_dev_poll(struct file *file, poll_table *wait)
 	return (data != 0) ? (POLLIN | POLLRDNORM) : 0;
 }
 
-static int rtc_dev_ioctl(struct inode *inode, struct file *file,
+static long rtc_dev_ioctl(struct file *file,
 		unsigned int cmd, unsigned long arg)
 {
 	int err = 0;
@@ -219,6 +219,10 @@ static int rtc_dev_ioctl(struct inode *inode, struct file *file,
 	struct rtc_wkalrm alarm;
 	void __user *uarg = (void __user *) arg;
 
+	err = mutex_lock_interruptible(&rtc->ops_lock);
+	if (err)
+		return -EBUSY;
+
 	/* check that the calling task has appropriate permissions
 	 * for certain ioctls. doing this check here is useful
 	 * to avoid duplicate code in each driver.
@@ -227,26 +231,31 @@ static int rtc_dev_ioctl(struct inode *inode, struct file *file,
 	case RTC_EPOCH_SET:
 	case RTC_SET_TIME:
 		if (!capable(CAP_SYS_TIME))
-			return -EACCES;
+			err = -EACCES;
 		break;
 
 	case RTC_IRQP_SET:
 		if (arg > rtc->max_user_freq && !capable(CAP_SYS_RESOURCE))
-			return -EACCES;
+			err = -EACCES;
 		break;
 
 	case RTC_PIE_ON:
 		if (rtc->irq_freq > rtc->max_user_freq &&
 				!capable(CAP_SYS_RESOURCE))
-			return -EACCES;
+			err = -EACCES;
 		break;
 	}
 
+	if (err)
+		goto done;
+
 	/* try the driver's ioctl interface */
 	if (ops->ioctl) {
 		err = ops->ioctl(rtc->dev.parent, cmd, arg);
-		if (err != -ENOIOCTLCMD)
+		if (err != -ENOIOCTLCMD) {
+			mutex_unlock(&rtc->ops_lock);
 			return err;
+		}
 	}
 
 	/* if the driver does not provide the ioctl interface
@@ -265,15 +274,19 @@ static int rtc_dev_ioctl(struct inode *inode, struct file *file,
 
 	switch (cmd) {
 	case RTC_ALM_READ:
+		mutex_unlock(&rtc->ops_lock);
+
 		err = rtc_read_alarm(rtc, &alarm);
 		if (err < 0)
 			return err;
 
 		if (copy_to_user(uarg, &alarm.time, sizeof(tm)))
-			return -EFAULT;
-		break;
+			err = -EFAULT;
+		return err;
 
 	case RTC_ALM_SET:
+		mutex_unlock(&rtc->ops_lock);
+
 		if (copy_from_user(&alarm.time, uarg, sizeof(tm)))
 			return -EFAULT;
 
@@ -321,24 +334,26 @@ static int rtc_dev_ioctl(struct inode *inode, struct file *file,
 			}
 		}
 
-		err = rtc_set_alarm(rtc, &alarm);
-		break;
+		return rtc_set_alarm(rtc, &alarm);
 
 	case RTC_RD_TIME:
+		mutex_unlock(&rtc->ops_lock);
+
 		err = rtc_read_time(rtc, &tm);
 		if (err < 0)
 			return err;
 
 		if (copy_to_user(uarg, &tm, sizeof(tm)))
-			return -EFAULT;
-		break;
+			err = -EFAULT;
+		return err;
 
 	case RTC_SET_TIME:
+		mutex_unlock(&rtc->ops_lock);
+
 		if (copy_from_user(&tm, uarg, sizeof(tm)))
 			return -EFAULT;
 
-		err = rtc_set_time(rtc, &tm);
-		break;
+		return rtc_set_time(rtc, &tm);
 
 	case RTC_PIE_ON:
 		err = rtc_irq_set_state(rtc, NULL, 1);
@@ -376,34 +391,37 @@ static int rtc_dev_ioctl(struct inode *inode, struct file *file,
 		break;
 #endif
 	case RTC_WKALM_SET:
+		mutex_unlock(&rtc->ops_lock);
 		if (copy_from_user(&alarm, uarg, sizeof(alarm)))
 			return -EFAULT;
 
-		err = rtc_set_alarm(rtc, &alarm);
-		break;
+		return rtc_set_alarm(rtc, &alarm);
 
 	case RTC_WKALM_RD:
+		mutex_unlock(&rtc->ops_lock);
 		err = rtc_read_alarm(rtc, &alarm);
 		if (err < 0)
 			return err;
 
 		if (copy_to_user(uarg, &alarm, sizeof(alarm)))
-			return -EFAULT;
-		break;
+			err = -EFAULT;
+		return err;
 
 #ifdef CONFIG_RTC_INTF_DEV_UIE_EMUL
 	case RTC_UIE_OFF:
 		clear_uie(rtc);
-		return 0;
+		break;
 
 	case RTC_UIE_ON:
-		return set_uie(rtc);
+		err = set_uie(rtc);
 #endif
 	default:
 		err = -ENOTTY;
 		break;
 	}
 
+done:
+	mutex_unlock(&rtc->ops_lock);
 	return err;
 }
 
@@ -432,7 +450,7 @@ static const struct file_operations rtc_dev_fops = {
 	.llseek		= no_llseek,
 	.read		= rtc_dev_read,
 	.poll		= rtc_dev_poll,
-	.ioctl		= rtc_dev_ioctl,
+	.unlocked_ioctl	= rtc_dev_ioctl,
 	.open		= rtc_dev_open,
 	.release	= rtc_dev_release,
 	.fasync		= rtc_dev_fasync,
diff --git a/include/linux/rtc.h b/include/linux/rtc.h
index f2d0d1527721..b01fe004cb5e 100644
--- a/include/linux/rtc.h
+++ b/include/linux/rtc.h
@@ -115,6 +115,23 @@ extern void rtc_time_to_tm(unsigned long time, struct rtc_time *tm);
 
 extern struct class *rtc_class;
 
+/*
+ * For these RTC methods the device parameter is the physical device
+ * on whatever bus holds the hardware (I2C, Platform, SPI, etc), which
+ * was passed to rtc_device_register().  Its driver_data normally holds
+ * device state, including the rtc_device pointer for the RTC.
+ *
+ * Most of these methods are called with rtc_device.ops_lock held,
+ * through the rtc_*(struct rtc_device *, ...) calls.
+ *
+ * The (current) exceptions are mostly filesystem hooks:
+ *   - the proc() hook for procfs
+ *   - non-ioctl() chardev hooks:  open(), release(), read_callback()
+ *   - periodic irq calls:  irq_set_state(), irq_set_freq()
+ *
+ * REVISIT those periodic irq calls *do* have ops_lock when they're
+ * issued through ioctl() ...
+ */
 struct rtc_class_ops {
 	int (*open)(struct device *);
 	void (*release)(struct device *);

From 8fc2c767b06067b417c565c4e75731e68ed41fd8 Mon Sep 17 00:00:00 2001
From: "Kim B. Heino" <Kim.Heino@bluegiga.com>
Date: Wed, 23 Jul 2008 21:30:34 -0700
Subject: [PATCH 0430/1171] rtc: add support for ST M41T94 SPI RTC

This patch adds kernel driver for M41T94 RTC chip connected via SPI.
I've tested it on two different AT91-based hardwares.

This is third revision of the patch: some comments made by
Alessandro Zummo fixed.

Revision two added support for century bit and fixes.

Signed-off-by: Kim B. Heino <Kim.Heino@bluegiga.com>
Signed-off-by: Alessandro Zummo <a.zummo@towertech.it>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/rtc/Kconfig      |   9 ++
 drivers/rtc/Makefile     |   1 +
 drivers/rtc/rtc-m41t94.c | 173 +++++++++++++++++++++++++++++++++++++++
 3 files changed, 183 insertions(+)
 create mode 100644 drivers/rtc/rtc-m41t94.c

diff --git a/drivers/rtc/Kconfig b/drivers/rtc/Kconfig
index fc85bf2e4a97..beffb834c440 100644
--- a/drivers/rtc/Kconfig
+++ b/drivers/rtc/Kconfig
@@ -273,6 +273,15 @@ comment "SPI RTC drivers"
 
 if SPI_MASTER
 
+config RTC_DRV_M41T94
+	tristate "ST M41T94"
+	help
+	  If you say yes here you will get support for the
+	  ST M41T94 SPI RTC chip.
+
+	  This driver can also be built as a module. If so, the module
+	  will be called rtc-m41t94.
+
 config RTC_DRV_MAX6902
 	tristate "Maxim MAX6902"
 	help
diff --git a/drivers/rtc/Makefile b/drivers/rtc/Makefile
index b5d9d67df887..b0e1af54f800 100644
--- a/drivers/rtc/Makefile
+++ b/drivers/rtc/Makefile
@@ -34,6 +34,7 @@ obj-$(CONFIG_RTC_DRV_EP93XX)	+= rtc-ep93xx.o
 obj-$(CONFIG_RTC_DRV_FM3130)	+= rtc-fm3130.o
 obj-$(CONFIG_RTC_DRV_ISL1208)	+= rtc-isl1208.o
 obj-$(CONFIG_RTC_DRV_M41T80)	+= rtc-m41t80.o
+obj-$(CONFIG_RTC_DRV_M41T94)	+= rtc-m41t94.o
 obj-$(CONFIG_RTC_DRV_M48T59)	+= rtc-m48t59.o
 obj-$(CONFIG_RTC_DRV_M48T86)	+= rtc-m48t86.o
 obj-$(CONFIG_RTC_DRV_MAX6900)	+= rtc-max6900.o
diff --git a/drivers/rtc/rtc-m41t94.c b/drivers/rtc/rtc-m41t94.c
new file mode 100644
index 000000000000..9b19499c829e
--- /dev/null
+++ b/drivers/rtc/rtc-m41t94.c
@@ -0,0 +1,173 @@
+/*
+ * Driver for ST M41T94 SPI RTC
+ *
+ * Copyright (C) 2008 Kim B. Heino
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/platform_device.h>
+#include <linux/rtc.h>
+#include <linux/spi/spi.h>
+#include <linux/bcd.h>
+
+#define M41T94_REG_SECONDS	0x01
+#define M41T94_REG_MINUTES	0x02
+#define M41T94_REG_HOURS	0x03
+#define M41T94_REG_WDAY		0x04
+#define M41T94_REG_DAY		0x05
+#define M41T94_REG_MONTH	0x06
+#define M41T94_REG_YEAR		0x07
+#define M41T94_REG_HT		0x0c
+
+#define M41T94_BIT_HALT		0x40
+#define M41T94_BIT_STOP		0x80
+#define M41T94_BIT_CB		0x40
+#define M41T94_BIT_CEB		0x80
+
+static int m41t94_set_time(struct device *dev, struct rtc_time *tm)
+{
+	struct spi_device *spi = to_spi_device(dev);
+	u8 buf[8]; /* write cmd + 7 registers */
+
+	dev_dbg(dev, "%s secs=%d, mins=%d, "
+		"hours=%d, mday=%d, mon=%d, year=%d, wday=%d\n",
+		"write", tm->tm_sec, tm->tm_min,
+		tm->tm_hour, tm->tm_mday,
+		tm->tm_mon, tm->tm_year, tm->tm_wday);
+
+	buf[0] = 0x80 | M41T94_REG_SECONDS; /* write time + date */
+	buf[M41T94_REG_SECONDS] = BIN2BCD(tm->tm_sec);
+	buf[M41T94_REG_MINUTES] = BIN2BCD(tm->tm_min);
+	buf[M41T94_REG_HOURS]   = BIN2BCD(tm->tm_hour);
+	buf[M41T94_REG_WDAY]    = BIN2BCD(tm->tm_wday + 1);
+	buf[M41T94_REG_DAY]     = BIN2BCD(tm->tm_mday);
+	buf[M41T94_REG_MONTH]   = BIN2BCD(tm->tm_mon + 1);
+
+	buf[M41T94_REG_HOURS] |= M41T94_BIT_CEB;
+	if (tm->tm_year >= 100)
+		buf[M41T94_REG_HOURS] |= M41T94_BIT_CB;
+	buf[M41T94_REG_YEAR] = BIN2BCD(tm->tm_year % 100);
+
+	return spi_write(spi, buf, 8);
+}
+
+static int m41t94_read_time(struct device *dev, struct rtc_time *tm)
+{
+	struct spi_device *spi = to_spi_device(dev);
+	u8 buf[2];
+	int ret, hour;
+
+	/* clear halt update bit */
+	ret = spi_w8r8(spi, M41T94_REG_HT);
+	if (ret < 0)
+		return ret;
+	if (ret & M41T94_BIT_HALT) {
+		buf[0] = 0x80 | M41T94_REG_HT;
+		buf[1] = ret & ~M41T94_BIT_HALT;
+		spi_write(spi, buf, 2);
+	}
+
+	/* clear stop bit */
+	ret = spi_w8r8(spi, M41T94_REG_SECONDS);
+	if (ret < 0)
+		return ret;
+	if (ret & M41T94_BIT_STOP) {
+		buf[0] = 0x80 | M41T94_REG_SECONDS;
+		buf[1] = ret & ~M41T94_BIT_STOP;
+		spi_write(spi, buf, 2);
+	}
+
+	tm->tm_sec  = BCD2BIN(spi_w8r8(spi, M41T94_REG_SECONDS));
+	tm->tm_min  = BCD2BIN(spi_w8r8(spi, M41T94_REG_MINUTES));
+	hour = spi_w8r8(spi, M41T94_REG_HOURS);
+	tm->tm_hour = BCD2BIN(hour & 0x3f);
+	tm->tm_wday = BCD2BIN(spi_w8r8(spi, M41T94_REG_WDAY)) - 1;
+	tm->tm_mday = BCD2BIN(spi_w8r8(spi, M41T94_REG_DAY));
+	tm->tm_mon  = BCD2BIN(spi_w8r8(spi, M41T94_REG_MONTH)) - 1;
+	tm->tm_year = BCD2BIN(spi_w8r8(spi, M41T94_REG_YEAR));
+	if ((hour & M41T94_BIT_CB) || !(hour & M41T94_BIT_CEB))
+		tm->tm_year += 100;
+
+	dev_dbg(dev, "%s secs=%d, mins=%d, "
+		"hours=%d, mday=%d, mon=%d, year=%d, wday=%d\n",
+		"read", tm->tm_sec, tm->tm_min,
+		tm->tm_hour, tm->tm_mday,
+		tm->tm_mon, tm->tm_year, tm->tm_wday);
+
+	/* initial clock setting can be undefined */
+	return rtc_valid_tm(tm);
+}
+
+static const struct rtc_class_ops m41t94_rtc_ops = {
+	.read_time	= m41t94_read_time,
+	.set_time	= m41t94_set_time,
+};
+
+static struct spi_driver m41t94_driver;
+
+static int __devinit m41t94_probe(struct spi_device *spi)
+{
+	struct rtc_device *rtc;
+	int res;
+
+	spi->bits_per_word = 8;
+	spi_setup(spi);
+
+	res = spi_w8r8(spi, M41T94_REG_SECONDS);
+	if (res < 0) {
+		dev_err(&spi->dev, "not found.\n");
+		return res;
+	}
+
+	rtc = rtc_device_register(m41t94_driver.driver.name,
+		&spi->dev, &m41t94_rtc_ops, THIS_MODULE);
+	if (IS_ERR(rtc))
+		return PTR_ERR(rtc);
+
+	dev_set_drvdata(&spi->dev, rtc);
+
+	return 0;
+}
+
+static int __devexit m41t94_remove(struct spi_device *spi)
+{
+	struct rtc_device *rtc = platform_get_drvdata(spi);
+
+	if (rtc)
+		rtc_device_unregister(rtc);
+
+	return 0;
+}
+
+static struct spi_driver m41t94_driver = {
+	.driver = {
+		.name	= "rtc-m41t94",
+		.bus	= &spi_bus_type,
+		.owner	= THIS_MODULE,
+	},
+	.probe	= m41t94_probe,
+	.remove = __devexit_p(m41t94_remove),
+};
+
+static __init int m41t94_init(void)
+{
+	return spi_register_driver(&m41t94_driver);
+}
+
+module_init(m41t94_init);
+
+static __exit void m41t94_exit(void)
+{
+	spi_unregister_driver(&m41t94_driver);
+}
+
+module_exit(m41t94_exit);
+
+MODULE_AUTHOR("Kim B. Heino <Kim.Heino@bluegiga.com>");
+MODULE_DESCRIPTION("Driver for ST M41T94 SPI RTC");
+MODULE_LICENSE("GPL");

From 53e84b672c1a8190af2b376c35c7a39cf1214f59 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Wed, 23 Jul 2008 21:30:36 -0700
Subject: [PATCH 0431/1171] rtc: ds1305/ds1306 driver

Support the Dallas/Maxim DS1305 and DS1306 RTC chips.  These use SPI, and
support alarms, NVRAM, and a trickle charger for use when their backup
power supply is a supercap or rechargeable cell.

This basic driver doesn't yet support suspend/resume or wakealarms.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/rtc/Kconfig        |  10 +
 drivers/rtc/Makefile       |   1 +
 drivers/rtc/rtc-ds1305.c   | 847 +++++++++++++++++++++++++++++++++++++
 include/linux/spi/ds1305.h |  35 ++
 4 files changed, 893 insertions(+)
 create mode 100644 drivers/rtc/rtc-ds1305.c
 create mode 100644 include/linux/spi/ds1305.h

diff --git a/drivers/rtc/Kconfig b/drivers/rtc/Kconfig
index beffb834c440..90ab73825401 100644
--- a/drivers/rtc/Kconfig
+++ b/drivers/rtc/Kconfig
@@ -282,6 +282,16 @@ config RTC_DRV_M41T94
 	  This driver can also be built as a module. If so, the module
 	  will be called rtc-m41t94.
 
+config RTC_DRV_DS1305
+	tristate "Dallas/Maxim DS1305/DS1306"
+	help
+	  Select this driver to get support for the Dallas/Maxim DS1305
+	  and DS1306 real time clock chips.  These support a trickle
+	  charger, alarms, and NVRAM in addition to the clock.
+
+	  This driver can also be built as a module. If so, the module
+	  will be called rtc-ds1305.
+
 config RTC_DRV_MAX6902
 	tristate "Maxim MAX6902"
 	help
diff --git a/drivers/rtc/Makefile b/drivers/rtc/Makefile
index b0e1af54f800..18622ef84cab 100644
--- a/drivers/rtc/Makefile
+++ b/drivers/rtc/Makefile
@@ -24,6 +24,7 @@ obj-$(CONFIG_RTC_DRV_BFIN)	+= rtc-bfin.o
 obj-$(CONFIG_RTC_DRV_CMOS)	+= rtc-cmos.o
 obj-$(CONFIG_RTC_DRV_DS1216)	+= rtc-ds1216.o
 obj-$(CONFIG_RTC_DRV_DS1302)	+= rtc-ds1302.o
+obj-$(CONFIG_RTC_DRV_DS1305)	+= rtc-ds1305.o
 obj-$(CONFIG_RTC_DRV_DS1307)	+= rtc-ds1307.o
 obj-$(CONFIG_RTC_DRV_DS1374)	+= rtc-ds1374.o
 obj-$(CONFIG_RTC_DRV_DS1511)	+= rtc-ds1511.o
diff --git a/drivers/rtc/rtc-ds1305.c b/drivers/rtc/rtc-ds1305.c
new file mode 100644
index 000000000000..b91d02a3ace9
--- /dev/null
+++ b/drivers/rtc/rtc-ds1305.c
@@ -0,0 +1,847 @@
+/*
+ * rtc-ds1305.c -- driver for DS1305 and DS1306 SPI RTC chips
+ *
+ * Copyright (C) 2008 David Brownell
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/bcd.h>
+#include <linux/rtc.h>
+#include <linux/workqueue.h>
+
+#include <linux/spi/spi.h>
+#include <linux/spi/ds1305.h>
+
+
+/*
+ * Registers ... mask DS1305_WRITE into register address to write,
+ * otherwise you're reading it.  All non-bitmask values are BCD.
+ */
+#define DS1305_WRITE		0x80
+
+
+/* RTC date/time ... the main special cases are that we:
+ *  - Need fancy "hours" encoding in 12hour mode
+ *  - Don't rely on the "day-of-week" field (or tm_wday)
+ *  - Are a 21st-century clock (2000 <= year < 2100)
+ */
+#define DS1305_RTC_LEN		7		/* bytes for RTC regs */
+
+#define DS1305_SEC		0x00		/* register addresses */
+#define DS1305_MIN		0x01
+#define DS1305_HOUR		0x02
+#	define DS1305_HR_12		0x40	/* set == 12 hr mode */
+#	define DS1305_HR_PM		0x20	/* set == PM (12hr mode) */
+#define DS1305_WDAY		0x03
+#define DS1305_MDAY		0x04
+#define DS1305_MON		0x05
+#define DS1305_YEAR		0x06
+
+
+/* The two alarms have only sec/min/hour/wday fields (ALM_LEN).
+ * DS1305_ALM_DISABLE disables a match field (some combos are bad).
+ *
+ * NOTE that since we don't use WDAY, we limit ourselves to alarms
+ * only one day into the future (vs potentially up to a week).
+ *
+ * NOTE ALSO that while we could generate once-a-second IRQs (UIE), we
+ * don't currently support them.  We'd either need to do it only when
+ * no alarm is pending (not the standard model), or to use the second
+ * alarm (implying that this is a DS1305 not DS1306, *and* that either
+ * it's wired up a second IRQ we know, or that INTCN is set)
+ */
+#define DS1305_ALM_LEN		4		/* bytes for ALM regs */
+#define DS1305_ALM_DISABLE	0x80
+
+#define DS1305_ALM0(r)		(0x07 + (r))	/* register addresses */
+#define DS1305_ALM1(r)		(0x0b + (r))
+
+
+/* three control registers */
+#define DS1305_CONTROL_LEN	3		/* bytes of control regs */
+
+#define DS1305_CONTROL		0x0f		/* register addresses */
+#	define DS1305_nEOSC		0x80	/* low enables oscillator */
+#	define DS1305_WP		0x40	/* write protect */
+#	define DS1305_INTCN		0x04	/* clear == only int0 used */
+#	define DS1306_1HZ		0x04	/* enable 1Hz output */
+#	define DS1305_AEI1		0x02	/* enable ALM1 IRQ */
+#	define DS1305_AEI0		0x01	/* enable ALM0 IRQ */
+#define DS1305_STATUS		0x10
+/* status has just AEIx bits, mirrored as IRQFx */
+#define DS1305_TRICKLE		0x11
+/* trickle bits are defined in <linux/spi/ds1305.h> */
+
+/* a bunch of NVRAM */
+#define DS1305_NVRAM_LEN	96		/* bytes of NVRAM */
+
+#define DS1305_NVRAM		0x20		/* register addresses */
+
+
+struct ds1305 {
+	struct spi_device	*spi;
+	struct rtc_device	*rtc;
+
+	struct work_struct	work;
+
+	unsigned long		flags;
+#define FLAG_EXITING	0
+
+	bool			hr12;
+	u8			ctrl[DS1305_CONTROL_LEN];
+};
+
+
+/*----------------------------------------------------------------------*/
+
+/*
+ * Utilities ...  tolerate 12-hour AM/PM notation in case of non-Linux
+ * software (like a bootloader) which may require it.
+ */
+
+static unsigned bcd2hour(u8 bcd)
+{
+	if (bcd & DS1305_HR_12) {
+		unsigned	hour = 0;
+
+		bcd &= ~DS1305_HR_12;
+		if (bcd & DS1305_HR_PM) {
+			hour = 12;
+			bcd &= ~DS1305_HR_PM;
+		}
+		hour += BCD2BIN(bcd);
+		return hour - 1;
+	}
+	return BCD2BIN(bcd);
+}
+
+static u8 hour2bcd(bool hr12, int hour)
+{
+	if (hr12) {
+		hour++;
+		if (hour <= 12)
+			return DS1305_HR_12 | BIN2BCD(hour);
+		hour -= 12;
+		return DS1305_HR_12 | DS1305_HR_PM | BIN2BCD(hour);
+	}
+	return BIN2BCD(hour);
+}
+
+/*----------------------------------------------------------------------*/
+
+/*
+ * Interface to RTC framework
+ */
+
+#ifdef CONFIG_RTC_INTF_DEV
+
+/*
+ * Context: caller holds rtc->ops_lock (to protect ds1305->ctrl)
+ */
+static int ds1305_ioctl(struct device *dev, unsigned cmd, unsigned long arg)
+{
+	struct ds1305	*ds1305 = dev_get_drvdata(dev);
+	u8		buf[2];
+	int		status = -ENOIOCTLCMD;
+
+	buf[0] = DS1305_WRITE | DS1305_CONTROL;
+	buf[1] = ds1305->ctrl[0];
+
+	switch (cmd) {
+	case RTC_AIE_OFF:
+		status = 0;
+		if (!(buf[1] & DS1305_AEI0))
+			goto done;
+		buf[1] &= ~DS1305_AEI0;
+		break;
+
+	case RTC_AIE_ON:
+		status = 0;
+		if (ds1305->ctrl[0] & DS1305_AEI0)
+			goto done;
+		buf[1] |= DS1305_AEI0;
+		break;
+	}
+	if (status == 0) {
+		status = spi_write_then_read(ds1305->spi, buf, sizeof buf,
+				NULL, 0);
+		if (status >= 0)
+			ds1305->ctrl[0] = buf[1];
+	}
+
+done:
+	return status;
+}
+
+#else
+#define ds1305_ioctl	NULL
+#endif
+
+/*
+ * Get/set of date and time is pretty normal.
+ */
+
+static int ds1305_get_time(struct device *dev, struct rtc_time *time)
+{
+	struct ds1305	*ds1305 = dev_get_drvdata(dev);
+	u8		addr = DS1305_SEC;
+	u8		buf[DS1305_RTC_LEN];
+	int		status;
+
+	/* Use write-then-read to get all the date/time registers
+	 * since dma from stack is nonportable
+	 */
+	status = spi_write_then_read(ds1305->spi, &addr, sizeof addr,
+			buf, sizeof buf);
+	if (status < 0)
+		return status;
+
+	dev_vdbg(dev, "%s: %02x %02x %02x, %02x %02x %02x %02x\n",
+		"read", buf[0], buf[1], buf[2], buf[3],
+		buf[4], buf[5], buf[6]);
+
+	/* Decode the registers */
+	time->tm_sec = BCD2BIN(buf[DS1305_SEC]);
+	time->tm_min = BCD2BIN(buf[DS1305_MIN]);
+	time->tm_hour = bcd2hour(buf[DS1305_HOUR]);
+	time->tm_wday = buf[DS1305_WDAY] - 1;
+	time->tm_mday = BCD2BIN(buf[DS1305_MDAY]);
+	time->tm_mon = BCD2BIN(buf[DS1305_MON]) - 1;
+	time->tm_year = BCD2BIN(buf[DS1305_YEAR]) + 100;
+
+	dev_vdbg(dev, "%s secs=%d, mins=%d, "
+		"hours=%d, mday=%d, mon=%d, year=%d, wday=%d\n",
+		"read", time->tm_sec, time->tm_min,
+		time->tm_hour, time->tm_mday,
+		time->tm_mon, time->tm_year, time->tm_wday);
+
+	/* Time may not be set */
+	return rtc_valid_tm(time);
+}
+
+static int ds1305_set_time(struct device *dev, struct rtc_time *time)
+{
+	struct ds1305	*ds1305 = dev_get_drvdata(dev);
+	u8		buf[1 + DS1305_RTC_LEN];
+	u8		*bp = buf;
+
+	dev_vdbg(dev, "%s secs=%d, mins=%d, "
+		"hours=%d, mday=%d, mon=%d, year=%d, wday=%d\n",
+		"write", time->tm_sec, time->tm_min,
+		time->tm_hour, time->tm_mday,
+		time->tm_mon, time->tm_year, time->tm_wday);
+
+	/* Write registers starting at the first time/date address. */
+	*bp++ = DS1305_WRITE | DS1305_SEC;
+
+	*bp++ = BIN2BCD(time->tm_sec);
+	*bp++ = BIN2BCD(time->tm_min);
+	*bp++ = hour2bcd(ds1305->hr12, time->tm_hour);
+	*bp++ = (time->tm_wday < 7) ? (time->tm_wday + 1) : 1;
+	*bp++ = BIN2BCD(time->tm_mday);
+	*bp++ = BIN2BCD(time->tm_mon + 1);
+	*bp++ = BIN2BCD(time->tm_year - 100);
+
+	dev_dbg(dev, "%s: %02x %02x %02x, %02x %02x %02x %02x\n",
+		"write", buf[1], buf[2], buf[3],
+		buf[4], buf[5], buf[6], buf[7]);
+
+	/* use write-then-read since dma from stack is nonportable */
+	return spi_write_then_read(ds1305->spi, buf, sizeof buf,
+			NULL, 0);
+}
+
+/*
+ * Get/set of alarm is a bit funky:
+ *
+ * - First there's the inherent raciness of getting the (partitioned)
+ *   status of an alarm that could trigger while we're reading parts
+ *   of that status.
+ *
+ * - Second there's its limited range (we could increase it a bit by
+ *   relying on WDAY), which means it will easily roll over.
+ *
+ * - Third there's the choice of two alarms and alarm signals.
+ *   Here we use ALM0 and expect that nINT0 (open drain) is used;
+ *   that's the only real option for DS1306 runtime alarms, and is
+ *   natural on DS1305.
+ *
+ * - Fourth, there's also ALM1, and a second interrupt signal:
+ *     + On DS1305 ALM1 uses nINT1 (when INTCN=1) else nINT0;
+ *     + On DS1306 ALM1 only uses INT1 (an active high pulse)
+ *       and it won't work when VCC1 is active.
+ *
+ *   So to be most general, we should probably set both alarms to the
+ *   same value, letting ALM1 be the wakeup event source on DS1306
+ *   and handling several wiring options on DS1305.
+ *
+ * - Fifth, we support the polled mode (as well as possible; why not?)
+ *   even when no interrupt line is wired to an IRQ.
+ */
+
+/*
+ * Context: caller holds rtc->ops_lock (to protect ds1305->ctrl)
+ */
+static int ds1305_get_alarm(struct device *dev, struct rtc_wkalrm *alm)
+{
+	struct ds1305	*ds1305 = dev_get_drvdata(dev);
+	struct spi_device *spi = ds1305->spi;
+	u8		addr;
+	int		status;
+	u8		buf[DS1305_ALM_LEN];
+
+	/* Refresh control register cache BEFORE reading ALM0 registers,
+	 * since reading alarm registers acks any pending IRQ.  That
+	 * makes returning "pending" status a bit of a lie, but that bit
+	 * of EFI status is at best fragile anyway (given IRQ handlers).
+	 */
+	addr = DS1305_CONTROL;
+	status = spi_write_then_read(spi, &addr, sizeof addr,
+			ds1305->ctrl, sizeof ds1305->ctrl);
+	if (status < 0)
+		return status;
+
+	alm->enabled = !!(ds1305->ctrl[0] & DS1305_AEI0);
+	alm->pending = !!(ds1305->ctrl[1] & DS1305_AEI0);
+
+	/* get and check ALM0 registers */
+	addr = DS1305_ALM0(DS1305_SEC);
+	status = spi_write_then_read(spi, &addr, sizeof addr,
+			buf, sizeof buf);
+	if (status < 0)
+		return status;
+
+	dev_vdbg(dev, "%s: %02x %02x %02x %02x\n",
+		"alm0 read", buf[DS1305_SEC], buf[DS1305_MIN],
+		buf[DS1305_HOUR], buf[DS1305_WDAY]);
+
+	if ((DS1305_ALM_DISABLE & buf[DS1305_SEC])
+			|| (DS1305_ALM_DISABLE & buf[DS1305_MIN])
+			|| (DS1305_ALM_DISABLE & buf[DS1305_HOUR]))
+		return -EIO;
+
+	/* Stuff these values into alm->time and let RTC framework code
+	 * fill in the rest ... and also handle rollover to tomorrow when
+	 * that's needed.
+	 */
+	alm->time.tm_sec = BCD2BIN(buf[DS1305_SEC]);
+	alm->time.tm_min = BCD2BIN(buf[DS1305_MIN]);
+	alm->time.tm_hour = bcd2hour(buf[DS1305_HOUR]);
+	alm->time.tm_mday = -1;
+	alm->time.tm_mon = -1;
+	alm->time.tm_year = -1;
+	/* next three fields are unused by Linux */
+	alm->time.tm_wday = -1;
+	alm->time.tm_mday = -1;
+	alm->time.tm_isdst = -1;
+
+	return 0;
+}
+
+/*
+ * Context: caller holds rtc->ops_lock (to protect ds1305->ctrl)
+ */
+static int ds1305_set_alarm(struct device *dev, struct rtc_wkalrm *alm)
+{
+	struct ds1305	*ds1305 = dev_get_drvdata(dev);
+	struct spi_device *spi = ds1305->spi;
+	unsigned long	now, later;
+	struct rtc_time	tm;
+	int		status;
+	u8		buf[1 + DS1305_ALM_LEN];
+
+	/* convert desired alarm to time_t */
+	status = rtc_tm_to_time(&alm->time, &later);
+	if (status < 0)
+		return status;
+
+	/* Read current time as time_t */
+	status = ds1305_get_time(dev, &tm);
+	if (status < 0)
+		return status;
+	status = rtc_tm_to_time(&tm, &now);
+	if (status < 0)
+		return status;
+
+	/* make sure alarm fires within the next 24 hours */
+	if (later <= now)
+		return -EINVAL;
+	if ((later - now) > 24 * 60 * 60)
+		return -EDOM;
+
+	/* disable alarm if needed */
+	if (ds1305->ctrl[0] & DS1305_AEI0) {
+		ds1305->ctrl[0] &= ~DS1305_AEI0;
+
+		buf[0] = DS1305_WRITE | DS1305_CONTROL;
+		buf[1] = ds1305->ctrl[0];
+		status = spi_write_then_read(ds1305->spi, buf, 2, NULL, 0);
+		if (status < 0)
+			return status;
+	}
+
+	/* write alarm */
+	buf[0] = DS1305_WRITE | DS1305_ALM0(DS1305_SEC);
+	buf[1 + DS1305_SEC] = BIN2BCD(alm->time.tm_sec);
+	buf[1 + DS1305_MIN] = BIN2BCD(alm->time.tm_min);
+	buf[1 + DS1305_HOUR] = hour2bcd(ds1305->hr12, alm->time.tm_hour);
+	buf[1 + DS1305_WDAY] = DS1305_ALM_DISABLE;
+
+	dev_dbg(dev, "%s: %02x %02x %02x %02x\n",
+		"alm0 write", buf[1 + DS1305_SEC], buf[1 + DS1305_MIN],
+		buf[1 + DS1305_HOUR], buf[1 + DS1305_WDAY]);
+
+	status = spi_write_then_read(spi, buf, sizeof buf, NULL, 0);
+	if (status < 0)
+		return status;
+
+	/* enable alarm if requested */
+	if (alm->enabled) {
+		ds1305->ctrl[0] |= DS1305_AEI0;
+
+		buf[0] = DS1305_WRITE | DS1305_CONTROL;
+		buf[1] = ds1305->ctrl[0];
+		status = spi_write_then_read(ds1305->spi, buf, 2, NULL, 0);
+	}
+
+	return status;
+}
+
+#ifdef CONFIG_PROC_FS
+
+static int ds1305_proc(struct device *dev, struct seq_file *seq)
+{
+	struct ds1305	*ds1305 = dev_get_drvdata(dev);
+	char		*diodes = "no";
+	char		*resistors = "";
+
+	/* ctrl[2] is treated as read-only; no locking needed */
+	if ((ds1305->ctrl[2] & 0xf0) == DS1305_TRICKLE_MAGIC) {
+		switch (ds1305->ctrl[2] & 0x0c) {
+		case DS1305_TRICKLE_DS2:
+			diodes = "2 diodes, ";
+			break;
+		case DS1305_TRICKLE_DS1:
+			diodes = "1 diode, ";
+			break;
+		default:
+			goto done;
+		}
+		switch (ds1305->ctrl[2] & 0x03) {
+		case DS1305_TRICKLE_2K:
+			resistors = "2k Ohm";
+			break;
+		case DS1305_TRICKLE_4K:
+			resistors = "4k Ohm";
+			break;
+		case DS1305_TRICKLE_8K:
+			resistors = "8k Ohm";
+			break;
+		default:
+			diodes = "no";
+			break;
+		}
+	}
+
+done:
+	return seq_printf(seq,
+			"trickle_charge\t: %s%s\n",
+			diodes, resistors);
+}
+
+#else
+#define ds1305_proc	NULL
+#endif
+
+static const struct rtc_class_ops ds1305_ops = {
+	.ioctl		= ds1305_ioctl,
+	.read_time	= ds1305_get_time,
+	.set_time	= ds1305_set_time,
+	.read_alarm	= ds1305_get_alarm,
+	.set_alarm	= ds1305_set_alarm,
+	.proc		= ds1305_proc,
+};
+
+static void ds1305_work(struct work_struct *work)
+{
+	struct ds1305	*ds1305 = container_of(work, struct ds1305, work);
+	struct mutex	*lock = &ds1305->rtc->ops_lock;
+	struct spi_device *spi = ds1305->spi;
+	u8		buf[3];
+	int		status;
+
+	/* lock to protect ds1305->ctrl */
+	mutex_lock(lock);
+
+	/* Disable the IRQ, and clear its status ... for now, we "know"
+	 * that if more than one alarm is active, they're in sync.
+	 * Note that reading ALM data registers also clears IRQ status.
+	 */
+	ds1305->ctrl[0] &= ~(DS1305_AEI1 | DS1305_AEI0);
+	ds1305->ctrl[1] = 0;
+
+	buf[0] = DS1305_WRITE | DS1305_CONTROL;
+	buf[1] = ds1305->ctrl[0];
+	buf[2] = 0;
+
+	status = spi_write_then_read(spi, buf, sizeof buf,
+			NULL, 0);
+	if (status < 0)
+		dev_dbg(&spi->dev, "clear irq --> %d\n", status);
+
+	mutex_unlock(lock);
+
+	if (!test_bit(FLAG_EXITING, &ds1305->flags))
+		enable_irq(spi->irq);
+
+	/* rtc_update_irq() requires an IRQ-disabled context */
+	local_irq_disable();
+	rtc_update_irq(ds1305->rtc, 1, RTC_AF | RTC_IRQF);
+	local_irq_enable();
+}
+
+/*
+ * This "real" IRQ handler hands off to a workqueue mostly to allow
+ * mutex locking for ds1305->ctrl ... unlike I2C, we could issue async
+ * I/O requests in IRQ context (to clear the IRQ status).
+ */
+static irqreturn_t ds1305_irq(int irq, void *p)
+{
+	struct ds1305		*ds1305 = p;
+
+	disable_irq(irq);
+	schedule_work(&ds1305->work);
+	return IRQ_HANDLED;
+}
+
+/*----------------------------------------------------------------------*/
+
+/*
+ * Interface for NVRAM
+ */
+
+static void msg_init(struct spi_message *m, struct spi_transfer *x,
+		u8 *addr, size_t count, char *tx, char *rx)
+{
+	spi_message_init(m);
+	memset(x, 0, 2 * sizeof(*x));
+
+	x->tx_buf = addr;
+	x->len = 1;
+	spi_message_add_tail(x, m);
+
+	x++;
+
+	x->tx_buf = tx;
+	x->rx_buf = rx;
+	x->len = count;
+	spi_message_add_tail(x, m);
+}
+
+static ssize_t
+ds1305_nvram_read(struct kobject *kobj, struct bin_attribute *attr,
+		char *buf, loff_t off, size_t count)
+{
+	struct spi_device	*spi;
+	u8			addr;
+	struct spi_message	m;
+	struct spi_transfer	x[2];
+	int			status;
+
+	spi = container_of(kobj, struct spi_device, dev.kobj);
+
+	if (unlikely(off >= DS1305_NVRAM_LEN))
+		return 0;
+	if (count >= DS1305_NVRAM_LEN)
+		count = DS1305_NVRAM_LEN;
+	if ((off + count) > DS1305_NVRAM_LEN)
+		count = DS1305_NVRAM_LEN - off;
+	if (unlikely(!count))
+		return count;
+
+	addr = DS1305_NVRAM + off;
+	msg_init(&m, x, &addr, count, NULL, buf);
+
+	status = spi_sync(spi, &m);
+	if (status < 0)
+		dev_err(&spi->dev, "nvram %s error %d\n", "read", status);
+	return (status < 0) ? status : count;
+}
+
+static ssize_t
+ds1305_nvram_write(struct kobject *kobj, struct bin_attribute *attr,
+		char *buf, loff_t off, size_t count)
+{
+	struct spi_device	*spi;
+	u8			addr;
+	struct spi_message	m;
+	struct spi_transfer	x[2];
+	int			status;
+
+	spi = container_of(kobj, struct spi_device, dev.kobj);
+
+	if (unlikely(off >= DS1305_NVRAM_LEN))
+		return -EFBIG;
+	if (count >= DS1305_NVRAM_LEN)
+		count = DS1305_NVRAM_LEN;
+	if ((off + count) > DS1305_NVRAM_LEN)
+		count = DS1305_NVRAM_LEN - off;
+	if (unlikely(!count))
+		return count;
+
+	addr = (DS1305_WRITE | DS1305_NVRAM) + off;
+	msg_init(&m, x, &addr, count, buf, NULL);
+
+	status = spi_sync(spi, &m);
+	if (status < 0)
+		dev_err(&spi->dev, "nvram %s error %d\n", "write", status);
+	return (status < 0) ? status : count;
+}
+
+static struct bin_attribute nvram = {
+	.attr.name	= "nvram",
+	.attr.mode	= S_IRUGO | S_IWUSR,
+	.attr.owner	= THIS_MODULE,
+	.read		= ds1305_nvram_read,
+	.write		= ds1305_nvram_write,
+	.size		= DS1305_NVRAM_LEN,
+};
+
+/*----------------------------------------------------------------------*/
+
+/*
+ * Interface to SPI stack
+ */
+
+static int __devinit ds1305_probe(struct spi_device *spi)
+{
+	struct ds1305			*ds1305;
+	struct rtc_device		*rtc;
+	int				status;
+	u8				addr, value;
+	struct ds1305_platform_data	*pdata = spi->dev.platform_data;
+	bool				write_ctrl = false;
+
+	/* Sanity check board setup data.  This may be hooked up
+	 * in 3wire mode, but we don't care.  Note that unless
+	 * there's an inverter in place, this needs SPI_CS_HIGH!
+	 */
+	if ((spi->bits_per_word && spi->bits_per_word != 8)
+			|| (spi->max_speed_hz > 2000000)
+			|| !(spi->mode & SPI_CPHA))
+		return -EINVAL;
+
+	/* set up driver data */
+	ds1305 = kzalloc(sizeof *ds1305, GFP_KERNEL);
+	if (!ds1305)
+		return -ENOMEM;
+	ds1305->spi = spi;
+	spi_set_drvdata(spi, ds1305);
+
+	/* read and cache control registers */
+	addr = DS1305_CONTROL;
+	status = spi_write_then_read(spi, &addr, sizeof addr,
+			ds1305->ctrl, sizeof ds1305->ctrl);
+	if (status < 0) {
+		dev_dbg(&spi->dev, "can't %s, %d\n",
+				"read", status);
+		goto fail0;
+	}
+
+	dev_dbg(&spi->dev, "ctrl %s: %02x %02x %02x\n",
+			"read", ds1305->ctrl[0],
+			ds1305->ctrl[1], ds1305->ctrl[2]);
+
+	/* Sanity check register values ... partially compensating for the
+	 * fact that SPI has no device handshake.  A pullup on MISO would
+	 * make these tests fail; but not all systems will have one.  If
+	 * some register is neither 0x00 nor 0xff, a chip is likely there.
+	 */
+	if ((ds1305->ctrl[0] & 0x38) != 0 || (ds1305->ctrl[1] & 0xfc) != 0) {
+		dev_dbg(&spi->dev, "RTC chip is not present\n");
+		status = -ENODEV;
+		goto fail0;
+	}
+	if (ds1305->ctrl[2] == 0)
+		dev_dbg(&spi->dev, "chip may not be present\n");
+
+	/* enable writes if needed ... if we were paranoid it would
+	 * make sense to enable them only when absolutely necessary.
+	 */
+	if (ds1305->ctrl[0] & DS1305_WP) {
+		u8		buf[2];
+
+		ds1305->ctrl[0] &= ~DS1305_WP;
+
+		buf[0] = DS1305_WRITE | DS1305_CONTROL;
+		buf[1] = ds1305->ctrl[0];
+		status = spi_write_then_read(spi, buf, sizeof buf, NULL, 0);
+
+		dev_dbg(&spi->dev, "clear WP --> %d\n", status);
+		if (status < 0)
+			goto fail0;
+	}
+
+	/* on DS1305, maybe start oscillator; like most low power
+	 * oscillators, it may take a second to stabilize
+	 */
+	if (ds1305->ctrl[0] & DS1305_nEOSC) {
+		ds1305->ctrl[0] &= ~DS1305_nEOSC;
+		write_ctrl = true;
+		dev_warn(&spi->dev, "SET TIME!\n");
+	}
+
+	/* ack any pending IRQs */
+	if (ds1305->ctrl[1]) {
+		ds1305->ctrl[1] = 0;
+		write_ctrl = true;
+	}
+
+	/* this may need one-time (re)init */
+	if (pdata) {
+		/* maybe enable trickle charge */
+		if (((ds1305->ctrl[2] & 0xf0) != DS1305_TRICKLE_MAGIC)) {
+			ds1305->ctrl[2] = DS1305_TRICKLE_MAGIC
+						| pdata->trickle;
+			write_ctrl = true;
+		}
+
+		/* on DS1306, configure 1 Hz signal */
+		if (pdata->is_ds1306) {
+			if (pdata->en_1hz) {
+				if (!(ds1305->ctrl[0] & DS1306_1HZ)) {
+					ds1305->ctrl[0] |= DS1306_1HZ;
+					write_ctrl = true;
+				}
+			} else {
+				if (ds1305->ctrl[0] & DS1306_1HZ) {
+					ds1305->ctrl[0] &= ~DS1306_1HZ;
+					write_ctrl = true;
+				}
+			}
+		}
+	}
+
+	if (write_ctrl) {
+		u8		buf[4];
+
+		buf[0] = DS1305_WRITE | DS1305_CONTROL;
+		buf[1] = ds1305->ctrl[0];
+		buf[2] = ds1305->ctrl[1];
+		buf[3] = ds1305->ctrl[2];
+		status = spi_write_then_read(spi, buf, sizeof buf, NULL, 0);
+		if (status < 0) {
+			dev_dbg(&spi->dev, "can't %s, %d\n",
+					"write", status);
+			goto fail0;
+		}
+
+		dev_dbg(&spi->dev, "ctrl %s: %02x %02x %02x\n",
+				"write", ds1305->ctrl[0],
+				ds1305->ctrl[1], ds1305->ctrl[2]);
+	}
+
+	/* see if non-Linux software set up AM/PM mode */
+	addr = DS1305_HOUR;
+	status = spi_write_then_read(spi, &addr, sizeof addr,
+				&value, sizeof value);
+	if (status < 0) {
+		dev_dbg(&spi->dev, "read HOUR --> %d\n", status);
+		goto fail0;
+	}
+
+	ds1305->hr12 = (DS1305_HR_12 & value) != 0;
+	if (ds1305->hr12)
+		dev_dbg(&spi->dev, "AM/PM\n");
+
+	/* register RTC ... from here on, ds1305->ctrl needs locking */
+	rtc = rtc_device_register("ds1305", &spi->dev,
+			&ds1305_ops, THIS_MODULE);
+	if (IS_ERR(rtc)) {
+		status = PTR_ERR(rtc);
+		dev_dbg(&spi->dev, "register rtc --> %d\n", status);
+		goto fail0;
+	}
+	ds1305->rtc = rtc;
+
+	/* Maybe set up alarm IRQ; be ready to handle it triggering right
+	 * away.  NOTE that we don't share this.  The signal is active low,
+	 * and we can't ack it before a SPI message delay.  We temporarily
+	 * disable the IRQ until it's acked, which lets us work with more
+	 * IRQ trigger modes (not all IRQ controllers can do falling edge).
+	 */
+	if (spi->irq) {
+		INIT_WORK(&ds1305->work, ds1305_work);
+		status = request_irq(spi->irq, ds1305_irq,
+				0, dev_name(&rtc->dev), ds1305);
+		if (status < 0) {
+			dev_dbg(&spi->dev, "request_irq %d --> %d\n",
+					spi->irq, status);
+			goto fail1;
+		}
+	}
+
+	/* export NVRAM */
+	status = sysfs_create_bin_file(&spi->dev.kobj, &nvram);
+	if (status < 0) {
+		dev_dbg(&spi->dev, "register nvram --> %d\n", status);
+		goto fail2;
+	}
+
+	return 0;
+
+fail2:
+	free_irq(spi->irq, ds1305);
+fail1:
+	rtc_device_unregister(rtc);
+fail0:
+	kfree(ds1305);
+	return status;
+}
+
+static int __devexit ds1305_remove(struct spi_device *spi)
+{
+	struct ds1305	*ds1305 = spi_get_drvdata(spi);
+
+	sysfs_remove_bin_file(&spi->dev.kobj, &nvram);
+
+	/* carefully shut down irq and workqueue, if present */
+	if (spi->irq) {
+		set_bit(FLAG_EXITING, &ds1305->flags);
+		free_irq(spi->irq, ds1305);
+		flush_scheduled_work();
+	}
+
+	rtc_device_unregister(ds1305->rtc);
+	spi_set_drvdata(spi, NULL);
+	kfree(ds1305);
+	return 0;
+}
+
+static struct spi_driver ds1305_driver = {
+	.driver.name	= "rtc-ds1305",
+	.driver.owner	= THIS_MODULE,
+	.probe		= ds1305_probe,
+	.remove		= __devexit_p(ds1305_remove),
+	/* REVISIT add suspend/resume */
+};
+
+static int __init ds1305_init(void)
+{
+	return spi_register_driver(&ds1305_driver);
+}
+module_init(ds1305_init);
+
+static void __exit ds1305_exit(void)
+{
+	spi_unregister_driver(&ds1305_driver);
+}
+module_exit(ds1305_exit);
+
+MODULE_DESCRIPTION("RTC driver for DS1305 and DS1306 chips");
+MODULE_LICENSE("GPL");
diff --git a/include/linux/spi/ds1305.h b/include/linux/spi/ds1305.h
new file mode 100644
index 000000000000..287ec830eab7
--- /dev/null
+++ b/include/linux/spi/ds1305.h
@@ -0,0 +1,35 @@
+#ifndef __LINUX_SPI_DS1305_H
+#define __LINUX_SPI_DS1305_H
+
+/*
+ * One-time configuration for ds1305 and ds1306 RTC chips.
+ *
+ * Put a pointer to this in spi_board_info.platform_data if you want to
+ * be sure that Linux (re)initializes this as needed ... after losing
+ * backup power, and potentially on the first boot.
+ */
+struct ds1305_platform_data {
+
+	/* Trickle charge configuration:  it's OK to leave out the MAGIC
+	 * bitmask; mask in either DS1 or DS2, and then one of 2K/4k/8K.
+	 */
+#define DS1305_TRICKLE_MAGIC	0xa0
+#define DS1305_TRICKLE_DS2	0x08	/* two diodes */
+#define DS1305_TRICKLE_DS1	0x04	/* one diode */
+#define DS1305_TRICKLE_2K	0x01	/* 2 KOhm resistance */
+#define DS1305_TRICKLE_4K	0x02	/* 4 KOhm resistance */
+#define DS1305_TRICKLE_8K	0x03	/* 8 KOhm resistance */
+	u8	trickle;
+
+	/* set only on ds1306 parts */
+	bool	is_ds1306;
+
+	/* ds1306 only:  enable 1 Hz output */
+	bool	en_1hz;
+
+	/* REVISIT:  the driver currently expects nINT0 to be wired
+	 * as the alarm IRQ.  ALM1 may also need to be set up ...
+	 */
+};
+
+#endif /* __LINUX_SPI_DS1305_H */

From d3de851a445123f24ad8ece18662014b5e8a8b4e Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Wed, 23 Jul 2008 21:30:37 -0700
Subject: [PATCH 0432/1171] rtc: BCD codeshrink

This updates <linux/bcd.h> to define the key routines as constant
functions, which the macros will then call.  Newer code can now call
bcd2bin() instead of SCREAMING BCD2BIN() TO THE FOUR WINDS.

This lets each driver shrink their codespace by using N function calls to
a single (global) copy of those routines, instead of N inlined copies of
these functions per driver.

These routines aren't used in speed-critical code.  Almost all callers are
in the RTC framework.  Typical per-driver savings is near 300 bytes.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Acked-by: Adrian Bunk <bunk@kernel.org>
Cc: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/bcd.h |  9 +++++++--
 lib/Makefile        |  2 +-
 lib/bcd.c           | 14 ++++++++++++++
 3 files changed, 22 insertions(+), 3 deletions(-)
 create mode 100644 lib/bcd.c

diff --git a/include/linux/bcd.h b/include/linux/bcd.h
index c545308125b0..7ac518e3c152 100644
--- a/include/linux/bcd.h
+++ b/include/linux/bcd.h
@@ -10,8 +10,13 @@
 #ifndef _BCD_H
 #define _BCD_H
 
-#define BCD2BIN(val)	(((val) & 0x0f) + ((val)>>4)*10)
-#define BIN2BCD(val)	((((val)/10)<<4) + (val)%10)
+#include <linux/compiler.h>
+
+unsigned bcd2bin(unsigned char val) __attribute_const__;
+unsigned char bin2bcd(unsigned val) __attribute_const__;
+
+#define BCD2BIN(val)	bcd2bin(val)
+#define BIN2BCD(val)	bin2bcd(val)
 
 /* backwards compat */
 #define BCD_TO_BIN(val) ((val)=BCD2BIN(val))
diff --git a/lib/Makefile b/lib/Makefile
index 818c4d455518..9085ad6fa53d 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -18,7 +18,7 @@ lib-$(CONFIG_SMP) += cpumask.o
 
 lib-y	+= kobject.o kref.o klist.o
 
-obj-y += div64.o sort.o parser.o halfmd4.o debug_locks.o random32.o \
+obj-y += bcd.o div64.o sort.o parser.o halfmd4.o debug_locks.o random32.o \
 	 bust_spinlocks.o hexdump.o kasprintf.o bitmap.o scatterlist.o
 
 ifeq ($(CONFIG_DEBUG_KOBJECT),y)
diff --git a/lib/bcd.c b/lib/bcd.c
new file mode 100644
index 000000000000..d74257fd0fe7
--- /dev/null
+++ b/lib/bcd.c
@@ -0,0 +1,14 @@
+#include <linux/bcd.h>
+#include <linux/module.h>
+
+unsigned bcd2bin(unsigned char val)
+{
+	return (val & 0x0f) + (val >> 4) * 10;
+}
+EXPORT_SYMBOL(bcd2bin);
+
+unsigned char bin2bcd(unsigned val)
+{
+	return ((val / 10) << 4) + val % 10;
+}
+EXPORT_SYMBOL(bin2bcd);

From 71fc822455ccb63a66be0b6e97a415aceb0062c6 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Wed, 23 Jul 2008 21:30:38 -0700
Subject: [PATCH 0433/1171] rtc: rtc-omap footprint shrinkage

Shrink the runtime footprint of the OMAP1 RTC driver a bunch by removing
some old hacks and switching to platform_driver_probe().

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/rtc/rtc-omap.c | 21 ++++-----------------
 1 file changed, 4 insertions(+), 17 deletions(-)

diff --git a/drivers/rtc/rtc-omap.c b/drivers/rtc/rtc-omap.c
index eb23d8423f42..8876605d4d4b 100644
--- a/drivers/rtc/rtc-omap.c
+++ b/drivers/rtc/rtc-omap.c
@@ -92,18 +92,6 @@
 #define rtc_write(val, addr)	omap_writeb(val, OMAP_RTC_BASE + (addr))
 
 
-/* platform_bus isn't hotpluggable, so for static linkage it'd be safe
- * to get rid of probe() and remove() code ... too bad the driver struct
- * remembers probe(), that's about 25% of the runtime footprint!!
- */
-#ifndef	MODULE
-#undef	__devexit
-#undef	__devexit_p
-#define	__devexit	__exit
-#define	__devexit_p	__exit_p
-#endif
-
-
 /* we rely on the rtc framework to handle locking (rtc->ops_lock),
  * so the only other requirement is that register accesses which
  * require BUSY to be clear are made with IRQs locally disabled
@@ -324,7 +312,7 @@ static struct rtc_class_ops omap_rtc_ops = {
 static int omap_rtc_alarm;
 static int omap_rtc_timer;
 
-static int __devinit omap_rtc_probe(struct platform_device *pdev)
+static int __init omap_rtc_probe(struct platform_device *pdev)
 {
 	struct resource		*res, *mem;
 	struct rtc_device	*rtc;
@@ -440,7 +428,7 @@ fail:
 	return -EIO;
 }
 
-static int __devexit omap_rtc_remove(struct platform_device *pdev)
+static int __exit omap_rtc_remove(struct platform_device *pdev)
 {
 	struct rtc_device	*rtc = platform_get_drvdata(pdev);;
 
@@ -498,8 +486,7 @@ static void omap_rtc_shutdown(struct platform_device *pdev)
 
 MODULE_ALIAS("platform:omap_rtc");
 static struct platform_driver omap_rtc_driver = {
-	.probe		= omap_rtc_probe,
-	.remove		= __devexit_p(omap_rtc_remove),
+	.remove		= __exit_p(omap_rtc_remove),
 	.suspend	= omap_rtc_suspend,
 	.resume		= omap_rtc_resume,
 	.shutdown	= omap_rtc_shutdown,
@@ -511,7 +498,7 @@ static struct platform_driver omap_rtc_driver = {
 
 static int __init rtc_init(void)
 {
-	return platform_driver_register(&omap_rtc_driver);
+	return platform_driver_probe(&omap_rtc_driver, omap_rtc_probe);
 }
 module_init(rtc_init);
 

From 02bb584f3b1cfc8188522a4d2c8881b65073a4f1 Mon Sep 17 00:00:00 2001
From: Wolfram Sang <w.sang@pengutronix.de>
Date: Wed, 23 Jul 2008 21:30:39 -0700
Subject: [PATCH 0434/1171] rtc: convert the PCF8583 driver to the new I2C
 style framework with device_ids

Convert the PCF8583 driver to the new I2C style framework with device_ids

Signed-off-by: Juergen Beisert <j.beisert@pengutronix.de>
Signed-off-by: Wolfram Sang <w.sang@pengutronix.de>
Signed-off-by: Alessandro Zummo <a.zummo@towertech.it>
Cc: David Brownell <david-b@pacbell.net>
Acked-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/rtc/rtc-pcf8583.c | 135 ++++++++++++--------------------------
 1 file changed, 42 insertions(+), 93 deletions(-)

diff --git a/drivers/rtc/rtc-pcf8583.c b/drivers/rtc/rtc-pcf8583.c
index 3d09d8f0b1f0..d388c662bf4b 100644
--- a/drivers/rtc/rtc-pcf8583.c
+++ b/drivers/rtc/rtc-pcf8583.c
@@ -2,6 +2,7 @@
  *  drivers/rtc/rtc-pcf8583.c
  *
  *  Copyright (C) 2000 Russell King
+ *  Copyright (C) 2008 Wolfram Sang & Juergen Beisert, Pengutronix
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -14,7 +15,6 @@
 #include <linux/module.h>
 #include <linux/i2c.h>
 #include <linux/slab.h>
-#include <linux/string.h>
 #include <linux/rtc.h>
 #include <linux/init.h>
 #include <linux/errno.h>
@@ -27,7 +27,6 @@ struct rtc_mem {
 };
 
 struct pcf8583 {
-	struct i2c_client client;
 	struct rtc_device *rtc;
 	unsigned char ctrl;
 };
@@ -40,10 +39,6 @@ struct pcf8583 {
 #define CTRL_ALARM	0x02
 #define CTRL_TIMER	0x01
 
-static const unsigned short normal_i2c[] = { 0x50, I2C_CLIENT_END };
-
-/* Module parameters */
-I2C_CLIENT_INSMOD;
 
 static struct i2c_driver pcf8583_driver;
 
@@ -269,107 +264,61 @@ static const struct rtc_class_ops pcf8583_rtc_ops = {
 	.set_time	= pcf8583_rtc_set_time,
 };
 
-static int pcf8583_probe(struct i2c_adapter *adap, int addr, int kind);
-
-static int pcf8583_attach(struct i2c_adapter *adap)
+static int pcf8583_probe(struct i2c_client *client,
+				const struct i2c_device_id *id)
 {
-	return i2c_probe(adap, &addr_data, pcf8583_probe);
+	struct pcf8583 *pcf8583;
+	int err;
+
+	if (!i2c_check_functionality(client->adapter, I2C_FUNC_I2C))
+		return -ENODEV;
+
+	pcf8583 = kzalloc(sizeof(struct pcf8583), GFP_KERNEL);
+	if (!pcf8583)
+		return -ENOMEM;
+
+	pcf8583->rtc = rtc_device_register(pcf8583_driver.driver.name,
+			&client->dev, &pcf8583_rtc_ops, THIS_MODULE);
+
+	if (IS_ERR(pcf8583->rtc)) {
+		err = PTR_ERR(pcf8583->rtc);
+		goto exit_kfree;
+	}
+
+	i2c_set_clientdata(client, pcf8583);
+	return 0;
+
+exit_kfree:
+	kfree(pcf8583);
+	return err;
 }
 
-static int pcf8583_detach(struct i2c_client *client)
+static int __devexit pcf8583_remove(struct i2c_client *client)
 {
-	int err;
-	struct pcf8583 *pcf = i2c_get_clientdata(client);
-	struct rtc_device *rtc = pcf->rtc;
+	struct pcf8583 *pcf8583 = i2c_get_clientdata(client);
 
-	if (rtc)
-		rtc_device_unregister(rtc);
-
-	if ((err = i2c_detach_client(client)))
-		return err;
-
-	kfree(pcf);
+	if (pcf8583->rtc)
+		rtc_device_unregister(pcf8583->rtc);
+	kfree(pcf8583);
 	return 0;
 }
 
+static const struct i2c_device_id pcf8583_id[] = {
+	{ "pcf8583", 0 },
+	{ }
+};
+MODULE_DEVICE_TABLE(i2c, pcf8583_id);
+
 static struct i2c_driver pcf8583_driver = {
 	.driver = {
 		.name	= "pcf8583",
+		.owner	= THIS_MODULE,
 	},
-	.id		= I2C_DRIVERID_PCF8583,
-	.attach_adapter	= pcf8583_attach,
-	.detach_client	= pcf8583_detach,
+	.probe		= pcf8583_probe,
+	.remove		= __devexit_p(pcf8583_remove),
+	.id_table	= pcf8583_id,
 };
 
-static int pcf8583_probe(struct i2c_adapter *adap, int addr, int kind)
-{
-	struct pcf8583 *pcf;
-	struct i2c_client *client;
-	struct rtc_device *rtc;
-	unsigned char buf[1], ad[1] = { 0 };
-	int err;
-	struct i2c_msg msgs[2] = {
-		{
-			.addr = addr,
-			.flags = 0,
-			.len = 1,
-			.buf = ad,
-		}, {
-			.addr = addr,
-			.flags = I2C_M_RD,
-			.len = 1,
-			.buf = buf,
-		}
-	};
-
-	if (!i2c_check_functionality(adap, I2C_FUNC_I2C))
-		return 0;
-
-	pcf = kzalloc(sizeof(*pcf), GFP_KERNEL);
-	if (!pcf)
-		return -ENOMEM;
-
-	client = &pcf->client;
-
-	client->addr		= addr;
-	client->adapter	= adap;
-	client->driver	= &pcf8583_driver;
-
-	strlcpy(client->name, pcf8583_driver.driver.name, I2C_NAME_SIZE);
-
-	if (i2c_transfer(client->adapter, msgs, 2) != 2) {
-		err = -EIO;
-		goto exit_kfree;
-	}
-
-	err = i2c_attach_client(client);
-
-	if (err)
-		goto exit_kfree;
-
-	rtc = rtc_device_register(pcf8583_driver.driver.name, &client->dev,
-				  &pcf8583_rtc_ops, THIS_MODULE);
-
-	if (IS_ERR(rtc)) {
-		err = PTR_ERR(rtc);
-		goto exit_detach;
-	}
-
-	pcf->rtc = rtc;
-	i2c_set_clientdata(client, pcf);
-	set_ctrl(client, buf[0]);
-
-	return 0;
-
-exit_detach:
-	i2c_detach_client(client);
-
-exit_kfree:
-	kfree(pcf);
-
-	return err;
-}
-
 static __init int pcf8583_init(void)
 {
 	return i2c_add_driver(&pcf8583_driver);

From c68d07b2da54c941bb36c9d6d35fe8f263ee10ef Mon Sep 17 00:00:00 2001
From: "Carlos R. Mafra" <crmafra@ift.unesp.br>
Date: Wed, 23 Jul 2008 21:30:40 -0700
Subject: [PATCH 0435/1171] rtc: remove and clarify unneeded externs

When CONFIG_HPET_EMULATE_RTC is defined the external declaration of
hpet_rtc_interrupt is redundant due to the inclusion of hpet.h.

When !CONFIG_HPET_EMULATE_RTC we make it clear that hpet_rtc_interrupt is
not used by defining it to return zero.

Signed-off-by: Carlos R. Mafra <crmafra@ift.unesp.br>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/rtc.c     | 2 --
 drivers/rtc/rtc-cmos.c | 5 ++++-
 2 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/char/rtc.c b/drivers/char/rtc.c
index d1569a0d0506..dbefbb30ed44 100644
--- a/drivers/char/rtc.c
+++ b/drivers/char/rtc.c
@@ -121,8 +121,6 @@ static irqreturn_t hpet_rtc_interrupt(int irq, void *dev_id)
 	return 0;
 }
 #endif
-#else
-extern irqreturn_t hpet_rtc_interrupt(int irq, void *dev_id);
 #endif
 
 /*
diff --git a/drivers/rtc/rtc-cmos.c b/drivers/rtc/rtc-cmos.c
index d7bb9bac71df..94b89a2d9c2e 100644
--- a/drivers/rtc/rtc-cmos.c
+++ b/drivers/rtc/rtc-cmos.c
@@ -52,7 +52,10 @@
 #define hpet_rtc_timer_init() 			do { } while (0)
 #define hpet_register_irq_handler(h) 		0
 #define hpet_unregister_irq_handler(h)		do { } while (0)
-extern irqreturn_t hpet_rtc_interrupt(int irq, void *dev_id);
+static irqreturn_t hpet_rtc_interrupt(int irq, void *dev_id)
+{
+	return 0;
+}
 #endif
 
 struct cmos_rtc {

From 35d3fdd5f304c06654c940921fc045c60df34693 Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Wed, 23 Jul 2008 21:30:43 -0700
Subject: [PATCH 0436/1171] rtc-cmos: improve HPET IRQ glue

Resolve http://bugzilla.kernel.org/show_bug.cgi?id=11051 and other bugs
related to the way the HPET glue code in rtc-cmos was incomplete and
inconsistent:

 * Switch the approach so that the basic driver code flow isn't
   changed by having HPET ... instead, just have HPET shadow the
   RTC_CONTROL irq enables and RTC_FREQ_SELECT data.  It's only
   coping with IRQ thievery, after all.

 * Do that consistently (!!) to avoid problems when the HPET code
   is out of sync with the real RTC intent.  Examples include:

   - cmos_procfs(), which now reports correct data

   - cmos_irq_set_state() ... also removing the previous PIE_{ON,OFF}
     ioctl support so only one code path manages "periodic" IRQs

   - cmos_do_shutdown() ... currently a "just in case" change.

   - cmos_suspend() and cmos_resume() ... also handling a bug that
     was specific to HPET's IRQ thievery, where the alarm wasn't
     disabled after waking the system

 * Always call that HPET code under the RTC spinlock (it doesn't do
   its own locking)

Also clean up the HPET glue:

 * Add some comments explaining what's going on.

 * Switch to having just one #ifdef for the HPET glue, and inline
   functions (not #defines) to avoid some compiler warnings.

 * Have the probe message also report when HPET IRQs are involved

This still leaves various holes in the HPET glue, like the emulated update
IRQs being out of sync with the RTC, alarms never using day or month
matches, and many extra IRQs (at 64 Hz).

[akpm@linux-foundation.org: fix build]
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Tomas Janousek <tomi@nomi.cz>
Cc: Bernhard Walle <bwalle@suse.de>
Cc: Carlos R. Mafra <crmafra@ift.unesp.br>
Acked-by: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/rtc/rtc-cmos.c | 193 ++++++++++++++++++++++++++++-------------
 1 file changed, 131 insertions(+), 62 deletions(-)

diff --git a/drivers/rtc/rtc-cmos.c b/drivers/rtc/rtc-cmos.c
index 94b89a2d9c2e..e9984650ea95 100644
--- a/drivers/rtc/rtc-cmos.c
+++ b/drivers/rtc/rtc-cmos.c
@@ -36,28 +36,9 @@
 #include <linux/platform_device.h>
 #include <linux/mod_devicetable.h>
 
-#ifdef CONFIG_HPET_EMULATE_RTC
-#include <asm/hpet.h>
-#endif
-
 /* this is for "generic access to PC-style RTC" using CMOS_READ/CMOS_WRITE */
 #include <asm-generic/rtc.h>
 
-#ifndef CONFIG_HPET_EMULATE_RTC
-#define is_hpet_enabled()			0
-#define hpet_set_alarm_time(hrs, min, sec) 	do { } while (0)
-#define hpet_set_periodic_freq(arg) 		0
-#define hpet_mask_rtc_irq_bit(arg) 		do { } while (0)
-#define hpet_set_rtc_irq_bit(arg) 		do { } while (0)
-#define hpet_rtc_timer_init() 			do { } while (0)
-#define hpet_register_irq_handler(h) 		0
-#define hpet_unregister_irq_handler(h)		do { } while (0)
-static irqreturn_t hpet_rtc_interrupt(int irq, void *dev_id)
-{
-	return 0;
-}
-#endif
-
 struct cmos_rtc {
 	struct rtc_device	*rtc;
 	struct device		*dev;
@@ -96,6 +77,72 @@ static inline int is_intr(u8 rtc_intr)
 
 /*----------------------------------------------------------------*/
 
+/* Much modern x86 hardware has HPETs (10+ MHz timers) which, because
+ * many BIOS programmers don't set up "sane mode" IRQ routing, are mostly
+ * used in a broken "legacy replacement" mode.  The breakage includes
+ * HPET #1 hijacking the IRQ for this RTC, and being unavailable for
+ * other (better) use.
+ *
+ * When that broken mode is in use, platform glue provides a partial
+ * emulation of hardware RTC IRQ facilities using HPET #1.  We don't
+ * want to use HPET for anything except those IRQs though...
+ */
+#ifdef CONFIG_HPET_EMULATE_RTC
+#include <asm/hpet.h>
+#else
+
+static inline int is_hpet_enabled(void)
+{
+	return 0;
+}
+
+static inline int hpet_mask_rtc_irq_bit(unsigned long mask)
+{
+	return 0;
+}
+
+static inline int hpet_set_rtc_irq_bit(unsigned long mask)
+{
+	return 0;
+}
+
+static inline int
+hpet_set_alarm_time(unsigned char hrs, unsigned char min, unsigned char sec)
+{
+	return 0;
+}
+
+static inline int hpet_set_periodic_freq(unsigned long freq)
+{
+	return 0;
+}
+
+static inline int hpet_rtc_dropped_irq(void)
+{
+	return 0;
+}
+
+static inline int hpet_rtc_timer_init(void)
+{
+	return 0;
+}
+
+extern irq_handler_t hpet_rtc_interrupt;
+
+static inline int hpet_register_irq_handler(irq_handler_t handler)
+{
+	return 0;
+}
+
+static inline int hpet_unregister_irq_handler(irq_handler_t handler)
+{
+	return 0;
+}
+
+#endif
+
+/*----------------------------------------------------------------*/
+
 static int cmos_read_time(struct device *dev, struct rtc_time *t)
 {
 	/* REVISIT:  if the clock has a "century" register, use
@@ -216,13 +263,14 @@ static int cmos_set_alarm(struct device *dev, struct rtc_wkalrm *t)
 	sec = t->time.tm_sec;
 	sec = (sec < 60) ? BIN2BCD(sec) : 0xff;
 
-	hpet_set_alarm_time(t->time.tm_hour, t->time.tm_min, t->time.tm_sec);
 	spin_lock_irq(&rtc_lock);
 
 	/* next rtc irq must not be from previous alarm setting */
 	rtc_control = CMOS_READ(RTC_CONTROL);
 	rtc_control &= ~RTC_AIE;
 	CMOS_WRITE(rtc_control, RTC_CONTROL);
+	hpet_mask_rtc_irq_bit(RTC_AIE);
+
 	rtc_intr = CMOS_READ(RTC_INTR_FLAGS);
 	rtc_intr &= (rtc_control & RTC_IRQMASK) | RTC_IRQF;
 	if (is_intr(rtc_intr))
@@ -240,9 +288,16 @@ static int cmos_set_alarm(struct device *dev, struct rtc_wkalrm *t)
 			CMOS_WRITE(mon, cmos->mon_alrm);
 	}
 
+	/* FIXME the HPET alarm glue currently ignores day_alrm
+	 * and mon_alrm ...
+	 */
+	hpet_set_alarm_time(t->time.tm_hour, t->time.tm_min, t->time.tm_sec);
+
 	if (t->enabled) {
 		rtc_control |= RTC_AIE;
 		CMOS_WRITE(rtc_control, RTC_CONTROL);
+		hpet_set_rtc_irq_bit(RTC_AIE);
+
 		rtc_intr = CMOS_READ(RTC_INTR_FLAGS);
 		rtc_intr &= (rtc_control & RTC_IRQMASK) | RTC_IRQF;
 		if (is_intr(rtc_intr))
@@ -270,8 +325,8 @@ static int cmos_irq_set_freq(struct device *dev, int freq)
 	f = 16 - f;
 
 	spin_lock_irqsave(&rtc_lock, flags);
-	if (!hpet_set_periodic_freq(freq))
-		CMOS_WRITE(RTC_REF_CLCK_32KHZ | f, RTC_FREQ_SELECT);
+	hpet_set_periodic_freq(freq);
+	CMOS_WRITE(RTC_REF_CLCK_32KHZ | f, RTC_FREQ_SELECT);
 	spin_unlock_irqrestore(&rtc_lock, flags);
 
 	return 0;
@@ -289,11 +344,13 @@ static int cmos_irq_set_state(struct device *dev, int enabled)
 	spin_lock_irqsave(&rtc_lock, flags);
 	rtc_control = CMOS_READ(RTC_CONTROL);
 
-	if (enabled)
+	if (enabled) {
 		rtc_control |= RTC_PIE;
-	else
+		hpet_set_rtc_irq_bit(RTC_PIE);
+	} else {
 		rtc_control &= ~RTC_PIE;
-
+		hpet_mask_rtc_irq_bit(RTC_PIE);
+	}
 	CMOS_WRITE(rtc_control, RTC_CONTROL);
 
 	rtc_intr = CMOS_READ(RTC_INTR_FLAGS);
@@ -319,11 +376,10 @@ cmos_rtc_ioctl(struct device *dev, unsigned int cmd, unsigned long arg)
 	case RTC_AIE_ON:
 	case RTC_UIE_OFF:
 	case RTC_UIE_ON:
-	case RTC_PIE_OFF:
-	case RTC_PIE_ON:
 		if (!is_valid_irq(cmos->irq))
 			return -EINVAL;
 		break;
+	/* PIE ON/OFF is handled by cmos_irq_set_state() */
 	default:
 		return -ENOIOCTLCMD;
 	}
@@ -347,17 +403,8 @@ cmos_rtc_ioctl(struct device *dev, unsigned int cmd, unsigned long arg)
 		rtc_control |= RTC_UIE;
 		hpet_set_rtc_irq_bit(RTC_UIE);
 		break;
-	case RTC_PIE_OFF:	/* periodic off */
-		rtc_control &= ~RTC_PIE;
-		hpet_mask_rtc_irq_bit(RTC_PIE);
-		break;
-	case RTC_PIE_ON:	/* periodic on */
-		rtc_control |= RTC_PIE;
-		hpet_set_rtc_irq_bit(RTC_PIE);
-		break;
 	}
-	if (!is_hpet_enabled())
-		CMOS_WRITE(rtc_control, RTC_CONTROL);
+	CMOS_WRITE(rtc_control, RTC_CONTROL);
 
 	rtc_intr = CMOS_READ(RTC_INTR_FLAGS);
 	rtc_intr &= (rtc_control & RTC_IRQMASK) | RTC_IRQF;
@@ -505,18 +552,19 @@ static irqreturn_t cmos_interrupt(int irq, void *p)
 	u8		rtc_control;
 
 	spin_lock(&rtc_lock);
-	/*
-	 * In this case it is HPET RTC interrupt handler
-	 * calling us, with the interrupt information
-	 * passed as arg1, instead of irq.
+
+	/* When the HPET interrupt handler calls us, the interrupt
+	 * status is passed as arg1 instead of the irq number.  But
+	 * always clear irq status, even when HPET is in the way.
+	 *
+	 * Note that HPET and RTC are almost certainly out of phase,
+	 * giving different IRQ status ...
 	 */
+	irqstat = CMOS_READ(RTC_INTR_FLAGS);
+	rtc_control = CMOS_READ(RTC_CONTROL);
 	if (is_hpet_enabled())
 		irqstat = (unsigned long)irq & 0xF0;
-	else {
-		irqstat = CMOS_READ(RTC_INTR_FLAGS);
-		rtc_control = CMOS_READ(RTC_CONTROL);
-		irqstat &= (rtc_control & RTC_IRQMASK) | RTC_IRQF;
-	}
+	irqstat &= (rtc_control & RTC_IRQMASK) | RTC_IRQF;
 
 	/* All Linux RTC alarms should be treated as if they were oneshot.
 	 * Similar code may be needed in system wakeup paths, in case the
@@ -526,6 +574,8 @@ static irqreturn_t cmos_interrupt(int irq, void *p)
 		rtc_control = CMOS_READ(RTC_CONTROL);
 		rtc_control &= ~RTC_AIE;
 		CMOS_WRITE(rtc_control, RTC_CONTROL);
+		hpet_mask_rtc_irq_bit(RTC_AIE);
+
 		CMOS_READ(RTC_INTR_FLAGS);
 	}
 	spin_unlock(&rtc_lock);
@@ -632,8 +682,8 @@ cmos_do_probe(struct device *dev, struct resource *ports, int rtc_irq)
 	 * do something about other clock frequencies.
 	 */
 	cmos_rtc.rtc->irq_freq = 1024;
-	if (!hpet_set_periodic_freq(cmos_rtc.rtc->irq_freq))
-		CMOS_WRITE(RTC_REF_CLCK_32KHZ | 0x06, RTC_FREQ_SELECT);
+	hpet_set_periodic_freq(cmos_rtc.rtc->irq_freq);
+	CMOS_WRITE(RTC_REF_CLCK_32KHZ | 0x06, RTC_FREQ_SELECT);
 
 	/* disable irqs.
 	 *
@@ -643,6 +693,8 @@ cmos_do_probe(struct device *dev, struct resource *ports, int rtc_irq)
 	rtc_control = CMOS_READ(RTC_CONTROL);
 	rtc_control &= ~(RTC_PIE | RTC_AIE | RTC_UIE);
 	CMOS_WRITE(rtc_control, RTC_CONTROL);
+	hpet_mask_rtc_irq_bit(RTC_PIE | RTC_AIE | RTC_UIE);
+
 	CMOS_READ(RTC_INTR_FLAGS);
 
 	spin_unlock_irq(&rtc_lock);
@@ -690,7 +742,7 @@ cmos_do_probe(struct device *dev, struct resource *ports, int rtc_irq)
 		goto cleanup2;
 	}
 
-	pr_info("%s: alarms up to one %s%s\n",
+	pr_info("%s: alarms up to one %s%s%s\n",
 			cmos_rtc.rtc->dev.bus_id,
 			is_valid_irq(rtc_irq)
 				?  (cmos_rtc.mon_alrm
@@ -698,8 +750,8 @@ cmos_do_probe(struct device *dev, struct resource *ports, int rtc_irq)
 					: (cmos_rtc.day_alrm
 						? "month" : "day"))
 				: "no",
-			cmos_rtc.century ? ", y3k" : ""
-			);
+			cmos_rtc.century ? ", y3k" : "",
+			is_hpet_enabled() ? ", hpet irqs" : "");
 
 	return 0;
 
@@ -720,8 +772,10 @@ static void cmos_do_shutdown(void)
 
 	spin_lock_irq(&rtc_lock);
 	rtc_control = CMOS_READ(RTC_CONTROL);
-	rtc_control &= ~(RTC_PIE|RTC_AIE|RTC_UIE);
+	rtc_control &= ~RTC_IRQMASK;
 	CMOS_WRITE(rtc_control, RTC_CONTROL);
+	hpet_mask_rtc_irq_bit(RTC_IRQMASK);
+
 	CMOS_READ(RTC_INTR_FLAGS);
 	spin_unlock_irq(&rtc_lock);
 }
@@ -764,12 +818,16 @@ static int cmos_suspend(struct device *dev, pm_message_t mesg)
 	cmos->suspend_ctrl = tmp = CMOS_READ(RTC_CONTROL);
 	if (tmp & (RTC_PIE|RTC_AIE|RTC_UIE)) {
 		unsigned char	irqstat;
+		unsigned char	mask;
 
 		if (do_wake)
-			tmp &= ~(RTC_PIE|RTC_UIE);
+			mask = RTC_IRQMASK & ~RTC_AIE;
 		else
-			tmp &= ~(RTC_PIE|RTC_AIE|RTC_UIE);
+			mask = RTC_IRQMASK;
+		tmp &= ~mask;
 		CMOS_WRITE(tmp, RTC_CONTROL);
+		hpet_mask_rtc_irq_bit(mask);
+
 		irqstat = CMOS_READ(RTC_INTR_FLAGS);
 		irqstat &= (tmp & RTC_IRQMASK) | RTC_IRQF;
 		if (is_intr(irqstat))
@@ -799,7 +857,8 @@ static int cmos_resume(struct device *dev)
 	unsigned char	tmp = cmos->suspend_ctrl;
 
 	/* re-enable any irqs previously active */
-	if (tmp & (RTC_PIE|RTC_AIE|RTC_UIE)) {
+	if (tmp & RTC_IRQMASK) {
+		unsigned char	mask;
 
 		if (cmos->enabled_wake) {
 			if (cmos->wake_off)
@@ -810,18 +869,28 @@ static int cmos_resume(struct device *dev)
 		}
 
 		spin_lock_irq(&rtc_lock);
-		CMOS_WRITE(tmp, RTC_CONTROL);
-		tmp = CMOS_READ(RTC_INTR_FLAGS);
-		tmp &= (cmos->suspend_ctrl & RTC_IRQMASK) | RTC_IRQF;
-		if (is_intr(tmp))
-			rtc_update_irq(cmos->rtc, 1, tmp);
+		do {
+			CMOS_WRITE(tmp, RTC_CONTROL);
+			hpet_set_rtc_irq_bit(tmp & RTC_IRQMASK);
+
+			mask = CMOS_READ(RTC_INTR_FLAGS);
+			mask &= (tmp & RTC_IRQMASK) | RTC_IRQF;
+			if (!is_intr(mask))
+				break;
+
+			/* force one-shot behavior if HPET blocked
+			 * the wake alarm's irq
+			 */
+			rtc_update_irq(cmos->rtc, 1, mask);
+			tmp &= ~RTC_AIE;
+			hpet_mask_rtc_irq_bit(RTC_AIE);
+		} while (mask & RTC_AIE);
 		spin_unlock_irq(&rtc_lock);
 	}
 
 	pr_debug("%s: resume, ctrl %02x\n",
 			cmos_rtc.rtc->dev.bus_id,
-			cmos->suspend_ctrl);
-
+			tmp);
 
 	return 0;
 }

From 4cd0c5c40b64ef9fd94fb8382dade2fd28f2b935 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Wed, 23 Jul 2008 21:30:44 -0700
Subject: [PATCH 0437/1171] rtc: rtc-s3c: add __devexit and __devinit markers

Add the relevant __devinit and __devexit attributes to the rtc-s3c driver.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Acked-by: Alessandro Zummo <a.zummo@towertech.it>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/rtc/rtc-s3c.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/rtc/rtc-s3c.c b/drivers/rtc/rtc-s3c.c
index fed86e507fdf..b81ba7020d92 100644
--- a/drivers/rtc/rtc-s3c.c
+++ b/drivers/rtc/rtc-s3c.c
@@ -430,7 +430,7 @@ static void s3c_rtc_enable(struct platform_device *pdev, int en)
 	}
 }
 
-static int s3c_rtc_remove(struct platform_device *dev)
+static int __devexit s3c_rtc_remove(struct platform_device *dev)
 {
 	struct rtc_device *rtc = platform_get_drvdata(dev);
 
@@ -447,7 +447,7 @@ static int s3c_rtc_remove(struct platform_device *dev)
 	return 0;
 }
 
-static int s3c_rtc_probe(struct platform_device *pdev)
+static int __devinit s3c_rtc_probe(struct platform_device *pdev)
 {
 	struct rtc_device *rtc;
 	struct resource *res;
@@ -560,7 +560,7 @@ static int s3c_rtc_resume(struct platform_device *pdev)
 
 static struct platform_driver s3c2410_rtcdrv = {
 	.probe		= s3c_rtc_probe,
-	.remove		= s3c_rtc_remove,
+	.remove		= __devexit_p(s3c_rtc_remove),
 	.suspend	= s3c_rtc_suspend,
 	.resume		= s3c_rtc_resume,
 	.driver		= {

From 773be7ee97c11fbb6b8a912a58b268dbe8a6a3fe Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Wed, 23 Jul 2008 21:30:45 -0700
Subject: [PATCH 0438/1171] rtc: rtc-s3c: update IRQ handling

The rtc-s3c.c driver has been using its own ioctl() handling to deal with
alarm and periodic interrupts to handle what should now be done with the
rtc core code.

Change to using the .irq_set_freq and .irq_set_state driver entries and
remove the .ioctl handling.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Acked-by: Alessandro Zummo <a.zummo@towertech.it>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/rtc/rtc-s3c.c | 83 +++++++++----------------------------------
 1 file changed, 16 insertions(+), 67 deletions(-)

diff --git a/drivers/rtc/rtc-s3c.c b/drivers/rtc/rtc-s3c.c
index b81ba7020d92..54b1ebb01502 100644
--- a/drivers/rtc/rtc-s3c.c
+++ b/drivers/rtc/rtc-s3c.c
@@ -36,10 +36,8 @@ static struct resource *s3c_rtc_mem;
 static void __iomem *s3c_rtc_base;
 static int s3c_rtc_alarmno = NO_IRQ;
 static int s3c_rtc_tickno  = NO_IRQ;
-static int s3c_rtc_freq    = 1;
 
 static DEFINE_SPINLOCK(s3c_rtc_pie_lock);
-static unsigned int tick_count;
 
 /* IRQ Handlers */
 
@@ -55,7 +53,7 @@ static irqreturn_t s3c_rtc_tickirq(int irq, void *id)
 {
 	struct rtc_device *rdev = id;
 
-	rtc_update_irq(rdev, tick_count++, RTC_PF | RTC_IRQF);
+	rtc_update_irq(rdev, 1, RTC_PF | RTC_IRQF);
 	return IRQ_HANDLED;
 }
 
@@ -74,35 +72,37 @@ static void s3c_rtc_setaie(int to)
 	writeb(tmp, s3c_rtc_base + S3C2410_RTCALM);
 }
 
-static void s3c_rtc_setpie(int to)
+static int s3c_rtc_setpie(struct device *dev, int enabled)
 {
 	unsigned int tmp;
 
-	pr_debug("%s: pie=%d\n", __func__, to);
+	pr_debug("%s: pie=%d\n", __func__, enabled);
 
 	spin_lock_irq(&s3c_rtc_pie_lock);
 	tmp = readb(s3c_rtc_base + S3C2410_TICNT) & ~S3C2410_TICNT_ENABLE;
 
-	if (to)
+	if (enabled)
 		tmp |= S3C2410_TICNT_ENABLE;
 
 	writeb(tmp, s3c_rtc_base + S3C2410_TICNT);
 	spin_unlock_irq(&s3c_rtc_pie_lock);
+
+	return 0;
 }
 
-static void s3c_rtc_setfreq(int freq)
+static int s3c_rtc_setfreq(struct device *dev, int freq)
 {
 	unsigned int tmp;
 
 	spin_lock_irq(&s3c_rtc_pie_lock);
+
 	tmp = readb(s3c_rtc_base + S3C2410_TICNT) & S3C2410_TICNT_ENABLE;
-
-	s3c_rtc_freq = freq;
-
 	tmp |= (128 / freq)-1;
 
 	writeb(tmp, s3c_rtc_base + S3C2410_TICNT);
 	spin_unlock_irq(&s3c_rtc_pie_lock);
+
+	return 0;
 }
 
 /* Time read/write */
@@ -267,12 +267,7 @@ static int s3c_rtc_setalarm(struct device *dev, struct rtc_wkalrm *alrm)
 
 	writeb(alrm_en, base + S3C2410_RTCALM);
 
-	if (0) {
-		alrm_en = readb(base + S3C2410_RTCALM);
-		alrm_en &= ~S3C2410_RTCALM_ALMEN;
-		writeb(alrm_en, base + S3C2410_RTCALM);
-		disable_irq_wake(s3c_rtc_alarmno);
-	}
+	s3c_rtc_setaie(alrm->enabled);
 
 	if (alrm->enabled)
 		enable_irq_wake(s3c_rtc_alarmno);
@@ -282,59 +277,12 @@ static int s3c_rtc_setalarm(struct device *dev, struct rtc_wkalrm *alrm)
 	return 0;
 }
 
-static int s3c_rtc_ioctl(struct device *dev,
-			 unsigned int cmd, unsigned long arg)
-{
-	unsigned int ret = -ENOIOCTLCMD;
-
-	switch (cmd) {
-	case RTC_AIE_OFF:
-	case RTC_AIE_ON:
-		s3c_rtc_setaie((cmd == RTC_AIE_ON) ? 1 : 0);
-		ret = 0;
-		break;
-
-	case RTC_PIE_OFF:
-	case RTC_PIE_ON:
-		tick_count = 0;
-		s3c_rtc_setpie((cmd == RTC_PIE_ON) ? 1 : 0);
-		ret = 0;
-		break;
-
-	case RTC_IRQP_READ:
-		ret = put_user(s3c_rtc_freq, (unsigned long __user *)arg);
-		break;
-
-	case RTC_IRQP_SET:
-		if (!is_power_of_2(arg)) {
-			ret = -EINVAL;
-			goto exit;
-		}
-
-		pr_debug("s3c2410_rtc: setting frequency %ld\n", arg);
-
-		s3c_rtc_setfreq(arg);
-		ret = 0;
-		break;
-
-	case RTC_UIE_ON:
-	case RTC_UIE_OFF:
-		ret = -EINVAL;
-	}
-
- exit:
-	return ret;
-}
-
 static int s3c_rtc_proc(struct device *dev, struct seq_file *seq)
 {
 	unsigned int ticnt = readb(s3c_rtc_base + S3C2410_TICNT);
 
 	seq_printf(seq, "periodic_IRQ\t: %s\n",
 		     (ticnt & S3C2410_TICNT_ENABLE) ? "yes" : "no" );
-
-	seq_printf(seq, "periodic_freq\t: %d\n", s3c_rtc_freq);
-
 	return 0;
 }
 
@@ -374,7 +322,7 @@ static void s3c_rtc_release(struct device *dev)
 
 	/* do not clear AIE here, it may be needed for wake */
 
-	s3c_rtc_setpie(0);
+	s3c_rtc_setpie(dev, 0);
 	free_irq(s3c_rtc_alarmno, rtc_dev);
 	free_irq(s3c_rtc_tickno, rtc_dev);
 }
@@ -382,11 +330,12 @@ static void s3c_rtc_release(struct device *dev)
 static const struct rtc_class_ops s3c_rtcops = {
 	.open		= s3c_rtc_open,
 	.release	= s3c_rtc_release,
-	.ioctl		= s3c_rtc_ioctl,
 	.read_time	= s3c_rtc_gettime,
 	.set_time	= s3c_rtc_settime,
 	.read_alarm	= s3c_rtc_getalarm,
 	.set_alarm	= s3c_rtc_setalarm,
+	.irq_set_freq	= s3c_rtc_setfreq,
+	.irq_set_state	= s3c_rtc_setpie,
 	.proc	        = s3c_rtc_proc,
 };
 
@@ -437,7 +386,7 @@ static int __devexit s3c_rtc_remove(struct platform_device *dev)
 	platform_set_drvdata(dev, NULL);
 	rtc_device_unregister(rtc);
 
-	s3c_rtc_setpie(0);
+	s3c_rtc_setpie(&dev->dev, 0);
 	s3c_rtc_setaie(0);
 
 	iounmap(s3c_rtc_base);
@@ -504,7 +453,7 @@ static int __devinit s3c_rtc_probe(struct platform_device *pdev)
  	pr_debug("s3c2410_rtc: RTCCON=%02x\n",
 		 readb(s3c_rtc_base + S3C2410_RTCCON));
 
-	s3c_rtc_setfreq(s3c_rtc_freq);
+	s3c_rtc_setfreq(&pdev->dev, 1);
 
 	/* register RTC and exit */
 

From 449321b39f6c6ebfa15d6da24f134240bd51db29 Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Wed, 23 Jul 2008 21:30:46 -0700
Subject: [PATCH 0439/1171] rtc-at91rm9200: avoid spurious irqs

This fixes kernel http://bugzilla.kernel.org/show_bug.cgi?id=11112 (bogus
RTC update IRQs reported) for rtc-at91rm9200 by scrubbing old IRQ status
before enabling IRQs.

It also removes nonfunctional periodic IRQ support from this driver;
only update IRQs are reported, or provided by the hardware.

I suspect some other RTCs probably have versions of #11112; it's easy to
overlook, since most non-RTC drivers don't care about spurious IRQs:
they're not reported to userspace.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Report-by: W Unruh <unruh@physics.ubc.ca>
Cc: Andrew Victor <avictor.za@gmail.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/rtc/rtc-at91rm9200.c | 16 ++++++----------
 1 file changed, 6 insertions(+), 10 deletions(-)

diff --git a/drivers/rtc/rtc-at91rm9200.c b/drivers/rtc/rtc-at91rm9200.c
index 9c3db934cc24..cd32d05db773 100644
--- a/drivers/rtc/rtc-at91rm9200.c
+++ b/drivers/rtc/rtc-at91rm9200.c
@@ -171,8 +171,10 @@ static int at91_rtc_setalarm(struct device *dev, struct rtc_wkalrm *alrm)
 		| BIN2BCD(tm.tm_mday) << 24
 		| AT91_RTC_DATEEN | AT91_RTC_MTHEN);
 
-	if (alrm->enabled)
+	if (alrm->enabled) {
+		at91_sys_write(AT91_RTC_SCCR, AT91_RTC_ALARM);
 		at91_sys_write(AT91_RTC_IER, AT91_RTC_ALARM);
+	}
 
 	pr_debug("%s(): %4d-%02d-%02d %02d:%02d:%02d\n", __func__,
 		at91_alarm_year, tm.tm_mon, tm.tm_mday, tm.tm_hour,
@@ -191,28 +193,22 @@ static int at91_rtc_ioctl(struct device *dev, unsigned int cmd,
 
 	pr_debug("%s(): cmd=%08x, arg=%08lx.\n", __func__, cmd, arg);
 
+	/* important:  scrub old status before enabling IRQs */
 	switch (cmd) {
 	case RTC_AIE_OFF:	/* alarm off */
 		at91_sys_write(AT91_RTC_IDR, AT91_RTC_ALARM);
 		break;
 	case RTC_AIE_ON:	/* alarm on */
+		at91_sys_write(AT91_RTC_SCCR, AT91_RTC_ALARM);
 		at91_sys_write(AT91_RTC_IER, AT91_RTC_ALARM);
 		break;
 	case RTC_UIE_OFF:	/* update off */
-	case RTC_PIE_OFF:	/* periodic off */
 		at91_sys_write(AT91_RTC_IDR, AT91_RTC_SECEV);
 		break;
 	case RTC_UIE_ON:	/* update on */
-	case RTC_PIE_ON:	/* periodic on */
+		at91_sys_write(AT91_RTC_SCCR, AT91_RTC_SECEV);
 		at91_sys_write(AT91_RTC_IER, AT91_RTC_SECEV);
 		break;
-	case RTC_IRQP_READ:	/* read periodic alarm frequency */
-		ret = put_user(AT91_RTC_FREQ, (unsigned long *) arg);
-		break;
-	case RTC_IRQP_SET:	/* set periodic alarm frequency */
-		if (arg != AT91_RTC_FREQ)
-			ret = -EINVAL;
-		break;
 	default:
 		ret = -ENOIOCTLCMD;
 		break;

From 7e2a31da854dcf8324012a83a31b40bc11e52589 Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Wed, 23 Jul 2008 21:30:47 -0700
Subject: [PATCH 0440/1171] rtc-cmos: avoid spurious irqs

This fixes kernel http://bugzilla.kernel.org/show_bug.cgi?id=11112 (bogus
RTC update IRQs reported) for rtc-cmos, in two ways:

  - When HPET is stealing the IRQs, use the first IRQ to grab
    the seconds counter which will be monitored (instead of
    using whatever was previously in that memory);

  - In sane IRQ handling modes, scrub out old IRQ status before
    enabling IRQs.

That latter is done by tightening up IRQ handling for rtc-cmos everywhere,
also ensuring that when HPET is used it's the only thing triggering IRQ
reports to userspace; net object shrink.

Also fix a bogus HPET message related to its RTC emulation.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Report-by: W Unruh <unruh@physics.ubc.ca>
Cc: Andrew Victor <avictor.za@gmail.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/kernel/hpet.c |  10 ++-
 drivers/rtc/rtc-cmos.c | 140 +++++++++++++++++++----------------------
 2 files changed, 70 insertions(+), 80 deletions(-)

diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c
index 0ea6a19bfdfe..ad2b15a1334d 100644
--- a/arch/x86/kernel/hpet.c
+++ b/arch/x86/kernel/hpet.c
@@ -468,7 +468,7 @@ void hpet_disable(void)
 #define RTC_NUM_INTS		1
 
 static unsigned long hpet_rtc_flags;
-static unsigned long hpet_prev_update_sec;
+static int hpet_prev_update_sec;
 static struct rtc_time hpet_alarm_time;
 static unsigned long hpet_pie_count;
 static unsigned long hpet_t1_cmp;
@@ -575,6 +575,9 @@ int hpet_set_rtc_irq_bit(unsigned long bit_mask)
 
 	hpet_rtc_flags |= bit_mask;
 
+	if ((bit_mask & RTC_UIE) && !(oldbits & RTC_UIE))
+		hpet_prev_update_sec = -1;
+
 	if (!oldbits)
 		hpet_rtc_timer_init();
 
@@ -652,7 +655,7 @@ static void hpet_rtc_timer_reinit(void)
 		if (hpet_rtc_flags & RTC_PIE)
 			hpet_pie_count += lost_ints;
 		if (printk_ratelimit())
-			printk(KERN_WARNING "rtc: lost %d interrupts\n",
+			printk(KERN_WARNING "hpet1: lost %d rtc interrupts\n",
 				lost_ints);
 	}
 }
@@ -670,7 +673,8 @@ irqreturn_t hpet_rtc_interrupt(int irq, void *dev_id)
 
 	if (hpet_rtc_flags & RTC_UIE &&
 	    curr_time.tm_sec != hpet_prev_update_sec) {
-		rtc_int_flag = RTC_UF;
+		if (hpet_prev_update_sec >= 0)
+			rtc_int_flag = RTC_UF;
 		hpet_prev_update_sec = curr_time.tm_sec;
 	}
 
diff --git a/drivers/rtc/rtc-cmos.c b/drivers/rtc/rtc-cmos.c
index e9984650ea95..6ea349aba3ba 100644
--- a/drivers/rtc/rtc-cmos.c
+++ b/drivers/rtc/rtc-cmos.c
@@ -235,11 +235,56 @@ static int cmos_read_alarm(struct device *dev, struct rtc_wkalrm *t)
 	return 0;
 }
 
+static void cmos_checkintr(struct cmos_rtc *cmos, unsigned char rtc_control)
+{
+	unsigned char	rtc_intr;
+
+	/* NOTE after changing RTC_xIE bits we always read INTR_FLAGS;
+	 * allegedly some older rtcs need that to handle irqs properly
+	 */
+	rtc_intr = CMOS_READ(RTC_INTR_FLAGS);
+
+	if (is_hpet_enabled())
+		return;
+
+	rtc_intr &= (rtc_control & RTC_IRQMASK) | RTC_IRQF;
+	if (is_intr(rtc_intr))
+		rtc_update_irq(cmos->rtc, 1, rtc_intr);
+}
+
+static void cmos_irq_enable(struct cmos_rtc *cmos, unsigned char mask)
+{
+	unsigned char	rtc_control;
+
+	/* flush any pending IRQ status, notably for update irqs,
+	 * before we enable new IRQs
+	 */
+	rtc_control = CMOS_READ(RTC_CONTROL);
+	cmos_checkintr(cmos, rtc_control);
+
+	rtc_control |= mask;
+	CMOS_WRITE(rtc_control, RTC_CONTROL);
+	hpet_set_rtc_irq_bit(mask);
+
+	cmos_checkintr(cmos, rtc_control);
+}
+
+static void cmos_irq_disable(struct cmos_rtc *cmos, unsigned char mask)
+{
+	unsigned char	rtc_control;
+
+	rtc_control = CMOS_READ(RTC_CONTROL);
+	rtc_control &= ~mask;
+	CMOS_WRITE(rtc_control, RTC_CONTROL);
+	hpet_mask_rtc_irq_bit(mask);
+
+	cmos_checkintr(cmos, rtc_control);
+}
+
 static int cmos_set_alarm(struct device *dev, struct rtc_wkalrm *t)
 {
 	struct cmos_rtc	*cmos = dev_get_drvdata(dev);
 	unsigned char	mon, mday, hrs, min, sec;
-	unsigned char	rtc_control, rtc_intr;
 
 	if (!is_valid_irq(cmos->irq))
 		return -EIO;
@@ -266,15 +311,7 @@ static int cmos_set_alarm(struct device *dev, struct rtc_wkalrm *t)
 	spin_lock_irq(&rtc_lock);
 
 	/* next rtc irq must not be from previous alarm setting */
-	rtc_control = CMOS_READ(RTC_CONTROL);
-	rtc_control &= ~RTC_AIE;
-	CMOS_WRITE(rtc_control, RTC_CONTROL);
-	hpet_mask_rtc_irq_bit(RTC_AIE);
-
-	rtc_intr = CMOS_READ(RTC_INTR_FLAGS);
-	rtc_intr &= (rtc_control & RTC_IRQMASK) | RTC_IRQF;
-	if (is_intr(rtc_intr))
-		rtc_update_irq(cmos->rtc, 1, rtc_intr);
+	cmos_irq_disable(cmos, RTC_AIE);
 
 	/* update alarm */
 	CMOS_WRITE(hrs, RTC_HOURS_ALARM);
@@ -293,16 +330,8 @@ static int cmos_set_alarm(struct device *dev, struct rtc_wkalrm *t)
 	 */
 	hpet_set_alarm_time(t->time.tm_hour, t->time.tm_min, t->time.tm_sec);
 
-	if (t->enabled) {
-		rtc_control |= RTC_AIE;
-		CMOS_WRITE(rtc_control, RTC_CONTROL);
-		hpet_set_rtc_irq_bit(RTC_AIE);
-
-		rtc_intr = CMOS_READ(RTC_INTR_FLAGS);
-		rtc_intr &= (rtc_control & RTC_IRQMASK) | RTC_IRQF;
-		if (is_intr(rtc_intr))
-			rtc_update_irq(cmos->rtc, 1, rtc_intr);
-	}
+	if (t->enabled)
+		cmos_irq_enable(cmos, RTC_AIE);
 
 	spin_unlock_irq(&rtc_lock);
 
@@ -335,28 +364,17 @@ static int cmos_irq_set_freq(struct device *dev, int freq)
 static int cmos_irq_set_state(struct device *dev, int enabled)
 {
 	struct cmos_rtc	*cmos = dev_get_drvdata(dev);
-	unsigned char	rtc_control, rtc_intr;
 	unsigned long	flags;
 
 	if (!is_valid_irq(cmos->irq))
 		return -ENXIO;
 
 	spin_lock_irqsave(&rtc_lock, flags);
-	rtc_control = CMOS_READ(RTC_CONTROL);
 
-	if (enabled) {
-		rtc_control |= RTC_PIE;
-		hpet_set_rtc_irq_bit(RTC_PIE);
-	} else {
-		rtc_control &= ~RTC_PIE;
-		hpet_mask_rtc_irq_bit(RTC_PIE);
-	}
-	CMOS_WRITE(rtc_control, RTC_CONTROL);
-
-	rtc_intr = CMOS_READ(RTC_INTR_FLAGS);
-	rtc_intr &= (rtc_control & RTC_IRQMASK) | RTC_IRQF;
-	if (is_intr(rtc_intr))
-		rtc_update_irq(cmos->rtc, 1, rtc_intr);
+	if (enabled)
+		cmos_irq_enable(cmos, RTC_PIE);
+	else
+		cmos_irq_disable(cmos, RTC_PIE);
 
 	spin_unlock_irqrestore(&rtc_lock, flags);
 	return 0;
@@ -368,7 +386,6 @@ static int
 cmos_rtc_ioctl(struct device *dev, unsigned int cmd, unsigned long arg)
 {
 	struct cmos_rtc	*cmos = dev_get_drvdata(dev);
-	unsigned char	rtc_control, rtc_intr;
 	unsigned long	flags;
 
 	switch (cmd) {
@@ -385,32 +402,20 @@ cmos_rtc_ioctl(struct device *dev, unsigned int cmd, unsigned long arg)
 	}
 
 	spin_lock_irqsave(&rtc_lock, flags);
-	rtc_control = CMOS_READ(RTC_CONTROL);
 	switch (cmd) {
 	case RTC_AIE_OFF:	/* alarm off */
-		rtc_control &= ~RTC_AIE;
-		hpet_mask_rtc_irq_bit(RTC_AIE);
+		cmos_irq_disable(cmos, RTC_AIE);
 		break;
 	case RTC_AIE_ON:	/* alarm on */
-		rtc_control |= RTC_AIE;
-		hpet_set_rtc_irq_bit(RTC_AIE);
+		cmos_irq_enable(cmos, RTC_AIE);
 		break;
 	case RTC_UIE_OFF:	/* update off */
-		rtc_control &= ~RTC_UIE;
-		hpet_mask_rtc_irq_bit(RTC_UIE);
+		cmos_irq_disable(cmos, RTC_UIE);
 		break;
 	case RTC_UIE_ON:	/* update on */
-		rtc_control |= RTC_UIE;
-		hpet_set_rtc_irq_bit(RTC_UIE);
+		cmos_irq_enable(cmos, RTC_UIE);
 		break;
 	}
-	CMOS_WRITE(rtc_control, RTC_CONTROL);
-
-	rtc_intr = CMOS_READ(RTC_INTR_FLAGS);
-	rtc_intr &= (rtc_control & RTC_IRQMASK) | RTC_IRQF;
-	if (is_intr(rtc_intr))
-		rtc_update_irq(cmos->rtc, 1, rtc_intr);
-
 	spin_unlock_irqrestore(&rtc_lock, flags);
 	return 0;
 }
@@ -571,7 +576,6 @@ static irqreturn_t cmos_interrupt(int irq, void *p)
 	 * alarm woke the system.
 	 */
 	if (irqstat & RTC_AIE) {
-		rtc_control = CMOS_READ(RTC_CONTROL);
 		rtc_control &= ~RTC_AIE;
 		CMOS_WRITE(rtc_control, RTC_CONTROL);
 		hpet_mask_rtc_irq_bit(RTC_AIE);
@@ -685,17 +689,10 @@ cmos_do_probe(struct device *dev, struct resource *ports, int rtc_irq)
 	hpet_set_periodic_freq(cmos_rtc.rtc->irq_freq);
 	CMOS_WRITE(RTC_REF_CLCK_32KHZ | 0x06, RTC_FREQ_SELECT);
 
-	/* disable irqs.
-	 *
-	 * NOTE after changing RTC_xIE bits we always read INTR_FLAGS;
-	 * allegedly some older rtcs need that to handle irqs properly
-	 */
-	rtc_control = CMOS_READ(RTC_CONTROL);
-	rtc_control &= ~(RTC_PIE | RTC_AIE | RTC_UIE);
-	CMOS_WRITE(rtc_control, RTC_CONTROL);
-	hpet_mask_rtc_irq_bit(RTC_PIE | RTC_AIE | RTC_UIE);
+	/* disable irqs */
+	cmos_irq_disable(&cmos_rtc, RTC_PIE | RTC_AIE | RTC_UIE);
 
-	CMOS_READ(RTC_INTR_FLAGS);
+	rtc_control = CMOS_READ(RTC_CONTROL);
 
 	spin_unlock_irq(&rtc_lock);
 
@@ -768,15 +765,8 @@ cleanup0:
 
 static void cmos_do_shutdown(void)
 {
-	unsigned char	rtc_control;
-
 	spin_lock_irq(&rtc_lock);
-	rtc_control = CMOS_READ(RTC_CONTROL);
-	rtc_control &= ~RTC_IRQMASK;
-	CMOS_WRITE(rtc_control, RTC_CONTROL);
-	hpet_mask_rtc_irq_bit(RTC_IRQMASK);
-
-	CMOS_READ(RTC_INTR_FLAGS);
+	cmos_irq_disable(&cmos_rtc, RTC_IRQMASK);
 	spin_unlock_irq(&rtc_lock);
 }
 
@@ -817,7 +807,6 @@ static int cmos_suspend(struct device *dev, pm_message_t mesg)
 	spin_lock_irq(&rtc_lock);
 	cmos->suspend_ctrl = tmp = CMOS_READ(RTC_CONTROL);
 	if (tmp & (RTC_PIE|RTC_AIE|RTC_UIE)) {
-		unsigned char	irqstat;
 		unsigned char	mask;
 
 		if (do_wake)
@@ -828,10 +817,7 @@ static int cmos_suspend(struct device *dev, pm_message_t mesg)
 		CMOS_WRITE(tmp, RTC_CONTROL);
 		hpet_mask_rtc_irq_bit(mask);
 
-		irqstat = CMOS_READ(RTC_INTR_FLAGS);
-		irqstat &= (tmp & RTC_IRQMASK) | RTC_IRQF;
-		if (is_intr(irqstat))
-			rtc_update_irq(cmos->rtc, 1, irqstat);
+		cmos_checkintr(cmos, tmp);
 	}
 	spin_unlock_irq(&rtc_lock);
 
@@ -875,7 +861,7 @@ static int cmos_resume(struct device *dev)
 
 			mask = CMOS_READ(RTC_INTR_FLAGS);
 			mask &= (tmp & RTC_IRQMASK) | RTC_IRQF;
-			if (!is_intr(mask))
+			if (!is_hpet_enabled() || !is_intr(mask))
 				break;
 
 			/* force one-shot behavior if HPET blocked

From 4cad4431fcd872a1b2efc093b0db6df943f5a898 Mon Sep 17 00:00:00 2001
From: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Date: Wed, 23 Jul 2008 21:30:48 -0700
Subject: [PATCH 0441/1171] rtc-vr41xx: add irq_set_freq() and irq_set_state()

Implement the ioctls RTC_PIE_ON, RTC_PIE_OFF, RTC_IRQP_SET and
RTC_IRQP_READ in the standard RTC way.

Thanks Dave for noticing it.

Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Cc: David Brownell <david-b@pacbell.net>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Alessandro Zummo <a.zummo@towertech.it>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/rtc/rtc-vr41xx.c | 65 +++++++++++++++++++++-------------------
 1 file changed, 34 insertions(+), 31 deletions(-)

diff --git a/drivers/rtc/rtc-vr41xx.c b/drivers/rtc/rtc-vr41xx.c
index be9c70d0b193..884b635f028b 100644
--- a/drivers/rtc/rtc-vr41xx.c
+++ b/drivers/rtc/rtc-vr41xx.c
@@ -1,7 +1,7 @@
 /*
  *  Driver for NEC VR4100 series Real Time Clock unit.
  *
- *  Copyright (C) 2003-2006  Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
+ *  Copyright (C) 2003-2008  Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
  *
  *  This program is free software; you can redistribute it and/or modify
  *  it under the terms of the GNU General Public License as published by
@@ -34,7 +34,7 @@
 
 MODULE_AUTHOR("Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>");
 MODULE_DESCRIPTION("NEC VR4100 series RTC driver");
-MODULE_LICENSE("GPL");
+MODULE_LICENSE("GPL v2");
 
 /* RTC 1 registers */
 #define ETIMELREG		0x00
@@ -82,7 +82,6 @@ static unsigned long epoch = 1970;	/* Jan 1 1970 00:00:00 */
 
 static DEFINE_SPINLOCK(rtc_lock);
 static char rtc_name[] = "RTC";
-static unsigned long periodic_frequency;
 static unsigned long periodic_count;
 static unsigned int alarm_enabled;
 static int aie_irq = -1;
@@ -207,10 +206,37 @@ static int vr41xx_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *wkalrm)
 	return 0;
 }
 
-static int vr41xx_rtc_ioctl(struct device *dev, unsigned int cmd, unsigned long arg)
+static int vr41xx_rtc_irq_set_freq(struct device *dev, int freq)
 {
 	unsigned long count;
 
+	count = RTC_FREQUENCY;
+	do_div(count, freq);
+
+	periodic_count = count;
+
+	spin_lock_irq(&rtc_lock);
+
+	rtc1_write(RTCL1LREG, count);
+	rtc1_write(RTCL1HREG, count >> 16);
+
+	spin_unlock_irq(&rtc_lock);
+
+	return 0;
+}
+
+static int vr41xx_rtc_irq_set_state(struct device *dev, int enabled)
+{
+	if (enabled)
+		enable_irq(pie_irq);
+	else
+		disable_irq(pie_irq);
+
+	return 0;
+}
+
+static int vr41xx_rtc_ioctl(struct device *dev, unsigned int cmd, unsigned long arg)
+{
 	switch (cmd) {
 	case RTC_AIE_ON:
 		spin_lock_irq(&rtc_lock);
@@ -230,33 +256,6 @@ static int vr41xx_rtc_ioctl(struct device *dev, unsigned int cmd, unsigned long
 			alarm_enabled = 0;
 		}
 
-		spin_unlock_irq(&rtc_lock);
-		break;
-	case RTC_PIE_ON:
-		enable_irq(pie_irq);
-		break;
-	case RTC_PIE_OFF:
-		disable_irq(pie_irq);
-		break;
-	case RTC_IRQP_READ:
-		return put_user(periodic_frequency, (unsigned long __user *)arg);
-		break;
-	case RTC_IRQP_SET:
-		if (arg > MAX_PERIODIC_RATE)
-			return -EINVAL;
-
-		periodic_frequency = arg;
-
-		count = RTC_FREQUENCY;
-		do_div(count, arg);
-
-		periodic_count = count;
-
-		spin_lock_irq(&rtc_lock);
-
-		rtc1_write(RTCL1LREG, count);
-		rtc1_write(RTCL1HREG, count >> 16);
-
 		spin_unlock_irq(&rtc_lock);
 		break;
 	case RTC_EPOCH_READ:
@@ -309,6 +308,8 @@ static const struct rtc_class_ops vr41xx_rtc_ops = {
 	.set_time	= vr41xx_rtc_set_time,
 	.read_alarm	= vr41xx_rtc_read_alarm,
 	.set_alarm	= vr41xx_rtc_set_alarm,
+	.irq_set_freq	= vr41xx_rtc_irq_set_freq,
+	.irq_set_state	= vr41xx_rtc_irq_set_state,
 };
 
 static int __devinit rtc_probe(struct platform_device *pdev)
@@ -346,6 +347,8 @@ static int __devinit rtc_probe(struct platform_device *pdev)
 		goto err_iounmap_all;
 	}
 
+	rtc->max_user_freq = MAX_PERIODIC_RATE;
+
 	spin_lock_irq(&rtc_lock);
 
 	rtc1_write(ECMPLREG, 0);

From 2ece5f43b041b96fa2a05107a10a6b0ea0c03a3b Mon Sep 17 00:00:00 2001
From: Sebastian Siewior <bigeasy@linutronix.de>
Date: Wed, 23 Jul 2008 21:30:49 -0700
Subject: [PATCH 0442/1171] fbdev: add the carmine FB driver

Basic FB driver for the carmine chip.  The driver registers two FB devices for
the two possible screens.  The DRAM settings can be be switched via Kconfig
(between eval board and custom).

Signed-off-by: Sebastian Siewior <bigeasy@linutronix.de>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/Kconfig          |  26 ++
 drivers/video/Makefile         |   1 +
 drivers/video/carminefb.c      | 790 +++++++++++++++++++++++++++++++++
 drivers/video/carminefb.h      |  64 +++
 drivers/video/carminefb_regs.h | 159 +++++++
 5 files changed, 1040 insertions(+)
 create mode 100644 drivers/video/carminefb.c
 create mode 100644 drivers/video/carminefb.h
 create mode 100644 drivers/video/carminefb_regs.h

diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig
index 9b887ef64ff1..7072d2c5a040 100644
--- a/drivers/video/Kconfig
+++ b/drivers/video/Kconfig
@@ -1658,6 +1658,32 @@ config FB_PM3
 	  similar boards, 3DLabs Permedia3 Create!, Appian Jeronimo 2000
 	  and maybe other boards.
 
+config FB_CARMINE
+	tristate "Fujitsu carmine frame buffer support"
+	depends on FB && PCI
+	select FB_CFB_FILLRECT
+	select FB_CFB_COPYAREA
+	select FB_CFB_IMAGEBLIT
+	help
+	  This is the frame buffer device driver for the Fujitsu Carmine chip.
+	  The driver provides two independent frame buffer devices.
+
+choice
+	depends on FB_CARMINE
+	prompt "DRAM timing"
+	default FB_CARMINE_DRAM_EVAL
+
+config FB_CARMINE_DRAM_EVAL
+	bool "Eval board timings"
+	help
+	  Use timings which work on the eval card.
+
+config CARMINE_DRAM_CUSTOM
+	bool "Custom board timings"
+	help
+	  Use custom board timings.
+endchoice
+
 config FB_AU1100
 	bool "Au1100 LCD Driver"
 	depends on (FB = y) && MIPS && SOC_AU1100
diff --git a/drivers/video/Makefile b/drivers/video/Makefile
index 04bca35403ff..7ee85c0d2e5f 100644
--- a/drivers/video/Makefile
+++ b/drivers/video/Makefile
@@ -117,6 +117,7 @@ obj-$(CONFIG_FB_SM501)            += sm501fb.o
 obj-$(CONFIG_FB_XILINX)           += xilinxfb.o
 obj-$(CONFIG_FB_OMAP)             += omap/
 obj-$(CONFIG_XEN_FBDEV_FRONTEND)  += xen-fbfront.o
+obj-$(CONFIG_FB_CARMINE)          += carminefb.o
 
 # Platform or fallback drivers go here
 obj-$(CONFIG_FB_UVESA)            += uvesafb.o
diff --git a/drivers/video/carminefb.c b/drivers/video/carminefb.c
new file mode 100644
index 000000000000..e15bb447440a
--- /dev/null
+++ b/drivers/video/carminefb.c
@@ -0,0 +1,790 @@
+/*
+ * Frame buffer driver for the Carmine GPU.
+ *
+ * The driver configures the GPU as follows
+ * - FB0 is display 0 with unique memory area
+ * - FB1 is display 1 with unique memory area
+ * - both display use 32 bit colors
+ */
+#include <linux/delay.h>
+#include <linux/errno.h>
+#include <linux/fb.h>
+#include <linux/interrupt.h>
+#include <linux/pci.h>
+
+#include "carminefb.h"
+#include "carminefb_regs.h"
+
+#if !defined(__LITTLE_ENDIAN) && !defined(__BIG_ENDIAN)
+#error  "The endianness of the target host has not been defined."
+#endif
+
+/*
+ * The initial video mode can be supplied via two different ways:
+ * - as a string that is passed to fb_find_mode() (module option fb_mode_str)
+ * - as an integer that picks the video mode from carmine_modedb[] (module
+ *   option fb_mode)
+ *
+ * If nothing is used than the initial video mode will be the
+ * CARMINEFB_DEFAULT_VIDEO_MODE member of the carmine_modedb[].
+ */
+#define CARMINEFB_DEFAULT_VIDEO_MODE	1
+
+static unsigned int fb_mode = CARMINEFB_DEFAULT_VIDEO_MODE;
+module_param(fb_mode, uint, 444);
+MODULE_PARM_DESC(fb_mode, "Initial video mode as integer.");
+
+static char *fb_mode_str;
+module_param(fb_mode_str, charp, 444);
+MODULE_PARM_DESC(fb_mode_str, "Initial video mode in characters.");
+
+/*
+ * Carminefb displays:
+ * 0b000 None
+ * 0b001 Display 0
+ * 0b010 Display 1
+ */
+static int fb_displays = CARMINE_USE_DISPLAY0 | CARMINE_USE_DISPLAY1;
+module_param(fb_displays, int, 444);
+MODULE_PARM_DESC(fb_displays, "Bit mode, which displays are used");
+
+struct carmine_hw {
+	void __iomem *v_regs;
+	void __iomem *screen_mem;
+	struct fb_info *fb[MAX_DISPLAY];
+};
+
+struct carmine_resolution {
+	u32 htp;
+	u32 hsp;
+	u32 hsw;
+	u32 hdp;
+	u32 vtr;
+	u32 vsp;
+	u32 vsw;
+	u32 vdp;
+	u32 disp_mode;
+};
+
+struct carmine_fb {
+	void __iomem *display_reg;
+	void __iomem *screen_base;
+	u32 smem_offset;
+	u32 cur_mode;
+	u32 new_mode;
+	struct carmine_resolution *res;
+	u32 pseudo_palette[16];
+};
+
+static struct fb_fix_screeninfo carminefb_fix __devinitdata = {
+	.id = "Carmine",
+	.type = FB_TYPE_PACKED_PIXELS,
+	.visual = FB_VISUAL_TRUECOLOR,
+	.accel = FB_ACCEL_NONE,
+};
+
+static const struct fb_videomode carmine_modedb[] = {
+	{
+		.name		= "640x480",
+		.xres		= 640,
+		.yres		= 480,
+	}, {
+		.name		= "800x600",
+		.xres		= 800,
+		.yres		= 600,
+	},
+};
+
+static struct carmine_resolution car_modes[] = {
+	{
+		/* 640x480 */
+		.htp = 800,
+		.hsp = 672,
+		.hsw = 96,
+		.hdp = 640,
+		.vtr = 525,
+		.vsp = 490,
+		.vsw = 2,
+		.vdp = 480,
+		.disp_mode = 0x1400,
+	},
+	{
+		/* 800x600 */
+		.htp = 1060,
+		.hsp = 864,
+		.hsw = 72,
+		.hdp = 800,
+		.vtr = 628,
+		.vsp = 601,
+		.vsw = 2,
+		.vdp = 600,
+		.disp_mode = 0x0d00,
+	}
+};
+
+static int carmine_find_mode(const struct fb_var_screeninfo *var)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(car_modes); i++)
+		if (car_modes[i].hdp == var->xres &&
+		    car_modes[i].vdp == var->yres)
+			return i;
+	return -EINVAL;
+}
+
+static void c_set_disp_reg(const struct carmine_fb *par,
+		u32 offset, u32 val)
+{
+	writel(val, par->display_reg + offset);
+}
+
+static u32 c_get_disp_reg(const struct carmine_fb *par,
+		u32 offset)
+{
+	return readl(par->display_reg + offset);
+}
+
+static void c_set_hw_reg(const struct carmine_hw *hw,
+		u32 offset, u32 val)
+{
+	writel(val, hw->v_regs + offset);
+}
+
+static u32 c_get_hw_reg(const struct carmine_hw *hw,
+		u32 offset)
+{
+	return readl(hw->v_regs + offset);
+}
+
+static int carmine_setcolreg(unsigned regno, unsigned red, unsigned green,
+		unsigned blue, unsigned transp, struct fb_info *info)
+{
+	if (regno >= 16)
+		return 1;
+
+	red >>= 8;
+	green >>= 8;
+	blue >>= 8;
+	transp >>= 8;
+
+	((u32 *)info->pseudo_palette)[regno] = be32_to_cpu(transp << 24 |
+		red << 0 | green << 8 | blue << 16);
+	return 0;
+}
+
+static int carmine_check_var(struct fb_var_screeninfo *var,
+		struct fb_info *info)
+{
+	int ret;
+
+	ret = carmine_find_mode(var);
+	if (ret < 0)
+		return ret;
+
+	if (var->grayscale || var->rotate || var->nonstd)
+		return -EINVAL;
+
+	var->xres_virtual = var->xres;
+	var->yres_virtual = var->yres;
+
+	var->bits_per_pixel = 32;
+
+#ifdef __BIG_ENDIAN
+	var->transp.offset = 24;
+	var->red.offset = 0;
+	var->green.offset = 8;
+	var->blue.offset = 16;
+#else
+	var->transp.offset = 24;
+	var->red.offset = 16;
+	var->green.offset = 8;
+	var->blue.offset = 0;
+#endif
+
+	var->red.length = 8;
+	var->green.length = 8;
+	var->blue.length = 8;
+	var->transp.length = 8;
+
+	var->red.msb_right = 0;
+	var->green.msb_right = 0;
+	var->blue.msb_right = 0;
+	var->transp.msb_right = 0;
+	return 0;
+}
+
+static void carmine_init_display_param(struct carmine_fb *par)
+{
+	u32 width;
+	u32 height;
+	u32 param;
+	u32 window_size;
+	u32 soffset = par->smem_offset;
+
+	c_set_disp_reg(par, CARMINE_DISP_REG_C_TRANS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_MLMR_TRANS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_CURSOR_MODE,
+			CARMINE_CURSOR0_PRIORITY_MASK |
+			CARMINE_CURSOR1_PRIORITY_MASK |
+			CARMINE_CURSOR_CUTZ_MASK);
+
+	/* Set default cursor position */
+	c_set_disp_reg(par, CARMINE_DISP_REG_CUR1_POS, 0 << 16 | 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_CUR2_POS, 0 << 16 | 0);
+
+	/* Set default display mode */
+	c_set_disp_reg(par, CARMINE_DISP_REG_L0_EXT_MODE, CARMINE_WINDOW_MODE |
+			CARMINE_EXT_CMODE_DIRECT24_RGBA);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L1_EXT_MODE,
+			CARMINE_EXT_CMODE_DIRECT24_RGBA);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L2_EXT_MODE, CARMINE_EXTEND_MODE |
+			CARMINE_EXT_CMODE_DIRECT24_RGBA);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L3_EXT_MODE, CARMINE_EXTEND_MODE |
+			CARMINE_EXT_CMODE_DIRECT24_RGBA);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L4_EXT_MODE, CARMINE_EXTEND_MODE |
+			CARMINE_EXT_CMODE_DIRECT24_RGBA);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L5_EXT_MODE, CARMINE_EXTEND_MODE |
+			CARMINE_EXT_CMODE_DIRECT24_RGBA);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L6_EXT_MODE, CARMINE_EXTEND_MODE |
+			CARMINE_EXT_CMODE_DIRECT24_RGBA);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L7_EXT_MODE, CARMINE_EXTEND_MODE |
+			CARMINE_EXT_CMODE_DIRECT24_RGBA);
+
+	/* Set default frame size to layer mode register */
+	width = par->res->hdp * 4 / CARMINE_DISP_WIDTH_UNIT;
+	width = width << CARMINE_DISP_WIDTH_SHIFT;
+
+	height = par->res->vdp - 1;
+	param = width | height;
+
+	c_set_disp_reg(par, CARMINE_DISP_REG_L0_MODE_W_H, param);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L1_WIDTH, width);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L2_MODE_W_H, param);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L3_MODE_W_H, param);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L4_MODE_W_H, param);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L5_MODE_W_H, param);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L6_MODE_W_H, param);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L7_MODE_W_H, param);
+
+	/* Set default pos and size */
+	window_size = (par->res->vdp - 1) << CARMINE_DISP_WIN_H_SHIFT;
+	window_size |= par->res->hdp;
+
+	c_set_disp_reg(par, CARMINE_DISP_REG_L0_WIN_POS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L0_WIN_SIZE, window_size);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L1_WIN_POS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L1_WIN_SIZE, window_size);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L2_WIN_POS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L2_WIN_SIZE, window_size);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L3_WIN_POS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L3_WIN_SIZE, window_size);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L4_WIN_POS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L4_WIN_SIZE, window_size);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L5_WIN_POS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L5_WIN_SIZE, window_size);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L6_WIN_POS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L6_WIN_SIZE, window_size);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L7_WIN_POS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L7_WIN_SIZE, window_size);
+
+	/* Set default origin address */
+	c_set_disp_reg(par, CARMINE_DISP_REG_L0_ORG_ADR, soffset);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L1_ORG_ADR, soffset);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L2_ORG_ADR1, soffset);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L3_ORG_ADR1, soffset);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L4_ORG_ADR1, soffset);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L5_ORG_ADR1, soffset);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L6_ORG_ADR1, soffset);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L7_ORG_ADR1, soffset);
+
+	/* Set default display address */
+	c_set_disp_reg(par, CARMINE_DISP_REG_L0_DISP_ADR, soffset);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L2_DISP_ADR1, soffset);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L3_DISP_ADR1, soffset);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L4_DISP_ADR1, soffset);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L5_DISP_ADR1, soffset);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L6_DISP_ADR0, soffset);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L7_DISP_ADR0, soffset);
+
+	/* Set default display position */
+	c_set_disp_reg(par, CARMINE_DISP_REG_L0_DISP_POS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L2_DISP_POS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L3_DISP_POS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L4_DISP_POS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L5_DISP_POS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L6_DISP_POS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L7_DISP_POS, 0);
+
+	/* Set default blend mode */
+	c_set_disp_reg(par, CARMINE_DISP_REG_BLEND_MODE_L0, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_BLEND_MODE_L1, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_BLEND_MODE_L2, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_BLEND_MODE_L3, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_BLEND_MODE_L4, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_BLEND_MODE_L5, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_BLEND_MODE_L6, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_BLEND_MODE_L7, 0);
+
+	/* default transparency mode */
+	c_set_disp_reg(par, CARMINE_DISP_REG_L0_TRANS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L1_TRANS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L2_TRANS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L3_TRANS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L4_TRANS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L5_TRANS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L6_TRANS, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L7_TRANS, 0);
+
+	/* Set default read skip parameter */
+	c_set_disp_reg(par, CARMINE_DISP_REG_L0RM, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L2RM, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L3RM, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L4RM, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L5RM, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L6RM, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L7RM, 0);
+
+	c_set_disp_reg(par, CARMINE_DISP_REG_L0PX, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L2PX, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L3PX, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L4PX, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L5PX, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L6PX, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L7PX, 0);
+
+	c_set_disp_reg(par, CARMINE_DISP_REG_L0PY, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L2PY, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L3PY, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L4PY, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L5PY, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L6PY, 0);
+	c_set_disp_reg(par, CARMINE_DISP_REG_L7PY, 0);
+}
+
+static void set_display_parameters(struct carmine_fb *par)
+{
+	u32 mode;
+	u32 hdp, vdp, htp, hsp, hsw, vtr, vsp, vsw;
+
+	/*
+	 * display timing. Parameters are decreased by one because hardware
+	 * spec is 0 to (n - 1)
+	 * */
+	hdp = par->res->hdp - 1;
+	vdp = par->res->vdp - 1;
+	htp = par->res->htp - 1;
+	hsp = par->res->hsp - 1;
+	hsw = par->res->hsw - 1;
+	vtr = par->res->vtr - 1;
+	vsp = par->res->vsp - 1;
+	vsw = par->res->vsw - 1;
+
+	c_set_disp_reg(par, CARMINE_DISP_REG_H_TOTAL,
+			htp << CARMINE_DISP_HTP_SHIFT);
+	c_set_disp_reg(par, CARMINE_DISP_REG_H_PERIOD,
+			(hdp << CARMINE_DISP_HDB_SHIFT)	| hdp);
+	c_set_disp_reg(par, CARMINE_DISP_REG_V_H_W_H_POS,
+			(vsw << CARMINE_DISP_VSW_SHIFT) |
+			(hsw << CARMINE_DISP_HSW_SHIFT) |
+			(hsp));
+	c_set_disp_reg(par, CARMINE_DISP_REG_V_TOTAL,
+			vtr << CARMINE_DISP_VTR_SHIFT);
+	c_set_disp_reg(par, CARMINE_DISP_REG_V_PERIOD_POS,
+			(vdp << CARMINE_DISP_VDP_SHIFT) | vsp);
+
+	/* clock */
+	mode = c_get_disp_reg(par, CARMINE_DISP_REG_DCM1);
+	mode = (mode & ~CARMINE_DISP_DCM_MASK) |
+		(par->res->disp_mode & CARMINE_DISP_DCM_MASK);
+	/* enable video output and layer 0 */
+	mode |= CARMINE_DEN | CARMINE_L0E;
+	c_set_disp_reg(par, CARMINE_DISP_REG_DCM1, mode);
+}
+
+static int carmine_set_par(struct fb_info *info)
+{
+	struct carmine_fb *par = info->par;
+	int ret;
+
+	ret = carmine_find_mode(&info->var);
+	if (ret < 0)
+		return ret;
+
+	par->new_mode = ret;
+	if (par->cur_mode != par->new_mode) {
+
+		par->cur_mode = par->new_mode;
+		par->res = &car_modes[par->new_mode];
+
+		carmine_init_display_param(par);
+		set_display_parameters(par);
+	}
+
+	info->fix.line_length = info->var.xres * info->var.bits_per_pixel / 8;
+	return 0;
+}
+
+static int init_hardware(struct carmine_hw *hw)
+{
+	u32 flags;
+	u32 loops;
+	u32 ret;
+
+	/* Initalize Carmine */
+	/* Sets internal clock */
+	c_set_hw_reg(hw, CARMINE_CTL_REG + CARMINE_CTL_REG_CLOCK_ENABLE,
+			CARMINE_DFLT_IP_CLOCK_ENABLE);
+
+	/* Video signal output is turned off */
+	c_set_hw_reg(hw, CARMINE_DISP0_REG + CARMINE_DISP_REG_DCM1, 0);
+	c_set_hw_reg(hw, CARMINE_DISP1_REG + CARMINE_DISP_REG_DCM1, 0);
+
+	/* Software reset */
+	c_set_hw_reg(hw, CARMINE_CTL_REG + CARMINE_CTL_REG_SOFTWARE_RESET, 1);
+	c_set_hw_reg(hw, CARMINE_CTL_REG + CARMINE_CTL_REG_SOFTWARE_RESET, 0);
+
+	/* I/O mode settings */
+	flags = CARMINE_DFLT_IP_DCTL_IO_CONT1 << 16 |
+		CARMINE_DFLT_IP_DCTL_IO_CONT0;
+	c_set_hw_reg(hw, CARMINE_DCTL_REG + CARMINE_DCTL_REG_IOCONT1_IOCONT0,
+			flags);
+
+	/* DRAM initial sequence */
+	flags = CARMINE_DFLT_IP_DCTL_MODE << 16 | CARMINE_DFLT_IP_DCTL_ADD;
+	c_set_hw_reg(hw, CARMINE_DCTL_REG + CARMINE_DCTL_REG_MODE_ADD,
+			flags);
+
+	flags = CARMINE_DFLT_IP_DCTL_SET_TIME1 << 16 |
+		CARMINE_DFLT_IP_DCTL_EMODE;
+	c_set_hw_reg(hw, CARMINE_DCTL_REG + CARMINE_DCTL_REG_SETTIME1_EMODE,
+			flags);
+
+	flags = CARMINE_DFLT_IP_DCTL_REFRESH << 16 |
+		CARMINE_DFLT_IP_DCTL_SET_TIME2;
+	c_set_hw_reg(hw, CARMINE_DCTL_REG + CARMINE_DCTL_REG_REFRESH_SETTIME2,
+			flags);
+
+	flags = CARMINE_DFLT_IP_DCTL_RESERVE2 << 16 |
+		CARMINE_DFLT_IP_DCTL_FIFO_DEPTH;
+	c_set_hw_reg(hw, CARMINE_DCTL_REG + CARMINE_DCTL_REG_RSV2_RSV1, flags);
+
+	flags = CARMINE_DFLT_IP_DCTL_DDRIF2 << 16 | CARMINE_DFLT_IP_DCTL_DDRIF1;
+	c_set_hw_reg(hw, CARMINE_DCTL_REG + CARMINE_DCTL_REG_DDRIF2_DDRIF1,
+			flags);
+
+	flags = CARMINE_DFLT_IP_DCTL_RESERVE0 << 16 |
+		CARMINE_DFLT_IP_DCTL_STATES;
+	c_set_hw_reg(hw, CARMINE_DCTL_REG + CARMINE_DCTL_REG_RSV0_STATES,
+			flags);
+
+	/* Executes DLL reset */
+	if (CARMINE_DCTL_DLL_RESET) {
+		for (loops = 0; loops < CARMINE_DCTL_INIT_WAIT_LIMIT; loops++) {
+
+			ret = c_get_hw_reg(hw, CARMINE_DCTL_REG +
+					CARMINE_DCTL_REG_RSV0_STATES);
+			ret &= CARMINE_DCTL_REG_STATES_MASK;
+			if (!ret)
+				break;
+
+			mdelay(CARMINE_DCTL_INIT_WAIT_INTERVAL);
+		}
+
+		if (loops >= CARMINE_DCTL_INIT_WAIT_LIMIT) {
+			printk(KERN_ERR "DRAM init failed\n");
+			return -EIO;
+		}
+	}
+
+	flags = CARMINE_DFLT_IP_DCTL_MODE_AFT_RST << 16 |
+		CARMINE_DFLT_IP_DCTL_ADD;
+	c_set_hw_reg(hw, CARMINE_DCTL_REG + CARMINE_DCTL_REG_MODE_ADD, flags);
+
+	flags = CARMINE_DFLT_IP_DCTL_RESERVE0 << 16 |
+		CARMINE_DFLT_IP_DCTL_STATES_AFT_RST;
+	c_set_hw_reg(hw, CARMINE_DCTL_REG + CARMINE_DCTL_REG_RSV0_STATES,
+			flags);
+
+	/* Initialize the write back register */
+	c_set_hw_reg(hw, CARMINE_WB_REG + CARMINE_WB_REG_WBM,
+			CARMINE_WB_REG_WBM_DEFAULT);
+
+	/* Initialize the Kottos registers */
+	c_set_hw_reg(hw, CARMINE_GRAPH_REG + CARMINE_GRAPH_REG_VRINTM, 0);
+	c_set_hw_reg(hw, CARMINE_GRAPH_REG + CARMINE_GRAPH_REG_VRERRM, 0);
+
+	/* Set DC offsets */
+	c_set_hw_reg(hw, CARMINE_GRAPH_REG + CARMINE_GRAPH_REG_DC_OFFSET_PX, 0);
+	c_set_hw_reg(hw, CARMINE_GRAPH_REG + CARMINE_GRAPH_REG_DC_OFFSET_PY, 0);
+	c_set_hw_reg(hw, CARMINE_GRAPH_REG + CARMINE_GRAPH_REG_DC_OFFSET_LX, 0);
+	c_set_hw_reg(hw, CARMINE_GRAPH_REG + CARMINE_GRAPH_REG_DC_OFFSET_LY, 0);
+	c_set_hw_reg(hw, CARMINE_GRAPH_REG + CARMINE_GRAPH_REG_DC_OFFSET_TX, 0);
+	c_set_hw_reg(hw, CARMINE_GRAPH_REG + CARMINE_GRAPH_REG_DC_OFFSET_TY, 0);
+	return 0;
+}
+
+static struct fb_ops carminefb_ops = {
+	.owner		= THIS_MODULE,
+	.fb_fillrect	= cfb_fillrect,
+	.fb_copyarea	= cfb_copyarea,
+	.fb_imageblit	= cfb_imageblit,
+
+	.fb_check_var	= carmine_check_var,
+	.fb_set_par	= carmine_set_par,
+	.fb_setcolreg	= carmine_setcolreg,
+};
+
+static int alloc_carmine_fb(void __iomem *regs, void __iomem *smem_base,
+		int smem_offset, struct device *device, struct fb_info **rinfo)
+{
+	int ret;
+	struct fb_info *info;
+	struct carmine_fb *par;
+
+	info = framebuffer_alloc(sizeof *par, device);
+	if (!info)
+		return -ENOMEM;
+
+	par = info->par;
+	par->display_reg = regs;
+	par->smem_offset = smem_offset;
+
+	info->screen_base = smem_base + smem_offset;
+	info->screen_size = CARMINE_DISPLAY_MEM;
+	info->fbops = &carminefb_ops;
+
+	info->fix = carminefb_fix;
+	info->pseudo_palette = par->pseudo_palette;
+	info->flags = FBINFO_DEFAULT;
+
+	ret = fb_alloc_cmap(&info->cmap, 256, 1);
+	if (ret < 0)
+		goto err_free_fb;
+
+	if (fb_mode > ARRAY_SIZE(carmine_modedb))
+		fb_mode = CARMINEFB_DEFAULT_VIDEO_MODE;
+
+	par->cur_mode = par->new_mode = ~0;
+
+	ret = fb_find_mode(&info->var, info, fb_mode_str, carmine_modedb,
+			ARRAY_SIZE(carmine_modedb),
+			&carmine_modedb[fb_mode], 32);
+	if (!ret || ret == 4) {
+		ret = -EINVAL;
+		goto err_dealloc_cmap;
+	}
+
+	fb_videomode_to_modelist(carmine_modedb, ARRAY_SIZE(carmine_modedb),
+			&info->modelist);
+
+	ret = register_framebuffer(info);
+	if (ret < 0)
+		goto err_dealloc_cmap;
+
+	printk(KERN_INFO "fb%d: %s frame buffer device\n", info->node,
+			info->fix.id);
+
+	*rinfo = info;
+	return 0;
+
+err_dealloc_cmap:
+	fb_dealloc_cmap(&info->cmap);
+err_free_fb:
+	framebuffer_release(info);
+	return ret;
+}
+
+static void cleanup_fb_device(struct fb_info *info)
+{
+	if (info) {
+		unregister_framebuffer(info);
+		fb_dealloc_cmap(&info->cmap);
+		framebuffer_release(info);
+	}
+}
+
+static int __devinit carminefb_probe(struct pci_dev *dev,
+		const struct pci_device_id *ent)
+{
+	struct carmine_hw *hw;
+	struct device *device = &dev->dev;
+	struct fb_info *info;
+	int ret;
+
+	ret = pci_enable_device(dev);
+	if (ret)
+		return ret;
+
+	ret = -ENOMEM;
+	hw = kzalloc(sizeof *hw, GFP_KERNEL);
+	if (!hw)
+		goto err_enable_pci;
+
+	carminefb_fix.mmio_start = pci_resource_start(dev, CARMINE_CONFIG_BAR);
+	carminefb_fix.mmio_len = pci_resource_len(dev, CARMINE_CONFIG_BAR);
+
+	if (!request_mem_region(carminefb_fix.mmio_start,
+				carminefb_fix.mmio_len,
+				"carminefb regbase")) {
+		printk(KERN_ERR "carminefb: Can't reserve regbase.\n");
+		ret = -EBUSY;
+		goto err_free_hw;
+	}
+	hw->v_regs = ioremap_nocache(carminefb_fix.mmio_start,
+			carminefb_fix.mmio_len);
+	if (!hw->v_regs) {
+		printk(KERN_ERR "carminefb: Can't remap %s register.\n",
+				carminefb_fix.id);
+		goto err_free_reg_mmio;
+	}
+
+	carminefb_fix.smem_start = pci_resource_start(dev, CARMINE_MEMORY_BAR);
+	carminefb_fix.smem_len = pci_resource_len(dev, CARMINE_MEMORY_BAR);
+
+	/* The memory area tends to be very large (256 MiB). Remap only what
+	 * is required for that largest resolution to avoid remaps at run
+	 * time
+	 */
+	if (carminefb_fix.smem_len > CARMINE_TOTAL_DIPLAY_MEM)
+		carminefb_fix.smem_len = CARMINE_TOTAL_DIPLAY_MEM;
+
+	else if (carminefb_fix.smem_len < CARMINE_TOTAL_DIPLAY_MEM) {
+		printk(KERN_ERR "carminefb: Memory bar is only %d bytes, %d "
+				"are required.", carminefb_fix.smem_len,
+				CARMINE_TOTAL_DIPLAY_MEM);
+		goto err_free_reg_mmio;
+	}
+
+	if (!request_mem_region(carminefb_fix.smem_start,
+				carminefb_fix.smem_len,	"carminefb smem")) {
+		printk(KERN_ERR "carminefb: Can't reserve smem.\n");
+		goto err_unmap_vregs;
+	}
+
+	hw->screen_mem = ioremap_nocache(carminefb_fix.smem_start,
+			carminefb_fix.smem_len);
+	if (!hw->screen_mem) {
+		printk(KERN_ERR "carmine: Can't ioremap smem area.\n");
+		release_mem_region(carminefb_fix.smem_start,
+				carminefb_fix.smem_len);
+		goto err_reg_smem;
+	}
+
+	ret = init_hardware(hw);
+	if (ret)
+		goto err_unmap_screen;
+
+	info = NULL;
+	if (fb_displays & CARMINE_USE_DISPLAY0) {
+		ret = alloc_carmine_fb(hw->v_regs + CARMINE_DISP0_REG,
+				hw->screen_mem, CARMINE_DISPLAY_MEM * 0,
+				device, &info);
+		if (ret)
+			goto err_deinit_hw;
+	}
+
+	hw->fb[0] = info;
+
+	info = NULL;
+	if (fb_displays & CARMINE_USE_DISPLAY1) {
+		ret = alloc_carmine_fb(hw->v_regs + CARMINE_DISP1_REG,
+				hw->screen_mem, CARMINE_DISPLAY_MEM * 1,
+				device, &info);
+		if (ret)
+			goto err_cleanup_fb0;
+	}
+
+	hw->fb[1] = info;
+	info = NULL;
+
+	pci_set_drvdata(dev, hw);
+	return 0;
+
+err_cleanup_fb0:
+	cleanup_fb_device(hw->fb[0]);
+err_deinit_hw:
+	/* disable clock, etc */
+	c_set_hw_reg(hw, CARMINE_CTL_REG + CARMINE_CTL_REG_CLOCK_ENABLE, 0);
+err_unmap_screen:
+	iounmap(hw->screen_mem);
+err_reg_smem:
+	release_mem_region(carminefb_fix.mmio_start, carminefb_fix.mmio_len);
+err_unmap_vregs:
+	iounmap(hw->v_regs);
+err_free_reg_mmio:
+	release_mem_region(carminefb_fix.mmio_start, carminefb_fix.mmio_len);
+err_free_hw:
+	kfree(hw);
+err_enable_pci:
+	pci_disable_device(dev);
+	return ret;
+}
+
+static void __devexit carminefb_remove(struct pci_dev *dev)
+{
+	struct carmine_hw *hw = pci_get_drvdata(dev);
+	struct fb_fix_screeninfo fix;
+	int i;
+
+	/* in case we use only fb1 and not fb1 */
+	if (hw->fb[0])
+		fix = hw->fb[0]->fix;
+	else
+		fix = hw->fb[1]->fix;
+
+	/* deactivate display(s) and switch clocks */
+	c_set_hw_reg(hw, CARMINE_DISP0_REG + CARMINE_DISP_REG_DCM1, 0);
+	c_set_hw_reg(hw, CARMINE_DISP1_REG + CARMINE_DISP_REG_DCM1, 0);
+	c_set_hw_reg(hw, CARMINE_CTL_REG + CARMINE_CTL_REG_CLOCK_ENABLE, 0);
+
+	for (i = 0; i < MAX_DISPLAY; i++)
+		cleanup_fb_device(hw->fb[i]);
+
+	iounmap(hw->screen_mem);
+	release_mem_region(fix.smem_start, fix.smem_len);
+	iounmap(hw->v_regs);
+	release_mem_region(fix.mmio_start, fix.mmio_len);
+
+	pci_set_drvdata(dev, NULL);
+	pci_disable_device(dev);
+	kfree(hw);
+}
+
+#define PCI_VENDOR_ID_FUJITU_LIMITED 0x10cf
+static struct pci_device_id carmine_devices[] __devinitdata = {
+{
+	PCI_DEVICE(PCI_VENDOR_ID_FUJITU_LIMITED, 0x202b)},
+	{0, 0, 0, 0, 0, 0, 0}
+};
+
+MODULE_DEVICE_TABLE(pci, carmine_devices);
+
+static struct pci_driver carmine_pci_driver = {
+	.name		= "carminefb",
+	.id_table	= carmine_devices,
+	.probe		= carminefb_probe,
+	.remove		= __devexit_p(carminefb_remove),
+};
+
+static int __init carminefb_init(void)
+{
+	if (!(fb_displays &
+		(CARMINE_USE_DISPLAY0 | CARMINE_USE_DISPLAY1))) {
+		printk(KERN_ERR "If you disable both displays than you don't "
+				"need the driver at all\n");
+		return -EINVAL;
+	}
+	return pci_register_driver(&carmine_pci_driver);
+}
+module_init(carminefb_init);
+
+static void __exit carminefb_cleanup(void)
+{
+	pci_unregister_driver(&carmine_pci_driver);
+}
+module_exit(carminefb_cleanup);
+
+MODULE_AUTHOR("Sebastian Siewior <bigeasy@linutronix.de>");
+MODULE_DESCRIPTION("Framebuffer driver for Fujitsu Carmine based devices");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/video/carminefb.h b/drivers/video/carminefb.h
new file mode 100644
index 000000000000..05306de0c6b6
--- /dev/null
+++ b/drivers/video/carminefb.h
@@ -0,0 +1,64 @@
+#ifndef CARMINE_CARMINE_H
+#define CARMINE_CARMINE_H
+
+#define CARMINE_MEMORY_BAR	2
+#define CARMINE_CONFIG_BAR	3
+
+#define MAX_DISPLAY	2
+#define CARMINE_DISPLAY_MEM	(800 * 600 * 4)
+#define CARMINE_TOTAL_DIPLAY_MEM	(CARMINE_DISPLAY_MEM * MAX_DISPLAY)
+
+#define CARMINE_USE_DISPLAY0	(1 << 0)
+#define CARMINE_USE_DISPLAY1	(1 << 1)
+
+/*
+ * This values work on the eval card. Custom boards may use different timings,
+ * here an example :)
+ */
+
+/* DRAM initialization values */
+#ifdef CONFIG_FB_CARMINE_DRAM_EVAL
+
+#define CARMINE_DFLT_IP_CLOCK_ENABLE		(0x03ff)
+#define CARMINE_DFLT_IP_DCTL_ADD		(0x05c3)
+#define CARMINE_DFLT_IP_DCTL_MODE		(0x0121)
+#define CARMINE_DFLT_IP_DCTL_EMODE		(0x8000)
+#define CARMINE_DFLT_IP_DCTL_SET_TIME1		(0x4749)
+#define CARMINE_DFLT_IP_DCTL_SET_TIME2		(0x2a22)
+#define CARMINE_DFLT_IP_DCTL_REFRESH		(0x0042)
+#define CARMINE_DFLT_IP_DCTL_STATES		(0x0003)
+#define CARMINE_DFLT_IP_DCTL_RESERVE0		(0x0020)
+#define CARMINE_DFLT_IP_DCTL_FIFO_DEPTH		(0x000f)
+#define CARMINE_DFLT_IP_DCTL_RESERVE2		(0x0000)
+#define CARMINE_DFLT_IP_DCTL_DDRIF1		(0x6646)
+#define CARMINE_DFLT_IP_DCTL_DDRIF2		(0x0055)
+#define CARMINE_DFLT_IP_DCTL_MODE_AFT_RST	(0x0021)
+#define CARMINE_DFLT_IP_DCTL_STATES_AFT_RST	(0x0002)
+#define CARMINE_DFLT_IP_DCTL_IO_CONT0		(0x0555)
+#define CARMINE_DFLT_IP_DCTL_IO_CONT1		(0x0555)
+#define CARMINE_DCTL_DLL_RESET			(1)
+#endif
+
+#ifdef CONFIG_CARMINE_DRAM_CUSTOM
+
+#define CARMINE_DFLT_IP_CLOCK_ENABLE		(0x03ff)
+#define CARMINE_DFLT_IP_DCTL_ADD		(0x03b2)
+#define CARMINE_DFLT_IP_DCTL_MODE		(0x0161)
+#define CARMINE_DFLT_IP_DCTL_EMODE		(0x8000)
+#define CARMINE_DFLT_IP_DCTL_SET_TIME1		(0x2628)
+#define CARMINE_DFLT_IP_DCTL_SET_TIME2		(0x1a09)
+#define CARMINE_DFLT_IP_DCTL_REFRESH		(0x00fe)
+#define CARMINE_DFLT_IP_DCTL_STATES		(0x0003)
+#define CARMINE_DFLT_IP_DCTL_RESERVE0		(0x0020)
+#define CARMINE_DFLT_IP_DCTL_FIFO_DEPTH		(0x000f)
+#define CARMINE_DFLT_IP_DCTL_RESERVE2		(0x0000)
+#define CARMINE_DFLT_IP_DCTL_DDRIF1		(0x0646)
+#define CARMINE_DFLT_IP_DCTL_DDRIF2		(0x55aa)
+#define CARMINE_DFLT_IP_DCTL_MODE_AFT_RST	(0x0061)
+#define CARMINE_DFLT_IP_DCTL_STATES_AFT_RST	(0x0002)
+#define CARMINE_DFLT_IP_DCTL_IO_CONT0		(0x0555)
+#define CARMINE_DFLT_IP_DCTL_IO_CONT1		(0x0555)
+#define CARMINE_DCTL_DLL_RESET			(1)
+#endif
+
+#endif
diff --git a/drivers/video/carminefb_regs.h b/drivers/video/carminefb_regs.h
new file mode 100644
index 000000000000..045215600b73
--- /dev/null
+++ b/drivers/video/carminefb_regs.h
@@ -0,0 +1,159 @@
+#ifndef _CARMINEFB_REGS_H
+#define _CARMINEFB_REGS_H
+
+#define CARMINE_OVERLAY_EXT_MODE	(0x00000002)
+#define CARMINE_GRAPH_REG		(0x00000000)
+#define CARMINE_DISP0_REG		(0x00100000)
+#define CARMINE_DISP1_REG		(0x00140000)
+#define CARMINE_WB_REG			(0x00180000)
+#define CARMINE_DCTL_REG		(0x00300000)
+#define CARMINE_CTL_REG			(0x00400000)
+#define CARMINE_WINDOW_MODE		(0x00000001)
+#define CARMINE_EXTEND_MODE		(CARMINE_WINDOW_MODE | \
+					CARMINE_OVERLAY_EXT_MODE)
+#define CARMINE_L0E			(1 << 16)
+#define CARMINE_L2E			(1 << 18)
+#define CARMINE_DEN			(1 << 31)
+
+#define CARMINE_EXT_CMODE_DIRECT24_RGBA		(0xC0000000)
+#define CARMINE_DCTL_REG_MODE_ADD		(0x00)
+#define CARMINE_DCTL_REG_SETTIME1_EMODE		(0x04)
+#define CARMINE_DCTL_REG_REFRESH_SETTIME2	(0x08)
+#define CARMINE_DCTL_REG_RSV0_STATES		(0x0C)
+#define CARMINE_DCTL_REG_RSV2_RSV1		(0x10)
+#define CARMINE_DCTL_REG_DDRIF2_DDRIF1		(0x14)
+#define CARMINE_DCTL_REG_IOCONT1_IOCONT0	(0x24)
+#define CARMINE_DCTL_REG_STATES_MASK		(0x000F)
+#define CARMINE_DCTL_INIT_WAIT_INTERVAL		(1)
+#define CARMINE_DCTL_INIT_WAIT_LIMIT		(5000)
+#define CARMINE_WB_REG_WBM_DEFAULT		(0x0001c020)
+#define CARMINE_DISP_REG_L0RM			(0x1880)
+#define CARMINE_DISP_REG_L0PX			(0x1884)
+#define CARMINE_DISP_REG_L0PY			(0x1888)
+#define CARMINE_DISP_REG_L2RM			(0x18A0)
+#define CARMINE_DISP_REG_L2PX			(0x18A4)
+#define CARMINE_DISP_REG_L2PY			(0x18A8)
+#define CARMINE_DISP_REG_L3RM			(0x18B0)
+#define CARMINE_DISP_REG_L3PX			(0x18B4)
+#define CARMINE_DISP_REG_L3PY			(0x18B8)
+#define CARMINE_DISP_REG_L4RM			(0x18C0)
+#define CARMINE_DISP_REG_L4PX			(0x18C4)
+#define CARMINE_DISP_REG_L4PY			(0x18C8)
+#define CARMINE_DISP_REG_L5RM			(0x18D0)
+#define CARMINE_DISP_REG_L5PX			(0x18D4)
+#define CARMINE_DISP_REG_L5PY			(0x18D8)
+#define CARMINE_DISP_REG_L6RM			(0x1924)
+#define CARMINE_DISP_REG_L6PX			(0x1928)
+#define CARMINE_DISP_REG_L6PY			(0x192C)
+#define CARMINE_DISP_REG_L7RM			(0x1964)
+#define CARMINE_DISP_REG_L7PX			(0x1968)
+#define CARMINE_DISP_REG_L7PY			(0x196C)
+#define CARMINE_WB_REG_WBM			(0x0004)
+#define CARMINE_DISP_HTP_SHIFT			(16)
+#define CARMINE_DISP_HDB_SHIFT			(16)
+#define CARMINE_DISP_HSW_SHIFT			(16)
+#define CARMINE_DISP_VSW_SHIFT			(24)
+#define CARMINE_DISP_VTR_SHIFT			(16)
+#define CARMINE_DISP_VDP_SHIFT			(16)
+#define CARMINE_CURSOR_CUTZ_MASK		(0x00000100)
+#define CARMINE_CURSOR0_PRIORITY_MASK		(0x00010000)
+#define CARMINE_CURSOR1_PRIORITY_MASK		(0x00020000)
+#define CARMINE_DISP_WIDTH_SHIFT		(16)
+#define CARMINE_DISP_WIN_H_SHIFT		(16)
+#define CARMINE_DISP_REG_H_TOTAL		(0x0004)
+#define CARMINE_DISP_REG_H_PERIOD		(0x0008)
+#define CARMINE_DISP_REG_V_H_W_H_POS		(0x000C)
+#define CARMINE_DISP_REG_V_TOTAL		(0x0010)
+#define CARMINE_DISP_REG_V_PERIOD_POS		(0x0014)
+#define CARMINE_DISP_REG_L0_MODE_W_H		(0x0020)
+#define CARMINE_DISP_REG_L0_ORG_ADR		(0x0024)
+#define CARMINE_DISP_REG_L0_DISP_ADR		(0x0028)
+#define CARMINE_DISP_REG_L0_DISP_POS		(0x002C)
+#define CARMINE_DISP_REG_L1_WIDTH		(0x0030)
+#define CARMINE_DISP_REG_L1_ORG_ADR		(0x0034)
+#define CARMINE_DISP_REG_L2_MODE_W_H		(0x0040)
+#define CARMINE_DISP_REG_L2_ORG_ADR1		(0x0044)
+#define CARMINE_DISP_REG_L2_DISP_ADR1		(0x0048)
+#define CARMINE_DISP_REG_L2_DISP_POS		(0x0054)
+#define CARMINE_DISP_REG_L3_MODE_W_H		(0x0058)
+#define CARMINE_DISP_REG_L3_ORG_ADR1		(0x005C)
+#define CARMINE_DISP_REG_L3_DISP_ADR1		(0x0060)
+#define CARMINE_DISP_REG_L3_DISP_POS		(0x006C)
+#define CARMINE_DISP_REG_L4_MODE_W_H		(0x0070)
+#define CARMINE_DISP_REG_L4_ORG_ADR1		(0x0074)
+#define CARMINE_DISP_REG_L4_DISP_ADR1		(0x0078)
+#define CARMINE_DISP_REG_L4_DISP_POS		(0x0084)
+#define CARMINE_DISP_REG_L5_MODE_W_H		(0x0088)
+#define CARMINE_DISP_REG_L5_ORG_ADR1		(0x008C)
+#define CARMINE_DISP_REG_L5_DISP_ADR1		(0x0090)
+#define CARMINE_DISP_REG_L5_DISP_POS		(0x009C)
+#define CARMINE_DISP_REG_CURSOR_MODE		(0x00A0)
+#define CARMINE_DISP_REG_CUR1_POS		(0x00A8)
+#define CARMINE_DISP_REG_CUR2_POS		(0x00B0)
+#define CARMINE_DISP_REG_C_TRANS		(0x00BC)
+#define CARMINE_DISP_REG_MLMR_TRANS		(0x00C0)
+#define CARMINE_DISP_REG_L0_EXT_MODE		(0x0110)
+#define CARMINE_DISP_REG_L0_WIN_POS		(0x0114)
+#define CARMINE_DISP_REG_L0_WIN_SIZE		(0x0118)
+#define CARMINE_DISP_REG_L1_EXT_MODE		(0x0120)
+#define CARMINE_DISP_REG_L1_WIN_POS		(0x0124)
+#define CARMINE_DISP_REG_L1_WIN_SIZE		(0x0128)
+#define CARMINE_DISP_REG_L2_EXT_MODE		(0x0130)
+#define CARMINE_DISP_REG_L2_WIN_POS		(0x0134)
+#define CARMINE_DISP_REG_L2_WIN_SIZE		(0x0138)
+#define CARMINE_DISP_REG_L3_EXT_MODE		(0x0140)
+#define CARMINE_DISP_REG_L3_WIN_POS		(0x0144)
+#define CARMINE_DISP_REG_L3_WIN_SIZE		(0x0148)
+#define CARMINE_DISP_REG_L4_EXT_MODE		(0x0150)
+#define CARMINE_DISP_REG_L4_WIN_POS		(0x0154)
+#define CARMINE_DISP_REG_L4_WIN_SIZE		(0x0158)
+#define CARMINE_DISP_REG_L5_EXT_MODE		(0x0160)
+#define CARMINE_DISP_REG_L5_WIN_POS		(0x0164)
+#define CARMINE_DISP_REG_L5_WIN_SIZE		(0x0168)
+#define CARMINE_DISP_REG_L6_EXT_MODE		(0x1918)
+#define CARMINE_DISP_REG_L6_WIN_POS		(0x191c)
+#define CARMINE_DISP_REG_L6_WIN_SIZE		(0x1920)
+#define CARMINE_DISP_REG_L7_EXT_MODE		(0x1958)
+#define CARMINE_DISP_REG_L7_WIN_POS		(0x195c)
+#define CARMINE_DISP_REG_L7_WIN_SIZE		(0x1960)
+#define CARMINE_DISP_REG_BLEND_MODE_L0		(0x00B4)
+#define CARMINE_DISP_REG_BLEND_MODE_L1		(0x0188)
+#define CARMINE_DISP_REG_BLEND_MODE_L2		(0x018C)
+#define CARMINE_DISP_REG_BLEND_MODE_L3		(0x0190)
+#define CARMINE_DISP_REG_BLEND_MODE_L4		(0x0194)
+#define CARMINE_DISP_REG_BLEND_MODE_L5		(0x0198)
+#define CARMINE_DISP_REG_BLEND_MODE_L6		(0x1990)
+#define CARMINE_DISP_REG_BLEND_MODE_L7		(0x1994)
+#define CARMINE_DISP_REG_L0_TRANS		(0x01A0)
+#define CARMINE_DISP_REG_L1_TRANS		(0x01A4)
+#define CARMINE_DISP_REG_L2_TRANS		(0x01A8)
+#define CARMINE_DISP_REG_L3_TRANS		(0x01AC)
+#define CARMINE_DISP_REG_L4_TRANS		(0x01B0)
+#define CARMINE_DISP_REG_L5_TRANS		(0x01B4)
+#define CARMINE_DISP_REG_L6_TRANS		(0x1998)
+#define CARMINE_DISP_REG_L7_TRANS		(0x199c)
+#define CARMINE_EXTEND_MODE_MASK		(0x00000003)
+#define CARMINE_DISP_DCM_MASK			(0x0000FFFF)
+#define CARMINE_DISP_REG_DCM1			(0x0100)
+#define CARMINE_DISP_WIDTH_UNIT			(64)
+#define CARMINE_DISP_REG_L6_MODE_W_H		(0x1900)
+#define CARMINE_DISP_REG_L6_ORG_ADR1		(0x1904)
+#define CARMINE_DISP_REG_L6_DISP_ADR0		(0x1908)
+#define CARMINE_DISP_REG_L6_DISP_POS		(0x1914)
+#define CARMINE_DISP_REG_L7_MODE_W_H		(0x1940)
+#define CARMINE_DISP_REG_L7_ORG_ADR1		(0x1944)
+#define CARMINE_DISP_REG_L7_DISP_ADR0		(0x1948)
+#define CARMINE_DISP_REG_L7_DISP_POS		(0x1954)
+#define CARMINE_CTL_REG_CLOCK_ENABLE		(0x000C)
+#define CARMINE_CTL_REG_SOFTWARE_RESET		(0x0010)
+#define CARMINE_CTL_REG_IST_MASK_ALL		(0x07FFFFFF)
+#define CARMINE_GRAPH_REG_VRINTM		(0x00028064)
+#define CARMINE_GRAPH_REG_VRERRM		(0x0002806C)
+#define CARMINE_GRAPH_REG_DC_OFFSET_PX		(0x0004005C)
+#define CARMINE_GRAPH_REG_DC_OFFSET_PY		(0x00040060)
+#define CARMINE_GRAPH_REG_DC_OFFSET_LX		(0x00040064)
+#define CARMINE_GRAPH_REG_DC_OFFSET_LY		(0x00040068)
+#define CARMINE_GRAPH_REG_DC_OFFSET_TX		(0x0004006C)
+#define CARMINE_GRAPH_REG_DC_OFFSET_TY		(0x00040070)
+
+#endif

From 306fa6f60a2870b7a9827a64e1b45cd35a9549aa Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:30:50 -0700
Subject: [PATCH 0443/1171] tridentfb: replace macros with functions

This patch replaces macros with static functions and puts tridentfb_par
pointer as the first argument of these functions.  These is a step toward
multihead support.

Additionally, bogus TRIDENT_MMIO define is removed as the driver supports
graphics cards only through the mmio mode.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 526 ++++++++++++++++++++------------------
 1 file changed, 283 insertions(+), 243 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index beefab2992c0..3e8a1ef892ce 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -137,28 +137,34 @@ static int iscyber(int id)
 
 #define CRT 0x3D0		/* CRTC registers offset for color display */
 
-#ifndef TRIDENT_MMIO
-	#define TRIDENT_MMIO 1
-#endif
-
-#if TRIDENT_MMIO
-	#define t_outb(val, reg)	writeb(val,((struct tridentfb_par *)(fb_info.par))->io_virt + reg)
-	#define t_inb(reg)	readb(((struct tridentfb_par*)(fb_info.par))->io_virt + reg)
-#else
-	#define t_outb(val, reg) outb(val, reg)
-	#define t_inb(reg) inb(reg)
-#endif
+static inline void t_outb(struct tridentfb_par *p, u8 val, u16 reg)
+{
+	fb_writeb(val, p->io_virt + reg);
+}
 
+static inline u8 t_inb(struct tridentfb_par *p, u16 reg)
+{
+	return fb_readb(p->io_virt + reg);
+}
 
 static struct accel_switch {
-	void (*init_accel) (int, int);
-	void (*wait_engine) (void);
-	void (*fill_rect) (u32, u32, u32, u32, u32, u32);
-	void (*copy_rect) (u32, u32, u32, u32, u32, u32);
+	void (*init_accel) (struct tridentfb_par *, int, int);
+	void (*wait_engine) (struct tridentfb_par *);
+	void (*fill_rect)
+		(struct tridentfb_par *par, u32, u32, u32, u32, u32, u32);
+	void (*copy_rect)
+		(struct tridentfb_par *par, u32, u32, u32, u32, u32, u32);
 } *acc;
 
-#define writemmr(r, v)	writel(v, ((struct tridentfb_par *)fb_info.par)->io_virt + r)
-#define readmmr(r)	readl(((struct tridentfb_par *)fb_info.par)->io_virt + r)
+static inline void writemmr(struct tridentfb_par *par, u16 r, u32 v)
+{
+	fb_writel(v, par->io_virt + r);
+}
+
+static inline u32 readmmr(struct tridentfb_par *par, u16 r)
+{
+	return fb_readl(par->io_virt + r);
+}
 
 /*
  * Blade specific acceleration.
@@ -176,7 +182,7 @@ static struct accel_switch {
 
 #define ROP_S	0xCC
 
-static void blade_init_accel(int pitch, int bpp)
+static void blade_init_accel(struct tridentfb_par *par, int pitch, int bpp)
 {
 	int v1 = (pitch >> 3) << 20;
 	int tmp = 0, v2;
@@ -196,33 +202,35 @@ static void blade_init_accel(int pitch, int bpp)
 		break;
 	}
 	v2 = v1 | (tmp << 29);
-	writemmr(0x21C0, v2);
-	writemmr(0x21C4, v2);
-	writemmr(0x21B8, v2);
-	writemmr(0x21BC, v2);
-	writemmr(0x21D0, v1);
-	writemmr(0x21D4, v1);
-	writemmr(0x21C8, v1);
-	writemmr(0x21CC, v1);
-	writemmr(0x216C, 0);
+	writemmr(par, 0x21C0, v2);
+	writemmr(par, 0x21C4, v2);
+	writemmr(par, 0x21B8, v2);
+	writemmr(par, 0x21BC, v2);
+	writemmr(par, 0x21D0, v1);
+	writemmr(par, 0x21D4, v1);
+	writemmr(par, 0x21C8, v1);
+	writemmr(par, 0x21CC, v1);
+	writemmr(par, 0x216C, 0);
 }
 
-static void blade_wait_engine(void)
+static void blade_wait_engine(struct tridentfb_par *par)
 {
-	while (readmmr(STA) & 0xFA800000) ;
+	while (readmmr(par, STA) & 0xFA800000) ;
 }
 
-static void blade_fill_rect(u32 x, u32 y, u32 w, u32 h, u32 c, u32 rop)
+static void blade_fill_rect(struct tridentfb_par *par,
+			    u32 x, u32 y, u32 w, u32 h, u32 c, u32 rop)
 {
-	writemmr(CLR, c);
-	writemmr(ROP, rop ? 0x66 : ROP_S);
-	writemmr(CMD, 0x20000000 | 1 << 19 | 1 << 4 | 2 << 2);
+	writemmr(par, CLR, c);
+	writemmr(par, ROP, rop ? 0x66 : ROP_S);
+	writemmr(par, CMD, 0x20000000 | 1 << 19 | 1 << 4 | 2 << 2);
 
-	writemmr(DR1, point(x, y));
-	writemmr(DR2, point(x + w - 1, y + h - 1));
+	writemmr(par, DR1, point(x, y));
+	writemmr(par, DR2, point(x + w - 1, y + h - 1));
 }
 
-static void blade_copy_rect(u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h)
+static void blade_copy_rect(struct tridentfb_par *par,
+			    u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h)
 {
 	u32 s1, s2, d1, d2;
 	int direction = 2;
@@ -234,13 +242,13 @@ static void blade_copy_rect(u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h)
 	if ((y1 > y2) || ((y1 == y2) && (x1 > x2)))
 		direction = 0;
 
-	writemmr(ROP, ROP_S);
-	writemmr(CMD, 0xE0000000 | 1 << 19 | 1 << 4 | 1 << 2 | direction);
+	writemmr(par, ROP, ROP_S);
+	writemmr(par, CMD, 0xE0000000 | 1 << 19 | 1 << 4 | 1 << 2 | direction);
 
-	writemmr(SR1, direction ? s2 : s1);
-	writemmr(SR2, direction ? s1 : s2);
-	writemmr(DR1, direction ? d2 : d1);
-	writemmr(DR2, direction ? d1 : d2);
+	writemmr(par, SR1, direction ? s2 : s1);
+	writemmr(par, SR2, direction ? s1 : s2);
+	writemmr(par, DR1, direction ? d2 : d1);
+	writemmr(par, DR2, direction ? d1 : d2);
 }
 
 static struct accel_switch accel_blade = {
@@ -257,7 +265,7 @@ static struct accel_switch accel_blade = {
 #define ROP_P 0xF0
 #define masked_point(x, y) ((y & 0xffff)<<16|(x & 0xffff))
 
-static void xp_init_accel(int pitch, int bpp)
+static void xp_init_accel(struct tridentfb_par *par, int pitch, int bpp)
 {
 	int tmp = 0, v1;
 	unsigned char x = 0;
@@ -293,7 +301,7 @@ static void xp_init_accel(int pitch, int bpp)
 		break;
 	}
 
-	t_outb(x, 0x2125);
+	t_outb(par, x, 0x2125);
 
 	eng_oper = x | 0x40;
 
@@ -313,12 +321,12 @@ static void xp_init_accel(int pitch, int bpp)
 
 	v1 = pitch << tmp;
 
-	writemmr(0x2154, v1);
-	writemmr(0x2150, v1);
-	t_outb(3, 0x2126);
+	writemmr(par, 0x2154, v1);
+	writemmr(par, 0x2150, v1);
+	t_outb(par, 3, 0x2126);
 }
 
-static void xp_wait_engine(void)
+static void xp_wait_engine(struct tridentfb_par *par)
 {
 	int busy;
 	int count, timeout;
@@ -326,7 +334,7 @@ static void xp_wait_engine(void)
 	count = 0;
 	timeout = 0;
 	for (;;) {
-		busy = t_inb(STA) & 0x80;
+		busy = t_inb(par, STA) & 0x80;
 		if (busy != 0x80)
 			return;
 		count++;
@@ -336,25 +344,27 @@ static void xp_wait_engine(void)
 			timeout++;
 			if (timeout == 8) {
 				/* Reset engine */
-				t_outb(0x00, 0x2120);
+				t_outb(par, 0x00, 0x2120);
 				return;
 			}
 		}
 	}
 }
 
-static void xp_fill_rect(u32 x, u32 y, u32 w, u32 h, u32 c, u32 rop)
+static void xp_fill_rect(struct tridentfb_par *par,
+			 u32 x, u32 y, u32 w, u32 h, u32 c, u32 rop)
 {
-	writemmr(0x2127, ROP_P);
-	writemmr(0x2158, c);
-	writemmr(0x2128, 0x4000);
-	writemmr(0x2140, masked_point(h, w));
-	writemmr(0x2138, masked_point(y, x));
-	t_outb(0x01, 0x2124);
-	t_outb(eng_oper, 0x2125);
+	writemmr(par, 0x2127, ROP_P);
+	writemmr(par, 0x2158, c);
+	writemmr(par, 0x2128, 0x4000);
+	writemmr(par, 0x2140, masked_point(h, w));
+	writemmr(par, 0x2138, masked_point(y, x));
+	t_outb(par, 0x01, 0x2124);
+	t_outb(par, eng_oper, 0x2125);
 }
 
-static void xp_copy_rect(u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h)
+static void xp_copy_rect(struct tridentfb_par *par,
+			 u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h)
 {
 	int direction;
 	u32 x1_tmp, x2_tmp, y1_tmp, y2_tmp;
@@ -379,12 +389,12 @@ static void xp_copy_rect(u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h)
 		y2_tmp = y2;
 	}
 
-	writemmr(0x2128, direction);
-	t_outb(ROP_S, 0x2127);
-	writemmr(0x213C, masked_point(y1_tmp, x1_tmp));
-	writemmr(0x2138, masked_point(y2_tmp, x2_tmp));
-	writemmr(0x2140, masked_point(h, w));
-	t_outb(0x01, 0x2124);
+	writemmr(par, 0x2128, direction);
+	t_outb(par, ROP_S, 0x2127);
+	writemmr(par, 0x213C, masked_point(y1_tmp, x1_tmp));
+	writemmr(par, 0x2138, masked_point(y2_tmp, x2_tmp));
+	writemmr(par, 0x2140, masked_point(h, w));
+	t_outb(par, 0x01, 0x2124);
 }
 
 static struct accel_switch accel_xp = {
@@ -397,7 +407,7 @@ static struct accel_switch accel_xp = {
 /*
  * Image specific acceleration functions
  */
-static void image_init_accel(int pitch, int bpp)
+static void image_init_accel(struct tridentfb_par *par, int pitch, int bpp)
 {
 	int tmp = 0;
 	switch (bpp) {
@@ -415,40 +425,42 @@ static void image_init_accel(int pitch, int bpp)
 		tmp = 2;
 		break;
 	}
-	writemmr(0x2120, 0xF0000000);
-	writemmr(0x2120, 0x40000000 | tmp);
-	writemmr(0x2120, 0x80000000);
-	writemmr(0x2144, 0x00000000);
-	writemmr(0x2148, 0x00000000);
-	writemmr(0x2150, 0x00000000);
-	writemmr(0x2154, 0x00000000);
-	writemmr(0x2120, 0x60000000 | (pitch << 16) | pitch);
-	writemmr(0x216C, 0x00000000);
-	writemmr(0x2170, 0x00000000);
-	writemmr(0x217C, 0x00000000);
-	writemmr(0x2120, 0x10000000);
-	writemmr(0x2130, (2047 << 16) | 2047);
+	writemmr(par, 0x2120, 0xF0000000);
+	writemmr(par, 0x2120, 0x40000000 | tmp);
+	writemmr(par, 0x2120, 0x80000000);
+	writemmr(par, 0x2144, 0x00000000);
+	writemmr(par, 0x2148, 0x00000000);
+	writemmr(par, 0x2150, 0x00000000);
+	writemmr(par, 0x2154, 0x00000000);
+	writemmr(par, 0x2120, 0x60000000 | (pitch << 16) | pitch);
+	writemmr(par, 0x216C, 0x00000000);
+	writemmr(par, 0x2170, 0x00000000);
+	writemmr(par, 0x217C, 0x00000000);
+	writemmr(par, 0x2120, 0x10000000);
+	writemmr(par, 0x2130, (2047 << 16) | 2047);
 }
 
-static void image_wait_engine(void)
+static void image_wait_engine(struct tridentfb_par *par)
 {
-	while (readmmr(0x2164) & 0xF0000000) ;
+	while (readmmr(par, 0x2164) & 0xF0000000) ;
 }
 
-static void image_fill_rect(u32 x, u32 y, u32 w, u32 h, u32 c, u32 rop)
+static void image_fill_rect(struct tridentfb_par *par,
+			    u32 x, u32 y, u32 w, u32 h, u32 c, u32 rop)
 {
-	writemmr(0x2120, 0x80000000);
-	writemmr(0x2120, 0x90000000 | ROP_S);
+	writemmr(par, 0x2120, 0x80000000);
+	writemmr(par, 0x2120, 0x90000000 | ROP_S);
 
-	writemmr(0x2144, c);
+	writemmr(par, 0x2144, c);
 
-	writemmr(DR1, point(x, y));
-	writemmr(DR2, point(x + w - 1, y + h - 1));
+	writemmr(par, DR1, point(x, y));
+	writemmr(par, DR2, point(x + w - 1, y + h - 1));
 
-	writemmr(0x2124, 0x80000000 | 3 << 22 | 1 << 10 | 1 << 9);
+	writemmr(par, 0x2124, 0x80000000 | 3 << 22 | 1 << 10 | 1 << 9);
 }
 
-static void image_copy_rect(u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h)
+static void image_copy_rect(struct tridentfb_par *par,
+			    u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h)
 {
 	u32 s1, s2, d1, d2;
 	int direction = 2;
@@ -460,14 +472,15 @@ static void image_copy_rect(u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h)
 	if ((y1 > y2) || ((y1 == y2) && (x1 > x2)))
 		direction = 0;
 
-	writemmr(0x2120, 0x80000000);
-	writemmr(0x2120, 0x90000000 | ROP_S);
+	writemmr(par, 0x2120, 0x80000000);
+	writemmr(par, 0x2120, 0x90000000 | ROP_S);
 
-	writemmr(SR1, direction ? s2 : s1);
-	writemmr(SR2, direction ? s1 : s2);
-	writemmr(DR1, direction ? d2 : d1);
-	writemmr(DR2, direction ? d1 : d2);
-	writemmr(0x2124, 0x80000000 | 1 << 22 | 1 << 10 | 1 << 7 | direction);
+	writemmr(par, SR1, direction ? s2 : s1);
+	writemmr(par, SR2, direction ? s1 : s2);
+	writemmr(par, DR1, direction ? d2 : d1);
+	writemmr(par, DR2, direction ? d1 : d2);
+	writemmr(par, 0x2124,
+		 0x80000000 | 1 << 22 | 1 << 10 | 1 << 7 | direction);
 }
 
 static struct accel_switch accel_image = {
@@ -484,6 +497,7 @@ static struct accel_switch accel_image = {
 static void tridentfb_fillrect(struct fb_info *info,
 			       const struct fb_fillrect *fr)
 {
+	struct tridentfb_par *par = info->par;
 	int bpp = info->var.bits_per_pixel;
 	int col = 0;
 
@@ -502,14 +516,18 @@ static void tridentfb_fillrect(struct fb_info *info,
 		break;
 	}
 
-	acc->fill_rect(fr->dx, fr->dy, fr->width, fr->height, col, fr->rop);
-	acc->wait_engine();
+	acc->fill_rect(par, fr->dx, fr->dy, fr->width,
+		       fr->height, col, fr->rop);
+	acc->wait_engine(par);
 }
 static void tridentfb_copyarea(struct fb_info *info,
 			       const struct fb_copyarea *ca)
 {
-	acc->copy_rect(ca->sx, ca->sy, ca->dx, ca->dy, ca->width, ca->height);
-	acc->wait_engine();
+	struct tridentfb_par *par = info->par;
+
+	acc->copy_rect(par, ca->sx, ca->sy, ca->dx, ca->dy,
+		       ca->width, ca->height);
+	acc->wait_engine(par);
 }
 #else /* !CONFIG_FB_TRIDENT_ACCEL */
 #define tridentfb_fillrect cfb_fillrect
@@ -521,49 +539,51 @@ static void tridentfb_copyarea(struct fb_info *info,
  * Hardware access functions
  */
 
-static inline unsigned char read3X4(int reg)
+static inline unsigned char read3X4(struct tridentfb_par *par, int reg)
 {
-	struct tridentfb_par *par = (struct tridentfb_par *)fb_info.par;
 	writeb(reg, par->io_virt + CRT + 4);
 	return readb(par->io_virt + CRT + 5);
 }
 
-static inline void write3X4(int reg, unsigned char val)
+static inline void write3X4(struct tridentfb_par *par, int reg,
+			    unsigned char val)
 {
-	struct tridentfb_par *par = (struct tridentfb_par *)fb_info.par;
 	writeb(reg, par->io_virt + CRT + 4);
 	writeb(val, par->io_virt + CRT + 5);
 }
 
-static inline unsigned char read3C4(int reg)
+static inline unsigned char read3C4(struct tridentfb_par *par, int reg)
 {
-	t_outb(reg, 0x3C4);
-	return t_inb(0x3C5);
+	t_outb(par, reg, 0x3C4);
+	return t_inb(par, 0x3C5);
 }
 
-static inline void write3C4(int reg, unsigned char val)
+static inline void write3C4(struct tridentfb_par *par, int reg,
+			    unsigned char val)
 {
-	t_outb(reg, 0x3C4);
-	t_outb(val, 0x3C5);
+	t_outb(par, reg, 0x3C4);
+	t_outb(par, val, 0x3C5);
 }
 
-static inline unsigned char read3CE(int reg)
+static inline unsigned char read3CE(struct tridentfb_par *par, int reg)
 {
-	t_outb(reg, 0x3CE);
-	return t_inb(0x3CF);
+	t_outb(par, reg, 0x3CE);
+	return t_inb(par, 0x3CF);
 }
 
-static inline void writeAttr(int reg, unsigned char val)
+static inline void writeAttr(struct tridentfb_par *par, int reg,
+			     unsigned char val)
 {
-	readb(((struct tridentfb_par *)fb_info.par)->io_virt + CRT + 0x0A);	/* flip-flop to index */
-	t_outb(reg, 0x3C0);
-	t_outb(val, 0x3C0);
+	fb_readb(par->io_virt + CRT + 0x0A);	/* flip-flop to index */
+	t_outb(par, reg, 0x3C0);
+	t_outb(par, val, 0x3C0);
 }
 
-static inline void write3CE(int reg, unsigned char val)
+static inline void write3CE(struct tridentfb_par *par, int reg,
+			    unsigned char val)
 {
-	t_outb(reg, 0x3CE);
-	t_outb(val, 0x3CF);
+	t_outb(par, reg, 0x3CE);
+	t_outb(par, val, 0x3CF);
 }
 
 static void enable_mmio(void)
@@ -581,32 +601,35 @@ static void enable_mmio(void)
 	outb(inb(0x3D5) | 0x01, 0x3D5);
 }
 
-static void disable_mmio(void)
+static void disable_mmio(struct tridentfb_par *par)
 {
 	/* Goto New Mode */
-	t_outb(0x0B, 0x3C4);
-	t_inb(0x3C5);
+	t_outb(par, 0x0B, 0x3C4);
+	t_inb(par, 0x3C5);
 
 	/* Unprotect registers */
-	t_outb(NewMode1, 0x3C4);
-	t_outb(0x80, 0x3C5);
+	t_outb(par, NewMode1, 0x3C4);
+	t_outb(par, 0x80, 0x3C5);
 
 	/* Disable MMIO */
-	t_outb(PCIReg, 0x3D4);
-	t_outb(t_inb(0x3D5) & ~0x01, 0x3D5);
+	t_outb(par, PCIReg, 0x3D4);
+	t_outb(par, t_inb(par, 0x3D5) & ~0x01, 0x3D5);
 }
 
-#define crtc_unlock()	write3X4(CRTVSyncEnd, read3X4(CRTVSyncEnd) & 0x7F)
+static void crtc_unlock(struct tridentfb_par *par)
+{
+	write3X4(par, CRTVSyncEnd, read3X4(par, CRTVSyncEnd) & 0x7F);
+}
 
 /*  Return flat panel's maximum x resolution */
-static int __devinit get_nativex(void)
+static int __devinit get_nativex(struct tridentfb_par *par)
 {
 	int x, y, tmp;
 
 	if (nativex)
 		return nativex;
 
-	tmp = (read3CE(VertStretch) >> 4) & 3;
+	tmp = (read3CE(par, VertStretch) >> 4) & 3;
 
 	switch (tmp) {
 	case 0:
@@ -632,44 +655,45 @@ static int __devinit get_nativex(void)
 }
 
 /* Set pitch */
-static void set_lwidth(int width)
+static void set_lwidth(struct tridentfb_par *par, int width)
 {
-	write3X4(Offset, width & 0xFF);
-	write3X4(AddColReg,
-		 (read3X4(AddColReg) & 0xCF) | ((width & 0x300) >> 4));
+	write3X4(par, Offset, width & 0xFF);
+	write3X4(par, AddColReg,
+		 (read3X4(par, AddColReg) & 0xCF) | ((width & 0x300) >> 4));
 }
 
 /* For resolutions smaller than FP resolution stretch */
-static void screen_stretch(void)
+static void screen_stretch(struct tridentfb_par *par)
 {
 	if (chip_id != CYBERBLADEXPAi1)
-		write3CE(BiosReg, 0);
+		write3CE(par, BiosReg, 0);
 	else
-		write3CE(BiosReg, 8);
-	write3CE(VertStretch, (read3CE(VertStretch) & 0x7C) | 1);
-	write3CE(HorStretch, (read3CE(HorStretch) & 0x7C) | 1);
+		write3CE(par, BiosReg, 8);
+	write3CE(par, VertStretch, (read3CE(par, VertStretch) & 0x7C) | 1);
+	write3CE(par, HorStretch, (read3CE(par, HorStretch) & 0x7C) | 1);
 }
 
 /* For resolutions smaller than FP resolution center */
-static void screen_center(void)
+static void screen_center(struct tridentfb_par *par)
 {
-	write3CE(VertStretch, (read3CE(VertStretch) & 0x7C) | 0x80);
-	write3CE(HorStretch, (read3CE(HorStretch) & 0x7C) | 0x80);
+	write3CE(par, VertStretch, (read3CE(par, VertStretch) & 0x7C) | 0x80);
+	write3CE(par, HorStretch, (read3CE(par, HorStretch) & 0x7C) | 0x80);
 }
 
 /* Address of first shown pixel in display memory */
-static void set_screen_start(int base)
+static void set_screen_start(struct tridentfb_par *par, int base)
 {
-	write3X4(StartAddrLow, base & 0xFF);
-	write3X4(StartAddrHigh, (base & 0xFF00) >> 8);
-	write3X4(CRTCModuleTest,
-		 (read3X4(CRTCModuleTest) & 0xDF) | ((base & 0x10000) >> 11));
-	write3X4(CRTHiOrd,
-		 (read3X4(CRTHiOrd) & 0xF8) | ((base & 0xE0000) >> 17));
+	u8 tmp;
+	write3X4(par, StartAddrLow, base & 0xFF);
+	write3X4(par, StartAddrHigh, (base & 0xFF00) >> 8);
+	tmp = read3X4(par, CRTCModuleTest) & 0xDF;
+	write3X4(par, CRTCModuleTest, tmp | ((base & 0x10000) >> 11));
+	tmp = read3X4(par, CRTHiOrd) & 0xF8;
+	write3X4(par, CRTHiOrd, tmp | ((base & 0xE0000) >> 17));
 }
 
 /* Set dotclock frequency */
-static void set_vclk(unsigned long freq)
+static void set_vclk(struct tridentfb_par *par, unsigned long freq)
 {
 	int m, n, k;
 	unsigned long f, fi, d, di;
@@ -690,8 +714,8 @@ static void set_vclk(unsigned long freq)
 					break;
 			}
 	if (chip3D) {
-		write3C4(ClockHigh, hi);
-		write3C4(ClockLow, lo);
+		write3C4(par, ClockHigh, hi);
+		write3C4(par, ClockLow, lo);
 	} else {
 		outb(lo, 0x43C8);
 		outb(hi, 0x43C9);
@@ -700,9 +724,9 @@ static void set_vclk(unsigned long freq)
 }
 
 /* Set number of lines for flat panels*/
-static void set_number_of_lines(int lines)
+static void set_number_of_lines(struct tridentfb_par *par, int lines)
 {
-	int tmp = read3CE(CyberEnhance) & 0x8F;
+	int tmp = read3CE(par, CyberEnhance) & 0x8F;
 	if (lines > 1024)
 		tmp |= 0x50;
 	else if (lines > 768)
@@ -711,24 +735,24 @@ static void set_number_of_lines(int lines)
 		tmp |= 0x20;
 	else if (lines > 480)
 		tmp |= 0x10;
-	write3CE(CyberEnhance, tmp);
+	write3CE(par, CyberEnhance, tmp);
 }
 
 /*
  * If we see that FP is active we assume we have one.
  * Otherwise we have a CRT display.User can override.
  */
-static unsigned int __devinit get_displaytype(void)
+static unsigned int __devinit get_displaytype(struct tridentfb_par *par)
 {
 	if (fp)
 		return DISPLAY_FP;
 	if (crt || !chipcyber)
 		return DISPLAY_CRT;
-	return (read3CE(FPConfig) & 0x10) ? DISPLAY_FP : DISPLAY_CRT;
+	return (read3CE(par, FPConfig) & 0x10) ? DISPLAY_FP : DISPLAY_CRT;
 }
 
 /* Try detecting the video memory size */
-static unsigned int __devinit get_memsize(void)
+static unsigned int __devinit get_memsize(struct tridentfb_par *par)
 {
 	unsigned char tmp, tmp2;
 	unsigned int k;
@@ -742,7 +766,7 @@ static unsigned int __devinit get_memsize(void)
 			k = 2560 * Kb;
 			break;
 		default:
-			tmp = read3X4(SPR) & 0x0F;
+			tmp = read3X4(par, SPR) & 0x0F;
 			switch (tmp) {
 
 			case 0x01:
@@ -774,7 +798,7 @@ static unsigned int __devinit get_memsize(void)
 				break;
 			case 0x0E:		/* XP */
 
-				tmp2 = read3C4(0xC1);
+				tmp2 = read3C4(par, 0xC1);
 				switch (tmp2) {
 				case 0x00:
 					k = 20 * Mb;
@@ -862,6 +886,7 @@ static int tridentfb_check_var(struct fb_var_screeninfo *var,
 static int tridentfb_pan_display(struct fb_var_screeninfo *var,
 				 struct fb_info *info)
 {
+	struct tridentfb_par *par = info->par;
 	unsigned int offset;
 
 	debug("enter\n");
@@ -869,13 +894,20 @@ static int tridentfb_pan_display(struct fb_var_screeninfo *var,
 		* var->bits_per_pixel / 32;
 	info->var.xoffset = var->xoffset;
 	info->var.yoffset = var->yoffset;
-	set_screen_start(offset);
+	set_screen_start(par, offset);
 	debug("exit\n");
 	return 0;
 }
 
-#define shadowmode_on()  write3CE(CyberControl, read3CE(CyberControl) | 0x81)
-#define shadowmode_off() write3CE(CyberControl, read3CE(CyberControl) & 0x7E)
+static void shadowmode_on(struct tridentfb_par *par)
+{
+	write3CE(par, CyberControl, read3CE(par, CyberControl) | 0x81);
+}
+
+static void shadowmode_off(struct tridentfb_par *par)
+{
+	write3CE(par, CyberControl, read3CE(par, CyberControl) & 0x7E);
+}
 
 /* Set the hardware to the requested video mode */
 static int tridentfb_set_par(struct fb_info *info)
@@ -905,8 +937,8 @@ static int tridentfb_set_par(struct fb_info *info)
 	vblankstart = var->yres;
 	vblankend = vtotal + 2;
 
-	crtc_unlock();
-	write3CE(CyberControl, 8);
+	crtc_unlock(par);
+	write3CE(par, CyberControl, 8);
 
 	if (flatpanel && var->xres < nativex) {
 		/*
@@ -914,35 +946,36 @@ static int tridentfb_set_par(struct fb_info *info)
 		 * than requested resolution decide whether
 		 * we stretch or center
 		 */
-		t_outb(0xEB, 0x3C2);
+		t_outb(par, 0xEB, 0x3C2);
 
-		shadowmode_on();
+		shadowmode_on(par);
 
 		if (center)
-			screen_center();
+			screen_center(par);
 		else if (stretch)
-			screen_stretch();
+			screen_stretch(par);
 
 	} else {
-		t_outb(0x2B, 0x3C2);
-		write3CE(CyberControl, 8);
+		t_outb(par, 0x2B, 0x3C2);
+		write3CE(par, CyberControl, 8);
 	}
 
 	/* vertical timing values */
-	write3X4(CRTVTotal, vtotal & 0xFF);
-	write3X4(CRTVDispEnd, vdispend & 0xFF);
-	write3X4(CRTVSyncStart, vsyncstart & 0xFF);
-	write3X4(CRTVSyncEnd, (vsyncend & 0x0F));
-	write3X4(CRTVBlankStart, vblankstart & 0xFF);
-	write3X4(CRTVBlankEnd, 0 /* p->vblankend & 0xFF */ );
+	write3X4(par, CRTVTotal, vtotal & 0xFF);
+	write3X4(par, CRTVDispEnd, vdispend & 0xFF);
+	write3X4(par, CRTVSyncStart, vsyncstart & 0xFF);
+	write3X4(par, CRTVSyncEnd, (vsyncend & 0x0F));
+	write3X4(par, CRTVBlankStart, vblankstart & 0xFF);
+	write3X4(par, CRTVBlankEnd, 0 /* p->vblankend & 0xFF */);
 
 	/* horizontal timing values */
-	write3X4(CRTHTotal, htotal & 0xFF);
-	write3X4(CRTHDispEnd, hdispend & 0xFF);
-	write3X4(CRTHSyncStart, hsyncstart & 0xFF);
-	write3X4(CRTHSyncEnd, (hsyncend & 0x1F) | ((hblankend & 0x20) << 2));
-	write3X4(CRTHBlankStart, hblankstart & 0xFF);
-	write3X4(CRTHBlankEnd, 0 /* (p->hblankend & 0x1F) */ );
+	write3X4(par, CRTHTotal, htotal & 0xFF);
+	write3X4(par, CRTHDispEnd, hdispend & 0xFF);
+	write3X4(par, CRTHSyncStart, hsyncstart & 0xFF);
+	write3X4(par, CRTHSyncEnd,
+		 (hsyncend & 0x1F) | ((hblankend & 0x20) << 2));
+	write3X4(par, CRTHBlankStart, hblankstart & 0xFF);
+	write3X4(par, CRTHBlankEnd, 0 /* (p->hblankend & 0x1F) */);
 
 	/* higher bits of vertical timing values */
 	tmp = 0x10;
@@ -954,38 +987,40 @@ static int tridentfb_set_par(struct fb_info *info)
 	if (vtotal & 0x200) tmp |= 0x20;
 	if (vdispend & 0x200) tmp |= 0x40;
 	if (vsyncstart & 0x200) tmp |= 0x80;
-	write3X4(CRTOverflow, tmp);
+	write3X4(par, CRTOverflow, tmp);
 
-	tmp = read3X4(CRTHiOrd) | 0x08;	/* line compare bit 10 */
+	tmp = read3X4(par, CRTHiOrd) | 0x08;	/* line compare bit 10 */
 	if (vtotal & 0x400) tmp |= 0x80;
 	if (vblankstart & 0x400) tmp |= 0x40;
 	if (vsyncstart & 0x400) tmp |= 0x20;
 	if (vdispend & 0x400) tmp |= 0x10;
-	write3X4(CRTHiOrd, tmp);
+	write3X4(par, CRTHiOrd, tmp);
 
 	tmp = 0;
 	if (htotal & 0x800) tmp |= 0x800 >> 11;
 	if (hblankstart & 0x800) tmp |= 0x800 >> 7;
-	write3X4(HorizOverflow, tmp);
+	write3X4(par, HorizOverflow, tmp);
 
 	tmp = 0x40;
 	if (vblankstart & 0x200) tmp |= 0x20;
 //FIXME	if (info->var.vmode & FB_VMODE_DOUBLE) tmp |= 0x80;  /* double scan for 200 line modes */
-	write3X4(CRTMaxScanLine, tmp);
+	write3X4(par, CRTMaxScanLine, tmp);
 
-	write3X4(CRTLineCompare, 0xFF);
-	write3X4(CRTPRowScan, 0);
-	write3X4(CRTModeControl, 0xC3);
+	write3X4(par, CRTLineCompare, 0xFF);
+	write3X4(par, CRTPRowScan, 0);
+	write3X4(par, CRTModeControl, 0xC3);
 
-	write3X4(LinearAddReg, 0x20);	/* enable linear addressing */
+	write3X4(par, LinearAddReg, 0x20);	/* enable linear addressing */
 
 	tmp = (info->var.vmode & FB_VMODE_INTERLACED) ? 0x84 : 0x80;
-	write3X4(CRTCModuleTest, tmp);	/* enable access extended memory */
+	/* enable access extended memory */
+	write3X4(par, CRTCModuleTest, tmp);
 
-	write3X4(GraphEngReg, 0x80);	/* enable GE for text acceleration */
+	/* enable GE for text acceleration */
+	write3X4(par, GraphEngReg, 0x80);
 
 #ifdef CONFIG_FB_TRIDENT_ACCEL
-	acc->init_accel(info->var.xres, bpp);
+	acc->init_accel(par, info->var.xres, bpp);
 #endif
 
 	switch (bpp) {
@@ -1003,49 +1038,52 @@ static int tridentfb_set_par(struct fb_info *info)
 		break;
 	}
 
-	write3X4(PixelBusReg, tmp);
+	write3X4(par, PixelBusReg, tmp);
 
 	tmp = 0x10;
 	if (chipcyber)
 		tmp |= 0x20;
-	write3X4(DRAMControl, tmp);	/* both IO, linear enable */
+	write3X4(par, DRAMControl, tmp);	/* both IO, linear enable */
 
-	write3X4(InterfaceSel, read3X4(InterfaceSel) | 0x40);
-	write3X4(Performance, 0x92);
-	write3X4(PCIReg, 0x07);		/* MMIO & PCI read and write burst enable */
+	write3X4(par, InterfaceSel, read3X4(par, InterfaceSel) | 0x40);
+	write3X4(par, Performance, 0x92);
+	/* MMIO & PCI read and write burst enable */
+	write3X4(par, PCIReg, 0x07);
 
 	/* convert from picoseconds to kHz */
 	vclk = PICOS2KHZ(info->var.pixclock);
 	if (bpp == 32)
 		vclk *= 2;
-	set_vclk(vclk);
+	set_vclk(par, vclk);
 
-	write3C4(0, 3);
-	write3C4(1, 1);		/* set char clock 8 dots wide */
-	write3C4(2, 0x0F);	/* enable 4 maps because needed in chain4 mode */
-	write3C4(3, 0);
-	write3C4(4, 0x0E);	/* memory mode enable bitmaps ?? */
+	write3C4(par, 0, 3);
+	write3C4(par, 1, 1);		/* set char clock 8 dots wide */
+	/* enable 4 maps because needed in chain4 mode */
+	write3C4(par, 2, 0x0F);
+	write3C4(par, 3, 0);
+	write3C4(par, 4, 0x0E);	/* memory mode enable bitmaps ?? */
 
-	write3CE(MiscExtFunc, (bpp == 32) ? 0x1A : 0x12);	/* divide clock by 2 if 32bpp */
-							/* chain4 mode display and CPU path */
-	write3CE(0x5, 0x40);	/* no CGA compat, allow 256 col */
-	write3CE(0x6, 0x05);	/* graphics mode */
-	write3CE(0x7, 0x0F);	/* planes? */
+	/* divide clock by 2 if 32bpp chain4 mode display and CPU path */
+	write3CE(par, MiscExtFunc, (bpp == 32) ? 0x1A : 0x12);
+	write3CE(par, 0x5, 0x40);	/* no CGA compat, allow 256 col */
+	write3CE(par, 0x6, 0x05);	/* graphics mode */
+	write3CE(par, 0x7, 0x0F);	/* planes? */
 
 	if (chip_id == CYBERBLADEXPAi1) {
 		/* This fixes snow-effect in 32 bpp */
-		write3X4(CRTHSyncStart, 0x84);
+		write3X4(par, CRTHSyncStart, 0x84);
 	}
 
-	writeAttr(0x10, 0x41);	/* graphics mode and support 256 color modes */
-	writeAttr(0x12, 0x0F);	/* planes */
-	writeAttr(0x13, 0);	/* horizontal pel panning */
+	/* graphics mode and support 256 color modes */
+	writeAttr(par, 0x10, 0x41);
+	writeAttr(par, 0x12, 0x0F);	/* planes */
+	writeAttr(par, 0x13, 0);	/* horizontal pel panning */
 
 	/* colors */
 	for (tmp = 0; tmp < 0x10; tmp++)
-		writeAttr(tmp, tmp);
-	readb(par->io_virt + CRT + 0x0A);	/* flip-flop to index */
-	t_outb(0x20, 0x3C0);			/* enable attr */
+		writeAttr(par, tmp, tmp);
+	fb_readb(par->io_virt + CRT + 0x0A);	/* flip-flop to index */
+	t_outb(par, 0x20, 0x3C0);		/* enable attr */
 
 	switch (bpp) {
 	case 8:
@@ -1063,17 +1101,17 @@ static int tridentfb_set_par(struct fb_info *info)
 		break;
 	}
 
-	t_inb(0x3C8);
-	t_inb(0x3C6);
-	t_inb(0x3C6);
-	t_inb(0x3C6);
-	t_inb(0x3C6);
-	t_outb(tmp, 0x3C6);
-	t_inb(0x3C8);
+	t_inb(par, 0x3C8);
+	t_inb(par, 0x3C6);
+	t_inb(par, 0x3C6);
+	t_inb(par, 0x3C6);
+	t_inb(par, 0x3C6);
+	t_outb(par, tmp, 0x3C6);
+	t_inb(par, 0x3C8);
 
 	if (flatpanel)
-		set_number_of_lines(info->var.yres);
-	set_lwidth(info->var.xres * bpp / (4 * 16));
+		set_number_of_lines(par, info->var.yres);
+	set_lwidth(par, info->var.xres * bpp / (4 * 16));
 	info->fix.visual = (bpp == 8) ? FB_VISUAL_PSEUDOCOLOR : FB_VISUAL_TRUECOLOR;
 	info->fix.line_length = info->var.xres * (bpp >> 3);
 	info->cmap.len = (bpp == 8) ? 256 : 16;
@@ -1087,17 +1125,18 @@ static int tridentfb_setcolreg(unsigned regno, unsigned red, unsigned green,
 			       struct fb_info *info)
 {
 	int bpp = info->var.bits_per_pixel;
+	struct tridentfb_par *par = info->par;
 
 	if (regno >= info->cmap.len)
 		return 1;
 
 	if (bpp == 8) {
-		t_outb(0xFF, 0x3C6);
-		t_outb(regno, 0x3C8);
+		t_outb(par, 0xFF, 0x3C6);
+		t_outb(par, regno, 0x3C8);
 
-		t_outb(red >> 10, 0x3C9);
-		t_outb(green >> 10, 0x3C9);
-		t_outb(blue >> 10, 0x3C9);
+		t_outb(par, red >> 10, 0x3C9);
+		t_outb(par, green >> 10, 0x3C9);
+		t_outb(par, blue >> 10, 0x3C9);
 
 	} else if (regno < 16) {
 		if (bpp == 16) {	/* RGB 565 */
@@ -1123,13 +1162,14 @@ static int tridentfb_setcolreg(unsigned regno, unsigned red, unsigned green,
 static int tridentfb_blank(int blank_mode, struct fb_info *info)
 {
 	unsigned char PMCont, DPMSCont;
+	struct tridentfb_par *par = info->par;
 
 	debug("enter\n");
 	if (flatpanel)
 		return 0;
-	t_outb(0x04, 0x83C8); /* Read DPMS Control */
-	PMCont = t_inb(0x83C6) & 0xFC;
-	DPMSCont = read3CE(PowerStatus) & 0xFC;
+	t_outb(par, 0x04, 0x83C8); /* Read DPMS Control */
+	PMCont = t_inb(par, 0x83C6) & 0xFC;
+	DPMSCont = read3CE(par, PowerStatus) & 0xFC;
 	switch (blank_mode) {
 	case FB_BLANK_UNBLANK:
 		/* Screen: On, HSync: On, VSync: On */
@@ -1155,9 +1195,9 @@ static int tridentfb_blank(int blank_mode, struct fb_info *info)
 		break;
 	}
 
-	write3CE(PowerStatus, DPMSCont);
-	t_outb(4, 0x83C8);
-	t_outb(PMCont, 0x83C6);
+	write3CE(par, PowerStatus, DPMSCont);
+	t_outb(par, 4, 0x83C8);
+	t_outb(par, PMCont, 0x83C6);
 
 	debug("exit\n");
 
@@ -1265,11 +1305,11 @@ static int __devinit trident_pci_probe(struct pci_dev * dev,
 
 	/* setup framebuffer memory */
 	tridentfb_fix.smem_start = pci_resource_start(dev, 0);
-	tridentfb_fix.smem_len = get_memsize();
+	tridentfb_fix.smem_len = get_memsize(&default_par);
 
 	if (!request_mem_region(tridentfb_fix.smem_start, tridentfb_fix.smem_len, "tridentfb")) {
 		debug("request_mem_region failed!\n");
-		disable_mmio();
+		disable_mmio(fb_info.par);
 		err = -1;
 		goto out_unmap1;
 	}
@@ -1284,10 +1324,10 @@ static int __devinit trident_pci_probe(struct pci_dev * dev,
 	}
 
 	output("%s board found\n", pci_name(dev));
-	displaytype = get_displaytype();
+	displaytype = get_displaytype(&default_par);
 
 	if (flatpanel)
-		nativex = get_nativex();
+		nativex = get_nativex(&default_par);
 
 	fb_info.fix = tridentfb_fix;
 	fb_info.fbops = &tridentfb_ops;
@@ -1330,7 +1370,7 @@ out_unmap2:
 	if (fb_info.screen_base)
 		iounmap(fb_info.screen_base);
 	release_mem_region(tridentfb_fix.smem_start, tridentfb_fix.smem_len);
-	disable_mmio();
+	disable_mmio(fb_info.par);
 out_unmap1:
 	if (default_par.io_virt)
 		iounmap(default_par.io_virt);

From e09ed099d0169ac3a22b17cfeece0fa54a9e43eb Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:30:51 -0700
Subject: [PATCH 0444/1171] tridentfb: convert fb_info into allocated one

This patch converts fb_info structure from global variable to allocatable one.

The global default_par is moved into function variable.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 82 ++++++++++++++++++++++-----------------
 1 file changed, 46 insertions(+), 36 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index 3e8a1ef892ce..cb37e10734b6 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -33,10 +33,7 @@ struct tridentfb_par {
 static unsigned char eng_oper;	/* engine operation... */
 static struct fb_ops tridentfb_ops;
 
-static struct tridentfb_par default_par;
-
 /* FIXME:kmalloc these 3 instead */
-static struct fb_info fb_info;
 static u32 pseudo_pal[16];
 
 static struct fb_var_screeninfo default_var;
@@ -1217,16 +1214,23 @@ static struct fb_ops tridentfb_ops = {
 	.fb_imageblit = cfb_imageblit,
 };
 
-static int __devinit trident_pci_probe(struct pci_dev * dev,
-				       const struct pci_device_id * id)
+static int __devinit trident_pci_probe(struct pci_dev *dev,
+				       const struct pci_device_id *id)
 {
 	int err;
 	unsigned char revision;
+	struct fb_info *info;
+	struct tridentfb_par *default_par;
 
 	err = pci_enable_device(dev);
 	if (err)
 		return err;
 
+	info = framebuffer_alloc(sizeof(struct tridentfb_par), &dev->dev);
+	if (!info)
+		return -ENOMEM;
+	default_par = info->par;
+
 	chip_id = id->device;
 
 	if (chip_id == CYBERBLADEi1)
@@ -1282,8 +1286,6 @@ static int __devinit trident_pci_probe(struct pci_dev * dev,
 	/* acceleration is on by default for 3D chips */
 	defaultaccel = chip3D && !noaccel;
 
-	fb_info.par = &default_par;
-
 	/* setup MMIO region */
 	tridentfb_fix.mmio_start = pci_resource_start(dev, 1);
 	tridentfb_fix.mmio_len = chip3D ? 0x20000 : 0x10000;
@@ -1293,9 +1295,10 @@ static int __devinit trident_pci_probe(struct pci_dev * dev,
 		return -1;
 	}
 
-	default_par.io_virt = ioremap_nocache(tridentfb_fix.mmio_start, tridentfb_fix.mmio_len);
+	default_par->io_virt = ioremap_nocache(tridentfb_fix.mmio_start,
+					       tridentfb_fix.mmio_len);
 
-	if (!default_par.io_virt) {
+	if (!default_par->io_virt) {
 		debug("ioremap failed\n");
 		err = -1;
 		goto out_unmap1;
@@ -1305,46 +1308,46 @@ static int __devinit trident_pci_probe(struct pci_dev * dev,
 
 	/* setup framebuffer memory */
 	tridentfb_fix.smem_start = pci_resource_start(dev, 0);
-	tridentfb_fix.smem_len = get_memsize(&default_par);
+	tridentfb_fix.smem_len = get_memsize(default_par);
 
 	if (!request_mem_region(tridentfb_fix.smem_start, tridentfb_fix.smem_len, "tridentfb")) {
 		debug("request_mem_region failed!\n");
-		disable_mmio(fb_info.par);
+		disable_mmio(info->par);
 		err = -1;
 		goto out_unmap1;
 	}
 
-	fb_info.screen_base = ioremap_nocache(tridentfb_fix.smem_start,
-					      tridentfb_fix.smem_len);
+	info->screen_base = ioremap_nocache(tridentfb_fix.smem_start,
+					    tridentfb_fix.smem_len);
 
-	if (!fb_info.screen_base) {
+	if (!info->screen_base) {
 		debug("ioremap failed\n");
 		err = -1;
 		goto out_unmap2;
 	}
 
 	output("%s board found\n", pci_name(dev));
-	displaytype = get_displaytype(&default_par);
+	displaytype = get_displaytype(default_par);
 
 	if (flatpanel)
-		nativex = get_nativex(&default_par);
+		nativex = get_nativex(default_par);
 
-	fb_info.fix = tridentfb_fix;
-	fb_info.fbops = &tridentfb_ops;
+	info->fix = tridentfb_fix;
+	info->fbops = &tridentfb_ops;
 
 
-	fb_info.flags = FBINFO_DEFAULT | FBINFO_HWACCEL_YPAN;
+	info->flags = FBINFO_DEFAULT | FBINFO_HWACCEL_YPAN;
 #ifdef CONFIG_FB_TRIDENT_ACCEL
-	fb_info.flags |= FBINFO_HWACCEL_COPYAREA | FBINFO_HWACCEL_FILLRECT;
+	info->flags |= FBINFO_HWACCEL_COPYAREA | FBINFO_HWACCEL_FILLRECT;
 #endif
-	fb_info.pseudo_palette = pseudo_pal;
+	info->pseudo_palette = pseudo_pal;
 
-	if (!fb_find_mode(&default_var, &fb_info,
+	if (!fb_find_mode(&default_var, info,
 			  mode_option, NULL, 0, NULL, bpp)) {
 		err = -EINVAL;
 		goto out_unmap2;
 	}
-	err = fb_alloc_cmap(&fb_info.cmap, 256, 0);
+	err = fb_alloc_cmap(&info->cmap, 256, 0);
 	if (err < 0)
 		goto out_unmap2;
 
@@ -1353,39 +1356,46 @@ static int __devinit trident_pci_probe(struct pci_dev * dev,
 	else
 		default_var.accel_flags &= ~FB_ACCELF_TEXT;
 	default_var.activate |= FB_ACTIVATE_NOW;
-	fb_info.var = default_var;
-	fb_info.device = &dev->dev;
-	if (register_framebuffer(&fb_info) < 0) {
+	info->var = default_var;
+	info->device = &dev->dev;
+	if (register_framebuffer(info) < 0) {
 		printk(KERN_ERR "tridentfb: could not register Trident framebuffer\n");
-		fb_dealloc_cmap(&fb_info.cmap);
+		fb_dealloc_cmap(&info->cmap);
 		err = -EINVAL;
 		goto out_unmap2;
 	}
 	output("fb%d: %s frame buffer device %dx%d-%dbpp\n",
-	   fb_info.node, fb_info.fix.id, default_var.xres,
+	   info->node, info->fix.id, default_var.xres,
 	   default_var.yres, default_var.bits_per_pixel);
+
+	pci_set_drvdata(dev, info);
 	return 0;
 
 out_unmap2:
-	if (fb_info.screen_base)
-		iounmap(fb_info.screen_base);
+	if (info->screen_base)
+		iounmap(info->screen_base);
 	release_mem_region(tridentfb_fix.smem_start, tridentfb_fix.smem_len);
-	disable_mmio(fb_info.par);
+	disable_mmio(info->par);
 out_unmap1:
-	if (default_par.io_virt)
-		iounmap(default_par.io_virt);
+	if (default_par->io_virt)
+		iounmap(default_par->io_virt);
 	release_mem_region(tridentfb_fix.mmio_start, tridentfb_fix.mmio_len);
+	framebuffer_release(info);
 	return err;
 }
 
 static void __devexit trident_pci_remove(struct pci_dev *dev)
 {
-	struct tridentfb_par *par = (struct tridentfb_par*)fb_info.par;
-	unregister_framebuffer(&fb_info);
+	struct fb_info *info = pci_get_drvdata(dev);
+	struct tridentfb_par *par = info->par;
+
+	unregister_framebuffer(info);
 	iounmap(par->io_virt);
-	iounmap(fb_info.screen_base);
+	iounmap(info->screen_base);
 	release_mem_region(tridentfb_fix.smem_start, tridentfb_fix.smem_len);
 	release_mem_region(tridentfb_fix.mmio_start, tridentfb_fix.mmio_len);
+	pci_set_drvdata(dev, NULL);
+	framebuffer_release(info);
 }
 
 /* List of boards that we are trying to support */

From ea8ee55c12f77cbbb6e067f91e0cd794baa692ab Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:30:51 -0700
Subject: [PATCH 0445/1171] tridentfb: move global pseudo palette into
 structure

This patch moves pseudo palette int tridentfb_par structure and removes global
default_var.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 21 +++++++--------------
 1 file changed, 7 insertions(+), 14 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index cb37e10734b6..0f6e4054c996 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -28,16 +28,12 @@
 
 struct tridentfb_par {
 	void __iomem *io_virt;	/* iospace virtual memory address */
+	u32 pseudo_pal[16];
 };
 
 static unsigned char eng_oper;	/* engine operation... */
 static struct fb_ops tridentfb_ops;
 
-/* FIXME:kmalloc these 3 instead */
-static u32 pseudo_pal[16];
-
-static struct fb_var_screeninfo default_var;
-
 static struct fb_fix_screeninfo tridentfb_fix = {
 	.id = "Trident",
 	.type = FB_TYPE_PACKED_PIXELS,
@@ -1340,9 +1336,7 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 #ifdef CONFIG_FB_TRIDENT_ACCEL
 	info->flags |= FBINFO_HWACCEL_COPYAREA | FBINFO_HWACCEL_FILLRECT;
 #endif
-	info->pseudo_palette = pseudo_pal;
-
-	if (!fb_find_mode(&default_var, info,
+	if (!fb_find_mode(&info->var, info,
 			  mode_option, NULL, 0, NULL, bpp)) {
 		err = -EINVAL;
 		goto out_unmap2;
@@ -1352,11 +1346,10 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 		goto out_unmap2;
 
 	if (defaultaccel && acc)
-		default_var.accel_flags |= FB_ACCELF_TEXT;
+		info->var.accel_flags |= FB_ACCELF_TEXT;
 	else
-		default_var.accel_flags &= ~FB_ACCELF_TEXT;
-	default_var.activate |= FB_ACTIVATE_NOW;
-	info->var = default_var;
+		info->var.accel_flags &= ~FB_ACCELF_TEXT;
+	info->var.activate |= FB_ACTIVATE_NOW;
 	info->device = &dev->dev;
 	if (register_framebuffer(info) < 0) {
 		printk(KERN_ERR "tridentfb: could not register Trident framebuffer\n");
@@ -1365,8 +1358,8 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 		goto out_unmap2;
 	}
 	output("fb%d: %s frame buffer device %dx%d-%dbpp\n",
-	   info->node, info->fix.id, default_var.xres,
-	   default_var.yres, default_var.bits_per_pixel);
+	   info->node, info->fix.id, info->var.xres,
+	   info->var.yres, info->var.bits_per_pixel);
 
 	pci_set_drvdata(dev, info);
 	return 0;

From 122e8ad3cbf172043ea93f2db8e107fa9f9b0192 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:30:52 -0700
Subject: [PATCH 0446/1171] tridentfb: move global chip_id into structure

This patch moves the chip_id into tridentfb_par structure and removes global
chip_id related constants.

It also bumps version of the driver to 0.7.9

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 27 +++++++++++++--------------
 1 file changed, 13 insertions(+), 14 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index 0f6e4054c996..dfe52b424c97 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -24,11 +24,12 @@
 #include <linux/delay.h>
 #include <video/trident.h>
 
-#define VERSION		"0.7.8-NEWAPI"
+#define VERSION		"0.7.9-NEWAPI"
 
 struct tridentfb_par {
 	void __iomem *io_virt;	/* iospace virtual memory address */
 	u32 pseudo_pal[16];
+	int chip_id;
 };
 
 static unsigned char eng_oper;	/* engine operation... */
@@ -42,9 +43,6 @@ static struct fb_fix_screeninfo tridentfb_fix = {
 	.accel = FB_ACCEL_NONE,
 };
 
-static int chip_id;
-
-static int defaultaccel;
 static int displaytype;
 
 /* defaults which are normally overriden by user values */
@@ -79,9 +77,6 @@ module_param(nativex, int, 0);
 module_param(fp, int, 0);
 module_param(crt, int, 0);
 
-static int chip3D;
-static int chipcyber;
-
 static int is3Dchip(int id)
 {
 	return ((id == BLADE3D) || (id == CYBERBLADEE4) ||
@@ -658,7 +653,7 @@ static void set_lwidth(struct tridentfb_par *par, int width)
 /* For resolutions smaller than FP resolution stretch */
 static void screen_stretch(struct tridentfb_par *par)
 {
-	if (chip_id != CYBERBLADEXPAi1)
+	if (par->chip_id != CYBERBLADEXPAi1)
 		write3CE(par, BiosReg, 0);
 	else
 		write3CE(par, BiosReg, 8);
@@ -706,7 +701,7 @@ static void set_vclk(struct tridentfb_par *par, unsigned long freq)
 				if (fi > freq)
 					break;
 			}
-	if (chip3D) {
+	if (is3Dchip(par->chip_id)) {
 		write3C4(par, ClockHigh, hi);
 		write3C4(par, ClockLow, lo);
 	} else {
@@ -739,7 +734,7 @@ static unsigned int __devinit get_displaytype(struct tridentfb_par *par)
 {
 	if (fp)
 		return DISPLAY_FP;
-	if (crt || !chipcyber)
+	if (crt || !iscyber(par->chip_id))
 		return DISPLAY_CRT;
 	return (read3CE(par, FPConfig) & 0x10) ? DISPLAY_FP : DISPLAY_CRT;
 }
@@ -754,7 +749,7 @@ static unsigned int __devinit get_memsize(struct tridentfb_par *par)
 	if (memsize)
 		k = memsize * Kb;
 	else
-		switch (chip_id) {
+		switch (par->chip_id) {
 		case CYBER9525DVD:
 			k = 2560 * Kb;
 			break;
@@ -1034,7 +1029,7 @@ static int tridentfb_set_par(struct fb_info *info)
 	write3X4(par, PixelBusReg, tmp);
 
 	tmp = 0x10;
-	if (chipcyber)
+	if (iscyber(par->chip_id))
 		tmp |= 0x20;
 	write3X4(par, DRAMControl, tmp);	/* both IO, linear enable */
 
@@ -1062,7 +1057,7 @@ static int tridentfb_set_par(struct fb_info *info)
 	write3CE(par, 0x6, 0x05);	/* graphics mode */
 	write3CE(par, 0x7, 0x0F);	/* planes? */
 
-	if (chip_id == CYBERBLADEXPAi1) {
+	if (par->chip_id == CYBERBLADEXPAi1) {
 		/* This fixes snow-effect in 32 bpp */
 		write3X4(par, CRTHSyncStart, 0x84);
 	}
@@ -1217,6 +1212,9 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 	unsigned char revision;
 	struct fb_info *info;
 	struct tridentfb_par *default_par;
+	int defaultaccel;
+	int chip3D;
+	int chip_id;
 
 	err = pci_enable_device(dev);
 	if (err)
@@ -1269,7 +1267,6 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 	}
 
 	chip3D = is3Dchip(chip_id);
-	chipcyber = iscyber(chip_id);
 
 	if (is_xp(chip_id)) {
 		acc = &accel_xp;
@@ -1279,6 +1276,8 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 		acc = &accel_image;
 	}
 
+	default_par->chip_id = chip_id;
+
 	/* acceleration is on by default for 3D chips */
 	defaultaccel = chip3D && !noaccel;
 

From 6eed8e1ec8532a6cd10c8b27236bde023c52c56a Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:30:53 -0700
Subject: [PATCH 0447/1171] tridentfb: move global flat panel variable into
 structure

This patch moves flat panel indicator into tridentfb_par structure and removes
related global variables and macros.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 44 ++++++++++++++++++++-------------------
 include/video/trident.h   |  6 ------
 2 files changed, 23 insertions(+), 27 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index dfe52b424c97..604b10a590da 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -30,6 +30,7 @@ struct tridentfb_par {
 	void __iomem *io_virt;	/* iospace virtual memory address */
 	u32 pseudo_pal[16];
 	int chip_id;
+	int flatpanel;
 };
 
 static unsigned char eng_oper;	/* engine operation... */
@@ -43,24 +44,22 @@ static struct fb_fix_screeninfo tridentfb_fix = {
 	.accel = FB_ACCEL_NONE,
 };
 
-static int displaytype;
-
 /* defaults which are normally overriden by user values */
 
 /* video mode */
 static char *mode_option __devinitdata = "640x480";
-static int bpp = 8;
+static int bpp __devinitdata = 8;
 
-static int noaccel;
+static int noaccel __devinitdata;
 
 static int center;
 static int stretch;
 
-static int fp;
-static int crt;
+static int fp __devinitdata;
+static int crt __devinitdata;
 
-static int memsize;
-static int memdiff;
+static int memsize __devinitdata;
+static int memdiff __devinitdata;
 static int nativex;
 
 module_param(mode_option, charp, 0);
@@ -75,7 +74,9 @@ module_param(memsize, int, 0);
 module_param(memdiff, int, 0);
 module_param(nativex, int, 0);
 module_param(fp, int, 0);
+MODULE_PARM_DESC(fp, "Define if flatpanel is connected");
 module_param(crt, int, 0);
+MODULE_PARM_DESC(crt, "Define if CRT is connected");
 
 static int is3Dchip(int id)
 {
@@ -728,15 +729,15 @@ static void set_number_of_lines(struct tridentfb_par *par, int lines)
 
 /*
  * If we see that FP is active we assume we have one.
- * Otherwise we have a CRT display.User can override.
+ * Otherwise we have a CRT display. User can override.
  */
-static unsigned int __devinit get_displaytype(struct tridentfb_par *par)
+static int __devinit is_flatpanel(struct tridentfb_par *par)
 {
 	if (fp)
-		return DISPLAY_FP;
+		return 1;
 	if (crt || !iscyber(par->chip_id))
-		return DISPLAY_CRT;
-	return (read3CE(par, FPConfig) & 0x10) ? DISPLAY_FP : DISPLAY_CRT;
+		return 0;
+	return (read3CE(par, FPConfig) & 0x10) ? 1 : 0;
 }
 
 /* Try detecting the video memory size */
@@ -824,6 +825,7 @@ static unsigned int __devinit get_memsize(struct tridentfb_par *par)
 static int tridentfb_check_var(struct fb_var_screeninfo *var,
 			       struct fb_info *info)
 {
+	struct tridentfb_par *par = info->par;
 	int bpp = var->bits_per_pixel;
 	debug("enter\n");
 
@@ -831,7 +833,7 @@ static int tridentfb_check_var(struct fb_var_screeninfo *var,
 	if (bpp == 24)
 		bpp = var->bits_per_pixel = 32;
 	/* check whether resolution fits on panel and in memory */
-	if (flatpanel && nativex && var->xres > nativex)
+	if (par->flatpanel && nativex && var->xres > nativex)
 		return -EINVAL;
 	if (var->xres * var->yres_virtual * bpp / 8 > info->fix.smem_len)
 		return -EINVAL;
@@ -928,7 +930,7 @@ static int tridentfb_set_par(struct fb_info *info)
 	crtc_unlock(par);
 	write3CE(par, CyberControl, 8);
 
-	if (flatpanel && var->xres < nativex) {
+	if (par->flatpanel && var->xres < nativex) {
 		/*
 		 * on flat panels with native size larger
 		 * than requested resolution decide whether
@@ -1097,7 +1099,7 @@ static int tridentfb_set_par(struct fb_info *info)
 	t_outb(par, tmp, 0x3C6);
 	t_inb(par, 0x3C8);
 
-	if (flatpanel)
+	if (par->flatpanel)
 		set_number_of_lines(par, info->var.yres);
 	set_lwidth(par, info->var.xres * bpp / (4 * 16));
 	info->fix.visual = (bpp == 8) ? FB_VISUAL_PSEUDOCOLOR : FB_VISUAL_TRUECOLOR;
@@ -1153,7 +1155,7 @@ static int tridentfb_blank(int blank_mode, struct fb_info *info)
 	struct tridentfb_par *par = info->par;
 
 	debug("enter\n");
-	if (flatpanel)
+	if (par->flatpanel)
 		return 0;
 	t_outb(par, 0x04, 0x83C8); /* Read DPMS Control */
 	PMCont = t_inb(par, 0x83C6) & 0xFC;
@@ -1322,9 +1324,9 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 	}
 
 	output("%s board found\n", pci_name(dev));
-	displaytype = get_displaytype(default_par);
+	default_par->flatpanel = is_flatpanel(default_par);
 
-	if (flatpanel)
+	if (default_par->flatpanel)
 		nativex = get_nativex(default_par);
 
 	info->fix = tridentfb_fix;
@@ -1441,9 +1443,9 @@ static int __init tridentfb_setup(char *options)
 		if (!strncmp(opt, "noaccel", 7))
 			noaccel = 1;
 		else if (!strncmp(opt, "fp", 2))
-			displaytype = DISPLAY_FP;
+			fp = 1;
 		else if (!strncmp(opt, "crt", 3))
-			displaytype = DISPLAY_CRT;
+			fp = 0;
 		else if (!strncmp(opt, "bpp=", 4))
 			bpp = simple_strtoul(opt + 4, NULL, 0);
 		else if (!strncmp(opt, "center", 6))
diff --git a/include/video/trident.h b/include/video/trident.h
index 200be2551681..d3dd737a1ba8 100644
--- a/include/video/trident.h
+++ b/include/video/trident.h
@@ -63,12 +63,6 @@
 #define LCD_CENTER	1
 #define LCD_BIOS	2
 
-/* display types */
-#define DISPLAY_CRT	0
-#define DISPLAY_FP	1
-
-#define flatpanel (displaytype == DISPLAY_FP)
-
 /* General Registers */
 #define SPR	0x1F		/* Software Programming Register (videoram) */
 

From e0759a5fbba12e0f2c9149d85bea1ec7df0178fd Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:30:53 -0700
Subject: [PATCH 0448/1171] tridentfb: convert is_blade and is_xp macros into
 functions

This patch converts the is_blade() and is_xp() macros into local functions.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 19 +++++++++++++++++++
 include/video/trident.h   | 19 -------------------
 2 files changed, 19 insertions(+), 19 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index 604b10a590da..e87f77093ec1 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -78,6 +78,25 @@ MODULE_PARM_DESC(fp, "Define if flatpanel is connected");
 module_param(crt, int, 0);
 MODULE_PARM_DESC(crt, "Define if CRT is connected");
 
+static int is_blade(int id)
+{
+	return	(id == BLADE3D) ||
+		(id == CYBERBLADEE4) ||
+		(id == CYBERBLADEi7) ||
+		(id == CYBERBLADEi7D) ||
+		(id == CYBERBLADEi1) ||
+		(id == CYBERBLADEi1D) ||
+		(id == CYBERBLADEAi1) ||
+		(id == CYBERBLADEAi1D);
+}
+
+static int is_xp(int id)
+{
+	return	(id == CYBERBLADEXPAi1) ||
+		(id == CYBERBLADEXPm8) ||
+		(id == CYBERBLADEXPm16);
+}
+
 static int is3Dchip(int id)
 {
 	return ((id == BLADE3D) || (id == CYBERBLADEE4) ||
diff --git a/include/video/trident.h b/include/video/trident.h
index d3dd737a1ba8..9c3670b28906 100644
--- a/include/video/trident.h
+++ b/include/video/trident.h
@@ -39,25 +39,6 @@
 #define CYBERBLADEXPm8  0x9910
 #define CYBERBLADEXPm16 0x9930
 
-/* acceleration families */
-#define IMAGE	0
-#define BLADE	1
-#define XP	2
-
-#define is_image(id)	
-#define is_xp(id)	((id == CYBERBLADEXPAi1) ||\
-			 (id == CYBERBLADEXPm8) ||\
-			 (id == CYBERBLADEXPm16)) 
-
-#define is_blade(id)	((id == BLADE3D) ||\
-			 (id == CYBERBLADEE4) ||\
-			 (id == CYBERBLADEi7) ||\
-			 (id == CYBERBLADEi7D) ||\
-			 (id == CYBERBLADEi1) ||\
-			 (id == CYBERBLADEi1D) ||\
-			 (id ==	CYBERBLADEAi1) ||\
-			 (id ==	CYBERBLADEAi1D))
-
 /* these defines are for 'lcd' variable */
 #define LCD_STRETCH	0
 #define LCD_CENTER	1

From d9cad04bcde00411976402eda726199ac13b29ca Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:30:54 -0700
Subject: [PATCH 0449/1171] tridentfb: move global acceleration hooks into
 structure

This patch moves acceleration hooks into the tridentfb_par structure and
removes global hooks.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 63 +++++++++++++++------------------------
 1 file changed, 24 insertions(+), 39 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index e87f77093ec1..d6796448c5ac 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -31,6 +31,12 @@ struct tridentfb_par {
 	u32 pseudo_pal[16];
 	int chip_id;
 	int flatpanel;
+	void (*init_accel) (struct tridentfb_par *, int, int);
+	void (*wait_engine) (struct tridentfb_par *);
+	void (*fill_rect)
+		(struct tridentfb_par *par, u32, u32, u32, u32, u32, u32);
+	void (*copy_rect)
+		(struct tridentfb_par *par, u32, u32, u32, u32, u32, u32);
 };
 
 static unsigned char eng_oper;	/* engine operation... */
@@ -155,15 +161,6 @@ static inline u8 t_inb(struct tridentfb_par *p, u16 reg)
 	return fb_readb(p->io_virt + reg);
 }
 
-static struct accel_switch {
-	void (*init_accel) (struct tridentfb_par *, int, int);
-	void (*wait_engine) (struct tridentfb_par *);
-	void (*fill_rect)
-		(struct tridentfb_par *par, u32, u32, u32, u32, u32, u32);
-	void (*copy_rect)
-		(struct tridentfb_par *par, u32, u32, u32, u32, u32, u32);
-} *acc;
-
 static inline void writemmr(struct tridentfb_par *par, u16 r, u32 v)
 {
 	fb_writel(v, par->io_virt + r);
@@ -259,13 +256,6 @@ static void blade_copy_rect(struct tridentfb_par *par,
 	writemmr(par, DR2, direction ? d1 : d2);
 }
 
-static struct accel_switch accel_blade = {
-	blade_init_accel,
-	blade_wait_engine,
-	blade_fill_rect,
-	blade_copy_rect,
-};
-
 /*
  * BladeXP specific acceleration functions
  */
@@ -405,13 +395,6 @@ static void xp_copy_rect(struct tridentfb_par *par,
 	t_outb(par, 0x01, 0x2124);
 }
 
-static struct accel_switch accel_xp = {
-	xp_init_accel,
-	xp_wait_engine,
-	xp_fill_rect,
-	xp_copy_rect,
-};
-
 /*
  * Image specific acceleration functions
  */
@@ -491,13 +474,6 @@ static void image_copy_rect(struct tridentfb_par *par,
 		 0x80000000 | 1 << 22 | 1 << 10 | 1 << 7 | direction);
 }
 
-static struct accel_switch accel_image = {
-	image_init_accel,
-	image_wait_engine,
-	image_fill_rect,
-	image_copy_rect,
-};
-
 /*
  * Accel functions called by the upper layers
  */
@@ -524,18 +500,18 @@ static void tridentfb_fillrect(struct fb_info *info,
 		break;
 	}
 
-	acc->fill_rect(par, fr->dx, fr->dy, fr->width,
+	par->fill_rect(par, fr->dx, fr->dy, fr->width,
 		       fr->height, col, fr->rop);
-	acc->wait_engine(par);
+	par->wait_engine(par);
 }
 static void tridentfb_copyarea(struct fb_info *info,
 			       const struct fb_copyarea *ca)
 {
 	struct tridentfb_par *par = info->par;
 
-	acc->copy_rect(par, ca->sx, ca->sy, ca->dx, ca->dy,
+	par->copy_rect(par, ca->sx, ca->sy, ca->dx, ca->dy,
 		       ca->width, ca->height);
-	acc->wait_engine(par);
+	par->wait_engine(par);
 }
 #else /* !CONFIG_FB_TRIDENT_ACCEL */
 #define tridentfb_fillrect cfb_fillrect
@@ -1029,7 +1005,7 @@ static int tridentfb_set_par(struct fb_info *info)
 	write3X4(par, GraphEngReg, 0x80);
 
 #ifdef CONFIG_FB_TRIDENT_ACCEL
-	acc->init_accel(par, info->var.xres, bpp);
+	par->init_accel(par, info->var.xres, bpp);
 #endif
 
 	switch (bpp) {
@@ -1290,11 +1266,20 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 	chip3D = is3Dchip(chip_id);
 
 	if (is_xp(chip_id)) {
-		acc = &accel_xp;
+		default_par->init_accel = xp_init_accel;
+		default_par->wait_engine = xp_wait_engine;
+		default_par->fill_rect = xp_fill_rect;
+		default_par->copy_rect = xp_copy_rect;
 	} else if (is_blade(chip_id)) {
-		acc = &accel_blade;
+		default_par->init_accel = blade_init_accel;
+		default_par->wait_engine = blade_wait_engine;
+		default_par->fill_rect = blade_fill_rect;
+		default_par->copy_rect = blade_copy_rect;
 	} else {
-		acc = &accel_image;
+		default_par->init_accel = image_init_accel;
+		default_par->wait_engine = image_wait_engine;
+		default_par->fill_rect = image_fill_rect;
+		default_par->copy_rect = image_copy_rect;
 	}
 
 	default_par->chip_id = chip_id;
@@ -1365,7 +1350,7 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 	if (err < 0)
 		goto out_unmap2;
 
-	if (defaultaccel && acc)
+	if (defaultaccel && default_par->init_accel)
 		info->var.accel_flags |= FB_ACCELF_TEXT;
 	else
 		info->var.accel_flags &= ~FB_ACCELF_TEXT;

From 10172ed6dc4d40ff42bf5ce2dd2f65f401a93696 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:30:54 -0700
Subject: [PATCH 0450/1171] tridentfb: make use of functions and constants from
 the vga.h

Make use of functions and constants from the vga.h header to compact the code
and make it more readable.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 144 ++++++++++++++++----------------------
 include/video/trident.h   |  26 -------
 2 files changed, 61 insertions(+), 109 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index d6796448c5ac..381e5853df6f 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -22,6 +22,7 @@
 #include <linux/pci.h>
 
 #include <linux/delay.h>
+#include <video/vga.h>
 #include <video/trident.h>
 
 #define VERSION		"0.7.9-NEWAPI"
@@ -149,8 +150,6 @@ static int iscyber(int id)
 	}
 }
 
-#define CRT 0x3D0		/* CRTC registers offset for color display */
-
 static inline void t_outb(struct tridentfb_par *p, u8 val, u16 reg)
 {
 	fb_writeb(val, p->io_virt + reg);
@@ -525,60 +524,41 @@ static void tridentfb_copyarea(struct fb_info *info,
 
 static inline unsigned char read3X4(struct tridentfb_par *par, int reg)
 {
-	writeb(reg, par->io_virt + CRT + 4);
-	return readb(par->io_virt + CRT + 5);
+	return vga_mm_rcrt(par->io_virt, reg);
 }
 
 static inline void write3X4(struct tridentfb_par *par, int reg,
 			    unsigned char val)
 {
-	writeb(reg, par->io_virt + CRT + 4);
-	writeb(val, par->io_virt + CRT + 5);
+	vga_mm_wcrt(par->io_virt, reg, val);
 }
 
-static inline unsigned char read3C4(struct tridentfb_par *par, int reg)
+static inline unsigned char read3CE(struct tridentfb_par *par,
+				    unsigned char reg)
 {
-	t_outb(par, reg, 0x3C4);
-	return t_inb(par, 0x3C5);
-}
-
-static inline void write3C4(struct tridentfb_par *par, int reg,
-			    unsigned char val)
-{
-	t_outb(par, reg, 0x3C4);
-	t_outb(par, val, 0x3C5);
-}
-
-static inline unsigned char read3CE(struct tridentfb_par *par, int reg)
-{
-	t_outb(par, reg, 0x3CE);
-	return t_inb(par, 0x3CF);
+	return vga_mm_rgfx(par->io_virt, reg);
 }
 
 static inline void writeAttr(struct tridentfb_par *par, int reg,
 			     unsigned char val)
 {
-	fb_readb(par->io_virt + CRT + 0x0A);	/* flip-flop to index */
-	t_outb(par, reg, 0x3C0);
-	t_outb(par, val, 0x3C0);
+	fb_readb(par->io_virt + VGA_IS1_RC);	/* flip-flop to index */
+	vga_mm_wattr(par->io_virt, reg, val);
 }
 
 static inline void write3CE(struct tridentfb_par *par, int reg,
 			    unsigned char val)
 {
-	t_outb(par, reg, 0x3CE);
-	t_outb(par, val, 0x3CF);
+	vga_mm_wgfx(par->io_virt, reg, val);
 }
 
 static void enable_mmio(void)
 {
 	/* Goto New Mode */
-	outb(0x0B, 0x3C4);
-	inb(0x3C5);
+	vga_io_rseq(0x0B);
 
 	/* Unprotect registers */
-	outb(NewMode1, 0x3C4);
-	outb(0x80, 0x3C5);
+	vga_io_wseq(NewMode1, 0x80);
 
 	/* Enable MMIO */
 	outb(PCIReg, 0x3D4);
@@ -588,12 +568,10 @@ static void enable_mmio(void)
 static void disable_mmio(struct tridentfb_par *par)
 {
 	/* Goto New Mode */
-	t_outb(par, 0x0B, 0x3C4);
-	t_inb(par, 0x3C5);
+	vga_mm_rseq(par->io_virt, 0x0B);
 
 	/* Unprotect registers */
-	t_outb(par, NewMode1, 0x3C4);
-	t_outb(par, 0x80, 0x3C5);
+	vga_mm_wseq(par->io_virt, NewMode1, 0x80);
 
 	/* Disable MMIO */
 	t_outb(par, PCIReg, 0x3D4);
@@ -602,7 +580,8 @@ static void disable_mmio(struct tridentfb_par *par)
 
 static void crtc_unlock(struct tridentfb_par *par)
 {
-	write3X4(par, CRTVSyncEnd, read3X4(par, CRTVSyncEnd) & 0x7F);
+	write3X4(par, VGA_CRTC_V_SYNC_END,
+		 read3X4(par, VGA_CRTC_V_SYNC_END) & 0x7F);
 }
 
 /*  Return flat panel's maximum x resolution */
@@ -641,7 +620,7 @@ static int __devinit get_nativex(struct tridentfb_par *par)
 /* Set pitch */
 static void set_lwidth(struct tridentfb_par *par, int width)
 {
-	write3X4(par, Offset, width & 0xFF);
+	write3X4(par, VGA_CRTC_OFFSET, width & 0xFF);
 	write3X4(par, AddColReg,
 		 (read3X4(par, AddColReg) & 0xCF) | ((width & 0x300) >> 4));
 }
@@ -668,8 +647,8 @@ static void screen_center(struct tridentfb_par *par)
 static void set_screen_start(struct tridentfb_par *par, int base)
 {
 	u8 tmp;
-	write3X4(par, StartAddrLow, base & 0xFF);
-	write3X4(par, StartAddrHigh, (base & 0xFF00) >> 8);
+	write3X4(par, VGA_CRTC_START_LO, base & 0xFF);
+	write3X4(par, VGA_CRTC_START_HI, (base & 0xFF00) >> 8);
 	tmp = read3X4(par, CRTCModuleTest) & 0xDF;
 	write3X4(par, CRTCModuleTest, tmp | ((base & 0x10000) >> 11));
 	tmp = read3X4(par, CRTHiOrd) & 0xF8;
@@ -698,8 +677,8 @@ static void set_vclk(struct tridentfb_par *par, unsigned long freq)
 					break;
 			}
 	if (is3Dchip(par->chip_id)) {
-		write3C4(par, ClockHigh, hi);
-		write3C4(par, ClockLow, lo);
+		vga_mm_wseq(par->io_virt, ClockHigh, hi);
+		vga_mm_wseq(par->io_virt, ClockLow, lo);
 	} else {
 		outb(lo, 0x43C8);
 		outb(hi, 0x43C9);
@@ -782,7 +761,7 @@ static unsigned int __devinit get_memsize(struct tridentfb_par *par)
 				break;
 			case 0x0E:		/* XP */
 
-				tmp2 = read3C4(par, 0xC1);
+				tmp2 = vga_mm_rseq(par->io_virt, 0xC1);
 				switch (tmp2) {
 				case 0x00:
 					k = 20 * Mb;
@@ -931,7 +910,7 @@ static int tridentfb_set_par(struct fb_info *info)
 		 * than requested resolution decide whether
 		 * we stretch or center
 		 */
-		t_outb(par, 0xEB, 0x3C2);
+		t_outb(par, 0xEB, VGA_MIS_W);
 
 		shadowmode_on(par);
 
@@ -941,26 +920,26 @@ static int tridentfb_set_par(struct fb_info *info)
 			screen_stretch(par);
 
 	} else {
-		t_outb(par, 0x2B, 0x3C2);
+		t_outb(par, 0x2B, VGA_MIS_W);
 		write3CE(par, CyberControl, 8);
 	}
 
 	/* vertical timing values */
-	write3X4(par, CRTVTotal, vtotal & 0xFF);
-	write3X4(par, CRTVDispEnd, vdispend & 0xFF);
-	write3X4(par, CRTVSyncStart, vsyncstart & 0xFF);
-	write3X4(par, CRTVSyncEnd, (vsyncend & 0x0F));
-	write3X4(par, CRTVBlankStart, vblankstart & 0xFF);
-	write3X4(par, CRTVBlankEnd, 0 /* p->vblankend & 0xFF */);
+	write3X4(par, VGA_CRTC_V_TOTAL, vtotal & 0xFF);
+	write3X4(par, VGA_CRTC_V_DISP_END, vdispend & 0xFF);
+	write3X4(par, VGA_CRTC_V_SYNC_START, vsyncstart & 0xFF);
+	write3X4(par, VGA_CRTC_V_SYNC_END, (vsyncend & 0x0F));
+	write3X4(par, VGA_CRTC_V_BLANK_START, vblankstart & 0xFF);
+	write3X4(par, VGA_CRTC_V_BLANK_END, 0 /* p->vblankend & 0xFF */);
 
 	/* horizontal timing values */
-	write3X4(par, CRTHTotal, htotal & 0xFF);
-	write3X4(par, CRTHDispEnd, hdispend & 0xFF);
-	write3X4(par, CRTHSyncStart, hsyncstart & 0xFF);
-	write3X4(par, CRTHSyncEnd,
+	write3X4(par, VGA_CRTC_H_TOTAL, htotal & 0xFF);
+	write3X4(par, VGA_CRTC_H_DISP, hdispend & 0xFF);
+	write3X4(par, VGA_CRTC_H_SYNC_START, hsyncstart & 0xFF);
+	write3X4(par, VGA_CRTC_H_SYNC_END,
 		 (hsyncend & 0x1F) | ((hblankend & 0x20) << 2));
-	write3X4(par, CRTHBlankStart, hblankstart & 0xFF);
-	write3X4(par, CRTHBlankEnd, 0 /* (p->hblankend & 0x1F) */);
+	write3X4(par, VGA_CRTC_H_BLANK_START, hblankstart & 0xFF);
+	write3X4(par, VGA_CRTC_H_BLANK_END, 0 /* (p->hblankend & 0x1F) */);
 
 	/* higher bits of vertical timing values */
 	tmp = 0x10;
@@ -972,7 +951,7 @@ static int tridentfb_set_par(struct fb_info *info)
 	if (vtotal & 0x200) tmp |= 0x20;
 	if (vdispend & 0x200) tmp |= 0x40;
 	if (vsyncstart & 0x200) tmp |= 0x80;
-	write3X4(par, CRTOverflow, tmp);
+	write3X4(par, VGA_CRTC_OVERFLOW, tmp);
 
 	tmp = read3X4(par, CRTHiOrd) | 0x08;	/* line compare bit 10 */
 	if (vtotal & 0x400) tmp |= 0x80;
@@ -989,11 +968,11 @@ static int tridentfb_set_par(struct fb_info *info)
 	tmp = 0x40;
 	if (vblankstart & 0x200) tmp |= 0x20;
 //FIXME	if (info->var.vmode & FB_VMODE_DOUBLE) tmp |= 0x80;  /* double scan for 200 line modes */
-	write3X4(par, CRTMaxScanLine, tmp);
+	write3X4(par, VGA_CRTC_MAX_SCAN, tmp);
 
-	write3X4(par, CRTLineCompare, 0xFF);
-	write3X4(par, CRTPRowScan, 0);
-	write3X4(par, CRTModeControl, 0xC3);
+	write3X4(par, VGA_CRTC_LINE_COMPARE, 0xFF);
+	write3X4(par, VGA_CRTC_PRESET_ROW, 0);
+	write3X4(par, VGA_CRTC_MODE, 0xC3);
 
 	write3X4(par, LinearAddReg, 0x20);	/* enable linear addressing */
 
@@ -1041,12 +1020,12 @@ static int tridentfb_set_par(struct fb_info *info)
 		vclk *= 2;
 	set_vclk(par, vclk);
 
-	write3C4(par, 0, 3);
-	write3C4(par, 1, 1);		/* set char clock 8 dots wide */
+	vga_mm_wseq(par->io_virt, 0, 3);
+	vga_mm_wseq(par->io_virt, 1, 1); /* set char clock 8 dots wide */
 	/* enable 4 maps because needed in chain4 mode */
-	write3C4(par, 2, 0x0F);
-	write3C4(par, 3, 0);
-	write3C4(par, 4, 0x0E);	/* memory mode enable bitmaps ?? */
+	vga_mm_wseq(par->io_virt, 2, 0x0F);
+	vga_mm_wseq(par->io_virt, 3, 0);
+	vga_mm_wseq(par->io_virt, 4, 0x0E); /* memory mode enable bitmaps ?? */
 
 	/* divide clock by 2 if 32bpp chain4 mode display and CPU path */
 	write3CE(par, MiscExtFunc, (bpp == 32) ? 0x1A : 0x12);
@@ -1056,7 +1035,7 @@ static int tridentfb_set_par(struct fb_info *info)
 
 	if (par->chip_id == CYBERBLADEXPAi1) {
 		/* This fixes snow-effect in 32 bpp */
-		write3X4(par, CRTHSyncStart, 0x84);
+		write3X4(par, VGA_CRTC_H_SYNC_START, 0x84);
 	}
 
 	/* graphics mode and support 256 color modes */
@@ -1067,8 +1046,8 @@ static int tridentfb_set_par(struct fb_info *info)
 	/* colors */
 	for (tmp = 0; tmp < 0x10; tmp++)
 		writeAttr(par, tmp, tmp);
-	fb_readb(par->io_virt + CRT + 0x0A);	/* flip-flop to index */
-	t_outb(par, 0x20, 0x3C0);		/* enable attr */
+	fb_readb(par->io_virt + VGA_IS1_RC);	/* flip-flop to index */
+	t_outb(par, 0x20, VGA_ATT_W);		/* enable attr */
 
 	switch (bpp) {
 	case 8:
@@ -1086,13 +1065,13 @@ static int tridentfb_set_par(struct fb_info *info)
 		break;
 	}
 
-	t_inb(par, 0x3C8);
-	t_inb(par, 0x3C6);
-	t_inb(par, 0x3C6);
-	t_inb(par, 0x3C6);
-	t_inb(par, 0x3C6);
-	t_outb(par, tmp, 0x3C6);
-	t_inb(par, 0x3C8);
+	t_inb(par, VGA_PEL_IW);
+	t_inb(par, VGA_PEL_MSK);
+	t_inb(par, VGA_PEL_MSK);
+	t_inb(par, VGA_PEL_MSK);
+	t_inb(par, VGA_PEL_MSK);
+	t_outb(par, tmp, VGA_PEL_MSK);
+	t_inb(par, VGA_PEL_IW);
 
 	if (par->flatpanel)
 		set_number_of_lines(par, info->var.yres);
@@ -1116,12 +1095,12 @@ static int tridentfb_setcolreg(unsigned regno, unsigned red, unsigned green,
 		return 1;
 
 	if (bpp == 8) {
-		t_outb(par, 0xFF, 0x3C6);
-		t_outb(par, regno, 0x3C8);
+		t_outb(par, 0xFF, VGA_PEL_MSK);
+		t_outb(par, regno, VGA_PEL_IW);
 
-		t_outb(par, red >> 10, 0x3C9);
-		t_outb(par, green >> 10, 0x3C9);
-		t_outb(par, blue >> 10, 0x3C9);
+		t_outb(par, red >> 10, VGA_PEL_D);
+		t_outb(par, green >> 10, VGA_PEL_D);
+		t_outb(par, blue >> 10, VGA_PEL_D);
 
 	} else if (regno < 16) {
 		if (bpp == 16) {	/* RGB 565 */
@@ -1232,8 +1211,7 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 	/* If PCI id is 0x9660 then further detect chip type */
 
 	if (chip_id == TGUI9660) {
-		outb(RevisionID, 0x3C4);
-		revision = inb(0x3C5);
+		revision = vga_io_rseq(RevisionID);
 
 		switch (revision) {
 		case 0x22:
diff --git a/include/video/trident.h b/include/video/trident.h
index 9c3670b28906..51ec5a95d1ad 100644
--- a/include/video/trident.h
+++ b/include/video/trident.h
@@ -63,33 +63,7 @@
 #define SKey 0x37
 #define SPKey 0x57
 
-/* 0x3x4 */
-#define CRTHTotal	0x00
-#define CRTHDispEnd	0x01
-#define CRTHBlankStart	0x02
-#define CRTHBlankEnd	0x03
-#define CRTHSyncStart	0x04
-#define CRTHSyncEnd	0x05
-
-#define CRTVTotal	0x06
-#define CRTVDispEnd	0x12
-#define CRTVBlankStart	0x15
-#define CRTVBlankEnd	0x16
-#define CRTVSyncStart	0x10
-#define CRTVSyncEnd	0x11
-
-#define CRTOverflow	0x07
-#define CRTPRowScan	0x08
-#define CRTMaxScanLine	0x09
-#define CRTModeControl	0x17
-#define CRTLineCompare	0x18
-
 /* 3x4 */
-#define StartAddrHigh 0x0C
-#define StartAddrLow 0x0D
-#define Offset 0x13
-#define Underline 0x14
-#define CRTCMode 0x17
 #define CRTCModuleTest 0x1E
 #define FIFOControl 0x20
 #define LinearAddReg 0x21

From 7f762d23e607af786bba8ff4a18059f43950c0e8 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:30:55 -0700
Subject: [PATCH 0451/1171] tridentfb: fix timing calculations

Fix broken timings calculations. This patch helps with following
problems:
 - no left part of screen visible (up to half of the screen)
 - monitor's frequencies are not the ones intended for selected modes
 - if mode with resoultion y > 1024 is selected at least once then
   all modes with y < 1024 are "out of sync" (no display)

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 35 ++++++++++++++++++-----------------
 1 file changed, 18 insertions(+), 17 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index 381e5853df6f..af02af11353c 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -886,20 +886,19 @@ static int tridentfb_set_par(struct fb_info *info)
 
 	debug("enter\n");
 	hdispend = var->xres / 8 - 1;
-	hsyncstart = (var->xres + var->right_margin) / 8;
-	hsyncend = var->hsync_len / 8;
-	htotal =
-		(var->xres + var->left_margin + var->right_margin +
-		 var->hsync_len) / 8 - 10;
-	hblankstart = hdispend + 1;
-	hblankend = htotal + 5;
+	hsyncstart = (var->xres + var->right_margin) / 8 - 1;
+	hsyncend = (var->xres + var->right_margin + var->hsync_len) / 8 - 1;
+	htotal = (var->xres + var->left_margin + var->right_margin +
+		  var->hsync_len) / 8 - 5;
+	hblankstart = hdispend + 2;
+	hblankend = htotal + 3;
 
 	vdispend = var->yres - 1;
 	vsyncstart = var->yres + var->lower_margin;
-	vsyncend = var->vsync_len;
-	vtotal = var->upper_margin + vsyncstart + vsyncend - 2;
-	vblankstart = var->yres;
-	vblankend = vtotal + 2;
+	vsyncend = vsyncstart + var->vsync_len;
+	vtotal = var->upper_margin + vsyncend - 2;
+	vblankstart = vdispend + 2;
+	vblankend = vtotal;
 
 	crtc_unlock(par);
 	write3CE(par, CyberControl, 8);
@@ -930,7 +929,7 @@ static int tridentfb_set_par(struct fb_info *info)
 	write3X4(par, VGA_CRTC_V_SYNC_START, vsyncstart & 0xFF);
 	write3X4(par, VGA_CRTC_V_SYNC_END, (vsyncend & 0x0F));
 	write3X4(par, VGA_CRTC_V_BLANK_START, vblankstart & 0xFF);
-	write3X4(par, VGA_CRTC_V_BLANK_END, 0 /* p->vblankend & 0xFF */);
+	write3X4(par, VGA_CRTC_V_BLANK_END, vblankend & 0xFF);
 
 	/* horizontal timing values */
 	write3X4(par, VGA_CRTC_H_TOTAL, htotal & 0xFF);
@@ -939,7 +938,7 @@ static int tridentfb_set_par(struct fb_info *info)
 	write3X4(par, VGA_CRTC_H_SYNC_END,
 		 (hsyncend & 0x1F) | ((hblankend & 0x20) << 2));
 	write3X4(par, VGA_CRTC_H_BLANK_START, hblankstart & 0xFF);
-	write3X4(par, VGA_CRTC_H_BLANK_END, 0 /* (p->hblankend & 0x1F) */);
+	write3X4(par, VGA_CRTC_H_BLANK_END, hblankend & 0x1F);
 
 	/* higher bits of vertical timing values */
 	tmp = 0x10;
@@ -953,16 +952,18 @@ static int tridentfb_set_par(struct fb_info *info)
 	if (vsyncstart & 0x200) tmp |= 0x80;
 	write3X4(par, VGA_CRTC_OVERFLOW, tmp);
 
-	tmp = read3X4(par, CRTHiOrd) | 0x08;	/* line compare bit 10 */
+	tmp = read3X4(par, CRTHiOrd) & 0x07;
+	tmp |= 0x08;	/* line compare bit 10 */
 	if (vtotal & 0x400) tmp |= 0x80;
 	if (vblankstart & 0x400) tmp |= 0x40;
 	if (vsyncstart & 0x400) tmp |= 0x20;
 	if (vdispend & 0x400) tmp |= 0x10;
 	write3X4(par, CRTHiOrd, tmp);
 
-	tmp = 0;
-	if (htotal & 0x800) tmp |= 0x800 >> 11;
-	if (hblankstart & 0x800) tmp |= 0x800 >> 7;
+	tmp = (htotal >> 8) & 0x01;
+	tmp |= (hdispend >> 7) & 0x02;
+	tmp |= (hsyncstart >> 5) & 0x08;
+	tmp |= (hblankstart >> 4) & 0x10;
 	write3X4(par, HorizOverflow, tmp);
 
 	tmp = 0x40;

From c1724fecabfed504a4cfb87319ad3b9d3a8baa92 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:30:56 -0700
Subject: [PATCH 0452/1171] tridentfb: use mmio access for clock setting

Use the mmio outb function instead of direct one.  The mmio registers are
already mapped (in the probe function).

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index af02af11353c..f3153a82f926 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -680,8 +680,8 @@ static void set_vclk(struct tridentfb_par *par, unsigned long freq)
 		vga_mm_wseq(par->io_virt, ClockHigh, hi);
 		vga_mm_wseq(par->io_virt, ClockLow, lo);
 	} else {
-		outb(lo, 0x43C8);
-		outb(hi, 0x43C9);
+		t_outb(par, lo, 0x43C8);
+		t_outb(par, hi, 0x43C9);
 	}
 	debug("VCLK = %X %X\n", hi, lo);
 }

From 6bdf1035602abf0564d24a7447eea1c149c4bcb1 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:30:56 -0700
Subject: [PATCH 0453/1171] tridentfb: fix clock settings for older Trident
 96XX chips

The Xorg code shows that Trident models 9660, 9680 and 9682 require a
different clock setting method.  Add the second clock setting method for older
models.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 31 +++++++++++++++++++++++--------
 1 file changed, 23 insertions(+), 8 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index f3153a82f926..8ee4261abf3b 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -85,6 +85,11 @@ MODULE_PARM_DESC(fp, "Define if flatpanel is connected");
 module_param(crt, int, 0);
 MODULE_PARM_DESC(crt, "Define if CRT is connected");
 
+static int is_oldclock(int id)
+{
+	return (id == TGUI9660);
+}
+
 static int is_blade(int id)
 {
 	return	(id == BLADE3D) ||
@@ -659,23 +664,33 @@ static void set_screen_start(struct tridentfb_par *par, int base)
 static void set_vclk(struct tridentfb_par *par, unsigned long freq)
 {
 	int m, n, k;
-	unsigned long f, fi, d, di;
-	unsigned char lo = 0, hi = 0;
+	unsigned long fi, d, di;
+	unsigned char best_m = 0, best_n = 0, best_k = 0;
+	unsigned char hi, lo;
 
 	d = 20000;
-	for (k = 2; k >= 0; k--)
-		for (m = 0; m < 63; m++)
-			for (n = 0; n < 128; n++) {
+	for (k = 1; k >= 0; k--)
+		for (m = 0; m < 32; m++)
+			for (n = 0; n < 122; n++) {
 				fi = ((14318l * (n + 8)) / (m + 2)) >> k;
 				if ((di = abs(fi - freq)) < d) {
 					d = di;
-					f = fi;
-					lo = n;
-					hi = (k << 6) | m;
+					best_n = n;
+					best_m = m;
+					best_k = k;
 				}
 				if (fi > freq)
 					break;
 			}
+
+	if (is_oldclock(par->chip_id)) {
+		lo = best_n | (best_m << 7);
+		hi = (best_m >> 1) | (best_k << 4);
+	} else {
+		lo = best_n;
+		hi = best_m | (best_k << 6);
+	}
+
 	if (is3Dchip(par->chip_id)) {
 		vga_mm_wseq(par->io_virt, ClockHigh, hi);
 		vga_mm_wseq(par->io_virt, ClockLow, lo);

From 3876ae8beb2c7c19e21279b9603b1244fcd744dd Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:30:57 -0700
Subject: [PATCH 0454/1171] tridentfb: improve probe function

Add missing release of allocated fb_info structure and move enable_mmio() to
fix error path.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index 8ee4261abf3b..e79788a778fe 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -1287,6 +1287,7 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 
 	if (!request_mem_region(tridentfb_fix.mmio_start, tridentfb_fix.mmio_len, "tridentfb")) {
 		debug("request_region failed!\n");
+		framebuffer_release(info);
 		return -1;
 	}
 
@@ -1299,8 +1300,6 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 		goto out_unmap1;
 	}
 
-	enable_mmio();
-
 	/* setup framebuffer memory */
 	tridentfb_fix.smem_start = pci_resource_start(dev, 0);
 	tridentfb_fix.smem_len = get_memsize(default_par);
@@ -1312,6 +1311,8 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 		goto out_unmap1;
 	}
 
+	enable_mmio();
+
 	info->screen_base = ioremap_nocache(tridentfb_fix.smem_start,
 					    tridentfb_fix.smem_len);
 

From 0e73a47f094a919e2edeaa88e840cd0400adc423 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:30:58 -0700
Subject: [PATCH 0455/1171] tridentfb: improved register values on TGUI 9680

Improved values for some registers after Xorg Trident driver.  The main
problem was that values set by BIOS have been ignored.

This patch completely remove random pixels ("snow") on the TGUI 9680 and
9440 (not supported yet by the driver).  It does not help with the "snow"
on 3DImage and Blade3D cards.

There is also small improvement in timing calculations (hblank start and
vblank start)

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 29 +++++++++++++++++++++++------
 include/video/trident.h   |  1 +
 2 files changed, 24 insertions(+), 6 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index e79788a778fe..9668be881fe3 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -87,7 +87,17 @@ MODULE_PARM_DESC(crt, "Define if CRT is connected");
 
 static int is_oldclock(int id)
 {
-	return (id == TGUI9660);
+	return	(id == TGUI9660) ||
+		(id == CYBER9320);
+}
+
+static int is_oldprotect(int id)
+{
+	return	(id == TGUI9660) ||
+		(id == PROVIDIA9685) ||
+		(id == CYBER9320) ||
+		(id == CYBER9382) ||
+		(id == CYBER9385);
 }
 
 static int is_blade(int id)
@@ -143,6 +153,7 @@ static int iscyber(int id)
 
 	case CYBER9320:
 	case TGUI9660:
+	case PROVIDIA9685:
 	case IMAGE975:
 	case IMAGE985:
 	case BLADE3D:
@@ -905,14 +916,14 @@ static int tridentfb_set_par(struct fb_info *info)
 	hsyncend = (var->xres + var->right_margin + var->hsync_len) / 8 - 1;
 	htotal = (var->xres + var->left_margin + var->right_margin +
 		  var->hsync_len) / 8 - 5;
-	hblankstart = hdispend + 2;
+	hblankstart = hdispend + 1;
 	hblankend = htotal + 3;
 
 	vdispend = var->yres - 1;
 	vsyncstart = var->yres + var->lower_margin;
 	vsyncend = vsyncstart + var->vsync_len;
 	vtotal = var->upper_margin + vsyncend - 2;
-	vblankstart = vdispend + 2;
+	vblankstart = vdispend + 1;
 	vblankend = vtotal;
 
 	crtc_unlock(par);
@@ -1020,15 +1031,18 @@ static int tridentfb_set_par(struct fb_info *info)
 
 	write3X4(par, PixelBusReg, tmp);
 
-	tmp = 0x10;
+	tmp = read3X4(par, DRAMControl);
+	if (!is_oldprotect(par->chip_id))
+		tmp |= 0x10;
 	if (iscyber(par->chip_id))
 		tmp |= 0x20;
 	write3X4(par, DRAMControl, tmp);	/* both IO, linear enable */
 
 	write3X4(par, InterfaceSel, read3X4(par, InterfaceSel) | 0x40);
-	write3X4(par, Performance, 0x92);
+	if (!is_xp(par->chip_id))
+		write3X4(par, Performance, read3X4(par, Performance) | 0x10);
 	/* MMIO & PCI read and write burst enable */
-	write3X4(par, PCIReg, 0x07);
+	write3X4(par, PCIReg, read3X4(par, PCIReg) | 0x06);
 
 	/* convert from picoseconds to kHz */
 	vclk = PICOS2KHZ(info->var.pixclock);
@@ -1230,6 +1244,9 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 		revision = vga_io_rseq(RevisionID);
 
 		switch (revision) {
+		case 0x21:
+			chip_id = PROVIDIA9685;
+			break;
 		case 0x22:
 		case 0x23:
 			chip_id = CYBER9397;
diff --git a/include/video/trident.h b/include/video/trident.h
index 51ec5a95d1ad..fa690b9fcc13 100644
--- a/include/video/trident.h
+++ b/include/video/trident.h
@@ -25,6 +25,7 @@
 #define CYBER9520	0x9520
 #define CYBER9525DVD	0x9525
 #define TGUI9660	0x9660
+#define PROVIDIA9685	0x9685
 #define IMAGE975	0x9750
 #define IMAGE985	0x9850
 #define BLADE3D		0x9880

From a0d922562d56073f147a4de2983bee499dd2a10e Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:30:58 -0700
Subject: [PATCH 0456/1171] tridentfb: add TGUI 9440 support

Add support for TGUI 9440 chip.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 10 +++++++---
 include/video/trident.h   |  1 +
 2 files changed, 8 insertions(+), 3 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index 9668be881fe3..9dedc1a5e421 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -87,13 +87,15 @@ MODULE_PARM_DESC(crt, "Define if CRT is connected");
 
 static int is_oldclock(int id)
 {
-	return	(id == TGUI9660) ||
+	return	(id == TGUI9440) ||
+		(id == TGUI9660) ||
 		(id == CYBER9320);
 }
 
 static int is_oldprotect(int id)
 {
-	return	(id == TGUI9660) ||
+	return	(id == TGUI9440) ||
+		(id == TGUI9660) ||
 		(id == PROVIDIA9685) ||
 		(id == CYBER9320) ||
 		(id == CYBER9382) ||
@@ -1042,7 +1044,8 @@ static int tridentfb_set_par(struct fb_info *info)
 	if (!is_xp(par->chip_id))
 		write3X4(par, Performance, read3X4(par, Performance) | 0x10);
 	/* MMIO & PCI read and write burst enable */
-	write3X4(par, PCIReg, read3X4(par, PCIReg) | 0x06);
+	if (par->chip_id != TGUI9440)
+		write3X4(par, PCIReg, read3X4(par, PCIReg) | 0x06);
 
 	/* convert from picoseconds to kHz */
 	vclk = PICOS2KHZ(info->var.pixclock);
@@ -1418,6 +1421,7 @@ static struct pci_device_id trident_devices[] = {
 	{PCI_VENDOR_ID_TRIDENT,	CYBERBLADEAi1, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
 	{PCI_VENDOR_ID_TRIDENT,	CYBERBLADEAi1D, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
 	{PCI_VENDOR_ID_TRIDENT,	CYBERBLADEE4, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
+	{PCI_VENDOR_ID_TRIDENT,	TGUI9440, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
 	{PCI_VENDOR_ID_TRIDENT,	TGUI9660, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
 	{PCI_VENDOR_ID_TRIDENT,	IMAGE975, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
 	{PCI_VENDOR_ID_TRIDENT,	IMAGE985, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},
diff --git a/include/video/trident.h b/include/video/trident.h
index fa690b9fcc13..76d81b2198dc 100644
--- a/include/video/trident.h
+++ b/include/video/trident.h
@@ -24,6 +24,7 @@
 #define CYBER9397DVD	0x939A
 #define CYBER9520	0x9520
 #define CYBER9525DVD	0x9525
+#define TGUI9440	0x9440
 #define TGUI9660	0x9660
 #define PROVIDIA9685	0x9685
 #define IMAGE975	0x9750

From aa0aa8ab2f28d8985daa79ecab51970376e17157 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:30:59 -0700
Subject: [PATCH 0457/1171] tridentfb: fix unitialized pseudo_palette

Initialize the pseudo_palette pointer properly.  This fixes crash when
16bpp or 32bpp mode is selected.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index 9dedc1a5e421..6b3bb1e20738 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -1350,7 +1350,7 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 
 	info->fix = tridentfb_fix;
 	info->fbops = &tridentfb_ops;
-
+	info->pseudo_palette = default_par->pseudo_pal;
 
 	info->flags = FBINFO_DEFAULT | FBINFO_HWACCEL_YPAN;
 #ifdef CONFIG_FB_TRIDENT_ACCEL

From 74a933feaf13f705e6c798d87efe6a9d758b3ca0 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:00 -0700
Subject: [PATCH 0458/1171] tridentfb: improve check_var function

Do some additional checks (like pixelclock versus ramdac speed) to
eliminate modes which do not work.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 34 ++++++++++++++++++++++++++++++++++
 1 file changed, 34 insertions(+)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index 6b3bb1e20738..62701c5570fd 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -829,6 +829,7 @@ static int tridentfb_check_var(struct fb_var_screeninfo *var,
 {
 	struct tridentfb_par *par = info->par;
 	int bpp = var->bits_per_pixel;
+	int ramdac = 230000; /* 230MHz for most 3D chips */
 	debug("enter\n");
 
 	/* check color depth */
@@ -837,6 +838,12 @@ static int tridentfb_check_var(struct fb_var_screeninfo *var,
 	/* check whether resolution fits on panel and in memory */
 	if (par->flatpanel && nativex && var->xres > nativex)
 		return -EINVAL;
+	/* various resolution checks */
+	var->xres = (var->xres + 7) & ~0x7;
+	if (var->xres != var->xres_virtual)
+		var->xres_virtual = var->xres;
+	if (var->yres > var->yres_virtual)
+		var->yres_virtual = var->yres;
 	if (var->xres * var->yres_virtual * bpp / 8 > info->fix.smem_len)
 		return -EINVAL;
 
@@ -868,6 +875,33 @@ static int tridentfb_check_var(struct fb_var_screeninfo *var,
 	default:
 		return -EINVAL;
 	}
+
+	if (is_xp(par->chip_id))
+		ramdac = 350000;
+
+	switch (par->chip_id) {
+	case TGUI9440:
+		ramdac = 90000;
+		break;
+	case CYBER9320:
+	case TGUI9660:
+		ramdac = 135000;
+		break;
+	case PROVIDIA9685:
+	case CYBER9388:
+	case CYBER9382:
+	case CYBER9385:
+		ramdac = 170000;
+		break;
+	}
+
+	/* The clock is doubled for 32 bpp */
+	if (bpp == 32)
+		ramdac /= 2;
+
+	if (PICOS2KHZ(var->pixclock) > ramdac)
+		return -EINVAL;
+
 	debug("exit\n");
 
 	return 0;

From 65e93e038c8a6eb65b6907d6aed22a8ff1029d3a Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:00 -0700
Subject: [PATCH 0459/1171] tridentfb: preserve memory type settings

Do not overwrite bits which contain memory type settings.  It removes
noise pixels ("snow") on Blade3D and 3DImage chips.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index 62701c5570fd..26bc4d75d4fd 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -1095,7 +1095,10 @@ static int tridentfb_set_par(struct fb_info *info)
 	vga_mm_wseq(par->io_virt, 4, 0x0E); /* memory mode enable bitmaps ?? */
 
 	/* divide clock by 2 if 32bpp chain4 mode display and CPU path */
-	write3CE(par, MiscExtFunc, (bpp == 32) ? 0x1A : 0x12);
+	tmp = read3CE(par, MiscExtFunc) & 0xF0;
+	if (bpp == 32)
+		tmp |= 8;
+	write3CE(par, MiscExtFunc, tmp | 0x12);
 	write3CE(par, 0x5, 0x40);	/* no CGA compat, allow 256 col */
 	write3CE(par, 0x6, 0x05);	/* graphics mode */
 	write3CE(par, 0x7, 0x0F);	/* planes? */

From 54f019e54244fef0ad927ce5501927d9033492de Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:01 -0700
Subject: [PATCH 0460/1171] tridentfb: fix hi-color modes for TGUI 9440

The TGUI 9440 requires doubling clock for 16bpp (hi-color) modes.

The patch also moves back enable_mmio() call to the right position.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 22 ++++++++++++----------
 1 file changed, 12 insertions(+), 10 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index 26bc4d75d4fd..ed1b32a1cef5 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -835,6 +835,8 @@ static int tridentfb_check_var(struct fb_var_screeninfo *var,
 	/* check color depth */
 	if (bpp == 24)
 		bpp = var->bits_per_pixel = 32;
+	if (par->chip_id == TGUI9440 && bpp == 32)
+		return -EINVAL;
 	/* check whether resolution fits on panel and in memory */
 	if (par->flatpanel && nativex && var->xres > nativex)
 		return -EINVAL;
@@ -881,7 +883,7 @@ static int tridentfb_check_var(struct fb_var_screeninfo *var,
 
 	switch (par->chip_id) {
 	case TGUI9440:
-		ramdac = 90000;
+		ramdac = (bpp >= 16) ? 45000 : 90000;
 		break;
 	case CYBER9320:
 	case TGUI9660:
@@ -1081,12 +1083,6 @@ static int tridentfb_set_par(struct fb_info *info)
 	if (par->chip_id != TGUI9440)
 		write3X4(par, PCIReg, read3X4(par, PCIReg) | 0x06);
 
-	/* convert from picoseconds to kHz */
-	vclk = PICOS2KHZ(info->var.pixclock);
-	if (bpp == 32)
-		vclk *= 2;
-	set_vclk(par, vclk);
-
 	vga_mm_wseq(par->io_virt, 0, 3);
 	vga_mm_wseq(par->io_virt, 1, 1); /* set char clock 8 dots wide */
 	/* enable 4 maps because needed in chain4 mode */
@@ -1094,10 +1090,16 @@ static int tridentfb_set_par(struct fb_info *info)
 	vga_mm_wseq(par->io_virt, 3, 0);
 	vga_mm_wseq(par->io_virt, 4, 0x0E); /* memory mode enable bitmaps ?? */
 
+	/* convert from picoseconds to kHz */
+	vclk = PICOS2KHZ(info->var.pixclock);
+
 	/* divide clock by 2 if 32bpp chain4 mode display and CPU path */
 	tmp = read3CE(par, MiscExtFunc) & 0xF0;
-	if (bpp == 32)
+	if (bpp == 32 || (par->chip_id == TGUI9440 && bpp == 16)) {
 		tmp |= 8;
+		vclk *= 2;
+	}
+	set_vclk(par, vclk);
 	write3CE(par, MiscExtFunc, tmp | 0x12);
 	write3CE(par, 0x5, 0x40);	/* no CGA compat, allow 256 col */
 	write3CE(par, 0x6, 0x05);	/* graphics mode */
@@ -1361,6 +1363,8 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 	tridentfb_fix.smem_start = pci_resource_start(dev, 0);
 	tridentfb_fix.smem_len = get_memsize(default_par);
 
+	enable_mmio();
+
 	if (!request_mem_region(tridentfb_fix.smem_start, tridentfb_fix.smem_len, "tridentfb")) {
 		debug("request_mem_region failed!\n");
 		disable_mmio(info->par);
@@ -1368,8 +1372,6 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 		goto out_unmap1;
 	}
 
-	enable_mmio();
-
 	info->screen_base = ioremap_nocache(tridentfb_fix.smem_start,
 					    tridentfb_fix.smem_len);
 

From bcac2d5fe36238dcfc955b49f9db10ad3ae3e53c Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:01 -0700
Subject: [PATCH 0461/1171] tridentfb: add acceleration for TGUI families

This patch adds acceleration for TGUI 9440 and 96xx chips.  These chips
requires line length to be power of 2, so this is also changed.

It also moves the troubling enable_mmio() function to its final
destination.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 135 ++++++++++++++++++++++++++++++++++----
 1 file changed, 123 insertions(+), 12 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index ed1b32a1cef5..9b87c08e517d 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -491,6 +491,95 @@ static void image_copy_rect(struct tridentfb_par *par,
 		 0x80000000 | 1 << 22 | 1 << 10 | 1 << 7 | direction);
 }
 
+/*
+ * TGUI 9440/96XX acceleration
+ */
+
+static void tgui_init_accel(struct tridentfb_par *par, int pitch, int bpp)
+{
+	unsigned char x = 0;
+
+	/* disable clipping */
+	writemmr(par, 0x2148, 0);
+	writemmr(par, 0x214C, point(4095, 2047));
+
+	switch (bpp) {
+	case 8:
+		x = 0;
+		break;
+	case 16:
+		x = 1;
+		break;
+	case 24:
+		x = 3;
+		break;
+	case 32:
+		x = 2;
+		break;
+	}
+
+	switch ((pitch * bpp) / 8) {
+	case 8192:
+	case 512:
+		x |= 0x00;
+		break;
+	case 1024:
+		x |= 0x04;
+		break;
+	case 2048:
+		x |= 0x08;
+		break;
+	case 4096:
+		x |= 0x0C;
+		break;
+	}
+
+	fb_writew(x, par->io_virt + 0x2122);
+}
+
+static void tgui_fill_rect(struct tridentfb_par *par,
+			   u32 x, u32 y, u32 w, u32 h, u32 c, u32 rop)
+{
+	t_outb(par, ROP_P, 0x2127);
+	writemmr(par, 0x212c, c);
+	writemmr(par, 0x2128, 0x4020);
+	writemmr(par, 0x2140, point(w - 1, h - 1));
+	writemmr(par, 0x2138, point(x, y));
+	t_outb(par, 1, 0x2124);
+}
+
+static void tgui_copy_rect(struct tridentfb_par *par,
+			   u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h)
+{
+	int flags = 0;
+	u16 x1_tmp, x2_tmp, y1_tmp, y2_tmp;
+
+	if ((x1 < x2) && (y1 == y2)) {
+		flags |= 0x0200;
+		x1_tmp = x1 + w - 1;
+		x2_tmp = x2 + w - 1;
+	} else {
+		x1_tmp = x1;
+		x2_tmp = x2;
+	}
+
+	if (y1 < y2) {
+		flags |= 0x0100;
+		y1_tmp = y1 + h - 1;
+		y2_tmp = y2 + h - 1;
+	} else {
+		y1_tmp = y1;
+		y2_tmp = y2;
+	}
+
+	writemmr(par, 0x2128, 0x4 | flags);
+	t_outb(par, ROP_S, 0x2127);
+	writemmr(par, 0x213C, point(x1_tmp, y1_tmp));
+	writemmr(par, 0x2138, point(x2_tmp, y2_tmp));
+	writemmr(par, 0x2140, point(w - 1, h - 1));
+	t_outb(par, 1, 0x2124);
+}
+
 /*
  * Accel functions called by the upper layers
  */
@@ -530,12 +619,8 @@ static void tridentfb_copyarea(struct fb_info *info,
 		       ca->width, ca->height);
 	par->wait_engine(par);
 }
-#else /* !CONFIG_FB_TRIDENT_ACCEL */
-#define tridentfb_fillrect cfb_fillrect
-#define tridentfb_copyarea cfb_copyarea
 #endif /* CONFIG_FB_TRIDENT_ACCEL */
 
-
 /*
  * Hardware access functions
  */
@@ -829,6 +914,7 @@ static int tridentfb_check_var(struct fb_var_screeninfo *var,
 {
 	struct tridentfb_par *par = info->par;
 	int bpp = var->bits_per_pixel;
+	int line_length;
 	int ramdac = 230000; /* 230MHz for most 3D chips */
 	debug("enter\n");
 
@@ -844,9 +930,27 @@ static int tridentfb_check_var(struct fb_var_screeninfo *var,
 	var->xres = (var->xres + 7) & ~0x7;
 	if (var->xres != var->xres_virtual)
 		var->xres_virtual = var->xres;
+	line_length = var->xres_virtual * bpp / 8;
+#ifdef CONFIG_FB_TRIDENT_ACCEL
+	if (!is3Dchip(par->chip_id)) {
+		/* acceleration requires line length to be power of 2 */
+		if (line_length <= 512)
+			var->xres_virtual = 512 * 8 / bpp;
+		else if (line_length <= 1024)
+			var->xres_virtual = 1024 * 8 / bpp;
+		else if (line_length <= 2048)
+			var->xres_virtual = 2048 * 8 / bpp;
+		else if (line_length <= 4096)
+			var->xres_virtual = 4096 * 8 / bpp;
+		else if (line_length <= 8192)
+			var->xres_virtual = 8192 * 8 / bpp;
+
+		line_length = var->xres_virtual * bpp / 8;
+	}
+#endif
 	if (var->yres > var->yres_virtual)
 		var->yres_virtual = var->yres;
-	if (var->xres * var->yres_virtual * bpp / 8 > info->fix.smem_len)
+	if (line_length * var->yres_virtual > info->fix.smem_len)
 		return -EINVAL;
 
 	switch (bpp) {
@@ -918,7 +1022,7 @@ static int tridentfb_pan_display(struct fb_var_screeninfo *var,
 	unsigned int offset;
 
 	debug("enter\n");
-	offset = (var->xoffset + (var->yoffset * var->xres))
+	offset = (var->xoffset + (var->yoffset * var->xres_virtual))
 		* var->bits_per_pixel / 32;
 	info->var.xoffset = var->xoffset;
 	info->var.yoffset = var->yoffset;
@@ -1049,7 +1153,7 @@ static int tridentfb_set_par(struct fb_info *info)
 	write3X4(par, GraphEngReg, 0x80);
 
 #ifdef CONFIG_FB_TRIDENT_ACCEL
-	par->init_accel(par, info->var.xres, bpp);
+	par->init_accel(par, info->var.xres_virtual, bpp);
 #endif
 
 	switch (bpp) {
@@ -1147,9 +1251,9 @@ static int tridentfb_set_par(struct fb_info *info)
 
 	if (par->flatpanel)
 		set_number_of_lines(par, info->var.yres);
-	set_lwidth(par, info->var.xres * bpp / (4 * 16));
+	info->fix.line_length = info->var.xres_virtual * bpp / 8;
+	set_lwidth(par, info->fix.line_length / 8);
 	info->fix.visual = (bpp == 8) ? FB_VISUAL_PSEUDOCOLOR : FB_VISUAL_TRUECOLOR;
-	info->fix.line_length = info->var.xres * (bpp >> 3);
 	info->cmap.len = (bpp == 8) ? 256 : 16;
 	debug("exit\n");
 	return 0;
@@ -1248,9 +1352,11 @@ static struct fb_ops tridentfb_ops = {
 	.fb_blank = tridentfb_blank,
 	.fb_check_var = tridentfb_check_var,
 	.fb_set_par = tridentfb_set_par,
+#ifdef CONFIG_FB_TRIDENT_ACCEL
 	.fb_fillrect = tridentfb_fillrect,
 	.fb_copyarea = tridentfb_copyarea,
 	.fb_imageblit = cfb_imageblit,
+#endif
 };
 
 static int __devinit trident_pci_probe(struct pci_dev *dev,
@@ -1328,11 +1434,16 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 		default_par->wait_engine = blade_wait_engine;
 		default_par->fill_rect = blade_fill_rect;
 		default_par->copy_rect = blade_copy_rect;
-	} else {
+	} else if (chip3D) {			/* 3DImage family left */
 		default_par->init_accel = image_init_accel;
 		default_par->wait_engine = image_wait_engine;
 		default_par->fill_rect = image_fill_rect;
 		default_par->copy_rect = image_copy_rect;
+	} else { 				/* TGUI 9440/96XX family */
+		default_par->init_accel = tgui_init_accel;
+		default_par->wait_engine = xp_wait_engine;
+		default_par->fill_rect = tgui_fill_rect;
+		default_par->copy_rect = tgui_copy_rect;
 	}
 
 	default_par->chip_id = chip_id;
@@ -1359,12 +1470,12 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 		goto out_unmap1;
 	}
 
+	enable_mmio();
+
 	/* setup framebuffer memory */
 	tridentfb_fix.smem_start = pci_resource_start(dev, 0);
 	tridentfb_fix.smem_len = get_memsize(default_par);
 
-	enable_mmio();
-
 	if (!request_mem_region(tridentfb_fix.smem_start, tridentfb_fix.smem_len, "tridentfb")) {
 		debug("request_mem_region failed!\n");
 		disable_mmio(info->par);

From 49b1f4b44bcdc47a10d2b354b269305043ef2a32 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:02 -0700
Subject: [PATCH 0462/1171] tridentfb: acceleration code improvements

This patch brings various acceleration improvements:
- set  copyarea/fillrect for non-accelerated framebuffer (fix)
- remove 15 bpp depth handling to simplify code as it hardly
  works (15 bpp handling was obviously missing in some switches)
- add fb_sync call and move waiting before accelerated function
  to make acceleration more asynchronous to cpu (few % of speed
  improvement)
- add cpu_relax() call in waiting loops
- make longer register names and name more registers
- move registers' definition to header
- general code improvements (shortening, simplifying)

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 263 +++++++++++++-------------------------
 include/video/trident.h   |  21 ++-
 2 files changed, 110 insertions(+), 174 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index 9b87c08e517d..4a1258f9509e 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -1,5 +1,5 @@
 /*
- * Frame buffer driver for Trident Blade and Image series
+ * Frame buffer driver for Trident TGUI, Blade and Image series
  *
  * Copyright 2001, 2002 - Jani Monoses   <jani@iv.ro>
  *
@@ -13,7 +13,6 @@
  *	code, suggestions
  * TODO:
  *	timing value tweaking so it looks good on every monitor in every mode
- *	TGUI acceleration
  */
 
 #include <linux/module.h>
@@ -193,37 +192,13 @@ static inline u32 readmmr(struct tridentfb_par *par, u16 r)
  */
 
 #define point(x, y) ((y) << 16 | (x))
-#define STA	0x2120
-#define CMD	0x2144
-#define ROP	0x2148
-#define CLR	0x2160
-#define SR1	0x2100
-#define SR2	0x2104
-#define DR1	0x2108
-#define DR2	0x210C
-
-#define ROP_S	0xCC
 
 static void blade_init_accel(struct tridentfb_par *par, int pitch, int bpp)
 {
 	int v1 = (pitch >> 3) << 20;
-	int tmp = 0, v2;
-	switch (bpp) {
-	case 8:
-		tmp = 0;
-		break;
-	case 15:
-		tmp = 5;
-		break;
-	case 16:
-		tmp = 1;
-		break;
-	case 24:
-	case 32:
-		tmp = 2;
-		break;
-	}
-	v2 = v1 | (tmp << 29);
+	int tmp = bpp == 24 ? 2 : (bpp >> 4);
+	int v2 = v1 | (tmp << 29);
+
 	writemmr(par, 0x21C0, v2);
 	writemmr(par, 0x21C4, v2);
 	writemmr(par, 0x21B8, v2);
@@ -237,29 +212,29 @@ static void blade_init_accel(struct tridentfb_par *par, int pitch, int bpp)
 
 static void blade_wait_engine(struct tridentfb_par *par)
 {
-	while (readmmr(par, STA) & 0xFA800000) ;
+	while (readmmr(par, STATUS) & 0xFA800000)
+		cpu_relax();
 }
 
 static void blade_fill_rect(struct tridentfb_par *par,
 			    u32 x, u32 y, u32 w, u32 h, u32 c, u32 rop)
 {
-	writemmr(par, CLR, c);
-	writemmr(par, ROP, rop ? 0x66 : ROP_S);
+	writemmr(par, COLOR, c);
+	writemmr(par, ROP, rop ? ROP_X : ROP_S);
 	writemmr(par, CMD, 0x20000000 | 1 << 19 | 1 << 4 | 2 << 2);
 
-	writemmr(par, DR1, point(x, y));
-	writemmr(par, DR2, point(x + w - 1, y + h - 1));
+	writemmr(par, DST1, point(x, y));
+	writemmr(par, DST2, point(x + w - 1, y + h - 1));
 }
 
 static void blade_copy_rect(struct tridentfb_par *par,
 			    u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h)
 {
-	u32 s1, s2, d1, d2;
 	int direction = 2;
-	s1 = point(x1, y1);
-	s2 = point(x1 + w - 1, y1 + h - 1);
-	d1 = point(x2, y2);
-	d2 = point(x2 + w - 1, y2 + h - 1);
+	u32 s1 = point(x1, y1);
+	u32 s2 = point(x1 + w - 1, y1 + h - 1);
+	u32 d1 = point(x2, y2);
+	u32 d2 = point(x2 + w - 1, y2 + h - 1);
 
 	if ((y1 > y2) || ((y1 == y2) && (x1 > x2)))
 		direction = 0;
@@ -267,38 +242,20 @@ static void blade_copy_rect(struct tridentfb_par *par,
 	writemmr(par, ROP, ROP_S);
 	writemmr(par, CMD, 0xE0000000 | 1 << 19 | 1 << 4 | 1 << 2 | direction);
 
-	writemmr(par, SR1, direction ? s2 : s1);
-	writemmr(par, SR2, direction ? s1 : s2);
-	writemmr(par, DR1, direction ? d2 : d1);
-	writemmr(par, DR2, direction ? d1 : d2);
+	writemmr(par, SRC1, direction ? s2 : s1);
+	writemmr(par, SRC2, direction ? s1 : s2);
+	writemmr(par, DST1, direction ? d2 : d1);
+	writemmr(par, DST2, direction ? d1 : d2);
 }
 
 /*
  * BladeXP specific acceleration functions
  */
 
-#define ROP_P 0xF0
-#define masked_point(x, y) ((y & 0xffff)<<16|(x & 0xffff))
-
 static void xp_init_accel(struct tridentfb_par *par, int pitch, int bpp)
 {
-	int tmp = 0, v1;
-	unsigned char x = 0;
-
-	switch (bpp) {
-	case 8:
-		x = 0;
-		break;
-	case 16:
-		x = 1;
-		break;
-	case 24:
-		x = 3;
-		break;
-	case 32:
-		x = 2;
-		break;
-	}
+	unsigned char x = bpp == 24 ? 3 : (bpp >> 4);
+	int v1 = pitch << (bpp == 24 ? 20 : (18 + x));
 
 	switch (pitch << (bpp >> 3)) {
 	case 8192:
@@ -320,22 +277,6 @@ static void xp_init_accel(struct tridentfb_par *par, int pitch, int bpp)
 
 	eng_oper = x | 0x40;
 
-	switch (bpp) {
-	case 8:
-		tmp = 18;
-		break;
-	case 15:
-	case 16:
-		tmp = 19;
-		break;
-	case 24:
-	case 32:
-		tmp = 20;
-		break;
-	}
-
-	v1 = pitch << tmp;
-
 	writemmr(par, 0x2154, v1);
 	writemmr(par, 0x2150, v1);
 	t_outb(par, 3, 0x2126);
@@ -343,15 +284,11 @@ static void xp_init_accel(struct tridentfb_par *par, int pitch, int bpp)
 
 static void xp_wait_engine(struct tridentfb_par *par)
 {
-	int busy;
 	int count, timeout;
 
 	count = 0;
 	timeout = 0;
-	for (;;) {
-		busy = t_inb(par, STA) & 0x80;
-		if (busy != 0x80)
-			return;
+	while (t_inb(par, STATUS) & 0x80) {
 		count++;
 		if (count == 10000000) {
 			/* Timeout */
@@ -359,10 +296,11 @@ static void xp_wait_engine(struct tridentfb_par *par)
 			timeout++;
 			if (timeout == 8) {
 				/* Reset engine */
-				t_outb(par, 0x00, 0x2120);
+				t_outb(par, 0x00, STATUS);
 				return;
 			}
 		}
+		cpu_relax();
 	}
 }
 
@@ -371,10 +309,10 @@ static void xp_fill_rect(struct tridentfb_par *par,
 {
 	writemmr(par, 0x2127, ROP_P);
 	writemmr(par, 0x2158, c);
-	writemmr(par, 0x2128, 0x4000);
-	writemmr(par, 0x2140, masked_point(h, w));
-	writemmr(par, 0x2138, masked_point(y, x));
-	t_outb(par, 0x01, 0x2124);
+	writemmr(par, DRAWFL, 0x4000);
+	writemmr(par, OLDDIM, point(h, w));
+	writemmr(par, OLDDST, point(y, x));
+	t_outb(par, 0x01, OLDCMD);
 	t_outb(par, eng_oper, 0x2125);
 }
 
@@ -404,12 +342,12 @@ static void xp_copy_rect(struct tridentfb_par *par,
 		y2_tmp = y2;
 	}
 
-	writemmr(par, 0x2128, direction);
+	writemmr(par, DRAWFL, direction);
 	t_outb(par, ROP_S, 0x2127);
-	writemmr(par, 0x213C, masked_point(y1_tmp, x1_tmp));
-	writemmr(par, 0x2138, masked_point(y2_tmp, x2_tmp));
-	writemmr(par, 0x2140, masked_point(h, w));
-	t_outb(par, 0x01, 0x2124);
+	writemmr(par, OLDSRC, point(y1_tmp, x1_tmp));
+	writemmr(par, OLDDST, point(y2_tmp, x2_tmp));
+	writemmr(par, OLDDIM, point(h, w));
+	t_outb(par, 0x01, OLDCMD);
 }
 
 /*
@@ -417,22 +355,8 @@ static void xp_copy_rect(struct tridentfb_par *par,
  */
 static void image_init_accel(struct tridentfb_par *par, int pitch, int bpp)
 {
-	int tmp = 0;
-	switch (bpp) {
-	case 8:
-		tmp = 0;
-		break;
-	case 15:
-		tmp = 5;
-		break;
-	case 16:
-		tmp = 1;
-		break;
-	case 24:
-	case 32:
-		tmp = 2;
-		break;
-	}
+	int tmp = bpp == 24 ? 2: (bpp >> 4);
+
 	writemmr(par, 0x2120, 0xF0000000);
 	writemmr(par, 0x2120, 0x40000000 | tmp);
 	writemmr(par, 0x2120, 0x80000000);
@@ -450,7 +374,8 @@ static void image_init_accel(struct tridentfb_par *par, int pitch, int bpp)
 
 static void image_wait_engine(struct tridentfb_par *par)
 {
-	while (readmmr(par, 0x2164) & 0xF0000000) ;
+	while (readmmr(par, 0x2164) & 0xF0000000)
+		cpu_relax();
 }
 
 static void image_fill_rect(struct tridentfb_par *par,
@@ -461,8 +386,8 @@ static void image_fill_rect(struct tridentfb_par *par,
 
 	writemmr(par, 0x2144, c);
 
-	writemmr(par, DR1, point(x, y));
-	writemmr(par, DR2, point(x + w - 1, y + h - 1));
+	writemmr(par, DST1, point(x, y));
+	writemmr(par, DST2, point(x + w - 1, y + h - 1));
 
 	writemmr(par, 0x2124, 0x80000000 | 3 << 22 | 1 << 10 | 1 << 9);
 }
@@ -470,12 +395,11 @@ static void image_fill_rect(struct tridentfb_par *par,
 static void image_copy_rect(struct tridentfb_par *par,
 			    u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h)
 {
-	u32 s1, s2, d1, d2;
 	int direction = 2;
-	s1 = point(x1, y1);
-	s2 = point(x1 + w - 1, y1 + h - 1);
-	d1 = point(x2, y2);
-	d2 = point(x2 + w - 1, y2 + h - 1);
+	u32 s1 = point(x1, y1);
+	u32 s2 = point(x1 + w - 1, y1 + h - 1);
+	u32 d1 = point(x2, y2);
+	u32 d2 = point(x2 + w - 1, y2 + h - 1);
 
 	if ((y1 > y2) || ((y1 == y2) && (x1 > x2)))
 		direction = 0;
@@ -483,10 +407,10 @@ static void image_copy_rect(struct tridentfb_par *par,
 	writemmr(par, 0x2120, 0x80000000);
 	writemmr(par, 0x2120, 0x90000000 | ROP_S);
 
-	writemmr(par, SR1, direction ? s2 : s1);
-	writemmr(par, SR2, direction ? s1 : s2);
-	writemmr(par, DR1, direction ? d2 : d1);
-	writemmr(par, DR2, direction ? d1 : d2);
+	writemmr(par, SRC1, direction ? s2 : s1);
+	writemmr(par, SRC2, direction ? s1 : s2);
+	writemmr(par, DST1, direction ? d2 : d1);
+	writemmr(par, DST2, direction ? d1 : d2);
 	writemmr(par, 0x2124,
 		 0x80000000 | 1 << 22 | 1 << 10 | 1 << 7 | direction);
 }
@@ -497,27 +421,12 @@ static void image_copy_rect(struct tridentfb_par *par,
 
 static void tgui_init_accel(struct tridentfb_par *par, int pitch, int bpp)
 {
-	unsigned char x = 0;
+	unsigned char x = bpp == 24 ? 3 : (bpp >> 4);
 
 	/* disable clipping */
 	writemmr(par, 0x2148, 0);
 	writemmr(par, 0x214C, point(4095, 2047));
 
-	switch (bpp) {
-	case 8:
-		x = 0;
-		break;
-	case 16:
-		x = 1;
-		break;
-	case 24:
-		x = 3;
-		break;
-	case 32:
-		x = 2;
-		break;
-	}
-
 	switch ((pitch * bpp) / 8) {
 	case 8192:
 	case 512:
@@ -541,11 +450,11 @@ static void tgui_fill_rect(struct tridentfb_par *par,
 			   u32 x, u32 y, u32 w, u32 h, u32 c, u32 rop)
 {
 	t_outb(par, ROP_P, 0x2127);
-	writemmr(par, 0x212c, c);
-	writemmr(par, 0x2128, 0x4020);
-	writemmr(par, 0x2140, point(w - 1, h - 1));
-	writemmr(par, 0x2138, point(x, y));
-	t_outb(par, 1, 0x2124);
+	writemmr(par, OLDCLR, c);
+	writemmr(par, DRAWFL, 0x4020);
+	writemmr(par, OLDDIM, point(w - 1, h - 1));
+	writemmr(par, OLDDST, point(x, y));
+	t_outb(par, 1, OLDCMD);
 }
 
 static void tgui_copy_rect(struct tridentfb_par *par,
@@ -572,12 +481,12 @@ static void tgui_copy_rect(struct tridentfb_par *par,
 		y2_tmp = y2;
 	}
 
-	writemmr(par, 0x2128, 0x4 | flags);
+	writemmr(par, DRAWFL, 0x4 | flags);
 	t_outb(par, ROP_S, 0x2127);
-	writemmr(par, 0x213C, point(x1_tmp, y1_tmp));
-	writemmr(par, 0x2138, point(x2_tmp, y2_tmp));
-	writemmr(par, 0x2140, point(w - 1, h - 1));
-	t_outb(par, 1, 0x2124);
+	writemmr(par, OLDSRC, point(x1_tmp, y1_tmp));
+	writemmr(par, OLDDST, point(x2_tmp, y2_tmp));
+	writemmr(par, OLDDIM, point(w - 1, h - 1));
+	t_outb(par, 1, OLDCMD);
 }
 
 /*
@@ -588,37 +497,40 @@ static void tridentfb_fillrect(struct fb_info *info,
 			       const struct fb_fillrect *fr)
 {
 	struct tridentfb_par *par = info->par;
-	int bpp = info->var.bits_per_pixel;
-	int col = 0;
+	int col;
 
-	switch (bpp) {
-	default:
-	case 8:
-		col |= fr->color;
+	if (info->var.bits_per_pixel == 8) {
+		col = fr->color;
 		col |= col << 8;
 		col |= col << 16;
-		break;
-	case 16:
+	} else
 		col = ((u32 *)(info->pseudo_palette))[fr->color];
-		break;
-	case 32:
-		col = ((u32 *)(info->pseudo_palette))[fr->color];
-		break;
-	}
 
+	par->wait_engine(par);
 	par->fill_rect(par, fr->dx, fr->dy, fr->width,
 		       fr->height, col, fr->rop);
-	par->wait_engine(par);
 }
+
 static void tridentfb_copyarea(struct fb_info *info,
 			       const struct fb_copyarea *ca)
 {
 	struct tridentfb_par *par = info->par;
 
+	par->wait_engine(par);
 	par->copy_rect(par, ca->sx, ca->sy, ca->dx, ca->dy,
 		       ca->width, ca->height);
-	par->wait_engine(par);
 }
+
+static int tridentfb_sync(struct fb_info *info)
+{
+	struct tridentfb_par *par = info->par;
+
+	par->wait_engine(par);
+	return 0;
+}
+#else
+#define tridentfb_fillrect cfb_fillrect
+#define tridentfb_copyarea cfb_copyarea
 #endif /* CONFIG_FB_TRIDENT_ACCEL */
 
 /*
@@ -921,6 +833,8 @@ static int tridentfb_check_var(struct fb_var_screeninfo *var,
 	/* check color depth */
 	if (bpp == 24)
 		bpp = var->bits_per_pixel = 32;
+	if (bpp != 8 && bpp != 16 && bpp != 32)
+		return -EINVAL;
 	if (par->chip_id == TGUI9440 && bpp == 32)
 		return -EINVAL;
 	/* check whether resolution fits on panel and in memory */
@@ -928,8 +842,15 @@ static int tridentfb_check_var(struct fb_var_screeninfo *var,
 		return -EINVAL;
 	/* various resolution checks */
 	var->xres = (var->xres + 7) & ~0x7;
-	if (var->xres != var->xres_virtual)
+	if (var->xres > var->xres_virtual)
 		var->xres_virtual = var->xres;
+	if (var->yres > var->yres_virtual)
+		var->yres_virtual = var->yres;
+	if (var->xres_virtual > 4095 || var->yres > 2048)
+		return -EINVAL;
+	/* prevent from position overflow for acceleration */
+	if (var->yres_virtual > 0xffff)
+		return -EINVAL;
 	line_length = var->xres_virtual * bpp / 8;
 #ifdef CONFIG_FB_TRIDENT_ACCEL
 	if (!is3Dchip(par->chip_id)) {
@@ -944,6 +865,8 @@ static int tridentfb_check_var(struct fb_var_screeninfo *var,
 			var->xres_virtual = 4096 * 8 / bpp;
 		else if (line_length <= 8192)
 			var->xres_virtual = 8192 * 8 / bpp;
+		else
+			return -EINVAL;
 
 		line_length = var->xres_virtual * bpp / 8;
 	}
@@ -1229,9 +1152,6 @@ static int tridentfb_set_par(struct fb_info *info)
 	case 8:
 		tmp = 0;
 		break;
-	case 15:
-		tmp = 0x10;
-		break;
 	case 16:
 		tmp = 0x30;
 		break;
@@ -1352,10 +1272,11 @@ static struct fb_ops tridentfb_ops = {
 	.fb_blank = tridentfb_blank,
 	.fb_check_var = tridentfb_check_var,
 	.fb_set_par = tridentfb_set_par,
-#ifdef CONFIG_FB_TRIDENT_ACCEL
 	.fb_fillrect = tridentfb_fillrect,
 	.fb_copyarea = tridentfb_copyarea,
 	.fb_imageblit = cfb_imageblit,
+#ifdef CONFIG_FB_TRIDENT_ACCEL
+	.fb_sync = tridentfb_sync,
 #endif
 };
 
@@ -1366,7 +1287,6 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 	unsigned char revision;
 	struct fb_info *info;
 	struct tridentfb_par *default_par;
-	int defaultaccel;
 	int chip3D;
 	int chip_id;
 
@@ -1448,9 +1368,6 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 
 	default_par->chip_id = chip_id;
 
-	/* acceleration is on by default for 3D chips */
-	defaultaccel = chip3D && !noaccel;
-
 	/* setup MMIO region */
 	tridentfb_fix.mmio_start = pci_resource_start(dev, 1);
 	tridentfb_fix.mmio_len = chip3D ? 0x20000 : 0x10000;
@@ -1515,7 +1432,7 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 	if (err < 0)
 		goto out_unmap2;
 
-	if (defaultaccel && default_par->init_accel)
+	if (!noaccel && default_par->init_accel)
 		info->var.accel_flags |= FB_ACCELF_TEXT;
 	else
 		info->var.accel_flags &= ~FB_ACCELF_TEXT;
diff --git a/include/video/trident.h b/include/video/trident.h
index 76d81b2198dc..7540501bc040 100644
--- a/include/video/trident.h
+++ b/include/video/trident.h
@@ -4,7 +4,7 @@
 #endif
 
 #if TRIDENTFB_DEBUG
-#define debug(f,a...)	printk("%s:" f,  __FUNCTION__ , ## a);mdelay(1000);
+#define debug(f,a...)	printk("%s:" f,  __FUNCTION__ , ## a);mdelay(100);
 #else
 #define debug(f,a...)
 #endif
@@ -124,3 +124,22 @@
 #define BiosMode     0x5c
 #define BiosReg      0x5d
 
+/* Graphics Engine */
+#define STATUS	0x2120
+#define OLDCMD	0x2124
+#define DRAWFL	0x2128
+#define OLDCLR	0x212C
+#define OLDDST	0x2138
+#define OLDSRC	0x213C
+#define OLDDIM	0x2140
+#define CMD	0x2144
+#define ROP	0x2148
+#define COLOR	0x2160
+#define SRC1	0x2100
+#define SRC2	0x2104
+#define DST1	0x2108
+#define DST2	0x210C
+
+#define ROP_S	0xCC
+#define ROP_P	0xF0
+#define ROP_X	0x66

From 2c86a0c26fbe8ea218f7a267645679fb78aba8a3 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:03 -0700
Subject: [PATCH 0463/1171] tridentfb: acceleration bug fixes

This patch fixes two problems when acceleration is enabled:

 - bit for bitblt direction is corrected
   so scrolling down works as expected on 3DImage chips

 - initialization of acceleration is done later
   this helps with initial console malfuntion (on Blade3D
   chips) well documented here:
   http://marc.info/?l=linux-fbdev-users&m=111386953124478&w=2

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index 4a1258f9509e..d896dee7b487 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -395,7 +395,7 @@ static void image_fill_rect(struct tridentfb_par *par,
 static void image_copy_rect(struct tridentfb_par *par,
 			    u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h)
 {
-	int direction = 2;
+	int direction = 0x4;
 	u32 s1 = point(x1, y1);
 	u32 s2 = point(x1 + w - 1, y1 + h - 1);
 	u32 d1 = point(x2, y2);
@@ -1075,10 +1075,6 @@ static int tridentfb_set_par(struct fb_info *info)
 	/* enable GE for text acceleration */
 	write3X4(par, GraphEngReg, 0x80);
 
-#ifdef CONFIG_FB_TRIDENT_ACCEL
-	par->init_accel(par, info->var.xres_virtual, bpp);
-#endif
-
 	switch (bpp) {
 	case 8:
 		tmp = 0x00;
@@ -1173,6 +1169,10 @@ static int tridentfb_set_par(struct fb_info *info)
 		set_number_of_lines(par, info->var.yres);
 	info->fix.line_length = info->var.xres_virtual * bpp / 8;
 	set_lwidth(par, info->fix.line_length / 8);
+#ifdef CONFIG_FB_TRIDENT_ACCEL
+	par->init_accel(par, info->var.xres_virtual, bpp);
+#endif
+
 	info->fix.visual = (bpp == 8) ? FB_VISUAL_PSEUDOCOLOR : FB_VISUAL_TRUECOLOR;
 	info->cmap.len = (bpp == 8) ? 256 : 16;
 	debug("exit\n");

From 34dec24317d6824b7db172bb0072b909a9c376f2 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:04 -0700
Subject: [PATCH 0464/1171] tridentfb: various pixclock and timing improvements

This patch fixes few issues related to timings and pixclock generation:

 - disallow the pixclocks with numerator lower than
   double denominator. This fixes display instability
   for some modes.
 - choose the pixelclock with the highest
   numerator and denominator values. This improve
   image quality and fixes display instability
   for some modes.
 - make interlaced modes work.
 - set synchronization pulses polarization
   correctly.
 - horizontal synchronization timing are now
   the same as generated by X.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 35 ++++++++++++++++++++++++++++-------
 1 file changed, 28 insertions(+), 7 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index d896dee7b487..ff82ec1e5e4d 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -680,10 +680,12 @@ static void set_vclk(struct tridentfb_par *par, unsigned long freq)
 
 	d = 20000;
 	for (k = 1; k >= 0; k--)
-		for (m = 0; m < 32; m++)
-			for (n = 0; n < 122; n++) {
+		for (m = 0; m < 32; m++) {
+			n = 2 * (m + 2) - 8;
+			for (n = (n < 0 ? 0 : n); n < 122; n++) {
 				fi = ((14318l * (n + 8)) / (m + 2)) >> k;
-				if ((di = abs(fi - freq)) < d) {
+				di = abs(fi - freq);
+				if (di <= d) {
 					d = di;
 					best_n = n;
 					best_m = m;
@@ -692,6 +694,7 @@ static void set_vclk(struct tridentfb_par *par, unsigned long freq)
 				if (fi > freq)
 					break;
 			}
+		}
 
 	if (is_oldclock(par->chip_id)) {
 		lo = best_n | (best_m << 7);
@@ -977,8 +980,8 @@ static int tridentfb_set_par(struct fb_info *info)
 
 	debug("enter\n");
 	hdispend = var->xres / 8 - 1;
-	hsyncstart = (var->xres + var->right_margin) / 8 - 1;
-	hsyncend = (var->xres + var->right_margin + var->hsync_len) / 8 - 1;
+	hsyncstart = (var->xres + var->right_margin) / 8;
+	hsyncend = (var->xres + var->right_margin + var->hsync_len) / 8;
 	htotal = (var->xres + var->left_margin + var->right_margin +
 		  var->hsync_len) / 8 - 5;
 	hblankstart = hdispend + 1;
@@ -991,8 +994,22 @@ static int tridentfb_set_par(struct fb_info *info)
 	vblankstart = vdispend + 1;
 	vblankend = vtotal;
 
+	if (info->var.vmode & FB_VMODE_INTERLACED) {
+		vtotal /= 2;
+		vdispend /= 2;
+		vsyncstart /= 2;
+		vsyncend /= 2;
+		vblankstart /= 2;
+		vblankend /= 2;
+	}
+
 	crtc_unlock(par);
 	write3CE(par, CyberControl, 8);
+	tmp = 0xEB;
+	if (var->sync & FB_SYNC_HOR_HIGH_ACT)
+		tmp &= ~0x40;
+	if (var->sync & FB_SYNC_VERT_HIGH_ACT)
+		tmp &= ~0x80;
 
 	if (par->flatpanel && var->xres < nativex) {
 		/*
@@ -1000,7 +1017,7 @@ static int tridentfb_set_par(struct fb_info *info)
 		 * than requested resolution decide whether
 		 * we stretch or center
 		 */
-		t_outb(par, 0xEB, VGA_MIS_W);
+		t_outb(par, tmp | 0xC0, VGA_MIS_W);
 
 		shadowmode_on(par);
 
@@ -1010,7 +1027,7 @@ static int tridentfb_set_par(struct fb_info *info)
 			screen_stretch(par);
 
 	} else {
-		t_outb(par, 0x2B, VGA_MIS_W);
+		t_outb(par, tmp, VGA_MIS_W);
 		write3CE(par, CyberControl, 8);
 	}
 
@@ -1071,6 +1088,10 @@ static int tridentfb_set_par(struct fb_info *info)
 	tmp = (info->var.vmode & FB_VMODE_INTERLACED) ? 0x84 : 0x80;
 	/* enable access extended memory */
 	write3X4(par, CRTCModuleTest, tmp);
+	tmp = read3CE(par, MiscIntContReg) & ~0x4;
+	if (info->var.vmode & FB_VMODE_INTERLACED)
+		tmp |= 0x4;
+	write3CE(par, MiscIntContReg, tmp);
 
 	/* enable GE for text acceleration */
 	write3X4(par, GraphEngReg, 0x80);

From 01a2d9ed85c945fc8a672622780533a1a0b7caf5 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:04 -0700
Subject: [PATCH 0465/1171] tridentfb: acceleration constants change

This patch replaces deprecated constant FB_ACCELF_TEXT with
FBINFO_HWACCEL_DISABLED and adds constants for Trident families of
accelerators.

The FBINFO_HWACCEL_DISABLED is correctly used so noaccel parameter works
now.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 45 +++++++++++++++++++++++++++------------
 include/linux/fb.h        |  4 ++++
 2 files changed, 35 insertions(+), 14 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index ff82ec1e5e4d..279411523ed9 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -499,6 +499,10 @@ static void tridentfb_fillrect(struct fb_info *info,
 	struct tridentfb_par *par = info->par;
 	int col;
 
+	if (info->flags & FBINFO_HWACCEL_DISABLED) {
+		cfb_fillrect(info, fr);
+		return;
+	}
 	if (info->var.bits_per_pixel == 8) {
 		col = fr->color;
 		col |= col << 8;
@@ -516,6 +520,10 @@ static void tridentfb_copyarea(struct fb_info *info,
 {
 	struct tridentfb_par *par = info->par;
 
+	if (info->flags & FBINFO_HWACCEL_DISABLED) {
+		cfb_copyarea(info, ca);
+		return;
+	}
 	par->wait_engine(par);
 	par->copy_rect(par, ca->sx, ca->sy, ca->dx, ca->dy,
 		       ca->width, ca->height);
@@ -525,7 +533,8 @@ static int tridentfb_sync(struct fb_info *info)
 {
 	struct tridentfb_par *par = info->par;
 
-	par->wait_engine(par);
+	if (!(info->flags & FBINFO_HWACCEL_DISABLED))
+		par->wait_engine(par);
 	return 0;
 }
 #else
@@ -855,8 +864,9 @@ static int tridentfb_check_var(struct fb_var_screeninfo *var,
 	if (var->yres_virtual > 0xffff)
 		return -EINVAL;
 	line_length = var->xres_virtual * bpp / 8;
-#ifdef CONFIG_FB_TRIDENT_ACCEL
-	if (!is3Dchip(par->chip_id)) {
+
+	if (!is3Dchip(par->chip_id) &&
+	    !(info->flags & FBINFO_HWACCEL_DISABLED)) {
 		/* acceleration requires line length to be power of 2 */
 		if (line_length <= 512)
 			var->xres_virtual = 512 * 8 / bpp;
@@ -873,7 +883,7 @@ static int tridentfb_check_var(struct fb_var_screeninfo *var,
 
 		line_length = var->xres_virtual * bpp / 8;
 	}
-#endif
+
 	if (var->yres > var->yres_virtual)
 		var->yres_virtual = var->yres;
 	if (line_length * var->yres_virtual > info->fix.smem_len)
@@ -1190,9 +1200,9 @@ static int tridentfb_set_par(struct fb_info *info)
 		set_number_of_lines(par, info->var.yres);
 	info->fix.line_length = info->var.xres_virtual * bpp / 8;
 	set_lwidth(par, info->fix.line_length / 8);
-#ifdef CONFIG_FB_TRIDENT_ACCEL
-	par->init_accel(par, info->var.xres_virtual, bpp);
-#endif
+
+	if (!(info->flags & FBINFO_HWACCEL_DISABLED))
+		par->init_accel(par, info->var.xres_virtual, bpp);
 
 	info->fix.visual = (bpp == 8) ? FB_VISUAL_PSEUDOCOLOR : FB_VISUAL_TRUECOLOR;
 	info->cmap.len = (bpp == 8) ? 256 : 16;
@@ -1326,6 +1336,9 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 		output("*** Please do use cyblafb, Cyberblade/i1 support "
 		       "will soon be removed from tridentfb!\n");
 
+#ifndef CONFIG_FB_TRIDENT_ACCEL
+	noaccel = 1;
+#endif
 
 	/* If PCI id is 0x9660 then further detect chip type */
 
@@ -1370,21 +1383,25 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 		default_par->wait_engine = xp_wait_engine;
 		default_par->fill_rect = xp_fill_rect;
 		default_par->copy_rect = xp_copy_rect;
+		tridentfb_fix.accel = FB_ACCEL_TRIDENT_BLADEXP;
 	} else if (is_blade(chip_id)) {
 		default_par->init_accel = blade_init_accel;
 		default_par->wait_engine = blade_wait_engine;
 		default_par->fill_rect = blade_fill_rect;
 		default_par->copy_rect = blade_copy_rect;
+		tridentfb_fix.accel = FB_ACCEL_TRIDENT_BLADE3D;
 	} else if (chip3D) {			/* 3DImage family left */
 		default_par->init_accel = image_init_accel;
 		default_par->wait_engine = image_wait_engine;
 		default_par->fill_rect = image_fill_rect;
 		default_par->copy_rect = image_copy_rect;
+		tridentfb_fix.accel = FB_ACCEL_TRIDENT_3DIMAGE;
 	} else { 				/* TGUI 9440/96XX family */
 		default_par->init_accel = tgui_init_accel;
 		default_par->wait_engine = xp_wait_engine;
 		default_par->fill_rect = tgui_fill_rect;
 		default_par->copy_rect = tgui_copy_rect;
+		tridentfb_fix.accel = FB_ACCEL_TRIDENT_TGUI;
 	}
 
 	default_par->chip_id = chip_id;
@@ -1441,9 +1458,13 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 	info->pseudo_palette = default_par->pseudo_pal;
 
 	info->flags = FBINFO_DEFAULT | FBINFO_HWACCEL_YPAN;
-#ifdef CONFIG_FB_TRIDENT_ACCEL
-	info->flags |= FBINFO_HWACCEL_COPYAREA | FBINFO_HWACCEL_FILLRECT;
-#endif
+	if (!noaccel && default_par->init_accel) {
+		info->flags &= ~FBINFO_HWACCEL_DISABLED;
+		info->flags |= FBINFO_HWACCEL_COPYAREA;
+		info->flags |= FBINFO_HWACCEL_FILLRECT;
+	} else
+		info->flags |= FBINFO_HWACCEL_DISABLED;
+
 	if (!fb_find_mode(&info->var, info,
 			  mode_option, NULL, 0, NULL, bpp)) {
 		err = -EINVAL;
@@ -1453,10 +1474,6 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 	if (err < 0)
 		goto out_unmap2;
 
-	if (!noaccel && default_par->init_accel)
-		info->var.accel_flags |= FB_ACCELF_TEXT;
-	else
-		info->var.accel_flags &= ~FB_ACCELF_TEXT;
 	info->var.activate |= FB_ACTIVATE_NOW;
 	info->device = &dev->dev;
 	if (register_framebuffer(info) < 0) {
diff --git a/include/linux/fb.h b/include/linux/fb.h
index 72295b099228..a084d1335869 100644
--- a/include/linux/fb.h
+++ b/include/linux/fb.h
@@ -120,6 +120,10 @@ struct dentry;
 #define FB_ACCEL_XGI_VOLARI_V	47	/* XGI Volari V3XT, V5, V8      */
 #define FB_ACCEL_XGI_VOLARI_Z	48	/* XGI Volari Z7                */
 #define FB_ACCEL_OMAP1610	49	/* TI OMAP16xx                  */
+#define FB_ACCEL_TRIDENT_TGUI	50	/* Trident TGUI			*/
+#define FB_ACCEL_TRIDENT_3DIMAGE 51	/* Trident 3DImage		*/
+#define FB_ACCEL_TRIDENT_BLADE3D 52	/* Trident Blade3D		*/
+#define FB_ACCEL_TRIDENT_BLADEXP 53	/* Trident BladeXP		*/
 #define FB_ACCEL_NEOMAGIC_NM2070 90	/* NeoMagic NM2070              */
 #define FB_ACCEL_NEOMAGIC_NM2090 91	/* NeoMagic NM2090              */
 #define FB_ACCEL_NEOMAGIC_NM2093 92	/* NeoMagic NM2093              */

From 5cf138457af20b0ef79d8c249381927718ca1417 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:05 -0700
Subject: [PATCH 0466/1171] tridentfb: source code improvements

This patch contains general source code improvments:
 - more simple functions are inline
 - removes some meaningless output and the VERSION
   string as it is no use
 - eng_par is moved into the tridentfb_par
 - removed small section of code for CyberBladeXPAi1
   which is maybe right for only one resolution
   and refresh rate and is probably redundant now
 - other minor improvements

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 83 ++++++++++++++-------------------------
 include/video/trident.h   |  4 +-
 2 files changed, 32 insertions(+), 55 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index 279411523ed9..b6065effc5e0 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -24,8 +24,6 @@
 #include <video/vga.h>
 #include <video/trident.h>
 
-#define VERSION		"0.7.9-NEWAPI"
-
 struct tridentfb_par {
 	void __iomem *io_virt;	/* iospace virtual memory address */
 	u32 pseudo_pal[16];
@@ -37,9 +35,9 @@ struct tridentfb_par {
 		(struct tridentfb_par *par, u32, u32, u32, u32, u32, u32);
 	void (*copy_rect)
 		(struct tridentfb_par *par, u32, u32, u32, u32, u32, u32);
+	unsigned char eng_oper;	/* engine operation... */
 };
 
-static unsigned char eng_oper;	/* engine operation... */
 static struct fb_ops tridentfb_ops;
 
 static struct fb_fix_screeninfo tridentfb_fix = {
@@ -53,7 +51,7 @@ static struct fb_fix_screeninfo tridentfb_fix = {
 /* defaults which are normally overriden by user values */
 
 /* video mode */
-static char *mode_option __devinitdata = "640x480";
+static char *mode_option __devinitdata = "640x480-8@60";
 static int bpp __devinitdata = 8;
 
 static int noaccel __devinitdata;
@@ -84,24 +82,22 @@ MODULE_PARM_DESC(fp, "Define if flatpanel is connected");
 module_param(crt, int, 0);
 MODULE_PARM_DESC(crt, "Define if CRT is connected");
 
-static int is_oldclock(int id)
+static inline int is_oldclock(int id)
 {
 	return	(id == TGUI9440) ||
 		(id == TGUI9660) ||
 		(id == CYBER9320);
 }
 
-static int is_oldprotect(int id)
+static inline int is_oldprotect(int id)
 {
-	return	(id == TGUI9440) ||
-		(id == TGUI9660) ||
+	return	is_oldclock(id) ||
 		(id == PROVIDIA9685) ||
-		(id == CYBER9320) ||
 		(id == CYBER9382) ||
 		(id == CYBER9385);
 }
 
-static int is_blade(int id)
+static inline int is_blade(int id)
 {
 	return	(id == BLADE3D) ||
 		(id == CYBERBLADEE4) ||
@@ -113,27 +109,22 @@ static int is_blade(int id)
 		(id == CYBERBLADEAi1D);
 }
 
-static int is_xp(int id)
+static inline int is_xp(int id)
 {
 	return	(id == CYBERBLADEXPAi1) ||
 		(id == CYBERBLADEXPm8) ||
 		(id == CYBERBLADEXPm16);
 }
 
-static int is3Dchip(int id)
+static inline int is3Dchip(int id)
 {
-	return ((id == BLADE3D) || (id == CYBERBLADEE4) ||
-		(id == CYBERBLADEi7) || (id == CYBERBLADEi7D) ||
+	return	is_blade(id) || is_xp(id) ||
 		(id == CYBER9397) || (id == CYBER9397DVD) ||
 		(id == CYBER9520) || (id == CYBER9525DVD) ||
-		(id == IMAGE975) || (id == IMAGE985) ||
-		(id == CYBERBLADEi1) || (id == CYBERBLADEi1D) ||
-		(id == CYBERBLADEAi1) || (id == CYBERBLADEAi1D) ||
-		(id == CYBERBLADEXPm8) || (id == CYBERBLADEXPm16) ||
-		(id == CYBERBLADEXPAi1));
+		(id == IMAGE975) || (id == IMAGE985);
 }
 
-static int iscyber(int id)
+static inline int iscyber(int id)
 {
 	switch (id) {
 	case CYBER9388:
@@ -153,13 +144,7 @@ static int iscyber(int id)
 		return 1;
 
 	case CYBER9320:
-	case TGUI9660:
-	case PROVIDIA9685:
-	case IMAGE975:
-	case IMAGE985:
-	case BLADE3D:
 	case CYBERBLADEi7:	/* VIA MPV4 integrated version */
-
 	default:
 		/* case CYBERBLDAEXPm8:  Strange */
 		/* case CYBERBLDAEXPm16: Strange */
@@ -275,7 +260,7 @@ static void xp_init_accel(struct tridentfb_par *par, int pitch, int bpp)
 
 	t_outb(par, x, 0x2125);
 
-	eng_oper = x | 0x40;
+	par->eng_oper = x | 0x40;
 
 	writemmr(par, 0x2154, v1);
 	writemmr(par, 0x2150, v1);
@@ -284,10 +269,9 @@ static void xp_init_accel(struct tridentfb_par *par, int pitch, int bpp)
 
 static void xp_wait_engine(struct tridentfb_par *par)
 {
-	int count, timeout;
+	int count = 0;
+	int timeout = 0;
 
-	count = 0;
-	timeout = 0;
 	while (t_inb(par, STATUS) & 0x80) {
 		count++;
 		if (count == 10000000) {
@@ -313,16 +297,14 @@ static void xp_fill_rect(struct tridentfb_par *par,
 	writemmr(par, OLDDIM, point(h, w));
 	writemmr(par, OLDDST, point(y, x));
 	t_outb(par, 0x01, OLDCMD);
-	t_outb(par, eng_oper, 0x2125);
+	t_outb(par, par->eng_oper, 0x2125);
 }
 
 static void xp_copy_rect(struct tridentfb_par *par,
 			 u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h)
 {
-	int direction;
 	u32 x1_tmp, x2_tmp, y1_tmp, y2_tmp;
-
-	direction = 0x0004;
+	int direction = 0x0004;
 
 	if ((x1 < x2) && (y1 == y2)) {
 		direction |= 0x0200;
@@ -602,7 +584,7 @@ static void disable_mmio(struct tridentfb_par *par)
 	t_outb(par, t_inb(par, 0x3D5) & ~0x01, 0x3D5);
 }
 
-static void crtc_unlock(struct tridentfb_par *par)
+static inline void crtc_unlock(struct tridentfb_par *par)
 {
 	write3X4(par, VGA_CRTC_V_SYNC_END,
 		 read3X4(par, VGA_CRTC_V_SYNC_END) & 0x7F);
@@ -642,7 +624,7 @@ static int __devinit get_nativex(struct tridentfb_par *par)
 }
 
 /* Set pitch */
-static void set_lwidth(struct tridentfb_par *par, int width)
+static inline void set_lwidth(struct tridentfb_par *par, int width)
 {
 	write3X4(par, VGA_CRTC_OFFSET, width & 0xFF);
 	write3X4(par, AddColReg,
@@ -661,7 +643,7 @@ static void screen_stretch(struct tridentfb_par *par)
 }
 
 /* For resolutions smaller than FP resolution center */
-static void screen_center(struct tridentfb_par *par)
+static inline void screen_center(struct tridentfb_par *par)
 {
 	write3CE(par, VertStretch, (read3CE(par, VertStretch) & 0x7C) | 0x80);
 	write3CE(par, HorStretch, (read3CE(par, HorStretch) & 0x7C) | 0x80);
@@ -967,12 +949,12 @@ static int tridentfb_pan_display(struct fb_var_screeninfo *var,
 	return 0;
 }
 
-static void shadowmode_on(struct tridentfb_par *par)
+static inline void shadowmode_on(struct tridentfb_par *par)
 {
 	write3CE(par, CyberControl, read3CE(par, CyberControl) | 0x81);
 }
 
-static void shadowmode_off(struct tridentfb_par *par)
+static inline void shadowmode_off(struct tridentfb_par *par)
 {
 	write3CE(par, CyberControl, read3CE(par, CyberControl) & 0x7E);
 }
@@ -980,7 +962,7 @@ static void shadowmode_off(struct tridentfb_par *par)
 /* Set the hardware to the requested video mode */
 static int tridentfb_set_par(struct fb_info *info)
 {
-	struct tridentfb_par *par = (struct tridentfb_par *)(info->par);
+	struct tridentfb_par *par = info->par;
 	u32 htotal, hdispend, hsyncstart, hsyncend, hblankstart, hblankend;
 	u32 vtotal, vdispend, vsyncstart, vsyncend, vblankstart, vblankend;
 	struct fb_var_screeninfo *var = &info->var;
@@ -1159,11 +1141,6 @@ static int tridentfb_set_par(struct fb_info *info)
 	write3CE(par, 0x6, 0x05);	/* graphics mode */
 	write3CE(par, 0x7, 0x0F);	/* planes? */
 
-	if (par->chip_id == CYBERBLADEXPAi1) {
-		/* This fixes snow-effect in 32 bpp */
-		write3X4(par, VGA_CRTC_H_SYNC_START, 0x84);
-	}
-
 	/* graphics mode and support 256 color modes */
 	writeAttr(par, 0x10, 0x41);
 	writeAttr(par, 0x12, 0x0F);	/* planes */
@@ -1238,7 +1215,7 @@ static int tridentfb_setcolreg(unsigned regno, unsigned red, unsigned green,
 			col |= col << 16;
 			((u32 *)(info->pseudo_palette))[regno] = col;
 		} else if (bpp == 32)		/* ARGB 8888 */
-			((u32*)info->pseudo_palette)[regno] =
+			((u32 *)info->pseudo_palette)[regno] =
 				((transp & 0xFF00) << 16)	|
 				((red & 0xFF00) << 8)		|
 				((green & 0xFF00))		|
@@ -1249,7 +1226,7 @@ static int tridentfb_setcolreg(unsigned regno, unsigned red, unsigned green,
 	return 0;
 }
 
-/* Try blanking the screen.For flat panels it does nothing */
+/* Try blanking the screen. For flat panels it does nothing */
 static int tridentfb_blank(int blank_mode, struct fb_info *info)
 {
 	unsigned char PMCont, DPMSCont;
@@ -1408,9 +1385,10 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 
 	/* setup MMIO region */
 	tridentfb_fix.mmio_start = pci_resource_start(dev, 1);
-	tridentfb_fix.mmio_len = chip3D ? 0x20000 : 0x10000;
+	tridentfb_fix.mmio_len = pci_resource_len(dev, 1);
 
-	if (!request_mem_region(tridentfb_fix.mmio_start, tridentfb_fix.mmio_len, "tridentfb")) {
+	if (!request_mem_region(tridentfb_fix.mmio_start,
+				tridentfb_fix.mmio_len, "tridentfb")) {
 		debug("request_region failed!\n");
 		framebuffer_release(info);
 		return -1;
@@ -1431,7 +1409,8 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 	tridentfb_fix.smem_start = pci_resource_start(dev, 0);
 	tridentfb_fix.smem_len = get_memsize(default_par);
 
-	if (!request_mem_region(tridentfb_fix.smem_start, tridentfb_fix.smem_len, "tridentfb")) {
+	if (!request_mem_region(tridentfb_fix.smem_start,
+				tridentfb_fix.smem_len, "tridentfb")) {
 		debug("request_mem_region failed!\n");
 		disable_mmio(info->par);
 		err = -1;
@@ -1447,7 +1426,6 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 		goto out_unmap2;
 	}
 
-	output("%s board found\n", pci_name(dev));
 	default_par->flatpanel = is_flatpanel(default_par);
 
 	if (default_par->flatpanel)
@@ -1477,7 +1455,7 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 	info->var.activate |= FB_ACTIVATE_NOW;
 	info->device = &dev->dev;
 	if (register_framebuffer(info) < 0) {
-		printk(KERN_ERR "tridentfb: could not register Trident framebuffer\n");
+		printk(KERN_ERR "tridentfb: could not register framebuffer\n");
 		fb_dealloc_cmap(&info->cmap);
 		err = -EINVAL;
 		goto out_unmap2;
@@ -1599,7 +1577,6 @@ static int __init tridentfb_init(void)
 		return -ENODEV;
 	tridentfb_setup(option);
 #endif
-	output("Trident framebuffer %s initializing\n", VERSION);
 	return pci_register_driver(&tridentfb_pci_driver);
 }
 
diff --git a/include/video/trident.h b/include/video/trident.h
index 7540501bc040..85ced9d91312 100644
--- a/include/video/trident.h
+++ b/include/video/trident.h
@@ -4,9 +4,9 @@
 #endif
 
 #if TRIDENTFB_DEBUG
-#define debug(f,a...)	printk("%s:" f,  __FUNCTION__ , ## a);mdelay(100);
+#define debug(f, a...)	printk("%s:" f,  __func__ , ## a);
 #else
-#define debug(f,a...)
+#define debug(f, a...)
 #endif
 
 #define output(f, a...) pr_info("tridentfb: " f, ## a)

From 13b0de49f52ec8638b3e3e59192a959b35214d9e Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:06 -0700
Subject: [PATCH 0467/1171] tridentfb: fix console freeze when switching from
 X11

This patch fixes two problems when acceleration is enabled:

 - console switch from the Xorg locks up the computer
   because the Xorg code locks some registers and disables
   the mmio mode, so reenable these in the
   tridentfb_set_par() and enable_mmio()

 - blacklist the Image975 chipset from setting PCI burst
   mode. This helps with random lock ups of the
   framebuffer on this chip. The same fix is probably
   needed for the Xorg as well.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index b6065effc5e0..da4b464cbdba 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -558,13 +558,15 @@ static inline void write3CE(struct tridentfb_par *par, int reg,
 	vga_mm_wgfx(par->io_virt, reg, val);
 }
 
-static void enable_mmio(void)
+static void enable_mmio(struct tridentfb_par *par)
 {
 	/* Goto New Mode */
 	vga_io_rseq(0x0B);
 
 	/* Unprotect registers */
 	vga_io_wseq(NewMode1, 0x80);
+	if (!is_oldprotect(par->chip_id))
+		vga_io_wseq(Protection, 0x92);
 
 	/* Enable MMIO */
 	outb(PCIReg, 0x3D4);
@@ -578,6 +580,8 @@ static void disable_mmio(struct tridentfb_par *par)
 
 	/* Unprotect registers */
 	vga_mm_wseq(par->io_virt, NewMode1, 0x80);
+	if (!is_oldprotect(par->chip_id))
+		vga_mm_wseq(par->io_virt, Protection, 0x92);
 
 	/* Disable MMIO */
 	t_outb(par, PCIReg, 0x3D4);
@@ -995,6 +999,7 @@ static int tridentfb_set_par(struct fb_info *info)
 		vblankend /= 2;
 	}
 
+	enable_mmio(par);
 	crtc_unlock(par);
 	write3CE(par, CyberControl, 8);
 	tmp = 0xEB;
@@ -1116,7 +1121,7 @@ static int tridentfb_set_par(struct fb_info *info)
 	if (!is_xp(par->chip_id))
 		write3X4(par, Performance, read3X4(par, Performance) | 0x10);
 	/* MMIO & PCI read and write burst enable */
-	if (par->chip_id != TGUI9440)
+	if (par->chip_id != TGUI9440 && par->chip_id != IMAGE975)
 		write3X4(par, PCIReg, read3X4(par, PCIReg) | 0x06);
 
 	vga_mm_wseq(par->io_virt, 0, 3);
@@ -1403,7 +1408,7 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 		goto out_unmap1;
 	}
 
-	enable_mmio();
+	enable_mmio(default_par);
 
 	/* setup framebuffer memory */
 	tridentfb_fix.smem_start = pci_resource_start(dev, 0);

From a4af1798d768ab2f12ab623e21ad68dc8c248005 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:06 -0700
Subject: [PATCH 0468/1171] tridentfb: fix 224 color logo at 8 bpp

Fix depth setting for 8 bpp mode.  The nice 224 color logo is not
displayed in 8 bpp depth without this fix.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index da4b464cbdba..1c3f0ba4b33a 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -878,11 +878,9 @@ static int tridentfb_check_var(struct fb_var_screeninfo *var,
 	switch (bpp) {
 	case 8:
 		var->red.offset = 0;
-		var->green.offset = 0;
-		var->blue.offset = 0;
-		var->red.length = 6;
-		var->green.length = 6;
-		var->blue.length = 6;
+		var->red.length = 8;
+		var->green = var->red;
+		var->blue = var->red;
 		break;
 	case 16:
 		var->red.offset = 11;

From f330c4b1961d730ef15ac184e4b7f1c25847d0ae Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:07 -0700
Subject: [PATCH 0469/1171] tridentfb: y-panning fixes

The Trident cards uses only 20-bit address of screen start in double
words.  This allows addressing for only 4MB of video memory so check this.

Also remove some redundant checks and assignments.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index 1c3f0ba4b33a..138140bdb495 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -870,8 +870,10 @@ static int tridentfb_check_var(struct fb_var_screeninfo *var,
 		line_length = var->xres_virtual * bpp / 8;
 	}
 
-	if (var->yres > var->yres_virtual)
-		var->yres_virtual = var->yres;
+	/* datasheet specifies how to set panning only up to 4 MB */
+	if (line_length * (var->yres_virtual - var->yres) > (4 << 20))
+		var->yres_virtual = ((4 << 20) / line_length) + var->yres;
+
 	if (line_length * var->yres_virtual > info->fix.smem_len)
 		return -EINVAL;
 
@@ -944,8 +946,6 @@ static int tridentfb_pan_display(struct fb_var_screeninfo *var,
 	debug("enter\n");
 	offset = (var->xoffset + (var->yoffset * var->xres_virtual))
 		* var->bits_per_pixel / 32;
-	info->var.xoffset = var->xoffset;
-	info->var.yoffset = var->yoffset;
 	set_screen_start(par, offset);
 	debug("exit\n");
 	return 0;
@@ -1225,7 +1225,6 @@ static int tridentfb_setcolreg(unsigned regno, unsigned red, unsigned green,
 				((blue & 0xFF00) >> 8);
 	}
 
-/* 	debug("exit\n"); */
 	return 0;
 }
 

From 6280fd4f9c2683a4d2f096320dd74ded4e5106ad Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:08 -0700
Subject: [PATCH 0470/1171] tridentfb: Blade3D clock fixes

This patch fixes following problems:
- does not allow the m parameter to reach 0 as
  it locks the graphics core (power cycle needed)
- for the newer chips (with new clock registers)
  does not allow of n / m ratio below 4 as it gives
  unstable image on the Blade3D core
- extend shift parameter (k) range  to 2 for the newer
  chips to cope with the n /m >= 4 limit at low resolution
  (bandwidth) modes
- prefer modes with higher n / m ratio (higher k values)

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index 138140bdb495..854e2e5af582 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -672,15 +672,16 @@ static void set_vclk(struct tridentfb_par *par, unsigned long freq)
 	unsigned long fi, d, di;
 	unsigned char best_m = 0, best_n = 0, best_k = 0;
 	unsigned char hi, lo;
+	unsigned char shift = !is_oldclock(par->chip_id) ? 2 : 1;
 
 	d = 20000;
-	for (k = 1; k >= 0; k--)
-		for (m = 0; m < 32; m++) {
-			n = 2 * (m + 2) - 8;
+	for (k = shift; k >= 0; k--)
+		for (m = 1; m < 32; m++) {
+			n = ((m + 2) << shift) - 8;
 			for (n = (n < 0 ? 0 : n); n < 122; n++) {
 				fi = ((14318l * (n + 8)) / (m + 2)) >> k;
 				di = abs(fi - freq);
-				if (di <= d) {
+				if (di < d || (di == d && k == best_k)) {
 					d = di;
 					best_n = n;
 					best_m = m;

From 0292be4a382957016e8b574dc292779cfb49e029 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:08 -0700
Subject: [PATCH 0471/1171] tridentfb: add imageblit acceleration for Blade3D
 family

Add imageblit acceleration for the Blade3D family of cores.  The code is
based on code from the cyblafb driver.

It is a step toward assimilating back the cyblafb driver into the
tridentfb driver.  The cyblafb driver handles a subfamily of the Trident
Blade3d cores.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 79 +++++++++++++++++++++++++++++++++++++--
 include/video/trident.h   |  1 +
 2 files changed, 77 insertions(+), 3 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index 854e2e5af582..b21f84239022 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -35,11 +35,12 @@ struct tridentfb_par {
 		(struct tridentfb_par *par, u32, u32, u32, u32, u32, u32);
 	void (*copy_rect)
 		(struct tridentfb_par *par, u32, u32, u32, u32, u32, u32);
+	void (*image_blit)
+		(struct tridentfb_par *par, const char*,
+		 u32, u32, u32, u32, u32, u32);
 	unsigned char eng_oper;	/* engine operation... */
 };
 
-static struct fb_ops tridentfb_ops;
-
 static struct fb_fix_screeninfo tridentfb_fix = {
 	.id = "Trident",
 	.type = FB_TYPE_PACKED_PIXELS,
@@ -212,6 +213,21 @@ static void blade_fill_rect(struct tridentfb_par *par,
 	writemmr(par, DST2, point(x + w - 1, y + h - 1));
 }
 
+static void blade_image_blit(struct tridentfb_par *par, const char *data,
+			     u32 x, u32 y, u32 w, u32 h, u32 c, u32 b)
+{
+	unsigned size = ((w + 31) >> 5) * h;
+
+	writemmr(par, COLOR, c);
+	writemmr(par, BGCOLOR, b);
+	writemmr(par, CMD, 0xa0000000 | 3 << 19);
+
+	writemmr(par, DST1, point(x, y));
+	writemmr(par, DST2, point(x + w - 1, y + h - 1));
+
+	memcpy(par->io_virt + 0x10000, data, 4 * size);
+}
+
 static void blade_copy_rect(struct tridentfb_par *par,
 			    u32 x1, u32 y1, u32 x2, u32 y2, u32 w, u32 h)
 {
@@ -497,6 +513,36 @@ static void tridentfb_fillrect(struct fb_info *info,
 		       fr->height, col, fr->rop);
 }
 
+static void tridentfb_imageblit(struct fb_info *info,
+				const struct fb_image *img)
+{
+	struct tridentfb_par *par = info->par;
+	int col, bgcol;
+
+	if ((info->flags & FBINFO_HWACCEL_DISABLED) || img->depth != 1) {
+		cfb_imageblit(info, img);
+		return;
+	}
+	if (info->var.bits_per_pixel == 8) {
+		col = img->fg_color;
+		col |= col << 8;
+		col |= col << 16;
+		bgcol = img->bg_color;
+		bgcol |= bgcol << 8;
+		bgcol |= bgcol << 16;
+	} else {
+		col = ((u32 *)(info->pseudo_palette))[img->fg_color];
+		bgcol = ((u32 *)(info->pseudo_palette))[img->bg_color];
+	}
+
+	par->wait_engine(par);
+	if (par->image_blit)
+		par->image_blit(par, img->data, img->dx, img->dy,
+				img->width, img->height, col, bgcol);
+	else
+		cfb_imageblit(info, img);
+}
+
 static void tridentfb_copyarea(struct fb_info *info,
 			       const struct fb_copyarea *ca)
 {
@@ -522,6 +568,7 @@ static int tridentfb_sync(struct fb_info *info)
 #else
 #define tridentfb_fillrect cfb_fillrect
 #define tridentfb_copyarea cfb_copyarea
+#define tridentfb_imageblit cfb_imageblit
 #endif /* CONFIG_FB_TRIDENT_ACCEL */
 
 /*
@@ -1285,7 +1332,7 @@ static struct fb_ops tridentfb_ops = {
 	.fb_set_par = tridentfb_set_par,
 	.fb_fillrect = tridentfb_fillrect,
 	.fb_copyarea = tridentfb_copyarea,
-	.fb_imageblit = cfb_imageblit,
+	.fb_imageblit = tridentfb_imageblit,
 #ifdef CONFIG_FB_TRIDENT_ACCEL
 	.fb_sync = tridentfb_sync,
 #endif
@@ -1369,6 +1416,7 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 		default_par->wait_engine = blade_wait_engine;
 		default_par->fill_rect = blade_fill_rect;
 		default_par->copy_rect = blade_copy_rect;
+		default_par->image_blit = blade_image_blit;
 		tridentfb_fix.accel = FB_ACCEL_TRIDENT_BLADE3D;
 	} else if (chip3D) {			/* 3DImage family left */
 		default_par->init_accel = image_init_accel;
@@ -1446,6 +1494,29 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 	} else
 		info->flags |= FBINFO_HWACCEL_DISABLED;
 
+	info->pixmap.addr = kmalloc(4096, GFP_KERNEL);
+	if (!info->pixmap.addr) {
+		err = -ENOMEM;
+		goto out_unmap2;
+	}
+
+	info->pixmap.size = 4096;
+	info->pixmap.buf_align = 4;
+	info->pixmap.scan_align = 1;
+	info->pixmap.access_align = 32;
+	info->pixmap.flags = FB_PIXMAP_SYSTEM;
+
+	if (default_par->image_blit) {
+		info->flags |= FBINFO_HWACCEL_IMAGEBLIT;
+		info->pixmap.scan_align = 4;
+	}
+
+	if (noaccel) {
+		printk(KERN_DEBUG "disabling acceleration\n");
+		info->flags |= FBINFO_HWACCEL_DISABLED;
+		info->pixmap.scan_align = 1;
+	}
+
 	if (!fb_find_mode(&info->var, info,
 			  mode_option, NULL, 0, NULL, bpp)) {
 		err = -EINVAL;
@@ -1471,6 +1542,7 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 	return 0;
 
 out_unmap2:
+	kfree(info->pixmap.addr);
 	if (info->screen_base)
 		iounmap(info->screen_base);
 	release_mem_region(tridentfb_fix.smem_start, tridentfb_fix.smem_len);
@@ -1494,6 +1566,7 @@ static void __devexit trident_pci_remove(struct pci_dev *dev)
 	release_mem_region(tridentfb_fix.smem_start, tridentfb_fix.smem_len);
 	release_mem_region(tridentfb_fix.mmio_start, tridentfb_fix.mmio_len);
 	pci_set_drvdata(dev, NULL);
+	kfree(info->pixmap.addr);
 	framebuffer_release(info);
 }
 
diff --git a/include/video/trident.h b/include/video/trident.h
index 85ced9d91312..b6ce19d1b61b 100644
--- a/include/video/trident.h
+++ b/include/video/trident.h
@@ -135,6 +135,7 @@
 #define CMD	0x2144
 #define ROP	0x2148
 #define COLOR	0x2160
+#define BGCOLOR	0x2164
 #define SRC1	0x2100
 #define SRC2	0x2104
 #define DST1	0x2108

From 663b0e15877293451bdfea619db45eafae9dec54 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:09 -0700
Subject: [PATCH 0472/1171] tridentfb: remove warning message that cyblafb
 driver should be used

The tridentfb driver should handle now all chipsets handled by the cyblafb
driver.  Remove the message which claims that support will be removed.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tridentfb.c | 4 ----
 1 file changed, 4 deletions(-)

diff --git a/drivers/video/tridentfb.c b/drivers/video/tridentfb.c
index b21f84239022..479b2e79ad68 100644
--- a/drivers/video/tridentfb.c
+++ b/drivers/video/tridentfb.c
@@ -1359,10 +1359,6 @@ static int __devinit trident_pci_probe(struct pci_dev *dev,
 
 	chip_id = id->device;
 
-	if (chip_id == CYBERBLADEi1)
-		output("*** Please do use cyblafb, Cyberblade/i1 support "
-		       "will soon be removed from tridentfb!\n");
-
 #ifndef CONFIG_FB_TRIDENT_ACCEL
 	noaccel = 1;
 #endif

From 0b9cf3aa6b1e934807b40b4d478d7e11f7c43f55 Mon Sep 17 00:00:00 2001
From: Roland Kletzing <devzero@web.de>
Date: Wed, 23 Jul 2008 21:31:10 -0700
Subject: [PATCH 0473/1171] mdacon messing up default vc's - set default to
 vc13-16 again

mdacon incorrectly detects MDA hardware on systems without such graphics card.

One may load this module by chance, for example when doing some systematical
module-testing, and if there is no Monochrome Display Adapter attached ,
module init renders vc1-16 completely unusable.

I and others have run into this more than once.  see [Bug 224522 - modprobe
mdacon freezes machine -> https://bugzilla.novell.com/show_bug.cgi?id=224522 ]
for example

Apparently  proper MDA detection seems to be broken for a long time - seems to
be related to those #ifdef TEST_MDA_B statements added by Edward Betts.

this commit back in 2002 made things even worse :
http://git.kernel.org/?p=linux/kernel/git/tglx/history.git;a=commit;h=c72757b49c88914433244757fb4967fc63546685

It changed default vc allocation from 13-16 to 1-16 for no apparent reason
(!?) , and with that (and without X), mdacon grabs the vc you`re currently
sitting on and locks you out.

this is from Kconfig :
>config MDA_CONSOLE
>        depends on !M68K && !PARISC && ISA
>        tristate "MDA text console (dual-headed) (EXPERIMENTAL)"
>        ---help---
>          Say Y here if you have an old MDA or monochrome Hercules graphics
>          adapter in your system acting as a second head ( = video card). You
>          will then be able to use two monitors with your Linux system. Do not
>          say Y here if your MDA card is the primary card in your system; the
>          normal VGA driver will handle it.

As we can see mdacon is just meant as an additional driver for dual-head
setup, and since kernel 2.4.36 still defaults to vc13-16 , setting the default
back to that value again shouldn`t do any harm.

Hereby i'm reverting that change, setting default back to to vc13-16 again.

Besides the fact that mdacon may be rarely or never be used these days and
could perhaps put to trash anyway (pre-dinosaur hardware!), indeed this is not
a real solution, but at least it removes the unfortunate side-effect of
messing up the vc you`re working on.

Signed-off-by: Roland Kletzing <devzero@web.de>
Cc: James Simmons <jsimmons@infradead.org>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Cc: Tim Schmielau <tim@physik3.uni-rostock.de>
Cc: Jan Engelhardt <jengelh@gmx.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/console/mdacon.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/video/console/mdacon.c b/drivers/video/console/mdacon.c
index 38a296bbdfc9..9901064199bd 100644
--- a/drivers/video/console/mdacon.c
+++ b/drivers/video/console/mdacon.c
@@ -71,13 +71,15 @@ static char *mda_type_name;
 
 /* console information */
 
-static int	mda_first_vc = 1;
+static int	mda_first_vc = 13;
 static int	mda_last_vc  = 16;
 
 static struct vc_data	*mda_display_fg = NULL;
 
 module_param(mda_first_vc, int, 0);
+MODULE_PARM_DESC(mda_first_vc, "First virtual console. Default: 13");
 module_param(mda_last_vc, int, 0);
+MODULE_PARM_DESC(mda_last_vc, "Last virtual console. Default: 16");
 
 /* MDA register values
  */

From 14aefd1b49ff3bd13caa37fb06bd53488d5d1486 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Wed, 23 Jul 2008 21:31:12 -0700
Subject: [PATCH 0474/1171] video/sis/: remove compat code

This patch removes compat code for older kernel versions.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Cc: <thomas@winischhofer.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/sis/init.h       |  1 -
 drivers/video/sis/init301.h    |  1 -
 drivers/video/sis/initextlfb.c |  1 -
 drivers/video/sis/osdef.h      |  1 -
 drivers/video/sis/sis.h        | 22 ++++-------------
 drivers/video/sis/sis_accel.c  |  1 -
 drivers/video/sis/sis_main.c   | 44 ++++++++--------------------------
 drivers/video/sis/sis_main.h   |  4 ++--
 drivers/video/sis/vgatypes.h   |  4 ----
 9 files changed, 17 insertions(+), 62 deletions(-)

diff --git a/drivers/video/sis/init.h b/drivers/video/sis/init.h
index f40a680df86f..b96005c39c67 100644
--- a/drivers/video/sis/init.h
+++ b/drivers/video/sis/init.h
@@ -73,7 +73,6 @@
 #ifdef SIS_CP
 #undef SIS_CP
 #endif
-#include <linux/version.h>
 #include <linux/types.h>
 #include <asm/io.h>
 #include <linux/fb.h>
diff --git a/drivers/video/sis/init301.h b/drivers/video/sis/init301.h
index 7708e1e1d99e..51d99222375d 100644
--- a/drivers/video/sis/init301.h
+++ b/drivers/video/sis/init301.h
@@ -67,7 +67,6 @@
 #ifdef SIS_CP
 #undef SIS_CP
 #endif
-#include <linux/version.h>
 #include <linux/types.h>
 #include <asm/io.h>
 #include <linux/fb.h>
diff --git a/drivers/video/sis/initextlfb.c b/drivers/video/sis/initextlfb.c
index 47a33501549d..99c04a4855d1 100644
--- a/drivers/video/sis/initextlfb.c
+++ b/drivers/video/sis/initextlfb.c
@@ -30,7 +30,6 @@
 #include "vgatypes.h"
 #include "vstruct.h"
 
-#include <linux/version.h>
 #include <linux/types.h>
 #include <linux/fb.h>
 
diff --git a/drivers/video/sis/osdef.h b/drivers/video/sis/osdef.h
index c1492782cb18..6ff8f988a1a7 100644
--- a/drivers/video/sis/osdef.h
+++ b/drivers/video/sis/osdef.h
@@ -87,7 +87,6 @@
 /**********************************************************************/
 
 #ifdef SIS_LINUX_KERNEL
-#include <linux/version.h>
 
 #ifdef CONFIG_FB_SIS_300
 #define SIS300
diff --git a/drivers/video/sis/sis.h b/drivers/video/sis/sis.h
index a14e82211037..7c5710e3fb56 100644
--- a/drivers/video/sis/sis.h
+++ b/drivers/video/sis/sis.h
@@ -24,8 +24,6 @@
 #ifndef _SIS_H_
 #define _SIS_H_
 
-#include <linux/version.h>
-
 #include "osdef.h"
 #include <video/sisfb.h>
 
@@ -42,16 +40,6 @@
 #define SIS_NEW_CONFIG_COMPAT
 #endif	/* CONFIG_COMPAT */
 
-#if LINUX_VERSION_CODE > KERNEL_VERSION(2,6,8)
-#define SIS_IOTYPE1 void __iomem
-#define SIS_IOTYPE2 __iomem
-#define SISINITSTATIC static
-#else
-#define SIS_IOTYPE1 unsigned char
-#define SIS_IOTYPE2
-#define SISINITSTATIC
-#endif
-
 #undef SISFBDEBUG
 
 #ifdef SISFBDEBUG
@@ -505,8 +493,8 @@ struct sis_video_info {
 
 	unsigned long	UMAsize, LFBsize;
 
-	SIS_IOTYPE1	*video_vbase;
-	SIS_IOTYPE1	*mmio_vbase;
+	void __iomem	*video_vbase;
+	void __iomem	*mmio_vbase;
 
 	unsigned char	*bios_abase;
 
@@ -533,8 +521,8 @@ struct sis_video_info {
 	int		sisfb_nocrt2rate;
 
 	u32		heapstart;		/* offset  */
-	SIS_IOTYPE1	*sisfb_heap_start;	/* address */
-	SIS_IOTYPE1	*sisfb_heap_end;	/* address */
+	void __iomem	*sisfb_heap_start;	/* address */
+	void __iomem	*sisfb_heap_end;	/* address */
 	u32		sisfb_heap_size;
 	int		havenoheap;
 
@@ -612,7 +600,7 @@ struct sis_video_info {
 	u8		detectedpdca;
 	u8		detectedlcda;
 
-	SIS_IOTYPE1	*hwcursor_vbase;
+	void __iomem	*hwcursor_vbase;
 
 	int		chronteltype;
 	int		tvxpos, tvypos;
diff --git a/drivers/video/sis/sis_accel.c b/drivers/video/sis/sis_accel.c
index 7addf91d2fea..ceb434c95c0d 100644
--- a/drivers/video/sis/sis_accel.c
+++ b/drivers/video/sis/sis_accel.c
@@ -28,7 +28,6 @@
  *			for more information and updates)
  */
 
-#include <linux/version.h>
 #include <linux/module.h>
 #include <linux/kernel.h>
 #include <linux/fb.h>
diff --git a/drivers/video/sis/sis_main.c b/drivers/video/sis/sis_main.c
index b9343844cd1f..346d6458cf76 100644
--- a/drivers/video/sis/sis_main.c
+++ b/drivers/video/sis/sis_main.c
@@ -33,7 +33,6 @@
  *
  */
 
-#include <linux/version.h>
 #include <linux/module.h>
 #include <linux/moduleparam.h>
 #include <linux/kernel.h>
@@ -41,13 +40,7 @@
 #include <linux/errno.h>
 #include <linux/string.h>
 #include <linux/mm.h>
-
-#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,17)
-#include <linux/tty.h>
-#else
 #include <linux/screen_info.h>
-#endif
-
 #include <linux/slab.h>
 #include <linux/fb.h>
 #include <linux/selection.h>
@@ -1167,11 +1160,7 @@ sisfb_set_mode(struct sis_video_info *ivideo, int clrscrn)
 	unsigned short modeno = ivideo->mode_no;
 
 	/* >=2.6.12's fbcon clears the screen anyway */
-#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,12)
-	if(!clrscrn) modeno |= 0x80;
-#else
 	modeno |= 0x80;
-#endif
 
 	outSISIDXREG(SISSR, IND_SIS_PASSWORD, SIS_PASSWORD);
 
@@ -1436,11 +1425,8 @@ sisfb_set_par(struct fb_info *info)
 	if((err = sisfb_do_set_var(&info->var, 1, info)))
 		return err;
 
-#if LINUX_VERSION_CODE < KERNEL_VERSION(2,6,10)
-	sisfb_get_fix(&info->fix, info->currcon, info);
-#else
 	sisfb_get_fix(&info->fix, -1, info);
-#endif
+
 	return 0;
 }
 
@@ -1676,14 +1662,8 @@ sisfb_blank(int blank, struct fb_info *info)
 
 /* ----------- FBDev related routines for all series ---------- */
 
-#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,15)
 static int	sisfb_ioctl(struct fb_info *info, unsigned int cmd,
 			    unsigned long arg)
-#else
-static int	sisfb_ioctl(struct inode *inode, struct file *file,
-				unsigned int cmd, unsigned long arg,
-				struct fb_info *info)
-#endif
 {
 	struct sis_video_info	*ivideo = (struct sis_video_info *)info->par;
 	struct sis_memreq	sismemreq;
@@ -3986,8 +3966,7 @@ sisfb_handle_command(struct sis_video_info *ivideo, struct sisfb_cmd *sisfb_comm
 }
 
 #ifndef MODULE
-SISINITSTATIC int __init
-sisfb_setup(char *options)
+static int __init sisfb_setup(char *options)
 {
 	char *this_opt;
 
@@ -4086,9 +4065,9 @@ sisfb_setup(char *options)
 #endif
 
 static int __devinit
-sisfb_check_rom(SIS_IOTYPE1 *rom_base, struct sis_video_info *ivideo)
+sisfb_check_rom(void __iomem *rom_base, struct sis_video_info *ivideo)
 {
-	SIS_IOTYPE1 *rom;
+	void __iomem *rom;
 	int romptr;
 
 	if((readb(rom_base) != 0x55) || (readb(rom_base + 1) != 0xaa))
@@ -4117,10 +4096,9 @@ static unsigned char * __devinit
 sisfb_find_rom(struct pci_dev *pdev)
 {
 	struct sis_video_info *ivideo = pci_get_drvdata(pdev);
-	SIS_IOTYPE1 *rom_base;
+	void __iomem *rom_base;
 	unsigned char *myrombase = NULL;
 	u32 temp;
-#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,11)
 	size_t romsize;
 
 	/* First, try the official pci ROM functions (except
@@ -4151,7 +4129,6 @@ sisfb_find_rom(struct pci_dev *pdev)
 	}
 
 	if(myrombase) return myrombase;
-#endif
 
 	/* Otherwise do it the conventional way. */
 
@@ -4225,7 +4202,7 @@ sisfb_post_map_vram(struct sis_video_info *ivideo, unsigned int *mapsize,
 static int __devinit
 sisfb_post_300_buswidth(struct sis_video_info *ivideo)
 {
-	SIS_IOTYPE1 *FBAddress = ivideo->video_vbase;
+	void __iomem *FBAddress = ivideo->video_vbase;
 	unsigned short temp;
 	unsigned char reg;
 	int i, j;
@@ -4273,7 +4250,7 @@ sisfb_post_300_rwtest(struct sis_video_info *ivideo, int iteration, int buswidth
 			int PseudoRankCapacity, int PseudoAdrPinCount,
 			unsigned int mapsize)
 {
-	SIS_IOTYPE1 *FBAddr = ivideo->video_vbase;
+	void __iomem *FBAddr = ivideo->video_vbase;
 	unsigned short sr14;
 	unsigned int k, RankCapacity, PageCapacity, BankNumHigh, BankNumMid;
 	unsigned int PhysicalAdrOtherPage, PhysicalAdrHigh, PhysicalAdrHalfPage;
@@ -5829,7 +5806,7 @@ sisfb_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	ivideo->engineok = 0;
 
 	ivideo->sisfb_was_boot_device = 0;
-#if (LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,12))
+
 	if(pdev->resource[PCI_ROM_RESOURCE].flags & IORESOURCE_ROM_SHADOW) {
 		if(ivideo->sisvga_enabled)
 			ivideo->sisfb_was_boot_device = 1;
@@ -5840,7 +5817,6 @@ sisfb_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 				"as the primary VGA device\n");
 		}
 	}
-#endif
 
 	ivideo->sisfb_parm_mem = sisfb_parm_mem;
 	ivideo->sisfb_accel = sisfb_accel;
@@ -6010,7 +5986,7 @@ sisfb_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 		ivideo->modeprechange = reg & 0x7f;
 	} else if(ivideo->sisvga_enabled) {
 #if defined(__i386__) || defined(__x86_64__)
-		unsigned char SIS_IOTYPE2 *tt = ioremap(0x400, 0x100);
+		unsigned char __iomem *tt = ioremap(0x400, 0x100);
 		if(tt) {
 			ivideo->modeprechange = readb(tt + 0x49);
 			iounmap(tt);
@@ -6503,7 +6479,7 @@ static struct pci_driver sisfb_driver = {
 	.remove 	= __devexit_p(sisfb_remove)
 };
 
-SISINITSTATIC int __init sisfb_init(void)
+static int __init sisfb_init(void)
 {
 #ifndef MODULE
 	char *options = NULL;
diff --git a/drivers/video/sis/sis_main.h b/drivers/video/sis/sis_main.h
index 3e3b7fa05d6c..9540e977270e 100644
--- a/drivers/video/sis/sis_main.h
+++ b/drivers/video/sis/sis_main.h
@@ -665,11 +665,11 @@ static struct _customttable {
 
 /* Interface used by the world */
 #ifndef MODULE
-SISINITSTATIC int sisfb_setup(char *options);
+static int sisfb_setup(char *options);
 #endif
 
 /* Interface to the low level console driver */
-SISINITSTATIC int sisfb_init(void);
+static int sisfb_init(void);
 
 /* fbdev routines */
 static int	sisfb_get_fix(struct fb_fix_screeninfo *fix, int con,
diff --git a/drivers/video/sis/vgatypes.h b/drivers/video/sis/vgatypes.h
index b532fbd2b04c..81a22eaabfde 100644
--- a/drivers/video/sis/vgatypes.h
+++ b/drivers/video/sis/vgatypes.h
@@ -53,10 +53,6 @@
 #ifndef _VGATYPES_H_
 #define _VGATYPES_H_
 
-#ifdef SIS_LINUX_KERNEL
-#include <linux/version.h>
-#endif
-
 #define SISIOMEMTYPE
 
 #ifdef SIS_LINUX_KERNEL

From b91dbce56a8dbf312f6255d5121b295553d2b4db Mon Sep 17 00:00:00 2001
From: Matthias Kaehlcke <matthias@kaehlcke.net>
Date: Wed, 23 Jul 2008 21:31:14 -0700
Subject: [PATCH 0475/1171] pxafb: convert ctrlr_sem in a mutex

The semaphore ctrlr_sem is used as a mutex.  Convert it to the mutex API.

Signed-off-by: Matthias Kaehlcke <matthias@kaehlcke.net>
Cc: Daniel Mack <daniel@caiaq.de>
Cc: Eric Miao <eric.miao@marvell.com>
Cc: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/pxafb.c | 7 ++++---
 drivers/video/pxafb.h | 2 +-
 2 files changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/video/pxafb.c b/drivers/video/pxafb.c
index 5e8a140399fc..2b707a8ce5de 100644
--- a/drivers/video/pxafb.c
+++ b/drivers/video/pxafb.c
@@ -41,6 +41,7 @@
 #include <linux/clk.h>
 #include <linux/err.h>
 #include <linux/completion.h>
+#include <linux/mutex.h>
 #include <linux/kthread.h>
 #include <linux/freezer.h>
 
@@ -1117,7 +1118,7 @@ static void set_ctrlr_state(struct pxafb_info *fbi, u_int state)
 {
 	u_int old_state;
 
-	down(&fbi->ctrlr_sem);
+	mutex_lock(&fbi->ctrlr_lock);
 
 	old_state = fbi->state;
 
@@ -1205,7 +1206,7 @@ static void set_ctrlr_state(struct pxafb_info *fbi, u_int state)
 		}
 		break;
 	}
-	up(&fbi->ctrlr_sem);
+	mutex_unlock(&fbi->ctrlr_lock);
 }
 
 /*
@@ -1458,7 +1459,7 @@ static struct pxafb_info * __devinit pxafb_init_fbinfo(struct device *dev)
 
 	init_waitqueue_head(&fbi->ctrlr_wait);
 	INIT_WORK(&fbi->task, pxafb_task);
-	init_MUTEX(&fbi->ctrlr_sem);
+	mutex_init(&fbi->ctrlr_lock);
 	init_completion(&fbi->disable_done);
 #ifdef CONFIG_FB_PXA_SMARTPANEL
 	init_completion(&fbi->command_done);
diff --git a/drivers/video/pxafb.h b/drivers/video/pxafb.h
index 8238dc826429..31541b86f13d 100644
--- a/drivers/video/pxafb.h
+++ b/drivers/video/pxafb.h
@@ -106,7 +106,7 @@ struct pxafb_info {
 
 	volatile u_char		state;
 	volatile u_char		task_state;
-	struct semaphore	ctrlr_sem;
+	struct mutex		ctrlr_lock;
 	wait_queue_head_t	ctrlr_wait;
 	struct work_struct	task;
 

From 7951ac91c7d45b61f54f1cdabc24b52b40785de6 Mon Sep 17 00:00:00 2001
From: Matthias Kaehlcke <matthias@kaehlcke.net>
Date: Wed, 23 Jul 2008 21:31:16 -0700
Subject: [PATCH 0476/1171] sa1100fb: convert ctrlr_sem in a mutex

The semaphore ctrlr_sem is used as a mutex.  Convert it to the mutex API

Signed-off-by: Matthias Kaehlcke <matthias@kaehlcke.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/sa1100fb.c | 7 ++++---
 drivers/video/sa1100fb.h | 2 +-
 2 files changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/video/sa1100fb.c b/drivers/video/sa1100fb.c
index 4a9f7e121807..78bcdbc3f484 100644
--- a/drivers/video/sa1100fb.c
+++ b/drivers/video/sa1100fb.c
@@ -175,6 +175,7 @@
 #include <linux/cpufreq.h>
 #include <linux/platform_device.h>
 #include <linux/dma-mapping.h>
+#include <linux/mutex.h>
 
 #include <asm/hardware.h>
 #include <asm/io.h>
@@ -1108,7 +1109,7 @@ static void set_ctrlr_state(struct sa1100fb_info *fbi, u_int state)
 {
 	u_int old_state;
 
-	down(&fbi->ctrlr_sem);
+	mutex_lock(&fbi->ctrlr_lock);
 
 	old_state = fbi->state;
 
@@ -1193,7 +1194,7 @@ static void set_ctrlr_state(struct sa1100fb_info *fbi, u_int state)
 		}
 		break;
 	}
-	up(&fbi->ctrlr_sem);
+	mutex_unlock(&fbi->ctrlr_lock);
 }
 
 /*
@@ -1445,7 +1446,7 @@ static struct sa1100fb_info * __init sa1100fb_init_fbinfo(struct device *dev)
 
 	init_waitqueue_head(&fbi->ctrlr_wait);
 	INIT_WORK(&fbi->task, sa1100fb_task);
-	init_MUTEX(&fbi->ctrlr_sem);
+	mutex_init(&fbi->ctrlr_lock);
 
 	return fbi;
 }
diff --git a/drivers/video/sa1100fb.h b/drivers/video/sa1100fb.h
index f465b27ed860..86831db9a042 100644
--- a/drivers/video/sa1100fb.h
+++ b/drivers/video/sa1100fb.h
@@ -100,7 +100,7 @@ struct sa1100fb_info {
 
 	volatile u_char		state;
 	volatile u_char		task_state;
-	struct semaphore	ctrlr_sem;
+	struct mutex		ctrlr_lock;
 	wait_queue_head_t	ctrlr_wait;
 	struct work_struct	task;
 

From fcea8030b3c2e71ad89f080901c63a04f07881c8 Mon Sep 17 00:00:00 2001
From: Tony Breeds <tony@bakeyournoodle.com>
Date: Wed, 23 Jul 2008 21:31:16 -0700
Subject: [PATCH 0477/1171] drivers/video/aty/radeon_base.c: notify user if
 sysfs_create_bin_file() failed

Current kernel builds warn about:
drivers/video/aty/radeon_base.c: In function 'radeonfb_pci_register':
drivers/video/aty/radeon_base.c:2334: warning: ignoring return value of 'sysfs_create_bin_file', declared with attribute warn_unused_result
drivers/video/aty/radeon_base.c:2336: warning: ignoring return value of 'sysfs_create_bin_file', declared with attribute warn_unused_result

Do minimal checking of these functions and issue a warning if either
fails.  They don't seem to be critical..

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Tony Breeds <tony@bakeyournoodle.com>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/aty/radeon_base.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/video/aty/radeon_base.c b/drivers/video/aty/radeon_base.c
index 400e9264e456..3c0a03f69d83 100644
--- a/drivers/video/aty/radeon_base.c
+++ b/drivers/video/aty/radeon_base.c
@@ -2161,6 +2161,7 @@ static int __devinit radeonfb_pci_register (struct pci_dev *pdev,
 	struct radeonfb_info *rinfo;
 	int ret;
 	unsigned char c1, c2;
+	int err = 0;
 
 	pr_debug("radeonfb_pci_register BEGIN\n");
 	
@@ -2340,9 +2341,14 @@ static int __devinit radeonfb_pci_register (struct pci_dev *pdev,
 
 	/* Register some sysfs stuff (should be done better) */
 	if (rinfo->mon1_EDID)
-		sysfs_create_bin_file(&rinfo->pdev->dev.kobj, &edid1_attr);
+		err |= sysfs_create_bin_file(&rinfo->pdev->dev.kobj,
+						&edid1_attr);
 	if (rinfo->mon2_EDID)
-		sysfs_create_bin_file(&rinfo->pdev->dev.kobj, &edid2_attr);
+		err |= sysfs_create_bin_file(&rinfo->pdev->dev.kobj,
+						&edid2_attr);
+	if (err)
+		pr_warning("%s() Creating sysfs files failed, continuing\n",
+			   __func__);
 
 	/* save current mode regs before we switch into the new one
 	 * so we can restore this upon __exit

From 816664f88707b03fde24fb09759d569ed42406cb Mon Sep 17 00:00:00 2001
From: Roel Kluin <12o3l@tiscali.nl>
Date: Wed, 23 Jul 2008 21:31:17 -0700
Subject: [PATCH 0478/1171] aty128fb: test below 0 on unsigned
 pll->post_divider

pll->post_divider is unsigned, so the test fails

Signed-off-by: Roel Kluin <12o3l@tiscali.nl>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/aty/aty128fb.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/video/aty/aty128fb.c b/drivers/video/aty/aty128fb.c
index 07b6addbb3c1..243ea4ab20c8 100644
--- a/drivers/video/aty/aty128fb.c
+++ b/drivers/video/aty/aty128fb.c
@@ -1339,10 +1339,8 @@ static int aty128_var_to_pll(u32 period_in_ps, struct aty128_pll *pll,
 	if (vclk * 12 < c.ppll_min)
 		vclk = c.ppll_min/12;
 
-	pll->post_divider = -1;
-
 	/* now, find an acceptable divider */
-	for (i = 0; i < sizeof(post_dividers); i++) {
+	for (i = 0; i < ARRAY_SIZE(post_dividers); i++) {
 		output_freq = post_dividers[i] * vclk;
 		if (output_freq >= c.ppll_min && output_freq <= c.ppll_max) {
 			pll->post_divider = post_dividers[i];
@@ -1350,7 +1348,7 @@ static int aty128_var_to_pll(u32 period_in_ps, struct aty128_pll *pll,
 		}
 	}
 
-	if (pll->post_divider < 0)
+	if (i == ARRAY_SIZE(post_dividers))
 		return -EINVAL;
 
 	/* calculate feedback divider */

From 091c82c01295719d47b89b38d24e41ad2066ead8 Mon Sep 17 00:00:00 2001
From: Roel Kluin <12o3l@tiscali.nl>
Date: Wed, 23 Jul 2008 21:31:18 -0700
Subject: [PATCH 0479/1171] amifb: test virtual screen range before subtraction
 on unsigned

dx and dy are u32's, so the test should occur before the subtraction

Signed-off-by: Roel Kluin <12o3l@tiscali.nl>
Cc: Antonino Daplas <adaplas@pol.net>
Cc: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/amifb.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/video/amifb.c b/drivers/video/amifb.c
index 45c154ade9ca..0c549aa1cf87 100644
--- a/drivers/video/amifb.c
+++ b/drivers/video/amifb.c
@@ -2048,13 +2048,16 @@ static void amifb_copyarea(struct fb_info *info,
 	width = x2 - dx;
 	height = y2 - dy;
 
+	if (area->sx + dx < area->dx || area->sy + dy < area->dy)
+		return;
+
 	/* update sx,sy */
 	sx = area->sx + (dx - area->dx);
 	sy = area->sy + (dy - area->dy);
 
 	/* the source must be completely inside the virtual screen */
-	if (sx < 0 || sy < 0 || (sx + width) > info->var.xres_virtual ||
-	    (sy + height) > info->var.yres_virtual)
+	if (sx + width > info->var.xres_virtual ||
+			sy + height > info->var.yres_virtual)
 		return;
 
 	if (dy > sy || (dy == sy && dx > sx)) {

From 1c0face9d4024bf942096297937759bdf0e1aeac Mon Sep 17 00:00:00 2001
From: Roel Kluin <12o3l@tiscali.nl>
Date: Wed, 23 Jul 2008 21:31:18 -0700
Subject: [PATCH 0480/1171] atafb: test virtual screen range before subtraction
 on unsigned

dx and dy are u32's, so the test should occur before the subtraction

Signed-off-by: Roel Kluin <12o3l@tiscali.nl>
Cc: Tim Schmielau <tim@physik3.uni-rostock.de>
Cc: Krzysztof Helt <krzysztof.h1@wp.pl>
Cc: Antonino Daplas <adaplas@pol.net>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/atafb.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/video/atafb.c b/drivers/video/atafb.c
index fa55d356b535..77eb8b34fbfa 100644
--- a/drivers/video/atafb.c
+++ b/drivers/video/atafb.c
@@ -2593,13 +2593,16 @@ static void atafb_copyarea(struct fb_info *info, const struct fb_copyarea *area)
 	width = x2 - dx;
 	height = y2 - dy;
 
+	if (area->sx + dx < area->dx || area->sy + dy < area->dy)
+		return;
+
 	/* update sx,sy */
 	sx = area->sx + (dx - area->dx);
 	sy = area->sy + (dy - area->dy);
 
 	/* the source must be completely inside the virtual screen */
-	if (sx < 0 || sy < 0 || (sx + width) > info->var.xres_virtual ||
-	    (sy + height) > info->var.yres_virtual)
+	if (sx + width > info->var.xres_virtual ||
+			sy + height > info->var.yres_virtual)
 		return;
 
 	if (dy > sy || (dy == sy && dx > sx)) {

From 77a6e7abb09de0e85a15e2fe42c21ffc59847759 Mon Sep 17 00:00:00 2001
From: Roel Kluin <12o3l@tiscali.nl>
Date: Wed, 23 Jul 2008 21:31:19 -0700
Subject: [PATCH 0481/1171] vga16fb: test virtual screen range before
 subtraction on unsigned

dx and dy are u32's, so the test should occur before the subtraction

Signed-off-by: Roel Kluin <12o3l@tiscali.nl>
Cc: Antonino Daplas <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/vga16fb.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/drivers/video/vga16fb.c b/drivers/video/vga16fb.c
index 9b3c5923365e..9d2751717896 100644
--- a/drivers/video/vga16fb.c
+++ b/drivers/video/vga16fb.c
@@ -1087,12 +1087,15 @@ static void vga16fb_copyarea(struct fb_info *info, const struct fb_copyarea *are
 	width = x2 - dx;
 	height = y2 - dy;
 
+	if (sx + dx < old_dx || sy + dy < old_dy)
+		return;
+
 	/* update sx1,sy1 */
 	sx += (dx - old_dx);
 	sy += (dy - old_dy);
 
 	/* the source must be completely inside the virtual screen */
-	if (sx < 0 || sy < 0 || (sx + width) > vxres || (sy + height) > vyres)
+	if (sx + width > vxres || sy + height > vyres)
 		return;
 
 	switch (info->fix.type) {

From d22579b837358cbef12ccca5adaf7e93ae09ab7a Mon Sep 17 00:00:00 2001
From: Nicolas Ferre <nicolas.ferre@atmel.com>
Date: Wed, 23 Jul 2008 21:31:20 -0700
Subject: [PATCH 0482/1171] atmel_lcdfb: FIFO underflow management

Manage atmel_lcdfb FIFO underflow

Resetting the LCD and DMA allows to fix screen shifting after a FIFO
underflow.  It follows reset sequence from errata "LCD Screen Shifting
After a Reset".

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Cc: Haavard Skinnemoen <hskinnemoen@atmel.com>
Cc: Andrew Victor <linux@maxim.org.za>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/atmel_lcdfb.c | 57 ++++++++++++++++++++++++++++++++++++-
 include/video/atmel_lcdc.h  |  1 +
 2 files changed, 57 insertions(+), 1 deletion(-)

diff --git a/drivers/video/atmel_lcdfb.c b/drivers/video/atmel_lcdfb.c
index b004036d4087..d335bb96b03b 100644
--- a/drivers/video/atmel_lcdfb.c
+++ b/drivers/video/atmel_lcdfb.c
@@ -379,6 +379,35 @@ static int atmel_lcdfb_check_var(struct fb_var_screeninfo *var,
 	return 0;
 }
 
+/*
+ * LCD reset sequence
+ */
+static void atmel_lcdfb_reset(struct atmel_lcdfb_info *sinfo)
+{
+	might_sleep();
+
+	/* LCD power off */
+	lcdc_writel(sinfo, ATMEL_LCDC_PWRCON, sinfo->guard_time << ATMEL_LCDC_GUARDT_OFFSET);
+
+	/* wait for the LCDC core to become idle */
+	while (lcdc_readl(sinfo, ATMEL_LCDC_PWRCON) & ATMEL_LCDC_BUSY)
+		msleep(10);
+
+	/* DMA disable */
+	lcdc_writel(sinfo, ATMEL_LCDC_DMACON, 0);
+
+	/* wait for DMA engine to become idle */
+	while (lcdc_readl(sinfo, ATMEL_LCDC_DMACON) & ATMEL_LCDC_DMABUSY)
+		msleep(10);
+
+	/* LCD power on */
+	lcdc_writel(sinfo, ATMEL_LCDC_PWRCON,
+		(sinfo->guard_time << ATMEL_LCDC_GUARDT_OFFSET) | ATMEL_LCDC_PWR);
+
+	/* DMA enable */
+	lcdc_writel(sinfo, ATMEL_LCDC_DMACON, sinfo->default_dmacon);
+}
+
 /**
  *      atmel_lcdfb_set_par - Alters the hardware state.
  *      @info: frame buffer structure that represents a single frame buffer
@@ -401,6 +430,8 @@ static int atmel_lcdfb_set_par(struct fb_info *info)
 	unsigned long clk_value_khz;
 	unsigned long bits_per_line;
 
+	might_sleep();
+
 	dev_dbg(info->device, "%s:\n", __func__);
 	dev_dbg(info->device, "  * resolution: %ux%u (%ux%u virtual)\n",
 		 info->var.xres, info->var.yres,
@@ -511,6 +542,8 @@ static int atmel_lcdfb_set_par(struct fb_info *info)
 
 	/* Disable all interrupts */
 	lcdc_writel(sinfo, ATMEL_LCDC_IDR, ~0UL);
+	/* Enable FIFO & DMA errors */
+	lcdc_writel(sinfo, ATMEL_LCDC_IER, ATMEL_LCDC_UFLWI | ATMEL_LCDC_OWRI | ATMEL_LCDC_MERI);
 
 	/* ...wait for DMA engine to become idle... */
 	while (lcdc_readl(sinfo, ATMEL_LCDC_DMACON) & ATMEL_LCDC_DMABUSY)
@@ -645,10 +678,26 @@ static irqreturn_t atmel_lcdfb_interrupt(int irq, void *dev_id)
 	u32 status;
 
 	status = lcdc_readl(sinfo, ATMEL_LCDC_ISR);
-	lcdc_writel(sinfo, ATMEL_LCDC_IDR, status);
+	if (status & ATMEL_LCDC_UFLWI) {
+		dev_warn(info->device, "FIFO underflow %#x\n", status);
+		/* reset DMA and FIFO to avoid screen shifting */
+		schedule_work(&sinfo->task);
+	}
+	lcdc_writel(sinfo, ATMEL_LCDC_ICR, status);
 	return IRQ_HANDLED;
 }
 
+/*
+ * LCD controller task (to reset the LCD)
+ */
+static void atmel_lcdfb_task(struct work_struct *work)
+{
+	struct atmel_lcdfb_info *sinfo =
+		container_of(work, struct atmel_lcdfb_info, task);
+
+	atmel_lcdfb_reset(sinfo);
+}
+
 static int __init atmel_lcdfb_init_fbinfo(struct atmel_lcdfb_info *sinfo)
 {
 	struct fb_info *info = sinfo->info;
@@ -824,6 +873,10 @@ static int __init atmel_lcdfb_probe(struct platform_device *pdev)
 		goto unmap_mmio;
 	}
 
+	/* Some operations on the LCDC might sleep and
+	 * require a preemptible task context */
+	INIT_WORK(&sinfo->task, atmel_lcdfb_task);
+
 	ret = atmel_lcdfb_init_fbinfo(sinfo);
 	if (ret < 0) {
 		dev_err(dev, "init fbinfo failed: %d\n", ret);
@@ -866,6 +919,7 @@ static int __init atmel_lcdfb_probe(struct platform_device *pdev)
 free_cmap:
 	fb_dealloc_cmap(&info->cmap);
 unregister_irqs:
+	cancel_work_sync(&sinfo->task);
 	free_irq(sinfo->irq_base, info);
 unmap_mmio:
 	exit_backlight(sinfo);
@@ -903,6 +957,7 @@ static int __exit atmel_lcdfb_remove(struct platform_device *pdev)
 	if (!sinfo)
 		return 0;
 
+	cancel_work_sync(&sinfo->task);
 	exit_backlight(sinfo);
 	if (sinfo->atmel_lcdfb_power_control)
 		sinfo->atmel_lcdfb_power_control(0);
diff --git a/include/video/atmel_lcdc.h b/include/video/atmel_lcdc.h
index ed64862c4e18..1ccf462b433a 100644
--- a/include/video/atmel_lcdc.h
+++ b/include/video/atmel_lcdc.h
@@ -37,6 +37,7 @@ struct atmel_lcdfb_info {
 	struct fb_info		*info;
 	void __iomem		*mmio;
 	unsigned long		irq_base;
+	struct work_struct	task;
 
 	unsigned int		guard_time;
 	struct platform_device	*pdev;

From 49a1d28f57adc9cb064572f0373e26363b0a412f Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:21 -0700
Subject: [PATCH 0483/1171] fbcon: make logo_height a local variable

Make logo_height variable local in the only function it is used.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/console/fbcon.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/video/console/fbcon.c b/drivers/video/console/fbcon.c
index 4be3b46c069b..3ccfa76d9b2a 100644
--- a/drivers/video/console/fbcon.c
+++ b/drivers/video/console/fbcon.c
@@ -107,9 +107,7 @@ static struct display fb_display[MAX_NR_CONSOLES];
 
 static signed char con2fb_map[MAX_NR_CONSOLES];
 static signed char con2fb_map_boot[MAX_NR_CONSOLES];
-#ifndef MODULE
-static int logo_height;
-#endif
+
 static int logo_lines;
 /* logo_shown is an index to vc_cons when >= 0; otherwise follows FBCON_LOGO
    enums.  */
@@ -607,6 +605,7 @@ static void fbcon_prepare_logo(struct vc_data *vc, struct fb_info *info,
 	struct fbcon_ops *ops = info->fbcon_par;
 	int cnt, erase = vc->vc_video_erase_char, step;
 	unsigned short *save = NULL, *r, *q;
+	int logo_height;
 
 	if (info->flags & FBINFO_MODULE) {
 		logo_shown = FBCON_LOGO_DONTSHOW;

From 012e26096b36bfeacaba2c9e31eaf32d6faa6567 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:21 -0700
Subject: [PATCH 0484/1171] uvesafb: change mode parameter to mode_option

Make more drivers use the "mode_option" parameter.  This one is quite new
so drop the old "mode" parameter before someone starts using it seriously.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/uvesafb.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/video/uvesafb.c b/drivers/video/uvesafb.c
index cdbb56edb6cb..50744229c7a9 100644
--- a/drivers/video/uvesafb.c
+++ b/drivers/video/uvesafb.c
@@ -2054,8 +2054,8 @@ MODULE_PARM_DESC(maxhf,
 module_param(maxvf, ushort, 0);
 MODULE_PARM_DESC(maxvf,
 	"Maximum vertical frequency [Hz], overrides EDID data");
-module_param_named(mode, mode_option, charp, 0);
-MODULE_PARM_DESC(mode,
+module_param(mode_option, charp, 0);
+MODULE_PARM_DESC(mode_option,
 	"Specify initial video mode as \"<xres>x<yres>[-<bpp>][@<refresh>]\"");
 module_param(vbemode, ushort, 0);
 MODULE_PARM_DESC(vbemode,

From a90ed92ed852a3d4b8a6f20b10bba771997f5ede Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:22 -0700
Subject: [PATCH 0485/1171] tridentfb: documentation update

Make the tridentfb documentation closer to current state of the tridentfb
driver.  Fix also some formatting.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/fb/tridentfb.txt | 44 +++++++++++++++++++++++-----------
 1 file changed, 30 insertions(+), 14 deletions(-)

diff --git a/Documentation/fb/tridentfb.txt b/Documentation/fb/tridentfb.txt
index 8a6c8a43e6a3..45d9de5b13a3 100644
--- a/Documentation/fb/tridentfb.txt
+++ b/Documentation/fb/tridentfb.txt
@@ -3,11 +3,25 @@ Tridentfb is a framebuffer driver for some Trident chip based cards.
 The following list of chips is thought to be supported although not all are
 tested:
 
-those from the Image series with Cyber in their names - accelerated
-those with Blade in their names (Blade3D,CyberBlade...) - accelerated
-the newer CyberBladeXP family  - nonaccelerated
+those from the TGUI series 9440/96XX and with Cyber in their names
+those from the Image series and with Cyber in their names
+those with Blade in their names (Blade3D,CyberBlade...)
+the newer CyberBladeXP family
 
-Only PCI/AGP based cards are supported, none of the older Tridents.
+All families are accelerated. Only PCI/AGP based cards are supported,
+none of the older Tridents.
+The driver supports 8, 16 and 32 bits per pixel depths.
+The TGUI family requires a line length to be power of 2 if acceleration
+is enabled. This means that range of possible resolutions and bpp is
+limited comparing to the range if acceleration is disabled (see list
+of parameters below).
+
+Known bugs:
+1. The driver randomly locks up on 3DImage975 chip with acceleration
+   enabled. The same happens in X11 (Xorg).
+2. The ramdac speeds require some more fine tuning. It is possible to
+   switch resolution which the chip does not support at some depths for
+   older chips.
 
 How to use it?
 ==============
@@ -17,12 +31,11 @@ video=tridentfb
 
 The parameters for tridentfb are concatenated with a ':' as in this example.
 
-video=tridentfb:800x600,bpp=16,noaccel
+video=tridentfb:800x600-16@75,noaccel
 
 The second level parameters that tridentfb understands are:
 
 noaccel - turns off acceleration (when it doesn't work for your card)
-accel - force text acceleration (for boards which by default are noacceled)
 
 fp	- use flat panel related stuff
 crt 	- assume monitor is present instead of fp
@@ -31,21 +44,24 @@ center 	- for flat panels and resolutions smaller than native size center the
 	  image, otherwise use
 stretch
 
-memsize - integer value in Kb, use if your card's memory size is misdetected.
+memsize - integer value in KB, use if your card's memory size is misdetected.
 	  look at the driver output to see what it says when initializing.
-memdiff - integer value in Kb,should be nonzero if your card reports
-	  more memory than it actually has.For instance mine is 192K less than
+
+memdiff - integer value in KB, should be nonzero if your card reports
+	  more memory than it actually has. For instance mine is 192K less than
 	  detection says in all three BIOS selectable situations 2M, 4M, 8M.
 	  Only use if your video memory is taken from main memory hence of
-	  configurable size.Otherwise use memsize.
-	  If in some modes which barely fit the memory you see garbage at the bottom
-	  this might help by not letting change to that mode anymore.
+	  configurable size. Otherwise use memsize.
+	  If in some modes which barely fit the memory you see garbage
+	  at the bottom this might help by not letting change to that mode
+	  anymore.
 
 nativex - the width in pixels of the flat panel.If you know it (usually 1024
 	  800 or 1280) and it is not what the driver seems to detect use it.
 
-bpp  - bits per pixel (8,16 or 32)
-mode - a mode name like 800x600 (as described in Documentation/fb/modedb.txt)
+bpp	- bits per pixel (8,16 or 32)
+mode	- a mode name like 800x600-8@75 as described in
+	  Documentation/fb/modedb.txt
 
 Using insane values for the above parameters will probably result in driver
 misbehaviour so take care(for instance memsize=12345678 or memdiff=23784 or

From ea9014bcacf236124d5e0ff971838049a98456cb Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:22 -0700
Subject: [PATCH 0486/1171] tdfxfb: add mode_option module parameter

Small step toward unification of mode setting parameter.  This is required
to fix the Bugzilla's bug 9847

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tdfxfb.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/video/tdfxfb.c b/drivers/video/tdfxfb.c
index ea9f19d25597..67664252868d 100644
--- a/drivers/video/tdfxfb.c
+++ b/drivers/video/tdfxfb.c
@@ -1426,6 +1426,8 @@ MODULE_LICENSE("GPL");
 module_param(hwcursor, int, 0644);
 MODULE_PARM_DESC(hwcursor, "Enable hardware cursor "
 			"(1=enable, 0=disable, default=1)");
+module_param(mode_option, charp, 0);
+MODULE_PARM_DESC(mode_option, "Initial video mode e.g. '648x480-8@60'");
 #ifdef CONFIG_MTRR
 module_param(nomtrr, bool, 0);
 MODULE_PARM_DESC(nomtrr, "Disable MTRR support (default: enabled)");

From 98219374d9ed2d257e56e8e1fcd9d16a083397bb Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:23 -0700
Subject: [PATCH 0487/1171] vga16fb: source code improvement

Use constants and functions from the vga.h file.  Also add module
description.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/vga16fb.c | 117 ++++++++++++++--------------------------
 1 file changed, 39 insertions(+), 78 deletions(-)

diff --git a/drivers/video/vga16fb.c b/drivers/video/vga16fb.c
index 9d2751717896..e31bca8a0cb2 100644
--- a/drivers/video/vga16fb.c
+++ b/drivers/video/vga16fb.c
@@ -26,18 +26,6 @@
 #include <asm/io.h>
 #include <video/vga.h>
 
-#define GRAPHICS_ADDR_REG VGA_GFX_I	/* Graphics address register. */
-#define GRAPHICS_DATA_REG VGA_GFX_D	/* Graphics data register. */
-
-#define SET_RESET_INDEX 	VGA_GFX_SR_VALUE	/* Set/Reset Register index. */
-#define ENABLE_SET_RESET_INDEX	VGA_GFX_SR_ENABLE	/* Enable Set/Reset Register index. */
-#define DATA_ROTATE_INDEX	VGA_GFX_DATA_ROTATE	/* Data Rotate Register index. */
-#define GRAPHICS_MODE_INDEX	VGA_GFX_MODE		/* Graphics Mode Register index. */
-#define BIT_MASK_INDEX		VGA_GFX_BIT_MASK	/* Bit Mask Register index. */
-
-#define dac_reg	(VGA_PEL_IW)
-#define dac_val	(VGA_PEL_D)
-
 #define VGA_FB_PHYS 0xA0000
 #define VGA_FB_PHYS_LEN 65536
 
@@ -108,7 +96,7 @@ static struct fb_fix_screeninfo vga16fb_fix __initdata = {
 	.visual		= FB_VISUAL_PSEUDOCOLOR,
 	.xpanstep	= 8,
 	.ypanstep	= 1,
-	.line_length	= 640/8,
+	.line_length	= 640 / 8,
 	.accel		= FB_ACCEL_NONE
 };
 
@@ -135,23 +123,22 @@ static inline int setmode(int mode)
 {
 	int oldmode;
 	
-	vga_io_w(GRAPHICS_ADDR_REG, GRAPHICS_MODE_INDEX);
-	oldmode = vga_io_r(GRAPHICS_DATA_REG);
-	vga_io_w(GRAPHICS_DATA_REG, mode);
+	oldmode = vga_io_rgfx(VGA_GFX_MODE);
+	vga_io_w(VGA_GFX_D, mode);
 	return oldmode;
 }
 
 /* Select the Bit Mask Register and return its value. */
 static inline int selectmask(void)
 {
-	return vga_io_rgfx(BIT_MASK_INDEX);
+	return vga_io_rgfx(VGA_GFX_BIT_MASK);
 }
 
 /* Set the value of the Bit Mask Register.  It must already have been
    selected with selectmask(). */
 static inline void setmask(int mask)
 {
-	vga_io_w(GRAPHICS_DATA_REG, mask);
+	vga_io_w(VGA_GFX_D, mask);
 }
 
 /* Set the Data Rotate Register and return its old value. 
@@ -161,9 +148,8 @@ static inline int setop(int op)
 {
 	int oldop;
 	
-	vga_io_w(GRAPHICS_ADDR_REG, DATA_ROTATE_INDEX);
-	oldop = vga_io_r(GRAPHICS_DATA_REG);
-	vga_io_w(GRAPHICS_DATA_REG, op);
+	oldop = vga_io_rgfx(VGA_GFX_DATA_ROTATE);
+	vga_io_w(VGA_GFX_D, op);
 	return oldop;
 }
 
@@ -173,9 +159,8 @@ static inline int setsr(int sr)
 {
 	int oldsr;
 
-	vga_io_w(GRAPHICS_ADDR_REG, ENABLE_SET_RESET_INDEX);
-	oldsr = vga_io_r(GRAPHICS_DATA_REG);
-	vga_io_w(GRAPHICS_DATA_REG, sr);
+	oldsr = vga_io_rgfx(VGA_GFX_SR_ENABLE);
+	vga_io_w(VGA_GFX_D, sr);
 	return oldsr;
 }
 
@@ -184,22 +169,21 @@ static inline int setcolor(int color)
 {
 	int oldcolor;
 
-	vga_io_w(GRAPHICS_ADDR_REG, SET_RESET_INDEX);
-	oldcolor = vga_io_r(GRAPHICS_DATA_REG);
-	vga_io_w(GRAPHICS_DATA_REG, color);
+	oldcolor = vga_io_rgfx(VGA_GFX_SR_VALUE);
+	vga_io_w(VGA_GFX_D, color);
 	return oldcolor;
 }
 
 /* Return the value in the Graphics Address Register. */
 static inline int getindex(void)
 {
-	return vga_io_r(GRAPHICS_ADDR_REG);
+	return vga_io_r(VGA_GFX_I);
 }
 
 /* Set the value in the Graphics Address Register. */
 static inline void setindex(int index)
 {
-	vga_io_w(GRAPHICS_ADDR_REG, index);
+	vga_io_w(VGA_GFX_I, index);
 }
 
 static void vga16fb_pan_var(struct fb_info *info, 
@@ -672,10 +656,10 @@ static void ega16_setpalette(int regno, unsigned red, unsigned green, unsigned b
 
 static void vga16_setpalette(int regno, unsigned red, unsigned green, unsigned blue)
 {
-	outb(regno,       dac_reg);
-	outb(red   >> 10, dac_val);
-	outb(green >> 10, dac_val);
-	outb(blue  >> 10, dac_val);
+	outb(regno,       VGA_PEL_IW);
+	outb(red   >> 10, VGA_PEL_D);
+	outb(green >> 10, VGA_PEL_D);
+	outb(blue  >> 10, VGA_PEL_D);
 }
 
 static int vga16fb_setcolreg(unsigned regno, unsigned red, unsigned green,
@@ -719,28 +703,15 @@ static int vga16fb_pan_display(struct fb_var_screeninfo *var,
    blanking code was originally by Huang shi chao, and modified by
    Christoph Rimek (chrimek@toppoint.de) and todd j. derr
    (tjd@barefoot.org) for Linux. */
-#define attrib_port		VGA_ATC_IW
-#define seq_port_reg		VGA_SEQ_I
-#define seq_port_val		VGA_SEQ_D
-#define gr_port_reg		VGA_GFX_I
-#define gr_port_val		VGA_GFX_D
-#define video_misc_rd		VGA_MIS_R
-#define video_misc_wr		VGA_MIS_W
-#define vga_video_port_reg	VGA_CRT_IC
-#define vga_video_port_val	VGA_CRT_DC
 
 static void vga_vesa_blank(struct vga16fb_par *par, int mode)
 {
-	unsigned char SeqCtrlIndex;
-	unsigned char CrtCtrlIndex;
+	unsigned char SeqCtrlIndex = vga_io_r(VGA_SEQ_I);
+	unsigned char CrtCtrlIndex = vga_io_r(VGA_CRT_IC);
 	
-	//cli();
-	SeqCtrlIndex = vga_io_r(seq_port_reg);
-	CrtCtrlIndex = vga_io_r(vga_video_port_reg);
-
 	/* save original values of VGA controller registers */
 	if(!par->vesa_blanked) {
-		par->vga_state.CrtMiscIO = vga_io_r(video_misc_rd);
+		par->vga_state.CrtMiscIO = vga_io_r(VGA_MIS_R);
 		//sti();
 
 		par->vga_state.HorizontalTotal = vga_io_rcrt(0x00);	/* HorizontalTotal */
@@ -756,12 +727,11 @@ static void vga_vesa_blank(struct vga16fb_par *par, int mode)
 
 	/* assure that video is enabled */
 	/* "0x20" is VIDEO_ENABLE_bit in register 01 of sequencer */
-	//cli();
 	vga_io_wseq(0x01, par->vga_state.ClockingMode | 0x20);
 
 	/* test for vertical retrace in process.... */
 	if ((par->vga_state.CrtMiscIO & 0x80) == 0x80)
-		vga_io_w(video_misc_wr, par->vga_state.CrtMiscIO & 0xef);
+		vga_io_w(VGA_MIS_W, par->vga_state.CrtMiscIO & 0xef);
 
 	/*
 	 * Set <End of vertical retrace> to minimum (0) and
@@ -769,12 +739,10 @@ static void vga_vesa_blank(struct vga16fb_par *par, int mode)
 	 * Result: turn off vertical sync (VSync) pulse.
 	 */
 	if (mode & FB_BLANK_VSYNC_SUSPEND) {
-		outb_p(0x10,vga_video_port_reg);	/* StartVertRetrace */
-		outb_p(0xff,vga_video_port_val); 	/* maximum value */
-		outb_p(0x11,vga_video_port_reg);	/* EndVertRetrace */
-		outb_p(0x40,vga_video_port_val);	/* minimum (bits 0..3)  */
-		outb_p(0x07,vga_video_port_reg);	/* Overflow */
-		outb_p(par->vga_state.Overflow | 0x84,vga_video_port_val); /* bits 9,10 of vert. retrace */
+		vga_io_wcrt(VGA_CRTC_V_SYNC_START, 0xff);
+		vga_io_wcrt(VGA_CRTC_V_SYNC_END, 0x40);
+		/* bits 9,10 of vert. retrace */
+		vga_io_wcrt(VGA_CRTC_OVERFLOW, par->vga_state.Overflow | 0x84);
 	}
 
 	if (mode & FB_BLANK_HSYNC_SUSPEND) {
@@ -783,29 +751,22 @@ static void vga_vesa_blank(struct vga16fb_par *par, int mode)
 		 *  <Start of horizontal Retrace> to maximum
 		 * Result: turn off horizontal sync (HSync) pulse.
 		 */
-		outb_p(0x04,vga_video_port_reg);	/* StartHorizRetrace */
-		outb_p(0xff,vga_video_port_val);	/* maximum */
-		outb_p(0x05,vga_video_port_reg);	/* EndHorizRetrace */
-		outb_p(0x00,vga_video_port_val);	/* minimum (0) */
+		vga_io_wcrt(VGA_CRTC_H_SYNC_START, 0xff);
+		vga_io_wcrt(VGA_CRTC_H_SYNC_END, 0x00);
 	}
 
 	/* restore both index registers */
-	outb_p(SeqCtrlIndex,seq_port_reg);
-	outb_p(CrtCtrlIndex,vga_video_port_reg);
-	//sti();
+	outb_p(SeqCtrlIndex, VGA_SEQ_I);
+	outb_p(CrtCtrlIndex, VGA_CRT_IC);
 }
 
 static void vga_vesa_unblank(struct vga16fb_par *par)
 {
-	unsigned char SeqCtrlIndex;
-	unsigned char CrtCtrlIndex;
+	unsigned char SeqCtrlIndex = vga_io_r(VGA_SEQ_I);
+	unsigned char CrtCtrlIndex = vga_io_r(VGA_CRT_IC);
 	
-	//cli();
-	SeqCtrlIndex = vga_io_r(seq_port_reg);
-	CrtCtrlIndex = vga_io_r(vga_video_port_reg);
-
 	/* restore original values of VGA controller registers */
-	vga_io_w(video_misc_wr, par->vga_state.CrtMiscIO);
+	vga_io_w(VGA_MIS_W, par->vga_state.CrtMiscIO);
 
 	/* HorizontalTotal */
 	vga_io_wcrt(0x00, par->vga_state.HorizontalTotal);
@@ -827,9 +788,8 @@ static void vga_vesa_unblank(struct vga16fb_par *par)
 	vga_io_wseq(0x01, par->vga_state.ClockingMode);
 
 	/* restore index/control registers */
-	vga_io_w(seq_port_reg, SeqCtrlIndex);
-	vga_io_w(vga_video_port_reg, CrtCtrlIndex);
-	//sti();
+	vga_io_w(VGA_SEQ_I, SeqCtrlIndex);
+	vga_io_w(VGA_CRT_IC, CrtCtrlIndex);
 }
 
 static void vga_pal_blank(void)
@@ -837,10 +797,10 @@ static void vga_pal_blank(void)
 	int i;
 
 	for (i=0; i<16; i++) {
-		outb_p (i, dac_reg) ;
-		outb_p (0, dac_val) ;
-		outb_p (0, dac_val) ;
-		outb_p (0, dac_val) ;
+		outb_p(i, VGA_PEL_IW);
+		outb_p(0, VGA_PEL_D);
+		outb_p(0, VGA_PEL_D);
+		outb_p(0, VGA_PEL_D);
 	}
 }
 
@@ -1485,6 +1445,7 @@ static void __exit vga16fb_exit(void)
 	platform_driver_unregister(&vga16fb_driver);
 }
 
+MODULE_DESCRIPTION("Legacy VGA framebuffer device driver");
 MODULE_LICENSE("GPL");
 module_init(vga16fb_init);
 module_exit(vga16fb_exit);

From c2c12155cf05bf3e25eeae5711beffc634505400 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:24 -0700
Subject: [PATCH 0488/1171] tdfxfb: remove ypan checks done by a higher layer

These checks and assignments are done by a higher layer so remove them
from the driver.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/tdfxfb.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

diff --git a/drivers/video/tdfxfb.c b/drivers/video/tdfxfb.c
index 67664252868d..77aafcfae037 100644
--- a/drivers/video/tdfxfb.c
+++ b/drivers/video/tdfxfb.c
@@ -836,16 +836,12 @@ static int tdfxfb_pan_display(struct fb_var_screeninfo *var,
 	struct tdfx_par *par = info->par;
 	u32 addr = var->yoffset * info->fix.line_length;
 
-	if (nopan || var->xoffset || (var->yoffset > var->yres_virtual))
-		return -EINVAL;
-	if ((var->yoffset + var->yres > var->yres_virtual && nowrap))
+	if (nopan || var->xoffset)
 		return -EINVAL;
 
 	banshee_make_room(par, 1);
 	tdfx_outl(par, VIDDESKSTART, addr);
 
-	info->var.xoffset = var->xoffset;
-	info->var.yoffset = var->yoffset;
 	return 0;
 }
 

From cfb4f5d1750e05f43902197713c50c29e7dfbc99 Mon Sep 17 00:00:00 2001
From: Magnus Damm <magnus.damm@gmail.com>
Date: Wed, 23 Jul 2008 21:31:24 -0700
Subject: [PATCH 0489/1171] fbdev: SuperH Mobile LCDC Driver

This is the SuperH Mobile LCDC frame buffer driver V2, adding support for
the LCDC block found in SuperH Mobile processors.  The hardware supports
up to two LCD panels per LCDC block, and both RGB and SYS interfaces can
be used to hook up LCD panels/modules.

The device driver is a regular platform driver, so LCD configuration and
board specific hooks are passed to the driver using platform data.  LCD
modules using SYS interface often require special configuration using the
SYS bus, and to solve this cleanly the driver provides SYS interface
operations to the board code.

Tested on sh7723 and sh7722 processors with a SYS16A QVGA panel and WVGA
panels using RGB16 and RGB18 interfaces.

Signed-off-by: Magnus Damm <damm@igel.co.jp>
Acked-by: Paul Mundt <lethal@linux-sh.org>
Reviewed-by: Krzysztof Helt <krzysztof.h1@poczta.fm>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/Kconfig            |  10 +
 drivers/video/Makefile           |   2 +
 drivers/video/sh_mobile_lcdcfb.c | 725 +++++++++++++++++++++++++++++++
 include/asm-sh/sh_mobile_lcdc.h  |  66 +++
 4 files changed, 803 insertions(+)
 create mode 100644 drivers/video/sh_mobile_lcdcfb.c
 create mode 100644 include/asm-sh/sh_mobile_lcdc.h

diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig
index 7072d2c5a040..80fa066416bf 100644
--- a/drivers/video/Kconfig
+++ b/drivers/video/Kconfig
@@ -1866,6 +1866,16 @@ config FB_W100
 
 	  If unsure, say N.
 
+config FB_SH_MOBILE_LCDC
+	tristate "SuperH Mobile LCDC framebuffer support"
+	depends on FB && SUPERH
+	select FB_CFB_FILLRECT
+	select FB_CFB_COPYAREA
+	select FB_CFB_IMAGEBLIT
+	default m
+	---help---
+	  Frame buffer driver for the on-chip SH-Mobile LCD controller.
+
 config FB_S3C2410
 	tristate "S3C2410 LCD framebuffer support"
 	depends on FB && ARCH_S3C2410
diff --git a/drivers/video/Makefile b/drivers/video/Makefile
index 7ee85c0d2e5f..4809f8b9bb21 100644
--- a/drivers/video/Makefile
+++ b/drivers/video/Makefile
@@ -115,6 +115,8 @@ obj-$(CONFIG_FB_IBM_GXT4500)	  += gxt4500.o
 obj-$(CONFIG_FB_PS3)		  += ps3fb.o
 obj-$(CONFIG_FB_SM501)            += sm501fb.o
 obj-$(CONFIG_FB_XILINX)           += xilinxfb.o
+obj-$(CONFIG_FB_SH_MOBILE_LCDC)	  += sh_mobile_lcdcfb.o
+obj-$(CONFIG_FB_SH7343VOU)	  += sh7343_voufb.o
 obj-$(CONFIG_FB_OMAP)             += omap/
 obj-$(CONFIG_XEN_FBDEV_FRONTEND)  += xen-fbfront.o
 obj-$(CONFIG_FB_CARMINE)          += carminefb.o
diff --git a/drivers/video/sh_mobile_lcdcfb.c b/drivers/video/sh_mobile_lcdcfb.c
new file mode 100644
index 000000000000..f6ef6cca73cd
--- /dev/null
+++ b/drivers/video/sh_mobile_lcdcfb.c
@@ -0,0 +1,725 @@
+/*
+ * SuperH Mobile LCDC Framebuffer
+ *
+ * Copyright (c) 2008 Magnus Damm
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ */
+
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/mm.h>
+#include <linux/fb.h>
+#include <linux/clk.h>
+#include <linux/platform_device.h>
+#include <linux/dma-mapping.h>
+#include <asm/sh_mobile_lcdc.h>
+
+#define PALETTE_NR 16
+
+struct sh_mobile_lcdc_priv;
+struct sh_mobile_lcdc_chan {
+	struct sh_mobile_lcdc_priv *lcdc;
+	unsigned long *reg_offs;
+	unsigned long ldmt1r_value;
+	unsigned long enabled; /* ME and SE in LDCNT2R */
+	struct sh_mobile_lcdc_chan_cfg cfg;
+	u32 pseudo_palette[PALETTE_NR];
+	struct fb_info info;
+	dma_addr_t dma_handle;
+};
+
+struct sh_mobile_lcdc_priv {
+	void __iomem *base;
+	struct clk *clk;
+	unsigned long lddckr;
+	struct sh_mobile_lcdc_chan ch[2];
+};
+
+/* shared registers */
+#define _LDDCKR 0x410
+#define _LDDCKSTPR 0x414
+#define _LDINTR 0x468
+#define _LDSR 0x46c
+#define _LDCNT1R 0x470
+#define _LDCNT2R 0x474
+#define _LDDDSR 0x47c
+#define _LDDWD0R 0x800
+#define _LDDRDR 0x840
+#define _LDDWAR 0x900
+#define _LDDRAR 0x904
+
+/* per-channel registers */
+enum { LDDCKPAT1R, LDDCKPAT2R, LDMT1R, LDMT2R, LDMT3R, LDDFR, LDSM1R,
+       LDSA1R, LDMLSR, LDHCNR, LDHSYNR, LDVLNR, LDVSYNR, LDPMR };
+
+static unsigned long lcdc_offs_mainlcd[] = {
+	[LDDCKPAT1R] = 0x400,
+	[LDDCKPAT2R] = 0x404,
+	[LDMT1R] = 0x418,
+	[LDMT2R] = 0x41c,
+	[LDMT3R] = 0x420,
+	[LDDFR] = 0x424,
+	[LDSM1R] = 0x428,
+	[LDSA1R] = 0x430,
+	[LDMLSR] = 0x438,
+	[LDHCNR] = 0x448,
+	[LDHSYNR] = 0x44c,
+	[LDVLNR] = 0x450,
+	[LDVSYNR] = 0x454,
+	[LDPMR] = 0x460,
+};
+
+static unsigned long lcdc_offs_sublcd[] = {
+	[LDDCKPAT1R] = 0x408,
+	[LDDCKPAT2R] = 0x40c,
+	[LDMT1R] = 0x600,
+	[LDMT2R] = 0x604,
+	[LDMT3R] = 0x608,
+	[LDDFR] = 0x60c,
+	[LDSM1R] = 0x610,
+	[LDSA1R] = 0x618,
+	[LDMLSR] = 0x620,
+	[LDHCNR] = 0x624,
+	[LDHSYNR] = 0x628,
+	[LDVLNR] = 0x62c,
+	[LDVSYNR] = 0x630,
+	[LDPMR] = 0x63c,
+};
+
+#define START_LCDC	0x00000001
+#define LCDC_RESET	0x00000100
+#define DISPLAY_BEU	0x00000008
+#define LCDC_ENABLE	0x00000001
+
+static void lcdc_write_chan(struct sh_mobile_lcdc_chan *chan,
+			    int reg_nr, unsigned long data)
+{
+	iowrite32(data, chan->lcdc->base + chan->reg_offs[reg_nr]);
+}
+
+static unsigned long lcdc_read_chan(struct sh_mobile_lcdc_chan *chan,
+				    int reg_nr)
+{
+	return ioread32(chan->lcdc->base + chan->reg_offs[reg_nr]);
+}
+
+static void lcdc_write(struct sh_mobile_lcdc_priv *priv,
+		       unsigned long reg_offs, unsigned long data)
+{
+	iowrite32(data, priv->base + reg_offs);
+}
+
+static unsigned long lcdc_read(struct sh_mobile_lcdc_priv *priv,
+			       unsigned long reg_offs)
+{
+	return ioread32(priv->base + reg_offs);
+}
+
+static void lcdc_wait_bit(struct sh_mobile_lcdc_priv *priv,
+			  unsigned long reg_offs,
+			  unsigned long mask, unsigned long until)
+{
+	while ((lcdc_read(priv, reg_offs) & mask) != until)
+		cpu_relax();
+}
+
+static int lcdc_chan_is_sublcd(struct sh_mobile_lcdc_chan *chan)
+{
+	return chan->cfg.chan == LCDC_CHAN_SUBLCD;
+}
+
+static void lcdc_sys_write_index(void *handle, unsigned long data)
+{
+	struct sh_mobile_lcdc_chan *ch = handle;
+
+	lcdc_write(ch->lcdc, _LDDWD0R, data | 0x10000000);
+	lcdc_wait_bit(ch->lcdc, _LDSR, 2, 0);
+	lcdc_write(ch->lcdc, _LDDWAR, 1 | (lcdc_chan_is_sublcd(ch) ? 2 : 0));
+}
+
+static void lcdc_sys_write_data(void *handle, unsigned long data)
+{
+	struct sh_mobile_lcdc_chan *ch = handle;
+
+	lcdc_write(ch->lcdc, _LDDWD0R, data | 0x11000000);
+	lcdc_wait_bit(ch->lcdc, _LDSR, 2, 0);
+	lcdc_write(ch->lcdc, _LDDWAR, 1 | (lcdc_chan_is_sublcd(ch) ? 2 : 0));
+}
+
+static unsigned long lcdc_sys_read_data(void *handle)
+{
+	struct sh_mobile_lcdc_chan *ch = handle;
+
+	lcdc_write(ch->lcdc, _LDDRDR, 0x01000000);
+	lcdc_wait_bit(ch->lcdc, _LDSR, 2, 0);
+	lcdc_write(ch->lcdc, _LDDRAR, 1 | (lcdc_chan_is_sublcd(ch) ? 2 : 0));
+	udelay(1);
+
+	return lcdc_read(ch->lcdc, _LDDRDR) & 0xffff;
+}
+
+struct sh_mobile_lcdc_sys_bus_ops sh_mobile_lcdc_sys_bus_ops = {
+	lcdc_sys_write_index,
+	lcdc_sys_write_data,
+	lcdc_sys_read_data,
+};
+
+static void sh_mobile_lcdc_start_stop(struct sh_mobile_lcdc_priv *priv,
+				      int start)
+{
+	unsigned long tmp = lcdc_read(priv, _LDCNT2R);
+	int k;
+
+	/* start or stop the lcdc */
+	if (start)
+		lcdc_write(priv, _LDCNT2R, tmp | START_LCDC);
+	else
+		lcdc_write(priv, _LDCNT2R, tmp & ~START_LCDC);
+
+	/* wait until power is applied/stopped on all channels */
+	for (k = 0; k < ARRAY_SIZE(priv->ch); k++)
+		if (lcdc_read(priv, _LDCNT2R) & priv->ch[k].enabled)
+			while (1) {
+				tmp = lcdc_read_chan(&priv->ch[k], LDPMR) & 3;
+				if (start && tmp == 3)
+					break;
+				if (!start && tmp == 0)
+					break;
+				cpu_relax();
+			}
+
+	if (!start)
+		lcdc_write(priv, _LDDCKSTPR, 1); /* stop dotclock */
+}
+
+static int sh_mobile_lcdc_start(struct sh_mobile_lcdc_priv *priv)
+{
+	struct sh_mobile_lcdc_chan *ch;
+	struct fb_videomode *lcd_cfg;
+	struct sh_mobile_lcdc_board_cfg	*board_cfg;
+	unsigned long tmp;
+	int k, m;
+	int ret = 0;
+
+	/* reset */
+	lcdc_write(priv, _LDCNT2R, lcdc_read(priv, _LDCNT2R) | LCDC_RESET);
+	lcdc_wait_bit(priv, _LDCNT2R, LCDC_RESET, 0);
+
+	/* enable LCDC channels */
+	tmp = lcdc_read(priv, _LDCNT2R);
+	tmp |= priv->ch[0].enabled;
+	tmp |= priv->ch[1].enabled;
+	lcdc_write(priv, _LDCNT2R, tmp);
+
+	/* read data from external memory, avoid using the BEU for now */
+	lcdc_write(priv, _LDCNT2R, lcdc_read(priv, _LDCNT2R) & ~DISPLAY_BEU);
+
+	/* stop the lcdc first */
+	sh_mobile_lcdc_start_stop(priv, 0);
+
+	/* configure clocks */
+	tmp = priv->lddckr;
+	for (k = 0; k < ARRAY_SIZE(priv->ch); k++) {
+		ch = &priv->ch[k];
+
+		if (!priv->ch[k].enabled)
+			continue;
+
+		m = ch->cfg.clock_divider;
+		if (!m)
+			continue;
+
+		if (m == 1)
+			m = 1 << 6;
+		tmp |= m << (lcdc_chan_is_sublcd(ch) ? 8 : 0);
+
+		lcdc_write_chan(ch, LDDCKPAT1R, 0x00000000);
+		lcdc_write_chan(ch, LDDCKPAT2R, (1 << (m/2)) - 1);
+	}
+
+	lcdc_write(priv, _LDDCKR, tmp);
+
+	/* start dotclock again */
+	lcdc_write(priv, _LDDCKSTPR, 0);
+	lcdc_wait_bit(priv, _LDDCKSTPR, ~0, 0);
+
+	/* interrupts are disabled */
+	lcdc_write(priv, _LDINTR, 0);
+
+	for (k = 0; k < ARRAY_SIZE(priv->ch); k++) {
+		ch = &priv->ch[k];
+		lcd_cfg = &ch->cfg.lcd_cfg;
+
+		if (!ch->enabled)
+			continue;
+
+		tmp = ch->ldmt1r_value;
+		tmp |= (lcd_cfg->sync & FB_SYNC_VERT_HIGH_ACT) ? 0 : 1 << 28;
+		tmp |= (lcd_cfg->sync & FB_SYNC_HOR_HIGH_ACT) ? 0 : 1 << 27;
+		lcdc_write_chan(ch, LDMT1R, tmp);
+
+		/* setup SYS bus */
+		lcdc_write_chan(ch, LDMT2R, ch->cfg.sys_bus_cfg.ldmt2r);
+		lcdc_write_chan(ch, LDMT3R, ch->cfg.sys_bus_cfg.ldmt3r);
+
+		/* horizontal configuration */
+		tmp = lcd_cfg->xres + lcd_cfg->hsync_len;
+		tmp += lcd_cfg->left_margin;
+		tmp += lcd_cfg->right_margin;
+		tmp /= 8; /* HTCN */
+		tmp |= (lcd_cfg->xres / 8) << 16; /* HDCN */
+		lcdc_write_chan(ch, LDHCNR, tmp);
+
+		tmp = lcd_cfg->xres;
+		tmp += lcd_cfg->right_margin;
+		tmp /= 8; /* HSYNP */
+		tmp |= (lcd_cfg->hsync_len / 8) << 16; /* HSYNW */
+		lcdc_write_chan(ch, LDHSYNR, tmp);
+
+		/* power supply */
+		lcdc_write_chan(ch, LDPMR, 0);
+
+		/* vertical configuration */
+		tmp = lcd_cfg->yres + lcd_cfg->vsync_len;
+		tmp += lcd_cfg->upper_margin;
+		tmp += lcd_cfg->lower_margin; /* VTLN */
+		tmp |= lcd_cfg->yres << 16; /* VDLN */
+		lcdc_write_chan(ch, LDVLNR, tmp);
+
+		tmp = lcd_cfg->yres;
+		tmp += lcd_cfg->lower_margin; /* VSYNP */
+		tmp |= lcd_cfg->vsync_len << 16; /* VSYNW */
+		lcdc_write_chan(ch, LDVSYNR, tmp);
+
+		board_cfg = &ch->cfg.board_cfg;
+		if (board_cfg->setup_sys)
+			ret = board_cfg->setup_sys(board_cfg->board_data, ch,
+						   &sh_mobile_lcdc_sys_bus_ops);
+		if (ret)
+			return ret;
+	}
+
+	/* --- display_lcdc_data() --- */
+	lcdc_write(priv, _LDINTR, 0x00000f00);
+
+	/* word and long word swap */
+	lcdc_write(priv, _LDDDSR, lcdc_read(priv, _LDDDSR) | 6);
+
+	for (k = 0; k < ARRAY_SIZE(priv->ch); k++) {
+		ch = &priv->ch[k];
+
+		if (!priv->ch[k].enabled)
+			continue;
+
+		/* set bpp format in PKF[4:0] */
+		tmp = lcdc_read_chan(ch, LDDFR);
+		tmp &= ~(0x0001001f);
+		tmp |= (priv->ch[k].info.var.bits_per_pixel == 16) ? 3 : 0;
+		lcdc_write_chan(ch, LDDFR, tmp);
+
+		/* point out our frame buffer */
+		lcdc_write_chan(ch, LDSA1R, ch->info.fix.smem_start);
+
+		/* set line size */
+		lcdc_write_chan(ch, LDMLSR, ch->info.fix.line_length);
+
+		/* continuous read mode */
+		lcdc_write_chan(ch, LDSM1R, 0);
+	}
+
+	/* display output */
+	lcdc_write(priv, _LDCNT1R, LCDC_ENABLE);
+
+	/* start the lcdc */
+	sh_mobile_lcdc_start_stop(priv, 1);
+
+	/* tell the board code to enable the panel */
+	for (k = 0; k < ARRAY_SIZE(priv->ch); k++) {
+		ch = &priv->ch[k];
+		board_cfg = &ch->cfg.board_cfg;
+		if (board_cfg->display_on)
+			board_cfg->display_on(board_cfg->board_data);
+	}
+
+	return 0;
+}
+
+static void sh_mobile_lcdc_stop(struct sh_mobile_lcdc_priv *priv)
+{
+	struct sh_mobile_lcdc_chan *ch;
+	struct sh_mobile_lcdc_board_cfg	*board_cfg;
+	int k;
+
+	/* tell the board code to disable the panel */
+	for (k = 0; k < ARRAY_SIZE(priv->ch); k++) {
+		ch = &priv->ch[k];
+		board_cfg = &ch->cfg.board_cfg;
+		if (board_cfg->display_off)
+			board_cfg->display_off(board_cfg->board_data);
+	}
+
+	/* stop the lcdc */
+	sh_mobile_lcdc_start_stop(priv, 0);
+}
+
+static int sh_mobile_lcdc_check_interface(struct sh_mobile_lcdc_chan *ch)
+{
+	int ifm, miftyp;
+
+	switch (ch->cfg.interface_type) {
+	case RGB8: ifm = 0; miftyp = 0; break;
+	case RGB9: ifm = 0; miftyp = 4; break;
+	case RGB12A: ifm = 0; miftyp = 5; break;
+	case RGB12B: ifm = 0; miftyp = 6; break;
+	case RGB16: ifm = 0; miftyp = 7; break;
+	case RGB18: ifm = 0; miftyp = 10; break;
+	case RGB24: ifm = 0; miftyp = 11; break;
+	case SYS8A: ifm = 1; miftyp = 0; break;
+	case SYS8B: ifm = 1; miftyp = 1; break;
+	case SYS8C: ifm = 1; miftyp = 2; break;
+	case SYS8D: ifm = 1; miftyp = 3; break;
+	case SYS9: ifm = 1; miftyp = 4; break;
+	case SYS12: ifm = 1; miftyp = 5; break;
+	case SYS16A: ifm = 1; miftyp = 7; break;
+	case SYS16B: ifm = 1; miftyp = 8; break;
+	case SYS16C: ifm = 1; miftyp = 9; break;
+	case SYS18: ifm = 1; miftyp = 10; break;
+	case SYS24: ifm = 1; miftyp = 11; break;
+	default: goto bad;
+	}
+
+	/* SUBLCD only supports SYS interface */
+	if (lcdc_chan_is_sublcd(ch)) {
+		if (ifm == 0)
+			goto bad;
+		else
+			ifm = 0;
+	}
+
+	ch->ldmt1r_value = (ifm << 12) | miftyp;
+	return 0;
+ bad:
+	return -EINVAL;
+}
+
+static int sh_mobile_lcdc_setup_clocks(struct device *dev, int clock_source,
+				       struct sh_mobile_lcdc_priv *priv)
+{
+	char *str;
+	int icksel;
+
+	switch (clock_source) {
+	case LCDC_CLK_BUS: str = "bus_clk"; icksel = 0; break;
+	case LCDC_CLK_PERIPHERAL: str = "peripheral_clk"; icksel = 1; break;
+	case LCDC_CLK_EXTERNAL: str = NULL; icksel = 2; break;
+	default:
+		return -EINVAL;
+	}
+
+	priv->lddckr = icksel << 16;
+
+	if (str) {
+		priv->clk = clk_get(dev, str);
+		if (IS_ERR(priv->clk)) {
+			dev_err(dev, "cannot get clock %s\n", str);
+			return PTR_ERR(priv->clk);
+		}
+
+		clk_enable(priv->clk);
+	}
+
+	return 0;
+}
+
+static int sh_mobile_lcdc_setcolreg(u_int regno,
+				    u_int red, u_int green, u_int blue,
+				    u_int transp, struct fb_info *info)
+{
+	u32 *palette = info->pseudo_palette;
+
+	if (regno >= PALETTE_NR)
+		return -EINVAL;
+
+	/* only FB_VISUAL_TRUECOLOR supported */
+
+	red >>= 16 - info->var.red.length;
+	green >>= 16 - info->var.green.length;
+	blue >>= 16 - info->var.blue.length;
+	transp >>= 16 - info->var.transp.length;
+
+	palette[regno] = (red << info->var.red.offset) |
+	  (green << info->var.green.offset) |
+	  (blue << info->var.blue.offset) |
+	  (transp << info->var.transp.offset);
+
+	return 0;
+}
+
+static struct fb_fix_screeninfo sh_mobile_lcdc_fix  = {
+	.id =		"SH Mobile LCDC",
+	.type =		FB_TYPE_PACKED_PIXELS,
+	.visual =	FB_VISUAL_TRUECOLOR,
+	.accel =	FB_ACCEL_NONE,
+};
+
+static struct fb_ops sh_mobile_lcdc_ops = {
+	.fb_setcolreg	= sh_mobile_lcdc_setcolreg,
+	.fb_fillrect	= cfb_fillrect,
+	.fb_copyarea	= cfb_copyarea,
+	.fb_imageblit	= cfb_imageblit,
+};
+
+static int sh_mobile_lcdc_set_bpp(struct fb_var_screeninfo *var, int bpp)
+{
+	switch (bpp) {
+	case 16: /* PKF[4:0] = 00011 - RGB 565 */
+		var->red.offset = 11;
+		var->red.length = 5;
+		var->green.offset = 5;
+		var->green.length = 6;
+		var->blue.offset = 0;
+		var->blue.length = 5;
+		var->transp.offset = 0;
+		var->transp.length = 0;
+		break;
+
+	case 32: /* PKF[4:0] = 00000 - RGB 888
+		  * sh7722 pdf says 00RRGGBB but reality is GGBB00RR
+		  * this may be because LDDDSR has word swap enabled..
+		  */
+		var->red.offset = 0;
+		var->red.length = 8;
+		var->green.offset = 24;
+		var->green.length = 8;
+		var->blue.offset = 16;
+		var->blue.length = 8;
+		var->transp.offset = 0;
+		var->transp.length = 0;
+		break;
+	default:
+		return -EINVAL;
+	}
+	var->bits_per_pixel = bpp;
+	var->red.msb_right = 0;
+	var->green.msb_right = 0;
+	var->blue.msb_right = 0;
+	var->transp.msb_right = 0;
+	return 0;
+}
+
+static int sh_mobile_lcdc_remove(struct platform_device *pdev);
+
+static int __init sh_mobile_lcdc_probe(struct platform_device *pdev)
+{
+	struct fb_info *info;
+	struct sh_mobile_lcdc_priv *priv;
+	struct sh_mobile_lcdc_info *pdata;
+	struct sh_mobile_lcdc_chan_cfg *cfg;
+	struct resource *res;
+	int error;
+	void *buf;
+	int i, j;
+
+	if (!pdev->dev.platform_data) {
+		dev_err(&pdev->dev, "no platform data defined\n");
+		error = -EINVAL;
+		goto err0;
+	}
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (res == NULL) {
+		dev_err(&pdev->dev, "cannot find IO resource\n");
+		error = -ENOENT;
+		goto err0;
+	}
+
+	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+	if (!priv) {
+		dev_err(&pdev->dev, "cannot allocate device data\n");
+		error = -ENOMEM;
+		goto err0;
+	}
+
+	platform_set_drvdata(pdev, priv);
+	pdata = pdev->dev.platform_data;
+
+	j = 0;
+	for (i = 0; i < ARRAY_SIZE(pdata->ch); i++) {
+		priv->ch[j].lcdc = priv;
+		memcpy(&priv->ch[j].cfg, &pdata->ch[i], sizeof(pdata->ch[i]));
+
+		error = sh_mobile_lcdc_check_interface(&priv->ch[i]);
+		if (error) {
+			dev_err(&pdev->dev, "unsupported interface type\n");
+			goto err1;
+		}
+
+		switch (pdata->ch[i].chan) {
+		case LCDC_CHAN_MAINLCD:
+			priv->ch[j].enabled = 1 << 1;
+			priv->ch[j].reg_offs = lcdc_offs_mainlcd;
+			j++;
+			break;
+		case LCDC_CHAN_SUBLCD:
+			priv->ch[j].enabled = 1 << 2;
+			priv->ch[j].reg_offs = lcdc_offs_sublcd;
+			j++;
+			break;
+		}
+	}
+
+	if (!j) {
+		dev_err(&pdev->dev, "no channels defined\n");
+		error = -EINVAL;
+		goto err1;
+	}
+
+	error = sh_mobile_lcdc_setup_clocks(&pdev->dev,
+					    pdata->clock_source, priv);
+	if (error) {
+		dev_err(&pdev->dev, "unable to setup clocks\n");
+		goto err1;
+	}
+
+	priv->lddckr = pdata->lddckr;
+	priv->base = ioremap_nocache(res->start, (res->end - res->start) + 1);
+
+	for (i = 0; i < j; i++) {
+		info = &priv->ch[i].info;
+		cfg = &priv->ch[i].cfg;
+
+		info->fbops = &sh_mobile_lcdc_ops;
+		info->var.xres = info->var.xres_virtual = cfg->lcd_cfg.xres;
+		info->var.yres = info->var.yres_virtual = cfg->lcd_cfg.yres;
+		info->var.activate = FB_ACTIVATE_NOW;
+		error = sh_mobile_lcdc_set_bpp(&info->var, cfg->bpp);
+		if (error)
+			break;
+
+		info->fix = sh_mobile_lcdc_fix;
+		info->fix.line_length = cfg->lcd_cfg.xres * (cfg->bpp / 8);
+		info->fix.smem_len = info->fix.line_length * cfg->lcd_cfg.yres;
+
+		buf = dma_alloc_coherent(&pdev->dev, info->fix.smem_len,
+					 &priv->ch[i].dma_handle, GFP_KERNEL);
+		if (!buf) {
+			dev_err(&pdev->dev, "unable to allocate buffer\n");
+			error = -ENOMEM;
+			break;
+		}
+
+		info->pseudo_palette = &priv->ch[i].pseudo_palette;
+		info->flags = FBINFO_FLAG_DEFAULT;
+
+		error = fb_alloc_cmap(&info->cmap, PALETTE_NR, 0);
+		if (error < 0) {
+			dev_err(&pdev->dev, "unable to allocate cmap\n");
+			dma_free_coherent(&pdev->dev, info->fix.smem_len,
+					  buf, priv->ch[i].dma_handle);
+			break;
+		}
+
+		memset(buf, 0, info->fix.smem_len);
+		info->fix.smem_start = priv->ch[i].dma_handle;
+		info->screen_base = buf;
+		info->device = &pdev->dev;
+	}
+
+	if (error)
+		goto err1;
+
+	error = sh_mobile_lcdc_start(priv);
+	if (error) {
+		dev_err(&pdev->dev, "unable to start hardware\n");
+		goto err1;
+	}
+
+	for (i = 0; i < j; i++) {
+		error = register_framebuffer(&priv->ch[i].info);
+		if (error < 0)
+			goto err1;
+	}
+
+	for (i = 0; i < j; i++) {
+		info = &priv->ch[i].info;
+		dev_info(info->dev,
+			 "registered %s/%s as %dx%d %dbpp.\n",
+			 pdev->name,
+			 (priv->ch[i].cfg.chan == LCDC_CHAN_MAINLCD) ?
+			 "mainlcd" : "sublcd",
+			 (int) priv->ch[i].cfg.lcd_cfg.xres,
+			 (int) priv->ch[i].cfg.lcd_cfg.yres,
+			 priv->ch[i].cfg.bpp);
+	}
+
+	return 0;
+ err1:
+	sh_mobile_lcdc_remove(pdev);
+ err0:
+	return error;
+}
+
+static int sh_mobile_lcdc_remove(struct platform_device *pdev)
+{
+	struct sh_mobile_lcdc_priv *priv = platform_get_drvdata(pdev);
+	struct fb_info *info;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(priv->ch); i++)
+		if (priv->ch[i].info.dev)
+			unregister_framebuffer(&priv->ch[i].info);
+
+	sh_mobile_lcdc_stop(priv);
+
+	for (i = 0; i < ARRAY_SIZE(priv->ch); i++) {
+		info = &priv->ch[i].info;
+
+		if (!info->device)
+			continue;
+
+		dma_free_coherent(&pdev->dev, info->fix.smem_len,
+				  info->screen_base, priv->ch[i].dma_handle);
+		fb_dealloc_cmap(&info->cmap);
+	}
+
+	if (priv->clk) {
+		clk_disable(priv->clk);
+		clk_put(priv->clk);
+	}
+
+	if (priv->base)
+		iounmap(priv->base);
+
+	kfree(priv);
+	return 0;
+}
+
+static struct platform_driver sh_mobile_lcdc_driver = {
+	.driver		= {
+		.name		= "sh_mobile_lcdc_fb",
+		.owner		= THIS_MODULE,
+	},
+	.probe		= sh_mobile_lcdc_probe,
+	.remove		= sh_mobile_lcdc_remove,
+};
+
+static int __init sh_mobile_lcdc_init(void)
+{
+	return platform_driver_register(&sh_mobile_lcdc_driver);
+}
+
+static void __exit sh_mobile_lcdc_exit(void)
+{
+	platform_driver_unregister(&sh_mobile_lcdc_driver);
+}
+
+module_init(sh_mobile_lcdc_init);
+module_exit(sh_mobile_lcdc_exit);
+
+MODULE_DESCRIPTION("SuperH Mobile LCDC Framebuffer driver");
+MODULE_AUTHOR("Magnus Damm <damm@opensource.se>");
+MODULE_LICENSE("GPL v2");
diff --git a/include/asm-sh/sh_mobile_lcdc.h b/include/asm-sh/sh_mobile_lcdc.h
new file mode 100644
index 000000000000..27677727df4d
--- /dev/null
+++ b/include/asm-sh/sh_mobile_lcdc.h
@@ -0,0 +1,66 @@
+#ifndef __ASM_SH_MOBILE_LCDC_H__
+#define __ASM_SH_MOBILE_LCDC_H__
+
+#include <linux/fb.h>
+
+enum { RGB8,   /* 24bpp, 8:8:8 */
+       RGB9,   /* 18bpp, 9:9 */
+       RGB12A, /* 24bpp, 12:12 */
+       RGB12B, /* 12bpp */
+       RGB16,  /* 16bpp */
+       RGB18,  /* 18bpp */
+       RGB24,  /* 24bpp */
+       SYS8A,  /* 24bpp, 8:8:8 */
+       SYS8B,  /* 18bpp, 8:8:2 */
+       SYS8C,  /* 18bpp, 2:8:8 */
+       SYS8D,  /* 16bpp, 8:8 */
+       SYS9,   /* 18bpp, 9:9 */
+       SYS12,  /* 24bpp, 12:12 */
+       SYS16A, /* 16bpp */
+       SYS16B, /* 18bpp, 16:2 */
+       SYS16C, /* 18bpp, 2:16 */
+       SYS18,  /* 18bpp */
+       SYS24 };/* 24bpp */
+
+enum { LCDC_CHAN_DISABLED = 0,
+       LCDC_CHAN_MAINLCD,
+       LCDC_CHAN_SUBLCD };
+
+enum { LCDC_CLK_BUS, LCDC_CLK_PERIPHERAL, LCDC_CLK_EXTERNAL };
+
+struct sh_mobile_lcdc_sys_bus_cfg {
+	unsigned long ldmt2r;
+	unsigned long ldmt3r;
+};
+
+struct sh_mobile_lcdc_sys_bus_ops {
+	void (*write_index)(void *handle, unsigned long data);
+	void (*write_data)(void *handle, unsigned long data);
+	unsigned long (*read_data)(void *handle);
+};
+
+struct sh_mobile_lcdc_board_cfg {
+	void *board_data;
+	int (*setup_sys)(void *board_data, void *sys_ops_handle,
+			 struct sh_mobile_lcdc_sys_bus_ops *sys_ops);
+	void (*display_on)(void *board_data);
+	void (*display_off)(void *board_data);
+};
+
+struct sh_mobile_lcdc_chan_cfg {
+	int chan;
+	int bpp;
+	int interface_type; /* selects RGBn or SYSn I/F, see above */
+	int clock_divider;
+	struct fb_videomode lcd_cfg;
+	struct sh_mobile_lcdc_board_cfg board_cfg;
+	struct sh_mobile_lcdc_sys_bus_cfg sys_bus_cfg; /* only for SYSn I/F */
+};
+
+struct sh_mobile_lcdc_info {
+	unsigned long lddckr;
+	int clock_source;
+	struct sh_mobile_lcdc_chan_cfg ch[2];
+};
+
+#endif /* __ASM_SH_MOBILE_LCDC_H__ */

From b604838ac6d233fd6bffc0e758a818133a01ff22 Mon Sep 17 00:00:00 2001
From: Frans Pop <elendil@planet.nl>
Date: Wed, 23 Jul 2008 21:31:26 -0700
Subject: [PATCH 0490/1171] vfb: only enable if explicitly requested when
 compiled in

The Kconfig help for the vfb driver says:
   Do NOT enable it for normal systems! To protect the innocent, it
   has to be enabled explicitly at boot time using the kernel option
   `video=vfb:'.

This change lets the code match the description.

Support for vfb:disable is kept for backwards compatibility; vfb:off works
because it is tested at a higher level.

Note: any undefined option (e.g. vfb:enable) will also enable this driver.

The relevant code has been unchanged since before the migration to
git (2.6.12).

This patch fixes bugzilla #9310 and was the root cause behind
http://lkml.org/lkml/2008/5/31/220.

Signed-off-by: Frans Pop <elendil@planet.nl>
Cc: Antonino A. Daplas <adaplas@gmail.com>
Acked-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/vfb.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)

diff --git a/drivers/video/vfb.c b/drivers/video/vfb.c
index 072638a9528a..93fe08d6c78f 100644
--- a/drivers/video/vfb.c
+++ b/drivers/video/vfb.c
@@ -443,19 +443,29 @@ static int vfb_mmap(struct fb_info *info,
 }
 
 #ifndef MODULE
+/*
+ * The virtual framebuffer driver is only enabled if explicitly
+ * requested by passing 'video=vfb:' (or any actual options).
+ */
 static int __init vfb_setup(char *options)
 {
 	char *this_opt;
 
+	vfb_enable = 0;
+
+	if (!options)
+		return 1;
+
 	vfb_enable = 1;
 
-	if (!options || !*options)
+	if (!*options)
 		return 1;
 
 	while ((this_opt = strsep(&options, ",")) != NULL) {
 		if (!*this_opt)
 			continue;
-		if (!strncmp(this_opt, "disable", 7))
+		/* Test disable for backwards compatibility */
+		if (!strcmp(this_opt, "disable"))
 			vfb_enable = 0;
 	}
 	return 1;

From 2870086e9f2032bdd95b8da9bd187e3c16fc6d49 Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:26 -0700
Subject: [PATCH 0491/1171] hgafb: convert to new platform driver API

Convert the hgafb driver to use new platform driver API.

Addresses http://bugzilla.kernel.org/show_bug.cgi?id=9689

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Cc: Anton Vorontsov <avorontsov@ru.mvista.com>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/hgafb.c | 36 +++++++++++++++++++-----------------
 1 file changed, 19 insertions(+), 17 deletions(-)

diff --git a/drivers/video/hgafb.c b/drivers/video/hgafb.c
index c18880d9db1f..0129c044f6d6 100644
--- a/drivers/video/hgafb.c
+++ b/drivers/video/hgafb.c
@@ -551,7 +551,7 @@ static struct fb_ops hgafb_ops = {
 	 *  Initialization
 	 */
 
-static int __init hgafb_probe(struct device *device)
+static int __init hgafb_probe(struct platform_device *pdev)
 {
 	struct fb_info *info;
 
@@ -565,7 +565,7 @@ static int __init hgafb_probe(struct device *device)
 	printk(KERN_INFO "hgafb: %s with %ldK of memory detected.\n",
 		hga_type_name, hga_vram_len/1024);
 
-	info = framebuffer_alloc(0, NULL);
+	info = framebuffer_alloc(0, &pdev->dev);
 	if (!info) {
 		iounmap(hga_vram);
 		return -ENOMEM;
@@ -593,13 +593,13 @@ static int __init hgafb_probe(struct device *device)
 
         printk(KERN_INFO "fb%d: %s frame buffer device\n",
                info->node, info->fix.id);
-	dev_set_drvdata(device, info);
+	platform_set_drvdata(pdev, info);
 	return 0;
 }
 
-static int hgafb_remove(struct device *device)
+static int hgafb_remove(struct platform_device *pdev)
 {
-	struct fb_info *info = dev_get_drvdata(device);
+	struct fb_info *info = platform_get_drvdata(pdev);
 
 	hga_txt_mode();
 	hga_clear_screen();
@@ -620,16 +620,15 @@ static int hgafb_remove(struct device *device)
 	return 0;
 }
 
-static struct device_driver hgafb_driver = {
-	.name = "hgafb",
-	.bus  = &platform_bus_type,
+static struct platform_driver hgafb_driver = {
 	.probe = hgafb_probe,
 	.remove = hgafb_remove,
+	.driver = {
+		.name = "hgafb",
+	},
 };
 
-static struct platform_device hgafb_device = {
-	.name = "hgafb",
-};
+static struct platform_device *hgafb_device;
 
 static int __init hgafb_init(void)
 {
@@ -638,12 +637,15 @@ static int __init hgafb_init(void)
 	if (fb_get_options("hgafb", NULL))
 		return -ENODEV;
 
-	ret = driver_register(&hgafb_driver);
+	ret = platform_driver_register(&hgafb_driver);
 
 	if (!ret) {
-		ret = platform_device_register(&hgafb_device);
-		if (ret)
-			driver_unregister(&hgafb_driver);
+		hgafb_device = platform_device_register_simple("hgafb", 0, NULL, 0);
+
+		if (IS_ERR(hgafb_device)) {
+			platform_driver_unregister(&hgafb_driver);
+			ret = PTR_ERR(hgafb_device);
+		}
 	}
 
 	return ret;
@@ -651,8 +653,8 @@ static int __init hgafb_init(void)
 
 static void __exit hgafb_exit(void)
 {
-	platform_device_unregister(&hgafb_device);
-	driver_unregister(&hgafb_driver);
+	platform_device_unregister(hgafb_device);
+	platform_driver_unregister(&hgafb_driver);
 }
 
 /* -------------------------------------------------------------------------

From 43a3abc6aca8505e708508e2c7c2f99a7f8f820b Mon Sep 17 00:00:00 2001
From: Ville Syrjala <syrjala@sci.fi>
Date: Wed, 23 Jul 2008 21:31:27 -0700
Subject: [PATCH 0492/1171] fbdev: width and height are unsigned

The width and height members of fb_var_screeninfo are __u32.  The code
initializes them to -1 which seems wrong, and 0 seems like an equally good
default value.

Signed-off-by: Ville Syrjala <syrjala@sci.fi>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/fbmon.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/video/fbmon.c b/drivers/video/fbmon.c
index 052e18058498..6a0aa180c266 100644
--- a/drivers/video/fbmon.c
+++ b/drivers/video/fbmon.c
@@ -879,7 +879,7 @@ int fb_parse_edid(unsigned char *edid, struct fb_var_screeninfo *var)
 		if (edid_is_timing_block(block)) {
 			var->xres = var->xres_virtual = H_ACTIVE;
 			var->yres = var->yres_virtual = V_ACTIVE;
-			var->height = var->width = -1;
+			var->height = var->width = 0;
 			var->right_margin = H_SYNC_OFFSET;
 			var->left_margin = (H_ACTIVE + H_BLANKING) -
 				(H_ACTIVE + H_SYNC_OFFSET + H_SYNC_WIDTH);

From 7572a1ea034a8fc45e57de28cc7573264975532a Mon Sep 17 00:00:00 2001
From: Ville Syrjala <syrjala@sci.fi>
Date: Wed, 23 Jul 2008 21:31:28 -0700
Subject: [PATCH 0493/1171] fbdev: xoffset, yoffset and yres are unsigned

The xoffset, yoffset and yres members of fb_var_screeninfo are __u32.
Make them unsigned in the code as well.

Signed-off-by: Ville Syrjala <syrjala@sci.fi>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/fbmem.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c
index 1cd5071e5362..6b487801eeae 100644
--- a/drivers/video/fbmem.c
+++ b/drivers/video/fbmem.c
@@ -848,9 +848,8 @@ int
 fb_pan_display(struct fb_info *info, struct fb_var_screeninfo *var)
 {
 	struct fb_fix_screeninfo *fix = &info->fix;
-        int xoffset = var->xoffset;
-        int yoffset = var->yoffset;
-        int err = 0, yres = info->var.yres;
+	unsigned int yres = info->var.yres;
+	int err = 0;
 
 	if (var->yoffset > 0) {
 		if (var->vmode & FB_VMODE_YWRAP) {
@@ -866,8 +865,8 @@ fb_pan_display(struct fb_info *info, struct fb_var_screeninfo *var)
 				 (var->xoffset % fix->xpanstep)))
 		err = -EINVAL;
 
-        if (err || !info->fbops->fb_pan_display || xoffset < 0 ||
-	    yoffset < 0 || var->yoffset + yres > info->var.yres_virtual ||
+	if (err || !info->fbops->fb_pan_display ||
+	    var->yoffset + yres > info->var.yres_virtual ||
 	    var->xoffset + info->var.xres > info->var.xres_virtual)
 		return -EINVAL;
 

From 50cd0221c9062ec5dac8a3620f36f568df052ac1 Mon Sep 17 00:00:00 2001
From: Olaf Hering <olaf@aepfle.de>
Date: Wed, 23 Jul 2008 21:31:29 -0700
Subject: [PATCH 0494/1171] atyfb: remove dead code

Remove dead code.  This will slightly change the behaviour of the driver
on systems that support backlight control.  Previously they would just
turn the backlight off using the backlight control but now the generic LCD
code will also turn off the LCD using the POWER_MANAGEMENT register.

Signed-off-by: Olaf Hering <olaf@aepfle.de>
Signed-off-by: Ville Syrjala <syrjala@sci.fi>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/aty/atyfb_base.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/video/aty/atyfb_base.c b/drivers/video/aty/atyfb_base.c
index bd4ac0bafecb..d7ba3f1a4d80 100644
--- a/drivers/video/aty/atyfb_base.c
+++ b/drivers/video/aty/atyfb_base.c
@@ -2709,8 +2709,7 @@ static int atyfb_blank(int blank, struct fb_info *info)
 	if (par->lock_blank || par->asleep)
 		return 0;
 
-#ifdef CONFIG_FB_ATY_BACKLIGHT
-#elif defined(CONFIG_FB_ATY_GENERIC_LCD)
+#ifdef CONFIG_FB_ATY_GENERIC_LCD
 	if (par->lcd_table && blank > FB_BLANK_NORMAL &&
 	    (aty_ld_lcd(LCD_GEN_CNTL, par) & LCD_ON)) {
 		u32 pm = aty_ld_lcd(POWER_MANAGEMENT, par);
@@ -2739,8 +2738,7 @@ static int atyfb_blank(int blank, struct fb_info *info)
 	}
 	aty_st_le32(CRTC_GEN_CNTL, gen_cntl, par);
 
-#ifdef CONFIG_FB_ATY_BACKLIGHT
-#elif defined(CONFIG_FB_ATY_GENERIC_LCD)
+#ifdef CONFIG_FB_ATY_GENERIC_LCD
 	if (par->lcd_table && blank <= FB_BLANK_NORMAL &&
 	    (aty_ld_lcd(LCD_GEN_CNTL, par) & LCD_ON)) {
 		u32 pm = aty_ld_lcd(POWER_MANAGEMENT, par);

From 3880b0b5297ae9bf58a7662d13a46b5d5f0b2af6 Mon Sep 17 00:00:00 2001
From: Ville Syrjala <syrjala@sci.fi>
Date: Wed, 23 Jul 2008 21:31:30 -0700
Subject: [PATCH 0495/1171] atyfb: correct_chipset() can fail

Atari probe code relies on correct_chipset() failing if the device is not
a mach64 GX/CX.  aty_chips[] array would be indexed with -1 in that case.

Signed-off-by: Ville Syrjala <syrjala@sci.fi>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/aty/atyfb_base.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/video/aty/atyfb_base.c b/drivers/video/aty/atyfb_base.c
index d7ba3f1a4d80..d6903c70bee5 100644
--- a/drivers/video/aty/atyfb_base.c
+++ b/drivers/video/aty/atyfb_base.c
@@ -424,7 +424,6 @@ static struct {
 #endif /* CONFIG_FB_ATY_CT */
 };
 
-/* can not fail */
 static int __devinit correct_chipset(struct atyfb_par *par)
 {
 	u8 rev;
@@ -437,6 +436,9 @@ static int __devinit correct_chipset(struct atyfb_par *par)
 		if (par->pci_id == aty_chips[i].pci_id)
 			break;
 
+	if (i < 0)
+		return -ENODEV;
+
 	name = aty_chips[i].name;
 	par->pll_limits.pll_max = aty_chips[i].pll;
 	par->pll_limits.mclk = aty_chips[i].mclk;

From 6cfafc15994ac2a2377b32b5a65cf62a90a80d49 Mon Sep 17 00:00:00 2001
From: Ville Syrjala <syrjala@sci.fi>
Date: Wed, 23 Jul 2008 21:31:31 -0700
Subject: [PATCH 0496/1171] atyfb: use a PCI device ID table

Convert atyfb to use a PCI device ID table.

Signed-off-by: Ville Syrjala <syrjala@sci.fi>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/aty/atyfb_base.c | 73 +++++++++++++++++++++++++---------
 1 file changed, 55 insertions(+), 18 deletions(-)

diff --git a/drivers/video/aty/atyfb_base.c b/drivers/video/aty/atyfb_base.c
index d6903c70bee5..872760accb93 100644
--- a/drivers/video/aty/atyfb_base.c
+++ b/drivers/video/aty/atyfb_base.c
@@ -3418,14 +3418,7 @@ static int __devinit atyfb_pci_probe(struct pci_dev *pdev, const struct pci_devi
 	struct fb_info *info;
 	struct resource *rp;
 	struct atyfb_par *par;
-	int i, rc = -ENOMEM;
-
-	for (i = ARRAY_SIZE(aty_chips) - 1; i >= 0; i--)
-		if (pdev->device == aty_chips[i].pci_id)
-			break;
-
-	if (i < 0)
-		return -ENODEV;
+	int rc = -ENOMEM;
 
 	/* Enable device in PCI config */
 	if (pci_enable_device(pdev)) {
@@ -3456,7 +3449,7 @@ static int __devinit atyfb_pci_probe(struct pci_dev *pdev, const struct pci_devi
 	par = info->par;
 	info->fix = atyfb_fix;
 	info->device = &pdev->dev;
-	par->pci_id = aty_chips[i].pci_id;
+	par->pci_id = pdev->device;
 	par->res_start = res_start;
 	par->res_size = res_size;
 	par->irq = pdev->irq;
@@ -3655,18 +3648,62 @@ static void __devexit atyfb_pci_remove(struct pci_dev *pdev)
 	atyfb_remove(info);
 }
 
-/*
- * This driver uses its own matching table. That will be more difficult
- * to fix, so for now, we just match against any ATI ID and let the
- * probe() function find out what's up. That also mean we don't have
- * a module ID table though.
- */
 static struct pci_device_id atyfb_pci_tbl[] = {
-	{ PCI_VENDOR_ID_ATI, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID,
-	  PCI_BASE_CLASS_DISPLAY << 16, 0xff0000, 0 },
-	{ 0, }
+#ifdef CONFIG_FB_ATY_GX
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64GX) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64CX) },
+#endif /* CONFIG_FB_ATY_GX */
+
+#ifdef CONFIG_FB_ATY_CT
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64CT) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64ET) },
+
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64LT) },
+
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64VT) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64GT) },
+
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64VU) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64GU) },
+
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64LG) },
+
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64VV) },
+
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64GV) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64GW) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64GY) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64GZ) },
+
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64GB) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64GD) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64GI) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64GP) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64GQ) },
+
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64LB) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64LD) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64LI) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64LP) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64LQ) },
+
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64GM) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64GN) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64GO) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64GL) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64GR) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64GS) },
+
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64LM) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64LN) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64LR) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_CHIP_MACH64LS) },
+#endif /* CONFIG_FB_ATY_CT */
+	{ }
 };
 
+MODULE_DEVICE_TABLE(pci, atyfb_pci_tbl);
+
 static struct pci_driver atyfb_driver = {
 	.name		= "atyfb",
 	.id_table	= atyfb_pci_tbl,

From 89c69d2b8eb3ee2338fded9d70a0795b4712f112 Mon Sep 17 00:00:00 2001
From: Ville Syrjala <syrjala@sci.fi>
Date: Wed, 23 Jul 2008 21:31:32 -0700
Subject: [PATCH 0497/1171] atyfb: report probe errors

Properly propagate errors to the probe function.

Signed-off-by: Ville Syrjala <syrjala@sci.fi>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/aty/atyfb_base.c | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)

diff --git a/drivers/video/aty/atyfb_base.c b/drivers/video/aty/atyfb_base.c
index 872760accb93..e38398f491f0 100644
--- a/drivers/video/aty/atyfb_base.c
+++ b/drivers/video/aty/atyfb_base.c
@@ -2231,6 +2231,7 @@ static int __devinit aty_init(struct fb_info *info)
 	const char *ramname = NULL, *xtal;
 	int gtb_memsize, has_var = 0;
 	struct fb_var_screeninfo var;
+	int ret;
 
 	init_waitqueue_head(&par->vblank.wait);
 	spin_lock_init(&par->int_lock);
@@ -2612,7 +2613,8 @@ static int __devinit aty_init(struct fb_info *info)
 			var.yres_virtual = var.yres;
 	}
 
-	if (atyfb_check_var(&var, info)) {
+	ret = atyfb_check_var(&var, info);
+	if (ret) {
 		PRINTKE("can't set default video mode\n");
 		goto aty_init_exit;
 	}
@@ -2623,10 +2625,12 @@ static int __devinit aty_init(struct fb_info *info)
 #endif /* CONFIG_FB_ATY_CT */
 	info->var = var;
 
-	if (fb_alloc_cmap(&info->cmap, 256, 0) < 0)
+	ret = fb_alloc_cmap(&info->cmap, 256, 0);
+	if (ret < 0)
 		goto aty_init_exit;
 
-	if (register_framebuffer(info) < 0) {
+	ret = register_framebuffer(info);
+	if (ret < 0) {
 		fb_dealloc_cmap(&info->cmap);
 		goto aty_init_exit;
 	}
@@ -2652,7 +2656,7 @@ aty_init_exit:
 	    par->mtrr_aper = -1;
 	}
 #endif
-	return -1;
+	return ret;
 }
 
 static void aty_resume_chip(struct fb_info *info)
@@ -3467,7 +3471,8 @@ static int __devinit atyfb_pci_probe(struct pci_dev *pdev, const struct pci_devi
 	pci_set_drvdata(pdev, info);
 
 	/* Init chip & register framebuffer */
-	if (aty_init(info))
+	rc = aty_init(info);
+	if (rc)
 		goto err_release_io;
 
 #ifdef __sparc__

From 1c554ff9554d67b4db0fb5e2f78c7cb4b2e0d627 Mon Sep 17 00:00:00 2001
From: Ville Syrjala <syrjala@sci.fi>
Date: Wed, 23 Jul 2008 21:31:32 -0700
Subject: [PATCH 0498/1171] atyfb: fix a cast

The argument to iounmap() is void __iomem *. Fix the cast.

Signed-off-by: Ville Syrjala <syrjala@sci.fi>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/aty/atyfb_base.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/video/aty/atyfb_base.c b/drivers/video/aty/atyfb_base.c
index e38398f491f0..620ba8120368 100644
--- a/drivers/video/aty/atyfb_base.c
+++ b/drivers/video/aty/atyfb_base.c
@@ -3335,7 +3335,7 @@ static int __devinit init_from_bios(struct atyfb_par *par)
 		PRINTKE("no BIOS frequency table found, use parameters\n");
 		ret = -ENXIO;
 	}
-	iounmap((void* __iomem )bios_base);
+	iounmap((void __iomem *)bios_base);
 
 	return ret;
 }

From a882ef47c7156e8cc47e72f2aa396f2514569c48 Mon Sep 17 00:00:00 2001
From: Akinobu Mita <akinobu.mita@gmail.com>
Date: Wed, 23 Jul 2008 21:31:33 -0700
Subject: [PATCH 0499/1171] aty: use memory_read_from_buffer()

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/aty/radeon_base.c | 10 +---------
 1 file changed, 1 insertion(+), 9 deletions(-)

diff --git a/drivers/video/aty/radeon_base.c b/drivers/video/aty/radeon_base.c
index 3c0a03f69d83..652273e9f5f9 100644
--- a/drivers/video/aty/radeon_base.c
+++ b/drivers/video/aty/radeon_base.c
@@ -2098,15 +2098,7 @@ static void radeon_identify_vram(struct radeonfb_info *rinfo)
 
 static ssize_t radeon_show_one_edid(char *buf, loff_t off, size_t count, const u8 *edid)
 {
-	if (off > EDID_LENGTH)
-		return 0;
-
-	if (off + count > EDID_LENGTH)
-		count = EDID_LENGTH - off;
-
-	memcpy(buf, edid + off, count);
-
-	return count;
+	return memory_read_from_buffer(buf, count, &off, edid, EDID_LENGTH);
 }
 
 

From 84c41ce83e9b2987ccef352f28ba0055b26c8f8e Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:34 -0700
Subject: [PATCH 0500/1171] skeletonfb: update to correct platform driver usage

It updates skeletonfb to new platform driver API.  The skeletonfb is
templates for creating new drivers.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/skeletonfb.c | 37 ++++++++++++++++++++-----------------
 1 file changed, 20 insertions(+), 17 deletions(-)

diff --git a/drivers/video/skeletonfb.c b/drivers/video/skeletonfb.c
index 62321458f71a..df5336561d13 100644
--- a/drivers/video/skeletonfb.c
+++ b/drivers/video/skeletonfb.c
@@ -675,13 +675,13 @@ static struct fb_ops xxxfb_ops = {
      *  Initialization
      */
 
-/* static int __init xxfb_probe (struct device *device) -- for platform devs */
+/* static int __init xxfb_probe (struct platform_device *pdev) -- for platform devs */
 static int __devinit xxxfb_probe(struct pci_dev *dev,
 			      const struct pci_device_id *ent)
 {
     struct fb_info *info;
     struct xxx_par *par;
-    struct device* device = &dev->dev; /* for pci drivers */
+    struct device *device = &dev->dev; /* or &pdev->dev */
     int cmap_len, retval;	
    
     /*
@@ -824,18 +824,18 @@ static int __devinit xxxfb_probe(struct pci_dev *dev,
 	return -EINVAL;
     printk(KERN_INFO "fb%d: %s frame buffer device\n", info->node,
 	   info->fix.id);
-    pci_set_drvdata(dev, info); /* or dev_set_drvdata(device, info) */
+    pci_set_drvdata(dev, info); /* or platform_set_drvdata(pdev, info) */
     return 0;
 }
 
     /*
      *  Cleanup
      */
-/* static void __devexit xxxfb_remove(struct device *device) */
+/* static void __devexit xxxfb_remove(struct platform_device *pdev) */
 static void __devexit xxxfb_remove(struct pci_dev *dev)
 {
 	struct fb_info *info = pci_get_drvdata(dev);
-	/* or dev_get_drvdata(device); */
+	/* or platform_get_drvdata(pdev); */
 
 	if (info) {
 		unregister_framebuffer(info);
@@ -961,18 +961,17 @@ static int xxxfb_resume(struct platform_dev *dev)
 #define xxxfb_resume NULL
 #endif /* CONFIG_PM */
 
-static struct device_driver xxxfb_driver = {
-	.name = "xxxfb",
-	.bus  = &platform_bus_type,
+static struct platform_device_driver xxxfb_driver = {
 	.probe = xxxfb_probe,
 	.remove = xxxfb_remove,
 	.suspend = xxxfb_suspend, /* optional but recommended */
 	.resume = xxxfb_resume,   /* optional but recommended */
+	.driver = {
+		.name = "xxxfb",
+	},
 };
 
-static struct platform_device xxxfb_device = {
-	.name = "xxxfb",
-};
+static struct platform_device *xxxfb_device;
 
 #ifndef MODULE
     /*
@@ -1002,12 +1001,16 @@ static int __init xxxfb_init(void)
 		return -ENODEV;
 	xxxfb_setup(option);
 #endif
-	ret = driver_register(&xxxfb_driver);
+	ret = platform_driver_register(&xxxfb_driver);
 
 	if (!ret) {
-		ret = platform_device_register(&xxxfb_device);
-		if (ret)
-			driver_unregister(&xxxfb_driver);
+		xxxfb_device = platform_device_register_simple("xxxfb", 0,
+								NULL, 0);
+
+		if (IS_ERR(xxxfb_device)) {
+			platform_driver_unregister(&xxxfb_driver);
+			ret = PTR_ERR(xxxfb_device);
+		}
 	}
 
 	return ret;
@@ -1015,8 +1018,8 @@ static int __init xxxfb_init(void)
 
 static void __exit xxxfb_exit(void)
 {
-	platform_device_unregister(&xxxfb_device);
-	driver_unregister(&xxxfb_driver);
+	platform_device_unregister(xxxfb_device);
+	platform_driver_unregister(&xxxfb_driver);
 }
 #endif /* CONFIG_PCI */
 

From 968910bd03b226ed410d092c2da59dffe5bfe8de Mon Sep 17 00:00:00 2001
From: Nicolas Ferre <nicolas.ferre@atmel.com>
Date: Wed, 23 Jul 2008 21:31:34 -0700
Subject: [PATCH 0501/1171] atmel_lcdfb: avoid division by zero

Avoid division by zero in  atmel_lcdfb_check_var() function.

If pixclock is not specified while passing a var structure in
the check_var() funtion, a division by zero occurs (when
translating pixclock to KHz).

This patch adds a checking of this value and try to choose a
video mode in the modelist.

The mode found in the probe function in added to the modelist.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
Cc: Haavard Skinnemoen <hskinnemoen@atmel.com>
Cc: Andrew Victor <linux@maxim.org.za>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/atmel_lcdfb.c | 35 +++++++++++++++++++++++++++++++++++
 1 file changed, 35 insertions(+)

diff --git a/drivers/video/atmel_lcdfb.c b/drivers/video/atmel_lcdfb.c
index d335bb96b03b..5b3a15dffb5f 100644
--- a/drivers/video/atmel_lcdfb.c
+++ b/drivers/video/atmel_lcdfb.c
@@ -256,6 +256,20 @@ static int atmel_lcdfb_alloc_video_memory(struct atmel_lcdfb_info *sinfo)
 	return 0;
 }
 
+static const struct fb_videomode *atmel_lcdfb_choose_mode(struct fb_var_screeninfo *var,
+						     struct fb_info *info)
+{
+	struct fb_videomode varfbmode;
+	const struct fb_videomode *fbmode = NULL;
+
+	fb_var_to_videomode(&varfbmode, var);
+	fbmode = fb_find_nearest_mode(&varfbmode, &info->modelist);
+	if (fbmode)
+		fb_videomode_to_var(var, fbmode);
+	return fbmode;
+}
+
+
 /**
  *      atmel_lcdfb_check_var - Validates a var passed in.
  *      @var: frame buffer variable screen structure
@@ -289,6 +303,15 @@ static int atmel_lcdfb_check_var(struct fb_var_screeninfo *var,
 	clk_value_khz = clk_get_rate(sinfo->lcdc_clk) / 1000;
 
 	dev_dbg(dev, "%s:\n", __func__);
+
+	if (!(var->pixclock && var->bits_per_pixel)) {
+		/* choose a suitable mode if possible */
+		if (!atmel_lcdfb_choose_mode(var, info)) {
+			dev_err(dev, "needed value not specified\n");
+			return -EINVAL;
+		}
+	}
+
 	dev_dbg(dev, "  resolution: %ux%u\n", var->xres, var->yres);
 	dev_dbg(dev, "  pixclk:     %lu KHz\n", PICOS2KHZ(var->pixclock));
 	dev_dbg(dev, "  bpp:        %u\n", var->bits_per_pixel);
@@ -299,6 +322,13 @@ static int atmel_lcdfb_check_var(struct fb_var_screeninfo *var,
 		return -EINVAL;
 	}
 
+	/* Do not allow to have real resoulution larger than virtual */
+	if (var->xres > var->xres_virtual)
+		var->xres_virtual = var->xres;
+
+	if (var->yres > var->yres_virtual)
+		var->yres_virtual = var->yres;
+
 	/* Force same alignment for each line */
 	var->xres = (var->xres + 3) & ~3UL;
 	var->xres_virtual = (var->xres_virtual + 3) & ~3UL;
@@ -740,6 +770,7 @@ static int __init atmel_lcdfb_probe(struct platform_device *pdev)
 	struct fb_info *info;
 	struct atmel_lcdfb_info *sinfo;
 	struct atmel_lcdfb_info *pdata_sinfo;
+	struct fb_videomode fbmode;
 	struct resource *regs = NULL;
 	struct resource *map = NULL;
 	int ret;
@@ -906,6 +937,10 @@ static int __init atmel_lcdfb_probe(struct platform_device *pdev)
 		goto free_cmap;
 	}
 
+	/* add selected videomode to modelist */
+	fb_var_to_videomode(&fbmode, &info->var);
+	fb_add_videomode(&fbmode, &info->modelist);
+
 	/* Power up the LCDC screen */
 	if (sinfo->atmel_lcdfb_power_control)
 		sinfo->atmel_lcdfb_power_control(1);

From 206c5d69d0540024faffd423fc703f1e457332d7 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Wed, 23 Jul 2008 21:31:35 -0700
Subject: [PATCH 0502/1171] sm501: add inversion controls for VBIASEN and FPEN

Add flags to allow the driver to invert the sense of both VBIASEN and FPEN
signals comming from the SM501.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/sm501fb.c | 26 ++++++++++++++++++++++----
 include/linux/sm501.h   |  2 ++
 2 files changed, 24 insertions(+), 4 deletions(-)

diff --git a/drivers/video/sm501fb.c b/drivers/video/sm501fb.c
index 15d4a768b1f6..122a0f8495c8 100644
--- a/drivers/video/sm501fb.c
+++ b/drivers/video/sm501fb.c
@@ -663,15 +663,25 @@ static void sm501fb_panel_power(struct sm501fb_info *fbi, int to)
 		sm501fb_sync_regs(fbi);
 		mdelay(10);
 
+		/* VBIASEN */
+
 		if (!(pd->flags & SM501FB_FLAG_PANEL_NO_VBIASEN)) {
-			control |= SM501_DC_PANEL_CONTROL_BIAS;	/* VBIASEN */
+			if (pd->flags & SM501FB_FLAG_PANEL_INV_VBIASEN)
+				control &= ~SM501_DC_PANEL_CONTROL_BIAS;
+			else
+				control |= SM501_DC_PANEL_CONTROL_BIAS;
+
 			writel(control, ctrl_reg);
 			sm501fb_sync_regs(fbi);
 			mdelay(10);
 		}
 
 		if (!(pd->flags & SM501FB_FLAG_PANEL_NO_FPEN)) {
-			control |= SM501_DC_PANEL_CONTROL_FPEN;
+			if (pd->flags & SM501FB_FLAG_PANEL_INV_FPEN)
+				control &= ~SM501_DC_PANEL_CONTROL_FPEN;
+			else
+				control |= SM501_DC_PANEL_CONTROL_FPEN;
+
 			writel(control, ctrl_reg);
 			sm501fb_sync_regs(fbi);
 			mdelay(10);
@@ -679,14 +689,22 @@ static void sm501fb_panel_power(struct sm501fb_info *fbi, int to)
 	} else if (!to && (control & SM501_DC_PANEL_CONTROL_VDD) != 0) {
 		/* disable panel power */
 		if (!(pd->flags & SM501FB_FLAG_PANEL_NO_FPEN)) {
-			control &= ~SM501_DC_PANEL_CONTROL_FPEN;
+			if (pd->flags & SM501FB_FLAG_PANEL_INV_FPEN)
+				control |= SM501_DC_PANEL_CONTROL_FPEN;
+			else
+				control &= ~SM501_DC_PANEL_CONTROL_FPEN;
+
 			writel(control, ctrl_reg);
 			sm501fb_sync_regs(fbi);
 			mdelay(10);
 		}
 
 		if (!(pd->flags & SM501FB_FLAG_PANEL_NO_VBIASEN)) {
-			control &= ~SM501_DC_PANEL_CONTROL_BIAS;
+			if (pd->flags & SM501FB_FLAG_PANEL_INV_VBIASEN)
+				control |= SM501_DC_PANEL_CONTROL_BIAS;
+			else
+				control &= ~SM501_DC_PANEL_CONTROL_BIAS;
+
 			writel(control, ctrl_reg);
 			sm501fb_sync_regs(fbi);
 			mdelay(10);
diff --git a/include/linux/sm501.h b/include/linux/sm501.h
index 95c1c39ba445..b530fa6a1d34 100644
--- a/include/linux/sm501.h
+++ b/include/linux/sm501.h
@@ -73,6 +73,8 @@ extern unsigned long sm501_gpio_get(struct device *dev,
 #define SM501FB_FLAG_USE_HWACCEL	(1<<3)
 #define SM501FB_FLAG_PANEL_NO_FPEN	(1<<4)
 #define SM501FB_FLAG_PANEL_NO_VBIASEN	(1<<5)
+#define SM501FB_FLAG_PANEL_INV_FPEN	(1<<6)
+#define SM501FB_FLAG_PANEL_INV_VBIASEN	(1<<7)
 
 struct sm501_platdata_fbsub {
 	struct fb_videomode	*def_mode;

From 9b599fb2fc23386dfd2965bf7d10b2b0f628b208 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Wed, 23 Jul 2008 21:31:36 -0700
Subject: [PATCH 0503/1171] sm501: restructure init to allow only 1 fb on an
 SM501

Add the ability to register only one of the two possible main framebuffer
devices on the SM501 by passing platform data for only the framebuffer
that you are interested in having.

As a side note, we update the init sequence to commonise the code that is
executed twice, and fix a pair of missing frees that we didn't do on
framebuffer exit, such as freeing the fb's cmap.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/sm501fb.c | 268 ++++++++++++++++++++++------------------
 1 file changed, 151 insertions(+), 117 deletions(-)

diff --git a/drivers/video/sm501fb.c b/drivers/video/sm501fb.c
index 122a0f8495c8..b473cf665d83 100644
--- a/drivers/video/sm501fb.c
+++ b/drivers/video/sm501fb.c
@@ -143,6 +143,8 @@ static int sm501_alloc_mem(struct sm501fb_info *inf, struct sm501_mem *mem,
 			   unsigned int why, size_t size)
 {
 	unsigned int ptr = 0;
+	unsigned int end;
+	struct fb_info *fbi;
 
 	switch (why) {
 	case SM501_MEMF_CURSOR:
@@ -152,7 +154,9 @@ static int sm501_alloc_mem(struct sm501fb_info *inf, struct sm501_mem *mem,
 
 	case SM501_MEMF_PANEL:
 		ptr = inf->fbmem_len - size;
-		if (ptr < inf->fb[0]->fix.smem_len)
+		fbi = inf->fb[0];
+
+		if (fbi && ptr < fbi->fix.smem_len)
 			return -ENOMEM;
 
 		break;
@@ -162,11 +166,18 @@ static int sm501_alloc_mem(struct sm501fb_info *inf, struct sm501_mem *mem,
 		break;
 
 	case SM501_MEMF_ACCEL:
-		ptr = inf->fb[0]->fix.smem_len;
+		fbi = inf->fb[0];
+		ptr = fbi ? fbi->fix.smem_len : 0;
 
-		if ((ptr + size) >
-		    (inf->fb[1]->fix.smem_start - inf->fbmem_res->start))
+		fbi = inf->fb[1];
+		if (fbi)
+			end = (fbi->fix.smem_start - inf->fbmem_res->start);
+		else
+			end = inf->fbmem_len;
+
+		if ((ptr + size) > end)
 			return -ENOMEM;
+
 		break;
 
 	default:
@@ -1228,39 +1239,6 @@ static struct fb_ops sm501fb_ops_pnl = {
 	.fb_imageblit	= cfb_imageblit,
 };
 
-/* sm501fb_info_alloc
- *
- * creates and initialises an sm501fb_info structure
-*/
-
-static struct sm501fb_info *sm501fb_info_alloc(struct fb_info *fbinfo_crt,
-					       struct fb_info *fbinfo_pnl)
-{
-	struct sm501fb_info *info;
-	struct sm501fb_par  *par;
-
-	info = kzalloc(sizeof(struct sm501fb_info), GFP_KERNEL);
-	if (info) {
-		/* set the references back */
-
-		par = fbinfo_crt->par;
-		par->info = info;
-		par->head = HEAD_CRT;
-		fbinfo_crt->pseudo_palette = &par->pseudo_palette;
-
-		par = fbinfo_pnl->par;
-		par->info = info;
-		par->head = HEAD_PANEL;
-		fbinfo_pnl->pseudo_palette = &par->pseudo_palette;
-
-		/* store the two fbs into our info */
-		info->fb[HEAD_CRT] = fbinfo_crt;
-		info->fb[HEAD_PANEL] = fbinfo_pnl;
-	}
-
-	return info;
-}
-
 /* sm501_init_cursor
  *
  * initialise hw cursor parameters
@@ -1268,10 +1246,16 @@ static struct sm501fb_info *sm501fb_info_alloc(struct fb_info *fbinfo_crt,
 
 static int sm501_init_cursor(struct fb_info *fbi, unsigned int reg_base)
 {
-	struct sm501fb_par *par = fbi->par;
-	struct sm501fb_info *info = par->info;
+	struct sm501fb_par *par;
+	struct sm501fb_info *info;
 	int ret;
 
+	if (fbi == NULL)
+		return 0;
+
+	par = fbi->par;
+	info = par->info;
+
 	par->cursor_regs = info->regs + reg_base;
 
 	ret = sm501_alloc_mem(info, &par->cursor, SM501_MEMF_CURSOR, 1024);
@@ -1299,13 +1283,10 @@ static int sm501fb_start(struct sm501fb_info *info,
 			 struct platform_device *pdev)
 {
 	struct resource	*res;
-	struct device *dev;
+	struct device *dev = &pdev->dev;
 	int k;
 	int ret;
 
-	info->dev = dev = &pdev->dev;
-	platform_set_drvdata(pdev, info);
-
 	info->irq = ret = platform_get_irq(pdev, 0);
 	if (ret < 0) {
 		/* we currently do not use the IRQ */
@@ -1408,11 +1389,6 @@ static void sm501fb_stop(struct sm501fb_info *info)
 	kfree(info->regs_res);
 }
 
-static void sm501fb_info_release(struct sm501fb_info *info)
-{
-	kfree(info);
-}
-
 static int sm501fb_init_fb(struct fb_info *fb,
 			   enum sm501_controller head,
 			   const char *fbname)
@@ -1557,36 +1533,93 @@ static struct sm501_platdata_fb sm501fb_def_pdata = {
 static char driver_name_crt[] = "sm501fb-crt";
 static char driver_name_pnl[] = "sm501fb-panel";
 
-static int __init sm501fb_probe(struct platform_device *pdev)
+static int __devinit sm501fb_probe_one(struct sm501fb_info *info,
+				       enum sm501_controller head)
 {
-	struct sm501fb_info *info;
-	struct device	    *dev = &pdev->dev;
-	struct fb_info	    *fbinfo_crt;
-	struct fb_info	    *fbinfo_pnl;
-	int		     ret;
+	unsigned char *name = (head == HEAD_CRT) ? "crt" : "panel";
+	struct sm501_platdata_fbsub *pd;
+	struct sm501fb_par *par;
+	struct fb_info *fbi;
 
-	/* allocate our framebuffers */
+	pd = (head == HEAD_CRT) ? info->pdata->fb_crt : info->pdata->fb_pnl;
 
-	fbinfo_crt = framebuffer_alloc(sizeof(struct sm501fb_par), dev);
-	if (fbinfo_crt == NULL) {
-		dev_err(dev, "cannot allocate crt framebuffer\n");
+	/* Do not initialise if we've not been given any platform data */
+	if (pd == NULL) {
+		dev_info(info->dev, "no data for fb %s (disabled)\n", name);
+		return 0;
+	}
+
+	fbi = framebuffer_alloc(sizeof(struct sm501fb_par), info->dev);
+	if (fbi == NULL) {
+		dev_err(info->dev, "cannot allocate %s framebuffer\n", name);
 		return -ENOMEM;
 	}
 
-	fbinfo_pnl = framebuffer_alloc(sizeof(struct sm501fb_par), dev);
-	if (fbinfo_pnl == NULL) {
-		dev_err(dev, "cannot allocate panel framebuffer\n");
-		ret = -ENOMEM;
-		goto fbinfo_crt_alloc_fail;
+	par = fbi->par;
+	par->info = info;
+	par->head = head;
+	fbi->pseudo_palette = &par->pseudo_palette;
+
+	info->fb[head] = fbi;
+
+	return 0;
+}
+
+/* Free up anything allocated by sm501fb_init_fb */
+
+static void sm501_free_init_fb(struct sm501fb_info *info,
+				enum sm501_controller head)
+{
+	struct fb_info *fbi = info->fb[head];
+
+	fb_dealloc_cmap(&fbi->cmap);
+}
+
+static int __devinit sm501fb_start_one(struct sm501fb_info *info,
+				       enum sm501_controller head,
+				       const char *drvname)
+{
+	struct fb_info *fbi = info->fb[head];
+	int ret;
+
+	if (!fbi)
+		return 0;
+
+	ret = sm501fb_init_fb(info->fb[head], head, drvname);
+	if (ret) {
+		dev_err(info->dev, "cannot initialise fb %s\n", drvname);
+		return ret;
 	}
 
-	info = sm501fb_info_alloc(fbinfo_crt, fbinfo_pnl);
-	if (info == NULL) {
-		dev_err(dev, "cannot allocate par\n");
-		ret = -ENOMEM;
-		goto sm501fb_alloc_fail;
+	ret = register_framebuffer(info->fb[head]);
+	if (ret) {
+		dev_err(info->dev, "failed to register fb %s\n", drvname);
+		sm501_free_init_fb(info, head);
+		return ret;
 	}
 
+	dev_info(info->dev, "fb%d: %s frame buffer\n", fbi->node, fbi->fix.id);
+
+	return 0;
+}
+
+static int __devinit sm501fb_probe(struct platform_device *pdev)
+{
+	struct sm501fb_info *info;
+	struct device *dev = &pdev->dev;
+	int ret;
+
+	/* allocate our framebuffers */
+
+	info = kzalloc(sizeof(struct sm501fb_info), GFP_KERNEL);
+	if (!info) {
+		dev_err(dev, "failed to allocate state\n");
+		return -ENOMEM;
+	}
+
+	info->dev = dev = &pdev->dev;
+	platform_set_drvdata(pdev, info);
+
 	if (dev->parent->platform_data) {
 		struct sm501_platdata *pd = dev->parent->platform_data;
 		info->pdata = pd->fb;
@@ -1597,90 +1630,88 @@ static int __init sm501fb_probe(struct platform_device *pdev)
 		info->pdata = &sm501fb_def_pdata;
 	}
 
-	/* start the framebuffers */
+	/* probe for the presence of each panel */
+
+	ret = sm501fb_probe_one(info, HEAD_CRT);
+	if (ret < 0) {
+		dev_err(dev, "failed to probe CRT\n");
+		goto err_alloc;
+	}
+
+	ret = sm501fb_probe_one(info, HEAD_PANEL);
+	if (ret < 0) {
+		dev_err(dev, "failed to probe PANEL\n");
+		goto err_probed_crt;
+	}
+
+	if (info->fb[HEAD_PANEL] == NULL &&
+	    info->fb[HEAD_CRT] == NULL) {
+		dev_err(dev, "no framebuffers found\n");
+		goto err_alloc;
+	}
+
+	/* get the resources for both of the framebuffers */
 
 	ret = sm501fb_start(info, pdev);
 	if (ret) {
 		dev_err(dev, "cannot initialise SM501\n");
-		goto sm501fb_start_fail;
+		goto err_probed_panel;
 	}
 
-	/* CRT framebuffer setup */
-
-	ret = sm501fb_init_fb(fbinfo_crt, HEAD_CRT, driver_name_crt);
+	ret = sm501fb_start_one(info, HEAD_CRT, driver_name_crt);
 	if (ret) {
-		dev_err(dev, "cannot initialise CRT fb\n");
-		goto sm501fb_start_fail;
+		dev_err(dev, "failed to start CRT\n");
+		goto err_started;
 	}
 
-	/* Panel framebuffer setup */
-
-	ret = sm501fb_init_fb(fbinfo_pnl, HEAD_PANEL, driver_name_pnl);
+	ret = sm501fb_start_one(info, HEAD_PANEL, driver_name_pnl);
 	if (ret) {
-		dev_err(dev, "cannot initialise Panel fb\n");
-		goto sm501fb_start_fail;
+		dev_err(dev, "failed to start Panel\n");
+		goto err_started_crt;
 	}
 
-	/* register framebuffers */
-
-	ret = register_framebuffer(fbinfo_crt);
-	if (ret < 0) {
-		dev_err(dev, "failed to register CRT fb (%d)\n", ret);
-		goto register_crt_fail;
-	}
-
-	ret = register_framebuffer(fbinfo_pnl);
-	if (ret < 0) {
-		dev_err(dev, "failed to register panel fb (%d)\n", ret);
-		goto register_pnl_fail;
-	}
-
-	dev_info(dev, "fb%d: %s frame buffer device\n",
-		 fbinfo_crt->node, fbinfo_crt->fix.id);
-
-	dev_info(dev, "fb%d: %s frame buffer device\n",
-	       fbinfo_pnl->node, fbinfo_pnl->fix.id);
-
 	/* create device files */
 
 	ret = device_create_file(dev, &dev_attr_crt_src);
 	if (ret)
-		goto crtsrc_fail;
+		goto err_started_panel;
 
 	ret = device_create_file(dev, &dev_attr_fbregs_pnl);
 	if (ret)
-		goto fbregs_pnl_fail;
+		goto err_attached_crtsrc_file;
 
 	ret = device_create_file(dev, &dev_attr_fbregs_crt);
 	if (ret)
-		goto fbregs_crt_fail;
+		goto err_attached_pnlregs_file;
 
 	/* we registered, return ok */
 	return 0;
 
- fbregs_crt_fail:
+err_attached_pnlregs_file:
 	device_remove_file(dev, &dev_attr_fbregs_pnl);
 
- fbregs_pnl_fail:
+err_attached_crtsrc_file:
 	device_remove_file(dev, &dev_attr_crt_src);
 
- crtsrc_fail:
-	unregister_framebuffer(fbinfo_pnl);
+err_started_panel:
+	unregister_framebuffer(info->fb[HEAD_PANEL]);
+	sm501_free_init_fb(info, HEAD_PANEL);
 
- register_pnl_fail:
-	unregister_framebuffer(fbinfo_crt);
+err_started_crt:
+	unregister_framebuffer(info->fb[HEAD_CRT]);
+	sm501_free_init_fb(info, HEAD_CRT);
 
- register_crt_fail:
+err_started:
 	sm501fb_stop(info);
 
- sm501fb_start_fail:
-	sm501fb_info_release(info);
+err_probed_panel:
+	framebuffer_release(info->fb[HEAD_PANEL]);
 
- sm501fb_alloc_fail:
-	framebuffer_release(fbinfo_pnl);
+err_probed_crt:
+	framebuffer_release(info->fb[HEAD_CRT]);
 
- fbinfo_crt_alloc_fail:
-	framebuffer_release(fbinfo_crt);
+err_alloc:
+	kfree(info);
 
 	return ret;
 }
@@ -1699,11 +1730,14 @@ static int sm501fb_remove(struct platform_device *pdev)
 	device_remove_file(&pdev->dev, &dev_attr_fbregs_pnl);
 	device_remove_file(&pdev->dev, &dev_attr_crt_src);
 
+	sm501_free_init_fb(info, HEAD_CRT);
+	sm501_free_init_fb(info, HEAD_PANEL);
+
 	unregister_framebuffer(fbinfo_crt);
 	unregister_framebuffer(fbinfo_pnl);
 
 	sm501fb_stop(info);
-	sm501fb_info_release(info);
+	kfree(info);
 
 	framebuffer_release(fbinfo_pnl);
 	framebuffer_release(fbinfo_crt);

From d05254190dd1a4751284f4a51efb70fcc16c45a4 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Wed, 23 Jul 2008 21:31:37 -0700
Subject: [PATCH 0504/1171] sm501: fixup allocation code to be 64bit resource
 compliant

As pointed out by Andrew Morton, we have a problem when setting the 64bit
resources option.  Alter the allocation routines to remove the need to use
the start and end fields, use the proper HEAD_PANEL/HEAD_CRT and update
the comments.

Note, we also fix the bug where we failed to check the size of the
CRT memory allocation.

[akpm@linux-foundation.org: cleanup]
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/sm501fb.c | 53 ++++++++++++++++++++++++++++++++---------
 1 file changed, 42 insertions(+), 11 deletions(-)

diff --git a/drivers/video/sm501fb.c b/drivers/video/sm501fb.c
index b473cf665d83..f94ae84a58cd 100644
--- a/drivers/video/sm501fb.c
+++ b/drivers/video/sm501fb.c
@@ -48,10 +48,15 @@ enum sm501_controller {
 	HEAD_PANEL	= 1,
 };
 
-/* SM501 memory address */
+/* SM501 memory address.
+ *
+ * This structure is used to track memory usage within the SM501 framebuffer
+ * allocation. The sm_addr field is stored as an offset as it is often used
+ * against both the physical and mapped addresses.
+ */
 struct sm501_mem {
 	unsigned long	 size;
-	unsigned long	 sm_addr;
+	unsigned long	 sm_addr;	/* offset from base of sm501 fb. */
 	void __iomem	*k_addr;
 };
 
@@ -142,37 +147,63 @@ static inline void sm501fb_sync_regs(struct sm501fb_info *info)
 static int sm501_alloc_mem(struct sm501fb_info *inf, struct sm501_mem *mem,
 			   unsigned int why, size_t size)
 {
-	unsigned int ptr = 0;
-	unsigned int end;
+	struct sm501fb_par *par;
 	struct fb_info *fbi;
+	unsigned int ptr;
+	unsigned int end;
 
 	switch (why) {
 	case SM501_MEMF_CURSOR:
 		ptr = inf->fbmem_len - size;
-		inf->fbmem_len = ptr;
+		inf->fbmem_len = ptr;	/* adjust available memory. */
 		break;
 
 	case SM501_MEMF_PANEL:
 		ptr = inf->fbmem_len - size;
-		fbi = inf->fb[0];
+		fbi = inf->fb[HEAD_CRT];
+
+		/* round down, some programs such as directfb do not draw
+		 * 0,0 correctly unless the start is aligned to a page start.
+		 */
+
+		if (ptr > 0)
+			ptr &= ~(PAGE_SIZE - 1);
 
 		if (fbi && ptr < fbi->fix.smem_len)
 			return -ENOMEM;
 
+		if (ptr < 0)
+			return -ENOMEM;
+
 		break;
 
 	case SM501_MEMF_CRT:
 		ptr = 0;
+
+		/* check to see if we have panel memory allocated
+		 * which would put an limit on available memory. */
+
+		fbi = inf->fb[HEAD_PANEL];
+		if (fbi) {
+			par = fbi->par;
+			end = par->screen.k_addr ? par->screen.sm_addr : inf->fbmem_len;
+		} else
+			end = inf->fbmem_len;
+
+		if ((ptr + size) > end)
+			return -ENOMEM;
+
 		break;
 
 	case SM501_MEMF_ACCEL:
-		fbi = inf->fb[0];
+		fbi = inf->fb[HEAD_CRT];
 		ptr = fbi ? fbi->fix.smem_len : 0;
 
-		fbi = inf->fb[1];
-		if (fbi)
-			end = (fbi->fix.smem_start - inf->fbmem_res->start);
-		else
+		fbi = inf->fb[HEAD_PANEL];
+		if (fbi) {
+			par = fbi->par;
+			end = par->screen.sm_addr;
+		} else
 			end = inf->fbmem_len;
 
 		if ((ptr + size) > end)

From cccb6d3c149603b9c15d3c460dff317455df1766 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Wed, 23 Jul 2008 21:31:37 -0700
Subject: [PATCH 0505/1171] fb: add support for the ILI9320 video display
 controller

Provide support for the ILI9320 display controller chip which is found in
many LCD displays.  Included with this is support for an example LCD using
this chip, the VGG2432A4.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/backlight/Kconfig     |  17 ++
 drivers/video/backlight/Makefile    |   4 +-
 drivers/video/backlight/ili9320.c   | 330 ++++++++++++++++++++++++++++
 drivers/video/backlight/ili9320.h   |  80 +++++++
 drivers/video/backlight/vgg2432a4.c | 284 ++++++++++++++++++++++++
 include/video/ili9320.h             | 201 +++++++++++++++++
 6 files changed, 915 insertions(+), 1 deletion(-)
 create mode 100644 drivers/video/backlight/ili9320.c
 create mode 100644 drivers/video/backlight/ili9320.h
 create mode 100644 drivers/video/backlight/vgg2432a4.c
 create mode 100644 include/video/ili9320.h

diff --git a/drivers/video/backlight/Kconfig b/drivers/video/backlight/Kconfig
index 30bf7f2f1635..a5b3a92ffdc1 100644
--- a/drivers/video/backlight/Kconfig
+++ b/drivers/video/backlight/Kconfig
@@ -36,6 +36,23 @@ config LCD_LTV350QV
 
 	  The LTV350QV panel is present on all ATSTK1000 boards.
 
+config LCD_ILI9320
+	tristate
+	depends on LCD_CLASS_DEVICE && BACKLIGHT_LCD_SUPPORT
+	default n
+	help
+	  If you have a panel based on the ILI9320 controller chip
+	  then say y to include a power driver for it.
+
+config LCD_VGG2432A4
+	tristate "VGG2432A4 LCM device support"
+	depends on BACKLIGHT_LCD_SUPPORT && LCD_CLASS_DEVICE && SPI_MASTER
+	select LCD_ILI9320
+	default n
+	help
+	  If you have a VGG2432A4 panel based on the ILI9320 controller chip
+	  then say y to include a power driver for it.
+
 #
 # Backlight
 #
diff --git a/drivers/video/backlight/Makefile b/drivers/video/backlight/Makefile
index b51a7cd12500..366d84e380cf 100644
--- a/drivers/video/backlight/Makefile
+++ b/drivers/video/backlight/Makefile
@@ -1,7 +1,9 @@
 # Backlight & LCD drivers
 
 obj-$(CONFIG_LCD_CLASS_DEVICE)     += lcd.o
-obj-$(CONFIG_LCD_LTV350QV)	+= ltv350qv.o
+obj-$(CONFIG_LCD_LTV350QV)	   += ltv350qv.o
+obj-$(CONFIG_LCD_ILI9320)	   += ili9320.o
+obj-$(CONFIG_LCD_VGG2432A4)	   += vgg2432a4.o
 
 obj-$(CONFIG_BACKLIGHT_CLASS_DEVICE) += backlight.o
 obj-$(CONFIG_BACKLIGHT_CORGI)	+= corgi_bl.o
diff --git a/drivers/video/backlight/ili9320.c b/drivers/video/backlight/ili9320.c
new file mode 100644
index 000000000000..ba89b41b639c
--- /dev/null
+++ b/drivers/video/backlight/ili9320.c
@@ -0,0 +1,330 @@
+/* drivers/video/backlight/ili9320.c
+ *
+ * ILI9320 LCD controller driver core.
+ *
+ * Copyright 2007 Simtec Electronics
+ *	http://armlinux.simtec.co.uk/
+ *	Ben Dooks <ben@simtec.co.uk>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+*/
+
+#include <linux/delay.h>
+#include <linux/err.h>
+#include <linux/fb.h>
+#include <linux/init.h>
+#include <linux/lcd.h>
+#include <linux/module.h>
+
+#include <linux/spi/spi.h>
+
+#include <video/ili9320.h>
+
+#include "ili9320.h"
+
+
+static inline int ili9320_write_spi(struct ili9320 *ili,
+				    unsigned int reg,
+				    unsigned int value)
+{
+	struct ili9320_spi *spi = &ili->access.spi;
+	unsigned char *addr = spi->buffer_addr;
+	unsigned char *data = spi->buffer_data;
+
+	/* spi message consits of:
+	 * first byte: ID and operation
+	 */
+
+	addr[0] = spi->id | ILI9320_SPI_INDEX | ILI9320_SPI_WRITE;
+	addr[1] = reg >> 8;
+	addr[2] = reg;
+
+	/* second message is the data to transfer */
+
+	data[0] = spi->id | ILI9320_SPI_DATA  | ILI9320_SPI_WRITE;
+ 	data[1] = value >> 8;
+	data[2] = value;
+
+	return spi_sync(spi->dev, &spi->message);
+}
+
+int ili9320_write(struct ili9320 *ili, unsigned int reg, unsigned int value)
+{
+	dev_dbg(ili->dev, "write: reg=%02x, val=%04x\n", reg, value);
+	return ili->write(ili, reg, value);
+}
+
+EXPORT_SYMBOL_GPL(ili9320_write);
+
+int ili9320_write_regs(struct ili9320 *ili,
+		       struct ili9320_reg *values,
+		       int nr_values)
+{
+	int index;
+	int ret;
+
+	for (index = 0; index < nr_values; index++, values++) {
+		ret = ili9320_write(ili, values->address, values->value);
+		if (ret != 0)
+			return ret;
+	}
+
+	return 0;
+}
+
+EXPORT_SYMBOL_GPL(ili9320_write_regs);
+
+static void ili9320_reset(struct ili9320 *lcd)
+{
+	struct ili9320_platdata *cfg = lcd->platdata;
+
+	cfg->reset(1);
+	mdelay(50);
+
+	cfg->reset(0);
+	mdelay(50);
+
+	cfg->reset(1);
+	mdelay(100);
+}
+
+static inline int ili9320_init_chip(struct ili9320 *lcd)
+{
+	int ret;
+
+	ili9320_reset(lcd);
+
+	ret = lcd->client->init(lcd, lcd->platdata);
+	if (ret != 0) {
+		dev_err(lcd->dev, "failed to initialise display\n");
+		return ret;
+	}
+
+	lcd->initialised = 1;
+	return 0;
+}
+
+static inline int ili9320_power_on(struct ili9320 *lcd)
+{
+	if (!lcd->initialised)
+		ili9320_init_chip(lcd);
+
+	lcd->display1 |= (ILI9320_DISPLAY1_D(3) | ILI9320_DISPLAY1_BASEE);
+	ili9320_write(lcd, ILI9320_DISPLAY1, lcd->display1);
+
+	return 0;
+}
+
+static inline int ili9320_power_off(struct ili9320 *lcd)
+{
+	lcd->display1 &= ~(ILI9320_DISPLAY1_D(3) | ILI9320_DISPLAY1_BASEE);
+	ili9320_write(lcd, ILI9320_DISPLAY1, lcd->display1);
+
+	return 0;
+}
+
+#define POWER_IS_ON(pwr)	((pwr) <= FB_BLANK_NORMAL)
+
+static int ili9320_power(struct ili9320 *lcd, int power)
+{
+	int ret = 0;
+
+	dev_dbg(lcd->dev, "power %d => %d\n", lcd->power, power);
+
+	if (POWER_IS_ON(power) && !POWER_IS_ON(lcd->power))
+		ret = ili9320_power_on(lcd);
+	else if (!POWER_IS_ON(power) && POWER_IS_ON(lcd->power))
+		ret = ili9320_power_off(lcd);
+
+	if (ret == 0)
+		lcd->power = power;
+	else
+		dev_warn(lcd->dev, "failed to set power mode %d\n", power);
+
+	return ret;
+}
+
+static inline struct ili9320 *to_our_lcd(struct lcd_device *lcd)
+{
+	return lcd_get_data(lcd);
+}
+
+static int ili9320_set_power(struct lcd_device *ld, int power)
+{
+	struct ili9320 *lcd = to_our_lcd(ld);
+
+	return ili9320_power(lcd, power);
+}
+
+static int ili9320_get_power(struct lcd_device *ld)
+{
+	struct ili9320 *lcd = to_our_lcd(ld);
+
+	return lcd->power;
+}
+
+static struct lcd_ops ili9320_ops = {
+	.get_power	= ili9320_get_power,
+	.set_power	= ili9320_set_power,
+};
+
+static void __devinit ili9320_setup_spi(struct ili9320 *ili,
+					struct spi_device *dev)
+{
+	struct ili9320_spi *spi = &ili->access.spi;
+
+	ili->write = ili9320_write_spi;
+	spi->dev = dev;
+
+	/* fill the two messages we are going to use to send the data
+	 * with, the first the address followed by the data. The datasheet
+	 * says they should be done as two distinct cycles of the SPI CS line.
+	 */
+
+	spi->xfer[0].tx_buf = spi->buffer_addr;
+	spi->xfer[1].tx_buf = spi->buffer_data;
+	spi->xfer[0].len = 3;
+	spi->xfer[1].len = 3;
+	spi->xfer[0].bits_per_word = 8;
+	spi->xfer[1].bits_per_word = 8;
+	spi->xfer[0].cs_change = 1;
+
+	spi_message_init(&spi->message);
+	spi_message_add_tail(&spi->xfer[0], &spi->message);
+	spi_message_add_tail(&spi->xfer[1], &spi->message);
+}
+
+int __devinit ili9320_probe_spi(struct spi_device *spi,
+				struct ili9320_client *client)
+{
+	struct ili9320_platdata *cfg = spi->dev.platform_data;
+	struct device *dev = &spi->dev;
+	struct ili9320 *ili;
+	struct lcd_device *lcd;
+	int ret = 0;
+
+	/* verify we where given some information */
+
+	if (cfg == NULL) {
+		dev_err(dev, "no platform data supplied\n");
+		return -EINVAL;
+	}
+
+	if (cfg->hsize <= 0 || cfg->vsize <= 0 || cfg->reset == NULL) {
+		dev_err(dev, "invalid platform data supplied\n");
+		return -EINVAL;
+	}
+
+	/* allocate and initialse our state */
+
+	ili = kzalloc(sizeof(struct ili9320), GFP_KERNEL);
+	if (ili == NULL) {
+		dev_err(dev, "no memory for device\n");
+		return -ENOMEM;
+	}
+
+	ili->access.spi.id = ILI9320_SPI_IDCODE | ILI9320_SPI_ID(1);
+
+	ili->dev = dev;
+	ili->client = client;
+	ili->power = FB_BLANK_POWERDOWN;
+	ili->platdata = cfg;
+
+	dev_set_drvdata(&spi->dev, ili);
+
+	ili9320_setup_spi(ili, spi);
+
+	lcd = lcd_device_register("ili9320", dev, ili, &ili9320_ops);
+	if (IS_ERR(lcd)) {
+		dev_err(dev, "failed to register lcd device\n");
+		ret = PTR_ERR(lcd);
+		goto err_free;
+	}
+
+	ili->lcd = lcd;
+
+	dev_info(dev, "initialising %s\n", client->name);
+
+	ret = ili9320_power(ili, FB_BLANK_UNBLANK);
+	if (ret != 0) {
+		dev_err(dev, "failed to set lcd power state\n");
+		goto err_unregister;
+	}
+
+	return 0;
+
+ err_unregister:
+	lcd_device_unregister(lcd);
+
+ err_free:
+	kfree(ili);
+
+	return ret;
+}
+
+EXPORT_SYMBOL_GPL(ili9320_probe_spi);
+
+int __devexit ili9320_remove(struct ili9320 *ili)
+{
+	ili9320_power(ili, FB_BLANK_POWERDOWN);
+
+	lcd_device_unregister(ili->lcd);
+	kfree(ili);
+
+	return 0;
+}
+
+EXPORT_SYMBOL_GPL(ili9320_remove);
+
+#ifdef CONFIG_PM
+int ili9320_suspend(struct ili9320 *lcd, pm_message_t state)
+{
+	int ret;
+
+	dev_dbg(lcd->dev, "%s: event %d\n", __func__, state.event);
+
+	if (state.event == PM_EVENT_SUSPEND) {
+		ret = ili9320_power(lcd, FB_BLANK_POWERDOWN);
+
+		if (lcd->platdata->suspend == ILI9320_SUSPEND_DEEP) {
+			ili9320_write(lcd, ILI9320_POWER1, lcd->power1 |
+				      ILI9320_POWER1_SLP |
+				      ILI9320_POWER1_DSTB);
+			lcd->initialised = 0;
+		}
+
+		return ret;
+	}
+
+	return 0;
+}
+
+EXPORT_SYMBOL_GPL(ili9320_suspend);
+
+int ili9320_resume(struct ili9320 *lcd)
+{
+	dev_info(lcd->dev, "resuming from power state %d\n", lcd->power);
+
+	if (lcd->platdata->suspend == ILI9320_SUSPEND_DEEP) {
+		ili9320_write(lcd, ILI9320_POWER1, 0x00);
+	}
+
+	return ili9320_power(lcd, FB_BLANK_UNBLANK);
+}
+
+EXPORT_SYMBOL_GPL(ili9320_resume);
+#endif
+
+/* Power down all displays on reboot, poweroff or halt */
+void ili9320_shutdown(struct ili9320 *lcd)
+{
+	ili9320_power(lcd, FB_BLANK_POWERDOWN);
+}
+
+EXPORT_SYMBOL_GPL(ili9320_shutdown);
+
+MODULE_AUTHOR("Ben Dooks <ben-linux@fluff.org>");
+MODULE_DESCRIPTION("ILI9320 LCD Driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/video/backlight/ili9320.h b/drivers/video/backlight/ili9320.h
new file mode 100644
index 000000000000..e388eca7cac5
--- /dev/null
+++ b/drivers/video/backlight/ili9320.h
@@ -0,0 +1,80 @@
+/* drivers/video/backlight/ili9320.h
+ *
+ * ILI9320 LCD controller driver core.
+ *
+ * Copyright 2007 Simtec Electronics
+ *	Ben Dooks <ben@simtec.co.uk>
+ *
+ * http://armlinux.simtec.co.uk/
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+*/
+
+/* Holder for register and value pairs. */
+struct ili9320_reg {
+	unsigned short		address;
+	unsigned short		value;
+};
+
+struct ili9320;
+
+struct ili9320_client {
+	const char	*name;
+	int	(*init)(struct ili9320 *ili, struct ili9320_platdata *cfg);
+
+};
+/* Device attached via an SPI bus. */
+struct  ili9320_spi {
+	struct spi_device	*dev;
+	struct spi_message	message;
+	struct spi_transfer	xfer[2];
+
+	unsigned char		id;
+	unsigned char		buffer_addr[4];
+	unsigned char		buffer_data[4];
+};
+
+/* ILI9320 device state. */
+struct ili9320 {
+	union {
+		struct ili9320_spi	spi;	/* SPI attachged device. */
+	} access;				/* Register access method. */
+
+	struct device			*dev;
+	struct lcd_device		*lcd;	/* LCD device we created. */
+	struct ili9320_client		*client;
+	struct ili9320_platdata		*platdata;
+
+	int				 power; /* current power state. */
+	int				 initialised;
+
+	unsigned short			 display1;
+	unsigned short			 power1;
+
+	int (*write)(struct ili9320 *ili, unsigned int reg, unsigned int val);
+};
+
+
+/* ILI9320 register access routines */
+
+extern int ili9320_write(struct ili9320 *ili,
+			 unsigned int reg, unsigned int value);
+
+extern int ili9320_write_regs(struct ili9320 *ili,
+			      struct ili9320_reg *values,
+			      int nr_values);
+
+/* Device probe */
+
+extern int ili9320_probe_spi(struct spi_device *spi,
+			     struct ili9320_client *cli);
+
+extern int ili9320_remove(struct ili9320 *lcd);
+extern void ili9320_shutdown(struct ili9320 *lcd);
+
+/* PM */
+
+extern int ili9320_suspend(struct ili9320 *lcd, pm_message_t state);
+extern int ili9320_resume(struct ili9320 *lcd);
diff --git a/drivers/video/backlight/vgg2432a4.c b/drivers/video/backlight/vgg2432a4.c
new file mode 100644
index 000000000000..593c7687d54a
--- /dev/null
+++ b/drivers/video/backlight/vgg2432a4.c
@@ -0,0 +1,284 @@
+/* drivers/video/backlight/vgg2432a4.c
+ *
+ * VGG2432A4 (ILI9320) LCD controller driver.
+ *
+ * Copyright 2007 Simtec Electronics
+ *	http://armlinux.simtec.co.uk/
+ *	Ben Dooks <ben@simtec.co.uk>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+*/
+
+#include <linux/delay.h>
+#include <linux/err.h>
+#include <linux/fb.h>
+#include <linux/init.h>
+#include <linux/lcd.h>
+#include <linux/module.h>
+
+#include <linux/spi/spi.h>
+
+#include <video/ili9320.h>
+
+#include "ili9320.h"
+
+/* Device initialisation sequences */
+
+static struct ili9320_reg vgg_init1[] = {
+	{
+		.address = ILI9320_POWER1,
+		.value	 = ILI9320_POWER1_AP(0) | ILI9320_POWER1_BT(0),
+	}, {
+		.address = ILI9320_POWER2,
+		.value	 = (ILI9320_POWER2_VC(7) |
+			    ILI9320_POWER2_DC0(0) | ILI9320_POWER2_DC1(0)),
+	}, {
+		.address = ILI9320_POWER3,
+		.value	 = ILI9320_POWER3_VRH(0),
+	}, {
+		.address = ILI9320_POWER4,
+		.value	 = ILI9320_POWER4_VREOUT(0),
+	},
+};
+
+static struct ili9320_reg vgg_init2[] = {
+	{
+		.address = ILI9320_POWER1,
+		.value   = (ILI9320_POWER1_AP(3) | ILI9320_POWER1_APE |
+			    ILI9320_POWER1_BT(7) | ILI9320_POWER1_SAP),
+	}, {
+		.address = ILI9320_POWER2,
+		.value   = ILI9320_POWER2_VC(7) | ILI9320_POWER2_DC0(3),
+	}
+};
+
+static struct ili9320_reg vgg_gamma[] = {
+	{
+		.address = ILI9320_GAMMA1,
+		.value	 = 0x0000,
+	}, {
+		.address = ILI9320_GAMMA2,
+		.value   = 0x0505,
+	}, {
+		.address = ILI9320_GAMMA3,
+		.value	 = 0x0004,
+	}, {
+		.address = ILI9320_GAMMA4,
+		.value	 = 0x0006,
+	}, {
+		.address = ILI9320_GAMMA5,
+		.value	 = 0x0707,
+	}, {
+		.address = ILI9320_GAMMA6,
+		.value	 = 0x0105,
+	}, {
+		.address = ILI9320_GAMMA7,
+		.value	 = 0x0002,
+	}, {
+		.address = ILI9320_GAMMA8,
+		.value	 = 0x0707,
+	}, {
+		.address = ILI9320_GAMMA9,
+		.value	 = 0x0704,
+	}, {
+		.address = ILI9320_GAMMA10,
+		.value	 = 0x807,
+	}
+
+};
+
+static struct ili9320_reg vgg_init0[] = {
+	[0]	= {
+		/* set direction and scan mode gate */
+		.address = ILI9320_DRIVER,
+		.value	 = ILI9320_DRIVER_SS,
+	}, {
+		.address = ILI9320_DRIVEWAVE,
+		.value	 = (ILI9320_DRIVEWAVE_MUSTSET |
+			    ILI9320_DRIVEWAVE_EOR | ILI9320_DRIVEWAVE_BC),
+	}, {
+		.address = ILI9320_ENTRYMODE,
+		.value	 = ILI9320_ENTRYMODE_ID(3) | ILI9320_ENTRYMODE_BGR,
+	}, {
+		.address = ILI9320_RESIZING,
+		.value	 = 0x0,
+	},
+};
+
+
+static int vgg2432a4_lcd_init(struct ili9320 *lcd,
+			      struct ili9320_platdata *cfg)
+{
+	unsigned int addr;
+	int ret;
+
+	/* Set VCore before anything else (VGG243237-6UFLWA) */
+	ret = ili9320_write(lcd, 0x00e5, 0x8000);
+	if (ret)
+		goto err_initial;
+
+	/* Start the oscillator up before we can do anything else. */
+	ret = ili9320_write(lcd, ILI9320_OSCILATION, ILI9320_OSCILATION_OSC);
+	if (ret)
+		goto err_initial;
+
+	/* must wait at-lesat 10ms after starting */
+	mdelay(15);
+
+	ret = ili9320_write_regs(lcd, vgg_init0, ARRAY_SIZE(vgg_init0));
+	if (ret != 0)
+		goto err_initial;
+
+	ili9320_write(lcd, ILI9320_DISPLAY2, cfg->display2);
+	ili9320_write(lcd, ILI9320_DISPLAY3, cfg->display3);
+	ili9320_write(lcd, ILI9320_DISPLAY4, cfg->display4);
+
+	ili9320_write(lcd, ILI9320_RGB_IF1, cfg->rgb_if1);
+	ili9320_write(lcd, ILI9320_FRAMEMAKER, 0x0);
+	ili9320_write(lcd, ILI9320_RGB_IF2, ILI9320_RGBIF2_DPL);
+
+	ret = ili9320_write_regs(lcd, vgg_init1, ARRAY_SIZE(vgg_init1));
+	if (ret != 0)
+		goto err_vgg;
+
+	mdelay(300);
+
+	ret = ili9320_write_regs(lcd, vgg_init2, ARRAY_SIZE(vgg_init2));
+	if (ret != 0)
+		goto err_vgg2;
+
+	mdelay(100);
+
+	ili9320_write(lcd, ILI9320_POWER3, 0x13c);
+
+	mdelay(100);
+
+	ili9320_write(lcd, ILI9320_POWER4, 0x1c00);
+	ili9320_write(lcd, ILI9320_POWER7, 0x000e);
+
+	mdelay(100);
+
+	ili9320_write(lcd, ILI9320_GRAM_HORIZ_ADDR, 0x00);
+	ili9320_write(lcd, ILI9320_GRAM_VERT_ADD, 0x00);
+
+	ret = ili9320_write_regs(lcd, vgg_gamma, ARRAY_SIZE(vgg_gamma));
+	if (ret != 0)
+		goto err_vgg3;
+
+	ili9320_write(lcd, ILI9320_HORIZ_START, 0x0);
+	ili9320_write(lcd, ILI9320_HORIZ_END, cfg->hsize - 1);
+	ili9320_write(lcd, ILI9320_VERT_START, 0x0);
+	ili9320_write(lcd, ILI9320_VERT_END, cfg->vsize - 1);
+
+	ili9320_write(lcd, ILI9320_DRIVER2,
+		      ILI9320_DRIVER2_NL(((cfg->vsize - 240) / 8) + 0x1D));
+
+	ili9320_write(lcd, ILI9320_BASE_IMAGE, 0x1);
+	ili9320_write(lcd, ILI9320_VERT_SCROLL, 0x00);
+
+	for (addr = ILI9320_PARTIAL1_POSITION; addr <= ILI9320_PARTIAL2_END;
+	     addr++) {
+		ili9320_write(lcd, addr, 0x0);
+	}
+
+	ili9320_write(lcd, ILI9320_INTERFACE1, 0x10);
+	ili9320_write(lcd, ILI9320_INTERFACE2, cfg->interface2);
+	ili9320_write(lcd, ILI9320_INTERFACE3, cfg->interface3);
+	ili9320_write(lcd, ILI9320_INTERFACE4, cfg->interface4);
+	ili9320_write(lcd, ILI9320_INTERFACE5, cfg->interface5);
+	ili9320_write(lcd, ILI9320_INTERFACE6, cfg->interface6);
+
+	lcd->display1 = (ILI9320_DISPLAY1_D(3) | ILI9320_DISPLAY1_DTE |
+			 ILI9320_DISPLAY1_GON | ILI9320_DISPLAY1_BASEE |
+			 0x40);
+
+	ili9320_write(lcd, ILI9320_DISPLAY1, lcd->display1);
+
+	return 0;
+
+ err_vgg3:
+ err_vgg2:
+ err_vgg:
+ err_initial:
+	return ret;
+}
+
+#ifdef CONFIG_PM
+static int vgg2432a4_suspend(struct spi_device *spi, pm_message_t state)
+{
+	return ili9320_suspend(dev_get_drvdata(&spi->dev), state);
+}
+
+static int vgg2432a4_resume(struct spi_device *spi)
+{
+	return ili9320_resume(dev_get_drvdata(&spi->dev));
+}
+#else
+#define vgg2432a4_suspend	NULL
+#define vgg2432a4_resume 	NULL
+#endif
+
+static struct ili9320_client vgg2432a4_client = {
+	.name	= "VGG2432A4",
+	.init	= vgg2432a4_lcd_init,
+};
+
+/* Device probe */
+
+static int __devinit vgg2432a4_probe(struct spi_device *spi)
+{
+	int ret;
+
+	ret = ili9320_probe_spi(spi, &vgg2432a4_client);
+	if (ret != 0) {
+		dev_err(&spi->dev, "failed to initialise ili9320\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static int __devexit vgg2432a4_remove(struct spi_device *spi)
+{
+	return ili9320_remove(dev_get_drvdata(&spi->dev));
+}
+
+static void vgg2432a4_shutdown(struct spi_device *spi)
+{
+	ili9320_shutdown(dev_get_drvdata(&spi->dev));
+}
+
+static struct spi_driver vgg2432a4_driver = {
+	.driver = {
+		.name		= "VGG2432A4",
+		.owner		= THIS_MODULE,
+	},
+	.probe		= vgg2432a4_probe,
+	.remove		= __devexit_p(vgg2432a4_remove),
+	.shutdown	= vgg2432a4_shutdown,
+	.suspend	= vgg2432a4_suspend,
+	.resume		= vgg2432a4_resume,
+};
+
+/* Device driver initialisation */
+
+static int __init vgg2432a4_init(void)
+{
+	return spi_register_driver(&vgg2432a4_driver);
+}
+
+static void __exit vgg2432a4_exit(void)
+{
+	spi_unregister_driver(&vgg2432a4_driver);
+}
+
+module_init(vgg2432a4_init);
+module_exit(vgg2432a4_exit);
+
+MODULE_AUTHOR("Ben Dooks <ben-linux@fluff.org>");
+MODULE_DESCRIPTION("VGG2432A4 LCD Driver");
+MODULE_LICENSE("GPL v2");
+
+
diff --git a/include/video/ili9320.h b/include/video/ili9320.h
new file mode 100644
index 000000000000..e5d1622e3f33
--- /dev/null
+++ b/include/video/ili9320.h
@@ -0,0 +1,201 @@
+/* include/video/ili9320.c
+ *
+ * ILI9320 LCD controller configuration control.
+ *
+ * Copyright 2007 Simtec Electronics
+ *	Ben Dooks <ben@simtec.co.uk>
+ *
+ * http://armlinux.simtec.co.uk/
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+*/
+
+#define ILI9320_REG(x)	(x)
+
+#define ILI9320_INDEX			ILI9320_REG(0x00)
+
+#define ILI9320_OSCILATION		ILI9320_REG(0x00)
+#define ILI9320_DRIVER			ILI9320_REG(0x01)
+#define ILI9320_DRIVEWAVE		ILI9320_REG(0x02)
+#define ILI9320_ENTRYMODE		ILI9320_REG(0x03)
+#define ILI9320_RESIZING		ILI9320_REG(0x04)
+#define ILI9320_DISPLAY1		ILI9320_REG(0x07)
+#define ILI9320_DISPLAY2		ILI9320_REG(0x08)
+#define ILI9320_DISPLAY3		ILI9320_REG(0x09)
+#define ILI9320_DISPLAY4		ILI9320_REG(0x0A)
+#define ILI9320_RGB_IF1			ILI9320_REG(0x0C)
+#define ILI9320_FRAMEMAKER		ILI9320_REG(0x0D)
+#define ILI9320_RGB_IF2			ILI9320_REG(0x0F)
+
+#define ILI9320_POWER1			ILI9320_REG(0x10)
+#define ILI9320_POWER2			ILI9320_REG(0x11)
+#define ILI9320_POWER3			ILI9320_REG(0x12)
+#define ILI9320_POWER4			ILI9320_REG(0x13)
+#define ILI9320_GRAM_HORIZ_ADDR		ILI9320_REG(0x20)
+#define ILI9320_GRAM_VERT_ADD		ILI9320_REG(0x21)
+#define ILI9320_POWER7			ILI9320_REG(0x29)
+#define ILI9320_FRAME_RATE_COLOUR	ILI9320_REG(0x2B)
+
+#define ILI9320_GAMMA1			ILI9320_REG(0x30)
+#define ILI9320_GAMMA2			ILI9320_REG(0x31)
+#define ILI9320_GAMMA3			ILI9320_REG(0x32)
+#define ILI9320_GAMMA4			ILI9320_REG(0x35)
+#define ILI9320_GAMMA5			ILI9320_REG(0x36)
+#define ILI9320_GAMMA6			ILI9320_REG(0x37)
+#define ILI9320_GAMMA7			ILI9320_REG(0x38)
+#define ILI9320_GAMMA8			ILI9320_REG(0x39)
+#define ILI9320_GAMMA9			ILI9320_REG(0x3C)
+#define ILI9320_GAMMA10			ILI9320_REG(0x3D)
+
+#define ILI9320_HORIZ_START		ILI9320_REG(0x50)
+#define ILI9320_HORIZ_END		ILI9320_REG(0x51)
+#define ILI9320_VERT_START		ILI9320_REG(0x52)
+#define ILI9320_VERT_END		ILI9320_REG(0x53)
+
+#define ILI9320_DRIVER2			ILI9320_REG(0x60)
+#define ILI9320_BASE_IMAGE		ILI9320_REG(0x61)
+#define ILI9320_VERT_SCROLL		ILI9320_REG(0x6a)
+
+#define ILI9320_PARTIAL1_POSITION	ILI9320_REG(0x80)
+#define ILI9320_PARTIAL1_START		ILI9320_REG(0x81)
+#define ILI9320_PARTIAL1_END		ILI9320_REG(0x82)
+#define ILI9320_PARTIAL2_POSITION	ILI9320_REG(0x83)
+#define ILI9320_PARTIAL2_START		ILI9320_REG(0x84)
+#define ILI9320_PARTIAL2_END		ILI9320_REG(0x85)
+
+#define ILI9320_INTERFACE1		ILI9320_REG(0x90)
+#define ILI9320_INTERFACE2		ILI9320_REG(0x92)
+#define ILI9320_INTERFACE3		ILI9320_REG(0x93)
+#define ILI9320_INTERFACE4		ILI9320_REG(0x95)
+#define ILI9320_INTERFACE5		ILI9320_REG(0x97)
+#define ILI9320_INTERFACE6		ILI9320_REG(0x98)
+
+/* Register contents definitions. */
+
+#define ILI9320_OSCILATION_OSC		(1 << 0)
+
+#define ILI9320_DRIVER_SS		(1 << 8)
+#define ILI9320_DRIVER_SM		(1 << 10)
+
+#define ILI9320_DRIVEWAVE_EOR		(1 << 8)
+#define ILI9320_DRIVEWAVE_BC		(1 << 9)
+#define ILI9320_DRIVEWAVE_MUSTSET	(1 << 10)
+
+#define ILI9320_ENTRYMODE_AM		(1 << 3)
+#define ILI9320_ENTRYMODE_ID(x)		((x) << 4)
+#define ILI9320_ENTRYMODE_ORG		(1 << 7)
+#define ILI9320_ENTRYMODE_HWM		(1 << 8)
+#define ILI9320_ENTRYMODE_BGR		(1 << 12)
+#define ILI9320_ENTRYMODE_DFM		(1 << 14)
+#define ILI9320_ENTRYMODE_TRI		(1 << 15)
+
+
+#define ILI9320_RESIZING_RSZ(x)		((x) << 0)
+#define ILI9320_RESIZING_RCH(x)		((x) << 4)
+#define ILI9320_RESIZING_RCV(x)		((x) << 8)
+
+
+#define ILI9320_DISPLAY1_D(x)		((x) << 0)
+#define ILI9320_DISPLAY1_CL		(1 << 3)
+#define ILI9320_DISPLAY1_DTE		(1 << 4)
+#define ILI9320_DISPLAY1_GON		(1 << 5)
+#define ILI9320_DISPLAY1_BASEE		(1 << 8)
+#define ILI9320_DISPLAY1_PTDE(x)	((x) << 12)
+
+
+#define ILI9320_DISPLAY2_BP(x)		((x) << 0)
+#define ILI9320_DISPLAY2_FP(x)		((x) << 8)
+
+
+#define ILI9320_RGBIF1_RIM_RGB18	(0 << 0)
+#define ILI9320_RGBIF1_RIM_RGB16	(1 << 0)
+#define ILI9320_RGBIF1_RIM_RGB6		(2 << 0)
+
+#define ILI9320_RGBIF1_CLK_INT		(0 << 4)
+#define ILI9320_RGBIF1_CLK_RGBIF	(1 << 4)
+#define ILI9320_RGBIF1_CLK_VSYNC	(2 << 4)
+
+#define ILI9320_RGBIF1_RM		(1 << 8)
+
+#define ILI9320_RGBIF1_ENC_FRAMES(x)	(((x) - 1)<< 13)
+
+#define ILI9320_RGBIF2_DPL		(1 << 0)
+#define ILI9320_RGBIF2_EPL		(1 << 1)
+#define ILI9320_RGBIF2_HSPL		(1 << 3)
+#define ILI9320_RGBIF2_VSPL		(1 << 4)
+
+
+#define ILI9320_POWER1_SLP		(1 << 1)
+#define ILI9320_POWER1_DSTB		(1 << 2)
+#define ILI9320_POWER1_AP(x)		((x) << 4)
+#define ILI9320_POWER1_APE		(1 << 7)
+#define ILI9320_POWER1_BT(x)		((x) << 8)
+#define ILI9320_POWER1_SAP		(1 << 12)
+
+
+#define ILI9320_POWER2_VC(x)		((x) << 0)
+#define ILI9320_POWER2_DC0(x)		((x) << 4)
+#define ILI9320_POWER2_DC1(x)		((x) << 8)
+
+
+#define ILI9320_POWER3_VRH(x)		((x) << 0)
+#define ILI9320_POWER3_PON		(1 << 4)
+#define ILI9320_POWER3_VCMR		(1 << 8)
+
+
+#define ILI9320_POWER4_VREOUT(x)	((x) << 8)
+
+
+#define ILI9320_DRIVER2_SCNL(x)		((x) << 0)
+#define ILI9320_DRIVER2_NL(x)		((x) << 8)
+#define ILI9320_DRIVER2_GS		(1 << 15)
+
+
+#define ILI9320_BASEIMAGE_REV		(1 << 0)
+#define ILI9320_BASEIMAGE_VLE		(1 << 1)
+#define ILI9320_BASEIMAGE_NDL		(1 << 2)
+
+
+#define ILI9320_INTERFACE4_RTNE(x)	(x)
+#define ILI9320_INTERFACE4_DIVE(x)	((x) << 8)
+
+/* SPI interface definitions */
+
+#define ILI9320_SPI_IDCODE		(0x70)
+#define ILI9320_SPI_ID(x)		((x) << 2)
+#define ILI9320_SPI_READ		(0x01)
+#define ILI9320_SPI_WRITE		(0x00)
+#define ILI9320_SPI_DATA		(0x02)
+#define ILI9320_SPI_INDEX		(0x00)
+
+/* platform data to pass configuration from lcd */
+
+enum ili9320_suspend {
+	ILI9320_SUSPEND_OFF,
+	ILI9320_SUSPEND_DEEP,
+};
+
+struct ili9320_platdata {
+	unsigned short	hsize;
+	unsigned short	vsize;
+
+	enum ili9320_suspend suspend;
+
+	/* set the reset line, 0 = reset asserted, 1 = normal */
+	void		(*reset)(unsigned int val);
+
+	unsigned short	entry_mode;
+	unsigned short	display2;
+	unsigned short	display3;
+	unsigned short	display4;
+	unsigned short	rgb_if1;
+	unsigned short	rgb_if2;
+	unsigned short	interface2;
+	unsigned short	interface3;
+	unsigned short	interface4;
+	unsigned short	interface5;
+	unsigned short	interface6;
+};
+

From 0c531360ed504aa0ce995fcb8ef08e82b6534d0b Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Wed, 23 Jul 2008 21:31:38 -0700
Subject: [PATCH 0506/1171] lcd: add lcd_device to check_fb() entry in lcd_ops

Add the lcd_device being checked to the check_fb entry of lcd_ops.  This
ensures that any driver using this to check against it's own state can do
so, and also makes all the calls in lcd_ops more orthogonal in their
arguments.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/backlight/lcd.c    | 2 +-
 drivers/video/bf54x-lq043fb.c    | 2 +-
 drivers/video/bfin-t350mcqb-fb.c | 2 +-
 include/linux/lcd.h              | 2 +-
 4 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/video/backlight/lcd.c b/drivers/video/backlight/lcd.c
index 299fd318dd45..b15b2b84a6f7 100644
--- a/drivers/video/backlight/lcd.c
+++ b/drivers/video/backlight/lcd.c
@@ -33,7 +33,7 @@ static int fb_notifier_callback(struct notifier_block *self,
 	ld = container_of(self, struct lcd_device, fb_notif);
 	mutex_lock(&ld->ops_lock);
 	if (ld->ops)
-		if (!ld->ops->check_fb || ld->ops->check_fb(evdata->info))
+		if (!ld->ops->check_fb || ld->ops->check_fb(ld, evdata->info))
 			ld->ops->set_power(ld, *(int *)evdata->data);
 	mutex_unlock(&ld->ops_lock);
 	return 0;
diff --git a/drivers/video/bf54x-lq043fb.c b/drivers/video/bf54x-lq043fb.c
index 49834a67a623..940467aed13f 100644
--- a/drivers/video/bf54x-lq043fb.c
+++ b/drivers/video/bf54x-lq043fb.c
@@ -478,7 +478,7 @@ static int bfin_lcd_set_contrast(struct lcd_device *dev, int contrast)
 	return 0;
 }
 
-static int bfin_lcd_check_fb(struct fb_info *fi)
+static int bfin_lcd_check_fb(struct lcd_device *dev, struct fb_info *fi)
 {
 	if (!fi || (fi == &bfin_bf54x_fb))
 		return 1;
diff --git a/drivers/video/bfin-t350mcqb-fb.c b/drivers/video/bfin-t350mcqb-fb.c
index 135d6dd7e672..7d1b819e501c 100644
--- a/drivers/video/bfin-t350mcqb-fb.c
+++ b/drivers/video/bfin-t350mcqb-fb.c
@@ -396,7 +396,7 @@ static int bfin_lcd_set_contrast(struct lcd_device *dev, int contrast)
 	return 0;
 }
 
-static int bfin_lcd_check_fb(struct fb_info *fi)
+static int bfin_lcd_check_fb(struct lcd_device *dev, struct fb_info *fi)
 {
 	if (!fi || (fi == &bfin_t350mcqb_fb))
 		return 1;
diff --git a/include/linux/lcd.h b/include/linux/lcd.h
index 1d379787f2e7..173febac6656 100644
--- a/include/linux/lcd.h
+++ b/include/linux/lcd.h
@@ -47,7 +47,7 @@ struct lcd_ops {
         int (*set_contrast)(struct lcd_device *, int contrast);
 	/* Check if given framebuffer device is the one LCD is bound to;
 	   return 0 if not, !=0 if it is. If NULL, lcd always matches the fb. */
-	int (*check_fb)(struct fb_info *);
+	int (*check_fb)(struct lcd_device *, struct fb_info *);
 };
 
 struct lcd_device {

From c25826a7cf1c61b5c6e6db8365172eb97ef39ef3 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Wed, 23 Jul 2008 21:31:38 -0700
Subject: [PATCH 0507/1171] lcd: add platform_lcd driver

Add a platform_lcd driver to allow boards with simple lcd power controls
to register themselves easily.

[akpm@linux-foundation.org: build fix]
Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Cc: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/backlight/Kconfig        |   7 +
 drivers/video/backlight/Makefile       |   1 +
 drivers/video/backlight/platform_lcd.c | 172 +++++++++++++++++++++++++
 include/video/platform_lcd.h           |  21 +++
 4 files changed, 201 insertions(+)
 create mode 100644 drivers/video/backlight/platform_lcd.c
 create mode 100644 include/video/platform_lcd.h

diff --git a/drivers/video/backlight/Kconfig b/drivers/video/backlight/Kconfig
index a5b3a92ffdc1..b289e197e55d 100644
--- a/drivers/video/backlight/Kconfig
+++ b/drivers/video/backlight/Kconfig
@@ -53,6 +53,13 @@ config LCD_VGG2432A4
 	  If you have a VGG2432A4 panel based on the ILI9320 controller chip
 	  then say y to include a power driver for it.
 
+config LCD_PLATFORM
+	tristate "Platform LCD controls"
+	depends on LCD_CLASS_DEVICE
+	help
+	  This driver provides a platform-device registered LCD power
+	  control interface.
+
 #
 # Backlight
 #
diff --git a/drivers/video/backlight/Makefile b/drivers/video/backlight/Makefile
index 366d84e380cf..7d31c14088aa 100644
--- a/drivers/video/backlight/Makefile
+++ b/drivers/video/backlight/Makefile
@@ -3,6 +3,7 @@
 obj-$(CONFIG_LCD_CLASS_DEVICE)     += lcd.o
 obj-$(CONFIG_LCD_LTV350QV)	   += ltv350qv.o
 obj-$(CONFIG_LCD_ILI9320)	   += ili9320.o
+obj-$(CONFIG_LCD_PLATFORM)	   += platform_lcd.o
 obj-$(CONFIG_LCD_VGG2432A4)	   += vgg2432a4.o
 
 obj-$(CONFIG_BACKLIGHT_CLASS_DEVICE) += backlight.o
diff --git a/drivers/video/backlight/platform_lcd.c b/drivers/video/backlight/platform_lcd.c
new file mode 100644
index 000000000000..72d44dbfce82
--- /dev/null
+++ b/drivers/video/backlight/platform_lcd.c
@@ -0,0 +1,172 @@
+/* drivers/video/backlight/platform_lcd.c
+ *
+ * Copyright 2008 Simtec Electronics
+ *	Ben Dooks <ben@simtec.co.uk>
+ *
+ * Generic platform-device LCD power control interface.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+*/
+
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/fb.h>
+#include <linux/backlight.h>
+#include <linux/lcd.h>
+
+#include <video/platform_lcd.h>
+
+struct platform_lcd {
+	struct device		*us;
+	struct lcd_device	*lcd;
+	struct plat_lcd_data	*pdata;
+
+	unsigned int		 power;
+	unsigned int		 suspended : 1;
+};
+
+static inline struct platform_lcd *to_our_lcd(struct lcd_device *lcd)
+{
+	return lcd_get_data(lcd);
+}
+
+static int platform_lcd_get_power(struct lcd_device *lcd)
+{
+	struct platform_lcd *plcd = to_our_lcd(lcd);
+
+	return plcd->power;
+}
+
+static int platform_lcd_set_power(struct lcd_device *lcd, int power)
+{
+	struct platform_lcd *plcd = to_our_lcd(lcd);
+	int lcd_power = 1;
+
+	if (power == FB_BLANK_POWERDOWN || plcd->suspended)
+		lcd_power = 0;
+
+	plcd->pdata->set_power(plcd->pdata, lcd_power);
+	plcd->power = power;
+
+	return 0;
+}
+
+static int platform_lcd_match(struct lcd_device *lcd, struct fb_info *info)
+{
+	struct platform_lcd *plcd = to_our_lcd(lcd);
+	struct plat_lcd_data *pdata = plcd->pdata;
+
+	if (pdata->match_fb)
+		return pdata->match_fb(pdata, info);
+
+	return plcd->us->parent == info->device;
+}
+
+static struct lcd_ops platform_lcd_ops = {
+	.get_power	= platform_lcd_get_power,
+	.set_power	= platform_lcd_set_power,
+	.check_fb	= platform_lcd_match,
+};
+
+static int __devinit platform_lcd_probe(struct platform_device *pdev)
+{
+	struct plat_lcd_data *pdata;
+	struct platform_lcd *plcd;
+	struct device *dev = &pdev->dev;
+	int err;
+
+	pdata = pdev->dev.platform_data;
+	if (!pdata) {
+		dev_err(dev, "no platform data supplied\n");
+		return -EINVAL;
+	}
+
+	plcd = kzalloc(sizeof(struct platform_lcd), GFP_KERNEL);
+	if (!plcd) {
+		dev_err(dev, "no memory for state\n");
+		return -ENOMEM;
+	}
+
+	plcd->us = dev;
+	plcd->pdata = pdata;
+	plcd->lcd = lcd_device_register("platform-lcd", dev,
+					plcd, &platform_lcd_ops);
+	if (IS_ERR(plcd->lcd)) {
+		dev_err(dev, "cannot register lcd device\n");
+		err = PTR_ERR(plcd->lcd);
+		goto err_mem;
+	}
+
+	platform_set_drvdata(pdev, plcd);
+	return 0;
+
+ err_mem:
+	kfree(plcd);
+	return err;
+}
+
+static int __devexit platform_lcd_remove(struct platform_device *pdev)
+{
+	struct platform_lcd *plcd = platform_get_drvdata(pdev);
+
+	lcd_device_unregister(plcd->lcd);
+	kfree(plcd);
+
+	return 0;
+}
+
+#ifdef CONFIG_PM
+static int platform_lcd_suspend(struct platform_device *pdev, pm_message_t st)
+{
+	struct platform_lcd *plcd = platform_get_drvdata(pdev);
+
+	plcd->suspended = 1;
+	platform_lcd_set_power(plcd->lcd, plcd->power);
+
+	return 0;
+}
+
+static int platform_lcd_resume(struct platform_device *pdev)
+{
+	struct platform_lcd *plcd = platform_get_drvdata(pdev);
+
+	plcd->suspended = 0;
+	platform_lcd_set_power(plcd->lcd, plcd->power);
+
+	return 0;
+}
+#else
+#define platform_lcd_suspend NULL
+#define platform_lcd_resume NULL
+#endif
+
+static struct platform_driver platform_lcd_driver = {
+	.driver		= {
+		.name	= "platform-lcd",
+		.owner	= THIS_MODULE,
+	},
+	.probe		= platform_lcd_probe,
+	.remove		= __devexit_p(platform_lcd_remove),
+	.suspend        = platform_lcd_suspend,
+	.resume         = platform_lcd_resume,
+};
+
+static int __init platform_lcd_init(void)
+{
+	return platform_driver_register(&platform_lcd_driver);
+}
+
+static void __exit platform_lcd_cleanup(void)
+{
+	platform_driver_unregister(&platform_lcd_driver);
+}
+
+module_init(platform_lcd_init);
+module_exit(platform_lcd_cleanup);
+
+MODULE_AUTHOR("Ben Dooks <ben-linux@fluff.org>");
+MODULE_LICENSE("GPL v2");
+MODULE_ALIAS("platform:platform-lcd");
diff --git a/include/video/platform_lcd.h b/include/video/platform_lcd.h
new file mode 100644
index 000000000000..ad3bdfe743b2
--- /dev/null
+++ b/include/video/platform_lcd.h
@@ -0,0 +1,21 @@
+/* include/video/platform_lcd.h
+ *
+ * Copyright 2008 Simtec Electronics
+ *	Ben Dooks <ben@simtec.co.uk>
+ *
+ * Generic platform-device LCD power control interface.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+*/
+
+struct plat_lcd_data;
+struct fb_info;
+
+struct plat_lcd_data {
+	void	(*set_power)(struct plat_lcd_data *, unsigned int power);
+	int	(*match_fb)(struct plat_lcd_data *, struct fb_info *);
+};
+

From 6b51d51a9d24719f905ba9657b29e04efd82a7ea Mon Sep 17 00:00:00 2001
From: Timur Tabi <timur@freescale.com>
Date: Wed, 23 Jul 2008 21:31:39 -0700
Subject: [PATCH 0508/1171] fsl-diu-fb: update Freescale DIU driver to use
 page_alloc_exact()

Update the Freescale DIU driver to use page_alloc_exact() to allocate a
DMA buffer.  This also eliminates the rheap-based memory allocator.  We
can do this now because commit 6ccf61f9 allows us to allocate 8MB
physically- contiguous memory blocks.

[akpm@linux-foundation.org: fix printk warnings]
Signed-off-by: Timur Tabi <timur@freescale.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/fsl-diu-fb.c | 60 ++++++++++++++------------------------
 1 file changed, 22 insertions(+), 38 deletions(-)

diff --git a/drivers/video/fsl-diu-fb.c b/drivers/video/fsl-diu-fb.c
index 09d7e22c6fef..9cd36c223d33 100644
--- a/drivers/video/fsl-diu-fb.c
+++ b/drivers/video/fsl-diu-fb.c
@@ -279,58 +279,42 @@ static struct diu_hw dr = {
 
 static struct diu_pool pool;
 
-/*	To allocate memory for framebuffer. First try __get_free_pages(). If it
- *	fails, try rh_alloc. The reason is __get_free_pages() cannot allocate
- *	very large memory (more than 4MB). We don't want to allocate all memory
- *	in rheap since small memory allocation/deallocation will fragment the
- *	rheap and make the furture large allocation fail.
+/**
+ * fsl_diu_alloc - allocate memory for the DIU
+ * @size: number of bytes to allocate
+ * @param: returned physical address of memory
+ *
+ * This function allocates a physically-contiguous block of memory.
  */
-
-static void *fsl_diu_alloc(unsigned long size, phys_addr_t *phys)
+static void *fsl_diu_alloc(size_t size, phys_addr_t *phys)
 {
 	void *virt;
 
-	pr_debug("size=%lu\n", size);
+	pr_debug("size=%zu\n", size);
 
-	virt = (void *)__get_free_pages(GFP_DMA | __GFP_ZERO, get_order(size));
+	virt = alloc_pages_exact(size, GFP_DMA | __GFP_ZERO);
 	if (virt) {
 		*phys = virt_to_phys(virt);
-		pr_debug("virt %p, phys=%llx\n", virt, (uint64_t) *phys);
-		return virt;
+		pr_debug("virt=%p phys=%llx\n", virt,
+			(unsigned long long)*phys);
 	}
-	if (!diu_ops.diu_mem) {
-		printk(KERN_INFO "%s: no diu_mem."
-			" To reserve more memory, put 'diufb=15M' "
-			"in the command line\n", __func__);
-		return NULL;
-	}
-
-	virt = (void *)rh_alloc(&diu_ops.diu_rh_info, size, "DIU");
-	if (virt) {
-		*phys = virt_to_bus(virt);
-		memset(virt, 0, size);
-	}
-
-	pr_debug("rh virt=%p phys=%llx\n", virt, (unsigned long long)*phys);
 
 	return virt;
 }
 
-static void fsl_diu_free(void *p, unsigned long size)
+/**
+ * fsl_diu_free - release DIU memory
+ * @virt: pointer returned by fsl_diu_alloc()
+ * @size: number of bytes allocated by fsl_diu_alloc()
+ *
+ * This function releases memory allocated by fsl_diu_alloc().
+ */
+static void fsl_diu_free(void *virt, size_t size)
 {
-	pr_debug("p=%p size=%lu\n", p, size);
+	pr_debug("virt=%p size=%zu\n", virt, size);
 
-	if (!p)
-		return;
-
-	if ((p >= diu_ops.diu_mem) &&
-	    (p < (diu_ops.diu_mem + diu_ops.diu_size))) {
-		pr_debug("rh\n");
-		rh_free(&diu_ops.diu_rh_info, (unsigned long) p);
-	} else {
-		pr_debug("dma\n");
-		free_pages((unsigned long)p, get_order(size));
-	}
+	if (virt && size)
+		free_pages_exact(virt, size);
 }
 
 static int fsl_diu_enable_panel(struct fb_info *info)

From 5abe3b4063f16245b8fafbff37bd93814eb8e363 Mon Sep 17 00:00:00 2001
From: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Date: Wed, 23 Jul 2008 21:31:40 -0700
Subject: [PATCH 0509/1171] fbdev: add new Cobalt LCD framebuffer driver

Add new Cobalt LCD framebuffer driver.

[akpm@linux-foundation.org: fix build]
Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/Kconfig        |   4 +
 drivers/video/Makefile       |   1 +
 drivers/video/cobalt_lcdfb.c | 371 +++++++++++++++++++++++++++++++++++
 3 files changed, 376 insertions(+)
 create mode 100644 drivers/video/cobalt_lcdfb.c

diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig
index 80fa066416bf..c1b7db843411 100644
--- a/drivers/video/Kconfig
+++ b/drivers/video/Kconfig
@@ -1987,6 +1987,10 @@ config FB_AM200EPD
          This enables support for the Metronome display controller used on
          the E-Ink AM-200 EPD devkit.
 
+config FB_COBALT
+	tristate "Cobalt server LCD frame buffer support"
+	depends on FB && MIPS_COBALT
+
 config FB_VIRTUAL
 	tristate "Virtual Frame Buffer support (ONLY FOR TESTING!)"
 	depends on FB
diff --git a/drivers/video/Makefile b/drivers/video/Makefile
index 4809f8b9bb21..358032597757 100644
--- a/drivers/video/Makefile
+++ b/drivers/video/Makefile
@@ -109,6 +109,7 @@ obj-$(CONFIG_FB_S1D13XXX)	  += s1d13xxxfb.o
 obj-$(CONFIG_FB_IMX)              += imxfb.o
 obj-$(CONFIG_FB_S3C2410)	  += s3c2410fb.o
 obj-$(CONFIG_FB_FSL_DIU)	  += fsl-diu-fb.o
+obj-$(CONFIG_FB_COBALT)           += cobalt_lcdfb.o
 obj-$(CONFIG_FB_PNX4008_DUM)	  += pnx4008/
 obj-$(CONFIG_FB_PNX4008_DUM_RGB)  += pnx4008/
 obj-$(CONFIG_FB_IBM_GXT4500)	  += gxt4500.o
diff --git a/drivers/video/cobalt_lcdfb.c b/drivers/video/cobalt_lcdfb.c
new file mode 100644
index 000000000000..7bad24ed04ef
--- /dev/null
+++ b/drivers/video/cobalt_lcdfb.c
@@ -0,0 +1,371 @@
+/*
+ *  Cobalt server LCD frame buffer driver.
+ *
+ *  Copyright (C) 2008  Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+#include <linux/delay.h>
+#include <linux/fb.h>
+#include <linux/init.h>
+#include <linux/io.h>
+#include <linux/ioport.h>
+#include <linux/uaccess.h>
+#include <linux/platform_device.h>
+
+/*
+ * Cursor position address
+ * \X  0    1    2  ...  14   15
+ * Y+----+----+----+---+----+----+
+ * 0|0x00|0x01|0x02|...|0x0e|0x0f|
+ *  +----+----+----+---+----+----+
+ * 1|0x40|0x41|0x42|...|0x4e|0x4f|
+ *  +----+----+----+---+----+----+
+ */
+#define LCD_DATA_REG_OFFSET	0x10
+#define LCD_XRES_MAX		16
+#define LCD_YRES_MAX		2
+#define LCD_CHARS_MAX		32
+
+#define LCD_CLEAR		0x01
+#define LCD_CURSOR_MOVE_HOME	0x02
+#define LCD_RESET		0x06
+#define LCD_OFF			0x08
+#define LCD_CURSOR_OFF		0x0c
+#define LCD_CURSOR_BLINK_OFF	0x0e
+#define LCD_CURSOR_ON		0x0f
+#define LCD_ON			LCD_CURSOR_ON
+#define LCD_CURSOR_MOVE_LEFT	0x10
+#define LCD_CURSOR_MOVE_RIGHT	0x14
+#define LCD_DISPLAY_LEFT	0x18
+#define LCD_DISPLAY_RIGHT	0x1c
+#define LCD_PRERESET		0x3f	/* execute 4 times continuously */
+#define LCD_BUSY		0x80
+
+#define LCD_GRAPHIC_MODE	0x40
+#define LCD_TEXT_MODE		0x80
+#define LCD_CUR_POS_MASK	0x7f
+
+#define LCD_CUR_POS(x)		((x) & LCD_CUR_POS_MASK)
+#define LCD_TEXT_POS(x)		((x) | LCD_TEXT_MODE)
+
+static inline void lcd_write_control(struct fb_info *info, u8 control)
+{
+	writel((u32)control << 24, info->screen_base);
+}
+
+static inline u8 lcd_read_control(struct fb_info *info)
+{
+	return readl(info->screen_base) >> 24;
+}
+
+static inline void lcd_write_data(struct fb_info *info, u8 data)
+{
+	writel((u32)data << 24, info->screen_base + LCD_DATA_REG_OFFSET);
+}
+
+static inline u8 lcd_read_data(struct fb_info *info)
+{
+	return readl(info->screen_base + LCD_DATA_REG_OFFSET) >> 24;
+}
+
+static int lcd_busy_wait(struct fb_info *info)
+{
+	u8 val = 0;
+	int timeout = 10, retval = 0;
+
+	do {
+		val = lcd_read_control(info);
+		val &= LCD_BUSY;
+		if (val != LCD_BUSY)
+			break;
+
+		if (msleep_interruptible(1))
+			return -EINTR;
+
+		timeout--;
+	} while (timeout);
+
+	if (val == LCD_BUSY)
+		retval = -EBUSY;
+
+	return retval;
+}
+
+static void lcd_clear(struct fb_info *info)
+{
+	int i;
+
+	for (i = 0; i < 4; i++) {
+		udelay(150);
+
+		lcd_write_control(info, LCD_PRERESET);
+	}
+
+	udelay(150);
+
+	lcd_write_control(info, LCD_CLEAR);
+
+	udelay(150);
+
+	lcd_write_control(info, LCD_RESET);
+}
+
+static struct fb_fix_screeninfo cobalt_lcdfb_fix __initdata = {
+	.id		= "cobalt-lcd",
+	.type		= FB_TYPE_TEXT,
+	.type_aux	= FB_AUX_TEXT_MDA,
+	.visual		= FB_VISUAL_MONO01,
+	.line_length	= LCD_XRES_MAX,
+	.accel		= FB_ACCEL_NONE,
+};
+
+static ssize_t cobalt_lcdfb_read(struct fb_info *info, char __user *buf,
+				 size_t count, loff_t *ppos)
+{
+	char src[LCD_CHARS_MAX];
+	unsigned long pos;
+	int len, retval = 0;
+
+	pos = *ppos;
+	if (pos >= LCD_CHARS_MAX || count == 0)
+		return 0;
+
+	if (count > LCD_CHARS_MAX)
+		count = LCD_CHARS_MAX;
+
+	if (pos + count > LCD_CHARS_MAX)
+		count = LCD_CHARS_MAX - pos;
+
+	for (len = 0; len < count; len++) {
+		retval = lcd_busy_wait(info);
+		if (retval < 0)
+			break;
+
+		lcd_write_control(info, LCD_TEXT_POS(pos));
+
+		retval = lcd_busy_wait(info);
+		if (retval < 0)
+			break;
+
+		src[len] = lcd_read_data(info);
+		if (pos == 0x0f)
+			pos = 0x40;
+		else
+			pos++;
+	}
+
+	if (retval < 0 && signal_pending(current))
+		return -ERESTARTSYS;
+
+	if (copy_to_user(buf, src, len))
+		return -EFAULT;
+
+	*ppos += len;
+
+	return len;
+}
+
+static ssize_t cobalt_lcdfb_write(struct fb_info *info, const char __user *buf,
+				  size_t count, loff_t *ppos)
+{
+	char dst[LCD_CHARS_MAX];
+	unsigned long pos;
+	int len, retval = 0;
+
+	pos = *ppos;
+	if (pos >= LCD_CHARS_MAX || count == 0)
+		return 0;
+
+	if (count > LCD_CHARS_MAX)
+		count = LCD_CHARS_MAX;
+
+	if (pos + count > LCD_CHARS_MAX)
+		count = LCD_CHARS_MAX - pos;
+
+	if (copy_from_user(dst, buf, count))
+		return -EFAULT;
+
+	for (len = 0; len < count; len++) {
+		retval = lcd_busy_wait(info);
+		if (retval < 0)
+			break;
+
+		lcd_write_control(info, LCD_TEXT_POS(pos));
+
+		retval = lcd_busy_wait(info);
+		if (retval < 0)
+			break;
+
+		lcd_write_data(info, dst[len]);
+		if (pos == 0x0f)
+			pos = 0x40;
+		else
+			pos++;
+	}
+
+	if (retval < 0 && signal_pending(current))
+		return -ERESTARTSYS;
+
+	*ppos += len;
+
+	return len;
+}
+
+static int cobalt_lcdfb_blank(int blank_mode, struct fb_info *info)
+{
+	int retval;
+
+	retval = lcd_busy_wait(info);
+	if (retval < 0)
+		return retval;
+
+	switch (blank_mode) {
+	case FB_BLANK_UNBLANK:
+		lcd_write_control(info, LCD_ON);
+		break;
+	default:
+		lcd_write_control(info, LCD_OFF);
+		break;
+	}
+
+	return 0;
+}
+
+static int cobalt_lcdfb_cursor(struct fb_info *info, struct fb_cursor *cursor)
+{
+	u32 x, y;
+	int retval;
+
+	switch (cursor->set) {
+	case FB_CUR_SETPOS:
+		x = cursor->image.dx;
+		y = cursor->image.dy;
+		if (x >= LCD_XRES_MAX || y >= LCD_YRES_MAX)
+			return -EINVAL;
+
+		retval = lcd_busy_wait(info);
+		if (retval < 0)
+			return retval;
+
+		lcd_write_control(info,
+				  LCD_TEXT_POS(info->fix.line_length * y + x));
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	retval = lcd_busy_wait(info);
+	if (retval < 0)
+		return retval;
+
+	if (cursor->enable)
+		lcd_write_control(info, LCD_CURSOR_ON);
+	else
+		lcd_write_control(info, LCD_CURSOR_OFF);
+
+	return 0;
+}
+
+static struct fb_ops cobalt_lcd_fbops = {
+	.owner		= THIS_MODULE,
+	.fb_read	= cobalt_lcdfb_read,
+	.fb_write	= cobalt_lcdfb_write,
+	.fb_blank	= cobalt_lcdfb_blank,
+	.fb_cursor	= cobalt_lcdfb_cursor,
+};
+
+static int __init cobalt_lcdfb_probe(struct platform_device *dev)
+{
+	struct fb_info *info;
+	struct resource *res;
+	int retval;
+
+	info = framebuffer_alloc(0, &dev->dev);
+	if (!info)
+		return -ENOMEM;
+
+	res = platform_get_resource(dev, IORESOURCE_MEM, 0);
+	if (!res) {
+		framebuffer_release(info);
+		return -EBUSY;
+	}
+
+	info->screen_size = res->end - res->start + 1;
+	info->screen_base = ioremap(res->start, info->screen_size);
+	info->fbops = &cobalt_lcd_fbops;
+	info->fix = cobalt_lcdfb_fix;
+	info->fix.smem_start = res->start;
+	info->fix.smem_len = info->screen_size;
+	info->pseudo_palette = NULL;
+	info->par = NULL;
+	info->flags = FBINFO_DEFAULT;
+
+	retval = register_framebuffer(info);
+	if (retval < 0) {
+		iounmap(info->screen_base);
+		framebuffer_release(info);
+		return retval;
+	}
+
+	platform_set_drvdata(dev, info);
+
+	lcd_clear(info);
+
+	printk(KERN_INFO "fb%d: Cobalt server LCD frame buffer device\n",
+		info->node);
+
+	return 0;
+}
+
+static int __devexit cobalt_lcdfb_remove(struct platform_device *dev)
+{
+	struct fb_info *info;
+
+	info = platform_get_drvdata(dev);
+	if (info) {
+		iounmap(info->screen_base);
+		unregister_framebuffer(info);
+		framebuffer_release(info);
+	}
+
+	return 0;
+}
+
+static struct platform_driver cobalt_lcdfb_driver = {
+	.probe	= cobalt_lcdfb_probe,
+	.remove	= __devexit_p(cobalt_lcdfb_remove),
+	.driver	= {
+		.name	= "cobalt-lcd",
+		.owner	= THIS_MODULE,
+	},
+};
+
+static int __init cobalt_lcdfb_init(void)
+{
+	return platform_driver_register(&cobalt_lcdfb_driver);
+}
+
+static void __exit cobalt_lcdfb_exit(void)
+{
+	platform_driver_unregister(&cobalt_lcdfb_driver);
+}
+
+module_init(cobalt_lcdfb_init);
+module_exit(cobalt_lcdfb_exit);
+
+MODULE_LICENSE("GPL v2");
+MODULE_AUTHOR("Yoichi Yuasa");
+MODULE_DESCRIPTION("Cobalt server LCD frame buffer driver");

From 18b095d4b847bb08bf8a1bace7711a93d27732c0 Mon Sep 17 00:00:00 2001
From: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Date: Wed, 23 Jul 2008 21:31:41 -0700
Subject: [PATCH 0510/1171] drivers/char: remove old broken Cobalt LCD driver

Remove old broken Cobalt LCD driver.

Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Acked-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/Kconfig  |   7 -
 drivers/char/Makefile |   1 -
 drivers/char/lcd.c    | 516 ------------------------------------------
 drivers/char/lcd.h    | 154 -------------
 4 files changed, 678 deletions(-)
 delete mode 100644 drivers/char/lcd.c
 delete mode 100644 drivers/char/lcd.h

diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig
index e0bbbfb6a36b..67b07576f8bf 100644
--- a/drivers/char/Kconfig
+++ b/drivers/char/Kconfig
@@ -857,13 +857,6 @@ config DS1302
 
 endif # RTC_LIB
 
-config COBALT_LCD
-	bool "Support for Cobalt LCD"
-	depends on MIPS_COBALT
-	help
-	  This option enables support for the LCD display and buttons found
-	  on Cobalt systems through a misc device.
-
 config DTLK
 	tristate "Double Talk PC internal speech card support"
 	depends on ISA
diff --git a/drivers/char/Makefile b/drivers/char/Makefile
index dc5a327d72d5..4b6e736cfa02 100644
--- a/drivers/char/Makefile
+++ b/drivers/char/Makefile
@@ -88,7 +88,6 @@ obj-$(CONFIG_TOSHIBA)		+= toshiba.o
 obj-$(CONFIG_I8K)		+= i8k.o
 obj-$(CONFIG_DS1620)		+= ds1620.o
 obj-$(CONFIG_HW_RANDOM)		+= hw_random/
-obj-$(CONFIG_COBALT_LCD)	+= lcd.o
 obj-$(CONFIG_PPDEV)		+= ppdev.o
 obj-$(CONFIG_NWBUTTON)		+= nwbutton.o
 obj-$(CONFIG_NWFLASH)		+= nwflash.o
diff --git a/drivers/char/lcd.c b/drivers/char/lcd.c
deleted file mode 100644
index 1c29b20e4f4c..000000000000
--- a/drivers/char/lcd.c
+++ /dev/null
@@ -1,516 +0,0 @@
-/*
- * LCD, LED and Button interface for Cobalt
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 1996, 1997 by Andrew Bose
- *
- * Linux kernel version history:
- *       March 2001: Ported from 2.0.34  by Liam Davies
- *
- */
-#include <linux/types.h>
-#include <linux/errno.h>
-#include <linux/miscdevice.h>
-#include <linux/slab.h>
-#include <linux/ioport.h>
-#include <linux/fcntl.h>
-#include <linux/mc146818rtc.h>
-#include <linux/netdevice.h>
-#include <linux/sched.h>
-#include <linux/smp_lock.h>
-#include <linux/delay.h>
-
-#include <asm/io.h>
-#include <asm/uaccess.h>
-#include <asm/system.h>
-
-#include "lcd.h"
-
-static int lcd_ioctl(struct inode *inode, struct file *file,
-		     unsigned int cmd, unsigned long arg);
-
-static unsigned int lcd_present = 1;
-
-/* used in arch/mips/cobalt/reset.c */
-int led_state = 0;
-
-#if defined(CONFIG_TULIP) && 0
-
-#define MAX_INTERFACES	8
-static linkcheck_func_t linkcheck_callbacks[MAX_INTERFACES];
-static void *linkcheck_cookies[MAX_INTERFACES];
-
-int lcd_register_linkcheck_func(int iface_num, void *func, void *cookie)
-{
-	if (iface_num < 0 ||
-	    iface_num >= MAX_INTERFACES ||
-	    linkcheck_callbacks[iface_num] != NULL)
-		return -1;
-	linkcheck_callbacks[iface_num] = (linkcheck_func_t) func;
-	linkcheck_cookies[iface_num] = cookie;
-	return 0;
-}
-#endif
-
-static int lcd_ioctl(struct inode *inode, struct file *file,
-		     unsigned int cmd, unsigned long arg)
-{
-	struct lcd_display button_display;
-	unsigned long address, a;
-
-	switch (cmd) {
-	case LCD_On:
-		udelay(150);
-		BusyCheck();
-		LCDWriteInst(0x0F);
-		break;
-
-	case LCD_Off:
-		udelay(150);
-		BusyCheck();
-		LCDWriteInst(0x08);
-		break;
-
-	case LCD_Reset:
-		udelay(150);
-		LCDWriteInst(0x3F);
-		udelay(150);
-		LCDWriteInst(0x3F);
-		udelay(150);
-		LCDWriteInst(0x3F);
-		udelay(150);
-		LCDWriteInst(0x3F);
-		udelay(150);
-		LCDWriteInst(0x01);
-		udelay(150);
-		LCDWriteInst(0x06);
-		break;
-
-	case LCD_Clear:
-		udelay(150);
-		BusyCheck();
-		LCDWriteInst(0x01);
-		break;
-
-	case LCD_Cursor_Left:
-		udelay(150);
-		BusyCheck();
-		LCDWriteInst(0x10);
-		break;
-
-	case LCD_Cursor_Right:
-		udelay(150);
-		BusyCheck();
-		LCDWriteInst(0x14);
-		break;
-
-	case LCD_Cursor_Off:
-		udelay(150);
-		BusyCheck();
-		LCDWriteInst(0x0C);
-		break;
-
-	case LCD_Cursor_On:
-		udelay(150);
-		BusyCheck();
-		LCDWriteInst(0x0F);
-		break;
-
-	case LCD_Blink_Off:
-		udelay(150);
-		BusyCheck();
-		LCDWriteInst(0x0E);
-		break;
-
-	case LCD_Get_Cursor_Pos:{
-			struct lcd_display display;
-
-			udelay(150);
-			BusyCheck();
-			display.cursor_address = (LCDReadInst);
-			display.cursor_address =
-			    (display.cursor_address & 0x07F);
-			if (copy_to_user
-			    ((struct lcd_display *) arg, &display,
-			     sizeof(struct lcd_display)))
-				return -EFAULT;
-
-			break;
-		}
-
-
-	case LCD_Set_Cursor_Pos:{
-			struct lcd_display display;
-
-			if (copy_from_user
-			    (&display, (struct lcd_display *) arg,
-			     sizeof(struct lcd_display)))
-				return -EFAULT;
-
-			a = (display.cursor_address | kLCD_Addr);
-
-			udelay(150);
-			BusyCheck();
-			LCDWriteInst(a);
-
-			break;
-		}
-
-	case LCD_Get_Cursor:{
-			struct lcd_display display;
-
-			udelay(150);
-			BusyCheck();
-			display.character = LCDReadData;
-
-			if (copy_to_user
-			    ((struct lcd_display *) arg, &display,
-			     sizeof(struct lcd_display)))
-				return -EFAULT;
-			udelay(150);
-			BusyCheck();
-			LCDWriteInst(0x10);
-
-			break;
-		}
-
-	case LCD_Set_Cursor:{
-			struct lcd_display display;
-
-			if (copy_from_user
-			    (&display, (struct lcd_display *) arg,
-			     sizeof(struct lcd_display)))
-				return -EFAULT;
-
-			udelay(150);
-			BusyCheck();
-			LCDWriteData(display.character);
-			udelay(150);
-			BusyCheck();
-			LCDWriteInst(0x10);
-
-			break;
-		}
-
-
-	case LCD_Disp_Left:
-		udelay(150);
-		BusyCheck();
-		LCDWriteInst(0x18);
-		break;
-
-	case LCD_Disp_Right:
-		udelay(150);
-		BusyCheck();
-		LCDWriteInst(0x1C);
-		break;
-
-	case LCD_Home:
-		udelay(150);
-		BusyCheck();
-		LCDWriteInst(0x02);
-		break;
-
-	case LCD_Write:{
-			struct lcd_display display;
-			unsigned int index;
-
-
-			if (copy_from_user
-			    (&display, (struct lcd_display *) arg,
-			     sizeof(struct lcd_display)))
-				return -EFAULT;
-
-			udelay(150);
-			BusyCheck();
-			LCDWriteInst(0x80);
-			udelay(150);
-			BusyCheck();
-
-			for (index = 0; index < (display.size1); index++) {
-				udelay(150);
-				BusyCheck();
-				LCDWriteData(display.line1[index]);
-				BusyCheck();
-			}
-
-			udelay(150);
-			BusyCheck();
-			LCDWriteInst(0xC0);
-			udelay(150);
-			BusyCheck();
-			for (index = 0; index < (display.size2); index++) {
-				udelay(150);
-				BusyCheck();
-				LCDWriteData(display.line2[index]);
-			}
-
-			break;
-		}
-
-	case LCD_Read:{
-			struct lcd_display display;
-
-			BusyCheck();
-			for (address = kDD_R00; address <= kDD_R01;
-			     address++) {
-				a = (address | kLCD_Addr);
-
-				udelay(150);
-				BusyCheck();
-				LCDWriteInst(a);
-				udelay(150);
-				BusyCheck();
-				display.line1[address] = LCDReadData;
-			}
-
-			display.line1[0x27] = '\0';
-
-			for (address = kDD_R10; address <= kDD_R11;
-			     address++) {
-				a = (address | kLCD_Addr);
-
-				udelay(150);
-				BusyCheck();
-				LCDWriteInst(a);
-
-				udelay(150);
-				BusyCheck();
-				display.line2[address - 0x40] =
-				    LCDReadData;
-			}
-
-			display.line2[0x27] = '\0';
-
-			if (copy_to_user
-			    ((struct lcd_display *) arg, &display,
-			     sizeof(struct lcd_display)))
-				return -EFAULT;
-			break;
-		}
-
-//  set all GPIO leds to led_display.leds
-
-	case LED_Set:{
-			struct lcd_display led_display;
-
-
-			if (copy_from_user
-			    (&led_display, (struct lcd_display *) arg,
-			     sizeof(struct lcd_display)))
-				return -EFAULT;
-
-			led_state = led_display.leds;
-			LEDSet(led_state);
-
-			break;
-		}
-
-
-//  set only bit led_display.leds
-
-	case LED_Bit_Set:{
-			unsigned int i;
-			int bit = 1;
-			struct lcd_display led_display;
-
-
-			if (copy_from_user
-			    (&led_display, (struct lcd_display *) arg,
-			     sizeof(struct lcd_display)))
-				return -EFAULT;
-
-			for (i = 0; i < (int) led_display.leds; i++) {
-				bit = 2 * bit;
-			}
-
-			led_state = led_state | bit;
-			LEDSet(led_state);
-			break;
-		}
-
-//  clear only bit led_display.leds
-
-	case LED_Bit_Clear:{
-			unsigned int i;
-			int bit = 1;
-			struct lcd_display led_display;
-
-
-			if (copy_from_user
-			    (&led_display, (struct lcd_display *) arg,
-			     sizeof(struct lcd_display)))
-				return -EFAULT;
-
-			for (i = 0; i < (int) led_display.leds; i++) {
-				bit = 2 * bit;
-			}
-
-			led_state = led_state & ~bit;
-			LEDSet(led_state);
-			break;
-		}
-
-
-	case BUTTON_Read:{
-			button_display.buttons = GPIRead;
-			if (copy_to_user
-			    ((struct lcd_display *) arg, &button_display,
-			     sizeof(struct lcd_display)))
-				return -EFAULT;
-			break;
-		}
-
-	case LINK_Check:{
-			button_display.buttons =
-			    *((volatile unsigned long *) (0xB0100060));
-			if (copy_to_user
-			    ((struct lcd_display *) arg, &button_display,
-			     sizeof(struct lcd_display)))
-				return -EFAULT;
-			break;
-		}
-
-	case LINK_Check_2:{
-			int iface_num;
-
-			/* panel-utils should pass in the desired interface status is wanted for
-			 * in "buttons" of the structure.  We will set this to non-zero if the
-			 * link is in fact up for the requested interface.  --DaveM
-			 */
-			if (copy_from_user
-			    (&button_display, (struct lcd_display *) arg,
-			     sizeof(button_display)))
-				return -EFAULT;
-			iface_num = button_display.buttons;
-#if defined(CONFIG_TULIP) && 0
-			if (iface_num >= 0 &&
-			    iface_num < MAX_INTERFACES &&
-			    linkcheck_callbacks[iface_num] != NULL) {
-				button_display.buttons =
-				    linkcheck_callbacks[iface_num]
-				    (linkcheck_cookies[iface_num]);
-			} else
-#endif
-				button_display.buttons = 0;
-
-			if (__copy_to_user
-			    ((struct lcd_display *) arg, &button_display,
-			     sizeof(struct lcd_display)))
-				return -EFAULT;
-			break;
-		}
-
-	default:
-		return -EINVAL;
-
-	}
-
-	return 0;
-
-}
-
-static int lcd_open(struct inode *inode, struct file *file)
-{
-	cycle_kernel_lock();
-
-	if (!lcd_present)
-		return -ENXIO;
-	else
-		return 0;
-}
-
-/* Only RESET or NEXT counts as button pressed */
-
-static inline int button_pressed(void)
-{
-	unsigned long buttons = GPIRead;
-
-	if ((buttons == BUTTON_Next) || (buttons == BUTTON_Next_B)
-	    || (buttons == BUTTON_Reset_B))
-		return buttons;
-	return 0;
-}
-
-/* LED daemon sits on this and we wake him up once a key is pressed. */
-
-static int lcd_waiters = 0;
-
-static ssize_t lcd_read(struct file *file, char *buf,
-		     size_t count, loff_t *ofs)
-{
-	long buttons_now;
-
-	if (lcd_waiters > 0)
-		return -EINVAL;
-
-	lcd_waiters++;
-	while (((buttons_now = (long) button_pressed()) == 0) &&
-	       !(signal_pending(current))) {
-		msleep_interruptible(2000);
-	}
-	lcd_waiters--;
-
-	if (signal_pending(current))
-		return -ERESTARTSYS;
-	return buttons_now;
-}
-
-/*
- *	The various file operations we support.
- */
-
-static const struct file_operations lcd_fops = {
-	.read = lcd_read,
-	.ioctl = lcd_ioctl,
-	.open = lcd_open,
-};
-
-static struct miscdevice lcd_dev = {
-	MISC_DYNAMIC_MINOR,
-	"lcd",
-	&lcd_fops
-};
-
-static int lcd_init(void)
-{
-	int ret;
-	unsigned long data;
-
-	pr_info("%s\n", LCD_DRIVER);
-	ret = misc_register(&lcd_dev);
-	if (ret) {
-		printk(KERN_WARNING LCD "Unable to register misc device.\n");
-		return ret;
-	}
-
-	/* Check region? Naaah! Just snarf it up. */
-/*	request_region(RTC_PORT(0), RTC_IO_EXTENT, "lcd");*/
-
-	udelay(150);
-	data = LCDReadData;
-	if ((data & 0x000000FF) == (0x00)) {
-		lcd_present = 0;
-		pr_info(LCD "LCD Not Present\n");
-	} else {
-		lcd_present = 1;
-		WRITE_GAL(kGal_DevBank2PReg, kGal_DevBank2Cfg);
-		WRITE_GAL(kGal_DevBank3PReg, kGal_DevBank3Cfg);
-	}
-
-	return 0;
-}
-
-static void __exit lcd_exit(void)
-{
-	misc_deregister(&lcd_dev);
-}
-
-module_init(lcd_init);
-module_exit(lcd_exit);
-
-MODULE_AUTHOR("Andrew Bose");
-MODULE_LICENSE("GPL");
diff --git a/drivers/char/lcd.h b/drivers/char/lcd.h
deleted file mode 100644
index 290b3ff23b03..000000000000
--- a/drivers/char/lcd.h
+++ /dev/null
@@ -1,154 +0,0 @@
-/*
- * LED, LCD and Button panel driver for Cobalt
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 1996, 1997 by Andrew Bose
- *
- * Linux kernel version history:
- *       March 2001: Ported from 2.0.34  by Liam Davies
- *
- */
-
-// function headers
-
-#define LCD_CHARS_PER_LINE 40
-#define MAX_IDLE_TIME 120
-
-struct lcd_display {
-        unsigned buttons;
-        int size1;
-        int size2;
-        unsigned char line1[LCD_CHARS_PER_LINE];
-        unsigned char line2[LCD_CHARS_PER_LINE];
-        unsigned char cursor_address;
-        unsigned char character;
-        unsigned char leds;
-        unsigned char *RomImage;
-};
-
-
-
-#define LCD_DRIVER	"Cobalt LCD Driver v2.10"
-
-#define LCD		"lcd: "
-
-#define kLCD_IR		0x0F000000
-#define kLCD_DR		0x0F000010
-#define kGPI		0x0D000000
-#define kLED		0x0C000000
-
-#define kDD_R00         0x00
-#define kDD_R01         0x27
-#define kDD_R10         0x40
-#define kDD_R11         0x67
-
-#define kLCD_Addr       0x00000080
-
-#define LCDTimeoutValue	0xfff
-
-
-// Macros
-
-#define LCDWriteData(x)	outl((x << 24), kLCD_DR)
-#define LCDWriteInst(x)	outl((x << 24), kLCD_IR)
-
-#define LCDReadData	(inl(kLCD_DR) >> 24)
-#define LCDReadInst	(inl(kLCD_IR) >> 24)
-
-#define GPIRead		(inl(kGPI) >> 24)
-
-#define LEDSet(x)	outb((char)x, kLED)
-
-#define WRITE_GAL(x,y)	outl(y, 0x04000000 | (x))
-#define BusyCheck()	while ((LCDReadInst & 0x80) == 0x80)
-
-
-
-/*
- * Function command codes for io_ctl.
- */
-#define LCD_On			1
-#define LCD_Off			2
-#define LCD_Clear		3
-#define LCD_Reset		4
-#define LCD_Cursor_Left		5
-#define LCD_Cursor_Right	6
-#define LCD_Disp_Left		7
-#define LCD_Disp_Right		8
-#define LCD_Get_Cursor		9
-#define LCD_Set_Cursor		10
-#define LCD_Home		11
-#define LCD_Read		12
-#define LCD_Write		13
-#define LCD_Cursor_Off		14
-#define LCD_Cursor_On		15
-#define LCD_Get_Cursor_Pos	16
-#define LCD_Set_Cursor_Pos	17
-#define LCD_Blink_Off           18
-
-#define LED_Set			40
-#define LED_Bit_Set		41
-#define LED_Bit_Clear		42
-
-
-//  Button defs
-#define BUTTON_Read             50
-
-
-// Ethernet LINK check hackaroo
-#define LINK_Check              90
-#define LINK_Check_2		91
-
-//  Button patterns  _B - single layer lcd boards
-
-#define BUTTON_NONE               0x3F
-#define BUTTON_NONE_B             0xFE
-
-#define BUTTON_Left               0x3B
-#define BUTTON_Left_B             0xFA
-
-#define BUTTON_Right              0x37
-#define BUTTON_Right_B            0xDE
-
-#define BUTTON_Up                 0x2F
-#define BUTTON_Up_B               0xF6
-
-#define BUTTON_Down               0x1F
-#define BUTTON_Down_B             0xEE
-
-#define BUTTON_Next               0x3D
-#define BUTTON_Next_B             0x7E
-
-#define BUTTON_Enter              0x3E
-#define BUTTON_Enter_B            0xBE
-
-#define BUTTON_Reset_B            0xFC
-
-
-// debounce constants
-
-#define BUTTON_SENSE            160000
-#define BUTTON_DEBOUNCE		5000
-
-
-//  Galileo register stuff
-
-#define kGal_DevBank2Cfg        0x1466DB33
-#define kGal_DevBank2PReg       0x464
-#define kGal_DevBank3Cfg        0x146FDFFB
-#define kGal_DevBank3PReg       0x468
-
-// Network
-
-#define kIPADDR			1
-#define kNETMASK		2
-#define kGATEWAY		3
-#define kDNS			4
-
-#define kClassA			5
-#define kClassB			6
-#define kClassC			7
-

From be935d5b6301865b4e9ec35d79d398cedb3c82b7 Mon Sep 17 00:00:00 2001
From: Andres Salomon <dilinger@queued.net>
Date: Wed, 23 Jul 2008 21:31:41 -0700
Subject: [PATCH 0511/1171] lxfb: drop dead declarations from header

We never sent the gamma stuff upstream, and don't really care about it.
However, lx_[gs]_et_gamma prototypes snuck into lxfb.h anyways; there are
no definitions for them.  Drop the dead code.

Signed-off-by: Andres Salomon <dilinger@debian.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/geode/lxfb.h | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/video/geode/lxfb.h b/drivers/video/geode/lxfb.h
index 3b9416f4ee20..6a51448fd3f7 100644
--- a/drivers/video/geode/lxfb.h
+++ b/drivers/video/geode/lxfb.h
@@ -51,8 +51,6 @@ static inline unsigned int lx_get_pitch(unsigned int xres, int bpp)
 }
 
 void lx_set_mode(struct fb_info *);
-void lx_get_gamma(struct fb_info *, unsigned int *, int);
-void lx_set_gamma(struct fb_info *, unsigned int *, int);
 unsigned int lx_framebuffer_size(void);
 int lx_blank_display(struct fb_info *, int);
 void lx_set_palette_reg(struct fb_info *, unsigned int, unsigned int,

From 104b198dd0b3b62a4fc4e9146f01f2abc718e926 Mon Sep 17 00:00:00 2001
From: Jordan Crouse <jordan.crouse@amd.com>
Date: Wed, 23 Jul 2008 21:31:43 -0700
Subject: [PATCH 0512/1171] lxfb: fix console blanking

Simply enabling DAC blanking without turning off the CRT seems to be resulting
in characters remaining on the screen when the monitor blanks.  This patch
turns off the CRT for all modes, and also powers down the DACs when vsync
and/or hsync are disabled.

Signed-off-by: Jordan Crouse <jordan.crouse@amd.com>
Acked-by: Andres Salomon <dilinger@debian.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/geode/lxfb_ops.c | 28 ++++++++++++++++++----------
 1 file changed, 18 insertions(+), 10 deletions(-)

diff --git a/drivers/video/geode/lxfb_ops.c b/drivers/video/geode/lxfb_ops.c
index aaef9165ec9b..b1cd49c99356 100644
--- a/drivers/video/geode/lxfb_ops.c
+++ b/drivers/video/geode/lxfb_ops.c
@@ -517,25 +517,25 @@ void lx_set_palette_reg(struct fb_info *info, unsigned regno,
 int lx_blank_display(struct fb_info *info, int blank_mode)
 {
 	struct lxfb_par *par = info->par;
-	u32 dcfg, fp_pm;
-	int blank, hsync, vsync, crt;
+	u32 dcfg, misc, fp_pm;
+	int blank, hsync, vsync;
 
 	/* CRT power saving modes. */
 	switch (blank_mode) {
 	case FB_BLANK_UNBLANK:
-		blank = 0; hsync = 1; vsync = 1; crt = 1;
+		blank = 0; hsync = 1; vsync = 1;
 		break;
 	case FB_BLANK_NORMAL:
-		blank = 1; hsync = 1; vsync = 1; crt = 1;
+		blank = 1; hsync = 1; vsync = 1;
 		break;
 	case FB_BLANK_VSYNC_SUSPEND:
-		blank = 1; hsync = 1; vsync = 0; crt = 1;
+		blank = 1; hsync = 1; vsync = 0;
 		break;
 	case FB_BLANK_HSYNC_SUSPEND:
-		blank = 1; hsync = 0; vsync = 1; crt = 1;
+		blank = 1; hsync = 0; vsync = 1;
 		break;
 	case FB_BLANK_POWERDOWN:
-		blank = 1; hsync = 0; vsync = 0; crt = 0;
+		blank = 1; hsync = 0; vsync = 0;
 		break;
 	default:
 		return -EINVAL;
@@ -545,15 +545,23 @@ int lx_blank_display(struct fb_info *info, int blank_mode)
 	dcfg &= ~(VP_DCFG_DAC_BL_EN | VP_DCFG_HSYNC_EN | VP_DCFG_VSYNC_EN |
 			VP_DCFG_CRT_EN);
 	if (!blank)
-		dcfg |= VP_DCFG_DAC_BL_EN;
+		dcfg |= VP_DCFG_DAC_BL_EN | VP_DCFG_CRT_EN;
 	if (hsync)
 		dcfg |= VP_DCFG_HSYNC_EN;
 	if (vsync)
 		dcfg |= VP_DCFG_VSYNC_EN;
-	if (crt)
-		dcfg |= VP_DCFG_CRT_EN;
+
 	write_vp(par, VP_DCFG, dcfg);
 
+	misc = read_vp(par, VP_MISC);
+
+	if (vsync && hsync)
+		misc &= ~VP_MISC_DACPWRDN;
+	else
+		misc |= VP_MISC_DACPWRDN;
+
+	write_vp(par, VP_MISC, misc);
+
 	/* Power on/off flat panel */
 
 	if (par->output & OUTPUT_PANEL) {

From 5798712d608f5ebad994487748a2ccf3cc613d78 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Wed, 23 Jul 2008 21:31:43 -0700
Subject: [PATCH 0513/1171] drivers/video/amifb.c cleanups

This patch contains the following cleanups:
- make the needlessly global amifb_init() static
- rename cleanup_module() to amifb_exit(),
  make it static __exit,
  use module_exit(),
  there's no need to #ifdef MODULE it

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/amifb.c | 17 +++++++----------
 1 file changed, 7 insertions(+), 10 deletions(-)

diff --git a/drivers/video/amifb.c b/drivers/video/amifb.c
index 0c549aa1cf87..b8e9a8682f2d 100644
--- a/drivers/video/amifb.c
+++ b/drivers/video/amifb.c
@@ -1136,7 +1136,6 @@ static int amifb_ioctl(struct fb_info *info, unsigned int cmd, unsigned long arg
 	 * Interface to the low level console driver
 	 */
 
-int amifb_init(void);
 static void amifb_deinit(void);
 
 	/*
@@ -2248,7 +2247,7 @@ static inline void chipfree(void)
 	 * Initialisation
 	 */
 
-int __init amifb_init(void)
+static int __init amifb_init(void)
 {
 	int tag, i, err = 0;
 	u_long chipptr;
@@ -3793,16 +3792,14 @@ static void ami_rebuild_copper(void)
 	}
 }
 
-
-module_init(amifb_init);
-
-#ifdef MODULE
-MODULE_LICENSE("GPL");
-
-void cleanup_module(void)
+static void __exit amifb_exit(void)
 {
 	unregister_framebuffer(&fb_info);
 	amifb_deinit();
 	amifb_video_off();
 }
-#endif /* MODULE */
+
+module_init(amifb_init);
+module_exit(amifb_exit);
+
+MODULE_LICENSE("GPL");

From 7fc80b7bd682b47825e806018cca8ff7dc6bb55a Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:44 -0700
Subject: [PATCH 0514/1171] neofb: simplify clock calculation

There is nothing to gain by converting value in kHz to fixed point MHz.
Just calculate everything in kHz.

A reorder of the loop allows reducing number of iterations (check if
frequency is not too high already).

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/neofb.c | 30 +++++++++++++-----------------
 1 file changed, 13 insertions(+), 17 deletions(-)

diff --git a/drivers/video/neofb.c b/drivers/video/neofb.c
index 5246b0402d76..b033e5a4dc86 100644
--- a/drivers/video/neofb.c
+++ b/drivers/video/neofb.c
@@ -201,7 +201,6 @@ static int neoFindMode(int xres, int yres, int depth)
  *
  * Determine the closest clock frequency to the one requested.
  */
-#define REF_FREQ 0xe517		/* 14.31818 in 20.12 fixed point */
 #define MAX_N 127
 #define MAX_D 31
 #define MAX_F 1
@@ -211,27 +210,24 @@ static void neoCalcVCLK(const struct fb_info *info,
 {
 	int n, d, f;
 	int n_best = 0, d_best = 0, f_best = 0;
-	long f_best_diff = (0x7ffff << 12);	/* 20.12 */
-	long f_target = (freq << 12) / 1000;	/* 20.12 */
+	long f_best_diff = 0x7ffff;
 
 	for (f = 0; f <= MAX_F; f++)
-		for (n = 0; n <= MAX_N; n++)
-			for (d = 0; d <= MAX_D; d++) {
-				long f_out;	/* 20.12 */
-				long f_diff;	/* 20.12 */
+		for (d = 0; d <= MAX_D; d++)
+			for (n = 0; n <= MAX_N; n++) {
+				long f_out;
+				long f_diff;
 
-				f_out =
-				    ((((n + 1) << 12) / ((d +
-							  1) *
-							 (1 << f))) >> 12)
-				    * REF_FREQ;
-				f_diff = abs(f_out - f_target);
-				if (f_diff < f_best_diff) {
+				f_out = ((14318 * (n + 1)) / (d + 1)) >> f;
+				f_diff = abs(f_out - freq);
+				if (f_diff <= f_best_diff) {
 					f_best_diff = f_diff;
 					n_best = n;
 					d_best = d;
 					f_best = f;
 				}
+				if (f_out > freq)
+					break;
 			}
 
 	if (info->fix.accel == FB_ACCEL_NEOMAGIC_NM2200 ||
@@ -248,11 +244,11 @@ static void neoCalcVCLK(const struct fb_info *info,
 	par->VCLK3Denominator = d_best;
 
 #ifdef NEOFB_DEBUG
-	printk("neoVCLK: f:%d NumLow=%d NumHi=%d Den=%d Df=%d\n",
-	       f_target >> 12,
+	printk(KERN_DEBUG "neoVCLK: f:%ld NumLow=%d NumHi=%d Den=%d Df=%ld\n",
+	       freq,
 	       par->VCLK3NumeratorLow,
 	       par->VCLK3NumeratorHigh,
-	       par->VCLK3Denominator, f_best_diff >> 12);
+	       par->VCLK3Denominator, f_best_diff);
 #endif
 }
 

From 1ca6b62f8ca668ccfab0da9112c0125ef82343bd Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:45 -0700
Subject: [PATCH 0515/1171] neofb: drop redundant code

Drop structure which is only set but never read.  Drop variables which are
only set and never read.  Convert one long switch into two shorter ones.

Add cpu_relax() in busy waiting loop.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/neofb.c | 95 ++++++++++---------------------------------
 1 file changed, 22 insertions(+), 73 deletions(-)

diff --git a/drivers/video/neofb.c b/drivers/video/neofb.c
index b033e5a4dc86..669b8c196bba 100644
--- a/drivers/video/neofb.c
+++ b/drivers/video/neofb.c
@@ -479,7 +479,8 @@ static inline int neo2200_sync(struct fb_info *info)
 {
 	struct neofb_par *par = info->par;
 
-	while (readl(&par->neo2200->bltStat) & 1);
+	while (readl(&par->neo2200->bltStat) & 1)
+		cpu_relax();
 	return 0;
 }
 
@@ -587,34 +588,14 @@ static int
 neofb_check_var(struct fb_var_screeninfo *var, struct fb_info *info)
 {
 	struct neofb_par *par = info->par;
-	unsigned int pixclock = var->pixclock;
-	struct xtimings timings;
 	int memlen, vramlen;
 	int mode_ok = 0;
 
 	DBG("neofb_check_var");
 
-	if (!pixclock)
-		pixclock = 10000;	/* 10ns = 100MHz */
-	timings.pixclock = 1000000000 / pixclock;
-	if (timings.pixclock < 1)
-		timings.pixclock = 1;
-
-	if (timings.pixclock > par->maxClock)
+	if (PICOS2KHZ(var->pixclock) > par->maxClock)
 		return -EINVAL;
 
-	timings.dblscan = var->vmode & FB_VMODE_DOUBLE;
-	timings.interlaced = var->vmode & FB_VMODE_INTERLACED;
-	timings.HDisplay = var->xres;
-	timings.HSyncStart = timings.HDisplay + var->right_margin;
-	timings.HSyncEnd = timings.HSyncStart + var->hsync_len;
-	timings.HTotal = timings.HSyncEnd + var->left_margin;
-	timings.VDisplay = var->yres;
-	timings.VSyncStart = timings.VDisplay + var->lower_margin;
-	timings.VSyncEnd = timings.VSyncStart + var->vsync_len;
-	timings.VTotal = timings.VSyncEnd + var->upper_margin;
-	timings.sync = var->sync;
-
 	/* Is the mode larger than the LCD panel? */
 	if (par->internal_display &&
             ((var->xres > par->NeoPanelWidth) ||
@@ -1923,9 +1904,6 @@ static int __devinit neo_init_hw(struct fb_info *info)
 	int maxClock = 65000;
 	int CursorMem = 1024;
 	int CursorOff = 0x100;
-	int linearSize = 1024;
-	int maxWidth = 1024;
-	int maxHeight = 1024;
 
 	DBG("neo_init_hw");
 
@@ -1944,81 +1922,52 @@ static int __devinit neo_init_hw(struct fb_info *info)
 	case FB_ACCEL_NEOMAGIC_NM2070:
 		videoRam = 896;
 		maxClock = 65000;
-		CursorMem = 2048;
-		CursorOff = 0x100;
-		linearSize = 1024;
-		maxWidth = 1024;
-		maxHeight = 1024;
 		break;
 	case FB_ACCEL_NEOMAGIC_NM2090:
 	case FB_ACCEL_NEOMAGIC_NM2093:
-		videoRam = 1152;
-		maxClock = 80000;
-		CursorMem = 2048;
-		CursorOff = 0x100;
-		linearSize = 2048;
-		maxWidth = 1024;
-		maxHeight = 1024;
-		break;
 	case FB_ACCEL_NEOMAGIC_NM2097:
 		videoRam = 1152;
 		maxClock = 80000;
-		CursorMem = 1024;
-		CursorOff = 0x100;
-		linearSize = 2048;
-		maxWidth = 1024;
-		maxHeight = 1024;
 		break;
 	case FB_ACCEL_NEOMAGIC_NM2160:
 		videoRam = 2048;
 		maxClock = 90000;
-		CursorMem = 1024;
-		CursorOff = 0x100;
-		linearSize = 2048;
-		maxWidth = 1024;
-		maxHeight = 1024;
 		break;
 	case FB_ACCEL_NEOMAGIC_NM2200:
 		videoRam = 2560;
 		maxClock = 110000;
-		CursorMem = 1024;
-		CursorOff = 0x1000;
-		linearSize = 4096;
-		maxWidth = 1280;
-		maxHeight = 1024;	/* ???? */
-
-		par->neo2200 = (Neo2200 __iomem *) par->mmio_vbase;
 		break;
 	case FB_ACCEL_NEOMAGIC_NM2230:
 		videoRam = 3008;
 		maxClock = 110000;
-		CursorMem = 1024;
-		CursorOff = 0x1000;
-		linearSize = 4096;
-		maxWidth = 1280;
-		maxHeight = 1024;	/* ???? */
-
-		par->neo2200 = (Neo2200 __iomem *) par->mmio_vbase;
 		break;
 	case FB_ACCEL_NEOMAGIC_NM2360:
 		videoRam = 4096;
 		maxClock = 110000;
-		CursorMem = 1024;
-		CursorOff = 0x1000;
-		linearSize = 4096;
-		maxWidth = 1280;
-		maxHeight = 1024;	/* ???? */
-
-		par->neo2200 = (Neo2200 __iomem *) par->mmio_vbase;
 		break;
 	case FB_ACCEL_NEOMAGIC_NM2380:
 		videoRam = 6144;
 		maxClock = 110000;
+		break;
+	}
+	switch (info->fix.accel) {
+	case FB_ACCEL_NEOMAGIC_NM2070:
+	case FB_ACCEL_NEOMAGIC_NM2090:
+	case FB_ACCEL_NEOMAGIC_NM2093:
+		CursorMem = 2048;
+		CursorOff = 0x100;
+		break;
+	case FB_ACCEL_NEOMAGIC_NM2097:
+	case FB_ACCEL_NEOMAGIC_NM2160:
+		CursorMem = 1024;
+		CursorOff = 0x100;
+		break;
+	case FB_ACCEL_NEOMAGIC_NM2200:
+	case FB_ACCEL_NEOMAGIC_NM2230:
+	case FB_ACCEL_NEOMAGIC_NM2360:
+	case FB_ACCEL_NEOMAGIC_NM2380:
 		CursorMem = 1024;
 		CursorOff = 0x1000;
-		linearSize = 8192;
-		maxWidth = 1280;
-		maxHeight = 1024;	/* ???? */
 
 		par->neo2200 = (Neo2200 __iomem *) par->mmio_vbase;
 		break;
@@ -2032,7 +1981,7 @@ static int __devinit neo_init_hw(struct fb_info *info)
 */
 	par->maxClock = maxClock;
 	par->cursorOff = CursorOff;
-	return ((videoRam * 1024));
+	return videoRam * 1024;
 }
 
 

From c6b044d6bab5e2878d408666469362fc200a889a Mon Sep 17 00:00:00 2001
From: Krzysztof Helt <krzysztof.h1@wp.pl>
Date: Wed, 23 Jul 2008 21:31:45 -0700
Subject: [PATCH 0516/1171] neofb: drop the xtimings structure

Remove the xtimings structure which only stored some values to be used
later (mostly once).  Calculate and use these values in places they are
needed.

Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/neofb.c    | 88 ++++++++++++++++++----------------------
 include/video/neomagic.h | 17 --------
 2 files changed, 40 insertions(+), 65 deletions(-)

diff --git a/drivers/video/neofb.c b/drivers/video/neofb.c
index 669b8c196bba..25172b2a2a94 100644
--- a/drivers/video/neofb.c
+++ b/drivers/video/neofb.c
@@ -259,15 +259,20 @@ static void neoCalcVCLK(const struct fb_info *info,
  */
 
 static int vgaHWInit(const struct fb_var_screeninfo *var,
-		     const struct fb_info *info,
-		     struct neofb_par *par, struct xtimings *timings)
+		     struct neofb_par *par)
 {
+	int hsync_end = var->xres + var->right_margin + var->hsync_len;
+	int htotal = (hsync_end + var->left_margin) >> 3;
+	int vsync_start = var->yres + var->lower_margin;
+	int vsync_end = vsync_start + var->vsync_len;
+	int vtotal = vsync_end + var->upper_margin;
+
 	par->MiscOutReg = 0x23;
 
-	if (!(timings->sync & FB_SYNC_HOR_HIGH_ACT))
+	if (!(var->sync & FB_SYNC_HOR_HIGH_ACT))
 		par->MiscOutReg |= 0x40;
 
-	if (!(timings->sync & FB_SYNC_VERT_HIGH_ACT))
+	if (!(var->sync & FB_SYNC_VERT_HIGH_ACT))
 		par->MiscOutReg |= 0x80;
 
 	/*
@@ -282,25 +287,25 @@ static int vgaHWInit(const struct fb_var_screeninfo *var,
 	/*
 	 * CRTC Controller
 	 */
-	par->CRTC[0] = (timings->HTotal >> 3) - 5;
-	par->CRTC[1] = (timings->HDisplay >> 3) - 1;
-	par->CRTC[2] = (timings->HDisplay >> 3) - 1;
-	par->CRTC[3] = (((timings->HTotal >> 3) - 1) & 0x1F) | 0x80;
-	par->CRTC[4] = (timings->HSyncStart >> 3);
-	par->CRTC[5] = ((((timings->HTotal >> 3) - 1) & 0x20) << 2)
-	    | (((timings->HSyncEnd >> 3)) & 0x1F);
-	par->CRTC[6] = (timings->VTotal - 2) & 0xFF;
-	par->CRTC[7] = (((timings->VTotal - 2) & 0x100) >> 8)
-	    | (((timings->VDisplay - 1) & 0x100) >> 7)
-	    | ((timings->VSyncStart & 0x100) >> 6)
-	    | (((timings->VDisplay - 1) & 0x100) >> 5)
-	    | 0x10 | (((timings->VTotal - 2) & 0x200) >> 4)
-	    | (((timings->VDisplay - 1) & 0x200) >> 3)
-	    | ((timings->VSyncStart & 0x200) >> 2);
+	par->CRTC[0] = htotal - 5;
+	par->CRTC[1] = (var->xres >> 3) - 1;
+	par->CRTC[2] = (var->xres >> 3) - 1;
+	par->CRTC[3] = ((htotal - 1) & 0x1F) | 0x80;
+	par->CRTC[4] = ((var->xres + var->right_margin) >> 3);
+	par->CRTC[5] = (((htotal - 1) & 0x20) << 2)
+	    | (((hsync_end >> 3)) & 0x1F);
+	par->CRTC[6] = (vtotal - 2) & 0xFF;
+	par->CRTC[7] = (((vtotal - 2) & 0x100) >> 8)
+	    | (((var->yres - 1) & 0x100) >> 7)
+	    | ((vsync_start & 0x100) >> 6)
+	    | (((var->yres - 1) & 0x100) >> 5)
+	    | 0x10 | (((vtotal - 2) & 0x200) >> 4)
+	    | (((var->yres - 1) & 0x200) >> 3)
+	    | ((vsync_start & 0x200) >> 2);
 	par->CRTC[8] = 0x00;
-	par->CRTC[9] = (((timings->VDisplay - 1) & 0x200) >> 4) | 0x40;
+	par->CRTC[9] = (((var->yres - 1) & 0x200) >> 4) | 0x40;
 
-	if (timings->dblscan)
+	if (var->vmode & FB_VMODE_DOUBLE)
 		par->CRTC[9] |= 0x80;
 
 	par->CRTC[10] = 0x00;
@@ -309,13 +314,13 @@ static int vgaHWInit(const struct fb_var_screeninfo *var,
 	par->CRTC[13] = 0x00;
 	par->CRTC[14] = 0x00;
 	par->CRTC[15] = 0x00;
-	par->CRTC[16] = timings->VSyncStart & 0xFF;
-	par->CRTC[17] = (timings->VSyncEnd & 0x0F) | 0x20;
-	par->CRTC[18] = (timings->VDisplay - 1) & 0xFF;
+	par->CRTC[16] = vsync_start & 0xFF;
+	par->CRTC[17] = (vsync_end & 0x0F) | 0x20;
+	par->CRTC[18] = (var->yres - 1) & 0xFF;
 	par->CRTC[19] = var->xres_virtual >> 4;
 	par->CRTC[20] = 0x00;
-	par->CRTC[21] = (timings->VDisplay - 1) & 0xFF;
-	par->CRTC[22] = (timings->VTotal - 1) & 0xFF;
+	par->CRTC[21] = (var->yres - 1) & 0xFF;
+	par->CRTC[22] = (vtotal - 1) & 0xFF;
 	par->CRTC[23] = 0xC3;
 	par->CRTC[24] = 0xFF;
 
@@ -736,11 +741,11 @@ neofb_check_var(struct fb_var_screeninfo *var, struct fb_info *info)
 static int neofb_set_par(struct fb_info *info)
 {
 	struct neofb_par *par = info->par;
-	struct xtimings timings;
 	unsigned char temp;
 	int i, clock_hi = 0;
 	int lcd_stretch;
 	int hoffset, voffset;
+	int vsync_start, vtotal;
 
 	DBG("neofb_set_par");
 
@@ -748,28 +753,15 @@ static int neofb_set_par(struct fb_info *info)
 
 	vgaHWProtect(1);	/* Blank the screen */
 
-	timings.dblscan = info->var.vmode & FB_VMODE_DOUBLE;
-	timings.interlaced = info->var.vmode & FB_VMODE_INTERLACED;
-	timings.HDisplay = info->var.xres;
-	timings.HSyncStart = timings.HDisplay + info->var.right_margin;
-	timings.HSyncEnd = timings.HSyncStart + info->var.hsync_len;
-	timings.HTotal = timings.HSyncEnd + info->var.left_margin;
-	timings.VDisplay = info->var.yres;
-	timings.VSyncStart = timings.VDisplay + info->var.lower_margin;
-	timings.VSyncEnd = timings.VSyncStart + info->var.vsync_len;
-	timings.VTotal = timings.VSyncEnd + info->var.upper_margin;
-	timings.sync = info->var.sync;
-	timings.pixclock = PICOS2KHZ(info->var.pixclock);
-
-	if (timings.pixclock < 1)
-		timings.pixclock = 1;
+	vsync_start = info->var.yres + info->var.lower_margin;
+	vtotal = vsync_start + info->var.vsync_len + info->var.upper_margin;
 
 	/*
 	 * This will allocate the datastructure and initialize all of the
 	 * generic VGA registers.
 	 */
 
-	if (vgaHWInit(&info->var, info, par, &timings))
+	if (vgaHWInit(&info->var, par))
 		return -EINVAL;
 
 	/*
@@ -808,10 +800,10 @@ static int neofb_set_par(struct fb_info *info)
 	par->ExtCRTDispAddr = 0x10;
 
 	/* Vertical Extension */
-	par->VerticalExt = (((timings.VTotal - 2) & 0x400) >> 10)
-	    | (((timings.VDisplay - 1) & 0x400) >> 9)
-	    | (((timings.VSyncStart) & 0x400) >> 8)
-	    | (((timings.VSyncStart) & 0x400) >> 7);
+	par->VerticalExt = (((vtotal - 2) & 0x400) >> 10)
+	    | (((info->var.yres - 1) & 0x400) >> 9)
+	    | (((vsync_start) & 0x400) >> 8)
+	    | (((vsync_start) & 0x400) >> 7);
 
 	/* Fast write bursts on unless disabled. */
 	if (par->pci_burst)
@@ -972,7 +964,7 @@ static int neofb_set_par(struct fb_info *info)
 	 * Calculate the VCLK that most closely matches the requested dot
 	 * clock.
 	 */
-	neoCalcVCLK(info, par, timings.pixclock);
+	neoCalcVCLK(info, par, PICOS2KHZ(info->var.pixclock));
 
 	/* Since we program the clocks ourselves, always use VCLK3. */
 	par->MiscOutReg |= 0x0C;
diff --git a/include/video/neomagic.h b/include/video/neomagic.h
index a9e118a1cd16..38910da0ae59 100644
--- a/include/video/neomagic.h
+++ b/include/video/neomagic.h
@@ -90,23 +90,6 @@
 #define PCI_CHIP_NM2360 0x0006
 #define PCI_CHIP_NM2380 0x0016
 
-
-struct xtimings {
-	unsigned int pixclock;
-	unsigned int HDisplay;
-	unsigned int HSyncStart;
-	unsigned int HSyncEnd;
-	unsigned int HTotal;
-	unsigned int VDisplay;
-	unsigned int VSyncStart;
-	unsigned int VSyncEnd;
-	unsigned int VTotal;
-	unsigned int sync;
-	int dblscan;
-	int interlaced;
-};
-
-
 /* --------------------------------------------------------------------- */
 
 typedef volatile struct {

From 4a25e41831ee851c1365d8b41decc22493b18e6d Mon Sep 17 00:00:00 2001
From: Nobuhiro Iwamatsu <iwamatsu.nobuhiro@renesas.com>
Date: Wed, 23 Jul 2008 21:31:46 -0700
Subject: [PATCH 0517/1171] video: sh7760fb: SH7760/SH7763 LCDC framebuffer
 driver

Framebuffer driver for the SH7760/SH7763 integrated LCD controller.

Signed-off-by: Manuel Lauss <mano@roarinelk.homelinux.net>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu.nobuhiro@renesas.com>
Reviewed-by: Paul Mundt <lethal@linux-sh.org>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Cc: Siegfried Schaefer <s.schaefer@schaefer-edv.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/fb/sh7760fb.txt | 131 +++++++
 drivers/video/Kconfig         |  13 +
 drivers/video/Makefile        |   1 +
 drivers/video/sh7760fb.c      | 658 ++++++++++++++++++++++++++++++++++
 include/asm-sh/sh7760fb.h     | 197 ++++++++++
 5 files changed, 1000 insertions(+)
 create mode 100644 Documentation/fb/sh7760fb.txt
 create mode 100644 drivers/video/sh7760fb.c
 create mode 100644 include/asm-sh/sh7760fb.h

diff --git a/Documentation/fb/sh7760fb.txt b/Documentation/fb/sh7760fb.txt
new file mode 100644
index 000000000000..c87bfe5c630a
--- /dev/null
+++ b/Documentation/fb/sh7760fb.txt
@@ -0,0 +1,131 @@
+SH7760/SH7763 integrated LCDC Framebuffer driver
+================================================
+
+0. Overwiew
+-----------
+The SH7760/SH7763 have an integrated LCD Display controller (LCDC) which
+supports (in theory) resolutions ranging from 1x1 to 1024x1024,
+with color depths ranging from 1 to 16 bits, on STN, DSTN and TFT Panels.
+
+Caveats:
+* Framebuffer memory must be a large chunk allocated at the top
+  of Area3 (HW requirement). Because of this requirement you should NOT
+  make the driver a module since at runtime it may become impossible to
+  get a large enough contiguous chunk of memory.
+
+* The driver does not support changing resolution while loaded
+  (displays aren't hotpluggable anyway)
+
+* Heavy flickering may be observed
+  a) if you're using 15/16bit color modes at >= 640x480 px resolutions,
+  b) during PCMCIA (or any other slow bus) activity.
+
+* Rotation works only 90degress clockwise, and only if horizontal
+  resolution is <= 320 pixels.
+
+files:   drivers/video/sh7760fb.c
+        include/asm-sh/sh7760fb.h
+        Documentation/fb/sh7760fb.txt
+
+1. Platform setup
+-----------------
+SH7760:
+ Video data is fetched via the DMABRG DMA engine, so you have to
+ configure the SH DMAC for DMABRG mode (write 0x94808080 to the
+ DMARSRA register somewhere at boot).
+
+ PFC registers PCCR and PCDR must be set to peripheral mode.
+ (write zeros to both).
+
+The driver does NOT do the above for you since board setup is, well, job
+of the board setup code.
+
+2. Panel definitions
+--------------------
+The LCDC must explicitly be told about the type of LCD panel
+attached.  Data must be wrapped in a "struct sh7760fb_platdata" and
+passed to the driver as platform_data.
+
+Suggest you take a closer look at the SH7760 Manual, Section 30.
+(http://documentation.renesas.com/eng/products/mpumcu/e602291_sh7760.pdf)
+
+The following code illustrates what needs to be done to
+get the framebuffer working on a 640x480 TFT:
+
+====================== cut here ======================================
+
+#include <linux/fb.h>
+#include <asm/sh7760fb.h>
+
+/*
+ * NEC NL6440bc26-01 640x480 TFT
+ * dotclock 25175 kHz
+ * Xres                640     Yres            480
+ * Htotal      800     Vtotal          525
+ * HsynStart   656     VsynStart       490
+ * HsynLenn    30      VsynLenn        2
+ *
+ * The linux framebuffer layer does not use the syncstart/synclen
+ * values but right/left/upper/lower margin values. The comments
+ * for the x_margin explain how to calculate those from given
+ * panel sync timings.
+ */
+static struct fb_videomode nl6448bc26 = {
+       .name           = "NL6448BC26",
+       .refresh        = 60,
+       .xres           = 640,
+       .yres           = 480,
+       .pixclock       = 39683,        /* in picoseconds! */
+       .hsync_len      = 30,
+       .vsync_len      = 2,
+       .left_margin    = 114,  /* HTOT - (HSYNSLEN + HSYNSTART) */
+       .right_margin   = 16,   /* HSYNSTART - XRES */
+       .upper_margin   = 33,   /* VTOT - (VSYNLEN + VSYNSTART) */
+       .lower_margin   = 10,   /* VSYNSTART - YRES */
+       .sync           = FB_SYNC_HOR_HIGH_ACT | FB_SYNC_VERT_HIGH_ACT,
+       .vmode          = FB_VMODE_NONINTERLACED,
+       .flag           = 0,
+};
+
+static struct sh7760fb_platdata sh7760fb_nl6448 = {
+       .def_mode       = &nl6448bc26,
+       .ldmtr          = LDMTR_TFT_COLOR_16,   /* 16bit TFT panel */
+       .lddfr          = LDDFR_8BPP,           /* we want 8bit output */
+       .ldpmmr         = 0x0070,
+       .ldpspr         = 0x0500,
+       .ldaclnr        = 0,
+       .ldickr         = LDICKR_CLKSRC(LCDC_CLKSRC_EXTERNAL) |
+                         LDICKR_CLKDIV(1),
+       .rotate         = 0,
+       .novsync        = 1,
+       .blank          = NULL,
+};
+
+/* SH7760:
+ * 0xFE300800: 256 * 4byte xRGB palette ram
+ * 0xFE300C00: 42 bytes ctrl registers
+ */
+static struct resource sh7760_lcdc_res[] = {
+       [0] = {
+               .start  = 0xFE300800,
+               .end    = 0xFE300CFF,
+               .flags  = IORESOURCE_MEM,
+       },
+       [1] = {
+               .start  = 65,
+               .end    = 65,
+               .flags  = IORESOURCE_IRQ,
+       },
+};
+
+static struct platform_device sh7760_lcdc_dev = {
+       .dev    = {
+               .platform_data = &sh7760fb_nl6448,
+       },
+       .name           = "sh7760-lcdc",
+       .id             = -1,
+       .resource       = sh7760_lcdc_res,
+       .num_resources  = ARRAY_SIZE(sh7760_lcdc_res),
+};
+
+====================== cut here ======================================
diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig
index c1b7db843411..70d135e0cc47 100644
--- a/drivers/video/Kconfig
+++ b/drivers/video/Kconfig
@@ -1991,6 +1991,19 @@ config FB_COBALT
 	tristate "Cobalt server LCD frame buffer support"
 	depends on FB && MIPS_COBALT
 
+config FB_SH7760
+       bool "SH7760/SH7763 LCDC support"
+       depends on FB && (CPU_SUBTYPE_SH7760 || CPU_SUBTYPE_SH7763)
+       select FB_CFB_FILLRECT
+       select FB_CFB_COPYAREA
+       select FB_CFB_IMAGEBLIT
+       help
+         Support for the SH7760/SH7763 integrated (D)STN/TFT LCD Controller.
+         Supports display resolutions up to 1024x1024 pixel, grayscale and
+         color operation, with depths ranging from 1 bpp to 8 bpp monochrome
+         and 8, 15 or 16 bpp color; 90 degrees clockwise display rotation for
+         panels <= 320 pixel horizontal resolution.
+
 config FB_VIRTUAL
 	tristate "Virtual Frame Buffer support (ONLY FOR TESTING!)"
 	depends on FB
diff --git a/drivers/video/Makefile b/drivers/video/Makefile
index 358032597757..0ebc1bfd2514 100644
--- a/drivers/video/Makefile
+++ b/drivers/video/Makefile
@@ -106,6 +106,7 @@ obj-$(CONFIG_FB_PMAGB_B)	  += pmagb-b-fb.o
 obj-$(CONFIG_FB_MAXINE)		  += maxinefb.o
 obj-$(CONFIG_FB_METRONOME)        += metronomefb.o
 obj-$(CONFIG_FB_S1D13XXX)	  += s1d13xxxfb.o
+obj-$(CONFIG_FB_SH7760)		  += sh7760fb.o
 obj-$(CONFIG_FB_IMX)              += imxfb.o
 obj-$(CONFIG_FB_S3C2410)	  += s3c2410fb.o
 obj-$(CONFIG_FB_FSL_DIU)	  += fsl-diu-fb.o
diff --git a/drivers/video/sh7760fb.c b/drivers/video/sh7760fb.c
new file mode 100644
index 000000000000..4d0e28c5790b
--- /dev/null
+++ b/drivers/video/sh7760fb.c
@@ -0,0 +1,658 @@
+/*
+ * SH7760/SH7763 LCDC Framebuffer driver.
+ *
+ * (c) 2006-2008 MSC Vertriebsges.m.b.H.,
+ *             Manuel Lauss <mano@roarinelk.homelinux.net>
+ * (c) 2008 Nobuhiro Iwamatsu <iwamatsu.nobuhiro@renesas.com>
+ *
+ *  This file is subject to the terms and conditions of the GNU General
+ *  Public License.  See the file COPYING in the main directory of this
+ *  archive for more details.
+ *
+ * PLEASE HAVE A LOOK AT Documentation/fb/sh7760fb.txt!
+ *
+ * Thanks to Siegfried Schaefer <s.schaefer at schaefer-edv.de>
+ *     for his original source and testing!
+ */
+
+#include <linux/completion.h>
+#include <linux/delay.h>
+#include <linux/dma-mapping.h>
+#include <linux/fb.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+
+#include <asm/sh7760fb.h>
+
+struct sh7760fb_par {
+	void __iomem *base;
+	int irq;
+
+	struct sh7760fb_platdata *pd;	/* display information */
+
+	dma_addr_t fbdma;	/* physical address */
+
+	int rot;		/* rotation enabled? */
+
+	u32 pseudo_palette[16];
+
+	struct platform_device *dev;
+	struct resource *ioarea;
+	struct completion vsync;	/* vsync irq event */
+};
+
+static irqreturn_t sh7760fb_irq(int irq, void *data)
+{
+	struct completion *c = data;
+
+	complete(c);
+
+	return IRQ_HANDLED;
+}
+
+static void sh7760fb_wait_vsync(struct fb_info *info)
+{
+	struct sh7760fb_par *par = info->par;
+
+	if (par->pd->novsync)
+		return;
+
+	iowrite16(ioread16(par->base + LDINTR) & ~VINT_CHECK,
+		  par->base + LDINTR);
+
+	if (par->irq < 0) {
+		/* poll for vert. retrace: status bit is sticky */
+		while (!(ioread16(par->base + LDINTR) & VINT_CHECK))
+			cpu_relax();
+	} else {
+		/* a "wait_for_irq_event(par->irq)" would be extremely nice */
+		init_completion(&par->vsync);
+		enable_irq(par->irq);
+		wait_for_completion(&par->vsync);
+		disable_irq_nosync(par->irq);
+	}
+}
+
+/* wait_for_lps - wait until power supply has reached a certain state. */
+static int wait_for_lps(struct sh7760fb_par *par, int val)
+{
+	int i = 100;
+	while (--i && ((ioread16(par->base + LDPMMR) & 3) != val))
+		msleep(1);
+
+	if (i <= 0)
+		return -ETIMEDOUT;
+
+	return 0;
+}
+
+/* en/disable the LCDC */
+static int sh7760fb_blank(int blank, struct fb_info *info)
+{
+	struct sh7760fb_par *par = info->par;
+	struct sh7760fb_platdata *pd = par->pd;
+	unsigned short cntr = ioread16(par->base + LDCNTR);
+	unsigned short intr = ioread16(par->base + LDINTR);
+	int lps;
+
+	if (blank == FB_BLANK_UNBLANK) {
+		intr |= VINT_START;
+		cntr = LDCNTR_DON2 | LDCNTR_DON;
+		lps = 3;
+	} else {
+		intr &= ~VINT_START;
+		cntr = LDCNTR_DON2;
+		lps = 0;
+	}
+
+	if (pd->blank)
+		pd->blank(blank);
+
+	iowrite16(intr, par->base + LDINTR);
+	iowrite16(cntr, par->base + LDCNTR);
+
+	return wait_for_lps(par, lps);
+}
+
+/* set color registers */
+static int sh7760fb_setcmap(struct fb_cmap *cmap, struct fb_info *info)
+{
+	struct sh7760fb_par *par = info->par;
+	u32 s = cmap->start;
+	u32 l = cmap->len;
+	u16 *r = cmap->red;
+	u16 *g = cmap->green;
+	u16 *b = cmap->blue;
+	u32 col, tmo;
+	int ret;
+
+	ret = 0;
+
+	sh7760fb_wait_vsync(info);
+
+	/* request palette access */
+	iowrite16(LDPALCR_PALEN, par->base + LDPALCR);
+
+	/* poll for access grant */
+	tmo = 100;
+	while (!(ioread16(par->base + LDPALCR) & LDPALCR_PALS) && (--tmo))
+		cpu_relax();
+
+	if (!tmo) {
+		ret = 1;
+		dev_dbg(info->dev, "no palette access!\n");
+		goto out;
+	}
+
+	while (l && (s < 256)) {
+		col = ((*r) & 0xff) << 16;
+		col |= ((*g) & 0xff) << 8;
+		col |= ((*b) & 0xff);
+		col &= SH7760FB_PALETTE_MASK;
+
+		if (s < 16)
+			((u32 *) (info->pseudo_palette))[s] = s;
+
+		s++;
+		l--;
+		r++;
+		g++;
+		b++;
+	}
+out:
+	iowrite16(0, par->base + LDPALCR);
+	return ret;
+}
+
+static void encode_fix(struct fb_fix_screeninfo *fix, struct fb_info *info,
+		       unsigned long stride)
+{
+	memset(fix, 0, sizeof(struct fb_fix_screeninfo));
+	strcpy(fix->id, "sh7760-lcdc");
+
+	fix->smem_start = (unsigned long)info->screen_base;
+	fix->smem_len = info->screen_size;
+
+	fix->line_length = stride;
+}
+
+static int sh7760fb_get_color_info(struct device *dev,
+				   u16 lddfr, int *bpp, int *gray)
+{
+	int lbpp, lgray;
+
+	lgray = lbpp = 0;
+
+	switch (lddfr & LDDFR_COLOR_MASK) {
+	case LDDFR_1BPP_MONO:
+		lgray = 1;
+		lbpp = 1;
+		break;
+	case LDDFR_2BPP_MONO:
+		lgray = 1;
+		lbpp = 2;
+		break;
+	case LDDFR_4BPP_MONO:
+		lgray = 1;
+	case LDDFR_4BPP:
+		lbpp = 4;
+		break;
+	case LDDFR_6BPP_MONO:
+		lgray = 1;
+	case LDDFR_8BPP:
+		lbpp = 8;
+		break;
+	case LDDFR_16BPP_RGB555:
+	case LDDFR_16BPP_RGB565:
+		lbpp = 16;
+		lgray = 0;
+		break;
+	default:
+		dev_dbg(dev, "unsupported LDDFR bit depth.\n");
+		return -EINVAL;
+	}
+
+	if (bpp)
+		*bpp = lbpp;
+	if (gray)
+		*gray = lgray;
+
+	return 0;
+}
+
+static int sh7760fb_check_var(struct fb_var_screeninfo *var,
+			      struct fb_info *info)
+{
+	struct fb_fix_screeninfo *fix = &info->fix;
+	struct sh7760fb_par *par = info->par;
+	int ret, bpp;
+
+	/* get color info from register value */
+	ret = sh7760fb_get_color_info(info->dev, par->pd->lddfr, &bpp, NULL);
+	if (ret)
+		return ret;
+
+	var->bits_per_pixel = bpp;
+
+	if ((var->grayscale) && (var->bits_per_pixel == 1))
+		fix->visual = FB_VISUAL_MONO10;
+	else if (var->bits_per_pixel >= 15)
+		fix->visual = FB_VISUAL_TRUECOLOR;
+	else
+		fix->visual = FB_VISUAL_PSEUDOCOLOR;
+
+	/* TODO: add some more validation here */
+	return 0;
+}
+
+/*
+ * sh7760fb_set_par - set videomode.
+ *
+ * NOTE: The rotation, grayscale and DSTN codepaths are
+ *     totally untested!
+ */
+static int sh7760fb_set_par(struct fb_info *info)
+{
+	struct sh7760fb_par *par = info->par;
+	struct fb_videomode *vm = par->pd->def_mode;
+	unsigned long sbase, dstn_off, ldsarl, stride;
+	unsigned short hsynp, hsynw, htcn, hdcn;
+	unsigned short vsynp, vsynw, vtln, vdln;
+	unsigned short lddfr, ldmtr;
+	int ret, bpp, gray;
+
+	par->rot = par->pd->rotate;
+
+	/* rotate only works with xres <= 320 */
+	if (par->rot && (vm->xres > 320)) {
+		dev_dbg(info->dev, "rotation disabled due to display size\n");
+		par->rot = 0;
+	}
+
+	/* calculate LCDC reg vals from display parameters */
+	hsynp = vm->right_margin + vm->xres;
+	hsynw = vm->hsync_len;
+	htcn = vm->left_margin + hsynp + hsynw;
+	hdcn = vm->xres;
+	vsynp = vm->lower_margin + vm->yres;
+	vsynw = vm->vsync_len;
+	vtln = vm->upper_margin + vsynp + vsynw;
+	vdln = vm->yres;
+
+	/* get color info from register value */
+	ret = sh7760fb_get_color_info(info->dev, par->pd->lddfr, &bpp, &gray);
+	if (ret)
+		return ret;
+
+	dev_dbg(info->dev, "%dx%d %dbpp %s (orientation %s)\n", hdcn,
+		vdln, bpp, gray ? "grayscale" : "color",
+		par->rot ? "rotated" : "normal");
+
+#ifdef CONFIG_CPU_LITTLE_ENDIAN
+	lddfr = par->pd->lddfr | (1 << 8);
+#else
+	lddfr = par->pd->lddfr & ~(1 << 8);
+#endif
+
+	ldmtr = par->pd->ldmtr;
+
+	if (!(vm->sync & FB_SYNC_HOR_HIGH_ACT))
+		ldmtr |= LDMTR_CL1POL;
+	if (!(vm->sync & FB_SYNC_VERT_HIGH_ACT))
+		ldmtr |= LDMTR_FLMPOL;
+
+	/* shut down LCDC before changing display parameters */
+	sh7760fb_blank(FB_BLANK_POWERDOWN, info);
+
+	iowrite16(par->pd->ldickr, par->base + LDICKR);	/* pixclock */
+	iowrite16(ldmtr, par->base + LDMTR);	/* polarities */
+	iowrite16(lddfr, par->base + LDDFR);	/* color/depth */
+	iowrite16((par->rot ? 1 << 13 : 0), par->base + LDSMR);	/* rotate */
+	iowrite16(par->pd->ldpmmr, par->base + LDPMMR);	/* Power Management */
+	iowrite16(par->pd->ldpspr, par->base + LDPSPR);	/* Power Supply Ctrl */
+
+	/* display resolution */
+	iowrite16(((htcn >> 3) - 1) | (((hdcn >> 3) - 1) << 8),
+		  par->base + LDHCNR);
+	iowrite16(vdln - 1, par->base + LDVDLNR);
+	iowrite16(vtln - 1, par->base + LDVTLNR);
+	/* h/v sync signals */
+	iowrite16((vsynp - 1) | ((vsynw - 1) << 12), par->base + LDVSYNR);
+	iowrite16(((hsynp >> 3) - 1) | (((hsynw >> 3) - 1) << 12),
+		  par->base + LDHSYNR);
+	/* AC modulation sig */
+	iowrite16(par->pd->ldaclnr, par->base + LDACLNR);
+
+	stride = (par->rot) ? vtln : hdcn;
+	if (!gray)
+		stride *= (bpp + 7) >> 3;
+	else {
+		if (bpp == 1)
+			stride >>= 3;
+		else if (bpp == 2)
+			stride >>= 2;
+		else if (bpp == 4)
+			stride >>= 1;
+		/* 6 bpp == 8 bpp */
+	}
+
+	/* if rotated, stride must be power of 2 */
+	if (par->rot) {
+		unsigned long bit = 1 << 31;
+		while (bit) {
+			if (stride & bit)
+				break;
+			bit >>= 1;
+		}
+		if (stride & ~bit)
+			stride = bit << 1;	/* not P-o-2, round up */
+	}
+	iowrite16(stride, par->base + LDLAOR);
+
+	/* set display mem start address */
+	sbase = (unsigned long)par->fbdma;
+	if (par->rot)
+		sbase += (hdcn - 1) * stride;
+
+	iowrite32(sbase, par->base + LDSARU);
+
+	/*
+	 * for DSTN need to set address for lower half.
+	 * I (mlau) don't know which address to set it to,
+	 * so I guessed at (stride * yres/2).
+	 */
+	if (((ldmtr & 0x003f) >= LDMTR_DSTN_MONO_8) &&
+	    ((ldmtr & 0x003f) <= LDMTR_DSTN_COLOR_16)) {
+
+		dev_dbg(info->dev, " ***** DSTN untested! *****\n");
+
+		dstn_off = stride;
+		if (par->rot)
+			dstn_off *= hdcn >> 1;
+		else
+			dstn_off *= vdln >> 1;
+
+		ldsarl = sbase + dstn_off;
+	} else
+		ldsarl = 0;
+
+	iowrite32(ldsarl, par->base + LDSARL);	/* mem for lower half of DSTN */
+
+	encode_fix(&info->fix, info, stride);
+	sh7760fb_check_var(&info->var, info);
+
+	sh7760fb_blank(FB_BLANK_UNBLANK, info);	/* panel on! */
+
+	dev_dbg(info->dev, "hdcn  : %6d htcn  : %6d\n", hdcn, htcn);
+	dev_dbg(info->dev, "hsynw : %6d hsynp : %6d\n", hsynw, hsynp);
+	dev_dbg(info->dev, "vdln  : %6d vtln  : %6d\n", vdln, vtln);
+	dev_dbg(info->dev, "vsynw : %6d vsynp : %6d\n", vsynw, vsynp);
+	dev_dbg(info->dev, "clksrc: %6d clkdiv: %6d\n",
+		(par->pd->ldickr >> 12) & 3, par->pd->ldickr & 0x1f);
+	dev_dbg(info->dev, "ldpmmr: 0x%04x ldpspr: 0x%04x\n", par->pd->ldpmmr,
+		par->pd->ldpspr);
+	dev_dbg(info->dev, "ldmtr : 0x%04x lddfr : 0x%04x\n", ldmtr, lddfr);
+	dev_dbg(info->dev, "ldlaor: %ld\n", stride);
+	dev_dbg(info->dev, "ldsaru: 0x%08lx ldsarl: 0x%08lx\n", sbase, ldsarl);
+
+	return 0;
+}
+
+static struct fb_ops sh7760fb_ops = {
+	.owner = THIS_MODULE,
+	.fb_blank = sh7760fb_blank,
+	.fb_check_var = sh7760fb_check_var,
+	.fb_setcmap = sh7760fb_setcmap,
+	.fb_set_par = sh7760fb_set_par,
+	.fb_fillrect = cfb_fillrect,
+	.fb_copyarea = cfb_copyarea,
+	.fb_imageblit = cfb_imageblit,
+};
+
+static void sh7760fb_free_mem(struct fb_info *info)
+{
+	struct sh7760fb_par *par = info->par;
+
+	if (!info->screen_base)
+		return;
+
+	dma_free_coherent(info->dev, info->screen_size,
+			  info->screen_base, par->fbdma);
+
+	par->fbdma = 0;
+	info->screen_base = NULL;
+	info->screen_size = 0;
+}
+
+/* allocate the framebuffer memory. This memory must be in Area3,
+ * (dictated by the DMA engine) and contiguous, at a 512 byte boundary.
+ */
+static int sh7760fb_alloc_mem(struct fb_info *info)
+{
+	struct sh7760fb_par *par = info->par;
+	void *fbmem;
+	unsigned long vram;
+	int ret, bpp;
+
+	if (info->screen_base)
+		return 0;
+
+	/* get color info from register value */
+	ret = sh7760fb_get_color_info(info->dev, par->pd->lddfr, &bpp, NULL);
+	if (ret) {
+		printk(KERN_ERR "colinfo\n");
+		return ret;
+	}
+
+	/* min VRAM: xres_min = 16, yres_min = 1, bpp = 1: 2byte -> 1 page
+	   max VRAM: xres_max = 1024, yres_max = 1024, bpp = 16: 2MB */
+
+	vram = info->var.xres * info->var.yres;
+	if (info->var.grayscale) {
+		if (bpp == 1)
+			vram >>= 3;
+		else if (bpp == 2)
+			vram >>= 2;
+		else if (bpp == 4)
+			vram >>= 1;
+	} else if (bpp > 8)
+		vram *= 2;
+	if ((vram < 1) || (vram > 1024 * 2048)) {
+		dev_dbg(info->dev, "too much VRAM required. Check settings\n");
+		return -ENODEV;
+	}
+
+	if (vram < PAGE_SIZE)
+		vram = PAGE_SIZE;
+
+	fbmem = dma_alloc_coherent(info->dev, vram, &par->fbdma, GFP_KERNEL);
+
+	if (!fbmem)
+		return -ENOMEM;
+
+	if ((par->fbdma & SH7760FB_DMA_MASK) != SH7760FB_DMA_MASK) {
+		sh7760fb_free_mem(info);
+		dev_err(info->dev, "kernel gave me memory at 0x%08lx, which is"
+			"unusable for the LCDC\n", (unsigned long)par->fbdma);
+		return -ENOMEM;
+	}
+
+	info->screen_base = fbmem;
+	info->screen_size = vram;
+
+	return 0;
+}
+
+static int __devinit sh7760fb_probe(struct platform_device *pdev)
+{
+	struct fb_info *info;
+	struct resource *res;
+	struct sh7760fb_par *par;
+	int ret;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (unlikely(res == NULL)) {
+		dev_err(&pdev->dev, "invalid resource\n");
+		return -EINVAL;
+	}
+
+	info = framebuffer_alloc(sizeof(struct sh7760fb_par), &pdev->dev);
+	if (!info)
+		return -ENOMEM;
+
+	par = info->par;
+	par->dev = pdev;
+
+	par->pd = pdev->dev.platform_data;
+	if (!par->pd) {
+		dev_dbg(info->dev, "no display setup data!\n");
+		ret = -ENODEV;
+		goto out_fb;
+	}
+
+	par->ioarea = request_mem_region(res->start,
+					 (res->end - res->start), pdev->name);
+	if (!par->ioarea) {
+		dev_err(&pdev->dev, "mmio area busy\n");
+		ret = -EBUSY;
+		goto out_fb;
+	}
+
+	par->base = ioremap_nocache(res->start, res->end - res->start + 1);
+	if (!par->base) {
+		dev_err(&pdev->dev, "cannot remap\n");
+		ret = -ENODEV;
+		goto out_res;
+	}
+
+	iowrite16(0, par->base + LDINTR);	/* disable vsync irq */
+	par->irq = platform_get_irq(pdev, 0);
+	if (par->irq >= 0) {
+		ret = request_irq(par->irq, sh7760fb_irq, 0,
+				  "sh7760-lcdc", &par->vsync);
+		if (ret) {
+			dev_err(&pdev->dev, "cannot grab IRQ\n");
+			par->irq = -ENXIO;
+		} else
+			disable_irq_nosync(par->irq);
+	}
+
+	fb_videomode_to_var(&info->var, par->pd->def_mode);
+
+	ret = sh7760fb_alloc_mem(info);
+	if (ret) {
+		dev_dbg(info->dev, "framebuffer memory allocation failed!\n");
+		goto out_unmap;
+	}
+
+	info->pseudo_palette = par->pseudo_palette;
+
+	/* fixup color register bitpositions. These are fixed by hardware */
+	info->var.red.offset = 11;
+	info->var.red.length = 5;
+	info->var.red.msb_right = 0;
+
+	info->var.green.offset = 5;
+	info->var.green.length = 6;
+	info->var.green.msb_right = 0;
+
+	info->var.blue.offset = 0;
+	info->var.blue.length = 5;
+	info->var.blue.msb_right = 0;
+
+	info->var.transp.offset = 0;
+	info->var.transp.length = 0;
+	info->var.transp.msb_right = 0;
+
+	/* set the DON2 bit now, before cmap allocation, as it will randomize
+	 * palette memory.
+	 */
+	iowrite16(LDCNTR_DON2, par->base + LDCNTR);
+	info->fbops = &sh7760fb_ops;
+
+	ret = fb_alloc_cmap(&info->cmap, 256, 0);
+	if (ret) {
+		dev_dbg(info->dev, "Unable to allocate cmap memory\n");
+		goto out_mem;
+	}
+
+	ret = register_framebuffer(info);
+	if (ret < 0) {
+		dev_dbg(info->dev, "cannot register fb!\n");
+		goto out_cmap;
+	}
+	platform_set_drvdata(pdev, info);
+
+	printk(KERN_INFO "%s: memory at phys 0x%08lx-0x%08lx, size %ld KiB\n",
+	       pdev->name,
+	       (unsigned long)par->fbdma,
+	       (unsigned long)(par->fbdma + info->screen_size - 1),
+	       info->screen_size >> 10);
+
+	return 0;
+
+out_cmap:
+	sh7760fb_blank(FB_BLANK_POWERDOWN, info);
+	fb_dealloc_cmap(&info->cmap);
+out_mem:
+	sh7760fb_free_mem(info);
+out_unmap:
+	if (par->irq >= 0)
+		free_irq(par->irq, &par->vsync);
+	iounmap(par->base);
+out_res:
+	release_resource(par->ioarea);
+	kfree(par->ioarea);
+out_fb:
+	framebuffer_release(info);
+	return ret;
+}
+
+static int __devexit sh7760fb_remove(struct platform_device *dev)
+{
+	struct fb_info *info = platform_get_drvdata(dev);
+	struct sh7760fb_par *par = info->par;
+
+	sh7760fb_blank(FB_BLANK_POWERDOWN, info);
+	unregister_framebuffer(info);
+	fb_dealloc_cmap(&info->cmap);
+	sh7760fb_free_mem(info);
+	if (par->irq >= 0)
+		free_irq(par->irq, par);
+	iounmap(par->base);
+	release_resource(par->ioarea);
+	kfree(par->ioarea);
+	framebuffer_release(info);
+	platform_set_drvdata(dev, NULL);
+
+	return 0;
+}
+
+static struct platform_driver sh7760_lcdc_driver = {
+	.driver = {
+		   .name = "sh7760-lcdc",
+		   .owner = THIS_MODULE,
+		   },
+	.probe = sh7760fb_probe,
+	.remove = __devexit_p(sh7760fb_remove),
+};
+
+static int __init sh7760fb_init(void)
+{
+	return platform_driver_register(&sh7760_lcdc_driver);
+}
+
+static void __exit sh7760fb_exit(void)
+{
+	platform_driver_unregister(&sh7760_lcdc_driver);
+}
+
+module_init(sh7760fb_init);
+module_exit(sh7760fb_exit);
+
+MODULE_AUTHOR("Nobuhiro Iwamatsu, Manuel Lauss");
+MODULE_DESCRIPTION("FBdev for SH7760/63 integrated LCD Controller");
+MODULE_LICENSE("GPL");
diff --git a/include/asm-sh/sh7760fb.h b/include/asm-sh/sh7760fb.h
new file mode 100644
index 000000000000..8767f61aceca
--- /dev/null
+++ b/include/asm-sh/sh7760fb.h
@@ -0,0 +1,197 @@
+/*
+ * sh7760fb.h -- platform data for SH7760/SH7763 LCDC framebuffer driver.
+ *
+ * (c) 2006-2008 MSC Vertriebsges.m.b.H.,
+ * 			Manuel Lauss <mano@roarinelk.homelinux.net>
+ * (c) 2008 Nobuhiro Iwamatsu <iwamatsu.nobuhiro@renesas.com>
+ */
+
+#ifndef _ASM_SH_SH7760FB_H
+#define _ASM_SH_SH7760FB_H
+
+/*
+ * some bits of the colormap registers should be written as zero.
+ * create a mask for that.
+ */
+#define SH7760FB_PALETTE_MASK 0x00f8fcf8
+
+/* The LCDC dma engine always sets bits 27-26 to 1: this is Area3 */
+#define SH7760FB_DMA_MASK 0x0C000000
+
+/* palette */
+#define LDPR(x) (((x) << 2))
+
+/* framebuffer registers and bits */
+#define LDICKR 0x400
+#define LDMTR 0x402
+/* see sh7760fb.h for LDMTR bits */
+#define LDDFR 0x404
+#define LDDFR_PABD (1 << 8)
+#define LDDFR_COLOR_MASK 0x7F
+#define LDSMR 0x406
+#define LDSMR_ROT (1 << 13)
+#define LDSARU 0x408
+#define LDSARL 0x40c
+#define LDLAOR 0x410
+#define LDPALCR 0x412
+#define LDPALCR_PALS (1 << 4)
+#define LDPALCR_PALEN (1 << 0)
+#define LDHCNR 0x414
+#define LDHSYNR 0x416
+#define LDVDLNR 0x418
+#define LDVTLNR 0x41a
+#define LDVSYNR 0x41c
+#define LDACLNR 0x41e
+#define LDINTR 0x420
+#define LDPMMR 0x424
+#define LDPSPR 0x426
+#define LDCNTR 0x428
+#define LDCNTR_DON (1 << 0)
+#define LDCNTR_DON2 (1 << 4)
+
+#ifdef CONFIG_CPU_SUBTYPE_SH7763
+# define LDLIRNR       0x440
+/* LDINTR bit */
+# define LDINTR_MINTEN (1 << 15)
+# define LDINTR_FINTEN (1 << 14)
+# define LDINTR_VSINTEN (1 << 13)
+# define LDINTR_VEINTEN (1 << 12)
+# define LDINTR_MINTS (1 << 11)
+# define LDINTR_FINTS (1 << 10)
+# define LDINTR_VSINTS (1 << 9)
+# define LDINTR_VEINTS (1 << 8)
+# define VINT_START (LDINTR_VSINTEN)
+# define VINT_CHECK (LDINTR_VSINTS)
+#else
+/* LDINTR bit */
+# define LDINTR_VINTSEL (1 << 12)
+# define LDINTR_VINTE (1 << 8)
+# define LDINTR_VINTS (1 << 0)
+# define VINT_START (LDINTR_VINTSEL)
+# define VINT_CHECK (LDINTR_VINTS)
+#endif
+
+/* HSYNC polarity inversion */
+#define LDMTR_FLMPOL (1 << 15)
+
+/* VSYNC polarity inversion */
+#define LDMTR_CL1POL (1 << 14)
+
+/* DISPLAY-ENABLE polarity inversion */
+#define LDMTR_DISPEN_LOWACT (1 << 13)
+
+/* DISPLAY DATA BUS polarity inversion */
+#define LDMTR_DPOL_LOWACT (1 << 12)
+
+/* AC modulation signal enable */
+#define LDMTR_MCNT (1 << 10)
+
+/* Disable output of HSYNC during VSYNC period */
+#define LDMTR_CL1CNT (1 << 9)
+
+/* Disable output of VSYNC during VSYNC period */
+#define LDMTR_CL2CNT (1 << 8)
+
+/* Display types supported by the LCDC */
+#define LDMTR_STN_MONO_4       0x00
+#define LDMTR_STN_MONO_8       0x01
+#define LDMTR_STN_COLOR_4      0x08
+#define LDMTR_STN_COLOR_8      0x09
+#define LDMTR_STN_COLOR_12     0x0A
+#define LDMTR_STN_COLOR_16     0x0B
+#define LDMTR_DSTN_MONO_8      0x11
+#define LDMTR_DSTN_MONO_16     0x13
+#define LDMTR_DSTN_COLOR_8     0x19
+#define LDMTR_DSTN_COLOR_12    0x1A
+#define LDMTR_DSTN_COLOR_16    0x1B
+#define LDMTR_TFT_COLOR_16     0x2B
+
+/* framebuffer color layout */
+#define LDDFR_1BPP_MONO 0x00
+#define LDDFR_2BPP_MONO 0x01
+#define LDDFR_4BPP_MONO 0x02
+#define LDDFR_6BPP_MONO 0x04
+#define LDDFR_4BPP 0x0A
+#define LDDFR_8BPP 0x0C
+#define LDDFR_16BPP_RGB555 0x1D
+#define LDDFR_16BPP_RGB565 0x2D
+
+/* LCDC Pixclock sources */
+#define LCDC_CLKSRC_BUSCLOCK 0
+#define LCDC_CLKSRC_PERIPHERAL 1
+#define LCDC_CLKSRC_EXTERNAL 2
+
+#define LDICKR_CLKSRC(x) \
+       (((x) & 3) << 12)
+
+/* LCDC pixclock input divider. Set to 1 at a minimum! */
+#define LDICKR_CLKDIV(x) \
+       ((x) & 0x1f)
+
+struct sh7760fb_platdata {
+
+	/* Set this member to a valid fb_videmode for the display you
+	 * wish to use.  The following members must be initialized:
+	 * xres, yres, hsync_len, vsync_len, sync,
+	 * {left,right,upper,lower}_margin.
+	 * The driver uses the above members to calculate register values
+	 * and memory requirements. Other members are ignored but may
+	 * be used by other framebuffer layer components.
+	 */
+	struct fb_videomode *def_mode;
+
+	/* LDMTR includes display type and signal polarity.  The
+	 * HSYNC/VSYNC polarities are derived from the fb_var_screeninfo
+	 * data above; however the polarities of the following signals
+	 * must be encoded in the ldmtr member:
+	 * Display Enable signal (default high-active)  DISPEN_LOWACT
+	 * Display Data signals (default high-active)   DPOL_LOWACT
+	 * AC Modulation signal (default off)           MCNT
+	 * Hsync-During-Vsync suppression (default off) CL1CNT
+	 * Vsync-during-vsync suppression (default off) CL2CNT
+	 * NOTE: also set a display type!
+	 * (one of LDMTR_{STN,DSTN,TFT}_{MONO,COLOR}_{4,8,12,16})
+	 */
+	u16 ldmtr;
+
+	/* LDDFR controls framebuffer image format (depth, organization)
+	 * Use ONE of the LDDFR_?BPP_* macros!
+	 */
+	u16 lddfr;
+
+	/* LDPMMR and LDPSPR control the timing of the power signals
+	 * for the display. Please read the SH7760 Hardware Manual,
+	 * Chapters 30.3.17, 30.3.18 and 30.4.6!
+	 */
+	u16 ldpmmr;
+	u16 ldpspr;
+
+	/* LDACLNR contains the line numbers after which the AC modulation
+	 * signal is to toggle. Set to ZERO for TFTs or displays which
+	 * do not need it. (Chapter 30.3.15 in SH7760 Hardware Manual).
+	 */
+	u16 ldaclnr;
+
+	/* LDICKR contains information on pixelclock source and config.
+	 * Please use the LDICKR_CLKSRC() and LDICKR_CLKDIV() macros.
+	 * minimal value for CLKDIV() must be 1!.
+	 */
+	u16 ldickr;
+
+	/* set this member to 1 if you wish to use the LCDC's hardware
+	 * rotation function.  This is limited to displays <= 320x200
+	 * pixels resolution!
+	 */
+	int rotate;		/* set to 1 to rotate 90 CCW */
+
+	/* set this to 1 to suppress vsync irq use. */
+	int novsync;
+
+	/* blanking hook for platform. Set this if your platform can do
+	 * more than the LCDC in terms of blanking (e.g. disable clock
+	 * generator / backlight power supply / etc.
+	 */
+	void (*blank) (int);
+};
+
+#endif /* _ASM_SH_SH7760FB_H */

From 2d04a4a72d7e1519b4838f24bdd4b5d0f3f426dc Mon Sep 17 00:00:00 2001
From: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Date: Wed, 23 Jul 2008 21:31:48 -0700
Subject: [PATCH 0518/1171] fbcon: bgcolor fix

The fourth bit of the background color is the blink property bit, not the
intensity bit, as for the foreground color.  Therefore it shouldn't be
included in the background color.

Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/console/fbcon.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/video/console/fbcon.h b/drivers/video/console/fbcon.h
index 0135e0395456..beb6a74fc7f2 100644
--- a/drivers/video/console/fbcon.h
+++ b/drivers/video/console/fbcon.h
@@ -92,7 +92,7 @@ struct fbcon_ops {
 #define attr_fgcol(fgshift,s)    \
 	(((s) >> (fgshift)) & 0x0f)
 #define attr_bgcol(bgshift,s)    \
-	(((s) >> (bgshift)) & 0x0f)
+	(((s) >> (bgshift)) & 0x07)
 
 /* Monochrome */
 #define attr_bold(s) \

From 3e074058d72486676f6fdf6fe803200c62dcb403 Mon Sep 17 00:00:00 2001
From: Hans-Christian Egtvedt <hans-christian.egtvedt@atmel.com>
Date: Wed, 23 Jul 2008 21:31:48 -0700
Subject: [PATCH 0519/1171] fbdev: LCD backlight driver using Atmel PWM driver

This patch adds a platform driver using the ATMEL PWM driver to control a
backlight which requires a PWM signal and optional GPIO signal for discrete
on/off signal.  It has been tested on Favr-32 board from EarthLCD.

The driver is configurable by supplying a struct with the platform data.  See
the include/linux/atmel-pwm-bl.h for details.

The board code for Favr-32 will be submitted to the AVR32 kernel list.

Signed-off-by: Hans-Christian Egtvedt <hans-christian.egtvedt@atmel.com>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Cc: Haavard Skinnemoen <hskinnemoen@atmel.com>
Cc: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/backlight/Kconfig        |  12 ++
 drivers/video/backlight/Makefile       |   1 +
 drivers/video/backlight/atmel-pwm-bl.c | 244 +++++++++++++++++++++++++
 include/linux/atmel-pwm-bl.h           |  43 +++++
 4 files changed, 300 insertions(+)
 create mode 100644 drivers/video/backlight/atmel-pwm-bl.c
 create mode 100644 include/linux/atmel-pwm-bl.h

diff --git a/drivers/video/backlight/Kconfig b/drivers/video/backlight/Kconfig
index b289e197e55d..98d9faf4970d 100644
--- a/drivers/video/backlight/Kconfig
+++ b/drivers/video/backlight/Kconfig
@@ -87,6 +87,18 @@ config BACKLIGHT_ATMEL_LCDC
 	  If in doubt, it's safe to enable this option; it doesn't kick
 	  in unless the board's description says it's wired that way.
 
+config BACKLIGHT_ATMEL_PWM
+	tristate "Atmel PWM backlight control"
+	depends on BACKLIGHT_CLASS_DEVICE && ATMEL_PWM
+	default n
+	help
+	  Say Y here if you want to use the PWM peripheral in Atmel AT91 and
+	  AVR32 devices. This driver will need additional platform data to know
+	  which PWM instance to use and how to configure it.
+
+	  To compile this driver as a module, choose M here: the module will be
+	  called atmel-pwm-bl.
+
 config BACKLIGHT_CORGI
 	tristate "Generic (aka Sharp Corgi) Backlight Driver"
 	depends on BACKLIGHT_CLASS_DEVICE
diff --git a/drivers/video/backlight/Makefile b/drivers/video/backlight/Makefile
index 7d31c14088aa..d8a08e468cc7 100644
--- a/drivers/video/backlight/Makefile
+++ b/drivers/video/backlight/Makefile
@@ -7,6 +7,7 @@ obj-$(CONFIG_LCD_PLATFORM)	   += platform_lcd.o
 obj-$(CONFIG_LCD_VGG2432A4)	   += vgg2432a4.o
 
 obj-$(CONFIG_BACKLIGHT_CLASS_DEVICE) += backlight.o
+obj-$(CONFIG_BACKLIGHT_ATMEL_PWM)    += atmel-pwm-bl.o
 obj-$(CONFIG_BACKLIGHT_CORGI)	+= corgi_bl.o
 obj-$(CONFIG_BACKLIGHT_HP680)	+= hp680_bl.o
 obj-$(CONFIG_BACKLIGHT_LOCOMO)	+= locomolcd.o
diff --git a/drivers/video/backlight/atmel-pwm-bl.c b/drivers/video/backlight/atmel-pwm-bl.c
new file mode 100644
index 000000000000..505c0823a105
--- /dev/null
+++ b/drivers/video/backlight/atmel-pwm-bl.c
@@ -0,0 +1,244 @@
+/*
+ * Copyright (C) 2008 Atmel Corporation
+ *
+ * Backlight driver using Atmel PWM peripheral.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ */
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/fb.h>
+#include <linux/clk.h>
+#include <linux/gpio.h>
+#include <linux/backlight.h>
+#include <linux/atmel_pwm.h>
+#include <linux/atmel-pwm-bl.h>
+
+struct atmel_pwm_bl {
+	const struct atmel_pwm_bl_platform_data	*pdata;
+	struct backlight_device			*bldev;
+	struct platform_device			*pdev;
+	struct pwm_channel			pwmc;
+	int					gpio_on;
+};
+
+static int atmel_pwm_bl_set_intensity(struct backlight_device *bd)
+{
+	struct atmel_pwm_bl *pwmbl = bl_get_data(bd);
+	int intensity = bd->props.brightness;
+	int pwm_duty;
+
+	if (bd->props.power != FB_BLANK_UNBLANK)
+		intensity = 0;
+	if (bd->props.fb_blank != FB_BLANK_UNBLANK)
+		intensity = 0;
+
+	if (pwmbl->pdata->pwm_active_low)
+		pwm_duty = pwmbl->pdata->pwm_duty_min + intensity;
+	else
+		pwm_duty = pwmbl->pdata->pwm_duty_max - intensity;
+
+	if (pwm_duty > pwmbl->pdata->pwm_duty_max)
+		pwm_duty = pwmbl->pdata->pwm_duty_max;
+	if (pwm_duty < pwmbl->pdata->pwm_duty_min)
+		pwm_duty = pwmbl->pdata->pwm_duty_min;
+
+	if (!intensity) {
+		if (pwmbl->gpio_on != -1) {
+			gpio_set_value(pwmbl->gpio_on,
+					0 ^ pwmbl->pdata->on_active_low);
+		}
+		pwm_channel_writel(&pwmbl->pwmc, PWM_CUPD, pwm_duty);
+		pwm_channel_disable(&pwmbl->pwmc);
+	} else {
+		pwm_channel_enable(&pwmbl->pwmc);
+		pwm_channel_writel(&pwmbl->pwmc, PWM_CUPD, pwm_duty);
+		if (pwmbl->gpio_on != -1) {
+			gpio_set_value(pwmbl->gpio_on,
+					1 ^ pwmbl->pdata->on_active_low);
+		}
+	}
+
+	return 0;
+}
+
+static int atmel_pwm_bl_get_intensity(struct backlight_device *bd)
+{
+	struct atmel_pwm_bl *pwmbl = bl_get_data(bd);
+	u8 intensity;
+
+	if (pwmbl->pdata->pwm_active_low) {
+		intensity = pwm_channel_readl(&pwmbl->pwmc, PWM_CDTY) -
+			pwmbl->pdata->pwm_duty_min;
+	} else {
+		intensity = pwmbl->pdata->pwm_duty_max -
+			pwm_channel_readl(&pwmbl->pwmc, PWM_CDTY);
+	}
+
+	return intensity;
+}
+
+static int atmel_pwm_bl_init_pwm(struct atmel_pwm_bl *pwmbl)
+{
+	unsigned long pwm_rate = pwmbl->pwmc.mck;
+	unsigned long prescale = DIV_ROUND_UP(pwm_rate,
+			(pwmbl->pdata->pwm_frequency *
+			 pwmbl->pdata->pwm_compare_max)) - 1;
+
+	/*
+	 * Prescale must be power of two and maximum 0xf in size because of
+	 * hardware limit. PWM speed will be:
+	 *	PWM module clock speed / (2 ^ prescale).
+	 */
+	prescale = fls(prescale);
+	if (prescale > 0xf)
+		prescale = 0xf;
+
+	pwm_channel_writel(&pwmbl->pwmc, PWM_CMR, prescale);
+	pwm_channel_writel(&pwmbl->pwmc, PWM_CDTY,
+			pwmbl->pdata->pwm_duty_min +
+			pwmbl->bldev->props.brightness);
+	pwm_channel_writel(&pwmbl->pwmc, PWM_CPRD,
+			pwmbl->pdata->pwm_compare_max);
+
+	dev_info(&pwmbl->pdev->dev, "Atmel PWM backlight driver "
+			"(%lu Hz)\n", pwmbl->pwmc.mck /
+			pwmbl->pdata->pwm_compare_max /
+			(1 << prescale));
+
+	return pwm_channel_enable(&pwmbl->pwmc);
+}
+
+static struct backlight_ops atmel_pwm_bl_ops = {
+	.get_brightness = atmel_pwm_bl_get_intensity,
+	.update_status  = atmel_pwm_bl_set_intensity,
+};
+
+static int atmel_pwm_bl_probe(struct platform_device *pdev)
+{
+	const struct atmel_pwm_bl_platform_data *pdata;
+	struct backlight_device *bldev;
+	struct atmel_pwm_bl *pwmbl;
+	int retval;
+
+	pwmbl = kzalloc(sizeof(struct atmel_pwm_bl), GFP_KERNEL);
+	if (!pwmbl)
+		return -ENOMEM;
+
+	pwmbl->pdev = pdev;
+
+	pdata = pdev->dev.platform_data;
+	if (!pdata) {
+		retval = -ENODEV;
+		goto err_free_mem;
+	}
+
+	if (pdata->pwm_compare_max < pdata->pwm_duty_max ||
+			pdata->pwm_duty_min > pdata->pwm_duty_max ||
+			pdata->pwm_frequency == 0) {
+		retval = -EINVAL;
+		goto err_free_mem;
+	}
+
+	pwmbl->pdata = pdata;
+	pwmbl->gpio_on = pdata->gpio_on;
+
+	retval = pwm_channel_alloc(pdata->pwm_channel, &pwmbl->pwmc);
+	if (retval)
+		goto err_free_mem;
+
+	if (pwmbl->gpio_on != -1) {
+		retval = gpio_request(pwmbl->gpio_on, "gpio_atmel_pwm_bl");
+		if (retval) {
+			pwmbl->gpio_on = -1;
+			goto err_free_pwm;
+		}
+
+		/* Turn display off by defatult. */
+		retval = gpio_direction_output(pwmbl->gpio_on,
+				0 ^ pdata->on_active_low);
+		if (retval)
+			goto err_free_gpio;
+	}
+
+	bldev = backlight_device_register("atmel-pwm-bl",
+			&pdev->dev, pwmbl, &atmel_pwm_bl_ops);
+	if (IS_ERR(bldev)) {
+		retval = PTR_ERR(bldev);
+		goto err_free_gpio;
+	}
+
+	pwmbl->bldev = bldev;
+
+	platform_set_drvdata(pdev, pwmbl);
+
+	/* Power up the backlight by default at middle intesity. */
+	bldev->props.power = FB_BLANK_UNBLANK;
+	bldev->props.max_brightness = pdata->pwm_duty_max - pdata->pwm_duty_min;
+	bldev->props.brightness = bldev->props.max_brightness / 2;
+
+	retval = atmel_pwm_bl_init_pwm(pwmbl);
+	if (retval)
+		goto err_free_bl_dev;
+
+	atmel_pwm_bl_set_intensity(bldev);
+
+	return 0;
+
+err_free_bl_dev:
+	platform_set_drvdata(pdev, NULL);
+	backlight_device_unregister(bldev);
+err_free_gpio:
+	if (pwmbl->gpio_on != -1)
+		gpio_free(pwmbl->gpio_on);
+err_free_pwm:
+	pwm_channel_free(&pwmbl->pwmc);
+err_free_mem:
+	kfree(pwmbl);
+	return retval;
+}
+
+static int __exit atmel_pwm_bl_remove(struct platform_device *pdev)
+{
+	struct atmel_pwm_bl *pwmbl = platform_get_drvdata(pdev);
+
+	if (pwmbl->gpio_on != -1) {
+		gpio_set_value(pwmbl->gpio_on, 0);
+		gpio_free(pwmbl->gpio_on);
+	}
+	pwm_channel_disable(&pwmbl->pwmc);
+	pwm_channel_free(&pwmbl->pwmc);
+	backlight_device_unregister(pwmbl->bldev);
+	platform_set_drvdata(pdev, NULL);
+	kfree(pwmbl);
+
+	return 0;
+}
+
+static struct platform_driver atmel_pwm_bl_driver = {
+	.driver = {
+		.name = "atmel-pwm-bl",
+	},
+	/* REVISIT add suspend() and resume() */
+	.remove = __exit_p(atmel_pwm_bl_remove),
+};
+
+static int __init atmel_pwm_bl_init(void)
+{
+	return platform_driver_probe(&atmel_pwm_bl_driver, atmel_pwm_bl_probe);
+}
+module_init(atmel_pwm_bl_init);
+
+static void __exit atmel_pwm_bl_exit(void)
+{
+	platform_driver_unregister(&atmel_pwm_bl_driver);
+}
+module_exit(atmel_pwm_bl_exit);
+
+MODULE_AUTHOR("Hans-Christian egtvedt <hans-christian.egtvedt@atmel.com>");
+MODULE_DESCRIPTION("Atmel PWM backlight driver");
+MODULE_LICENSE("GPL");
diff --git a/include/linux/atmel-pwm-bl.h b/include/linux/atmel-pwm-bl.h
new file mode 100644
index 000000000000..0153a47806c2
--- /dev/null
+++ b/include/linux/atmel-pwm-bl.h
@@ -0,0 +1,43 @@
+/*
+ * Copyright (C) 2007 Atmel Corporation
+ *
+ * Driver for the AT32AP700X PS/2 controller (PSIF).
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation.
+ */
+
+#ifndef __INCLUDE_ATMEL_PWM_BL_H
+#define __INCLUDE_ATMEL_PWM_BL_H
+
+/**
+ * struct atmel_pwm_bl_platform_data
+ * @pwm_channel: which PWM channel in the PWM module to use.
+ * @pwm_frequency: PWM frequency to generate, the driver will try to be as
+ *	close as the prescaler allows.
+ * @pwm_compare_max: value to use in the PWM channel compare register.
+ * @pwm_duty_max: maximum duty cycle value, must be less than or equal to
+ *	pwm_compare_max.
+ * @pwm_duty_min: minimum duty cycle value, must be less than pwm_duty_max.
+ * @pwm_active_low: set to one if the low part of the PWM signal increases the
+ *	brightness of the backlight.
+ * @gpio_on: GPIO line to control the backlight on/off, set to -1 if not used.
+ * @on_active_low: set to one if the on/off signal is on when GPIO is low.
+ *
+ * This struct must be added to the platform device in the board code. It is
+ * used by the atmel-pwm-bl driver to setup the GPIO to control on/off and the
+ * PWM device.
+ */
+struct atmel_pwm_bl_platform_data {
+	unsigned int pwm_channel;
+	unsigned int pwm_frequency;
+	unsigned int pwm_compare_max;
+	unsigned int pwm_duty_max;
+	unsigned int pwm_duty_min;
+	unsigned int pwm_active_low;
+	int gpio_on;
+	unsigned int on_active_low;
+};
+
+#endif /* __INCLUDE_ATMEL_PWM_BL_H */

From cba603bf514c101bf48f6adf393c3d00ed457a57 Mon Sep 17 00:00:00 2001
From: Jan Beulich <jbeulich@novell.com>
Date: Wed, 23 Jul 2008 21:31:49 -0700
Subject: [PATCH 0520/1171] fbcon: remove stray semicolons

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/console/fbcon.h | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/video/console/fbcon.h b/drivers/video/console/fbcon.h
index beb6a74fc7f2..de1b1365279b 100644
--- a/drivers/video/console/fbcon.h
+++ b/drivers/video/console/fbcon.h
@@ -146,10 +146,8 @@ static inline int attr_col_ec(int shift, struct vc_data *vc,
 	return is_fg ? fg : bg;
 }
 
-#define attr_bgcol_ec(bgshift,vc,info)		\
-	attr_col_ec(bgshift,vc,info,0);
-#define attr_fgcol_ec(fgshift,vc,info)		\
-	attr_col_ec(fgshift,vc,info,1);
+#define attr_bgcol_ec(bgshift, vc, info) attr_col_ec(bgshift, vc, info, 0)
+#define attr_fgcol_ec(fgshift, vc, info) attr_col_ec(fgshift, vc, info, 1)
 
 /* Font */
 #define REFCOUNT(fd)	(((int *)(fd))[-1])

From 5bb49fcd501aa9fd3d321a22b7c01d9b0db7ab36 Mon Sep 17 00:00:00 2001
From: Philippe De Muyter <phdm@macqel.be>
Date: Wed, 23 Jul 2008 21:31:50 -0700
Subject: [PATCH 0521/1171] video/fb: cleanup FB_MAJOR usage

Currently, linux/major.h defines a GRAPHDEV_MAJOR (29) that nobody uses,
and linux/fb.h defines the real FB_MAJOR (also 29), that only fbmem.c
needs.  Drop GRAPHDEV_MAJOR from major.h, move FB_MAJOR definition from
fb.h to major.h, and fix fbmem.c to use major.h's definition.

Signed-off-by: Philippe De Muyter <phdm@macqel.be>
Cc: Krzysztof Helt <krzysztof.h1@poczta.fm>
Cc: "Antonino A. Daplas" <adaplas@pol.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/fbmem.c | 1 +
 include/linux/fb.h    | 1 -
 include/linux/major.h | 2 +-
 3 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c
index 6b487801eeae..5d84b3431098 100644
--- a/drivers/video/fbmem.c
+++ b/drivers/video/fbmem.c
@@ -35,6 +35,7 @@
 #include <linux/device.h>
 #include <linux/efi.h>
 #include <linux/fb.h>
+#include <linux/major.h>
 
 #include <asm/fb.h>
 
diff --git a/include/linux/fb.h b/include/linux/fb.h
index a084d1335869..3b8870e32afd 100644
--- a/include/linux/fb.h
+++ b/include/linux/fb.h
@@ -8,7 +8,6 @@ struct dentry;
 
 /* Definitions of frame buffers						*/
 
-#define FB_MAJOR		29
 #define FB_MAX			32	/* sufficient for now */
 
 /* ioctls
diff --git a/include/linux/major.h b/include/linux/major.h
index 0cb98053537a..53d5fafd85c3 100644
--- a/include/linux/major.h
+++ b/include/linux/major.h
@@ -53,7 +53,7 @@
 #define STL_SIOMEMMAJOR		28
 #define ACSI_MAJOR		28
 #define AZTECH_CDROM_MAJOR	29
-#define GRAPHDEV_MAJOR		29   /* SparcLinux & Linux/68k /dev/fb */
+#define FB_MAJOR		29   /* /dev/fb* framebuffers */
 #define CM206_CDROM_MAJOR	32
 #define IDE2_MAJOR		33
 #define IDE3_MAJOR		34

From b340e8a57ef381e69c99a7a8ede61a6bf71a8014 Mon Sep 17 00:00:00 2001
From: Akinobu Mita <akinobu.mita@gmail.com>
Date: Wed, 23 Jul 2008 21:31:51 -0700
Subject: [PATCH 0522/1171] auxdisplay: small cleanups

- Use BUILD_BUG_ON for CFAG12864B_SIZE instead of runtime-check

- Use get_zeroed_page()

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Miguel Ojeda Sandonis <maxextreme@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/auxdisplay/cfag12864b.c | 13 ++-----------
 1 file changed, 2 insertions(+), 11 deletions(-)

diff --git a/drivers/auxdisplay/cfag12864b.c b/drivers/auxdisplay/cfag12864b.c
index 683509f013ab..eacb175f6bd3 100644
--- a/drivers/auxdisplay/cfag12864b.c
+++ b/drivers/auxdisplay/cfag12864b.c
@@ -336,16 +336,9 @@ static int __init cfag12864b_init(void)
 			"ks0108 is not initialized\n");
 		goto none;
 	}
+	BUILD_BUG_ON(PAGE_SIZE < CFAG12864B_SIZE);
 
-	if (PAGE_SIZE < CFAG12864B_SIZE) {
-		printk(KERN_ERR CFAG12864B_NAME ": ERROR: "
-			"page size (%i) < cfag12864b size (%i)\n",
-			(unsigned int)PAGE_SIZE, CFAG12864B_SIZE);
-		ret = -ENOMEM;
-		goto none;
-	}
-
-	cfag12864b_buffer = (unsigned char *) __get_free_page(GFP_KERNEL);
+	cfag12864b_buffer = (unsigned char *) get_zeroed_page(GFP_KERNEL);
 	if (cfag12864b_buffer == NULL) {
 		printk(KERN_ERR CFAG12864B_NAME ": ERROR: "
 			"can't get a free page\n");
@@ -367,8 +360,6 @@ static int __init cfag12864b_init(void)
 	if (cfag12864b_workqueue == NULL)
 		goto cachealloced;
 
-	memset(cfag12864b_buffer, 0, CFAG12864B_SIZE);
-
 	cfag12864b_clear();
 	cfag12864b_on();
 

From f9247273cb69ba101877e946d2d83044409cc8c5 Mon Sep 17 00:00:00 2001
From: Steven Whitehouse <swhiteho@redhat.com>
Date: Thu, 24 Jul 2008 17:22:13 +0100
Subject: [PATCH 0523/1171] UFS: add const to parser token table

This patch adds a "const" to the parser token table. I've done an
allmodconfig build to see if this produces any warnings/failures and the
patch includes a fix for the only warning that was produced.

Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
Acked-by: Alexander Viro <aviro@redhat.com>
Acked-by: Evgeniy Dushistov <dushistov@mail.ru>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ufs/super.c         | 2 +-
 include/linux/parser.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/fs/ufs/super.c b/fs/ufs/super.c
index 85b22b5977fa..506f724055c2 100644
--- a/fs/ufs/super.c
+++ b/fs/ufs/super.c
@@ -1232,7 +1232,7 @@ static int ufs_show_options(struct seq_file *seq, struct vfsmount *vfs)
 {
 	struct ufs_sb_info *sbi = UFS_SB(vfs->mnt_sb);
 	unsigned mval = sbi->s_mount_opt & UFS_MOUNT_UFSTYPE;
-	struct match_token *tp = tokens;
+	const struct match_token *tp = tokens;
 
 	while (tp->token != Opt_onerror_panic && tp->token != mval)
 		++tp;
diff --git a/include/linux/parser.h b/include/linux/parser.h
index 7dcd05075756..cc554ca8bc78 100644
--- a/include/linux/parser.h
+++ b/include/linux/parser.h
@@ -14,7 +14,7 @@ struct match_token {
 	const char *pattern;
 };
 
-typedef struct match_token match_table_t[];
+typedef const struct match_token match_table_t[];
 
 /* Maximum number of arguments that match_token will find in a pattern */
 enum {MAX_OPT_ARGS = 3};

From 6209ed9d8443b63c36d340908530fa470c4d4fff Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Thu, 24 Jul 2008 12:49:26 -0700
Subject: [PATCH 0524/1171] x86-64: make BUILD_IRQ() also reset section back

Commit 9d25d4db81833029d30b7b03cc1000cbbe09e192 ("x86: BUILD_IRQ say
.text to avoid .data.percpu") added a ".text" specifier to make sure
that BUILD_IRQ() builds the irq trampoline in the text segment rather
than in some random left-over segment that the compiler happened to
leave the asm in.

However, we should also make sure that we switch back by adding a
".previous" at the end, so that there are no subtle issues with
subsequent compiler-generated code.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/kernel/irqinit_64.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/x86/kernel/irqinit_64.c b/arch/x86/kernel/irqinit_64.c
index 9414125f19ce..1f26fd9ec4f4 100644
--- a/arch/x86/kernel/irqinit_64.c
+++ b/arch/x86/kernel/irqinit_64.c
@@ -46,7 +46,8 @@
 	asm("\n.text\n.p2align\n"		\
 	    "IRQ" #nr "_interrupt:\n\t"		\
 	    "push $~(" #nr ") ; "		\
-	    "jmp common_interrupt");
+	    "jmp common_interrupt\n"		\
+	    ".previous");
 
 #define BI(x,y) \
 	BUILD_IRQ(x##y)

From e2d2867ff8700d7431c68c089ff5f5ed7f2d5b40 Mon Sep 17 00:00:00 2001
From: Chris Wright <chrisw@sous-sol.org>
Date: Thu, 24 Jul 2008 20:43:34 +0000
Subject: [PATCH 0525/1171] When verifying the decoded header before decoding
 the object identifier (expecting a SPNEGO pseudo-mechanism oid), the test to
 verify it is a primitive encoding is compared against the asn1 class. 
 Primitive is not a class.  This brings check in line with similar check for
 krb/ntlmssp oid.

Signed-off-by: Chris Wright <chrisw@sous-sol.org>
Cc: Steven French <sfrench@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Steve French <sfrench@us.ibm.com>
---
 fs/cifs/asn1.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/cifs/asn1.c b/fs/cifs/asn1.c
index 42765465701c..6bb440b257b0 100644
--- a/fs/cifs/asn1.c
+++ b/fs/cifs/asn1.c
@@ -494,7 +494,7 @@ decode_negTokenInit(unsigned char *security_blob, int length,
 		/*      remember to free obj->oid */
 		rc = asn1_header_decode(&ctx, &end, &cls, &con, &tag);
 		if (rc) {
-			if ((tag == ASN1_OJI) && (cls == ASN1_PRI)) {
+			if ((tag == ASN1_OJI) && (con == ASN1_PRI)) {
 				rc = asn1_oid_decode(&ctx, end, &oid, &oidlen);
 				if (rc) {
 					rc = compare_oid(oid, oidlen,

From 708e5f9eb68589b87724af3f0fb4e681dfdfd69f Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:11 +0200
Subject: [PATCH 0526/1171] ide: always call ->init_chipset method in
 do_ide_setup_pci_device()

Call ->init_chipset method also for 'tried_config' / '!pciirq' conditions.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/setup-pci.c | 24 ++++++++++--------------
 1 file changed, 10 insertions(+), 14 deletions(-)

diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
index b15cad58dc81..a95d51080237 100644
--- a/drivers/ide/setup-pci.c
+++ b/drivers/ide/setup-pci.c
@@ -499,20 +499,21 @@ static int do_ide_setup_pci_device(struct pci_dev *dev,
 	 */
 	pciirq = dev->irq;
 
+	/*
+	 * This allows offboard ide-pci cards the enable a BIOS,
+	 * verify interrupt settings of split-mirror pci-config
+	 * space, place chipset into init-mode, and/or preserve
+	 * an interrupt if the card is not native ide support.
+	 */
+	ret = d->init_chipset ? d->init_chipset(dev, d->name) : 0;
+	if (ret < 0)
+		goto out;
+
 	/* Is it an "IDE storage" device in non-PCI mode? */
 	if ((dev->class >> 8) == PCI_CLASS_STORAGE_IDE && (dev->class & 5) != 5) {
 		if (noisy)
 			printk(KERN_INFO "%s: not 100%% native mode: "
 				"will probe irqs later\n", d->name);
-		/*
-		 * This allows offboard ide-pci cards the enable a BIOS,
-		 * verify interrupt settings of split-mirror pci-config
-		 * space, place chipset into init-mode, and/or preserve
-		 * an interrupt if the card is not native ide support.
-		 */
-		ret = d->init_chipset ? d->init_chipset(dev, d->name) : 0;
-		if (ret < 0)
-			goto out;
 		pciirq = ret;
 	} else if (tried_config) {
 		if (noisy)
@@ -524,11 +525,6 @@ static int do_ide_setup_pci_device(struct pci_dev *dev,
 				d->name, pciirq);
 		pciirq = 0;
 	} else {
-		if (d->init_chipset) {
-			ret = d->init_chipset(dev, d->name);
-			if (ret < 0)
-				goto out;
-		}
 		if (noisy)
 			printk(KERN_INFO "%s: 100%% native mode on irq %d\n",
 				d->name, pciirq);

From a95925a309cd9a2e7f5a5713fd70e0dadb09890c Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:11 +0200
Subject: [PATCH 0527/1171] ide: respect dev->irq in do_ide_setup_pci_device()
 also if 'tried_config'

* If device is in the PCI native mode respect dev->irq regardless of
  'tried_config' in do_ide_setup_pci_device().

* Drop no longer needed 'config' argument from ide_setup_pci_controller().

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/setup-pci.c | 12 +++---------
 1 file changed, 3 insertions(+), 9 deletions(-)

diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
index a95d51080237..58f4a95f6f5c 100644
--- a/drivers/ide/setup-pci.c
+++ b/drivers/ide/setup-pci.c
@@ -393,14 +393,14 @@ int ide_hwif_setup_dma(ide_hwif_t *hwif, const struct ide_port_info *d)
  *	@dev: PCI device
  *	@d: IDE port info
  *	@noisy: verbose flag
- *	@config: returned as 1 if we configured the hardware
  *
  *	Set up the PCI and controller side of the IDE interface. This brings
  *	up the PCI side of the device, checks that the device is enabled
  *	and enables it if need be
  */
 
-static int ide_setup_pci_controller(struct pci_dev *dev, const struct ide_port_info *d, int noisy, int *config)
+static int ide_setup_pci_controller(struct pci_dev *dev,
+				    const struct ide_port_info *d, int noisy)
 {
 	int ret;
 	u16 pcicmd;
@@ -421,7 +421,6 @@ static int ide_setup_pci_controller(struct pci_dev *dev, const struct ide_port_i
 		ret = ide_pci_configure(dev, d);
 		if (ret < 0)
 			goto out;
-		*config = 1;
 		printk(KERN_INFO "%s: device enabled (Linux)\n", d->name);
 	}
 
@@ -487,10 +486,9 @@ static int do_ide_setup_pci_device(struct pci_dev *dev,
 				   const struct ide_port_info *d,
 				   u8 noisy)
 {
-	int tried_config = 0;
 	int pciirq, ret;
 
-	ret = ide_setup_pci_controller(dev, d, noisy, &tried_config);
+	ret = ide_setup_pci_controller(dev, d, noisy);
 	if (ret < 0)
 		goto out;
 
@@ -515,10 +513,6 @@ static int do_ide_setup_pci_device(struct pci_dev *dev,
 			printk(KERN_INFO "%s: not 100%% native mode: "
 				"will probe irqs later\n", d->name);
 		pciirq = ret;
-	} else if (tried_config) {
-		if (noisy)
-			printk(KERN_INFO "%s: will probe irqs later\n", d->name);
-		pciirq = 0;
 	} else if (!pciirq) {
 		if (noisy)
 			printk(KERN_WARNING "%s: bad irq (%d): will probe later\n",

From a742d6cf0b37b1a96a1549b1fda0d6b19e0185c2 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:12 +0200
Subject: [PATCH 0528/1171] ide: move ide_setup_pci_controller() call to
 ide_setup_pci_device[s]()

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/setup-pci.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
index 58f4a95f6f5c..98c663d62b9f 100644
--- a/drivers/ide/setup-pci.c
+++ b/drivers/ide/setup-pci.c
@@ -488,10 +488,6 @@ static int do_ide_setup_pci_device(struct pci_dev *dev,
 {
 	int pciirq, ret;
 
-	ret = ide_setup_pci_controller(dev, d, noisy);
-	if (ret < 0)
-		goto out;
-
 	/*
 	 * Can we trust the reported IRQ?
 	 */
@@ -534,6 +530,10 @@ int ide_setup_pci_device(struct pci_dev *dev, const struct ide_port_info *d)
 	hw_regs_t hw[4], *hws[] = { NULL, NULL, NULL, NULL };
 	int ret;
 
+	ret = ide_setup_pci_controller(dev, d, 1);
+	if (ret < 0)
+		goto out;
+
 	ret = do_ide_setup_pci_device(dev, d, 1);
 
 	if (ret >= 0) {
@@ -542,7 +542,7 @@ int ide_setup_pci_device(struct pci_dev *dev, const struct ide_port_info *d)
 
 		ret = ide_host_add(d, hws, NULL);
 	}
-
+out:
 	return ret;
 }
 EXPORT_SYMBOL_GPL(ide_setup_pci_device);
@@ -555,6 +555,10 @@ int ide_setup_pci_devices(struct pci_dev *dev1, struct pci_dev *dev2,
 	hw_regs_t hw[4], *hws[] = { NULL, NULL, NULL, NULL };
 
 	for (i = 0; i < 2; i++) {
+		ret = ide_setup_pci_controller(pdev[i], d, !i);
+		if (ret < 0)
+			goto out;
+
 		ret = do_ide_setup_pci_device(pdev[i], d, !i);
 
 		/*

From 8c2eece50a368c7986bae0b3e52739558dd71b51 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:12 +0200
Subject: [PATCH 0529/1171] ide: call ide_pci_setup_ports() before
 do_ide_setup_pci_device()

* Call ide_pci_setup_ports() before do_ide_setup_pci_device()
  in ide_setup_pci_device[s]().

While at it:

* Remove stale FIXMEs.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/setup-pci.c | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)

diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
index 98c663d62b9f..b85de71fdc88 100644
--- a/drivers/ide/setup-pci.c
+++ b/drivers/ide/setup-pci.c
@@ -534,14 +534,16 @@ int ide_setup_pci_device(struct pci_dev *dev, const struct ide_port_info *d)
 	if (ret < 0)
 		goto out;
 
+	ide_pci_setup_ports(dev, d, 0, &hw[0], &hws[0]);
+
 	ret = do_ide_setup_pci_device(dev, d, 1);
+	if (ret < 0)
+		goto out;
 
-	if (ret >= 0) {
-		/* FIXME: silent failure can happen */
-		ide_pci_setup_ports(dev, d, ret, &hw[0], &hws[0]);
+	/* fixup IRQ */
+	hw[1].irq = hw[0].irq = ret;
 
-		ret = ide_host_add(d, hws, NULL);
-	}
+	ret = ide_host_add(d, hws, NULL);
 out:
 	return ret;
 }
@@ -559,6 +561,8 @@ int ide_setup_pci_devices(struct pci_dev *dev1, struct pci_dev *dev2,
 		if (ret < 0)
 			goto out;
 
+		ide_pci_setup_ports(pdev[i], d, 0, &hw[i*2], &hws[i*2]);
+
 		ret = do_ide_setup_pci_device(pdev[i], d, !i);
 
 		/*
@@ -568,8 +572,8 @@ int ide_setup_pci_devices(struct pci_dev *dev1, struct pci_dev *dev2,
 		if (ret < 0)
 			goto out;
 
-		/* FIXME: silent failure can happen */
-		ide_pci_setup_ports(pdev[i], d, ret, &hw[i*2], &hws[i*2]);
+		/* fixup IRQ */
+		hw[i*2 + 1].irq = hw[i*2].irq = ret;
 	}
 
 	ret = ide_host_add(d, hws, NULL);

From 6cdf6eb357c2681596b7b1672b92396ba82333d4 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:14 +0200
Subject: [PATCH 0530/1171] ide: add ->dev and ->host_priv fields to struct
 ide_host

* Add 'struct device *dev[2]' and 'void *host_priv' fields
  to struct ide_host.

* Set ->dev[] in ide_host_alloc_all()/ide_setup_pci_device[s]().

* Pass 'void *priv' argument to ide_setup_pci_device[s]()
  and use it to set ->host_priv.

* Set PCI dev's ->driver_data to point to the struct ide_host
  instance if PCI host driver wants to use ->host_priv.

* Rename ide_setup_pci_device[s]() to ide_pci_init_{one,two}().

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-probe.c        |  3 ++
 drivers/ide/pci/aec62xx.c      |  2 +-
 drivers/ide/pci/alim15x3.c     |  2 +-
 drivers/ide/pci/amd74xx.c      |  2 +-
 drivers/ide/pci/atiixp.c       |  2 +-
 drivers/ide/pci/cmd64x.c       |  2 +-
 drivers/ide/pci/cs5530.c       |  2 +-
 drivers/ide/pci/cs5535.c       |  2 +-
 drivers/ide/pci/cy82c693.c     |  2 +-
 drivers/ide/pci/generic.c      |  2 +-
 drivers/ide/pci/hpt34x.c       |  2 +-
 drivers/ide/pci/hpt366.c       |  4 +--
 drivers/ide/pci/it8213.c       |  2 +-
 drivers/ide/pci/it821x.c       |  2 +-
 drivers/ide/pci/jmicron.c      |  2 +-
 drivers/ide/pci/ns87415.c      |  2 +-
 drivers/ide/pci/opti621.c      |  2 +-
 drivers/ide/pci/pdc202xx_new.c |  4 +--
 drivers/ide/pci/pdc202xx_old.c |  2 +-
 drivers/ide/pci/piix.c         |  2 +-
 drivers/ide/pci/rz1000.c       |  2 +-
 drivers/ide/pci/sc1200.c       |  2 +-
 drivers/ide/pci/serverworks.c  |  2 +-
 drivers/ide/pci/siimage.c      |  2 +-
 drivers/ide/pci/sis5513.c      |  2 +-
 drivers/ide/pci/sl82c105.c     |  2 +-
 drivers/ide/pci/slc90e66.c     |  2 +-
 drivers/ide/pci/tc86c001.c     |  2 +-
 drivers/ide/pci/triflex.c      |  2 +-
 drivers/ide/pci/trm290.c       |  2 +-
 drivers/ide/pci/via82cxxx.c    |  2 +-
 drivers/ide/setup-pci.c        | 52 +++++++++++++++++++++++++++++-----
 include/linux/ide.h            |  7 +++--
 33 files changed, 85 insertions(+), 41 deletions(-)

diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 4aa76c453755..890c15b1b3ae 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1604,6 +1604,9 @@ struct ide_host *ide_host_alloc_all(const struct ide_port_info *d,
 		return NULL;
 	}
 
+	if (hws[0])
+		host->dev[0] = hws[0]->dev;
+
 	return host;
 }
 EXPORT_SYMBOL_GPL(ide_host_alloc_all);
diff --git a/drivers/ide/pci/aec62xx.c b/drivers/ide/pci/aec62xx.c
index fbc43e121e6b..7a5d246fe9b1 100644
--- a/drivers/ide/pci/aec62xx.c
+++ b/drivers/ide/pci/aec62xx.c
@@ -273,7 +273,7 @@ static int __devinit aec62xx_init_one(struct pci_dev *dev, const struct pci_devi
 		}
 	}
 
-	err = ide_setup_pci_device(dev, &d);
+	err = ide_pci_init_one(dev, &d, NULL);
 	if (err)
 		pci_disable_device(dev);
 
diff --git a/drivers/ide/pci/alim15x3.c b/drivers/ide/pci/alim15x3.c
index 5ef7817ac64f..7f96e7ca3864 100644
--- a/drivers/ide/pci/alim15x3.c
+++ b/drivers/ide/pci/alim15x3.c
@@ -565,7 +565,7 @@ static int __devinit alim15x3_init_one(struct pci_dev *dev, const struct pci_dev
 	if (idx == 0)
 		d.host_flags |= IDE_HFLAG_CLEAR_SIMPLEX;
 
-	return ide_setup_pci_device(dev, &d);
+	return ide_pci_init_one(dev, &d, NULL);
 }
 
 
diff --git a/drivers/ide/pci/amd74xx.c b/drivers/ide/pci/amd74xx.c
index ef7d971031ee..b6a475313c7c 100644
--- a/drivers/ide/pci/amd74xx.c
+++ b/drivers/ide/pci/amd74xx.c
@@ -302,7 +302,7 @@ static int __devinit amd74xx_probe(struct pci_dev *dev, const struct pci_device_
 			 d.name, pci_name(dev), dev->revision,
 			 amd_dma[fls(d.udma_mask) - 1]);
 
-	return ide_setup_pci_device(dev, &d);
+	return ide_pci_init_one(dev, &d, NULL);
 }
 
 static const struct pci_device_id amd74xx_pci_tbl[] = {
diff --git a/drivers/ide/pci/atiixp.c b/drivers/ide/pci/atiixp.c
index 8b637181681a..b483a68b39f6 100644
--- a/drivers/ide/pci/atiixp.c
+++ b/drivers/ide/pci/atiixp.c
@@ -167,7 +167,7 @@ static const struct ide_port_info atiixp_pci_info[] __devinitdata = {
 
 static int __devinit atiixp_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
-	return ide_setup_pci_device(dev, &atiixp_pci_info[id->driver_data]);
+	return ide_pci_init_one(dev, &atiixp_pci_info[id->driver_data], NULL);
 }
 
 static const struct pci_device_id atiixp_pci_tbl[] = {
diff --git a/drivers/ide/pci/cmd64x.c b/drivers/ide/pci/cmd64x.c
index ce58bfcdb3c6..fc0333c9a4e5 100644
--- a/drivers/ide/pci/cmd64x.c
+++ b/drivers/ide/pci/cmd64x.c
@@ -507,7 +507,7 @@ static int __devinit cmd64x_init_one(struct pci_dev *dev, const struct pci_devic
 		}
 	}
 
-	return ide_setup_pci_device(dev, &d);
+	return ide_pci_init_one(dev, &d, NULL);
 }
 
 static const struct pci_device_id cmd64x_pci_tbl[] = {
diff --git a/drivers/ide/pci/cs5530.c b/drivers/ide/pci/cs5530.c
index f5534c1ff349..ba82bad8bf4e 100644
--- a/drivers/ide/pci/cs5530.c
+++ b/drivers/ide/pci/cs5530.c
@@ -256,7 +256,7 @@ static const struct ide_port_info cs5530_chipset __devinitdata = {
 
 static int __devinit cs5530_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
-	return ide_setup_pci_device(dev, &cs5530_chipset);
+	return ide_pci_init_one(dev, &cs5530_chipset, NULL);
 }
 
 static const struct pci_device_id cs5530_pci_tbl[] = {
diff --git a/drivers/ide/pci/cs5535.c b/drivers/ide/pci/cs5535.c
index 5404fe4f701d..2161f43ca1b8 100644
--- a/drivers/ide/pci/cs5535.c
+++ b/drivers/ide/pci/cs5535.c
@@ -180,7 +180,7 @@ static const struct ide_port_info cs5535_chipset __devinitdata = {
 static int __devinit cs5535_init_one(struct pci_dev *dev,
 					const struct pci_device_id *id)
 {
-	return ide_setup_pci_device(dev, &cs5535_chipset);
+	return ide_pci_init_one(dev, &cs5535_chipset, NULL);
 }
 
 static const struct pci_device_id cs5535_pci_tbl[] = {
diff --git a/drivers/ide/pci/cy82c693.c b/drivers/ide/pci/cy82c693.c
index e14ad5530fa4..abd27ed7c30c 100644
--- a/drivers/ide/pci/cy82c693.c
+++ b/drivers/ide/pci/cy82c693.c
@@ -419,7 +419,7 @@ static int __devinit cy82c693_init_one(struct pci_dev *dev, const struct pci_dev
 	if ((dev->class >> 8) == PCI_CLASS_STORAGE_IDE &&
 	    PCI_FUNC(dev->devfn) == 1) {
 		dev2 = pci_get_slot(dev->bus, dev->devfn + 1);
-		ret = ide_setup_pci_devices(dev, dev2, &cy82c693_chipset);
+		ret = ide_pci_init_two(dev, dev2, &cy82c693_chipset, NULL);
 		/* We leak pci refs here but thats ok - we can't be unloaded */
 	}
 	return ret;
diff --git a/drivers/ide/pci/generic.c b/drivers/ide/pci/generic.c
index 041720e22762..dd0caea5e4f3 100644
--- a/drivers/ide/pci/generic.c
+++ b/drivers/ide/pci/generic.c
@@ -139,7 +139,7 @@ static int __devinit generic_init_one(struct pci_dev *dev, const struct pci_devi
 			goto out;
 		}
 	}
-	ret = ide_setup_pci_device(dev, d);
+	ret = ide_pci_init_one(dev, d, NULL);
 out:
 	return ret;
 }
diff --git a/drivers/ide/pci/hpt34x.c b/drivers/ide/pci/hpt34x.c
index 9e1d1c4741da..3d70c5150ac6 100644
--- a/drivers/ide/pci/hpt34x.c
+++ b/drivers/ide/pci/hpt34x.c
@@ -156,7 +156,7 @@ static int __devinit hpt34x_init_one(struct pci_dev *dev, const struct pci_devic
 
 	d = &hpt34x_chipsets[(pcicmd & PCI_COMMAND_MEMORY) ? 1 : 0];
 
-	return ide_setup_pci_device(dev, d);
+	return ide_pci_init_one(dev, d, NULL);
 }
 
 static const struct pci_device_id hpt34x_pci_tbl[] = {
diff --git a/drivers/ide/pci/hpt366.c b/drivers/ide/pci/hpt366.c
index 1f1135ce7cd6..b23b7a278005 100644
--- a/drivers/ide/pci/hpt366.c
+++ b/drivers/ide/pci/hpt366.c
@@ -1608,13 +1608,13 @@ static int __devinit hpt366_init_one(struct pci_dev *dev, const struct pci_devic
 				d.host_flags &= ~IDE_HFLAG_NON_BOOTABLE;
 		}
 
-		ret = ide_setup_pci_devices(dev, dev2, &d);
+		ret = ide_pci_init_two(dev, dev2, &d, NULL);
 		if (ret < 0)
 			pci_dev_put(dev2);
 		return ret;
 	}
 
-	return ide_setup_pci_device(dev, &d);
+	return ide_pci_init_one(dev, &d, NULL);
 }
 
 static const struct pci_device_id hpt366_pci_tbl[] __devinitconst = {
diff --git a/drivers/ide/pci/it8213.c b/drivers/ide/pci/it8213.c
index 2b71bdf74e73..18219fa9ef01 100644
--- a/drivers/ide/pci/it8213.c
+++ b/drivers/ide/pci/it8213.c
@@ -184,7 +184,7 @@ static const struct ide_port_info it8213_chipsets[] __devinitdata = {
 
 static int __devinit it8213_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
-	return ide_setup_pci_device(dev, &it8213_chipsets[id->driver_data]);
+	return ide_pci_init_one(dev, &it8213_chipsets[id->driver_data], NULL);
 }
 
 static const struct pci_device_id it8213_pci_tbl[] = {
diff --git a/drivers/ide/pci/it821x.c b/drivers/ide/pci/it821x.c
index cbf647202994..40186f9e56aa 100644
--- a/drivers/ide/pci/it821x.c
+++ b/drivers/ide/pci/it821x.c
@@ -664,7 +664,7 @@ static int __devinit it821x_init_one(struct pci_dev *dev, const struct pci_devic
 
 	pci_set_drvdata(dev, itdevs);
 
-	return ide_setup_pci_device(dev, &it821x_chipsets[id->driver_data]);
+	return ide_pci_init_one(dev, &it821x_chipsets[id->driver_data], NULL);
 }
 
 static const struct pci_device_id it821x_pci_tbl[] = {
diff --git a/drivers/ide/pci/jmicron.c b/drivers/ide/pci/jmicron.c
index 96ef7394f283..a7e3c14f7b07 100644
--- a/drivers/ide/pci/jmicron.c
+++ b/drivers/ide/pci/jmicron.c
@@ -121,7 +121,7 @@ static const struct ide_port_info jmicron_chipset __devinitdata = {
 
 static int __devinit jmicron_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
-	return ide_setup_pci_device(dev, &jmicron_chipset);
+	return ide_pci_init_one(dev, &jmicron_chipset, NULL);
 }
 
 /* All JMB PATA controllers have and will continue to have the same
diff --git a/drivers/ide/pci/ns87415.c b/drivers/ide/pci/ns87415.c
index 5cd2b32ff0ef..a45c33c0c792 100644
--- a/drivers/ide/pci/ns87415.c
+++ b/drivers/ide/pci/ns87415.c
@@ -324,7 +324,7 @@ static int __devinit ns87415_init_one(struct pci_dev *dev, const struct pci_devi
 		d.tp_ops = &superio_tp_ops;
 	}
 #endif
-	return ide_setup_pci_device(dev, &d);
+	return ide_pci_init_one(dev, &d, NULL);
 }
 
 static const struct pci_device_id ns87415_pci_tbl[] = {
diff --git a/drivers/ide/pci/opti621.c b/drivers/ide/pci/opti621.c
index 725c80508d90..edb9132ffbe4 100644
--- a/drivers/ide/pci/opti621.c
+++ b/drivers/ide/pci/opti621.c
@@ -209,7 +209,7 @@ static const struct ide_port_info opti621_chipset __devinitdata = {
 
 static int __devinit opti621_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
-	return ide_setup_pci_device(dev, &opti621_chipset);
+	return ide_pci_init_one(dev, &opti621_chipset, NULL);
 }
 
 static const struct pci_device_id opti621_pci_tbl[] = {
diff --git a/drivers/ide/pci/pdc202xx_new.c b/drivers/ide/pci/pdc202xx_new.c
index 070df8ab3b21..71a420feb981 100644
--- a/drivers/ide/pci/pdc202xx_new.c
+++ b/drivers/ide/pci/pdc202xx_new.c
@@ -524,7 +524,7 @@ static int __devinit pdc202new_init_one(struct pci_dev *dev, const struct pci_de
 		dev2 = pdc20270_get_dev2(dev);
 
 		if (dev2) {
-			int ret = ide_setup_pci_devices(dev, dev2, d);
+			int ret = ide_pci_init_two(dev, dev2, d, NULL);
 			if (ret < 0)
 				pci_dev_put(dev2);
 			return ret;
@@ -540,7 +540,7 @@ static int __devinit pdc202new_init_one(struct pci_dev *dev, const struct pci_de
 		return -ENODEV;
 	}
 
-	return ide_setup_pci_device(dev, d);
+	return ide_pci_init_one(dev, d, NULL);
 }
 
 static const struct pci_device_id pdc202new_pci_tbl[] = {
diff --git a/drivers/ide/pci/pdc202xx_old.c b/drivers/ide/pci/pdc202xx_old.c
index e54dc653b8c4..eba1d60a73a0 100644
--- a/drivers/ide/pci/pdc202xx_old.c
+++ b/drivers/ide/pci/pdc202xx_old.c
@@ -412,7 +412,7 @@ static int __devinit pdc202xx_init_one(struct pci_dev *dev, const struct pci_dev
 		}
 	}
 
-	return ide_setup_pci_device(dev, d);
+	return ide_pci_init_one(dev, d, NULL);
 }
 
 static const struct pci_device_id pdc202xx_pci_tbl[] = {
diff --git a/drivers/ide/pci/piix.c b/drivers/ide/pci/piix.c
index 0ce41b4dddaf..359f65ddcbf9 100644
--- a/drivers/ide/pci/piix.c
+++ b/drivers/ide/pci/piix.c
@@ -394,7 +394,7 @@ static const struct ide_port_info piix_pci_info[] __devinitdata = {
  
 static int __devinit piix_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
-	return ide_setup_pci_device(dev, &piix_pci_info[id->driver_data]);
+	return ide_pci_init_one(dev, &piix_pci_info[id->driver_data], NULL);
 }
 
 /**
diff --git a/drivers/ide/pci/rz1000.c b/drivers/ide/pci/rz1000.c
index 532154adba29..860ffdeca095 100644
--- a/drivers/ide/pci/rz1000.c
+++ b/drivers/ide/pci/rz1000.c
@@ -48,7 +48,7 @@ static const struct ide_port_info rz1000_chipset __devinitdata = {
 
 static int __devinit rz1000_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
-	return ide_setup_pci_device(dev, &rz1000_chipset);
+	return ide_pci_init_one(dev, &rz1000_chipset, NULL);
 }
 
 static const struct pci_device_id rz1000_pci_tbl[] = {
diff --git a/drivers/ide/pci/sc1200.c b/drivers/ide/pci/sc1200.c
index 14c787b5d95f..8fd9cc2119d6 100644
--- a/drivers/ide/pci/sc1200.c
+++ b/drivers/ide/pci/sc1200.c
@@ -317,7 +317,7 @@ static const struct ide_port_info sc1200_chipset __devinitdata = {
 
 static int __devinit sc1200_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
-	return ide_setup_pci_device(dev, &sc1200_chipset);
+	return ide_pci_init_one(dev, &sc1200_chipset, NULL);
 }
 
 static const struct pci_device_id sc1200_pci_tbl[] = {
diff --git a/drivers/ide/pci/serverworks.c b/drivers/ide/pci/serverworks.c
index 127ccb45e261..34abdfc8d567 100644
--- a/drivers/ide/pci/serverworks.c
+++ b/drivers/ide/pci/serverworks.c
@@ -422,7 +422,7 @@ static int __devinit svwks_init_one(struct pci_dev *dev, const struct pci_device
 			d.host_flags &= ~IDE_HFLAG_SINGLE;
 	}
 
-	return ide_setup_pci_device(dev, &d);
+	return ide_pci_init_one(dev, &d, NULL);
 }
 
 static const struct pci_device_id svwks_pci_tbl[] = {
diff --git a/drivers/ide/pci/siimage.c b/drivers/ide/pci/siimage.c
index 5965a35d94ae..48124133601a 100644
--- a/drivers/ide/pci/siimage.c
+++ b/drivers/ide/pci/siimage.c
@@ -795,7 +795,7 @@ static int __devinit siimage_init_one(struct pci_dev *dev,
 		d.host_flags |= IDE_HFLAG_NO_ATAPI_DMA;
 	}
 
-	return ide_setup_pci_device(dev, &d);
+	return ide_pci_init_one(dev, &d, NULL);
 }
 
 static const struct pci_device_id siimage_pci_tbl[] = {
diff --git a/drivers/ide/pci/sis5513.c b/drivers/ide/pci/sis5513.c
index 2389945ca95d..a2330c4ac75b 100644
--- a/drivers/ide/pci/sis5513.c
+++ b/drivers/ide/pci/sis5513.c
@@ -583,7 +583,7 @@ static int __devinit sis5513_init_one(struct pci_dev *dev, const struct pci_devi
 
 	d.udma_mask = udma_rates[chipset_family];
 
-	return ide_setup_pci_device(dev, &d);
+	return ide_pci_init_one(dev, &d, NULL);
 }
 
 static const struct pci_device_id sis5513_pci_tbl[] = {
diff --git a/drivers/ide/pci/sl82c105.c b/drivers/ide/pci/sl82c105.c
index f82a6502c1b7..be22f8125d71 100644
--- a/drivers/ide/pci/sl82c105.c
+++ b/drivers/ide/pci/sl82c105.c
@@ -335,7 +335,7 @@ static int __devinit sl82c105_init_one(struct pci_dev *dev, const struct pci_dev
 		d.host_flags &= ~IDE_HFLAG_SERIALIZE_DMA;
 	}
 
-	return ide_setup_pci_device(dev, &d);
+	return ide_pci_init_one(dev, &d, NULL);
 }
 
 static const struct pci_device_id sl82c105_pci_tbl[] = {
diff --git a/drivers/ide/pci/slc90e66.c b/drivers/ide/pci/slc90e66.c
index dae6e2c94d86..2fc2f2cf2206 100644
--- a/drivers/ide/pci/slc90e66.c
+++ b/drivers/ide/pci/slc90e66.c
@@ -144,7 +144,7 @@ static const struct ide_port_info slc90e66_chipset __devinitdata = {
 
 static int __devinit slc90e66_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
-	return ide_setup_pci_device(dev, &slc90e66_chipset);
+	return ide_pci_init_one(dev, &slc90e66_chipset, NULL);
 }
 
 static const struct pci_device_id slc90e66_pci_tbl[] = {
diff --git a/drivers/ide/pci/tc86c001.c b/drivers/ide/pci/tc86c001.c
index 477e19790102..e16e79d21772 100644
--- a/drivers/ide/pci/tc86c001.c
+++ b/drivers/ide/pci/tc86c001.c
@@ -215,7 +215,7 @@ static const struct ide_port_info tc86c001_chipset __devinitdata = {
 static int __devinit tc86c001_init_one(struct pci_dev *dev,
 				       const struct pci_device_id *id)
 {
-	return ide_setup_pci_device(dev, &tc86c001_chipset);
+	return ide_pci_init_one(dev, &tc86c001_chipset, NULL);
 }
 
 static const struct pci_device_id tc86c001_pci_tbl[] = {
diff --git a/drivers/ide/pci/triflex.c b/drivers/ide/pci/triflex.c
index db65a558d4ec..60dcb645d1b0 100644
--- a/drivers/ide/pci/triflex.c
+++ b/drivers/ide/pci/triflex.c
@@ -104,7 +104,7 @@ static const struct ide_port_info triflex_device __devinitdata = {
 static int __devinit triflex_init_one(struct pci_dev *dev, 
 		const struct pci_device_id *id)
 {
-	return ide_setup_pci_device(dev, &triflex_device);
+	return ide_pci_init_one(dev, &triflex_device, NULL);
 }
 
 static const struct pci_device_id triflex_pci_tbl[] = {
diff --git a/drivers/ide/pci/trm290.c b/drivers/ide/pci/trm290.c
index a8a3138682ef..d8127b51a542 100644
--- a/drivers/ide/pci/trm290.c
+++ b/drivers/ide/pci/trm290.c
@@ -340,7 +340,7 @@ static const struct ide_port_info trm290_chipset __devinitdata = {
 
 static int __devinit trm290_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
-	return ide_setup_pci_device(dev, &trm290_chipset);
+	return ide_pci_init_one(dev, &trm290_chipset, NULL);
 }
 
 static const struct pci_device_id trm290_pci_tbl[] = {
diff --git a/drivers/ide/pci/via82cxxx.c b/drivers/ide/pci/via82cxxx.c
index 09dc4803ef9d..2f22abfe003b 100644
--- a/drivers/ide/pci/via82cxxx.c
+++ b/drivers/ide/pci/via82cxxx.c
@@ -466,7 +466,7 @@ static int __devinit via_init_one(struct pci_dev *dev, const struct pci_device_i
 
 	d.udma_mask = via_config->udma_mask;
 
-	return ide_setup_pci_device(dev, &d);
+	return ide_pci_init_one(dev, &d, NULL);
 }
 
 static const struct pci_device_id via_pci_tbl[] = {
diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
index b85de71fdc88..ca17bf8896df 100644
--- a/drivers/ide/setup-pci.c
+++ b/drivers/ide/setup-pci.c
@@ -525,8 +525,10 @@ out:
 	return ret;
 }
 
-int ide_setup_pci_device(struct pci_dev *dev, const struct ide_port_info *d)
+int ide_pci_init_one(struct pci_dev *dev, const struct ide_port_info *d,
+		     void *priv)
 {
+	struct ide_host *host;
 	hw_regs_t hw[4], *hws[] = { NULL, NULL, NULL, NULL };
 	int ret;
 
@@ -536,6 +538,19 @@ int ide_setup_pci_device(struct pci_dev *dev, const struct ide_port_info *d)
 
 	ide_pci_setup_ports(dev, d, 0, &hw[0], &hws[0]);
 
+	host = ide_host_alloc(d, hws);
+	if (host == NULL) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	host->dev[0] = &dev->dev;
+
+	host->host_priv = priv;
+
+	if (priv)
+		pci_set_drvdata(dev, host);
+
 	ret = do_ide_setup_pci_device(dev, d, 1);
 	if (ret < 0)
 		goto out;
@@ -543,16 +558,19 @@ int ide_setup_pci_device(struct pci_dev *dev, const struct ide_port_info *d)
 	/* fixup IRQ */
 	hw[1].irq = hw[0].irq = ret;
 
-	ret = ide_host_add(d, hws, NULL);
+	ret = ide_host_register(host, d, hws);
+	if (ret)
+		ide_host_free(host);
 out:
 	return ret;
 }
-EXPORT_SYMBOL_GPL(ide_setup_pci_device);
+EXPORT_SYMBOL_GPL(ide_pci_init_one);
 
-int ide_setup_pci_devices(struct pci_dev *dev1, struct pci_dev *dev2,
-			  const struct ide_port_info *d)
+int ide_pci_init_two(struct pci_dev *dev1, struct pci_dev *dev2,
+		     const struct ide_port_info *d, void *priv)
 {
 	struct pci_dev *pdev[] = { dev1, dev2 };
+	struct ide_host *host;
 	int ret, i;
 	hw_regs_t hw[4], *hws[] = { NULL, NULL, NULL, NULL };
 
@@ -562,7 +580,25 @@ int ide_setup_pci_devices(struct pci_dev *dev1, struct pci_dev *dev2,
 			goto out;
 
 		ide_pci_setup_ports(pdev[i], d, 0, &hw[i*2], &hws[i*2]);
+	}
 
+	host = ide_host_alloc(d, hws);
+	if (host == NULL) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	host->dev[0] = &dev1->dev;
+	host->dev[1] = &dev2->dev;
+
+	host->host_priv = priv;
+
+	if (priv) {
+		pci_set_drvdata(pdev[0], host);
+		pci_set_drvdata(pdev[1], host);
+	}
+
+	for (i = 0; i < 2; i++) {
 		ret = do_ide_setup_pci_device(pdev[i], d, !i);
 
 		/*
@@ -576,8 +612,10 @@ int ide_setup_pci_devices(struct pci_dev *dev1, struct pci_dev *dev2,
 		hw[i*2 + 1].irq = hw[i*2].irq = ret;
 	}
 
-	ret = ide_host_add(d, hws, NULL);
+	ret = ide_host_register(host, d, hws);
+	if (ret)
+		ide_host_free(host);
 out:
 	return ret;
 }
-EXPORT_SYMBOL_GPL(ide_setup_pci_devices);
+EXPORT_SYMBOL_GPL(ide_pci_init_two);
diff --git a/include/linux/ide.h b/include/linux/ide.h
index d67ccca2b964..776c574c9640 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -626,6 +626,8 @@ typedef struct hwif_s {
 struct ide_host {
 	ide_hwif_t	*ports[MAX_HWIFS];
 	unsigned int	n_ports;
+	struct device	*dev[2];
+	void		*host_priv;
 };
 
 /*
@@ -1201,8 +1203,9 @@ struct ide_port_info {
 	u8			udma_mask;
 };
 
-int ide_setup_pci_device(struct pci_dev *, const struct ide_port_info *);
-int ide_setup_pci_devices(struct pci_dev *, struct pci_dev *, const struct ide_port_info *);
+int ide_pci_init_one(struct pci_dev *, const struct ide_port_info *, void *);
+int ide_pci_init_two(struct pci_dev *, struct pci_dev *,
+		     const struct ide_port_info *, void *);
 
 void ide_map_sg(ide_drive_t *, struct request *);
 void ide_init_sg_cmd(ide_drive_t *, struct request *);

From 08da591e14cf87247ec09b17c350235157a92fc3 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:15 +0200
Subject: [PATCH 0531/1171] ide: add ide_device_{get,put}() helpers

* Add 'struct ide_host *host' field to ide_hwif_t and set it
  in ide_host_alloc_all().

* Add ide_device_{get,put}() helpers loosely based on SCSI's
  scsi_device_{get,put}() ones.

* Convert IDE device drivers to use ide_device_{get,put}().

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-cd.c     | 12 +++++++---
 drivers/ide/ide-disk.c   | 12 +++++++---
 drivers/ide/ide-floppy.c | 12 +++++++---
 drivers/ide/ide-probe.c  |  2 ++
 drivers/ide/ide-tape.c   | 12 +++++++---
 drivers/ide/ide.c        | 47 ++++++++++++++++++++++++++++++++++++++++
 drivers/scsi/ide-scsi.c  |  8 ++++++-
 include/linux/ide.h      |  7 ++++++
 8 files changed, 99 insertions(+), 13 deletions(-)

diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index 4e73aeee4053..8f253e5f26a8 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -57,23 +57,29 @@ static DEFINE_MUTEX(idecd_ref_mutex);
 #define ide_cd_g(disk) \
 	container_of((disk)->private_data, struct cdrom_info, driver)
 
+static void ide_cd_release(struct kref *);
+
 static struct cdrom_info *ide_cd_get(struct gendisk *disk)
 {
 	struct cdrom_info *cd = NULL;
 
 	mutex_lock(&idecd_ref_mutex);
 	cd = ide_cd_g(disk);
-	if (cd)
+	if (cd) {
 		kref_get(&cd->kref);
+		if (ide_device_get(cd->drive)) {
+			kref_put(&cd->kref, ide_cd_release);
+			cd = NULL;
+		}
+	}
 	mutex_unlock(&idecd_ref_mutex);
 	return cd;
 }
 
-static void ide_cd_release(struct kref *);
-
 static void ide_cd_put(struct cdrom_info *cd)
 {
 	mutex_lock(&idecd_ref_mutex);
+	ide_device_put(cd->drive);
 	kref_put(&cd->kref, ide_cd_release);
 	mutex_unlock(&idecd_ref_mutex);
 }
diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index df5fe5756871..28d85b410f7c 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -56,23 +56,29 @@ static DEFINE_MUTEX(idedisk_ref_mutex);
 #define ide_disk_g(disk) \
 	container_of((disk)->private_data, struct ide_disk_obj, driver)
 
+static void ide_disk_release(struct kref *);
+
 static struct ide_disk_obj *ide_disk_get(struct gendisk *disk)
 {
 	struct ide_disk_obj *idkp = NULL;
 
 	mutex_lock(&idedisk_ref_mutex);
 	idkp = ide_disk_g(disk);
-	if (idkp)
+	if (idkp) {
 		kref_get(&idkp->kref);
+		if (ide_device_get(idkp->drive)) {
+			kref_put(&idkp->kref, ide_disk_release);
+			idkp = NULL;
+		}
+	}
 	mutex_unlock(&idedisk_ref_mutex);
 	return idkp;
 }
 
-static void ide_disk_release(struct kref *);
-
 static void ide_disk_put(struct ide_disk_obj *idkp)
 {
 	mutex_lock(&idedisk_ref_mutex);
+	ide_device_put(idkp->drive);
 	kref_put(&idkp->kref, ide_disk_release);
 	mutex_unlock(&idedisk_ref_mutex);
 }
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 3d8e6dd0f41e..ca11a26746f1 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -158,23 +158,29 @@ static DEFINE_MUTEX(idefloppy_ref_mutex);
 #define ide_floppy_g(disk) \
 	container_of((disk)->private_data, struct ide_floppy_obj, driver)
 
+static void idefloppy_cleanup_obj(struct kref *);
+
 static struct ide_floppy_obj *ide_floppy_get(struct gendisk *disk)
 {
 	struct ide_floppy_obj *floppy = NULL;
 
 	mutex_lock(&idefloppy_ref_mutex);
 	floppy = ide_floppy_g(disk);
-	if (floppy)
+	if (floppy) {
 		kref_get(&floppy->kref);
+		if (ide_device_get(floppy->drive)) {
+			kref_put(&floppy->kref, idefloppy_cleanup_obj);
+			floppy = NULL;
+		}
+	}
 	mutex_unlock(&idefloppy_ref_mutex);
 	return floppy;
 }
 
-static void idefloppy_cleanup_obj(struct kref *);
-
 static void ide_floppy_put(struct ide_floppy_obj *floppy)
 {
 	mutex_lock(&idefloppy_ref_mutex);
+	ide_device_put(floppy->drive);
 	kref_put(&floppy->kref, idefloppy_cleanup_obj);
 	mutex_unlock(&idefloppy_ref_mutex);
 }
diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 890c15b1b3ae..9ab5892eaea1 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1595,6 +1595,8 @@ struct ide_host *ide_host_alloc_all(const struct ide_port_info *d,
 
 		ide_init_port_data(hwif, idx);
 
+		hwif->host = host;
+
 		host->ports[i] = hwif;
 		host->n_ports++;
 	}
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index 6962ca4891a1..789f3428f072 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -322,23 +322,29 @@ static struct class *idetape_sysfs_class;
 #define ide_tape_g(disk) \
 	container_of((disk)->private_data, struct ide_tape_obj, driver)
 
+static void ide_tape_release(struct kref *);
+
 static struct ide_tape_obj *ide_tape_get(struct gendisk *disk)
 {
 	struct ide_tape_obj *tape = NULL;
 
 	mutex_lock(&idetape_ref_mutex);
 	tape = ide_tape_g(disk);
-	if (tape)
+	if (tape) {
 		kref_get(&tape->kref);
+		if (ide_device_get(tape->drive)) {
+			kref_put(&tape->kref, ide_tape_release);
+			tape = NULL;
+		}
+	}
 	mutex_unlock(&idetape_ref_mutex);
 	return tape;
 }
 
-static void ide_tape_release(struct kref *);
-
 static void ide_tape_put(struct ide_tape_obj *tape)
 {
 	mutex_lock(&idetape_ref_mutex);
+	ide_device_put(tape->drive);
 	kref_put(&tape->kref, ide_tape_release);
 	mutex_unlock(&idetape_ref_mutex);
 }
diff --git a/drivers/ide/ide.c b/drivers/ide/ide.c
index 60f0ca66aa93..772451600e4d 100644
--- a/drivers/ide/ide.c
+++ b/drivers/ide/ide.c
@@ -618,6 +618,53 @@ set_val:
 
 EXPORT_SYMBOL(generic_ide_ioctl);
 
+/**
+ * ide_device_get	-	get an additional reference to a ide_drive_t
+ * @drive:	device to get a reference to
+ *
+ * Gets a reference to the ide_drive_t and increments the use count of the
+ * underlying LLDD module.
+ */
+int ide_device_get(ide_drive_t *drive)
+{
+	struct device *host_dev;
+	struct module *module;
+
+	if (!get_device(&drive->gendev))
+		return -ENXIO;
+
+	host_dev = drive->hwif->host->dev[0];
+	module = host_dev ? host_dev->driver->owner : NULL;
+
+	if (module && !try_module_get(module)) {
+		put_device(&drive->gendev);
+		return -ENXIO;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ide_device_get);
+
+/**
+ * ide_device_put	-	release a reference to a ide_drive_t
+ * @drive:	device to release a reference on
+ *
+ * Release a reference to the ide_drive_t and decrements the use count of
+ * the underlying LLDD module.
+ */
+void ide_device_put(ide_drive_t *drive)
+{
+#ifdef CONFIG_MODULE_UNLOAD
+	struct device *host_dev = drive->hwif->host->dev[0];
+	struct module *module = host_dev ? host_dev->driver->owner : NULL;
+
+	if (module)
+		module_put(module);
+#endif
+	put_device(&drive->gendev);
+}
+EXPORT_SYMBOL_GPL(ide_device_put);
+
 static int ide_bus_match(struct device *dev, struct device_driver *drv)
 {
 	return 1;
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index 538552495d48..318ef382448f 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -101,8 +101,13 @@ static struct ide_scsi_obj *ide_scsi_get(struct gendisk *disk)
 
 	mutex_lock(&idescsi_ref_mutex);
 	scsi = ide_scsi_g(disk);
-	if (scsi)
+	if (scsi) {
 		scsi_host_get(scsi->host);
+		if (ide_device_get(scsi->drive)) {
+			scsi_host_put(scsi->host);
+			scsi = NULL;
+		}
+	}
 	mutex_unlock(&idescsi_ref_mutex);
 	return scsi;
 }
@@ -110,6 +115,7 @@ static struct ide_scsi_obj *ide_scsi_get(struct gendisk *disk)
 static void ide_scsi_put(struct ide_scsi_obj *scsi)
 {
 	mutex_lock(&idescsi_ref_mutex);
+	ide_device_put(scsi->drive);
 	scsi_host_put(scsi->host);
 	mutex_unlock(&idescsi_ref_mutex);
 }
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 776c574c9640..3eccac0a2a36 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -532,12 +532,16 @@ struct ide_dma_ops {
 	void	(*dma_timeout)(struct ide_drive_s *);
 };
 
+struct ide_host;
+
 typedef struct hwif_s {
 	struct hwif_s *next;		/* for linked-list in ide_hwgroup_t */
 	struct hwif_s *mate;		/* other hwif from same PCI chip */
 	struct hwgroup_s *hwgroup;	/* actually (ide_hwgroup_t *) */
 	struct proc_dir_entry *proc;	/* /proc/ide/ directory entry */
 
+	struct ide_host *host;
+
 	char name[6];			/* name of interface, eg. "ide0" */
 
 	struct ide_io_ports	io_ports;
@@ -876,6 +880,9 @@ struct ide_driver_s {
 
 #define to_ide_driver(drv) container_of(drv, ide_driver_t, gen_driver)
 
+int ide_device_get(ide_drive_t *);
+void ide_device_put(ide_drive_t *);
+
 int generic_ide_ioctl(ide_drive_t *, struct file *, struct block_device *, unsigned, unsigned long);
 
 extern int ide_vlb_clk;

From 60e57ed7c12917932a01d1679d92a7a8735afbce Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:15 +0200
Subject: [PATCH 0532/1171] aec62xx: convert to use ->host_priv

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/aec62xx.c | 32 ++++++++++++++++----------------
 1 file changed, 16 insertions(+), 16 deletions(-)

diff --git a/drivers/ide/pci/aec62xx.c b/drivers/ide/pci/aec62xx.c
index 7a5d246fe9b1..7ca7989bc35d 100644
--- a/drivers/ide/pci/aec62xx.c
+++ b/drivers/ide/pci/aec62xx.c
@@ -59,10 +59,6 @@ static const struct chipset_bus_clock_list_entry aec6xxx_34_base [] = {
 	{	0,		0x00,	0x00	}
 };
 
-#define BUSCLOCK(D)	\
-	((struct chipset_bus_clock_list_entry *) pci_get_drvdata((D)))
-
-
 /*
  * TO DO: active tuning and correction of cards without a bios.
  */
@@ -88,6 +84,8 @@ static void aec6210_set_mode(ide_drive_t *drive, const u8 speed)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
+	struct ide_host *host	= pci_get_drvdata(dev);
+	struct chipset_bus_clock_list_entry *bus_clock = host->host_priv;
 	u16 d_conf		= 0;
 	u8 ultra = 0, ultra_conf = 0;
 	u8 tmp0 = 0, tmp1 = 0, tmp2 = 0;
@@ -96,7 +94,7 @@ static void aec6210_set_mode(ide_drive_t *drive, const u8 speed)
 	local_irq_save(flags);
 	/* 0x40|(2*drive->dn): Active, 0x41|(2*drive->dn): Recovery */
 	pci_read_config_word(dev, 0x40|(2*drive->dn), &d_conf);
-	tmp0 = pci_bus_clock_list(speed, BUSCLOCK(dev));
+	tmp0 = pci_bus_clock_list(speed, bus_clock);
 	d_conf = ((tmp0 & 0xf0) << 4) | (tmp0 & 0xf);
 	pci_write_config_word(dev, 0x40|(2*drive->dn), d_conf);
 
@@ -104,7 +102,7 @@ static void aec6210_set_mode(ide_drive_t *drive, const u8 speed)
 	tmp2 = 0x00;
 	pci_read_config_byte(dev, 0x54, &ultra);
 	tmp1 = ((0x00 << (2*drive->dn)) | (ultra & ~(3 << (2*drive->dn))));
-	ultra_conf = pci_bus_clock_list_ultra(speed, BUSCLOCK(dev));
+	ultra_conf = pci_bus_clock_list_ultra(speed, bus_clock);
 	tmp2 = ((ultra_conf << (2*drive->dn)) | (tmp1 & ~(3 << (2*drive->dn))));
 	pci_write_config_byte(dev, 0x54, tmp2);
 	local_irq_restore(flags);
@@ -114,6 +112,8 @@ static void aec6260_set_mode(ide_drive_t *drive, const u8 speed)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
+	struct ide_host *host	= pci_get_drvdata(dev);
+	struct chipset_bus_clock_list_entry *bus_clock = host->host_priv;
 	u8 unit		= (drive->select.b.unit & 0x01);
 	u8 tmp1 = 0, tmp2 = 0;
 	u8 ultra = 0, drive_conf = 0, ultra_conf = 0;
@@ -122,12 +122,12 @@ static void aec6260_set_mode(ide_drive_t *drive, const u8 speed)
 	local_irq_save(flags);
 	/* high 4-bits: Active, low 4-bits: Recovery */
 	pci_read_config_byte(dev, 0x40|drive->dn, &drive_conf);
-	drive_conf = pci_bus_clock_list(speed, BUSCLOCK(dev));
+	drive_conf = pci_bus_clock_list(speed, bus_clock);
 	pci_write_config_byte(dev, 0x40|drive->dn, drive_conf);
 
 	pci_read_config_byte(dev, (0x44|hwif->channel), &ultra);
 	tmp1 = ((0x00 << (4*unit)) | (ultra & ~(7 << (4*unit))));
-	ultra_conf = pci_bus_clock_list_ultra(speed, BUSCLOCK(dev));
+	ultra_conf = pci_bus_clock_list_ultra(speed, bus_clock);
 	tmp2 = ((ultra_conf << (4*unit)) | (tmp1 & ~(7 << (4*unit))));
 	pci_write_config_byte(dev, (0x44|hwif->channel), tmp2);
 	local_irq_restore(flags);
@@ -140,13 +140,6 @@ static void aec_set_pio_mode(ide_drive_t *drive, const u8 pio)
 
 static unsigned int __devinit init_chipset_aec62xx(struct pci_dev *dev, const char *name)
 {
-	int bus_speed = ide_pci_clk ? ide_pci_clk : 33;
-
-	if (bus_speed <= 33)
-		pci_set_drvdata(dev, (void *) aec6xxx_33_base);
-	else
-		pci_set_drvdata(dev, (void *) aec6xxx_34_base);
-
 	/* These are necessary to get AEC6280 Macintosh cards to work */
 	if ((dev->device == PCI_DEVICE_ID_ARTOP_ATP865) ||
 	    (dev->device == PCI_DEVICE_ID_ARTOP_ATP865R)) {
@@ -254,10 +247,17 @@ static const struct ide_port_info aec62xx_chipsets[] __devinitdata = {
 
 static int __devinit aec62xx_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
+	const struct chipset_bus_clock_list_entry *bus_clock;
 	struct ide_port_info d;
 	u8 idx = id->driver_data;
+	int bus_speed = ide_pci_clk ? ide_pci_clk : 33;
 	int err;
 
+	if (bus_speed <= 33)
+		bus_clock = aec6xxx_33_base;
+	else
+		bus_clock = aec6xxx_34_base;
+
 	err = pci_enable_device(dev);
 	if (err)
 		return err;
@@ -273,7 +273,7 @@ static int __devinit aec62xx_init_one(struct pci_dev *dev, const struct pci_devi
 		}
 	}
 
-	err = ide_pci_init_one(dev, &d, NULL);
+	err = ide_pci_init_one(dev, &d, (void *)bus_clock);
 	if (err)
 		pci_disable_device(dev);
 

From 74811f355f4f69a187fa74892dcf2a684b84ce99 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:15 +0200
Subject: [PATCH 0533/1171] hpt366: convert to use ->host_priv

While at it:

* Allocate both struct hpt_info instances at once.

Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/hpt366.c | 72 +++++++++++++++++++++++-----------------
 1 file changed, 41 insertions(+), 31 deletions(-)

diff --git a/drivers/ide/pci/hpt366.c b/drivers/ide/pci/hpt366.c
index b23b7a278005..8f29571345a2 100644
--- a/drivers/ide/pci/hpt366.c
+++ b/drivers/ide/pci/hpt366.c
@@ -620,7 +620,8 @@ static u8 hpt3xx_udma_filter(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
-	struct hpt_info *info	= pci_get_drvdata(dev);
+	struct ide_host *host	= pci_get_drvdata(dev);
+	struct hpt_info *info	= host->host_priv + (hwif->dev == host->dev[1]);
 	u8 mask 		= hwif->ultra_mask;
 
 	switch (info->chip_type) {
@@ -660,7 +661,8 @@ static u8 hpt3xx_mdma_filter(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
-	struct hpt_info *info	= pci_get_drvdata(dev);
+	struct ide_host *host	= pci_get_drvdata(dev);
+	struct hpt_info *info	= host->host_priv + (hwif->dev == host->dev[1]);
 
 	switch (info->chip_type) {
 	case HPT372 :
@@ -694,8 +696,10 @@ static u32 get_speed_setting(u8 speed, struct hpt_info *info)
 
 static void hpt3xx_set_mode(ide_drive_t *drive, const u8 speed)
 {
-	struct pci_dev  *dev	= to_pci_dev(drive->hwif->dev);
-	struct hpt_info	*info	= pci_get_drvdata(dev);
+	ide_hwif_t *hwif	= drive->hwif;
+	struct pci_dev *dev	= to_pci_dev(hwif->dev);
+	struct ide_host *host	= pci_get_drvdata(dev);
+	struct hpt_info *info	= host->host_priv + (hwif->dev == host->dev[1]);
 	struct hpt_timings *t	= info->timings;
 	u8  itr_addr		= 0x40 + (drive->dn * 4);
 	u32 old_itr		= 0;
@@ -738,7 +742,8 @@ static void hpt3xx_maskproc(ide_drive_t *drive, int mask)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
 	struct pci_dev	*dev	= to_pci_dev(hwif->dev);
-	struct hpt_info *info	= pci_get_drvdata(dev);
+	struct ide_host *host	= pci_get_drvdata(dev);
+	struct hpt_info *info	= host->host_priv + (hwif->dev == host->dev[1]);
 
 	if (drive->quirk_list) {
 		if (info->chip_type >= HPT370) {
@@ -965,22 +970,13 @@ static int __devinit hpt37x_calibrate_dpll(struct pci_dev *dev, u16 f_low, u16 f
 
 static unsigned int __devinit init_chipset_hpt366(struct pci_dev *dev, const char *name)
 {
-	struct hpt_info *info	= kmalloc(sizeof(struct hpt_info), GFP_KERNEL);
 	unsigned long io_base	= pci_resource_start(dev, 4);
+	struct ide_host *host	= pci_get_drvdata(dev);
+	struct hpt_info *info	= host->host_priv + (&dev->dev == host->dev[1]);
 	u8 pci_clk,  dpll_clk	= 0;	/* PCI and DPLL clock in MHz */
 	u8 chip_type;
 	enum ata_clock	clock;
 
-	if (info == NULL) {
-		printk(KERN_ERR "%s: out of memory!\n", name);
-		return -ENOMEM;
-	}
-
-	/*
-	 * Copy everything from a static "template" structure
-	 * to just allocated per-chip hpt_info structure.
-	 */
-	memcpy(info, pci_get_drvdata(dev), sizeof(struct hpt_info));
 	chip_type = info->chip_type;
 
 	pci_write_config_byte(dev, PCI_CACHE_LINE_SIZE, (L1_CACHE_BYTES / 4));
@@ -1142,7 +1138,6 @@ static unsigned int __devinit init_chipset_hpt366(struct pci_dev *dev, const cha
 
 		if (info->timings->clock_table[clock] == NULL) {
 			printk(KERN_ERR "%s: unknown bus timing!\n", name);
-			kfree(info);
 			return -EIO;
 		}
 
@@ -1169,7 +1164,6 @@ static unsigned int __devinit init_chipset_hpt366(struct pci_dev *dev, const cha
 		}
 		if (adjust == 8) {
 			printk(KERN_ERR "%s: DPLL did not stabilize!\n", name);
-			kfree(info);
 			return -EIO;
 		}
 
@@ -1186,9 +1180,6 @@ static unsigned int __devinit init_chipset_hpt366(struct pci_dev *dev, const cha
 	info->pci_clk	= pci_clk;
 	info->clock	= clock;
 
-	/* Point to this chip's own instance of the hpt_info structure. */
-	pci_set_drvdata(dev, info);
-
 	if (chip_type >= HPT370) {
 		u8  mcr1, mcr4;
 
@@ -1218,7 +1209,8 @@ static unsigned int __devinit init_chipset_hpt366(struct pci_dev *dev, const cha
 static u8 __devinit hpt3xx_cable_detect(ide_hwif_t *hwif)
 {
 	struct pci_dev	*dev	= to_pci_dev(hwif->dev);
-	struct hpt_info *info	= pci_get_drvdata(dev);
+	struct ide_host *host	= pci_get_drvdata(dev);
+	struct hpt_info *info	= host->host_priv + (hwif->dev == host->dev[1]);
 	u8 chip_type		= info->chip_type;
 	u8 scr1 = 0, ata66	= hwif->channel ? 0x01 : 0x02;
 
@@ -1262,7 +1254,8 @@ static u8 __devinit hpt3xx_cable_detect(ide_hwif_t *hwif)
 static void __devinit init_hwif_hpt366(ide_hwif_t *hwif)
 {
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
-	struct hpt_info *info	= pci_get_drvdata(dev);
+	struct ide_host *host	= pci_get_drvdata(dev);
+	struct hpt_info *info	= host->host_priv + (hwif->dev == host->dev[1]);
 	int serialize		= HPT_SERIALIZE_IO;
 	u8  chip_type		= info->chip_type;
 	u8  new_mcr, old_mcr	= 0;
@@ -1542,10 +1535,12 @@ static const struct ide_port_info hpt366_chipsets[] __devinitdata = {
 static int __devinit hpt366_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
 	const struct hpt_info *info = NULL;
+	struct hpt_info *dyn_info;
 	struct pci_dev *dev2 = NULL;
 	struct ide_port_info d;
 	u8 idx = id->driver_data;
 	u8 rev = dev->revision;
+	int ret;
 
 	if ((idx == 0 || idx == 4) && (PCI_FUNC(dev->devfn) & 1))
 		return -ENODEV;
@@ -1591,15 +1586,24 @@ static int __devinit hpt366_init_one(struct pci_dev *dev, const struct pci_devic
 	if (info == &hpt370 || info == &hpt370a)
 		d.dma_ops = &hpt370_dma_ops;
 
-	pci_set_drvdata(dev, (void *)info);
-
 	if (info == &hpt36x || info == &hpt374)
 		dev2 = pci_get_slot(dev->bus, dev->devfn + 1);
 
-	if (dev2) {
-		int ret;
+	dyn_info = kzalloc(sizeof(*dyn_info) * (dev2 ? 2 : 1), GFP_KERNEL);
+	if (dyn_info == NULL) {
+		printk(KERN_ERR "%s: out of memory!\n", d.name);
+		pci_dev_put(dev2);
+		return -ENOMEM;
+	}
 
-		pci_set_drvdata(dev2, (void *)info);
+	/*
+	 * Copy everything from a static "template" structure
+	 * to just allocated per-chip hpt_info structure.
+	 */
+	memcpy(dyn_info, info, sizeof(*dyn_info));
+
+	if (dev2) {
+		memcpy(dyn_info + 1, info, sizeof(*dyn_info));
 
 		if (info == &hpt374)
 			hpt374_init(dev, dev2);
@@ -1608,13 +1612,19 @@ static int __devinit hpt366_init_one(struct pci_dev *dev, const struct pci_devic
 				d.host_flags &= ~IDE_HFLAG_NON_BOOTABLE;
 		}
 
-		ret = ide_pci_init_two(dev, dev2, &d, NULL);
-		if (ret < 0)
+		ret = ide_pci_init_two(dev, dev2, &d, dyn_info);
+		if (ret < 0) {
 			pci_dev_put(dev2);
+			kfree(dyn_info);
+		}
 		return ret;
 	}
 
-	return ide_pci_init_one(dev, &d, NULL);
+	ret = ide_pci_init_one(dev, &d, dyn_info);
+	if (ret < 0)
+		kfree(dyn_info);
+
+	return ret;
 }
 
 static const struct pci_device_id hpt366_pci_tbl[] __devinitconst = {

From 1d76d9dc448d5a6fc7b49ba06c634aa6927bcc3d Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:16 +0200
Subject: [PATCH 0534/1171] it821x: convert to use ->host_priv

While at it:

* Allocate both struct it821x_dev instances at once.

* Don't leak itdevs on ide_pci_init_one() failure.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/it821x.c | 28 +++++++++++++---------------
 1 file changed, 13 insertions(+), 15 deletions(-)

diff --git a/drivers/ide/pci/it821x.c b/drivers/ide/pci/it821x.c
index 40186f9e56aa..e63fdafe18fc 100644
--- a/drivers/ide/pci/it821x.c
+++ b/drivers/ide/pci/it821x.c
@@ -534,8 +534,9 @@ static struct ide_dma_ops it821x_pass_through_dma_ops = {
 static void __devinit init_hwif_it821x(ide_hwif_t *hwif)
 {
 	struct pci_dev *dev = to_pci_dev(hwif->dev);
-	struct it821x_dev **itdevs = (struct it821x_dev **)pci_get_drvdata(dev);
-	struct it821x_dev *idev = itdevs[hwif->channel];
+	struct ide_host *host = pci_get_drvdata(dev);
+	struct it821x_dev *itdevs = host->host_priv;
+	struct it821x_dev *idev = itdevs + hwif->channel;
 	u8 conf;
 
 	ide_set_hwifdata(hwif, idev);
@@ -648,23 +649,20 @@ static const struct ide_port_info it821x_chipsets[] __devinitdata = {
 
 static int __devinit it821x_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
-	struct it821x_dev *itdevs[2] = { NULL, NULL} , *itdev;
-	unsigned int i;
+	struct it821x_dev *itdevs;
+	int rc;
 
-	for (i = 0; i < 2; i++) {
-		itdev = kzalloc(sizeof(*itdev), GFP_KERNEL);
-		if (itdev == NULL) {
-			kfree(itdevs[0]);
-			printk(KERN_ERR "it821x: out of memory\n");
-			return -ENOMEM;
-		}
-
-		itdevs[i] = itdev;
+	itdevs = kzalloc(2 * sizeof(*itdevs), GFP_KERNEL);
+	if (itdevs == NULL) {
+		printk(KERN_ERR "it821x: out of memory\n");
+		return -ENOMEM;
 	}
 
-	pci_set_drvdata(dev, itdevs);
+	rc = ide_pci_init_one(dev, &it821x_chipsets[id->driver_data], itdevs);
+	if (rc)
+		kfree(itdevs);
 
-	return ide_pci_init_one(dev, &it821x_chipsets[id->driver_data], NULL);
+	return rc;
 }
 
 static const struct pci_device_id it821x_pci_tbl[] = {

From 96776f3b57eb7beb889a4368937cc9d74082a47e Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:16 +0200
Subject: [PATCH 0535/1171] sc1200: convert to use ->host_priv

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/sc1200.c | 39 +++++++++++++++++++--------------------
 1 file changed, 19 insertions(+), 20 deletions(-)

diff --git a/drivers/ide/pci/sc1200.c b/drivers/ide/pci/sc1200.c
index 8fd9cc2119d6..fa2ce76837d0 100644
--- a/drivers/ide/pci/sc1200.c
+++ b/drivers/ide/pci/sc1200.c
@@ -234,20 +234,10 @@ static int sc1200_suspend (struct pci_dev *dev, pm_message_t state)
 	 * we only save state when going from full power to less
 	 */
 	if (state.event == PM_EVENT_ON) {
-		struct sc1200_saved_state *ss;
+		struct ide_host *host = pci_get_drvdata(dev);
+		struct sc1200_saved_state *ss = host->host_priv;
 		unsigned int r;
 
-		/*
-		 * allocate a permanent save area, if not already allocated
-		 */
-		ss = (struct sc1200_saved_state *)pci_get_drvdata(dev);
-		if (ss == NULL) {
-			ss = kmalloc(sizeof(*ss), GFP_KERNEL);
-			if (ss == NULL)
-				return -ENOMEM;
-			pci_set_drvdata(dev, ss);
-		}
-
 		/*
 		 * save timing registers
 		 * (this may be unnecessary if BIOS also does it)
@@ -263,7 +253,8 @@ static int sc1200_suspend (struct pci_dev *dev, pm_message_t state)
 
 static int sc1200_resume (struct pci_dev *dev)
 {
-	struct sc1200_saved_state *ss;
+	struct ide_host *host = pci_get_drvdata(dev);
+	struct sc1200_saved_state *ss = host->host_priv;
 	unsigned int r;
 	int i;
 
@@ -271,16 +262,12 @@ static int sc1200_resume (struct pci_dev *dev)
 	if (i)
 		return i;
 
-	ss = (struct sc1200_saved_state *)pci_get_drvdata(dev);
-
 	/*
 	 * restore timing registers
 	 * (this may be unnecessary if BIOS also does it)
 	 */
-	if (ss) {
-		for (r = 0; r < 8; r++)
-			pci_write_config_dword(dev, 0x40 + r * 4, ss->regs[r]);
-	}
+	for (r = 0; r < 8; r++)
+		pci_write_config_dword(dev, 0x40 + r * 4, ss->regs[r]);
 
 	return 0;
 }
@@ -317,7 +304,19 @@ static const struct ide_port_info sc1200_chipset __devinitdata = {
 
 static int __devinit sc1200_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
-	return ide_pci_init_one(dev, &sc1200_chipset, NULL);
+	struct sc1200_saved_state *ss = NULL;
+	int rc;
+
+#ifdef CONFIG_PM
+	ss = kmalloc(sizeof(*ss), GFP_KERNEL);
+	if (ss == NULL)
+		return -ENOMEM;
+#endif
+	rc = ide_pci_init_one(dev, &sc1200_chipset, ss);
+	if (rc)
+		kfree(ss);
+
+	return rc;
 }
 
 static const struct pci_device_id sc1200_pci_tbl[] = {

From 4c674235d667d7ddc6b0c95a228a507eb94da2d6 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:16 +0200
Subject: [PATCH 0536/1171] siimage: convert to use ->host_priv

While at it:

* Reserve PCI BAR 5 in siimage_init_one() and remove no longer needed
  setup_mmio_siimage().

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/siimage.c | 114 +++++++++++++++++++-------------------
 1 file changed, 58 insertions(+), 56 deletions(-)

diff --git a/drivers/ide/pci/siimage.c b/drivers/ide/pci/siimage.c
index 48124133601a..1388ffa97644 100644
--- a/drivers/ide/pci/siimage.c
+++ b/drivers/ide/pci/siimage.c
@@ -127,9 +127,10 @@ static inline unsigned long siimage_seldev(ide_drive_t *drive, int r)
 
 static u8 sil_ioread8(struct pci_dev *dev, unsigned long addr)
 {
+	struct ide_host *host = pci_get_drvdata(dev);
 	u8 tmp = 0;
 
-	if (pci_get_drvdata(dev))
+	if (host->host_priv)
 		tmp = readb((void __iomem *)addr);
 	else
 		pci_read_config_byte(dev, addr, &tmp);
@@ -139,9 +140,10 @@ static u8 sil_ioread8(struct pci_dev *dev, unsigned long addr)
 
 static u16 sil_ioread16(struct pci_dev *dev, unsigned long addr)
 {
+	struct ide_host *host = pci_get_drvdata(dev);
 	u16 tmp = 0;
 
-	if (pci_get_drvdata(dev))
+	if (host->host_priv)
 		tmp = readw((void __iomem *)addr);
 	else
 		pci_read_config_word(dev, addr, &tmp);
@@ -151,7 +153,9 @@ static u16 sil_ioread16(struct pci_dev *dev, unsigned long addr)
 
 static void sil_iowrite8(struct pci_dev *dev, u8 val, unsigned long addr)
 {
-	if (pci_get_drvdata(dev))
+	struct ide_host *host = pci_get_drvdata(dev);
+
+	if (host->host_priv)
 		writeb(val, (void __iomem *)addr);
 	else
 		pci_write_config_byte(dev, addr, val);
@@ -159,7 +163,9 @@ static void sil_iowrite8(struct pci_dev *dev, u8 val, unsigned long addr)
 
 static void sil_iowrite16(struct pci_dev *dev, u16 val, unsigned long addr)
 {
-	if (pci_get_drvdata(dev))
+	struct ide_host *host = pci_get_drvdata(dev);
+
+	if (host->host_priv)
 		writew(val, (void __iomem *)addr);
 	else
 		pci_write_config_word(dev, addr, val);
@@ -167,7 +173,9 @@ static void sil_iowrite16(struct pci_dev *dev, u16 val, unsigned long addr)
 
 static void sil_iowrite32(struct pci_dev *dev, u32 val, unsigned long addr)
 {
-	if (pci_get_drvdata(dev))
+	struct ide_host *host = pci_get_drvdata(dev);
+
+	if (host->host_priv)
 		writel(val, (void __iomem *)addr);
 	else
 		pci_write_config_dword(dev, addr, val);
@@ -444,44 +452,6 @@ static void sil_sata_pre_reset(ide_drive_t *drive)
 	}
 }
 
-/**
- *	setup_mmio_siimage	-	switch controller into MMIO mode
- *	@dev: PCI device we are configuring
- *	@name: device name
- *
- *	Attempt to put the device into MMIO mode. There are some slight
- *	complications here with certain systems where the MMIO BAR isn't
- *	mapped, so we have to be sure that we can fall back to I/O.
- */
-
-static unsigned int setup_mmio_siimage(struct pci_dev *dev, const char *name)
-{
-	resource_size_t bar5	= pci_resource_start(dev, 5);
-	unsigned long barsize	= pci_resource_len(dev, 5);
-	void __iomem *ioaddr;
-
-	/*
-	 *	Drop back to PIO if we can't map the MMIO. Some	systems
-	 *	seem to get terminally confused in the PCI spaces.
-	 */
-	if (!request_mem_region(bar5, barsize, name)) {
-		printk(KERN_WARNING "siimage: IDE controller MMIO ports not "
-				    "available.\n");
-		return 0;
-	}
-
-	ioaddr = ioremap(bar5, barsize);
-	if (ioaddr == NULL) {
-		release_mem_region(bar5, barsize);
-		return 0;
-	}
-
-	pci_set_master(dev);
-	pci_set_drvdata(dev, (void *) ioaddr);
-
-	return 1;
-}
-
 /**
  *	init_chipset_siimage	-	set up an SI device
  *	@dev: PCI device
@@ -494,17 +464,15 @@ static unsigned int setup_mmio_siimage(struct pci_dev *dev, const char *name)
 static unsigned int __devinit init_chipset_siimage(struct pci_dev *dev,
 						   const char *name)
 {
+	struct ide_host *host = pci_get_drvdata(dev);
+	void __iomem *ioaddr = host->host_priv;
 	unsigned long base, scsc_addr;
-	void __iomem *ioaddr = NULL;
-	u8 rev = dev->revision, tmp, BA5_EN;
+	u8 rev = dev->revision, tmp;
 
 	pci_write_config_byte(dev, PCI_CACHE_LINE_SIZE, rev ? 1 : 255);
 
-	pci_read_config_byte(dev, 0x8A, &BA5_EN);
-
-	if ((BA5_EN & 0x01) || pci_resource_start(dev, 5))
-		if (setup_mmio_siimage(dev, name))
-			ioaddr = pci_get_drvdata(dev);
+	if (ioaddr)
+		pci_set_master(dev);
 
 	base = (unsigned long)ioaddr;
 
@@ -592,7 +560,8 @@ static unsigned int __devinit init_chipset_siimage(struct pci_dev *dev,
 static void __devinit init_mmio_iops_siimage(ide_hwif_t *hwif)
 {
 	struct pci_dev *dev	= to_pci_dev(hwif->dev);
-	void *addr		= pci_get_drvdata(dev);
+	struct ide_host *host	= pci_get_drvdata(dev);
+	void *addr		= host->host_priv;
 	u8 ch			= hwif->channel;
 	struct ide_io_ports *io_ports = &hwif->io_ports;
 	unsigned long base;
@@ -691,16 +660,15 @@ static void __devinit sil_quirkproc(ide_drive_t *drive)
 static void __devinit init_iops_siimage(ide_hwif_t *hwif)
 {
 	struct pci_dev *dev = to_pci_dev(hwif->dev);
+	struct ide_host *host = pci_get_drvdata(dev);
 
 	hwif->hwif_data = NULL;
 
 	/* Pessimal until we finish probing */
 	hwif->rqsize = 15;
 
-	if (pci_get_drvdata(dev) == NULL)
-		return;
-
-	init_mmio_iops_siimage(hwif);
+	if (host->host_priv)
+		init_mmio_iops_siimage(hwif);
 }
 
 /**
@@ -778,8 +746,13 @@ static const struct ide_port_info siimage_chipsets[] __devinitdata = {
 static int __devinit siimage_init_one(struct pci_dev *dev,
 				      const struct pci_device_id *id)
 {
+	void __iomem *ioaddr = NULL;
+	resource_size_t bar5 = pci_resource_start(dev, 5);
+	unsigned long barsize = pci_resource_len(dev, 5);
+	int rc;
 	struct ide_port_info d;
 	u8 idx = id->driver_data;
+	u8 BA5_EN;
 
 	d = siimage_chipsets[idx];
 
@@ -795,7 +768,36 @@ static int __devinit siimage_init_one(struct pci_dev *dev,
 		d.host_flags |= IDE_HFLAG_NO_ATAPI_DMA;
 	}
 
-	return ide_pci_init_one(dev, &d, NULL);
+	rc = pci_enable_device(dev);
+	if (rc)
+		return rc;
+
+	pci_read_config_byte(dev, 0x8A, &BA5_EN);
+	if ((BA5_EN & 0x01) || bar5) {
+		/*
+		* Drop back to PIO if we can't map the MMIO. Some systems
+		* seem to get terminally confused in the PCI spaces.
+		*/
+		if (!request_mem_region(bar5, barsize, d.name)) {
+			printk(KERN_WARNING "siimage: IDE controller MMIO "
+					    "ports not available.\n");
+		} else {
+			ioaddr = ioremap(bar5, barsize);
+			if (ioaddr == NULL)
+				release_mem_region(bar5, barsize);
+		}
+	}
+
+	rc = ide_pci_init_one(dev, &d, ioaddr);
+	if (rc) {
+		if (ioaddr) {
+			iounmap(ioaddr);
+			release_mem_region(bar5, barsize);
+		}
+		pci_disable_device(dev);
+	}
+
+	return rc;
 }
 
 static const struct pci_device_id siimage_pci_tbl[] = {

From ee77325b074a73694b66ec9eca4f7e55dad58b84 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:17 +0200
Subject: [PATCH 0537/1171] via82cxxx: convert to use ->host_priv

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/via82cxxx.c | 33 +++++++++++++++++++++------------
 1 file changed, 21 insertions(+), 12 deletions(-)

diff --git a/drivers/ide/pci/via82cxxx.c b/drivers/ide/pci/via82cxxx.c
index 2f22abfe003b..a8c050b462bc 100644
--- a/drivers/ide/pci/via82cxxx.c
+++ b/drivers/ide/pci/via82cxxx.c
@@ -113,7 +113,8 @@ struct via82cxxx_dev
 static void via_set_speed(ide_hwif_t *hwif, u8 dn, struct ide_timing *timing)
 {
 	struct pci_dev *dev = to_pci_dev(hwif->dev);
-	struct via82cxxx_dev *vdev = pci_get_drvdata(dev);
+	struct ide_host *host = pci_get_drvdata(dev);
+	struct via82cxxx_dev *vdev = host->host_priv;
 	u8 t;
 
 	if (~vdev->via_config->flags & VIA_BAD_AST) {
@@ -153,7 +154,8 @@ static void via_set_drive(ide_drive_t *drive, const u8 speed)
 	ide_hwif_t *hwif = drive->hwif;
 	ide_drive_t *peer = hwif->drives + (~drive->dn & 1);
 	struct pci_dev *dev = to_pci_dev(hwif->dev);
-	struct via82cxxx_dev *vdev = pci_get_drvdata(dev);
+	struct ide_host *host = pci_get_drvdata(dev);
+	struct via82cxxx_dev *vdev = host->host_priv;
 	struct ide_timing t, p;
 	unsigned int T, UT;
 
@@ -266,19 +268,13 @@ static void __devinit via_cable_detect(struct via82cxxx_dev *vdev, u32 u)
 
 static unsigned int __devinit init_chipset_via82cxxx(struct pci_dev *dev, const char *name)
 {
+	struct ide_host *host = pci_get_drvdata(dev);
+	struct via82cxxx_dev *vdev = host->host_priv;
 	struct pci_dev *isa = NULL;
-	struct via82cxxx_dev *vdev;
 	struct via_isa_bridge *via_config;
 	u8 t, v;
 	u32 u;
 
-	vdev = kzalloc(sizeof(*vdev), GFP_KERNEL);
-	if (!vdev) {
-		printk(KERN_ERR "VP_IDE: out of memory :(\n");
-		return -ENOMEM;
-	}
-	pci_set_drvdata(dev, vdev);
-
 	/*
 	 * Find the ISA bridge to see how good the IDE is.
 	 */
@@ -402,7 +398,8 @@ static int via_cable_override(struct pci_dev *pdev)
 static u8 __devinit via82cxxx_cable_detect(ide_hwif_t *hwif)
 {
 	struct pci_dev *pdev = to_pci_dev(hwif->dev);
-	struct via82cxxx_dev *vdev = pci_get_drvdata(pdev);
+	struct ide_host *host = pci_get_drvdata(pdev);
+	struct via82cxxx_dev *vdev = host->host_priv;
 
 	if (via_cable_override(pdev))
 		return ATA_CBL_PATA40_SHORT;
@@ -436,6 +433,8 @@ static int __devinit via_init_one(struct pci_dev *dev, const struct pci_device_i
 {
 	struct pci_dev *isa = NULL;
 	struct via_isa_bridge *via_config;
+	struct via82cxxx_dev *vdev;
+	int rc;
 	u8 idx = id->driver_data;
 	struct ide_port_info d;
 
@@ -466,7 +465,17 @@ static int __devinit via_init_one(struct pci_dev *dev, const struct pci_device_i
 
 	d.udma_mask = via_config->udma_mask;
 
-	return ide_pci_init_one(dev, &d, NULL);
+	vdev = kzalloc(sizeof(*vdev), GFP_KERNEL);
+	if (!vdev) {
+		printk(KERN_ERR "VP_IDE: out of memory :(\n");
+		return -ENOMEM;
+	}
+
+	rc = ide_pci_init_one(dev, &d, vdev);
+	if (rc)
+		kfree(vdev);
+
+	return rc;
 }
 
 static const struct pci_device_id via_pci_tbl[] = {

From b16040b14e766d390138b04c8829c816f4c1d95b Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:17 +0200
Subject: [PATCH 0538/1171] tc86c001: remove ->init_chipset method

* Reserve PCI BAR 5 in tc86c001_init_one() and remove no longer needed
  init_chipset_tc86c001().

While at it:

* Add & use DRV_NAME define.

Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/tc86c001.c | 38 ++++++++++++++++++++++++++------------
 1 file changed, 26 insertions(+), 12 deletions(-)

diff --git a/drivers/ide/pci/tc86c001.c b/drivers/ide/pci/tc86c001.c
index e16e79d21772..bb3292199356 100644
--- a/drivers/ide/pci/tc86c001.c
+++ b/drivers/ide/pci/tc86c001.c
@@ -11,6 +11,8 @@
 #include <linux/pci.h>
 #include <linux/ide.h>
 
+#define DRV_NAME "TC86C001"
+
 static void tc86c001_set_mode(ide_drive_t *drive, const u8 speed)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
@@ -173,16 +175,6 @@ static void __devinit init_hwif_tc86c001(ide_hwif_t *hwif)
 	hwif->rqsize	 = 0xffff;
 }
 
-static unsigned int __devinit init_chipset_tc86c001(struct pci_dev *dev,
-							const char *name)
-{
-	int err = pci_request_region(dev, 5, name);
-
-	if (err)
-		printk(KERN_ERR "%s: system control regs already in use", name);
-	return err;
-}
-
 static const struct ide_port_ops tc86c001_port_ops = {
 	.set_pio_mode		= tc86c001_set_pio_mode,
 	.set_dma_mode		= tc86c001_set_mode,
@@ -202,7 +194,6 @@ static const struct ide_dma_ops tc86c001_dma_ops = {
 
 static const struct ide_port_info tc86c001_chipset __devinitdata = {
 	.name		= "TC86C001",
-	.init_chipset	= init_chipset_tc86c001,
 	.init_hwif	= init_hwif_tc86c001,
 	.port_ops	= &tc86c001_port_ops,
 	.dma_ops	= &tc86c001_dma_ops,
@@ -215,7 +206,30 @@ static const struct ide_port_info tc86c001_chipset __devinitdata = {
 static int __devinit tc86c001_init_one(struct pci_dev *dev,
 				       const struct pci_device_id *id)
 {
-	return ide_pci_init_one(dev, &tc86c001_chipset, NULL);
+	int rc;
+
+	rc = pci_enable_device(dev);
+	if (rc)
+		goto out;
+
+	rc = pci_request_region(dev, 5, DRV_NAME);
+	if (rc) {
+		printk(KERN_ERR DRV_NAME ": system control regs already in use");
+		goto out_disable;
+	}
+
+	rc = ide_pci_init_one(dev, &tc86c001_chipset, NULL);
+	if (rc)
+		goto out_release;
+
+	goto out;
+
+out_release:
+	pci_release_region(dev, 5);
+out_disable:
+	pci_disable_device(dev);
+out:
+	return rc;
 }
 
 static const struct pci_device_id tc86c001_pci_tbl[] = {

From d51f19c86583ca70468883d8137a92689f1a80c1 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:17 +0200
Subject: [PATCH 0539/1171] amd74xx: cleanup ->init_chipset method

Move amd_clock setup from init_chipset_amd74xx() to amd74xx_probe().

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/amd74xx.c | 36 ++++++++++++++++++------------------
 1 file changed, 18 insertions(+), 18 deletions(-)

diff --git a/drivers/ide/pci/amd74xx.c b/drivers/ide/pci/amd74xx.c
index b6a475313c7c..5115d448ff83 100644
--- a/drivers/ide/pci/amd74xx.c
+++ b/drivers/ide/pci/amd74xx.c
@@ -173,24 +173,6 @@ static unsigned int __devinit init_chipset_amd74xx(struct pci_dev *dev,
 		t |= 0xf0;
 	pci_write_config_byte(dev, AMD_IDE_CONFIG + offset, t);
 
-/*
- * Determine the system bus clock.
- */
-
-	amd_clock = (ide_pci_clk ? ide_pci_clk : 33) * 1000;
-
-	switch (amd_clock) {
-		case 33000: amd_clock = 33333; break;
-		case 37000: amd_clock = 37500; break;
-		case 41000: amd_clock = 41666; break;
-	}
-
-	if (amd_clock < 20000 || amd_clock > 50000) {
-		printk(KERN_WARNING "%s: User given PCI clock speed impossible (%d), using 33 MHz instead.\n",
-				    name, amd_clock);
-		amd_clock = 33333;
-	}
-
 	return dev->irq;
 }
 
@@ -302,6 +284,24 @@ static int __devinit amd74xx_probe(struct pci_dev *dev, const struct pci_device_
 			 d.name, pci_name(dev), dev->revision,
 			 amd_dma[fls(d.udma_mask) - 1]);
 
+	/*
+	* Determine the system bus clock.
+	*/
+	amd_clock = (ide_pci_clk ? ide_pci_clk : 33) * 1000;
+
+	switch (amd_clock) {
+	case 33000: amd_clock = 33333; break;
+	case 37000: amd_clock = 37500; break;
+	case 41000: amd_clock = 41666; break;
+	}
+
+	if (amd_clock < 20000 || amd_clock > 50000) {
+		printk(KERN_WARNING "%s: User given PCI clock speed impossible"
+				    " (%d), using 33 MHz instead.\n",
+				    d.name, amd_clock);
+		amd_clock = 33333;
+	}
+
 	return ide_pci_init_one(dev, &d, NULL);
 }
 

From 0794230fd4b1bf61af8aabd7e987a595d6dbc430 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:17 +0200
Subject: [PATCH 0540/1171] cmd64x: cleanup ->init_chipset method

Remove verbose reporting for CMD646 (PCI device revision is always
logged by IDE PCI layer).

Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/cmd64x.c | 18 ------------------
 1 file changed, 18 deletions(-)

diff --git a/drivers/ide/pci/cmd64x.c b/drivers/ide/pci/cmd64x.c
index fc0333c9a4e5..4eebcf09e0d8 100644
--- a/drivers/ide/pci/cmd64x.c
+++ b/drivers/ide/pci/cmd64x.c
@@ -334,24 +334,6 @@ static unsigned int __devinit init_chipset_cmd64x(struct pci_dev *dev, const cha
 {
 	u8 mrdmode = 0;
 
-	if (dev->device == PCI_DEVICE_ID_CMD_646) {
-
-		switch (dev->revision) {
-		case 0x07:
-		case 0x05:
-			printk("%s: UltraDMA capable\n", name);
-			break;
-		case 0x03:
-		default:
-			printk("%s: MultiWord DMA force limited\n", name);
-			break;
-		case 0x01:
-			printk("%s: MultiWord DMA limited, "
-			       "IRQ workaround enabled\n", name);
-			break;
-		}
-	}
-
 	/* Set a good latency timer and cache line size value. */
 	(void) pci_write_config_byte(dev, PCI_LATENCY_TIMER, 64);
 	/* FIXME: pci_set_master() to ensure a good latency timer value */

From 37525bebcfc15a1fe5a9cb50bf49b21bf43559c1 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:18 +0200
Subject: [PATCH 0541/1171] via82cxxx: cleanup ->init_chipset method

* Move the boot message and via_clock setup from
  init_chipset_via82cxxx() to via_init_one().

* Set vdev->via_config in via_init_one() and cleanup
  init_chipset_via82cxxx() accordingly.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/via82cxxx.c | 81 ++++++++++++++++---------------------
 1 file changed, 35 insertions(+), 46 deletions(-)

diff --git a/drivers/ide/pci/via82cxxx.c b/drivers/ide/pci/via82cxxx.c
index a8c050b462bc..bc1675da37d1 100644
--- a/drivers/ide/pci/via82cxxx.c
+++ b/drivers/ide/pci/via82cxxx.c
@@ -270,21 +270,10 @@ static unsigned int __devinit init_chipset_via82cxxx(struct pci_dev *dev, const
 {
 	struct ide_host *host = pci_get_drvdata(dev);
 	struct via82cxxx_dev *vdev = host->host_priv;
-	struct pci_dev *isa = NULL;
-	struct via_isa_bridge *via_config;
+	struct via_isa_bridge *via_config = vdev->via_config;
 	u8 t, v;
 	u32 u;
 
-	/*
-	 * Find the ISA bridge to see how good the IDE is.
-	 */
-	vdev->via_config = via_config = via_config_find(&isa);
-
-	/* We checked this earlier so if it fails here deeep badness
-	   is involved */
-
-	BUG_ON(!via_config->id);
-
 	/*
 	 * Detect cable and configure Clk66
 	 */
@@ -330,39 +319,6 @@ static unsigned int __devinit init_chipset_via82cxxx(struct pci_dev *dev, const
 
 	pci_write_config_byte(dev, VIA_FIFO_CONFIG, t);
 
-	/*
-	 * Determine system bus clock.
-	 */
-
-	via_clock = (ide_pci_clk ? ide_pci_clk : 33) * 1000;
-
-	switch (via_clock) {
-		case 33000: via_clock = 33333; break;
-		case 37000: via_clock = 37500; break;
-		case 41000: via_clock = 41666; break;
-	}
-
-	if (via_clock < 20000 || via_clock > 50000) {
-		printk(KERN_WARNING "VP_IDE: User given PCI clock speed "
-			"impossible (%d), using 33 MHz instead.\n", via_clock);
-		printk(KERN_WARNING "VP_IDE: Use ide0=ata66 if you want "
-			"to assume 80-wire cable.\n");
-		via_clock = 33333;
-	}
-
-	/*
-	 * Print the boot message.
-	 */
-
-	printk(KERN_INFO "VP_IDE: VIA %s (rev %02x) IDE %sDMA%s "
-		"controller on pci%s\n",
-		via_config->name, isa->revision,
-		via_config->udma_mask ? "U" : "MW",
-		via_dma[via_config->udma_mask ?
-			(fls(via_config->udma_mask) - 1) : 0],
-		pci_name(dev));
-
-	pci_dev_put(isa);
 	return 0;
 }
 
@@ -444,12 +400,43 @@ static int __devinit via_init_one(struct pci_dev *dev, const struct pci_device_i
 	 * Find the ISA bridge and check we know what it is.
 	 */
 	via_config = via_config_find(&isa);
-	pci_dev_put(isa);
 	if (!via_config->id) {
 		printk(KERN_WARNING "VP_IDE: Unknown VIA SouthBridge, disabling DMA.\n");
 		return -ENODEV;
 	}
 
+	/*
+	 * Print the boot message.
+	 */
+	printk(KERN_INFO "VP_IDE: VIA %s (rev %02x) IDE %sDMA%s "
+		"controller on pci%s\n",
+		via_config->name, isa->revision,
+		via_config->udma_mask ? "U" : "MW",
+		via_dma[via_config->udma_mask ?
+			(fls(via_config->udma_mask) - 1) : 0],
+		pci_name(dev));
+
+	pci_dev_put(isa);
+
+	/*
+	 * Determine system bus clock.
+	 */
+	via_clock = (ide_pci_clk ? ide_pci_clk : 33) * 1000;
+
+	switch (via_clock) {
+	case 33000: via_clock = 33333; break;
+	case 37000: via_clock = 37500; break;
+	case 41000: via_clock = 41666; break;
+	}
+
+	if (via_clock < 20000 || via_clock > 50000) {
+		printk(KERN_WARNING "VP_IDE: User given PCI clock speed "
+			"impossible (%d), using 33 MHz instead.\n", via_clock);
+		printk(KERN_WARNING "VP_IDE: Use ide0=ata66 if you want "
+			"to assume 80-wire cable.\n");
+		via_clock = 33333;
+	}
+
 	if (idx == 0)
 		d.host_flags |= IDE_HFLAG_NO_AUTODMA;
 	else
@@ -471,6 +458,8 @@ static int __devinit via_init_one(struct pci_dev *dev, const struct pci_device_i
 		return -ENOMEM;
 	}
 
+	vdev->via_config = via_config;
+
 	rc = ide_pci_init_one(dev, &d, vdev);
 	if (rc)
 		kfree(vdev);

From ef0b04276d8f719d754c092434fbd62c2aeb5307 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:19 +0200
Subject: [PATCH 0542/1171] ide: add ide_pci_remove() helper

* Add 'unsigned long host_flags' field to struct ide_host.

* Set ->host_flags in ide_host_alloc_all().

* Always set PCI dev's ->driver_data in ide_pci_init_{one,two}().

* Add ide_pci_remove() helper (the default implementation for
  struct pci_driver's ->remove method).

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-probe.c |  3 +++
 drivers/ide/setup-pci.c | 39 +++++++++++++++++++++++++++++++++------
 include/linux/ide.h     |  2 ++
 3 files changed, 38 insertions(+), 6 deletions(-)

diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index 9ab5892eaea1..f0c162488ec4 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -1609,6 +1609,9 @@ struct ide_host *ide_host_alloc_all(const struct ide_port_info *d,
 	if (hws[0])
 		host->dev[0] = hws[0]->dev;
 
+	if (d)
+		host->host_flags = d->host_flags;
+
 	return host;
 }
 EXPORT_SYMBOL_GPL(ide_host_alloc_all);
diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
index ca17bf8896df..20f0ee004695 100644
--- a/drivers/ide/setup-pci.c
+++ b/drivers/ide/setup-pci.c
@@ -548,8 +548,7 @@ int ide_pci_init_one(struct pci_dev *dev, const struct ide_port_info *d,
 
 	host->host_priv = priv;
 
-	if (priv)
-		pci_set_drvdata(dev, host);
+	pci_set_drvdata(dev, host);
 
 	ret = do_ide_setup_pci_device(dev, d, 1);
 	if (ret < 0)
@@ -593,10 +592,8 @@ int ide_pci_init_two(struct pci_dev *dev1, struct pci_dev *dev2,
 
 	host->host_priv = priv;
 
-	if (priv) {
-		pci_set_drvdata(pdev[0], host);
-		pci_set_drvdata(pdev[1], host);
-	}
+	pci_set_drvdata(pdev[0], host);
+	pci_set_drvdata(pdev[1], host);
 
 	for (i = 0; i < 2; i++) {
 		ret = do_ide_setup_pci_device(pdev[i], d, !i);
@@ -619,3 +616,33 @@ out:
 	return ret;
 }
 EXPORT_SYMBOL_GPL(ide_pci_init_two);
+
+void ide_pci_remove(struct pci_dev *dev)
+{
+	struct ide_host *host = pci_get_drvdata(dev);
+	struct pci_dev *dev2 = host->dev[1] ? to_pci_dev(host->dev[1]) : NULL;
+	int bars;
+
+	if (host->host_flags & IDE_HFLAG_SINGLE)
+		bars = (1 << 2) - 1;
+	else
+		bars = (1 << 4) - 1;
+
+	if ((host->host_flags & IDE_HFLAG_NO_DMA) == 0) {
+		if (host->host_flags & IDE_HFLAG_CS5520)
+			bars |= (1 << 2);
+		else
+			bars |= (1 << 4);
+	}
+
+	ide_host_remove(host);
+
+	if (dev2)
+		pci_release_selected_regions(dev2, bars);
+	pci_release_selected_regions(dev, bars);
+
+	if (dev2)
+		pci_disable_device(dev2);
+	pci_disable_device(dev);
+}
+EXPORT_SYMBOL_GPL(ide_pci_remove);
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 3eccac0a2a36..dbd0aeb3a56d 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -631,6 +631,7 @@ struct ide_host {
 	ide_hwif_t	*ports[MAX_HWIFS];
 	unsigned int	n_ports;
 	struct device	*dev[2];
+	unsigned long	host_flags;
 	void		*host_priv;
 };
 
@@ -1213,6 +1214,7 @@ struct ide_port_info {
 int ide_pci_init_one(struct pci_dev *, const struct ide_port_info *, void *);
 int ide_pci_init_two(struct pci_dev *, struct pci_dev *,
 		     const struct ide_port_info *, void *);
+void ide_pci_remove(struct pci_dev *);
 
 void ide_map_sg(ide_drive_t *, struct request *);
 void ide_init_sg_cmd(ide_drive_t *, struct request *);

From eb7cb98b1cc8be1d4395d9accf49ae3924cd68f1 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:19 +0200
Subject: [PATCH 0543/1171] aec62xx: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/aec62xx.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/drivers/ide/pci/aec62xx.c b/drivers/ide/pci/aec62xx.c
index 7ca7989bc35d..253299961a14 100644
--- a/drivers/ide/pci/aec62xx.c
+++ b/drivers/ide/pci/aec62xx.c
@@ -280,6 +280,12 @@ static int __devinit aec62xx_init_one(struct pci_dev *dev, const struct pci_devi
 	return err;
 }
 
+static void __devexit aec62xx_remove(struct pci_dev *dev)
+{
+	ide_pci_remove(dev);
+	pci_disable_device(dev);
+}
+
 static const struct pci_device_id aec62xx_pci_tbl[] = {
 	{ PCI_VDEVICE(ARTOP, PCI_DEVICE_ID_ARTOP_ATP850UF), 0 },
 	{ PCI_VDEVICE(ARTOP, PCI_DEVICE_ID_ARTOP_ATP860),   1 },
@@ -294,6 +300,7 @@ static struct pci_driver driver = {
 	.name		= "AEC62xx_IDE",
 	.id_table	= aec62xx_pci_tbl,
 	.probe		= aec62xx_init_one,
+	.remove		= aec62xx_remove,
 };
 
 static int __init aec62xx_ide_init(void)
@@ -301,7 +308,13 @@ static int __init aec62xx_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit aec62xx_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(aec62xx_ide_init);
+module_exit(aec62xx_ide_exit);
 
 MODULE_AUTHOR("Andre Hedrick");
 MODULE_DESCRIPTION("PCI driver module for ARTOP AEC62xx IDE");

From 8ee3f3b69d9c37f86a45862f53451699ec77fe12 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:19 +0200
Subject: [PATCH 0544/1171] alim15x3: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/alim15x3.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/ide/pci/alim15x3.c b/drivers/ide/pci/alim15x3.c
index 7f96e7ca3864..ac171502b997 100644
--- a/drivers/ide/pci/alim15x3.c
+++ b/drivers/ide/pci/alim15x3.c
@@ -580,6 +580,7 @@ static struct pci_driver driver = {
 	.name		= "ALI15x3_IDE",
 	.id_table	= alim15x3_pci_tbl,
 	.probe		= alim15x3_init_one,
+	.remove		= ide_pci_remove,
 };
 
 static int __init ali15x3_ide_init(void)
@@ -587,7 +588,13 @@ static int __init ali15x3_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit ali15x3_ide_exit(void)
+{
+	return pci_unregister_driver(&driver);
+}
+
 module_init(ali15x3_ide_init);
+module_exit(ali15x3_ide_exit);
 
 MODULE_AUTHOR("Michael Aubry, Andrzej Krzysztofowicz, CJ, Andre Hedrick, Alan Cox");
 MODULE_DESCRIPTION("PCI driver module for ALi 15x3 IDE");

From b2509ac1d9dbe7a9d3a9915afbe108978002c95b Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:19 +0200
Subject: [PATCH 0545/1171] amd74xx: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/amd74xx.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/ide/pci/amd74xx.c b/drivers/ide/pci/amd74xx.c
index 5115d448ff83..a91f2e873ba4 100644
--- a/drivers/ide/pci/amd74xx.c
+++ b/drivers/ide/pci/amd74xx.c
@@ -341,6 +341,7 @@ static struct pci_driver driver = {
 	.name		= "AMD_IDE",
 	.id_table	= amd74xx_pci_tbl,
 	.probe		= amd74xx_probe,
+	.remove		= ide_pci_remove,
 };
 
 static int __init amd74xx_ide_init(void)
@@ -348,7 +349,13 @@ static int __init amd74xx_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit amd74xx_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(amd74xx_ide_init);
+module_exit(amd74xx_ide_exit);
 
 MODULE_AUTHOR("Vojtech Pavlik");
 MODULE_DESCRIPTION("AMD PCI IDE driver");

From f354fbc4b45a730aa0f876322ea4f096b47d1013 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:20 +0200
Subject: [PATCH 0546/1171] atiixp: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/atiixp.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/ide/pci/atiixp.c b/drivers/ide/pci/atiixp.c
index b483a68b39f6..4c49c1ba6182 100644
--- a/drivers/ide/pci/atiixp.c
+++ b/drivers/ide/pci/atiixp.c
@@ -184,6 +184,7 @@ static struct pci_driver driver = {
 	.name		= "ATIIXP_IDE",
 	.id_table	= atiixp_pci_tbl,
 	.probe		= atiixp_init_one,
+	.remove		= ide_pci_remove,
 };
 
 static int __init atiixp_ide_init(void)
@@ -191,7 +192,13 @@ static int __init atiixp_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit atiixp_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(atiixp_ide_init);
+module_exit(atiixp_ide_exit);
 
 MODULE_AUTHOR("HUI YU");
 MODULE_DESCRIPTION("PCI driver module for ATI IXP IDE");

From e2b15b4765ca032d0837dfc8c195ecd3bc56a433 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:20 +0200
Subject: [PATCH 0547/1171] cmd64x: add ->remove method and module_exit()

Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/cmd64x.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/ide/pci/cmd64x.c b/drivers/ide/pci/cmd64x.c
index 4eebcf09e0d8..0a4d194bc35f 100644
--- a/drivers/ide/pci/cmd64x.c
+++ b/drivers/ide/pci/cmd64x.c
@@ -505,6 +505,7 @@ static struct pci_driver driver = {
 	.name		= "CMD64x_IDE",
 	.id_table	= cmd64x_pci_tbl,
 	.probe		= cmd64x_init_one,
+	.remove		= ide_pci_remove,
 };
 
 static int __init cmd64x_ide_init(void)
@@ -512,7 +513,13 @@ static int __init cmd64x_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit cmd64x_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(cmd64x_ide_init);
+module_exit(cmd64x_ide_exit);
 
 MODULE_AUTHOR("Eddie Dost, David Miller, Andre Hedrick");
 MODULE_DESCRIPTION("PCI driver module for CMD64x IDE");

From d16492a9789982955e627a7ffdcd1c3b945f7e85 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:20 +0200
Subject: [PATCH 0548/1171] cs5530: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/cs5530.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/ide/pci/cs5530.c b/drivers/ide/pci/cs5530.c
index ba82bad8bf4e..dff345c763e5 100644
--- a/drivers/ide/pci/cs5530.c
+++ b/drivers/ide/pci/cs5530.c
@@ -269,6 +269,7 @@ static struct pci_driver driver = {
 	.name		= "CS5530 IDE",
 	.id_table	= cs5530_pci_tbl,
 	.probe		= cs5530_init_one,
+	.remove		= ide_pci_remove,
 };
 
 static int __init cs5530_ide_init(void)
@@ -276,7 +277,13 @@ static int __init cs5530_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit cs5530_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(cs5530_ide_init);
+module_exit(cs5530_ide_exit);
 
 MODULE_AUTHOR("Mark Lord");
 MODULE_DESCRIPTION("PCI driver module for Cyrix/NS 5530 IDE");

From 40c8a7f67d38de87f97a548b81b6cd0621a3ff9a Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:20 +0200
Subject: [PATCH 0549/1171] cs5535: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/cs5535.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/ide/pci/cs5535.c b/drivers/ide/pci/cs5535.c
index 2161f43ca1b8..a73001391536 100644
--- a/drivers/ide/pci/cs5535.c
+++ b/drivers/ide/pci/cs5535.c
@@ -194,6 +194,7 @@ static struct pci_driver driver = {
 	.name       = "CS5535_IDE",
 	.id_table   = cs5535_pci_tbl,
 	.probe      = cs5535_init_one,
+	.remove     = ide_pci_remove,
 };
 
 static int __init cs5535_ide_init(void)
@@ -201,7 +202,13 @@ static int __init cs5535_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit cs5535_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(cs5535_ide_init);
+module_exit(cs5535_ide_exit);
 
 MODULE_AUTHOR("AMD");
 MODULE_DESCRIPTION("PCI driver module for AMD/NS CS5535 IDE");

From cd68841b854e24076d41c32eae3ccfce6ae60a59 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:21 +0200
Subject: [PATCH 0550/1171] cy82c693: add ->remove method and module_exit()

Fix the refcounting for dev2 while at it.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/cy82c693.c | 19 ++++++++++++++++++-
 1 file changed, 18 insertions(+), 1 deletion(-)

diff --git a/drivers/ide/pci/cy82c693.c b/drivers/ide/pci/cy82c693.c
index abd27ed7c30c..04f268866b6d 100644
--- a/drivers/ide/pci/cy82c693.c
+++ b/drivers/ide/pci/cy82c693.c
@@ -420,11 +420,21 @@ static int __devinit cy82c693_init_one(struct pci_dev *dev, const struct pci_dev
 	    PCI_FUNC(dev->devfn) == 1) {
 		dev2 = pci_get_slot(dev->bus, dev->devfn + 1);
 		ret = ide_pci_init_two(dev, dev2, &cy82c693_chipset, NULL);
-		/* We leak pci refs here but thats ok - we can't be unloaded */
+		if (ret)
+			pci_dev_put(dev2);
 	}
 	return ret;
 }
 
+static void __devexit cy82c693_remove(struct pci_dev *dev)
+{
+	struct ide_host *host = pci_get_drvdata(dev);
+	struct pci_dev *dev2 = host->dev[1] ? to_pci_dev(host->dev[1]) : NULL;
+
+	ide_pci_remove(dev);
+	pci_dev_put(dev2);
+}
+
 static const struct pci_device_id cy82c693_pci_tbl[] = {
 	{ PCI_VDEVICE(CONTAQ, PCI_DEVICE_ID_CONTAQ_82C693), 0 },
 	{ 0, },
@@ -435,6 +445,7 @@ static struct pci_driver driver = {
 	.name		= "Cypress_IDE",
 	.id_table	= cy82c693_pci_tbl,
 	.probe		= cy82c693_init_one,
+	.remove		= cy82c693_remove,
 };
 
 static int __init cy82c693_ide_init(void)
@@ -442,7 +453,13 @@ static int __init cy82c693_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit cy82c693_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(cy82c693_ide_init);
+module_exit(cy82c693_ide_exit);
 
 MODULE_AUTHOR("Andreas Krebs, Andre Hedrick");
 MODULE_DESCRIPTION("PCI driver module for the Cypress CY82C693 IDE");

From f566bcae9fb39b108e39a2f31594c028d6ee2e77 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:21 +0200
Subject: [PATCH 0551/1171] ide/pci/generic: add ->remove method and
 module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/generic.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/ide/pci/generic.c b/drivers/ide/pci/generic.c
index dd0caea5e4f3..81fc171f8ffc 100644
--- a/drivers/ide/pci/generic.c
+++ b/drivers/ide/pci/generic.c
@@ -174,6 +174,7 @@ static struct pci_driver driver = {
 	.name		= "PCI_IDE",
 	.id_table	= generic_pci_tbl,
 	.probe		= generic_init_one,
+	.remove		= ide_pci_remove,
 };
 
 static int __init generic_ide_init(void)
@@ -181,7 +182,13 @@ static int __init generic_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit generic_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(generic_ide_init);
+module_exit(generic_ide_exit);
 
 MODULE_AUTHOR("Andre Hedrick");
 MODULE_DESCRIPTION("PCI driver module for generic PCI IDE");

From 741ac62f6fca55ddbef52513fbc687ba6b04f99e Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:21 +0200
Subject: [PATCH 0552/1171] hpt34x: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/hpt34x.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/ide/pci/hpt34x.c b/drivers/ide/pci/hpt34x.c
index 3d70c5150ac6..b52f8339102e 100644
--- a/drivers/ide/pci/hpt34x.c
+++ b/drivers/ide/pci/hpt34x.c
@@ -169,6 +169,7 @@ static struct pci_driver driver = {
 	.name		= "HPT34x_IDE",
 	.id_table	= hpt34x_pci_tbl,
 	.probe		= hpt34x_init_one,
+	.remove		= ide_pci_remove,
 };
 
 static int __init hpt34x_ide_init(void)
@@ -176,7 +177,13 @@ static int __init hpt34x_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit hpt34x_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(hpt34x_ide_init);
+module_exit(hpt34x_ide_exit);
 
 MODULE_AUTHOR("Andre Hedrick");
 MODULE_DESCRIPTION("PCI driver module for Highpoint 34x IDE");

From a6c43a2be9721d00ef9d6ef5b7b0e8113444577b Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:21 +0200
Subject: [PATCH 0553/1171] hpt366: add ->remove method and module_exit()

Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/hpt366.c | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/drivers/ide/pci/hpt366.c b/drivers/ide/pci/hpt366.c
index 8f29571345a2..398808905f94 100644
--- a/drivers/ide/pci/hpt366.c
+++ b/drivers/ide/pci/hpt366.c
@@ -1627,6 +1627,17 @@ static int __devinit hpt366_init_one(struct pci_dev *dev, const struct pci_devic
 	return ret;
 }
 
+static void __devexit hpt366_remove(struct pci_dev *dev)
+{
+	struct ide_host *host = pci_get_drvdata(dev);
+	struct ide_info *info = host->host_priv;
+	struct pci_dev *dev2 = host->dev[1] ? to_pci_dev(host->dev[1]) : NULL;
+
+	ide_pci_remove(dev);
+	pci_dev_put(dev2);
+	kfree(info);
+}
+
 static const struct pci_device_id hpt366_pci_tbl[] __devinitconst = {
 	{ PCI_VDEVICE(TTI, PCI_DEVICE_ID_TTI_HPT366),  0 },
 	{ PCI_VDEVICE(TTI, PCI_DEVICE_ID_TTI_HPT372),  1 },
@@ -1642,6 +1653,7 @@ static struct pci_driver driver = {
 	.name		= "HPT366_IDE",
 	.id_table	= hpt366_pci_tbl,
 	.probe		= hpt366_init_one,
+	.remove		= hpt366_remove,
 };
 
 static int __init hpt366_ide_init(void)
@@ -1649,7 +1661,13 @@ static int __init hpt366_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit hpt366_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(hpt366_ide_init);
+module_exit(hpt366_ide_exit);
 
 MODULE_AUTHOR("Andre Hedrick");
 MODULE_DESCRIPTION("PCI driver module for Highpoint HPT366 IDE");

From 5102f768570b3486979afb68c595b71cfb7f026f Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:22 +0200
Subject: [PATCH 0554/1171] it8213: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/it8213.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/ide/pci/it8213.c b/drivers/ide/pci/it8213.c
index 18219fa9ef01..575bf2386f8e 100644
--- a/drivers/ide/pci/it8213.c
+++ b/drivers/ide/pci/it8213.c
@@ -198,6 +198,7 @@ static struct pci_driver driver = {
 	.name		= "ITE8213_IDE",
 	.id_table	= it8213_pci_tbl,
 	.probe		= it8213_init_one,
+	.remove		= ide_pci_remove,
 };
 
 static int __init it8213_ide_init(void)
@@ -205,7 +206,13 @@ static int __init it8213_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit it8213_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(it8213_ide_init);
+module_exit(it8213_ide_exit);
 
 MODULE_AUTHOR("Jack Lee, Alan Cox");
 MODULE_DESCRIPTION("PCI driver module for the ITE 8213");

From 87d8b61356108835f5e91c0fb32b830ec585978c Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:22 +0200
Subject: [PATCH 0555/1171] it821x: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/it821x.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/drivers/ide/pci/it821x.c b/drivers/ide/pci/it821x.c
index e63fdafe18fc..c5c91f2109d4 100644
--- a/drivers/ide/pci/it821x.c
+++ b/drivers/ide/pci/it821x.c
@@ -665,6 +665,15 @@ static int __devinit it821x_init_one(struct pci_dev *dev, const struct pci_devic
 	return rc;
 }
 
+static void __devexit it821x_remove(struct pci_dev *dev)
+{
+	struct ide_host *host = pci_get_drvdata(dev);
+	struct it821x_dev *itdevs = host->host_priv;
+
+	ide_pci_remove(dev);
+	kfree(itdevs);
+}
+
 static const struct pci_device_id it821x_pci_tbl[] = {
 	{ PCI_VDEVICE(ITE, PCI_DEVICE_ID_ITE_8211), 0 },
 	{ PCI_VDEVICE(ITE, PCI_DEVICE_ID_ITE_8212), 0 },
@@ -677,6 +686,7 @@ static struct pci_driver driver = {
 	.name		= "ITE821x IDE",
 	.id_table	= it821x_pci_tbl,
 	.probe		= it821x_init_one,
+	.remove		= it821x_remove,
 };
 
 static int __init it821x_ide_init(void)
@@ -684,7 +694,13 @@ static int __init it821x_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit it821x_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(it821x_ide_init);
+module_exit(it821x_ide_exit);
 
 module_param_named(noraid, it8212_noraid, int, S_IRUGO);
 MODULE_PARM_DESC(noraid, "Force card into bypass mode");

From 1bcaaba7749dce7c0506cff0e811c9bed8121f38 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:22 +0200
Subject: [PATCH 0556/1171] jmicron: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/jmicron.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/ide/pci/jmicron.c b/drivers/ide/pci/jmicron.c
index a7e3c14f7b07..39e221b076e2 100644
--- a/drivers/ide/pci/jmicron.c
+++ b/drivers/ide/pci/jmicron.c
@@ -152,6 +152,7 @@ static struct pci_driver driver = {
 	.name		= "JMicron IDE",
 	.id_table	= jmicron_pci_tbl,
 	.probe		= jmicron_init_one,
+	.remove		= ide_pci_remove,
 };
 
 static int __init jmicron_ide_init(void)
@@ -159,7 +160,13 @@ static int __init jmicron_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit jmicron_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(jmicron_ide_init);
+module_exit(jmicron_ide_exit);
 
 MODULE_AUTHOR("Alan Cox");
 MODULE_DESCRIPTION("PCI driver module for the JMicron in legacy modes");

From aa6e518d75742fd3ac3d2cb4c2bcbae850319fc1 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:23 +0200
Subject: [PATCH 0557/1171] ns87415: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/ns87415.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/ide/pci/ns87415.c b/drivers/ide/pci/ns87415.c
index a45c33c0c792..afcc742a63ac 100644
--- a/drivers/ide/pci/ns87415.c
+++ b/drivers/ide/pci/ns87415.c
@@ -337,6 +337,7 @@ static struct pci_driver driver = {
 	.name		= "NS87415_IDE",
 	.id_table	= ns87415_pci_tbl,
 	.probe		= ns87415_init_one,
+	.remove		= ide_pci_remove,
 };
 
 static int __init ns87415_ide_init(void)
@@ -344,7 +345,13 @@ static int __init ns87415_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit ns87415_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(ns87415_ide_init);
+module_exit(ns87415_ide_exit);
 
 MODULE_AUTHOR("Mark Lord, Eddie Dost, Andre Hedrick");
 MODULE_DESCRIPTION("PCI driver module for NS87415 IDE");

From adc7f85ae68bd2e8db2e0136dcd4679891e5c321 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:23 +0200
Subject: [PATCH 0558/1171] opti621: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/opti621.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/ide/pci/opti621.c b/drivers/ide/pci/opti621.c
index edb9132ffbe4..4895f2ff3f05 100644
--- a/drivers/ide/pci/opti621.c
+++ b/drivers/ide/pci/opti621.c
@@ -223,6 +223,7 @@ static struct pci_driver driver = {
 	.name		= "Opti621_IDE",
 	.id_table	= opti621_pci_tbl,
 	.probe		= opti621_init_one,
+	.remove		= ide_pci_remove,
 };
 
 static int __init opti621_ide_init(void)
@@ -230,7 +231,13 @@ static int __init opti621_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit opti621_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(opti621_ide_init);
+module_exit(opti621_ide_exit);
 
 MODULE_AUTHOR("Jaromir Koutek, Jan Harkes, Mark Lord");
 MODULE_DESCRIPTION("PCI driver module for Opti621 IDE");

From d69c8f8c0068b9fc7f5a5082d8a891618b732e2d Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:23 +0200
Subject: [PATCH 0559/1171] pdc202xx_new: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/pdc202xx_new.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/drivers/ide/pci/pdc202xx_new.c b/drivers/ide/pci/pdc202xx_new.c
index 71a420feb981..9881a1a8b89c 100644
--- a/drivers/ide/pci/pdc202xx_new.c
+++ b/drivers/ide/pci/pdc202xx_new.c
@@ -543,6 +543,15 @@ static int __devinit pdc202new_init_one(struct pci_dev *dev, const struct pci_de
 	return ide_pci_init_one(dev, d, NULL);
 }
 
+static void __devexit pdc202new_remove(struct pci_dev *dev)
+{
+	struct ide_host *host = pci_get_drvdata(dev);
+	struct pci_dev *dev2 = host->dev[1] ? to_pci_dev(host->dev[1]) : NULL;
+
+	ide_pci_remove(dev);
+	pci_dev_put(dev2);
+}
+
 static const struct pci_device_id pdc202new_pci_tbl[] = {
 	{ PCI_VDEVICE(PROMISE, PCI_DEVICE_ID_PROMISE_20268), 0 },
 	{ PCI_VDEVICE(PROMISE, PCI_DEVICE_ID_PROMISE_20269), 1 },
@@ -559,6 +568,7 @@ static struct pci_driver driver = {
 	.name		= "Promise_IDE",
 	.id_table	= pdc202new_pci_tbl,
 	.probe		= pdc202new_init_one,
+	.remove		= pdc202new_remove,
 };
 
 static int __init pdc202new_ide_init(void)
@@ -566,7 +576,13 @@ static int __init pdc202new_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit pdc202new_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(pdc202new_ide_init);
+module_exit(pdc202new_ide_exit);
 
 MODULE_AUTHOR("Andre Hedrick, Frank Tiernan");
 MODULE_DESCRIPTION("PCI driver module for Promise PDC20268 and higher");

From 574a1c24b63fdb584935b4924a38b451eeb0880e Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:24 +0200
Subject: [PATCH 0560/1171] pdc202xx_old: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/pdc202xx_old.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/ide/pci/pdc202xx_old.c b/drivers/ide/pci/pdc202xx_old.c
index eba1d60a73a0..17d99ce273af 100644
--- a/drivers/ide/pci/pdc202xx_old.c
+++ b/drivers/ide/pci/pdc202xx_old.c
@@ -429,6 +429,7 @@ static struct pci_driver driver = {
 	.name		= "Promise_Old_IDE",
 	.id_table	= pdc202xx_pci_tbl,
 	.probe		= pdc202xx_init_one,
+	.remove		= ide_pci_remove,
 };
 
 static int __init pdc202xx_ide_init(void)
@@ -436,7 +437,13 @@ static int __init pdc202xx_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit pdc202xx_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(pdc202xx_ide_init);
+module_exit(pdc202xx_ide_exit);
 
 MODULE_AUTHOR("Andre Hedrick, Frank Tiernan");
 MODULE_DESCRIPTION("PCI driver module for older Promise IDE");

From da8c3e0d21c5dbb2815d7c8f1f09e0c68f626ed1 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:24 +0200
Subject: [PATCH 0561/1171] piix: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/piix.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/ide/pci/piix.c b/drivers/ide/pci/piix.c
index 359f65ddcbf9..c16b1ab4d4ff 100644
--- a/drivers/ide/pci/piix.c
+++ b/drivers/ide/pci/piix.c
@@ -462,6 +462,7 @@ static struct pci_driver driver = {
 	.name		= "PIIX_IDE",
 	.id_table	= piix_pci_tbl,
 	.probe		= piix_init_one,
+	.remove		= ide_pci_remove,
 };
 
 static int __init piix_ide_init(void)
@@ -470,7 +471,13 @@ static int __init piix_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit piix_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(piix_ide_init);
+module_exit(piix_ide_exit);
 
 MODULE_AUTHOR("Andre Hedrick, Andrzej Krzysztofowicz");
 MODULE_DESCRIPTION("PCI driver module for Intel PIIX IDE");

From 0fd188047ca75df85191cc55f929cb2889631430 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:24 +0200
Subject: [PATCH 0562/1171] rz1000: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/rz1000.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/ide/pci/rz1000.c b/drivers/ide/pci/rz1000.c
index 860ffdeca095..f7a3b9aff29d 100644
--- a/drivers/ide/pci/rz1000.c
+++ b/drivers/ide/pci/rz1000.c
@@ -62,6 +62,7 @@ static struct pci_driver driver = {
 	.name		= "RZ1000_IDE",
 	.id_table	= rz1000_pci_tbl,
 	.probe		= rz1000_init_one,
+	.remove		= ide_pci_remove,
 };
 
 static int __init rz1000_ide_init(void)
@@ -69,7 +70,13 @@ static int __init rz1000_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit rz1000_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(rz1000_ide_init);
+module_exit(rz1000_ide_exit);
 
 MODULE_AUTHOR("Andre Hedrick");
 MODULE_DESCRIPTION("PCI driver module for RZ1000 IDE");

From 991f5e69c512b284aaec81432dff0440b2a2f418 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:24 +0200
Subject: [PATCH 0563/1171] sc1200: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/sc1200.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/ide/pci/sc1200.c b/drivers/ide/pci/sc1200.c
index fa2ce76837d0..6509560ba665 100644
--- a/drivers/ide/pci/sc1200.c
+++ b/drivers/ide/pci/sc1200.c
@@ -329,6 +329,7 @@ static struct pci_driver driver = {
 	.name		= "SC1200_IDE",
 	.id_table	= sc1200_pci_tbl,
 	.probe		= sc1200_init_one,
+	.remove		= ide_pci_remove,
 #ifdef CONFIG_PM
 	.suspend	= sc1200_suspend,
 	.resume		= sc1200_resume,
@@ -340,7 +341,13 @@ static int __init sc1200_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit sc1200_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(sc1200_ide_init);
+module_exit(sc1200_ide_exit);
 
 MODULE_AUTHOR("Mark Lord");
 MODULE_DESCRIPTION("PCI driver module for NS SC1200 IDE");

From bc2c9a8025921972f0774859b8f19b324734e824 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:25 +0200
Subject: [PATCH 0564/1171] serverworks: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/serverworks.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/ide/pci/serverworks.c b/drivers/ide/pci/serverworks.c
index 34abdfc8d567..eca52fede281 100644
--- a/drivers/ide/pci/serverworks.c
+++ b/drivers/ide/pci/serverworks.c
@@ -439,6 +439,7 @@ static struct pci_driver driver = {
 	.name		= "Serverworks_IDE",
 	.id_table	= svwks_pci_tbl,
 	.probe		= svwks_init_one,
+	.remove		= ide_pci_remove,
 };
 
 static int __init svwks_ide_init(void)
@@ -446,7 +447,13 @@ static int __init svwks_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit svwks_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(svwks_ide_init);
+module_exit(svwks_ide_exit);
 
 MODULE_AUTHOR("Michael Aubry. Andrzej Krzysztofowicz, Andre Hedrick");
 MODULE_DESCRIPTION("PCI driver module for Serverworks OSB4/CSB5/CSB6 IDE");

From fe3825808ad67af02bd826a0d2ca6831e947e80e Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:25 +0200
Subject: [PATCH 0565/1171] siimage: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/siimage.c | 25 +++++++++++++++++++++++++
 1 file changed, 25 insertions(+)

diff --git a/drivers/ide/pci/siimage.c b/drivers/ide/pci/siimage.c
index 1388ffa97644..f2e4dac9b719 100644
--- a/drivers/ide/pci/siimage.c
+++ b/drivers/ide/pci/siimage.c
@@ -800,6 +800,24 @@ static int __devinit siimage_init_one(struct pci_dev *dev,
 	return rc;
 }
 
+static void __devexit siimage_remove(struct pci_dev *dev)
+{
+	struct ide_host *host = pci_get_drvdata(dev);
+	void __iomem *ioaddr = host->host_priv;
+
+	ide_pci_remove(dev);
+
+	if (ioaddr) {
+		resource_size_t bar5 = pci_resource_start(dev, 5);
+		unsigned long barsize = pci_resource_len(dev, 5);
+
+		iounmap(ioaddr);
+		release_mem_region(bar5, barsize);
+	}
+
+	pci_disable_device(dev);
+}
+
 static const struct pci_device_id siimage_pci_tbl[] = {
 	{ PCI_VDEVICE(CMD, PCI_DEVICE_ID_SII_680),    0 },
 #ifdef CONFIG_BLK_DEV_IDE_SATA
@@ -814,6 +832,7 @@ static struct pci_driver driver = {
 	.name		= "SiI_IDE",
 	.id_table	= siimage_pci_tbl,
 	.probe		= siimage_init_one,
+	.remove		= siimage_remove,
 };
 
 static int __init siimage_ide_init(void)
@@ -821,7 +840,13 @@ static int __init siimage_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit siimage_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(siimage_ide_init);
+module_exit(siimage_ide_exit);
 
 MODULE_AUTHOR("Andre Hedrick, Alan Cox");
 MODULE_DESCRIPTION("PCI driver module for SiI IDE");

From 1ceb906b4062954e92295191402e9214345ee0e9 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:25 +0200
Subject: [PATCH 0566/1171] sis5513: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/sis5513.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/drivers/ide/pci/sis5513.c b/drivers/ide/pci/sis5513.c
index a2330c4ac75b..f3cf7c6c0cc3 100644
--- a/drivers/ide/pci/sis5513.c
+++ b/drivers/ide/pci/sis5513.c
@@ -586,6 +586,12 @@ static int __devinit sis5513_init_one(struct pci_dev *dev, const struct pci_devi
 	return ide_pci_init_one(dev, &d, NULL);
 }
 
+static void __devexit sis5513_remove(struct pci_dev *dev)
+{
+	ide_pci_remove(dev);
+	pci_disable_device(dev);
+}
+
 static const struct pci_device_id sis5513_pci_tbl[] = {
 	{ PCI_VDEVICE(SI, PCI_DEVICE_ID_SI_5513), 0 },
 	{ PCI_VDEVICE(SI, PCI_DEVICE_ID_SI_5518), 0 },
@@ -598,6 +604,7 @@ static struct pci_driver driver = {
 	.name		= "SIS_IDE",
 	.id_table	= sis5513_pci_tbl,
 	.probe		= sis5513_init_one,
+	.remove		= sis5513_remove,
 };
 
 static int __init sis5513_ide_init(void)
@@ -605,7 +612,13 @@ static int __init sis5513_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit sis5513_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(sis5513_ide_init);
+module_exit(sis5513_ide_exit);
 
 MODULE_AUTHOR("Lionel Bouton, L C Chang, Andre Hedrick, Vojtech Pavlik");
 MODULE_DESCRIPTION("PCI driver module for SIS IDE");

From 6ce7199897bcbad05ecd06a4df22795fb37f4d0a Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:25 +0200
Subject: [PATCH 0567/1171] sl82c105: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/sl82c105.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/ide/pci/sl82c105.c b/drivers/ide/pci/sl82c105.c
index be22f8125d71..72899f85c5ba 100644
--- a/drivers/ide/pci/sl82c105.c
+++ b/drivers/ide/pci/sl82c105.c
@@ -348,6 +348,7 @@ static struct pci_driver driver = {
 	.name		= "W82C105_IDE",
 	.id_table	= sl82c105_pci_tbl,
 	.probe		= sl82c105_init_one,
+	.remove		= ide_pci_remove,
 };
 
 static int __init sl82c105_ide_init(void)
@@ -355,7 +356,13 @@ static int __init sl82c105_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit sl82c105_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(sl82c105_ide_init);
+module_exit(sl82c105_ide_exit);
 
 MODULE_DESCRIPTION("PCI driver module for W82C105 IDE");
 MODULE_LICENSE("GPL");

From 64b0fed31d6704e4e2e42e9a1ac5995b0a1b54e4 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:25 +0200
Subject: [PATCH 0568/1171] slc90e66: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/slc90e66.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/ide/pci/slc90e66.c b/drivers/ide/pci/slc90e66.c
index 2fc2f2cf2206..fee5ebe4bb99 100644
--- a/drivers/ide/pci/slc90e66.c
+++ b/drivers/ide/pci/slc90e66.c
@@ -157,6 +157,7 @@ static struct pci_driver driver = {
 	.name		= "SLC90e66_IDE",
 	.id_table	= slc90e66_pci_tbl,
 	.probe		= slc90e66_init_one,
+	.remove		= ide_pci_remove,
 };
 
 static int __init slc90e66_ide_init(void)
@@ -164,7 +165,13 @@ static int __init slc90e66_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit slc90e66_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(slc90e66_ide_init);
+module_exit(slc90e66_ide_exit);
 
 MODULE_AUTHOR("Andre Hedrick");
 MODULE_DESCRIPTION("PCI driver module for SLC90E66 IDE");

From ea881d6d6c58aa6d56105d1faba7432243ea7118 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:26 +0200
Subject: [PATCH 0569/1171] tc86c001: add ->remove method and module_exit()

Cc: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/tc86c001.c | 17 ++++++++++++++++-
 1 file changed, 16 insertions(+), 1 deletion(-)

diff --git a/drivers/ide/pci/tc86c001.c b/drivers/ide/pci/tc86c001.c
index bb3292199356..102cd7c40cdd 100644
--- a/drivers/ide/pci/tc86c001.c
+++ b/drivers/ide/pci/tc86c001.c
@@ -232,6 +232,13 @@ out:
 	return rc;
 }
 
+static void __devexit tc86c001_remove(struct pci_dev *dev)
+{
+	ide_pci_remove(dev);
+	pci_release_region(dev, 5);
+	pci_disable_device(dev);
+}
+
 static const struct pci_device_id tc86c001_pci_tbl[] = {
 	{ PCI_VDEVICE(TOSHIBA_2, PCI_DEVICE_ID_TOSHIBA_TC86C001_IDE), 0 },
 	{ 0, }
@@ -241,14 +248,22 @@ MODULE_DEVICE_TABLE(pci, tc86c001_pci_tbl);
 static struct pci_driver driver = {
 	.name		= "TC86C001",
 	.id_table	= tc86c001_pci_tbl,
-	.probe		= tc86c001_init_one
+	.probe		= tc86c001_init_one,
+	.remove		= tc86c001_remove,
 };
 
 static int __init tc86c001_ide_init(void)
 {
 	return ide_pci_register_driver(&driver);
 }
+
+static void __exit tc86c001_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(tc86c001_ide_init);
+module_exit(tc86c001_ide_exit);
 
 MODULE_AUTHOR("MontaVista Software, Inc. <source@mvista.com>");
 MODULE_DESCRIPTION("PCI driver module for TC86C001 IDE");

From 29d72f2df933ea5ecf294b170b2f02af2af88120 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:26 +0200
Subject: [PATCH 0570/1171] triflex: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/triflex.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/ide/pci/triflex.c b/drivers/ide/pci/triflex.c
index 60dcb645d1b0..78e24ac80972 100644
--- a/drivers/ide/pci/triflex.c
+++ b/drivers/ide/pci/triflex.c
@@ -117,6 +117,7 @@ static struct pci_driver driver = {
 	.name		= "TRIFLEX_IDE",
 	.id_table	= triflex_pci_tbl,
 	.probe		= triflex_init_one,
+	.remove		= ide_pci_remove,
 };
 
 static int __init triflex_ide_init(void)
@@ -124,7 +125,13 @@ static int __init triflex_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit triflex_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(triflex_ide_init);
+module_exit(triflex_ide_exit);
 
 MODULE_AUTHOR("Torben Mathiasen");
 MODULE_DESCRIPTION("PCI driver module for Compaq Triflex IDE");

From fc2c32b737fa370683f8c44d74f41febe33b9c23 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:26 +0200
Subject: [PATCH 0571/1171] trm290: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/trm290.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/ide/pci/trm290.c b/drivers/ide/pci/trm290.c
index d8127b51a542..449f4de9a702 100644
--- a/drivers/ide/pci/trm290.c
+++ b/drivers/ide/pci/trm290.c
@@ -353,6 +353,7 @@ static struct pci_driver driver = {
 	.name		= "TRM290_IDE",
 	.id_table	= trm290_pci_tbl,
 	.probe		= trm290_init_one,
+	.remove		= ide_pci_remove,
 };
 
 static int __init trm290_ide_init(void)
@@ -360,7 +361,13 @@ static int __init trm290_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit trm290_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(trm290_ide_init);
+module_exit(trm290_ide_exit);
 
 MODULE_AUTHOR("Mark Lord");
 MODULE_DESCRIPTION("PCI driver module for Tekram TRM290 IDE");

From 585f67e736eece4cdf96b628042170273221e770 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:26 +0200
Subject: [PATCH 0572/1171] via82cxxx: add ->remove method and module_exit()

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/via82cxxx.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/drivers/ide/pci/via82cxxx.c b/drivers/ide/pci/via82cxxx.c
index bc1675da37d1..5f07eeb736ab 100644
--- a/drivers/ide/pci/via82cxxx.c
+++ b/drivers/ide/pci/via82cxxx.c
@@ -467,6 +467,15 @@ static int __devinit via_init_one(struct pci_dev *dev, const struct pci_device_i
 	return rc;
 }
 
+static void __devexit via_remove(struct pci_dev *dev)
+{
+	struct ide_host *host = pci_get_drvdata(dev);
+	struct via82cxxx_dev *vdev = host->host_priv;
+
+	ide_pci_remove(dev);
+	kfree(vdev);
+}
+
 static const struct pci_device_id via_pci_tbl[] = {
 	{ PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_82C576_1),  0 },
 	{ PCI_VDEVICE(VIA, PCI_DEVICE_ID_VIA_82C586_1),  0 },
@@ -481,6 +490,7 @@ static struct pci_driver driver = {
 	.name 		= "VIA_IDE",
 	.id_table 	= via_pci_tbl,
 	.probe 		= via_init_one,
+	.remove		= via_remove,
 };
 
 static int __init via_ide_init(void)
@@ -488,7 +498,13 @@ static int __init via_ide_init(void)
 	return ide_pci_register_driver(&driver);
 }
 
+static void __exit via_ide_exit(void)
+{
+	pci_unregister_driver(&driver);
+}
+
 module_init(via_ide_init);
+module_exit(via_ide_exit);
 
 MODULE_AUTHOR("Vojtech Pavlik, Michel Aubry, Jeff Garzik, Andre Hedrick");
 MODULE_DESCRIPTION("PCI driver module for VIA IDE");

From 8e27cb1135de4cc69bf358209f91e1f7ba81eca1 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:27 +0200
Subject: [PATCH 0573/1171] icside: add module_exit()

Cc: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/arm/icside.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/ide/arm/icside.c b/drivers/ide/arm/icside.c
index f575e8341aec..df4af4083954 100644
--- a/drivers/ide/arm/icside.c
+++ b/drivers/ide/arm/icside.c
@@ -710,8 +710,14 @@ static int __init icside_init(void)
 	return ecard_register_driver(&icside_driver);
 }
 
+static void __exit icside_exit(void);
+{
+	ecard_unregister_driver(&icside_driver);
+}
+
 MODULE_AUTHOR("Russell King <rmk@arm.linux.org.uk>");
 MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("ICS IDE driver");
 
 module_init(icside_init);
+module_exit(icside_exit);

From 37c5ef56989717d871d048f98fb6411e7a17c43d Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:27 +0200
Subject: [PATCH 0574/1171] rapide: add module_exit()

Cc: Russell King <rmk@arm.linux.org.uk>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/arm/rapide.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/ide/arm/rapide.c b/drivers/ide/arm/rapide.c
index 2bdd8b734afb..78d27d9ae430 100644
--- a/drivers/ide/arm/rapide.c
+++ b/drivers/ide/arm/rapide.c
@@ -95,7 +95,13 @@ static int __init rapide_init(void)
 	return ecard_register_driver(&rapide_driver);
 }
 
+static void __exit rapide_exit(void)
+{
+	ecard_unregister_driver(&rapide_driver);
+}
+
 MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("Yellowstone RAPIDE driver");
 
 module_init(rapide_init);
+module_exit(rapide_exit);

From b0a62817961796f6dcef5f316134d8bc7279bf6e Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:27 +0200
Subject: [PATCH 0575/1171] ide: fix <asm-xtensa/ide.h>

* Add missing <asm-generic/ide_iops.h> include.

While at it:

* Remove needless ide_default_{irq,io_base}() inlines.

Cc: Chris Zankel <chris@zankel.net>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 include/asm-xtensa/ide.h | 13 ++-----------
 1 file changed, 2 insertions(+), 11 deletions(-)

diff --git a/include/asm-xtensa/ide.h b/include/asm-xtensa/ide.h
index 6b912742a42d..cb995701c42d 100644
--- a/include/asm-xtensa/ide.h
+++ b/include/asm-xtensa/ide.h
@@ -19,17 +19,8 @@
 # define MAX_HWIFS	1
 #endif
 
-static __inline__ int ide_default_irq(unsigned long base)
-{
-	/* Unsupported! */
-  	return 0;
-}
-
-static __inline__ unsigned long ide_default_io_base(int index)
-{
-	/* Unsupported! */
-  	return 0;
-}
+#include <asm-generic/ide_iops.h>
 
 #endif	/* __KERNEL__ */
+
 #endif	/* _XTENSA_IDE_H */

From ac32f3238c1d95a6ebea2c312160dbdbd61bf91c Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:27 +0200
Subject: [PATCH 0576/1171] ide-generic: fix ide_default_io_base() for m32r

Fix ide_default_io_base() to match ide_default_irq().

Cc: Hirokazu Takata <takata@linux-m32r.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 include/asm-m32r/ide.h | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/include/asm-m32r/ide.h b/include/asm-m32r/ide.h
index 1e7f6474d130..72798d624221 100644
--- a/include/asm-m32r/ide.h
+++ b/include/asm-m32r/ide.h
@@ -52,12 +52,20 @@ static __inline__ int ide_default_irq(unsigned long base)
 static __inline__ unsigned long ide_default_io_base(int index)
 {
 	switch (index) {
+#if defined(CONFIG_PLAT_M32700UT) || defined(CONFIG_PLAT_MAPPI2) \
+	|| defined(CONFIG_PLAT_OPSPUT)
+		case 0:	return 0x1f0;
+#elif defined(CONFIG_PLAT_MAPPI3)
+		case 0:	return 0x1f0;
+		case 1:	return 0x170;
+#else
 		case 0:	return 0x1f0;
 		case 1:	return 0x170;
 		case 2: return 0x1e8;
 		case 3: return 0x168;
 		case 4: return 0x1e0;
 		case 5: return 0x160;
+#endif
 		default:
 			return 0;
 	}

From dbdec839c4c2bfc8f2da8e50c06b9947e5ad0394 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:28 +0200
Subject: [PATCH 0577/1171] ide-generic: minor fix for mips

Move ide_probe_legacy() call to ide_generic_init() so it fails
early if necessary and returns the proper error value (nowadays
ide_default_io_base() is used only by ide-generic).

Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-generic.c           | 4 ++++
 include/asm-mips/mach-generic/ide.h | 2 --
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/ide/ide-generic.c b/drivers/ide/ide-generic.c
index 31d98fec775f..567fd843c7ff 100644
--- a/drivers/ide/ide-generic.c
+++ b/drivers/ide/ide-generic.c
@@ -87,6 +87,10 @@ static int __init ide_generic_init(void)
 	unsigned long io_addr;
 	int i, rc;
 
+#ifdef CONFIG_MIPS
+	if (!ide_probe_legacy())
+		return -ENODEV;
+#endif
 	printk(KERN_INFO DRV_NAME ": please use \"probe_mask=0x3f\" module "
 			 "parameter for probing all legacy ISA IDE ports\n");
 
diff --git a/include/asm-mips/mach-generic/ide.h b/include/asm-mips/mach-generic/ide.h
index 0f6c251f5fec..71a01c5aec16 100644
--- a/include/asm-mips/mach-generic/ide.h
+++ b/include/asm-mips/mach-generic/ide.h
@@ -72,8 +72,6 @@ static __inline__ int ide_default_irq(unsigned long base)
 
 static __inline__ unsigned long ide_default_io_base(int index)
 {
-	if (!ide_probe_legacy())
-		return 0;
 	/*
 	 *      If PCI is present then it is not safe to poke around
 	 *      the other legacy IDE ports. Only 0x1f0 and 0x170 are

From b6cd7da5be2522b62bbc48d41b36c828b88e02fe Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:28 +0200
Subject: [PATCH 0578/1171] ide-generic: remove "no_pci_devices()" quirk from
 ide_default_io_base()

Since the decision to probe for ISA ide2-6 is now left to the user
"no_pci_devices()" quirk is no longer needed and may be removed.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 include/asm-mips/mach-generic/ide.h | 18 ++++--------------
 include/asm-x86/ide.h               | 18 ++++--------------
 2 files changed, 8 insertions(+), 28 deletions(-)

diff --git a/include/asm-mips/mach-generic/ide.h b/include/asm-mips/mach-generic/ide.h
index 71a01c5aec16..f34740ee6775 100644
--- a/include/asm-mips/mach-generic/ide.h
+++ b/include/asm-mips/mach-generic/ide.h
@@ -72,23 +72,13 @@ static __inline__ int ide_default_irq(unsigned long base)
 
 static __inline__ unsigned long ide_default_io_base(int index)
 {
-	/*
-	 *      If PCI is present then it is not safe to poke around
-	 *      the other legacy IDE ports. Only 0x1f0 and 0x170 are
-	 *      defined compatibility mode ports for PCI. A user can
-	 *      override this using ide= but we must default safe.
-	 */
-	if (no_pci_devices()) {
-		switch (index) {
-		case 2: return 0x1e8;
-		case 3: return 0x168;
-		case 4: return 0x1e0;
-		case 5: return 0x160;
-		}
-	}
 	switch (index) {
 	case 0: return 0x1f0;
 	case 1: return 0x170;
+	case 2: return 0x1e8;
+	case 3: return 0x168;
+	case 4: return 0x1e0;
+	case 5: return 0x160;
 	default:
 		return 0;
 	}
diff --git a/include/asm-x86/ide.h b/include/asm-x86/ide.h
index cf9c98e5bdb5..34050747f38c 100644
--- a/include/asm-x86/ide.h
+++ b/include/asm-x86/ide.h
@@ -36,23 +36,13 @@ static __inline__ int ide_default_irq(unsigned long base)
 
 static __inline__ unsigned long ide_default_io_base(int index)
 {
-	/*
-	 *	If PCI is present then it is not safe to poke around
-	 *	the other legacy IDE ports. Only 0x1f0 and 0x170 are
-	 *	defined compatibility mode ports for PCI. A user can 
-	 *	override this using ide= but we must default safe.
-	 */
-	if (no_pci_devices()) {
-		switch(index) {
-			case 2: return 0x1e8;
-			case 3: return 0x168;
-			case 4: return 0x1e0;
-			case 5: return 0x160;
-		}
-	}
 	switch (index) {
 		case 0:	return 0x1f0;
 		case 1:	return 0x170;
+		case 2: return 0x1e8;
+		case 3: return 0x168;
+		case 4: return 0x1e0;
+		case 5: return 0x160;
 		default:
 			return 0;
 	}

From 2c9d86438a0104800da2a8ecdc1e27baf38ba6a4 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:29 +0200
Subject: [PATCH 0579/1171] ide: remove <asm-cris/ide.h>

Remove <asm-cris/arch-v{10,32}/ide.h> and <asm-cris/ide.h>.

This has been a broken code for some time now and needs rewrite
to match IDE core code / host driver model anyway.

Cc: Jesper Nilsson <Jesper.Nilsson@axis.com>
Cc: Mikael Starvik <mikael.starvik@axis.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 include/asm-cris/arch-v10/ide.h | 91 ---------------------------------
 include/asm-cris/arch-v32/ide.h | 56 --------------------
 include/asm-cris/ide.h          |  1 -
 3 files changed, 148 deletions(-)
 delete mode 100644 include/asm-cris/arch-v10/ide.h
 delete mode 100644 include/asm-cris/arch-v32/ide.h
 delete mode 100644 include/asm-cris/ide.h

diff --git a/include/asm-cris/arch-v10/ide.h b/include/asm-cris/arch-v10/ide.h
deleted file mode 100644
index 5366e6239328..000000000000
--- a/include/asm-cris/arch-v10/ide.h
+++ /dev/null
@@ -1,91 +0,0 @@
-/*
- *  linux/include/asm-cris/ide.h
- *
- *  Copyright (C) 2000, 2001, 2002  Axis Communications AB
- *
- *  Authors:    Bjorn Wesen
- *
- */
-
-/*
- *  This file contains the ETRAX 100LX specific IDE code.
- */
-
-#ifndef __ASMCRIS_IDE_H
-#define __ASMCRIS_IDE_H
-
-#ifdef __KERNEL__
-
-#include <asm/arch/svinto.h>
-#include <asm/io.h>
-#include <asm-generic/ide_iops.h>
-
-
-/* ETRAX 100 can support 4 IDE busses on the same pins (serialized) */
-
-#define MAX_HWIFS	4
-
-static inline int ide_default_irq(unsigned long base)
-{
-	/* all IDE busses share the same IRQ, number 4.
-	 * this has the side-effect that ide-probe.c will cluster our 4 interfaces
-	 * together in a hwgroup, and will serialize accesses. this is good, because
-	 * we can't access more than one interface at the same time on ETRAX100.
-	 */
-	return 4;
-}
-
-static inline unsigned long ide_default_io_base(int index)
-{
-	/* we have no real I/O base address per interface, since all go through the
-	 * same register. but in a bitfield in that register, we have the i/f number.
-	 * so we can use the io_base to remember that bitfield.
-	 */
-	static const unsigned long io_bases[MAX_HWIFS] = {
-		IO_FIELD(R_ATA_CTRL_DATA, sel, 0),
-		IO_FIELD(R_ATA_CTRL_DATA, sel, 1),
-		IO_FIELD(R_ATA_CTRL_DATA, sel, 2),
-		IO_FIELD(R_ATA_CTRL_DATA, sel, 3)
-	};
-	return io_bases[index];
-}
-
-/* this is called once for each interface, to setup the port addresses. data_port is the result
- * of the ide_default_io_base call above. ctrl_port will be 0, but that is don't care for us.
- */
-
-static inline void ide_init_hwif_ports(hw_regs_t *hw, unsigned long data_port, unsigned long ctrl_port, int *irq)
-{
-	int i;
-
-	/* fill in ports for ATA addresses 0 to 7 */
-	for (i = 0; i <= 7; i++) {
-		hw->io_ports_array[i] = data_port |
-			IO_FIELD(R_ATA_CTRL_DATA, addr, i) |
-			IO_STATE(R_ATA_CTRL_DATA, cs0, active);
-	}
-
-	/* the IDE control register is at ATA address 6, with CS1 active instead of CS0 */
-	hw->io_ports.ctl_addr = data_port |
-			IO_FIELD(R_ATA_CTRL_DATA, addr, 6) |
-			IO_STATE(R_ATA_CTRL_DATA, cs1, active);
-
-	/* whats this for ? */
-	hw->io_ports.irq_addr = 0;
-}
-
-static inline void ide_init_default_hwifs(void)
-{
-	hw_regs_t hw;
-	int index;
-
-	for(index = 0; index < MAX_HWIFS; index++) {
-		ide_init_hwif_ports(&hw, ide_default_io_base(index), 0, NULL);
-		hw.irq = ide_default_irq(ide_default_io_base(index));
-		ide_register_hw(&hw, NULL);
-	}
-}
-
-#endif /* __KERNEL__ */
-
-#endif /* __ASMCRIS_IDE_H */
diff --git a/include/asm-cris/arch-v32/ide.h b/include/asm-cris/arch-v32/ide.h
deleted file mode 100644
index fb9c3627a5b4..000000000000
--- a/include/asm-cris/arch-v32/ide.h
+++ /dev/null
@@ -1,56 +0,0 @@
-/*
- *  linux/include/asm-cris/ide.h
- *
- *  Copyright (C) 2000-2004  Axis Communications AB
- *
- *  Authors:    Bjorn Wesen, Mikael Starvik
- *
- */
-
-/*
- *  This file contains the ETRAX FS specific IDE code.
- */
-
-#ifndef __ASMCRIS_IDE_H
-#define __ASMCRIS_IDE_H
-
-#ifdef __KERNEL__
-
-#include <asm/arch/hwregs/intr_vect.h>
-#include <asm/arch/hwregs/ata_defs.h>
-#include <asm/io.h>
-#include <asm-generic/ide_iops.h>
-
-
-/* ETRAX FS can support 4 IDE busses on the same pins (serialized) */
-
-#define MAX_HWIFS	4
-
-static inline int ide_default_irq(unsigned long base)
-{
-	/* all IDE busses share the same IRQ,
-	 * this has the side-effect that ide-probe.c will cluster our 4 interfaces
-	 * together in a hwgroup, and will serialize accesses. this is good, because
-	 * we can't access more than one interface at the same time on ETRAX100.
-	 */
-	return ATA_INTR_VECT;
-}
-
-static inline unsigned long ide_default_io_base(int index)
-{
-	reg_ata_rw_ctrl2 ctrl2 = {.sel = index};
-	/* we have no real I/O base address per interface, since all go through the
-	 * same register. but in a bitfield in that register, we have the i/f number.
-	 * so we can use the io_base to remember that bitfield.
-	 */
-        ctrl2.sel = index;
-
-	return REG_TYPE_CONV(unsigned long, reg_ata_rw_ctrl2, ctrl2);
-}
-
-#define IDE_ARCH_ACK_INTR
-#define ide_ack_intr(hwif)	((hwif)->ack_intr(hwif))
-
-#endif /* __KERNEL__ */
-
-#endif /* __ASMCRIS_IDE_H */
diff --git a/include/asm-cris/ide.h b/include/asm-cris/ide.h
deleted file mode 100644
index a894f66665f8..000000000000
--- a/include/asm-cris/ide.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm/arch/ide.h>

From d83b8b85cd56a083d30df73f3fd5e4714591b910 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:30 +0200
Subject: [PATCH 0580/1171] ide: define MAX_HWIFS in <linux/ide.h>

* Now that ide_hwif_t instances are allocated dynamically
  the difference between MAX_HWIFS == 2 and MAX_HWIFS == 10
  is ~100 bytes (x86-32) so use MAX_HWIFS == 10 on all archs
  except these ones that use MAX_HWIFS == 1.

* Define MAX_HWIFS in <linux/ide.h> instead of <asm/ide.h>.

[ Please note that avr32/cris/v850 have no <asm/ide.h>
  and alpha/ia64/sh always define CONFIG_IDE_MAX_HWIFS. ]

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 include/asm-arm/ide.h               | 4 ----
 include/asm-blackfin/ide.h          | 2 --
 include/asm-frv/ide.h               | 4 ----
 include/asm-h8300/ide.h             | 2 --
 include/asm-m32r/ide.h              | 8 --------
 include/asm-m68k/ide.h              | 4 ----
 include/asm-mips/mach-generic/ide.h | 8 --------
 include/asm-mn10300/ide.h           | 4 ----
 include/asm-parisc/ide.h            | 4 ----
 include/asm-powerpc/ide.h           | 8 --------
 include/asm-sparc/ide.h             | 3 ---
 include/asm-x86/ide.h               | 9 ---------
 include/asm-xtensa/ide.h            | 5 -----
 include/linux/ide.h                 | 8 ++++++++
 14 files changed, 8 insertions(+), 65 deletions(-)

diff --git a/include/asm-arm/ide.h b/include/asm-arm/ide.h
index 88f4d231ce4f..a48019f99d08 100644
--- a/include/asm-arm/ide.h
+++ b/include/asm-arm/ide.h
@@ -13,10 +13,6 @@
 
 #ifdef __KERNEL__
 
-#ifndef MAX_HWIFS
-#define MAX_HWIFS	4
-#endif
-
 #define __ide_mm_insw(port,addr,len)	readsw(port,addr,len)
 #define __ide_mm_insl(port,addr,len)	readsl(port,addr,len)
 #define __ide_mm_outsw(port,addr,len)	writesw(port,addr,len)
diff --git a/include/asm-blackfin/ide.h b/include/asm-blackfin/ide.h
index 5b88de115bf4..90bc50bd22e9 100644
--- a/include/asm-blackfin/ide.h
+++ b/include/asm-blackfin/ide.h
@@ -17,8 +17,6 @@
 #ifdef __KERNEL__
 /****************************************************************************/
 
-#define MAX_HWIFS	1
-
 #include <asm-generic/ide_iops.h>
 
 /****************************************************************************/
diff --git a/include/asm-frv/ide.h b/include/asm-frv/ide.h
index 8c9a540d4344..7ebcc56a2229 100644
--- a/include/asm-frv/ide.h
+++ b/include/asm-frv/ide.h
@@ -18,10 +18,6 @@
 #include <asm/io.h>
 #include <asm/irq.h>
 
-#ifndef MAX_HWIFS
-#define MAX_HWIFS 8
-#endif
-
 /****************************************************************************/
 /*
  * some bits needed for parts of the IDE subsystem to compile
diff --git a/include/asm-h8300/ide.h b/include/asm-h8300/ide.h
index f8535ce7476e..8f79ba2ff929 100644
--- a/include/asm-h8300/ide.h
+++ b/include/asm-h8300/ide.h
@@ -16,8 +16,6 @@
 #ifdef __KERNEL__
 /****************************************************************************/
 
-#define MAX_HWIFS	1
-
 #include <asm-generic/ide_iops.h>
 
 /****************************************************************************/
diff --git a/include/asm-m32r/ide.h b/include/asm-m32r/ide.h
index 72798d624221..d755d41b9931 100644
--- a/include/asm-m32r/ide.h
+++ b/include/asm-m32r/ide.h
@@ -15,14 +15,6 @@
 
 #include <asm/m32r.h>
 
-#ifndef MAX_HWIFS
-# ifdef CONFIG_BLK_DEV_IDEPCI
-#define MAX_HWIFS	10
-# else
-#define MAX_HWIFS	2
-# endif
-#endif
-
 static __inline__ int ide_default_irq(unsigned long base)
 {
 	switch (base) {
diff --git a/include/asm-m68k/ide.h b/include/asm-m68k/ide.h
index 909c6dfd3851..1daf6cbdd9f0 100644
--- a/include/asm-m68k/ide.h
+++ b/include/asm-m68k/ide.h
@@ -45,10 +45,6 @@
 #include <asm/macints.h>
 #endif
 
-#ifndef MAX_HWIFS
-#define MAX_HWIFS	4	/* same as the other archs */
-#endif
-
 /*
  * Get rid of defs from io.h - ide has its private and conflicting versions
  * Since so far no single m68k platform uses ISA/PCI I/O space for IDE, we
diff --git a/include/asm-mips/mach-generic/ide.h b/include/asm-mips/mach-generic/ide.h
index f34740ee6775..8ee6bff030d9 100644
--- a/include/asm-mips/mach-generic/ide.h
+++ b/include/asm-mips/mach-generic/ide.h
@@ -19,14 +19,6 @@
 #include <linux/stddef.h>
 #include <asm/processor.h>
 
-#ifndef MAX_HWIFS
-# ifdef CONFIG_BLK_DEV_IDEPCI
-#define MAX_HWIFS	10
-# else
-#define MAX_HWIFS	6
-# endif
-#endif
-
 static __inline__ int ide_probe_legacy(void)
 {
 #ifdef CONFIG_PCI
diff --git a/include/asm-mn10300/ide.h b/include/asm-mn10300/ide.h
index dc235121ec42..6adcdd92e83d 100644
--- a/include/asm-mn10300/ide.h
+++ b/include/asm-mn10300/ide.h
@@ -23,10 +23,6 @@
 #undef SUPPORT_VLB_SYNC
 #define SUPPORT_VLB_SYNC 0
 
-#ifndef MAX_HWIFS
-#define MAX_HWIFS 8
-#endif
-
 /*
  * some bits needed for parts of the IDE subsystem to compile
  */
diff --git a/include/asm-parisc/ide.h b/include/asm-parisc/ide.h
index db0c94410095..c246ef75017d 100644
--- a/include/asm-parisc/ide.h
+++ b/include/asm-parisc/ide.h
@@ -13,10 +13,6 @@
 
 #ifdef __KERNEL__
 
-#ifndef MAX_HWIFS
-#define MAX_HWIFS	2
-#endif
-
 #define ide_request_irq(irq,hand,flg,dev,id)	request_irq((irq),(hand),(flg),(dev),(id))
 #define ide_free_irq(irq,dev_id)		free_irq((irq), (dev_id))
 #define ide_request_region(from,extent,name)	request_region((from), (extent), (name))
diff --git a/include/asm-powerpc/ide.h b/include/asm-powerpc/ide.h
index 3d90bf7d3d73..262def6a9f0a 100644
--- a/include/asm-powerpc/ide.h
+++ b/include/asm-powerpc/ide.h
@@ -14,14 +14,6 @@
 #endif
 #include <asm/io.h>
 
-#ifndef MAX_HWIFS
-#ifdef __powerpc64__
-#define MAX_HWIFS	10
-#else
-#define MAX_HWIFS	8
-#endif
-#endif
-
 #define __ide_mm_insw(p, a, c)	readsw((void __iomem *)(p), (a), (c))
 #define __ide_mm_insl(p, a, c)	readsl((void __iomem *)(p), (a), (c))
 #define __ide_mm_outsw(p, a, c)	writesw((void __iomem *)(p), (a), (c))
diff --git a/include/asm-sparc/ide.h b/include/asm-sparc/ide.h
index 879fcec72dc1..b7af3d658239 100644
--- a/include/asm-sparc/ide.h
+++ b/include/asm-sparc/ide.h
@@ -21,9 +21,6 @@
 #include <asm/psr.h>
 #endif
 
-#undef  MAX_HWIFS
-#define MAX_HWIFS	2
-
 #define __ide_insl(data_reg, buffer, wcount) \
 	__ide_insw(data_reg, buffer, (wcount)<<1)
 #define __ide_outsl(data_reg, buffer, wcount) \
diff --git a/include/asm-x86/ide.h b/include/asm-x86/ide.h
index 34050747f38c..bc54879daed1 100644
--- a/include/asm-x86/ide.h
+++ b/include/asm-x86/ide.h
@@ -11,15 +11,6 @@
 
 #ifdef __KERNEL__
 
-
-#ifndef MAX_HWIFS
-# ifdef CONFIG_BLK_DEV_IDEPCI
-#define MAX_HWIFS	10
-# else
-#define MAX_HWIFS	6
-# endif
-#endif
-
 static __inline__ int ide_default_irq(unsigned long base)
 {
 	switch (base) {
diff --git a/include/asm-xtensa/ide.h b/include/asm-xtensa/ide.h
index cb995701c42d..18342a2cc774 100644
--- a/include/asm-xtensa/ide.h
+++ b/include/asm-xtensa/ide.h
@@ -14,11 +14,6 @@
 
 #ifdef __KERNEL__
 
-
-#ifndef MAX_HWIFS
-# define MAX_HWIFS	1
-#endif
-
 #include <asm-generic/ide_iops.h>
 
 #endif	/* __KERNEL__ */
diff --git a/include/linux/ide.h b/include/linux/ide.h
index dbd0aeb3a56d..76fe00b24b51 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -213,6 +213,14 @@ static inline int __ide_default_irq(unsigned long base)
 
 #include <asm/ide.h>
 
+#ifndef MAX_HWIFS
+#if defined(CONFIG_BLACKFIN) || defined(CONFIG_H8300) || defined(CONFIG_XTENSA)
+# define MAX_HWIFS	1
+#else
+# define MAX_HWIFS	10
+#endif
+#endif
+
 #if !defined(MAX_HWIFS) || defined(CONFIG_EMBEDDED)
 #undef MAX_HWIFS
 #define MAX_HWIFS	CONFIG_IDE_MAX_HWIFS

From ffed0b6e1a6f5132681d4b521531d992f893190b Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:30 +0200
Subject: [PATCH 0581/1171] ide-generic: remove broken PPC_PREP support

PPC_PREP has been depending on BROKEN for some time now.

Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 include/asm-powerpc/ide.h | 18 ------------------
 1 file changed, 18 deletions(-)

diff --git a/include/asm-powerpc/ide.h b/include/asm-powerpc/ide.h
index 262def6a9f0a..1aaf27be8741 100644
--- a/include/asm-powerpc/ide.h
+++ b/include/asm-powerpc/ide.h
@@ -31,16 +31,6 @@ static __inline__ int ide_default_irq(unsigned long base)
 	case 0x1f0:	return 14;
 	case 0x170:	return 15;
 	}
-#endif
-#ifdef CONFIG_PPC_PREP
-	switch (base) {
-	case 0x1f0:	return 13;
-	case 0x170:	return 13;
-	case 0x1e8:	return 11;
-	case 0x168:	return 10;
-	case 0xfff0:	return 14;	/* MCP(N)750 ide0 */
-	case 0xffe0:	return 15;	/* MCP(N)750 ide1 */
-	}
 #endif
 	return 0;
 }
@@ -53,14 +43,6 @@ static __inline__ unsigned long ide_default_io_base(int index)
 	case 0:		return 0x1f0;
 	case 1:		return 0x170;
 	}
-#endif
-#ifdef CONFIG_PPC_PREP
-	switch (index) {
-	case 0:		return 0x1f0;
-	case 1:		return 0x170;
-	case 2:		return 0x1e8;
-	case 3:		return 0x168;
-	}
 #endif
 	return 0;
 }

From 35bbac9a2f73a7e0967d0a1d3e3673e2590ef716 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:30 +0200
Subject: [PATCH 0582/1171] ide-generic: is no longer needed on ppc32

Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/ide/Kconfig b/drivers/ide/Kconfig
index 04d9c4d459d0..130ef64b44f7 100644
--- a/drivers/ide/Kconfig
+++ b/drivers/ide/Kconfig
@@ -314,7 +314,7 @@ comment "IDE chipset support/bugfixes"
 
 config IDE_GENERIC
 	tristate "generic/default IDE chipset support"
-	depends on ALPHA || X86 || IA64 || M32R || MIPS || PPC32
+	depends on ALPHA || X86 || IA64 || M32R || MIPS
 	help
 	  If unsure, say N.
 

From f01d35d87f39ab794ddcdefadb79c11054bcbfbc Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:31 +0200
Subject: [PATCH 0583/1171] ide-generic: remove ide_default_{io_base,irq}()
 inlines (take 3)

Replace ide_default_{io_base,irq}() inlines by legacy_{bases,irqs}[].

v2:
Add missing zero-ing of hws[] (caught during testing by Borislav Petkov).

v3:
Fix zero-oing of hws[] for _real_ this time.

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-generic.c           | 32 ++++++++++++++++--
 include/asm-alpha/ide.h             | 24 --------------
 include/asm-ia64/ide.h              | 29 -----------------
 include/asm-m32r/ide.h              | 50 -----------------------------
 include/asm-mips/mach-generic/ide.h | 28 ----------------
 include/asm-x86/ide.h               | 28 ----------------
 6 files changed, 29 insertions(+), 162 deletions(-)

diff --git a/drivers/ide/ide-generic.c b/drivers/ide/ide-generic.c
index 567fd843c7ff..8fe8b5b9cf7d 100644
--- a/drivers/ide/ide-generic.c
+++ b/drivers/ide/ide-generic.c
@@ -20,6 +20,11 @@
 #include <linux/module.h>
 #include <linux/ide.h>
 
+/* FIXME: convert m32r to use ide_platform host driver */
+#ifdef CONFIG_M32R
+#include <asm/m32r.h>
+#endif
+
 #define DRV_NAME	"ide_generic"
 
 static int probe_mask = 0x03;
@@ -80,6 +85,21 @@ static int __init ide_generic_sysfs_init(void)
 	return 0;
 }
 
+#if defined(CONFIG_PLAT_M32700UT) || defined(CONFIG_PLAT_MAPPI2) \
+	|| defined(CONFIG_PLAT_OPSPUT)
+static const u16 legacy_bases[] = { 0x1f0 };
+static const int legacy_irqs[]  = { PLD_IRQ_CFIREQ };
+#elif defined(CONFIG_PLAT_MAPPI3)
+static const u16 legacy_bases[] = { 0x1f0, 0x170 };
+static const int legacy_irqs[]  = { PLD_IRQ_CFIREQ, PLD_IRQ_IDEIREQ };
+#elif defined(CONFIG_ALPHA)
+static const u16 legacy_bases[] = { 0x1f0, 0x170, 0x1e8, 0x168 };
+static const int legacy_irqs[]  = { 14, 15, 11, 10 };
+#else
+static const u16 legacy_bases[] = { 0x1f0, 0x170, 0x1e8, 0x168, 0x1e0, 0x160 };
+static const int legacy_irqs[]  = { 14, 15, 11, 10, 8, 12 };
+#endif
+
 static int __init ide_generic_init(void)
 {
 	hw_regs_t hw[MAX_HWIFS], *hws[MAX_HWIFS];
@@ -94,8 +114,10 @@ static int __init ide_generic_init(void)
 	printk(KERN_INFO DRV_NAME ": please use \"probe_mask=0x3f\" module "
 			 "parameter for probing all legacy ISA IDE ports\n");
 
-	for (i = 0; i < MAX_HWIFS; i++) {
-		io_addr = ide_default_io_base(i);
+	memset(hws, 0, sizeof(hw_regs_t *) * MAX_HWIFS);
+
+	for (i = 0; i < ARRAY_SIZE(legacy_bases); i++) {
+		io_addr = legacy_bases[i];
 
 		hws[i] = NULL;
 
@@ -117,7 +139,11 @@ static int __init ide_generic_init(void)
 
 			memset(&hw[i], 0, sizeof(hw[i]));
 			ide_std_init_ports(&hw[i], io_addr, io_addr + 0x206);
-			hw[i].irq = ide_default_irq(io_addr);
+#ifdef CONFIG_IA64
+			hw[i].irq = isa_irq_to_vector(legacy_irqs[i]);
+#else
+			hw[i].irq = legacy_irqs[i];
+#endif
 			hw[i].chipset = ide_generic;
 
 			hws[i] = &hw[i];
diff --git a/include/asm-alpha/ide.h b/include/asm-alpha/ide.h
index f44129abc02c..55f9f6870249 100644
--- a/include/asm-alpha/ide.h
+++ b/include/asm-alpha/ide.h
@@ -13,30 +13,6 @@
 
 #ifdef __KERNEL__
 
-static inline int ide_default_irq(unsigned long base)
-{
-	switch (base) {
-		case 0x1f0: return 14;
-		case 0x170: return 15;
-		case 0x1e8: return 11;
-		case 0x168: return 10;
-		default:
-			return 0;
-	}
-}
-
-static inline unsigned long ide_default_io_base(int index)
-{
-	switch (index) {
-		case 0:	return 0x1f0;
-		case 1:	return 0x170;
-		case 2: return 0x1e8;
-		case 3: return 0x168;
-		default:
-			return 0;
-	}
-}
-
 #include <asm-generic/ide_iops.h>
 
 #endif /* __KERNEL__ */
diff --git a/include/asm-ia64/ide.h b/include/asm-ia64/ide.h
index 8fa3f8cd067a..5a0aedea4764 100644
--- a/include/asm-ia64/ide.h
+++ b/include/asm-ia64/ide.h
@@ -13,37 +13,8 @@
 
 #ifdef __KERNEL__
 
-
 #include <linux/irq.h>
 
-static inline int ide_default_irq(unsigned long base)
-{
-	switch (base) {
-	      case 0x1f0: return isa_irq_to_vector(14);
-	      case 0x170: return isa_irq_to_vector(15);
-	      case 0x1e8: return isa_irq_to_vector(11);
-	      case 0x168: return isa_irq_to_vector(10);
-	      case 0x1e0: return isa_irq_to_vector(8);
-	      case 0x160: return isa_irq_to_vector(12);
-	      default:
-		return 0;
-	}
-}
-
-static inline unsigned long ide_default_io_base(int index)
-{
-	switch (index) {
-	      case 0: return 0x1f0;
-	      case 1: return 0x170;
-	      case 2: return 0x1e8;
-	      case 3: return 0x168;
-	      case 4: return 0x1e0;
-	      case 5: return 0x160;
-	      default:
-		return 0;
-	}
-}
-
 #include <asm-generic/ide_iops.h>
 
 #endif /* __KERNEL__ */
diff --git a/include/asm-m32r/ide.h b/include/asm-m32r/ide.h
index d755d41b9931..0f1ec6973879 100644
--- a/include/asm-m32r/ide.h
+++ b/include/asm-m32r/ide.h
@@ -13,56 +13,6 @@
 
 #ifdef __KERNEL__
 
-#include <asm/m32r.h>
-
-static __inline__ int ide_default_irq(unsigned long base)
-{
-	switch (base) {
-#if defined(CONFIG_PLAT_M32700UT) || defined(CONFIG_PLAT_MAPPI2) \
-	|| defined(CONFIG_PLAT_OPSPUT)
-		case 0x1f0: return PLD_IRQ_CFIREQ;
-		default:
-			return 0;
-#elif defined(CONFIG_PLAT_MAPPI3)
-		case 0x1f0: return PLD_IRQ_CFIREQ;
-		case 0x170: return PLD_IRQ_IDEIREQ;
-		default:
-			return 0;
-#else
-		case 0x1f0: return 14;
-		case 0x170: return 15;
-		case 0x1e8: return 11;
-		case 0x168: return 10;
-		case 0x1e0: return 8;
-		case 0x160: return 12;
-		default:
-			return 0;
-#endif
-	}
-}
-
-static __inline__ unsigned long ide_default_io_base(int index)
-{
-	switch (index) {
-#if defined(CONFIG_PLAT_M32700UT) || defined(CONFIG_PLAT_MAPPI2) \
-	|| defined(CONFIG_PLAT_OPSPUT)
-		case 0:	return 0x1f0;
-#elif defined(CONFIG_PLAT_MAPPI3)
-		case 0:	return 0x1f0;
-		case 1:	return 0x170;
-#else
-		case 0:	return 0x1f0;
-		case 1:	return 0x170;
-		case 2: return 0x1e8;
-		case 3: return 0x168;
-		case 4: return 0x1e0;
-		case 5: return 0x160;
-#endif
-		default:
-			return 0;
-	}
-}
-
 #include <asm-generic/ide_iops.h>
 
 #endif /* __KERNEL__ */
diff --git a/include/asm-mips/mach-generic/ide.h b/include/asm-mips/mach-generic/ide.h
index 8ee6bff030d9..73008f7bdc93 100644
--- a/include/asm-mips/mach-generic/ide.h
+++ b/include/asm-mips/mach-generic/ide.h
@@ -48,34 +48,6 @@ found:
 #endif
 }
 
-static __inline__ int ide_default_irq(unsigned long base)
-{
-	switch (base) {
-		case 0x1f0: return 14;
-		case 0x170: return 15;
-		case 0x1e8: return 11;
-		case 0x168: return 10;
-		case 0x1e0: return 8;
-		case 0x160: return 12;
-		default:
-			return 0;
-	}
-}
-
-static __inline__ unsigned long ide_default_io_base(int index)
-{
-	switch (index) {
-	case 0: return 0x1f0;
-	case 1: return 0x170;
-	case 2: return 0x1e8;
-	case 3: return 0x168;
-	case 4: return 0x1e0;
-	case 5: return 0x160;
-	default:
-		return 0;
-	}
-}
-
 /* MIPS port and memory-mapped I/O string operations.  */
 static inline void __ide_flush_prologue(void)
 {
diff --git a/include/asm-x86/ide.h b/include/asm-x86/ide.h
index bc54879daed1..0289baf9ce0a 100644
--- a/include/asm-x86/ide.h
+++ b/include/asm-x86/ide.h
@@ -11,34 +11,6 @@
 
 #ifdef __KERNEL__
 
-static __inline__ int ide_default_irq(unsigned long base)
-{
-	switch (base) {
-		case 0x1f0: return 14;
-		case 0x170: return 15;
-		case 0x1e8: return 11;
-		case 0x168: return 10;
-		case 0x1e0: return 8;
-		case 0x160: return 12;
-		default:
-			return 0;
-	}
-}
-
-static __inline__ unsigned long ide_default_io_base(int index)
-{
-	switch (index) {
-		case 0:	return 0x1f0;
-		case 1:	return 0x170;
-		case 2: return 0x1e8;
-		case 3: return 0x168;
-		case 4: return 0x1e0;
-		case 5: return 0x160;
-		default:
-			return 0;
-	}
-}
-
 #include <asm-generic/ide_iops.h>
 
 #endif /* __KERNEL__ */

From 2a8f7450f828eaee49d66f41f99ac2e54f1160a6 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:31 +0200
Subject: [PATCH 0584/1171] ide: remove <asm/ide.h> for some archs

* Remove <linux/irq.h> include from <asm-ia64.h> (<linux/ide.h> includes
  <linux/interrupt.h> which is enough).

* Remove <asm/ide.h> for alpha/blackfin/h8300/ia64/m32r/sh/x86/xtensa
  (this leaves us with arm/frv/m68k/mips/mn10300/parisc/powerpc/sparc[64]).

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 include/asm-alpha/ide.h    | 20 --------------------
 include/asm-blackfin/ide.h | 25 -------------------------
 include/asm-h8300/ide.h    | 24 ------------------------
 include/asm-ia64/ide.h     | 22 ----------------------
 include/asm-m32r/ide.h     | 20 --------------------
 include/asm-sh/ide.h       | 21 ---------------------
 include/asm-x86/ide.h      | 18 ------------------
 include/asm-xtensa/ide.h   | 21 ---------------------
 include/linux/ide.h        |  6 ++++++
 9 files changed, 6 insertions(+), 171 deletions(-)
 delete mode 100644 include/asm-alpha/ide.h
 delete mode 100644 include/asm-blackfin/ide.h
 delete mode 100644 include/asm-h8300/ide.h
 delete mode 100644 include/asm-ia64/ide.h
 delete mode 100644 include/asm-m32r/ide.h
 delete mode 100644 include/asm-sh/ide.h
 delete mode 100644 include/asm-x86/ide.h
 delete mode 100644 include/asm-xtensa/ide.h

diff --git a/include/asm-alpha/ide.h b/include/asm-alpha/ide.h
deleted file mode 100644
index 55f9f6870249..000000000000
--- a/include/asm-alpha/ide.h
+++ /dev/null
@@ -1,20 +0,0 @@
-/*
- *  linux/include/asm-alpha/ide.h
- *
- *  Copyright (C) 1994-1996  Linus Torvalds & authors
- */
-
-/*
- *  This file contains the alpha architecture specific IDE code.
- */
-
-#ifndef __ASMalpha_IDE_H
-#define __ASMalpha_IDE_H
-
-#ifdef __KERNEL__
-
-#include <asm-generic/ide_iops.h>
-
-#endif /* __KERNEL__ */
-
-#endif /* __ASMalpha_IDE_H */
diff --git a/include/asm-blackfin/ide.h b/include/asm-blackfin/ide.h
deleted file mode 100644
index 90bc50bd22e9..000000000000
--- a/include/asm-blackfin/ide.h
+++ /dev/null
@@ -1,25 +0,0 @@
-/****************************************************************************/
-
-/*
- *  linux/include/asm-blackfin/ide.h
- *
- *  Copyright (C) 1994-1996  Linus Torvalds & authors
- *  Copyright (C) 2001       Lineo Inc., davidm@snapgear.com
- *  Copyright (C) 2002       Greg Ungerer (gerg@snapgear.com)
- *  Copyright (C) 2002       Yoshinori Sato (ysato@users.sourceforge.jp)
- *  Copyright (C) 2005       Hennerich Michael (hennerich@blackfin.uclinux.org)
- */
-
-/****************************************************************************/
-#ifndef _BLACKFIN_IDE_H
-#define _BLACKFIN_IDE_H
-/****************************************************************************/
-#ifdef __KERNEL__
-/****************************************************************************/
-
-#include <asm-generic/ide_iops.h>
-
-/****************************************************************************/
-#endif				/* __KERNEL__ */
-#endif				/* _BLACKFIN_IDE_H */
-/****************************************************************************/
diff --git a/include/asm-h8300/ide.h b/include/asm-h8300/ide.h
deleted file mode 100644
index 8f79ba2ff929..000000000000
--- a/include/asm-h8300/ide.h
+++ /dev/null
@@ -1,24 +0,0 @@
-/****************************************************************************/
-
-/*
- *  linux/include/asm-h8300/ide.h
- *
- *  Copyright (C) 1994-1996  Linus Torvalds & authors
- *  Copyright (C) 2001       Lineo Inc., davidm@snapgear.com
- *  Copyright (C) 2002       Greg Ungerer (gerg@snapgear.com)
- *  Copyright (C) 2002       Yoshinori Sato (ysato@users.sourceforge.jp)
- */
-
-/****************************************************************************/
-#ifndef _H8300_IDE_H
-#define _H8300_IDE_H
-/****************************************************************************/
-#ifdef __KERNEL__
-/****************************************************************************/
-
-#include <asm-generic/ide_iops.h>
-
-/****************************************************************************/
-#endif /* __KERNEL__ */
-#endif /* _H8300_IDE_H */
-/****************************************************************************/
diff --git a/include/asm-ia64/ide.h b/include/asm-ia64/ide.h
deleted file mode 100644
index 5a0aedea4764..000000000000
--- a/include/asm-ia64/ide.h
+++ /dev/null
@@ -1,22 +0,0 @@
-/*
- *  linux/include/asm-ia64/ide.h
- *
- *  Copyright (C) 1994-1996  Linus Torvalds & authors
- */
-
-/*
- *  This file contains the ia64 architecture specific IDE code.
- */
-
-#ifndef __ASM_IA64_IDE_H
-#define __ASM_IA64_IDE_H
-
-#ifdef __KERNEL__
-
-#include <linux/irq.h>
-
-#include <asm-generic/ide_iops.h>
-
-#endif /* __KERNEL__ */
-
-#endif /* __ASM_IA64_IDE_H */
diff --git a/include/asm-m32r/ide.h b/include/asm-m32r/ide.h
deleted file mode 100644
index 0f1ec6973879..000000000000
--- a/include/asm-m32r/ide.h
+++ /dev/null
@@ -1,20 +0,0 @@
-#ifndef _ASM_M32R_IDE_H
-#define _ASM_M32R_IDE_H
-
-/*
- *  linux/include/asm-m32r/ide.h
- *
- *  Copyright (C) 1994-1996  Linus Torvalds & authors
- */
-
-/*
- *  This file contains the i386 architecture specific IDE code.
- */
-
-#ifdef __KERNEL__
-
-#include <asm-generic/ide_iops.h>
-
-#endif /* __KERNEL__ */
-
-#endif /* _ASM_M32R_IDE_H */
diff --git a/include/asm-sh/ide.h b/include/asm-sh/ide.h
deleted file mode 100644
index 58e0bdd52be4..000000000000
--- a/include/asm-sh/ide.h
+++ /dev/null
@@ -1,21 +0,0 @@
-/*
- *  linux/include/asm-sh/ide.h
- *
- *  Copyright (C) 1994-1996  Linus Torvalds & authors
- */
-
-/*
- *  This file contains the i386 architecture specific IDE code.
- *  In future, SuperH code.
- */
-
-#ifndef __ASM_SH_IDE_H
-#define __ASM_SH_IDE_H
-
-#ifdef __KERNEL__
-
-#include <asm-generic/ide_iops.h>
-
-#endif /* __KERNEL__ */
-
-#endif /* __ASM_SH_IDE_H */
diff --git a/include/asm-x86/ide.h b/include/asm-x86/ide.h
deleted file mode 100644
index 0289baf9ce0a..000000000000
--- a/include/asm-x86/ide.h
+++ /dev/null
@@ -1,18 +0,0 @@
-/*
- *  Copyright (C) 1994-1996  Linus Torvalds & authors
- */
-
-/*
- *  This file contains the i386 architecture specific IDE code.
- */
-
-#ifndef __ASMi386_IDE_H
-#define __ASMi386_IDE_H
-
-#ifdef __KERNEL__
-
-#include <asm-generic/ide_iops.h>
-
-#endif /* __KERNEL__ */
-
-#endif /* __ASMi386_IDE_H */
diff --git a/include/asm-xtensa/ide.h b/include/asm-xtensa/ide.h
deleted file mode 100644
index 18342a2cc774..000000000000
--- a/include/asm-xtensa/ide.h
+++ /dev/null
@@ -1,21 +0,0 @@
-/*
- * include/asm-xtensa/ide.h
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 1994 - 1996  Linus Torvalds & authors
- * Copyright (C) 2001 - 2005 Tensilica Inc.
- */
-
-#ifndef _XTENSA_IDE_H
-#define _XTENSA_IDE_H
-
-#ifdef __KERNEL__
-
-#include <asm-generic/ide_iops.h>
-
-#endif	/* __KERNEL__ */
-
-#endif	/* _XTENSA_IDE_H */
diff --git a/include/linux/ide.h b/include/linux/ide.h
index 76fe00b24b51..fd78b401b036 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -211,7 +211,13 @@ static inline int __ide_default_irq(unsigned long base)
 	return 0;
 }
 
+#if defined(CONFIG_ARM) || defined(CONFIG_FRV) || defined(CONFIG_M68K) || \
+    defined(CONFIG_MIPS) || defined(CONFIG_MN10300) || defined(CONFIG_PARISC) \
+    || defined(CONFIG_PPC) || defined(CONFIG_SPARC) || defined(CONFIG_SPARC64)
 #include <asm/ide.h>
+#else
+#include <asm-generic/ide_iops.h>
+#endif
 
 #ifndef MAX_HWIFS
 #if defined(CONFIG_BLACKFIN) || defined(CONFIG_H8300) || defined(CONFIG_XTENSA)

From 28cfd8af52a9ed4e5bd1751ea6bc0b8c870f68ec Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:31 +0200
Subject: [PATCH 0585/1171] ide: include PCI device name in messages from IDE
 PCI host drivers

While at it:

* Apply small fixes to messages (s/dma/DMA/, remove trailing '.', etc).

* Fix printk() call in ide_setup_pci_baseregs() to use KERN_INFO.

* Move printk() call from ide_pci_clear_simplex() to the caller.

* Cleanup do_ide_setup_pci_device() a bit.

* amd74xx.c: remove superfluous PCI device revision information.

* hpt366.c: fix two printk() calls in ->init_chipset to use KERN_INFO.

* pdc202xx_new.c: fix printk() call in ->init_chipset to use KERN_INFO.

* pdc202xx_old.c: fix driver message in pdc202xx_init_one().

* via82cxxx.c: fix driver warning message in via_init_one().

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/amd74xx.c      | 11 ++--
 drivers/ide/pci/generic.c      |  4 +-
 drivers/ide/pci/hpt366.c       | 31 ++++++-----
 drivers/ide/pci/it821x.c       | 11 ++--
 drivers/ide/pci/pdc202xx_new.c | 21 ++++----
 drivers/ide/pci/pdc202xx_old.c |  9 ++--
 drivers/ide/pci/serverworks.c  |  3 +-
 drivers/ide/pci/siimage.c      |  7 +--
 drivers/ide/pci/sis5513.c      | 17 +++---
 drivers/ide/pci/trm290.c       |  4 +-
 drivers/ide/pci/via82cxxx.c    | 13 +++--
 drivers/ide/setup-pci.c        | 99 +++++++++++++++++++---------------
 12 files changed, 131 insertions(+), 99 deletions(-)

diff --git a/drivers/ide/pci/amd74xx.c b/drivers/ide/pci/amd74xx.c
index a91f2e873ba4..2b118f80fab6 100644
--- a/drivers/ide/pci/amd74xx.c
+++ b/drivers/ide/pci/amd74xx.c
@@ -129,9 +129,9 @@ static void __devinit amd7411_cable_detect(struct pci_dev *dev,
 	amd_80w = ((t & 0x3) ? 1 : 0) | ((t & 0xc) ? 2 : 0);
 	for (i = 24; i >= 0; i -= 8)
 		if (((u >> i) & 4) && !(amd_80w & (1 << (1 - (i >> 4))))) {
-			printk(KERN_WARNING "%s: BIOS didn't set cable bits "
-					    "correctly. Enabling workaround.\n",
-					    name);
+			printk(KERN_WARNING "%s %s: BIOS didn't set cable bits "
+				"correctly. Enabling workaround.\n",
+				name, pci_name(dev));
 			amd_80w |= (1 << (1 - (i >> 4)));
 		}
 }
@@ -280,9 +280,8 @@ static int __devinit amd74xx_probe(struct pci_dev *dev, const struct pci_device_
 			d.udma_mask = ATA_UDMA5;
 	}
 
-	printk(KERN_INFO "%s: %s (rev %02x) UDMA%s controller\n",
-			 d.name, pci_name(dev), dev->revision,
-			 amd_dma[fls(d.udma_mask) - 1]);
+	printk(KERN_INFO "%s %s: UDMA%s controller\n",
+		d.name, pci_name(dev), amd_dma[fls(d.udma_mask) - 1]);
 
 	/*
 	* Determine the system bus clock.
diff --git a/drivers/ide/pci/generic.c b/drivers/ide/pci/generic.c
index 81fc171f8ffc..e034e21af7fe 100644
--- a/drivers/ide/pci/generic.c
+++ b/drivers/ide/pci/generic.c
@@ -134,8 +134,8 @@ static int __devinit generic_init_one(struct pci_dev *dev, const struct pci_devi
 		u16 command;
 		pci_read_config_word(dev, PCI_COMMAND, &command);
 		if (!(command & PCI_COMMAND_IO)) {
-			printk(KERN_INFO "Skipping disabled %s IDE "
-					"controller.\n", d->name);
+			printk(KERN_INFO "%s %s: skipping disabled "
+				"controller\n", d->name, pci_name(dev));
 			goto out;
 		}
 	}
diff --git a/drivers/ide/pci/hpt366.c b/drivers/ide/pci/hpt366.c
index 398808905f94..b8004c331edb 100644
--- a/drivers/ide/pci/hpt366.c
+++ b/drivers/ide/pci/hpt366.c
@@ -1044,8 +1044,8 @@ static unsigned int __devinit init_chipset_hpt366(struct pci_dev *dev, const cha
 		if ((temp & 0xFFFFF000) != 0xABCDE000) {
 			int i;
 
-			printk(KERN_WARNING "%s: no clock data saved by BIOS\n",
-			       name);
+			printk(KERN_WARNING "%s %s: no clock data saved by "
+				"BIOS\n", name, pci_name(dev));
 
 			/* Calculate the average value of f_CNT. */
 			for (temp = i = 0; i < 128; i++) {
@@ -1070,8 +1070,9 @@ static unsigned int __devinit init_chipset_hpt366(struct pci_dev *dev, const cha
 		else
 			pci_clk = 66;
 
-		printk(KERN_INFO "%s: DPLL base: %d MHz, f_CNT: %d, "
-		       "assuming %d MHz PCI\n", name, dpll_clk, f_cnt, pci_clk);
+		printk(KERN_INFO "%s %s: DPLL base: %d MHz, f_CNT: %d, "
+			"assuming %d MHz PCI\n", name, pci_name(dev),
+			dpll_clk, f_cnt, pci_clk);
 	} else {
 		u32 itr1 = 0;
 
@@ -1137,7 +1138,8 @@ static unsigned int __devinit init_chipset_hpt366(struct pci_dev *dev, const cha
 		}
 
 		if (info->timings->clock_table[clock] == NULL) {
-			printk(KERN_ERR "%s: unknown bus timing!\n", name);
+			printk(KERN_ERR "%s %s: unknown bus timing!\n",
+				name, pci_name(dev));
 			return -EIO;
 		}
 
@@ -1163,16 +1165,19 @@ static unsigned int __devinit init_chipset_hpt366(struct pci_dev *dev, const cha
 				f_low += adjust >> 1;
 		}
 		if (adjust == 8) {
-			printk(KERN_ERR "%s: DPLL did not stabilize!\n", name);
+			printk(KERN_ERR "%s %s: DPLL did not stabilize!\n",
+				name, pci_name(dev));
 			return -EIO;
 		}
 
-		printk("%s: using %d MHz DPLL clock\n", name, dpll_clk);
+		printk(KERN_INFO "%s %s: using %d MHz DPLL clock\n",
+			name, pci_name(dev), dpll_clk);
 	} else {
 		/* Mark the fact that we're not using the DPLL. */
 		dpll_clk = 0;
 
-		printk("%s: using %d MHz PCI clock\n", name, pci_clk);
+		printk(KERN_INFO "%s %s: using %d MHz PCI clock\n",
+			name, pci_name(dev), pci_clk);
 	}
 
 	/* Store the clock frequencies. */
@@ -1357,7 +1362,8 @@ static void __devinit hpt374_init(struct pci_dev *dev, struct pci_dev *dev2)
 	if (dev2->irq != dev->irq) {
 		/* FIXME: we need a core pci_set_interrupt() */
 		dev2->irq = dev->irq;
-		printk(KERN_INFO "HPT374: PCI config space interrupt fixed\n");
+		printk(KERN_INFO "HPT374 %s: PCI config space interrupt "
+			"fixed\n", pci_name(dev2));
 	}
 }
 
@@ -1392,8 +1398,8 @@ static int __devinit hpt36x_init(struct pci_dev *dev, struct pci_dev *dev2)
 	pci_read_config_byte(dev2, PCI_INTERRUPT_PIN, &pin2);
 
 	if (pin1 != pin2 && dev->irq == dev2->irq) {
-		printk(KERN_INFO "HPT36x: onboard version of chipset, "
-				 "pin1=%d pin2=%d\n", pin1, pin2);
+		printk(KERN_INFO "HPT36x %s: onboard version of chipset, "
+			"pin1=%d pin2=%d\n", pci_name(dev), pin1, pin2);
 		return 1;
 	}
 
@@ -1591,7 +1597,8 @@ static int __devinit hpt366_init_one(struct pci_dev *dev, const struct pci_devic
 
 	dyn_info = kzalloc(sizeof(*dyn_info) * (dev2 ? 2 : 1), GFP_KERNEL);
 	if (dyn_info == NULL) {
-		printk(KERN_ERR "%s: out of memory!\n", d.name);
+		printk(KERN_ERR "%s %s: out of memory!\n",
+			d.name, pci_name(dev));
 		pci_dev_put(dev2);
 		return -ENOMEM;
 	}
diff --git a/drivers/ide/pci/it821x.c b/drivers/ide/pci/it821x.c
index c5c91f2109d4..4ce5db98f89a 100644
--- a/drivers/ide/pci/it821x.c
+++ b/drivers/ide/pci/it821x.c
@@ -569,7 +569,8 @@ static void __devinit init_hwif_it821x(ide_hwif_t *hwif)
 		idev->timing10 = 1;
 		hwif->host_flags |= IDE_HFLAG_NO_ATAPI_DMA;
 		if (idev->smart == 0)
-			printk(KERN_WARNING "it821x: Revision 0x10, workarounds activated.\n");
+			printk(KERN_WARNING "it821x %s: revision 0x10, "
+				"workarounds activated\n", pci_name(dev));
 	}
 
 	if (idev->smart == 0) {
@@ -609,11 +610,13 @@ static unsigned int __devinit init_chipset_it821x(struct pci_dev *dev, const cha
 
 	/* Force the card into bypass mode if so requested */
 	if (it8212_noraid) {
-		printk(KERN_INFO "it8212: forcing bypass mode.\n");
+		printk(KERN_INFO "it821x %s: forcing bypass mode\n",
+			pci_name(dev));
 		it8212_disable_raid(dev);
 	}
 	pci_read_config_byte(dev, 0x50, &conf);
-	printk(KERN_INFO "it821x: controller in %s mode.\n", mode[conf & 1]);
+	printk(KERN_INFO "it821x %s: controller in %s mode\n",
+		pci_name(dev), mode[conf & 1]);
 	return 0;
 }
 
@@ -654,7 +657,7 @@ static int __devinit it821x_init_one(struct pci_dev *dev, const struct pci_devic
 
 	itdevs = kzalloc(2 * sizeof(*itdevs), GFP_KERNEL);
 	if (itdevs == NULL) {
-		printk(KERN_ERR "it821x: out of memory\n");
+		printk(KERN_ERR "it821x %s: out of memory\n", pci_name(dev));
 		return -ENOMEM;
 	}
 
diff --git a/drivers/ide/pci/pdc202xx_new.c b/drivers/ide/pci/pdc202xx_new.c
index 9881a1a8b89c..4c2b669d7de6 100644
--- a/drivers/ide/pci/pdc202xx_new.c
+++ b/drivers/ide/pci/pdc202xx_new.c
@@ -358,12 +358,13 @@ static unsigned int __devinit init_chipset_pdcnew(struct pci_dev *dev, const cha
 	 * registers setting.
 	 */
 	pll_input = detect_pll_input_clock(dma_base);
-	printk("%s: PLL input clock is %ld kHz\n", name, pll_input / 1000);
+	printk(KERN_INFO "%s %s: PLL input clock is %ld kHz\n",
+		name, pci_name(dev), pll_input / 1000);
 
 	/* Sanity check */
 	if (unlikely(pll_input < 5000000L || pll_input > 70000000L)) {
-		printk(KERN_ERR "%s: Bad PLL input clock %ld Hz, giving up!\n",
-		       name, pll_input);
+		printk(KERN_ERR "%s %s: Bad PLL input clock %ld Hz, giving up!"
+			"\n", name, pci_name(dev), pll_input);
 		goto out;
 	}
 
@@ -399,7 +400,8 @@ static unsigned int __devinit init_chipset_pdcnew(struct pci_dev *dev, const cha
 		r = 0x00;
 	} else {
 		/* Invalid ratio */
-		printk(KERN_ERR "%s: Bad ratio %ld, giving up!\n", name, ratio);
+		printk(KERN_ERR "%s %s: Bad ratio %ld, giving up!\n",
+			name, pci_name(dev), ratio);
 		goto out;
 	}
 
@@ -409,7 +411,8 @@ static unsigned int __devinit init_chipset_pdcnew(struct pci_dev *dev, const cha
 
 	if (unlikely(f < 0 || f > 127)) {
 		/* Invalid F */
-		printk(KERN_ERR "%s: F[%d] invalid!\n", name, f);
+		printk(KERN_ERR "%s %s: F[%d] invalid!\n",
+			name, pci_name(dev), f);
 		goto out;
 	}
 
@@ -455,8 +458,8 @@ static struct pci_dev * __devinit pdc20270_get_dev2(struct pci_dev *dev)
 
 		if (dev2->irq != dev->irq) {
 			dev2->irq = dev->irq;
-			printk(KERN_INFO "PDC20270: PCI config space "
-					 "interrupt fixed\n");
+			printk(KERN_INFO "PDC20270 %s: PCI config space "
+				"interrupt fixed\n", pci_name(dev));
 		}
 
 		return dev2;
@@ -535,8 +538,8 @@ static int __devinit pdc202new_init_one(struct pci_dev *dev, const struct pci_de
 	    bridge->vendor == PCI_VENDOR_ID_INTEL &&
 	    (bridge->device == PCI_DEVICE_ID_INTEL_I960 ||
 	     bridge->device == PCI_DEVICE_ID_INTEL_I960RM)) {
-		printk(KERN_INFO "PDC20276: attached to I2O RAID controller, "
-				 "skipping\n");
+		printk(KERN_INFO "PDC20276 %s: attached to I2O RAID controller,"
+			" skipping\n", pci_name(dev));
 		return -ENODEV;
 	}
 
diff --git a/drivers/ide/pci/pdc202xx_old.c b/drivers/ide/pci/pdc202xx_old.c
index 17d99ce273af..5cb2731047e9 100644
--- a/drivers/ide/pci/pdc202xx_old.c
+++ b/drivers/ide/pci/pdc202xx_old.c
@@ -304,8 +304,8 @@ static void __devinit pdc202ata4_fixup_irq(struct pci_dev *dev,
 		if (irq != irq2) {
 			pci_write_config_byte(dev,
 				(PCI_INTERRUPT_LINE)|0x80, irq);     /* 0xbc */
-			printk(KERN_INFO "%s: PCI config space interrupt "
-					 "mirror fixed\n", name);
+			printk(KERN_INFO "%s %s: PCI config space interrupt "
+				"mirror fixed\n", name, pci_name(dev));
 		}
 	}
 }
@@ -406,8 +406,9 @@ static int __devinit pdc202xx_init_one(struct pci_dev *dev, const struct pci_dev
 		    bridge->vendor == PCI_VENDOR_ID_INTEL &&
 		    (bridge->device == PCI_DEVICE_ID_INTEL_I960 ||
 		     bridge->device == PCI_DEVICE_ID_INTEL_I960RM)) {
-			printk(KERN_INFO "ide: Skipping Promise PDC20265 "
-				"attached to I2O RAID controller\n");
+			printk(KERN_INFO "pdc202xx_old %s: skipping Promise "
+				"PDC20265 attached to I2O RAID controller\n",
+				pci_name(dev));
 			return -ENODEV;
 		}
 	}
diff --git a/drivers/ide/pci/serverworks.c b/drivers/ide/pci/serverworks.c
index eca52fede281..1106ff44cdee 100644
--- a/drivers/ide/pci/serverworks.c
+++ b/drivers/ide/pci/serverworks.c
@@ -188,7 +188,8 @@ static unsigned int __devinit init_chipset_svwks (struct pci_dev *dev, const cha
 			pci_read_config_dword(isa_dev, 0x64, &reg);
 			reg &= ~0x00002000; /* disable 600ns interrupt mask */
 			if(!(reg & 0x00004000))
-				printk(KERN_DEBUG "%s: UDMA not BIOS enabled.\n", name);
+				printk(KERN_DEBUG "%s %s: UDMA not BIOS "
+					"enabled.\n", name, pci_name(dev));
 			reg |=  0x00004000; /* enable UDMA/33 support */
 			pci_write_config_dword(isa_dev, 0x64, reg);
 		}
diff --git a/drivers/ide/pci/siimage.c b/drivers/ide/pci/siimage.c
index f2e4dac9b719..fc29f1ca503a 100644
--- a/drivers/ide/pci/siimage.c
+++ b/drivers/ide/pci/siimage.c
@@ -539,7 +539,8 @@ static unsigned int __devinit init_chipset_siimage(struct pci_dev *dev,
 			{ "== 100", "== 133", "== 2X PCI", "DISABLED!" };
 
 		tmp >>= 4;
-		printk(KERN_INFO "%s: BASE CLOCK %s\n", name, clk_str[tmp & 3]);
+		printk(KERN_INFO "%s %s: BASE CLOCK %s\n",
+			name, pci_name(dev), clk_str[tmp & 3]);
 	}
 
 	return 0;
@@ -779,8 +780,8 @@ static int __devinit siimage_init_one(struct pci_dev *dev,
 		* seem to get terminally confused in the PCI spaces.
 		*/
 		if (!request_mem_region(bar5, barsize, d.name)) {
-			printk(KERN_WARNING "siimage: IDE controller MMIO "
-					    "ports not available.\n");
+			printk(KERN_WARNING "siimage %s: MMIO ports not "
+				"available\n", pci_name(dev));
 		} else {
 			ioaddr = ioremap(bar5, barsize);
 			if (ioaddr == NULL)
diff --git a/drivers/ide/pci/sis5513.c b/drivers/ide/pci/sis5513.c
index f3cf7c6c0cc3..518d8ab413f3 100644
--- a/drivers/ide/pci/sis5513.c
+++ b/drivers/ide/pci/sis5513.c
@@ -380,8 +380,9 @@ static int __devinit sis_find_family(struct pci_dev *dev)
 		}
 		pci_dev_put(host);
 
-		printk(KERN_INFO "SIS5513: %s %s controller\n",
-			 SiSHostChipInfo[i].name, chipset_capability[chipset_family]);
+		printk(KERN_INFO "SIS5513 %s: %s %s controller\n",
+			pci_name(dev), SiSHostChipInfo[i].name,
+			chipset_capability[chipset_family]);
 	}
 
 	if (!chipset_family) { /* Belongs to pci-quirks */
@@ -396,7 +397,8 @@ static int __devinit sis_find_family(struct pci_dev *dev)
 			pci_write_config_dword(dev, 0x54, idemisc);
 
 			if (trueid == 0x5518) {
-				printk(KERN_INFO "SIS5513: SiS 962/963 MuTIOL IDE UDMA133 controller\n");
+				printk(KERN_INFO "SIS5513 %s: SiS 962/963 MuTIOL IDE UDMA133 controller\n",
+					pci_name(dev));
 				chipset_family = ATA_133;
 
 				/* Check for 5513 compability mapping
@@ -405,7 +407,8 @@ static int __devinit sis_find_family(struct pci_dev *dev)
 				 */
 				if ((idemisc & 0x40000000) == 0) {
 					pci_write_config_dword(dev, 0x54, idemisc | 0x40000000);
-					printk(KERN_INFO "SIS5513: Switching to 5513 register mapping\n");
+					printk(KERN_INFO "SIS5513 %s: Switching to 5513 register mapping\n",
+						pci_name(dev));
 				}
 			}
 	}
@@ -429,10 +432,12 @@ static int __devinit sis_find_family(struct pci_dev *dev)
 				pci_dev_put(lpc_bridge);
 
 				if (lpc_bridge->revision == 0x10 && (prefctl & 0x80)) {
-					printk(KERN_INFO "SIS5513: SiS 961B MuTIOL IDE UDMA133 controller\n");
+					printk(KERN_INFO "SIS5513 %s: SiS 961B MuTIOL IDE UDMA133 controller\n",
+						pci_name(dev));
 					chipset_family = ATA_133a;
 				} else {
-					printk(KERN_INFO "SIS5513: SiS 961 MuTIOL IDE UDMA100 controller\n");
+					printk(KERN_INFO "SIS5513 %s: SiS 961 MuTIOL IDE UDMA100 controller\n",
+						pci_name(dev));
 					chipset_family = ATA_100;
 				}
 			}
diff --git a/drivers/ide/pci/trm290.c b/drivers/ide/pci/trm290.c
index 449f4de9a702..7bda5ed92e1e 100644
--- a/drivers/ide/pci/trm290.c
+++ b/drivers/ide/pci/trm290.c
@@ -245,10 +245,10 @@ static void __devinit init_hwif_trm290(ide_hwif_t *hwif)
 	u8 reg = 0;
 
 	if ((dev->class & 5) && cfg_base)
-		printk(KERN_INFO "TRM290: chip");
+		printk(KERN_INFO "TRM290 %s: chip", pci_name(dev));
 	else {
 		cfg_base = 0x3df0;
-		printk(KERN_INFO "TRM290: using default");
+		printk(KERN_INFO "TRM290 %s: using default", pci_name(dev));
 	}
 	printk(KERN_CONT " config base at 0x%04x\n", cfg_base);
 	hwif->config_data = cfg_base;
diff --git a/drivers/ide/pci/via82cxxx.c b/drivers/ide/pci/via82cxxx.c
index 5f07eeb736ab..23332556e616 100644
--- a/drivers/ide/pci/via82cxxx.c
+++ b/drivers/ide/pci/via82cxxx.c
@@ -401,20 +401,19 @@ static int __devinit via_init_one(struct pci_dev *dev, const struct pci_device_i
 	 */
 	via_config = via_config_find(&isa);
 	if (!via_config->id) {
-		printk(KERN_WARNING "VP_IDE: Unknown VIA SouthBridge, disabling DMA.\n");
+		printk(KERN_WARNING "VP_IDE %s: unknown chipset, skipping\n",
+			pci_name(dev));
 		return -ENODEV;
 	}
 
 	/*
 	 * Print the boot message.
 	 */
-	printk(KERN_INFO "VP_IDE: VIA %s (rev %02x) IDE %sDMA%s "
-		"controller on pci%s\n",
-		via_config->name, isa->revision,
+	printk(KERN_INFO "VP_IDE %s: VIA %s (rev %02x) IDE %sDMA%s\n",
+		pci_name(dev), via_config->name, isa->revision,
 		via_config->udma_mask ? "U" : "MW",
 		via_dma[via_config->udma_mask ?
-			(fls(via_config->udma_mask) - 1) : 0],
-		pci_name(dev));
+			(fls(via_config->udma_mask) - 1) : 0]);
 
 	pci_dev_put(isa);
 
@@ -454,7 +453,7 @@ static int __devinit via_init_one(struct pci_dev *dev, const struct pci_device_i
 
 	vdev = kzalloc(sizeof(*vdev), GFP_KERNEL);
 	if (!vdev) {
-		printk(KERN_ERR "VP_IDE: out of memory :(\n");
+		printk(KERN_ERR "VP_IDE %s: out of memory :(\n", pci_name(dev));
 		return -ENOMEM;
 	}
 
diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
index 20f0ee004695..d9655aeb013b 100644
--- a/drivers/ide/setup-pci.c
+++ b/drivers/ide/setup-pci.c
@@ -39,17 +39,18 @@ static int ide_setup_pci_baseregs(struct pci_dev *dev, const char *name)
 	if (pci_read_config_byte(dev, PCI_CLASS_PROG, &progif) ||
 			 (progif & 5) != 5) {
 		if ((progif & 0xa) != 0xa) {
-			printk(KERN_INFO "%s: device not capable of full "
-				"native PCI mode\n", name);
+			printk(KERN_INFO "%s %s: device not capable of full "
+				"native PCI mode\n", name, pci_name(dev));
 			return -EOPNOTSUPP;
 		}
-		printk("%s: placing both ports into native PCI mode\n", name);
+		printk(KERN_INFO "%s %s: placing both ports into native PCI "
+			"mode\n", name, pci_name(dev));
 		(void) pci_write_config_byte(dev, PCI_CLASS_PROG, progif|5);
 		if (pci_read_config_byte(dev, PCI_CLASS_PROG, &progif) ||
 		    (progif & 5) != 5) {
-			printk(KERN_ERR "%s: rewrite of PROGIF failed, wanted "
-				"0x%04x, got 0x%04x\n",
-				name, progif|5, progif);
+			printk(KERN_ERR "%s %s: rewrite of PROGIF failed, "
+				"wanted 0x%04x, got 0x%04x\n",
+				name, pci_name(dev), progif | 5, progif);
 			return -EOPNOTSUPP;
 		}
 	}
@@ -57,14 +58,14 @@ static int ide_setup_pci_baseregs(struct pci_dev *dev, const char *name)
 }
 
 #ifdef CONFIG_BLK_DEV_IDEDMA_PCI
-static void ide_pci_clear_simplex(unsigned long dma_base, const char *name)
+static int ide_pci_clear_simplex(unsigned long dma_base, const char *name)
 {
 	u8 dma_stat = inb(dma_base + 2);
 
 	outb(dma_stat & 0x60, dma_base + 2);
 	dma_stat = inb(dma_base + 2);
-	if (dma_stat & 0x80)
-		printk(KERN_INFO "%s: simplex device: DMA forced\n", name);
+
+	return (dma_stat & 0x80) ? 1 : 0;
 }
 
 /**
@@ -91,7 +92,8 @@ unsigned long ide_pci_dma_base(ide_hwif_t *hwif, const struct ide_port_info *d)
 		dma_base = pci_resource_start(dev, baridx);
 
 		if (dma_base == 0) {
-			printk(KERN_ERR "%s: DMA base is invalid\n", d->name);
+			printk(KERN_ERR "%s %s: DMA base is invalid\n",
+				d->name, pci_name(dev));
 			return 0;
 		}
 	}
@@ -105,13 +107,16 @@ EXPORT_SYMBOL_GPL(ide_pci_dma_base);
 
 int ide_pci_check_simplex(ide_hwif_t *hwif, const struct ide_port_info *d)
 {
+	struct pci_dev *dev = to_pci_dev(hwif->dev);
 	u8 dma_stat;
 
 	if (d->host_flags & (IDE_HFLAG_MMIO | IDE_HFLAG_CS5520))
 		goto out;
 
 	if (d->host_flags & IDE_HFLAG_CLEAR_SIMPLEX) {
-		ide_pci_clear_simplex(hwif->dma_base, d->name);
+		if (ide_pci_clear_simplex(hwif->dma_base, d->name))
+			printk(KERN_INFO "%s %s: simplex device: DMA forced\n",
+				d->name, pci_name(dev));
 		goto out;
 	}
 
@@ -127,7 +132,8 @@ int ide_pci_check_simplex(ide_hwif_t *hwif, const struct ide_port_info *d)
 	 */
 	dma_stat = hwif->tp_ops->read_sff_dma_status(hwif);
 	if ((dma_stat & 0x80) && hwif->mate && hwif->mate->dma_base) {
-		printk(KERN_INFO "%s: simplex device: DMA disabled\n", d->name);
+		printk(KERN_INFO "%s %s: simplex device: DMA disabled\n",
+			d->name, pci_name(dev));
 		return -1;
 	}
 out:
@@ -149,8 +155,8 @@ int ide_pci_set_master(struct pci_dev *dev, const char *name)
 
 		if (pci_read_config_word(dev, PCI_COMMAND, &pcicmd) ||
 		    (pcicmd & PCI_COMMAND_MASTER) == 0) {
-			printk(KERN_ERR "%s: error updating PCICMD on %s\n",
-					name, pci_name(dev));
+			printk(KERN_ERR "%s %s: error updating PCICMD\n",
+				name, pci_name(dev));
 			return -EIO;
 		}
 	}
@@ -162,9 +168,9 @@ EXPORT_SYMBOL_GPL(ide_pci_set_master);
 
 void ide_setup_pci_noise(struct pci_dev *dev, const struct ide_port_info *d)
 {
-	printk(KERN_INFO "%s: IDE controller (0x%04x:0x%04x rev 0x%02x) at "
-			 " PCI slot %s\n", d->name, dev->vendor, dev->device,
-			 dev->revision, pci_name(dev));
+	printk(KERN_INFO "%s %s: IDE controller (0x%04x:0x%04x rev 0x%02x)\n",
+		d->name, pci_name(dev),
+		dev->vendor, dev->device, dev->revision);
 }
 EXPORT_SYMBOL_GPL(ide_setup_pci_noise);
 
@@ -189,11 +195,12 @@ static int ide_pci_enable(struct pci_dev *dev, const struct ide_port_info *d)
 	if (pci_enable_device(dev)) {
 		ret = pci_enable_device_io(dev);
 		if (ret < 0) {
-			printk(KERN_WARNING "%s: (ide_setup_pci_device:) "
-				"Could not enable device.\n", d->name);
+			printk(KERN_WARNING "%s %s: couldn't enable device\n",
+				d->name, pci_name(dev));
 			goto out;
 		}
-		printk(KERN_WARNING "%s: BIOS configuration fixed.\n", d->name);
+		printk(KERN_WARNING "%s %s: BIOS configuration fixed\n",
+			d->name, pci_name(dev));
 	}
 
 	/*
@@ -203,7 +210,8 @@ static int ide_pci_enable(struct pci_dev *dev, const struct ide_port_info *d)
 	 */
 	ret = pci_set_dma_mask(dev, DMA_32BIT_MASK);
 	if (ret < 0) {
-		printk(KERN_ERR "%s: can't set dma mask\n", d->name);
+		printk(KERN_ERR "%s %s: can't set DMA mask\n",
+			d->name, pci_name(dev));
 		goto out;
 	}
 
@@ -221,7 +229,8 @@ static int ide_pci_enable(struct pci_dev *dev, const struct ide_port_info *d)
 
 	ret = pci_request_selected_regions(dev, bars, d->name);
 	if (ret < 0)
-		printk(KERN_ERR "%s: can't reserve resources\n", d->name);
+		printk(KERN_ERR "%s %s: can't reserve resources\n",
+			d->name, pci_name(dev));
 out:
 	return ret;
 }
@@ -247,15 +256,18 @@ static int ide_pci_configure(struct pci_dev *dev, const struct ide_port_info *d)
 	 */
 	if (ide_setup_pci_baseregs(dev, d->name) ||
 	    pci_write_config_word(dev, PCI_COMMAND, pcicmd | PCI_COMMAND_IO)) {
-		printk(KERN_INFO "%s: device disabled (BIOS)\n", d->name);
+		printk(KERN_INFO "%s %s: device disabled (BIOS)\n",
+			d->name, pci_name(dev));
 		return -ENODEV;
 	}
 	if (pci_read_config_word(dev, PCI_COMMAND, &pcicmd)) {
-		printk(KERN_ERR "%s: error accessing PCI regs\n", d->name);
+		printk(KERN_ERR "%s %s: error accessing PCI regs\n",
+			d->name, pci_name(dev));
 		return -EIO;
 	}
 	if (!(pcicmd & PCI_COMMAND_IO)) {
-		printk(KERN_ERR "%s: unable to enable IDE controller\n", d->name);
+		printk(KERN_ERR "%s %s: unable to enable IDE controller\n",
+			d->name, pci_name(dev));
 		return -ENXIO;
 	}
 	return 0;
@@ -311,8 +323,9 @@ static int ide_hw_configure(struct pci_dev *dev, const struct ide_port_info *d,
 	if ((d->host_flags & IDE_HFLAG_ISA_PORTS) == 0) {
 		if (ide_pci_check_iomem(dev, d, 2 * port) ||
 		    ide_pci_check_iomem(dev, d, 2 * port + 1)) {
-			printk(KERN_ERR "%s: I/O baseregs (BIOS) are reported "
-					"as MEM for port %d!\n", d->name, port);
+			printk(KERN_ERR "%s %s: I/O baseregs (BIOS) are "
+				"reported as MEM for port %d!\n",
+				d->name, pci_name(dev), port);
 			return -EINVAL;
 		}
 
@@ -325,8 +338,8 @@ static int ide_hw_configure(struct pci_dev *dev, const struct ide_port_info *d,
 	}
 
 	if (!base || !ctl) {
-		printk(KERN_ERR "%s: bad PCI BARs for port %d, skipping\n",
-				d->name, port);
+		printk(KERN_ERR "%s %s: bad PCI BARs for port %d, skipping\n",
+			d->name, pci_name(dev), port);
 		return -EINVAL;
 	}
 
@@ -414,14 +427,16 @@ static int ide_setup_pci_controller(struct pci_dev *dev,
 
 	ret = pci_read_config_word(dev, PCI_COMMAND, &pcicmd);
 	if (ret < 0) {
-		printk(KERN_ERR "%s: error accessing PCI regs\n", d->name);
+		printk(KERN_ERR "%s %s: error accessing PCI regs\n",
+			d->name, pci_name(dev));
 		goto out;
 	}
 	if (!(pcicmd & PCI_COMMAND_IO)) {	/* is device disabled? */
 		ret = ide_pci_configure(dev, d);
 		if (ret < 0)
 			goto out;
-		printk(KERN_INFO "%s: device enabled (Linux)\n", d->name);
+		printk(KERN_INFO "%s %s: device enabled (Linux)\n",
+			d->name, pci_name(dev));
 	}
 
 out:
@@ -460,7 +475,8 @@ void ide_pci_setup_ports(struct pci_dev *dev, const struct ide_port_info *d,
 
 		if (e->reg && (pci_read_config_byte(dev, e->reg, &tmp) ||
 		    (tmp & e->mask) != e->val)) {
-			printk(KERN_INFO "%s: IDE port disabled\n", d->name);
+			printk(KERN_INFO "%s %s: IDE port disabled\n",
+				d->name, pci_name(dev));
 			continue;	/* port not enabled */
 		}
 
@@ -506,18 +522,15 @@ static int do_ide_setup_pci_device(struct pci_dev *dev,
 	/* Is it an "IDE storage" device in non-PCI mode? */
 	if ((dev->class >> 8) == PCI_CLASS_STORAGE_IDE && (dev->class & 5) != 5) {
 		if (noisy)
-			printk(KERN_INFO "%s: not 100%% native mode: "
-				"will probe irqs later\n", d->name);
+			printk(KERN_INFO "%s %s: not 100%% native mode: will "
+				"probe irqs later\n", d->name, pci_name(dev));
 		pciirq = ret;
-	} else if (!pciirq) {
-		if (noisy)
-			printk(KERN_WARNING "%s: bad irq (%d): will probe later\n",
-				d->name, pciirq);
-		pciirq = 0;
-	} else {
-		if (noisy)
-			printk(KERN_INFO "%s: 100%% native mode on irq %d\n",
-				d->name, pciirq);
+	} else if (!pciirq && noisy) {
+		printk(KERN_WARNING "%s %s: bad irq (%d): will probe later\n",
+			d->name, pci_name(dev), pciirq);
+	} else if (noisy) {
+		printk(KERN_INFO "%s %s: 100%% native mode on irq %d\n",
+			d->name, pci_name(dev), pciirq);
 	}
 
 	ret = pciirq;

From 29f1ca920cb8d65b979f7edf2fc7d11095461306 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:32 +0200
Subject: [PATCH 0586/1171] it8213: remove DECLARE_ITE_DEV() macro

While at it:

* it8213_chipsets[] -> it8213_chipset.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/it8213.c | 26 ++++++++++----------------
 1 file changed, 10 insertions(+), 16 deletions(-)

diff --git a/drivers/ide/pci/it8213.c b/drivers/ide/pci/it8213.c
index 575bf2386f8e..451b87fd8217 100644
--- a/drivers/ide/pci/it8213.c
+++ b/drivers/ide/pci/it8213.c
@@ -155,23 +155,17 @@ static const struct ide_port_ops it8213_port_ops = {
 	.cable_detect		= it8213_cable_detect,
 };
 
-#define DECLARE_ITE_DEV(name_str)			\
-	{						\
-		.name		= name_str,		\
-		.enablebits	= { {0x41, 0x80, 0x80} }, \
-		.port_ops	= &it8213_port_ops,	\
-		.host_flags	= IDE_HFLAG_SINGLE,	\
-		.pio_mask	= ATA_PIO4,		\
-		.swdma_mask	= ATA_SWDMA2_ONLY,	\
-		.mwdma_mask	= ATA_MWDMA12_ONLY,	\
-		.udma_mask	= ATA_UDMA6,		\
-	}
-
-static const struct ide_port_info it8213_chipsets[] __devinitdata = {
-	/* 0 */ DECLARE_ITE_DEV("IT8213"),
+static const struct ide_port_info it8213_chipset __devinitdata = {
+	.name		= "IT8213",
+	.enablebits	= { {0x41, 0x80, 0x80} },
+	.port_ops	= &it8213_port_ops,
+	.host_flags	= IDE_HFLAG_SINGLE,
+	.pio_mask	= ATA_PIO4,
+	.swdma_mask	= ATA_SWDMA2_ONLY,
+	.mwdma_mask	= ATA_MWDMA12_ONLY,
+	.udma_mask	= ATA_UDMA6,
 };
 
-
 /**
  *	it8213_init_one	-	pci layer discovery entry
  *	@dev: PCI device
@@ -184,7 +178,7 @@ static const struct ide_port_info it8213_chipsets[] __devinitdata = {
 
 static int __devinit it8213_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
-	return ide_pci_init_one(dev, &it8213_chipsets[id->driver_data], NULL);
+	return ide_pci_init_one(dev, &it8213_chipset, NULL);
 }
 
 static const struct pci_device_id it8213_pci_tbl[] = {

From 04ba6e739e9c0623c25f94b191fd20dfbd1b26e3 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:32 +0200
Subject: [PATCH 0587/1171] it821x: remove DECLARE_ITE_DEV() macro

While at it:

* it821x_chipsets[] -> it821x_chipset.

* Fix it821x_chipset's name field (as it is used for IT8211/8212).

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/it821x.c | 19 +++++++------------
 1 file changed, 7 insertions(+), 12 deletions(-)

diff --git a/drivers/ide/pci/it821x.c b/drivers/ide/pci/it821x.c
index 4ce5db98f89a..a3d8959436c5 100644
--- a/drivers/ide/pci/it821x.c
+++ b/drivers/ide/pci/it821x.c
@@ -628,17 +628,12 @@ static const struct ide_port_ops it821x_port_ops = {
 	.cable_detect		= it821x_cable_detect,
 };
 
-#define DECLARE_ITE_DEV(name_str)			\
-	{						\
-		.name		= name_str,		\
-		.init_chipset	= init_chipset_it821x,	\
-		.init_hwif	= init_hwif_it821x,	\
-		.port_ops	= &it821x_port_ops,	\
-		.pio_mask	= ATA_PIO4,		\
-	}
-
-static const struct ide_port_info it821x_chipsets[] __devinitdata = {
-	/* 0 */ DECLARE_ITE_DEV("IT8212"),
+static const struct ide_port_info it821x_chipset __devinitdata = {
+	.name		= "IT821X",
+	.init_chipset	= init_chipset_it821x,
+	.init_hwif	= init_hwif_it821x,
+	.port_ops	= &it821x_port_ops,
+	.pio_mask	= ATA_PIO4,
 };
 
 /**
@@ -661,7 +656,7 @@ static int __devinit it821x_init_one(struct pci_dev *dev, const struct pci_devic
 		return -ENOMEM;
 	}
 
-	rc = ide_pci_init_one(dev, &it821x_chipsets[id->driver_data], itdevs);
+	rc = ide_pci_init_one(dev, &it821x_chipset, itdevs);
 	if (rc)
 		kfree(itdevs);
 

From ced3ec8aa7d0fa3300187ee47c144a22ccfc974e Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:32 +0200
Subject: [PATCH 0588/1171] ide: prefix messages from IDE PCI host drivers by
 driver name

Prefix messages from IDE PCI host drivers by driver name instead of marketed
chipset name (it is still possible to exactly identify the particular chipset
basing on driver messages).

As a bonus this provides nice code savings for some drivers:

   text    data     bss     dec     hex filename
   3826     112       8    3946     f6a drivers/ide/pci/amd74xx.o.before
   2786     112       8    2906     b5a drivers/ide/pci/amd74xx.o.after
    764     108       0     872     368 drivers/ide/pci/cs5520.o.before
    680     108       0     788     314 drivers/ide/pci/cs5520.o.after
   1680     112       4    1796     704 drivers/ide/pci/generic.o.before
   1155     112       4    1271     4f7 drivers/ide/pci/generic.o.after
   7128     792       0    7920    1ef0 drivers/ide/pci/hpt366.o.before
   6984     792       0    7776    1e60 drivers/ide/pci/hpt366.o.after
   2800     148       0    2948     b84 drivers/ide/pci/pdc202xx_new.o.before
   2523     148       0    2671     a6f drivers/ide/pci/pdc202xx_new.o.after
   2831     148       0    2979     ba3 drivers/ide/pci/pdc202xx_old.o.before
   2683     148       0    2831     b0f drivers/ide/pci/pdc202xx_old.o.after
   3776     112       4    3892     f34 drivers/ide/pci/piix.o.before
   2804     112       4    2920     b68 drivers/ide/pci/piix.o.after
   4693     116       0    4809    12c9 drivers/ide/pci/siimage.o.before
   4600     116       0    4716    126c drivers/ide/pci/siimage.o.after

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/aec62xx.c      |  29 +++++----
 drivers/ide/pci/alim15x3.c     |   4 +-
 drivers/ide/pci/amd74xx.c      |  82 +++++++++++--------------
 drivers/ide/pci/atiixp.c       |  11 ++--
 drivers/ide/pci/cmd64x.c       |  21 ++++---
 drivers/ide/pci/cs5520.c       |  24 ++++----
 drivers/ide/pci/cs5530.c       |   4 +-
 drivers/ide/pci/cs5535.c       |   4 +-
 drivers/ide/pci/cy82c693.c     |   4 +-
 drivers/ide/pci/generic.c      |  63 ++++++++++----------
 drivers/ide/pci/hpt34x.c       |  10 ++--
 drivers/ide/pci/hpt366.c       |  65 ++++----------------
 drivers/ide/pci/it8213.c       |   4 +-
 drivers/ide/pci/it821x.c       |  12 ++--
 drivers/ide/pci/jmicron.c      |   4 +-
 drivers/ide/pci/ns87415.c      |   4 +-
 drivers/ide/pci/opti621.c      |   4 +-
 drivers/ide/pci/pdc202xx_new.c |  38 +++++-------
 drivers/ide/pci/pdc202xx_old.c |  30 +++++-----
 drivers/ide/pci/piix.c         | 106 +++++++++++++++------------------
 drivers/ide/pci/rz1000.c       |   4 +-
 drivers/ide/pci/sc1200.c       |   4 +-
 drivers/ide/pci/serverworks.c  |  26 ++++----
 drivers/ide/pci/siimage.c      |  17 +++---
 drivers/ide/pci/sis5513.c      |  14 +++--
 drivers/ide/pci/sl82c105.c     |   6 +-
 drivers/ide/pci/slc90e66.c     |   4 +-
 drivers/ide/pci/tc86c001.c     |   4 +-
 drivers/ide/pci/triflex.c      |   4 +-
 drivers/ide/pci/trm290.c       |   8 ++-
 drivers/ide/pci/via82cxxx.c    |  15 +++--
 31 files changed, 305 insertions(+), 324 deletions(-)

diff --git a/drivers/ide/pci/aec62xx.c b/drivers/ide/pci/aec62xx.c
index 253299961a14..f6dc6c20f3af 100644
--- a/drivers/ide/pci/aec62xx.c
+++ b/drivers/ide/pci/aec62xx.c
@@ -13,6 +13,8 @@
 
 #include <asm/io.h>
 
+#define DRV_NAME "aec62xx"
+
 struct chipset_bus_clock_list_entry {
 	u8 xfer_speed;
 	u8 chipset_settings;
@@ -180,8 +182,8 @@ static const struct ide_port_ops atp86x_port_ops = {
 };
 
 static const struct ide_port_info aec62xx_chipsets[] __devinitdata = {
-	{	/* 0 */
-		.name		= "AEC6210",
+	{	/* 0: AEC6210 */
+		.name		= DRV_NAME,
 		.init_chipset	= init_chipset_aec62xx,
 		.enablebits	= {{0x4a,0x02,0x02}, {0x4a,0x04,0x04}},
 		.port_ops	= &atp850_port_ops,
@@ -192,8 +194,9 @@ static const struct ide_port_info aec62xx_chipsets[] __devinitdata = {
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 		.udma_mask	= ATA_UDMA2,
-	},{	/* 1 */
-		.name		= "AEC6260",
+	},
+	{	/* 1: AEC6260 */
+		.name		= DRV_NAME,
 		.init_chipset	= init_chipset_aec62xx,
 		.port_ops	= &atp86x_port_ops,
 		.host_flags	= IDE_HFLAG_NO_ATAPI_DMA | IDE_HFLAG_NO_AUTODMA |
@@ -201,8 +204,9 @@ static const struct ide_port_info aec62xx_chipsets[] __devinitdata = {
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 		.udma_mask	= ATA_UDMA4,
-	},{	/* 2 */
-		.name		= "AEC6260R",
+	},
+	{	/* 2: AEC6260R */
+		.name		= DRV_NAME,
 		.init_chipset	= init_chipset_aec62xx,
 		.enablebits	= {{0x4a,0x02,0x02}, {0x4a,0x04,0x04}},
 		.port_ops	= &atp86x_port_ops,
@@ -211,8 +215,9 @@ static const struct ide_port_info aec62xx_chipsets[] __devinitdata = {
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 		.udma_mask	= ATA_UDMA4,
-	},{	/* 3 */
-		.name		= "AEC6280",
+	},
+	{	/* 3: AEC6280 */
+		.name		= DRV_NAME,
 		.init_chipset	= init_chipset_aec62xx,
 		.port_ops	= &atp86x_port_ops,
 		.host_flags	= IDE_HFLAG_NO_ATAPI_DMA |
@@ -220,8 +225,9 @@ static const struct ide_port_info aec62xx_chipsets[] __devinitdata = {
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 		.udma_mask	= ATA_UDMA5,
-	},{	/* 4 */
-		.name		= "AEC6280R",
+	},
+	{	/* 4: AEC6280R */
+		.name		= DRV_NAME,
 		.init_chipset	= init_chipset_aec62xx,
 		.enablebits	= {{0x4a,0x02,0x02}, {0x4a,0x04,0x04}},
 		.port_ops	= &atp86x_port_ops,
@@ -268,7 +274,8 @@ static int __devinit aec62xx_init_one(struct pci_dev *dev, const struct pci_devi
 		unsigned long dma_base = pci_resource_start(dev, 4);
 
 		if (inb(dma_base + 2) & 0x10) {
-			d.name = (idx == 4) ? "AEC6880R" : "AEC6880";
+			printk(KERN_INFO DRV_NAME " %s: AEC6880%s card detected"
+				"\n", pci_name(dev), (idx == 4) ? "R" : "");
 			d.udma_mask = ATA_UDMA6;
 		}
 	}
diff --git a/drivers/ide/pci/alim15x3.c b/drivers/ide/pci/alim15x3.c
index ac171502b997..a099c4dd599d 100644
--- a/drivers/ide/pci/alim15x3.c
+++ b/drivers/ide/pci/alim15x3.c
@@ -38,6 +38,8 @@
 
 #include <asm/io.h>
 
+#define DRV_NAME "alim15x3"
+
 /*
  * Allow UDMA on M1543C-E chipset for WDC disks that ignore CRC checking
  * (this is DANGEROUS and could result in data corruption).
@@ -515,7 +517,7 @@ static const struct ide_dma_ops ali_dma_ops = {
 };
 
 static const struct ide_port_info ali15x3_chipset __devinitdata = {
-	.name		= "ALI15X3",
+	.name		= DRV_NAME,
 	.init_chipset	= init_chipset_ali15x3,
 	.init_hwif	= init_hwif_ali15x3,
 	.init_dma	= init_dma_ali15x3,
diff --git a/drivers/ide/pci/amd74xx.c b/drivers/ide/pci/amd74xx.c
index 2b118f80fab6..cbf78edfe00b 100644
--- a/drivers/ide/pci/amd74xx.c
+++ b/drivers/ide/pci/amd74xx.c
@@ -21,6 +21,8 @@
 #include <linux/init.h>
 #include <linux/ide.h>
 
+#define DRV_NAME "amd74xx"
+
 enum {
 	AMD_IDE_CONFIG		= 0x41,
 	AMD_CABLE_DETECT	= 0x42,
@@ -204,9 +206,9 @@ static const struct ide_port_ops amd_port_ops = {
 	 IDE_HFLAG_IO_32BIT | \
 	 IDE_HFLAG_UNMASK_IRQS)
 
-#define DECLARE_AMD_DEV(name_str, swdma, udma)				\
+#define DECLARE_AMD_DEV(swdma, udma)				\
 	{								\
-		.name		= name_str,				\
+		.name		= DRV_NAME,				\
 		.init_chipset	= init_chipset_amd74xx,			\
 		.init_hwif	= init_hwif_amd74xx,			\
 		.enablebits	= {{0x40,0x02,0x02}, {0x40,0x01,0x01}},	\
@@ -218,9 +220,9 @@ static const struct ide_port_ops amd_port_ops = {
 		.udma_mask	= udma,					\
 	}
 
-#define DECLARE_NV_DEV(name_str, udma)					\
+#define DECLARE_NV_DEV(udma)					\
 	{								\
-		.name		= name_str,				\
+		.name		= DRV_NAME,				\
 		.init_chipset	= init_chipset_amd74xx,			\
 		.init_hwif	= init_hwif_amd74xx,			\
 		.enablebits	= {{0x50,0x02,0x02}, {0x50,0x01,0x01}},	\
@@ -233,31 +235,15 @@ static const struct ide_port_ops amd_port_ops = {
 	}
 
 static const struct ide_port_info amd74xx_chipsets[] __devinitdata = {
-	/*  0 */ DECLARE_AMD_DEV("AMD7401",	  0x00, ATA_UDMA2),
-	/*  1 */ DECLARE_AMD_DEV("AMD7409", ATA_SWDMA2, ATA_UDMA4),
-	/*  2 */ DECLARE_AMD_DEV("AMD7411", ATA_SWDMA2, ATA_UDMA5),
-	/*  3 */ DECLARE_AMD_DEV("AMD7441", ATA_SWDMA2, ATA_UDMA5),
-	/*  4 */ DECLARE_AMD_DEV("AMD8111", ATA_SWDMA2, ATA_UDMA6),
+	/* 0: AMD7401 */	DECLARE_AMD_DEV(0x00, ATA_UDMA2),
+	/* 1: AMD7409 */	DECLARE_AMD_DEV(ATA_SWDMA2, ATA_UDMA4),
+	/* 2: AMD7411/7441 */	DECLARE_AMD_DEV(ATA_SWDMA2, ATA_UDMA5),
+	/* 3: AMD8111 */	DECLARE_AMD_DEV(ATA_SWDMA2, ATA_UDMA6),
 
-	/*  5 */ DECLARE_NV_DEV("NFORCE",		ATA_UDMA5),
-	/*  6 */ DECLARE_NV_DEV("NFORCE2",		ATA_UDMA6),
-	/*  7 */ DECLARE_NV_DEV("NFORCE2-U400R",	ATA_UDMA6),
-	/*  8 */ DECLARE_NV_DEV("NFORCE2-U400R-SATA",	ATA_UDMA6),
-	/*  9 */ DECLARE_NV_DEV("NFORCE3-150",		ATA_UDMA6),
-	/* 10 */ DECLARE_NV_DEV("NFORCE3-250",		ATA_UDMA6),
-	/* 11 */ DECLARE_NV_DEV("NFORCE3-250-SATA",	ATA_UDMA6),
-	/* 12 */ DECLARE_NV_DEV("NFORCE3-250-SATA2",	ATA_UDMA6),
-	/* 13 */ DECLARE_NV_DEV("NFORCE-CK804",		ATA_UDMA6),
-	/* 14 */ DECLARE_NV_DEV("NFORCE-MCP04",		ATA_UDMA6),
-	/* 15 */ DECLARE_NV_DEV("NFORCE-MCP51",		ATA_UDMA6),
-	/* 16 */ DECLARE_NV_DEV("NFORCE-MCP55",		ATA_UDMA6),
-	/* 17 */ DECLARE_NV_DEV("NFORCE-MCP61",		ATA_UDMA6),
-	/* 18 */ DECLARE_NV_DEV("NFORCE-MCP65",		ATA_UDMA6),
-	/* 19 */ DECLARE_NV_DEV("NFORCE-MCP67",		ATA_UDMA6),
-	/* 20 */ DECLARE_NV_DEV("NFORCE-MCP73",		ATA_UDMA6),
-	/* 21 */ DECLARE_NV_DEV("NFORCE-MCP77",		ATA_UDMA6),
+	/* 4: NFORCE */		DECLARE_NV_DEV(ATA_UDMA5),
+	/* 5: >= NFORCE2 */	DECLARE_NV_DEV(ATA_UDMA6),
 
-	/* 22 */ DECLARE_AMD_DEV("AMD5536", ATA_SWDMA2, ATA_UDMA5),
+	/* 6: AMD5536 */	DECLARE_AMD_DEV(ATA_SWDMA2, ATA_UDMA5),
 };
 
 static int __devinit amd74xx_probe(struct pci_dev *dev, const struct pci_device_id *id)
@@ -274,7 +260,7 @@ static int __devinit amd74xx_probe(struct pci_dev *dev, const struct pci_device_
 		if (dev->revision <= 7)
 			d.swdma_mask = 0;
 		d.host_flags |= IDE_HFLAG_CLEAR_SIMPLEX;
-	} else if (idx == 4) {
+	} else if (idx == 3) {
 		if (dev->subsystem_vendor == PCI_VENDOR_ID_AMD &&
 		    dev->subsystem_device == PCI_DEVICE_ID_AMD_SERENADE)
 			d.udma_mask = ATA_UDMA5;
@@ -308,30 +294,30 @@ static const struct pci_device_id amd74xx_pci_tbl[] = {
 	{ PCI_VDEVICE(AMD,	PCI_DEVICE_ID_AMD_COBRA_7401),		 0 },
 	{ PCI_VDEVICE(AMD,	PCI_DEVICE_ID_AMD_VIPER_7409),		 1 },
 	{ PCI_VDEVICE(AMD,	PCI_DEVICE_ID_AMD_VIPER_7411),		 2 },
-	{ PCI_VDEVICE(AMD,	PCI_DEVICE_ID_AMD_OPUS_7441),		 3 },
-	{ PCI_VDEVICE(AMD,	PCI_DEVICE_ID_AMD_8111_IDE),		 4 },
-	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE_IDE),	 5 },
-	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE2_IDE),	 6 },
-	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE2S_IDE),	 7 },
+	{ PCI_VDEVICE(AMD,	PCI_DEVICE_ID_AMD_OPUS_7441),		 2 },
+	{ PCI_VDEVICE(AMD,	PCI_DEVICE_ID_AMD_8111_IDE),		 3 },
+	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE_IDE),	 4 },
+	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE2_IDE),	 5 },
+	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE2S_IDE),	 5 },
 #ifdef CONFIG_BLK_DEV_IDE_SATA
-	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE2S_SATA),	 8 },
+	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE2S_SATA),	 5 },
 #endif
-	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE3_IDE),	 9 },
-	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE3S_IDE),	10 },
+	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE3_IDE),	 5 },
+	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE3S_IDE),	 5 },
 #ifdef CONFIG_BLK_DEV_IDE_SATA
-	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE3S_SATA),	11 },
-	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE3S_SATA2),	12 },
+	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE3S_SATA),	 5 },
+	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE3S_SATA2),	 5 },
 #endif
-	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE_CK804_IDE),	13 },
-	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE_MCP04_IDE),	14 },
-	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_IDE),	15 },
-	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE_MCP55_IDE),	16 },
-	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE_MCP61_IDE),	17 },
-	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE_MCP65_IDE),	18 },
-	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE_MCP67_IDE),	19 },
-	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE_MCP73_IDE),	20 },
-	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE_MCP77_IDE),	21 },
-	{ PCI_VDEVICE(AMD,	PCI_DEVICE_ID_AMD_CS5536_IDE),		22 },
+	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE_CK804_IDE),	 5 },
+	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE_MCP04_IDE),	 5 },
+	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE_MCP51_IDE),	 5 },
+	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE_MCP55_IDE),	 5 },
+	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE_MCP61_IDE),	 5 },
+	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE_MCP65_IDE),	 5 },
+	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE_MCP67_IDE),	 5 },
+	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE_MCP73_IDE),	 5 },
+	{ PCI_VDEVICE(NVIDIA,	PCI_DEVICE_ID_NVIDIA_NFORCE_MCP77_IDE),	 5 },
+	{ PCI_VDEVICE(AMD,	PCI_DEVICE_ID_AMD_CS5536_IDE),		 6 },
 	{ 0, },
 };
 MODULE_DEVICE_TABLE(pci, amd74xx_pci_tbl);
diff --git a/drivers/ide/pci/atiixp.c b/drivers/ide/pci/atiixp.c
index 4c49c1ba6182..332f08f43b56 100644
--- a/drivers/ide/pci/atiixp.c
+++ b/drivers/ide/pci/atiixp.c
@@ -11,6 +11,8 @@
 #include <linux/ide.h>
 #include <linux/init.h>
 
+#define DRV_NAME "atiixp"
+
 #define ATIIXP_IDE_PIO_TIMING		0x40
 #define ATIIXP_IDE_MDMA_TIMING		0x44
 #define ATIIXP_IDE_PIO_CONTROL		0x48
@@ -137,16 +139,17 @@ static const struct ide_port_ops atiixp_port_ops = {
 };
 
 static const struct ide_port_info atiixp_pci_info[] __devinitdata = {
-	{	/* 0 */
-		.name		= "ATIIXP",
+	{	/* 0: IXP200/300/400/700 */
+		.name		= DRV_NAME,
 		.enablebits	= {{0x48,0x01,0x00}, {0x48,0x08,0x00}},
 		.port_ops	= &atiixp_port_ops,
 		.host_flags	= IDE_HFLAG_LEGACY_IRQS,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 		.udma_mask	= ATA_UDMA5,
-	},{	/* 1 */
-		.name		= "SB600_PATA",
+	},
+	{	/* 1: IXP600 */
+		.name		= DRV_NAME,
 		.enablebits	= {{0x48,0x01,0x00}, {0x00,0x00,0x00}},
 		.port_ops	= &atiixp_port_ops,
 		.host_flags	= IDE_HFLAG_SINGLE | IDE_HFLAG_LEGACY_IRQS,
diff --git a/drivers/ide/pci/cmd64x.c b/drivers/ide/pci/cmd64x.c
index 0a4d194bc35f..3d84debaf81f 100644
--- a/drivers/ide/pci/cmd64x.c
+++ b/drivers/ide/pci/cmd64x.c
@@ -19,6 +19,8 @@
 
 #include <asm/io.h>
 
+#define DRV_NAME "cmd64x"
+
 #define CMD_DEBUG 0
 
 #if CMD_DEBUG
@@ -407,8 +409,8 @@ static const struct ide_dma_ops cmd648_dma_ops = {
 };
 
 static const struct ide_port_info cmd64x_chipsets[] __devinitdata = {
-	{	/* 0 */
-		.name		= "CMD643",
+	{	/* 0: CMD643 */
+		.name		= DRV_NAME,
 		.init_chipset	= init_chipset_cmd64x,
 		.enablebits	= {{0x00,0x00,0x00}, {0x51,0x08,0x08}},
 		.port_ops	= &cmd64x_port_ops,
@@ -418,8 +420,9 @@ static const struct ide_port_info cmd64x_chipsets[] __devinitdata = {
 		.pio_mask	= ATA_PIO5,
 		.mwdma_mask	= ATA_MWDMA2,
 		.udma_mask	= 0x00, /* no udma */
-	},{	/* 1 */
-		.name		= "CMD646",
+	},
+	{	/* 1: CMD646 */
+		.name		= DRV_NAME,
 		.init_chipset	= init_chipset_cmd64x,
 		.enablebits	= {{0x51,0x04,0x04}, {0x51,0x08,0x08}},
 		.chipset	= ide_cmd646,
@@ -429,8 +432,9 @@ static const struct ide_port_info cmd64x_chipsets[] __devinitdata = {
 		.pio_mask	= ATA_PIO5,
 		.mwdma_mask	= ATA_MWDMA2,
 		.udma_mask	= ATA_UDMA2,
-	},{	/* 2 */
-		.name		= "CMD648",
+	},
+	{	/* 2: CMD648 */
+		.name		= DRV_NAME,
 		.init_chipset	= init_chipset_cmd64x,
 		.enablebits	= {{0x51,0x04,0x04}, {0x51,0x08,0x08}},
 		.port_ops	= &cmd64x_port_ops,
@@ -439,8 +443,9 @@ static const struct ide_port_info cmd64x_chipsets[] __devinitdata = {
 		.pio_mask	= ATA_PIO5,
 		.mwdma_mask	= ATA_MWDMA2,
 		.udma_mask	= ATA_UDMA4,
-	},{	/* 3 */
-		.name		= "CMD649",
+	},
+	{	/* 3: CMD649 */
+		.name		= DRV_NAME,
 		.init_chipset	= init_chipset_cmd64x,
 		.enablebits	= {{0x51,0x04,0x04}, {0x51,0x08,0x08}},
 		.port_ops	= &cmd64x_port_ops,
diff --git a/drivers/ide/pci/cs5520.c b/drivers/ide/pci/cs5520.c
index b03d8ae947e6..c0364b287f17 100644
--- a/drivers/ide/pci/cs5520.c
+++ b/drivers/ide/pci/cs5520.c
@@ -41,6 +41,8 @@
 #include <linux/ide.h>
 #include <linux/dma-mapping.h>
 
+#define DRV_NAME "cs5520"
+
 struct pio_clocks
 {
 	int address;
@@ -92,18 +94,11 @@ static const struct ide_port_ops cs5520_port_ops = {
 	.set_dma_mode		= cs5520_set_dma_mode,
 };
 
-#define DECLARE_CS_DEV(name_str)				\
-	{							\
-		.name		= name_str,			\
-		.port_ops	= &cs5520_port_ops,		\
-		.host_flags	= IDE_HFLAG_ISA_PORTS |		\
-				  IDE_HFLAG_CS5520,		\
-		.pio_mask	= ATA_PIO4,			\
-	}
-
-static const struct ide_port_info cyrix_chipsets[] __devinitdata = {
-	/* 0 */ DECLARE_CS_DEV("Cyrix 5510"),
-	/* 1 */ DECLARE_CS_DEV("Cyrix 5520")
+static const struct ide_port_info cyrix_chipset __devinitdata = {
+	.name		= DRV_NAME,
+	.port_ops	= &cs5520_port_ops,
+	.host_flags	= IDE_HFLAG_ISA_PORTS | IDE_HFLAG_CS5520,
+	.pio_mask	= ATA_PIO4,
 };
 
 /*
@@ -114,7 +109,7 @@ static const struct ide_port_info cyrix_chipsets[] __devinitdata = {
  
 static int __devinit cs5520_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
-	const struct ide_port_info *d = &cyrix_chipsets[id->driver_data];
+	const struct ide_port_info *d = &cyrix_chipset;
 	hw_regs_t hw[4], *hws[] = { NULL, NULL, NULL, NULL };
 
 	ide_setup_pci_noise(dev, d);
@@ -128,7 +123,8 @@ static int __devinit cs5520_init_one(struct pci_dev *dev, const struct pci_devic
 	}
 	pci_set_master(dev);
 	if (pci_set_dma_mask(dev, DMA_32BIT_MASK)) {
-		printk(KERN_WARNING "cs5520: No suitable DMA available.\n");
+		printk(KERN_WARNING "%s: No suitable DMA available.\n",
+			d->name);
 		return -ENODEV;
 	}
 
diff --git a/drivers/ide/pci/cs5530.c b/drivers/ide/pci/cs5530.c
index dff345c763e5..5543c8677a5a 100644
--- a/drivers/ide/pci/cs5530.c
+++ b/drivers/ide/pci/cs5530.c
@@ -22,6 +22,8 @@
 
 #include <asm/io.h>
 
+#define DRV_NAME "cs5530"
+
 /*
  * Here are the standard PIO mode 0-4 timings for each "format".
  * Format-0 uses fast data reg timings, with slower command reg timings.
@@ -243,7 +245,7 @@ static const struct ide_port_ops cs5530_port_ops = {
 };
 
 static const struct ide_port_info cs5530_chipset __devinitdata = {
-	.name		= "CS5530",
+	.name		= DRV_NAME,
 	.init_chipset	= init_chipset_cs5530,
 	.init_hwif	= init_hwif_cs5530,
 	.port_ops	= &cs5530_port_ops,
diff --git a/drivers/ide/pci/cs5535.c b/drivers/ide/pci/cs5535.c
index a73001391536..f7b50cdeefa6 100644
--- a/drivers/ide/pci/cs5535.c
+++ b/drivers/ide/pci/cs5535.c
@@ -26,6 +26,8 @@
 #include <linux/pci.h>
 #include <linux/ide.h>
 
+#define DRV_NAME "cs5535"
+
 #define MSR_ATAC_BASE		0x51300000
 #define ATAC_GLD_MSR_CAP	(MSR_ATAC_BASE+0)
 #define ATAC_GLD_MSR_CONFIG	(MSR_ATAC_BASE+0x01)
@@ -169,7 +171,7 @@ static const struct ide_port_ops cs5535_port_ops = {
 };
 
 static const struct ide_port_info cs5535_chipset __devinitdata = {
-	.name		= "CS5535",
+	.name		= DRV_NAME,
 	.port_ops	= &cs5535_port_ops,
 	.host_flags	= IDE_HFLAG_SINGLE | IDE_HFLAG_POST_SET_MODE,
 	.pio_mask	= ATA_PIO4,
diff --git a/drivers/ide/pci/cy82c693.c b/drivers/ide/pci/cy82c693.c
index 04f268866b6d..41c7f3351eb6 100644
--- a/drivers/ide/pci/cy82c693.c
+++ b/drivers/ide/pci/cy82c693.c
@@ -48,6 +48,8 @@
 
 #include <asm/io.h>
 
+#define DRV_NAME "cy82c693"
+
 /* the current version */
 #define CY82_VERSION	"CY82C693U driver v0.34 99-13-12 Andreas S. Krebs (akrebs@altavista.net)"
 
@@ -398,7 +400,7 @@ static const struct ide_port_ops cy82c693_port_ops = {
 };
 
 static const struct ide_port_info cy82c693_chipset __devinitdata = {
-	.name		= "CY82C693",
+	.name		= DRV_NAME,
 	.init_chipset	= init_chipset_cy82c693,
 	.init_iops	= init_iops_cy82c693,
 	.port_ops	= &cy82c693_port_ops,
diff --git a/drivers/ide/pci/generic.c b/drivers/ide/pci/generic.c
index e034e21af7fe..b07d4f4273b3 100644
--- a/drivers/ide/pci/generic.c
+++ b/drivers/ide/pci/generic.c
@@ -27,6 +27,8 @@
 #include <linux/ide.h>
 #include <linux/init.h>
 
+#define DRV_NAME "ide_pci_generic"
+
 static int ide_generic_all;		/* Set to claim all devices */
 
 module_param_named(all_generic_ide, ide_generic_all, bool, 0444);
@@ -34,9 +36,9 @@ MODULE_PARM_DESC(all_generic_ide, "IDE generic will claim all unknown PCI IDE st
 
 #define IDE_HFLAGS_UMC (IDE_HFLAG_NO_DMA | IDE_HFLAG_FORCE_LEGACY_IRQS)
 
-#define DECLARE_GENERIC_PCI_DEV(name_str, extra_flags) \
+#define DECLARE_GENERIC_PCI_DEV(extra_flags) \
 	{ \
-		.name		= name_str, \
+		.name		= DRV_NAME, \
 		.host_flags	= IDE_HFLAG_TRUST_BIOS_FOR_DMA | \
 				  extra_flags, \
 		.swdma_mask	= ATA_SWDMA2, \
@@ -45,10 +47,11 @@ MODULE_PARM_DESC(all_generic_ide, "IDE generic will claim all unknown PCI IDE st
 	}
 
 static const struct ide_port_info generic_chipsets[] __devinitdata = {
-	/*  0 */ DECLARE_GENERIC_PCI_DEV("Unknown",	0),
+	/*  0: Unknown */
+	DECLARE_GENERIC_PCI_DEV(0),
 
-	{	/* 1 */
-		.name		= "NS87410",
+	{	/* 1: NS87410 */
+		.name		= DRV_NAME,
 		.enablebits	= { {0x43, 0x08, 0x08}, {0x47, 0x08, 0x08} },
 		.host_flags	= IDE_HFLAG_TRUST_BIOS_FOR_DMA,
 		.swdma_mask	= ATA_SWDMA2,
@@ -56,17 +59,15 @@ static const struct ide_port_info generic_chipsets[] __devinitdata = {
 		.udma_mask	= ATA_UDMA6,
 	},
 
-	/*  2 */ DECLARE_GENERIC_PCI_DEV("SAMURAI",	0),
-	/*  3 */ DECLARE_GENERIC_PCI_DEV("HT6565",	0),
-	/*  4 */ DECLARE_GENERIC_PCI_DEV("UM8673F",	IDE_HFLAGS_UMC),
-	/*  5 */ DECLARE_GENERIC_PCI_DEV("UM8886A",	IDE_HFLAGS_UMC),
-	/*  6 */ DECLARE_GENERIC_PCI_DEV("UM8886BF",	IDE_HFLAGS_UMC),
-	/*  7 */ DECLARE_GENERIC_PCI_DEV("HINT_IDE",	0),
-	/*  8 */ DECLARE_GENERIC_PCI_DEV("VIA_IDE",	IDE_HFLAG_NO_AUTODMA),
-	/*  9 */ DECLARE_GENERIC_PCI_DEV("OPTI621V",	IDE_HFLAG_NO_AUTODMA),
+	/*  2: SAMURAI / HT6565 / HINT_IDE */
+	DECLARE_GENERIC_PCI_DEV(0),
+	/*  3: UM8673F / UM8886A / UM8886BF */
+	DECLARE_GENERIC_PCI_DEV(IDE_HFLAGS_UMC),
+	/*  4: VIA_IDE / OPTI621V / Piccolo010{2,3,5} */
+	DECLARE_GENERIC_PCI_DEV(IDE_HFLAG_NO_AUTODMA),
 
-	{	/* 10 */
-		.name		= "VIA8237SATA",
+	{	/* 5: VIA8237SATA */
+		.name		= DRV_NAME,
 		.host_flags	= IDE_HFLAG_TRUST_BIOS_FOR_DMA |
 				  IDE_HFLAG_OFF_BOARD,
 		.swdma_mask	= ATA_SWDMA2,
@@ -74,12 +75,8 @@ static const struct ide_port_info generic_chipsets[] __devinitdata = {
 		.udma_mask	= ATA_UDMA6,
 	},
 
-	/* 11 */ DECLARE_GENERIC_PCI_DEV("Piccolo0102",	IDE_HFLAG_NO_AUTODMA),
-	/* 12 */ DECLARE_GENERIC_PCI_DEV("Piccolo0103",	IDE_HFLAG_NO_AUTODMA),
-	/* 13 */ DECLARE_GENERIC_PCI_DEV("Piccolo0105",	IDE_HFLAG_NO_AUTODMA),
-
-	{	/* 14 */
-		.name		= "Revolution",
+	{	/* 6: Revolution */
+		.name		= DRV_NAME,
 		.host_flags	= IDE_HFLAG_CLEAR_SIMPLEX |
 				  IDE_HFLAG_TRUST_BIOS_FOR_DMA |
 				  IDE_HFLAG_OFF_BOARD,
@@ -147,20 +144,20 @@ out:
 static const struct pci_device_id generic_pci_tbl[] = {
 	{ PCI_VDEVICE(NS,	PCI_DEVICE_ID_NS_87410),		 1 },
 	{ PCI_VDEVICE(PCTECH,	PCI_DEVICE_ID_PCTECH_SAMURAI_IDE),	 2 },
-	{ PCI_VDEVICE(HOLTEK,	PCI_DEVICE_ID_HOLTEK_6565),		 3 },
-	{ PCI_VDEVICE(UMC,	PCI_DEVICE_ID_UMC_UM8673F),		 4 },
-	{ PCI_VDEVICE(UMC,	PCI_DEVICE_ID_UMC_UM8886A),		 5 },
-	{ PCI_VDEVICE(UMC,	PCI_DEVICE_ID_UMC_UM8886BF),		 6 },
-	{ PCI_VDEVICE(HINT,	PCI_DEVICE_ID_HINT_VXPROII_IDE),	 7 },
-	{ PCI_VDEVICE(VIA,	PCI_DEVICE_ID_VIA_82C561),		 8 },
-	{ PCI_VDEVICE(OPTI,	PCI_DEVICE_ID_OPTI_82C558),		 9 },
+	{ PCI_VDEVICE(HOLTEK,	PCI_DEVICE_ID_HOLTEK_6565),		 2 },
+	{ PCI_VDEVICE(UMC,	PCI_DEVICE_ID_UMC_UM8673F),		 3 },
+	{ PCI_VDEVICE(UMC,	PCI_DEVICE_ID_UMC_UM8886A),		 3 },
+	{ PCI_VDEVICE(UMC,	PCI_DEVICE_ID_UMC_UM8886BF),		 3 },
+	{ PCI_VDEVICE(HINT,	PCI_DEVICE_ID_HINT_VXPROII_IDE),	 2 },
+	{ PCI_VDEVICE(VIA,	PCI_DEVICE_ID_VIA_82C561),		 4 },
+	{ PCI_VDEVICE(OPTI,	PCI_DEVICE_ID_OPTI_82C558),		 4 },
 #ifdef CONFIG_BLK_DEV_IDE_SATA
-	{ PCI_VDEVICE(VIA,	PCI_DEVICE_ID_VIA_8237_SATA),		10 },
+	{ PCI_VDEVICE(VIA,	PCI_DEVICE_ID_VIA_8237_SATA),		 5 },
 #endif
-	{ PCI_VDEVICE(TOSHIBA,	PCI_DEVICE_ID_TOSHIBA_PICCOLO),		11 },
-	{ PCI_VDEVICE(TOSHIBA,	PCI_DEVICE_ID_TOSHIBA_PICCOLO_1),	12 },
-	{ PCI_VDEVICE(TOSHIBA,	PCI_DEVICE_ID_TOSHIBA_PICCOLO_2),	13 },
-	{ PCI_VDEVICE(NETCELL,	PCI_DEVICE_ID_REVOLUTION),		14 },
+	{ PCI_VDEVICE(TOSHIBA,	PCI_DEVICE_ID_TOSHIBA_PICCOLO),		 4 },
+	{ PCI_VDEVICE(TOSHIBA,	PCI_DEVICE_ID_TOSHIBA_PICCOLO_1),	 4 },
+	{ PCI_VDEVICE(TOSHIBA,	PCI_DEVICE_ID_TOSHIBA_PICCOLO_2),	 4 },
+	{ PCI_VDEVICE(NETCELL,	PCI_DEVICE_ID_REVOLUTION),		 6 },
 	/*
 	 * Must come last.  If you add entries adjust
 	 * this table and generic_chipsets[] appropriately.
diff --git a/drivers/ide/pci/hpt34x.c b/drivers/ide/pci/hpt34x.c
index b52f8339102e..baabb4ce0d78 100644
--- a/drivers/ide/pci/hpt34x.c
+++ b/drivers/ide/pci/hpt34x.c
@@ -33,6 +33,8 @@
 #include <linux/init.h>
 #include <linux/ide.h>
 
+#define DRV_NAME "hpt34x"
+
 #define HPT343_DEBUG_DRIVE_INFO		0
 
 static void hpt34x_set_mode(ide_drive_t *drive, const u8 speed)
@@ -126,15 +128,15 @@ static const struct ide_port_ops hpt34x_port_ops = {
 	 IDE_HFLAG_NO_AUTODMA)
 
 static const struct ide_port_info hpt34x_chipsets[] __devinitdata = {
-	{ /* 0 */
-		.name		= "HPT343",
+	{ /* 0: HPT343 */
+		.name		= DRV_NAME,
 		.init_chipset	= init_chipset_hpt34x,
 		.port_ops	= &hpt34x_port_ops,
 		.host_flags	= IDE_HFLAGS_HPT34X | IDE_HFLAG_NON_BOOTABLE,
 		.pio_mask	= ATA_PIO5,
 	},
-	{ /* 1 */
-		.name		= "HPT345",
+	{ /* 1: HPT345 */
+		.name		= DRV_NAME,
 		.init_chipset	= init_chipset_hpt34x,
 		.port_ops	= &hpt34x_port_ops,
 		.host_flags	= IDE_HFLAGS_HPT34X | IDE_HFLAG_OFF_BOARD,
diff --git a/drivers/ide/pci/hpt366.c b/drivers/ide/pci/hpt366.c
index b8004c331edb..6a1c65c3be3e 100644
--- a/drivers/ide/pci/hpt366.c
+++ b/drivers/ide/pci/hpt366.c
@@ -131,6 +131,8 @@
 #include <asm/uaccess.h>
 #include <asm/io.h>
 
+#define DRV_NAME "hpt366"
+
 /* various tuning parameters */
 #define HPT_RESET_STATE_ENGINE
 #undef	HPT_DELAY_INTERRUPT
@@ -1362,7 +1364,7 @@ static void __devinit hpt374_init(struct pci_dev *dev, struct pci_dev *dev2)
 	if (dev2->irq != dev->irq) {
 		/* FIXME: we need a core pci_set_interrupt() */
 		dev2->irq = dev->irq;
-		printk(KERN_INFO "HPT374 %s: PCI config space interrupt "
+		printk(KERN_INFO DRV_NAME " %s: PCI config space interrupt "
 			"fixed\n", pci_name(dev2));
 	}
 }
@@ -1398,7 +1400,7 @@ static int __devinit hpt36x_init(struct pci_dev *dev, struct pci_dev *dev2)
 	pci_read_config_byte(dev2, PCI_INTERRUPT_PIN, &pin2);
 
 	if (pin1 != pin2 && dev->irq == dev2->irq) {
-		printk(KERN_INFO "HPT36x %s: onboard version of chipset, "
+		printk(KERN_INFO DRV_NAME " %s: onboard version of chipset, "
 			"pin1=%d pin2=%d\n", pci_name(dev), pin1, pin2);
 		return 1;
 	}
@@ -1454,8 +1456,8 @@ static const struct ide_dma_ops hpt36x_dma_ops = {
 };
 
 static const struct ide_port_info hpt366_chipsets[] __devinitdata = {
-	{	/* 0 */
-		.name		= "HPT36x",
+	{	/* 0: HPT36x */
+		.name		= DRV_NAME,
 		.init_chipset	= init_chipset_hpt366,
 		.init_hwif	= init_hwif_hpt366,
 		.init_dma	= init_dma_hpt366,
@@ -1471,53 +1473,9 @@ static const struct ide_port_info hpt366_chipsets[] __devinitdata = {
 		.host_flags	= IDE_HFLAGS_HPT3XX | IDE_HFLAG_SINGLE,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
-	},{	/* 1 */
-		.name		= "HPT372A",
-		.init_chipset	= init_chipset_hpt366,
-		.init_hwif	= init_hwif_hpt366,
-		.init_dma	= init_dma_hpt366,
-		.enablebits	= {{0x50,0x04,0x04}, {0x54,0x04,0x04}},
-		.port_ops	= &hpt3xx_port_ops,
-		.dma_ops	= &hpt37x_dma_ops,
-		.host_flags	= IDE_HFLAGS_HPT3XX,
-		.pio_mask	= ATA_PIO4,
-		.mwdma_mask	= ATA_MWDMA2,
-	},{	/* 2 */
-		.name		= "HPT302",
-		.init_chipset	= init_chipset_hpt366,
-		.init_hwif	= init_hwif_hpt366,
-		.init_dma	= init_dma_hpt366,
-		.enablebits	= {{0x50,0x04,0x04}, {0x54,0x04,0x04}},
-		.port_ops	= &hpt3xx_port_ops,
-		.dma_ops	= &hpt37x_dma_ops,
-		.host_flags	= IDE_HFLAGS_HPT3XX,
-		.pio_mask	= ATA_PIO4,
-		.mwdma_mask	= ATA_MWDMA2,
-	},{	/* 3 */
-		.name		= "HPT371",
-		.init_chipset	= init_chipset_hpt366,
-		.init_hwif	= init_hwif_hpt366,
-		.init_dma	= init_dma_hpt366,
-		.enablebits	= {{0x50,0x04,0x04}, {0x54,0x04,0x04}},
-		.port_ops	= &hpt3xx_port_ops,
-		.dma_ops	= &hpt37x_dma_ops,
-		.host_flags	= IDE_HFLAGS_HPT3XX,
-		.pio_mask	= ATA_PIO4,
-		.mwdma_mask	= ATA_MWDMA2,
-	},{	/* 4 */
-		.name		= "HPT374",
-		.init_chipset	= init_chipset_hpt366,
-		.init_hwif	= init_hwif_hpt366,
-		.init_dma	= init_dma_hpt366,
-		.enablebits	= {{0x50,0x04,0x04}, {0x54,0x04,0x04}},
-		.udma_mask	= ATA_UDMA5,
-		.port_ops	= &hpt3xx_port_ops,
-		.dma_ops	= &hpt37x_dma_ops,
-		.host_flags	= IDE_HFLAGS_HPT3XX,
-		.pio_mask	= ATA_PIO4,
-		.mwdma_mask	= ATA_MWDMA2,
-	},{	/* 5 */
-		.name		= "HPT372N",
+	},
+	{	/* 1: HPT3xx */
+		.name		= DRV_NAME,
 		.init_chipset	= init_chipset_hpt366,
 		.init_hwif	= init_hwif_hpt366,
 		.init_dma	= init_dma_hpt366,
@@ -1583,9 +1541,10 @@ static int __devinit hpt366_init_one(struct pci_dev *dev, const struct pci_devic
 		break;
 	}
 
-	d = hpt366_chipsets[idx];
+	printk(KERN_INFO DRV_NAME ": %s chipset detected\n", info->chip_name);
+
+	d = hpt366_chipsets[min_t(u8, idx, 1)];
 
-	d.name = info->chip_name;
 	d.udma_mask = info->udma_mask;
 
 	/* fixup ->dma_ops for HPT370/HPT370A */
diff --git a/drivers/ide/pci/it8213.c b/drivers/ide/pci/it8213.c
index 451b87fd8217..6eba8f188264 100644
--- a/drivers/ide/pci/it8213.c
+++ b/drivers/ide/pci/it8213.c
@@ -14,6 +14,8 @@
 #include <linux/ide.h>
 #include <linux/init.h>
 
+#define DRV_NAME "it8213"
+
 /**
  *	it8213_set_pio_mode	-	set host controller for PIO mode
  *	@drive: drive
@@ -156,7 +158,7 @@ static const struct ide_port_ops it8213_port_ops = {
 };
 
 static const struct ide_port_info it8213_chipset __devinitdata = {
-	.name		= "IT8213",
+	.name		= DRV_NAME,
 	.enablebits	= { {0x41, 0x80, 0x80} },
 	.port_ops	= &it8213_port_ops,
 	.host_flags	= IDE_HFLAG_SINGLE,
diff --git a/drivers/ide/pci/it821x.c b/drivers/ide/pci/it821x.c
index a3d8959436c5..74173352741f 100644
--- a/drivers/ide/pci/it821x.c
+++ b/drivers/ide/pci/it821x.c
@@ -67,6 +67,8 @@
 #include <linux/ide.h>
 #include <linux/init.h>
 
+#define DRV_NAME "it821x"
+
 struct it821x_dev
 {
 	unsigned int smart:1,		/* Are we in smart raid mode */
@@ -569,7 +571,7 @@ static void __devinit init_hwif_it821x(ide_hwif_t *hwif)
 		idev->timing10 = 1;
 		hwif->host_flags |= IDE_HFLAG_NO_ATAPI_DMA;
 		if (idev->smart == 0)
-			printk(KERN_WARNING "it821x %s: revision 0x10, "
+			printk(KERN_WARNING DRV_NAME " %s: revision 0x10, "
 				"workarounds activated\n", pci_name(dev));
 	}
 
@@ -610,12 +612,12 @@ static unsigned int __devinit init_chipset_it821x(struct pci_dev *dev, const cha
 
 	/* Force the card into bypass mode if so requested */
 	if (it8212_noraid) {
-		printk(KERN_INFO "it821x %s: forcing bypass mode\n",
+		printk(KERN_INFO DRV_NAME " %s: forcing bypass mode\n",
 			pci_name(dev));
 		it8212_disable_raid(dev);
 	}
 	pci_read_config_byte(dev, 0x50, &conf);
-	printk(KERN_INFO "it821x %s: controller in %s mode\n",
+	printk(KERN_INFO DRV_NAME " %s: controller in %s mode\n",
 		pci_name(dev), mode[conf & 1]);
 	return 0;
 }
@@ -629,7 +631,7 @@ static const struct ide_port_ops it821x_port_ops = {
 };
 
 static const struct ide_port_info it821x_chipset __devinitdata = {
-	.name		= "IT821X",
+	.name		= DRV_NAME,
 	.init_chipset	= init_chipset_it821x,
 	.init_hwif	= init_hwif_it821x,
 	.port_ops	= &it821x_port_ops,
@@ -652,7 +654,7 @@ static int __devinit it821x_init_one(struct pci_dev *dev, const struct pci_devic
 
 	itdevs = kzalloc(2 * sizeof(*itdevs), GFP_KERNEL);
 	if (itdevs == NULL) {
-		printk(KERN_ERR "it821x %s: out of memory\n", pci_name(dev));
+		printk(KERN_ERR DRV_NAME " %s: out of memory\n", pci_name(dev));
 		return -ENOMEM;
 	}
 
diff --git a/drivers/ide/pci/jmicron.c b/drivers/ide/pci/jmicron.c
index 39e221b076e2..545b6e172d9b 100644
--- a/drivers/ide/pci/jmicron.c
+++ b/drivers/ide/pci/jmicron.c
@@ -12,6 +12,8 @@
 #include <linux/ide.h>
 #include <linux/init.h>
 
+#define DRV_NAME "jmicron"
+
 typedef enum {
 	PORT_PATA0 = 0,
 	PORT_PATA1 = 1,
@@ -102,7 +104,7 @@ static const struct ide_port_ops jmicron_port_ops = {
 };
 
 static const struct ide_port_info jmicron_chipset __devinitdata = {
-	.name		= "JMB",
+	.name		= DRV_NAME,
 	.enablebits	= { { 0x40, 0x01, 0x01 }, { 0x40, 0x10, 0x10 } },
 	.port_ops	= &jmicron_port_ops,
 	.pio_mask	= ATA_PIO5,
diff --git a/drivers/ide/pci/ns87415.c b/drivers/ide/pci/ns87415.c
index afcc742a63ac..ffefcd15196c 100644
--- a/drivers/ide/pci/ns87415.c
+++ b/drivers/ide/pci/ns87415.c
@@ -19,6 +19,8 @@
 
 #include <asm/io.h>
 
+#define DRV_NAME "ns87415"
+
 #ifdef CONFIG_SUPERIO
 /* SUPERIO 87560 is a PoS chip that NatSem denies exists.
  * Unfortunately, it's built-in on all Astro-based PA-RISC workstations
@@ -305,7 +307,7 @@ static const struct ide_dma_ops ns87415_dma_ops = {
 };
 
 static const struct ide_port_info ns87415_chipset __devinitdata = {
-	.name		= "NS87415",
+	.name		= DRV_NAME,
 	.init_hwif	= init_hwif_ns87415,
 	.port_ops	= &ns87415_port_ops,
 	.dma_ops	= &ns87415_dma_ops,
diff --git a/drivers/ide/pci/opti621.c b/drivers/ide/pci/opti621.c
index 4895f2ff3f05..e28e672ddafc 100644
--- a/drivers/ide/pci/opti621.c
+++ b/drivers/ide/pci/opti621.c
@@ -90,6 +90,8 @@
 
 #include <asm/io.h>
 
+#define DRV_NAME "opti621"
+
 #define READ_REG 0	/* index of Read cycle timing register */
 #define WRITE_REG 1	/* index of Write cycle timing register */
 #define CNTRL_REG 3	/* index of Control register */
@@ -200,7 +202,7 @@ static const struct ide_port_ops opti621_port_ops = {
 };
 
 static const struct ide_port_info opti621_chipset __devinitdata = {
-	.name		= "OPTI621/X",
+	.name		= DRV_NAME,
 	.enablebits	= { {0x45, 0x80, 0x00}, {0x40, 0x08, 0x00} },
 	.port_ops	= &opti621_port_ops,
 	.host_flags	= IDE_HFLAG_NO_DMA,
diff --git a/drivers/ide/pci/pdc202xx_new.c b/drivers/ide/pci/pdc202xx_new.c
index 4c2b669d7de6..1f6791957227 100644
--- a/drivers/ide/pci/pdc202xx_new.c
+++ b/drivers/ide/pci/pdc202xx_new.c
@@ -31,6 +31,8 @@
 #include <asm/pci-bridge.h>
 #endif
 
+#define DRV_NAME "pdc202xx_new"
+
 #undef DEBUG
 
 #ifdef DEBUG
@@ -458,7 +460,7 @@ static struct pci_dev * __devinit pdc20270_get_dev2(struct pci_dev *dev)
 
 		if (dev2->irq != dev->irq) {
 			dev2->irq = dev->irq;
-			printk(KERN_INFO "PDC20270 %s: PCI config space "
+			printk(KERN_INFO DRV_NAME " %s: PCI config space "
 				"interrupt fixed\n", pci_name(dev));
 		}
 
@@ -476,9 +478,9 @@ static const struct ide_port_ops pdcnew_port_ops = {
 	.cable_detect		= pdcnew_cable_detect,
 };
 
-#define DECLARE_PDCNEW_DEV(name_str, udma) \
+#define DECLARE_PDCNEW_DEV(udma) \
 	{ \
-		.name		= name_str, \
+		.name		= DRV_NAME, \
 		.init_chipset	= init_chipset_pdcnew, \
 		.port_ops	= &pdcnew_port_ops, \
 		.host_flags	= IDE_HFLAG_POST_SET_MODE | \
@@ -490,13 +492,8 @@ static const struct ide_port_ops pdcnew_port_ops = {
 	}
 
 static const struct ide_port_info pdcnew_chipsets[] __devinitdata = {
-	/* 0 */ DECLARE_PDCNEW_DEV("PDC20268", ATA_UDMA5),
-	/* 1 */ DECLARE_PDCNEW_DEV("PDC20269", ATA_UDMA6),
-	/* 2 */ DECLARE_PDCNEW_DEV("PDC20270", ATA_UDMA5),
-	/* 3 */ DECLARE_PDCNEW_DEV("PDC20271", ATA_UDMA6),
-	/* 4 */ DECLARE_PDCNEW_DEV("PDC20275", ATA_UDMA6),
-	/* 5 */ DECLARE_PDCNEW_DEV("PDC20276", ATA_UDMA6),
-	/* 6 */ DECLARE_PDCNEW_DEV("PDC20277", ATA_UDMA6),
+	/* 0: PDC202{68,70} */		DECLARE_PDCNEW_DEV(ATA_UDMA5),
+	/* 1: PDC202{69,71,75,76,77} */	DECLARE_PDCNEW_DEV(ATA_UDMA6),
 };
 
 /**
@@ -510,13 +507,10 @@ static const struct ide_port_info pdcnew_chipsets[] __devinitdata = {
  
 static int __devinit pdc202new_init_one(struct pci_dev *dev, const struct pci_device_id *id)
 {
-	const struct ide_port_info *d;
+	const struct ide_port_info *d = &pdcnew_chipsets[id->driver_data];
 	struct pci_dev *bridge = dev->bus->self;
-	u8 idx = id->driver_data;
 
-	d = &pdcnew_chipsets[idx];
-
-	if (idx == 2 && bridge &&
+	if (dev->device == PCI_DEVICE_ID_PROMISE_20270 && bridge &&
 	    bridge->vendor == PCI_VENDOR_ID_DEC &&
 	    bridge->device == PCI_DEVICE_ID_DEC_21150) {
 		struct pci_dev *dev2;
@@ -534,11 +528,11 @@ static int __devinit pdc202new_init_one(struct pci_dev *dev, const struct pci_de
 		}
 	}
 
-	if (idx == 5 && bridge &&
+	if (dev->device == PCI_DEVICE_ID_PROMISE_20276 && bridge &&
 	    bridge->vendor == PCI_VENDOR_ID_INTEL &&
 	    (bridge->device == PCI_DEVICE_ID_INTEL_I960 ||
 	     bridge->device == PCI_DEVICE_ID_INTEL_I960RM)) {
-		printk(KERN_INFO "PDC20276 %s: attached to I2O RAID controller,"
+		printk(KERN_INFO DRV_NAME " %s: attached to I2O RAID controller,"
 			" skipping\n", pci_name(dev));
 		return -ENODEV;
 	}
@@ -558,11 +552,11 @@ static void __devexit pdc202new_remove(struct pci_dev *dev)
 static const struct pci_device_id pdc202new_pci_tbl[] = {
 	{ PCI_VDEVICE(PROMISE, PCI_DEVICE_ID_PROMISE_20268), 0 },
 	{ PCI_VDEVICE(PROMISE, PCI_DEVICE_ID_PROMISE_20269), 1 },
-	{ PCI_VDEVICE(PROMISE, PCI_DEVICE_ID_PROMISE_20270), 2 },
-	{ PCI_VDEVICE(PROMISE, PCI_DEVICE_ID_PROMISE_20271), 3 },
-	{ PCI_VDEVICE(PROMISE, PCI_DEVICE_ID_PROMISE_20275), 4 },
-	{ PCI_VDEVICE(PROMISE, PCI_DEVICE_ID_PROMISE_20276), 5 },
-	{ PCI_VDEVICE(PROMISE, PCI_DEVICE_ID_PROMISE_20277), 6 },
+	{ PCI_VDEVICE(PROMISE, PCI_DEVICE_ID_PROMISE_20270), 0 },
+	{ PCI_VDEVICE(PROMISE, PCI_DEVICE_ID_PROMISE_20271), 1 },
+	{ PCI_VDEVICE(PROMISE, PCI_DEVICE_ID_PROMISE_20275), 1 },
+	{ PCI_VDEVICE(PROMISE, PCI_DEVICE_ID_PROMISE_20276), 1 },
+	{ PCI_VDEVICE(PROMISE, PCI_DEVICE_ID_PROMISE_20277), 1 },
 	{ 0, },
 };
 MODULE_DEVICE_TABLE(pci, pdc202new_pci_tbl);
diff --git a/drivers/ide/pci/pdc202xx_old.c b/drivers/ide/pci/pdc202xx_old.c
index 5cb2731047e9..da92d127868f 100644
--- a/drivers/ide/pci/pdc202xx_old.c
+++ b/drivers/ide/pci/pdc202xx_old.c
@@ -20,6 +20,8 @@
 
 #include <asm/io.h>
 
+#define DRV_NAME "pdc202xx_old"
+
 #define PDC202XX_DEBUG_DRIVE_INFO	0
 
 static const char *pdc_quirk_drives[] = {
@@ -350,9 +352,9 @@ static const struct ide_dma_ops pdc2026x_dma_ops = {
 	.dma_timeout		= pdc202xx_dma_timeout,
 };
 
-#define DECLARE_PDC2026X_DEV(name_str, udma, extra_flags) \
+#define DECLARE_PDC2026X_DEV(udma, extra_flags) \
 	{ \
-		.name		= name_str, \
+		.name		= DRV_NAME, \
 		.init_chipset	= init_chipset_pdc202xx, \
 		.port_ops	= &pdc2026x_port_ops, \
 		.dma_ops	= &pdc2026x_dma_ops, \
@@ -363,8 +365,8 @@ static const struct ide_dma_ops pdc2026x_dma_ops = {
 	}
 
 static const struct ide_port_info pdc202xx_chipsets[] __devinitdata = {
-	{	/* 0 */
-		.name		= "PDC20246",
+	{	/* 0: PDC20246 */
+		.name		= DRV_NAME,
 		.init_chipset	= init_chipset_pdc202xx,
 		.port_ops	= &pdc20246_port_ops,
 		.dma_ops	= &pdc20246_dma_ops,
@@ -374,10 +376,10 @@ static const struct ide_port_info pdc202xx_chipsets[] __devinitdata = {
 		.udma_mask	= ATA_UDMA2,
 	},
 
-	/* 1 */ DECLARE_PDC2026X_DEV("PDC20262", ATA_UDMA4, 0),
-	/* 2 */ DECLARE_PDC2026X_DEV("PDC20263", ATA_UDMA4, 0),
-	/* 3 */ DECLARE_PDC2026X_DEV("PDC20265", ATA_UDMA5, IDE_HFLAG_RQSIZE_256),
-	/* 4 */ DECLARE_PDC2026X_DEV("PDC20267", ATA_UDMA5, IDE_HFLAG_RQSIZE_256),
+	/* 1: PDC2026{2,3} */
+	DECLARE_PDC2026X_DEV(ATA_UDMA4, 0),
+	/* 2: PDC2026{5,7} */
+	DECLARE_PDC2026X_DEV(ATA_UDMA5, IDE_HFLAG_RQSIZE_256),
 };
 
 /**
@@ -396,17 +398,17 @@ static int __devinit pdc202xx_init_one(struct pci_dev *dev, const struct pci_dev
 
 	d = &pdc202xx_chipsets[idx];
 
-	if (idx < 3)
+	if (idx < 2)
 		pdc202ata4_fixup_irq(dev, d->name);
 
-	if (idx == 3) {
+	if (dev->vendor == PCI_DEVICE_ID_PROMISE_20265) {
 		struct pci_dev *bridge = dev->bus->self;
 
 		if (bridge &&
 		    bridge->vendor == PCI_VENDOR_ID_INTEL &&
 		    (bridge->device == PCI_DEVICE_ID_INTEL_I960 ||
 		     bridge->device == PCI_DEVICE_ID_INTEL_I960RM)) {
-			printk(KERN_INFO "pdc202xx_old %s: skipping Promise "
+			printk(KERN_INFO DRV_NAME " %s: skipping Promise "
 				"PDC20265 attached to I2O RAID controller\n",
 				pci_name(dev));
 			return -ENODEV;
@@ -419,9 +421,9 @@ static int __devinit pdc202xx_init_one(struct pci_dev *dev, const struct pci_dev
 static const struct pci_device_id pdc202xx_pci_tbl[] = {
 	{ PCI_VDEVICE(PROMISE, PCI_DEVICE_ID_PROMISE_20246), 0 },
 	{ PCI_VDEVICE(PROMISE, PCI_DEVICE_ID_PROMISE_20262), 1 },
-	{ PCI_VDEVICE(PROMISE, PCI_DEVICE_ID_PROMISE_20263), 2 },
-	{ PCI_VDEVICE(PROMISE, PCI_DEVICE_ID_PROMISE_20265), 3 },
-	{ PCI_VDEVICE(PROMISE, PCI_DEVICE_ID_PROMISE_20267), 4 },
+	{ PCI_VDEVICE(PROMISE, PCI_DEVICE_ID_PROMISE_20263), 1 },
+	{ PCI_VDEVICE(PROMISE, PCI_DEVICE_ID_PROMISE_20265), 2 },
+	{ PCI_VDEVICE(PROMISE, PCI_DEVICE_ID_PROMISE_20267), 2 },
 	{ 0, },
 };
 MODULE_DEVICE_TABLE(pci, pdc202xx_pci_tbl);
diff --git a/drivers/ide/pci/piix.c b/drivers/ide/pci/piix.c
index c16b1ab4d4ff..9eb411f5c358 100644
--- a/drivers/ide/pci/piix.c
+++ b/drivers/ide/pci/piix.c
@@ -54,6 +54,8 @@
 
 #include <asm/io.h>
 
+#define DRV_NAME "piix"
+
 static int no_piix_dma;
 
 /**
@@ -314,9 +316,9 @@ static const struct ide_port_ops piix_port_ops = {
  #define IDE_HFLAGS_PIIX 0
 #endif
 
-#define DECLARE_PIIX_DEV(name_str, udma) \
+#define DECLARE_PIIX_DEV(udma) \
 	{						\
-		.name		= name_str,		\
+		.name		= DRV_NAME,		\
 		.init_hwif	= init_hwif_piix,	\
 		.enablebits	= {{0x41,0x80,0x80}, {0x43,0x80,0x80}}, \
 		.port_ops	= &piix_port_ops,	\
@@ -327,9 +329,9 @@ static const struct ide_port_ops piix_port_ops = {
 		.udma_mask	= udma,			\
 	}
 
-#define DECLARE_ICH_DEV(name_str, udma) \
+#define DECLARE_ICH_DEV(udma) \
 	{ \
-		.name		= name_str, \
+		.name		= DRV_NAME, \
 		.init_chipset	= init_chipset_ich, \
 		.init_hwif	= init_hwif_ich, \
 		.enablebits	= {{0x41,0x80,0x80}, {0x43,0x80,0x80}}, \
@@ -342,45 +344,31 @@ static const struct ide_port_ops piix_port_ops = {
 	}
 
 static const struct ide_port_info piix_pci_info[] __devinitdata = {
-	/*  0 */ DECLARE_PIIX_DEV("PIIXa",	0x00),	/* no udma */
-	/*  1 */ DECLARE_PIIX_DEV("PIIXb",	0x00),	/* no udma */
-
-	/*  2 */
+	/* 0: MPIIX */
 	{	/*
 		 * MPIIX actually has only a single IDE channel mapped to
 		 * the primary or secondary ports depending on the value
 		 * of the bit 14 of the IDETIM register at offset 0x6c
 		 */
-		.name		= "MPIIX",
+		.name		= DRV_NAME,
 		.enablebits	= {{0x6d,0xc0,0x80}, {0x6d,0xc0,0xc0}},
 		.host_flags	= IDE_HFLAG_ISA_PORTS | IDE_HFLAG_NO_DMA |
 				  IDE_HFLAGS_PIIX,
 		.pio_mask	= ATA_PIO4,
 		/* This is a painful system best to let it self tune for now */
 	},
-
-	/*  3 */ DECLARE_PIIX_DEV("PIIX3",	0x00),	/* no udma */
-	/*  4 */ DECLARE_PIIX_DEV("PIIX4",	ATA_UDMA2),
-	/*  5 */ DECLARE_ICH_DEV("ICH0",	ATA_UDMA2),
-	/*  6 */ DECLARE_PIIX_DEV("PIIX4",	ATA_UDMA2),
-	/*  7 */ DECLARE_ICH_DEV("ICH",		ATA_UDMA4),
-	/*  8 */ DECLARE_PIIX_DEV("PIIX4",	ATA_UDMA4),
-	/*  9 */ DECLARE_PIIX_DEV("PIIX4",	ATA_UDMA2),
-	/* 10 */ DECLARE_ICH_DEV("ICH2",	ATA_UDMA5),
-	/* 11 */ DECLARE_ICH_DEV("ICH2M",	ATA_UDMA5),
-	/* 12 */ DECLARE_ICH_DEV("ICH3M",	ATA_UDMA5),
-	/* 13 */ DECLARE_ICH_DEV("ICH3",	ATA_UDMA5),
-	/* 14 */ DECLARE_ICH_DEV("ICH4",	ATA_UDMA5),
-	/* 15 */ DECLARE_ICH_DEV("ICH5",	ATA_UDMA5),
-	/* 16 */ DECLARE_ICH_DEV("C-ICH",	ATA_UDMA5),
-	/* 17 */ DECLARE_ICH_DEV("ICH4",	ATA_UDMA5),
-	/* 18 */ DECLARE_ICH_DEV("ICH5-SATA",	ATA_UDMA5),
-	/* 19 */ DECLARE_ICH_DEV("ICH5",	ATA_UDMA5),
-	/* 20 */ DECLARE_ICH_DEV("ICH6",	ATA_UDMA5),
-	/* 21 */ DECLARE_ICH_DEV("ICH7",	ATA_UDMA5),
-	/* 22 */ DECLARE_ICH_DEV("ICH4",	ATA_UDMA5),
-	/* 23 */ DECLARE_ICH_DEV("ESB2",	ATA_UDMA5),
-	/* 24 */ DECLARE_ICH_DEV("ICH8M",	ATA_UDMA5),
+	/* 1: PIIXa/PIIXb/PIIX3 */
+	DECLARE_PIIX_DEV(0x00), /* no udma */
+	/* 2: PIIX4 */
+	DECLARE_PIIX_DEV(ATA_UDMA2),
+	/* 3: ICH0 */
+	DECLARE_ICH_DEV(ATA_UDMA2),
+	/* 4: ICH */
+	DECLARE_ICH_DEV(ATA_UDMA4),
+	/* 5: PIIX4 */
+	DECLARE_PIIX_DEV(ATA_UDMA4),
+	/* 6: ICH[2-7]/ICH[2-3]M/C-ICH/ICH5-SATA/ESB2/ICH8M */
+	DECLARE_ICH_DEV(ATA_UDMA5),
 };
 
 /**
@@ -421,39 +409,39 @@ static void __devinit piix_check_450nx(void)
 			no_piix_dma = 2;
 	}
 	if(no_piix_dma)
-		printk(KERN_WARNING "piix: 450NX errata present, disabling IDE DMA.\n");
+		printk(KERN_WARNING DRV_NAME ": 450NX errata present, disabling IDE DMA.\n");
 	if(no_piix_dma == 2)
-		printk(KERN_WARNING "piix: A BIOS update may resolve this.\n");
+		printk(KERN_WARNING DRV_NAME ": A BIOS update may resolve this.\n");
 }		
 
 static const struct pci_device_id piix_pci_tbl[] = {
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82371FB_0),   0 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82371FB_1),   1 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82371MX),     2 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82371SB_1),   3 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82371AB),     4 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801AB_1),   5 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82443MX_1),   6 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801AA_1),   7 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82372FB_1),   8 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82451NX),     9 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801BA_9),  10 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801BA_8),  11 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801CA_10), 12 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801CA_11), 13 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801DB_11), 14 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801EB_11), 15 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801E_11),  16 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801DB_10), 17 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82371FB_0),  1 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82371FB_1),  1 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82371MX),    0 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82371SB_1),  1 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82371AB),    2 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801AB_1),  3 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82443MX_1),  2 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801AA_1),  4 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82372FB_1),  5 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82451NX),    2 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801BA_9),  6 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801BA_8),  6 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801CA_10), 6 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801CA_11), 6 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801DB_11), 6 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801EB_11), 6 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801E_11),  6 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801DB_10), 6 },
 #ifdef CONFIG_BLK_DEV_IDE_SATA
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801EB_1),  18 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801EB_1),  6 },
 #endif
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ESB_2),      19 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ICH6_19),    20 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ICH7_21),    21 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801DB_1),  22 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ESB2_18),    23 },
-	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ICH8_6),     24 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ESB_2),      6 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ICH6_19),    6 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ICH7_21),    6 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_82801DB_1),  6 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ESB2_18),    6 },
+	{ PCI_VDEVICE(INTEL, PCI_DEVICE_ID_INTEL_ICH8_6),     6 },
 	{ 0, },
 };
 MODULE_DEVICE_TABLE(pci, piix_pci_tbl);
diff --git a/drivers/ide/pci/rz1000.c b/drivers/ide/pci/rz1000.c
index f7a3b9aff29d..8d11ee838a2a 100644
--- a/drivers/ide/pci/rz1000.c
+++ b/drivers/ide/pci/rz1000.c
@@ -21,6 +21,8 @@
 #include <linux/ide.h>
 #include <linux/init.h>
 
+#define DRV_NAME "rz1000"
+
 static void __devinit init_hwif_rz1000 (ide_hwif_t *hwif)
 {
 	struct pci_dev *dev = to_pci_dev(hwif->dev);
@@ -40,7 +42,7 @@ static void __devinit init_hwif_rz1000 (ide_hwif_t *hwif)
 }
 
 static const struct ide_port_info rz1000_chipset __devinitdata = {
-	.name		= "RZ100x",
+	.name		= DRV_NAME,
 	.init_hwif	= init_hwif_rz1000,
 	.chipset	= ide_rz1000,
 	.host_flags	= IDE_HFLAG_NO_DMA,
diff --git a/drivers/ide/pci/sc1200.c b/drivers/ide/pci/sc1200.c
index 6509560ba665..8efaed16fea3 100644
--- a/drivers/ide/pci/sc1200.c
+++ b/drivers/ide/pci/sc1200.c
@@ -22,6 +22,8 @@
 
 #include <asm/io.h>
 
+#define DRV_NAME "sc1200"
+
 #define SC1200_REV_A	0x00
 #define SC1200_REV_B1	0x01
 #define SC1200_REV_B3	0x02
@@ -291,7 +293,7 @@ static const struct ide_dma_ops sc1200_dma_ops = {
 };
 
 static const struct ide_port_info sc1200_chipset __devinitdata = {
-	.name		= "SC1200",
+	.name		= DRV_NAME,
 	.port_ops	= &sc1200_port_ops,
 	.dma_ops	= &sc1200_dma_ops,
 	.host_flags	= IDE_HFLAG_SERIALIZE |
diff --git a/drivers/ide/pci/serverworks.c b/drivers/ide/pci/serverworks.c
index 1106ff44cdee..e26bc8326dbb 100644
--- a/drivers/ide/pci/serverworks.c
+++ b/drivers/ide/pci/serverworks.c
@@ -38,6 +38,8 @@
 
 #include <asm/io.h>
 
+#define DRV_NAME "serverworks"
+
 #define SVWKS_CSB5_REVISION_NEW	0x92 /* min PCI_REVISION_ID for UDMA5 (A2.0) */
 #define SVWKS_CSB6_REVISION	0xa0 /* min PCI_REVISION_ID for UDMA4 (A1.0) */
 
@@ -353,40 +355,44 @@ static const struct ide_port_ops svwks_port_ops = {
 #define IDE_HFLAGS_SVWKS IDE_HFLAG_LEGACY_IRQS
 
 static const struct ide_port_info serverworks_chipsets[] __devinitdata = {
-	{	/* 0 */
-		.name		= "SvrWks OSB4",
+	{	/* 0: OSB4 */
+		.name		= DRV_NAME,
 		.init_chipset	= init_chipset_svwks,
 		.port_ops	= &osb4_port_ops,
 		.host_flags	= IDE_HFLAGS_SVWKS,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 		.udma_mask	= 0x00, /* UDMA is problematic on OSB4 */
-	},{	/* 1 */
-		.name		= "SvrWks CSB5",
+	},
+	{	/* 1: CSB5 */
+		.name		= DRV_NAME,
 		.init_chipset	= init_chipset_svwks,
 		.port_ops	= &svwks_port_ops,
 		.host_flags	= IDE_HFLAGS_SVWKS,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 		.udma_mask	= ATA_UDMA5,
-	},{	/* 2 */
-		.name		= "SvrWks CSB6",
+	},
+	{	/* 2: CSB6 */
+		.name		= DRV_NAME,
 		.init_chipset	= init_chipset_svwks,
 		.port_ops	= &svwks_port_ops,
 		.host_flags	= IDE_HFLAGS_SVWKS,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 		.udma_mask	= ATA_UDMA5,
-	},{	/* 3 */
-		.name		= "SvrWks CSB6",
+	},
+	{	/* 3: CSB6-2 */
+		.name		= DRV_NAME,
 		.init_chipset	= init_chipset_svwks,
 		.port_ops	= &svwks_port_ops,
 		.host_flags	= IDE_HFLAGS_SVWKS | IDE_HFLAG_SINGLE,
 		.pio_mask	= ATA_PIO4,
 		.mwdma_mask	= ATA_MWDMA2,
 		.udma_mask	= ATA_UDMA5,
-	},{	/* 4 */
-		.name		= "SvrWks HT1000",
+	},
+	{	/* 4: HT1000 */
+		.name		= DRV_NAME,
 		.init_chipset	= init_chipset_svwks,
 		.port_ops	= &svwks_port_ops,
 		.host_flags	= IDE_HFLAGS_SVWKS | IDE_HFLAG_SINGLE,
diff --git a/drivers/ide/pci/siimage.c b/drivers/ide/pci/siimage.c
index fc29f1ca503a..572b479a3922 100644
--- a/drivers/ide/pci/siimage.c
+++ b/drivers/ide/pci/siimage.c
@@ -44,6 +44,8 @@
 #include <linux/init.h>
 #include <linux/io.h>
 
+#define DRV_NAME "siimage"
+
 /**
  *	pdev_is_sata		-	check if device is SATA
  *	@pdev:	PCI device to check
@@ -717,9 +719,9 @@ static const struct ide_dma_ops sil_dma_ops = {
 	.dma_lost_irq		= ide_dma_lost_irq,
 };
 
-#define DECLARE_SII_DEV(name_str, p_ops)		\
+#define DECLARE_SII_DEV(p_ops)				\
 	{						\
-		.name		= name_str,		\
+		.name		= DRV_NAME,		\
 		.init_chipset	= init_chipset_siimage,	\
 		.init_iops	= init_iops_siimage,	\
 		.port_ops	= p_ops,		\
@@ -730,9 +732,8 @@ static const struct ide_dma_ops sil_dma_ops = {
 	}
 
 static const struct ide_port_info siimage_chipsets[] __devinitdata = {
-	/* 0 */ DECLARE_SII_DEV("SiI680",		&sil_pata_port_ops),
-	/* 1 */ DECLARE_SII_DEV("SiI3112 Serial ATA",	&sil_sata_port_ops),
-	/* 2 */ DECLARE_SII_DEV("Adaptec AAR-1210SA",	&sil_sata_port_ops)
+	/* 0: SiI680 */  DECLARE_SII_DEV(&sil_pata_port_ops),
+	/* 1: SiI3112 */ DECLARE_SII_DEV(&sil_sata_port_ops)
 };
 
 /**
@@ -761,7 +762,7 @@ static int __devinit siimage_init_one(struct pci_dev *dev,
 		static int first = 1;
 
 		if (first) {
-			printk(KERN_INFO "siimage: For full SATA support you "
+			printk(KERN_INFO DRV_NAME ": For full SATA support you "
 				"should use the libata sata_sil module.\n");
 			first = 0;
 		}
@@ -780,7 +781,7 @@ static int __devinit siimage_init_one(struct pci_dev *dev,
 		* seem to get terminally confused in the PCI spaces.
 		*/
 		if (!request_mem_region(bar5, barsize, d.name)) {
-			printk(KERN_WARNING "siimage %s: MMIO ports not "
+			printk(KERN_WARNING DRV_NAME " %s: MMIO ports not "
 				"available\n", pci_name(dev));
 		} else {
 			ioaddr = ioremap(bar5, barsize);
@@ -823,7 +824,7 @@ static const struct pci_device_id siimage_pci_tbl[] = {
 	{ PCI_VDEVICE(CMD, PCI_DEVICE_ID_SII_680),    0 },
 #ifdef CONFIG_BLK_DEV_IDE_SATA
 	{ PCI_VDEVICE(CMD, PCI_DEVICE_ID_SII_3112),   1 },
-	{ PCI_VDEVICE(CMD, PCI_DEVICE_ID_SII_1210SA), 2 },
+	{ PCI_VDEVICE(CMD, PCI_DEVICE_ID_SII_1210SA), 1 },
 #endif
 	{ 0, },
 };
diff --git a/drivers/ide/pci/sis5513.c b/drivers/ide/pci/sis5513.c
index 518d8ab413f3..6fcb46c87871 100644
--- a/drivers/ide/pci/sis5513.c
+++ b/drivers/ide/pci/sis5513.c
@@ -52,6 +52,8 @@
 #include <linux/init.h>
 #include <linux/ide.h>
 
+#define DRV_NAME "sis5513"
+
 /* registers layout and init values are chipset family dependant */
 
 #define ATA_16		0x01
@@ -380,7 +382,7 @@ static int __devinit sis_find_family(struct pci_dev *dev)
 		}
 		pci_dev_put(host);
 
-		printk(KERN_INFO "SIS5513 %s: %s %s controller\n",
+		printk(KERN_INFO DRV_NAME " %s: %s %s controller\n",
 			pci_name(dev), SiSHostChipInfo[i].name,
 			chipset_capability[chipset_family]);
 	}
@@ -397,7 +399,7 @@ static int __devinit sis_find_family(struct pci_dev *dev)
 			pci_write_config_dword(dev, 0x54, idemisc);
 
 			if (trueid == 0x5518) {
-				printk(KERN_INFO "SIS5513 %s: SiS 962/963 MuTIOL IDE UDMA133 controller\n",
+				printk(KERN_INFO DRV_NAME " %s: SiS 962/963 MuTIOL IDE UDMA133 controller\n",
 					pci_name(dev));
 				chipset_family = ATA_133;
 
@@ -407,7 +409,7 @@ static int __devinit sis_find_family(struct pci_dev *dev)
 				 */
 				if ((idemisc & 0x40000000) == 0) {
 					pci_write_config_dword(dev, 0x54, idemisc | 0x40000000);
-					printk(KERN_INFO "SIS5513 %s: Switching to 5513 register mapping\n",
+					printk(KERN_INFO DRV_NAME " %s: Switching to 5513 register mapping\n",
 						pci_name(dev));
 				}
 			}
@@ -432,11 +434,11 @@ static int __devinit sis_find_family(struct pci_dev *dev)
 				pci_dev_put(lpc_bridge);
 
 				if (lpc_bridge->revision == 0x10 && (prefctl & 0x80)) {
-					printk(KERN_INFO "SIS5513 %s: SiS 961B MuTIOL IDE UDMA133 controller\n",
+					printk(KERN_INFO DRV_NAME " %s: SiS 961B MuTIOL IDE UDMA133 controller\n",
 						pci_name(dev));
 					chipset_family = ATA_133a;
 				} else {
-					printk(KERN_INFO "SIS5513 %s: SiS 961 MuTIOL IDE UDMA100 controller\n",
+					printk(KERN_INFO DRV_NAME " %s: SiS 961 MuTIOL IDE UDMA100 controller\n",
 						pci_name(dev));
 					chipset_family = ATA_100;
 				}
@@ -560,7 +562,7 @@ static const struct ide_port_ops sis_ata133_port_ops = {
 };
 
 static const struct ide_port_info sis5513_chipset __devinitdata = {
-	.name		= "SIS5513",
+	.name		= DRV_NAME,
 	.init_chipset	= init_chipset_sis5513,
 	.enablebits	= { {0x4a, 0x02, 0x02}, {0x4a, 0x04, 0x04} },
 	.host_flags	= IDE_HFLAG_LEGACY_IRQS | IDE_HFLAG_NO_AUTODMA,
diff --git a/drivers/ide/pci/sl82c105.c b/drivers/ide/pci/sl82c105.c
index 72899f85c5ba..fa720db3de10 100644
--- a/drivers/ide/pci/sl82c105.c
+++ b/drivers/ide/pci/sl82c105.c
@@ -23,6 +23,8 @@
 
 #include <asm/io.h>
 
+#define DRV_NAME "sl82c105"
+
 #undef DEBUG
 
 #ifdef DEBUG
@@ -301,7 +303,7 @@ static const struct ide_dma_ops sl82c105_dma_ops = {
 };
 
 static const struct ide_port_info sl82c105_chipset __devinitdata = {
-	.name		= "W82C105",
+	.name		= DRV_NAME,
 	.init_chipset	= init_chipset_sl82c105,
 	.enablebits	= {{0x40,0x01,0x01}, {0x40,0x10,0x10}},
 	.port_ops	= &sl82c105_port_ops,
@@ -328,7 +330,7 @@ static int __devinit sl82c105_init_one(struct pci_dev *dev, const struct pci_dev
 		 * Never ever EVER under any circumstances enable
 		 * DMA when the bridge is this old.
 		 */
-		printk(KERN_INFO "W82C105_IDE: Winbond W83C553 bridge "
+		printk(KERN_INFO DRV_NAME ": Winbond W83C553 bridge "
 				 "revision %d, BM-DMA disabled\n", rev);
 		d.dma_ops = NULL;
 		d.mwdma_mask = 0;
diff --git a/drivers/ide/pci/slc90e66.c b/drivers/ide/pci/slc90e66.c
index fee5ebe4bb99..13d1fa491f26 100644
--- a/drivers/ide/pci/slc90e66.c
+++ b/drivers/ide/pci/slc90e66.c
@@ -15,6 +15,8 @@
 #include <linux/ide.h>
 #include <linux/init.h>
 
+#define DRV_NAME "slc90e66"
+
 static DEFINE_SPINLOCK(slc90e66_lock);
 
 static void slc90e66_set_pio_mode(ide_drive_t *drive, const u8 pio)
@@ -132,7 +134,7 @@ static const struct ide_port_ops slc90e66_port_ops = {
 };
 
 static const struct ide_port_info slc90e66_chipset __devinitdata = {
-	.name		= "SLC90E66",
+	.name		= DRV_NAME,
 	.enablebits	= { {0x41, 0x80, 0x80}, {0x43, 0x80, 0x80} },
 	.port_ops	= &slc90e66_port_ops,
 	.host_flags	= IDE_HFLAG_LEGACY_IRQS,
diff --git a/drivers/ide/pci/tc86c001.c b/drivers/ide/pci/tc86c001.c
index 102cd7c40cdd..b1cb8a9ce5a9 100644
--- a/drivers/ide/pci/tc86c001.c
+++ b/drivers/ide/pci/tc86c001.c
@@ -11,7 +11,7 @@
 #include <linux/pci.h>
 #include <linux/ide.h>
 
-#define DRV_NAME "TC86C001"
+#define DRV_NAME "tc86c001"
 
 static void tc86c001_set_mode(ide_drive_t *drive, const u8 speed)
 {
@@ -193,7 +193,7 @@ static const struct ide_dma_ops tc86c001_dma_ops = {
 };
 
 static const struct ide_port_info tc86c001_chipset __devinitdata = {
-	.name		= "TC86C001",
+	.name		= DRV_NAME,
 	.init_hwif	= init_hwif_tc86c001,
 	.port_ops	= &tc86c001_port_ops,
 	.dma_ops	= &tc86c001_dma_ops,
diff --git a/drivers/ide/pci/triflex.c b/drivers/ide/pci/triflex.c
index 78e24ac80972..b77ec35151b3 100644
--- a/drivers/ide/pci/triflex.c
+++ b/drivers/ide/pci/triflex.c
@@ -33,6 +33,8 @@
 #include <linux/ide.h>
 #include <linux/init.h>
 
+#define DRV_NAME "triflex"
+
 static void triflex_set_mode(ide_drive_t *drive, const u8 speed)
 {
 	ide_hwif_t *hwif = HWIF(drive);
@@ -93,7 +95,7 @@ static const struct ide_port_ops triflex_port_ops = {
 };
 
 static const struct ide_port_info triflex_device __devinitdata = {
-	.name		= "TRIFLEX",
+	.name		= DRV_NAME,
 	.enablebits	= {{0x80, 0x01, 0x01}, {0x80, 0x02, 0x02}},
 	.port_ops	= &triflex_port_ops,
 	.pio_mask	= ATA_PIO4,
diff --git a/drivers/ide/pci/trm290.c b/drivers/ide/pci/trm290.c
index 7bda5ed92e1e..fd28b49977fd 100644
--- a/drivers/ide/pci/trm290.c
+++ b/drivers/ide/pci/trm290.c
@@ -141,6 +141,8 @@
 
 #include <asm/io.h>
 
+#define DRV_NAME "trm290"
+
 static void trm290_prepare_drive (ide_drive_t *drive, unsigned int use_dma)
 {
 	ide_hwif_t *hwif = HWIF(drive);
@@ -245,10 +247,10 @@ static void __devinit init_hwif_trm290(ide_hwif_t *hwif)
 	u8 reg = 0;
 
 	if ((dev->class & 5) && cfg_base)
-		printk(KERN_INFO "TRM290 %s: chip", pci_name(dev));
+		printk(KERN_INFO DRV_NAME " %s: chip", pci_name(dev));
 	else {
 		cfg_base = 0x3df0;
-		printk(KERN_INFO "TRM290 %s: using default", pci_name(dev));
+		printk(KERN_INFO DRV_NAME " %s: using default", pci_name(dev));
 	}
 	printk(KERN_CONT " config base at 0x%04x\n", cfg_base);
 	hwif->config_data = cfg_base;
@@ -325,7 +327,7 @@ static struct ide_dma_ops trm290_dma_ops = {
 };
 
 static const struct ide_port_info trm290_chipset __devinitdata = {
-	.name		= "TRM290",
+	.name		= DRV_NAME,
 	.init_hwif	= init_hwif_trm290,
 	.chipset	= ide_trm290,
 	.port_ops	= &trm290_port_ops,
diff --git a/drivers/ide/pci/via82cxxx.c b/drivers/ide/pci/via82cxxx.c
index 23332556e616..170e058f1fbd 100644
--- a/drivers/ide/pci/via82cxxx.c
+++ b/drivers/ide/pci/via82cxxx.c
@@ -35,6 +35,8 @@
 #include <asm/processor.h>
 #endif
 
+#define DRV_NAME "via82cxxx"
+
 #define VIA_IDE_ENABLE		0x40
 #define VIA_IDE_CONFIG		0x41
 #define VIA_FIFO_CONFIG		0x43
@@ -373,7 +375,7 @@ static const struct ide_port_ops via_port_ops = {
 };
 
 static const struct ide_port_info via82cxxx_chipset __devinitdata = {
-	.name		= "VP_IDE",
+	.name		= DRV_NAME,
 	.init_chipset	= init_chipset_via82cxxx,
 	.enablebits	= { { 0x40, 0x02, 0x02 }, { 0x40, 0x01, 0x01 } },
 	.port_ops	= &via_port_ops,
@@ -401,7 +403,7 @@ static int __devinit via_init_one(struct pci_dev *dev, const struct pci_device_i
 	 */
 	via_config = via_config_find(&isa);
 	if (!via_config->id) {
-		printk(KERN_WARNING "VP_IDE %s: unknown chipset, skipping\n",
+		printk(KERN_WARNING DRV_NAME " %s: unknown chipset, skipping\n",
 			pci_name(dev));
 		return -ENODEV;
 	}
@@ -409,7 +411,7 @@ static int __devinit via_init_one(struct pci_dev *dev, const struct pci_device_i
 	/*
 	 * Print the boot message.
 	 */
-	printk(KERN_INFO "VP_IDE %s: VIA %s (rev %02x) IDE %sDMA%s\n",
+	printk(KERN_INFO DRV_NAME " %s: VIA %s (rev %02x) IDE %sDMA%s\n",
 		pci_name(dev), via_config->name, isa->revision,
 		via_config->udma_mask ? "U" : "MW",
 		via_dma[via_config->udma_mask ?
@@ -429,9 +431,9 @@ static int __devinit via_init_one(struct pci_dev *dev, const struct pci_device_i
 	}
 
 	if (via_clock < 20000 || via_clock > 50000) {
-		printk(KERN_WARNING "VP_IDE: User given PCI clock speed "
+		printk(KERN_WARNING DRV_NAME ": User given PCI clock speed "
 			"impossible (%d), using 33 MHz instead.\n", via_clock);
-		printk(KERN_WARNING "VP_IDE: Use ide0=ata66 if you want "
+		printk(KERN_WARNING DRV_NAME ": Use ide0=ata66 if you want "
 			"to assume 80-wire cable.\n");
 		via_clock = 33333;
 	}
@@ -453,7 +455,8 @@ static int __devinit via_init_one(struct pci_dev *dev, const struct pci_device_i
 
 	vdev = kzalloc(sizeof(*vdev), GFP_KERNEL);
 	if (!vdev) {
-		printk(KERN_ERR "VP_IDE %s: out of memory :(\n", pci_name(dev));
+		printk(KERN_ERR DRV_NAME " %s: out of memory :(\n",
+			pci_name(dev));
 		return -ENOMEM;
 	}
 

From a326b02b0c576001353dbc489154959b0889c6bf Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:33 +0200
Subject: [PATCH 0589/1171] ide: drop 'name' parameter from ->init_chipset
 method

There should be no functional changes caused by this patch.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/pci/aec62xx.c      |  2 +-
 drivers/ide/pci/alim15x3.c     |  5 ++---
 drivers/ide/pci/amd74xx.c      | 19 ++++++++-----------
 drivers/ide/pci/cmd64x.c       |  2 +-
 drivers/ide/pci/cs5530.c       |  7 +++----
 drivers/ide/pci/cy82c693.c     | 10 +++++-----
 drivers/ide/pci/hpt34x.c       |  2 +-
 drivers/ide/pci/hpt366.c       |  3 ++-
 drivers/ide/pci/it821x.c       |  2 +-
 drivers/ide/pci/pdc202xx_new.c |  3 ++-
 drivers/ide/pci/pdc202xx_old.c |  3 +--
 drivers/ide/pci/piix.c         |  3 +--
 drivers/ide/pci/serverworks.c  |  6 +++---
 drivers/ide/pci/siimage.c      |  8 +++-----
 drivers/ide/pci/sis5513.c      |  3 +--
 drivers/ide/pci/sl82c105.c     |  2 +-
 drivers/ide/pci/via82cxxx.c    |  3 +--
 drivers/ide/setup-pci.c        |  2 +-
 include/linux/ide.h            |  2 +-
 19 files changed, 39 insertions(+), 48 deletions(-)

diff --git a/drivers/ide/pci/aec62xx.c b/drivers/ide/pci/aec62xx.c
index f6dc6c20f3af..e0c8fe7d9fea 100644
--- a/drivers/ide/pci/aec62xx.c
+++ b/drivers/ide/pci/aec62xx.c
@@ -140,7 +140,7 @@ static void aec_set_pio_mode(ide_drive_t *drive, const u8 pio)
 	drive->hwif->port_ops->set_dma_mode(drive, pio + XFER_PIO_0);
 }
 
-static unsigned int __devinit init_chipset_aec62xx(struct pci_dev *dev, const char *name)
+static unsigned int __devinit init_chipset_aec62xx(struct pci_dev *dev)
 {
 	/* These are necessary to get AEC6280 Macintosh cards to work */
 	if ((dev->device == PCI_DEVICE_ID_ARTOP_ATP865) ||
diff --git a/drivers/ide/pci/alim15x3.c b/drivers/ide/pci/alim15x3.c
index a099c4dd599d..b582687e0cd4 100644
--- a/drivers/ide/pci/alim15x3.c
+++ b/drivers/ide/pci/alim15x3.c
@@ -209,13 +209,12 @@ static int ali15x3_dma_setup(ide_drive_t *drive)
 /**
  *	init_chipset_ali15x3	-	Initialise an ALi IDE controller
  *	@dev: PCI device
- *	@name: Name of the controller
  *
  *	This function initializes the ALI IDE controller and where 
  *	appropriate also sets up the 1533 southbridge.
  */
-  
-static unsigned int __devinit init_chipset_ali15x3 (struct pci_dev *dev, const char *name)
+
+static unsigned int __devinit init_chipset_ali15x3(struct pci_dev *dev)
 {
 	unsigned long flags;
 	u8 tmpbyte;
diff --git a/drivers/ide/pci/amd74xx.c b/drivers/ide/pci/amd74xx.c
index cbf78edfe00b..2cea7bf51a0f 100644
--- a/drivers/ide/pci/amd74xx.c
+++ b/drivers/ide/pci/amd74xx.c
@@ -112,15 +112,13 @@ static void amd_set_pio_mode(ide_drive_t *drive, const u8 pio)
 	amd_set_drive(drive, XFER_PIO_0 + pio);
 }
 
-static void __devinit amd7409_cable_detect(struct pci_dev *dev,
-					   const char *name)
+static void __devinit amd7409_cable_detect(struct pci_dev *dev)
 {
 	/* no host side cable detection */
 	amd_80w = 0x03;
 }
 
-static void __devinit amd7411_cable_detect(struct pci_dev *dev,
-					   const char *name)
+static void __devinit amd7411_cable_detect(struct pci_dev *dev)
 {
 	int i;
 	u32 u = 0;
@@ -131,9 +129,9 @@ static void __devinit amd7411_cable_detect(struct pci_dev *dev,
 	amd_80w = ((t & 0x3) ? 1 : 0) | ((t & 0xc) ? 2 : 0);
 	for (i = 24; i >= 0; i -= 8)
 		if (((u >> i) & 4) && !(amd_80w & (1 << (1 - (i >> 4))))) {
-			printk(KERN_WARNING "%s %s: BIOS didn't set cable bits "
-				"correctly. Enabling workaround.\n",
-				name, pci_name(dev));
+			printk(KERN_WARNING DRV_NAME " %s: BIOS didn't set "
+				"cable bits correctly. Enabling workaround.\n",
+				pci_name(dev));
 			amd_80w |= (1 << (1 - (i >> 4)));
 		}
 }
@@ -142,8 +140,7 @@ static void __devinit amd7411_cable_detect(struct pci_dev *dev,
  * The initialization callback.  Initialize drive independent registers.
  */
 
-static unsigned int __devinit init_chipset_amd74xx(struct pci_dev *dev,
-						   const char *name)
+static unsigned int __devinit init_chipset_amd74xx(struct pci_dev *dev)
 {
 	u8 t = 0, offset = amd_offset(dev);
 
@@ -156,9 +153,9 @@ static unsigned int __devinit init_chipset_amd74xx(struct pci_dev *dev,
 		; /* no UDMA > 2 */
 	else if (dev->vendor == PCI_VENDOR_ID_AMD &&
 		 dev->device == PCI_DEVICE_ID_AMD_VIPER_7409)
-		amd7409_cable_detect(dev, name);
+		amd7409_cable_detect(dev);
 	else
-		amd7411_cable_detect(dev, name);
+		amd7411_cable_detect(dev);
 
 /*
  * Take care of prefetch & postwrite.
diff --git a/drivers/ide/pci/cmd64x.c b/drivers/ide/pci/cmd64x.c
index 3d84debaf81f..1360b4fa9fd3 100644
--- a/drivers/ide/pci/cmd64x.c
+++ b/drivers/ide/pci/cmd64x.c
@@ -332,7 +332,7 @@ static int cmd646_1_dma_end(ide_drive_t *drive)
 	return (dma_stat & 7) != 4;
 }
 
-static unsigned int __devinit init_chipset_cmd64x(struct pci_dev *dev, const char *name)
+static unsigned int __devinit init_chipset_cmd64x(struct pci_dev *dev)
 {
 	u8 mrdmode = 0;
 
diff --git a/drivers/ide/pci/cs5530.c b/drivers/ide/pci/cs5530.c
index 5543c8677a5a..f235db8c678b 100644
--- a/drivers/ide/pci/cs5530.c
+++ b/drivers/ide/pci/cs5530.c
@@ -129,12 +129,11 @@ static void cs5530_set_dma_mode(ide_drive_t *drive, const u8 mode)
 /**
  *	init_chipset_5530	-	set up 5530 bridge
  *	@dev: PCI device
- *	@name: device name
  *
  *	Initialize the cs5530 bridge for reliable IDE DMA operation.
  */
 
-static unsigned int __devinit init_chipset_cs5530 (struct pci_dev *dev, const char *name)
+static unsigned int __devinit init_chipset_cs5530(struct pci_dev *dev)
 {
 	struct pci_dev *master_0 = NULL, *cs5530_0 = NULL;
 
@@ -153,11 +152,11 @@ static unsigned int __devinit init_chipset_cs5530 (struct pci_dev *dev, const ch
 		}
 	}
 	if (!master_0) {
-		printk(KERN_ERR "%s: unable to locate PCI MASTER function\n", name);
+		printk(KERN_ERR DRV_NAME ": unable to locate PCI MASTER function\n");
 		goto out;
 	}
 	if (!cs5530_0) {
-		printk(KERN_ERR "%s: unable to locate CS5530 LEGACY function\n", name);
+		printk(KERN_ERR DRV_NAME ": unable to locate CS5530 LEGACY function\n");
 		goto out;
 	}
 
diff --git a/drivers/ide/pci/cy82c693.c b/drivers/ide/pci/cy82c693.c
index 41c7f3351eb6..bfae2f882f48 100644
--- a/drivers/ide/pci/cy82c693.c
+++ b/drivers/ide/pci/cy82c693.c
@@ -332,7 +332,7 @@ static void cy82c693_set_pio_mode(ide_drive_t *drive, const u8 pio)
 /*
  * this function is called during init and is used to setup the cy82c693 chip
  */
-static unsigned int __devinit init_chipset_cy82c693(struct pci_dev *dev, const char *name)
+static unsigned int __devinit init_chipset_cy82c693(struct pci_dev *dev)
 {
 	if (PCI_FUNC(dev->devfn) != 1)
 		return 0;
@@ -351,8 +351,8 @@ static unsigned int __devinit init_chipset_cy82c693(struct pci_dev *dev, const c
 	data = inb(CY82_DATA_PORT);
 
 #if CY82C693_DEBUG_INFO
-	printk(KERN_INFO "%s: Peripheral Configuration Register: 0x%X\n",
-		name, data);
+	printk(KERN_INFO DRV_NAME ": Peripheral Configuration Register: 0x%X\n",
+		data);
 #endif /* CY82C693_DEBUG_INFO */
 
 	/*
@@ -373,8 +373,8 @@ static unsigned int __devinit init_chipset_cy82c693(struct pci_dev *dev, const c
 	outb(data, CY82_DATA_PORT);
 
 #if CY82C693_DEBUG_INFO
-	printk(KERN_INFO "%s: New Peripheral Configuration Register: 0x%X\n",
-		name, data);
+	printk(KERN_INFO ": New Peripheral Configuration Register: 0x%X\n",
+		data);
 #endif /* CY82C693_DEBUG_INFO */
 
 #endif /* CY82C693_SETDMA_CLOCK */
diff --git a/drivers/ide/pci/hpt34x.c b/drivers/ide/pci/hpt34x.c
index baabb4ce0d78..6009b0b9655d 100644
--- a/drivers/ide/pci/hpt34x.c
+++ b/drivers/ide/pci/hpt34x.c
@@ -79,7 +79,7 @@ static void hpt34x_set_pio_mode(ide_drive_t *drive, const u8 pio)
  */
 #define	HPT34X_PCI_INIT_REG		0x80
 
-static unsigned int __devinit init_chipset_hpt34x(struct pci_dev *dev, const char *name)
+static unsigned int __devinit init_chipset_hpt34x(struct pci_dev *dev)
 {
 	int i = 0;
 	unsigned long hpt34xIoBase = pci_resource_start(dev, 4);
diff --git a/drivers/ide/pci/hpt366.c b/drivers/ide/pci/hpt366.c
index 6a1c65c3be3e..5271b246b88c 100644
--- a/drivers/ide/pci/hpt366.c
+++ b/drivers/ide/pci/hpt366.c
@@ -970,11 +970,12 @@ static int __devinit hpt37x_calibrate_dpll(struct pci_dev *dev, u16 f_low, u16 f
 	return 1;
 }
 
-static unsigned int __devinit init_chipset_hpt366(struct pci_dev *dev, const char *name)
+static unsigned int __devinit init_chipset_hpt366(struct pci_dev *dev)
 {
 	unsigned long io_base	= pci_resource_start(dev, 4);
 	struct ide_host *host	= pci_get_drvdata(dev);
 	struct hpt_info *info	= host->host_priv + (&dev->dev == host->dev[1]);
+	const char *name	= DRV_NAME;
 	u8 pci_clk,  dpll_clk	= 0;	/* PCI and DPLL clock in MHz */
 	u8 chip_type;
 	enum ata_clock	clock;
diff --git a/drivers/ide/pci/it821x.c b/drivers/ide/pci/it821x.c
index 74173352741f..e16a1d113a2a 100644
--- a/drivers/ide/pci/it821x.c
+++ b/drivers/ide/pci/it821x.c
@@ -605,7 +605,7 @@ static void __devinit it8212_disable_raid(struct pci_dev *dev)
 	pci_write_config_byte(dev, PCI_LATENCY_TIMER, 0x20);
 }
 
-static unsigned int __devinit init_chipset_it821x(struct pci_dev *dev, const char *name)
+static unsigned int __devinit init_chipset_it821x(struct pci_dev *dev)
 {
 	u8 conf;
 	static char *mode[2] = { "pass through", "smart" };
diff --git a/drivers/ide/pci/pdc202xx_new.c b/drivers/ide/pci/pdc202xx_new.c
index 1f6791957227..998615fa285f 100644
--- a/drivers/ide/pci/pdc202xx_new.c
+++ b/drivers/ide/pci/pdc202xx_new.c
@@ -326,8 +326,9 @@ static void __devinit apple_kiwi_init(struct pci_dev *pdev)
 }
 #endif /* CONFIG_PPC_PMAC */
 
-static unsigned int __devinit init_chipset_pdcnew(struct pci_dev *dev, const char *name)
+static unsigned int __devinit init_chipset_pdcnew(struct pci_dev *dev)
 {
+	const char *name = DRV_NAME;
 	unsigned long dma_base = pci_resource_start(dev, 4);
 	unsigned long sec_dma_base = dma_base + 0x08;
 	long pll_input, pll_output, ratio;
diff --git a/drivers/ide/pci/pdc202xx_old.c b/drivers/ide/pci/pdc202xx_old.c
index da92d127868f..6ff2def58da0 100644
--- a/drivers/ide/pci/pdc202xx_old.c
+++ b/drivers/ide/pci/pdc202xx_old.c
@@ -265,8 +265,7 @@ static void pdc202xx_dma_timeout(ide_drive_t *drive)
 	ide_dma_timeout(drive);
 }
 
-static unsigned int __devinit init_chipset_pdc202xx(struct pci_dev *dev,
-						    const char *name)
+static unsigned int __devinit init_chipset_pdc202xx(struct pci_dev *dev)
 {
 	unsigned long dmabase = pci_resource_start(dev, 4);
 	u8 udma_speed_flag = 0, primary_mode = 0, secondary_mode = 0;
diff --git a/drivers/ide/pci/piix.c b/drivers/ide/pci/piix.c
index 9eb411f5c358..7fc3022dcf68 100644
--- a/drivers/ide/pci/piix.c
+++ b/drivers/ide/pci/piix.c
@@ -200,13 +200,12 @@ static void piix_set_dma_mode(ide_drive_t *drive, const u8 speed)
 /**
  *	init_chipset_ich	-	set up the ICH chipset
  *	@dev: PCI device to set up
- *	@name: Name of the device
  *
  *	Initialize the PCI device as required.  For the ICH this turns
  *	out to be nice and simple.
  */
 
-static unsigned int __devinit init_chipset_ich(struct pci_dev *dev, const char *name)
+static unsigned int __devinit init_chipset_ich(struct pci_dev *dev)
 {
 	u32 extra = 0;
 
diff --git a/drivers/ide/pci/serverworks.c b/drivers/ide/pci/serverworks.c
index e26bc8326dbb..d173f2937722 100644
--- a/drivers/ide/pci/serverworks.c
+++ b/drivers/ide/pci/serverworks.c
@@ -174,7 +174,7 @@ static void svwks_set_dma_mode(ide_drive_t *drive, const u8 speed)
 	pci_write_config_byte(dev, 0x54, ultra_enable);
 }
 
-static unsigned int __devinit init_chipset_svwks (struct pci_dev *dev, const char *name)
+static unsigned int __devinit init_chipset_svwks(struct pci_dev *dev)
 {
 	unsigned int reg;
 	u8 btr;
@@ -190,8 +190,8 @@ static unsigned int __devinit init_chipset_svwks (struct pci_dev *dev, const cha
 			pci_read_config_dword(isa_dev, 0x64, &reg);
 			reg &= ~0x00002000; /* disable 600ns interrupt mask */
 			if(!(reg & 0x00004000))
-				printk(KERN_DEBUG "%s %s: UDMA not BIOS "
-					"enabled.\n", name, pci_name(dev));
+				printk(KERN_DEBUG DRV_NAME " %s: UDMA not BIOS "
+					"enabled.\n", pci_name(dev));
 			reg |=  0x00004000; /* enable UDMA/33 support */
 			pci_write_config_dword(isa_dev, 0x64, reg);
 		}
diff --git a/drivers/ide/pci/siimage.c b/drivers/ide/pci/siimage.c
index 572b479a3922..b8ad9ad6cf0d 100644
--- a/drivers/ide/pci/siimage.c
+++ b/drivers/ide/pci/siimage.c
@@ -457,14 +457,12 @@ static void sil_sata_pre_reset(ide_drive_t *drive)
 /**
  *	init_chipset_siimage	-	set up an SI device
  *	@dev: PCI device
- *	@name: device name
  *
  *	Perform the initial PCI set up for this device. Attempt to switch
  *	to 133 MHz clocking if the system isn't already set up to do it.
  */
 
-static unsigned int __devinit init_chipset_siimage(struct pci_dev *dev,
-						   const char *name)
+static unsigned int __devinit init_chipset_siimage(struct pci_dev *dev)
 {
 	struct ide_host *host = pci_get_drvdata(dev);
 	void __iomem *ioaddr = host->host_priv;
@@ -541,8 +539,8 @@ static unsigned int __devinit init_chipset_siimage(struct pci_dev *dev,
 			{ "== 100", "== 133", "== 2X PCI", "DISABLED!" };
 
 		tmp >>= 4;
-		printk(KERN_INFO "%s %s: BASE CLOCK %s\n",
-			name, pci_name(dev), clk_str[tmp & 3]);
+		printk(KERN_INFO DRV_NAME " %s: BASE CLOCK %s\n",
+			pci_name(dev), clk_str[tmp & 3]);
 	}
 
 	return 0;
diff --git a/drivers/ide/pci/sis5513.c b/drivers/ide/pci/sis5513.c
index 6fcb46c87871..cc95f90b53b7 100644
--- a/drivers/ide/pci/sis5513.c
+++ b/drivers/ide/pci/sis5513.c
@@ -448,8 +448,7 @@ static int __devinit sis_find_family(struct pci_dev *dev)
 	return chipset_family;
 }
 
-static unsigned int __devinit init_chipset_sis5513(struct pci_dev *dev,
-						   const char *name)
+static unsigned int __devinit init_chipset_sis5513(struct pci_dev *dev)
 {
 	/* Make general config ops here
 	   1/ tell IDE channels to operate in Compatibility mode only
diff --git a/drivers/ide/pci/sl82c105.c b/drivers/ide/pci/sl82c105.c
index fa720db3de10..73905bcc08fb 100644
--- a/drivers/ide/pci/sl82c105.c
+++ b/drivers/ide/pci/sl82c105.c
@@ -272,7 +272,7 @@ static u8 sl82c105_bridge_revision(struct pci_dev *dev)
  * channel 0 here at least, but channel 1 has to be enabled by
  * firmware or arch code. We still set both to 16 bits mode.
  */
-static unsigned int __devinit init_chipset_sl82c105(struct pci_dev *dev, const char *msg)
+static unsigned int __devinit init_chipset_sl82c105(struct pci_dev *dev)
 {
 	u32 val;
 
diff --git a/drivers/ide/pci/via82cxxx.c b/drivers/ide/pci/via82cxxx.c
index 170e058f1fbd..454d2bf62dce 100644
--- a/drivers/ide/pci/via82cxxx.c
+++ b/drivers/ide/pci/via82cxxx.c
@@ -262,13 +262,12 @@ static void __devinit via_cable_detect(struct via82cxxx_dev *vdev, u32 u)
 /**
  *	init_chipset_via82cxxx	-	initialization handler
  *	@dev: PCI device
- *	@name: Name of interface
  *
  *	The initialization callback. Here we determine the IDE chip type
  *	and initialize its drive independent registers.
  */
 
-static unsigned int __devinit init_chipset_via82cxxx(struct pci_dev *dev, const char *name)
+static unsigned int __devinit init_chipset_via82cxxx(struct pci_dev *dev)
 {
 	struct ide_host *host = pci_get_drvdata(dev);
 	struct via82cxxx_dev *vdev = host->host_priv;
diff --git a/drivers/ide/setup-pci.c b/drivers/ide/setup-pci.c
index d9655aeb013b..a8e9e8a69a52 100644
--- a/drivers/ide/setup-pci.c
+++ b/drivers/ide/setup-pci.c
@@ -515,7 +515,7 @@ static int do_ide_setup_pci_device(struct pci_dev *dev,
 	 * space, place chipset into init-mode, and/or preserve
 	 * an interrupt if the card is not native ide support.
 	 */
-	ret = d->init_chipset ? d->init_chipset(dev, d->name) : 0;
+	ret = d->init_chipset ? d->init_chipset(dev) : 0;
 	if (ret < 0)
 		goto out;
 
diff --git a/include/linux/ide.h b/include/linux/ide.h
index fd78b401b036..b846bc44a27e 100644
--- a/include/linux/ide.h
+++ b/include/linux/ide.h
@@ -1206,7 +1206,7 @@ enum {
 
 struct ide_port_info {
 	char			*name;
-	unsigned int		(*init_chipset)(struct pci_dev *, const char *);
+	unsigned int		(*init_chipset)(struct pci_dev *);
 	void			(*init_iops)(ide_hwif_t *);
 	void                    (*init_hwif)(ide_hwif_t *);
 	int			(*init_dma)(ide_hwif_t *,

From cd740ab0f69f6c94d9c7f916758e308f30a439fa Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Thu, 24 Jul 2008 22:53:33 +0200
Subject: [PATCH 0590/1171] ide: ide-tape.c sparse annotations and unaligned
 access removal

If this is actually unaligned the access of speed/max_speed above
is already broken and needs a get_unaligned.  Otherwise it is
aligned and they can be removed.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Cc: Borislav Petkov <petkovbb@googlemail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-tape.c | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index 789f3428f072..82c2afe4d28a 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -655,10 +655,10 @@ static void ide_tape_callback(ide_drive_t *drive)
 			uptodate = 0;
 		} else {
 			debug_log(DBG_SENSE, "Block Location - %u\n",
-					be32_to_cpu(*(u32 *)&readpos[4]));
+					be32_to_cpup((__be32 *)&readpos[4]));
 
 			tape->partition = readpos[1];
-			tape->first_frame = be32_to_cpu(*(u32 *)&readpos[4]);
+			tape->first_frame = be32_to_cpup((__be32 *)&readpos[4]);
 			set_bit(IDE_AFLAG_ADDRESS_VALID, &drive->atapi_flags);
 		}
 	}
@@ -2381,23 +2381,23 @@ static void idetape_get_mode_sense_results(ide_drive_t *drive)
 	caps = pc.buf + 4 + pc.buf[3];
 
 	/* convert to host order and save for later use */
-	speed = be16_to_cpu(*(u16 *)&caps[14]);
-	max_speed = be16_to_cpu(*(u16 *)&caps[8]);
+	speed = be16_to_cpup((__be16 *)&caps[14]);
+	max_speed = be16_to_cpup((__be16 *)&caps[8]);
 
-	put_unaligned(max_speed, (u16 *)&caps[8]);
-	put_unaligned(be16_to_cpu(*(u16 *)&caps[12]), (u16 *)&caps[12]);
-	put_unaligned(speed, (u16 *)&caps[14]);
-	put_unaligned(be16_to_cpu(*(u16 *)&caps[16]), (u16 *)&caps[16]);
+	*(u16 *)&caps[8] = max_speed;
+	*(u16 *)&caps[12] = be16_to_cpup((__be16 *)&caps[12]);
+	*(u16 *)&caps[14] = speed;
+	*(u16 *)&caps[16] = be16_to_cpup((__be16 *)&caps[16]);
 
 	if (!speed) {
 		printk(KERN_INFO "ide-tape: %s: invalid tape speed "
 				"(assuming 650KB/sec)\n", drive->name);
-		put_unaligned(650, (u16 *)&caps[14]);
+		*(u16 *)&caps[14] = 650;
 	}
 	if (!max_speed) {
 		printk(KERN_INFO "ide-tape: %s: invalid max_speed "
 				"(assuming 650KB/sec)\n", drive->name);
-		put_unaligned(650, (u16 *)&caps[8]);
+		*(u16 *)&caps[8] = 650;
 	}
 
 	memcpy(&tape->caps, caps, 20);

From 7fa897b91a3ea0f16c2873b869d7a0eef05acff4 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Thu, 24 Jul 2008 22:53:34 +0200
Subject: [PATCH 0591/1171] ide: trivial sparse annotations

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-dma.c  | 2 +-
 drivers/ide/ide-iops.c | 6 ++----
 drivers/ide/ide-proc.c | 4 ++--
 3 files changed, 5 insertions(+), 7 deletions(-)

diff --git a/drivers/ide/ide-dma.c b/drivers/ide/ide-dma.c
index be99d463dcc7..71c377a7bcf2 100644
--- a/drivers/ide/ide-dma.c
+++ b/drivers/ide/ide-dma.c
@@ -173,7 +173,7 @@ EXPORT_SYMBOL_GPL(ide_build_sglist);
 int ide_build_dmatable (ide_drive_t *drive, struct request *rq)
 {
 	ide_hwif_t *hwif	= HWIF(drive);
-	unsigned int *table	= hwif->dmatable_cpu;
+	__le32 *table = (__le32 *)hwif->dmatable_cpu;
 	unsigned int is_trm290	= (hwif->chipset == ide_trm290) ? 1 : 0;
 	unsigned int count = 0;
 	int i;
diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
index 07da5fb9eaff..8aae91764513 100644
--- a/drivers/ide/ide-iops.c
+++ b/drivers/ide/ide-iops.c
@@ -510,10 +510,8 @@ void ide_fixstring (u8 *s, const int bytecount, const int byteswap)
 
 	if (byteswap) {
 		/* convert from big-endian to host byte order */
-		for (p = end ; p != s;) {
-			unsigned short *pp = (unsigned short *) (p -= 2);
-			*pp = ntohs(*pp);
-		}
+		for (p = end ; p != s;)
+			be16_to_cpus((u16 *)(p -= 2));
 	}
 	/* strip leading blanks */
 	while (s != end && *s == ' ')
diff --git a/drivers/ide/ide-proc.c b/drivers/ide/ide-proc.c
index 151c91e933da..f66c9c3f6fc6 100644
--- a/drivers/ide/ide-proc.c
+++ b/drivers/ide/ide-proc.c
@@ -105,7 +105,7 @@ static int proc_ide_read_identify
 	len = sprintf(page, "\n");
 
 	if (drive) {
-		unsigned short *val = (unsigned short *) page;
+		__le16 *val = (__le16 *)page;
 
 		err = taskfile_lib_get_identify(drive, page);
 		if (!err) {
@@ -113,7 +113,7 @@ static int proc_ide_read_identify
 			page = out;
 			do {
 				out += sprintf(out, "%04x%c",
-					le16_to_cpu(*val), (++i & 7) ? ' ' : '\n');
+					le16_to_cpup(val), (++i & 7) ? ' ' : '\n');
 				val += 1;
 			} while (i < (SECTOR_WORDS * 2));
 			len = out - page;

From a0f403bc58dcaa118f02ec70c3ecfec1bc26e445 Mon Sep 17 00:00:00 2001
From: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Date: Thu, 24 Jul 2008 22:53:34 +0200
Subject: [PATCH 0592/1171] palm_bk3710: add UltraDMA/100 support

This controller supports UltraDMA up to mode 5 but it should be clocked with
at least twice the data strobe frequency, so enable mode 5 for 100+ MHz IDECLK.

While at it, start passing the correct device to clk_get() -- it worked anyway
but WTF? :-/

Signed-off-by: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/arm/palm_bk3710.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/ide/arm/palm_bk3710.c b/drivers/ide/arm/palm_bk3710.c
index 65bb4b8fd570..3e842d60eae9 100644
--- a/drivers/ide/arm/palm_bk3710.c
+++ b/drivers/ide/arm/palm_bk3710.c
@@ -82,6 +82,7 @@ static const struct palm_bk3710_udmatiming palm_bk3710_udmatimings[6] = {
 	{100, 120},		/* UDMA Mode 2 */
 	{100, 90},		/* UDMA Mode 3 */
 	{100, 60},		/* UDMA Mode 4 */
+	{85,  40},		/* UDMA Mode 5 */
 };
 
 static void palm_bk3710_setudmamode(void __iomem *base, unsigned int dev,
@@ -334,12 +335,11 @@ static const struct ide_port_ops palm_bk3710_ports_ops = {
 	.cable_detect		= palm_bk3710_cable_detect,
 };
 
-static const struct ide_port_info __devinitdata palm_bk3710_port_info = {
+static struct ide_port_info __devinitdata palm_bk3710_port_info = {
 	.init_dma		= palm_bk3710_init_dma,
 	.port_ops		= &palm_bk3710_ports_ops,
 	.host_flags		= IDE_HFLAG_MMIO,
 	.pio_mask		= ATA_PIO4,
-	.udma_mask		= ATA_UDMA4,	/* (input clk 99MHz) */
 	.mwdma_mask		= ATA_MWDMA2,
 };
 
@@ -352,7 +352,7 @@ static int __devinit palm_bk3710_probe(struct platform_device *pdev)
 	int i, rc;
 	hw_regs_t hw, *hws[] = { &hw, NULL, NULL, NULL };
 
-	clk = clk_get(NULL, "IDECLK");
+	clk = clk_get(&pdev->dev, "IDECLK");
 	if (IS_ERR(clk))
 		return -ENODEV;
 
@@ -392,6 +392,9 @@ static int __devinit palm_bk3710_probe(struct platform_device *pdev)
 	hw.irq = irq->start;
 	hw.chipset = ide_palm3710;
 
+	palm_bk3710_port_info.udma_mask = rate < 100000000 ? ATA_UDMA4 :
+							     ATA_UDMA5;
+
 	rc = ide_host_add(&palm_bk3710_port_info, hws, NULL);
 	if (rc)
 		goto out;

From 96cc112c09b3c6674da01ef8b377f7a916883ea2 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:34 +0200
Subject: [PATCH 0593/1171] gayle: release resources on ide_host_add() failure

"gayle: reserve memory resources at once" patch temporary removed
freeing of resources on failure (to ease convertion to ide_host_add()
interface).  This patch fixes it.

Thanks to Geert for noticing the issue.

Noticed-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/legacy/gayle.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/ide/legacy/gayle.c b/drivers/ide/legacy/gayle.c
index dd5c467d8dd0..51ba085d7aa8 100644
--- a/drivers/ide/legacy/gayle.c
+++ b/drivers/ide/legacy/gayle.c
@@ -127,7 +127,7 @@ static int __init gayle_init(void)
     unsigned long phys_base, res_start, res_n;
     unsigned long base, ctrlport, irqport;
     ide_ack_intr_t *ack_intr;
-    int a4000, i;
+    int a4000, i, rc;
     hw_regs_t hw[GAYLE_NUM_HWIFS], *hws[] = { NULL, NULL, NULL, NULL };
 
     if (!MACH_IS_AMIGA)
@@ -179,7 +179,11 @@ found:
 	hws[i] = &hw[i];
     }
 
-    return ide_host_add(NULL, hws, NULL);
+    rc = ide_host_add(NULL, hws, NULL);
+    if (rc)
+	release_mem_region(res_start, res_n);
+
+    return rc;
 }
 
 module_init(gayle_init);

From e8e7b9eb11c34ee18bde8b7011af41938d1ad667 Mon Sep 17 00:00:00 2001
From: Jens Axboe <jens.axboe@oracle.com>
Date: Thu, 24 Jul 2008 22:53:35 +0200
Subject: [PATCH 0594/1171] ide-cd: fix oops when using growisofs

cdrom_read_capacity() will blindly return the capacity from the device
without sanity-checking it.  This later causes code in fs/buffer.c to
oops.

Fix this by checking that the device is telling us sensible things.

From: Jens Axboe <jens.axboe@oracle.com>
Cc: Michael Buesch <mb@bu3sch.de>
Cc: Jan Kara <jack@suse.cz>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: <stable@kernel.org>
Cc: Borislav Petkov <petkovbb@googlemail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
[bart: print device name instead of driver name]
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
[harvey: blocklen is a big-endian value]
Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-cd.c | 27 ++++++++++++++++++++++-----
 1 file changed, 22 insertions(+), 5 deletions(-)

diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index 8f253e5f26a8..e617cf08aef6 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -1311,13 +1311,30 @@ static int cdrom_read_capacity(ide_drive_t *drive, unsigned long *capacity,
 
 	stat = ide_cd_queue_pc(drive, cmd, 0, &capbuf, &len, sense, 0,
 			       REQ_QUIET);
-	if (stat == 0) {
-		*capacity = 1 + be32_to_cpu(capbuf.lba);
-		*sectors_per_frame =
-			be32_to_cpu(capbuf.blocklen) >> SECTOR_BITS;
+	if (stat)
+		return stat;
+
+	/*
+	 * Sanity check the given block size
+	 */
+	switch (capbuf.blocklen) {
+	case __constant_cpu_to_be32(512):
+	case __constant_cpu_to_be32(1024):
+	case __constant_cpu_to_be32(2048):
+	case __constant_cpu_to_be32(4096):
+		break;
+	default:
+		printk(KERN_ERR "%s: weird block size %u\n",
+			drive->name, capbuf.blocklen);
+		printk(KERN_ERR "%s: default to 2kb block size\n",
+			drive->name);
+		capbuf.blocklen = __constant_cpu_to_be32(2048);
+		break;
 	}
 
-	return stat;
+	*capacity = 1 + be32_to_cpu(capbuf.lba);
+	*sectors_per_frame = be32_to_cpu(capbuf.blocklen) >> SECTOR_BITS;
+	return 0;
 }
 
 static int cdrom_read_tocentry(ide_drive_t *drive, int trackno, int msf_flag,

From eb34b2d90e71380ad19695188934230b06a3668b Mon Sep 17 00:00:00 2001
From: Jan Evert van Grootheest <j.e.van.grootheest@caiway.nl>
Date: Thu, 24 Jul 2008 22:53:35 +0200
Subject: [PATCH 0595/1171] ht6560b: update email address

Update email address.

From: Jan Evert van Grootheest <j.e.van.grootheest@caiway.nl>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/legacy/ht6560b.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/ide/legacy/ht6560b.c b/drivers/ide/legacy/ht6560b.c
index 7bc8fd59ea9e..7262b3ad45c6 100644
--- a/drivers/ide/legacy/ht6560b.c
+++ b/drivers/ide/legacy/ht6560b.c
@@ -30,7 +30,7 @@
  *  Use hdparm utility to enable PIO mode support.
  *
  *  Author:    Mikko Ala-Fossi            <maf@iki.fi>
- *             Jan Evert van Grootheest   <janevert@caiway.nl>
+ *             Jan Evert van Grootheest   <j.e.van.grootheest@caiway.nl>
  *
  *  Try:  http://www.maf.iki.fi/~maf/ht6560b/
  */

From 216f9a88feabf5ed574c3aa78447a6bd872910bc Mon Sep 17 00:00:00 2001
From: Jan Evert van Grootheest <j.e.van.grootheest@caiway.nl>
Date: Thu, 24 Jul 2008 22:53:35 +0200
Subject: [PATCH 0596/1171] ht6560b: remove old history

Remove the ancient version history. Git does a better job.

From: Jan Evert van Grootheest <j.e.van.grootheest@caiway.nl>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/legacy/ht6560b.c | 22 ----------------------
 1 file changed, 22 deletions(-)

diff --git a/drivers/ide/legacy/ht6560b.c b/drivers/ide/legacy/ht6560b.c
index 7262b3ad45c6..98f7c95e39ed 100644
--- a/drivers/ide/legacy/ht6560b.c
+++ b/drivers/ide/legacy/ht6560b.c
@@ -3,28 +3,6 @@
  */
 
 /*
- *
- *  Version 0.01        Initial version hacked out of ide.c
- *
- *  Version 0.02        Added support for PIO modes, auto-tune
- *
- *  Version 0.03        Some cleanups
- *
- *  Version 0.05        PIO mode cycle timings auto-tune using bus-speed
- *
- *  Version 0.06        Prefetch mode now defaults no OFF. To set
- *                      prefetch mode OFF/ON use "hdparm -p8/-p9".
- *                      Unmask irq is disabled when prefetch mode
- *                      is enabled.
- *
- *  Version 0.07        Trying to fix CD-ROM detection problem.
- *                      "Prefetch" mode bit OFF for ide disks and
- *                      ON for anything else.
- *
- *  Version 0.08        Need to force prefetch for CDs and other non-disk
- *                      devices. (not sure which devices exactly need
- *                      prefetch)
- *
  *  HT-6560B EIDE-controller support
  *  To activate controller support use kernel parameter "ide0=ht6560b".
  *  Use hdparm utility to enable PIO mode support.

From e27420d046600cd3e4139ea1b6cba59a8b4050eb Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Thu, 24 Jul 2008 22:53:35 +0200
Subject: [PATCH 0597/1171] ide-scsi: remove kmalloced struct request

This converts ide-scsi to use blk_get/put_request instead of
kmalloc/kfree.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/scsi/ide-scsi.c | 25 +++++++++++++++----------
 1 file changed, 15 insertions(+), 10 deletions(-)

diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index 318ef382448f..b40a673985aa 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -207,15 +207,15 @@ static int idescsi_check_condition(ide_drive_t *drive,
 
 	/* stuff a sense request in front of our current request */
 	pc = kzalloc(sizeof(struct ide_atapi_pc), GFP_ATOMIC);
-	rq = kmalloc(sizeof(struct request), GFP_ATOMIC);
+	rq = blk_get_request(drive->queue, READ, GFP_ATOMIC);
 	buf = kzalloc(SCSI_SENSE_BUFFERSIZE, GFP_ATOMIC);
 	if (!pc || !rq || !buf) {
 		kfree(buf);
-		kfree(rq);
+		if (rq)
+			blk_put_request(rq);
 		kfree(pc);
 		return -ENOMEM;
 	}
-	blk_rq_init(NULL, rq);
 	rq->special = (char *) pc;
 	pc->rq = rq;
 	pc->buf = buf;
@@ -232,6 +232,7 @@ static int idescsi_check_condition(ide_drive_t *drive,
 		ide_scsi_hex_dump(pc->c, 6);
 	}
 	rq->rq_disk = scsi->disk;
+	rq->ref_count++;
 	memcpy(rq->cmd, pc->c, 12);
 	ide_do_drive_cmd(drive, rq);
 	return 0;
@@ -278,7 +279,7 @@ static int idescsi_end_request (ide_drive_t *drive, int uptodate, int nrsecs)
 			SCSI_SENSE_BUFFERSIZE);
 		kfree(pc->buf);
 		kfree(pc);
-		kfree(rq);
+		blk_put_request(rq);
 		pc = opc;
 		rq = pc->rq;
 		pc->scsi_cmd->result = (CHECK_CONDITION << 1) |
@@ -309,7 +310,7 @@ static int idescsi_end_request (ide_drive_t *drive, int uptodate, int nrsecs)
 	pc->done(pc->scsi_cmd);
 	spin_unlock_irqrestore(host->host_lock, flags);
 	kfree(pc);
-	kfree(rq);
+	blk_put_request(rq);
 	scsi->pc = NULL;
 	return 0;
 }
@@ -583,6 +584,7 @@ static int idescsi_queue (struct scsi_cmnd *cmd,
 	ide_drive_t *drive = scsi->drive;
 	struct request *rq = NULL;
 	struct ide_atapi_pc *pc = NULL;
+	int write = cmd->sc_data_direction == DMA_TO_DEVICE;
 
 	if (!drive) {
 		scmd_printk (KERN_ERR, cmd, "drive not present\n");
@@ -590,7 +592,7 @@ static int idescsi_queue (struct scsi_cmnd *cmd,
 	}
 	scsi = drive_to_idescsi(drive);
 	pc = kmalloc(sizeof(struct ide_atapi_pc), GFP_ATOMIC);
-	rq = kmalloc(sizeof(struct request), GFP_ATOMIC);
+	rq = blk_get_request(drive->queue, write, GFP_ATOMIC);
 	if (rq == NULL || pc == NULL) {
 		printk (KERN_ERR "ide-scsi: %s: out of memory\n", drive->name);
 		goto abort;
@@ -620,17 +622,18 @@ static int idescsi_queue (struct scsi_cmnd *cmd,
 		}
 	}
 
-	blk_rq_init(NULL, rq);
 	rq->special = (char *) pc;
 	rq->cmd_type = REQ_TYPE_SPECIAL;
 	spin_unlock_irq(host->host_lock);
+	rq->ref_count++;
 	memcpy(rq->cmd, pc->c, 12);
 	blk_execute_rq_nowait(drive->queue, scsi->disk, rq, 0, NULL);
 	spin_lock_irq(host->host_lock);
 	return 0;
 abort:
 	kfree (pc);
-	kfree (rq);
+	if (rq)
+		blk_put_request(rq);
 	cmd->result = DID_ERROR << 16;
 	done(cmd);
 	return 0;
@@ -678,7 +681,9 @@ static int idescsi_eh_abort (struct scsi_cmnd *cmd)
 
 		if (blk_sense_request(scsi->pc->rq))
 			kfree(scsi->pc->buf);
-		kfree(scsi->pc->rq);
+		/* we need to call blk_put_request twice. */
+		blk_put_request(scsi->pc->rq);
+		blk_put_request(scsi->pc->rq);
 		kfree(scsi->pc);
 		scsi->pc = NULL;
 
@@ -730,7 +735,7 @@ static int idescsi_eh_reset (struct scsi_cmnd *cmd)
 		kfree(scsi->pc->buf);
 	kfree(scsi->pc);
 	scsi->pc = NULL;
-	kfree(req);
+	blk_put_request(req);
 
 	/* now nuke the drive queue */
 	while ((req = elv_next_request(drive->queue))) {

From 90d2c6bc68745d67cdbf00bab43818d90aa0dfb6 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:36 +0200
Subject: [PATCH 0598/1171] ide: enable local IRQs in all handlers for
 TASKFILE_NO_DATA data phase

It is already done by task_no_data_intr() and there is no reason
not to do it in other TASKFILE_NO_DATA data phase handlers.

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-taskfile.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
index aeddbbd69e86..7fb6f1c86272 100644
--- a/drivers/ide/ide-taskfile.c
+++ b/drivers/ide/ide-taskfile.c
@@ -126,7 +126,10 @@ EXPORT_SYMBOL_GPL(do_rw_taskfile);
 static ide_startstop_t set_multmode_intr(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	u8 stat = hwif->tp_ops->read_status(hwif);
+	u8 stat;
+
+	local_irq_enable_in_hardirq();
+	stat = hwif->tp_ops->read_status(hwif);
 
 	if (OK_STAT(stat, READY_STAT, BAD_STAT))
 		drive->mult_count = drive->mult_req;
@@ -147,6 +150,8 @@ static ide_startstop_t set_geometry_intr(ide_drive_t *drive)
 	int retries = 5;
 	u8 stat;
 
+	local_irq_enable_in_hardirq();
+
 	while (1) {
 		stat = hwif->tp_ops->read_status(hwif);
 		if ((stat & BUSY_STAT) == 0 || retries-- == 0)
@@ -170,7 +175,10 @@ static ide_startstop_t set_geometry_intr(ide_drive_t *drive)
 static ide_startstop_t recal_intr(ide_drive_t *drive)
 {
 	ide_hwif_t *hwif = drive->hwif;
-	u8 stat = hwif->tp_ops->read_status(hwif);
+	u8 stat;
+
+	local_irq_enable_in_hardirq();
+	stat = hwif->tp_ops->read_status(hwif);
 
 	if (!OK_STAT(stat, READY_STAT, BAD_STAT))
 		return ide_error(drive, "recal_intr", stat);

From d0b53f6866fa185da94968e62ae97923db18298c Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:36 +0200
Subject: [PATCH 0599/1171] ide: remove stale comments from
 drivers/ide/Makefile

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/Makefile | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

diff --git a/drivers/ide/Makefile b/drivers/ide/Makefile
index 5d414e301a5a..64e0ecdc4ed5 100644
--- a/drivers/ide/Makefile
+++ b/drivers/ide/Makefile
@@ -1,13 +1,6 @@
 #
-# Makefile for the kernel ata, atapi, and ide block device drivers.
-#
-# 12 September 2000, Bartlomiej Zolnierkiewicz <bkz@linux-ide.org>
-# Rewritten to use lists instead of if-statements.
-#
-# Note : at this point, these files are compiled on all systems.
-# In the future, some of these should be built conditionally.
-#
 # link order is important here
+#
 
 EXTRA_CFLAGS				+= -Idrivers/ide
 

From 52f3a771feafe3e9c56f8d00c8eb53fd8f578f2d Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:36 +0200
Subject: [PATCH 0600/1171] ide: fix for EATA SCSI HBA in ATA emulating mode

IDE probing code used to skip devices attached to EATA SCSI HBA
in ATA emulating mode but because of warm-plug support port I/O
resources are no longer freed if no devices are detected on a port
and the decision about the driver to use is left up to the user.

Remove no longer valid EATA SCSI HBA quirk from do_identify().

Noticed-by: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-probe.c | 12 ------------
 1 file changed, 12 deletions(-)

diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index f0c162488ec4..d227fe425d2f 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -134,18 +134,6 @@ static inline void do_identify (ide_drive_t *drive, u8 cmd)
 #endif
 	ide_fix_driveid(id);
 
-#if defined (CONFIG_SCSI_EATA_PIO) || defined (CONFIG_SCSI_EATA)
-	/*
-	 * EATA SCSI controllers do a hardware ATA emulation:
-	 * Ignore them if there is a driver for them available.
-	 */
-	if ((id->model[0] == 'P' && id->model[1] == 'M') ||
-	    (id->model[0] == 'S' && id->model[1] == 'K')) {
-		printk("%s: EATA SCSI HBA %.10s\n", drive->name, id->model);
-		goto err_misc;
-	}
-#endif /* CONFIG_SCSI_EATA || CONFIG_SCSI_EATA_PIO */
-
 	/*
 	 *  WIN_IDENTIFY returns little-endian info,
 	 *  WIN_PIDENTIFY *usually* returns little-endian info.

From 1b8ebad87b459e2e1333fbf28005977245ff5402 Mon Sep 17 00:00:00 2001
From: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Date: Thu, 24 Jul 2008 22:53:36 +0200
Subject: [PATCH 0601/1171] ide: use proper printk() KERN_* levels in
 ide-probe.c

While at it:

- fixup printk() messages in save_match() and hwif_init().

Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
---
 drivers/ide/ide-probe.c | 51 ++++++++++++++++++++++-------------------
 1 file changed, 28 insertions(+), 23 deletions(-)

diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
index d227fe425d2f..994e41099b42 100644
--- a/drivers/ide/ide-probe.c
+++ b/drivers/ide/ide-probe.c
@@ -155,7 +155,8 @@ static inline void do_identify (ide_drive_t *drive, u8 cmd)
 	if (strstr(id->model, "E X A B Y T E N E S T"))
 		goto err_misc;
 
-	printk("%s: %s, ", drive->name, id->model);
+	printk(KERN_INFO "%s: %s, ", drive->name, id->model);
+
 	drive->present = 1;
 	drive->dead = 0;
 
@@ -164,16 +165,17 @@ static inline void do_identify (ide_drive_t *drive, u8 cmd)
 	 */
 	if (cmd == WIN_PIDENTIFY) {
 		u8 type = (id->config >> 8) & 0x1f;
-		printk("ATAPI ");
+
+		printk(KERN_CONT "ATAPI ");
 		switch (type) {
 			case ide_floppy:
 				if (!strstr(id->model, "CD-ROM")) {
 					if (!strstr(id->model, "oppy") &&
 					    !strstr(id->model, "poyp") &&
 					    !strstr(id->model, "ZIP"))
-						printk("cdrom or floppy?, assuming ");
+						printk(KERN_CONT "cdrom or floppy?, assuming ");
 					if (drive->media != ide_cdrom) {
-						printk ("FLOPPY");
+						printk(KERN_CONT "FLOPPY");
 						drive->removable = 1;
 						break;
 					}
@@ -186,25 +188,25 @@ static inline void do_identify (ide_drive_t *drive, u8 cmd)
 				/* kludge for Apple PowerBook internal zip */
 				if (!strstr(id->model, "CD-ROM") &&
 				    strstr(id->model, "ZIP")) {
-					printk ("FLOPPY");
+					printk(KERN_CONT "FLOPPY");
 					type = ide_floppy;
 					break;
 				}
 #endif
-				printk ("CD/DVD-ROM");
+				printk(KERN_CONT "CD/DVD-ROM");
 				break;
 			case ide_tape:
-				printk ("TAPE");
+				printk(KERN_CONT "TAPE");
 				break;
 			case ide_optical:
-				printk ("OPTICAL");
+				printk(KERN_CONT "OPTICAL");
 				drive->removable = 1;
 				break;
 			default:
-				printk("UNKNOWN (type %d)", type);
+				printk(KERN_CONT "UNKNOWN (type %d)", type);
 				break;
 		}
-		printk (" drive\n");
+		printk(KERN_CONT " drive\n");
 		drive->media = type;
 		/* an ATAPI device ignores DRDY */
 		drive->ready_stat = 0;
@@ -224,7 +226,9 @@ static inline void do_identify (ide_drive_t *drive, u8 cmd)
 		drive->removable = 1;
 
 	drive->media = ide_disk;
-	printk("%s DISK drive\n", (id->config == 0x848a) ? "CFA" : "ATA" );
+
+	printk(KERN_CONT "%s DISK drive\n",
+		(id->config == 0x848a) ? "CFA" : "ATA");
 
 	return;
 
@@ -375,7 +379,7 @@ static int try_to_identify (ide_drive_t *drive, u8 cmd)
 				/* Mmmm.. multiple IRQs..
 				 * don't know which was ours
 				 */
-				printk("%s: IRQ probe failed (0x%lx)\n",
+				printk(KERN_ERR "%s: IRQ probe failed (0x%lx)\n",
 					drive->name, cookie);
 			}
 		}
@@ -444,7 +448,7 @@ static int do_probe (ide_drive_t *drive, u8 cmd)
 			return 4;
 	}
 #ifdef DEBUG
-	printk("probing for %s: present=%d, media=%d, probetype=%s\n",
+	printk(KERN_INFO "probing for %s: present=%d, media=%d, probetype=%s\n",
 		drive->name, drive->present, drive->media,
 		(cmd == WIN_IDENTIFY) ? "ATA" : "ATAPI");
 #endif
@@ -522,7 +526,8 @@ static void enable_nest (ide_drive_t *drive)
 	const struct ide_tp_ops *tp_ops = hwif->tp_ops;
 	u8 stat;
 
-	printk("%s: enabling %s -- ", hwif->name, drive->id->model);
+	printk(KERN_INFO "%s: enabling %s -- ", hwif->name, drive->id->model);
+
 	SELECT_DRIVE(drive);
 	msleep(50);
 	tp_ops->exec_command(hwif, EXABYTE_ENABLE_NEST);
@@ -871,7 +876,7 @@ static void save_match(ide_hwif_t *hwif, ide_hwif_t *new, ide_hwif_t **match)
 	if (m && m->hwgroup && m->hwgroup != new->hwgroup) {
 		if (!new->hwgroup)
 			return;
-		printk("%s: potential irq problem with %s and %s\n",
+		printk(KERN_WARNING "%s: potential IRQ problem with %s and %s\n",
 			hwif->name, new->name, m->name);
 	}
 	if (!m || m->irq != hwif->irq) /* don't undo a prior perfect match */
@@ -1130,17 +1135,17 @@ static int init_irq (ide_hwif_t *hwif)
 	}
 
 #if !defined(__mc68000__)
-	printk("%s at 0x%03lx-0x%03lx,0x%03lx on irq %d", hwif->name,
+	printk(KERN_INFO "%s at 0x%03lx-0x%03lx,0x%03lx on irq %d", hwif->name,
 		io_ports->data_addr, io_ports->status_addr,
 		io_ports->ctl_addr, hwif->irq);
 #else
-	printk("%s at 0x%08lx on irq %d", hwif->name,
+	printk(KERN_INFO "%s at 0x%08lx on irq %d", hwif->name,
 		io_ports->data_addr, hwif->irq);
 #endif /* __mc68000__ */
 	if (match)
-		printk(" (%sed with %s)",
+		printk(KERN_CONT " (%sed with %s)",
 			hwif->sharing_irq ? "shar" : "serializ", match->name);
-	printk("\n");
+	printk(KERN_CONT "\n");
 
 	mutex_unlock(&ide_cfg_mtx);
 	return 0;
@@ -1275,7 +1280,7 @@ static int hwif_init(ide_hwif_t *hwif)
 	if (!hwif->irq) {
 		hwif->irq = __ide_default_irq(hwif->io_ports.data_addr);
 		if (!hwif->irq) {
-			printk("%s: DISABLED, NO IRQ\n", hwif->name);
+			printk(KERN_ERR "%s: disabled, no IRQ\n", hwif->name);
 			return 0;
 		}
 	}
@@ -1305,16 +1310,16 @@ static int hwif_init(ide_hwif_t *hwif)
 	 */
 	hwif->irq = __ide_default_irq(hwif->io_ports.data_addr);
 	if (!hwif->irq) {
-		printk("%s: Disabled unable to get IRQ %d.\n",
+		printk(KERN_ERR "%s: disabled, unable to get IRQ %d\n",
 			hwif->name, old_irq);
 		goto out;
 	}
 	if (init_irq(hwif)) {
-		printk("%s: probed IRQ %d and default IRQ %d failed.\n",
+		printk(KERN_ERR "%s: probed IRQ %d and default IRQ %d failed\n",
 			hwif->name, old_irq, hwif->irq);
 		goto out;
 	}
-	printk("%s: probed IRQ %d failed, using default.\n",
+	printk(KERN_WARNING "%s: probed IRQ %d failed, using default\n",
 		hwif->name, hwif->irq);
 
 done:

From b30f3ae50cd03ef2ff433a5030fbf88dd8323528 Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Thu, 24 Jul 2008 15:43:44 -0700
Subject: [PATCH 0602/1171] x86-64: Clean up 'save/restore_i387()' usage

Suresh Siddha wants to fix a possible FPU leakage in error conditions,
but the fact that save/restore_i387() are inlines in a header file makes
that harder to do than necessary.  So start off with an obvious cleanup.

This just moves the x86-64 version of save/restore_i387() out of the
header file, and moves it to the only file that it is actually used in:
arch/x86/kernel/signal_64.c.  So exposing it in a header file was wrong
to begin with.

[ Side note: I'd like to fix up some of the games we play with the
  32-bit version of these functions too, but that's a separate
  matter.  The 32-bit versions are shared - under different names
  at that! - by both the native x86-32 code and the x86-64 32-bit
  compatibility code ]

Acked-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/kernel/signal_64.c | 53 ++++++++++++++++++++++++++++++++++++
 include/asm-x86/i387.h      | 54 -------------------------------------
 2 files changed, 53 insertions(+), 54 deletions(-)

diff --git a/arch/x86/kernel/signal_64.c b/arch/x86/kernel/signal_64.c
index 47c3d249e638..b45ef8ddd651 100644
--- a/arch/x86/kernel/signal_64.c
+++ b/arch/x86/kernel/signal_64.c
@@ -53,6 +53,59 @@ sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss,
 	return do_sigaltstack(uss, uoss, regs->sp);
 }
 
+/*
+ * Signal frame handlers.
+ */
+
+static inline int save_i387(struct _fpstate __user *buf)
+{
+	struct task_struct *tsk = current;
+	int err = 0;
+
+	BUILD_BUG_ON(sizeof(struct user_i387_struct) !=
+			sizeof(tsk->thread.xstate->fxsave));
+
+	if ((unsigned long)buf % 16)
+		printk("save_i387: bad fpstate %p\n", buf);
+
+	if (!used_math())
+		return 0;
+	clear_used_math(); /* trigger finit */
+	if (task_thread_info(tsk)->status & TS_USEDFPU) {
+		err = save_i387_checking((struct i387_fxsave_struct __user *)
+					 buf);
+		if (err)
+			return err;
+		task_thread_info(tsk)->status &= ~TS_USEDFPU;
+		stts();
+	} else {
+		if (__copy_to_user(buf, &tsk->thread.xstate->fxsave,
+				   sizeof(struct i387_fxsave_struct)))
+			return -1;
+	}
+	return 1;
+}
+
+/*
+ * This restores directly out of user space. Exceptions are handled.
+ */
+static inline int restore_i387(struct _fpstate __user *buf)
+{
+	struct task_struct *tsk = current;
+	int err;
+
+	if (!used_math()) {
+		err = init_fpu(tsk);
+		if (err)
+			return err;
+	}
+
+	if (!(task_thread_info(current)->status & TS_USEDFPU)) {
+		clts();
+		task_thread_info(current)->status |= TS_USEDFPU;
+	}
+	return restore_fpu_checking((__force struct i387_fxsave_struct *)buf);
+}
 
 /*
  * Do a signal return; undo the signal stack.
diff --git a/include/asm-x86/i387.h b/include/asm-x86/i387.h
index 37672f79dcc8..96fa8449ff11 100644
--- a/include/asm-x86/i387.h
+++ b/include/asm-x86/i387.h
@@ -137,60 +137,6 @@ static inline void __save_init_fpu(struct task_struct *tsk)
 	task_thread_info(tsk)->status &= ~TS_USEDFPU;
 }
 
-/*
- * Signal frame handlers.
- */
-
-static inline int save_i387(struct _fpstate __user *buf)
-{
-	struct task_struct *tsk = current;
-	int err = 0;
-
-	BUILD_BUG_ON(sizeof(struct user_i387_struct) !=
-			sizeof(tsk->thread.xstate->fxsave));
-
-	if ((unsigned long)buf % 16)
-		printk("save_i387: bad fpstate %p\n", buf);
-
-	if (!used_math())
-		return 0;
-	clear_used_math(); /* trigger finit */
-	if (task_thread_info(tsk)->status & TS_USEDFPU) {
-		err = save_i387_checking((struct i387_fxsave_struct __user *)
-					 buf);
-		if (err)
-			return err;
-		task_thread_info(tsk)->status &= ~TS_USEDFPU;
-		stts();
-	} else {
-		if (__copy_to_user(buf, &tsk->thread.xstate->fxsave,
-				   sizeof(struct i387_fxsave_struct)))
-			return -1;
-	}
-	return 1;
-}
-
-/*
- * This restores directly out of user space. Exceptions are handled.
- */
-static inline int restore_i387(struct _fpstate __user *buf)
-{
-	struct task_struct *tsk = current;
-	int err;
-
-	if (!used_math()) {
-		err = init_fpu(tsk);
-		if (err)
-			return err;
-	}
-
-	if (!(task_thread_info(current)->status & TS_USEDFPU)) {
-		clts();
-		task_thread_info(current)->status |= TS_USEDFPU;
-	}
-	return restore_fpu_checking((__force struct i387_fxsave_struct *)buf);
-}
-
 #else  /* CONFIG_X86_32 */
 
 extern void finit(void);

From 4b9f12a3779c548b68bc9af7d94030868ad3aa1b Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Thu, 24 Jul 2008 17:29:00 -0700
Subject: [PATCH 0603/1171] x86/oprofile/nmi_int: add Nehalem to list of ppro
 cores

..otherwise oprofile will fall back on that poor timer interrupt.

Also replace the unreadable chain of if-statements with a "switch()"
statement instead. It generates better code, and is a lot clearer.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/oprofile/nmi_int.c | 40 +++++++++++++++++++++++++------------
 1 file changed, 27 insertions(+), 13 deletions(-)

diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c
index 7f3329b55d2e..3f90289410e6 100644
--- a/arch/x86/oprofile/nmi_int.c
+++ b/arch/x86/oprofile/nmi_int.c
@@ -369,20 +369,34 @@ static int __init ppro_init(char **cpu_type)
 {
 	__u8 cpu_model = boot_cpu_data.x86_model;
 
-	if (cpu_model == 14)
-		*cpu_type = "i386/core";
-	else if (cpu_model == 15 || cpu_model == 23)
-		*cpu_type = "i386/core_2";
-	else if (cpu_model > 0xd)
-		return 0;
-	else if (cpu_model == 9) {
-		*cpu_type = "i386/p6_mobile";
-	} else if (cpu_model > 5) {
-		*cpu_type = "i386/piii";
-	} else if (cpu_model > 2) {
-		*cpu_type = "i386/pii";
-	} else {
+	switch (cpu_model) {
+	case 0 ... 2:
 		*cpu_type = "i386/ppro";
+		break;
+	case 3 ... 5:
+		*cpu_type = "i386/pii";
+		break;
+	case 6 ... 8:
+		*cpu_type = "i386/piii";
+		break;
+	case 9:
+		*cpu_type = "i386/p6_mobile";
+		break;
+	case 10 ... 13:
+		*cpu_type = "i386/p6";
+		break;
+	case 14:
+		*cpu_type = "i386/core";
+		break;
+	case 15: case 23:
+		*cpu_type = "i386/core_2";
+		break;
+	case 26:
+		*cpu_type = "i386/core_2";
+		break;
+	default:
+		/* Unknown */
+		return 0;
 	}
 
 	model = &op_ppro_spec;

From fb2e405fc1fc8b20d9c78eaa1c7fd5a297efde43 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 02:55:49 +0300
Subject: [PATCH 0604/1171] fix fs/nfs/nfsroot.c compilation

This fixes the following compile error caused by commit
f9247273cb69ba101877e946d2d83044409cc8c5 ("UFS: add const to parser
token table"):

    CC      fs/nfs/nfsroot.o
  /home/bunk/linux/kernel-2.6/git/linux-2.6/fs/nfs/nfsroot.c:130: error: tokens causes a section type conflict
  make[3]: *** [fs/nfs/nfsroot.o] Error 1

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/nfs/nfsroot.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/nfs/nfsroot.c b/fs/nfs/nfsroot.c
index 46763d1cd397..8478fc25daee 100644
--- a/fs/nfs/nfsroot.c
+++ b/fs/nfs/nfsroot.c
@@ -127,7 +127,7 @@ enum {
 	Opt_err
 };
 
-static match_table_t __initdata tokens = {
+static match_table_t __initconst tokens = {
 	{Opt_port, "port=%u"},
 	{Opt_rsize, "rsize=%u"},
 	{Opt_wsize, "wsize=%u"},

From 9953ca6cb757fb317bb7cdd2fcbf9b88312e241b Mon Sep 17 00:00:00 2001
From: Mark McLoughlin <markmc@redhat.com>
Date: Tue, 27 May 2008 12:06:26 +0100
Subject: [PATCH 0605/1171] virtio: fix virtio_net xmit of freed skb bug

On Mon, 2008-05-26 at 17:42 +1000, Rusty Russell wrote:
> If we fail to transmit a packet, we assume the queue is full and put
> the skb into last_xmit_skb.  However, if more space frees up before we
> xmit it, we loop, and the result can be transmitting the same skb twice.
>
> Fix is simple: set skb to NULL if we've used it in some way, and check
> before sending.
...
> diff -r 564237b31993 drivers/net/virtio_net.c
> --- a/drivers/net/virtio_net.c	Mon May 19 12:22:00 2008 +1000
> +++ b/drivers/net/virtio_net.c	Mon May 19 12:24:58 2008 +1000
> @@ -287,21 +287,25 @@ again:
>  	free_old_xmit_skbs(vi);
>
>  	/* If we has a buffer left over from last time, send it now. */
> -	if (vi->last_xmit_skb) {
> +	if (unlikely(vi->last_xmit_skb)) {
>  		if (xmit_skb(vi, vi->last_xmit_skb) != 0) {
>  			/* Drop this skb: we only queue one. */
>  			vi->dev->stats.tx_dropped++;
>  			kfree_skb(skb);
> +			skb = NULL;
>  			goto stop_queue;
>  		}
>  		vi->last_xmit_skb = NULL;

With this, may drop an skb and then later in the function discover that
we could have sent it after all. Poor wee skb :)

How about the incremental patch below?

Cheers,
Mark.

Subject: [PATCH] virtio_net: Delay dropping tx skbs

Currently we drop the skb in start_xmit() if we have a
queued buffer and fail to transmit it.

However, if we delay dropping it until we've stopped the
queue and enabled the tx notification callback, then there
is a chance space might become available for it.

Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 drivers/net/virtio_net.c | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index c28d7cb2035b..06d5c43bb207 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -335,16 +335,11 @@ again:
 	free_old_xmit_skbs(vi);
 
 	/* If we has a buffer left over from last time, send it now. */
-	if (unlikely(vi->last_xmit_skb)) {
-		if (xmit_skb(vi, vi->last_xmit_skb) != 0) {
-			/* Drop this skb: we only queue one. */
-			vi->dev->stats.tx_dropped++;
-			kfree_skb(skb);
-			skb = NULL;
-			goto stop_queue;
-		}
-		vi->last_xmit_skb = NULL;
-	}
+	if (unlikely(vi->last_xmit_skb) &&
+	    xmit_skb(vi, vi->last_xmit_skb) != 0)
+		goto stop_queue;
+
+	vi->last_xmit_skb = NULL;
 
 	/* Put new one in send queue and do transmit */
 	if (likely(skb)) {
@@ -370,6 +365,11 @@ stop_queue:
 		netif_start_queue(dev);
 		goto again;
 	}
+	if (skb) {
+		/* Drop this skb: we only queue one. */
+		vi->dev->stats.tx_dropped++;
+		kfree_skb(skb);
+	}
 	goto done;
 }
 

From a9ea3fc6f2654a7407864fec983d1671d775b5ee Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Fri, 18 Apr 2008 11:21:42 +0800
Subject: [PATCH 0606/1171] virtio net: Add ethtool ops for SG/GSO

This patch adds some basic ethtool operations to virtio_net so
I could test SG without GSO (which was really useful because TSO
turned out to be buggy :)

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (remove MTU setting)
---
 drivers/net/virtio_net.c | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 06d5c43bb207..ce37a7e9541c 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -19,6 +19,7 @@
 //#define DEBUG
 #include <linux/netdevice.h>
 #include <linux/etherdevice.h>
+#include <linux/ethtool.h>
 #include <linux/module.h>
 #include <linux/virtio.h>
 #include <linux/virtio_net.h>
@@ -408,6 +409,22 @@ static int virtnet_close(struct net_device *dev)
 	return 0;
 }
 
+static int virtnet_set_tx_csum(struct net_device *dev, u32 data)
+{
+	struct virtnet_info *vi = netdev_priv(dev);
+	struct virtio_device *vdev = vi->vdev;
+
+	if (data && !virtio_has_feature(vdev, VIRTIO_NET_F_CSUM))
+		return -ENOSYS;
+
+	return ethtool_op_set_tx_hw_csum(dev, data);
+}
+
+static struct ethtool_ops virtnet_ethtool_ops = {
+	.set_tx_csum = virtnet_set_tx_csum,
+	.set_sg = ethtool_op_set_sg,
+};
+
 static int virtnet_probe(struct virtio_device *vdev)
 {
 	int err;
@@ -427,6 +444,7 @@ static int virtnet_probe(struct virtio_device *vdev)
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	dev->poll_controller = virtnet_netpoll;
 #endif
+	SET_ETHTOOL_OPS(dev, &virtnet_ethtool_ops);
 	SET_NETDEV_DEV(dev, &vdev->dev);
 
 	/* Do we support "hardware" checksums? */

From 97402b96f87c6e32f75f1bffdd91a5ee144b679d Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Fri, 18 Apr 2008 11:24:27 +0800
Subject: [PATCH 0607/1171] virtio net: Allow receiving SG packets

Finally this patch lets virtio_net receive GSO packets in addition
to sending them.  This can definitely be optimised for the non-GSO
case.  For comparison the Xen approach stores one page in each skb
and uses subsequent skb's pages to construct an SG skb instead of
preallocating the maximum amount of pages per skb.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (added feature bits)
---
 drivers/net/virtio_net.c | 44 +++++++++++++++++++++++++++++++++++-----
 1 file changed, 39 insertions(+), 5 deletions(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index ce37a7e9541c..0886b8a2d92d 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -55,6 +55,9 @@ struct virtnet_info
 	struct tasklet_struct tasklet;
 	bool free_in_tasklet;
 
+	/* I like... big packets and I cannot lie! */
+	bool big_packets;
+
 	/* Receive & send queues. */
 	struct sk_buff_head recv;
 	struct sk_buff_head send;
@@ -89,6 +92,7 @@ static void receive_skb(struct net_device *dev, struct sk_buff *skb,
 			unsigned len)
 {
 	struct virtio_net_hdr *hdr = skb_vnet_hdr(skb);
+	int err;
 
 	if (unlikely(len < sizeof(struct virtio_net_hdr) + ETH_HLEN)) {
 		pr_debug("%s: short packet %i\n", dev->name, len);
@@ -96,10 +100,14 @@ static void receive_skb(struct net_device *dev, struct sk_buff *skb,
 		goto drop;
 	}
 	len -= sizeof(struct virtio_net_hdr);
-	BUG_ON(len > MAX_PACKET_LEN);
-
-	skb_trim(skb, len);
 
+	err = pskb_trim(skb, len);
+	if (err) {
+		pr_debug("%s: pskb_trim failed %i %d\n", dev->name, len, err);
+		dev->stats.rx_dropped++;
+		goto drop;
+	}
+	skb->truesize += skb->data_len;
 	dev->stats.rx_bytes += skb->len;
 	dev->stats.rx_packets++;
 
@@ -161,7 +169,7 @@ static void try_fill_recv(struct virtnet_info *vi)
 {
 	struct sk_buff *skb;
 	struct scatterlist sg[2+MAX_SKB_FRAGS];
-	int num, err;
+	int num, err, i;
 
 	sg_init_table(sg, 2+MAX_SKB_FRAGS);
 	for (;;) {
@@ -171,6 +179,24 @@ static void try_fill_recv(struct virtnet_info *vi)
 
 		skb_put(skb, MAX_PACKET_LEN);
 		vnet_hdr_to_sg(sg, skb);
+
+		if (vi->big_packets) {
+			for (i = 0; i < MAX_SKB_FRAGS; i++) {
+				skb_frag_t *f = &skb_shinfo(skb)->frags[i];
+				f->page = alloc_page(GFP_ATOMIC);
+				if (!f->page)
+					break;
+
+				f->page_offset = 0;
+				f->size = PAGE_SIZE;
+
+				skb->data_len += PAGE_SIZE;
+				skb->len += PAGE_SIZE;
+
+				skb_shinfo(skb)->nr_frags++;
+			}
+		}
+
 		num = skb_to_sgvec(skb, sg+1, 0, skb->len) + 1;
 		skb_queue_head(&vi->recv, skb);
 
@@ -485,6 +511,12 @@ static int virtnet_probe(struct virtio_device *vdev)
 	 * the timer. */
 	vi->free_in_tasklet = virtio_has_feature(vdev,VIRTIO_F_NOTIFY_ON_EMPTY);
 
+	/* If we can receive ANY GSO packets, we must allocate large ones. */
+	if (virtio_has_feature(vdev, VIRTIO_NET_F_GUEST_TSO4)
+	    || virtio_has_feature(vdev, VIRTIO_NET_F_GUEST_TSO6)
+	    || virtio_has_feature(vdev, VIRTIO_NET_F_GUEST_ECN))
+		vi->big_packets = true;
+
 	/* We expect two virtqueues, receive then send. */
 	vi->rvq = vdev->config->find_vq(vdev, 0, skb_recv_done);
 	if (IS_ERR(vi->rvq)) {
@@ -571,7 +603,9 @@ static unsigned int features[] = {
 	VIRTIO_NET_F_CSUM, VIRTIO_NET_F_GUEST_CSUM,
 	VIRTIO_NET_F_GSO, VIRTIO_NET_F_MAC,
 	VIRTIO_NET_F_HOST_TSO4, VIRTIO_NET_F_HOST_UFO, VIRTIO_NET_F_HOST_TSO6,
-	VIRTIO_NET_F_HOST_ECN, VIRTIO_F_NOTIFY_ON_EMPTY,
+	VIRTIO_NET_F_HOST_ECN, VIRTIO_NET_F_GUEST_TSO4, VIRTIO_NET_F_GUEST_TSO6,
+	VIRTIO_NET_F_GUEST_ECN, /* We don't yet handle UFO input. */
+	VIRTIO_F_NOTIFY_ON_EMPTY,
 };
 
 static struct virtio_driver virtio_net = {

From fb6813f480806d62361719e84777c8e00d3e86a8 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Fri, 25 Jul 2008 12:06:01 -0500
Subject: [PATCH 0608/1171] virtio: Recycle unused recv buffer pages for large
 skbs in net driver

If we hack the virtio_net driver to always allocate full-sized (64k+)
skbuffs, the driver slows down (lguest numbers):

  Time to receive 1GB (small buffers): 10.85 seconds
  Time to receive 1GB (64k+ buffers): 24.75 seconds

Of course, large buffers use up more space in the ring, so we increase
that from 128 to 2048:

  Time to receive 1GB (64k+ buffers, 2k ring): 16.61 seconds

If we recycle pages rather than using alloc_page/free_page:

  Time to receive 1GB (64k+ buffers, 2k ring, recycle pages): 10.81 seconds

This demonstrates that with efficient allocation, we don't need to
have a separate "small buffer" queue.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 drivers/net/virtio_net.c | 36 +++++++++++++++++++++++++++++++++++-
 1 file changed, 35 insertions(+), 1 deletion(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 0886b8a2d92d..0196a0df9021 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -61,6 +61,9 @@ struct virtnet_info
 	/* Receive & send queues. */
 	struct sk_buff_head recv;
 	struct sk_buff_head send;
+
+	/* Chain pages by the private ptr. */
+	struct page *pages;
 };
 
 static inline struct virtio_net_hdr *skb_vnet_hdr(struct sk_buff *skb)
@@ -73,6 +76,23 @@ static inline void vnet_hdr_to_sg(struct scatterlist *sg, struct sk_buff *skb)
 	sg_init_one(sg, skb_vnet_hdr(skb), sizeof(struct virtio_net_hdr));
 }
 
+static void give_a_page(struct virtnet_info *vi, struct page *page)
+{
+	page->private = (unsigned long)vi->pages;
+	vi->pages = page;
+}
+
+static struct page *get_a_page(struct virtnet_info *vi, gfp_t gfp_mask)
+{
+	struct page *p = vi->pages;
+
+	if (p)
+		vi->pages = (struct page *)p->private;
+	else
+		p = alloc_page(gfp_mask);
+	return p;
+}
+
 static void skb_xmit_done(struct virtqueue *svq)
 {
 	struct virtnet_info *vi = svq->vdev->priv;
@@ -101,6 +121,15 @@ static void receive_skb(struct net_device *dev, struct sk_buff *skb,
 	}
 	len -= sizeof(struct virtio_net_hdr);
 
+	if (len <= MAX_PACKET_LEN) {
+		unsigned int i;
+
+		for (i = 0; i < skb_shinfo(skb)->nr_frags; i++)
+			give_a_page(dev->priv, skb_shinfo(skb)->frags[i].page);
+		skb->data_len = 0;
+		skb_shinfo(skb)->nr_frags = 0;
+	}
+
 	err = pskb_trim(skb, len);
 	if (err) {
 		pr_debug("%s: pskb_trim failed %i %d\n", dev->name, len, err);
@@ -183,7 +212,7 @@ static void try_fill_recv(struct virtnet_info *vi)
 		if (vi->big_packets) {
 			for (i = 0; i < MAX_SKB_FRAGS; i++) {
 				skb_frag_t *f = &skb_shinfo(skb)->frags[i];
-				f->page = alloc_page(GFP_ATOMIC);
+				f->page = get_a_page(vi, GFP_ATOMIC);
 				if (!f->page)
 					break;
 
@@ -506,6 +535,7 @@ static int virtnet_probe(struct virtio_device *vdev)
 	vi->dev = dev;
 	vi->vdev = vdev;
 	vdev->priv = vi;
+	vi->pages = NULL;
 
 	/* If they give us a callback when all buffers are done, we don't need
 	 * the timer. */
@@ -591,6 +621,10 @@ static void virtnet_remove(struct virtio_device *vdev)
 	vdev->config->del_vq(vi->svq);
 	vdev->config->del_vq(vi->rvq);
 	unregister_netdev(vi->dev);
+
+	while (vi->pages)
+		__free_pages(get_a_page(vi, GFP_KERNEL), 0);
+
 	free_netdev(vi->dev);
 }
 

From 674bfc23c585b34c42263d73fb51710d49762a23 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Fri, 25 Jul 2008 12:06:03 -0500
Subject: [PATCH 0609/1171] virtio: clarify that ABI is usable by any
 implementations

We want others to implement and use virtio, so it makes sense to BSD
license the non-__KERNEL__ parts of the headers to make this crystal
clear.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Acked-by: Mark McLoughlin <markmc@redhat.com>
Acked-by: Ryan Harper <ryanh@us.ibm.com>
Acked-by: Eric Van Hensbergen <ericvh@gmail.com>
Acked-by: Anthony Liguori <aliguori@us.ibm.com>
---
 include/linux/virtio_9p.h      | 2 ++
 include/linux/virtio_balloon.h | 2 ++
 include/linux/virtio_blk.h     | 2 ++
 include/linux/virtio_config.h  | 3 +++
 include/linux/virtio_console.h | 2 ++
 include/linux/virtio_net.h     | 2 ++
 include/linux/virtio_pci.h     | 5 ++---
 include/linux/virtio_rng.h     | 2 ++
 8 files changed, 17 insertions(+), 3 deletions(-)

diff --git a/include/linux/virtio_9p.h b/include/linux/virtio_9p.h
index 8eff0b53910b..b3c4a60ceeb3 100644
--- a/include/linux/virtio_9p.h
+++ b/include/linux/virtio_9p.h
@@ -1,5 +1,7 @@
 #ifndef _LINUX_VIRTIO_9P_H
 #define _LINUX_VIRTIO_9P_H
+/* This header is BSD licensed so anyone can use the definitions to implement
+ * compatible drivers/servers. */
 #include <linux/virtio_config.h>
 
 /* The ID for virtio console */
diff --git a/include/linux/virtio_balloon.h b/include/linux/virtio_balloon.h
index 979524ee75b7..c30c7bfbf39b 100644
--- a/include/linux/virtio_balloon.h
+++ b/include/linux/virtio_balloon.h
@@ -1,5 +1,7 @@
 #ifndef _LINUX_VIRTIO_BALLOON_H
 #define _LINUX_VIRTIO_BALLOON_H
+/* This header is BSD licensed so anyone can use the definitions to implement
+ * compatible drivers/servers. */
 #include <linux/virtio_config.h>
 
 /* The ID for virtio_balloon */
diff --git a/include/linux/virtio_blk.h b/include/linux/virtio_blk.h
index 5f79a5f9de79..6a66c7f30bcb 100644
--- a/include/linux/virtio_blk.h
+++ b/include/linux/virtio_blk.h
@@ -1,5 +1,7 @@
 #ifndef _LINUX_VIRTIO_BLK_H
 #define _LINUX_VIRTIO_BLK_H
+/* This header is BSD licensed so anyone can use the definitions to implement
+ * compatible drivers/servers. */
 #include <linux/virtio_config.h>
 
 /* The ID for virtio_block */
diff --git a/include/linux/virtio_config.h b/include/linux/virtio_config.h
index f364bbf63c34..7eb4b34d13bb 100644
--- a/include/linux/virtio_config.h
+++ b/include/linux/virtio_config.h
@@ -1,5 +1,8 @@
 #ifndef _LINUX_VIRTIO_CONFIG_H
 #define _LINUX_VIRTIO_CONFIG_H
+/* This header, excluding the #ifdef __KERNEL__ part, is BSD licensed so
+ * anyone can use the definitions to implement compatible drivers/servers. */
+
 /* Virtio devices use a standardized configuration space to define their
  * features and pass configuration information, but each implementation can
  * store and access that space differently. */
diff --git a/include/linux/virtio_console.h b/include/linux/virtio_console.h
index ed2d4ead7eb7..19a0da0dba41 100644
--- a/include/linux/virtio_console.h
+++ b/include/linux/virtio_console.h
@@ -1,6 +1,8 @@
 #ifndef _LINUX_VIRTIO_CONSOLE_H
 #define _LINUX_VIRTIO_CONSOLE_H
 #include <linux/virtio_config.h>
+/* This header, excluding the #ifdef __KERNEL__ part, is BSD licensed so
+ * anyone can use the definitions to implement compatible drivers/servers. */
 
 /* The ID for virtio console */
 #define VIRTIO_ID_CONSOLE	3
diff --git a/include/linux/virtio_net.h b/include/linux/virtio_net.h
index 38c0571820fb..5e33761b9b8a 100644
--- a/include/linux/virtio_net.h
+++ b/include/linux/virtio_net.h
@@ -1,5 +1,7 @@
 #ifndef _LINUX_VIRTIO_NET_H
 #define _LINUX_VIRTIO_NET_H
+/* This header is BSD licensed so anyone can use the definitions to implement
+ * compatible drivers/servers. */
 #include <linux/virtio_config.h>
 
 /* The ID for virtio_net */
diff --git a/include/linux/virtio_pci.h b/include/linux/virtio_pci.h
index b3151659cf49..cdef35742932 100644
--- a/include/linux/virtio_pci.h
+++ b/include/linux/virtio_pci.h
@@ -9,9 +9,8 @@
  * Authors:
  *  Anthony Liguori  <aliguori@us.ibm.com>
  *
- * This work is licensed under the terms of the GNU GPL, version 2 or later.
- * See the COPYING file in the top-level directory.
- *
+ * This header is BSD licensed so anyone can use the definitions to implement
+ * compatible drivers/servers.
  */
 
 #ifndef _LINUX_VIRTIO_PCI_H
diff --git a/include/linux/virtio_rng.h b/include/linux/virtio_rng.h
index 331afb6c9f62..1a85dab8a940 100644
--- a/include/linux/virtio_rng.h
+++ b/include/linux/virtio_rng.h
@@ -1,5 +1,7 @@
 #ifndef _LINUX_VIRTIO_RNG_H
 #define _LINUX_VIRTIO_RNG_H
+/* This header is BSD licensed so anyone can use the definitions to implement
+ * compatible drivers/servers. */
 #include <linux/virtio_config.h>
 
 /* The ID for virtio_rng */

From 44653eae1407f79dff6f52fcf594ae84cb165ec4 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Fri, 25 Jul 2008 12:06:04 -0500
Subject: [PATCH 0610/1171] virtio: don't always force a notification when ring
 is full

We force notification when the ring is full, even if the host has
indicated it doesn't want to know.  This seemed like a good idea at
the time: if we fill the transmit ring, we should tell the host
immediately.

Unfortunately this logic also applies to the receiving ring, which is
refilled constantly.  We should introduce real notification thesholds
to replace this logic.  Meanwhile, removing the logic altogether breaks
the heuristics which KVM uses, so we use a hack: only notify if there are
outgoing parts of the new buffer.

Here are the number of exits with lguest's crappy network implementation:
Before:
	network xmit 7859051 recv 236420
After:
	network xmit 7858610 recv 118136

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 drivers/virtio/virtio_ring.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c
index 72bf8bc09014..21d9a62767af 100644
--- a/drivers/virtio/virtio_ring.c
+++ b/drivers/virtio/virtio_ring.c
@@ -87,8 +87,11 @@ static int vring_add_buf(struct virtqueue *_vq,
 	if (vq->num_free < out + in) {
 		pr_debug("Can't add buf len %i - avail = %i\n",
 			 out + in, vq->num_free);
-		/* We notify *even if* VRING_USED_F_NO_NOTIFY is set here. */
-		vq->notify(&vq->vq);
+		/* FIXME: for historical reasons, we force a notify here if
+		 * there are outgoing parts to the buffer.  Presumably the
+		 * host should service the ring ASAP. */
+		if (out)
+			vq->notify(&vq->vq);
 		END_USE(vq);
 		return -ENOSPC;
 	}

From e962fa660d391fc9b90988e6538c94c858c099f9 Mon Sep 17 00:00:00 2001
From: Mark McLoughlin <markmc@redhat.com>
Date: Fri, 13 Jun 2008 13:46:40 +0100
Subject: [PATCH 0611/1171] virtio: Use bus_type probe and remove methods

Hook up to the probe() and remove() methods in bus_type
rather than device_driver. The latter has been preferred
since 2.6.16.

Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 drivers/virtio/virtio.c | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/virtio/virtio.c b/drivers/virtio/virtio.c
index 7084e7e146c0..fc85cba64578 100644
--- a/drivers/virtio/virtio.c
+++ b/drivers/virtio/virtio.c
@@ -71,13 +71,6 @@ static int virtio_uevent(struct device *_dv, struct kobj_uevent_env *env)
 			      dev->id.device, dev->id.vendor);
 }
 
-static struct bus_type virtio_bus = {
-	.name  = "virtio",
-	.match = virtio_dev_match,
-	.dev_attrs = virtio_dev_attrs,
-	.uevent = virtio_uevent,
-};
-
 static void add_status(struct virtio_device *dev, unsigned status)
 {
 	dev->config->set_status(dev, dev->config->get_status(dev) | status);
@@ -147,13 +140,20 @@ static int virtio_dev_remove(struct device *_d)
 	return 0;
 }
 
+static struct bus_type virtio_bus = {
+	.name  = "virtio",
+	.match = virtio_dev_match,
+	.dev_attrs = virtio_dev_attrs,
+	.uevent = virtio_uevent,
+	.probe = virtio_dev_probe,
+	.remove = virtio_dev_remove,
+};
+
 int register_virtio_driver(struct virtio_driver *driver)
 {
 	/* Catch this early. */
 	BUG_ON(driver->feature_table_size && !driver->feature_table);
 	driver->driver.bus = &virtio_bus;
-	driver->driver.probe = virtio_dev_probe;
-	driver->driver.remove = virtio_dev_remove;
 	return driver_register(&driver->driver);
 }
 EXPORT_SYMBOL_GPL(register_virtio_driver);

From 066f4d82a67f621ddd547bfa4b9c94631d8457b0 Mon Sep 17 00:00:00 2001
From: Christian Borntraeger <borntraeger@de.ibm.com>
Date: Thu, 29 May 2008 11:08:26 +0200
Subject: [PATCH 0612/1171] virtio_blk: check for hardsector size from host

Currently virtio_blk assumes a 512 byte hard sector size. This can cause
trouble / performance issues if the backing has a different block size
(like a file on an ext3 file system formatted with 4k block size or a dasd).

Lets add a feature flag that tells the guest to use a different hard sector
size than 512 byte.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 drivers/block/virtio_blk.c | 10 +++++++++-
 include/linux/virtio_blk.h |  3 +++
 2 files changed, 12 insertions(+), 1 deletion(-)

diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c
index dd7ea203f940..42251095134f 100644
--- a/drivers/block/virtio_blk.c
+++ b/drivers/block/virtio_blk.c
@@ -196,6 +196,7 @@ static int virtblk_probe(struct virtio_device *vdev)
 	int err;
 	u64 cap;
 	u32 v;
+	u32 blk_size;
 
 	if (index_to_minor(index) >= 1 << MINORBITS)
 		return -ENOSPC;
@@ -290,6 +291,13 @@ static int virtblk_probe(struct virtio_device *vdev)
 	if (!err)
 		blk_queue_max_hw_segments(vblk->disk->queue, v);
 
+	/* Host can optionally specify the block size of the device */
+	err = virtio_config_val(vdev, VIRTIO_BLK_F_BLK_SIZE,
+				offsetof(struct virtio_blk_config, blk_size),
+				&blk_size);
+	if (!err)
+		blk_queue_hardsect_size(vblk->disk->queue, blk_size);
+
 	add_disk(vblk->disk);
 	return 0;
 
@@ -330,7 +338,7 @@ static struct virtio_device_id id_table[] = {
 
 static unsigned int features[] = {
 	VIRTIO_BLK_F_BARRIER, VIRTIO_BLK_F_SEG_MAX, VIRTIO_BLK_F_SIZE_MAX,
-	VIRTIO_BLK_F_GEOMETRY, VIRTIO_BLK_F_RO,
+	VIRTIO_BLK_F_GEOMETRY, VIRTIO_BLK_F_RO, VIRTIO_BLK_F_BLK_SIZE,
 };
 
 static struct virtio_driver virtio_blk = {
diff --git a/include/linux/virtio_blk.h b/include/linux/virtio_blk.h
index 6a66c7f30bcb..c1aef85243bf 100644
--- a/include/linux/virtio_blk.h
+++ b/include/linux/virtio_blk.h
@@ -13,6 +13,7 @@
 #define VIRTIO_BLK_F_SEG_MAX	2	/* Indicates maximum # of segments */
 #define VIRTIO_BLK_F_GEOMETRY	4	/* Legacy geometry available  */
 #define VIRTIO_BLK_F_RO		5	/* Disk is read-only */
+#define VIRTIO_BLK_F_BLK_SIZE	6	/* Block size of disk is available*/
 
 struct virtio_blk_config
 {
@@ -28,6 +29,8 @@ struct virtio_blk_config
 		__u8 heads;
 		__u8 sectors;
 	} geometry;
+	/* block size of device (if VIRTIO_BLK_F_BLK_SIZE) */
+	__u32 blk_size;
 } __attribute__((packed));
 
 /* These two define direction. */

From 611e097d7707741a336a0677d9d69bec40f29f3d Mon Sep 17 00:00:00 2001
From: Christian Borntraeger <borntraeger@de.ibm.com>
Date: Fri, 20 Jun 2008 15:24:08 +0200
Subject: [PATCH 0613/1171] hvc_console: rework setup to replace irq functions
 with callbacks

This patch tries to change hvc_console to not use request_irq/free_irq if
the backend does not use irqs. This allows virtio_console to use hvc_console
without having a linker reference to request_irq/free_irq.

In addition, together with patch 2/3 it improves the performance for virtio
console input. (an earlier version of this patch was tested by Yajin on lguest)

The irq specific code is moved to hvc_irq.c and selected by the drivers that
use irqs (System p, System i, XEN).

I replaced "int irq" with the opaque "int data". The request_irq and
free_irq calls are replaced with notifier_add and notifier_del. I have also
changed the code a bit to call the notifier_add and notifier_del inside the
spinlock area as the callbacks are found via hp->ops.

Changes since last version:
o remove ifdef
o reintroduce "irq_requested" as "notified"
o cleanups, sparse..

I did not move the timer based polling into a separate polling scheme. I
played with several variants, but it seems we need to sleep/schedule in
a thread even for irq based consoles, as there are throttleing and buffer
size constraints.

I also kept hvc_struct defined in hvc_console.h so that hvc_irq.c can access
the irq_requested element.

Feedback is appreciated. virtio_console is currently the only available console
for kvm on s390. I plan to push this change as soon as all affected parties
agree on it. I would love to get test results from System p, Xen etc.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 drivers/char/Kconfig       |  5 +++
 drivers/char/Makefile      |  1 +
 drivers/char/hvc_console.c | 81 +++++++++-----------------------------
 drivers/char/hvc_console.h | 35 ++++++++++++++--
 drivers/char/hvc_irq.c     | 44 +++++++++++++++++++++
 drivers/char/hvc_iseries.c |  2 +
 drivers/char/hvc_vio.c     |  2 +
 drivers/char/hvc_xen.c     |  2 +
 8 files changed, 105 insertions(+), 67 deletions(-)
 create mode 100644 drivers/char/hvc_irq.c

diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig
index 67b07576f8bf..d825361a6baf 100644
--- a/drivers/char/Kconfig
+++ b/drivers/char/Kconfig
@@ -578,11 +578,14 @@ config HVC_DRIVER
 	  It will automatically be selected if one of the back-end console drivers
 	  is selected.
 
+config HVC_IRQ
+	bool
 
 config HVC_CONSOLE
 	bool "pSeries Hypervisor Virtual Console support"
 	depends on PPC_PSERIES
 	select HVC_DRIVER
+	select HVC_IRQ
 	help
 	  pSeries machines when partitioned support a hypervisor virtual
 	  console. This driver allows each pSeries partition to have a console
@@ -593,6 +596,7 @@ config HVC_ISERIES
 	depends on PPC_ISERIES
 	default y
 	select HVC_DRIVER
+	select HVC_IRQ
 	help
 	  iSeries machines support a hypervisor virtual console.
 
@@ -614,6 +618,7 @@ config HVC_XEN
 	bool "Xen Hypervisor Console support"
 	depends on XEN
 	select HVC_DRIVER
+	select HVC_IRQ
 	default y
 	help
 	  Xen virtual console device driver
diff --git a/drivers/char/Makefile b/drivers/char/Makefile
index 4b6e736cfa02..eb02c3506800 100644
--- a/drivers/char/Makefile
+++ b/drivers/char/Makefile
@@ -48,6 +48,7 @@ obj-$(CONFIG_HVC_ISERIES)	+= hvc_iseries.o
 obj-$(CONFIG_HVC_RTAS)		+= hvc_rtas.o
 obj-$(CONFIG_HVC_BEAT)		+= hvc_beat.o
 obj-$(CONFIG_HVC_DRIVER)	+= hvc_console.o
+obj-$(CONFIG_HVC_IRQ)		+= hvc_irq.o
 obj-$(CONFIG_HVC_XEN)		+= hvc_xen.o
 obj-$(CONFIG_VIRTIO_CONSOLE)	+= virtio_console.o
 obj-$(CONFIG_RAW_DRIVER)	+= raw.o
diff --git a/drivers/char/hvc_console.c b/drivers/char/hvc_console.c
index 2f9759d625cc..2f5b7fb67045 100644
--- a/drivers/char/hvc_console.c
+++ b/drivers/char/hvc_console.c
@@ -27,7 +27,6 @@
 #include <linux/init.h>
 #include <linux/kbd_kern.h>
 #include <linux/kernel.h>
-#include <linux/kref.h>
 #include <linux/kthread.h>
 #include <linux/list.h>
 #include <linux/module.h>
@@ -75,23 +74,6 @@ static int hvc_init(void);
 static int sysrq_pressed;
 #endif
 
-struct hvc_struct {
-	spinlock_t lock;
-	int index;
-	struct tty_struct *tty;
-	unsigned int count;
-	int do_wakeup;
-	char *outbuf;
-	int outbuf_size;
-	int n_outbuf;
-	uint32_t vtermno;
-	struct hv_ops *ops;
-	int irq_requested;
-	int irq;
-	struct list_head next;
-	struct kref kref; /* ref count & hvc_struct lifetime */
-};
-
 /* dynamic list of hvc_struct instances */
 static LIST_HEAD(hvc_structs);
 
@@ -300,26 +282,12 @@ int hvc_instantiate(uint32_t vtermno, int index, struct hv_ops *ops)
 }
 
 /* Wake the sleeping khvcd */
-static void hvc_kick(void)
+void hvc_kick(void)
 {
 	hvc_kicked = 1;
 	wake_up_process(hvc_task);
 }
 
-static int hvc_poll(struct hvc_struct *hp);
-
-/*
- * NOTE: This API isn't used if the console adapter doesn't support interrupts.
- * In this case the console is poll driven.
- */
-static irqreturn_t hvc_handle_interrupt(int irq, void *dev_instance)
-{
-	/* if hvc_poll request a repoll, then kick the hvcd thread */
-	if (hvc_poll(dev_instance))
-		hvc_kick();
-	return IRQ_HANDLED;
-}
-
 static void hvc_unthrottle(struct tty_struct *tty)
 {
 	hvc_kick();
@@ -333,7 +301,6 @@ static int hvc_open(struct tty_struct *tty, struct file * filp)
 {
 	struct hvc_struct *hp;
 	unsigned long flags;
-	int irq = 0;
 	int rc = 0;
 
 	/* Auto increments kref reference if found. */
@@ -352,18 +319,15 @@ static int hvc_open(struct tty_struct *tty, struct file * filp)
 	tty->low_latency = 1; /* Makes flushes to ldisc synchronous. */
 
 	hp->tty = tty;
-	/* Save for request_irq outside of spin_lock. */
-	irq = hp->irq;
-	if (irq)
-		hp->irq_requested = 1;
+
+	if (hp->ops->notifier_add)
+		rc = hp->ops->notifier_add(hp, hp->data);
 
 	spin_unlock_irqrestore(&hp->lock, flags);
-	/* check error, fallback to non-irq */
-	if (irq)
-		rc = request_irq(irq, hvc_handle_interrupt, IRQF_DISABLED, "hvc_console", hp);
+
 
 	/*
-	 * If the request_irq() fails and we return an error.  The tty layer
+	 * If the notifier fails we return an error.  The tty layer
 	 * will call hvc_close() after a failed open but we don't want to clean
 	 * up there so we'll clean up here and clear out the previously set
 	 * tty fields and return the kref reference.
@@ -371,7 +335,6 @@ static int hvc_open(struct tty_struct *tty, struct file * filp)
 	if (rc) {
 		spin_lock_irqsave(&hp->lock, flags);
 		hp->tty = NULL;
-		hp->irq_requested = 0;
 		spin_unlock_irqrestore(&hp->lock, flags);
 		tty->driver_data = NULL;
 		kref_put(&hp->kref, destroy_hvc_struct);
@@ -386,7 +349,6 @@ static int hvc_open(struct tty_struct *tty, struct file * filp)
 static void hvc_close(struct tty_struct *tty, struct file * filp)
 {
 	struct hvc_struct *hp;
-	int irq = 0;
 	unsigned long flags;
 
 	if (tty_hung_up_p(filp))
@@ -404,9 +366,8 @@ static void hvc_close(struct tty_struct *tty, struct file * filp)
 	spin_lock_irqsave(&hp->lock, flags);
 
 	if (--hp->count == 0) {
-		if (hp->irq_requested)
-			irq = hp->irq;
-		hp->irq_requested = 0;
+		if (hp->ops->notifier_del)
+			hp->ops->notifier_del(hp, hp->data);
 
 		/* We are done with the tty pointer now. */
 		hp->tty = NULL;
@@ -418,10 +379,6 @@ static void hvc_close(struct tty_struct *tty, struct file * filp)
 		 * waking periodically to check chars_in_buffer().
 		 */
 		tty_wait_until_sent(tty, HVC_CLOSE_WAIT);
-
-		if (irq)
-			free_irq(irq, hp);
-
 	} else {
 		if (hp->count < 0)
 			printk(KERN_ERR "hvc_close %X: oops, count is %d\n",
@@ -436,7 +393,6 @@ static void hvc_hangup(struct tty_struct *tty)
 {
 	struct hvc_struct *hp = tty->driver_data;
 	unsigned long flags;
-	int irq = 0;
 	int temp_open_count;
 
 	if (!hp)
@@ -458,13 +414,12 @@ static void hvc_hangup(struct tty_struct *tty)
 	hp->count = 0;
 	hp->n_outbuf = 0;
 	hp->tty = NULL;
-	if (hp->irq_requested)
-		/* Saved for use outside of spin_lock. */
-		irq = hp->irq;
-	hp->irq_requested = 0;
+
+	if (hp->ops->notifier_del)
+			hp->ops->notifier_del(hp, hp->data);
+
 	spin_unlock_irqrestore(&hp->lock, flags);
-	if (irq)
-		free_irq(irq, hp);
+
 	while(temp_open_count) {
 		--temp_open_count;
 		kref_put(&hp->kref, destroy_hvc_struct);
@@ -575,7 +530,7 @@ static u32 timeout = MIN_TIMEOUT;
 #define HVC_POLL_READ	0x00000001
 #define HVC_POLL_WRITE	0x00000002
 
-static int hvc_poll(struct hvc_struct *hp)
+int hvc_poll(struct hvc_struct *hp)
 {
 	struct tty_struct *tty;
 	int i, n, poll_mask = 0;
@@ -602,10 +557,10 @@ static int hvc_poll(struct hvc_struct *hp)
 	if (test_bit(TTY_THROTTLED, &tty->flags))
 		goto throttled;
 
-	/* If we aren't interrupt driven and aren't throttled, we always
+	/* If we aren't notifier driven and aren't throttled, we always
 	 * request a reschedule
 	 */
-	if (hp->irq == 0)
+	if (!hp->irq_requested)
 		poll_mask |= HVC_POLL_READ;
 
 	/* Read data if any */
@@ -733,7 +688,7 @@ static const struct tty_operations hvc_ops = {
 	.chars_in_buffer = hvc_chars_in_buffer,
 };
 
-struct hvc_struct __devinit *hvc_alloc(uint32_t vtermno, int irq,
+struct hvc_struct __devinit *hvc_alloc(uint32_t vtermno, int data,
 					struct hv_ops *ops, int outbuf_size)
 {
 	struct hvc_struct *hp;
@@ -754,7 +709,7 @@ struct hvc_struct __devinit *hvc_alloc(uint32_t vtermno, int irq,
 	memset(hp, 0x00, sizeof(*hp));
 
 	hp->vtermno = vtermno;
-	hp->irq = irq;
+	hp->data = data;
 	hp->ops = ops;
 	hp->outbuf_size = outbuf_size;
 	hp->outbuf = &((char *)hp)[ALIGN(sizeof(*hp), sizeof(long))];
diff --git a/drivers/char/hvc_console.h b/drivers/char/hvc_console.h
index 42ffb17e15df..d9ce10915625 100644
--- a/drivers/char/hvc_console.h
+++ b/drivers/char/hvc_console.h
@@ -26,6 +26,7 @@
 
 #ifndef HVC_CONSOLE_H
 #define HVC_CONSOLE_H
+#include <linux/kref.h>
 
 /*
  * This is the max number of console adapters that can/will be found as
@@ -42,24 +43,50 @@
  */
 #define HVC_ALLOC_TTY_ADAPTERS	8
 
+struct hvc_struct {
+	spinlock_t lock;
+	int index;
+	struct tty_struct *tty;
+	unsigned int count;
+	int do_wakeup;
+	char *outbuf;
+	int outbuf_size;
+	int n_outbuf;
+	uint32_t vtermno;
+	struct hv_ops *ops;
+	int irq_requested;
+	int data;
+	struct list_head next;
+	struct kref kref; /* ref count & hvc_struct lifetime */
+};
 
 /* implemented by a low level driver */
 struct hv_ops {
 	int (*get_chars)(uint32_t vtermno, char *buf, int count);
 	int (*put_chars)(uint32_t vtermno, const char *buf, int count);
-};
 
-struct hvc_struct;
+	/* Callbacks for notification. Called in open and close */
+	int (*notifier_add)(struct hvc_struct *hp, int irq);
+	void (*notifier_del)(struct hvc_struct *hp, int irq);
+};
 
 /* Register a vterm and a slot index for use as a console (console_init) */
 extern int hvc_instantiate(uint32_t vtermno, int index, struct hv_ops *ops);
 
 /* register a vterm for hvc tty operation (module_init or hotplug add) */
-extern struct hvc_struct * __devinit hvc_alloc(uint32_t vtermno, int irq,
+extern struct hvc_struct * __devinit hvc_alloc(uint32_t vtermno, int data,
 				struct hv_ops *ops, int outbuf_size);
-/* remove a vterm from hvc tty operation (modele_exit or hotplug remove) */
+/* remove a vterm from hvc tty operation (module_exit or hotplug remove) */
 extern int __devexit hvc_remove(struct hvc_struct *hp);
 
+/* data available */
+int hvc_poll(struct hvc_struct *hp);
+void hvc_kick(void);
+
+/* default notifier for irq based notification */
+extern int notifier_add_irq(struct hvc_struct *hp, int data);
+extern void notifier_del_irq(struct hvc_struct *hp, int data);
+
 
 #if defined(CONFIG_XMON) && defined(CONFIG_SMP)
 #include <asm/xmon.h>
diff --git a/drivers/char/hvc_irq.c b/drivers/char/hvc_irq.c
new file mode 100644
index 000000000000..73a59cdb8947
--- /dev/null
+++ b/drivers/char/hvc_irq.c
@@ -0,0 +1,44 @@
+/*
+ * Copyright IBM Corp. 2001,2008
+ *
+ * This file contains the IRQ specific code for hvc_console
+ *
+ */
+
+#include <linux/interrupt.h>
+
+#include "hvc_console.h"
+
+static irqreturn_t hvc_handle_interrupt(int irq, void *dev_instance)
+{
+	/* if hvc_poll request a repoll, then kick the hvcd thread */
+	if (hvc_poll(dev_instance))
+		hvc_kick();
+	return IRQ_HANDLED;
+}
+
+/*
+ * For IRQ based systems these callbacks can be used
+ */
+int notifier_add_irq(struct hvc_struct *hp, int irq)
+{
+	int rc;
+
+	if (!irq) {
+		hp->irq_requested = 0;
+		return 0;
+	}
+	rc = request_irq(irq, hvc_handle_interrupt, IRQF_DISABLED,
+			   "hvc_console", hp);
+	if (!rc)
+		hp->irq_requested = 1;
+	return rc;
+}
+
+void notifier_del_irq(struct hvc_struct *hp, int irq)
+{
+	if (!irq)
+		return;
+	free_irq(irq, hp);
+	hp->irq_requested = 0;
+}
diff --git a/drivers/char/hvc_iseries.c b/drivers/char/hvc_iseries.c
index a08f8f981c11..b71c610fe5ae 100644
--- a/drivers/char/hvc_iseries.c
+++ b/drivers/char/hvc_iseries.c
@@ -200,6 +200,8 @@ done:
 static struct hv_ops hvc_get_put_ops = {
 	.get_chars = get_chars,
 	.put_chars = put_chars,
+	.notifier_add = notifier_add_irq,
+	.notifier_del = notifier_del_irq,
 };
 
 static int __devinit hvc_vio_probe(struct vio_dev *vdev,
diff --git a/drivers/char/hvc_vio.c b/drivers/char/hvc_vio.c
index 79711aa4b41d..93f3840c1682 100644
--- a/drivers/char/hvc_vio.c
+++ b/drivers/char/hvc_vio.c
@@ -80,6 +80,8 @@ static int filtered_get_chars(uint32_t vtermno, char *buf, int count)
 static struct hv_ops hvc_get_put_ops = {
 	.get_chars = filtered_get_chars,
 	.put_chars = hvc_put_chars,
+	.notifier_add = notifier_add_irq,
+	.notifier_del = notifier_del_irq,
 };
 
 static int __devinit hvc_vio_probe(struct vio_dev *vdev,
diff --git a/drivers/char/hvc_xen.c b/drivers/char/hvc_xen.c
index db2ae4216279..6b70aa66a587 100644
--- a/drivers/char/hvc_xen.c
+++ b/drivers/char/hvc_xen.c
@@ -100,6 +100,8 @@ static int read_console(uint32_t vtermno, char *buf, int len)
 static struct hv_ops hvc_ops = {
 	.get_chars = read_console,
 	.put_chars = write_console,
+	.notifier_add = notifier_add_irq,
+	.notifier_del = notifier_del_irq,
 };
 
 static int __init xen_init(void)

From 91fcad19d03ed67cb50fd0e1913a8b89cc3ed3ec Mon Sep 17 00:00:00 2001
From: Christian Borntraeger <borntraeger@de.ibm.com>
Date: Fri, 20 Jun 2008 15:24:15 +0200
Subject: [PATCH 0614/1171] virtio_console: use virtqueue notification for
 hvc_console

This patch exploits the new notifier callbacks of the hvc_console. We can
use the virtio callbacks instead of the polling code.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 drivers/char/virtio_console.c | 40 +++++++++++++++++++++++++++++------
 1 file changed, 33 insertions(+), 7 deletions(-)

diff --git a/drivers/char/virtio_console.c b/drivers/char/virtio_console.c
index dc17fe3a88bc..d0f4eb6fdb7f 100644
--- a/drivers/char/virtio_console.c
+++ b/drivers/char/virtio_console.c
@@ -46,6 +46,9 @@ static char *in, *inbuf;
 /* The operations for our console. */
 static struct hv_ops virtio_cons;
 
+/* The hvc device */
+static struct hvc_struct *hvc;
+
 /*D:310 The put_chars() callback is pretty straightforward.
  *
  * We turn the characters into a scatter-gather list, add it to the output
@@ -134,6 +137,27 @@ int __init virtio_cons_early_init(int (*put_chars)(u32, const char *, int))
 	return hvc_instantiate(0, 0, &virtio_cons);
 }
 
+/*
+ * we support only one console, the hvc struct is a global var
+ * There is no need to do anything
+ */
+static int notifier_add_vio(struct hvc_struct *hp, int data)
+{
+	hp->irq_requested = 1;
+	return 0;
+}
+
+static void notifier_del_vio(struct hvc_struct *hp, int data)
+{
+	hp->irq_requested = 0;
+}
+
+static void hvc_handle_input(struct virtqueue *vq)
+{
+	if (hvc_poll(hvc))
+		hvc_kick();
+}
+
 /*D:370 Once we're further in boot, we get probed like any other virtio device.
  * At this stage we set up the output virtqueue.
  *
@@ -144,7 +168,6 @@ int __init virtio_cons_early_init(int (*put_chars)(u32, const char *, int))
 static int __devinit virtcons_probe(struct virtio_device *dev)
 {
 	int err;
-	struct hvc_struct *hvc;
 
 	vdev = dev;
 
@@ -158,7 +181,7 @@ static int __devinit virtcons_probe(struct virtio_device *dev)
 	/* Find the input queue. */
 	/* FIXME: This is why we want to wean off hvc: we do nothing
 	 * when input comes in. */
-	in_vq = vdev->config->find_vq(vdev, 0, NULL);
+	in_vq = vdev->config->find_vq(vdev, 0, hvc_handle_input);
 	if (IS_ERR(in_vq)) {
 		err = PTR_ERR(in_vq);
 		goto free;
@@ -173,15 +196,18 @@ static int __devinit virtcons_probe(struct virtio_device *dev)
 	/* Start using the new console output. */
 	virtio_cons.get_chars = get_chars;
 	virtio_cons.put_chars = put_chars;
+	virtio_cons.notifier_add = notifier_add_vio;
+	virtio_cons.notifier_del = notifier_del_vio;
 
 	/* The first argument of hvc_alloc() is the virtual console number, so
-	 * we use zero.  The second argument is the interrupt number; we
-	 * currently leave this as zero: it would be better not to use the
-	 * hvc mechanism and fix this (FIXME!).
+	 * we use zero.  The second argument is the parameter for the
+	 * notification mechanism (like irq number). We currently leave this
+	 * as zero, virtqueues have implicit notifications.
 	 *
 	 * The third argument is a "struct hv_ops" containing the put_chars()
-	 * and get_chars() pointers.  The final argument is the output buffer
-	 * size: we can do any size, so we put PAGE_SIZE here. */
+	 * get_chars(), notifier_add() and notifier_del() pointers.
+	 * The final argument is the output buffer size: we can do any size,
+	 * so we put PAGE_SIZE here. */
 	hvc = hvc_alloc(0, 0, &virtio_cons, PAGE_SIZE);
 	if (IS_ERR(hvc)) {
 		err = PTR_ERR(hvc);

From 7721c494a28e06543a3d6aa412957aa783a4a531 Mon Sep 17 00:00:00 2001
From: Christian Borntraeger <borntraeger@de.ibm.com>
Date: Fri, 25 Jul 2008 12:06:06 -0500
Subject: [PATCH 0615/1171] virtio: console as a config option

I also added a small Kconfig change that allows the user to specify the
virtio console in menuconfig.

(Fixes to export symbols from Stephen Rothwell <sfr@canb.auug.org.au>)
(Fixes for CONFIG_VIRTIO_CONSOLE=y vs CONFIG_VIRTIO=m from Christian himself)

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
---
 drivers/char/Kconfig       | 6 +++++-
 drivers/char/hvc_console.c | 4 ++++
 2 files changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/char/Kconfig b/drivers/char/Kconfig
index d825361a6baf..6c070dc5f2d4 100644
--- a/drivers/char/Kconfig
+++ b/drivers/char/Kconfig
@@ -624,8 +624,12 @@ config HVC_XEN
 	  Xen virtual console device driver
 
 config VIRTIO_CONSOLE
-	bool
+	tristate "Virtio console"
+	depends on VIRTIO
 	select HVC_DRIVER
+	help
+	  Virtio console for use with lguest and other hypervisors.
+
 
 config HVCS
 	tristate "IBM Hypervisor Virtual Console Server support"
diff --git a/drivers/char/hvc_console.c b/drivers/char/hvc_console.c
index 2f5b7fb67045..02aac104842d 100644
--- a/drivers/char/hvc_console.c
+++ b/drivers/char/hvc_console.c
@@ -280,6 +280,7 @@ int hvc_instantiate(uint32_t vtermno, int index, struct hv_ops *ops)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(hvc_instantiate);
 
 /* Wake the sleeping khvcd */
 void hvc_kick(void)
@@ -287,6 +288,7 @@ void hvc_kick(void)
 	hvc_kicked = 1;
 	wake_up_process(hvc_task);
 }
+EXPORT_SYMBOL_GPL(hvc_kick);
 
 static void hvc_unthrottle(struct tty_struct *tty)
 {
@@ -629,6 +631,7 @@ int hvc_poll(struct hvc_struct *hp)
 
 	return poll_mask;
 }
+EXPORT_SYMBOL_GPL(hvc_poll);
 
 /*
  * This kthread is either polling or interrupt driven.  This is determined by
@@ -739,6 +742,7 @@ struct hvc_struct __devinit *hvc_alloc(uint32_t vtermno, int data,
 
 	return hp;
 }
+EXPORT_SYMBOL_GPL(hvc_alloc);
 
 int __devexit hvc_remove(struct hvc_struct *hp)
 {

From faeba830b086bc9e58748869054e994cb09693cd Mon Sep 17 00:00:00 2001
From: Christian Borntraeger <borntraeger@de.ibm.com>
Date: Fri, 20 Jun 2008 15:24:18 +0200
Subject: [PATCH 0616/1171] s390: use virtio_console for KVM on s390

This patch enables virtio_console as the default console on kvm for
s390. We currently use the same notify hack as lguest for early
console output. I will try to address this for lguest and s390 later.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 arch/s390/Kconfig             |  1 +
 arch/s390/kernel/setup.c      |  4 +++-
 drivers/s390/kvm/kvm_virtio.c | 20 ++++++++++++++++++++
 include/asm-s390/kvm_virtio.h | 10 ++++++++++
 4 files changed, 34 insertions(+), 1 deletion(-)

diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index eb530b4128ba..2ed88122be93 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -565,6 +565,7 @@ bool "s390 guest support (EXPERIMENTAL)"
 	depends on 64BIT && EXPERIMENTAL
 	select VIRTIO
 	select VIRTIO_RING
+	select VIRTIO_CONSOLE
 	help
 	  Select this option if you want to run the kernel under s390 linux
 endmenu
diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c
index b358e18273b0..62122bad1e33 100644
--- a/arch/s390/kernel/setup.c
+++ b/arch/s390/kernel/setup.c
@@ -54,6 +54,7 @@
 #include <asm/sections.h>
 #include <asm/ebcdic.h>
 #include <asm/compat.h>
+#include <asm/kvm_virtio.h>
 
 long psw_kernel_bits	= (PSW_BASE_BITS | PSW_MASK_DAT | PSW_ASC_PRIMARY |
 			   PSW_MASK_MCHECK | PSW_DEFAULT_KEY);
@@ -766,7 +767,8 @@ setup_arch(char **cmdline_p)
 		printk("We are running under VM (64 bit mode)\n");
 	else if (MACHINE_IS_KVM) {
 		printk("We are running under KVM (64 bit mode)\n");
-		add_preferred_console("ttyS", 1, NULL);
+		add_preferred_console("hvc", 0, NULL);
+		s390_virtio_console_init();
 	} else
 		printk("We are running native (64 bit mode)\n");
 #endif /* CONFIG_64BIT */
diff --git a/drivers/s390/kvm/kvm_virtio.c b/drivers/s390/kvm/kvm_virtio.c
index 5ab34340919b..d41f234bb2c2 100644
--- a/drivers/s390/kvm/kvm_virtio.c
+++ b/drivers/s390/kvm/kvm_virtio.c
@@ -15,6 +15,7 @@
 #include <linux/err.h>
 #include <linux/virtio.h>
 #include <linux/virtio_config.h>
+#include <linux/virtio_console.h>
 #include <linux/interrupt.h>
 #include <linux/virtio_ring.h>
 #include <linux/pfn.h>
@@ -333,6 +334,25 @@ static int __init kvm_devices_init(void)
 	return 0;
 }
 
+/* code for early console output with virtio_console */
+static __init int early_put_chars(u32 vtermno, const char *buf, int count)
+{
+	char scratch[17];
+	unsigned int len = count;
+
+	if (len > sizeof(scratch) - 1)
+		len = sizeof(scratch) - 1;
+	scratch[len] = '\0';
+	memcpy(scratch, buf, len);
+	kvm_hypercall1(KVM_S390_VIRTIO_NOTIFY, __pa(scratch));
+	return len;
+}
+
+void s390_virtio_console_init(void)
+{
+	virtio_cons_early_init(early_put_chars);
+}
+
 /*
  * We do this after core stuff, but before the drivers.
  */
diff --git a/include/asm-s390/kvm_virtio.h b/include/asm-s390/kvm_virtio.h
index 5c871a990c29..146100224def 100644
--- a/include/asm-s390/kvm_virtio.h
+++ b/include/asm-s390/kvm_virtio.h
@@ -50,4 +50,14 @@ struct kvm_vqconfig {
 #define KVM_S390_VIRTIO_RESET		1
 #define KVM_S390_VIRTIO_SET_STATUS	2
 
+#ifdef __KERNEL__
+/* early virtio console setup */
+#ifdef CONFIG_VIRTIO_CONSOLE
+extern void s390_virtio_console_init(void);
+#else
+static inline void s390_virtio_console_init(void)
+{
+}
+#endif /* CONFIG_VIRTIO_CONSOLE */
+#endif /* __KERNEL__ */
 #endif

From dd7c7bc46211785a1aa7d70feb15830f62682b3c Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Fri, 25 Jul 2008 12:06:07 -0500
Subject: [PATCH 0617/1171] virtio: Formally reserve bits 28-31 to be
 'transport' features.

We assign feature bits as required, but it makes sense to reserve some
for the particular transport, rather than the particular device.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 drivers/virtio/virtio.c       | 5 +++++
 include/linux/virtio_config.h | 6 ++++++
 2 files changed, 11 insertions(+)

diff --git a/drivers/virtio/virtio.c b/drivers/virtio/virtio.c
index fc85cba64578..baf103361e3a 100644
--- a/drivers/virtio/virtio.c
+++ b/drivers/virtio/virtio.c
@@ -113,6 +113,11 @@ static int virtio_dev_probe(struct device *_d)
 			set_bit(f, dev->features);
 	}
 
+	/* Transport features are always preserved to pass to set_features. */
+	for (i = VIRTIO_TRANSPORT_F_START; i < VIRTIO_TRANSPORT_F_END; i++)
+		if (device_features & (1 << i))
+			set_bit(i, dev->features);
+
 	err = drv->probe(dev);
 	if (err)
 		add_status(dev, VIRTIO_CONFIG_S_FAILED);
diff --git a/include/linux/virtio_config.h b/include/linux/virtio_config.h
index 7eb4b34d13bb..5a30cfb7934b 100644
--- a/include/linux/virtio_config.h
+++ b/include/linux/virtio_config.h
@@ -18,6 +18,12 @@
 /* We've given up on this device. */
 #define VIRTIO_CONFIG_S_FAILED		0x80
 
+/* Some virtio feature bits (currently bits 28 through 31) are reserved for the
+ * transport being used (eg. virtio_ring), the rest are per-device feature
+ * bits. */
+#define VIRTIO_TRANSPORT_F_START	28
+#define VIRTIO_TRANSPORT_F_END		32
+
 /* Do we get callbacks when the ring is completely used, even if we've
  * suppressed them? */
 #define VIRTIO_F_NOTIFY_ON_EMPTY	24

From c624896e488ba2bff5ae497782cfb265c8b00646 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Fri, 25 Jul 2008 12:06:07 -0500
Subject: [PATCH 0618/1171] virtio: Rename set_features to finalize_features

Rather than explicitly handing the features to the lower-level, we just
hand the virtio_device and have it set the features.  This make it clear
that it has the chance to manipulate the features of the device at this
point (and that all feature negotiation is already done).

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 drivers/lguest/lguest_device.c | 11 ++++++-----
 drivers/s390/kvm/kvm_virtio.c  | 11 ++++++-----
 drivers/virtio/virtio.c        |  5 ++---
 drivers/virtio/virtio_pci.c    | 10 ++++++----
 include/linux/virtio_config.h  |  7 ++++---
 5 files changed, 24 insertions(+), 20 deletions(-)

diff --git a/drivers/lguest/lguest_device.c b/drivers/lguest/lguest_device.c
index 1a8de57289eb..54fdc2aa4806 100644
--- a/drivers/lguest/lguest_device.c
+++ b/drivers/lguest/lguest_device.c
@@ -98,16 +98,17 @@ static u32 lg_get_features(struct virtio_device *vdev)
 	return features;
 }
 
-static void lg_set_features(struct virtio_device *vdev, u32 features)
+static void lg_finalize_features(struct virtio_device *vdev)
 {
-	unsigned int i;
+	unsigned int i, bits;
 	struct lguest_device_desc *desc = to_lgdev(vdev)->desc;
 	/* Second half of bitmap is features we accept. */
 	u8 *out_features = lg_features(desc) + desc->feature_len;
 
 	memset(out_features, 0, desc->feature_len);
-	for (i = 0; i < min(desc->feature_len * 8, 32); i++) {
-		if (features & (1 << i))
+	bits = min_t(unsigned, desc->feature_len, sizeof(vdev->features)) * 8;
+	for (i = 0; i < bits; i++) {
+		if (test_bit(i, vdev->features))
 			out_features[i / 8] |= (1 << (i % 8));
 	}
 }
@@ -297,7 +298,7 @@ static void lg_del_vq(struct virtqueue *vq)
 /* The ops structure which hooks everything together. */
 static struct virtio_config_ops lguest_config_ops = {
 	.get_features = lg_get_features,
-	.set_features = lg_set_features,
+	.finalize_features = lg_finalize_features,
 	.get = lg_get,
 	.set = lg_set,
 	.get_status = lg_get_status,
diff --git a/drivers/s390/kvm/kvm_virtio.c b/drivers/s390/kvm/kvm_virtio.c
index d41f234bb2c2..5953510e7d5f 100644
--- a/drivers/s390/kvm/kvm_virtio.c
+++ b/drivers/s390/kvm/kvm_virtio.c
@@ -88,16 +88,17 @@ static u32 kvm_get_features(struct virtio_device *vdev)
 	return features;
 }
 
-static void kvm_set_features(struct virtio_device *vdev, u32 features)
+static void kvm_finalize_features(struct virtio_device *vdev)
 {
-	unsigned int i;
+	unsigned int i, bits;
 	struct kvm_device_desc *desc = to_kvmdev(vdev)->desc;
 	/* Second half of bitmap is features we accept. */
 	u8 *out_features = kvm_vq_features(desc) + desc->feature_len;
 
 	memset(out_features, 0, desc->feature_len);
-	for (i = 0; i < min(desc->feature_len * 8, 32); i++) {
-		if (features & (1 << i))
+	bits = min_t(unsigned, desc->feature_len, sizeof(vdev->features)) * 8;
+	for (i = 0; i < bits; i++) {
+		if (test_bit(i, vdev->features))
 			out_features[i / 8] |= (1 << (i % 8));
 	}
 }
@@ -223,7 +224,7 @@ static void kvm_del_vq(struct virtqueue *vq)
  */
 static struct virtio_config_ops kvm_vq_configspace_ops = {
 	.get_features = kvm_get_features,
-	.set_features = kvm_set_features,
+	.finalize_features = kvm_finalize_features,
 	.get = kvm_get,
 	.set = kvm_set,
 	.get_status = kvm_get_status,
diff --git a/drivers/virtio/virtio.c b/drivers/virtio/virtio.c
index baf103361e3a..5b78fd0aff0a 100644
--- a/drivers/virtio/virtio.c
+++ b/drivers/virtio/virtio.c
@@ -113,7 +113,7 @@ static int virtio_dev_probe(struct device *_d)
 			set_bit(f, dev->features);
 	}
 
-	/* Transport features are always preserved to pass to set_features. */
+	/* Transport features always preserved to pass to finalize_features. */
 	for (i = VIRTIO_TRANSPORT_F_START; i < VIRTIO_TRANSPORT_F_END; i++)
 		if (device_features & (1 << i))
 			set_bit(i, dev->features);
@@ -122,8 +122,7 @@ static int virtio_dev_probe(struct device *_d)
 	if (err)
 		add_status(dev, VIRTIO_CONFIG_S_FAILED);
 	else {
-		/* They should never have set feature bits beyond 32 */
-		dev->config->set_features(dev, dev->features[0]);
+		dev->config->finalize_features(dev);
 		add_status(dev, VIRTIO_CONFIG_S_DRIVER_OK);
 	}
 	return err;
diff --git a/drivers/virtio/virtio_pci.c b/drivers/virtio/virtio_pci.c
index eae7236310e4..9855975a72a3 100644
--- a/drivers/virtio/virtio_pci.c
+++ b/drivers/virtio/virtio_pci.c
@@ -94,12 +94,14 @@ static u32 vp_get_features(struct virtio_device *vdev)
 	return ioread32(vp_dev->ioaddr + VIRTIO_PCI_HOST_FEATURES);
 }
 
-/* virtio config->set_features() implementation */
-static void vp_set_features(struct virtio_device *vdev, u32 features)
+/* virtio config->finalize_features() implementation */
+static void vp_finalize_features(struct virtio_device *vdev)
 {
 	struct virtio_pci_device *vp_dev = to_vp_device(vdev);
 
-	iowrite32(features, vp_dev->ioaddr + VIRTIO_PCI_GUEST_FEATURES);
+	/* We only support 32 feature bits. */
+	BUILD_BUG_ON(ARRAY_SIZE(vdev->features) != 1);
+	iowrite32(vdev->features[0], vp_dev->ioaddr+VIRTIO_PCI_GUEST_FEATURES);
 }
 
 /* virtio config->get() implementation */
@@ -297,7 +299,7 @@ static struct virtio_config_ops virtio_pci_config_ops = {
 	.find_vq	= vp_find_vq,
 	.del_vq		= vp_del_vq,
 	.get_features	= vp_get_features,
-	.set_features	= vp_set_features,
+	.finalize_features = vp_finalize_features,
 };
 
 /* the PCI probing function */
diff --git a/include/linux/virtio_config.h b/include/linux/virtio_config.h
index 5a30cfb7934b..bf8ec283b232 100644
--- a/include/linux/virtio_config.h
+++ b/include/linux/virtio_config.h
@@ -61,9 +61,10 @@
  * @get_features: get the array of feature bits for this device.
  *	vdev: the virtio_device
  *	Returns the first 32 feature bits (all we currently need).
- * @set_features: confirm what device features we'll be using.
+ * @finalize_features: confirm what device features we'll be using.
  *	vdev: the virtio_device
- *	feature: the first 32 feature bits
+ *	This gives the final feature bits for the device: it can change
+ *	the dev->feature bits if it wants.
  */
 struct virtio_config_ops
 {
@@ -79,7 +80,7 @@ struct virtio_config_ops
 				     void (*callback)(struct virtqueue *));
 	void (*del_vq)(struct virtqueue *vq);
 	u32 (*get_features)(struct virtio_device *vdev);
-	void (*set_features)(struct virtio_device *vdev, u32 features);
+	void (*finalize_features)(struct virtio_device *vdev);
 };
 
 /* If driver didn't advertise the feature, it will never appear. */

From e34f87256794b87e7f4a8f1812538be7b7b5214c Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Fri, 25 Jul 2008 12:06:13 -0500
Subject: [PATCH 0619/1171] virtio: Add transport feature handling stub for
 virtio_ring.

To prepare for virtio_ring transport feature bits, hook in a call in
all the users to manipulate them.  This currently just clears all the
bits, since it doesn't understand any features.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
---
 drivers/lguest/lguest_device.c |  3 +++
 drivers/s390/kvm/kvm_virtio.c  |  3 +++
 drivers/virtio/virtio_pci.c    |  3 +++
 drivers/virtio/virtio_ring.c   | 16 ++++++++++++++++
 include/linux/virtio_ring.h    |  2 ++
 5 files changed, 27 insertions(+)

diff --git a/drivers/lguest/lguest_device.c b/drivers/lguest/lguest_device.c
index 54fdc2aa4806..37344aaee22f 100644
--- a/drivers/lguest/lguest_device.c
+++ b/drivers/lguest/lguest_device.c
@@ -105,6 +105,9 @@ static void lg_finalize_features(struct virtio_device *vdev)
 	/* Second half of bitmap is features we accept. */
 	u8 *out_features = lg_features(desc) + desc->feature_len;
 
+	/* Give virtio_ring a chance to accept features. */
+	vring_transport_features(vdev);
+
 	memset(out_features, 0, desc->feature_len);
 	bits = min_t(unsigned, desc->feature_len, sizeof(vdev->features)) * 8;
 	for (i = 0; i < bits; i++) {
diff --git a/drivers/s390/kvm/kvm_virtio.c b/drivers/s390/kvm/kvm_virtio.c
index 5953510e7d5f..79954bd6bfa5 100644
--- a/drivers/s390/kvm/kvm_virtio.c
+++ b/drivers/s390/kvm/kvm_virtio.c
@@ -95,6 +95,9 @@ static void kvm_finalize_features(struct virtio_device *vdev)
 	/* Second half of bitmap is features we accept. */
 	u8 *out_features = kvm_vq_features(desc) + desc->feature_len;
 
+	/* Give virtio_ring a chance to accept features. */
+	vring_transport_features(vdev);
+
 	memset(out_features, 0, desc->feature_len);
 	bits = min_t(unsigned, desc->feature_len, sizeof(vdev->features)) * 8;
 	for (i = 0; i < bits; i++) {
diff --git a/drivers/virtio/virtio_pci.c b/drivers/virtio/virtio_pci.c
index 9855975a72a3..c7dc37c7cce9 100644
--- a/drivers/virtio/virtio_pci.c
+++ b/drivers/virtio/virtio_pci.c
@@ -99,6 +99,9 @@ static void vp_finalize_features(struct virtio_device *vdev)
 {
 	struct virtio_pci_device *vp_dev = to_vp_device(vdev);
 
+	/* Give virtio_ring a chance to accept features. */
+	vring_transport_features(vdev);
+
 	/* We only support 32 feature bits. */
 	BUILD_BUG_ON(ARRAY_SIZE(vdev->features) != 1);
 	iowrite32(vdev->features[0], vp_dev->ioaddr+VIRTIO_PCI_GUEST_FEATURES);
diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c
index 21d9a62767af..6eb5303fed11 100644
--- a/drivers/virtio/virtio_ring.c
+++ b/drivers/virtio/virtio_ring.c
@@ -18,6 +18,7 @@
  */
 #include <linux/virtio.h>
 #include <linux/virtio_ring.h>
+#include <linux/virtio_config.h>
 #include <linux/device.h>
 
 #ifdef DEBUG
@@ -323,4 +324,19 @@ void vring_del_virtqueue(struct virtqueue *vq)
 }
 EXPORT_SYMBOL_GPL(vring_del_virtqueue);
 
+/* Manipulates transport-specific feature bits. */
+void vring_transport_features(struct virtio_device *vdev)
+{
+	unsigned int i;
+
+	for (i = VIRTIO_TRANSPORT_F_START; i < VIRTIO_TRANSPORT_F_END; i++) {
+		switch (i) {
+		default:
+			/* We don't understand this bit. */
+			clear_bit(i, vdev->features);
+		}
+	}
+}
+EXPORT_SYMBOL_GPL(vring_transport_features);
+
 MODULE_LICENSE("GPL");
diff --git a/include/linux/virtio_ring.h b/include/linux/virtio_ring.h
index abe481ed990e..c4a598fb3826 100644
--- a/include/linux/virtio_ring.h
+++ b/include/linux/virtio_ring.h
@@ -120,6 +120,8 @@ struct virtqueue *vring_new_virtqueue(unsigned int num,
 				      void (*notify)(struct virtqueue *vq),
 				      void (*callback)(struct virtqueue *vq));
 void vring_del_virtqueue(struct virtqueue *vq);
+/* Filter out transport-specific feature bits. */
+void vring_transport_features(struct virtio_device *vdev);
 
 irqreturn_t vring_interrupt(int irq, void *_vq);
 #endif /* __KERNEL__ */

From ed9559d38a87a44e3bda87d73a50aab92471d7dc Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Fri, 25 Jul 2008 12:11:09 +1000
Subject: [PATCH 0620/1171] Label kthread_create() with printf attribute tag.

Obvious misc patch been in my queue (& linux-next) for over a cycle.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/kthread.h | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/include/linux/kthread.h b/include/linux/kthread.h
index 00dd957e245b..aabc8a13ba71 100644
--- a/include/linux/kthread.h
+++ b/include/linux/kthread.h
@@ -6,7 +6,8 @@
 
 struct task_struct *kthread_create(int (*threadfn)(void *data),
 				   void *data,
-				   const char namefmt[], ...);
+				   const char namefmt[], ...)
+	__attribute__((format(printf, 3, 4)));
 
 /**
  * kthread_run - create and wake a thread.

From f7a6117ee59c0001d58e830a82d7e205ed602bdd Mon Sep 17 00:00:00 2001
From: Roland Dreier <rolandd@cisco.com>
Date: Thu, 24 Jul 2008 20:36:59 -0700
Subject: [PATCH 0621/1171] RDMA/ucma: BKL is not needed for ucma_open()

Remove explicit lock_kernel() calls and document why the code is safe.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/core/ucma.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/drivers/infiniband/core/ucma.c b/drivers/infiniband/core/ucma.c
index 195f97302fe5..b41dd26bbfa1 100644
--- a/drivers/infiniband/core/ucma.c
+++ b/drivers/infiniband/core/ucma.c
@@ -38,7 +38,6 @@
 #include <linux/in.h>
 #include <linux/in6.h>
 #include <linux/miscdevice.h>
-#include <linux/smp_lock.h>
 
 #include <rdma/rdma_user_cm.h>
 #include <rdma/ib_marshall.h>
@@ -1149,6 +1148,14 @@ static unsigned int ucma_poll(struct file *filp, struct poll_table_struct *wait)
 	return mask;
 }
 
+/*
+ * ucma_open() does not need the BKL:
+ *
+ *  - no global state is referred to;
+ *  - there is no ioctl method to race against;
+ *  - no further module initialization is required for open to work
+ *    after the device is registered.
+ */
 static int ucma_open(struct inode *inode, struct file *filp)
 {
 	struct ucma_file *file;
@@ -1157,7 +1164,6 @@ static int ucma_open(struct inode *inode, struct file *filp)
 	if (!file)
 		return -ENOMEM;
 
-	lock_kernel();
 	INIT_LIST_HEAD(&file->event_list);
 	INIT_LIST_HEAD(&file->ctx_list);
 	init_waitqueue_head(&file->poll_wait);
@@ -1165,7 +1171,6 @@ static int ucma_open(struct inode *inode, struct file *filp)
 
 	filp->private_data = file;
 	file->filp = filp;
-	unlock_kernel();
 	return 0;
 }
 

From 5ba18b186c979283a2bf75a28b7ea325184b0c08 Mon Sep 17 00:00:00 2001
From: Roland Dreier <rolandd@cisco.com>
Date: Thu, 24 Jul 2008 20:36:59 -0700
Subject: [PATCH 0622/1171] RDMA/ucm: BKL is not needed for ib_ucm_open()

Remove explicit cycle_kernel_lock() call and document why the code is safe.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/core/ucm.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/drivers/infiniband/core/ucm.c b/drivers/infiniband/core/ucm.c
index 9494005d1c9a..e603736682bf 100644
--- a/drivers/infiniband/core/ucm.c
+++ b/drivers/infiniband/core/ucm.c
@@ -43,7 +43,6 @@
 #include <linux/cdev.h>
 #include <linux/idr.h>
 #include <linux/mutex.h>
-#include <linux/smp_lock.h>
 
 #include <asm/uaccess.h>
 
@@ -1154,11 +1153,18 @@ static unsigned int ib_ucm_poll(struct file *filp,
 	return mask;
 }
 
+/*
+ * ib_ucm_open() does not need the BKL:
+ *
+ *  - no global state is referred to;
+ *  - there is no ioctl method to race against;
+ *  - no further module initialization is required for open to work
+ *    after the device is registered.
+ */
 static int ib_ucm_open(struct inode *inode, struct file *filp)
 {
 	struct ib_ucm_file *file;
 
-	cycle_kernel_lock();
 	file = kmalloc(sizeof(*file), GFP_KERNEL);
 	if (!file)
 		return -ENOMEM;

From 99c3a5a9e388e0ac166c617aaf02150e778d2779 Mon Sep 17 00:00:00 2001
From: Roland Dreier <rolandd@cisco.com>
Date: Thu, 24 Jul 2008 20:37:25 -0700
Subject: [PATCH 0623/1171] IPoIB/cm: Connected mode is no longer EXPERIMENTAL

Connected mode is now tested and used by lots of people.  No need to
hide it under CONFIG_EXPERIMENTAL.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/ulp/ipoib/Kconfig | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)

diff --git a/drivers/infiniband/ulp/ipoib/Kconfig b/drivers/infiniband/ulp/ipoib/Kconfig
index 691525cf394a..ccc7180818a3 100644
--- a/drivers/infiniband/ulp/ipoib/Kconfig
+++ b/drivers/infiniband/ulp/ipoib/Kconfig
@@ -11,16 +11,17 @@ config INFINIBAND_IPOIB
 
 config INFINIBAND_IPOIB_CM
 	bool "IP-over-InfiniBand Connected Mode support"
-	depends on INFINIBAND_IPOIB && EXPERIMENTAL
+	depends on INFINIBAND_IPOIB
 	default n
 	---help---
-	  This option enables experimental support for IPoIB connected mode.
-	  After enabling this option, you need to switch to connected mode through
-	  /sys/class/net/ibXXX/mode to actually create connections, and then increase
-	  the interface MTU with e.g. ifconfig ib0 mtu 65520.
+	  This option enables support for IPoIB connected mode.  After
+	  enabling this option, you need to switch to connected mode
+	  through /sys/class/net/ibXXX/mode to actually create
+	  connections, and then increase the interface MTU with
+	  e.g. ifconfig ib0 mtu 65520.
 
-	  WARNING: Enabling connected mode will trigger some
-	  packet drops for multicast and UD mode traffic from this interface,
+	  WARNING: Enabling connected mode will trigger some packet
+	  drops for multicast and UD mode traffic from this interface,
 	  unless you limit mtu for these destinations to 2044.
 
 config INFINIBAND_IPOIB_DEBUG

From 9905922446f6dc02fd4650c8f59114d6bdb5b777 Mon Sep 17 00:00:00 2001
From: Roland Dreier <rolandd@cisco.com>
Date: Thu, 24 Jul 2008 20:37:25 -0700
Subject: [PATCH 0624/1171] IPoIB: Correct help text for INFINIBAND_IPOIB_DEBUG

The help text for INFINIBAND_IPOIB_DEBUG refers to "ipoib_debugfs,"
which no longer exists.  Correct this to talk about the files under
debugfs that are really created.

Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/ulp/ipoib/Kconfig | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/infiniband/ulp/ipoib/Kconfig b/drivers/infiniband/ulp/ipoib/Kconfig
index ccc7180818a3..9d9a9dc51f18 100644
--- a/drivers/infiniband/ulp/ipoib/Kconfig
+++ b/drivers/infiniband/ulp/ipoib/Kconfig
@@ -34,9 +34,10 @@ config INFINIBAND_IPOIB_DEBUG
 	  debug_level and mcast_debug_level module parameters (which
 	  can also be set after the driver is loaded through sysfs).
 
-	  This option also creates an "ipoib_debugfs," which can be
-	  mounted to expose debugging information about IB multicast
-	  groups used by the IPoIB driver.
+	  This option also creates a directory tree under ipoib/ in
+	  debugfs, which contains files that expose debugging
+	  information about IB multicast groups used by the IPoIB
+	  driver.
 
 config INFINIBAND_IPOIB_DEBUG_DATA
 	bool "IP-over-InfiniBand data path debugging"

From 6492cdf3a24fd620660c399745b5e169a0ed27d6 Mon Sep 17 00:00:00 2001
From: Faisal Latif <flatif@neteffect.com>
Date: Thu, 24 Jul 2008 20:50:45 -0700
Subject: [PATCH 0625/1171] RDMA/nes: CM connection setup/teardown rework

Major rework of CM connection setup/teardown.  We had a number of issues
with MPI applications not starting/terminating properly over time.
With these changes we were able to run longer on larger clusters.

* Remove memory allocation from nes_connect() and nes_cm_connect().
* Fix mini_cm_dec_refcnt_listen() when destroying listener.
* Remove unnecessary code from schedule_nes_timer() and nes_cm_timer_tick().
* Functionalize mini_cm_recv_pkt() and process_packet().
* Clean up cm_node->ref_count usage.
* Reuse skbs if available.

Signed-off-by: Faisal Latif <flatif@neteffect.com>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/nes/nes.c       |    4 +-
 drivers/infiniband/hw/nes/nes_cm.c    | 2054 ++++++++++++++-----------
 drivers/infiniband/hw/nes/nes_cm.h    |   23 +-
 drivers/infiniband/hw/nes/nes_hw.c    |    9 -
 drivers/infiniband/hw/nes/nes_verbs.c |   15 -
 5 files changed, 1166 insertions(+), 939 deletions(-)

diff --git a/drivers/infiniband/hw/nes/nes.c b/drivers/infiniband/hw/nes/nes.c
index d2884e778098..b0cab64e5e3d 100644
--- a/drivers/infiniband/hw/nes/nes.c
+++ b/drivers/infiniband/hw/nes/nes.c
@@ -276,6 +276,7 @@ static void nes_cqp_rem_ref_callback(struct nes_device *nesdev, struct nes_cqp_r
 	}
 	nes_free_resource(nesadapter, nesadapter->allocated_qps, nesqp->hwqp.qp_id);
 
+	nesadapter->qp_table[nesqp->hwqp.qp_id-NES_FIRST_QPN] = NULL;
 	kfree(nesqp->allocated_buffer);
 
 }
@@ -289,7 +290,6 @@ void nes_rem_ref(struct ib_qp *ibqp)
 	struct nes_qp *nesqp;
 	struct nes_vnic *nesvnic = to_nesvnic(ibqp->device);
 	struct nes_device *nesdev = nesvnic->nesdev;
-	struct nes_adapter *nesadapter = nesdev->nesadapter;
 	struct nes_hw_cqp_wqe *cqp_wqe;
 	struct nes_cqp_request *cqp_request;
 	u32 opcode;
@@ -303,8 +303,6 @@ void nes_rem_ref(struct ib_qp *ibqp)
 	}
 
 	if (atomic_dec_and_test(&nesqp->refcount)) {
-		nesadapter->qp_table[nesqp->hwqp.qp_id-NES_FIRST_QPN] = NULL;
-
 		/* Destroy the QP */
 		cqp_request = nes_get_cqp_request(nesdev);
 		if (cqp_request == NULL) {
diff --git a/drivers/infiniband/hw/nes/nes_cm.c b/drivers/infiniband/hw/nes/nes_cm.c
index 6aa531d5276d..9f0b964b2c99 100644
--- a/drivers/infiniband/hw/nes/nes_cm.c
+++ b/drivers/infiniband/hw/nes/nes_cm.c
@@ -74,36 +74,59 @@ atomic_t cm_nodes_destroyed;
 atomic_t cm_accel_dropped_pkts;
 atomic_t cm_resets_recvd;
 
-static inline int mini_cm_accelerated(struct nes_cm_core *, struct nes_cm_node *);
+static inline int mini_cm_accelerated(struct nes_cm_core *,
+	struct nes_cm_node *);
 static struct nes_cm_listener *mini_cm_listen(struct nes_cm_core *,
-		struct nes_vnic *, struct nes_cm_info *);
-static int add_ref_cm_node(struct nes_cm_node *);
-static int rem_ref_cm_node(struct nes_cm_core *, struct nes_cm_node *);
+	struct nes_vnic *, struct nes_cm_info *);
 static int mini_cm_del_listen(struct nes_cm_core *, struct nes_cm_listener *);
-static struct sk_buff *form_cm_frame(struct sk_buff *, struct nes_cm_node *,
-				     void *, u32, void *, u32, u8);
-static struct sk_buff *get_free_pkt(struct nes_cm_node *cm_node);
-
 static struct nes_cm_node *mini_cm_connect(struct nes_cm_core *,
-					   struct nes_vnic *,
-					   struct ietf_mpa_frame *,
-					   struct nes_cm_info *);
-static int mini_cm_accept(struct nes_cm_core *, struct ietf_mpa_frame *,
-			  struct nes_cm_node *);
-static int mini_cm_reject(struct nes_cm_core *, struct ietf_mpa_frame *,
-			  struct nes_cm_node *);
+	struct nes_vnic *, u16, void *, struct nes_cm_info *);
 static int mini_cm_close(struct nes_cm_core *, struct nes_cm_node *);
-static int mini_cm_recv_pkt(struct nes_cm_core *, struct nes_vnic *,
-			    struct sk_buff *);
+static int mini_cm_accept(struct nes_cm_core *, struct ietf_mpa_frame *,
+	struct nes_cm_node *);
+static int mini_cm_reject(struct nes_cm_core *, struct ietf_mpa_frame *,
+	struct nes_cm_node *);
+static void mini_cm_recv_pkt(struct nes_cm_core *, struct nes_vnic *,
+	struct sk_buff *);
 static int mini_cm_dealloc_core(struct nes_cm_core *);
 static int mini_cm_get(struct nes_cm_core *);
 static int mini_cm_set(struct nes_cm_core *, u32, u32);
+
+static struct sk_buff *form_cm_frame(struct sk_buff *, struct nes_cm_node *,
+	void *, u32, void *, u32, u8);
+static struct sk_buff *get_free_pkt(struct nes_cm_node *cm_node);
+static int add_ref_cm_node(struct nes_cm_node *);
+static int rem_ref_cm_node(struct nes_cm_core *, struct nes_cm_node *);
+
 static int nes_cm_disconn_true(struct nes_qp *);
 static int nes_cm_post_event(struct nes_cm_event *event);
 static int nes_disconnect(struct nes_qp *nesqp, int abrupt);
 static void nes_disconnect_worker(struct work_struct *work);
-static int send_ack(struct nes_cm_node *cm_node);
+
+static int send_mpa_request(struct nes_cm_node *, struct sk_buff *);
+static int send_syn(struct nes_cm_node *, u32, struct sk_buff *);
+static int send_reset(struct nes_cm_node *, struct sk_buff *);
+static int send_ack(struct nes_cm_node *cm_node, struct sk_buff *skb);
 static int send_fin(struct nes_cm_node *cm_node, struct sk_buff *skb);
+static void process_packet(struct nes_cm_node *, struct sk_buff *,
+	struct nes_cm_core *);
+
+static void active_open_err(struct nes_cm_node *, struct sk_buff *, int);
+static void passive_open_err(struct nes_cm_node *, struct sk_buff *, int);
+static void cleanup_retrans_entry(struct nes_cm_node *);
+static void handle_rcv_mpa(struct nes_cm_node *, struct sk_buff *,
+	enum nes_cm_event_type);
+static void free_retrans_entry(struct nes_cm_node *cm_node);
+static int handle_tcp_options(struct nes_cm_node *cm_node, struct tcphdr *tcph,
+	struct sk_buff *skb, int optionsize, int passive);
+
+/* CM event handler functions */
+static void cm_event_connected(struct nes_cm_event *);
+static void cm_event_connect_error(struct nes_cm_event *);
+static void cm_event_reset(struct nes_cm_event *);
+static void cm_event_mpa_req(struct nes_cm_event *);
+
+static void print_core(struct nes_cm_core *core);
 
 /* External CM API Interface */
 /* instance of function pointers for client API */
@@ -158,11 +181,11 @@ static struct nes_cm_event *create_event(struct nes_cm_node *cm_node,
 	event->cm_info.loc_port = cm_node->loc_port;
 	event->cm_info.cm_id = cm_node->cm_id;
 
-	nes_debug(NES_DBG_CM, "Created event=%p, type=%u, dst_addr=%08x[%x],"
-			" src_addr=%08x[%x]\n",
-			event, type,
-			event->cm_info.loc_addr, event->cm_info.loc_port,
-			event->cm_info.rem_addr, event->cm_info.rem_port);
+	nes_debug(NES_DBG_CM, "cm_node=%p Created event=%p, type=%u, "
+		"dst_addr=%08x[%x], src_addr=%08x[%x]\n",
+		cm_node, event, type, event->cm_info.loc_addr,
+		event->cm_info.loc_port, event->cm_info.rem_addr,
+		event->cm_info.rem_port);
 
 	nes_cm_post_event(event);
 	return event;
@@ -172,14 +195,11 @@ static struct nes_cm_event *create_event(struct nes_cm_node *cm_node,
 /**
  * send_mpa_request
  */
-static int send_mpa_request(struct nes_cm_node *cm_node)
+static int send_mpa_request(struct nes_cm_node *cm_node, struct sk_buff *skb)
 {
-	struct sk_buff *skb;
 	int ret;
-
-	skb = get_free_pkt(cm_node);
 	if (!skb) {
-		nes_debug(NES_DBG_CM, "Failed to get a Free pkt\n");
+		nes_debug(NES_DBG_CM, "skb set to NULL\n");
 		return -1;
 	}
 
@@ -188,9 +208,8 @@ static int send_mpa_request(struct nes_cm_node *cm_node)
 			cm_node->mpa_frame_size, SET_ACK);
 
 	ret = schedule_nes_timer(cm_node, skb, NES_TIMER_TYPE_SEND, 1, 0);
-	if (ret < 0) {
+	if (ret < 0)
 		return ret;
-	}
 
 	return 0;
 }
@@ -228,47 +247,13 @@ static int parse_mpa(struct nes_cm_node *cm_node, u8 *buffer, u32 len)
 }
 
 
-/**
- * handle_exception_pkt - process an exception packet.
- * We have been in a TSA state, and we have now received SW
- * TCP/IP traffic should be a FIN request or IP pkt with options
- */
-static int handle_exception_pkt(struct nes_cm_node *cm_node, struct sk_buff *skb)
-{
-	int ret = 0;
-	struct tcphdr *tcph = tcp_hdr(skb);
-
-	/* first check to see if this a FIN pkt */
-	if (tcph->fin) {
-		/* we need to ACK the FIN request */
-		send_ack(cm_node);
-
-		/* check which side we are (client/server) and set next state accordingly */
-		if (cm_node->tcp_cntxt.client)
-			cm_node->state = NES_CM_STATE_CLOSING;
-		else {
-			/* we are the server side */
-			cm_node->state = NES_CM_STATE_CLOSE_WAIT;
-			/* since this is a self contained CM we don't wait for */
-			/* an APP to close us, just send final FIN immediately */
-			ret = send_fin(cm_node, NULL);
-			cm_node->state = NES_CM_STATE_LAST_ACK;
-		}
-	} else {
-		ret = -EINVAL;
-	}
-
-	return ret;
-}
-
-
 /**
  * form_cm_frame - get a free packet and build empty frame Use
  * node info to build.
  */
-static struct sk_buff *form_cm_frame(struct sk_buff *skb, struct nes_cm_node *cm_node,
-				     void *options, u32 optionsize, void *data,
-				     u32 datasize, u8 flags)
+static struct sk_buff *form_cm_frame(struct sk_buff *skb,
+	struct nes_cm_node *cm_node, void *options, u32 optionsize,
+	void *data, u32 datasize, u8 flags)
 {
 	struct tcphdr *tcph;
 	struct iphdr *iph;
@@ -332,10 +317,12 @@ static struct sk_buff *form_cm_frame(struct sk_buff *skb, struct nes_cm_node *cm
 		cm_node->tcp_cntxt.loc_seq_num++;
 		tcph->syn = 1;
 	} else
-		cm_node->tcp_cntxt.loc_seq_num += datasize;	/* data (no headers) */
+		cm_node->tcp_cntxt.loc_seq_num += datasize;
 
-	if (flags & SET_FIN)
+	if (flags & SET_FIN) {
+		cm_node->tcp_cntxt.loc_seq_num++;
 		tcph->fin = 1;
+	}
 
 	if (flags & SET_RST)
 		tcph->rst = 1;
@@ -389,7 +376,7 @@ int schedule_nes_timer(struct nes_cm_node *cm_node, struct sk_buff *skb,
 		int close_when_complete)
 {
 	unsigned long  flags;
-	struct nes_cm_core *cm_core;
+	struct nes_cm_core *cm_core = cm_node->cm_core;
 	struct nes_timer_entry *new_send;
 	int ret = 0;
 	u32 was_timer_set;
@@ -411,7 +398,7 @@ int schedule_nes_timer(struct nes_cm_node *cm_node, struct sk_buff *skb,
 	new_send->close_when_complete = close_when_complete;
 
 	if (type == NES_TIMER_TYPE_CLOSE) {
-		new_send->timetosend += (HZ/2);	/* TODO: decide on the correct value here */
+		new_send->timetosend += (HZ/10);
 		spin_lock_irqsave(&cm_node->recv_list_lock, flags);
 		list_add_tail(&new_send->list, &cm_node->recv_list);
 		spin_unlock_irqrestore(&cm_node->recv_list_lock, flags);
@@ -420,36 +407,28 @@ int schedule_nes_timer(struct nes_cm_node *cm_node, struct sk_buff *skb,
 	if (type == NES_TIMER_TYPE_SEND) {
 		new_send->seq_num = ntohl(tcp_hdr(skb)->seq);
 		atomic_inc(&new_send->skb->users);
+		spin_lock_irqsave(&cm_node->retrans_list_lock, flags);
+		cm_node->send_entry = new_send;
+		add_ref_cm_node(cm_node);
+		spin_unlock_irqrestore(&cm_node->retrans_list_lock, flags);
+		new_send->timetosend = jiffies + NES_RETRY_TIMEOUT;
 
 		ret = nes_nic_cm_xmit(new_send->skb, cm_node->netdev);
 		if (ret != NETDEV_TX_OK) {
-			nes_debug(NES_DBG_CM, "Error sending packet %p (jiffies = %lu)\n",
-					new_send, jiffies);
+			nes_debug(NES_DBG_CM, "Error sending packet %p "
+				"(jiffies = %lu)\n", new_send, jiffies);
 			atomic_dec(&new_send->skb->users);
 			new_send->timetosend = jiffies;
 		} else {
 			cm_packets_sent++;
 			if (!send_retrans) {
+				cleanup_retrans_entry(cm_node);
 				if (close_when_complete)
-					rem_ref_cm_node(cm_node->cm_core, cm_node);
-				dev_kfree_skb_any(new_send->skb);
-				kfree(new_send);
+					rem_ref_cm_node(cm_core, cm_node);
 				return ret;
 			}
-			new_send->timetosend = jiffies + NES_RETRY_TIMEOUT;
 		}
-		spin_lock_irqsave(&cm_node->retrans_list_lock, flags);
-		list_add_tail(&new_send->list, &cm_node->retrans_list);
-		spin_unlock_irqrestore(&cm_node->retrans_list_lock, flags);
 	}
-	if (type == NES_TIMER_TYPE_RECV) {
-		new_send->seq_num = ntohl(tcp_hdr(skb)->seq);
-		new_send->timetosend = jiffies;
-		spin_lock_irqsave(&cm_node->recv_list_lock, flags);
-		list_add_tail(&new_send->list, &cm_node->recv_list);
-		spin_unlock_irqrestore(&cm_node->recv_list_lock, flags);
-	}
-	cm_core = cm_node->cm_core;
 
 	was_timer_set = timer_pending(&cm_core->tcp_timer);
 
@@ -476,23 +455,27 @@ static void nes_cm_timer_tick(unsigned long pass)
 	struct list_head *list_node, *list_node_temp;
 	struct nes_cm_core *cm_core = g_cm_core;
 	struct nes_qp *nesqp;
-	struct sk_buff *skb;
 	u32 settimer = 0;
 	int ret = NETDEV_TX_OK;
-	int    node_done;
+	enum nes_cm_node_state last_state;
 
 	spin_lock_irqsave(&cm_core->ht_lock, flags);
 
-	list_for_each_safe(list_node, list_core_temp, &cm_core->connected_nodes) {
+	list_for_each_safe(list_node, list_core_temp,
+		&cm_core->connected_nodes) {
 		cm_node = container_of(list_node, struct nes_cm_node, list);
 		add_ref_cm_node(cm_node);
 		spin_unlock_irqrestore(&cm_core->ht_lock, flags);
 		spin_lock_irqsave(&cm_node->recv_list_lock, flags);
-		list_for_each_safe(list_core, list_node_temp, &cm_node->recv_list) {
-			recv_entry = container_of(list_core, struct nes_timer_entry, list);
-			if ((time_after(recv_entry->timetosend, jiffies)) &&
-					(recv_entry->type == NES_TIMER_TYPE_CLOSE)) {
-				if (nexttimeout > recv_entry->timetosend || !settimer) {
+		list_for_each_safe(list_core, list_node_temp,
+			&cm_node->recv_list) {
+			recv_entry = container_of(list_core,
+				struct nes_timer_entry, list);
+			if (!recv_entry)
+				break;
+			if (time_after(recv_entry->timetosend, jiffies)) {
+				if (nexttimeout > recv_entry->timetosend ||
+					!settimer) {
 					nexttimeout = recv_entry->timetosend;
 					settimer = 1;
 				}
@@ -501,157 +484,143 @@ static void nes_cm_timer_tick(unsigned long pass)
 			list_del(&recv_entry->list);
 			cm_id = cm_node->cm_id;
 			spin_unlock_irqrestore(&cm_node->recv_list_lock, flags);
-			if (recv_entry->type == NES_TIMER_TYPE_CLOSE) {
-				nesqp = (struct nes_qp *)recv_entry->skb;
-				spin_lock_irqsave(&nesqp->lock, qplockflags);
-				if (nesqp->cm_id) {
-					nes_debug(NES_DBG_CM, "QP%u: cm_id = %p, refcount = %d: "
-							"****** HIT A NES_TIMER_TYPE_CLOSE"
-							" with something to do!!! ******\n",
-							nesqp->hwqp.qp_id, cm_id,
-							atomic_read(&nesqp->refcount));
-					nesqp->hw_tcp_state = NES_AEQE_TCP_STATE_CLOSED;
-					nesqp->last_aeq = NES_AEQE_AEID_RESET_SENT;
-					nesqp->ibqp_state = IB_QPS_ERR;
-					spin_unlock_irqrestore(&nesqp->lock, qplockflags);
-					nes_cm_disconn(nesqp);
-				} else {
-					spin_unlock_irqrestore(&nesqp->lock, qplockflags);
-					nes_debug(NES_DBG_CM, "QP%u: cm_id = %p, refcount = %d:"
-							" ****** HIT A NES_TIMER_TYPE_CLOSE"
-							" with nothing to do!!! ******\n",
-							nesqp->hwqp.qp_id, cm_id,
-							atomic_read(&nesqp->refcount));
-					nes_rem_ref(&nesqp->ibqp);
-				}
-				if (cm_id)
-					cm_id->rem_ref(cm_id);
+			nesqp = (struct nes_qp *)recv_entry->skb;
+			spin_lock_irqsave(&nesqp->lock, qplockflags);
+			if (nesqp->cm_id) {
+				nes_debug(NES_DBG_CM, "QP%u: cm_id = %p, "
+					"refcount = %d: HIT A "
+					"NES_TIMER_TYPE_CLOSE with something "
+					"to do!!!\n", nesqp->hwqp.qp_id, cm_id,
+					atomic_read(&nesqp->refcount));
+				nesqp->hw_tcp_state = NES_AEQE_TCP_STATE_CLOSED;
+				nesqp->last_aeq = NES_AEQE_AEID_RESET_SENT;
+				nesqp->ibqp_state = IB_QPS_ERR;
+				spin_unlock_irqrestore(&nesqp->lock,
+					qplockflags);
+				nes_cm_disconn(nesqp);
+			} else {
+				spin_unlock_irqrestore(&nesqp->lock,
+					qplockflags);
+				nes_debug(NES_DBG_CM, "QP%u: cm_id = %p, "
+					"refcount = %d: HIT A "
+					"NES_TIMER_TYPE_CLOSE with nothing "
+					"to do!!!\n", nesqp->hwqp.qp_id, cm_id,
+					atomic_read(&nesqp->refcount));
 			}
+			if (cm_id)
+				cm_id->rem_ref(cm_id);
+
 			kfree(recv_entry);
 			spin_lock_irqsave(&cm_node->recv_list_lock, flags);
 		}
 		spin_unlock_irqrestore(&cm_node->recv_list_lock, flags);
 
 		spin_lock_irqsave(&cm_node->retrans_list_lock, flags);
-		node_done = 0;
-		list_for_each_safe(list_core, list_node_temp, &cm_node->retrans_list) {
-			if (node_done) {
-				break;
-			}
-			send_entry = container_of(list_core, struct nes_timer_entry, list);
+		do {
+			send_entry = cm_node->send_entry;
+			if (!send_entry)
+				continue;
 			if (time_after(send_entry->timetosend, jiffies)) {
 				if (cm_node->state != NES_CM_STATE_TSA) {
-					if ((nexttimeout > send_entry->timetosend) || !settimer) {
-						nexttimeout = send_entry->timetosend;
+					if ((nexttimeout >
+						send_entry->timetosend) ||
+						!settimer) {
+						nexttimeout =
+							send_entry->timetosend;
 						settimer = 1;
+						continue;
 					}
-					node_done = 1;
-					continue;
 				} else {
-					list_del(&send_entry->list);
-					skb = send_entry->skb;
-					spin_unlock_irqrestore(&cm_node->retrans_list_lock, flags);
-					dev_kfree_skb_any(skb);
-					kfree(send_entry);
-					spin_lock_irqsave(&cm_node->retrans_list_lock, flags);
+					free_retrans_entry(cm_node);
 					continue;
 				}
 			}
-			if (send_entry->type == NES_TIMER_NODE_CLEANUP) {
-				list_del(&send_entry->list);
-				spin_unlock_irqrestore(&cm_node->retrans_list_lock, flags);
-				kfree(send_entry);
-				spin_lock_irqsave(&cm_node->retrans_list_lock, flags);
-				continue;
-			}
-			if ((send_entry->seq_num < cm_node->tcp_cntxt.rem_ack_num) ||
-					(cm_node->state == NES_CM_STATE_TSA) ||
-					(cm_node->state == NES_CM_STATE_CLOSED)) {
-				skb = send_entry->skb;
-				list_del(&send_entry->list);
-				spin_unlock_irqrestore(&cm_node->retrans_list_lock, flags);
-				kfree(send_entry);
-				dev_kfree_skb_any(skb);
-				spin_lock_irqsave(&cm_node->retrans_list_lock, flags);
+
+			if ((cm_node->state == NES_CM_STATE_TSA) ||
+				(cm_node->state == NES_CM_STATE_CLOSED)) {
+				free_retrans_entry(cm_node);
 				continue;
 			}
 
-			if (!send_entry->retranscount || !send_entry->retrycount) {
+			if (!send_entry->retranscount ||
+				!send_entry->retrycount) {
 				cm_packets_dropped++;
-				skb = send_entry->skb;
-				list_del(&send_entry->list);
-				spin_unlock_irqrestore(&cm_node->retrans_list_lock, flags);
-				dev_kfree_skb_any(skb);
-				kfree(send_entry);
-				if (cm_node->state == NES_CM_STATE_SYN_RCVD) {
-					/* this node never even generated an indication up to the cm */
+				last_state = cm_node->state;
+				cm_node->state = NES_CM_STATE_CLOSED;
+				free_retrans_entry(cm_node);
+				spin_unlock_irqrestore(
+					&cm_node->retrans_list_lock, flags);
+				if (last_state == NES_CM_STATE_SYN_RCVD)
 					rem_ref_cm_node(cm_core, cm_node);
-				} else {
-					cm_node->state = NES_CM_STATE_CLOSED;
-					create_event(cm_node, NES_CM_EVENT_ABORTED);
-				}
-				spin_lock_irqsave(&cm_node->retrans_list_lock, flags);
+				else
+					create_event(cm_node,
+						NES_CM_EVENT_ABORTED);
+				spin_lock_irqsave(&cm_node->retrans_list_lock,
+					flags);
 				continue;
 			}
-			/* this seems like the correct place, but leave send entry unprotected */
-			/* spin_unlock_irqrestore(&cm_node->retrans_list_lock, flags); */
 			atomic_inc(&send_entry->skb->users);
 			cm_packets_retrans++;
-			nes_debug(NES_DBG_CM, "Retransmitting send_entry %p for node %p,"
-					" jiffies = %lu, time to send =  %lu, retranscount = %u, "
-					"send_entry->seq_num = 0x%08X, cm_node->tcp_cntxt.rem_ack_num = 0x%08X\n",
-					send_entry, cm_node, jiffies, send_entry->timetosend, send_entry->retranscount,
-					send_entry->seq_num, cm_node->tcp_cntxt.rem_ack_num);
+			nes_debug(NES_DBG_CM, "Retransmitting send_entry %p "
+				"for node %p, jiffies = %lu, time to send = "
+				"%lu, retranscount = %u, send_entry->seq_num = "
+				"0x%08X, cm_node->tcp_cntxt.rem_ack_num = "
+				"0x%08X\n", send_entry, cm_node, jiffies,
+				send_entry->timetosend,
+				send_entry->retranscount,
+				send_entry->seq_num,
+				cm_node->tcp_cntxt.rem_ack_num);
 
-			spin_unlock_irqrestore(&cm_node->retrans_list_lock, flags);
+			spin_unlock_irqrestore(&cm_node->retrans_list_lock,
+				flags);
 			ret = nes_nic_cm_xmit(send_entry->skb, cm_node->netdev);
+			spin_lock_irqsave(&cm_node->retrans_list_lock, flags);
 			if (ret != NETDEV_TX_OK) {
+				nes_debug(NES_DBG_CM, "rexmit failed for "
+					"node=%p\n", cm_node);
 				cm_packets_bounced++;
 				atomic_dec(&send_entry->skb->users);
 				send_entry->retrycount--;
 				nexttimeout = jiffies + NES_SHORT_TIME;
 				settimer = 1;
-				node_done = 1;
-				spin_lock_irqsave(&cm_node->retrans_list_lock, flags);
 				continue;
 			} else {
 				cm_packets_sent++;
 			}
-			spin_lock_irqsave(&cm_node->retrans_list_lock, flags);
-			list_del(&send_entry->list);
-			nes_debug(NES_DBG_CM, "Packet Sent: retrans count = %u, retry count = %u.\n",
-					send_entry->retranscount, send_entry->retrycount);
+			nes_debug(NES_DBG_CM, "Packet Sent: retrans count = "
+				"%u, retry count = %u.\n",
+				send_entry->retranscount,
+				send_entry->retrycount);
 			if (send_entry->send_retrans) {
 				send_entry->retranscount--;
-				send_entry->timetosend = jiffies + NES_RETRY_TIMEOUT;
-				if (nexttimeout > send_entry->timetosend || !settimer) {
+				send_entry->timetosend = jiffies +
+					NES_RETRY_TIMEOUT;
+				if (nexttimeout > send_entry->timetosend ||
+					!settimer) {
 					nexttimeout = send_entry->timetosend;
 					settimer = 1;
 				}
-				list_add(&send_entry->list, &cm_node->retrans_list);
-				continue;
 			} else {
 				int close_when_complete;
-				skb = send_entry->skb;
-				close_when_complete = send_entry->close_when_complete;
-				spin_unlock_irqrestore(&cm_node->retrans_list_lock, flags);
-				if (close_when_complete) {
-					BUG_ON(atomic_read(&cm_node->ref_count) == 1);
-					rem_ref_cm_node(cm_core, cm_node);
-				}
-				dev_kfree_skb_any(skb);
-				kfree(send_entry);
-				spin_lock_irqsave(&cm_node->retrans_list_lock, flags);
-				continue;
+				close_when_complete =
+					send_entry->close_when_complete;
+				nes_debug(NES_DBG_CM, "cm_node=%p state=%d\n",
+					cm_node, cm_node->state);
+				free_retrans_entry(cm_node);
+				if (close_when_complete)
+					rem_ref_cm_node(cm_node->cm_core,
+						cm_node);
 			}
-		}
+		} while (0);
+
 		spin_unlock_irqrestore(&cm_node->retrans_list_lock, flags);
-
-		rem_ref_cm_node(cm_core, cm_node);
-
+		rem_ref_cm_node(cm_node->cm_core, cm_node);
 		spin_lock_irqsave(&cm_core->ht_lock, flags);
-		if (ret != NETDEV_TX_OK)
+		if (ret != NETDEV_TX_OK) {
+			nes_debug(NES_DBG_CM, "rexmit failed for cm_node=%p\n",
+				cm_node);
 			break;
+		}
 	}
 	spin_unlock_irqrestore(&cm_core->ht_lock, flags);
 
@@ -667,14 +636,14 @@ static void nes_cm_timer_tick(unsigned long pass)
 /**
  * send_syn
  */
-static int send_syn(struct nes_cm_node *cm_node, u32 sendack)
+static int send_syn(struct nes_cm_node *cm_node, u32 sendack,
+	struct sk_buff *skb)
 {
 	int ret;
 	int flags = SET_SYN;
-	struct sk_buff *skb;
 	char optionsbuffer[sizeof(struct option_mss) +
-			sizeof(struct option_windowscale) +
-			sizeof(struct option_base) + 1];
+		sizeof(struct option_windowscale) + sizeof(struct option_base) +
+		TCP_OPTIONS_PADDING];
 
 	int optionssize = 0;
 	/* Sending MSS option */
@@ -695,8 +664,7 @@ static int send_syn(struct nes_cm_node *cm_node, u32 sendack)
 	options->as_windowscale.shiftcount = cm_node->tcp_cntxt.rcv_wscale;
 	optionssize += sizeof(struct option_windowscale);
 
-	if (sendack && !(NES_DRV_OPT_SUPRESS_OPTION_BC & nes_drv_opt)
-			) {
+	if (sendack && !(NES_DRV_OPT_SUPRESS_OPTION_BC & nes_drv_opt)) {
 		options = (union all_known_options *)&optionsbuffer[optionssize];
 		options->as_base.optionnum = OPTION_NUMBER_WRITE0;
 		options->as_base.length = sizeof(struct option_base);
@@ -714,7 +682,8 @@ static int send_syn(struct nes_cm_node *cm_node, u32 sendack)
 	options->as_end = OPTION_NUMBER_END;
 	optionssize += 1;
 
-	skb = get_free_pkt(cm_node);
+	if (!skb)
+		skb = get_free_pkt(cm_node);
 	if (!skb) {
 		nes_debug(NES_DBG_CM, "Failed to get a Free pkt\n");
 		return -1;
@@ -733,18 +702,18 @@ static int send_syn(struct nes_cm_node *cm_node, u32 sendack)
 /**
  * send_reset
  */
-static int send_reset(struct nes_cm_node *cm_node)
+static int send_reset(struct nes_cm_node *cm_node, struct sk_buff *skb)
 {
 	int ret;
-	struct sk_buff *skb = get_free_pkt(cm_node);
 	int flags = SET_RST | SET_ACK;
 
+	if (!skb)
+		skb = get_free_pkt(cm_node);
 	if (!skb) {
 		nes_debug(NES_DBG_CM, "Failed to get a Free pkt\n");
 		return -1;
 	}
 
-	add_ref_cm_node(cm_node);
 	form_cm_frame(skb, cm_node, NULL, 0, NULL, 0, flags);
 	ret = schedule_nes_timer(cm_node, skb, NES_TIMER_TYPE_SEND, 0, 1);
 
@@ -755,10 +724,12 @@ static int send_reset(struct nes_cm_node *cm_node)
 /**
  * send_ack
  */
-static int send_ack(struct nes_cm_node *cm_node)
+static int send_ack(struct nes_cm_node *cm_node, struct sk_buff *skb)
 {
 	int ret;
-	struct sk_buff *skb = get_free_pkt(cm_node);
+
+	if (!skb)
+		skb = get_free_pkt(cm_node);
 
 	if (!skb) {
 		nes_debug(NES_DBG_CM, "Failed to get a Free pkt\n");
@@ -922,7 +893,8 @@ static int add_hte_node(struct nes_cm_core *cm_core, struct nes_cm_node *cm_node
 	if (!cm_node || !cm_core)
 		return -EINVAL;
 
-	nes_debug(NES_DBG_CM, "Adding Node to Active Connection HT\n");
+	nes_debug(NES_DBG_CM, "Adding Node %p to Active Connection HT\n",
+		cm_node);
 
 	/* first, make an index into our hash table */
 	hashkey = make_hashkey(cm_node->loc_port, cm_node->loc_addr,
@@ -946,10 +918,35 @@ static int add_hte_node(struct nes_cm_core *cm_core, struct nes_cm_node *cm_node
  * mini_cm_dec_refcnt_listen
  */
 static int mini_cm_dec_refcnt_listen(struct nes_cm_core *cm_core,
-		struct nes_cm_listener *listener, int free_hanging_nodes)
+	struct nes_cm_listener *listener, int free_hanging_nodes)
 {
 	int ret = 1;
 	unsigned long flags;
+	struct list_head *list_pos = NULL;
+	struct list_head *list_temp = NULL;
+	struct nes_cm_node *cm_node = NULL;
+
+	nes_debug(NES_DBG_CM, "attempting listener= %p free_nodes= %d, "
+		"refcnt=%d\n", listener, free_hanging_nodes,
+		atomic_read(&listener->ref_count));
+	/* free non-accelerated child nodes for this listener */
+	if (free_hanging_nodes) {
+		spin_lock_irqsave(&cm_core->ht_lock, flags);
+		list_for_each_safe(list_pos, list_temp,
+			&g_cm_core->connected_nodes) {
+			cm_node = container_of(list_pos, struct nes_cm_node,
+				list);
+			if ((cm_node->listener == listener) &&
+				(!cm_node->accelerated)) {
+				cleanup_retrans_entry(cm_node);
+				spin_unlock_irqrestore(&cm_core->ht_lock,
+					flags);
+				send_reset(cm_node, NULL);
+				spin_lock_irqsave(&cm_core->ht_lock, flags);
+			}
+		}
+		spin_unlock_irqrestore(&cm_core->ht_lock, flags);
+	}
 	spin_lock_irqsave(&cm_core->listen_list_lock, flags);
 	if (!atomic_dec_return(&listener->ref_count)) {
 		list_del(&listener->list);
@@ -1067,18 +1064,18 @@ static struct nes_cm_node *make_cm_node(struct nes_cm_core *cm_core,
 	cm_node->loc_port = cm_info->loc_port;
 	cm_node->rem_port = cm_info->rem_port;
 	cm_node->send_write0 = send_first;
-	nes_debug(NES_DBG_CM, "Make node addresses : loc = " NIPQUAD_FMT ":%x, rem = " NIPQUAD_FMT ":%x\n",
-		  HIPQUAD(cm_node->loc_addr), cm_node->loc_port,
-		  HIPQUAD(cm_node->rem_addr), cm_node->rem_port);
+	nes_debug(NES_DBG_CM, "Make node addresses : loc = " NIPQUAD_FMT
+			":%x, rem = " NIPQUAD_FMT ":%x\n",
+			HIPQUAD(cm_node->loc_addr), cm_node->loc_port,
+			HIPQUAD(cm_node->rem_addr), cm_node->rem_port);
 	cm_node->listener = listener;
 	cm_node->netdev = nesvnic->netdev;
 	cm_node->cm_id = cm_info->cm_id;
 	memcpy(cm_node->loc_mac, nesvnic->netdev->dev_addr, ETH_ALEN);
 
-	nes_debug(NES_DBG_CM, "listener=%p, cm_id=%p\n",
-			cm_node->listener, cm_node->cm_id);
+	nes_debug(NES_DBG_CM, "listener=%p, cm_id=%p\n", cm_node->listener,
+			cm_node->cm_id);
 
-	INIT_LIST_HEAD(&cm_node->retrans_list);
 	spin_lock_init(&cm_node->retrans_list_lock);
 	INIT_LIST_HEAD(&cm_node->recv_list);
 	spin_lock_init(&cm_node->recv_list_lock);
@@ -1142,10 +1139,9 @@ static int add_ref_cm_node(struct nes_cm_node *cm_node)
  * rem_ref_cm_node - destroy an instance of a cm node
  */
 static int rem_ref_cm_node(struct nes_cm_core *cm_core,
-		struct nes_cm_node *cm_node)
+	struct nes_cm_node *cm_node)
 {
 	unsigned long flags, qplockflags;
-	struct nes_timer_entry *send_entry;
 	struct nes_timer_entry *recv_entry;
 	struct iw_cm_id *cm_id;
 	struct list_head *list_core, *list_node_temp;
@@ -1169,48 +1165,33 @@ static int rem_ref_cm_node(struct nes_cm_core *cm_core,
 		atomic_dec(&cm_node->listener->pend_accepts_cnt);
 		BUG_ON(atomic_read(&cm_node->listener->pend_accepts_cnt) < 0);
 	}
-
-	spin_lock_irqsave(&cm_node->retrans_list_lock, flags);
-	list_for_each_safe(list_core, list_node_temp, &cm_node->retrans_list) {
-		send_entry = container_of(list_core, struct nes_timer_entry, list);
-		list_del(&send_entry->list);
-		spin_unlock_irqrestore(&cm_node->retrans_list_lock, flags);
-		dev_kfree_skb_any(send_entry->skb);
-		kfree(send_entry);
-		spin_lock_irqsave(&cm_node->retrans_list_lock, flags);
-		continue;
-	}
-	spin_unlock_irqrestore(&cm_node->retrans_list_lock, flags);
-
+	BUG_ON(cm_node->send_entry);
 	spin_lock_irqsave(&cm_node->recv_list_lock, flags);
 	list_for_each_safe(list_core, list_node_temp, &cm_node->recv_list) {
-		recv_entry = container_of(list_core, struct nes_timer_entry, list);
+		recv_entry = container_of(list_core, struct nes_timer_entry,
+				list);
 		list_del(&recv_entry->list);
 		cm_id = cm_node->cm_id;
 		spin_unlock_irqrestore(&cm_node->recv_list_lock, flags);
-		if (recv_entry->type == NES_TIMER_TYPE_CLOSE) {
-			nesqp = (struct nes_qp *)recv_entry->skb;
-			spin_lock_irqsave(&nesqp->lock, qplockflags);
-			if (nesqp->cm_id) {
-				nes_debug(NES_DBG_CM, "QP%u: cm_id = %p: ****** HIT A NES_TIMER_TYPE_CLOSE"
-						" with something to do!!! ******\n",
-						nesqp->hwqp.qp_id, cm_id);
-				nesqp->hw_tcp_state = NES_AEQE_TCP_STATE_CLOSED;
-				nesqp->last_aeq = NES_AEQE_AEID_RESET_SENT;
-				nesqp->ibqp_state = IB_QPS_ERR;
-				spin_unlock_irqrestore(&nesqp->lock, qplockflags);
-				nes_cm_disconn(nesqp);
-			} else {
-				spin_unlock_irqrestore(&nesqp->lock, qplockflags);
-				nes_debug(NES_DBG_CM, "QP%u: cm_id = %p: ****** HIT A NES_TIMER_TYPE_CLOSE"
-						" with nothing to do!!! ******\n",
-						nesqp->hwqp.qp_id, cm_id);
-				nes_rem_ref(&nesqp->ibqp);
-			}
-			cm_id->rem_ref(cm_id);
-		} else if (recv_entry->type == NES_TIMER_TYPE_RECV) {
-			dev_kfree_skb_any(recv_entry->skb);
+		nesqp = (struct nes_qp *)recv_entry->skb;
+		spin_lock_irqsave(&nesqp->lock, qplockflags);
+		if (nesqp->cm_id) {
+			nes_debug(NES_DBG_CM, "QP%u: cm_id = %p: HIT A "
+				"NES_TIMER_TYPE_CLOSE with something to do!\n",
+				nesqp->hwqp.qp_id, cm_id);
+			nesqp->hw_tcp_state = NES_AEQE_TCP_STATE_CLOSED;
+			nesqp->last_aeq = NES_AEQE_AEID_RESET_SENT;
+			nesqp->ibqp_state = IB_QPS_ERR;
+			spin_unlock_irqrestore(&nesqp->lock, qplockflags);
+			nes_cm_disconn(nesqp);
+		} else {
+			spin_unlock_irqrestore(&nesqp->lock, qplockflags);
+			nes_debug(NES_DBG_CM, "QP%u: cm_id = %p: HIT A "
+				"NES_TIMER_TYPE_CLOSE with nothing to do!\n",
+				nesqp->hwqp.qp_id, cm_id);
 		}
+		cm_id->rem_ref(cm_id);
+
 		kfree(recv_entry);
 		spin_lock_irqsave(&cm_node->recv_list_lock, flags);
 	}
@@ -1221,23 +1202,31 @@ static int rem_ref_cm_node(struct nes_cm_core *cm_core,
 	} else {
 		if (cm_node->apbvt_set && cm_node->nesvnic) {
 			nes_manage_apbvt(cm_node->nesvnic, cm_node->loc_port,
-					PCI_FUNC(cm_node->nesvnic->nesdev->pcidev->devfn),
-					NES_MANAGE_APBVT_DEL);
+				PCI_FUNC(
+				cm_node->nesvnic->nesdev->pcidev->devfn),
+				NES_MANAGE_APBVT_DEL);
 		}
 	}
 
-	kfree(cm_node);
 	atomic_dec(&cm_core->node_cnt);
 	atomic_inc(&cm_nodes_destroyed);
+	nesqp = cm_node->nesqp;
+	if (nesqp) {
+		nesqp->cm_node = NULL;
+		nes_rem_ref(&nesqp->ibqp);
+		cm_node->nesqp = NULL;
+	}
 
+	cm_node->freed = 1;
+	kfree(cm_node);
 	return 0;
 }
 
-
 /**
  * process_options
  */
-static int process_options(struct nes_cm_node *cm_node, u8 *optionsloc, u32 optionsize, u32 syn_packet)
+static int process_options(struct nes_cm_node *cm_node, u8 *optionsloc,
+	u32 optionsize, u32 syn_packet)
 {
 	u32 tmp;
 	u32 offset = 0;
@@ -1247,35 +1236,37 @@ static int process_options(struct nes_cm_node *cm_node, u8 *optionsloc, u32 opti
 	while (offset < optionsize) {
 		all_options = (union all_known_options *)(optionsloc + offset);
 		switch (all_options->as_base.optionnum) {
-			case OPTION_NUMBER_END:
-				offset = optionsize;
-				break;
-			case OPTION_NUMBER_NONE:
-				offset += 1;
-				continue;
-			case OPTION_NUMBER_MSS:
-				nes_debug(NES_DBG_CM, "%s: MSS Length: %d Offset: %d Size: %d\n",
-						__func__,
-						all_options->as_mss.length, offset, optionsize);
-				got_mss_option = 1;
-				if (all_options->as_mss.length != 4) {
-					return 1;
-				} else {
-					tmp = ntohs(all_options->as_mss.mss);
-					if (tmp > 0 && tmp < cm_node->tcp_cntxt.mss)
-						cm_node->tcp_cntxt.mss = tmp;
-				}
-				break;
-			case OPTION_NUMBER_WINDOW_SCALE:
-				cm_node->tcp_cntxt.snd_wscale = all_options->as_windowscale.shiftcount;
-				break;
-			case OPTION_NUMBER_WRITE0:
-				cm_node->send_write0 = 1;
-				break;
-			default:
-				nes_debug(NES_DBG_CM, "TCP Option not understood: %x\n",
-						all_options->as_base.optionnum);
-				break;
+		case OPTION_NUMBER_END:
+			offset = optionsize;
+			break;
+		case OPTION_NUMBER_NONE:
+			offset += 1;
+			continue;
+		case OPTION_NUMBER_MSS:
+			nes_debug(NES_DBG_CM, "%s: MSS Length: %d Offset: %d "
+				"Size: %d\n", __func__,
+				all_options->as_mss.length, offset, optionsize);
+			got_mss_option = 1;
+			if (all_options->as_mss.length != 4) {
+				return 1;
+			} else {
+				tmp = ntohs(all_options->as_mss.mss);
+				if (tmp > 0 && tmp <
+					cm_node->tcp_cntxt.mss)
+					cm_node->tcp_cntxt.mss = tmp;
+			}
+			break;
+		case OPTION_NUMBER_WINDOW_SCALE:
+			cm_node->tcp_cntxt.snd_wscale =
+				all_options->as_windowscale.shiftcount;
+			break;
+		case OPTION_NUMBER_WRITE0:
+			cm_node->send_write0 = 1;
+			break;
+		default:
+			nes_debug(NES_DBG_CM, "TCP Option not understood: %x\n",
+				all_options->as_base.optionnum);
+			break;
 		}
 		offset += all_options->as_base.length;
 	}
@@ -1284,300 +1275,491 @@ static int process_options(struct nes_cm_node *cm_node, u8 *optionsloc, u32 opti
 	return 0;
 }
 
-
-/**
- * process_packet
- */
-static int process_packet(struct nes_cm_node *cm_node, struct sk_buff *skb,
-			  struct nes_cm_core *cm_core)
+static void drop_packet(struct sk_buff *skb)
 {
-	int optionsize;
-	int datasize;
-	int ret = 0;
-	struct tcphdr *tcph = tcp_hdr(skb);
-	u32 inc_sequence;
-	if (cm_node->state == NES_CM_STATE_SYN_SENT && tcph->syn) {
-		inc_sequence = ntohl(tcph->seq);
-		cm_node->tcp_cntxt.rcv_nxt = inc_sequence;
+	atomic_inc(&cm_accel_dropped_pkts);
+	dev_kfree_skb_any(skb);
+}
+
+static void handle_fin_pkt(struct nes_cm_node *cm_node, struct sk_buff *skb,
+	struct tcphdr *tcph)
+{
+	atomic_inc(&cm_resets_recvd);
+	nes_debug(NES_DBG_CM, "Received FIN, cm_node = %p, state = %u. "
+		"refcnt=%d\n", cm_node, cm_node->state,
+		atomic_read(&cm_node->ref_count));
+	cm_node->tcp_cntxt.rcv_nxt++;
+	cleanup_retrans_entry(cm_node);
+	switch (cm_node->state) {
+	case NES_CM_STATE_SYN_RCVD:
+	case NES_CM_STATE_SYN_SENT:
+	case NES_CM_STATE_ESTABLISHED:
+	case NES_CM_STATE_MPAREQ_SENT:
+		cm_node->state = NES_CM_STATE_LAST_ACK;
+		send_fin(cm_node, skb);
+		break;
+	case NES_CM_STATE_FIN_WAIT1:
+		cm_node->state = NES_CM_STATE_CLOSING;
+		send_ack(cm_node, skb);
+		break;
+	case NES_CM_STATE_FIN_WAIT2:
+		cm_node->state = NES_CM_STATE_TIME_WAIT;
+		send_ack(cm_node, skb);
+		cm_node->state = NES_CM_STATE_CLOSED;
+		break;
+	case NES_CM_STATE_TSA:
+	default:
+		nes_debug(NES_DBG_CM, "Error Rcvd FIN for node-%p state = %d\n",
+			cm_node, cm_node->state);
+		drop_packet(skb);
+		break;
 	}
+}
 
-	if ((!tcph) || (cm_node->state == NES_CM_STATE_TSA)) {
-		BUG_ON(!tcph);
-		atomic_inc(&cm_accel_dropped_pkts);
-		return -1;
+
+static void handle_rst_pkt(struct nes_cm_node *cm_node, struct sk_buff *skb,
+	struct tcphdr *tcph)
+{
+
+	int	reset = 0;	/* whether to send reset in case of err.. */
+	atomic_inc(&cm_resets_recvd);
+	nes_debug(NES_DBG_CM, "Received Reset, cm_node = %p, state = %u."
+			" refcnt=%d\n", cm_node, cm_node->state,
+			atomic_read(&cm_node->ref_count));
+	cleanup_retrans_entry(cm_node);
+	switch (cm_node->state) {
+	case NES_CM_STATE_SYN_SENT:
+	case NES_CM_STATE_MPAREQ_SENT:
+		nes_debug(NES_DBG_CM, "%s[%u] create abort for cm_node=%p "
+			"listener=%p state=%d\n", __func__, __LINE__, cm_node,
+			cm_node->listener, cm_node->state);
+		active_open_err(cm_node, skb, reset);
+		break;
+	/* For PASSIVE open states, remove the cm_node event */
+	case NES_CM_STATE_ESTABLISHED:
+	case NES_CM_STATE_SYN_RCVD:
+	case NES_CM_STATE_LISTENING:
+		nes_debug(NES_DBG_CM, "Bad state %s[%u]\n", __func__, __LINE__);
+		passive_open_err(cm_node, skb, reset);
+		break;
+	case NES_CM_STATE_TSA:
+	default:
+		break;
 	}
+}
 
-	if (tcph->rst) {
-		atomic_inc(&cm_resets_recvd);
-		nes_debug(NES_DBG_CM, "Received Reset, cm_node = %p, state = %u. refcnt=%d\n",
-				cm_node, cm_node->state, atomic_read(&cm_node->ref_count));
-		switch (cm_node->state) {
-			case NES_CM_STATE_LISTENING:
-				rem_ref_cm_node(cm_core, cm_node);
-				break;
-			case NES_CM_STATE_TSA:
-			case NES_CM_STATE_CLOSED:
-				break;
-			case NES_CM_STATE_SYN_RCVD:
-					nes_debug(NES_DBG_CM, "Received a reset for local 0x%08X:%04X,"
-							" remote 0x%08X:%04X, node state = %u\n",
-							cm_node->loc_addr, cm_node->loc_port,
-							cm_node->rem_addr, cm_node->rem_port,
-							cm_node->state);
-				rem_ref_cm_node(cm_core, cm_node);
-				break;
-			case NES_CM_STATE_ONE_SIDE_ESTABLISHED:
-			case NES_CM_STATE_ESTABLISHED:
-			case NES_CM_STATE_MPAREQ_SENT:
-			default:
-					nes_debug(NES_DBG_CM, "Received a reset for local 0x%08X:%04X,"
-							" remote 0x%08X:%04X, node state = %u refcnt=%d\n",
-							cm_node->loc_addr, cm_node->loc_port,
-							cm_node->rem_addr, cm_node->rem_port,
-							cm_node->state, atomic_read(&cm_node->ref_count));
-				/* create event */
-				cm_node->state = NES_CM_STATE_CLOSED;
-
-				create_event(cm_node, NES_CM_EVENT_ABORTED);
-				break;
+static void handle_rcv_mpa(struct nes_cm_node *cm_node, struct sk_buff *skb,
+	enum nes_cm_event_type type)
+{
 
+	int	ret;
+	int datasize = skb->len;
+	u8 *dataloc = skb->data;
+	ret = parse_mpa(cm_node, dataloc, datasize);
+	if (ret < 0) {
+		nes_debug(NES_DBG_CM, "didn't like MPA Request\n");
+		if (type == NES_CM_EVENT_CONNECTED) {
+			nes_debug(NES_DBG_CM, "%s[%u] create abort for "
+				"cm_node=%p listener=%p state=%d\n", __func__,
+				__LINE__, cm_node, cm_node->listener,
+				cm_node->state);
+			active_open_err(cm_node, skb, 1);
+		} else {
+			passive_open_err(cm_node, skb, 1);
 		}
-		return -1;
+	} else {
+		cleanup_retrans_entry(cm_node);
+		dev_kfree_skb_any(skb);
+		if (type == NES_CM_EVENT_CONNECTED)
+			cm_node->state = NES_CM_STATE_TSA;
+		create_event(cm_node, type);
+
 	}
+	return ;
+}
+
+static void indicate_pkt_err(struct nes_cm_node *cm_node, struct sk_buff *skb)
+{
+	switch (cm_node->state) {
+	case NES_CM_STATE_SYN_SENT:
+	case NES_CM_STATE_MPAREQ_SENT:
+		nes_debug(NES_DBG_CM, "%s[%u] create abort for cm_node=%p "
+			"listener=%p state=%d\n", __func__, __LINE__, cm_node,
+			cm_node->listener, cm_node->state);
+		active_open_err(cm_node, skb, 1);
+		break;
+	case NES_CM_STATE_ESTABLISHED:
+	case NES_CM_STATE_SYN_RCVD:
+		passive_open_err(cm_node, skb, 1);
+		break;
+	case NES_CM_STATE_TSA:
+	default:
+		drop_packet(skb);
+	}
+}
+
+static int check_syn(struct nes_cm_node *cm_node, struct tcphdr *tcph,
+	struct sk_buff *skb)
+{
+	int err;
+
+	err = ((ntohl(tcph->ack_seq) == cm_node->tcp_cntxt.loc_seq_num))? 0 : 1;
+	if (err)
+		active_open_err(cm_node, skb, 1);
+
+	return err;
+}
+
+static int check_seq(struct nes_cm_node *cm_node, struct tcphdr *tcph,
+	struct sk_buff *skb)
+{
+	int err = 0;
+	u32 seq;
+	u32 ack_seq;
+	u32 loc_seq_num = cm_node->tcp_cntxt.loc_seq_num;
+	u32 rcv_nxt = cm_node->tcp_cntxt.rcv_nxt;
+	u32 rcv_wnd;
+	seq = ntohl(tcph->seq);
+	ack_seq = ntohl(tcph->ack_seq);
+	rcv_wnd = cm_node->tcp_cntxt.rcv_wnd;
+	if (ack_seq != loc_seq_num)
+		err = 1;
+	else if ((seq + rcv_wnd) < rcv_nxt)
+		err = 1;
+	if (err) {
+		nes_debug(NES_DBG_CM, "%s[%u] create abort for cm_node=%p "
+			"listener=%p state=%d\n", __func__, __LINE__, cm_node,
+			cm_node->listener, cm_node->state);
+		indicate_pkt_err(cm_node, skb);
+		nes_debug(NES_DBG_CM, "seq ERROR cm_node =%p seq=0x%08X "
+			"rcv_nxt=0x%08X rcv_wnd=0x%x\n", cm_node, seq, rcv_nxt,
+			rcv_wnd);
+	}
+	return err;
+}
+
+/*
+ * handle_syn_pkt() is for Passive node. The syn packet is received when a node
+ * is created with a listener or it may comein as rexmitted packet which in
+ * that case will be just dropped.
+ */
+
+static void handle_syn_pkt(struct nes_cm_node *cm_node, struct sk_buff *skb,
+	struct tcphdr *tcph)
+{
+	int ret;
+	u32 inc_sequence;
+	int optionsize;
 
 	optionsize = (tcph->doff << 2) - sizeof(struct tcphdr);
-
-	skb_pull(skb, ip_hdr(skb)->ihl << 2);
 	skb_pull(skb, tcph->doff << 2);
-
-	datasize = skb->len;
 	inc_sequence = ntohl(tcph->seq);
-	nes_debug(NES_DBG_CM, "datasize = %u, sequence = 0x%08X, ack_seq = 0x%08X,"
-			" rcv_nxt = 0x%08X Flags: %s %s.\n",
-			datasize, inc_sequence, ntohl(tcph->ack_seq),
-			cm_node->tcp_cntxt.rcv_nxt, (tcph->syn ? "SYN":""),
-			(tcph->ack ? "ACK":""));
 
-	if (!tcph->syn && (inc_sequence != cm_node->tcp_cntxt.rcv_nxt)
-		) {
-		nes_debug(NES_DBG_CM, "dropping packet, datasize = %u, sequence = 0x%08X,"
-				" ack_seq = 0x%08X, rcv_nxt = 0x%08X Flags: %s.\n",
-				datasize, inc_sequence, ntohl(tcph->ack_seq),
-				cm_node->tcp_cntxt.rcv_nxt, (tcph->ack ? "ACK":""));
-		if (cm_node->state == NES_CM_STATE_LISTENING) {
-			rem_ref_cm_node(cm_core, cm_node);
+	switch (cm_node->state) {
+	case NES_CM_STATE_SYN_SENT:
+	case NES_CM_STATE_MPAREQ_SENT:
+		/* Rcvd syn on active open connection*/
+		active_open_err(cm_node, skb, 1);
+		break;
+	case NES_CM_STATE_LISTENING:
+		/* Passive OPEN */
+		cm_node->accept_pend = 1;
+		atomic_inc(&cm_node->listener->pend_accepts_cnt);
+		if (atomic_read(&cm_node->listener->pend_accepts_cnt) >
+				cm_node->listener->backlog) {
+			nes_debug(NES_DBG_CM, "drop syn due to backlog "
+				"pressure \n");
+			cm_backlog_drops++;
+			passive_open_err(cm_node, skb, 0);
+			break;
 		}
-		return -1;
+		ret = handle_tcp_options(cm_node, tcph, skb, optionsize,
+			1);
+		if (ret) {
+			passive_open_err(cm_node, skb, 0);
+			/* drop pkt */
+			break;
+		}
+		cm_node->tcp_cntxt.rcv_nxt = inc_sequence + 1;
+		BUG_ON(cm_node->send_entry);
+		cm_node->state = NES_CM_STATE_SYN_RCVD;
+		send_syn(cm_node, 1, skb);
+		break;
+	case NES_CM_STATE_TSA:
+	case NES_CM_STATE_ESTABLISHED:
+	case NES_CM_STATE_FIN_WAIT1:
+	case NES_CM_STATE_FIN_WAIT2:
+	case NES_CM_STATE_MPAREQ_RCVD:
+	case NES_CM_STATE_LAST_ACK:
+	case NES_CM_STATE_CLOSING:
+	case NES_CM_STATE_UNKNOWN:
+	case NES_CM_STATE_CLOSED:
+	default:
+		drop_packet(skb);
+		break;
 	}
+}
 
-		cm_node->tcp_cntxt.rcv_nxt = inc_sequence + datasize;
+static void handle_synack_pkt(struct nes_cm_node *cm_node, struct sk_buff *skb,
+	struct tcphdr *tcph)
+{
 
+	int ret;
+	u32 inc_sequence;
+	int optionsize;
 
-	if (optionsize) {
-		u8 *optionsloc = (u8 *)&tcph[1];
-		if (process_options(cm_node, optionsloc, optionsize, (u32)tcph->syn)) {
-			nes_debug(NES_DBG_CM, "%s: Node %p, Sending RESET\n", __func__, cm_node);
-			send_reset(cm_node);
-			if (cm_node->state != NES_CM_STATE_SYN_SENT)
-			rem_ref_cm_node(cm_core, cm_node);
-			return 0;
+	optionsize = (tcph->doff << 2) - sizeof(struct tcphdr);
+	skb_pull(skb, tcph->doff << 2);
+	inc_sequence = ntohl(tcph->seq);
+	switch (cm_node->state) {
+	case NES_CM_STATE_SYN_SENT:
+		/* active open */
+		if (check_syn(cm_node, tcph, skb))
+			return;
+		cm_node->tcp_cntxt.rem_ack_num = ntohl(tcph->ack_seq);
+		/* setup options */
+		ret = handle_tcp_options(cm_node, tcph, skb, optionsize, 0);
+		if (ret) {
+			nes_debug(NES_DBG_CM, "cm_node=%p tcp_options failed\n",
+				cm_node);
+			break;
 		}
-	} else if (tcph->syn)
-		cm_node->tcp_cntxt.mss = NES_CM_DEFAULT_MSS;
+		cleanup_retrans_entry(cm_node);
+		cm_node->tcp_cntxt.rcv_nxt = inc_sequence + 1;
+		send_mpa_request(cm_node, skb);
+		cm_node->state = NES_CM_STATE_MPAREQ_SENT;
+		break;
+	case NES_CM_STATE_MPAREQ_RCVD:
+		/* passive open, so should not be here */
+		passive_open_err(cm_node, skb, 1);
+		break;
+	case NES_CM_STATE_ESTABLISHED:
+	case NES_CM_STATE_FIN_WAIT1:
+	case NES_CM_STATE_FIN_WAIT2:
+	case NES_CM_STATE_LAST_ACK:
+	case NES_CM_STATE_TSA:
+	case NES_CM_STATE_CLOSING:
+	case NES_CM_STATE_UNKNOWN:
+	case NES_CM_STATE_CLOSED:
+	case NES_CM_STATE_MPAREQ_SENT:
+	default:
+		drop_packet(skb);
+		break;
+	}
+}
+
+static void handle_ack_pkt(struct nes_cm_node *cm_node, struct sk_buff *skb,
+	struct tcphdr *tcph)
+{
+	int datasize = 0;
+	u32 inc_sequence;
+	u32 rem_seq_ack;
+	u32 rem_seq;
+	if (check_seq(cm_node, tcph, skb))
+		return;
+
+	skb_pull(skb, tcph->doff << 2);
+	inc_sequence = ntohl(tcph->seq);
+	rem_seq = ntohl(tcph->seq);
+	rem_seq_ack =  ntohl(tcph->ack_seq);
+	datasize = skb->len;
+
+	switch (cm_node->state) {
+	case NES_CM_STATE_SYN_RCVD:
+		/* Passive OPEN */
+		cm_node->tcp_cntxt.rem_ack_num = ntohl(tcph->ack_seq);
+		cm_node->state = NES_CM_STATE_ESTABLISHED;
+		if (datasize) {
+			cm_node->tcp_cntxt.rcv_nxt = inc_sequence + datasize;
+			cm_node->state = NES_CM_STATE_MPAREQ_RCVD;
+			handle_rcv_mpa(cm_node, skb, NES_CM_EVENT_MPA_REQ);
+		 } else { /* rcvd ACK only */
+			dev_kfree_skb_any(skb);
+			cleanup_retrans_entry(cm_node);
+		 }
+		break;
+	case NES_CM_STATE_ESTABLISHED:
+		/* Passive OPEN */
+		/* We expect mpa frame to be received only */
+		if (datasize) {
+			cm_node->tcp_cntxt.rcv_nxt = inc_sequence + datasize;
+			cm_node->state = NES_CM_STATE_MPAREQ_RCVD;
+			handle_rcv_mpa(cm_node, skb,
+				NES_CM_EVENT_MPA_REQ);
+		} else
+			drop_packet(skb);
+		break;
+	case NES_CM_STATE_MPAREQ_SENT:
+		cm_node->tcp_cntxt.rem_ack_num = ntohl(tcph->ack_seq);
+		if (datasize) {
+			cm_node->tcp_cntxt.rcv_nxt = inc_sequence + datasize;
+			handle_rcv_mpa(cm_node, skb, NES_CM_EVENT_CONNECTED);
+		} else { /* Could be just an ack pkt.. */
+			cleanup_retrans_entry(cm_node);
+			dev_kfree_skb_any(skb);
+		}
+		break;
+	case NES_CM_STATE_FIN_WAIT1:
+	case NES_CM_STATE_SYN_SENT:
+	case NES_CM_STATE_FIN_WAIT2:
+	case NES_CM_STATE_TSA:
+	case NES_CM_STATE_CLOSED:
+	case NES_CM_STATE_MPAREQ_RCVD:
+	case NES_CM_STATE_LAST_ACK:
+	case NES_CM_STATE_CLOSING:
+	case NES_CM_STATE_UNKNOWN:
+	default:
+		drop_packet(skb);
+		break;
+	}
+}
+
+
+
+static int handle_tcp_options(struct nes_cm_node *cm_node, struct tcphdr *tcph,
+	struct sk_buff *skb, int optionsize, int passive)
+{
+	u8 *optionsloc = (u8 *)&tcph[1];
+	if (optionsize) {
+		if (process_options(cm_node, optionsloc, optionsize,
+			(u32)tcph->syn)) {
+			nes_debug(NES_DBG_CM, "%s: Node %p, Sending RESET\n",
+				__func__, cm_node);
+			if (passive)
+				passive_open_err(cm_node, skb, 0);
+			else
+				active_open_err(cm_node, skb, 0);
+			return 1;
+		}
+	}
 
 	cm_node->tcp_cntxt.snd_wnd = ntohs(tcph->window) <<
 			cm_node->tcp_cntxt.snd_wscale;
 
-	if (cm_node->tcp_cntxt.snd_wnd > cm_node->tcp_cntxt.max_snd_wnd) {
+	if (cm_node->tcp_cntxt.snd_wnd > cm_node->tcp_cntxt.max_snd_wnd)
 		cm_node->tcp_cntxt.max_snd_wnd = cm_node->tcp_cntxt.snd_wnd;
-	}
-
-	if (tcph->ack) {
-		cm_node->tcp_cntxt.rem_ack_num = ntohl(tcph->ack_seq);
-		switch (cm_node->state) {
-			case NES_CM_STATE_SYN_RCVD:
-			case NES_CM_STATE_SYN_SENT:
-				/* read and stash current sequence number */
-				if (cm_node->tcp_cntxt.rem_ack_num != cm_node->tcp_cntxt.loc_seq_num) {
-					nes_debug(NES_DBG_CM, "ERROR - cm_node->tcp_cntxt.rem_ack_num !="
-							" cm_node->tcp_cntxt.loc_seq_num\n");
-					send_reset(cm_node);
-					return 0;
-				}
-				if (cm_node->state == NES_CM_STATE_SYN_SENT)
-					cm_node->state = NES_CM_STATE_ONE_SIDE_ESTABLISHED;
-				else {
-						cm_node->state = NES_CM_STATE_ESTABLISHED;
-				}
-				break;
-			case NES_CM_STATE_LAST_ACK:
-				cm_node->state = NES_CM_STATE_CLOSED;
-				break;
-			case NES_CM_STATE_FIN_WAIT1:
-				cm_node->state = NES_CM_STATE_FIN_WAIT2;
-				break;
-			case NES_CM_STATE_CLOSING:
-				cm_node->state = NES_CM_STATE_TIME_WAIT;
-				/* need to schedule this to happen in 2MSL timeouts */
-				cm_node->state = NES_CM_STATE_CLOSED;
-				break;
-			case NES_CM_STATE_ONE_SIDE_ESTABLISHED:
-			case NES_CM_STATE_ESTABLISHED:
-			case NES_CM_STATE_MPAREQ_SENT:
-			case NES_CM_STATE_CLOSE_WAIT:
-			case NES_CM_STATE_TIME_WAIT:
-			case NES_CM_STATE_CLOSED:
-				break;
-			case NES_CM_STATE_LISTENING:
-				nes_debug(NES_DBG_CM, "Received an ACK on a listening port (SYN %d)\n", tcph->syn);
-				cm_node->tcp_cntxt.loc_seq_num = ntohl(tcph->ack_seq);
-				send_reset(cm_node);
-				/* send_reset bumps refcount, this should have been a new node */
-				rem_ref_cm_node(cm_core, cm_node);
-				return -1;
-				break;
-			case NES_CM_STATE_TSA:
-				nes_debug(NES_DBG_CM, "Received a packet with the ack bit set while in TSA state\n");
-				break;
-			case NES_CM_STATE_UNKNOWN:
-			case NES_CM_STATE_INITED:
-			case NES_CM_STATE_ACCEPTING:
-			case NES_CM_STATE_FIN_WAIT2:
-			default:
-				nes_debug(NES_DBG_CM, "Received ack from unknown state: %x\n",
-						cm_node->state);
-				send_reset(cm_node);
-				break;
-		}
-	}
-
-	if (tcph->syn) {
-		if (cm_node->state == NES_CM_STATE_LISTENING) {
-			/* do not exceed backlog */
-			atomic_inc(&cm_node->listener->pend_accepts_cnt);
-			if (atomic_read(&cm_node->listener->pend_accepts_cnt) >
-					cm_node->listener->backlog) {
-				nes_debug(NES_DBG_CM, "drop syn due to backlog pressure \n");
-				cm_backlog_drops++;
-				atomic_dec(&cm_node->listener->pend_accepts_cnt);
-				rem_ref_cm_node(cm_core, cm_node);
-				return 0;
-			}
-			cm_node->accept_pend = 1;
-
-		}
-		if (datasize == 0)
-			cm_node->tcp_cntxt.rcv_nxt ++;
-
-		if (cm_node->state == NES_CM_STATE_LISTENING) {
-			cm_node->state = NES_CM_STATE_SYN_RCVD;
-			send_syn(cm_node, 1);
-		}
-		if (cm_node->state == NES_CM_STATE_ONE_SIDE_ESTABLISHED) {
-			cm_node->state = NES_CM_STATE_ESTABLISHED;
-			/* send final handshake ACK */
-			ret = send_ack(cm_node);
-			if (ret < 0)
-				return ret;
-
-				cm_node->state = NES_CM_STATE_MPAREQ_SENT;
-				ret = send_mpa_request(cm_node);
-				if (ret < 0)
-					return ret;
-		}
-	}
-
-	if (tcph->fin) {
-		cm_node->tcp_cntxt.rcv_nxt++;
-		switch (cm_node->state) {
-			case NES_CM_STATE_SYN_RCVD:
-			case NES_CM_STATE_SYN_SENT:
-			case NES_CM_STATE_ONE_SIDE_ESTABLISHED:
-			case NES_CM_STATE_ESTABLISHED:
-			case NES_CM_STATE_ACCEPTING:
-			case NES_CM_STATE_MPAREQ_SENT:
-				cm_node->state = NES_CM_STATE_CLOSE_WAIT;
-				cm_node->state = NES_CM_STATE_LAST_ACK;
-				ret = send_fin(cm_node, NULL);
-				break;
-			case NES_CM_STATE_FIN_WAIT1:
-				cm_node->state = NES_CM_STATE_CLOSING;
-				ret = send_ack(cm_node);
-				break;
-			case NES_CM_STATE_FIN_WAIT2:
-				cm_node->state = NES_CM_STATE_TIME_WAIT;
-				cm_node->tcp_cntxt.loc_seq_num ++;
-				ret = send_ack(cm_node);
-				/* need to schedule this to happen in 2MSL timeouts */
-				cm_node->state = NES_CM_STATE_CLOSED;
-				break;
-			case NES_CM_STATE_CLOSE_WAIT:
-			case NES_CM_STATE_LAST_ACK:
-			case NES_CM_STATE_CLOSING:
-			case NES_CM_STATE_TSA:
-			default:
-				nes_debug(NES_DBG_CM, "Received a fin while in %x state\n",
-						cm_node->state);
-				ret = -EINVAL;
-				break;
-		}
-	}
-
-	if (datasize) {
-		u8 *dataloc = skb->data;
-		/* figure out what state we are in and handle transition to next state */
-		switch (cm_node->state) {
-			case NES_CM_STATE_LISTENING:
-			case NES_CM_STATE_SYN_RCVD:
-			case NES_CM_STATE_SYN_SENT:
-			case NES_CM_STATE_FIN_WAIT1:
-			case NES_CM_STATE_FIN_WAIT2:
-			case NES_CM_STATE_CLOSE_WAIT:
-			case NES_CM_STATE_LAST_ACK:
-			case NES_CM_STATE_CLOSING:
-				break;
-			case  NES_CM_STATE_MPAREQ_SENT:
-				/* recv the mpa res frame, ret=frame len (incl priv data) */
-				ret = parse_mpa(cm_node, dataloc, datasize);
-				if (ret < 0)
-					break;
-				/* set the req frame payload len in skb */
-				/* we are done handling this state, set node to a TSA state */
-				cm_node->state = NES_CM_STATE_TSA;
-				send_ack(cm_node);
-				create_event(cm_node, NES_CM_EVENT_CONNECTED);
-				break;
-
-			case  NES_CM_STATE_ESTABLISHED:
-				/* we are expecting an MPA req frame */
-				ret = parse_mpa(cm_node, dataloc, datasize);
-				if (ret < 0) {
-					break;
-				}
-				cm_node->state = NES_CM_STATE_TSA;
-				send_ack(cm_node);
-				/* we got a valid MPA request, create an event */
-				create_event(cm_node, NES_CM_EVENT_MPA_REQ);
-				break;
-			case  NES_CM_STATE_TSA:
-				handle_exception_pkt(cm_node, skb);
-				break;
-			case NES_CM_STATE_UNKNOWN:
-			case NES_CM_STATE_INITED:
-			default:
-				ret = -1;
-		}
-	}
-
-	return ret;
+	return 0;
 }
 
+/*
+ * active_open_err() will send reset() if flag set..
+ * It will also send ABORT event.
+ */
+
+static void active_open_err(struct nes_cm_node *cm_node, struct sk_buff *skb,
+	int reset)
+{
+	cleanup_retrans_entry(cm_node);
+	if (reset) {
+		nes_debug(NES_DBG_CM, "ERROR active err called for cm_node=%p, "
+				"state=%d\n", cm_node, cm_node->state);
+		add_ref_cm_node(cm_node);
+		send_reset(cm_node, skb);
+	} else
+		dev_kfree_skb_any(skb);
+
+	cm_node->state = NES_CM_STATE_CLOSED;
+	create_event(cm_node, NES_CM_EVENT_ABORTED);
+}
+
+/*
+ * passive_open_err() will either do a reset() or will free up the skb and
+ * remove the cm_node.
+ */
+
+static void passive_open_err(struct nes_cm_node *cm_node, struct sk_buff *skb,
+	int reset)
+{
+	cleanup_retrans_entry(cm_node);
+	cm_node->state = NES_CM_STATE_CLOSED;
+	if (reset) {
+		nes_debug(NES_DBG_CM, "passive_open_err sending RST for "
+			"cm_node=%p state =%d\n", cm_node, cm_node->state);
+		send_reset(cm_node, skb);
+	} else {
+		dev_kfree_skb_any(skb);
+		rem_ref_cm_node(cm_node->cm_core, cm_node);
+	}
+}
+
+/*
+ * free_retrans_entry() routines assumes that the retrans_list_lock has
+ * been acquired before calling.
+ */
+static void free_retrans_entry(struct nes_cm_node *cm_node)
+{
+	struct nes_timer_entry *send_entry;
+	send_entry = cm_node->send_entry;
+	if (send_entry) {
+		cm_node->send_entry = NULL;
+		dev_kfree_skb_any(send_entry->skb);
+		kfree(send_entry);
+		rem_ref_cm_node(cm_node->cm_core, cm_node);
+	}
+}
+
+static void cleanup_retrans_entry(struct nes_cm_node *cm_node)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&cm_node->retrans_list_lock, flags);
+	free_retrans_entry(cm_node);
+	spin_unlock_irqrestore(&cm_node->retrans_list_lock, flags);
+}
+
+/**
+ * process_packet
+ * Returns skb if to be freed, else it will return NULL if already used..
+ */
+static void process_packet(struct nes_cm_node *cm_node, struct sk_buff *skb,
+	struct nes_cm_core *cm_core)
+{
+	enum nes_tcpip_pkt_type	pkt_type = NES_PKT_TYPE_UNKNOWN;
+	struct tcphdr *tcph = tcp_hdr(skb);
+	skb_pull(skb, ip_hdr(skb)->ihl << 2);
+
+	nes_debug(NES_DBG_CM, "process_packet: cm_node=%p state =%d syn=%d "
+		"ack=%d rst=%d fin=%d\n", cm_node, cm_node->state, tcph->syn,
+		tcph->ack, tcph->rst, tcph->fin);
+
+	if (tcph->rst)
+		pkt_type = NES_PKT_TYPE_RST;
+	else if (tcph->syn) {
+		pkt_type = NES_PKT_TYPE_SYN;
+		if (tcph->ack)
+			pkt_type = NES_PKT_TYPE_SYNACK;
+	} else if (tcph->fin)
+		pkt_type = NES_PKT_TYPE_FIN;
+	else if (tcph->ack)
+		pkt_type = NES_PKT_TYPE_ACK;
+
+	switch (pkt_type) {
+	case NES_PKT_TYPE_SYN:
+		handle_syn_pkt(cm_node, skb, tcph);
+		break;
+	case NES_PKT_TYPE_SYNACK:
+		handle_synack_pkt(cm_node, skb, tcph);
+		break;
+	case NES_PKT_TYPE_ACK:
+		handle_ack_pkt(cm_node, skb, tcph);
+		break;
+	case NES_PKT_TYPE_RST:
+		handle_rst_pkt(cm_node, skb, tcph);
+		break;
+	case NES_PKT_TYPE_FIN:
+		handle_fin_pkt(cm_node, skb, tcph);
+		break;
+	default:
+		drop_packet(skb);
+		break;
+	}
+}
 
 /**
  * mini_cm_listen - create a listen node with params
  */
 static struct nes_cm_listener *mini_cm_listen(struct nes_cm_core *cm_core,
-		struct nes_vnic *nesvnic, struct nes_cm_info *cm_info)
+	struct nes_vnic *nesvnic, struct nes_cm_info *cm_info)
 {
 	struct nes_cm_listener *listener;
 	unsigned long flags;
@@ -1644,37 +1826,36 @@ static struct nes_cm_listener *mini_cm_listen(struct nes_cm_core *cm_core,
 /**
  * mini_cm_connect - make a connection node with params
  */
-static struct nes_cm_node *mini_cm_connect(struct nes_cm_core *cm_core,
-					   struct nes_vnic *nesvnic,
-					   struct ietf_mpa_frame *mpa_frame,
-					   struct nes_cm_info *cm_info)
+struct nes_cm_node *mini_cm_connect(struct nes_cm_core *cm_core,
+	struct nes_vnic *nesvnic, u16 private_data_len,
+	void *private_data, struct nes_cm_info *cm_info)
 {
 	int ret = 0;
 	struct nes_cm_node *cm_node;
 	struct nes_cm_listener *loopbackremotelistener;
 	struct nes_cm_node *loopbackremotenode;
 	struct nes_cm_info loopback_cm_info;
-
-	u16 mpa_frame_size = sizeof(struct ietf_mpa_frame) +
-			ntohs(mpa_frame->priv_data_len);
-
-	cm_info->loc_addr = htonl(cm_info->loc_addr);
-	cm_info->rem_addr = htonl(cm_info->rem_addr);
-	cm_info->loc_port = htons(cm_info->loc_port);
-	cm_info->rem_port = htons(cm_info->rem_port);
+	u16 mpa_frame_size = sizeof(struct ietf_mpa_frame) + private_data_len;
+	struct ietf_mpa_frame *mpa_frame = NULL;
 
 	/* create a CM connection node */
 	cm_node = make_cm_node(cm_core, nesvnic, cm_info, NULL);
 	if (!cm_node)
 		return NULL;
+	mpa_frame = &cm_node->mpa_frame;
+	strcpy(mpa_frame->key, IEFT_MPA_KEY_REQ);
+	mpa_frame->flags = IETF_MPA_FLAGS_CRC;
+	mpa_frame->rev =  IETF_MPA_VERSION;
+	mpa_frame->priv_data_len = htons(private_data_len);
 
 	/* set our node side to client (active) side */
 	cm_node->tcp_cntxt.client = 1;
 	cm_node->tcp_cntxt.rcv_wscale = NES_CM_DEFAULT_RCV_WND_SCALE;
 
 	if (cm_info->loc_addr == cm_info->rem_addr) {
-		loopbackremotelistener = find_listener(cm_core, cm_node->rem_addr,
-				cm_node->rem_port, NES_CM_LISTENER_ACTIVE_STATE);
+		loopbackremotelistener = find_listener(cm_core,
+				ntohl(nesvnic->local_ipaddr), cm_node->rem_port,
+				NES_CM_LISTENER_ACTIVE_STATE);
 		if (loopbackremotelistener == NULL) {
 			create_event(cm_node, NES_CM_EVENT_ABORTED);
 		} else {
@@ -1683,26 +1864,35 @@ static struct nes_cm_node *mini_cm_connect(struct nes_cm_core *cm_core,
 			loopback_cm_info.loc_port = cm_info->rem_port;
 			loopback_cm_info.rem_port = cm_info->loc_port;
 			loopback_cm_info.cm_id = loopbackremotelistener->cm_id;
-			loopbackremotenode = make_cm_node(cm_core, nesvnic, &loopback_cm_info,
-					loopbackremotelistener);
+			loopbackremotenode = make_cm_node(cm_core, nesvnic,
+				&loopback_cm_info, loopbackremotelistener);
 			loopbackremotenode->loopbackpartner = cm_node;
-			loopbackremotenode->tcp_cntxt.rcv_wscale = NES_CM_DEFAULT_RCV_WND_SCALE;
+			loopbackremotenode->tcp_cntxt.rcv_wscale =
+				NES_CM_DEFAULT_RCV_WND_SCALE;
 			cm_node->loopbackpartner = loopbackremotenode;
-			memcpy(loopbackremotenode->mpa_frame_buf, &mpa_frame->priv_data,
-					mpa_frame_size);
-			loopbackremotenode->mpa_frame_size = mpa_frame_size -
-					sizeof(struct ietf_mpa_frame);
+			memcpy(loopbackremotenode->mpa_frame_buf, private_data,
+				private_data_len);
+			loopbackremotenode->mpa_frame_size = private_data_len;
 
-			/* we are done handling this state, set node to a TSA state */
+			/* we are done handling this state. */
+			/* set node to a TSA state */
 			cm_node->state = NES_CM_STATE_TSA;
-			cm_node->tcp_cntxt.rcv_nxt = loopbackremotenode->tcp_cntxt.loc_seq_num;
-			loopbackremotenode->tcp_cntxt.rcv_nxt = cm_node->tcp_cntxt.loc_seq_num;
-			cm_node->tcp_cntxt.max_snd_wnd = loopbackremotenode->tcp_cntxt.rcv_wnd;
-			loopbackremotenode->tcp_cntxt.max_snd_wnd = cm_node->tcp_cntxt.rcv_wnd;
-			cm_node->tcp_cntxt.snd_wnd = loopbackremotenode->tcp_cntxt.rcv_wnd;
-			loopbackremotenode->tcp_cntxt.snd_wnd = cm_node->tcp_cntxt.rcv_wnd;
-			cm_node->tcp_cntxt.snd_wscale = loopbackremotenode->tcp_cntxt.rcv_wscale;
-			loopbackremotenode->tcp_cntxt.snd_wscale = cm_node->tcp_cntxt.rcv_wscale;
+			cm_node->tcp_cntxt.rcv_nxt =
+				loopbackremotenode->tcp_cntxt.loc_seq_num;
+			loopbackremotenode->tcp_cntxt.rcv_nxt =
+				cm_node->tcp_cntxt.loc_seq_num;
+			cm_node->tcp_cntxt.max_snd_wnd =
+				loopbackremotenode->tcp_cntxt.rcv_wnd;
+			loopbackremotenode->tcp_cntxt.max_snd_wnd =
+				cm_node->tcp_cntxt.rcv_wnd;
+			cm_node->tcp_cntxt.snd_wnd =
+				loopbackremotenode->tcp_cntxt.rcv_wnd;
+			loopbackremotenode->tcp_cntxt.snd_wnd =
+				cm_node->tcp_cntxt.rcv_wnd;
+			cm_node->tcp_cntxt.snd_wscale =
+				loopbackremotenode->tcp_cntxt.rcv_wscale;
+			loopbackremotenode->tcp_cntxt.snd_wscale =
+				cm_node->tcp_cntxt.rcv_wscale;
 
 			create_event(loopbackremotenode, NES_CM_EVENT_MPA_REQ);
 		}
@@ -1712,16 +1902,29 @@ static struct nes_cm_node *mini_cm_connect(struct nes_cm_core *cm_core,
 	/* set our node side to client (active) side */
 	cm_node->tcp_cntxt.client = 1;
 	/* init our MPA frame ptr */
-	memcpy(&cm_node->mpa_frame, mpa_frame, mpa_frame_size);
+	memcpy(mpa_frame->priv_data, private_data, private_data_len);
+
 	cm_node->mpa_frame_size = mpa_frame_size;
 
 	/* send a syn and goto syn sent state */
 	cm_node->state = NES_CM_STATE_SYN_SENT;
-	ret = send_syn(cm_node, 0);
+	ret = send_syn(cm_node, 0, NULL);
 
-	nes_debug(NES_DBG_CM, "Api - connect(): dest addr=0x%08X, port=0x%04x,"
-			" cm_node=%p, cm_id = %p.\n",
-			cm_node->rem_addr, cm_node->rem_port, cm_node, cm_node->cm_id);
+	if (ret) {
+		/* error in sending the syn free up the cm_node struct */
+		nes_debug(NES_DBG_CM, "Api - connect() FAILED: dest "
+			"addr=0x%08X, port=0x%04x, cm_node=%p, cm_id = %p.\n",
+			cm_node->rem_addr, cm_node->rem_port, cm_node,
+			cm_node->cm_id);
+		rem_ref_cm_node(cm_node->cm_core, cm_node);
+		cm_node = NULL;
+	}
+
+	if (cm_node)
+		nes_debug(NES_DBG_CM, "Api - connect(): dest addr=0x%08X,"
+			"port=0x%04x, cm_node=%p, cm_id = %p.\n",
+			cm_node->rem_addr, cm_node->rem_port, cm_node,
+			cm_node->cm_id);
 
 	return cm_node;
 }
@@ -1731,8 +1934,8 @@ static struct nes_cm_node *mini_cm_connect(struct nes_cm_core *cm_core,
  * mini_cm_accept - accept a connection
  * This function is never called
  */
-static int mini_cm_accept(struct nes_cm_core *cm_core, struct ietf_mpa_frame *mpa_frame,
-			  struct nes_cm_node *cm_node)
+static int mini_cm_accept(struct nes_cm_core *cm_core,
+	struct ietf_mpa_frame *mpa_frame, struct nes_cm_node *cm_node)
 {
 	return 0;
 }
@@ -1742,32 +1945,26 @@ static int mini_cm_accept(struct nes_cm_core *cm_core, struct ietf_mpa_frame *mp
  * mini_cm_reject - reject and teardown a connection
  */
 static int mini_cm_reject(struct nes_cm_core *cm_core,
-			  struct ietf_mpa_frame *mpa_frame,
-			  struct nes_cm_node *cm_node)
+	struct ietf_mpa_frame *mpa_frame, struct nes_cm_node *cm_node)
 {
 	int ret = 0;
-	struct sk_buff *skb;
-	u16 mpa_frame_size = sizeof(struct ietf_mpa_frame) +
-			ntohs(mpa_frame->priv_data_len);
 
-	skb = get_free_pkt(cm_node);
-	if (!skb) {
-		nes_debug(NES_DBG_CM, "Failed to get a Free pkt\n");
-		return -1;
-	}
-
-	/* send an MPA Request frame */
-	form_cm_frame(skb, cm_node, NULL, 0, mpa_frame, mpa_frame_size, SET_ACK | SET_FIN);
-	ret = schedule_nes_timer(cm_node, skb, NES_TIMER_TYPE_SEND, 1, 0);
+	nes_debug(NES_DBG_CM, "%s cm_node=%p type=%d state=%d\n",
+		__func__, cm_node, cm_node->tcp_cntxt.client, cm_node->state);
 
+	if (cm_node->tcp_cntxt.client)
+		return ret;
+	cleanup_retrans_entry(cm_node);
 	cm_node->state = NES_CM_STATE_CLOSED;
 	ret = send_fin(cm_node, NULL);
 
-	if (ret < 0) {
-		printk(KERN_INFO PFX "failed to send MPA Reply (reject)\n");
-		return ret;
+	if (cm_node->accept_pend) {
+		BUG_ON(!cm_node->listener);
+		atomic_dec(&cm_node->listener->pend_accepts_cnt);
+		BUG_ON(atomic_read(&cm_node->listener->pend_accepts_cnt) < 0);
 	}
 
+	ret = send_reset(cm_node, NULL);
 	return ret;
 }
 
@@ -1783,35 +1980,39 @@ static int mini_cm_close(struct nes_cm_core *cm_core, struct nes_cm_node *cm_nod
 		return -EINVAL;
 
 	switch (cm_node->state) {
-		/* if passed in node is null, create a reference key node for node search */
-		/* check if we found an owner node for this pkt */
-		case NES_CM_STATE_SYN_RCVD:
-		case NES_CM_STATE_SYN_SENT:
-		case NES_CM_STATE_ONE_SIDE_ESTABLISHED:
-		case NES_CM_STATE_ESTABLISHED:
-		case NES_CM_STATE_ACCEPTING:
-		case NES_CM_STATE_MPAREQ_SENT:
-			cm_node->state = NES_CM_STATE_FIN_WAIT1;
-			send_fin(cm_node, NULL);
-			break;
-		case NES_CM_STATE_CLOSE_WAIT:
-			cm_node->state = NES_CM_STATE_LAST_ACK;
-			send_fin(cm_node, NULL);
-			break;
-		case NES_CM_STATE_FIN_WAIT1:
-		case NES_CM_STATE_FIN_WAIT2:
-		case NES_CM_STATE_LAST_ACK:
-		case NES_CM_STATE_TIME_WAIT:
-		case NES_CM_STATE_CLOSING:
-			ret = -1;
-			break;
-		case NES_CM_STATE_LISTENING:
-		case NES_CM_STATE_UNKNOWN:
-		case NES_CM_STATE_INITED:
-		case NES_CM_STATE_CLOSED:
-		case NES_CM_STATE_TSA:
-			ret = rem_ref_cm_node(cm_core, cm_node);
-			break;
+	case NES_CM_STATE_SYN_RCVD:
+	case NES_CM_STATE_SYN_SENT:
+	case NES_CM_STATE_ONE_SIDE_ESTABLISHED:
+	case NES_CM_STATE_ESTABLISHED:
+	case NES_CM_STATE_ACCEPTING:
+	case NES_CM_STATE_MPAREQ_SENT:
+	case NES_CM_STATE_MPAREQ_RCVD:
+		cleanup_retrans_entry(cm_node);
+		send_reset(cm_node, NULL);
+		break;
+	case NES_CM_STATE_CLOSE_WAIT:
+		cm_node->state = NES_CM_STATE_LAST_ACK;
+		send_fin(cm_node, NULL);
+		break;
+	case NES_CM_STATE_FIN_WAIT1:
+	case NES_CM_STATE_FIN_WAIT2:
+	case NES_CM_STATE_LAST_ACK:
+	case NES_CM_STATE_TIME_WAIT:
+	case NES_CM_STATE_CLOSING:
+		ret = -1;
+		break;
+	case NES_CM_STATE_LISTENING:
+	case NES_CM_STATE_UNKNOWN:
+	case NES_CM_STATE_INITED:
+	case NES_CM_STATE_CLOSED:
+		ret = rem_ref_cm_node(cm_core, cm_node);
+		break;
+	case NES_CM_STATE_TSA:
+		if (cm_node->send_entry)
+			printk(KERN_ERR "ERROR Close got called from STATE_TSA "
+				"send_entry=%p\n", cm_node->send_entry);
+		ret = rem_ref_cm_node(cm_core, cm_node);
+		break;
 	}
 	cm_node->cm_id = NULL;
 	return ret;
@@ -1822,25 +2023,30 @@ static int mini_cm_close(struct nes_cm_core *cm_core, struct nes_cm_node *cm_nod
  * recv_pkt - recv an ETHERNET packet, and process it through CM
  * node state machine
  */
-static int mini_cm_recv_pkt(struct nes_cm_core *cm_core, struct nes_vnic *nesvnic,
-			    struct sk_buff *skb)
+static void mini_cm_recv_pkt(struct nes_cm_core *cm_core,
+	struct nes_vnic *nesvnic, struct sk_buff *skb)
 {
 	struct nes_cm_node *cm_node = NULL;
 	struct nes_cm_listener *listener = NULL;
 	struct iphdr *iph;
 	struct tcphdr *tcph;
 	struct nes_cm_info nfo;
-	int ret = 0;
 
-	if (!skb || skb->len < sizeof(struct iphdr) + sizeof(struct tcphdr)) {
-		ret = -EINVAL;
-		goto out;
+	if (!skb)
+		return;
+	if (skb->len < sizeof(struct iphdr) + sizeof(struct tcphdr)) {
+		dev_kfree_skb_any(skb);
+		return;
 	}
 
 	iph = (struct iphdr *)skb->data;
 	tcph = (struct tcphdr *)(skb->data + sizeof(struct iphdr));
 	skb_reset_network_header(skb);
 	skb_set_transport_header(skb, sizeof(*tcph));
+	if (!tcph) {
+		dev_kfree_skb_any(skb);
+		return;
+	}
 	skb->len = ntohs(iph->tot_len);
 
 	nfo.loc_addr = ntohl(iph->daddr);
@@ -1853,61 +2059,60 @@ static int mini_cm_recv_pkt(struct nes_cm_core *cm_core, struct nes_vnic *nesvni
 		  NIPQUAD(iph->daddr), tcph->dest,
 		  NIPQUAD(iph->saddr), tcph->source);
 
-	/* note: this call is going to increment cm_node ref count */
-	cm_node = find_node(cm_core,
+	do {
+		cm_node = find_node(cm_core,
 			nfo.rem_port, nfo.rem_addr,
 			nfo.loc_port, nfo.loc_addr);
 
-	if (!cm_node) {
-		listener = find_listener(cm_core, nfo.loc_addr, nfo.loc_port,
-				NES_CM_LISTENER_ACTIVE_STATE);
-		if (listener) {
-			nfo.cm_id = listener->cm_id;
-			nfo.conn_type = listener->conn_type;
-		} else {
-			nfo.cm_id = NULL;
-			nfo.conn_type = 0;
-		}
-
-		cm_node = make_cm_node(cm_core, nesvnic, &nfo, listener);
 		if (!cm_node) {
-			nes_debug(NES_DBG_CM, "Unable to allocate node\n");
+			/* Only type of packet accepted are for */
+			/* the PASSIVE open (syn only) */
+			if ((!tcph->syn) || (tcph->ack)) {
+				cm_packets_dropped++;
+				break;
+			}
+			listener = find_listener(cm_core, nfo.loc_addr,
+				nfo.loc_port,
+				NES_CM_LISTENER_ACTIVE_STATE);
 			if (listener) {
-				nes_debug(NES_DBG_CM, "unable to allocate node and decrementing listener refcount\n");
+				nfo.cm_id = listener->cm_id;
+				nfo.conn_type = listener->conn_type;
+			} else {
+				nes_debug(NES_DBG_CM, "Unable to find listener "
+					"for the pkt\n");
+				cm_packets_dropped++;
+				dev_kfree_skb_any(skb);
+				break;
+			}
+
+			cm_node = make_cm_node(cm_core, nesvnic, &nfo,
+				listener);
+			if (!cm_node) {
+				nes_debug(NES_DBG_CM, "Unable to allocate "
+					"node\n");
+				cm_packets_dropped++;
 				atomic_dec(&listener->ref_count);
+				dev_kfree_skb_any(skb);
+				break;
 			}
-			ret = -1;
-			goto out;
-		}
-		if (!listener) {
-			nes_debug(NES_DBG_CM, "Packet found for unknown port %x refcnt=%d\n",
-					nfo.loc_port, atomic_read(&cm_node->ref_count));
-			if (!tcph->rst) {
-				nes_debug(NES_DBG_CM, "Packet found for unknown port=%d"
-						" rem_port=%d refcnt=%d\n",
-						nfo.loc_port, nfo.rem_port, atomic_read(&cm_node->ref_count));
-
-				cm_node->tcp_cntxt.rcv_nxt = ntohl(tcph->seq);
-				cm_node->tcp_cntxt.loc_seq_num = ntohl(tcph->ack_seq);
-				send_reset(cm_node);
+			if (!tcph->rst && !tcph->fin) {
+				cm_node->state = NES_CM_STATE_LISTENING;
+			} else {
+				cm_packets_dropped++;
+				rem_ref_cm_node(cm_core, cm_node);
+				dev_kfree_skb_any(skb);
+				break;
 			}
+			add_ref_cm_node(cm_node);
+		} else if (cm_node->state == NES_CM_STATE_TSA) {
 			rem_ref_cm_node(cm_core, cm_node);
-			ret = -1;
-			goto out;
+			atomic_inc(&cm_accel_dropped_pkts);
+			dev_kfree_skb_any(skb);
+			break;
 		}
-		add_ref_cm_node(cm_node);
-		cm_node->state = NES_CM_STATE_LISTENING;
-	}
-
-	nes_debug(NES_DBG_CM, "Processing Packet for node %p, data = (%p):\n",
-			cm_node, skb->data);
-	process_packet(cm_node, skb, cm_core);
-
-	rem_ref_cm_node(cm_core, cm_node);
-	out:
-	if (skb)
-		dev_kfree_skb_any(skb);
-	return ret;
+		process_packet(cm_node, skb, cm_core);
+		rem_ref_cm_node(cm_core, cm_node);
+	} while (0);
 }
 
 
@@ -2107,15 +2312,12 @@ int nes_cm_disconn(struct nes_qp *nesqp)
 	if (nesqp->disconn_pending == 0) {
 		nesqp->disconn_pending++;
 		spin_unlock_irqrestore(&nesqp->lock, flags);
-		/* nes_add_ref(&nesqp->ibqp); */
 		/* init our disconnect work element, to */
 		INIT_WORK(&nesqp->disconn_work, nes_disconnect_worker);
 
 		queue_work(g_cm_core->disconn_wq, &nesqp->disconn_work);
-	} else {
+	} else
 		spin_unlock_irqrestore(&nesqp->lock, flags);
-		nes_rem_ref(&nesqp->ibqp);
-	}
 
 	return 0;
 }
@@ -2161,7 +2363,6 @@ static int nes_cm_disconn_true(struct nes_qp *nesqp)
 		nes_debug(NES_DBG_CM, "QP%u disconnect_worker cmid is NULL\n",
 				nesqp->hwqp.qp_id);
 		spin_unlock_irqrestore(&nesqp->lock, flags);
-		nes_rem_ref(&nesqp->ibqp);
 		return -1;
 	}
 
@@ -2182,30 +2383,31 @@ static int nes_cm_disconn_true(struct nes_qp *nesqp)
 			atomic_inc(&cm_disconnects);
 			cm_event.event = IW_CM_EVENT_DISCONNECT;
 			if (last_ae == NES_AEQE_AEID_LLP_CONNECTION_RESET) {
-				issued_disconnect_reset = 1;
 				cm_event.status = IW_CM_EVENT_STATUS_RESET;
-				nes_debug(NES_DBG_CM, "Generating a CM Disconnect Event (status reset) for "
-						" QP%u, cm_id = %p. \n",
-						nesqp->hwqp.qp_id, cm_id);
-			} else {
+				nes_debug(NES_DBG_CM, "Generating a CM "
+					"Disconnect Event (status reset) for "
+					"QP%u, cm_id = %p. \n",
+					nesqp->hwqp.qp_id, cm_id);
+			} else
 				cm_event.status = IW_CM_EVENT_STATUS_OK;
-			}
 
 			cm_event.local_addr = cm_id->local_addr;
 			cm_event.remote_addr = cm_id->remote_addr;
 			cm_event.private_data = NULL;
 			cm_event.private_data_len = 0;
 
-			nes_debug(NES_DBG_CM, "Generating a CM Disconnect Event for "
-					" QP%u, SQ Head = %u, SQ Tail = %u. cm_id = %p, refcount = %u.\n",
-					nesqp->hwqp.qp_id,
-					nesqp->hwqp.sq_head, nesqp->hwqp.sq_tail, cm_id,
-					atomic_read(&nesqp->refcount));
+			nes_debug(NES_DBG_CM, "Generating a CM Disconnect Event"
+				" for  QP%u, SQ Head = %u, SQ Tail = %u. "
+				"cm_id = %p, refcount = %u.\n",
+				nesqp->hwqp.qp_id, nesqp->hwqp.sq_head,
+				nesqp->hwqp.sq_tail, cm_id,
+				atomic_read(&nesqp->refcount));
 
 			spin_unlock_irqrestore(&nesqp->lock, flags);
 			ret = cm_id->event_handler(cm_id, &cm_event);
 			if (ret)
-				nes_debug(NES_DBG_CM, "OFA CM event_handler returned, ret=%d\n", ret);
+				nes_debug(NES_DBG_CM, "OFA CM event_handler "
+					"returned, ret=%d\n", ret);
 			spin_lock_irqsave(&nesqp->lock, flags);
 		}
 
@@ -2247,31 +2449,24 @@ static int nes_cm_disconn_true(struct nes_qp *nesqp)
 			if (nesqp->flush_issued == 0) {
 				nesqp->flush_issued = 1;
 				spin_unlock_irqrestore(&nesqp->lock, flags);
-				flush_wqes(nesvnic->nesdev, nesqp, NES_CQP_FLUSH_RQ, 1);
-			} else {
+				flush_wqes(nesvnic->nesdev, nesqp,
+					NES_CQP_FLUSH_RQ, 1);
+			} else
 				spin_unlock_irqrestore(&nesqp->lock, flags);
-			}
-
-			/* This reference is from either ModifyQP or the AE processing,
-					there is still a race here with modifyqp */
-			nes_rem_ref(&nesqp->ibqp);
-
 		} else {
 			cm_id = nesqp->cm_id;
 			spin_unlock_irqrestore(&nesqp->lock, flags);
 			/* check to see if the inbound reset beat the outbound reset */
 			if ((!cm_id) && (last_ae==NES_AEQE_AEID_RESET_SENT)) {
-				nes_debug(NES_DBG_CM, "QP%u: Decing refcount due to inbound reset"
-						" beating the outbound reset.\n",
-						nesqp->hwqp.qp_id);
-				nes_rem_ref(&nesqp->ibqp);
+				nes_debug(NES_DBG_CM, "QP%u: Decing refcount "
+					"due to inbound reset beating the "
+					"outbound reset.\n", nesqp->hwqp.qp_id);
 			}
 		}
 	} else {
 		nesqp->disconn_pending = 0;
 		spin_unlock_irqrestore(&nesqp->lock, flags);
 	}
-	nes_rem_ref(&nesqp->ibqp);
 
 	return 0;
 }
@@ -2349,71 +2544,82 @@ int nes_accept(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	nesdev = nesvnic->nesdev;
 	adapter = nesdev->nesadapter;
 
-	nes_debug(NES_DBG_CM, "nesvnic=%p, netdev=%p, %s\n",
-			nesvnic, nesvnic->netdev, nesvnic->netdev->name);
-
-	/* since this is from a listen, we were able to put node handle into cm_id */
 	cm_node = (struct nes_cm_node *)cm_id->provider_data;
+	nes_debug(NES_DBG_CM, "nes_accept: cm_node= %p nesvnic=%p, netdev=%p,"
+		"%s\n", cm_node, nesvnic, nesvnic->netdev,
+		nesvnic->netdev->name);
 
 	/* associate the node with the QP */
 	nesqp->cm_node = (void *)cm_node;
+	cm_node->nesqp = nesqp;
+	nes_add_ref(&nesqp->ibqp);
 
-	nes_debug(NES_DBG_CM, "QP%u, cm_node=%p, jiffies = %lu\n",
-			nesqp->hwqp.qp_id, cm_node, jiffies);
+	nes_debug(NES_DBG_CM, "QP%u, cm_node=%p, jiffies = %lu listener = %p\n",
+		nesqp->hwqp.qp_id, cm_node, jiffies, cm_node->listener);
 	atomic_inc(&cm_accepts);
 
 	nes_debug(NES_DBG_CM, "netdev refcnt = %u.\n",
 			atomic_read(&nesvnic->netdev->refcnt));
 
-		/* allocate the ietf frame and space for private data */
-		nesqp->ietf_frame = pci_alloc_consistent(nesdev->pcidev,
-				sizeof(struct ietf_mpa_frame) + conn_param->private_data_len,
-				&nesqp->ietf_frame_pbase);
+	/* allocate the ietf frame and space for private data */
+	nesqp->ietf_frame = pci_alloc_consistent(nesdev->pcidev,
+		sizeof(struct ietf_mpa_frame) + conn_param->private_data_len,
+		&nesqp->ietf_frame_pbase);
 
-		if (!nesqp->ietf_frame) {
-			nes_debug(NES_DBG_CM, "Unable to allocate memory for private data\n");
-			return -ENOMEM;
-		}
+	if (!nesqp->ietf_frame) {
+		nes_debug(NES_DBG_CM, "Unable to allocate memory for private "
+			"data\n");
+		return -ENOMEM;
+	}
 
 
-		/* setup the MPA frame */
-		nesqp->private_data_len = conn_param->private_data_len;
-		memcpy(nesqp->ietf_frame->key, IEFT_MPA_KEY_REP, IETF_MPA_KEY_SIZE);
+	/* setup the MPA frame */
+	nesqp->private_data_len = conn_param->private_data_len;
+	memcpy(nesqp->ietf_frame->key, IEFT_MPA_KEY_REP, IETF_MPA_KEY_SIZE);
 
-		memcpy(nesqp->ietf_frame->priv_data, conn_param->private_data,
-				conn_param->private_data_len);
+	memcpy(nesqp->ietf_frame->priv_data, conn_param->private_data,
+			conn_param->private_data_len);
 
-		nesqp->ietf_frame->priv_data_len = cpu_to_be16(conn_param->private_data_len);
-		nesqp->ietf_frame->rev = mpa_version;
-		nesqp->ietf_frame->flags = IETF_MPA_FLAGS_CRC;
+	nesqp->ietf_frame->priv_data_len =
+		cpu_to_be16(conn_param->private_data_len);
+	nesqp->ietf_frame->rev = mpa_version;
+	nesqp->ietf_frame->flags = IETF_MPA_FLAGS_CRC;
 
-		/* setup our first outgoing iWarp send WQE (the IETF frame response) */
-		wqe = &nesqp->hwqp.sq_vbase[0];
+	/* setup our first outgoing iWarp send WQE (the IETF frame response) */
+	wqe = &nesqp->hwqp.sq_vbase[0];
 
-		if (cm_id->remote_addr.sin_addr.s_addr != cm_id->local_addr.sin_addr.s_addr) {
-			u64temp = (unsigned long)nesqp;
-			u64temp |= NES_SW_CONTEXT_ALIGN>>1;
-			set_wqe_64bit_value(wqe->wqe_words, NES_IWARP_SQ_WQE_COMP_CTX_LOW_IDX,
-					    u64temp);
-			wqe->wqe_words[NES_IWARP_SQ_WQE_MISC_IDX] =
-					cpu_to_le32(NES_IWARP_SQ_WQE_STREAMING | NES_IWARP_SQ_WQE_WRPDU);
-			wqe->wqe_words[NES_IWARP_SQ_WQE_TOTAL_PAYLOAD_IDX] =
-					cpu_to_le32(conn_param->private_data_len + sizeof(struct ietf_mpa_frame));
-			wqe->wqe_words[NES_IWARP_SQ_WQE_FRAG0_LOW_IDX] =
-					cpu_to_le32((u32)nesqp->ietf_frame_pbase);
-			wqe->wqe_words[NES_IWARP_SQ_WQE_FRAG0_HIGH_IDX] =
-					cpu_to_le32((u32)((u64)nesqp->ietf_frame_pbase >> 32));
-			wqe->wqe_words[NES_IWARP_SQ_WQE_LENGTH0_IDX] =
-					cpu_to_le32(conn_param->private_data_len + sizeof(struct ietf_mpa_frame));
-			wqe->wqe_words[NES_IWARP_SQ_WQE_STAG0_IDX] = 0;
+	if (cm_id->remote_addr.sin_addr.s_addr !=
+			cm_id->local_addr.sin_addr.s_addr) {
+		u64temp = (unsigned long)nesqp;
+		u64temp |= NES_SW_CONTEXT_ALIGN>>1;
+		set_wqe_64bit_value(wqe->wqe_words,
+			NES_IWARP_SQ_WQE_COMP_CTX_LOW_IDX,
+			u64temp);
+		wqe->wqe_words[NES_IWARP_SQ_WQE_MISC_IDX] =
+			cpu_to_le32(NES_IWARP_SQ_WQE_STREAMING |
+			NES_IWARP_SQ_WQE_WRPDU);
+		wqe->wqe_words[NES_IWARP_SQ_WQE_TOTAL_PAYLOAD_IDX] =
+			cpu_to_le32(conn_param->private_data_len +
+			sizeof(struct ietf_mpa_frame));
+		wqe->wqe_words[NES_IWARP_SQ_WQE_FRAG0_LOW_IDX] =
+			cpu_to_le32((u32)nesqp->ietf_frame_pbase);
+		wqe->wqe_words[NES_IWARP_SQ_WQE_FRAG0_HIGH_IDX] =
+			cpu_to_le32((u32)((u64)nesqp->ietf_frame_pbase >> 32));
+		wqe->wqe_words[NES_IWARP_SQ_WQE_LENGTH0_IDX] =
+			cpu_to_le32(conn_param->private_data_len +
+			sizeof(struct ietf_mpa_frame));
+		wqe->wqe_words[NES_IWARP_SQ_WQE_STAG0_IDX] = 0;
 
-			nesqp->nesqp_context->ird_ord_sizes |= cpu_to_le32(
-					NES_QPCONTEXT_ORDIRD_LSMM_PRESENT | NES_QPCONTEXT_ORDIRD_WRPDU);
-		} else {
-			nesqp->nesqp_context->ird_ord_sizes |= cpu_to_le32((NES_QPCONTEXT_ORDIRD_LSMM_PRESENT |
-					NES_QPCONTEXT_ORDIRD_WRPDU | NES_QPCONTEXT_ORDIRD_ALSMM));
-		}
-		nesqp->skip_lsmm = 1;
+		nesqp->nesqp_context->ird_ord_sizes |=
+			cpu_to_le32(NES_QPCONTEXT_ORDIRD_LSMM_PRESENT |
+			NES_QPCONTEXT_ORDIRD_WRPDU);
+	} else {
+		nesqp->nesqp_context->ird_ord_sizes |=
+			cpu_to_le32((NES_QPCONTEXT_ORDIRD_LSMM_PRESENT |
+			NES_QPCONTEXT_ORDIRD_WRPDU |
+			NES_QPCONTEXT_ORDIRD_ALSMM));
+	}
+	nesqp->skip_lsmm = 1;
 
 
 	/* Cache the cm_id in the qp */
@@ -2424,55 +2630,75 @@ int nes_accept(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	cm_id->provider_data = nesqp;
 	nesqp->active_conn   = 0;
 
+	if (cm_node->state == NES_CM_STATE_TSA)
+		nes_debug(NES_DBG_CM, "Already state = TSA for cm_node=%p\n",
+			cm_node);
+
 	nes_cm_init_tsa_conn(nesqp, cm_node);
 
-	nesqp->nesqp_context->tcpPorts[0] = cpu_to_le16(ntohs(cm_id->local_addr.sin_port));
-	nesqp->nesqp_context->tcpPorts[1] = cpu_to_le16(ntohs(cm_id->remote_addr.sin_port));
-	nesqp->nesqp_context->ip0 = cpu_to_le32(ntohl(cm_id->remote_addr.sin_addr.s_addr));
+	nesqp->nesqp_context->tcpPorts[0] =
+		cpu_to_le16(ntohs(cm_id->local_addr.sin_port));
+	nesqp->nesqp_context->tcpPorts[1] =
+		cpu_to_le16(ntohs(cm_id->remote_addr.sin_port));
+
+	if (ipv4_is_loopback(cm_id->remote_addr.sin_addr.s_addr))
+		nesqp->nesqp_context->ip0 =
+			cpu_to_le32(ntohl(nesvnic->local_ipaddr));
+	else
+		nesqp->nesqp_context->ip0 =
+			cpu_to_le32(ntohl(cm_id->remote_addr.sin_addr.s_addr));
 
 	nesqp->nesqp_context->misc2 |= cpu_to_le32(
-			(u32)PCI_FUNC(nesdev->pcidev->devfn) << NES_QPCONTEXT_MISC2_SRC_IP_SHIFT);
+			(u32)PCI_FUNC(nesdev->pcidev->devfn) <<
+			NES_QPCONTEXT_MISC2_SRC_IP_SHIFT);
 
-	nesqp->nesqp_context->arp_index_vlan |= cpu_to_le32(
-			nes_arp_table(nesdev, le32_to_cpu(nesqp->nesqp_context->ip0), NULL,
+	nesqp->nesqp_context->arp_index_vlan |=
+		cpu_to_le32(nes_arp_table(nesdev,
+			le32_to_cpu(nesqp->nesqp_context->ip0), NULL,
 			NES_ARP_RESOLVE) << 16);
 
 	nesqp->nesqp_context->ts_val_delta = cpu_to_le32(
-			jiffies - nes_read_indexed(nesdev, NES_IDX_TCP_NOW));
+		jiffies - nes_read_indexed(nesdev, NES_IDX_TCP_NOW));
 
 	nesqp->nesqp_context->ird_index = cpu_to_le32(nesqp->hwqp.qp_id);
 
 	nesqp->nesqp_context->ird_ord_sizes |= cpu_to_le32(
-			((u32)1 << NES_QPCONTEXT_ORDIRD_IWARP_MODE_SHIFT));
-	nesqp->nesqp_context->ird_ord_sizes |= cpu_to_le32((u32)conn_param->ord);
+		((u32)1 << NES_QPCONTEXT_ORDIRD_IWARP_MODE_SHIFT));
+	nesqp->nesqp_context->ird_ord_sizes |=
+		cpu_to_le32((u32)conn_param->ord);
 
 	memset(&nes_quad, 0, sizeof(nes_quad));
-	nes_quad.DstIpAdrIndex = cpu_to_le32((u32)PCI_FUNC(nesdev->pcidev->devfn) << 24);
-	nes_quad.SrcIpadr      = cm_id->remote_addr.sin_addr.s_addr;
-	nes_quad.TcpPorts[0]   = cm_id->remote_addr.sin_port;
-	nes_quad.TcpPorts[1]   = cm_id->local_addr.sin_port;
+	nes_quad.DstIpAdrIndex =
+		cpu_to_le32((u32)PCI_FUNC(nesdev->pcidev->devfn) << 24);
+	if (ipv4_is_loopback(cm_id->remote_addr.sin_addr.s_addr))
+		nes_quad.SrcIpadr = nesvnic->local_ipaddr;
+	else
+		nes_quad.SrcIpadr = cm_id->remote_addr.sin_addr.s_addr;
+	nes_quad.TcpPorts[0] = cm_id->remote_addr.sin_port;
+	nes_quad.TcpPorts[1] = cm_id->local_addr.sin_port;
 
 	/* Produce hash key */
 	crc_value = get_crc_value(&nes_quad);
 	nesqp->hte_index = cpu_to_be32(crc_value ^ 0xffffffff);
 	nes_debug(NES_DBG_CM, "HTE Index = 0x%08X, CRC = 0x%08X\n",
-			nesqp->hte_index, nesqp->hte_index & adapter->hte_index_mask);
+		nesqp->hte_index, nesqp->hte_index & adapter->hte_index_mask);
 
 	nesqp->hte_index &= adapter->hte_index_mask;
 	nesqp->nesqp_context->hte_index = cpu_to_le32(nesqp->hte_index);
 
 	cm_node->cm_core->api->accelerated(cm_node->cm_core, cm_node);
 
-	nes_debug(NES_DBG_CM, "QP%u, Destination IP = 0x%08X:0x%04X, local = 0x%08X:0x%04X,"
-			" rcv_nxt=0x%08X, snd_nxt=0x%08X, mpa + private data length=%zu.\n",
-			nesqp->hwqp.qp_id,
+	nes_debug(NES_DBG_CM, "QP%u, Destination IP = 0x%08X:0x%04X, local = "
+			"0x%08X:0x%04X, rcv_nxt=0x%08X, snd_nxt=0x%08X, mpa + "
+			"private data length=%zu.\n", nesqp->hwqp.qp_id,
 			ntohl(cm_id->remote_addr.sin_addr.s_addr),
 			ntohs(cm_id->remote_addr.sin_port),
 			ntohl(cm_id->local_addr.sin_addr.s_addr),
 			ntohs(cm_id->local_addr.sin_port),
 			le32_to_cpu(nesqp->nesqp_context->rcv_nxt),
 			le32_to_cpu(nesqp->nesqp_context->snd_nxt),
-			conn_param->private_data_len+sizeof(struct ietf_mpa_frame));
+			conn_param->private_data_len +
+			sizeof(struct ietf_mpa_frame));
 
 	attr.qp_state = IB_QPS_RTS;
 	nes_modify_qp(&nesqp->ibqp, &attr, IB_QP_STATE, NULL);
@@ -2489,15 +2715,16 @@ int nes_accept(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	cm_event.private_data_len = 0;
 	ret = cm_id->event_handler(cm_id, &cm_event);
 	if (cm_node->loopbackpartner) {
-		cm_node->loopbackpartner->mpa_frame_size = nesqp->private_data_len;
+		cm_node->loopbackpartner->mpa_frame_size =
+			nesqp->private_data_len;
 		/* copy entire MPA frame to our cm_node's frame */
-		memcpy(cm_node->loopbackpartner->mpa_frame_buf, nesqp->ietf_frame->priv_data,
-			   nesqp->private_data_len);
+		memcpy(cm_node->loopbackpartner->mpa_frame_buf,
+			nesqp->ietf_frame->priv_data, nesqp->private_data_len);
 		create_event(cm_node->loopbackpartner, NES_CM_EVENT_CONNECTED);
 	}
 	if (ret)
-		printk("%s[%u] OFA CM event_handler returned, ret=%d\n",
-				__func__, __LINE__, ret);
+		printk(KERN_ERR "%s[%u] OFA CM event_handler returned, "
+			"ret=%d\n", __func__, __LINE__, ret);
 
 	return 0;
 }
@@ -2555,74 +2782,61 @@ int nes_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	if (!nesdev)
 		return -EINVAL;
 
+	nes_debug(NES_DBG_CM, "QP%u, current IP = 0x%08X, Destination IP = "
+		"0x%08X:0x%04X, local = 0x%08X:0x%04X.\n", nesqp->hwqp.qp_id,
+		ntohl(nesvnic->local_ipaddr),
+		ntohl(cm_id->remote_addr.sin_addr.s_addr),
+		ntohs(cm_id->remote_addr.sin_port),
+		ntohl(cm_id->local_addr.sin_addr.s_addr),
+		ntohs(cm_id->local_addr.sin_port));
+
 	atomic_inc(&cm_connects);
-
-	nesqp->ietf_frame = kzalloc(sizeof(struct ietf_mpa_frame) +
-			conn_param->private_data_len, GFP_KERNEL);
-	if (!nesqp->ietf_frame)
-		return -ENOMEM;
-
-	/* set qp as having an active connection */
 	nesqp->active_conn = 1;
 
-	nes_debug(NES_DBG_CM, "QP%u, Destination IP = 0x%08X:0x%04X, local = 0x%08X:0x%04X.\n",
-			nesqp->hwqp.qp_id,
-			ntohl(cm_id->remote_addr.sin_addr.s_addr),
-			ntohs(cm_id->remote_addr.sin_port),
-			ntohl(cm_id->local_addr.sin_addr.s_addr),
-			ntohs(cm_id->local_addr.sin_port));
-
 	/* cache the cm_id in the qp */
 	nesqp->cm_id = cm_id;
 
 	cm_id->provider_data = nesqp;
 
-	/* copy the private data */
-	if (conn_param->private_data_len) {
-		memcpy(nesqp->ietf_frame->priv_data, conn_param->private_data,
-				conn_param->private_data_len);
-	}
-
 	nesqp->private_data_len = conn_param->private_data_len;
 	nesqp->nesqp_context->ird_ord_sizes |= cpu_to_le32((u32)conn_param->ord);
 	nes_debug(NES_DBG_CM, "requested ord = 0x%08X.\n", (u32)conn_param->ord);
-	nes_debug(NES_DBG_CM, "mpa private data len =%u\n", conn_param->private_data_len);
+	nes_debug(NES_DBG_CM, "mpa private data len =%u\n",
+		conn_param->private_data_len);
 
-	strcpy(&nesqp->ietf_frame->key[0], IEFT_MPA_KEY_REQ);
-	nesqp->ietf_frame->flags = IETF_MPA_FLAGS_CRC;
-	nesqp->ietf_frame->rev = IETF_MPA_VERSION;
-	nesqp->ietf_frame->priv_data_len = htons(conn_param->private_data_len);
-
-	if (cm_id->local_addr.sin_addr.s_addr != cm_id->remote_addr.sin_addr.s_addr)
+	if (cm_id->local_addr.sin_addr.s_addr !=
+		cm_id->remote_addr.sin_addr.s_addr)
 		nes_manage_apbvt(nesvnic, ntohs(cm_id->local_addr.sin_port),
-				PCI_FUNC(nesdev->pcidev->devfn), NES_MANAGE_APBVT_ADD);
+			PCI_FUNC(nesdev->pcidev->devfn), NES_MANAGE_APBVT_ADD);
 
 	/* set up the connection params for the node */
-	cm_info.loc_addr = (cm_id->local_addr.sin_addr.s_addr);
-	cm_info.loc_port = (cm_id->local_addr.sin_port);
-	cm_info.rem_addr = (cm_id->remote_addr.sin_addr.s_addr);
-	cm_info.rem_port = (cm_id->remote_addr.sin_port);
+	cm_info.loc_addr = htonl(cm_id->local_addr.sin_addr.s_addr);
+	cm_info.loc_port = htons(cm_id->local_addr.sin_port);
+	cm_info.rem_addr = htonl(cm_id->remote_addr.sin_addr.s_addr);
+	cm_info.rem_port = htons(cm_id->remote_addr.sin_port);
 	cm_info.cm_id = cm_id;
 	cm_info.conn_type = NES_CM_IWARP_CONN_TYPE;
 
 	cm_id->add_ref(cm_id);
-	nes_add_ref(&nesqp->ibqp);
 
 	/* create a connect CM node connection */
-	cm_node = g_cm_core->api->connect(g_cm_core, nesvnic, nesqp->ietf_frame, &cm_info);
+	cm_node = g_cm_core->api->connect(g_cm_core, nesvnic,
+		conn_param->private_data_len, (void *)conn_param->private_data,
+		&cm_info);
 	if (!cm_node) {
-		if (cm_id->local_addr.sin_addr.s_addr != cm_id->remote_addr.sin_addr.s_addr)
+		if (cm_id->local_addr.sin_addr.s_addr !=
+				cm_id->remote_addr.sin_addr.s_addr)
 			nes_manage_apbvt(nesvnic, ntohs(cm_id->local_addr.sin_port),
-					PCI_FUNC(nesdev->pcidev->devfn), NES_MANAGE_APBVT_DEL);
-		nes_rem_ref(&nesqp->ibqp);
-		kfree(nesqp->ietf_frame);
-		nesqp->ietf_frame = NULL;
+				PCI_FUNC(nesdev->pcidev->devfn),
+				NES_MANAGE_APBVT_DEL);
+
 		cm_id->rem_ref(cm_id);
 		return -ENOMEM;
 	}
 
 	cm_node->apbvt_set = 1;
 	nesqp->cm_node = cm_node;
+	cm_node->nesqp = nesqp;
 
 	return 0;
 }
@@ -2664,7 +2878,7 @@ int nes_create_listen(struct iw_cm_id *cm_id, int backlog)
 
 	cm_node = g_cm_core->api->listen(g_cm_core, nesvnic, &cm_info);
 	if (!cm_node) {
-		printk("%s[%u] Error returned from listen API call\n",
+		printk(KERN_ERR "%s[%u] Error returned from listen API call\n",
 				__func__, __LINE__);
 		return -ENOMEM;
 	}
@@ -2672,10 +2886,13 @@ int nes_create_listen(struct iw_cm_id *cm_id, int backlog)
 	cm_id->provider_data = cm_node;
 
 	if (!cm_node->reused_node) {
-		err = nes_manage_apbvt(nesvnic, ntohs(cm_id->local_addr.sin_port),
-				PCI_FUNC(nesvnic->nesdev->pcidev->devfn), NES_MANAGE_APBVT_ADD);
+		err = nes_manage_apbvt(nesvnic,
+			ntohs(cm_id->local_addr.sin_port),
+			PCI_FUNC(nesvnic->nesdev->pcidev->devfn),
+			NES_MANAGE_APBVT_ADD);
 		if (err) {
-			printk("nes_manage_apbvt call returned %d.\n", err);
+			printk(KERN_ERR "nes_manage_apbvt call returned %d.\n",
+				err);
 			g_cm_core->api->stop_listener(g_cm_core, (void *)cm_node);
 			return err;
 		}
@@ -2795,53 +3012,70 @@ static void cm_event_connected(struct nes_cm_event *event)
 	nes_cm_init_tsa_conn(nesqp, cm_node);
 
 	/* set the QP tsa context */
-	nesqp->nesqp_context->tcpPorts[0] = cpu_to_le16(ntohs(cm_id->local_addr.sin_port));
-	nesqp->nesqp_context->tcpPorts[1] = cpu_to_le16(ntohs(cm_id->remote_addr.sin_port));
-	nesqp->nesqp_context->ip0 = cpu_to_le32(ntohl(cm_id->remote_addr.sin_addr.s_addr));
+	nesqp->nesqp_context->tcpPorts[0] =
+		cpu_to_le16(ntohs(cm_id->local_addr.sin_port));
+	nesqp->nesqp_context->tcpPorts[1] =
+		cpu_to_le16(ntohs(cm_id->remote_addr.sin_port));
+	if (ipv4_is_loopback(cm_id->remote_addr.sin_addr.s_addr))
+		nesqp->nesqp_context->ip0 =
+			cpu_to_le32(ntohl(nesvnic->local_ipaddr));
+	else
+		nesqp->nesqp_context->ip0 =
+			cpu_to_le32(ntohl(cm_id->remote_addr.sin_addr.s_addr));
 
 	nesqp->nesqp_context->misc2 |= cpu_to_le32(
-			(u32)PCI_FUNC(nesdev->pcidev->devfn) << NES_QPCONTEXT_MISC2_SRC_IP_SHIFT);
+			(u32)PCI_FUNC(nesdev->pcidev->devfn) <<
+			NES_QPCONTEXT_MISC2_SRC_IP_SHIFT);
 	nesqp->nesqp_context->arp_index_vlan |= cpu_to_le32(
-			nes_arp_table(nesdev, le32_to_cpu(nesqp->nesqp_context->ip0),
+			nes_arp_table(nesdev,
+			le32_to_cpu(nesqp->nesqp_context->ip0),
 			NULL, NES_ARP_RESOLVE) << 16);
 	nesqp->nesqp_context->ts_val_delta = cpu_to_le32(
 			jiffies - nes_read_indexed(nesdev, NES_IDX_TCP_NOW));
 	nesqp->nesqp_context->ird_index = cpu_to_le32(nesqp->hwqp.qp_id);
 	nesqp->nesqp_context->ird_ord_sizes |=
-			cpu_to_le32((u32)1 << NES_QPCONTEXT_ORDIRD_IWARP_MODE_SHIFT);
+			cpu_to_le32((u32)1 <<
+			NES_QPCONTEXT_ORDIRD_IWARP_MODE_SHIFT);
 
 	/* Adjust tail for not having a LSMM */
 	nesqp->hwqp.sq_tail = 1;
 
 #if defined(NES_SEND_FIRST_WRITE)
-		if (cm_node->send_write0) {
-			nes_debug(NES_DBG_CM, "Sending first write.\n");
-			wqe = &nesqp->hwqp.sq_vbase[0];
-			u64temp = (unsigned long)nesqp;
-			u64temp |= NES_SW_CONTEXT_ALIGN>>1;
-			set_wqe_64bit_value(wqe->wqe_words, NES_IWARP_SQ_WQE_COMP_CTX_LOW_IDX,
-					    u64temp);
-			wqe->wqe_words[NES_IWARP_SQ_WQE_MISC_IDX] = cpu_to_le32(NES_IWARP_SQ_OP_RDMAW);
-			wqe->wqe_words[NES_IWARP_SQ_WQE_TOTAL_PAYLOAD_IDX] = 0;
-			wqe->wqe_words[NES_IWARP_SQ_WQE_FRAG0_LOW_IDX] = 0;
-			wqe->wqe_words[NES_IWARP_SQ_WQE_FRAG0_HIGH_IDX] = 0;
-			wqe->wqe_words[NES_IWARP_SQ_WQE_LENGTH0_IDX] = 0;
-			wqe->wqe_words[NES_IWARP_SQ_WQE_STAG0_IDX] = 0;
+	if (cm_node->send_write0) {
+		nes_debug(NES_DBG_CM, "Sending first write.\n");
+		wqe = &nesqp->hwqp.sq_vbase[0];
+		u64temp = (unsigned long)nesqp;
+		u64temp |= NES_SW_CONTEXT_ALIGN>>1;
+		set_wqe_64bit_value(wqe->wqe_words,
+				NES_IWARP_SQ_WQE_COMP_CTX_LOW_IDX, u64temp);
+		wqe->wqe_words[NES_IWARP_SQ_WQE_MISC_IDX] =
+			cpu_to_le32(NES_IWARP_SQ_OP_RDMAW);
+		wqe->wqe_words[NES_IWARP_SQ_WQE_TOTAL_PAYLOAD_IDX] = 0;
+		wqe->wqe_words[NES_IWARP_SQ_WQE_FRAG0_LOW_IDX] = 0;
+		wqe->wqe_words[NES_IWARP_SQ_WQE_FRAG0_HIGH_IDX] = 0;
+		wqe->wqe_words[NES_IWARP_SQ_WQE_LENGTH0_IDX] = 0;
+		wqe->wqe_words[NES_IWARP_SQ_WQE_STAG0_IDX] = 0;
 
-			/* use the reserved spot on the WQ for the extra first WQE */
-			nesqp->nesqp_context->ird_ord_sizes &= cpu_to_le32(~(NES_QPCONTEXT_ORDIRD_LSMM_PRESENT |
-					NES_QPCONTEXT_ORDIRD_WRPDU | NES_QPCONTEXT_ORDIRD_ALSMM));
-			nesqp->skip_lsmm = 1;
-			nesqp->hwqp.sq_tail = 0;
-			nes_write32(nesdev->regs + NES_WQE_ALLOC,
-					(1 << 24) | 0x00800000 | nesqp->hwqp.qp_id);
-		}
+		/* use the reserved spot on the WQ for the extra first WQE */
+		nesqp->nesqp_context->ird_ord_sizes &=
+			cpu_to_le32(~(NES_QPCONTEXT_ORDIRD_LSMM_PRESENT |
+						NES_QPCONTEXT_ORDIRD_WRPDU |
+						NES_QPCONTEXT_ORDIRD_ALSMM));
+		nesqp->skip_lsmm = 1;
+		nesqp->hwqp.sq_tail = 0;
+		nes_write32(nesdev->regs + NES_WQE_ALLOC,
+				(1 << 24) | 0x00800000 | nesqp->hwqp.qp_id);
+	}
 #endif
 
 	memset(&nes_quad, 0, sizeof(nes_quad));
 
-	nes_quad.DstIpAdrIndex = cpu_to_le32((u32)PCI_FUNC(nesdev->pcidev->devfn) << 24);
-	nes_quad.SrcIpadr = cm_id->remote_addr.sin_addr.s_addr;
+	nes_quad.DstIpAdrIndex =
+		cpu_to_le32((u32)PCI_FUNC(nesdev->pcidev->devfn) << 24);
+	if (ipv4_is_loopback(cm_id->remote_addr.sin_addr.s_addr))
+		nes_quad.SrcIpadr = nesvnic->local_ipaddr;
+	else
+		nes_quad.SrcIpadr = cm_id->remote_addr.sin_addr.s_addr;
 	nes_quad.TcpPorts[0] = cm_id->remote_addr.sin_port;
 	nes_quad.TcpPorts[1] = cm_id->local_addr.sin_port;
 
@@ -2858,10 +3092,6 @@ static void cm_event_connected(struct nes_cm_event *event)
 	nesqp->private_data_len = (u8) cm_node->mpa_frame_size;
 	cm_node->cm_core->api->accelerated(cm_node->cm_core, cm_node);
 
-	/* modify QP state to rts */
-	attr.qp_state = IB_QPS_RTS;
-	nes_modify_qp(&nesqp->ibqp, &attr, IB_QP_STATE, NULL);
-
 	/* notify OF layer we successfully created the requested connection */
 	cm_event.event = IW_CM_EVENT_CONNECT_REPLY;
 	cm_event.status = IW_CM_EVENT_STATUS_ACCEPTED;
@@ -2870,20 +3100,21 @@ static void cm_event_connected(struct nes_cm_event *event)
 	cm_event.local_addr.sin_port = cm_id->local_addr.sin_port;
 	cm_event.remote_addr = cm_id->remote_addr;
 
-		cm_event.private_data = (void *)event->cm_node->mpa_frame_buf;
-		cm_event.private_data_len = (u8) event->cm_node->mpa_frame_size;
+	cm_event.private_data = (void *)event->cm_node->mpa_frame_buf;
+	cm_event.private_data_len = (u8) event->cm_node->mpa_frame_size;
 
 	cm_event.local_addr.sin_addr.s_addr = event->cm_info.rem_addr;
 	ret = cm_id->event_handler(cm_id, &cm_event);
 	nes_debug(NES_DBG_CM, "OFA CM event_handler returned, ret=%d\n", ret);
 
 	if (ret)
-		printk("%s[%u] OFA CM event_handler returned, ret=%d\n",
-				__func__, __LINE__, ret);
-	nes_debug(NES_DBG_CM, "Exiting connect thread for QP%u. jiffies = %lu\n",
-			nesqp->hwqp.qp_id, jiffies );
+		printk(KERN_ERR "%s[%u] OFA CM event_handler returned, "
+			"ret=%d\n", __func__, __LINE__, ret);
+	attr.qp_state = IB_QPS_RTS;
+	nes_modify_qp(&nesqp->ibqp, &attr, IB_QP_STATE, NULL);
 
-	nes_rem_ref(&nesqp->ibqp);
+	nes_debug(NES_DBG_CM, "Exiting connect thread for QP%u. jiffies = "
+		"%lu\n", nesqp->hwqp.qp_id, jiffies);
 
 	return;
 }
@@ -2927,17 +3158,19 @@ static void cm_event_connect_error(struct nes_cm_event *event)
 	cm_event.private_data = NULL;
 	cm_event.private_data_len = 0;
 
-	nes_debug(NES_DBG_CM, "call CM_EVENT REJECTED, local_addr=%08x, remove_addr=%08x\n",
-			cm_event.local_addr.sin_addr.s_addr, cm_event.remote_addr.sin_addr.s_addr);
+	nes_debug(NES_DBG_CM, "call CM_EVENT REJECTED, local_addr=%08x, "
+		"remove_addr=%08x\n", cm_event.local_addr.sin_addr.s_addr,
+		cm_event.remote_addr.sin_addr.s_addr);
 
 	ret = cm_id->event_handler(cm_id, &cm_event);
 	nes_debug(NES_DBG_CM, "OFA CM event_handler returned, ret=%d\n", ret);
 	if (ret)
-		printk("%s[%u] OFA CM event_handler returned, ret=%d\n",
-				__func__, __LINE__, ret);
+		printk(KERN_ERR "%s[%u] OFA CM event_handler returned, "
+			"ret=%d\n", __func__, __LINE__, ret);
 	nes_rem_ref(&nesqp->ibqp);
-		cm_id->rem_ref(cm_id);
+	cm_id->rem_ref(cm_id);
 
+	rem_ref_cm_node(event->cm_node->cm_core, event->cm_node);
 	return;
 }
 
@@ -3040,7 +3273,8 @@ static int nes_cm_post_event(struct nes_cm_event *event)
 	add_ref_cm_node(event->cm_node);
 	event->cm_info.cm_id->add_ref(event->cm_info.cm_id);
 	INIT_WORK(&event->event_work, nes_cm_event_handler);
-	nes_debug(NES_DBG_CM, "queue_work, event=%p\n", event);
+	nes_debug(NES_DBG_CM, "cm_node=%p queue_work, event=%p\n",
+		event->cm_node, event);
 
 	queue_work(event->cm_node->cm_core->event_wq, &event->event_work);
 
@@ -3056,46 +3290,48 @@ static int nes_cm_post_event(struct nes_cm_event *event)
  */
 static void nes_cm_event_handler(struct work_struct *work)
 {
-	struct nes_cm_event *event = container_of(work, struct nes_cm_event, event_work);
+	struct nes_cm_event *event = container_of(work, struct nes_cm_event,
+			event_work);
 	struct nes_cm_core *cm_core;
 
-	if ((!event) || (!event->cm_node) || (!event->cm_node->cm_core)) {
+	if ((!event) || (!event->cm_node) || (!event->cm_node->cm_core))
 		return;
-	}
+
 	cm_core = event->cm_node->cm_core;
 	nes_debug(NES_DBG_CM, "event=%p, event->type=%u, events posted=%u\n",
-			event, event->type, atomic_read(&cm_core->events_posted));
+		event, event->type, atomic_read(&cm_core->events_posted));
 
 	switch (event->type) {
-		case NES_CM_EVENT_MPA_REQ:
-			cm_event_mpa_req(event);
-			nes_debug(NES_DBG_CM, "CM Event: MPA REQUEST\n");
+	case NES_CM_EVENT_MPA_REQ:
+		cm_event_mpa_req(event);
+		nes_debug(NES_DBG_CM, "cm_node=%p CM Event: MPA REQUEST\n",
+			event->cm_node);
+		break;
+	case NES_CM_EVENT_RESET:
+		nes_debug(NES_DBG_CM, "cm_node = %p CM Event: RESET\n",
+			event->cm_node);
+		cm_event_reset(event);
+		break;
+	case NES_CM_EVENT_CONNECTED:
+		if ((!event->cm_node->cm_id) ||
+			(event->cm_node->state != NES_CM_STATE_TSA))
 			break;
-		case NES_CM_EVENT_RESET:
-			nes_debug(NES_DBG_CM, "CM Event: RESET\n");
-			cm_event_reset(event);
-			break;
-		case NES_CM_EVENT_CONNECTED:
-			if ((!event->cm_node->cm_id) ||
-				(event->cm_node->state != NES_CM_STATE_TSA)) {
-				break;
-			}
-			cm_event_connected(event);
-			nes_debug(NES_DBG_CM, "CM Event: CONNECTED\n");
-			break;
-		case NES_CM_EVENT_ABORTED:
-			if ((!event->cm_node->cm_id) || (event->cm_node->state == NES_CM_STATE_TSA)) {
-				break;
-			}
-			cm_event_connect_error(event);
-			nes_debug(NES_DBG_CM, "CM Event: ABORTED\n");
-			break;
-		case NES_CM_EVENT_DROPPED_PKT:
-			nes_debug(NES_DBG_CM, "CM Event: DROPPED PKT\n");
-			break;
-		default:
-			nes_debug(NES_DBG_CM, "CM Event: UNKNOWN EVENT TYPE\n");
+		cm_event_connected(event);
+		nes_debug(NES_DBG_CM, "CM Event: CONNECTED\n");
+		break;
+	case NES_CM_EVENT_ABORTED:
+		if ((!event->cm_node->cm_id) ||
+			(event->cm_node->state == NES_CM_STATE_TSA))
 			break;
+		cm_event_connect_error(event);
+		nes_debug(NES_DBG_CM, "CM Event: ABORTED\n");
+		break;
+	case NES_CM_EVENT_DROPPED_PKT:
+		nes_debug(NES_DBG_CM, "CM Event: DROPPED PKT\n");
+		break;
+	default:
+		nes_debug(NES_DBG_CM, "CM Event: UNKNOWN EVENT TYPE\n");
+		break;
 	}
 
 	atomic_dec(&cm_core->events_posted);
diff --git a/drivers/infiniband/hw/nes/nes_cm.h b/drivers/infiniband/hw/nes/nes_cm.h
index 7717cb2ab500..367b3d290140 100644
--- a/drivers/infiniband/hw/nes/nes_cm.h
+++ b/drivers/infiniband/hw/nes/nes_cm.h
@@ -83,6 +83,8 @@ enum nes_timer_type {
 #define SET_FIN 4
 #define SET_RST 8
 
+#define TCP_OPTIONS_PADDING	3
+
 struct option_base {
 	u8 optionnum;
 	u8 length;
@@ -177,6 +179,7 @@ enum nes_cm_node_state {
 	NES_CM_STATE_ESTABLISHED,
 	NES_CM_STATE_ACCEPTING,
 	NES_CM_STATE_MPAREQ_SENT,
+	NES_CM_STATE_MPAREQ_RCVD,
 	NES_CM_STATE_TSA,
 	NES_CM_STATE_FIN_WAIT1,
 	NES_CM_STATE_FIN_WAIT2,
@@ -187,6 +190,16 @@ enum nes_cm_node_state {
 	NES_CM_STATE_CLOSED
 };
 
+enum nes_tcpip_pkt_type {
+	NES_PKT_TYPE_UNKNOWN,
+	NES_PKT_TYPE_SYN,
+	NES_PKT_TYPE_SYNACK,
+	NES_PKT_TYPE_ACK,
+	NES_PKT_TYPE_FIN,
+	NES_PKT_TYPE_RST
+};
+
+
 /* type of nes connection */
 enum nes_cm_conn_type {
 	NES_CM_IWARP_CONN_TYPE,
@@ -257,7 +270,9 @@ struct nes_cm_node {
 	struct net_device         *netdev;
 
 	struct nes_cm_node        *loopbackpartner;
-	struct list_head          retrans_list;
+
+	struct nes_timer_entry	*send_entry;
+
 	spinlock_t                retrans_list_lock;
 	struct list_head          recv_list;
 	spinlock_t                recv_list_lock;
@@ -276,6 +291,8 @@ struct nes_cm_node {
 	struct nes_vnic           *nesvnic;
 	int                       apbvt_set;
 	int                       accept_pend;
+	int			freed;
+	struct nes_qp		*nesqp;
 };
 
 /* structure for client or CM to fill when making CM api calls. */
@@ -366,14 +383,14 @@ struct nes_cm_ops {
 			struct nes_cm_info *);
 	int (*stop_listener)(struct nes_cm_core *, struct nes_cm_listener *);
 	struct nes_cm_node * (*connect)(struct nes_cm_core *,
-			struct nes_vnic *, struct ietf_mpa_frame *,
+			struct nes_vnic *, u16, void *,
 			struct nes_cm_info *);
 	int (*close)(struct nes_cm_core *, struct nes_cm_node *);
 	int (*accept)(struct nes_cm_core *, struct ietf_mpa_frame *,
 			struct nes_cm_node *);
 	int (*reject)(struct nes_cm_core *, struct ietf_mpa_frame *,
 			struct nes_cm_node *);
-	int (*recv_pkt)(struct nes_cm_core *, struct nes_vnic *,
+	void (*recv_pkt)(struct nes_cm_core *, struct nes_vnic *,
 			struct sk_buff *);
 	int (*destroy_cm_core)(struct nes_cm_core *);
 	int (*get)(struct nes_cm_core *);
diff --git a/drivers/infiniband/hw/nes/nes_hw.c b/drivers/infiniband/hw/nes/nes_hw.c
index 85f26d19a32b..1513d4066f1b 100644
--- a/drivers/infiniband/hw/nes/nes_hw.c
+++ b/drivers/infiniband/hw/nes/nes_hw.c
@@ -2814,7 +2814,6 @@ static void nes_process_iwarp_aeqe(struct nes_device *nesdev,
 			nesqp = *((struct nes_qp **)&context);
 			if (atomic_inc_return(&nesqp->close_timer_started) == 1) {
 				nesqp->cm_id->add_ref(nesqp->cm_id);
-				nes_add_ref(&nesqp->ibqp);
 				schedule_nes_timer(nesqp->cm_node, (struct sk_buff *)nesqp,
 						NES_TIMER_TYPE_CLOSE, 1, 0);
 				nes_debug(NES_DBG_AEQ, "QP%u Not decrementing QP refcount (%d),"
@@ -2838,7 +2837,6 @@ static void nes_process_iwarp_aeqe(struct nes_device *nesdev,
 			if (async_event_id == NES_AEQE_AEID_RESET_SENT) {
 				tcp_state = NES_AEQE_TCP_STATE_CLOSED;
 			}
-			nes_add_ref(&nesqp->ibqp);
 			spin_lock_irqsave(&nesqp->lock, flags);
 			nesqp->hw_iwarp_state = iwarp_state;
 			nesqp->hw_tcp_state = tcp_state;
@@ -2876,7 +2874,6 @@ static void nes_process_iwarp_aeqe(struct nes_device *nesdev,
 				}
 				spin_unlock_irqrestore(&nesqp->lock, flags);
 				if (next_iwarp_state) {
-					nes_add_ref(&nesqp->ibqp);
 					nes_debug(NES_DBG_AEQ, "issuing hw modifyqp for QP%u. next state = 0x%08X,"
 							" also added another reference\n",
 							nesqp->hwqp.qp_id, next_iwarp_state);
@@ -2888,7 +2885,6 @@ static void nes_process_iwarp_aeqe(struct nes_device *nesdev,
 					/* FIN Received but ib state not RTS,
 							close complete will be on its way */
 					spin_unlock_irqrestore(&nesqp->lock, flags);
-					nes_rem_ref(&nesqp->ibqp);
 					return;
 				}
 				spin_unlock_irqrestore(&nesqp->lock, flags);
@@ -2922,7 +2918,6 @@ static void nes_process_iwarp_aeqe(struct nes_device *nesdev,
 			if ((tcp_state == NES_AEQE_TCP_STATE_CLOSE_WAIT) ||
 					((nesqp->ibqp_state == IB_QPS_RTS)&&
 					(async_event_id == NES_AEQE_AEID_LLP_CONNECTION_RESET))) {
-				nes_add_ref(&nesqp->ibqp);
 				nes_cm_disconn(nesqp);
 			} else {
 				nesqp->in_disconnect = 0;
@@ -2931,7 +2926,6 @@ static void nes_process_iwarp_aeqe(struct nes_device *nesdev,
 			break;
 		case NES_AEQE_AEID_LLP_TOO_MANY_RETRIES:
 			nesqp = *((struct nes_qp **)&context);
-			nes_add_ref(&nesqp->ibqp);
 			spin_lock_irqsave(&nesqp->lock, flags);
 			nesqp->hw_iwarp_state = NES_AEQE_IWARP_STATE_ERROR;
 			nesqp->hw_tcp_state = NES_AEQE_TCP_STATE_CLOSED;
@@ -3042,7 +3036,6 @@ static void nes_process_iwarp_aeqe(struct nes_device *nesdev,
 				nesqp->ibqp.event_handler(&ibevent, nesqp->ibqp.qp_context);
 			}
 			/* tell cm to disconnect, cm will queue work to thread */
-			nes_add_ref(&nesqp->ibqp);
 			nes_cm_disconn(nesqp);
 			break;
 		case NES_AEQE_AEID_DDP_UBE_INVALID_MSN_NO_BUFFER_AVAILABLE:
@@ -3062,7 +3055,6 @@ static void nes_process_iwarp_aeqe(struct nes_device *nesdev,
 				nesqp->ibqp.event_handler(&ibevent, nesqp->ibqp.qp_context);
 			}
 			/* tell cm to disconnect, cm will queue work to thread */
-			nes_add_ref(&nesqp->ibqp);
 			nes_cm_disconn(nesqp);
 			break;
 		case NES_AEQE_AEID_LLP_RECEIVED_MPA_CRC_ERROR:
@@ -3082,7 +3074,6 @@ static void nes_process_iwarp_aeqe(struct nes_device *nesdev,
 				nesqp->ibqp.event_handler(&ibevent, nesqp->ibqp.qp_context);
 			}
 			/* tell cm to disconnect, cm will queue work to thread */
-			nes_add_ref(&nesqp->ibqp);
 			nes_cm_disconn(nesqp);
 			break;
 			/* TODO: additional AEs need to be here */
diff --git a/drivers/infiniband/hw/nes/nes_verbs.c b/drivers/infiniband/hw/nes/nes_verbs.c
index e3939d13484e..d79942e84979 100644
--- a/drivers/infiniband/hw/nes/nes_verbs.c
+++ b/drivers/infiniband/hw/nes/nes_verbs.c
@@ -2867,7 +2867,6 @@ int nes_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 			nesqp->hwqp.qp_id, attr->qp_state, nesqp->ibqp_state,
 			nesqp->iwarp_state, atomic_read(&nesqp->refcount));
 
-	nes_add_ref(&nesqp->ibqp);
 	spin_lock_irqsave(&nesqp->lock, qplockflags);
 
 	nes_debug(NES_DBG_MOD_QP, "QP%u: hw_iwarp_state=0x%X, hw_tcp_state=0x%X,"
@@ -2882,7 +2881,6 @@ int nes_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 						nesqp->hwqp.qp_id);
 				if (nesqp->iwarp_state > (u32)NES_CQP_QP_IWARP_STATE_IDLE) {
 					spin_unlock_irqrestore(&nesqp->lock, qplockflags);
-					nes_rem_ref(&nesqp->ibqp);
 					return -EINVAL;
 				}
 				next_iwarp_state = NES_CQP_QP_IWARP_STATE_IDLE;
@@ -2893,7 +2891,6 @@ int nes_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 						nesqp->hwqp.qp_id);
 				if (nesqp->iwarp_state>(u32)NES_CQP_QP_IWARP_STATE_IDLE) {
 					spin_unlock_irqrestore(&nesqp->lock, qplockflags);
-					nes_rem_ref(&nesqp->ibqp);
 					return -EINVAL;
 				}
 				next_iwarp_state = NES_CQP_QP_IWARP_STATE_IDLE;
@@ -2904,14 +2901,12 @@ int nes_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 						nesqp->hwqp.qp_id);
 				if (nesqp->iwarp_state>(u32)NES_CQP_QP_IWARP_STATE_RTS) {
 					spin_unlock_irqrestore(&nesqp->lock, qplockflags);
-					nes_rem_ref(&nesqp->ibqp);
 					return -EINVAL;
 				}
 				if (nesqp->cm_id == NULL) {
 					nes_debug(NES_DBG_MOD_QP, "QP%u: Failing attempt to move QP to RTS without a CM_ID. \n",
 							nesqp->hwqp.qp_id );
 					spin_unlock_irqrestore(&nesqp->lock, qplockflags);
-					nes_rem_ref(&nesqp->ibqp);
 					return -EINVAL;
 				}
 				next_iwarp_state = NES_CQP_QP_IWARP_STATE_RTS;
@@ -2929,7 +2924,6 @@ int nes_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 						nesqp->hwqp.qp_id, nesqp->hwqp.sq_head, nesqp->hwqp.sq_tail);
 				if (nesqp->iwarp_state == (u32)NES_CQP_QP_IWARP_STATE_CLOSING) {
 					spin_unlock_irqrestore(&nesqp->lock, qplockflags);
-					nes_rem_ref(&nesqp->ibqp);
 					return 0;
 				} else {
 					if (nesqp->iwarp_state > (u32)NES_CQP_QP_IWARP_STATE_CLOSING) {
@@ -2937,7 +2931,6 @@ int nes_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 								" ignored due to current iWARP state\n",
 								nesqp->hwqp.qp_id);
 						spin_unlock_irqrestore(&nesqp->lock, qplockflags);
-						nes_rem_ref(&nesqp->ibqp);
 						return -EINVAL;
 					}
 					if (nesqp->hw_iwarp_state != NES_AEQE_IWARP_STATE_RTS) {
@@ -2969,7 +2962,6 @@ int nes_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 						nesqp->hwqp.qp_id);
 				if (nesqp->iwarp_state>=(u32)NES_CQP_QP_IWARP_STATE_TERMINATE) {
 					spin_unlock_irqrestore(&nesqp->lock, qplockflags);
-					nes_rem_ref(&nesqp->ibqp);
 					return -EINVAL;
 				}
 				/* next_iwarp_state = (NES_CQP_QP_IWARP_STATE_TERMINATE | 0x02000000); */
@@ -2982,7 +2974,6 @@ int nes_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 			case IB_QPS_RESET:
 				if (nesqp->iwarp_state == (u32)NES_CQP_QP_IWARP_STATE_ERROR) {
 					spin_unlock_irqrestore(&nesqp->lock, qplockflags);
-					nes_rem_ref(&nesqp->ibqp);
 					return -EINVAL;
 				}
 				nes_debug(NES_DBG_MOD_QP, "QP%u: new state = error\n",
@@ -3008,7 +2999,6 @@ int nes_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 				break;
 			default:
 				spin_unlock_irqrestore(&nesqp->lock, qplockflags);
-				nes_rem_ref(&nesqp->ibqp);
 				return -EINVAL;
 				break;
 		}
@@ -3088,7 +3078,6 @@ int nes_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 							nesqp->hwqp.qp_id, atomic_read(&nesqp->refcount),
 							original_last_aeq, nesqp->last_aeq);
 					/* this one is for the cm_disconnect thread */
-					nes_add_ref(&nesqp->ibqp);
 					spin_lock_irqsave(&nesqp->lock, qplockflags);
 					nesqp->hw_tcp_state = NES_AEQE_TCP_STATE_CLOSED;
 					nesqp->last_aeq = NES_AEQE_AEID_RESET_SENT;
@@ -3097,14 +3086,12 @@ int nes_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 				} else {
 					nes_debug(NES_DBG_MOD_QP, "QP%u No fake disconnect, QP refcount=%d\n",
 							nesqp->hwqp.qp_id, atomic_read(&nesqp->refcount));
-					nes_rem_ref(&nesqp->ibqp);
 				}
 			} else {
 				spin_lock_irqsave(&nesqp->lock, qplockflags);
 				if (nesqp->cm_id) {
 					/* These two are for the timer thread */
 					if (atomic_inc_return(&nesqp->close_timer_started) == 1) {
-						nes_add_ref(&nesqp->ibqp);
 						nesqp->cm_id->add_ref(nesqp->cm_id);
 						nes_debug(NES_DBG_MOD_QP, "QP%u Not decrementing QP refcount (%d),"
 								" need ae to finish up, original_last_aeq = 0x%04X."
@@ -3128,14 +3115,12 @@ int nes_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 					" original_last_aeq = 0x%04X. last_aeq = 0x%04X.\n",
 					nesqp->hwqp.qp_id, atomic_read(&nesqp->refcount),
 					original_last_aeq, nesqp->last_aeq);
-			nes_rem_ref(&nesqp->ibqp);
 		}
 	} else {
 		nes_debug(NES_DBG_MOD_QP, "QP%u Decrementing QP refcount (%d), No ae to finish up,"
 				" original_last_aeq = 0x%04X. last_aeq = 0x%04X.\n",
 				nesqp->hwqp.qp_id, atomic_read(&nesqp->refcount),
 				original_last_aeq, nesqp->last_aeq);
-		nes_rem_ref(&nesqp->ibqp);
 	}
 
 	err = 0;

From 483fad1c3fa1060d7e6710e84a065ad514571739 Mon Sep 17 00:00:00 2001
From: Nathan Lynch <ntl@pobox.com>
Date: Tue, 22 Jul 2008 04:48:46 +1000
Subject: [PATCH 0626/1171] ELF loader support for auxvec base platform string

Some IBM POWER-based platforms have the ability to run in a
mode which mostly appears to the OS as a different processor from the
actual hardware.  For example, a Power6 system may appear to be a
Power5+, which makes the AT_PLATFORM value "power5+".  This means that
programs are restricted to the ISA supported by Power5+;
Power6-specific instructions are treated as illegal.

However, some applications (virtual machines, optimized libraries) can
benefit from knowledge of the underlying CPU model.  A new aux vector
entry, AT_BASE_PLATFORM, will denote the actual hardware.  For
example, on a Power6 system in Power5+ compatibility mode, AT_PLATFORM
will be "power5+" and AT_BASE_PLATFORM will be "power6".  The idea is
that AT_PLATFORM indicates the instruction set supported, while
AT_BASE_PLATFORM indicates the underlying microarchitecture.

If the architecture has defined ELF_BASE_PLATFORM, copy that value to
the user stack in the same manner as ELF_PLATFORM.

Signed-off-by: Nathan Lynch <ntl@pobox.com>
Acked-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 fs/binfmt_elf.c        | 28 ++++++++++++++++++++++++++++
 include/linux/auxvec.h |  6 +++++-
 2 files changed, 33 insertions(+), 1 deletion(-)

diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index 639d2d8b5710..742c8f530481 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -131,6 +131,15 @@ static int padzero(unsigned long elf_bss)
 #define STACK_ALLOC(sp, len) ({ sp -= len ; sp; })
 #endif
 
+#ifndef ELF_BASE_PLATFORM
+/*
+ * AT_BASE_PLATFORM indicates the "real" hardware/microarchitecture.
+ * If the arch defines ELF_BASE_PLATFORM (in asm/elf.h), the value
+ * will be copied to the user stack in the same manner as AT_PLATFORM.
+ */
+#define ELF_BASE_PLATFORM NULL
+#endif
+
 static int
 create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec,
 		unsigned long load_addr, unsigned long interp_load_addr)
@@ -142,7 +151,9 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec,
 	elf_addr_t __user *envp;
 	elf_addr_t __user *sp;
 	elf_addr_t __user *u_platform;
+	elf_addr_t __user *u_base_platform;
 	const char *k_platform = ELF_PLATFORM;
+	const char *k_base_platform = ELF_BASE_PLATFORM;
 	int items;
 	elf_addr_t *elf_info;
 	int ei_index = 0;
@@ -172,6 +183,19 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec,
 			return -EFAULT;
 	}
 
+	/*
+	 * If this architecture has a "base" platform capability
+	 * string, copy it to userspace.
+	 */
+	u_base_platform = NULL;
+	if (k_base_platform) {
+		size_t len = strlen(k_base_platform) + 1;
+
+		u_base_platform = (elf_addr_t __user *)STACK_ALLOC(p, len);
+		if (__copy_to_user(u_base_platform, k_base_platform, len))
+			return -EFAULT;
+	}
+
 	/* Create the ELF interpreter info */
 	elf_info = (elf_addr_t *)current->mm->saved_auxv;
 	/* update AT_VECTOR_SIZE_BASE if the number of NEW_AUX_ENT() changes */
@@ -209,6 +233,10 @@ create_elf_tables(struct linux_binprm *bprm, struct elfhdr *exec,
 		NEW_AUX_ENT(AT_PLATFORM,
 			    (elf_addr_t)(unsigned long)u_platform);
 	}
+	if (k_base_platform) {
+		NEW_AUX_ENT(AT_BASE_PLATFORM,
+			    (elf_addr_t)(unsigned long)u_base_platform);
+	}
 	if (bprm->interp_flags & BINPRM_FLAGS_EXECFD) {
 		NEW_AUX_ENT(AT_EXECFD, bprm->interp_data);
 	}
diff --git a/include/linux/auxvec.h b/include/linux/auxvec.h
index 0da17d14fd13..d7afa9dd6635 100644
--- a/include/linux/auxvec.h
+++ b/include/linux/auxvec.h
@@ -26,9 +26,13 @@
 
 #define AT_SECURE 23   /* secure mode boolean */
 
+#define AT_BASE_PLATFORM 24	/* string identifying real platform, may
+				 * differ from AT_PLATFORM. */
+
 #define AT_EXECFN  31	/* filename of program */
+
 #ifdef __KERNEL__
-#define AT_VECTOR_SIZE_BASE 17 /* NEW_AUX_ENT entries in auxiliary table */
+#define AT_VECTOR_SIZE_BASE 18 /* NEW_AUX_ENT entries in auxiliary table */
   /* number of "#define AT_.*" above, minus {AT_NULL, AT_IGNORE, AT_NOTELF} */
 #endif
 

From 9115d13453dee22473a1e8cacc90a8d64a9c4bc9 Mon Sep 17 00:00:00 2001
From: Nathan Lynch <ntl@pobox.com>
Date: Wed, 16 Jul 2008 09:58:51 +1000
Subject: [PATCH 0627/1171] powerpc: Enable AT_BASE_PLATFORM aux vector

Stash the first platform string matched by identify_cpu() in
powerpc_base_platform, and supply that to the ELF loader for the value
of AT_BASE_PLATFORM.

Signed-off-by: Nathan Lynch <ntl@pobox.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/cputable.c | 11 +++++++++++
 include/asm-powerpc/cputable.h |  2 ++
 include/asm-powerpc/elf.h      |  8 ++++++++
 3 files changed, 21 insertions(+)

diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c
index b936a1dd0a50..25a052c16754 100644
--- a/arch/powerpc/kernel/cputable.c
+++ b/arch/powerpc/kernel/cputable.c
@@ -23,6 +23,9 @@
 struct cpu_spec* cur_cpu_spec = NULL;
 EXPORT_SYMBOL(cur_cpu_spec);
 
+/* The platform string corresponding to the real PVR */
+const char *powerpc_base_platform;
+
 /* NOTE:
  * Unlike ppc32, ppc64 will only call this once for the boot CPU, it's
  * the responsibility of the appropriate CPU save/restore functions to
@@ -1652,6 +1655,14 @@ struct cpu_spec * __init identify_cpu(unsigned long offset, unsigned int pvr)
 			} else
 				*t = *s;
 			*PTRRELOC(&cur_cpu_spec) = &the_cpu_spec;
+
+			/*
+			 * Set the base platform string once; assumes
+			 * we're called with real pvr first.
+			 */
+			if (powerpc_base_platform == NULL)
+				powerpc_base_platform = t->platform;
+
 #if defined(CONFIG_PPC64) || defined(CONFIG_BOOKE)
 			/* ppc64 and booke expect identify_cpu to also call
 			 * setup_cpu for that processor. I will consolidate
diff --git a/include/asm-powerpc/cputable.h b/include/asm-powerpc/cputable.h
index 2a3e9075a5a0..ef8a248dfd55 100644
--- a/include/asm-powerpc/cputable.h
+++ b/include/asm-powerpc/cputable.h
@@ -127,6 +127,8 @@ extern struct cpu_spec *identify_cpu(unsigned long offset, unsigned int pvr);
 extern void do_feature_fixups(unsigned long value, void *fixup_start,
 			      void *fixup_end);
 
+extern const char *powerpc_base_platform;
+
 #endif /* __ASSEMBLY__ */
 
 /* CPU kernel features */
diff --git a/include/asm-powerpc/elf.h b/include/asm-powerpc/elf.h
index 89664675b469..80d1f399ee51 100644
--- a/include/asm-powerpc/elf.h
+++ b/include/asm-powerpc/elf.h
@@ -217,6 +217,14 @@ typedef elf_vrregset_t elf_fpxregset_t;
 
 #define ELF_PLATFORM	(cur_cpu_spec->platform)
 
+/* While ELF_PLATFORM indicates the ISA supported by the platform, it
+ * may not accurately reflect the underlying behavior of the hardware
+ * (as in the case of running in Power5+ compatibility mode on a
+ * Power6 machine).  ELF_BASE_PLATFORM allows ld.so to load libraries
+ * that are tuned for the real hardware.
+ */
+#define ELF_BASE_PLATFORM (powerpc_base_platform)
+
 #ifdef __powerpc64__
 # define ELF_PLAT_INIT(_r, load_addr)	do {	\
 	_r->gpr[2] = load_addr; 		\

From 00bf6e906156b07cd641fe154ad0efe78f989692 Mon Sep 17 00:00:00 2001
From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Wed, 23 Jul 2008 10:44:58 +1000
Subject: [PATCH 0628/1171] powerpc: Fallout from sysdev API changes

A struct sysdev_attribute * parameter was added to the show routine by
commit 4a0b2b4dbe1335b8b9886ba3dc85a145d5d938ed "sysdev: Pass the
attribute to the low level sysdev show/store function".

This eliminates a warning:

arch/powerpc/kernel/sysfs.c:538: warning: initialization from incompatible pointer type

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/sysfs.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/powerpc/kernel/sysfs.c b/arch/powerpc/kernel/sysfs.c
index aba0ba95f062..800e5e9a087b 100644
--- a/arch/powerpc/kernel/sysfs.c
+++ b/arch/powerpc/kernel/sysfs.c
@@ -529,7 +529,8 @@ static void register_nodes(void)
 #endif
 
 /* Only valid if CPU is present. */
-static ssize_t show_physical_id(struct sys_device *dev, char *buf)
+static ssize_t show_physical_id(struct sys_device *dev,
+				struct sysdev_attribute *attr, char *buf)
 {
 	struct cpu *cpu = container_of(dev, struct cpu, sysdev);
 

From d6a61bfc06d6f2248f3e75f208d64e794082013c Mon Sep 17 00:00:00 2001
From: Luis Machado <luisgpm@linux.vnet.ibm.com>
Date: Thu, 24 Jul 2008 02:10:41 +1000
Subject: [PATCH 0629/1171] powerpc: BookE hardware watchpoint support

This patch implements support for HW based watchpoint via the
DBSR_DAC (Data Address Compare) facility of the BookE processors.

It does so by interfacing with the existing DABR breakpoint code
and adding the necessary bits and pieces for the new bits to
be properly set or cleared

Signed-off-by: Luis Machado <luisgpm@br.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/entry_32.S |  6 +--
 arch/powerpc/kernel/process.c  | 46 ++++++++++++++++++++++
 arch/powerpc/kernel/ptrace.c   | 72 +++++++++++++++++++++++++++++++---
 arch/powerpc/kernel/signal.c   |  6 ++-
 arch/powerpc/kernel/traps.c    | 16 ++++++++
 arch/powerpc/mm/fault.c        | 25 ------------
 include/asm-powerpc/system.h   |  2 +
 7 files changed, 138 insertions(+), 35 deletions(-)

diff --git a/arch/powerpc/kernel/entry_32.S b/arch/powerpc/kernel/entry_32.S
index da52269aec1e..81c8324a4a3c 100644
--- a/arch/powerpc/kernel/entry_32.S
+++ b/arch/powerpc/kernel/entry_32.S
@@ -148,7 +148,7 @@ transfer_to_handler:
 	/* Check to see if the dbcr0 register is set up to debug.  Use the
 	   internal debug mode bit to do this. */
 	lwz	r12,THREAD_DBCR0(r12)
-	andis.	r12,r12,DBCR0_IDM@h
+	andis.	r12,r12,(DBCR0_IDM  | DBSR_DAC1R | DBSR_DAC1W)@h
 	beq+	3f
 	/* From user and task is ptraced - load up global dbcr0 */
 	li	r12,-1			/* clear all pending debug events */
@@ -292,7 +292,7 @@ syscall_exit_cont:
 	/* If the process has its own DBCR0 value, load it up.  The internal
 	   debug mode bit tells us that dbcr0 should be loaded. */
 	lwz	r0,THREAD+THREAD_DBCR0(r2)
-	andis.	r10,r0,DBCR0_IDM@h
+	andis.	r10,r0,(DBCR0_IDM  | DBSR_DAC1R | DBSR_DAC1W)@h
 	bnel-	load_dbcr0
 #endif
 #ifdef CONFIG_44x
@@ -720,7 +720,7 @@ restore_user:
 	/* Check whether this process has its own DBCR0 value.  The internal
 	   debug mode bit tells us that dbcr0 should be loaded. */
 	lwz	r0,THREAD+THREAD_DBCR0(r2)
-	andis.	r10,r0,DBCR0_IDM@h
+	andis.	r10,r0,(DBCR0_IDM  | DBSR_DAC1R | DBSR_DAC1W)@h
 	bnel-	load_dbcr0
 #endif
 
diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c
index 219f3634115e..db2497ccc111 100644
--- a/arch/powerpc/kernel/process.c
+++ b/arch/powerpc/kernel/process.c
@@ -47,6 +47,8 @@
 #ifdef CONFIG_PPC64
 #include <asm/firmware.h>
 #endif
+#include <linux/kprobes.h>
+#include <linux/kdebug.h>
 
 extern unsigned long _get_SP(void);
 
@@ -239,6 +241,35 @@ void discard_lazy_cpu_state(void)
 }
 #endif /* CONFIG_SMP */
 
+void do_dabr(struct pt_regs *regs, unsigned long address,
+		    unsigned long error_code)
+{
+	siginfo_t info;
+
+	if (notify_die(DIE_DABR_MATCH, "dabr_match", regs, error_code,
+			11, SIGSEGV) == NOTIFY_STOP)
+		return;
+
+	if (debugger_dabr_match(regs))
+		return;
+
+	/* Clear the DAC and struct entries.  One shot trigger */
+#if (defined(CONFIG_44x) || defined(CONFIG_BOOKE))
+	mtspr(SPRN_DBCR0, mfspr(SPRN_DBCR0) & ~(DBSR_DAC1R | DBSR_DAC1W
+							| DBCR0_IDM));
+#endif
+
+	/* Clear the DABR */
+	set_dabr(0);
+
+	/* Deliver the signal to userspace */
+	info.si_signo = SIGTRAP;
+	info.si_errno = 0;
+	info.si_code = TRAP_HWBKPT;
+	info.si_addr = (void __user *)address;
+	force_sig_info(SIGTRAP, &info, current);
+}
+
 static DEFINE_PER_CPU(unsigned long, current_dabr);
 
 int set_dabr(unsigned long dabr)
@@ -254,6 +285,11 @@ int set_dabr(unsigned long dabr)
 #if defined(CONFIG_PPC64) || defined(CONFIG_6xx)
 	mtspr(SPRN_DABR, dabr);
 #endif
+
+#if defined(CONFIG_44x) || defined(CONFIG_BOOKE)
+	mtspr(SPRN_DAC1, dabr);
+#endif
+
 	return 0;
 }
 
@@ -337,6 +373,12 @@ struct task_struct *__switch_to(struct task_struct *prev,
 	if (unlikely(__get_cpu_var(current_dabr) != new->thread.dabr))
 		set_dabr(new->thread.dabr);
 
+#if defined(CONFIG_44x) || defined(CONFIG_BOOKE)
+	/* If new thread DAC (HW breakpoint) is the same then leave it */
+	if (new->thread.dabr)
+		set_dabr(new->thread.dabr);
+#endif
+
 	new_thread = &new->thread;
 	old_thread = &current->thread;
 
@@ -525,6 +567,10 @@ void flush_thread(void)
 	if (current->thread.dabr) {
 		current->thread.dabr = 0;
 		set_dabr(0);
+
+#if defined(CONFIG_44x) || defined(CONFIG_BOOKE)
+		current->thread.dbcr0 &= ~(DBSR_DAC1R | DBSR_DAC1W);
+#endif
 	}
 }
 
diff --git a/arch/powerpc/kernel/ptrace.c b/arch/powerpc/kernel/ptrace.c
index 8feb93e7890c..a5d0e78779c8 100644
--- a/arch/powerpc/kernel/ptrace.c
+++ b/arch/powerpc/kernel/ptrace.c
@@ -703,7 +703,7 @@ void user_enable_single_step(struct task_struct *task)
 
 	if (regs != NULL) {
 #if defined(CONFIG_40x) || defined(CONFIG_BOOKE)
-		task->thread.dbcr0 = DBCR0_IDM | DBCR0_IC;
+		task->thread.dbcr0 |= DBCR0_IDM | DBCR0_IC;
 		regs->msr |= MSR_DE;
 #else
 		regs->msr |= MSR_SE;
@@ -716,9 +716,16 @@ void user_disable_single_step(struct task_struct *task)
 {
 	struct pt_regs *regs = task->thread.regs;
 
+
+#if defined(CONFIG_44x) || defined(CONFIG_BOOKE)
+	/* If DAC then do not single step, skip */
+	if (task->thread.dabr)
+		return;
+#endif
+
 	if (regs != NULL) {
 #if defined(CONFIG_40x) || defined(CONFIG_BOOKE)
-		task->thread.dbcr0 = 0;
+		task->thread.dbcr0 &= ~(DBCR0_IC | DBCR0_IDM);
 		regs->msr &= ~MSR_DE;
 #else
 		regs->msr &= ~MSR_SE;
@@ -727,22 +734,75 @@ void user_disable_single_step(struct task_struct *task)
 	clear_tsk_thread_flag(task, TIF_SINGLESTEP);
 }
 
-static int ptrace_set_debugreg(struct task_struct *task, unsigned long addr,
+int ptrace_set_debugreg(struct task_struct *task, unsigned long addr,
 			       unsigned long data)
 {
-	/* We only support one DABR and no IABRS at the moment */
+	/* For ppc64 we support one DABR and no IABR's at the moment (ppc64).
+	 *  For embedded processors we support one DAC and no IAC's at the
+	 *  moment.
+	 */
 	if (addr > 0)
 		return -EINVAL;
 
-	/* The bottom 3 bits are flags */
 	if ((data & ~0x7UL) >= TASK_SIZE)
 		return -EIO;
 
-	/* Ensure translation is on */
+#ifdef CONFIG_PPC64
+
+	/* For processors using DABR (i.e. 970), the bottom 3 bits are flags.
+	 *  It was assumed, on previous implementations, that 3 bits were
+	 *  passed together with the data address, fitting the design of the
+	 *  DABR register, as follows:
+	 *
+	 *  bit 0: Read flag
+	 *  bit 1: Write flag
+	 *  bit 2: Breakpoint translation
+	 *
+	 *  Thus, we use them here as so.
+	 */
+
+	/* Ensure breakpoint translation bit is set */
 	if (data && !(data & DABR_TRANSLATION))
 		return -EIO;
 
+	/* Move contents to the DABR register */
 	task->thread.dabr = data;
+
+#endif
+#if defined(CONFIG_44x) || defined(CONFIG_BOOKE)
+
+	/* As described above, it was assumed 3 bits were passed with the data
+	 *  address, but we will assume only the mode bits will be passed
+	 *  as to not cause alignment restrictions for DAC-based processors.
+	 */
+
+	/* DAC's hold the whole address without any mode flags */
+	task->thread.dabr = data & ~0x3UL;
+
+	if (task->thread.dabr == 0) {
+		task->thread.dbcr0 &= ~(DBSR_DAC1R | DBSR_DAC1W | DBCR0_IDM);
+		task->thread.regs->msr &= ~MSR_DE;
+		return 0;
+	}
+
+	/* Read or Write bits must be set */
+
+	if (!(data & 0x3UL))
+		return -EINVAL;
+
+	/* Set the Internal Debugging flag (IDM bit 1) for the DBCR0
+	   register */
+	task->thread.dbcr0 = DBCR0_IDM;
+
+	/* Check for write and read flags and set DBCR0
+	   accordingly */
+	if (data & 0x1UL)
+		task->thread.dbcr0 |= DBSR_DAC1R;
+	if (data & 0x2UL)
+		task->thread.dbcr0 |= DBSR_DAC1W;
+
+	task->thread.regs->msr |= MSR_DE;
+#endif
 	return 0;
 }
 
diff --git a/arch/powerpc/kernel/signal.c b/arch/powerpc/kernel/signal.c
index ad55488939c3..7aada783ec6a 100644
--- a/arch/powerpc/kernel/signal.c
+++ b/arch/powerpc/kernel/signal.c
@@ -145,8 +145,12 @@ int do_signal(sigset_t *oldset, struct pt_regs *regs)
 	 * user space. The DABR will have been cleared if it
 	 * triggered inside the kernel.
 	 */
-	if (current->thread.dabr)
+	if (current->thread.dabr) {
 		set_dabr(current->thread.dabr);
+#if defined(CONFIG_44x) || defined(CONFIG_BOOKE)
+		mtspr(SPRN_DBCR0, current->thread.dbcr0);
+#endif
+	}
 
 	if (is32) {
         	if (ka.sa.sa_flags & SA_SIGINFO)
diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c
index 878fbddb6ae1..81ccb8dd1a54 100644
--- a/arch/powerpc/kernel/traps.c
+++ b/arch/powerpc/kernel/traps.c
@@ -1067,6 +1067,22 @@ void __kprobes DebugException(struct pt_regs *regs, unsigned long debug_status)
 		}
 
 		_exception(SIGTRAP, regs, TRAP_TRACE, regs->nip);
+	} else if (debug_status & (DBSR_DAC1R | DBSR_DAC1W)) {
+		regs->msr &= ~MSR_DE;
+
+		if (user_mode(regs)) {
+			current->thread.dbcr0 &= ~(DBSR_DAC1R | DBSR_DAC1W |
+								DBCR0_IDM);
+		} else {
+			/* Disable DAC interupts */
+			mtspr(SPRN_DBCR0, mfspr(SPRN_DBCR0) & ~(DBSR_DAC1R |
+						DBSR_DAC1W | DBCR0_IDM));
+
+			/* Clear the DAC event */
+			mtspr(SPRN_DBSR, (DBSR_DAC1R | DBSR_DAC1W));
+		}
+		/* Setup and send the trap to the handler */
+		do_dabr(regs, mfspr(SPRN_DAC1), debug_status);
 	}
 }
 #endif /* CONFIG_4xx || CONFIG_BOOKE */
diff --git a/arch/powerpc/mm/fault.c b/arch/powerpc/mm/fault.c
index 1707d00331fc..565b7a237c84 100644
--- a/arch/powerpc/mm/fault.c
+++ b/arch/powerpc/mm/fault.c
@@ -100,31 +100,6 @@ static int store_updates_sp(struct pt_regs *regs)
 	return 0;
 }
 
-#if !(defined(CONFIG_4xx) || defined(CONFIG_BOOKE))
-static void do_dabr(struct pt_regs *regs, unsigned long address,
-		    unsigned long error_code)
-{
-	siginfo_t info;
-
-	if (notify_die(DIE_DABR_MATCH, "dabr_match", regs, error_code,
-			11, SIGSEGV) == NOTIFY_STOP)
-		return;
-
-	if (debugger_dabr_match(regs))
-		return;
-
-	/* Clear the DABR */
-	set_dabr(0);
-
-	/* Deliver the signal to userspace */
-	info.si_signo = SIGTRAP;
-	info.si_errno = 0;
-	info.si_code = TRAP_HWBKPT;
-	info.si_addr = (void __user *)address;
-	force_sig_info(SIGTRAP, &info, current);
-}
-#endif /* !(CONFIG_4xx || CONFIG_BOOKE)*/
-
 /*
  * For 600- and 800-family processors, the error_code parameter is DSISR
  * for a data fault, SRR1 for an instruction fault. For 400-family processors
diff --git a/include/asm-powerpc/system.h b/include/asm-powerpc/system.h
index e6e25e2364eb..d6648c143322 100644
--- a/include/asm-powerpc/system.h
+++ b/include/asm-powerpc/system.h
@@ -110,6 +110,8 @@ static inline int debugger_fault_handler(struct pt_regs *regs) { return 0; }
 #endif
 
 extern int set_dabr(unsigned long dabr);
+extern void do_dabr(struct pt_regs *regs, unsigned long address,
+		    unsigned long error_code);
 extern void print_backtrace(unsigned long *);
 extern void show_regs(struct pt_regs * regs);
 extern void flush_instruction_cache(void);

From 7886250e9d71b24d0205ac6798ee855fb3836318 Mon Sep 17 00:00:00 2001
From: Mark Nelson <markn@au1.ibm.com>
Date: Thu, 24 Jul 2008 14:28:48 +1000
Subject: [PATCH 0630/1171] powerpc/cell: Fixed IOMMU mapping uses weak
 ordering for a pcie endpoint

At the moment the fixed mapping is by default strongly ordered (the
iommu_fixed=weak boot option must be used to make the fixed mapping weakly
ordered). If we're on a setup where the southbridge is being used in
endpoint mode (triblade and CAB boards) the default should be a weakly
ordered fixed mapping.

This adds a check so that if a node of type pcie-endpoint can be found in
the device tree the fixed mapping is set to be weak by default (but can be
overridden using iommu_fixed=strong).

Signed-off-by: Mark Nelson <markn@au1.ibm.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/platforms/cell/iommu.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/arch/powerpc/platforms/cell/iommu.c b/arch/powerpc/platforms/cell/iommu.c
index 208005ca262c..031124a8e37b 100644
--- a/arch/powerpc/platforms/cell/iommu.c
+++ b/arch/powerpc/platforms/cell/iommu.c
@@ -1150,12 +1150,23 @@ static int iommu_fixed_disabled;
 
 static int __init setup_iommu_fixed(char *str)
 {
+	struct device_node *pciep;
+
 	if (strcmp(str, "off") == 0)
 		iommu_fixed_disabled = 1;
 
-	else if (strcmp(str, "weak") == 0)
+	/* If we can find a pcie-endpoint in the device tree assume that
+	 * we're on a triblade or a CAB so by default the fixed mapping
+	 * should be set to be weakly ordered; but only if the boot
+	 * option WASN'T set for strong ordering
+	 */
+	pciep = of_find_node_by_type(NULL, "pcie-endpoint");
+
+	if (strcmp(str, "weak") == 0 || (pciep && strcmp(str, "strong") != 0))
 		iommu_fixed_is_weak = 1;
 
+	of_node_put(pciep);
+
 	return 1;
 }
 __setup("iommu_fixed=", setup_iommu_fixed);

From 80c60bf9b96f6108c630d90efc073cd520801e6c Mon Sep 17 00:00:00 2001
From: Segher Boessenkool <segher@kernel.crashing.org>
Date: Fri, 25 Jul 2008 10:08:41 +1000
Subject: [PATCH 0631/1171] powerpc: Fix compile error with binutils 2.15

My previous patch to fix compilation with binutils-2.17 causes
a "file truncated" build error from ld with binutils 2.15 (and
possibly older), and a warning with 2.16 and 2.17.

This fixes it.

Signed-off-by: Segher Boessenkool <segher@kernel.crashing.org>
Acked-by: Chuck Meade <chuckmeade@mindspring.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/vmlinux.lds.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/powerpc/kernel/vmlinux.lds.S b/arch/powerpc/kernel/vmlinux.lds.S
index a914411bced5..4a8ce62fe112 100644
--- a/arch/powerpc/kernel/vmlinux.lds.S
+++ b/arch/powerpc/kernel/vmlinux.lds.S
@@ -85,7 +85,7 @@ SECTIONS
 
 	/* The dummy segment contents for the bug workaround mentioned above
 	   near PHDRS.  */
-	.dummy : {
+	.dummy : AT(ADDR(.dummy) - LOAD_OFFSET) {
 		LONG(0xf177)
 	} :kernel :dummy
 

From 545500b307658ad5783e0f3a52a32b97b2dfaed2 Mon Sep 17 00:00:00 2001
From: Nathan Fontenot <nfont@austin.ibm.com>
Date: Thu, 24 Jul 2008 04:25:00 +1000
Subject: [PATCH 0632/1171] powerpc/pseries: Remove extraneous error reporting
 for hcall failures in lparcfg

Remove the extraneous error reporting used when a hcall made from lparcfg fails.

Signed-off-by: Nathan Fontenot <nfont@austin.ibm.com>
Signed-off-by: Robert Jennings <rcj@linux.vnet.ibm.com>
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/lparcfg.c | 32 --------------------------------
 1 file changed, 32 deletions(-)

diff --git a/arch/powerpc/kernel/lparcfg.c b/arch/powerpc/kernel/lparcfg.c
index 827a5726a035..20278ece31dc 100644
--- a/arch/powerpc/kernel/lparcfg.c
+++ b/arch/powerpc/kernel/lparcfg.c
@@ -129,33 +129,6 @@ static int iseries_lparcfg_data(struct seq_file *m, void *v)
 /*
  * Methods used to fetch LPAR data when running on a pSeries platform.
  */
-static void log_plpar_hcall_return(unsigned long rc, char *tag)
-{
-	switch(rc) {
-	case 0:
-		return;
-	case H_HARDWARE:
-		printk(KERN_INFO "plpar-hcall (%s) "
-				"Hardware fault\n", tag);
-		return;
-	case H_FUNCTION:
-		printk(KERN_INFO "plpar-hcall (%s) "
-				"Function not allowed\n", tag);
-		return;
-	case H_AUTHORITY:
-		printk(KERN_INFO "plpar-hcall (%s) "
-				"Not authorized to this function\n", tag);
-		return;
-	case H_PARAMETER:
-		printk(KERN_INFO "plpar-hcall (%s) "
-				"Bad parameter(s)\n",tag);
-		return;
-	default:
-		printk(KERN_INFO "plpar-hcall (%s) "
-				"Unexpected rc(0x%lx)\n", tag, rc);
-	}
-}
-
 /*
  * H_GET_PPP hcall returns info in 4 parms.
  *  entitled_capacity,unallocated_capacity,
@@ -191,8 +164,6 @@ static unsigned int h_get_ppp(unsigned long *entitled,
 	*aggregation = retbuf[2];
 	*resource = retbuf[3];
 
-	log_plpar_hcall_return(rc, "H_GET_PPP");
-
 	return rc;
 }
 
@@ -205,9 +176,6 @@ static void h_pic(unsigned long *pool_idle_time, unsigned long *num_procs)
 
 	*pool_idle_time = retbuf[0];
 	*num_procs = retbuf[1];
-
-	if (rc != H_AUTHORITY)
-		log_plpar_hcall_return(rc, "H_PIC");
 }
 
 #define SPLPAR_CHARACTERISTICS_TOKEN 20

From 11529396ea3190113173f7a15e59a58dbcaa36c8 Mon Sep 17 00:00:00 2001
From: Nathan Fotenot <nfont@austin.ibm.com>
Date: Thu, 24 Jul 2008 04:25:16 +1000
Subject: [PATCH 0633/1171] powerpc/pseries: Split processor entitlement
 retrieval and gathering to helper routines

Split the retrieval and setting of processor entitlement and weight into
helper routines.  This also removes the printing of the raw values
returned from h_get_ppp, the values are already parsed and printed.

Signed-off-by: Nathan Fontenot <nfont@austin.ibm.com>
Signed-off-by: Robert Jennings <rcj@linux.vnet.ibm.com>
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/lparcfg.c | 166 ++++++++++++++++++----------------
 1 file changed, 88 insertions(+), 78 deletions(-)

diff --git a/arch/powerpc/kernel/lparcfg.c b/arch/powerpc/kernel/lparcfg.c
index 20278ece31dc..a0ca90ab5e39 100644
--- a/arch/powerpc/kernel/lparcfg.c
+++ b/arch/powerpc/kernel/lparcfg.c
@@ -167,7 +167,8 @@ static unsigned int h_get_ppp(unsigned long *entitled,
 	return rc;
 }
 
-static void h_pic(unsigned long *pool_idle_time, unsigned long *num_procs)
+static unsigned h_pic(unsigned long *pool_idle_time,
+		      unsigned long *num_procs)
 {
 	unsigned long rc;
 	unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
@@ -176,6 +177,51 @@ static void h_pic(unsigned long *pool_idle_time, unsigned long *num_procs)
 
 	*pool_idle_time = retbuf[0];
 	*num_procs = retbuf[1];
+
+	return rc;
+}
+
+/*
+ * parse_ppp_data
+ * Parse out the data returned from h_get_ppp and h_pic
+ */
+static void parse_ppp_data(struct seq_file *m)
+{
+	unsigned long h_entitled, h_unallocated;
+	unsigned long h_aggregation, h_resource;
+	int rc;
+
+	rc = h_get_ppp(&h_entitled, &h_unallocated, &h_aggregation,
+		       &h_resource);
+	if (rc)
+		return;
+
+	seq_printf(m, "partition_entitled_capacity=%ld\n", h_entitled);
+	seq_printf(m, "group=%ld\n", (h_aggregation >> 2 * 8) & 0xffff);
+	seq_printf(m, "system_active_processors=%ld\n",
+		   (h_resource >> 0 * 8) & 0xffff);
+
+	/* pool related entries are apropriate for shared configs */
+	if (lppaca[0].shared_proc) {
+		unsigned long pool_idle_time, pool_procs;
+
+		seq_printf(m, "pool=%ld\n", (h_aggregation >> 0 * 8) & 0xffff);
+
+		/* report pool_capacity in percentage */
+		seq_printf(m, "pool_capacity=%ld\n",
+			   ((h_resource >> 2 * 8) & 0xffff) * 100);
+
+		h_pic(&pool_idle_time, &pool_procs);
+		seq_printf(m, "pool_idle_time=%ld\n", pool_idle_time);
+		seq_printf(m, "pool_num_procs=%ld\n", pool_procs);
+	}
+
+	seq_printf(m, "unallocated_capacity_weight=%ld\n",
+		   (h_resource >> 4 * 8) & 0xFF);
+
+	seq_printf(m, "capacity_weight=%ld\n", (h_resource >> 5 * 8) & 0xFF);
+	seq_printf(m, "capped=%ld\n", (h_resource >> 6 * 8) & 0x01);
+	seq_printf(m, "unallocated_capacity=%ld\n", h_unallocated);
 }
 
 #define SPLPAR_CHARACTERISTICS_TOKEN 20
@@ -302,60 +348,11 @@ static int pseries_lparcfg_data(struct seq_file *m, void *v)
 	partition_active_processors = lparcfg_count_active_processors();
 
 	if (firmware_has_feature(FW_FEATURE_SPLPAR)) {
-		unsigned long h_entitled, h_unallocated;
-		unsigned long h_aggregation, h_resource;
-		unsigned long pool_idle_time, pool_procs;
-		unsigned long purr;
-
-		h_get_ppp(&h_entitled, &h_unallocated, &h_aggregation,
-			  &h_resource);
-
-		seq_printf(m, "R4=0x%lx\n", h_entitled);
-		seq_printf(m, "R5=0x%lx\n", h_unallocated);
-		seq_printf(m, "R6=0x%lx\n", h_aggregation);
-		seq_printf(m, "R7=0x%lx\n", h_resource);
-
-		purr = get_purr();
-
 		/* this call handles the ibm,get-system-parameter contents */
 		parse_system_parameter_string(m);
+		parse_ppp_data(m);
 
-		seq_printf(m, "partition_entitled_capacity=%ld\n", h_entitled);
-
-		seq_printf(m, "group=%ld\n", (h_aggregation >> 2 * 8) & 0xffff);
-
-		seq_printf(m, "system_active_processors=%ld\n",
-			   (h_resource >> 0 * 8) & 0xffff);
-
-		/* pool related entries are apropriate for shared configs */
-		if (lppaca[0].shared_proc) {
-
-			h_pic(&pool_idle_time, &pool_procs);
-
-			seq_printf(m, "pool=%ld\n",
-				   (h_aggregation >> 0 * 8) & 0xffff);
-
-			/* report pool_capacity in percentage */
-			seq_printf(m, "pool_capacity=%ld\n",
-				   ((h_resource >> 2 * 8) & 0xffff) * 100);
-
-			seq_printf(m, "pool_idle_time=%ld\n", pool_idle_time);
-
-			seq_printf(m, "pool_num_procs=%ld\n", pool_procs);
-		}
-
-		seq_printf(m, "unallocated_capacity_weight=%ld\n",
-			   (h_resource >> 4 * 8) & 0xFF);
-
-		seq_printf(m, "capacity_weight=%ld\n",
-			   (h_resource >> 5 * 8) & 0xFF);
-
-		seq_printf(m, "capped=%ld\n", (h_resource >> 6 * 8) & 0x01);
-
-		seq_printf(m, "unallocated_capacity=%ld\n", h_unallocated);
-
-		seq_printf(m, "purr=%ld\n", purr);
-
+		seq_printf(m, "purr=%ld\n", get_purr());
 	} else {		/* non SPLPAR case */
 
 		seq_printf(m, "system_active_processors=%d\n",
@@ -382,6 +379,41 @@ static int pseries_lparcfg_data(struct seq_file *m, void *v)
 	return 0;
 }
 
+static ssize_t update_ppp(u64 *entitlement, u8 *weight)
+{
+	unsigned long current_entitled;
+	unsigned long dummy;
+	unsigned long resource;
+	u8 current_weight, new_weight;
+	u64 new_entitled;
+	ssize_t retval;
+
+	/* Get our current parameters */
+	retval = h_get_ppp(&current_entitled, &dummy, &dummy, &resource);
+	if (retval)
+		return retval;
+
+	current_weight = (resource >> 5 * 8) & 0xFF;
+
+	if (entitlement) {
+		new_weight = current_weight;
+		new_entitled = *entitlement;
+	} else if (weight) {
+		new_weight = *weight;
+		new_entitled = current_entitled;
+	} else
+		return -EINVAL;
+
+	pr_debug("%s: current_entitled = %lu, current_weight = %u\n",
+		 __FUNCTION__, current_entitled, current_weight);
+
+	pr_debug("%s: new_entitled = %lu, new_weight = %u\n",
+		 __FUNCTION__, new_entitled, new_weight);
+
+	retval = plpar_hcall_norets(H_SET_PPP, new_entitled, new_weight);
+	return retval;
+}
+
 /*
  * Interface for changing system parameters (variable capacity weight
  * and entitled capacity).  Format of input is "param_name=value";
@@ -399,12 +431,6 @@ static ssize_t lparcfg_write(struct file *file, const char __user * buf,
 	char *tmp;
 	u64 new_entitled, *new_entitled_ptr = &new_entitled;
 	u8 new_weight, *new_weight_ptr = &new_weight;
-
-	unsigned long current_entitled;	/* parameters for h_get_ppp */
-	unsigned long dummy;
-	unsigned long resource;
-	u8 current_weight;
-
 	ssize_t retval = -ENOMEM;
 
 	if (!firmware_has_feature(FW_FEATURE_SPLPAR) ||
@@ -432,34 +458,18 @@ static ssize_t lparcfg_write(struct file *file, const char __user * buf,
 		*new_entitled_ptr = (u64) simple_strtoul(tmp, &endp, 10);
 		if (endp == tmp)
 			goto out;
-		new_weight_ptr = &current_weight;
+
+		retval = update_ppp(new_entitled_ptr, NULL);
 	} else if (!strcmp(kbuf, "capacity_weight")) {
 		char *endp;
 		*new_weight_ptr = (u8) simple_strtoul(tmp, &endp, 10);
 		if (endp == tmp)
 			goto out;
-		new_entitled_ptr = &current_entitled;
+
+		retval = update_ppp(NULL, new_weight_ptr);
 	} else
 		goto out;
 
-	/* Get our current parameters */
-	retval = h_get_ppp(&current_entitled, &dummy, &dummy, &resource);
-	if (retval) {
-		retval = -EIO;
-		goto out;
-	}
-
-	current_weight = (resource >> 5 * 8) & 0xFF;
-
-	pr_debug("%s: current_entitled = %lu, current_weight = %u\n",
-		 __func__, current_entitled, current_weight);
-
-	pr_debug("%s: new_entitled = %lu, new_weight = %u\n",
-		 __func__, *new_entitled_ptr, *new_weight_ptr);
-
-	retval = plpar_hcall_norets(H_SET_PPP, *new_entitled_ptr,
-				    *new_weight_ptr);
-
 	if (retval == H_SUCCESS || retval == H_CONSTRAINED) {
 		retval = count;
 	} else if (retval == H_BUSY) {

From dfc3403f0e5ffb94ee29942f313b87d4061d951b Mon Sep 17 00:00:00 2001
From: Nathan Fontenot <nfont@austin.ibm.com>
Date: Thu, 24 Jul 2008 04:27:30 +1000
Subject: [PATCH 0634/1171] powerpc/pseries: Add memory entitlement
 capabilities to /proc/ppc64/lparcfg

Update /proc/ppc64/lparcfg to display Cooperative Memory
Overcommitment statistics as reported by the H_GET_MPP hcall.  This
also updates the lparcfg interface to allow setting memory entitlement
and weight.

Signed-off-by: Nathan Fontenot <nfont@austin.ibm.com>
Signed-off-by: Robert Jennings <rcj@linux.vnet.ibm.com>
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/lparcfg.c | 121 +++++++++++++++++++++++++++++++++-
 include/asm-powerpc/hvcall.h  |  18 ++++-
 2 files changed, 137 insertions(+), 2 deletions(-)

diff --git a/arch/powerpc/kernel/lparcfg.c b/arch/powerpc/kernel/lparcfg.c
index a0ca90ab5e39..86e5b3ed10d8 100644
--- a/arch/powerpc/kernel/lparcfg.c
+++ b/arch/powerpc/kernel/lparcfg.c
@@ -35,7 +35,7 @@
 #include <asm/prom.h>
 #include <asm/vdso_datapage.h>
 
-#define MODULE_VERS "1.7"
+#define MODULE_VERS "1.8"
 #define MODULE_NAME "lparcfg"
 
 /* #define LPARCFG_DEBUG */
@@ -129,6 +129,35 @@ static int iseries_lparcfg_data(struct seq_file *m, void *v)
 /*
  * Methods used to fetch LPAR data when running on a pSeries platform.
  */
+/**
+ * h_get_mpp
+ * H_GET_MPP hcall returns info in 7 parms
+ */
+int h_get_mpp(struct hvcall_mpp_data *mpp_data)
+{
+	int rc;
+	unsigned long retbuf[PLPAR_HCALL9_BUFSIZE];
+
+	rc = plpar_hcall9(H_GET_MPP, retbuf);
+
+	mpp_data->entitled_mem = retbuf[0];
+	mpp_data->mapped_mem = retbuf[1];
+
+	mpp_data->group_num = (retbuf[2] >> 2 * 8) & 0xffff;
+	mpp_data->pool_num = retbuf[2] & 0xffff;
+
+	mpp_data->mem_weight = (retbuf[3] >> 7 * 8) & 0xff;
+	mpp_data->unallocated_mem_weight = (retbuf[3] >> 6 * 8) & 0xff;
+	mpp_data->unallocated_entitlement = retbuf[3] & 0xffffffffffff;
+
+	mpp_data->pool_size = retbuf[4];
+	mpp_data->loan_request = retbuf[5];
+	mpp_data->backing_mem = retbuf[6];
+
+	return rc;
+}
+EXPORT_SYMBOL(h_get_mpp);
+
 /*
  * H_GET_PPP hcall returns info in 4 parms.
  *  entitled_capacity,unallocated_capacity,
@@ -224,6 +253,44 @@ static void parse_ppp_data(struct seq_file *m)
 	seq_printf(m, "unallocated_capacity=%ld\n", h_unallocated);
 }
 
+/**
+ * parse_mpp_data
+ * Parse out data returned from h_get_mpp
+ */
+static void parse_mpp_data(struct seq_file *m)
+{
+	struct hvcall_mpp_data mpp_data;
+	int rc;
+
+	rc = h_get_mpp(&mpp_data);
+	if (rc)
+		return;
+
+	seq_printf(m, "entitled_memory=%ld\n", mpp_data.entitled_mem);
+
+	if (mpp_data.mapped_mem != -1)
+		seq_printf(m, "mapped_entitled_memory=%ld\n",
+		           mpp_data.mapped_mem);
+
+	seq_printf(m, "entitled_memory_group_number=%d\n", mpp_data.group_num);
+	seq_printf(m, "entitled_memory_pool_number=%d\n", mpp_data.pool_num);
+
+	seq_printf(m, "entitled_memory_weight=%d\n", mpp_data.mem_weight);
+	seq_printf(m, "unallocated_entitled_memory_weight=%d\n",
+	           mpp_data.unallocated_mem_weight);
+	seq_printf(m, "unallocated_io_mapping_entitlement=%ld\n",
+	           mpp_data.unallocated_entitlement);
+
+	if (mpp_data.pool_size != -1)
+		seq_printf(m, "entitled_memory_pool_size=%ld bytes\n",
+		           mpp_data.pool_size);
+
+	seq_printf(m, "entitled_memory_loan_request=%ld\n",
+	           mpp_data.loan_request);
+
+	seq_printf(m, "backing_memory=%ld bytes\n", mpp_data.backing_mem);
+}
+
 #define SPLPAR_CHARACTERISTICS_TOKEN 20
 #define SPLPAR_MAXLENGTH 1026*(sizeof(char))
 
@@ -351,6 +418,7 @@ static int pseries_lparcfg_data(struct seq_file *m, void *v)
 		/* this call handles the ibm,get-system-parameter contents */
 		parse_system_parameter_string(m);
 		parse_ppp_data(m);
+		parse_mpp_data(m);
 
 		seq_printf(m, "purr=%ld\n", get_purr());
 	} else {		/* non SPLPAR case */
@@ -414,6 +482,43 @@ static ssize_t update_ppp(u64 *entitlement, u8 *weight)
 	return retval;
 }
 
+/**
+ * update_mpp
+ *
+ * Update the memory entitlement and weight for the partition.  Caller must
+ * specify either a new entitlement or weight, not both, to be updated
+ * since the h_set_mpp call takes both entitlement and weight as parameters.
+ */
+static ssize_t update_mpp(u64 *entitlement, u8 *weight)
+{
+	struct hvcall_mpp_data mpp_data;
+	u64 new_entitled;
+	u8 new_weight;
+	ssize_t rc;
+
+	rc = h_get_mpp(&mpp_data);
+	if (rc)
+		return rc;
+
+	if (entitlement) {
+		new_weight = mpp_data.mem_weight;
+		new_entitled = *entitlement;
+	} else if (weight) {
+		new_weight = *weight;
+		new_entitled = mpp_data.entitled_mem;
+	} else
+		return -EINVAL;
+
+	pr_debug("%s: current_entitled = %lu, current_weight = %u\n",
+	         __FUNCTION__, mpp_data.entitled_mem, mpp_data.mem_weight);
+
+	pr_debug("%s: new_entitled = %lu, new_weight = %u\n",
+	         __FUNCTION__, new_entitled, new_weight);
+
+	rc = plpar_hcall_norets(H_SET_MPP, new_entitled, new_weight);
+	return rc;
+}
+
 /*
  * Interface for changing system parameters (variable capacity weight
  * and entitled capacity).  Format of input is "param_name=value";
@@ -467,6 +572,20 @@ static ssize_t lparcfg_write(struct file *file, const char __user * buf,
 			goto out;
 
 		retval = update_ppp(NULL, new_weight_ptr);
+	} else if (!strcmp(kbuf, "entitled_memory")) {
+		char *endp;
+		*new_entitled_ptr = (u64) simple_strtoul(tmp, &endp, 10);
+		if (endp == tmp)
+			goto out;
+
+		retval = update_mpp(new_entitled_ptr, NULL);
+	} else if (!strcmp(kbuf, "entitled_memory_weight")) {
+		char *endp;
+		*new_weight_ptr = (u8) simple_strtoul(tmp, &endp, 10);
+		if (endp == tmp)
+			goto out;
+
+		retval = update_mpp(NULL, new_weight_ptr);
 	} else
 		goto out;
 
diff --git a/include/asm-powerpc/hvcall.h b/include/asm-powerpc/hvcall.h
index bf6cd7cb996c..46e76456cbbd 100644
--- a/include/asm-powerpc/hvcall.h
+++ b/include/asm-powerpc/hvcall.h
@@ -210,7 +210,9 @@
 #define H_JOIN			0x298
 #define H_VASI_STATE            0x2A4
 #define H_ENABLE_CRQ		0x2B0
-#define MAX_HCALL_OPCODE	H_ENABLE_CRQ
+#define H_SET_MPP		0x2D0
+#define H_GET_MPP		0x2D4
+#define MAX_HCALL_OPCODE	H_GET_MPP
 
 #ifndef __ASSEMBLY__
 
@@ -270,6 +272,20 @@ struct hcall_stats {
 };
 #define HCALL_STAT_ARRAY_SIZE	((MAX_HCALL_OPCODE >> 2) + 1)
 
+struct hvcall_mpp_data {
+	unsigned long entitled_mem;
+	unsigned long mapped_mem;
+	unsigned short group_num;
+	unsigned short pool_num;
+	unsigned char mem_weight;
+	unsigned char unallocated_mem_weight;
+	unsigned long unallocated_entitlement;  /* value in bytes */
+	unsigned long pool_size;
+	signed long loan_request;
+	unsigned long backing_mem;
+};
+
+int h_get_mpp(struct hvcall_mpp_data *);
 #endif /* __ASSEMBLY__ */
 #endif /* __KERNEL__ */
 #endif /* _ASM_POWERPC_HVCALL_H */

From 398778f78b76fb72cb18411487af01dea202709e Mon Sep 17 00:00:00 2001
From: Robert Jennings <rcj@linux.vnet.ibm.com>
Date: Thu, 24 Jul 2008 04:28:05 +1000
Subject: [PATCH 0635/1171] powerpc/pseries: Split retrieval of processor
 entitlement data into a helper routine

Split the retrieval of processor entitlement data returned in the H_GET_PPP
hcall into its own helper routine.

Signed-off-by: Nathan Fontenot <nfont@austin.ibm.com>
Signed-off-by: Robert Jennings <rcj@linux.vnet.ibm.com>
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/lparcfg.c | 81 ++++++++++++++++++++---------------
 1 file changed, 46 insertions(+), 35 deletions(-)

diff --git a/arch/powerpc/kernel/lparcfg.c b/arch/powerpc/kernel/lparcfg.c
index 86e5b3ed10d8..d82e1fa5ce2a 100644
--- a/arch/powerpc/kernel/lparcfg.c
+++ b/arch/powerpc/kernel/lparcfg.c
@@ -158,6 +158,18 @@ int h_get_mpp(struct hvcall_mpp_data *mpp_data)
 }
 EXPORT_SYMBOL(h_get_mpp);
 
+struct hvcall_ppp_data {
+	u64	entitlement;
+	u64	unallocated_entitlement;
+	u16	group_num;
+	u16	pool_num;
+	u8	capped;
+	u8	weight;
+	u8	unallocated_weight;
+	u16	active_procs_in_pool;
+	u16	active_system_procs;
+};
+
 /*
  * H_GET_PPP hcall returns info in 4 parms.
  *  entitled_capacity,unallocated_capacity,
@@ -178,20 +190,24 @@ EXPORT_SYMBOL(h_get_mpp);
  *              XXXX - Active processors in Physical Processor Pool.
  *                  XXXX  - Processors active on platform.
  */
-static unsigned int h_get_ppp(unsigned long *entitled,
-			      unsigned long *unallocated,
-			      unsigned long *aggregation,
-			      unsigned long *resource)
+static unsigned int h_get_ppp(struct hvcall_ppp_data *ppp_data)
 {
 	unsigned long rc;
 	unsigned long retbuf[PLPAR_HCALL_BUFSIZE];
 
 	rc = plpar_hcall(H_GET_PPP, retbuf);
 
-	*entitled = retbuf[0];
-	*unallocated = retbuf[1];
-	*aggregation = retbuf[2];
-	*resource = retbuf[3];
+	ppp_data->entitlement = retbuf[0];
+	ppp_data->unallocated_entitlement = retbuf[1];
+
+	ppp_data->group_num = (retbuf[2] >> 2 * 8) & 0xffff;
+	ppp_data->pool_num = retbuf[2] & 0xffff;
+
+	ppp_data->capped = (retbuf[3] >> 6 * 8) & 0x01;
+	ppp_data->weight = (retbuf[3] >> 5 * 8) & 0xff;
+	ppp_data->unallocated_weight = (retbuf[3] >> 4 * 8) & 0xff;
+	ppp_data->active_procs_in_pool = (retbuf[3] >> 2 * 8) & 0xffff;
+	ppp_data->active_system_procs = retbuf[3] & 0xffff;
 
 	return rc;
 }
@@ -216,41 +232,40 @@ static unsigned h_pic(unsigned long *pool_idle_time,
  */
 static void parse_ppp_data(struct seq_file *m)
 {
-	unsigned long h_entitled, h_unallocated;
-	unsigned long h_aggregation, h_resource;
+	struct hvcall_ppp_data ppp_data;
 	int rc;
 
-	rc = h_get_ppp(&h_entitled, &h_unallocated, &h_aggregation,
-		       &h_resource);
+	rc = h_get_ppp(&ppp_data);
 	if (rc)
 		return;
 
-	seq_printf(m, "partition_entitled_capacity=%ld\n", h_entitled);
-	seq_printf(m, "group=%ld\n", (h_aggregation >> 2 * 8) & 0xffff);
-	seq_printf(m, "system_active_processors=%ld\n",
-		   (h_resource >> 0 * 8) & 0xffff);
+	seq_printf(m, "partition_entitled_capacity=%ld\n",
+	           ppp_data.entitlement);
+	seq_printf(m, "group=%d\n", ppp_data.group_num);
+	seq_printf(m, "system_active_processors=%d\n",
+	           ppp_data.active_system_procs);
 
 	/* pool related entries are apropriate for shared configs */
 	if (lppaca[0].shared_proc) {
 		unsigned long pool_idle_time, pool_procs;
 
-		seq_printf(m, "pool=%ld\n", (h_aggregation >> 0 * 8) & 0xffff);
+		seq_printf(m, "pool=%d\n", ppp_data.pool_num);
 
 		/* report pool_capacity in percentage */
-		seq_printf(m, "pool_capacity=%ld\n",
-			   ((h_resource >> 2 * 8) & 0xffff) * 100);
+		seq_printf(m, "pool_capacity=%d\n",
+			   ppp_data.active_procs_in_pool * 100);
 
 		h_pic(&pool_idle_time, &pool_procs);
 		seq_printf(m, "pool_idle_time=%ld\n", pool_idle_time);
 		seq_printf(m, "pool_num_procs=%ld\n", pool_procs);
 	}
 
-	seq_printf(m, "unallocated_capacity_weight=%ld\n",
-		   (h_resource >> 4 * 8) & 0xFF);
-
-	seq_printf(m, "capacity_weight=%ld\n", (h_resource >> 5 * 8) & 0xFF);
-	seq_printf(m, "capped=%ld\n", (h_resource >> 6 * 8) & 0x01);
-	seq_printf(m, "unallocated_capacity=%ld\n", h_unallocated);
+	seq_printf(m, "unallocated_capacity_weight=%d\n",
+		   ppp_data.unallocated_weight);
+	seq_printf(m, "capacity_weight=%d\n", ppp_data.weight);
+	seq_printf(m, "capped=%d\n", ppp_data.capped);
+	seq_printf(m, "unallocated_capacity=%ld\n",
+		   ppp_data.unallocated_entitlement);
 }
 
 /**
@@ -449,31 +464,27 @@ static int pseries_lparcfg_data(struct seq_file *m, void *v)
 
 static ssize_t update_ppp(u64 *entitlement, u8 *weight)
 {
-	unsigned long current_entitled;
-	unsigned long dummy;
-	unsigned long resource;
-	u8 current_weight, new_weight;
+	struct hvcall_ppp_data ppp_data;
+	u8 new_weight;
 	u64 new_entitled;
 	ssize_t retval;
 
 	/* Get our current parameters */
-	retval = h_get_ppp(&current_entitled, &dummy, &dummy, &resource);
+	retval = h_get_ppp(&ppp_data);
 	if (retval)
 		return retval;
 
-	current_weight = (resource >> 5 * 8) & 0xFF;
-
 	if (entitlement) {
-		new_weight = current_weight;
+		new_weight = ppp_data.weight;
 		new_entitled = *entitlement;
 	} else if (weight) {
 		new_weight = *weight;
-		new_entitled = current_entitled;
+		new_entitled = ppp_data.entitlement;
 	} else
 		return -EINVAL;
 
 	pr_debug("%s: current_entitled = %lu, current_weight = %u\n",
-		 __FUNCTION__, current_entitled, current_weight);
+	         __FUNCTION__, ppp_data.entitlement, ppp_data.weight);
 
 	pr_debug("%s: new_entitled = %lu, new_weight = %u\n",
 		 __FUNCTION__, new_entitled, new_weight);

From e46de429cb954d30a5642fba81d516ede518c65e Mon Sep 17 00:00:00 2001
From: Robert Jennings <rcj@linux.vnet.ibm.com>
Date: Thu, 24 Jul 2008 04:29:03 +1000
Subject: [PATCH 0636/1171] powerpc/pseries: Enable CMO feature during platform
 setup

For Cooperative Memory Overcommitment (CMO), set the FW_FEATURE_CMO
flag in powerpc_firmware_features from the rtas ibm,get-system-parameters
table prior to calling iommu_init_early_pSeries.

With this, any CMO specific functionality can be controlled by checking:
 firmware_has_feature(FW_FEATURE_CMO)

Signed-off-by: Robert Jennings <rcj@linux.vnet.ibm.com>
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/platforms/pseries/setup.c | 71 ++++++++++++++++++++++++++
 include/asm-powerpc/firmware.h         |  3 +-
 2 files changed, 73 insertions(+), 1 deletion(-)

diff --git a/arch/powerpc/platforms/pseries/setup.c b/arch/powerpc/platforms/pseries/setup.c
index 90beb444e1dd..063a0d2fba30 100644
--- a/arch/powerpc/platforms/pseries/setup.c
+++ b/arch/powerpc/platforms/pseries/setup.c
@@ -314,6 +314,76 @@ static int pseries_set_xdabr(unsigned long dabr)
 			H_DABRX_KERNEL | H_DABRX_USER);
 }
 
+#define CMO_CHARACTERISTICS_TOKEN 44
+#define CMO_MAXLENGTH 1026
+
+/**
+ * fw_cmo_feature_init - FW_FEATURE_CMO is not stored in ibm,hypertas-functions,
+ * handle that here. (Stolen from parse_system_parameter_string)
+ */
+void pSeries_cmo_feature_init(void)
+{
+	char *ptr, *key, *value, *end;
+	int call_status;
+	int PrPSP = -1;
+	int SecPSP = -1;
+
+	pr_debug(" -> fw_cmo_feature_init()\n");
+	spin_lock(&rtas_data_buf_lock);
+	memset(rtas_data_buf, 0, RTAS_DATA_BUF_SIZE);
+	call_status = rtas_call(rtas_token("ibm,get-system-parameter"), 3, 1,
+				NULL,
+				CMO_CHARACTERISTICS_TOKEN,
+				__pa(rtas_data_buf),
+				RTAS_DATA_BUF_SIZE);
+
+	if (call_status != 0) {
+		spin_unlock(&rtas_data_buf_lock);
+		pr_debug("CMO not available\n");
+		pr_debug(" <- fw_cmo_feature_init()\n");
+		return;
+	}
+
+	end = rtas_data_buf + CMO_MAXLENGTH - 2;
+	ptr = rtas_data_buf + 2;	/* step over strlen value */
+	key = value = ptr;
+
+	while (*ptr && (ptr <= end)) {
+		/* Separate the key and value by replacing '=' with '\0' and
+		 * point the value at the string after the '='
+		 */
+		if (ptr[0] == '=') {
+			ptr[0] = '\0';
+			value = ptr + 1;
+		} else if (ptr[0] == '\0' || ptr[0] == ',') {
+			/* Terminate the string containing the key/value pair */
+			ptr[0] = '\0';
+
+			if (key == value) {
+				pr_debug("Malformed key/value pair\n");
+				/* Never found a '=', end processing */
+				break;
+			}
+
+			if (0 == strcmp(key, "PrPSP"))
+				PrPSP = simple_strtol(value, NULL, 10);
+			else if (0 == strcmp(key, "SecPSP"))
+				SecPSP = simple_strtol(value, NULL, 10);
+			value = key = ptr + 1;
+		}
+		ptr++;
+	}
+
+	if (PrPSP != -1 || SecPSP != -1) {
+		pr_info("CMO enabled\n");
+		pr_debug("CMO enabled, PrPSP=%d, SecPSP=%d\n", PrPSP, SecPSP);
+		powerpc_firmware_features |= FW_FEATURE_CMO;
+	} else
+		pr_debug("CMO not enabled, PrPSP=%d, SecPSP=%d\n", PrPSP, SecPSP);
+	spin_unlock(&rtas_data_buf_lock);
+	pr_debug(" <- fw_cmo_feature_init()\n");
+}
+
 /*
  * Early initialization.  Relocation is on but do not reference unbolted pages
  */
@@ -329,6 +399,7 @@ static void __init pSeries_init_early(void)
 	else if (firmware_has_feature(FW_FEATURE_XDABR))
 		ppc_md.set_dabr = pseries_set_xdabr;
 
+	pSeries_cmo_feature_init();
 	iommu_init_early_pSeries();
 
 	pr_debug(" <- pSeries_init_early()\n");
diff --git a/include/asm-powerpc/firmware.h b/include/asm-powerpc/firmware.h
index ef328995ba9d..3a179827528d 100644
--- a/include/asm-powerpc/firmware.h
+++ b/include/asm-powerpc/firmware.h
@@ -46,6 +46,7 @@
 #define FW_FEATURE_PS3_LV1	ASM_CONST(0x0000000000800000)
 #define FW_FEATURE_BEAT		ASM_CONST(0x0000000001000000)
 #define FW_FEATURE_BULK_REMOVE	ASM_CONST(0x0000000002000000)
+#define FW_FEATURE_CMO		ASM_CONST(0x0000000004000000)
 
 #ifndef __ASSEMBLY__
 
@@ -58,7 +59,7 @@ enum {
 		FW_FEATURE_MIGRATE | FW_FEATURE_PERFMON | FW_FEATURE_CRQ |
 		FW_FEATURE_VIO | FW_FEATURE_RDMA | FW_FEATURE_LLAN |
 		FW_FEATURE_BULK | FW_FEATURE_XDABR | FW_FEATURE_MULTITCE |
-		FW_FEATURE_SPLPAR | FW_FEATURE_LPAR,
+		FW_FEATURE_SPLPAR | FW_FEATURE_LPAR | FW_FEATURE_CMO,
 	FW_FEATURE_PSERIES_ALWAYS = 0,
 	FW_FEATURE_ISERIES_POSSIBLE = FW_FEATURE_ISERIES | FW_FEATURE_LPAR,
 	FW_FEATURE_ISERIES_ALWAYS = FW_FEATURE_ISERIES | FW_FEATURE_LPAR,

From 86630a32320f83736c4c24e2c8bae218e4c56c7c Mon Sep 17 00:00:00 2001
From: Brian King <brking@linux.vnet.ibm.com>
Date: Thu, 24 Jul 2008 04:29:16 +1000
Subject: [PATCH 0637/1171] powerpc/pseries: Utilities to set firmware page
 state

Newer versions of firmware support page states, which are used by the
collaborative memory manager (future patch) to "loan" pages to the
hypervisor for use by other partitions.

Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: Robert Jennings <rcj@linux.vnet.ibm.com>
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/platforms/pseries/plpar_wrappers.h | 10 ++++++++++
 include/asm-powerpc/hvcall.h                    |  5 +++++
 2 files changed, 15 insertions(+)

diff --git a/arch/powerpc/platforms/pseries/plpar_wrappers.h b/arch/powerpc/platforms/pseries/plpar_wrappers.h
index d8680b589dc9..a437267c6bf8 100644
--- a/arch/powerpc/platforms/pseries/plpar_wrappers.h
+++ b/arch/powerpc/platforms/pseries/plpar_wrappers.h
@@ -42,6 +42,16 @@ static inline long register_slb_shadow(unsigned long cpu, unsigned long vpa)
 	return vpa_call(0x3, cpu, vpa);
 }
 
+static inline long plpar_page_set_loaned(unsigned long vpa)
+{
+	return plpar_hcall_norets(H_PAGE_INIT, H_PAGE_SET_LOANED, vpa, 0);
+}
+
+static inline long plpar_page_set_active(unsigned long vpa)
+{
+	return plpar_hcall_norets(H_PAGE_INIT, H_PAGE_SET_ACTIVE, vpa, 0);
+}
+
 extern void vpa_init(int cpu);
 
 static inline long plpar_pte_enter(unsigned long flags,
diff --git a/include/asm-powerpc/hvcall.h b/include/asm-powerpc/hvcall.h
index 46e76456cbbd..fbe2932fa9e9 100644
--- a/include/asm-powerpc/hvcall.h
+++ b/include/asm-powerpc/hvcall.h
@@ -92,6 +92,11 @@
 #define H_EXACT			(1UL<<(63-24))	/* Use exact PTE or return H_PTEG_FULL */
 #define H_R_XLATE		(1UL<<(63-25))	/* include a valid logical page num in the pte if the valid bit is set */
 #define H_READ_4		(1UL<<(63-26))	/* Return 4 PTEs */
+#define H_PAGE_STATE_CHANGE	(1UL<<(63-28))
+#define H_PAGE_UNUSED		((1UL<<(63-29)) | (1UL<<(63-30)))
+#define H_PAGE_SET_UNUSED	(H_PAGE_STATE_CHANGE | H_PAGE_UNUSED)
+#define H_PAGE_SET_LOANED	(H_PAGE_SET_UNUSED | (1UL<<(63-31)))
+#define H_PAGE_SET_ACTIVE	H_PAGE_STATE_CHANGE
 #define H_AVPN			(1UL<<(63-32))	/* An avpn is provided as a sanity test */
 #define H_ANDCOND		(1UL<<(63-33))
 #define H_ICACHE_INVALIDATE	(1UL<<(63-40))	/* icbi, etc.  (ignored for IO pages) */

From 84af458bb23bf5f0ba1af4320dd2a57f7c4363e5 Mon Sep 17 00:00:00 2001
From: Brian King <brking@linux.vnet.ibm.com>
Date: Thu, 24 Jul 2008 04:30:29 +1000
Subject: [PATCH 0638/1171] powerpc/pseries: Add collaborative memory manager

Adds a collaborative memory manager, which acts as a simple balloon driver
for System p machines that support cooperative memory overcommitment
(CMO).

Adds a platform configuration option for CMO called PPC_SMLPAR.

Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: Robert Jennings <rcj@linux.vnet.ibm.com>
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/platforms/pseries/Kconfig  |  23 ++
 arch/powerpc/platforms/pseries/Makefile |   1 +
 arch/powerpc/platforms/pseries/cmm.c    | 468 ++++++++++++++++++++++++
 3 files changed, 492 insertions(+)
 create mode 100644 arch/powerpc/platforms/pseries/cmm.c

diff --git a/arch/powerpc/platforms/pseries/Kconfig b/arch/powerpc/platforms/pseries/Kconfig
index 757c0296e0b8..97619fd51e39 100644
--- a/arch/powerpc/platforms/pseries/Kconfig
+++ b/arch/powerpc/platforms/pseries/Kconfig
@@ -40,3 +40,26 @@ config PPC_PSERIES_DEBUG
 	depends on PPC_PSERIES && PPC_EARLY_DEBUG
 	bool "Enable extra debug logging in platforms/pseries"
 	default y
+
+config PPC_SMLPAR
+	bool "Support for shared-memory logical partitions"
+	depends on PPC_PSERIES
+	select LPARCFG
+	default n
+	help
+	  Select this option to enable shared memory partition support.
+	  With this option a system running in an LPAR can be given more
+	  memory than physically available and will allow firmware to
+	  balance memory across many LPARs.
+
+config CMM
+	tristate "Collaborative memory management"
+	depends on PPC_SMLPAR
+	default y
+	help
+	  Select this option, if you want to enable the kernel interface
+	  to reduce the memory size of the system. This is accomplished
+	  by allocating pages of memory and put them "on hold". This only
+	  makes sense for a system running in an LPAR where the unused pages
+	  will be reused for other LPARs. The interface allows firmware to
+	  balance memory across many LPARs.
diff --git a/arch/powerpc/platforms/pseries/Makefile b/arch/powerpc/platforms/pseries/Makefile
index 554c6e42ef2a..dfe574af2dc0 100644
--- a/arch/powerpc/platforms/pseries/Makefile
+++ b/arch/powerpc/platforms/pseries/Makefile
@@ -24,3 +24,4 @@ obj-$(CONFIG_HVC_CONSOLE)	+= hvconsole.o
 obj-$(CONFIG_HVCS)		+= hvcserver.o
 obj-$(CONFIG_HCALL_STATS)	+= hvCall_inst.o
 obj-$(CONFIG_PHYP_DUMP)	+= phyp_dump.o
+obj-$(CONFIG_CMM)		+= cmm.o
diff --git a/arch/powerpc/platforms/pseries/cmm.c b/arch/powerpc/platforms/pseries/cmm.c
new file mode 100644
index 000000000000..c6b3be03168b
--- /dev/null
+++ b/arch/powerpc/platforms/pseries/cmm.c
@@ -0,0 +1,468 @@
+/*
+ * Collaborative memory management interface.
+ *
+ * Copyright (C) 2008 IBM Corporation
+ * Author(s): Brian King (brking@linux.vnet.ibm.com),
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ *
+ */
+
+#include <linux/ctype.h>
+#include <linux/delay.h>
+#include <linux/errno.h>
+#include <linux/fs.h>
+#include <linux/init.h>
+#include <linux/kthread.h>
+#include <linux/module.h>
+#include <linux/oom.h>
+#include <linux/sched.h>
+#include <linux/stringify.h>
+#include <linux/swap.h>
+#include <linux/sysdev.h>
+#include <asm/firmware.h>
+#include <asm/hvcall.h>
+#include <asm/mmu.h>
+#include <asm/pgalloc.h>
+#include <asm/uaccess.h>
+
+#include "plpar_wrappers.h"
+
+#define CMM_DRIVER_VERSION	"1.0.0"
+#define CMM_DEFAULT_DELAY	1
+#define CMM_DEBUG			0
+#define CMM_DISABLE		0
+#define CMM_OOM_KB		1024
+#define CMM_MIN_MEM_MB		256
+#define KB2PAGES(_p)		((_p)>>(PAGE_SHIFT-10))
+#define PAGES2KB(_p)		((_p)<<(PAGE_SHIFT-10))
+
+static unsigned int delay = CMM_DEFAULT_DELAY;
+static unsigned int oom_kb = CMM_OOM_KB;
+static unsigned int cmm_debug = CMM_DEBUG;
+static unsigned int cmm_disabled = CMM_DISABLE;
+static unsigned long min_mem_mb = CMM_MIN_MEM_MB;
+static struct sys_device cmm_sysdev;
+
+MODULE_AUTHOR("Brian King <brking@linux.vnet.ibm.com>");
+MODULE_DESCRIPTION("IBM System p Collaborative Memory Manager");
+MODULE_LICENSE("GPL");
+MODULE_VERSION(CMM_DRIVER_VERSION);
+
+module_param_named(delay, delay, uint, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(delay, "Delay (in seconds) between polls to query hypervisor paging requests. "
+		 "[Default=" __stringify(CMM_DEFAULT_DELAY) "]");
+module_param_named(oom_kb, oom_kb, uint, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(oom_kb, "Amount of memory in kb to free on OOM. "
+		 "[Default=" __stringify(CMM_OOM_KB) "]");
+module_param_named(min_mem_mb, min_mem_mb, ulong, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(min_mem_mb, "Minimum amount of memory (in MB) to not balloon. "
+		 "[Default=" __stringify(CMM_MIN_MEM_MB) "]");
+module_param_named(debug, cmm_debug, uint, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(debug, "Enable module debugging logging. Set to 1 to enable. "
+		 "[Default=" __stringify(CMM_DEBUG) "]");
+
+#define CMM_NR_PAGES ((PAGE_SIZE - sizeof(void *) - sizeof(unsigned long)) / sizeof(unsigned long))
+
+#define cmm_dbg(...) if (cmm_debug) { printk(KERN_INFO "cmm: "__VA_ARGS__); }
+
+struct cmm_page_array {
+	struct cmm_page_array *next;
+	unsigned long index;
+	unsigned long page[CMM_NR_PAGES];
+};
+
+static unsigned long loaned_pages;
+static unsigned long loaned_pages_target;
+static unsigned long oom_freed_pages;
+
+static struct cmm_page_array *cmm_page_list;
+static DEFINE_SPINLOCK(cmm_lock);
+
+static struct task_struct *cmm_thread_ptr;
+
+/**
+ * cmm_alloc_pages - Allocate pages and mark them as loaned
+ * @nr:	number of pages to allocate
+ *
+ * Return value:
+ * 	number of pages requested to be allocated which were not
+ **/
+static long cmm_alloc_pages(long nr)
+{
+	struct cmm_page_array *pa, *npa;
+	unsigned long addr;
+	long rc;
+
+	cmm_dbg("Begin request for %ld pages\n", nr);
+
+	while (nr) {
+		addr = __get_free_page(GFP_NOIO | __GFP_NOWARN |
+				       __GFP_NORETRY | __GFP_NOMEMALLOC);
+		if (!addr)
+			break;
+		spin_lock(&cmm_lock);
+		pa = cmm_page_list;
+		if (!pa || pa->index >= CMM_NR_PAGES) {
+			/* Need a new page for the page list. */
+			spin_unlock(&cmm_lock);
+			npa = (struct cmm_page_array *)__get_free_page(GFP_NOIO | __GFP_NOWARN |
+								       __GFP_NORETRY | __GFP_NOMEMALLOC);
+			if (!npa) {
+				pr_info("%s: Can not allocate new page list\n", __FUNCTION__);
+				free_page(addr);
+				break;
+			}
+			spin_lock(&cmm_lock);
+			pa = cmm_page_list;
+
+			if (!pa || pa->index >= CMM_NR_PAGES) {
+				npa->next = pa;
+				npa->index = 0;
+				pa = npa;
+				cmm_page_list = pa;
+			} else
+				free_page((unsigned long) npa);
+		}
+
+		if ((rc = plpar_page_set_loaned(__pa(addr)))) {
+			pr_err("%s: Can not set page to loaned. rc=%ld\n", __FUNCTION__, rc);
+			spin_unlock(&cmm_lock);
+			free_page(addr);
+			break;
+		}
+
+		pa->page[pa->index++] = addr;
+		loaned_pages++;
+		totalram_pages--;
+		spin_unlock(&cmm_lock);
+		nr--;
+	}
+
+	cmm_dbg("End request with %ld pages unfulfilled\n", nr);
+	return nr;
+}
+
+/**
+ * cmm_free_pages - Free pages and mark them as active
+ * @nr:	number of pages to free
+ *
+ * Return value:
+ * 	number of pages requested to be freed which were not
+ **/
+static long cmm_free_pages(long nr)
+{
+	struct cmm_page_array *pa;
+	unsigned long addr;
+
+	cmm_dbg("Begin free of %ld pages.\n", nr);
+	spin_lock(&cmm_lock);
+	pa = cmm_page_list;
+	while (nr) {
+		if (!pa || pa->index <= 0)
+			break;
+		addr = pa->page[--pa->index];
+
+		if (pa->index == 0) {
+			pa = pa->next;
+			free_page((unsigned long) cmm_page_list);
+			cmm_page_list = pa;
+		}
+
+		plpar_page_set_active(__pa(addr));
+		free_page(addr);
+		loaned_pages--;
+		nr--;
+		totalram_pages++;
+	}
+	spin_unlock(&cmm_lock);
+	cmm_dbg("End request with %ld pages unfulfilled\n", nr);
+	return nr;
+}
+
+/**
+ * cmm_oom_notify - OOM notifier
+ * @self:	notifier block struct
+ * @dummy:	not used
+ * @parm:	returned - number of pages freed
+ *
+ * Return value:
+ * 	NOTIFY_OK
+ **/
+static int cmm_oom_notify(struct notifier_block *self,
+			  unsigned long dummy, void *parm)
+{
+	unsigned long *freed = parm;
+	long nr = KB2PAGES(oom_kb);
+
+	cmm_dbg("OOM processing started\n");
+	nr = cmm_free_pages(nr);
+	loaned_pages_target = loaned_pages;
+	*freed += KB2PAGES(oom_kb) - nr;
+	oom_freed_pages += KB2PAGES(oom_kb) - nr;
+	cmm_dbg("OOM processing complete\n");
+	return NOTIFY_OK;
+}
+
+/**
+ * cmm_get_mpp - Read memory performance parameters
+ *
+ * Makes hcall to query the current page loan request from the hypervisor.
+ *
+ * Return value:
+ * 	nothing
+ **/
+static void cmm_get_mpp(void)
+{
+	int rc;
+	struct hvcall_mpp_data mpp_data;
+	unsigned long active_pages_target;
+	signed long page_loan_request;
+
+	rc = h_get_mpp(&mpp_data);
+
+	if (rc != H_SUCCESS)
+		return;
+
+	page_loan_request = div_s64((s64)mpp_data.loan_request, PAGE_SIZE);
+	loaned_pages_target = page_loan_request + loaned_pages;
+	if (loaned_pages_target > oom_freed_pages)
+		loaned_pages_target -= oom_freed_pages;
+	else
+		loaned_pages_target = 0;
+
+	active_pages_target = totalram_pages + loaned_pages - loaned_pages_target;
+
+	if ((min_mem_mb * 1024 * 1024) > (active_pages_target * PAGE_SIZE))
+		loaned_pages_target = totalram_pages + loaned_pages -
+			((min_mem_mb * 1024 * 1024) / PAGE_SIZE);
+
+	cmm_dbg("delta = %ld, loaned = %lu, target = %lu, oom = %lu, totalram = %lu\n",
+		page_loan_request, loaned_pages, loaned_pages_target,
+		oom_freed_pages, totalram_pages);
+}
+
+static struct notifier_block cmm_oom_nb = {
+	.notifier_call = cmm_oom_notify
+};
+
+/**
+ * cmm_thread - CMM task thread
+ * @dummy:	not used
+ *
+ * Return value:
+ * 	0
+ **/
+static int cmm_thread(void *dummy)
+{
+	unsigned long timeleft;
+
+	while (1) {
+		timeleft = msleep_interruptible(delay * 1000);
+
+		if (kthread_should_stop() || timeleft) {
+			loaned_pages_target = loaned_pages;
+			break;
+		}
+
+		cmm_get_mpp();
+
+		if (loaned_pages_target > loaned_pages) {
+			if (cmm_alloc_pages(loaned_pages_target - loaned_pages))
+				loaned_pages_target = loaned_pages;
+		} else if (loaned_pages_target < loaned_pages)
+			cmm_free_pages(loaned_pages - loaned_pages_target);
+	}
+	return 0;
+}
+
+#define CMM_SHOW(name, format, args...)			\
+	static ssize_t show_##name(struct sys_device *dev, char *buf)	\
+	{							\
+		return sprintf(buf, format, ##args);		\
+	}							\
+	static SYSDEV_ATTR(name, S_IRUGO, show_##name, NULL)
+
+CMM_SHOW(loaned_kb, "%lu\n", PAGES2KB(loaned_pages));
+CMM_SHOW(loaned_target_kb, "%lu\n", PAGES2KB(loaned_pages_target));
+
+static ssize_t show_oom_pages(struct sys_device *dev, char *buf)
+{
+	return sprintf(buf, "%lu\n", PAGES2KB(oom_freed_pages));
+}
+
+static ssize_t store_oom_pages(struct sys_device *dev,
+			       const char *buf, size_t count)
+{
+	unsigned long val = simple_strtoul (buf, NULL, 10);
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+	if (val != 0)
+		return -EBADMSG;
+
+	oom_freed_pages = 0;
+	return count;
+}
+
+static SYSDEV_ATTR(oom_freed_kb, S_IWUSR| S_IRUGO,
+		   show_oom_pages, store_oom_pages);
+
+static struct sysdev_attribute *cmm_attrs[] = {
+	&attr_loaned_kb,
+	&attr_loaned_target_kb,
+	&attr_oom_freed_kb,
+};
+
+static struct sysdev_class cmm_sysdev_class = {
+	.name = "cmm",
+};
+
+/**
+ * cmm_sysfs_register - Register with sysfs
+ *
+ * Return value:
+ * 	0 on success / other on failure
+ **/
+static int cmm_sysfs_register(struct sys_device *sysdev)
+{
+	int i, rc;
+
+	if ((rc = sysdev_class_register(&cmm_sysdev_class)))
+		return rc;
+
+	sysdev->id = 0;
+	sysdev->cls = &cmm_sysdev_class;
+
+	if ((rc = sysdev_register(sysdev)))
+		goto class_unregister;
+
+	for (i = 0; i < ARRAY_SIZE(cmm_attrs); i++) {
+		if ((rc = sysdev_create_file(sysdev, cmm_attrs[i])))
+			goto fail;
+	}
+
+	return 0;
+
+fail:
+	while (--i >= 0)
+		sysdev_remove_file(sysdev, cmm_attrs[i]);
+	sysdev_unregister(sysdev);
+class_unregister:
+	sysdev_class_unregister(&cmm_sysdev_class);
+	return rc;
+}
+
+/**
+ * cmm_unregister_sysfs - Unregister from sysfs
+ *
+ **/
+static void cmm_unregister_sysfs(struct sys_device *sysdev)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(cmm_attrs); i++)
+		sysdev_remove_file(sysdev, cmm_attrs[i]);
+	sysdev_unregister(sysdev);
+	sysdev_class_unregister(&cmm_sysdev_class);
+}
+
+/**
+ * cmm_init - Module initialization
+ *
+ * Return value:
+ * 	0 on success / other on failure
+ **/
+static int cmm_init(void)
+{
+	int rc = -ENOMEM;
+
+	if (!firmware_has_feature(FW_FEATURE_CMO))
+		return -EOPNOTSUPP;
+
+	if ((rc = register_oom_notifier(&cmm_oom_nb)) < 0)
+		return rc;
+
+	if ((rc = cmm_sysfs_register(&cmm_sysdev)))
+		goto out_oom_notifier;
+
+	if (cmm_disabled)
+		return rc;
+
+	cmm_thread_ptr = kthread_run(cmm_thread, NULL, "cmmthread");
+	if (IS_ERR(cmm_thread_ptr)) {
+		rc = PTR_ERR(cmm_thread_ptr);
+		goto out_unregister_sysfs;
+	}
+
+	return rc;
+
+out_unregister_sysfs:
+	cmm_unregister_sysfs(&cmm_sysdev);
+out_oom_notifier:
+	unregister_oom_notifier(&cmm_oom_nb);
+	return rc;
+}
+
+/**
+ * cmm_exit - Module exit
+ *
+ * Return value:
+ * 	nothing
+ **/
+static void cmm_exit(void)
+{
+	if (cmm_thread_ptr)
+		kthread_stop(cmm_thread_ptr);
+	unregister_oom_notifier(&cmm_oom_nb);
+	cmm_free_pages(loaned_pages);
+	cmm_unregister_sysfs(&cmm_sysdev);
+}
+
+/**
+ * cmm_set_disable - Disable/Enable CMM
+ *
+ * Return value:
+ * 	0 on success / other on failure
+ **/
+static int cmm_set_disable(const char *val, struct kernel_param *kp)
+{
+	int disable = simple_strtoul(val, NULL, 10);
+
+	if (disable != 0 && disable != 1)
+		return -EINVAL;
+
+	if (disable && !cmm_disabled) {
+		if (cmm_thread_ptr)
+			kthread_stop(cmm_thread_ptr);
+		cmm_thread_ptr = NULL;
+		cmm_free_pages(loaned_pages);
+	} else if (!disable && cmm_disabled) {
+		cmm_thread_ptr = kthread_run(cmm_thread, NULL, "cmmthread");
+		if (IS_ERR(cmm_thread_ptr))
+			return PTR_ERR(cmm_thread_ptr);
+	}
+
+	cmm_disabled = disable;
+	return 0;
+}
+
+module_param_call(disable, cmm_set_disable, param_get_uint,
+		  &cmm_disabled, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(disable, "Disable CMM. Set to 1 to disable. "
+		 "[Default=" __stringify(CMM_DISABLE) "]");
+
+module_init(cmm_init);
+module_exit(cmm_exit);

From ffa5abbd0c399b32fc13a1b4718d87ee7a716999 Mon Sep 17 00:00:00 2001
From: Brian King <brking@linux.vnet.ibm.com>
Date: Thu, 24 Jul 2008 04:30:58 +1000
Subject: [PATCH 0639/1171] powerpc/pseries: Add CMO paging statistics

With the addition of Cooperative Memory Overcommitment (CMO) support
for IBM Power Systems, two fields have been added to the VPA to report
paging statistics.  Add support in lparcfg to report them to userspace.

Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: Robert Jennings <rcj@linux.vnet.ibm.com>
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/lparcfg.c | 20 ++++++++++++++++++++
 include/asm-powerpc/lppaca.h  |  5 ++++-
 2 files changed, 24 insertions(+), 1 deletion(-)

diff --git a/arch/powerpc/kernel/lparcfg.c b/arch/powerpc/kernel/lparcfg.c
index d82e1fa5ce2a..848c3e5a6370 100644
--- a/arch/powerpc/kernel/lparcfg.c
+++ b/arch/powerpc/kernel/lparcfg.c
@@ -409,6 +409,25 @@ static int lparcfg_count_active_processors(void)
 	return count;
 }
 
+static void pseries_cmo_data(struct seq_file *m)
+{
+	int cpu;
+	unsigned long cmo_faults = 0;
+	unsigned long cmo_fault_time = 0;
+
+	if (!firmware_has_feature(FW_FEATURE_CMO))
+		return;
+
+	for_each_possible_cpu(cpu) {
+		cmo_faults += lppaca[cpu].cmo_faults;
+		cmo_fault_time += lppaca[cpu].cmo_fault_time;
+	}
+
+	seq_printf(m, "cmo_faults=%lu\n", cmo_faults);
+	seq_printf(m, "cmo_fault_time_usec=%lu\n",
+		   cmo_fault_time / tb_ticks_per_usec);
+}
+
 static int pseries_lparcfg_data(struct seq_file *m, void *v)
 {
 	int partition_potential_processors;
@@ -434,6 +453,7 @@ static int pseries_lparcfg_data(struct seq_file *m, void *v)
 		parse_system_parameter_string(m);
 		parse_ppp_data(m);
 		parse_mpp_data(m);
+		pseries_cmo_data(m);
 
 		seq_printf(m, "purr=%ld\n", get_purr());
 	} else {		/* non SPLPAR case */
diff --git a/include/asm-powerpc/lppaca.h b/include/asm-powerpc/lppaca.h
index 567ed92cd91f..2fe268b10333 100644
--- a/include/asm-powerpc/lppaca.h
+++ b/include/asm-powerpc/lppaca.h
@@ -125,7 +125,10 @@ struct lppaca {
 	// NOTE: This value will ALWAYS be zero for dedicated processors and
 	// will NEVER be zero for shared processors (ie, initialized to a 1).
 	volatile u32 yield_count;	// PLIC increments each dispatchx00-x03
-	u8	reserved6[124];		// Reserved                     x04-x7F
+	u32 reserved6;
+	volatile u64 cmo_faults;	// CMO page fault count         x08-x0F
+	volatile u64 cmo_fault_time;	// CMO page fault time          x10-x17
+	u8	reserved7[104];		// Reserved                     x18-x7F
 
 //=============================================================================
 // CACHE_LINE_4-5 0x0180 - 0x027F Contains PMC interrupt data

From 6490c4903d12f242bec4454301f76f6a7520e399 Mon Sep 17 00:00:00 2001
From: Robert Jennings <rcj@linux.vnet.ibm.com>
Date: Thu, 24 Jul 2008 04:31:16 +1000
Subject: [PATCH 0640/1171] powerpc/pseries: iommu enablement for CMO

To support Cooperative Memory Overcommitment (CMO), we need to check
for failure from some of the tce hcalls.

These changes for the pseries platform affect the powerpc architecture;
patches for the other affected platforms are included in this patch.

pSeries platform IOMMU code changes:
 * platform TCE functions must handle H_NOT_ENOUGH_RESOURCES errors and
   return an error.

Architecture IOMMU code changes:
 * Calls to ppc_md.tce_build need to check return values and return
   DMA_MAPPING_ERROR for transient errors.

Architecture changes:
 * struct machdep_calls for tce_build*_pSeriesLP functions need to change
   to indicate failure.
 * all other platforms will need updates to iommu functions to match the new
   calling semantics; they will return 0 on success.  The other platforms
   default configs have been built, but no further testing was performed.

Signed-off-by: Robert Jennings <rcj@linux.vnet.ibm.com>
Acked-by: Olof Johansson <olof@lixom.net>
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/iommu.c            | 28 ++++++++++++++---
 arch/powerpc/platforms/cell/iommu.c    |  3 +-
 arch/powerpc/platforms/iseries/iommu.c |  3 +-
 arch/powerpc/platforms/pasemi/iommu.c  |  3 +-
 arch/powerpc/platforms/pseries/iommu.c | 42 ++++++++++++++++++++------
 arch/powerpc/sysdev/dart_iommu.c       |  3 +-
 include/asm-powerpc/machdep.h          |  2 +-
 7 files changed, 64 insertions(+), 20 deletions(-)

diff --git a/arch/powerpc/kernel/iommu.c b/arch/powerpc/kernel/iommu.c
index 2385f68c1751..550a19399bfa 100644
--- a/arch/powerpc/kernel/iommu.c
+++ b/arch/powerpc/kernel/iommu.c
@@ -49,6 +49,8 @@ static int novmerge = 1;
 
 static int protect4gb = 1;
 
+static void __iommu_free(struct iommu_table *, dma_addr_t, unsigned int);
+
 static inline unsigned long iommu_num_pages(unsigned long vaddr,
 					    unsigned long slen)
 {
@@ -191,6 +193,7 @@ static dma_addr_t iommu_alloc(struct device *dev, struct iommu_table *tbl,
 {
 	unsigned long entry, flags;
 	dma_addr_t ret = DMA_ERROR_CODE;
+	int build_fail;
 
 	spin_lock_irqsave(&(tbl->it_lock), flags);
 
@@ -205,9 +208,21 @@ static dma_addr_t iommu_alloc(struct device *dev, struct iommu_table *tbl,
 	ret = entry << IOMMU_PAGE_SHIFT;	/* Set the return dma address */
 
 	/* Put the TCEs in the HW table */
-	ppc_md.tce_build(tbl, entry, npages, (unsigned long)page & IOMMU_PAGE_MASK,
-			 direction, attrs);
+	build_fail = ppc_md.tce_build(tbl, entry, npages,
+	                              (unsigned long)page & IOMMU_PAGE_MASK,
+	                              direction, attrs);
 
+	/* ppc_md.tce_build() only returns non-zero for transient errors.
+	 * Clean up the table bitmap in this case and return
+	 * DMA_ERROR_CODE. For all other errors the functionality is
+	 * not altered.
+	 */
+	if (unlikely(build_fail)) {
+		__iommu_free(tbl, ret, npages);
+
+		spin_unlock_irqrestore(&(tbl->it_lock), flags);
+		return DMA_ERROR_CODE;
+	}
 
 	/* Flush/invalidate TLB caches if necessary */
 	if (ppc_md.tce_flush)
@@ -276,7 +291,7 @@ int iommu_map_sg(struct device *dev, struct iommu_table *tbl,
 	dma_addr_t dma_next = 0, dma_addr;
 	unsigned long flags;
 	struct scatterlist *s, *outs, *segstart;
-	int outcount, incount, i;
+	int outcount, incount, i, build_fail = 0;
 	unsigned int align;
 	unsigned long handle;
 	unsigned int max_seg_size;
@@ -337,8 +352,11 @@ int iommu_map_sg(struct device *dev, struct iommu_table *tbl,
 			    npages, entry, dma_addr);
 
 		/* Insert into HW table */
-		ppc_md.tce_build(tbl, entry, npages, vaddr & IOMMU_PAGE_MASK,
-				 direction, attrs);
+		build_fail = ppc_md.tce_build(tbl, entry, npages,
+		                              vaddr & IOMMU_PAGE_MASK,
+		                              direction, attrs);
+		if(unlikely(build_fail))
+			goto failure;
 
 		/* If we are in an open segment, try merging */
 		if (segstart != s) {
diff --git a/arch/powerpc/platforms/cell/iommu.c b/arch/powerpc/platforms/cell/iommu.c
index 031124a8e37b..e06420af5fe9 100644
--- a/arch/powerpc/platforms/cell/iommu.c
+++ b/arch/powerpc/platforms/cell/iommu.c
@@ -172,7 +172,7 @@ static void invalidate_tce_cache(struct cbe_iommu *iommu, unsigned long *pte,
 	}
 }
 
-static void tce_build_cell(struct iommu_table *tbl, long index, long npages,
+static int tce_build_cell(struct iommu_table *tbl, long index, long npages,
 		unsigned long uaddr, enum dma_data_direction direction,
 		struct dma_attrs *attrs)
 {
@@ -213,6 +213,7 @@ static void tce_build_cell(struct iommu_table *tbl, long index, long npages,
 
 	pr_debug("tce_build_cell(index=%lx,n=%lx,dir=%d,base_pte=%lx)\n",
 		 index, npages, direction, base_pte);
+	return 0;
 }
 
 static void tce_free_cell(struct iommu_table *tbl, long index, long npages)
diff --git a/arch/powerpc/platforms/iseries/iommu.c b/arch/powerpc/platforms/iseries/iommu.c
index bc818e4e2033..bb464d1211b2 100644
--- a/arch/powerpc/platforms/iseries/iommu.c
+++ b/arch/powerpc/platforms/iseries/iommu.c
@@ -41,7 +41,7 @@
 #include <asm/iseries/hv_call_event.h>
 #include <asm/iseries/iommu.h>
 
-static void tce_build_iSeries(struct iommu_table *tbl, long index, long npages,
+static int tce_build_iSeries(struct iommu_table *tbl, long index, long npages,
 		unsigned long uaddr, enum dma_data_direction direction,
 		struct dma_attrs *attrs)
 {
@@ -71,6 +71,7 @@ static void tce_build_iSeries(struct iommu_table *tbl, long index, long npages,
 		index++;
 		uaddr += TCE_PAGE_SIZE;
 	}
+	return 0;
 }
 
 static void tce_free_iSeries(struct iommu_table *tbl, long index, long npages)
diff --git a/arch/powerpc/platforms/pasemi/iommu.c b/arch/powerpc/platforms/pasemi/iommu.c
index 70541b7a5013..a0ff03a3d8da 100644
--- a/arch/powerpc/platforms/pasemi/iommu.c
+++ b/arch/powerpc/platforms/pasemi/iommu.c
@@ -83,7 +83,7 @@ static u32 *iob_l2_base;
 static struct iommu_table iommu_table_iobmap;
 static int iommu_table_iobmap_inited;
 
-static void iobmap_build(struct iommu_table *tbl, long index,
+static int iobmap_build(struct iommu_table *tbl, long index,
 			 long npages, unsigned long uaddr,
 			 enum dma_data_direction direction,
 			 struct dma_attrs *attrs)
@@ -108,6 +108,7 @@ static void iobmap_build(struct iommu_table *tbl, long index,
 		uaddr += IOBMAP_PAGE_SIZE;
 		bus_addr += IOBMAP_PAGE_SIZE;
 	}
+	return 0;
 }
 
 
diff --git a/arch/powerpc/platforms/pseries/iommu.c b/arch/powerpc/platforms/pseries/iommu.c
index 5377dd4b849a..a8c446697f9e 100644
--- a/arch/powerpc/platforms/pseries/iommu.c
+++ b/arch/powerpc/platforms/pseries/iommu.c
@@ -48,7 +48,7 @@
 #include "plpar_wrappers.h"
 
 
-static void tce_build_pSeries(struct iommu_table *tbl, long index,
+static int tce_build_pSeries(struct iommu_table *tbl, long index,
 			      long npages, unsigned long uaddr,
 			      enum dma_data_direction direction,
 			      struct dma_attrs *attrs)
@@ -72,6 +72,7 @@ static void tce_build_pSeries(struct iommu_table *tbl, long index,
 		uaddr += TCE_PAGE_SIZE;
 		tcep++;
 	}
+	return 0;
 }
 
 
@@ -94,14 +95,19 @@ static unsigned long tce_get_pseries(struct iommu_table *tbl, long index)
 	return *tcep;
 }
 
-static void tce_build_pSeriesLP(struct iommu_table *tbl, long tcenum,
+static void tce_free_pSeriesLP(struct iommu_table*, long, long);
+static void tce_freemulti_pSeriesLP(struct iommu_table*, long, long);
+
+static int tce_build_pSeriesLP(struct iommu_table *tbl, long tcenum,
 				long npages, unsigned long uaddr,
 				enum dma_data_direction direction,
 				struct dma_attrs *attrs)
 {
-	u64 rc;
+	u64 rc = 0;
 	u64 proto_tce, tce;
 	u64 rpn;
+	int ret = 0;
+	long tcenum_start = tcenum, npages_start = npages;
 
 	rpn = (virt_to_abs(uaddr)) >> TCE_SHIFT;
 	proto_tce = TCE_PCI_READ;
@@ -112,6 +118,13 @@ static void tce_build_pSeriesLP(struct iommu_table *tbl, long tcenum,
 		tce = proto_tce | (rpn & TCE_RPN_MASK) << TCE_RPN_SHIFT;
 		rc = plpar_tce_put((u64)tbl->it_index, (u64)tcenum << 12, tce);
 
+		if (unlikely(rc == H_NOT_ENOUGH_RESOURCES)) {
+			ret = (int)rc;
+			tce_free_pSeriesLP(tbl, tcenum_start,
+			                   (npages_start - (npages + 1)));
+			break;
+		}
+
 		if (rc && printk_ratelimit()) {
 			printk("tce_build_pSeriesLP: plpar_tce_put failed. rc=%ld\n", rc);
 			printk("\tindex   = 0x%lx\n", (u64)tbl->it_index);
@@ -123,25 +136,27 @@ static void tce_build_pSeriesLP(struct iommu_table *tbl, long tcenum,
 		tcenum++;
 		rpn++;
 	}
+	return ret;
 }
 
 static DEFINE_PER_CPU(u64 *, tce_page) = NULL;
 
-static void tce_buildmulti_pSeriesLP(struct iommu_table *tbl, long tcenum,
+static int tce_buildmulti_pSeriesLP(struct iommu_table *tbl, long tcenum,
 				     long npages, unsigned long uaddr,
 				     enum dma_data_direction direction,
 				     struct dma_attrs *attrs)
 {
-	u64 rc;
+	u64 rc = 0;
 	u64 proto_tce;
 	u64 *tcep;
 	u64 rpn;
 	long l, limit;
+	long tcenum_start = tcenum, npages_start = npages;
+	int ret = 0;
 
 	if (npages == 1) {
-		tce_build_pSeriesLP(tbl, tcenum, npages, uaddr,
-				    direction, attrs);
-		return;
+		return tce_build_pSeriesLP(tbl, tcenum, npages, uaddr,
+		                           direction, attrs);
 	}
 
 	tcep = __get_cpu_var(tce_page);
@@ -153,9 +168,8 @@ static void tce_buildmulti_pSeriesLP(struct iommu_table *tbl, long tcenum,
 		tcep = (u64 *)__get_free_page(GFP_ATOMIC);
 		/* If allocation fails, fall back to the loop implementation */
 		if (!tcep) {
-			tce_build_pSeriesLP(tbl, tcenum, npages, uaddr,
+			return tce_build_pSeriesLP(tbl, tcenum, npages, uaddr,
 					    direction, attrs);
-			return;
 		}
 		__get_cpu_var(tce_page) = tcep;
 	}
@@ -187,6 +201,13 @@ static void tce_buildmulti_pSeriesLP(struct iommu_table *tbl, long tcenum,
 		tcenum += limit;
 	} while (npages > 0 && !rc);
 
+	if (unlikely(rc == H_NOT_ENOUGH_RESOURCES)) {
+		ret = (int)rc;
+		tce_freemulti_pSeriesLP(tbl, tcenum_start,
+		                        (npages_start - (npages + limit)));
+		return ret;
+	}
+
 	if (rc && printk_ratelimit()) {
 		printk("tce_buildmulti_pSeriesLP: plpar_tce_put failed. rc=%ld\n", rc);
 		printk("\tindex   = 0x%lx\n", (u64)tbl->it_index);
@@ -194,6 +215,7 @@ static void tce_buildmulti_pSeriesLP(struct iommu_table *tbl, long tcenum,
 		printk("\ttce[0] val = 0x%lx\n", tcep[0]);
 		show_stack(current, (unsigned long *)__get_SP());
 	}
+	return ret;
 }
 
 static void tce_free_pSeriesLP(struct iommu_table *tbl, long tcenum, long npages)
diff --git a/arch/powerpc/sysdev/dart_iommu.c b/arch/powerpc/sysdev/dart_iommu.c
index de8c8b542cfa..89639ecbf381 100644
--- a/arch/powerpc/sysdev/dart_iommu.c
+++ b/arch/powerpc/sysdev/dart_iommu.c
@@ -147,7 +147,7 @@ static void dart_flush(struct iommu_table *tbl)
 	}
 }
 
-static void dart_build(struct iommu_table *tbl, long index,
+static int dart_build(struct iommu_table *tbl, long index,
 		       long npages, unsigned long uaddr,
 		       enum dma_data_direction direction,
 		       struct dma_attrs *attrs)
@@ -184,6 +184,7 @@ static void dart_build(struct iommu_table *tbl, long index,
 	} else {
 		dart_dirty = 1;
 	}
+	return 0;
 }
 
 
diff --git a/include/asm-powerpc/machdep.h b/include/asm-powerpc/machdep.h
index 1233d735fd28..893aafd87fde 100644
--- a/include/asm-powerpc/machdep.h
+++ b/include/asm-powerpc/machdep.h
@@ -76,7 +76,7 @@ struct machdep_calls {
 	 * destroyed as well */
 	void		(*hpte_clear_all)(void);
 
-	void		(*tce_build)(struct iommu_table * tbl,
+	int		(*tce_build)(struct iommu_table *tbl,
 				     long index,
 				     long npages,
 				     unsigned long uaddr,

From a90ab95a9576d35de0d05f9f4fc435edcccafaa9 Mon Sep 17 00:00:00 2001
From: Robert Jennings <rcj@linux.vnet.ibm.com>
Date: Thu, 24 Jul 2008 04:31:33 +1000
Subject: [PATCH 0641/1171] powerpc/pseries: vio bus support for CMO

This is a large patch but the normal code path is not affected.  For
non-pSeries platforms the code is ifdef'ed out and for non-CMO enabled
pSeries systems this does not affect the normal code path.  Devices that
do not perform DMA operations do not need modification with this patch.
The function get_desired_dma was renamed from get_io_entitlement for
clarity.

Overview

Cooperative Memory Overcommitment (CMO) allows for a set of OS partitions
to be run with less RAM than the aggregate needs of the group of
partitions.  The firmware will balance memory between the partitions
and page in/out memory as needed.  Based on the number and type of IO
adpaters preset each partition is allocated an amount of memory for
DMA operations and this allocation will be guaranteed to the partition;
this is referred to as the partition's 'entitlement'.

Partitions running in a CMO environment can only have virtual IO devices
present.  The VIO bus layer will manage the IO entitlement for the system.
Accounting, at a system and per-device level, is tracked in the VIO bus
code and exposed via sysfs.  A set of dma_ops functions are added to
the bus to allow for this accounting.

Bus initialization

At initialization, the bus will calculate the minimum needs of the system
based on providing each device present with a standard minimum entitlement
along with a spare allocation for the bus to handle hotplug events.
If the minimum needs can not be met the system boot will be halted.

Device changes

The significant changes for devices while running under CMO are that the
devices must specify how much dedicated IO entitlement they desire and
must also handle DMA mapping errors that can occur due to constrained
IO memory.  The virtual IO drivers are modified to silence errors when
DMA mappings fail for CMO and handle these failures gracefully.

Each devices will be guaranteed a minimum entitlement that can always
be mapped.  Devices will specify how much entitlement they desire and
the VIO bus will attempt to provide for this.  Devices can change their
desired entitlement level at any point in time to address particular needs
(via vio_cmo_set_dev_desired()), not just at device probe time.

VIO bus changes

The system will have a particular entitlement level available from which
it can provide memory to the devices.  The bus defines two pools of memory
within this entitlement, the reserved and excess pools.  Each device is
provided with it's own entitlement no less than a system defined minimum
entitlement and no greater than what the device has specified as it's
desired entitlement.  The entitlement provided to devices comes from the
reserve pool.  The reserve pool can also contain a spare allocation as
large as the system defined minimum entitlement which is used for device
hotplug events.  Any entitlement not needed to fulfill the needs of a
reserve pool is placed in the excess pool.  Each device is guaranteed
that it can map up to it's entitled level; additional mapping are possible
as long as there is unmapped memory in the excess pool.

Bus probe

As the system starts, each device is given an entitlement equal only
to the system defined minimum entitlement.  The reserve pool is equal
to the sum of these entitlements, plus a spare allocation.  The VIO bus
also tracks the aggregate desired entitlement of all the devices.  If the
system desired entitlement is greater than the size of the reserve pool,
when devices unmap IO memory it will be reserved and a balance operation
will be scheduled for some time in the future.

Entitlement balancing

The balance function tries to fairly distribute entitlement between the
devices in the system with the goal of providing each device with it's
desired amount of entitlement.  Devices using more than what would be
ideal will have their entitled set-point adjusted; this will effectively
set a goal for lower IO memory usage as future mappings can fail and
deallocations will trigger a balance operation to distribute the newly
unmapped memory.  A fair distribution of entitlement can take several
balance operations to achieve.  Entitlement changes and device DLPAR
events will alter the state of CMO and will trigger balance operations.

Hotplug events

The VIO bus allows for changes in system entitlement at run-time via
'vio_cmo_entitlement_update()'.  When devices are added the hotplug
device event will be preceded by a system entitlement increase and this
is reversed when devices are removed.

The following changes are made that the VIO bus layer for CMO:
 * add IO memory accounting per device structure.
 * add IO memory entitlement query function to driver structure.
 * during vio bus probe, if CMO is enabled, check that driver has
   memory entitlement query function defined.  Fail if function not defined.
 * fail to register driver if io entitlement function not defined.
 * create set of dma_ops at vio level for CMO that will track allocations
   and return DMA failures once entitlement is reached.  Entitlement will
   limited by overall system entitlement.  Devices will have a reserved
   quantity of memory that is guaranteed, the rest can be used as available.
 * expose entitlement, current allocation, desired allocation, and the
   allocation error counter for devices to the user through sysfs
 * provide mechanism for changing a device's desired entitlement at run time
   for devices as an exported function and sysfs tunable
 * track any DMA failures for entitled IO memory for each vio device.
 * check entitlement against available system entitlement on device add
 * track entitlement metrics (high water mark, current usage)
 * provide function to reset high water mark
 * provide minimum and desired entitlement numbers at a bus level
 * provide drivers with a minimum guaranteed entitlement
 * balance available entitlement between devices to satisfy their needs
 * handle system entitlement changes and device hotplug

Signed-off-by: Robert Jennings <rcj@linux.vnet.ibm.com>
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/vio.c | 1033 ++++++++++++++++++++++++++++++++++++-
 include/asm-powerpc/vio.h |   27 +-
 2 files changed, 1052 insertions(+), 8 deletions(-)

diff --git a/arch/powerpc/kernel/vio.c b/arch/powerpc/kernel/vio.c
index b77f8af7ddde..ade8aeaa2e70 100644
--- a/arch/powerpc/kernel/vio.c
+++ b/arch/powerpc/kernel/vio.c
@@ -1,11 +1,12 @@
 /*
  * IBM PowerPC Virtual I/O Infrastructure Support.
  *
- *    Copyright (c) 2003-2005 IBM Corp.
+ *    Copyright (c) 2003,2008 IBM Corp.
  *     Dave Engebretsen engebret@us.ibm.com
  *     Santiago Leon santil@us.ibm.com
  *     Hollis Blanchard <hollisb@us.ibm.com>
  *     Stephen Rothwell
+ *     Robert Jennings <rcjenn@us.ibm.com>
  *
  *      This program is free software; you can redistribute it and/or
  *      modify it under the terms of the GNU General Public License
@@ -46,6 +47,996 @@ static struct vio_dev vio_bus_device  = { /* fake "parent" device */
 	.dev.bus = &vio_bus_type,
 };
 
+#ifdef CONFIG_PPC_SMLPAR
+/**
+ * vio_cmo_pool - A pool of IO memory for CMO use
+ *
+ * @size: The size of the pool in bytes
+ * @free: The amount of free memory in the pool
+ */
+struct vio_cmo_pool {
+	size_t size;
+	size_t free;
+};
+
+/* How many ms to delay queued balance work */
+#define VIO_CMO_BALANCE_DELAY 100
+
+/* Portion out IO memory to CMO devices by this chunk size */
+#define VIO_CMO_BALANCE_CHUNK 131072
+
+/**
+ * vio_cmo_dev_entry - A device that is CMO-enabled and requires entitlement
+ *
+ * @vio_dev: struct vio_dev pointer
+ * @list: pointer to other devices on bus that are being tracked
+ */
+struct vio_cmo_dev_entry {
+	struct vio_dev *viodev;
+	struct list_head list;
+};
+
+/**
+ * vio_cmo - VIO bus accounting structure for CMO entitlement
+ *
+ * @lock: spinlock for entire structure
+ * @balance_q: work queue for balancing system entitlement
+ * @device_list: list of CMO-enabled devices requiring entitlement
+ * @entitled: total system entitlement in bytes
+ * @reserve: pool of memory from which devices reserve entitlement, incl. spare
+ * @excess: pool of excess entitlement not needed for device reserves or spare
+ * @spare: IO memory for device hotplug functionality
+ * @min: minimum necessary for system operation
+ * @desired: desired memory for system operation
+ * @curr: bytes currently allocated
+ * @high: high water mark for IO data usage
+ */
+struct vio_cmo {
+	spinlock_t lock;
+	struct delayed_work balance_q;
+	struct list_head device_list;
+	size_t entitled;
+	struct vio_cmo_pool reserve;
+	struct vio_cmo_pool excess;
+	size_t spare;
+	size_t min;
+	size_t desired;
+	size_t curr;
+	size_t high;
+} vio_cmo;
+
+/**
+ * vio_cmo_OF_devices - Count the number of OF devices that have DMA windows
+ */
+static int vio_cmo_num_OF_devs(void)
+{
+	struct device_node *node_vroot;
+	int count = 0;
+
+	/*
+	 * Count the number of vdevice entries with an
+	 * ibm,my-dma-window OF property
+	 */
+	node_vroot = of_find_node_by_name(NULL, "vdevice");
+	if (node_vroot) {
+		struct device_node *of_node;
+		struct property *prop;
+
+		for_each_child_of_node(node_vroot, of_node) {
+			prop = of_find_property(of_node, "ibm,my-dma-window",
+			                       NULL);
+			if (prop)
+				count++;
+		}
+	}
+	of_node_put(node_vroot);
+	return count;
+}
+
+/**
+ * vio_cmo_alloc - allocate IO memory for CMO-enable devices
+ *
+ * @viodev: VIO device requesting IO memory
+ * @size: size of allocation requested
+ *
+ * Allocations come from memory reserved for the devices and any excess
+ * IO memory available to all devices.  The spare pool used to service
+ * hotplug must be equal to %VIO_CMO_MIN_ENT for the excess pool to be
+ * made available.
+ *
+ * Return codes:
+ *  0 for successful allocation and -ENOMEM for a failure
+ */
+static inline int vio_cmo_alloc(struct vio_dev *viodev, size_t size)
+{
+	unsigned long flags;
+	size_t reserve_free = 0;
+	size_t excess_free = 0;
+	int ret = -ENOMEM;
+
+	spin_lock_irqsave(&vio_cmo.lock, flags);
+
+	/* Determine the amount of free entitlement available in reserve */
+	if (viodev->cmo.entitled > viodev->cmo.allocated)
+		reserve_free = viodev->cmo.entitled - viodev->cmo.allocated;
+
+	/* If spare is not fulfilled, the excess pool can not be used. */
+	if (vio_cmo.spare >= VIO_CMO_MIN_ENT)
+		excess_free = vio_cmo.excess.free;
+
+	/* The request can be satisfied */
+	if ((reserve_free + excess_free) >= size) {
+		vio_cmo.curr += size;
+		if (vio_cmo.curr > vio_cmo.high)
+			vio_cmo.high = vio_cmo.curr;
+		viodev->cmo.allocated += size;
+		size -= min(reserve_free, size);
+		vio_cmo.excess.free -= size;
+		ret = 0;
+	}
+
+	spin_unlock_irqrestore(&vio_cmo.lock, flags);
+	return ret;
+}
+
+/**
+ * vio_cmo_dealloc - deallocate IO memory from CMO-enable devices
+ * @viodev: VIO device freeing IO memory
+ * @size: size of deallocation
+ *
+ * IO memory is freed by the device back to the correct memory pools.
+ * The spare pool is replenished first from either memory pool, then
+ * the reserve pool is used to reduce device entitlement, the excess
+ * pool is used to increase the reserve pool toward the desired entitlement
+ * target, and then the remaining memory is returned to the pools.
+ *
+ */
+static inline void vio_cmo_dealloc(struct vio_dev *viodev, size_t size)
+{
+	unsigned long flags;
+	size_t spare_needed = 0;
+	size_t excess_freed = 0;
+	size_t reserve_freed = size;
+	size_t tmp;
+	int balance = 0;
+
+	spin_lock_irqsave(&vio_cmo.lock, flags);
+	vio_cmo.curr -= size;
+
+	/* Amount of memory freed from the excess pool */
+	if (viodev->cmo.allocated > viodev->cmo.entitled) {
+		excess_freed = min(reserve_freed, (viodev->cmo.allocated -
+		                                   viodev->cmo.entitled));
+		reserve_freed -= excess_freed;
+	}
+
+	/* Remove allocation from device */
+	viodev->cmo.allocated -= (reserve_freed + excess_freed);
+
+	/* Spare is a subset of the reserve pool, replenish it first. */
+	spare_needed = VIO_CMO_MIN_ENT - vio_cmo.spare;
+
+	/*
+	 * Replenish the spare in the reserve pool from the excess pool.
+	 * This moves entitlement into the reserve pool.
+	 */
+	if (spare_needed && excess_freed) {
+		tmp = min(excess_freed, spare_needed);
+		vio_cmo.excess.size -= tmp;
+		vio_cmo.reserve.size += tmp;
+		vio_cmo.spare += tmp;
+		excess_freed -= tmp;
+		spare_needed -= tmp;
+		balance = 1;
+	}
+
+	/*
+	 * Replenish the spare in the reserve pool from the reserve pool.
+	 * This removes entitlement from the device down to VIO_CMO_MIN_ENT,
+	 * if needed, and gives it to the spare pool. The amount of used
+	 * memory in this pool does not change.
+	 */
+	if (spare_needed && reserve_freed) {
+		tmp = min(spare_needed, min(reserve_freed,
+		                            (viodev->cmo.entitled -
+		                             VIO_CMO_MIN_ENT)));
+
+		vio_cmo.spare += tmp;
+		viodev->cmo.entitled -= tmp;
+		reserve_freed -= tmp;
+		spare_needed -= tmp;
+		balance = 1;
+	}
+
+	/*
+	 * Increase the reserve pool until the desired allocation is met.
+	 * Move an allocation freed from the excess pool into the reserve
+	 * pool and schedule a balance operation.
+	 */
+	if (excess_freed && (vio_cmo.desired > vio_cmo.reserve.size)) {
+		tmp = min(excess_freed, (vio_cmo.desired - vio_cmo.reserve.size));
+
+		vio_cmo.excess.size -= tmp;
+		vio_cmo.reserve.size += tmp;
+		excess_freed -= tmp;
+		balance = 1;
+	}
+
+	/* Return memory from the excess pool to that pool */
+	if (excess_freed)
+		vio_cmo.excess.free += excess_freed;
+
+	if (balance)
+		schedule_delayed_work(&vio_cmo.balance_q, VIO_CMO_BALANCE_DELAY);
+	spin_unlock_irqrestore(&vio_cmo.lock, flags);
+}
+
+/**
+ * vio_cmo_entitlement_update - Manage system entitlement changes
+ *
+ * @new_entitlement: new system entitlement to attempt to accommodate
+ *
+ * Increases in entitlement will be used to fulfill the spare entitlement
+ * and the rest is given to the excess pool.  Decreases, if they are
+ * possible, come from the excess pool and from unused device entitlement
+ *
+ * Returns: 0 on success, -ENOMEM when change can not be made
+ */
+int vio_cmo_entitlement_update(size_t new_entitlement)
+{
+	struct vio_dev *viodev;
+	struct vio_cmo_dev_entry *dev_ent;
+	unsigned long flags;
+	size_t avail, delta, tmp;
+
+	spin_lock_irqsave(&vio_cmo.lock, flags);
+
+	/* Entitlement increases */
+	if (new_entitlement > vio_cmo.entitled) {
+		delta = new_entitlement - vio_cmo.entitled;
+
+		/* Fulfill spare allocation */
+		if (vio_cmo.spare < VIO_CMO_MIN_ENT) {
+			tmp = min(delta, (VIO_CMO_MIN_ENT - vio_cmo.spare));
+			vio_cmo.spare += tmp;
+			vio_cmo.reserve.size += tmp;
+			delta -= tmp;
+		}
+
+		/* Remaining new allocation goes to the excess pool */
+		vio_cmo.entitled += delta;
+		vio_cmo.excess.size += delta;
+		vio_cmo.excess.free += delta;
+
+		goto out;
+	}
+
+	/* Entitlement decreases */
+	delta = vio_cmo.entitled - new_entitlement;
+	avail = vio_cmo.excess.free;
+
+	/*
+	 * Need to check how much unused entitlement each device can
+	 * sacrifice to fulfill entitlement change.
+	 */
+	list_for_each_entry(dev_ent, &vio_cmo.device_list, list) {
+		if (avail >= delta)
+			break;
+
+		viodev = dev_ent->viodev;
+		if ((viodev->cmo.entitled > viodev->cmo.allocated) &&
+		    (viodev->cmo.entitled > VIO_CMO_MIN_ENT))
+				avail += viodev->cmo.entitled -
+				         max_t(size_t, viodev->cmo.allocated,
+				               VIO_CMO_MIN_ENT);
+	}
+
+	if (delta <= avail) {
+		vio_cmo.entitled -= delta;
+
+		/* Take entitlement from the excess pool first */
+		tmp = min(vio_cmo.excess.free, delta);
+		vio_cmo.excess.size -= tmp;
+		vio_cmo.excess.free -= tmp;
+		delta -= tmp;
+
+		/*
+		 * Remove all but VIO_CMO_MIN_ENT bytes from devices
+		 * until entitlement change is served
+		 */
+		list_for_each_entry(dev_ent, &vio_cmo.device_list, list) {
+			if (!delta)
+				break;
+
+			viodev = dev_ent->viodev;
+			tmp = 0;
+			if ((viodev->cmo.entitled > viodev->cmo.allocated) &&
+			    (viodev->cmo.entitled > VIO_CMO_MIN_ENT))
+				tmp = viodev->cmo.entitled -
+				      max_t(size_t, viodev->cmo.allocated,
+				            VIO_CMO_MIN_ENT);
+			viodev->cmo.entitled -= min(tmp, delta);
+			delta -= min(tmp, delta);
+		}
+	} else {
+		spin_unlock_irqrestore(&vio_cmo.lock, flags);
+		return -ENOMEM;
+	}
+
+out:
+	schedule_delayed_work(&vio_cmo.balance_q, 0);
+	spin_unlock_irqrestore(&vio_cmo.lock, flags);
+	return 0;
+}
+
+/**
+ * vio_cmo_balance - Balance entitlement among devices
+ *
+ * @work: work queue structure for this operation
+ *
+ * Any system entitlement above the minimum needed for devices, or
+ * already allocated to devices, can be distributed to the devices.
+ * The list of devices is iterated through to recalculate the desired
+ * entitlement level and to determine how much entitlement above the
+ * minimum entitlement is allocated to devices.
+ *
+ * Small chunks of the available entitlement are given to devices until
+ * their requirements are fulfilled or there is no entitlement left to give.
+ * Upon completion sizes of the reserve and excess pools are calculated.
+ *
+ * The system minimum entitlement level is also recalculated here.
+ * Entitlement will be reserved for devices even after vio_bus_remove to
+ * accommodate reloading the driver.  The OF tree is walked to count the
+ * number of devices present and this will remove entitlement for devices
+ * that have actually left the system after having vio_bus_remove called.
+ */
+static void vio_cmo_balance(struct work_struct *work)
+{
+	struct vio_cmo *cmo;
+	struct vio_dev *viodev;
+	struct vio_cmo_dev_entry *dev_ent;
+	unsigned long flags;
+	size_t avail = 0, level, chunk, need;
+	int devcount = 0, fulfilled;
+
+	cmo = container_of(work, struct vio_cmo, balance_q.work);
+
+	spin_lock_irqsave(&vio_cmo.lock, flags);
+
+	/* Calculate minimum entitlement and fulfill spare */
+	cmo->min = vio_cmo_num_OF_devs() * VIO_CMO_MIN_ENT;
+	BUG_ON(cmo->min > cmo->entitled);
+	cmo->spare = min_t(size_t, VIO_CMO_MIN_ENT, (cmo->entitled - cmo->min));
+	cmo->min += cmo->spare;
+	cmo->desired = cmo->min;
+
+	/*
+	 * Determine how much entitlement is available and reset device
+	 * entitlements
+	 */
+	avail = cmo->entitled - cmo->spare;
+	list_for_each_entry(dev_ent, &vio_cmo.device_list, list) {
+		viodev = dev_ent->viodev;
+		devcount++;
+		viodev->cmo.entitled = VIO_CMO_MIN_ENT;
+		cmo->desired += (viodev->cmo.desired - VIO_CMO_MIN_ENT);
+		avail -= max_t(size_t, viodev->cmo.allocated, VIO_CMO_MIN_ENT);
+	}
+
+	/*
+	 * Having provided each device with the minimum entitlement, loop
+	 * over the devices portioning out the remaining entitlement
+	 * until there is nothing left.
+	 */
+	level = VIO_CMO_MIN_ENT;
+	while (avail) {
+		fulfilled = 0;
+		list_for_each_entry(dev_ent, &vio_cmo.device_list, list) {
+			viodev = dev_ent->viodev;
+
+			if (viodev->cmo.desired <= level) {
+				fulfilled++;
+				continue;
+			}
+
+			/*
+			 * Give the device up to VIO_CMO_BALANCE_CHUNK
+			 * bytes of entitlement, but do not exceed the
+			 * desired level of entitlement for the device.
+			 */
+			chunk = min_t(size_t, avail, VIO_CMO_BALANCE_CHUNK);
+			chunk = min(chunk, (viodev->cmo.desired -
+			                    viodev->cmo.entitled));
+			viodev->cmo.entitled += chunk;
+
+			/*
+			 * If the memory for this entitlement increase was
+			 * already allocated to the device it does not come
+			 * from the available pool being portioned out.
+			 */
+			need = max(viodev->cmo.allocated, viodev->cmo.entitled)-
+			       max(viodev->cmo.allocated, level);
+			avail -= need;
+
+		}
+		if (fulfilled == devcount)
+			break;
+		level += VIO_CMO_BALANCE_CHUNK;
+	}
+
+	/* Calculate new reserve and excess pool sizes */
+	cmo->reserve.size = cmo->min;
+	cmo->excess.free = 0;
+	cmo->excess.size = 0;
+	need = 0;
+	list_for_each_entry(dev_ent, &vio_cmo.device_list, list) {
+		viodev = dev_ent->viodev;
+		/* Calculated reserve size above the minimum entitlement */
+		if (viodev->cmo.entitled)
+			cmo->reserve.size += (viodev->cmo.entitled -
+			                      VIO_CMO_MIN_ENT);
+		/* Calculated used excess entitlement */
+		if (viodev->cmo.allocated > viodev->cmo.entitled)
+			need += viodev->cmo.allocated - viodev->cmo.entitled;
+	}
+	cmo->excess.size = cmo->entitled - cmo->reserve.size;
+	cmo->excess.free = cmo->excess.size - need;
+
+	cancel_delayed_work(container_of(work, struct delayed_work, work));
+	spin_unlock_irqrestore(&vio_cmo.lock, flags);
+}
+
+static void *vio_dma_iommu_alloc_coherent(struct device *dev, size_t size,
+                                          dma_addr_t *dma_handle, gfp_t flag)
+{
+	struct vio_dev *viodev = to_vio_dev(dev);
+	void *ret;
+
+	if (vio_cmo_alloc(viodev, roundup(size, IOMMU_PAGE_SIZE))) {
+		atomic_inc(&viodev->cmo.allocs_failed);
+		return NULL;
+	}
+
+	ret = dma_iommu_ops.alloc_coherent(dev, size, dma_handle, flag);
+	if (unlikely(ret == NULL)) {
+		vio_cmo_dealloc(viodev, roundup(size, IOMMU_PAGE_SIZE));
+		atomic_inc(&viodev->cmo.allocs_failed);
+	}
+
+	return ret;
+}
+
+static void vio_dma_iommu_free_coherent(struct device *dev, size_t size,
+                                        void *vaddr, dma_addr_t dma_handle)
+{
+	struct vio_dev *viodev = to_vio_dev(dev);
+
+	dma_iommu_ops.free_coherent(dev, size, vaddr, dma_handle);
+
+	vio_cmo_dealloc(viodev, roundup(size, IOMMU_PAGE_SIZE));
+}
+
+static dma_addr_t vio_dma_iommu_map_single(struct device *dev, void *vaddr,
+                                           size_t size,
+                                           enum dma_data_direction direction,
+                                           struct dma_attrs *attrs)
+{
+	struct vio_dev *viodev = to_vio_dev(dev);
+	dma_addr_t ret = DMA_ERROR_CODE;
+
+	if (vio_cmo_alloc(viodev, roundup(size, IOMMU_PAGE_SIZE))) {
+		atomic_inc(&viodev->cmo.allocs_failed);
+		return ret;
+	}
+
+	ret = dma_iommu_ops.map_single(dev, vaddr, size, direction, attrs);
+	if (unlikely(dma_mapping_error(ret))) {
+		vio_cmo_dealloc(viodev, roundup(size, IOMMU_PAGE_SIZE));
+		atomic_inc(&viodev->cmo.allocs_failed);
+	}
+
+	return ret;
+}
+
+static void vio_dma_iommu_unmap_single(struct device *dev,
+		dma_addr_t dma_handle, size_t size,
+		enum dma_data_direction direction,
+		struct dma_attrs *attrs)
+{
+	struct vio_dev *viodev = to_vio_dev(dev);
+
+	dma_iommu_ops.unmap_single(dev, dma_handle, size, direction, attrs);
+
+	vio_cmo_dealloc(viodev, roundup(size, IOMMU_PAGE_SIZE));
+}
+
+static int vio_dma_iommu_map_sg(struct device *dev, struct scatterlist *sglist,
+                                int nelems, enum dma_data_direction direction,
+                                struct dma_attrs *attrs)
+{
+	struct vio_dev *viodev = to_vio_dev(dev);
+	struct scatterlist *sgl;
+	int ret, count = 0;
+	size_t alloc_size = 0;
+
+	for (sgl = sglist; count < nelems; count++, sgl++)
+		alloc_size += roundup(sgl->length, IOMMU_PAGE_SIZE);
+
+	if (vio_cmo_alloc(viodev, alloc_size)) {
+		atomic_inc(&viodev->cmo.allocs_failed);
+		return 0;
+	}
+
+	ret = dma_iommu_ops.map_sg(dev, sglist, nelems, direction, attrs);
+
+	if (unlikely(!ret)) {
+		vio_cmo_dealloc(viodev, alloc_size);
+		atomic_inc(&viodev->cmo.allocs_failed);
+	}
+
+	for (sgl = sglist, count = 0; count < ret; count++, sgl++)
+		alloc_size -= roundup(sgl->dma_length, IOMMU_PAGE_SIZE);
+	if (alloc_size)
+		vio_cmo_dealloc(viodev, alloc_size);
+
+	return ret;
+}
+
+static void vio_dma_iommu_unmap_sg(struct device *dev,
+		struct scatterlist *sglist, int nelems,
+		enum dma_data_direction direction,
+		struct dma_attrs *attrs)
+{
+	struct vio_dev *viodev = to_vio_dev(dev);
+	struct scatterlist *sgl;
+	size_t alloc_size = 0;
+	int count = 0;
+
+	for (sgl = sglist; count < nelems; count++, sgl++)
+		alloc_size += roundup(sgl->dma_length, IOMMU_PAGE_SIZE);
+
+	dma_iommu_ops.unmap_sg(dev, sglist, nelems, direction, attrs);
+
+	vio_cmo_dealloc(viodev, alloc_size);
+}
+
+struct dma_mapping_ops vio_dma_mapping_ops = {
+	.alloc_coherent = vio_dma_iommu_alloc_coherent,
+	.free_coherent  = vio_dma_iommu_free_coherent,
+	.map_single     = vio_dma_iommu_map_single,
+	.unmap_single   = vio_dma_iommu_unmap_single,
+	.map_sg         = vio_dma_iommu_map_sg,
+	.unmap_sg       = vio_dma_iommu_unmap_sg,
+};
+
+/**
+ * vio_cmo_set_dev_desired - Set desired entitlement for a device
+ *
+ * @viodev: struct vio_dev for device to alter
+ * @new_desired: new desired entitlement level in bytes
+ *
+ * For use by devices to request a change to their entitlement at runtime or
+ * through sysfs.  The desired entitlement level is changed and a balancing
+ * of system resources is scheduled to run in the future.
+ */
+void vio_cmo_set_dev_desired(struct vio_dev *viodev, size_t desired)
+{
+	unsigned long flags;
+	struct vio_cmo_dev_entry *dev_ent;
+	int found = 0;
+
+	if (!firmware_has_feature(FW_FEATURE_CMO))
+		return;
+
+	spin_lock_irqsave(&vio_cmo.lock, flags);
+	if (desired < VIO_CMO_MIN_ENT)
+		desired = VIO_CMO_MIN_ENT;
+
+	/*
+	 * Changes will not be made for devices not in the device list.
+	 * If it is not in the device list, then no driver is loaded
+	 * for the device and it can not receive entitlement.
+	 */
+	list_for_each_entry(dev_ent, &vio_cmo.device_list, list)
+		if (viodev == dev_ent->viodev) {
+			found = 1;
+			break;
+		}
+	if (!found)
+		return;
+
+	/* Increase/decrease in desired device entitlement */
+	if (desired >= viodev->cmo.desired) {
+		/* Just bump the bus and device values prior to a balance*/
+		vio_cmo.desired += desired - viodev->cmo.desired;
+		viodev->cmo.desired = desired;
+	} else {
+		/* Decrease bus and device values for desired entitlement */
+		vio_cmo.desired -= viodev->cmo.desired - desired;
+		viodev->cmo.desired = desired;
+		/*
+		 * If less entitlement is desired than current entitlement, move
+		 * any reserve memory in the change region to the excess pool.
+		 */
+		if (viodev->cmo.entitled > desired) {
+			vio_cmo.reserve.size -= viodev->cmo.entitled - desired;
+			vio_cmo.excess.size += viodev->cmo.entitled - desired;
+			/*
+			 * If entitlement moving from the reserve pool to the
+			 * excess pool is currently unused, add to the excess
+			 * free counter.
+			 */
+			if (viodev->cmo.allocated < viodev->cmo.entitled)
+				vio_cmo.excess.free += viodev->cmo.entitled -
+				                       max(viodev->cmo.allocated, desired);
+			viodev->cmo.entitled = desired;
+		}
+	}
+	schedule_delayed_work(&vio_cmo.balance_q, 0);
+	spin_unlock_irqrestore(&vio_cmo.lock, flags);
+}
+
+/**
+ * vio_cmo_bus_probe - Handle CMO specific bus probe activities
+ *
+ * @viodev - Pointer to struct vio_dev for device
+ *
+ * Determine the devices IO memory entitlement needs, attempting
+ * to satisfy the system minimum entitlement at first and scheduling
+ * a balance operation to take care of the rest at a later time.
+ *
+ * Returns: 0 on success, -EINVAL when device doesn't support CMO, and
+ *          -ENOMEM when entitlement is not available for device or
+ *          device entry.
+ *
+ */
+static int vio_cmo_bus_probe(struct vio_dev *viodev)
+{
+	struct vio_cmo_dev_entry *dev_ent;
+	struct device *dev = &viodev->dev;
+	struct vio_driver *viodrv = to_vio_driver(dev->driver);
+	unsigned long flags;
+	size_t size;
+
+	/*
+	 * Check to see that device has a DMA window and configure
+	 * entitlement for the device.
+	 */
+	if (of_get_property(viodev->dev.archdata.of_node,
+	                    "ibm,my-dma-window", NULL)) {
+		/* Check that the driver is CMO enabled and get desired DMA */
+		if (!viodrv->get_desired_dma) {
+			dev_err(dev, "%s: device driver does not support CMO\n",
+			        __func__);
+			return -EINVAL;
+		}
+
+		viodev->cmo.desired = IOMMU_PAGE_ALIGN(viodrv->get_desired_dma(viodev));
+		if (viodev->cmo.desired < VIO_CMO_MIN_ENT)
+			viodev->cmo.desired = VIO_CMO_MIN_ENT;
+		size = VIO_CMO_MIN_ENT;
+
+		dev_ent = kmalloc(sizeof(struct vio_cmo_dev_entry),
+		                  GFP_KERNEL);
+		if (!dev_ent)
+			return -ENOMEM;
+
+		dev_ent->viodev = viodev;
+		spin_lock_irqsave(&vio_cmo.lock, flags);
+		list_add(&dev_ent->list, &vio_cmo.device_list);
+	} else {
+		viodev->cmo.desired = 0;
+		size = 0;
+		spin_lock_irqsave(&vio_cmo.lock, flags);
+	}
+
+	/*
+	 * If the needs for vio_cmo.min have not changed since they
+	 * were last set, the number of devices in the OF tree has
+	 * been constant and the IO memory for this is already in
+	 * the reserve pool.
+	 */
+	if (vio_cmo.min == ((vio_cmo_num_OF_devs() + 1) *
+	                    VIO_CMO_MIN_ENT)) {
+		/* Updated desired entitlement if device requires it */
+		if (size)
+			vio_cmo.desired += (viodev->cmo.desired -
+		                        VIO_CMO_MIN_ENT);
+	} else {
+		size_t tmp;
+
+		tmp = vio_cmo.spare + vio_cmo.excess.free;
+		if (tmp < size) {
+			dev_err(dev, "%s: insufficient free "
+			        "entitlement to add device. "
+			        "Need %lu, have %lu\n", __func__,
+				size, (vio_cmo.spare + tmp));
+			spin_unlock_irqrestore(&vio_cmo.lock, flags);
+			return -ENOMEM;
+		}
+
+		/* Use excess pool first to fulfill request */
+		tmp = min(size, vio_cmo.excess.free);
+		vio_cmo.excess.free -= tmp;
+		vio_cmo.excess.size -= tmp;
+		vio_cmo.reserve.size += tmp;
+
+		/* Use spare if excess pool was insufficient */
+		vio_cmo.spare -= size - tmp;
+
+		/* Update bus accounting */
+		vio_cmo.min += size;
+		vio_cmo.desired += viodev->cmo.desired;
+	}
+	spin_unlock_irqrestore(&vio_cmo.lock, flags);
+	return 0;
+}
+
+/**
+ * vio_cmo_bus_remove - Handle CMO specific bus removal activities
+ *
+ * @viodev - Pointer to struct vio_dev for device
+ *
+ * Remove the device from the cmo device list.  The minimum entitlement
+ * will be reserved for the device as long as it is in the system.  The
+ * rest of the entitlement the device had been allocated will be returned
+ * to the system.
+ */
+static void vio_cmo_bus_remove(struct vio_dev *viodev)
+{
+	struct vio_cmo_dev_entry *dev_ent;
+	unsigned long flags;
+	size_t tmp;
+
+	spin_lock_irqsave(&vio_cmo.lock, flags);
+	if (viodev->cmo.allocated) {
+		dev_err(&viodev->dev, "%s: device had %lu bytes of IO "
+		        "allocated after remove operation.\n",
+		        __func__, viodev->cmo.allocated);
+		BUG();
+	}
+
+	/*
+	 * Remove the device from the device list being maintained for
+	 * CMO enabled devices.
+	 */
+	list_for_each_entry(dev_ent, &vio_cmo.device_list, list)
+		if (viodev == dev_ent->viodev) {
+			list_del(&dev_ent->list);
+			kfree(dev_ent);
+			break;
+		}
+
+	/*
+	 * Devices may not require any entitlement and they do not need
+	 * to be processed.  Otherwise, return the device's entitlement
+	 * back to the pools.
+	 */
+	if (viodev->cmo.entitled) {
+		/*
+		 * This device has not yet left the OF tree, it's
+		 * minimum entitlement remains in vio_cmo.min and
+		 * vio_cmo.desired
+		 */
+		vio_cmo.desired -= (viodev->cmo.desired - VIO_CMO_MIN_ENT);
+
+		/*
+		 * Save min allocation for device in reserve as long
+		 * as it exists in OF tree as determined by later
+		 * balance operation
+		 */
+		viodev->cmo.entitled -= VIO_CMO_MIN_ENT;
+
+		/* Replenish spare from freed reserve pool */
+		if (viodev->cmo.entitled && (vio_cmo.spare < VIO_CMO_MIN_ENT)) {
+			tmp = min(viodev->cmo.entitled, (VIO_CMO_MIN_ENT -
+			                                 vio_cmo.spare));
+			vio_cmo.spare += tmp;
+			viodev->cmo.entitled -= tmp;
+		}
+
+		/* Remaining reserve goes to excess pool */
+		vio_cmo.excess.size += viodev->cmo.entitled;
+		vio_cmo.excess.free += viodev->cmo.entitled;
+		vio_cmo.reserve.size -= viodev->cmo.entitled;
+
+		/*
+		 * Until the device is removed it will keep a
+		 * minimum entitlement; this will guarantee that
+		 * a module unload/load will result in a success.
+		 */
+		viodev->cmo.entitled = VIO_CMO_MIN_ENT;
+		viodev->cmo.desired = VIO_CMO_MIN_ENT;
+		atomic_set(&viodev->cmo.allocs_failed, 0);
+	}
+
+	spin_unlock_irqrestore(&vio_cmo.lock, flags);
+}
+
+static void vio_cmo_set_dma_ops(struct vio_dev *viodev)
+{
+	vio_dma_mapping_ops.dma_supported = dma_iommu_ops.dma_supported;
+	viodev->dev.archdata.dma_ops = &vio_dma_mapping_ops;
+}
+
+/**
+ * vio_cmo_bus_init - CMO entitlement initialization at bus init time
+ *
+ * Set up the reserve and excess entitlement pools based on available
+ * system entitlement and the number of devices in the OF tree that
+ * require entitlement in the reserve pool.
+ */
+static void vio_cmo_bus_init(void)
+{
+	struct hvcall_mpp_data mpp_data;
+	int err;
+
+	memset(&vio_cmo, 0, sizeof(struct vio_cmo));
+	spin_lock_init(&vio_cmo.lock);
+	INIT_LIST_HEAD(&vio_cmo.device_list);
+	INIT_DELAYED_WORK(&vio_cmo.balance_q, vio_cmo_balance);
+
+	/* Get current system entitlement */
+	err = h_get_mpp(&mpp_data);
+
+	/*
+	 * On failure, continue with entitlement set to 0, will panic()
+	 * later when spare is reserved.
+	 */
+	if (err != H_SUCCESS) {
+		printk(KERN_ERR "%s: unable to determine system IO "\
+		       "entitlement. (%d)\n", __func__, err);
+		vio_cmo.entitled = 0;
+	} else {
+		vio_cmo.entitled = mpp_data.entitled_mem;
+	}
+
+	/* Set reservation and check against entitlement */
+	vio_cmo.spare = VIO_CMO_MIN_ENT;
+	vio_cmo.reserve.size = vio_cmo.spare;
+	vio_cmo.reserve.size += (vio_cmo_num_OF_devs() *
+	                         VIO_CMO_MIN_ENT);
+	if (vio_cmo.reserve.size > vio_cmo.entitled) {
+		printk(KERN_ERR "%s: insufficient system entitlement\n",
+		       __func__);
+		panic("%s: Insufficient system entitlement", __func__);
+	}
+
+	/* Set the remaining accounting variables */
+	vio_cmo.excess.size = vio_cmo.entitled - vio_cmo.reserve.size;
+	vio_cmo.excess.free = vio_cmo.excess.size;
+	vio_cmo.min = vio_cmo.reserve.size;
+	vio_cmo.desired = vio_cmo.reserve.size;
+}
+
+/* sysfs device functions and data structures for CMO */
+
+#define viodev_cmo_rd_attr(name)                                        \
+static ssize_t viodev_cmo_##name##_show(struct device *dev,             \
+                                        struct device_attribute *attr,  \
+                                         char *buf)                     \
+{                                                                       \
+	return sprintf(buf, "%lu\n", to_vio_dev(dev)->cmo.name);        \
+}
+
+static ssize_t viodev_cmo_allocs_failed_show(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	struct vio_dev *viodev = to_vio_dev(dev);
+	return sprintf(buf, "%d\n", atomic_read(&viodev->cmo.allocs_failed));
+}
+
+static ssize_t viodev_cmo_allocs_failed_reset(struct device *dev,
+		struct device_attribute *attr, const char *buf, size_t count)
+{
+	struct vio_dev *viodev = to_vio_dev(dev);
+	atomic_set(&viodev->cmo.allocs_failed, 0);
+	return count;
+}
+
+static ssize_t viodev_cmo_desired_set(struct device *dev,
+		struct device_attribute *attr, const char *buf, size_t count)
+{
+	struct vio_dev *viodev = to_vio_dev(dev);
+	size_t new_desired;
+	int ret;
+
+	ret = strict_strtoul(buf, 10, &new_desired);
+	if (ret)
+		return ret;
+
+	vio_cmo_set_dev_desired(viodev, new_desired);
+	return count;
+}
+
+viodev_cmo_rd_attr(desired);
+viodev_cmo_rd_attr(entitled);
+viodev_cmo_rd_attr(allocated);
+
+static ssize_t name_show(struct device *, struct device_attribute *, char *);
+static ssize_t devspec_show(struct device *, struct device_attribute *, char *);
+static struct device_attribute vio_cmo_dev_attrs[] = {
+	__ATTR_RO(name),
+	__ATTR_RO(devspec),
+	__ATTR(cmo_desired,       S_IWUSR|S_IRUSR|S_IWGRP|S_IRGRP|S_IROTH,
+	       viodev_cmo_desired_show, viodev_cmo_desired_set),
+	__ATTR(cmo_entitled,      S_IRUGO, viodev_cmo_entitled_show,      NULL),
+	__ATTR(cmo_allocated,     S_IRUGO, viodev_cmo_allocated_show,     NULL),
+	__ATTR(cmo_allocs_failed, S_IWUSR|S_IRUSR|S_IWGRP|S_IRGRP|S_IROTH,
+	       viodev_cmo_allocs_failed_show, viodev_cmo_allocs_failed_reset),
+	__ATTR_NULL
+};
+
+/* sysfs bus functions and data structures for CMO */
+
+#define viobus_cmo_rd_attr(name)                                        \
+static ssize_t                                                          \
+viobus_cmo_##name##_show(struct bus_type *bt, char *buf)                \
+{                                                                       \
+	return sprintf(buf, "%lu\n", vio_cmo.name);                     \
+}
+
+#define viobus_cmo_pool_rd_attr(name, var)                              \
+static ssize_t                                                          \
+viobus_cmo_##name##_pool_show_##var(struct bus_type *bt, char *buf)     \
+{                                                                       \
+	return sprintf(buf, "%lu\n", vio_cmo.name.var);                 \
+}
+
+static ssize_t viobus_cmo_high_reset(struct bus_type *bt, const char *buf,
+                                     size_t count)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&vio_cmo.lock, flags);
+	vio_cmo.high = vio_cmo.curr;
+	spin_unlock_irqrestore(&vio_cmo.lock, flags);
+
+	return count;
+}
+
+viobus_cmo_rd_attr(entitled);
+viobus_cmo_pool_rd_attr(reserve, size);
+viobus_cmo_pool_rd_attr(excess, size);
+viobus_cmo_pool_rd_attr(excess, free);
+viobus_cmo_rd_attr(spare);
+viobus_cmo_rd_attr(min);
+viobus_cmo_rd_attr(desired);
+viobus_cmo_rd_attr(curr);
+viobus_cmo_rd_attr(high);
+
+static struct bus_attribute vio_cmo_bus_attrs[] = {
+	__ATTR(cmo_entitled, S_IRUGO, viobus_cmo_entitled_show, NULL),
+	__ATTR(cmo_reserve_size, S_IRUGO, viobus_cmo_reserve_pool_show_size, NULL),
+	__ATTR(cmo_excess_size, S_IRUGO, viobus_cmo_excess_pool_show_size, NULL),
+	__ATTR(cmo_excess_free, S_IRUGO, viobus_cmo_excess_pool_show_free, NULL),
+	__ATTR(cmo_spare,   S_IRUGO, viobus_cmo_spare_show,   NULL),
+	__ATTR(cmo_min,     S_IRUGO, viobus_cmo_min_show,     NULL),
+	__ATTR(cmo_desired, S_IRUGO, viobus_cmo_desired_show, NULL),
+	__ATTR(cmo_curr,    S_IRUGO, viobus_cmo_curr_show,    NULL),
+	__ATTR(cmo_high,    S_IWUSR|S_IRUSR|S_IWGRP|S_IRGRP|S_IROTH,
+	       viobus_cmo_high_show, viobus_cmo_high_reset),
+	__ATTR_NULL
+};
+
+static void vio_cmo_sysfs_init(void)
+{
+	vio_bus_type.dev_attrs = vio_cmo_dev_attrs;
+	vio_bus_type.bus_attrs = vio_cmo_bus_attrs;
+}
+#else /* CONFIG_PPC_SMLPAR */
+/* Dummy functions for iSeries platform */
+int vio_cmo_entitlement_update(size_t new_entitlement) { return 0; }
+void vio_cmo_set_dev_desired(struct vio_dev *viodev, size_t desired) {}
+static int vio_cmo_bus_probe(struct vio_dev *viodev) { return 0; }
+static void vio_cmo_bus_remove(struct vio_dev *viodev) {}
+static void vio_cmo_set_dma_ops(struct vio_dev *viodev) {}
+static void vio_cmo_bus_init() {}
+static void vio_cmo_sysfs_init() { }
+#endif /* CONFIG_PPC_SMLPAR */
+EXPORT_SYMBOL(vio_cmo_entitlement_update);
+EXPORT_SYMBOL(vio_cmo_set_dev_desired);
+
 static struct iommu_table *vio_build_iommu_table(struct vio_dev *dev)
 {
 	const unsigned char *dma_window;
@@ -114,8 +1105,17 @@ static int vio_bus_probe(struct device *dev)
 		return error;
 
 	id = vio_match_device(viodrv->id_table, viodev);
-	if (id)
+	if (id) {
+		memset(&viodev->cmo, 0, sizeof(viodev->cmo));
+		if (firmware_has_feature(FW_FEATURE_CMO)) {
+			error = vio_cmo_bus_probe(viodev);
+			if (error)
+				return error;
+		}
 		error = viodrv->probe(viodev, id);
+		if (error)
+			vio_cmo_bus_remove(viodev);
+	}
 
 	return error;
 }
@@ -125,12 +1125,23 @@ static int vio_bus_remove(struct device *dev)
 {
 	struct vio_dev *viodev = to_vio_dev(dev);
 	struct vio_driver *viodrv = to_vio_driver(dev->driver);
+	struct device *devptr;
+	int ret = 1;
+
+	/*
+	 * Hold a reference to the device after the remove function is called
+	 * to allow for CMO accounting cleanup for the device.
+	 */
+	devptr = get_device(dev);
 
 	if (viodrv->remove)
-		return viodrv->remove(viodev);
+		ret = viodrv->remove(viodev);
 
-	/* driver can't remove */
-	return 1;
+	if (!ret && firmware_has_feature(FW_FEATURE_CMO))
+		vio_cmo_bus_remove(viodev);
+
+	put_device(devptr);
+	return ret;
 }
 
 /**
@@ -215,7 +1226,11 @@ struct vio_dev *vio_register_device_node(struct device_node *of_node)
 			viodev->unit_address = *unit_address;
 	}
 	viodev->dev.archdata.of_node = of_node_get(of_node);
-	viodev->dev.archdata.dma_ops = &dma_iommu_ops;
+
+	if (firmware_has_feature(FW_FEATURE_CMO))
+		vio_cmo_set_dma_ops(viodev);
+	else
+		viodev->dev.archdata.dma_ops = &dma_iommu_ops;
 	viodev->dev.archdata.dma_data = vio_build_iommu_table(viodev);
 	viodev->dev.archdata.numa_node = of_node_to_nid(of_node);
 
@@ -245,6 +1260,9 @@ static int __init vio_bus_init(void)
 	int err;
 	struct device_node *node_vroot;
 
+	if (firmware_has_feature(FW_FEATURE_CMO))
+		vio_cmo_sysfs_init();
+
 	err = bus_register(&vio_bus_type);
 	if (err) {
 		printk(KERN_ERR "failed to register VIO bus\n");
@@ -262,6 +1280,9 @@ static int __init vio_bus_init(void)
 		return err;
 	}
 
+	if (firmware_has_feature(FW_FEATURE_CMO))
+		vio_cmo_bus_init();
+
 	node_vroot = of_find_node_by_name(NULL, "vdevice");
 	if (node_vroot) {
 		struct device_node *of_node;
diff --git a/include/asm-powerpc/vio.h b/include/asm-powerpc/vio.h
index 56512a968dab..0a290a195946 100644
--- a/include/asm-powerpc/vio.h
+++ b/include/asm-powerpc/vio.h
@@ -39,16 +39,32 @@
 #define VIO_IRQ_DISABLE		0UL
 #define VIO_IRQ_ENABLE		1UL
 
+/*
+ * VIO CMO minimum entitlement for all devices and spare entitlement
+ */
+#define VIO_CMO_MIN_ENT 1562624
+
 struct iommu_table;
 
-/*
- * The vio_dev structure is used to describe virtual I/O devices.
+/**
+ * vio_dev - This structure is used to describe virtual I/O devices.
+ *
+ * @desired: set from return of driver's get_desired_dma() function
+ * @entitled: bytes of IO data that has been reserved for this device.
+ * @allocated: bytes of IO data currently in use by the device.
+ * @allocs_failed: number of DMA failures due to insufficient entitlement.
  */
 struct vio_dev {
 	const char *name;
 	const char *type;
 	uint32_t unit_address;
 	unsigned int irq;
+	struct {
+		size_t desired;
+		size_t entitled;
+		size_t allocated;
+		atomic_t allocs_failed;
+	} cmo;
 	struct device dev;
 };
 
@@ -56,12 +72,19 @@ struct vio_driver {
 	const struct vio_device_id *id_table;
 	int (*probe)(struct vio_dev *dev, const struct vio_device_id *id);
 	int (*remove)(struct vio_dev *dev);
+	/* A driver must have a get_desired_dma() function to
+	 * be loaded in a CMO environment if it uses DMA.
+	 */
+	unsigned long (*get_desired_dma)(struct vio_dev *dev);
 	struct device_driver driver;
 };
 
 extern int vio_register_driver(struct vio_driver *drv);
 extern void vio_unregister_driver(struct vio_driver *drv);
 
+extern int vio_cmo_entitlement_update(size_t);
+extern void vio_cmo_set_dev_desired(struct vio_dev *viodev, size_t desired);
+
 extern void __devinit vio_unregister_device(struct vio_dev *dev);
 
 struct device_node;

From 22e1a4dd3f2a9009d1d8896a5e833b6094877008 Mon Sep 17 00:00:00 2001
From: Nathan Fontenot <nfont@austin.ibm.com>
Date: Thu, 24 Jul 2008 04:31:52 +1000
Subject: [PATCH 0642/1171] powerpc/pseries: Verify CMO memory entitlement
 updates with virtual I/O

Verify memory entitlement updates can be handled by vio.

Signed-off-by: Nathan Fontenot <nfont@austin.ibm.com>
Signed-off-by: Robert Jennings <rcj@linux.vnet.ibm.com>
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/lparcfg.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/arch/powerpc/kernel/lparcfg.c b/arch/powerpc/kernel/lparcfg.c
index 848c3e5a6370..64381a204a58 100644
--- a/arch/powerpc/kernel/lparcfg.c
+++ b/arch/powerpc/kernel/lparcfg.c
@@ -34,6 +34,7 @@
 #include <asm/time.h>
 #include <asm/prom.h>
 #include <asm/vdso_datapage.h>
+#include <asm/vio.h>
 
 #define MODULE_VERS "1.8"
 #define MODULE_NAME "lparcfg"
@@ -527,6 +528,15 @@ static ssize_t update_mpp(u64 *entitlement, u8 *weight)
 	u8 new_weight;
 	ssize_t rc;
 
+	if (entitlement) {
+		/* Check with vio to ensure the new memory entitlement
+		 * can be handled.
+		 */
+		rc = vio_cmo_entitlement_update(*entitlement);
+		if (rc)
+			return rc;
+	}
+
 	rc = h_get_mpp(&mpp_data);
 	if (rc)
 		return rc;

From ea866e6526b8a2ead92875732d41b26fdb470312 Mon Sep 17 00:00:00 2001
From: Santiago Leon <santil@us.ibm.com>
Date: Thu, 24 Jul 2008 04:34:23 +1000
Subject: [PATCH 0643/1171] ibmveth: Automatically enable larger rx buffer
 pools for larger mtu

Activates larger rx buffer pools when the MTU is changed to a larger
value.  This patch de-activates the large rx buffer pools when the MTU
changes to a smaller value.

Signed-off-by: Santiago Leon <santil@us.ibm.com>
Signed-off-by: Robert Jennings <rcj@linux.vnet.ibm.com>
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 drivers/net/ibmveth.c | 22 ++++++++++++++--------
 1 file changed, 14 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ibmveth.c b/drivers/net/ibmveth.c
index 00527805e4f1..007ca8735a94 100644
--- a/drivers/net/ibmveth.c
+++ b/drivers/net/ibmveth.c
@@ -1054,7 +1054,6 @@ static int ibmveth_change_mtu(struct net_device *dev, int new_mtu)
 {
 	struct ibmveth_adapter *adapter = dev->priv;
 	int new_mtu_oh = new_mtu + IBMVETH_BUFF_OH;
-	int reinit = 0;
 	int i, rc;
 
 	if (new_mtu < IBMVETH_MAX_MTU)
@@ -1067,15 +1066,21 @@ static int ibmveth_change_mtu(struct net_device *dev, int new_mtu)
 	if (i == IbmVethNumBufferPools)
 		return -EINVAL;
 
-	/* Look for an active buffer pool that can hold the new MTU */
-	for(i = 0; i<IbmVethNumBufferPools; i++) {
-		if (!adapter->rx_buff_pool[i].active) {
-			adapter->rx_buff_pool[i].active = 1;
-			reinit = 1;
+	/* Deactivate all the buffer pools so that the next loop can activate
+	   only the buffer pools necessary to hold the new MTU */
+	for (i = 0; i < IbmVethNumBufferPools; i++)
+		if (adapter->rx_buff_pool[i].active) {
+			ibmveth_free_buffer_pool(adapter,
+						 &adapter->rx_buff_pool[i]);
+			adapter->rx_buff_pool[i].active = 0;
 		}
 
+	/* Look for an active buffer pool that can hold the new MTU */
+	for(i = 0; i<IbmVethNumBufferPools; i++) {
+		adapter->rx_buff_pool[i].active = 1;
+
 		if (new_mtu_oh < adapter->rx_buff_pool[i].buff_size) {
-			if (reinit && netif_running(adapter->netdev)) {
+			if (netif_running(adapter->netdev)) {
 				adapter->pool_config = 1;
 				ibmveth_close(adapter->netdev);
 				adapter->pool_config = 0;
@@ -1402,14 +1407,15 @@ const char * buf, size_t count)
 				return -EPERM;
 			}
 
-			pool->active = 0;
 			if (netif_running(netdev)) {
 				adapter->pool_config = 1;
 				ibmveth_close(netdev);
+				pool->active = 0;
 				adapter->pool_config = 0;
 				if ((rc = ibmveth_open(netdev)))
 					return rc;
 			}
+			pool->active = 0;
 		}
 	} else if (attr == &veth_num_attr) {
 		if (value <= 0 || value > IBMVETH_MAX_POOL_COUNT)

From 1096d63d8e7d226630706e15648705d0187787e4 Mon Sep 17 00:00:00 2001
From: Robert Jennings <rcj@linux.vnet.ibm.com>
Date: Thu, 24 Jul 2008 04:34:52 +1000
Subject: [PATCH 0644/1171] ibmveth: enable driver for CMO

Enable ibmveth for Cooperative Memory Overcommitment (CMO).  For this driver
it means calculating a desired amount of IO memory based on the current MTU
and updating this value with the bus when MTU changes occur.  Because DMA
mappings can fail, we have added a bounce buffer for temporary cases where
the driver can not map IO memory for the buffer pool.

The following changes are made to enable the driver for CMO:
 * DMA mapping errors will not result in error messages if entitlement has
   been exceeded and resources were not available.
 * DMA mapping errors are handled gracefully, ibmveth_replenish_buffer_pool()
   is corrected to check the return from dma_map_single and fail gracefully.
 * The driver will have a get_desired_dma function defined to function
   in a CMO environment.
 * When the MTU is changed, the driver will update the device IO entitlement

Signed-off-by: Robert Jennings <rcj@linux.vnet.ibm.com>
Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: Santiago Leon <santil@us.ibm.com>
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 drivers/net/ibmveth.c | 169 +++++++++++++++++++++++++++++++++---------
 drivers/net/ibmveth.h |   5 ++
 2 files changed, 140 insertions(+), 34 deletions(-)

diff --git a/drivers/net/ibmveth.c b/drivers/net/ibmveth.c
index 007ca8735a94..e5a6e2e84540 100644
--- a/drivers/net/ibmveth.c
+++ b/drivers/net/ibmveth.c
@@ -33,6 +33,7 @@
 */
 
 #include <linux/module.h>
+#include <linux/moduleparam.h>
 #include <linux/types.h>
 #include <linux/errno.h>
 #include <linux/ioport.h>
@@ -52,7 +53,9 @@
 #include <asm/hvcall.h>
 #include <asm/atomic.h>
 #include <asm/vio.h>
+#include <asm/iommu.h>
 #include <asm/uaccess.h>
+#include <asm/firmware.h>
 #include <linux/seq_file.h>
 
 #include "ibmveth.h"
@@ -94,8 +97,10 @@ static void ibmveth_proc_register_adapter(struct ibmveth_adapter *adapter);
 static void ibmveth_proc_unregister_adapter(struct ibmveth_adapter *adapter);
 static irqreturn_t ibmveth_interrupt(int irq, void *dev_instance);
 static void ibmveth_rxq_harvest_buffer(struct ibmveth_adapter *adapter);
+static unsigned long ibmveth_get_desired_dma(struct vio_dev *vdev);
 static struct kobj_type ktype_veth_pool;
 
+
 #ifdef CONFIG_PROC_FS
 #define IBMVETH_PROC_DIR "ibmveth"
 static struct proc_dir_entry *ibmveth_proc_dir;
@@ -226,16 +231,16 @@ static void ibmveth_replenish_buffer_pool(struct ibmveth_adapter *adapter, struc
 	u32 i;
 	u32 count = pool->size - atomic_read(&pool->available);
 	u32 buffers_added = 0;
+	struct sk_buff *skb;
+	unsigned int free_index, index;
+	u64 correlator;
+	unsigned long lpar_rc;
+	dma_addr_t dma_addr;
 
 	mb();
 
 	for(i = 0; i < count; ++i) {
-		struct sk_buff *skb;
-		unsigned int free_index, index;
-		u64 correlator;
 		union ibmveth_buf_desc desc;
-		unsigned long lpar_rc;
-		dma_addr_t dma_addr;
 
 		skb = alloc_skb(pool->buff_size, GFP_ATOMIC);
 
@@ -255,6 +260,9 @@ static void ibmveth_replenish_buffer_pool(struct ibmveth_adapter *adapter, struc
 		dma_addr = dma_map_single(&adapter->vdev->dev, skb->data,
 				pool->buff_size, DMA_FROM_DEVICE);
 
+		if (dma_mapping_error(dma_addr))
+			goto failure;
+
 		pool->free_map[free_index] = IBM_VETH_INVALID_MAP;
 		pool->dma_addr[index] = dma_addr;
 		pool->skbuff[index] = skb;
@@ -267,25 +275,32 @@ static void ibmveth_replenish_buffer_pool(struct ibmveth_adapter *adapter, struc
 
 		lpar_rc = h_add_logical_lan_buffer(adapter->vdev->unit_address, desc.desc);
 
-		if(lpar_rc != H_SUCCESS) {
-			pool->free_map[free_index] = index;
-			pool->skbuff[index] = NULL;
-			if (pool->consumer_index == 0)
-				pool->consumer_index = pool->size - 1;
-			else
-				pool->consumer_index--;
-			dma_unmap_single(&adapter->vdev->dev,
-					pool->dma_addr[index], pool->buff_size,
-					DMA_FROM_DEVICE);
-			dev_kfree_skb_any(skb);
-			adapter->replenish_add_buff_failure++;
-			break;
-		} else {
+		if (lpar_rc != H_SUCCESS)
+			goto failure;
+		else {
 			buffers_added++;
 			adapter->replenish_add_buff_success++;
 		}
 	}
 
+	mb();
+	atomic_add(buffers_added, &(pool->available));
+	return;
+
+failure:
+	pool->free_map[free_index] = index;
+	pool->skbuff[index] = NULL;
+	if (pool->consumer_index == 0)
+		pool->consumer_index = pool->size - 1;
+	else
+		pool->consumer_index--;
+	if (!dma_mapping_error(dma_addr))
+		dma_unmap_single(&adapter->vdev->dev,
+		                 pool->dma_addr[index], pool->buff_size,
+		                 DMA_FROM_DEVICE);
+	dev_kfree_skb_any(skb);
+	adapter->replenish_add_buff_failure++;
+
 	mb();
 	atomic_add(buffers_added, &(pool->available));
 }
@@ -297,7 +312,7 @@ static void ibmveth_replenish_task(struct ibmveth_adapter *adapter)
 
 	adapter->replenish_task_cycles++;
 
-	for(i = 0; i < IbmVethNumBufferPools; i++)
+	for (i = (IbmVethNumBufferPools - 1); i >= 0; i--)
 		if(adapter->rx_buff_pool[i].active)
 			ibmveth_replenish_buffer_pool(adapter,
 						     &adapter->rx_buff_pool[i]);
@@ -472,6 +487,18 @@ static void ibmveth_cleanup(struct ibmveth_adapter *adapter)
 		if (adapter->rx_buff_pool[i].active)
 			ibmveth_free_buffer_pool(adapter,
 						 &adapter->rx_buff_pool[i]);
+
+	if (adapter->bounce_buffer != NULL) {
+		if (!dma_mapping_error(adapter->bounce_buffer_dma)) {
+			dma_unmap_single(&adapter->vdev->dev,
+					adapter->bounce_buffer_dma,
+					adapter->netdev->mtu + IBMVETH_BUFF_OH,
+					DMA_BIDIRECTIONAL);
+			adapter->bounce_buffer_dma = DMA_ERROR_CODE;
+		}
+		kfree(adapter->bounce_buffer);
+		adapter->bounce_buffer = NULL;
+	}
 }
 
 static int ibmveth_register_logical_lan(struct ibmveth_adapter *adapter,
@@ -607,6 +634,24 @@ static int ibmveth_open(struct net_device *netdev)
 		return rc;
 	}
 
+	adapter->bounce_buffer =
+	    kmalloc(netdev->mtu + IBMVETH_BUFF_OH, GFP_KERNEL);
+	if (!adapter->bounce_buffer) {
+		ibmveth_error_printk("unable to allocate bounce buffer\n");
+		ibmveth_cleanup(adapter);
+		napi_disable(&adapter->napi);
+		return -ENOMEM;
+	}
+	adapter->bounce_buffer_dma =
+	    dma_map_single(&adapter->vdev->dev, adapter->bounce_buffer,
+			   netdev->mtu + IBMVETH_BUFF_OH, DMA_BIDIRECTIONAL);
+	if (dma_mapping_error(adapter->bounce_buffer_dma)) {
+		ibmveth_error_printk("unable to map bounce buffer\n");
+		ibmveth_cleanup(adapter);
+		napi_disable(&adapter->napi);
+		return -ENOMEM;
+	}
+
 	ibmveth_debug_printk("initial replenish cycle\n");
 	ibmveth_interrupt(netdev->irq, netdev);
 
@@ -853,10 +898,12 @@ static int ibmveth_start_xmit(struct sk_buff *skb, struct net_device *netdev)
 	unsigned int tx_packets = 0;
 	unsigned int tx_send_failed = 0;
 	unsigned int tx_map_failed = 0;
+	int used_bounce = 0;
+	unsigned long data_dma_addr;
 
 	desc.fields.flags_len = IBMVETH_BUF_VALID | skb->len;
-	desc.fields.address = dma_map_single(&adapter->vdev->dev, skb->data,
-					     skb->len, DMA_TO_DEVICE);
+	data_dma_addr = dma_map_single(&adapter->vdev->dev, skb->data,
+				       skb->len, DMA_TO_DEVICE);
 
 	if (skb->ip_summed == CHECKSUM_PARTIAL &&
 	    ip_hdr(skb)->protocol != IPPROTO_TCP && skb_checksum_help(skb)) {
@@ -875,12 +922,16 @@ static int ibmveth_start_xmit(struct sk_buff *skb, struct net_device *netdev)
 		buf[1] = 0;
 	}
 
-	if (dma_mapping_error(desc.fields.address)) {
-		ibmveth_error_printk("tx: unable to map xmit buffer\n");
+	if (dma_mapping_error(data_dma_addr)) {
+		if (!firmware_has_feature(FW_FEATURE_CMO))
+			ibmveth_error_printk("tx: unable to map xmit buffer\n");
+		skb_copy_from_linear_data(skb, adapter->bounce_buffer,
+					  skb->len);
+		desc.fields.address = adapter->bounce_buffer_dma;
 		tx_map_failed++;
-		tx_dropped++;
-		goto out;
-	}
+		used_bounce = 1;
+	} else
+		desc.fields.address = data_dma_addr;
 
 	/* send the frame. Arbitrarily set retrycount to 1024 */
 	correlator = 0;
@@ -904,8 +955,9 @@ static int ibmveth_start_xmit(struct sk_buff *skb, struct net_device *netdev)
 		netdev->trans_start = jiffies;
 	}
 
-	dma_unmap_single(&adapter->vdev->dev, desc.fields.address,
-			 skb->len, DMA_TO_DEVICE);
+	if (!used_bounce)
+		dma_unmap_single(&adapter->vdev->dev, data_dma_addr,
+				 skb->len, DMA_TO_DEVICE);
 
 out:	spin_lock_irqsave(&adapter->stats_lock, flags);
 	netdev->stats.tx_dropped += tx_dropped;
@@ -1053,8 +1105,9 @@ static void ibmveth_set_multicast_list(struct net_device *netdev)
 static int ibmveth_change_mtu(struct net_device *dev, int new_mtu)
 {
 	struct ibmveth_adapter *adapter = dev->priv;
+	struct vio_dev *viodev = adapter->vdev;
 	int new_mtu_oh = new_mtu + IBMVETH_BUFF_OH;
-	int i, rc;
+	int i;
 
 	if (new_mtu < IBMVETH_MAX_MTU)
 		return -EINVAL;
@@ -1085,10 +1138,15 @@ static int ibmveth_change_mtu(struct net_device *dev, int new_mtu)
 				ibmveth_close(adapter->netdev);
 				adapter->pool_config = 0;
 				dev->mtu = new_mtu;
-				if ((rc = ibmveth_open(adapter->netdev)))
-					return rc;
-			} else
-				dev->mtu = new_mtu;
+				vio_cmo_set_dev_desired(viodev,
+						ibmveth_get_desired_dma
+						(viodev));
+				return ibmveth_open(adapter->netdev);
+			}
+			dev->mtu = new_mtu;
+			vio_cmo_set_dev_desired(viodev,
+						ibmveth_get_desired_dma
+						(viodev));
 			return 0;
 		}
 	}
@@ -1103,6 +1161,46 @@ static void ibmveth_poll_controller(struct net_device *dev)
 }
 #endif
 
+/**
+ * ibmveth_get_desired_dma - Calculate IO memory desired by the driver
+ *
+ * @vdev: struct vio_dev for the device whose desired IO mem is to be returned
+ *
+ * Return value:
+ *	Number of bytes of IO data the driver will need to perform well.
+ */
+static unsigned long ibmveth_get_desired_dma(struct vio_dev *vdev)
+{
+	struct net_device *netdev = dev_get_drvdata(&vdev->dev);
+	struct ibmveth_adapter *adapter;
+	unsigned long ret;
+	int i;
+	int rxqentries = 1;
+
+	/* netdev inits at probe time along with the structures we need below*/
+	if (netdev == NULL)
+		return IOMMU_PAGE_ALIGN(IBMVETH_IO_ENTITLEMENT_DEFAULT);
+
+	adapter = netdev_priv(netdev);
+
+	ret = IBMVETH_BUFF_LIST_SIZE + IBMVETH_FILT_LIST_SIZE;
+	ret += IOMMU_PAGE_ALIGN(netdev->mtu);
+
+	for (i = 0; i < IbmVethNumBufferPools; i++) {
+		/* add the size of the active receive buffers */
+		if (adapter->rx_buff_pool[i].active)
+			ret +=
+			    adapter->rx_buff_pool[i].size *
+			    IOMMU_PAGE_ALIGN(adapter->rx_buff_pool[i].
+			            buff_size);
+		rxqentries += adapter->rx_buff_pool[i].size;
+	}
+	/* add the size of the receive queue entries */
+	ret += IOMMU_PAGE_ALIGN(rxqentries * sizeof(struct ibmveth_rx_q_entry));
+
+	return ret;
+}
+
 static int __devinit ibmveth_probe(struct vio_dev *dev, const struct vio_device_id *id)
 {
 	int rc, i;
@@ -1247,6 +1345,8 @@ static int __devexit ibmveth_remove(struct vio_dev *dev)
 	ibmveth_proc_unregister_adapter(adapter);
 
 	free_netdev(netdev);
+	dev_set_drvdata(&dev->dev, NULL);
+
 	return 0;
 }
 
@@ -1491,6 +1591,7 @@ static struct vio_driver ibmveth_driver = {
 	.id_table	= ibmveth_device_table,
 	.probe		= ibmveth_probe,
 	.remove		= ibmveth_remove,
+	.get_desired_dma = ibmveth_get_desired_dma,
 	.driver		= {
 		.name	= ibmveth_driver_name,
 		.owner	= THIS_MODULE,
diff --git a/drivers/net/ibmveth.h b/drivers/net/ibmveth.h
index 41f61cd18852..d28186948752 100644
--- a/drivers/net/ibmveth.h
+++ b/drivers/net/ibmveth.h
@@ -93,9 +93,12 @@ static inline long h_illan_attributes(unsigned long unit_address,
   plpar_hcall_norets(H_CHANGE_LOGICAL_LAN_MAC, ua, mac)
 
 #define IbmVethNumBufferPools 5
+#define IBMVETH_IO_ENTITLEMENT_DEFAULT 4243456 /* MTU of 1500 needs 4.2Mb */
 #define IBMVETH_BUFF_OH 22 /* Overhead: 14 ethernet header + 8 opaque handle */
 #define IBMVETH_MAX_MTU 68
 #define IBMVETH_MAX_POOL_COUNT 4096
+#define IBMVETH_BUFF_LIST_SIZE 4096
+#define IBMVETH_FILT_LIST_SIZE 4096
 #define IBMVETH_MAX_BUF_SIZE (1024 * 128)
 
 static int pool_size[] = { 512, 1024 * 2, 1024 * 16, 1024 * 32, 1024 * 64 };
@@ -143,6 +146,8 @@ struct ibmveth_adapter {
     struct ibmveth_rx_q rx_queue;
     int pool_config;
     int rx_csum;
+    void *bounce_buffer;
+    dma_addr_t bounce_buffer_dma;
 
     /* adapter specific stats */
     u64 replenish_task_cycles;

From 7912a0ac5907df1f8b214b3ca15ccf96129daae0 Mon Sep 17 00:00:00 2001
From: Robert Jennings <rcj@linux.vnet.ibm.com>
Date: Thu, 24 Jul 2008 04:35:27 +1000
Subject: [PATCH 0645/1171] ibmvscsi: driver enablement for CMO

Enable the driver to function in a Cooperative Memory Overcommitment (CMO)
environment.

The following changes are made to enable the driver for CMO:
 * DMA mapping errors will not result in error messages if entitlement has
   been exceeded and resources were not available.
 * The driver has a get_desired_dma function defined to function
   in a CMO environment. It will indicate how much IO memory it would like
   to function.

Signed-off-by: Robert Jennings <rcj@linux.vnet.ibm.com>
Acked by: Brian King <brking@linux.vnet.ibm.com>
Acked-by: Paul Mackerras <paulus@samba.org>
Acked-by: James Bottomley <James.Bottomley@HansenPartnership.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 drivers/scsi/ibmvscsi/ibmvscsi.c | 45 +++++++++++++++++++++++++++-----
 drivers/scsi/ibmvscsi/ibmvscsi.h |  2 ++
 2 files changed, 40 insertions(+), 7 deletions(-)

diff --git a/drivers/scsi/ibmvscsi/ibmvscsi.c b/drivers/scsi/ibmvscsi/ibmvscsi.c
index 5d23368a1bce..20000ec79b04 100644
--- a/drivers/scsi/ibmvscsi/ibmvscsi.c
+++ b/drivers/scsi/ibmvscsi/ibmvscsi.c
@@ -72,6 +72,7 @@
 #include <linux/delay.h>
 #include <asm/firmware.h>
 #include <asm/vio.h>
+#include <asm/firmware.h>
 #include <scsi/scsi.h>
 #include <scsi/scsi_cmnd.h>
 #include <scsi/scsi_host.h>
@@ -426,8 +427,10 @@ static int map_sg_data(struct scsi_cmnd *cmd,
 					   SG_ALL * sizeof(struct srp_direct_buf),
 					   &evt_struct->ext_list_token, 0);
 		if (!evt_struct->ext_list) {
-			sdev_printk(KERN_ERR, cmd->device,
-				    "Can't allocate memory for indirect table\n");
+			if (!firmware_has_feature(FW_FEATURE_CMO))
+				sdev_printk(KERN_ERR, cmd->device,
+				            "Can't allocate memory "
+				            "for indirect table\n");
 			return 0;
 		}
 	}
@@ -743,7 +746,9 @@ static int ibmvscsi_queuecommand(struct scsi_cmnd *cmnd,
 	srp_cmd->lun = ((u64) lun) << 48;
 
 	if (!map_data_for_srp_cmd(cmnd, evt_struct, srp_cmd, hostdata->dev)) {
-		sdev_printk(KERN_ERR, cmnd->device, "couldn't convert cmd to srp_cmd\n");
+		if (!firmware_has_feature(FW_FEATURE_CMO))
+			sdev_printk(KERN_ERR, cmnd->device,
+			            "couldn't convert cmd to srp_cmd\n");
 		free_event_struct(&hostdata->pool, evt_struct);
 		return SCSI_MLQUEUE_HOST_BUSY;
 	}
@@ -855,7 +860,10 @@ static void send_mad_adapter_info(struct ibmvscsi_host_data *hostdata)
 					    DMA_BIDIRECTIONAL);
 
 	if (dma_mapping_error(req->buffer)) {
-		dev_err(hostdata->dev, "Unable to map request_buffer for adapter_info!\n");
+		if (!firmware_has_feature(FW_FEATURE_CMO))
+			dev_err(hostdata->dev,
+			        "Unable to map request_buffer for "
+			        "adapter_info!\n");
 		free_event_struct(&hostdata->pool, evt_struct);
 		return;
 	}
@@ -1400,7 +1408,9 @@ static int ibmvscsi_do_host_config(struct ibmvscsi_host_data *hostdata,
 						    DMA_BIDIRECTIONAL);
 
 	if (dma_mapping_error(host_config->buffer)) {
-		dev_err(hostdata->dev, "dma_mapping error getting host config\n");
+		if (!firmware_has_feature(FW_FEATURE_CMO))
+			dev_err(hostdata->dev,
+			        "dma_mapping error getting host config\n");
 		free_event_struct(&hostdata->pool, evt_struct);
 		return -1;
 	}
@@ -1604,7 +1614,7 @@ static struct scsi_host_template driver_template = {
 	.eh_host_reset_handler = ibmvscsi_eh_host_reset_handler,
 	.slave_configure = ibmvscsi_slave_configure,
 	.change_queue_depth = ibmvscsi_change_queue_depth,
-	.cmd_per_lun = 16,
+	.cmd_per_lun = IBMVSCSI_CMDS_PER_LUN_DEFAULT,
 	.can_queue = IBMVSCSI_MAX_REQUESTS_DEFAULT,
 	.this_id = -1,
 	.sg_tablesize = SG_ALL,
@@ -1612,6 +1622,26 @@ static struct scsi_host_template driver_template = {
 	.shost_attrs = ibmvscsi_attrs,
 };
 
+/**
+ * ibmvscsi_get_desired_dma - Calculate IO memory desired by the driver
+ *
+ * @vdev: struct vio_dev for the device whose desired IO mem is to be returned
+ *
+ * Return value:
+ *	Number of bytes of IO data the driver will need to perform well.
+ */
+static unsigned long ibmvscsi_get_desired_dma(struct vio_dev *vdev)
+{
+	/* iu_storage data allocated in initialize_event_pool */
+	unsigned long desired_io = max_requests * sizeof(union viosrp_iu);
+
+	/* add io space for sg data */
+	desired_io += (IBMVSCSI_MAX_SECTORS_DEFAULT *
+	                     IBMVSCSI_CMDS_PER_LUN_DEFAULT);
+
+	return desired_io;
+}
+
 /**
  * Called by bus code for each adapter
  */
@@ -1641,7 +1671,7 @@ static int ibmvscsi_probe(struct vio_dev *vdev, const struct vio_device_id *id)
 	hostdata->host = host;
 	hostdata->dev = dev;
 	atomic_set(&hostdata->request_limit, -1);
-	hostdata->host->max_sectors = 32 * 8; /* default max I/O 32 pages */
+	hostdata->host->max_sectors = IBMVSCSI_MAX_SECTORS_DEFAULT;
 
 	rc = ibmvscsi_ops->init_crq_queue(&hostdata->queue, hostdata, max_requests);
 	if (rc != 0 && rc != H_RESOURCE) {
@@ -1735,6 +1765,7 @@ static struct vio_driver ibmvscsi_driver = {
 	.id_table = ibmvscsi_device_table,
 	.probe = ibmvscsi_probe,
 	.remove = ibmvscsi_remove,
+	.get_desired_dma = ibmvscsi_get_desired_dma,
 	.driver = {
 		.name = "ibmvscsi",
 		.owner = THIS_MODULE,
diff --git a/drivers/scsi/ibmvscsi/ibmvscsi.h b/drivers/scsi/ibmvscsi/ibmvscsi.h
index 46e850e302c7..2d4339d5e16e 100644
--- a/drivers/scsi/ibmvscsi/ibmvscsi.h
+++ b/drivers/scsi/ibmvscsi/ibmvscsi.h
@@ -45,6 +45,8 @@ struct Scsi_Host;
 #define MAX_INDIRECT_BUFS 10
 
 #define IBMVSCSI_MAX_REQUESTS_DEFAULT 100
+#define IBMVSCSI_CMDS_PER_LUN_DEFAULT 16
+#define IBMVSCSI_MAX_SECTORS_DEFAULT 256 /* 32 * 8 = default max I/O 32 pages */
 #define IBMVSCSI_MAX_CMDS_PER_LUN 64
 
 /* ------------------------------------------------------------

From 39c1ffecc6aabcc8105602a95ce769f27bcf6048 Mon Sep 17 00:00:00 2001
From: Brian King <brking@linux.vnet.ibm.com>
Date: Thu, 24 Jul 2008 04:35:48 +1000
Subject: [PATCH 0646/1171] ibmvfc: Add support for collaborative memory
 overcommit

Adds support to the ibmvfc driver for collaborative memory overcommit.

Signed-off-by: Brian King <brking@linux.vnet.ibm.com>
Signed-off-by: Robert Jennings <rcj@linux.vnet.ibm.com>
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 drivers/scsi/ibmvscsi/ibmvfc.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/drivers/scsi/ibmvscsi/ibmvfc.c b/drivers/scsi/ibmvscsi/ibmvfc.c
index eb702b96d57c..c4a7c06793c5 100644
--- a/drivers/scsi/ibmvscsi/ibmvfc.c
+++ b/drivers/scsi/ibmvscsi/ibmvfc.c
@@ -3819,6 +3819,20 @@ static int ibmvfc_remove(struct vio_dev *vdev)
 	return 0;
 }
 
+/**
+ * ibmvfc_get_desired_dma - Calculate DMA resources needed by the driver
+ * @vdev:	vio device struct
+ *
+ * Return value:
+ *	Number of bytes the driver will need to DMA map at the same time in
+ *	order to perform well.
+ */
+static unsigned long ibmvfc_get_desired_dma(struct vio_dev *vdev)
+{
+	unsigned long pool_dma = max_requests * sizeof(union ibmvfc_iu);
+	return pool_dma + ((512 * 1024) * driver_template.cmd_per_lun);
+}
+
 static struct vio_device_id ibmvfc_device_table[] __devinitdata = {
 	{"fcp", "IBM,vfc-client"},
 	{ "", "" }
@@ -3829,6 +3843,7 @@ static struct vio_driver ibmvfc_driver = {
 	.id_table = ibmvfc_device_table,
 	.probe = ibmvfc_probe,
 	.remove = ibmvfc_remove,
+	.get_desired_dma = ibmvfc_get_desired_dma,
 	.driver = {
 		.name = IBMVFC_NAME,
 		.owner = THIS_MODULE,

From 8391e42a5c1f3d757faa5e7f46a4a68f9aa6cb12 Mon Sep 17 00:00:00 2001
From: Nathan Fontenot <nfont@austin.ibm.com>
Date: Thu, 24 Jul 2008 04:36:38 +1000
Subject: [PATCH 0647/1171] powerpc/pseries: Update arch vector to indicate
 support for CMO

Update the architecture vector to indicate that Cooperative Memory
Overcommitment is supported if CONFIG_PPC_SMLPAR is set.

Signed-off-by: Nathan Fontenot <nfont@austin.ibm.com>
Signed-off-by: Robert Jennings <rcj@linux.vnet.ibm.com>
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/prom_init.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c
index 1ea8c8d3ce89..c4ab2195b9cb 100644
--- a/arch/powerpc/kernel/prom_init.c
+++ b/arch/powerpc/kernel/prom_init.c
@@ -643,6 +643,11 @@ static void __init early_cmdline_parse(void)
 #else
 #define OV5_MSI			0x00
 #endif /* CONFIG_PCI_MSI */
+#ifdef CONFIG_PPC_SMLPAR
+#define OV5_CMO			0x80	/* Cooperative Memory Overcommitment */
+#else
+#define OV5_CMO			0x00
+#endif
 
 /*
  * The architecture vector has an array of PVR mask/value pairs,
@@ -687,10 +692,12 @@ static unsigned char ibm_architecture_vec[] = {
 	0,				/* don't halt */
 
 	/* option vector 5: PAPR/OF options */
-	3 - 2,				/* length */
+	5 - 2,				/* length */
 	0,				/* don't ignore, don't halt */
 	OV5_LPAR | OV5_SPLPAR | OV5_LARGE_PAGES | OV5_DRCONF_MEMORY |
 	OV5_DONATE_DEDICATE_CPU | OV5_MSI,
+	0,
+	OV5_CMO,
 };
 
 /* Old method - ELF header with PT_NOTE sections */

From 16c14b4621c7b6fc4611abf1f86cd78cdb1b2b03 Mon Sep 17 00:00:00 2001
From: Nathan Fontenot <nfont@austin.ibm.com>
Date: Thu, 24 Jul 2008 05:10:46 +1000
Subject: [PATCH 0648/1171] powerpc/pseries: Remove kmalloc call in handling
 writes to lparcfg

There are only 4 valid name=value pairs for writes to
/proc/ppc64/lparcfg.  Current code allocates a buffer to copy
this information in from the user.  Since the longest name=value
pair will easily fit into a buffer of 64 characters, simply
put the buffer on the stack instead of allocating the buffer.

Signed-off-by: Nathan Fotenot <nfont@austin.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/lparcfg.c | 28 ++++++++++++----------------
 1 file changed, 12 insertions(+), 16 deletions(-)

diff --git a/arch/powerpc/kernel/lparcfg.c b/arch/powerpc/kernel/lparcfg.c
index 64381a204a58..9f856a0c3e38 100644
--- a/arch/powerpc/kernel/lparcfg.c
+++ b/arch/powerpc/kernel/lparcfg.c
@@ -573,29 +573,27 @@ static ssize_t update_mpp(u64 *entitlement, u8 *weight)
 static ssize_t lparcfg_write(struct file *file, const char __user * buf,
 			     size_t count, loff_t * off)
 {
-	char *kbuf;
+	int kbuf_sz = 64;
+	char kbuf[kbuf_sz];
 	char *tmp;
 	u64 new_entitled, *new_entitled_ptr = &new_entitled;
 	u8 new_weight, *new_weight_ptr = &new_weight;
-	ssize_t retval = -ENOMEM;
+	ssize_t retval;
 
 	if (!firmware_has_feature(FW_FEATURE_SPLPAR) ||
 			firmware_has_feature(FW_FEATURE_ISERIES))
 		return -EINVAL;
 
-	kbuf = kmalloc(count, GFP_KERNEL);
-	if (!kbuf)
-		goto out;
+	if (count > kbuf_sz)
+		return -EINVAL;
 
-	retval = -EFAULT;
 	if (copy_from_user(kbuf, buf, count))
-		goto out;
+		return -EFAULT;
 
-	retval = -EINVAL;
 	kbuf[count - 1] = '\0';
 	tmp = strchr(kbuf, '=');
 	if (!tmp)
-		goto out;
+		return -EINVAL;
 
 	*tmp++ = '\0';
 
@@ -603,32 +601,32 @@ static ssize_t lparcfg_write(struct file *file, const char __user * buf,
 		char *endp;
 		*new_entitled_ptr = (u64) simple_strtoul(tmp, &endp, 10);
 		if (endp == tmp)
-			goto out;
+			return -EINVAL;
 
 		retval = update_ppp(new_entitled_ptr, NULL);
 	} else if (!strcmp(kbuf, "capacity_weight")) {
 		char *endp;
 		*new_weight_ptr = (u8) simple_strtoul(tmp, &endp, 10);
 		if (endp == tmp)
-			goto out;
+			return -EINVAL;
 
 		retval = update_ppp(NULL, new_weight_ptr);
 	} else if (!strcmp(kbuf, "entitled_memory")) {
 		char *endp;
 		*new_entitled_ptr = (u64) simple_strtoul(tmp, &endp, 10);
 		if (endp == tmp)
-			goto out;
+			return -EINVAL;
 
 		retval = update_mpp(new_entitled_ptr, NULL);
 	} else if (!strcmp(kbuf, "entitled_memory_weight")) {
 		char *endp;
 		*new_weight_ptr = (u8) simple_strtoul(tmp, &endp, 10);
 		if (endp == tmp)
-			goto out;
+			return -EINVAL;
 
 		retval = update_mpp(NULL, new_weight_ptr);
 	} else
-		goto out;
+		return -EINVAL;
 
 	if (retval == H_SUCCESS || retval == H_CONSTRAINED) {
 		retval = count;
@@ -644,8 +642,6 @@ static ssize_t lparcfg_write(struct file *file, const char __user * buf,
 		retval = -EIO;
 	}
 
-out:
-	kfree(kbuf);
 	return retval;
 }
 

From 1e3519f8e1baec0b733cd42684fcd3d9681662f1 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Fri, 25 Jul 2008 16:21:11 +1000
Subject: [PATCH 0649/1171] Move update_mmu_cache() declaration from tlbflush.h
 to pgtable.h where it belongs. This fixes some build problems on some configs

Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 include/asm-powerpc/pgtable.h  | 13 +++++++++++++
 include/asm-powerpc/tlbflush.h | 11 -----------
 2 files changed, 13 insertions(+), 11 deletions(-)

diff --git a/include/asm-powerpc/pgtable.h b/include/asm-powerpc/pgtable.h
index d18ffe7bc7c4..dbb8ca172e44 100644
--- a/include/asm-powerpc/pgtable.h
+++ b/include/asm-powerpc/pgtable.h
@@ -38,6 +38,19 @@ extern void paging_init(void);
 		remap_pfn_range(vma, vaddr, pfn, size, prot)
 
 #include <asm-generic/pgtable.h>
+
+
+/*
+ * This gets called at the end of handling a page fault, when
+ * the kernel has put a new PTE into the page table for the process.
+ * We use it to ensure coherency between the i-cache and d-cache
+ * for the page which has just been mapped in.
+ * On machines which use an MMU hash table, we use this to put a
+ * corresponding HPTE into the hash table ahead of time, instead of
+ * waiting for the inevitable extra hash-table miss exception.
+ */
+extern void update_mmu_cache(struct vm_area_struct *, unsigned long, pte_t);
+
 #endif /* __ASSEMBLY__ */
 
 #endif /* __KERNEL__ */
diff --git a/include/asm-powerpc/tlbflush.h b/include/asm-powerpc/tlbflush.h
index 5c9108147644..361cd5c7a32b 100644
--- a/include/asm-powerpc/tlbflush.h
+++ b/include/asm-powerpc/tlbflush.h
@@ -162,16 +162,5 @@ extern void __flush_hash_table_range(struct mm_struct *mm, unsigned long start,
 
 #endif
 
-/*
- * This gets called at the end of handling a page fault, when
- * the kernel has put a new PTE into the page table for the process.
- * We use it to ensure coherency between the i-cache and d-cache
- * for the page which has just been mapped in.
- * On machines which use an MMU hash table, we use this to put a
- * corresponding HPTE into the hash table ahead of time, instead of
- * waiting for the inevitable extra hash-table miss exception.
- */
-extern void update_mmu_cache(struct vm_area_struct *, unsigned long, pte_t);
-
 #endif /*__KERNEL__ */
 #endif /* _ASM_POWERPC_TLBFLUSH_H */

From 973b7d83ebeb1e34b8bee69208916e5f0e2353c3 Mon Sep 17 00:00:00 2001
From: Tony Breeds <tony@bakeyournoodle.com>
Date: Fri, 25 Jul 2008 16:21:51 +1000
Subject: [PATCH 0650/1171] powerpc: Wireup new syscalls

signalfd4, eventfd2, epoll_create1, dup3, pipe2 and inotify_init1

Signed-off-by: Tony Breeds <tony@bakeyournoodle.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 include/asm-powerpc/syscalls.h | 1 +
 include/asm-powerpc/systbl.h   | 6 ++++++
 include/asm-powerpc/unistd.h   | 8 +++++++-
 3 files changed, 14 insertions(+), 1 deletion(-)

diff --git a/include/asm-powerpc/syscalls.h b/include/asm-powerpc/syscalls.h
index 2b8a458f990a..eb8eb400c664 100644
--- a/include/asm-powerpc/syscalls.h
+++ b/include/asm-powerpc/syscalls.h
@@ -31,6 +31,7 @@ asmlinkage int sys_vfork(unsigned long p1, unsigned long p2,
 		unsigned long p3, unsigned long p4, unsigned long p5,
 		unsigned long p6, struct pt_regs *regs);
 asmlinkage long sys_pipe(int __user *fildes);
+asmlinkage long sys_pipe2(int __user *fildes, int flags);
 asmlinkage long sys_rt_sigaction(int sig,
 		const struct sigaction __user *act,
 		struct sigaction __user *oact, size_t sigsetsize);
diff --git a/include/asm-powerpc/systbl.h b/include/asm-powerpc/systbl.h
index ae7085c65692..e084272ed1c2 100644
--- a/include/asm-powerpc/systbl.h
+++ b/include/asm-powerpc/systbl.h
@@ -316,3 +316,9 @@ COMPAT_SYS(fallocate)
 SYSCALL(subpage_prot)
 COMPAT_SYS_SPU(timerfd_settime)
 COMPAT_SYS_SPU(timerfd_gettime)
+COMPAT_SYS_SPU(signalfd4)
+SYSCALL_SPU(eventfd2)
+SYSCALL_SPU(epoll_create1)
+SYSCALL_SPU(dup3)
+SYSCALL_SPU(pipe2)
+SYSCALL(inotify_init1)
diff --git a/include/asm-powerpc/unistd.h b/include/asm-powerpc/unistd.h
index ce91bb662063..e07d0c76ed77 100644
--- a/include/asm-powerpc/unistd.h
+++ b/include/asm-powerpc/unistd.h
@@ -335,10 +335,16 @@
 #define __NR_subpage_prot	310
 #define __NR_timerfd_settime	311
 #define __NR_timerfd_gettime	312
+#define __NR_signalfd4		313
+#define __NR_eventfd2		314
+#define __NR_epoll_create1	315
+#define __NR_dup3		316
+#define __NR_pipe2		317
+#define __NR_inotify_init1	318
 
 #ifdef __KERNEL__
 
-#define __NR_syscalls		313
+#define __NR_syscalls		319
 
 #define __NR__exit __NR_exit
 #define NR_syscalls	__NR_syscalls

From cffe1c5d7a5a1e54f7c2c6d0510f651a965bccc3 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Fri, 25 Jul 2008 01:25:04 -0700
Subject: [PATCH 0651/1171] pkt_sched: Fix locking in
 shutdown_scheduler_queue()

Qdisc locks need to be held with BH disabled.

Tested-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sched/sch_generic.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 4ac7e3a8c253..43abd4d27ea6 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -736,9 +736,9 @@ static void shutdown_scheduler_queue(struct net_device *dev,
 		dev_queue->qdisc = qdisc_default;
 		dev_queue->qdisc_sleeping = qdisc_default;
 
-		spin_lock(root_lock);
+		spin_lock_bh(root_lock);
 		qdisc_destroy(qdisc);
-		spin_unlock(root_lock);
+		spin_unlock_bh(root_lock);
 	}
 }
 

From d37e6bf68fc1eb34a4ad21d9ae8890ed37ea80e7 Mon Sep 17 00:00:00 2001
From: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Date: Thu, 24 Jul 2008 18:28:11 +0300
Subject: [PATCH 0652/1171] UBI: always start the background thread

This fix only affects UBI debugging.

If the the background thread is disabled for debugging purposes,
start it anyway, because otherwise we see tonns of kernel debugging
complaints like this:

INFO: task ubi_bgt0d:26857 blocked for more than 120 seconds.
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
ubi_bgt0d     D dd37bf94     0 26857      2
       dd37bfcc 00000086 f8e17cea dd37bf94 00000046 00000000 00000000 f5c62430
       f5c62430 f5c62590 c2a09c80 f6cbd498 dd8e9cbc 00000296 dd37bfb0 00000296
       dd8e9cb8 dd8e9cbc dd37bfcc c0119774 00000000 00000000 c0132e89 f6961560
Call Trace:
 [<f8e17cea>] ? ubi_thread+0x0/0x127 [ubi]
 [<c0119774>] ? complete+0x43/0x4b
 [<c0132e89>] ? kthread+0x0/0x5b
 [<f8e17cea>] ? ubi_thread+0x0/0x127 [ubi]
 [<c0132eae>] kthread+0x25/0x5b
 [<c0132e89>] ? kthread+0x0/0x5b
 [<c0104953>] kernel_thread_helper+0x7/0x14
 =======================

So start it, and go sleep inside it, instead of creating it and never
start.

Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
---
 drivers/mtd/ubi/build.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c
index eba760b3b8c7..c7630a228310 100644
--- a/drivers/mtd/ubi/build.c
+++ b/drivers/mtd/ubi/build.c
@@ -870,11 +870,9 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset)
 		ubi->beb_rsvd_pebs);
 	ubi_msg("max/mean erase counter: %d/%d", ubi->max_ec, ubi->mean_ec);
 
-	/* Enable the background thread */
-	if (!DBG_DISABLE_BGT) {
+	if (!DBG_DISABLE_BGT)
 		ubi->thread_enabled = 1;
-		wake_up_process(ubi->bgt_thread);
-	}
+	wake_up_process(ubi->bgt_thread);
 
 	ubi_devices[ubi_num] = ubi;
 	return ubi_num;

From 6fccab671f2f0a24b799f29a4ec878f62d34656c Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Fri, 25 Jul 2008 02:54:40 -0700
Subject: [PATCH 0653/1171] ipsec: ipcomp - Merge IPComp implementations

This patch merges the IPv4/IPv6 IPComp implementations since most
of the code is identical.  As a result future enhancements will no
longer need to be duplicated.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/ipcomp.h   |   6 +
 net/ipv4/Kconfig       |   4 +-
 net/ipv4/ipcomp.c      | 315 +------------------------------------
 net/ipv6/Kconfig       |   4 +-
 net/ipv6/ipcomp6.c     | 298 +----------------------------------
 net/xfrm/Kconfig       |   6 +
 net/xfrm/Makefile      |   1 +
 net/xfrm/xfrm_ipcomp.c | 349 +++++++++++++++++++++++++++++++++++++++++
 8 files changed, 377 insertions(+), 606 deletions(-)
 create mode 100644 net/xfrm/xfrm_ipcomp.c

diff --git a/include/net/ipcomp.h b/include/net/ipcomp.h
index 330b74e813a9..2a1092abaa07 100644
--- a/include/net/ipcomp.h
+++ b/include/net/ipcomp.h
@@ -14,6 +14,12 @@ struct ipcomp_data {
 
 struct ip_comp_hdr;
 struct sk_buff;
+struct xfrm_state;
+
+int ipcomp_input(struct xfrm_state *x, struct sk_buff *skb);
+int ipcomp_output(struct xfrm_state *x, struct sk_buff *skb);
+void ipcomp_destroy(struct xfrm_state *x);
+int ipcomp_init_state(struct xfrm_state *x);
 
 static inline struct ip_comp_hdr *ip_comp_hdr(const struct sk_buff *skb)
 {
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index 4670683b4688..591ea23639ca 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -356,10 +356,8 @@ config INET_ESP
 
 config INET_IPCOMP
 	tristate "IP: IPComp transformation"
-	select XFRM
 	select INET_XFRM_TUNNEL
-	select CRYPTO
-	select CRYPTO_DEFLATE
+	select XFRM_IPCOMP
 	---help---
 	  Support for IP Payload Compression Protocol (IPComp) (RFC3173),
 	  typically needed for IPsec.
diff --git a/net/ipv4/ipcomp.c b/net/ipv4/ipcomp.c
index a75807b971b3..a42b64d040c4 100644
--- a/net/ipv4/ipcomp.c
+++ b/net/ipv4/ipcomp.c
@@ -14,153 +14,14 @@
  *   - Adaptive compression.
  */
 #include <linux/module.h>
-#include <linux/crypto.h>
 #include <linux/err.h>
-#include <linux/pfkeyv2.h>
-#include <linux/percpu.h>
-#include <linux/smp.h>
-#include <linux/list.h>
-#include <linux/vmalloc.h>
 #include <linux/rtnetlink.h>
-#include <linux/mutex.h>
 #include <net/ip.h>
 #include <net/xfrm.h>
 #include <net/icmp.h>
 #include <net/ipcomp.h>
 #include <net/protocol.h>
-
-struct ipcomp_tfms {
-	struct list_head list;
-	struct crypto_comp **tfms;
-	int users;
-};
-
-static DEFINE_MUTEX(ipcomp_resource_mutex);
-static void **ipcomp_scratches;
-static int ipcomp_scratch_users;
-static LIST_HEAD(ipcomp_tfms_list);
-
-static int ipcomp_decompress(struct xfrm_state *x, struct sk_buff *skb)
-{
-	struct ipcomp_data *ipcd = x->data;
-	const int plen = skb->len;
-	int dlen = IPCOMP_SCRATCH_SIZE;
-	const u8 *start = skb->data;
-	const int cpu = get_cpu();
-	u8 *scratch = *per_cpu_ptr(ipcomp_scratches, cpu);
-	struct crypto_comp *tfm = *per_cpu_ptr(ipcd->tfms, cpu);
-	int err = crypto_comp_decompress(tfm, start, plen, scratch, &dlen);
-
-	if (err)
-		goto out;
-
-	if (dlen < (plen + sizeof(struct ip_comp_hdr))) {
-		err = -EINVAL;
-		goto out;
-	}
-
-	err = pskb_expand_head(skb, 0, dlen - plen, GFP_ATOMIC);
-	if (err)
-		goto out;
-
-	skb->truesize += dlen - plen;
-	__skb_put(skb, dlen - plen);
-	skb_copy_to_linear_data(skb, scratch, dlen);
-out:
-	put_cpu();
-	return err;
-}
-
-static int ipcomp_input(struct xfrm_state *x, struct sk_buff *skb)
-{
-	int nexthdr;
-	int err = -ENOMEM;
-	struct ip_comp_hdr *ipch;
-
-	if (skb_linearize_cow(skb))
-		goto out;
-
-	skb->ip_summed = CHECKSUM_NONE;
-
-	/* Remove ipcomp header and decompress original payload */
-	ipch = (void *)skb->data;
-	nexthdr = ipch->nexthdr;
-
-	skb->transport_header = skb->network_header + sizeof(*ipch);
-	__skb_pull(skb, sizeof(*ipch));
-	err = ipcomp_decompress(x, skb);
-	if (err)
-		goto out;
-
-	err = nexthdr;
-
-out:
-	return err;
-}
-
-static int ipcomp_compress(struct xfrm_state *x, struct sk_buff *skb)
-{
-	struct ipcomp_data *ipcd = x->data;
-	const int plen = skb->len;
-	int dlen = IPCOMP_SCRATCH_SIZE;
-	u8 *start = skb->data;
-	const int cpu = get_cpu();
-	u8 *scratch = *per_cpu_ptr(ipcomp_scratches, cpu);
-	struct crypto_comp *tfm = *per_cpu_ptr(ipcd->tfms, cpu);
-	int err;
-
-	local_bh_disable();
-	err = crypto_comp_compress(tfm, start, plen, scratch, &dlen);
-	local_bh_enable();
-	if (err)
-		goto out;
-
-	if ((dlen + sizeof(struct ip_comp_hdr)) >= plen) {
-		err = -EMSGSIZE;
-		goto out;
-	}
-
-	memcpy(start + sizeof(struct ip_comp_hdr), scratch, dlen);
-	put_cpu();
-
-	pskb_trim(skb, dlen + sizeof(struct ip_comp_hdr));
-	return 0;
-
-out:
-	put_cpu();
-	return err;
-}
-
-static int ipcomp_output(struct xfrm_state *x, struct sk_buff *skb)
-{
-	int err;
-	struct ip_comp_hdr *ipch;
-	struct ipcomp_data *ipcd = x->data;
-
-	if (skb->len < ipcd->threshold) {
-		/* Don't bother compressing */
-		goto out_ok;
-	}
-
-	if (skb_linearize_cow(skb))
-		goto out_ok;
-
-	err = ipcomp_compress(x, skb);
-
-	if (err) {
-		goto out_ok;
-	}
-
-	/* Install ipcomp header, convert into ipcomp datagram. */
-	ipch = ip_comp_hdr(skb);
-	ipch->nexthdr = *skb_mac_header(skb);
-	ipch->flags = 0;
-	ipch->cpi = htons((u16 )ntohl(x->id.spi));
-	*skb_mac_header(skb) = IPPROTO_COMP;
-out_ok:
-	skb_push(skb, -skb_network_offset(skb));
-	return 0;
-}
+#include <net/sock.h>
 
 static void ipcomp4_err(struct sk_buff *skb, u32 info)
 {
@@ -241,156 +102,12 @@ out:
 	return err;
 }
 
-static void ipcomp_free_scratches(void)
-{
-	int i;
-	void **scratches;
-
-	if (--ipcomp_scratch_users)
-		return;
-
-	scratches = ipcomp_scratches;
-	if (!scratches)
-		return;
-
-	for_each_possible_cpu(i)
-		vfree(*per_cpu_ptr(scratches, i));
-
-	free_percpu(scratches);
-}
-
-static void **ipcomp_alloc_scratches(void)
-{
-	int i;
-	void **scratches;
-
-	if (ipcomp_scratch_users++)
-		return ipcomp_scratches;
-
-	scratches = alloc_percpu(void *);
-	if (!scratches)
-		return NULL;
-
-	ipcomp_scratches = scratches;
-
-	for_each_possible_cpu(i) {
-		void *scratch = vmalloc(IPCOMP_SCRATCH_SIZE);
-		if (!scratch)
-			return NULL;
-		*per_cpu_ptr(scratches, i) = scratch;
-	}
-
-	return scratches;
-}
-
-static void ipcomp_free_tfms(struct crypto_comp **tfms)
-{
-	struct ipcomp_tfms *pos;
-	int cpu;
-
-	list_for_each_entry(pos, &ipcomp_tfms_list, list) {
-		if (pos->tfms == tfms)
-			break;
-	}
-
-	BUG_TRAP(pos);
-
-	if (--pos->users)
-		return;
-
-	list_del(&pos->list);
-	kfree(pos);
-
-	if (!tfms)
-		return;
-
-	for_each_possible_cpu(cpu) {
-		struct crypto_comp *tfm = *per_cpu_ptr(tfms, cpu);
-		crypto_free_comp(tfm);
-	}
-	free_percpu(tfms);
-}
-
-static struct crypto_comp **ipcomp_alloc_tfms(const char *alg_name)
-{
-	struct ipcomp_tfms *pos;
-	struct crypto_comp **tfms;
-	int cpu;
-
-	/* This can be any valid CPU ID so we don't need locking. */
-	cpu = raw_smp_processor_id();
-
-	list_for_each_entry(pos, &ipcomp_tfms_list, list) {
-		struct crypto_comp *tfm;
-
-		tfms = pos->tfms;
-		tfm = *per_cpu_ptr(tfms, cpu);
-
-		if (!strcmp(crypto_comp_name(tfm), alg_name)) {
-			pos->users++;
-			return tfms;
-		}
-	}
-
-	pos = kmalloc(sizeof(*pos), GFP_KERNEL);
-	if (!pos)
-		return NULL;
-
-	pos->users = 1;
-	INIT_LIST_HEAD(&pos->list);
-	list_add(&pos->list, &ipcomp_tfms_list);
-
-	pos->tfms = tfms = alloc_percpu(struct crypto_comp *);
-	if (!tfms)
-		goto error;
-
-	for_each_possible_cpu(cpu) {
-		struct crypto_comp *tfm = crypto_alloc_comp(alg_name, 0,
-							    CRYPTO_ALG_ASYNC);
-		if (IS_ERR(tfm))
-			goto error;
-		*per_cpu_ptr(tfms, cpu) = tfm;
-	}
-
-	return tfms;
-
-error:
-	ipcomp_free_tfms(tfms);
-	return NULL;
-}
-
-static void ipcomp_free_data(struct ipcomp_data *ipcd)
-{
-	if (ipcd->tfms)
-		ipcomp_free_tfms(ipcd->tfms);
-	ipcomp_free_scratches();
-}
-
-static void ipcomp_destroy(struct xfrm_state *x)
-{
-	struct ipcomp_data *ipcd = x->data;
-	if (!ipcd)
-		return;
-	xfrm_state_delete_tunnel(x);
-	mutex_lock(&ipcomp_resource_mutex);
-	ipcomp_free_data(ipcd);
-	mutex_unlock(&ipcomp_resource_mutex);
-	kfree(ipcd);
-}
-
-static int ipcomp_init_state(struct xfrm_state *x)
+static int ipcomp4_init_state(struct xfrm_state *x)
 {
 	int err;
 	struct ipcomp_data *ipcd;
 	struct xfrm_algo_desc *calg_desc;
 
-	err = -EINVAL;
-	if (!x->calg)
-		goto out;
-
-	if (x->encap)
-		goto out;
-
 	x->props.header_len = 0;
 	switch (x->props.mode) {
 	case XFRM_MODE_TRANSPORT:
@@ -402,40 +119,22 @@ static int ipcomp_init_state(struct xfrm_state *x)
 		goto out;
 	}
 
-	err = -ENOMEM;
-	ipcd = kzalloc(sizeof(*ipcd), GFP_KERNEL);
-	if (!ipcd)
+	err = ipcomp_init_state(x);
+	if (err)
 		goto out;
 
-	mutex_lock(&ipcomp_resource_mutex);
-	if (!ipcomp_alloc_scratches())
-		goto error;
-
-	ipcd->tfms = ipcomp_alloc_tfms(x->calg->alg_name);
-	if (!ipcd->tfms)
-		goto error;
-	mutex_unlock(&ipcomp_resource_mutex);
-
 	if (x->props.mode == XFRM_MODE_TUNNEL) {
 		err = ipcomp_tunnel_attach(x);
 		if (err)
 			goto error_tunnel;
 	}
 
-	calg_desc = xfrm_calg_get_byname(x->calg->alg_name, 0);
-	BUG_ON(!calg_desc);
-	ipcd->threshold = calg_desc->uinfo.comp.threshold;
-	x->data = ipcd;
 	err = 0;
 out:
 	return err;
 
 error_tunnel:
-	mutex_lock(&ipcomp_resource_mutex);
-error:
-	ipcomp_free_data(ipcd);
-	mutex_unlock(&ipcomp_resource_mutex);
-	kfree(ipcd);
+	ipcomp_destroy(x);
 	goto out;
 }
 
@@ -443,7 +142,7 @@ static const struct xfrm_type ipcomp_type = {
 	.description	= "IPCOMP4",
 	.owner		= THIS_MODULE,
 	.proto	     	= IPPROTO_COMP,
-	.init_state	= ipcomp_init_state,
+	.init_state	= ipcomp4_init_state,
 	.destructor	= ipcomp_destroy,
 	.input		= ipcomp_input,
 	.output		= ipcomp_output
@@ -481,7 +180,7 @@ module_init(ipcomp4_init);
 module_exit(ipcomp4_fini);
 
 MODULE_LICENSE("GPL");
-MODULE_DESCRIPTION("IP Payload Compression Protocol (IPComp) - RFC3173");
+MODULE_DESCRIPTION("IP Payload Compression Protocol (IPComp/IPv4) - RFC3173");
 MODULE_AUTHOR("James Morris <jmorris@intercode.com.au>");
 
 MODULE_ALIAS_XFRM_TYPE(AF_INET, XFRM_PROTO_COMP);
diff --git a/net/ipv6/Kconfig b/net/ipv6/Kconfig
index 42814a2ec9d7..ec992159b5f8 100644
--- a/net/ipv6/Kconfig
+++ b/net/ipv6/Kconfig
@@ -96,10 +96,8 @@ config INET6_ESP
 
 config INET6_IPCOMP
 	tristate "IPv6: IPComp transformation"
-	select XFRM
 	select INET6_XFRM_TUNNEL
-	select CRYPTO
-	select CRYPTO_DEFLATE
+	select XFRM_IPCOMP
 	---help---
 	  Support for IP Payload Compression Protocol (IPComp) (RFC3173),
 	  typically needed for IPsec.
diff --git a/net/ipv6/ipcomp6.c b/net/ipv6/ipcomp6.c
index ee6de425ce6b..0cfcea42153a 100644
--- a/net/ipv6/ipcomp6.c
+++ b/net/ipv6/ipcomp6.c
@@ -50,125 +50,6 @@
 #include <linux/icmpv6.h>
 #include <linux/mutex.h>
 
-struct ipcomp6_tfms {
-	struct list_head list;
-	struct crypto_comp **tfms;
-	int users;
-};
-
-static DEFINE_MUTEX(ipcomp6_resource_mutex);
-static void **ipcomp6_scratches;
-static int ipcomp6_scratch_users;
-static LIST_HEAD(ipcomp6_tfms_list);
-
-static int ipcomp6_input(struct xfrm_state *x, struct sk_buff *skb)
-{
-	int nexthdr;
-	int err = -ENOMEM;
-	struct ip_comp_hdr *ipch;
-	int plen, dlen;
-	struct ipcomp_data *ipcd = x->data;
-	u8 *start, *scratch;
-	struct crypto_comp *tfm;
-	int cpu;
-
-	if (skb_linearize_cow(skb))
-		goto out;
-
-	skb->ip_summed = CHECKSUM_NONE;
-
-	/* Remove ipcomp header and decompress original payload */
-	ipch = (void *)skb->data;
-	nexthdr = ipch->nexthdr;
-
-	skb->transport_header = skb->network_header + sizeof(*ipch);
-	__skb_pull(skb, sizeof(*ipch));
-
-	/* decompression */
-	plen = skb->len;
-	dlen = IPCOMP_SCRATCH_SIZE;
-	start = skb->data;
-
-	cpu = get_cpu();
-	scratch = *per_cpu_ptr(ipcomp6_scratches, cpu);
-	tfm = *per_cpu_ptr(ipcd->tfms, cpu);
-
-	err = crypto_comp_decompress(tfm, start, plen, scratch, &dlen);
-	if (err)
-		goto out_put_cpu;
-
-	if (dlen < (plen + sizeof(*ipch))) {
-		err = -EINVAL;
-		goto out_put_cpu;
-	}
-
-	err = pskb_expand_head(skb, 0, dlen - plen, GFP_ATOMIC);
-	if (err) {
-		goto out_put_cpu;
-	}
-
-	skb->truesize += dlen - plen;
-	__skb_put(skb, dlen - plen);
-	skb_copy_to_linear_data(skb, scratch, dlen);
-	err = nexthdr;
-
-out_put_cpu:
-	put_cpu();
-out:
-	return err;
-}
-
-static int ipcomp6_output(struct xfrm_state *x, struct sk_buff *skb)
-{
-	int err;
-	struct ip_comp_hdr *ipch;
-	struct ipcomp_data *ipcd = x->data;
-	int plen, dlen;
-	u8 *start, *scratch;
-	struct crypto_comp *tfm;
-	int cpu;
-
-	/* check whether datagram len is larger than threshold */
-	if (skb->len < ipcd->threshold) {
-		goto out_ok;
-	}
-
-	if (skb_linearize_cow(skb))
-		goto out_ok;
-
-	/* compression */
-	plen = skb->len;
-	dlen = IPCOMP_SCRATCH_SIZE;
-	start = skb->data;
-
-	cpu = get_cpu();
-	scratch = *per_cpu_ptr(ipcomp6_scratches, cpu);
-	tfm = *per_cpu_ptr(ipcd->tfms, cpu);
-
-	local_bh_disable();
-	err = crypto_comp_compress(tfm, start, plen, scratch, &dlen);
-	local_bh_enable();
-	if (err || (dlen + sizeof(*ipch)) >= plen) {
-		put_cpu();
-		goto out_ok;
-	}
-	memcpy(start + sizeof(struct ip_comp_hdr), scratch, dlen);
-	put_cpu();
-	pskb_trim(skb, dlen + sizeof(struct ip_comp_hdr));
-
-	/* insert ipcomp header and replace datagram */
-	ipch = ip_comp_hdr(skb);
-	ipch->nexthdr = *skb_mac_header(skb);
-	ipch->flags = 0;
-	ipch->cpi = htons((u16 )ntohl(x->id.spi));
-	*skb_mac_header(skb) = IPPROTO_COMP;
-
-out_ok:
-	skb_push(skb, -skb_network_offset(skb));
-
-	return 0;
-}
-
 static void ipcomp6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 				int type, int code, int offset, __be32 info)
 {
@@ -251,161 +132,12 @@ out:
 	return err;
 }
 
-static void ipcomp6_free_scratches(void)
-{
-	int i;
-	void **scratches;
-
-	if (--ipcomp6_scratch_users)
-		return;
-
-	scratches = ipcomp6_scratches;
-	if (!scratches)
-		return;
-
-	for_each_possible_cpu(i) {
-		void *scratch = *per_cpu_ptr(scratches, i);
-
-		vfree(scratch);
-	}
-
-	free_percpu(scratches);
-}
-
-static void **ipcomp6_alloc_scratches(void)
-{
-	int i;
-	void **scratches;
-
-	if (ipcomp6_scratch_users++)
-		return ipcomp6_scratches;
-
-	scratches = alloc_percpu(void *);
-	if (!scratches)
-		return NULL;
-
-	ipcomp6_scratches = scratches;
-
-	for_each_possible_cpu(i) {
-		void *scratch = vmalloc(IPCOMP_SCRATCH_SIZE);
-		if (!scratch)
-			return NULL;
-		*per_cpu_ptr(scratches, i) = scratch;
-	}
-
-	return scratches;
-}
-
-static void ipcomp6_free_tfms(struct crypto_comp **tfms)
-{
-	struct ipcomp6_tfms *pos;
-	int cpu;
-
-	list_for_each_entry(pos, &ipcomp6_tfms_list, list) {
-		if (pos->tfms == tfms)
-			break;
-	}
-
-	BUG_TRAP(pos);
-
-	if (--pos->users)
-		return;
-
-	list_del(&pos->list);
-	kfree(pos);
-
-	if (!tfms)
-		return;
-
-	for_each_possible_cpu(cpu) {
-		struct crypto_comp *tfm = *per_cpu_ptr(tfms, cpu);
-		crypto_free_comp(tfm);
-	}
-	free_percpu(tfms);
-}
-
-static struct crypto_comp **ipcomp6_alloc_tfms(const char *alg_name)
-{
-	struct ipcomp6_tfms *pos;
-	struct crypto_comp **tfms;
-	int cpu;
-
-	/* This can be any valid CPU ID so we don't need locking. */
-	cpu = raw_smp_processor_id();
-
-	list_for_each_entry(pos, &ipcomp6_tfms_list, list) {
-		struct crypto_comp *tfm;
-
-		tfms = pos->tfms;
-		tfm = *per_cpu_ptr(tfms, cpu);
-
-		if (!strcmp(crypto_comp_name(tfm), alg_name)) {
-			pos->users++;
-			return tfms;
-		}
-	}
-
-	pos = kmalloc(sizeof(*pos), GFP_KERNEL);
-	if (!pos)
-		return NULL;
-
-	pos->users = 1;
-	INIT_LIST_HEAD(&pos->list);
-	list_add(&pos->list, &ipcomp6_tfms_list);
-
-	pos->tfms = tfms = alloc_percpu(struct crypto_comp *);
-	if (!tfms)
-		goto error;
-
-	for_each_possible_cpu(cpu) {
-		struct crypto_comp *tfm = crypto_alloc_comp(alg_name, 0,
-							    CRYPTO_ALG_ASYNC);
-		if (IS_ERR(tfm))
-			goto error;
-		*per_cpu_ptr(tfms, cpu) = tfm;
-	}
-
-	return tfms;
-
-error:
-	ipcomp6_free_tfms(tfms);
-	return NULL;
-}
-
-static void ipcomp6_free_data(struct ipcomp_data *ipcd)
-{
-	if (ipcd->tfms)
-		ipcomp6_free_tfms(ipcd->tfms);
-	ipcomp6_free_scratches();
-}
-
-static void ipcomp6_destroy(struct xfrm_state *x)
-{
-	struct ipcomp_data *ipcd = x->data;
-	if (!ipcd)
-		return;
-	xfrm_state_delete_tunnel(x);
-	mutex_lock(&ipcomp6_resource_mutex);
-	ipcomp6_free_data(ipcd);
-	mutex_unlock(&ipcomp6_resource_mutex);
-	kfree(ipcd);
-
-	xfrm6_tunnel_free_spi((xfrm_address_t *)&x->props.saddr);
-}
-
 static int ipcomp6_init_state(struct xfrm_state *x)
 {
 	int err;
 	struct ipcomp_data *ipcd;
 	struct xfrm_algo_desc *calg_desc;
 
-	err = -EINVAL;
-	if (!x->calg)
-		goto out;
-
-	if (x->encap)
-		goto out;
-
 	x->props.header_len = 0;
 	switch (x->props.mode) {
 	case XFRM_MODE_TRANSPORT:
@@ -417,39 +149,21 @@ static int ipcomp6_init_state(struct xfrm_state *x)
 		goto out;
 	}
 
-	err = -ENOMEM;
-	ipcd = kzalloc(sizeof(*ipcd), GFP_KERNEL);
-	if (!ipcd)
+	err = ipcomp_init_state(x);
+	if (err)
 		goto out;
 
-	mutex_lock(&ipcomp6_resource_mutex);
-	if (!ipcomp6_alloc_scratches())
-		goto error;
-
-	ipcd->tfms = ipcomp6_alloc_tfms(x->calg->alg_name);
-	if (!ipcd->tfms)
-		goto error;
-	mutex_unlock(&ipcomp6_resource_mutex);
-
 	if (x->props.mode == XFRM_MODE_TUNNEL) {
 		err = ipcomp6_tunnel_attach(x);
 		if (err)
 			goto error_tunnel;
 	}
 
-	calg_desc = xfrm_calg_get_byname(x->calg->alg_name, 0);
-	BUG_ON(!calg_desc);
-	ipcd->threshold = calg_desc->uinfo.comp.threshold;
-	x->data = ipcd;
 	err = 0;
 out:
 	return err;
 error_tunnel:
-	mutex_lock(&ipcomp6_resource_mutex);
-error:
-	ipcomp6_free_data(ipcd);
-	mutex_unlock(&ipcomp6_resource_mutex);
-	kfree(ipcd);
+	ipcomp_destroy(x);
 
 	goto out;
 }
@@ -460,9 +174,9 @@ static const struct xfrm_type ipcomp6_type =
 	.owner		= THIS_MODULE,
 	.proto		= IPPROTO_COMP,
 	.init_state	= ipcomp6_init_state,
-	.destructor	= ipcomp6_destroy,
-	.input		= ipcomp6_input,
-	.output		= ipcomp6_output,
+	.destructor	= ipcomp_destroy,
+	.input		= ipcomp_input,
+	.output		= ipcomp_output,
 	.hdr_offset	= xfrm6_find_1stfragopt,
 };
 
diff --git a/net/xfrm/Kconfig b/net/xfrm/Kconfig
index 9201ef8ad90e..6d081674515f 100644
--- a/net/xfrm/Kconfig
+++ b/net/xfrm/Kconfig
@@ -46,6 +46,12 @@ config XFRM_STATISTICS
 
 	  If unsure, say N.
 
+config XFRM_IPCOMP
+	tristate
+	select XFRM
+	select CRYPTO
+	select CRYPTO_DEFLATE
+
 config NET_KEY
 	tristate "PF_KEY sockets"
 	select XFRM
diff --git a/net/xfrm/Makefile b/net/xfrm/Makefile
index 332cfb0ff566..0f439a72ccab 100644
--- a/net/xfrm/Makefile
+++ b/net/xfrm/Makefile
@@ -6,4 +6,5 @@ obj-$(CONFIG_XFRM) := xfrm_policy.o xfrm_state.o xfrm_hash.o \
 		      xfrm_input.o xfrm_output.o xfrm_algo.o
 obj-$(CONFIG_XFRM_STATISTICS) += xfrm_proc.o
 obj-$(CONFIG_XFRM_USER) += xfrm_user.o
+obj-$(CONFIG_XFRM_IPCOMP) += xfrm_ipcomp.o
 
diff --git a/net/xfrm/xfrm_ipcomp.c b/net/xfrm/xfrm_ipcomp.c
new file mode 100644
index 000000000000..b51e804fbbad
--- /dev/null
+++ b/net/xfrm/xfrm_ipcomp.c
@@ -0,0 +1,349 @@
+/*
+ * IP Payload Compression Protocol (IPComp) - RFC3173.
+ *
+ * Copyright (c) 2003 James Morris <jmorris@intercode.com.au>
+ * Copyright (c) 2003-2008 Herbert Xu <herbert@gondor.apana.org.au>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.
+ *
+ * Todo:
+ *   - Tunable compression parameters.
+ *   - Compression stats.
+ *   - Adaptive compression.
+ */
+
+#include <linux/crypto.h>
+#include <linux/err.h>
+#include <linux/list.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/percpu.h>
+#include <linux/rtnetlink.h>
+#include <linux/smp.h>
+#include <linux/vmalloc.h>
+#include <net/ip.h>
+#include <net/ipcomp.h>
+#include <net/xfrm.h>
+
+struct ipcomp_tfms {
+	struct list_head list;
+	struct crypto_comp **tfms;
+	int users;
+};
+
+static DEFINE_MUTEX(ipcomp_resource_mutex);
+static void **ipcomp_scratches;
+static int ipcomp_scratch_users;
+static LIST_HEAD(ipcomp_tfms_list);
+
+static int ipcomp_decompress(struct xfrm_state *x, struct sk_buff *skb)
+{
+	struct ipcomp_data *ipcd = x->data;
+	const int plen = skb->len;
+	int dlen = IPCOMP_SCRATCH_SIZE;
+	const u8 *start = skb->data;
+	const int cpu = get_cpu();
+	u8 *scratch = *per_cpu_ptr(ipcomp_scratches, cpu);
+	struct crypto_comp *tfm = *per_cpu_ptr(ipcd->tfms, cpu);
+	int err = crypto_comp_decompress(tfm, start, plen, scratch, &dlen);
+
+	if (err)
+		goto out;
+
+	if (dlen < (plen + sizeof(struct ip_comp_hdr))) {
+		err = -EINVAL;
+		goto out;
+	}
+
+	err = pskb_expand_head(skb, 0, dlen - plen, GFP_ATOMIC);
+	if (err)
+		goto out;
+
+	skb->truesize += dlen - plen;
+	__skb_put(skb, dlen - plen);
+	skb_copy_to_linear_data(skb, scratch, dlen);
+out:
+	put_cpu();
+	return err;
+}
+
+int ipcomp_input(struct xfrm_state *x, struct sk_buff *skb)
+{
+	int nexthdr;
+	int err = -ENOMEM;
+	struct ip_comp_hdr *ipch;
+
+	if (skb_linearize_cow(skb))
+		goto out;
+
+	skb->ip_summed = CHECKSUM_NONE;
+
+	/* Remove ipcomp header and decompress original payload */
+	ipch = (void *)skb->data;
+	nexthdr = ipch->nexthdr;
+
+	skb->transport_header = skb->network_header + sizeof(*ipch);
+	__skb_pull(skb, sizeof(*ipch));
+	err = ipcomp_decompress(x, skb);
+	if (err)
+		goto out;
+
+	err = nexthdr;
+
+out:
+	return err;
+}
+EXPORT_SYMBOL_GPL(ipcomp_input);
+
+static int ipcomp_compress(struct xfrm_state *x, struct sk_buff *skb)
+{
+	struct ipcomp_data *ipcd = x->data;
+	const int plen = skb->len;
+	int dlen = IPCOMP_SCRATCH_SIZE;
+	u8 *start = skb->data;
+	const int cpu = get_cpu();
+	u8 *scratch = *per_cpu_ptr(ipcomp_scratches, cpu);
+	struct crypto_comp *tfm = *per_cpu_ptr(ipcd->tfms, cpu);
+	int err;
+
+	local_bh_disable();
+	err = crypto_comp_compress(tfm, start, plen, scratch, &dlen);
+	local_bh_enable();
+	if (err)
+		goto out;
+
+	if ((dlen + sizeof(struct ip_comp_hdr)) >= plen) {
+		err = -EMSGSIZE;
+		goto out;
+	}
+
+	memcpy(start + sizeof(struct ip_comp_hdr), scratch, dlen);
+	put_cpu();
+
+	pskb_trim(skb, dlen + sizeof(struct ip_comp_hdr));
+	return 0;
+
+out:
+	put_cpu();
+	return err;
+}
+
+int ipcomp_output(struct xfrm_state *x, struct sk_buff *skb)
+{
+	int err;
+	struct ip_comp_hdr *ipch;
+	struct ipcomp_data *ipcd = x->data;
+
+	if (skb->len < ipcd->threshold) {
+		/* Don't bother compressing */
+		goto out_ok;
+	}
+
+	if (skb_linearize_cow(skb))
+		goto out_ok;
+
+	err = ipcomp_compress(x, skb);
+
+	if (err) {
+		goto out_ok;
+	}
+
+	/* Install ipcomp header, convert into ipcomp datagram. */
+	ipch = ip_comp_hdr(skb);
+	ipch->nexthdr = *skb_mac_header(skb);
+	ipch->flags = 0;
+	ipch->cpi = htons((u16 )ntohl(x->id.spi));
+	*skb_mac_header(skb) = IPPROTO_COMP;
+out_ok:
+	skb_push(skb, -skb_network_offset(skb));
+	return 0;
+}
+EXPORT_SYMBOL_GPL(ipcomp_output);
+
+static void ipcomp_free_scratches(void)
+{
+	int i;
+	void **scratches;
+
+	if (--ipcomp_scratch_users)
+		return;
+
+	scratches = ipcomp_scratches;
+	if (!scratches)
+		return;
+
+	for_each_possible_cpu(i)
+		vfree(*per_cpu_ptr(scratches, i));
+
+	free_percpu(scratches);
+}
+
+static void **ipcomp_alloc_scratches(void)
+{
+	int i;
+	void **scratches;
+
+	if (ipcomp_scratch_users++)
+		return ipcomp_scratches;
+
+	scratches = alloc_percpu(void *);
+	if (!scratches)
+		return NULL;
+
+	ipcomp_scratches = scratches;
+
+	for_each_possible_cpu(i) {
+		void *scratch = vmalloc(IPCOMP_SCRATCH_SIZE);
+		if (!scratch)
+			return NULL;
+		*per_cpu_ptr(scratches, i) = scratch;
+	}
+
+	return scratches;
+}
+
+static void ipcomp_free_tfms(struct crypto_comp **tfms)
+{
+	struct ipcomp_tfms *pos;
+	int cpu;
+
+	list_for_each_entry(pos, &ipcomp_tfms_list, list) {
+		if (pos->tfms == tfms)
+			break;
+	}
+
+	BUG_TRAP(pos);
+
+	if (--pos->users)
+		return;
+
+	list_del(&pos->list);
+	kfree(pos);
+
+	if (!tfms)
+		return;
+
+	for_each_possible_cpu(cpu) {
+		struct crypto_comp *tfm = *per_cpu_ptr(tfms, cpu);
+		crypto_free_comp(tfm);
+	}
+	free_percpu(tfms);
+}
+
+static struct crypto_comp **ipcomp_alloc_tfms(const char *alg_name)
+{
+	struct ipcomp_tfms *pos;
+	struct crypto_comp **tfms;
+	int cpu;
+
+	/* This can be any valid CPU ID so we don't need locking. */
+	cpu = raw_smp_processor_id();
+
+	list_for_each_entry(pos, &ipcomp_tfms_list, list) {
+		struct crypto_comp *tfm;
+
+		tfms = pos->tfms;
+		tfm = *per_cpu_ptr(tfms, cpu);
+
+		if (!strcmp(crypto_comp_name(tfm), alg_name)) {
+			pos->users++;
+			return tfms;
+		}
+	}
+
+	pos = kmalloc(sizeof(*pos), GFP_KERNEL);
+	if (!pos)
+		return NULL;
+
+	pos->users = 1;
+	INIT_LIST_HEAD(&pos->list);
+	list_add(&pos->list, &ipcomp_tfms_list);
+
+	pos->tfms = tfms = alloc_percpu(struct crypto_comp *);
+	if (!tfms)
+		goto error;
+
+	for_each_possible_cpu(cpu) {
+		struct crypto_comp *tfm = crypto_alloc_comp(alg_name, 0,
+							    CRYPTO_ALG_ASYNC);
+		if (IS_ERR(tfm))
+			goto error;
+		*per_cpu_ptr(tfms, cpu) = tfm;
+	}
+
+	return tfms;
+
+error:
+	ipcomp_free_tfms(tfms);
+	return NULL;
+}
+
+static void ipcomp_free_data(struct ipcomp_data *ipcd)
+{
+	if (ipcd->tfms)
+		ipcomp_free_tfms(ipcd->tfms);
+	ipcomp_free_scratches();
+}
+
+void ipcomp_destroy(struct xfrm_state *x)
+{
+	struct ipcomp_data *ipcd = x->data;
+	if (!ipcd)
+		return;
+	xfrm_state_delete_tunnel(x);
+	mutex_lock(&ipcomp_resource_mutex);
+	ipcomp_free_data(ipcd);
+	mutex_unlock(&ipcomp_resource_mutex);
+	kfree(ipcd);
+}
+EXPORT_SYMBOL_GPL(ipcomp_destroy);
+
+int ipcomp_init_state(struct xfrm_state *x)
+{
+	int err;
+	struct ipcomp_data *ipcd;
+	struct xfrm_algo_desc *calg_desc;
+
+	err = -EINVAL;
+	if (!x->calg)
+		goto out;
+
+	if (x->encap)
+		goto out;
+
+	err = -ENOMEM;
+	ipcd = kzalloc(sizeof(*ipcd), GFP_KERNEL);
+	if (!ipcd)
+		goto out;
+
+	mutex_lock(&ipcomp_resource_mutex);
+	if (!ipcomp_alloc_scratches())
+		goto error;
+
+	ipcd->tfms = ipcomp_alloc_tfms(x->calg->alg_name);
+	if (!ipcd->tfms)
+		goto error;
+	mutex_unlock(&ipcomp_resource_mutex);
+
+	calg_desc = xfrm_calg_get_byname(x->calg->alg_name, 0);
+	BUG_ON(!calg_desc);
+	ipcd->threshold = calg_desc->uinfo.comp.threshold;
+	x->data = ipcd;
+	err = 0;
+out:
+	return err;
+
+error:
+	ipcomp_free_data(ipcd);
+	mutex_unlock(&ipcomp_resource_mutex);
+	kfree(ipcd);
+	goto out;
+}
+EXPORT_SYMBOL_GPL(ipcomp_init_state);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("IP Payload Compression Protocol (IPComp) - RFC3173");
+MODULE_AUTHOR("James Morris <jmorris@intercode.com.au>");

From 7d7e5a60c62e88cb8782760bb6c4d3bd1577a6c6 Mon Sep 17 00:00:00 2001
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Fri, 25 Jul 2008 02:55:33 -0700
Subject: [PATCH 0654/1171] ipsec: ipcomp - Decompress into frags if necessary

When decompressing extremely large packets allocating them through
kmalloc is prone to failure.  Therefore it's better to use page
frags instead.

Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/xfrm/xfrm_ipcomp.c | 48 ++++++++++++++++++++++++++++++++++++------
 1 file changed, 42 insertions(+), 6 deletions(-)

diff --git a/net/xfrm/xfrm_ipcomp.c b/net/xfrm/xfrm_ipcomp.c
index b51e804fbbad..800f669083fb 100644
--- a/net/xfrm/xfrm_ipcomp.c
+++ b/net/xfrm/xfrm_ipcomp.c
@@ -17,6 +17,7 @@
 
 #include <linux/crypto.h>
 #include <linux/err.h>
+#include <linux/gfp.h>
 #include <linux/list.h>
 #include <linux/module.h>
 #include <linux/mutex.h>
@@ -49,6 +50,7 @@ static int ipcomp_decompress(struct xfrm_state *x, struct sk_buff *skb)
 	u8 *scratch = *per_cpu_ptr(ipcomp_scratches, cpu);
 	struct crypto_comp *tfm = *per_cpu_ptr(ipcd->tfms, cpu);
 	int err = crypto_comp_decompress(tfm, start, plen, scratch, &dlen);
+	int len;
 
 	if (err)
 		goto out;
@@ -58,13 +60,47 @@ static int ipcomp_decompress(struct xfrm_state *x, struct sk_buff *skb)
 		goto out;
 	}
 
-	err = pskb_expand_head(skb, 0, dlen - plen, GFP_ATOMIC);
-	if (err)
-		goto out;
+	len = dlen - plen;
+	if (len > skb_tailroom(skb))
+		len = skb_tailroom(skb);
+
+	skb->truesize += len;
+	__skb_put(skb, len);
+
+	len += plen;
+	skb_copy_to_linear_data(skb, scratch, len);
+
+	while ((scratch += len, dlen -= len) > 0) {
+		skb_frag_t *frag;
+
+		err = -EMSGSIZE;
+		if (WARN_ON(skb_shinfo(skb)->nr_frags >= MAX_SKB_FRAGS))
+			goto out;
+
+		frag = skb_shinfo(skb)->frags + skb_shinfo(skb)->nr_frags;
+		frag->page = alloc_page(GFP_ATOMIC);
+
+		err = -ENOMEM;
+		if (!frag->page)
+			goto out;
+
+		len = PAGE_SIZE;
+		if (dlen < len)
+			len = dlen;
+
+		memcpy(page_address(frag->page), scratch, len);
+
+		frag->page_offset = 0;
+		frag->size = len;
+		skb->truesize += len;
+		skb->data_len += len;
+		skb->len += len;
+
+		skb_shinfo(skb)->nr_frags++;
+	}
+
+	err = 0;
 
-	skb->truesize += dlen - plen;
-	__skb_put(skb, dlen - plen);
-	skb_copy_to_linear_data(skb, scratch, dlen);
 out:
 	put_cpu();
 	return err;

From 55679df30dfa37886cd9e22d8dea0e6974a552df Mon Sep 17 00:00:00 2001
From: Anton Vorontsov <avorontsov@ru.mvista.com>
Date: Mon, 14 Jul 2008 19:20:37 +0400
Subject: [PATCH 0655/1171] [MTD] [NAND] fsl_elbc_nand: fix section mismatch
 with CONFIG_MTD_OF_PARTS=y

With CONFIG_MTD_OF_PARTS=y I'm getting this new section mismatch in reference
from the function fsl_elbc_chip_probe() to the function
.devinit.text:of_mtd_parse_partitions()

This patch fixes the mismatch by providing __devinit annotation to the
fsl_elbc_chip_probe() function.

Signed-off-by: Anton Vorontsov <avorontsov@ru.mvista.com>
Acked-By: Scott Wood <scottwood@freescale.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/fsl_elbc_nand.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/nand/fsl_elbc_nand.c b/drivers/mtd/nand/fsl_elbc_nand.c
index d6d1ff55c4e5..9dff51351f4f 100644
--- a/drivers/mtd/nand/fsl_elbc_nand.c
+++ b/drivers/mtd/nand/fsl_elbc_nand.c
@@ -836,8 +836,8 @@ static int fsl_elbc_chip_remove(struct fsl_elbc_mtd *priv)
 	return 0;
 }
 
-static int fsl_elbc_chip_probe(struct fsl_elbc_ctrl *ctrl,
-                               struct device_node *node)
+static int __devinit fsl_elbc_chip_probe(struct fsl_elbc_ctrl *ctrl,
+					 struct device_node *node)
 {
 	struct fsl_lbc_regs __iomem *lbc = ctrl->regs;
 	struct fsl_elbc_mtd *priv;

From 998453fbf2e0709bf65ac419718ad284401b2b4f Mon Sep 17 00:00:00 2001
From: Alexey Korolev <akorolev@infradead.org>
Date: Wed, 16 Jul 2008 15:28:56 +0100
Subject: [PATCH 0656/1171] [MTD] [NOR] Fix -ETIMEO errors in CFI driver

Existing CFI driver has problems with excessive writes during erase.
If CFI driver does many writes during one erase cycle we may face the
messages with -ETIMEO error on erase operation.  It may cause the
following data corruption and kernel panics.

The reason of the issue is related to specifics of suspend operation:
if we write to flash during erase, suspend operation will cost some time
to erase procedure (for P30 it could be significant). In current version of
cfi driver the problem of many suspends is partially workarounded by adding
some time reserv to any operation (8xerase_time) but if we have many writes
during one erase the problem appears.

This patch detects the suspend and resets timer if suspend occured. It
has been well verified on different chips. No problems were found.
Could you please include the patch as it is simple and fixes bad issue.

Signed-off-by: Alexey Korolev <akorolev@infradead.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/chips/cfi_cmdset_0001.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/drivers/mtd/chips/cfi_cmdset_0001.c b/drivers/mtd/chips/cfi_cmdset_0001.c
index 324ff82a3cd9..5f1b472137a0 100644
--- a/drivers/mtd/chips/cfi_cmdset_0001.c
+++ b/drivers/mtd/chips/cfi_cmdset_0001.c
@@ -1149,7 +1149,7 @@ static int inval_cache_and_wait_for_operation(
 	struct cfi_private *cfi = map->fldrv_priv;
 	map_word status, status_OK = CMD(0x80);
 	int chip_state = chip->state;
-	unsigned int timeo, sleep_time;
+	unsigned int timeo, sleep_time, reset_timeo;
 
 	spin_unlock(chip->mutex);
 	if (inval_len)
@@ -1160,6 +1160,7 @@ static int inval_cache_and_wait_for_operation(
 	timeo = chip_op_time * 8;
 	if (!timeo)
 		timeo = 500000;
+	reset_timeo = timeo;
 	sleep_time = chip_op_time / 2;
 
 	for (;;) {
@@ -1201,6 +1202,12 @@ static int inval_cache_and_wait_for_operation(
 			remove_wait_queue(&chip->wq, &wait);
 			spin_lock(chip->mutex);
 		}
+		if (chip->erase_suspended || chip->write_suspended)  {
+			/* Suspend has occured while sleep: reset timeout */
+			timeo = reset_timeo;
+			chip->erase_suspended = 0;
+			chip->write_suspended = 0;
+		}
 	}
 
 	/* Done and happy. */

From 7b2491911540e4904498622fbee2e1a9e3120d2f Mon Sep 17 00:00:00 2001
From: Robert Jarzmik <rjarzmik@free.fr>
Date: Tue, 22 Jul 2008 09:39:00 +0200
Subject: [PATCH 0657/1171] [MTD] physmap: Fix suspend/resume/shutdown bugs.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Don't call suspend/resume functions if they have not been
defined.

Signed-off-by: Robert Jarzmik <rjarzmik@free.fr>
Acked-By: Jörn Engel <joern@logfs.org>
Signed-off-by: Uwe Kleine-König <Uwe.Kleine-Koenig@digi.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/maps/physmap.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/mtd/maps/physmap.c b/drivers/mtd/maps/physmap.c
index 1f6b9066b63e..7c8cdf49deb6 100644
--- a/drivers/mtd/maps/physmap.c
+++ b/drivers/mtd/maps/physmap.c
@@ -201,7 +201,8 @@ static int physmap_flash_suspend(struct platform_device *dev, pm_message_t state
 	int i;
 
 	for (i = 0; i < MAX_RESOURCES && info->mtd[i]; i++)
-		ret |= info->mtd[i]->suspend(info->mtd[i]);
+		if (info->mtd[i]->suspend)
+			ret |= info->mtd[i]->suspend(info->mtd[i]);
 
 	return ret;
 }
@@ -212,7 +213,8 @@ static int physmap_flash_resume(struct platform_device *dev)
 	int i;
 
 	for (i = 0; i < MAX_RESOURCES && info->mtd[i]; i++)
-		info->mtd[i]->resume(info->mtd[i]);
+		if (info->mtd[i]->resume)
+			info->mtd[i]->resume(info->mtd[i]);
 
 	return 0;
 }
@@ -223,8 +225,9 @@ static void physmap_flash_shutdown(struct platform_device *dev)
 	int i;
 
 	for (i = 0; i < MAX_RESOURCES && info->mtd[i]; i++)
-		if (info->mtd[i]->suspend(info->mtd[i]) == 0)
-			info->mtd[i]->resume(info->mtd[i]);
+		if (info->mtd[i]->suspend && info->mtd[i]->resume)
+			if (info->mtd[i]->suspend(info->mtd[i]) == 0)
+				info->mtd[i]->resume(info->mtd[i]);
 }
 #else
 #define physmap_flash_suspend NULL

From 4b5e33a7bf185c8d8568a807d9805fb155bcedd9 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig?= <Uwe.Kleine-Koenig@digi.com>
Date: Tue, 22 Jul 2008 09:39:01 +0200
Subject: [PATCH 0658/1171] [MTD] physmap: resume already suspended chips on
 failure to suspend
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

A nice side effect of this patch is that the return value of
physmap_flash_suspend in the error path is the value of the first failing
suspend callback and not the bitwise OR of all of them.

Signed-off-by: Uwe Kleine-König <Uwe.Kleine-Koenig@digi.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/maps/physmap.c | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

diff --git a/drivers/mtd/maps/physmap.c b/drivers/mtd/maps/physmap.c
index 7c8cdf49deb6..42d844f8f6bf 100644
--- a/drivers/mtd/maps/physmap.c
+++ b/drivers/mtd/maps/physmap.c
@@ -201,8 +201,19 @@ static int physmap_flash_suspend(struct platform_device *dev, pm_message_t state
 	int i;
 
 	for (i = 0; i < MAX_RESOURCES && info->mtd[i]; i++)
-		if (info->mtd[i]->suspend)
-			ret |= info->mtd[i]->suspend(info->mtd[i]);
+		if (info->mtd[i]->suspend) {
+			ret = info->mtd[i]->suspend(info->mtd[i]);
+			if (ret)
+				goto fail;
+		}
+
+	return 0;
+fail:
+	for (--i; i >= 0; --i)
+		if (info->mtd[i]->suspend) {
+			BUG_ON(!info->mtd[i]->resume);
+			info->mtd[i]->resume(info->mtd[i]);
+		}
 
 	return ret;
 }

From 7788ba71a6046de1b70e7dd45ed0bc5768a4bbd9 Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Sat, 19 Jul 2008 01:00:18 +0900
Subject: [PATCH 0659/1171] [MTD][MTDPART] Seperate main loop from
 per-partition code in add_mtd_partition
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

add_mtd_partition was a 150+ line monster consisting mostly of a single
loop.  Seperate the loop from most of the body.  Now it should be
obvious which variables are carried around from iteration to iteration.

Signed-off-by: Jörn Engel <joern@logfs.org>
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/mtdpart.c | 324 ++++++++++++++++++++++--------------------
 1 file changed, 166 insertions(+), 158 deletions(-)

diff --git a/drivers/mtd/mtdpart.c b/drivers/mtd/mtdpart.c
index 56a760a736a9..45c6f32b0bf1 100644
--- a/drivers/mtd/mtdpart.c
+++ b/drivers/mtd/mtdpart.c
@@ -313,6 +313,170 @@ int del_mtd_partitions(struct mtd_info *master)
 	return 0;
 }
 
+static struct mtd_part *add_one_partition(struct mtd_info *master,
+		const struct mtd_partition *part, int partno,
+		u_int32_t cur_offset)
+{
+	struct mtd_part *slave;
+
+	/* allocate the partition structure */
+	slave = kzalloc (sizeof(*slave), GFP_KERNEL);
+	if (!slave) {
+		printk("memory allocation error while creating partitions for \"%s\"\n",
+			master->name);
+		del_mtd_partitions(master);
+		return NULL;
+	}
+	list_add(&slave->list, &mtd_partitions);
+
+	/* set up the MTD object for this partition */
+	slave->mtd.type = master->type;
+	slave->mtd.flags = master->flags & ~part->mask_flags;
+	slave->mtd.size = part->size;
+	slave->mtd.writesize = master->writesize;
+	slave->mtd.oobsize = master->oobsize;
+	slave->mtd.oobavail = master->oobavail;
+	slave->mtd.subpage_sft = master->subpage_sft;
+
+	slave->mtd.name = part->name;
+	slave->mtd.owner = master->owner;
+
+	slave->mtd.read = part_read;
+	slave->mtd.write = part_write;
+
+	if (master->panic_write)
+		slave->mtd.panic_write = part_panic_write;
+
+	if(master->point && master->unpoint){
+		slave->mtd.point = part_point;
+		slave->mtd.unpoint = part_unpoint;
+	}
+
+	if (master->read_oob)
+		slave->mtd.read_oob = part_read_oob;
+	if (master->write_oob)
+		slave->mtd.write_oob = part_write_oob;
+	if(master->read_user_prot_reg)
+		slave->mtd.read_user_prot_reg = part_read_user_prot_reg;
+	if(master->read_fact_prot_reg)
+		slave->mtd.read_fact_prot_reg = part_read_fact_prot_reg;
+	if(master->write_user_prot_reg)
+		slave->mtd.write_user_prot_reg = part_write_user_prot_reg;
+	if(master->lock_user_prot_reg)
+		slave->mtd.lock_user_prot_reg = part_lock_user_prot_reg;
+	if(master->get_user_prot_info)
+		slave->mtd.get_user_prot_info = part_get_user_prot_info;
+	if(master->get_fact_prot_info)
+		slave->mtd.get_fact_prot_info = part_get_fact_prot_info;
+	if (master->sync)
+		slave->mtd.sync = part_sync;
+	if (!partno && master->suspend && master->resume) {
+			slave->mtd.suspend = part_suspend;
+			slave->mtd.resume = part_resume;
+	}
+	if (master->writev)
+		slave->mtd.writev = part_writev;
+	if (master->lock)
+		slave->mtd.lock = part_lock;
+	if (master->unlock)
+		slave->mtd.unlock = part_unlock;
+	if (master->block_isbad)
+		slave->mtd.block_isbad = part_block_isbad;
+	if (master->block_markbad)
+		slave->mtd.block_markbad = part_block_markbad;
+	slave->mtd.erase = part_erase;
+	slave->master = master;
+	slave->offset = part->offset;
+	slave->index = partno;
+
+	if (slave->offset == MTDPART_OFS_APPEND)
+		slave->offset = cur_offset;
+	if (slave->offset == MTDPART_OFS_NXTBLK) {
+		slave->offset = cur_offset;
+		if ((cur_offset % master->erasesize) != 0) {
+			/* Round up to next erasesize */
+			slave->offset = ((cur_offset / master->erasesize) + 1) * master->erasesize;
+			printk(KERN_NOTICE "Moving partition %d: "
+			       "0x%08x -> 0x%08x\n", partno,
+			       cur_offset, slave->offset);
+		}
+	}
+	if (slave->mtd.size == MTDPART_SIZ_FULL)
+		slave->mtd.size = master->size - slave->offset;
+
+	printk (KERN_NOTICE "0x%08x-0x%08x : \"%s\"\n", slave->offset,
+		slave->offset + slave->mtd.size, slave->mtd.name);
+
+	/* let's do some sanity checks */
+	if (slave->offset >= master->size) {
+			/* let's register it anyway to preserve ordering */
+		slave->offset = 0;
+		slave->mtd.size = 0;
+		printk ("mtd: partition \"%s\" is out of reach -- disabled\n",
+			part->name);
+	}
+	if (slave->offset + slave->mtd.size > master->size) {
+		slave->mtd.size = master->size - slave->offset;
+		printk ("mtd: partition \"%s\" extends beyond the end of device \"%s\" -- size truncated to %#x\n",
+			part->name, master->name, slave->mtd.size);
+	}
+	if (master->numeraseregions>1) {
+		/* Deal with variable erase size stuff */
+		int i;
+		struct mtd_erase_region_info *regions = master->eraseregions;
+
+		/* Find the first erase regions which is part of this partition. */
+		for (i=0; i < master->numeraseregions && regions[i].offset <= slave->offset; i++)
+			;
+
+		for (i--; i < master->numeraseregions && regions[i].offset < slave->offset + slave->mtd.size; i++) {
+			if (slave->mtd.erasesize < regions[i].erasesize) {
+				slave->mtd.erasesize = regions[i].erasesize;
+			}
+		}
+	} else {
+		/* Single erase size */
+		slave->mtd.erasesize = master->erasesize;
+	}
+
+	if ((slave->mtd.flags & MTD_WRITEABLE) &&
+	    (slave->offset % slave->mtd.erasesize)) {
+		/* Doesn't start on a boundary of major erase size */
+		/* FIXME: Let it be writable if it is on a boundary of _minor_ erase size though */
+		slave->mtd.flags &= ~MTD_WRITEABLE;
+		printk ("mtd: partition \"%s\" doesn't start on an erase block boundary -- force read-only\n",
+			part->name);
+	}
+	if ((slave->mtd.flags & MTD_WRITEABLE) &&
+	    (slave->mtd.size % slave->mtd.erasesize)) {
+		slave->mtd.flags &= ~MTD_WRITEABLE;
+		printk ("mtd: partition \"%s\" doesn't end on an erase block -- force read-only\n",
+			part->name);
+	}
+
+	slave->mtd.ecclayout = master->ecclayout;
+	if (master->block_isbad) {
+		uint32_t offs = 0;
+
+		while(offs < slave->mtd.size) {
+			if (master->block_isbad(master,
+						offs + slave->offset))
+				slave->mtd.ecc_stats.badblocks++;
+			offs += slave->mtd.erasesize;
+		}
+	}
+
+	if(part->mtdp) {	/* store the object pointer (caller may or may not register it */
+		*part->mtdp = &slave->mtd;
+		slave->registered = 0;
+	} else {
+		/* register our partition */
+		add_mtd_device(&slave->mtd);
+		slave->registered = 1;
+	}
+	return slave;
+}
+
 /*
  * This function, given a master MTD object and a partition table, creates
  * and registers slave MTD objects which are bound to the master according to
@@ -331,166 +495,10 @@ int add_mtd_partitions(struct mtd_info *master,
 	printk (KERN_NOTICE "Creating %d MTD partitions on \"%s\":\n", nbparts, master->name);
 
 	for (i = 0; i < nbparts; i++) {
-
-		/* allocate the partition structure */
-		slave = kzalloc (sizeof(*slave), GFP_KERNEL);
-		if (!slave) {
-			printk ("memory allocation error while creating partitions for \"%s\"\n",
-				master->name);
-			del_mtd_partitions(master);
+		slave = add_one_partition(master, parts + i, i, cur_offset);
+		if (!slave)
 			return -ENOMEM;
-		}
-		list_add(&slave->list, &mtd_partitions);
-
-		/* set up the MTD object for this partition */
-		slave->mtd.type = master->type;
-		slave->mtd.flags = master->flags & ~parts[i].mask_flags;
-		slave->mtd.size = parts[i].size;
-		slave->mtd.writesize = master->writesize;
-		slave->mtd.oobsize = master->oobsize;
-		slave->mtd.oobavail = master->oobavail;
-		slave->mtd.subpage_sft = master->subpage_sft;
-
-		slave->mtd.name = parts[i].name;
-		slave->mtd.owner = master->owner;
-
-		slave->mtd.read = part_read;
-		slave->mtd.write = part_write;
-
-		if (master->panic_write)
-			slave->mtd.panic_write = part_panic_write;
-
-		if(master->point && master->unpoint){
-			slave->mtd.point = part_point;
-			slave->mtd.unpoint = part_unpoint;
-		}
-
-		if (master->read_oob)
-			slave->mtd.read_oob = part_read_oob;
-		if (master->write_oob)
-			slave->mtd.write_oob = part_write_oob;
-		if(master->read_user_prot_reg)
-			slave->mtd.read_user_prot_reg = part_read_user_prot_reg;
-		if(master->read_fact_prot_reg)
-			slave->mtd.read_fact_prot_reg = part_read_fact_prot_reg;
-		if(master->write_user_prot_reg)
-			slave->mtd.write_user_prot_reg = part_write_user_prot_reg;
-		if(master->lock_user_prot_reg)
-			slave->mtd.lock_user_prot_reg = part_lock_user_prot_reg;
-		if(master->get_user_prot_info)
-			slave->mtd.get_user_prot_info = part_get_user_prot_info;
-		if(master->get_fact_prot_info)
-			slave->mtd.get_fact_prot_info = part_get_fact_prot_info;
-		if (master->sync)
-			slave->mtd.sync = part_sync;
-		if (!i && master->suspend && master->resume) {
-				slave->mtd.suspend = part_suspend;
-				slave->mtd.resume = part_resume;
-		}
-		if (master->writev)
-			slave->mtd.writev = part_writev;
-		if (master->lock)
-			slave->mtd.lock = part_lock;
-		if (master->unlock)
-			slave->mtd.unlock = part_unlock;
-		if (master->block_isbad)
-			slave->mtd.block_isbad = part_block_isbad;
-		if (master->block_markbad)
-			slave->mtd.block_markbad = part_block_markbad;
-		slave->mtd.erase = part_erase;
-		slave->master = master;
-		slave->offset = parts[i].offset;
-		slave->index = i;
-
-		if (slave->offset == MTDPART_OFS_APPEND)
-			slave->offset = cur_offset;
-		if (slave->offset == MTDPART_OFS_NXTBLK) {
-			slave->offset = cur_offset;
-			if ((cur_offset % master->erasesize) != 0) {
-				/* Round up to next erasesize */
-				slave->offset = ((cur_offset / master->erasesize) + 1) * master->erasesize;
-				printk(KERN_NOTICE "Moving partition %d: "
-				       "0x%08x -> 0x%08x\n", i,
-				       cur_offset, slave->offset);
-			}
-		}
-		if (slave->mtd.size == MTDPART_SIZ_FULL)
-			slave->mtd.size = master->size - slave->offset;
 		cur_offset = slave->offset + slave->mtd.size;
-
-		printk (KERN_NOTICE "0x%08x-0x%08x : \"%s\"\n", slave->offset,
-			slave->offset + slave->mtd.size, slave->mtd.name);
-
-		/* let's do some sanity checks */
-		if (slave->offset >= master->size) {
-				/* let's register it anyway to preserve ordering */
-			slave->offset = 0;
-			slave->mtd.size = 0;
-			printk ("mtd: partition \"%s\" is out of reach -- disabled\n",
-				parts[i].name);
-		}
-		if (slave->offset + slave->mtd.size > master->size) {
-			slave->mtd.size = master->size - slave->offset;
-			printk ("mtd: partition \"%s\" extends beyond the end of device \"%s\" -- size truncated to %#x\n",
-				parts[i].name, master->name, slave->mtd.size);
-		}
-		if (master->numeraseregions>1) {
-			/* Deal with variable erase size stuff */
-			int i;
-			struct mtd_erase_region_info *regions = master->eraseregions;
-
-			/* Find the first erase regions which is part of this partition. */
-			for (i=0; i < master->numeraseregions && slave->offset >= regions[i].offset; i++)
-				;
-
-			for (i--; i < master->numeraseregions && slave->offset + slave->mtd.size > regions[i].offset; i++) {
-				if (slave->mtd.erasesize < regions[i].erasesize) {
-					slave->mtd.erasesize = regions[i].erasesize;
-				}
-			}
-		} else {
-			/* Single erase size */
-			slave->mtd.erasesize = master->erasesize;
-		}
-
-		if ((slave->mtd.flags & MTD_WRITEABLE) &&
-		    (slave->offset % slave->mtd.erasesize)) {
-			/* Doesn't start on a boundary of major erase size */
-			/* FIXME: Let it be writable if it is on a boundary of _minor_ erase size though */
-			slave->mtd.flags &= ~MTD_WRITEABLE;
-			printk ("mtd: partition \"%s\" doesn't start on an erase block boundary -- force read-only\n",
-				parts[i].name);
-		}
-		if ((slave->mtd.flags & MTD_WRITEABLE) &&
-		    (slave->mtd.size % slave->mtd.erasesize)) {
-			slave->mtd.flags &= ~MTD_WRITEABLE;
-			printk ("mtd: partition \"%s\" doesn't end on an erase block -- force read-only\n",
-				parts[i].name);
-		}
-
-		slave->mtd.ecclayout = master->ecclayout;
-		if (master->block_isbad) {
-			uint32_t offs = 0;
-
-			while(offs < slave->mtd.size) {
-				if (master->block_isbad(master,
-							offs + slave->offset))
-					slave->mtd.ecc_stats.badblocks++;
-				offs += slave->mtd.erasesize;
-			}
-		}
-
-		if(parts[i].mtdp)
-		{	/* store the object pointer (caller may or may not register it */
-			*parts[i].mtdp = &slave->mtd;
-			slave->registered = 0;
-		}
-		else
-		{
-			/* register our partition */
-			add_mtd_device(&slave->mtd);
-			slave->registered = 1;
-		}
 	}
 
 	return 0;

From b33a2887396a1a5207e56459f62c4b132294ca58 Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Sat, 19 Jul 2008 01:00:33 +0900
Subject: [PATCH 0660/1171] [MTD][MTDPART] Handle most checkpatch findings
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Remaining are 12 warnings about long lines and 1 about braces that
could be argued about.

Signed-off-by: Jörn Engel <joern@logfs.org>
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/mtdpart.c | 129 +++++++++++++++++++++---------------------
 1 file changed, 65 insertions(+), 64 deletions(-)

diff --git a/drivers/mtd/mtdpart.c b/drivers/mtd/mtdpart.c
index 45c6f32b0bf1..9cf73d360e78 100644
--- a/drivers/mtd/mtdpart.c
+++ b/drivers/mtd/mtdpart.c
@@ -44,8 +44,8 @@ struct mtd_part {
  * to the _real_ device.
  */
 
-static int part_read (struct mtd_info *mtd, loff_t from, size_t len,
-			size_t *retlen, u_char *buf)
+static int part_read(struct mtd_info *mtd, loff_t from, size_t len,
+		size_t *retlen, u_char *buf)
 {
 	struct mtd_part *part = PART(mtd);
 	int res;
@@ -54,7 +54,7 @@ static int part_read (struct mtd_info *mtd, loff_t from, size_t len,
 		len = 0;
 	else if (from + len > mtd->size)
 		len = mtd->size - from;
-	res = part->master->read (part->master, from + part->offset,
+	res = part->master->read(part->master, from + part->offset,
 				   len, retlen, buf);
 	if (unlikely(res)) {
 		if (res == -EUCLEAN)
@@ -65,8 +65,8 @@ static int part_read (struct mtd_info *mtd, loff_t from, size_t len,
 	return res;
 }
 
-static int part_point (struct mtd_info *mtd, loff_t from, size_t len,
-			size_t *retlen, void **virt, resource_size_t *phys)
+static int part_point(struct mtd_info *mtd, loff_t from, size_t len,
+		size_t *retlen, void **virt, resource_size_t *phys)
 {
 	struct mtd_part *part = PART(mtd);
 	if (from >= mtd->size)
@@ -85,7 +85,7 @@ static void part_unpoint(struct mtd_info *mtd, loff_t from, size_t len)
 }
 
 static int part_read_oob(struct mtd_info *mtd, loff_t from,
-			 struct mtd_oob_ops *ops)
+		struct mtd_oob_ops *ops)
 {
 	struct mtd_part *part = PART(mtd);
 	int res;
@@ -105,38 +105,38 @@ static int part_read_oob(struct mtd_info *mtd, loff_t from,
 	return res;
 }
 
-static int part_read_user_prot_reg (struct mtd_info *mtd, loff_t from, size_t len,
-			size_t *retlen, u_char *buf)
+static int part_read_user_prot_reg(struct mtd_info *mtd, loff_t from,
+		size_t len, size_t *retlen, u_char *buf)
 {
 	struct mtd_part *part = PART(mtd);
-	return part->master->read_user_prot_reg (part->master, from,
+	return part->master->read_user_prot_reg(part->master, from,
 					len, retlen, buf);
 }
 
-static int part_get_user_prot_info (struct mtd_info *mtd,
-				    struct otp_info *buf, size_t len)
+static int part_get_user_prot_info(struct mtd_info *mtd,
+		struct otp_info *buf, size_t len)
 {
 	struct mtd_part *part = PART(mtd);
-	return part->master->get_user_prot_info (part->master, buf, len);
+	return part->master->get_user_prot_info(part->master, buf, len);
 }
 
-static int part_read_fact_prot_reg (struct mtd_info *mtd, loff_t from, size_t len,
-			size_t *retlen, u_char *buf)
+static int part_read_fact_prot_reg(struct mtd_info *mtd, loff_t from,
+		size_t len, size_t *retlen, u_char *buf)
 {
 	struct mtd_part *part = PART(mtd);
-	return part->master->read_fact_prot_reg (part->master, from,
+	return part->master->read_fact_prot_reg(part->master, from,
 					len, retlen, buf);
 }
 
-static int part_get_fact_prot_info (struct mtd_info *mtd,
-				    struct otp_info *buf, size_t len)
+static int part_get_fact_prot_info(struct mtd_info *mtd, struct otp_info *buf,
+		size_t len)
 {
 	struct mtd_part *part = PART(mtd);
-	return part->master->get_fact_prot_info (part->master, buf, len);
+	return part->master->get_fact_prot_info(part->master, buf, len);
 }
 
-static int part_write (struct mtd_info *mtd, loff_t to, size_t len,
-			size_t *retlen, const u_char *buf)
+static int part_write(struct mtd_info *mtd, loff_t to, size_t len,
+		size_t *retlen, const u_char *buf)
 {
 	struct mtd_part *part = PART(mtd);
 	if (!(mtd->flags & MTD_WRITEABLE))
@@ -145,12 +145,12 @@ static int part_write (struct mtd_info *mtd, loff_t to, size_t len,
 		len = 0;
 	else if (to + len > mtd->size)
 		len = mtd->size - to;
-	return part->master->write (part->master, to + part->offset,
+	return part->master->write(part->master, to + part->offset,
 				    len, retlen, buf);
 }
 
-static int part_panic_write (struct mtd_info *mtd, loff_t to, size_t len,
-			size_t *retlen, const u_char *buf)
+static int part_panic_write(struct mtd_info *mtd, loff_t to, size_t len,
+		size_t *retlen, const u_char *buf)
 {
 	struct mtd_part *part = PART(mtd);
 	if (!(mtd->flags & MTD_WRITEABLE))
@@ -159,12 +159,12 @@ static int part_panic_write (struct mtd_info *mtd, loff_t to, size_t len,
 		len = 0;
 	else if (to + len > mtd->size)
 		len = mtd->size - to;
-	return part->master->panic_write (part->master, to + part->offset,
+	return part->master->panic_write(part->master, to + part->offset,
 				    len, retlen, buf);
 }
 
 static int part_write_oob(struct mtd_info *mtd, loff_t to,
-			 struct mtd_oob_ops *ops)
+		struct mtd_oob_ops *ops)
 {
 	struct mtd_part *part = PART(mtd);
 
@@ -178,31 +178,32 @@ static int part_write_oob(struct mtd_info *mtd, loff_t to,
 	return part->master->write_oob(part->master, to + part->offset, ops);
 }
 
-static int part_write_user_prot_reg (struct mtd_info *mtd, loff_t from, size_t len,
-			size_t *retlen, u_char *buf)
+static int part_write_user_prot_reg(struct mtd_info *mtd, loff_t from,
+		size_t len, size_t *retlen, u_char *buf)
 {
 	struct mtd_part *part = PART(mtd);
-	return part->master->write_user_prot_reg (part->master, from,
+	return part->master->write_user_prot_reg(part->master, from,
 					len, retlen, buf);
 }
 
-static int part_lock_user_prot_reg (struct mtd_info *mtd, loff_t from, size_t len)
+static int part_lock_user_prot_reg(struct mtd_info *mtd, loff_t from,
+		size_t len)
 {
 	struct mtd_part *part = PART(mtd);
-	return part->master->lock_user_prot_reg (part->master, from, len);
+	return part->master->lock_user_prot_reg(part->master, from, len);
 }
 
-static int part_writev (struct mtd_info *mtd,  const struct kvec *vecs,
-			 unsigned long count, loff_t to, size_t *retlen)
+static int part_writev(struct mtd_info *mtd, const struct kvec *vecs,
+		unsigned long count, loff_t to, size_t *retlen)
 {
 	struct mtd_part *part = PART(mtd);
 	if (!(mtd->flags & MTD_WRITEABLE))
 		return -EROFS;
-	return part->master->writev (part->master, vecs, count,
+	return part->master->writev(part->master, vecs, count,
 					to + part->offset, retlen);
 }
 
-static int part_erase (struct mtd_info *mtd, struct erase_info *instr)
+static int part_erase(struct mtd_info *mtd, struct erase_info *instr)
 {
 	struct mtd_part *part = PART(mtd);
 	int ret;
@@ -234,7 +235,7 @@ void mtd_erase_callback(struct erase_info *instr)
 }
 EXPORT_SYMBOL_GPL(mtd_erase_callback);
 
-static int part_lock (struct mtd_info *mtd, loff_t ofs, size_t len)
+static int part_lock(struct mtd_info *mtd, loff_t ofs, size_t len)
 {
 	struct mtd_part *part = PART(mtd);
 	if ((len + ofs) > mtd->size)
@@ -242,7 +243,7 @@ static int part_lock (struct mtd_info *mtd, loff_t ofs, size_t len)
 	return part->master->lock(part->master, ofs + part->offset, len);
 }
 
-static int part_unlock (struct mtd_info *mtd, loff_t ofs, size_t len)
+static int part_unlock(struct mtd_info *mtd, loff_t ofs, size_t len)
 {
 	struct mtd_part *part = PART(mtd);
 	if ((len + ofs) > mtd->size)
@@ -268,7 +269,7 @@ static void part_resume(struct mtd_info *mtd)
 	part->master->resume(part->master);
 }
 
-static int part_block_isbad (struct mtd_info *mtd, loff_t ofs)
+static int part_block_isbad(struct mtd_info *mtd, loff_t ofs)
 {
 	struct mtd_part *part = PART(mtd);
 	if (ofs >= mtd->size)
@@ -277,7 +278,7 @@ static int part_block_isbad (struct mtd_info *mtd, loff_t ofs)
 	return part->master->block_isbad(part->master, ofs);
 }
 
-static int part_block_markbad (struct mtd_info *mtd, loff_t ofs)
+static int part_block_markbad(struct mtd_info *mtd, loff_t ofs)
 {
 	struct mtd_part *part = PART(mtd);
 	int res;
@@ -305,13 +306,14 @@ int del_mtd_partitions(struct mtd_info *master)
 	list_for_each_entry_safe(slave, next, &mtd_partitions, list)
 		if (slave->master == master) {
 			list_del(&slave->list);
-			if(slave->registered)
+			if (slave->registered)
 				del_mtd_device(&slave->mtd);
 			kfree(slave);
 		}
 
 	return 0;
 }
+EXPORT_SYMBOL(del_mtd_partitions);
 
 static struct mtd_part *add_one_partition(struct mtd_info *master,
 		const struct mtd_partition *part, int partno,
@@ -320,9 +322,9 @@ static struct mtd_part *add_one_partition(struct mtd_info *master,
 	struct mtd_part *slave;
 
 	/* allocate the partition structure */
-	slave = kzalloc (sizeof(*slave), GFP_KERNEL);
+	slave = kzalloc(sizeof(*slave), GFP_KERNEL);
 	if (!slave) {
-		printk("memory allocation error while creating partitions for \"%s\"\n",
+		printk(KERN_ERR"memory allocation error while creating partitions for \"%s\"\n",
 			master->name);
 		del_mtd_partitions(master);
 		return NULL;
@@ -347,7 +349,7 @@ static struct mtd_part *add_one_partition(struct mtd_info *master,
 	if (master->panic_write)
 		slave->mtd.panic_write = part_panic_write;
 
-	if(master->point && master->unpoint){
+	if (master->point && master->unpoint) {
 		slave->mtd.point = part_point;
 		slave->mtd.unpoint = part_unpoint;
 	}
@@ -356,17 +358,17 @@ static struct mtd_part *add_one_partition(struct mtd_info *master,
 		slave->mtd.read_oob = part_read_oob;
 	if (master->write_oob)
 		slave->mtd.write_oob = part_write_oob;
-	if(master->read_user_prot_reg)
+	if (master->read_user_prot_reg)
 		slave->mtd.read_user_prot_reg = part_read_user_prot_reg;
-	if(master->read_fact_prot_reg)
+	if (master->read_fact_prot_reg)
 		slave->mtd.read_fact_prot_reg = part_read_fact_prot_reg;
-	if(master->write_user_prot_reg)
+	if (master->write_user_prot_reg)
 		slave->mtd.write_user_prot_reg = part_write_user_prot_reg;
-	if(master->lock_user_prot_reg)
+	if (master->lock_user_prot_reg)
 		slave->mtd.lock_user_prot_reg = part_lock_user_prot_reg;
-	if(master->get_user_prot_info)
+	if (master->get_user_prot_info)
 		slave->mtd.get_user_prot_info = part_get_user_prot_info;
-	if(master->get_fact_prot_info)
+	if (master->get_fact_prot_info)
 		slave->mtd.get_fact_prot_info = part_get_fact_prot_info;
 	if (master->sync)
 		slave->mtd.sync = part_sync;
@@ -404,7 +406,7 @@ static struct mtd_part *add_one_partition(struct mtd_info *master,
 	if (slave->mtd.size == MTDPART_SIZ_FULL)
 		slave->mtd.size = master->size - slave->offset;
 
-	printk (KERN_NOTICE "0x%08x-0x%08x : \"%s\"\n", slave->offset,
+	printk(KERN_NOTICE "0x%08x-0x%08x : \"%s\"\n", slave->offset,
 		slave->offset + slave->mtd.size, slave->mtd.name);
 
 	/* let's do some sanity checks */
@@ -412,21 +414,21 @@ static struct mtd_part *add_one_partition(struct mtd_info *master,
 			/* let's register it anyway to preserve ordering */
 		slave->offset = 0;
 		slave->mtd.size = 0;
-		printk ("mtd: partition \"%s\" is out of reach -- disabled\n",
+		printk(KERN_ERR"mtd: partition \"%s\" is out of reach -- disabled\n",
 			part->name);
 	}
 	if (slave->offset + slave->mtd.size > master->size) {
 		slave->mtd.size = master->size - slave->offset;
-		printk ("mtd: partition \"%s\" extends beyond the end of device \"%s\" -- size truncated to %#x\n",
+		printk(KERN_WARNING"mtd: partition \"%s\" extends beyond the end of device \"%s\" -- size truncated to %#x\n",
 			part->name, master->name, slave->mtd.size);
 	}
-	if (master->numeraseregions>1) {
+	if (master->numeraseregions > 1) {
 		/* Deal with variable erase size stuff */
 		int i;
 		struct mtd_erase_region_info *regions = master->eraseregions;
 
 		/* Find the first erase regions which is part of this partition. */
-		for (i=0; i < master->numeraseregions && regions[i].offset <= slave->offset; i++)
+		for (i = 0; i < master->numeraseregions && regions[i].offset <= slave->offset; i++)
 			;
 
 		for (i--; i < master->numeraseregions && regions[i].offset < slave->offset + slave->mtd.size; i++) {
@@ -442,15 +444,16 @@ static struct mtd_part *add_one_partition(struct mtd_info *master,
 	if ((slave->mtd.flags & MTD_WRITEABLE) &&
 	    (slave->offset % slave->mtd.erasesize)) {
 		/* Doesn't start on a boundary of major erase size */
-		/* FIXME: Let it be writable if it is on a boundary of _minor_ erase size though */
+		/* FIXME: Let it be writable if it is on a boundary of
+		 * _minor_ erase size though */
 		slave->mtd.flags &= ~MTD_WRITEABLE;
-		printk ("mtd: partition \"%s\" doesn't start on an erase block boundary -- force read-only\n",
+		printk(KERN_WARNING"mtd: partition \"%s\" doesn't start on an erase block boundary -- force read-only\n",
 			part->name);
 	}
 	if ((slave->mtd.flags & MTD_WRITEABLE) &&
 	    (slave->mtd.size % slave->mtd.erasesize)) {
 		slave->mtd.flags &= ~MTD_WRITEABLE;
-		printk ("mtd: partition \"%s\" doesn't end on an erase block -- force read-only\n",
+		printk(KERN_WARNING"mtd: partition \"%s\" doesn't end on an erase block -- force read-only\n",
 			part->name);
 	}
 
@@ -458,7 +461,7 @@ static struct mtd_part *add_one_partition(struct mtd_info *master,
 	if (master->block_isbad) {
 		uint32_t offs = 0;
 
-		while(offs < slave->mtd.size) {
+		while (offs < slave->mtd.size) {
 			if (master->block_isbad(master,
 						offs + slave->offset))
 				slave->mtd.ecc_stats.badblocks++;
@@ -466,7 +469,8 @@ static struct mtd_part *add_one_partition(struct mtd_info *master,
 		}
 	}
 
-	if(part->mtdp) {	/* store the object pointer (caller may or may not register it */
+	if (part->mtdp) {
+		/* store the object pointer (caller may or may not register it*/
 		*part->mtdp = &slave->mtd;
 		slave->registered = 0;
 	} else {
@@ -492,7 +496,7 @@ int add_mtd_partitions(struct mtd_info *master,
 	u_int32_t cur_offset = 0;
 	int i;
 
-	printk (KERN_NOTICE "Creating %d MTD partitions on \"%s\":\n", nbparts, master->name);
+	printk(KERN_NOTICE "Creating %d MTD partitions on \"%s\":\n", nbparts, master->name);
 
 	for (i = 0; i < nbparts; i++) {
 		slave = add_one_partition(master, parts + i, i, cur_offset);
@@ -503,9 +507,7 @@ int add_mtd_partitions(struct mtd_info *master,
 
 	return 0;
 }
-
 EXPORT_SYMBOL(add_mtd_partitions);
-EXPORT_SYMBOL(del_mtd_partitions);
 
 static DEFINE_SPINLOCK(part_parser_lock);
 static LIST_HEAD(part_parsers);
@@ -535,6 +537,7 @@ int register_mtd_parser(struct mtd_part_parser *p)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(register_mtd_parser);
 
 int deregister_mtd_parser(struct mtd_part_parser *p)
 {
@@ -543,6 +546,7 @@ int deregister_mtd_parser(struct mtd_part_parser *p)
 	spin_unlock(&part_parser_lock);
 	return 0;
 }
+EXPORT_SYMBOL_GPL(deregister_mtd_parser);
 
 int parse_mtd_partitions(struct mtd_info *master, const char **types,
 			 struct mtd_partition **pparts, unsigned long origin)
@@ -570,7 +574,4 @@ int parse_mtd_partitions(struct mtd_info *master, const char **types,
 	}
 	return ret;
 }
-
 EXPORT_SYMBOL_GPL(parse_mtd_partitions);
-EXPORT_SYMBOL_GPL(register_mtd_parser);
-EXPORT_SYMBOL_GPL(deregister_mtd_parser);

From 6910c1368104d50e6b6afc6c8b7e9d1670a374e7 Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Sat, 19 Jul 2008 01:00:57 +0900
Subject: [PATCH 0661/1171] [MTD][MTDPART] Cleanup and document the erase
 region handling
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Mostly simplifying the loops.  Now everything fits into 80 columns,
is easier to read and the finer details have extra comments.

Signed-off-by: Jörn Engel <joern@logfs.org>
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/mtdpart.c | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

diff --git a/drivers/mtd/mtdpart.c b/drivers/mtd/mtdpart.c
index 9cf73d360e78..5aac59c21ead 100644
--- a/drivers/mtd/mtdpart.c
+++ b/drivers/mtd/mtdpart.c
@@ -424,18 +424,24 @@ static struct mtd_part *add_one_partition(struct mtd_info *master,
 	}
 	if (master->numeraseregions > 1) {
 		/* Deal with variable erase size stuff */
-		int i;
+		int i, max = master->numeraseregions;
+		u32 end = slave->offset + slave->mtd.size;
 		struct mtd_erase_region_info *regions = master->eraseregions;
 
-		/* Find the first erase regions which is part of this partition. */
-		for (i = 0; i < master->numeraseregions && regions[i].offset <= slave->offset; i++)
+		/* Find the first erase regions which is part of this
+		 * partition. */
+		for (i = 0; i < max && regions[i].offset <= slave->offset; i++)
 			;
+		/* The loop searched for the region _behind_ the first one */
+		i--;
 
-		for (i--; i < master->numeraseregions && regions[i].offset < slave->offset + slave->mtd.size; i++) {
+		/* Pick biggest erasesize */
+		for (; i < max && regions[i].offset < end; i++) {
 			if (slave->mtd.erasesize < regions[i].erasesize) {
 				slave->mtd.erasesize = regions[i].erasesize;
 			}
 		}
+		BUG_ON(slave->mtd.erasesize == 0);
 	} else {
 		/* Single erase size */
 		slave->mtd.erasesize = master->erasesize;

From f636ffb420f0f9059c1d0b841afd691657246ad6 Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Sat, 19 Jul 2008 01:01:22 +0900
Subject: [PATCH 0662/1171] [MTD][MTDPART] Fix a division by zero bug
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

When detecting a partition beyond the end of the device, skip most of
the initialisation, in particular those bits causing a division by zero.

Signed-off-by: Jörn Engel <joern@logfs.org>
Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/mtdpart.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/mtd/mtdpart.c b/drivers/mtd/mtdpart.c
index 5aac59c21ead..edb90b58a9b1 100644
--- a/drivers/mtd/mtdpart.c
+++ b/drivers/mtd/mtdpart.c
@@ -411,11 +411,12 @@ static struct mtd_part *add_one_partition(struct mtd_info *master,
 
 	/* let's do some sanity checks */
 	if (slave->offset >= master->size) {
-			/* let's register it anyway to preserve ordering */
+		/* let's register it anyway to preserve ordering */
 		slave->offset = 0;
 		slave->mtd.size = 0;
 		printk(KERN_ERR"mtd: partition \"%s\" is out of reach -- disabled\n",
 			part->name);
+		goto out_register;
 	}
 	if (slave->offset + slave->mtd.size > master->size) {
 		slave->mtd.size = master->size - slave->offset;
@@ -475,6 +476,7 @@ static struct mtd_part *add_one_partition(struct mtd_info *master,
 		}
 	}
 
+out_register:
 	if (part->mtdp) {
 		/* store the object pointer (caller may or may not register it*/
 		*part->mtdp = &slave->mtd;

From ca6f12c67ed19718cf37d0f531af9438de85b70c Mon Sep 17 00:00:00 2001
From: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Date: Wed, 16 Jul 2008 00:09:15 +0900
Subject: [PATCH 0663/1171] [MTD] jedec_probe: Fix SST 16-bit chip detection

The unlock_addr rework in kernel 2.6.25 breaks 16-bit SST chips.  SST
39LF160 and SST 39VF1601 are both 16-bit only chip (do not have BYTE#
pin) and new uaddr value is not correct for them.  Add
MTD_UADDR_0xAAAA_0x5555 for those chips.  Tested with SST 39VF1601
chip.

Signed-off-by: Atsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/chips/jedec_probe.c | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

diff --git a/drivers/mtd/chips/jedec_probe.c b/drivers/mtd/chips/jedec_probe.c
index afb35e3a3cee..dbba5abf0db8 100644
--- a/drivers/mtd/chips/jedec_probe.c
+++ b/drivers/mtd/chips/jedec_probe.c
@@ -200,6 +200,7 @@ enum uaddr {
 	MTD_UADDR_0x0555_0x0AAA,
 	MTD_UADDR_0x5555_0x2AAA,
 	MTD_UADDR_0x0AAA_0x0555,
+	MTD_UADDR_0xAAAA_0x5555,
 	MTD_UADDR_DONT_CARE,		/* Requires an arbitrary address */
 	MTD_UADDR_UNNECESSARY,		/* Does not require any address */
 };
@@ -247,6 +248,11 @@ static const struct unlock_addr  unlock_addrs[] = {
 		.addr2 = 0x0555
 	},
 
+	[MTD_UADDR_0xAAAA_0x5555] = {
+		.addr1 = 0xaaaa,
+		.addr2 = 0x5555
+	},
+
 	[MTD_UADDR_DONT_CARE] = {
 		.addr1 = 0x0000,      /* Doesn't matter which address */
 		.addr2 = 0x0000       /* is used - must be last entry */
@@ -1461,8 +1467,8 @@ static const struct amd_flash_info jedec_table[] = {
 		.mfr_id		= MANUFACTURER_SST,     /* should be CFI */
 		.dev_id		= SST39LF160,
 		.name		= "SST 39LF160",
-		.devtypes	= CFI_DEVICETYPE_X16|CFI_DEVICETYPE_X8,
-		.uaddr		= MTD_UADDR_0x5555_0x2AAA,	/* ???? */
+		.devtypes	= CFI_DEVICETYPE_X16,
+		.uaddr		= MTD_UADDR_0xAAAA_0x5555,
 		.dev_size	= SIZE_2MiB,
 		.cmd_set	= P_ID_AMD_STD,
 		.nr_regions	= 2,
@@ -1474,8 +1480,8 @@ static const struct amd_flash_info jedec_table[] = {
 		.mfr_id		= MANUFACTURER_SST,     /* should be CFI */
 		.dev_id		= SST39VF1601,
 		.name		= "SST 39VF1601",
-		.devtypes	= CFI_DEVICETYPE_X16|CFI_DEVICETYPE_X8,
-		.uaddr		= MTD_UADDR_0x5555_0x2AAA,	/* ???? */
+		.devtypes	= CFI_DEVICETYPE_X16,
+		.uaddr		= MTD_UADDR_0xAAAA_0x5555,
 		.dev_size	= SIZE_2MiB,
 		.cmd_set	= P_ID_AMD_STD,
 		.nr_regions	= 2,

From ee39a0e61b8a307576b5e26057f8257444b5c9c1 Mon Sep 17 00:00:00 2001
From: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Date: Tue, 15 Jul 2008 23:04:35 +0900
Subject: [PATCH 0664/1171] [MTD][NAND] au1550nd: remove unused variable

Remove unused variable from au1550 NAND driver.

Signed-off-by: Yoichi Yuasa <yoichi_yuasa@tripeaks.co.jp>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/au1550nd.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/mtd/nand/au1550nd.c b/drivers/mtd/nand/au1550nd.c
index 22ad9f367760..761946ea45b1 100644
--- a/drivers/mtd/nand/au1550nd.c
+++ b/drivers/mtd/nand/au1550nd.c
@@ -602,8 +602,6 @@ module_init(au1xxx_nand_init);
  */
 static void __exit au1550_cleanup(void)
 {
-	struct nand_chip *this = (struct nand_chip *)&au1550_mtd[1];
-
 	/* Release resources, unregister device */
 	nand_release(au1550_mtd);
 

From 30821fee4f0cb3e6d241d9f7ddc37742212e3eb7 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-mtd@fluff.org>
Date: Tue, 15 Jul 2008 11:58:31 +0100
Subject: [PATCH 0665/1171] CPUFREQ: S3C24XX NAND driver frequency scaling
 support.

Add support for CPU frequency scalling to the S3C24XX NAND driver.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/s3c2410.c | 143 +++++++++++++++++++++++++++++++------
 1 file changed, 122 insertions(+), 21 deletions(-)

diff --git a/drivers/mtd/nand/s3c2410.c b/drivers/mtd/nand/s3c2410.c
index 35c6db54c985..556139ed1fdf 100644
--- a/drivers/mtd/nand/s3c2410.c
+++ b/drivers/mtd/nand/s3c2410.c
@@ -36,6 +36,7 @@
 #include <linux/err.h>
 #include <linux/slab.h>
 #include <linux/clk.h>
+#include <linux/cpufreq.h>
 
 #include <linux/mtd/mtd.h>
 #include <linux/mtd/nand.h>
@@ -104,8 +105,13 @@ struct s3c2410_nand_info {
 	int				sel_bit;
 	int				mtd_count;
 	unsigned long			save_sel;
+	unsigned long			clk_rate;
 
 	enum s3c_cpu_type		cpu_type;
+
+#ifdef CONFIG_CPU_FREQ
+	struct notifier_block	freq_transition;
+#endif
 };
 
 /* conversion functions */
@@ -163,17 +169,18 @@ static int s3c_nand_calc_rate(int wanted, unsigned long clk, int max)
 
 /* controller setup */
 
-static int s3c2410_nand_inithw(struct s3c2410_nand_info *info,
-			       struct platform_device *pdev)
+static int s3c2410_nand_setrate(struct s3c2410_nand_info *info)
 {
-	struct s3c2410_platform_nand *plat = to_nand_plat(pdev);
-	unsigned long clkrate = clk_get_rate(info->clk);
+	struct s3c2410_platform_nand *plat = info->platform;
 	int tacls_max = (info->cpu_type == TYPE_S3C2412) ? 8 : 4;
 	int tacls, twrph0, twrph1;
-	unsigned long cfg = 0;
+	unsigned long clkrate = clk_get_rate(info->clk);
+	unsigned long set, cfg, mask;
+	unsigned long flags;
 
 	/* calculate the timing information for the controller */
 
+	info->clk_rate = clkrate;
 	clkrate /= 1000;	/* turn clock into kHz for ease of use */
 
 	if (plat != NULL) {
@@ -195,28 +202,69 @@ static int s3c2410_nand_inithw(struct s3c2410_nand_info *info,
 	dev_info(info->device, "Tacls=%d, %dns Twrph0=%d %dns, Twrph1=%d %dns\n",
 	       tacls, to_ns(tacls, clkrate), twrph0, to_ns(twrph0, clkrate), twrph1, to_ns(twrph1, clkrate));
 
+	switch (info->cpu_type) {
+	case TYPE_S3C2410:
+		mask = (S3C2410_NFCONF_TACLS(3) |
+			S3C2410_NFCONF_TWRPH0(7) |
+			S3C2410_NFCONF_TWRPH1(7));
+		set = S3C2410_NFCONF_EN;
+		set |= S3C2410_NFCONF_TACLS(tacls - 1);
+		set |= S3C2410_NFCONF_TWRPH0(twrph0 - 1);
+		set |= S3C2410_NFCONF_TWRPH1(twrph1 - 1);
+		break;
+
+	case TYPE_S3C2440:
+	case TYPE_S3C2412:
+		mask = (S3C2410_NFCONF_TACLS(tacls_max - 1) |
+			S3C2410_NFCONF_TWRPH0(7) |
+			S3C2410_NFCONF_TWRPH1(7));
+
+		set = S3C2440_NFCONF_TACLS(tacls - 1);
+		set |= S3C2440_NFCONF_TWRPH0(twrph0 - 1);
+		set |= S3C2440_NFCONF_TWRPH1(twrph1 - 1);
+		break;
+
+	default:
+		/* keep compiler happy */
+		mask = 0;
+		set = 0;
+		BUG();
+	}
+
+	dev_dbg(info->device, "NF_CONF is 0x%lx\n", cfg);
+
+	local_irq_save(flags);
+
+	cfg = readl(info->regs + S3C2410_NFCONF);
+	cfg &= ~mask;
+	cfg |= set;
+	writel(cfg, info->regs + S3C2410_NFCONF);
+
+	local_irq_restore(flags);
+
+	return 0;
+}
+
+static int s3c2410_nand_inithw(struct s3c2410_nand_info *info)
+{
+	int ret;
+
+	ret = s3c2410_nand_setrate(info);
+	if (ret < 0)
+		return ret;
+
  	switch (info->cpu_type) {
  	case TYPE_S3C2410:
-		cfg = S3C2410_NFCONF_EN;
-		cfg |= S3C2410_NFCONF_TACLS(tacls - 1);
-		cfg |= S3C2410_NFCONF_TWRPH0(twrph0 - 1);
-		cfg |= S3C2410_NFCONF_TWRPH1(twrph1 - 1);
+	default:
 		break;
 
  	case TYPE_S3C2440:
  	case TYPE_S3C2412:
-		cfg = S3C2440_NFCONF_TACLS(tacls - 1);
-		cfg |= S3C2440_NFCONF_TWRPH0(twrph0 - 1);
-		cfg |= S3C2440_NFCONF_TWRPH1(twrph1 - 1);
-
 		/* enable the controller and de-assert nFCE */
 
 		writel(S3C2440_NFCONT_ENABLE, info->regs + S3C2440_NFCONT);
 	}
 
-	dev_dbg(info->device, "NF_CONF is 0x%lx\n", cfg);
-
-	writel(cfg, info->regs + S3C2410_NFCONF);
 	return 0;
 }
 
@@ -497,6 +545,52 @@ static void s3c2440_nand_write_buf(struct mtd_info *mtd, const u_char *buf, int
 	writesl(info->regs + S3C2440_NFDATA, buf, len / 4);
 }
 
+/* cpufreq driver support */
+
+#ifdef CONFIG_CPU_FREQ
+
+static int s3c2410_nand_cpufreq_transition(struct notifier_block *nb,
+					  unsigned long val, void *data)
+{
+	struct s3c2410_nand_info *info;
+	unsigned long newclk;
+
+	info = container_of(nb, struct s3c2410_nand_info, freq_transition);
+	newclk = clk_get_rate(info->clk);
+
+	if ((val == CPUFREQ_POSTCHANGE && newclk < info->clk_rate) ||
+	    (val == CPUFREQ_PRECHANGE && newclk > info->clk_rate)) {
+		s3c2410_nand_setrate(info);
+	}
+
+	return 0;
+}
+
+static inline int s3c2410_nand_cpufreq_register(struct s3c2410_nand_info *info)
+{
+	info->freq_transition.notifier_call = s3c2410_nand_cpufreq_transition;
+
+	return cpufreq_register_notifier(&info->freq_transition,
+					 CPUFREQ_TRANSITION_NOTIFIER);
+}
+
+static inline void s3c2410_nand_cpufreq_deregister(struct s3c2410_nand_info *info)
+{
+	cpufreq_unregister_notifier(&info->freq_transition,
+				    CPUFREQ_TRANSITION_NOTIFIER);
+}
+
+#else
+static inline int s3c2410_nand_cpufreq_register(struct s3c2410_nand_info *info)
+{
+	return 0;
+}
+
+static inline void s3c2410_nand_cpufreq_deregister(struct s3c2410_nand_info *info)
+{
+}
+#endif
+
 /* device management functions */
 
 static int s3c2410_nand_remove(struct platform_device *pdev)
@@ -508,9 +602,10 @@ static int s3c2410_nand_remove(struct platform_device *pdev)
 	if (info == NULL)
 		return 0;
 
-	/* first thing we need to do is release all our mtds
-	 * and their partitions, then go through freeing the
-	 * resources used
+	s3c2410_nand_cpufreq_deregister(info);
+
+	/* Release all our mtds  and their partitions, then go through
+	 * freeing the resources used
 	 */
 
 	if (info->mtds != NULL) {
@@ -769,7 +864,7 @@ static int s3c24xx_nand_probe(struct platform_device *pdev,
 
 	/* initialise the hardware */
 
-	err = s3c2410_nand_inithw(info, pdev);
+	err = s3c2410_nand_inithw(info);
 	if (err != 0)
 		goto exit_error;
 
@@ -812,6 +907,12 @@ static int s3c24xx_nand_probe(struct platform_device *pdev,
 			sets++;
 	}
 
+	err = s3c2410_nand_cpufreq_register(info);
+	if (err < 0) {
+		dev_err(&pdev->dev, "failed to init cpufreq support\n");
+		goto exit_error;
+	}
+
 	if (allow_clk_stop(info)) {
 		dev_info(&pdev->dev, "clock idle support enabled\n");
 		clk_disable(info->clk);
@@ -859,7 +960,7 @@ static int s3c24xx_nand_resume(struct platform_device *dev)
 
 	if (info) {
 		clk_enable(info->clk);
-		s3c2410_nand_inithw(info, dev);
+		s3c2410_nand_inithw(info);
 
 		/* Restore the state of the nFCE line. */
 

From 3d45955962496879dead8d4dd70bb9a23b07154b Mon Sep 17 00:00:00 2001
From: Alexey Korolev <akorolev@infradead.org>
Date: Thu, 15 May 2008 17:23:18 +0100
Subject: [PATCH 0666/1171] [MTD] [NAND] subpage read feature as a way to
 increase performance.
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

This patch enables NAND subpage read functionality.
If upper layer drivers are requesting to read non page aligned data NAND
subpage-read functionality reads the only whose ECC regions which include
requested data when original code reads whole page.
This significantly improves performance in many cases.

Here are some digits :

UBI volume mount time
No subpage reads: 5.75 seconds
Subpage read patch: 2.42 seconds

Open/stat time for files on JFFS2 volume:
No subpage read  0m 5.36s
Subpage read     0m 2.88s

Signed-off-by Alexey Korolev <akorolev@infradead.org>
Acked-by: Artem Bityutskiy <dedekind@infradead.org>
Acked-by: Jörn Engel <joern@logfs.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 drivers/mtd/nand/nand_base.c | 87 +++++++++++++++++++++++++++++++++++-
 include/linux/mtd/nand.h     |  5 +++
 2 files changed, 91 insertions(+), 1 deletion(-)

diff --git a/drivers/mtd/nand/nand_base.c b/drivers/mtd/nand/nand_base.c
index ba1bdf787323..d1129bae6c27 100644
--- a/drivers/mtd/nand/nand_base.c
+++ b/drivers/mtd/nand/nand_base.c
@@ -797,6 +797,87 @@ static int nand_read_page_swecc(struct mtd_info *mtd, struct nand_chip *chip,
 	return 0;
 }
 
+/**
+ * nand_read_subpage - [REPLACABLE] software ecc based sub-page read function
+ * @mtd:	mtd info structure
+ * @chip:	nand chip info structure
+ * @dataofs	offset of requested data within the page
+ * @readlen	data length
+ * @buf:	buffer to store read data
+ */
+static int nand_read_subpage(struct mtd_info *mtd, struct nand_chip *chip, uint32_t data_offs, uint32_t readlen, uint8_t *bufpoi)
+{
+	int start_step, end_step, num_steps;
+	uint32_t *eccpos = chip->ecc.layout->eccpos;
+	uint8_t *p;
+	int data_col_addr, i, gaps = 0;
+	int datafrag_len, eccfrag_len, aligned_len, aligned_pos;
+	int busw = (chip->options & NAND_BUSWIDTH_16) ? 2 : 1;
+
+	/* Column address wihin the page aligned to ECC size (256bytes). */
+	start_step = data_offs / chip->ecc.size;
+	end_step = (data_offs + readlen - 1) / chip->ecc.size;
+	num_steps = end_step - start_step + 1;
+
+	/* Data size aligned to ECC ecc.size*/
+	datafrag_len = num_steps * chip->ecc.size;
+	eccfrag_len = num_steps * chip->ecc.bytes;
+
+	data_col_addr = start_step * chip->ecc.size;
+	/* If we read not a page aligned data */
+	if (data_col_addr != 0)
+		chip->cmdfunc(mtd, NAND_CMD_RNDOUT, data_col_addr, -1);
+
+	p = bufpoi + data_col_addr;
+	chip->read_buf(mtd, p, datafrag_len);
+
+	/* Calculate  ECC */
+	for (i = 0; i < eccfrag_len ; i += chip->ecc.bytes, p += chip->ecc.size)
+		chip->ecc.calculate(mtd, p, &chip->buffers->ecccalc[i]);
+
+	/* The performance is faster if to position offsets
+	   according to ecc.pos. Let make sure here that
+	   there are no gaps in ecc positions */
+	for (i = 0; i < eccfrag_len - 1; i++) {
+		if (eccpos[i + start_step * chip->ecc.bytes] + 1 !=
+			eccpos[i + start_step * chip->ecc.bytes + 1]) {
+			gaps = 1;
+			break;
+		}
+	}
+	if (gaps) {
+		chip->cmdfunc(mtd, NAND_CMD_RNDOUT, mtd->writesize, -1);
+		chip->read_buf(mtd, chip->oob_poi, mtd->oobsize);
+	} else {
+		/* send the command to read the particular ecc bytes */
+		/* take care about buswidth alignment in read_buf */
+		aligned_pos = eccpos[start_step * chip->ecc.bytes] & ~(busw - 1);
+		aligned_len = eccfrag_len;
+		if (eccpos[start_step * chip->ecc.bytes] & (busw - 1))
+			aligned_len++;
+		if (eccpos[(start_step + num_steps) * chip->ecc.bytes] & (busw - 1))
+			aligned_len++;
+
+		chip->cmdfunc(mtd, NAND_CMD_RNDOUT, mtd->writesize + aligned_pos, -1);
+		chip->read_buf(mtd, &chip->oob_poi[aligned_pos], aligned_len);
+	}
+
+	for (i = 0; i < eccfrag_len; i++)
+		chip->buffers->ecccode[i] = chip->oob_poi[eccpos[i + start_step * chip->ecc.bytes]];
+
+	p = bufpoi + data_col_addr;
+	for (i = 0; i < eccfrag_len ; i += chip->ecc.bytes, p += chip->ecc.size) {
+		int stat;
+
+		stat = chip->ecc.correct(mtd, p, &chip->buffers->ecccode[i], &chip->buffers->ecccalc[i]);
+		if (stat == -1)
+			mtd->ecc_stats.failed++;
+		else
+			mtd->ecc_stats.corrected += stat;
+	}
+	return 0;
+}
+
 /**
  * nand_read_page_hwecc - [REPLACABLE] hardware ecc based page read function
  * @mtd:	mtd info structure
@@ -994,6 +1075,8 @@ static int nand_do_read_ops(struct mtd_info *mtd, loff_t from,
 			/* Now read the page into the buffer */
 			if (unlikely(ops->mode == MTD_OOB_RAW))
 				ret = chip->ecc.read_page_raw(mtd, chip, bufpoi);
+			else if (!aligned && NAND_SUBPAGE_READ(chip) && !oob)
+				ret = chip->ecc.read_subpage(mtd, chip, col, bytes, bufpoi);
 			else
 				ret = chip->ecc.read_page(mtd, chip, bufpoi);
 			if (ret < 0)
@@ -1001,7 +1084,8 @@ static int nand_do_read_ops(struct mtd_info *mtd, loff_t from,
 
 			/* Transfer not aligned data */
 			if (!aligned) {
-				chip->pagebuf = realpage;
+				if (!NAND_SUBPAGE_READ(chip) && !oob)
+					chip->pagebuf = realpage;
 				memcpy(buf, chip->buffers->databuf + col, bytes);
 			}
 
@@ -2521,6 +2605,7 @@ int nand_scan_tail(struct mtd_info *mtd)
 		chip->ecc.calculate = nand_calculate_ecc;
 		chip->ecc.correct = nand_correct_data;
 		chip->ecc.read_page = nand_read_page_swecc;
+		chip->ecc.read_subpage = nand_read_subpage;
 		chip->ecc.write_page = nand_write_page_swecc;
 		chip->ecc.read_oob = nand_read_oob_std;
 		chip->ecc.write_oob = nand_write_oob_std;
diff --git a/include/linux/mtd/nand.h b/include/linux/mtd/nand.h
index 863e22a0ddb5..83f678702dff 100644
--- a/include/linux/mtd/nand.h
+++ b/include/linux/mtd/nand.h
@@ -177,6 +177,7 @@ typedef enum {
 #define NAND_MUST_PAD(chip) (!(chip->options & NAND_NO_PADDING))
 #define NAND_HAS_CACHEPROG(chip) ((chip->options & NAND_CACHEPRG))
 #define NAND_HAS_COPYBACK(chip) ((chip->options & NAND_COPYBACK))
+#define NAND_SUBPAGE_READ(chip) ((chip->ecc.mode == NAND_ECC_SOFT))
 
 /* Mask to zero out the chip options, which come from the id table */
 #define NAND_CHIPOPTIONS_MSK	(0x0000ffff & ~NAND_NO_AUTOINCR)
@@ -274,6 +275,10 @@ struct nand_ecc_ctrl {
 	int			(*read_page)(struct mtd_info *mtd,
 					     struct nand_chip *chip,
 					     uint8_t *buf);
+	int			(*read_subpage)(struct mtd_info *mtd,
+					     struct nand_chip *chip,
+					     uint32_t offs, uint32_t len,
+					     uint8_t *buf);
 	void			(*write_page)(struct mtd_info *mtd,
 					      struct nand_chip *chip,
 					      const uint8_t *buf);

From 29b309e52d3d51ef8a15bd15590903cf272beb93 Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Fri, 25 Jul 2008 09:19:36 -0700
Subject: [PATCH 0667/1171] Undo duplicate "m68k: drivers/input/serio/hp_sdc.c
 needs <linux/semaphore.h>"

Both commits 0f17e4c796e89d1f69f13b653aba60e6ccfb8ae0 ("Add missing
semaphore.h includes") and 4933d07531711e399d8d578036aa9fc1be2f9b20
("m68k: drivers/input/serio/hp_sdc.c needs <linux/semaphore.h>") added a

We only really need one ;)

Reported-by: Huang Weiyi <weiyi.huang@gmail.com>
Requested-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/input/serio/hp_sdc.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/input/serio/hp_sdc.c b/drivers/input/serio/hp_sdc.c
index aad664d5259f..0d395979b2d1 100644
--- a/drivers/input/serio/hp_sdc.c
+++ b/drivers/input/serio/hp_sdc.c
@@ -70,7 +70,6 @@
 #include <linux/semaphore.h>
 #include <linux/slab.h>
 #include <linux/hil.h>
-#include <linux/semaphore.h>
 #include <asm/io.h>
 #include <asm/system.h>
 

From 43de804df8d6002059bf4af4522fa9273a19b8aa Mon Sep 17 00:00:00 2001
From: Huang Weiyi <weiyi.huang@gmail.com>
Date: Fri, 25 Jul 2008 23:30:15 +0800
Subject: [PATCH 0668/1171] char/xilinx_hwicap/xilinx_hwicap.c: Removed
 duplicated include

Removed duplicated include file <linux/version.h> in
char/xilinx_hwicap/xilinx_hwicap.c.

Signed-off-by: Huang Weiyi <weiyi.huang@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/xilinx_hwicap/xilinx_hwicap.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/char/xilinx_hwicap/xilinx_hwicap.c b/drivers/char/xilinx_hwicap/xilinx_hwicap.c
index 51966ccf4ea3..8bfee5fb7223 100644
--- a/drivers/char/xilinx_hwicap/xilinx_hwicap.c
+++ b/drivers/char/xilinx_hwicap/xilinx_hwicap.c
@@ -87,7 +87,6 @@
 #include <linux/mutex.h>
 #include <linux/smp_lock.h>
 #include <linux/sysctl.h>
-#include <linux/version.h>
 #include <linux/fs.h>
 #include <linux/cdev.h>
 #include <linux/platform_device.h>

From 3d6f4a20cc287a8980c6186624834cf10a70752b Mon Sep 17 00:00:00 2001
From: David Miller <davem@davemloft.net>
Date: Thu, 24 Jul 2008 23:38:31 -0700
Subject: [PATCH 0669/1171] endian: Always evaluate arguments.

Changeset 7fa897b91a3ea0f16c2873b869d7a0eef05acff4 ("ide: trivial sparse
annotations") created an IDE bootup regression on big-endian systems.

In drivers/ide/ide-iops.c, function ide_fixstring() we now have the
loop:

		for (p = end ; p != s;)
			be16_to_cpus((u16 *)(p -= 2));

which will never terminate on big-endian because in such
a configuration be16_to_cpus() evaluates to "do { } while (0)"

Therefore, always evaluate the arguments to nop endian transformation
operations.

Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/byteorder/big_endian.h    | 12 ++++++------
 include/linux/byteorder/little_endian.h | 12 ++++++------
 2 files changed, 12 insertions(+), 12 deletions(-)

diff --git a/include/linux/byteorder/big_endian.h b/include/linux/byteorder/big_endian.h
index 961ed4b48d8e..44f95b92393b 100644
--- a/include/linux/byteorder/big_endian.h
+++ b/include/linux/byteorder/big_endian.h
@@ -94,12 +94,12 @@ static inline __u16 __be16_to_cpup(const __be16 *p)
 #define __le32_to_cpus(x) __swab32s((x))
 #define __cpu_to_le16s(x) __swab16s((x))
 #define __le16_to_cpus(x) __swab16s((x))
-#define __cpu_to_be64s(x) do {} while (0)
-#define __be64_to_cpus(x) do {} while (0)
-#define __cpu_to_be32s(x) do {} while (0)
-#define __be32_to_cpus(x) do {} while (0)
-#define __cpu_to_be16s(x) do {} while (0)
-#define __be16_to_cpus(x) do {} while (0)
+#define __cpu_to_be64s(x) do { (void)(x); } while (0)
+#define __be64_to_cpus(x) do { (void)(x); } while (0)
+#define __cpu_to_be32s(x) do { (void)(x); } while (0)
+#define __be32_to_cpus(x) do { (void)(x); } while (0)
+#define __cpu_to_be16s(x) do { (void)(x); } while (0)
+#define __be16_to_cpus(x) do { (void)(x); } while (0)
 
 #ifdef __KERNEL__
 #include <linux/byteorder/generic.h>
diff --git a/include/linux/byteorder/little_endian.h b/include/linux/byteorder/little_endian.h
index 05dc7c35b3b2..4cc170a31762 100644
--- a/include/linux/byteorder/little_endian.h
+++ b/include/linux/byteorder/little_endian.h
@@ -88,12 +88,12 @@ static inline __u16 __be16_to_cpup(const __be16 *p)
 {
 	return __swab16p((__u16 *)p);
 }
-#define __cpu_to_le64s(x) do {} while (0)
-#define __le64_to_cpus(x) do {} while (0)
-#define __cpu_to_le32s(x) do {} while (0)
-#define __le32_to_cpus(x) do {} while (0)
-#define __cpu_to_le16s(x) do {} while (0)
-#define __le16_to_cpus(x) do {} while (0)
+#define __cpu_to_le64s(x) do { (void)(x); } while (0)
+#define __le64_to_cpus(x) do { (void)(x); } while (0)
+#define __cpu_to_le32s(x) do { (void)(x); } while (0)
+#define __le32_to_cpus(x) do { (void)(x); } while (0)
+#define __cpu_to_le16s(x) do { (void)(x); } while (0)
+#define __le16_to_cpus(x) do { (void)(x); } while (0)
 #define __cpu_to_be64s(x) __swab64s((x))
 #define __be64_to_cpus(x) __swab64s((x))
 #define __cpu_to_be32s(x) __swab32s((x))

From 3e4d0cab61c88a9ae3e61151a857960397e26403 Mon Sep 17 00:00:00 2001
From: Tony Luck <tony.luck@intel.com>
Date: Fri, 25 Jul 2008 10:10:28 -0700
Subject: [PATCH 0670/1171] [IA64] Wire up new system calls

Six new system calls: signalfd4, eventfd2, epoll_create1,
dup3, pipe2 and inotify_init1.

Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/ia64/kernel/entry.S  | 6 ++++++
 include/asm-ia64/unistd.h | 8 +++++++-
 2 files changed, 13 insertions(+), 1 deletion(-)

diff --git a/arch/ia64/kernel/entry.S b/arch/ia64/kernel/entry.S
index 56ab156c48ae..0dd6c1419d8d 100644
--- a/arch/ia64/kernel/entry.S
+++ b/arch/ia64/kernel/entry.S
@@ -1691,6 +1691,12 @@ sys_call_table:
 	data8 sys_timerfd_create		// 1310
 	data8 sys_timerfd_settime
 	data8 sys_timerfd_gettime
+	data8 sys_signalfd4
+	data8 sys_eventfd2
+	data8 sys_epoll_create1			// 1315
+	data8 sys_dup3
+	data8 sys_pipe2
+	data8 sys_inotify_init1
 
 	.org sys_call_table + 8*NR_syscalls	// guard against failures to increase NR_syscalls
 #endif /* __IA64_ASM_PARAVIRTUALIZED_NATIVE */
diff --git a/include/asm-ia64/unistd.h b/include/asm-ia64/unistd.h
index e60314716122..d535833aab5e 100644
--- a/include/asm-ia64/unistd.h
+++ b/include/asm-ia64/unistd.h
@@ -302,11 +302,17 @@
 #define __NR_timerfd_create		1310
 #define __NR_timerfd_settime		1311
 #define __NR_timerfd_gettime		1312
+#define __NR_signalfd4			1313
+#define __NR_eventfd2			1314
+#define __NR_epoll_create1		1315
+#define __NR_dup3			1316
+#define __NR_pipe2			1317
+#define __NR_inotify_init1		1318
 
 #ifdef __KERNEL__
 
 
-#define NR_syscalls			289 /* length of syscall table */
+#define NR_syscalls			295 /* length of syscall table */
 
 /*
  * The following defines stop scripts/checksyscalls.sh from complaining about

From 25c94d010a8ae8605dc4d5453e0c82fa97da5d12 Mon Sep 17 00:00:00 2001
From: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Date: Fri, 25 Jul 2008 10:30:06 -0700
Subject: [PATCH 0671/1171] mlx4_core: Add VLAN tag field to WQE control
 segment struct

Add fields for VLAN tag and insert VLAN tag flag to the control
section struct.  These fields will be used for sending ethernet
packets.

Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 include/linux/mlx4/qp.h | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/include/linux/mlx4/qp.h b/include/linux/mlx4/qp.h
index e27082cd650e..bf8f11982dae 100644
--- a/include/linux/mlx4/qp.h
+++ b/include/linux/mlx4/qp.h
@@ -164,11 +164,13 @@ enum {
 	MLX4_WQE_CTRL_SOLICITED		= 1 << 1,
 	MLX4_WQE_CTRL_IP_CSUM		= 1 << 4,
 	MLX4_WQE_CTRL_TCP_UDP_CSUM	= 1 << 5,
+	MLX4_WQE_CTRL_INS_VLAN		= 1 << 6,
 };
 
 struct mlx4_wqe_ctrl_seg {
 	__be32			owner_opcode;
-	u8			reserved2[3];
+	__be16			vlan_tag;
+	u8			ins_vlan;
 	u8			fence_size;
 	/*
 	 * High 24 bits are SRC remote buffer; low 8 bits are flags:

From 51a379d0c8f7a6db7c9e3c9c770d90a6d2d1ef9b Mon Sep 17 00:00:00 2001
From: Jack Morgenstein <jackm@dev.mellanox.co.il>
Date: Fri, 25 Jul 2008 10:32:52 -0700
Subject: [PATCH 0672/1171] mlx4: Update/add Mellanox Technologies copyright
 lines to mlx4 driver files

Update existing Mellanox copyright lines to 2008, and add such lines
to files where they are missing.

Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <rolandd@cisco.com>
---
 drivers/infiniband/hw/mlx4/cq.c      | 1 +
 drivers/infiniband/hw/mlx4/main.c    | 1 +
 drivers/infiniband/hw/mlx4/mlx4_ib.h | 1 +
 drivers/infiniband/hw/mlx4/mr.c      | 1 +
 drivers/infiniband/hw/mlx4/qp.c      | 1 +
 drivers/infiniband/hw/mlx4/srq.c     | 1 +
 drivers/infiniband/hw/mlx4/user.h    | 1 +
 drivers/net/mlx4/alloc.c             | 1 +
 drivers/net/mlx4/catas.c             | 1 +
 drivers/net/mlx4/cmd.c               | 2 +-
 drivers/net/mlx4/cq.c                | 2 +-
 drivers/net/mlx4/eq.c                | 2 +-
 drivers/net/mlx4/fw.c                | 2 +-
 drivers/net/mlx4/fw.h                | 2 +-
 drivers/net/mlx4/icm.c               | 2 +-
 drivers/net/mlx4/icm.h               | 2 +-
 drivers/net/mlx4/intf.c              | 1 +
 drivers/net/mlx4/main.c              | 2 +-
 drivers/net/mlx4/mcg.c               | 1 +
 drivers/net/mlx4/mlx4.h              | 2 +-
 drivers/net/mlx4/mr.c                | 2 +-
 drivers/net/mlx4/qp.c                | 2 +-
 drivers/net/mlx4/reset.c             | 1 +
 drivers/net/mlx4/srq.c               | 1 +
 24 files changed, 24 insertions(+), 11 deletions(-)

diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c
index 0b191a4842ce..a1464574bfdd 100644
--- a/drivers/infiniband/hw/mlx4/cq.c
+++ b/drivers/infiniband/hw/mlx4/cq.c
@@ -1,5 +1,6 @@
 /*
  * Copyright (c) 2007 Cisco Systems, Inc. All rights reserved.
+ * Copyright (c) 2007, 2008 Mellanox Technologies. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index 38d6907ab521..a3c2851c0545 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -1,5 +1,6 @@
 /*
  * Copyright (c) 2006, 2007 Cisco Systems, Inc. All rights reserved.
+ * Copyright (c) 2007, 2008 Mellanox Technologies. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
diff --git a/drivers/infiniband/hw/mlx4/mlx4_ib.h b/drivers/infiniband/hw/mlx4/mlx4_ib.h
index d26a91317d4d..6e2b0dc21b61 100644
--- a/drivers/infiniband/hw/mlx4/mlx4_ib.h
+++ b/drivers/infiniband/hw/mlx4/mlx4_ib.h
@@ -1,5 +1,6 @@
 /*
  * Copyright (c) 2006, 2007 Cisco Systems.  All rights reserved.
+ * Copyright (c) 2007, 2008 Mellanox Technologies. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
diff --git a/drivers/infiniband/hw/mlx4/mr.c b/drivers/infiniband/hw/mlx4/mr.c
index db2086faa4ed..a4cdb465cd1d 100644
--- a/drivers/infiniband/hw/mlx4/mr.c
+++ b/drivers/infiniband/hw/mlx4/mr.c
@@ -1,5 +1,6 @@
 /*
  * Copyright (c) 2007 Cisco Systems, Inc. All rights reserved.
+ * Copyright (c) 2007, 2008 Mellanox Technologies. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
index 02a99bc4442e..f7bc7dd8578a 100644
--- a/drivers/infiniband/hw/mlx4/qp.c
+++ b/drivers/infiniband/hw/mlx4/qp.c
@@ -1,5 +1,6 @@
 /*
  * Copyright (c) 2007 Cisco Systems, Inc. All rights reserved.
+ * Copyright (c) 2007, 2008 Mellanox Technologies. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
diff --git a/drivers/infiniband/hw/mlx4/srq.c b/drivers/infiniband/hw/mlx4/srq.c
index 12d6bc6f8007..d42565258fb7 100644
--- a/drivers/infiniband/hw/mlx4/srq.c
+++ b/drivers/infiniband/hw/mlx4/srq.c
@@ -1,5 +1,6 @@
 /*
  * Copyright (c) 2007 Cisco Systems, Inc. All rights reserved.
+ * Copyright (c) 2007, 2008 Mellanox Technologies. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
diff --git a/drivers/infiniband/hw/mlx4/user.h b/drivers/infiniband/hw/mlx4/user.h
index e2d11be4525c..13beedeeef9f 100644
--- a/drivers/infiniband/hw/mlx4/user.h
+++ b/drivers/infiniband/hw/mlx4/user.h
@@ -1,5 +1,6 @@
 /*
  * Copyright (c) 2007 Cisco Systems, Inc. All rights reserved.
+ * Copyright (c) 2007, 2008 Mellanox Technologies. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
diff --git a/drivers/net/mlx4/alloc.c b/drivers/net/mlx4/alloc.c
index f9d6b4dca180..096bca54bcf7 100644
--- a/drivers/net/mlx4/alloc.c
+++ b/drivers/net/mlx4/alloc.c
@@ -1,5 +1,6 @@
 /*
  * Copyright (c) 2006, 2007 Cisco Systems, Inc.  All rights reserved.
+ * Copyright (c) 2007, 2008 Mellanox Technologies. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
diff --git a/drivers/net/mlx4/catas.c b/drivers/net/mlx4/catas.c
index aa9528779044..f094ee00c416 100644
--- a/drivers/net/mlx4/catas.c
+++ b/drivers/net/mlx4/catas.c
@@ -1,5 +1,6 @@
 /*
  * Copyright (c) 2007 Cisco Systems, Inc. All rights reserved.
+ * Copyright (c) 2007, 2008 Mellanox Technologies. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
diff --git a/drivers/net/mlx4/cmd.c b/drivers/net/mlx4/cmd.c
index 04d5bc69a6f8..2845a0560b84 100644
--- a/drivers/net/mlx4/cmd.c
+++ b/drivers/net/mlx4/cmd.c
@@ -1,6 +1,6 @@
 /*
  * Copyright (c) 2004, 2005 Topspin Communications.  All rights reserved.
- * Copyright (c) 2005 Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2005, 2006, 2007, 2008 Mellanox Technologies. All rights reserved.
  * Copyright (c) 2005, 2006, 2007 Cisco Systems, Inc.  All rights reserved.
  *
  * This software is available to you under a choice of one of two
diff --git a/drivers/net/mlx4/cq.c b/drivers/net/mlx4/cq.c
index 95e87a2f8896..9bb50e3f8974 100644
--- a/drivers/net/mlx4/cq.c
+++ b/drivers/net/mlx4/cq.c
@@ -2,7 +2,7 @@
  * Copyright (c) 2004, 2005 Topspin Communications.  All rights reserved.
  * Copyright (c) 2005 Sun Microsystems, Inc. All rights reserved.
  * Copyright (c) 2005, 2006, 2007 Cisco Systems, Inc. All rights reserved.
- * Copyright (c) 2005 Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2005, 2006, 2007, 2008 Mellanox Technologies. All rights reserved.
  * Copyright (c) 2004 Voltaire, Inc. All rights reserved.
  *
  * This software is available to you under a choice of one of two
diff --git a/drivers/net/mlx4/eq.c b/drivers/net/mlx4/eq.c
index ea3a09aaa844..bd34bf08830d 100644
--- a/drivers/net/mlx4/eq.c
+++ b/drivers/net/mlx4/eq.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2005 Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2005, 2006, 2007, 2008 Mellanox Technologies. All rights reserved.
  * Copyright (c) 2005, 2006, 2007 Cisco Systems, Inc. All rights reserved.
  *
  * This software is available to you under a choice of one of two
diff --git a/drivers/net/mlx4/fw.c b/drivers/net/mlx4/fw.c
index 57278224ba1e..7e32955da982 100644
--- a/drivers/net/mlx4/fw.c
+++ b/drivers/net/mlx4/fw.c
@@ -1,6 +1,6 @@
 /*
  * Copyright (c) 2004, 2005 Topspin Communications.  All rights reserved.
- * Copyright (c) 2005 Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2005, 2006, 2007, 2008 Mellanox Technologies. All rights reserved.
  * Copyright (c) 2005, 2006, 2007 Cisco Systems, Inc.  All rights reserved.
  *
  * This software is available to you under a choice of one of two
diff --git a/drivers/net/mlx4/fw.h b/drivers/net/mlx4/fw.h
index fbf0e22be122..decbb5c2ad41 100644
--- a/drivers/net/mlx4/fw.h
+++ b/drivers/net/mlx4/fw.h
@@ -1,6 +1,6 @@
 /*
  * Copyright (c) 2004, 2005 Topspin Communications.  All rights reserved.
- * Copyright (c) 2005 Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2005, 2006, 2007, 2008 Mellanox Technologies. All rights reserved.
  * Copyright (c) 2006, 2007 Cisco Systems.  All rights reserved.
  *
  * This software is available to you under a choice of one of two
diff --git a/drivers/net/mlx4/icm.c b/drivers/net/mlx4/icm.c
index 2a5bef6388fe..baf4bf66062c 100644
--- a/drivers/net/mlx4/icm.c
+++ b/drivers/net/mlx4/icm.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2005 Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2005, 2006, 2007, 2008 Mellanox Technologies. All rights reserved.
  * Copyright (c) 2006, 2007 Cisco Systems, Inc.  All rights reserved.
  *
  * This software is available to you under a choice of one of two
diff --git a/drivers/net/mlx4/icm.h b/drivers/net/mlx4/icm.h
index 6c44edf35847..ab56a2f89b65 100644
--- a/drivers/net/mlx4/icm.h
+++ b/drivers/net/mlx4/icm.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2005 Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2005, 2006, 2007, 2008 Mellanox Technologies. All rights reserved.
  * Copyright (c) 2006, 2007 Cisco Systems, Inc.  All rights reserved.
  *
  * This software is available to you under a choice of one of two
diff --git a/drivers/net/mlx4/intf.c b/drivers/net/mlx4/intf.c
index 4a6c4d526f1b..0e7eb1038f9f 100644
--- a/drivers/net/mlx4/intf.c
+++ b/drivers/net/mlx4/intf.c
@@ -1,5 +1,6 @@
 /*
  * Copyright (c) 2006, 2007 Cisco Systems, Inc. All rights reserved.
+ * Copyright (c) 2007, 2008 Mellanox Technologies. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
diff --git a/drivers/net/mlx4/main.c b/drivers/net/mlx4/main.c
index 8e1d24cda1b0..1252a919de2e 100644
--- a/drivers/net/mlx4/main.c
+++ b/drivers/net/mlx4/main.c
@@ -1,7 +1,7 @@
 /*
  * Copyright (c) 2004, 2005 Topspin Communications.  All rights reserved.
  * Copyright (c) 2005 Sun Microsystems, Inc. All rights reserved.
- * Copyright (c) 2005 Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2005, 2006, 2007, 2008 Mellanox Technologies. All rights reserved.
  * Copyright (c) 2006, 2007 Cisco Systems, Inc. All rights reserved.
  *
  * This software is available to you under a choice of one of two
diff --git a/drivers/net/mlx4/mcg.c b/drivers/net/mlx4/mcg.c
index b4b57870ddfd..c83f88ce0736 100644
--- a/drivers/net/mlx4/mcg.c
+++ b/drivers/net/mlx4/mcg.c
@@ -1,5 +1,6 @@
 /*
  * Copyright (c) 2006, 2007 Cisco Systems, Inc.  All rights reserved.
+ * Copyright (c) 2007, 2008 Mellanox Technologies. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
diff --git a/drivers/net/mlx4/mlx4.h b/drivers/net/mlx4/mlx4.h
index 78038499cff5..5337e3ac3e78 100644
--- a/drivers/net/mlx4/mlx4.h
+++ b/drivers/net/mlx4/mlx4.h
@@ -2,7 +2,7 @@
  * Copyright (c) 2004, 2005 Topspin Communications.  All rights reserved.
  * Copyright (c) 2005 Sun Microsystems, Inc. All rights reserved.
  * Copyright (c) 2005, 2006, 2007 Cisco Systems.  All rights reserved.
- * Copyright (c) 2005 Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2005, 2006, 2007, 2008 Mellanox Technologies. All rights reserved.
  * Copyright (c) 2004 Voltaire, Inc. All rights reserved.
  *
  * This software is available to you under a choice of one of two
diff --git a/drivers/net/mlx4/mr.c b/drivers/net/mlx4/mr.c
index a3c04c5f12c2..62071d9c4a55 100644
--- a/drivers/net/mlx4/mr.c
+++ b/drivers/net/mlx4/mr.c
@@ -1,6 +1,6 @@
 /*
  * Copyright (c) 2004 Topspin Communications.  All rights reserved.
- * Copyright (c) 2005 Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2005, 2006, 2007, 2008 Mellanox Technologies. All rights reserved.
  * Copyright (c) 2006, 2007 Cisco Systems, Inc.  All rights reserved.
  *
  * This software is available to you under a choice of one of two
diff --git a/drivers/net/mlx4/qp.c b/drivers/net/mlx4/qp.c
index ee5484c44a18..c49a86044bf7 100644
--- a/drivers/net/mlx4/qp.c
+++ b/drivers/net/mlx4/qp.c
@@ -1,7 +1,7 @@
 /*
  * Copyright (c) 2004 Topspin Communications.  All rights reserved.
  * Copyright (c) 2005, 2006, 2007 Cisco Systems, Inc. All rights reserved.
- * Copyright (c) 2005 Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2005, 2006, 2007, 2008 Mellanox Technologies. All rights reserved.
  * Copyright (c) 2004 Voltaire, Inc. All rights reserved.
  *
  * This software is available to you under a choice of one of two
diff --git a/drivers/net/mlx4/reset.c b/drivers/net/mlx4/reset.c
index e199715fabd0..3951b884c0fb 100644
--- a/drivers/net/mlx4/reset.c
+++ b/drivers/net/mlx4/reset.c
@@ -1,5 +1,6 @@
 /*
  * Copyright (c) 2006, 2007 Cisco Systems, Inc.  All rights reserved.
+ * Copyright (c) 2007, 2008 Mellanox Technologies. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU
diff --git a/drivers/net/mlx4/srq.c b/drivers/net/mlx4/srq.c
index d23f46d692ef..533eb6db24b3 100644
--- a/drivers/net/mlx4/srq.c
+++ b/drivers/net/mlx4/srq.c
@@ -1,5 +1,6 @@
 /*
  * Copyright (c) 2006, 2007 Cisco Systems, Inc. All rights reserved.
+ * Copyright (c) 2007, 2008 Mellanox Technologies. All rights reserved.
  *
  * This software is available to you under a choice of one of two
  * licenses.  You may choose to be licensed under the terms of the GNU

From c82dd5321cf779f1f536ef26b383cbe8c9de7f10 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@linux-foundation.org>
Date: Fri, 25 Jul 2008 01:45:22 -0700
Subject: [PATCH 0673/1171] mfd: don't use memzero

For it doesn't exist on i386.

Cc: Ian Molton <spyro@f2s.com>
Cc: Dmitry Baryshkov <dbaryshkov@gmail.com>
Cc: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/mfd/mfd-core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mfd/mfd-core.c b/drivers/mfd/mfd-core.c
index d7d88ce053a6..0454be4266c1 100644
--- a/drivers/mfd/mfd-core.c
+++ b/drivers/mfd/mfd-core.c
@@ -36,7 +36,7 @@ static int mfd_add_device(struct platform_device *parent,
 	if (ret)
 		goto fail_device;
 
-	memzero(res, sizeof(res));
+	memset(res, 0, sizeof(res));
 	for (r = 0; r < cell->num_resources; r++) {
 		res[r].name = cell->resources[r].name;
 		res[r].flags = cell->resources[r].flags;

From 5df439ef06d4173357711a04740aa8bfcf50d621 Mon Sep 17 00:00:00 2001
From: Wang Chen <wangchen@cn.fujitsu.com>
Date: Fri, 25 Jul 2008 01:45:23 -0700
Subject: [PATCH 0674/1171] flag parameters: fix compile error of
 sys_epoll_create1

GEN     .version
  CHK     include/linux/compile.h
  UPD     include/linux/compile.h
  CC      init/version.o
  LD      init/built-in.o
  LD      vmlinux
arch/x86/kernel/built-in.o: In function `sys_call_table':
(.rodata+0x8a4): undefined reference to `sys_epoll_create1'
make: *** [vmlinux] Error 1

Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Cc: Ulrich Drepper <drepper@redhat.com>
Cc: Davide Libenzi <davidel@xmailserver.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/sys_ni.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c
index bd66ac5406f3..55eca1594da9 100644
--- a/kernel/sys_ni.c
+++ b/kernel/sys_ni.c
@@ -57,6 +57,7 @@ cond_syscall(compat_sys_set_robust_list);
 cond_syscall(sys_get_robust_list);
 cond_syscall(compat_sys_get_robust_list);
 cond_syscall(sys_epoll_create);
+cond_syscall(sys_epoll_create1);
 cond_syscall(sys_epoll_ctl);
 cond_syscall(sys_epoll_wait);
 cond_syscall(sys_epoll_pwait);

From e0deaff470900a4c3222ca7139f6c9639e26a2f5 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@linux-foundation.org>
Date: Fri, 25 Jul 2008 01:45:24 -0700
Subject: [PATCH 0675/1171] split the typecheck macros out of
 include/linux/kernel.h

Needed to fix up a recursive include snafu in
locking-add-typecheck-on-irqsave-and-friends-for-correct-flags.patch

Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/kernel.h    | 21 +--------------------
 include/linux/typecheck.h | 24 ++++++++++++++++++++++++
 2 files changed, 25 insertions(+), 20 deletions(-)
 create mode 100644 include/linux/typecheck.h

diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index f9cd7a513f9c..5c4b1251e110 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -14,6 +14,7 @@
 #include <linux/compiler.h>
 #include <linux/bitops.h>
 #include <linux/log2.h>
+#include <linux/typecheck.h>
 #include <asm/byteorder.h>
 #include <asm/bug.h>
 
@@ -441,26 +442,6 @@ static inline char *pack_hex_byte(char *buf, u8 byte)
 	const typeof( ((type *)0)->member ) *__mptr = (ptr);	\
 	(type *)( (char *)__mptr - offsetof(type,member) );})
 
-/*
- * Check at compile time that something is of a particular type.
- * Always evaluates to 1 so you may use it easily in comparisons.
- */
-#define typecheck(type,x) \
-({	type __dummy; \
-	typeof(x) __dummy2; \
-	(void)(&__dummy == &__dummy2); \
-	1; \
-})
-
-/*
- * Check at compile time that 'function' is a certain type, or is a pointer
- * to that type (needs to use typedef for the function type.)
- */
-#define typecheck_fn(type,function) \
-({	typeof(type) __tmp = function; \
-	(void)__tmp; \
-})
-
 struct sysinfo;
 extern int do_sysinfo(struct sysinfo *info);
 
diff --git a/include/linux/typecheck.h b/include/linux/typecheck.h
new file mode 100644
index 000000000000..eb5b74a575be
--- /dev/null
+++ b/include/linux/typecheck.h
@@ -0,0 +1,24 @@
+#ifndef TYPECHECK_H_INCLUDED
+#define TYPECHECK_H_INCLUDED
+
+/*
+ * Check at compile time that something is of a particular type.
+ * Always evaluates to 1 so you may use it easily in comparisons.
+ */
+#define typecheck(type,x) \
+({	type __dummy; \
+	typeof(x) __dummy2; \
+	(void)(&__dummy == &__dummy2); \
+	1; \
+})
+
+/*
+ * Check at compile time that 'function' is a certain type, or is a pointer
+ * to that type (needs to use typedef for the function type.)
+ */
+#define typecheck_fn(type,function) \
+({	typeof(type) __tmp = function; \
+	(void)__tmp; \
+})
+
+#endif		/* TYPECHECK_H_INCLUDED */

From 3f307891ce0e7b0438c432af1aacd656a092ff45 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <rostedt@goodmis.org>
Date: Fri, 25 Jul 2008 01:45:25 -0700
Subject: [PATCH 0676/1171] locking: add typecheck on irqsave and friends for
 correct flags

There haave been several areas in the kernel where an int has been used for
flags in local_irq_save() and friends instead of a long.  This can cause some
hard to debug problems on some architectures.

This patch adds a typecheck inside the irqsave and restore functions to flag
these cases.

[akpm@linux-foundation.org: coding-style fixes]
[akpm@linux-foundation.org: build fix]
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/irqflags.h | 54 +++++++++++++++++++++---------
 include/linux/spinlock.h | 72 +++++++++++++++++++++++++++++++---------
 2 files changed, 95 insertions(+), 31 deletions(-)

diff --git a/include/linux/irqflags.h b/include/linux/irqflags.h
index 2b1c2e58566e..74bde13224c9 100644
--- a/include/linux/irqflags.h
+++ b/include/linux/irqflags.h
@@ -11,6 +11,8 @@
 #ifndef _LINUX_TRACE_IRQFLAGS_H
 #define _LINUX_TRACE_IRQFLAGS_H
 
+#include <linux/typecheck.h>
+
 #ifdef CONFIG_TRACE_IRQFLAGS
   extern void trace_softirqs_on(unsigned long ip);
   extern void trace_softirqs_off(unsigned long ip);
@@ -58,18 +60,24 @@
 	do { trace_hardirqs_on(); raw_local_irq_enable(); } while (0)
 #define local_irq_disable() \
 	do { raw_local_irq_disable(); trace_hardirqs_off(); } while (0)
-#define local_irq_save(flags) \
-	do { raw_local_irq_save(flags); trace_hardirqs_off(); } while (0)
+#define local_irq_save(flags)				\
+	do {						\
+		typecheck(unsigned long, flags);	\
+		raw_local_irq_save(flags);		\
+		trace_hardirqs_off();			\
+	} while (0)
 
-#define local_irq_restore(flags)				\
-	do {							\
-		if (raw_irqs_disabled_flags(flags)) {		\
-			raw_local_irq_restore(flags);		\
-			trace_hardirqs_off();			\
-		} else {					\
-			trace_hardirqs_on();			\
-			raw_local_irq_restore(flags);		\
-		}						\
+
+#define local_irq_restore(flags)			\
+	do {						\
+		typecheck(unsigned long, flags);	\
+		if (raw_irqs_disabled_flags(flags)) {	\
+			raw_local_irq_restore(flags);	\
+			trace_hardirqs_off();		\
+		} else {				\
+			trace_hardirqs_on();		\
+			raw_local_irq_restore(flags);	\
+		}					\
 	} while (0)
 #else /* !CONFIG_TRACE_IRQFLAGS_SUPPORT */
 /*
@@ -78,8 +86,16 @@
  */
 # define raw_local_irq_disable()	local_irq_disable()
 # define raw_local_irq_enable()		local_irq_enable()
-# define raw_local_irq_save(flags)	local_irq_save(flags)
-# define raw_local_irq_restore(flags)	local_irq_restore(flags)
+# define raw_local_irq_save(flags)			\
+	do {						\
+		typecheck(unsigned long, flags);	\
+		local_irq_save(flags);			\
+	} while (0)
+# define raw_local_irq_restore(flags)			\
+	do {						\
+		typecheck(unsigned long, flags);	\
+		local_irq_restore(flags);		\
+	} while (0)
 #endif /* CONFIG_TRACE_IRQFLAGS_SUPPORT */
 
 #ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT
@@ -89,7 +105,11 @@
 		raw_safe_halt();				\
 	} while (0)
 
-#define local_save_flags(flags)		raw_local_save_flags(flags)
+#define local_save_flags(flags)				\
+	do {						\
+		typecheck(unsigned long, flags);	\
+		raw_local_save_flags(flags);		\
+	} while (0)
 
 #define irqs_disabled()						\
 ({								\
@@ -99,7 +119,11 @@
 	raw_irqs_disabled_flags(_flags);			\
 })
 
-#define irqs_disabled_flags(flags)	raw_irqs_disabled_flags(flags)
+#define irqs_disabled_flags(flags)		\
+({						\
+	typecheck(unsigned long, flags);	\
+	raw_irqs_disabled_flags(flags);		\
+})
 #endif		/* CONFIG_X86 */
 
 #endif
diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h
index d311a090fae7..61e5610ad165 100644
--- a/include/linux/spinlock.h
+++ b/include/linux/spinlock.h
@@ -46,6 +46,7 @@
  *  linux/spinlock.h:     builds the final spin_*() APIs.
  */
 
+#include <linux/typecheck.h>
 #include <linux/preempt.h>
 #include <linux/linkage.h>
 #include <linux/compiler.h>
@@ -191,23 +192,53 @@ do {								\
 
 #if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK)
 
-#define spin_lock_irqsave(lock, flags)	flags = _spin_lock_irqsave(lock)
-#define read_lock_irqsave(lock, flags)	flags = _read_lock_irqsave(lock)
-#define write_lock_irqsave(lock, flags)	flags = _write_lock_irqsave(lock)
+#define spin_lock_irqsave(lock, flags)			\
+	do {						\
+		typecheck(unsigned long, flags);	\
+		flags = _spin_lock_irqsave(lock);	\
+	} while (0)
+#define read_lock_irqsave(lock, flags)			\
+	do {						\
+		typecheck(unsigned long, flags);	\
+		flags = _read_lock_irqsave(lock);	\
+	} while (0)
+#define write_lock_irqsave(lock, flags)			\
+	do {						\
+		typecheck(unsigned long, flags);	\
+		flags = _write_lock_irqsave(lock);	\
+	} while (0)
 
 #ifdef CONFIG_DEBUG_LOCK_ALLOC
-#define spin_lock_irqsave_nested(lock, flags, subclass) \
-	flags = _spin_lock_irqsave_nested(lock, subclass)
+#define spin_lock_irqsave_nested(lock, flags, subclass)			\
+	do {								\
+		typecheck(unsigned long, flags);			\
+		flags = _spin_lock_irqsave_nested(lock, subclass);	\
+	} while (0)
 #else
-#define spin_lock_irqsave_nested(lock, flags, subclass) \
-	flags = _spin_lock_irqsave(lock)
+#define spin_lock_irqsave_nested(lock, flags, subclass)			\
+	do {								\
+		typecheck(unsigned long, flags);			\
+		flags = _spin_lock_irqsave(lock);			\
+	} while (0)
 #endif
 
 #else
 
-#define spin_lock_irqsave(lock, flags)	_spin_lock_irqsave(lock, flags)
-#define read_lock_irqsave(lock, flags)	_read_lock_irqsave(lock, flags)
-#define write_lock_irqsave(lock, flags)	_write_lock_irqsave(lock, flags)
+#define spin_lock_irqsave(lock, flags)			\
+	do {						\
+		typecheck(unsigned long, flags);	\
+		_spin_lock_irqsave(lock, flags);	\
+	} while (0)
+#define read_lock_irqsave(lock, flags)			\
+	do {						\
+		typecheck(unsigned long, flags);	\
+		_read_lock_irqsave(lock, flags);	\
+	} while (0)
+#define write_lock_irqsave(lock, flags)			\
+	do {						\
+		typecheck(unsigned long, flags);	\
+		_write_lock_irqsave(lock, flags);	\
+	} while (0)
 #define spin_lock_irqsave_nested(lock, flags, subclass)	\
 	spin_lock_irqsave(lock, flags)
 
@@ -260,16 +291,25 @@ do {						\
 } while (0)
 #endif
 
-#define spin_unlock_irqrestore(lock, flags) \
-					_spin_unlock_irqrestore(lock, flags)
+#define spin_unlock_irqrestore(lock, flags)		\
+	do {						\
+		typecheck(unsigned long, flags);	\
+		_spin_unlock_irqrestore(lock, flags);	\
+	} while (0)
 #define spin_unlock_bh(lock)		_spin_unlock_bh(lock)
 
-#define read_unlock_irqrestore(lock, flags) \
-					_read_unlock_irqrestore(lock, flags)
+#define read_unlock_irqrestore(lock, flags)		\
+	do {						\
+		typecheck(unsigned long, flags);	\
+		_read_unlock_irqrestore(lock, flags);	\
+	} while (0)
 #define read_unlock_bh(lock)		_read_unlock_bh(lock)
 
-#define write_unlock_irqrestore(lock, flags) \
-					_write_unlock_irqrestore(lock, flags)
+#define write_unlock_irqrestore(lock, flags)		\
+	do {						\
+		typecheck(unsigned long, flags);	\
+		_write_unlock_irqrestore(lock, flags);	\
+	} while (0)
 #define write_unlock_bh(lock)		_write_unlock_bh(lock)
 
 #define spin_trylock_bh(lock)	__cond_lock(lock, _spin_trylock_bh(lock))

From b7bbf8fa6ba329b3552b75a0716f5fbc6f839499 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Fri, 25 Jul 2008 01:45:25 -0700
Subject: [PATCH 0677/1171] fs: ldm.[ch] use get_unaligned_* helpers

Replace the private BE16/BE32/BE64 macros with direct calls to
get_unaligned_be16/32/64.

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/partitions/ldm.c | 70 ++++++++++++++++++++++-----------------------
 fs/partitions/ldm.h |  5 ----
 2 files changed, 35 insertions(+), 40 deletions(-)

diff --git a/fs/partitions/ldm.c b/fs/partitions/ldm.c
index 0fdda2e8a4cc..8652fb99e962 100644
--- a/fs/partitions/ldm.c
+++ b/fs/partitions/ldm.c
@@ -133,17 +133,17 @@ static bool ldm_parse_privhead(const u8 *data, struct privhead *ph)
 	bool is_vista = false;
 
 	BUG_ON(!data || !ph);
-	if (MAGIC_PRIVHEAD != BE64(data)) {
+	if (MAGIC_PRIVHEAD != get_unaligned_be64(data)) {
 		ldm_error("Cannot find PRIVHEAD structure. LDM database is"
 			" corrupt. Aborting.");
 		return false;
 	}
-	ph->ver_major = BE16(data + 0x000C);
-	ph->ver_minor = BE16(data + 0x000E);
-	ph->logical_disk_start = BE64(data + 0x011B);
-	ph->logical_disk_size = BE64(data + 0x0123);
-	ph->config_start = BE64(data + 0x012B);
-	ph->config_size = BE64(data + 0x0133);
+	ph->ver_major = get_unaligned_be16(data + 0x000C);
+	ph->ver_minor = get_unaligned_be16(data + 0x000E);
+	ph->logical_disk_start = get_unaligned_be64(data + 0x011B);
+	ph->logical_disk_size = get_unaligned_be64(data + 0x0123);
+	ph->config_start = get_unaligned_be64(data + 0x012B);
+	ph->config_size = get_unaligned_be64(data + 0x0133);
 	/* Version 2.11 is Win2k/XP and version 2.12 is Vista. */
 	if (ph->ver_major == 2 && ph->ver_minor == 12)
 		is_vista = true;
@@ -191,14 +191,14 @@ static bool ldm_parse_tocblock (const u8 *data, struct tocblock *toc)
 {
 	BUG_ON (!data || !toc);
 
-	if (MAGIC_TOCBLOCK != BE64 (data)) {
+	if (MAGIC_TOCBLOCK != get_unaligned_be64(data)) {
 		ldm_crit ("Cannot find TOCBLOCK, database may be corrupt.");
 		return false;
 	}
 	strncpy (toc->bitmap1_name, data + 0x24, sizeof (toc->bitmap1_name));
 	toc->bitmap1_name[sizeof (toc->bitmap1_name) - 1] = 0;
-	toc->bitmap1_start = BE64 (data + 0x2E);
-	toc->bitmap1_size  = BE64 (data + 0x36);
+	toc->bitmap1_start = get_unaligned_be64(data + 0x2E);
+	toc->bitmap1_size  = get_unaligned_be64(data + 0x36);
 
 	if (strncmp (toc->bitmap1_name, TOC_BITMAP1,
 			sizeof (toc->bitmap1_name)) != 0) {
@@ -208,8 +208,8 @@ static bool ldm_parse_tocblock (const u8 *data, struct tocblock *toc)
 	}
 	strncpy (toc->bitmap2_name, data + 0x46, sizeof (toc->bitmap2_name));
 	toc->bitmap2_name[sizeof (toc->bitmap2_name) - 1] = 0;
-	toc->bitmap2_start = BE64 (data + 0x50);
-	toc->bitmap2_size  = BE64 (data + 0x58);
+	toc->bitmap2_start = get_unaligned_be64(data + 0x50);
+	toc->bitmap2_size  = get_unaligned_be64(data + 0x58);
 	if (strncmp (toc->bitmap2_name, TOC_BITMAP2,
 			sizeof (toc->bitmap2_name)) != 0) {
 		ldm_crit ("TOCBLOCK's second bitmap is '%s', should be '%s'.",
@@ -237,22 +237,22 @@ static bool ldm_parse_vmdb (const u8 *data, struct vmdb *vm)
 {
 	BUG_ON (!data || !vm);
 
-	if (MAGIC_VMDB != BE32 (data)) {
+	if (MAGIC_VMDB != get_unaligned_be32(data)) {
 		ldm_crit ("Cannot find the VMDB, database may be corrupt.");
 		return false;
 	}
 
-	vm->ver_major = BE16 (data + 0x12);
-	vm->ver_minor = BE16 (data + 0x14);
+	vm->ver_major = get_unaligned_be16(data + 0x12);
+	vm->ver_minor = get_unaligned_be16(data + 0x14);
 	if ((vm->ver_major != 4) || (vm->ver_minor != 10)) {
 		ldm_error ("Expected VMDB version %d.%d, got %d.%d. "
 			"Aborting.", 4, 10, vm->ver_major, vm->ver_minor);
 		return false;
 	}
 
-	vm->vblk_size     = BE32 (data + 0x08);
-	vm->vblk_offset   = BE32 (data + 0x0C);
-	vm->last_vblk_seq = BE32 (data + 0x04);
+	vm->vblk_size     = get_unaligned_be32(data + 0x08);
+	vm->vblk_offset   = get_unaligned_be32(data + 0x0C);
+	vm->last_vblk_seq = get_unaligned_be32(data + 0x04);
 
 	ldm_debug ("Parsed VMDB successfully.");
 	return true;
@@ -507,7 +507,7 @@ static bool ldm_validate_vmdb (struct block_device *bdev, unsigned long base,
 		goto out;				/* Already logged */
 
 	/* Are there uncommitted transactions? */
-	if (BE16(data + 0x10) != 0x01) {
+	if (get_unaligned_be16(data + 0x10) != 0x01) {
 		ldm_crit ("Database is not in a consistent state.  Aborting.");
 		goto out;
 	}
@@ -802,7 +802,7 @@ static bool ldm_parse_cmp3 (const u8 *buffer, int buflen, struct vblk *vb)
 		return false;
 
 	len += VBLK_SIZE_CMP3;
-	if (len != BE32 (buffer + 0x14))
+	if (len != get_unaligned_be32(buffer + 0x14))
 		return false;
 
 	comp = &vb->vblk.comp;
@@ -851,7 +851,7 @@ static int ldm_parse_dgr3 (const u8 *buffer, int buflen, struct vblk *vb)
 		return false;
 
 	len += VBLK_SIZE_DGR3;
-	if (len != BE32 (buffer + 0x14))
+	if (len != get_unaligned_be32(buffer + 0x14))
 		return false;
 
 	dgrp = &vb->vblk.dgrp;
@@ -895,7 +895,7 @@ static bool ldm_parse_dgr4 (const u8 *buffer, int buflen, struct vblk *vb)
 		return false;
 
 	len += VBLK_SIZE_DGR4;
-	if (len != BE32 (buffer + 0x14))
+	if (len != get_unaligned_be32(buffer + 0x14))
 		return false;
 
 	dgrp = &vb->vblk.dgrp;
@@ -931,7 +931,7 @@ static bool ldm_parse_dsk3 (const u8 *buffer, int buflen, struct vblk *vb)
 		return false;
 
 	len += VBLK_SIZE_DSK3;
-	if (len != BE32 (buffer + 0x14))
+	if (len != get_unaligned_be32(buffer + 0x14))
 		return false;
 
 	disk = &vb->vblk.disk;
@@ -968,7 +968,7 @@ static bool ldm_parse_dsk4 (const u8 *buffer, int buflen, struct vblk *vb)
 		return false;
 
 	len += VBLK_SIZE_DSK4;
-	if (len != BE32 (buffer + 0x14))
+	if (len != get_unaligned_be32(buffer + 0x14))
 		return false;
 
 	disk = &vb->vblk.disk;
@@ -1034,14 +1034,14 @@ static bool ldm_parse_prt3(const u8 *buffer, int buflen, struct vblk *vb)
 		return false;
 	}
 	len += VBLK_SIZE_PRT3;
-	if (len > BE32(buffer + 0x14)) {
+	if (len > get_unaligned_be32(buffer + 0x14)) {
 		ldm_error("len %d > BE32(buffer + 0x14) %d", len,
-				BE32(buffer + 0x14));
+				get_unaligned_be32(buffer + 0x14));
 		return false;
 	}
 	part = &vb->vblk.part;
-	part->start = BE64(buffer + 0x24 + r_name);
-	part->volume_offset = BE64(buffer + 0x2C + r_name);
+	part->start = get_unaligned_be64(buffer + 0x24 + r_name);
+	part->volume_offset = get_unaligned_be64(buffer + 0x2C + r_name);
 	part->size = ldm_get_vnum(buffer + 0x34 + r_name);
 	part->parent_id = ldm_get_vnum(buffer + 0x34 + r_size);
 	part->disk_id = ldm_get_vnum(buffer + 0x34 + r_parent);
@@ -1139,9 +1139,9 @@ static bool ldm_parse_vol5(const u8 *buffer, int buflen, struct vblk *vb)
 		return false;
 	}
 	len += VBLK_SIZE_VOL5;
-	if (len > BE32(buffer + 0x14)) {
+	if (len > get_unaligned_be32(buffer + 0x14)) {
 		ldm_error("len %d > BE32(buffer + 0x14) %d", len,
-				BE32(buffer + 0x14));
+				get_unaligned_be32(buffer + 0x14));
 		return false;
 	}
 	volu = &vb->vblk.volu;
@@ -1294,9 +1294,9 @@ static bool ldm_frag_add (const u8 *data, int size, struct list_head *frags)
 
 	BUG_ON (!data || !frags);
 
-	group = BE32 (data + 0x08);
-	rec   = BE16 (data + 0x0C);
-	num   = BE16 (data + 0x0E);
+	group = get_unaligned_be32(data + 0x08);
+	rec   = get_unaligned_be16(data + 0x0C);
+	num   = get_unaligned_be16(data + 0x0E);
 	if ((num < 1) || (num > 4)) {
 		ldm_error ("A VBLK claims to have %d parts.", num);
 		return false;
@@ -1425,12 +1425,12 @@ static bool ldm_get_vblks (struct block_device *bdev, unsigned long base,
 		}
 
 		for (v = 0; v < perbuf; v++, data+=size) {  /* For each vblk */
-			if (MAGIC_VBLK != BE32 (data)) {
+			if (MAGIC_VBLK != get_unaligned_be32(data)) {
 				ldm_error ("Expected to find a VBLK.");
 				goto out;
 			}
 
-			recs = BE16 (data + 0x0E);	/* Number of records */
+			recs = get_unaligned_be16(data + 0x0E);	/* Number of records */
 			if (recs == 1) {
 				if (!ldm_ldmdb_add (data, size, ldb))
 					goto out;	/* Already logged */
diff --git a/fs/partitions/ldm.h b/fs/partitions/ldm.h
index 80f63b5fdd9f..30e08e809c1d 100644
--- a/fs/partitions/ldm.h
+++ b/fs/partitions/ldm.h
@@ -98,11 +98,6 @@ struct parsed_partitions;
 #define TOC_BITMAP1		"config"	/* Names of the two defined */
 #define TOC_BITMAP2		"log"		/* bitmaps in the TOCBLOCK. */
 
-/* Most numbers we deal with are big-endian and won't be aligned. */
-#define BE16(x)			((u16)be16_to_cpu(get_unaligned((__be16*)(x))))
-#define BE32(x)			((u32)be32_to_cpu(get_unaligned((__be32*)(x))))
-#define BE64(x)			((u64)be64_to_cpu(get_unaligned((__be64*)(x))))
-
 /* Borrowed from msdos.c */
 #define SYS_IND(p)		(get_unaligned(&(p)->sys_ind))
 

From 8b5ac31e27135a6f2c210c40d03bf8f1b3a86b77 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Fri, 25 Jul 2008 01:45:26 -0700
Subject: [PATCH 0678/1171] include: use get/put_unaligned_* helpers

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Cc: "John W. Linville" <linville@tuxdriver.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/reiserfs_fs.h      |  4 ++--
 include/linux/smb_fs.h           | 19 +++++++------------
 include/net/ieee80211_radiotap.h |  2 +-
 3 files changed, 10 insertions(+), 15 deletions(-)

diff --git a/include/linux/reiserfs_fs.h b/include/linux/reiserfs_fs.h
index 4aacaeecb56f..e9963af16cda 100644
--- a/include/linux/reiserfs_fs.h
+++ b/include/linux/reiserfs_fs.h
@@ -526,8 +526,8 @@ struct item_head {
 ** p is the array of __u32, i is the index into the array, v is the value
 ** to store there.
 */
-#define get_block_num(p, i) le32_to_cpu(get_unaligned((p) + (i)))
-#define put_block_num(p, i, v) put_unaligned(cpu_to_le32(v), (p) + (i))
+#define get_block_num(p, i) get_unaligned_le32((p) + (i))
+#define put_block_num(p, i, v) put_unaligned_le32((v), (p) + (i))
 
 //
 // in old version uniqueness field shows key type
diff --git a/include/linux/smb_fs.h b/include/linux/smb_fs.h
index 2c5cd55f44ff..923cd8a247b1 100644
--- a/include/linux/smb_fs.h
+++ b/include/linux/smb_fs.h
@@ -43,18 +43,13 @@ static inline struct smb_inode_info *SMB_I(struct inode *inode)
 }
 
 /* macro names are short for word, double-word, long value (?) */
-#define WVAL(buf,pos) \
-	(le16_to_cpu(get_unaligned((__le16 *)((u8 *)(buf) + (pos)))))
-#define DVAL(buf,pos) \
-	(le32_to_cpu(get_unaligned((__le32 *)((u8 *)(buf) + (pos)))))
-#define LVAL(buf,pos) \
-	(le64_to_cpu(get_unaligned((__le64 *)((u8 *)(buf) + (pos)))))
-#define WSET(buf,pos,val) \
-	put_unaligned(cpu_to_le16((u16)(val)), (__le16 *)((u8 *)(buf) + (pos)))
-#define DSET(buf,pos,val) \
-	put_unaligned(cpu_to_le32((u32)(val)), (__le32 *)((u8 *)(buf) + (pos)))
-#define LSET(buf,pos,val) \
-	put_unaligned(cpu_to_le64((u64)(val)), (__le64 *)((u8 *)(buf) + (pos)))
+#define WVAL(buf, pos) (get_unaligned_le16((u8 *)(buf) + (pos)))
+#define DVAL(buf, pos) (get_unaligned_le32((u8 *)(buf) + (pos)))
+#define LVAL(buf, pos) (get_unaligned_le64((u8 *)(buf) + (pos)))
+
+#define WSET(buf, pos, val) put_unaligned_le16((val), (u8 *)(buf) + (pos))
+#define DSET(buf, pos, val) put_unaligned_le32((val), (u8 *)(buf) + (pos))
+#define LSET(buf, pos, val) put_unaligned_le64((val), (u8 *)(buf) + (pos))
 
 /* where to find the base of the SMB packet proper */
 #define smb_base(buf) ((u8 *)(((u8 *)(buf))+4))
diff --git a/include/net/ieee80211_radiotap.h b/include/net/ieee80211_radiotap.h
index dfd8bf66ce27..d364fd594ea4 100644
--- a/include/net/ieee80211_radiotap.h
+++ b/include/net/ieee80211_radiotap.h
@@ -262,7 +262,7 @@ static inline int ieee80211_get_radiotap_len(unsigned char *data)
 	struct ieee80211_radiotap_header *hdr =
 		(struct ieee80211_radiotap_header *)data;
 
-	return le16_to_cpu(get_unaligned(&hdr->it_len));
+	return get_unaligned_le16(&hdr->it_len);
 }
 
 #endif				/* IEEE80211_RADIOTAP_H */

From 545e400619b24b6b17b7f1f1e838e9ff6d036949 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Fri, 25 Jul 2008 01:45:27 -0700
Subject: [PATCH 0679/1171] lzo: use get/put_unaligned_* helpers

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Acked-by: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 lib/lzo/lzo1x_decompress.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/lib/lzo/lzo1x_decompress.c b/lib/lzo/lzo1x_decompress.c
index 77f0f9b775a9..5dc6b29c1575 100644
--- a/lib/lzo/lzo1x_decompress.c
+++ b/lib/lzo/lzo1x_decompress.c
@@ -138,8 +138,7 @@ match:
 					t += 31 + *ip++;
 				}
 				m_pos = op - 1;
-				m_pos -= le16_to_cpu(get_unaligned(
-					(const unsigned short *)ip)) >> 2;
+				m_pos -= get_unaligned_le16(ip) >> 2;
 				ip += 2;
 			} else if (t >= 16) {
 				m_pos = op;
@@ -157,8 +156,7 @@ match:
 					}
 					t += 7 + *ip++;
 				}
-				m_pos -= le16_to_cpu(get_unaligned(
-					(const unsigned short *)ip)) >> 2;
+				m_pos -= get_unaligned_le16(ip) >> 2;
 				ip += 2;
 				if (m_pos == op)
 					goto eof_found;

From 585e93ae83b80c874bf4eb50a239027cef5db4af Mon Sep 17 00:00:00 2001
From: Eric Sandeen <sandeen@redhat.com>
Date: Fri, 25 Jul 2008 01:45:27 -0700
Subject: [PATCH 0680/1171] find dynamic stack allocations in checkstack.pl

Currently, checkstack.pl only looks for fixed subtractions from the stack
pointer.  However, things like this:

void function(int size)
{
        char stackbuster[size << 2];
...

are certainly worth pointing out, I think.

This could perhaps be done more cleanly, and the following patch only
adds "dynamic" REs for x86 and x86_64, but it works:

0x00b0 crypto_cbc_decrypt_inplace [cbc]:                Dynamic (%rax)
0x00ad crypto_pcbc_decrypt_inplace [pcbc]:              Dynamic (%rax)
0x02f6 crypto_pcbc_encrypt_inplace [pcbc]:              Dynamic (%rax)
0x036c _crypto_xcbc_digest_setkey [xcbc]:               Dynamic (%rax)
...

(Inspired by Keith Owens' old stack-check script)

Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkstack.pl | 24 +++++++++++++++++++++++-
 1 file changed, 23 insertions(+), 1 deletion(-)

diff --git a/scripts/checkstack.pl b/scripts/checkstack.pl
index 340ad6920511..358f96c75b43 100755
--- a/scripts/checkstack.pl
+++ b/scripts/checkstack.pl
@@ -26,8 +26,12 @@
 # $& (whole re) matches the complete objdump line with the stack growth
 # $1 (first bracket) matches the size of the stack growth
 #
+# $dre is similar, but for dynamic stack redutions:
+# $& (whole re) matches the complete objdump line with the stack growth
+# $1 (first bracket) matches the dynamic amount of the stack growth
+#
 # use anything else and feel the pain ;)
-my (@stack, $re, $x, $xs);
+my (@stack, $re, $dre, $x, $xs);
 {
 	my $arch = shift;
 	if ($arch eq "") {
@@ -46,9 +50,11 @@ my (@stack, $re, $x, $xs);
 	} elsif ($arch =~ /^i[3456]86$/) {
 		#c0105234:       81 ec ac 05 00 00       sub    $0x5ac,%esp
 		$re = qr/^.*[as][du][db]    \$(0x$x{1,8}),\%esp$/o;
+		$dre = qr/^.*[as][du][db]    (%.*),\%esp$/o;
 	} elsif ($arch eq 'x86_64') {
 		#    2f60:	48 81 ec e8 05 00 00 	sub    $0x5e8,%rsp
 		$re = qr/^.*[as][du][db]    \$(0x$x{1,8}),\%rsp$/o;
+		$dre = qr/^.*[as][du][db]    (\%.*),\%rsp$/o;
 	} elsif ($arch eq 'ia64') {
 		#e0000000044011fc:       01 0f fc 8c     adds r12=-384,r12
 		$re = qr/.*adds.*r12=-(([0-9]{2}|[3-9])[0-9]{2}),r12/o;
@@ -141,6 +147,22 @@ while (my $line = <STDIN>) {
 		next if ($size < 100);
 		push @stack, "$intro$size\n";
 	}
+	elsif (defined $dre && $line =~ m/$dre/) {
+		my $size = "Dynamic ($1)";
+
+		next if $line !~ m/^($xs*)/;
+		my $addr = $1;
+		$addr =~ s/ /0/g;
+		$addr = "0x$addr";
+
+		my $intro = "$addr $func [$file]:";
+		my $padlen = 56 - length($intro);
+		while ($padlen > 0) {
+			$intro .= '	';
+			$padlen -= 8;
+		}
+		push @stack, "$intro$size\n";
+	}
 }
 
 print sort bysize @stack;

From abddaec56ebb7911bbf0578a4636a74bd7376d92 Mon Sep 17 00:00:00 2001
From: Eric Sandeen <sandeen@redhat.com>
Date: Fri, 25 Jul 2008 01:45:28 -0700
Subject: [PATCH 0681/1171] fix checkstack.pl arch detection

uname -m was leaving a newline in $arch, and not passing the tests.

Also, printing the unknown arch on failure is probably helpful.

Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 scripts/checkstack.pl | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/scripts/checkstack.pl b/scripts/checkstack.pl
index 358f96c75b43..3eca62566d6b 100755
--- a/scripts/checkstack.pl
+++ b/scripts/checkstack.pl
@@ -36,6 +36,7 @@ my (@stack, $re, $dre, $x, $xs);
 	my $arch = shift;
 	if ($arch eq "") {
 		$arch = `uname -m`;
+		chomp($arch);
 	}
 
 	$x	= "[0-9a-f]";	# hex character
@@ -91,7 +92,7 @@ my (@stack, $re, $dre, $x, $xs);
 		#   0:   00 e8 38 01     LINK 0x4e0;
 		$re = qr/.*[[:space:]]LINK[[:space:]]*(0x$x{1,8})/o;
 	} else {
-		print("wrong or unknown architecture\n");
+		print("wrong or unknown architecture \"$arch\"\n");
 		exit
 	}
 }

From 82c8253ac27291d6c70114eb445c714359812a10 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 01:45:29 -0700
Subject: [PATCH 0682/1171] init/do_mounts.c should #include <linux/initrd.h>

Every file should include the headers containing the externs for its
global code (in this case for rd_doload).

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 init/do_mounts.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/init/do_mounts.c b/init/do_mounts.c
index a1de1bf3d6b9..f769fac4f4c0 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -12,6 +12,7 @@
 #include <linux/device.h>
 #include <linux/init.h>
 #include <linux/fs.h>
+#include <linux/initrd.h>
 
 #include <linux/nfs_fs.h>
 #include <linux/nfs_fs_sb.h>

From b39c08cb692cb8898c30e0d8187c7cbe27cc905c Mon Sep 17 00:00:00 2001
From: "Robert P. J. Day" <rpjday@crashcourse.ca>
Date: Fri, 25 Jul 2008 01:45:29 -0700
Subject: [PATCH 0683/1171] Remove apparently unused fd1772.h header file.

This header file has been unused for quite some time, and the
corresponding source files appear to have been removed back in commit
99eb8a550dbccc0e1f6c7e866fe421810e0585f6 ("Remove the arm26 port")

Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Cc: Adrian Bunk <bunk@stusta.de>
Cc: Ian Molton <spyro@f2s.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/fd1772.h | 80 ------------------------------------------
 1 file changed, 80 deletions(-)
 delete mode 100644 include/linux/fd1772.h

diff --git a/include/linux/fd1772.h b/include/linux/fd1772.h
deleted file mode 100644
index 871d6e4c677e..000000000000
--- a/include/linux/fd1772.h
+++ /dev/null
@@ -1,80 +0,0 @@
-#ifndef _LINUX_FD1772REG_H
-#define _LINUX_FD1772REG_H
-
-/*
-** WD1772 stuff - originally from the M68K Linux
- * Modified for Archimedes by Dave Gilbert (gilbertd@cs.man.ac.uk)
- */
-
-/* register codes */
-
-#define FDC1772SELREG_STP   (0x80)   /* command/status register */
-#define FDC1772SELREG_TRA   (0x82)   /* track register */
-#define FDC1772SELREG_SEC   (0x84)   /* sector register */
-#define FDC1772SELREG_DTA   (0x86)   /* data register */
-
-/* register names for FDC1772_READ/WRITE macros */
-
-#define FDC1772REG_CMD         0
-#define FDC1772REG_STATUS      0
-#define FDC1772REG_TRACK       2
-#define FDC1772REG_SECTOR      4
-#define FDC1772REG_DATA                6
-
-/* command opcodes */
-
-#define FDC1772CMD_RESTORE  (0x00)   /*  -                   */
-#define FDC1772CMD_SEEK     (0x10)   /*   |                  */
-#define FDC1772CMD_STEP     (0x20)   /*   |  TYP 1 Commands  */
-#define FDC1772CMD_STIN     (0x40)   /*   |                  */
-#define FDC1772CMD_STOT     (0x60)   /*  -                   */
-#define FDC1772CMD_RDSEC    (0x80)   /*  -   TYP 2 Commands  */
-#define FDC1772CMD_WRSEC    (0xa0)   /*  -          "        */
-#define FDC1772CMD_RDADR    (0xc0)   /*  -                   */
-#define FDC1772CMD_RDTRA    (0xe0)   /*   |  TYP 3 Commands  */
-#define FDC1772CMD_WRTRA    (0xf0)   /*  -                   */
-#define FDC1772CMD_FORCI    (0xd0)   /*  -   TYP 4 Command   */
-
-/* command modifier bits */
-
-#define FDC1772CMDADD_SR6   (0x00)   /* step rate settings */
-#define FDC1772CMDADD_SR12  (0x01)
-#define FDC1772CMDADD_SR2   (0x02)
-#define FDC1772CMDADD_SR3   (0x03)
-#define FDC1772CMDADD_V     (0x04)   /* verify */
-#define FDC1772CMDADD_H     (0x08)   /* wait for spin-up */
-#define FDC1772CMDADD_U     (0x10)   /* update track register */
-#define FDC1772CMDADD_M     (0x10)   /* multiple sector access */
-#define FDC1772CMDADD_E     (0x04)   /* head settling flag */
-#define FDC1772CMDADD_P     (0x02)   /* precompensation */
-#define FDC1772CMDADD_A0    (0x01)   /* DAM flag */
-
-/* status register bits */
-
-#define        FDC1772STAT_MOTORON     (0x80)   /* motor on */
-#define        FDC1772STAT_WPROT       (0x40)   /* write protected (FDC1772CMD_WR*) */
-#define        FDC1772STAT_SPINUP      (0x20)   /* motor speed stable (Type I) */
-#define        FDC1772STAT_DELDAM      (0x20)   /* sector has deleted DAM (Type II+III) */
-#define        FDC1772STAT_RECNF       (0x10)   /* record not found */
-#define        FDC1772STAT_CRC         (0x08)   /* CRC error */
-#define        FDC1772STAT_TR00        (0x04)   /* Track 00 flag (Type I) */
-#define        FDC1772STAT_LOST        (0x04)   /* Lost Data (Type II+III) */
-#define        FDC1772STAT_IDX         (0x02)   /* Index status (Type I) */
-#define        FDC1772STAT_DRQ         (0x02)   /* DRQ status (Type II+III) */
-#define        FDC1772STAT_BUSY        (0x01)   /* FDC1772 is busy */
-
-
-/* PSG Port A Bit Nr 0 .. Side Sel .. 0 -> Side 1  1 -> Side 2 */
-#define DSKSIDE     (0x01)
-        
-#define DSKDRVNONE  (0x06)
-#define DSKDRV0     (0x02)
-#define DSKDRV1     (0x04)
-
-/* step rates */
-#define        FDC1772STEP_6   0x00
-#define        FDC1772STEP_12  0x01
-#define        FDC1772STEP_2   0x02
-#define        FDC1772STEP_3   0x03
-
-#endif

From cb345d7352aa9e692ef4b83c41d3e6e1cdb2f846 Mon Sep 17 00:00:00 2001
From: "Robert P. J. Day" <rpjday@crashcourse.ca>
Date: Fri, 25 Jul 2008 01:45:30 -0700
Subject: [PATCH 0684/1171] init/: delete hard-coded setting and testing of
 BUILD_CRAMDISK

There seems to be little point in explicitly setting, then testing the macro
BUILD_CRAMDISK within the context of a single source file.

Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 init/do_mounts_rd.c | 12 ------------
 1 file changed, 12 deletions(-)

diff --git a/init/do_mounts_rd.c b/init/do_mounts_rd.c
index 46dfd64ae8fb..470a328d1454 100644
--- a/init/do_mounts_rd.c
+++ b/init/do_mounts_rd.c
@@ -10,8 +10,6 @@
 
 #include "do_mounts.h"
 
-#define BUILD_CRAMDISK
-
 int __initdata rd_prompt = 1;/* 1 = prompt for RAM disk, 0 = don't prompt */
 
 static int __init prompt_ramdisk(char *str)
@@ -162,14 +160,8 @@ int __init rd_load_image(char *from)
 		goto done;
 
 	if (nblocks == 0) {
-#ifdef BUILD_CRAMDISK
 		if (crd_load(in_fd, out_fd) == 0)
 			goto successful_load;
-#else
-		printk(KERN_NOTICE
-		       "RAMDISK: Kernel does not support compressed "
-		       "RAM disk images\n");
-#endif
 		goto done;
 	}
 
@@ -267,8 +259,6 @@ int __init rd_load_disk(int n)
 	return rd_load_image("/dev/root");
 }
 
-#ifdef BUILD_CRAMDISK
-
 /*
  * gzip declarations
  */
@@ -425,5 +415,3 @@ static int __init crd_load(int in_fd, int out_fd)
 	kfree(window);
 	return result;
 }
-
-#endif  /* BUILD_CRAMDISK */

From fd193829744bc77392395cf8f47889235c97f0a3 Mon Sep 17 00:00:00 2001
From: "Robert P. J. Day" <rpjday@crashcourse.ca>
Date: Fri, 25 Jul 2008 01:45:31 -0700
Subject: [PATCH 0685/1171] lib: allow memparse() to accept a NULL and
 ignorable second parm

Extend memparse() to allow the caller to use a NULL second parameter, which
would represent no interest in returning the address of the end of the parsed
string.

In numerous cases, callers invoke memparse() to parse a possibly-suffixed
string (such as "64K" or "2G" or whatever) and define a character pointer to
accept the end pointer being returned by memparse() even though they have no
interest in it and promptly throw it away.

This (backward-compatible) enhancement allows callers to use NULL in the cases
where they just don't care about getting back that end pointer.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 lib/cmdline.c | 16 +++++++++++-----
 1 file changed, 11 insertions(+), 5 deletions(-)

diff --git a/lib/cmdline.c b/lib/cmdline.c
index f596c08d213a..5ba8a942a478 100644
--- a/lib/cmdline.c
+++ b/lib/cmdline.c
@@ -116,7 +116,7 @@ char *get_options(const char *str, int nints, int *ints)
 /**
  *	memparse - parse a string with mem suffixes into a number
  *	@ptr: Where parse begins
- *	@retptr: (output) Pointer to next char after parse completes
+ *	@retptr: (output) Optional pointer to next char after parse completes
  *
  *	Parses a string into a number.  The number stored at @ptr is
  *	potentially suffixed with %K (for kilobytes, or 1024 bytes),
@@ -126,11 +126,13 @@ char *get_options(const char *str, int nints, int *ints)
  *	megabyte, or one gigabyte, respectively.
  */
 
-unsigned long long memparse (char *ptr, char **retptr)
+unsigned long long memparse(char *ptr, char **retptr)
 {
-	unsigned long long ret = simple_strtoull (ptr, retptr, 0);
+	char *endptr;	/* local pointer to end of parsed string */
 
-	switch (**retptr) {
+	unsigned long long ret = simple_strtoull(ptr, &endptr, 0);
+
+	switch (*endptr) {
 	case 'G':
 	case 'g':
 		ret <<= 10;
@@ -140,10 +142,14 @@ unsigned long long memparse (char *ptr, char **retptr)
 	case 'K':
 	case 'k':
 		ret <<= 10;
-		(*retptr)++;
+		endptr++;
 	default:
 		break;
 	}
+
+	if (retptr)
+		*retptr = endptr;
+
 	return ret;
 }
 

From e0ce0da9fefcc723dc006c35a7f91a32750abd40 Mon Sep 17 00:00:00 2001
From: "Robert P. J. Day" <rpjday@crashcourse.ca>
Date: Fri, 25 Jul 2008 01:45:32 -0700
Subject: [PATCH 0686/1171] lists: remove a redundant conditional definition of
 list_add()

Remove the conditional surrounding the definition of list_add() from list.h
since, if you define CONFIG_DEBUG_LIST, the definition you will subsequently
pick up from lib/list_debug.c will be absolutely identical, at which point you
can remove that redundant definition from list_debug.c as well.

Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Cc: Dave Jones <davej@codemonkey.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/list.h |  4 ----
 lib/list_debug.c     | 14 --------------
 2 files changed, 18 deletions(-)

diff --git a/include/linux/list.h b/include/linux/list.h
index 139ec41d9c2e..453916bc0412 100644
--- a/include/linux/list.h
+++ b/include/linux/list.h
@@ -61,14 +61,10 @@ extern void __list_add(struct list_head *new,
  * Insert a new entry after the specified head.
  * This is good for implementing stacks.
  */
-#ifndef CONFIG_DEBUG_LIST
 static inline void list_add(struct list_head *new, struct list_head *head)
 {
 	__list_add(new, head, head->next);
 }
-#else
-extern void list_add(struct list_head *new, struct list_head *head);
-#endif
 
 
 /**
diff --git a/lib/list_debug.c b/lib/list_debug.c
index 4350ba9655bd..45c03fd608dd 100644
--- a/lib/list_debug.c
+++ b/lib/list_debug.c
@@ -39,20 +39,6 @@ void __list_add(struct list_head *new,
 }
 EXPORT_SYMBOL(__list_add);
 
-/**
- * list_add - add a new entry
- * @new: new entry to be added
- * @head: list head to add it after
- *
- * Insert a new entry after the specified head.
- * This is good for implementing stacks.
- */
-void list_add(struct list_head *new, struct list_head *head)
-{
-	__list_add(new, head, head->next);
-}
-EXPORT_SYMBOL(list_add);
-
 /**
  * list_del - deletes entry from list.
  * @entry: the element to delete from the list.

From 58340a07c194e0aed7bc58b61ff24330bb2a409f Mon Sep 17 00:00:00 2001
From: Johannes Berg <johannes@sipsolutions.net>
Date: Fri, 25 Jul 2008 01:45:33 -0700
Subject: [PATCH 0687/1171] introduce HAVE_EFFICIENT_UNALIGNED_ACCESS Kconfig
 symbol

In many cases, especially in networking, it can be beneficial to know at
compile time whether the architecture can do unaligned accesses efficiently.
This patch introduces a new Kconfig symbol

	HAVE_EFFICIENT_UNALIGNED_ACCESS

for that purpose and adds it to the powerpc and x86 architectures.  Also add
some documentation about alignment and networking, and especially one intended
use of this symbol.

Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Acked-by: Ingo Molnar <mingo@elte.hu> [x86 architecture part]
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/unaligned-memory-access.txt | 36 +++++++++++++++++++----
 arch/Kconfig                              | 19 ++++++++++++
 arch/powerpc/Kconfig                      |  1 +
 arch/x86/Kconfig                          |  1 +
 4 files changed, 52 insertions(+), 5 deletions(-)

diff --git a/Documentation/unaligned-memory-access.txt b/Documentation/unaligned-memory-access.txt
index b0472ac5226a..f866c72291bf 100644
--- a/Documentation/unaligned-memory-access.txt
+++ b/Documentation/unaligned-memory-access.txt
@@ -218,9 +218,35 @@ If use of such macros is not convenient, another option is to use memcpy(),
 where the source or destination (or both) are of type u8* or unsigned char*.
 Due to the byte-wise nature of this operation, unaligned accesses are avoided.
 
---
-Author: Daniel Drake <dsd@gentoo.org>
-With help from: Alan Cox, Avuton Olrich, Heikki Orsila, Jan Engelhardt,
-Johannes Berg, Kyle McMartin, Kyle Moffett, Randy Dunlap, Robert Hancock,
-Uli Kunitz, Vadim Lobanov
+
+Alignment vs. Networking
+========================
+
+On architectures that require aligned loads, networking requires that the IP
+header is aligned on a four-byte boundary to optimise the IP stack. For
+regular ethernet hardware, the constant NET_IP_ALIGN is used. On most
+architectures this constant has the value 2 because the normal ethernet
+header is 14 bytes long, so in order to get proper alignment one needs to
+DMA to an address which can be expressed as 4*n + 2. One notable exception
+here is powerpc which defines NET_IP_ALIGN to 0 because DMA to unaligned
+addresses can be very expensive and dwarf the cost of unaligned loads.
+
+For some ethernet hardware that cannot DMA to unaligned addresses like
+4*n+2 or non-ethernet hardware, this can be a problem, and it is then
+required to copy the incoming frame into an aligned buffer. Because this is
+unnecessary on architectures that can do unaligned accesses, the code can be
+made dependent on CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS like so:
+
+#ifdef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
+	skb = original skb
+#else
+	skb = copy skb
+#endif
+
+--
+Authors: Daniel Drake <dsd@gentoo.org>,
+         Johannes Berg <johannes@sipsolutions.net>
+With help from: Alan Cox, Avuton Olrich, Heikki Orsila, Jan Engelhardt,
+Kyle McMartin, Kyle Moffett, Randy Dunlap, Robert Hancock, Uli Kunitz,
+Vadim Lobanov
 
diff --git a/arch/Kconfig b/arch/Kconfig
index 6093c0be58b0..b0fabfa864ff 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -27,6 +27,25 @@ config KPROBES
 	  for kernel debugging, non-intrusive instrumentation and testing.
 	  If in doubt, say "N".
 
+config HAVE_EFFICIENT_UNALIGNED_ACCESS
+	def_bool n
+	help
+	  Some architectures are unable to perform unaligned accesses
+	  without the use of get_unaligned/put_unaligned. Others are
+	  unable to perform such accesses efficiently (e.g. trap on
+	  unaligned access and require fixing it up in the exception
+	  handler.)
+
+	  This symbol should be selected by an architecture if it can
+	  perform unaligned accesses efficiently to allow different
+	  code paths to be selected for these cases. Some network
+	  drivers, for example, could opt to not fix up alignment
+	  problems with received packets if doing so would not help
+	  much.
+
+	  See Documentation/unaligned-memory-access.txt for more
+	  information on the topic of unaligned memory accesses.
+
 config KRETPROBES
 	def_bool y
 	depends on KPROBES && HAVE_KRETPROBES
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index a487671c282f..de6b49cd6be8 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -112,6 +112,7 @@ config PPC
 	select HAVE_FTRACE
 	select HAVE_IDE
 	select HAVE_IOREMAP_PROT
+	select HAVE_EFFICIENT_UNALIGNED_ACCESS
 	select HAVE_KPROBES
 	select HAVE_ARCH_KGDB
 	select HAVE_KRETPROBES
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index b2ddfcf01728..66f3ab05b18c 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -28,6 +28,7 @@ config X86
 	select HAVE_FTRACE
 	select HAVE_KVM if ((X86_32 && !X86_VOYAGER && !X86_VISWS && !X86_NUMAQ) || X86_64)
 	select HAVE_ARCH_KGDB if !X86_VOYAGER
+	select HAVE_EFFICIENT_UNALIGNED_ACCESS
 
 config ARCH_DEFCONFIG
 	string

From 2fc9c4e18f94431e7eb77d97edb2a995b46fba55 Mon Sep 17 00:00:00 2001
From: Vegard Nossum <vegard.nossum@gmail.com>
Date: Fri, 25 Jul 2008 01:45:34 -0700
Subject: [PATCH 0688/1171] kallsyms: fix potential overflow in binary search

This will probably never trigger... but it won't hurt to be careful.

http://googleresearch.blogspot.com/2006/06/extra-extra-read-all-about-it-nearly.html

Signed-off-by: Vegard Nossum <vegard.nossum@gmail.com>
Cc: Joshua Bloch <jjb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/kallsyms.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/kallsyms.c b/kernel/kallsyms.c
index 6fc0040f3e3a..38fc10ac7541 100644
--- a/kernel/kallsyms.c
+++ b/kernel/kallsyms.c
@@ -176,7 +176,7 @@ static unsigned long get_symbol_pos(unsigned long addr,
 	high = kallsyms_num_syms;
 
 	while (high - low > 1) {
-		mid = (low + high) / 2;
+		mid = low + (high - low) / 2;
 		if (kallsyms_addresses[mid] <= addr)
 			low = mid;
 		else

From 696adfe84c11c571a1e0863460ff0ec142b4e5a9 Mon Sep 17 00:00:00 2001
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Date: Fri, 25 Jul 2008 01:45:34 -0700
Subject: [PATCH 0689/1171] list_for_each_rcu must die: networking

All uses of list_for_each_rcu() can be profitably replaced by the
easier-to-use list_for_each_entry_rcu().  This patch makes this change for
networking, in preparation for removing the list_for_each_rcu() API
entirely.

Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 net/802/psnap.c     | 4 +---
 net/ipv4/af_inet.c  | 9 +++------
 net/ipv6/af_inet6.c | 9 +++------
 3 files changed, 7 insertions(+), 15 deletions(-)

diff --git a/net/802/psnap.c b/net/802/psnap.c
index ea4643931446..b3cfe5a14fca 100644
--- a/net/802/psnap.c
+++ b/net/802/psnap.c
@@ -31,11 +31,9 @@ static struct llc_sap *snap_sap;
  */
 static struct datalink_proto *find_snap_client(unsigned char *desc)
 {
-	struct list_head *entry;
 	struct datalink_proto *proto = NULL, *p;
 
-	list_for_each_rcu(entry, &snap_list) {
-		p = list_entry(entry, struct datalink_proto, node);
+	list_for_each_entry_rcu(p, &snap_list, node) {
 		if (!memcmp(p->type, desc, 5)) {
 			proto = p;
 			break;
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index dd919d84285f..f440a9f54924 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -264,7 +264,6 @@ static inline int inet_netns_ok(struct net *net, int protocol)
 static int inet_create(struct net *net, struct socket *sock, int protocol)
 {
 	struct sock *sk;
-	struct list_head *p;
 	struct inet_protosw *answer;
 	struct inet_sock *inet;
 	struct proto *answer_prot;
@@ -281,13 +280,12 @@ static int inet_create(struct net *net, struct socket *sock, int protocol)
 	sock->state = SS_UNCONNECTED;
 
 	/* Look for the requested type/protocol pair. */
-	answer = NULL;
 lookup_protocol:
 	err = -ESOCKTNOSUPPORT;
 	rcu_read_lock();
-	list_for_each_rcu(p, &inetsw[sock->type]) {
-		answer = list_entry(p, struct inet_protosw, list);
+	list_for_each_entry_rcu(answer, &inetsw[sock->type], list) {
 
+		err = 0;
 		/* Check the non-wild match. */
 		if (protocol == answer->protocol) {
 			if (protocol != IPPROTO_IP)
@@ -302,10 +300,9 @@ lookup_protocol:
 				break;
 		}
 		err = -EPROTONOSUPPORT;
-		answer = NULL;
 	}
 
-	if (unlikely(answer == NULL)) {
+	if (unlikely(err)) {
 		if (try_loading_module < 2) {
 			rcu_read_unlock();
 			/*
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 3d828bc4b1cf..60461ad7fa6f 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -83,7 +83,6 @@ static int inet6_create(struct net *net, struct socket *sock, int protocol)
 	struct inet_sock *inet;
 	struct ipv6_pinfo *np;
 	struct sock *sk;
-	struct list_head *p;
 	struct inet_protosw *answer;
 	struct proto *answer_prot;
 	unsigned char answer_flags;
@@ -97,13 +96,12 @@ static int inet6_create(struct net *net, struct socket *sock, int protocol)
 		build_ehash_secret();
 
 	/* Look for the requested type/protocol pair. */
-	answer = NULL;
 lookup_protocol:
 	err = -ESOCKTNOSUPPORT;
 	rcu_read_lock();
-	list_for_each_rcu(p, &inetsw6[sock->type]) {
-		answer = list_entry(p, struct inet_protosw, list);
+	list_for_each_entry_rcu(answer, &inetsw6[sock->type], list) {
 
+		err = 0;
 		/* Check the non-wild match. */
 		if (protocol == answer->protocol) {
 			if (protocol != IPPROTO_IP)
@@ -118,10 +116,9 @@ lookup_protocol:
 				break;
 		}
 		err = -EPROTONOSUPPORT;
-		answer = NULL;
 	}
 
-	if (!answer) {
+	if (err) {
 		if (try_loading_module < 2) {
 			rcu_read_unlock();
 			/*

From b03f6489f9f27dc519a4c60ebf39cc7b8a58eae7 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 01:45:35 -0700
Subject: [PATCH 0690/1171] build kernel/profile.o only when requested

Build kernel/profile.o only if CONFIG_PROFILING is enabled.

This makes CONFIG_PROFILING=n kernels smaller.

As a bonus, some profile_tick() calls and one branch from schedule() are
now eliminated with CONFIG_PROFILING=n (but I doubt these are
measurable effects).

This patch changes the effects of CONFIG_PROFILING=n, but I don't think
having more than two choices would be the better choice.

This patch also adds the name of the first parameter to the prototypes
of profile_{hits,tick}() since I anyway had to add them for the dummy
functions.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/profile.h | 56 ++++++++++++++++++++++++++++-------------
 kernel/Makefile         |  3 ++-
 kernel/profile.c        |  4 ---
 3 files changed, 41 insertions(+), 22 deletions(-)

diff --git a/include/linux/profile.h b/include/linux/profile.h
index 05c1cc736937..4081fa31081f 100644
--- a/include/linux/profile.h
+++ b/include/linux/profile.h
@@ -8,8 +8,6 @@
 
 #include <asm/errno.h>
 
-extern int prof_on __read_mostly;
-
 #define CPU_PROFILING	1
 #define SCHED_PROFILING	2
 #define SLEEP_PROFILING	3
@@ -19,14 +17,29 @@ struct proc_dir_entry;
 struct pt_regs;
 struct notifier_block;
 
+#if defined(CONFIG_PROFILING) && defined(CONFIG_PROC_FS)
+void create_prof_cpu_mask(struct proc_dir_entry *);
+#else
+#define create_prof_cpu_mask(x)			do { (void)(x); } while (0)
+#endif
+
+enum profile_type {
+	PROFILE_TASK_EXIT,
+	PROFILE_MUNMAP
+};
+
+#ifdef CONFIG_PROFILING
+
+extern int prof_on __read_mostly;
+
 /* init basic kernel profiler */
 void __init profile_init(void);
-void profile_tick(int);
+void profile_tick(int type);
 
 /*
  * Add multiple profiler hits to a given address:
  */
-void profile_hits(int, void *ip, unsigned int nr_hits);
+void profile_hits(int type, void *ip, unsigned int nr_hits);
 
 /*
  * Single profiler hit:
@@ -40,19 +53,6 @@ static inline void profile_hit(int type, void *ip)
 		profile_hits(type, ip, 1);
 }
 
-#ifdef CONFIG_PROC_FS
-void create_prof_cpu_mask(struct proc_dir_entry *);
-#else
-#define create_prof_cpu_mask(x)			do { (void)(x); } while (0)
-#endif
-
-enum profile_type {
-	PROFILE_TASK_EXIT,
-	PROFILE_MUNMAP
-};
-
-#ifdef CONFIG_PROFILING
-
 struct task_struct;
 struct mm_struct;
 
@@ -80,6 +80,28 @@ struct pt_regs;
 
 #else
 
+#define prof_on 0
+
+static inline void profile_init(void)
+{
+	return;
+}
+
+static inline void profile_tick(int type)
+{
+	return;
+}
+
+static inline void profile_hits(int type, void *ip, unsigned int nr_hits)
+{
+	return;
+}
+
+static inline void profile_hit(int type, void *ip)
+{
+	return;
+}
+
 static inline int task_handoff_register(struct notifier_block * n)
 {
 	return -ENOSYS;
diff --git a/kernel/Makefile b/kernel/Makefile
index 15ab63ffe64d..54f69837d35a 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -2,7 +2,7 @@
 # Makefile for the linux kernel.
 #
 
-obj-y     = sched.o fork.o exec_domain.o panic.o printk.o profile.o \
+obj-y     = sched.o fork.o exec_domain.o panic.o printk.o \
 	    cpu.o exit.o itimer.o time.o softirq.o resource.o \
 	    sysctl.o capability.o ptrace.o timer.o user.o \
 	    signal.o sys.o kmod.o workqueue.o pid.o \
@@ -24,6 +24,7 @@ CFLAGS_REMOVE_sched_clock.o = -pg
 CFLAGS_REMOVE_sched.o = -mno-spe -pg
 endif
 
+obj-$(CONFIG_PROFILING) += profile.o
 obj-$(CONFIG_SYSCTL_SYSCALL_CHECK) += sysctl_check.o
 obj-$(CONFIG_STACKTRACE) += stacktrace.o
 obj-y += time/
diff --git a/kernel/profile.c b/kernel/profile.c
index 58926411eb2a..cd26bed4cc26 100644
--- a/kernel/profile.c
+++ b/kernel/profile.c
@@ -112,8 +112,6 @@ void __init profile_init(void)
 
 /* Profile event notifications */
 
-#ifdef CONFIG_PROFILING
-
 static BLOCKING_NOTIFIER_HEAD(task_exit_notifier);
 static ATOMIC_NOTIFIER_HEAD(task_free_notifier);
 static BLOCKING_NOTIFIER_HEAD(munmap_notifier);
@@ -203,8 +201,6 @@ void unregister_timer_hook(int (*hook)(struct pt_regs *))
 }
 EXPORT_SYMBOL_GPL(unregister_timer_hook);
 
-#endif /* CONFIG_PROFILING */
-
 
 #ifdef CONFIG_SMP
 /*

From cebbd3fb803603b12408458ba17c29ce1e15a5f2 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@linux-foundation.org>
Date: Fri, 25 Jul 2008 01:45:35 -0700
Subject: [PATCH 0691/1171] build-kernel-profileo-only-when-requested-cleanups

Cc: Adrian Bunk <bunk@kernel.org>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/profile.h | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/include/linux/profile.h b/include/linux/profile.h
index 4081fa31081f..7e7087239af5 100644
--- a/include/linux/profile.h
+++ b/include/linux/profile.h
@@ -18,9 +18,11 @@ struct pt_regs;
 struct notifier_block;
 
 #if defined(CONFIG_PROFILING) && defined(CONFIG_PROC_FS)
-void create_prof_cpu_mask(struct proc_dir_entry *);
+void create_prof_cpu_mask(struct proc_dir_entry *de);
 #else
-#define create_prof_cpu_mask(x)			do { (void)(x); } while (0)
+static inline void create_prof_cpu_mask(struct proc_dir_entry *de)
+{
+}
 #endif
 
 enum profile_type {

From f16695f4ac088cf7593e113574046d2d7e5af5eb Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 01:45:36 -0700
Subject: [PATCH 0692/1171] asm-generic/int-ll64.h: always provide __{s,u}64

Several compilers offer "long long" without claiming to support C99.

Considering how frequent __s64/__u64 are used our userspace headers are
anyway unusable without __s64/__u64 available.

Always offer __s64/__u64 to non-gcc non-C99 compilers - if they provide
"long long" that makes the headers compiling and if they don't they are
anyway screwed.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Acked-by: H. Peter Anvin <hpa@zytor.com>
Cc: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/asm-generic/int-ll64.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/asm-generic/int-ll64.h b/include/asm-generic/int-ll64.h
index 260948905e4e..f9bc9ac29b36 100644
--- a/include/asm-generic/int-ll64.h
+++ b/include/asm-generic/int-ll64.h
@@ -26,7 +26,7 @@ typedef unsigned int __u32;
 #ifdef __GNUC__
 __extension__ typedef __signed__ long long __s64;
 __extension__ typedef unsigned long long __u64;
-#elif defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L
+#else
 typedef __signed__ long long __s64;
 typedef unsigned long long __u64;
 #endif

From f557d0996a6f9c06912528ea85e1dba0fb7d485f Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 01:45:37 -0700
Subject: [PATCH 0693/1171] remove some more tipar bits

Some bits were missed when the tipar driver was removed.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/00-INDEX              | 2 --
 Documentation/kernel-parameters.txt | 7 -------
 drivers/char/Makefile               | 1 -
 3 files changed, 10 deletions(-)

diff --git a/Documentation/00-INDEX b/Documentation/00-INDEX
index 1977fab38656..6de71308a906 100644
--- a/Documentation/00-INDEX
+++ b/Documentation/00-INDEX
@@ -361,8 +361,6 @@ telephony/
 	- directory with info on telephony (e.g. voice over IP) support.
 time_interpolators.txt
 	- info on time interpolators.
-tipar.txt
-	- information about Parallel link cable for Texas Instruments handhelds.
 tty.txt
 	- guide to the locking policies of the tty layer.
 uml/
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 497a98dafdaa..e7bea3e85304 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -2159,13 +2159,6 @@ and is between 256 and 4096 characters. It is defined in the file
 			<deci-seconds>: poll all this frequency
 			0: no polling (default)
 
-	tipar.timeout=	[HW,PPT]
-			Set communications timeout in tenths of a second
-			(default 15).
-
-	tipar.delay=	[HW,PPT]
-			Set inter-bit delay in microseconds (default 10).
-
 	tmscsim=	[HW,SCSI]
 			See comment before function dc390_setup() in
 			drivers/scsi/tmscsim.c.
diff --git a/drivers/char/Makefile b/drivers/char/Makefile
index eb02c3506800..f7a0d1a754fc 100644
--- a/drivers/char/Makefile
+++ b/drivers/char/Makefile
@@ -64,7 +64,6 @@ obj-$(CONFIG_BRIQ_PANEL)	+= briq_panel.o
 obj-$(CONFIG_BFIN_OTP)		+= bfin-otp.o
 
 obj-$(CONFIG_PRINTER)		+= lp.o
-obj-$(CONFIG_TIPAR)		+= tipar.o
 
 obj-$(CONFIG_APM_EMULATION)	+= apm-emulation.o
 

From ac331d158e198d2a91a5b0a3ec4ca9991fdb57af Mon Sep 17 00:00:00 2001
From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Date: Fri, 25 Jul 2008 01:45:38 -0700
Subject: [PATCH 0694/1171] call_usermodehelper(): increase reliability

Presently call_usermodehelper_setup() uses GFP_ATOMIC.  but it can return
NULL _very_ easily.

GFP_ATOMIC is needed only when we can't sleep.  and, GFP_KERNEL is robust
and better.

thus, I add gfp_mask argument to call_usermodehelper_setup().

So, its callers pass the gfp_t as below:

call_usermodehelper() and call_usermodehelper_keys():
	depend on 'wait' argument.
call_usermodehelper_pipe():
	always GFP_KERNEL because always run under process context.
orderly_poweroff():
	pass to GFP_ATOMIC because may run under interrupt context.

Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: "Paul Menage" <menage@google.com>
Reviewed-by: Li Zefan <lizf@cn.fujitsu.com>
Acked-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/kmod.h | 11 +++++++----
 kernel/kmod.c        |  9 +++++----
 kernel/sys.c         |  2 +-
 3 files changed, 13 insertions(+), 9 deletions(-)

diff --git a/include/linux/kmod.h b/include/linux/kmod.h
index 0509c4ce4857..a1a91577813c 100644
--- a/include/linux/kmod.h
+++ b/include/linux/kmod.h
@@ -19,6 +19,7 @@
  *      Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  */
 
+#include <linux/gfp.h>
 #include <linux/stddef.h>
 #include <linux/errno.h>
 #include <linux/compiler.h>
@@ -41,8 +42,8 @@ struct file;
 struct subprocess_info;
 
 /* Allocate a subprocess_info structure */
-struct subprocess_info *call_usermodehelper_setup(char *path,
-						  char **argv, char **envp);
+struct subprocess_info *call_usermodehelper_setup(char *path, char **argv,
+						  char **envp, gfp_t gfp_mask);
 
 /* Set various pieces of state into the subprocess_info structure */
 void call_usermodehelper_setkeys(struct subprocess_info *info,
@@ -69,8 +70,9 @@ static inline int
 call_usermodehelper(char *path, char **argv, char **envp, enum umh_wait wait)
 {
 	struct subprocess_info *info;
+	gfp_t gfp_mask = (wait == UMH_NO_WAIT) ? GFP_ATOMIC : GFP_KERNEL;
 
-	info = call_usermodehelper_setup(path, argv, envp);
+	info = call_usermodehelper_setup(path, argv, envp, gfp_mask);
 	if (info == NULL)
 		return -ENOMEM;
 	return call_usermodehelper_exec(info, wait);
@@ -81,8 +83,9 @@ call_usermodehelper_keys(char *path, char **argv, char **envp,
 			 struct key *session_keyring, enum umh_wait wait)
 {
 	struct subprocess_info *info;
+	gfp_t gfp_mask = (wait == UMH_NO_WAIT) ? GFP_ATOMIC : GFP_KERNEL;
 
-	info = call_usermodehelper_setup(path, argv, envp);
+	info = call_usermodehelper_setup(path, argv, envp, gfp_mask);
 	if (info == NULL)
 		return -ENOMEM;
 
diff --git a/kernel/kmod.c b/kernel/kmod.c
index 2989f67c4446..2456d1a0befb 100644
--- a/kernel/kmod.c
+++ b/kernel/kmod.c
@@ -352,16 +352,17 @@ static inline void register_pm_notifier_callback(void) {}
  * @path: path to usermode executable
  * @argv: arg vector for process
  * @envp: environment for process
+ * @gfp_mask: gfp mask for memory allocation
  *
  * Returns either %NULL on allocation failure, or a subprocess_info
  * structure.  This should be passed to call_usermodehelper_exec to
  * exec the process and free the structure.
  */
-struct subprocess_info *call_usermodehelper_setup(char *path,
-						  char **argv, char **envp)
+struct subprocess_info *call_usermodehelper_setup(char *path, char **argv,
+						  char **envp, gfp_t gfp_mask)
 {
 	struct subprocess_info *sub_info;
-	sub_info = kzalloc(sizeof(struct subprocess_info),  GFP_ATOMIC);
+	sub_info = kzalloc(sizeof(struct subprocess_info), gfp_mask);
 	if (!sub_info)
 		goto out;
 
@@ -494,7 +495,7 @@ int call_usermodehelper_pipe(char *path, char **argv, char **envp,
 	struct subprocess_info *sub_info;
 	int ret;
 
-	sub_info = call_usermodehelper_setup(path, argv, envp);
+	sub_info = call_usermodehelper_setup(path, argv, envp, GFP_KERNEL);
 	if (sub_info == NULL)
 		return -ENOMEM;
 
diff --git a/kernel/sys.c b/kernel/sys.c
index 14e97282eb6c..6c2188046048 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -1795,7 +1795,7 @@ int orderly_poweroff(bool force)
 		goto out;
 	}
 
-	info = call_usermodehelper_setup(argv[0], argv, envp);
+	info = call_usermodehelper_setup(argv[0], argv, envp, GFP_ATOMIC);
 	if (info == NULL) {
 		argv_free(argv);
 		goto out;

From 62ec30d45ecbb85b5991474c8f04192697687495 Mon Sep 17 00:00:00 2001
From: Matthew Garrett <mjg59@srcf.ucam.org>
Date: Fri, 25 Jul 2008 01:45:39 -0700
Subject: [PATCH 0695/1171] misc: add HP WMI laptop extras driver

This driver adds support for reading and configuring certain information
on modern HP laptops with WMI BIOS interfaces.  It supports enabling and
disabling the ambient light sensor, querying attached displays and hard
drive temperature, sending events on docking and querying the state of the
dock and toggling the state of the wifi, bluetooth and wwan hardware via
rfkill.  It also makes the little "(i)" button work on machines that send
that via WMI rather than via the keyboard controller.

Signed-off-by: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/misc/Kconfig  |  12 +
 drivers/misc/Makefile |   1 +
 drivers/misc/hp-wmi.c | 494 ++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 507 insertions(+)
 create mode 100644 drivers/misc/hp-wmi.c

diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig
index d5bc288b1b0d..1689c051f68f 100644
--- a/drivers/misc/Kconfig
+++ b/drivers/misc/Kconfig
@@ -212,6 +212,18 @@ config TC1100_WMI
 	  This is a driver for the WMI extensions (wireless and bluetooth power
 	  control) of the HP Compaq TC1100 tablet.
 
+config HP_WMI
+       tristate "HP WMI extras"
+       depends on ACPI_WMI
+       depends on INPUT
+       depends on RFKILL
+       help
+         Say Y here if you want to support WMI-based hotkeys on HP laptops and
+	 to read data from WMI such as docking or ambient light sensor state.
+
+         To compile this driver as a module, choose M here: the module will
+         be called hp-wmi.
+
 config MSI_LAPTOP
         tristate "MSI Laptop Extras"
         depends on X86
diff --git a/drivers/misc/Makefile b/drivers/misc/Makefile
index 688fe76135e0..f5e273420c09 100644
--- a/drivers/misc/Makefile
+++ b/drivers/misc/Makefile
@@ -13,6 +13,7 @@ obj-$(CONFIG_ACER_WMI)		+= acer-wmi.o
 obj-$(CONFIG_ATMEL_PWM)		+= atmel_pwm.o
 obj-$(CONFIG_ATMEL_SSC)		+= atmel-ssc.o
 obj-$(CONFIG_ATMEL_TCLIB)	+= atmel_tclib.o
+obj-$(CONFIG_HP_WMI)		+= hp-wmi.o
 obj-$(CONFIG_TC1100_WMI)	+= tc1100-wmi.o
 obj-$(CONFIG_LKDTM)		+= lkdtm.o
 obj-$(CONFIG_TIFM_CORE)       	+= tifm_core.o
diff --git a/drivers/misc/hp-wmi.c b/drivers/misc/hp-wmi.c
new file mode 100644
index 000000000000..1dbcbcb323a2
--- /dev/null
+++ b/drivers/misc/hp-wmi.c
@@ -0,0 +1,494 @@
+/*
+ * HP WMI hotkeys
+ *
+ * Copyright (C) 2008 Red Hat <mjg@redhat.com>
+ *
+ * Portions based on wistron_btns.c:
+ * Copyright (C) 2005 Miloslav Trmac <mitr@volny.cz>
+ * Copyright (C) 2005 Bernhard Rosenkraenzer <bero@arklinux.org>
+ * Copyright (C) 2005 Dmitry Torokhov <dtor@mail.ru>
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; either version 2 of the License, or
+ *  (at your option) any later version.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/types.h>
+#include <linux/input.h>
+#include <acpi/acpi_drivers.h>
+#include <linux/platform_device.h>
+#include <linux/acpi.h>
+#include <linux/rfkill.h>
+#include <linux/string.h>
+
+MODULE_AUTHOR("Matthew Garrett <mjg59@srcf.ucam.org>");
+MODULE_DESCRIPTION("HP laptop WMI hotkeys driver");
+MODULE_LICENSE("GPL");
+
+MODULE_ALIAS("wmi:95F24279-4D7B-4334-9387-ACCDC67EF61C");
+MODULE_ALIAS("wmi:5FB7F034-2C63-45e9-BE91-3D44E2C707E4");
+
+#define HPWMI_EVENT_GUID "95F24279-4D7B-4334-9387-ACCDC67EF61C"
+#define HPWMI_BIOS_GUID "5FB7F034-2C63-45e9-BE91-3D44E2C707E4"
+
+#define HPWMI_DISPLAY_QUERY 0x1
+#define HPWMI_HDDTEMP_QUERY 0x2
+#define HPWMI_ALS_QUERY 0x3
+#define HPWMI_DOCK_QUERY 0x4
+#define HPWMI_WIRELESS_QUERY 0x5
+
+static int __init hp_wmi_bios_setup(struct platform_device *device);
+static int __exit hp_wmi_bios_remove(struct platform_device *device);
+
+struct bios_args {
+	u32 signature;
+	u32 command;
+	u32 commandtype;
+	u32 datasize;
+	u32 data;
+};
+
+struct bios_return {
+	u32 sigpass;
+	u32 return_code;
+	u32 value;
+};
+
+struct key_entry {
+	char type;		/* See KE_* below */
+	u8 code;
+	u16 keycode;
+};
+
+enum { KE_KEY, KE_SW, KE_END };
+
+static struct key_entry hp_wmi_keymap[] = {
+	{KE_SW, 0x01, SW_DOCK},
+	{KE_KEY, 0x02, KEY_BRIGHTNESSUP},
+	{KE_KEY, 0x03, KEY_BRIGHTNESSDOWN},
+	{KE_KEY, 0x04, KEY_HELP},
+	{KE_END, 0}
+};
+
+static struct input_dev *hp_wmi_input_dev;
+static struct platform_device *hp_wmi_platform_dev;
+
+static struct rfkill *wifi_rfkill;
+static struct rfkill *bluetooth_rfkill;
+static struct rfkill *wwan_rfkill;
+
+static struct platform_driver hp_wmi_driver = {
+	.driver = {
+		   .name = "hp-wmi",
+		   .owner = THIS_MODULE,
+	},
+	.probe = hp_wmi_bios_setup,
+	.remove = hp_wmi_bios_remove,
+};
+
+static int hp_wmi_perform_query(int query, int write, int value)
+{
+	struct bios_return bios_return;
+	acpi_status status;
+	union acpi_object *obj;
+	struct bios_args args = {
+		.signature = 0x55434553,
+		.command = write ? 0x2 : 0x1,
+		.commandtype = query,
+		.datasize = write ? 0x4 : 0,
+		.data = value,
+	};
+	struct acpi_buffer input = { sizeof(struct bios_args), &args };
+	struct acpi_buffer output = { ACPI_ALLOCATE_BUFFER, NULL };
+
+	status = wmi_evaluate_method(HPWMI_BIOS_GUID, 0, 0x3, &input, &output);
+
+	obj = output.pointer;
+
+	if (!obj || obj->type != ACPI_TYPE_BUFFER)
+		return -EINVAL;
+
+	bios_return = *((struct bios_return *)obj->buffer.pointer);
+	if (bios_return.return_code > 0)
+		return bios_return.return_code * -1;
+	else
+		return bios_return.value;
+}
+
+static int hp_wmi_display_state(void)
+{
+	return hp_wmi_perform_query(HPWMI_DISPLAY_QUERY, 0, 0);
+}
+
+static int hp_wmi_hddtemp_state(void)
+{
+	return hp_wmi_perform_query(HPWMI_HDDTEMP_QUERY, 0, 0);
+}
+
+static int hp_wmi_als_state(void)
+{
+	return hp_wmi_perform_query(HPWMI_ALS_QUERY, 0, 0);
+}
+
+static int hp_wmi_dock_state(void)
+{
+	return hp_wmi_perform_query(HPWMI_DOCK_QUERY, 0, 0);
+}
+
+static int hp_wmi_wifi_set(void *data, enum rfkill_state state)
+{
+	if (state)
+		return hp_wmi_perform_query(HPWMI_WIRELESS_QUERY, 1, 0x101);
+	else
+		return hp_wmi_perform_query(HPWMI_WIRELESS_QUERY, 1, 0x100);
+}
+
+static int hp_wmi_bluetooth_set(void *data, enum rfkill_state state)
+{
+	if (state)
+		return hp_wmi_perform_query(HPWMI_WIRELESS_QUERY, 1, 0x202);
+	else
+		return hp_wmi_perform_query(HPWMI_WIRELESS_QUERY, 1, 0x200);
+}
+
+static int hp_wmi_wwan_set(void *data, enum rfkill_state state)
+{
+	if (state)
+		return hp_wmi_perform_query(HPWMI_WIRELESS_QUERY, 1, 0x404);
+	else
+		return hp_wmi_perform_query(HPWMI_WIRELESS_QUERY, 1, 0x400);
+}
+
+static int hp_wmi_wifi_state(void)
+{
+	int wireless = hp_wmi_perform_query(HPWMI_WIRELESS_QUERY, 0, 0);
+
+	if (wireless & 0x100)
+		return 1;
+	else
+		return 0;
+}
+
+static int hp_wmi_bluetooth_state(void)
+{
+	int wireless = hp_wmi_perform_query(HPWMI_WIRELESS_QUERY, 0, 0);
+
+	if (wireless & 0x10000)
+		return 1;
+	else
+		return 0;
+}
+
+static int hp_wmi_wwan_state(void)
+{
+	int wireless = hp_wmi_perform_query(HPWMI_WIRELESS_QUERY, 0, 0);
+
+	if (wireless & 0x1000000)
+		return 1;
+	else
+		return 0;
+}
+
+static ssize_t show_display(struct device *dev, struct device_attribute *attr,
+			    char *buf)
+{
+	int value = hp_wmi_display_state();
+	if (value < 0)
+		return -EINVAL;
+	return sprintf(buf, "%d\n", value);
+}
+
+static ssize_t show_hddtemp(struct device *dev, struct device_attribute *attr,
+			    char *buf)
+{
+	int value = hp_wmi_hddtemp_state();
+	if (value < 0)
+		return -EINVAL;
+	return sprintf(buf, "%d\n", value);
+}
+
+static ssize_t show_als(struct device *dev, struct device_attribute *attr,
+			char *buf)
+{
+	int value = hp_wmi_als_state();
+	if (value < 0)
+		return -EINVAL;
+	return sprintf(buf, "%d\n", value);
+}
+
+static ssize_t show_dock(struct device *dev, struct device_attribute *attr,
+			 char *buf)
+{
+	int value = hp_wmi_dock_state();
+	if (value < 0)
+		return -EINVAL;
+	return sprintf(buf, "%d\n", value);
+}
+
+static ssize_t set_als(struct device *dev, struct device_attribute *attr,
+		       const char *buf, size_t count)
+{
+	u32 tmp = simple_strtoul(buf, NULL, 10);
+	hp_wmi_perform_query(HPWMI_ALS_QUERY, 1, tmp);
+	return count;
+}
+
+static DEVICE_ATTR(display, S_IRUGO, show_display, NULL);
+static DEVICE_ATTR(hddtemp, S_IRUGO, show_hddtemp, NULL);
+static DEVICE_ATTR(als, S_IRUGO | S_IWUSR, show_als, set_als);
+static DEVICE_ATTR(dock, S_IRUGO, show_dock, NULL);
+
+static struct key_entry *hp_wmi_get_entry_by_scancode(int code)
+{
+	struct key_entry *key;
+
+	for (key = hp_wmi_keymap; key->type != KE_END; key++)
+		if (code == key->code)
+			return key;
+
+	return NULL;
+}
+
+static struct key_entry *hp_wmi_get_entry_by_keycode(int keycode)
+{
+	struct key_entry *key;
+
+	for (key = hp_wmi_keymap; key->type != KE_END; key++)
+		if (key->type == KE_KEY && keycode == key->keycode)
+			return key;
+
+	return NULL;
+}
+
+static int hp_wmi_getkeycode(struct input_dev *dev, int scancode, int *keycode)
+{
+	struct key_entry *key = hp_wmi_get_entry_by_scancode(scancode);
+
+	if (key && key->type == KE_KEY) {
+		*keycode = key->keycode;
+		return 0;
+	}
+
+	return -EINVAL;
+}
+
+static int hp_wmi_setkeycode(struct input_dev *dev, int scancode, int keycode)
+{
+	struct key_entry *key;
+	int old_keycode;
+
+	if (keycode < 0 || keycode > KEY_MAX)
+		return -EINVAL;
+
+	key = hp_wmi_get_entry_by_scancode(scancode);
+	if (key && key->type == KE_KEY) {
+		old_keycode = key->keycode;
+		key->keycode = keycode;
+		set_bit(keycode, dev->keybit);
+		if (!hp_wmi_get_entry_by_keycode(old_keycode))
+			clear_bit(old_keycode, dev->keybit);
+		return 0;
+	}
+
+	return -EINVAL;
+}
+
+void hp_wmi_notify(u32 value, void *context)
+{
+	struct acpi_buffer response = { ACPI_ALLOCATE_BUFFER, NULL };
+	static struct key_entry *key;
+	union acpi_object *obj;
+
+	wmi_get_event_data(value, &response);
+
+	obj = (union acpi_object *)response.pointer;
+
+	if (obj && obj->type == ACPI_TYPE_BUFFER && obj->buffer.length == 8) {
+		int eventcode = *((u8 *) obj->buffer.pointer);
+		key = hp_wmi_get_entry_by_scancode(eventcode);
+		if (key) {
+			switch (key->type) {
+			case KE_KEY:
+				input_report_key(hp_wmi_input_dev,
+						 key->keycode, 1);
+				input_sync(hp_wmi_input_dev);
+				input_report_key(hp_wmi_input_dev,
+						 key->keycode, 0);
+				input_sync(hp_wmi_input_dev);
+				break;
+			case KE_SW:
+				input_report_switch(hp_wmi_input_dev,
+						    key->keycode,
+						    hp_wmi_dock_state());
+				input_sync(hp_wmi_input_dev);
+				break;
+			}
+		} else if (eventcode == 0x5) {
+			if (wifi_rfkill)
+				wifi_rfkill->state = hp_wmi_wifi_state();
+			if (bluetooth_rfkill)
+				bluetooth_rfkill->state =
+				    hp_wmi_bluetooth_state();
+			if (wwan_rfkill)
+				wwan_rfkill->state = hp_wmi_wwan_state();
+		} else
+			printk(KERN_INFO "HP WMI: Unknown key pressed - %x\n",
+			       eventcode);
+	} else
+		printk(KERN_INFO "HP WMI: Unknown response received\n");
+}
+
+static int __init hp_wmi_input_setup(void)
+{
+	struct key_entry *key;
+	int err;
+
+	hp_wmi_input_dev = input_allocate_device();
+
+	hp_wmi_input_dev->name = "HP WMI hotkeys";
+	hp_wmi_input_dev->phys = "wmi/input0";
+	hp_wmi_input_dev->id.bustype = BUS_HOST;
+	hp_wmi_input_dev->getkeycode = hp_wmi_getkeycode;
+	hp_wmi_input_dev->setkeycode = hp_wmi_setkeycode;
+
+	for (key = hp_wmi_keymap; key->type != KE_END; key++) {
+		switch (key->type) {
+		case KE_KEY:
+			set_bit(EV_KEY, hp_wmi_input_dev->evbit);
+			set_bit(key->keycode, hp_wmi_input_dev->keybit);
+			break;
+		case KE_SW:
+			set_bit(EV_SW, hp_wmi_input_dev->evbit);
+			set_bit(key->keycode, hp_wmi_input_dev->swbit);
+			break;
+		}
+	}
+
+	err = input_register_device(hp_wmi_input_dev);
+
+	if (err) {
+		input_free_device(hp_wmi_input_dev);
+		return err;
+	}
+
+	return 0;
+}
+
+static void cleanup_sysfs(struct platform_device *device)
+{
+	device_remove_file(&device->dev, &dev_attr_display);
+	device_remove_file(&device->dev, &dev_attr_hddtemp);
+	device_remove_file(&device->dev, &dev_attr_als);
+	device_remove_file(&device->dev, &dev_attr_dock);
+}
+
+static int __init hp_wmi_bios_setup(struct platform_device *device)
+{
+	int err;
+
+	err = device_create_file(&device->dev, &dev_attr_display);
+	if (err)
+		goto add_sysfs_error;
+	err = device_create_file(&device->dev, &dev_attr_hddtemp);
+	if (err)
+		goto add_sysfs_error;
+	err = device_create_file(&device->dev, &dev_attr_als);
+	if (err)
+		goto add_sysfs_error;
+	err = device_create_file(&device->dev, &dev_attr_dock);
+	if (err)
+		goto add_sysfs_error;
+
+	wifi_rfkill = rfkill_allocate(&device->dev, RFKILL_TYPE_WLAN);
+	wifi_rfkill->name = "hp-wifi";
+	wifi_rfkill->state = hp_wmi_wifi_state();
+	wifi_rfkill->toggle_radio = hp_wmi_wifi_set;
+	wifi_rfkill->user_claim_unsupported = 1;
+
+	bluetooth_rfkill = rfkill_allocate(&device->dev,
+					   RFKILL_TYPE_BLUETOOTH);
+	bluetooth_rfkill->name = "hp-bluetooth";
+	bluetooth_rfkill->state = hp_wmi_bluetooth_state();
+	bluetooth_rfkill->toggle_radio = hp_wmi_bluetooth_set;
+	bluetooth_rfkill->user_claim_unsupported = 1;
+
+	wwan_rfkill = rfkill_allocate(&device->dev, RFKILL_TYPE_WIMAX);
+	wwan_rfkill->name = "hp-wwan";
+	wwan_rfkill->state = hp_wmi_wwan_state();
+	wwan_rfkill->toggle_radio = hp_wmi_wwan_set;
+	wwan_rfkill->user_claim_unsupported = 1;
+
+	rfkill_register(wifi_rfkill);
+	rfkill_register(bluetooth_rfkill);
+	rfkill_register(wwan_rfkill);
+
+	return 0;
+add_sysfs_error:
+	cleanup_sysfs(device);
+	return err;
+}
+
+static int __exit hp_wmi_bios_remove(struct platform_device *device)
+{
+	cleanup_sysfs(device);
+
+	rfkill_unregister(wifi_rfkill);
+	rfkill_unregister(bluetooth_rfkill);
+	rfkill_unregister(wwan_rfkill);
+
+	return 0;
+}
+
+static int __init hp_wmi_init(void)
+{
+	int err;
+
+	if (wmi_has_guid(HPWMI_EVENT_GUID)) {
+		err = wmi_install_notify_handler(HPWMI_EVENT_GUID,
+						 hp_wmi_notify, NULL);
+		if (!err)
+			hp_wmi_input_setup();
+	}
+
+	if (wmi_has_guid(HPWMI_BIOS_GUID)) {
+		err = platform_driver_register(&hp_wmi_driver);
+		if (err)
+			return 0;
+		hp_wmi_platform_dev = platform_device_alloc("hp-wmi", -1);
+		if (!hp_wmi_platform_dev) {
+			platform_driver_unregister(&hp_wmi_driver);
+			return 0;
+		}
+		platform_device_add(hp_wmi_platform_dev);
+	}
+
+	return 0;
+}
+
+static void __exit hp_wmi_exit(void)
+{
+	if (wmi_has_guid(HPWMI_EVENT_GUID)) {
+		wmi_remove_notify_handler(HPWMI_EVENT_GUID);
+		input_unregister_device(hp_wmi_input_dev);
+	}
+	if (hp_wmi_platform_dev) {
+		platform_device_del(hp_wmi_platform_dev);
+		platform_driver_unregister(&hp_wmi_driver);
+	}
+}
+
+module_init(hp_wmi_init);
+module_exit(hp_wmi_exit);

From b69c49b78457f681ecfb3147bd968434ee6559c1 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Fri, 25 Jul 2008 01:45:40 -0700
Subject: [PATCH 0696/1171] clean up duplicated alloc/free_thread_info

We duplicate alloc/free_thread_info defines on many platforms (the
majority uses __get_free_pages/free_pages).  This patch defines common
defines and removes these duplicated defines.
__HAVE_ARCH_THREAD_INFO_ALLOCATOR is introduced for platforms that do
something different.

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Acked-by: Russell King <rmk+kernel@arm.linux.org.uk>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: <linux-arch@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/asm-alpha/thread_info.h     |  4 +---
 include/asm-arm/thread_info.h       | 13 -------------
 include/asm-avr32/thread_info.h     |  4 ----
 include/asm-blackfin/thread_info.h  |  5 +----
 include/asm-cris/thread_info.h      |  2 ++
 include/asm-frv/thread_info.h       |  2 ++
 include/asm-h8300/thread_info.h     |  5 +----
 include/asm-ia64/thread_info.h      |  2 ++
 include/asm-m32r/thread_info.h      |  2 ++
 include/asm-m68k/thread_info.h      |  8 +-------
 include/asm-m68knommu/thread_info.h |  4 ----
 include/asm-mips/thread_info.h      |  2 ++
 include/asm-mn10300/thread_info.h   |  2 ++
 include/asm-parisc/thread_info.h    | 10 +++-------
 include/asm-powerpc/thread_info.h   | 14 +++-----------
 include/asm-s390/thread_info.h      |  5 +----
 include/asm-sh/thread_info.h        |  2 ++
 include/asm-sparc/thread_info_32.h  |  2 ++
 include/asm-sparc/thread_info_64.h  |  2 ++
 include/asm-um/thread_info.h        | 16 +---------------
 include/asm-x86/thread_info.h       |  2 ++
 include/asm-xtensa/thread_info.h    |  5 +----
 kernel/fork.c                       | 17 +++++++++++++++++
 23 files changed, 50 insertions(+), 80 deletions(-)

diff --git a/include/asm-alpha/thread_info.h b/include/asm-alpha/thread_info.h
index fb3185196298..15fda4344424 100644
--- a/include/asm-alpha/thread_info.h
+++ b/include/asm-alpha/thread_info.h
@@ -50,10 +50,8 @@ register struct thread_info *__current_thread_info __asm__("$8");
 #define current_thread_info()  __current_thread_info
 
 /* Thread information allocation.  */
+#define THREAD_SIZE_ORDER 1
 #define THREAD_SIZE (2*PAGE_SIZE)
-#define alloc_thread_info(tsk) \
-  ((struct thread_info *) __get_free_pages(GFP_KERNEL,1))
-#define free_thread_info(ti) free_pages((unsigned long) (ti), 1)
 
 #endif /* __ASSEMBLY__ */
 
diff --git a/include/asm-arm/thread_info.h b/include/asm-arm/thread_info.h
index f5a664786311..d4be2d646160 100644
--- a/include/asm-arm/thread_info.h
+++ b/include/asm-arm/thread_info.h
@@ -97,19 +97,6 @@ static inline struct thread_info *current_thread_info(void)
 	return (struct thread_info *)(sp & ~(THREAD_SIZE - 1));
 }
 
-/* thread information allocation */
-#ifdef CONFIG_DEBUG_STACK_USAGE
-#define alloc_thread_info(tsk) \
-	((struct thread_info *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, \
-		THREAD_SIZE_ORDER))
-#else
-#define alloc_thread_info(tsk) \
-	((struct thread_info *)__get_free_pages(GFP_KERNEL, THREAD_SIZE_ORDER))
-#endif
-
-#define free_thread_info(info) \
-	free_pages((unsigned long)info, THREAD_SIZE_ORDER);
-
 #define thread_saved_pc(tsk)	\
 	((unsigned long)(pc_pointer(task_thread_info(tsk)->cpu_context.pc)))
 #define thread_saved_fp(tsk)	\
diff --git a/include/asm-avr32/thread_info.h b/include/asm-avr32/thread_info.h
index df68631b7b27..294b25f9323d 100644
--- a/include/asm-avr32/thread_info.h
+++ b/include/asm-avr32/thread_info.h
@@ -61,10 +61,6 @@ static inline struct thread_info *current_thread_info(void)
 	return (struct thread_info *)addr;
 }
 
-/* thread information allocation */
-#define alloc_thread_info(ti) \
-	((struct thread_info *) __get_free_pages(GFP_KERNEL, THREAD_SIZE_ORDER))
-#define free_thread_info(ti) free_pages((unsigned long)(ti), 1)
 #define get_thread_info(ti) get_task_struct((ti)->task)
 #define put_thread_info(ti) put_task_struct((ti)->task)
 
diff --git a/include/asm-blackfin/thread_info.h b/include/asm-blackfin/thread_info.h
index bc2fe5accf20..642769329d12 100644
--- a/include/asm-blackfin/thread_info.h
+++ b/include/asm-blackfin/thread_info.h
@@ -42,6 +42,7 @@
 /*
  * Size of kernel stack for each process. This must be a power of 2...
  */
+#define THREAD_SIZE_ORDER	1
 #define THREAD_SIZE		8192	/* 2 pages */
 
 #ifndef __ASSEMBLY__
@@ -94,10 +95,6 @@ static inline struct thread_info *current_thread_info(void)
 	return (struct thread_info *)((long)ti & ~((long)THREAD_SIZE-1));
 }
 
-/* thread information allocation */
-#define alloc_thread_info(tsk) ((struct thread_info *) \
-				__get_free_pages(GFP_KERNEL, 1))
-#define free_thread_info(ti)	free_pages((unsigned long) (ti), 1)
 #endif				/* __ASSEMBLY__ */
 
 /*
diff --git a/include/asm-cris/thread_info.h b/include/asm-cris/thread_info.h
index 784668ab0fa2..7efe1000f99d 100644
--- a/include/asm-cris/thread_info.h
+++ b/include/asm-cris/thread_info.h
@@ -11,6 +11,8 @@
 
 #ifdef __KERNEL__
 
+#define __HAVE_ARCH_THREAD_INFO_ALLOCATOR
+
 #ifndef __ASSEMBLY__
 #include <asm/types.h>
 #include <asm/processor.h>
diff --git a/include/asm-frv/thread_info.h b/include/asm-frv/thread_info.h
index 348b8f1df17e..b7ac6bf2844c 100644
--- a/include/asm-frv/thread_info.h
+++ b/include/asm-frv/thread_info.h
@@ -82,6 +82,8 @@ register struct thread_info *__current_thread_info asm("gr15");
 
 #define current_thread_info() ({ __current_thread_info; })
 
+#define __HAVE_ARCH_THREAD_INFO_ALLOCATOR
+
 /* thread information allocation */
 #ifdef CONFIG_DEBUG_STACK_USAGE
 #define alloc_thread_info(tsk)					\
diff --git a/include/asm-h8300/thread_info.h b/include/asm-h8300/thread_info.h
index 27bb95e2944c..aafd4d322ec3 100644
--- a/include/asm-h8300/thread_info.h
+++ b/include/asm-h8300/thread_info.h
@@ -49,6 +49,7 @@ struct thread_info {
 /*
  * Size of kernel stack for each process. This must be a power of 2...
  */
+#define THREAD_SIZE_ORDER	1
 #define THREAD_SIZE		8192	/* 2 pages */
 
 
@@ -65,10 +66,6 @@ static inline struct thread_info *current_thread_info(void)
 	return ti;
 }
 
-/* thread information allocation */
-#define alloc_thread_info(tsk) ((struct thread_info *) \
-				__get_free_pages(GFP_KERNEL, 1))
-#define free_thread_info(ti)	free_pages((unsigned long) (ti), 1)
 #endif /* __ASSEMBLY__ */
 
 /*
diff --git a/include/asm-ia64/thread_info.h b/include/asm-ia64/thread_info.h
index 2422ac61658a..7c60fcdd2efd 100644
--- a/include/asm-ia64/thread_info.h
+++ b/include/asm-ia64/thread_info.h
@@ -54,6 +54,8 @@ struct thread_info {
 	},					\
 }
 
+#define __HAVE_ARCH_THREAD_INFO_ALLOCATOR
+
 #ifndef ASM_OFFSETS_C
 /* how to get the thread information struct from C */
 #define current_thread_info()	((struct thread_info *) ((char *) current + IA64_TASK_SIZE))
diff --git a/include/asm-m32r/thread_info.h b/include/asm-m32r/thread_info.h
index 1effcd0f5e63..8589d462df27 100644
--- a/include/asm-m32r/thread_info.h
+++ b/include/asm-m32r/thread_info.h
@@ -94,6 +94,8 @@ static inline struct thread_info *current_thread_info(void)
 	return ti;
 }
 
+#define __HAVE_ARCH_THREAD_INFO_ALLOCATOR
+
 /* thread information allocation */
 #ifdef CONFIG_DEBUG_STACK_USAGE
 #define alloc_thread_info(tsk)					\
diff --git a/include/asm-m68k/thread_info.h b/include/asm-m68k/thread_info.h
index d635a3752488..abc002798a2b 100644
--- a/include/asm-m68k/thread_info.h
+++ b/include/asm-m68k/thread_info.h
@@ -25,13 +25,7 @@ struct thread_info {
 }
 
 /* THREAD_SIZE should be 8k, so handle differently for 4k and 8k machines */
-#if PAGE_SHIFT == 13 /* 8k machines */
-#define alloc_thread_info(tsk)   ((struct thread_info *)__get_free_pages(GFP_KERNEL,0))
-#define free_thread_info(ti)  free_pages((unsigned long)(ti),0)
-#else /* otherwise assume 4k pages */
-#define alloc_thread_info(tsk)   ((struct thread_info *)__get_free_pages(GFP_KERNEL,1))
-#define free_thread_info(ti)  free_pages((unsigned long)(ti),1)
-#endif /* PAGE_SHIFT == 13 */
+#define THREAD_SIZE_ORDER (13 - PAGE_SHIFT)
 
 #define init_thread_info	(init_task.thread.info)
 #define init_stack		(init_thread_union.stack)
diff --git a/include/asm-m68knommu/thread_info.h b/include/asm-m68knommu/thread_info.h
index 95996d978bed..0c9bc095f3f0 100644
--- a/include/asm-m68knommu/thread_info.h
+++ b/include/asm-m68knommu/thread_info.h
@@ -71,10 +71,6 @@ static inline struct thread_info *current_thread_info(void)
 	return ti;
 }
 
-/* thread information allocation */
-#define alloc_thread_info(tsk) ((struct thread_info *) \
-				__get_free_pages(GFP_KERNEL, THREAD_SIZE_ORDER))
-#define free_thread_info(ti)	free_pages((unsigned long) (ti), THREAD_SIZE_ORDER)
 #endif /* __ASSEMBLY__ */
 
 #define	PREEMPT_ACTIVE	0x4000000
diff --git a/include/asm-mips/thread_info.h b/include/asm-mips/thread_info.h
index b2772df1a1bd..bb3060699df2 100644
--- a/include/asm-mips/thread_info.h
+++ b/include/asm-mips/thread_info.h
@@ -82,6 +82,8 @@ register struct thread_info *__current_thread_info __asm__("$28");
 #define THREAD_SIZE (PAGE_SIZE << THREAD_SIZE_ORDER)
 #define THREAD_MASK (THREAD_SIZE - 1UL)
 
+#define __HAVE_ARCH_THREAD_INFO_ALLOCATOR
+
 #ifdef CONFIG_DEBUG_STACK_USAGE
 #define alloc_thread_info(tsk)					\
 ({								\
diff --git a/include/asm-mn10300/thread_info.h b/include/asm-mn10300/thread_info.h
index e397e7192785..78a3881f3c12 100644
--- a/include/asm-mn10300/thread_info.h
+++ b/include/asm-mn10300/thread_info.h
@@ -112,6 +112,8 @@ static inline unsigned long current_stack_pointer(void)
 	return sp;
 }
 
+#define __HAVE_ARCH_THREAD_INFO_ALLOCATOR
+
 /* thread information allocation */
 #ifdef CONFIG_DEBUG_STACK_USAGE
 #define alloc_thread_info(tsk) kzalloc(THREAD_SIZE, GFP_KERNEL)
diff --git a/include/asm-parisc/thread_info.h b/include/asm-parisc/thread_info.h
index 2d9c7500867b..9f812741c355 100644
--- a/include/asm-parisc/thread_info.h
+++ b/include/asm-parisc/thread_info.h
@@ -34,15 +34,11 @@ struct thread_info {
 
 /* thread information allocation */
 
-#define THREAD_ORDER            2
+#define THREAD_SIZE_ORDER            2
 /* Be sure to hunt all references to this down when you change the size of
  * the kernel stack */
-#define THREAD_SIZE             (PAGE_SIZE << THREAD_ORDER)
-#define THREAD_SHIFT            (PAGE_SHIFT + THREAD_ORDER)
-
-#define alloc_thread_info(tsk) ((struct thread_info *) \
-			__get_free_pages(GFP_KERNEL, THREAD_ORDER))
-#define free_thread_info(ti)    free_pages((unsigned long) (ti), THREAD_ORDER)
+#define THREAD_SIZE             (PAGE_SIZE << THREAD_SIZE_ORDER)
+#define THREAD_SHIFT            (PAGE_SHIFT + THREAD_SIZE_ORDER)
 
 /* how to get the thread information struct from C */
 #define current_thread_info()	((struct thread_info *)mfctl(30))
diff --git a/include/asm-powerpc/thread_info.h b/include/asm-powerpc/thread_info.h
index b705c2a7651a..a9db562df69a 100644
--- a/include/asm-powerpc/thread_info.h
+++ b/include/asm-powerpc/thread_info.h
@@ -66,20 +66,12 @@ struct thread_info {
 
 #if THREAD_SHIFT >= PAGE_SHIFT
 
-#define THREAD_ORDER	(THREAD_SHIFT - PAGE_SHIFT)
-
-#ifdef CONFIG_DEBUG_STACK_USAGE
-#define alloc_thread_info(tsk)	\
-	((struct thread_info *)__get_free_pages(GFP_KERNEL | \
-		__GFP_ZERO, THREAD_ORDER))
-#else
-#define alloc_thread_info(tsk)	\
-	((struct thread_info *)__get_free_pages(GFP_KERNEL, THREAD_ORDER))
-#endif
-#define free_thread_info(ti)	free_pages((unsigned long)ti, THREAD_ORDER)
+#define THREAD_SIZE_ORDER	(THREAD_SHIFT - PAGE_SHIFT)
 
 #else /* THREAD_SHIFT < PAGE_SHIFT */
 
+#define __HAVE_ARCH_THREAD_INFO_ALLOCATOR
+
 extern struct thread_info *alloc_thread_info(struct task_struct *tsk);
 extern void free_thread_info(struct thread_info *ti);
 
diff --git a/include/asm-s390/thread_info.h b/include/asm-s390/thread_info.h
index 99bbed99a3b2..91a8f93ad355 100644
--- a/include/asm-s390/thread_info.h
+++ b/include/asm-s390/thread_info.h
@@ -78,10 +78,7 @@ static inline struct thread_info *current_thread_info(void)
 	return (struct thread_info *)((*(unsigned long *) __LC_KERNEL_STACK)-THREAD_SIZE);
 }
 
-/* thread information allocation */
-#define alloc_thread_info(tsk) ((struct thread_info *) \
-	__get_free_pages(GFP_KERNEL,THREAD_ORDER))
-#define free_thread_info(ti) free_pages((unsigned long) (ti),THREAD_ORDER)
+#define THREAD_SIZE_ORDER THREAD_ORDER
 
 #endif
 
diff --git a/include/asm-sh/thread_info.h b/include/asm-sh/thread_info.h
index c50e5d35fe84..5131e3907525 100644
--- a/include/asm-sh/thread_info.h
+++ b/include/asm-sh/thread_info.h
@@ -92,6 +92,8 @@ static inline struct thread_info *current_thread_info(void)
 	return ti;
 }
 
+#define __HAVE_ARCH_THREAD_INFO_ALLOCATOR
+
 /* thread information allocation */
 #ifdef CONFIG_DEBUG_STACK_USAGE
 #define alloc_thread_info(ti)	kzalloc(THREAD_SIZE, GFP_KERNEL)
diff --git a/include/asm-sparc/thread_info_32.h b/include/asm-sparc/thread_info_32.h
index 91b9f5888c85..2cf9db044055 100644
--- a/include/asm-sparc/thread_info_32.h
+++ b/include/asm-sparc/thread_info_32.h
@@ -86,6 +86,8 @@ register struct thread_info *current_thread_info_reg asm("g6");
 #define THREAD_INFO_ORDER  1
 #endif
 
+#define __HAVE_ARCH_THREAD_INFO_ALLOCATOR
+
 BTFIXUPDEF_CALL(struct thread_info *, alloc_thread_info, void)
 #define alloc_thread_info(tsk) BTFIXUP_CALL(alloc_thread_info)()
 
diff --git a/include/asm-sparc/thread_info_64.h b/include/asm-sparc/thread_info_64.h
index c6d2e6c7f844..960969d5ad06 100644
--- a/include/asm-sparc/thread_info_64.h
+++ b/include/asm-sparc/thread_info_64.h
@@ -155,6 +155,8 @@ register struct thread_info *current_thread_info_reg asm("g6");
 #define __THREAD_INFO_ORDER	0
 #endif /* PAGE_SHIFT == 13 */
 
+#define __HAVE_ARCH_THREAD_INFO_ALLOCATOR
+
 #ifdef CONFIG_DEBUG_STACK_USAGE
 #define alloc_thread_info(tsk)					\
 ({								\
diff --git a/include/asm-um/thread_info.h b/include/asm-um/thread_info.h
index 356b83e2c22e..e07e72846c7a 100644
--- a/include/asm-um/thread_info.h
+++ b/include/asm-um/thread_info.h
@@ -53,21 +53,7 @@ static inline struct thread_info *current_thread_info(void)
 	return ti;
 }
 
-#ifdef CONFIG_DEBUG_STACK_USAGE
-
-#define alloc_thread_info(tsk) \
-	((struct thread_info *) __get_free_pages(GFP_KERNEL | __GFP_ZERO, \
-						 CONFIG_KERNEL_STACK_ORDER))
-#else
-
-/* thread information allocation */
-#define alloc_thread_info(tsk) \
-	((struct thread_info *) __get_free_pages(GFP_KERNEL, \
-						 CONFIG_KERNEL_STACK_ORDER))
-#endif
-
-#define free_thread_info(ti) \
-	free_pages((unsigned long)(ti),CONFIG_KERNEL_STACK_ORDER)
+#define THREAD_SIZE_ORDER CONFIG_KERNEL_STACK_ORDER
 
 #endif
 
diff --git a/include/asm-x86/thread_info.h b/include/asm-x86/thread_info.h
index 3f2de1050988..da0a675adf94 100644
--- a/include/asm-x86/thread_info.h
+++ b/include/asm-x86/thread_info.h
@@ -152,6 +152,8 @@ struct thread_info {
 #define THREAD_FLAGS GFP_KERNEL
 #endif
 
+#define __HAVE_ARCH_THREAD_INFO_ALLOCATOR
+
 #define alloc_thread_info(tsk)						\
 	((struct thread_info *)__get_free_pages(THREAD_FLAGS, THREAD_ORDER))
 
diff --git a/include/asm-xtensa/thread_info.h b/include/asm-xtensa/thread_info.h
index a2c640682ed9..7e4131dd546c 100644
--- a/include/asm-xtensa/thread_info.h
+++ b/include/asm-xtensa/thread_info.h
@@ -111,10 +111,6 @@ static inline struct thread_info *current_thread_info(void)
 	return ti;
 }
 
-/* thread information allocation */
-#define alloc_thread_info(tsk) ((struct thread_info *) __get_free_pages(GFP_KERNEL,1))
-#define free_thread_info(ti) free_pages((unsigned long) (ti), 1)
-
 #else /* !__ASSEMBLY__ */
 
 /* how to get the thread information struct from ASM */
@@ -160,6 +156,7 @@ static inline struct thread_info *current_thread_info(void)
 #define TS_USEDFPU		0x0001	/* FPU was used by this task this quantum (SMP) */
 
 #define THREAD_SIZE 8192	//(2*PAGE_SIZE)
+#define THREAD_SIZE_ORDER 1
 
 #endif	/* __KERNEL__ */
 #endif	/* _XTENSA_THREAD_INFO */
diff --git a/kernel/fork.c b/kernel/fork.c
index 552c8d8e77ad..5a5d6fef341d 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -93,6 +93,23 @@ int nr_processes(void)
 static struct kmem_cache *task_struct_cachep;
 #endif
 
+#ifndef __HAVE_ARCH_THREAD_INFO_ALLOCATOR
+static inline struct thread_info *alloc_thread_info(struct task_struct *tsk)
+{
+#ifdef CONFIG_DEBUG_STACK_USAGE
+	gfp_t mask = GFP_KERNEL | __GFP_ZERO;
+#else
+	gfp_t mask = GFP_KERNEL;
+#endif
+	return (struct thread_info *)__get_free_pages(mask, THREAD_SIZE_ORDER);
+}
+
+static inline void free_thread_info(struct thread_info *ti)
+{
+	free_pages((unsigned long)ti, THREAD_SIZE_ORDER);
+}
+#endif
+
 /* SLAB cache for signal_struct structures (tsk->signal) */
 static struct kmem_cache *signal_cachep;
 

From 2b4bc46052ea8cd7c370b67ca0b9c26586f1439a Mon Sep 17 00:00:00 2001
From: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Date: Fri, 25 Jul 2008 01:45:42 -0700
Subject: [PATCH 0697/1171] pdflush: use time_after() instead of open-coding it

Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/pdflush.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/mm/pdflush.c b/mm/pdflush.c
index 9d834aa4b979..0cbe0c60c6bf 100644
--- a/mm/pdflush.c
+++ b/mm/pdflush.c
@@ -130,7 +130,7 @@ static int __pdflush(struct pdflush_work *my_work)
 		 * Thread creation: For how long have there been zero
 		 * available threads?
 		 */
-		if (jiffies - last_empty_jifs > 1 * HZ) {
+		if (time_after(jiffies, last_empty_jifs + 1 * HZ)) {
 			/* unlocked list_empty() test is OK here */
 			if (list_empty(&pdflush_list)) {
 				/* unlocked test is OK here */
@@ -151,7 +151,7 @@ static int __pdflush(struct pdflush_work *my_work)
 		if (nr_pdflush_threads <= MIN_PDFLUSH_THREADS)
 			continue;
 		pdf = list_entry(pdflush_list.prev, struct pdflush_work, list);
-		if (jiffies - pdf->when_i_went_to_sleep > 1 * HZ) {
+		if (time_after(jiffies, pdf->when_i_went_to_sleep + 1 * HZ)) {
 			/* Limit exit rate */
 			pdf->when_i_went_to_sleep = jiffies;
 			break;					/* exeunt */

From ba92a43dbaee339cf5915ef766d3d3ffbaaf103c Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Fri, 25 Jul 2008 01:45:43 -0700
Subject: [PATCH 0698/1171] exec: remove some includes

fs/exec.c used to need mman.h pagemap.h swap.h and rmap.h when it did
mm-ish stuff in install_arg_page(); but no need for them after 2.6.22.

[akpm@linux-foundation.org: unbreak arm]
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/exec.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/fs/exec.c b/fs/exec.c
index 190ed1f92774..e41aef0fb351 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -25,19 +25,18 @@
 #include <linux/slab.h>
 #include <linux/file.h>
 #include <linux/fdtable.h>
-#include <linux/mman.h>
+#include <linux/mm.h>
 #include <linux/stat.h>
 #include <linux/fcntl.h>
 #include <linux/smp_lock.h>
+#include <linux/swap.h>
 #include <linux/string.h>
 #include <linux/init.h>
-#include <linux/pagemap.h>
 #include <linux/highmem.h>
 #include <linux/spinlock.h>
 #include <linux/key.h>
 #include <linux/personality.h>
 #include <linux/binfmts.h>
-#include <linux/swap.h>
 #include <linux/utsname.h>
 #include <linux/pid_namespace.h>
 #include <linux/module.h>
@@ -47,7 +46,6 @@
 #include <linux/mount.h>
 #include <linux/security.h>
 #include <linux/syscalls.h>
-#include <linux/rmap.h>
 #include <linux/tsacct_kern.h>
 #include <linux/cn_proc.h>
 #include <linux/audit.h>

From 2d6ffcca623a9a16df6cdfbe8250b7a5904a5f5e Mon Sep 17 00:00:00 2001
From: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Date: Fri, 25 Jul 2008 01:45:44 -0700
Subject: [PATCH 0699/1171] inflate: refactor inflate malloc code

Inflate requires some dynamic memory allocation very early in the boot
process and this is provided with a set of four functions:
malloc/free/gzip_mark/gzip_release.

The old inflate code used a mark/release strategy rather than implement
free.  This new version instead keeps a count on the number of outstanding
allocations and when it hits zero, it resets the malloc arena.

This allows removing all the mark and release implementations and unifying
all the malloc/free implementations.

The architecture-dependent code must define two addresses:
 - free_mem_ptr, the address of the beginning of the area in which
   allocations should be made
 - free_mem_end_ptr, the address of the end of the area in which
   allocations should be made. If set to 0, then no check is made on
   the number of allocations, it just grows as much as needed

The architecture-dependent code can also provide an arch_decomp_wdog()
function call.  This function will be called several times during the
decompression process, and allow to notify the watchdog that the system is
still running.  If an architecture provides such a call, then it must
define ARCH_HAS_DECOMP_WDOG so that the generic inflate code calls
arch_decomp_wdog().

Work initially done by Matt Mackall, updated to a recent version of the
kernel and improved by me.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Matt Mackall <mpm@selenic.com>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Mikael Starvik <mikael.starvik@axis.com>
Cc: Jesper Nilsson <jesper.nilsson@axis.com>
Cc: Haavard Skinnemoen <hskinnemoen@atmel.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Acked-by: Paul Mundt <lethal@linux-sh.org>
Acked-by: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/alpha/boot/misc.c                    | 39 +--------------
 arch/arm/boot/compressed/misc.c           | 59 +++--------------------
 arch/cris/arch-v10/boot/compressed/misc.c | 36 +-------------
 arch/cris/arch-v32/boot/compressed/misc.c | 39 ++-------------
 arch/h8300/boot/compressed/misc.c         | 38 ---------------
 arch/m32r/boot/compressed/misc.c          | 37 --------------
 arch/mn10300/boot/compressed/misc.c       | 37 --------------
 arch/sh/boot/compressed/misc_32.c         | 38 ---------------
 arch/sh/boot/compressed/misc_64.c         | 40 ---------------
 arch/x86/boot/compressed/misc.c           | 39 ---------------
 init/do_mounts_rd.c                       | 25 +---------
 init/initramfs.c                          | 22 +--------
 lib/inflate.c                             | 52 +++++++++++++++++---
 13 files changed, 62 insertions(+), 439 deletions(-)

diff --git a/arch/alpha/boot/misc.c b/arch/alpha/boot/misc.c
index c00646b25f6e..3047a1b3a517 100644
--- a/arch/alpha/boot/misc.c
+++ b/arch/alpha/boot/misc.c
@@ -78,8 +78,6 @@ static unsigned outcnt;		/* bytes in output buffer */
 static int  fill_inbuf(void);
 static void flush_window(void);
 static void error(char *m);
-static void gzip_mark(void **);
-static void gzip_release(void **);
 
 static char *input_data;
 static int  input_data_size;
@@ -88,51 +86,18 @@ static uch *output_data;
 static ulg output_ptr;
 static ulg bytes_out;
 
-static void *malloc(int size);
-static void free(void *where);
 static void error(char *m);
 static void gzip_mark(void **);
 static void gzip_release(void **);
 
 extern int end;
 static ulg free_mem_ptr;
-static ulg free_mem_ptr_end;
+static ulg free_mem_end_ptr;
 
 #define HEAP_SIZE 0x3000
 
 #include "../../../lib/inflate.c"
 
-static void *malloc(int size)
-{
-	void *p;
-
-	if (size <0) error("Malloc error");
-	if (free_mem_ptr <= 0) error("Memory error");
-
-	free_mem_ptr = (free_mem_ptr + 3) & ~3;	/* Align */
-
-	p = (void *)free_mem_ptr;
-	free_mem_ptr += size;
-
-	if (free_mem_ptr >= free_mem_ptr_end)
-		error("Out of memory");
-	return p;
-}
-
-static void free(void *where)
-{ /* gzip_mark & gzip_release do the free */
-}
-
-static void gzip_mark(void **ptr)
-{
-	*ptr = (void *) free_mem_ptr;
-}
-
-static void gzip_release(void **ptr)
-{
-	free_mem_ptr = (long) *ptr;
-}
-
 /* ===========================================================================
  * Fill the input buffer. This is called only when the buffer is empty
  * and at least one byte is really needed.
@@ -193,7 +158,7 @@ decompress_kernel(void *output_start,
 
 	/* FIXME FIXME FIXME */
 	free_mem_ptr		= (ulg)output_start + ksize;
-	free_mem_ptr_end	= (ulg)output_start + ksize + 0x200000;
+	free_mem_end_ptr	= (ulg)output_start + ksize + 0x200000;
 	/* FIXME FIXME FIXME */
 
 	/* put in temp area to reduce initial footprint */
diff --git a/arch/arm/boot/compressed/misc.c b/arch/arm/boot/compressed/misc.c
index 9b444022cb9b..7145cc7c04f0 100644
--- a/arch/arm/boot/compressed/misc.c
+++ b/arch/arm/boot/compressed/misc.c
@@ -217,8 +217,6 @@ static unsigned outcnt;		/* bytes in output buffer */
 static int  fill_inbuf(void);
 static void flush_window(void);
 static void error(char *m);
-static void gzip_mark(void **);
-static void gzip_release(void **);
 
 extern char input_data[];
 extern char input_data_end[];
@@ -227,65 +225,22 @@ static uch *output_data;
 static ulg output_ptr;
 static ulg bytes_out;
 
-static void *malloc(int size);
-static void free(void *where);
 static void error(char *m);
-static void gzip_mark(void **);
-static void gzip_release(void **);
 
 static void putstr(const char *);
 
 extern int end;
 static ulg free_mem_ptr;
-static ulg free_mem_ptr_end;
+static ulg free_mem_end_ptr;
 
-#define HEAP_SIZE 0x3000
+#ifdef STANDALONE_DEBUG
+#define NO_INFLATE_MALLOC
+#endif
+
+#define ARCH_HAS_DECOMP_WDOG
 
 #include "../../../../lib/inflate.c"
 
-#ifndef STANDALONE_DEBUG
-static void *malloc(int size)
-{
-	void *p;
-
-	if (size <0) error("Malloc error");
-	if (free_mem_ptr <= 0) error("Memory error");
-
-	free_mem_ptr = (free_mem_ptr + 3) & ~3;	/* Align */
-
-	p = (void *)free_mem_ptr;
-	free_mem_ptr += size;
-
-	if (free_mem_ptr >= free_mem_ptr_end)
-		error("Out of memory");
-	return p;
-}
-
-static void free(void *where)
-{ /* gzip_mark & gzip_release do the free */
-}
-
-static void gzip_mark(void **ptr)
-{
-	arch_decomp_wdog();
-	*ptr = (void *) free_mem_ptr;
-}
-
-static void gzip_release(void **ptr)
-{
-	arch_decomp_wdog();
-	free_mem_ptr = (long) *ptr;
-}
-#else
-static void gzip_mark(void **ptr)
-{
-}
-
-static void gzip_release(void **ptr)
-{
-}
-#endif
-
 /* ===========================================================================
  * Fill the input buffer. This is called only when the buffer is empty
  * and at least one byte is really needed.
@@ -348,7 +303,7 @@ decompress_kernel(ulg output_start, ulg free_mem_ptr_p, ulg free_mem_ptr_end_p,
 {
 	output_data		= (uch *)output_start;	/* Points to kernel start */
 	free_mem_ptr		= free_mem_ptr_p;
-	free_mem_ptr_end	= free_mem_ptr_end_p;
+	free_mem_end_ptr	= free_mem_ptr_end_p;
 	__machine_arch_type	= arch_id;
 
 	arch_decomp_setup();
diff --git a/arch/cris/arch-v10/boot/compressed/misc.c b/arch/cris/arch-v10/boot/compressed/misc.c
index 18e13bce1400..d933c89889db 100644
--- a/arch/cris/arch-v10/boot/compressed/misc.c
+++ b/arch/cris/arch-v10/boot/compressed/misc.c
@@ -102,50 +102,16 @@ extern char *input_data;  /* lives in head.S */
 static long bytes_out = 0;
 static uch *output_data;
 static unsigned long output_ptr = 0;
-
-static void *malloc(int size);
-static void free(void *where);
-static void gzip_mark(void **);
-static void gzip_release(void **);
-
 static void puts(const char *);
 
 /* the "heap" is put directly after the BSS ends, at end */
 
 extern int _end;
 static long free_mem_ptr = (long)&_end;
+static long free_mem_end_ptr;
 
 #include "../../../../../lib/inflate.c"
 
-static void *malloc(int size)
-{
-	void *p;
-
-	if (size < 0)
-		error("Malloc error");
-
-	free_mem_ptr = (free_mem_ptr + 3) & ~3;	/* Align */
-
-	p = (void *)free_mem_ptr;
-	free_mem_ptr += size;
-
-	return p;
-}
-
-static void free(void *where)
-{	/* Don't care */
-}
-
-static void gzip_mark(void **ptr)
-{
-	*ptr = (void *) free_mem_ptr;
-}
-
-static void gzip_release(void **ptr)
-{
-	free_mem_ptr = (long) *ptr;
-}
-
 /* decompressor info and error messages to serial console */
 
 static void
diff --git a/arch/cris/arch-v32/boot/compressed/misc.c b/arch/cris/arch-v32/boot/compressed/misc.c
index 55b2695c5d70..3595e16e82bc 100644
--- a/arch/cris/arch-v32/boot/compressed/misc.c
+++ b/arch/cris/arch-v32/boot/compressed/misc.c
@@ -89,20 +89,14 @@ static unsigned outcnt = 0;  /* bytes in output buffer */
 
 static void flush_window(void);
 static void error(char *m);
-static void gzip_mark(void **);
-static void gzip_release(void **);
 
 extern char *input_data;  /* lives in head.S */
 
-static long bytes_out = 0;
+static long bytes_out;
 static uch *output_data;
-static unsigned long output_ptr = 0;
+static unsigned long output_ptr;
 
-static void *malloc(int size);
-static void free(void *where);
 static void error(char *m);
-static void gzip_mark(void **);
-static void gzip_release(void **);
 
 static void puts(const char *);
 
@@ -110,37 +104,10 @@ static void puts(const char *);
 
 extern int _end;
 static long free_mem_ptr = (long)&_end;
+static long free_mem_end_ptr;
 
 #include "../../../../../lib/inflate.c"
 
-static void *malloc(int size)
-{
-	void *p;
-
-	if (size <0) error("Malloc error");
-
-	free_mem_ptr = (free_mem_ptr + 3) & ~3;	/* Align */
-
-	p = (void *)free_mem_ptr;
-	free_mem_ptr += size;
-
-	return p;
-}
-
-static void free(void *where)
-{	/* Don't care */
-}
-
-static void gzip_mark(void **ptr)
-{
-	*ptr = (void *) free_mem_ptr;
-}
-
-static void gzip_release(void **ptr)
-{
-	free_mem_ptr = (long) *ptr;
-}
-
 /* decompressor info and error messages to serial console */
 
 static inline void
diff --git a/arch/h8300/boot/compressed/misc.c b/arch/h8300/boot/compressed/misc.c
index 845074588af0..51ab6cbd030f 100644
--- a/arch/h8300/boot/compressed/misc.c
+++ b/arch/h8300/boot/compressed/misc.c
@@ -67,8 +67,6 @@ static unsigned outcnt = 0;  /* bytes in output buffer */
 static int  fill_inbuf(void);
 static void flush_window(void);
 static void error(char *m);
-static void gzip_mark(void **);
-static void gzip_release(void **);
 
 extern char input_data[];
 extern int input_len;
@@ -77,11 +75,7 @@ static long bytes_out = 0;
 static uch *output_data;
 static unsigned long output_ptr = 0;
 
-static void *malloc(int size);
-static void free(void *where);
 static void error(char *m);
-static void gzip_mark(void **);
-static void gzip_release(void **);
 
 int puts(const char *);
 
@@ -98,38 +92,6 @@ static unsigned long free_mem_end_ptr;
 #define TDR *((volatile unsigned char *)0xffff8b)
 #define SSR *((volatile unsigned char *)0xffff8c)
 
-static void *malloc(int size)
-{
-	void *p;
-
-	if (size <0) error("Malloc error");
-	if (free_mem_ptr == 0) error("Memory error");
-
-	free_mem_ptr = (free_mem_ptr + 3) & ~3;	/* Align */
-
-	p = (void *)free_mem_ptr;
-	free_mem_ptr += size;
-
-	if (free_mem_ptr >= free_mem_end_ptr)
-		error("Out of memory");
-
-	return p;
-}
-
-static void free(void *where)
-{	/* Don't care */
-}
-
-static void gzip_mark(void **ptr)
-{
-	*ptr = (void *) free_mem_ptr;
-}
-
-static void gzip_release(void **ptr)
-{
-	free_mem_ptr = (long) *ptr;
-}
-
 int puts(const char *s)
 {
 	return 0;
diff --git a/arch/m32r/boot/compressed/misc.c b/arch/m32r/boot/compressed/misc.c
index 600d40e33495..d394292498c0 100644
--- a/arch/m32r/boot/compressed/misc.c
+++ b/arch/m32r/boot/compressed/misc.c
@@ -70,8 +70,6 @@ static unsigned outcnt = 0;  /* bytes in output buffer */
 static int  fill_inbuf(void);
 static void flush_window(void);
 static void error(char *m);
-static void gzip_mark(void **);
-static void gzip_release(void **);
 
 static unsigned char *input_data;
 static int input_len;
@@ -82,9 +80,6 @@ static unsigned long output_ptr = 0;
 
 #include "m32r_sio.c"
 
-static void *malloc(int size);
-static void free(void *where);
-
 static unsigned long free_mem_ptr;
 static unsigned long free_mem_end_ptr;
 
@@ -92,38 +87,6 @@ static unsigned long free_mem_end_ptr;
 
 #include "../../../../lib/inflate.c"
 
-static void *malloc(int size)
-{
-	void *p;
-
-	if (size <0) error("Malloc error");
-	if (free_mem_ptr == 0) error("Memory error");
-
-	free_mem_ptr = (free_mem_ptr + 3) & ~3;	/* Align */
-
-	p = (void *)free_mem_ptr;
-	free_mem_ptr += size;
-
-	if (free_mem_ptr >= free_mem_end_ptr)
-		error("Out of memory");
-
-	return p;
-}
-
-static void free(void *where)
-{	/* Don't care */
-}
-
-static void gzip_mark(void **ptr)
-{
-	*ptr = (void *) free_mem_ptr;
-}
-
-static void gzip_release(void **ptr)
-{
-	free_mem_ptr = (long) *ptr;
-}
-
 void* memset(void* s, int c, size_t n)
 {
 	int i;
diff --git a/arch/mn10300/boot/compressed/misc.c b/arch/mn10300/boot/compressed/misc.c
index ded207efc97a..f673383518e4 100644
--- a/arch/mn10300/boot/compressed/misc.c
+++ b/arch/mn10300/boot/compressed/misc.c
@@ -153,26 +153,9 @@ static uch *output_data;
 static unsigned long output_ptr;
 
 
-static void *malloc(int size);
-
-static inline void free(void *where)
-{	/* Don't care */
-}
-
 static unsigned long free_mem_ptr = (unsigned long) &end;
 static unsigned long free_mem_end_ptr = (unsigned long) &end + 0x90000;
 
-static inline void gzip_mark(void **ptr)
-{
-	kputs(".");
-	*ptr = (void *) free_mem_ptr;
-}
-
-static inline void gzip_release(void **ptr)
-{
-	free_mem_ptr = (unsigned long) *ptr;
-}
-
 #define INPLACE_MOVE_ROUTINE	0x1000
 #define LOW_BUFFER_START	0x2000
 #define LOW_BUFFER_END		0x90000
@@ -186,26 +169,6 @@ static int lines, cols;
 
 #include "../../../../lib/inflate.c"
 
-static void *malloc(int size)
-{
-	void *p;
-
-	if (size < 0)
-		error("Malloc error\n");
-	if (!free_mem_ptr)
-		error("Memory error\n");
-
-	free_mem_ptr = (free_mem_ptr + 3) & ~3;	/* Align */
-
-	p = (void *) free_mem_ptr;
-	free_mem_ptr += size;
-
-	if (free_mem_ptr >= free_mem_end_ptr)
-		error("\nOut of memory\n");
-
-	return p;
-}
-
 static inline void scroll(void)
 {
 	int i;
diff --git a/arch/sh/boot/compressed/misc_32.c b/arch/sh/boot/compressed/misc_32.c
index adcea31e663e..f386997e4d9c 100644
--- a/arch/sh/boot/compressed/misc_32.c
+++ b/arch/sh/boot/compressed/misc_32.c
@@ -74,8 +74,6 @@ static unsigned outcnt = 0;  /* bytes in output buffer */
 static int  fill_inbuf(void);
 static void flush_window(void);
 static void error(char *m);
-static void gzip_mark(void **);
-static void gzip_release(void **);
 
 extern char input_data[];
 extern int input_len;
@@ -84,11 +82,7 @@ static long bytes_out = 0;
 static uch *output_data;
 static unsigned long output_ptr = 0;
 
-static void *malloc(int size);
-static void free(void *where);
 static void error(char *m);
-static void gzip_mark(void **);
-static void gzip_release(void **);
 
 int puts(const char *);
 
@@ -101,38 +95,6 @@ static unsigned long free_mem_end_ptr;
 
 #include "../../../../lib/inflate.c"
 
-static void *malloc(int size)
-{
-	void *p;
-
-	if (size <0) error("Malloc error");
-	if (free_mem_ptr == 0) error("Memory error");
-
-	free_mem_ptr = (free_mem_ptr + 3) & ~3;	/* Align */
-
-	p = (void *)free_mem_ptr;
-	free_mem_ptr += size;
-
-	if (free_mem_ptr >= free_mem_end_ptr)
-		error("Out of memory");
-
-	return p;
-}
-
-static void free(void *where)
-{	/* Don't care */
-}
-
-static void gzip_mark(void **ptr)
-{
-	*ptr = (void *) free_mem_ptr;
-}
-
-static void gzip_release(void **ptr)
-{
-	free_mem_ptr = (long) *ptr;
-}
-
 #ifdef CONFIG_SH_STANDARD_BIOS
 size_t strlen(const char *s)
 {
diff --git a/arch/sh/boot/compressed/misc_64.c b/arch/sh/boot/compressed/misc_64.c
index a006ef89b9dd..2941657e18aa 100644
--- a/arch/sh/boot/compressed/misc_64.c
+++ b/arch/sh/boot/compressed/misc_64.c
@@ -72,8 +72,6 @@ static unsigned outcnt = 0;	/* bytes in output buffer */
 static int fill_inbuf(void);
 static void flush_window(void);
 static void error(char *m);
-static void gzip_mark(void **);
-static void gzip_release(void **);
 
 extern char input_data[];
 extern int input_len;
@@ -82,11 +80,7 @@ static long bytes_out = 0;
 static uch *output_data;
 static unsigned long output_ptr = 0;
 
-static void *malloc(int size);
-static void free(void *where);
 static void error(char *m);
-static void gzip_mark(void **);
-static void gzip_release(void **);
 
 static void puts(const char *);
 
@@ -99,40 +93,6 @@ static unsigned long free_mem_end_ptr;
 
 #include "../../../../lib/inflate.c"
 
-static void *malloc(int size)
-{
-	void *p;
-
-	if (size < 0)
-		error("Malloc error\n");
-	if (free_mem_ptr == 0)
-		error("Memory error\n");
-
-	free_mem_ptr = (free_mem_ptr + 3) & ~3;	/* Align */
-
-	p = (void *) free_mem_ptr;
-	free_mem_ptr += size;
-
-	if (free_mem_ptr >= free_mem_end_ptr)
-		error("\nOut of memory\n");
-
-	return p;
-}
-
-static void free(void *where)
-{				/* Don't care */
-}
-
-static void gzip_mark(void **ptr)
-{
-	*ptr = (void *) free_mem_ptr;
-}
-
-static void gzip_release(void **ptr)
-{
-	free_mem_ptr = (long) *ptr;
-}
-
 void puts(const char *s)
 {
 }
diff --git a/arch/x86/boot/compressed/misc.c b/arch/x86/boot/compressed/misc.c
index bc5553b496f7..9fea73706479 100644
--- a/arch/x86/boot/compressed/misc.c
+++ b/arch/x86/boot/compressed/misc.c
@@ -182,8 +182,6 @@ static unsigned		outcnt;
 static int  fill_inbuf(void);
 static void flush_window(void);
 static void error(char *m);
-static void gzip_mark(void **);
-static void gzip_release(void **);
 
 /*
  * This is set up by the setup-routine at boot-time
@@ -196,9 +194,6 @@ extern int input_len;
 
 static long bytes_out;
 
-static void *malloc(int size);
-static void free(void *where);
-
 static void *memset(void *s, int c, unsigned n);
 static void *memcpy(void *dest, const void *src, unsigned n);
 
@@ -220,40 +215,6 @@ static int lines, cols;
 
 #include "../../../../lib/inflate.c"
 
-static void *malloc(int size)
-{
-	void *p;
-
-	if (size < 0)
-		error("Malloc error");
-	if (free_mem_ptr <= 0)
-		error("Memory error");
-
-	free_mem_ptr = (free_mem_ptr + 3) & ~3;	/* Align */
-
-	p = (void *)free_mem_ptr;
-	free_mem_ptr += size;
-
-	if (free_mem_ptr >= free_mem_end_ptr)
-		error("Out of memory");
-
-	return p;
-}
-
-static void free(void *where)
-{	/* Don't care */
-}
-
-static void gzip_mark(void **ptr)
-{
-	*ptr = (void *) free_mem_ptr;
-}
-
-static void gzip_release(void **ptr)
-{
-	free_mem_ptr = (memptr) *ptr;
-}
-
 static void scroll(void)
 {
 	int i;
diff --git a/init/do_mounts_rd.c b/init/do_mounts_rd.c
index 470a328d1454..fedef93b586f 100644
--- a/init/do_mounts_rd.c
+++ b/init/do_mounts_rd.c
@@ -303,33 +303,12 @@ static int crd_infd, crd_outfd;
 
 static int  __init fill_inbuf(void);
 static void __init flush_window(void);
-static void __init *malloc(size_t size);
-static void __init free(void *where);
 static void __init error(char *m);
-static void __init gzip_mark(void **);
-static void __init gzip_release(void **);
+
+#define NO_INFLATE_MALLOC
 
 #include "../lib/inflate.c"
 
-static void __init *malloc(size_t size)
-{
-	return kmalloc(size, GFP_KERNEL);
-}
-
-static void __init free(void *where)
-{
-	kfree(where);
-}
-
-static void __init gzip_mark(void **ptr)
-{
-}
-
-static void __init gzip_release(void **ptr)
-{
-}
-
-
 /* ===========================================================================
  * Fill the input buffer. This is called only when the buffer is empty
  * and at least one byte is really needed.
diff --git a/init/initramfs.c b/init/initramfs.c
index 8eeeccb328c9..644fc01ad5f0 100644
--- a/init/initramfs.c
+++ b/init/initramfs.c
@@ -14,16 +14,6 @@ static void __init error(char *x)
 		message = x;
 }
 
-static void __init *malloc(size_t size)
-{
-	return kmalloc(size, GFP_KERNEL);
-}
-
-static void __init free(void *where)
-{
-	kfree(where);
-}
-
 /* link hash */
 
 #define N_ALIGN(len) ((((len) + 1) & ~3) + 2)
@@ -407,19 +397,11 @@ static long bytes_out;
 
 static void __init flush_window(void);
 static void __init error(char *m);
-static void __init gzip_mark(void **);
-static void __init gzip_release(void **);
+
+#define NO_INFLATE_MALLOC
 
 #include "../lib/inflate.c"
 
-static void __init gzip_mark(void **ptr)
-{
-}
-
-static void __init gzip_release(void **ptr)
-{
-}
-
 /* ===========================================================================
  * Write the output window window[0..outcnt-1] and update crc and bytes_out.
  * (Used for the decompressed data only.)
diff --git a/lib/inflate.c b/lib/inflate.c
index 9762294be062..1a8e8a978128 100644
--- a/lib/inflate.c
+++ b/lib/inflate.c
@@ -230,6 +230,45 @@ STATIC const ush mask_bits[] = {
 #define NEEDBITS(n) {while(k<(n)){b|=((ulg)NEXTBYTE())<<k;k+=8;}}
 #define DUMPBITS(n) {b>>=(n);k-=(n);}
 
+#ifndef NO_INFLATE_MALLOC
+/* A trivial malloc implementation, adapted from
+ *  malloc by Hannu Savolainen 1993 and Matthias Urlichs 1994
+ */
+
+static unsigned long malloc_ptr;
+static int malloc_count;
+
+static void *malloc(int size)
+{
+       void *p;
+
+       if (size < 0)
+		error("Malloc error");
+       if (!malloc_ptr)
+		malloc_ptr = free_mem_ptr;
+
+       malloc_ptr = (malloc_ptr + 3) & ~3;     /* Align */
+
+       p = (void *)malloc_ptr;
+       malloc_ptr += size;
+
+       if (free_mem_end_ptr && malloc_ptr >= free_mem_end_ptr)
+		error("Out of memory");
+
+       malloc_count++;
+       return p;
+}
+
+static void free(void *where)
+{
+       malloc_count--;
+       if (!malloc_count)
+		malloc_ptr = free_mem_ptr;
+}
+#else
+#define malloc(a) kmalloc(a, GFP_KERNEL)
+#define free(a) kfree(a)
+#endif
 
 /*
    Huffman code decoding is performed using a multi-level table lookup.
@@ -1045,7 +1084,6 @@ STATIC int INIT inflate(void)
   int e;                /* last block flag */
   int r;                /* result code */
   unsigned h;           /* maximum struct huft's malloc'ed */
-  void *ptr;
 
   /* initialize window, bit buffer */
   wp = 0;
@@ -1057,12 +1095,12 @@ STATIC int INIT inflate(void)
   h = 0;
   do {
     hufts = 0;
-    gzip_mark(&ptr);
-    if ((r = inflate_block(&e)) != 0) {
-      gzip_release(&ptr);	    
-      return r;
-    }
-    gzip_release(&ptr);
+#ifdef ARCH_HAS_DECOMP_WDOG
+    arch_decomp_wdog();
+#endif
+    r = inflate_block(&e);
+    if (r)
+	    return r;
     if (hufts > h)
       h = hufts;
   } while (!e);

From 2f5a5cf93fae7b8354b45b8443dcc3448a8fc276 Mon Sep 17 00:00:00 2001
From: Kay Sievers <kay.sievers@vrfy.org>
Date: Fri, 25 Jul 2008 01:45:46 -0700
Subject: [PATCH 0700/1171] drivers/power: fix platform driver hotplug/coldplug

Since 43cc71eed1250755986da4c0f9898f9a635cb3bf ("platform: prefix MODALIAS
with "platform:"), the platform modalias is prefixed with "platform:".
Add MODULE_ALIAS() to the hotpluggable "power" drivers drivers, to
re-enable auto loading.

[dbrownell@users.sourceforge.net: one was missing]
Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Greg KH <greg@kroah.com>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/power/ds2760_battery.c | 2 ++
 drivers/power/pda_power.c      | 2 ++
 2 files changed, 4 insertions(+)

diff --git a/drivers/power/ds2760_battery.c b/drivers/power/ds2760_battery.c
index 71be36f18709..308ddb201b66 100644
--- a/drivers/power/ds2760_battery.c
+++ b/drivers/power/ds2760_battery.c
@@ -433,6 +433,8 @@ static int ds2760_battery_resume(struct platform_device *pdev)
 
 #endif /* CONFIG_PM */
 
+MODULE_ALIAS("platform:ds2760-battery");
+
 static struct platform_driver ds2760_battery_driver = {
 	.driver = {
 		.name = "ds2760-battery",
diff --git a/drivers/power/pda_power.c b/drivers/power/pda_power.c
index 82810b7bff9c..0471ec743ab9 100644
--- a/drivers/power/pda_power.c
+++ b/drivers/power/pda_power.c
@@ -362,6 +362,8 @@ static int pda_power_resume(struct platform_device *pdev)
 #define pda_power_resume NULL
 #endif /* CONFIG_PM */
 
+MODULE_ALIAS("platform:pda-power");
+
 static struct platform_driver pda_power_pdrv = {
 	.driver = {
 		.name = "pda-power",

From 4f46d6e7e5ffbce0ee1d1a80767fdf45e56cc863 Mon Sep 17 00:00:00 2001
From: Kay Sievers <kay.sievers@vrfy.org>
Date: Fri, 25 Jul 2008 01:45:47 -0700
Subject: [PATCH 0701/1171] mfd: fix platform driver hotplug/coldplug

Since 43cc71eed1250755986da4c0f9898f9a635cb3bf (platform: prefix MODALIAS
with "platform:"), the platform modalias is prefixed with "platform:".
Add MODULE_ALIAS() to the MFD platform drivers, to re-enable auto loading.

[dbrownell@users.sourceforge.net: one was missing]
Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Greg KH <greg@kroah.com>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Samuel Ortiz <samuel@sortiz.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/mfd/htc-pasic3.c | 2 ++
 drivers/mfd/mcp-sa11x0.c | 2 ++
 drivers/mfd/sm501.c      | 2 ++
 3 files changed, 6 insertions(+)

diff --git a/drivers/mfd/htc-pasic3.c b/drivers/mfd/htc-pasic3.c
index 633cbba072f0..91b294dcc133 100644
--- a/drivers/mfd/htc-pasic3.c
+++ b/drivers/mfd/htc-pasic3.c
@@ -238,6 +238,8 @@ static int pasic3_remove(struct platform_device *pdev)
 	return 0;
 }
 
+MODULE_ALIAS("platform:pasic3");
+
 static struct platform_driver pasic3_driver = {
 	.driver		= {
 		.name	= "pasic3",
diff --git a/drivers/mfd/mcp-sa11x0.c b/drivers/mfd/mcp-sa11x0.c
index 1eab7cffceaa..b5272b5ce3fa 100644
--- a/drivers/mfd/mcp-sa11x0.c
+++ b/drivers/mfd/mcp-sa11x0.c
@@ -242,6 +242,8 @@ static int mcp_sa11x0_resume(struct platform_device *dev)
 /*
  * The driver for the SA11x0 MCP port.
  */
+MODULE_ALIAS("platform:sa11x0-mcp");
+
 static struct platform_driver mcp_sa11x0_driver = {
 	.probe		= mcp_sa11x0_probe,
 	.remove		= mcp_sa11x0_remove,
diff --git a/drivers/mfd/sm501.c b/drivers/mfd/sm501.c
index 2fe64734d8af..e2530df4d85c 100644
--- a/drivers/mfd/sm501.c
+++ b/drivers/mfd/sm501.c
@@ -1378,6 +1378,8 @@ static struct pci_driver sm501_pci_drv = {
 	.remove		= sm501_pci_remove,
 };
 
+MODULE_ALIAS("platform:sm501");
+
 static struct platform_driver sm501_plat_drv = {
 	.driver		= {
 		.name	= "sm501",

From db358b40e0674fd4079204d8e3e1c8ab3829a1b9 Mon Sep 17 00:00:00 2001
From: Kay Sievers <kay.sievers@vrfy.org>
Date: Fri, 25 Jul 2008 01:45:48 -0700
Subject: [PATCH 0702/1171] parport: fix platform driver hotplug/coldplug

Since 43cc71eed1250755986da4c0f9898f9a635cb3bf (platform: prefix MODALIAS
with "platform:"), the platform modalias is prefixed with "platform:".
Add MODULE_ALIAS() to the hotpluggable parport platform drivers, to
re-enable auto loading.

Signed-off-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Greg KH <greg@kroah.com>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Acked-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/parport/parport_ax88796.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/parport/parport_ax88796.c b/drivers/parport/parport_ax88796.c
index 4ec220b2eae7..6938d2e9f18f 100644
--- a/drivers/parport/parport_ax88796.c
+++ b/drivers/parport/parport_ax88796.c
@@ -406,6 +406,8 @@ static int parport_ax88796_resume(struct platform_device *dev)
 #define parport_ax88796_resume  NULL
 #endif
 
+MODULE_ALIAS("platform:ax88796-pp");
+
 static struct platform_driver axdrv = {
 	.driver		= {
 		.name	= "ax88796-pp",

From 4500d067eeb3d00679335d9cf5c6536e79cd3ef4 Mon Sep 17 00:00:00 2001
From: "Robert P. J. Day" <rpjday@crashcourse.ca>
Date: Fri, 25 Jul 2008 01:45:49 -0700
Subject: [PATCH 0703/1171] init.h: remove obsolete content

Remove apparently obsolete content from init.h referring to gcc 2.9x
and to "no_module_init".

Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/init.h | 8 +-------
 1 file changed, 1 insertion(+), 7 deletions(-)

diff --git a/include/linux/init.h b/include/linux/init.h
index 21d658cdfa27..42ae95411a93 100644
--- a/include/linux/init.h
+++ b/include/linux/init.h
@@ -275,13 +275,7 @@ void __init parse_early_param(void);
 
 #define security_initcall(fn)		module_init(fn)
 
-/* These macros create a dummy inline: gcc 2.9x does not count alias
- as usage, hence the `unused function' warning when __init functions
- are declared static. We use the dummy __*_module_inline functions
- both to kill the warning and check the type of the init/cleanup
- function. */
-
-/* Each module must use one module_init(), or one no_module_init */
+/* Each module must use one module_init(). */
 #define module_init(initfn)					\
 	static inline initcall_t __inittest(void)		\
 	{ return initfn; }					\

From 277e2c695907a70b316a31769cd891dc4d43b7f3 Mon Sep 17 00:00:00 2001
From: Daniel Guilak <guilak@linux.vnet.ibm.com>
Date: Fri, 25 Jul 2008 01:45:49 -0700
Subject: [PATCH 0704/1171] init/version.c: silence sparse warning by declaring
 the version string

Signed-off-by: Daniel Guilak <daniel@danielguilak.com>
Cc: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 init/version.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/init/version.c b/init/version.c
index 9d17d70ee02d..041fd822ce24 100644
--- a/init/version.c
+++ b/init/version.c
@@ -16,6 +16,7 @@
 #define version(a) Version_ ## a
 #define version_string(a) version(a)
 
+extern int version_string(LINUX_VERSION_CODE);
 int version_string(LINUX_VERSION_CODE);
 
 struct uts_namespace init_uts_ns = {

From 197dcffc8ba0ea943fee86e28e99cd9575799772 Mon Sep 17 00:00:00 2001
From: Daniel Guilak <guilak@linux.vnet.ibm.com>
Date: Fri, 25 Jul 2008 01:45:50 -0700
Subject: [PATCH 0705/1171] init/version.c: define version_string only if
 CONFIG_KALLSYMS is not defined

int Version_* is only used with ksymoops, which is only needed (according
to README and Documentation/Changes) if CONFIG_KALLSYMS is NOT defined.
Therefore this patch defines version_string only if CONFIG_KALLSYMS is not
defined.

Signed-off-by: Daniel Guilak <daniel@danielguilak.com>
Cc: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 init/version.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/init/version.c b/init/version.c
index 041fd822ce24..52a8b98642b8 100644
--- a/init/version.c
+++ b/init/version.c
@@ -13,11 +13,13 @@
 #include <linux/utsrelease.h>
 #include <linux/version.h>
 
+#ifndef CONFIG_KALLSYMS
 #define version(a) Version_ ## a
 #define version_string(a) version(a)
 
 extern int version_string(LINUX_VERSION_CODE);
 int version_string(LINUX_VERSION_CODE);
+#endif
 
 struct uts_namespace init_uts_ns = {
 	.kref = {

From a7f371e54fac49ff62bb640d4a7276fca01527e8 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Fri, 25 Jul 2008 01:45:51 -0700
Subject: [PATCH 0706/1171] documentation: update CodingStyle tips for Emacs
 users

Describe a setup that integrates better with Emacs' cc-mode and also fixes
up the alignment of continuation lines to really only use tabs.

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/CodingStyle | 38 +++++++++++++++++++++-----------------
 1 file changed, 21 insertions(+), 17 deletions(-)

diff --git a/Documentation/CodingStyle b/Documentation/CodingStyle
index 6caa14615578..1875e502f872 100644
--- a/Documentation/CodingStyle
+++ b/Documentation/CodingStyle
@@ -474,25 +474,29 @@ make a good program).
 So, you can either get rid of GNU emacs, or change it to use saner
 values.  To do the latter, you can stick the following in your .emacs file:
 
-(defun linux-c-mode ()
-  "C mode with adjusted defaults for use with the Linux kernel."
-  (interactive)
-  (c-mode)
-  (c-set-style "K&R")
-  (setq tab-width 8)
-  (setq indent-tabs-mode t)
-  (setq c-basic-offset 8))
+(defun c-lineup-arglist-tabs-only (ignored)
+  "Line up argument lists by tabs, not spaces"
+  (let* ((anchor (c-langelem-pos c-syntactic-element))
+	 (column (c-langelem-2nd-pos c-syntactic-element))
+	 (offset (- (1+ column) anchor))
+	 (steps (floor offset c-basic-offset)))
+    (* (max steps 1)
+       c-basic-offset)))
 
-This will define the M-x linux-c-mode command.  When hacking on a
-module, if you put the string -*- linux-c -*- somewhere on the first
-two lines, this mode will be automatically invoked. Also, you may want
-to add
+(add-hook 'c-mode-hook
+          (lambda ()
+            (let ((filename (buffer-file-name)))
+              ;; Enable kernel mode for the appropriate files
+              (when (and filename
+                         (string-match "~/src/linux-trees" filename))
+                (setq indent-tabs-mode t)
+                (c-set-style "linux")
+                (c-set-offset 'arglist-cont-nonempty
+                              '(c-lineup-gcc-asm-reg
+                                c-lineup-arglist-tabs-only))))))
 
-(setq auto-mode-alist (cons '("/usr/src/linux.*/.*\\.[ch]$" . linux-c-mode)
-			auto-mode-alist))
-
-to your .emacs file if you want to have linux-c-mode switched on
-automagically when you edit source files under /usr/src/linux.
+This will make emacs go better with the kernel coding style for C
+files below ~/src/linux-trees.
 
 But even if you fail in getting emacs to do sane formatting, not
 everything is lost: use "indent".

From f38954c93c4a548f55d73ac5c1cf5e7f4023bb6c Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@linux-foundation.org>
Date: Fri, 25 Jul 2008 01:45:52 -0700
Subject: [PATCH 0707/1171] drivers/misc/hpilo.c needs CONFIG_PCI

m68k allmodconfig:

drivers/misc/hpilo.c: In function 'ilo_ccb_close':
drivers/misc/hpilo.c:225: error: implicit declaration of function 'pci_free_consistent'
drivers/misc/hpilo.c: In function 'ilo_ccb_open':
drivers/misc/hpilo.c:244: error: implicit declaration of function 'pci_alloc_consistent'
drivers/misc/hpilo.c:245: warning: assignment makes pointer from integer without a cast

Cc: David Altobelli <david.altobelli@hp.com>
Cc: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/misc/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig
index 1689c051f68f..7e37ba5afe30 100644
--- a/drivers/misc/Kconfig
+++ b/drivers/misc/Kconfig
@@ -436,6 +436,7 @@ config SGI_XP
 
 config HP_ILO
 	tristate "Channel interface driver for HP iLO/iLO2 processor"
+	depends on PCI
 	default n
 	help
 	  The channel interface driver allows applications to communicate

From b6c63937001889af6fe431aaba97e59d04e028e7 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Fri, 25 Jul 2008 01:45:52 -0700
Subject: [PATCH 0708/1171] Rename WARN() to WARNING() to clear the namespace

We want to use WARN() as a variant of WARN_ON(), however a few drivers are
using WARN() internally.  This patch renames these to WARNING() to avoid the
namespace clash.  A few cases were defining but not using the thing, for those
cases I just deleted the definition.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Acked-by: Greg KH <greg@kroah.com>
Cc: Karsten Keil <kkeil@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/isdn/hisax/st5481.h       |  4 ++--
 drivers/isdn/hisax/st5481_b.c     |  4 ++--
 drivers/isdn/hisax/st5481_d.c     |  6 +++---
 drivers/isdn/hisax/st5481_usb.c   | 18 +++++++++---------
 drivers/usb/gadget/at91_udc.h     |  2 +-
 drivers/usb/gadget/cdc2.c         |  2 +-
 drivers/usb/gadget/ether.c        |  2 +-
 drivers/usb/gadget/file_storage.c | 14 +++++++-------
 drivers/usb/gadget/fsl_usb2_udc.c |  2 +-
 drivers/usb/gadget/fsl_usb2_udc.h |  2 +-
 drivers/usb/gadget/gmidi.c        |  2 --
 drivers/usb/gadget/goku_udc.c     |  2 +-
 drivers/usb/gadget/goku_udc.h     |  2 +-
 drivers/usb/gadget/inode.c        |  2 --
 drivers/usb/gadget/net2280.c      |  2 +-
 drivers/usb/gadget/net2280.h      |  2 +-
 drivers/usb/gadget/omap_udc.c     |  6 +++---
 drivers/usb/gadget/omap_udc.h     |  2 +-
 drivers/usb/gadget/printer.c      |  2 +-
 drivers/usb/gadget/pxa25x_udc.c   |  6 +++---
 drivers/usb/gadget/pxa25x_udc.h   |  2 +-
 drivers/usb/gadget/u_ether.c      |  3 ---
 drivers/usb/host/isp116x-hcd.c    |  2 +-
 drivers/usb/host/isp116x.h        |  2 +-
 drivers/usb/host/sl811-hcd.c      |  2 +-
 drivers/usb/host/sl811.h          |  2 +-
 drivers/usb/misc/usbtest.c        |  4 ++--
 include/linux/usb/composite.h     |  2 +-
 28 files changed, 48 insertions(+), 55 deletions(-)

diff --git a/drivers/isdn/hisax/st5481.h b/drivers/isdn/hisax/st5481.h
index 2044e7173ab4..cff7a6354334 100644
--- a/drivers/isdn/hisax/st5481.h
+++ b/drivers/isdn/hisax/st5481.h
@@ -220,7 +220,7 @@ enum {
 #define ERR(format, arg...) \
 printk(KERN_ERR "%s:%s: " format "\n" , __FILE__,  __func__ , ## arg)
 
-#define WARN(format, arg...) \
+#define WARNING(format, arg...) \
 printk(KERN_WARNING "%s:%s: " format "\n" , __FILE__,  __func__ , ## arg)
 
 #define INFO(format, arg...) \
@@ -412,7 +412,7 @@ struct st5481_adapter {
 ({ \
 	int status; \
 	if ((status = usb_submit_urb(urb, mem_flags)) < 0) { \
-		WARN("usb_submit_urb failed,status=%d", status); \
+		WARNING("usb_submit_urb failed,status=%d", status); \
 	} \
         status; \
 })
diff --git a/drivers/isdn/hisax/st5481_b.c b/drivers/isdn/hisax/st5481_b.c
index fa64115cd7c7..0074b600a0ef 100644
--- a/drivers/isdn/hisax/st5481_b.c
+++ b/drivers/isdn/hisax/st5481_b.c
@@ -180,7 +180,7 @@ static void usb_b_out_complete(struct urb *urb)
 				DBG(4,"urb killed status %d", urb->status);
 				return; // Give up
 			default: 
-				WARN("urb status %d",urb->status);
+				WARNING("urb status %d",urb->status);
 				if (b_out->busy == 0) {
 					st5481_usb_pipe_reset(adapter, (bcs->channel+1)*2 | USB_DIR_OUT, NULL, NULL);
 				}
@@ -372,6 +372,6 @@ void st5481_b_l2l1(struct hisax_if *ifc, int pr, void *arg)
 		B_L1L2(bcs, PH_DEACTIVATE | INDICATION, NULL);
 		break;
 	default:
-		WARN("pr %#x\n", pr);
+		WARNING("pr %#x\n", pr);
 	}
 }
diff --git a/drivers/isdn/hisax/st5481_d.c b/drivers/isdn/hisax/st5481_d.c
index b8c4855cc889..077991c1cd05 100644
--- a/drivers/isdn/hisax/st5481_d.c
+++ b/drivers/isdn/hisax/st5481_d.c
@@ -389,7 +389,7 @@ static void usb_d_out_complete(struct urb *urb)
 				DBG(1,"urb killed status %d", urb->status);
 				break;
 			default: 
-				WARN("urb status %d",urb->status);
+				WARNING("urb status %d",urb->status);
 				if (d_out->busy == 0) {
 					st5481_usb_pipe_reset(adapter, EP_D_OUT | USB_DIR_OUT, fifo_reseted, adapter);
 				}
@@ -420,7 +420,7 @@ static void dout_start_xmit(struct FsmInst *fsm, int event, void *arg)
 	isdnhdlc_out_init(&d_out->hdlc_state, 1, 0);
 
 	if (test_and_set_bit(buf_nr, &d_out->busy)) {
-		WARN("ep %d urb %d busy %#lx", EP_D_OUT, buf_nr, d_out->busy);
+		WARNING("ep %d urb %d busy %#lx", EP_D_OUT, buf_nr, d_out->busy);
 		return;
 	}
 	urb = d_out->urb[buf_nr];
@@ -601,7 +601,7 @@ void st5481_d_l2l1(struct hisax_if *hisax_d_if, int pr, void *arg)
 		FsmEvent(&adapter->d_out.fsm, EV_DOUT_START_XMIT, NULL);
 		break;
 	default:
-		WARN("pr %#x\n", pr);
+		WARNING("pr %#x\n", pr);
 		break;
 	}
 }
diff --git a/drivers/isdn/hisax/st5481_usb.c b/drivers/isdn/hisax/st5481_usb.c
index 427a8b0520f5..ec3c0e507669 100644
--- a/drivers/isdn/hisax/st5481_usb.c
+++ b/drivers/isdn/hisax/st5481_usb.c
@@ -66,7 +66,7 @@ static void usb_ctrl_msg(struct st5481_adapter *adapter,
 	struct ctrl_msg *ctrl_msg;
 	
 	if ((w_index = fifo_add(&ctrl->msg_fifo.f)) < 0) {
-		WARN("control msg FIFO full");
+		WARNING("control msg FIFO full");
 		return;
 	}
 	ctrl_msg = &ctrl->msg_fifo.data[w_index]; 
@@ -139,7 +139,7 @@ static void usb_ctrl_complete(struct urb *urb)
 				DBG(1,"urb killed status %d", urb->status);
 				return; // Give up
 			default: 
-				WARN("urb status %d",urb->status);
+				WARNING("urb status %d",urb->status);
 				break;
 		}
 	}
@@ -198,7 +198,7 @@ static void usb_int_complete(struct urb *urb)
 			DBG(2, "urb shutting down with status: %d", urb->status);
 			return;
 		default:
-			WARN("nonzero urb status received: %d", urb->status);
+			WARNING("nonzero urb status received: %d", urb->status);
 			goto exit;
 	}
 
@@ -235,7 +235,7 @@ static void usb_int_complete(struct urb *urb)
 exit:
 	status = usb_submit_urb (urb, GFP_ATOMIC);
 	if (status)
-		WARN("usb_submit_urb failed with result %d", status);
+		WARNING("usb_submit_urb failed with result %d", status);
 }
 
 /* ======================================================================
@@ -257,7 +257,7 @@ int st5481_setup_usb(struct st5481_adapter *adapter)
 	DBG(2,"");
 	
 	if ((status = usb_reset_configuration (dev)) < 0) {
-		WARN("reset_configuration failed,status=%d",status);
+		WARNING("reset_configuration failed,status=%d",status);
 		return status;
 	}
 
@@ -269,7 +269,7 @@ int st5481_setup_usb(struct st5481_adapter *adapter)
 
 	// Check if the config is sane
 	if ( altsetting->desc.bNumEndpoints != 7 ) {
-		WARN("expecting 7 got %d endpoints!", altsetting->desc.bNumEndpoints);
+		WARNING("expecting 7 got %d endpoints!", altsetting->desc.bNumEndpoints);
 		return -EINVAL;
 	}
 
@@ -279,7 +279,7 @@ int st5481_setup_usb(struct st5481_adapter *adapter)
 
 	// Use alternative setting 3 on interface 0 to have 2B+D
 	if ((status = usb_set_interface (dev, 0, 3)) < 0) {
-		WARN("usb_set_interface failed,status=%d",status);
+		WARNING("usb_set_interface failed,status=%d",status);
 		return status;
 	}
 
@@ -497,7 +497,7 @@ static void usb_in_complete(struct urb *urb)
 				DBG(1,"urb killed status %d", urb->status);
 				return; // Give up
 			default: 
-				WARN("urb status %d",urb->status);
+				WARNING("urb status %d",urb->status);
 				break;
 		}
 	}
@@ -523,7 +523,7 @@ static void usb_in_complete(struct urb *urb)
 			DBG(4,"count=%d",status);
 			DBG_PACKET(0x400, in->rcvbuf, status);
 			if (!(skb = dev_alloc_skb(status))) {
-				WARN("receive out of memory\n");
+				WARNING("receive out of memory\n");
 				break;
 			}
 			memcpy(skb_put(skb, status), in->rcvbuf, status);
diff --git a/drivers/usb/gadget/at91_udc.h b/drivers/usb/gadget/at91_udc.h
index a973f2a50fb9..c65d62295890 100644
--- a/drivers/usb/gadget/at91_udc.h
+++ b/drivers/usb/gadget/at91_udc.h
@@ -171,7 +171,7 @@ struct at91_request {
 #endif
 
 #define ERR(stuff...)		pr_err("udc: " stuff)
-#define WARN(stuff...)		pr_warning("udc: " stuff)
+#define WARNING(stuff...)	pr_warning("udc: " stuff)
 #define INFO(stuff...)		pr_info("udc: " stuff)
 #define DBG(stuff...)		pr_debug("udc: " stuff)
 
diff --git a/drivers/usb/gadget/cdc2.c b/drivers/usb/gadget/cdc2.c
index d490d0289507..a39a4b940c33 100644
--- a/drivers/usb/gadget/cdc2.c
+++ b/drivers/usb/gadget/cdc2.c
@@ -170,7 +170,7 @@ static int __init cdc_bind(struct usb_composite_dev *cdev)
 		 * but if the controller isn't recognized at all then
 		 * that assumption is a bit more likely to be wrong.
 		 */
-		WARN(cdev, "controller '%s' not recognized; trying %s\n",
+		WARNING(cdev, "controller '%s' not recognized; trying %s\n",
 				gadget->name,
 				cdc_config_driver.label);
 		device_desc.bcdDevice =
diff --git a/drivers/usb/gadget/ether.c b/drivers/usb/gadget/ether.c
index d7aaaa29b1e1..bcac2e68660d 100644
--- a/drivers/usb/gadget/ether.c
+++ b/drivers/usb/gadget/ether.c
@@ -293,7 +293,7 @@ static int __init eth_bind(struct usb_composite_dev *cdev)
 		 * but if the controller isn't recognized at all then
 		 * that assumption is a bit more likely to be wrong.
 		 */
-		WARN(cdev, "controller '%s' not recognized; trying %s\n",
+		WARNING(cdev, "controller '%s' not recognized; trying %s\n",
 				gadget->name,
 				eth_config_driver.label);
 		device_desc.bcdDevice =
diff --git a/drivers/usb/gadget/file_storage.c b/drivers/usb/gadget/file_storage.c
index 15c24edbb61a..ea2c31d18080 100644
--- a/drivers/usb/gadget/file_storage.c
+++ b/drivers/usb/gadget/file_storage.c
@@ -308,7 +308,7 @@ MODULE_LICENSE("Dual BSD/GPL");
 	dev_vdbg(&(d)->gadget->dev , fmt , ## args)
 #define ERROR(d, fmt, args...) \
 	dev_err(&(d)->gadget->dev , fmt , ## args)
-#define WARN(d, fmt, args...) \
+#define WARNING(d, fmt, args...) \
 	dev_warn(&(d)->gadget->dev , fmt , ## args)
 #define INFO(d, fmt, args...) \
 	dev_info(&(d)->gadget->dev , fmt , ## args)
@@ -1091,7 +1091,7 @@ static int ep0_queue(struct fsg_dev *fsg)
 	if (rc != 0 && rc != -ESHUTDOWN) {
 
 		/* We can't do much more than wait for a reset */
-		WARN(fsg, "error in submission: %s --> %d\n",
+		WARNING(fsg, "error in submission: %s --> %d\n",
 				fsg->ep0->name, rc);
 	}
 	return rc;
@@ -1227,7 +1227,7 @@ static void received_cbi_adsc(struct fsg_dev *fsg, struct fsg_buffhd *bh)
 
 	/* Save the command for later */
 	if (fsg->cbbuf_cmnd_size)
-		WARN(fsg, "CB[I] overwriting previous command\n");
+		WARNING(fsg, "CB[I] overwriting previous command\n");
 	fsg->cbbuf_cmnd_size = req->actual;
 	memcpy(fsg->cbbuf_cmnd, req->buf, fsg->cbbuf_cmnd_size);
 
@@ -1506,7 +1506,7 @@ static void start_transfer(struct fsg_dev *fsg, struct usb_ep *ep,
 		 * submissions if DMA is enabled. */
 		if (rc != -ESHUTDOWN && !(rc == -EOPNOTSUPP &&
 						req->length == 0))
-			WARN(fsg, "error in submission: %s --> %d\n",
+			WARNING(fsg, "error in submission: %s --> %d\n",
 					ep->name, rc);
 	}
 }
@@ -2294,7 +2294,7 @@ static int halt_bulk_in_endpoint(struct fsg_dev *fsg)
 		VDBG(fsg, "delayed bulk-in endpoint halt\n");
 	while (rc != 0) {
 		if (rc != -EAGAIN) {
-			WARN(fsg, "usb_ep_set_halt -> %d\n", rc);
+			WARNING(fsg, "usb_ep_set_halt -> %d\n", rc);
 			rc = 0;
 			break;
 		}
@@ -2317,7 +2317,7 @@ static int wedge_bulk_in_endpoint(struct fsg_dev *fsg)
 		VDBG(fsg, "delayed bulk-in endpoint wedge\n");
 	while (rc != 0) {
 		if (rc != -EAGAIN) {
-			WARN(fsg, "usb_ep_set_wedge -> %d\n", rc);
+			WARNING(fsg, "usb_ep_set_wedge -> %d\n", rc);
 			rc = 0;
 			break;
 		}
@@ -3755,7 +3755,7 @@ static int __init check_parameters(struct fsg_dev *fsg)
 		if (gcnum >= 0)
 			mod_data.release = 0x0300 + gcnum;
 		else {
-			WARN(fsg, "controller '%s' not recognized\n",
+			WARNING(fsg, "controller '%s' not recognized\n",
 				fsg->gadget->name);
 			mod_data.release = 0x0399;
 		}
diff --git a/drivers/usb/gadget/fsl_usb2_udc.c b/drivers/usb/gadget/fsl_usb2_udc.c
index 1695382f30fe..1cfccf102a2d 100644
--- a/drivers/usb/gadget/fsl_usb2_udc.c
+++ b/drivers/usb/gadget/fsl_usb2_udc.c
@@ -1538,7 +1538,7 @@ static void dtd_complete_irq(struct fsl_udc *udc)
 
 		/* If the ep is configured */
 		if (curr_ep->name == NULL) {
-			WARN("Invalid EP?");
+			WARNING("Invalid EP?");
 			continue;
 		}
 
diff --git a/drivers/usb/gadget/fsl_usb2_udc.h b/drivers/usb/gadget/fsl_usb2_udc.h
index 98b1483ef6a5..6131752a38bc 100644
--- a/drivers/usb/gadget/fsl_usb2_udc.h
+++ b/drivers/usb/gadget/fsl_usb2_udc.h
@@ -552,7 +552,7 @@ static void dump_msg(const char *label, const u8 * buf, unsigned int length)
 #endif
 
 #define ERR(stuff...)		pr_err("udc: " stuff)
-#define WARN(stuff...)		pr_warning("udc: " stuff)
+#define WARNING(stuff...)		pr_warning("udc: " stuff)
 #define INFO(stuff...)		pr_info("udc: " stuff)
 
 /*-------------------------------------------------------------------------*/
diff --git a/drivers/usb/gadget/gmidi.c b/drivers/usb/gadget/gmidi.c
index 7f4d4828e3aa..ea8651e3da1a 100644
--- a/drivers/usb/gadget/gmidi.c
+++ b/drivers/usb/gadget/gmidi.c
@@ -138,8 +138,6 @@ static void gmidi_transmit(struct gmidi_device* dev, struct usb_request* req);
 	dev_vdbg(&(d)->gadget->dev , fmt , ## args)
 #define ERROR(d, fmt, args...) \
 	dev_err(&(d)->gadget->dev , fmt , ## args)
-#define WARN(d, fmt, args...) \
-	dev_warn(&(d)->gadget->dev , fmt , ## args)
 #define INFO(d, fmt, args...) \
 	dev_info(&(d)->gadget->dev , fmt , ## args)
 
diff --git a/drivers/usb/gadget/goku_udc.c b/drivers/usb/gadget/goku_udc.c
index 48f1c63b7013..60aa04847b18 100644
--- a/drivers/usb/gadget/goku_udc.c
+++ b/drivers/usb/gadget/goku_udc.c
@@ -1768,7 +1768,7 @@ static int goku_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	 * usb_gadget_driver_{register,unregister}() must change.
 	 */
 	if (the_controller) {
-		WARN(dev, "ignoring %s\n", pci_name(pdev));
+		WARNING(dev, "ignoring %s\n", pci_name(pdev));
 		return -EBUSY;
 	}
 	if (!pdev->irq) {
diff --git a/drivers/usb/gadget/goku_udc.h b/drivers/usb/gadget/goku_udc.h
index bc4eb1e0b507..566cb2319056 100644
--- a/drivers/usb/gadget/goku_udc.h
+++ b/drivers/usb/gadget/goku_udc.h
@@ -285,7 +285,7 @@ struct goku_udc {
 
 #define ERROR(dev,fmt,args...) \
 	xprintk(dev , KERN_ERR , fmt , ## args)
-#define WARN(dev,fmt,args...) \
+#define WARNING(dev,fmt,args...) \
 	xprintk(dev , KERN_WARNING , fmt , ## args)
 #define INFO(dev,fmt,args...) \
 	xprintk(dev , KERN_INFO , fmt , ## args)
diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c
index 04692d59fc1c..f4585d3e90d7 100644
--- a/drivers/usb/gadget/inode.c
+++ b/drivers/usb/gadget/inode.c
@@ -262,8 +262,6 @@ static const char *CHIP;
 
 #define ERROR(dev,fmt,args...) \
 	xprintk(dev , KERN_ERR , fmt , ## args)
-#define WARN(dev,fmt,args...) \
-	xprintk(dev , KERN_WARNING , fmt , ## args)
 #define INFO(dev,fmt,args...) \
 	xprintk(dev , KERN_INFO , fmt , ## args)
 
diff --git a/drivers/usb/gadget/net2280.c b/drivers/usb/gadget/net2280.c
index b67ab677af72..5cfb5ebf3881 100644
--- a/drivers/usb/gadget/net2280.c
+++ b/drivers/usb/gadget/net2280.c
@@ -1007,7 +1007,7 @@ static void scan_dma_completions (struct net2280_ep *ep)
 			 * 0122, and 0124; not all cases trigger the warning.
 			 */
 			if ((tmp & (1 << NAK_OUT_PACKETS)) == 0) {
-				WARN (ep->dev, "%s lost packet sync!\n",
+				WARNING (ep->dev, "%s lost packet sync!\n",
 						ep->ep.name);
 				req->req.status = -EOVERFLOW;
 			} else if ((tmp = readl (&ep->regs->ep_avail)) != 0) {
diff --git a/drivers/usb/gadget/net2280.h b/drivers/usb/gadget/net2280.h
index 1f2af398a9a4..81a71dbdc2c6 100644
--- a/drivers/usb/gadget/net2280.h
+++ b/drivers/usb/gadget/net2280.h
@@ -272,7 +272,7 @@ static inline void net2280_led_shutdown (struct net2280 *dev)
 
 #define ERROR(dev,fmt,args...) \
 	xprintk(dev , KERN_ERR , fmt , ## args)
-#define WARN(dev,fmt,args...) \
+#define WARNING(dev,fmt,args...) \
 	xprintk(dev , KERN_WARNING , fmt , ## args)
 #define INFO(dev,fmt,args...) \
 	xprintk(dev , KERN_INFO , fmt , ## args)
diff --git a/drivers/usb/gadget/omap_udc.c b/drivers/usb/gadget/omap_udc.c
index 4b79a8509e84..395bd1844482 100644
--- a/drivers/usb/gadget/omap_udc.c
+++ b/drivers/usb/gadget/omap_udc.c
@@ -1120,7 +1120,7 @@ static int omap_ep_set_halt(struct usb_ep *_ep, int value)
 			status = -EINVAL;
 		else if (value) {
 			if (ep->udc->ep0_set_config) {
-				WARN("error changing config?\n");
+				WARNING("error changing config?\n");
 				omap_writew(UDC_CLR_CFG, UDC_SYSCON2);
 			}
 			omap_writew(UDC_STALL_CMD, UDC_SYSCON2);
@@ -1764,7 +1764,7 @@ do_stall:
 					u.r.bRequestType, u.r.bRequest, status);
 			if (udc->ep0_set_config) {
 				if (udc->ep0_reset_config)
-					WARN("error resetting config?\n");
+					WARNING("error resetting config?\n");
 				else
 					omap_writew(UDC_CLR_CFG, UDC_SYSCON2);
 			}
@@ -3076,7 +3076,7 @@ static int omap_udc_suspend(struct platform_device *dev, pm_message_t message)
 	 * which would prevent entry to deep sleep...
 	 */
 	if ((devstat & UDC_ATT) != 0 && (devstat & UDC_SUS) == 0) {
-		WARN("session active; suspend requires disconnect\n");
+		WARNING("session active; suspend requires disconnect\n");
 		omap_pullup(&udc->gadget, 0);
 	}
 
diff --git a/drivers/usb/gadget/omap_udc.h b/drivers/usb/gadget/omap_udc.h
index 8522bbb12278..29edc51b6b22 100644
--- a/drivers/usb/gadget/omap_udc.h
+++ b/drivers/usb/gadget/omap_udc.h
@@ -188,7 +188,7 @@ struct omap_udc {
 #endif
 
 #define ERR(stuff...)		pr_err("udc: " stuff)
-#define WARN(stuff...)		pr_warning("udc: " stuff)
+#define WARNING(stuff...)	pr_warning("udc: " stuff)
 #define INFO(stuff...)		pr_info("udc: " stuff)
 #define DBG(stuff...)		pr_debug("udc: " stuff)
 
diff --git a/drivers/usb/gadget/printer.c b/drivers/usb/gadget/printer.c
index 49cd9e145a9b..e0090085b78e 100644
--- a/drivers/usb/gadget/printer.c
+++ b/drivers/usb/gadget/printer.c
@@ -179,7 +179,7 @@ module_param(qlen, uint, S_IRUGO|S_IWUSR);
 
 #define ERROR(dev, fmt, args...) \
 	xprintk(dev, KERN_ERR, fmt, ## args)
-#define WARN(dev, fmt, args...) \
+#define WARNING(dev, fmt, args...) \
 	xprintk(dev, KERN_WARNING, fmt, ## args)
 #define INFO(dev, fmt, args...) \
 	xprintk(dev, KERN_INFO, fmt, ## args)
diff --git a/drivers/usb/gadget/pxa25x_udc.c b/drivers/usb/gadget/pxa25x_udc.c
index 8fb0066609bb..7e6725d89976 100644
--- a/drivers/usb/gadget/pxa25x_udc.c
+++ b/drivers/usb/gadget/pxa25x_udc.c
@@ -342,7 +342,7 @@ pxa25x_ep_free_request (struct usb_ep *_ep, struct usb_request *_req)
 	struct pxa25x_request	*req;
 
 	req = container_of (_req, struct pxa25x_request, req);
-	WARN_ON (!list_empty (&req->queue));
+	WARN_ON(!list_empty (&req->queue));
 	kfree(req);
 }
 
@@ -1556,7 +1556,7 @@ config_change:
 					 * tell us about config change events,
 					 * so later ones may fail...
 					 */
-					WARN("config change %02x fail %d?\n",
+					WARNING("config change %02x fail %d?\n",
 						u.r.bRequest, i);
 					return;
 					/* TODO experiment:  if has_cfr,
@@ -2330,7 +2330,7 @@ static int pxa25x_udc_suspend(struct platform_device *dev, pm_message_t state)
 	unsigned long flags;
 
 	if (!udc->mach->gpio_pullup && !udc->mach->udc_command)
-		WARN("USB host won't detect disconnect!\n");
+		WARNING("USB host won't detect disconnect!\n");
 	udc->suspended = 1;
 
 	local_irq_save(flags);
diff --git a/drivers/usb/gadget/pxa25x_udc.h b/drivers/usb/gadget/pxa25x_udc.h
index 4d11ece7c95f..c8a13215e02c 100644
--- a/drivers/usb/gadget/pxa25x_udc.h
+++ b/drivers/usb/gadget/pxa25x_udc.h
@@ -259,7 +259,7 @@ dump_state(struct pxa25x_udc *dev)
 #define DBG(lvl, stuff...) do{if ((lvl) <= UDC_DEBUG) DMSG(stuff);}while(0)
 
 #define ERR(stuff...)		pr_err("udc: " stuff)
-#define WARN(stuff...)		pr_warning("udc: " stuff)
+#define WARNING(stuff...)	pr_warning("udc: " stuff)
 #define INFO(stuff...)		pr_info("udc: " stuff)
 
 
diff --git a/drivers/usb/gadget/u_ether.c b/drivers/usb/gadget/u_ether.c
index 5458f43a8668..3791e6271903 100644
--- a/drivers/usb/gadget/u_ether.c
+++ b/drivers/usb/gadget/u_ether.c
@@ -116,7 +116,6 @@ static inline int qlen(struct usb_gadget *gadget)
 #undef DBG
 #undef VDBG
 #undef ERROR
-#undef WARN
 #undef INFO
 
 #define xprintk(d, level, fmt, args...) \
@@ -140,8 +139,6 @@ static inline int qlen(struct usb_gadget *gadget)
 
 #define ERROR(dev, fmt, args...) \
 	xprintk(dev , KERN_ERR , fmt , ## args)
-#define WARN(dev, fmt, args...) \
-	xprintk(dev , KERN_WARNING , fmt , ## args)
 #define INFO(dev, fmt, args...) \
 	xprintk(dev , KERN_INFO , fmt , ## args)
 
diff --git a/drivers/usb/host/isp116x-hcd.c b/drivers/usb/host/isp116x-hcd.c
index 31178e10cbbe..ce1ca0ba0515 100644
--- a/drivers/usb/host/isp116x-hcd.c
+++ b/drivers/usb/host/isp116x-hcd.c
@@ -882,7 +882,7 @@ static void isp116x_endpoint_disable(struct usb_hcd *hcd,
 	for (i = 0; i < 100 && !list_empty(&hep->urb_list); i++)
 		msleep(3);
 	if (!list_empty(&hep->urb_list))
-		WARN("ep %p not empty?\n", ep);
+		WARNING("ep %p not empty?\n", ep);
 
 	kfree(ep);
 	hep->hcpriv = NULL;
diff --git a/drivers/usb/host/isp116x.h b/drivers/usb/host/isp116x.h
index 595b90a99848..aa211bafcff9 100644
--- a/drivers/usb/host/isp116x.h
+++ b/drivers/usb/host/isp116x.h
@@ -338,7 +338,7 @@ struct isp116x_ep {
 #endif
 
 #define ERR(stuff...)		printk(KERN_ERR "116x: " stuff)
-#define WARN(stuff...)		printk(KERN_WARNING "116x: " stuff)
+#define WARNING(stuff...)	printk(KERN_WARNING "116x: " stuff)
 #define INFO(stuff...)		printk(KERN_INFO "116x: " stuff)
 
 /* ------------------------------------------------- */
diff --git a/drivers/usb/host/sl811-hcd.c b/drivers/usb/host/sl811-hcd.c
index 340d72da554a..8a74bbb57d08 100644
--- a/drivers/usb/host/sl811-hcd.c
+++ b/drivers/usb/host/sl811-hcd.c
@@ -1026,7 +1026,7 @@ sl811h_endpoint_disable(struct usb_hcd *hcd, struct usb_host_endpoint *hep)
 	if (!list_empty(&hep->urb_list))
 		msleep(3);
 	if (!list_empty(&hep->urb_list))
-		WARN("ep %p not empty?\n", ep);
+		WARNING("ep %p not empty?\n", ep);
 
 	kfree(ep);
 	hep->hcpriv = NULL;
diff --git a/drivers/usb/host/sl811.h b/drivers/usb/host/sl811.h
index 7690d98e42a7..b6b8c1f233dd 100644
--- a/drivers/usb/host/sl811.h
+++ b/drivers/usb/host/sl811.h
@@ -261,6 +261,6 @@ sl811_read_buf(struct sl811 *sl811, int addr, void *buf, size_t count)
 #endif
 
 #define ERR(stuff...)		printk(KERN_ERR "sl811: " stuff)
-#define WARN(stuff...)		printk(KERN_WARNING "sl811: " stuff)
+#define WARNING(stuff...)	printk(KERN_WARNING "sl811: " stuff)
 #define INFO(stuff...)		printk(KERN_INFO "sl811: " stuff)
 
diff --git a/drivers/usb/misc/usbtest.c b/drivers/usb/misc/usbtest.c
index 054dedd28127..b358c4e1cf21 100644
--- a/drivers/usb/misc/usbtest.c
+++ b/drivers/usb/misc/usbtest.c
@@ -81,7 +81,7 @@ static struct usb_device *testdev_to_usbdev (struct usbtest_dev *test)
 
 #define ERROR(tdev, fmt, args...) \
 	dev_err(&(tdev)->intf->dev , fmt , ## args)
-#define WARN(tdev, fmt, args...) \
+#define WARNING(tdev, fmt, args...) \
 	dev_warn(&(tdev)->intf->dev , fmt , ## args)
 
 /*-------------------------------------------------------------------------*/
@@ -1946,7 +1946,7 @@ usbtest_probe (struct usb_interface *intf, const struct usb_device_id *id)
 
 			status = get_endpoints (dev, intf);
 			if (status < 0) {
-				WARN(dev, "couldn't get endpoints, %d\n",
+				WARNING(dev, "couldn't get endpoints, %d\n",
 						status);
 				return status;
 			}
diff --git a/include/linux/usb/composite.h b/include/linux/usb/composite.h
index 747c3a49cdc9..c932390c6da0 100644
--- a/include/linux/usb/composite.h
+++ b/include/linux/usb/composite.h
@@ -330,7 +330,7 @@ extern int usb_string_id(struct usb_composite_dev *c);
 	dev_vdbg(&(d)->gadget->dev , fmt , ## args)
 #define ERROR(d, fmt, args...) \
 	dev_err(&(d)->gadget->dev , fmt , ## args)
-#define WARN(d, fmt, args...) \
+#define WARNING(d, fmt, args...) \
 	dev_warn(&(d)->gadget->dev , fmt , ## args)
 #define INFO(d, fmt, args...) \
 	dev_info(&(d)->gadget->dev , fmt , ## args)

From a8f18b909c0a3f22630846207035c8b84bb252b8 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Fri, 25 Jul 2008 01:45:53 -0700
Subject: [PATCH 0709/1171] Add a WARN() macro; this is WARN_ON() + printk
 arguments

Add a WARN() macro that acts like WARN_ON(), with the added feature that it
takes a printk like argument that is printed as part of the warning message.

[akpm@linux-foundation.org: fix printk arguments]
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Cc: Greg KH <greg@kroah.com>
Cc: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/asm-generic/bug.h | 22 ++++++++++++++++++++++
 kernel/panic.c            | 22 ++++++++++++++++++++++
 2 files changed, 44 insertions(+)

diff --git a/include/asm-generic/bug.h b/include/asm-generic/bug.h
index 2632328d8646..a346e744e770 100644
--- a/include/asm-generic/bug.h
+++ b/include/asm-generic/bug.h
@@ -34,9 +34,14 @@ struct bug_entry {
 #ifndef __WARN
 #ifndef __ASSEMBLY__
 extern void warn_on_slowpath(const char *file, const int line);
+extern void warn_slowpath(const char *file, const int line,
+		const char *fmt, ...) __attribute__((format(printf, 3, 4)));
 #define WANT_WARN_ON_SLOWPATH
 #endif
 #define __WARN() warn_on_slowpath(__FILE__, __LINE__)
+#define __WARN_printf(arg...) warn_slowpath(__FILE__, __LINE__, arg)
+#else
+#define __WARN_printf(arg...) __WARN()
 #endif
 
 #ifndef WARN_ON
@@ -48,6 +53,15 @@ extern void warn_on_slowpath(const char *file, const int line);
 })
 #endif
 
+#ifndef WARN
+#define WARN(condition, format...) ({						\
+	int __ret_warn_on = !!(condition);				\
+	if (unlikely(__ret_warn_on))					\
+		__WARN_printf(format);					\
+	unlikely(__ret_warn_on);					\
+})
+#endif
+
 #else /* !CONFIG_BUG */
 #ifndef HAVE_ARCH_BUG
 #define BUG()
@@ -63,6 +77,14 @@ extern void warn_on_slowpath(const char *file, const int line);
 	unlikely(__ret_warn_on);					\
 })
 #endif
+
+#ifndef WARN
+#define WARN(condition, format...) ({					\
+	int __ret_warn_on = !!(condition);				\
+	unlikely(__ret_warn_on);					\
+})
+#endif
+
 #endif
 
 #define WARN_ON_ONCE(condition)	({				\
diff --git a/kernel/panic.c b/kernel/panic.c
index 425567f45b9f..12c5a0a6c89b 100644
--- a/kernel/panic.c
+++ b/kernel/panic.c
@@ -318,6 +318,28 @@ void warn_on_slowpath(const char *file, int line)
 	add_taint(TAINT_WARN);
 }
 EXPORT_SYMBOL(warn_on_slowpath);
+
+
+void warn_slowpath(const char *file, int line, const char *fmt, ...)
+{
+	va_list args;
+	char function[KSYM_SYMBOL_LEN];
+	unsigned long caller = (unsigned long)__builtin_return_address(0);
+	sprint_symbol(function, caller);
+
+	printk(KERN_WARNING "------------[ cut here ]------------\n");
+	printk(KERN_WARNING "WARNING: at %s:%d %s()\n", file,
+		line, function);
+	va_start(args, fmt);
+	vprintk(fmt, args);
+	va_end(args);
+
+	print_modules();
+	dump_stack();
+	print_oops_end_marker();
+	add_taint(TAINT_WARN);
+}
+EXPORT_SYMBOL(warn_slowpath);
 #endif
 
 #ifdef CONFIG_CC_STACKPROTECTOR

From 7a2c477069fbd32f91598f05334003979b987a39 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Fri, 25 Jul 2008 01:45:54 -0700
Subject: [PATCH 0710/1171] kernel/irq/manage.c: replace a printk + WARN_ON()
 to a WARN()

Replace a printk+WARN_ON() by a WARN(); this increases the chance of the
string making it into the bugreport (ie: it goes inside the
---[ cut here ]--- section)

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/irq/manage.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index 5bc6e5ecc493..f8914b92b664 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -260,9 +260,7 @@ int set_irq_wake(unsigned int irq, unsigned int on)
 		}
 	} else {
 		if (desc->wake_depth == 0) {
-			printk(KERN_WARNING "Unbalanced IRQ %d "
-					"wake disable\n", irq);
-			WARN_ON(1);
+			WARN(1, "Unbalanced IRQ %d wake disable\n", irq);
 		} else if (--desc->wake_depth == 0) {
 			ret = set_irq_wake_real(irq, on);
 			if (ret)

From d955c78ac4699ac9c3fe07be62982cda13d13267 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Fri, 25 Jul 2008 01:45:55 -0700
Subject: [PATCH 0711/1171] Example use of WARN()

Now that WARN() exists, we can fold some of the printk's into it.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 lib/kobject.c | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)

diff --git a/lib/kobject.c b/lib/kobject.c
index 744401571ed7..bd732ffebc85 100644
--- a/lib/kobject.c
+++ b/lib/kobject.c
@@ -164,9 +164,8 @@ static int kobject_add_internal(struct kobject *kobj)
 		return -ENOENT;
 
 	if (!kobj->name || !kobj->name[0]) {
-		pr_debug("kobject: (%p): attempted to be registered with empty "
+		WARN(1, "kobject: (%p): attempted to be registered with empty "
 			 "name!\n", kobj);
-		WARN_ON(1);
 		return -EINVAL;
 	}
 
@@ -583,12 +582,10 @@ static void kobject_release(struct kref *kref)
 void kobject_put(struct kobject *kobj)
 {
 	if (kobj) {
-		if (!kobj->state_initialized) {
-			printk(KERN_WARNING "kobject: '%s' (%p): is not "
+		if (!kobj->state_initialized)
+			WARN(1, KERN_WARNING "kobject: '%s' (%p): is not "
 			       "initialized, yet kobject_put() is being "
 			       "called.\n", kobject_name(kobj), kobj);
-			WARN_ON(1);
-		}
 		kref_put(&kobj->kref, kobject_release);
 	}
 }

From 924d9addb9b1474fc81a78a5c6706755efea7aaa Mon Sep 17 00:00:00 2001
From: Dave Jones <davej@redhat.com>
Date: Fri, 25 Jul 2008 01:45:55 -0700
Subject: [PATCH 0712/1171] list debugging: use WARN() instead of BUG()

Arjan noted that the list_head debugging is BUG'ing when it detects
corruption.  By causing the box to panic immediately, we're possibly
losing some bug reports.  Changing this to a WARN() should mean we at the
least start seeing reports collected at kerneloops.org

Signed-off-by: Dave Jones <davej@redhat.com>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Arjan van de Ven <arjan@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 lib/list_debug.c | 36 ++++++++++++++----------------------
 1 file changed, 14 insertions(+), 22 deletions(-)

diff --git a/lib/list_debug.c b/lib/list_debug.c
index 45c03fd608dd..1a39f4e3ae1f 100644
--- a/lib/list_debug.c
+++ b/lib/list_debug.c
@@ -20,18 +20,14 @@ void __list_add(struct list_head *new,
 			      struct list_head *prev,
 			      struct list_head *next)
 {
-	if (unlikely(next->prev != prev)) {
-		printk(KERN_ERR "list_add corruption. next->prev should be "
-			"prev (%p), but was %p. (next=%p).\n",
-			prev, next->prev, next);
-		BUG();
-	}
-	if (unlikely(prev->next != next)) {
-		printk(KERN_ERR "list_add corruption. prev->next should be "
-			"next (%p), but was %p. (prev=%p).\n",
-			next, prev->next, prev);
-		BUG();
-	}
+	WARN(next->prev != prev,
+		"list_add corruption. next->prev should be "
+		"prev (%p), but was %p. (next=%p).\n",
+		prev, next->prev, next);
+	WARN(prev->next != next,
+		"list_add corruption. prev->next should be "
+		"next (%p), but was %p. (prev=%p).\n",
+		next, prev->next, prev);
 	next->prev = new;
 	new->next = next;
 	new->prev = prev;
@@ -47,16 +43,12 @@ EXPORT_SYMBOL(__list_add);
  */
 void list_del(struct list_head *entry)
 {
-	if (unlikely(entry->prev->next != entry)) {
-		printk(KERN_ERR "list_del corruption. prev->next should be %p, "
-				"but was %p\n", entry, entry->prev->next);
-		BUG();
-	}
-	if (unlikely(entry->next->prev != entry)) {
-		printk(KERN_ERR "list_del corruption. next->prev should be %p, "
-				"but was %p\n", entry, entry->next->prev);
-		BUG();
-	}
+	WARN(entry->prev->next != entry,
+		"list_del corruption. prev->next should be %p, "
+		"but was %p\n", entry, entry->prev->next);
+	WARN(entry->next->prev != entry,
+		"list_del corruption. next->prev should be %p, "
+		"but was %p\n", entry, entry->next->prev);
 	__list_del(entry->prev, entry->next);
 	entry->next = LIST_POISON1;
 	entry->prev = LIST_POISON2;

From 2711b793eb62a5873a0ba583a69252040aef176e Mon Sep 17 00:00:00 2001
From: Vegard Nossum <vegard.nossum@gmail.com>
Date: Fri, 25 Jul 2008 01:45:56 -0700
Subject: [PATCH 0713/1171] kallsyms: unify 32- and 64-bit code

Use the %p format string which already accounts for the padding you need
with a pointer type on a particular architecture.

Also replace the macro with a static inline function to match the rest of
the file.

Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Arjan van de Ven <arjan@infradead.org>
Signed-off-by: Vegard Nossum <vegard.nossum@gmail.com>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/kallsyms.h | 19 ++++++-------------
 1 file changed, 6 insertions(+), 13 deletions(-)

diff --git a/include/linux/kallsyms.h b/include/linux/kallsyms.h
index 00c1801099fa..57aefa160a92 100644
--- a/include/linux/kallsyms.h
+++ b/include/linux/kallsyms.h
@@ -6,6 +6,7 @@
 #define _LINUX_KALLSYMS_H
 
 #include <linux/errno.h>
+#include <linux/kernel.h>
 #include <linux/stddef.h>
 
 #define KSYM_NAME_LEN 128
@@ -105,18 +106,10 @@ static inline void print_fn_descriptor_symbol(const char *fmt, void *addr)
 	print_symbol(fmt, (unsigned long)addr);
 }
 
-#ifndef CONFIG_64BIT
-#define print_ip_sym(ip)		\
-do {					\
-	printk("[<%08lx>]", ip);	\
-	print_symbol(" %s\n", ip);	\
-} while(0)
-#else
-#define print_ip_sym(ip)		\
-do {					\
-	printk("[<%016lx>]", ip);	\
-	print_symbol(" %s\n", ip);	\
-} while(0)
-#endif
+static inline void print_ip_sym(unsigned long ip)
+{
+	printk("[<%p>]", (void *) ip);
+	print_symbol(" %s\n", ip);
+}
 
 #endif /*_LINUX_KALLSYMS_H*/

From 717115e1a5856b57af0f71e1df7149108294fc10 Mon Sep 17 00:00:00 2001
From: Dave Young <hidave.darkstar@gmail.com>
Date: Fri, 25 Jul 2008 01:45:58 -0700
Subject: [PATCH 0714/1171] printk ratelimiting rewrite

All ratelimit user use same jiffies and burst params, so some messages
(callbacks) will be lost.

For example:
a call printk_ratelimit(5 * HZ, 1)
b call printk_ratelimit(5 * HZ, 1) before the 5*HZ timeout of a, then b will
will be supressed.

- rewrite __ratelimit, and use a ratelimit_state as parameter.  Thanks for
  hints from andrew.

- Add WARN_ON_RATELIMIT, update rcupreempt.h

- remove __printk_ratelimit

- use __ratelimit in net_ratelimit

Signed-off-by: Dave Young <hidave.darkstar@gmail.com>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Cc: Dave Young <hidave.darkstar@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/asm-generic/bug.h  |  3 +++
 include/linux/kernel.h     |  8 ++----
 include/linux/net.h        |  3 +--
 include/linux/ratelimit.h  | 27 +++++++++++++++++++
 include/linux/rcupreempt.h |  9 +++++--
 kernel/printk.c            | 17 +++---------
 kernel/sysctl.c            |  4 +--
 lib/ratelimit.c            | 55 +++++++++++++++++++++-----------------
 net/core/sysctl_net_core.c |  4 +--
 net/core/utils.c           |  5 ++--
 10 files changed, 79 insertions(+), 56 deletions(-)
 create mode 100644 include/linux/ratelimit.h

diff --git a/include/asm-generic/bug.h b/include/asm-generic/bug.h
index a346e744e770..a3f738cffdb6 100644
--- a/include/asm-generic/bug.h
+++ b/include/asm-generic/bug.h
@@ -97,6 +97,9 @@ extern void warn_slowpath(const char *file, const int line,
 	unlikely(__ret_warn_once);				\
 })
 
+#define WARN_ON_RATELIMIT(condition, state)			\
+		WARN_ON((condition) && __ratelimit(state))
+
 #ifdef CONFIG_SMP
 # define WARN_ON_SMP(x)			WARN_ON(x)
 #else
diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 5c4b1251e110..fdbbf72ca2eb 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -15,6 +15,7 @@
 #include <linux/bitops.h>
 #include <linux/log2.h>
 #include <linux/typecheck.h>
+#include <linux/ratelimit.h>
 #include <asm/byteorder.h>
 #include <asm/bug.h>
 
@@ -189,11 +190,8 @@ asmlinkage int vprintk(const char *fmt, va_list args)
 asmlinkage int printk(const char * fmt, ...)
 	__attribute__ ((format (printf, 1, 2))) __cold;
 
-extern int printk_ratelimit_jiffies;
-extern int printk_ratelimit_burst;
+extern struct ratelimit_state printk_ratelimit_state;
 extern int printk_ratelimit(void);
-extern int __ratelimit(int ratelimit_jiffies, int ratelimit_burst);
-extern int __printk_ratelimit(int ratelimit_jiffies, int ratelimit_burst);
 extern bool printk_timed_ratelimit(unsigned long *caller_jiffies,
 				   unsigned int interval_msec);
 #else
@@ -204,8 +202,6 @@ static inline int printk(const char *s, ...)
 	__attribute__ ((format (printf, 1, 2)));
 static inline int __cold printk(const char *s, ...) { return 0; }
 static inline int printk_ratelimit(void) { return 0; }
-static inline int __printk_ratelimit(int ratelimit_jiffies, \
-				     int ratelimit_burst) { return 0; }
 static inline bool printk_timed_ratelimit(unsigned long *caller_jiffies, \
 					  unsigned int interval_msec)	\
 		{ return false; }
diff --git a/include/linux/net.h b/include/linux/net.h
index 2f999fbb188d..4a9a30f2d68f 100644
--- a/include/linux/net.h
+++ b/include/linux/net.h
@@ -351,8 +351,7 @@ static const struct proto_ops name##_ops = {			\
 
 #ifdef CONFIG_SYSCTL
 #include <linux/sysctl.h>
-extern int net_msg_cost;
-extern int net_msg_burst;
+extern struct ratelimit_state net_ratelimit_state;
 #endif
 
 #endif /* __KERNEL__ */
diff --git a/include/linux/ratelimit.h b/include/linux/ratelimit.h
new file mode 100644
index 000000000000..18a5b9ba9d40
--- /dev/null
+++ b/include/linux/ratelimit.h
@@ -0,0 +1,27 @@
+#ifndef _LINUX_RATELIMIT_H
+#define _LINUX_RATELIMIT_H
+#include <linux/param.h>
+
+#define DEFAULT_RATELIMIT_INTERVAL (5 * HZ)
+#define DEFAULT_RATELIMIT_BURST 10
+
+struct ratelimit_state {
+	int interval;
+	int burst;
+	int printed;
+	int missed;
+	unsigned long begin;
+};
+
+#define DEFINE_RATELIMIT_STATE(name, interval, burst)		\
+		struct ratelimit_state name = {interval, burst,}
+
+extern int __ratelimit(struct ratelimit_state *rs);
+
+static inline int ratelimit(void)
+{
+	static DEFINE_RATELIMIT_STATE(rs, DEFAULT_RATELIMIT_INTERVAL,
+					DEFAULT_RATELIMIT_BURST);
+	return __ratelimit(&rs);
+}
+#endif
diff --git a/include/linux/rcupreempt.h b/include/linux/rcupreempt.h
index f04b64eca636..0967f03b0705 100644
--- a/include/linux/rcupreempt.h
+++ b/include/linux/rcupreempt.h
@@ -115,16 +115,21 @@ DECLARE_PER_CPU(struct rcu_dyntick_sched, rcu_dyntick_sched);
 
 static inline void rcu_enter_nohz(void)
 {
+	static DEFINE_RATELIMIT_STATE(rs, 10 * HZ, 1);
+
 	smp_mb(); /* CPUs seeing ++ must see prior RCU read-side crit sects */
 	__get_cpu_var(rcu_dyntick_sched).dynticks++;
-	WARN_ON(__get_cpu_var(rcu_dyntick_sched).dynticks & 0x1);
+	WARN_ON_RATELIMIT(__get_cpu_var(rcu_dyntick_sched).dynticks & 0x1, &rs);
 }
 
 static inline void rcu_exit_nohz(void)
 {
+	static DEFINE_RATELIMIT_STATE(rs, 10 * HZ, 1);
+
 	smp_mb(); /* CPUs seeing ++ must see later RCU read-side crit sects */
 	__get_cpu_var(rcu_dyntick_sched).dynticks++;
-	WARN_ON(!(__get_cpu_var(rcu_dyntick_sched).dynticks & 0x1));
+	WARN_ON_RATELIMIT(!(__get_cpu_var(rcu_dyntick_sched).dynticks & 0x1),
+				&rs);
 }
 
 #else /* CONFIG_NO_HZ */
diff --git a/kernel/printk.c b/kernel/printk.c
index 3f7a2a94583b..a7f7559c5f6c 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -1308,6 +1308,8 @@ void tty_write_message(struct tty_struct *tty, char *msg)
 }
 
 #if defined CONFIG_PRINTK
+
+DEFINE_RATELIMIT_STATE(printk_ratelimit_state, 5 * HZ, 10);
 /*
  * printk rate limiting, lifted from the networking subsystem.
  *
@@ -1315,22 +1317,9 @@ void tty_write_message(struct tty_struct *tty, char *msg)
  * every printk_ratelimit_jiffies to make a denial-of-service
  * attack impossible.
  */
-int __printk_ratelimit(int ratelimit_jiffies, int ratelimit_burst)
-{
-	return __ratelimit(ratelimit_jiffies, ratelimit_burst);
-}
-EXPORT_SYMBOL(__printk_ratelimit);
-
-/* minimum time in jiffies between messages */
-int printk_ratelimit_jiffies = 5 * HZ;
-
-/* number of messages we send before ratelimiting */
-int printk_ratelimit_burst = 10;
-
 int printk_ratelimit(void)
 {
-	return __printk_ratelimit(printk_ratelimit_jiffies,
-				printk_ratelimit_burst);
+	return __ratelimit(&printk_ratelimit_state);
 }
 EXPORT_SYMBOL(printk_ratelimit);
 
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 1a8299d1fe59..35a50db9b6ce 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -624,7 +624,7 @@ static struct ctl_table kern_table[] = {
 	{
 		.ctl_name	= KERN_PRINTK_RATELIMIT,
 		.procname	= "printk_ratelimit",
-		.data		= &printk_ratelimit_jiffies,
+		.data		= &printk_ratelimit_state.interval,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec_jiffies,
@@ -633,7 +633,7 @@ static struct ctl_table kern_table[] = {
 	{
 		.ctl_name	= KERN_PRINTK_RATELIMIT_BURST,
 		.procname	= "printk_ratelimit_burst",
-		.data		= &printk_ratelimit_burst,
+		.data		= &printk_ratelimit_state.burst,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec,
diff --git a/lib/ratelimit.c b/lib/ratelimit.c
index 485e3040dcd4..35136671b215 100644
--- a/lib/ratelimit.c
+++ b/lib/ratelimit.c
@@ -3,6 +3,9 @@
  *
  * Isolated from kernel/printk.c by Dave Young <hidave.darkstar@gmail.com>
  *
+ * 2008-05-01 rewrite the function and use a ratelimit_state data struct as
+ * parameter. Now every user can use their own standalone ratelimit_state.
+ *
  * This file is released under the GPLv2.
  *
  */
@@ -11,41 +14,43 @@
 #include <linux/jiffies.h>
 #include <linux/module.h>
 
+static DEFINE_SPINLOCK(ratelimit_lock);
+static unsigned long flags;
+
 /*
  * __ratelimit - rate limiting
- * @ratelimit_jiffies: minimum time in jiffies between two callbacks
- * @ratelimit_burst: number of callbacks we do before ratelimiting
+ * @rs: ratelimit_state data
  *
- * This enforces a rate limit: not more than @ratelimit_burst callbacks
- * in every ratelimit_jiffies
+ * This enforces a rate limit: not more than @rs->ratelimit_burst callbacks
+ * in every @rs->ratelimit_jiffies
  */
-int __ratelimit(int ratelimit_jiffies, int ratelimit_burst)
+int __ratelimit(struct ratelimit_state *rs)
 {
-	static DEFINE_SPINLOCK(ratelimit_lock);
-	static unsigned toks = 10 * 5 * HZ;
-	static unsigned long last_msg;
-	static int missed;
-	unsigned long flags;
-	unsigned long now = jiffies;
+	if (!rs->interval)
+		return 1;
 
 	spin_lock_irqsave(&ratelimit_lock, flags);
-	toks += now - last_msg;
-	last_msg = now;
-	if (toks > (ratelimit_burst * ratelimit_jiffies))
-		toks = ratelimit_burst * ratelimit_jiffies;
-	if (toks >= ratelimit_jiffies) {
-		int lost = missed;
+	if (!rs->begin)
+		rs->begin = jiffies;
 
-		missed = 0;
-		toks -= ratelimit_jiffies;
-		spin_unlock_irqrestore(&ratelimit_lock, flags);
-		if (lost)
-			printk(KERN_WARNING "%s: %d messages suppressed\n",
-				__func__, lost);
-		return 1;
+	if (time_is_before_jiffies(rs->begin + rs->interval)) {
+		if (rs->missed)
+			printk(KERN_WARNING "%s: %d callbacks suppressed\n",
+				__func__, rs->missed);
+		rs->begin = 0;
+		rs->printed = 0;
+		rs->missed = 0;
 	}
-	missed++;
+	if (rs->burst && rs->burst > rs->printed)
+		goto print;
+
+	rs->missed++;
 	spin_unlock_irqrestore(&ratelimit_lock, flags);
 	return 0;
+
+print:
+	rs->printed++;
+	spin_unlock_irqrestore(&ratelimit_lock, flags);
+	return 1;
 }
 EXPORT_SYMBOL(__ratelimit);
diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c
index a570e2af22cb..f686467ff12b 100644
--- a/net/core/sysctl_net_core.c
+++ b/net/core/sysctl_net_core.c
@@ -67,7 +67,7 @@ static struct ctl_table net_core_table[] = {
 	{
 		.ctl_name	= NET_CORE_MSG_COST,
 		.procname	= "message_cost",
-		.data		= &net_msg_cost,
+		.data		= &net_ratelimit_state.interval,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec_jiffies,
@@ -76,7 +76,7 @@ static struct ctl_table net_core_table[] = {
 	{
 		.ctl_name	= NET_CORE_MSG_BURST,
 		.procname	= "message_burst",
-		.data		= &net_msg_burst,
+		.data		= &net_ratelimit_state.burst,
 		.maxlen		= sizeof(int),
 		.mode		= 0644,
 		.proc_handler	= &proc_dointvec,
diff --git a/net/core/utils.c b/net/core/utils.c
index 8031eb59054e..72e0ebe964a0 100644
--- a/net/core/utils.c
+++ b/net/core/utils.c
@@ -31,17 +31,16 @@
 #include <asm/system.h>
 #include <asm/uaccess.h>
 
-int net_msg_cost __read_mostly = 5*HZ;
-int net_msg_burst __read_mostly = 10;
 int net_msg_warn __read_mostly = 1;
 EXPORT_SYMBOL(net_msg_warn);
 
+DEFINE_RATELIMIT_STATE(net_ratelimit_state, 5 * HZ, 10);
 /*
  * All net warning printk()s should be guarded by this function.
  */
 int net_ratelimit(void)
 {
-	return __printk_ratelimit(net_msg_cost, net_msg_burst);
+	return __ratelimit(&net_ratelimit_state);
 }
 EXPORT_SYMBOL(net_ratelimit);
 

From 472dba7d117844c746be97db6be26c2810d79b62 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Fri, 25 Jul 2008 01:45:58 -0700
Subject: [PATCH 0715/1171] sm501: add power control callback

Add callback to get or set the power control if the device has the sleep
connected to some form of GPIO.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Cc: Arnaud Patard <apatard@mandriva.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/mfd/sm501.c   | 31 +++++++++++++++++++++++++++++++
 include/linux/sm501.h |  7 +++++++
 2 files changed, 38 insertions(+)

diff --git a/drivers/mfd/sm501.c b/drivers/mfd/sm501.c
index e2530df4d85c..9296b2673b52 100644
--- a/drivers/mfd/sm501.c
+++ b/drivers/mfd/sm501.c
@@ -1138,8 +1138,31 @@ static int sm501_plat_probe(struct platform_device *dev)
 }
 
 #ifdef CONFIG_PM
+
 /* power management support */
 
+static void sm501_set_power(struct sm501_devdata *sm, int on)
+{
+	struct sm501_platdata *pd = sm->platdata;
+
+	if (pd == NULL)
+		return;
+
+	if (pd->get_power) {
+		if (pd->get_power(sm->dev) == on) {
+			dev_dbg(sm->dev, "is already %d\n", on);
+			return;
+		}
+	}
+
+	if (pd->set_power) {
+		dev_dbg(sm->dev, "setting power to %d\n", on);
+
+		pd->set_power(sm->dev, on);
+		sm501_mdelay(sm, 10);
+	}
+}
+
 static int sm501_plat_suspend(struct platform_device *pdev, pm_message_t state)
 {
 	struct sm501_devdata *sm = platform_get_drvdata(pdev);
@@ -1148,6 +1171,12 @@ static int sm501_plat_suspend(struct platform_device *pdev, pm_message_t state)
 	sm->pm_misc = readl(sm->regs + SM501_MISC_CONTROL);
 
 	sm501_dump_regs(sm);
+
+	if (sm->platdata) {
+		if (sm->platdata->flags & SM501_FLAG_SUSPEND_OFF)
+			sm501_set_power(sm, 0);
+	}
+
 	return 0;
 }
 
@@ -1155,6 +1184,8 @@ static int sm501_plat_resume(struct platform_device *pdev)
 {
 	struct sm501_devdata *sm = platform_get_drvdata(pdev);
 
+	sm501_set_power(sm, 1);
+
 	sm501_dump_regs(sm);
 	sm501_dump_gate(sm);
 	sm501_dump_clk(sm);
diff --git a/include/linux/sm501.h b/include/linux/sm501.h
index b530fa6a1d34..145405bf9efa 100644
--- a/include/linux/sm501.h
+++ b/include/linux/sm501.h
@@ -157,6 +157,8 @@ struct sm501_init_gpio {
 	struct sm501_reg_init	gpio_ddr_high;
 };
 
+#define SM501_FLAG_SUSPEND_OFF		(1<<4)
+
 /* sm501_platdata
  *
  * This is passed with the platform device to allow the board
@@ -170,6 +172,11 @@ struct sm501_platdata {
 	struct sm501_init_gpio		*init_gpiop;
 	struct sm501_platdata_fb	*fb;
 
+	int				 flags;
+
+	int	(*get_power)(struct device *dev);
+	int	(*set_power)(struct device *dev, unsigned int on);
+
 	struct sm501_platdata_gpio_i2c	*gpio_i2c;
 	unsigned int			 gpio_i2c_nr;
 };

From f61be273d3699d174bc1438e6804f9f9e52bb932 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Fri, 25 Jul 2008 01:45:59 -0700
Subject: [PATCH 0716/1171] sm501: add gpiolib support

Add support for exporting the GPIOs on the SM501 via gpiolib.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Cc: Arnaud Patard <apatard@mandriva.com>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/mfd/Kconfig   |   8 ++
 drivers/mfd/sm501.c   | 299 ++++++++++++++++++++++++++++++++++--------
 include/linux/sm501.h |  20 +--
 3 files changed, 257 insertions(+), 70 deletions(-)

diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig
index 9f93c29fed35..bac9e973ece0 100644
--- a/drivers/mfd/Kconfig
+++ b/drivers/mfd/Kconfig
@@ -19,6 +19,14 @@ config MFD_SM501
 	  interface. The device may be connected by PCI or local bus with
 	  varying functions enabled.
 
+config MFD_SM501_GPIO
+	bool "Export GPIO via GPIO layer"
+	depends on MFD_SM501 && HAVE_GPIO_LIB
+	 ---help---
+	 This option uses the gpio library layer to export the 64 GPIO
+	 lines on the SM501. The platform data is used to supply the
+	 base number for the first GPIO line to register.
+
 config MFD_ASIC3
 	bool "Support for Compaq ASIC3"
 	depends on GENERIC_HARDIRQS && HAVE_GPIO_LIB && ARM
diff --git a/drivers/mfd/sm501.c b/drivers/mfd/sm501.c
index 9296b2673b52..be8713908125 100644
--- a/drivers/mfd/sm501.c
+++ b/drivers/mfd/sm501.c
@@ -19,6 +19,7 @@
 #include <linux/device.h>
 #include <linux/platform_device.h>
 #include <linux/pci.h>
+#include <linux/gpio.h>
 
 #include <linux/sm501.h>
 #include <linux/sm501-regs.h>
@@ -31,10 +32,29 @@ struct sm501_device {
 	struct platform_device		pdev;
 };
 
+struct sm501_gpio;
+
+struct sm501_gpio_chip {
+	struct gpio_chip	gpio;
+	struct sm501_gpio	*ourgpio;	/* to get back to parent. */
+	void __iomem		*regbase;
+};
+
+struct sm501_gpio {
+	struct sm501_gpio_chip	low;
+	struct sm501_gpio_chip	high;
+	spinlock_t		lock;
+
+	unsigned int		 registered : 1;
+	void __iomem		*regs;
+	struct resource		*regs_res;
+};
+
 struct sm501_devdata {
 	spinlock_t			 reg_lock;
 	struct mutex			 clock_lock;
 	struct list_head		 devices;
+	struct sm501_gpio		 gpio;
 
 	struct device			*dev;
 	struct resource			*io_res;
@@ -42,6 +62,7 @@ struct sm501_devdata {
 	struct resource			*regs_claim;
 	struct sm501_platdata		*platdata;
 
+
 	unsigned int			 in_suspend;
 	unsigned long			 pm_misc;
 
@@ -52,6 +73,7 @@ struct sm501_devdata {
 	unsigned int			 rev;
 };
 
+
 #define MHZ (1000 * 1000)
 
 #ifdef DEBUG
@@ -276,58 +298,6 @@ unsigned long sm501_modify_reg(struct device *dev,
 
 EXPORT_SYMBOL_GPL(sm501_modify_reg);
 
-unsigned long sm501_gpio_get(struct device *dev,
-			     unsigned long gpio)
-{
-	struct sm501_devdata *sm = dev_get_drvdata(dev);
-	unsigned long result;
-	unsigned long reg;
-
-	reg = (gpio > 32) ? SM501_GPIO_DATA_HIGH : SM501_GPIO_DATA_LOW;
-	result = readl(sm->regs + reg);
-
-	result >>= (gpio & 31);
-	return result & 1UL;
-}
-
-EXPORT_SYMBOL_GPL(sm501_gpio_get);
-
-void sm501_gpio_set(struct device *dev,
-		    unsigned long gpio,
-		    unsigned int to,
-		    unsigned int dir)
-{
-	struct sm501_devdata *sm = dev_get_drvdata(dev);
-
-	unsigned long bit = 1 << (gpio & 31);
-	unsigned long base;
-	unsigned long save;
-	unsigned long val;
-
-	base = (gpio > 32) ? SM501_GPIO_DATA_HIGH : SM501_GPIO_DATA_LOW;
-	base += SM501_GPIO;
-
-	spin_lock_irqsave(&sm->reg_lock, save);
-
-	val = readl(sm->regs + base) & ~bit;
-	if (to)
-		val |= bit;
-	writel(val, sm->regs + base);
-
-	val = readl(sm->regs + SM501_GPIO_DDR_LOW) & ~bit;
-	if (dir)
-		val |= bit;
-
-	writel(val, sm->regs + SM501_GPIO_DDR_LOW);
-	sm501_sync_regs(sm);
-
-	spin_unlock_irqrestore(&sm->reg_lock, save);
-
-}
-
-EXPORT_SYMBOL_GPL(sm501_gpio_set);
-
-
 /* sm501_unit_power
  *
  * alters the power active gate to set specific units on or off
@@ -906,6 +876,226 @@ static int sm501_register_display(struct sm501_devdata *sm,
 	return sm501_register_device(sm, pdev);
 }
 
+#ifdef CONFIG_MFD_SM501_GPIO
+
+static inline struct sm501_gpio_chip *to_sm501_gpio(struct gpio_chip *gc)
+{
+	return container_of(gc, struct sm501_gpio_chip, gpio);
+}
+
+static inline struct sm501_devdata *sm501_gpio_to_dev(struct sm501_gpio *gpio)
+{
+	return container_of(gpio, struct sm501_devdata, gpio);
+}
+
+static int sm501_gpio_get(struct gpio_chip *chip, unsigned offset)
+
+{
+	struct sm501_gpio_chip *smgpio = to_sm501_gpio(chip);
+	unsigned long result;
+
+	result = readl(smgpio->regbase + SM501_GPIO_DATA_LOW);
+	result >>= offset;
+
+	return result & 1UL;
+}
+
+static void sm501_gpio_set(struct gpio_chip *chip, unsigned offset, int value)
+
+{
+	struct sm501_gpio_chip *smchip = to_sm501_gpio(chip);
+	struct sm501_gpio *smgpio = smchip->ourgpio;
+	unsigned long bit = 1 << offset;
+	void __iomem *regs = smchip->regbase;
+	unsigned long save;
+	unsigned long val;
+
+	dev_dbg(sm501_gpio_to_dev(smgpio)->dev, "%s(%p,%d)\n",
+		__func__, chip, offset);
+
+	spin_lock_irqsave(&smgpio->lock, save);
+
+	val = readl(regs + SM501_GPIO_DATA_LOW) & ~bit;
+	if (value)
+		val |= bit;
+	writel(val, regs);
+
+	sm501_sync_regs(sm501_gpio_to_dev(smgpio));
+	spin_unlock_irqrestore(&smgpio->lock, save);
+}
+
+static int sm501_gpio_input(struct gpio_chip *chip, unsigned offset)
+{
+	struct sm501_gpio_chip *smchip = to_sm501_gpio(chip);
+	struct sm501_gpio *smgpio = smchip->ourgpio;
+	void __iomem *regs = smchip->regbase;
+	unsigned long bit = 1 << offset;
+	unsigned long save;
+	unsigned long ddr;
+
+	dev_info(sm501_gpio_to_dev(smgpio)->dev, "%s(%p,%d)\n",
+		 __func__, chip, offset);
+
+	spin_lock_irqsave(&smgpio->lock, save);
+
+	ddr = readl(regs + SM501_GPIO_DDR_LOW);
+	writel(ddr & ~bit, regs + SM501_GPIO_DDR_LOW);
+
+	sm501_sync_regs(sm501_gpio_to_dev(smgpio));
+	spin_unlock_irqrestore(&smgpio->lock, save);
+
+	return 0;
+}
+
+static int sm501_gpio_output(struct gpio_chip *chip,
+			     unsigned offset, int value)
+{
+	struct sm501_gpio_chip *smchip = to_sm501_gpio(chip);
+	struct sm501_gpio *smgpio = smchip->ourgpio;
+	unsigned long bit = 1 << offset;
+	void __iomem *regs = smchip->regbase;
+	unsigned long save;
+	unsigned long val;
+	unsigned long ddr;
+
+	dev_dbg(sm501_gpio_to_dev(smgpio)->dev, "%s(%p,%d,%d)\n",
+		__func__, chip, offset, value);
+
+	spin_lock_irqsave(&smgpio->lock, save);
+
+	val = readl(regs + SM501_GPIO_DATA_LOW);
+	if (value)
+		val |= bit;
+	else
+		val &= ~bit;
+	writel(val, regs);
+
+	ddr = readl(regs + SM501_GPIO_DDR_LOW);
+	writel(ddr | bit, regs + SM501_GPIO_DDR_LOW);
+
+	sm501_sync_regs(sm501_gpio_to_dev(smgpio));
+	writel(val, regs + SM501_GPIO_DATA_LOW);
+
+	sm501_sync_regs(sm501_gpio_to_dev(smgpio));
+	spin_unlock_irqrestore(&smgpio->lock, save);
+
+	return 0;
+}
+
+static struct gpio_chip gpio_chip_template = {
+	.ngpio			= 32,
+	.direction_input	= sm501_gpio_input,
+	.direction_output	= sm501_gpio_output,
+	.set			= sm501_gpio_set,
+	.get			= sm501_gpio_get,
+};
+
+static int __devinit sm501_gpio_register_chip(struct sm501_devdata *sm,
+					      struct sm501_gpio *gpio,
+					      struct sm501_gpio_chip *chip)
+{
+	struct sm501_platdata *pdata = sm->platdata;
+	struct gpio_chip *gchip = &chip->gpio;
+	unsigned base = pdata->gpio_base;
+
+	memcpy(chip, &gpio_chip_template, sizeof(struct gpio_chip));
+
+	if (chip == &gpio->high) {
+		base += 32;
+		chip->regbase = gpio->regs + SM501_GPIO_DATA_HIGH;
+		gchip->label  = "SM501-HIGH";
+	} else {
+		chip->regbase = gpio->regs + SM501_GPIO_DATA_LOW;
+		gchip->label  = "SM501-LOW";
+	}
+
+	gchip->base   = base;
+	chip->ourgpio = gpio;
+
+	return gpiochip_add(gchip);
+}
+
+static int sm501_register_gpio(struct sm501_devdata *sm)
+{
+	struct sm501_gpio *gpio = &sm->gpio;
+	resource_size_t iobase = sm->io_res->start + SM501_GPIO;
+	int ret;
+	int tmp;
+
+	dev_dbg(sm->dev, "registering gpio block %08llx\n",
+		(unsigned long long)iobase);
+
+	spin_lock_init(&gpio->lock);
+
+	gpio->regs_res = request_mem_region(iobase, 0x20, "sm501-gpio");
+	if (gpio->regs_res == NULL) {
+		dev_err(sm->dev, "gpio: failed to request region\n");
+		return -ENXIO;
+	}
+
+	gpio->regs = ioremap(iobase, 0x20);
+	if (gpio->regs == NULL) {
+		dev_err(sm->dev, "gpio: failed to remap registers\n");
+		ret = -ENXIO;
+		goto err_mapped;
+	}
+
+	/* Register both our chips. */
+
+	ret = sm501_gpio_register_chip(sm, gpio, &gpio->low);
+	if (ret) {
+		dev_err(sm->dev, "failed to add low chip\n");
+		goto err_mapped;
+	}
+
+	ret = sm501_gpio_register_chip(sm, gpio, &gpio->high);
+	if (ret) {
+		dev_err(sm->dev, "failed to add high chip\n");
+		goto err_low_chip;
+	}
+
+	gpio->registered = 1;
+
+	return 0;
+
+ err_low_chip:
+	tmp = gpiochip_remove(&gpio->low.gpio);
+	if (tmp) {
+		dev_err(sm->dev, "cannot remove low chip, cannot tidy up\n");
+		return ret;
+	}
+
+ err_mapped:
+	release_resource(gpio->regs_res);
+	kfree(gpio->regs_res);
+
+	return ret;
+}
+
+static void sm501_gpio_remove(struct sm501_devdata *sm)
+{
+	int ret;
+
+	ret = gpiochip_remove(&sm->gpio.low.gpio);
+	if (ret)
+		dev_err(sm->dev, "cannot remove low chip, cannot tidy up\n");
+
+	ret = gpiochip_remove(&sm->gpio.high.gpio);
+	if (ret)
+		dev_err(sm->dev, "cannot remove high chip, cannot tidy up\n");
+}
+
+#else
+static int sm501_register_gpio(struct sm501_devdata *sm)
+{
+	return 0;
+}
+
+static void sm501_gpio_remove(struct sm501_devdata *sm)
+{
+}
+#endif
+
 /* sm501_dbg_regs
  *
  * Debug attribute to attach to parent device to show core registers
@@ -1059,6 +1249,8 @@ static int sm501_init_dev(struct sm501_devdata *sm)
 			sm501_register_usbhost(sm, &mem_avail);
 		if (idata->devices & (SM501_USE_UART0 | SM501_USE_UART1))
 			sm501_register_uart(sm, idata->devices);
+		if (idata->devices & SM501_USE_GPIO)
+			sm501_register_gpio(sm);
 	}
 
 	ret = sm501_check_clocks(sm);
@@ -1366,6 +1558,9 @@ static void sm501_dev_remove(struct sm501_devdata *sm)
 		sm501_remove_sub(sm, smdev);
 
 	device_remove_file(sm->dev, &dev_attr_dbg_regs);
+
+	if (sm->gpio.registered)
+		sm501_gpio_remove(sm);
 }
 
 static void sm501_pci_remove(struct pci_dev *dev)
diff --git a/include/linux/sm501.h b/include/linux/sm501.h
index 145405bf9efa..6ea39007c8a3 100644
--- a/include/linux/sm501.h
+++ b/include/linux/sm501.h
@@ -46,24 +46,6 @@ extern unsigned long sm501_modify_reg(struct device *dev,
 				      unsigned long set,
 				      unsigned long clear);
 
-/* sm501_gpio_set
- *
- * set the state of the given GPIO line
-*/
-
-extern void sm501_gpio_set(struct device *dev,
-			   unsigned long gpio,
-			   unsigned int to,
-			   unsigned int dir);
-
-/* sm501_gpio_get
- *
- * get the state of the given GPIO line
-*/
-
-extern unsigned long sm501_gpio_get(struct device *dev,
-				    unsigned long gpio);
-
 
 /* Platform data definitions */
 
@@ -131,6 +113,7 @@ struct sm501_reg_init {
 #define SM501_USE_FBACCEL	(1<<6)
 #define SM501_USE_AC97		(1<<7)
 #define SM501_USE_I2S		(1<<8)
+#define SM501_USE_GPIO		(1<<9)
 
 #define SM501_USE_ALL		(0xffffffff)
 
@@ -173,6 +156,7 @@ struct sm501_platdata {
 	struct sm501_platdata_fb	*fb;
 
 	int				 flags;
+	unsigned			 gpio_base;
 
 	int	(*get_power)(struct device *dev);
 	int	(*set_power)(struct device *dev, unsigned int on);

From 60e540d617b40eb3d37f1dd99c97af588ff9b70b Mon Sep 17 00:00:00 2001
From: Arnaud Patard <apatard@mandriva.com>
Date: Fri, 25 Jul 2008 01:46:00 -0700
Subject: [PATCH 0717/1171] sm501: gpio dynamic registration for PCI devices

The SM501 PCI card requires a dyanmic gpio allocation as the number of
cards is not known at compile time.  Fixup the platform data and
registration to deal with this.

Acked-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Arnaud Patard <apatard@mandriva.com>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/mfd/sm501.c   | 6 ++++--
 include/linux/sm501.h | 2 +-
 2 files changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/mfd/sm501.c b/drivers/mfd/sm501.c
index be8713908125..c3e5a48f6148 100644
--- a/drivers/mfd/sm501.c
+++ b/drivers/mfd/sm501.c
@@ -996,12 +996,13 @@ static int __devinit sm501_gpio_register_chip(struct sm501_devdata *sm,
 {
 	struct sm501_platdata *pdata = sm->platdata;
 	struct gpio_chip *gchip = &chip->gpio;
-	unsigned base = pdata->gpio_base;
+	int base = pdata->gpio_base;
 
 	memcpy(chip, &gpio_chip_template, sizeof(struct gpio_chip));
 
 	if (chip == &gpio->high) {
-		base += 32;
+		if (base > 0)
+			base += 32;
 		chip->regbase = gpio->regs + SM501_GPIO_DATA_HIGH;
 		gchip->label  = "SM501-HIGH";
 	} else {
@@ -1452,6 +1453,7 @@ static struct sm501_platdata_fb sm501_fb_pdata = {
 static struct sm501_platdata sm501_pci_platdata = {
 	.init		= &sm501_pci_initdata,
 	.fb		= &sm501_fb_pdata,
+	.gpio_base	= -1,
 };
 
 static int sm501_pci_probe(struct pci_dev *dev,
diff --git a/include/linux/sm501.h b/include/linux/sm501.h
index 6ea39007c8a3..a8d02f36ad32 100644
--- a/include/linux/sm501.h
+++ b/include/linux/sm501.h
@@ -156,7 +156,7 @@ struct sm501_platdata {
 	struct sm501_platdata_fb	*fb;
 
 	int				 flags;
-	unsigned			 gpio_base;
+	int				 gpio_base;
 
 	int	(*get_power)(struct device *dev);
 	int	(*set_power)(struct device *dev, unsigned int on);

From 42cd2366fb9b58cdfc1855be32b31a78e40b2079 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Fri, 25 Jul 2008 01:46:01 -0700
Subject: [PATCH 0718/1171] sm501: gpio I2C support

Add support for adding the GPIO based I2C resources.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Cc: Arnaud Patard <apatard@mandriva.com>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/mfd/sm501.c   | 76 ++++++++++++++++++++++++++++++++++++++++++-
 include/linux/sm501.h | 10 +++++-
 2 files changed, 84 insertions(+), 2 deletions(-)

diff --git a/drivers/mfd/sm501.c b/drivers/mfd/sm501.c
index c3e5a48f6148..107215b28805 100644
--- a/drivers/mfd/sm501.c
+++ b/drivers/mfd/sm501.c
@@ -20,6 +20,7 @@
 #include <linux/platform_device.h>
 #include <linux/pci.h>
 #include <linux/gpio.h>
+#include <linux/i2c-gpio.h>
 
 #include <linux/sm501.h>
 #include <linux/sm501-regs.h>
@@ -1086,6 +1087,11 @@ static void sm501_gpio_remove(struct sm501_devdata *sm)
 		dev_err(sm->dev, "cannot remove high chip, cannot tidy up\n");
 }
 
+static int sm501_gpio_pin2nr(struct sm501_devdata *sm, unsigned int pin)
+{
+	struct sm501_gpio *gpio = &sm->gpio;
+	return pin + (pin < 32) ? gpio->low.gpio.base : gpio->high.gpio.base;
+}
 #else
 static int sm501_register_gpio(struct sm501_devdata *sm)
 {
@@ -1095,8 +1101,66 @@ static int sm501_register_gpio(struct sm501_devdata *sm)
 static void sm501_gpio_remove(struct sm501_devdata *sm)
 {
 }
+
+static int sm501_gpio_pin2nr(struct sm501_devdata *sm, unsigned int pin)
+{
+	return -1;
+}
 #endif
 
+static int sm501_register_gpio_i2c_instance(struct sm501_devdata *sm,
+					    struct sm501_platdata_gpio_i2c *iic)
+{
+	struct i2c_gpio_platform_data *icd;
+	struct platform_device *pdev;
+
+	pdev = sm501_create_subdev(sm, "i2c-gpio", 0,
+				   sizeof(struct i2c_gpio_platform_data));
+	if (!pdev)
+		return -ENOMEM;
+
+	icd = pdev->dev.platform_data;
+
+	/* We keep the pin_sda and pin_scl fields relative in case the
+	 * same platform data is passed to >1 SM501.
+	 */
+
+	icd->sda_pin = sm501_gpio_pin2nr(sm, iic->pin_sda);
+	icd->scl_pin = sm501_gpio_pin2nr(sm, iic->pin_scl);
+	icd->timeout = iic->timeout;
+	icd->udelay = iic->udelay;
+
+	/* note, we can't use either of the pin numbers, as the i2c-gpio
+	 * driver uses the platform.id field to generate the bus number
+	 * to register with the i2c core; The i2c core doesn't have enough
+	 * entries to deal with anything we currently use.
+	*/
+
+	pdev->id = iic->bus_num;
+
+	dev_info(sm->dev, "registering i2c-%d: sda=%d (%d), scl=%d (%d)\n",
+		 iic->bus_num,
+		 icd->sda_pin, iic->pin_sda, icd->scl_pin, iic->pin_scl);
+
+	return sm501_register_device(sm, pdev);
+}
+
+static int sm501_register_gpio_i2c(struct sm501_devdata *sm,
+				   struct sm501_platdata *pdata)
+{
+	struct sm501_platdata_gpio_i2c *iic = pdata->gpio_i2c;
+	int index;
+	int ret;
+
+	for (index = 0; index < pdata->gpio_i2c_nr; index++, iic++) {
+		ret = sm501_register_gpio_i2c_instance(sm, iic);
+		if (ret < 0)
+			return ret;
+	}
+
+	return 0;
+}
+
 /* sm501_dbg_regs
  *
  * Debug attribute to attach to parent device to show core registers
@@ -1204,6 +1268,7 @@ static unsigned int sm501_mem_local[] = {
 static int sm501_init_dev(struct sm501_devdata *sm)
 {
 	struct sm501_initdata *idata;
+	struct sm501_platdata *pdata;
 	resource_size_t mem_avail;
 	unsigned long dramctrl;
 	unsigned long devid;
@@ -1242,7 +1307,9 @@ static int sm501_init_dev(struct sm501_devdata *sm)
 
 	/* check to see if we have some device initialisation */
 
-	idata = sm->platdata ? sm->platdata->init : NULL;
+	pdata = sm->platdata;
+	idata = pdata ? pdata->init : NULL;
+
 	if (idata) {
 		sm501_init_regs(sm, idata);
 
@@ -1254,6 +1321,13 @@ static int sm501_init_dev(struct sm501_devdata *sm)
 			sm501_register_gpio(sm);
 	}
 
+	if (pdata->gpio_i2c != NULL && pdata->gpio_i2c_nr > 0) {
+		if (!sm->gpio.registered)
+			dev_err(sm->dev, "no gpio registered for i2c gpio.\n");
+		else
+			sm501_register_gpio_i2c(sm, pdata);
+	}
+
 	ret = sm501_check_clocks(sm);
 	if (ret) {
 		dev_err(sm->dev, "M1X and M clocks sourced from different "
diff --git a/include/linux/sm501.h b/include/linux/sm501.h
index a8d02f36ad32..214f93209b8c 100644
--- a/include/linux/sm501.h
+++ b/include/linux/sm501.h
@@ -86,11 +86,19 @@ struct sm501_platdata_fb {
 	struct sm501_platdata_fbsub	*fb_pnl;
 };
 
-/* gpio i2c */
+/* gpio i2c
+ *
+ * Note, we have to pass in the bus number, as the number used will be
+ * passed to the i2c-gpio driver's platform_device.id, subsequently used
+ * to register the i2c bus.
+*/
 
 struct sm501_platdata_gpio_i2c {
+	unsigned int		bus_num;
 	unsigned int		pin_sda;
 	unsigned int		pin_scl;
+	int			udelay;
+	int			timeout;
 };
 
 /* sm501_initdata

From 28130bea3bcfefe3437b0a5dcab786f1f0296953 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Fri, 25 Jul 2008 01:46:02 -0700
Subject: [PATCH 0719/1171] sm501: fixes for akpms comments on gpiolib addition

Fixup the comments from the patch that added the gpiolib support from
Andrew Morton.  These include spotting some missing frees on error or
release, and changing a memcpy for a type-safe assingment.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/mfd/sm501.c | 24 ++++++++++++++++--------
 1 file changed, 16 insertions(+), 8 deletions(-)

diff --git a/drivers/mfd/sm501.c b/drivers/mfd/sm501.c
index 107215b28805..2dfb41aabcaf 100644
--- a/drivers/mfd/sm501.c
+++ b/drivers/mfd/sm501.c
@@ -999,7 +999,7 @@ static int __devinit sm501_gpio_register_chip(struct sm501_devdata *sm,
 	struct gpio_chip *gchip = &chip->gpio;
 	int base = pdata->gpio_base;
 
-	memcpy(chip, &gpio_chip_template, sizeof(struct gpio_chip));
+	chip->gpio = gpio_chip_template;
 
 	if (chip == &gpio->high) {
 		if (base > 0)
@@ -1039,7 +1039,7 @@ static int sm501_register_gpio(struct sm501_devdata *sm)
 	if (gpio->regs == NULL) {
 		dev_err(sm->dev, "gpio: failed to remap registers\n");
 		ret = -ENXIO;
-		goto err_mapped;
+		goto err_claimed;
 	}
 
 	/* Register both our chips. */
@@ -1068,6 +1068,9 @@ static int sm501_register_gpio(struct sm501_devdata *sm)
 	}
 
  err_mapped:
+	iounmap(gpio->regs);
+
+ err_claimed:
 	release_resource(gpio->regs_res);
 	kfree(gpio->regs_res);
 
@@ -1076,33 +1079,38 @@ static int sm501_register_gpio(struct sm501_devdata *sm)
 
 static void sm501_gpio_remove(struct sm501_devdata *sm)
 {
+	struct sm501_gpio *gpio = &sm->gpio;
 	int ret;
 
-	ret = gpiochip_remove(&sm->gpio.low.gpio);
+	ret = gpiochip_remove(&gpio->low.gpio);
 	if (ret)
 		dev_err(sm->dev, "cannot remove low chip, cannot tidy up\n");
 
-	ret = gpiochip_remove(&sm->gpio.high.gpio);
+	ret = gpiochip_remove(&gpio->high.gpio);
 	if (ret)
 		dev_err(sm->dev, "cannot remove high chip, cannot tidy up\n");
+
+	iounmap(gpio->regs);
+	release_resource(gpio->regs_res);
+	kfree(gpio->regs_res);
 }
 
-static int sm501_gpio_pin2nr(struct sm501_devdata *sm, unsigned int pin)
+static inline int sm501_gpio_pin2nr(struct sm501_devdata *sm, unsigned int pin)
 {
 	struct sm501_gpio *gpio = &sm->gpio;
 	return pin + (pin < 32) ? gpio->low.gpio.base : gpio->high.gpio.base;
 }
 #else
-static int sm501_register_gpio(struct sm501_devdata *sm)
+static inline int sm501_register_gpio(struct sm501_devdata *sm)
 {
 	return 0;
 }
 
-static void sm501_gpio_remove(struct sm501_devdata *sm)
+static inline void sm501_gpio_remove(struct sm501_devdata *sm)
 {
 }
 
-static int sm501_gpio_pin2nr(struct sm501_devdata *sm, unsigned int pin)
+static inline int sm501_gpio_pin2nr(struct sm501_devdata *sm, unsigned int pin)
 {
 	return -1;
 }

From f2999209d779573e17468b680f5f267d8cb2a9c7 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Fri, 25 Jul 2008 01:46:02 -0700
Subject: [PATCH 0720/1171] mfd: sm501 build fixes when CONFIG_MFD_SM501_GPIO
 unset

Fix the build problems if CONFIG_MFD_SM501_GPIO is not set, which is
generally when there is no gpiolib support available as currently happens
on x86 when building PCI SM501.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Tested-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/mfd/sm501.c | 29 ++++++++++++++++++++++++-----
 1 file changed, 24 insertions(+), 5 deletions(-)

diff --git a/drivers/mfd/sm501.c b/drivers/mfd/sm501.c
index 2dfb41aabcaf..79d7aea5510b 100644
--- a/drivers/mfd/sm501.c
+++ b/drivers/mfd/sm501.c
@@ -19,7 +19,6 @@
 #include <linux/device.h>
 #include <linux/platform_device.h>
 #include <linux/pci.h>
-#include <linux/gpio.h>
 #include <linux/i2c-gpio.h>
 
 #include <linux/sm501.h>
@@ -35,6 +34,9 @@ struct sm501_device {
 
 struct sm501_gpio;
 
+#ifdef CONFIG_MFD_SM501_GPIO
+#include <linux/gpio.h>
+
 struct sm501_gpio_chip {
 	struct gpio_chip	gpio;
 	struct sm501_gpio	*ourgpio;	/* to get back to parent. */
@@ -50,6 +52,11 @@ struct sm501_gpio {
 	void __iomem		*regs;
 	struct resource		*regs_res;
 };
+#else
+struct sm501_gpio {
+	/* no gpio support, empty definition for sm501_devdata. */
+};
+#endif
 
 struct sm501_devdata {
 	spinlock_t			 reg_lock;
@@ -1082,6 +1089,9 @@ static void sm501_gpio_remove(struct sm501_devdata *sm)
 	struct sm501_gpio *gpio = &sm->gpio;
 	int ret;
 
+	if (!sm->gpio.registered)
+		return;
+
 	ret = gpiochip_remove(&gpio->low.gpio);
 	if (ret)
 		dev_err(sm->dev, "cannot remove low chip, cannot tidy up\n");
@@ -1100,6 +1110,11 @@ static inline int sm501_gpio_pin2nr(struct sm501_devdata *sm, unsigned int pin)
 	struct sm501_gpio *gpio = &sm->gpio;
 	return pin + (pin < 32) ? gpio->low.gpio.base : gpio->high.gpio.base;
 }
+
+static inline int sm501_gpio_isregistered(struct sm501_devdata *sm)
+{
+	return sm->gpio.registered;
+}
 #else
 static inline int sm501_register_gpio(struct sm501_devdata *sm)
 {
@@ -1114,6 +1129,11 @@ static inline int sm501_gpio_pin2nr(struct sm501_devdata *sm, unsigned int pin)
 {
 	return -1;
 }
+
+static inline int sm501_gpio_isregistered(struct sm501_devdata *sm)
+{
+	return 0;
+}
 #endif
 
 static int sm501_register_gpio_i2c_instance(struct sm501_devdata *sm,
@@ -1330,8 +1350,8 @@ static int sm501_init_dev(struct sm501_devdata *sm)
 	}
 
 	if (pdata->gpio_i2c != NULL && pdata->gpio_i2c_nr > 0) {
-		if (!sm->gpio.registered)
-			dev_err(sm->dev, "no gpio registered for i2c gpio.\n");
+		if (!sm501_gpio_isregistered(sm))
+			dev_err(sm->dev, "no gpio available for i2c gpio.\n");
 		else
 			sm501_register_gpio_i2c(sm, pdata);
 	}
@@ -1643,8 +1663,7 @@ static void sm501_dev_remove(struct sm501_devdata *sm)
 
 	device_remove_file(sm->dev, &dev_attr_dbg_regs);
 
-	if (sm->gpio.registered)
-		sm501_gpio_remove(sm);
+	sm501_gpio_remove(sm);
 }
 
 static void sm501_pci_remove(struct pci_dev *dev)

From 53a9600c634e3bfd6230e0597aca159bf4d4d010 Mon Sep 17 00:00:00 2001
From: Ben Dooks <ben-linux@fluff.org>
Date: Fri, 25 Jul 2008 01:46:03 -0700
Subject: [PATCH 0721/1171] mfd: sm501 fix gpio number calculation for upper
 bank

The sm501_gpio_pin2nr() routine returns the wrong values for gpios in the
upper bank.

Signed-off-by: Ben Dooks <ben-linux@fluff.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/mfd/sm501.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/mfd/sm501.c b/drivers/mfd/sm501.c
index 79d7aea5510b..7aebad4c06ff 100644
--- a/drivers/mfd/sm501.c
+++ b/drivers/mfd/sm501.c
@@ -1108,7 +1108,9 @@ static void sm501_gpio_remove(struct sm501_devdata *sm)
 static inline int sm501_gpio_pin2nr(struct sm501_devdata *sm, unsigned int pin)
 {
 	struct sm501_gpio *gpio = &sm->gpio;
-	return pin + (pin < 32) ? gpio->low.gpio.base : gpio->high.gpio.base;
+	int base = (pin < 32) ? gpio->low.gpio.base : gpio->high.gpio.base;
+
+	return (pin % 32) + base;
 }
 
 static inline int sm501_gpio_isregistered(struct sm501_devdata *sm)

From ef53d9c5e4da147ecaa43c44c5e5945eb83970a2 Mon Sep 17 00:00:00 2001
From: Srinivasa D S <srinivasa@in.ibm.com>
Date: Fri, 25 Jul 2008 01:46:04 -0700
Subject: [PATCH 0722/1171] kprobes: improve kretprobe scalability with hashed
 locking

Currently list of kretprobe instances are stored in kretprobe object (as
used_instances,free_instances) and in kretprobe hash table.  We have one
global kretprobe lock to serialise the access to these lists.  This causes
only one kretprobe handler to execute at a time.  Hence affects system
performance, particularly on SMP systems and when return probe is set on
lot of functions (like on all systemcalls).

Solution proposed here gives fine-grain locks that performs better on SMP
system compared to present kretprobe implementation.

Solution:

 1) Instead of having one global lock to protect kretprobe instances
    present in kretprobe object and kretprobe hash table.  We will have
    two locks, one lock for protecting kretprobe hash table and another
    lock for kretporbe object.

 2) We hold lock present in kretprobe object while we modify kretprobe
    instance in kretprobe object and we hold per-hash-list lock while
    modifying kretprobe instances present in that hash list.  To prevent
    deadlock, we never grab a per-hash-list lock while holding a kretprobe
    lock.

 3) We can remove used_instances from struct kretprobe, as we can
    track used instances of kretprobe instances using kretprobe hash
    table.

Time duration for kernel compilation ("make -j 8") on a 8-way ppc64 system
with return probes set on all systemcalls looks like this.

cacheline              non-cacheline             Un-patched kernel
aligned patch 	       aligned patch
===============================================================================
real    9m46.784s       9m54.412s                  10m2.450s
user    40m5.715s       40m7.142s                  40m4.273s
sys     2m57.754s       2m58.583s                  3m17.430s
===========================================================

Time duration for kernel compilation ("make -j 8) on the same system, when
kernel is not probed.
=========================
real    9m26.389s
user    40m8.775s
sys     2m7.283s
=========================

Signed-off-by: Srinivasa DS <srinivasa@in.ibm.com>
Signed-off-by: Jim Keniston <jkenisto@us.ibm.com>
Acked-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Masami Hiramatsu <mhiramat@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/arm/kernel/kprobes.c     |   6 +-
 arch/ia64/kernel/kprobes.c    |   6 +-
 arch/powerpc/kernel/kprobes.c |   6 +-
 arch/s390/kernel/kprobes.c    |   6 +-
 arch/sparc64/kernel/kprobes.c |  11 ++-
 arch/x86/kernel/kprobes.c     |   6 +-
 include/linux/kprobes.h       |   7 +-
 kernel/kprobes.c              | 127 ++++++++++++++++++++++++----------
 8 files changed, 108 insertions(+), 67 deletions(-)

diff --git a/arch/arm/kernel/kprobes.c b/arch/arm/kernel/kprobes.c
index 5ee39e10c8d1..d28513f14d05 100644
--- a/arch/arm/kernel/kprobes.c
+++ b/arch/arm/kernel/kprobes.c
@@ -296,8 +296,7 @@ static __used __kprobes void *trampoline_handler(struct pt_regs *regs)
 	unsigned long trampoline_address = (unsigned long)&kretprobe_trampoline;
 
 	INIT_HLIST_HEAD(&empty_rp);
-	spin_lock_irqsave(&kretprobe_lock, flags);
-	head = kretprobe_inst_table_head(current);
+	kretprobe_hash_lock(current, &head, &flags);
 
 	/*
 	 * It is possible to have multiple instances associated with a given
@@ -337,7 +336,7 @@ static __used __kprobes void *trampoline_handler(struct pt_regs *regs)
 	}
 
 	kretprobe_assert(ri, orig_ret_address, trampoline_address);
-	spin_unlock_irqrestore(&kretprobe_lock, flags);
+	kretprobe_hash_unlock(current, &flags);
 
 	hlist_for_each_entry_safe(ri, node, tmp, &empty_rp, hlist) {
 		hlist_del(&ri->hlist);
@@ -347,7 +346,6 @@ static __used __kprobes void *trampoline_handler(struct pt_regs *regs)
 	return (void *)orig_ret_address;
 }
 
-/* Called with kretprobe_lock held. */
 void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri,
 				      struct pt_regs *regs)
 {
diff --git a/arch/ia64/kernel/kprobes.c b/arch/ia64/kernel/kprobes.c
index 233434f4f88f..f07688da947c 100644
--- a/arch/ia64/kernel/kprobes.c
+++ b/arch/ia64/kernel/kprobes.c
@@ -429,8 +429,7 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
 		((struct fnptr *)kretprobe_trampoline)->ip;
 
 	INIT_HLIST_HEAD(&empty_rp);
-	spin_lock_irqsave(&kretprobe_lock, flags);
-	head = kretprobe_inst_table_head(current);
+	kretprobe_hash_lock(current, &head, &flags);
 
 	/*
 	 * It is possible to have multiple instances associated with a given
@@ -485,7 +484,7 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
 	kretprobe_assert(ri, orig_ret_address, trampoline_address);
 
 	reset_current_kprobe();
-	spin_unlock_irqrestore(&kretprobe_lock, flags);
+	kretprobe_hash_unlock(current, &flags);
 	preempt_enable_no_resched();
 
 	hlist_for_each_entry_safe(ri, node, tmp, &empty_rp, hlist) {
@@ -500,7 +499,6 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
 	return 1;
 }
 
-/* Called with kretprobe_lock held */
 void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri,
 				      struct pt_regs *regs)
 {
diff --git a/arch/powerpc/kernel/kprobes.c b/arch/powerpc/kernel/kprobes.c
index 4ba2af125450..de79915452c8 100644
--- a/arch/powerpc/kernel/kprobes.c
+++ b/arch/powerpc/kernel/kprobes.c
@@ -144,7 +144,6 @@ static void __kprobes set_current_kprobe(struct kprobe *p, struct pt_regs *regs,
 	kcb->kprobe_saved_msr = regs->msr;
 }
 
-/* Called with kretprobe_lock held */
 void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri,
 				      struct pt_regs *regs)
 {
@@ -312,8 +311,7 @@ static int __kprobes trampoline_probe_handler(struct kprobe *p,
 	unsigned long trampoline_address =(unsigned long)&kretprobe_trampoline;
 
 	INIT_HLIST_HEAD(&empty_rp);
-	spin_lock_irqsave(&kretprobe_lock, flags);
-	head = kretprobe_inst_table_head(current);
+	kretprobe_hash_lock(current, &head, &flags);
 
 	/*
 	 * It is possible to have multiple instances associated with a given
@@ -352,7 +350,7 @@ static int __kprobes trampoline_probe_handler(struct kprobe *p,
 	regs->nip = orig_ret_address;
 
 	reset_current_kprobe();
-	spin_unlock_irqrestore(&kretprobe_lock, flags);
+	kretprobe_hash_unlock(current, &flags);
 	preempt_enable_no_resched();
 
 	hlist_for_each_entry_safe(ri, node, tmp, &empty_rp, hlist) {
diff --git a/arch/s390/kernel/kprobes.c b/arch/s390/kernel/kprobes.c
index 288ad490a6dd..4f82e5b5f879 100644
--- a/arch/s390/kernel/kprobes.c
+++ b/arch/s390/kernel/kprobes.c
@@ -270,7 +270,6 @@ static void __kprobes set_current_kprobe(struct kprobe *p, struct pt_regs *regs,
 	__ctl_store(kcb->kprobe_saved_ctl, 9, 11);
 }
 
-/* Called with kretprobe_lock held */
 void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri,
 					struct pt_regs *regs)
 {
@@ -377,8 +376,7 @@ static int __kprobes trampoline_probe_handler(struct kprobe *p,
 	unsigned long trampoline_address = (unsigned long)&kretprobe_trampoline;
 
 	INIT_HLIST_HEAD(&empty_rp);
-	spin_lock_irqsave(&kretprobe_lock, flags);
-	head = kretprobe_inst_table_head(current);
+	kretprobe_hash_lock(current, &head, &flags);
 
 	/*
 	 * It is possible to have multiple instances associated with a given
@@ -417,7 +415,7 @@ static int __kprobes trampoline_probe_handler(struct kprobe *p,
 	regs->psw.addr = orig_ret_address | PSW_ADDR_AMODE;
 
 	reset_current_kprobe();
-	spin_unlock_irqrestore(&kretprobe_lock, flags);
+	kretprobe_hash_unlock(current, &flags);
 	preempt_enable_no_resched();
 
 	hlist_for_each_entry_safe(ri, node, tmp, &empty_rp, hlist) {
diff --git a/arch/sparc64/kernel/kprobes.c b/arch/sparc64/kernel/kprobes.c
index f43b5d755354..201a6e547e4a 100644
--- a/arch/sparc64/kernel/kprobes.c
+++ b/arch/sparc64/kernel/kprobes.c
@@ -478,9 +478,9 @@ int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 	return 0;
 }
 
-/* Called with kretprobe_lock held.  The value stored in the return
- * address register is actually 2 instructions before where the
- * callee will return to.  Sequences usually look something like this
+/* The value stored in the return address register is actually 2
+ * instructions before where the callee will return to.
+ * Sequences usually look something like this
  *
  *		call	some_function	<--- return register points here
  *		 nop			<--- call delay slot
@@ -512,8 +512,7 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
 	unsigned long trampoline_address =(unsigned long)&kretprobe_trampoline;
 
 	INIT_HLIST_HEAD(&empty_rp);
-	spin_lock_irqsave(&kretprobe_lock, flags);
-	head = kretprobe_inst_table_head(current);
+	kretprobe_hash_lock(current, &head, &flags);
 
 	/*
 	 * It is possible to have multiple instances associated with a given
@@ -553,7 +552,7 @@ int __kprobes trampoline_probe_handler(struct kprobe *p, struct pt_regs *regs)
 	regs->tnpc = orig_ret_address + 4;
 
 	reset_current_kprobe();
-	spin_unlock_irqrestore(&kretprobe_lock, flags);
+	kretprobe_hash_unlock(current, &flags);
 	preempt_enable_no_resched();
 
 	hlist_for_each_entry_safe(ri, node, tmp, &empty_rp, hlist) {
diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
index 43c019f85f0d..6c27679ec6aa 100644
--- a/arch/x86/kernel/kprobes.c
+++ b/arch/x86/kernel/kprobes.c
@@ -431,7 +431,6 @@ static void __kprobes prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
 		regs->ip = (unsigned long)p->ainsn.insn;
 }
 
-/* Called with kretprobe_lock held */
 void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri,
 				      struct pt_regs *regs)
 {
@@ -682,8 +681,7 @@ static __used __kprobes void *trampoline_handler(struct pt_regs *regs)
 	unsigned long trampoline_address = (unsigned long)&kretprobe_trampoline;
 
 	INIT_HLIST_HEAD(&empty_rp);
-	spin_lock_irqsave(&kretprobe_lock, flags);
-	head = kretprobe_inst_table_head(current);
+	kretprobe_hash_lock(current, &head, &flags);
 	/* fixup registers */
 #ifdef CONFIG_X86_64
 	regs->cs = __KERNEL_CS;
@@ -732,7 +730,7 @@ static __used __kprobes void *trampoline_handler(struct pt_regs *regs)
 
 	kretprobe_assert(ri, orig_ret_address, trampoline_address);
 
-	spin_unlock_irqrestore(&kretprobe_lock, flags);
+	kretprobe_hash_unlock(current, &flags);
 
 	hlist_for_each_entry_safe(ri, node, tmp, &empty_rp, hlist) {
 		hlist_del(&ri->hlist);
diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h
index 04a3556bdea6..0be7795655fa 100644
--- a/include/linux/kprobes.h
+++ b/include/linux/kprobes.h
@@ -157,11 +157,10 @@ struct kretprobe {
 	int nmissed;
 	size_t data_size;
 	struct hlist_head free_instances;
-	struct hlist_head used_instances;
+	spinlock_t lock;
 };
 
 struct kretprobe_instance {
-	struct hlist_node uflist; /* either on free list or used list */
 	struct hlist_node hlist;
 	struct kretprobe *rp;
 	kprobe_opcode_t *ret_addr;
@@ -201,7 +200,6 @@ static inline int init_test_probes(void)
 }
 #endif /* CONFIG_KPROBES_SANITY_TEST */
 
-extern spinlock_t kretprobe_lock;
 extern struct mutex kprobe_mutex;
 extern int arch_prepare_kprobe(struct kprobe *p);
 extern void arch_arm_kprobe(struct kprobe *p);
@@ -214,6 +212,9 @@ extern void kprobes_inc_nmissed_count(struct kprobe *p);
 
 /* Get the kprobe at this addr (if any) - called with preemption disabled */
 struct kprobe *get_kprobe(void *addr);
+void kretprobe_hash_lock(struct task_struct *tsk,
+			 struct hlist_head **head, unsigned long *flags);
+void kretprobe_hash_unlock(struct task_struct *tsk, unsigned long *flags);
 struct hlist_head * kretprobe_inst_table_head(struct task_struct *tsk);
 
 /* kprobe_running() will just return the current_kprobe on this CPU */
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index 1485ca8d0e00..cb0b3bde3617 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -62,6 +62,7 @@
 	addr = ((kprobe_opcode_t *)(kallsyms_lookup_name(name)))
 #endif
 
+static int kprobes_initialized;
 static struct hlist_head kprobe_table[KPROBE_TABLE_SIZE];
 static struct hlist_head kretprobe_inst_table[KPROBE_TABLE_SIZE];
 
@@ -69,8 +70,15 @@ static struct hlist_head kretprobe_inst_table[KPROBE_TABLE_SIZE];
 static bool kprobe_enabled;
 
 DEFINE_MUTEX(kprobe_mutex);		/* Protects kprobe_table */
-DEFINE_SPINLOCK(kretprobe_lock);	/* Protects kretprobe_inst_table */
 static DEFINE_PER_CPU(struct kprobe *, kprobe_instance) = NULL;
+static struct {
+	spinlock_t lock ____cacheline_aligned;
+} kretprobe_table_locks[KPROBE_TABLE_SIZE];
+
+static spinlock_t *kretprobe_table_lock_ptr(unsigned long hash)
+{
+	return &(kretprobe_table_locks[hash].lock);
+}
 
 /*
  * Normally, functions that we'd want to prohibit kprobes in, are marked
@@ -368,26 +376,53 @@ void __kprobes kprobes_inc_nmissed_count(struct kprobe *p)
 	return;
 }
 
-/* Called with kretprobe_lock held */
 void __kprobes recycle_rp_inst(struct kretprobe_instance *ri,
 				struct hlist_head *head)
 {
+	struct kretprobe *rp = ri->rp;
+
 	/* remove rp inst off the rprobe_inst_table */
 	hlist_del(&ri->hlist);
-	if (ri->rp) {
-		/* remove rp inst off the used list */
-		hlist_del(&ri->uflist);
-		/* put rp inst back onto the free list */
-		INIT_HLIST_NODE(&ri->uflist);
-		hlist_add_head(&ri->uflist, &ri->rp->free_instances);
+	INIT_HLIST_NODE(&ri->hlist);
+	if (likely(rp)) {
+		spin_lock(&rp->lock);
+		hlist_add_head(&ri->hlist, &rp->free_instances);
+		spin_unlock(&rp->lock);
 	} else
 		/* Unregistering */
 		hlist_add_head(&ri->hlist, head);
 }
 
-struct hlist_head __kprobes *kretprobe_inst_table_head(struct task_struct *tsk)
+void kretprobe_hash_lock(struct task_struct *tsk,
+			 struct hlist_head **head, unsigned long *flags)
 {
-	return &kretprobe_inst_table[hash_ptr(tsk, KPROBE_HASH_BITS)];
+	unsigned long hash = hash_ptr(tsk, KPROBE_HASH_BITS);
+	spinlock_t *hlist_lock;
+
+	*head = &kretprobe_inst_table[hash];
+	hlist_lock = kretprobe_table_lock_ptr(hash);
+	spin_lock_irqsave(hlist_lock, *flags);
+}
+
+void kretprobe_table_lock(unsigned long hash, unsigned long *flags)
+{
+	spinlock_t *hlist_lock = kretprobe_table_lock_ptr(hash);
+	spin_lock_irqsave(hlist_lock, *flags);
+}
+
+void kretprobe_hash_unlock(struct task_struct *tsk, unsigned long *flags)
+{
+	unsigned long hash = hash_ptr(tsk, KPROBE_HASH_BITS);
+	spinlock_t *hlist_lock;
+
+	hlist_lock = kretprobe_table_lock_ptr(hash);
+	spin_unlock_irqrestore(hlist_lock, *flags);
+}
+
+void kretprobe_table_unlock(unsigned long hash, unsigned long *flags)
+{
+	spinlock_t *hlist_lock = kretprobe_table_lock_ptr(hash);
+	spin_unlock_irqrestore(hlist_lock, *flags);
 }
 
 /*
@@ -401,17 +436,21 @@ void __kprobes kprobe_flush_task(struct task_struct *tk)
 	struct kretprobe_instance *ri;
 	struct hlist_head *head, empty_rp;
 	struct hlist_node *node, *tmp;
-	unsigned long flags = 0;
+	unsigned long hash, flags = 0;
 
-	INIT_HLIST_HEAD(&empty_rp);
-	spin_lock_irqsave(&kretprobe_lock, flags);
-	head = kretprobe_inst_table_head(tk);
+	if (unlikely(!kprobes_initialized))
+		/* Early boot.  kretprobe_table_locks not yet initialized. */
+		return;
+
+	hash = hash_ptr(tk, KPROBE_HASH_BITS);
+	head = &kretprobe_inst_table[hash];
+	kretprobe_table_lock(hash, &flags);
 	hlist_for_each_entry_safe(ri, node, tmp, head, hlist) {
 		if (ri->task == tk)
 			recycle_rp_inst(ri, &empty_rp);
 	}
-	spin_unlock_irqrestore(&kretprobe_lock, flags);
-
+	kretprobe_table_unlock(hash, &flags);
+	INIT_HLIST_HEAD(&empty_rp);
 	hlist_for_each_entry_safe(ri, node, tmp, &empty_rp, hlist) {
 		hlist_del(&ri->hlist);
 		kfree(ri);
@@ -423,24 +462,29 @@ static inline void free_rp_inst(struct kretprobe *rp)
 	struct kretprobe_instance *ri;
 	struct hlist_node *pos, *next;
 
-	hlist_for_each_entry_safe(ri, pos, next, &rp->free_instances, uflist) {
-		hlist_del(&ri->uflist);
+	hlist_for_each_entry_safe(ri, pos, next, &rp->free_instances, hlist) {
+		hlist_del(&ri->hlist);
 		kfree(ri);
 	}
 }
 
 static void __kprobes cleanup_rp_inst(struct kretprobe *rp)
 {
-	unsigned long flags;
+	unsigned long flags, hash;
 	struct kretprobe_instance *ri;
 	struct hlist_node *pos, *next;
+	struct hlist_head *head;
+
 	/* No race here */
-	spin_lock_irqsave(&kretprobe_lock, flags);
-	hlist_for_each_entry_safe(ri, pos, next, &rp->used_instances, uflist) {
-		ri->rp = NULL;
-		hlist_del(&ri->uflist);
+	for (hash = 0; hash < KPROBE_TABLE_SIZE; hash++) {
+		kretprobe_table_lock(hash, &flags);
+		head = &kretprobe_inst_table[hash];
+		hlist_for_each_entry_safe(ri, pos, next, head, hlist) {
+			if (ri->rp == rp)
+				ri->rp = NULL;
+		}
+		kretprobe_table_unlock(hash, &flags);
 	}
-	spin_unlock_irqrestore(&kretprobe_lock, flags);
 	free_rp_inst(rp);
 }
 
@@ -831,32 +875,37 @@ static int __kprobes pre_handler_kretprobe(struct kprobe *p,
 					   struct pt_regs *regs)
 {
 	struct kretprobe *rp = container_of(p, struct kretprobe, kp);
-	unsigned long flags = 0;
+	unsigned long hash, flags = 0;
+	struct kretprobe_instance *ri;
 
 	/*TODO: consider to only swap the RA after the last pre_handler fired */
-	spin_lock_irqsave(&kretprobe_lock, flags);
+	hash = hash_ptr(current, KPROBE_HASH_BITS);
+	spin_lock_irqsave(&rp->lock, flags);
 	if (!hlist_empty(&rp->free_instances)) {
-		struct kretprobe_instance *ri;
-
 		ri = hlist_entry(rp->free_instances.first,
-				 struct kretprobe_instance, uflist);
+				struct kretprobe_instance, hlist);
+		hlist_del(&ri->hlist);
+		spin_unlock_irqrestore(&rp->lock, flags);
+
 		ri->rp = rp;
 		ri->task = current;
 
 		if (rp->entry_handler && rp->entry_handler(ri, regs)) {
-			spin_unlock_irqrestore(&kretprobe_lock, flags);
+			spin_unlock_irqrestore(&rp->lock, flags);
 			return 0;
 		}
 
 		arch_prepare_kretprobe(ri, regs);
 
 		/* XXX(hch): why is there no hlist_move_head? */
-		hlist_del(&ri->uflist);
-		hlist_add_head(&ri->uflist, &ri->rp->used_instances);
-		hlist_add_head(&ri->hlist, kretprobe_inst_table_head(ri->task));
-	} else
+		INIT_HLIST_NODE(&ri->hlist);
+		kretprobe_table_lock(hash, &flags);
+		hlist_add_head(&ri->hlist, &kretprobe_inst_table[hash]);
+		kretprobe_table_unlock(hash, &flags);
+	} else {
 		rp->nmissed++;
-	spin_unlock_irqrestore(&kretprobe_lock, flags);
+		spin_unlock_irqrestore(&rp->lock, flags);
+	}
 	return 0;
 }
 
@@ -892,7 +941,7 @@ static int __kprobes __register_kretprobe(struct kretprobe *rp,
 		rp->maxactive = NR_CPUS;
 #endif
 	}
-	INIT_HLIST_HEAD(&rp->used_instances);
+	spin_lock_init(&rp->lock);
 	INIT_HLIST_HEAD(&rp->free_instances);
 	for (i = 0; i < rp->maxactive; i++) {
 		inst = kmalloc(sizeof(struct kretprobe_instance) +
@@ -901,8 +950,8 @@ static int __kprobes __register_kretprobe(struct kretprobe *rp,
 			free_rp_inst(rp);
 			return -ENOMEM;
 		}
-		INIT_HLIST_NODE(&inst->uflist);
-		hlist_add_head(&inst->uflist, &rp->free_instances);
+		INIT_HLIST_NODE(&inst->hlist);
+		hlist_add_head(&inst->hlist, &rp->free_instances);
 	}
 
 	rp->nmissed = 0;
@@ -1009,6 +1058,7 @@ static int __init init_kprobes(void)
 	for (i = 0; i < KPROBE_TABLE_SIZE; i++) {
 		INIT_HLIST_HEAD(&kprobe_table[i]);
 		INIT_HLIST_HEAD(&kretprobe_inst_table[i]);
+		spin_lock_init(&(kretprobe_table_locks[i].lock));
 	}
 
 	/*
@@ -1050,6 +1100,7 @@ static int __init init_kprobes(void)
 	err = arch_init_kprobes();
 	if (!err)
 		err = register_die_notifier(&kprobe_exceptions_nb);
+	kprobes_initialized = (err == 0);
 
 	if (!err)
 		init_test_probes();

From 8b6dd986823a8d92ed9f54baa5cef8604d9d9d44 Mon Sep 17 00:00:00 2001
From: Abhishek Sagar <sagar.abhishek@gmail.com>
Date: Fri, 25 Jul 2008 01:46:05 -0700
Subject: [PATCH 0723/1171] kprobes: remove redundant config check

I noticed that there's a CONFIG_KPROBES check inside kernel/kprobes.c,
which is redundant.

Signed-off-by: Abhishek Sagar <sagar.abhishek@gmail.com>
Acked-by: Masami Hiramatsu <mhiramat@redhat.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Cc: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/kprobes.c | 5 -----
 1 file changed, 5 deletions(-)

diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index cb0b3bde3617..75bc2cd9ebc6 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -1337,13 +1337,8 @@ EXPORT_SYMBOL_GPL(register_jprobe);
 EXPORT_SYMBOL_GPL(unregister_jprobe);
 EXPORT_SYMBOL_GPL(register_jprobes);
 EXPORT_SYMBOL_GPL(unregister_jprobes);
-#ifdef CONFIG_KPROBES
 EXPORT_SYMBOL_GPL(jprobe_return);
-#endif
-
-#ifdef CONFIG_KPROBES
 EXPORT_SYMBOL_GPL(register_kretprobe);
 EXPORT_SYMBOL_GPL(unregister_kretprobe);
 EXPORT_SYMBOL_GPL(register_kretprobes);
 EXPORT_SYMBOL_GPL(unregister_kretprobes);
-#endif

From d8f388d8dc8d4f36539dd37c1fff62cc404ea0fc Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Fri, 25 Jul 2008 01:46:07 -0700
Subject: [PATCH 0724/1171] gpio: sysfs interface

This adds a simple sysfs interface for GPIOs.

    /sys/class/gpio
    	/export ... asks the kernel to export a GPIO to userspace
    	/unexport ... to return a GPIO to the kernel
        /gpioN ... for each exported GPIO #N
	    /value ... always readable, writes fail for input GPIOs
	    /direction ... r/w as: in, out (default low); write high, low
	/gpiochipN ... for each gpiochip; #N is its first GPIO
	    /base ... (r/o) same as N
	    /label ... (r/o) descriptive, not necessarily unique
	    /ngpio ... (r/o) number of GPIOs; numbered N .. N+(ngpio - 1)

GPIOs claimed by kernel code may be exported by its owner using a new
gpio_export() call, which should be most useful for driver debugging.
Such exports may optionally be done without a "direction" attribute.

Userspace may ask to take over a GPIO by writing to a sysfs control file,
helping to cope with incomplete board support or other "one-off"
requirements that don't merit full kernel support:

  echo 23 > /sys/class/gpio/export
	... will gpio_request(23, "sysfs") and gpio_export(23);
	use /sys/class/gpio/gpio-23/direction to (re)configure it,
	when that GPIO can be used as both input and output.
  echo 23 > /sys/class/gpio/unexport
	... will gpio_free(23), when it was exported as above

The extra D-space footprint is a few hundred bytes, except for the sysfs
resources associated with each exported GPIO.  The additional I-space
footprint is about two thirds of the current size of gpiolib (!).  Since
no /dev node creation is involved, no "udev" support is needed.

Related changes:

  * This adds a device pointer to "struct gpio_chip".  When GPIO
    providers initialize that, sysfs gpio class devices become children of
    that device instead of being "virtual" devices.

  * The (few) gpio_chip providers which have such a device node have
    been updated.

  * Some gpio_chip drivers also needed to update their module "owner"
    field ...  for which missing kerneldoc was added.

  * Some gpio_chips don't support input GPIOs.  Those GPIOs are now
    flagged appropriately when the chip is registered.

Based on previous patches, and discussion both on and off LKML.

A Documentation/ABI/testing/sysfs-gpio update is ready to submit once this
merges to mainline.

[akpm@linux-foundation.org: a few maintenance build fixes]
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Cc: Guennadi Liakhovetski <g.liakhovetski@pengutronix.de>
Cc: Greg KH <greg@kroah.com>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/gpio.txt       | 123 +++++++-
 arch/arm/plat-omap/gpio.c    |   3 +
 arch/avr32/mach-at32ap/pio.c |   2 +
 drivers/gpio/Kconfig         |  15 +
 drivers/gpio/gpiolib.c       | 536 ++++++++++++++++++++++++++++++++++-
 drivers/gpio/mcp23s08.c      |   1 +
 drivers/gpio/pca953x.c       |   1 +
 drivers/gpio/pcf857x.c       |   1 +
 drivers/i2c/chips/tps65010.c |   2 +
 drivers/mfd/htc-egpio.c      |   2 +
 include/asm-generic/gpio.h   |  33 ++-
 include/linux/gpio.h         |  13 +
 12 files changed, 712 insertions(+), 20 deletions(-)

diff --git a/Documentation/gpio.txt b/Documentation/gpio.txt
index c35ca9e40d4c..8b69811a9642 100644
--- a/Documentation/gpio.txt
+++ b/Documentation/gpio.txt
@@ -347,15 +347,12 @@ necessarily be nonportable.
 Dynamic definition of GPIOs is not currently standard; for example, as
 a side effect of configuring an add-on board with some GPIO expanders.
 
-These calls are purely for kernel space, but a userspace API could be built
-on top of them.
-
 
 GPIO implementor's framework (OPTIONAL)
 =======================================
 As noted earlier, there is an optional implementation framework making it
 easier for platforms to support different kinds of GPIO controller using
-the same programming interface.
+the same programming interface.  This framework is called "gpiolib".
 
 As a debugging aid, if debugfs is available a /sys/kernel/debug/gpio file
 will be found there.  That will list all the controllers registered through
@@ -439,4 +436,120 @@ becomes available.  That may mean the device should not be registered until
 calls for that GPIO can work.  One way to address such dependencies is for
 such gpio_chip controllers to provide setup() and teardown() callbacks to
 board specific code; those board specific callbacks would register devices
-once all the necessary resources are available.
+once all the necessary resources are available, and remove them later when
+the GPIO controller device becomes unavailable.
+
+
+Sysfs Interface for Userspace (OPTIONAL)
+========================================
+Platforms which use the "gpiolib" implementors framework may choose to
+configure a sysfs user interface to GPIOs.  This is different from the
+debugfs interface, since it provides control over GPIO direction and
+value instead of just showing a gpio state summary.  Plus, it could be
+present on production systems without debugging support.
+
+Given approprate hardware documentation for the system, userspace could
+know for example that GPIO #23 controls the write protect line used to
+protect boot loader segments in flash memory.  System upgrade procedures
+may need to temporarily remove that protection, first importing a GPIO,
+then changing its output state, then updating the code before re-enabling
+the write protection.  In normal use, GPIO #23 would never be touched,
+and the kernel would have no need to know about it.
+
+Again depending on appropriate hardware documentation, on some systems
+userspace GPIO can be used to determine system configuration data that
+standard kernels won't know about.  And for some tasks, simple userspace
+GPIO drivers could be all that the system really needs.
+
+Note that standard kernel drivers exist for common "LEDs and Buttons"
+GPIO tasks:  "leds-gpio" and "gpio_keys", respectively.  Use those
+instead of talking directly to the GPIOs; they integrate with kernel
+frameworks better than your userspace code could.
+
+
+Paths in Sysfs
+--------------
+There are three kinds of entry in /sys/class/gpio:
+
+   -	Control interfaces used to get userspace control over GPIOs;
+
+   -	GPIOs themselves; and
+
+   -	GPIO controllers ("gpio_chip" instances).
+
+That's in addition to standard files including the "device" symlink.
+
+The control interfaces are write-only:
+
+    /sys/class/gpio/
+
+    	"export" ... Userspace may ask the kernel to export control of
+		a GPIO to userspace by writing its number to this file.
+
+		Example:  "echo 19 > export" will create a "gpio19" node
+		for GPIO #19, if that's not requested by kernel code.
+
+    	"unexport" ... Reverses the effect of exporting to userspace.
+
+		Example:  "echo 19 > unexport" will remove a "gpio19"
+		node exported using the "export" file.
+
+GPIO signals have paths like /sys/class/gpio/gpio42/ (for GPIO #42)
+and have the following read/write attributes:
+
+    /sys/class/gpio/gpioN/
+
+	"direction" ... reads as either "in" or "out".  This value may
+		normally be written.  Writing as "out" defaults to
+		initializing the value as low.  To ensure glitch free
+		operation, values "low" and "high" may be written to
+		configure the GPIO as an output with that initial value.
+
+		Note that this attribute *will not exist* if the kernel
+		doesn't support changing the direction of a GPIO, or
+		it was exported by kernel code that didn't explicitly
+		allow userspace to reconfigure this GPIO's direction.
+
+	"value" ... reads as either 0 (low) or 1 (high).  If the GPIO
+		is configured as an output, this value may be written;
+		any nonzero value is treated as high.
+
+GPIO controllers have paths like /sys/class/gpio/chipchip42/ (for the
+controller implementing GPIOs starting at #42) and have the following
+read-only attributes:
+
+    /sys/class/gpio/gpiochipN/
+
+    	"base" ... same as N, the first GPIO managed by this chip
+
+    	"label" ... provided for diagnostics (not always unique)
+
+    	"ngpio" ... how many GPIOs this manges (N to N + ngpio - 1)
+
+Board documentation should in most cases cover what GPIOs are used for
+what purposes.  However, those numbers are not always stable; GPIOs on
+a daughtercard might be different depending on the base board being used,
+or other cards in the stack.  In such cases, you may need to use the
+gpiochip nodes (possibly in conjunction with schematics) to determine
+the correct GPIO number to use for a given signal.
+
+
+Exporting from Kernel code
+--------------------------
+Kernel code can explicitly manage exports of GPIOs which have already been
+requested using gpio_request():
+
+	/* export the GPIO to userspace */
+	int gpio_export(unsigned gpio, bool direction_may_change);
+
+	/* reverse gpio_export() */
+	void gpio_unexport();
+
+After a kernel driver requests a GPIO, it may only be made available in
+the sysfs interface by gpio_export().  The driver can control whether the
+signal direction may change.  This helps drivers prevent userspace code
+from accidentally clobbering important system state.
+
+This explicit exporting can help with debugging (by making some kinds
+of experiments easier), or can provide an always-there interface that's
+suitable for documenting as part of a board support package.
diff --git a/arch/arm/plat-omap/gpio.c b/arch/arm/plat-omap/gpio.c
index 1903a3491ee9..d8e9c2c3f0f6 100644
--- a/arch/arm/plat-omap/gpio.c
+++ b/arch/arm/plat-omap/gpio.c
@@ -1488,6 +1488,9 @@ static int __init _omap_gpio_init(void)
 		bank->chip.set = gpio_set;
 		if (bank_is_mpuio(bank)) {
 			bank->chip.label = "mpuio";
+#ifdef CONFIG_ARCH_OMAP1
+			bank->chip.dev = &omap_mpuio_device.dev;
+#endif
 			bank->chip.base = OMAP_MPUIO(0);
 		} else {
 			bank->chip.label = "gpio";
diff --git a/arch/avr32/mach-at32ap/pio.c b/arch/avr32/mach-at32ap/pio.c
index 60da03ba7117..296294f8ed81 100644
--- a/arch/avr32/mach-at32ap/pio.c
+++ b/arch/avr32/mach-at32ap/pio.c
@@ -360,6 +360,8 @@ static int __init pio_probe(struct platform_device *pdev)
 	pio->chip.label = pio->name;
 	pio->chip.base = pdev->id * 32;
 	pio->chip.ngpio = 32;
+	pio->chip.dev = &pdev->dev;
+	pio->chip.owner = THIS_MODULE;
 
 	pio->chip.direction_input = direction_input;
 	pio->chip.get = gpio_get;
diff --git a/drivers/gpio/Kconfig b/drivers/gpio/Kconfig
index fced1909cbba..6ec0e35b98e3 100644
--- a/drivers/gpio/Kconfig
+++ b/drivers/gpio/Kconfig
@@ -23,6 +23,21 @@ config DEBUG_GPIO
 	  slower.  The diagnostics help catch the type of setup errors
 	  that are most common when setting up new platforms or boards.
 
+config GPIO_SYSFS
+	bool "/sys/class/gpio/... (sysfs interface)"
+	depends on SYSFS && EXPERIMENTAL
+	help
+	  Say Y here to add a sysfs interface for GPIOs.
+
+	  This is mostly useful to work around omissions in a system's
+	  kernel support.  Those are common in custom and semicustom
+	  hardware assembled using standard kernels with a minimum of
+	  custom patches.  In those cases, userspace code may import
+	  a given GPIO from the kernel, if no kernel driver requested it.
+
+	  Kernel drivers may also request that a particular GPIO be
+	  exported to userspace; this can be useful when debugging.
+
 # put expanders in the right section, in alphabetical order
 
 comment "I2C GPIO expanders:"
diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c
index beaf6b3a37dc..8d2940517c99 100644
--- a/drivers/gpio/gpiolib.c
+++ b/drivers/gpio/gpiolib.c
@@ -2,8 +2,11 @@
 #include <linux/module.h>
 #include <linux/irq.h>
 #include <linux/spinlock.h>
-
-#include <asm/gpio.h>
+#include <linux/device.h>
+#include <linux/err.h>
+#include <linux/debugfs.h>
+#include <linux/seq_file.h>
+#include <linux/gpio.h>
 
 
 /* Optional implementation infrastructure for GPIO interfaces.
@@ -44,6 +47,8 @@ struct gpio_desc {
 #define FLAG_REQUESTED	0
 #define FLAG_IS_OUT	1
 #define FLAG_RESERVED	2
+#define FLAG_EXPORT	3	/* protected by sysfs_lock */
+#define FLAG_SYSFS	4	/* exported via /sys/class/gpio/control */
 
 #ifdef CONFIG_DEBUG_FS
 	const char		*label;
@@ -151,6 +156,482 @@ err:
 	return ret;
 }
 
+#ifdef CONFIG_GPIO_SYSFS
+
+/* lock protects against unexport_gpio() being called while
+ * sysfs files are active.
+ */
+static DEFINE_MUTEX(sysfs_lock);
+
+/*
+ * /sys/class/gpio/gpioN... only for GPIOs that are exported
+ *   /direction
+ *      * MAY BE OMITTED if kernel won't allow direction changes
+ *      * is read/write as "in" or "out"
+ *      * may also be written as "high" or "low", initializing
+ *        output value as specified ("out" implies "low")
+ *   /value
+ *      * always readable, subject to hardware behavior
+ *      * may be writable, as zero/nonzero
+ *
+ * REVISIT there will likely be an attribute for configuring async
+ * notifications, e.g. to specify polling interval or IRQ trigger type
+ * that would for example trigger a poll() on the "value".
+ */
+
+static ssize_t gpio_direction_show(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	const struct gpio_desc	*desc = dev_get_drvdata(dev);
+	ssize_t			status;
+
+	mutex_lock(&sysfs_lock);
+
+	if (!test_bit(FLAG_EXPORT, &desc->flags))
+		status = -EIO;
+	else
+		status = sprintf(buf, "%s\n",
+			test_bit(FLAG_IS_OUT, &desc->flags)
+				? "out" : "in");
+
+	mutex_unlock(&sysfs_lock);
+	return status;
+}
+
+static ssize_t gpio_direction_store(struct device *dev,
+		struct device_attribute *attr, const char *buf, size_t size)
+{
+	const struct gpio_desc	*desc = dev_get_drvdata(dev);
+	unsigned		gpio = desc - gpio_desc;
+	ssize_t			status;
+
+	mutex_lock(&sysfs_lock);
+
+	if (!test_bit(FLAG_EXPORT, &desc->flags))
+		status = -EIO;
+	else if (sysfs_streq(buf, "high"))
+		status = gpio_direction_output(gpio, 1);
+	else if (sysfs_streq(buf, "out") || sysfs_streq(buf, "low"))
+		status = gpio_direction_output(gpio, 0);
+	else if (sysfs_streq(buf, "in"))
+		status = gpio_direction_input(gpio);
+	else
+		status = -EINVAL;
+
+	mutex_unlock(&sysfs_lock);
+	return status ? : size;
+}
+
+static const DEVICE_ATTR(direction, 0644,
+		gpio_direction_show, gpio_direction_store);
+
+static ssize_t gpio_value_show(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	const struct gpio_desc	*desc = dev_get_drvdata(dev);
+	unsigned		gpio = desc - gpio_desc;
+	ssize_t			status;
+
+	mutex_lock(&sysfs_lock);
+
+	if (!test_bit(FLAG_EXPORT, &desc->flags))
+		status = -EIO;
+	else
+		status = sprintf(buf, "%d\n", gpio_get_value_cansleep(gpio));
+
+	mutex_unlock(&sysfs_lock);
+	return status;
+}
+
+static ssize_t gpio_value_store(struct device *dev,
+		struct device_attribute *attr, const char *buf, size_t size)
+{
+	const struct gpio_desc	*desc = dev_get_drvdata(dev);
+	unsigned		gpio = desc - gpio_desc;
+	ssize_t			status;
+
+	mutex_lock(&sysfs_lock);
+
+	if (!test_bit(FLAG_EXPORT, &desc->flags))
+		status = -EIO;
+	else if (!test_bit(FLAG_IS_OUT, &desc->flags))
+		status = -EPERM;
+	else {
+		long		value;
+
+		status = strict_strtol(buf, 0, &value);
+		if (status == 0) {
+			gpio_set_value_cansleep(gpio, value != 0);
+			status = size;
+		}
+	}
+
+	mutex_unlock(&sysfs_lock);
+	return status;
+}
+
+static /*const*/ DEVICE_ATTR(value, 0644,
+		gpio_value_show, gpio_value_store);
+
+static const struct attribute *gpio_attrs[] = {
+	&dev_attr_direction.attr,
+	&dev_attr_value.attr,
+	NULL,
+};
+
+static const struct attribute_group gpio_attr_group = {
+	.attrs = (struct attribute **) gpio_attrs,
+};
+
+/*
+ * /sys/class/gpio/gpiochipN/
+ *   /base ... matching gpio_chip.base (N)
+ *   /label ... matching gpio_chip.label
+ *   /ngpio ... matching gpio_chip.ngpio
+ */
+
+static ssize_t chip_base_show(struct device *dev,
+			       struct device_attribute *attr, char *buf)
+{
+	const struct gpio_chip	*chip = dev_get_drvdata(dev);
+
+	return sprintf(buf, "%d\n", chip->base);
+}
+static DEVICE_ATTR(base, 0444, chip_base_show, NULL);
+
+static ssize_t chip_label_show(struct device *dev,
+			       struct device_attribute *attr, char *buf)
+{
+	const struct gpio_chip	*chip = dev_get_drvdata(dev);
+
+	return sprintf(buf, "%s\n", chip->label ? : "");
+}
+static DEVICE_ATTR(label, 0444, chip_label_show, NULL);
+
+static ssize_t chip_ngpio_show(struct device *dev,
+			       struct device_attribute *attr, char *buf)
+{
+	const struct gpio_chip	*chip = dev_get_drvdata(dev);
+
+	return sprintf(buf, "%u\n", chip->ngpio);
+}
+static DEVICE_ATTR(ngpio, 0444, chip_ngpio_show, NULL);
+
+static const struct attribute *gpiochip_attrs[] = {
+	&dev_attr_base.attr,
+	&dev_attr_label.attr,
+	&dev_attr_ngpio.attr,
+	NULL,
+};
+
+static const struct attribute_group gpiochip_attr_group = {
+	.attrs = (struct attribute **) gpiochip_attrs,
+};
+
+/*
+ * /sys/class/gpio/export ... write-only
+ *	integer N ... number of GPIO to export (full access)
+ * /sys/class/gpio/unexport ... write-only
+ *	integer N ... number of GPIO to unexport
+ */
+static ssize_t export_store(struct class *class, const char *buf, size_t len)
+{
+	long	gpio;
+	int	status;
+
+	status = strict_strtol(buf, 0, &gpio);
+	if (status < 0)
+		goto done;
+
+	/* No extra locking here; FLAG_SYSFS just signifies that the
+	 * request and export were done by on behalf of userspace, so
+	 * they may be undone on its behalf too.
+	 */
+
+	status = gpio_request(gpio, "sysfs");
+	if (status < 0)
+		goto done;
+
+	status = gpio_export(gpio, true);
+	if (status < 0)
+		gpio_free(gpio);
+	else
+		set_bit(FLAG_SYSFS, &gpio_desc[gpio].flags);
+
+done:
+	if (status)
+		pr_debug("%s: status %d\n", __func__, status);
+	return status ? : len;
+}
+
+static ssize_t unexport_store(struct class *class, const char *buf, size_t len)
+{
+	long	gpio;
+	int	status;
+
+	status = strict_strtol(buf, 0, &gpio);
+	if (status < 0)
+		goto done;
+
+	status = -EINVAL;
+
+	/* reject bogus commands (gpio_unexport ignores them) */
+	if (!gpio_is_valid(gpio))
+		goto done;
+
+	/* No extra locking here; FLAG_SYSFS just signifies that the
+	 * request and export were done by on behalf of userspace, so
+	 * they may be undone on its behalf too.
+	 */
+	if (test_and_clear_bit(FLAG_SYSFS, &gpio_desc[gpio].flags)) {
+		status = 0;
+		gpio_free(gpio);
+	}
+done:
+	if (status)
+		pr_debug("%s: status %d\n", __func__, status);
+	return status ? : len;
+}
+
+static struct class_attribute gpio_class_attrs[] = {
+	__ATTR(export, 0200, NULL, export_store),
+	__ATTR(unexport, 0200, NULL, unexport_store),
+	__ATTR_NULL,
+};
+
+static struct class gpio_class = {
+	.name =		"gpio",
+	.owner =	THIS_MODULE,
+
+	.class_attrs =	gpio_class_attrs,
+};
+
+
+/**
+ * gpio_export - export a GPIO through sysfs
+ * @gpio: gpio to make available, already requested
+ * @direction_may_change: true if userspace may change gpio direction
+ * Context: arch_initcall or later
+ *
+ * When drivers want to make a GPIO accessible to userspace after they
+ * have requested it -- perhaps while debugging, or as part of their
+ * public interface -- they may use this routine.  If the GPIO can
+ * change direction (some can't) and the caller allows it, userspace
+ * will see "direction" sysfs attribute which may be used to change
+ * the gpio's direction.  A "value" attribute will always be provided.
+ *
+ * Returns zero on success, else an error.
+ */
+int gpio_export(unsigned gpio, bool direction_may_change)
+{
+	unsigned long		flags;
+	struct gpio_desc	*desc;
+	int			status = -EINVAL;
+
+	/* can't export until sysfs is available ... */
+	if (!gpio_class.p) {
+		pr_debug("%s: called too early!\n", __func__);
+		return -ENOENT;
+	}
+
+	if (!gpio_is_valid(gpio))
+		goto done;
+
+	mutex_lock(&sysfs_lock);
+
+	spin_lock_irqsave(&gpio_lock, flags);
+	desc = &gpio_desc[gpio];
+	if (test_bit(FLAG_REQUESTED, &desc->flags)
+			&& !test_bit(FLAG_EXPORT, &desc->flags)) {
+		status = 0;
+		if (!desc->chip->direction_input
+				|| !desc->chip->direction_output)
+			direction_may_change = false;
+	}
+	spin_unlock_irqrestore(&gpio_lock, flags);
+
+	if (status == 0) {
+		struct device	*dev;
+
+		dev = device_create(&gpio_class, desc->chip->dev, MKDEV(0, 0),
+					desc, "gpio%d", gpio);
+		if (dev) {
+			if (direction_may_change)
+				status = sysfs_create_group(&dev->kobj,
+						&gpio_attr_group);
+			else
+				status = device_create_file(dev,
+						&dev_attr_value);
+			if (status != 0)
+				device_unregister(dev);
+		} else
+			status = -ENODEV;
+		if (status == 0)
+			set_bit(FLAG_EXPORT, &desc->flags);
+	}
+
+	mutex_unlock(&sysfs_lock);
+
+done:
+	if (status)
+		pr_debug("%s: gpio%d status %d\n", __func__, gpio, status);
+
+	return status;
+}
+EXPORT_SYMBOL_GPL(gpio_export);
+
+static int match_export(struct device *dev, void *data)
+{
+	return dev_get_drvdata(dev) == data;
+}
+
+/**
+ * gpio_unexport - reverse effect of gpio_export()
+ * @gpio: gpio to make unavailable
+ *
+ * This is implicit on gpio_free().
+ */
+void gpio_unexport(unsigned gpio)
+{
+	struct gpio_desc	*desc;
+	int			status = -EINVAL;
+
+	if (!gpio_is_valid(gpio))
+		goto done;
+
+	mutex_lock(&sysfs_lock);
+
+	desc = &gpio_desc[gpio];
+	if (test_bit(FLAG_EXPORT, &desc->flags)) {
+		struct device	*dev = NULL;
+
+		dev = class_find_device(&gpio_class, NULL, desc, match_export);
+		if (dev) {
+			clear_bit(FLAG_EXPORT, &desc->flags);
+			put_device(dev);
+			device_unregister(dev);
+			status = 0;
+		} else
+			status = -ENODEV;
+	}
+
+	mutex_unlock(&sysfs_lock);
+done:
+	if (status)
+		pr_debug("%s: gpio%d status %d\n", __func__, gpio, status);
+}
+EXPORT_SYMBOL_GPL(gpio_unexport);
+
+static int gpiochip_export(struct gpio_chip *chip)
+{
+	int		status;
+	struct device	*dev;
+
+	/* Many systems register gpio chips for SOC support very early,
+	 * before driver model support is available.  In those cases we
+	 * export this later, in gpiolib_sysfs_init() ... here we just
+	 * verify that _some_ field of gpio_class got initialized.
+	 */
+	if (!gpio_class.p)
+		return 0;
+
+	/* use chip->base for the ID; it's already known to be unique */
+	mutex_lock(&sysfs_lock);
+	dev = device_create(&gpio_class, chip->dev, MKDEV(0, 0), chip,
+				"gpiochip%d", chip->base);
+	if (dev) {
+		status = sysfs_create_group(&dev->kobj,
+				&gpiochip_attr_group);
+	} else
+		status = -ENODEV;
+	chip->exported = (status == 0);
+	mutex_unlock(&sysfs_lock);
+
+	if (status) {
+		unsigned long	flags;
+		unsigned	gpio;
+
+		spin_lock_irqsave(&gpio_lock, flags);
+		gpio = chip->base;
+		while (gpio_desc[gpio].chip == chip)
+			gpio_desc[gpio++].chip = NULL;
+		spin_unlock_irqrestore(&gpio_lock, flags);
+
+		pr_debug("%s: chip %s status %d\n", __func__,
+				chip->label, status);
+	}
+
+	return status;
+}
+
+static void gpiochip_unexport(struct gpio_chip *chip)
+{
+	int			status;
+	struct device		*dev;
+
+	mutex_lock(&sysfs_lock);
+	dev = class_find_device(&gpio_class, NULL, chip, match_export);
+	if (dev) {
+		put_device(dev);
+		device_unregister(dev);
+		chip->exported = 0;
+		status = 0;
+	} else
+		status = -ENODEV;
+	mutex_unlock(&sysfs_lock);
+
+	if (status)
+		pr_debug("%s: chip %s status %d\n", __func__,
+				chip->label, status);
+}
+
+static int __init gpiolib_sysfs_init(void)
+{
+	int		status;
+	unsigned long	flags;
+	unsigned	gpio;
+
+	status = class_register(&gpio_class);
+	if (status < 0)
+		return status;
+
+	/* Scan and register the gpio_chips which registered very
+	 * early (e.g. before the class_register above was called).
+	 *
+	 * We run before arch_initcall() so chip->dev nodes can have
+	 * registered, and so arch_initcall() can always gpio_export().
+	 */
+	spin_lock_irqsave(&gpio_lock, flags);
+	for (gpio = 0; gpio < ARCH_NR_GPIOS; gpio++) {
+		struct gpio_chip	*chip;
+
+		chip = gpio_desc[gpio].chip;
+		if (!chip || chip->exported)
+			continue;
+
+		spin_unlock_irqrestore(&gpio_lock, flags);
+		status = gpiochip_export(chip);
+		spin_lock_irqsave(&gpio_lock, flags);
+	}
+	spin_unlock_irqrestore(&gpio_lock, flags);
+
+
+	return status;
+}
+postcore_initcall(gpiolib_sysfs_init);
+
+#else
+static inline int gpiochip_export(struct gpio_chip *chip)
+{
+	return 0;
+}
+
+static inline void gpiochip_unexport(struct gpio_chip *chip)
+{
+}
+
+#endif /* CONFIG_GPIO_SYSFS */
+
 /**
  * gpiochip_add() - register a gpio_chip
  * @chip: the chip to register, with chip->base initialized
@@ -160,6 +641,11 @@ err:
  * because the chip->base is invalid or already associated with a
  * different chip.  Otherwise it returns zero as a success code.
  *
+ * When gpiochip_add() is called very early during boot, so that GPIOs
+ * can be freely used, the chip->dev device must be registered before
+ * the gpio framework's arch_initcall().  Otherwise sysfs initialization
+ * for GPIOs will fail rudely.
+ *
  * If chip->base is negative, this requests dynamic assignment of
  * a range of valid GPIOs.
  */
@@ -182,7 +668,7 @@ int gpiochip_add(struct gpio_chip *chip)
 		base = gpiochip_find_base(chip->ngpio);
 		if (base < 0) {
 			status = base;
-			goto fail_unlock;
+			goto unlock;
 		}
 		chip->base = base;
 	}
@@ -197,12 +683,23 @@ int gpiochip_add(struct gpio_chip *chip)
 	if (status == 0) {
 		for (id = base; id < base + chip->ngpio; id++) {
 			gpio_desc[id].chip = chip;
-			gpio_desc[id].flags = 0;
+
+			/* REVISIT:  most hardware initializes GPIOs as
+			 * inputs (often with pullups enabled) so power
+			 * usage is minimized.  Linux code should set the
+			 * gpio direction first thing; but until it does,
+			 * we may expose the wrong direction in sysfs.
+			 */
+			gpio_desc[id].flags = !chip->direction_input
+				? (1 << FLAG_IS_OUT)
+				: 0;
 		}
 	}
 
-fail_unlock:
+unlock:
 	spin_unlock_irqrestore(&gpio_lock, flags);
+	if (status == 0)
+		status = gpiochip_export(chip);
 fail:
 	/* failures here can mean systems won't boot... */
 	if (status)
@@ -239,6 +736,10 @@ int gpiochip_remove(struct gpio_chip *chip)
 	}
 
 	spin_unlock_irqrestore(&gpio_lock, flags);
+
+	if (status == 0)
+		gpiochip_unexport(chip);
+
 	return status;
 }
 EXPORT_SYMBOL_GPL(gpiochip_remove);
@@ -296,6 +797,8 @@ void gpio_free(unsigned gpio)
 		return;
 	}
 
+	gpio_unexport(gpio);
+
 	spin_lock_irqsave(&gpio_lock, flags);
 
 	desc = &gpio_desc[gpio];
@@ -534,10 +1037,6 @@ EXPORT_SYMBOL_GPL(gpio_set_value_cansleep);
 
 #ifdef CONFIG_DEBUG_FS
 
-#include <linux/debugfs.h>
-#include <linux/seq_file.h>
-
-
 static void gpiolib_dbg_show(struct seq_file *s, struct gpio_chip *chip)
 {
 	unsigned		i;
@@ -614,17 +1113,28 @@ static int gpiolib_show(struct seq_file *s, void *unused)
 	/* REVISIT this isn't locked against gpio_chip removal ... */
 
 	for (gpio = 0; gpio_is_valid(gpio); gpio++) {
+		struct device *dev;
+
 		if (chip == gpio_desc[gpio].chip)
 			continue;
 		chip = gpio_desc[gpio].chip;
 		if (!chip)
 			continue;
 
-		seq_printf(s, "%sGPIOs %d-%d, %s%s:\n",
+		seq_printf(s, "%sGPIOs %d-%d",
 				started ? "\n" : "",
-				chip->base, chip->base + chip->ngpio - 1,
-				chip->label ? : "generic",
-				chip->can_sleep ? ", can sleep" : "");
+				chip->base, chip->base + chip->ngpio - 1);
+		dev = chip->dev;
+		if (dev)
+			seq_printf(s, ", %s/%s",
+				dev->bus ? dev->bus->name : "no-bus",
+				dev->bus_id);
+		if (chip->label)
+			seq_printf(s, ", %s", chip->label);
+		if (chip->can_sleep)
+			seq_printf(s, ", can sleep");
+		seq_printf(s, ":\n");
+
 		started = 1;
 		if (chip->dbg_show)
 			chip->dbg_show(s, chip);
diff --git a/drivers/gpio/mcp23s08.c b/drivers/gpio/mcp23s08.c
index 7f92fdd5f0e2..7efd7d3a81f9 100644
--- a/drivers/gpio/mcp23s08.c
+++ b/drivers/gpio/mcp23s08.c
@@ -239,6 +239,7 @@ static int mcp23s08_probe(struct spi_device *spi)
 	mcp->chip.base = pdata->base;
 	mcp->chip.ngpio = 8;
 	mcp->chip.can_sleep = 1;
+	mcp->chip.dev = &spi->dev;
 	mcp->chip.owner = THIS_MODULE;
 
 	spi_set_drvdata(spi, mcp);
diff --git a/drivers/gpio/pca953x.c b/drivers/gpio/pca953x.c
index a380730b61ab..cc8468692ae0 100644
--- a/drivers/gpio/pca953x.c
+++ b/drivers/gpio/pca953x.c
@@ -188,6 +188,7 @@ static void pca953x_setup_gpio(struct pca953x_chip *chip, int gpios)
 	gc->base = chip->gpio_start;
 	gc->ngpio = gpios;
 	gc->label = chip->client->name;
+	gc->dev = &chip->client->dev;
 	gc->owner = THIS_MODULE;
 }
 
diff --git a/drivers/gpio/pcf857x.c b/drivers/gpio/pcf857x.c
index d25d356c4f20..fc9c6ae739ee 100644
--- a/drivers/gpio/pcf857x.c
+++ b/drivers/gpio/pcf857x.c
@@ -200,6 +200,7 @@ static int pcf857x_probe(struct i2c_client *client,
 
 	gpio->chip.base = pdata->gpio_base;
 	gpio->chip.can_sleep = 1;
+	gpio->chip.dev = &client->dev;
 	gpio->chip.owner = THIS_MODULE;
 
 	/* NOTE:  the OnSemi jlc1562b is also largely compatible with
diff --git a/drivers/i2c/chips/tps65010.c b/drivers/i2c/chips/tps65010.c
index 85949685191b..cf02e8fceb42 100644
--- a/drivers/i2c/chips/tps65010.c
+++ b/drivers/i2c/chips/tps65010.c
@@ -636,6 +636,8 @@ static int tps65010_probe(struct i2c_client *client,
 		tps->outmask = board->outmask;
 
 		tps->chip.label = client->name;
+		tps->chip.dev = &client->dev;
+		tps->chip.owner = THIS_MODULE;
 
 		tps->chip.set = tps65010_gpio_set;
 		tps->chip.direction_output = tps65010_output;
diff --git a/drivers/mfd/htc-egpio.c b/drivers/mfd/htc-egpio.c
index 8872cc077519..6be43172dc65 100644
--- a/drivers/mfd/htc-egpio.c
+++ b/drivers/mfd/htc-egpio.c
@@ -318,6 +318,8 @@ static int __init egpio_probe(struct platform_device *pdev)
 		ei->chip[i].dev = &(pdev->dev);
 		chip = &(ei->chip[i].chip);
 		chip->label           = "htc-egpio";
+		chip->dev             = &pdev->dev;
+		chip->owner           = THIS_MODULE;
 		chip->get             = egpio_get;
 		chip->set             = egpio_set;
 		chip->direction_input = egpio_direction_input;
diff --git a/include/asm-generic/gpio.h b/include/asm-generic/gpio.h
index 6be061d09da9..1beff5166e53 100644
--- a/include/asm-generic/gpio.h
+++ b/include/asm-generic/gpio.h
@@ -32,6 +32,8 @@ struct module;
 /**
  * struct gpio_chip - abstract a GPIO controller
  * @label: for diagnostics
+ * @dev: optional device providing the GPIOs
+ * @owner: helps prevent removal of modules exporting active GPIOs
  * @direction_input: configures signal "offset" as input, or returns error
  * @get: returns value for signal "offset"; for output signals this
  *	returns either the value actually sensed, or zero
@@ -59,6 +61,7 @@ struct module;
  */
 struct gpio_chip {
 	char			*label;
+	struct device		*dev;
 	struct module		*owner;
 
 	int			(*direction_input)(struct gpio_chip *chip,
@@ -74,6 +77,7 @@ struct gpio_chip {
 	int			base;
 	u16			ngpio;
 	unsigned		can_sleep:1;
+	unsigned		exported:1;
 };
 
 extern const char *gpiochip_is_requested(struct gpio_chip *chip,
@@ -108,7 +112,18 @@ extern void __gpio_set_value(unsigned gpio, int value);
 extern int __gpio_cansleep(unsigned gpio);
 
 
-#else
+#ifdef CONFIG_GPIO_SYSFS
+
+/*
+ * A sysfs interface can be exported by individual drivers if they want,
+ * but more typically is configured entirely from userspace.
+ */
+extern int gpio_export(unsigned gpio, bool direction_may_change);
+extern void gpio_unexport(unsigned gpio);
+
+#endif	/* CONFIG_GPIO_SYSFS */
+
+#else	/* !CONFIG_HAVE_GPIO_LIB */
 
 static inline int gpio_is_valid(int number)
 {
@@ -137,6 +152,20 @@ static inline void gpio_set_value_cansleep(unsigned gpio, int value)
 	gpio_set_value(gpio, value);
 }
 
-#endif
+#endif /* !CONFIG_HAVE_GPIO_LIB */
+
+#ifndef CONFIG_GPIO_SYSFS
+
+/* sysfs support is only available with gpiolib, where it's optional */
+
+static inline int gpio_export(unsigned gpio, bool direction_may_change)
+{
+	return -ENOSYS;
+}
+
+static inline void gpio_unexport(unsigned gpio)
+{
+}
+#endif	/* CONFIG_GPIO_SYSFS */
 
 #endif /* _ASM_GENERIC_GPIO_H */
diff --git a/include/linux/gpio.h b/include/linux/gpio.h
index 98be6c5762b9..730a20b83576 100644
--- a/include/linux/gpio.h
+++ b/include/linux/gpio.h
@@ -79,6 +79,19 @@ static inline void gpio_set_value_cansleep(unsigned gpio, int value)
 	WARN_ON(1);
 }
 
+static inline int gpio_export(unsigned gpio, bool direction_may_change)
+{
+	/* GPIO can never have been requested or set as {in,out}put */
+	WARN_ON(1);
+	return -EINVAL;
+}
+
+static inline void gpio_unexport(unsigned gpio)
+{
+	/* GPIO can never have been exported */
+	WARN_ON(1);
+}
+
 static inline int gpio_to_irq(unsigned gpio)
 {
 	/* GPIO can never have been requested or set as input */

From 8f1cc3b10e6ee0c5c7c8ed27f8771c4f252b4862 Mon Sep 17 00:00:00 2001
From: David Brownell <david-b@pacbell.net>
Date: Fri, 25 Jul 2008 01:46:09 -0700
Subject: [PATCH 0725/1171] gpio: mcp23s08 handles multiple chips per
 chipselect

Teach the mcp23s08 driver about a curious feature of these chips: up to
four of them can share the same chipselect, with the SPI signals wired in
parallel, by matching two bits in the first protocol byte against two
address lines on the chip.

This is handled by three software changes:

  * Platform data now holds an array of per-chip structs, not
    just one chip's address and pullup configuration.

  * Probe() and remove() now use another level of structure,
    wrapping an instance of the original structure for each
    mcp23s08 chip sharing that chipselect.

  * The HAEN bit is set, so that the hardware address bits can no
    longer be ignored (boot firmware may not have enabled them).

The "one struct per chip" preserves the guts of the current code,
but platform_data will need minor changes.

    OLD:
	/* incorrect "slave" ID may not have mattered */
	.slave = 3,
	.pullups = BIT(3) | BIT(1) | BIT(0),

    NEW:
	/* slave address _must_ match chip's wiring */
	.chip[3] = {
		.is_present = true,
		.pullups = BIT(3) | BIT(1) | BIT(0),
	},

There's no change in how things _behave_ for spi_device nodes with a
single mcp23s08 chip.  New multi-chip configurations assign GPIOs in
sequence, without holes.  The spi_device just resembles a bigger
controller, but internally it has multiple gpio_chip instances.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/gpio/mcp23s08.c      | 133 +++++++++++++++++++++++++++--------
 include/linux/spi/mcp23s08.h |  25 ++++---
 2 files changed, 118 insertions(+), 40 deletions(-)

diff --git a/drivers/gpio/mcp23s08.c b/drivers/gpio/mcp23s08.c
index 7efd7d3a81f9..8a1b405fefda 100644
--- a/drivers/gpio/mcp23s08.c
+++ b/drivers/gpio/mcp23s08.c
@@ -40,15 +40,26 @@ struct mcp23s08 {
 	struct spi_device	*spi;
 	u8			addr;
 
+	u8			cache[11];
 	/* lock protects the cached values */
 	struct mutex		lock;
-	u8			cache[11];
 
 	struct gpio_chip	chip;
 
 	struct work_struct	work;
 };
 
+/* A given spi_device can represent up to four mcp23s08 chips
+ * sharing the same chipselect but using different addresses
+ * (e.g. chips #0 and #3 might be populated, but not #1 or $2).
+ * Driver data holds all the per-chip data.
+ */
+struct mcp23s08_driver_data {
+	unsigned		ngpio;
+	struct mcp23s08		*mcp[4];
+	struct mcp23s08		chip[];
+};
+
 static int mcp23s08_read(struct mcp23s08 *mcp, unsigned reg)
 {
 	u8	tx[2], rx[1];
@@ -208,25 +219,18 @@ done:
 
 /*----------------------------------------------------------------------*/
 
-static int mcp23s08_probe(struct spi_device *spi)
+static int mcp23s08_probe_one(struct spi_device *spi, unsigned addr,
+		unsigned base, unsigned pullups)
 {
-	struct mcp23s08			*mcp;
-	struct mcp23s08_platform_data	*pdata;
+	struct mcp23s08_driver_data	*data = spi_get_drvdata(spi);
+	struct mcp23s08			*mcp = data->mcp[addr];
 	int				status;
 	int				do_update = 0;
 
-	pdata = spi->dev.platform_data;
-	if (!pdata || pdata->slave > 3 || !pdata->base)
-		return -ENODEV;
-
-	mcp = kzalloc(sizeof *mcp, GFP_KERNEL);
-	if (!mcp)
-		return -ENOMEM;
-
 	mutex_init(&mcp->lock);
 
 	mcp->spi = spi;
-	mcp->addr = 0x40 | (pdata->slave << 1);
+	mcp->addr = 0x40 | (addr << 1);
 
 	mcp->chip.label = "mcp23s08",
 
@@ -236,27 +240,28 @@ static int mcp23s08_probe(struct spi_device *spi)
 	mcp->chip.set = mcp23s08_set;
 	mcp->chip.dbg_show = mcp23s08_dbg_show;
 
-	mcp->chip.base = pdata->base;
+	mcp->chip.base = base;
 	mcp->chip.ngpio = 8;
 	mcp->chip.can_sleep = 1;
 	mcp->chip.dev = &spi->dev;
 	mcp->chip.owner = THIS_MODULE;
 
-	spi_set_drvdata(spi, mcp);
-
-	/* verify MCP_IOCON.SEQOP = 0, so sequential reads work */
+	/* verify MCP_IOCON.SEQOP = 0, so sequential reads work,
+	 * and MCP_IOCON.HAEN = 1, so we work with all chips.
+	 */
 	status = mcp23s08_read(mcp, MCP_IOCON);
 	if (status < 0)
 		goto fail;
-	if (status & IOCON_SEQOP) {
+	if ((status & IOCON_SEQOP) || !(status & IOCON_HAEN)) {
 		status &= ~IOCON_SEQOP;
+		status |= IOCON_HAEN;
 		status = mcp23s08_write(mcp, MCP_IOCON, (u8) status);
 		if (status < 0)
 			goto fail;
 	}
 
 	/* configure ~100K pullups */
-	status = mcp23s08_write(mcp, MCP_GPPU, pdata->pullups);
+	status = mcp23s08_write(mcp, MCP_GPPU, pullups);
 	if (status < 0)
 		goto fail;
 
@@ -283,11 +288,58 @@ static int mcp23s08_probe(struct spi_device *spi)
 		tx[1] = MCP_IPOL;
 		memcpy(&tx[2], &mcp->cache[MCP_IPOL], sizeof(tx) - 2);
 		status = spi_write_then_read(mcp->spi, tx, sizeof tx, NULL, 0);
-
-		/* FIXME check status... */
+		if (status < 0)
+			goto fail;
 	}
 
 	status = gpiochip_add(&mcp->chip);
+fail:
+	if (status < 0)
+		dev_dbg(&spi->dev, "can't setup chip %d, --> %d\n",
+				addr, status);
+	return status;
+}
+
+static int mcp23s08_probe(struct spi_device *spi)
+{
+	struct mcp23s08_platform_data	*pdata;
+	unsigned			addr;
+	unsigned			chips = 0;
+	struct mcp23s08_driver_data	*data;
+	int				status;
+	unsigned			base;
+
+	pdata = spi->dev.platform_data;
+	if (!pdata || !gpio_is_valid(pdata->base))
+		return -ENODEV;
+
+	for (addr = 0; addr < 4; addr++) {
+		if (!pdata->chip[addr].is_present)
+			continue;
+		chips++;
+	}
+	if (!chips)
+		return -ENODEV;
+
+	data = kzalloc(sizeof *data + chips * sizeof(struct mcp23s08),
+			GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
+	spi_set_drvdata(spi, data);
+
+	base = pdata->base;
+	for (addr = 0; addr < 4; addr++) {
+		if (!pdata->chip[addr].is_present)
+			continue;
+		chips--;
+		data->mcp[addr] = &data->chip[chips];
+		status = mcp23s08_probe_one(spi, addr, base,
+				pdata->chip[addr].pullups);
+		if (status < 0)
+			goto fail;
+		base += 8;
+	}
+	data->ngpio = base - pdata->base;
 
 	/* NOTE:  these chips have a relatively sane IRQ framework, with
 	 * per-signal masking and level/edge triggering.  It's not yet
@@ -295,8 +347,9 @@ static int mcp23s08_probe(struct spi_device *spi)
 	 */
 
 	if (pdata->setup) {
-		status = pdata->setup(spi, mcp->chip.base,
-				mcp->chip.ngpio, pdata->context);
+		status = pdata->setup(spi,
+				pdata->base, data->ngpio,
+				pdata->context);
 		if (status < 0)
 			dev_dbg(&spi->dev, "setup --> %d\n", status);
 	}
@@ -304,19 +357,29 @@ static int mcp23s08_probe(struct spi_device *spi)
 	return 0;
 
 fail:
-	kfree(mcp);
+	for (addr = 0; addr < 4; addr++) {
+		int tmp;
+
+		if (!data->mcp[addr])
+			continue;
+		tmp = gpiochip_remove(&data->mcp[addr]->chip);
+		if (tmp < 0)
+			dev_err(&spi->dev, "%s --> %d\n", "remove", tmp);
+	}
+	kfree(data);
 	return status;
 }
 
 static int mcp23s08_remove(struct spi_device *spi)
 {
-	struct mcp23s08			*mcp = spi_get_drvdata(spi);
+	struct mcp23s08_driver_data	*data = spi_get_drvdata(spi);
 	struct mcp23s08_platform_data	*pdata = spi->dev.platform_data;
+	unsigned			addr;
 	int				status = 0;
 
 	if (pdata->teardown) {
 		status = pdata->teardown(spi,
-				mcp->chip.base, mcp->chip.ngpio,
+				pdata->base, data->ngpio,
 				pdata->context);
 		if (status < 0) {
 			dev_err(&spi->dev, "%s --> %d\n", "teardown", status);
@@ -324,11 +387,20 @@ static int mcp23s08_remove(struct spi_device *spi)
 		}
 	}
 
-	status = gpiochip_remove(&mcp->chip);
+	for (addr = 0; addr < 4; addr++) {
+		int tmp;
+
+		if (!data->mcp[addr])
+			continue;
+
+		tmp = gpiochip_remove(&data->mcp[addr]->chip);
+		if (tmp < 0) {
+			dev_err(&spi->dev, "%s --> %d\n", "remove", tmp);
+			status = tmp;
+		}
+	}
 	if (status == 0)
-		kfree(mcp);
-	else
-		dev_err(&spi->dev, "%s --> %d\n", "remove", status);
+		kfree(data);
 	return status;
 }
 
@@ -356,4 +428,3 @@ static void __exit mcp23s08_exit(void)
 module_exit(mcp23s08_exit);
 
 MODULE_LICENSE("GPL");
-
diff --git a/include/linux/spi/mcp23s08.h b/include/linux/spi/mcp23s08.h
index 835ddf47d45c..22ef107d7704 100644
--- a/include/linux/spi/mcp23s08.h
+++ b/include/linux/spi/mcp23s08.h
@@ -1,18 +1,25 @@
 
-/* FIXME driver should be able to handle all four slaves that
- * can be hooked up to each chipselect, as well as IRQs...
- */
+/* FIXME driver should be able to handle IRQs...  */
+
+struct mcp23s08_chip_info {
+	bool	is_present;		/* true iff populated */
+	u8	pullups;		/* BIT(x) means enable pullup x */
+};
 
 struct mcp23s08_platform_data {
-	/* four slaves can share one SPI chipselect */
-	u8		slave;
+	/* Four slaves (numbered 0..3) can share one SPI chipselect, and
+	 * will provide 8..32 GPIOs using 1..4 gpio_chip instances.
+	 */
+	struct mcp23s08_chip_info	chip[4];
 
-	/* number assigned to the first GPIO */
+	/* "base" is the number of the first GPIO.  Dynamic assignment is
+	 * not currently supported, and even if there are gaps in chip
+	 * addressing the GPIO numbers are sequential .. so for example
+	 * if only slaves 0 and 3 are present, their GPIOs range from
+	 * base to base+15.
+	 */
 	unsigned	base;
 
-	/* pins with pullups */
-	u8		pullups;
-
 	void		*context;	/* param to setup/teardown */
 
 	int		(*setup)(struct spi_device *spi,

From ff1d5c2f0268f4e32103536e2e65480b5b7b6530 Mon Sep 17 00:00:00 2001
From: Michael Buesch <mb@bu3sch.de>
Date: Fri, 25 Jul 2008 01:46:10 -0700
Subject: [PATCH 0726/1171] gpio: add bt8xxgpio driver

This adds the bt8xxgpio driver.  The purpose of the bt8xxgpio driver is to
export all of the 24 GPIO pins available on Brooktree 8xx chips to the
kernel GPIO infrastructure.

This makes it possible to use a physically modified BT8xx card as
cheap digital GPIO card.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Michael Buesch <mb@bu3sch.de>
Cc: David Brownell <david-b@pacbell.net>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Mauro Carvalho Chehab <mchehab@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/bt8xxgpio.txt |  67 +++++++
 MAINTAINERS                 |   6 +
 drivers/gpio/Kconfig        |  18 ++
 drivers/gpio/Makefile       |   1 +
 drivers/gpio/bt8xxgpio.c    | 348 ++++++++++++++++++++++++++++++++++++
 5 files changed, 440 insertions(+)
 create mode 100644 Documentation/bt8xxgpio.txt
 create mode 100644 drivers/gpio/bt8xxgpio.c

diff --git a/Documentation/bt8xxgpio.txt b/Documentation/bt8xxgpio.txt
new file mode 100644
index 000000000000..d8297e4ebd26
--- /dev/null
+++ b/Documentation/bt8xxgpio.txt
@@ -0,0 +1,67 @@
+===============================================================
+==  BT8XXGPIO driver                                         ==
+==                                                           ==
+==  A driver for a selfmade cheap BT8xx based PCI GPIO-card  ==
+==                                                           ==
+==  For advanced documentation, see                          ==
+==  http://www.bu3sch.de/btgpio.php                          ==
+===============================================================
+
+
+A generic digital 24-port PCI GPIO card can be built out of an ordinary
+Brooktree bt848, bt849, bt878 or bt879 based analog TV tuner card. The
+Brooktree chip is used in old analog Hauppauge WinTV PCI cards. You can easily
+find them used for low prices on the net.
+
+The bt8xx chip does have 24 digital GPIO ports.
+These ports are accessible via 24 pins on the SMD chip package.
+
+
+==============================================
+==  How to physically access the GPIO pins  ==
+==============================================
+
+The are several ways to access these pins. One might unsolder the whole chip
+and put it on a custom PCI board, or one might only unsolder each individual
+GPIO pin and solder that to some tiny wire. As the chip package really is tiny
+there are some advanced soldering skills needed in any case.
+
+The physical pinouts are drawn in the following ASCII art.
+The GPIO pins are marked with G00-G23
+
+                                           G G G G G G G G G G G G     G G G G G G
+                                           0 0 0 0 0 0 0 0 0 0 1 1     1 1 1 1 1 1
+                                           0 1 2 3 4 5 6 7 8 9 0 1     2 3 4 5 6 7
+           | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
+           ---------------------------------------------------------------------------
+         --|                               ^                                     ^   |--
+         --|                               pin 86                           pin 67   |--
+         --|                                                                         |--
+         --|                                                               pin 61 >  |-- G18
+         --|                                                                         |-- G19
+         --|                                                                         |-- G20
+         --|                                                                         |-- G21
+         --|                                                                         |-- G22
+         --|                                                               pin 56 >  |-- G23
+         --|                                                                         |--
+         --|                           Brooktree 878/879                             |--
+         --|                                                                         |--
+         --|                                                                         |--
+         --|                                                                         |--
+         --|                                                                         |--
+         --|                                                                         |--
+         --|                                                                         |--
+         --|                                                                         |--
+         --|                                                                         |--
+         --|                                                                         |--
+         --|                                                                         |--
+         --|                                                                         |--
+         --|                                                                         |--
+         --|                                                                         |--
+         --|   O                                                                     |--
+         --|                                                                         |--
+           ---------------------------------------------------------------------------
+           | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | |
+           ^
+           This is pin 1
+
diff --git a/MAINTAINERS b/MAINTAINERS
index be05ef9b7b42..4cbf6016a9b9 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -1043,6 +1043,12 @@ M:	fujita.tomonori@lab.ntt.co.jp
 L:	linux-scsi@vger.kernel.org
 S:	Supported
 
+BT8XXGPIO DRIVER
+P:	Michael Buesch
+M:	mb@bu3sch.de
+W:	http://bu3sch.de/btgpio.php
+S:	Maintained
+
 BTTV VIDEO4LINUX DRIVER
 P:	Mauro Carvalho Chehab
 M:	mchehab@infradead.org
diff --git a/drivers/gpio/Kconfig b/drivers/gpio/Kconfig
index 6ec0e35b98e3..de202dbe5300 100644
--- a/drivers/gpio/Kconfig
+++ b/drivers/gpio/Kconfig
@@ -83,6 +83,24 @@ config GPIO_PCF857X
 	  This driver provides an in-kernel interface to those GPIOs using
 	  platform-neutral GPIO calls.
 
+comment "PCI GPIO expanders:"
+
+config GPIO_BT8XX
+	tristate "BT8XX GPIO abuser"
+	depends on PCI && VIDEO_BT848=n
+	help
+	  The BT8xx frame grabber chip has 24 GPIO pins than can be abused
+	  as a cheap PCI GPIO card.
+
+	  This chip can be found on Miro, Hauppauge and STB TV-cards.
+
+	  The card needs to be physically altered for using it as a
+	  GPIO card. For more information on how to build a GPIO card
+	  from a BT8xx TV card, see the documentation file at
+	  Documentation/bt8xxgpio.txt
+
+	  If unsure, say N.
+
 comment "SPI GPIO expanders:"
 
 config GPIO_MAX7301
diff --git a/drivers/gpio/Makefile b/drivers/gpio/Makefile
index 16e796dc5410..eeb2f2b20282 100644
--- a/drivers/gpio/Makefile
+++ b/drivers/gpio/Makefile
@@ -8,3 +8,4 @@ obj-$(CONFIG_GPIO_MAX7301)	+= max7301.o
 obj-$(CONFIG_GPIO_MCP23S08)	+= mcp23s08.o
 obj-$(CONFIG_GPIO_PCA953X)	+= pca953x.o
 obj-$(CONFIG_GPIO_PCF857X)	+= pcf857x.o
+obj-$(CONFIG_GPIO_BT8XX)	+= bt8xxgpio.o
diff --git a/drivers/gpio/bt8xxgpio.c b/drivers/gpio/bt8xxgpio.c
new file mode 100644
index 000000000000..7a1168249dd5
--- /dev/null
+++ b/drivers/gpio/bt8xxgpio.c
@@ -0,0 +1,348 @@
+/*
+
+    bt8xx GPIO abuser
+
+    Copyright (C) 2008 Michael Buesch <mb@bu3sch.de>
+
+    Please do _only_ contact the people listed _above_ with issues related to this driver.
+    All the other people listed below are not related to this driver. Their names
+    are only here, because this driver is derived from the bt848 driver.
+
+
+    Derived from the bt848 driver:
+
+    Copyright (C) 1996,97,98 Ralph  Metzler
+			   & Marcus Metzler
+    (c) 1999-2002 Gerd Knorr
+
+    some v4l2 code lines are taken from Justin's bttv2 driver which is
+    (c) 2000 Justin Schoeman
+
+    V4L1 removal from:
+    (c) 2005-2006 Nickolay V. Shmyrev
+
+    Fixes to be fully V4L2 compliant by
+    (c) 2006 Mauro Carvalho Chehab
+
+    Cropping and overscan support
+    Copyright (C) 2005, 2006 Michael H. Schimek
+    Sponsored by OPQ Systems AB
+
+    This program is free software; you can redistribute it and/or modify
+    it under the terms of the GNU General Public License as published by
+    the Free Software Foundation; either version 2 of the License, or
+    (at your option) any later version.
+
+    This program is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU General Public License for more details.
+
+    You should have received a copy of the GNU General Public License
+    along with this program; if not, write to the Free Software
+    Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+*/
+
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/spinlock.h>
+
+#include <asm/gpio.h>
+
+/* Steal the hardware definitions from the bttv driver. */
+#include "../media/video/bt8xx/bt848.h"
+
+
+#define BT8XXGPIO_NR_GPIOS		24 /* We have 24 GPIO pins */
+
+
+struct bt8xxgpio {
+	spinlock_t lock;
+
+	void __iomem *mmio;
+	struct pci_dev *pdev;
+	struct gpio_chip gpio;
+
+#ifdef CONFIG_PM
+	u32 saved_outen;
+	u32 saved_data;
+#endif
+};
+
+#define bgwrite(dat, adr)	writel((dat), bg->mmio+(adr))
+#define bgread(adr)		readl(bg->mmio+(adr))
+
+
+static int modparam_gpiobase = -1/* dynamic */;
+module_param_named(gpiobase, modparam_gpiobase, int, 0444);
+MODULE_PARM_DESC(gpiobase, "The GPIO number base. -1 means dynamic, which is the default.");
+
+
+static int bt8xxgpio_gpio_direction_input(struct gpio_chip *gpio, unsigned nr)
+{
+	struct bt8xxgpio *bg = container_of(gpio, struct bt8xxgpio, gpio);
+	unsigned long flags;
+	u32 outen, data;
+
+	spin_lock_irqsave(&bg->lock, flags);
+
+	data = bgread(BT848_GPIO_DATA);
+	data &= ~(1 << nr);
+	bgwrite(data, BT848_GPIO_DATA);
+
+	outen = bgread(BT848_GPIO_OUT_EN);
+	outen &= ~(1 << nr);
+	bgwrite(outen, BT848_GPIO_OUT_EN);
+
+	spin_unlock_irqrestore(&bg->lock, flags);
+
+	return 0;
+}
+
+static int bt8xxgpio_gpio_get(struct gpio_chip *gpio, unsigned nr)
+{
+	struct bt8xxgpio *bg = container_of(gpio, struct bt8xxgpio, gpio);
+	unsigned long flags;
+	u32 val;
+
+	spin_lock_irqsave(&bg->lock, flags);
+	val = bgread(BT848_GPIO_DATA);
+	spin_unlock_irqrestore(&bg->lock, flags);
+
+	return !!(val & (1 << nr));
+}
+
+static int bt8xxgpio_gpio_direction_output(struct gpio_chip *gpio,
+					unsigned nr, int val)
+{
+	struct bt8xxgpio *bg = container_of(gpio, struct bt8xxgpio, gpio);
+	unsigned long flags;
+	u32 outen, data;
+
+	spin_lock_irqsave(&bg->lock, flags);
+
+	outen = bgread(BT848_GPIO_OUT_EN);
+	outen |= (1 << nr);
+	bgwrite(outen, BT848_GPIO_OUT_EN);
+
+	data = bgread(BT848_GPIO_DATA);
+	if (val)
+		data |= (1 << nr);
+	else
+		data &= ~(1 << nr);
+	bgwrite(data, BT848_GPIO_DATA);
+
+	spin_unlock_irqrestore(&bg->lock, flags);
+
+	return 0;
+}
+
+static void bt8xxgpio_gpio_set(struct gpio_chip *gpio,
+			    unsigned nr, int val)
+{
+	struct bt8xxgpio *bg = container_of(gpio, struct bt8xxgpio, gpio);
+	unsigned long flags;
+	u32 data;
+
+	spin_lock_irqsave(&bg->lock, flags);
+
+	data = bgread(BT848_GPIO_DATA);
+	if (val)
+		data |= (1 << nr);
+	else
+		data &= ~(1 << nr);
+	bgwrite(data, BT848_GPIO_DATA);
+
+	spin_unlock_irqrestore(&bg->lock, flags);
+}
+
+static void bt8xxgpio_gpio_setup(struct bt8xxgpio *bg)
+{
+	struct gpio_chip *c = &bg->gpio;
+
+	c->label = bg->pdev->dev.bus_id;
+	c->owner = THIS_MODULE;
+	c->direction_input = bt8xxgpio_gpio_direction_input;
+	c->get = bt8xxgpio_gpio_get;
+	c->direction_output = bt8xxgpio_gpio_direction_output;
+	c->set = bt8xxgpio_gpio_set;
+	c->dbg_show = NULL;
+	c->base = modparam_gpiobase;
+	c->ngpio = BT8XXGPIO_NR_GPIOS;
+	c->can_sleep = 0;
+}
+
+static int bt8xxgpio_probe(struct pci_dev *dev,
+			const struct pci_device_id *pci_id)
+{
+	struct bt8xxgpio *bg;
+	int err;
+
+	bg = kzalloc(sizeof(*bg), GFP_KERNEL);
+	if (!bg)
+		return -ENOMEM;
+
+	bg->pdev = dev;
+	spin_lock_init(&bg->lock);
+
+	err = pci_enable_device(dev);
+	if (err) {
+		printk(KERN_ERR "bt8xxgpio: Can't enable device.\n");
+		goto err_freebg;
+	}
+	if (!request_mem_region(pci_resource_start(dev, 0),
+				pci_resource_len(dev, 0),
+				"bt8xxgpio")) {
+		printk(KERN_WARNING "bt8xxgpio: Can't request iomem (0x%llx).\n",
+		       (unsigned long long)pci_resource_start(dev, 0));
+		err = -EBUSY;
+		goto err_disable;
+	}
+	pci_set_master(dev);
+	pci_set_drvdata(dev, bg);
+
+	bg->mmio = ioremap(pci_resource_start(dev, 0), 0x1000);
+	if (!bg->mmio) {
+		printk(KERN_ERR "bt8xxgpio: ioremap() failed\n");
+		err = -EIO;
+		goto err_release_mem;
+	}
+
+	/* Disable interrupts */
+	bgwrite(0, BT848_INT_MASK);
+
+	/* gpio init */
+	bgwrite(0, BT848_GPIO_DMA_CTL);
+	bgwrite(0, BT848_GPIO_REG_INP);
+	bgwrite(0, BT848_GPIO_OUT_EN);
+
+	bt8xxgpio_gpio_setup(bg);
+	err = gpiochip_add(&bg->gpio);
+	if (err) {
+		printk(KERN_ERR "bt8xxgpio: Failed to register GPIOs\n");
+		goto err_release_mem;
+	}
+
+	printk(KERN_INFO "bt8xxgpio: Abusing BT8xx card for GPIOs %d to %d\n",
+	       bg->gpio.base, bg->gpio.base + BT8XXGPIO_NR_GPIOS - 1);
+
+	return 0;
+
+err_release_mem:
+	release_mem_region(pci_resource_start(dev, 0),
+			   pci_resource_len(dev, 0));
+	pci_set_drvdata(dev, NULL);
+err_disable:
+	pci_disable_device(dev);
+err_freebg:
+	kfree(bg);
+
+	return err;
+}
+
+static void bt8xxgpio_remove(struct pci_dev *pdev)
+{
+	struct bt8xxgpio *bg = pci_get_drvdata(pdev);
+
+	gpiochip_remove(&bg->gpio);
+
+	bgwrite(0, BT848_INT_MASK);
+	bgwrite(~0x0, BT848_INT_STAT);
+	bgwrite(0x0, BT848_GPIO_OUT_EN);
+
+	iounmap(bg->mmio);
+	release_mem_region(pci_resource_start(pdev, 0),
+			   pci_resource_len(pdev, 0));
+	pci_disable_device(pdev);
+
+	pci_set_drvdata(pdev, NULL);
+	kfree(bg);
+}
+
+#ifdef CONFIG_PM
+static int bt8xxgpio_suspend(struct pci_dev *pdev, pm_message_t state)
+{
+	struct bt8xxgpio *bg = pci_get_drvdata(pdev);
+	unsigned long flags;
+
+	spin_lock_irqsave(&bg->lock, flags);
+
+	bg->saved_outen = bgread(BT848_GPIO_OUT_EN);
+	bg->saved_data = bgread(BT848_GPIO_DATA);
+
+	bgwrite(0, BT848_INT_MASK);
+	bgwrite(~0x0, BT848_INT_STAT);
+	bgwrite(0x0, BT848_GPIO_OUT_EN);
+
+	spin_unlock_irqrestore(&bg->lock, flags);
+
+	pci_save_state(pdev);
+	pci_disable_device(pdev);
+	pci_set_power_state(pdev, pci_choose_state(pdev, state));
+
+	return 0;
+}
+
+static int bt8xxgpio_resume(struct pci_dev *pdev)
+{
+	struct bt8xxgpio *bg = pci_get_drvdata(pdev);
+	unsigned long flags;
+	int err;
+
+	pci_set_power_state(pdev, 0);
+	err = pci_enable_device(pdev);
+	if (err)
+		return err;
+	pci_restore_state(pdev);
+
+	spin_lock_irqsave(&bg->lock, flags);
+
+	bgwrite(0, BT848_INT_MASK);
+	bgwrite(0, BT848_GPIO_DMA_CTL);
+	bgwrite(0, BT848_GPIO_REG_INP);
+	bgwrite(bg->saved_outen, BT848_GPIO_OUT_EN);
+	bgwrite(bg->saved_data & bg->saved_outen,
+		BT848_GPIO_DATA);
+
+	spin_unlock_irqrestore(&bg->lock, flags);
+
+	return 0;
+}
+#else
+#define bt8xxgpio_suspend NULL
+#define bt8xxgpio_resume NULL
+#endif /* CONFIG_PM */
+
+static struct pci_device_id bt8xxgpio_pci_tbl[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_BROOKTREE, PCI_DEVICE_ID_BT848) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_BROOKTREE, PCI_DEVICE_ID_BT849) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_BROOKTREE, PCI_DEVICE_ID_BT878) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_BROOKTREE, PCI_DEVICE_ID_BT879) },
+	{ 0, },
+};
+MODULE_DEVICE_TABLE(pci, bt8xxgpio_pci_tbl);
+
+static struct pci_driver bt8xxgpio_pci_driver = {
+	.name		= "bt8xxgpio",
+	.id_table	= bt8xxgpio_pci_tbl,
+	.probe		= bt8xxgpio_probe,
+	.remove		= bt8xxgpio_remove,
+	.suspend	= bt8xxgpio_suspend,
+	.resume		= bt8xxgpio_resume,
+};
+
+static int bt8xxgpio_init(void)
+{
+	return pci_register_driver(&bt8xxgpio_pci_driver);
+}
+module_init(bt8xxgpio_init)
+
+static void bt8xxgpio_exit(void)
+{
+	pci_unregister_driver(&bt8xxgpio_pci_driver);
+}
+module_exit(bt8xxgpio_exit)
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Michael Buesch");
+MODULE_DESCRIPTION("Abuse a BT8xx framegrabber card as generic GPIO card");

From 7444a72effa632fcd8edc566f880d96fe213c73b Mon Sep 17 00:00:00 2001
From: Michael Buesch <mb@bu3sch.de>
Date: Fri, 25 Jul 2008 01:46:11 -0700
Subject: [PATCH 0727/1171] gpiolib: allow user-selection

This patch adds functionality to the gpio-lib subsystem to make it
possible to enable the gpio-lib code even if the architecture code didn't
request to get it built in.

The archtitecture code does still need to implement the gpiolib accessor
functions in its asm/gpio.h file.  This patch adds the implementations for
x86 and PPC.

With these changes it is possible to run generic GPIO expansion cards on
every architecture that implements the trivial wrapper functions.  Support
for more architectures can easily be added.

Signed-off-by: Michael Buesch <mb@bu3sch.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: David Brownell <david-b@pacbell.net>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Haavard Skinnemoen <hskinnemoen@atmel.com>
Cc: Jesper Nilsson <jesper.nilsson@axis.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Jean Delvare <khali@linux-fr.org>
Cc: Samuel Ortiz <sameo@openedhand.com>
Cc: Kumar Gala <galak@gate.crashing.org>
Cc: Sam Ravnborg <sam@ravnborg.org>
Cc: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/gpio.txt               | 12 +++++-
 arch/arm/Kconfig                     |  8 ++--
 arch/avr32/Kconfig                   |  2 +-
 arch/mips/Kconfig                    |  2 +-
 arch/powerpc/Kconfig                 |  1 +
 arch/powerpc/platforms/52xx/Kconfig  |  2 +-
 arch/powerpc/sysdev/qe_lib/Kconfig   |  2 +-
 arch/x86/Kconfig                     |  1 +
 drivers/Makefile                     |  2 +-
 drivers/gpio/Kconfig                 | 33 ++++++++++++++--
 drivers/gpio/Makefile                |  2 +-
 drivers/i2c/chips/Kconfig            |  2 +-
 drivers/mfd/Kconfig                  |  4 +-
 drivers/of/Kconfig                   |  2 +-
 include/asm-generic/gpio.h           |  2 +-
 include/asm-mips/mach-generic/gpio.h |  2 +-
 include/asm-powerpc/gpio.h           |  4 +-
 include/asm-x86/gpio.h               | 56 ++++++++++++++++++++++++++++
 18 files changed, 116 insertions(+), 23 deletions(-)

diff --git a/Documentation/gpio.txt b/Documentation/gpio.txt
index 8b69811a9642..18022e249c53 100644
--- a/Documentation/gpio.txt
+++ b/Documentation/gpio.txt
@@ -389,11 +389,21 @@ either NULL or the label associated with that GPIO when it was requested.
 
 Platform Support
 ----------------
-To support this framework, a platform's Kconfig will "select HAVE_GPIO_LIB"
+To support this framework, a platform's Kconfig will "select" either
+ARCH_REQUIRE_GPIOLIB or ARCH_WANT_OPTIONAL_GPIOLIB
 and arrange that its <asm/gpio.h> includes <asm-generic/gpio.h> and defines
 three functions: gpio_get_value(), gpio_set_value(), and gpio_cansleep().
 They may also want to provide a custom value for ARCH_NR_GPIOS.
 
+ARCH_REQUIRE_GPIOLIB means that the gpio-lib code will always get compiled
+into the kernel on that architecture.
+
+ARCH_WANT_OPTIONAL_GPIOLIB means the gpio-lib code defaults to off and the user
+can enable it and build it into the kernel optionally.
+
+If neither of these options are selected, the platform does not support
+GPIOs through GPIO-lib and the code cannot be enabled by the user.
+
 Trivial implementations of those functions can directly use framework
 code, which always dispatches through the gpio_chip:
 
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 6fb4f03369f2..dabb015aa40b 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -268,7 +268,7 @@ config ARCH_EP93XX
 	select GENERIC_GPIO
 	select HAVE_CLK
 	select HAVE_CLK
-	select HAVE_GPIO_LIB
+	select ARCH_REQUIRE_GPIOLIB
 	help
 	  This enables support for the Cirrus EP93xx series of CPUs.
 
@@ -447,7 +447,7 @@ config ARCH_PXA
 	select ARCH_MTD_XIP
 	select GENERIC_GPIO
 	select HAVE_CLK
-	select HAVE_GPIO_LIB
+	select ARCH_REQUIRE_GPIOLIB
 	select GENERIC_TIME
 	select GENERIC_CLOCKEVENTS
 	select TICK_ONESHOT
@@ -479,7 +479,7 @@ config ARCH_SA1100
 	select GENERIC_CLOCKEVENTS
 	select HAVE_CLK
 	select TICK_ONESHOT
-	select HAVE_GPIO_LIB
+	select ARCH_REQUIRE_GPIOLIB
 	help
 	  Support for StrongARM 11x0 based boards.
 
@@ -522,7 +522,7 @@ config ARCH_OMAP
 	bool "TI OMAP"
 	select GENERIC_GPIO
 	select HAVE_CLK
-	select HAVE_GPIO_LIB
+	select ARCH_REQUIRE_GPIOLIB
 	select GENERIC_TIME
 	select GENERIC_CLOCKEVENTS
 	help
diff --git a/arch/avr32/Kconfig b/arch/avr32/Kconfig
index df4adefedb42..7c239a916275 100644
--- a/arch/avr32/Kconfig
+++ b/arch/avr32/Kconfig
@@ -88,7 +88,7 @@ config PLATFORM_AT32AP
 	select SUBARCH_AVR32B
 	select MMU
 	select PERFORMANCE_COUNTERS
-	select HAVE_GPIO_LIB
+	select ARCH_REQUIRE_GPIOLIB
 	select GENERIC_ALLOCATOR
 
 #
diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index b9c754f4070c..b4c4eaa5dd26 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -713,7 +713,7 @@ config CSRC_SB1250
 
 config GPIO_TXX9
 	select GENERIC_GPIO
-	select HAVE_GPIO_LIB
+	select ARCH_REQUIRE_GPIOLIB
 	bool
 
 config CFE
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index de6b49cd6be8..fe88418167c5 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -110,6 +110,7 @@ config PPC
 	default y
 	select HAVE_DYNAMIC_FTRACE
 	select HAVE_FTRACE
+	select ARCH_WANT_OPTIONAL_GPIOLIB
 	select HAVE_IDE
 	select HAVE_IOREMAP_PROT
 	select HAVE_EFFICIENT_UNALIGNED_ACCESS
diff --git a/arch/powerpc/platforms/52xx/Kconfig b/arch/powerpc/platforms/52xx/Kconfig
index d664b1bce381..ccbd4958412e 100644
--- a/arch/powerpc/platforms/52xx/Kconfig
+++ b/arch/powerpc/platforms/52xx/Kconfig
@@ -48,6 +48,6 @@ config PPC_MPC5200_BUGFIX
 config PPC_MPC5200_GPIO
 	bool "MPC5200 GPIO support"
 	depends on PPC_MPC52xx
-	select HAVE_GPIO_LIB
+	select ARCH_REQUIRE_GPIOLIB
 	help
 	  Enable gpiolib support for mpc5200 based boards
diff --git a/arch/powerpc/sysdev/qe_lib/Kconfig b/arch/powerpc/sysdev/qe_lib/Kconfig
index 4bb18f57901e..1ce546462be5 100644
--- a/arch/powerpc/sysdev/qe_lib/Kconfig
+++ b/arch/powerpc/sysdev/qe_lib/Kconfig
@@ -29,7 +29,7 @@ config QE_GPIO
 	bool "QE GPIO support"
 	depends on QUICC_ENGINE
 	select GENERIC_GPIO
-	select HAVE_GPIO_LIB
+	select ARCH_REQUIRE_GPIOLIB
 	help
 	  Say Y here if you're going to use hardware that connects to the
 	  QE GPIOs.
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 66f3ab05b18c..e3cba0b45600 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -23,6 +23,7 @@ config X86
 	select HAVE_OPROFILE
 	select HAVE_IOREMAP_PROT
 	select HAVE_KPROBES
+	select ARCH_WANT_OPTIONAL_GPIOLIB if !X86_RDC321X
 	select HAVE_KRETPROBES
 	select HAVE_DYNAMIC_FTRACE
 	select HAVE_FTRACE
diff --git a/drivers/Makefile b/drivers/Makefile
index 808e0ae66aa8..54ec5e718c0e 100644
--- a/drivers/Makefile
+++ b/drivers/Makefile
@@ -5,7 +5,7 @@
 # Rewritten to use lists instead of if-statements.
 #
 
-obj-$(CONFIG_HAVE_GPIO_LIB)	+= gpio/
+obj-y				+= gpio/
 obj-$(CONFIG_PCI)		+= pci/
 obj-$(CONFIG_PARISC)		+= parisc/
 obj-$(CONFIG_RAPIDIO)		+= rapidio/
diff --git a/drivers/gpio/Kconfig b/drivers/gpio/Kconfig
index de202dbe5300..5a355f829167 100644
--- a/drivers/gpio/Kconfig
+++ b/drivers/gpio/Kconfig
@@ -2,15 +2,40 @@
 # GPIO infrastructure and expanders
 #
 
-config HAVE_GPIO_LIB
+config ARCH_WANT_OPTIONAL_GPIOLIB
 	bool
+	help
+	  Select this config option from the architecture Kconfig, if
+	  it is possible to use gpiolib on the architecture, but let the
+	  user decide whether to actually build it or not.
+	  Select this instead of ARCH_REQUIRE_GPIOLIB, if your architecture does
+	  not depend on GPIOs being available, but rather let the user
+	  decide whether he needs it or not.
+
+config ARCH_REQUIRE_GPIOLIB
+	bool
+	select GPIOLIB
 	help
 	  Platforms select gpiolib if they use this infrastructure
 	  for all their GPIOs, usually starting with ones integrated
 	  into SOC processors.
+	  Selecting this from the architecture code will cause the gpiolib
+	  code to always get built in.
 
-menu "GPIO Support"
-	depends on HAVE_GPIO_LIB
+
+
+menuconfig GPIOLIB
+	bool "GPIO Support"
+	depends on ARCH_WANT_OPTIONAL_GPIOLIB || ARCH_REQUIRE_GPIOLIB
+	select GENERIC_GPIO
+	help
+	  This enables GPIO support through the generic GPIO library.
+	  You only need to enable this, if you also want to enable
+	  one or more of the GPIO expansion card drivers below.
+
+	  If unsure, say N.
+
+if GPIOLIB
 
 config DEBUG_GPIO
 	bool "Debug GPIO calls"
@@ -116,4 +141,4 @@ config GPIO_MCP23S08
 	  SPI driver for Microchip MCP23S08 I/O expander.  This provides
 	  a GPIO interface supporting inputs and outputs.
 
-endmenu
+endif
diff --git a/drivers/gpio/Makefile b/drivers/gpio/Makefile
index eeb2f2b20282..8c45948d1fe7 100644
--- a/drivers/gpio/Makefile
+++ b/drivers/gpio/Makefile
@@ -2,7 +2,7 @@
 
 ccflags-$(CONFIG_DEBUG_GPIO)	+= -DDEBUG
 
-obj-$(CONFIG_HAVE_GPIO_LIB)	+= gpiolib.o
+obj-$(CONFIG_GPIOLIB)		+= gpiolib.o
 
 obj-$(CONFIG_GPIO_MAX7301)	+= max7301.o
 obj-$(CONFIG_GPIO_MCP23S08)	+= mcp23s08.o
diff --git a/drivers/i2c/chips/Kconfig b/drivers/i2c/chips/Kconfig
index 50e0a4653741..a95cb9465d65 100644
--- a/drivers/i2c/chips/Kconfig
+++ b/drivers/i2c/chips/Kconfig
@@ -126,7 +126,7 @@ config ISP1301_OMAP
 
 config TPS65010
 	tristate "TPS6501x Power Management chips"
-	depends on HAVE_GPIO_LIB
+	depends on GPIOLIB
 	default y if MACH_OMAP_H2 || MACH_OMAP_H3 || MACH_OMAP_OSK
 	help
 	  If you say yes here you get support for the TPS6501x series of
diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig
index bac9e973ece0..1f57a99fd968 100644
--- a/drivers/mfd/Kconfig
+++ b/drivers/mfd/Kconfig
@@ -36,7 +36,7 @@ config MFD_ASIC3
 
 config HTC_EGPIO
 	bool "HTC EGPIO support"
-	depends on GENERIC_HARDIRQS && HAVE_GPIO_LIB && ARM
+	depends on GENERIC_HARDIRQS && GPIOLIB && ARM
 	help
 	    This driver supports the CPLD egpio chip present on
 	    several HTC phones.  It provides basic support for input
@@ -52,7 +52,7 @@ config HTC_PASIC3
 
 config MFD_TC6393XB
 	bool "Support Toshiba TC6393XB"
-	depends on HAVE_GPIO_LIB
+	depends on GPIOLIB
 	select MFD_CORE
 	help
 	  Support for Toshiba Mobile IO Controller TC6393XB
diff --git a/drivers/of/Kconfig b/drivers/of/Kconfig
index 3a7a11a75fb4..1d7ec3129349 100644
--- a/drivers/of/Kconfig
+++ b/drivers/of/Kconfig
@@ -4,7 +4,7 @@ config OF_DEVICE
 
 config OF_GPIO
 	def_bool y
-	depends on OF && PPC_OF && HAVE_GPIO_LIB
+	depends on OF && PPC_OF && GPIOLIB
 	help
 	  OpenFirmware GPIO accessors
 
diff --git a/include/asm-generic/gpio.h b/include/asm-generic/gpio.h
index 1beff5166e53..a3034d20ebd5 100644
--- a/include/asm-generic/gpio.h
+++ b/include/asm-generic/gpio.h
@@ -3,7 +3,7 @@
 
 #include <linux/types.h>
 
-#ifdef CONFIG_HAVE_GPIO_LIB
+#ifdef CONFIG_GPIOLIB
 
 #include <linux/compiler.h>
 
diff --git a/include/asm-mips/mach-generic/gpio.h b/include/asm-mips/mach-generic/gpio.h
index e6b376bd9d06..b4e70208da64 100644
--- a/include/asm-mips/mach-generic/gpio.h
+++ b/include/asm-mips/mach-generic/gpio.h
@@ -1,7 +1,7 @@
 #ifndef __ASM_MACH_GENERIC_GPIO_H
 #define __ASM_MACH_GENERIC_GPIO_H
 
-#ifdef CONFIG_HAVE_GPIO_LIB
+#ifdef CONFIG_GPIOLIB
 #define gpio_get_value	__gpio_get_value
 #define gpio_set_value	__gpio_set_value
 #define gpio_cansleep	__gpio_cansleep
diff --git a/include/asm-powerpc/gpio.h b/include/asm-powerpc/gpio.h
index 77ad3a890f30..ea04632399d8 100644
--- a/include/asm-powerpc/gpio.h
+++ b/include/asm-powerpc/gpio.h
@@ -17,7 +17,7 @@
 #include <linux/errno.h>
 #include <asm-generic/gpio.h>
 
-#ifdef CONFIG_HAVE_GPIO_LIB
+#ifdef CONFIG_GPIOLIB
 
 /*
  * We don't (yet) implement inlined/rapid versions for on-chip gpios.
@@ -51,6 +51,6 @@ static inline int irq_to_gpio(unsigned int irq)
 	return -EINVAL;
 }
 
-#endif /* CONFIG_HAVE_GPIO_LIB */
+#endif /* CONFIG_GPIOLIB */
 
 #endif /* __ASM_POWERPC_GPIO_H */
diff --git a/include/asm-x86/gpio.h b/include/asm-x86/gpio.h
index ff87fca0caf9..116e9147fe66 100644
--- a/include/asm-x86/gpio.h
+++ b/include/asm-x86/gpio.h
@@ -1,6 +1,62 @@
+/*
+ * Generic GPIO API implementation for x86.
+ *
+ * Derived from the generic GPIO API for powerpc:
+ *
+ * Copyright (c) 2007-2008  MontaVista Software, Inc.
+ *
+ * Author: Anton Vorontsov <avorontsov@ru.mvista.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ */
+
 #ifndef _ASM_I386_GPIO_H
 #define _ASM_I386_GPIO_H
 
+#ifdef CONFIG_X86_RDC321X
 #include <gpio.h>
+#else /* CONFIG_X86_RDC321X */
+
+#include <asm-generic/gpio.h>
+
+#ifdef CONFIG_GPIOLIB
+
+/*
+ * Just call gpiolib.
+ */
+static inline int gpio_get_value(unsigned int gpio)
+{
+	return __gpio_get_value(gpio);
+}
+
+static inline void gpio_set_value(unsigned int gpio, int value)
+{
+	__gpio_set_value(gpio, value);
+}
+
+static inline int gpio_cansleep(unsigned int gpio)
+{
+	return __gpio_cansleep(gpio);
+}
+
+/*
+ * Not implemented, yet.
+ */
+static inline int gpio_to_irq(unsigned int gpio)
+{
+	return -ENOSYS;
+}
+
+static inline int irq_to_gpio(unsigned int irq)
+{
+	return -EINVAL;
+}
+
+#endif /* CONFIG_GPIOLIB */
+
+#endif /* CONFIG_X86_RDC321X */
 
 #endif /* _ASM_I386_GPIO_H */

From bbcd6d543de335bf81e96477f46a60a8bf51039c Mon Sep 17 00:00:00 2001
From: Eric Miao <eric.miao@marvell.com>
Date: Fri, 25 Jul 2008 01:46:14 -0700
Subject: [PATCH 0728/1171] gpio: max732x driver

This adds a driver supporting a family of I2C port expanders from Maxim,
which includes the MAX7319 and MAX7320-7327 chips.

[dbrownell@users.sourceforge.net: minor fixes]
Signed-off-by: Jack Ren <jack.ren@marvell.com>
Signed-off-by: Eric Miao <eric.miao@marvell.com>
Acked-by: Jean Delvare <khali@linux-fr.org>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/gpio/Kconfig        |  19 ++
 drivers/gpio/Makefile       |   1 +
 drivers/gpio/max732x.c      | 385 ++++++++++++++++++++++++++++++++++++
 include/linux/i2c/max732x.h |  19 ++
 4 files changed, 424 insertions(+)
 create mode 100644 drivers/gpio/max732x.c
 create mode 100644 include/linux/i2c/max732x.h

diff --git a/drivers/gpio/Kconfig b/drivers/gpio/Kconfig
index 5a355f829167..dbd42d6c93a7 100644
--- a/drivers/gpio/Kconfig
+++ b/drivers/gpio/Kconfig
@@ -67,6 +67,25 @@ config GPIO_SYSFS
 
 comment "I2C GPIO expanders:"
 
+config GPIO_MAX732X
+	tristate "MAX7319, MAX7320-7327 I2C Port Expanders"
+	depends on I2C
+	help
+	  Say yes here to support the MAX7319, MAX7320-7327 series of I2C
+	  Port Expanders. Each IO port on these chips has a fixed role of
+	  Input (designated by 'I'), Push-Pull Output ('O'), or Open-Drain
+	  Input and Output (designed by 'P'). The combinations are listed
+	  below:
+
+	  8 bits:	max7319 (8I), max7320 (8O), max7321 (8P),
+		  	max7322 (4I4O), max7323 (4P4O)
+
+	  16 bits:	max7324 (8I8O), max7325 (8P8O),
+		  	max7326 (4I12O), max7327 (4P12O)
+
+	  Board setup code must specify the model to use, and the start
+	  number for these GPIOs.
+
 config GPIO_PCA953X
 	tristate "PCA953x, PCA955x, and MAX7310 I/O ports"
 	depends on I2C
diff --git a/drivers/gpio/Makefile b/drivers/gpio/Makefile
index 8c45948d1fe7..01b4bbde1956 100644
--- a/drivers/gpio/Makefile
+++ b/drivers/gpio/Makefile
@@ -5,6 +5,7 @@ ccflags-$(CONFIG_DEBUG_GPIO)	+= -DDEBUG
 obj-$(CONFIG_GPIOLIB)		+= gpiolib.o
 
 obj-$(CONFIG_GPIO_MAX7301)	+= max7301.o
+obj-$(CONFIG_GPIO_MAX732X)	+= max732x.o
 obj-$(CONFIG_GPIO_MCP23S08)	+= mcp23s08.o
 obj-$(CONFIG_GPIO_PCA953X)	+= pca953x.o
 obj-$(CONFIG_GPIO_PCF857X)	+= pcf857x.o
diff --git a/drivers/gpio/max732x.c b/drivers/gpio/max732x.c
new file mode 100644
index 000000000000..b51c8135ca28
--- /dev/null
+++ b/drivers/gpio/max732x.c
@@ -0,0 +1,385 @@
+/*
+ *  max732x.c - I2C Port Expander with 8/16 I/O
+ *
+ *  Copyright (C) 2007 Marvell International Ltd.
+ *  Copyright (C) 2008 Jack Ren <jack.ren@marvell.com>
+ *  Copyright (C) 2008 Eric Miao <eric.miao@marvell.com>
+ *
+ *  Derived from drivers/gpio/pca953x.c
+ *
+ *  This program is free software; you can redistribute it and/or modify
+ *  it under the terms of the GNU General Public License as published by
+ *  the Free Software Foundation; version 2 of the License.
+ */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/slab.h>
+#include <linux/string.h>
+#include <linux/gpio.h>
+
+#include <linux/i2c.h>
+#include <linux/i2c/max732x.h>
+
+
+/*
+ * Each port of MAX732x (including MAX7319) falls into one of the
+ * following three types:
+ *
+ *   - Push Pull Output
+ *   - Input
+ *   - Open Drain I/O
+ *
+ * designated by 'O', 'I' and 'P' individually according to MAXIM's
+ * datasheets.
+ *
+ * There are two groups of I/O ports, each group usually includes
+ * up to 8 I/O ports, and is accessed by a specific I2C address:
+ *
+ *   - Group A : by I2C address 0b'110xxxx
+ *   - Group B : by I2C address 0b'101xxxx
+ *
+ * where 'xxxx' is decided by the connections of pin AD2/AD0.  The
+ * address used also affects the initial state of output signals.
+ *
+ * Within each group of ports, there are five known combinations of
+ * I/O ports: 4I4O, 4P4O, 8I, 8P, 8O, see the definitions below for
+ * the detailed organization of these ports.
+ *
+ * GPIO numbers start from 'gpio_base + 0' to 'gpio_base + 8/16',
+ * and GPIOs from GROUP_A are numbered before those from GROUP_B
+ * (if there are two groups).
+ *
+ * NOTE: MAX7328/MAX7329 are drop-in replacements for PCF8574/a, so
+ * they are not supported by this driver.
+ */
+
+#define PORT_NONE	0x0	/* '/' No Port */
+#define PORT_OUTPUT	0x1	/* 'O' Push-Pull, Output Only */
+#define PORT_INPUT	0x2	/* 'I' Input Only */
+#define PORT_OPENDRAIN	0x3	/* 'P' Open-Drain, I/O */
+
+#define IO_4I4O		0x5AA5	/* O7 O6 I5 I4 I3 I2 O1 O0 */
+#define IO_4P4O		0x5FF5	/* O7 O6 P5 P4 P3 P2 O1 O0 */
+#define IO_8I		0xAAAA	/* I7 I6 I5 I4 I3 I2 I1 I0 */
+#define IO_8P		0xFFFF	/* P7 P6 P5 P4 P3 P2 P1 P0 */
+#define IO_8O		0x5555	/* O7 O6 O5 O4 O3 O2 O1 O0 */
+
+#define GROUP_A(x)	((x) & 0xffff)	/* I2C Addr: 0b'110xxxx */
+#define GROUP_B(x)	((x) << 16)	/* I2C Addr: 0b'101xxxx */
+
+static const struct i2c_device_id max732x_id[] = {
+	{ "max7319", GROUP_A(IO_8I) },
+	{ "max7320", GROUP_B(IO_8O) },
+	{ "max7321", GROUP_A(IO_8P) },
+	{ "max7322", GROUP_A(IO_4I4O) },
+	{ "max7323", GROUP_A(IO_4P4O) },
+	{ "max7324", GROUP_A(IO_8I) | GROUP_B(IO_8O) },
+	{ "max7325", GROUP_A(IO_8P) | GROUP_B(IO_8O) },
+	{ "max7326", GROUP_A(IO_4I4O) | GROUP_B(IO_8O) },
+	{ "max7327", GROUP_A(IO_4P4O) | GROUP_B(IO_8O) },
+	{ },
+};
+MODULE_DEVICE_TABLE(i2c, max732x_id);
+
+struct max732x_chip {
+	struct gpio_chip gpio_chip;
+
+	struct i2c_client *client;	/* "main" client */
+	struct i2c_client *client_dummy;
+	struct i2c_client *client_group_a;
+	struct i2c_client *client_group_b;
+
+	unsigned int	mask_group_a;
+	unsigned int	dir_input;
+	unsigned int	dir_output;
+
+	struct mutex	lock;
+	uint8_t		reg_out[2];
+};
+
+static int max732x_write(struct max732x_chip *chip, int group_a, uint8_t val)
+{
+	struct i2c_client *client;
+	int ret;
+
+	client = group_a ? chip->client_group_a : chip->client_group_b;
+	ret = i2c_smbus_write_byte(client, val);
+	if (ret < 0) {
+		dev_err(&client->dev, "failed writing\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static int max732x_read(struct max732x_chip *chip, int group_a, uint8_t *val)
+{
+	struct i2c_client *client;
+	int ret;
+
+	client = group_a ? chip->client_group_a : chip->client_group_b;
+	ret = i2c_smbus_read_byte(client);
+	if (ret < 0) {
+		dev_err(&client->dev, "failed reading\n");
+		return ret;
+	}
+
+	*val = (uint8_t)ret;
+	return 0;
+}
+
+static inline int is_group_a(struct max732x_chip *chip, unsigned off)
+{
+	return (1u << off) & chip->mask_group_a;
+}
+
+static int max732x_gpio_get_value(struct gpio_chip *gc, unsigned off)
+{
+	struct max732x_chip *chip;
+	uint8_t reg_val;
+	int ret;
+
+	chip = container_of(gc, struct max732x_chip, gpio_chip);
+
+	ret = max732x_read(chip, is_group_a(chip, off), &reg_val);
+	if (ret < 0)
+		return 0;
+
+	return reg_val & (1u << (off & 0x7));
+}
+
+static void max732x_gpio_set_value(struct gpio_chip *gc, unsigned off, int val)
+{
+	struct max732x_chip *chip;
+	uint8_t reg_out, mask = 1u << (off & 0x7);
+	int ret;
+
+	chip = container_of(gc, struct max732x_chip, gpio_chip);
+
+	mutex_lock(&chip->lock);
+
+	reg_out = (off > 7) ? chip->reg_out[1] : chip->reg_out[0];
+	reg_out = (val) ? reg_out | mask : reg_out & ~mask;
+
+	ret = max732x_write(chip, is_group_a(chip, off), reg_out);
+	if (ret < 0)
+		goto out;
+
+	/* update the shadow register then */
+	if (off > 7)
+		chip->reg_out[1] = reg_out;
+	else
+		chip->reg_out[0] = reg_out;
+out:
+	mutex_unlock(&chip->lock);
+}
+
+static int max732x_gpio_direction_input(struct gpio_chip *gc, unsigned off)
+{
+	struct max732x_chip *chip;
+	unsigned int mask = 1u << off;
+
+	chip = container_of(gc, struct max732x_chip, gpio_chip);
+
+	if ((mask & chip->dir_input) == 0) {
+		dev_dbg(&chip->client->dev, "%s port %d is output only\n",
+			chip->client->name, off);
+		return -EACCES;
+	}
+
+	return 0;
+}
+
+static int max732x_gpio_direction_output(struct gpio_chip *gc,
+		unsigned off, int val)
+{
+	struct max732x_chip *chip;
+	unsigned int mask = 1u << off;
+
+	chip = container_of(gc, struct max732x_chip, gpio_chip);
+
+	if ((mask & chip->dir_output) == 0) {
+		dev_dbg(&chip->client->dev, "%s port %d is input only\n",
+			chip->client->name, off);
+		return -EACCES;
+	}
+
+	max732x_gpio_set_value(gc, off, val);
+	return 0;
+}
+
+static int __devinit max732x_setup_gpio(struct max732x_chip *chip,
+					const struct i2c_device_id *id,
+					unsigned gpio_start)
+{
+	struct gpio_chip *gc = &chip->gpio_chip;
+	uint32_t id_data = id->driver_data;
+	int i, port = 0;
+
+	for (i = 0; i < 16; i++, id_data >>= 2) {
+		unsigned int mask = 1 << port;
+
+		switch (id_data & 0x3) {
+		case PORT_OUTPUT:
+			chip->dir_output |= mask;
+			break;
+		case PORT_INPUT:
+			chip->dir_input |= mask;
+			break;
+		case PORT_OPENDRAIN:
+			chip->dir_output |= mask;
+			chip->dir_input |= mask;
+			break;
+		default:
+			continue;
+		}
+
+		if (i < 8)
+			chip->mask_group_a |= mask;
+		port++;
+	}
+
+	if (chip->dir_input)
+		gc->direction_input = max732x_gpio_direction_input;
+	if (chip->dir_output) {
+		gc->direction_output = max732x_gpio_direction_output;
+		gc->set = max732x_gpio_set_value;
+	}
+	gc->get = max732x_gpio_get_value;
+	gc->can_sleep = 1;
+
+	gc->base = gpio_start;
+	gc->ngpio = port;
+	gc->label = chip->client->name;
+	gc->owner = THIS_MODULE;
+
+	return port;
+}
+
+static int __devinit max732x_probe(struct i2c_client *client,
+				   const struct i2c_device_id *id)
+{
+	struct max732x_platform_data *pdata;
+	struct max732x_chip *chip;
+	struct i2c_client *c;
+	uint16_t addr_a, addr_b;
+	int ret, nr_port;
+
+	pdata = client->dev.platform_data;
+	if (pdata == NULL)
+		return -ENODEV;
+
+	chip = kzalloc(sizeof(struct max732x_chip), GFP_KERNEL);
+	if (chip == NULL)
+		return -ENOMEM;
+	chip->client = client;
+
+	nr_port = max732x_setup_gpio(chip, id, pdata->gpio_base);
+
+	addr_a = (client->addr & 0x0f) | 0x60;
+	addr_b = (client->addr & 0x0f) | 0x50;
+
+	switch (client->addr & 0x70) {
+	case 0x60:
+		chip->client_group_a = client;
+		if (nr_port > 7) {
+			c = i2c_new_dummy(client->adapter, addr_b);
+			chip->client_group_b = chip->client_dummy = c;
+		}
+		break;
+	case 0x50:
+		chip->client_group_b = client;
+		if (nr_port > 7) {
+			c = i2c_new_dummy(client->adapter, addr_a);
+			chip->client_group_a = chip->client_dummy = c;
+		}
+		break;
+	default:
+		dev_err(&client->dev, "invalid I2C address specified %02x\n",
+				client->addr);
+		ret = -EINVAL;
+		goto out_failed;
+	}
+
+	mutex_init(&chip->lock);
+
+	max732x_read(chip, is_group_a(chip, 0), &chip->reg_out[0]);
+	if (nr_port > 7)
+		max732x_read(chip, is_group_a(chip, 8), &chip->reg_out[1]);
+
+	ret = gpiochip_add(&chip->gpio_chip);
+	if (ret)
+		goto out_failed;
+
+	if (pdata->setup) {
+		ret = pdata->setup(client, chip->gpio_chip.base,
+				chip->gpio_chip.ngpio, pdata->context);
+		if (ret < 0)
+			dev_warn(&client->dev, "setup failed, %d\n", ret);
+	}
+
+	i2c_set_clientdata(client, chip);
+	return 0;
+
+out_failed:
+	kfree(chip);
+	return ret;
+}
+
+static int __devexit max732x_remove(struct i2c_client *client)
+{
+	struct max732x_platform_data *pdata = client->dev.platform_data;
+	struct max732x_chip *chip = i2c_get_clientdata(client);
+	int ret;
+
+	if (pdata->teardown) {
+		ret = pdata->teardown(client, chip->gpio_chip.base,
+				chip->gpio_chip.ngpio, pdata->context);
+		if (ret < 0) {
+			dev_err(&client->dev, "%s failed, %d\n",
+					"teardown", ret);
+			return ret;
+		}
+	}
+
+	ret = gpiochip_remove(&chip->gpio_chip);
+	if (ret) {
+		dev_err(&client->dev, "%s failed, %d\n",
+				"gpiochip_remove()", ret);
+		return ret;
+	}
+
+	/* unregister any dummy i2c_client */
+	if (chip->client_dummy)
+		i2c_unregister_device(chip->client_dummy);
+
+	kfree(chip);
+	return 0;
+}
+
+static struct i2c_driver max732x_driver = {
+	.driver = {
+		.name	= "max732x",
+		.owner	= THIS_MODULE,
+	},
+	.probe		= max732x_probe,
+	.remove		= __devexit_p(max732x_remove),
+	.id_table	= max732x_id,
+};
+
+static int __init max732x_init(void)
+{
+	return i2c_add_driver(&max732x_driver);
+}
+module_init(max732x_init);
+
+static void __exit max732x_exit(void)
+{
+	i2c_del_driver(&max732x_driver);
+}
+module_exit(max732x_exit);
+
+MODULE_AUTHOR("Eric Miao <eric.miao@marvell.com>");
+MODULE_DESCRIPTION("GPIO expander driver for MAX732X");
+MODULE_LICENSE("GPL");
diff --git a/include/linux/i2c/max732x.h b/include/linux/i2c/max732x.h
new file mode 100644
index 000000000000..e10336631c62
--- /dev/null
+++ b/include/linux/i2c/max732x.h
@@ -0,0 +1,19 @@
+#ifndef __LINUX_I2C_MAX732X_H
+#define __LINUX_I2C_MAX732X_H
+
+/* platform data for the MAX732x 8/16-bit I/O expander driver */
+
+struct max732x_platform_data {
+	/* number of the first GPIO */
+	unsigned	gpio_base;
+
+	void		*context;	/* param to setup/teardown */
+
+	int		(*setup)(struct i2c_client *client,
+				unsigned gpio, unsigned ngpio,
+				void *context);
+	int		(*teardown)(struct i2c_client *client,
+				unsigned gpio, unsigned ngpio,
+				void *context);
+};
+#endif /* __LINUX_I2C_MAX732X_H */

From fb523f32275344282f20ef3352cbf03e599241e6 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 01:46:14 -0700
Subject: [PATCH 0729/1171] minix: remove !NO_TRUNCATE code

This patch removes the !NO_TRUNCATE code that anyway required a manual
editing of the code for being used.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/minix/inode.c |  3 ---
 fs/minix/minix.h |  6 ------
 fs/minix/namei.c | 24 ------------------------
 3 files changed, 33 deletions(-)

diff --git a/fs/minix/inode.c b/fs/minix/inode.c
index 84f6242ba6fc..523d73713418 100644
--- a/fs/minix/inode.c
+++ b/fs/minix/inode.c
@@ -256,9 +256,6 @@ static int minix_fill_super(struct super_block *s, void *data, int silent)
 	if (!s->s_root)
 		goto out_iput;
 
-	if (!NO_TRUNCATE)
-		s->s_root->d_op = &minix_dentry_operations;
-
 	if (!(s->s_flags & MS_RDONLY)) {
 		if (sbi->s_version != MINIX_V3) /* s_state is now out from V3 sb */
 			ms->s_state &= ~MINIX_VALID_FS;
diff --git a/fs/minix/minix.h b/fs/minix/minix.h
index 326edfe96108..e6a0b193bea4 100644
--- a/fs/minix/minix.h
+++ b/fs/minix/minix.h
@@ -2,11 +2,6 @@
 #include <linux/pagemap.h>
 #include <linux/minix_fs.h>
 
-/*
- * change the define below to 0 if you want names > info->s_namelen chars to be
- * truncated. Else they will be disallowed (ENAMETOOLONG).
- */
-#define NO_TRUNCATE 1
 #define INODE_VERSION(inode)	minix_sb(inode->i_sb)->s_version
 #define MINIX_V1		0x0001		/* original minix fs */
 #define MINIX_V2		0x0002		/* minix V2 fs */
@@ -83,7 +78,6 @@ extern const struct inode_operations minix_file_inode_operations;
 extern const struct inode_operations minix_dir_inode_operations;
 extern const struct file_operations minix_file_operations;
 extern const struct file_operations minix_dir_operations;
-extern struct dentry_operations minix_dentry_operations;
 
 static inline struct minix_sb_info *minix_sb(struct super_block *sb)
 {
diff --git a/fs/minix/namei.c b/fs/minix/namei.c
index 102241bc9c79..32b131cd6121 100644
--- a/fs/minix/namei.c
+++ b/fs/minix/namei.c
@@ -18,30 +18,6 @@ static int add_nondir(struct dentry *dentry, struct inode *inode)
 	return err;
 }
 
-static int minix_hash(struct dentry *dentry, struct qstr *qstr)
-{
-	unsigned long hash;
-	int i;
-	const unsigned char *name;
-
-	i = minix_sb(dentry->d_inode->i_sb)->s_namelen;
-	if (i >= qstr->len)
-		return 0;
-	/* Truncate the name in place, avoids having to define a compare
-	   function. */
-	qstr->len = i;
-	name = qstr->name;
-	hash = init_name_hash();
-	while (i--)
-		hash = partial_name_hash(*name++, hash);
-	qstr->hash = end_name_hash(hash);
-	return 0;
-}
-
-struct dentry_operations minix_dentry_operations = {
-	.d_hash		= minix_hash,
-};
-
 static struct dentry *minix_lookup(struct inode * dir, struct dentry *dentry, struct nameidata *nd)
 {
 	struct inode * inode = NULL;

From f905f06fca5d3949eca12f5a43e251a404b3470a Mon Sep 17 00:00:00 2001
From: Shen Feng <shen@cn.fujitsu.com>
Date: Fri, 25 Jul 2008 01:46:15 -0700
Subject: [PATCH 0730/1171] ext2: remove double definitions of xattr macros

remove the definitions of macros:
XATTR_TRUSTED_PREFIX
XATTR_USER_PREFIX
since they are defined in linux/xattr.h

Signed-off-by: Shen Feng <shen@cn.fujitsu.com>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ext2/xattr_security.c | 2 +-
 fs/ext2/xattr_trusted.c  | 4 +---
 fs/ext2/xattr_user.c     | 4 +---
 3 files changed, 3 insertions(+), 7 deletions(-)

diff --git a/fs/ext2/xattr_security.c b/fs/ext2/xattr_security.c
index eaa23d2d5213..70c0dbdcdcb7 100644
--- a/fs/ext2/xattr_security.c
+++ b/fs/ext2/xattr_security.c
@@ -14,7 +14,7 @@ static size_t
 ext2_xattr_security_list(struct inode *inode, char *list, size_t list_size,
 			 const char *name, size_t name_len)
 {
-	const int prefix_len = sizeof(XATTR_SECURITY_PREFIX)-1;
+	const int prefix_len = XATTR_SECURITY_PREFIX_LEN;
 	const size_t total_len = prefix_len + name_len + 1;
 
 	if (list && total_len <= list_size) {
diff --git a/fs/ext2/xattr_trusted.c b/fs/ext2/xattr_trusted.c
index 83ee149f353d..e8219f8eae9f 100644
--- a/fs/ext2/xattr_trusted.c
+++ b/fs/ext2/xattr_trusted.c
@@ -12,13 +12,11 @@
 #include <linux/ext2_fs.h>
 #include "xattr.h"
 
-#define XATTR_TRUSTED_PREFIX "trusted."
-
 static size_t
 ext2_xattr_trusted_list(struct inode *inode, char *list, size_t list_size,
 			const char *name, size_t name_len)
 {
-	const int prefix_len = sizeof(XATTR_TRUSTED_PREFIX)-1;
+	const int prefix_len = XATTR_TRUSTED_PREFIX_LEN;
 	const size_t total_len = prefix_len + name_len + 1;
 
 	if (!capable(CAP_SYS_ADMIN))
diff --git a/fs/ext2/xattr_user.c b/fs/ext2/xattr_user.c
index f383e7c3a7b5..92495d28c62f 100644
--- a/fs/ext2/xattr_user.c
+++ b/fs/ext2/xattr_user.c
@@ -11,13 +11,11 @@
 #include "ext2.h"
 #include "xattr.h"
 
-#define XATTR_USER_PREFIX "user."
-
 static size_t
 ext2_xattr_user_list(struct inode *inode, char *list, size_t list_size,
 		     const char *name, size_t name_len)
 {
-	const size_t prefix_len = sizeof(XATTR_USER_PREFIX)-1;
+	const size_t prefix_len = XATTR_USER_PREFIX_LEN;
 	const size_t total_len = prefix_len + name_len + 1;
 
 	if (!test_opt(inode->i_sb, XATTR_USER))

From 50c33a84db4aa5082e3af8d873b22344ae2ebea8 Mon Sep 17 00:00:00 2001
From: Samuel Thibault <samuel.thibault@ens-lyon.org>
Date: Fri, 25 Jul 2008 01:46:16 -0700
Subject: [PATCH 0731/1171] ext2: fix typo in Hurd part of
 include/linux/ext2_fs.h

Fix typo in Hurd part of include/linux/ext2_fs.h

The ';' here is redundant or can even pose problem.  This is actually not
used by the Linux kernel, but it is exposed in GNU/Hurd.

Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/ext2_fs.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/linux/ext2_fs.h b/include/linux/ext2_fs.h
index 84cec2aa9f1e..2efe7b863cff 100644
--- a/include/linux/ext2_fs.h
+++ b/include/linux/ext2_fs.h
@@ -284,8 +284,8 @@ struct ext2_inode {
 
 #ifdef	__hurd__
 #define i_translator	osd1.hurd1.h_i_translator
-#define i_frag		osd2.hurd2.h_i_frag;
-#define i_fsize		osd2.hurd2.h_i_fsize;
+#define i_frag		osd2.hurd2.h_i_frag
+#define i_fsize		osd2.hurd2.h_i_fsize
 #define i_uid_high	osd2.hurd2.h_i_uid_high
 #define i_gid_high	osd2.hurd2.h_i_gid_high
 #define i_author	osd2.hurd2.h_i_author

From 9cfe7b9010aa66da5f3b2bc33d9e30a4d53bd274 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Fri, 25 Jul 2008 01:46:16 -0700
Subject: [PATCH 0732/1171] ext3: fix synchronization of quota files in
 journal=data mode

In journal=data mode, it is not enough to do write_inode_now as done in
vfs_quota_on() to write all data to their final location (which is needed for
quota_read to work correctly).  Calling journal_flush() does its job.

Reported-by: Nick <gentuu@gmail.com>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ext3/super.c | 35 +++++++++++++++++++++++++++--------
 1 file changed, 27 insertions(+), 8 deletions(-)

diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 2845425077e8..50796e90d070 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -2759,23 +2759,42 @@ static int ext3_quota_on(struct super_block *sb, int type, int format_id,
 
 	if (!test_opt(sb, QUOTA))
 		return -EINVAL;
-	/* Not journalling quota or remount? */
-	if ((!EXT3_SB(sb)->s_qf_names[USRQUOTA] &&
-	    !EXT3_SB(sb)->s_qf_names[GRPQUOTA]) || remount)
+	/* When remounting, no checks are needed and in fact, path is NULL */
+	if (remount)
 		return vfs_quota_on(sb, type, format_id, path, remount);
+
 	err = path_lookup(path, LOOKUP_FOLLOW, &nd);
 	if (err)
 		return err;
+
 	/* Quotafile not on the same filesystem? */
 	if (nd.path.mnt->mnt_sb != sb) {
 		path_put(&nd.path);
 		return -EXDEV;
 	}
-	/* Quotafile not in fs root? */
-	if (nd.path.dentry->d_parent->d_inode != sb->s_root->d_inode)
-		printk(KERN_WARNING
-			"EXT3-fs: Quota file not on filesystem root. "
-			"Journalled quota will not work.\n");
+	/* Journaling quota? */
+	if (EXT3_SB(sb)->s_qf_names[type]) {
+		/* Quotafile not of fs root? */
+		if (nd.path.dentry->d_parent->d_inode != sb->s_root->d_inode)
+			printk(KERN_WARNING
+				"EXT3-fs: Quota file not on filesystem root. "
+				"Journaled quota will not work.\n");
+	}
+
+	/*
+	 * When we journal data on quota file, we have to flush journal to see
+	 * all updates to the file when we bypass pagecache...
+	 */
+	if (ext3_should_journal_data(nd.path.dentry->d_inode)) {
+		/*
+		 * We don't need to lock updates but journal_flush() could
+		 * otherwise be livelocked...
+		 */
+		journal_lock_updates(EXT3_SB(sb)->s_journal);
+		journal_flush(EXT3_SB(sb)->s_journal);
+		journal_unlock_updates(EXT3_SB(sb)->s_journal);
+	}
+
 	path_put(&nd.path);
 	return vfs_quota_on(sb, type, format_id, path, remount);
 }

From 99aeaf639f61ab6be1967e5f92e2e28dafad8383 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Fri, 25 Jul 2008 01:46:17 -0700
Subject: [PATCH 0733/1171] ext3: fix typos in messages and comments
 (journalled -> journaled)

Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ext3/super.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 50796e90d070..0a1bf82845c3 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -1020,7 +1020,7 @@ static int parse_options (char *options, struct super_block *sb,
 set_qf_name:
 			if (sb_any_quota_enabled(sb)) {
 				printk(KERN_ERR
-					"EXT3-fs: Cannot change journalled "
+					"EXT3-fs: Cannot change journaled "
 					"quota options when quota turned on.\n");
 				return 0;
 			}
@@ -1058,7 +1058,7 @@ set_qf_name:
 clear_qf_name:
 			if (sb_any_quota_enabled(sb)) {
 				printk(KERN_ERR "EXT3-fs: Cannot change "
-					"journalled quota options when "
+					"journaled quota options when "
 					"quota turned on.\n");
 				return 0;
 			}
@@ -1169,14 +1169,14 @@ clear_qf_name:
 		}
 
 		if (!sbi->s_jquota_fmt) {
-			printk(KERN_ERR "EXT3-fs: journalled quota format "
+			printk(KERN_ERR "EXT3-fs: journaled quota format "
 					"not specified.\n");
 			return 0;
 		}
 	} else {
 		if (sbi->s_jquota_fmt) {
-			printk(KERN_ERR "EXT3-fs: journalled quota format "
-					"specified with no journalling "
+			printk(KERN_ERR "EXT3-fs: journaled quota format "
+					"specified with no journaling "
 					"enabled.\n");
 			return 0;
 		}
@@ -1370,7 +1370,7 @@ static void ext3_orphan_cleanup (struct super_block * sb,
 			int ret = ext3_quota_on_mount(sb, i);
 			if (ret < 0)
 				printk(KERN_ERR
-					"EXT3-fs: Cannot turn on journalled "
+					"EXT3-fs: Cannot turn on journaled "
 					"quota: error %d\n", ret);
 		}
 	}
@@ -2712,7 +2712,7 @@ static int ext3_release_dquot(struct dquot *dquot)
 
 static int ext3_mark_dquot_dirty(struct dquot *dquot)
 {
-	/* Are we journalling quotas? */
+	/* Are we journaling quotas? */
 	if (EXT3_SB(dquot->dq_sb)->s_qf_names[USRQUOTA] ||
 	    EXT3_SB(dquot->dq_sb)->s_qf_names[GRPQUOTA]) {
 		dquot_mark_dquot_dirty(dquot);

From d06bf1d252fe16f5f0d13e04da7a9913420aa1cf Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Fri, 25 Jul 2008 01:46:18 -0700
Subject: [PATCH 0734/1171] ext3: correct mount option parsing to detect when
 quota options can be changed

We should not allow user to change quota mount options when quota is just
suspended.  I would make mount options and internal quota state inconsistent.
Also we should not allow user to change quota format when quota is turned on.
On the other hand we can just silently ignore when some option is set to the
value it already has (mount does this on remount).

Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ext3/super.c | 29 ++++++++++++++++++++++-------
 1 file changed, 22 insertions(+), 7 deletions(-)

diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 0a1bf82845c3..615788c6843a 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -842,7 +842,7 @@ static int parse_options (char *options, struct super_block *sb,
 	int data_opt = 0;
 	int option;
 #ifdef CONFIG_QUOTA
-	int qtype;
+	int qtype, qfmt;
 	char *qname;
 #endif
 
@@ -1018,7 +1018,9 @@ static int parse_options (char *options, struct super_block *sb,
 		case Opt_grpjquota:
 			qtype = GRPQUOTA;
 set_qf_name:
-			if (sb_any_quota_enabled(sb)) {
+			if ((sb_any_quota_enabled(sb) ||
+			     sb_any_quota_suspended(sb)) &&
+			    !sbi->s_qf_names[qtype]) {
 				printk(KERN_ERR
 					"EXT3-fs: Cannot change journaled "
 					"quota options when quota turned on.\n");
@@ -1056,7 +1058,9 @@ set_qf_name:
 		case Opt_offgrpjquota:
 			qtype = GRPQUOTA;
 clear_qf_name:
-			if (sb_any_quota_enabled(sb)) {
+			if ((sb_any_quota_enabled(sb) ||
+			     sb_any_quota_suspended(sb)) &&
+			    sbi->s_qf_names[qtype]) {
 				printk(KERN_ERR "EXT3-fs: Cannot change "
 					"journaled quota options when "
 					"quota turned on.\n");
@@ -1069,10 +1073,20 @@ clear_qf_name:
 			sbi->s_qf_names[qtype] = NULL;
 			break;
 		case Opt_jqfmt_vfsold:
-			sbi->s_jquota_fmt = QFMT_VFS_OLD;
-			break;
+			qfmt = QFMT_VFS_OLD;
+			goto set_qf_format;
 		case Opt_jqfmt_vfsv0:
-			sbi->s_jquota_fmt = QFMT_VFS_V0;
+			qfmt = QFMT_VFS_V0;
+set_qf_format:
+			if ((sb_any_quota_enabled(sb) ||
+			     sb_any_quota_suspended(sb)) &&
+			    sbi->s_jquota_fmt != qfmt) {
+				printk(KERN_ERR "EXT3-fs: Cannot change "
+					"journaled quota options when "
+					"quota turned on.\n");
+				return 0;
+			}
+			sbi->s_jquota_fmt = qfmt;
 			break;
 		case Opt_quota:
 		case Opt_usrquota:
@@ -1084,7 +1098,8 @@ clear_qf_name:
 			set_opt(sbi->s_mount_opt, GRPQUOTA);
 			break;
 		case Opt_noquota:
-			if (sb_any_quota_enabled(sb)) {
+			if (sb_any_quota_enabled(sb) ||
+			    sb_any_quota_suspended(sb)) {
 				printk(KERN_ERR "EXT3-fs: Cannot change quota "
 					"options when quota turned on.\n");
 				return 0;

From 3850f7a521dc17659ef6758a219f083418788490 Mon Sep 17 00:00:00 2001
From: Duane Griffin <duaneg@dghda.com>
Date: Fri, 25 Jul 2008 01:46:19 -0700
Subject: [PATCH 0735/1171] jbd: replace potentially false assertion with if
 block

If an error occurs during jbd cache initialisation it is possible for the
journal_head_cache to be NULL when journal_destroy_journal_head_cache is
called.  Replace the J_ASSERT with an if block to handle the situation
correctly.

Note that even with this fix things will break badly if jbd is statically
compiled in and cache initialisation fails.

Signed-off-by: Duane Griffin <duaneg@dghda.com
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/jbd/journal.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c
index b99c3b3654c4..15ea16ad8662 100644
--- a/fs/jbd/journal.c
+++ b/fs/jbd/journal.c
@@ -1636,9 +1636,10 @@ static int journal_init_journal_head_cache(void)
 
 static void journal_destroy_journal_head_cache(void)
 {
-	J_ASSERT(journal_head_cache != NULL);
-	kmem_cache_destroy(journal_head_cache);
-	journal_head_cache = NULL;
+	if (journal_head_cache) {
+		kmem_cache_destroy(journal_head_cache);
+		journal_head_cache = NULL;
+	}
 }
 
 /*

From f4d79ca2fa211cffc07306eeed7013448e77d7ec Mon Sep 17 00:00:00 2001
From: Duane Griffin <duaneg@dghda.com>
Date: Fri, 25 Jul 2008 01:46:20 -0700
Subject: [PATCH 0736/1171] jbd: eliminate duplicated code in revocation table
 init/destroy functions

The revocation table initialisation/destruction code is repeated for each
of the two revocation tables stored in the journal.  Refactoring the
duplicated code into functions is tidier, simplifies the logic in
initialisation in particular, and slightly reduces the code size.

There should not be any functional change.

Signed-off-by: Duane Griffin <duaneg@dghda.com>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/jbd/revoke.c | 135 ++++++++++++++++++++----------------------------
 1 file changed, 55 insertions(+), 80 deletions(-)

diff --git a/fs/jbd/revoke.c b/fs/jbd/revoke.c
index 1bb43e987f4b..8ff5a7b89b9a 100644
--- a/fs/jbd/revoke.c
+++ b/fs/jbd/revoke.c
@@ -195,109 +195,84 @@ void journal_destroy_revoke_caches(void)
 	revoke_table_cache = NULL;
 }
 
-/* Initialise the revoke table for a given journal to a given size. */
-
-int journal_init_revoke(journal_t *journal, int hash_size)
+static struct jbd_revoke_table_s *journal_init_revoke_table(int hash_size)
 {
-	int shift, tmp;
+	int shift = 0;
+	int tmp = hash_size;
+	struct jbd_revoke_table_s *table;
 
-	J_ASSERT (journal->j_revoke_table[0] == NULL);
+	table = kmem_cache_alloc(revoke_table_cache, GFP_KERNEL);
+	if (!table)
+		goto out;
 
-	shift = 0;
-	tmp = hash_size;
 	while((tmp >>= 1UL) != 0UL)
 		shift++;
 
-	journal->j_revoke_table[0] = kmem_cache_alloc(revoke_table_cache, GFP_KERNEL);
-	if (!journal->j_revoke_table[0])
-		return -ENOMEM;
-	journal->j_revoke = journal->j_revoke_table[0];
-
-	/* Check that the hash_size is a power of two */
-	J_ASSERT(is_power_of_2(hash_size));
-
-	journal->j_revoke->hash_size = hash_size;
-
-	journal->j_revoke->hash_shift = shift;
-
-	journal->j_revoke->hash_table =
+	table->hash_size = hash_size;
+	table->hash_shift = shift;
+	table->hash_table =
 		kmalloc(hash_size * sizeof(struct list_head), GFP_KERNEL);
-	if (!journal->j_revoke->hash_table) {
-		kmem_cache_free(revoke_table_cache, journal->j_revoke_table[0]);
-		journal->j_revoke = NULL;
-		return -ENOMEM;
+	if (!table->hash_table) {
+		kmem_cache_free(revoke_table_cache, table);
+		table = NULL;
+		goto out;
 	}
 
 	for (tmp = 0; tmp < hash_size; tmp++)
-		INIT_LIST_HEAD(&journal->j_revoke->hash_table[tmp]);
+		INIT_LIST_HEAD(&table->hash_table[tmp]);
 
-	journal->j_revoke_table[1] = kmem_cache_alloc(revoke_table_cache, GFP_KERNEL);
-	if (!journal->j_revoke_table[1]) {
-		kfree(journal->j_revoke_table[0]->hash_table);
-		kmem_cache_free(revoke_table_cache, journal->j_revoke_table[0]);
-		return -ENOMEM;
+out:
+	return table;
+}
+
+static void journal_destroy_revoke_table(struct jbd_revoke_table_s *table)
+{
+	int i;
+	struct list_head *hash_list;
+
+	for (i = 0; i < table->hash_size; i++) {
+		hash_list = &table->hash_table[i];
+		J_ASSERT(list_empty(hash_list));
 	}
 
+	kfree(table->hash_table);
+	kmem_cache_free(revoke_table_cache, table);
+}
+
+/* Initialise the revoke table for a given journal to a given size. */
+int journal_init_revoke(journal_t *journal, int hash_size)
+{
+	J_ASSERT(journal->j_revoke_table[0] == NULL);
+	J_ASSERT(is_power_of_2(hash_size));
+
+	journal->j_revoke_table[0] = journal_init_revoke_table(hash_size);
+	if (!journal->j_revoke_table[0])
+		goto fail0;
+
+	journal->j_revoke_table[1] = journal_init_revoke_table(hash_size);
+	if (!journal->j_revoke_table[1])
+		goto fail1;
+
 	journal->j_revoke = journal->j_revoke_table[1];
 
-	/* Check that the hash_size is a power of two */
-	J_ASSERT(is_power_of_2(hash_size));
-
-	journal->j_revoke->hash_size = hash_size;
-
-	journal->j_revoke->hash_shift = shift;
-
-	journal->j_revoke->hash_table =
-		kmalloc(hash_size * sizeof(struct list_head), GFP_KERNEL);
-	if (!journal->j_revoke->hash_table) {
-		kfree(journal->j_revoke_table[0]->hash_table);
-		kmem_cache_free(revoke_table_cache, journal->j_revoke_table[0]);
-		kmem_cache_free(revoke_table_cache, journal->j_revoke_table[1]);
-		journal->j_revoke = NULL;
-		return -ENOMEM;
-	}
-
-	for (tmp = 0; tmp < hash_size; tmp++)
-		INIT_LIST_HEAD(&journal->j_revoke->hash_table[tmp]);
-
 	spin_lock_init(&journal->j_revoke_lock);
 
 	return 0;
+
+fail1:
+	journal_destroy_revoke_table(journal->j_revoke_table[0]);
+fail0:
+	return -ENOMEM;
 }
 
-/* Destoy a journal's revoke table.  The table must already be empty! */
-
+/* Destroy a journal's revoke table.  The table must already be empty! */
 void journal_destroy_revoke(journal_t *journal)
 {
-	struct jbd_revoke_table_s *table;
-	struct list_head *hash_list;
-	int i;
-
-	table = journal->j_revoke_table[0];
-	if (!table)
-		return;
-
-	for (i=0; i<table->hash_size; i++) {
-		hash_list = &table->hash_table[i];
-		J_ASSERT (list_empty(hash_list));
-	}
-
-	kfree(table->hash_table);
-	kmem_cache_free(revoke_table_cache, table);
-	journal->j_revoke = NULL;
-
-	table = journal->j_revoke_table[1];
-	if (!table)
-		return;
-
-	for (i=0; i<table->hash_size; i++) {
-		hash_list = &table->hash_table[i];
-		J_ASSERT (list_empty(hash_list));
-	}
-
-	kfree(table->hash_table);
-	kmem_cache_free(revoke_table_cache, table);
 	journal->j_revoke = NULL;
+	if (journal->j_revoke_table[0])
+		journal_destroy_revoke_table(journal->j_revoke_table[0]);
+	if (journal->j_revoke_table[1])
+		journal_destroy_revoke_table(journal->j_revoke_table[1]);
 }
 
 

From 1984bb763c2e50d0ebfb0cf56d1b319bd7afe63a Mon Sep 17 00:00:00 2001
From: Duane Griffin <duaneg@dghda.com>
Date: Fri, 25 Jul 2008 01:46:21 -0700
Subject: [PATCH 0737/1171] jbd: tidy up revoke cache initialisation and
 destruction

Make revocation cache destruction safe to call if initialisation fails
partially or entirely.  This allows it to be used to cleanup in the case
of initialisation failure, simplifying that code slightly.

Signed-off-by: Duane Griffin <duaneg@dghda.com>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/jbd/revoke.c | 38 ++++++++++++++++++++++++--------------
 1 file changed, 24 insertions(+), 14 deletions(-)

diff --git a/fs/jbd/revoke.c b/fs/jbd/revoke.c
index 8ff5a7b89b9a..c7bd649bbbdc 100644
--- a/fs/jbd/revoke.c
+++ b/fs/jbd/revoke.c
@@ -166,33 +166,43 @@ static struct jbd_revoke_record_s *find_revoke_record(journal_t *journal,
 	return NULL;
 }
 
+void journal_destroy_revoke_caches(void)
+{
+	if (revoke_record_cache) {
+		kmem_cache_destroy(revoke_record_cache);
+		revoke_record_cache = NULL;
+	}
+	if (revoke_table_cache) {
+		kmem_cache_destroy(revoke_table_cache);
+		revoke_table_cache = NULL;
+	}
+}
+
 int __init journal_init_revoke_caches(void)
 {
+	J_ASSERT(!revoke_record_cache);
+	J_ASSERT(!revoke_table_cache);
+
 	revoke_record_cache = kmem_cache_create("revoke_record",
 					   sizeof(struct jbd_revoke_record_s),
 					   0,
 					   SLAB_HWCACHE_ALIGN|SLAB_TEMPORARY,
 					   NULL);
 	if (!revoke_record_cache)
-		return -ENOMEM;
+		goto record_cache_failure;
 
 	revoke_table_cache = kmem_cache_create("revoke_table",
 					   sizeof(struct jbd_revoke_table_s),
 					   0, SLAB_TEMPORARY, NULL);
-	if (!revoke_table_cache) {
-		kmem_cache_destroy(revoke_record_cache);
-		revoke_record_cache = NULL;
-		return -ENOMEM;
-	}
-	return 0;
-}
+	if (!revoke_table_cache)
+		goto table_cache_failure;
 
-void journal_destroy_revoke_caches(void)
-{
-	kmem_cache_destroy(revoke_record_cache);
-	revoke_record_cache = NULL;
-	kmem_cache_destroy(revoke_table_cache);
-	revoke_table_cache = NULL;
+	return 0;
+
+table_cache_failure:
+	journal_destroy_revoke_caches();
+record_cache_failure:
+	return -ENOMEM;
 }
 
 static struct jbd_revoke_table_s *journal_init_revoke_table(int hash_size)

From 9ebfbe9f926553eabc21b4400918d1216b27ed0c Mon Sep 17 00:00:00 2001
From: Shen Feng <shen@cn.fujitsu.com>
Date: Fri, 25 Jul 2008 01:46:21 -0700
Subject: [PATCH 0738/1171] ext3: improve some code in rb tree part of dir.c

- remove unnecessary code in free_rb_tree_fname
 - rename free_rb_tree_fname to ext3_htree_create_dir_info
   since it and ext3_htree_free_dir_info are a pair
 - replace kmalloc with kzalloc in ext3_htree_free_dir_info

Signed-off-by: Shen Feng <shen@cn.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ext3/dir.c | 14 ++++----------
 1 file changed, 4 insertions(+), 10 deletions(-)

diff --git a/fs/ext3/dir.c b/fs/ext3/dir.c
index 8ca3bfd72427..2eea96ec78ed 100644
--- a/fs/ext3/dir.c
+++ b/fs/ext3/dir.c
@@ -272,7 +272,7 @@ static void free_rb_tree_fname(struct rb_root *root)
 
 	while (n) {
 		/* Do the node's children first */
-		if ((n)->rb_left) {
+		if (n->rb_left) {
 			n = n->rb_left;
 			continue;
 		}
@@ -301,24 +301,18 @@ static void free_rb_tree_fname(struct rb_root *root)
 			parent->rb_right = NULL;
 		n = parent;
 	}
-	root->rb_node = NULL;
 }
 
 
-static struct dir_private_info *create_dir_info(loff_t pos)
+static struct dir_private_info *ext3_htree_create_dir_info(loff_t pos)
 {
 	struct dir_private_info *p;
 
-	p = kmalloc(sizeof(struct dir_private_info), GFP_KERNEL);
+	p = kzalloc(sizeof(struct dir_private_info), GFP_KERNEL);
 	if (!p)
 		return NULL;
-	p->root.rb_node = NULL;
-	p->curr_node = NULL;
-	p->extra_fname = NULL;
-	p->last_pos = 0;
 	p->curr_hash = pos2maj_hash(pos);
 	p->curr_minor_hash = pos2min_hash(pos);
-	p->next_hash = 0;
 	return p;
 }
 
@@ -433,7 +427,7 @@ static int ext3_dx_readdir(struct file * filp,
 	int	ret;
 
 	if (!info) {
-		info = create_dir_info(filp->f_pos);
+		info = ext3_htree_create_dir_info(filp->f_pos);
 		if (!info)
 			return -ENOMEM;
 		filp->private_data = info;

From 3f31fddfa26b7594b44ff2b34f9a04ba409e0f91 Mon Sep 17 00:00:00 2001
From: Mingming Cao <cmm@us.ibm.com>
Date: Fri, 25 Jul 2008 01:46:22 -0700
Subject: [PATCH 0739/1171] jbd: fix race between free buffer and commit
 transaction

journal_try_to_free_buffers() could race with jbd commit transaction when
the later is holding the buffer reference while waiting for the data
buffer to flush to disk.  If the caller of journal_try_to_free_buffers()
request tries hard to release the buffers, it will treat the failure as
error and return back to the caller.  We have seen the directo IO failed
due to this race.  Some of the caller of releasepage() also expecting the
buffer to be dropped when passed with GFP_KERNEL mask to the
releasepage()->journal_try_to_free_buffers().

With this patch, if the caller is passing the __GFP_WAIT and __GFP_FS to
indicating this call could wait, in case of try_to_free_buffers() failed,
let's waiting for journal_commit_transaction() to finish commit the
current committing transaction, then try to free those buffers again.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Reviewed-by: Badari Pulavarty <pbadari@us.ibm.com>
Acked-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/jbd/transaction.c | 57 ++++++++++++++++++++++++++++++++++++++++++--
 mm/filemap.c         |  3 +--
 2 files changed, 56 insertions(+), 4 deletions(-)

diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c
index 67ff2024c23c..8dee32007500 100644
--- a/fs/jbd/transaction.c
+++ b/fs/jbd/transaction.c
@@ -1648,12 +1648,42 @@ out:
 	return;
 }
 
+/*
+ * journal_try_to_free_buffers() could race with journal_commit_transaction()
+ * The latter might still hold the a count on buffers when inspecting
+ * them on t_syncdata_list or t_locked_list.
+ *
+ * journal_try_to_free_buffers() will call this function to
+ * wait for the current transaction to finish syncing data buffers, before
+ * tryinf to free that buffer.
+ *
+ * Called with journal->j_state_lock held.
+ */
+static void journal_wait_for_transaction_sync_data(journal_t *journal)
+{
+	transaction_t *transaction = NULL;
+	tid_t tid;
+
+	spin_lock(&journal->j_state_lock);
+	transaction = journal->j_committing_transaction;
+
+	if (!transaction) {
+		spin_unlock(&journal->j_state_lock);
+		return;
+	}
+
+	tid = transaction->t_tid;
+	spin_unlock(&journal->j_state_lock);
+	log_wait_commit(journal, tid);
+}
 
 /**
  * int journal_try_to_free_buffers() - try to free page buffers.
  * @journal: journal for operation
  * @page: to try and free
- * @unused_gfp_mask: unused
+ * @gfp_mask: we use the mask to detect how hard should we try to release
+ * buffers. If __GFP_WAIT and __GFP_FS is set, we wait for commit code to
+ * release the buffers.
  *
  *
  * For all the buffers on this page,
@@ -1682,9 +1712,11 @@ out:
  * journal_try_to_free_buffer() is changing its state.  But that
  * cannot happen because we never reallocate freed data as metadata
  * while the data is part of a transaction.  Yes?
+ *
+ * Return 0 on failure, 1 on success
  */
 int journal_try_to_free_buffers(journal_t *journal,
-				struct page *page, gfp_t unused_gfp_mask)
+				struct page *page, gfp_t gfp_mask)
 {
 	struct buffer_head *head;
 	struct buffer_head *bh;
@@ -1713,7 +1745,28 @@ int journal_try_to_free_buffers(journal_t *journal,
 		if (buffer_jbd(bh))
 			goto busy;
 	} while ((bh = bh->b_this_page) != head);
+
 	ret = try_to_free_buffers(page);
+
+	/*
+	 * There are a number of places where journal_try_to_free_buffers()
+	 * could race with journal_commit_transaction(), the later still
+	 * holds the reference to the buffers to free while processing them.
+	 * try_to_free_buffers() failed to free those buffers. Some of the
+	 * caller of releasepage() request page buffers to be dropped, otherwise
+	 * treat the fail-to-free as errors (such as generic_file_direct_IO())
+	 *
+	 * So, if the caller of try_to_release_page() wants the synchronous
+	 * behaviour(i.e make sure buffers are dropped upon return),
+	 * let's wait for the current transaction to finish flush of
+	 * dirty data buffers, then try to free those buffers again,
+	 * with the journal locked.
+	 */
+	if (ret == 0 && (gfp_mask & __GFP_WAIT) && (gfp_mask & __GFP_FS)) {
+		journal_wait_for_transaction_sync_data(journal);
+		ret = try_to_free_buffers(page);
+	}
+
 busy:
 	return ret;
 }
diff --git a/mm/filemap.c b/mm/filemap.c
index 7675b91f4f63..5d4c880d7cd9 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -2563,9 +2563,8 @@ EXPORT_SYMBOL(generic_file_aio_write);
  * Otherwise return zero.
  *
  * The @gfp_mask argument specifies whether I/O may be performed to release
- * this page (__GFP_IO), and whether the call may block (__GFP_WAIT).
+ * this page (__GFP_IO), and whether the call may block (__GFP_WAIT & __GFP_FS).
  *
- * NOTE: @gfp_mask may go away, and this function may become non-blocking.
  */
 int try_to_release_page(struct page *page, gfp_t gfp_mask)
 {

From ef1afd39519b74fbe1f63c9ab5a14490effec0e3 Mon Sep 17 00:00:00 2001
From: Shen Feng <shen@cn.fujitsu.com>
Date: Fri, 25 Jul 2008 01:46:23 -0700
Subject: [PATCH 0740/1171] ext3: remove double definitions of xattr macros

remove the definitions of macros:
XATTR_TRUSTED_PREFIX
XATTR_USER_PREFIX
since they are defined in linux/xattr.h

Signed-off-by: Shen Feng <shen@cn.fujitsu.com>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ext3/xattr_security.c | 2 +-
 fs/ext3/xattr_trusted.c  | 4 +---
 fs/ext3/xattr_user.c     | 4 +---
 3 files changed, 3 insertions(+), 7 deletions(-)

diff --git a/fs/ext3/xattr_security.c b/fs/ext3/xattr_security.c
index 821efaf2b94e..37b81097bdf2 100644
--- a/fs/ext3/xattr_security.c
+++ b/fs/ext3/xattr_security.c
@@ -15,7 +15,7 @@ static size_t
 ext3_xattr_security_list(struct inode *inode, char *list, size_t list_size,
 			 const char *name, size_t name_len)
 {
-	const size_t prefix_len = sizeof(XATTR_SECURITY_PREFIX)-1;
+	const size_t prefix_len = XATTR_SECURITY_PREFIX_LEN;
 	const size_t total_len = prefix_len + name_len + 1;
 
 
diff --git a/fs/ext3/xattr_trusted.c b/fs/ext3/xattr_trusted.c
index 0327497a55ce..c7c41a410c4b 100644
--- a/fs/ext3/xattr_trusted.c
+++ b/fs/ext3/xattr_trusted.c
@@ -13,13 +13,11 @@
 #include <linux/ext3_fs.h>
 #include "xattr.h"
 
-#define XATTR_TRUSTED_PREFIX "trusted."
-
 static size_t
 ext3_xattr_trusted_list(struct inode *inode, char *list, size_t list_size,
 			const char *name, size_t name_len)
 {
-	const size_t prefix_len = sizeof(XATTR_TRUSTED_PREFIX)-1;
+	const size_t prefix_len = XATTR_TRUSTED_PREFIX_LEN;
 	const size_t total_len = prefix_len + name_len + 1;
 
 	if (!capable(CAP_SYS_ADMIN))
diff --git a/fs/ext3/xattr_user.c b/fs/ext3/xattr_user.c
index 1abd8f92c440..430fe63b31b3 100644
--- a/fs/ext3/xattr_user.c
+++ b/fs/ext3/xattr_user.c
@@ -12,13 +12,11 @@
 #include <linux/ext3_fs.h>
 #include "xattr.h"
 
-#define XATTR_USER_PREFIX "user."
-
 static size_t
 ext3_xattr_user_list(struct inode *inode, char *list, size_t list_size,
 		     const char *name, size_t name_len)
 {
-	const size_t prefix_len = sizeof(XATTR_USER_PREFIX)-1;
+	const size_t prefix_len = XATTR_USER_PREFIX_LEN;
 	const size_t total_len = prefix_len + name_len + 1;
 
 	if (!test_opt(inode->i_sb, XATTR_USER))

From ae76dd9a6b5bbe5315fb7028e03f68f75b8538f3 Mon Sep 17 00:00:00 2001
From: Duane Griffin <duaneg@dghda.com>
Date: Fri, 25 Jul 2008 01:46:23 -0700
Subject: [PATCH 0741/1171] ext3: handle corrupted orphan list at mount

If the orphan node list includes valid, untruncatable nodes with nlink > 0
the ext3_orphan_cleanup loop which attempts to delete them will not do so,
causing it to loop forever. Fix by checking for such nodes in the
ext3_orphan_get function.

This patch fixes the second case (image hdb.20000009.softlockup.gz)
reported in http://bugzilla.kernel.org/show_bug.cgi?id=10882.

[akpm@linux-foundation.org: coding-style fixes]
[akpm@linux-foundation.org: printk warning fix]
Signed-off-by: Duane Griffin <duaneg@dghda.com>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ext3/ialloc.c        |  9 +++++++++
 fs/ext3/inode.c         | 20 ++++++++++++++------
 include/linux/ext3_fs.h |  1 +
 3 files changed, 24 insertions(+), 6 deletions(-)

diff --git a/fs/ext3/ialloc.c b/fs/ext3/ialloc.c
index 77126821b2e9..47b678d73e7a 100644
--- a/fs/ext3/ialloc.c
+++ b/fs/ext3/ialloc.c
@@ -669,6 +669,14 @@ struct inode *ext3_orphan_get(struct super_block *sb, unsigned long ino)
 	if (IS_ERR(inode))
 		goto iget_failed;
 
+	/*
+	 * If the orphans has i_nlinks > 0 then it should be able to be
+	 * truncated, otherwise it won't be removed from the orphan list
+	 * during processing and an infinite loop will result.
+	 */
+	if (inode->i_nlink && !ext3_can_truncate(inode))
+		goto bad_orphan;
+
 	if (NEXT_ORPHAN(inode) > max_ino)
 		goto bad_orphan;
 	brelse(bitmap_bh);
@@ -690,6 +698,7 @@ bad_orphan:
 		printk(KERN_NOTICE "NEXT_ORPHAN(inode)=%u\n",
 		       NEXT_ORPHAN(inode));
 		printk(KERN_NOTICE "max_ino=%lu\n", max_ino);
+		printk(KERN_NOTICE "i_nlink=%u\n", inode->i_nlink);
 		/* Avoid freeing blocks if we got a bad deleted inode */
 		if (inode->i_nlink == 0)
 			inode->i_blocks = 0;
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index 6ae4ecf3ce40..74b432fa166b 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -2253,6 +2253,19 @@ static void ext3_free_branches(handle_t *handle, struct inode *inode,
 	}
 }
 
+int ext3_can_truncate(struct inode *inode)
+{
+	if (IS_APPEND(inode) || IS_IMMUTABLE(inode))
+		return 0;
+	if (S_ISREG(inode->i_mode))
+		return 1;
+	if (S_ISDIR(inode->i_mode))
+		return 1;
+	if (S_ISLNK(inode->i_mode))
+		return !ext3_inode_is_fast_symlink(inode);
+	return 0;
+}
+
 /*
  * ext3_truncate()
  *
@@ -2297,12 +2310,7 @@ void ext3_truncate(struct inode *inode)
 	unsigned blocksize = inode->i_sb->s_blocksize;
 	struct page *page;
 
-	if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) ||
-	    S_ISLNK(inode->i_mode)))
-		return;
-	if (ext3_inode_is_fast_symlink(inode))
-		return;
-	if (IS_APPEND(inode) || IS_IMMUTABLE(inode))
+	if (!ext3_can_truncate(inode))
 		return;
 
 	/*
diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h
index 36c540396377..80171ee89a22 100644
--- a/include/linux/ext3_fs.h
+++ b/include/linux/ext3_fs.h
@@ -832,6 +832,7 @@ extern void ext3_discard_reservation (struct inode *);
 extern void ext3_dirty_inode(struct inode *);
 extern int ext3_change_inode_journal_flag(struct inode *, int);
 extern int ext3_get_inode_loc(struct inode *, struct ext3_iloc *);
+extern int ext3_can_truncate(struct inode *inode);
 extern void ext3_truncate (struct inode *);
 extern void ext3_set_inode_flags(struct inode *);
 extern void ext3_get_inode_flags(struct ext3_inode_info *);

From 95450f5a7e53d5752ce1a0d0b8282e10fe745ae0 Mon Sep 17 00:00:00 2001
From: Hidehiro Kawai <hidehiro.kawai.ez@hitachi.com>
Date: Fri, 25 Jul 2008 01:46:24 -0700
Subject: [PATCH 0742/1171] ext3: don't read inode block if the buffer has a
 write error

A transient I/O error can corrupt inode data.  Here is the scenario:

(1) update inode_A at the block_B
(2) pdflush writes out new inode_A to the filesystem, but it results
    in write I/O error, at this point, BH_Uptodate flag of the buffer
    for block_B is cleared and BH_Write_EIO is set
(3) create new inode_C which located at block_B, and
    __ext3_get_inode_loc() tries to read on-disk block_B because the
    buffer is not uptodate
(4) if it can read on-disk block_B successfully, inode_A is
    overwritten by old data

This patch makes __ext3_get_inode_loc() not read the inode block if the
buffer has BH_Write_EIO flag.  In this case, the buffer should have the
latest information, so setting the uptodate flag to the buffer (this
avoids WARN_ON_ONCE() in mark_buffer_dirty().)

According to this change, we would need to test BH_Write_EIO flag for the
error checking.  Currently nobody checks write I/O errors on metadata
buffers, but it will be done in other patches I'm working on.

Signed-off-by: Hidehiro Kawai <hidehiro.kawai.ez@hitachi.com>
Cc: sugita <yumiko.sugita.yf@hitachi.com>
Cc: Satoshi OSHIMA <satoshi.oshima.fk@hitachi.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Jan Kara <jack@ucw.cz>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ext3/inode.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index 74b432fa166b..36f74f17a11c 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -2521,6 +2521,16 @@ static int __ext3_get_inode_loc(struct inode *inode,
 	}
 	if (!buffer_uptodate(bh)) {
 		lock_buffer(bh);
+
+		/*
+		 * If the buffer has the write error flag, we have failed
+		 * to write out another inode in the same block.  In this
+		 * case, we don't have to read the block because we may
+		 * read the old inode data successfully.
+		 */
+		if (buffer_write_io_error(bh) && !buffer_uptodate(bh))
+			set_buffer_uptodate(bh);
+
 		if (buffer_uptodate(bh)) {
 			/* someone brought it uptodate while we waited */
 			unlock_buffer(bh);

From 3ccc3167b0e5d46ab3bf03e22fbdb7616ce038cd Mon Sep 17 00:00:00 2001
From: Duane Griffin <duaneg@dghda.com>
Date: Fri, 25 Jul 2008 01:46:26 -0700
Subject: [PATCH 0743/1171] ext3: handle deleting corrupted indirect blocks

While freeing indirect blocks we attach a journal head to the parent
buffer head, free the blocks, then journal the parent.  If the indirect
block list is corrupted and points to the parent the journal head will be
detached when the block is cleared, causing an OOPS.

Check for that explicitly and handle it gracefully.

This patch fixes the third case (image hdb.20000057.nullderef.gz)
reported in http://bugzilla.kernel.org/show_bug.cgi?id=10882.

Immediately above the change, in the ext3_free_data function, we call
ext3_clear_blocks to clear the indirect blocks in this parent block.  If
one of those blocks happens to actually be the parent block it will clear
b_private / BH_JBD.

I did the check at the end rather than earlier as it seemed more elegant.
I don't think there should be much practical difference, although it is
possible the FS may not be quite so badly corrupted if we did it the other
way (and didn't clear the block at all).  To be honest, I'm not convinced
there aren't other similar failure modes lurking in this code, although I
couldn't find any with a quick review.

[akpm@linux-foundation.org: fix printk warning]
Signed-off-by: Duane Griffin <duaneg@dghda.com>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ext3/inode.c | 16 +++++++++++++++-
 1 file changed, 15 insertions(+), 1 deletion(-)

diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index 36f74f17a11c..3bf07d70b914 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -2127,7 +2127,21 @@ static void ext3_free_data(handle_t *handle, struct inode *inode,
 
 	if (this_bh) {
 		BUFFER_TRACE(this_bh, "call ext3_journal_dirty_metadata");
-		ext3_journal_dirty_metadata(handle, this_bh);
+
+		/*
+		 * The buffer head should have an attached journal head at this
+		 * point. However, if the data is corrupted and an indirect
+		 * block pointed to itself, it would have been detached when
+		 * the block was cleared. Check for this instead of OOPSing.
+		 */
+		if (bh2jh(this_bh))
+			ext3_journal_dirty_metadata(handle, this_bh);
+		else
+			ext3_error(inode->i_sb, "ext3_free_data",
+				   "circular indirect block detected, "
+				   "inode=%lu, block=%llu",
+				   inode->i_ino,
+				   (unsigned long long)this_bh->b_blocknr);
 	}
 }
 

From a10320e8f7c4dcfa050aac566092f29b40458d5a Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 01:46:26 -0700
Subject: [PATCH 0744/1171] jbd: unexport journal_update_superblock

Remove the unused EXPORT_SYMBOL(journal_update_superblock).

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/jbd/journal.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c
index 15ea16ad8662..aa7143a8349b 100644
--- a/fs/jbd/journal.c
+++ b/fs/jbd/journal.c
@@ -68,7 +68,6 @@ EXPORT_SYMBOL(journal_set_features);
 EXPORT_SYMBOL(journal_create);
 EXPORT_SYMBOL(journal_load);
 EXPORT_SYMBOL(journal_destroy);
-EXPORT_SYMBOL(journal_update_superblock);
 EXPORT_SYMBOL(journal_abort);
 EXPORT_SYMBOL(journal_errno);
 EXPORT_SYMBOL(journal_ack_err);

From fc80c44277b3c92d808b73e9d40e120229aa4b6a Mon Sep 17 00:00:00 2001
From: Toshiyuki Okajima <toshi.okajima@jp.fujitsu.com>
Date: Fri, 25 Jul 2008 01:46:29 -0700
Subject: [PATCH 0745/1171] jbd: positively dispose the unmapped data buffers
 in journal_commit_transaction()

After ext3-ordered files are truncated, there is a possibility that the
pages which cannot be estimated still remain.  Remaining pages can be
released when the system has really few memory.  So, it is not memory
leakage.  But the resource management software etc.  may not work
correctly.

It is possible that journal_unmap_buffer() cannot release the buffers, and
the pages to which they belong because they are attached to a commiting
transaction and journal_unmap_buffer() cannot release them.  To release
such the buffers and the pages later, journal_unmap_buffer() leaves it to
journal_commit_transaction().  (journal_unmap_buffer() puts the mark
'BH_Freed' to the buffers so that journal_commit_transaction() can
identify whether they can be released or not.)

In the journalled mode and the writeback mode, jbd does with only metadata
buffers.  But in the ordered mode, jbd does with metadata buffers and also
data buffers.

Actually, journal_commit_transaction() releases only the metadata buffers
of which release is demanded by journal_unmap_buffer(), and also releases
the pages to which they belong if possible.

As a result, the data buffers of which release is demanded by
journal_unmap_buffer() remain after a transaction commits.  And also the
pages to which they belong remain.

Such the remained pages don't have mapping any longer.  Due to this fact,
there is a possibility that the pages which cannot be estimated remain.

The metadata buffers marked 'BH_Freed' and the pages to which
they belong can be released at 'JBD: commit phase 7'.

Therefore, by applying the same code into 'JBD: commit phase 2' (where the
data buffers are done with), journal_commit_transaction() can also release
the data buffers marked 'BH_Freed' and the pages to which they belong.

As a result, all the buffers marked 'BH_Freed' can be released, and also
all the pages to which these buffers belong can be released at
journal_commit_transaction().  So, the page which cannot be estimated is
lost.

<<Excerpt of code at 'JBD: commit phase 7'>>
 >         spin_lock(&journal->j_list_lock);
 >         while (commit_transaction->t_forget) {
 >                 transaction_t *cp_transaction;
 >                 struct buffer_head *bh;
 >
 >                 jh = commit_transaction->t_forget;
 >...
 >                 if (buffer_freed(bh)) {
 >                 ^^^^^^^^^^^^^^^^^^^^^^^^
 >                         clear_buffer_freed(bh);
 >                        ^^^^^^^^^^^^^^^^^^^^^^^^
 >                         clear_buffer_jbddirty(bh);
 >                 }
 >
 >                 if (buffer_jbddirty(bh)) {
 >                         JBUFFER_TRACE(jh, "add to new checkpointing trans");
 >                         __journal_insert_checkpoint(jh, commit_transaction);
 >                         JBUFFER_TRACE(jh, "refile for checkpoint writeback");
 >                         __journal_refile_buffer(jh);
 >                         jbd_unlock_bh_state(bh);
 >                 } else {
 >                         J_ASSERT_BH(bh, !buffer_dirty(bh));
 > ...
 >                         JBUFFER_TRACE(jh, "refile or unfile freed buffer");
 >                         __journal_refile_buffer(jh);
 >                         if (!jh->b_transaction) {
 >                                 jbd_unlock_bh_state(bh);
 >                                  /* needs a brelse */
 >                                 journal_remove_journal_head(bh);
 >                                 release_buffer_page(bh);
 >                                 ^^^^^^^^^^^^^^^^^^^^^^^^
 >                         } else
 >                 }
****************************************************************
* Apply the code of "^^^^^^" lines into 'JBD: commit phase 2' *
****************************************************************

At journal_commit_transaction() code, there is one extra message in the
series of jbd debug messages.  ("JBD: commit phase 2") This patch fixes
it, too.

Signed-off-by: Toshiyuki Okajima <toshi.okajima@jp.fujitsu.com>
Acked-by: Jan Kara <jack@suse.cz>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/jbd/commit.c | 29 ++++++++++++++++++++---------
 1 file changed, 20 insertions(+), 9 deletions(-)

diff --git a/fs/jbd/commit.c b/fs/jbd/commit.c
index 5a8ca61498ca..f943b9b3f208 100644
--- a/fs/jbd/commit.c
+++ b/fs/jbd/commit.c
@@ -36,7 +36,7 @@ static void journal_end_buffer_io_sync(struct buffer_head *bh, int uptodate)
 
 /*
  * When an ext3-ordered file is truncated, it is possible that many pages are
- * not sucessfully freed, because they are attached to a committing transaction.
+ * not successfully freed, because they are attached to a committing transaction.
  * After the transaction commits, these pages are left on the LRU, with no
  * ->mapping, and with attached buffers.  These pages are trivially reclaimable
  * by the VM, but their apparent absence upsets the VM accounting, and it makes
@@ -45,8 +45,8 @@ static void journal_end_buffer_io_sync(struct buffer_head *bh, int uptodate)
  * So here, we have a buffer which has just come off the forget list.  Look to
  * see if we can strip all buffers from the backing page.
  *
- * Called under lock_journal(), and possibly under journal_datalist_lock.  The
- * caller provided us with a ref against the buffer, and we drop that here.
+ * Called under journal->j_list_lock.  The caller provided us with a ref
+ * against the buffer, and we drop that here.
  */
 static void release_buffer_page(struct buffer_head *bh)
 {
@@ -77,6 +77,19 @@ nope:
 	__brelse(bh);
 }
 
+/*
+ * Decrement reference counter for data buffer. If it has been marked
+ * 'BH_Freed', release it and the page to which it belongs if possible.
+ */
+static void release_data_buffer(struct buffer_head *bh)
+{
+	if (buffer_freed(bh)) {
+		clear_buffer_freed(bh);
+		release_buffer_page(bh);
+	} else
+		put_bh(bh);
+}
+
 /*
  * Try to acquire jbd_lock_bh_state() against the buffer, when j_list_lock is
  * held.  For ranking reasons we must trylock.  If we lose, schedule away and
@@ -231,7 +244,7 @@ write_out_data:
 			if (locked)
 				unlock_buffer(bh);
 			BUFFER_TRACE(bh, "already cleaned up");
-			put_bh(bh);
+			release_data_buffer(bh);
 			continue;
 		}
 		if (locked && test_clear_buffer_dirty(bh)) {
@@ -258,10 +271,10 @@ write_out_data:
 			if (locked)
 				unlock_buffer(bh);
 			journal_remove_journal_head(bh);
-			/* Once for our safety reference, once for
+			/* One for our safety reference, other for
 			 * journal_remove_journal_head() */
 			put_bh(bh);
-			put_bh(bh);
+			release_data_buffer(bh);
 		}
 
 		if (need_resched() || spin_needbreak(&journal->j_list_lock)) {
@@ -443,7 +456,7 @@ void journal_commit_transaction(journal_t *journal)
 		} else {
 			jbd_unlock_bh_state(bh);
 		}
-		put_bh(bh);
+		release_data_buffer(bh);
 		cond_resched_lock(&journal->j_list_lock);
 	}
 	spin_unlock(&journal->j_list_lock);
@@ -453,8 +466,6 @@ void journal_commit_transaction(journal_t *journal)
 
 	journal_write_revoke_records(journal, commit_transaction);
 
-	jbd_debug(3, "JBD: commit phase 2\n");
-
 	/*
 	 * If we found any dirty or locked buffers, then we should have
 	 * looped back up to the write_out_data label.  If there weren't

From 8ef2720397bb813d4985405a5ae7b8ad6474188b Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Fri, 25 Jul 2008 01:46:29 -0700
Subject: [PATCH 0746/1171] ext3: kill 2 useless magic numbers

dx_root_limit() will never return 20, and I can't figure out what 20
stands for.  This function has never changed since htree directory
indexing was merged.

Similar for dx_node_limit() and the magic 22.

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Acked-by: Andreas Dilger <adilger@sun.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ext3/namei.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c
index 0b8cf80154f1..d282ea870081 100644
--- a/fs/ext3/namei.c
+++ b/fs/ext3/namei.c
@@ -240,13 +240,13 @@ static inline unsigned dx_root_limit (struct inode *dir, unsigned infosize)
 {
 	unsigned entry_space = dir->i_sb->s_blocksize - EXT3_DIR_REC_LEN(1) -
 		EXT3_DIR_REC_LEN(2) - infosize;
-	return 0? 20: entry_space / sizeof(struct dx_entry);
+	return entry_space / sizeof(struct dx_entry);
 }
 
 static inline unsigned dx_node_limit (struct inode *dir)
 {
 	unsigned entry_space = dir->i_sb->s_blocksize - EXT3_DIR_REC_LEN(0);
-	return 0? 22: entry_space / sizeof(struct dx_entry);
+	return entry_space / sizeof(struct dx_entry);
 }
 
 /*

From cbe5f466f6995e10a10c7ae66d6dc8608f08a6b8 Mon Sep 17 00:00:00 2001
From: Hidehiro Kawai <hidehiro.kawai.ez@hitachi.com>
Date: Fri, 25 Jul 2008 01:46:30 -0700
Subject: [PATCH 0747/1171] jbd: don't abort if flushing file data failed

In ordered mode, the current jbd aborts the journal if a file data buffer
has an error.  But this behavior is unintended, and we found that it has
been adopted accidentally.

This patch undoes it and just calls printk() instead of aborting the
journal.  Additionally, set AS_EIO into the address_space object of the
failed buffer which is submitted by journal_do_submit_data() so that
fsync() can get -EIO.

Missing error checkings are also added to inform errors on file data
buffers to the user.  The following buffers are targeted.

  (a) the buffer which has already been written out by pdflush
  (b) the buffer which has been unlocked before scanned in the
      t_locked_list loop

[akpm@linux-foundation.org: improve grammar in a printk]
Signed-off-by: Hidehiro Kawai <hidehiro.kawai.ez@hitachi.com>
Acked-by: Jan Kara <jack@suse.cz>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/jbd/commit.c | 35 ++++++++++++++++++++++++++++-------
 1 file changed, 28 insertions(+), 7 deletions(-)

diff --git a/fs/jbd/commit.c b/fs/jbd/commit.c
index f943b9b3f208..2eccbfaa1d48 100644
--- a/fs/jbd/commit.c
+++ b/fs/jbd/commit.c
@@ -185,7 +185,7 @@ static void journal_do_submit_data(struct buffer_head **wbuf, int bufs)
 /*
  *  Submit all the data buffers to disk
  */
-static void journal_submit_data_buffers(journal_t *journal,
+static int journal_submit_data_buffers(journal_t *journal,
 				transaction_t *commit_transaction)
 {
 	struct journal_head *jh;
@@ -193,6 +193,7 @@ static void journal_submit_data_buffers(journal_t *journal,
 	int locked;
 	int bufs = 0;
 	struct buffer_head **wbuf = journal->j_wbuf;
+	int err = 0;
 
 	/*
 	 * Whenever we unlock the journal and sleep, things can get added
@@ -266,6 +267,8 @@ write_out_data:
 			put_bh(bh);
 		} else {
 			BUFFER_TRACE(bh, "writeout complete: unfile");
+			if (unlikely(!buffer_uptodate(bh)))
+				err = -EIO;
 			__journal_unfile_buffer(jh);
 			jbd_unlock_bh_state(bh);
 			if (locked)
@@ -284,6 +287,8 @@ write_out_data:
 	}
 	spin_unlock(&journal->j_list_lock);
 	journal_do_submit_data(wbuf, bufs);
+
+	return err;
 }
 
 /*
@@ -423,8 +428,7 @@ void journal_commit_transaction(journal_t *journal)
 	 * Now start flushing things to disk, in the order they appear
 	 * on the transaction lists.  Data blocks go first.
 	 */
-	err = 0;
-	journal_submit_data_buffers(journal, commit_transaction);
+	err = journal_submit_data_buffers(journal, commit_transaction);
 
 	/*
 	 * Wait for all previously submitted IO to complete.
@@ -439,10 +443,21 @@ void journal_commit_transaction(journal_t *journal)
 		if (buffer_locked(bh)) {
 			spin_unlock(&journal->j_list_lock);
 			wait_on_buffer(bh);
-			if (unlikely(!buffer_uptodate(bh)))
-				err = -EIO;
 			spin_lock(&journal->j_list_lock);
 		}
+		if (unlikely(!buffer_uptodate(bh))) {
+			if (TestSetPageLocked(bh->b_page)) {
+				spin_unlock(&journal->j_list_lock);
+				lock_page(bh->b_page);
+				spin_lock(&journal->j_list_lock);
+			}
+			if (bh->b_page->mapping)
+				set_bit(AS_EIO, &bh->b_page->mapping->flags);
+
+			unlock_page(bh->b_page);
+			SetPageError(bh->b_page);
+			err = -EIO;
+		}
 		if (!inverted_lock(journal, bh)) {
 			put_bh(bh);
 			spin_lock(&journal->j_list_lock);
@@ -461,8 +476,14 @@ void journal_commit_transaction(journal_t *journal)
 	}
 	spin_unlock(&journal->j_list_lock);
 
-	if (err)
-		journal_abort(journal, err);
+	if (err) {
+		char b[BDEVNAME_SIZE];
+
+		printk(KERN_WARNING
+			"JBD: Detected IO errors while flushing file data "
+			"on %s\n", bdevname(journal->j_fs_dev, b));
+		err = 0;
+	}
 
 	journal_write_revoke_records(journal, commit_transaction);
 

From 275c0a8f1253a7542ad9726956c918d8a1f694c4 Mon Sep 17 00:00:00 2001
From: Duane Griffin <duaneg@dghda.com>
Date: Fri, 25 Jul 2008 01:46:31 -0700
Subject: [PATCH 0748/1171] ext3: validate directory entry data before use

ext3_dx_find_entry uses ext3_next_entry without verifying that the entry
is valid.  If its rec_len == 0 this causes an infinite loop.  Refactor the
loop to check the validity of entries before checking whether they match
and moving onto the next one.

There are other uses of ext3_next_entry in this file which also look
problematic.  They should be reviewed and fixed if/when we have a
test-case that triggers them.

This patch fixes the first case (image hdb.25.softlockup.gz) reported in
http://bugzilla.kernel.org/show_bug.cgi?id=10882.

Signed-off-by: Duane Griffin <duaneg@dghda.com>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ext3/namei.c | 22 ++++++++++++----------
 1 file changed, 12 insertions(+), 10 deletions(-)

diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c
index d282ea870081..de13e919cd81 100644
--- a/fs/ext3/namei.c
+++ b/fs/ext3/namei.c
@@ -991,19 +991,21 @@ static struct buffer_head * ext3_dx_find_entry(struct dentry *dentry,
 		de = (struct ext3_dir_entry_2 *) bh->b_data;
 		top = (struct ext3_dir_entry_2 *) ((char *) de + sb->s_blocksize -
 				       EXT3_DIR_REC_LEN(0));
-		for (; de < top; de = ext3_next_entry(de))
-		if (ext3_match (namelen, name, de)) {
-			if (!ext3_check_dir_entry("ext3_find_entry",
-						  dir, de, bh,
-				  (block<<EXT3_BLOCK_SIZE_BITS(sb))
-					  +((char *)de - bh->b_data))) {
-				brelse (bh);
+		for (; de < top; de = ext3_next_entry(de)) {
+			int off = (block << EXT3_BLOCK_SIZE_BITS(sb))
+				  + ((char *) de - bh->b_data);
+
+			if (!ext3_check_dir_entry(__func__, dir, de, bh, off)) {
+				brelse(bh);
 				*err = ERR_BAD_DX_DIR;
 				goto errout;
 			}
-			*res_dir = de;
-			dx_release (frames);
-			return bh;
+
+			if (ext3_match(namelen, name, de)) {
+				*res_dir = de;
+				dx_release(frames);
+				return bh;
+			}
 		}
 		brelse (bh);
 		/* Check to see if we should continue to search */

From c0a1633b6201ef79e31b7da464d44fdf5953054d Mon Sep 17 00:00:00 2001
From: Adam Greenblatt <adam.greenblatt@gmail.com>
Date: Fri, 25 Jul 2008 01:46:32 -0700
Subject: [PATCH 0749/1171] isofs: fix minor filesystem corruption

Some iso9660 images contain files with rockridge data that is either
incorrect or incompletely parsed.  Prior to commit
f2966632a134e865db3c819346a1dc7d96e05309 ("[PATCH] rock: handle directory
overflows") (included with kernel 2.6.13) the kernel ignored the rockridge
data for these files, while still allowing the files to be accessed under
their non-rockridge names.  That commit inadvertently changed things so
that files with invalid rockridge data could not be accessed at all.  (I
ran across the problem when comparing some old CDs with hard disk copies I
had made long ago under kernel 2.4: a few of the files on the hard disk
copies were no longer visible on the CDs.)

This change reverts to the pre-2.6.13 behavior.

Signed-off-by: Adam Greenblatt <adam.greenblatt@gmail.com>
Reviewed-by: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: <stable@kernel.org>		[2.6.25.x, 2.6.26.x]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/isofs/rock.c | 22 ++++++++++++++++++++--
 1 file changed, 20 insertions(+), 2 deletions(-)

diff --git a/fs/isofs/rock.c b/fs/isofs/rock.c
index 6bd48f0a7047..c2fb2dd0131f 100644
--- a/fs/isofs/rock.c
+++ b/fs/isofs/rock.c
@@ -209,6 +209,11 @@ repeat:
 
 	while (rs.len > 2) { /* There may be one byte for padding somewhere */
 		rr = (struct rock_ridge *)rs.chr;
+		/*
+		 * Ignore rock ridge info if rr->len is out of range, but
+		 * don't return -EIO because that would make the file
+		 * invisible.
+		 */
 		if (rr->len < 3)
 			goto out;	/* Something got screwed up here */
 		sig = isonum_721(rs.chr);
@@ -216,8 +221,12 @@ repeat:
 			goto eio;
 		rs.chr += rr->len;
 		rs.len -= rr->len;
+		/*
+		 * As above, just ignore the rock ridge info if rr->len
+		 * is bogus.
+		 */
 		if (rs.len < 0)
-			goto eio;	/* corrupted isofs */
+			goto out;	/* Something got screwed up here */
 
 		switch (sig) {
 		case SIG('R', 'R'):
@@ -307,6 +316,11 @@ parse_rock_ridge_inode_internal(struct iso_directory_record *de,
 repeat:
 	while (rs.len > 2) { /* There may be one byte for padding somewhere */
 		rr = (struct rock_ridge *)rs.chr;
+		/*
+		 * Ignore rock ridge info if rr->len is out of range, but
+		 * don't return -EIO because that would make the file
+		 * invisible.
+		 */
 		if (rr->len < 3)
 			goto out;	/* Something got screwed up here */
 		sig = isonum_721(rs.chr);
@@ -314,8 +328,12 @@ repeat:
 			goto eio;
 		rs.chr += rr->len;
 		rs.len -= rr->len;
+		/*
+		 * As above, just ignore the rock ridge info if rr->len
+		 * is bogus.
+		 */
 		if (rs.len < 0)
-			goto eio;	/* corrupted isofs */
+			goto out;	/* Something got screwed up here */
 
 		switch (sig) {
 #ifndef CONFIG_ZISOFS		/* No flag for SF or ZF */

From de0ca06a99c33df8333955642843331ab6b6e7ff Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 01:46:34 -0700
Subject: [PATCH 0750/1171] coda: remove CODA_FS_OLD_API

While fixing CONFIG_ leakages to the userspace kernel headers I ran into
CODA_FS_OLD_API.

After five years, are there still people using the old API left?
Especially considering that you have to choose at compile time which API
to support in the kernel (and distributions tend to offer the new API for
some time).

Jan: "The old API can definitely go.  Around the time the new
      interface went in there were some non-Coda userspace file system
      implementations that took a while longer to convert to the new API,
      but by now they all switched to the new interface or in some cases
      to a FUSE-based solution."

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Acked-by: Jan Harkes <jaharkes@cs.cmu.edu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/Kconfig           | 14 --------------
 fs/coda/coda_linux.c |  6 ++----
 fs/coda/psdev.c      |  4 ----
 fs/coda/upcall.c     | 15 +--------------
 include/linux/coda.h | 43 -------------------------------------------
 5 files changed, 3 insertions(+), 79 deletions(-)

diff --git a/fs/Kconfig b/fs/Kconfig
index 37db79a2ff95..ed563b9e352a 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -2093,20 +2093,6 @@ config CODA_FS
 	  To compile the coda client support as a module, choose M here: the
 	  module will be called coda.
 
-config CODA_FS_OLD_API
-	bool "Use 96-bit Coda file identifiers"
-	depends on CODA_FS
-	help
-	  A new kernel-userspace API had to be introduced for Coda v6.0
-	  to support larger 128-bit file identifiers as needed by the
-	  new realms implementation.
-
-	  However this new API is not backward compatible with older
-	  clients. If you really need to run the old Coda userspace
-	  cache manager then say Y.
-
-	  For most cases you probably want to say N.
-
 config AFS_FS
 	tristate "Andrew File System support (AFS) (EXPERIMENTAL)"
 	depends on INET && EXPERIMENTAL
diff --git a/fs/coda/coda_linux.c b/fs/coda/coda_linux.c
index e1c854890f94..bf4a3fd3c8e3 100644
--- a/fs/coda/coda_linux.c
+++ b/fs/coda/coda_linux.c
@@ -28,11 +28,9 @@ int coda_fake_statfs;
 char * coda_f2s(struct CodaFid *f)
 {
 	static char s[60];
-#ifdef CONFIG_CODA_FS_OLD_API
- 	sprintf(s, "(%08x.%08x.%08x)", f->opaque[0], f->opaque[1], f->opaque[2]);
-#else
+
  	sprintf(s, "(%08x.%08x.%08x.%08x)", f->opaque[0], f->opaque[1], f->opaque[2], f->opaque[3]);
-#endif
+
 	return s;
 }
 
diff --git a/fs/coda/psdev.c b/fs/coda/psdev.c
index 40c36f7352a6..0d9b80ec689c 100644
--- a/fs/coda/psdev.c
+++ b/fs/coda/psdev.c
@@ -378,11 +378,7 @@ MODULE_AUTHOR("Jan Harkes, Peter J. Braam");
 MODULE_DESCRIPTION("Coda Distributed File System VFS interface");
 MODULE_ALIAS_CHARDEV_MAJOR(CODA_PSDEV_MAJOR);
 MODULE_LICENSE("GPL");
-#ifdef CONFIG_CODA_FS_OLD_API
-MODULE_VERSION("5.3.21");
-#else
 MODULE_VERSION("6.6");
-#endif
 
 static int __init init_coda(void)
 {
diff --git a/fs/coda/upcall.c b/fs/coda/upcall.c
index 359e531094dd..ce432bca95d1 100644
--- a/fs/coda/upcall.c
+++ b/fs/coda/upcall.c
@@ -52,12 +52,8 @@ static void *alloc_upcall(int opcode, int size)
         inp->ih.opcode = opcode;
 	inp->ih.pid = current->pid;
 	inp->ih.pgid = task_pgrp_nr(current);
-#ifdef CONFIG_CODA_FS_OLD_API
-	memset(&inp->ih.cred, 0, sizeof(struct coda_cred));
-	inp->ih.cred.cr_fsuid = current->fsuid;
-#else
 	inp->ih.uid = current->fsuid;
-#endif
+
 	return (void*)inp;
 }
 
@@ -166,20 +162,11 @@ int venus_close(struct super_block *sb, struct CodaFid *fid, int flags,
 	union inputArgs *inp;
 	union outputArgs *outp;
 	int insize, outsize, error;
-#ifdef CONFIG_CODA_FS_OLD_API
-	struct coda_cred cred = { 0, };
-	cred.cr_fsuid = uid;
-#endif
 	
 	insize = SIZE(release);
 	UPARG(CODA_CLOSE);
 	
-#ifdef CONFIG_CODA_FS_OLD_API
-	memcpy(&(inp->ih.cred), &cred, sizeof(cred));
-#else
 	inp->ih.uid = uid;
-#endif
-	
         inp->coda_close.VFid = *fid;
         inp->coda_close.flags = flags;
 
diff --git a/include/linux/coda.h b/include/linux/coda.h
index b5cf0780c51a..96c87693800b 100644
--- a/include/linux/coda.h
+++ b/include/linux/coda.h
@@ -199,28 +199,6 @@ typedef u_int32_t vuid_t;
 typedef u_int32_t vgid_t;
 #endif /*_VUID_T_ */
 
-#ifdef CONFIG_CODA_FS_OLD_API
-struct CodaFid {
-	u_int32_t opaque[3];
-};
-
-static __inline__ ino_t  coda_f2i(struct CodaFid *fid)
-{
-	if ( ! fid ) 
-		return 0; 
-	if (fid->opaque[1] == 0xfffffffe || fid->opaque[1] == 0xffffffff)
-		return ((fid->opaque[0] << 20) | (fid->opaque[2] & 0xfffff));
-	else
-		return (fid->opaque[2] + (fid->opaque[1]<<10) + (fid->opaque[0]<<20));
-}
-
-struct coda_cred {
-    vuid_t cr_uid, cr_euid, cr_suid, cr_fsuid; /* Real, efftve, set, fs uid*/
-    vgid_t cr_groupid, cr_egid, cr_sgid, cr_fsgid; /* same for groups */
-};
-
-#else /* not defined(CONFIG_CODA_FS_OLD_API) */
-
 struct CodaFid {
 	u_int32_t opaque[4];
 };
@@ -228,8 +206,6 @@ struct CodaFid {
 #define coda_f2i(fid)\
 	(fid ? (fid->opaque[3] ^ (fid->opaque[2]<<10) ^ (fid->opaque[1]<<20) ^ fid->opaque[0]) : 0)
 
-#endif
-
 #ifndef _VENUS_VATTR_T_
 #define _VENUS_VATTR_T_
 /*
@@ -313,15 +289,7 @@ struct coda_statfs {
 
 #define CIOC_KERNEL_VERSION _IOWR('c', 10, size_t)
 
-#if 0
-#define CODA_KERNEL_VERSION 0 /* don't care about kernel version number */
-#define CODA_KERNEL_VERSION 1 /* The old venus 4.6 compatible interface */
-#endif
-#ifdef CONFIG_CODA_FS_OLD_API
-#define CODA_KERNEL_VERSION 2 /* venus_lookup got an extra parameter */
-#else
 #define CODA_KERNEL_VERSION 3 /* 128-bit file identifiers */
-#endif
 
 /*
  *        Venus <-> Coda  RPC arguments
@@ -329,16 +297,9 @@ struct coda_statfs {
 struct coda_in_hdr {
     u_int32_t opcode;
     u_int32_t unique;	    /* Keep multiple outstanding msgs distinct */
-#ifdef CONFIG_CODA_FS_OLD_API
-    u_int16_t pid;	    /* Common to all */
-    u_int16_t pgid;	    /* Common to all */
-    u_int16_t sid;          /* Common to all */
-    struct coda_cred cred;  /* Common to all */
-#else
     pid_t pid;
     pid_t pgid;
     vuid_t uid;
-#endif
 };
 
 /* Really important that opcode and unique are 1st two fields! */
@@ -613,11 +574,7 @@ struct coda_vget_out {
 /* CODA_PURGEUSER is a venus->kernel call */
 struct coda_purgeuser_out {
     struct coda_out_hdr oh;
-#ifdef CONFIG_CODA_FS_OLD_API
-    struct coda_cred cred;
-#else
     vuid_t uid;
-#endif
 };
 
 /* coda_zapfile: */

From 3084b72de73a6f8af0f16989ffb348068bd066d4 Mon Sep 17 00:00:00 2001
From: Matthias Kaehlcke <matthias@kaehlcke.net>
Date: Fri, 25 Jul 2008 01:46:34 -0700
Subject: [PATCH 0751/1171] hfs: convert bitmap_lock in a mutex

Apple Macintosh file system: The semaphore bitmap_lock is used as a mutex.
Convert it to the mutex API

Signed-off-by: Matthias Kaehlcke <matthias@kaehlcke.net>
Cc: Roman Zippel <zippel@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/hfs/bitmap.c | 8 ++++----
 fs/hfs/hfs_fs.h | 3 ++-
 fs/hfs/super.c  | 2 +-
 3 files changed, 7 insertions(+), 6 deletions(-)

diff --git a/fs/hfs/bitmap.c b/fs/hfs/bitmap.c
index 24e75798ddf0..c6e97366e8ac 100644
--- a/fs/hfs/bitmap.c
+++ b/fs/hfs/bitmap.c
@@ -145,7 +145,7 @@ u32 hfs_vbm_search_free(struct super_block *sb, u32 goal, u32 *num_bits)
 	if (!*num_bits)
 		return 0;
 
-	down(&HFS_SB(sb)->bitmap_lock);
+	mutex_lock(&HFS_SB(sb)->bitmap_lock);
 	bitmap = HFS_SB(sb)->bitmap;
 
 	pos = hfs_find_set_zero_bits(bitmap, HFS_SB(sb)->fs_ablocks, goal, num_bits);
@@ -162,7 +162,7 @@ u32 hfs_vbm_search_free(struct super_block *sb, u32 goal, u32 *num_bits)
 	HFS_SB(sb)->free_ablocks -= *num_bits;
 	hfs_bitmap_dirty(sb);
 out:
-	up(&HFS_SB(sb)->bitmap_lock);
+	mutex_unlock(&HFS_SB(sb)->bitmap_lock);
 	return pos;
 }
 
@@ -205,7 +205,7 @@ int hfs_clear_vbm_bits(struct super_block *sb, u16 start, u16 count)
 	if ((start + count) > HFS_SB(sb)->fs_ablocks)
 		return -2;
 
-	down(&HFS_SB(sb)->bitmap_lock);
+	mutex_lock(&HFS_SB(sb)->bitmap_lock);
 	/* bitmap is always on a 32-bit boundary */
 	curr = HFS_SB(sb)->bitmap + (start / 32);
 	len = count;
@@ -236,7 +236,7 @@ int hfs_clear_vbm_bits(struct super_block *sb, u16 start, u16 count)
 	}
 out:
 	HFS_SB(sb)->free_ablocks += len;
-	up(&HFS_SB(sb)->bitmap_lock);
+	mutex_unlock(&HFS_SB(sb)->bitmap_lock);
 	hfs_bitmap_dirty(sb);
 
 	return 0;
diff --git a/fs/hfs/hfs_fs.h b/fs/hfs/hfs_fs.h
index 147374b6f675..ad6528819117 100644
--- a/fs/hfs/hfs_fs.h
+++ b/fs/hfs/hfs_fs.h
@@ -11,6 +11,7 @@
 
 #include <linux/slab.h>
 #include <linux/types.h>
+#include <linux/mutex.h>
 #include <linux/buffer_head.h>
 #include <linux/fs.h>
 
@@ -139,7 +140,7 @@ struct hfs_sb_info {
 
 	struct nls_table *nls_io, *nls_disk;
 
-	struct semaphore bitmap_lock;
+	struct mutex bitmap_lock;
 
 	unsigned long flags;
 
diff --git a/fs/hfs/super.c b/fs/hfs/super.c
index 8cf67974adf6..ac2ec5ef66e4 100644
--- a/fs/hfs/super.c
+++ b/fs/hfs/super.c
@@ -372,7 +372,7 @@ static int hfs_fill_super(struct super_block *sb, void *data, int silent)
 
 	sb->s_op = &hfs_super_operations;
 	sb->s_flags |= MS_NODIRATIME;
-	init_MUTEX(&sbi->bitmap_lock);
+	mutex_init(&sbi->bitmap_lock);
 
 	res = hfs_mdb_get(sb);
 	if (res) {

From 39f8d472f280dee6503a364d1d911b9e20ce3ec9 Mon Sep 17 00:00:00 2001
From: Matthias Kaehlcke <matthias@kaehlcke.net>
Date: Fri, 25 Jul 2008 01:46:35 -0700
Subject: [PATCH 0752/1171] hfs: convert extents_lock in a mutex

Apple Macintosh file system: The semaphore extens_lock is used as a mutex.
Convert it to the mutex API

Signed-off-by: Matthias Kaehlcke <matthias@kaehlcke.net>
Cc: Roman Zippel <zippel@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/hfs/btree.c  |  2 +-
 fs/hfs/extent.c | 14 +++++++-------
 fs/hfs/hfs_fs.h |  2 +-
 fs/hfs/inode.c  |  4 ++--
 4 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/fs/hfs/btree.c b/fs/hfs/btree.c
index f6621a785202..9b9d6395bad3 100644
--- a/fs/hfs/btree.c
+++ b/fs/hfs/btree.c
@@ -40,7 +40,7 @@ struct hfs_btree *hfs_btree_open(struct super_block *sb, u32 id, btree_keycmp ke
 	{
 	struct hfs_mdb *mdb = HFS_SB(sb)->mdb;
 	HFS_I(tree->inode)->flags = 0;
-	init_MUTEX(&HFS_I(tree->inode)->extents_lock);
+	mutex_init(&HFS_I(tree->inode)->extents_lock);
 	switch (id) {
 	case HFS_EXT_CNID:
 		hfs_inode_read_fork(tree->inode, mdb->drXTExtRec, mdb->drXTFlSize,
diff --git a/fs/hfs/extent.c b/fs/hfs/extent.c
index c176f67ba0a5..2c16316d2917 100644
--- a/fs/hfs/extent.c
+++ b/fs/hfs/extent.c
@@ -343,16 +343,16 @@ int hfs_get_block(struct inode *inode, sector_t block,
 		goto done;
 	}
 
-	down(&HFS_I(inode)->extents_lock);
+	mutex_lock(&HFS_I(inode)->extents_lock);
 	res = hfs_ext_read_extent(inode, ablock);
 	if (!res)
 		dblock = hfs_ext_find_block(HFS_I(inode)->cached_extents,
 					    ablock - HFS_I(inode)->cached_start);
 	else {
-		up(&HFS_I(inode)->extents_lock);
+		mutex_unlock(&HFS_I(inode)->extents_lock);
 		return -EIO;
 	}
-	up(&HFS_I(inode)->extents_lock);
+	mutex_unlock(&HFS_I(inode)->extents_lock);
 
 done:
 	map_bh(bh_result, sb, HFS_SB(sb)->fs_start +
@@ -375,7 +375,7 @@ int hfs_extend_file(struct inode *inode)
 	u32 start, len, goal;
 	int res;
 
-	down(&HFS_I(inode)->extents_lock);
+	mutex_lock(&HFS_I(inode)->extents_lock);
 	if (HFS_I(inode)->alloc_blocks == HFS_I(inode)->first_blocks)
 		goal = hfs_ext_lastblock(HFS_I(inode)->first_extents);
 	else {
@@ -425,7 +425,7 @@ int hfs_extend_file(struct inode *inode)
 			goto insert_extent;
 	}
 out:
-	up(&HFS_I(inode)->extents_lock);
+	mutex_unlock(&HFS_I(inode)->extents_lock);
 	if (!res) {
 		HFS_I(inode)->alloc_blocks += len;
 		mark_inode_dirty(inode);
@@ -487,7 +487,7 @@ void hfs_file_truncate(struct inode *inode)
 	if (blk_cnt == alloc_cnt)
 		goto out;
 
-	down(&HFS_I(inode)->extents_lock);
+	mutex_lock(&HFS_I(inode)->extents_lock);
 	hfs_find_init(HFS_SB(sb)->ext_tree, &fd);
 	while (1) {
 		if (alloc_cnt == HFS_I(inode)->first_blocks) {
@@ -514,7 +514,7 @@ void hfs_file_truncate(struct inode *inode)
 		hfs_brec_remove(&fd);
 	}
 	hfs_find_exit(&fd);
-	up(&HFS_I(inode)->extents_lock);
+	mutex_unlock(&HFS_I(inode)->extents_lock);
 
 	HFS_I(inode)->alloc_blocks = blk_cnt;
 out:
diff --git a/fs/hfs/hfs_fs.h b/fs/hfs/hfs_fs.h
index ad6528819117..9955232fdf8c 100644
--- a/fs/hfs/hfs_fs.h
+++ b/fs/hfs/hfs_fs.h
@@ -54,7 +54,7 @@ struct hfs_inode_info {
 	struct list_head open_dir_list;
 	struct inode *rsrc_inode;
 
-	struct semaphore extents_lock;
+	struct mutex extents_lock;
 
 	u16 alloc_blocks, clump_blocks;
 	sector_t fs_blocks;
diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c
index 97f8446c4ff4..dc4ec640e875 100644
--- a/fs/hfs/inode.c
+++ b/fs/hfs/inode.c
@@ -150,7 +150,7 @@ struct inode *hfs_new_inode(struct inode *dir, struct qstr *name, int mode)
 	if (!inode)
 		return NULL;
 
-	init_MUTEX(&HFS_I(inode)->extents_lock);
+	mutex_init(&HFS_I(inode)->extents_lock);
 	INIT_LIST_HEAD(&HFS_I(inode)->open_dir_list);
 	hfs_cat_build_key(sb, (btree_key *)&HFS_I(inode)->cat_key, dir->i_ino, name);
 	inode->i_ino = HFS_SB(sb)->next_id++;
@@ -281,7 +281,7 @@ static int hfs_read_inode(struct inode *inode, void *data)
 
 	HFS_I(inode)->flags = 0;
 	HFS_I(inode)->rsrc_inode = NULL;
-	init_MUTEX(&HFS_I(inode)->extents_lock);
+	mutex_init(&HFS_I(inode)->extents_lock);
 	INIT_LIST_HEAD(&HFS_I(inode)->open_dir_list);
 
 	/* Initialize the inode */

From 895c23f8c39c0c8d7b536bb2566d4aa968d78be2 Mon Sep 17 00:00:00 2001
From: Matthias Kaehlcke <matthias@kaehlcke.net>
Date: Fri, 25 Jul 2008 01:46:36 -0700
Subject: [PATCH 0753/1171] hfsplus: convert the extents_lock in a mutex

Apple Extended HFS file system: The semaphore extents lock is used as a
mutex.  Convert it to the mutex API.

Signed-off-by: Matthias Kaehlcke <matthias@kaehlcke.net>
Cc: Roman Zippel <zippel@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/hfsplus/extents.c    | 14 +++++++-------
 fs/hfsplus/hfsplus_fs.h |  3 ++-
 fs/hfsplus/inode.c      |  4 ++--
 fs/hfsplus/super.c      |  2 +-
 4 files changed, 12 insertions(+), 11 deletions(-)

diff --git a/fs/hfsplus/extents.c b/fs/hfsplus/extents.c
index 12e899cd7886..fec8f61227ff 100644
--- a/fs/hfsplus/extents.c
+++ b/fs/hfsplus/extents.c
@@ -199,16 +199,16 @@ int hfsplus_get_block(struct inode *inode, sector_t iblock,
 		goto done;
 	}
 
-	down(&HFSPLUS_I(inode).extents_lock);
+	mutex_lock(&HFSPLUS_I(inode).extents_lock);
 	res = hfsplus_ext_read_extent(inode, ablock);
 	if (!res) {
 		dblock = hfsplus_ext_find_block(HFSPLUS_I(inode).cached_extents, ablock -
 					     HFSPLUS_I(inode).cached_start);
 	} else {
-		up(&HFSPLUS_I(inode).extents_lock);
+		mutex_unlock(&HFSPLUS_I(inode).extents_lock);
 		return -EIO;
 	}
-	up(&HFSPLUS_I(inode).extents_lock);
+	mutex_unlock(&HFSPLUS_I(inode).extents_lock);
 
 done:
 	dprint(DBG_EXTENT, "get_block(%lu): %llu - %u\n", inode->i_ino, (long long)iblock, dblock);
@@ -355,7 +355,7 @@ int hfsplus_file_extend(struct inode *inode)
 		return -ENOSPC;
 	}
 
-	down(&HFSPLUS_I(inode).extents_lock);
+	mutex_lock(&HFSPLUS_I(inode).extents_lock);
 	if (HFSPLUS_I(inode).alloc_blocks == HFSPLUS_I(inode).first_blocks)
 		goal = hfsplus_ext_lastblock(HFSPLUS_I(inode).first_extents);
 	else {
@@ -408,7 +408,7 @@ int hfsplus_file_extend(struct inode *inode)
 			goto insert_extent;
 	}
 out:
-	up(&HFSPLUS_I(inode).extents_lock);
+	mutex_unlock(&HFSPLUS_I(inode).extents_lock);
 	if (!res) {
 		HFSPLUS_I(inode).alloc_blocks += len;
 		mark_inode_dirty(inode);
@@ -465,7 +465,7 @@ void hfsplus_file_truncate(struct inode *inode)
 	if (blk_cnt == alloc_cnt)
 		goto out;
 
-	down(&HFSPLUS_I(inode).extents_lock);
+	mutex_lock(&HFSPLUS_I(inode).extents_lock);
 	hfs_find_init(HFSPLUS_SB(sb).ext_tree, &fd);
 	while (1) {
 		if (alloc_cnt == HFSPLUS_I(inode).first_blocks) {
@@ -492,7 +492,7 @@ void hfsplus_file_truncate(struct inode *inode)
 		hfs_brec_remove(&fd);
 	}
 	hfs_find_exit(&fd);
-	up(&HFSPLUS_I(inode).extents_lock);
+	mutex_unlock(&HFSPLUS_I(inode).extents_lock);
 
 	HFSPLUS_I(inode).alloc_blocks = blk_cnt;
 out:
diff --git a/fs/hfsplus/hfsplus_fs.h b/fs/hfsplus/hfsplus_fs.h
index 9e59537b43d5..f027a905225f 100644
--- a/fs/hfsplus/hfsplus_fs.h
+++ b/fs/hfsplus/hfsplus_fs.h
@@ -11,6 +11,7 @@
 #define _LINUX_HFSPLUS_FS_H
 
 #include <linux/fs.h>
+#include <linux/mutex.h>
 #include <linux/buffer_head.h>
 #include "hfsplus_raw.h"
 
@@ -154,7 +155,7 @@ struct hfsplus_sb_info {
 
 
 struct hfsplus_inode_info {
-	struct semaphore extents_lock;
+	struct mutex extents_lock;
 	u32 clump_blocks, alloc_blocks;
 	sector_t fs_blocks;
 	/* Allocation extents from catalog record or volume header */
diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c
index 67e1c8b467c4..cc3b5e24339b 100644
--- a/fs/hfsplus/inode.c
+++ b/fs/hfsplus/inode.c
@@ -163,7 +163,7 @@ static struct dentry *hfsplus_file_lookup(struct inode *dir, struct dentry *dent
 
 	inode->i_ino = dir->i_ino;
 	INIT_LIST_HEAD(&HFSPLUS_I(inode).open_dir_list);
-	init_MUTEX(&HFSPLUS_I(inode).extents_lock);
+	mutex_init(&HFSPLUS_I(inode).extents_lock);
 	HFSPLUS_I(inode).flags = HFSPLUS_FLG_RSRC;
 
 	hfs_find_init(HFSPLUS_SB(sb).cat_tree, &fd);
@@ -316,7 +316,7 @@ struct inode *hfsplus_new_inode(struct super_block *sb, int mode)
 	inode->i_nlink = 1;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC;
 	INIT_LIST_HEAD(&HFSPLUS_I(inode).open_dir_list);
-	init_MUTEX(&HFSPLUS_I(inode).extents_lock);
+	mutex_init(&HFSPLUS_I(inode).extents_lock);
 	atomic_set(&HFSPLUS_I(inode).opencnt, 0);
 	HFSPLUS_I(inode).flags = 0;
 	memset(HFSPLUS_I(inode).first_extents, 0, sizeof(hfsplus_extent_rec));
diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c
index ce97a54518d8..3859118531c7 100644
--- a/fs/hfsplus/super.c
+++ b/fs/hfsplus/super.c
@@ -34,7 +34,7 @@ struct inode *hfsplus_iget(struct super_block *sb, unsigned long ino)
 		return inode;
 
 	INIT_LIST_HEAD(&HFSPLUS_I(inode).open_dir_list);
-	init_MUTEX(&HFSPLUS_I(inode).extents_lock);
+	mutex_init(&HFSPLUS_I(inode).extents_lock);
 	HFSPLUS_I(inode).flags = 0;
 	HFSPLUS_I(inode).rsrc_inode = NULL;
 	atomic_set(&HFSPLUS_I(inode).opencnt, 0);

From 5d4f7fddf8882b214e4aabb3bdb37f90a72b2537 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Fri, 25 Jul 2008 01:46:36 -0700
Subject: [PATCH 0754/1171] reiserfs: fix synchronization of quota files in
 journal=data mode

In journal=data mode, it is not enough to do write_inode_now() as done in
vfs_quota_on() to write all data to their final location (which is needed for
quota_read to work correctly).  Calling journal_end_sync() before calling
vfs_quota_on() does it's job because transactions are committed to the journal
and data marked as dirty in memory so write_inode_now() writes them to their
final locations.

Cc: <reiserfs-devel@vger.kernel.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/reiserfs/super.c | 30 +++++++++++++++++++++---------
 1 file changed, 21 insertions(+), 9 deletions(-)

diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index 1d40f2bd1970..0cbf4cd11147 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -2026,6 +2026,7 @@ static int reiserfs_quota_on(struct super_block *sb, int type, int format_id,
 	int err;
 	struct nameidata nd;
 	struct inode *inode;
+	struct reiserfs_transaction_handle th;
 
 	if (!(REISERFS_SB(sb)->s_mount_opt & (1 << REISERFS_QUOTA)))
 		return -EINVAL;
@@ -2053,17 +2054,28 @@ static int reiserfs_quota_on(struct super_block *sb, int type, int format_id,
 		}
 		mark_inode_dirty(inode);
 	}
-	/* Not journalling quota? No more tests needed... */
-	if (!REISERFS_SB(sb)->s_qf_names[USRQUOTA] &&
-	    !REISERFS_SB(sb)->s_qf_names[GRPQUOTA]) {
-		path_put(&nd.path);
-		return vfs_quota_on(sb, type, format_id, path, 0);
-	}
-	/* Quotafile not of fs root? */
-	if (nd.path.dentry->d_parent->d_inode != sb->s_root->d_inode)
-		reiserfs_warning(sb,
+	/* Journaling quota? */
+	if (REISERFS_SB(sb)->s_qf_names[type]) {
+		/* Quotafile not of fs root? */
+		if (nd.path.dentry->d_parent->d_inode != sb->s_root->d_inode)
+			reiserfs_warning(sb,
 				 "reiserfs: Quota file not on filesystem root. "
 				 "Journalled quota will not work.");
+	}
+
+	/*
+	 * When we journal data on quota file, we have to flush journal to see
+	 * all updates to the file when we bypass pagecache...
+	 */
+	if (reiserfs_file_data_log(inode)) {
+		/* Just start temporary transaction and finish it */
+		err = journal_begin(&th, sb, 1);
+		if (err)
+			return err;
+		err = journal_end_sync(&th, sb, 1);
+		if (err)
+			return err;
+	}
 	path_put(&nd.path);
 	return vfs_quota_on(sb, type, format_id, path, 0);
 }

From 4506567b24d3ea707e46e8aad64caef539382f4b Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Fri, 25 Jul 2008 01:46:37 -0700
Subject: [PATCH 0755/1171] reiserfs: fix typos in messages and comments
 (journalled -> journaled)

Cc: <reiserfs-devel@vger.kernel.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/reiserfs/super.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index 0cbf4cd11147..f723604c5d9b 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -182,7 +182,7 @@ static int finish_unfinished(struct super_block *s)
 			int ret = reiserfs_quota_on_mount(s, i);
 			if (ret < 0)
 				reiserfs_warning(s,
-						 "reiserfs: cannot turn on journalled quota: error %d",
+						 "reiserfs: cannot turn on journaled quota: error %d",
 						 ret);
 		}
 	}
@@ -994,7 +994,7 @@ static int reiserfs_parse_options(struct super_block *s, char *options,	/* strin
 
 			if (sb_any_quota_enabled(s)) {
 				reiserfs_warning(s,
-						 "reiserfs_parse_options: cannot change journalled quota options when quota turned on.");
+						 "reiserfs_parse_options: cannot change journaled quota options when quota turned on.");
 				return 0;
 			}
 			if (*arg) {	/* Some filename specified? */
@@ -1039,7 +1039,7 @@ static int reiserfs_parse_options(struct super_block *s, char *options,	/* strin
 #else
 		if (c == 'u' || c == 'g' || c == 'f') {
 			reiserfs_warning(s,
-					 "reiserfs_parse_options: journalled quota options not supported.");
+					 "reiserfs_parse_options: journaled quota options not supported.");
 			return 0;
 		}
 #endif
@@ -1050,7 +1050,7 @@ static int reiserfs_parse_options(struct super_block *s, char *options,	/* strin
 	    && (REISERFS_SB(s)->s_qf_names[USRQUOTA]
 		|| REISERFS_SB(s)->s_qf_names[GRPQUOTA])) {
 		reiserfs_warning(s,
-				 "reiserfs_parse_options: journalled quota format not specified.");
+				 "reiserfs_parse_options: journaled quota format not specified.");
 		return 0;
 	}
 	/* This checking is not precise wrt the quota type but for our purposes it is sufficient */
@@ -1980,7 +1980,7 @@ static int reiserfs_release_dquot(struct dquot *dquot)
 
 static int reiserfs_mark_dquot_dirty(struct dquot *dquot)
 {
-	/* Are we journalling quotas? */
+	/* Are we journaling quotas? */
 	if (REISERFS_SB(dquot->dq_sb)->s_qf_names[USRQUOTA] ||
 	    REISERFS_SB(dquot->dq_sb)->s_qf_names[GRPQUOTA]) {
 		dquot_mark_dquot_dirty(dquot);

From 00b441970a0ab48185244300ac7d4e4eb76df692 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Fri, 25 Jul 2008 01:46:38 -0700
Subject: [PATCH 0756/1171] reiserfs: correct mount option parsing to detect
 when quota options can be changed

We should not allow user to change quota mount options when quota is just
suspended.  It would make mount options and internal quota state inconsistent.

Also we should not allow user to change quota format when quota is turned on.
On the other hand we can just silently ignore when some option is set to the
value it already has (some mount versions do this on remount).  Finally, we
should not discard current quota options if parsing of mount options fails.

Cc: <reiserfs-devel@vger.kernel.org>
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/reiserfs/super.c | 83 ++++++++++++++++++++++++++++++++-------------
 1 file changed, 60 insertions(+), 23 deletions(-)

diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index f723604c5d9b..a10a6d2a8870 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -876,7 +876,9 @@ static int reiserfs_parse_options(struct super_block *s, char *options,	/* strin
 				     mount options were selected. */
 				  unsigned long *blocks,	/* strtol-ed from NNN of resize=NNN */
 				  char **jdev_name,
-				  unsigned int *commit_max_age)
+				  unsigned int *commit_max_age,
+				  char **qf_names,
+				  unsigned int *qfmt)
 {
 	int c;
 	char *arg = NULL;
@@ -992,7 +994,9 @@ static int reiserfs_parse_options(struct super_block *s, char *options,	/* strin
 		if (c == 'u' || c == 'g') {
 			int qtype = c == 'u' ? USRQUOTA : GRPQUOTA;
 
-			if (sb_any_quota_enabled(s)) {
+			if ((sb_any_quota_enabled(s) ||
+			     sb_any_quota_suspended(s)) &&
+			    (!*arg != !REISERFS_SB(s)->s_qf_names[qtype])) {
 				reiserfs_warning(s,
 						 "reiserfs_parse_options: cannot change journaled quota options when quota turned on.");
 				return 0;
@@ -1011,30 +1015,39 @@ static int reiserfs_parse_options(struct super_block *s, char *options,	/* strin
 							 "reiserfs_parse_options: quotafile must be on filesystem root.");
 					return 0;
 				}
-				REISERFS_SB(s)->s_qf_names[qtype] =
+				qf_names[qtype] =
 				    kmalloc(strlen(arg) + 1, GFP_KERNEL);
-				if (!REISERFS_SB(s)->s_qf_names[qtype]) {
+				if (!qf_names[qtype]) {
 					reiserfs_warning(s,
 							 "reiserfs_parse_options: not enough memory for storing quotafile name.");
 					return 0;
 				}
-				strcpy(REISERFS_SB(s)->s_qf_names[qtype], arg);
+				strcpy(qf_names[qtype], arg);
 				*mount_options |= 1 << REISERFS_QUOTA;
 			} else {
-				kfree(REISERFS_SB(s)->s_qf_names[qtype]);
-				REISERFS_SB(s)->s_qf_names[qtype] = NULL;
+				if (qf_names[qtype] !=
+				    REISERFS_SB(s)->s_qf_names[qtype])
+					kfree(qf_names[qtype]);
+				qf_names[qtype] = NULL;
 			}
 		}
 		if (c == 'f') {
 			if (!strcmp(arg, "vfsold"))
-				REISERFS_SB(s)->s_jquota_fmt = QFMT_VFS_OLD;
+				*qfmt = QFMT_VFS_OLD;
 			else if (!strcmp(arg, "vfsv0"))
-				REISERFS_SB(s)->s_jquota_fmt = QFMT_VFS_V0;
+				*qfmt = QFMT_VFS_V0;
 			else {
 				reiserfs_warning(s,
 						 "reiserfs_parse_options: unknown quota format specified.");
 				return 0;
 			}
+			if ((sb_any_quota_enabled(s) ||
+			     sb_any_quota_suspended(s)) &&
+			    *qfmt != REISERFS_SB(s)->s_jquota_fmt) {
+				reiserfs_warning(s,
+						 "reiserfs_parse_options: cannot change journaled quota options when quota turned on.");
+				return 0;
+			}
 		}
 #else
 		if (c == 'u' || c == 'g' || c == 'f') {
@@ -1046,9 +1059,8 @@ static int reiserfs_parse_options(struct super_block *s, char *options,	/* strin
 	}
 
 #ifdef CONFIG_QUOTA
-	if (!REISERFS_SB(s)->s_jquota_fmt
-	    && (REISERFS_SB(s)->s_qf_names[USRQUOTA]
-		|| REISERFS_SB(s)->s_qf_names[GRPQUOTA])) {
+	if (!REISERFS_SB(s)->s_jquota_fmt && !*qfmt
+	    && (qf_names[USRQUOTA] || qf_names[GRPQUOTA])) {
 		reiserfs_warning(s,
 				 "reiserfs_parse_options: journaled quota format not specified.");
 		return 0;
@@ -1130,6 +1142,21 @@ static void handle_attrs(struct super_block *s)
 	}
 }
 
+#ifdef CONFIG_QUOTA
+static void handle_quota_files(struct super_block *s, char **qf_names,
+			       unsigned int *qfmt)
+{
+	int i;
+
+	for (i = 0; i < MAXQUOTAS; i++) {
+		if (qf_names[i] != REISERFS_SB(s)->s_qf_names[i])
+			kfree(REISERFS_SB(s)->s_qf_names[i]);
+		REISERFS_SB(s)->s_qf_names[i] = qf_names[i];
+	}
+	REISERFS_SB(s)->s_jquota_fmt = *qfmt;
+}
+#endif
+
 static int reiserfs_remount(struct super_block *s, int *mount_flags, char *arg)
 {
 	struct reiserfs_super_block *rs;
@@ -1141,23 +1168,30 @@ static int reiserfs_remount(struct super_block *s, int *mount_flags, char *arg)
 	struct reiserfs_journal *journal = SB_JOURNAL(s);
 	char *new_opts = kstrdup(arg, GFP_KERNEL);
 	int err;
+	char *qf_names[MAXQUOTAS];
+	unsigned int qfmt = 0;
 #ifdef CONFIG_QUOTA
 	int i;
+
+	memcpy(qf_names, REISERFS_SB(s)->s_qf_names, sizeof(qf_names));
 #endif
 
 	rs = SB_DISK_SUPER_BLOCK(s);
 
 	if (!reiserfs_parse_options
-	    (s, arg, &mount_options, &blocks, NULL, &commit_max_age)) {
+	    (s, arg, &mount_options, &blocks, NULL, &commit_max_age,
+	    qf_names, &qfmt)) {
 #ifdef CONFIG_QUOTA
-		for (i = 0; i < MAXQUOTAS; i++) {
-			kfree(REISERFS_SB(s)->s_qf_names[i]);
-			REISERFS_SB(s)->s_qf_names[i] = NULL;
-		}
+		for (i = 0; i < MAXQUOTAS; i++)
+			if (qf_names[i] != REISERFS_SB(s)->s_qf_names[i])
+				kfree(qf_names[i]);
 #endif
 		err = -EINVAL;
 		goto out_err;
 	}
+#ifdef CONFIG_QUOTA
+	handle_quota_files(s, qf_names, &qfmt);
+#endif
 
 	handle_attrs(s);
 
@@ -1570,6 +1604,8 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent)
 	char *jdev_name;
 	struct reiserfs_sb_info *sbi;
 	int errval = -EINVAL;
+	char *qf_names[MAXQUOTAS] = {};
+	unsigned int qfmt = 0;
 
 	save_mount_options(s, data);
 
@@ -1597,9 +1633,12 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent)
 	jdev_name = NULL;
 	if (reiserfs_parse_options
 	    (s, (char *)data, &(sbi->s_mount_opt), &blocks, &jdev_name,
-	     &commit_max_age) == 0) {
+	     &commit_max_age, qf_names, &qfmt) == 0) {
 		goto error;
 	}
+#ifdef CONFIG_QUOTA
+	handle_quota_files(s, qf_names, &qfmt);
+#endif
 
 	if (blocks) {
 		SWARN(silent, s, "jmacd-7: reiserfs_fill_super: resize option "
@@ -1819,7 +1858,7 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent)
 
 	return (0);
 
-      error:
+error:
 	if (jinit_done) {	/* kill the commit thread, free journal ram */
 		journal_release_error(NULL, s);
 	}
@@ -1830,10 +1869,8 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent)
 #ifdef CONFIG_QUOTA
 	{
 		int j;
-		for (j = 0; j < MAXQUOTAS; j++) {
-			kfree(sbi->s_qf_names[j]);
-			sbi->s_qf_names[j] = NULL;
-		}
+		for (j = 0; j < MAXQUOTAS; j++)
+			kfree(qf_names[j]);
 	}
 #endif
 	kfree(sbi);

From f68215c4640a38d66429014e524a627bf572d26a Mon Sep 17 00:00:00 2001
From: Jeff Mahoney <jeffm@suse.com>
Date: Fri, 25 Jul 2008 01:46:38 -0700
Subject: [PATCH 0757/1171] reiserfs: convert j_lock to mutex

j_lock is a semaphore but uses it as if it were a mutex.  This patch converts
it to a mutex.

Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Chris Mason <chris.mason@oracle.com>
Cc: Edward Shishkin <edward.shishkin@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/reiserfs/journal.c          | 6 +++---
 include/linux/reiserfs_fs_sb.h | 2 +-
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c
index e396b2fa4743..0f7b1e807e60 100644
--- a/fs/reiserfs/journal.c
+++ b/fs/reiserfs/journal.c
@@ -558,13 +558,13 @@ static inline void insert_journal_hash(struct reiserfs_journal_cnode **table,
 static inline void lock_journal(struct super_block *p_s_sb)
 {
 	PROC_INFO_INC(p_s_sb, journal.lock_journal);
-	down(&SB_JOURNAL(p_s_sb)->j_lock);
+	mutex_lock(&SB_JOURNAL(p_s_sb)->j_mutex);
 }
 
 /* unlock the current transaction */
 static inline void unlock_journal(struct super_block *p_s_sb)
 {
-	up(&SB_JOURNAL(p_s_sb)->j_lock);
+	mutex_unlock(&SB_JOURNAL(p_s_sb)->j_mutex);
 }
 
 static inline void get_journal_list(struct reiserfs_journal_list *jl)
@@ -2837,7 +2837,7 @@ int journal_init(struct super_block *p_s_sb, const char *j_dev_name,
 	journal->j_last = NULL;
 	journal->j_first = NULL;
 	init_waitqueue_head(&(journal->j_join_wait));
-	sema_init(&journal->j_lock, 1);
+	mutex_init(&journal->j_mutex);
 	sema_init(&journal->j_flush_sem, 1);
 
 	journal->j_trans_id = 10;
diff --git a/include/linux/reiserfs_fs_sb.h b/include/linux/reiserfs_fs_sb.h
index 336ee43ed7d8..49b639b88bac 100644
--- a/include/linux/reiserfs_fs_sb.h
+++ b/include/linux/reiserfs_fs_sb.h
@@ -193,7 +193,7 @@ struct reiserfs_journal {
 	struct buffer_head *j_header_bh;
 
 	time_t j_trans_start_time;	/* time this transaction started */
-	struct semaphore j_lock;
+	struct mutex j_mutex;
 	struct semaphore j_flush_sem;
 	wait_queue_head_t j_join_wait;	/* wait for current transaction to finish before starting new one */
 	atomic_t j_jlock;	/* lock for j_join_wait */

From afe70259076fff0446001eaa1a287f615241a357 Mon Sep 17 00:00:00 2001
From: Jeff Mahoney <jeffm@suse.com>
Date: Fri, 25 Jul 2008 01:46:39 -0700
Subject: [PATCH 0758/1171] reiserfs: convert j_flush_sem to mutex

j_flush_sem is a semaphore but uses it as if it were a mutex.  This patch
converts it to a mutex.

[akpm@linux-foundation.org: fix mutex_trylock retval treatment]
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Chris Mason <chris.mason@oracle.com>
Cc: Edward Shishkin <edward.shishkin@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/reiserfs/journal.c          | 14 +++++++-------
 include/linux/reiserfs_fs_sb.h |  2 +-
 2 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c
index 0f7b1e807e60..3cb4a562030e 100644
--- a/fs/reiserfs/journal.c
+++ b/fs/reiserfs/journal.c
@@ -1411,8 +1411,8 @@ static int flush_journal_list(struct super_block *s,
 
 	/* if flushall == 0, the lock is already held */
 	if (flushall) {
-		down(&journal->j_flush_sem);
-	} else if (!down_trylock(&journal->j_flush_sem)) {
+		mutex_lock(&journal->j_flush_mutex);
+	} else if (mutex_trylock(&journal->j_flush_mutex)) {
 		BUG();
 	}
 
@@ -1642,7 +1642,7 @@ static int flush_journal_list(struct super_block *s,
 	jl->j_state = 0;
 	put_journal_list(s, jl);
 	if (flushall)
-		up(&journal->j_flush_sem);
+		mutex_unlock(&journal->j_flush_mutex);
 	put_fs_excl();
 	return err;
 }
@@ -1772,12 +1772,12 @@ static int kupdate_transactions(struct super_block *s,
 	struct reiserfs_journal *journal = SB_JOURNAL(s);
 	chunk.nr = 0;
 
-	down(&journal->j_flush_sem);
+	mutex_lock(&journal->j_flush_mutex);
 	if (!journal_list_still_alive(s, orig_trans_id)) {
 		goto done;
 	}
 
-	/* we've got j_flush_sem held, nobody is going to delete any
+	/* we've got j_flush_mutex held, nobody is going to delete any
 	 * of these lists out from underneath us
 	 */
 	while ((num_trans && transactions_flushed < num_trans) ||
@@ -1812,7 +1812,7 @@ static int kupdate_transactions(struct super_block *s,
 	}
 
       done:
-	up(&journal->j_flush_sem);
+	mutex_unlock(&journal->j_flush_mutex);
 	return ret;
 }
 
@@ -2838,7 +2838,7 @@ int journal_init(struct super_block *p_s_sb, const char *j_dev_name,
 	journal->j_first = NULL;
 	init_waitqueue_head(&(journal->j_join_wait));
 	mutex_init(&journal->j_mutex);
-	sema_init(&journal->j_flush_sem, 1);
+	mutex_init(&journal->j_flush_mutex);
 
 	journal->j_trans_id = 10;
 	journal->j_mount_id = 10;
diff --git a/include/linux/reiserfs_fs_sb.h b/include/linux/reiserfs_fs_sb.h
index 49b639b88bac..c0751724ee64 100644
--- a/include/linux/reiserfs_fs_sb.h
+++ b/include/linux/reiserfs_fs_sb.h
@@ -194,7 +194,7 @@ struct reiserfs_journal {
 
 	time_t j_trans_start_time;	/* time this transaction started */
 	struct mutex j_mutex;
-	struct semaphore j_flush_sem;
+	struct mutex j_flush_mutex;
 	wait_queue_head_t j_join_wait;	/* wait for current transaction to finish before starting new one */
 	atomic_t j_jlock;	/* lock for j_join_wait */
 	int j_list_bitmap_index;	/* number of next list bitmap to use */

From 90415deac75a761a25239af6f56381546f8d2201 Mon Sep 17 00:00:00 2001
From: Jeff Mahoney <jeffm@suse.com>
Date: Fri, 25 Jul 2008 01:46:40 -0700
Subject: [PATCH 0759/1171] reiserfs: convert j_commit_lock to mutex

j_commit_lock is a semaphore but uses it as if it were a mutex.  This patch
converts it to a mutex.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Chris Mason <chris.mason@oracle.com>
Cc: Edward Shishkin <edward.shishkin@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/reiserfs/journal.c          | 22 ++++++++++------------
 include/linux/reiserfs_fs_sb.h |  2 +-
 2 files changed, 11 insertions(+), 13 deletions(-)

diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c
index 3cb4a562030e..c8f60ee183b5 100644
--- a/fs/reiserfs/journal.c
+++ b/fs/reiserfs/journal.c
@@ -34,15 +34,10 @@
 **		        from within kupdate, it will ignore the immediate flag
 */
 
-#include <asm/uaccess.h>
-#include <asm/system.h>
-
 #include <linux/time.h>
 #include <linux/semaphore.h>
-
 #include <linux/vmalloc.h>
 #include <linux/reiserfs_fs.h>
-
 #include <linux/kernel.h>
 #include <linux/errno.h>
 #include <linux/fcntl.h>
@@ -54,6 +49,9 @@
 #include <linux/writeback.h>
 #include <linux/blkdev.h>
 #include <linux/backing-dev.h>
+#include <linux/uaccess.h>
+
+#include <asm/system.h>
 
 /* gets a struct reiserfs_journal_list * from a list head */
 #define JOURNAL_LIST_ENTRY(h) (list_entry((h), struct reiserfs_journal_list, \
@@ -1045,9 +1043,9 @@ static int flush_commit_list(struct super_block *s,
 	}
 
 	/* make sure nobody is trying to flush this one at the same time */
-	down(&jl->j_commit_lock);
+	mutex_lock(&jl->j_commit_mutex);
 	if (!journal_list_still_alive(s, trans_id)) {
-		up(&jl->j_commit_lock);
+		mutex_unlock(&jl->j_commit_mutex);
 		goto put_jl;
 	}
 	BUG_ON(jl->j_trans_id == 0);
@@ -1057,7 +1055,7 @@ static int flush_commit_list(struct super_block *s,
 		if (flushall) {
 			atomic_set(&(jl->j_older_commits_done), 1);
 		}
-		up(&jl->j_commit_lock);
+		mutex_unlock(&jl->j_commit_mutex);
 		goto put_jl;
 	}
 
@@ -1181,7 +1179,7 @@ static int flush_commit_list(struct super_block *s,
 	if (flushall) {
 		atomic_set(&(jl->j_older_commits_done), 1);
 	}
-	up(&jl->j_commit_lock);
+	mutex_unlock(&jl->j_commit_mutex);
       put_jl:
 	put_journal_list(s, jl);
 
@@ -2556,7 +2554,7 @@ static struct reiserfs_journal_list *alloc_journal_list(struct super_block *s)
 	INIT_LIST_HEAD(&jl->j_working_list);
 	INIT_LIST_HEAD(&jl->j_tail_bh_list);
 	INIT_LIST_HEAD(&jl->j_bh_list);
-	sema_init(&jl->j_commit_lock, 1);
+	mutex_init(&jl->j_commit_mutex);
 	SB_JOURNAL(s)->j_num_lists++;
 	get_journal_list(jl);
 	return jl;
@@ -4030,7 +4028,7 @@ static int do_journal_end(struct reiserfs_transaction_handle *th,
 	 * the new transaction is fully setup, and we've already flushed the
 	 * ordered bh list
 	 */
-	down(&jl->j_commit_lock);
+	mutex_lock(&jl->j_commit_mutex);
 
 	/* save the transaction id in case we need to commit it later */
 	commit_trans_id = jl->j_trans_id;
@@ -4196,7 +4194,7 @@ static int do_journal_end(struct reiserfs_transaction_handle *th,
 		lock_kernel();
 	}
 	BUG_ON(!list_empty(&jl->j_tail_bh_list));
-	up(&jl->j_commit_lock);
+	mutex_unlock(&jl->j_commit_mutex);
 
 	/* honor the flush wishes from the caller, simple commits can
 	 ** be done outside the journal lock, they are done below
diff --git a/include/linux/reiserfs_fs_sb.h b/include/linux/reiserfs_fs_sb.h
index c0751724ee64..315517e8bfa1 100644
--- a/include/linux/reiserfs_fs_sb.h
+++ b/include/linux/reiserfs_fs_sb.h
@@ -152,7 +152,7 @@ struct reiserfs_journal_list {
 	atomic_t j_nonzerolen;
 	atomic_t j_commit_left;
 	atomic_t j_older_commits_done;	/* all commits older than this on disk */
-	struct semaphore j_commit_lock;
+	struct mutex j_commit_mutex;
 	unsigned long j_trans_id;
 	time_t j_timestamp;
 	struct reiserfs_list_bitmap *j_list_bitmap;

From 3264d4ded4d916d294d776b77b72d477c63ac3be Mon Sep 17 00:00:00 2001
From: Shen Feng <shen@cn.fujitsu.com>
Date: Fri, 25 Jul 2008 01:46:41 -0700
Subject: [PATCH 0760/1171] reiserfs: remove double definitions of xattr macros

remove the definitions of macros:
XATTR_SECURITY_PREFIX
XATTR_TRUSTED_PREFIX
XATTR_USER_PREFIX
since they are defined in linux/xattr.h

Signed-off-by: Shen Feng <shen@cn.fujitsu.com>
Signed-off-by: Mingming Cao <cmm@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/reiserfs/xattr_security.c | 2 --
 fs/reiserfs/xattr_trusted.c  | 2 --
 fs/reiserfs/xattr_user.c     | 2 --
 3 files changed, 6 deletions(-)

diff --git a/fs/reiserfs/xattr_security.c b/fs/reiserfs/xattr_security.c
index 5e90a95ad60b..056008db1377 100644
--- a/fs/reiserfs/xattr_security.c
+++ b/fs/reiserfs/xattr_security.c
@@ -6,8 +6,6 @@
 #include <linux/reiserfs_xattr.h>
 #include <asm/uaccess.h>
 
-#define XATTR_SECURITY_PREFIX "security."
-
 static int
 security_get(struct inode *inode, const char *name, void *buffer, size_t size)
 {
diff --git a/fs/reiserfs/xattr_trusted.c b/fs/reiserfs/xattr_trusted.c
index 024a938ca60f..60abe2bb1f98 100644
--- a/fs/reiserfs/xattr_trusted.c
+++ b/fs/reiserfs/xattr_trusted.c
@@ -7,8 +7,6 @@
 #include <linux/reiserfs_xattr.h>
 #include <asm/uaccess.h>
 
-#define XATTR_TRUSTED_PREFIX "trusted."
-
 static int
 trusted_get(struct inode *inode, const char *name, void *buffer, size_t size)
 {
diff --git a/fs/reiserfs/xattr_user.c b/fs/reiserfs/xattr_user.c
index 073f39364b11..1384efcb938e 100644
--- a/fs/reiserfs/xattr_user.c
+++ b/fs/reiserfs/xattr_user.c
@@ -10,8 +10,6 @@
 # include <linux/reiserfs_acl.h>
 #endif
 
-#define XATTR_USER_PREFIX "user."
-
 static int
 user_get(struct inode *inode, const char *name, void *buffer, size_t size)
 {

From 8d44d9741f6808c107a144f469fb89e6fe7c55e3 Mon Sep 17 00:00:00 2001
From: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Date: Fri, 25 Jul 2008 01:46:41 -0700
Subject: [PATCH 0761/1171] fat: fix parse_options()

Current parse_options() exits too early.  We need to run the code of
bottom in this function even if users doesn't specify options.

Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/fat/inode.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index 46a4508ffd2e..60deb5fd1188 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -950,7 +950,7 @@ static int parse_options(char *options, int is_vfat, int silent, int *debug,
 	*debug = 0;
 
 	if (!options)
-		return 0;
+		goto out;
 
 	while ((p = strsep(&options, ",")) != NULL) {
 		int token;
@@ -1104,10 +1104,13 @@ static int parse_options(char *options, int is_vfat, int silent, int *debug,
 			return -EINVAL;
 		}
 	}
+
+out:
 	/* UTF-8 doesn't provide FAT semantics */
 	if (!strcmp(opts->iocharset, "utf8")) {
 		printk(KERN_ERR "FAT: utf8 is not a recommended IO charset"
-		       " for FAT filesystems, filesystem will be case sensitive!\n");
+		       " for FAT filesystems, filesystem will be "
+		       "case sensitive!\n");
 	}
 
 	/* If user doesn't specify allow_utime, it's initialized from dmask. */

From 4596c8aaf96e8634ca755c9f34b91420a39bebd4 Mon Sep 17 00:00:00 2001
From: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Date: Fri, 25 Jul 2008 01:46:42 -0700
Subject: [PATCH 0762/1171] fat: fix VFAT_IOCTL_READDIR_xxx and cleanup for
 userland

"struct dirent" is a kernel type here, but is a **different type** in
userspace!  This means both the structure and the IOCTL number is wrong!

So, this adds new "struct __fat_dirent" to generate correct IOCTL number.
And kernel stuff moves to under __KERNEL__.

Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/msdos_fs.h | 47 +++++++++++++++++++++++-----------------
 1 file changed, 27 insertions(+), 20 deletions(-)

diff --git a/include/linux/msdos_fs.h b/include/linux/msdos_fs.h
index 81cd36b735b0..5161394c7894 100644
--- a/include/linux/msdos_fs.h
+++ b/include/linux/msdos_fs.h
@@ -2,11 +2,11 @@
 #define _LINUX_MSDOS_FS_H
 
 #include <linux/magic.h>
+#include <asm/byteorder.h>
 
 /*
  * The MS-DOS filesystem constants/structures
  */
-#include <asm/byteorder.h>
 
 #define SECTOR_SIZE	512		/* sector size (bytes) */
 #define SECTOR_BITS	9		/* log2(SECTOR_SIZE) */
@@ -89,24 +89,22 @@
 #define IS_FSINFO(x)	(le32_to_cpu((x)->signature1) == FAT_FSINFO_SIG1 \
 			 && le32_to_cpu((x)->signature2) == FAT_FSINFO_SIG2)
 
+struct __fat_dirent {
+	long		d_ino;
+	__kernel_off_t	d_off;
+	unsigned short	d_reclen;
+	char		d_name[256]; /* We must not include limits.h! */
+};
+
 /*
  * ioctl commands
  */
-#define VFAT_IOCTL_READDIR_BOTH		_IOR('r', 1, struct dirent [2])
-#define VFAT_IOCTL_READDIR_SHORT	_IOR('r', 2, struct dirent [2])
+#define VFAT_IOCTL_READDIR_BOTH		_IOR('r', 1, struct __fat_dirent[2])
+#define VFAT_IOCTL_READDIR_SHORT	_IOR('r', 2, struct __fat_dirent[2])
 /* <linux/videotext.h> has used 0x72 ('r') in collision, so skip a few */
 #define FAT_IOCTL_GET_ATTRIBUTES	_IOR('r', 0x10, __u32)
 #define FAT_IOCTL_SET_ATTRIBUTES	_IOW('r', 0x11, __u32)
 
-/*
- * vfat shortname flags
- */
-#define VFAT_SFN_DISPLAY_LOWER	0x0001 /* convert to lowercase for display */
-#define VFAT_SFN_DISPLAY_WIN95	0x0002 /* emulate win95 rule for display */
-#define VFAT_SFN_DISPLAY_WINNT	0x0004 /* emulate winnt rule for display */
-#define VFAT_SFN_CREATE_WIN95	0x0100 /* emulate win95 rule for create */
-#define VFAT_SFN_CREATE_WINNT	0x0200 /* emulate winnt rule for create */
-
 struct fat_boot_sector {
 	__u8	ignored[3];	/* Boot strap short or near jump */
 	__u8	system_id[8];	/* Name - can be used to special case
@@ -168,14 +166,6 @@ struct msdos_dir_slot {
 	__u8    name11_12[4];	/* last 2 characters in name */
 };
 
-struct fat_slot_info {
-	loff_t i_pos;		/* on-disk position of directory entry */
-	loff_t slot_off;	/* offset for slot or de start */
-	int nr_slots;		/* number of slots + 1(de) in filename */
-	struct msdos_dir_entry *de;
-	struct buffer_head *bh;
-};
-
 #ifdef __KERNEL__
 
 #include <linux/buffer_head.h>
@@ -184,6 +174,15 @@ struct fat_slot_info {
 #include <linux/fs.h>
 #include <linux/mutex.h>
 
+/*
+ * vfat shortname flags
+ */
+#define VFAT_SFN_DISPLAY_LOWER	0x0001 /* convert to lowercase for display */
+#define VFAT_SFN_DISPLAY_WIN95	0x0002 /* emulate win95 rule for display */
+#define VFAT_SFN_DISPLAY_WINNT	0x0004 /* emulate winnt rule for display */
+#define VFAT_SFN_CREATE_WIN95	0x0100 /* emulate win95 rule for create */
+#define VFAT_SFN_CREATE_WINNT	0x0200 /* emulate winnt rule for create */
+
 struct fat_mount_options {
 	uid_t fs_uid;
 	gid_t fs_gid;
@@ -267,6 +266,14 @@ struct msdos_inode_info {
 	struct inode vfs_inode;
 };
 
+struct fat_slot_info {
+	loff_t i_pos;		/* on-disk position of directory entry */
+	loff_t slot_off;	/* offset for slot or de start */
+	int nr_slots;		/* number of slots + 1(de) in filename */
+	struct msdos_dir_entry *de;
+	struct buffer_head *bh;
+};
+
 static inline struct msdos_sb_info *MSDOS_SB(struct super_block *sb)
 {
 	return sb->s_fs_info;

From 531f710f8e68fd2bad7516a090bff372f5f9cf6d Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 01:46:43 -0700
Subject: [PATCH 0763/1171] fat/dir.c: switch to struct __fat_dirent

struct __fat_dirent is what was formerly the kernel struct dirent (that
was different from the userspace struct dirent).

Converting all fat users to struct __fat_dirent will allow us to get rid
of the conflicting struct dirent definition.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/fat/dir.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/fs/fat/dir.c b/fs/fat/dir.c
index 34541d06e626..b57c4b1db636 100644
--- a/fs/fat/dir.c
+++ b/fs/fat/dir.c
@@ -17,7 +17,6 @@
 #include <linux/slab.h>
 #include <linux/time.h>
 #include <linux/msdos_fs.h>
-#include <linux/dirent.h>
 #include <linux/smp_lock.h>
 #include <linux/buffer_head.h>
 #include <linux/compat.h>
@@ -715,7 +714,7 @@ efault:									   \
 	return -EFAULT;							   \
 }
 
-FAT_IOCTL_FILLDIR_FUNC(fat_ioctl_filldir, dirent)
+FAT_IOCTL_FILLDIR_FUNC(fat_ioctl_filldir, __fat_dirent)
 
 static int fat_ioctl_readdir(struct inode *inode, struct file *filp,
 			     void __user *dirent, filldir_t filldir,
@@ -741,7 +740,7 @@ static int fat_ioctl_readdir(struct inode *inode, struct file *filp,
 static int fat_dir_ioctl(struct inode *inode, struct file *filp,
 			 unsigned int cmd, unsigned long arg)
 {
-	struct dirent __user *d1 = (struct dirent __user *)arg;
+	struct __fat_dirent __user *d1 = (struct __fat_dirent __user *)arg;
 	int short_only, both;
 
 	switch (cmd) {
@@ -757,7 +756,7 @@ static int fat_dir_ioctl(struct inode *inode, struct file *filp,
 		return fat_generic_ioctl(inode, filp, cmd, arg);
 	}
 
-	if (!access_ok(VERIFY_WRITE, d1, sizeof(struct dirent[2])))
+	if (!access_ok(VERIFY_WRITE, d1, sizeof(struct __fat_dirent[2])))
 		return -EFAULT;
 	/*
 	 * Yes, we don't need this put_user() absolutely. However old

From d688611674cc9c265ee67e89d2ea8bf060c17e8d Mon Sep 17 00:00:00 2001
From: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Date: Fri, 25 Jul 2008 01:46:43 -0700
Subject: [PATCH 0764/1171] fat: cleanup fs/fat/dir.c

This is no logic changes, just cleans fs/fat/dir.c up.

Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/fat/dir.c | 133 ++++++++++++++++++++++++++-------------------------
 1 file changed, 68 insertions(+), 65 deletions(-)

diff --git a/fs/fat/dir.c b/fs/fat/dir.c
index b57c4b1db636..a4410740627f 100644
--- a/fs/fat/dir.c
+++ b/fs/fat/dir.c
@@ -123,10 +123,11 @@ static inline int fat_get_entry(struct inode *dir, loff_t *pos,
  * but ignore that right now.
  * Ahem... Stack smashing in ring 0 isn't fun. Fixed.
  */
-static int uni16_to_x8(unsigned char *ascii, wchar_t *uni, int len,
+static int uni16_to_x8(unsigned char *ascii, const wchar_t *uni, int len,
 		       int uni_xlate, struct nls_table *nls)
 {
-	wchar_t *ip, ec;
+	const wchar_t *ip;
+	wchar_t ec;
 	unsigned char *op, nc;
 	int charlen;
 	int k;
@@ -166,6 +167,16 @@ static int uni16_to_x8(unsigned char *ascii, wchar_t *uni, int len,
 	return (op - ascii);
 }
 
+static inline int fat_uni_to_x8(struct msdos_sb_info *sbi, const wchar_t *uni,
+				unsigned char *buf, int size)
+{
+	if (sbi->options.utf8)
+		return utf8_wcstombs(buf, uni, size);
+	else
+		return uni16_to_x8(buf, uni, size, sbi->options.unicode_xlate,
+				   sbi->nls_io);
+}
+
 static inline int
 fat_short2uni(struct nls_table *t, unsigned char *c, int clen, wchar_t *uni)
 {
@@ -226,6 +237,19 @@ fat_shortname2uni(struct nls_table *nls, unsigned char *buf, int buf_size,
 	return len;
 }
 
+static inline int fat_name_match(struct msdos_sb_info *sbi,
+				 const unsigned char *a, int a_len,
+				 const unsigned char *b, int b_len)
+{
+	if (a_len != b_len)
+		return 0;
+
+	if (sbi->options.name_check != 's')
+		return !nls_strnicmp(sbi->nls_io, a, b, a_len);
+	else
+		return !memcmp(a, b, a_len);
+}
+
 enum { PARSE_INVALID = 1, PARSE_NOT_LONGNAME, PARSE_EOF, };
 
 /**
@@ -311,29 +335,24 @@ int fat_search_long(struct inode *inode, const unsigned char *name,
 	struct msdos_sb_info *sbi = MSDOS_SB(sb);
 	struct buffer_head *bh = NULL;
 	struct msdos_dir_entry *de;
-	struct nls_table *nls_io = sbi->nls_io;
 	struct nls_table *nls_disk = sbi->nls_disk;
-	wchar_t bufuname[14];
 	unsigned char nr_slots;
-	int xlate_len;
+	wchar_t bufuname[14];
 	wchar_t *unicode = NULL;
 	unsigned char work[MSDOS_NAME];
 	unsigned char *bufname = NULL;
-	int uni_xlate = sbi->options.unicode_xlate;
-	int utf8 = sbi->options.utf8;
-	int anycase = (sbi->options.name_check != 's');
 	unsigned short opt_shortname = sbi->options.shortname;
 	loff_t cpos = 0;
-	int chl, i, j, last_u, err;
+	int chl, i, j, last_u, err, len;
 
 	bufname = __getname();
 	if (!bufname)
 		return -ENOMEM;
 
 	err = -ENOENT;
-	while(1) {
+	while (1) {
 		if (fat_get_entry(inode, &cpos, &bh, &de) == -1)
-			goto EODir;
+			goto end_of_dir;
 parse_record:
 		nr_slots = 0;
 		if (de->name[0] == DELETED_FLAG)
@@ -352,7 +371,7 @@ parse_record:
 			else if (status == PARSE_NOT_LONGNAME)
 				goto parse_record;
 			else if (status == PARSE_EOF)
-				goto EODir;
+				goto end_of_dir;
 		}
 
 		memcpy(work, de->name, sizeof(de->name));
@@ -393,30 +412,21 @@ parse_record:
 		if (!last_u)
 			continue;
 
+		/* Compare shortname */
 		bufuname[last_u] = 0x0000;
-		xlate_len = utf8
-			?utf8_wcstombs(bufname, bufuname, PATH_MAX)
-			:uni16_to_x8(bufname, bufuname, PATH_MAX, uni_xlate, nls_io);
-		if (xlate_len == name_len)
-			if ((!anycase && !memcmp(name, bufname, xlate_len)) ||
-			    (anycase && !nls_strnicmp(nls_io, name, bufname,
-								xlate_len)))
-				goto Found;
+		len = fat_uni_to_x8(sbi, bufuname, bufname, PATH_MAX);
+		if (fat_name_match(sbi, name, name_len, bufname, len))
+			goto found;
 
 		if (nr_slots) {
-			xlate_len = utf8
-				?utf8_wcstombs(bufname, unicode, PATH_MAX)
-				:uni16_to_x8(bufname, unicode, PATH_MAX, uni_xlate, nls_io);
-			if (xlate_len != name_len)
-				continue;
-			if ((!anycase && !memcmp(name, bufname, xlate_len)) ||
-			    (anycase && !nls_strnicmp(nls_io, name, bufname,
-								xlate_len)))
-				goto Found;
+			/* Compare longname */
+			len = fat_uni_to_x8(sbi, unicode, bufname, PATH_MAX);
+			if (fat_name_match(sbi, name, name_len, bufname, len))
+				goto found;
 		}
 	}
 
-Found:
+found:
 	nr_slots++;	/* include the de */
 	sinfo->slot_off = cpos - nr_slots * sizeof(*de);
 	sinfo->nr_slots = nr_slots;
@@ -424,7 +434,7 @@ Found:
 	sinfo->bh = bh;
 	sinfo->i_pos = fat_make_i_pos(sb, sinfo->bh, sinfo->de);
 	err = 0;
-EODir:
+end_of_dir:
 	if (bufname)
 		__putname(bufname);
 	if (unicode)
@@ -452,23 +462,19 @@ static int __fat_readdir(struct inode *inode, struct file *filp, void *dirent,
 	struct msdos_sb_info *sbi = MSDOS_SB(sb);
 	struct buffer_head *bh;
 	struct msdos_dir_entry *de;
-	struct nls_table *nls_io = sbi->nls_io;
 	struct nls_table *nls_disk = sbi->nls_disk;
-	unsigned char long_slots;
-	const char *fill_name;
-	int fill_len;
+	unsigned char nr_slots;
 	wchar_t bufuname[14];
 	wchar_t *unicode = NULL;
 	unsigned char c, work[MSDOS_NAME], bufname[56], *ptname = bufname;
-	unsigned long lpos, dummy, *furrfu = &lpos;
-	int uni_xlate = sbi->options.unicode_xlate;
-	int isvfat = sbi->options.isvfat;
-	int utf8 = sbi->options.utf8;
-	int nocase = sbi->options.nocase;
 	unsigned short opt_shortname = sbi->options.shortname;
+	int isvfat = sbi->options.isvfat;
+	int nocase = sbi->options.nocase;
+	const char *fill_name;
 	unsigned long inum;
-	int chi, chl, i, i2, j, last, last_u, dotoffset = 0;
+	unsigned long lpos, dummy, *furrfu = &lpos;
 	loff_t cpos;
+	int chi, chl, i, i2, j, last, last_u, dotoffset = 0, fill_len;
 	int ret = 0;
 
 	lock_super(sb);
@@ -488,43 +494,43 @@ static int __fat_readdir(struct inode *inode, struct file *filp, void *dirent,
 			cpos = 0;
 		}
 	}
-	if (cpos & (sizeof(struct msdos_dir_entry)-1)) {
+	if (cpos & (sizeof(struct msdos_dir_entry) - 1)) {
 		ret = -ENOENT;
 		goto out;
 	}
 
 	bh = NULL;
-GetNew:
+get_new:
 	if (fat_get_entry(inode, &cpos, &bh, &de) == -1)
-		goto EODir;
+		goto end_of_dir;
 parse_record:
-	long_slots = 0;
+	nr_slots = 0;
 	/* Check for long filename entry */
 	if (isvfat) {
 		if (de->name[0] == DELETED_FLAG)
-			goto RecEnd;
+			goto record_end;
 		if (de->attr != ATTR_EXT && (de->attr & ATTR_VOLUME))
-			goto RecEnd;
+			goto record_end;
 		if (de->attr != ATTR_EXT && IS_FREE(de->name))
-			goto RecEnd;
+			goto record_end;
 	} else {
 		if ((de->attr & ATTR_VOLUME) || IS_FREE(de->name))
-			goto RecEnd;
+			goto record_end;
 	}
 
 	if (isvfat && de->attr == ATTR_EXT) {
 		int status = fat_parse_long(inode, &cpos, &bh, &de,
-					    &unicode, &long_slots);
+					    &unicode, &nr_slots);
 		if (status < 0) {
 			filp->f_pos = cpos;
 			ret = status;
 			goto out;
 		} else if (status == PARSE_INVALID)
-			goto RecEnd;
+			goto record_end;
 		else if (status == PARSE_NOT_LONGNAME)
 			goto parse_record;
 		else if (status == PARSE_EOF)
-			goto EODir;
+			goto end_of_dir;
 	}
 
 	if (sbi->options.dotsOK) {
@@ -586,12 +592,12 @@ parse_record:
 		}
 	}
 	if (!last)
-		goto RecEnd;
+		goto record_end;
 
 	i = last + dotoffset;
 	j = last_u;
 
-	lpos = cpos - (long_slots+1)*sizeof(struct msdos_dir_entry);
+	lpos = cpos - (nr_slots + 1) * sizeof(struct msdos_dir_entry);
 	if (!memcmp(de->name, MSDOS_DOT, MSDOS_NAME))
 		inum = inode->i_ino;
 	else if (!memcmp(de->name, MSDOS_DOTDOT, MSDOS_NAME)) {
@@ -608,20 +614,17 @@ parse_record:
 
 	if (isvfat) {
 		bufuname[j] = 0x0000;
-		i = utf8 ? utf8_wcstombs(bufname, bufuname, sizeof(bufname))
-			 : uni16_to_x8(bufname, bufuname, sizeof(bufname), uni_xlate, nls_io);
+		i = fat_uni_to_x8(sbi, bufuname, bufname, sizeof(bufname));
 	}
 
 	fill_name = bufname;
 	fill_len = i;
-	if (!short_only && long_slots) {
+	if (!short_only && nr_slots) {
 		/* convert the unicode long name. 261 is maximum size
 		 * of unicode buffer. (13 * slots + nul) */
 		void *longname = unicode + 261;
 		int buf_size = PATH_MAX - (261 * sizeof(unicode[0]));
-		int long_len = utf8
-			? utf8_wcstombs(longname, unicode, buf_size)
-			: uni16_to_x8(longname, unicode, buf_size, uni_xlate, nls_io);
+		int long_len = fat_uni_to_x8(sbi, unicode, longname, buf_size);
 
 		if (!both) {
 			fill_name = longname;
@@ -640,15 +643,15 @@ parse_record:
 	}
 	if (filldir(dirent, fill_name, fill_len, *furrfu, inum,
 		    (de->attr & ATTR_DIR) ? DT_DIR : DT_REG) < 0)
-		goto FillFailed;
+		goto fill_failed;
 
-RecEnd:
+record_end:
 	furrfu = &lpos;
 	filp->f_pos = cpos;
-	goto GetNew;
-EODir:
+	goto get_new;
+end_of_dir:
 	filp->f_pos = cpos;
-FillFailed:
+fill_failed:
 	brelse(bh);
 	if (unicode)
 		__putname(unicode);

From 98a15160049fc1a0f822047f33ff513906a35567 Mon Sep 17 00:00:00 2001
From: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Date: Fri, 25 Jul 2008 01:46:44 -0700
Subject: [PATCH 0765/1171] fat: use same logic in fat_search_long() and
 __fat_readdir()

This uses uses stack for shortname, and uses __getname() for longname in
fat_search_long() and __fat_readdir().  By this, it removes unneeded
__getname() for shortname.

Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/fat/dir.c | 42 ++++++++++++++++++++++++++----------------
 1 file changed, 26 insertions(+), 16 deletions(-)

diff --git a/fs/fat/dir.c b/fs/fat/dir.c
index a4410740627f..96a1cad30da4 100644
--- a/fs/fat/dir.c
+++ b/fs/fat/dir.c
@@ -324,6 +324,19 @@ parse_long:
 	return 0;
 }
 
+/*
+ * Maximum buffer size of short name.
+ * [(MSDOS_NAME + '.') * max one char + nul]
+ * For msdos style, ['.' (hidden) + MSDOS_NAME + '.' + nul]
+ */
+#define FAT_MAX_SHORT_SIZE	((MSDOS_NAME + 1) * NLS_MAX_CHARSET_SIZE + 1)
+/*
+ * Maximum buffer size of unicode chars from slots.
+ * [(max longname slots * 13 (size in a slot) + nul) * sizeof(wchar_t)]
+ */
+#define FAT_MAX_UNI_CHARS	((MSDOS_SLOTS - 1) * 13 + 1)
+#define FAT_MAX_UNI_SIZE	(FAT_MAX_UNI_CHARS * sizeof(wchar_t))
+
 /*
  * Return values: negative -> error, 0 -> not found, positive -> found,
  * value is the total amount of slots, including the shortname entry.
@@ -340,15 +353,11 @@ int fat_search_long(struct inode *inode, const unsigned char *name,
 	wchar_t bufuname[14];
 	wchar_t *unicode = NULL;
 	unsigned char work[MSDOS_NAME];
-	unsigned char *bufname = NULL;
+	unsigned char bufname[FAT_MAX_SHORT_SIZE];
 	unsigned short opt_shortname = sbi->options.shortname;
 	loff_t cpos = 0;
 	int chl, i, j, last_u, err, len;
 
-	bufname = __getname();
-	if (!bufname)
-		return -ENOMEM;
-
 	err = -ENOENT;
 	while (1) {
 		if (fat_get_entry(inode, &cpos, &bh, &de) == -1)
@@ -414,14 +423,17 @@ parse_record:
 
 		/* Compare shortname */
 		bufuname[last_u] = 0x0000;
-		len = fat_uni_to_x8(sbi, bufuname, bufname, PATH_MAX);
+		len = fat_uni_to_x8(sbi, bufuname, bufname, sizeof(bufname));
 		if (fat_name_match(sbi, name, name_len, bufname, len))
 			goto found;
 
 		if (nr_slots) {
+			void *longname = unicode + FAT_MAX_UNI_CHARS;
+			int size = PATH_MAX - FAT_MAX_UNI_SIZE;
+
 			/* Compare longname */
-			len = fat_uni_to_x8(sbi, unicode, bufname, PATH_MAX);
-			if (fat_name_match(sbi, name, name_len, bufname, len))
+			len = fat_uni_to_x8(sbi, unicode, longname, size);
+			if (fat_name_match(sbi, name, name_len, longname, len))
 				goto found;
 		}
 	}
@@ -435,8 +447,6 @@ found:
 	sinfo->i_pos = fat_make_i_pos(sb, sinfo->bh, sinfo->de);
 	err = 0;
 end_of_dir:
-	if (bufname)
-		__putname(bufname);
 	if (unicode)
 		__putname(unicode);
 
@@ -466,7 +476,8 @@ static int __fat_readdir(struct inode *inode, struct file *filp, void *dirent,
 	unsigned char nr_slots;
 	wchar_t bufuname[14];
 	wchar_t *unicode = NULL;
-	unsigned char c, work[MSDOS_NAME], bufname[56], *ptname = bufname;
+	unsigned char c, work[MSDOS_NAME];
+	unsigned char bufname[FAT_MAX_SHORT_SIZE], *ptname = bufname;
 	unsigned short opt_shortname = sbi->options.shortname;
 	int isvfat = sbi->options.isvfat;
 	int nocase = sbi->options.nocase;
@@ -620,11 +631,10 @@ parse_record:
 	fill_name = bufname;
 	fill_len = i;
 	if (!short_only && nr_slots) {
-		/* convert the unicode long name. 261 is maximum size
-		 * of unicode buffer. (13 * slots + nul) */
-		void *longname = unicode + 261;
-		int buf_size = PATH_MAX - (261 * sizeof(unicode[0]));
-		int long_len = fat_uni_to_x8(sbi, unicode, longname, buf_size);
+		void *longname = unicode + FAT_MAX_UNI_CHARS;
+		int long_len, size = PATH_MAX - FAT_MAX_UNI_SIZE;
+
+		long_len = fat_uni_to_x8(sbi, unicode, longname, size);
 
 		if (!both) {
 			fill_name = longname;

From dcd8c53f13f068ee039589d84fbd0baf686abc41 Mon Sep 17 00:00:00 2001
From: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Date: Fri, 25 Jul 2008 01:46:44 -0700
Subject: [PATCH 0766/1171] fat: small optimization to __fat_readdir()

This removes unnecessary parsing for directory entries.

If short_only, we don't need to parse longname.  And if !both and it found
the longname, we don't need shortname.

Signed-off-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/fat/dir.c | 71 +++++++++++++++++++++++++++++-----------------------
 1 file changed, 39 insertions(+), 32 deletions(-)

diff --git a/fs/fat/dir.c b/fs/fat/dir.c
index 96a1cad30da4..4c35477bc94c 100644
--- a/fs/fat/dir.c
+++ b/fs/fat/dir.c
@@ -481,11 +481,11 @@ static int __fat_readdir(struct inode *inode, struct file *filp, void *dirent,
 	unsigned short opt_shortname = sbi->options.shortname;
 	int isvfat = sbi->options.isvfat;
 	int nocase = sbi->options.nocase;
-	const char *fill_name;
+	const char *fill_name = NULL;
 	unsigned long inum;
 	unsigned long lpos, dummy, *furrfu = &lpos;
 	loff_t cpos;
-	int chi, chl, i, i2, j, last, last_u, dotoffset = 0, fill_len;
+	int chi, chl, i, i2, j, last, last_u, dotoffset = 0, fill_len = 0;
 	int ret = 0;
 
 	lock_super(sb);
@@ -516,8 +516,11 @@ get_new:
 		goto end_of_dir;
 parse_record:
 	nr_slots = 0;
-	/* Check for long filename entry */
-	if (isvfat) {
+	/*
+	 * Check for long filename entry, but if short_only, we don't
+	 * need to parse long filename.
+	 */
+	if (isvfat && !short_only) {
 		if (de->name[0] == DELETED_FLAG)
 			goto record_end;
 		if (de->attr != ATTR_EXT && (de->attr & ATTR_VOLUME))
@@ -542,6 +545,18 @@ parse_record:
 			goto parse_record;
 		else if (status == PARSE_EOF)
 			goto end_of_dir;
+
+		if (nr_slots) {
+			void *longname = unicode + FAT_MAX_UNI_CHARS;
+			int size = PATH_MAX - FAT_MAX_UNI_SIZE;
+			int len = fat_uni_to_x8(sbi, unicode, longname, size);
+
+			fill_name = longname;
+			fill_len = len;
+			/* !both && !short_only, so we don't need shortname. */
+			if (!both)
+				goto start_filldir;
+		}
 	}
 
 	if (sbi->options.dotsOK) {
@@ -608,6 +623,26 @@ parse_record:
 	i = last + dotoffset;
 	j = last_u;
 
+	if (isvfat) {
+		bufuname[j] = 0x0000;
+		i = fat_uni_to_x8(sbi, bufuname, bufname, sizeof(bufname));
+	}
+	if (nr_slots) {
+		/* hack for fat_ioctl_filldir() */
+		struct fat_ioctl_filldir_callback *p = dirent;
+
+		p->longname = fill_name;
+		p->long_len = fill_len;
+		p->shortname = bufname;
+		p->short_len = i;
+		fill_name = NULL;
+		fill_len = 0;
+	} else {
+		fill_name = bufname;
+		fill_len = i;
+	}
+
+start_filldir:
 	lpos = cpos - (nr_slots + 1) * sizeof(struct msdos_dir_entry);
 	if (!memcmp(de->name, MSDOS_DOT, MSDOS_NAME))
 		inum = inode->i_ino;
@@ -623,34 +658,6 @@ parse_record:
 			inum = iunique(sb, MSDOS_ROOT_INO);
 	}
 
-	if (isvfat) {
-		bufuname[j] = 0x0000;
-		i = fat_uni_to_x8(sbi, bufuname, bufname, sizeof(bufname));
-	}
-
-	fill_name = bufname;
-	fill_len = i;
-	if (!short_only && nr_slots) {
-		void *longname = unicode + FAT_MAX_UNI_CHARS;
-		int long_len, size = PATH_MAX - FAT_MAX_UNI_SIZE;
-
-		long_len = fat_uni_to_x8(sbi, unicode, longname, size);
-
-		if (!both) {
-			fill_name = longname;
-			fill_len = long_len;
-		} else {
-			/* hack for fat_ioctl_filldir() */
-			struct fat_ioctl_filldir_callback *p = dirent;
-
-			p->longname = longname;
-			p->long_len = long_len;
-			p->shortname = bufname;
-			p->short_len = i;
-			fill_name = NULL;
-			fill_len = 0;
-		}
-	}
 	if (filldir(dirent, fill_name, fill_len, *furrfu, inum,
 		    (de->attr & ATTR_DIR) ? DT_DIR : DT_REG) < 0)
 		goto fill_failed;

From 7557bc66be629d19a402e752673708bfbb8b5e86 Mon Sep 17 00:00:00 2001
From: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Date: Fri, 25 Jul 2008 01:46:45 -0700
Subject: [PATCH 0767/1171] msdos fs: remove unsettable atari option

It has been impossible to set the option 'atari' of the MSDOS filesystem
for several years.  Since nobody seems to have missed it, let's remove its
remains.

Signed-off-by: Rene Scharfe <rene.scharfe@lsrfire.ath.cx>
Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/msdos/namei.c         | 18 ++++++------------
 include/linux/msdos_fs.h |  1 -
 2 files changed, 6 insertions(+), 13 deletions(-)

diff --git a/fs/msdos/namei.c b/fs/msdos/namei.c
index 1f7f2956412a..e4ad6c6b753e 100644
--- a/fs/msdos/namei.c
+++ b/fs/msdos/namei.c
@@ -14,12 +14,7 @@
 
 /* Characters that are undesirable in an MS-DOS file name */
 static unsigned char bad_chars[] = "*?<>|\"";
-static unsigned char bad_if_strict_pc[] = "+=,; ";
-/* GEMDOS is less restrictive */
-static unsigned char bad_if_strict_atari[] = " ";
-
-#define bad_if_strict(opts) \
-	((opts)->atari ? bad_if_strict_atari : bad_if_strict_pc)
+static unsigned char bad_if_strict[] = "+=,; ";
 
 /***** Formats an MS-DOS file name. Rejects invalid names. */
 static int msdos_format_name(const unsigned char *name, int len,
@@ -40,21 +35,20 @@ static int msdos_format_name(const unsigned char *name, int len,
 			/* Get rid of dot - test for it elsewhere */
 			name++;
 			len--;
-		} else if (!opts->atari)
+		} else
 			return -EINVAL;
 	}
 	/*
-	 * disallow names that _really_ start with a dot for MS-DOS,
-	 * GEMDOS does not care
+	 * disallow names that _really_ start with a dot
 	 */
-	space = !opts->atari;
+	space = 1;
 	c = 0;
 	for (walk = res; len && walk - res < 8; walk++) {
 		c = *name++;
 		len--;
 		if (opts->name_check != 'r' && strchr(bad_chars, c))
 			return -EINVAL;
-		if (opts->name_check == 's' && strchr(bad_if_strict(opts), c))
+		if (opts->name_check == 's' && strchr(bad_if_strict, c))
 			return -EINVAL;
 		if (c >= 'A' && c <= 'Z' && opts->name_check == 's')
 			return -EINVAL;
@@ -94,7 +88,7 @@ static int msdos_format_name(const unsigned char *name, int len,
 			if (opts->name_check != 'r' && strchr(bad_chars, c))
 				return -EINVAL;
 			if (opts->name_check == 's' &&
-			    strchr(bad_if_strict(opts), c))
+			    strchr(bad_if_strict, c))
 				return -EINVAL;
 			if (c < ' ' || c == ':' || c == '\\')
 				return -EINVAL;
diff --git a/include/linux/msdos_fs.h b/include/linux/msdos_fs.h
index 5161394c7894..3346c9c8f17a 100644
--- a/include/linux/msdos_fs.h
+++ b/include/linux/msdos_fs.h
@@ -201,7 +201,6 @@ struct fat_mount_options {
 		 utf8:1,	  /* Use of UTF-8 character set (Default) */
 		 unicode_xlate:1, /* create escape sequences for unhandled Unicode */
 		 numtail:1,       /* Does first alias have a numeric '~1' type tail? */
-		 atari:1,         /* Use Atari GEMDOS variation of MS-DOS fs */
 		 flush:1,	  /* write things quickly */
 		 nocase:1,	  /* Does this need case conversion? 0=need case conversion*/
 		 usefree:1;	  /* Use free_clusters for FAT32 */

From cf6ae8b50e0ee3f764392dadd1970e3f03c40773 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 01:46:46 -0700
Subject: [PATCH 0768/1171] remove the in-kernel struct dirent{,64}

The kernel struct dirent{,64} were different from the ones in
userspace.

Even worse, we exported the kernel ones to userspace.

But after the fat usages are fixed we can remove the conflicting
kernel versions.

Reviewed-by: H. Peter Anvin <hpa@kernel.org>
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Cc: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/Kbuild   |  1 -
 include/linux/dirent.h | 20 --------------------
 2 files changed, 21 deletions(-)

diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index 71d70d1fbce2..a18008ce7aba 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -189,7 +189,6 @@ unifdef-y += connector.h
 unifdef-y += cuda.h
 unifdef-y += cyclades.h
 unifdef-y += dccp.h
-unifdef-y += dirent.h
 unifdef-y += dlm.h
 unifdef-y += dlm_plock.h
 unifdef-y += edd.h
diff --git a/include/linux/dirent.h b/include/linux/dirent.h
index 5d6023b87800..f072fb8d10a3 100644
--- a/include/linux/dirent.h
+++ b/include/linux/dirent.h
@@ -1,23 +1,6 @@
 #ifndef _LINUX_DIRENT_H
 #define _LINUX_DIRENT_H
 
-struct dirent {
-	long		d_ino;
-	__kernel_off_t	d_off;
-	unsigned short	d_reclen;
-	char		d_name[256]; /* We must not include limits.h! */
-};
-
-struct dirent64 {
-	__u64		d_ino;
-	__s64		d_off;
-	unsigned short	d_reclen;
-	unsigned char	d_type;
-	char		d_name[256];
-};
-
-#ifdef __KERNEL__
-
 struct linux_dirent64 {
 	u64		d_ino;
 	s64		d_off;
@@ -26,7 +9,4 @@ struct linux_dirent64 {
 	char		d_name[0];
 };
 
-#endif	/* __KERNEL__ */
-
-
 #endif

From e8938a62a85d1f487e02c3b01955b47c9598f6d2 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 01:46:46 -0700
Subject: [PATCH 0769/1171] remove unused #include <linux/dirent.h>'s

Remove some unused #include <linux/dirent.h>'s.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/mips/kernel/linux32.c | 1 -
 fs/compat_ioctl.c          | 1 -
 fs/smbfs/cache.c           | 1 -
 fs/smbfs/proc.c            | 1 -
 include/linux/nfsd/nfsd.h  | 1 -
 5 files changed, 5 deletions(-)

diff --git a/arch/mips/kernel/linux32.c b/arch/mips/kernel/linux32.c
index c266211ed653..2fefb14414b7 100644
--- a/arch/mips/kernel/linux32.c
+++ b/arch/mips/kernel/linux32.c
@@ -11,7 +11,6 @@
 #include <linux/file.h>
 #include <linux/smp_lock.h>
 #include <linux/highuid.h>
-#include <linux/dirent.h>
 #include <linux/resource.h>
 #include <linux/highmem.h>
 #include <linux/time.h>
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index 18e2c548161d..5235c67e7594 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -25,7 +25,6 @@
 #include <linux/slab.h>
 #include <linux/raid/md.h>
 #include <linux/kd.h>
-#include <linux/dirent.h>
 #include <linux/route.h>
 #include <linux/in6.h>
 #include <linux/ipv6_route.h>
diff --git a/fs/smbfs/cache.c b/fs/smbfs/cache.c
index 8182f0542a21..8c177eb7e344 100644
--- a/fs/smbfs/cache.c
+++ b/fs/smbfs/cache.c
@@ -13,7 +13,6 @@
 #include <linux/errno.h>
 #include <linux/kernel.h>
 #include <linux/mm.h>
-#include <linux/dirent.h>
 #include <linux/smb_fs.h>
 #include <linux/pagemap.h>
 #include <linux/net.h>
diff --git a/fs/smbfs/proc.c b/fs/smbfs/proc.c
index d517a27b7f4b..ee536e8a649a 100644
--- a/fs/smbfs/proc.c
+++ b/fs/smbfs/proc.c
@@ -16,7 +16,6 @@
 #include <linux/stat.h>
 #include <linux/fcntl.h>
 #include <linux/dcache.h>
-#include <linux/dirent.h>
 #include <linux/nls.h>
 #include <linux/smp_lock.h>
 #include <linux/net.h>
diff --git a/include/linux/nfsd/nfsd.h b/include/linux/nfsd/nfsd.h
index a2861d95ecc3..108f47e5fd95 100644
--- a/include/linux/nfsd/nfsd.h
+++ b/include/linux/nfsd/nfsd.h
@@ -12,7 +12,6 @@
 
 #include <linux/types.h>
 #include <linux/unistd.h>
-#include <linux/dirent.h>
 #include <linux/fs.h>
 #include <linux/posix_acl.h>
 #include <linux/mount.h>

From b271e067c896ad4082b15e96077675d08db40625 Mon Sep 17 00:00:00 2001
From: Joe Peterson <joe@skyrush.com>
Date: Fri, 25 Jul 2008 01:46:47 -0700
Subject: [PATCH 0770/1171] fatfs: add UTC timestamp option

Provide a new mount option ("tz=UTC") for DOS (vfat/msdos) filesystems,
allowing timestamps to be in coordinated universal time (UTC) rather than
local time in applications where doing this is advantageous.

In particular, portable devices that use fat/vfat (such as digital
cameras) can benefit from using UTC in their internal clocks, thus
avoiding daylight saving time errors and general time ambiguity issues.
The user of the device does not have to worry about changing the time when
moving from place or when daylight saving changes.

The new mount option, when set, disables the counter-adjustment that Linux
currently makes to FAT timestamp info in anticipation of the normal
userspace time zone correction.  When used in this new mode, all daylight
saving time and time zone handling is done in userspace as is normal for
many other filesystems (like ext3).  The default mode, which remains
unchanged, is still appropriate when mounting volumes written in Windows
(because of its use of local time).

I originally based this patch on one submitted last year by Paul Collins,
but I updated it to work with current source and changed variable/option
naming.  Ogawa Hirofumi (who maintains these filesystems) and I discussed
this patch at length on lkml, and he suggested using the option name in
the attached version of the patch.  Barry Bouwsma pointed out a good
addition to the patch as well.

Signed-off-by: Joe Peterson <joe@skyrush.com>
Signed-off-by: Paul Collins <paul@ondioline.org>
Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Cc: Barry Bouwsma <free_beer_for_all@yahoo.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/fat/dir.c             |  2 +-
 fs/fat/inode.c           | 27 ++++++++++++++++++++-------
 fs/fat/misc.c            | 10 ++++++----
 fs/msdos/namei.c         |  3 ++-
 fs/vfat/namei.c          |  2 +-
 include/linux/msdos_fs.h |  8 +++++---
 6 files changed, 35 insertions(+), 17 deletions(-)

diff --git a/fs/fat/dir.c b/fs/fat/dir.c
index 4c35477bc94c..cd4a0162e10d 100644
--- a/fs/fat/dir.c
+++ b/fs/fat/dir.c
@@ -1101,7 +1101,7 @@ int fat_alloc_new_dir(struct inode *dir, struct timespec *ts)
 		goto error_free;
 	}
 
-	fat_date_unix2dos(ts->tv_sec, &time, &date);
+	fat_date_unix2dos(ts->tv_sec, &time, &date, sbi->options.tz_utc);
 
 	de = (struct msdos_dir_entry *)bhs[0]->b_data;
 	/* filling the new directory slots ("." and ".." entries) */
diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index 60deb5fd1188..23676f9d79ce 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -382,17 +382,20 @@ static int fat_fill_inode(struct inode *inode, struct msdos_dir_entry *de)
 	inode->i_blocks = ((inode->i_size + (sbi->cluster_size - 1))
 			   & ~((loff_t)sbi->cluster_size - 1)) >> 9;
 	inode->i_mtime.tv_sec =
-		date_dos2unix(le16_to_cpu(de->time), le16_to_cpu(de->date));
+		date_dos2unix(le16_to_cpu(de->time), le16_to_cpu(de->date),
+			      sbi->options.tz_utc);
 	inode->i_mtime.tv_nsec = 0;
 	if (sbi->options.isvfat) {
 		int secs = de->ctime_cs / 100;
 		int csecs = de->ctime_cs % 100;
 		inode->i_ctime.tv_sec  =
 			date_dos2unix(le16_to_cpu(de->ctime),
-				      le16_to_cpu(de->cdate)) + secs;
+				      le16_to_cpu(de->cdate),
+				      sbi->options.tz_utc) + secs;
 		inode->i_ctime.tv_nsec = csecs * 10000000;
 		inode->i_atime.tv_sec =
-			date_dos2unix(0, le16_to_cpu(de->adate));
+			date_dos2unix(0, le16_to_cpu(de->adate),
+				      sbi->options.tz_utc);
 		inode->i_atime.tv_nsec = 0;
 	} else
 		inode->i_ctime = inode->i_atime = inode->i_mtime;
@@ -591,11 +594,14 @@ retry:
 	raw_entry->attr = fat_attr(inode);
 	raw_entry->start = cpu_to_le16(MSDOS_I(inode)->i_logstart);
 	raw_entry->starthi = cpu_to_le16(MSDOS_I(inode)->i_logstart >> 16);
-	fat_date_unix2dos(inode->i_mtime.tv_sec, &raw_entry->time, &raw_entry->date);
+	fat_date_unix2dos(inode->i_mtime.tv_sec, &raw_entry->time,
+			  &raw_entry->date, sbi->options.tz_utc);
 	if (sbi->options.isvfat) {
 		__le16 atime;
-		fat_date_unix2dos(inode->i_ctime.tv_sec,&raw_entry->ctime,&raw_entry->cdate);
-		fat_date_unix2dos(inode->i_atime.tv_sec,&atime,&raw_entry->adate);
+		fat_date_unix2dos(inode->i_ctime.tv_sec, &raw_entry->ctime,
+				  &raw_entry->cdate, sbi->options.tz_utc);
+		fat_date_unix2dos(inode->i_atime.tv_sec, &atime,
+				  &raw_entry->adate, sbi->options.tz_utc);
 		raw_entry->ctime_cs = (inode->i_ctime.tv_sec & 1) * 100 +
 			inode->i_ctime.tv_nsec / 10000000;
 	}
@@ -836,6 +842,8 @@ static int fat_show_options(struct seq_file *m, struct vfsmount *mnt)
 	}
 	if (sbi->options.flush)
 		seq_puts(m, ",flush");
+	if (opts->tz_utc)
+		seq_puts(m, ",tz=UTC");
 
 	return 0;
 }
@@ -848,7 +856,7 @@ enum {
 	Opt_charset, Opt_shortname_lower, Opt_shortname_win95,
 	Opt_shortname_winnt, Opt_shortname_mixed, Opt_utf8_no, Opt_utf8_yes,
 	Opt_uni_xl_no, Opt_uni_xl_yes, Opt_nonumtail_no, Opt_nonumtail_yes,
-	Opt_obsolate, Opt_flush, Opt_err,
+	Opt_obsolate, Opt_flush, Opt_tz_utc, Opt_err,
 };
 
 static match_table_t fat_tokens = {
@@ -883,6 +891,7 @@ static match_table_t fat_tokens = {
 	{Opt_obsolate, "cvf_options=%100s"},
 	{Opt_obsolate, "posix"},
 	{Opt_flush, "flush"},
+	{Opt_tz_utc, "tz=UTC"},
 	{Opt_err, NULL},
 };
 static match_table_t msdos_tokens = {
@@ -947,6 +956,7 @@ static int parse_options(char *options, int is_vfat, int silent, int *debug,
 	opts->utf8 = opts->unicode_xlate = 0;
 	opts->numtail = 1;
 	opts->usefree = opts->nocase = 0;
+	opts->tz_utc = 0;
 	*debug = 0;
 
 	if (!options)
@@ -1036,6 +1046,9 @@ static int parse_options(char *options, int is_vfat, int silent, int *debug,
 		case Opt_flush:
 			opts->flush = 1;
 			break;
+		case Opt_tz_utc:
+			opts->tz_utc = 1;
+			break;
 
 		/* msdos specific */
 		case Opt_dots:
diff --git a/fs/fat/misc.c b/fs/fat/misc.c
index 61f23511eacf..79fb98ad36d4 100644
--- a/fs/fat/misc.c
+++ b/fs/fat/misc.c
@@ -142,7 +142,7 @@ static int day_n[] = {
 };
 
 /* Convert a MS-DOS time/date pair to a UNIX date (seconds since 1 1 70). */
-int date_dos2unix(unsigned short time, unsigned short date)
+int date_dos2unix(unsigned short time, unsigned short date, int tz_utc)
 {
 	int month, year, secs;
 
@@ -156,16 +156,18 @@ int date_dos2unix(unsigned short time, unsigned short date)
 	    ((date & 31)-1+day_n[month]+(year/4)+year*365-((year & 3) == 0 &&
 	    month < 2 ? 1 : 0)+3653);
 			/* days since 1.1.70 plus 80's leap day */
-	secs += sys_tz.tz_minuteswest*60;
+	if (!tz_utc)
+		secs += sys_tz.tz_minuteswest*60;
 	return secs;
 }
 
 /* Convert linear UNIX date to a MS-DOS time/date pair. */
-void fat_date_unix2dos(int unix_date, __le16 *time, __le16 *date)
+void fat_date_unix2dos(int unix_date, __le16 *time, __le16 *date, int tz_utc)
 {
 	int day, year, nl_day, month;
 
-	unix_date -= sys_tz.tz_minuteswest*60;
+	if (!tz_utc)
+		unix_date -= sys_tz.tz_minuteswest*60;
 
 	/* Jan 1 GMT 00:00:00 1980. But what about another time zone? */
 	if (unix_date < 315532800)
diff --git a/fs/msdos/namei.c b/fs/msdos/namei.c
index e4ad6c6b753e..e844b9809d27 100644
--- a/fs/msdos/namei.c
+++ b/fs/msdos/namei.c
@@ -237,6 +237,7 @@ static int msdos_add_entry(struct inode *dir, const unsigned char *name,
 			   int is_dir, int is_hid, int cluster,
 			   struct timespec *ts, struct fat_slot_info *sinfo)
 {
+	struct msdos_sb_info *sbi = MSDOS_SB(dir->i_sb);
 	struct msdos_dir_entry de;
 	__le16 time, date;
 	int err;
@@ -246,7 +247,7 @@ static int msdos_add_entry(struct inode *dir, const unsigned char *name,
 	if (is_hid)
 		de.attr |= ATTR_HIDDEN;
 	de.lcase = 0;
-	fat_date_unix2dos(ts->tv_sec, &time, &date);
+	fat_date_unix2dos(ts->tv_sec, &time, &date, sbi->options.tz_utc);
 	de.cdate = de.adate = 0;
 	de.ctime = 0;
 	de.ctime_cs = 0;
diff --git a/fs/vfat/namei.c b/fs/vfat/namei.c
index b546ba69be82..155c10b4adbd 100644
--- a/fs/vfat/namei.c
+++ b/fs/vfat/namei.c
@@ -621,7 +621,7 @@ shortname:
 	memcpy(de->name, msdos_name, MSDOS_NAME);
 	de->attr = is_dir ? ATTR_DIR : ATTR_ARCH;
 	de->lcase = lcase;
-	fat_date_unix2dos(ts->tv_sec, &time, &date);
+	fat_date_unix2dos(ts->tv_sec, &time, &date, sbi->options.tz_utc);
 	de->time = de->ctime = time;
 	de->date = de->cdate = de->adate = date;
 	de->ctime_cs = 0;
diff --git a/include/linux/msdos_fs.h b/include/linux/msdos_fs.h
index 3346c9c8f17a..ba63858056c7 100644
--- a/include/linux/msdos_fs.h
+++ b/include/linux/msdos_fs.h
@@ -203,7 +203,8 @@ struct fat_mount_options {
 		 numtail:1,       /* Does first alias have a numeric '~1' type tail? */
 		 flush:1,	  /* write things quickly */
 		 nocase:1,	  /* Does this need case conversion? 0=need case conversion*/
-		 usefree:1;	  /* Use free_clusters for FAT32 */
+		 usefree:1,	  /* Use free_clusters for FAT32 */
+		 tz_utc:1;	  /* Filesystem timestamps are in UTC */
 };
 
 #define FAT_HASH_BITS	8
@@ -434,8 +435,9 @@ extern int fat_flush_inodes(struct super_block *sb, struct inode *i1,
 extern void fat_fs_panic(struct super_block *s, const char *fmt, ...);
 extern void fat_clusters_flush(struct super_block *sb);
 extern int fat_chain_add(struct inode *inode, int new_dclus, int nr_cluster);
-extern int date_dos2unix(unsigned short time, unsigned short date);
-extern void fat_date_unix2dos(int unix_date, __le16 *time, __le16 *date);
+extern int date_dos2unix(unsigned short time, unsigned short date, int tz_utc);
+extern void fat_date_unix2dos(int unix_date, __le16 *time, __le16 *date,
+			      int tz_utc);
 extern int fat_sync_bhs(struct buffer_head **bhs, int nr_bhs);
 
 int fat_cache_init(void);

From 41003cde95e7e976d3876dbdcdc83dd0a9059279 Mon Sep 17 00:00:00 2001
From: Joe Peterson <joe@skyrush.com>
Date: Fri, 25 Jul 2008 01:46:48 -0700
Subject: [PATCH 0771/1171] UTC timestamp option for FAT filesystems fix

Signed-off-by: Joe Peterson <joe@skyrush.com>
Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/filesystems/vfat.txt | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/Documentation/filesystems/vfat.txt b/Documentation/filesystems/vfat.txt
index 2d5e1e582e13..bbac4f1d9056 100644
--- a/Documentation/filesystems/vfat.txt
+++ b/Documentation/filesystems/vfat.txt
@@ -96,6 +96,14 @@ shortname=lower|win95|winnt|mixed
 			emulate the Windows 95 rule for create.
 		 Default setting is `lower'.
 
+tz=UTC        -- Interpret timestamps as UTC rather than local time.
+                 This option disables the conversion of timestamps
+                 between local time (as used by Windows on FAT) and UTC
+                 (which Linux uses internally).  This is particuluarly
+                 useful when mounting devices (like digital cameras)
+                 that are set to UTC in order to avoid the pitfalls of
+                 local time.
+
 <bool>: 0,1,yes,no,true,false
 
 TODO

From b48d380541f634663b71766005838edbb7261685 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Fri, 25 Jul 2008 01:46:49 -0700
Subject: [PATCH 0772/1171] quota: fix possible infinite loop in quota code

When quota structure is going to be dropped and it is dirty, quota code tries
to write it.  If the write fails for some reason (e.  g.  transaction cannot
be started because the journal is aborted), we try writing again and again and
again...  Fix the problem by clearing the dirty bit even if the write failed.

(akpm: for 2.6.27, 2.6.26.x and 2.6.25.x)

Signed-off-by: Jan Kara <jack@suse.cz>
Reviewed-by: dingdinghua <dingdinghua85@gmail.com>
Cc: <stable@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/dquot.c | 16 +++++++++++++++-
 1 file changed, 15 insertions(+), 1 deletion(-)

diff --git a/fs/dquot.c b/fs/dquot.c
index 5ac77da19959..ad88cf6fcbaf 100644
--- a/fs/dquot.c
+++ b/fs/dquot.c
@@ -562,6 +562,8 @@ static struct shrinker dqcache_shrinker = {
  */
 static void dqput(struct dquot *dquot)
 {
+	int ret;
+
 	if (!dquot)
 		return;
 #ifdef __DQUOT_PARANOIA
@@ -594,7 +596,19 @@ we_slept:
 	if (test_bit(DQ_ACTIVE_B, &dquot->dq_flags) && dquot_dirty(dquot)) {
 		spin_unlock(&dq_list_lock);
 		/* Commit dquot before releasing */
-		dquot->dq_sb->dq_op->write_dquot(dquot);
+		ret = dquot->dq_sb->dq_op->write_dquot(dquot);
+		if (ret < 0) {
+			printk(KERN_ERR "VFS: cannot write quota structure on "
+				"device %s (error %d). Quota may get out of "
+				"sync!\n", dquot->dq_sb->s_id, ret);
+			/*
+			 * We clear dirty bit anyway, so that we avoid
+			 * infinite loop here
+			 */
+			spin_lock(&dq_list_lock);
+			clear_dquot_dirty(dquot);
+			spin_unlock(&dq_list_lock);
+		}
 		goto we_slept;
 	}
 	/* Clear flag in case dquot was inactive (something bad happened) */

From b85f4b87a511bea86dac68c4f0fabaee2cac6c4c Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Fri, 25 Jul 2008 01:46:50 -0700
Subject: [PATCH 0773/1171] quota: rename quota functions from upper case, make
 bigger ones non-inline

Cleanup quotaops.h: Rename functions from uppercase to lowercase (and
define backward compatibility macros), move larger functions to dquot.c
and make them non-inline.

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/dquot.c               |  53 ++++++++++
 include/linux/quotaops.h | 216 ++++++++++++++++++---------------------
 2 files changed, 155 insertions(+), 114 deletions(-)

diff --git a/fs/dquot.c b/fs/dquot.c
index ad88cf6fcbaf..0bcaf970bbb4 100644
--- a/fs/dquot.c
+++ b/fs/dquot.c
@@ -1153,6 +1153,28 @@ int dquot_drop(struct inode *inode)
 	return 0;
 }
 
+/* Wrapper to remove references to quota structures from inode */
+void vfs_dq_drop(struct inode *inode)
+{
+	/* Here we can get arbitrary inode from clear_inode() so we have
+	 * to be careful. OTOH we don't need locking as quota operations
+	 * are allowed to change only at mount time */
+	if (!IS_NOQUOTA(inode) && inode->i_sb && inode->i_sb->dq_op
+	    && inode->i_sb->dq_op->drop) {
+		int cnt;
+		/* Test before calling to rule out calls from proc and such
+                 * where we are not allowed to block. Note that this is
+		 * actually reliable test even without the lock - the caller
+		 * must assure that nobody can come after the DQUOT_DROP and
+		 * add quota pointers back anyway */
+		for (cnt = 0; cnt < MAXQUOTAS; cnt++)
+			if (inode->i_dquot[cnt] != NODQUOT)
+				break;
+		if (cnt < MAXQUOTAS)
+			inode->i_sb->dq_op->drop(inode);
+	}
+}
+
 /*
  * Following four functions update i_blocks+i_bytes fields and
  * quota information (together with appropriate checks)
@@ -1426,6 +1448,18 @@ warn_put_all:
 	return ret;
 }
 
+/* Wrapper for transferring ownership of an inode */
+int vfs_dq_transfer(struct inode *inode, struct iattr *iattr)
+{
+	if (sb_any_quota_enabled(inode->i_sb) && !IS_NOQUOTA(inode)) {
+		vfs_dq_init(inode);
+		if (inode->i_sb->dq_op->transfer(inode, iattr) == NO_QUOTA)
+			return 1;
+	}
+	return 0;
+}
+
+
 /*
  * Write info of quota file to disk
  */
@@ -1766,6 +1800,22 @@ out:
 	return error;
 }
 
+/* Wrapper to turn on quotas when remounting rw */
+int vfs_dq_quota_on_remount(struct super_block *sb)
+{
+	int cnt;
+	int ret = 0, err;
+
+	if (!sb->s_qcop || !sb->s_qcop->quota_on)
+		return -ENOSYS;
+	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
+		err = sb->s_qcop->quota_on(sb, cnt, 0, NULL, 1);
+		if (err < 0 && !ret)
+			ret = err;
+	}
+	return ret;
+}
+
 /* Generic routine for getting common part of quota structure */
 static void do_get_dqblk(struct dquot *dquot, struct if_dqblk *di)
 {
@@ -2101,8 +2151,11 @@ EXPORT_SYMBOL(dquot_release);
 EXPORT_SYMBOL(dquot_mark_dquot_dirty);
 EXPORT_SYMBOL(dquot_initialize);
 EXPORT_SYMBOL(dquot_drop);
+EXPORT_SYMBOL(vfs_dq_drop);
 EXPORT_SYMBOL(dquot_alloc_space);
 EXPORT_SYMBOL(dquot_alloc_inode);
 EXPORT_SYMBOL(dquot_free_space);
 EXPORT_SYMBOL(dquot_free_inode);
 EXPORT_SYMBOL(dquot_transfer);
+EXPORT_SYMBOL(vfs_dq_transfer);
+EXPORT_SYMBOL(vfs_dq_quota_on_remount);
diff --git a/include/linux/quotaops.h b/include/linux/quotaops.h
index f86702053853..0c8f9fe462af 100644
--- a/include/linux/quotaops.h
+++ b/include/linux/quotaops.h
@@ -19,34 +19,38 @@
 /*
  * declaration of quota_function calls in kernel.
  */
-extern void sync_dquots(struct super_block *sb, int type);
+void sync_dquots(struct super_block *sb, int type);
 
-extern int dquot_initialize(struct inode *inode, int type);
-extern int dquot_drop(struct inode *inode);
+int dquot_initialize(struct inode *inode, int type);
+int dquot_drop(struct inode *inode);
 
-extern int dquot_alloc_space(struct inode *inode, qsize_t number, int prealloc);
-extern int dquot_alloc_inode(const struct inode *inode, unsigned long number);
+int dquot_alloc_space(struct inode *inode, qsize_t number, int prealloc);
+int dquot_alloc_inode(const struct inode *inode, unsigned long number);
 
-extern int dquot_free_space(struct inode *inode, qsize_t number);
-extern int dquot_free_inode(const struct inode *inode, unsigned long number);
+int dquot_free_space(struct inode *inode, qsize_t number);
+int dquot_free_inode(const struct inode *inode, unsigned long number);
 
-extern int dquot_transfer(struct inode *inode, struct iattr *iattr);
-extern int dquot_commit(struct dquot *dquot);
-extern int dquot_acquire(struct dquot *dquot);
-extern int dquot_release(struct dquot *dquot);
-extern int dquot_commit_info(struct super_block *sb, int type);
-extern int dquot_mark_dquot_dirty(struct dquot *dquot);
+int dquot_transfer(struct inode *inode, struct iattr *iattr);
+int dquot_commit(struct dquot *dquot);
+int dquot_acquire(struct dquot *dquot);
+int dquot_release(struct dquot *dquot);
+int dquot_commit_info(struct super_block *sb, int type);
+int dquot_mark_dquot_dirty(struct dquot *dquot);
 
-extern int vfs_quota_on(struct super_block *sb, int type, int format_id,
-		char *path, int remount);
-extern int vfs_quota_on_mount(struct super_block *sb, char *qf_name,
-		int format_id, int type);
-extern int vfs_quota_off(struct super_block *sb, int type, int remount);
-extern int vfs_quota_sync(struct super_block *sb, int type);
-extern int vfs_get_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii);
-extern int vfs_set_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii);
-extern int vfs_get_dqblk(struct super_block *sb, int type, qid_t id, struct if_dqblk *di);
-extern int vfs_set_dqblk(struct super_block *sb, int type, qid_t id, struct if_dqblk *di);
+int vfs_quota_on(struct super_block *sb, int type, int format_id,
+ 	char *path, int remount);
+int vfs_quota_on_mount(struct super_block *sb, char *qf_name,
+ 	int format_id, int type);
+int vfs_quota_off(struct super_block *sb, int type, int remount);
+int vfs_quota_sync(struct super_block *sb, int type);
+int vfs_get_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii);
+int vfs_set_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii);
+int vfs_get_dqblk(struct super_block *sb, int type, qid_t id, struct if_dqblk *di);
+int vfs_set_dqblk(struct super_block *sb, int type, qid_t id, struct if_dqblk *di);
+
+void vfs_dq_drop(struct inode *inode);
+int vfs_dq_transfer(struct inode *inode, struct iattr *iattr);
+int vfs_dq_quota_on_remount(struct super_block *sb);
 
 /*
  * Operations supported for diskquotas.
@@ -59,38 +63,16 @@ extern struct quotactl_ops vfs_quotactl_ops;
 
 /* It is better to call this function outside of any transaction as it might
  * need a lot of space in journal for dquot structure allocation. */
-static inline void DQUOT_INIT(struct inode *inode)
+static inline void vfs_dq_init(struct inode *inode)
 {
 	BUG_ON(!inode->i_sb);
 	if (sb_any_quota_enabled(inode->i_sb) && !IS_NOQUOTA(inode))
 		inode->i_sb->dq_op->initialize(inode, -1);
 }
 
-/* The same as with DQUOT_INIT */
-static inline void DQUOT_DROP(struct inode *inode)
-{
-	/* Here we can get arbitrary inode from clear_inode() so we have
-	 * to be careful. OTOH we don't need locking as quota operations
-	 * are allowed to change only at mount time */
-	if (!IS_NOQUOTA(inode) && inode->i_sb && inode->i_sb->dq_op
-	    && inode->i_sb->dq_op->drop) {
-		int cnt;
-		/* Test before calling to rule out calls from proc and such
-                 * where we are not allowed to block. Note that this is
-		 * actually reliable test even without the lock - the caller
-		 * must assure that nobody can come after the DQUOT_DROP and
-		 * add quota pointers back anyway */
-		for (cnt = 0; cnt < MAXQUOTAS; cnt++)
-			if (inode->i_dquot[cnt] != NODQUOT)
-				break;
-		if (cnt < MAXQUOTAS)
-			inode->i_sb->dq_op->drop(inode);
-	}
-}
-
 /* The following allocation/freeing/transfer functions *must* be called inside
  * a transaction (deadlocks possible otherwise) */
-static inline int DQUOT_PREALLOC_SPACE_NODIRTY(struct inode *inode, qsize_t nr)
+static inline int vfs_dq_prealloc_space_nodirty(struct inode *inode, qsize_t nr)
 {
 	if (sb_any_quota_enabled(inode->i_sb)) {
 		/* Used space is updated in alloc_space() */
@@ -102,15 +84,15 @@ static inline int DQUOT_PREALLOC_SPACE_NODIRTY(struct inode *inode, qsize_t nr)
 	return 0;
 }
 
-static inline int DQUOT_PREALLOC_SPACE(struct inode *inode, qsize_t nr)
+static inline int vfs_dq_prealloc_space(struct inode *inode, qsize_t nr)
 {
 	int ret;
-        if (!(ret =  DQUOT_PREALLOC_SPACE_NODIRTY(inode, nr)))
+        if (!(ret =  vfs_dq_prealloc_space_nodirty(inode, nr)))
 		mark_inode_dirty(inode);
 	return ret;
 }
 
-static inline int DQUOT_ALLOC_SPACE_NODIRTY(struct inode *inode, qsize_t nr)
+static inline int vfs_dq_alloc_space_nodirty(struct inode *inode, qsize_t nr)
 {
 	if (sb_any_quota_enabled(inode->i_sb)) {
 		/* Used space is updated in alloc_space() */
@@ -122,25 +104,25 @@ static inline int DQUOT_ALLOC_SPACE_NODIRTY(struct inode *inode, qsize_t nr)
 	return 0;
 }
 
-static inline int DQUOT_ALLOC_SPACE(struct inode *inode, qsize_t nr)
+static inline int vfs_dq_alloc_space(struct inode *inode, qsize_t nr)
 {
 	int ret;
-	if (!(ret = DQUOT_ALLOC_SPACE_NODIRTY(inode, nr)))
+	if (!(ret = vfs_dq_alloc_space_nodirty(inode, nr)))
 		mark_inode_dirty(inode);
 	return ret;
 }
 
-static inline int DQUOT_ALLOC_INODE(struct inode *inode)
+static inline int vfs_dq_alloc_inode(struct inode *inode)
 {
 	if (sb_any_quota_enabled(inode->i_sb)) {
-		DQUOT_INIT(inode);
+		vfs_dq_init(inode);
 		if (inode->i_sb->dq_op->alloc_inode(inode, 1) == NO_QUOTA)
 			return 1;
 	}
 	return 0;
 }
 
-static inline void DQUOT_FREE_SPACE_NODIRTY(struct inode *inode, qsize_t nr)
+static inline void vfs_dq_free_space_nodirty(struct inode *inode, qsize_t nr)
 {
 	if (sb_any_quota_enabled(inode->i_sb))
 		inode->i_sb->dq_op->free_space(inode, nr);
@@ -148,35 +130,25 @@ static inline void DQUOT_FREE_SPACE_NODIRTY(struct inode *inode, qsize_t nr)
 		inode_sub_bytes(inode, nr);
 }
 
-static inline void DQUOT_FREE_SPACE(struct inode *inode, qsize_t nr)
+static inline void vfs_dq_free_space(struct inode *inode, qsize_t nr)
 {
-	DQUOT_FREE_SPACE_NODIRTY(inode, nr);
+	vfs_dq_free_space_nodirty(inode, nr);
 	mark_inode_dirty(inode);
 }
 
-static inline void DQUOT_FREE_INODE(struct inode *inode)
+static inline void vfs_dq_free_inode(struct inode *inode)
 {
 	if (sb_any_quota_enabled(inode->i_sb))
 		inode->i_sb->dq_op->free_inode(inode, 1);
 }
 
-static inline int DQUOT_TRANSFER(struct inode *inode, struct iattr *iattr)
-{
-	if (sb_any_quota_enabled(inode->i_sb) && !IS_NOQUOTA(inode)) {
-		DQUOT_INIT(inode);
-		if (inode->i_sb->dq_op->transfer(inode, iattr) == NO_QUOTA)
-			return 1;
-	}
-	return 0;
-}
-
 /* The following two functions cannot be called inside a transaction */
-static inline void DQUOT_SYNC(struct super_block *sb)
+static inline void vfs_dq_sync(struct super_block *sb)
 {
 	sync_dquots(sb, -1);
 }
 
-static inline int DQUOT_OFF(struct super_block *sb, int remount)
+static inline int vfs_dq_off(struct super_block *sb, int remount)
 {
 	int ret = -ENOSYS;
 
@@ -185,21 +157,6 @@ static inline int DQUOT_OFF(struct super_block *sb, int remount)
 	return ret;
 }
 
-static inline int DQUOT_ON_REMOUNT(struct super_block *sb)
-{
-	int cnt;
-	int ret = 0, err;
-
-	if (!sb->s_qcop || !sb->s_qcop->quota_on)
-		return -ENOSYS;
-	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
-		err = sb->s_qcop->quota_on(sb, cnt, 0, NULL, 1);
-		if (err < 0 && !ret)
-			ret = err;
-	}
-	return ret;
-}
-
 #else
 
 /*
@@ -208,113 +165,144 @@ static inline int DQUOT_ON_REMOUNT(struct super_block *sb)
 #define sb_dquot_ops				(NULL)
 #define sb_quotactl_ops				(NULL)
 
-static inline void DQUOT_INIT(struct inode *inode)
+static inline void vfs_dq_init(struct inode *inode)
 {
 }
 
-static inline void DQUOT_DROP(struct inode *inode)
+static inline void vfs_dq_drop(struct inode *inode)
 {
 }
 
-static inline int DQUOT_ALLOC_INODE(struct inode *inode)
+static inline int vfs_dq_alloc_inode(struct inode *inode)
 {
 	return 0;
 }
 
-static inline void DQUOT_FREE_INODE(struct inode *inode)
+static inline void vfs_dq_free_inode(struct inode *inode)
 {
 }
 
-static inline void DQUOT_SYNC(struct super_block *sb)
+static inline void vfs_dq_sync(struct super_block *sb)
 {
 }
 
-static inline int DQUOT_OFF(struct super_block *sb, int remount)
+static inline int vfs_dq_off(struct super_block *sb, int remount)
 {
 	return 0;
 }
 
-static inline int DQUOT_ON_REMOUNT(struct super_block *sb)
+static inline int vfs_dq_quota_on_remount(struct super_block *sb)
 {
 	return 0;
 }
 
-static inline int DQUOT_TRANSFER(struct inode *inode, struct iattr *iattr)
+static inline int vfs_dq_transfer(struct inode *inode, struct iattr *iattr)
 {
 	return 0;
 }
 
-static inline int DQUOT_PREALLOC_SPACE_NODIRTY(struct inode *inode, qsize_t nr)
+static inline int vfs_dq_prealloc_space_nodirty(struct inode *inode, qsize_t nr)
 {
 	inode_add_bytes(inode, nr);
 	return 0;
 }
 
-static inline int DQUOT_PREALLOC_SPACE(struct inode *inode, qsize_t nr)
+static inline int vfs_dq_prealloc_space(struct inode *inode, qsize_t nr)
 {
-	DQUOT_PREALLOC_SPACE_NODIRTY(inode, nr);
+	vfs_dq_prealloc_space_nodirty(inode, nr);
 	mark_inode_dirty(inode);
 	return 0;
 }
 
-static inline int DQUOT_ALLOC_SPACE_NODIRTY(struct inode *inode, qsize_t nr)
+static inline int vfs_dq_alloc_space_nodirty(struct inode *inode, qsize_t nr)
 {
 	inode_add_bytes(inode, nr);
 	return 0;
 }
 
-static inline int DQUOT_ALLOC_SPACE(struct inode *inode, qsize_t nr)
+static inline int vfs_dq_alloc_space(struct inode *inode, qsize_t nr)
 {
-	DQUOT_ALLOC_SPACE_NODIRTY(inode, nr);
+	vfs_dq_alloc_space_nodirty(inode, nr);
 	mark_inode_dirty(inode);
 	return 0;
 }
 
-static inline void DQUOT_FREE_SPACE_NODIRTY(struct inode *inode, qsize_t nr)
+static inline void vfs_dq_free_space_nodirty(struct inode *inode, qsize_t nr)
 {
 	inode_sub_bytes(inode, nr);
 }
 
-static inline void DQUOT_FREE_SPACE(struct inode *inode, qsize_t nr)
+static inline void vfs_dq_free_space(struct inode *inode, qsize_t nr)
 {
-	DQUOT_FREE_SPACE_NODIRTY(inode, nr);
+	vfs_dq_free_space_nodirty(inode, nr);
 	mark_inode_dirty(inode);
 }	
 
 #endif /* CONFIG_QUOTA */
 
-static inline int DQUOT_PREALLOC_BLOCK_NODIRTY(struct inode *inode, qsize_t nr)
+static inline int vfs_dq_prealloc_block_nodirty(struct inode *inode, qsize_t nr)
 {
-	return DQUOT_PREALLOC_SPACE_NODIRTY(inode,
+	return vfs_dq_prealloc_space_nodirty(inode,
 			nr << inode->i_sb->s_blocksize_bits);
 }
 
-static inline int DQUOT_PREALLOC_BLOCK(struct inode *inode, qsize_t nr)
+static inline int vfs_dq_prealloc_block(struct inode *inode, qsize_t nr)
 {
-	return DQUOT_PREALLOC_SPACE(inode,
+	return vfs_dq_prealloc_space(inode,
 			nr << inode->i_sb->s_blocksize_bits);
 }
 
-static inline int DQUOT_ALLOC_BLOCK_NODIRTY(struct inode *inode, qsize_t nr)
+static inline int vfs_dq_alloc_block_nodirty(struct inode *inode, qsize_t nr)
 {
-	return DQUOT_ALLOC_SPACE_NODIRTY(inode,
+ 	return vfs_dq_alloc_space_nodirty(inode,
 			nr << inode->i_sb->s_blocksize_bits);
 }
 
-static inline int DQUOT_ALLOC_BLOCK(struct inode *inode, qsize_t nr)
+static inline int vfs_dq_alloc_block(struct inode *inode, qsize_t nr)
 {
-	return DQUOT_ALLOC_SPACE(inode,
+	return vfs_dq_alloc_space(inode,
 			nr << inode->i_sb->s_blocksize_bits);
 }
 
-static inline void DQUOT_FREE_BLOCK_NODIRTY(struct inode *inode, qsize_t nr)
+static inline void vfs_dq_free_block_nodirty(struct inode *inode, qsize_t nr)
 {
-	DQUOT_FREE_SPACE_NODIRTY(inode, nr << inode->i_sb->s_blocksize_bits);
+	vfs_dq_free_space_nodirty(inode, nr << inode->i_sb->s_blocksize_bits);
 }
 
-static inline void DQUOT_FREE_BLOCK(struct inode *inode, qsize_t nr)
+static inline void vfs_dq_free_block(struct inode *inode, qsize_t nr)
 {
-	DQUOT_FREE_SPACE(inode, nr << inode->i_sb->s_blocksize_bits);
+	vfs_dq_free_space(inode, nr << inode->i_sb->s_blocksize_bits);
 }
 
+/*
+ * Define uppercase equivalents for compatibility with old function names
+ * Can go away when we think all users have been converted (15/04/2008)
+ */
+#define DQUOT_INIT(inode) vfs_dq_init(inode)
+#define DQUOT_DROP(inode) vfs_dq_drop(inode)
+#define DQUOT_PREALLOC_SPACE_NODIRTY(inode, nr) \
+				vfs_dq_prealloc_space_nodirty(inode, nr)
+#define DQUOT_PREALLOC_SPACE(inode, nr) vfs_dq_prealloc_space(inode, nr)
+#define DQUOT_ALLOC_SPACE_NODIRTY(inode, nr) \
+				vfs_dq_alloc_space_nodirty(inode, nr)
+#define DQUOT_ALLOC_SPACE(inode, nr) vfs_dq_alloc_space(inode, nr)
+#define DQUOT_PREALLOC_BLOCK_NODIRTY(inode, nr) \
+				vfs_dq_prealloc_block_nodirty(inode, nr)
+#define DQUOT_PREALLOC_BLOCK(inode, nr) vfs_dq_prealloc_block(inode, nr)
+#define DQUOT_ALLOC_BLOCK_NODIRTY(inode, nr) \
+				vfs_dq_alloc_block_nodirty(inode, nr)
+#define DQUOT_ALLOC_BLOCK(inode, nr) vfs_dq_alloc_block(inode, nr)
+#define DQUOT_ALLOC_INODE(inode) vfs_dq_alloc_inode(inode)
+#define DQUOT_FREE_SPACE_NODIRTY(inode, nr) \
+				vfs_dq_free_space_nodirty(inode, nr)
+#define DQUOT_FREE_SPACE(inode, nr) vfs_dq_free_space(inode, nr)
+#define DQUOT_FREE_BLOCK_NODIRTY(inode, nr) \
+				vfs_dq_free_block_nodirty(inode, nr)
+#define DQUOT_FREE_BLOCK(inode, nr) vfs_dq_free_block(inode, nr)
+#define DQUOT_FREE_INODE(inode) vfs_dq_free_inode(inode)
+#define DQUOT_TRANSFER(inode, iattr) vfs_dq_transfer(inode, iattr)
+#define DQUOT_SYNC(sb) vfs_dq_sync(sb)
+#define DQUOT_OFF(sb, remount) vfs_dq_off(sb, remount)
+#define DQUOT_ON_REMOUNT(sb) vfs_dq_quota_on_remount(sb)
+
 #endif /* _LINUX_QUOTAOPS_ */

From 02a55ca87185e114e5d298a8d00608501dbabf67 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Fri, 25 Jul 2008 01:46:50 -0700
Subject: [PATCH 0774/1171] quota: cleanup loop in sync_dquots()

Make loop in sync_dquots() checking whether there's something to write
more readable, remove useless variable and macro info_any_dirty() which
is used only in this place.

Signed-off-by: Jan Kara <jack@suse.cz>
Cc: "Vegard Nossum" <vegard.nossum@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/quota.c            | 18 ++++++++++++------
 include/linux/quota.h |  2 --
 2 files changed, 12 insertions(+), 8 deletions(-)

diff --git a/fs/quota.c b/fs/quota.c
index db1cc9f3c7aa..7f4386ebc23a 100644
--- a/fs/quota.c
+++ b/fs/quota.c
@@ -186,7 +186,7 @@ static void quota_sync_sb(struct super_block *sb, int type)
 
 void sync_dquots(struct super_block *sb, int type)
 {
-	int cnt, dirty;
+	int cnt;
 
 	if (sb) {
 		if (sb->s_qcop->quota_sync)
@@ -198,11 +198,17 @@ void sync_dquots(struct super_block *sb, int type)
 restart:
 	list_for_each_entry(sb, &super_blocks, s_list) {
 		/* This test just improves performance so it needn't be reliable... */
-		for (cnt = 0, dirty = 0; cnt < MAXQUOTAS; cnt++)
-			if ((type == cnt || type == -1) && sb_has_quota_enabled(sb, cnt)
-			    && info_any_dirty(&sb_dqopt(sb)->info[cnt]))
-				dirty = 1;
-		if (!dirty)
+		for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
+			if (type != -1 && type != cnt)
+				continue;
+			if (!sb_has_quota_enabled(sb, cnt))
+				continue;
+			if (!info_dirty(&sb_dqopt(sb)->info[cnt]) &&
+			    list_empty(&sb_dqopt(sb)->info[cnt].dqi_dirty_list))
+				continue;
+			break;
+		}
+		if (cnt == MAXQUOTAS)
 			continue;
 		sb->s_count++;
 		spin_unlock(&sb_lock);
diff --git a/include/linux/quota.h b/include/linux/quota.h
index dcddfb200947..6f1d97ddf828 100644
--- a/include/linux/quota.h
+++ b/include/linux/quota.h
@@ -224,8 +224,6 @@ struct super_block;
 
 extern void mark_info_dirty(struct super_block *sb, int type);
 #define info_dirty(info) test_bit(DQF_INFO_DIRTY_B, &(info)->dqi_flags)
-#define info_any_dquot_dirty(info) (!list_empty(&(info)->dqi_dirty_list))
-#define info_any_dirty(info) (info_dirty(info) || info_any_dquot_dirty(info))
 
 #define sb_dqopt(sb) (&(sb)->s_dquot)
 #define sb_dqinfo(sb, type) (sb_dqopt(sb)->info+(type))

From 74abb9890dafb12a50dc140de215ed477beb1b88 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Fri, 25 Jul 2008 01:46:51 -0700
Subject: [PATCH 0775/1171] quota: move function-macros from quota.h to
 quotaops.h

Move declarations of some macros, which should be in fact functions to
quotaops.h.  This way they can be later converted to inline functions
because we can now use declarations from quota.h.  Also add necessary
includes of quotaops.h to a few files.

[akpm@linux-foundation.org: fix JFS build]
[akpm@linux-foundation.org: fix UFS build]
[vegard.nossum@gmail.com: fix QUOTA=n build]
Signed-off-by: Jan Kara <jack@suse.cz>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
Cc: Arjen Pool <arjenpool@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/ext2/super.c          |  1 +
 fs/jfs/super.c           |  1 +
 fs/quota_v1.c            |  1 +
 fs/quota_v2.c            |  1 +
 fs/reiserfs/super.c      |  1 +
 fs/ufs/super.c           |  1 +
 include/linux/quota.h    | 22 +++-------------------
 include/linux/quotaops.h | 26 ++++++++++++++++++++++++++
 8 files changed, 35 insertions(+), 19 deletions(-)

diff --git a/fs/ext2/super.c b/fs/ext2/super.c
index ef50cbc792db..31308a3b0b8b 100644
--- a/fs/ext2/super.c
+++ b/fs/ext2/super.c
@@ -31,6 +31,7 @@
 #include <linux/seq_file.h>
 #include <linux/mount.h>
 #include <linux/log2.h>
+#include <linux/quotaops.h>
 #include <asm/uaccess.h>
 #include "ext2.h"
 #include "xattr.h"
diff --git a/fs/jfs/super.c b/fs/jfs/super.c
index 0288e6d7936a..359c091d8965 100644
--- a/fs/jfs/super.c
+++ b/fs/jfs/super.c
@@ -22,6 +22,7 @@
 #include <linux/parser.h>
 #include <linux/completion.h>
 #include <linux/vfs.h>
+#include <linux/quotaops.h>
 #include <linux/mount.h>
 #include <linux/moduleparam.h>
 #include <linux/kthread.h>
diff --git a/fs/quota_v1.c b/fs/quota_v1.c
index a6cf9269105c..5ae15b13eeb0 100644
--- a/fs/quota_v1.c
+++ b/fs/quota_v1.c
@@ -1,6 +1,7 @@
 #include <linux/errno.h>
 #include <linux/fs.h>
 #include <linux/quota.h>
+#include <linux/quotaops.h>
 #include <linux/dqblk_v1.h>
 #include <linux/quotaio_v1.h>
 #include <linux/kernel.h>
diff --git a/fs/quota_v2.c b/fs/quota_v2.c
index 234ada903633..b53827dc02d9 100644
--- a/fs/quota_v2.c
+++ b/fs/quota_v2.c
@@ -11,6 +11,7 @@
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/slab.h>
+#include <linux/quotaops.h>
 
 #include <asm/byteorder.h>
 
diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index a10a6d2a8870..2ec748ba0bd3 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -22,6 +22,7 @@
 #include <linux/blkdev.h>
 #include <linux/buffer_head.h>
 #include <linux/exportfs.h>
+#include <linux/quotaops.h>
 #include <linux/vfs.h>
 #include <linux/mnt_namespace.h>
 #include <linux/mount.h>
diff --git a/fs/ufs/super.c b/fs/ufs/super.c
index 506f724055c2..227c9d700040 100644
--- a/fs/ufs/super.c
+++ b/fs/ufs/super.c
@@ -76,6 +76,7 @@
 
 #include <linux/errno.h>
 #include <linux/fs.h>
+#include <linux/quotaops.h>
 #include <linux/slab.h>
 #include <linux/time.h>
 #include <linux/stat.h>
diff --git a/include/linux/quota.h b/include/linux/quota.h
index 6f1d97ddf828..f9983ea0ff88 100644
--- a/include/linux/quota.h
+++ b/include/linux/quota.h
@@ -41,9 +41,6 @@
 #define __DQUOT_VERSION__	"dquot_6.5.1"
 #define __DQUOT_NUM_VERSION__	6*10000+5*100+1
 
-typedef __kernel_uid32_t qid_t; /* Type in which we store ids in memory */
-typedef __u64 qsize_t;          /* Type in which we store sizes */
-
 /* Size of blocks in which are counted size limits */
 #define QUOTABLOCK_BITS 10
 #define QUOTABLOCK_SIZE (1 << QUOTABLOCK_BITS)
@@ -172,6 +169,9 @@ enum {
 
 #include <asm/atomic.h>
 
+typedef __kernel_uid32_t qid_t; /* Type in which we store ids in memory */
+typedef __u64 qsize_t;          /* Type in which we store sizes */
+
 extern spinlock_t dq_data_lock;
 
 /* Maximal numbers of writes for quota operation (insert/delete/update)
@@ -225,9 +225,6 @@ struct super_block;
 extern void mark_info_dirty(struct super_block *sb, int type);
 #define info_dirty(info) test_bit(DQF_INFO_DIRTY_B, &(info)->dqi_flags)
 
-#define sb_dqopt(sb) (&(sb)->s_dquot)
-#define sb_dqinfo(sb, type) (sb_dqopt(sb)->info+(type))
-
 struct dqstats {
 	int lookups;
 	int drops;
@@ -335,19 +332,6 @@ struct quota_info {
 	struct quota_format_ops *ops[MAXQUOTAS];	/* Operations for each type */
 };
 
-#define sb_has_quota_enabled(sb, type) ((type)==USRQUOTA ? \
-	(sb_dqopt(sb)->flags & DQUOT_USR_ENABLED) : (sb_dqopt(sb)->flags & DQUOT_GRP_ENABLED))
-
-#define sb_any_quota_enabled(sb) (sb_has_quota_enabled(sb, USRQUOTA) | \
-				  sb_has_quota_enabled(sb, GRPQUOTA))
-
-#define sb_has_quota_suspended(sb, type) \
-	((type) == USRQUOTA ? (sb_dqopt(sb)->flags & DQUOT_USR_SUSPENDED) : \
-			      (sb_dqopt(sb)->flags & DQUOT_GRP_SUSPENDED))
-
-#define sb_any_quota_suspended(sb) (sb_has_quota_suspended(sb, USRQUOTA) | \
-				  sb_has_quota_suspended(sb, GRPQUOTA))
-
 int register_quota_format(struct quota_format_type *fmt);
 void unregister_quota_format(struct quota_format_type *fmt);
 
diff --git a/include/linux/quotaops.h b/include/linux/quotaops.h
index 0c8f9fe462af..38218c1334b1 100644
--- a/include/linux/quotaops.h
+++ b/include/linux/quotaops.h
@@ -14,6 +14,8 @@
 
 #include <linux/fs.h>
 
+#define sb_dqopt(sb) (&(sb)->s_dquot)
+
 #if defined(CONFIG_QUOTA)
 
 /*
@@ -52,6 +54,25 @@ void vfs_dq_drop(struct inode *inode);
 int vfs_dq_transfer(struct inode *inode, struct iattr *iattr);
 int vfs_dq_quota_on_remount(struct super_block *sb);
 
+#define sb_dqinfo(sb, type) (sb_dqopt(sb)->info+(type))
+
+/*
+ * Functions for checking status of quota
+ */
+
+#define sb_has_quota_enabled(sb, type) ((type)==USRQUOTA ? \
+	(sb_dqopt(sb)->flags & DQUOT_USR_ENABLED) : (sb_dqopt(sb)->flags & DQUOT_GRP_ENABLED))
+
+#define sb_any_quota_enabled(sb) (sb_has_quota_enabled(sb, USRQUOTA) | \
+				  sb_has_quota_enabled(sb, GRPQUOTA))
+
+#define sb_has_quota_suspended(sb, type) \
+	((type) == USRQUOTA ? (sb_dqopt(sb)->flags & DQUOT_USR_SUSPENDED) : \
+			      (sb_dqopt(sb)->flags & DQUOT_GRP_SUSPENDED))
+
+#define sb_any_quota_suspended(sb) (sb_has_quota_suspended(sb, USRQUOTA) | \
+				  sb_has_quota_suspended(sb, GRPQUOTA))
+
 /*
  * Operations supported for diskquotas.
  */
@@ -159,6 +180,11 @@ static inline int vfs_dq_off(struct super_block *sb, int remount)
 
 #else
 
+#define sb_has_quota_enabled(sb, type) 0
+#define sb_any_quota_enabled(sb) 0
+#define sb_has_quota_suspended(sb, type) 0
+#define sb_any_quota_suspended(sb) 0
+
 /*
  * NO-OP when quota not configured.
  */

From 03b063436ca1076301de58d9d628f610ab5404ad Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Fri, 25 Jul 2008 01:46:52 -0700
Subject: [PATCH 0776/1171] quota: convert macros to inline functions

Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/quota.h    |  5 +++-
 include/linux/quotaops.h | 65 ++++++++++++++++++++++++++++++----------
 2 files changed, 53 insertions(+), 17 deletions(-)

diff --git a/include/linux/quota.h b/include/linux/quota.h
index f9983ea0ff88..4e004fef8134 100644
--- a/include/linux/quota.h
+++ b/include/linux/quota.h
@@ -223,7 +223,10 @@ struct super_block;
 #define DQF_INFO_DIRTY (1 << DQF_INFO_DIRTY_B)	/* Is info dirty? */
 
 extern void mark_info_dirty(struct super_block *sb, int type);
-#define info_dirty(info) test_bit(DQF_INFO_DIRTY_B, &(info)->dqi_flags)
+static inline int info_dirty(struct mem_dqinfo *info)
+{
+	return test_bit(DQF_INFO_DIRTY_B, &info->dqi_flags);
+}
 
 struct dqstats {
 	int lookups;
diff --git a/include/linux/quotaops.h b/include/linux/quotaops.h
index 38218c1334b1..742187f7a05c 100644
--- a/include/linux/quotaops.h
+++ b/include/linux/quotaops.h
@@ -11,10 +11,12 @@
 #define _LINUX_QUOTAOPS_
 
 #include <linux/smp_lock.h>
-
 #include <linux/fs.h>
 
-#define sb_dqopt(sb) (&(sb)->s_dquot)
+static inline struct quota_info *sb_dqopt(struct super_block *sb)
+{
+	return &sb->s_dquot;
+}
 
 #if defined(CONFIG_QUOTA)
 
@@ -54,24 +56,40 @@ void vfs_dq_drop(struct inode *inode);
 int vfs_dq_transfer(struct inode *inode, struct iattr *iattr);
 int vfs_dq_quota_on_remount(struct super_block *sb);
 
-#define sb_dqinfo(sb, type) (sb_dqopt(sb)->info+(type))
+static inline struct mem_dqinfo *sb_dqinfo(struct super_block *sb, int type)
+{
+	return sb_dqopt(sb)->info + type;
+}
 
 /*
  * Functions for checking status of quota
  */
 
-#define sb_has_quota_enabled(sb, type) ((type)==USRQUOTA ? \
-	(sb_dqopt(sb)->flags & DQUOT_USR_ENABLED) : (sb_dqopt(sb)->flags & DQUOT_GRP_ENABLED))
+static inline int sb_has_quota_enabled(struct super_block *sb, int type)
+{
+	if (type == USRQUOTA)
+		return sb_dqopt(sb)->flags & DQUOT_USR_ENABLED;
+	return sb_dqopt(sb)->flags & DQUOT_GRP_ENABLED;
+}
 
-#define sb_any_quota_enabled(sb) (sb_has_quota_enabled(sb, USRQUOTA) | \
-				  sb_has_quota_enabled(sb, GRPQUOTA))
+static inline int sb_any_quota_enabled(struct super_block *sb)
+{
+	return sb_has_quota_enabled(sb, USRQUOTA) ||
+		sb_has_quota_enabled(sb, GRPQUOTA);
+}
 
-#define sb_has_quota_suspended(sb, type) \
-	((type) == USRQUOTA ? (sb_dqopt(sb)->flags & DQUOT_USR_SUSPENDED) : \
-			      (sb_dqopt(sb)->flags & DQUOT_GRP_SUSPENDED))
+static inline int sb_has_quota_suspended(struct super_block *sb, int type)
+{
+	if (type == USRQUOTA)
+		return sb_dqopt(sb)->flags & DQUOT_USR_SUSPENDED;
+	return sb_dqopt(sb)->flags & DQUOT_GRP_SUSPENDED;
+}
 
-#define sb_any_quota_suspended(sb) (sb_has_quota_suspended(sb, USRQUOTA) | \
-				  sb_has_quota_suspended(sb, GRPQUOTA))
+static inline int sb_any_quota_suspended(struct super_block *sb)
+{
+	return sb_has_quota_suspended(sb, USRQUOTA) ||
+		sb_has_quota_suspended(sb, GRPQUOTA);
+}
 
 /*
  * Operations supported for diskquotas.
@@ -180,10 +198,25 @@ static inline int vfs_dq_off(struct super_block *sb, int remount)
 
 #else
 
-#define sb_has_quota_enabled(sb, type) 0
-#define sb_any_quota_enabled(sb) 0
-#define sb_has_quota_suspended(sb, type) 0
-#define sb_any_quota_suspended(sb) 0
+static inline int sb_has_quota_enabled(struct super_block *sb, int type)
+{
+	return 0;
+}
+
+static inline int sb_any_quota_enabled(struct super_block *sb)
+{
+	return 0;
+}
+
+static inline int sb_has_quota_suspended(struct super_block *sb, int type)
+{
+	return 0;
+}
+
+static inline int sb_any_quota_suspended(struct super_block *sb)
+{
+	return 0;
+}
 
 /*
  * NO-OP when quota not configured.

From 657d3bfa98e542271b449f8cd84c7501ae2b2255 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Fri, 25 Jul 2008 01:46:52 -0700
Subject: [PATCH 0777/1171] quota: implement sending information via netlink
 about user below quota

Sometimes it may be useful for userspace to know (e.g.  for some hosting
guys) that some user stopped exceeding his hardlimit or softlimit in
quotas.  Implement sending of such events to userspace via quota netlink
protocol so that they don't have to poll for such events.  Based on idea
and initial implementation by Vladislav Bogdanov.

Cc: Vladislav Bogdanov <slava@nsys.by>
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/dquot.c            | 60 ++++++++++++++++++++++++++++++++++++++-----
 include/linux/quota.h |  4 +++
 2 files changed, 58 insertions(+), 6 deletions(-)

diff --git a/fs/dquot.c b/fs/dquot.c
index 0bcaf970bbb4..1346eebe74ce 100644
--- a/fs/dquot.c
+++ b/fs/dquot.c
@@ -889,7 +889,10 @@ static void print_warning(struct dquot *dquot, const int warntype)
 	char *msg = NULL;
 	struct tty_struct *tty;
 
-	if (!need_print_warning(dquot))
+	if (warntype == QUOTA_NL_IHARDBELOW ||
+	    warntype == QUOTA_NL_ISOFTBELOW ||
+	    warntype == QUOTA_NL_BHARDBELOW ||
+	    warntype == QUOTA_NL_BSOFTBELOW || !need_print_warning(dquot))
 		return;
 
 	mutex_lock(&tty_mutex);
@@ -1097,6 +1100,35 @@ static int check_bdq(struct dquot *dquot, qsize_t space, int prealloc, char *war
 	return QUOTA_OK;
 }
 
+static int info_idq_free(struct dquot *dquot, ulong inodes)
+{
+	if (test_bit(DQ_FAKE_B, &dquot->dq_flags) ||
+	    dquot->dq_dqb.dqb_curinodes <= dquot->dq_dqb.dqb_isoftlimit)
+		return QUOTA_NL_NOWARN;
+
+	if (dquot->dq_dqb.dqb_curinodes - inodes <= dquot->dq_dqb.dqb_isoftlimit)
+		return QUOTA_NL_ISOFTBELOW;
+	if (dquot->dq_dqb.dqb_curinodes >= dquot->dq_dqb.dqb_ihardlimit &&
+	    dquot->dq_dqb.dqb_curinodes - inodes < dquot->dq_dqb.dqb_ihardlimit)
+		return QUOTA_NL_IHARDBELOW;
+	return QUOTA_NL_NOWARN;
+}
+
+static int info_bdq_free(struct dquot *dquot, qsize_t space)
+{
+	if (test_bit(DQ_FAKE_B, &dquot->dq_flags) ||
+	    toqb(dquot->dq_dqb.dqb_curspace) <= dquot->dq_dqb.dqb_bsoftlimit)
+		return QUOTA_NL_NOWARN;
+
+	if (toqb(dquot->dq_dqb.dqb_curspace - space) <=
+	    dquot->dq_dqb.dqb_bsoftlimit)
+		return QUOTA_NL_BSOFTBELOW;
+	if (toqb(dquot->dq_dqb.dqb_curspace) >= dquot->dq_dqb.dqb_bhardlimit &&
+	    toqb(dquot->dq_dqb.dqb_curspace - space) <
+						dquot->dq_dqb.dqb_bhardlimit)
+		return QUOTA_NL_BHARDBELOW;
+	return QUOTA_NL_NOWARN;
+}
 /*
  *	Initialize quota pointers in inode
  *	Transaction must be started at entry
@@ -1284,6 +1316,7 @@ warn_put_all:
 int dquot_free_space(struct inode *inode, qsize_t number)
 {
 	unsigned int cnt;
+	char warntype[MAXQUOTAS];
 
 	/* First test before acquiring mutex - solves deadlocks when we
          * re-enter the quota code and are already holding the mutex */
@@ -1292,6 +1325,7 @@ out_sub:
 		inode_sub_bytes(inode, number);
 		return QUOTA_OK;
 	}
+
 	down_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
 	/* Now recheck reliably when holding dqptr_sem */
 	if (IS_NOQUOTA(inode)) {
@@ -1302,6 +1336,7 @@ out_sub:
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
 		if (inode->i_dquot[cnt] == NODQUOT)
 			continue;
+		warntype[cnt] = info_bdq_free(inode->i_dquot[cnt], number);
 		dquot_decr_space(inode->i_dquot[cnt], number);
 	}
 	inode_sub_bytes(inode, number);
@@ -1310,6 +1345,7 @@ out_sub:
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
 		if (inode->i_dquot[cnt])
 			mark_dquot_dirty(inode->i_dquot[cnt]);
+	flush_warnings(inode->i_dquot, warntype);
 	up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
 	return QUOTA_OK;
 }
@@ -1320,11 +1356,13 @@ out_sub:
 int dquot_free_inode(const struct inode *inode, unsigned long number)
 {
 	unsigned int cnt;
+	char warntype[MAXQUOTAS];
 
 	/* First test before acquiring mutex - solves deadlocks when we
          * re-enter the quota code and are already holding the mutex */
 	if (IS_NOQUOTA(inode))
 		return QUOTA_OK;
+
 	down_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
 	/* Now recheck reliably when holding dqptr_sem */
 	if (IS_NOQUOTA(inode)) {
@@ -1335,6 +1373,7 @@ int dquot_free_inode(const struct inode *inode, unsigned long number)
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
 		if (inode->i_dquot[cnt] == NODQUOT)
 			continue;
+		warntype[cnt] = info_idq_free(inode->i_dquot[cnt], number);
 		dquot_decr_inodes(inode->i_dquot[cnt], number);
 	}
 	spin_unlock(&dq_data_lock);
@@ -1342,6 +1381,7 @@ int dquot_free_inode(const struct inode *inode, unsigned long number)
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++)
 		if (inode->i_dquot[cnt])
 			mark_dquot_dirty(inode->i_dquot[cnt]);
+	flush_warnings(inode->i_dquot, warntype);
 	up_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
 	return QUOTA_OK;
 }
@@ -1359,7 +1399,8 @@ int dquot_transfer(struct inode *inode, struct iattr *iattr)
 	struct dquot *transfer_to[MAXQUOTAS];
 	int cnt, ret = NO_QUOTA, chuid = (iattr->ia_valid & ATTR_UID) && inode->i_uid != iattr->ia_uid,
 	    chgid = (iattr->ia_valid & ATTR_GID) && inode->i_gid != iattr->ia_gid;
-	char warntype[MAXQUOTAS];
+	char warntype_to[MAXQUOTAS];
+	char warntype_from_inodes[MAXQUOTAS], warntype_from_space[MAXQUOTAS];
 
 	/* First test before acquiring mutex - solves deadlocks when we
          * re-enter the quota code and are already holding the mutex */
@@ -1368,7 +1409,7 @@ int dquot_transfer(struct inode *inode, struct iattr *iattr)
 	/* Clear the arrays */
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
 		transfer_to[cnt] = transfer_from[cnt] = NODQUOT;
-		warntype[cnt] = QUOTA_NL_NOWARN;
+		warntype_to[cnt] = QUOTA_NL_NOWARN;
 	}
 	down_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
 	/* Now recheck reliably when holding dqptr_sem */
@@ -1400,8 +1441,9 @@ int dquot_transfer(struct inode *inode, struct iattr *iattr)
 		if (transfer_to[cnt] == NODQUOT)
 			continue;
 		transfer_from[cnt] = inode->i_dquot[cnt];
-		if (check_idq(transfer_to[cnt], 1, warntype+cnt) == NO_QUOTA ||
-		    check_bdq(transfer_to[cnt], space, 0, warntype+cnt) == NO_QUOTA)
+		if (check_idq(transfer_to[cnt], 1, warntype_to + cnt) ==
+		    NO_QUOTA || check_bdq(transfer_to[cnt], space, 0,
+		    warntype_to + cnt) == NO_QUOTA)
 			goto warn_put_all;
 	}
 
@@ -1417,6 +1459,10 @@ int dquot_transfer(struct inode *inode, struct iattr *iattr)
 
 		/* Due to IO error we might not have transfer_from[] structure */
 		if (transfer_from[cnt]) {
+			warntype_from_inodes[cnt] =
+				info_idq_free(transfer_from[cnt], 1);
+			warntype_from_space[cnt] =
+				info_bdq_free(transfer_from[cnt], space);
 			dquot_decr_inodes(transfer_from[cnt], 1);
 			dquot_decr_space(transfer_from[cnt], space);
 		}
@@ -1436,7 +1482,9 @@ warn_put_all:
 		if (transfer_to[cnt])
 			mark_dquot_dirty(transfer_to[cnt]);
 	}
-	flush_warnings(transfer_to, warntype);
+	flush_warnings(transfer_to, warntype_to);
+	flush_warnings(transfer_from, warntype_from_inodes);
+	flush_warnings(transfer_from, warntype_from_space);
 	
 	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
 		if (ret == QUOTA_OK && transfer_from[cnt] != NODQUOT)
diff --git a/include/linux/quota.h b/include/linux/quota.h
index 4e004fef8134..376a05048bc5 100644
--- a/include/linux/quota.h
+++ b/include/linux/quota.h
@@ -135,6 +135,10 @@ struct if_dqinfo {
 #define QUOTA_NL_BHARDWARN 4		/* Block hardlimit reached */
 #define QUOTA_NL_BSOFTLONGWARN 5	/* Block grace time expired */
 #define QUOTA_NL_BSOFTWARN 6		/* Block softlimit reached */
+#define QUOTA_NL_IHARDBELOW 7		/* Usage got below inode hardlimit */
+#define QUOTA_NL_ISOFTBELOW 8		/* Usage got below inode softlimit */
+#define QUOTA_NL_BHARDBELOW 9		/* Usage got below block hardlimit */
+#define QUOTA_NL_BSOFTBELOW 10		/* Usage got below block softlimit */
 
 enum {
 	QUOTA_NL_C_UNSPEC,

From 9d96d82da437ed5f2053821779ed5d7797ed1f81 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier@gentoo.org>
Date: Fri, 25 Jul 2008 01:46:53 -0700
Subject: [PATCH 0778/1171] procfs-guide: drop pointless &nbsp; entities

Having trailing &nbsp; entities in a revision numer seems pretty pointless
to me.  More so, it's causing me pains, so just drop them since no other
guide is doing this.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Acked-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/DocBook/procfs-guide.tmpl | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/Documentation/DocBook/procfs-guide.tmpl b/Documentation/DocBook/procfs-guide.tmpl
index 1fd6a1ec7591..8a5dc6e021ff 100644
--- a/Documentation/DocBook/procfs-guide.tmpl
+++ b/Documentation/DocBook/procfs-guide.tmpl
@@ -29,12 +29,12 @@
 
     <revhistory>
       <revision>
-	<revnumber>1.0&nbsp;</revnumber>
+	<revnumber>1.0</revnumber>
 	<date>May 30, 2001</date>
 	<revremark>Initial revision posted to linux-kernel</revremark>
       </revision>
       <revision>
-	<revnumber>1.1&nbsp;</revnumber>
+	<revnumber>1.1</revnumber>
 	<date>June 3, 2001</date>
 	<revremark>Revised after comments from linux-kernel</revremark>
       </revision>

From 7e9abd89cbdf9b73d327d8173343abce9022609b Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Fri, 25 Jul 2008 01:46:54 -0700
Subject: [PATCH 0779/1171] cgroup: use read lock to guard
 find_existing_css_set()

The function does not modify anything (except the temporary css template), so
it's sufficient to hold read lock.

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Acked-by: Paul Menage <menage@google.com>
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/cgroup.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index 15ac0e1e4f4d..f50edadfdd86 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -415,11 +415,11 @@ static struct css_set *find_css_set(
 
 	/* First see if we already have a cgroup group that matches
 	 * the desired set */
-	write_lock(&css_set_lock);
+	read_lock(&css_set_lock);
 	res = find_existing_css_set(oldcg, cgrp, template);
 	if (res)
 		get_css_set(res);
-	write_unlock(&css_set_lock);
+	read_unlock(&css_set_lock);
 
 	if (res)
 		return res;

From f2992db2a4f7ae10f61d5bc68c7c1528cec639e2 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 25 Jul 2008 01:46:55 -0700
Subject: [PATCH 0780/1171] Mark res_counter_charge(_locked) with __must_check

Ignoring their return values may result in counter underflow in the future -
when the value charged will be uncharged (or in "leaks" - when the value is
not uncharged).

This also prevents from using charging routines to decrement the
counter value (i.e. uncharge it) ;)

(Current code works OK with res_counter, however :) )

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Cc: Paul Menage <menage@google.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/res_counter.h | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/include/linux/res_counter.h b/include/linux/res_counter.h
index 6d9e1fca098c..125660e7793f 100644
--- a/include/linux/res_counter.h
+++ b/include/linux/res_counter.h
@@ -95,8 +95,10 @@ void res_counter_init(struct res_counter *counter);
  * counter->limit _locked call expects the counter->lock to be taken
  */
 
-int res_counter_charge_locked(struct res_counter *counter, unsigned long val);
-int res_counter_charge(struct res_counter *counter, unsigned long val);
+int __must_check res_counter_charge_locked(struct res_counter *counter,
+		unsigned long val);
+int __must_check res_counter_charge(struct res_counter *counter,
+		unsigned long val);
 
 /*
  * uncharge - tell that some portion of the resource is released

From 71cbb949d17d4d776abd547135feb7f3282405c8 Mon Sep 17 00:00:00 2001
From: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Date: Fri, 25 Jul 2008 01:46:55 -0700
Subject: [PATCH 0781/1171] cgroup: list_for_each cleanup

--------------------------
while() {
	list_entry();
	...
}
--------------------------

is equivalent to following code.

--------------------------
list_for_each_entry(){
	...
}
--------------------------

later can review easily more.

this patch is just clean up.
it doesn't have any behavor change.

Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Paul Menage <menage@google.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/cgroup.c | 44 +++++++++++++++++++++-----------------------
 1 file changed, 21 insertions(+), 23 deletions(-)

diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index f50edadfdd86..6836a9063634 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -241,17 +241,20 @@ static int use_task_css_set_links;
  */
 static void unlink_css_set(struct css_set *cg)
 {
+	struct cg_cgroup_link *link;
+	struct cg_cgroup_link *saved_link;
+
 	write_lock(&css_set_lock);
 	hlist_del(&cg->hlist);
 	css_set_count--;
-	while (!list_empty(&cg->cg_links)) {
-		struct cg_cgroup_link *link;
-		link = list_entry(cg->cg_links.next,
-				  struct cg_cgroup_link, cg_link_list);
+
+	list_for_each_entry_safe(link, saved_link, &cg->cg_links,
+				 cg_link_list) {
 		list_del(&link->cg_link_list);
 		list_del(&link->cgrp_link_list);
 		kfree(link);
 	}
+
 	write_unlock(&css_set_lock);
 }
 
@@ -363,15 +366,14 @@ static struct css_set *find_existing_css_set(
 static int allocate_cg_links(int count, struct list_head *tmp)
 {
 	struct cg_cgroup_link *link;
+	struct cg_cgroup_link *saved_link;
 	int i;
 	INIT_LIST_HEAD(tmp);
 	for (i = 0; i < count; i++) {
 		link = kmalloc(sizeof(*link), GFP_KERNEL);
 		if (!link) {
-			while (!list_empty(tmp)) {
-				link = list_entry(tmp->next,
-						  struct cg_cgroup_link,
-						  cgrp_link_list);
+			list_for_each_entry_safe(link, saved_link, tmp,
+						 cgrp_link_list) {
 				list_del(&link->cgrp_link_list);
 				kfree(link);
 			}
@@ -384,11 +386,10 @@ static int allocate_cg_links(int count, struct list_head *tmp)
 
 static void free_cg_links(struct list_head *tmp)
 {
-	while (!list_empty(tmp)) {
-		struct cg_cgroup_link *link;
-		link = list_entry(tmp->next,
-				  struct cg_cgroup_link,
-				  cgrp_link_list);
+	struct cg_cgroup_link *link;
+	struct cg_cgroup_link *saved_link;
+
+	list_for_each_entry_safe(link, saved_link, tmp, cgrp_link_list) {
 		list_del(&link->cgrp_link_list);
 		kfree(link);
 	}
@@ -1093,6 +1094,8 @@ static void cgroup_kill_sb(struct super_block *sb) {
 	struct cgroupfs_root *root = sb->s_fs_info;
 	struct cgroup *cgrp = &root->top_cgroup;
 	int ret;
+	struct cg_cgroup_link *link;
+	struct cg_cgroup_link *saved_link;
 
 	BUG_ON(!root);
 
@@ -1112,10 +1115,9 @@ static void cgroup_kill_sb(struct super_block *sb) {
 	 * root cgroup
 	 */
 	write_lock(&css_set_lock);
-	while (!list_empty(&cgrp->css_sets)) {
-		struct cg_cgroup_link *link;
-		link = list_entry(cgrp->css_sets.next,
-				  struct cg_cgroup_link, cgrp_link_list);
+
+	list_for_each_entry_safe(link, saved_link, &cgrp->css_sets,
+				 cgrp_link_list) {
 		list_del(&link->cg_link_list);
 		list_del(&link->cgrp_link_list);
 		kfree(link);
@@ -1756,15 +1758,11 @@ int cgroup_add_files(struct cgroup *cgrp,
 int cgroup_task_count(const struct cgroup *cgrp)
 {
 	int count = 0;
-	struct list_head *l;
+	struct cg_cgroup_link *link;
 
 	read_lock(&css_set_lock);
-	l = cgrp->css_sets.next;
-	while (l != &cgrp->css_sets) {
-		struct cg_cgroup_link *link =
-			list_entry(l, struct cg_cgroup_link, cgrp_link_list);
+	list_for_each_entry(link, &cgrp->css_sets, cgrp_link_list) {
 		count += atomic_read(&link->cg->ref.refcount);
-		l = l->next;
 	}
 	read_unlock(&css_set_lock);
 	return count;

From 8947f9d5b361ce927be6d5c11fed57905b7a4100 Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Fri, 25 Jul 2008 01:46:56 -0700
Subject: [PATCH 0782/1171] cgroups: annotate two variables with __read_mostly

- need_forkexit_callback will be read only after system boot.
- use_task_css_set_links will be read only after it's set.

And these 2 variables are checked when a new process is forked.

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Acked-by: Paul Menage <menage@google.com>
Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/cgroup.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index 6836a9063634..70d083c6fb6b 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -118,7 +118,7 @@ static int root_count;
  * extra work in the fork/exit path if none of the subsystems need to
  * be called.
  */
-static int need_forkexit_callback;
+static int need_forkexit_callback __read_mostly;
 static int need_mm_owner_callback __read_mostly;
 
 /* convenient tests for these bits */
@@ -220,7 +220,7 @@ static struct hlist_head *css_set_hash(struct cgroup_subsys_state *css[])
  * task until after the first call to cgroup_iter_start(). This
  * reduces the fork()/exit() overhead for people who have cgroups
  * compiled into their kernel but not actually in use */
-static int use_task_css_set_links;
+static int use_task_css_set_links __read_mostly;
 
 /* When we create or destroy a css_set, the operation simply
  * takes/releases a reference count on all the cgroups referenced

From ce16b49d37e748574f7fabc2726268d542d0aa1a Mon Sep 17 00:00:00 2001
From: Paul Menage <menage@google.com>
Date: Fri, 25 Jul 2008 01:46:57 -0700
Subject: [PATCH 0783/1171] cgroup files: clean up whitespace in struct cftype

This patch removes some extraneous spaces from method declarations in
struct cftype, to fit in with conventional kernel style.

Signed-off-by: Paul Menage <menage@google.com>
Cc: Paul Jackson <pj@sgi.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Serge Hallyn <serue@us.ibm.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/cgroup.h | 32 ++++++++++++++++----------------
 1 file changed, 16 insertions(+), 16 deletions(-)

diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h
index e155aa78d859..88a734edccbc 100644
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -205,48 +205,48 @@ struct cftype {
 	 * subsystem, followed by a period */
 	char name[MAX_CFTYPE_NAME];
 	int private;
-	int (*open) (struct inode *inode, struct file *file);
-	ssize_t (*read) (struct cgroup *cgrp, struct cftype *cft,
-			 struct file *file,
-			 char __user *buf, size_t nbytes, loff_t *ppos);
+	int (*open)(struct inode *inode, struct file *file);
+	ssize_t (*read)(struct cgroup *cgrp, struct cftype *cft,
+			struct file *file,
+			char __user *buf, size_t nbytes, loff_t *ppos);
 	/*
 	 * read_u64() is a shortcut for the common case of returning a
 	 * single integer. Use it in place of read()
 	 */
-	u64 (*read_u64) (struct cgroup *cgrp, struct cftype *cft);
+	u64 (*read_u64)(struct cgroup *cgrp, struct cftype *cft);
 	/*
 	 * read_s64() is a signed version of read_u64()
 	 */
-	s64 (*read_s64) (struct cgroup *cgrp, struct cftype *cft);
+	s64 (*read_s64)(struct cgroup *cgrp, struct cftype *cft);
 	/*
 	 * read_map() is used for defining a map of key/value
 	 * pairs. It should call cb->fill(cb, key, value) for each
 	 * entry. The key/value pairs (and their ordering) should not
 	 * change between reboots.
 	 */
-	int (*read_map) (struct cgroup *cont, struct cftype *cft,
-			 struct cgroup_map_cb *cb);
+	int (*read_map)(struct cgroup *cont, struct cftype *cft,
+			struct cgroup_map_cb *cb);
 	/*
 	 * read_seq_string() is used for outputting a simple sequence
 	 * using seqfile.
 	 */
-	int (*read_seq_string) (struct cgroup *cont, struct cftype *cft,
-			 struct seq_file *m);
+	int (*read_seq_string)(struct cgroup *cont, struct cftype *cft,
+			       struct seq_file *m);
 
-	ssize_t (*write) (struct cgroup *cgrp, struct cftype *cft,
-			  struct file *file,
-			  const char __user *buf, size_t nbytes, loff_t *ppos);
+	ssize_t (*write)(struct cgroup *cgrp, struct cftype *cft,
+			 struct file *file,
+			 const char __user *buf, size_t nbytes, loff_t *ppos);
 
 	/*
 	 * write_u64() is a shortcut for the common case of accepting
 	 * a single integer (as parsed by simple_strtoull) from
 	 * userspace. Use in place of write(); return 0 or error.
 	 */
-	int (*write_u64) (struct cgroup *cgrp, struct cftype *cft, u64 val);
+	int (*write_u64)(struct cgroup *cgrp, struct cftype *cft, u64 val);
 	/*
 	 * write_s64() is a signed version of write_u64()
 	 */
-	int (*write_s64) (struct cgroup *cgrp, struct cftype *cft, s64 val);
+	int (*write_s64)(struct cgroup *cgrp, struct cftype *cft, s64 val);
 
 	/*
 	 * trigger() callback can be used to get some kick from the
@@ -256,7 +256,7 @@ struct cftype {
 	 */
 	int (*trigger)(struct cgroup *cgrp, unsigned int event);
 
-	int (*release) (struct inode *inode, struct file *file);
+	int (*release)(struct inode *inode, struct file *file);
 };
 
 struct cgroup_scanner {

From db3b14978abc02041046ed8353f0899cb58ffffc Mon Sep 17 00:00:00 2001
From: Paul Menage <menage@google.com>
Date: Fri, 25 Jul 2008 01:46:58 -0700
Subject: [PATCH 0784/1171] cgroup files: add write_string cgroup control file
 method

This patch adds a write_string() method for cgroups control files. The
semantics are that a buffer is copied from userspace to kernelspace
and the handler function invoked on that buffer.  The buffer is
guaranteed to be nul-terminated, and no longer than max_write_len
(defaulting to 64 bytes if unspecified). Later patches will convert
existing raw file write handlers in control group subsystems to use
this method.

Signed-off-by: Paul Menage <menage@google.com>
Cc: Paul Jackson <pj@sgi.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Balbir Singh <balbir@in.ibm.com>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/cgroup.h | 14 ++++++++++++++
 kernel/cgroup.c        | 35 +++++++++++++++++++++++++++++++++++
 2 files changed, 49 insertions(+)

diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h
index 88a734edccbc..f5379455bb59 100644
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -205,6 +205,13 @@ struct cftype {
 	 * subsystem, followed by a period */
 	char name[MAX_CFTYPE_NAME];
 	int private;
+
+	/*
+	 * If non-zero, defines the maximum length of string that can
+	 * be passed to write_string; defaults to 64
+	 */
+	size_t max_write_len;
+
 	int (*open)(struct inode *inode, struct file *file);
 	ssize_t (*read)(struct cgroup *cgrp, struct cftype *cft,
 			struct file *file,
@@ -248,6 +255,13 @@ struct cftype {
 	 */
 	int (*write_s64)(struct cgroup *cgrp, struct cftype *cft, s64 val);
 
+	/*
+	 * write_string() is passed a nul-terminated kernelspace
+	 * buffer of maximum length determined by max_write_len.
+	 * Returns 0 or -ve error code.
+	 */
+	int (*write_string)(struct cgroup *cgrp, struct cftype *cft,
+			    const char *buffer);
 	/*
 	 * trigger() callback can be used to get some kick from the
 	 * userspace, when the actual string written is not important
diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index 70d083c6fb6b..3a99cc2df860 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -1363,6 +1363,39 @@ static ssize_t cgroup_write_X64(struct cgroup *cgrp, struct cftype *cft,
 	return retval;
 }
 
+static ssize_t cgroup_write_string(struct cgroup *cgrp, struct cftype *cft,
+				   struct file *file,
+				   const char __user *userbuf,
+				   size_t nbytes, loff_t *unused_ppos)
+{
+	char local_buffer[64];
+	int retval = 0;
+	size_t max_bytes = cft->max_write_len;
+	char *buffer = local_buffer;
+
+	if (!max_bytes)
+		max_bytes = sizeof(local_buffer) - 1;
+	if (nbytes >= max_bytes)
+		return -E2BIG;
+	/* Allocate a dynamic buffer if we need one */
+	if (nbytes >= sizeof(local_buffer)) {
+		buffer = kmalloc(nbytes + 1, GFP_KERNEL);
+		if (buffer == NULL)
+			return -ENOMEM;
+	}
+	if (nbytes && copy_from_user(buffer, userbuf, nbytes))
+		return -EFAULT;
+
+	buffer[nbytes] = 0;     /* nul-terminate */
+	strstrip(buffer);
+	retval = cft->write_string(cgrp, cft, buffer);
+	if (!retval)
+		retval = nbytes;
+	if (buffer != local_buffer)
+		kfree(buffer);
+	return retval;
+}
+
 static ssize_t cgroup_common_file_write(struct cgroup *cgrp,
 					   struct cftype *cft,
 					   struct file *file,
@@ -1440,6 +1473,8 @@ static ssize_t cgroup_file_write(struct file *file, const char __user *buf,
 		return cft->write(cgrp, cft, file, buf, nbytes, ppos);
 	if (cft->write_u64 || cft->write_s64)
 		return cgroup_write_X64(cgrp, cft, file, buf, nbytes, ppos);
+	if (cft->write_string)
+		return cgroup_write_string(cgrp, cft, file, buf, nbytes, ppos);
 	if (cft->trigger) {
 		int ret = cft->trigger(cgrp, (unsigned int)cft->private);
 		return ret ? ret : nbytes;

From e788e066c651b1bbf4a927dc95395c1aa13be436 Mon Sep 17 00:00:00 2001
From: Paul Menage <menage@google.com>
Date: Fri, 25 Jul 2008 01:46:59 -0700
Subject: [PATCH 0785/1171] cgroup files: move the release_agent file to use
 typed handlers

Adds cgroup_release_agent_write() and cgroup_release_agent_show()
methods to handle writing/reading the path to a cgroup hierarchy's
release agent. As a result, cgroup_common_file_read() is now unnecessary.

As part of the change, a previously-tolerated race in
cgroup_release_agent() is avoided by copying the current
release_agent_path prior to calling call_usermode_helper().

Signed-off-by: Paul Menage <menage@google.com>
Cc: Paul Jackson <pj@sgi.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: Balbir Singh <balbir@in.ibm.com>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/cgroup.h |   2 +
 kernel/cgroup.c        | 125 +++++++++++++++++++----------------------
 2 files changed, 59 insertions(+), 68 deletions(-)

diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h
index f5379455bb59..e78377a91a74 100644
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -295,6 +295,8 @@ int cgroup_add_files(struct cgroup *cgrp,
 
 int cgroup_is_removed(const struct cgroup *cgrp);
 
+int cgroup_lock_live_group(struct cgroup *cgrp);
+
 int cgroup_path(const struct cgroup *cgrp, char *buf, int buflen);
 
 int cgroup_task_count(const struct cgroup *cgrp);
diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index 3a99cc2df860..0120b5d67a73 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -89,11 +89,7 @@ struct cgroupfs_root {
 	/* Hierarchy-specific flags */
 	unsigned long flags;
 
-	/* The path to use for release notifications. No locking
-	 * between setting and use - so if userspace updates this
-	 * while child cgroups exist, you could miss a
-	 * notification. We ensure that it's always a valid
-	 * NUL-terminated string */
+	/* The path to use for release notifications. */
 	char release_agent_path[PATH_MAX];
 };
 
@@ -1329,6 +1325,45 @@ enum cgroup_filetype {
 	FILE_RELEASE_AGENT,
 };
 
+/**
+ * cgroup_lock_live_group - take cgroup_mutex and check that cgrp is alive.
+ * @cgrp: the cgroup to be checked for liveness
+ *
+ * Returns true (with lock held) on success, or false (with no lock
+ * held) on failure.
+ */
+int cgroup_lock_live_group(struct cgroup *cgrp)
+{
+	mutex_lock(&cgroup_mutex);
+	if (cgroup_is_removed(cgrp)) {
+		mutex_unlock(&cgroup_mutex);
+		return false;
+	}
+	return true;
+}
+
+static int cgroup_release_agent_write(struct cgroup *cgrp, struct cftype *cft,
+				      const char *buffer)
+{
+	BUILD_BUG_ON(sizeof(cgrp->root->release_agent_path) < PATH_MAX);
+	if (!cgroup_lock_live_group(cgrp))
+		return -ENODEV;
+	strcpy(cgrp->root->release_agent_path, buffer);
+	mutex_unlock(&cgroup_mutex);
+	return 0;
+}
+
+static int cgroup_release_agent_show(struct cgroup *cgrp, struct cftype *cft,
+				     struct seq_file *seq)
+{
+	if (!cgroup_lock_live_group(cgrp))
+		return -ENODEV;
+	seq_puts(seq, cgrp->root->release_agent_path);
+	seq_putc(seq, '\n');
+	mutex_unlock(&cgroup_mutex);
+	return 0;
+}
+
 static ssize_t cgroup_write_X64(struct cgroup *cgrp, struct cftype *cft,
 				struct file *file,
 				const char __user *userbuf,
@@ -1443,10 +1478,6 @@ static ssize_t cgroup_common_file_write(struct cgroup *cgrp,
 		else
 			clear_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags);
 		break;
-	case FILE_RELEASE_AGENT:
-		BUILD_BUG_ON(sizeof(cgrp->root->release_agent_path) < PATH_MAX);
-		strcpy(cgrp->root->release_agent_path, buffer);
-		break;
 	default:
 		retval = -EINVAL;
 		goto out2;
@@ -1506,49 +1537,6 @@ static ssize_t cgroup_read_s64(struct cgroup *cgrp, struct cftype *cft,
 	return simple_read_from_buffer(buf, nbytes, ppos, tmp, len);
 }
 
-static ssize_t cgroup_common_file_read(struct cgroup *cgrp,
-					  struct cftype *cft,
-					  struct file *file,
-					  char __user *buf,
-					  size_t nbytes, loff_t *ppos)
-{
-	enum cgroup_filetype type = cft->private;
-	char *page;
-	ssize_t retval = 0;
-	char *s;
-
-	if (!(page = (char *)__get_free_page(GFP_KERNEL)))
-		return -ENOMEM;
-
-	s = page;
-
-	switch (type) {
-	case FILE_RELEASE_AGENT:
-	{
-		struct cgroupfs_root *root;
-		size_t n;
-		mutex_lock(&cgroup_mutex);
-		root = cgrp->root;
-		n = strnlen(root->release_agent_path,
-			    sizeof(root->release_agent_path));
-		n = min(n, (size_t) PAGE_SIZE);
-		strncpy(s, root->release_agent_path, n);
-		mutex_unlock(&cgroup_mutex);
-		s += n;
-		break;
-	}
-	default:
-		retval = -EINVAL;
-		goto out;
-	}
-	*s++ = '\n';
-
-	retval = simple_read_from_buffer(buf, nbytes, ppos, page, s - page);
-out:
-	free_page((unsigned long)page);
-	return retval;
-}
-
 static ssize_t cgroup_file_read(struct file *file, char __user *buf,
 				   size_t nbytes, loff_t *ppos)
 {
@@ -1606,6 +1594,7 @@ int cgroup_seqfile_release(struct inode *inode, struct file *file)
 
 static struct file_operations cgroup_seqfile_operations = {
 	.read = seq_read,
+	.write = cgroup_file_write,
 	.llseek = seq_lseek,
 	.release = cgroup_seqfile_release,
 };
@@ -2283,8 +2272,9 @@ static struct cftype files[] = {
 
 static struct cftype cft_release_agent = {
 	.name = "release_agent",
-	.read = cgroup_common_file_read,
-	.write = cgroup_common_file_write,
+	.read_seq_string = cgroup_release_agent_show,
+	.write_string = cgroup_release_agent_write,
+	.max_write_len = PATH_MAX,
 	.private = FILE_RELEASE_AGENT,
 };
 
@@ -3111,27 +3101,24 @@ static void cgroup_release_agent(struct work_struct *work)
 	while (!list_empty(&release_list)) {
 		char *argv[3], *envp[3];
 		int i;
-		char *pathbuf;
+		char *pathbuf = NULL, *agentbuf = NULL;
 		struct cgroup *cgrp = list_entry(release_list.next,
 						    struct cgroup,
 						    release_list);
 		list_del_init(&cgrp->release_list);
 		spin_unlock(&release_list_lock);
 		pathbuf = kmalloc(PAGE_SIZE, GFP_KERNEL);
-		if (!pathbuf) {
-			spin_lock(&release_list_lock);
-			continue;
-		}
-
-		if (cgroup_path(cgrp, pathbuf, PAGE_SIZE) < 0) {
-			kfree(pathbuf);
-			spin_lock(&release_list_lock);
-			continue;
-		}
+		if (!pathbuf)
+			goto continue_free;
+		if (cgroup_path(cgrp, pathbuf, PAGE_SIZE) < 0)
+			goto continue_free;
+		agentbuf = kstrdup(cgrp->root->release_agent_path, GFP_KERNEL);
+		if (!agentbuf)
+			goto continue_free;
 
 		i = 0;
-		argv[i++] = cgrp->root->release_agent_path;
-		argv[i++] = (char *)pathbuf;
+		argv[i++] = agentbuf;
+		argv[i++] = pathbuf;
 		argv[i] = NULL;
 
 		i = 0;
@@ -3145,8 +3132,10 @@ static void cgroup_release_agent(struct work_struct *work)
 		 * be a slow process */
 		mutex_unlock(&cgroup_mutex);
 		call_usermodehelper(argv[0], argv, envp, UMH_WAIT_EXEC);
-		kfree(pathbuf);
 		mutex_lock(&cgroup_mutex);
+ continue_free:
+		kfree(pathbuf);
+		kfree(agentbuf);
 		spin_lock(&release_list_lock);
 	}
 	spin_unlock(&release_list_lock);

From 84eea842886ac35020be6043e04748ed22014359 Mon Sep 17 00:00:00 2001
From: Paul Menage <menage@google.com>
Date: Fri, 25 Jul 2008 01:47:00 -0700
Subject: [PATCH 0786/1171] cgroups: misc cleanups to write_string patchset

This patch contains cleanups suggested by reviewers for the recent
write_string() patchset:

- pair cgroup_lock_live_group() with cgroup_unlock() in cgroup.c for
  clarity, rather than directly unlocking cgroup_mutex.

- make the return type of cgroup_lock_live_group() a bool

- use a #define'd constant for the local buffer size in read/write functions

Signed-off-by: Paul Menage <menage@google.com>
Cc: Paul Jackson <pj@sgi.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: Balbir Singh <balbir@in.ibm.com>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/cgroup.h |  4 ++--
 kernel/cgroup.c        | 21 ++++++++++++---------
 2 files changed, 14 insertions(+), 11 deletions(-)

diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h
index e78377a91a74..cc59d3a21d87 100644
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -21,11 +21,13 @@
 struct cgroupfs_root;
 struct cgroup_subsys;
 struct inode;
+struct cgroup;
 
 extern int cgroup_init_early(void);
 extern int cgroup_init(void);
 extern void cgroup_init_smp(void);
 extern void cgroup_lock(void);
+extern bool cgroup_lock_live_group(struct cgroup *cgrp);
 extern void cgroup_unlock(void);
 extern void cgroup_fork(struct task_struct *p);
 extern void cgroup_fork_callbacks(struct task_struct *p);
@@ -295,8 +297,6 @@ int cgroup_add_files(struct cgroup *cgrp,
 
 int cgroup_is_removed(const struct cgroup *cgrp);
 
-int cgroup_lock_live_group(struct cgroup *cgrp);
-
 int cgroup_path(const struct cgroup *cgrp, char *buf, int buflen);
 
 int cgroup_task_count(const struct cgroup *cgrp);
diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index 0120b5d67a73..a14122ecaa5e 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -1329,10 +1329,10 @@ enum cgroup_filetype {
  * cgroup_lock_live_group - take cgroup_mutex and check that cgrp is alive.
  * @cgrp: the cgroup to be checked for liveness
  *
- * Returns true (with lock held) on success, or false (with no lock
- * held) on failure.
+ * On success, returns true; the lock should be later released with
+ * cgroup_unlock(). On failure returns false with no lock held.
  */
-int cgroup_lock_live_group(struct cgroup *cgrp)
+bool cgroup_lock_live_group(struct cgroup *cgrp)
 {
 	mutex_lock(&cgroup_mutex);
 	if (cgroup_is_removed(cgrp)) {
@@ -1349,7 +1349,7 @@ static int cgroup_release_agent_write(struct cgroup *cgrp, struct cftype *cft,
 	if (!cgroup_lock_live_group(cgrp))
 		return -ENODEV;
 	strcpy(cgrp->root->release_agent_path, buffer);
-	mutex_unlock(&cgroup_mutex);
+	cgroup_unlock();
 	return 0;
 }
 
@@ -1360,16 +1360,19 @@ static int cgroup_release_agent_show(struct cgroup *cgrp, struct cftype *cft,
 		return -ENODEV;
 	seq_puts(seq, cgrp->root->release_agent_path);
 	seq_putc(seq, '\n');
-	mutex_unlock(&cgroup_mutex);
+	cgroup_unlock();
 	return 0;
 }
 
+/* A buffer size big enough for numbers or short strings */
+#define CGROUP_LOCAL_BUFFER_SIZE 64
+
 static ssize_t cgroup_write_X64(struct cgroup *cgrp, struct cftype *cft,
 				struct file *file,
 				const char __user *userbuf,
 				size_t nbytes, loff_t *unused_ppos)
 {
-	char buffer[64];
+	char buffer[CGROUP_LOCAL_BUFFER_SIZE];
 	int retval = 0;
 	char *end;
 
@@ -1403,7 +1406,7 @@ static ssize_t cgroup_write_string(struct cgroup *cgrp, struct cftype *cft,
 				   const char __user *userbuf,
 				   size_t nbytes, loff_t *unused_ppos)
 {
-	char local_buffer[64];
+	char local_buffer[CGROUP_LOCAL_BUFFER_SIZE];
 	int retval = 0;
 	size_t max_bytes = cft->max_write_len;
 	char *buffer = local_buffer;
@@ -1518,7 +1521,7 @@ static ssize_t cgroup_read_u64(struct cgroup *cgrp, struct cftype *cft,
 			       char __user *buf, size_t nbytes,
 			       loff_t *ppos)
 {
-	char tmp[64];
+	char tmp[CGROUP_LOCAL_BUFFER_SIZE];
 	u64 val = cft->read_u64(cgrp, cft);
 	int len = sprintf(tmp, "%llu\n", (unsigned long long) val);
 
@@ -1530,7 +1533,7 @@ static ssize_t cgroup_read_s64(struct cgroup *cgrp, struct cftype *cft,
 			       char __user *buf, size_t nbytes,
 			       loff_t *ppos)
 {
-	char tmp[64];
+	char tmp[CGROUP_LOCAL_BUFFER_SIZE];
 	s64 val = cft->read_s64(cgrp, cft);
 	int len = sprintf(tmp, "%lld\n", (long long) val);
 

From 6379c106152388f7ea45d6dda63edda0e9181fc8 Mon Sep 17 00:00:00 2001
From: Paul Menage <menage@google.com>
Date: Fri, 25 Jul 2008 01:47:01 -0700
Subject: [PATCH 0787/1171] cgroup files: move notify_on_release file to
 separate write handler

This patch moves the write handler for the cgroups notify_on_release
file into a separate handler. This handler requires no cgroups locking
since it relies on atomic bitops for synchronization.

Signed-off-by: Paul Menage <menage@google.com>
Cc: Paul Jackson <pj@sgi.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Serge Hallyn <serue@us.ibm.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/cgroup.c | 21 +++++++++++++--------
 1 file changed, 13 insertions(+), 8 deletions(-)

diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index a14122ecaa5e..d597d3015786 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -1474,13 +1474,6 @@ static ssize_t cgroup_common_file_write(struct cgroup *cgrp,
 	case FILE_TASKLIST:
 		retval = attach_task_by_pid(cgrp, buffer);
 		break;
-	case FILE_NOTIFY_ON_RELEASE:
-		clear_bit(CGRP_RELEASABLE, &cgrp->flags);
-		if (simple_strtoul(buffer, NULL, 10) != 0)
-			set_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags);
-		else
-			clear_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags);
-		break;
 	default:
 		retval = -EINVAL;
 		goto out2;
@@ -2252,6 +2245,18 @@ static u64 cgroup_read_notify_on_release(struct cgroup *cgrp,
 	return notify_on_release(cgrp);
 }
 
+static int cgroup_write_notify_on_release(struct cgroup *cgrp,
+					  struct cftype *cft,
+					  u64 val)
+{
+	clear_bit(CGRP_RELEASABLE, &cgrp->flags);
+	if (val)
+		set_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags);
+	else
+		clear_bit(CGRP_NOTIFY_ON_RELEASE, &cgrp->flags);
+	return 0;
+}
+
 /*
  * for the common functions, 'private' gives the type of file
  */
@@ -2268,7 +2273,7 @@ static struct cftype files[] = {
 	{
 		.name = "notify_on_release",
 		.read_u64 = cgroup_read_notify_on_release,
-		.write = cgroup_common_file_write,
+		.write_u64 = cgroup_write_notify_on_release,
 		.private = FILE_NOTIFY_ON_RELEASE,
 	},
 };

From af351026aafc8da16518a02b41c66d3e0c1cdef4 Mon Sep 17 00:00:00 2001
From: Paul Menage <menage@google.com>
Date: Fri, 25 Jul 2008 01:47:01 -0700
Subject: [PATCH 0788/1171] cgroup files: turn attach_task_by_pid directly into
 a cgroup write handler

This patch changes attach_task_by_pid() to take a u64 rather than a
string; as a result it can be called directly as a control groups
write_u64 handler, and cgroup_common_file_write() can be removed.

Signed-off-by: Paul Menage <menage@google.com>
Cc: Paul Jackson <pj@sgi.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Serge Hallyn <serue@us.ibm.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/cgroup.c | 80 +++++++++----------------------------------------
 1 file changed, 14 insertions(+), 66 deletions(-)

diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index d597d3015786..86b71e714e13 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -504,10 +504,6 @@ static struct css_set *find_css_set(
  * knows that the cgroup won't be removed, as cgroup_rmdir()
  * needs that mutex.
  *
- * The cgroup_common_file_write handler for operations that modify
- * the cgroup hierarchy holds cgroup_mutex across the entire operation,
- * single threading all such cgroup modifications across the system.
- *
  * The fork and exit callbacks cgroup_fork() and cgroup_exit(), don't
  * (usually) take cgroup_mutex.  These are the two most performance
  * critical pieces of code here.  The exception occurs on cgroup_exit(),
@@ -1279,18 +1275,14 @@ int cgroup_attach_task(struct cgroup *cgrp, struct task_struct *tsk)
 }
 
 /*
- * Attach task with pid 'pid' to cgroup 'cgrp'. Call with
- * cgroup_mutex, may take task_lock of task
+ * Attach task with pid 'pid' to cgroup 'cgrp'. Call with cgroup_mutex
+ * held. May take task_lock of task
  */
-static int attach_task_by_pid(struct cgroup *cgrp, char *pidbuf)
+static int attach_task_by_pid(struct cgroup *cgrp, u64 pid)
 {
-	pid_t pid;
 	struct task_struct *tsk;
 	int ret;
 
-	if (sscanf(pidbuf, "%d", &pid) != 1)
-		return -EIO;
-
 	if (pid) {
 		rcu_read_lock();
 		tsk = find_task_by_vpid(pid);
@@ -1316,6 +1308,16 @@ static int attach_task_by_pid(struct cgroup *cgrp, char *pidbuf)
 	return ret;
 }
 
+static int cgroup_tasks_write(struct cgroup *cgrp, struct cftype *cft, u64 pid)
+{
+	int ret;
+	if (!cgroup_lock_live_group(cgrp))
+		return -ENODEV;
+	ret = attach_task_by_pid(cgrp, pid);
+	cgroup_unlock();
+	return ret;
+}
+
 /* The various types of files and directories in a cgroup file system */
 enum cgroup_filetype {
 	FILE_ROOT,
@@ -1434,60 +1436,6 @@ static ssize_t cgroup_write_string(struct cgroup *cgrp, struct cftype *cft,
 	return retval;
 }
 
-static ssize_t cgroup_common_file_write(struct cgroup *cgrp,
-					   struct cftype *cft,
-					   struct file *file,
-					   const char __user *userbuf,
-					   size_t nbytes, loff_t *unused_ppos)
-{
-	enum cgroup_filetype type = cft->private;
-	char *buffer;
-	int retval = 0;
-
-	if (nbytes >= PATH_MAX)
-		return -E2BIG;
-
-	/* +1 for nul-terminator */
-	buffer = kmalloc(nbytes + 1, GFP_KERNEL);
-	if (buffer == NULL)
-		return -ENOMEM;
-
-	if (copy_from_user(buffer, userbuf, nbytes)) {
-		retval = -EFAULT;
-		goto out1;
-	}
-	buffer[nbytes] = 0;	/* nul-terminate */
-	strstrip(buffer);	/* strip -just- trailing whitespace */
-
-	mutex_lock(&cgroup_mutex);
-
-	/*
-	 * This was already checked for in cgroup_file_write(), but
-	 * check again now we're holding cgroup_mutex.
-	 */
-	if (cgroup_is_removed(cgrp)) {
-		retval = -ENODEV;
-		goto out2;
-	}
-
-	switch (type) {
-	case FILE_TASKLIST:
-		retval = attach_task_by_pid(cgrp, buffer);
-		break;
-	default:
-		retval = -EINVAL;
-		goto out2;
-	}
-
-	if (retval == 0)
-		retval = nbytes;
-out2:
-	mutex_unlock(&cgroup_mutex);
-out1:
-	kfree(buffer);
-	return retval;
-}
-
 static ssize_t cgroup_file_write(struct file *file, const char __user *buf,
 						size_t nbytes, loff_t *ppos)
 {
@@ -2265,7 +2213,7 @@ static struct cftype files[] = {
 		.name = "tasks",
 		.open = cgroup_tasks_open,
 		.read = cgroup_tasks_read,
-		.write = cgroup_common_file_write,
+		.write_u64 = cgroup_tasks_write,
 		.release = cgroup_tasks_release,
 		.private = FILE_TASKLIST,
 	},

From e37123953292146445c8629b3950d0513fd10ae2 Mon Sep 17 00:00:00 2001
From: Paul Menage <menage@google.com>
Date: Fri, 25 Jul 2008 01:47:02 -0700
Subject: [PATCH 0789/1171] cgroup files: remove cpuset_common_file_write()

This patch tweaks the signatures of the update_cpumask() and
update_nodemask() functions so that they can be called directly as
handlers for the new cgroups write_string() method.

This allows cpuset_common_file_write() to be removed.

Signed-off-by: Paul Menage <menage@google.com>
Cc: Paul Jackson <pj@sgi.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Serge Hallyn <serue@us.ibm.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/cpuset.c | 109 ++++++++++++++++--------------------------------
 1 file changed, 35 insertions(+), 74 deletions(-)

diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index d5738910c34c..276ce7e4f1ab 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -227,10 +227,6 @@ static struct cpuset top_cpuset = {
  * The task_struct fields mems_allowed and mems_generation may only
  * be accessed in the context of that task, so require no locks.
  *
- * The cpuset_common_file_write handler for operations that modify
- * the cpuset hierarchy holds cgroup_mutex across the entire operation,
- * single threading all such cpuset modifications across the system.
- *
  * The cpuset_common_file_read() handlers only hold callback_mutex across
  * small pieces of code, such as when reading out possibly multi-word
  * cpumasks and nodemasks.
@@ -772,7 +768,7 @@ static void cpuset_change_cpumask(struct task_struct *tsk,
  * @cs: the cpuset to consider
  * @buf: buffer of cpu numbers written to this cpuset
  */
-static int update_cpumask(struct cpuset *cs, char *buf)
+static int update_cpumask(struct cpuset *cs, const char *buf)
 {
 	struct cpuset trialcs;
 	struct cgroup_scanner scan;
@@ -792,7 +788,6 @@ static int update_cpumask(struct cpuset *cs, char *buf)
 	 * that parsing.  The validate_change() call ensures that cpusets
 	 * with tasks have cpus.
 	 */
-	buf = strstrip(buf);
 	if (!*buf) {
 		cpus_clear(trialcs.cpus_allowed);
 	} else {
@@ -902,7 +897,7 @@ static void cpuset_migrate_mm(struct mm_struct *mm, const nodemask_t *from,
 
 static void *cpuset_being_rebound;
 
-static int update_nodemask(struct cpuset *cs, char *buf)
+static int update_nodemask(struct cpuset *cs, const char *buf)
 {
 	struct cpuset trialcs;
 	nodemask_t oldmem;
@@ -929,7 +924,6 @@ static int update_nodemask(struct cpuset *cs, char *buf)
 	 * that parsing.  The validate_change() call ensures that cpusets
 	 * with tasks have memory.
 	 */
-	buf = strstrip(buf);
 	if (!*buf) {
 		nodes_clear(trialcs.mems_allowed);
 	} else {
@@ -1256,72 +1250,14 @@ typedef enum {
 	FILE_SPREAD_SLAB,
 } cpuset_filetype_t;
 
-static ssize_t cpuset_common_file_write(struct cgroup *cont,
-					struct cftype *cft,
-					struct file *file,
-					const char __user *userbuf,
-					size_t nbytes, loff_t *unused_ppos)
-{
-	struct cpuset *cs = cgroup_cs(cont);
-	cpuset_filetype_t type = cft->private;
-	char *buffer;
-	int retval = 0;
-
-	/* Crude upper limit on largest legitimate cpulist user might write. */
-	if (nbytes > 100U + 6 * max(NR_CPUS, MAX_NUMNODES))
-		return -E2BIG;
-
-	/* +1 for nul-terminator */
-	buffer = kmalloc(nbytes + 1, GFP_KERNEL);
-	if (!buffer)
-		return -ENOMEM;
-
-	if (copy_from_user(buffer, userbuf, nbytes)) {
-		retval = -EFAULT;
-		goto out1;
-	}
-	buffer[nbytes] = 0;	/* nul-terminate */
-
-	cgroup_lock();
-
-	if (cgroup_is_removed(cont)) {
-		retval = -ENODEV;
-		goto out2;
-	}
-
-	switch (type) {
-	case FILE_CPULIST:
-		retval = update_cpumask(cs, buffer);
-		break;
-	case FILE_MEMLIST:
-		retval = update_nodemask(cs, buffer);
-		break;
-	default:
-		retval = -EINVAL;
-		goto out2;
-	}
-
-	if (retval == 0)
-		retval = nbytes;
-out2:
-	cgroup_unlock();
-out1:
-	kfree(buffer);
-	return retval;
-}
-
 static int cpuset_write_u64(struct cgroup *cgrp, struct cftype *cft, u64 val)
 {
 	int retval = 0;
 	struct cpuset *cs = cgroup_cs(cgrp);
 	cpuset_filetype_t type = cft->private;
 
-	cgroup_lock();
-
-	if (cgroup_is_removed(cgrp)) {
-		cgroup_unlock();
+	if (!cgroup_lock_live_group(cgrp))
 		return -ENODEV;
-	}
 
 	switch (type) {
 	case FILE_CPU_EXCLUSIVE:
@@ -1367,12 +1303,9 @@ static int cpuset_write_s64(struct cgroup *cgrp, struct cftype *cft, s64 val)
 	struct cpuset *cs = cgroup_cs(cgrp);
 	cpuset_filetype_t type = cft->private;
 
-	cgroup_lock();
-
-	if (cgroup_is_removed(cgrp)) {
-		cgroup_unlock();
+	if (!cgroup_lock_live_group(cgrp))
 		return -ENODEV;
-	}
+
 	switch (type) {
 	case FILE_SCHED_RELAX_DOMAIN_LEVEL:
 		retval = update_relax_domain_level(cs, val);
@@ -1385,6 +1318,32 @@ static int cpuset_write_s64(struct cgroup *cgrp, struct cftype *cft, s64 val)
 	return retval;
 }
 
+/*
+ * Common handling for a write to a "cpus" or "mems" file.
+ */
+static int cpuset_write_resmask(struct cgroup *cgrp, struct cftype *cft,
+				const char *buf)
+{
+	int retval = 0;
+
+	if (!cgroup_lock_live_group(cgrp))
+		return -ENODEV;
+
+	switch (cft->private) {
+	case FILE_CPULIST:
+		retval = update_cpumask(cgroup_cs(cgrp), buf);
+		break;
+	case FILE_MEMLIST:
+		retval = update_nodemask(cgroup_cs(cgrp), buf);
+		break;
+	default:
+		retval = -EINVAL;
+		break;
+	}
+	cgroup_unlock();
+	return retval;
+}
+
 /*
  * These ascii lists should be read in a single call, by using a user
  * buffer large enough to hold the entire map.  If read in smaller
@@ -1504,14 +1463,16 @@ static struct cftype files[] = {
 	{
 		.name = "cpus",
 		.read = cpuset_common_file_read,
-		.write = cpuset_common_file_write,
+		.write_string = cpuset_write_resmask,
+		.max_write_len = (100U + 6 * NR_CPUS),
 		.private = FILE_CPULIST,
 	},
 
 	{
 		.name = "mems",
 		.read = cpuset_common_file_read,
-		.write = cpuset_common_file_write,
+		.write_string = cpuset_write_resmask,
+		.max_write_len = (100U + 6 * MAX_NUMNODES),
 		.private = FILE_MEMLIST,
 	},
 

From f92523e3a7861f5dbd76021e0719a35fe8771f2d Mon Sep 17 00:00:00 2001
From: Paul Menage <menage@google.com>
Date: Fri, 25 Jul 2008 01:47:03 -0700
Subject: [PATCH 0790/1171] cgroup files: convert devcgroup_access_write() into
 a cgroup write_string() handler

This patch converts devcgroup_access_write() from a raw file handler
into a handler for the cgroup write_string() method. This allows some
boilerplate copying/locking/checking to be removed and simplifies the
cleanup path, since these functions are performed by the cgroups
framework before calling the handler.

Signed-off-by: Paul Menage <menage@google.com>
Cc: Paul Jackson <pj@sgi.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: Balbir Singh <balbir@in.ibm.com>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 security/device_cgroup.c | 101 +++++++++++++++------------------------
 1 file changed, 38 insertions(+), 63 deletions(-)

diff --git a/security/device_cgroup.c b/security/device_cgroup.c
index ddd92cec78ed..236fffa9d05e 100644
--- a/security/device_cgroup.c
+++ b/security/device_cgroup.c
@@ -59,6 +59,11 @@ static inline struct dev_cgroup *cgroup_to_devcgroup(struct cgroup *cgroup)
 	return css_to_devcgroup(cgroup_subsys_state(cgroup, devices_subsys_id));
 }
 
+static inline struct dev_cgroup *task_devcgroup(struct task_struct *task)
+{
+	return css_to_devcgroup(task_subsys_state(task, devices_subsys_id));
+}
+
 struct cgroup_subsys devices_subsys;
 
 static int devcgroup_can_attach(struct cgroup_subsys *ss,
@@ -312,10 +317,10 @@ static int may_access_whitelist(struct dev_cgroup *c,
  * when adding a new allow rule to a device whitelist, the rule
  * must be allowed in the parent device
  */
-static int parent_has_perm(struct cgroup *childcg,
+static int parent_has_perm(struct dev_cgroup *childcg,
 				  struct dev_whitelist_item *wh)
 {
-	struct cgroup *pcg = childcg->parent;
+	struct cgroup *pcg = childcg->css.cgroup->parent;
 	struct dev_cgroup *parent;
 	int ret;
 
@@ -341,39 +346,18 @@ static int parent_has_perm(struct cgroup *childcg,
  * new access is only allowed if you're in the top-level cgroup, or your
  * parent cgroup has the access you're asking for.
  */
-static ssize_t devcgroup_access_write(struct cgroup *cgroup, struct cftype *cft,
-				struct file *file, const char __user *userbuf,
-				size_t nbytes, loff_t *ppos)
+static int devcgroup_update_access(struct dev_cgroup *devcgroup,
+				   int filetype, const char *buffer)
 {
-	struct cgroup *cur_cgroup;
-	struct dev_cgroup *devcgroup, *cur_devcgroup;
-	int filetype = cft->private;
-	char *buffer, *b;
+	struct dev_cgroup *cur_devcgroup;
+	const char *b;
 	int retval = 0, count;
 	struct dev_whitelist_item wh;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
-	devcgroup = cgroup_to_devcgroup(cgroup);
-	cur_cgroup = task_cgroup(current, devices_subsys.subsys_id);
-	cur_devcgroup = cgroup_to_devcgroup(cur_cgroup);
-
-	buffer = kmalloc(nbytes+1, GFP_KERNEL);
-	if (!buffer)
-		return -ENOMEM;
-
-	if (copy_from_user(buffer, userbuf, nbytes)) {
-		retval = -EFAULT;
-		goto out1;
-	}
-	buffer[nbytes] = 0;	/* nul-terminate */
-
-	cgroup_lock();
-	if (cgroup_is_removed(cgroup)) {
-		retval = -ENODEV;
-		goto out2;
-	}
+	cur_devcgroup = task_devcgroup(current);
 
 	memset(&wh, 0, sizeof(wh));
 	b = buffer;
@@ -392,14 +376,11 @@ static ssize_t devcgroup_access_write(struct cgroup *cgroup, struct cftype *cft,
 		wh.type = DEV_CHAR;
 		break;
 	default:
-		retval = -EINVAL;
-		goto out2;
+		return -EINVAL;
 	}
 	b++;
-	if (!isspace(*b)) {
-		retval = -EINVAL;
-		goto out2;
-	}
+	if (!isspace(*b))
+		return -EINVAL;
 	b++;
 	if (*b == '*') {
 		wh.major = ~0;
@@ -411,13 +392,10 @@ static ssize_t devcgroup_access_write(struct cgroup *cgroup, struct cftype *cft,
 			b++;
 		}
 	} else {
-		retval = -EINVAL;
-		goto out2;
-	}
-	if (*b != ':') {
-		retval = -EINVAL;
-		goto out2;
+		return -EINVAL;
 	}
+	if (*b != ':')
+		return -EINVAL;
 	b++;
 
 	/* read minor */
@@ -431,13 +409,10 @@ static ssize_t devcgroup_access_write(struct cgroup *cgroup, struct cftype *cft,
 			b++;
 		}
 	} else {
-		retval = -EINVAL;
-		goto out2;
-	}
-	if (!isspace(*b)) {
-		retval = -EINVAL;
-		goto out2;
+		return -EINVAL;
 	}
+	if (!isspace(*b))
+		return -EINVAL;
 	for (b++, count = 0; count < 3; count++, b++) {
 		switch (*b) {
 		case 'r':
@@ -454,8 +429,7 @@ static ssize_t devcgroup_access_write(struct cgroup *cgroup, struct cftype *cft,
 			count = 3;
 			break;
 		default:
-			retval = -EINVAL;
-			goto out2;
+			return -EINVAL;
 		}
 	}
 
@@ -463,38 +437,39 @@ handle:
 	retval = 0;
 	switch (filetype) {
 	case DEVCG_ALLOW:
-		if (!parent_has_perm(cgroup, &wh))
-			retval = -EPERM;
-		else
-			retval = dev_whitelist_add(devcgroup, &wh);
-		break;
+		if (!parent_has_perm(devcgroup, &wh))
+			return -EPERM;
+		return dev_whitelist_add(devcgroup, &wh);
 	case DEVCG_DENY:
 		dev_whitelist_rm(devcgroup, &wh);
 		break;
 	default:
-		retval = -EINVAL;
-		goto out2;
+		return -EINVAL;
 	}
+	return 0;
+}
 
-	if (retval == 0)
-		retval = nbytes;
-
-out2:
+static int devcgroup_access_write(struct cgroup *cgrp, struct cftype *cft,
+				  const char *buffer)
+{
+	int retval;
+	if (!cgroup_lock_live_group(cgrp))
+		return -ENODEV;
+	retval = devcgroup_update_access(cgroup_to_devcgroup(cgrp),
+					 cft->private, buffer);
 	cgroup_unlock();
-out1:
-	kfree(buffer);
 	return retval;
 }
 
 static struct cftype dev_cgroup_files[] = {
 	{
 		.name = "allow",
-		.write  = devcgroup_access_write,
+		.write_string  = devcgroup_access_write,
 		.private = DEVCG_ALLOW,
 	},
 	{
 		.name = "deny",
-		.write = devcgroup_access_write,
+		.write_string = devcgroup_access_write,
 		.private = DEVCG_DENY,
 	},
 	{

From 856c13aa1ff6136c1968414fdea5938ea9d5ebf2 Mon Sep 17 00:00:00 2001
From: Paul Menage <menage@google.com>
Date: Fri, 25 Jul 2008 01:47:04 -0700
Subject: [PATCH 0791/1171] cgroup files: convert res_counter_write() to be a
 cgroups write_string() handler

Currently res_counter_write() is a raw file handler even though it's
ultimately taking a number, since in some cases it wants to
pre-process the string when converting it to a number.

This patch converts res_counter_write() from a raw file handler to a
write_string() handler; this allows some of the boilerplate
copying/locking/checking to be removed, and simplies the cleanup path,
since these functions are now performed by the cgroups framework.

[lizf@cn.fujitsu.com: build fix]
Signed-off-by: Paul Menage <menage@google.com>
Cc: Paul Jackson <pj@sgi.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Serge Hallyn <serue@us.ibm.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/res_counter.h | 11 ++++++---
 kernel/res_counter.c        | 48 ++++++++++++++++---------------------
 mm/memcontrol.c             | 24 ++++---------------
 3 files changed, 34 insertions(+), 49 deletions(-)

diff --git a/include/linux/res_counter.h b/include/linux/res_counter.h
index 125660e7793f..290205dfe094 100644
--- a/include/linux/res_counter.h
+++ b/include/linux/res_counter.h
@@ -63,9 +63,14 @@ u64 res_counter_read_u64(struct res_counter *counter, int member);
 ssize_t res_counter_read(struct res_counter *counter, int member,
 		const char __user *buf, size_t nbytes, loff_t *pos,
 		int (*read_strategy)(unsigned long long val, char *s));
-ssize_t res_counter_write(struct res_counter *counter, int member,
-		const char __user *buf, size_t nbytes, loff_t *pos,
-		int (*write_strategy)(char *buf, unsigned long long *val));
+
+typedef int (*write_strategy_fn)(const char *buf, unsigned long long *val);
+
+int res_counter_memparse_write_strategy(const char *buf,
+					unsigned long long *res);
+
+int res_counter_write(struct res_counter *counter, int member,
+		      const char *buffer, write_strategy_fn write_strategy);
 
 /*
  * the field descriptors. one for each member of res_counter
diff --git a/kernel/res_counter.c b/kernel/res_counter.c
index d3c61b4ebef2..f275c8eca772 100644
--- a/kernel/res_counter.c
+++ b/kernel/res_counter.c
@@ -13,6 +13,7 @@
 #include <linux/slab.h>
 #include <linux/res_counter.h>
 #include <linux/uaccess.h>
+#include <linux/mm.h>
 
 void res_counter_init(struct res_counter *counter)
 {
@@ -102,44 +103,37 @@ u64 res_counter_read_u64(struct res_counter *counter, int member)
 	return *res_counter_member(counter, member);
 }
 
-ssize_t res_counter_write(struct res_counter *counter, int member,
-		const char __user *userbuf, size_t nbytes, loff_t *pos,
-		int (*write_strategy)(char *st_buf, unsigned long long *val))
+int res_counter_memparse_write_strategy(const char *buf,
+					unsigned long long *res)
 {
-	int ret;
-	char *buf, *end;
+	char *end;
+	/* FIXME - make memparse() take const char* args */
+	*res = memparse((char *)buf, &end);
+	if (*end != '\0')
+		return -EINVAL;
+
+	*res = PAGE_ALIGN(*res);
+	return 0;
+}
+
+int res_counter_write(struct res_counter *counter, int member,
+		      const char *buf, write_strategy_fn write_strategy)
+{
+	char *end;
 	unsigned long flags;
 	unsigned long long tmp, *val;
 
-	buf = kmalloc(nbytes + 1, GFP_KERNEL);
-	ret = -ENOMEM;
-	if (buf == NULL)
-		goto out;
-
-	buf[nbytes] = '\0';
-	ret = -EFAULT;
-	if (copy_from_user(buf, userbuf, nbytes))
-		goto out_free;
-
-	ret = -EINVAL;
-
-	strstrip(buf);
 	if (write_strategy) {
-		if (write_strategy(buf, &tmp)) {
-			goto out_free;
-		}
+		if (write_strategy(buf, &tmp))
+			return -EINVAL;
 	} else {
 		tmp = simple_strtoull(buf, &end, 10);
 		if (*end != '\0')
-			goto out_free;
+			return -EINVAL;
 	}
 	spin_lock_irqsave(&counter->lock, flags);
 	val = res_counter_member(counter, member);
 	*val = tmp;
 	spin_unlock_irqrestore(&counter->lock, flags);
-	ret = nbytes;
-out_free:
-	kfree(buf);
-out:
-	return ret;
+	return 0;
 }
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index e46451e1d9b7..7385d58fb061 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -838,32 +838,18 @@ out:
 	return ret;
 }
 
-static int mem_cgroup_write_strategy(char *buf, unsigned long long *tmp)
-{
-	*tmp = memparse(buf, &buf);
-	if (*buf != '\0')
-		return -EINVAL;
-
-	/*
-	 * Round up the value to the closest page size
-	 */
-	*tmp = ((*tmp + PAGE_SIZE - 1) >> PAGE_SHIFT) << PAGE_SHIFT;
-	return 0;
-}
-
 static u64 mem_cgroup_read(struct cgroup *cont, struct cftype *cft)
 {
 	return res_counter_read_u64(&mem_cgroup_from_cont(cont)->res,
 				    cft->private);
 }
 
-static ssize_t mem_cgroup_write(struct cgroup *cont, struct cftype *cft,
-				struct file *file, const char __user *userbuf,
-				size_t nbytes, loff_t *ppos)
+static int mem_cgroup_write(struct cgroup *cont, struct cftype *cft,
+			    const char *buffer)
 {
 	return res_counter_write(&mem_cgroup_from_cont(cont)->res,
-				cft->private, userbuf, nbytes, ppos,
-				mem_cgroup_write_strategy);
+				 cft->private, buffer,
+				 res_counter_memparse_write_strategy);
 }
 
 static int mem_cgroup_reset(struct cgroup *cont, unsigned int event)
@@ -940,7 +926,7 @@ static struct cftype mem_cgroup_files[] = {
 	{
 		.name = "limit_in_bytes",
 		.private = RES_LIMIT,
-		.write = mem_cgroup_write,
+		.write_string = mem_cgroup_write,
 		.read_u64 = mem_cgroup_read,
 	},
 	{

From e885dcde75685e09f23cffae1f6d5169c105b8a0 Mon Sep 17 00:00:00 2001
From: "Serge E. Hallyn" <serue@us.ibm.com>
Date: Fri, 25 Jul 2008 01:47:06 -0700
Subject: [PATCH 0792/1171] cgroup_clone: use pid of newly created task for new
 cgroup

cgroup_clone creates a new cgroup with the pid of the task.  This works
correctly for unshare, but for clone cgroup_clone is called from
copy_namespaces inside copy_process, which happens before the new pid is
created.  As a result, the new cgroup was created with current's pid.
This patch:

	1. Moves the call inside copy_process to after the new pid
	   is created
	2. Passes the struct pid into ns_cgroup_clone (as it is not
	   yet attached to the task)
	3. Passes a name from ns_cgroup_clone() into cgroup_clone()
	   so as to keep cgroup_clone() itself simpler
	4. Uses pid_vnr() to get the process id value, so that the
	   pid used to name the new cgroup is always the pid as it
	   would be known to the task which did the cloning or
	   unsharing.  I think that is the most intuitive thing to
	   do.  This way, task t1 does clone(CLONE_NEWPID) to get
	   t2, which does clone(CLONE_NEWPID) to get t3, then the
	   cgroup for t3 will be named for the pid by which t2 knows
	   t3.

(Thanks to Dan Smith for finding the main bug)

Changelog:
	June 11: Incorporate Paul Menage's feedback:  don't pass
	         NULL to ns_cgroup_clone from unshare, and reduce
		 patch size by using 'nodename' in cgroup_clone.
	June 10: Original version

[akpm@linux-foundation.org: build fix]
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Serge Hallyn <serge@us.ibm.com>
Acked-by: Paul Menage <menage@google.com>
Tested-by: Dan Smith <danms@us.ibm.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/cgroup.h  | 3 ++-
 include/linux/nsproxy.h | 7 +++++--
 kernel/cgroup.c         | 7 +++----
 kernel/fork.c           | 6 ++++++
 kernel/ns_cgroup.c      | 8 ++++++--
 kernel/nsproxy.c        | 8 +-------
 6 files changed, 23 insertions(+), 16 deletions(-)

diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h
index cc59d3a21d87..c98dd7cb7076 100644
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -364,7 +364,8 @@ static inline struct cgroup* task_cgroup(struct task_struct *task,
 	return task_subsys_state(task, subsys_id)->cgroup;
 }
 
-int cgroup_clone(struct task_struct *tsk, struct cgroup_subsys *ss);
+int cgroup_clone(struct task_struct *tsk, struct cgroup_subsys *ss,
+							char *nodename);
 
 /* A cgroup_iter should be treated as an opaque object */
 struct cgroup_iter {
diff --git a/include/linux/nsproxy.h b/include/linux/nsproxy.h
index 0e66b57631fc..c8a768e59640 100644
--- a/include/linux/nsproxy.h
+++ b/include/linux/nsproxy.h
@@ -82,9 +82,12 @@ static inline void get_nsproxy(struct nsproxy *ns)
 }
 
 #ifdef CONFIG_CGROUP_NS
-int ns_cgroup_clone(struct task_struct *tsk);
+int ns_cgroup_clone(struct task_struct *tsk, struct pid *pid);
 #else
-static inline int ns_cgroup_clone(struct task_struct *tsk) { return 0; }
+static inline int ns_cgroup_clone(struct task_struct *tsk, struct pid *pid)
+{
+	return 0;
+}
 #endif
 
 #endif
diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index 86b71e714e13..66ec9fd21e0c 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -2848,16 +2848,17 @@ void cgroup_exit(struct task_struct *tsk, int run_callbacks)
  * cgroup_clone - clone the cgroup the given subsystem is attached to
  * @tsk: the task to be moved
  * @subsys: the given subsystem
+ * @nodename: the name for the new cgroup
  *
  * Duplicate the current cgroup in the hierarchy that the given
  * subsystem is attached to, and move this task into the new
  * child.
  */
-int cgroup_clone(struct task_struct *tsk, struct cgroup_subsys *subsys)
+int cgroup_clone(struct task_struct *tsk, struct cgroup_subsys *subsys,
+							char *nodename)
 {
 	struct dentry *dentry;
 	int ret = 0;
-	char nodename[MAX_CGROUP_TYPE_NAMELEN];
 	struct cgroup *parent, *child;
 	struct inode *inode;
 	struct css_set *cg;
@@ -2882,8 +2883,6 @@ int cgroup_clone(struct task_struct *tsk, struct cgroup_subsys *subsys)
 	cg = tsk->cgroups;
 	parent = task_cgroup(tsk, subsys->subsys_id);
 
-	snprintf(nodename, MAX_CGROUP_TYPE_NAMELEN, "%d", tsk->pid);
-
 	/* Pin the hierarchy */
 	atomic_inc(&parent->root->sb->s_active);
 
diff --git a/kernel/fork.c b/kernel/fork.c
index 5a5d6fef341d..228f80c9155a 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1107,6 +1107,12 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 	if (clone_flags & CLONE_THREAD)
 		p->tgid = current->tgid;
 
+	if (current->nsproxy != p->nsproxy) {
+		retval = ns_cgroup_clone(p, pid);
+		if (retval)
+			goto bad_fork_free_pid;
+	}
+
 	p->set_child_tid = (clone_flags & CLONE_CHILD_SETTID) ? child_tidptr : NULL;
 	/*
 	 * Clear TID on mm_release()?
diff --git a/kernel/ns_cgroup.c b/kernel/ns_cgroup.c
index 48d7ed6fc3a4..43c2111cd54d 100644
--- a/kernel/ns_cgroup.c
+++ b/kernel/ns_cgroup.c
@@ -7,6 +7,7 @@
 #include <linux/module.h>
 #include <linux/cgroup.h>
 #include <linux/fs.h>
+#include <linux/proc_fs.h>
 #include <linux/slab.h>
 #include <linux/nsproxy.h>
 
@@ -24,9 +25,12 @@ static inline struct ns_cgroup *cgroup_to_ns(
 			    struct ns_cgroup, css);
 }
 
-int ns_cgroup_clone(struct task_struct *task)
+int ns_cgroup_clone(struct task_struct *task, struct pid *pid)
 {
-	return cgroup_clone(task, &ns_subsys);
+	char name[PROC_NUMBUF];
+
+	snprintf(name, PROC_NUMBUF, "%d", pid_vnr(pid));
+	return cgroup_clone(task, &ns_subsys, name);
 }
 
 /*
diff --git a/kernel/nsproxy.c b/kernel/nsproxy.c
index adc785146a1c..21575fc46d05 100644
--- a/kernel/nsproxy.c
+++ b/kernel/nsproxy.c
@@ -157,12 +157,6 @@ int copy_namespaces(unsigned long flags, struct task_struct *tsk)
 		goto out;
 	}
 
-	err = ns_cgroup_clone(tsk);
-	if (err) {
-		put_nsproxy(new_ns);
-		goto out;
-	}
-
 	tsk->nsproxy = new_ns;
 
 out:
@@ -209,7 +203,7 @@ int unshare_nsproxy_namespaces(unsigned long unshare_flags,
 		goto out;
 	}
 
-	err = ns_cgroup_clone(current);
+	err = ns_cgroup_clone(current, task_pid(current));
 	if (err)
 		put_nsproxy(*new_nsp);
 

From 4efd1a1b2f09a4b746dd9dc057986c6dadcb1317 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 25 Jul 2008 01:47:07 -0700
Subject: [PATCH 0793/1171] devcgroup: relax white-list protection down to RCU

Currently this list is protected with a simple spinlock, even for reading
from one.  This is OK, but can be better.

Actually I want it to be better very much, since after replacing the
OpenVZ device permissions engine with the cgroup-based one I noticed, that
we set 12 default device permissions for each newly created container (for
/dev/null, full, terminals, ect devices), and people sometimes have up to
20 perms more, so traversing the ~30-40 elements list under a spinlock
doesn't seem very good.

Here's the RCU protection for white-list - dev_whitelist_item-s are added
and removed under the devcg->lock, but are looked up in permissions
checking under the rcu_read_lock.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Paul Menage <menage@google.com>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 security/device_cgroup.c | 35 ++++++++++++++++++++++-------------
 1 file changed, 22 insertions(+), 13 deletions(-)

diff --git a/security/device_cgroup.c b/security/device_cgroup.c
index 236fffa9d05e..9da3532726ff 100644
--- a/security/device_cgroup.c
+++ b/security/device_cgroup.c
@@ -41,6 +41,7 @@ struct dev_whitelist_item {
 	short type;
 	short access;
 	struct list_head list;
+	struct rcu_head rcu;
 };
 
 struct dev_cgroup {
@@ -133,11 +134,19 @@ static int dev_whitelist_add(struct dev_cgroup *dev_cgroup,
 	}
 
 	if (whcopy != NULL)
-		list_add_tail(&whcopy->list, &dev_cgroup->whitelist);
+		list_add_tail_rcu(&whcopy->list, &dev_cgroup->whitelist);
 	spin_unlock(&dev_cgroup->lock);
 	return 0;
 }
 
+static void whitelist_item_free(struct rcu_head *rcu)
+{
+	struct dev_whitelist_item *item;
+
+	item = container_of(rcu, struct dev_whitelist_item, rcu);
+	kfree(item);
+}
+
 /*
  * called under cgroup_lock()
  * since the list is visible to other tasks, we need the spinlock also
@@ -161,8 +170,8 @@ static void dev_whitelist_rm(struct dev_cgroup *dev_cgroup,
 remove:
 		walk->access &= ~wh->access;
 		if (!walk->access) {
-			list_del(&walk->list);
-			kfree(walk);
+			list_del_rcu(&walk->list);
+			call_rcu(&walk->rcu, whitelist_item_free);
 		}
 	}
 	spin_unlock(&dev_cgroup->lock);
@@ -269,15 +278,15 @@ static int devcgroup_seq_read(struct cgroup *cgroup, struct cftype *cft,
 	struct dev_whitelist_item *wh;
 	char maj[MAJMINLEN], min[MAJMINLEN], acc[ACCLEN];
 
-	spin_lock(&devcgroup->lock);
-	list_for_each_entry(wh, &devcgroup->whitelist, list) {
+	rcu_read_lock();
+	list_for_each_entry_rcu(wh, &devcgroup->whitelist, list) {
 		set_access(acc, wh->access);
 		set_majmin(maj, wh->major);
 		set_majmin(min, wh->minor);
 		seq_printf(m, "%c %s:%s %s\n", type_to_char(wh->type),
 			   maj, min, acc);
 	}
-	spin_unlock(&devcgroup->lock);
+	rcu_read_unlock();
 
 	return 0;
 }
@@ -510,8 +519,8 @@ int devcgroup_inode_permission(struct inode *inode, int mask)
 	if (!dev_cgroup)
 		return 0;
 
-	spin_lock(&dev_cgroup->lock);
-	list_for_each_entry(wh, &dev_cgroup->whitelist, list) {
+	rcu_read_lock();
+	list_for_each_entry_rcu(wh, &dev_cgroup->whitelist, list) {
 		if (wh->type & DEV_ALL)
 			goto acc_check;
 		if ((wh->type & DEV_BLOCK) && !S_ISBLK(inode->i_mode))
@@ -527,10 +536,10 @@ acc_check:
 			continue;
 		if ((mask & MAY_READ) && !(wh->access & ACC_READ))
 			continue;
-		spin_unlock(&dev_cgroup->lock);
+		rcu_read_unlock();
 		return 0;
 	}
-	spin_unlock(&dev_cgroup->lock);
+	rcu_read_unlock();
 
 	return -EPERM;
 }
@@ -545,7 +554,7 @@ int devcgroup_inode_mknod(int mode, dev_t dev)
 	if (!dev_cgroup)
 		return 0;
 
-	spin_lock(&dev_cgroup->lock);
+	rcu_read_lock();
 	list_for_each_entry(wh, &dev_cgroup->whitelist, list) {
 		if (wh->type & DEV_ALL)
 			goto acc_check;
@@ -560,9 +569,9 @@ int devcgroup_inode_mknod(int mode, dev_t dev)
 acc_check:
 		if (!(wh->access & ACC_MKNOD))
 			continue;
-		spin_unlock(&dev_cgroup->lock);
+		rcu_read_unlock();
 		return 0;
 	}
-	spin_unlock(&dev_cgroup->lock);
+	rcu_read_unlock();
 	return -EPERM;
 }

From 7759fc9d10d3559f365cb122d81e0c0a185fe0fe Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Fri, 25 Jul 2008 01:47:08 -0700
Subject: [PATCH 0794/1171] devcgroup: code cleanup

- clean up set_majmin()
- use simple_strtoul() to parse major/minor

[akpm@linux-foundation.org: fix simple_strtoul() usage]
[kosaki.motohiro@jp.fujitsu.com: fix warnings]
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Acked-by: Serge Hallyn <serue@us.ibm.com>
Cc: Serge Hallyn <serue@us.ibm.com>
Cc: Paul Menage <menage@google.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 security/device_cgroup.c | 22 ++++++++--------------
 1 file changed, 8 insertions(+), 14 deletions(-)

diff --git a/security/device_cgroup.c b/security/device_cgroup.c
index 9da3532726ff..7bd296cca041 100644
--- a/security/device_cgroup.c
+++ b/security/device_cgroup.c
@@ -202,7 +202,7 @@ static struct cgroup_subsys_state *devcgroup_create(struct cgroup_subsys *ss,
 		}
 		wh->minor = wh->major = ~0;
 		wh->type = DEV_ALL;
-		wh->access = ACC_MKNOD | ACC_READ | ACC_WRITE;
+		wh->access = ACC_MASK;
 		list_add(&wh->list, &dev_cgroup->whitelist);
 	} else {
 		parent_dev_cgroup = cgroup_to_devcgroup(parent_cgroup);
@@ -264,11 +264,10 @@ static char type_to_char(short type)
 
 static void set_majmin(char *str, unsigned m)
 {
-	memset(str, 0, MAJMINLEN);
 	if (m == ~0)
-		sprintf(str, "*");
+		strcpy(str, "*");
 	else
-		snprintf(str, MAJMINLEN, "%u", m);
+		sprintf(str, "%u", m);
 }
 
 static int devcgroup_seq_read(struct cgroup *cgroup, struct cftype *cft,
@@ -360,6 +359,7 @@ static int devcgroup_update_access(struct dev_cgroup *devcgroup,
 {
 	struct dev_cgroup *cur_devcgroup;
 	const char *b;
+	char *endp;
 	int retval = 0, count;
 	struct dev_whitelist_item wh;
 
@@ -395,11 +395,8 @@ static int devcgroup_update_access(struct dev_cgroup *devcgroup,
 		wh.major = ~0;
 		b++;
 	} else if (isdigit(*b)) {
-		wh.major = 0;
-		while (isdigit(*b)) {
-			wh.major = wh.major*10+(*b-'0');
-			b++;
-		}
+		wh.major = simple_strtoul(b, &endp, 10);
+		b = endp;
 	} else {
 		return -EINVAL;
 	}
@@ -412,11 +409,8 @@ static int devcgroup_update_access(struct dev_cgroup *devcgroup,
 		wh.minor = ~0;
 		b++;
 	} else if (isdigit(*b)) {
-		wh.minor = 0;
-		while (isdigit(*b)) {
-			wh.minor = wh.minor*10+(*b-'0');
-			b++;
-		}
+		wh.minor = simple_strtoul(b, &endp, 10);
+		b = endp;
 	} else {
 		return -EINVAL;
 	}

From a181b0e888a1d917edcab57cd73ccf7d8e75a46c Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Fri, 25 Jul 2008 01:47:08 -0700
Subject: [PATCH 0795/1171] memcg: make global var read_mostly

mem_cgroup_subsys and page_cgroup_cache should be read_mostly and
MEM_CGROUP_RECLAIM_RETRIES can be just a fixed number.

Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Acked-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/memcontrol.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 7385d58fb061..c52c045f5152 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -35,9 +35,9 @@
 
 #include <asm/uaccess.h>
 
-struct cgroup_subsys mem_cgroup_subsys;
-static const int MEM_CGROUP_RECLAIM_RETRIES = 5;
-static struct kmem_cache *page_cgroup_cache;
+struct cgroup_subsys mem_cgroup_subsys __read_mostly;
+static struct kmem_cache *page_cgroup_cache __read_mostly;
+#define MEM_CGROUP_RECLAIM_RETRIES	5
 
 /*
  * Statistics for memory cgroup.

From 508b7be0a5b06b64203512ed9b34191cddc83f56 Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Fri, 25 Jul 2008 01:47:09 -0700
Subject: [PATCH 0796/1171] memcg: avoid unnecessary initialization

* remove over-killing initialization (in fast path)
* makeing the condition for PAGE_CGROUP_FLAG_ACTIVE be more obvious.

Signed-off-by: KAMEAZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Reviewed-by: Li Zefan <lizf@cn.fujitsu.com>
Acked-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Acked-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/memcontrol.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index c52c045f5152..90ccc1326356 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -296,7 +296,7 @@ static void __mem_cgroup_remove_list(struct mem_cgroup_per_zone *mz,
 		MEM_CGROUP_ZSTAT(mz, MEM_CGROUP_ZSTAT_INACTIVE) -= 1;
 
 	mem_cgroup_charge_statistics(pc->mem_cgroup, pc->flags, false);
-	list_del_init(&pc->lru);
+	list_del(&pc->lru);
 }
 
 static void __mem_cgroup_add_list(struct mem_cgroup_per_zone *mz,
@@ -559,7 +559,7 @@ retry:
 	}
 	unlock_page_cgroup(page);
 
-	pc = kmem_cache_zalloc(page_cgroup_cache, gfp_mask);
+	pc = kmem_cache_alloc(page_cgroup_cache, gfp_mask);
 	if (pc == NULL)
 		goto err;
 
@@ -606,9 +606,14 @@ retry:
 	pc->ref_cnt = 1;
 	pc->mem_cgroup = mem;
 	pc->page = page;
-	pc->flags = PAGE_CGROUP_FLAG_ACTIVE;
+	/*
+	 * If a page is accounted as a page cache, insert to inactive list.
+	 * If anon, insert to active list.
+	 */
 	if (ctype == MEM_CGROUP_CHARGE_TYPE_CACHE)
 		pc->flags = PAGE_CGROUP_FLAG_CACHE;
+	else
+		pc->flags = PAGE_CGROUP_FLAG_ACTIVE;
 
 	lock_page_cgroup(page);
 	if (page_get_page_cgroup(page)) {

From e8589cc189f96b87348ae83ea4db38eaac624135 Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Fri, 25 Jul 2008 01:47:10 -0700
Subject: [PATCH 0797/1171] memcg: better migration handling

This patch changes page migration under memory controller to use a
different algorithm.  (thanks to Christoph for new idea.)

Before:
 - page_cgroup is migrated from an old page to a new page.
After:
 - a new page is accounted , no reuse of page_cgroup.

Pros:

 - We can avoid compliated lock depndencies and races in migration.

Cons:

 - new param to mem_cgroup_charge_common().

 - mem_cgroup_getref() is added for handling ref_cnt ping-pong.

This version simplifies complicated lock dependency in page migraiton
under memory resource controller.

  new refcnt sequence is following.

a mapped page:
  prepage_migration() ..... +1 to NEW page
  try_to_unmap()      ..... all refs to OLD page is gone.
  move_pages()        ..... +1 to NEW page if page cache.
  remap...            ..... all refs from *map* is added to NEW one.
  end_migration()     ..... -1 to New page.

  page's mapcount + (page_is_cache) refs are added to NEW one.

Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: YAMAMOTO Takashi <yamamoto@valinux.co.jp>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Christoph Lameter <cl@linux-foundation.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/memcontrol.h |  11 ++--
 mm/memcontrol.c            | 130 +++++++++++++++++++------------------
 mm/migrate.c               |  22 +++++--
 3 files changed, 87 insertions(+), 76 deletions(-)

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index e6608776bc96..84ead2aa6f18 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -50,9 +50,10 @@ extern struct mem_cgroup *mem_cgroup_from_task(struct task_struct *p);
 #define mm_match_cgroup(mm, cgroup)	\
 	((cgroup) == mem_cgroup_from_task((mm)->owner))
 
-extern int mem_cgroup_prepare_migration(struct page *page);
+extern int
+mem_cgroup_prepare_migration(struct page *page, struct page *newpage);
 extern void mem_cgroup_end_migration(struct page *page);
-extern void mem_cgroup_page_migration(struct page *page, struct page *newpage);
+extern int mem_cgroup_getref(struct page *page);
 
 /*
  * For memory reclaim.
@@ -112,7 +113,8 @@ static inline int task_in_mem_cgroup(struct task_struct *task,
 	return 1;
 }
 
-static inline int mem_cgroup_prepare_migration(struct page *page)
+static inline int
+mem_cgroup_prepare_migration(struct page *page, struct page *newpage)
 {
 	return 0;
 }
@@ -121,8 +123,7 @@ static inline void mem_cgroup_end_migration(struct page *page)
 {
 }
 
-static inline void
-mem_cgroup_page_migration(struct page *page, struct page *newpage)
+static inline void mem_cgroup_getref(struct page *page)
 {
 }
 
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 90ccc1326356..da5912b84551 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -524,7 +524,8 @@ unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan,
  * < 0 if the cgroup is over its limit
  */
 static int mem_cgroup_charge_common(struct page *page, struct mm_struct *mm,
-				gfp_t gfp_mask, enum charge_type ctype)
+				gfp_t gfp_mask, enum charge_type ctype,
+				struct mem_cgroup *memcg)
 {
 	struct mem_cgroup *mem;
 	struct page_cgroup *pc;
@@ -569,16 +570,21 @@ retry:
 	 * thread group leader migrates. It's possible that mm is not
 	 * set, if so charge the init_mm (happens for pagecache usage).
 	 */
-	if (!mm)
-		mm = &init_mm;
+	if (!memcg) {
+		if (!mm)
+			mm = &init_mm;
 
-	rcu_read_lock();
-	mem = mem_cgroup_from_task(rcu_dereference(mm->owner));
-	/*
-	 * For every charge from the cgroup, increment reference count
-	 */
-	css_get(&mem->css);
-	rcu_read_unlock();
+		rcu_read_lock();
+		mem = mem_cgroup_from_task(rcu_dereference(mm->owner));
+		/*
+		 * For every charge from the cgroup, increment reference count
+		 */
+		css_get(&mem->css);
+		rcu_read_unlock();
+	} else {
+		mem = memcg;
+		css_get(&memcg->css);
+	}
 
 	while (res_counter_charge(&mem->res, PAGE_SIZE)) {
 		if (!(gfp_mask & __GFP_WAIT))
@@ -648,7 +654,7 @@ err:
 int mem_cgroup_charge(struct page *page, struct mm_struct *mm, gfp_t gfp_mask)
 {
 	return mem_cgroup_charge_common(page, mm, gfp_mask,
-				MEM_CGROUP_CHARGE_TYPE_MAPPED);
+				MEM_CGROUP_CHARGE_TYPE_MAPPED, NULL);
 }
 
 int mem_cgroup_cache_charge(struct page *page, struct mm_struct *mm,
@@ -657,7 +663,22 @@ int mem_cgroup_cache_charge(struct page *page, struct mm_struct *mm,
 	if (!mm)
 		mm = &init_mm;
 	return mem_cgroup_charge_common(page, mm, gfp_mask,
-				MEM_CGROUP_CHARGE_TYPE_CACHE);
+				MEM_CGROUP_CHARGE_TYPE_CACHE, NULL);
+}
+
+int mem_cgroup_getref(struct page *page)
+{
+	struct page_cgroup *pc;
+
+	if (mem_cgroup_subsys.disabled)
+		return 0;
+
+	lock_page_cgroup(page);
+	pc = page_get_page_cgroup(page);
+	VM_BUG_ON(!pc);
+	pc->ref_cnt++;
+	unlock_page_cgroup(page);
+	return 0;
 }
 
 /*
@@ -707,65 +728,39 @@ unlock:
 }
 
 /*
- * Returns non-zero if a page (under migration) has valid page_cgroup member.
- * Refcnt of page_cgroup is incremented.
+ * Before starting migration, account against new page.
  */
-int mem_cgroup_prepare_migration(struct page *page)
+int mem_cgroup_prepare_migration(struct page *page, struct page *newpage)
 {
 	struct page_cgroup *pc;
+	struct mem_cgroup *mem = NULL;
+	enum charge_type ctype = MEM_CGROUP_CHARGE_TYPE_MAPPED;
+	int ret = 0;
 
 	if (mem_cgroup_subsys.disabled)
 		return 0;
 
 	lock_page_cgroup(page);
 	pc = page_get_page_cgroup(page);
-	if (pc)
-		pc->ref_cnt++;
-	unlock_page_cgroup(page);
-	return pc != NULL;
-}
-
-void mem_cgroup_end_migration(struct page *page)
-{
-	mem_cgroup_uncharge_page(page);
-}
-
-/*
- * We know both *page* and *newpage* are now not-on-LRU and PG_locked.
- * And no race with uncharge() routines because page_cgroup for *page*
- * has extra one reference by mem_cgroup_prepare_migration.
- */
-void mem_cgroup_page_migration(struct page *page, struct page *newpage)
-{
-	struct page_cgroup *pc;
-	struct mem_cgroup_per_zone *mz;
-	unsigned long flags;
-
-	lock_page_cgroup(page);
-	pc = page_get_page_cgroup(page);
-	if (!pc) {
-		unlock_page_cgroup(page);
-		return;
+	if (pc) {
+		mem = pc->mem_cgroup;
+		css_get(&mem->css);
+		if (pc->flags & PAGE_CGROUP_FLAG_CACHE)
+			ctype = MEM_CGROUP_CHARGE_TYPE_CACHE;
 	}
-
-	mz = page_cgroup_zoneinfo(pc);
-	spin_lock_irqsave(&mz->lru_lock, flags);
-	__mem_cgroup_remove_list(mz, pc);
-	spin_unlock_irqrestore(&mz->lru_lock, flags);
-
-	page_assign_page_cgroup(page, NULL);
 	unlock_page_cgroup(page);
+	if (mem) {
+		ret = mem_cgroup_charge_common(newpage, NULL, GFP_KERNEL,
+			ctype, mem);
+		css_put(&mem->css);
+	}
+	return ret;
+}
 
-	pc->page = newpage;
-	lock_page_cgroup(newpage);
-	page_assign_page_cgroup(newpage, pc);
-
-	mz = page_cgroup_zoneinfo(pc);
-	spin_lock_irqsave(&mz->lru_lock, flags);
-	__mem_cgroup_add_list(mz, pc);
-	spin_unlock_irqrestore(&mz->lru_lock, flags);
-
-	unlock_page_cgroup(newpage);
+/* remove redundant charge */
+void mem_cgroup_end_migration(struct page *newpage)
+{
+	mem_cgroup_uncharge_page(newpage);
 }
 
 /*
@@ -795,12 +790,19 @@ static void mem_cgroup_force_empty_list(struct mem_cgroup *mem,
 		page = pc->page;
 		get_page(page);
 		spin_unlock_irqrestore(&mz->lru_lock, flags);
-		mem_cgroup_uncharge_page(page);
-		put_page(page);
-		if (--count <= 0) {
-			count = FORCE_UNCHARGE_BATCH;
+		/*
+		 * Check if this page is on LRU. !LRU page can be found
+		 * if it's under page migration.
+		 */
+		if (PageLRU(page)) {
+			mem_cgroup_uncharge_page(page);
+			put_page(page);
+			if (--count <= 0) {
+				count = FORCE_UNCHARGE_BATCH;
+				cond_resched();
+			}
+		} else
 			cond_resched();
-		}
 		spin_lock_irqsave(&mz->lru_lock, flags);
 	}
 	spin_unlock_irqrestore(&mz->lru_lock, flags);
diff --git a/mm/migrate.c b/mm/migrate.c
index 376cceba82f9..f6d7f8efd1a8 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -358,6 +358,10 @@ static int migrate_page_move_mapping(struct address_space *mapping,
 	__inc_zone_page_state(newpage, NR_FILE_PAGES);
 
 	write_unlock_irq(&mapping->tree_lock);
+	if (!PageSwapCache(newpage)) {
+		mem_cgroup_uncharge_page(page);
+		mem_cgroup_getref(newpage);
+	}
 
 	return 0;
 }
@@ -611,7 +615,6 @@ static int move_to_new_page(struct page *newpage, struct page *page)
 		rc = fallback_migrate_page(mapping, newpage, page);
 
 	if (!rc) {
-		mem_cgroup_page_migration(page, newpage);
 		remove_migration_ptes(page, newpage);
 	} else
 		newpage->mapping = NULL;
@@ -641,6 +644,14 @@ static int unmap_and_move(new_page_t get_new_page, unsigned long private,
 		/* page was freed from under us. So we are done. */
 		goto move_newpage;
 
+	charge = mem_cgroup_prepare_migration(page, newpage);
+	if (charge == -ENOMEM) {
+		rc = -ENOMEM;
+		goto move_newpage;
+	}
+	/* prepare cgroup just returns 0 or -ENOMEM */
+	BUG_ON(charge);
+
 	rc = -EAGAIN;
 	if (TestSetPageLocked(page)) {
 		if (!force)
@@ -692,19 +703,14 @@ static int unmap_and_move(new_page_t get_new_page, unsigned long private,
 		goto rcu_unlock;
 	}
 
-	charge = mem_cgroup_prepare_migration(page);
 	/* Establish migration ptes or remove ptes */
 	try_to_unmap(page, 1);
 
 	if (!page_mapped(page))
 		rc = move_to_new_page(newpage, page);
 
-	if (rc) {
+	if (rc)
 		remove_migration_ptes(page, page);
-		if (charge)
-			mem_cgroup_end_migration(page);
-	} else if (charge)
- 		mem_cgroup_end_migration(newpage);
 rcu_unlock:
 	if (rcu_locked)
 		rcu_read_unlock();
@@ -725,6 +731,8 @@ unlock:
 	}
 
 move_newpage:
+	if (!charge)
+		mem_cgroup_end_migration(newpage);
 	/*
 	 * Move the new page to the LRU. If migration was not successful
 	 * then this will free the page.

From 69029cd550284e32de13d6dd2f77b723c8a0e444 Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Fri, 25 Jul 2008 01:47:14 -0700
Subject: [PATCH 0798/1171] memcg: remove refcnt from page_cgroup

memcg: performance improvements

Patch Description
 1/5 ... remove refcnt fron page_cgroup patch (shmem handling is fixed)
 2/5 ... swapcache handling patch
 3/5 ... add helper function for shmem's memory reclaim patch
 4/5 ... optimize by likely/unlikely ppatch
 5/5 ... remove redundunt check patch (shmem handling is fixed.)

Unix bench result.

== 2.6.26-rc2-mm1 + memory resource controller
Execl Throughput                           2915.4 lps   (29.6 secs, 3 samples)
C Compiler Throughput                      1019.3 lpm   (60.0 secs, 3 samples)
Shell Scripts (1 concurrent)               5796.0 lpm   (60.0 secs, 3 samples)
Shell Scripts (8 concurrent)               1097.7 lpm   (60.0 secs, 3 samples)
Shell Scripts (16 concurrent)               565.3 lpm   (60.0 secs, 3 samples)
File Read 1024 bufsize 2000 maxblocks    1022128.0 KBps  (30.0 secs, 3 samples)
File Write 1024 bufsize 2000 maxblocks   544057.0 KBps  (30.0 secs, 3 samples)
File Copy 1024 bufsize 2000 maxblocks    346481.0 KBps  (30.0 secs, 3 samples)
File Read 256 bufsize 500 maxblocks      319325.0 KBps  (30.0 secs, 3 samples)
File Write 256 bufsize 500 maxblocks     148788.0 KBps  (30.0 secs, 3 samples)
File Copy 256 bufsize 500 maxblocks       99051.0 KBps  (30.0 secs, 3 samples)
File Read 4096 bufsize 8000 maxblocks    2058917.0 KBps  (30.0 secs, 3 samples)
File Write 4096 bufsize 8000 maxblocks   1606109.0 KBps  (30.0 secs, 3 samples)
File Copy 4096 bufsize 8000 maxblocks    854789.0 KBps  (30.0 secs, 3 samples)
Dc: sqrt(2) to 99 decimal places         126145.2 lpm   (30.0 secs, 3 samples)

                     INDEX VALUES
TEST                                        BASELINE     RESULT      INDEX

Execl Throughput                                43.0     2915.4      678.0
File Copy 1024 bufsize 2000 maxblocks         3960.0   346481.0      875.0
File Copy 256 bufsize 500 maxblocks           1655.0    99051.0      598.5
File Copy 4096 bufsize 8000 maxblocks         5800.0   854789.0     1473.8
Shell Scripts (8 concurrent)                     6.0     1097.7     1829.5
                                                                 =========
     FINAL SCORE                                                     991.3

== 2.6.26-rc2-mm1 + this set ==
Execl Throughput                           3012.9 lps   (29.9 secs, 3 samples)
C Compiler Throughput                       981.0 lpm   (60.0 secs, 3 samples)
Shell Scripts (1 concurrent)               5872.0 lpm   (60.0 secs, 3 samples)
Shell Scripts (8 concurrent)               1120.3 lpm   (60.0 secs, 3 samples)
Shell Scripts (16 concurrent)               578.0 lpm   (60.0 secs, 3 samples)
File Read 1024 bufsize 2000 maxblocks    1003993.0 KBps  (30.0 secs, 3 samples)
File Write 1024 bufsize 2000 maxblocks   550452.0 KBps  (30.0 secs, 3 samples)
File Copy 1024 bufsize 2000 maxblocks    347159.0 KBps  (30.0 secs, 3 samples)
File Read 256 bufsize 500 maxblocks      314644.0 KBps  (30.0 secs, 3 samples)
File Write 256 bufsize 500 maxblocks     151852.0 KBps  (30.0 secs, 3 samples)
File Copy 256 bufsize 500 maxblocks      101000.0 KBps  (30.0 secs, 3 samples)
File Read 4096 bufsize 8000 maxblocks    2033256.0 KBps  (30.0 secs, 3 samples)
File Write 4096 bufsize 8000 maxblocks   1611814.0 KBps  (30.0 secs, 3 samples)
File Copy 4096 bufsize 8000 maxblocks    847979.0 KBps  (30.0 secs, 3 samples)
Dc: sqrt(2) to 99 decimal places         128148.7 lpm   (30.0 secs, 3 samples)

                     INDEX VALUES
TEST                                        BASELINE     RESULT      INDEX

Execl Throughput                                43.0     3012.9      700.7
File Copy 1024 bufsize 2000 maxblocks         3960.0   347159.0      876.7
File Copy 256 bufsize 500 maxblocks           1655.0   101000.0      610.3
File Copy 4096 bufsize 8000 maxblocks         5800.0   847979.0     1462.0
Shell Scripts (8 concurrent)                     6.0     1120.3     1867.2
                                                                 =========
     FINAL SCORE                                                    1004.6

This patch:

Remove refcnt from page_cgroup().

After this,

 * A page is charged only when !page_mapped() && no page_cgroup is assigned.
	* Anon page is newly mapped.
	* File page is added to mapping->tree.

 * A page is uncharged only when
	* Anon page is fully unmapped.
	* File page is removed from LRU.

There is no change in behavior from user's view.

This patch also removes unnecessary calls in rmap.c which was used only for
refcnt mangement.

[akpm@linux-foundation.org: fix warning]
[hugh@veritas.com: fix shmem_unuse_inode charging]
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: YAMAMOTO Takashi <yamamoto@valinux.co.jp>
Cc: Paul Menage <menage@google.com>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/memcontrol.h |  10 ++--
 mm/filemap.c               |   6 +-
 mm/memcontrol.c            | 109 +++++++++++++++++++++----------------
 mm/migrate.c               |   3 +-
 mm/rmap.c                  |  14 +----
 mm/shmem.c                 |  35 ++++++++----
 6 files changed, 97 insertions(+), 80 deletions(-)

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 84ead2aa6f18..b4980b8f048e 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -35,6 +35,7 @@ extern int mem_cgroup_charge(struct page *page, struct mm_struct *mm,
 extern int mem_cgroup_cache_charge(struct page *page, struct mm_struct *mm,
 					gfp_t gfp_mask);
 extern void mem_cgroup_uncharge_page(struct page *page);
+extern void mem_cgroup_uncharge_cache_page(struct page *page);
 extern void mem_cgroup_move_lists(struct page *page, bool active);
 extern unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan,
 					struct list_head *dst,
@@ -53,7 +54,6 @@ extern struct mem_cgroup *mem_cgroup_from_task(struct task_struct *p);
 extern int
 mem_cgroup_prepare_migration(struct page *page, struct page *newpage);
 extern void mem_cgroup_end_migration(struct page *page);
-extern int mem_cgroup_getref(struct page *page);
 
 /*
  * For memory reclaim.
@@ -98,6 +98,10 @@ static inline void mem_cgroup_uncharge_page(struct page *page)
 {
 }
 
+static inline void mem_cgroup_uncharge_cache_page(struct page *page)
+{
+}
+
 static inline void mem_cgroup_move_lists(struct page *page, bool active)
 {
 }
@@ -123,10 +127,6 @@ static inline void mem_cgroup_end_migration(struct page *page)
 {
 }
 
-static inline void mem_cgroup_getref(struct page *page)
-{
-}
-
 static inline int mem_cgroup_calc_mapped_ratio(struct mem_cgroup *mem)
 {
 	return 0;
diff --git a/mm/filemap.c b/mm/filemap.c
index 5d4c880d7cd9..2d3ec1ffc66e 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -115,7 +115,7 @@ void __remove_from_page_cache(struct page *page)
 {
 	struct address_space *mapping = page->mapping;
 
-	mem_cgroup_uncharge_page(page);
+	mem_cgroup_uncharge_cache_page(page);
 	radix_tree_delete(&mapping->page_tree, page->index);
 	page->mapping = NULL;
 	mapping->nrpages--;
@@ -474,12 +474,12 @@ int add_to_page_cache(struct page *page, struct address_space *mapping,
 			mapping->nrpages++;
 			__inc_zone_page_state(page, NR_FILE_PAGES);
 		} else
-			mem_cgroup_uncharge_page(page);
+			mem_cgroup_uncharge_cache_page(page);
 
 		write_unlock_irq(&mapping->tree_lock);
 		radix_tree_preload_end();
 	} else
-		mem_cgroup_uncharge_page(page);
+		mem_cgroup_uncharge_cache_page(page);
 out:
 	return error;
 }
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index da5912b84551..a61706193c31 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -166,7 +166,6 @@ struct page_cgroup {
 	struct list_head lru;		/* per cgroup LRU list */
 	struct page *page;
 	struct mem_cgroup *mem_cgroup;
-	int ref_cnt;			/* cached, mapped, migrating */
 	int flags;
 };
 #define PAGE_CGROUP_FLAG_CACHE	(0x1)	/* charged as cache */
@@ -185,6 +184,7 @@ static enum zone_type page_cgroup_zid(struct page_cgroup *pc)
 enum charge_type {
 	MEM_CGROUP_CHARGE_TYPE_CACHE = 0,
 	MEM_CGROUP_CHARGE_TYPE_MAPPED,
+	MEM_CGROUP_CHARGE_TYPE_FORCE,	/* used by force_empty */
 };
 
 /*
@@ -552,9 +552,7 @@ retry:
 	 */
 	if (pc) {
 		VM_BUG_ON(pc->page != page);
-		VM_BUG_ON(pc->ref_cnt <= 0);
-
-		pc->ref_cnt++;
+		VM_BUG_ON(!pc->mem_cgroup);
 		unlock_page_cgroup(page);
 		goto done;
 	}
@@ -570,10 +568,7 @@ retry:
 	 * thread group leader migrates. It's possible that mm is not
 	 * set, if so charge the init_mm (happens for pagecache usage).
 	 */
-	if (!memcg) {
-		if (!mm)
-			mm = &init_mm;
-
+	if (likely(!memcg)) {
 		rcu_read_lock();
 		mem = mem_cgroup_from_task(rcu_dereference(mm->owner));
 		/*
@@ -609,7 +604,6 @@ retry:
 		}
 	}
 
-	pc->ref_cnt = 1;
 	pc->mem_cgroup = mem;
 	pc->page = page;
 	/*
@@ -653,6 +647,17 @@ err:
 
 int mem_cgroup_charge(struct page *page, struct mm_struct *mm, gfp_t gfp_mask)
 {
+	/*
+	 * If already mapped, we don't have to account.
+	 * If page cache, page->mapping has address_space.
+	 * But page->mapping may have out-of-use anon_vma pointer,
+	 * detecit it by PageAnon() check. newly-mapped-anon's page->mapping
+	 * is NULL.
+  	 */
+	if (page_mapped(page) || (page->mapping && !PageAnon(page)))
+		return 0;
+	if (unlikely(!mm))
+		mm = &init_mm;
 	return mem_cgroup_charge_common(page, mm, gfp_mask,
 				MEM_CGROUP_CHARGE_TYPE_MAPPED, NULL);
 }
@@ -660,32 +665,17 @@ int mem_cgroup_charge(struct page *page, struct mm_struct *mm, gfp_t gfp_mask)
 int mem_cgroup_cache_charge(struct page *page, struct mm_struct *mm,
 				gfp_t gfp_mask)
 {
-	if (!mm)
+	if (unlikely(!mm))
 		mm = &init_mm;
 	return mem_cgroup_charge_common(page, mm, gfp_mask,
 				MEM_CGROUP_CHARGE_TYPE_CACHE, NULL);
 }
 
-int mem_cgroup_getref(struct page *page)
-{
-	struct page_cgroup *pc;
-
-	if (mem_cgroup_subsys.disabled)
-		return 0;
-
-	lock_page_cgroup(page);
-	pc = page_get_page_cgroup(page);
-	VM_BUG_ON(!pc);
-	pc->ref_cnt++;
-	unlock_page_cgroup(page);
-	return 0;
-}
-
 /*
- * Uncharging is always a welcome operation, we never complain, simply
- * uncharge.
+ * uncharge if !page_mapped(page)
  */
-void mem_cgroup_uncharge_page(struct page *page)
+static void
+__mem_cgroup_uncharge_common(struct page *page, enum charge_type ctype)
 {
 	struct page_cgroup *pc;
 	struct mem_cgroup *mem;
@@ -704,29 +694,41 @@ void mem_cgroup_uncharge_page(struct page *page)
 		goto unlock;
 
 	VM_BUG_ON(pc->page != page);
-	VM_BUG_ON(pc->ref_cnt <= 0);
 
-	if (--(pc->ref_cnt) == 0) {
-		mz = page_cgroup_zoneinfo(pc);
-		spin_lock_irqsave(&mz->lru_lock, flags);
-		__mem_cgroup_remove_list(mz, pc);
-		spin_unlock_irqrestore(&mz->lru_lock, flags);
+	if ((ctype == MEM_CGROUP_CHARGE_TYPE_MAPPED)
+	    && ((pc->flags & PAGE_CGROUP_FLAG_CACHE)
+		|| page_mapped(page)))
+		goto unlock;
 
-		page_assign_page_cgroup(page, NULL);
-		unlock_page_cgroup(page);
+	mz = page_cgroup_zoneinfo(pc);
+	spin_lock_irqsave(&mz->lru_lock, flags);
+	__mem_cgroup_remove_list(mz, pc);
+	spin_unlock_irqrestore(&mz->lru_lock, flags);
 
-		mem = pc->mem_cgroup;
-		res_counter_uncharge(&mem->res, PAGE_SIZE);
-		css_put(&mem->css);
+	page_assign_page_cgroup(page, NULL);
+	unlock_page_cgroup(page);
 
-		kmem_cache_free(page_cgroup_cache, pc);
-		return;
-	}
+	mem = pc->mem_cgroup;
+	res_counter_uncharge(&mem->res, PAGE_SIZE);
+	css_put(&mem->css);
 
+	kmem_cache_free(page_cgroup_cache, pc);
+	return;
 unlock:
 	unlock_page_cgroup(page);
 }
 
+void mem_cgroup_uncharge_page(struct page *page)
+{
+	__mem_cgroup_uncharge_common(page, MEM_CGROUP_CHARGE_TYPE_MAPPED);
+}
+
+void mem_cgroup_uncharge_cache_page(struct page *page)
+{
+	VM_BUG_ON(page_mapped(page));
+	__mem_cgroup_uncharge_common(page, MEM_CGROUP_CHARGE_TYPE_CACHE);
+}
+
 /*
  * Before starting migration, account against new page.
  */
@@ -757,15 +759,29 @@ int mem_cgroup_prepare_migration(struct page *page, struct page *newpage)
 	return ret;
 }
 
-/* remove redundant charge */
+/* remove redundant charge if migration failed*/
 void mem_cgroup_end_migration(struct page *newpage)
 {
-	mem_cgroup_uncharge_page(newpage);
+	/*
+	 * At success, page->mapping is not NULL.
+	 * special rollback care is necessary when
+	 * 1. at migration failure. (newpage->mapping is cleared in this case)
+	 * 2. the newpage was moved but not remapped again because the task
+	 *    exits and the newpage is obsolete. In this case, the new page
+	 *    may be a swapcache. So, we just call mem_cgroup_uncharge_page()
+	 *    always for avoiding mess. The  page_cgroup will be removed if
+	 *    unnecessary. File cache pages is still on radix-tree. Don't
+	 *    care it.
+	 */
+	if (!newpage->mapping)
+		__mem_cgroup_uncharge_common(newpage,
+					 MEM_CGROUP_CHARGE_TYPE_FORCE);
+	else if (PageAnon(newpage))
+		mem_cgroup_uncharge_page(newpage);
 }
 
 /*
  * This routine traverse page_cgroup in given list and drop them all.
- * This routine ignores page_cgroup->ref_cnt.
  * *And* this routine doesn't reclaim page itself, just removes page_cgroup.
  */
 #define FORCE_UNCHARGE_BATCH	(128)
@@ -795,7 +811,8 @@ static void mem_cgroup_force_empty_list(struct mem_cgroup *mem,
 		 * if it's under page migration.
 		 */
 		if (PageLRU(page)) {
-			mem_cgroup_uncharge_page(page);
+			__mem_cgroup_uncharge_common(page,
+					MEM_CGROUP_CHARGE_TYPE_FORCE);
 			put_page(page);
 			if (--count <= 0) {
 				count = FORCE_UNCHARGE_BATCH;
diff --git a/mm/migrate.c b/mm/migrate.c
index f6d7f8efd1a8..d8c65a65c61d 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -359,8 +359,7 @@ static int migrate_page_move_mapping(struct address_space *mapping,
 
 	write_unlock_irq(&mapping->tree_lock);
 	if (!PageSwapCache(newpage)) {
-		mem_cgroup_uncharge_page(page);
-		mem_cgroup_getref(newpage);
+		mem_cgroup_uncharge_cache_page(page);
 	}
 
 	return 0;
diff --git a/mm/rmap.c b/mm/rmap.c
index bf0a5b7cfb8e..abbd29f7c43f 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -576,14 +576,8 @@ void page_add_anon_rmap(struct page *page,
 	VM_BUG_ON(address < vma->vm_start || address >= vma->vm_end);
 	if (atomic_inc_and_test(&page->_mapcount))
 		__page_set_anon_rmap(page, vma, address);
-	else {
+	else
 		__page_check_anon_rmap(page, vma, address);
-		/*
-		 * We unconditionally charged during prepare, we uncharge here
-		 * This takes care of balancing the reference counts
-		 */
-		mem_cgroup_uncharge_page(page);
-	}
 }
 
 /**
@@ -614,12 +608,6 @@ void page_add_file_rmap(struct page *page)
 {
 	if (atomic_inc_and_test(&page->_mapcount))
 		__inc_zone_page_state(page, NR_FILE_MAPPED);
-	else
-		/*
-		 * We unconditionally charged during prepare, we uncharge here
-		 * This takes care of balancing the reference counts
-		 */
-		mem_cgroup_uncharge_page(page);
 }
 
 #ifdef CONFIG_DEBUG_VM
diff --git a/mm/shmem.c b/mm/shmem.c
index 9ffbea9b79e1..d58305e8a484 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -922,20 +922,26 @@ found:
 	error = 1;
 	if (!inode)
 		goto out;
-	/* Precharge page while we can wait, compensate afterwards */
+	/* Precharge page using GFP_KERNEL while we can wait */
 	error = mem_cgroup_cache_charge(page, current->mm, GFP_KERNEL);
 	if (error)
 		goto out;
 	error = radix_tree_preload(GFP_KERNEL);
-	if (error)
-		goto uncharge;
+	if (error) {
+		mem_cgroup_uncharge_cache_page(page);
+		goto out;
+	}
 	error = 1;
 
 	spin_lock(&info->lock);
 	ptr = shmem_swp_entry(info, idx, NULL);
-	if (ptr && ptr->val == entry.val)
+	if (ptr && ptr->val == entry.val) {
 		error = add_to_page_cache(page, inode->i_mapping,
 						idx, GFP_NOWAIT);
+		/* does mem_cgroup_uncharge_cache_page on error */
+	} else	/* we must compensate for our precharge above */
+		mem_cgroup_uncharge_cache_page(page);
+
 	if (error == -EEXIST) {
 		struct page *filepage = find_get_page(inode->i_mapping, idx);
 		error = 1;
@@ -961,8 +967,6 @@ found:
 		shmem_swp_unmap(ptr);
 	spin_unlock(&info->lock);
 	radix_tree_preload_end();
-uncharge:
-	mem_cgroup_uncharge_page(page);
 out:
 	unlock_page(page);
 	page_cache_release(page);
@@ -1319,7 +1323,7 @@ repeat:
 					page_cache_release(swappage);
 					goto failed;
 				}
-				mem_cgroup_uncharge_page(swappage);
+				mem_cgroup_uncharge_cache_page(swappage);
 			}
 			page_cache_release(swappage);
 			goto repeat;
@@ -1358,6 +1362,8 @@ repeat:
 		}
 
 		if (!filepage) {
+			int ret;
+
 			spin_unlock(&info->lock);
 			filepage = shmem_alloc_page(gfp, info, idx);
 			if (!filepage) {
@@ -1386,10 +1392,18 @@ repeat:
 				swap = *entry;
 				shmem_swp_unmap(entry);
 			}
-			if (error || swap.val || 0 != add_to_page_cache_lru(
-					filepage, mapping, idx, GFP_NOWAIT)) {
+			ret = error || swap.val;
+			if (ret)
+				mem_cgroup_uncharge_cache_page(filepage);
+			else
+				ret = add_to_page_cache_lru(filepage, mapping,
+						idx, GFP_NOWAIT);
+			/*
+			 * At add_to_page_cache_lru() failure, uncharge will
+			 * be done automatically.
+			 */
+			if (ret) {
 				spin_unlock(&info->lock);
-				mem_cgroup_uncharge_page(filepage);
 				page_cache_release(filepage);
 				shmem_unacct_blocks(info->flags, 1);
 				shmem_free_blocks(inode, 1);
@@ -1398,7 +1412,6 @@ repeat:
 					goto failed;
 				goto repeat;
 			}
-			mem_cgroup_uncharge_page(filepage);
 			info->flags |= SHMEM_PAGEIN;
 		}
 

From c9b0ed51483cc2fc42bb801b6675c4231b0e4634 Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Fri, 25 Jul 2008 01:47:15 -0700
Subject: [PATCH 0799/1171] memcg: helper function for relcaim from shmem.

A new call, mem_cgroup_shrink_usage() is added for shmem handling and
relacing non-standard usage of mem_cgroup_charge/uncharge.

Now, shmem calls mem_cgroup_charge() just for reclaim some pages from
mem_cgroup.  In general, shmem is used by some process group and not for
global resource (like file caches).  So, it's reasonable to reclaim pages
from mem_cgroup where shmem is mainly used.

[hugh@veritas.com: shmem_getpage release page sooner]
[hugh@veritas.com: mem_cgroup_shrink_usage css_put]
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: YAMAMOTO Takashi <yamamoto@valinux.co.jp>
Cc: Paul Menage <menage@google.com>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/memcontrol.h |  7 +++++++
 mm/memcontrol.c            | 26 ++++++++++++++++++++++++++
 mm/shmem.c                 | 11 ++++-------
 3 files changed, 37 insertions(+), 7 deletions(-)

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index b4980b8f048e..fdf3967e1397 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -37,6 +37,8 @@ extern int mem_cgroup_cache_charge(struct page *page, struct mm_struct *mm,
 extern void mem_cgroup_uncharge_page(struct page *page);
 extern void mem_cgroup_uncharge_cache_page(struct page *page);
 extern void mem_cgroup_move_lists(struct page *page, bool active);
+extern int mem_cgroup_shrink_usage(struct mm_struct *mm, gfp_t gfp_mask);
+
 extern unsigned long mem_cgroup_isolate_pages(unsigned long nr_to_scan,
 					struct list_head *dst,
 					unsigned long *scanned, int order,
@@ -102,6 +104,11 @@ static inline void mem_cgroup_uncharge_cache_page(struct page *page)
 {
 }
 
+static inline int mem_cgroup_shrink_usage(struct mm_struct *mm, gfp_t gfp_mask)
+{
+	return 0;
+}
+
 static inline void mem_cgroup_move_lists(struct page *page, bool active)
 {
 }
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index a61706193c31..f46b8615de6c 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -780,6 +780,32 @@ void mem_cgroup_end_migration(struct page *newpage)
 		mem_cgroup_uncharge_page(newpage);
 }
 
+/*
+ * A call to try to shrink memory usage under specified resource controller.
+ * This is typically used for page reclaiming for shmem for reducing side
+ * effect of page allocation from shmem, which is used by some mem_cgroup.
+ */
+int mem_cgroup_shrink_usage(struct mm_struct *mm, gfp_t gfp_mask)
+{
+	struct mem_cgroup *mem;
+	int progress = 0;
+	int retry = MEM_CGROUP_RECLAIM_RETRIES;
+
+	rcu_read_lock();
+	mem = mem_cgroup_from_task(rcu_dereference(mm->owner));
+	css_get(&mem->css);
+	rcu_read_unlock();
+
+	do {
+		progress = try_to_free_mem_cgroup_pages(mem, gfp_mask);
+	} while (!progress && --retry);
+
+	css_put(&mem->css);
+	if (!retry)
+		return -ENOMEM;
+	return 0;
+}
+
 /*
  * This routine traverse page_cgroup in given list and drop them all.
  * *And* this routine doesn't reclaim page itself, just removes page_cgroup.
diff --git a/mm/shmem.c b/mm/shmem.c
index d58305e8a484..f92fea94d037 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -1315,17 +1315,14 @@ repeat:
 			shmem_swp_unmap(entry);
 			spin_unlock(&info->lock);
 			unlock_page(swappage);
+			page_cache_release(swappage);
 			if (error == -ENOMEM) {
 				/* allow reclaim from this memory cgroup */
-				error = mem_cgroup_cache_charge(swappage,
-					current->mm, gfp & ~__GFP_HIGHMEM);
-				if (error) {
-					page_cache_release(swappage);
+				error = mem_cgroup_shrink_usage(current->mm,
+								gfp);
+				if (error)
 					goto failed;
-				}
-				mem_cgroup_uncharge_cache_page(swappage);
 			}
-			page_cache_release(swappage);
 			goto repeat;
 		}
 	} else if (sgp == SGP_READ && !filepage) {

From b76734e5e34e1889ab9fc5f3756570b1129f0f50 Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Fri, 25 Jul 2008 01:47:16 -0700
Subject: [PATCH 0800/1171] memcg: add hints for branch

Showing brach direction for obvious conditions.

Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: YAMAMOTO Takashi <yamamoto@valinux.co.jp>
Cc: Paul Menage <menage@google.com>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/memcontrol.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index f46b8615de6c..04ded27f6226 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -550,7 +550,7 @@ retry:
 	 * The page_cgroup exists and
 	 * the page has already been accounted.
 	 */
-	if (pc) {
+	if (unlikely(pc)) {
 		VM_BUG_ON(pc->page != page);
 		VM_BUG_ON(!pc->mem_cgroup);
 		unlock_page_cgroup(page);
@@ -559,7 +559,7 @@ retry:
 	unlock_page_cgroup(page);
 
 	pc = kmem_cache_alloc(page_cgroup_cache, gfp_mask);
-	if (pc == NULL)
+	if (unlikely(pc == NULL))
 		goto err;
 
 	/*
@@ -616,7 +616,7 @@ retry:
 		pc->flags = PAGE_CGROUP_FLAG_ACTIVE;
 
 	lock_page_cgroup(page);
-	if (page_get_page_cgroup(page)) {
+	if (unlikely(page_get_page_cgroup(page))) {
 		unlock_page_cgroup(page);
 		/*
 		 * Another charge has been added to this page already.
@@ -690,7 +690,7 @@ __mem_cgroup_uncharge_common(struct page *page, enum charge_type ctype)
 	 */
 	lock_page_cgroup(page);
 	pc = page_get_page_cgroup(page);
-	if (!pc)
+	if (unlikely(!pc))
 		goto unlock;
 
 	VM_BUG_ON(pc->page != page);

From accf163e6ab729f1fc5fffaa0310e498270bf4e7 Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Fri, 25 Jul 2008 01:47:17 -0700
Subject: [PATCH 0801/1171] memcg: remove a redundant check

Because of remove refcnt patch, it's very rare case to that
mem_cgroup_charge_common() is called against a page which is accounted.

mem_cgroup_charge_common() is called when.
 1. a page is added into file cache.
 2. an anon page is _newly_ mapped.

A racy case is that a newly-swapped-in anonymous page is referred from
prural threads in do_swap_page() at the same time.
(a page is not Locked when mem_cgroup_charge() is called from do_swap_page.)

Another case is shmem. It charges its page before calling add_to_page_cache().
Then, mem_cgroup_charge_cache() is called twice. This case is handled in
mem_cgroup_cache_charge(). But this check may be too hacky...

Signed-off-by : KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: YAMAMOTO Takashi <yamamoto@valinux.co.jp>
Cc: Paul Menage <menage@google.com>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/memcontrol.c | 53 +++++++++++++++++++++++--------------------------
 1 file changed, 25 insertions(+), 28 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 04ded27f6226..5b3759bd5494 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -536,28 +536,6 @@ static int mem_cgroup_charge_common(struct page *page, struct mm_struct *mm,
 	if (mem_cgroup_subsys.disabled)
 		return 0;
 
-	/*
-	 * Should page_cgroup's go to their own slab?
-	 * One could optimize the performance of the charging routine
-	 * by saving a bit in the page_flags and using it as a lock
-	 * to see if the cgroup page already has a page_cgroup associated
-	 * with it
-	 */
-retry:
-	lock_page_cgroup(page);
-	pc = page_get_page_cgroup(page);
-	/*
-	 * The page_cgroup exists and
-	 * the page has already been accounted.
-	 */
-	if (unlikely(pc)) {
-		VM_BUG_ON(pc->page != page);
-		VM_BUG_ON(!pc->mem_cgroup);
-		unlock_page_cgroup(page);
-		goto done;
-	}
-	unlock_page_cgroup(page);
-
 	pc = kmem_cache_alloc(page_cgroup_cache, gfp_mask);
 	if (unlikely(pc == NULL))
 		goto err;
@@ -618,15 +596,10 @@ retry:
 	lock_page_cgroup(page);
 	if (unlikely(page_get_page_cgroup(page))) {
 		unlock_page_cgroup(page);
-		/*
-		 * Another charge has been added to this page already.
-		 * We take lock_page_cgroup(page) again and read
-		 * page->cgroup, increment refcnt.... just retry is OK.
-		 */
 		res_counter_uncharge(&mem->res, PAGE_SIZE);
 		css_put(&mem->css);
 		kmem_cache_free(page_cgroup_cache, pc);
-		goto retry;
+		goto done;
 	}
 	page_assign_page_cgroup(page, pc);
 
@@ -665,8 +638,32 @@ int mem_cgroup_charge(struct page *page, struct mm_struct *mm, gfp_t gfp_mask)
 int mem_cgroup_cache_charge(struct page *page, struct mm_struct *mm,
 				gfp_t gfp_mask)
 {
+	/*
+	 * Corner case handling. This is called from add_to_page_cache()
+	 * in usual. But some FS (shmem) precharges this page before calling it
+	 * and call add_to_page_cache() with GFP_NOWAIT.
+	 *
+	 * For GFP_NOWAIT case, the page may be pre-charged before calling
+	 * add_to_page_cache(). (See shmem.c) check it here and avoid to call
+	 * charge twice. (It works but has to pay a bit larger cost.)
+	 */
+	if (!(gfp_mask & __GFP_WAIT)) {
+		struct page_cgroup *pc;
+
+		lock_page_cgroup(page);
+		pc = page_get_page_cgroup(page);
+		if (pc) {
+			VM_BUG_ON(pc->page != page);
+			VM_BUG_ON(!pc->mem_cgroup);
+			unlock_page_cgroup(page);
+			return 0;
+		}
+		unlock_page_cgroup(page);
+	}
+
 	if (unlikely(!mm))
 		mm = &init_mm;
+
 	return mem_cgroup_charge_common(page, mm, gfp_mask,
 				MEM_CGROUP_CHARGE_TYPE_CACHE, NULL);
 }

From cede86acd8bd5d2205dec28db8ac86410a3a19e8 Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Fri, 25 Jul 2008 01:47:18 -0700
Subject: [PATCH 0802/1171] memcg: clean up checking of the disabled flag

Those checks are unnecessary, because when the subsystem is disabled
it can't be mounted, so those functions won't get called.

The check is needed in functions which will be called in other places
except cgroup.

[hugh@veritas.com: further checking of disabled flag]
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Acked-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/memcontrol.c | 23 ++++++++++++-----------
 1 file changed, 12 insertions(+), 11 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 5b3759bd5494..0c035647d36a 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -354,6 +354,9 @@ void mem_cgroup_move_lists(struct page *page, bool active)
 	struct mem_cgroup_per_zone *mz;
 	unsigned long flags;
 
+	if (mem_cgroup_subsys.disabled)
+		return;
+
 	/*
 	 * We cannot lock_page_cgroup while holding zone's lru_lock,
 	 * because other holders of lock_page_cgroup can be interrupted
@@ -533,9 +536,6 @@ static int mem_cgroup_charge_common(struct page *page, struct mm_struct *mm,
 	unsigned long nr_retries = MEM_CGROUP_RECLAIM_RETRIES;
 	struct mem_cgroup_per_zone *mz;
 
-	if (mem_cgroup_subsys.disabled)
-		return 0;
-
 	pc = kmem_cache_alloc(page_cgroup_cache, gfp_mask);
 	if (unlikely(pc == NULL))
 		goto err;
@@ -620,6 +620,9 @@ err:
 
 int mem_cgroup_charge(struct page *page, struct mm_struct *mm, gfp_t gfp_mask)
 {
+	if (mem_cgroup_subsys.disabled)
+		return 0;
+
 	/*
 	 * If already mapped, we don't have to account.
 	 * If page cache, page->mapping has address_space.
@@ -638,6 +641,9 @@ int mem_cgroup_charge(struct page *page, struct mm_struct *mm, gfp_t gfp_mask)
 int mem_cgroup_cache_charge(struct page *page, struct mm_struct *mm,
 				gfp_t gfp_mask)
 {
+	if (mem_cgroup_subsys.disabled)
+		return 0;
+
 	/*
 	 * Corner case handling. This is called from add_to_page_cache()
 	 * in usual. But some FS (shmem) precharges this page before calling it
@@ -788,6 +794,9 @@ int mem_cgroup_shrink_usage(struct mm_struct *mm, gfp_t gfp_mask)
 	int progress = 0;
 	int retry = MEM_CGROUP_RECLAIM_RETRIES;
 
+	if (mem_cgroup_subsys.disabled)
+		return 0;
+
 	rcu_read_lock();
 	mem = mem_cgroup_from_task(rcu_dereference(mm->owner));
 	css_get(&mem->css);
@@ -857,9 +866,6 @@ static int mem_cgroup_force_empty(struct mem_cgroup *mem)
 	int ret = -EBUSY;
 	int node, zid;
 
-	if (mem_cgroup_subsys.disabled)
-		return 0;
-
 	css_get(&mem->css);
 	/*
 	 * page reclaim code (kswapd etc..) will move pages between
@@ -1103,8 +1109,6 @@ static void mem_cgroup_destroy(struct cgroup_subsys *ss,
 static int mem_cgroup_populate(struct cgroup_subsys *ss,
 				struct cgroup *cont)
 {
-	if (mem_cgroup_subsys.disabled)
-		return 0;
 	return cgroup_add_files(cont, ss, mem_cgroup_files,
 					ARRAY_SIZE(mem_cgroup_files));
 }
@@ -1117,9 +1121,6 @@ static void mem_cgroup_move_task(struct cgroup_subsys *ss,
 	struct mm_struct *mm;
 	struct mem_cgroup *mem, *old_mem;
 
-	if (mem_cgroup_subsys.disabled)
-		return;
-
 	mm = get_task_mm(p);
 	if (mm == NULL)
 		return;

From 12b9804419cfb1c1bdac413f6c373af3b88d154b Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Fri, 25 Jul 2008 01:47:19 -0700
Subject: [PATCH 0803/1171] res_counter: limit change support ebusy

Add an interface to set limit.  This is necessary to memory resource
controller because it shrinks usage at set limit.

Other controllers may not need this interface to shrink usage because
shrinking is not necessary or impossible.

Acked-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Acked-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Paul Menage <menage@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/res_counter.h | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/include/linux/res_counter.h b/include/linux/res_counter.h
index 290205dfe094..fdeadd9740dc 100644
--- a/include/linux/res_counter.h
+++ b/include/linux/res_counter.h
@@ -158,4 +158,20 @@ static inline void res_counter_reset_failcnt(struct res_counter *cnt)
 	cnt->failcnt = 0;
 	spin_unlock_irqrestore(&cnt->lock, flags);
 }
+
+static inline int res_counter_set_limit(struct res_counter *cnt,
+		unsigned long long limit)
+{
+	unsigned long flags;
+	int ret = -EBUSY;
+
+	spin_lock_irqsave(&cnt->lock, flags);
+	if (cnt->usage < limit) {
+		cnt->limit = limit;
+		ret = 0;
+	}
+	spin_unlock_irqrestore(&cnt->lock, flags);
+	return ret;
+}
+
 #endif

From 628f42355389cfb596ca3a5a5f64fb9054a2a06a Mon Sep 17 00:00:00 2001
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Date: Fri, 25 Jul 2008 01:47:20 -0700
Subject: [PATCH 0804/1171] memcg: limit change shrink usage

Shrinking memory usage at limit change.

[akpm@linux-foundation.org: coding-style fixes]
Acked-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Acked-by: Pavel Emelyanov <xemul@openvz.org>
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Paul Menage <menage@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/controllers/memory.txt |  3 +-
 mm/memcontrol.c                      | 48 +++++++++++++++++++++++++---
 2 files changed, 45 insertions(+), 6 deletions(-)

diff --git a/Documentation/controllers/memory.txt b/Documentation/controllers/memory.txt
index 866b9cd9a959..9b53d5827361 100644
--- a/Documentation/controllers/memory.txt
+++ b/Documentation/controllers/memory.txt
@@ -242,8 +242,7 @@ rmdir() if there are no tasks.
 1. Add support for accounting huge pages (as a separate controller)
 2. Make per-cgroup scanner reclaim not-shared pages first
 3. Teach controller to account for shared-pages
-4. Start reclamation when the limit is lowered
-5. Start reclamation in the background when the limit is
+4. Start reclamation in the background when the limit is
    not yet hit but the usage is getting closer
 
 Summary
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 0c035647d36a..fba566c51322 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -812,6 +812,30 @@ int mem_cgroup_shrink_usage(struct mm_struct *mm, gfp_t gfp_mask)
 	return 0;
 }
 
+int mem_cgroup_resize_limit(struct mem_cgroup *memcg, unsigned long long val)
+{
+
+	int retry_count = MEM_CGROUP_RECLAIM_RETRIES;
+	int progress;
+	int ret = 0;
+
+	while (res_counter_set_limit(&memcg->res, val)) {
+		if (signal_pending(current)) {
+			ret = -EINTR;
+			break;
+		}
+		if (!retry_count) {
+			ret = -EBUSY;
+			break;
+		}
+		progress = try_to_free_mem_cgroup_pages(memcg, GFP_KERNEL);
+		if (!progress)
+			retry_count--;
+	}
+	return ret;
+}
+
+
 /*
  * This routine traverse page_cgroup in given list and drop them all.
  * *And* this routine doesn't reclaim page itself, just removes page_cgroup.
@@ -896,13 +920,29 @@ static u64 mem_cgroup_read(struct cgroup *cont, struct cftype *cft)
 	return res_counter_read_u64(&mem_cgroup_from_cont(cont)->res,
 				    cft->private);
 }
-
+/*
+ * The user of this function is...
+ * RES_LIMIT.
+ */
 static int mem_cgroup_write(struct cgroup *cont, struct cftype *cft,
 			    const char *buffer)
 {
-	return res_counter_write(&mem_cgroup_from_cont(cont)->res,
-				 cft->private, buffer,
-				 res_counter_memparse_write_strategy);
+	struct mem_cgroup *memcg = mem_cgroup_from_cont(cont);
+	unsigned long long val;
+	int ret;
+
+	switch (cft->private) {
+	case RES_LIMIT:
+		/* This function does all necessary parse...reuse it */
+		ret = res_counter_memparse_write_strategy(buffer, &val);
+		if (!ret)
+			ret = mem_cgroup_resize_limit(memcg, val);
+		break;
+	default:
+		ret = -EINVAL; /* should be BUG() ? */
+		break;
+	}
+	return ret;
 }
 
 static int mem_cgroup_reset(struct cgroup *cont, unsigned int event)

From 0b2f630a28d53b5a2082a5275bc3334b10373508 Mon Sep 17 00:00:00 2001
From: Miao Xie <miaox@cn.fujitsu.com>
Date: Fri, 25 Jul 2008 01:47:21 -0700
Subject: [PATCH 0805/1171] cpusets: restructure the function update_cpumask()
 and update_nodemask()

Extract two functions from update_cpumask() and update_nodemask().They
will be used later for updating tasks' cpus_allowed and mems_allowed after
CPU/NODE offline/online.

[lizf@cn.fujitsu.com: build fix]
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Acked-by: Paul Jackson <pj@sgi.com>
Cc: David Rientjes <rientjes@google.com>
Cc:  Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/cpuset.c | 181 +++++++++++++++++++++++++++++-------------------
 1 file changed, 109 insertions(+), 72 deletions(-)

diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 276ce7e4f1ab..7326d51eefe1 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -763,6 +763,37 @@ static void cpuset_change_cpumask(struct task_struct *tsk,
 	set_cpus_allowed_ptr(tsk, &((cgroup_cs(scan->cg))->cpus_allowed));
 }
 
+/**
+ * update_tasks_cpumask - Update the cpumasks of tasks in the cpuset.
+ * @cs: the cpuset in which each task's cpus_allowed mask needs to be changed
+ *
+ * Called with cgroup_mutex held
+ *
+ * The cgroup_scan_tasks() function will scan all the tasks in a cgroup,
+ * calling callback functions for each.
+ *
+ * Return 0 if successful, -errno if not.
+ */
+static int update_tasks_cpumask(struct cpuset *cs)
+{
+	struct cgroup_scanner scan;
+	struct ptr_heap heap;
+	int retval;
+
+	retval = heap_init(&heap, PAGE_SIZE, GFP_KERNEL, &started_after);
+	if (retval)
+		return retval;
+
+	scan.cg = cs->css.cgroup;
+	scan.test_task = cpuset_test_cpumask;
+	scan.process_task = cpuset_change_cpumask;
+	scan.heap = &heap;
+	retval = cgroup_scan_tasks(&scan);
+
+	heap_free(&heap);
+	return retval;
+}
+
 /**
  * update_cpumask - update the cpus_allowed mask of a cpuset and all tasks in it
  * @cs: the cpuset to consider
@@ -771,8 +802,6 @@ static void cpuset_change_cpumask(struct task_struct *tsk,
 static int update_cpumask(struct cpuset *cs, const char *buf)
 {
 	struct cpuset trialcs;
-	struct cgroup_scanner scan;
-	struct ptr_heap heap;
 	int retval;
 	int is_load_balanced;
 
@@ -806,10 +835,6 @@ static int update_cpumask(struct cpuset *cs, const char *buf)
 	if (cpus_equal(cs->cpus_allowed, trialcs.cpus_allowed))
 		return 0;
 
-	retval = heap_init(&heap, PAGE_SIZE, GFP_KERNEL, &started_after);
-	if (retval)
-		return retval;
-
 	is_load_balanced = is_sched_load_balance(&trialcs);
 
 	mutex_lock(&callback_mutex);
@@ -820,12 +845,9 @@ static int update_cpumask(struct cpuset *cs, const char *buf)
 	 * Scan tasks in the cpuset, and update the cpumasks of any
 	 * that need an update.
 	 */
-	scan.cg = cs->css.cgroup;
-	scan.test_task = cpuset_test_cpumask;
-	scan.process_task = cpuset_change_cpumask;
-	scan.heap = &heap;
-	cgroup_scan_tasks(&scan);
-	heap_free(&heap);
+	retval = update_tasks_cpumask(cs);
+	if (retval < 0)
+		return retval;
 
 	if (is_load_balanced)
 		rebuild_sched_domains();
@@ -881,73 +903,25 @@ static void cpuset_migrate_mm(struct mm_struct *mm, const nodemask_t *from,
 	mutex_unlock(&callback_mutex);
 }
 
-/*
- * Handle user request to change the 'mems' memory placement
- * of a cpuset.  Needs to validate the request, update the
- * cpusets mems_allowed and mems_generation, and for each
- * task in the cpuset, rebind any vma mempolicies and if
- * the cpuset is marked 'memory_migrate', migrate the tasks
- * pages to the new memory.
- *
- * Call with cgroup_mutex held.  May take callback_mutex during call.
- * Will take tasklist_lock, scan tasklist for tasks in cpuset cs,
- * lock each such tasks mm->mmap_sem, scan its vma's and rebind
- * their mempolicies to the cpusets new mems_allowed.
- */
-
 static void *cpuset_being_rebound;
 
-static int update_nodemask(struct cpuset *cs, const char *buf)
+/**
+ * update_tasks_nodemask - Update the nodemasks of tasks in the cpuset.
+ * @cs: the cpuset in which each task's mems_allowed mask needs to be changed
+ * @oldmem: old mems_allowed of cpuset cs
+ *
+ * Called with cgroup_mutex held
+ * Return 0 if successful, -errno if not.
+ */
+static int update_tasks_nodemask(struct cpuset *cs, const nodemask_t *oldmem)
 {
-	struct cpuset trialcs;
-	nodemask_t oldmem;
 	struct task_struct *p;
 	struct mm_struct **mmarray;
 	int i, n, ntasks;
 	int migrate;
 	int fudge;
-	int retval;
 	struct cgroup_iter it;
-
-	/*
-	 * top_cpuset.mems_allowed tracks node_stats[N_HIGH_MEMORY];
-	 * it's read-only
-	 */
-	if (cs == &top_cpuset)
-		return -EACCES;
-
-	trialcs = *cs;
-
-	/*
-	 * An empty mems_allowed is ok iff there are no tasks in the cpuset.
-	 * Since nodelist_parse() fails on an empty mask, we special case
-	 * that parsing.  The validate_change() call ensures that cpusets
-	 * with tasks have memory.
-	 */
-	if (!*buf) {
-		nodes_clear(trialcs.mems_allowed);
-	} else {
-		retval = nodelist_parse(buf, trialcs.mems_allowed);
-		if (retval < 0)
-			goto done;
-
-		if (!nodes_subset(trialcs.mems_allowed,
-				node_states[N_HIGH_MEMORY]))
-			return -EINVAL;
-	}
-	oldmem = cs->mems_allowed;
-	if (nodes_equal(oldmem, trialcs.mems_allowed)) {
-		retval = 0;		/* Too easy - nothing to do */
-		goto done;
-	}
-	retval = validate_change(cs, &trialcs);
-	if (retval < 0)
-		goto done;
-
-	mutex_lock(&callback_mutex);
-	cs->mems_allowed = trialcs.mems_allowed;
-	cs->mems_generation = cpuset_mems_generation++;
-	mutex_unlock(&callback_mutex);
+	int retval;
 
 	cpuset_being_rebound = cs;		/* causes mpol_dup() rebind */
 
@@ -1014,7 +988,7 @@ static int update_nodemask(struct cpuset *cs, const char *buf)
 
 		mpol_rebind_mm(mm, &cs->mems_allowed);
 		if (migrate)
-			cpuset_migrate_mm(mm, &oldmem, &cs->mems_allowed);
+			cpuset_migrate_mm(mm, oldmem, &cs->mems_allowed);
 		mmput(mm);
 	}
 
@@ -1026,6 +1000,70 @@ done:
 	return retval;
 }
 
+/*
+ * Handle user request to change the 'mems' memory placement
+ * of a cpuset.  Needs to validate the request, update the
+ * cpusets mems_allowed and mems_generation, and for each
+ * task in the cpuset, rebind any vma mempolicies and if
+ * the cpuset is marked 'memory_migrate', migrate the tasks
+ * pages to the new memory.
+ *
+ * Call with cgroup_mutex held.  May take callback_mutex during call.
+ * Will take tasklist_lock, scan tasklist for tasks in cpuset cs,
+ * lock each such tasks mm->mmap_sem, scan its vma's and rebind
+ * their mempolicies to the cpusets new mems_allowed.
+ */
+static int update_nodemask(struct cpuset *cs, const char *buf)
+{
+	struct cpuset trialcs;
+	nodemask_t oldmem;
+	int retval;
+
+	/*
+	 * top_cpuset.mems_allowed tracks node_stats[N_HIGH_MEMORY];
+	 * it's read-only
+	 */
+	if (cs == &top_cpuset)
+		return -EACCES;
+
+	trialcs = *cs;
+
+	/*
+	 * An empty mems_allowed is ok iff there are no tasks in the cpuset.
+	 * Since nodelist_parse() fails on an empty mask, we special case
+	 * that parsing.  The validate_change() call ensures that cpusets
+	 * with tasks have memory.
+	 */
+	if (!*buf) {
+		nodes_clear(trialcs.mems_allowed);
+	} else {
+		retval = nodelist_parse(buf, trialcs.mems_allowed);
+		if (retval < 0)
+			goto done;
+
+		if (!nodes_subset(trialcs.mems_allowed,
+				node_states[N_HIGH_MEMORY]))
+			return -EINVAL;
+	}
+	oldmem = cs->mems_allowed;
+	if (nodes_equal(oldmem, trialcs.mems_allowed)) {
+		retval = 0;		/* Too easy - nothing to do */
+		goto done;
+	}
+	retval = validate_change(cs, &trialcs);
+	if (retval < 0)
+		goto done;
+
+	mutex_lock(&callback_mutex);
+	cs->mems_allowed = trialcs.mems_allowed;
+	cs->mems_generation = cpuset_mems_generation++;
+	mutex_unlock(&callback_mutex);
+
+	retval = update_tasks_nodemask(cs, &oldmem);
+done:
+	return retval;
+}
+
 int current_cpuset_is_being_rebound(void)
 {
 	return task_cs(current) == cpuset_being_rebound;
@@ -1935,7 +1973,6 @@ void __init cpuset_init_smp(void)
 }
 
 /**
-
  * cpuset_cpus_allowed - return cpus_allowed mask from a tasks cpuset.
  * @tsk: pointer to task_struct from which to obtain cpuset->cpus_allowed.
  * @pmask: pointer to cpumask_t variable to receive cpus_allowed set.

From f9b4fb8dabf38fb456c97f01aace07cb6e7c1723 Mon Sep 17 00:00:00 2001
From: Miao Xie <miaox@cn.fujitsu.com>
Date: Fri, 25 Jul 2008 01:47:22 -0700
Subject: [PATCH 0806/1171] cpusets: update task's cpus_allowed and
 mems_allowed after CPU/NODE offline/online

The bug is that a task may run on the cpu/node which is not in its
cpuset.cpus/ cpuset.mems.

It can be reproduced by the following commands:
-----------------------------------
# mkdir /dev/cpuset
# mount -t cpuset xxx /dev/cpuset
# mkdir /dev/cpuset/0
# echo 0-1 > /dev/cpuset/0/cpus
# echo 0 > /dev/cpuset/0/mems
# echo $$ > /dev/cpuset/0/tasks
# echo 0 > /sys/devices/system/cpu/cpu1/online
# echo 1 > /sys/devices/system/cpu/cpu1/online
-----------------------------------

There is only CPU0 in cpuset.cpus, but the task in this cpuset runs on
both CPU0 and CPU1.

It is because the task's cpu_allowed didn't get updated after we did CPU
offline/online manipulation.  Similar for mem_allowed.

This patch fixes this bug expect for root cpuset.  Because there is a
problem about root cpuset, in that whether it is necessary to update all
the tasks in root cpuset or not after cpu/node offline/online.

If updating, some kernel threads which is bound into a specified cpu will
be unbound.

If not updating, there is a bug in root cpuset.  This bug is also caused
by offline/online manipulation.  For example, there is a dual-cpu machine.
 we create a sub cpuset in root cpuset and assign 1 to its cpus.  And then
we attach some tasks into this sub cpuset.  After this, we offline CPU1.
Now, the tasks in this new cpuset are moved into root cpuset automatically
because there is no cpu in sub cpuset.  Then we online CPU1, we find all
the tasks which doesn't belong to root cpuset originally just run on CPU0.

Maybe we need to add a flag in the task_struct to mark which task can't be
unbound?

Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Acked-by: Paul Jackson <pj@sgi.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Paul Jackson <pj@sgi.com>
Cc: Paul Menage <menage@google.com>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/cpuset.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 7326d51eefe1..6eae6639e851 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -1851,6 +1851,7 @@ static void scan_for_empty_cpusets(const struct cpuset *root)
 	struct cpuset *child;	/* scans child cpusets of cp */
 	struct list_head queue;
 	struct cgroup *cont;
+	nodemask_t oldmems;
 
 	INIT_LIST_HEAD(&queue);
 
@@ -1870,6 +1871,8 @@ static void scan_for_empty_cpusets(const struct cpuset *root)
 		    nodes_subset(cp->mems_allowed, node_states[N_HIGH_MEMORY]))
 			continue;
 
+		oldmems = cp->mems_allowed;
+
 		/* Remove offline cpus and mems from this cpuset. */
 		mutex_lock(&callback_mutex);
 		cpus_and(cp->cpus_allowed, cp->cpus_allowed, cpu_online_map);
@@ -1881,6 +1884,10 @@ static void scan_for_empty_cpusets(const struct cpuset *root)
 		if (cpus_empty(cp->cpus_allowed) ||
 		     nodes_empty(cp->mems_allowed))
 			remove_tasks_in_empty_cpuset(cp);
+		else {
+			update_tasks_cpumask(cp);
+			update_tasks_nodemask(cp, &oldmems);
+		}
 	}
 }
 

From c372e817afc629fea9ff6321313325ed0b4a855b Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Fri, 25 Jul 2008 01:47:23 -0700
Subject: [PATCH 0807/1171] cpuset: avoid unnecessary sched domains rebuilding

When changing 'sched_relax_domain_level', don't rebuild sched domains if
'cpus' is empty or 'sched_load_balance' is not set.

Also make the comments of rebuild_sched_domains() more readable.

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Cc: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Cc: Paul Jackson <pj@sgi.com>
Cc: Paul Menage <menage@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/cpuset.c | 18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)

diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 6eae6639e851..60d2c4702c6c 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -496,11 +496,16 @@ update_domain_attr(struct sched_domain_attr *dattr, struct cpuset *c)
 /*
  * rebuild_sched_domains()
  *
- * If the flag 'sched_load_balance' of any cpuset with non-empty
- * 'cpus' changes, or if the 'cpus' allowed changes in any cpuset
- * which has that flag enabled, or if any cpuset with a non-empty
- * 'cpus' is removed, then call this routine to rebuild the
- * scheduler's dynamic sched domains.
+ * This routine will be called to rebuild the scheduler's dynamic
+ * sched domains:
+ * - if the flag 'sched_load_balance' of any cpuset with non-empty
+ *   'cpus' changes,
+ * - or if the 'cpus' allowed changes in any cpuset which has that
+ *   flag enabled,
+ * - or if the 'sched_relax_domain_level' of any cpuset which has
+ *   that flag enabled and with non-empty 'cpus' changes,
+ * - or if any cpuset with non-empty 'cpus' is removed,
+ * - or if a cpu gets offlined.
  *
  * This routine builds a partial partition of the systems CPUs
  * (the set of non-overlappping cpumask_t's in the array 'part'
@@ -1076,7 +1081,8 @@ static int update_relax_domain_level(struct cpuset *cs, s64 val)
 
 	if (val != cs->relax_domain_level) {
 		cs->relax_domain_level = val;
-		rebuild_sched_domains();
+		if (!cpus_empty(cs->cpus_allowed) && is_sched_load_balance(cs))
+			rebuild_sched_domains();
 	}
 
 	return 0;

From 489a5393a20dcbf91104052120eb2eff8791b61b Mon Sep 17 00:00:00 2001
From: Lai Jiangshan <laijs@cn.fujitsu.com>
Date: Fri, 25 Jul 2008 01:47:23 -0700
Subject: [PATCH 0808/1171] cpuset: don't pass empty cpumasks to
 partition_sched_domains()

I create lots of empty cpusets(empty cpumasks) and turn off the
"sched_load_balance" in top cpuset.

I found that all these empty cpumasks are passed to
partition_sched_domains() in rebuild_sched_domains(), it's very
time-consuming for partition_sched_domains() and it's not need.

It also reduce memory consumed and some works in rebuild_sched_domains()
too.

Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Acked-by: Paul Menage <menage@google.com>
Cc: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/cpuset.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 60d2c4702c6c..531b235e546f 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -610,8 +610,13 @@ void rebuild_sched_domains(void)
 	while (__kfifo_get(q, (void *)&cp, sizeof(cp))) {
 		struct cgroup *cont;
 		struct cpuset *child;   /* scans child cpusets of cp */
+
+		if (cpus_empty(cp->cpus_allowed))
+			continue;
+
 		if (is_sched_load_balance(cp))
 			csa[csn++] = cp;
+
 		list_for_each_entry(cont, &cp->css.cgroup->children, sibling) {
 			child = cgroup_cs(cont);
 			__kfifo_put(q, (void *)&child, sizeof(cp));

From 02412483777651a26b19a75e49c2a451a174ca9c Mon Sep 17 00:00:00 2001
From: Lai Jiangshan <laijs@cn.fujitsu.com>
Date: Fri, 25 Jul 2008 01:47:24 -0700
Subject: [PATCH 0809/1171] cpuset: code-cleanup for started_after

cgroup(cgroup_scan_tasks) will initialize heap->gt for us.  This patch
removes started_after() and its helper-function.

Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Acked-by: Paul Menage <menage@google.com>
Cc: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/cpuset.c | 37 ++++++-------------------------------
 1 file changed, 6 insertions(+), 31 deletions(-)

diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 531b235e546f..ebbc9b082e48 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -709,36 +709,6 @@ done:
 	/* Don't kfree(dattr) -- partition_sched_domains() does that. */
 }
 
-static inline int started_after_time(struct task_struct *t1,
-				     struct timespec *time,
-				     struct task_struct *t2)
-{
-	int start_diff = timespec_compare(&t1->start_time, time);
-	if (start_diff > 0) {
-		return 1;
-	} else if (start_diff < 0) {
-		return 0;
-	} else {
-		/*
-		 * Arbitrarily, if two processes started at the same
-		 * time, we'll say that the lower pointer value
-		 * started first. Note that t2 may have exited by now
-		 * so this may not be a valid pointer any longer, but
-		 * that's fine - it still serves to distinguish
-		 * between two tasks started (effectively)
-		 * simultaneously.
-		 */
-		return t1 > t2;
-	}
-}
-
-static inline int started_after(void *p1, void *p2)
-{
-	struct task_struct *t1 = p1;
-	struct task_struct *t2 = p2;
-	return started_after_time(t1, &t2->start_time, t2);
-}
-
 /**
  * cpuset_test_cpumask - test a task's cpus_allowed versus its cpuset's
  * @tsk: task to test
@@ -790,7 +760,12 @@ static int update_tasks_cpumask(struct cpuset *cs)
 	struct ptr_heap heap;
 	int retval;
 
-	retval = heap_init(&heap, PAGE_SIZE, GFP_KERNEL, &started_after);
+	/*
+	 * cgroup_scan_tasks() will initialize heap->gt for us.
+	 * heap_init() is still needed here for we should not change
+	 * cs->cpus_allowed when heap_init() fails.
+	 */
+	retval = heap_init(&heap, PAGE_SIZE, GFP_KERNEL, NULL);
 	if (retval)
 		return retval;
 

From da5ef6bb96158b0fc0d808704237a453af449124 Mon Sep 17 00:00:00 2001
From: Lai Jiangshan <laijs@cn.fujitsu.com>
Date: Fri, 25 Jul 2008 01:47:25 -0700
Subject: [PATCH 0810/1171] cpuset: two minor code-cleanups

In cpuset_update_task_memory_state() local variable struct task_struct
*tsk = current;

And local variable tsk is used 14 times and statement task_cs(tsk) is used
twice in this function.  So using task_cs(tsk) instead of task_cs(current)
is better for readability.

And "(struct cgroup_scanner *)&scan" is not good for readability also.
(and "container_of" is used in cpuset_do_move_task(), not
"(cpuset_hotplug_scanner *)scan")

Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Acked-by: Paul Menage <menage@google.com>
Cc: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/cpuset.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index ebbc9b082e48..91cf85b36dd5 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -365,7 +365,7 @@ void cpuset_update_task_memory_state(void)
 		my_cpusets_mem_gen = top_cpuset.mems_generation;
 	} else {
 		rcu_read_lock();
-		my_cpusets_mem_gen = task_cs(current)->mems_generation;
+		my_cpusets_mem_gen = task_cs(tsk)->mems_generation;
 		rcu_read_unlock();
 	}
 
@@ -1777,7 +1777,7 @@ static void move_member_tasks_to_cpuset(struct cpuset *from, struct cpuset *to)
 	scan.scan.heap = NULL;
 	scan.to = to->css.cgroup;
 
-	if (cgroup_scan_tasks((struct cgroup_scanner *)&scan))
+	if (cgroup_scan_tasks(&scan.scan))
 		printk(KERN_ERR "move_member_tasks_to_cpuset: "
 				"cgroup_scan_tasks failed\n");
 }

From 4b7a1304267bff68260ae861784b27130e805be3 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:26 -0700
Subject: [PATCH 0811/1171] posix timers: timer_delete: remove the bogus
 "->it_process != NULL" check

sys_timer_delete() and itimer_delete() check "timer->it_process != NULL",
this looks completely bogus.  ->it_process == NULL means that this timer
is already under destruction or it is not fully initialized, this must not
happen.

	sys_timer_delete: the timer is locked, and lock_timer() can't succeed
	if ->it_process == NULL.

	itimer_delete: it is called by exit_itimers() when there are no other
	threads which can play with signal_struct->posix_timers.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: Roland McGrath <roland@redhat.com>
Cc: john stultz <johnstul@us.ibm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/posix-timers.c | 18 ++++++++----------
 1 file changed, 8 insertions(+), 10 deletions(-)

diff --git a/kernel/posix-timers.c b/kernel/posix-timers.c
index dbd8398ddb0b..17f53266fb67 100644
--- a/kernel/posix-timers.c
+++ b/kernel/posix-timers.c
@@ -856,11 +856,10 @@ retry_delete:
 	 * This keeps any tasks waiting on the spin lock from thinking
 	 * they got something (see the lock code above).
 	 */
-	if (timer->it_process) {
-		if (timer->it_sigev_notify == (SIGEV_SIGNAL|SIGEV_THREAD_ID))
-			put_task_struct(timer->it_process);
-		timer->it_process = NULL;
-	}
+	if (timer->it_sigev_notify == (SIGEV_SIGNAL|SIGEV_THREAD_ID))
+		put_task_struct(timer->it_process);
+	timer->it_process = NULL;
+
 	unlock_timer(timer, flags);
 	release_posix_timer(timer, IT_ID_SET);
 	return 0;
@@ -885,11 +884,10 @@ retry_delete:
 	 * This keeps any tasks waiting on the spin lock from thinking
 	 * they got something (see the lock code above).
 	 */
-	if (timer->it_process) {
-		if (timer->it_sigev_notify == (SIGEV_SIGNAL|SIGEV_THREAD_ID))
-			put_task_struct(timer->it_process);
-		timer->it_process = NULL;
-	}
+	if (timer->it_sigev_notify == (SIGEV_SIGNAL|SIGEV_THREAD_ID))
+		put_task_struct(timer->it_process);
+	timer->it_process = NULL;
+
 	unlock_timer(timer, flags);
 	release_posix_timer(timer, IT_ID_SET);
 }

From 96347e7759e2e433c427defa0fa1adfc8cce6226 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:27 -0700
Subject: [PATCH 0812/1171] posix timers: release_posix_timer: kill the bogus
 put_task_struct(->it_process);

release_posix_timer() can't be called with ->it_process != NULL.  Once
sys_timer_create() sets ->it_process it must not call
release_posix_timer(), otherwise we can race with another thread doing
sys_timer_delete(), this timer is visible to idr_find() and unlocked.

The same is true for two other callers (actually, for any possible
caller), sys_timer_delete() and itimer_delete().  They must clear
->it_process before unlock_timer() + release_posix_timer().

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: Roland McGrath <roland@redhat.com>
Cc: john stultz <johnstul@us.ibm.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/posix-timers.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/kernel/posix-timers.c b/kernel/posix-timers.c
index 17f53266fb67..9a21681aa80f 100644
--- a/kernel/posix-timers.c
+++ b/kernel/posix-timers.c
@@ -449,9 +449,6 @@ static void release_posix_timer(struct k_itimer *tmr, int it_id_set)
 		spin_unlock_irqrestore(&idr_lock, flags);
 	}
 	sigqueue_free(tmr->sigq);
-	if (unlikely(tmr->it_process) &&
-	    tmr->it_sigev_notify == (SIGEV_SIGNAL|SIGEV_THREAD_ID))
-		put_task_struct(tmr->it_process);
 	kmem_cache_free(posix_timers_cache, tmr);
 }
 

From 6715ca451cfff1c9ce4b33ad9918a1dacf43997c Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:27 -0700
Subject: [PATCH 0813/1171] signals: collect_signal: remove the unneeded
 sigismember() check

collect_signal() checks sigismember(&list->signal, sig), this is not
needed.  This "sig" was just found by next_signal(), so it must be valid.

We have a (completely broken) call to ->notifier in between, but it must
not play with sigpending->signal bits or unlock ->siglock.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/signal.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/kernel/signal.c b/kernel/signal.c
index 6c0958e52ea7..c5b9aabb1550 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -343,9 +343,6 @@ static int collect_signal(int sig, struct sigpending *list, siginfo_t *info)
 	struct sigqueue *q, *first = NULL;
 	int still_pending = 0;
 
-	if (unlikely(!sigismember(&list->signal, sig)))
-		return 0;
-
 	/*
 	 * Collect the siginfo appropriate to this signal.  Check if
 	 * there is another siginfo for the same signal.

From d4434207616980885205c605697868c0f07e4378 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:28 -0700
Subject: [PATCH 0814/1171] signals: collect_signal: simplify the
 "still_pending" logic

Factor out sigdelset() calls and remove the "still_pending" variable.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/signal.c | 15 ++++++---------
 1 file changed, 6 insertions(+), 9 deletions(-)

diff --git a/kernel/signal.c b/kernel/signal.c
index c5b9aabb1550..50ad439377b2 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -341,7 +341,6 @@ unblock_all_signals(void)
 static int collect_signal(int sig, struct sigpending *list, siginfo_t *info)
 {
 	struct sigqueue *q, *first = NULL;
-	int still_pending = 0;
 
 	/*
 	 * Collect the siginfo appropriate to this signal.  Check if
@@ -349,26 +348,24 @@ static int collect_signal(int sig, struct sigpending *list, siginfo_t *info)
 	*/
 	list_for_each_entry(q, &list->list, list) {
 		if (q->info.si_signo == sig) {
-			if (first) {
-				still_pending = 1;
-				break;
-			}
+			if (first)
+				goto still_pending;
 			first = q;
 		}
 	}
+
+	sigdelset(&list->signal, sig);
+
 	if (first) {
+still_pending:
 		list_del_init(&first->list);
 		copy_siginfo(info, &first->info);
 		__sigqueue_free(first);
-		if (!still_pending)
-			sigdelset(&list->signal, sig);
 	} else {
-
 		/* Ok, it wasn't in the queue.  This must be
 		   a fast-pathed signal or we must have been
 		   out of queue space.  So zero out the info.
 		 */
-		sigdelset(&list->signal, sig);
 		info->si_signo = sig;
 		info->si_errno = 0;
 		info->si_code = 0;

From 100360f03077663b7bef3af44805b6cf700c3bee Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:29 -0700
Subject: [PATCH 0815/1171] signals: change collect_signal() to return void

With the recent changes collect_signal() always returns true.  Change it
to return void and update the single caller.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/signal.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/kernel/signal.c b/kernel/signal.c
index 50ad439377b2..fea236fe0b50 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -338,7 +338,7 @@ unblock_all_signals(void)
 	spin_unlock_irqrestore(&current->sighand->siglock, flags);
 }
 
-static int collect_signal(int sig, struct sigpending *list, siginfo_t *info)
+static void collect_signal(int sig, struct sigpending *list, siginfo_t *info)
 {
 	struct sigqueue *q, *first = NULL;
 
@@ -372,7 +372,6 @@ still_pending:
 		info->si_pid = 0;
 		info->si_uid = 0;
 	}
-	return 1;
 }
 
 static int __dequeue_signal(struct sigpending *pending, sigset_t *mask,
@@ -390,8 +389,7 @@ static int __dequeue_signal(struct sigpending *pending, sigset_t *mask,
 			}
 		}
 
-		if (!collect_signal(sig, pending, info))
-			sig = 0;
+		collect_signal(sig, pending, info);
 	}
 
 	return sig;

From 3854a771821c970065e3203a0b40ddc4101538cc Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:29 -0700
Subject: [PATCH 0816/1171] __exit_signal: don't take rcu lock

There is no reason for rcu_read_lock() in __exit_signal().  tsk->sighand
can only be changed if tsk does exec, obviously this is not possible.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/exit.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/kernel/exit.c b/kernel/exit.c
index 93d2711b9381..a7799d8a6404 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -85,7 +85,6 @@ static void __exit_signal(struct task_struct *tsk)
 	BUG_ON(!sig);
 	BUG_ON(!atomic_read(&sig->count));
 
-	rcu_read_lock();
 	sighand = rcu_dereference(tsk->sighand);
 	spin_lock(&sighand->siglock);
 
@@ -136,7 +135,6 @@ static void __exit_signal(struct task_struct *tsk)
 	tsk->signal = NULL;
 	tsk->sighand = NULL;
 	spin_unlock(&sighand->siglock);
-	rcu_read_unlock();
 
 	__cleanup_sighand(sighand);
 	clear_tsk_thread_flag(tsk,TIF_SIGPENDING);

From 92413d771e7123304fb4b9efd2a00cccc946e383 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:30 -0700
Subject: [PATCH 0817/1171] signals: dequeue_signal: don't check
 SIGNAL_GROUP_EXIT when setting SIGNAL_STOP_DEQUEUED

dequeue_signal() checks SIGNAL_GROUP_EXIT before setting
SIGNAL_STOP_DEQUEUED.  This was added by
788e05a67c343fa22f2ae1d3ca264e7f15c25eaf a long ago to avoid the
coredump/SIGSTOP race.

Since then the related code was changed, and now this subtle check is both
incomplete and unneeded at the same time.  It is incomplete because
nowadays exec() doesn't set SIGNAL_GROUP_EXIT, so in fact we should check
signal_group_exit() to avoid a similar race.  Fortunately, we doesn't need
the check at all.  The only function which relies on SIGNAL_STOP_DEQUEUED
is do_signal_stop(), and it ignores this flag if signal_group_exit() == T,
this covers the SIGNAL_GROUP_EXIT case.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/signal.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/kernel/signal.c b/kernel/signal.c
index fea236fe0b50..15f901a26ec7 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -454,8 +454,7 @@ int dequeue_signal(struct task_struct *tsk, sigset_t *mask, siginfo_t *info)
 		 * is to alert stop-signal processing code when another
 		 * processor has come along and cleared the flag.
 		 */
-		if (!(tsk->signal->flags & SIGNAL_GROUP_EXIT))
-			tsk->signal->flags |= SIGNAL_STOP_DEQUEUED;
+		tsk->signal->flags |= SIGNAL_STOP_DEQUEUED;
 	}
 	if ((info->si_code & __SI_MASK) == __SI_TIMER && info->si_sys_private) {
 		/*

From 2b201a9eddf509e8e935b45e573648e36f4b623f Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:31 -0700
Subject: [PATCH 0818/1171] signals: do_signal_stop: kill the SIGNAL_UNKILLABLE
 check

fae5fa44f1fd079ffbed8e0add929dd7bbd1347f changed do_signal_stop() to check
SIGNAL_UNKILLABLE, this wasn't needed.  If signal_group_exit() == F, the
signal sent to SIGNAL_UNKILLABLE task must be already filtered out by the
caller, get_signal_to_deliver().  And if signal_group_exit() == T we are
not going to stop.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/signal.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/kernel/signal.c b/kernel/signal.c
index 15f901a26ec7..0514da573f22 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -1649,8 +1649,7 @@ static int do_signal_stop(int signr)
 	} else {
 		struct task_struct *t;
 
-		if (unlikely((sig->flags & (SIGNAL_STOP_DEQUEUED | SIGNAL_UNKILLABLE))
-					 != SIGNAL_STOP_DEQUEUED) ||
+		if (!likely(sig->flags & SIGNAL_STOP_DEQUEUED) ||
 		    unlikely(signal_group_exit(sig)))
 			return 0;
 		/*

From e4901f92a8dbe843e76651a50f7a2a6dd3d53474 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:31 -0700
Subject: [PATCH 0819/1171] coredump: zap_threads: comments && use
 while_each_thread()

No changes in fs/exec.o

The for_each_process() loop in zap_threads() is very subtle, it is not
clear why we don't race with fork/exit/exec.  Add the fat comment.

Also, change the code to use while_each_thread().

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/exec.c | 39 ++++++++++++++++++++++++++++++++-------
 1 file changed, 32 insertions(+), 7 deletions(-)

diff --git a/fs/exec.c b/fs/exec.c
index e41aef0fb351..af249af4ccab 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1517,7 +1517,7 @@ static void zap_process(struct task_struct *start)
 			sigaddset(&t->pending.signal, SIGKILL);
 			signal_wake_up(t, 1);
 		}
-	} while ((t = next_thread(t)) != start);
+	} while_each_thread(start, t);
 }
 
 static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm,
@@ -1539,7 +1539,36 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm,
 
 	if (atomic_read(&mm->mm_users) == mm->core_waiters + 1)
 		goto done;
-
+	/*
+	 * We should find and kill all tasks which use this mm, and we should
+	 * count them correctly into mm->core_waiters. We don't take tasklist
+	 * lock, but this is safe wrt:
+	 *
+	 * fork:
+	 *	None of sub-threads can fork after zap_process(leader). All
+	 *	processes which were created before this point should be
+	 *	visible to zap_threads() because copy_process() adds the new
+	 *	process to the tail of init_task.tasks list, and lock/unlock
+	 *	of ->siglock provides a memory barrier.
+	 *
+	 * do_exit:
+	 *	The caller holds mm->mmap_sem. This means that the task which
+	 *	uses this mm can't pass exit_mm(), so it can't exit or clear
+	 *	its ->mm.
+	 *
+	 * de_thread:
+	 *	It does list_replace_rcu(&leader->tasks, &current->tasks),
+	 *	we must see either old or new leader, this does not matter.
+	 *	However, it can change p->sighand, so lock_task_sighand(p)
+	 *	must be used. Since p->mm != NULL and we hold ->mmap_sem
+	 *	it can't fail.
+	 *
+	 *	Note also that "g" can be the old leader with ->mm == NULL
+	 *	and already unhashed and thus removed from ->thread_group.
+	 *	This is OK, __unhash_process()->list_del_rcu() does not
+	 *	clear the ->next pointer, we will find the new leader via
+	 *	next_thread().
+	 */
 	rcu_read_lock();
 	for_each_process(g) {
 		if (g == tsk->group_leader)
@@ -1549,17 +1578,13 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm,
 		do {
 			if (p->mm) {
 				if (p->mm == mm) {
-					/*
-					 * p->sighand can't disappear, but
-					 * may be changed by de_thread()
-					 */
 					lock_task_sighand(p, &flags);
 					zap_process(p);
 					unlock_task_sighand(p, &flags);
 				}
 				break;
 			}
-		} while ((p = next_thread(p)) != g);
+		} while_each_thread(g, p);
 	}
 	rcu_read_unlock();
 done:

From d8878ba3f05ae5bbfad5a6e72e5121c0ea35f989 Mon Sep 17 00:00:00 2001
From: Michael Kerrisk <mtk.manpages@googlemail.com>
Date: Fri, 25 Jul 2008 01:47:32 -0700
Subject: [PATCH 0820/1171] signals: make siginfo_t si_utime + si_sstime report
 times in USER_HZ, not HZ

In the switch to configurable HZ in 2.6, the treatment of the si_utime and
si_stime fields that are exposed to userland via the siginfo structure
looks to have been botched.  As things stand, these fields report times in
units of HZ, so that userland gets information that varies depending on
the HZ that the kernel was configured with.  This patch changes the
reported values to use USER_HZ units.

Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
Acked-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/signal.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/kernel/signal.c b/kernel/signal.c
index 0514da573f22..ba60eeeb63aa 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -1370,10 +1370,9 @@ void do_notify_parent(struct task_struct *tsk, int sig)
 
 	info.si_uid = tsk->uid;
 
-	/* FIXME: find out whether or not this is supposed to be c*time. */
-	info.si_utime = cputime_to_jiffies(cputime_add(tsk->utime,
+	info.si_utime = cputime_to_clock_t(cputime_add(tsk->utime,
 						       tsk->signal->utime));
-	info.si_stime = cputime_to_jiffies(cputime_add(tsk->stime,
+	info.si_stime = cputime_to_clock_t(cputime_add(tsk->stime,
 						       tsk->signal->stime));
 
 	info.si_status = tsk->exit_code & 0x7f;
@@ -1441,9 +1440,8 @@ static void do_notify_parent_cldstop(struct task_struct *tsk, int why)
 
 	info.si_uid = tsk->uid;
 
-	/* FIXME: find out whether or not this is supposed to be c*time. */
-	info.si_utime = cputime_to_jiffies(tsk->utime);
-	info.si_stime = cputime_to_jiffies(tsk->stime);
+	info.si_utime = cputime_to_clock_t(tsk->utime);
+	info.si_stime = cputime_to_clock_t(tsk->stime);
 
  	info.si_code = why;
  	switch (why) {

From bc64efd220dcd4449aef8dd2564d73127b583b09 Mon Sep 17 00:00:00 2001
From: Gustavo Fernando Padovan <gustavo@las.ic.unicamp.br>
Date: Fri, 25 Jul 2008 01:47:33 -0700
Subject: [PATCH 0821/1171] kernel/signal.c: change vars pid and tgid types to
 pid_t

Change the type of pid and tgid variables from int to the POSIX type
pid_t.

Signed-off-by: Gustavo F. Padovan <gustavo@las.ic.unicamp.br>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/signal.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

diff --git a/kernel/signal.c b/kernel/signal.c
index ba60eeeb63aa..fdab7b363fa7 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -1116,7 +1116,7 @@ EXPORT_SYMBOL_GPL(kill_pid_info_as_uid);
  * is probably wrong.  Should make it like BSD or SYSV.
  */
 
-static int kill_something_info(int sig, struct siginfo *info, int pid)
+static int kill_something_info(int sig, struct siginfo *info, pid_t pid)
 {
 	int ret;
 
@@ -2184,7 +2184,7 @@ sys_rt_sigtimedwait(const sigset_t __user *uthese,
 }
 
 asmlinkage long
-sys_kill(int pid, int sig)
+sys_kill(pid_t pid, int sig)
 {
 	struct siginfo info;
 
@@ -2197,7 +2197,7 @@ sys_kill(int pid, int sig)
 	return kill_something_info(sig, &info, pid);
 }
 
-static int do_tkill(int tgid, int pid, int sig)
+static int do_tkill(pid_t tgid, pid_t pid, int sig)
 {
 	int error;
 	struct siginfo info;
@@ -2243,7 +2243,7 @@ static int do_tkill(int tgid, int pid, int sig)
  *  exists but it's not belonging to the target process anymore. This
  *  method solves the problem of threads exiting and PIDs getting reused.
  */
-asmlinkage long sys_tgkill(int tgid, int pid, int sig)
+asmlinkage long sys_tgkill(pid_t tgid, pid_t pid, int sig)
 {
 	/* This is only valid for single tasks */
 	if (pid <= 0 || tgid <= 0)
@@ -2256,7 +2256,7 @@ asmlinkage long sys_tgkill(int tgid, int pid, int sig)
  *  Send a signal to only one task, even if it's a CLONE_THREAD task.
  */
 asmlinkage long
-sys_tkill(int pid, int sig)
+sys_tkill(pid_t pid, int sig)
 {
 	/* This is only valid for single tasks */
 	if (pid <= 0)
@@ -2266,7 +2266,7 @@ sys_tkill(int pid, int sig)
 }
 
 asmlinkage long
-sys_rt_sigqueueinfo(int pid, int sig, siginfo_t __user *uinfo)
+sys_rt_sigqueueinfo(pid_t pid, int sig, siginfo_t __user *uinfo)
 {
 	siginfo_t info;
 

From f22ab814a24e654b1de24db0c5f8b57b5ab2026a Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 01:47:34 -0700
Subject: [PATCH 0822/1171] include/asm/ptrace.h userspace headers cleanup

This patch contains the following cleanups for the asm/ptrace.h
userspace headers:

- include/asm-generic/Kbuild.asm already lists ptrace.h, remove
  the superfluous listings in the Kbuild files of the following
  architectures:
  - cris
  - frv
  - powerpc
  - x86
- don't expose function prototypes and macros to userspace:
  - arm
  - blackfin
  - cris
  - mn10300
  - parisc
- remove #ifdef CONFIG_'s around #define's:
  - blackfin
  - m68knommu
- sh: AFAIK __SH5__ should work in both kernel and userspace,
      no need to leak CONFIG_SUPERH64 to userspace
- xtensa: cosmetical change to remove empty
            #ifndef __ASSEMBLY__ #else #endif
          from the userspace headers

Not changed by this patch is the fact that the following architectures
have a different struct pt_regs depending on CONFIG_ variables:
- h8300
- m68knommu
- mips

This does not work in userspace.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Cc: <linux-arch@vger.kernel.org>
Cc: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: Greg Ungerer <gerg@uclinux.org>
Acked-by: Paul Mundt <lethal@linux-sh.org>
Acked-by: Grant Grundler <grundler@parisc-linux.org>
Acked-by: Jesper Nilsson <jesper.nilsson@axis.com>
Acked-by: Chris Zankel <chris@zankel.net>
Acked-by: David Howells <dhowells@redhat.com>
Acked-by: Paul Mackerras <paulus@samba.org>
Acked-by: Russell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/asm-arm/ptrace.h           |  6 ++----
 include/asm-blackfin/ptrace.h      |  6 ++++--
 include/asm-cris/arch-v10/Kbuild   |  1 -
 include/asm-cris/arch-v10/ptrace.h |  4 ++++
 include/asm-cris/arch-v32/Kbuild   |  1 -
 include/asm-cris/arch-v32/ptrace.h |  4 ++++
 include/asm-cris/ptrace.h          |  4 +++-
 include/asm-frv/Kbuild             |  1 -
 include/asm-m68knommu/ptrace.h     |  2 --
 include/asm-mn10300/ptrace.h       |  8 ++++++--
 include/asm-parisc/ptrace.h        |  4 +++-
 include/asm-powerpc/Kbuild         |  1 -
 include/asm-sh/ptrace.h            |  2 +-
 include/asm-x86/Kbuild             |  1 -
 include/asm-xtensa/ptrace.h        | 10 +++++-----
 15 files changed, 32 insertions(+), 23 deletions(-)

diff --git a/include/asm-arm/ptrace.h b/include/asm-arm/ptrace.h
index 7aaa206cb54e..8382b7510f94 100644
--- a/include/asm-arm/ptrace.h
+++ b/include/asm-arm/ptrace.h
@@ -139,8 +139,6 @@ static inline int valid_user_regs(struct pt_regs *regs)
 	return 0;
 }
 
-#endif	/* __KERNEL__ */
-
 #define pc_pointer(v) \
 	((v) & ~PCMASK)
 
@@ -153,10 +151,10 @@ extern unsigned long profile_pc(struct pt_regs *regs);
 #define profile_pc(regs) instruction_pointer(regs)
 #endif
 
-#ifdef __KERNEL__
 #define predicate(x)		((x) & 0xf0000000)
 #define PREDICATE_ALWAYS	0xe0000000
-#endif
+
+#endif /* __KERNEL__ */
 
 #endif /* __ASSEMBLY__ */
 
diff --git a/include/asm-blackfin/ptrace.h b/include/asm-blackfin/ptrace.h
index b8346cd3a6f6..a45a80e54adc 100644
--- a/include/asm-blackfin/ptrace.h
+++ b/include/asm-blackfin/ptrace.h
@@ -83,14 +83,14 @@ struct pt_regs {
 #define PTRACE_GETREGS            12
 #define PTRACE_SETREGS            13	/* ptrace signal  */
 
-#ifdef CONFIG_BINFMT_ELF_FDPIC
 #define PTRACE_GETFDPIC           31
 #define PTRACE_GETFDPIC_EXEC      0
 #define PTRACE_GETFDPIC_INTERP    1
-#endif
 
 #define PS_S  (0x0002)
 
+#ifdef __KERNEL__
+
 /* user_mode returns true if only one bit is set in IPEND, other than the
    master interrupt enable.  */
 #define user_mode(regs) (!(((regs)->ipend & ~0x10) & (((regs)->ipend & ~0x10) - 1)))
@@ -98,6 +98,8 @@ struct pt_regs {
 #define profile_pc(regs) instruction_pointer(regs)
 extern void show_regs(struct pt_regs *);
 
+#endif  /*  __KERNEL__  */
+
 #endif				/* __ASSEMBLY__ */
 
 /*
diff --git a/include/asm-cris/arch-v10/Kbuild b/include/asm-cris/arch-v10/Kbuild
index 60e7e1b73cec..7a192e1290b1 100644
--- a/include/asm-cris/arch-v10/Kbuild
+++ b/include/asm-cris/arch-v10/Kbuild
@@ -1,4 +1,3 @@
-header-y += ptrace.h
 header-y += user.h
 header-y += svinto.h
 header-y += sv_addr_ag.h
diff --git a/include/asm-cris/arch-v10/ptrace.h b/include/asm-cris/arch-v10/ptrace.h
index fb14c5ee37f9..2f464eab3a51 100644
--- a/include/asm-cris/arch-v10/ptrace.h
+++ b/include/asm-cris/arch-v10/ptrace.h
@@ -106,10 +106,14 @@ struct switch_stack {
 	unsigned long return_ip; /* ip that _resume will return to */
 };
 
+#ifdef __KERNEL__
+
 /* bit 8 is user-mode flag */
 #define user_mode(regs) (((regs)->dccr & 0x100) != 0)
 #define instruction_pointer(regs) ((regs)->irp)
 #define profile_pc(regs) instruction_pointer(regs)
 extern void show_regs(struct pt_regs *);
 
+#endif  /*  __KERNEL__  */
+
 #endif
diff --git a/include/asm-cris/arch-v32/Kbuild b/include/asm-cris/arch-v32/Kbuild
index a0ec545e242e..35f2fc4f993e 100644
--- a/include/asm-cris/arch-v32/Kbuild
+++ b/include/asm-cris/arch-v32/Kbuild
@@ -1,3 +1,2 @@
-header-y += ptrace.h
 header-y += user.h
 header-y += cryptocop.h
diff --git a/include/asm-cris/arch-v32/ptrace.h b/include/asm-cris/arch-v32/ptrace.h
index 516cc7062d94..41f4e8662bc2 100644
--- a/include/asm-cris/arch-v32/ptrace.h
+++ b/include/asm-cris/arch-v32/ptrace.h
@@ -106,9 +106,13 @@ struct switch_stack {
 	unsigned long return_ip; /* ip that _resume will return to */
 };
 
+#ifdef __KERNEL__
+
 #define user_mode(regs) (((regs)->ccs & (1 << (U_CCS_BITNR + CCS_SHIFT))) != 0)
 #define instruction_pointer(regs) ((regs)->erp)
 extern void show_regs(struct pt_regs *);
 #define profile_pc(regs) instruction_pointer(regs)
 
+#endif  /*  __KERNEL__  */
+
 #endif
diff --git a/include/asm-cris/ptrace.h b/include/asm-cris/ptrace.h
index 1ec69a7ea836..d910925e3174 100644
--- a/include/asm-cris/ptrace.h
+++ b/include/asm-cris/ptrace.h
@@ -4,11 +4,13 @@
 #include <asm/arch/ptrace.h>
 
 #ifdef __KERNEL__
+
 /* Arbitrarily choose the same ptrace numbers as used by the Sparc code. */
 #define PTRACE_GETREGS            12
 #define PTRACE_SETREGS            13
-#endif
 
 #define profile_pc(regs) instruction_pointer(regs)
 
+#endif /* __KERNEL__ */
+
 #endif /* _CRIS_PTRACE_H */
diff --git a/include/asm-frv/Kbuild b/include/asm-frv/Kbuild
index bc3f12c5b7e0..0f8956def738 100644
--- a/include/asm-frv/Kbuild
+++ b/include/asm-frv/Kbuild
@@ -3,4 +3,3 @@ include include/asm-generic/Kbuild.asm
 header-y += registers.h
 
 unifdef-y += termios.h
-unifdef-y += ptrace.h
diff --git a/include/asm-m68knommu/ptrace.h b/include/asm-m68knommu/ptrace.h
index 47258e86e8c4..8c9194b98548 100644
--- a/include/asm-m68knommu/ptrace.h
+++ b/include/asm-m68knommu/ptrace.h
@@ -68,10 +68,8 @@ struct switch_stack {
 /* Arbitrarily choose the same ptrace numbers as used by the Sparc code. */
 #define PTRACE_GETREGS            12
 #define PTRACE_SETREGS            13
-#ifdef CONFIG_FPU
 #define PTRACE_GETFPREGS          14
 #define PTRACE_SETFPREGS          15
-#endif
 
 #ifdef __KERNEL__
 
diff --git a/include/asm-mn10300/ptrace.h b/include/asm-mn10300/ptrace.h
index b3684689fcce..7b06cc623d8b 100644
--- a/include/asm-mn10300/ptrace.h
+++ b/include/asm-mn10300/ptrace.h
@@ -88,12 +88,16 @@ extern struct pt_regs *__frame; /* current frame pointer */
 /* options set using PTRACE_SETOPTIONS */
 #define PTRACE_O_TRACESYSGOOD     0x00000001
 
-#if defined(__KERNEL__) && !defined(__ASSEMBLY__)
+#if defined(__KERNEL__)
+
+#if !defined(__ASSEMBLY__)
 #define user_mode(regs)			(((regs)->epsw & EPSW_nSL) == EPSW_nSL)
 #define instruction_pointer(regs)	((regs)->pc)
 extern void show_regs(struct pt_regs *);
-#endif
+#endif  /*  !__ASSEMBLY  */
 
 #define profile_pc(regs) ((regs)->pc)
 
+#endif  /*  __KERNEL__  */
+
 #endif /* _ASM_PTRACE_H */
diff --git a/include/asm-parisc/ptrace.h b/include/asm-parisc/ptrace.h
index 93f990e418f1..3e94c5d85ff5 100644
--- a/include/asm-parisc/ptrace.h
+++ b/include/asm-parisc/ptrace.h
@@ -33,7 +33,6 @@ struct pt_regs {
 	unsigned long ipsw;	/* CR22 */
 };
 
-#define task_regs(task) ((struct pt_regs *) ((char *)(task) + TASK_REGS))
 /*
  * The numbers chosen here are somewhat arbitrary but absolutely MUST
  * not overlap with any of the number assigned in <linux/ptrace.h>.
@@ -43,8 +42,11 @@ struct pt_regs {
  * since we have taken branch traps too)
  */
 #define PTRACE_SINGLEBLOCK	12	/* resume execution until next branch */
+
 #ifdef __KERNEL__
 
+#define task_regs(task) ((struct pt_regs *) ((char *)(task) + TASK_REGS))
+
 /* XXX should we use iaoq[1] or iaoq[0] ? */
 #define user_mode(regs)			(((regs)->iaoq[0] & 3) ? 1 : 0)
 #define user_space(regs)		(((regs)->iasq[1] != 0) ? 1 : 0)
diff --git a/include/asm-powerpc/Kbuild b/include/asm-powerpc/Kbuild
index 04ce8f8a2ee7..5ab7d7fe198c 100644
--- a/include/asm-powerpc/Kbuild
+++ b/include/asm-powerpc/Kbuild
@@ -29,7 +29,6 @@ unifdef-y += elf.h
 unifdef-y += nvram.h
 unifdef-y += param.h
 unifdef-y += posix_types.h
-unifdef-y += ptrace.h
 unifdef-y += seccomp.h
 unifdef-y += signal.h
 unifdef-y += spu_info.h
diff --git a/include/asm-sh/ptrace.h b/include/asm-sh/ptrace.h
index 8d6c92b3e770..7d36dc3bee69 100644
--- a/include/asm-sh/ptrace.h
+++ b/include/asm-sh/ptrace.h
@@ -5,7 +5,7 @@
  * Copyright (C) 1999, 2000  Niibe Yutaka
  *
  */
-#if defined(__SH5__) || defined(CONFIG_SUPERH64)
+#if defined(__SH5__)
 struct pt_regs {
 	unsigned long long pc;
 	unsigned long long sr;
diff --git a/include/asm-x86/Kbuild b/include/asm-x86/Kbuild
index 1e3554596f72..00473f7dd819 100644
--- a/include/asm-x86/Kbuild
+++ b/include/asm-x86/Kbuild
@@ -19,7 +19,6 @@ unifdef-y += msr.h
 unifdef-y += mtrr.h
 unifdef-y += posix_types_32.h
 unifdef-y += posix_types_64.h
-unifdef-y += ptrace.h
 unifdef-y += unistd_32.h
 unifdef-y += unistd_64.h
 unifdef-y += vm86.h
diff --git a/include/asm-xtensa/ptrace.h b/include/asm-xtensa/ptrace.h
index 422c73e26937..089b0db44816 100644
--- a/include/asm-xtensa/ptrace.h
+++ b/include/asm-xtensa/ptrace.h
@@ -73,10 +73,10 @@
 #define PTRACE_GETXTREGS	18
 #define PTRACE_SETXTREGS	19
 
-#ifndef __ASSEMBLY__
-
 #ifdef __KERNEL__
 
+#ifndef __ASSEMBLY__
+
 /*
  * This struct defines the way the registers are stored on the
  * kernel stack during a system call or other kernel entry.
@@ -122,14 +122,14 @@ extern void show_regs(struct pt_regs *);
 # ifndef CONFIG_SMP
 #  define profile_pc(regs) instruction_pointer(regs)
 # endif
-#endif /* __KERNEL__ */
 
 #else	/* __ASSEMBLY__ */
 
-#ifdef __KERNEL__
 # include <asm/asm-offsets.h>
 #define PT_REGS_OFFSET	  (KERNEL_STACK_SIZE - PT_USER_SIZE)
-#endif
 
 #endif	/* !__ASSEMBLY__ */
+
+#endif  /* __KERNEL__ */
+
 #endif	/* _XTENSA_PTRACE_H */

From 364d3c13c17f45da6d638011078d4c4d3070d719 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:36 -0700
Subject: [PATCH 0823/1171] ptrace: give more respect to SIGKILL

ptrace_stop() has some complicated checks to prevent the scheduling in the
TASK_TRACED state with the pending SIGKILL, but these checks are racy, and
they depend on arch_ptrace_stop_needed().

This patch assumes that the traced task should die asap if it was killed by
SIGKILL, in that case schedule()->signal_pending_state() has no reason to
ignore the TASK_WAKEKILL part of TASK_TRACED, and we can kill this nasty
special case.

Note: do_exit()->ptrace_notify() is special, the killed task can already
dequeue SIGKILL at this point. Another indication that fatal_signal_pending()
is not exactly right.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/sched.h | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 6aca4a16e377..79e749dbf81e 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -2054,9 +2054,6 @@ static inline int signal_pending_state(long state, struct task_struct *p)
 	if (!signal_pending(p))
 		return 0;
 
-	if (state & (__TASK_STOPPED | __TASK_TRACED))
-		return 0;
-
 	return (state & TASK_INTERRUPTIBLE) || __fatal_signal_pending(p);
 }
 

From 3d749b9e676b26584a47e75c235aa6f69d0697ae Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:37 -0700
Subject: [PATCH 0824/1171] ptrace: simplify ptrace_stop()->sigkill_pending()
 path

1. SIGKILL can't be blocked, remove this check from sigkill_pending().

2. When ptrace_stop() sees sigkill_pending() == T, it can just return.
   Kill "int killed" and simplify the code. This also is more correct,
   the tracer shouldn't see us in TASK_TRACED if we are not going to
   stop.

I strongly believe this code needs further changes.  We should do the "was
this task killed" check unconditionally, currently it depends on
arch_ptrace_stop_needed().  On the other hand, sigkill_pending() isn't
very clever.  If the task was killed tkill(SIGKILL), the signal can be
already dequeued if the caller is do_exit().

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/signal.c | 12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

diff --git a/kernel/signal.c b/kernel/signal.c
index fdab7b363fa7..39c1706edf03 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -1496,9 +1496,8 @@ static inline int may_ptrace_stop(void)
  */
 static int sigkill_pending(struct task_struct *tsk)
 {
-	return ((sigismember(&tsk->pending.signal, SIGKILL) ||
-		 sigismember(&tsk->signal->shared_pending.signal, SIGKILL)) &&
-		!unlikely(sigismember(&tsk->blocked, SIGKILL)));
+	return	sigismember(&tsk->pending.signal, SIGKILL) ||
+		sigismember(&tsk->signal->shared_pending.signal, SIGKILL);
 }
 
 /*
@@ -1514,8 +1513,6 @@ static int sigkill_pending(struct task_struct *tsk)
  */
 static void ptrace_stop(int exit_code, int clear_code, siginfo_t *info)
 {
-	int killed = 0;
-
 	if (arch_ptrace_stop_needed(exit_code, info)) {
 		/*
 		 * The arch code has something special to do before a
@@ -1531,7 +1528,8 @@ static void ptrace_stop(int exit_code, int clear_code, siginfo_t *info)
 		spin_unlock_irq(&current->sighand->siglock);
 		arch_ptrace_stop(exit_code, info);
 		spin_lock_irq(&current->sighand->siglock);
-		killed = sigkill_pending(current);
+		if (sigkill_pending(current))
+			return;
 	}
 
 	/*
@@ -1548,7 +1546,7 @@ static void ptrace_stop(int exit_code, int clear_code, siginfo_t *info)
 	__set_current_state(TASK_TRACED);
 	spin_unlock_irq(&current->sighand->siglock);
 	read_lock(&tasklist_lock);
-	if (!unlikely(killed) && may_ptrace_stop()) {
+	if (may_ptrace_stop()) {
 		do_notify_parent_cldstop(current, CLD_TRAPPED);
 		read_unlock(&tasklist_lock);
 		schedule();

From 7b34e4283c685f5cc6ba6d30e939906eee0d4bcf Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:37 -0700
Subject: [PATCH 0825/1171] introduce PF_KTHREAD flag

Introduce the new PF_KTHREAD flag to mark the kernel threads.  It is set
by INIT_TASK() and copied to the forked childs (we could set it in
kthreadd() along with PF_NOFREEZE instead).

daemonize() was changed as well.  In that case testing of PF_KTHREAD is
racy, but daemonize() is hopeless anyway.

This flag is cleared in do_execve(), before search_binary_handler().
Probably not the best place, we can do this in exec_mmap() or in
start_thread(), or clear it along with PF_FORKNOEXEC.  But I think this
doesn't matter in practice, and if do_execve() fails kthread should die
soon.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/exec.c                 | 1 +
 include/linux/init_task.h | 2 +-
 include/linux/sched.h     | 1 +
 kernel/exit.c             | 2 +-
 4 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/fs/exec.c b/fs/exec.c
index af249af4ccab..cd2e8c9b1249 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1326,6 +1326,7 @@ int do_execve(char * filename,
 	if (retval < 0)
 		goto out;
 
+	current->flags &= ~PF_KTHREAD;
 	retval = search_binary_handler(bprm,regs);
 	if (retval >= 0) {
 		/* execve success */
diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index 93c45acf249a..021d8e720c79 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -122,7 +122,7 @@ extern struct group_info init_groups;
 	.state		= 0,						\
 	.stack		= &init_thread_info,				\
 	.usage		= ATOMIC_INIT(2),				\
-	.flags		= 0,						\
+	.flags		= PF_KTHREAD,					\
 	.lock_depth	= -1,						\
 	.prio		= MAX_PRIO-20,					\
 	.static_prio	= MAX_PRIO-20,					\
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 79e749dbf81e..eec64a4adb9d 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1483,6 +1483,7 @@ static inline void put_task_struct(struct task_struct *t)
 #define PF_EXITING	0x00000004	/* getting shut down */
 #define PF_EXITPIDONE	0x00000008	/* pi exit done on shut down */
 #define PF_VCPU		0x00000010	/* I'm a virtual CPU */
+#define PF_KTHREAD	0x00000020	/* I am a kernel thread */
 #define PF_FORKNOEXEC	0x00000040	/* forked but didn't exec */
 #define PF_SUPERPRIV	0x00000100	/* used super-user privileges */
 #define PF_DUMPCORE	0x00000200	/* dumped core */
diff --git a/kernel/exit.c b/kernel/exit.c
index a7799d8a6404..28a44a2612dc 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -430,7 +430,7 @@ void daemonize(const char *name, ...)
 	 * We don't want to have TIF_FREEZE set if the system-wide hibernation
 	 * or suspend transition begins right now.
 	 */
-	current->flags |= PF_NOFREEZE;
+	current->flags |= (PF_NOFREEZE | PF_KTHREAD);
 
 	if (current->nsproxy != &init_nsproxy) {
 		get_nsproxy(&init_nsproxy);

From 246bb0b1deb29726990620d8b5e55ca29f331362 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:38 -0700
Subject: [PATCH 0826/1171] kill PF_BORROWED_MM in favour of PF_KTHREAD

Kill PF_BORROWED_MM.  Change use_mm/unuse_mm to not play with ->flags, and
do s/PF_BORROWED_MM/PF_KTHREAD/ for a couple of other users.

No functional changes yet.  But this allows us to do further
fixes/cleanups.

oom_kill/ptrace/etc often check "p->mm != NULL" to filter out the
kthreads, this is wrong because of use_mm().  The problem with
PF_BORROWED_MM is that we need task_lock() to avoid races.  With this
patch we can check PF_KTHREAD directly, or use a simple lockless helper:

	/* The result must not be dereferenced !!! */
	struct mm_struct *__get_task_mm(struct task_struct *tsk)
	{
		if (tsk->flags & PF_KTHREAD)
			return NULL;
		return tsk->mm;
	}

Note also ecard_task().  It runs with ->mm != NULL, but it's the kernel
thread without PF_BORROWED_MM.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/aio.c              | 2 --
 include/linux/sched.h | 3 +--
 kernel/fork.c         | 4 ++--
 3 files changed, 3 insertions(+), 6 deletions(-)

diff --git a/fs/aio.c b/fs/aio.c
index 0fb3117ddd93..0051fd94b44e 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -586,7 +586,6 @@ static void use_mm(struct mm_struct *mm)
 	struct task_struct *tsk = current;
 
 	task_lock(tsk);
-	tsk->flags |= PF_BORROWED_MM;
 	active_mm = tsk->active_mm;
 	atomic_inc(&mm->mm_count);
 	tsk->mm = mm;
@@ -610,7 +609,6 @@ static void unuse_mm(struct mm_struct *mm)
 	struct task_struct *tsk = current;
 
 	task_lock(tsk);
-	tsk->flags &= ~PF_BORROWED_MM;
 	tsk->mm = NULL;
 	/* active_mm is still 'mm' */
 	enter_lazy_tlb(mm, tsk);
diff --git a/include/linux/sched.h b/include/linux/sched.h
index eec64a4adb9d..0560999eb1db 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1483,7 +1483,6 @@ static inline void put_task_struct(struct task_struct *t)
 #define PF_EXITING	0x00000004	/* getting shut down */
 #define PF_EXITPIDONE	0x00000008	/* pi exit done on shut down */
 #define PF_VCPU		0x00000010	/* I'm a virtual CPU */
-#define PF_KTHREAD	0x00000020	/* I am a kernel thread */
 #define PF_FORKNOEXEC	0x00000040	/* forked but didn't exec */
 #define PF_SUPERPRIV	0x00000100	/* used super-user privileges */
 #define PF_DUMPCORE	0x00000200	/* dumped core */
@@ -1497,7 +1496,7 @@ static inline void put_task_struct(struct task_struct *t)
 #define PF_KSWAPD	0x00040000	/* I am kswapd */
 #define PF_SWAPOFF	0x00080000	/* I am in swapoff */
 #define PF_LESS_THROTTLE 0x00100000	/* Throttle me less: I clean memory */
-#define PF_BORROWED_MM	0x00200000	/* I am a kthread doing use_mm */
+#define PF_KTHREAD	0x00200000	/* I am a kernel thread */
 #define PF_RANDOMIZE	0x00400000	/* randomize virtual address space */
 #define PF_SWAPWRITE	0x00800000	/* Allowed to write to swap */
 #define PF_SPREAD_PAGE	0x01000000	/* Spread page cache over cpuset */
diff --git a/kernel/fork.c b/kernel/fork.c
index 228f80c9155a..eeaec6893b0d 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -474,7 +474,7 @@ EXPORT_SYMBOL_GPL(mmput);
 /**
  * get_task_mm - acquire a reference to the task's mm
  *
- * Returns %NULL if the task has no mm.  Checks PF_BORROWED_MM (meaning
+ * Returns %NULL if the task has no mm.  Checks PF_KTHREAD (meaning
  * this kernel workthread has transiently adopted a user mm with use_mm,
  * to do its AIO) is not set and if so returns a reference to it, after
  * bumping up the use count.  User must release the mm via mmput()
@@ -487,7 +487,7 @@ struct mm_struct *get_task_mm(struct task_struct *task)
 	task_lock(task);
 	mm = task->mm;
 	if (mm) {
-		if (task->flags & PF_BORROWED_MM)
+		if (task->flags & PF_KTHREAD)
 			mm = NULL;
 		else
 			atomic_inc(&mm->mm_users);

From 15b9f360c0316c06d37c09b02d85565edbaf9dd3 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:39 -0700
Subject: [PATCH 0827/1171] coredump: zap_threads() must skip kernel threads

The main loop in zap_threads() must skip kthreads which may use the same
mm.  Otherwise we "kill" this thread erroneously (for example, it can not
fork or exec after that), and the coredumping task stucks in the
TASK_UNINTERRUPTIBLE state forever because of the wrong ->core_waiters
count.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/exec.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/fs/exec.c b/fs/exec.c
index cd2e8c9b1249..e347e6ed1617 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1574,11 +1574,12 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm,
 	for_each_process(g) {
 		if (g == tsk->group_leader)
 			continue;
-
+		if (g->flags & PF_KTHREAD)
+			continue;
 		p = g;
 		do {
 			if (p->mm) {
-				if (p->mm == mm) {
+				if (unlikely(p->mm == mm)) {
 					lock_task_sighand(p, &flags);
 					zap_process(p);
 					unlock_task_sighand(p, &flags);

From 24d5288f06ed8b3a368eba967d587cdb012dfdf7 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:40 -0700
Subject: [PATCH 0828/1171] coredump: elf_core_dump: skip kernel threads

linux_binfmt->core_dump() runs before the process does exit_aio(), this
means that we can hit the kernel thread which shares the same ->mm.
Afaics, nothing really bad can happen, but perhaps it makes sense to fix
this minor bug.

It is sad we have to iterate over all threads in system and use
GFP_ATOMIC.  Hopefully we can kill theses ugly do_each_thread()s, but this
needs some nontrivial changes in mm_struct and do_coredump.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/binfmt_elf.c       | 6 ++++++
 fs/binfmt_elf_fdpic.c | 3 +++
 2 files changed, 9 insertions(+)

diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index 639d2d8b5710..bad7d8770d72 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -1520,6 +1520,9 @@ static int fill_note_info(struct elfhdr *elf, int phdrs,
 	rcu_read_lock();
 	do_each_thread(g, p)
 		if (p->mm == dump_task->mm) {
+			if (p->flags & PF_KTHREAD)
+				continue;
+
 			t = kzalloc(offsetof(struct elf_thread_core_info,
 					     notes[info->thread_notes]),
 				    GFP_ATOMIC);
@@ -1724,6 +1727,9 @@ static int fill_note_info(struct elfhdr *elf, int phdrs,
 		rcu_read_lock();
 		do_each_thread(g, p)
 			if (current->mm == p->mm && current != p) {
+				if (p->flags & PF_KTHREAD)
+					continue;
+
 				ets = kzalloc(sizeof(*ets), GFP_ATOMIC);
 				if (!ets) {
 					rcu_read_unlock();
diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c
index d051a32e6270..71bcc4b4d088 100644
--- a/fs/binfmt_elf_fdpic.c
+++ b/fs/binfmt_elf_fdpic.c
@@ -1626,6 +1626,9 @@ static int elf_fdpic_core_dump(long signr, struct pt_regs *regs,
 		rcu_read_lock();
 		do_each_thread(g,p)
 			if (current->mm == p->mm && current != p) {
+				if (p->flags & PF_KTHREAD)
+					continue;
+
 				tmp = kzalloc(sizeof(*tmp), GFP_ATOMIC);
 				if (!tmp) {
 					rcu_read_unlock();

From 32ecb1f26dd50eeaac4e3f4dea4541c97848e459 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:41 -0700
Subject: [PATCH 0829/1171] coredump: turn mm->core_startup_done into the
 pointer to struct core_state

mm->core_startup_done points to "struct completion startup_done" allocated
on the coredump_wait()'s stack.  Introduce the new structure, core_state,
which holds this "struct completion".  This way we can add more info
visible to the threads participating in coredump without enlarging
mm_struct.

No changes in affected .o files.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/exec.c                | 8 ++++----
 include/linux/mm_types.h | 7 ++++++-
 kernel/exit.c            | 2 +-
 3 files changed, 11 insertions(+), 6 deletions(-)

diff --git a/fs/exec.c b/fs/exec.c
index e347e6ed1617..71734568f018 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1597,13 +1597,13 @@ static int coredump_wait(int exit_code)
 {
 	struct task_struct *tsk = current;
 	struct mm_struct *mm = tsk->mm;
-	struct completion startup_done;
+	struct core_state core_state;
 	struct completion *vfork_done;
 	int core_waiters;
 
 	init_completion(&mm->core_done);
-	init_completion(&startup_done);
-	mm->core_startup_done = &startup_done;
+	init_completion(&core_state.startup);
+	mm->core_state = &core_state;
 
 	core_waiters = zap_threads(tsk, mm, exit_code);
 	up_write(&mm->mmap_sem);
@@ -1622,7 +1622,7 @@ static int coredump_wait(int exit_code)
 	}
 
 	if (core_waiters)
-		wait_for_completion(&startup_done);
+		wait_for_completion(&core_state.startup);
 fail:
 	BUG_ON(mm->core_waiters);
 	return core_waiters;
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index 02a27ae78539..97819efd2333 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -159,6 +159,10 @@ struct vm_area_struct {
 #endif
 };
 
+struct core_state {
+	struct completion startup;
+};
+
 struct mm_struct {
 	struct vm_area_struct * mmap;		/* list of VMAs */
 	struct rb_root mm_rb;
@@ -220,7 +224,8 @@ struct mm_struct {
 	unsigned long flags; /* Must use atomic bitops to access the bits */
 
 	/* coredumping support */
-	struct completion *core_startup_done, core_done;
+	struct core_state *core_state;
+	struct completion core_done;
 
 	/* aio bits */
 	rwlock_t		ioctx_list_lock;	/* aio lock */
diff --git a/kernel/exit.c b/kernel/exit.c
index 28a44a2612dc..f7fa21dbced4 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -680,7 +680,7 @@ static void exit_mm(struct task_struct * tsk)
 		up_read(&mm->mmap_sem);
 		down_write(&mm->mmap_sem);
 		if (!--mm->core_waiters)
-			complete(mm->core_startup_done);
+			complete(&mm->core_state->startup);
 		up_write(&mm->mmap_sem);
 
 		wait_for_completion(&mm->core_done);

From 999d9fc1670bc082928b93b11d1f2e0e417d973c Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:41 -0700
Subject: [PATCH 0830/1171] coredump: move mm->core_waiters into struct
 core_state

Move mm->core_waiters into "struct core_state" allocated on stack.  This
shrinks mm_struct a little bit and allows further changes.

This patch mostly does s/core_waiters/core_state.  The only essential
change is that coredump_wait() must clear mm->core_state before return.

The coredump_wait()'s path is uglified and .text grows by 30 bytes, this
is fixed by the next patch.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/exec.c                | 21 +++++++++++----------
 include/linux/mm_types.h |  2 +-
 kernel/exit.c            |  8 ++++----
 kernel/fork.c            |  2 +-
 kernel/signal.c          |  4 ++--
 5 files changed, 19 insertions(+), 18 deletions(-)

diff --git a/fs/exec.c b/fs/exec.c
index 71734568f018..50de3aaff4d0 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -722,12 +722,10 @@ static int exec_mmap(struct mm_struct *mm)
 		 * Make sure that if there is a core dump in progress
 		 * for the old mm, we get out and die instead of going
 		 * through with the exec.  We must hold mmap_sem around
-		 * checking core_waiters and changing tsk->mm.  The
-		 * core-inducing thread will increment core_waiters for
-		 * each thread whose ->mm == old_mm.
+		 * checking core_state and changing tsk->mm.
 		 */
 		down_read(&old_mm->mmap_sem);
-		if (unlikely(old_mm->core_waiters)) {
+		if (unlikely(old_mm->core_state)) {
 			up_read(&old_mm->mmap_sem);
 			return -EINTR;
 		}
@@ -1514,7 +1512,7 @@ static void zap_process(struct task_struct *start)
 	t = start;
 	do {
 		if (t != current && t->mm) {
-			t->mm->core_waiters++;
+			t->mm->core_state->nr_threads++;
 			sigaddset(&t->pending.signal, SIGKILL);
 			signal_wake_up(t, 1);
 		}
@@ -1538,11 +1536,11 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm,
 	if (err)
 		return err;
 
-	if (atomic_read(&mm->mm_users) == mm->core_waiters + 1)
+	if (atomic_read(&mm->mm_users) == mm->core_state->nr_threads + 1)
 		goto done;
 	/*
 	 * We should find and kill all tasks which use this mm, and we should
-	 * count them correctly into mm->core_waiters. We don't take tasklist
+	 * count them correctly into ->nr_threads. We don't take tasklist
 	 * lock, but this is safe wrt:
 	 *
 	 * fork:
@@ -1590,7 +1588,7 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm,
 	}
 	rcu_read_unlock();
 done:
-	return mm->core_waiters;
+	return mm->core_state->nr_threads;
 }
 
 static int coredump_wait(int exit_code)
@@ -1603,9 +1601,12 @@ static int coredump_wait(int exit_code)
 
 	init_completion(&mm->core_done);
 	init_completion(&core_state.startup);
+	core_state.nr_threads = 0;
 	mm->core_state = &core_state;
 
 	core_waiters = zap_threads(tsk, mm, exit_code);
+	if (core_waiters < 0)
+		mm->core_state = NULL;
 	up_write(&mm->mmap_sem);
 
 	if (unlikely(core_waiters < 0))
@@ -1623,8 +1624,8 @@ static int coredump_wait(int exit_code)
 
 	if (core_waiters)
 		wait_for_completion(&core_state.startup);
+	mm->core_state = NULL;
 fail:
-	BUG_ON(mm->core_waiters);
 	return core_waiters;
 }
 
@@ -1702,7 +1703,7 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs)
 	/*
 	 * If another thread got here first, or we are not dumpable, bail out.
 	 */
-	if (mm->core_waiters || !get_dumpable(mm)) {
+	if (mm->core_state || !get_dumpable(mm)) {
 		up_write(&mm->mmap_sem);
 		goto fail;
 	}
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index 97819efd2333..c0b1747b61a5 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -160,6 +160,7 @@ struct vm_area_struct {
 };
 
 struct core_state {
+	int nr_threads;
 	struct completion startup;
 };
 
@@ -179,7 +180,6 @@ struct mm_struct {
 	atomic_t mm_users;			/* How many users with user space? */
 	atomic_t mm_count;			/* How many references to "struct mm_struct" (users count as 1) */
 	int map_count;				/* number of VMAs */
-	int core_waiters;
 	struct rw_semaphore mmap_sem;
 	spinlock_t page_table_lock;		/* Protects page tables and some counters */
 
diff --git a/kernel/exit.c b/kernel/exit.c
index f7fa21dbced4..988e232254e9 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -670,16 +670,16 @@ static void exit_mm(struct task_struct * tsk)
 		return;
 	/*
 	 * Serialize with any possible pending coredump.
-	 * We must hold mmap_sem around checking core_waiters
+	 * We must hold mmap_sem around checking core_state
 	 * and clearing tsk->mm.  The core-inducing thread
-	 * will increment core_waiters for each thread in the
+	 * will increment ->nr_threads for each thread in the
 	 * group with ->mm != NULL.
 	 */
 	down_read(&mm->mmap_sem);
-	if (mm->core_waiters) {
+	if (mm->core_state) {
 		up_read(&mm->mmap_sem);
 		down_write(&mm->mmap_sem);
-		if (!--mm->core_waiters)
+		if (!--mm->core_state->nr_threads)
 			complete(&mm->core_state->startup);
 		up_write(&mm->mmap_sem);
 
diff --git a/kernel/fork.c b/kernel/fork.c
index eeaec6893b0d..813d5c89b9d5 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -400,7 +400,7 @@ static struct mm_struct * mm_init(struct mm_struct * mm, struct task_struct *p)
 	INIT_LIST_HEAD(&mm->mmlist);
 	mm->flags = (current->mm) ? current->mm->flags
 				  : MMF_DUMP_FILTER_DEFAULT;
-	mm->core_waiters = 0;
+	mm->core_state = NULL;
 	mm->nr_ptes = 0;
 	set_mm_counter(mm, file_rss, 0);
 	set_mm_counter(mm, anon_rss, 0);
diff --git a/kernel/signal.c b/kernel/signal.c
index 39c1706edf03..5c7b7eaa0dc6 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -1480,10 +1480,10 @@ static inline int may_ptrace_stop(void)
 	 * is a deadlock situation, and pointless because our tracer
 	 * is dead so don't allow us to stop.
 	 * If SIGKILL was already sent before the caller unlocked
-	 * ->siglock we must see ->core_waiters != 0. Otherwise it
+	 * ->siglock we must see ->core_state != NULL. Otherwise it
 	 * is safe to enter schedule().
 	 */
-	if (unlikely(current->mm->core_waiters) &&
+	if (unlikely(current->mm->core_state) &&
 	    unlikely(current->mm == current->parent->mm))
 		return 0;
 

From 8cd9c249128a59e8e833d454a784b0cbd338d468 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:42 -0700
Subject: [PATCH 0831/1171] coredump: simplify core_state->nr_threads
 calculation

Change zap_process() to return int instead of incrementing
mm->core_state->nr_threads directly.  Change zap_threads() to set
mm->core_state only on success.

This patch restores the original size of .text, and more importantly now
->nr_threads is used in two places only.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/exec.c | 33 ++++++++++++++++-----------------
 1 file changed, 16 insertions(+), 17 deletions(-)

diff --git a/fs/exec.c b/fs/exec.c
index 50de3aaff4d0..c74bb34eeeff 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1502,9 +1502,10 @@ out:
 	return ispipe;
 }
 
-static void zap_process(struct task_struct *start)
+static int zap_process(struct task_struct *start)
 {
 	struct task_struct *t;
+	int nr = 0;
 
 	start->signal->flags = SIGNAL_GROUP_EXIT;
 	start->signal->group_stop_count = 0;
@@ -1512,31 +1513,33 @@ static void zap_process(struct task_struct *start)
 	t = start;
 	do {
 		if (t != current && t->mm) {
-			t->mm->core_state->nr_threads++;
 			sigaddset(&t->pending.signal, SIGKILL);
 			signal_wake_up(t, 1);
+			nr++;
 		}
 	} while_each_thread(start, t);
+
+	return nr;
 }
 
 static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm,
-				int exit_code)
+				struct core_state *core_state, int exit_code)
 {
 	struct task_struct *g, *p;
 	unsigned long flags;
-	int err = -EAGAIN;
+	int nr = -EAGAIN;
 
 	spin_lock_irq(&tsk->sighand->siglock);
 	if (!signal_group_exit(tsk->signal)) {
+		mm->core_state = core_state;
 		tsk->signal->group_exit_code = exit_code;
-		zap_process(tsk);
-		err = 0;
+		nr = zap_process(tsk);
 	}
 	spin_unlock_irq(&tsk->sighand->siglock);
-	if (err)
-		return err;
+	if (unlikely(nr < 0))
+		return nr;
 
-	if (atomic_read(&mm->mm_users) == mm->core_state->nr_threads + 1)
+	if (atomic_read(&mm->mm_users) == nr + 1)
 		goto done;
 	/*
 	 * We should find and kill all tasks which use this mm, and we should
@@ -1579,7 +1582,7 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm,
 			if (p->mm) {
 				if (unlikely(p->mm == mm)) {
 					lock_task_sighand(p, &flags);
-					zap_process(p);
+					nr += zap_process(p);
 					unlock_task_sighand(p, &flags);
 				}
 				break;
@@ -1588,7 +1591,8 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm,
 	}
 	rcu_read_unlock();
 done:
-	return mm->core_state->nr_threads;
+	core_state->nr_threads = nr;
+	return nr;
 }
 
 static int coredump_wait(int exit_code)
@@ -1601,12 +1605,7 @@ static int coredump_wait(int exit_code)
 
 	init_completion(&mm->core_done);
 	init_completion(&core_state.startup);
-	core_state.nr_threads = 0;
-	mm->core_state = &core_state;
-
-	core_waiters = zap_threads(tsk, mm, exit_code);
-	if (core_waiters < 0)
-		mm->core_state = NULL;
+	core_waiters = zap_threads(tsk, mm, &core_state, exit_code);
 	up_write(&mm->mmap_sem);
 
 	if (unlikely(core_waiters < 0))

From c5f1cc8c1828486a61ab3e575da6e2c62b34d399 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:42 -0700
Subject: [PATCH 0832/1171] coredump: turn core_state->nr_threads into atomic_t

Turn core_state->nr_threads into atomic_t and kill now unneeded
down_write(&mm->mmap_sem) in exit_mm().

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/exec.c                | 2 +-
 include/linux/mm_types.h | 2 +-
 kernel/exit.c            | 5 ++---
 3 files changed, 4 insertions(+), 5 deletions(-)

diff --git a/fs/exec.c b/fs/exec.c
index c74bb34eeeff..15d493fe8aa3 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1591,7 +1591,7 @@ static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm,
 	}
 	rcu_read_unlock();
 done:
-	core_state->nr_threads = nr;
+	atomic_set(&core_state->nr_threads, nr);
 	return nr;
 }
 
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index c0b1747b61a5..ae99a28ba6ae 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -160,7 +160,7 @@ struct vm_area_struct {
 };
 
 struct core_state {
-	int nr_threads;
+	atomic_t nr_threads;
 	struct completion startup;
 };
 
diff --git a/kernel/exit.c b/kernel/exit.c
index 988e232254e9..63d82957baae 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -678,10 +678,9 @@ static void exit_mm(struct task_struct * tsk)
 	down_read(&mm->mmap_sem);
 	if (mm->core_state) {
 		up_read(&mm->mmap_sem);
-		down_write(&mm->mmap_sem);
-		if (!--mm->core_state->nr_threads)
+
+		if (atomic_dec_and_test(&mm->core_state->nr_threads))
 			complete(&mm->core_state->startup);
-		up_write(&mm->mmap_sem);
 
 		wait_for_completion(&mm->core_done);
 		down_read(&mm->mmap_sem);

From 9d5b327bf198d2720666de958dcc2ae219d86952 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:43 -0700
Subject: [PATCH 0833/1171] coredump: make mm->core_state visible to
 ->core_dump()

Move the "struct core_state core_state" from coredump_wait() to
do_coredump(), this makes mm->core_state visible to binfmt->core_dump().

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/exec.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/fs/exec.c b/fs/exec.c
index 15d493fe8aa3..b8ee842d93cd 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1595,17 +1595,16 @@ done:
 	return nr;
 }
 
-static int coredump_wait(int exit_code)
+static int coredump_wait(int exit_code, struct core_state *core_state)
 {
 	struct task_struct *tsk = current;
 	struct mm_struct *mm = tsk->mm;
-	struct core_state core_state;
 	struct completion *vfork_done;
 	int core_waiters;
 
 	init_completion(&mm->core_done);
-	init_completion(&core_state.startup);
-	core_waiters = zap_threads(tsk, mm, &core_state, exit_code);
+	init_completion(&core_state->startup);
+	core_waiters = zap_threads(tsk, mm, core_state, exit_code);
 	up_write(&mm->mmap_sem);
 
 	if (unlikely(core_waiters < 0))
@@ -1622,8 +1621,7 @@ static int coredump_wait(int exit_code)
 	}
 
 	if (core_waiters)
-		wait_for_completion(&core_state.startup);
-	mm->core_state = NULL;
+		wait_for_completion(&core_state->startup);
 fail:
 	return core_waiters;
 }
@@ -1679,6 +1677,7 @@ int get_dumpable(struct mm_struct *mm)
 
 int do_coredump(long signr, int exit_code, struct pt_regs * regs)
 {
+	struct core_state core_state;
 	char corename[CORENAME_MAX_SIZE + 1];
 	struct mm_struct *mm = current->mm;
 	struct linux_binfmt * binfmt;
@@ -1717,7 +1716,7 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs)
 		current->fsuid = 0;	/* Dump root private */
 	}
 
-	retval = coredump_wait(exit_code);
+	retval = coredump_wait(exit_code, &core_state);
 	if (retval < 0)
 		goto fail;
 
@@ -1812,6 +1811,7 @@ fail_unlock:
 
 	current->fsuid = fsuid;
 	complete_all(&mm->core_done);
+	mm->core_state = NULL;
 fail:
 	return retval;
 }

From b564daf806d492dd4f7afe9b6c83b8d35d137669 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:44 -0700
Subject: [PATCH 0834/1171] coredump: construct the list of coredumping threads
 at startup time

binfmt->core_dump() has to iterate over the all threads in system in order
to find the coredumping threads and construct the list using the
GFP_ATOMIC allocations.

With this patch each thread allocates the list node on exit_mm()'s stack and
adds itself to the list.

This allows us to do further changes:

	- simplify ->core_dump()

	- change exit_mm() to clear ->mm first, then wait for ->core_done.
	  this makes the coredumping process visible to oom_kill

	- kill mm->core_done

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/exec.c                |  2 ++
 include/linux/mm_types.h |  6 ++++++
 kernel/exit.c            | 15 ++++++++++++---
 3 files changed, 20 insertions(+), 3 deletions(-)

diff --git a/fs/exec.c b/fs/exec.c
index b8ee842d93cd..fe2873b8037f 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1604,6 +1604,8 @@ static int coredump_wait(int exit_code, struct core_state *core_state)
 
 	init_completion(&mm->core_done);
 	init_completion(&core_state->startup);
+	core_state->dumper.task = tsk;
+	core_state->dumper.next = NULL;
 	core_waiters = zap_threads(tsk, mm, core_state, exit_code);
 	up_write(&mm->mmap_sem);
 
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index ae99a28ba6ae..4d0d0abc79fe 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -159,8 +159,14 @@ struct vm_area_struct {
 #endif
 };
 
+struct core_thread {
+	struct task_struct *task;
+	struct core_thread *next;
+};
+
 struct core_state {
 	atomic_t nr_threads;
+	struct core_thread dumper;
 	struct completion startup;
 };
 
diff --git a/kernel/exit.c b/kernel/exit.c
index 63d82957baae..b66f0d55c791 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -664,6 +664,7 @@ assign_new_owner:
 static void exit_mm(struct task_struct * tsk)
 {
 	struct mm_struct *mm = tsk->mm;
+	struct core_state *core_state;
 
 	mm_release(tsk, mm);
 	if (!mm)
@@ -676,11 +677,19 @@ static void exit_mm(struct task_struct * tsk)
 	 * group with ->mm != NULL.
 	 */
 	down_read(&mm->mmap_sem);
-	if (mm->core_state) {
+	core_state = mm->core_state;
+	if (core_state) {
+		struct core_thread self;
 		up_read(&mm->mmap_sem);
 
-		if (atomic_dec_and_test(&mm->core_state->nr_threads))
-			complete(&mm->core_state->startup);
+		self.task = tsk;
+		self.next = xchg(&core_state->dumper.next, &self);
+		/*
+		 * Implies mb(), the result of xchg() must be visible
+		 * to core_state->dumper.
+		 */
+		if (atomic_dec_and_test(&core_state->nr_threads))
+			complete(&core_state->startup);
 
 		wait_for_completion(&mm->core_done);
 		down_read(&mm->mmap_sem);

From 83914441f94c6f2cd468ca97365f6c34f418706e Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:45 -0700
Subject: [PATCH 0835/1171] coredump: elf_core_dump: use core_state->dumper
 list

Kill the nasty rcu_read_lock() + do_each_thread() loop, use the list
encoded in mm->core_state instead, s/GFP_ATOMIC/GFP_KERNEL/.

This patch allows futher cleanups in binfmt_elf.c, in particular we can
kill the parallel info->threads list.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/binfmt_elf.c | 73 ++++++++++++++++++++-----------------------------
 1 file changed, 30 insertions(+), 43 deletions(-)

diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index bad7d8770d72..88d180306cf9 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -1478,7 +1478,7 @@ static int fill_note_info(struct elfhdr *elf, int phdrs,
 	const struct user_regset_view *view = task_user_regset_view(dump_task);
 	struct elf_thread_core_info *t;
 	struct elf_prpsinfo *psinfo;
-	struct task_struct *g, *p;
+	struct core_thread *ct;
 	unsigned int i;
 
 	info->size = 0;
@@ -1517,34 +1517,26 @@ static int fill_note_info(struct elfhdr *elf, int phdrs,
 	/*
 	 * Allocate a structure for each thread.
 	 */
-	rcu_read_lock();
-	do_each_thread(g, p)
-		if (p->mm == dump_task->mm) {
-			if (p->flags & PF_KTHREAD)
-				continue;
+	for (ct = &dump_task->mm->core_state->dumper; ct; ct = ct->next) {
+		t = kzalloc(offsetof(struct elf_thread_core_info,
+				     notes[info->thread_notes]),
+			    GFP_KERNEL);
+		if (unlikely(!t))
+			return 0;
 
-			t = kzalloc(offsetof(struct elf_thread_core_info,
-					     notes[info->thread_notes]),
-				    GFP_ATOMIC);
-			if (unlikely(!t)) {
-				rcu_read_unlock();
-				return 0;
-			}
-			t->task = p;
-			if (p == dump_task || !info->thread) {
-				t->next = info->thread;
-				info->thread = t;
-			} else {
-				/*
-				 * Make sure to keep the original task at
-				 * the head of the list.
-				 */
-				t->next = info->thread->next;
-				info->thread->next = t;
-			}
+		t->task = ct->task;
+		if (ct->task == dump_task || !info->thread) {
+			t->next = info->thread;
+			info->thread = t;
+		} else {
+			/*
+			 * Make sure to keep the original task at
+			 * the head of the list.
+			 */
+			t->next = info->thread->next;
+			info->thread->next = t;
 		}
-	while_each_thread(g, p);
-	rcu_read_unlock();
+	}
 
 	/*
 	 * Now fill in each thread's information.
@@ -1691,7 +1683,6 @@ static int fill_note_info(struct elfhdr *elf, int phdrs,
 {
 #define	NUM_NOTES	6
 	struct list_head *t;
-	struct task_struct *g, *p;
 
 	info->notes = NULL;
 	info->prstatus = NULL;
@@ -1723,23 +1714,19 @@ static int fill_note_info(struct elfhdr *elf, int phdrs,
 
 	info->thread_status_size = 0;
 	if (signr) {
+		struct core_thread *ct;
 		struct elf_thread_status *ets;
-		rcu_read_lock();
-		do_each_thread(g, p)
-			if (current->mm == p->mm && current != p) {
-				if (p->flags & PF_KTHREAD)
-					continue;
 
-				ets = kzalloc(sizeof(*ets), GFP_ATOMIC);
-				if (!ets) {
-					rcu_read_unlock();
-					return 0;
-				}
-				ets->thread = p;
-				list_add(&ets->list, &info->thread_list);
-			}
-		while_each_thread(g, p);
-		rcu_read_unlock();
+		for (ct = current->mm->core_state->dumper.next;
+						ct; ct = ct->next) {
+			ets = kzalloc(sizeof(*ets), GFP_KERNEL);
+			if (!ets)
+				return 0;
+
+			ets->thread = ct->task;
+			list_add(&ets->list, &info->thread_list);
+		}
+
 		list_for_each(t, &info->thread_list) {
 			int sz;
 

From 182c515fd2a942623aed4e4e0e0b37fe96571b05 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:45 -0700
Subject: [PATCH 0836/1171] coredump: elf_fdpic_core_dump: use
 core_state->dumper list

Kill the nasty rcu_read_lock() + do_each_thread() loop, use the list
encoded in mm->core_state instead, s/GFP_ATOMIC/GFP_KERNEL/.

This patch allows futher cleanups in binfmt_elf_fdpic.c.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: Roland McGrath <roland@redhat.com>
Cc: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/binfmt_elf_fdpic.c | 27 +++++++++++----------------
 1 file changed, 11 insertions(+), 16 deletions(-)

diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c
index 71bcc4b4d088..1b59b1edf26d 100644
--- a/fs/binfmt_elf_fdpic.c
+++ b/fs/binfmt_elf_fdpic.c
@@ -1573,7 +1573,6 @@ static int elf_fdpic_core_dump(long signr, struct pt_regs *regs,
 	struct memelfnote *notes = NULL;
 	struct elf_prstatus *prstatus = NULL;	/* NT_PRSTATUS */
 	struct elf_prpsinfo *psinfo = NULL;	/* NT_PRPSINFO */
- 	struct task_struct *g, *p;
  	LIST_HEAD(thread_list);
  	struct list_head *t;
 	elf_fpregset_t *fpu = NULL;
@@ -1622,23 +1621,19 @@ static int elf_fdpic_core_dump(long signr, struct pt_regs *regs,
 #endif
 
 	if (signr) {
+		struct core_thread *ct;
 		struct elf_thread_status *tmp;
-		rcu_read_lock();
-		do_each_thread(g,p)
-			if (current->mm == p->mm && current != p) {
-				if (p->flags & PF_KTHREAD)
-					continue;
 
-				tmp = kzalloc(sizeof(*tmp), GFP_ATOMIC);
-				if (!tmp) {
-					rcu_read_unlock();
-					goto cleanup;
-				}
-				tmp->thread = p;
-				list_add(&tmp->list, &thread_list);
-			}
-		while_each_thread(g,p);
-		rcu_read_unlock();
+		for (ct = current->mm->core_state->dumper.next;
+						ct; ct = ct->next) {
+			tmp = kzalloc(sizeof(*tmp), GFP_KERNEL);
+			if (!tmp)
+				goto cleanup;
+
+			tmp->thread = ct->task;
+			list_add(&tmp->list, &thread_list);
+		}
+
 		list_for_each(t, &thread_list) {
 			struct elf_thread_status *tmp;
 			int sz;

From a94e2d408eaedbd85aae259621d46fafc10479a2 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:46 -0700
Subject: [PATCH 0837/1171] coredump: kill mm->core_done

Now that we have core_state->dumper list we can use it to wake up the
sub-threads waiting for the coredump completion.

This uglifies the code and .text grows by 47 bytes, but otoh mm_struct
lessens by sizeof(struct completion).  Also, with this change we can
decouple exit_mm() from the coredumping code.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/exec.c                | 25 ++++++++++++++++++++++---
 include/linux/mm_types.h |  4 +---
 kernel/exit.c            |  8 +++++++-
 3 files changed, 30 insertions(+), 7 deletions(-)

diff --git a/fs/exec.c b/fs/exec.c
index fe2873b8037f..bff43aeb235e 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1602,7 +1602,6 @@ static int coredump_wait(int exit_code, struct core_state *core_state)
 	struct completion *vfork_done;
 	int core_waiters;
 
-	init_completion(&mm->core_done);
 	init_completion(&core_state->startup);
 	core_state->dumper.task = tsk;
 	core_state->dumper.next = NULL;
@@ -1628,6 +1627,27 @@ fail:
 	return core_waiters;
 }
 
+static void coredump_finish(struct mm_struct *mm)
+{
+	struct core_thread *curr, *next;
+	struct task_struct *task;
+
+	next = mm->core_state->dumper.next;
+	while ((curr = next) != NULL) {
+		next = curr->next;
+		task = curr->task;
+		/*
+		 * see exit_mm(), curr->task must not see
+		 * ->task == NULL before we read ->next.
+		 */
+		smp_mb();
+		curr->task = NULL;
+		wake_up_process(task);
+	}
+
+	mm->core_state = NULL;
+}
+
 /*
  * set_dumpable converts traditional three-value dumpable to two flags and
  * stores them into mm->flags.  It modifies lower two bits of mm->flags, but
@@ -1812,8 +1832,7 @@ fail_unlock:
 		argv_free(helper_argv);
 
 	current->fsuid = fsuid;
-	complete_all(&mm->core_done);
-	mm->core_state = NULL;
+	coredump_finish(mm);
 fail:
 	return retval;
 }
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index 4d0d0abc79fe..746f975b58ef 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -229,9 +229,7 @@ struct mm_struct {
 
 	unsigned long flags; /* Must use atomic bitops to access the bits */
 
-	/* coredumping support */
-	struct core_state *core_state;
-	struct completion core_done;
+	struct core_state *core_state; /* coredumping support */
 
 	/* aio bits */
 	rwlock_t		ioctx_list_lock;	/* aio lock */
diff --git a/kernel/exit.c b/kernel/exit.c
index b66f0d55c791..8a4d4d12e294 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -691,7 +691,13 @@ static void exit_mm(struct task_struct * tsk)
 		if (atomic_dec_and_test(&core_state->nr_threads))
 			complete(&core_state->startup);
 
-		wait_for_completion(&mm->core_done);
+		for (;;) {
+			set_task_state(tsk, TASK_UNINTERRUPTIBLE);
+			if (!self.task) /* see coredump_finish() */
+				break;
+			schedule();
+		}
+		__set_task_state(tsk, TASK_RUNNING);
 		down_read(&mm->mmap_sem);
 	}
 	atomic_inc(&mm->mm_count);

From 565b9b14e7f48131bca58840aa404bbef058fa89 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:47 -0700
Subject: [PATCH 0838/1171] coredump: format_corename: fix the "core_uses_pid"
 logic

I don't understand why the multi-thread coredump implies the core_uses_pid
behaviour, but we shouldn't use mm->mm_users for that.  This counter can
be incremented by get_task_mm().  Use the valued returned by
coredump_wait() instead.

Also, remove the "const char *pattern" argument, format_corename() can use
core_pattern directly.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Roland McGrath <roland@redhat.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/exec.c | 13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)

diff --git a/fs/exec.c b/fs/exec.c
index bff43aeb235e..5e559013e303 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1379,17 +1379,14 @@ EXPORT_SYMBOL(set_binfmt);
  * name into corename, which must have space for at least
  * CORENAME_MAX_SIZE bytes plus one byte for the zero terminator.
  */
-static int format_corename(char *corename, const char *pattern, long signr)
+static int format_corename(char *corename, int nr_threads, long signr)
 {
-	const char *pat_ptr = pattern;
+	const char *pat_ptr = core_pattern;
+	int ispipe = (*pat_ptr == '|');
 	char *out_ptr = corename;
 	char *const out_end = corename + CORENAME_MAX_SIZE;
 	int rc;
 	int pid_in_pattern = 0;
-	int ispipe = 0;
-
-	if (*pattern == '|')
-		ispipe = 1;
 
 	/* Repeat as long as we have more pattern to process and more output
 	   space */
@@ -1490,7 +1487,7 @@ static int format_corename(char *corename, const char *pattern, long signr)
 	 * and core_uses_pid is set, then .%pid will be appended to
 	 * the filename. Do not do this for piped commands. */
 	if (!ispipe && !pid_in_pattern
-            && (core_uses_pid || atomic_read(&current->mm->mm_users) != 1)) {
+	    && (core_uses_pid || nr_threads)) {
 		rc = snprintf(out_ptr, out_end - out_ptr,
 			      ".%d", task_tgid_vnr(current));
 		if (rc > out_end - out_ptr)
@@ -1753,7 +1750,7 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs)
 	 * uses lock_kernel()
 	 */
  	lock_kernel();
-	ispipe = format_corename(corename, core_pattern, signr);
+	ispipe = format_corename(corename, retval, signr);
 	unlock_kernel();
 	/*
 	 * Don't bother to check the RLIMIT_CORE value if core_pattern points

From 1a4d9b0aa0d3c50314e57525a5e5ec2cfc48b4c8 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:47 -0700
Subject: [PATCH 0839/1171] workqueues: insert_work: use "list_head *" instead
 of "int tail"

insert_work() inserts the new work_struct before or after cwq->worklist,
depending on the "int tail" parameter. Change it to accept "list_head *"
instead, this shrinks .text a bit and allows us to insert the barrier
after specific work_struct.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Jarek Poplawski <jarkao2@gmail.com>
Cc: Max Krasnyansky <maxk@qualcomm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/workqueue.c | 17 +++++++----------
 1 file changed, 7 insertions(+), 10 deletions(-)

diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index 6fd158b21026..d9a2d65cc63e 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -125,7 +125,7 @@ struct cpu_workqueue_struct *get_wq_data(struct work_struct *work)
 }
 
 static void insert_work(struct cpu_workqueue_struct *cwq,
-				struct work_struct *work, int tail)
+			struct work_struct *work, struct list_head *head)
 {
 	set_wq_data(work, cwq);
 	/*
@@ -133,10 +133,7 @@ static void insert_work(struct cpu_workqueue_struct *cwq,
 	 * result of list_add() below, see try_to_grab_pending().
 	 */
 	smp_wmb();
-	if (tail)
-		list_add_tail(&work->entry, &cwq->worklist);
-	else
-		list_add(&work->entry, &cwq->worklist);
+	list_add_tail(&work->entry, head);
 	wake_up(&cwq->more_work);
 }
 
@@ -146,7 +143,7 @@ static void __queue_work(struct cpu_workqueue_struct *cwq,
 	unsigned long flags;
 
 	spin_lock_irqsave(&cwq->lock, flags);
-	insert_work(cwq, work, 1);
+	insert_work(cwq, work, &cwq->worklist);
 	spin_unlock_irqrestore(&cwq->lock, flags);
 }
 
@@ -361,14 +358,14 @@ static void wq_barrier_func(struct work_struct *work)
 }
 
 static void insert_wq_barrier(struct cpu_workqueue_struct *cwq,
-					struct wq_barrier *barr, int tail)
+			struct wq_barrier *barr, struct list_head *head)
 {
 	INIT_WORK(&barr->work, wq_barrier_func);
 	__set_bit(WORK_STRUCT_PENDING, work_data_bits(&barr->work));
 
 	init_completion(&barr->done);
 
-	insert_work(cwq, &barr->work, tail);
+	insert_work(cwq, &barr->work, head);
 }
 
 static int flush_cpu_workqueue(struct cpu_workqueue_struct *cwq)
@@ -388,7 +385,7 @@ static int flush_cpu_workqueue(struct cpu_workqueue_struct *cwq)
 		active = 0;
 		spin_lock_irq(&cwq->lock);
 		if (!list_empty(&cwq->worklist) || cwq->current_work != NULL) {
-			insert_wq_barrier(cwq, &barr, 1);
+			insert_wq_barrier(cwq, &barr, &cwq->worklist);
 			active = 1;
 		}
 		spin_unlock_irq(&cwq->lock);
@@ -473,7 +470,7 @@ static void wait_on_cpu_work(struct cpu_workqueue_struct *cwq,
 
 	spin_lock_irq(&cwq->lock);
 	if (unlikely(cwq->current_work == work)) {
-		insert_wq_barrier(cwq, &barr, 0);
+		insert_wq_barrier(cwq, &barr, cwq->worklist.next);
 		running = 1;
 	}
 	spin_unlock_irq(&cwq->lock);

From db700897224b5ebdf852f2d38920ce428940d059 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:49 -0700
Subject: [PATCH 0840/1171] workqueues: implement flush_work()

Most of users of flush_workqueue() can be changed to use cancel_work_sync(),
but sometimes we really need to wait for the completion and cancelling is not
an option. schedule_on_each_cpu() is good example.

Add the new helper, flush_work(work), which waits for the completion of the
specific work_struct. More precisely, it "flushes" the result of of the last
queue_work() which is visible to the caller.

For example, this code

	queue_work(wq, work);
	/* WINDOW */
	queue_work(wq, work);

	flush_work(work);

doesn't necessary work "as expected". What can happen in the WINDOW above is

	- wq starts the execution of work->func()

	- the caller migrates to another CPU

now, after the 2nd queue_work() this work is active on the previous CPU, and
at the same time it is queued on another. In this case flush_work(work) may
return before the first work->func() completes.

It is trivial to add another helper

	int flush_work_sync(struct work_struct *work)
	{
		return flush_work(work) || wait_on_work(work);
	}

which works "more correctly", but it has to iterate over all CPUs and thus
it much slower than flush_work().

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: Max Krasnyansky <maxk@qualcomm.com>
Acked-by: Jarek Poplawski <jarkao2@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/workqueue.h |  2 ++
 kernel/workqueue.c        | 46 +++++++++++++++++++++++++++++++++++++++
 2 files changed, 48 insertions(+)

diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h
index 14d47120682b..5c158c477ac7 100644
--- a/include/linux/workqueue.h
+++ b/include/linux/workqueue.h
@@ -201,6 +201,8 @@ extern int keventd_up(void);
 extern void init_workqueues(void);
 int execute_in_process_context(work_func_t fn, struct execute_work *);
 
+extern int flush_work(struct work_struct *work);
+
 extern int cancel_work_sync(struct work_struct *work);
 
 /*
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index d9a2d65cc63e..ee41cf857d55 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -423,6 +423,52 @@ void flush_workqueue(struct workqueue_struct *wq)
 }
 EXPORT_SYMBOL_GPL(flush_workqueue);
 
+/**
+ * flush_work - block until a work_struct's callback has terminated
+ * @work: the work which is to be flushed
+ *
+ * It is expected that, prior to calling flush_work(), the caller has
+ * arranged for the work to not be requeued, otherwise it doesn't make
+ * sense to use this function.
+ */
+int flush_work(struct work_struct *work)
+{
+	struct cpu_workqueue_struct *cwq;
+	struct list_head *prev;
+	struct wq_barrier barr;
+
+	might_sleep();
+	cwq = get_wq_data(work);
+	if (!cwq)
+		return 0;
+
+	prev = NULL;
+	spin_lock_irq(&cwq->lock);
+	if (!list_empty(&work->entry)) {
+		/*
+		 * See the comment near try_to_grab_pending()->smp_rmb().
+		 * If it was re-queued under us we are not going to wait.
+		 */
+		smp_rmb();
+		if (unlikely(cwq != get_wq_data(work)))
+			goto out;
+		prev = &work->entry;
+	} else {
+		if (cwq->current_work != work)
+			goto out;
+		prev = &cwq->worklist;
+	}
+	insert_wq_barrier(cwq, &barr, prev->next);
+out:
+	spin_unlock_irq(&cwq->lock);
+	if (!prev)
+		return 0;
+
+	wait_for_completion(&barr.done);
+	return 1;
+}
+EXPORT_SYMBOL_GPL(flush_work);
+
 /*
  * Upon a successful return (>= 0), the caller "owns" WORK_STRUCT_PENDING bit,
  * so this work can't be re-armed in any way.

From 8616a89ab761239c963eea3a63be383f127cc7e8 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:49 -0700
Subject: [PATCH 0841/1171] workqueues: schedule_on_each_cpu: use flush_work()

Change schedule_on_each_cpu() to use flush_work() instead of
flush_workqueue(), this way we don't wait for other work_struct's which
can be queued meanwhile.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Jarek Poplawski <jarkao2@gmail.com>
Cc: Max Krasnyansky <maxk@qualcomm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/workqueue.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index ee41cf857d55..5fbffd302eb5 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -690,7 +690,8 @@ int schedule_on_each_cpu(work_func_t func)
 		set_bit(WORK_STRUCT_PENDING, work_data_bits(work));
 		__queue_work(per_cpu_ptr(keventd_wq->cpu_wq, cpu), work);
 	}
-	flush_workqueue(keventd_wq);
+	for_each_online_cpu(cpu)
+		flush_work(per_cpu_ptr(works, cpu));
 	put_online_cpus();
 	free_percpu(works);
 	return 0;

From 3da1c84c00c7e5fa8348336bd8c342f9128b0f14 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:50 -0700
Subject: [PATCH 0842/1171] workqueues: make get_online_cpus() useable for
 work->func()

workqueue_cpu_callback(CPU_DEAD) flushes cwq->thread under
cpu_maps_update_begin().  This means that the multithreaded workqueues
can't use get_online_cpus() due to the possible deadlock, very bad and
very old problem.

Introduce the new state, CPU_POST_DEAD, which is called after
cpu_hotplug_done() but before cpu_maps_update_done().

Change workqueue_cpu_callback() to use CPU_POST_DEAD instead of CPU_DEAD.
This means that create/destroy functions can't rely on get_online_cpus()
any longer and should take cpu_add_remove_lock instead.

[akpm@linux-foundation.org: fix CONFIG_SMP=n]
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Acked-by: Gautham R Shenoy <ego@in.ibm.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Max Krasnyansky <maxk@qualcomm.com>
Cc: Paul Jackson <pj@sgi.com>
Cc: Paul Menage <menage@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/cpu.h      | 15 +++++++++++----
 include/linux/notifier.h |  2 ++
 kernel/cpu.c             |  5 +++++
 kernel/workqueue.c       | 18 +++++++++---------
 4 files changed, 27 insertions(+), 13 deletions(-)

diff --git a/include/linux/cpu.h b/include/linux/cpu.h
index 7464ba3b4333..d7faf8808497 100644
--- a/include/linux/cpu.h
+++ b/include/linux/cpu.h
@@ -69,10 +69,11 @@ static inline void unregister_cpu_notifier(struct notifier_block *nb)
 #endif
 
 int cpu_up(unsigned int cpu);
-
 extern void cpu_hotplug_init(void);
+extern void cpu_maps_update_begin(void);
+extern void cpu_maps_update_done(void);
 
-#else
+#else	/* CONFIG_SMP */
 
 static inline int register_cpu_notifier(struct notifier_block *nb)
 {
@@ -87,10 +88,16 @@ static inline void cpu_hotplug_init(void)
 {
 }
 
+static inline void cpu_maps_update_begin(void)
+{
+}
+
+static inline void cpu_maps_update_done(void)
+{
+}
+
 #endif /* CONFIG_SMP */
 extern struct sysdev_class cpu_sysdev_class;
-extern void cpu_maps_update_begin(void);
-extern void cpu_maps_update_done(void);
 
 #ifdef CONFIG_HOTPLUG_CPU
 /* Stop CPUs going up and down. */
diff --git a/include/linux/notifier.h b/include/linux/notifier.h
index bd3d72ddf333..da2698b0fdd1 100644
--- a/include/linux/notifier.h
+++ b/include/linux/notifier.h
@@ -214,6 +214,8 @@ static inline int notifier_to_errno(int ret)
 #define CPU_DEAD		0x0007 /* CPU (unsigned)v dead */
 #define CPU_DYING		0x0008 /* CPU (unsigned)v not running any task,
 				        * not handling interrupts, soon dead */
+#define CPU_POST_DEAD		0x0009 /* CPU (unsigned)v dead, cpu_hotplug
+					* lock is dropped */
 
 /* Used for CPU hotplug events occuring while tasks are frozen due to a suspend
  * operation in progress
diff --git a/kernel/cpu.c b/kernel/cpu.c
index 2cc409ce0a8f..10ba5f1004a5 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -285,6 +285,11 @@ out_allowed:
 	set_cpus_allowed_ptr(current, &old_allowed);
 out_release:
 	cpu_hotplug_done();
+	if (!err) {
+		if (raw_notifier_call_chain(&cpu_chain, CPU_POST_DEAD | mod,
+					    hcpu) == NOTIFY_BAD)
+			BUG();
+	}
 	return err;
 }
 
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index 5fbffd302eb5..828e58230cbc 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -828,7 +828,7 @@ struct workqueue_struct *__create_workqueue_key(const char *name,
 		err = create_workqueue_thread(cwq, singlethread_cpu);
 		start_workqueue_thread(cwq, -1);
 	} else {
-		get_online_cpus();
+		cpu_maps_update_begin();
 		spin_lock(&workqueue_lock);
 		list_add(&wq->list, &workqueues);
 		spin_unlock(&workqueue_lock);
@@ -840,7 +840,7 @@ struct workqueue_struct *__create_workqueue_key(const char *name,
 			err = create_workqueue_thread(cwq, cpu);
 			start_workqueue_thread(cwq, cpu);
 		}
-		put_online_cpus();
+		cpu_maps_update_done();
 	}
 
 	if (err) {
@@ -854,8 +854,8 @@ EXPORT_SYMBOL_GPL(__create_workqueue_key);
 static void cleanup_workqueue_thread(struct cpu_workqueue_struct *cwq)
 {
 	/*
-	 * Our caller is either destroy_workqueue() or CPU_DEAD,
-	 * get_online_cpus() protects cwq->thread.
+	 * Our caller is either destroy_workqueue() or CPU_POST_DEAD,
+	 * cpu_add_remove_lock protects cwq->thread.
 	 */
 	if (cwq->thread == NULL)
 		return;
@@ -865,7 +865,7 @@ static void cleanup_workqueue_thread(struct cpu_workqueue_struct *cwq)
 
 	flush_cpu_workqueue(cwq);
 	/*
-	 * If the caller is CPU_DEAD and cwq->worklist was not empty,
+	 * If the caller is CPU_POST_DEAD and cwq->worklist was not empty,
 	 * a concurrent flush_workqueue() can insert a barrier after us.
 	 * However, in that case run_workqueue() won't return and check
 	 * kthread_should_stop() until it flushes all work_struct's.
@@ -889,14 +889,14 @@ void destroy_workqueue(struct workqueue_struct *wq)
 	const cpumask_t *cpu_map = wq_cpu_map(wq);
 	int cpu;
 
-	get_online_cpus();
+	cpu_maps_update_begin();
 	spin_lock(&workqueue_lock);
 	list_del(&wq->list);
 	spin_unlock(&workqueue_lock);
 
 	for_each_cpu_mask_nr(cpu, *cpu_map)
 		cleanup_workqueue_thread(per_cpu_ptr(wq->cpu_wq, cpu));
-	put_online_cpus();
+ 	cpu_maps_update_done();
 
 	free_percpu(wq->cpu_wq);
 	kfree(wq);
@@ -935,7 +935,7 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb,
 
 		case CPU_UP_CANCELED:
 			start_workqueue_thread(cwq, -1);
-		case CPU_DEAD:
+		case CPU_POST_DEAD:
 			cleanup_workqueue_thread(cwq);
 			break;
 		}
@@ -943,7 +943,7 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb,
 
 	switch (action) {
 	case CPU_UP_CANCELED:
-	case CPU_DEAD:
+	case CPU_POST_DEAD:
 		cpu_clear(cpu, cpu_populated_map);
 	}
 

From 69b895fd13d73aebf62b75502eb6513d43057ba3 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:51 -0700
Subject: [PATCH 0843/1171] S390 topology: don't use kthread() for
 arch_reinit_sched_domains()

Now that it is safe to use get_online_cpus() we can revert

	[S390] cpu topology: Fix possible deadlock.
	commit: fd781fa25c9e9c6fd1599df060b05e7c4ad724e5

and call arch_reinit_sched_domains() directly from topology_work_fn().

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Gautham R Shenoy <ego@in.ibm.com>
Tested-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Max Krasnyansky <maxk@qualcomm.com>
Cc: Paul Jackson <pj@sgi.com>
Cc: Paul Menage <menage@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Vegard Nossum <vegard.nossum@gmail.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/s390/kernel/topology.c | 14 +-------------
 1 file changed, 1 insertion(+), 13 deletions(-)

diff --git a/arch/s390/kernel/topology.c b/arch/s390/kernel/topology.c
index 212d618b0095..632b13e10053 100644
--- a/arch/s390/kernel/topology.c
+++ b/arch/s390/kernel/topology.c
@@ -9,7 +9,6 @@
 #include <linux/device.h>
 #include <linux/bootmem.h>
 #include <linux/sched.h>
-#include <linux/kthread.h>
 #include <linux/workqueue.h>
 #include <linux/cpu.h>
 #include <linux/smp.h>
@@ -230,20 +229,9 @@ void arch_update_cpu_topology(void)
 	}
 }
 
-static int topology_kthread(void *data)
-{
-	arch_reinit_sched_domains();
-	return 0;
-}
-
 static void topology_work_fn(struct work_struct *work)
 {
-	/* We can't call arch_reinit_sched_domains() from a multi-threaded
-	 * workqueue context since it may deadlock in case of cpu hotplug.
-	 * So we have to create a kernel thread in order to call
-	 * arch_reinit_sched_domains().
-	 */
-	kthread_run(topology_kthread, NULL, "topology_update");
+	arch_reinit_sched_domains();
 }
 
 void topology_schedule_update(void)

From a67da70dc0955580665f5444f318b92e69a3c272 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:52 -0700
Subject: [PATCH 0844/1171] workqueues: lockdep annotations for flush_work()

Add lockdep annotations to flush_work() and update the comment.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Jarek Poplawski <jarkao2@o2.pl>
Acked-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/workqueue.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index 828e58230cbc..4fcb75b98443 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -427,6 +427,8 @@ EXPORT_SYMBOL_GPL(flush_workqueue);
  * flush_work - block until a work_struct's callback has terminated
  * @work: the work which is to be flushed
  *
+ * Returns false if @work has already terminated.
+ *
  * It is expected that, prior to calling flush_work(), the caller has
  * arranged for the work to not be requeued, otherwise it doesn't make
  * sense to use this function.
@@ -442,6 +444,9 @@ int flush_work(struct work_struct *work)
 	if (!cwq)
 		return 0;
 
+	lock_acquire(&cwq->wq->lockdep_map, 0, 0, 0, 2, _THIS_IP_);
+	lock_release(&cwq->wq->lockdep_map, 1, _THIS_IP_);
+
 	prev = NULL;
 	spin_lock_irq(&cwq->lock);
 	if (!list_empty(&work->entry)) {

From ef1ca236b8d645349ed6569598ae3f6c1b9511c0 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:53 -0700
Subject: [PATCH 0845/1171] workqueues: queue_work() can use queue_work_on()

queue_work() can use queue_work_on() to avoid the code duplication.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/workqueue.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index 4fcb75b98443..fe08a8512ddd 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -159,14 +159,11 @@ static void __queue_work(struct cpu_workqueue_struct *cwq,
  */
 int queue_work(struct workqueue_struct *wq, struct work_struct *work)
 {
-	int ret = 0;
+	int ret;
+
+	ret = queue_work_on(get_cpu(), wq, work);
+	put_cpu();
 
-	if (!test_and_set_bit(WORK_STRUCT_PENDING, work_data_bits(work))) {
-		BUG_ON(!list_empty(&work->entry));
-		__queue_work(wq_per_cpu(wq, get_cpu()), work);
-		put_cpu();
-		ret = 1;
-	}
 	return ret;
 }
 EXPORT_SYMBOL_GPL(queue_work);

From 8de6d308bab4f67fcf953562f9f08f9527cad72d Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:53 -0700
Subject: [PATCH 0846/1171] workqueues: schedule_on_each_cpu() can use
 schedule_work_on()

schedule_on_each_cpu() can use schedule_work_on() to avoid the code
duplication.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/workqueue.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index fe08a8512ddd..7cf430372f89 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -689,8 +689,7 @@ int schedule_on_each_cpu(work_func_t func)
 		struct work_struct *work = per_cpu_ptr(works, cpu);
 
 		INIT_WORK(work, func);
-		set_bit(WORK_STRUCT_PENDING, work_data_bits(work));
-		__queue_work(per_cpu_ptr(keventd_wq->cpu_wq, cpu), work);
+		schedule_work_on(cpu, work);
 	}
 	for_each_online_cpu(cpu)
 		flush_work(per_cpu_ptr(works, cpu));

From 8448502cfc915f70e3f8923849ade27d472044cb Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@tv-sign.ru>
Date: Fri, 25 Jul 2008 01:47:54 -0700
Subject: [PATCH 0847/1171] workqueues: do CPU_UP_CANCELED if CPU_UP_PREPARE
 fails

The bug was pointed out by Akinobu Mita <akinobu.mita@gmail.com>, and this
patch is based on his original patch.

workqueue_cpu_callback(CPU_UP_PREPARE) expects that if it returns
NOTIFY_BAD, _cpu_up() will send CPU_UP_CANCELED then.

However, this is not true since

	"cpu hotplug: cpu: deliver CPU_UP_CANCELED only to NOTIFY_OKed callbacks with CPU_UP_PREPARE"
	commit: a0d8cdb652d35af9319a9e0fb7134de2a276c636

The callback which has returned NOTIFY_BAD will not receive
CPU_UP_CANCELED.  Change the code to fulfil the CPU_UP_CANCELED logic if
CPU_UP_PREPARE fails.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Reported-by: Akinobu Mita <akinobu.mita@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/workqueue.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index 7cf430372f89..ec7e4f62aaff 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -911,6 +911,7 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb,
 	unsigned int cpu = (unsigned long)hcpu;
 	struct cpu_workqueue_struct *cwq;
 	struct workqueue_struct *wq;
+	int ret = NOTIFY_OK;
 
 	action &= ~CPU_TASKS_FROZEN;
 
@@ -918,7 +919,7 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb,
 	case CPU_UP_PREPARE:
 		cpu_set(cpu, cpu_populated_map);
 	}
-
+undo:
 	list_for_each_entry(wq, &workqueues, list) {
 		cwq = per_cpu_ptr(wq->cpu_wq, cpu);
 
@@ -928,7 +929,9 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb,
 				break;
 			printk(KERN_ERR "workqueue [%s] for %i failed\n",
 				wq->name, cpu);
-			return NOTIFY_BAD;
+			action = CPU_UP_CANCELED;
+			ret = NOTIFY_BAD;
+			goto undo;
 
 		case CPU_ONLINE:
 			start_workqueue_thread(cwq, cpu);
@@ -948,7 +951,7 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb,
 		cpu_clear(cpu, cpu_populated_map);
 	}
 
-	return NOTIFY_OK;
+	return ret;
 }
 
 void __init init_workqueues(void)

From 95b68dec0d52c7b8fea3698b3938cf3ab936436b Mon Sep 17 00:00:00 2001
From: Chandru <chandru@in.ibm.com>
Date: Fri, 25 Jul 2008 01:47:55 -0700
Subject: [PATCH 0848/1171] calgary iommu: use the first kernels TCE tables in
 kdump

kdump kernel fails to boot with calgary iommu and aacraid driver on a x366
box.  The ongoing dma's of aacraid from the first kernel continue to exist
until the driver is loaded in the kdump kernel.  Calgary is initialized
prior to aacraid and creation of new tce tables causes wrong dma's to
occur.  Here we try to get the tce tables of the first kernel in kdump
kernel and use them.  While in the kdump kernel we do not allocate new tce
tables but instead read the base address register contents of calgary
iommu and use the tables that the registers point to.  With these changes
the kdump kernel and hence aacraid now boots normally.

Signed-off-by: Chandru Siddalingappa <chandru@in.ibm.com>
Acked-by: Muli Ben-Yehuda <muli@il.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/kernel/pci-calgary_64.c | 85 +++++++++++++++++++++++++++++---
 include/linux/crash_dump.h       |  8 +++
 2 files changed, 87 insertions(+), 6 deletions(-)

diff --git a/arch/x86/kernel/pci-calgary_64.c b/arch/x86/kernel/pci-calgary_64.c
index 151f2d171f7c..19e7fc7c2c4f 100644
--- a/arch/x86/kernel/pci-calgary_64.c
+++ b/arch/x86/kernel/pci-calgary_64.c
@@ -29,6 +29,7 @@
 #include <linux/mm.h>
 #include <linux/spinlock.h>
 #include <linux/string.h>
+#include <linux/crash_dump.h>
 #include <linux/dma-mapping.h>
 #include <linux/bitops.h>
 #include <linux/pci_ids.h>
@@ -167,6 +168,8 @@ static void calgary_dump_error_regs(struct iommu_table *tbl);
 static void calioc2_handle_quirks(struct iommu_table *tbl, struct pci_dev *dev);
 static void calioc2_tce_cache_blast(struct iommu_table *tbl);
 static void calioc2_dump_error_regs(struct iommu_table *tbl);
+static void calgary_init_bitmap_from_tce_table(struct iommu_table *tbl);
+static void get_tce_space_from_tar(void);
 
 static struct cal_chipset_ops calgary_chip_ops = {
 	.handle_quirks = calgary_handle_quirks,
@@ -830,7 +833,11 @@ static int __init calgary_setup_tar(struct pci_dev *dev, void __iomem *bbar)
 
 	tbl = pci_iommu(dev->bus);
 	tbl->it_base = (unsigned long)bus_info[dev->bus->number].tce_space;
-	tce_free(tbl, 0, tbl->it_size);
+
+	if (is_kdump_kernel())
+		calgary_init_bitmap_from_tce_table(tbl);
+	else
+		tce_free(tbl, 0, tbl->it_size);
 
 	if (is_calgary(dev->device))
 		tbl->chip_ops = &calgary_chip_ops;
@@ -1209,6 +1216,10 @@ static int __init calgary_init(void)
 	if (ret)
 		return ret;
 
+	/* Purely for kdump kernel case */
+	if (is_kdump_kernel())
+		get_tce_space_from_tar();
+
 	do {
 		dev = pci_get_device(PCI_VENDOR_ID_IBM, PCI_ANY_ID, dev);
 		if (!dev)
@@ -1339,6 +1350,61 @@ static int __init calgary_bus_has_devices(int bus, unsigned short pci_dev)
 	return (val != 0xffffffff);
 }
 
+/*
+ * calgary_init_bitmap_from_tce_table():
+ * Funtion for kdump case. In the second/kdump kernel initialize
+ * the bitmap based on the tce table entries obtained from first kernel
+ */
+static void calgary_init_bitmap_from_tce_table(struct iommu_table *tbl)
+{
+	u64 *tp;
+	unsigned int index;
+	tp = ((u64 *)tbl->it_base);
+	for (index = 0 ; index < tbl->it_size; index++) {
+		if (*tp != 0x0)
+			set_bit(index, tbl->it_map);
+		tp++;
+	}
+}
+
+/*
+ * get_tce_space_from_tar():
+ * Function for kdump case. Get the tce tables from first kernel
+ * by reading the contents of the base adress register of calgary iommu
+ */
+static void get_tce_space_from_tar()
+{
+	int bus;
+	void __iomem *target;
+	unsigned long tce_space;
+
+	for (bus = 0; bus < MAX_PHB_BUS_NUM; bus++) {
+		struct calgary_bus_info *info = &bus_info[bus];
+		unsigned short pci_device;
+		u32 val;
+
+		val = read_pci_config(bus, 0, 0, 0);
+		pci_device = (val & 0xFFFF0000) >> 16;
+
+		if (!is_cal_pci_dev(pci_device))
+			continue;
+		if (info->translation_disabled)
+			continue;
+
+		if (calgary_bus_has_devices(bus, pci_device) ||
+						translate_empty_slots) {
+			target = calgary_reg(bus_info[bus].bbar,
+						tar_offset(bus));
+			tce_space = be64_to_cpu(readq(target));
+			tce_space = tce_space & TAR_SW_BITS;
+
+			tce_space = tce_space & (~specified_table_size);
+			info->tce_space = (u64 *)__va(tce_space);
+		}
+	}
+	return;
+}
+
 void __init detect_calgary(void)
 {
 	int bus;
@@ -1394,7 +1460,8 @@ void __init detect_calgary(void)
 		return;
 	}
 
-	specified_table_size = determine_tce_table_size(max_pfn * PAGE_SIZE);
+	specified_table_size = determine_tce_table_size((is_kdump_kernel() ?
+					saved_max_pfn : max_pfn) * PAGE_SIZE);
 
 	for (bus = 0; bus < MAX_PHB_BUS_NUM; bus++) {
 		struct calgary_bus_info *info = &bus_info[bus];
@@ -1412,10 +1479,16 @@ void __init detect_calgary(void)
 
 		if (calgary_bus_has_devices(bus, pci_device) ||
 		    translate_empty_slots) {
-			tbl = alloc_tce_table();
-			if (!tbl)
-				goto cleanup;
-			info->tce_space = tbl;
+			/*
+			 * If it is kdump kernel, find and use tce tables
+			 * from first kernel, else allocate tce tables here
+			 */
+			if (!is_kdump_kernel()) {
+				tbl = alloc_tce_table();
+				if (!tbl)
+					goto cleanup;
+				info->tce_space = tbl;
+			}
 			calgary_found = 1;
 		}
 	}
diff --git a/include/linux/crash_dump.h b/include/linux/crash_dump.h
index 22c7ac5cd80c..6cd39a927e1f 100644
--- a/include/linux/crash_dump.h
+++ b/include/linux/crash_dump.h
@@ -22,5 +22,13 @@ extern struct proc_dir_entry *proc_vmcore;
 
 #define vmcore_elf_check_arch(x) (elf_check_arch(x) || vmcore_elf_check_arch_cross(x))
 
+static inline int is_kdump_kernel(void)
+{
+	return (elfcorehdr_addr != ELFCORE_ADDR_MAX) ? 1 : 0;
+}
+#else /* !CONFIG_CRASH_DUMP */
+static inline int is_kdump_kernel(void) { return 0; }
 #endif /* CONFIG_CRASH_DUMP */
+
+extern unsigned long saved_max_pfn;
 #endif /* LINUX_CRASHDUMP_H */

From 2027d1abc25ff770cc3bc936abd33570ce85d85a Mon Sep 17 00:00:00 2001
From: Nadia Derbey <Nadia.Derbey@bull.net>
Date: Fri, 25 Jul 2008 01:47:57 -0700
Subject: [PATCH 0849/1171] idr: change the idr structure

After scalability problems have been detected when using the sysV ipcs, I have
proposed to use an RCU based implementation of the IDR api instead (see
threads http://lkml.org/lkml/2008/4/11/212 and
http://lkml.org/lkml/2008/4/29/295).

This resulted in many people asking to convert the idr API and make it rcu
safe (because most of the code was duplicated and thus unmaintanable and
unreviewable).

So here is a first attempt.

The important change wrt to the idr API itself is during idr removes: idr
layers are freed after a grace period, instead of being moved to the free
list.

The important change wrt to ipcs, is that idr_find() can now be called
locklessly inside a rcu read critical section.

Here are the results I've got for the pmsg test sent by Manfred:

   2.6.25-rc3-mm1   2.6.25-rc3-mm1+   2.6.25-mm1   Patched 2.6.25-mm1
1         1168441           1064021       876000               947488
2         1094264            921059      1549592              1730685
3         2082520           1738165      1694370              2324880
4         2079929           1695521       404553              2400408
5         2898758            406566       391283              3246580
6         2921417            261275       263249              3752148
7         3308761            126056       191742              4243142
8         3329456            100129       141722              4275780

1st column: stock 2.6.25-rc3-mm1
2nd column: 2.6.25-rc3-mm1 + ipc patches (store ipcs into idrs)
3nd column: stock 2.6.25-mm1
4th column: 2.6.25-mm1 + this pacth series.

This patch:

Add an rcu_head to the idr_layer structure in order to free it after a grace
period.

Signed-off-by: Nadia Derbey <Nadia.Derbey@bull.net>
Reviewed-by: "Paul E. McKenney" <paulmck@us.ibm.com>
Cc: Manfred Spraul <manfred@colorfullife.com>
Cc: Jim Houston <jim.houston@comcast.net>
Cc: Pierre Peiffer <peifferp@gmail.com>
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/idr.h | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/include/linux/idr.h b/include/linux/idr.h
index 9a2d762124de..1af61d23be36 100644
--- a/include/linux/idr.h
+++ b/include/linux/idr.h
@@ -15,6 +15,7 @@
 #include <linux/types.h>
 #include <linux/bitops.h>
 #include <linux/init.h>
+#include <linux/rcupdate.h>
 
 #if BITS_PER_LONG == 32
 # define IDR_BITS 5
@@ -51,6 +52,7 @@ struct idr_layer {
 	unsigned long		 bitmap; /* A zero bit means "space here" */
 	struct idr_layer	*ary[1<<IDR_BITS];
 	int			 count;	 /* When zero, we can release it */
+	struct rcu_head		 rcu_head;
 };
 
 struct idr {

From 4ae537892ab9858f71c78701f4651ad1ca531a1b Mon Sep 17 00:00:00 2001
From: Nadia Derbey <Nadia.Derbey@bull.net>
Date: Fri, 25 Jul 2008 01:47:58 -0700
Subject: [PATCH 0850/1171] idr: rename some of the idr APIs internal routines

This is a trivial patch that renames:

   . alloc_layer to get_from_free_list since it idr_pre_get that actually
     allocates memory.
   . free_layer to move_to_free_list since memory is not actually freed there.

This makes things more clear for the next patches.

Signed-off-by: Nadia Derbey <Nadia.Derbey@bull.net>
Reviewed-by: "Paul E. McKenney" <paulmck@us.ibm.com>
Cc: Manfred Spraul <manfred@colorfullife.com>
Cc: Jim Houston <jim.houston@comcast.net>
Cc: Pierre Peiffer <peifferp@gmail.com>
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 lib/idr.c | 31 ++++++++++++++++---------------
 1 file changed, 16 insertions(+), 15 deletions(-)

diff --git a/lib/idr.c b/lib/idr.c
index 7a02e173f027..8170ace154fb 100644
--- a/lib/idr.c
+++ b/lib/idr.c
@@ -35,7 +35,7 @@
 
 static struct kmem_cache *idr_layer_cache;
 
-static struct idr_layer *alloc_layer(struct idr *idp)
+static struct idr_layer *get_from_free_list(struct idr *idp)
 {
 	struct idr_layer *p;
 	unsigned long flags;
@@ -51,14 +51,14 @@ static struct idr_layer *alloc_layer(struct idr *idp)
 }
 
 /* only called when idp->lock is held */
-static void __free_layer(struct idr *idp, struct idr_layer *p)
+static void __move_to_free_list(struct idr *idp, struct idr_layer *p)
 {
 	p->ary[0] = idp->id_free;
 	idp->id_free = p;
 	idp->id_free_cnt++;
 }
 
-static void free_layer(struct idr *idp, struct idr_layer *p)
+static void move_to_free_list(struct idr *idp, struct idr_layer *p)
 {
 	unsigned long flags;
 
@@ -66,7 +66,7 @@ static void free_layer(struct idr *idp, struct idr_layer *p)
 	 * Depends on the return element being zeroed.
 	 */
 	spin_lock_irqsave(&idp->lock, flags);
-	__free_layer(idp, p);
+	__move_to_free_list(idp, p);
 	spin_unlock_irqrestore(&idp->lock, flags);
 }
 
@@ -109,7 +109,7 @@ int idr_pre_get(struct idr *idp, gfp_t gfp_mask)
 		new = kmem_cache_alloc(idr_layer_cache, gfp_mask);
 		if (new == NULL)
 			return (0);
-		free_layer(idp, new);
+		move_to_free_list(idp, new);
 	}
 	return 1;
 }
@@ -167,7 +167,8 @@ static int sub_alloc(struct idr *idp, int *starting_id, struct idr_layer **pa)
 		 * Create the layer below if it is missing.
 		 */
 		if (!p->ary[m]) {
-			if (!(new = alloc_layer(idp)))
+			new = get_from_free_list(idp);
+			if (!new)
 				return -1;
 			p->ary[m] = new;
 			p->count++;
@@ -192,7 +193,7 @@ build_up:
 	p = idp->top;
 	layers = idp->layers;
 	if (unlikely(!p)) {
-		if (!(p = alloc_layer(idp)))
+		if (!(p = get_from_free_list(idp)))
 			return -1;
 		layers = 1;
 	}
@@ -204,7 +205,7 @@ build_up:
 		layers++;
 		if (!p->count)
 			continue;
-		if (!(new = alloc_layer(idp))) {
+		if (!(new = get_from_free_list(idp))) {
 			/*
 			 * The allocation failed.  If we built part of
 			 * the structure tear it down.
@@ -214,7 +215,7 @@ build_up:
 				p = p->ary[0];
 				new->ary[0] = NULL;
 				new->bitmap = new->count = 0;
-				__free_layer(idp, new);
+				__move_to_free_list(idp, new);
 			}
 			spin_unlock_irqrestore(&idp->lock, flags);
 			return -1;
@@ -351,7 +352,7 @@ static void sub_remove(struct idr *idp, int shift, int id)
 		__clear_bit(n, &p->bitmap);
 		p->ary[n] = NULL;
 		while(*paa && ! --((**paa)->count)){
-			free_layer(idp, **paa);
+			move_to_free_list(idp, **paa);
 			**paa-- = NULL;
 		}
 		if (!*paa)
@@ -378,12 +379,12 @@ void idr_remove(struct idr *idp, int id)
 
 		p = idp->top->ary[0];
 		idp->top->bitmap = idp->top->count = 0;
-		free_layer(idp, idp->top);
+		move_to_free_list(idp, idp->top);
 		idp->top = p;
 		--idp->layers;
 	}
 	while (idp->id_free_cnt >= IDR_FREE_MAX) {
-		p = alloc_layer(idp);
+		p = get_from_free_list(idp);
 		kmem_cache_free(idr_layer_cache, p);
 	}
 	return;
@@ -426,7 +427,7 @@ void idr_remove_all(struct idr *idp)
 		while (n < fls(id)) {
 			if (p) {
 				memset(p, 0, sizeof *p);
-				free_layer(idp, p);
+				move_to_free_list(idp, p);
 			}
 			n += IDR_BITS;
 			p = *--paa;
@@ -444,7 +445,7 @@ EXPORT_SYMBOL(idr_remove_all);
 void idr_destroy(struct idr *idp)
 {
 	while (idp->id_free_cnt) {
-		struct idr_layer *p = alloc_layer(idp);
+		struct idr_layer *p = get_from_free_list(idp);
 		kmem_cache_free(idr_layer_cache, p);
 	}
 }
@@ -749,7 +750,7 @@ int ida_get_new_above(struct ida *ida, int starting_id, int *p_id)
 	 * allocation.
 	 */
 	if (ida->idr.id_free_cnt || ida->free_bitmap) {
-		struct idr_layer *p = alloc_layer(&ida->idr);
+		struct idr_layer *p = get_from_free_list(&ida->idr);
 		if (p)
 			kmem_cache_free(idr_layer_cache, p);
 	}

From f098ad655f4dd8e3da98ffbeda9cedcc4459c01a Mon Sep 17 00:00:00 2001
From: Nadia Derbey <Nadia.Derbey@bull.net>
Date: Fri, 25 Jul 2008 01:47:59 -0700
Subject: [PATCH 0851/1171] idr: fix a printk call

Fix the incomplete printk call.

Signed-off-by: Nadia Derbey <Nadia.Derbey@bull.net>
Reviewed-by: "Paul E. McKenney" <paulmck@us.ibm.com>
Cc: Manfred Spraul <manfred@colorfullife.com>
Cc: Jim Houston <jim.houston@comcast.net>
Cc: Pierre Peiffer <peifferp@gmail.com>
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 lib/idr.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/lib/idr.c b/lib/idr.c
index 8170ace154fb..9d905b131ecb 100644
--- a/lib/idr.c
+++ b/lib/idr.c
@@ -326,7 +326,8 @@ EXPORT_SYMBOL(idr_get_new);
 
 static void idr_remove_warning(int id)
 {
-	printk("idr_remove called for id=%d which is not allocated.\n", id);
+	printk(KERN_WARNING
+		"idr_remove called for id=%d which is not allocated.\n", id);
 	dump_stack();
 }
 

From 944ca05c7b4972f2ebf37262e0f4933d178ad6db Mon Sep 17 00:00:00 2001
From: Nadia Derbey <Nadia.Derbey@bull.net>
Date: Fri, 25 Jul 2008 01:47:59 -0700
Subject: [PATCH 0852/1171] idr: error checking factorization

Do some code factorization in the return code analysis.

Signed-off-by: Nadia Derbey <Nadia.Derbey@bull.net>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Cc: Manfred Spraul <manfred@colorfullife.com>
Cc: Jim Houston <jim.houston@comcast.net>
Cc: Pierre Peiffer <peifferp@gmail.com>
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/idr.h |  6 ++++++
 lib/idr.c           | 30 +++++++++---------------------
 2 files changed, 15 insertions(+), 21 deletions(-)

diff --git a/include/linux/idr.h b/include/linux/idr.h
index 1af61d23be36..762c3f2c631d 100644
--- a/include/linux/idr.h
+++ b/include/linux/idr.h
@@ -73,6 +73,12 @@ struct idr {
 }
 #define DEFINE_IDR(name)	struct idr name = IDR_INIT(name)
 
+/* Actions to be taken after a call to _idr_sub_alloc */
+#define IDR_NEED_TO_GROW -2
+#define IDR_NOMORE_SPACE -3
+
+#define _idr_rc_to_errno(rc) ((rc) == -1 ? -EAGAIN : -ENOSPC)
+
 /*
  * This is what we export.
  */
diff --git a/lib/idr.c b/lib/idr.c
index 9d905b131ecb..80ba06f29d36 100644
--- a/lib/idr.c
+++ b/lib/idr.c
@@ -143,7 +143,7 @@ static int sub_alloc(struct idr *idp, int *starting_id, struct idr_layer **pa)
 			/* if already at the top layer, we need to grow */
 			if (!(p = pa[l])) {
 				*starting_id = id;
-				return -2;
+				return IDR_NEED_TO_GROW;
 			}
 
 			/* If we need to go up one layer, continue the
@@ -160,7 +160,7 @@ static int sub_alloc(struct idr *idp, int *starting_id, struct idr_layer **pa)
 			id = ((id >> sh) ^ n ^ m) << sh;
 		}
 		if ((id >= MAX_ID_BIT) || (id < 0))
-			return -3;
+			return IDR_NOMORE_SPACE;
 		if (l == 0)
 			break;
 		/*
@@ -229,7 +229,7 @@ build_up:
 	idp->top = p;
 	idp->layers = layers;
 	v = sub_alloc(idp, &id, pa);
-	if (v == -2)
+	if (v == IDR_NEED_TO_GROW)
 		goto build_up;
 	return(v);
 }
@@ -278,12 +278,8 @@ int idr_get_new_above(struct idr *idp, void *ptr, int starting_id, int *id)
 	 * This is a cheap hack until the IDR code can be fixed to
 	 * return proper error values.
 	 */
-	if (rv < 0) {
-		if (rv == -1)
-			return -EAGAIN;
-		else /* Will be -3 */
-			return -ENOSPC;
-	}
+	if (rv < 0)
+		return _idr_rc_to_errno(rv);
 	*id = rv;
 	return 0;
 }
@@ -313,12 +309,8 @@ int idr_get_new(struct idr *idp, void *ptr, int *id)
 	 * This is a cheap hack until the IDR code can be fixed to
 	 * return proper error values.
 	 */
-	if (rv < 0) {
-		if (rv == -1)
-			return -EAGAIN;
-		else /* Will be -3 */
-			return -ENOSPC;
-	}
+	if (rv < 0)
+		return _idr_rc_to_errno(rv);
 	*id = rv;
 	return 0;
 }
@@ -696,12 +688,8 @@ int ida_get_new_above(struct ida *ida, int starting_id, int *p_id)
  restart:
 	/* get vacant slot */
 	t = idr_get_empty_slot(&ida->idr, idr_id, pa);
-	if (t < 0) {
-		if (t == -1)
-			return -EAGAIN;
-		else /* will be -3 */
-			return -ENOSPC;
-	}
+	if (t < 0)
+		return _idr_rc_to_errno(t);
 
 	if (t * IDA_BITMAP_BITS >= MAX_ID_BIT)
 		return -ENOSPC;

From 3219b3b7456d5cf15ba7b1fe7b1bcf15ce8840e2 Mon Sep 17 00:00:00 2001
From: Nadia Derbey <Nadia.Derbey@bull.net>
Date: Fri, 25 Jul 2008 01:48:00 -0700
Subject: [PATCH 0853/1171] idr: make idr_get_new* rcu-safe

Make the idr_get_new* routines rcu-safe.

Signed-off-by: Nadia Derbey <Nadia.Derbey@bull.net>
Reviewed-by: "Paul E. McKenney" <paulmck@us.ibm.com>
Cc: Manfred Spraul <manfred@colorfullife.com>
Cc: Jim Houston <jim.houston@comcast.net>
Cc: Pierre Peiffer <peifferp@gmail.com>
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 lib/idr.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/lib/idr.c b/lib/idr.c
index 80ba06f29d36..44ab3b2a4eba 100644
--- a/lib/idr.c
+++ b/lib/idr.c
@@ -6,6 +6,8 @@
  * Modified by George Anzinger to reuse immediately and to use
  * find bit instructions.  Also removed _irq on spinlocks.
  *
+ * Modified by Nadia Derbey to make it RCU safe.
+ *
  * Small id to pointer translation service.
  *
  * It uses a radix tree like structure as a sparse array indexed
@@ -96,7 +98,7 @@ static void idr_mark_full(struct idr_layer **pa, int id)
  * @gfp_mask:	memory allocation flags
  *
  * This function should be called prior to locking and calling the
- * following function.  It preallocates enough memory to satisfy
+ * idr_get_new* functions. It preallocates enough memory to satisfy
  * the worst possible allocation.
  *
  * If the system is REALLY out of memory this function returns 0,
@@ -170,7 +172,7 @@ static int sub_alloc(struct idr *idp, int *starting_id, struct idr_layer **pa)
 			new = get_from_free_list(idp);
 			if (!new)
 				return -1;
-			p->ary[m] = new;
+			rcu_assign_pointer(p->ary[m], new);
 			p->count++;
 		}
 		pa[l--] = p;
@@ -226,7 +228,7 @@ build_up:
 			__set_bit(0, &new->bitmap);
 		p = new;
 	}
-	idp->top = p;
+	rcu_assign_pointer(idp->top, p);
 	idp->layers = layers;
 	v = sub_alloc(idp, &id, pa);
 	if (v == IDR_NEED_TO_GROW)
@@ -245,7 +247,8 @@ static int idr_get_new_above_int(struct idr *idp, void *ptr, int starting_id)
 		 * Successfully found an empty slot.  Install the user
 		 * pointer and mark the slot full.
 		 */
-		pa[0]->ary[id & IDR_MASK] = (struct idr_layer *)ptr;
+		rcu_assign_pointer(pa[0]->ary[id & IDR_MASK],
+				(struct idr_layer *)ptr);
 		pa[0]->count++;
 		idr_mark_full(pa, id);
 	}
@@ -710,7 +713,8 @@ int ida_get_new_above(struct ida *ida, int starting_id, int *p_id)
 			return -EAGAIN;
 
 		memset(bitmap, 0, sizeof(struct ida_bitmap));
-		pa[0]->ary[idr_id & IDR_MASK] = (void *)bitmap;
+		rcu_assign_pointer(pa[0]->ary[idr_id & IDR_MASK],
+				(void *)bitmap);
 		pa[0]->count++;
 	}
 

From f9c46d6ea5ce138a886c3a0f10a46130afab75f5 Mon Sep 17 00:00:00 2001
From: Nadia Derbey <Nadia.Derbey@bull.net>
Date: Fri, 25 Jul 2008 01:48:01 -0700
Subject: [PATCH 0854/1171] idr: make idr_find rcu-safe

Make idr_find rcu-safe: it can now be called inside an rcu_read critical
section.

Signed-off-by: Nadia Derbey <Nadia.Derbey@bull.net>
Reviewed-by: "Paul E. McKenney" <paulmck@us.ibm.com>
Cc: Manfred Spraul <manfred@colorfullife.com>
Cc: Jim Houston <jim.houston@comcast.net>
Cc: Pierre Peiffer <peifferp@gmail.com>
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/idr.h | 16 ++++++++++++++++
 lib/idr.c           | 11 ++++++-----
 2 files changed, 22 insertions(+), 5 deletions(-)

diff --git a/include/linux/idr.h b/include/linux/idr.h
index 762c3f2c631d..fa035f96f2a3 100644
--- a/include/linux/idr.h
+++ b/include/linux/idr.h
@@ -79,6 +79,22 @@ struct idr {
 
 #define _idr_rc_to_errno(rc) ((rc) == -1 ? -EAGAIN : -ENOSPC)
 
+/**
+ * idr synchronization (stolen from radix-tree.h)
+ *
+ * idr_find() is able to be called locklessly, using RCU. The caller must
+ * ensure calls to this function are made within rcu_read_lock() regions.
+ * Other readers (lock-free or otherwise) and modifications may be running
+ * concurrently.
+ *
+ * It is still required that the caller manage the synchronization and
+ * lifetimes of the items. So if RCU lock-free lookups are used, typically
+ * this would mean that the items have their own locks, or are amenable to
+ * lock-free access; and that the items are freed by RCU (or only freed after
+ * having been deleted from the idr tree *and* a synchronize_rcu() grace
+ * period).
+ */
+
 /*
  * This is what we export.
  */
diff --git a/lib/idr.c b/lib/idr.c
index 44ab3b2a4eba..21e12af1f231 100644
--- a/lib/idr.c
+++ b/lib/idr.c
@@ -456,7 +456,8 @@ EXPORT_SYMBOL(idr_destroy);
  * return indicates that @id is not valid or you passed %NULL in
  * idr_get_new().
  *
- * The caller must serialize idr_find() vs idr_get_new() and idr_remove().
+ * This function can be called under rcu_read_lock(), given that the leaf
+ * pointers lifetimes are correctly managed.
  */
 void *idr_find(struct idr *idp, int id)
 {
@@ -464,7 +465,7 @@ void *idr_find(struct idr *idp, int id)
 	struct idr_layer *p;
 
 	n = idp->layers * IDR_BITS;
-	p = idp->top;
+	p = rcu_dereference(idp->top);
 
 	/* Mask off upper bits we don't use for the search. */
 	id &= MAX_ID_MASK;
@@ -474,7 +475,7 @@ void *idr_find(struct idr *idp, int id)
 
 	while (n > 0 && p) {
 		n -= IDR_BITS;
-		p = p->ary[(id >> n) & IDR_MASK];
+		p = rcu_dereference(p->ary[(id >> n) & IDR_MASK]);
 	}
 	return((void *)p);
 }
@@ -507,7 +508,7 @@ int idr_for_each(struct idr *idp,
 	struct idr_layer **paa = &pa[0];
 
 	n = idp->layers * IDR_BITS;
-	p = idp->top;
+	p = rcu_dereference(idp->top);
 	max = 1 << n;
 
 	id = 0;
@@ -515,7 +516,7 @@ int idr_for_each(struct idr *idp,
 		while (n > 0 && p) {
 			n -= IDR_BITS;
 			*paa++ = p;
-			p = p->ary[(id >> n) & IDR_MASK];
+			p = rcu_dereference(p->ary[(id >> n) & IDR_MASK]);
 		}
 
 		if (p) {

From cf481c20c476ad2c0febdace9ce23f5a4db19582 Mon Sep 17 00:00:00 2001
From: Nadia Derbey <Nadia.Derbey@bull.net>
Date: Fri, 25 Jul 2008 01:48:02 -0700
Subject: [PATCH 0855/1171] idr: make idr_remove rcu-safe

Introduce the free_layer() routine: it is the one that actually frees memory
after a grace period has elapsed.

Signed-off-by: Nadia Derbey <Nadia.Derbey@bull.net>
Reviewed-by: "Paul E. McKenney" <paulmck@us.ibm.com>
Cc: Manfred Spraul <manfred@colorfullife.com>
Cc: Jim Houston <jim.houston@comcast.net>
Cc: Pierre Peiffer <peifferp@gmail.com>
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 lib/idr.c | 57 +++++++++++++++++++++++++++++++++++++++++--------------
 1 file changed, 43 insertions(+), 14 deletions(-)

diff --git a/lib/idr.c b/lib/idr.c
index 21e12af1f231..3476f8203e97 100644
--- a/lib/idr.c
+++ b/lib/idr.c
@@ -52,6 +52,19 @@ static struct idr_layer *get_from_free_list(struct idr *idp)
 	return(p);
 }
 
+static void idr_layer_rcu_free(struct rcu_head *head)
+{
+	struct idr_layer *layer;
+
+	layer = container_of(head, struct idr_layer, rcu_head);
+	kmem_cache_free(idr_layer_cache, layer);
+}
+
+static inline void free_layer(struct idr_layer *p)
+{
+	call_rcu(&p->rcu_head, idr_layer_rcu_free);
+}
+
 /* only called when idp->lock is held */
 static void __move_to_free_list(struct idr *idp, struct idr_layer *p)
 {
@@ -331,6 +344,7 @@ static void sub_remove(struct idr *idp, int shift, int id)
 	struct idr_layer *p = idp->top;
 	struct idr_layer **pa[MAX_LEVEL];
 	struct idr_layer ***paa = &pa[0];
+	struct idr_layer *to_free;
 	int n;
 
 	*paa = NULL;
@@ -346,13 +360,18 @@ static void sub_remove(struct idr *idp, int shift, int id)
 	n = id & IDR_MASK;
 	if (likely(p != NULL && test_bit(n, &p->bitmap))){
 		__clear_bit(n, &p->bitmap);
-		p->ary[n] = NULL;
+		rcu_assign_pointer(p->ary[n], NULL);
+		to_free = NULL;
 		while(*paa && ! --((**paa)->count)){
-			move_to_free_list(idp, **paa);
+			if (to_free)
+				free_layer(to_free);
+			to_free = **paa;
 			**paa-- = NULL;
 		}
 		if (!*paa)
 			idp->layers = 0;
+		if (to_free)
+			free_layer(to_free);
 	} else
 		idr_remove_warning(id);
 }
@@ -365,22 +384,34 @@ static void sub_remove(struct idr *idp, int shift, int id)
 void idr_remove(struct idr *idp, int id)
 {
 	struct idr_layer *p;
+	struct idr_layer *to_free;
 
 	/* Mask off upper bits we don't use for the search. */
 	id &= MAX_ID_MASK;
 
 	sub_remove(idp, (idp->layers - 1) * IDR_BITS, id);
 	if (idp->top && idp->top->count == 1 && (idp->layers > 1) &&
-	    idp->top->ary[0]) {  // We can drop a layer
-
+	    idp->top->ary[0]) {
+		/*
+		 * Single child at leftmost slot: we can shrink the tree.
+		 * This level is not needed anymore since when layers are
+		 * inserted, they are inserted at the top of the existing
+		 * tree.
+		 */
+		to_free = idp->top;
 		p = idp->top->ary[0];
-		idp->top->bitmap = idp->top->count = 0;
-		move_to_free_list(idp, idp->top);
-		idp->top = p;
+		rcu_assign_pointer(idp->top, p);
 		--idp->layers;
+		to_free->bitmap = to_free->count = 0;
+		free_layer(to_free);
 	}
 	while (idp->id_free_cnt >= IDR_FREE_MAX) {
 		p = get_from_free_list(idp);
+		/*
+		 * Note: we don't call the rcu callback here, since the only
+		 * layers that fall into the freelist are those that have been
+		 * preallocated.
+		 */
 		kmem_cache_free(idr_layer_cache, p);
 	}
 	return;
@@ -421,15 +452,13 @@ void idr_remove_all(struct idr *idp)
 
 		id += 1 << n;
 		while (n < fls(id)) {
-			if (p) {
-				memset(p, 0, sizeof *p);
-				move_to_free_list(idp, p);
-			}
+			if (p)
+				free_layer(p);
 			n += IDR_BITS;
 			p = *--paa;
 		}
 	}
-	idp->top = NULL;
+	rcu_assign_pointer(idp->top, NULL);
 	idp->layers = 0;
 }
 EXPORT_SYMBOL(idr_remove_all);
@@ -546,7 +575,7 @@ EXPORT_SYMBOL(idr_for_each);
  * A -ENOENT return indicates that @id was not found.
  * A -EINVAL return indicates that @id was not within valid constraints.
  *
- * The caller must serialize vs idr_find(), idr_get_new(), and idr_remove().
+ * The caller must serialize with writers.
  */
 void *idr_replace(struct idr *idp, void *ptr, int id)
 {
@@ -572,7 +601,7 @@ void *idr_replace(struct idr *idp, void *ptr, int id)
 		return ERR_PTR(-ENOENT);
 
 	old_p = p->ary[n];
-	p->ary[n] = ptr;
+	rcu_assign_pointer(p->ary[n], ptr);
 
 	return old_p;
 }

From 983bfb7db303cfde56ae5bbf4e0f2f46e38c9576 Mon Sep 17 00:00:00 2001
From: Nadia Derbey <Nadia.Derbey@bull.net>
Date: Fri, 25 Jul 2008 01:48:03 -0700
Subject: [PATCH 0856/1171] ipc: call idr_find() without locking in ipc_lock()

Call idr_find() locklessly from ipc_lock(), since the idr tree is now RCU
protected.

Signed-off-by: Nadia Derbey <Nadia.Derbey@bull.net>
Acked-by: "Paul E. McKenney" <paulmck@us.ibm.com>
Cc: Manfred Spraul <manfred@colorfullife.com>
Cc: Jim Houston <jim.houston@comcast.net>
Cc: Pierre Peiffer <peifferp@gmail.com>
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 ipc/util.c | 9 ---------
 1 file changed, 9 deletions(-)

diff --git a/ipc/util.c b/ipc/util.c
index 3339177b336c..0f468c34e83c 100644
--- a/ipc/util.c
+++ b/ipc/util.c
@@ -688,10 +688,6 @@ void ipc64_perm_to_ipc_perm (struct ipc64_perm *in, struct ipc_perm *out)
  * Look for an id in the ipc ids idr and lock the associated ipc object.
  *
  * The ipc object is locked on exit.
- *
- * This is the routine that should be called when the rw_mutex is not already
- * held, i.e. idr tree not protected: it protects the idr tree in read mode
- * during the idr_find().
  */
 
 struct kern_ipc_perm *ipc_lock(struct ipc_ids *ids, int id)
@@ -699,18 +695,13 @@ struct kern_ipc_perm *ipc_lock(struct ipc_ids *ids, int id)
 	struct kern_ipc_perm *out;
 	int lid = ipcid_to_idx(id);
 
-	down_read(&ids->rw_mutex);
-
 	rcu_read_lock();
 	out = idr_find(&ids->ipcs_idr, lid);
 	if (out == NULL) {
 		rcu_read_unlock();
-		up_read(&ids->rw_mutex);
 		return ERR_PTR(-EINVAL);
 	}
 
-	up_read(&ids->rw_mutex);
-
 	spin_lock(&out->lock);
 	
 	/* ipc_rmid() may have already freed the ID while ipc_lock

From 00c2bf85d8febfcfddde63822043462b026134ff Mon Sep 17 00:00:00 2001
From: Nadia Derbey <Nadia.Derbey@bull.net>
Date: Fri, 25 Jul 2008 01:48:03 -0700
Subject: [PATCH 0857/1171] ipc: get rid of ipc_lock_down()

Remove the ipc_lock_down() routines: they used to call idr_find() locklessly
(given that the ipc ids lock was already held), so they are not needed
anymore.

Signed-off-by: Nadia Derbey <Nadia.Derbey@bull.net>
Acked-by: "Paul E. McKenney" <paulmck@us.ibm.com>
Cc: Manfred Spraul <manfred@colorfullife.com>
Cc: Jim Houston <jim.houston@comcast.net>
Cc: Pierre Peiffer <peifferp@gmail.com>
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 ipc/shm.c  | 21 +++------------------
 ipc/util.c | 52 +---------------------------------------------------
 ipc/util.h |  6 ------
 3 files changed, 4 insertions(+), 75 deletions(-)

diff --git a/ipc/shm.c b/ipc/shm.c
index a726aebce7d7..e77ec698cf40 100644
--- a/ipc/shm.c
+++ b/ipc/shm.c
@@ -111,24 +111,9 @@ void __init shm_init (void)
 				IPC_SHM_IDS, sysvipc_shm_proc_show);
 }
 
-/*
- * shm_lock_(check_)down routines are called in the paths where the rw_mutex
- * is held to protect access to the idr tree.
- */
-static inline struct shmid_kernel *shm_lock_down(struct ipc_namespace *ns,
-						int id)
-{
-	struct kern_ipc_perm *ipcp = ipc_lock_down(&shm_ids(ns), id);
-
-	if (IS_ERR(ipcp))
-		return (struct shmid_kernel *)ipcp;
-
-	return container_of(ipcp, struct shmid_kernel, shm_perm);
-}
-
 /*
  * shm_lock_(check_) routines are called in the paths where the rw_mutex
- * is not held.
+ * is not necessarily held.
  */
 static inline struct shmid_kernel *shm_lock(struct ipc_namespace *ns, int id)
 {
@@ -211,7 +196,7 @@ static void shm_close(struct vm_area_struct *vma)
 
 	down_write(&shm_ids(ns).rw_mutex);
 	/* remove from the list of attaches of the shm segment */
-	shp = shm_lock_down(ns, sfd->id);
+	shp = shm_lock(ns, sfd->id);
 	BUG_ON(IS_ERR(shp));
 	shp->shm_lprid = task_tgid_vnr(current);
 	shp->shm_dtim = get_seconds();
@@ -932,7 +917,7 @@ invalid:
 
 out_nattch:
 	down_write(&shm_ids(ns).rw_mutex);
-	shp = shm_lock_down(ns, shmid);
+	shp = shm_lock(ns, shmid);
 	BUG_ON(IS_ERR(shp));
 	shp->shm_nattch--;
 	if(shp->shm_nattch == 0 &&
diff --git a/ipc/util.c b/ipc/util.c
index 0f468c34e83c..49b3ea615dc5 100644
--- a/ipc/util.c
+++ b/ipc/util.c
@@ -716,56 +716,6 @@ struct kern_ipc_perm *ipc_lock(struct ipc_ids *ids, int id)
 	return out;
 }
 
-/**
- * ipc_lock_down - Lock an ipc structure with rw_sem held
- * @ids: IPC identifier set
- * @id: ipc id to look for
- *
- * Look for an id in the ipc ids idr and lock the associated ipc object.
- *
- * The ipc object is locked on exit.
- *
- * This is the routine that should be called when the rw_mutex is already
- * held, i.e. idr tree protected.
- */
-
-struct kern_ipc_perm *ipc_lock_down(struct ipc_ids *ids, int id)
-{
-	struct kern_ipc_perm *out;
-	int lid = ipcid_to_idx(id);
-
-	rcu_read_lock();
-	out = idr_find(&ids->ipcs_idr, lid);
-	if (out == NULL) {
-		rcu_read_unlock();
-		return ERR_PTR(-EINVAL);
-	}
-
-	spin_lock(&out->lock);
-
-	/*
-	 * No need to verify that the structure is still valid since the
-	 * rw_mutex is held.
-	 */
-	return out;
-}
-
-struct kern_ipc_perm *ipc_lock_check_down(struct ipc_ids *ids, int id)
-{
-	struct kern_ipc_perm *out;
-
-	out = ipc_lock_down(ids, id);
-	if (IS_ERR(out))
-		return out;
-
-	if (ipc_checkid(out, id)) {
-		ipc_unlock(out);
-		return ERR_PTR(-EIDRM);
-	}
-
-	return out;
-}
-
 struct kern_ipc_perm *ipc_lock_check(struct ipc_ids *ids, int id)
 {
 	struct kern_ipc_perm *out;
@@ -837,7 +787,7 @@ struct kern_ipc_perm *ipcctl_pre_down(struct ipc_ids *ids, int id, int cmd,
 	int err;
 
 	down_write(&ids->rw_mutex);
-	ipcp = ipc_lock_check_down(ids, id);
+	ipcp = ipc_lock_check(ids, id);
 	if (IS_ERR(ipcp)) {
 		err = PTR_ERR(ipcp);
 		goto out_up;
diff --git a/ipc/util.h b/ipc/util.h
index cdb966aebe07..3646b45a03c9 100644
--- a/ipc/util.h
+++ b/ipc/util.h
@@ -102,11 +102,6 @@ void* ipc_rcu_alloc(int size);
 void ipc_rcu_getref(void *ptr);
 void ipc_rcu_putref(void *ptr);
 
-/*
- * ipc_lock_down: called with rw_mutex held
- * ipc_lock: called without that lock held
- */
-struct kern_ipc_perm *ipc_lock_down(struct ipc_ids *, int);
 struct kern_ipc_perm *ipc_lock(struct ipc_ids *, int);
 
 void kernel_to_ipc64_perm(struct kern_ipc_perm *in, struct ipc64_perm *out);
@@ -155,7 +150,6 @@ static inline void ipc_unlock(struct kern_ipc_perm *perm)
 	rcu_read_unlock();
 }
 
-struct kern_ipc_perm *ipc_lock_check_down(struct ipc_ids *ids, int id);
 struct kern_ipc_perm *ipc_lock_check(struct ipc_ids *ids, int id);
 int ipcget(struct ipc_namespace *ns, struct ipc_ids *ids,
 			struct ipc_ops *ops, struct ipc_params *params);

From 4daa28f6d8f5cda8ea0f55048e3c8811c384cbdd Mon Sep 17 00:00:00 2001
From: Manfred Spraul <manfred@colorfullife.com>
Date: Fri, 25 Jul 2008 01:48:04 -0700
Subject: [PATCH 0858/1171] ipc/sem.c: convert undo structures to struct
 list_head

The undo structures contain two linked lists, the attached patch replaces
them with generic struct list_head lists.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Manfred Spraul <manfred@colorfullife.com>
Cc: Nadia Derbey <Nadia.Derbey@bull.net>
Cc: Pierre Peiffer <peifferp@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/sem.h |  12 ++--
 ipc/sem.c           | 159 ++++++++++++++++++++++++--------------------
 2 files changed, 93 insertions(+), 78 deletions(-)

diff --git a/include/linux/sem.h b/include/linux/sem.h
index c8eaad9e4b72..6a1af1b49a13 100644
--- a/include/linux/sem.h
+++ b/include/linux/sem.h
@@ -95,7 +95,7 @@ struct sem_array {
 	struct sem		*sem_base;	/* ptr to first semaphore in array */
 	struct sem_queue	*sem_pending;	/* pending operations to be processed */
 	struct sem_queue	**sem_pending_last; /* last pending operation */
-	struct sem_undo		*undo;		/* undo requests on this array */
+	struct list_head	list_id;	/* undo requests on this array */
 	unsigned long		sem_nsems;	/* no. of semaphores in array */
 };
 
@@ -118,8 +118,8 @@ struct sem_queue {
  * when the process exits.
  */
 struct sem_undo {
-	struct sem_undo *	proc_next;	/* next entry on this process */
-	struct sem_undo *	id_next;	/* next entry on this semaphore set */
+	struct list_head	list_proc;	/* per-process list: all undos from one process */
+	struct list_head	list_id;	/* per semaphore array list: all undos for one array */
 	int			semid;		/* semaphore set identifier */
 	short *			semadj;		/* array of adjustments, one per semaphore */
 };
@@ -128,9 +128,9 @@ struct sem_undo {
  * that may be shared among all a CLONE_SYSVSEM task group.
  */ 
 struct sem_undo_list {
-	atomic_t	refcnt;
-	spinlock_t	lock;
-	struct sem_undo	*proc_list;
+	atomic_t		refcnt;
+	spinlock_t		lock;
+	struct list_head	list_proc;
 };
 
 struct sysv_sem {
diff --git a/ipc/sem.c b/ipc/sem.c
index e9418df5ff3e..4f26c7157356 100644
--- a/ipc/sem.c
+++ b/ipc/sem.c
@@ -274,7 +274,7 @@ static int newary(struct ipc_namespace *ns, struct ipc_params *params)
 	sma->sem_base = (struct sem *) &sma[1];
 	/* sma->sem_pending = NULL; */
 	sma->sem_pending_last = &sma->sem_pending;
-	/* sma->undo = NULL; */
+	INIT_LIST_HEAD(&sma->list_id);
 	sma->sem_nsems = nsems;
 	sma->sem_ctime = get_seconds();
 	sem_unlock(sma);
@@ -536,7 +536,8 @@ static void freeary(struct ipc_namespace *ns, struct kern_ipc_perm *ipcp)
 	 * (They will be freed without any further action in exit_sem()
 	 * or during the next semop.)
 	 */
-	for (un = sma->undo; un; un = un->id_next)
+	assert_spin_locked(&sma->sem_perm.lock);
+	list_for_each_entry(un, &sma->list_id, list_id)
 		un->semid = -1;
 
 	/* Wake up all pending processes and let them fail with EIDRM. */
@@ -763,9 +764,12 @@ static int semctl_main(struct ipc_namespace *ns, int semid, int semnum,
 
 		for (i = 0; i < nsems; i++)
 			sma->sem_base[i].semval = sem_io[i];
-		for (un = sma->undo; un; un = un->id_next)
+
+		assert_spin_locked(&sma->sem_perm.lock);
+		list_for_each_entry(un, &sma->list_id, list_id) {
 			for (i = 0; i < nsems; i++)
 				un->semadj[i] = 0;
+		}
 		sma->sem_ctime = get_seconds();
 		/* maybe some queued-up processes were waiting for this */
 		update_queue(sma);
@@ -797,12 +801,15 @@ static int semctl_main(struct ipc_namespace *ns, int semid, int semnum,
 	{
 		int val = arg.val;
 		struct sem_undo *un;
+
 		err = -ERANGE;
 		if (val > SEMVMX || val < 0)
 			goto out_unlock;
 
-		for (un = sma->undo; un; un = un->id_next)
+		assert_spin_locked(&sma->sem_perm.lock);
+		list_for_each_entry(un, &sma->list_id, list_id)
 			un->semadj[semnum] = 0;
+
 		curr->semval = val;
 		curr->sempid = task_tgid_vnr(current);
 		sma->sem_ctime = get_seconds();
@@ -952,6 +959,8 @@ static inline int get_undo_list(struct sem_undo_list **undo_listp)
 			return -ENOMEM;
 		spin_lock_init(&undo_list->lock);
 		atomic_set(&undo_list->refcnt, 1);
+		INIT_LIST_HEAD(&undo_list->list_proc);
+
 		current->sysvsem.undo_list = undo_list;
 	}
 	*undo_listp = undo_list;
@@ -960,25 +969,30 @@ static inline int get_undo_list(struct sem_undo_list **undo_listp)
 
 static struct sem_undo *lookup_undo(struct sem_undo_list *ulp, int semid)
 {
-	struct sem_undo **last, *un;
+	struct sem_undo *walk, *tmp;
 
-	last = &ulp->proc_list;
-	un = *last;
-	while(un != NULL) {
-		if(un->semid==semid)
-			break;
-		if(un->semid==-1) {
-			*last=un->proc_next;
-			kfree(un);
-		} else {
-			last=&un->proc_next;
+	assert_spin_locked(&ulp->lock);
+	list_for_each_entry_safe(walk, tmp, &ulp->list_proc, list_proc) {
+		if (walk->semid == semid)
+			return walk;
+		if (walk->semid == -1) {
+			list_del(&walk->list_proc);
+			kfree(walk);
 		}
-		un=*last;
 	}
-	return un;
+	return NULL;
 }
 
-static struct sem_undo *find_undo(struct ipc_namespace *ns, int semid)
+/**
+ * find_alloc_undo - Lookup (and if not present create) undo array
+ * @ns: namespace
+ * @semid: semaphore array id
+ *
+ * The function looks up (and if not present creates) the undo structure.
+ * The size of the undo structure depends on the size of the semaphore
+ * array, thus the alloc path is not that straightforward.
+ */
+static struct sem_undo *find_alloc_undo(struct ipc_namespace *ns, int semid)
 {
 	struct sem_array *sma;
 	struct sem_undo_list *ulp;
@@ -997,6 +1011,7 @@ static struct sem_undo *find_undo(struct ipc_namespace *ns, int semid)
 		goto out;
 
 	/* no undo structure around - allocate one. */
+	/* step 1: figure out the size of the semaphore array */
 	sma = sem_lock_check(ns, semid);
 	if (IS_ERR(sma))
 		return ERR_PTR(PTR_ERR(sma));
@@ -1004,15 +1019,19 @@ static struct sem_undo *find_undo(struct ipc_namespace *ns, int semid)
 	nsems = sma->sem_nsems;
 	sem_getref_and_unlock(sma);
 
+	/* step 2: allocate new undo structure */
 	new = kzalloc(sizeof(struct sem_undo) + sizeof(short)*nsems, GFP_KERNEL);
 	if (!new) {
 		sem_putref(sma);
 		return ERR_PTR(-ENOMEM);
 	}
-	new->semadj = (short *) &new[1];
-	new->semid = semid;
 
+	/* step 3: Acquire the lock on the undo list pointer */
 	spin_lock(&ulp->lock);
+
+	/* step 4: check for races: someone else allocated the undo struct,
+	 *         semaphore array was destroyed.
+	 */
 	un = lookup_undo(ulp, semid);
 	if (un) {
 		spin_unlock(&ulp->lock);
@@ -1028,13 +1047,17 @@ static struct sem_undo *find_undo(struct ipc_namespace *ns, int semid)
 		un = ERR_PTR(-EIDRM);
 		goto out;
 	}
-	new->proc_next = ulp->proc_list;
-	ulp->proc_list = new;
-	new->id_next = sma->undo;
-	sma->undo = new;
+	/* step 5: initialize & link new undo structure */
+	new->semadj = (short *) &new[1];
+	new->semid = semid;
+	assert_spin_locked(&ulp->lock);
+	list_add(&new->list_proc, &ulp->list_proc);
+	assert_spin_locked(&sma->sem_perm.lock);
+	list_add(&new->list_id, &sma->list_id);
+
 	sem_unlock(sma);
-	un = new;
 	spin_unlock(&ulp->lock);
+	un = new;
 out:
 	return un;
 }
@@ -1090,9 +1113,8 @@ asmlinkage long sys_semtimedop(int semid, struct sembuf __user *tsops,
 			alter = 1;
 	}
 
-retry_undos:
 	if (undos) {
-		un = find_undo(ns, semid);
+		un = find_alloc_undo(ns, semid);
 		if (IS_ERR(un)) {
 			error = PTR_ERR(un);
 			goto out_free;
@@ -1107,14 +1129,14 @@ retry_undos:
 	}
 
 	/*
-	 * semid identifiers are not unique - find_undo may have
+	 * semid identifiers are not unique - find_alloc_undo may have
 	 * allocated an undo structure, it was invalidated by an RMID
-	 * and now a new array with received the same id. Check and retry.
+	 * and now a new array with received the same id. Check and fail.
 	 */
-	if (un && un->semid == -1) {
-		sem_unlock(sma);
-		goto retry_undos;
-	}
+	error = -EIDRM;
+	if (un && un->semid == -1)
+		goto out_unlock_free;
+
 	error = -EFBIG;
 	if (max >= sma->sem_nsems)
 		goto out_unlock_free;
@@ -1243,56 +1265,44 @@ int copy_semundo(unsigned long clone_flags, struct task_struct *tsk)
  */
 void exit_sem(struct task_struct *tsk)
 {
-	struct sem_undo_list *undo_list;
-	struct sem_undo *u, **up;
-	struct ipc_namespace *ns;
+	struct sem_undo_list *ulp;
+	struct sem_undo *un, *tmp;
 
-	undo_list = tsk->sysvsem.undo_list;
-	if (!undo_list)
+	ulp = tsk->sysvsem.undo_list;
+	if (!ulp)
 		return;
 	tsk->sysvsem.undo_list = NULL;
 
-	if (!atomic_dec_and_test(&undo_list->refcnt))
+	if (!atomic_dec_and_test(&ulp->refcnt))
 		return;
 
-	ns = tsk->nsproxy->ipc_ns;
-	/* There's no need to hold the semundo list lock, as current
-         * is the last task exiting for this undo list.
-	 */
-	for (up = &undo_list->proc_list; (u = *up); *up = u->proc_next, kfree(u)) {
+	spin_lock(&ulp->lock);
+
+	list_for_each_entry_safe(un, tmp, &ulp->list_proc, list_proc) {
 		struct sem_array *sma;
-		int nsems, i;
-		struct sem_undo *un, **unp;
-		int semid;
-	       
-		semid = u->semid;
+		int i;
 
-		if(semid == -1)
-			continue;
-		sma = sem_lock(ns, semid);
+		if (un->semid == -1)
+			goto free;
+
+		sma = sem_lock(tsk->nsproxy->ipc_ns, un->semid);
 		if (IS_ERR(sma))
-			continue;
+			goto free;
 
-		if (u->semid == -1)
-			goto next_entry;
+		if (un->semid == -1)
+			goto unlock_free;
 
-		BUG_ON(sem_checkid(sma, u->semid));
+		BUG_ON(sem_checkid(sma, un->semid));
 
-		/* remove u from the sma->undo list */
-		for (unp = &sma->undo; (un = *unp); unp = &un->id_next) {
-			if (u == un)
-				goto found;
-		}
-		printk ("exit_sem undo list error id=%d\n", u->semid);
-		goto next_entry;
-found:
-		*unp = un->id_next;
-		/* perform adjustments registered in u */
-		nsems = sma->sem_nsems;
-		for (i = 0; i < nsems; i++) {
+		/* remove un from sma->list_id */
+		assert_spin_locked(&sma->sem_perm.lock);
+		list_del(&un->list_id);
+
+		/* perform adjustments registered in un */
+		for (i = 0; i < sma->sem_nsems; i++) {
 			struct sem * semaphore = &sma->sem_base[i];
-			if (u->semadj[i]) {
-				semaphore->semval += u->semadj[i];
+			if (un->semadj[i]) {
+				semaphore->semval += un->semadj[i];
 				/*
 				 * Range checks of the new semaphore value,
 				 * not defined by sus:
@@ -1316,10 +1326,15 @@ found:
 		sma->sem_otime = get_seconds();
 		/* maybe some queued-up processes were waiting for this */
 		update_queue(sma);
-next_entry:
+unlock_free:
 		sem_unlock(sma);
+free:
+		assert_spin_locked(&ulp->lock);
+		list_del(&un->list_proc);
+		kfree(un);
 	}
-	kfree(undo_list);
+	spin_unlock(&ulp->lock);
+	kfree(ulp);
 }
 
 #ifdef CONFIG_PROC_FS

From 2c0c29d414087f3b021059673c20a7088f5f1fff Mon Sep 17 00:00:00 2001
From: Manfred Spraul <manfred@colorfullife.com>
Date: Fri, 25 Jul 2008 01:48:05 -0700
Subject: [PATCH 0859/1171] ipc/sem.c: remove unused entries from struct
 sem_queue

sem_queue.sma and sem_queue.id were never used, the attached patch removes
them.

Signed-off-by: Manfred Spraul <manfred@colorfullife.com>
Reviewed-by: Nadia Derbey <Nadia.Derbey@bull.net>
Cc: Pierre Peiffer <peifferp@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/sem.h | 2 --
 ipc/sem.c           | 2 --
 2 files changed, 4 deletions(-)

diff --git a/include/linux/sem.h b/include/linux/sem.h
index 6a1af1b49a13..87756ef1198e 100644
--- a/include/linux/sem.h
+++ b/include/linux/sem.h
@@ -107,8 +107,6 @@ struct sem_queue {
 	struct sem_undo *	undo;	 /* undo structure */
 	int    			pid;	 /* process id of requesting process */
 	int    			status;	 /* completion status of operation */
-	struct sem_array *	sma;	 /* semaphore array for operations */
-	int			id;	 /* internal sem id */
 	struct sembuf *		sops;	 /* array of pending operations */
 	int			nsops;	 /* number of operations */
 	int			alter;   /* does the operation alter the array? */
diff --git a/ipc/sem.c b/ipc/sem.c
index 4f26c7157356..d5ce4000ca17 100644
--- a/ipc/sem.c
+++ b/ipc/sem.c
@@ -1160,12 +1160,10 @@ asmlinkage long sys_semtimedop(int semid, struct sembuf __user *tsops,
 	 * task into the pending queue and go to sleep.
 	 */
 		
-	queue.sma = sma;
 	queue.sops = sops;
 	queue.nsops = nsops;
 	queue.undo = un;
 	queue.pid = task_tgid_vnr(current);
-	queue.id = semid;
 	queue.alter = alter;
 	if (alter)
 		append_to_queue(sma ,&queue);

From a1193f8ec091cd8fd309cc2982abe4499f6f2b4d Mon Sep 17 00:00:00 2001
From: Manfred Spraul <manfred@colorfullife.com>
Date: Fri, 25 Jul 2008 01:48:06 -0700
Subject: [PATCH 0860/1171] ipc/sem.c: convert sem_array.sem_pending to struct
 list_head

sem_array.sem_pending is a double linked list, the attached patch converts
it to struct list_head.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Manfred Spraul <manfred@colorfullife.com>
Reviewed-by: Nadia Derbey <Nadia.Derbey@bull.net>
Cc: Pierre Peiffer <peifferp@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/sem.h | 12 +++---
 ipc/sem.c           | 92 +++++++++++++++++----------------------------
 2 files changed, 40 insertions(+), 64 deletions(-)

diff --git a/include/linux/sem.h b/include/linux/sem.h
index 87756ef1198e..d42599395d79 100644
--- a/include/linux/sem.h
+++ b/include/linux/sem.h
@@ -93,21 +93,19 @@ struct sem_array {
 	time_t			sem_otime;	/* last semop time */
 	time_t			sem_ctime;	/* last change time */
 	struct sem		*sem_base;	/* ptr to first semaphore in array */
-	struct sem_queue	*sem_pending;	/* pending operations to be processed */
-	struct sem_queue	**sem_pending_last; /* last pending operation */
+	struct list_head	sem_pending;	/* pending operations to be processed */
 	struct list_head	list_id;	/* undo requests on this array */
 	unsigned long		sem_nsems;	/* no. of semaphores in array */
 };
 
 /* One queue for each sleeping process in the system. */
 struct sem_queue {
-	struct sem_queue *	next;	 /* next entry in the queue */
-	struct sem_queue **	prev;	 /* previous entry in the queue, *(q->prev) == q */
-	struct task_struct*	sleeper; /* this process */
-	struct sem_undo *	undo;	 /* undo structure */
+	struct list_head	list;	 /* queue of pending operations */
+	struct task_struct	*sleeper; /* this process */
+	struct sem_undo		*undo;	 /* undo structure */
 	int    			pid;	 /* process id of requesting process */
 	int    			status;	 /* completion status of operation */
-	struct sembuf *		sops;	 /* array of pending operations */
+	struct sembuf		*sops;	 /* array of pending operations */
 	int			nsops;	 /* number of operations */
 	int			alter;   /* does the operation alter the array? */
 };
diff --git a/ipc/sem.c b/ipc/sem.c
index d5ce4000ca17..3ca232736b31 100644
--- a/ipc/sem.c
+++ b/ipc/sem.c
@@ -272,8 +272,7 @@ static int newary(struct ipc_namespace *ns, struct ipc_params *params)
 	ns->used_sems += nsems;
 
 	sma->sem_base = (struct sem *) &sma[1];
-	/* sma->sem_pending = NULL; */
-	sma->sem_pending_last = &sma->sem_pending;
+	INIT_LIST_HEAD(&sma->sem_pending);
 	INIT_LIST_HEAD(&sma->list_id);
 	sma->sem_nsems = nsems;
 	sma->sem_ctime = get_seconds();
@@ -331,38 +330,6 @@ asmlinkage long sys_semget(key_t key, int nsems, int semflg)
 	return ipcget(ns, &sem_ids(ns), &sem_ops, &sem_params);
 }
 
-/* Manage the doubly linked list sma->sem_pending as a FIFO:
- * insert new queue elements at the tail sma->sem_pending_last.
- */
-static inline void append_to_queue (struct sem_array * sma,
-				    struct sem_queue * q)
-{
-	*(q->prev = sma->sem_pending_last) = q;
-	*(sma->sem_pending_last = &q->next) = NULL;
-}
-
-static inline void prepend_to_queue (struct sem_array * sma,
-				     struct sem_queue * q)
-{
-	q->next = sma->sem_pending;
-	*(q->prev = &sma->sem_pending) = q;
-	if (q->next)
-		q->next->prev = &q->next;
-	else /* sma->sem_pending_last == &sma->sem_pending */
-		sma->sem_pending_last = &q->next;
-}
-
-static inline void remove_from_queue (struct sem_array * sma,
-				      struct sem_queue * q)
-{
-	*(q->prev) = q->next;
-	if (q->next)
-		q->next->prev = q->prev;
-	else /* sma->sem_pending_last == &q->next */
-		sma->sem_pending_last = q->prev;
-	q->prev = NULL; /* mark as removed */
-}
-
 /*
  * Determine whether a sequence of semaphore operations would succeed
  * all at once. Return 0 if yes, 1 if need to sleep, else return error code.
@@ -438,16 +405,15 @@ static void update_queue (struct sem_array * sma)
 	int error;
 	struct sem_queue * q;
 
-	q = sma->sem_pending;
-	while(q) {
+	q = list_entry(sma->sem_pending.next, struct sem_queue, list);
+	while (&q->list != &sma->sem_pending) {
 		error = try_atomic_semop(sma, q->sops, q->nsops,
 					 q->undo, q->pid);
 
 		/* Does q->sleeper still need to sleep? */
 		if (error <= 0) {
 			struct sem_queue *n;
-			remove_from_queue(sma,q);
-			q->status = IN_WAKEUP;
+
 			/*
 			 * Continue scanning. The next operation
 			 * that must be checked depends on the type of the
@@ -458,11 +424,26 @@ static void update_queue (struct sem_array * sma)
 			 *   for semaphore values to become 0.
 			 * - if the operation didn't modify the array,
 			 *   then just continue.
+			 * The order of list_del() and reading ->next
+			 * is crucial: In the former case, the list_del()
+			 * must be done first [because we might be the
+			 * first entry in ->sem_pending], in the latter
+			 * case the list_del() must be done last
+			 * [because the list is invalid after the list_del()]
 			 */
-			if (q->alter)
-				n = sma->sem_pending;
-			else
-				n = q->next;
+			if (q->alter) {
+				list_del(&q->list);
+				n = list_entry(sma->sem_pending.next,
+						struct sem_queue, list);
+			} else {
+				n = list_entry(q->list.next, struct sem_queue,
+						list);
+				list_del(&q->list);
+			}
+
+			/* wake up the waiting thread */
+			q->status = IN_WAKEUP;
+
 			wake_up_process(q->sleeper);
 			/* hands-off: q will disappear immediately after
 			 * writing q->status.
@@ -471,7 +452,7 @@ static void update_queue (struct sem_array * sma)
 			q->status = error;
 			q = n;
 		} else {
-			q = q->next;
+			q = list_entry(q->list.next, struct sem_queue, list);
 		}
 	}
 }
@@ -491,7 +472,7 @@ static int count_semncnt (struct sem_array * sma, ushort semnum)
 	struct sem_queue * q;
 
 	semncnt = 0;
-	for (q = sma->sem_pending; q; q = q->next) {
+	list_for_each_entry(q, &sma->sem_pending, list) {
 		struct sembuf * sops = q->sops;
 		int nsops = q->nsops;
 		int i;
@@ -503,13 +484,14 @@ static int count_semncnt (struct sem_array * sma, ushort semnum)
 	}
 	return semncnt;
 }
+
 static int count_semzcnt (struct sem_array * sma, ushort semnum)
 {
 	int semzcnt;
 	struct sem_queue * q;
 
 	semzcnt = 0;
-	for (q = sma->sem_pending; q; q = q->next) {
+	list_for_each_entry(q, &sma->sem_pending, list) {
 		struct sembuf * sops = q->sops;
 		int nsops = q->nsops;
 		int i;
@@ -529,7 +511,7 @@ static int count_semzcnt (struct sem_array * sma, ushort semnum)
 static void freeary(struct ipc_namespace *ns, struct kern_ipc_perm *ipcp)
 {
 	struct sem_undo *un;
-	struct sem_queue *q;
+	struct sem_queue *q, *t;
 	struct sem_array *sma = container_of(ipcp, struct sem_array, sem_perm);
 
 	/* Invalidate the existing undo structures for this semaphore set.
@@ -541,17 +523,14 @@ static void freeary(struct ipc_namespace *ns, struct kern_ipc_perm *ipcp)
 		un->semid = -1;
 
 	/* Wake up all pending processes and let them fail with EIDRM. */
-	q = sma->sem_pending;
-	while(q) {
-		struct sem_queue *n;
-		/* lazy remove_from_queue: we are killing the whole queue */
-		q->prev = NULL;
-		n = q->next;
+
+	list_for_each_entry_safe(q, t, &sma->sem_pending, list) {
+		list_del(&q->list);
+
 		q->status = IN_WAKEUP;
 		wake_up_process(q->sleeper); /* doesn't sleep */
 		smp_wmb();
 		q->status = -EIDRM;	/* hands-off q */
-		q = n;
 	}
 
 	/* Remove the semaphore set from the IDR */
@@ -1166,9 +1145,9 @@ asmlinkage long sys_semtimedop(int semid, struct sembuf __user *tsops,
 	queue.pid = task_tgid_vnr(current);
 	queue.alter = alter;
 	if (alter)
-		append_to_queue(sma ,&queue);
+		list_add_tail(&queue.list, &sma->sem_pending);
 	else
-		prepend_to_queue(sma ,&queue);
+		list_add(&queue.list, &sma->sem_pending);
 
 	queue.status = -EINTR;
 	queue.sleeper = current;
@@ -1194,7 +1173,6 @@ asmlinkage long sys_semtimedop(int semid, struct sembuf __user *tsops,
 
 	sma = sem_lock(ns, semid);
 	if (IS_ERR(sma)) {
-		BUG_ON(queue.prev != NULL);
 		error = -EIDRM;
 		goto out_free;
 	}
@@ -1212,7 +1190,7 @@ asmlinkage long sys_semtimedop(int semid, struct sembuf __user *tsops,
 	 */
 	if (timeout && jiffies_left == 0)
 		error = -EAGAIN;
-	remove_from_queue(sma,&queue);
+	list_del(&queue.list);
 	goto out_unlock_free;
 
 out_unlock_free:

From 380af1b33b3ff92df5cda96329b58f5d1b6b5a53 Mon Sep 17 00:00:00 2001
From: Manfred Spraul <manfred@colorfullife.com>
Date: Fri, 25 Jul 2008 01:48:06 -0700
Subject: [PATCH 0861/1171] ipc/sem.c: rewrite undo list locking

The attached patch:
- reverses the locking order of ulp->lock and sem_lock:
  Previously, it was first ulp->lock, then inside sem_lock.
  Now it's the other way around.
- converts the undo structure to rcu.

Benefits:
- With the old locking order, IPC_RMID could not kfree the undo structures.
  The stale entries remained in the linked lists and were released later.
- The patch fixes a a race in semtimedop(): if both IPC_RMID and a semget() that
  recreates exactly the same id happen between find_alloc_undo() and sem_lock,
  then semtimedop() would access already kfree'd memory.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Manfred Spraul <manfred@colorfullife.com>
Reviewed-by: Nadia Derbey <Nadia.Derbey@bull.net>
Cc: Pierre Peiffer <peifferp@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/sem.h |   6 +-
 ipc/sem.c           | 149 ++++++++++++++++++++++++++++----------------
 2 files changed, 99 insertions(+), 56 deletions(-)

diff --git a/include/linux/sem.h b/include/linux/sem.h
index d42599395d79..1b191c176bcd 100644
--- a/include/linux/sem.h
+++ b/include/linux/sem.h
@@ -78,6 +78,7 @@ struct  seminfo {
 
 #ifdef __KERNEL__
 #include <asm/atomic.h>
+#include <linux/rcupdate.h>
 
 struct task_struct;
 
@@ -114,7 +115,10 @@ struct sem_queue {
  * when the process exits.
  */
 struct sem_undo {
-	struct list_head	list_proc;	/* per-process list: all undos from one process */
+	struct list_head	list_proc;	/* per-process list: all undos from one process. */
+						/* rcu protected */
+	struct rcu_head		rcu;		/* rcu struct for sem_undo() */
+	struct sem_undo_list	*ulp;		/* sem_undo_list for the process */
 	struct list_head	list_id;	/* per semaphore array list: all undos for one array */
 	int			semid;		/* semaphore set identifier */
 	short *			semadj;		/* array of adjustments, one per semaphore */
diff --git a/ipc/sem.c b/ipc/sem.c
index 3ca232736b31..bf1bc36cb7ee 100644
--- a/ipc/sem.c
+++ b/ipc/sem.c
@@ -504,27 +504,35 @@ static int count_semzcnt (struct sem_array * sma, ushort semnum)
 	return semzcnt;
 }
 
+void free_un(struct rcu_head *head)
+{
+	struct sem_undo *un = container_of(head, struct sem_undo, rcu);
+	kfree(un);
+}
+
 /* Free a semaphore set. freeary() is called with sem_ids.rw_mutex locked
  * as a writer and the spinlock for this semaphore set hold. sem_ids.rw_mutex
  * remains locked on exit.
  */
 static void freeary(struct ipc_namespace *ns, struct kern_ipc_perm *ipcp)
 {
-	struct sem_undo *un;
-	struct sem_queue *q, *t;
+	struct sem_undo *un, *tu;
+	struct sem_queue *q, *tq;
 	struct sem_array *sma = container_of(ipcp, struct sem_array, sem_perm);
 
-	/* Invalidate the existing undo structures for this semaphore set.
-	 * (They will be freed without any further action in exit_sem()
-	 * or during the next semop.)
-	 */
+	/* Free the existing undo structures for this semaphore set.  */
 	assert_spin_locked(&sma->sem_perm.lock);
-	list_for_each_entry(un, &sma->list_id, list_id)
+	list_for_each_entry_safe(un, tu, &sma->list_id, list_id) {
+		list_del(&un->list_id);
+		spin_lock(&un->ulp->lock);
 		un->semid = -1;
+		list_del_rcu(&un->list_proc);
+		spin_unlock(&un->ulp->lock);
+		call_rcu(&un->rcu, free_un);
+	}
 
 	/* Wake up all pending processes and let them fail with EIDRM. */
-
-	list_for_each_entry_safe(q, t, &sma->sem_pending, list) {
+	list_for_each_entry_safe(q, tq, &sma->sem_pending, list) {
 		list_del(&q->list);
 
 		q->status = IN_WAKEUP;
@@ -948,16 +956,11 @@ static inline int get_undo_list(struct sem_undo_list **undo_listp)
 
 static struct sem_undo *lookup_undo(struct sem_undo_list *ulp, int semid)
 {
-	struct sem_undo *walk, *tmp;
+	struct sem_undo *walk;
 
-	assert_spin_locked(&ulp->lock);
-	list_for_each_entry_safe(walk, tmp, &ulp->list_proc, list_proc) {
+	list_for_each_entry_rcu(walk, &ulp->list_proc, list_proc) {
 		if (walk->semid == semid)
 			return walk;
-		if (walk->semid == -1) {
-			list_del(&walk->list_proc);
-			kfree(walk);
-		}
 	}
 	return NULL;
 }
@@ -970,6 +973,8 @@ static struct sem_undo *lookup_undo(struct sem_undo_list *ulp, int semid)
  * The function looks up (and if not present creates) the undo structure.
  * The size of the undo structure depends on the size of the semaphore
  * array, thus the alloc path is not that straightforward.
+ * Lifetime-rules: sem_undo is rcu-protected, on success, the function
+ * performs a rcu_read_lock().
  */
 static struct sem_undo *find_alloc_undo(struct ipc_namespace *ns, int semid)
 {
@@ -983,11 +988,13 @@ static struct sem_undo *find_alloc_undo(struct ipc_namespace *ns, int semid)
 	if (error)
 		return ERR_PTR(error);
 
+	rcu_read_lock();
 	spin_lock(&ulp->lock);
 	un = lookup_undo(ulp, semid);
 	spin_unlock(&ulp->lock);
 	if (likely(un!=NULL))
 		goto out;
+	rcu_read_unlock();
 
 	/* no undo structure around - allocate one. */
 	/* step 1: figure out the size of the semaphore array */
@@ -1005,38 +1012,38 @@ static struct sem_undo *find_alloc_undo(struct ipc_namespace *ns, int semid)
 		return ERR_PTR(-ENOMEM);
 	}
 
-	/* step 3: Acquire the lock on the undo list pointer */
-	spin_lock(&ulp->lock);
-
-	/* step 4: check for races: someone else allocated the undo struct,
-	 *         semaphore array was destroyed.
-	 */
-	un = lookup_undo(ulp, semid);
-	if (un) {
-		spin_unlock(&ulp->lock);
-		kfree(new);
-		sem_putref(sma);
-		goto out;
-	}
+	/* step 3: Acquire the lock on semaphore array */
 	sem_lock_and_putref(sma);
 	if (sma->sem_perm.deleted) {
 		sem_unlock(sma);
-		spin_unlock(&ulp->lock);
 		kfree(new);
 		un = ERR_PTR(-EIDRM);
 		goto out;
 	}
+	spin_lock(&ulp->lock);
+
+	/*
+	 * step 4: check for races: did someone else allocate the undo struct?
+	 */
+	un = lookup_undo(ulp, semid);
+	if (un) {
+		kfree(new);
+		goto success;
+	}
 	/* step 5: initialize & link new undo structure */
 	new->semadj = (short *) &new[1];
+	new->ulp = ulp;
 	new->semid = semid;
 	assert_spin_locked(&ulp->lock);
-	list_add(&new->list_proc, &ulp->list_proc);
+	list_add_rcu(&new->list_proc, &ulp->list_proc);
 	assert_spin_locked(&sma->sem_perm.lock);
 	list_add(&new->list_id, &sma->list_id);
-
-	sem_unlock(sma);
-	spin_unlock(&ulp->lock);
 	un = new;
+
+success:
+	spin_unlock(&ulp->lock);
+	rcu_read_lock();
+	sem_unlock(sma);
 out:
 	return un;
 }
@@ -1103,6 +1110,8 @@ asmlinkage long sys_semtimedop(int semid, struct sembuf __user *tsops,
 
 	sma = sem_lock_check(ns, semid);
 	if (IS_ERR(sma)) {
+		if (un)
+			rcu_read_unlock();
 		error = PTR_ERR(sma);
 		goto out_free;
 	}
@@ -1111,10 +1120,26 @@ asmlinkage long sys_semtimedop(int semid, struct sembuf __user *tsops,
 	 * semid identifiers are not unique - find_alloc_undo may have
 	 * allocated an undo structure, it was invalidated by an RMID
 	 * and now a new array with received the same id. Check and fail.
+	 * This case can be detected checking un->semid. The existance of
+	 * "un" itself is guaranteed by rcu.
 	 */
 	error = -EIDRM;
-	if (un && un->semid == -1)
-		goto out_unlock_free;
+	if (un) {
+		if (un->semid == -1) {
+			rcu_read_unlock();
+			goto out_unlock_free;
+		} else {
+			/*
+			 * rcu lock can be released, "un" cannot disappear:
+			 * - sem_lock is acquired, thus IPC_RMID is
+			 *   impossible.
+			 * - exit_sem is impossible, it always operates on
+			 *   current (or a dead task).
+			 */
+
+			rcu_read_unlock();
+		}
+	}
 
 	error = -EFBIG;
 	if (max >= sma->sem_nsems)
@@ -1242,7 +1267,6 @@ int copy_semundo(unsigned long clone_flags, struct task_struct *tsk)
 void exit_sem(struct task_struct *tsk)
 {
 	struct sem_undo_list *ulp;
-	struct sem_undo *un, *tmp;
 
 	ulp = tsk->sysvsem.undo_list;
 	if (!ulp)
@@ -1252,28 +1276,47 @@ void exit_sem(struct task_struct *tsk)
 	if (!atomic_dec_and_test(&ulp->refcnt))
 		return;
 
-	spin_lock(&ulp->lock);
-
-	list_for_each_entry_safe(un, tmp, &ulp->list_proc, list_proc) {
+	for (;;) {
 		struct sem_array *sma;
+		struct sem_undo *un;
+		int semid;
 		int i;
 
-		if (un->semid == -1)
-			goto free;
+		rcu_read_lock();
+		un = list_entry(rcu_dereference(ulp->list_proc.next),
+					struct sem_undo, list_proc);
+		if (&un->list_proc == &ulp->list_proc)
+			semid = -1;
+		 else
+			semid = un->semid;
+		rcu_read_unlock();
 
-		sma = sem_lock(tsk->nsproxy->ipc_ns, un->semid);
+		if (semid == -1)
+			break;
+
+		sma = sem_lock_check(tsk->nsproxy->ipc_ns, un->semid);
+
+		/* exit_sem raced with IPC_RMID, nothing to do */
 		if (IS_ERR(sma))
-			goto free;
+			continue;
 
-		if (un->semid == -1)
-			goto unlock_free;
+		un = lookup_undo(ulp, semid);
+		if (un == NULL) {
+			/* exit_sem raced with IPC_RMID+semget() that created
+			 * exactly the same semid. Nothing to do.
+			 */
+			sem_unlock(sma);
+			continue;
+		}
 
-		BUG_ON(sem_checkid(sma, un->semid));
-
-		/* remove un from sma->list_id */
+		/* remove un from the linked lists */
 		assert_spin_locked(&sma->sem_perm.lock);
 		list_del(&un->list_id);
 
+		spin_lock(&ulp->lock);
+		list_del_rcu(&un->list_proc);
+		spin_unlock(&ulp->lock);
+
 		/* perform adjustments registered in un */
 		for (i = 0; i < sma->sem_nsems; i++) {
 			struct sem * semaphore = &sma->sem_base[i];
@@ -1302,14 +1345,10 @@ void exit_sem(struct task_struct *tsk)
 		sma->sem_otime = get_seconds();
 		/* maybe some queued-up processes were waiting for this */
 		update_queue(sma);
-unlock_free:
 		sem_unlock(sma);
-free:
-		assert_spin_locked(&ulp->lock);
-		list_del(&un->list_proc);
-		kfree(un);
+
+		call_rcu(&un->rcu, free_un);
 	}
-	spin_unlock(&ulp->lock);
 	kfree(ulp);
 }
 

From f1a43f93f0f3bab418800eaccb9e2e3b5427e173 Mon Sep 17 00:00:00 2001
From: Akinobu Mita <akinobu.mita@gmail.com>
Date: Fri, 25 Jul 2008 01:48:07 -0700
Subject: [PATCH 0862/1171] ipc: use simple_read_from_buffer()

Also this patch kills unneccesary trailing NULL character.

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Nadia Derbey <Nadia.Derbey@bull.net>
Cc: Manfred Spraul <manfred@colorfullife.com>
Cc: Pierre Peiffer <peifferp@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 ipc/mqueue.c | 25 +++++++------------------
 1 file changed, 7 insertions(+), 18 deletions(-)

diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index 3e84b958186b..1fdc2eb2f6d8 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -314,15 +314,11 @@ static int mqueue_unlink(struct inode *dir, struct dentry *dentry)
 *	through std routines)
 */
 static ssize_t mqueue_read_file(struct file *filp, char __user *u_data,
-				size_t count, loff_t * off)
+				size_t count, loff_t *off)
 {
 	struct mqueue_inode_info *info = MQUEUE_I(filp->f_path.dentry->d_inode);
 	char buffer[FILENT_SIZE];
-	size_t slen;
-	loff_t o;
-
-	if (!count)
-		return 0;
+	ssize_t ret;
 
 	spin_lock(&info->lock);
 	snprintf(buffer, sizeof(buffer),
@@ -335,21 +331,14 @@ static ssize_t mqueue_read_file(struct file *filp, char __user *u_data,
 			pid_vnr(info->notify_owner));
 	spin_unlock(&info->lock);
 	buffer[sizeof(buffer)-1] = '\0';
-	slen = strlen(buffer)+1;
 
-	o = *off;
-	if (o > slen)
-		return 0;
+	ret = simple_read_from_buffer(u_data, count, off, buffer,
+				strlen(buffer));
+	if (ret <= 0)
+		return ret;
 
-	if (o + count > slen)
-		count = slen - o;
-
-	if (copy_to_user(u_data, buffer + o, count))
-		return -EFAULT;
-
-	*off = o + count;
 	filp->f_path.dentry->d_inode->i_atime = filp->f_path.dentry->d_inode->i_ctime = CURRENT_TIME;
-	return count;
+	return ret;
 }
 
 static int mqueue_flush_file(struct file *filp, fl_owner_t id)

From 9eefe520c814f6f62c5d36a2ddcd3fb99dfdb30e Mon Sep 17 00:00:00 2001
From: Nadia Derbey <Nadia.Derbey@bull.net>
Date: Fri, 25 Jul 2008 01:48:08 -0700
Subject: [PATCH 0863/1171] ipc: do not use a negative value to re-enable
 msgmni automatic recomputing

This patch proposes an alternative to the "magical
positive-versus-negative number trick" Andrew complained about last week
in http://lkml.org/lkml/2008/6/24/418.

This had been introduced with the patches that scale msgmni to the amount
of lowmem.  With these patches, msgmni has a registered notification
routine that recomputes msgmni value upon memory add/remove or ipc
namespace creation/ removal.

When msgmni is changed from user space (i.e.  value written to the proc
file), that notification routine is unregistered, and the way to make it
registered back is to write a negative value into the proc file.  This is
the "magical positive-versus-negative number trick".

To fix this, a new proc file is introduced: /proc/sys/kernel/auto_msgmni.
This file acts as ON/OFF for msgmni automatic recomputing.

With this patch, the process is the following:
1) kernel boots in "automatic recomputing mode"
   /proc/sys/kernel/msgmni contains the value that has been computed (depends
                           on lowmem)
   /proc/sys/kernel/automatic_msgmni contains "1"

2) echo <val> > /proc/sys/kernel/msgmni
   . sets msg_ctlmni to <val>
   . de-activates automatic recomputing (i.e. if, say, some memory is added
     msgmni won't be recomputed anymore)
   . /proc/sys/kernel/automatic_msgmni now contains "0"

3) echo "0" > /proc/sys/kernel/automatic_msgmni
   . de-activates msgmni automatic recomputing
     this has the same effect as 2) except that msg_ctlmni's value stays
     blocked at its current value)

3) echo "1" > /proc/sys/kernel/automatic_msgmni
   . recomputes msgmni's value based on the current available memory size
     and number of ipc namespaces
   . re-activates automatic recomputing for msgmni.

Signed-off-by: Nadia Derbey <Nadia.Derbey@bull.net>
Cc: Solofo Ramangalahy <Solofo.Ramangalahy@bull.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/ipc_namespace.h |  3 +-
 ipc/ipc_sysctl.c              | 72 ++++++++++++++++++++++++++++-------
 ipc/ipcns_notifier.c          | 20 +++++++---
 3 files changed, 76 insertions(+), 19 deletions(-)

diff --git a/include/linux/ipc_namespace.h b/include/linux/ipc_namespace.h
index ea6c18a8b0d4..ea330f9e7100 100644
--- a/include/linux/ipc_namespace.h
+++ b/include/linux/ipc_namespace.h
@@ -36,6 +36,7 @@ struct ipc_namespace {
 	int		msg_ctlmni;
 	atomic_t	msg_bytes;
 	atomic_t	msg_hdrs;
+	int		auto_msgmni;
 
 	size_t		shm_ctlmax;
 	size_t		shm_ctlall;
@@ -53,7 +54,7 @@ extern atomic_t nr_ipc_ns;
 
 extern int register_ipcns_notifier(struct ipc_namespace *);
 extern int cond_register_ipcns_notifier(struct ipc_namespace *);
-extern int unregister_ipcns_notifier(struct ipc_namespace *);
+extern void unregister_ipcns_notifier(struct ipc_namespace *);
 extern int ipcns_notify(unsigned long);
 
 #else /* CONFIG_SYSVIPC */
diff --git a/ipc/ipc_sysctl.c b/ipc/ipc_sysctl.c
index d3497465cc0a..69bc85978ba0 100644
--- a/ipc/ipc_sysctl.c
+++ b/ipc/ipc_sysctl.c
@@ -27,15 +27,17 @@ static void *get_ipc(ctl_table *table)
 }
 
 /*
- * Routine that is called when a tunable has successfully been changed by
- * hand and it has a callback routine registered on the ipc namespace notifier
- * chain: we don't want such tunables to be recomputed anymore upon memory
- * add/remove or ipc namespace creation/removal.
- * They can come back to a recomputable state by being set to a <0 value.
+ * Routine that is called when the file "auto_msgmni" has successfully been
+ * written.
+ * Two values are allowed:
+ * 0: unregister msgmni's callback routine from the ipc namespace notifier
+ *    chain. This means that msgmni won't be recomputed anymore upon memory
+ *    add/remove or ipc namespace creation/removal.
+ * 1: register back the callback routine.
  */
-static void tunable_set_callback(int val)
+static void ipc_auto_callback(int val)
 {
-	if (val >= 0)
+	if (!val)
 		unregister_ipcns_notifier(current->nsproxy->ipc_ns);
 	else {
 		/*
@@ -71,7 +73,12 @@ static int proc_ipc_callback_dointvec(ctl_table *table, int write,
 	rc = proc_dointvec(&ipc_table, write, filp, buffer, lenp, ppos);
 
 	if (write && !rc && lenp_bef == *lenp)
-		tunable_set_callback(*((int *)(ipc_table.data)));
+		/*
+		 * Tunable has successfully been changed by hand. Disable its
+		 * automatic adjustment. This simply requires unregistering
+		 * the notifiers that trigger recalculation.
+		 */
+		unregister_ipcns_notifier(current->nsproxy->ipc_ns);
 
 	return rc;
 }
@@ -87,10 +94,39 @@ static int proc_ipc_doulongvec_minmax(ctl_table *table, int write,
 					lenp, ppos);
 }
 
+static int proc_ipcauto_dointvec_minmax(ctl_table *table, int write,
+	struct file *filp, void __user *buffer, size_t *lenp, loff_t *ppos)
+{
+	struct ctl_table ipc_table;
+	size_t lenp_bef = *lenp;
+	int oldval;
+	int rc;
+
+	memcpy(&ipc_table, table, sizeof(ipc_table));
+	ipc_table.data = get_ipc(table);
+	oldval = *((int *)(ipc_table.data));
+
+	rc = proc_dointvec_minmax(&ipc_table, write, filp, buffer, lenp, ppos);
+
+	if (write && !rc && lenp_bef == *lenp) {
+		int newval = *((int *)(ipc_table.data));
+		/*
+		 * The file "auto_msgmni" has correctly been set.
+		 * React by (un)registering the corresponding tunable, if the
+		 * value has changed.
+		 */
+		if (newval != oldval)
+			ipc_auto_callback(newval);
+	}
+
+	return rc;
+}
+
 #else
 #define proc_ipc_doulongvec_minmax NULL
 #define proc_ipc_dointvec	   NULL
 #define proc_ipc_callback_dointvec NULL
+#define proc_ipcauto_dointvec_minmax NULL
 #endif
 
 #ifdef CONFIG_SYSCTL_SYSCALL
@@ -142,14 +178,11 @@ static int sysctl_ipc_registered_data(ctl_table *table, int __user *name,
 	rc = sysctl_ipc_data(table, name, nlen, oldval, oldlenp, newval,
 		newlen);
 
-	if (newval && newlen && rc > 0) {
+	if (newval && newlen && rc > 0)
 		/*
 		 * Tunable has successfully been changed from userland
 		 */
-		int *data = get_ipc(table);
-
-		tunable_set_callback(*data);
-	}
+		unregister_ipcns_notifier(current->nsproxy->ipc_ns);
 
 	return rc;
 }
@@ -158,6 +191,9 @@ static int sysctl_ipc_registered_data(ctl_table *table, int __user *name,
 #define sysctl_ipc_registered_data NULL
 #endif
 
+static int zero;
+static int one = 1;
+
 static struct ctl_table ipc_kern_table[] = {
 	{
 		.ctl_name	= KERN_SHMMAX,
@@ -222,6 +258,16 @@ static struct ctl_table ipc_kern_table[] = {
 		.proc_handler	= proc_ipc_dointvec,
 		.strategy	= sysctl_ipc_data,
 	},
+	{
+		.ctl_name	= CTL_UNNUMBERED,
+		.procname	= "auto_msgmni",
+		.data		= &init_ipc_ns.auto_msgmni,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_ipcauto_dointvec_minmax,
+		.extra1		= &zero,
+		.extra2		= &one,
+	},
 	{}
 };
 
diff --git a/ipc/ipcns_notifier.c b/ipc/ipcns_notifier.c
index 70ff09183f7b..b9b31a4f77e1 100644
--- a/ipc/ipcns_notifier.c
+++ b/ipc/ipcns_notifier.c
@@ -55,25 +55,35 @@ static int ipcns_callback(struct notifier_block *self,
 
 int register_ipcns_notifier(struct ipc_namespace *ns)
 {
+	int rc;
+
 	memset(&ns->ipcns_nb, 0, sizeof(ns->ipcns_nb));
 	ns->ipcns_nb.notifier_call = ipcns_callback;
 	ns->ipcns_nb.priority = IPCNS_CALLBACK_PRI;
-	return blocking_notifier_chain_register(&ipcns_chain, &ns->ipcns_nb);
+	rc = blocking_notifier_chain_register(&ipcns_chain, &ns->ipcns_nb);
+	if (!rc)
+		ns->auto_msgmni = 1;
+	return rc;
 }
 
 int cond_register_ipcns_notifier(struct ipc_namespace *ns)
 {
+	int rc;
+
 	memset(&ns->ipcns_nb, 0, sizeof(ns->ipcns_nb));
 	ns->ipcns_nb.notifier_call = ipcns_callback;
 	ns->ipcns_nb.priority = IPCNS_CALLBACK_PRI;
-	return blocking_notifier_chain_cond_register(&ipcns_chain,
+	rc = blocking_notifier_chain_cond_register(&ipcns_chain,
 							&ns->ipcns_nb);
+	if (!rc)
+		ns->auto_msgmni = 1;
+	return rc;
 }
 
-int unregister_ipcns_notifier(struct ipc_namespace *ns)
+void unregister_ipcns_notifier(struct ipc_namespace *ns)
 {
-	return blocking_notifier_chain_unregister(&ipcns_chain,
-						&ns->ipcns_nb);
+	blocking_notifier_chain_unregister(&ipcns_chain, &ns->ipcns_nb);
+	ns->auto_msgmni = 0;
 }
 
 int ipcns_notify(unsigned long val)

From 7833351b5260b3a58b54a0c2e7065001d986d749 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 01:48:09 -0700
Subject: [PATCH 0864/1171] pty: remove unused UNIX98_PTY_COUNT options

The h8300 and sparc options somehow survived when the code stopped using
CONFIG_UNIX98_PTY_COUNT.

Reviewed-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/h8300/Kconfig | 14 --------------
 arch/sparc/Kconfig | 14 --------------
 2 files changed, 28 deletions(-)

diff --git a/arch/h8300/Kconfig b/arch/h8300/Kconfig
index 085dc6ec152b..396ab059efa3 100644
--- a/arch/h8300/Kconfig
+++ b/arch/h8300/Kconfig
@@ -203,20 +203,6 @@ config UNIX98_PTYS
 	  Read the instructions in <file:Documentation/Changes> pertaining to
 	  pseudo terminals. It's safe to say N.
 
-config UNIX98_PTY_COUNT
-	int "Maximum number of Unix98 PTYs in use (0-2048)"
-	depends on UNIX98_PTYS
-	default "256"
-	help
-	  The maximum number of Unix98 PTYs that can be used at any one time.
-	  The default is 256, and should be enough for desktop systems. Server
-	  machines which support incoming telnet/rlogin/ssh connections and/or
-	  serve several X terminals may want to increase this: every incoming
-	  connection and every xterm uses up one PTY.
-
-	  When not in use, each additional set of 256 PTYs occupy
-	  approximately 8 KB of kernel memory on 32-bit architectures.
-
 source "drivers/char/pcmcia/Kconfig"
 
 source "drivers/serial/Kconfig"
diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
index 789724e61e83..375de7c6d082 100644
--- a/arch/sparc/Kconfig
+++ b/arch/sparc/Kconfig
@@ -298,20 +298,6 @@ config UNIX98_PTYS
 	  Read the instructions in <file:Documentation/Changes> pertaining to
 	  pseudo terminals. It's safe to say N.
 
-config UNIX98_PTY_COUNT
-	int "Maximum number of Unix98 PTYs in use (0-2048)"
-	depends on UNIX98_PTYS
-	default "256"
-	help
-	  The maximum number of Unix98 PTYs that can be used at any one time.
-	  The default is 256, and should be enough for desktop systems. Server
-	  machines which support incoming telnet/rlogin/ssh connections and/or
-	  serve several X terminals may want to increase this: every incoming
-	  connection and every xterm uses up one PTY.
-
-	  When not in use, each additional set of 256 PTYs occupy
-	  approximately 8 KB of kernel memory on 32-bit architectures.
-
 endmenu
 
 source "fs/Kconfig"

From 79885b227740b9c7d3057f2de556f4098d37cc8f Mon Sep 17 00:00:00 2001
From: "Edgar E. Iglesias" <edgar.iglesias@axis.com>
Date: Fri, 25 Jul 2008 01:48:10 -0700
Subject: [PATCH 0865/1171] elf: use ELF_CORE_EFLAGS for kcore ELF header flags

ELF_CORE_EFLAGS is already used by the binfmt_elf coredumper to set correct
arch specific ELF header flags on coredumps.  Use it for kcore dumps as well.
At the moment, this affects the CRIS and the H8300 arch.

Signed-off-by: Edgar E. Iglesias <edgar@axis.com>
Cc: Mikael Starvik <starvik@axis.com>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Cc: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/proc/kcore.c         | 10 +++++-----
 include/asm-h8300/elf.h |  4 ++--
 2 files changed, 7 insertions(+), 7 deletions(-)

diff --git a/fs/proc/kcore.c b/fs/proc/kcore.c
index e78c81fcf547..c2370c76fb71 100644
--- a/fs/proc/kcore.c
+++ b/fs/proc/kcore.c
@@ -23,6 +23,10 @@
 
 #define CORE_STR "CORE"
 
+#ifndef ELF_CORE_EFLAGS
+#define ELF_CORE_EFLAGS	0
+#endif
+
 static int open_kcore(struct inode * inode, struct file * filp)
 {
 	return capable(CAP_SYS_RAWIO) ? 0 : -EPERM;
@@ -164,11 +168,7 @@ static void elf_kcore_store_hdr(char *bufp, int nphdr, int dataoff)
 	elf->e_entry	= 0;
 	elf->e_phoff	= sizeof(struct elfhdr);
 	elf->e_shoff	= 0;
-#if defined(CONFIG_H8300)
-	elf->e_flags	= ELF_FLAGS;
-#else
-	elf->e_flags	= 0;
-#endif
+	elf->e_flags	= ELF_CORE_EFLAGS;
 	elf->e_ehsize	= sizeof(struct elfhdr);
 	elf->e_phentsize= sizeof(struct elf_phdr);
 	elf->e_phnum	= nphdr;
diff --git a/include/asm-h8300/elf.h b/include/asm-h8300/elf.h
index 26bfc7e641da..a8b57d1f4128 100644
--- a/include/asm-h8300/elf.h
+++ b/include/asm-h8300/elf.h
@@ -26,10 +26,10 @@ typedef unsigned long elf_fpregset_t;
 #define ELF_DATA	ELFDATA2MSB
 #define ELF_ARCH	EM_H8_300
 #if defined(__H8300H__)
-#define ELF_FLAGS       0x810000
+#define ELF_CORE_EFLAGS 0x810000
 #endif
 #if defined(__H8300S__)
-#define ELF_FLAGS       0x820000
+#define ELF_CORE_EFLAGS 0x820000
 #endif
 
 #define ELF_PLAT_INIT(_r)	_r->er1 = 0

From 8d1e120f695e9bcf01585e052577dc1e099033f9 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 01:48:11 -0700
Subject: [PATCH 0866/1171] proper extern for mwave_s_mdd

This patch adds a proper extern for mwave_s_mdd in
drivers/char/mwave/mwavedd.h

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/mwave/mwavedd.h | 2 ++
 drivers/char/mwave/tp3780i.c | 2 --
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/char/mwave/mwavedd.h b/drivers/char/mwave/mwavedd.h
index 8eca61e0a19c..7e0d530e2e07 100644
--- a/drivers/char/mwave/mwavedd.h
+++ b/drivers/char/mwave/mwavedd.h
@@ -147,4 +147,6 @@ typedef struct _MWAVE_DEVICE_DATA {
 
 } MWAVE_DEVICE_DATA, *pMWAVE_DEVICE_DATA;
 
+extern MWAVE_DEVICE_DATA mwave_s_mdd;
+
 #endif
diff --git a/drivers/char/mwave/tp3780i.c b/drivers/char/mwave/tp3780i.c
index f282976daaac..c68969708068 100644
--- a/drivers/char/mwave/tp3780i.c
+++ b/drivers/char/mwave/tp3780i.c
@@ -57,8 +57,6 @@
 #include "3780i.h"
 #include "mwavepub.h"
 
-extern MWAVE_DEVICE_DATA mwave_s_mdd;
-
 static unsigned short s_ausThinkpadIrqToField[16] =
 	{ 0xFFFF, 0xFFFF, 0xFFFF, 0x0001, 0x0002, 0x0003, 0xFFFF, 0x0004,
 	0xFFFF, 0xFFFF, 0x0005, 0x0006, 0xFFFF, 0xFFFF, 0xFFFF, 0x0007 };

From 372572e9b1dcc5e36091199be63766d13e5a8ae0 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 01:48:11 -0700
Subject: [PATCH 0867/1171] #if 0 hpet_unregister()

This patch #if 0's the unused hpet_unregister().

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Acked-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/hpet.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/char/hpet.c b/drivers/char/hpet.c
index fb0a85a1eb36..b3f5dbc6d880 100644
--- a/drivers/char/hpet.c
+++ b/drivers/char/hpet.c
@@ -623,6 +623,7 @@ static inline int hpet_tpcheck(struct hpet_task *tp)
 	return -ENXIO;
 }
 
+#if 0
 int hpet_unregister(struct hpet_task *tp)
 {
 	struct hpet_dev *devp;
@@ -652,6 +653,7 @@ int hpet_unregister(struct hpet_task *tp)
 
 	return 0;
 }
+#endif  /*  0  */
 
 static ctl_table hpet_table[] = {
 	{

From 76528a42e2c5199a1208909318a9c9948d25d0b7 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Fri, 25 Jul 2008 01:48:12 -0700
Subject: [PATCH 0868/1171] efirtc: push down the BKL

Push it down as far as the EFI method calls.  Someone who knows EFI can do
the other bits.  Also fix another wrong unknown ioctl return.

Signed-off-by: Alan Cox <alan@redhat.com>
Cc: Joe Perches <joe@perches.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/efirtc.c | 35 +++++++++++++++++++----------------
 1 file changed, 19 insertions(+), 16 deletions(-)

diff --git a/drivers/char/efirtc.c b/drivers/char/efirtc.c
index d57ca3e4e534..67fbd7aab5db 100644
--- a/drivers/char/efirtc.c
+++ b/drivers/char/efirtc.c
@@ -37,8 +37,9 @@
 #include <linux/rtc.h>
 #include <linux/proc_fs.h>
 #include <linux/efi.h>
+#include <linux/smp_lock.h>
+#include <linux/uaccess.h>
 
-#include <asm/uaccess.h>
 #include <asm/system.h>
 
 #define EFI_RTC_VERSION		"0.4"
@@ -51,8 +52,8 @@
 
 static DEFINE_SPINLOCK(efi_rtc_lock);
 
-static int efi_rtc_ioctl(struct inode *inode, struct file *file,
-		     unsigned int cmd, unsigned long arg);
+static long efi_rtc_ioctl(struct file *file, unsigned int cmd,
+							unsigned long arg);
 
 #define is_leap(year) \
           ((year) % 4 == 0 && ((year) % 100 != 0 || (year) % 400 == 0))
@@ -146,9 +147,8 @@ convert_from_efi_time(efi_time_t *eft, struct rtc_time *wtime)
 	}
 }
 
-static int
-efi_rtc_ioctl(struct inode *inode, struct file *file, unsigned int cmd,
-		     unsigned long arg)
+static long efi_rtc_ioctl(struct file *file, unsigned int cmd,
+							unsigned long arg)
 {
 
 	efi_status_t	status;
@@ -175,13 +175,13 @@ efi_rtc_ioctl(struct inode *inode, struct file *file, unsigned int cmd,
 			return -EINVAL;
 
 		case RTC_RD_TIME:
-
+			lock_kernel();
 			spin_lock_irqsave(&efi_rtc_lock, flags);
 
 			status = efi.get_time(&eft, &cap);
 
 			spin_unlock_irqrestore(&efi_rtc_lock,flags);
-
+			unlock_kernel();
 			if (status != EFI_SUCCESS) {
 				/* should never happen */
 				printk(KERN_ERR "efitime: can't read time\n");
@@ -203,11 +203,13 @@ efi_rtc_ioctl(struct inode *inode, struct file *file, unsigned int cmd,
 
 			convert_to_efi_time(&wtime, &eft);
 
+			lock_kernel();
 			spin_lock_irqsave(&efi_rtc_lock, flags);
 
 			status = efi.set_time(&eft);
 
 			spin_unlock_irqrestore(&efi_rtc_lock,flags);
+			unlock_kernel();
 
 			return status == EFI_SUCCESS ? 0 : -EINVAL;
 
@@ -223,6 +225,7 @@ efi_rtc_ioctl(struct inode *inode, struct file *file, unsigned int cmd,
 
 			convert_to_efi_time(&wtime, &eft);
 
+			lock_kernel();
 			spin_lock_irqsave(&efi_rtc_lock, flags);
 			/*
 			 * XXX Fixme:
@@ -233,16 +236,19 @@ efi_rtc_ioctl(struct inode *inode, struct file *file, unsigned int cmd,
 			status = efi.set_wakeup_time((efi_bool_t)enabled, &eft);
 
 			spin_unlock_irqrestore(&efi_rtc_lock,flags);
+			unlock_kernel();
 
 			return status == EFI_SUCCESS ? 0 : -EINVAL;
 
 		case RTC_WKALM_RD:
 
+			lock_kernel();
 			spin_lock_irqsave(&efi_rtc_lock, flags);
 
 			status = efi.get_wakeup_time((efi_bool_t *)&enabled, (efi_bool_t *)&pending, &eft);
 
 			spin_unlock_irqrestore(&efi_rtc_lock,flags);
+			unlock_kernel();
 
 			if (status != EFI_SUCCESS) return -EINVAL;
 
@@ -256,7 +262,7 @@ efi_rtc_ioctl(struct inode *inode, struct file *file, unsigned int cmd,
 			return copy_to_user(&ewp->time, &wtime,
 					    sizeof(struct rtc_time)) ? -EFAULT : 0;
 	}
-	return -EINVAL;
+	return -ENOTTY;
 }
 
 /*
@@ -265,8 +271,7 @@ efi_rtc_ioctl(struct inode *inode, struct file *file, unsigned int cmd,
  *	up things on a close.
  */
 
-static int
-efi_rtc_open(struct inode *inode, struct file *file)
+static int efi_rtc_open(struct inode *inode, struct file *file)
 {
 	/*
 	 * nothing special to do here
@@ -277,8 +282,7 @@ efi_rtc_open(struct inode *inode, struct file *file)
 	return 0;
 }
 
-static int
-efi_rtc_close(struct inode *inode, struct file *file)
+static int efi_rtc_close(struct inode *inode, struct file *file)
 {
 	return 0;
 }
@@ -289,13 +293,12 @@ efi_rtc_close(struct inode *inode, struct file *file)
 
 static const struct file_operations efi_rtc_fops = {
 	.owner		= THIS_MODULE,
-	.ioctl		= efi_rtc_ioctl,
+	.unlocked_ioctl	= efi_rtc_ioctl,
 	.open		= efi_rtc_open,
 	.release	= efi_rtc_close,
 };
 
-static struct miscdevice efi_rtc_dev=
-{
+static struct miscdevice efi_rtc_dev= {
 	EFI_RTC_MINOR,
 	"efirtc",
 	&efi_rtc_fops

From 47be36a24defbd19aea1354c416ec99f291c7ab8 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Fri, 25 Jul 2008 01:48:13 -0700
Subject: [PATCH 0869/1171] ip2: push BKL down for the firmware interface

(The tty side is already done)

Signed-off-by: Alan Cox <alan@redhat.com>
Cc: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/ip2/ip2main.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/drivers/char/ip2/ip2main.c b/drivers/char/ip2/ip2main.c
index 9cb48fcd316c..689f9dcd3b86 100644
--- a/drivers/char/ip2/ip2main.c
+++ b/drivers/char/ip2/ip2main.c
@@ -203,7 +203,7 @@ static int set_serial_info(i2ChanStrPtr, struct serial_struct __user *);
 
 static ssize_t ip2_ipl_read(struct file *, char __user *, size_t, loff_t *);
 static ssize_t ip2_ipl_write(struct file *, const char __user *, size_t, loff_t *);
-static int ip2_ipl_ioctl(struct inode *, struct file *, UINT, ULONG);
+static long ip2_ipl_ioctl(struct file *, UINT, ULONG);
 static int ip2_ipl_open(struct inode *, struct file *);
 
 static int DumpTraceBuffer(char __user *, int);
@@ -236,7 +236,7 @@ static const struct file_operations ip2_ipl = {
 	.owner		= THIS_MODULE,
 	.read		= ip2_ipl_read,
 	.write		= ip2_ipl_write,
-	.ioctl		= ip2_ipl_ioctl,
+	.unlocked_ioctl	= ip2_ipl_ioctl,
 	.open		= ip2_ipl_open,
 }; 
 
@@ -2845,10 +2845,10 @@ ip2_ipl_write(struct file *pFile, const char __user *pData, size_t count, loff_t
 /*                                                                            */
 /*                                                                            */
 /******************************************************************************/
-static int
-ip2_ipl_ioctl ( struct inode *pInode, struct file *pFile, UINT cmd, ULONG arg )
+static long
+ip2_ipl_ioctl (struct file *pFile, UINT cmd, ULONG arg )
 {
-	unsigned int iplminor = iminor(pInode);
+	unsigned int iplminor = iminor(pFile->f_path.dentry->d_inode);
 	int rc = 0;
 	void __user *argp = (void __user *)arg;
 	ULONG __user *pIndex = argp;
@@ -2859,6 +2859,8 @@ ip2_ipl_ioctl ( struct inode *pInode, struct file *pFile, UINT cmd, ULONG arg )
 	printk (KERN_DEBUG "IP2IPL: ioctl cmd %d, arg %ld\n", cmd, arg );
 #endif
 
+	lock_kernel();
+
 	switch ( iplminor ) {
 	case 0:	    // IPL device
 		rc = -EINVAL;
@@ -2919,6 +2921,7 @@ ip2_ipl_ioctl ( struct inode *pInode, struct file *pFile, UINT cmd, ULONG arg )
 		rc = -ENODEV;
 		break;
 	}
+	unlock_kernel();
 	return rc;
 }
 

From 909d145f0decbc4f17955e1fc4122a669a51fbc0 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Fri, 25 Jul 2008 01:48:14 -0700
Subject: [PATCH 0870/1171] mwave: ioctl BKL pushdown

Push the BKL down to the point it wraps the actual mwave method handlers

Signed-off-by: Alan Cox <alan@redhat.com>
Cc: Eric Sesterhenn <snakebyte@gmx.de>
Cc: Yani Ioannou <yani.ioannou@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/mwave/mwavedd.c | 39 ++++++++++++++++++++++++++----------
 1 file changed, 28 insertions(+), 11 deletions(-)

diff --git a/drivers/char/mwave/mwavedd.c b/drivers/char/mwave/mwavedd.c
index 50243fcd87e8..4f8d67fed292 100644
--- a/drivers/char/mwave/mwavedd.c
+++ b/drivers/char/mwave/mwavedd.c
@@ -86,8 +86,8 @@ module_param(mwave_uart_io, int, 0);
 
 static int mwave_open(struct inode *inode, struct file *file);
 static int mwave_close(struct inode *inode, struct file *file);
-static int mwave_ioctl(struct inode *inode, struct file *filp,
-                       unsigned int iocmd, unsigned long ioarg);
+static long mwave_ioctl(struct file *filp, unsigned int iocmd,
+							unsigned long ioarg);
 
 MWAVE_DEVICE_DATA mwave_s_mdd;
 
@@ -119,16 +119,16 @@ static int mwave_close(struct inode *inode, struct file *file)
 	return retval;
 }
 
-static int mwave_ioctl(struct inode *inode, struct file *file,
-                       unsigned int iocmd, unsigned long ioarg)
+static long mwave_ioctl(struct file *file, unsigned int iocmd,
+							unsigned long ioarg)
 {
 	unsigned int retval = 0;
 	pMWAVE_DEVICE_DATA pDrvData = &mwave_s_mdd;
 	void __user *arg = (void __user *)ioarg;
 
-	PRINTK_5(TRACE_MWAVE,
-		"mwavedd::mwave_ioctl, entry inode %p file %p cmd %x arg %x\n",
-		 inode,  file, iocmd, (int) ioarg);
+	PRINTK_4(TRACE_MWAVE,
+		"mwavedd::mwave_ioctl, entry file %p cmd %x arg %x\n",
+		file, iocmd, (int) ioarg);
 
 	switch (iocmd) {
 
@@ -136,7 +136,9 @@ static int mwave_ioctl(struct inode *inode, struct file *file,
 			PRINTK_1(TRACE_MWAVE,
 				"mwavedd::mwave_ioctl, IOCTL_MW_RESET"
 				" calling tp3780I_ResetDSP\n");
+			lock_kernel();
 			retval = tp3780I_ResetDSP(&pDrvData->rBDData);
+			unlock_kernel();
 			PRINTK_2(TRACE_MWAVE,
 				"mwavedd::mwave_ioctl, IOCTL_MW_RESET"
 				" retval %x from tp3780I_ResetDSP\n",
@@ -147,7 +149,9 @@ static int mwave_ioctl(struct inode *inode, struct file *file,
 			PRINTK_1(TRACE_MWAVE,
 				"mwavedd::mwave_ioctl, IOCTL_MW_RUN"
 				" calling tp3780I_StartDSP\n");
+			lock_kernel();
 			retval = tp3780I_StartDSP(&pDrvData->rBDData);
+			unlock_kernel();
 			PRINTK_2(TRACE_MWAVE,
 				"mwavedd::mwave_ioctl, IOCTL_MW_RUN"
 				" retval %x from tp3780I_StartDSP\n",
@@ -161,8 +165,10 @@ static int mwave_ioctl(struct inode *inode, struct file *file,
 				"mwavedd::mwave_ioctl,"
 				" IOCTL_MW_DSP_ABILITIES calling"
 				" tp3780I_QueryAbilities\n");
+			lock_kernel();
 			retval = tp3780I_QueryAbilities(&pDrvData->rBDData,
 					&rAbilities);
+			unlock_kernel();
 			PRINTK_2(TRACE_MWAVE,
 				"mwavedd::mwave_ioctl, IOCTL_MW_DSP_ABILITIES"
 				" retval %x from tp3780I_QueryAbilities\n",
@@ -193,11 +199,13 @@ static int mwave_ioctl(struct inode *inode, struct file *file,
 				"mwavedd::mwave_ioctl IOCTL_MW_READ_DATA,"
 				" size %lx, ioarg %lx pusBuffer %p\n",
 				rReadData.ulDataLength, ioarg, pusBuffer);
+			lock_kernel();
 			retval = tp3780I_ReadWriteDspDStore(&pDrvData->rBDData,
 					iocmd,
 					pusBuffer,
 					rReadData.ulDataLength,
 					rReadData.usDspAddress);
+			unlock_kernel();
 		}
 			break;
 	
@@ -215,10 +223,12 @@ static int mwave_ioctl(struct inode *inode, struct file *file,
 				" size %lx, ioarg %lx pusBuffer %p\n",
 				rReadData.ulDataLength / 2, ioarg,
 				pusBuffer);
+			lock_kernel();
 			retval = tp3780I_ReadWriteDspDStore(&pDrvData->rBDData,
 				iocmd, pusBuffer,
 				rReadData.ulDataLength / 2,
 				rReadData.usDspAddress);
+			unlock_kernel();
 		}
 			break;
 	
@@ -236,10 +246,12 @@ static int mwave_ioctl(struct inode *inode, struct file *file,
 				" size %lx, ioarg %lx pusBuffer %p\n",
 				rWriteData.ulDataLength, ioarg,
 				pusBuffer);
+			lock_kernel();
 			retval = tp3780I_ReadWriteDspDStore(&pDrvData->rBDData,
 					iocmd, pusBuffer,
 					rWriteData.ulDataLength,
 					rWriteData.usDspAddress);
+			unlock_kernel();
 		}
 			break;
 	
@@ -257,10 +269,12 @@ static int mwave_ioctl(struct inode *inode, struct file *file,
 				" size %lx, ioarg %lx pusBuffer %p\n",
 				rWriteData.ulDataLength, ioarg,
 				pusBuffer);
+			lock_kernel();
 			retval = tp3780I_ReadWriteDspIStore(&pDrvData->rBDData,
 					iocmd, pusBuffer,
 					rWriteData.ulDataLength,
 					rWriteData.usDspAddress);
+			unlock_kernel();
 		}
 			break;
 	
@@ -281,8 +295,10 @@ static int mwave_ioctl(struct inode *inode, struct file *file,
 						ipcnum);
 				return -EINVAL;
 			}
+			lock_kernel();
 			pDrvData->IPCs[ipcnum].bIsHere = FALSE;
 			pDrvData->IPCs[ipcnum].bIsEnabled = TRUE;
+			unlock_kernel();
 	
 			PRINTK_2(TRACE_MWAVE,
 				"mwavedd::mwave_ioctl IOCTL_MW_REGISTER_IPC"
@@ -307,6 +323,7 @@ static int mwave_ioctl(struct inode *inode, struct file *file,
 				return -EINVAL;
 			}
 	
+			lock_kernel();
 			if (pDrvData->IPCs[ipcnum].bIsEnabled == TRUE) {
 				DECLARE_WAITQUEUE(wait, current);
 
@@ -347,6 +364,7 @@ static int mwave_ioctl(struct inode *inode, struct file *file,
 					" processing\n",
 					ipcnum);
 			}
+			unlock_kernel();
 		}
 			break;
 	
@@ -365,19 +383,18 @@ static int mwave_ioctl(struct inode *inode, struct file *file,
 						ipcnum);
 				return -EINVAL;
 			}
+			lock_kernel();
 			if (pDrvData->IPCs[ipcnum].bIsEnabled == TRUE) {
 				pDrvData->IPCs[ipcnum].bIsEnabled = FALSE;
 				if (pDrvData->IPCs[ipcnum].bIsHere == TRUE) {
 					wake_up_interruptible(&pDrvData->IPCs[ipcnum].ipc_wait_queue);
 				}
 			}
+			unlock_kernel();
 		}
 			break;
 	
 		default:
-			PRINTK_ERROR(KERN_ERR_MWAVE "mwavedd::mwave_ioctl:"
-					" Error: Unrecognized iocmd %x\n",
-					iocmd);
 			return -ENOTTY;
 			break;
 	} /* switch */
@@ -460,7 +477,7 @@ static const struct file_operations mwave_fops = {
 	.owner		= THIS_MODULE,
 	.read		= mwave_read,
 	.write		= mwave_write,
-	.ioctl		= mwave_ioctl,
+	.unlocked_ioctl	= mwave_ioctl,
 	.open		= mwave_open,
 	.release	= mwave_close
 };

From f6759fdcfd79ff1827fd5d4ddfe876164466d30d Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Fri, 25 Jul 2008 01:48:14 -0700
Subject: [PATCH 0871/1171] rio: push down the BKL into the firmware ioctl
 handler

TTY side is already done.

Signed-off-by: Alan Cox <alan@redhat.com>
Cc: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/rio/rio_linux.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/char/rio/rio_linux.c b/drivers/char/rio/rio_linux.c
index 0cdfee152916..a8f68a3f14dd 100644
--- a/drivers/char/rio/rio_linux.c
+++ b/drivers/char/rio/rio_linux.c
@@ -179,7 +179,7 @@ static int rio_set_real_termios(void *ptr);
 static void rio_hungup(void *ptr);
 static void rio_close(void *ptr);
 static int rio_chars_in_buffer(void *ptr);
-static int rio_fw_ioctl(struct inode *inode, struct file *filp, unsigned int cmd, unsigned long arg);
+static long rio_fw_ioctl(struct file *filp, unsigned int cmd, unsigned long arg);
 static int rio_init_drivers(void);
 
 static void my_hd(void *addr, int len);
@@ -240,7 +240,7 @@ static struct real_driver rio_real_driver = {
 
 static const struct file_operations rio_fw_fops = {
 	.owner = THIS_MODULE,
-	.ioctl = rio_fw_ioctl,
+	.unlocked_ioctl = rio_fw_ioctl,
 };
 
 static struct miscdevice rio_fw_device = {
@@ -560,13 +560,15 @@ static void rio_close(void *ptr)
 
 
 
-static int rio_fw_ioctl(struct inode *inode, struct file *filp, unsigned int cmd, unsigned long arg)
+static long rio_fw_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 {
 	int rc = 0;
 	func_enter();
 
 	/* The "dev" argument isn't used. */
+	lock_kernel();
 	rc = riocontrol(p, 0, cmd, arg, capable(CAP_SYS_ADMIN));
+	unlock_kernel();
 
 	func_exit();
 	return rc;

From 11af7478addd34c42999b3b84095903ed9e67038 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Fri, 25 Jul 2008 01:48:15 -0700
Subject: [PATCH 0872/1171] sx: push BKL down into the firmware ioctl handler

Also fix the capability checking for firmware load.

Signed-off-by: Alan Cox <alan@redhat.com>
Cc: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/sx.c | 73 ++++++++++++++++++++++++++++-------------------
 1 file changed, 43 insertions(+), 30 deletions(-)

diff --git a/drivers/char/sx.c b/drivers/char/sx.c
index 2162439bbe48..c385206f9db5 100644
--- a/drivers/char/sx.c
+++ b/drivers/char/sx.c
@@ -286,8 +286,8 @@ static void sx_close(void *ptr);
 static int sx_chars_in_buffer(void *ptr);
 static int sx_init_board(struct sx_board *board);
 static int sx_init_portstructs(int nboards, int nports);
-static int sx_fw_ioctl(struct inode *inode, struct file *filp,
-		unsigned int cmd, unsigned long arg);
+static long sx_fw_ioctl(struct file *filp, unsigned int cmd,
+						unsigned long arg);
 static int sx_init_drivers(void);
 
 static struct tty_driver *sx_driver;
@@ -396,7 +396,7 @@ static struct real_driver sx_real_driver = {
 
 static const struct file_operations sx_fw_fops = {
 	.owner = THIS_MODULE,
-	.ioctl = sx_fw_ioctl,
+	.unlocked_ioctl = sx_fw_ioctl,
 };
 
 static struct miscdevice sx_fw_device = {
@@ -1686,10 +1686,10 @@ static int do_memtest_w(struct sx_board *board, int min, int max)
 }
 #endif
 
-static int sx_fw_ioctl(struct inode *inode, struct file *filp,
-		unsigned int cmd, unsigned long arg)
+static long sx_fw_ioctl(struct file *filp, unsigned int cmd,
+							unsigned long arg)
 {
-	int rc = 0;
+	long rc = 0;
 	int __user *descr = (int __user *)arg;
 	int i;
 	static struct sx_board *board = NULL;
@@ -1699,13 +1699,10 @@ static int sx_fw_ioctl(struct inode *inode, struct file *filp,
 
 	func_enter();
 
-#if 0
-	/* Removed superuser check: Sysops can use the permissions on the device
-	   file to restrict access. Recommendation: Root only. (root.root 600) */
-	if (!capable(CAP_SYS_ADMIN)) {
+	if (!capable(CAP_SYS_RAWIO))
 		return -EPERM;
-	}
-#endif
+
+	lock_kernel();
 
 	sx_dprintk(SX_DEBUG_FIRMWARE, "IOCTL %x: %lx\n", cmd, arg);
 
@@ -1720,19 +1717,23 @@ static int sx_fw_ioctl(struct inode *inode, struct file *filp,
 		for (i = 0; i < SX_NBOARDS; i++)
 			sx_dprintk(SX_DEBUG_FIRMWARE, "<%x> ", boards[i].flags);
 		sx_dprintk(SX_DEBUG_FIRMWARE, "\n");
+		unlock_kernel();
 		return -EIO;
 	}
 
 	switch (cmd) {
 	case SXIO_SET_BOARD:
 		sx_dprintk(SX_DEBUG_FIRMWARE, "set board to %ld\n", arg);
+		rc = -EIO;
 		if (arg >= SX_NBOARDS)
-			return -EIO;
+			break;
 		sx_dprintk(SX_DEBUG_FIRMWARE, "not out of range\n");
 		if (!(boards[arg].flags & SX_BOARD_PRESENT))
-			return -EIO;
+			break;
 		sx_dprintk(SX_DEBUG_FIRMWARE, ".. and present!\n");
 		board = &boards[arg];
+		rc = 0;
+		/* FIXME: And this does ... nothing?? */
 		break;
 	case SXIO_GET_TYPE:
 		rc = -ENOENT;	/* If we manage to miss one, return error. */
@@ -1746,7 +1747,7 @@ static int sx_fw_ioctl(struct inode *inode, struct file *filp,
 			rc = SX_TYPE_SI;
 		if (IS_EISA_BOARD(board))
 			rc = SX_TYPE_SI;
-		sx_dprintk(SX_DEBUG_FIRMWARE, "returning type= %d\n", rc);
+		sx_dprintk(SX_DEBUG_FIRMWARE, "returning type= %ld\n", rc);
 		break;
 	case SXIO_DO_RAMTEST:
 		if (sx_initialized)	/* Already initialized: better not ramtest the board.  */
@@ -1760,19 +1761,26 @@ static int sx_fw_ioctl(struct inode *inode, struct file *filp,
 			rc = do_memtest(board, 0, 0x7ff8);
 			/* if (!rc) rc = do_memtest_w (board, 0, 0x7ff8); */
 		}
-		sx_dprintk(SX_DEBUG_FIRMWARE, "returning memtest result= %d\n",
-			   rc);
+		sx_dprintk(SX_DEBUG_FIRMWARE,
+				"returning memtest result= %ld\n", rc);
 		break;
 	case SXIO_DOWNLOAD:
-		if (sx_initialized)	/* Already initialized */
-			return -EEXIST;
-		if (!sx_reset(board))
-			return -EIO;
+		if (sx_initialized) {/* Already initialized */
+			rc = -EEXIST;
+			break;
+		}
+		if (!sx_reset(board)) {
+			rc = -EIO;
+			break;
+		}
 		sx_dprintk(SX_DEBUG_INIT, "reset the board...\n");
 
 		tmp = kmalloc(SX_CHUNK_SIZE, GFP_USER);
-		if (!tmp)
-			return -ENOMEM;
+		if (!tmp) {
+			rc = -ENOMEM;
+			break;
+		}
+		/* FIXME: check returns */
 		get_user(nbytes, descr++);
 		get_user(offset, descr++);
 		get_user(data, descr++);
@@ -1782,7 +1790,8 @@ static int sx_fw_ioctl(struct inode *inode, struct file *filp,
 						(i + SX_CHUNK_SIZE > nbytes) ?
 						nbytes - i : SX_CHUNK_SIZE)) {
 					kfree(tmp);
-					return -EFAULT;
+					rc = -EFAULT;
+					break;
 				}
 				memcpy_toio(board->base2 + offset + i, tmp,
 						(i + SX_CHUNK_SIZE > nbytes) ?
@@ -1798,13 +1807,17 @@ static int sx_fw_ioctl(struct inode *inode, struct file *filp,
 		rc = sx_nports;
 		break;
 	case SXIO_INIT:
-		if (sx_initialized)	/* Already initialized */
-			return -EEXIST;
+		if (sx_initialized) {	/* Already initialized */
+			rc = -EEXIST;
+			break;
+		}
 		/* This is not allowed until all boards are initialized... */
 		for (i = 0; i < SX_NBOARDS; i++) {
 			if ((boards[i].flags & SX_BOARD_PRESENT) &&
-				!(boards[i].flags & SX_BOARD_INITIALIZED))
-				return -EIO;
+				!(boards[i].flags & SX_BOARD_INITIALIZED)) {
+				rc = -EIO;
+				break;
+			}
 		}
 		for (i = 0; i < SX_NBOARDS; i++)
 			if (!(boards[i].flags & SX_BOARD_PRESENT))
@@ -1832,10 +1845,10 @@ static int sx_fw_ioctl(struct inode *inode, struct file *filp,
 		rc = sx_nports;
 		break;
 	default:
-		printk(KERN_WARNING "Unknown ioctl on firmware device (%x).\n",
-				cmd);
+		rc = -ENOTTY;
 		break;
 	}
+	unlock_kernel();
 	func_exit();
 	return rc;
 }

From e05e9f7c4aeb82eaa23e46b29580ff514590c641 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Fri, 25 Jul 2008 01:48:16 -0700
Subject: [PATCH 0873/1171] ixj: push BKL into driver and wrap ioctls

Signed-off-by: Alan Cox <alan@redhat.com>
Cc: Nishanth Aravamudan <nacc@us.ibm.com>
Cc: Domen Puncer <domen@coderock.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/telephony/ixj.c | 17 +++++++++++++----
 1 file changed, 13 insertions(+), 4 deletions(-)

diff --git a/drivers/telephony/ixj.c b/drivers/telephony/ixj.c
index 49cd9793404f..ec7aeb502d15 100644
--- a/drivers/telephony/ixj.c
+++ b/drivers/telephony/ixj.c
@@ -6095,15 +6095,15 @@ static int capabilities_check(IXJ *j, struct phone_capability *pcreq)
 	return retval;
 }
 
-static int ixj_ioctl(struct inode *inode, struct file *file_p, unsigned int cmd, unsigned long arg)
+static long do_ixj_ioctl(struct file *file_p, unsigned int cmd, unsigned long arg)
 {
 	IXJ_TONE ti;
 	IXJ_FILTER jf;
 	IXJ_FILTER_RAW jfr;
 	void __user *argp = (void __user *)arg;
-
-	unsigned int raise, mant;
+	struct inode *inode = file_p->f_path.dentry->d_inode;
 	unsigned int minor = iminor(inode);
+	unsigned int raise, mant;
 	int board = NUM(inode);
 
 	IXJ *j = get_ixj(NUM(inode));
@@ -6661,6 +6661,15 @@ static int ixj_ioctl(struct inode *inode, struct file *file_p, unsigned int cmd,
 	return retval;
 }
 
+static long ixj_ioctl(struct file *file_p, unsigned int cmd, unsigned long arg)
+{
+	long ret;
+	lock_kernel();
+	ret = do_ixj_ioctl(file_p, cmd, arg);
+	unlock_kernel();
+	return ret;
+}
+
 static int ixj_fasync(int fd, struct file *file_p, int mode)
 {
 	IXJ *j = get_ixj(NUM(file_p->f_path.dentry->d_inode));
@@ -6674,7 +6683,7 @@ static const struct file_operations ixj_fops =
         .read           = ixj_enhanced_read,
         .write          = ixj_enhanced_write,
         .poll           = ixj_poll,
-        .ioctl          = ixj_ioctl,
+        .unlocked_ioctl = ixj_ioctl,
         .release        = ixj_release,
         .fasync         = ixj_fasync
 };

From 6d535d3e6ad395345750c361bd2b7f1b9429455d Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Fri, 25 Jul 2008 01:48:16 -0700
Subject: [PATCH 0874/1171] ppdev: wrap ioctl handler in driver and push lock
 down

Signed-off-by: Alan Cox <alan@redhat.com>
Cc: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/ppdev.c | 18 +++++++++++++-----
 1 file changed, 13 insertions(+), 5 deletions(-)

diff --git a/drivers/char/ppdev.c b/drivers/char/ppdev.c
index 7af7a7e6b9c2..bee39fdfba73 100644
--- a/drivers/char/ppdev.c
+++ b/drivers/char/ppdev.c
@@ -67,7 +67,7 @@
 #include <linux/major.h>
 #include <linux/ppdev.h>
 #include <linux/smp_lock.h>
-#include <asm/uaccess.h>
+#include <linux/uaccess.h>
 
 #define PP_VERSION "ppdev: user-space parallel port driver"
 #define CHRDEV "ppdev"
@@ -328,10 +328,9 @@ static enum ieee1284_phase init_phase (int mode)
 	return IEEE1284_PH_FWD_IDLE;
 }
 
-static int pp_ioctl(struct inode *inode, struct file *file,
-		    unsigned int cmd, unsigned long arg)
+static int pp_do_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 {
-	unsigned int minor = iminor(inode);
+	unsigned int minor = iminor(file->f_path.dentry->d_inode);
 	struct pp_struct *pp = file->private_data;
 	struct parport * port;
 	void __user *argp = (void __user *)arg;
@@ -634,6 +633,15 @@ static int pp_ioctl(struct inode *inode, struct file *file,
 	return 0;
 }
 
+static long pp_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+	long ret;
+	lock_kernel();
+	ret = pp_do_ioctl(file, cmd, arg);
+	unlock_kernel();
+	return ret;
+}
+
 static int pp_open (struct inode * inode, struct file * file)
 {
 	unsigned int minor = iminor(inode);
@@ -745,7 +753,7 @@ static const struct file_operations pp_fops = {
 	.read		= pp_read,
 	.write		= pp_write,
 	.poll		= pp_poll,
-	.ioctl		= pp_ioctl,
+	.unlocked_ioctl	= pp_ioctl,
 	.open		= pp_open,
 	.release	= pp_release,
 };

From b8e35919653d76e7dceb8d3b8569c4ec1004d546 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Fri, 25 Jul 2008 01:48:17 -0700
Subject: [PATCH 0875/1171] ds1302: push down the BKL into the driver ioctl
 code

Signed-off-by: Alan Cox <alan@redhat.com>
Cc: Jiri Kosina <jkosina@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/ds1302.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

diff --git a/drivers/char/ds1302.c b/drivers/char/ds1302.c
index fada6ddefbae..c5e67a623951 100644
--- a/drivers/char/ds1302.c
+++ b/drivers/char/ds1302.c
@@ -20,10 +20,11 @@
 #include <linux/miscdevice.h>
 #include <linux/delay.h>
 #include <linux/bcd.h>
+#include <linux/smp_lock.h>
+#include <linux/uaccess.h>
+#include <linux/io.h>
 
-#include <asm/uaccess.h>
 #include <asm/system.h>
-#include <asm/io.h>
 #include <asm/rtc.h>
 #if defined(CONFIG_M32R)
 #include <asm/m32r.h>
@@ -153,9 +154,7 @@ static unsigned char days_in_mo[] =
 
 /* ioctl that supports RTC_RD_TIME and RTC_SET_TIME (read and set time/date). */
 
-static int
-rtc_ioctl(struct inode *inode, struct file *file, unsigned int cmd,
-	  unsigned long arg)
+static long rtc_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 {
 	unsigned long flags;
 
@@ -165,7 +164,9 @@ rtc_ioctl(struct inode *inode, struct file *file, unsigned int cmd,
 			struct rtc_time rtc_tm;
 
 			memset(&rtc_tm, 0, sizeof (struct rtc_time));
+			lock_kernel();
 			get_rtc_time(&rtc_tm);
+			unlock_kernel();
 			if (copy_to_user((struct rtc_time*)arg, &rtc_tm, sizeof(struct rtc_time)))
 				return -EFAULT;
 			return 0;
@@ -217,6 +218,7 @@ rtc_ioctl(struct inode *inode, struct file *file, unsigned int cmd,
 			BIN_TO_BCD(mon);
 			BIN_TO_BCD(yrs);
 
+			lock_kernel();
 			local_irq_save(flags);
 			CMOS_WRITE(yrs, RTC_YEAR);
 			CMOS_WRITE(mon, RTC_MONTH);
@@ -225,6 +227,7 @@ rtc_ioctl(struct inode *inode, struct file *file, unsigned int cmd,
 			CMOS_WRITE(min, RTC_MINUTES);
 			CMOS_WRITE(sec, RTC_SECONDS);
 			local_irq_restore(flags);
+			unlock_kernel();
 
 			/* Notice that at this point, the RTC is updated but
 			 * the kernel is still running with the old time.
@@ -244,8 +247,10 @@ rtc_ioctl(struct inode *inode, struct file *file, unsigned int cmd,
 			if(copy_from_user(&tcs_val, (int*)arg, sizeof(int)))
 				return -EFAULT;
 
+			lock_kernel();
 			tcs_val = RTC_TCR_PATTERN | (tcs_val & 0x0F);
 			ds1302_writereg(RTC_TRICKLECHARGER, tcs_val);
+			unlock_kernel();
 			return 0;
 		}
 		default:
@@ -282,7 +287,7 @@ get_rtc_status(char *buf)
 
 static const struct file_operations rtc_fops = {
 	.owner		= THIS_MODULE,
-	.ioctl		= rtc_ioctl,
+	.unlocked_ioctl	= rtc_ioctl,
 };
 
 /* Probe for the chip by writing something to its RAM and try reading it back. */

From 236b8756a2b6f90498d45b2c36d43e5372f2d4b8 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
Date: Fri, 25 Jul 2008 01:48:17 -0700
Subject: [PATCH 0876/1171] dsp56k: BKL pushdown

Push the BKL down into the driver ioctl methods

Signed-off-by: Alan Cox <alan@redhat.com>
Cc: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/dsp56k.c | 20 ++++++++++++++------
 1 file changed, 14 insertions(+), 6 deletions(-)

diff --git a/drivers/char/dsp56k.c b/drivers/char/dsp56k.c
index 33c466a4888f..19b88504e960 100644
--- a/drivers/char/dsp56k.c
+++ b/drivers/char/dsp56k.c
@@ -36,10 +36,10 @@
 #include <linux/smp_lock.h>
 #include <linux/firmware.h>
 #include <linux/platform_device.h>
+#include <linux/uaccess.h>	/* For put_user and get_user */
 
 #include <asm/atarihw.h>
 #include <asm/traps.h>
-#include <asm/uaccess.h>	/* For put_user and get_user */
 
 #include <asm/dsp56k.h>
 
@@ -303,8 +303,8 @@ static ssize_t dsp56k_write(struct file *file, const char __user *buf, size_t co
 	}
 }
 
-static int dsp56k_ioctl(struct inode *inode, struct file *file,
-			unsigned int cmd, unsigned long arg)
+static long dsp56k_ioctl(struct file *file, unsigned int cmd,
+							unsigned long arg)
 {
 	int dev = iminor(inode) & 0x0f;
 	void __user *argp = (void __user *)arg;
@@ -331,8 +331,9 @@ static int dsp56k_ioctl(struct inode *inode, struct file *file,
 			if (len > DSP56K_MAX_BINARY_LENGTH) {
 				return -EINVAL;
 			}
-    
+			lock_kernel();
 			r = dsp56k_upload(bin, len);
+			unlock_kernel();
 			if (r < 0) {
 				return r;
 			}
@@ -342,12 +343,16 @@ static int dsp56k_ioctl(struct inode *inode, struct file *file,
 		case DSP56K_SET_TX_WSIZE:
 			if (arg > 4 || arg < 1)
 				return -EINVAL;
+			lock_kernel();
 			dsp56k.tx_wsize = (int) arg;
+			unlock_kernel();
 			break;
 		case DSP56K_SET_RX_WSIZE:
 			if (arg > 4 || arg < 1)
 				return -EINVAL;
+			lock_kernel();
 			dsp56k.rx_wsize = (int) arg;
+			unlock_kernel();
 			break;
 		case DSP56K_HOST_FLAGS:
 		{
@@ -359,6 +364,7 @@ static int dsp56k_ioctl(struct inode *inode, struct file *file,
 			if(get_user(out, &hf->out) < 0)
 				return -EFAULT;
 
+			lock_kernel();
 			if ((dir & 0x1) && (out & 0x1))
 				dsp56k_host_interface.icr |= DSP56K_ICR_HF0;
 			else if (dir & 0x1)
@@ -373,14 +379,16 @@ static int dsp56k_ioctl(struct inode *inode, struct file *file,
 			if (dsp56k_host_interface.icr & DSP56K_ICR_HF1) status |= 0x2;
 			if (dsp56k_host_interface.isr & DSP56K_ISR_HF2) status |= 0x4;
 			if (dsp56k_host_interface.isr & DSP56K_ISR_HF3) status |= 0x8;
-
+			unlock_kernel();
 			return put_user(status, &hf->status);
 		}
 		case DSP56K_HOST_CMD:
 			if (arg > 31 || arg < 0)
 				return -EINVAL;
+			lock_kernel();
 			dsp56k_host_interface.cvr = (u_char)((arg & DSP56K_CVR_HV_MASK) |
 							     DSP56K_CVR_HC);
+			unlock_kernel();
 			break;
 		default:
 			return -EINVAL;
@@ -472,7 +480,7 @@ static const struct file_operations dsp56k_fops = {
 	.owner		= THIS_MODULE,
 	.read		= dsp56k_read,
 	.write		= dsp56k_write,
-	.ioctl		= dsp56k_ioctl,
+	.unlocked_ioctl	= dsp56k_ioctl,
 	.open		= dsp56k_open,
 	.release	= dsp56k_release,
 };

From 6ee8928d94841aa764aeaf645ad16daff811dc26 Mon Sep 17 00:00:00 2001
From: Akinobu Mita <akinobu.mita@gmail.com>
Date: Fri, 25 Jul 2008 01:48:18 -0700
Subject: [PATCH 0877/1171] nwflash: use simple_read_from_buffer()

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Tim Schmielau <tim@physik3.uni-rostock.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/nwflash.c | 31 ++++++++-----------------------
 1 file changed, 8 insertions(+), 23 deletions(-)

diff --git a/drivers/char/nwflash.c b/drivers/char/nwflash.c
index ba012c2bdf7a..f9f72a211292 100644
--- a/drivers/char/nwflash.c
+++ b/drivers/char/nwflash.c
@@ -122,35 +122,20 @@ static int flash_ioctl(struct inode *inodep, struct file *filep, unsigned int cm
 static ssize_t flash_read(struct file *file, char __user *buf, size_t size,
 			  loff_t *ppos)
 {
-	unsigned long p = *ppos;
-	unsigned int count = size;
-	int ret = 0;
+	ssize_t ret;
 
 	if (flashdebug)
 		printk(KERN_DEBUG "flash_read: flash_read: offset=0x%lX, "
 		       "buffer=%p, count=0x%X.\n", p, buf, count);
+	/*
+	 * We now lock against reads and writes. --rmk
+	 */
+	if (mutex_lock_interruptible(&nwflash_mutex))
+		return -ERESTARTSYS;
 
-	if (count)
-		ret = -ENXIO;
+	ret = simple_read_from_buffer(buf, size, ppos, FLASH_BASE, gbFlashSize);
+	mutex_unlock(&nwflash_mutex);
 
-	if (p < gbFlashSize) {
-		if (count > gbFlashSize - p)
-			count = gbFlashSize - p;
-
-		/*
-		 * We now lock against reads and writes. --rmk
-		 */
-		if (mutex_lock_interruptible(&nwflash_mutex))
-			return -ERESTARTSYS;
-
-		ret = copy_to_user(buf, (void *)(FLASH_BASE + p), count);
-		if (ret == 0) {
-			ret = count;
-			*ppos += count;
-		} else
-			ret = -EFAULT;
-		mutex_unlock(&nwflash_mutex);
-	}
 	return ret;
 }
 

From 41aee9a121fd0c31ae22dfe57e8f9ee9d6d85c25 Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Fri, 25 Jul 2008 01:48:19 -0700
Subject: [PATCH 0878/1171] Char: mxser, ioctl cleanup

- remove break ctl from ioctl handler, it's never reached, since
  tty_ops->break_ctl is defined (mxser break handling is done in software)
- mark MOXA_GET_MAJOR as deprecated
- fix TIOCGICOUNT (some retval non-checks of put_user). Use copy_to_user
  to whole structure instead.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Acked-by: Alan Cox <alan@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/mxser.c | 42 +++++++++++++++++++-----------------------
 1 file changed, 19 insertions(+), 23 deletions(-)

diff --git a/drivers/char/mxser.c b/drivers/char/mxser.c
index 4c756bbba948..e5029b149c5b 100644
--- a/drivers/char/mxser.c
+++ b/drivers/char/mxser.c
@@ -16,7 +16,6 @@
  *	Fed through a cleanup, indent and remove of non 2.6 code by Alan Cox
  *	<alan@redhat.com>. The original 1.8 code is available on www.moxa.com.
  *	- Fixed x86_64 cleanness
- *	- Fixed sleep with spinlock held in mxser_send_break
  */
 
 #include <linux/module.h>
@@ -1634,6 +1633,8 @@ static int mxser_ioctl_special(unsigned int cmd, void __user *argp)
 
 	switch (cmd) {
 	case MOXA_GET_MAJOR:
+		printk(KERN_WARNING "mxser: '%s' uses deprecated ioctl %x, fix "
+				"your userspace\n", current->comm, cmd);
 		return put_user(ttymajor, (int __user *)argp);
 
 	case MOXA_CHKPORTENABLE:
@@ -1804,7 +1805,6 @@ static int mxser_ioctl(struct tty_struct *tty, struct file *file,
 {
 	struct mxser_port *info = tty->driver_data;
 	struct async_icount cnow;
-	struct serial_icounter_struct __user *p_cuser;
 	unsigned long flags;
 	void __user *argp = (void __user *)arg;
 	int retval;
@@ -1884,30 +1884,26 @@ static int mxser_ioctl(struct tty_struct *tty, struct file *file,
 	 * NB: both 1->0 and 0->1 transitions are counted except for
 	 *     RI where only 0->1 is counted.
 	 */
-	case TIOCGICOUNT:
+	case TIOCGICOUNT: {
+		struct serial_icounter_struct icnt = { 0 };
 		spin_lock_irqsave(&info->slock, flags);
 		cnow = info->icount;
 		spin_unlock_irqrestore(&info->slock, flags);
-		p_cuser = argp;
-		if (put_user(cnow.frame, &p_cuser->frame))
-			return -EFAULT;
-		if (put_user(cnow.brk, &p_cuser->brk))
-			return -EFAULT;
-		if (put_user(cnow.overrun, &p_cuser->overrun))
-			return -EFAULT;
-		if (put_user(cnow.buf_overrun, &p_cuser->buf_overrun))
-			return -EFAULT;
-		if (put_user(cnow.parity, &p_cuser->parity))
-			return -EFAULT;
-		if (put_user(cnow.rx, &p_cuser->rx))
-			return -EFAULT;
-		if (put_user(cnow.tx, &p_cuser->tx))
-			return -EFAULT;
-		put_user(cnow.cts, &p_cuser->cts);
-		put_user(cnow.dsr, &p_cuser->dsr);
-		put_user(cnow.rng, &p_cuser->rng);
-		put_user(cnow.dcd, &p_cuser->dcd);
-		return 0;
+
+		icnt.frame = cnow.frame;
+		icnt.brk = cnow.brk;
+		icnt.overrun = cnow.overrun;
+		icnt.buf_overrun = cnow.buf_overrun;
+		icnt.parity = cnow.parity;
+		icnt.rx = cnow.rx;
+		icnt.tx = cnow.tx;
+		icnt.cts = cnow.cts;
+		icnt.dsr = cnow.dsr;
+		icnt.rng = cnow.rng;
+		icnt.dcd = cnow.dcd;
+
+		return copy_to_user(argp, &icnt, sizeof(icnt)) ? -EFAULT : 0;
+	}
 	case MOXA_HighSpeedOn:
 		return put_user(info->baud_base != 115200 ? 1 : 0, (int __user *)argp);
 	case MOXA_SDS_RSTICOUNTER:

From 72800df9ba3199df02a95b3830c49fbf16ec4a6d Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Fri, 25 Jul 2008 01:48:20 -0700
Subject: [PATCH 0879/1171] Char: mxser, globals cleanup

- remove unused mxvar_diagflag
- move mxser_msr into the only user/function
- GMStatus, hmm, fix race-prone access to it. We need only one instance for
  real, not MXSER_PORTS. Move it to MOXA_GETMSTATUS ioctl.
- mxser_mon_ext, almost the same, but alloc it on heap, since it has more than
  2 kilos.
- fix indexing, `i' is not the index value, `i * MXSER_PORTS_PER_BOARD + j' is

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Acked-by: Alan Cox <alan@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/mxser.c | 130 ++++++++++++++++++-------------------------
 1 file changed, 55 insertions(+), 75 deletions(-)

diff --git a/drivers/char/mxser.c b/drivers/char/mxser.c
index e5029b149c5b..3d7f2a970495 100644
--- a/drivers/char/mxser.c
+++ b/drivers/char/mxser.c
@@ -286,8 +286,6 @@ struct mxser_mstatus {
 	int dcd;
 };
 
-static struct mxser_mstatus GMStatus[MXSER_PORTS];
-
 static int mxserBoardCAP[MXSER_BOARDS] = {
 	0, 0, 0, 0
 	/*  0x180, 0x280, 0x200, 0x320 */
@@ -296,9 +294,6 @@ static int mxserBoardCAP[MXSER_BOARDS] = {
 static struct mxser_board mxser_boards[MXSER_BOARDS];
 static struct tty_driver *mxvar_sdriver;
 static struct mxser_log mxvar_log;
-static int mxvar_diagflag;
-static unsigned char mxser_msr[MXSER_PORTS + 1];
-static struct mxser_mon_ext mon_data_ext;
 static int mxser_set_baud_method[MXSER_PORTS + 1];
 
 static void mxser_enable_must_enchance_mode(unsigned long baseio)
@@ -542,6 +537,7 @@ static void process_txrx_fifo(struct mxser_port *info)
 
 static unsigned char mxser_get_msr(int baseaddr, int mode, int port)
 {
+	static unsigned char mxser_msr[MXSER_PORTS + 1];
 	unsigned char status = 0;
 
 	status = inb(baseaddr + UART_MSR);
@@ -1652,62 +1648,60 @@ static int mxser_ioctl_special(unsigned int cmd, void __user *argp)
 			ret = -EFAULT;
 		unlock_kernel();
 		return ret;
-	case MOXA_GETMSTATUS:
+	case MOXA_GETMSTATUS: {
+		struct mxser_mstatus ms, __user *msu = argp;
 		lock_kernel();
 		for (i = 0; i < MXSER_BOARDS; i++)
 			for (j = 0; j < MXSER_PORTS_PER_BOARD; j++) {
 				port = &mxser_boards[i].ports[j];
+				memset(&ms, 0, sizeof(ms));
 
-				GMStatus[i].ri = 0;
-				if (!port->ioaddr) {
-					GMStatus[i].dcd = 0;
-					GMStatus[i].dsr = 0;
-					GMStatus[i].cts = 0;
-					continue;
-				}
+				if (!port->ioaddr)
+					goto copy;
 
 				if (!port->port.tty || !port->port.tty->termios)
-					GMStatus[i].cflag =
-						port->normal_termios.c_cflag;
+					ms.cflag = port->normal_termios.c_cflag;
 				else
-					GMStatus[i].cflag =
-						port->port.tty->termios->c_cflag;
+					ms.cflag = port->port.tty->termios->c_cflag;
 
 				status = inb(port->ioaddr + UART_MSR);
-				if (status & 0x80 /*UART_MSR_DCD */ )
-					GMStatus[i].dcd = 1;
-				else
-					GMStatus[i].dcd = 0;
-
-				if (status & 0x20 /*UART_MSR_DSR */ )
-					GMStatus[i].dsr = 1;
-				else
-					GMStatus[i].dsr = 0;
-
-
-				if (status & 0x10 /*UART_MSR_CTS */ )
-					GMStatus[i].cts = 1;
-				else
-					GMStatus[i].cts = 0;
+				if (status & UART_MSR_DCD)
+					ms.dcd = 1;
+				if (status & UART_MSR_DSR)
+					ms.dsr = 1;
+				if (status & UART_MSR_CTS)
+					ms.cts = 1;
+			copy:
+				if (copy_to_user(msu, &ms, sizeof(ms))) {
+					unlock_kernel();
+					return -EFAULT;
+				}
+				msu++;
 			}
 		unlock_kernel();
-		if (copy_to_user(argp, GMStatus,
-				sizeof(struct mxser_mstatus) * MXSER_PORTS))
-			return -EFAULT;
 		return 0;
+	}
 	case MOXA_ASPP_MON_EXT: {
-		int p, shiftbit;
-		unsigned long opmode;
-		unsigned cflag, iflag;
+		struct mxser_mon_ext *me; /* it's 2k, stack unfriendly */
+		unsigned int cflag, iflag, p;
+		u8 opmode;
+
+		me = kzalloc(sizeof(*me), GFP_KERNEL);
+		if (!me)
+			return -ENOMEM;
 
 		lock_kernel();
-		for (i = 0; i < MXSER_BOARDS; i++) {
-			for (j = 0; j < MXSER_PORTS_PER_BOARD; j++) {
+		for (i = 0, p = 0; i < MXSER_BOARDS; i++) {
+			for (j = 0; j < MXSER_PORTS_PER_BOARD; j++, p++) {
+				if (p >= ARRAY_SIZE(me->rx_cnt)) {
+					i = MXSER_BOARDS;
+					break;
+				}
 				port = &mxser_boards[i].ports[j];
 				if (!port->ioaddr)
 					continue;
 
-				status = mxser_get_msr(port->ioaddr, 0, i);
+				status = mxser_get_msr(port->ioaddr, 0, p);
 
 				if (status & UART_MSR_TERI)
 					port->icount.rng++;
@@ -1719,16 +1713,13 @@ static int mxser_ioctl_special(unsigned int cmd, void __user *argp)
 					port->icount.cts++;
 
 				port->mon_data.modem_status = status;
-				mon_data_ext.rx_cnt[i] = port->mon_data.rxcnt;
-				mon_data_ext.tx_cnt[i] = port->mon_data.txcnt;
-				mon_data_ext.up_rxcnt[i] =
-					port->mon_data.up_rxcnt;
-				mon_data_ext.up_txcnt[i] =
-					port->mon_data.up_txcnt;
-				mon_data_ext.modem_status[i] =
+				me->rx_cnt[p] = port->mon_data.rxcnt;
+				me->tx_cnt[p] = port->mon_data.txcnt;
+				me->up_rxcnt[p] = port->mon_data.up_rxcnt;
+				me->up_txcnt[p] = port->mon_data.up_txcnt;
+				me->modem_status[p] =
 					port->mon_data.modem_status;
-				mon_data_ext.baudrate[i] =
-					tty_get_baud_rate(port->port.tty);
+				me->baudrate[p] = tty_get_baud_rate(port->port.tty);
 
 				if (!port->port.tty || !port->port.tty->termios) {
 					cflag = port->normal_termios.c_cflag;
@@ -1738,40 +1729,31 @@ static int mxser_ioctl_special(unsigned int cmd, void __user *argp)
 					iflag = port->port.tty->termios->c_iflag;
 				}
 
-				mon_data_ext.databits[i] = cflag & CSIZE;
-
-				mon_data_ext.stopbits[i] = cflag & CSTOPB;
-
-				mon_data_ext.parity[i] =
-					cflag & (PARENB | PARODD | CMSPAR);
-
-				mon_data_ext.flowctrl[i] = 0x00;
+				me->databits[p] = cflag & CSIZE;
+				me->stopbits[p] = cflag & CSTOPB;
+				me->parity[p] = cflag & (PARENB | PARODD |
+						CMSPAR);
 
 				if (cflag & CRTSCTS)
-					mon_data_ext.flowctrl[i] |= 0x03;
+					me->flowctrl[p] |= 0x03;
 
 				if (iflag & (IXON | IXOFF))
-					mon_data_ext.flowctrl[i] |= 0x0C;
+					me->flowctrl[p] |= 0x0C;
 
 				if (port->type == PORT_16550A)
-					mon_data_ext.fifo[i] = 1;
-				else
-					mon_data_ext.fifo[i] = 0;
+					me->fifo[p] = 1;
 
-				p = i % 4;
-				shiftbit = p * 2;
-				opmode = inb(port->opmode_ioaddr) >> shiftbit;
+				opmode = inb(port->opmode_ioaddr) >>
+						((p % 4) * 2);
 				opmode &= OP_MODE_MASK;
-
-				mon_data_ext.iftype[i] = opmode;
-
+				me->iftype[p] = opmode;
 			}
 		}
 		unlock_kernel();
-		if (copy_to_user(argp, &mon_data_ext,
-					sizeof(mon_data_ext)))
-			return -EFAULT;
-		return 0;
+		if (copy_to_user(argp, me, sizeof(*me)))
+			ret = -EFAULT;
+		kfree(me);
+		return ret;
 	}
 	default:
 		return -ENOIOCTLCMD;
@@ -2802,8 +2784,6 @@ static int __init mxser_module_init(void)
 		goto err_put;
 	}
 
-	mxvar_diagflag = 0;
-
 	m = 0;
 	/* Start finding ISA boards here */
 	for (isaloop = 0; isaloop < 2; isaloop++)

From 729f0edbecd0c59c82ee9bf92009acc7e984c425 Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Fri, 25 Jul 2008 01:48:20 -0700
Subject: [PATCH 0880/1171] Char: mxser, update documentation

Update Documentation/moxa-smartio to the later document from the mxser
package.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Acked-by: Alan Cox <alan@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/moxa-smartio | 386 ++++++++++++++++++++++++-------------
 1 file changed, 249 insertions(+), 137 deletions(-)

diff --git a/Documentation/moxa-smartio b/Documentation/moxa-smartio
index fe24ecc6372e..5337e80a5b96 100644
--- a/Documentation/moxa-smartio
+++ b/Documentation/moxa-smartio
@@ -1,14 +1,22 @@
 =============================================================================
+          MOXA Smartio/Industio Family Device Driver Installation Guide
+		    for Linux Kernel 2.4.x, 2.6.x
+	       Copyright (C) 2008, Moxa Inc.
+=============================================================================
+Date: 01/21/2008
 
-	MOXA Smartio Family Device Driver Ver 1.1 Installation Guide
-		    for Linux Kernel 2.2.x and 2.0.3x
-	       Copyright (C) 1999, Moxa Technologies Co, Ltd.
-=============================================================================
 Content
 
 1. Introduction
 2. System Requirement
 3. Installation
+   3.1 Hardware installation
+   3.2 Driver files
+   3.3 Device naming convention
+   3.4 Module driver configuration
+   3.5 Static driver configuration for Linux kernel 2.4.x and 2.6.x.
+   3.6 Custom configuration
+   3.7 Verify driver installation
 4. Utilities
 5. Setserial
 6. Troubleshooting
@@ -16,27 +24,48 @@ Content
 -----------------------------------------------------------------------------
 1. Introduction
 
-   The Smartio family Linux driver, Ver. 1.1, supports following multiport
+   The Smartio/Industio/UPCI family Linux driver supports following multiport
    boards.
 
-    -C104P/H/HS, C104H/PCI, C104HS/PCI, CI-104J 4 port multiport board.
-    -C168P/H/HS, C168H/PCI 8 port multiport board.
+    - 2 ports multiport board
+	CP-102U, CP-102UL, CP-102UF
+	CP-132U-I, CP-132UL,
+	CP-132, CP-132I, CP132S, CP-132IS,
+	CI-132, CI-132I, CI-132IS,
+	(C102H, C102HI, C102HIS, C102P, CP-102, CP-102S)
 
-   This driver has been modified a little and cleaned up from the Moxa
-   contributed driver code and merged into Linux 2.2.14pre. In particular
-   official major/minor numbers have been assigned which are different to
-   those the original Moxa supplied driver used.
+    - 4 ports multiport board
+	CP-104EL,
+	CP-104UL, CP-104JU,
+	CP-134U, CP-134U-I,
+	C104H/PCI, C104HS/PCI,
+	CP-114, CP-114I, CP-114S, CP-114IS, CP-114UL,
+	C104H, C104HS,
+	CI-104J, CI-104JS,
+	CI-134, CI-134I, CI-134IS,
+	(C114HI, CT-114I, C104P)
+	POS-104UL,
+	CB-114,
+	CB-134I
+
+    - 8 ports multiport board
+	CP-118EL, CP-168EL,
+	CP-118U, CP-168U,
+	C168H/PCI,
+	C168H, C168HS,
+	(C168P),
+	CB-108
 
    This driver and installation procedure have been developed upon Linux Kernel
-   2.2.5 and backward compatible to 2.0.3x. This driver supports Intel x86 and
-   Alpha hardware platform. In order to maintain compatibility, this version
-   has also been properly tested with RedHat, OpenLinux, TurboLinux and
-   S.u.S.E Linux. However, if compatibility problem occurs, please contact
-   Moxa at support@moxa.com.tw.
+   2.4.x and 2.6.x. This driver supports Intel x86 hardware platform. In order
+   to maintain compatibility, this version has also been properly tested with
+   RedHat, Mandrake, Fedora and S.u.S.E Linux. However, if compatibility problem
+   occurs, please contact Moxa at support@moxa.com.tw.
 
    In addition to device driver, useful utilities are also provided in this
    version. They are
-    - msdiag     Diagnostic program for detecting installed Moxa Smartio boards.
+    - msdiag     Diagnostic program for displaying installed Moxa
+                 Smartio/Industio boards.
     - msmon      Monitor program to observe data count and line status signals.
     - msterm     A simple terminal program which is useful in testing serial
 	         ports.
@@ -47,8 +76,7 @@ Content
    GNU General Public License in this version. Please refer to GNU General
    Public License announcement in each source code file for more detail.
 
-   In Moxa's ftp sites, you may always find latest driver at
-   ftp://ftp.moxa.com  or ftp://ftp.moxa.com.tw.
+   In Moxa's Web sites, you may always find latest driver at http://web.moxa.com.
 
    This version of driver can be installed as Loadable Module (Module driver)
    or built-in into kernel (Static driver). You may refer to following
@@ -61,8 +89,8 @@ Content
 
 -----------------------------------------------------------------------------
 2. System Requirement
-   - Hardware platform: Intel x86 or Alpha machine
-   - Kernel version: 2.0.3x or 2.2.x
+   - Hardware platform: Intel x86 machine
+   - Kernel version: 2.4.x or 2.6.x
    - gcc version 2.72 or later
    - Maximum 4 boards can be installed in combination
 
@@ -70,9 +98,18 @@ Content
 3. Installation
 
    3.1 Hardware installation
+   3.2 Driver files
+   3.3 Device naming convention
+   3.4 Module driver configuration
+   3.5 Static driver configuration for Linux kernel 2.4.x, 2.6.x.
+   3.6 Custom configuration
+   3.7 Verify driver installation
 
-       There are two types of buses, ISA and PCI, for Smartio family multiport
-       board.
+
+   3.1 Hardware installation
+
+       There are two types of buses, ISA and PCI, for Smartio/Industio
+       family multiport board.
 
        ISA board
        ---------
@@ -81,47 +118,57 @@ Content
        installation procedure in User's Manual before proceed any further.
        Please make sure the JP1 is open after the ISA board is set properly.
 
-       PCI board
-       ---------
+       PCI/UPCI board
+       --------------
        You may need to adjust IRQ usage in BIOS to avoid from IRQ conflict
        with other ISA devices. Please refer to hardware installation
        procedure in User's Manual in advance.
 
-       IRQ Sharing
+       PCI IRQ Sharing
        -----------
        Each port within the same multiport board shares the same IRQ. Up to
-       4 Moxa Smartio Family multiport boards can be installed together on
-       one system and they can share the same IRQ.
+       4 Moxa Smartio/Industio PCI Family multiport boards can be installed
+       together on one system and they can share the same IRQ.
 
-   3.2 Driver files and device naming convention
+
+   3.2 Driver files
 
        The driver file may be obtained from ftp, CD-ROM or floppy disk. The
        first step, anyway, is to copy driver file "mxser.tgz" into specified
        directory. e.g. /moxa. The execute commands as below.
 
+       # cd /
+       # mkdir moxa
        # cd /moxa
-       # tar xvf /dev/fd0 
+       # tar xvf /dev/fd0
+
        or
+
+       # cd /
+       # mkdir moxa
        # cd /moxa
        # cp /mnt/cdrom/<driver directory>/mxser.tgz .
        # tar xvfz mxser.tgz
 
+
+   3.3 Device naming convention
+
        You may find all the driver and utilities files in /moxa/mxser.
        Following installation procedure depends on the model you'd like to
-       run the driver. If you prefer module driver, please refer to 3.3.
-       If static driver is required, please refer to 3.4.
+       run the driver. If you prefer module driver, please refer to 3.4.
+       If static driver is required, please refer to 3.5.
 
        Dialin and callout port
        -----------------------
-       This driver remains traditional serial device properties. There're
+       This driver remains traditional serial device properties. There are
        two special file name for each serial port. One is dial-in port
        which is named "ttyMxx". For callout port, the naming convention
        is "cumxx".
 
        Device naming when more than 2 boards installed
        -----------------------------------------------
-       Naming convention for each Smartio multiport board is pre-defined
-       as below.
+       Naming convention for each Smartio/Industio multiport board is
+       pre-defined as below.
 
        Board Num.	 Dial-in Port	      Callout port
        1st board	ttyM0  - ttyM7	      cum0  - cum7
@@ -129,6 +176,12 @@ Content
        3rd board	ttyM16 - ttyM23       cum16 - cum23
        4th board	ttyM24 - ttym31       cum24 - cum31
 
+
+       !!!!!!!!!!!!!!!!!!!! NOTE !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
+       Under Kernel 2.6 the cum Device is Obsolete. So use ttyM*
+       device instead.
+       !!!!!!!!!!!!!!!!!!!! NOTE !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
+
        Board sequence
        --------------
        This driver will activate ISA boards according to the parameter set
@@ -138,69 +191,131 @@ Content
        For PCI boards, their sequence will be after ISA boards and C168H/PCI
        has higher priority than C104H/PCI boards.
 
-   3.3 Module driver configuration
+   3.4 Module driver configuration
        Module driver is easiest way to install. If you prefer static driver
        installation, please skip this paragraph.
-       1. Find "Makefile" in /moxa/mxser, then run
 
-	  # make install
 
-	  The driver files "mxser.o" and utilities will be properly compiled
-	  and copied to system directories respectively.Then run
+       ------------- Prepare to use the MOXA driver--------------------
+       3.4.1 Create tty device with correct major number
+          Before using MOXA driver, your system must have the tty devices
+          which are created with driver's major number. We offer one shell
+          script "msmknod" to simplify the procedure.
+          This step is only needed to be executed once. But you still
+          need to do this procedure when:
+          a. You change the driver's major number. Please refer the "3.7"
+             section.
+          b. Your total installed MOXA boards number is changed. Maybe you
+             add/delete one MOXA board.
+          c. You want to change the tty name. This needs to modify the
+             shell script "msmknod"
 
-	  # insmod mxser
-
-	  to activate the modular driver. You may run "lsmod" to check
-	  if "mxser.o" is activated.
-
-       2. Create special files by executing "msmknod".
+          The procedure is:
 	  # cd /moxa/mxser/driver
 	  # ./msmknod
 
-	  Default major numbers for dial-in device and callout device are
-	  174, 175. Msmknod will delete any special files occupying the same
-	  device naming.
+          This shell script will require the major number for dial-in
+          device and callout device to create tty device. You also need
+          to specify the total installed MOXA board number. Default major
+          numbers for dial-in device and callout device are 30, 35. If
+          you need to change to other number, please refer section "3.7"
+          for more detailed procedure.
+          Msmknod will delete any special files occupying the same device
+          naming.
 
-       3. Up to now, you may manually execute "insmod mxser" to activate
-	  this driver and run "rmmod mxser" to remove it. However, it's
-	  better to have a boot time configuration to eliminate manual
-	  operation.
-	  Boot time configuration can be achieved by rc file. Run following
-	  command for setting rc files.
+       3.4.2 Build the MOXA driver and utilities
+          Before using the MOXA driver and utilities, you need compile the
+          all the source code. This step is only need to be executed once.
+          But you still re-compile the source code if you modify the source
+          code. For example, if you change the driver's major number (see
+          "3.7" section), then you need to do this step again.
+
+          Find "Makefile" in /moxa/mxser, then run
+
+	  # make clean; make install
+
+          !!!!!!!!!! NOTE !!!!!!!!!!!!!!!!!
+	  For Red Hat 9, Red Hat Enterprise Linux AS3/ES3/WS3 & Fedora Core1:
+	  # make clean; make installsp1
+
+	  For Red Hat Enterprise Linux AS4/ES4/WS4:
+	  # make clean; make installsp2
+          !!!!!!!!!! NOTE !!!!!!!!!!!!!!!!!
+
+	  The driver files "mxser.o" and utilities will be properly compiled
+	  and copied to system directories respectively.
+
+       ------------- Load MOXA driver--------------------
+       3.4.3 Load the MOXA driver
+
+	  # modprobe mxser <argument>
+
+	  will activate the module driver. You may run "lsmod" to check
+	  if "mxser" is activated. If the MOXA board is ISA board, the
+          <argument> is needed. Please refer to section "3.4.5" for more
+          information.
+
+
+       ------------- Load MOXA driver on boot --------------------
+       3.4.4 For the above description, you may manually execute
+          "modprobe mxser" to activate this driver and run
+	  "rmmod mxser" to remove it.
+          However, it's better to have a boot time configuration to
+          eliminate manual operation. Boot time configuration can be
+          achieved by rc file. We offer one "rc.mxser" file to simplify
+          the procedure under "moxa/mxser/driver".
+
+          But if you use ISA board, please modify the "modprobe ..." command
+          to add the argument (see "3.4.5" section). After modifying the
+          rc.mxser, please try to execute "/moxa/mxser/driver/rc.mxser"
+          manually to make sure the modification is ok. If any error
+          encountered, please try to modify again. If the modification is
+          completed, follow the below step.
+
+	  Run following command for setting rc files.
 
 	  # cd /moxa/mxser/driver
 	  # cp ./rc.mxser /etc/rc.d
 	  # cd /etc/rc.d
 
-	  You may have to modify part of the content in rc.mxser to specify
-          parameters for ISA board. Please refer to rc.mxser for more detail.
-          Find "rc.serial". If "rc.serial" doesn't exist, create it by vi.
-	  Add "rc.mxser" in last line. Next, open rc.local by vi
-	  and append following content.
+	  Check "rc.serial" is existed or not. If "rc.serial" doesn't exist,
+	  create it by vi, run "chmod 755 rc.serial" to change the permission.
+	  Add "/etc/rc.d/rc.mxser" in last line,
 
-	  if [ -f /etc/rc.d/rc.serial ]; then
-	     sh /etc/rc.d/rc.serial
-	  fi
+          Reboot and check if moxa.o activated by "lsmod" command.
 
-       4. Reboot and check if mxser.o activated by "lsmod" command.
-       5. If you'd like to drive Smartio ISA boards in the system, you'll
-	  have to add parameter to specify CAP address of given board while
-          activating "mxser.o". The format for parameters are as follows.
+       3.4.5. If you'd like to drive Smartio/Industio ISA boards in the system,
+          you'll have to add parameter to specify CAP address of given
+	  board while activating "mxser.o". The format for parameters are
+	  as follows.
 
-	  insmod mxser ioaddr=0x???,0x???,0x???,0x???
+	  modprobe mxser ioaddr=0x???,0x???,0x???,0x???
 				|      |     |	  |
 				|      |     |	  +- 4th ISA board
 				|      |     +------ 3rd ISA board
 				|      +------------ 2nd ISA board
 				+------------------- 1st ISA board
 
-   3.4 Static driver configuration
+   3.5 Static driver configuration for Linux kernel 2.4.x and 2.6.x
 
-       1. Create link
+       Note: To use static driver, you must install the linux kernel
+             source package.
+
+       3.5.1 Backup the built-in driver in the kernel.
+          # cd /usr/src/linux/drivers/char
+          # mv mxser.c mxser.c.old
+
+          For Red Hat 7.x user, you need to create link:
+          # cd /usr/src
+          # ln -s linux-2.4 linux
+
+       3.5.2 Create link
 	  # cd /usr/src/linux/drivers/char
 	  # ln -s /moxa/mxser/driver/mxser.c mxser.c
 
-       2. Add CAP address list for ISA boards
+       3.5.3 Add CAP address list for ISA boards. For PCI boards user,
+          please skip this step.
+
 	  In module mode, the CAP address for ISA board is given by
 	  parameter. In static driver configuration, you'll have to
 	  assign it within driver's source code. If you will not
@@ -222,73 +337,55 @@ Content
 	     static int mxserBoardCAP[]
 	     = {0x280, 0x180, 0x00, 0x00};
 
-       3. Modify tty_io.c
-	  # cd /usr/src/linux/drivers/char/
-	  # vi tty_io.c
-	    Find pty_init(), insert "mxser_init()" as
+       3.5.4 Setup kernel configuration
 
-	    pty_init();
-	    mxser_init();
+          Configure the kernel:
 
-       4. Modify tty.h
-	  # cd /usr/src/linux/include/linux
-	  # vi tty.h
-	    Find extern int tty_init(void), insert "mxser_init()" as
+            # cd /usr/src/linux
+            # make menuconfig
 
-	    extern int tty_init(void);
-	    extern int mxser_init(void);
-     
-       5. Modify Makefile
-	  # cd /usr/src/linux/drivers/char
-	  # vi Makefile
-	    Find L_OBJS := tty_io.o ...... random.o, add
-	    "mxser.o" at last of this line as
-	    L_OBJS := tty_io.o ....... mxser.o
+          You will go into a menu-driven system. Please select [Character
+          devices][Non-standard serial port support], enable the [Moxa
+          SmartIO support] driver with "[*]" for built-in (not "[M]"), then
+          select [Exit] to exit this program.
 
-       6. Rebuild kernel
-	  The following are for Linux kernel rebuilding,for your reference only.
+       3.5.5 Rebuild kernel
+	  The following are for Linux kernel rebuilding, for your
+          reference only.
 	  For appropriate details, please refer to the Linux document.
 
-	  If 'lilo' utility is installed, please use 'make zlilo' to rebuild
-	  kernel. If 'lilo' is not installed, please follow the following steps.
-
 	   a. cd /usr/src/linux
-	   b. make clean			     /* take a few minutes */
-	   c. make bzImage		   /* take probably 10-20 minutes */
-	   d. Backup original boot kernel.		  /* optional step */
-	   e. cp /usr/src/linux/arch/i386/boot/bzImage /boot/vmlinuz
+	   b. make clean	     /* take a few minutes */
+	   c. make dep		     /* take a few minutes */
+	   d. make bzImage	     /* take probably 10-20 minutes */
+	   e. make install	     /* copy boot image to correct position */
 	   f. Please make sure the boot kernel (vmlinuz) is in the
-	      correct position. If you use 'lilo' utility, you should
-	      check /etc/lilo.conf 'image' item specified the path
-	      which is the 'vmlinuz' path, or you will load wrong
-	      (or old) boot kernel image (vmlinuz).
-	   g. chmod 400 /vmlinuz
-	   h. lilo
-	   i. rdev -R /vmlinuz 1
-	   j. sync
+	      correct position.
+	   g. If you use 'lilo' utility, you should check /etc/lilo.conf
+	      'image' item specified the path which is the 'vmlinuz' path,
+	      or you will load wrong (or old) boot kernel image (vmlinuz).
+	      After checking /etc/lilo.conf, please run "lilo".
 
-	  Note that if the result of "make zImage" is ERROR, then you have to
-	  go back to Linux configuration Setup. Type "make config" in directory
-	  /usr/src/linux or "setup".
+	  Note that if the result of "make bzImage" is ERROR, then you have to
+	  go back to Linux configuration Setup. Type "make menuconfig" in
+          directory /usr/src/linux.
 
-	  Since system include file, /usr/src/linux/include/linux/interrupt.h,
-	  is modified each time the MOXA driver is installed, kernel rebuilding
-	  is inevitable. And it takes about 10 to 20 minutes depends on the
-	  machine.
 
-       7. Make utility
-	  # cd /moxa/mxser/utility
-	  # make install
-       
-       8. Make special file
+       3.5.6 Make tty device and special file
           # cd /moxa/mxser/driver
           # ./msmknod
 
-       9. Reboot
+       3.5.7 Make utility
+	  # cd /moxa/mxser/utility
+	  # make clean; make install
 
-   3.5 Custom configuration
+       3.5.8 Reboot
+
+
+
+   3.6 Custom configuration
        Although this driver already provides you default configuration, you
-       still can change the device name and major number.The instruction to
+       still can change the device name and major number. The instruction to
        change these parameters are shown as below.
 
        Change Device name
@@ -306,33 +403,37 @@ Content
        2 free major numbers for this driver. There are 3 steps to change
        major numbers.
 
-       1. Find free major numbers
+       3.6.1 Find free major numbers
 	  In /proc/devices, you may find all the major numbers occupied
 	  in the system. Please select 2 major numbers that are available.
 	  e.g. 40, 45.
-       2. Create special files
+       3.6.2 Create special files
 	  Run /moxa/mxser/driver/msmknod to create special files with
 	  specified major numbers.
-       3. Modify driver with new major number
+       3.6.3 Modify driver with new major number
 	  Run vi to open /moxa/mxser/driver/mxser.c. Locate the line
 	  contains "MXSERMAJOR". Change the content as below.
 	  #define	  MXSERMAJOR		  40
 	  #define	  MXSERCUMAJOR		  45
-       4. Run # make install in /moxa/mxser/driver.
+       3.6.4 Run "make clean; make install" in /moxa/mxser/driver.
 
-   3.6 Verify driver installation
+   3.7 Verify driver installation
        You may refer to /var/log/messages to check the latest status
        log reported by this driver whenever it's activated.
+
 -----------------------------------------------------------------------------
 4. Utilities
    There are 3 utilities contained in this driver. They are msdiag, msmon and
    msterm. These 3 utilities are released in form of source code. They should
    be compiled into executable file and copied into /usr/bin.
 
+   Before using these utilities, please load driver (refer 3.4 & 3.5) and
+   make sure you had run the "msmknod" utility.
+
    msdiag - Diagnostic
    --------------------
-   This utility provides the function to detect what Moxa Smartio multiport
-   board exists in the system.
+   This utility provides the function to display what Moxa Smartio/Industio
+   board found by driver in the system.
 
    msmon - Port Monitoring
    -----------------------
@@ -353,12 +454,13 @@ Content
    application, for example, sending AT command to a modem connected to the
    port or used as a terminal for login purpose. Note that this is only a
    dumb terminal emulation without handling full screen operation.
+
 -----------------------------------------------------------------------------
 5. Setserial
 
    Supported Setserial parameters are listed as below.
 
-   uart 	  set UART type(16450-->disable FIFO, 16550A-->enable FIFO)
+   uart		  set UART type(16450-->disable FIFO, 16550A-->enable FIFO)
    close_delay	  set the amount of time(in 1/100 of a second) that DTR
 		  should be kept low while being closed.
    closing_wait   set the amount of time(in 1/100 of a second) that the
@@ -366,7 +468,13 @@ Content
 		  being closed, before the receiver is disable.
    spd_hi	  Use  57.6kb  when  the application requests 38.4kb.
    spd_vhi	  Use  115.2kb	when  the application requests 38.4kb.
+   spd_shi	  Use  230.4kb	when  the application requests 38.4kb.
+   spd_warp	  Use  460.8kb	when  the application requests 38.4kb.
    spd_normal	  Use  38.4kb  when  the application requests 38.4kb.
+   spd_cust	  Use  the custom divisor to set the speed when  the
+		  application requests 38.4kb.
+   divisor	  This option set the custom divison.
+   baud_base	  This option set the base baud rate.
 
 -----------------------------------------------------------------------------
 6. Troubleshooting
@@ -375,8 +483,9 @@ Content
    possible. If all the possible solutions fail, please contact our technical
    support team to get more help.
 
-   Error msg: More than 4 Moxa Smartio family boards found. Fifth board and
-	      after are ignored.
+
+   Error msg: More than 4 Moxa Smartio/Industio family boards found. Fifth board
+              and after are ignored.
    Solution:
    To avoid this problem, please unplug fifth and after board, because Moxa
    driver supports up to 4 boards.
@@ -384,7 +493,7 @@ Content
    Error msg: Request_irq fail, IRQ(?) may be conflict with another device.
    Solution:
    Other PCI or ISA devices occupy the assigned IRQ. If you are not sure
-   which device causes the situation,please check /proc/interrupts to find
+   which device causes the situation, please check /proc/interrupts to find
    free IRQ and simply change another free IRQ for Moxa board.
 
    Error msg: Board #: C1xx Series(CAP=xxx) interrupt number invalid.
@@ -397,15 +506,18 @@ Content
    Moxa ISA board needs an interrupt vector.Please refer to user's manual
    "Hardware Installation" chapter to set interrupt vector.
 
-   Error msg: Couldn't install MOXA Smartio family driver!
+   Error msg: Couldn't install MOXA Smartio/Industio family driver!
    Solution:
    Load Moxa driver fail, the major number may conflict with other devices.
-   Please refer to previous section 3.5 to change a free major number for
+   Please refer to previous section 3.7 to change a free major number for
    Moxa driver.
 
-   Error msg: Couldn't install MOXA Smartio family callout driver!
+   Error msg: Couldn't install MOXA Smartio/Industio family callout driver!
    Solution:
    Load Moxa callout driver fail, the callout device major number may
-   conflict with other devices. Please refer to previous section 3.5 to
+   conflict with other devices. Please refer to previous section 3.7 to
    change a free callout device major number for Moxa driver.
+
+
 -----------------------------------------------------------------------------
+

From 83766bc63f7e49b0215811026e7802bd09a9c7e1 Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Fri, 25 Jul 2008 01:48:21 -0700
Subject: [PATCH 0881/1171] Char: mxser, prints cleanup

- use dev_* for printing in pci probe function
- move ISA p[rints directly into isa find function, do not postpone it.
  Remove macros bound to it then.
- prepend some prints by "mxser: " to know what it belongs to

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Acked-by: Alan Cox <alan@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/mxser.c | 80 ++++++++++++++++++++------------------------
 1 file changed, 37 insertions(+), 43 deletions(-)

diff --git a/drivers/char/mxser.c b/drivers/char/mxser.c
index 3d7f2a970495..57570f7db2be 100644
--- a/drivers/char/mxser.c
+++ b/drivers/char/mxser.c
@@ -55,11 +55,6 @@
 #define MXSER_PORTS		(MXSER_BOARDS * MXSER_PORTS_PER_BOARD)
 #define MXSER_ISR_PASS_LIMIT	100
 
-#define	MXSER_ERR_IOADDR	-1
-#define	MXSER_ERR_IRQ		-2
-#define	MXSER_ERR_IRQ_CONFLIT	-3
-#define	MXSER_ERR_VECTOR	-4
-
 /*CheckIsMoxaMust return value*/
 #define MOXA_OTHER_UART		0x00
 #define MOXA_MUST_MU150_HWID	0x01
@@ -2481,7 +2476,8 @@ static int __devinit mxser_initbrd(struct mxser_board *brd,
 	unsigned int i;
 	int retval;
 
-	printk(KERN_INFO "max. baud rate = %d bps.\n", brd->ports[0].max_baud);
+	printk(KERN_INFO "mxser: max. baud rate = %d bps\n",
+			brd->ports[0].max_baud);
 
 	for (i = 0; i < brd->info->nports; i++) {
 		info = &brd->ports[i];
@@ -2564,28 +2560,32 @@ static int __init mxser_get_ISA_conf(int cap, struct mxser_board *brd)
 		irq = regs[9] & 0xF000;
 		irq = irq | (irq >> 4);
 		if (irq != (regs[9] & 0xFF00))
-			return MXSER_ERR_IRQ_CONFLIT;
+			goto err_irqconflict;
 	} else if (brd->info->nports == 4) {
 		irq = regs[9] & 0xF000;
 		irq = irq | (irq >> 4);
 		irq = irq | (irq >> 8);
 		if (irq != regs[9])
-			return MXSER_ERR_IRQ_CONFLIT;
+			goto err_irqconflict;
 	} else if (brd->info->nports == 8) {
 		irq = regs[9] & 0xF000;
 		irq = irq | (irq >> 4);
 		irq = irq | (irq >> 8);
 		if ((irq != regs[9]) || (irq != regs[10]))
-			return MXSER_ERR_IRQ_CONFLIT;
+			goto err_irqconflict;
 	}
 
-	if (!irq)
-		return MXSER_ERR_IRQ;
+	if (!irq) {
+		printk(KERN_ERR "mxser: interrupt number unset\n");
+		return -EIO;
+	}
 	brd->irq = ((int)(irq & 0xF000) >> 12);
 	for (i = 0; i < 8; i++)
 		brd->ports[i].ioaddr = (int) regs[i + 1] & 0xFFF8;
-	if ((regs[12] & 0x80) == 0)
-		return MXSER_ERR_VECTOR;
+	if ((regs[12] & 0x80) == 0) {
+		printk(KERN_ERR "mxser: invalid interrupt vector\n");
+		return -EIO;
+	}
 	brd->vector = (int)regs[11];	/* interrupt vector */
 	if (id == 1)
 		brd->vector_mask = 0x00FF;
@@ -2612,13 +2612,26 @@ static int __init mxser_get_ISA_conf(int cap, struct mxser_board *brd)
 	else
 		brd->uart_type = PORT_16450;
 	if (!request_region(brd->ports[0].ioaddr, 8 * brd->info->nports,
-			"mxser(IO)"))
-		return MXSER_ERR_IOADDR;
+			"mxser(IO)")) {
+		printk(KERN_ERR "mxser: can't request ports I/O region: "
+				"0x%.8lx-0x%.8lx\n",
+				brd->ports[0].ioaddr, brd->ports[0].ioaddr +
+				8 * brd->info->nports - 1);
+		return -EIO;
+	}
 	if (!request_region(brd->vector, 1, "mxser(vector)")) {
 		release_region(brd->ports[0].ioaddr, 8 * brd->info->nports);
-		return MXSER_ERR_VECTOR;
+		printk(KERN_ERR "mxser: can't request interrupt vector region: "
+				"0x%.8lx-0x%.8lx\n",
+				brd->ports[0].ioaddr, brd->ports[0].ioaddr +
+				8 * brd->info->nports - 1);
+		return -EIO;
 	}
 	return brd->info->nports;
+
+err_irqconflict:
+	printk(KERN_ERR "mxser: invalid interrupt number\n");
+	return -EIO;
 }
 
 static int __devinit mxser_probe(struct pci_dev *pdev,
@@ -2635,20 +2648,20 @@ static int __devinit mxser_probe(struct pci_dev *pdev,
 			break;
 
 	if (i >= MXSER_BOARDS) {
-		printk(KERN_ERR "Too many Smartio/Industio family boards found "
-			"(maximum %d), board not configured\n", MXSER_BOARDS);
+		dev_err(&pdev->dev, "too many boards found (maximum %d), board "
+				"not configured\n", MXSER_BOARDS);
 		goto err;
 	}
 
 	brd = &mxser_boards[i];
 	brd->idx = i * MXSER_PORTS_PER_BOARD;
-	printk(KERN_INFO "Found MOXA %s board (BusNo=%d, DevNo=%d)\n",
+	dev_info(&pdev->dev, "found MOXA %s board (BusNo=%d, DevNo=%d)\n",
 		mxser_cards[ent->driver_data].name,
 		pdev->bus->number, PCI_SLOT(pdev->devfn));
 
 	retval = pci_enable_device(pdev);
 	if (retval) {
-		printk(KERN_ERR "Moxa SmartI/O PCI enable fail !\n");
+		dev_err(&pdev->dev, "PCI enable failed\n");
 		goto err;
 	}
 
@@ -2798,33 +2811,14 @@ static int __init mxser_module_init(void)
 
 			brd = &mxser_boards[m];
 			retval = mxser_get_ISA_conf(cap, brd);
-
-			if (retval != 0)
-				printk(KERN_INFO "Found MOXA %s board "
-					"(CAP=0x%x)\n",
-					brd->info->name, ioaddr[b]);
-
 			if (retval <= 0) {
-				if (retval == MXSER_ERR_IRQ)
-					printk(KERN_ERR "Invalid interrupt "
-						"number, board not "
-						"configured\n");
-				else if (retval == MXSER_ERR_IRQ_CONFLIT)
-					printk(KERN_ERR "Invalid interrupt "
-						"number, board not "
-						"configured\n");
-				else if (retval == MXSER_ERR_VECTOR)
-					printk(KERN_ERR "Invalid interrupt "
-						"vector, board not "
-						"configured\n");
-				else if (retval == MXSER_ERR_IOADDR)
-					printk(KERN_ERR "Invalid I/O address, "
-						"board not configured\n");
-
 				brd->info = NULL;
 				continue;
 			}
 
+			printk(KERN_INFO "mxser: found MOXA %s board "
+				"(CAP=0x%x)\n",	brd->info->name, ioaddr[b]);
+
 			/* mxser_initbrd will hook ISR. */
 			if (mxser_initbrd(brd, NULL) < 0) {
 				brd->info = NULL;
@@ -2841,7 +2835,7 @@ static int __init mxser_module_init(void)
 
 	retval = pci_register_driver(&mxser_driver);
 	if (retval) {
-		printk(KERN_ERR "Can't register pci driver\n");
+		printk(KERN_ERR "mxser: can't register pci driver\n");
 		if (!m) {
 			retval = -ENODEV;
 			goto err_unr;

From 1df0092477b8b2df605812e298624f5c35bb4805 Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Fri, 25 Jul 2008 01:48:22 -0700
Subject: [PATCH 0882/1171] Char: mxser, remove predefined isa support

Remove a support of ISA addresses predefined at compile time.  It is
unused (filled by zeroes) and prolongs the code.  Don't initialize global
array and add `ioaddr' module param description.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Acked-by: Alan Cox <alan@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/mxser.c | 71 ++++++++++++++++++--------------------------
 1 file changed, 29 insertions(+), 42 deletions(-)

diff --git a/drivers/char/mxser.c b/drivers/char/mxser.c
index 57570f7db2be..9b4d03cf4e1d 100644
--- a/drivers/char/mxser.c
+++ b/drivers/char/mxser.c
@@ -173,14 +173,15 @@ static struct pci_device_id mxser_pcibrds[] = {
 };
 MODULE_DEVICE_TABLE(pci, mxser_pcibrds);
 
-static int ioaddr[MXSER_BOARDS] = { 0, 0, 0, 0 };
+static unsigned long ioaddr[MXSER_BOARDS];
 static int ttymajor = MXSERMAJOR;
 
 /* Variables for insmod */
 
 MODULE_AUTHOR("Casper Yang");
 MODULE_DESCRIPTION("MOXA Smartio/Industio Family Multiport Board Device Driver");
-module_param_array(ioaddr, int, NULL, 0);
+module_param_array(ioaddr, ulong, NULL, 0);
+MODULE_PARM_DESC(ioaddr, "ISA io addresses to look for a moxa board");
 module_param(ttymajor, int, 0);
 MODULE_LICENSE("GPL");
 
@@ -281,11 +282,6 @@ struct mxser_mstatus {
 	int dcd;
 };
 
-static int mxserBoardCAP[MXSER_BOARDS] = {
-	0, 0, 0, 0
-	/*  0x180, 0x280, 0x200, 0x320 */
-};
-
 static struct mxser_board mxser_boards[MXSER_BOARDS];
 static struct tty_driver *mxvar_sdriver;
 static struct mxser_log mxvar_log;
@@ -2763,9 +2759,8 @@ static struct pci_driver mxser_driver = {
 static int __init mxser_module_init(void)
 {
 	struct mxser_board *brd;
-	unsigned long cap;
-	unsigned int i, m, isaloop;
-	int retval, b;
+	unsigned int b, i, m;
+	int retval;
 
 	pr_debug("Loading module mxser ...\n");
 
@@ -2797,42 +2792,34 @@ static int __init mxser_module_init(void)
 		goto err_put;
 	}
 
-	m = 0;
 	/* Start finding ISA boards here */
-	for (isaloop = 0; isaloop < 2; isaloop++)
-		for (b = 0; b < MXSER_BOARDS && m < MXSER_BOARDS; b++) {
-			if (!isaloop)
-				cap = mxserBoardCAP[b]; /* predefined */
-			else
-				cap = ioaddr[b]; /* module param */
+	for (m = 0, b = 0; b < MXSER_BOARDS; b++) {
+		if (!ioaddr[b])
+			continue;
 
-			if (!cap)
-				continue;
-
-			brd = &mxser_boards[m];
-			retval = mxser_get_ISA_conf(cap, brd);
-			if (retval <= 0) {
-				brd->info = NULL;
-				continue;
-			}
-
-			printk(KERN_INFO "mxser: found MOXA %s board "
-				"(CAP=0x%x)\n",	brd->info->name, ioaddr[b]);
-
-			/* mxser_initbrd will hook ISR. */
-			if (mxser_initbrd(brd, NULL) < 0) {
-				brd->info = NULL;
-				continue;
-			}
-
-			brd->idx = m * MXSER_PORTS_PER_BOARD;
-			for (i = 0; i < brd->info->nports; i++)
-				tty_register_device(mxvar_sdriver, brd->idx + i,
-						NULL);
-
-			m++;
+		brd = &mxser_boards[m];
+		retval = mxser_get_ISA_conf(!ioaddr[b], brd);
+		if (retval <= 0) {
+			brd->info = NULL;
+			continue;
 		}
 
+		printk(KERN_INFO "mxser: found MOXA %s board (CAP=0x%lx)\n",
+				brd->info->name, ioaddr[b]);
+
+		/* mxser_initbrd will hook ISR. */
+		if (mxser_initbrd(brd, NULL) < 0) {
+			brd->info = NULL;
+			continue;
+		}
+
+		brd->idx = m * MXSER_PORTS_PER_BOARD;
+		for (i = 0; i < brd->info->nports; i++)
+			tty_register_device(mxvar_sdriver, brd->idx + i, NULL);
+
+		m++;
+	}
+
 	retval = pci_register_driver(&mxser_driver);
 	if (retval) {
 		printk(KERN_ERR "mxser: can't register pci driver\n");

From ace7dd96695769f9d76980c7e52139e73228221c Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Fri, 25 Jul 2008 01:48:22 -0700
Subject: [PATCH 0883/1171] Char: mxser, various cleanups

- remove unused macro
- some whitespace cleanup
- useless debug prints removal

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Acked-by: Alan Cox <alan@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/mxser.c | 24 ++++--------------------
 1 file changed, 4 insertions(+), 20 deletions(-)

diff --git a/drivers/char/mxser.c b/drivers/char/mxser.c
index 9b4d03cf4e1d..e30575e87648 100644
--- a/drivers/char/mxser.c
+++ b/drivers/char/mxser.c
@@ -48,7 +48,6 @@
 
 #define	MXSER_VERSION	"2.0.4"		/* 1.12 */
 #define	MXSERMAJOR	 174
-#define	MXSERCUMAJOR	 175
 
 #define MXSER_BOARDS		4	/* Max. boards */
 #define MXSER_PORTS_PER_BOARD	8	/* Max. ports per board */
@@ -191,7 +190,6 @@ struct mxser_log {
 	unsigned long txcnt[MXSER_PORTS];
 };
 
-
 struct mxser_mon {
 	unsigned long rxcnt;
 	unsigned long txcnt;
@@ -1305,13 +1303,9 @@ static void mxser_flush_chars(struct tty_struct *tty)
 	struct mxser_port *info = tty->driver_data;
 	unsigned long flags;
 
-	if (info->xmit_cnt <= 0 ||
-			tty->stopped ||
-			!info->port.xmit_buf ||
-			(tty->hw_stopped &&
-			 (info->type != PORT_16550A) &&
-			 (!info->board->chip_flag)
-			))
+	if (info->xmit_cnt <= 0 || tty->stopped || !info->port.xmit_buf ||
+			(tty->hw_stopped && info->type != PORT_16550A &&
+			 !info->board->chip_flag))
 		return;
 
 	spin_lock_irqsave(&info->slock, flags);
@@ -1329,9 +1323,7 @@ static int mxser_write_room(struct tty_struct *tty)
 	int ret;
 
 	ret = SERIAL_XMIT_SIZE - info->xmit_cnt - 1;
-	if (ret < 0)
-		ret = 0;
-	return ret;
+	return ret < 0 ? 0 : ret;
 }
 
 static int mxser_chars_in_buffer(struct tty_struct *tty)
@@ -2762,8 +2754,6 @@ static int __init mxser_module_init(void)
 	unsigned int b, i, m;
 	int retval;
 
-	pr_debug("Loading module mxser ...\n");
-
 	mxvar_sdriver = alloc_tty_driver(MXSER_PORTS + 1);
 	if (!mxvar_sdriver)
 		return -ENOMEM;
@@ -2829,8 +2819,6 @@ static int __init mxser_module_init(void)
 		} /* else: we have some ISA cards under control */
 	}
 
-	pr_debug("Done.\n");
-
 	return 0;
 err_unr:
 	tty_unregister_driver(mxvar_sdriver);
@@ -2843,8 +2831,6 @@ static void __exit mxser_module_exit(void)
 {
 	unsigned int i, j;
 
-	pr_debug("Unloading module mxser ...\n");
-
 	pci_unregister_driver(&mxser_driver);
 
 	for (i = 0; i < MXSER_BOARDS; i++) /* ISA remains */
@@ -2858,8 +2844,6 @@ static void __exit mxser_module_exit(void)
 	for (i = 0; i < MXSER_BOARDS; i++)
 		if (mxser_boards[i].info != NULL)
 			mxser_release_res(&mxser_boards[i], NULL, 1);
-
-	pr_debug("Done.\n");
 }
 
 module_init(mxser_module_init);

From ec905a18656daa4d9300bad2bebc02d5dba7883d Mon Sep 17 00:00:00 2001
From: Jiri Slaby <jirislaby@gmail.com>
Date: Fri, 25 Jul 2008 01:48:23 -0700
Subject: [PATCH 0884/1171] drivers/misc/phantom: note PCI

Tell users that the driver is only for PCI devices to stop asking for
support of firewire and parallel devices.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/misc/Kconfig   | 4 +++-
 drivers/misc/phantom.c | 2 +-
 2 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig
index 7e37ba5afe30..321eb9134635 100644
--- a/drivers/misc/Kconfig
+++ b/drivers/misc/Kconfig
@@ -77,11 +77,13 @@ config IBM_ASM
 	  for your IBM server.
 
 config PHANTOM
-	tristate "Sensable PHANToM"
+	tristate "Sensable PHANToM (PCI)"
 	depends on PCI
 	help
 	  Say Y here if you want to build a driver for Sensable PHANToM device.
 
+	  This driver is only for PCI PHANToMs.
+
 	  If you choose to build module, its name will be phantom. If unsure,
 	  say N here.
 
diff --git a/drivers/misc/phantom.c b/drivers/misc/phantom.c
index 4ce3bdc2f959..daf585689ce3 100644
--- a/drivers/misc/phantom.c
+++ b/drivers/misc/phantom.c
@@ -563,6 +563,6 @@ module_init(phantom_init);
 module_exit(phantom_exit);
 
 MODULE_AUTHOR("Jiri Slaby <jirislaby@gmail.com>");
-MODULE_DESCRIPTION("Sensable Phantom driver");
+MODULE_DESCRIPTION("Sensable Phantom driver (PCI devices)");
 MODULE_LICENSE("GPL");
 MODULE_VERSION(PHANTOM_VERSION);

From f37e66173e0cc09b4e5a89eb0294abbefc15f435 Mon Sep 17 00:00:00 2001
From: Akinobu Mita <akinobu.mita@gmail.com>
Date: Fri, 25 Jul 2008 01:48:23 -0700
Subject: [PATCH 0885/1171] firmware: use memory_read_from_buffer()

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Greg Kroah-Hartman <gregkh@suse.de>
Cc: Markus Rechberger <markus.rechberger@amd.com>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/base/firmware_class.c | 12 +++---------
 1 file changed, 3 insertions(+), 9 deletions(-)

diff --git a/drivers/base/firmware_class.c b/drivers/base/firmware_class.c
index b0be1d18fee2..c9c92b00fd55 100644
--- a/drivers/base/firmware_class.c
+++ b/drivers/base/firmware_class.c
@@ -184,7 +184,7 @@ firmware_data_read(struct kobject *kobj, struct bin_attribute *bin_attr,
 	struct device *dev = to_dev(kobj);
 	struct firmware_priv *fw_priv = dev_get_drvdata(dev);
 	struct firmware *fw;
-	ssize_t ret_count = count;
+	ssize_t ret_count;
 
 	mutex_lock(&fw_lock);
 	fw = fw_priv->fw;
@@ -192,14 +192,8 @@ firmware_data_read(struct kobject *kobj, struct bin_attribute *bin_attr,
 		ret_count = -ENODEV;
 		goto out;
 	}
-	if (offset > fw->size) {
-		ret_count = 0;
-		goto out;
-	}
-	if (offset + ret_count > fw->size)
-		ret_count = fw->size - offset;
-
-	memcpy(buffer, fw->data + offset, ret_count);
+	ret_count = memory_read_from_buffer(buffer, count, &offset,
+						fw->data, fw->size);
 out:
 	mutex_unlock(&fw_lock);
 	return ret_count;

From abe19b7b822a8fdbe3dbfd6e066d0698b4eefb06 Mon Sep 17 00:00:00 2001
From: Akinobu Mita <akinobu.mita@gmail.com>
Date: Fri, 25 Jul 2008 01:48:24 -0700
Subject: [PATCH 0886/1171] dcdbas: use memory_read_from_buffer()

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Doug Warzecha <Douglas_Warzecha@dell.com>
Cc: Zhang Rui <rui.zhang@intel.com>
Cc: Matt Domsch <Matt_Domsch@dell.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/firmware/dcdbas.c | 13 ++-----------
 1 file changed, 2 insertions(+), 11 deletions(-)

diff --git a/drivers/firmware/dcdbas.c b/drivers/firmware/dcdbas.c
index 0b624e927a6f..c66817e7717b 100644
--- a/drivers/firmware/dcdbas.c
+++ b/drivers/firmware/dcdbas.c
@@ -152,20 +152,11 @@ static ssize_t smi_data_read(struct kobject *kobj,
 			     struct bin_attribute *bin_attr,
 			     char *buf, loff_t pos, size_t count)
 {
-	size_t max_read;
 	ssize_t ret;
 
 	mutex_lock(&smi_data_lock);
-
-	if (pos >= smi_data_buf_size) {
-		ret = 0;
-		goto out;
-	}
-
-	max_read = smi_data_buf_size - pos;
-	ret = min(max_read, count);
-	memcpy(buf, smi_data_buf + pos, ret);
-out:
+	ret = memory_read_from_buffer(buf, count, &pos, smi_data_buf,
+					smi_data_buf_size);
 	mutex_unlock(&smi_data_lock);
 	return ret;
 }

From d805dda412346225a50af2d399d958a4bc676c38 Mon Sep 17 00:00:00 2001
From: Abdel Benamrouche <draconux@gmail.com>
Date: Fri, 25 Jul 2008 01:48:25 -0700
Subject: [PATCH 0887/1171] fs/partition/check.c: fix return value warning

fs/partitions/check.c:381: warning: ignoring return value of ___device_add___,
  declared with attribute warn_unused_result

[akpm@linux-foundation.org: multiple-return-statements-per-function are evil]
Signed-off-by: Abdel Benamrouche <draconux@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/partitions/check.c | 28 ++++++++++++++++++++++------
 include/linux/genhd.h |  2 +-
 2 files changed, 23 insertions(+), 7 deletions(-)

diff --git a/fs/partitions/check.c b/fs/partitions/check.c
index efef715135d3..2e6413fbd2d8 100644
--- a/fs/partitions/check.c
+++ b/fs/partitions/check.c
@@ -344,18 +344,18 @@ static ssize_t whole_disk_show(struct device *dev,
 static DEVICE_ATTR(whole_disk, S_IRUSR | S_IRGRP | S_IROTH,
 		   whole_disk_show, NULL);
 
-void add_partition(struct gendisk *disk, int part, sector_t start, sector_t len, int flags)
+int add_partition(struct gendisk *disk, int part, sector_t start, sector_t len, int flags)
 {
 	struct hd_struct *p;
 	int err;
 
 	p = kzalloc(sizeof(*p), GFP_KERNEL);
 	if (!p)
-		return;
+		return -ENOMEM;
 
 	if (!init_part_stats(p)) {
-		kfree(p);
-		return;
+		err = -ENOMEM;
+		goto out0;
 	}
 	p->start_sect = start;
 	p->nr_sects = len;
@@ -378,15 +378,31 @@ void add_partition(struct gendisk *disk, int part, sector_t start, sector_t len,
 
 	/* delay uevent until 'holders' subdir is created */
 	p->dev.uevent_suppress = 1;
-	device_add(&p->dev);
+	err = device_add(&p->dev);
+	if (err)
+		goto out1;
 	partition_sysfs_add_subdir(p);
 	p->dev.uevent_suppress = 0;
-	if (flags & ADDPART_FLAG_WHOLEDISK)
+	if (flags & ADDPART_FLAG_WHOLEDISK) {
 		err = device_create_file(&p->dev, &dev_attr_whole_disk);
+		if (err)
+			goto out2;
+	}
 
 	/* suppress uevent if the disk supresses it */
 	if (!disk->dev.uevent_suppress)
 		kobject_uevent(&p->dev.kobj, KOBJ_ADD);
+
+	return 0;
+
+out2:
+	device_del(&p->dev);
+out1:
+	put_device(&p->dev);
+	free_part_stats(p);
+out0:
+	kfree(p);
+	return err;
 }
 
 /* Not exported, helper to add_disk(). */
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index e8787417f65a..118216f1bd3c 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -541,7 +541,7 @@ extern dev_t blk_lookup_devt(const char *name, int part);
 extern char *disk_name (struct gendisk *hd, int part, char *buf);
 
 extern int rescan_partitions(struct gendisk *disk, struct block_device *bdev);
-extern void add_partition(struct gendisk *, int, sector_t, sector_t, int);
+extern int __must_check add_partition(struct gendisk *, int, sector_t, sector_t, int);
 extern void delete_partition(struct gendisk *, int);
 extern void printk_all_partitions(void);
 

From 04ebd4aee52b06a2c38127d9208546e5b96f3a19 Mon Sep 17 00:00:00 2001
From: Abdel Benamrouche <draconux@gmail.com>
Date: Fri, 25 Jul 2008 01:48:26 -0700
Subject: [PATCH 0888/1171] block/ioctl.c and fs/partition/check.c: check value
 returned by add_partition()

Now that add_partition() has been aught to propagate errors, let's check them.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Abdel Benamrouche <draconux@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 block/ioctl.c         |  5 +++--
 fs/partitions/check.c | 10 ++++++++--
 2 files changed, 11 insertions(+), 4 deletions(-)

diff --git a/block/ioctl.c b/block/ioctl.c
index 52d6385216ad..77185e5c026a 100644
--- a/block/ioctl.c
+++ b/block/ioctl.c
@@ -17,6 +17,7 @@ static int blkpg_ioctl(struct block_device *bdev, struct blkpg_ioctl_arg __user
 	long long start, length;
 	int part;
 	int i;
+	int err;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EACCES;
@@ -61,9 +62,9 @@ static int blkpg_ioctl(struct block_device *bdev, struct blkpg_ioctl_arg __user
 				}
 			}
 			/* all seems OK */
-			add_partition(disk, part, start, length, ADDPART_FLAG_NONE);
+			err = add_partition(disk, part, start, length, ADDPART_FLAG_NONE);
 			mutex_unlock(&bdev->bd_mutex);
-			return 0;
+			return err;
 		case BLKPG_DEL_PARTITION:
 			if (!disk->part[part-1])
 				return -ENXIO;
diff --git a/fs/partitions/check.c b/fs/partitions/check.c
index 2e6413fbd2d8..7d6b34e201db 100644
--- a/fs/partitions/check.c
+++ b/fs/partitions/check.c
@@ -499,10 +499,16 @@ int rescan_partitions(struct gendisk *disk, struct block_device *bdev)
 		if (!size)
 			continue;
 		if (from + size > get_capacity(disk)) {
-			printk(" %s: p%d exceeds device capacity\n",
+			printk(KERN_ERR " %s: p%d exceeds device capacity\n",
 				disk->disk_name, p);
+			continue;
+		}
+		res = add_partition(disk, p, from, size, state->parts[p].flags);
+		if (res) {
+			printk(KERN_ERR " %s: p%d could not be added: %d\n",
+				disk->disk_name, p, -res);
+			continue;
 		}
-		add_partition(disk, p, from, size, state->parts[p].flags);
 #ifdef CONFIG_BLK_DEV_MD
 		if (state->parts[p].flags & ADDPART_FLAG_RAID)
 			md_autodetect_dev(bdev->bd_dev+p);

From d991696263a704be7f41ac186f1a0ed17963c260 Mon Sep 17 00:00:00 2001
From: Thomas Gleixner <tglx@linutronix.de>
Date: Fri, 25 Jul 2008 01:48:26 -0700
Subject: [PATCH 0889/1171] fs/partitions/efi: convert to pr_debug

convert the local Dprintk() compile time debug printk wrappers to the
generic pr_debug() wrapper.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Matt Domsch <Matt_Domsch@dell.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/partitions/efi.c | 42 +++++++++++++++++-------------------------
 1 file changed, 17 insertions(+), 25 deletions(-)

diff --git a/fs/partitions/efi.c b/fs/partitions/efi.c
index e7b07006bc41..038a6022152f 100644
--- a/fs/partitions/efi.c
+++ b/fs/partitions/efi.c
@@ -95,13 +95,6 @@
 #include "check.h"
 #include "efi.h"
 
-#undef EFI_DEBUG
-#ifdef EFI_DEBUG
-#define Dprintk(x...) printk(KERN_DEBUG x)
-#else
-#define Dprintk(x...)
-#endif
-
 /* This allows a kernel command line option 'gpt' to override
  * the test for invalid PMBR.  Not __initdata because reloading
  * the partition tables happens after init too.
@@ -305,10 +298,10 @@ is_gpt_valid(struct block_device *bdev, u64 lba,
 
 	/* Check the GUID Partition Table signature */
 	if (le64_to_cpu((*gpt)->signature) != GPT_HEADER_SIGNATURE) {
-		Dprintk("GUID Partition Table Header signature is wrong:"
-			"%lld != %lld\n",
-			(unsigned long long)le64_to_cpu((*gpt)->signature),
-			(unsigned long long)GPT_HEADER_SIGNATURE);
+		pr_debug("GUID Partition Table Header signature is wrong:"
+			 "%lld != %lld\n",
+			 (unsigned long long)le64_to_cpu((*gpt)->signature),
+			 (unsigned long long)GPT_HEADER_SIGNATURE);
 		goto fail;
 	}
 
@@ -318,9 +311,8 @@ is_gpt_valid(struct block_device *bdev, u64 lba,
 	crc = efi_crc32((const unsigned char *) (*gpt), le32_to_cpu((*gpt)->header_size));
 
 	if (crc != origcrc) {
-		Dprintk
-		    ("GUID Partition Table Header CRC is wrong: %x != %x\n",
-		     crc, origcrc);
+		pr_debug("GUID Partition Table Header CRC is wrong: %x != %x\n",
+			 crc, origcrc);
 		goto fail;
 	}
 	(*gpt)->header_crc32 = cpu_to_le32(origcrc);
@@ -328,9 +320,9 @@ is_gpt_valid(struct block_device *bdev, u64 lba,
 	/* Check that the my_lba entry points to the LBA that contains
 	 * the GUID Partition Table */
 	if (le64_to_cpu((*gpt)->my_lba) != lba) {
-		Dprintk("GPT my_lba incorrect: %lld != %lld\n",
-			(unsigned long long)le64_to_cpu((*gpt)->my_lba),
-			(unsigned long long)lba);
+		pr_debug("GPT my_lba incorrect: %lld != %lld\n",
+			 (unsigned long long)le64_to_cpu((*gpt)->my_lba),
+			 (unsigned long long)lba);
 		goto fail;
 	}
 
@@ -339,15 +331,15 @@ is_gpt_valid(struct block_device *bdev, u64 lba,
 	 */
 	lastlba = last_lba(bdev);
 	if (le64_to_cpu((*gpt)->first_usable_lba) > lastlba) {
-		Dprintk("GPT: first_usable_lba incorrect: %lld > %lld\n",
-			(unsigned long long)le64_to_cpu((*gpt)->first_usable_lba),
-			(unsigned long long)lastlba);
+		pr_debug("GPT: first_usable_lba incorrect: %lld > %lld\n",
+			 (unsigned long long)le64_to_cpu((*gpt)->first_usable_lba),
+			 (unsigned long long)lastlba);
 		goto fail;
 	}
 	if (le64_to_cpu((*gpt)->last_usable_lba) > lastlba) {
-		Dprintk("GPT: last_usable_lba incorrect: %lld > %lld\n",
-			(unsigned long long)le64_to_cpu((*gpt)->last_usable_lba),
-			(unsigned long long)lastlba);
+		pr_debug("GPT: last_usable_lba incorrect: %lld > %lld\n",
+			 (unsigned long long)le64_to_cpu((*gpt)->last_usable_lba),
+			 (unsigned long long)lastlba);
 		goto fail;
 	}
 
@@ -360,7 +352,7 @@ is_gpt_valid(struct block_device *bdev, u64 lba,
 			le32_to_cpu((*gpt)->sizeof_partition_entry));
 
 	if (crc != le32_to_cpu((*gpt)->partition_entry_array_crc32)) {
-		Dprintk("GUID Partitition Entry Array CRC check failed.\n");
+		pr_debug("GUID Partitition Entry Array CRC check failed.\n");
 		goto fail_ptes;
 	}
 
@@ -616,7 +608,7 @@ efi_partition(struct parsed_partitions *state, struct block_device *bdev)
 		return 0;
 	}
 
-	Dprintk("GUID Partition Table is valid!  Yea!\n");
+	pr_debug("GUID Partition Table is valid!  Yea!\n");
 
 	for (i = 0; i < le32_to_cpu(gpt->num_partition_entries) && i < state->limit-1; i++) {
 		if (!is_pte_valid(&ptes[i], last_lba(bdev)))

From 25377479de7539fdc871a0f0ecaa39da42353bbc Mon Sep 17 00:00:00 2001
From: Akinobu Mita <akinobu.mita@gmail.com>
Date: Fri, 25 Jul 2008 01:48:27 -0700
Subject: [PATCH 0890/1171] dell_rbu: use memory_read_from_buffer()

Signed-off-by: Akinobu Mita <akinobu.mita@gmail.com>
Cc: Abhay Salunke <Abhay_Salunke@dell.com>
Cc: Zhang Rui <rui.zhang@intel.com>
Cc: Matt Domsch <Matt_Domsch@dell.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/firmware/dell_rbu.c | 28 +++-------------------------
 1 file changed, 3 insertions(+), 25 deletions(-)

diff --git a/drivers/firmware/dell_rbu.c b/drivers/firmware/dell_rbu.c
index 7430e218cda6..13946ebd77d6 100644
--- a/drivers/firmware/dell_rbu.c
+++ b/drivers/firmware/dell_rbu.c
@@ -507,11 +507,6 @@ static ssize_t read_packet_data(char *buffer, loff_t pos, size_t count)
 
 static ssize_t read_rbu_mono_data(char *buffer, loff_t pos, size_t count)
 {
-	unsigned char *ptemp = NULL;
-	size_t bytes_left = 0;
-	size_t data_length = 0;
-	ssize_t ret_count = 0;
-
 	/* check to see if we have something to return */
 	if ((rbu_data.image_update_buffer == NULL) ||
 		(rbu_data.bios_image_size == 0)) {
@@ -519,28 +514,11 @@ static ssize_t read_rbu_mono_data(char *buffer, loff_t pos, size_t count)
 			"bios_image_size %lu\n",
 			rbu_data.image_update_buffer,
 			rbu_data.bios_image_size);
-		ret_count = -ENOMEM;
-		goto read_rbu_data_exit;
+		return -ENOMEM;
 	}
 
-	if (pos > rbu_data.bios_image_size) {
-		ret_count = 0;
-		goto read_rbu_data_exit;
-	}
-
-	bytes_left = rbu_data.bios_image_size - pos;
-	data_length = min(bytes_left, count);
-
-	ptemp = rbu_data.image_update_buffer;
-	memcpy(buffer, (ptemp + pos), data_length);
-
-	if ((pos + count) > rbu_data.bios_image_size)
-		/* this was the last copy */
-		ret_count = bytes_left;
-	else
-		ret_count = count;
-      read_rbu_data_exit:
-	return ret_count;
+	return memory_read_from_buffer(buffer, count, &pos,
+			rbu_data.image_update_buffer, rbu_data.bios_image_size);
 }
 
 static ssize_t read_rbu_data(struct kobject *kobj,

From cd9a6f1078ed07fe919667b73e829f3bac485573 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 01:48:28 -0700
Subject: [PATCH 0891/1171] unexport proc_clear_tty

With the removal of the Solaris binary emulation the export of
proc_clear_tty became unused.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Acked-by: David S. Miller <davem@davemloft.net>
Acked-by: Alan Cox <alan@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/tty_io.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index 6f4d856df987..e1b46bc7e43c 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -3580,7 +3580,6 @@ void proc_clear_tty(struct task_struct *p)
 	p->signal->tty = NULL;
 	spin_unlock_irq(&p->sighand->siglock);
 }
-EXPORT_SYMBOL(proc_clear_tty);
 
 /* Called under the sighand lock */
 

From 6e644c3126149b65460610fe5a00d8a162092abe Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 01:48:28 -0700
Subject: [PATCH 0892/1171] move proc_kmsg_operations to fs/proc/internal.h

This patch moves the extern of struct proc_kmsg_operations to
fs/proc/internal.h and adds an #include "internal.h" to fs/proc/kmsg.c
so that the latter sees the former.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/proc/internal.h      | 1 +
 fs/proc/kmsg.c          | 2 ++
 include/linux/proc_fs.h | 1 -
 3 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/fs/proc/internal.h b/fs/proc/internal.h
index 28cbca805905..8d67616e7bb0 100644
--- a/fs/proc/internal.h
+++ b/fs/proc/internal.h
@@ -63,6 +63,7 @@ extern const struct file_operations proc_smaps_operations;
 extern const struct file_operations proc_clear_refs_operations;
 extern const struct file_operations proc_pagemap_operations;
 extern const struct file_operations proc_net_operations;
+extern const struct file_operations proc_kmsg_operations;
 extern const struct inode_operations proc_net_inode_operations;
 
 void free_proc_entry(struct proc_dir_entry *de);
diff --git a/fs/proc/kmsg.c b/fs/proc/kmsg.c
index ff3b90b56e9d..9fd5df3f40ce 100644
--- a/fs/proc/kmsg.c
+++ b/fs/proc/kmsg.c
@@ -15,6 +15,8 @@
 #include <asm/uaccess.h>
 #include <asm/io.h>
 
+#include "internal.h"
+
 extern wait_queue_head_t log_wait;
 
 extern int do_syslog(int type, char __user *bug, int count);
diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h
index 15a9eaf4a802..cdabc2fc02f7 100644
--- a/include/linux/proc_fs.h
+++ b/include/linux/proc_fs.h
@@ -138,7 +138,6 @@ extern int proc_readdir(struct file *, void *, filldir_t);
 extern struct dentry *proc_lookup(struct inode *, struct dentry *, struct nameidata *);
 
 extern const struct file_operations proc_kcore_operations;
-extern const struct file_operations proc_kmsg_operations;
 extern const struct file_operations ppc_htab_operations;
 
 extern int pid_ns_prepare_proc(struct pid_namespace *ns);

From 881adb85358309ea9c6f707394002719982ec607 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Fri, 25 Jul 2008 01:48:29 -0700
Subject: [PATCH 0893/1171] proc: always do ->release

Current two-stage scheme of removing PDE emphasizes one bug in proc:

		open
				rmmod
				remove_proc_entry
		close

->release won't be called because ->proc_fops were cleared.  In simple
cases it's small memory leak.

For every ->open, ->release has to be done.  List of openers is introduced
which is traversed at remove_proc_entry() if neeeded.

Discussions with Al long ago (sigh).

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/proc/generic.c       | 14 ++++++++
 fs/proc/inode.c         | 74 ++++++++++++++++++++++++++++++++++++++---
 fs/proc/internal.h      |  7 ++++
 include/linux/proc_fs.h |  1 +
 4 files changed, 92 insertions(+), 4 deletions(-)

diff --git a/fs/proc/generic.c b/fs/proc/generic.c
index 43e54e86cefd..bc0a0dd2d844 100644
--- a/fs/proc/generic.c
+++ b/fs/proc/generic.c
@@ -597,6 +597,7 @@ static struct proc_dir_entry *__proc_create(struct proc_dir_entry **parent,
 	ent->pde_users = 0;
 	spin_lock_init(&ent->pde_unload_lock);
 	ent->pde_unload_completion = NULL;
+	INIT_LIST_HEAD(&ent->pde_openers);
  out:
 	return ent;
 }
@@ -789,6 +790,19 @@ void remove_proc_entry(const char *name, struct proc_dir_entry *parent)
 	spin_unlock(&de->pde_unload_lock);
 
 continue_removing:
+	spin_lock(&de->pde_unload_lock);
+	while (!list_empty(&de->pde_openers)) {
+		struct pde_opener *pdeo;
+
+		pdeo = list_first_entry(&de->pde_openers, struct pde_opener, lh);
+		list_del(&pdeo->lh);
+		spin_unlock(&de->pde_unload_lock);
+		pdeo->release(pdeo->inode, pdeo->file);
+		kfree(pdeo);
+		spin_lock(&de->pde_unload_lock);
+	}
+	spin_unlock(&de->pde_unload_lock);
+
 	if (S_ISDIR(de->mode))
 		parent->nlink--;
 	de->nlink = 0;
diff --git a/fs/proc/inode.c b/fs/proc/inode.c
index b08d10017911..354c08485825 100644
--- a/fs/proc/inode.c
+++ b/fs/proc/inode.c
@@ -126,12 +126,17 @@ static const struct super_operations proc_sops = {
 	.remount_fs	= proc_remount,
 };
 
-static void pde_users_dec(struct proc_dir_entry *pde)
+static void __pde_users_dec(struct proc_dir_entry *pde)
 {
-	spin_lock(&pde->pde_unload_lock);
 	pde->pde_users--;
 	if (pde->pde_unload_completion && pde->pde_users == 0)
 		complete(pde->pde_unload_completion);
+}
+
+static void pde_users_dec(struct proc_dir_entry *pde)
+{
+	spin_lock(&pde->pde_unload_lock);
+	__pde_users_dec(pde);
 	spin_unlock(&pde->pde_unload_lock);
 }
 
@@ -318,36 +323,97 @@ static int proc_reg_open(struct inode *inode, struct file *file)
 	struct proc_dir_entry *pde = PDE(inode);
 	int rv = 0;
 	int (*open)(struct inode *, struct file *);
+	int (*release)(struct inode *, struct file *);
+	struct pde_opener *pdeo;
+
+	/*
+	 * What for, you ask? Well, we can have open, rmmod, remove_proc_entry
+	 * sequence. ->release won't be called because ->proc_fops will be
+	 * cleared. Depending on complexity of ->release, consequences vary.
+	 *
+	 * We can't wait for mercy when close will be done for real, it's
+	 * deadlockable: rmmod foo </proc/foo . So, we're going to do ->release
+	 * by hand in remove_proc_entry(). For this, save opener's credentials
+	 * for later.
+	 */
+	pdeo = kmalloc(sizeof(struct pde_opener), GFP_KERNEL);
+	if (!pdeo)
+		return -ENOMEM;
 
 	spin_lock(&pde->pde_unload_lock);
 	if (!pde->proc_fops) {
 		spin_unlock(&pde->pde_unload_lock);
+		kfree(pdeo);
 		return rv;
 	}
 	pde->pde_users++;
 	open = pde->proc_fops->open;
+	release = pde->proc_fops->release;
 	spin_unlock(&pde->pde_unload_lock);
 
 	if (open)
 		rv = open(inode, file);
 
-	pde_users_dec(pde);
+	spin_lock(&pde->pde_unload_lock);
+	if (rv == 0 && release) {
+		/* To know what to release. */
+		pdeo->inode = inode;
+		pdeo->file = file;
+		/* Strictly for "too late" ->release in proc_reg_release(). */
+		pdeo->release = release;
+		list_add(&pdeo->lh, &pde->pde_openers);
+	} else
+		kfree(pdeo);
+	__pde_users_dec(pde);
+	spin_unlock(&pde->pde_unload_lock);
 	return rv;
 }
 
+static struct pde_opener *find_pde_opener(struct proc_dir_entry *pde,
+					struct inode *inode, struct file *file)
+{
+	struct pde_opener *pdeo;
+
+	list_for_each_entry(pdeo, &pde->pde_openers, lh) {
+		if (pdeo->inode == inode && pdeo->file == file)
+			return pdeo;
+	}
+	return NULL;
+}
+
 static int proc_reg_release(struct inode *inode, struct file *file)
 {
 	struct proc_dir_entry *pde = PDE(inode);
 	int rv = 0;
 	int (*release)(struct inode *, struct file *);
+	struct pde_opener *pdeo;
 
 	spin_lock(&pde->pde_unload_lock);
+	pdeo = find_pde_opener(pde, inode, file);
 	if (!pde->proc_fops) {
-		spin_unlock(&pde->pde_unload_lock);
+		/*
+		 * Can't simply exit, __fput() will think that everything is OK,
+		 * and move on to freeing struct file. remove_proc_entry() will
+		 * find slacker in opener's list and will try to do non-trivial
+		 * things with struct file. Therefore, remove opener from list.
+		 *
+		 * But if opener is removed from list, who will ->release it?
+		 */
+		if (pdeo) {
+			list_del(&pdeo->lh);
+			spin_unlock(&pde->pde_unload_lock);
+			rv = pdeo->release(inode, file);
+			kfree(pdeo);
+		} else
+			spin_unlock(&pde->pde_unload_lock);
 		return rv;
 	}
 	pde->pde_users++;
 	release = pde->proc_fops->release;
+	if (pdeo) {
+		list_del(&pdeo->lh);
+		kfree(pdeo);
+	}
 	spin_unlock(&pde->pde_unload_lock);
 
 	if (release)
diff --git a/fs/proc/internal.h b/fs/proc/internal.h
index 8d67616e7bb0..442202314d53 100644
--- a/fs/proc/internal.h
+++ b/fs/proc/internal.h
@@ -89,3 +89,10 @@ struct dentry *proc_lookup_de(struct proc_dir_entry *de, struct inode *ino,
 		struct dentry *dentry);
 int proc_readdir_de(struct proc_dir_entry *de, struct file *filp, void *dirent,
 		filldir_t filldir);
+
+struct pde_opener {
+	struct inode *inode;
+	struct file *file;
+	int (*release)(struct inode *, struct file *);
+	struct list_head lh;
+};
diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h
index cdabc2fc02f7..f560d1705afe 100644
--- a/include/linux/proc_fs.h
+++ b/include/linux/proc_fs.h
@@ -79,6 +79,7 @@ struct proc_dir_entry {
 	int pde_users;	/* number of callers into module in progress */
 	spinlock_t pde_unload_lock; /* proc_fops checks and pde_users bumps */
 	struct completion *pde_unload_completion;
+	struct list_head pde_openers;	/* who did ->open, but not ->release */
 };
 
 struct kcore_list {

From a9bd4a3e070ba7494f154e1a11687a8a957d88dc Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Fri, 25 Jul 2008 01:48:30 -0700
Subject: [PATCH 0894/1171] proc: remove pathetic remount code

MS_RMT_MASK will unmask changes in do_remount_sb() anyway.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/proc/inode.c | 7 -------
 1 file changed, 7 deletions(-)

diff --git a/fs/proc/inode.c b/fs/proc/inode.c
index 354c08485825..02eca2ed9dd7 100644
--- a/fs/proc/inode.c
+++ b/fs/proc/inode.c
@@ -111,19 +111,12 @@ int __init proc_init_inodecache(void)
 	return 0;
 }
 
-static int proc_remount(struct super_block *sb, int *flags, char *data)
-{
-	*flags |= MS_NODIRATIME;
-	return 0;
-}
-
 static const struct super_operations proc_sops = {
 	.alloc_inode	= proc_alloc_inode,
 	.destroy_inode	= proc_destroy_inode,
 	.drop_inode	= generic_delete_inode,
 	.delete_inode	= proc_delete_inode,
 	.statfs		= simple_statfs,
-	.remount_fs	= proc_remount,
 };
 
 static void __pde_users_dec(struct proc_dir_entry *pde)

From 6eedf8d30d2b48e86fbcee1a32fb2fa5f42219ee Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Fri, 25 Jul 2008 01:48:30 -0700
Subject: [PATCH 0895/1171] proc: move Kconfig to fs/proc/Kconfig

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/Kconfig      | 60 +------------------------------------------------
 fs/proc/Kconfig | 59 ++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 60 insertions(+), 59 deletions(-)
 create mode 100644 fs/proc/Kconfig

diff --git a/fs/Kconfig b/fs/Kconfig
index ed563b9e352a..97e3bdedb1e6 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -902,65 +902,7 @@ endif # BLOCK
 
 menu "Pseudo filesystems"
 
-config PROC_FS
-	bool "/proc file system support" if EMBEDDED
-	default y
-	help
-	  This is a virtual file system providing information about the status
-	  of the system. "Virtual" means that it doesn't take up any space on
-	  your hard disk: the files are created on the fly by the kernel when
-	  you try to access them. Also, you cannot read the files with older
-	  version of the program less: you need to use more or cat.
-
-	  It's totally cool; for example, "cat /proc/interrupts" gives
-	  information about what the different IRQs are used for at the moment
-	  (there is a small number of Interrupt ReQuest lines in your computer
-	  that are used by the attached devices to gain the CPU's attention --
-	  often a source of trouble if two devices are mistakenly configured
-	  to use the same IRQ). The program procinfo to display some
-	  information about your system gathered from the /proc file system.
-
-	  Before you can use the /proc file system, it has to be mounted,
-	  meaning it has to be given a location in the directory hierarchy.
-	  That location should be /proc. A command such as "mount -t proc proc
-	  /proc" or the equivalent line in /etc/fstab does the job.
-
-	  The /proc file system is explained in the file
-	  <file:Documentation/filesystems/proc.txt> and on the proc(5) manpage
-	  ("man 5 proc").
-
-	  This option will enlarge your kernel by about 67 KB. Several
-	  programs depend on this, so everyone should say Y here.
-
-config PROC_KCORE
-	bool "/proc/kcore support" if !ARM
-	depends on PROC_FS && MMU
-
-config PROC_VMCORE
-        bool "/proc/vmcore support (EXPERIMENTAL)"
-        depends on PROC_FS && CRASH_DUMP
-	default y
-        help
-        Exports the dump image of crashed kernel in ELF format.
-
-config PROC_SYSCTL
-	bool "Sysctl support (/proc/sys)" if EMBEDDED
-	depends on PROC_FS
-	select SYSCTL
-	default y
-	---help---
-	  The sysctl interface provides a means of dynamically changing
-	  certain kernel parameters and variables on the fly without requiring
-	  a recompile of the kernel or reboot of the system.  The primary
-	  interface is through /proc/sys.  If you say Y here a tree of
-	  modifiable sysctl entries will be generated beneath the
-          /proc/sys directory. They are explained in the files
-	  in <file:Documentation/sysctl/>.  Note that enabling this
-	  option will enlarge the kernel by at least 8 KB.
-
-	  As it is generally a good thing, you should say Y here unless
-	  building a kernel for install/rescue disks or your system is very
-	  limited in memory.
+source "fs/proc/Kconfig"
 
 config SYSFS
 	bool "sysfs file system support" if EMBEDDED
diff --git a/fs/proc/Kconfig b/fs/proc/Kconfig
new file mode 100644
index 000000000000..73cd7a418f06
--- /dev/null
+++ b/fs/proc/Kconfig
@@ -0,0 +1,59 @@
+config PROC_FS
+	bool "/proc file system support" if EMBEDDED
+	default y
+	help
+	  This is a virtual file system providing information about the status
+	  of the system. "Virtual" means that it doesn't take up any space on
+	  your hard disk: the files are created on the fly by the kernel when
+	  you try to access them. Also, you cannot read the files with older
+	  version of the program less: you need to use more or cat.
+
+	  It's totally cool; for example, "cat /proc/interrupts" gives
+	  information about what the different IRQs are used for at the moment
+	  (there is a small number of Interrupt ReQuest lines in your computer
+	  that are used by the attached devices to gain the CPU's attention --
+	  often a source of trouble if two devices are mistakenly configured
+	  to use the same IRQ). The program procinfo to display some
+	  information about your system gathered from the /proc file system.
+
+	  Before you can use the /proc file system, it has to be mounted,
+	  meaning it has to be given a location in the directory hierarchy.
+	  That location should be /proc. A command such as "mount -t proc proc
+	  /proc" or the equivalent line in /etc/fstab does the job.
+
+	  The /proc file system is explained in the file
+	  <file:Documentation/filesystems/proc.txt> and on the proc(5) manpage
+	  ("man 5 proc").
+
+	  This option will enlarge your kernel by about 67 KB. Several
+	  programs depend on this, so everyone should say Y here.
+
+config PROC_KCORE
+	bool "/proc/kcore support" if !ARM
+	depends on PROC_FS && MMU
+
+config PROC_VMCORE
+        bool "/proc/vmcore support (EXPERIMENTAL)"
+        depends on PROC_FS && CRASH_DUMP
+	default y
+        help
+        Exports the dump image of crashed kernel in ELF format.
+
+config PROC_SYSCTL
+	bool "Sysctl support (/proc/sys)" if EMBEDDED
+	depends on PROC_FS
+	select SYSCTL
+	default y
+	---help---
+	  The sysctl interface provides a means of dynamically changing
+	  certain kernel parameters and variables on the fly without requiring
+	  a recompile of the kernel or reboot of the system.  The primary
+	  interface is through /proc/sys.  If you say Y here a tree of
+	  modifiable sysctl entries will be generated beneath the
+          /proc/sys directory. They are explained in the files
+	  in <file:Documentation/sysctl/>.  Note that enabling this
+	  option will enlarge the kernel by at least 8 KB.
+
+	  As it is generally a good thing, you should say Y here unless
+	  building a kernel for install/rescue disks or your system is very
+	  limited in memory.

From 339caf2a224fc9af0f01686bf287dda32c6efca6 Mon Sep 17 00:00:00 2001
From: David Sterba <dsterba@suse.cz>
Date: Fri, 25 Jul 2008 01:48:31 -0700
Subject: [PATCH 0896/1171] proc: misplaced export of find_get_pid

Move EXPORT_SYMBOL right after the func

Signed-off-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/pid.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/pid.c b/kernel/pid.c
index 30bd5d4b2ac7..753fd90d9ec1 100644
--- a/kernel/pid.c
+++ b/kernel/pid.c
@@ -435,6 +435,7 @@ struct pid *find_get_pid(pid_t nr)
 
 	return pid;
 }
+EXPORT_SYMBOL_GPL(find_get_pid);
 
 pid_t pid_nr_ns(struct pid *pid, struct pid_namespace *ns)
 {
@@ -497,7 +498,6 @@ struct pid *find_ge_pid(int nr, struct pid_namespace *ns)
 
 	return pid;
 }
-EXPORT_SYMBOL_GPL(find_get_pid);
 
 /*
  * The pid hash table is scaled according to the amount of memory in the

From 99541c23cd32bacf1a591ca537a7c0cb9053ad7e Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@parallels.com>
Date: Fri, 25 Jul 2008 01:48:31 -0700
Subject: [PATCH 0897/1171] sysctl: check for bogus modes

Catch, e. g., 644/0644 typo.

Signed-off-by: Alexey Dobriyan <adobriyan@parallels.com>
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/sysctl_check.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/kernel/sysctl_check.c b/kernel/sysctl_check.c
index c09350d564f2..c35da23ab8fb 100644
--- a/kernel/sysctl_check.c
+++ b/kernel/sysctl_check.c
@@ -1532,6 +1532,8 @@ int sysctl_check_table(struct nsproxy *namespaces, struct ctl_table *table)
 			sysctl_check_leaf(namespaces, table, &fail);
 		}
 		sysctl_check_bin_path(table, &fail);
+		if (table->mode > 0777)
+			set_fail(&fail, table, "bogus .mode");
 		if (fail) {
 			set_fail(&fail, table, NULL);
 			error = -EINVAL;

From 4ecb90090c84210a8bd2a9d7a5906e616735873c Mon Sep 17 00:00:00 2001
From: Stephen Hemminger <shemminger@vyatta.com>
Date: Fri, 25 Jul 2008 01:48:32 -0700
Subject: [PATCH 0898/1171] sysctl: allow override of /proc/sys/net with
 CAP_NET_ADMIN

Extend the permission check for networking sysctl's to allow modification
when current process has CAP_NET_ADMIN capability and is not root.  This
version uses the until now unused permissions hook to override the mode
value for /proc/sys/net if accessed by a user with capabilities.

Found while working with Quagga.  It is impossible to turn forwarding
on/off through the command interface because Quagga uses secure coding
practice of dropping privledges during initialization and only raising via
capabilities when necessary.  Since the dameon has reset real/effective
uid after initialization, all attempts to access /proc/sys/net variables
will fail.

Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Chris Wright <chrisw@sous-sol.org>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Andrew Morgan <morgan@kernel.org>
Cc: Pavel Emelyanov <xemul@openvz.org>
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 net/sysctl_net.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/net/sysctl_net.c b/net/sysctl_net.c
index 007c1a6708ee..63ada437fc2f 100644
--- a/net/sysctl_net.c
+++ b/net/sysctl_net.c
@@ -35,8 +35,22 @@ net_ctl_header_lookup(struct ctl_table_root *root, struct nsproxy *namespaces)
 	return &namespaces->net_ns->sysctl_table_headers;
 }
 
+/* Return standard mode bits for table entry. */
+static int net_ctl_permissions(struct ctl_table_root *root,
+			       struct nsproxy *nsproxy,
+			       struct ctl_table *table)
+{
+	/* Allow network administrator to have same access as root. */
+	if (capable(CAP_NET_ADMIN)) {
+		int mode = (table->mode >> 6) & 7;
+		return (mode << 6) | (mode << 3) | mode;
+	}
+	return table->mode;
+}
+
 static struct ctl_table_root net_sysctl_root = {
 	.lookup = net_ctl_header_lookup,
+	.permissions = net_ctl_permissions,
 };
 
 static LIST_HEAD(net_sysctl_ro_tables);

From 3ae4eed34be0177a8e003411a84e4ee212adbced Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 01:48:34 -0700
Subject: [PATCH 0899/1171] proper pid{hash,map}_init() prototypes

This patch adds proper prototypes for pid{hash,map}_init() in
include/linux/pid_namespace.h

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/pid_namespace.h | 3 +++
 init/main.c                   | 2 --
 2 files changed, 3 insertions(+), 2 deletions(-)

diff --git a/include/linux/pid_namespace.h b/include/linux/pid_namespace.h
index caff5283d15c..1a49ab5ec7b9 100644
--- a/include/linux/pid_namespace.h
+++ b/include/linux/pid_namespace.h
@@ -85,4 +85,7 @@ static inline struct task_struct *task_child_reaper(struct task_struct *tsk)
 	return tsk->nsproxy->pid_ns->child_reaper;
 }
 
+void pidhash_init(void);
+void pidmap_init(void);
+
 #endif /* _LINUX_PID_NS_H */
diff --git a/init/main.c b/init/main.c
index 2769dc031c62..0604cbcaf1e4 100644
--- a/init/main.c
+++ b/init/main.c
@@ -87,8 +87,6 @@ extern void init_IRQ(void);
 extern void fork_init(unsigned long);
 extern void mca_init(void);
 extern void sbus_init(void);
-extern void pidhash_init(void);
-extern void pidmap_init(void);
 extern void prio_tree_init(void);
 extern void radix_tree_init(void);
 extern void free_initmem(void);

From 33166b1ffca5e1945246bcaa77d72a22b0d3e531 Mon Sep 17 00:00:00 2001
From: Richard Kennedy <richard@rsk.demon.co.uk>
Date: Fri, 25 Jul 2008 01:48:35 -0700
Subject: [PATCH 0900/1171] shrink struct pid by removing padding on 64 bit
 builds

When struct pid is built on a 64 bit platform gcc has to insert padding to
maintain the correct alignment, by simply reordering its members the
memory usage shrinks from 88 bytes to 80.

I've successfully run with this patch on my desktop AMD64 machine.

There are no significant kernel size changes to a default config.X86_64
on the latest git v2.6.26-rc1

   text    data     bss     dec     hex filename
5404828  976760  734280 7115868  6c945c vmlinux
5404811  976760  734280 7115851  6c944b vmlinux.pid-patch

Acked-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/pid.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/linux/pid.h b/include/linux/pid.h
index c21c7e8124a7..6f084b9e2c40 100644
--- a/include/linux/pid.h
+++ b/include/linux/pid.h
@@ -57,10 +57,10 @@ struct upid {
 struct pid
 {
 	atomic_t count;
+	unsigned int level;
 	/* lists of tasks that use this pid */
 	struct hlist_head tasks[PIDTYPE_MAX];
 	struct rcu_head rcu;
-	unsigned int level;
 	struct upid numbers[1];
 };
 

From 19b0cfcca41dd772065671ad0584e1cea0f3fd13 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 25 Jul 2008 01:48:35 -0700
Subject: [PATCH 0901/1171] pidns: remove now unused kill_proc function

This function operated on a pid_t to kill a task, which is no longer valid
in a containerized system.

It has finally lost all its users and we can safely remove it from the
tree.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/sched.h |  1 -
 kernel/signal.c       | 12 ------------
 2 files changed, 13 deletions(-)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 0560999eb1db..134cb5cb506c 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1800,7 +1800,6 @@ extern void force_sig(int, struct task_struct *);
 extern void force_sig_specific(int, struct task_struct *);
 extern int send_sig(int, struct task_struct *, int);
 extern void zap_other_threads(struct task_struct *p);
-extern int kill_proc(pid_t, int, int);
 extern struct sigqueue *sigqueue_alloc(void);
 extern void sigqueue_free(struct sigqueue *);
 extern int send_sigqueue(struct sigqueue *,  struct task_struct *, int group);
diff --git a/kernel/signal.c b/kernel/signal.c
index 5c7b7eaa0dc6..82c3545596c5 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -1228,17 +1228,6 @@ int kill_pid(struct pid *pid, int sig, int priv)
 }
 EXPORT_SYMBOL(kill_pid);
 
-int
-kill_proc(pid_t pid, int sig, int priv)
-{
-	int ret;
-
-	rcu_read_lock();
-	ret = kill_pid_info(sig, __si_special(priv), find_pid(pid));
-	rcu_read_unlock();
-	return ret;
-}
-
 /*
  * These functions support sending signals using preallocated sigqueue
  * structures.  This is needed "because realtime applications cannot
@@ -1906,7 +1895,6 @@ EXPORT_SYMBOL(recalc_sigpending);
 EXPORT_SYMBOL_GPL(dequeue_signal);
 EXPORT_SYMBOL(flush_signals);
 EXPORT_SYMBOL(force_sig);
-EXPORT_SYMBOL(kill_proc);
 EXPORT_SYMBOL(ptrace_notify);
 EXPORT_SYMBOL(send_sig);
 EXPORT_SYMBOL(send_sig_info);

From e49859e71e0318b564de1546bdc30fab738f9deb Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 25 Jul 2008 01:48:36 -0700
Subject: [PATCH 0902/1171] pidns: remove now unused find_pid function.

This one had the only users so far - the kill_proc, which is removed, so
drop this (invalid in namespaced world) call too.

And of course - erase all references on it from comments.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/pid.h   | 4 +---
 include/linux/sched.h | 2 +-
 kernel/pid.c          | 8 +-------
 3 files changed, 3 insertions(+), 11 deletions(-)

diff --git a/include/linux/pid.h b/include/linux/pid.h
index 6f084b9e2c40..ff1b2a5814d4 100644
--- a/include/linux/pid.h
+++ b/include/linux/pid.h
@@ -48,7 +48,7 @@ enum pid_type
  */
 
 struct upid {
-	/* Try to keep pid_chain in the same cacheline as nr for find_pid */
+	/* Try to keep pid_chain in the same cacheline as nr for find_vpid */
 	int nr;
 	struct pid_namespace *ns;
 	struct hlist_node pid_chain;
@@ -105,14 +105,12 @@ extern struct pid_namespace init_pid_ns;
  * or rcu_read_lock() held.
  *
  * find_pid_ns() finds the pid in the namespace specified
- * find_pid() find the pid by its global id, i.e. in the init namespace
  * find_vpid() finr the pid by its virtual id, i.e. in the current namespace
  *
  * see also find_task_by_pid() set in include/linux/sched.h
  */
 extern struct pid *find_pid_ns(int nr, struct pid_namespace *ns);
 extern struct pid *find_vpid(int nr);
-extern struct pid *find_pid(int nr);
 
 /*
  * Lookup a PID in the hash table, and return with it's count elevated.
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 134cb5cb506c..182da1550fad 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1718,7 +1718,7 @@ extern struct pid_namespace init_pid_ns;
  * find_task_by_pid():
  *      finds a task by its global pid
  *
- * see also find_pid() etc in include/linux/pid.h
+ * see also find_vpid() etc in include/linux/pid.h
  */
 
 extern struct task_struct *find_task_by_pid_type_ns(int type, int pid,
diff --git a/kernel/pid.c b/kernel/pid.c
index 753fd90d9ec1..064e76afa507 100644
--- a/kernel/pid.c
+++ b/kernel/pid.c
@@ -309,12 +309,6 @@ struct pid *find_vpid(int nr)
 }
 EXPORT_SYMBOL_GPL(find_vpid);
 
-struct pid *find_pid(int nr)
-{
-	return find_pid_ns(nr, &init_pid_ns);
-}
-EXPORT_SYMBOL_GPL(find_pid);
-
 /*
  * attach_pid() must be called with the tasklist_lock write-held.
  */
@@ -483,7 +477,7 @@ EXPORT_SYMBOL(task_session_nr_ns);
 /*
  * Used by proc to find the first pid that is greater then or equal to nr.
  *
- * If there is a pid at nr this function is exactly the same as find_pid.
+ * If there is a pid at nr this function is exactly the same as find_pid_ns.
  */
 struct pid *find_ge_pid(int nr, struct pid_namespace *ns)
 {

From dbda0de52618d13d1b927c7ba7bb839cfddc4e8c Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 25 Jul 2008 01:48:37 -0700
Subject: [PATCH 0903/1171] pidns: remove find_task_by_pid, unused for a long
 time

It seems to me that it was a mistake marking this function as deprecated
and scheduling it for removal, rather than resolutely removing it after
the last caller's death.

Anyway - better late, then never.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/feature-removal-schedule.txt | 18 ------------------
 include/linux/pid.h                        |  2 +-
 include/linux/sched.h                      |  6 ------
 3 files changed, 1 insertion(+), 25 deletions(-)

diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt
index 09c4a1efb8e3..721c71b86e06 100644
--- a/Documentation/feature-removal-schedule.txt
+++ b/Documentation/feature-removal-schedule.txt
@@ -138,24 +138,6 @@ Who:	Kay Sievers <kay.sievers@suse.de>
 
 ---------------------------
 
-What:	find_task_by_pid
-When:	2.6.26
-Why:	With pid namespaces, calling this funciton will return the
-	wrong task when called from inside a namespace.
-
-	The best way to save a task pid and find a task by this
-	pid later, is to find this task's struct pid pointer (or get
-	it directly from the task) and call pid_task() later.
-
-	If someone really needs to get a task by its pid_t, then
-	he most likely needs the find_task_by_vpid() to get the
-	task from the same namespace as the current task is in, but
-	this may be not so in general.
-
-Who:	Pavel Emelyanov <xemul@openvz.org>
-
----------------------------
-
 What:	ACPI procfs interface
 When:	July 2008
 Why:	ACPI sysfs conversion should be finished by January 2008.
diff --git a/include/linux/pid.h b/include/linux/pid.h
index ff1b2a5814d4..22921ac4cfd9 100644
--- a/include/linux/pid.h
+++ b/include/linux/pid.h
@@ -107,7 +107,7 @@ extern struct pid_namespace init_pid_ns;
  * find_pid_ns() finds the pid in the namespace specified
  * find_vpid() finr the pid by its virtual id, i.e. in the current namespace
  *
- * see also find_task_by_pid() set in include/linux/sched.h
+ * see also find_task_by_vpid() set in include/linux/sched.h
  */
 extern struct pid *find_pid_ns(int nr, struct pid_namespace *ns);
 extern struct pid *find_vpid(int nr);
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 182da1550fad..354ef478a80d 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1715,8 +1715,6 @@ extern struct pid_namespace init_pid_ns;
  *      finds a task by its pid in the specified namespace
  * find_task_by_vpid():
  *      finds a task by its virtual pid
- * find_task_by_pid():
- *      finds a task by its global pid
  *
  * see also find_vpid() etc in include/linux/pid.h
  */
@@ -1724,10 +1722,6 @@ extern struct pid_namespace init_pid_ns;
 extern struct task_struct *find_task_by_pid_type_ns(int type, int pid,
 		struct pid_namespace *ns);
 
-static inline struct task_struct *__deprecated find_task_by_pid(pid_t nr)
-{
-	return find_task_by_pid_type_ns(PIDTYPE_PID, nr, &init_pid_ns);
-}
 extern struct task_struct *find_task_by_vpid(pid_t nr);
 extern struct task_struct *find_task_by_pid_ns(pid_t nr,
 		struct pid_namespace *ns);

From 24879a8e3e68f146d4d85528cc0b5dea712b77c5 Mon Sep 17 00:00:00 2001
From: Matthias Kaehlcke <matthias@kaehlcke.net>
Date: Fri, 25 Jul 2008 01:48:38 -0700
Subject: [PATCH 0904/1171] aoe: convert emsgs_sema into a completion

ATA over Ethernet: The semaphore emsgs_sema is used for signalling an
event, convert it in a completion.

Signed-off-by: Matthias Kaehlcke <matthias@kaehlcke.net>
Cc: "Ed L. Cashin" <ecashin@coraid.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/block/aoe/aoechr.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/block/aoe/aoechr.c b/drivers/block/aoe/aoechr.c
index c04440cd6a32..181ebb85f0be 100644
--- a/drivers/block/aoe/aoechr.c
+++ b/drivers/block/aoe/aoechr.c
@@ -6,6 +6,7 @@
 
 #include <linux/hdreg.h>
 #include <linux/blkdev.h>
+#include <linux/completion.h>
 #include <linux/delay.h>
 #include <linux/smp_lock.h>
 #include "aoe.h"
@@ -36,7 +37,7 @@ struct ErrMsg {
 
 static struct ErrMsg emsgs[NMSG];
 static int emsgs_head_idx, emsgs_tail_idx;
-static struct semaphore emsgs_sema;
+static struct completion emsgs_comp;
 static spinlock_t emsgs_lock;
 static int nblocked_emsgs_readers;
 static struct class *aoe_class;
@@ -141,7 +142,7 @@ bail:		spin_unlock_irqrestore(&emsgs_lock, flags);
 	spin_unlock_irqrestore(&emsgs_lock, flags);
 
 	if (nblocked_emsgs_readers)
-		up(&emsgs_sema);
+		complete(&emsgs_comp);
 }
 
 static ssize_t
@@ -221,7 +222,7 @@ aoechr_read(struct file *filp, char __user *buf, size_t cnt, loff_t *off)
 
 		spin_unlock_irqrestore(&emsgs_lock, flags);
 
-		n = down_interruptible(&emsgs_sema);
+		n = wait_for_completion_interruptible(&emsgs_comp);
 
 		spin_lock_irqsave(&emsgs_lock, flags);
 
@@ -269,7 +270,7 @@ aoechr_init(void)
 		printk(KERN_ERR "aoe: can't register char device\n");
 		return n;
 	}
-	sema_init(&emsgs_sema, 0);
+	init_completion(&emsgs_comp);
 	spin_lock_init(&emsgs_lock);
 	aoe_class = class_create(THIS_MODULE, "aoe");
 	if (IS_ERR(aoe_class)) {

From 28325df0d9339b7f3aba9c45174d4586223ef46b Mon Sep 17 00:00:00 2001
From: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Date: Fri, 25 Jul 2008 01:48:38 -0700
Subject: [PATCH 0905/1171] markers: use rcu_barrier_sched() and
 call_rcu_sched()

rcu_barrier_sched() and call_rcu_sched() were introduced in 2.6.26 for the
Markers.  Change the marker code to use them.

It can be seen as a fix since the marker code was using an ugly,
temporary, #ifdef hack to work around CONFIG_PREEMPT_RCU.

Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Acked-by: Paul McKenney <paulmck@us.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/marker.c | 25 ++++++++-----------------
 1 file changed, 8 insertions(+), 17 deletions(-)

diff --git a/kernel/marker.c b/kernel/marker.c
index 1abfb923b761..971da5317903 100644
--- a/kernel/marker.c
+++ b/kernel/marker.c
@@ -441,7 +441,7 @@ static int remove_marker(const char *name)
 	hlist_del(&e->hlist);
 	/* Make sure the call_rcu has been executed */
 	if (e->rcu_pending)
-		rcu_barrier();
+		rcu_barrier_sched();
 	kfree(e);
 	return 0;
 }
@@ -476,7 +476,7 @@ static int marker_set_format(struct marker_entry **entry, const char *format)
 	hlist_del(&(*entry)->hlist);
 	/* Make sure the call_rcu has been executed */
 	if ((*entry)->rcu_pending)
-		rcu_barrier();
+		rcu_barrier_sched();
 	kfree(*entry);
 	*entry = e;
 	trace_mark(core_marker_format, "name %s format %s",
@@ -655,7 +655,7 @@ int marker_probe_register(const char *name, const char *format,
 	 * make sure it's executed now.
 	 */
 	if (entry->rcu_pending)
-		rcu_barrier();
+		rcu_barrier_sched();
 	old = marker_entry_add_probe(entry, probe, probe_private);
 	if (IS_ERR(old)) {
 		ret = PTR_ERR(old);
@@ -670,10 +670,7 @@ int marker_probe_register(const char *name, const char *format,
 	entry->rcu_pending = 1;
 	/* write rcu_pending before calling the RCU callback */
 	smp_wmb();
-#ifdef CONFIG_PREEMPT_RCU
-	synchronize_sched();	/* Until we have the call_rcu_sched() */
-#endif
-	call_rcu(&entry->rcu, free_old_closure);
+	call_rcu_sched(&entry->rcu, free_old_closure);
 end:
 	mutex_unlock(&markers_mutex);
 	return ret;
@@ -704,7 +701,7 @@ int marker_probe_unregister(const char *name,
 	if (!entry)
 		goto end;
 	if (entry->rcu_pending)
-		rcu_barrier();
+		rcu_barrier_sched();
 	old = marker_entry_remove_probe(entry, probe, probe_private);
 	mutex_unlock(&markers_mutex);
 	marker_update_probes();		/* may update entry */
@@ -716,10 +713,7 @@ int marker_probe_unregister(const char *name,
 	entry->rcu_pending = 1;
 	/* write rcu_pending before calling the RCU callback */
 	smp_wmb();
-#ifdef CONFIG_PREEMPT_RCU
-	synchronize_sched();	/* Until we have the call_rcu_sched() */
-#endif
-	call_rcu(&entry->rcu, free_old_closure);
+	call_rcu_sched(&entry->rcu, free_old_closure);
 	remove_marker(name);	/* Ignore busy error message */
 	ret = 0;
 end:
@@ -786,7 +780,7 @@ int marker_probe_unregister_private_data(marker_probe_func *probe,
 		goto end;
 	}
 	if (entry->rcu_pending)
-		rcu_barrier();
+		rcu_barrier_sched();
 	old = marker_entry_remove_probe(entry, NULL, probe_private);
 	mutex_unlock(&markers_mutex);
 	marker_update_probes();		/* may update entry */
@@ -797,10 +791,7 @@ int marker_probe_unregister_private_data(marker_probe_func *probe,
 	entry->rcu_pending = 1;
 	/* write rcu_pending before calling the RCU callback */
 	smp_wmb();
-#ifdef CONFIG_PREEMPT_RCU
-	synchronize_sched();	/* Until we have the call_rcu_sched() */
-#endif
-	call_rcu(&entry->rcu, free_old_closure);
+	call_rcu_sched(&entry->rcu, free_old_closure);
 	remove_marker(entry->name);	/* Ignore busy error message */
 end:
 	mutex_unlock(&markers_mutex);

From a89cc1959d0ea5f36bf7421dc97b34f03809637d Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Fri, 25 Jul 2008 01:48:39 -0700
Subject: [PATCH 0906/1171] markers: fix sparse integer as NULL pointer warning

kernel/trace/trace_sysprof.c:164:20: warning: Using plain integer as NULL pointer

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/trace/trace_sysprof.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/trace/trace_sysprof.c b/kernel/trace/trace_sysprof.c
index 63528086337c..ce2d723c10e1 100644
--- a/kernel/trace/trace_sysprof.c
+++ b/kernel/trace/trace_sysprof.c
@@ -161,7 +161,7 @@ static void timer_notify(struct pt_regs *regs, int cpu)
 		__trace_special(tr, data, 2, regs->ip, 0);
 
 		while (i < sample_max_depth) {
-			frame.next_fp = 0;
+			frame.next_fp = NULL;
 			frame.return_address = 0;
 			if (!copy_stack_frame(fp, &frame))
 				break;

From 7394f0f6c0baab650ea9194cb1be847df646fb57 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 01:48:40 -0700
Subject: [PATCH 0907/1171] unexport uts_sem

With the removal of the Solaris binary emulation the export of
uts_sem became unused.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/sys.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/kernel/sys.c b/kernel/sys.c
index 6c2188046048..0c9d3fa1f5ff 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -1343,8 +1343,6 @@ EXPORT_SYMBOL(in_egroup_p);
 
 DECLARE_RWSEM(uts_sem);
 
-EXPORT_SYMBOL(uts_sem);
-
 asmlinkage long sys_newuname(struct new_utsname __user * name)
 {
 	int errno = 0;

From 49b5cf34727a6c1be1568ab28e89a2d9a6bf51e0 Mon Sep 17 00:00:00 2001
From: Jonathan Lim <jlim@sgi.com>
Date: Fri, 25 Jul 2008 01:48:40 -0700
Subject: [PATCH 0908/1171] accounting: account for user time when updating
 memory integrals

Adapt acct_update_integrals() to include user time when calculating the time
difference.  The units of acct_rss_mem1 and acct_vm_mem1 are also changed from
pages-jiffies to pages-usecs to avoid calling jiffies_to_usecs() in
xacct_add_tsk() which might overflow.

Signed-off-by: Jonathan Lim <jlim@sgi.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/sched.h |  2 +-
 kernel/sched.c        |  2 ++
 kernel/tsacct.c       | 21 ++++++++++++++-------
 3 files changed, 17 insertions(+), 8 deletions(-)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 354ef478a80d..af780f299c7c 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1257,7 +1257,7 @@ struct task_struct {
 #if defined(CONFIG_TASK_XACCT)
 	u64 acct_rss_mem1;	/* accumulated rss usage */
 	u64 acct_vm_mem1;	/* accumulated virtual memory usage */
-	cputime_t acct_stimexpd;/* stime since last update */
+	cputime_t acct_timexpd;	/* stime + utime since last update */
 #endif
 #ifdef CONFIG_CPUSETS
 	nodemask_t mems_allowed;
diff --git a/kernel/sched.c b/kernel/sched.c
index 6acf749d3336..0047bd9b96aa 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -4046,6 +4046,8 @@ void account_user_time(struct task_struct *p, cputime_t cputime)
 		cpustat->nice = cputime64_add(cpustat->nice, tmp);
 	else
 		cpustat->user = cputime64_add(cpustat->user, tmp);
+	/* Account for user time used */
+	acct_update_integrals(p);
 }
 
 /*
diff --git a/kernel/tsacct.c b/kernel/tsacct.c
index 4ab1b584961b..1da6990af8e0 100644
--- a/kernel/tsacct.c
+++ b/kernel/tsacct.c
@@ -84,9 +84,9 @@ void xacct_add_tsk(struct taskstats *stats, struct task_struct *p)
 {
 	struct mm_struct *mm;
 
-	/* convert pages-jiffies to Mbyte-usec */
-	stats->coremem = jiffies_to_usecs(p->acct_rss_mem1) * PAGE_SIZE / MB;
-	stats->virtmem = jiffies_to_usecs(p->acct_vm_mem1) * PAGE_SIZE / MB;
+	/* convert pages-usec to Mbyte-usec */
+	stats->coremem = p->acct_rss_mem1 * PAGE_SIZE / MB;
+	stats->virtmem = p->acct_vm_mem1 * PAGE_SIZE / MB;
 	mm = get_task_mm(p);
 	if (mm) {
 		/* adjust to KB unit */
@@ -118,12 +118,19 @@ void xacct_add_tsk(struct taskstats *stats, struct task_struct *p)
 void acct_update_integrals(struct task_struct *tsk)
 {
 	if (likely(tsk->mm)) {
-		long delta = cputime_to_jiffies(
-			cputime_sub(tsk->stime, tsk->acct_stimexpd));
+		cputime_t time, dtime;
+		struct timeval value;
+		u64 delta;
+
+		time = tsk->stime + tsk->utime;
+		dtime = cputime_sub(time, tsk->acct_timexpd);
+		jiffies_to_timeval(cputime_to_jiffies(dtime), &value);
+		delta = value.tv_sec;
+		delta = delta * USEC_PER_SEC + value.tv_usec;
 
 		if (delta == 0)
 			return;
-		tsk->acct_stimexpd = tsk->stime;
+		tsk->acct_timexpd = time;
 		tsk->acct_rss_mem1 += delta * get_mm_rss(tsk->mm);
 		tsk->acct_vm_mem1 += delta * tsk->mm->total_vm;
 	}
@@ -135,7 +142,7 @@ void acct_update_integrals(struct task_struct *tsk)
  */
 void acct_clear_integrals(struct task_struct *tsk)
 {
-	tsk->acct_stimexpd = 0;
+	tsk->acct_timexpd = 0;
 	tsk->acct_rss_mem1 = 0;
 	tsk->acct_vm_mem1 = 0;
 }

From 081e4c8a75692c21f3a119a81ca3270081879d0e Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 25 Jul 2008 01:48:42 -0700
Subject: [PATCH 0909/1171] bsdacct: rename acct_gbls to bsd_acct_struct

After I fixed access to task->tgid in kernel/acct.c, Oleg pointed out some
bad side effects with this accounting vs pid namespaces interaction.  I.e.
 when some task in pid namespace sets this accounting up, this blocks all
the others from doing the same.  Restricting this to init namespace only
could help, but didn't look a graceful solution.

So here is the approach to make this accounting work with pid namespaces
properly.

The idea is simple - when a task dies it accounts itself in each namespace
it is visible from and which set the accounting up.

For example here are the commands run and the output of lastcomm from init
and sub namespaces:

init_ns# accton pacct
 sub_ns# accton pacct (this is a different file - sub ns is run in
                       a chroot-ed environment)
init_ns# cat /dev/null
 sub_ns# ls /dev/null
init_ns# accton
 sub_ns# accton

 sub_ns#  lastcomm -f pacct
ls                      0        [136,0]    0.00 secs Thu May 15 10:30
accton                  0        [136,0]    0.00 secs Thu May 15 10:30

init_ns# lastcomm -f pacct
accton                  root     pts/0      0.00 secs Thu May 15 14:30 << got from sub
cat                     root     pts/1      0.00 secs Thu May 15 14:30
ls                      root     pts/0      0.00 secs Thu May 15 14:30 << got from sub
accton                  root     pts/1      0.00 secs Thu May 15 14:30

That was the summary, the details are in patches.

This patch:

It will be visible in pid_namespace.h file, so fix its name to look better
outside the acct.c file.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/acct.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/acct.c b/kernel/acct.c
index 91e1cfd734d2..ee3e605190f9 100644
--- a/kernel/acct.c
+++ b/kernel/acct.c
@@ -82,7 +82,7 @@ static void do_acct_process(struct pid_namespace *ns, struct file *);
  * can be placed in the same cache line as the lock.  This primes
  * the cache line to have the data after getting the lock.
  */
-struct acct_glbs {
+struct bsd_acct_struct {
 	spinlock_t		lock;
 	volatile int		active;
 	volatile int		needcheck;
@@ -91,7 +91,7 @@ struct acct_glbs {
 	struct timer_list	timer;
 };
 
-static struct acct_glbs acct_globals __cacheline_aligned =
+static struct bsd_acct_struct acct_globals __cacheline_aligned =
 	{__SPIN_LOCK_UNLOCKED(acct_globals.lock)};
 
 /*

From 84406c153a5bfa5d8b428a0933e9d39db6b59a75 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 25 Jul 2008 01:48:42 -0700
Subject: [PATCH 0910/1171] pidns: use kzalloc when allocating new
 pid_namespace struct

It makes many fields initialization implicit helping in auto-setting
#ifdef-ed fields (bsd-acct related pointer will be such).

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/pid_namespace.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/kernel/pid_namespace.c b/kernel/pid_namespace.c
index 98702b4b8851..06331cc1c3f5 100644
--- a/kernel/pid_namespace.c
+++ b/kernel/pid_namespace.c
@@ -71,7 +71,7 @@ static struct pid_namespace *create_pid_namespace(unsigned int level)
 	struct pid_namespace *ns;
 	int i;
 
-	ns = kmem_cache_alloc(pid_ns_cachep, GFP_KERNEL);
+	ns = kmem_cache_zalloc(pid_ns_cachep, GFP_KERNEL);
 	if (ns == NULL)
 		goto out;
 
@@ -84,17 +84,13 @@ static struct pid_namespace *create_pid_namespace(unsigned int level)
 		goto out_free_map;
 
 	kref_init(&ns->kref);
-	ns->last_pid = 0;
-	ns->child_reaper = NULL;
 	ns->level = level;
 
 	set_bit(0, ns->pidmap[0].page);
 	atomic_set(&ns->pidmap[0].nr_free, BITS_PER_PAGE - 1);
 
-	for (i = 1; i < PIDMAP_ENTRIES; i++) {
-		ns->pidmap[i].page = NULL;
+	for (i = 1; i < PIDMAP_ENTRIES; i++)
 		atomic_set(&ns->pidmap[i].nr_free, BITS_PER_PAGE);
-	}
 
 	return ns;
 

From 20fad13ac66ac001c19220d3d08b4de5b6cca6e1 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 25 Jul 2008 01:48:43 -0700
Subject: [PATCH 0911/1171] pidns: add the struct bsd_acct_struct pointer on
 pid_namespace struct

All the bsdacct-related info will be stored in the area, pointer by this
one.

It will be NULL automatically for all new namespaces.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/pid_namespace.h | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/include/linux/pid_namespace.h b/include/linux/pid_namespace.h
index 1a49ab5ec7b9..1af82c4e17d4 100644
--- a/include/linux/pid_namespace.h
+++ b/include/linux/pid_namespace.h
@@ -14,6 +14,8 @@ struct pidmap {
 
 #define PIDMAP_ENTRIES         ((PID_MAX_LIMIT + 8*PAGE_SIZE - 1)/PAGE_SIZE/8)
 
+struct bsd_acct_struct;
+
 struct pid_namespace {
 	struct kref kref;
 	struct pidmap pidmap[PIDMAP_ENTRIES];
@@ -25,6 +27,9 @@ struct pid_namespace {
 #ifdef CONFIG_PROC_FS
 	struct vfsmount *proc_mnt;
 #endif
+#ifdef CONFIG_BSD_PROCESS_ACCT
+	struct bsd_acct_struct *bacct;
+#endif
 };
 
 extern struct pid_namespace init_pid_ns;

From 1c552858ac2b1732a99d234d46b98098baef41ff Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 25 Jul 2008 01:48:44 -0700
Subject: [PATCH 0912/1171] bsdacct: "truthify" a comment near acct_process

The acct_process does not accept any arguments actually.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/acct.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/kernel/acct.c b/kernel/acct.c
index ee3e605190f9..d9ee1838b4d4 100644
--- a/kernel/acct.c
+++ b/kernel/acct.c
@@ -579,7 +579,6 @@ void acct_collect(long exitcode, int group_dead)
 
 /**
  * acct_process - now just a wrapper around do_acct_process
- * @exitcode: task exit code
  *
  * handles process accounting for an exiting task
  */

From e59a04a7aa5ce2483470aee4f2eb79ba6b9afe8b Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 25 Jul 2008 01:48:44 -0700
Subject: [PATCH 0913/1171] bsdacct: make check timer accept a bsd_acct_struct
 argument

We're going to have many bsd_acct_struct instances, not just one, so the
timer (currently working with a global one) has to know which one to work
with.

Use a handy setup_timer macro for it (thanks to Oleg for one).

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/acct.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/kernel/acct.c b/kernel/acct.c
index d9ee1838b4d4..05f8bc094a4b 100644
--- a/kernel/acct.c
+++ b/kernel/acct.c
@@ -97,9 +97,10 @@ static struct bsd_acct_struct acct_globals __cacheline_aligned =
 /*
  * Called whenever the timer says to check the free space.
  */
-static void acct_timeout(unsigned long unused)
+static void acct_timeout(unsigned long x)
 {
-	acct_globals.needcheck = 1;
+	struct bsd_acct_struct *acct = (struct bsd_acct_struct *)x;
+	acct->needcheck = 1;
 }
 
 /*
@@ -193,8 +194,8 @@ static void acct_file_reopen(struct file *file)
 		acct_globals.needcheck = 0;
 		acct_globals.active = 1;
 		/* It's been deleted if it was used before so this is safe */
-		init_timer(&acct_globals.timer);
-		acct_globals.timer.function = acct_timeout;
+		setup_timer(&acct_globals.timer, acct_timeout,
+				(unsigned long)&acct_globals);
 		acct_globals.timer.expires = jiffies + ACCT_TIMEOUT*HZ;
 		add_timer(&acct_globals.timer);
 	}

From a75d97976517dcda69150fd81d6be86ae63324a1 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 25 Jul 2008 01:48:45 -0700
Subject: [PATCH 0914/1171] bsdacct: turn the acct_lock from on-the-struct to
 global

Don't use per-bsd-acct-struct lock, but work with a global one.

This lock is taken for short periods, so it doesn't seem it'll become a
bottleneck, but it will allow us to easily avoid many locking difficulties
in the future.

So this is a mostly s/acct_globals.lock/acct_lock/ over the file.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/acct.c | 42 +++++++++++++++++++++---------------------
 1 file changed, 21 insertions(+), 21 deletions(-)

diff --git a/kernel/acct.c b/kernel/acct.c
index 05f8bc094a4b..fc71c1304977 100644
--- a/kernel/acct.c
+++ b/kernel/acct.c
@@ -83,7 +83,6 @@ static void do_acct_process(struct pid_namespace *ns, struct file *);
  * the cache line to have the data after getting the lock.
  */
 struct bsd_acct_struct {
-	spinlock_t		lock;
 	volatile int		active;
 	volatile int		needcheck;
 	struct file		*file;
@@ -91,8 +90,9 @@ struct bsd_acct_struct {
 	struct timer_list	timer;
 };
 
-static struct bsd_acct_struct acct_globals __cacheline_aligned =
-	{__SPIN_LOCK_UNLOCKED(acct_globals.lock)};
+static DEFINE_SPINLOCK(acct_lock);
+
+static struct bsd_acct_struct acct_globals __cacheline_aligned;
 
 /*
  * Called whenever the timer says to check the free space.
@@ -114,11 +114,11 @@ static int check_free_space(struct file *file)
 	sector_t resume;
 	sector_t suspend;
 
-	spin_lock(&acct_globals.lock);
+	spin_lock(&acct_lock);
 	res = acct_globals.active;
 	if (!file || !acct_globals.needcheck)
 		goto out;
-	spin_unlock(&acct_globals.lock);
+	spin_unlock(&acct_lock);
 
 	/* May block */
 	if (vfs_statfs(file->f_path.dentry, &sbuf))
@@ -140,7 +140,7 @@ static int check_free_space(struct file *file)
 	 * If some joker switched acct_globals.file under us we'ld better be
 	 * silent and _not_ touch anything.
 	 */
-	spin_lock(&acct_globals.lock);
+	spin_lock(&acct_lock);
 	if (file != acct_globals.file) {
 		if (act)
 			res = act>0;
@@ -165,7 +165,7 @@ static int check_free_space(struct file *file)
 	add_timer(&acct_globals.timer);
 	res = acct_globals.active;
 out:
-	spin_unlock(&acct_globals.lock);
+	spin_unlock(&acct_lock);
 	return res;
 }
 
@@ -173,7 +173,7 @@ out:
  * Close the old accounting file (if currently open) and then replace
  * it with file (if non-NULL).
  *
- * NOTE: acct_globals.lock MUST be held on entry and exit.
+ * NOTE: acct_lock MUST be held on entry and exit.
  */
 static void acct_file_reopen(struct file *file)
 {
@@ -201,11 +201,11 @@ static void acct_file_reopen(struct file *file)
 	}
 	if (old_acct) {
 		mnt_unpin(old_acct->f_path.mnt);
-		spin_unlock(&acct_globals.lock);
+		spin_unlock(&acct_lock);
 		do_acct_process(old_ns, old_acct);
 		filp_close(old_acct, NULL);
 		put_pid_ns(old_ns);
-		spin_lock(&acct_globals.lock);
+		spin_lock(&acct_lock);
 	}
 }
 
@@ -235,10 +235,10 @@ static int acct_on(char *name)
 		return error;
 	}
 
-	spin_lock(&acct_globals.lock);
+	spin_lock(&acct_lock);
 	mnt_pin(file->f_path.mnt);
 	acct_file_reopen(file);
-	spin_unlock(&acct_globals.lock);
+	spin_unlock(&acct_lock);
 
 	mntput(file->f_path.mnt); /* it's pinned, now give up active reference */
 
@@ -272,9 +272,9 @@ asmlinkage long sys_acct(const char __user *name)
 	} else {
 		error = security_acct(NULL);
 		if (!error) {
-			spin_lock(&acct_globals.lock);
+			spin_lock(&acct_lock);
 			acct_file_reopen(NULL);
-			spin_unlock(&acct_globals.lock);
+			spin_unlock(&acct_lock);
 		}
 	}
 	return error;
@@ -289,10 +289,10 @@ asmlinkage long sys_acct(const char __user *name)
  */
 void acct_auto_close_mnt(struct vfsmount *m)
 {
-	spin_lock(&acct_globals.lock);
+	spin_lock(&acct_lock);
 	if (acct_globals.file && acct_globals.file->f_path.mnt == m)
 		acct_file_reopen(NULL);
-	spin_unlock(&acct_globals.lock);
+	spin_unlock(&acct_lock);
 }
 
 /**
@@ -304,12 +304,12 @@ void acct_auto_close_mnt(struct vfsmount *m)
  */
 void acct_auto_close(struct super_block *sb)
 {
-	spin_lock(&acct_globals.lock);
+	spin_lock(&acct_lock);
 	if (acct_globals.file &&
 	    acct_globals.file->f_path.mnt->mnt_sb == sb) {
 		acct_file_reopen(NULL);
 	}
-	spin_unlock(&acct_globals.lock);
+	spin_unlock(&acct_lock);
 }
 
 /*
@@ -594,15 +594,15 @@ void acct_process(void)
 	if (!acct_globals.file)
 		return;
 
-	spin_lock(&acct_globals.lock);
+	spin_lock(&acct_lock);
 	file = acct_globals.file;
 	if (unlikely(!file)) {
-		spin_unlock(&acct_globals.lock);
+		spin_unlock(&acct_lock);
 		return;
 	}
 	get_file(file);
 	ns = get_pid_ns(acct_globals.ns);
-	spin_unlock(&acct_globals.lock);
+	spin_unlock(&acct_lock);
 
 	do_acct_process(ns, file);
 	fput(file);

From 6248b1b342005a428b1247b4e89249da1528d88d Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 25 Jul 2008 01:48:46 -0700
Subject: [PATCH 0915/1171] bsdacct: make internal code work with passed
 bsd_acct_struct, not global

This adds the appropriate pointer to all the internal (i.e.  static)
functions that work with global acct instance.  API calls pass a global
instance to them (while we still have such).

Mostly this is a s/acct_globals./acct->/ over the file.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/acct.c | 77 ++++++++++++++++++++++++++-------------------------
 1 file changed, 39 insertions(+), 38 deletions(-)

diff --git a/kernel/acct.c b/kernel/acct.c
index fc71c1304977..72d4760c8da8 100644
--- a/kernel/acct.c
+++ b/kernel/acct.c
@@ -75,7 +75,8 @@ int acct_parm[3] = {4, 2, 30};
 /*
  * External references and all of the globals.
  */
-static void do_acct_process(struct pid_namespace *ns, struct file *);
+static void do_acct_process(struct bsd_acct_struct *acct,
+		struct pid_namespace *ns, struct file *);
 
 /*
  * This structure is used so that all the data protected by lock
@@ -106,7 +107,7 @@ static void acct_timeout(unsigned long x)
 /*
  * Check the amount of free space and suspend/resume accordingly.
  */
-static int check_free_space(struct file *file)
+static int check_free_space(struct bsd_acct_struct *acct, struct file *file)
 {
 	struct kstatfs sbuf;
 	int res;
@@ -115,8 +116,8 @@ static int check_free_space(struct file *file)
 	sector_t suspend;
 
 	spin_lock(&acct_lock);
-	res = acct_globals.active;
-	if (!file || !acct_globals.needcheck)
+	res = acct->active;
+	if (!file || !acct->needcheck)
 		goto out;
 	spin_unlock(&acct_lock);
 
@@ -137,33 +138,33 @@ static int check_free_space(struct file *file)
 		act = 0;
 
 	/*
-	 * If some joker switched acct_globals.file under us we'ld better be
+	 * If some joker switched acct->file under us we'ld better be
 	 * silent and _not_ touch anything.
 	 */
 	spin_lock(&acct_lock);
-	if (file != acct_globals.file) {
+	if (file != acct->file) {
 		if (act)
 			res = act>0;
 		goto out;
 	}
 
-	if (acct_globals.active) {
+	if (acct->active) {
 		if (act < 0) {
-			acct_globals.active = 0;
+			acct->active = 0;
 			printk(KERN_INFO "Process accounting paused\n");
 		}
 	} else {
 		if (act > 0) {
-			acct_globals.active = 1;
+			acct->active = 1;
 			printk(KERN_INFO "Process accounting resumed\n");
 		}
 	}
 
-	del_timer(&acct_globals.timer);
-	acct_globals.needcheck = 0;
-	acct_globals.timer.expires = jiffies + ACCT_TIMEOUT*HZ;
-	add_timer(&acct_globals.timer);
-	res = acct_globals.active;
+	del_timer(&acct->timer);
+	acct->needcheck = 0;
+	acct->timer.expires = jiffies + ACCT_TIMEOUT*HZ;
+	add_timer(&acct->timer);
+	res = acct->active;
 out:
 	spin_unlock(&acct_lock);
 	return res;
@@ -175,34 +176,33 @@ out:
  *
  * NOTE: acct_lock MUST be held on entry and exit.
  */
-static void acct_file_reopen(struct file *file)
+static void acct_file_reopen(struct bsd_acct_struct *acct, struct file *file)
 {
 	struct file *old_acct = NULL;
 	struct pid_namespace *old_ns = NULL;
 
-	if (acct_globals.file) {
-		old_acct = acct_globals.file;
-		old_ns = acct_globals.ns;
-		del_timer(&acct_globals.timer);
-		acct_globals.active = 0;
-		acct_globals.needcheck = 0;
-		acct_globals.file = NULL;
+	if (acct->file) {
+		old_acct = acct->file;
+		old_ns = acct->ns;
+		del_timer(&acct->timer);
+		acct->active = 0;
+		acct->needcheck = 0;
+		acct->file = NULL;
 	}
 	if (file) {
-		acct_globals.file = file;
-		acct_globals.ns = get_pid_ns(task_active_pid_ns(current));
-		acct_globals.needcheck = 0;
-		acct_globals.active = 1;
+		acct->file = file;
+		acct->ns = get_pid_ns(task_active_pid_ns(current));
+		acct->needcheck = 0;
+		acct->active = 1;
 		/* It's been deleted if it was used before so this is safe */
-		setup_timer(&acct_globals.timer, acct_timeout,
-				(unsigned long)&acct_globals);
-		acct_globals.timer.expires = jiffies + ACCT_TIMEOUT*HZ;
-		add_timer(&acct_globals.timer);
+		setup_timer(&acct->timer, acct_timeout, (unsigned long)acct);
+		acct->timer.expires = jiffies + ACCT_TIMEOUT*HZ;
+		add_timer(&acct->timer);
 	}
 	if (old_acct) {
 		mnt_unpin(old_acct->f_path.mnt);
 		spin_unlock(&acct_lock);
-		do_acct_process(old_ns, old_acct);
+		do_acct_process(acct, old_ns, old_acct);
 		filp_close(old_acct, NULL);
 		put_pid_ns(old_ns);
 		spin_lock(&acct_lock);
@@ -237,7 +237,7 @@ static int acct_on(char *name)
 
 	spin_lock(&acct_lock);
 	mnt_pin(file->f_path.mnt);
-	acct_file_reopen(file);
+	acct_file_reopen(&acct_globals, file);
 	spin_unlock(&acct_lock);
 
 	mntput(file->f_path.mnt); /* it's pinned, now give up active reference */
@@ -273,7 +273,7 @@ asmlinkage long sys_acct(const char __user *name)
 		error = security_acct(NULL);
 		if (!error) {
 			spin_lock(&acct_lock);
-			acct_file_reopen(NULL);
+			acct_file_reopen(&acct_globals, NULL);
 			spin_unlock(&acct_lock);
 		}
 	}
@@ -291,7 +291,7 @@ void acct_auto_close_mnt(struct vfsmount *m)
 {
 	spin_lock(&acct_lock);
 	if (acct_globals.file && acct_globals.file->f_path.mnt == m)
-		acct_file_reopen(NULL);
+		acct_file_reopen(&acct_globals, NULL);
 	spin_unlock(&acct_lock);
 }
 
@@ -307,7 +307,7 @@ void acct_auto_close(struct super_block *sb)
 	spin_lock(&acct_lock);
 	if (acct_globals.file &&
 	    acct_globals.file->f_path.mnt->mnt_sb == sb) {
-		acct_file_reopen(NULL);
+		acct_file_reopen(&acct_globals, NULL);
 	}
 	spin_unlock(&acct_lock);
 }
@@ -426,7 +426,8 @@ static u32 encode_float(u64 value)
 /*
  *  do_acct_process does all actual work. Caller holds the reference to file.
  */
-static void do_acct_process(struct pid_namespace *ns, struct file *file)
+static void do_acct_process(struct bsd_acct_struct *acct,
+		struct pid_namespace *ns, struct file *file)
 {
 	struct pacct_struct *pacct = &current->signal->pacct;
 	acct_t ac;
@@ -441,7 +442,7 @@ static void do_acct_process(struct pid_namespace *ns, struct file *file)
 	 * First check to see if there is enough free_space to continue
 	 * the process accounting system.
 	 */
-	if (!check_free_space(file))
+	if (!check_free_space(acct, file))
 		return;
 
 	/*
@@ -604,7 +605,7 @@ void acct_process(void)
 	ns = get_pid_ns(acct_globals.ns);
 	spin_unlock(&acct_lock);
 
-	do_acct_process(ns, file);
+	do_acct_process(&acct_globals, ns, file);
 	fput(file);
 	put_pid_ns(ns);
 }

From 0b6b030fc30d169bb406b34b4fc60d99dde4a9c6 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 25 Jul 2008 01:48:47 -0700
Subject: [PATCH 0916/1171] bsdacct: switch from global bsd_acct_struct
 instance to per-pidns one

Allocate the structure on the first call to sys_acct().  After this each
namespace, that ordered the accounting, will live with this structure till
its own death.

Two notes
- routines, that close the accounting on fs umount time use
  the init_pid_ns's acct by now;
- accounting routine accounts to dying task's namespace
  (also by now).

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/acct.h   |  3 ++
 kernel/acct.c          | 84 +++++++++++++++++++++++++++++++++---------
 kernel/pid_namespace.c |  2 +
 3 files changed, 71 insertions(+), 18 deletions(-)

diff --git a/include/linux/acct.h b/include/linux/acct.h
index e8cae54e8d88..882dc7248766 100644
--- a/include/linux/acct.h
+++ b/include/linux/acct.h
@@ -120,17 +120,20 @@ struct acct_v3
 struct vfsmount;
 struct super_block;
 struct pacct_struct;
+struct pid_namespace;
 extern void acct_auto_close_mnt(struct vfsmount *m);
 extern void acct_auto_close(struct super_block *sb);
 extern void acct_init_pacct(struct pacct_struct *pacct);
 extern void acct_collect(long exitcode, int group_dead);
 extern void acct_process(void);
+extern void acct_exit_ns(struct pid_namespace *);
 #else
 #define acct_auto_close_mnt(x)	do { } while (0)
 #define acct_auto_close(x)	do { } while (0)
 #define acct_init_pacct(x)	do { } while (0)
 #define acct_collect(x,y)	do { } while (0)
 #define acct_process()		do { } while (0)
+#define acct_exit_ns(ns)	do { } while (0)
 #endif
 
 /*
diff --git a/kernel/acct.c b/kernel/acct.c
index 72d4760c8da8..febbbc67157e 100644
--- a/kernel/acct.c
+++ b/kernel/acct.c
@@ -93,8 +93,6 @@ struct bsd_acct_struct {
 
 static DEFINE_SPINLOCK(acct_lock);
 
-static struct bsd_acct_struct acct_globals __cacheline_aligned;
-
 /*
  * Called whenever the timer says to check the free space.
  */
@@ -176,7 +174,8 @@ out:
  *
  * NOTE: acct_lock MUST be held on entry and exit.
  */
-static void acct_file_reopen(struct bsd_acct_struct *acct, struct file *file)
+static void acct_file_reopen(struct bsd_acct_struct *acct, struct file *file,
+		struct pid_namespace *ns)
 {
 	struct file *old_acct = NULL;
 	struct pid_namespace *old_ns = NULL;
@@ -188,10 +187,11 @@ static void acct_file_reopen(struct bsd_acct_struct *acct, struct file *file)
 		acct->active = 0;
 		acct->needcheck = 0;
 		acct->file = NULL;
+		acct->ns = NULL;
 	}
 	if (file) {
 		acct->file = file;
-		acct->ns = get_pid_ns(task_active_pid_ns(current));
+		acct->ns = ns;
 		acct->needcheck = 0;
 		acct->active = 1;
 		/* It's been deleted if it was used before so this is safe */
@@ -204,7 +204,6 @@ static void acct_file_reopen(struct bsd_acct_struct *acct, struct file *file)
 		spin_unlock(&acct_lock);
 		do_acct_process(acct, old_ns, old_acct);
 		filp_close(old_acct, NULL);
-		put_pid_ns(old_ns);
 		spin_lock(&acct_lock);
 	}
 }
@@ -213,6 +212,8 @@ static int acct_on(char *name)
 {
 	struct file *file;
 	int error;
+	struct pid_namespace *ns;
+	struct bsd_acct_struct *acct = NULL;
 
 	/* Difference from BSD - they don't do O_APPEND */
 	file = filp_open(name, O_WRONLY|O_APPEND|O_LARGEFILE, 0);
@@ -229,18 +230,34 @@ static int acct_on(char *name)
 		return -EIO;
 	}
 
+	ns = task_active_pid_ns(current);
+	if (ns->bacct == NULL) {
+		acct = kzalloc(sizeof(struct bsd_acct_struct), GFP_KERNEL);
+		if (acct == NULL) {
+			filp_close(file, NULL);
+			return -ENOMEM;
+		}
+	}
+
 	error = security_acct(file);
 	if (error) {
+		kfree(acct);
 		filp_close(file, NULL);
 		return error;
 	}
 
 	spin_lock(&acct_lock);
+	if (ns->bacct == NULL) {
+		ns->bacct = acct;
+		acct = NULL;
+	}
+
 	mnt_pin(file->f_path.mnt);
-	acct_file_reopen(&acct_globals, file);
+	acct_file_reopen(ns->bacct, file, ns);
 	spin_unlock(&acct_lock);
 
 	mntput(file->f_path.mnt); /* it's pinned, now give up active reference */
+	kfree(acct);
 
 	return 0;
 }
@@ -270,10 +287,16 @@ asmlinkage long sys_acct(const char __user *name)
 		error = acct_on(tmp);
 		putname(tmp);
 	} else {
+		struct bsd_acct_struct *acct;
+
+		acct = task_active_pid_ns(current)->bacct;
+		if (acct == NULL)
+			return 0;
+
 		error = security_acct(NULL);
 		if (!error) {
 			spin_lock(&acct_lock);
-			acct_file_reopen(&acct_globals, NULL);
+			acct_file_reopen(acct, NULL, NULL);
 			spin_unlock(&acct_lock);
 		}
 	}
@@ -289,9 +312,15 @@ asmlinkage long sys_acct(const char __user *name)
  */
 void acct_auto_close_mnt(struct vfsmount *m)
 {
+	struct bsd_acct_struct *acct;
+
+	acct = init_pid_ns.bacct;
+	if (acct == NULL)
+		return;
+
 	spin_lock(&acct_lock);
-	if (acct_globals.file && acct_globals.file->f_path.mnt == m)
-		acct_file_reopen(&acct_globals, NULL);
+	if (acct->file && acct->file->f_path.mnt == m)
+		acct_file_reopen(acct, NULL, NULL);
 	spin_unlock(&acct_lock);
 }
 
@@ -304,10 +333,29 @@ void acct_auto_close_mnt(struct vfsmount *m)
  */
 void acct_auto_close(struct super_block *sb)
 {
+	struct bsd_acct_struct *acct;
+
+	acct = init_pid_ns.bacct;
+	if (acct == NULL)
+		return;
+
 	spin_lock(&acct_lock);
-	if (acct_globals.file &&
-	    acct_globals.file->f_path.mnt->mnt_sb == sb) {
-		acct_file_reopen(&acct_globals, NULL);
+	if (acct->file && acct->file->f_path.mnt->mnt_sb == sb)
+		acct_file_reopen(acct, NULL, NULL);
+	spin_unlock(&acct_lock);
+}
+
+void acct_exit_ns(struct pid_namespace *ns)
+{
+	struct bsd_acct_struct *acct;
+
+	spin_lock(&acct_lock);
+	acct = ns->bacct;
+	if (acct != NULL) {
+		if (acct->file != NULL)
+			acct_file_reopen(acct, NULL, NULL);
+
+		kfree(acct);
 	}
 	spin_unlock(&acct_lock);
 }
@@ -587,25 +635,25 @@ void acct_collect(long exitcode, int group_dead)
 void acct_process(void)
 {
 	struct file *file = NULL;
-	struct pid_namespace *ns;
+	struct pid_namespace *ns = task_active_pid_ns(current);
+	struct bsd_acct_struct *acct;
 
+	acct = ns->bacct;
 	/*
 	 * accelerate the common fastpath:
 	 */
-	if (!acct_globals.file)
+	if (!acct || !acct->file)
 		return;
 
 	spin_lock(&acct_lock);
-	file = acct_globals.file;
+	file = acct->file;
 	if (unlikely(!file)) {
 		spin_unlock(&acct_lock);
 		return;
 	}
 	get_file(file);
-	ns = get_pid_ns(acct_globals.ns);
 	spin_unlock(&acct_lock);
 
-	do_acct_process(&acct_globals, ns, file);
+	do_acct_process(acct, ns, file);
 	fput(file);
-	put_pid_ns(ns);
 }
diff --git a/kernel/pid_namespace.c b/kernel/pid_namespace.c
index 06331cc1c3f5..ea567b78d1aa 100644
--- a/kernel/pid_namespace.c
+++ b/kernel/pid_namespace.c
@@ -12,6 +12,7 @@
 #include <linux/pid_namespace.h>
 #include <linux/syscalls.h>
 #include <linux/err.h>
+#include <linux/acct.h>
 
 #define BITS_PER_PAGE		(PAGE_SIZE*8)
 
@@ -181,6 +182,7 @@ void zap_pid_ns_processes(struct pid_namespace *pid_ns)
 
 	/* Child reaper for the pid namespace is going away */
 	pid_ns->child_reaper = NULL;
+	acct_exit_ns(pid_ns);
 	return;
 }
 

From b5a7174875ea570cc675f2c503e800db8efdd6a7 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 25 Jul 2008 01:48:47 -0700
Subject: [PATCH 0917/1171] bsdacct: turn acct off for all pidns-s on umount
 time

All the bsd_acct_strcts with opened accounting are linked into a global
list.  So, the acct_auto_close(_mnt) walks one and drops the accounting
for each.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/acct.c | 28 ++++++++++++++++------------
 1 file changed, 16 insertions(+), 12 deletions(-)

diff --git a/kernel/acct.c b/kernel/acct.c
index febbbc67157e..7fc9f9dd1e9e 100644
--- a/kernel/acct.c
+++ b/kernel/acct.c
@@ -89,9 +89,11 @@ struct bsd_acct_struct {
 	struct file		*file;
 	struct pid_namespace	*ns;
 	struct timer_list	timer;
+	struct list_head	list;
 };
 
 static DEFINE_SPINLOCK(acct_lock);
+static LIST_HEAD(acct_list);
 
 /*
  * Called whenever the timer says to check the free space.
@@ -188,12 +190,14 @@ static void acct_file_reopen(struct bsd_acct_struct *acct, struct file *file,
 		acct->needcheck = 0;
 		acct->file = NULL;
 		acct->ns = NULL;
+		list_del(&acct->list);
 	}
 	if (file) {
 		acct->file = file;
 		acct->ns = ns;
 		acct->needcheck = 0;
 		acct->active = 1;
+		list_add(&acct->list, &acct_list);
 		/* It's been deleted if it was used before so this is safe */
 		setup_timer(&acct->timer, acct_timeout, (unsigned long)acct);
 		acct->timer.expires = jiffies + ACCT_TIMEOUT*HZ;
@@ -314,13 +318,13 @@ void acct_auto_close_mnt(struct vfsmount *m)
 {
 	struct bsd_acct_struct *acct;
 
-	acct = init_pid_ns.bacct;
-	if (acct == NULL)
-		return;
-
 	spin_lock(&acct_lock);
-	if (acct->file && acct->file->f_path.mnt == m)
-		acct_file_reopen(acct, NULL, NULL);
+restart:
+	list_for_each_entry(acct, &acct_list, list)
+		if (acct->file && acct->file->f_path.mnt == m) {
+			acct_file_reopen(acct, NULL, NULL);
+			goto restart;
+		}
 	spin_unlock(&acct_lock);
 }
 
@@ -335,13 +339,13 @@ void acct_auto_close(struct super_block *sb)
 {
 	struct bsd_acct_struct *acct;
 
-	acct = init_pid_ns.bacct;
-	if (acct == NULL)
-		return;
-
 	spin_lock(&acct_lock);
-	if (acct->file && acct->file->f_path.mnt->mnt_sb == sb)
-		acct_file_reopen(acct, NULL, NULL);
+restart:
+	list_for_each_entry(acct, &acct_list, list)
+		if (acct->file && acct->file->f_path.mnt->mnt_sb == sb) {
+			acct_file_reopen(acct, NULL, NULL);
+			goto restart;
+		}
 	spin_unlock(&acct_lock);
 }
 

From 7d1e13505be8c2bd2207894f4e0f069e1f9b51c9 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 25 Jul 2008 01:48:48 -0700
Subject: [PATCH 0918/1171] bsdacct: account dying tasks in all relevant
 namespaces

This just makes the acct_proces walk the pid namespaces from current up to
the top and account a task in each with the accounting turned on.

ns->parent access if safe lockless, since current it still alive and holds
its namespace, which in turn holds its parent.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/acct.c | 21 ++++++++++++++-------
 1 file changed, 14 insertions(+), 7 deletions(-)

diff --git a/kernel/acct.c b/kernel/acct.c
index 7fc9f9dd1e9e..0feba97e114e 100644
--- a/kernel/acct.c
+++ b/kernel/acct.c
@@ -631,15 +631,9 @@ void acct_collect(long exitcode, int group_dead)
 	spin_unlock_irq(&current->sighand->siglock);
 }
 
-/**
- * acct_process - now just a wrapper around do_acct_process
- *
- * handles process accounting for an exiting task
- */
-void acct_process(void)
+static void acct_process_in_ns(struct pid_namespace *ns)
 {
 	struct file *file = NULL;
-	struct pid_namespace *ns = task_active_pid_ns(current);
 	struct bsd_acct_struct *acct;
 
 	acct = ns->bacct;
@@ -661,3 +655,16 @@ void acct_process(void)
 	do_acct_process(acct, ns, file);
 	fput(file);
 }
+
+/**
+ * acct_process - now just a wrapper around do_acct_process
+ *
+ * handles process accounting for an exiting task
+ */
+void acct_process(void)
+{
+	struct pid_namespace *ns;
+
+	for (ns = task_active_pid_ns(current); ns != NULL; ns = ns->parent)
+		acct_process_in_ns(ns);
+}

From 0c18d7a5df82524e634637c3aec24d4cba096442 Mon Sep 17 00:00:00 2001
From: Pavel Emelyanov <xemul@openvz.org>
Date: Fri, 25 Jul 2008 01:48:49 -0700
Subject: [PATCH 0919/1171] bsdacct: fix and add comments around acct_process()

Fix the one describing what this function is and add one more - about
locking absence around pid namespaces loop.

Signed-off-by: Pavel Emelyanov <xemul@openvz.org>
Cc: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/acct.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/kernel/acct.c b/kernel/acct.c
index 0feba97e114e..dd68b9059418 100644
--- a/kernel/acct.c
+++ b/kernel/acct.c
@@ -657,7 +657,8 @@ static void acct_process_in_ns(struct pid_namespace *ns)
 }
 
 /**
- * acct_process - now just a wrapper around do_acct_process
+ * acct_process - now just a wrapper around acct_process_in_ns,
+ * which in turn is a wrapper around do_acct_process.
  *
  * handles process accounting for an exiting task
  */
@@ -665,6 +666,11 @@ void acct_process(void)
 {
 	struct pid_namespace *ns;
 
+	/*
+	 * This loop is safe lockless, since current is still
+	 * alive and holds its namespace, which in turn holds
+	 * its parent.
+	 */
 	for (ns = task_active_pid_ns(current); ns != NULL; ns = ns->parent)
 		acct_process_in_ns(ns);
 }

From 297c5d92634c809cef23d73e7b2556f2528ff7e2 Mon Sep 17 00:00:00 2001
From: Andrea Righi <righi.andrea@gmail.com>
Date: Fri, 25 Jul 2008 01:48:49 -0700
Subject: [PATCH 0920/1171] task IO accounting: provide distinct tgid/tid I/O
 statistics

Report per-thread I/O statistics in /proc/pid/task/tid/io and aggregate
parent I/O statistics in /proc/pid/io.  This approach follows the same
model used to account per-process and per-thread CPU times.

As a practial application, this allows for example to quickly find the top
I/O consumer when a process spawns many child threads that perform the
actual I/O work, because the aggregated I/O statistics can always be found
in /proc/pid/io.

[ Oleg Nesterov points out that we should check that the task is still
  alive before we iterate over the threads, but also says that we can do
  that fixup on top of this later.  - Linus ]

Acked-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Signed-off-by: Andrea Righi <righi.andrea@gmail.com>
Cc: Matt Heaton <matt@hostmonster.com>
Cc: Shailabh Nagar <nagar@watson.ibm.com>
Acked-by-with-comments: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/proc/base.c        | 84 +++++++++++++++++++++++++++++++++++--------
 include/linux/sched.h |  4 +++
 kernel/exit.c         | 27 ++++++++++++++
 kernel/fork.c         |  6 ++++
 4 files changed, 107 insertions(+), 14 deletions(-)

diff --git a/fs/proc/base.c b/fs/proc/base.c
index 58c3e6a8e15e..a891fe4cb43b 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -2376,29 +2376,82 @@ static int proc_base_fill_cache(struct file *filp, void *dirent,
 }
 
 #ifdef CONFIG_TASK_IO_ACCOUNTING
-static int proc_pid_io_accounting(struct task_struct *task, char *buffer)
+static int do_io_accounting(struct task_struct *task, char *buffer, int whole)
 {
+	u64 rchar, wchar, syscr, syscw;
+	struct task_io_accounting ioac;
+
+	if (!whole) {
+		rchar = task->rchar;
+		wchar = task->wchar;
+		syscr = task->syscr;
+		syscw = task->syscw;
+		memcpy(&ioac, &task->ioac, sizeof(ioac));
+	} else {
+		unsigned long flags;
+		struct task_struct *t = task;
+		rchar = wchar = syscr = syscw = 0;
+		memset(&ioac, 0, sizeof(ioac));
+
+		rcu_read_lock();
+		do {
+			rchar += t->rchar;
+			wchar += t->wchar;
+			syscr += t->syscr;
+			syscw += t->syscw;
+
+			ioac.read_bytes += t->ioac.read_bytes;
+			ioac.write_bytes += t->ioac.write_bytes;
+			ioac.cancelled_write_bytes +=
+					t->ioac.cancelled_write_bytes;
+			t = next_thread(t);
+		} while (t != task);
+		rcu_read_unlock();
+
+		if (lock_task_sighand(task, &flags)) {
+			struct signal_struct *sig = task->signal;
+
+			rchar += sig->rchar;
+			wchar += sig->wchar;
+			syscr += sig->syscr;
+			syscw += sig->syscw;
+
+			ioac.read_bytes += sig->ioac.read_bytes;
+			ioac.write_bytes += sig->ioac.write_bytes;
+			ioac.cancelled_write_bytes +=
+					sig->ioac.cancelled_write_bytes;
+
+			unlock_task_sighand(task, &flags);
+		}
+	}
+
 	return sprintf(buffer,
-#ifdef CONFIG_TASK_XACCT
 			"rchar: %llu\n"
 			"wchar: %llu\n"
 			"syscr: %llu\n"
 			"syscw: %llu\n"
-#endif
 			"read_bytes: %llu\n"
 			"write_bytes: %llu\n"
 			"cancelled_write_bytes: %llu\n",
-#ifdef CONFIG_TASK_XACCT
-			(unsigned long long)task->rchar,
-			(unsigned long long)task->wchar,
-			(unsigned long long)task->syscr,
-			(unsigned long long)task->syscw,
-#endif
-			(unsigned long long)task->ioac.read_bytes,
-			(unsigned long long)task->ioac.write_bytes,
-			(unsigned long long)task->ioac.cancelled_write_bytes);
+			(unsigned long long)rchar,
+			(unsigned long long)wchar,
+			(unsigned long long)syscr,
+			(unsigned long long)syscw,
+			(unsigned long long)ioac.read_bytes,
+			(unsigned long long)ioac.write_bytes,
+			(unsigned long long)ioac.cancelled_write_bytes);
 }
-#endif
+
+static int proc_tid_io_accounting(struct task_struct *task, char *buffer)
+{
+	return do_io_accounting(task, buffer, 0);
+}
+
+static int proc_tgid_io_accounting(struct task_struct *task, char *buffer)
+{
+	return do_io_accounting(task, buffer, 1);
+}
+#endif /* CONFIG_TASK_IO_ACCOUNTING */
 
 /*
  * Thread groups
@@ -2470,7 +2523,7 @@ static const struct pid_entry tgid_base_stuff[] = {
 	REG("coredump_filter", S_IRUGO|S_IWUSR, coredump_filter),
 #endif
 #ifdef CONFIG_TASK_IO_ACCOUNTING
-	INF("io",	S_IRUGO, pid_io_accounting),
+	INF("io",	S_IRUGO, tgid_io_accounting),
 #endif
 };
 
@@ -2797,6 +2850,9 @@ static const struct pid_entry tid_base_stuff[] = {
 #ifdef CONFIG_FAULT_INJECTION
 	REG("make-it-fail", S_IRUGO|S_IWUSR, fault_inject),
 #endif
+#ifdef CONFIG_TASK_IO_ACCOUNTING
+	INF("io",	S_IRUGO, tid_io_accounting),
+#endif
 };
 
 static int proc_tid_base_readdir(struct file * filp,
diff --git a/include/linux/sched.h b/include/linux/sched.h
index af780f299c7c..d22ffe06d0eb 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -506,6 +506,10 @@ struct signal_struct {
 	unsigned long nvcsw, nivcsw, cnvcsw, cnivcsw;
 	unsigned long min_flt, maj_flt, cmin_flt, cmaj_flt;
 	unsigned long inblock, oublock, cinblock, coublock;
+#ifdef CONFIG_TASK_XACCT
+	u64 rchar, wchar, syscr, syscw;
+#endif
+	struct task_io_accounting ioac;
 
 	/*
 	 * Cumulative ns of scheduled CPU time for dead threads in the
diff --git a/kernel/exit.c b/kernel/exit.c
index 8a4d4d12e294..ad933bb29ec7 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -120,6 +120,18 @@ static void __exit_signal(struct task_struct *tsk)
 		sig->nivcsw += tsk->nivcsw;
 		sig->inblock += task_io_get_inblock(tsk);
 		sig->oublock += task_io_get_oublock(tsk);
+#ifdef CONFIG_TASK_XACCT
+		sig->rchar += tsk->rchar;
+		sig->wchar += tsk->wchar;
+		sig->syscr += tsk->syscr;
+		sig->syscw += tsk->syscw;
+#endif /* CONFIG_TASK_XACCT */
+#ifdef CONFIG_TASK_IO_ACCOUNTING
+		sig->ioac.read_bytes += tsk->ioac.read_bytes;
+		sig->ioac.write_bytes += tsk->ioac.write_bytes;
+		sig->ioac.cancelled_write_bytes +=
+					tsk->ioac.cancelled_write_bytes;
+#endif /* CONFIG_TASK_IO_ACCOUNTING */
 		sig->sum_sched_runtime += tsk->se.sum_exec_runtime;
 		sig = NULL; /* Marker for below. */
 	}
@@ -1366,6 +1378,21 @@ static int wait_task_zombie(struct task_struct *p, int options,
 		psig->coublock +=
 			task_io_get_oublock(p) +
 			sig->oublock + sig->coublock;
+#ifdef CONFIG_TASK_XACCT
+		psig->rchar += p->rchar + sig->rchar;
+		psig->wchar += p->wchar + sig->wchar;
+		psig->syscr += p->syscr + sig->syscr;
+		psig->syscw += p->syscw + sig->syscw;
+#endif /* CONFIG_TASK_XACCT */
+#ifdef CONFIG_TASK_IO_ACCOUNTING
+		psig->ioac.read_bytes +=
+			p->ioac.read_bytes + sig->ioac.read_bytes;
+		psig->ioac.write_bytes +=
+			p->ioac.write_bytes + sig->ioac.write_bytes;
+		psig->ioac.cancelled_write_bytes +=
+				p->ioac.cancelled_write_bytes +
+				sig->ioac.cancelled_write_bytes;
+#endif /* CONFIG_TASK_IO_ACCOUNTING */
 		spin_unlock_irq(&p->parent->sighand->siglock);
 	}
 
diff --git a/kernel/fork.c b/kernel/fork.c
index 813d5c89b9d5..b99d73e971a4 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -812,6 +812,12 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk)
 	sig->nvcsw = sig->nivcsw = sig->cnvcsw = sig->cnivcsw = 0;
 	sig->min_flt = sig->maj_flt = sig->cmin_flt = sig->cmaj_flt = 0;
 	sig->inblock = sig->oublock = sig->cinblock = sig->coublock = 0;
+#ifdef CONFIG_TASK_XACCT
+	sig->rchar = sig->wchar = sig->syscr = sig->syscw = 0;
+#endif
+#ifdef CONFIG_TASK_IO_ACCOUNTING
+	memset(&sig->ioac, 0, sizeof(sig->ioac));
+#endif
 	sig->sum_sched_runtime = 0;
 	INIT_LIST_HEAD(&sig->cpu_timers[0]);
 	INIT_LIST_HEAD(&sig->cpu_timers[1]);

From 3e85ba034deec351f02cb55ff225bbd616463841 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Fri, 25 Jul 2008 01:48:50 -0700
Subject: [PATCH 0921/1171] tsacct: fix bacct_add_tsk()'s use of do_div()

Fix bacct_add_tsk()'s use of do_div() on an s64 by making ac_etime a u64
instead and dividing that.

Possibly this should be guarded lest the interval calculation turn up
negative, but the possible negativity of the result of the division is
cast away, and it shouldn't end up negative anyway.

This was introduced by patch f3cef7a99469afc159fec3a61b42dc7ca5b6824f.

Signed-off-by: David Howells <dhowells@redhat.com>
Cc: Jay Lan <jlan@engr.sgi.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/tsacct.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/tsacct.c b/kernel/tsacct.c
index 1da6990af8e0..3da47ccdc5e5 100644
--- a/kernel/tsacct.c
+++ b/kernel/tsacct.c
@@ -28,14 +28,14 @@
 void bacct_add_tsk(struct taskstats *stats, struct task_struct *tsk)
 {
 	struct timespec uptime, ts;
-	s64 ac_etime;
+	u64 ac_etime;
 
 	BUILD_BUG_ON(TS_COMM_LEN < TASK_COMM_LEN);
 
 	/* calculate task elapsed time in timespec */
 	do_posix_clock_monotonic_gettime(&uptime);
 	ts = timespec_sub(uptime, tsk->start_time);
-	/* rebase elapsed time to usec */
+	/* rebase elapsed time to usec (should never be negative) */
 	ac_etime = timespec_to_ns(&ts);
 	do_div(ac_etime, NSEC_PER_USEC);
 	stats->ac_etime = ac_etime;

From 873b47717732c2f33a4b14de02571a4295a02f0c Mon Sep 17 00:00:00 2001
From: Keika Kobayashi <kobayashi.kk@ncos.nec.co.jp>
Date: Fri, 25 Jul 2008 01:48:52 -0700
Subject: [PATCH 0922/1171] per-task-delay-accounting: add memory reclaim delay

Sometimes, application responses become bad under heavy memory load.
Applications take a bit time to reclaim memory.  The statistics, how long
memory reclaim takes, will be useful to measure memory usage.

This patch adds accounting memory reclaim to per-task-delay-accounting for
accounting the time of do_try_to_free_pages().

<i.e>

- When System is under low memory load,
  memory reclaim may not occur.

$ free
             total       used       free     shared    buffers     cached
Mem:       8197800    1577300    6620500          0       4808    1516724
-/+ buffers/cache:      55768    8142032
Swap:     16386292          0   16386292

$ vmstat 1
procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa
 0  0      0 5069748  10612 3014060    0    0     0     0    3   26  0  0 100  0
 0  0      0 5069748  10612 3014060    0    0     0     0    4   22  0  0 100  0
 0  0      0 5069748  10612 3014060    0    0     0     0    3   18  0  0 100  0

Measure the time of tar command.

$ ls -s test.dat
1501472 test.dat

$ time tar cvf test.tar test.dat
real    0m13.388s
user    0m0.116s
sys     0m5.304s

$ ./delayget -d -p <pid>
CPU             count     real total  virtual total    delay total
                  428     5528345500     5477116080       62749891
IO              count    delay total
                  338     8078977189
SWAP            count    delay total
                    0              0
RECLAIM         count    delay total
                    0              0

- When system is under heavy memory load
  memory reclaim may occur.

$ vmstat 1
procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa
 0  0 7159032  49724   1812   3012    0    0     0     0    3   24  0  0 100  0
 0  0 7159032  49724   1812   3012    0    0     0     0    4   24  0  0 100  0
 0  0 7159032  49848   1812   3012    0    0     0     0    3   22  0  0 100  0

In this case, one process uses more 8G memory
by execution of malloc() and memset().

$ time tar cvf test.tar test.dat
real    1m38.563s        <-  increased by 85 sec
user    0m0.140s
sys     0m7.060s

$ ./delayget -d -p <pid>
CPU             count     real total  virtual total    delay total
                 9021     7140446250     7315277975      923201824
IO              count    delay total
                 8965    90466349669
SWAP            count    delay total
                    3       21036367
RECLAIM         count    delay total
                  740    61011951153

In the later case, the value of RECLAIM is increasing.
So, taskstats can show how much memory reclaim influences TAT.

Signed-off-by: Keika Kobayashi <kobayashi.kk@ncos.nec.co.jp>
Acked-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujistu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/delayacct.h | 19 +++++++++++++++++++
 include/linux/sched.h     |  4 ++++
 kernel/delayacct.c        | 13 +++++++++++++
 mm/vmscan.c               |  5 +++++
 4 files changed, 41 insertions(+)

diff --git a/include/linux/delayacct.h b/include/linux/delayacct.h
index ab94bc083558..f352f06fa063 100644
--- a/include/linux/delayacct.h
+++ b/include/linux/delayacct.h
@@ -39,6 +39,8 @@ extern void __delayacct_blkio_start(void);
 extern void __delayacct_blkio_end(void);
 extern int __delayacct_add_tsk(struct taskstats *, struct task_struct *);
 extern __u64 __delayacct_blkio_ticks(struct task_struct *);
+extern void __delayacct_freepages_start(void);
+extern void __delayacct_freepages_end(void);
 
 static inline int delayacct_is_task_waiting_on_io(struct task_struct *p)
 {
@@ -107,6 +109,18 @@ static inline __u64 delayacct_blkio_ticks(struct task_struct *tsk)
 	return 0;
 }
 
+static inline void delayacct_freepages_start(void)
+{
+	if (current->delays)
+		__delayacct_freepages_start();
+}
+
+static inline void delayacct_freepages_end(void)
+{
+	if (current->delays)
+		__delayacct_freepages_end();
+}
+
 #else
 static inline void delayacct_set_flag(int flag)
 {}
@@ -129,6 +143,11 @@ static inline __u64 delayacct_blkio_ticks(struct task_struct *tsk)
 { return 0; }
 static inline int delayacct_is_task_waiting_on_io(struct task_struct *p)
 { return 0; }
+static inline void delayacct_freepages_start(void)
+{}
+static inline void delayacct_freepages_end(void)
+{}
+
 #endif /* CONFIG_TASK_DELAY_ACCT */
 
 #endif
diff --git a/include/linux/sched.h b/include/linux/sched.h
index d22ffe06d0eb..42036ffe6b00 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -672,6 +672,10 @@ struct task_delay_info {
 				/* io operations performed */
 	u32 swapin_count;	/* total count of the number of swapin block */
 				/* io operations performed */
+
+	struct timespec freepages_start, freepages_end;
+	u64 freepages_delay;	/* wait for memory reclaim */
+	u32 freepages_count;	/* total count of memory reclaim */
 };
 #endif	/* CONFIG_TASK_DELAY_ACCT */
 
diff --git a/kernel/delayacct.c b/kernel/delayacct.c
index 10e43fd8b721..84b6782a2ce4 100644
--- a/kernel/delayacct.c
+++ b/kernel/delayacct.c
@@ -165,3 +165,16 @@ __u64 __delayacct_blkio_ticks(struct task_struct *tsk)
 	return ret;
 }
 
+void __delayacct_freepages_start(void)
+{
+	delayacct_start(&current->delays->freepages_start);
+}
+
+void __delayacct_freepages_end(void)
+{
+	delayacct_end(&current->delays->freepages_start,
+			&current->delays->freepages_end,
+			&current->delays->freepages_delay,
+			&current->delays->freepages_count);
+}
+
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 967d30ccd92b..26672c6cd3ce 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -38,6 +38,7 @@
 #include <linux/kthread.h>
 #include <linux/freezer.h>
 #include <linux/memcontrol.h>
+#include <linux/delayacct.h>
 
 #include <asm/tlbflush.h>
 #include <asm/div64.h>
@@ -1316,6 +1317,8 @@ static unsigned long do_try_to_free_pages(struct zonelist *zonelist,
 	struct zone *zone;
 	enum zone_type high_zoneidx = gfp_zone(sc->gfp_mask);
 
+	delayacct_freepages_start();
+
 	if (scan_global_lru(sc))
 		count_vm_event(ALLOCSTALL);
 	/*
@@ -1396,6 +1399,8 @@ out:
 	} else
 		mem_cgroup_record_reclaim_priority(sc->mem_cgroup, priority);
 
+	delayacct_freepages_end();
+
 	return ret;
 }
 

From 016ae219b920c4e606088761d3d6070cdf8ba706 Mon Sep 17 00:00:00 2001
From: Keika Kobayashi <kobayashi.kk@ncos.nec.co.jp>
Date: Fri, 25 Jul 2008 01:48:53 -0700
Subject: [PATCH 0923/1171] per-task-delay-accounting: update taskstats for
 memory reclaim delay

Add members for memory reclaim delay to taskstats, and accumulate them in
__delayacct_add_tsk() .

Signed-off-by: Keika Kobayashi <kobayashi.kk@ncos.nec.co.jp>
Cc: Hiroshi Shimamoto <h-shimamoto@ct.jp.nec.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/accounting/taskstats-struct.txt | 7 +++++++
 include/linux/taskstats.h                     | 6 +++++-
 kernel/delayacct.c                            | 3 +++
 3 files changed, 15 insertions(+), 1 deletion(-)

diff --git a/Documentation/accounting/taskstats-struct.txt b/Documentation/accounting/taskstats-struct.txt
index cd784f46bf8a..b988d110db59 100644
--- a/Documentation/accounting/taskstats-struct.txt
+++ b/Documentation/accounting/taskstats-struct.txt
@@ -26,6 +26,8 @@ There are three different groups of fields in the struct taskstats:
 
 5) Time accounting for SMT machines
 
+6) Extended delay accounting fields for memory reclaim
+
 Future extension should add fields to the end of the taskstats struct, and
 should not change the relative position of each field within the struct.
 
@@ -170,4 +172,9 @@ struct taskstats {
 	__u64	ac_utimescaled;		/* utime scaled on frequency etc */
 	__u64	ac_stimescaled;		/* stime scaled on frequency etc */
 	__u64	cpu_scaled_run_real_total; /* scaled cpu_run_real_total */
+
+6) Extended delay accounting fields for memory reclaim
+	/* Delay waiting for memory reclaim */
+	__u64	freepages_count;
+	__u64	freepages_delay_total;
 }
diff --git a/include/linux/taskstats.h b/include/linux/taskstats.h
index 5d69c0744fff..18269e956a71 100644
--- a/include/linux/taskstats.h
+++ b/include/linux/taskstats.h
@@ -31,7 +31,7 @@
  */
 
 
-#define TASKSTATS_VERSION	6
+#define TASKSTATS_VERSION	7
 #define TS_COMM_LEN		32	/* should be >= TASK_COMM_LEN
 					 * in linux/sched.h */
 
@@ -157,6 +157,10 @@ struct taskstats {
 	__u64	ac_utimescaled;		/* utime scaled on frequency etc */
 	__u64	ac_stimescaled;		/* stime scaled on frequency etc */
 	__u64	cpu_scaled_run_real_total; /* scaled cpu_run_real_total */
+
+	/* Delay waiting for memory reclaim */
+	__u64	freepages_count;
+	__u64	freepages_delay_total;
 };
 
 
diff --git a/kernel/delayacct.c b/kernel/delayacct.c
index 84b6782a2ce4..b3179dad71be 100644
--- a/kernel/delayacct.c
+++ b/kernel/delayacct.c
@@ -145,8 +145,11 @@ int __delayacct_add_tsk(struct taskstats *d, struct task_struct *tsk)
 	d->blkio_delay_total = (tmp < d->blkio_delay_total) ? 0 : tmp;
 	tmp = d->swapin_delay_total + tsk->delays->swapin_delay;
 	d->swapin_delay_total = (tmp < d->swapin_delay_total) ? 0 : tmp;
+	tmp = d->freepages_delay_total + tsk->delays->freepages_delay;
+	d->freepages_delay_total = (tmp < d->freepages_delay_total) ? 0 : tmp;
 	d->blkio_count += tsk->delays->blkio_count;
 	d->swapin_count += tsk->delays->swapin_count;
+	d->freepages_count += tsk->delays->freepages_count;
 	spin_unlock_irqrestore(&tsk->delays->lock, flags);
 
 done:

From 9b0975a20af1ff2f367e3b6b7c150eb114c6b500 Mon Sep 17 00:00:00 2001
From: Keika Kobayashi <kobayashi.kk@ncos.nec.co.jp>
Date: Fri, 25 Jul 2008 01:48:54 -0700
Subject: [PATCH 0924/1171] per-task-delay-accounting: update document and
 getdelays.c for memory reclaim

Update document and make getdelays.c show delay accounting for memory reclaim.

For making a distinction between "swapping in pages" and "memory reclaim"
in getdelays.c, MEM is changed to SWAP.

Signed-off-by: Keika Kobayashi <kobayashi.kk@ncos.nec.co.jp>
Acked-by: Balbir Singh <balbir@linux.vnet.ibm.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/accounting/delay-accounting.txt | 11 ++++++++---
 Documentation/accounting/getdelays.c          |  8 ++++++--
 2 files changed, 14 insertions(+), 5 deletions(-)

diff --git a/Documentation/accounting/delay-accounting.txt b/Documentation/accounting/delay-accounting.txt
index 1443cd71d263..8a12f0730c94 100644
--- a/Documentation/accounting/delay-accounting.txt
+++ b/Documentation/accounting/delay-accounting.txt
@@ -11,6 +11,7 @@ the delays experienced by a task while
 a) waiting for a CPU (while being runnable)
 b) completion of synchronous block I/O initiated by the task
 c) swapping in pages
+d) memory reclaim
 
 and makes these statistics available to userspace through
 the taskstats interface.
@@ -41,7 +42,7 @@ this structure. See
      include/linux/taskstats.h
 for a description of the fields pertaining to delay accounting.
 It will generally be in the form of counters returning the cumulative
-delay seen for cpu, sync block I/O, swapin etc.
+delay seen for cpu, sync block I/O, swapin, memory reclaim etc.
 
 Taking the difference of two successive readings of a given
 counter (say cpu_delay_total) for a task will give the delay
@@ -94,7 +95,9 @@ CPU	count	real total	virtual total	delay total
 	7876	92005750	100000000	24001500
 IO	count	delay total
 	0	0
-MEM	count	delay total
+SWAP	count	delay total
+	0	0
+RECLAIM	count	delay total
 	0	0
 
 Get delays seen in executing a given simple command
@@ -108,5 +111,7 @@ CPU	count	real total	virtual total	delay total
 	6	4000250		4000000		0
 IO	count	delay total
 	0	0
-MEM	count	delay total
+SWAP	count	delay total
+	0	0
+RECLAIM	count	delay total
 	0	0
diff --git a/Documentation/accounting/getdelays.c b/Documentation/accounting/getdelays.c
index 40121b5cca14..3f7755f3963f 100644
--- a/Documentation/accounting/getdelays.c
+++ b/Documentation/accounting/getdelays.c
@@ -196,14 +196,18 @@ void print_delayacct(struct taskstats *t)
 	       "      %15llu%15llu%15llu%15llu\n"
 	       "IO    %15s%15s\n"
 	       "      %15llu%15llu\n"
-	       "MEM   %15s%15s\n"
+	       "SWAP  %15s%15s\n"
+	       "      %15llu%15llu\n"
+	       "RECLAIM  %12s%15s\n"
 	       "      %15llu%15llu\n",
 	       "count", "real total", "virtual total", "delay total",
 	       t->cpu_count, t->cpu_run_real_total, t->cpu_run_virtual_total,
 	       t->cpu_delay_total,
 	       "count", "delay total",
 	       t->blkio_count, t->blkio_delay_total,
-	       "count", "delay total", t->swapin_count, t->swapin_delay_total);
+	       "count", "delay total", t->swapin_count, t->swapin_delay_total,
+	       "count", "delay total",
+	       t->freepages_count, t->freepages_delay_total);
 }
 
 void task_context_switch_counts(struct taskstats *t)

From b81f3ea92ba1fa676775677679889dc2a7f03c8b Mon Sep 17 00:00:00 2001
From: Vegard Nossum <vegard.nossum@gmail.com>
Date: Fri, 25 Jul 2008 01:48:55 -0700
Subject: [PATCH 0925/1171] taskstats: remove initialization of static per-cpu
 variable

Cc: Shailabh Nagar <nagar@watson.ibm.com>
Signed-off-by: Vegard Nossum <vegard.nossum@gmail.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/taskstats.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/taskstats.c b/kernel/taskstats.c
index 06b17547f4e7..bd6be76303cf 100644
--- a/kernel/taskstats.c
+++ b/kernel/taskstats.c
@@ -35,7 +35,7 @@
  */
 #define TASKSTATS_CPUMASK_MAXLEN	(100+6*NR_CPUS)
 
-static DEFINE_PER_CPU(__u32, taskstats_seqnum) = { 0 };
+static DEFINE_PER_CPU(__u32, taskstats_seqnum);
 static int family_registered;
 struct kmem_cache *taskstats_cache;
 

From cc77b1521d06be07c9bb1a4a3e1f775dcaa15093 Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Fri, 25 Jul 2008 01:48:55 -0700
Subject: [PATCH 0926/1171] lockd: dont return EAGAIN for a permanent error

Fix nlm_fopen() to return NLM_FAILED (or NLM_LCK_DENIED_NOLOCKS) instead
of NLM_LCK_DENIED.  The latter means the lock request failed because of a
conflicting lock (i.e.  a temporary error), which is wrong in this case.

Also fix the client to return ENOLCK instead of EAGAIN if a blocking lock
request returns with NLM_LOCK_DENIED.

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: "J. Bruce Fields" <bfields@fieldses.org>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: David Teigland <teigland@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/lockd/clntproc.c | 10 +++++++++-
 fs/nfsd/lockd.c     | 13 +++++++++----
 2 files changed, 18 insertions(+), 5 deletions(-)

diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c
index 1f6dc518505c..31668b690e03 100644
--- a/fs/lockd/clntproc.c
+++ b/fs/lockd/clntproc.c
@@ -582,7 +582,15 @@ again:
 	}
 	if (status < 0)
 		goto out_unlock;
-	status = nlm_stat_to_errno(resp->status);
+	/*
+	 * EAGAIN doesn't make sense for sleeping locks, and in some
+	 * cases NLM_LCK_DENIED is returned for a permanent error.  So
+	 * turn it into an ENOLCK.
+	 */
+	if (resp->status == nlm_lck_denied && (fl_flags & FL_SLEEP))
+		status = -ENOLCK;
+	else
+		status = nlm_stat_to_errno(resp->status);
 out_unblock:
 	nlmclnt_finish_block(block);
 out:
diff --git a/fs/nfsd/lockd.c b/fs/nfsd/lockd.c
index 6b6225ac4926..15c6faeec77c 100644
--- a/fs/nfsd/lockd.c
+++ b/fs/nfsd/lockd.c
@@ -19,6 +19,13 @@
 
 #define NFSDDBG_FACILITY		NFSDDBG_LOCKD
 
+#ifdef CONFIG_LOCKD_V4
+#define nlm_stale_fh	nlm4_stale_fh
+#define nlm_failed	nlm4_failed
+#else
+#define nlm_stale_fh	nlm_lck_denied_nolocks
+#define nlm_failed	nlm_lck_denied_nolocks
+#endif
 /*
  * Note: we hold the dentry use count while the file is open.
  */
@@ -47,12 +54,10 @@ nlm_fopen(struct svc_rqst *rqstp, struct nfs_fh *f, struct file **filp)
 		return 0;
 	case nfserr_dropit:
 		return nlm_drop_reply;
-#ifdef CONFIG_LOCKD_V4
 	case nfserr_stale:
-		return nlm4_stale_fh;
-#endif
+		return nlm_stale_fh;
 	default:
-		return nlm_lck_denied;
+		return nlm_failed;
 	}
 }
 

From bde74e4bc64415b142e556a34d295a52a1b7da9d Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Fri, 25 Jul 2008 01:48:57 -0700
Subject: [PATCH 0927/1171] locks: add special return value for asynchronous
 locks

Use a special error value FILE_LOCK_DEFERRED to mean that a locking
operation returned asynchronously.  This is returned by

  posix_lock_file() for sleeping locks to mean that the lock has been
  queued on the block list, and will be woken up when it might become
  available and needs to be retried (either fl_lmops->fl_notify() is
  called or fl_wait is woken up).

  f_op->lock() to mean either the above, or that the filesystem will
  call back with fl_lmops->fl_grant() when the result of the locking
  operation is known.  The filesystem can do this for sleeping as well
  as non-sleeping locks.

This is to make sure, that return values of -EAGAIN and -EINPROGRESS by
filesystems are not mistaken to mean an asynchronous locking.

This also makes error handling in fs/locks.c and lockd/svclock.c slightly
cleaner.

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: "J. Bruce Fields" <bfields@fieldses.org>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: David Teigland <teigland@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/dlm/plock.c     |  2 +-
 fs/lockd/svclock.c | 13 ++++---------
 fs/locks.c         | 28 ++++++++++++++--------------
 include/linux/fs.h |  6 ++++++
 4 files changed, 25 insertions(+), 24 deletions(-)

diff --git a/fs/dlm/plock.c b/fs/dlm/plock.c
index 78878c5781ca..eba87ff3177b 100644
--- a/fs/dlm/plock.c
+++ b/fs/dlm/plock.c
@@ -116,7 +116,7 @@ int dlm_posix_lock(dlm_lockspace_t *lockspace, u64 number, struct file *file,
 	if (xop->callback == NULL)
 		wait_event(recv_wq, (op->done != 0));
 	else {
-		rv = -EINPROGRESS;
+		rv = FILE_LOCK_DEFERRED;
 		goto out;
 	}
 
diff --git a/fs/lockd/svclock.c b/fs/lockd/svclock.c
index 821b9acdfb66..cf0d5c2c318d 100644
--- a/fs/lockd/svclock.c
+++ b/fs/lockd/svclock.c
@@ -418,8 +418,8 @@ nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file,
 			goto out;
 		case -EAGAIN:
 			ret = nlm_lck_denied;
-			break;
-		case -EINPROGRESS:
+			goto out;
+		case FILE_LOCK_DEFERRED:
 			if (wait)
 				break;
 			/* Filesystem lock operation is in progress
@@ -434,10 +434,6 @@ nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file,
 			goto out;
 	}
 
-	ret = nlm_lck_denied;
-	if (!wait)
-		goto out;
-
 	ret = nlm_lck_blocked;
 
 	/* Append to list of blocked */
@@ -507,7 +503,7 @@ nlmsvc_testlock(struct svc_rqst *rqstp, struct nlm_file *file,
 	}
 
 	error = vfs_test_lock(file->f_file, &lock->fl);
-	if (error == -EINPROGRESS) {
+	if (error == FILE_LOCK_DEFERRED) {
 		ret = nlmsvc_defer_lock_rqst(rqstp, block);
 		goto out;
 	}
@@ -731,8 +727,7 @@ nlmsvc_grant_blocked(struct nlm_block *block)
 	switch (error) {
 	case 0:
 		break;
-	case -EAGAIN:
-	case -EINPROGRESS:
+	case FILE_LOCK_DEFERRED:
 		dprintk("lockd: lock still blocked error %d\n", error);
 		nlmsvc_insert_block(block, NLM_NEVER);
 		nlmsvc_release_block(block);
diff --git a/fs/locks.c b/fs/locks.c
index dce8c747371c..1ce57b4b362c 100644
--- a/fs/locks.c
+++ b/fs/locks.c
@@ -779,8 +779,10 @@ find_conflict:
 		if (!flock_locks_conflict(request, fl))
 			continue;
 		error = -EAGAIN;
-		if (request->fl_flags & FL_SLEEP)
-			locks_insert_block(fl, request);
+		if (!(request->fl_flags & FL_SLEEP))
+			goto out;
+		error = FILE_LOCK_DEFERRED;
+		locks_insert_block(fl, request);
 		goto out;
 	}
 	if (request->fl_flags & FL_ACCESS)
@@ -836,7 +838,7 @@ static int __posix_lock_file(struct inode *inode, struct file_lock *request, str
 			error = -EDEADLK;
 			if (posix_locks_deadlock(request, fl))
 				goto out;
-			error = -EAGAIN;
+			error = FILE_LOCK_DEFERRED;
 			locks_insert_block(fl, request);
 			goto out;
   		}
@@ -1035,7 +1037,7 @@ int posix_lock_file_wait(struct file *filp, struct file_lock *fl)
 	might_sleep ();
 	for (;;) {
 		error = posix_lock_file(filp, fl, NULL);
-		if ((error != -EAGAIN) || !(fl->fl_flags & FL_SLEEP))
+		if (error != FILE_LOCK_DEFERRED)
 			break;
 		error = wait_event_interruptible(fl->fl_wait, !fl->fl_next);
 		if (!error)
@@ -1107,9 +1109,7 @@ int locks_mandatory_area(int read_write, struct inode *inode,
 
 	for (;;) {
 		error = __posix_lock_file(inode, &fl, NULL);
-		if (error != -EAGAIN)
-			break;
-		if (!(fl.fl_flags & FL_SLEEP))
+		if (error != FILE_LOCK_DEFERRED)
 			break;
 		error = wait_event_interruptible(fl.fl_wait, !fl.fl_next);
 		if (!error) {
@@ -1531,7 +1531,7 @@ int flock_lock_file_wait(struct file *filp, struct file_lock *fl)
 	might_sleep();
 	for (;;) {
 		error = flock_lock_file(filp, fl);
-		if ((error != -EAGAIN) || !(fl->fl_flags & FL_SLEEP))
+		if (error != FILE_LOCK_DEFERRED)
 			break;
 		error = wait_event_interruptible(fl->fl_wait, !fl->fl_next);
 		if (!error)
@@ -1716,17 +1716,17 @@ out:
  * fl_grant is set. Callers expecting ->lock() to return asynchronously
  * will only use F_SETLK, not F_SETLKW; they will set FL_SLEEP if (and only if)
  * the request is for a blocking lock. When ->lock() does return asynchronously,
- * it must return -EINPROGRESS, and call ->fl_grant() when the lock
+ * it must return FILE_LOCK_DEFERRED, and call ->fl_grant() when the lock
  * request completes.
  * If the request is for non-blocking lock the file system should return
- * -EINPROGRESS then try to get the lock and call the callback routine with
- * the result. If the request timed out the callback routine will return a
+ * FILE_LOCK_DEFERRED then try to get the lock and call the callback routine
+ * with the result. If the request timed out the callback routine will return a
  * nonzero return code and the file system should release the lock. The file
  * system is also responsible to keep a corresponding posix lock when it
  * grants a lock so the VFS can find out which locks are locally held and do
  * the correct lock cleanup when required.
  * The underlying filesystem must not drop the kernel lock or call
- * ->fl_grant() before returning to the caller with a -EINPROGRESS
+ * ->fl_grant() before returning to the caller with a FILE_LOCK_DEFERRED
  * return code.
  */
 int vfs_lock_file(struct file *filp, unsigned int cmd, struct file_lock *fl, struct file_lock *conf)
@@ -1804,7 +1804,7 @@ again:
 	else {
 		for (;;) {
 			error = posix_lock_file(filp, file_lock, NULL);
-			if (error != -EAGAIN || cmd == F_SETLK)
+			if (error != FILE_LOCK_DEFERRED)
 				break;
 			error = wait_event_interruptible(file_lock->fl_wait,
 					!file_lock->fl_next);
@@ -1941,7 +1941,7 @@ again:
 	else {
 		for (;;) {
 			error = posix_lock_file(filp, file_lock, NULL);
-			if (error != -EAGAIN || cmd == F_SETLK64)
+			if (error != FILE_LOCK_DEFERRED)
 				break;
 			error = wait_event_interruptible(file_lock->fl_wait,
 					!file_lock->fl_next);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 4b86f806014c..49d8eb7a71be 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -885,6 +885,12 @@ static inline int file_check_writeable(struct file *filp)
 #define FL_CLOSE	64	/* unlock on close */
 #define FL_SLEEP	128	/* A blocking lock */
 
+/*
+ * Special return value from posix_lock_file() and vfs_lock_file() for
+ * asynchronous locking.
+ */
+#define FILE_LOCK_DEFERRED 1
+
 /*
  * The POSIX file lock owner is determined by
  * the "struct files_struct" in the thread group

From b648a6de00770cc325c22f43bdd4e935f6a2ee55 Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Fri, 25 Jul 2008 01:48:57 -0700
Subject: [PATCH 0928/1171] locks: cleanup code duplication

Extract common code into a function.

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Cc: "J. Bruce Fields" <bfields@fieldses.org>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: David Teigland <teigland@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/locks.c | 71 ++++++++++++++++++++++++------------------------------
 1 file changed, 31 insertions(+), 40 deletions(-)

diff --git a/fs/locks.c b/fs/locks.c
index 1ce57b4b362c..6222e4b580e2 100644
--- a/fs/locks.c
+++ b/fs/locks.c
@@ -1738,6 +1738,35 @@ int vfs_lock_file(struct file *filp, unsigned int cmd, struct file_lock *fl, str
 }
 EXPORT_SYMBOL_GPL(vfs_lock_file);
 
+static int do_lock_file_wait(struct file *filp, unsigned int cmd,
+			     struct file_lock *fl)
+{
+	int error;
+
+	error = security_file_lock(filp, fl->fl_type);
+	if (error)
+		return error;
+
+	if (filp->f_op && filp->f_op->lock != NULL)
+		error = filp->f_op->lock(filp, cmd, fl);
+	else {
+		for (;;) {
+			error = posix_lock_file(filp, fl, NULL);
+			if (error != FILE_LOCK_DEFERRED)
+				break;
+			error = wait_event_interruptible(fl->fl_wait,
+							 !fl->fl_next);
+			if (!error)
+				continue;
+
+			locks_delete_block(fl);
+			break;
+		}
+	}
+
+	return error;
+}
+
 /* Apply the lock described by l to an open file descriptor.
  * This implements both the F_SETLK and F_SETLKW commands of fcntl().
  */
@@ -1795,26 +1824,7 @@ again:
 		goto out;
 	}
 
-	error = security_file_lock(filp, file_lock->fl_type);
-	if (error)
-		goto out;
-
-	if (filp->f_op && filp->f_op->lock != NULL)
-		error = filp->f_op->lock(filp, cmd, file_lock);
-	else {
-		for (;;) {
-			error = posix_lock_file(filp, file_lock, NULL);
-			if (error != FILE_LOCK_DEFERRED)
-				break;
-			error = wait_event_interruptible(file_lock->fl_wait,
-					!file_lock->fl_next);
-			if (!error)
-				continue;
-
-			locks_delete_block(file_lock);
-			break;
-		}
-	}
+	error = do_lock_file_wait(filp, cmd, file_lock);
 
 	/*
 	 * Attempt to detect a close/fcntl race and recover by
@@ -1932,26 +1942,7 @@ again:
 		goto out;
 	}
 
-	error = security_file_lock(filp, file_lock->fl_type);
-	if (error)
-		goto out;
-
-	if (filp->f_op && filp->f_op->lock != NULL)
-		error = filp->f_op->lock(filp, cmd, file_lock);
-	else {
-		for (;;) {
-			error = posix_lock_file(filp, file_lock, NULL);
-			if (error != FILE_LOCK_DEFERRED)
-				break;
-			error = wait_event_interruptible(file_lock->fl_wait,
-					!file_lock->fl_next);
-			if (!error)
-				continue;
-
-			locks_delete_block(file_lock);
-			break;
-		}
-	}
+	error = do_lock_file_wait(filp, cmd, file_lock);
 
 	/*
 	 * Attempt to detect a close/fcntl race and recover by

From 764c76b371722e0cba5c24d91225f0f954b69d44 Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Fri, 25 Jul 2008 01:48:58 -0700
Subject: [PATCH 0929/1171] locks: allow ->lock() to return FILE_LOCK_DEFERRED

Allow filesystem's ->lock() method to call posix_lock_file() instead of
posix_lock_file_wait(), and return FILE_LOCK_DEFERRED.  This makes it
possible to implement a such a ->lock() function, that works with the lock
manager, which needs the call to be asynchronous.

Now the vfs_lock_file() helper can be used, so this is a cleanup as well.

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Cc: "J. Bruce Fields" <bfields@fieldses.org>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: David Teigland <teigland@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/locks.c | 23 +++++++++--------------
 1 file changed, 9 insertions(+), 14 deletions(-)

diff --git a/fs/locks.c b/fs/locks.c
index 6222e4b580e2..01490300f7cb 100644
--- a/fs/locks.c
+++ b/fs/locks.c
@@ -1747,21 +1747,16 @@ static int do_lock_file_wait(struct file *filp, unsigned int cmd,
 	if (error)
 		return error;
 
-	if (filp->f_op && filp->f_op->lock != NULL)
-		error = filp->f_op->lock(filp, cmd, fl);
-	else {
-		for (;;) {
-			error = posix_lock_file(filp, fl, NULL);
-			if (error != FILE_LOCK_DEFERRED)
-				break;
-			error = wait_event_interruptible(fl->fl_wait,
-							 !fl->fl_next);
-			if (!error)
-				continue;
-
-			locks_delete_block(fl);
+	for (;;) {
+		error = vfs_lock_file(filp, cmd, fl, NULL);
+		if (error != FILE_LOCK_DEFERRED)
 			break;
-		}
+		error = wait_event_interruptible(fl->fl_wait, !fl->fl_next);
+		if (!error)
+			continue;
+
+		locks_delete_block(fl);
+		break;
 	}
 
 	return error;

From 0de6256daafa3a97a269995e9b29f956bd419bbf Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Fri, 25 Jul 2008 01:48:59 -0700
Subject: [PATCH 0930/1171] fuse: prepare lookup for nfs export

Use d_splice_alias() instead of d_add() in fuse lookup code, to allow NFS
exporting.

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/fuse/dir.c | 24 ++++++++++++++----------
 1 file changed, 14 insertions(+), 10 deletions(-)

diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index 2060bf06b906..e5217b213b44 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -239,18 +239,20 @@ int fuse_valid_type(int m)
  * Add a directory inode to a dentry, ensuring that no other dentry
  * refers to this inode.  Called with fc->inst_mutex.
  */
-static int fuse_d_add_directory(struct dentry *entry, struct inode *inode)
+static struct dentry *fuse_d_add_directory(struct dentry *entry,
+					   struct inode *inode)
 {
 	struct dentry *alias = d_find_alias(inode);
-	if (alias) {
+	if (alias && !(alias->d_flags & DCACHE_DISCONNECTED)) {
 		/* This tries to shrink the subtree below alias */
 		fuse_invalidate_entry(alias);
 		dput(alias);
 		if (!list_empty(&inode->i_dentry))
-			return -EBUSY;
+			return ERR_PTR(-EBUSY);
+	} else {
+		dput(alias);
 	}
-	d_add(entry, inode);
-	return 0;
+	return d_splice_alias(inode, entry);
 }
 
 static struct dentry *fuse_lookup(struct inode *dir, struct dentry *entry,
@@ -259,6 +261,7 @@ static struct dentry *fuse_lookup(struct inode *dir, struct dentry *entry,
 	int err;
 	struct fuse_entry_out outarg;
 	struct inode *inode = NULL;
+	struct dentry *newent;
 	struct fuse_conn *fc = get_fuse_conn(dir);
 	struct fuse_req *req;
 	struct fuse_req *forget_req;
@@ -303,21 +306,22 @@ static struct dentry *fuse_lookup(struct inode *dir, struct dentry *entry,
 
 	if (inode && S_ISDIR(inode->i_mode)) {
 		mutex_lock(&fc->inst_mutex);
-		err = fuse_d_add_directory(entry, inode);
+		newent = fuse_d_add_directory(entry, inode);
 		mutex_unlock(&fc->inst_mutex);
-		if (err) {
+		if (IS_ERR(newent)) {
 			iput(inode);
-			return ERR_PTR(err);
+			return newent;
 		}
 	} else
-		d_add(entry, inode);
+		newent = d_splice_alias(inode, entry);
 
+	entry = newent ? newent : entry;
 	entry->d_op = &fuse_dentry_operations;
 	if (!err)
 		fuse_change_entry_timeout(entry, &outarg);
 	else
 		fuse_invalidate_entry_cache(entry);
-	return NULL;
+	return newent;
 }
 
 /*

From dbd561d236ff16f8143bc727d91758ddd190e8cb Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Fri, 25 Jul 2008 01:49:00 -0700
Subject: [PATCH 0931/1171] fuse: add export operations

Implement export_operations, to allow fuse filesystems to be exported to
NFS.  This feature has been in the out-of-tree fuse module, and is widely
used and tested.

It has not been originally merged into mainline, because doing the NFS
export in userspace was thought to be a cleaner and more efficient way of
doing it, than through the kernel.

While that is true, it would also have involved a lot of duplicated effort
at reimplementing NFS exporting (all the different versions of the
protocol).  This effort was unfortunately not undertaken by anyone, so we
are left with doing it the easy but less efficient way.

If this feature goes in, the out-of-tree fuse module can go away,
which would have several advantages:

  - not having to maintain two versions
  - less confusion for users
  - no bugs due to kernel API changes

Comment from hch:
 - Use the same fh_type values as XFS, since we use the same fh encoding.

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/fuse/dir.c    |   4 +-
 fs/fuse/fuse_i.h |   4 ++
 fs/fuse/inode.c  | 115 +++++++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 121 insertions(+), 2 deletions(-)

diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index e5217b213b44..be5450dd6387 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -97,7 +97,7 @@ void fuse_invalidate_attr(struct inode *inode)
  * timeout is unknown (unlink, rmdir, rename and in some cases
  * lookup)
  */
-static void fuse_invalidate_entry_cache(struct dentry *entry)
+void fuse_invalidate_entry_cache(struct dentry *entry)
 {
 	fuse_dentry_settime(entry, 0);
 }
@@ -225,7 +225,7 @@ static int invalid_nodeid(u64 nodeid)
 	return !nodeid || nodeid == FUSE_ROOT_ID;
 }
 
-static struct dentry_operations fuse_dentry_operations = {
+struct dentry_operations fuse_dentry_operations = {
 	.d_revalidate	= fuse_dentry_revalidate,
 };
 
diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h
index bae948657c4f..5d3146da64e6 100644
--- a/fs/fuse/fuse_i.h
+++ b/fs/fuse/fuse_i.h
@@ -464,6 +464,8 @@ static inline u64 get_node_id(struct inode *inode)
 /** Device operations */
 extern const struct file_operations fuse_dev_operations;
 
+extern struct dentry_operations fuse_dentry_operations;
+
 /**
  * Get a filled in inode
  */
@@ -604,6 +606,8 @@ void fuse_abort_conn(struct fuse_conn *fc);
  */
 void fuse_invalidate_attr(struct inode *inode);
 
+void fuse_invalidate_entry_cache(struct dentry *entry);
+
 /**
  * Acquire reference to fuse_conn
  */
diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index 3141690558c8..71fa76a48a31 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -18,6 +18,7 @@
 #include <linux/statfs.h>
 #include <linux/random.h>
 #include <linux/sched.h>
+#include <linux/exportfs.h>
 
 MODULE_AUTHOR("Miklos Szeredi <miklos@szeredi.hu>");
 MODULE_DESCRIPTION("Filesystem in Userspace");
@@ -552,6 +553,119 @@ static struct inode *get_root_inode(struct super_block *sb, unsigned mode)
 	return fuse_iget(sb, 1, 0, &attr, 0, 0);
 }
 
+struct fuse_inode_handle
+{
+	u64 nodeid;
+	u32 generation;
+};
+
+static struct dentry *fuse_get_dentry(struct super_block *sb,
+				      struct fuse_inode_handle *handle)
+{
+	struct inode *inode;
+	struct dentry *entry;
+	int err = -ESTALE;
+
+	if (handle->nodeid == 0)
+		goto out_err;
+
+	inode = ilookup5(sb, handle->nodeid, fuse_inode_eq, &handle->nodeid);
+	if (!inode)
+		goto out_err;
+	err = -ESTALE;
+	if (inode->i_generation != handle->generation)
+		goto out_iput;
+
+	entry = d_alloc_anon(inode);
+	err = -ENOMEM;
+	if (!entry)
+		goto out_iput;
+
+	if (get_node_id(inode) != FUSE_ROOT_ID) {
+		entry->d_op = &fuse_dentry_operations;
+		fuse_invalidate_entry_cache(entry);
+	}
+
+	return entry;
+
+ out_iput:
+	iput(inode);
+ out_err:
+	return ERR_PTR(err);
+}
+
+static int fuse_encode_fh(struct dentry *dentry, u32 *fh, int *max_len,
+			   int connectable)
+{
+	struct inode *inode = dentry->d_inode;
+	bool encode_parent = connectable && !S_ISDIR(inode->i_mode);
+	int len = encode_parent ? 6 : 3;
+	u64 nodeid;
+	u32 generation;
+
+	if (*max_len < len)
+		return  255;
+
+	nodeid = get_fuse_inode(inode)->nodeid;
+	generation = inode->i_generation;
+
+	fh[0] = (u32)(nodeid >> 32);
+	fh[1] = (u32)(nodeid & 0xffffffff);
+	fh[2] = generation;
+
+	if (encode_parent) {
+		struct inode *parent;
+
+		spin_lock(&dentry->d_lock);
+		parent = dentry->d_parent->d_inode;
+		nodeid = get_fuse_inode(parent)->nodeid;
+		generation = parent->i_generation;
+		spin_unlock(&dentry->d_lock);
+
+		fh[3] = (u32)(nodeid >> 32);
+		fh[4] = (u32)(nodeid & 0xffffffff);
+		fh[5] = generation;
+	}
+
+	*max_len = len;
+	return encode_parent ? 0x82 : 0x81;
+}
+
+static struct dentry *fuse_fh_to_dentry(struct super_block *sb,
+		struct fid *fid, int fh_len, int fh_type)
+{
+	struct fuse_inode_handle handle;
+
+	if ((fh_type != 0x81 && fh_type != 0x82) || fh_len < 3)
+		return NULL;
+
+	handle.nodeid = (u64) fid->raw[0] << 32;
+	handle.nodeid |= (u64) fid->raw[1];
+	handle.generation = fid->raw[2];
+	return fuse_get_dentry(sb, &handle);
+}
+
+static struct dentry *fuse_fh_to_parent(struct super_block *sb,
+		struct fid *fid, int fh_len, int fh_type)
+{
+	struct fuse_inode_handle parent;
+
+	if (fh_type != 0x82 || fh_len < 6)
+		return NULL;
+
+	parent.nodeid = (u64) fid->raw[3] << 32;
+	parent.nodeid |= (u64) fid->raw[4];
+	parent.generation = fid->raw[5];
+	return fuse_get_dentry(sb, &parent);
+}
+
+
+static const struct export_operations fuse_export_operations = {
+	.fh_to_dentry	= fuse_fh_to_dentry,
+	.fh_to_parent	= fuse_fh_to_parent,
+	.encode_fh	= fuse_encode_fh,
+};
+
 static const struct super_operations fuse_super_operations = {
 	.alloc_inode    = fuse_alloc_inode,
 	.destroy_inode  = fuse_destroy_inode,
@@ -652,6 +766,7 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent)
 	sb->s_magic = FUSE_SUPER_MAGIC;
 	sb->s_op = &fuse_super_operations;
 	sb->s_maxbytes = MAX_LFS_FILESIZE;
+	sb->s_export_op = &fuse_export_operations;
 
 	file = fget(d.fd);
 	if (!file)

From c180eebe1390c2076ead6a9bc95a02efb994edb7 Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Fri, 25 Jul 2008 01:49:01 -0700
Subject: [PATCH 0932/1171] fuse: add fuse_lookup_name() helper

Add a new helper function which sends a LOOKUP request with the supplied
name.  This will be used by the next patch to send special LOOKUP requests
with "." and ".." as the name.

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/fuse/dir.c | 141 +++++++++++++++++++++++++++++++-------------------
 1 file changed, 89 insertions(+), 52 deletions(-)

diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index be5450dd6387..51d0035ff07e 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -112,18 +112,16 @@ static void fuse_invalidate_entry(struct dentry *entry)
 	fuse_invalidate_entry_cache(entry);
 }
 
-static void fuse_lookup_init(struct fuse_req *req, struct inode *dir,
-			     struct dentry *entry,
+static void fuse_lookup_init(struct fuse_conn *fc, struct fuse_req *req,
+			     u64 nodeid, struct qstr *name,
 			     struct fuse_entry_out *outarg)
 {
-	struct fuse_conn *fc = get_fuse_conn(dir);
-
 	memset(outarg, 0, sizeof(struct fuse_entry_out));
 	req->in.h.opcode = FUSE_LOOKUP;
-	req->in.h.nodeid = get_node_id(dir);
+	req->in.h.nodeid = nodeid;
 	req->in.numargs = 1;
-	req->in.args[0].size = entry->d_name.len + 1;
-	req->in.args[0].value = entry->d_name.name;
+	req->in.args[0].size = name->len + 1;
+	req->in.args[0].value = name->name;
 	req->out.numargs = 1;
 	if (fc->minor < 9)
 		req->out.args[0].size = FUSE_COMPAT_ENTRY_OUT_SIZE;
@@ -189,7 +187,8 @@ static int fuse_dentry_revalidate(struct dentry *entry, struct nameidata *nd)
 		attr_version = fuse_get_attr_version(fc);
 
 		parent = dget_parent(entry);
-		fuse_lookup_init(req, parent->d_inode, entry, &outarg);
+		fuse_lookup_init(fc, req, get_node_id(parent->d_inode),
+				 &entry->d_name, &outarg);
 		request_send(fc, req);
 		dput(parent);
 		err = req->out.h.error;
@@ -255,73 +254,111 @@ static struct dentry *fuse_d_add_directory(struct dentry *entry,
 	return d_splice_alias(inode, entry);
 }
 
+int fuse_lookup_name(struct super_block *sb, u64 nodeid, struct qstr *name,
+		     struct fuse_entry_out *outarg, struct inode **inode)
+{
+	struct fuse_conn *fc = get_fuse_conn_super(sb);
+	struct fuse_req *req;
+	struct fuse_req *forget_req;
+	u64 attr_version;
+	int err;
+
+	*inode = NULL;
+	err = -ENAMETOOLONG;
+	if (name->len > FUSE_NAME_MAX)
+		goto out;
+
+	req = fuse_get_req(fc);
+	err = PTR_ERR(req);
+	if (IS_ERR(req))
+		goto out;
+
+	forget_req = fuse_get_req(fc);
+	err = PTR_ERR(forget_req);
+	if (IS_ERR(forget_req)) {
+		fuse_put_request(fc, req);
+		goto out;
+	}
+
+	attr_version = fuse_get_attr_version(fc);
+
+	fuse_lookup_init(fc, req, nodeid, name, outarg);
+	request_send(fc, req);
+	err = req->out.h.error;
+	fuse_put_request(fc, req);
+	/* Zero nodeid is same as -ENOENT, but with valid timeout */
+	if (err || !outarg->nodeid)
+		goto out_put_forget;
+
+	err = -EIO;
+	if (!outarg->nodeid)
+		goto out_put_forget;
+	if (!fuse_valid_type(outarg->attr.mode))
+		goto out_put_forget;
+
+	*inode = fuse_iget(sb, outarg->nodeid, outarg->generation,
+			   &outarg->attr, entry_attr_timeout(outarg),
+			   attr_version);
+	err = -ENOMEM;
+	if (!*inode) {
+		fuse_send_forget(fc, forget_req, outarg->nodeid, 1);
+		goto out;
+	}
+	err = 0;
+
+ out_put_forget:
+	fuse_put_request(fc, forget_req);
+ out:
+	return err;
+}
+
 static struct dentry *fuse_lookup(struct inode *dir, struct dentry *entry,
 				  struct nameidata *nd)
 {
 	int err;
 	struct fuse_entry_out outarg;
-	struct inode *inode = NULL;
+	struct inode *inode;
 	struct dentry *newent;
 	struct fuse_conn *fc = get_fuse_conn(dir);
-	struct fuse_req *req;
-	struct fuse_req *forget_req;
-	u64 attr_version;
+	bool outarg_valid = true;
 
-	if (entry->d_name.len > FUSE_NAME_MAX)
-		return ERR_PTR(-ENAMETOOLONG);
-
-	req = fuse_get_req(fc);
-	if (IS_ERR(req))
-		return ERR_CAST(req);
-
-	forget_req = fuse_get_req(fc);
-	if (IS_ERR(forget_req)) {
-		fuse_put_request(fc, req);
-		return ERR_CAST(forget_req);
+	err = fuse_lookup_name(dir->i_sb, get_node_id(dir), &entry->d_name,
+			       &outarg, &inode);
+	if (err == -ENOENT) {
+		outarg_valid = false;
+		err = 0;
 	}
+	if (err)
+		goto out_err;
 
-	attr_version = fuse_get_attr_version(fc);
-
-	fuse_lookup_init(req, dir, entry, &outarg);
-	request_send(fc, req);
-	err = req->out.h.error;
-	fuse_put_request(fc, req);
-	/* Zero nodeid is same as -ENOENT, but with valid timeout */
-	if (!err && outarg.nodeid &&
-	    (invalid_nodeid(outarg.nodeid) ||
-	     !fuse_valid_type(outarg.attr.mode)))
-		err = -EIO;
-	if (!err && outarg.nodeid) {
-		inode = fuse_iget(dir->i_sb, outarg.nodeid, outarg.generation,
-				  &outarg.attr, entry_attr_timeout(&outarg),
-				  attr_version);
-		if (!inode) {
-			fuse_send_forget(fc, forget_req, outarg.nodeid, 1);
-			return ERR_PTR(-ENOMEM);
-		}
-	}
-	fuse_put_request(fc, forget_req);
-	if (err && err != -ENOENT)
-		return ERR_PTR(err);
+	err = -EIO;
+	if (inode && get_node_id(inode) == FUSE_ROOT_ID)
+		goto out_iput;
 
 	if (inode && S_ISDIR(inode->i_mode)) {
 		mutex_lock(&fc->inst_mutex);
 		newent = fuse_d_add_directory(entry, inode);
 		mutex_unlock(&fc->inst_mutex);
-		if (IS_ERR(newent)) {
-			iput(inode);
-			return newent;
-		}
-	} else
+		err = PTR_ERR(newent);
+		if (IS_ERR(newent))
+			goto out_iput;
+	} else {
 		newent = d_splice_alias(inode, entry);
+	}
 
 	entry = newent ? newent : entry;
 	entry->d_op = &fuse_dentry_operations;
-	if (!err)
+	if (outarg_valid)
 		fuse_change_entry_timeout(entry, &outarg);
 	else
 		fuse_invalidate_entry_cache(entry);
+
 	return newent;
+
+ out_iput:
+	iput(inode);
+ out_err:
+	return ERR_PTR(err);
 }
 
 /*

From 33670fa296860283f04a7975b8c790f101e43a6e Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Fri, 25 Jul 2008 01:49:02 -0700
Subject: [PATCH 0933/1171] fuse: nfs export special lookups

Implement the get_parent export operation by sending a LOOKUP request with
".." as the name.

Implement looking up an inode by node ID after it has been evicted from
the cache.  This is done by seding a LOOKUP request with "." as the name
(for all file types, not just directories).

The filesystem can set the FUSE_EXPORT_SUPPORT flag in the INIT reply, to
indicate that it supports these special lookups.

Thanks to John Muir for the original implementation of this feature.

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Cc: "J. Bruce Fields" <bfields@fieldses.org>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: David Teigland <teigland@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/fuse/fuse_i.h     |  6 ++++
 fs/fuse/inode.c      | 66 ++++++++++++++++++++++++++++++++++++++++++--
 include/linux/fuse.h |  3 ++
 3 files changed, 72 insertions(+), 3 deletions(-)

diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h
index 5d3146da64e6..3a876076bdd1 100644
--- a/fs/fuse/fuse_i.h
+++ b/fs/fuse/fuse_i.h
@@ -363,6 +363,9 @@ struct fuse_conn {
 	/** Do not send separate SETATTR request before open(O_TRUNC)  */
 	unsigned atomic_o_trunc : 1;
 
+	/** Filesystem supports NFS exporting.  Only set in INIT */
+	unsigned export_support : 1;
+
 	/*
 	 * The following bitfields are only for optimization purposes
 	 * and hence races in setting them will not cause malfunction
@@ -473,6 +476,9 @@ struct inode *fuse_iget(struct super_block *sb, u64 nodeid,
 			int generation, struct fuse_attr *attr,
 			u64 attr_valid, u64 attr_version);
 
+int fuse_lookup_name(struct super_block *sb, u64 nodeid, struct qstr *name,
+		     struct fuse_entry_out *outarg, struct inode **inode);
+
 /**
  * Send FORGET command
  */
diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index 71fa76a48a31..7d2f7d6e22e2 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -562,6 +562,7 @@ struct fuse_inode_handle
 static struct dentry *fuse_get_dentry(struct super_block *sb,
 				      struct fuse_inode_handle *handle)
 {
+	struct fuse_conn *fc = get_fuse_conn_super(sb);
 	struct inode *inode;
 	struct dentry *entry;
 	int err = -ESTALE;
@@ -570,8 +571,27 @@ static struct dentry *fuse_get_dentry(struct super_block *sb,
 		goto out_err;
 
 	inode = ilookup5(sb, handle->nodeid, fuse_inode_eq, &handle->nodeid);
-	if (!inode)
-		goto out_err;
+	if (!inode) {
+		struct fuse_entry_out outarg;
+		struct qstr name;
+
+		if (!fc->export_support)
+			goto out_err;
+
+		name.len = 1;
+		name.name = ".";
+		err = fuse_lookup_name(sb, handle->nodeid, &name, &outarg,
+				       &inode);
+		if (err && err != -ENOENT)
+			goto out_err;
+		if (err || !inode) {
+			err = -ESTALE;
+			goto out_err;
+		}
+		err = -EIO;
+		if (get_node_id(inode) != handle->nodeid)
+			goto out_iput;
+	}
 	err = -ESTALE;
 	if (inode->i_generation != handle->generation)
 		goto out_iput;
@@ -659,11 +679,46 @@ static struct dentry *fuse_fh_to_parent(struct super_block *sb,
 	return fuse_get_dentry(sb, &parent);
 }
 
+static struct dentry *fuse_get_parent(struct dentry *child)
+{
+	struct inode *child_inode = child->d_inode;
+	struct fuse_conn *fc = get_fuse_conn(child_inode);
+	struct inode *inode;
+	struct dentry *parent;
+	struct fuse_entry_out outarg;
+	struct qstr name;
+	int err;
+
+	if (!fc->export_support)
+		return ERR_PTR(-ESTALE);
+
+	name.len = 2;
+	name.name = "..";
+	err = fuse_lookup_name(child_inode->i_sb, get_node_id(child_inode),
+			       &name, &outarg, &inode);
+	if (err && err != -ENOENT)
+		return ERR_PTR(err);
+	if (err || !inode)
+		return ERR_PTR(-ESTALE);
+
+	parent = d_alloc_anon(inode);
+	if (!parent) {
+		iput(inode);
+		return ERR_PTR(-ENOMEM);
+	}
+	if (get_node_id(inode) != FUSE_ROOT_ID) {
+		parent->d_op = &fuse_dentry_operations;
+		fuse_invalidate_entry_cache(parent);
+	}
+
+	return parent;
+}
 
 static const struct export_operations fuse_export_operations = {
 	.fh_to_dentry	= fuse_fh_to_dentry,
 	.fh_to_parent	= fuse_fh_to_parent,
 	.encode_fh	= fuse_encode_fh,
+	.get_parent	= fuse_get_parent,
 };
 
 static const struct super_operations fuse_super_operations = {
@@ -695,6 +750,11 @@ static void process_init_reply(struct fuse_conn *fc, struct fuse_req *req)
 				fc->no_lock = 1;
 			if (arg->flags & FUSE_ATOMIC_O_TRUNC)
 				fc->atomic_o_trunc = 1;
+			if (arg->minor >= 9) {
+				/* LOOKUP has dependency on proto version */
+				if (arg->flags & FUSE_EXPORT_SUPPORT)
+					fc->export_support = 1;
+			}
 			if (arg->flags & FUSE_BIG_WRITES)
 				fc->big_writes = 1;
 		} else {
@@ -721,7 +781,7 @@ static void fuse_send_init(struct fuse_conn *fc, struct fuse_req *req)
 	arg->minor = FUSE_KERNEL_MINOR_VERSION;
 	arg->max_readahead = fc->bdi.ra_pages * PAGE_CACHE_SIZE;
 	arg->flags |= FUSE_ASYNC_READ | FUSE_POSIX_LOCKS | FUSE_ATOMIC_O_TRUNC |
-		FUSE_BIG_WRITES;
+		FUSE_EXPORT_SUPPORT | FUSE_BIG_WRITES;
 	req->in.h.opcode = FUSE_INIT;
 	req->in.numargs = 1;
 	req->in.args[0].size = sizeof(*arg);
diff --git a/include/linux/fuse.h b/include/linux/fuse.h
index d48282197696..265635dc9908 100644
--- a/include/linux/fuse.h
+++ b/include/linux/fuse.h
@@ -104,11 +104,14 @@ struct fuse_file_lock {
 
 /**
  * INIT request/reply flags
+ *
+ * FUSE_EXPORT_SUPPORT: filesystem handles lookups of "." and ".."
  */
 #define FUSE_ASYNC_READ		(1 << 0)
 #define FUSE_POSIX_LOCKS	(1 << 1)
 #define FUSE_FILE_OPS		(1 << 2)
 #define FUSE_ATOMIC_O_TRUNC	(1 << 3)
+#define FUSE_EXPORT_SUPPORT	(1 << 4)
 #define FUSE_BIG_WRITES		(1 << 5)
 
 /**

From 48e90761b570ff57f58b726229d229729949c5bb Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Fri, 25 Jul 2008 01:49:02 -0700
Subject: [PATCH 0934/1171] fuse: lockd support

If fuse filesystem doesn't define it's own lock operations, then allow the
lock manager to work with fuse.

Adding lockd support for remote locking is also possible, but more rarely
used, so leave it till later.

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Cc: "J. Bruce Fields" <bfields@fieldses.org>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: David Teigland <teigland@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/fuse/file.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index 8092f0d9fd1f..67ff2c6a8f63 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -1341,6 +1341,11 @@ static int fuse_setlk(struct file *file, struct file_lock *fl, int flock)
 	pid_t pid = fl->fl_type != F_UNLCK ? current->tgid : 0;
 	int err;
 
+	if (fl->fl_lmops && fl->fl_lmops->fl_grant) {
+		/* NLM needs asynchronous locks, which we don't support yet */
+		return -ENOLCK;
+	}
+
 	/* Unlock on close is handled by the flush method */
 	if (fl->fl_flags & FL_CLOSE)
 		return 0;
@@ -1365,7 +1370,9 @@ static int fuse_file_lock(struct file *file, int cmd, struct file_lock *fl)
 	struct fuse_conn *fc = get_fuse_conn(inode);
 	int err;
 
-	if (cmd == F_GETLK) {
+	if (cmd == F_CANCELLK) {
+		err = 0;
+	} else if (cmd == F_GETLK) {
 		if (fc->no_lock) {
 			posix_test_lock(file, fl);
 			err = 0;
@@ -1373,7 +1380,7 @@ static int fuse_file_lock(struct file *file, int cmd, struct file_lock *fl)
 			err = fuse_getlk(file, fl);
 	} else {
 		if (fc->no_lock)
-			err = posix_lock_file_wait(file, fl);
+			err = posix_lock_file(file, fl, NULL);
 		else
 			err = fuse_setlk(file, fl, 0);
 	}

From 8f421c595a9145959d8aab09172743132abdffdb Mon Sep 17 00:00:00 2001
From: Arthur Jones <ajones@riverbed.com>
Date: Fri, 25 Jul 2008 01:49:04 -0700
Subject: [PATCH 0935/1171] edac: i5100 new intel chipset driver

Preliminary support for the Intel 5100 MCH.  CE and UE errors are reported
along with the current DIMM label information and other memory parameters.

Reasons why this is preliminary:

1) This chip has 2 independent memory controllers which, for best
   perforance, use interleaved accesses to the DDR2 memory.  This
   architecture does not map very well to the current edac data structures
   which depend on symmetric channel access to the interleaved data.
   Without core changes, the best I could do for now is to map both memory
   controllers to different csrows (first all ranks of controller 0, then
   all ranks of controller 1).  Someone much more familiar with the edac
   core than I will probably need to come up with a more general data
   structure to handle the interleaving and de-interleaving of the two
   memory controllers.

2) I have not yet tackled the de-interleaving of the rank/controller
   address space into the physical address space of the CPU.  There is
   nothing fundamentally missing, it is just ending up to be a lot of
   code, and I'd rather keep it separate for now, esp since it doesn't
   work yet...

3) The code depends on a particular i5100 chip select to DIMM mainboard
   chip select mapping.  This mapping seems obvious to me in order to
   support dual and single ranked memory, but it is not unique and DIMM
   labels could be wrong on other mainboards.  There is no way to query
   this mapping that I know of.

4) The code requires that the i5100 is in 32GB mode.  Only 4 ranks per
   controller, 2 ranks per DIMM are supported.  I do not have hardware
   (nor do I expect to have hardware anytime soon) for the 48GB (6 ranks
   per controller) mode.

5) The serial presence detect code should be broken out into a "real"
   i2c driver so that decode-dimms.pl can work.

Signed-off-by: Arthur Jones <ajones@riverbed.com>
Signed-off-by: Doug Thompson <dougthompson@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/edac/Kconfig      |   7 +
 drivers/edac/Makefile     |   1 +
 drivers/edac/i5100_edac.c | 827 ++++++++++++++++++++++++++++++++++++++
 include/linux/pci_ids.h   |   3 +
 4 files changed, 838 insertions(+)
 create mode 100644 drivers/edac/i5100_edac.c

diff --git a/drivers/edac/Kconfig b/drivers/edac/Kconfig
index 6e6c3c4aea6b..5a11e3cbcae2 100644
--- a/drivers/edac/Kconfig
+++ b/drivers/edac/Kconfig
@@ -123,6 +123,13 @@ config EDAC_I5000
 	  Support for error detection and correction the Intel
 	  Greekcreek/Blackford chipsets.
 
+config EDAC_I5100
+	tristate "Intel San Clemente MCH"
+	depends on EDAC_MM_EDAC && X86 && PCI
+	help
+	  Support for error detection and correction the Intel
+	  San Clemente MCH.
+
 config EDAC_MPC85XX
 	tristate "Freescale MPC85xx"
 	depends on EDAC_MM_EDAC && FSL_SOC && MPC85xx
diff --git a/drivers/edac/Makefile b/drivers/edac/Makefile
index 83807731d4a9..e5e9104b5520 100644
--- a/drivers/edac/Makefile
+++ b/drivers/edac/Makefile
@@ -19,6 +19,7 @@ endif
 
 obj-$(CONFIG_EDAC_AMD76X)		+= amd76x_edac.o
 obj-$(CONFIG_EDAC_I5000)		+= i5000_edac.o
+obj-$(CONFIG_EDAC_I5100)		+= i5100_edac.o
 obj-$(CONFIG_EDAC_E7XXX)		+= e7xxx_edac.o
 obj-$(CONFIG_EDAC_E752X)		+= e752x_edac.o
 obj-$(CONFIG_EDAC_I82443BXGX)		+= i82443bxgx_edac.o
diff --git a/drivers/edac/i5100_edac.c b/drivers/edac/i5100_edac.c
new file mode 100644
index 000000000000..43430bf70181
--- /dev/null
+++ b/drivers/edac/i5100_edac.c
@@ -0,0 +1,827 @@
+/*
+ * Intel 5100 Memory Controllers kernel module
+ *
+ * This file may be distributed under the terms of the
+ * GNU General Public License.
+ *
+ * This module is based on the following document:
+ *
+ * Intel 5100X Chipset Memory Controller Hub (MCH) - Datasheet
+ *      http://download.intel.com/design/chipsets/datashts/318378.pdf
+ *
+ */
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/pci.h>
+#include <linux/pci_ids.h>
+#include <linux/slab.h>
+#include <linux/edac.h>
+#include <linux/delay.h>
+#include <linux/mmzone.h>
+
+#include "edac_core.h"
+
+/* register addresses and bit field accessors... */
+
+/* device 16, func 1 */
+#define I5100_MS		0x44	/* Memory Status Register */
+#define I5100_SPDDATA		0x48	/* Serial Presence Detect Status Reg */
+#define		I5100_SPDDATA_RDO(a)	((a) >> 15 & 1)
+#define		I5100_SPDDATA_SBE(a)	((a) >> 13 & 1)
+#define		I5100_SPDDATA_BUSY(a)	((a) >> 12 & 1)
+#define		I5100_SPDDATA_DATA(a)	((a)       & ((1 << 8) - 1))
+#define I5100_SPDCMD		0x4c	/* Serial Presence Detect Command Reg */
+#define		I5100_SPDCMD_DTI(a)	(((a) & ((1 << 4) - 1)) << 28)
+#define		I5100_SPDCMD_CKOVRD(a)	(((a) & 1)              << 27)
+#define		I5100_SPDCMD_SA(a)	(((a) & ((1 << 3) - 1)) << 24)
+#define		I5100_SPDCMD_BA(a)	(((a) & ((1 << 8) - 1)) << 16)
+#define		I5100_SPDCMD_DATA(a)	(((a) & ((1 << 8) - 1)) <<  8)
+#define		I5100_SPDCMD_CMD(a)	((a) & 1)
+#define I5100_TOLM		0x6c	/* Top of Low Memory */
+#define		I5100_TOLM_TOLM(a)	((a) >> 12 & ((1 << 4) - 1))
+#define I5100_MIR0		0x80	/* Memory Interleave Range 0 */
+#define I5100_MIR1		0x84	/* Memory Interleave Range 1 */
+#define I5100_AMIR_0		0x8c	/* Adjusted Memory Interleave Range 0 */
+#define I5100_AMIR_1		0x90	/* Adjusted Memory Interleave Range 1 */
+#define		I5100_MIR_LIMIT(a)	((a) >> 4 & ((1 << 12) - 1))
+#define		I5100_MIR_WAY1(a)	((a) >> 1 & 1)
+#define		I5100_MIR_WAY0(a)	((a)      & 1)
+#define I5100_FERR_NF_MEM	0xa0	/* MC First Non Fatal Errors */
+#define		I5100_FERR_NF_MEM_CHAN_INDX(a)	((a) >> 28 & 1)
+#define		I5100_FERR_NF_MEM_SPD_MASK	(1 << 18)
+#define		I5100_FERR_NF_MEM_M16ERR_MASK	(1 << 16)
+#define		I5100_FERR_NF_MEM_M15ERR_MASK	(1 << 15)
+#define		I5100_FERR_NF_MEM_M14ERR_MASK	(1 << 14)
+#define		I5100_FERR_NF_MEM_
+#define		I5100_FERR_NF_MEM_
+#define		I5100_FERR_NF_MEM_ANY_MASK	\
+			(I5100_FERR_NF_MEM_M16ERR_MASK | \
+			I5100_FERR_NF_MEM_M15ERR_MASK | \
+			I5100_FERR_NF_MEM_M14ERR_MASK)
+#define		I5100_FERR_NF_MEM_ANY(a)  ((a) & I5100_FERR_NF_MEM_ANY_MASK)
+#define	I5100_NERR_NF_MEM	0xa4	/* MC Next Non-Fatal Errors */
+#define		I5100_NERR_NF_MEM_ANY(a)  I5100_FERR_NF_MEM_ANY(a)
+
+/* device 21 and 22, func 0 */
+#define I5100_MTR_0	0x154	/* Memory Technology Registers 0-3 */
+#define I5100_DMIR	0x15c	/* DIMM Interleave Range */
+#define		I5100_DMIR_LIMIT(a)	((a) >> 16 & ((1 << 11) - 1))
+#define		I5100_DMIR_RANK(a, i)	((a) >> (4 * i) & ((1 <<  2) - 1))
+#define I5100_MTR_4	0x1b0	/* Memory Technology Registers 4,5 */
+#define		I5100_MTR_PRESENT(a)	((a) >> 10 & 1)
+#define		I5100_MTR_ETHROTTLE(a)	((a) >>  9 & 1)
+#define		I5100_MTR_WIDTH(a)	((a) >>  8 & 1)
+#define		I5100_MTR_NUMBANK(a)	((a) >>  6 & 1)
+#define		I5100_MTR_NUMROW(a)	((a) >>  2 & ((1 << 2) - 1))
+#define		I5100_MTR_NUMCOL(a)	((a)       & ((1 << 2) - 1))
+#define	I5100_VALIDLOG	0x18c	/* Valid Log Markers */
+#define		I5100_VALIDLOG_REDMEMVALID(a)	((a) >> 2 & 1)
+#define		I5100_VALIDLOG_RECMEMVALID(a)	((a) >> 1 & 1)
+#define		I5100_VALIDLOG_NRECMEMVALID(a)	((a)      & 1)
+#define	I5100_NRECMEMA	0x190	/* Non-Recoverable Memory Error Log Reg A */
+#define		I5100_NRECMEMA_MERR(a)		((a) >> 15 & ((1 << 5) - 1))
+#define		I5100_NRECMEMA_BANK(a)		((a) >> 12 & ((1 << 3) - 1))
+#define		I5100_NRECMEMA_RANK(a)		((a) >>  8 & ((1 << 3) - 1))
+#define		I5100_NRECMEMA_DM_BUF_ID(a)	((a)       & ((1 << 8) - 1))
+#define	I5100_NRECMEMB	0x194	/* Non-Recoverable Memory Error Log Reg B */
+#define		I5100_NRECMEMB_CAS(a)		((a) >> 16 & ((1 << 13) - 1))
+#define		I5100_NRECMEMB_RAS(a)		((a)       & ((1 << 16) - 1))
+#define	I5100_REDMEMA	0x198	/* Recoverable Memory Data Error Log Reg A */
+#define		I5100_REDMEMA_SYNDROME(a)	(a)
+#define	I5100_REDMEMB	0x19c	/* Recoverable Memory Data Error Log Reg B */
+#define		I5100_REDMEMB_ECC_LOCATOR(a)	((a) & ((1 << 18) - 1))
+#define	I5100_RECMEMA	0x1a0	/* Recoverable Memory Error Log Reg A */
+#define		I5100_RECMEMA_MERR(a)		I5100_NRECMEMA_MERR(a)
+#define		I5100_RECMEMA_BANK(a)		I5100_NRECMEMA_BANK(a)
+#define		I5100_RECMEMA_RANK(a)		I5100_NRECMEMA_RANK(a)
+#define		I5100_RECMEMA_DM_BUF_ID(a)	I5100_NRECMEMA_DM_BUF_ID(a)
+#define	I5100_RECMEMB	0x1a4	/* Recoverable Memory Error Log Reg B */
+#define		I5100_RECMEMB_CAS(a)		I5100_NRECMEMB_CAS(a)
+#define		I5100_RECMEMB_RAS(a)		I5100_NRECMEMB_RAS(a)
+
+/* some generic limits */
+#define I5100_MAX_RANKS_PER_CTLR	6
+#define I5100_MAX_CTLRS			2
+#define I5100_MAX_RANKS_PER_DIMM	4
+#define I5100_DIMM_ADDR_LINES		(6 - 3)	/* 64 bits / 8 bits per byte */
+#define I5100_MAX_DIMM_SLOTS_PER_CTLR	4
+#define I5100_MAX_RANK_INTERLEAVE	4
+#define I5100_MAX_DMIRS			5
+
+struct i5100_priv {
+	/* ranks on each dimm -- 0 maps to not present -- obtained via SPD */
+	int dimm_numrank[I5100_MAX_CTLRS][I5100_MAX_DIMM_SLOTS_PER_CTLR];
+
+	/*
+	 * mainboard chip select map -- maps i5100 chip selects to
+	 * DIMM slot chip selects.  In the case of only 4 ranks per
+	 * controller, the mapping is fairly obvious but not unique.
+	 * we map -1 -> NC and assume both controllers use the same
+	 * map...
+	 *
+	 */
+	int dimm_csmap[I5100_MAX_DIMM_SLOTS_PER_CTLR][I5100_MAX_RANKS_PER_DIMM];
+
+	/* memory interleave range */
+	struct {
+		u64	 limit;
+		unsigned way[2];
+	} mir[I5100_MAX_CTLRS];
+
+	/* adjusted memory interleave range register */
+	unsigned amir[I5100_MAX_CTLRS];
+
+	/* dimm interleave range */
+	struct {
+		unsigned rank[I5100_MAX_RANK_INTERLEAVE];
+		u64	 limit;
+	} dmir[I5100_MAX_CTLRS][I5100_MAX_DMIRS];
+
+	/* memory technology registers... */
+	struct {
+		unsigned present;	/* 0 or 1 */
+		unsigned ethrottle;	/* 0 or 1 */
+		unsigned width;		/* 4 or 8 bits  */
+		unsigned numbank;	/* 2 or 3 lines */
+		unsigned numrow;	/* 13 .. 16 lines */
+		unsigned numcol;	/* 11 .. 12 lines */
+	} mtr[I5100_MAX_CTLRS][I5100_MAX_RANKS_PER_CTLR];
+
+	u64 tolm;		/* top of low memory in bytes */
+	unsigned ranksperctlr;	/* number of ranks per controller */
+
+	struct pci_dev *mc;	/* device 16 func 1 */
+	struct pci_dev *ch0mm;	/* device 21 func 0 */
+	struct pci_dev *ch1mm;	/* device 22 func 0 */
+};
+
+/* map a rank/ctlr to a slot number on the mainboard */
+static int i5100_rank_to_slot(const struct mem_ctl_info *mci,
+			      int ctlr, int rank)
+{
+	const struct i5100_priv *priv = mci->pvt_info;
+	int i;
+
+	for (i = 0; i < I5100_MAX_DIMM_SLOTS_PER_CTLR; i++) {
+		int j;
+		const int numrank = priv->dimm_numrank[ctlr][i];
+
+		for (j = 0; j < numrank; j++)
+			if (priv->dimm_csmap[i][j] == rank)
+				return i * 2 + ctlr;
+	}
+
+	return -1;
+}
+
+/*
+ * The processor bus memory addresses are broken into three
+ * pieces, whereas the controller addresses are contiguous.
+ *
+ * here we map from the controller address space to the
+ * processor address space:
+ *
+ *    Processor Address Space
+ * +-----------------------------+
+ * |                             |
+ * |  "high" memory addresses    |
+ * |                             |
+ * +-----------------------------+ <- 4GB on the i5100
+ * |                             |
+ * |  other non-memory addresses |
+ * |                             |
+ * +-----------------------------+ <- top of low memory
+ * |                             |
+ * | "low" memory addresses      |
+ * |                             |
+ * +-----------------------------+
+ */
+static unsigned long i5100_ctl_page_to_phys(struct mem_ctl_info *mci,
+					    unsigned long cntlr_addr)
+{
+	const struct i5100_priv *priv = mci->pvt_info;
+
+	if (cntlr_addr < priv->tolm)
+		return cntlr_addr;
+
+	return (1ULL << 32) + (cntlr_addr - priv->tolm);
+}
+
+static const char *i5100_err_msg(unsigned err)
+{
+	const char *merrs[] = {
+		"unknown", /* 0 */
+		"uncorrectable data ECC on replay", /* 1 */
+		"unknown", /* 2 */
+		"unknown", /* 3 */
+		"aliased uncorrectable demand data ECC", /* 4 */
+		"aliased uncorrectable spare-copy data ECC", /* 5 */
+		"aliased uncorrectable patrol data ECC", /* 6 */
+		"unknown", /* 7 */
+		"unknown", /* 8 */
+		"unknown", /* 9 */
+		"non-aliased uncorrectable demand data ECC", /* 10 */
+		"non-aliased uncorrectable spare-copy data ECC", /* 11 */
+		"non-aliased uncorrectable patrol data ECC", /* 12 */
+		"unknown", /* 13 */
+		"correctable demand data ECC", /* 14 */
+		"correctable spare-copy data ECC", /* 15 */
+		"correctable patrol data ECC", /* 16 */
+		"unknown", /* 17 */
+		"SPD protocol error", /* 18 */
+		"unknown", /* 19 */
+		"spare copy initiated", /* 20 */
+		"spare copy completed", /* 21 */
+	};
+	unsigned i;
+
+	for (i = 0; i < ARRAY_SIZE(merrs); i++)
+		if (1 << i & err)
+			return merrs[i];
+
+	return "none";
+}
+
+/* convert csrow index into a rank (per controller -- 0..5) */
+static int i5100_csrow_to_rank(const struct mem_ctl_info *mci, int csrow)
+{
+	const struct i5100_priv *priv = mci->pvt_info;
+
+	return csrow % priv->ranksperctlr;
+}
+
+/* convert csrow index into a controller (0..1) */
+static int i5100_csrow_to_cntlr(const struct mem_ctl_info *mci, int csrow)
+{
+	const struct i5100_priv *priv = mci->pvt_info;
+
+	return csrow / priv->ranksperctlr;
+}
+
+static unsigned i5100_rank_to_csrow(const struct mem_ctl_info *mci,
+				    int ctlr, int rank)
+{
+	const struct i5100_priv *priv = mci->pvt_info;
+
+	return ctlr * priv->ranksperctlr + rank;
+}
+
+static void i5100_handle_ce(struct mem_ctl_info *mci,
+			    int ctlr,
+			    unsigned bank,
+			    unsigned rank,
+			    unsigned long syndrome,
+			    unsigned cas,
+			    unsigned ras,
+			    const char *msg)
+{
+	const int csrow = i5100_rank_to_csrow(mci, ctlr, rank);
+
+	printk(KERN_ERR
+		"CE ctlr %d, bank %u, rank %u, syndrome 0x%lx, "
+		"cas %u, ras %u, csrow %u, label \"%s\": %s\n",
+		ctlr, bank, rank, syndrome, cas, ras,
+		csrow, mci->csrows[csrow].channels[0].label, msg);
+
+	mci->ce_count++;
+	mci->csrows[csrow].ce_count++;
+	mci->csrows[csrow].channels[0].ce_count++;
+}
+
+static void i5100_handle_ue(struct mem_ctl_info *mci,
+			    int ctlr,
+			    unsigned bank,
+			    unsigned rank,
+			    unsigned long syndrome,
+			    unsigned cas,
+			    unsigned ras,
+			    const char *msg)
+{
+	const int csrow = i5100_rank_to_csrow(mci, ctlr, rank);
+
+	printk(KERN_ERR
+		"UE ctlr %d, bank %u, rank %u, syndrome 0x%lx, "
+		"cas %u, ras %u, csrow %u, label \"%s\": %s\n",
+		ctlr, bank, rank, syndrome, cas, ras,
+		csrow, mci->csrows[csrow].channels[0].label, msg);
+
+	mci->ue_count++;
+	mci->csrows[csrow].ue_count++;
+}
+
+static void i5100_read_log(struct mem_ctl_info *mci, int ctlr,
+			   u32 ferr, u32 nerr)
+{
+	struct i5100_priv *priv = mci->pvt_info;
+	struct pci_dev *pdev = (ctlr) ? priv->ch1mm : priv->ch0mm;
+	u32 dw;
+	u32 dw2;
+	unsigned syndrome = 0;
+	unsigned ecc_loc = 0;
+	unsigned merr;
+	unsigned bank;
+	unsigned rank;
+	unsigned cas;
+	unsigned ras;
+
+	pci_read_config_dword(pdev, I5100_VALIDLOG, &dw);
+
+	if (I5100_VALIDLOG_REDMEMVALID(dw)) {
+		pci_read_config_dword(pdev, I5100_REDMEMA, &dw2);
+		syndrome = I5100_REDMEMA_SYNDROME(dw2);
+		pci_read_config_dword(pdev, I5100_REDMEMB, &dw2);
+		ecc_loc = I5100_REDMEMB_ECC_LOCATOR(dw2);
+	}
+
+	if (I5100_VALIDLOG_RECMEMVALID(dw)) {
+		const char *msg;
+
+		pci_read_config_dword(pdev, I5100_RECMEMA, &dw2);
+		merr = I5100_RECMEMA_MERR(dw2);
+		bank = I5100_RECMEMA_BANK(dw2);
+		rank = I5100_RECMEMA_RANK(dw2);
+
+		pci_read_config_dword(pdev, I5100_RECMEMB, &dw2);
+		cas = I5100_RECMEMB_CAS(dw2);
+		ras = I5100_RECMEMB_RAS(dw2);
+
+		/* FIXME:  not really sure if this is what merr is...
+		 */
+		if (!merr)
+			msg = i5100_err_msg(ferr);
+		else
+			msg = i5100_err_msg(nerr);
+
+		i5100_handle_ce(mci, ctlr, bank, rank, syndrome, cas, ras, msg);
+	}
+
+	if (I5100_VALIDLOG_NRECMEMVALID(dw)) {
+		const char *msg;
+
+		pci_read_config_dword(pdev, I5100_NRECMEMA, &dw2);
+		merr = I5100_NRECMEMA_MERR(dw2);
+		bank = I5100_NRECMEMA_BANK(dw2);
+		rank = I5100_NRECMEMA_RANK(dw2);
+
+		pci_read_config_dword(pdev, I5100_NRECMEMB, &dw2);
+		cas = I5100_NRECMEMB_CAS(dw2);
+		ras = I5100_NRECMEMB_RAS(dw2);
+
+		/* FIXME:  not really sure if this is what merr is...
+		 */
+		if (!merr)
+			msg = i5100_err_msg(ferr);
+		else
+			msg = i5100_err_msg(nerr);
+
+		i5100_handle_ue(mci, ctlr, bank, rank, syndrome, cas, ras, msg);
+	}
+
+	pci_write_config_dword(pdev, I5100_VALIDLOG, dw);
+}
+
+static void i5100_check_error(struct mem_ctl_info *mci)
+{
+	struct i5100_priv *priv = mci->pvt_info;
+	u32 dw;
+
+
+	pci_read_config_dword(priv->mc, I5100_FERR_NF_MEM, &dw);
+	if (I5100_FERR_NF_MEM_ANY(dw)) {
+		u32 dw2;
+
+		pci_read_config_dword(priv->mc, I5100_NERR_NF_MEM, &dw2);
+		if (dw2)
+			pci_write_config_dword(priv->mc, I5100_NERR_NF_MEM,
+					       dw2);
+		pci_write_config_dword(priv->mc, I5100_FERR_NF_MEM, dw);
+
+		i5100_read_log(mci, I5100_FERR_NF_MEM_CHAN_INDX(dw),
+			       I5100_FERR_NF_MEM_ANY(dw),
+			       I5100_NERR_NF_MEM_ANY(dw2));
+	}
+}
+
+static struct pci_dev *pci_get_device_func(unsigned vendor,
+					   unsigned device,
+					   unsigned func)
+{
+	struct pci_dev *ret = NULL;
+
+	while (1) {
+		ret = pci_get_device(vendor, device, ret);
+
+		if (!ret)
+			break;
+
+		if (PCI_FUNC(ret->devfn) == func)
+			break;
+	}
+
+	return ret;
+}
+
+static unsigned long __devinit i5100_npages(struct mem_ctl_info *mci,
+					    int csrow)
+{
+	struct i5100_priv *priv = mci->pvt_info;
+	const unsigned ctlr_rank = i5100_csrow_to_rank(mci, csrow);
+	const unsigned ctlr = i5100_csrow_to_cntlr(mci, csrow);
+	unsigned addr_lines;
+
+	/* dimm present? */
+	if (!priv->mtr[ctlr][ctlr_rank].present)
+		return 0ULL;
+
+	addr_lines =
+		I5100_DIMM_ADDR_LINES +
+		priv->mtr[ctlr][ctlr_rank].numcol +
+		priv->mtr[ctlr][ctlr_rank].numrow +
+		priv->mtr[ctlr][ctlr_rank].numbank;
+
+	return (unsigned long)
+		((unsigned long long) (1ULL << addr_lines) / PAGE_SIZE);
+}
+
+static void __devinit i5100_init_mtr(struct mem_ctl_info *mci)
+{
+	struct i5100_priv *priv = mci->pvt_info;
+	struct pci_dev *mms[2] = { priv->ch0mm, priv->ch1mm };
+	int i;
+
+	for (i = 0; i < I5100_MAX_CTLRS; i++) {
+		int j;
+		struct pci_dev *pdev = mms[i];
+
+		for (j = 0; j < I5100_MAX_RANKS_PER_CTLR; j++) {
+			const unsigned addr =
+				(j < 4) ? I5100_MTR_0 + j * 2 :
+					  I5100_MTR_4 + (j - 4) * 2;
+			u16 w;
+
+			pci_read_config_word(pdev, addr, &w);
+
+			priv->mtr[i][j].present = I5100_MTR_PRESENT(w);
+			priv->mtr[i][j].ethrottle = I5100_MTR_ETHROTTLE(w);
+			priv->mtr[i][j].width = 4 + 4 * I5100_MTR_WIDTH(w);
+			priv->mtr[i][j].numbank = 2 + I5100_MTR_NUMBANK(w);
+			priv->mtr[i][j].numrow = 13 + I5100_MTR_NUMROW(w);
+			priv->mtr[i][j].numcol = 10 + I5100_MTR_NUMCOL(w);
+		}
+	}
+}
+
+/*
+ * FIXME: make this into a real i2c adapter (so that dimm-decode
+ * will work)?
+ */
+static int i5100_read_spd_byte(const struct mem_ctl_info *mci,
+			       u8 ch, u8 slot, u8 addr, u8 *byte)
+{
+	struct i5100_priv *priv = mci->pvt_info;
+	u16 w;
+	u32 dw;
+	unsigned long et;
+
+	pci_read_config_word(priv->mc, I5100_SPDDATA, &w);
+	if (I5100_SPDDATA_BUSY(w))
+		return -1;
+
+	dw =	I5100_SPDCMD_DTI(0xa) |
+		I5100_SPDCMD_CKOVRD(1) |
+		I5100_SPDCMD_SA(ch * 4 + slot) |
+		I5100_SPDCMD_BA(addr) |
+		I5100_SPDCMD_DATA(0) |
+		I5100_SPDCMD_CMD(0);
+	pci_write_config_dword(priv->mc, I5100_SPDCMD, dw);
+
+	/* wait up to 100ms */
+	et = jiffies + HZ / 10;
+	udelay(100);
+	while (1) {
+		pci_read_config_word(priv->mc, I5100_SPDDATA, &w);
+		if (!I5100_SPDDATA_BUSY(w))
+			break;
+		udelay(100);
+	}
+
+	if (!I5100_SPDDATA_RDO(w) || I5100_SPDDATA_SBE(w))
+		return -1;
+
+	*byte = I5100_SPDDATA_DATA(w);
+
+	return 0;
+}
+
+/*
+ * fill dimm chip select map
+ *
+ * FIXME:
+ *   o only valid for 4 ranks per controller
+ *   o not the only way to may chip selects to dimm slots
+ *   o investigate if there is some way to obtain this map from the bios
+ */
+static void __devinit i5100_init_dimm_csmap(struct mem_ctl_info *mci)
+{
+	struct i5100_priv *priv = mci->pvt_info;
+	int i;
+
+	WARN_ON(priv->ranksperctlr != 4);
+
+	for (i = 0; i < I5100_MAX_DIMM_SLOTS_PER_CTLR; i++) {
+		int j;
+
+		for (j = 0; j < I5100_MAX_RANKS_PER_DIMM; j++)
+			priv->dimm_csmap[i][j] = -1; /* default NC */
+	}
+
+	/* only 2 chip selects per slot... */
+	priv->dimm_csmap[0][0] = 0;
+	priv->dimm_csmap[0][1] = 3;
+	priv->dimm_csmap[1][0] = 1;
+	priv->dimm_csmap[1][1] = 2;
+	priv->dimm_csmap[2][0] = 2;
+	priv->dimm_csmap[3][0] = 3;
+}
+
+static void __devinit i5100_init_dimm_layout(struct pci_dev *pdev,
+					     struct mem_ctl_info *mci)
+{
+	struct i5100_priv *priv = mci->pvt_info;
+	int i;
+
+	for (i = 0; i < I5100_MAX_CTLRS; i++) {
+		int j;
+
+		for (j = 0; j < I5100_MAX_DIMM_SLOTS_PER_CTLR; j++) {
+			u8 rank;
+
+			if (i5100_read_spd_byte(mci, i, j, 5, &rank) < 0)
+				priv->dimm_numrank[i][j] = 0;
+			else
+				priv->dimm_numrank[i][j] = (rank & 3) + 1;
+		}
+	}
+
+	i5100_init_dimm_csmap(mci);
+}
+
+static void __devinit i5100_init_interleaving(struct pci_dev *pdev,
+					      struct mem_ctl_info *mci)
+{
+	u16 w;
+	u32 dw;
+	struct i5100_priv *priv = mci->pvt_info;
+	struct pci_dev *mms[2] = { priv->ch0mm, priv->ch1mm };
+	int i;
+
+	pci_read_config_word(pdev, I5100_TOLM, &w);
+	priv->tolm = (u64) I5100_TOLM_TOLM(w) * 256 * 1024 * 1024;
+
+	pci_read_config_word(pdev, I5100_MIR0, &w);
+	priv->mir[0].limit = (u64) I5100_MIR_LIMIT(w) << 28;
+	priv->mir[0].way[1] = I5100_MIR_WAY1(w);
+	priv->mir[0].way[0] = I5100_MIR_WAY0(w);
+
+	pci_read_config_word(pdev, I5100_MIR1, &w);
+	priv->mir[1].limit = (u64) I5100_MIR_LIMIT(w) << 28;
+	priv->mir[1].way[1] = I5100_MIR_WAY1(w);
+	priv->mir[1].way[0] = I5100_MIR_WAY0(w);
+
+	pci_read_config_word(pdev, I5100_AMIR_0, &w);
+	priv->amir[0] = w;
+	pci_read_config_word(pdev, I5100_AMIR_1, &w);
+	priv->amir[1] = w;
+
+	for (i = 0; i < I5100_MAX_CTLRS; i++) {
+		int j;
+
+		for (j = 0; j < 5; j++) {
+			int k;
+
+			pci_read_config_dword(mms[i], I5100_DMIR + j * 4, &dw);
+
+			priv->dmir[i][j].limit =
+				(u64) I5100_DMIR_LIMIT(dw) << 28;
+			for (k = 0; k < I5100_MAX_RANKS_PER_DIMM; k++)
+				priv->dmir[i][j].rank[k] =
+					I5100_DMIR_RANK(dw, k);
+		}
+	}
+
+	i5100_init_mtr(mci);
+}
+
+static void __devinit i5100_init_csrows(struct mem_ctl_info *mci)
+{
+	int i;
+	unsigned long total_pages = 0UL;
+	struct i5100_priv *priv = mci->pvt_info;
+
+	for (i = 0; i < mci->nr_csrows; i++) {
+		const unsigned long npages = i5100_npages(mci, i);
+		const unsigned cntlr = i5100_csrow_to_cntlr(mci, i);
+		const unsigned rank = i5100_csrow_to_rank(mci, i);
+
+		if (!npages)
+			continue;
+
+		/*
+		 * FIXME: these two are totally bogus -- I don't see how to
+		 * map them correctly to this structure...
+		 */
+		mci->csrows[i].first_page = total_pages;
+		mci->csrows[i].last_page = total_pages + npages - 1;
+		mci->csrows[i].page_mask = 0UL;
+
+		mci->csrows[i].nr_pages = npages;
+		mci->csrows[i].grain = 32;
+		mci->csrows[i].csrow_idx = i;
+		mci->csrows[i].dtype =
+			(priv->mtr[cntlr][rank].width == 4) ? DEV_X4 : DEV_X8;
+		mci->csrows[i].ue_count = 0;
+		mci->csrows[i].ce_count = 0;
+		mci->csrows[i].mtype = MEM_RDDR2;
+		mci->csrows[i].edac_mode = EDAC_SECDED;
+		mci->csrows[i].mci = mci;
+		mci->csrows[i].nr_channels = 1;
+		mci->csrows[i].channels[0].chan_idx = 0;
+		mci->csrows[i].channels[0].ce_count = 0;
+		mci->csrows[i].channels[0].csrow = mci->csrows + i;
+		snprintf(mci->csrows[i].channels[0].label,
+			 sizeof(mci->csrows[i].channels[0].label),
+			 "DIMM%u", i5100_rank_to_slot(mci, cntlr, rank));
+
+		total_pages += npages;
+	}
+}
+
+static int __devinit i5100_init_one(struct pci_dev *pdev,
+				    const struct pci_device_id *id)
+{
+	int rc;
+	struct mem_ctl_info *mci;
+	struct i5100_priv *priv;
+	struct pci_dev *ch0mm, *ch1mm;
+	int ret = 0;
+	u32 dw;
+	int ranksperch;
+
+	if (PCI_FUNC(pdev->devfn) != 1)
+		return -ENODEV;
+
+	rc = pci_enable_device(pdev);
+	if (rc < 0) {
+		ret = rc;
+		goto bail;
+	}
+
+	/* figure out how many ranks, from strapped state of 48GB_Mode input */
+	pci_read_config_dword(pdev, I5100_MS, &dw);
+	ranksperch = !!(dw & (1 << 8)) * 2 + 4;
+
+	if (ranksperch != 4) {
+		/* FIXME: get 6 ranks / controller to work - need hw... */
+		printk(KERN_INFO "i5100_edac: unsupported configuration.\n");
+		ret = -ENODEV;
+		goto bail;
+	}
+
+	/* device 21, func 0, Channel 0 Memory Map, Error Flag/Mask, etc... */
+	ch0mm = pci_get_device_func(PCI_VENDOR_ID_INTEL,
+				    PCI_DEVICE_ID_INTEL_5100_21, 0);
+	if (!ch0mm)
+		return -ENODEV;
+
+	rc = pci_enable_device(ch0mm);
+	if (rc < 0) {
+		ret = rc;
+		goto bail_ch0;
+	}
+
+	/* device 22, func 0, Channel 1 Memory Map, Error Flag/Mask, etc... */
+	ch1mm = pci_get_device_func(PCI_VENDOR_ID_INTEL,
+				    PCI_DEVICE_ID_INTEL_5100_22, 0);
+	if (!ch1mm) {
+		ret = -ENODEV;
+		goto bail_ch0;
+	}
+
+	rc = pci_enable_device(ch1mm);
+	if (rc < 0) {
+		ret = rc;
+		goto bail_ch1;
+	}
+
+	mci = edac_mc_alloc(sizeof(*priv), ranksperch * 2, 1, 0);
+	if (!mci) {
+		ret = -ENOMEM;
+		goto bail_ch1;
+	}
+
+	mci->dev = &pdev->dev;
+
+	priv = mci->pvt_info;
+	priv->ranksperctlr = ranksperch;
+	priv->mc = pdev;
+	priv->ch0mm = ch0mm;
+	priv->ch1mm = ch1mm;
+
+	i5100_init_dimm_layout(pdev, mci);
+	i5100_init_interleaving(pdev, mci);
+
+	mci->mtype_cap = MEM_FLAG_FB_DDR2;
+	mci->edac_ctl_cap = EDAC_FLAG_SECDED;
+	mci->edac_cap = EDAC_FLAG_SECDED;
+	mci->mod_name = "i5100_edac.c";
+	mci->mod_ver = "not versioned";
+	mci->ctl_name = "i5100";
+	mci->dev_name = pci_name(pdev);
+	mci->ctl_page_to_phys = i5100_ctl_page_to_phys;
+
+	mci->edac_check = i5100_check_error;
+
+	i5100_init_csrows(mci);
+
+	/* this strange construction seems to be in every driver, dunno why */
+	switch (edac_op_state) {
+	case EDAC_OPSTATE_POLL:
+	case EDAC_OPSTATE_NMI:
+		break;
+	default:
+		edac_op_state = EDAC_OPSTATE_POLL;
+		break;
+	}
+
+	if (edac_mc_add_mc(mci)) {
+		ret = -ENODEV;
+		goto bail_mc;
+	}
+
+	goto bail;
+
+bail_mc:
+	edac_mc_free(mci);
+
+bail_ch1:
+	pci_dev_put(ch1mm);
+
+bail_ch0:
+	pci_dev_put(ch0mm);
+
+bail:
+	return ret;
+}
+
+static void __devexit i5100_remove_one(struct pci_dev *pdev)
+{
+	struct mem_ctl_info *mci;
+	struct i5100_priv *priv;
+
+	mci = edac_mc_del_mc(&pdev->dev);
+
+	if (!mci)
+		return;
+
+	priv = mci->pvt_info;
+	pci_dev_put(priv->ch0mm);
+	pci_dev_put(priv->ch1mm);
+
+	edac_mc_free(mci);
+}
+
+static const struct pci_device_id i5100_pci_tbl[] __devinitdata = {
+	/* Device 16, Function 0, Channel 0 Memory Map, Error Flag/Mask, ... */
+	{ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_5100_16) },
+	{ 0, }
+};
+MODULE_DEVICE_TABLE(pci, i5100_pci_tbl);
+
+static struct pci_driver i5100_driver = {
+	.name = KBUILD_BASENAME,
+	.probe = i5100_init_one,
+	.remove = __devexit_p(i5100_remove_one),
+	.id_table = i5100_pci_tbl,
+};
+
+static int __init i5100_init(void)
+{
+	int pci_rc;
+
+	pci_rc = pci_register_driver(&i5100_driver);
+
+	return (pci_rc < 0) ? pci_rc : 0;
+}
+
+static void __exit i5100_exit(void)
+{
+	pci_unregister_driver(&i5100_driver);
+}
+
+module_init(i5100_init);
+module_exit(i5100_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR
+    ("Arthur Jones <ajones@riverbed.com>");
+MODULE_DESCRIPTION("MC Driver for Intel I5100 memory controllers");
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index 119ae7b8f028..c3b1761aba26 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -2400,6 +2400,9 @@
 #define PCI_DEVICE_ID_INTEL_ICH10_4	0x3a30
 #define PCI_DEVICE_ID_INTEL_ICH10_5	0x3a60
 #define PCI_DEVICE_ID_INTEL_IOAT_SNB	0x402f
+#define PCI_DEVICE_ID_INTEL_5100_16	0x65f0
+#define PCI_DEVICE_ID_INTEL_5100_21	0x65f5
+#define PCI_DEVICE_ID_INTEL_5100_22	0x65f6
 #define PCI_DEVICE_ID_INTEL_5400_ERR	0x4030
 #define PCI_DEVICE_ID_INTEL_5400_FBD0	0x4035
 #define PCI_DEVICE_ID_INTEL_5400_FBD1	0x4036

From f7952ffcffa88c9a3fa92c26081f4ec9143c680f Mon Sep 17 00:00:00 2001
From: Arthur Jones <ajones@riverbed.com>
Date: Fri, 25 Jul 2008 01:49:05 -0700
Subject: [PATCH 0936/1171] edac: i5100 fix missing bits

The error mask we use to trigger ECC notifications is missing many bits of
interest.  We add these bits here so that all possible ECC errors can be
reported.

Signed-off-by: Arthur Jones <ajones@riverbed.com>
Signed-off-by: Doug Thompson <dougthompson@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/edac/i5100_edac.c | 18 +++++++++++++++---
 1 file changed, 15 insertions(+), 3 deletions(-)

diff --git a/drivers/edac/i5100_edac.c b/drivers/edac/i5100_edac.c
index 43430bf70181..a8767a6c1481 100644
--- a/drivers/edac/i5100_edac.c
+++ b/drivers/edac/i5100_edac.c
@@ -52,12 +52,24 @@
 #define		I5100_FERR_NF_MEM_M16ERR_MASK	(1 << 16)
 #define		I5100_FERR_NF_MEM_M15ERR_MASK	(1 << 15)
 #define		I5100_FERR_NF_MEM_M14ERR_MASK	(1 << 14)
-#define		I5100_FERR_NF_MEM_
-#define		I5100_FERR_NF_MEM_
+#define		I5100_FERR_NF_MEM_M12ERR_MASK	(1 << 12)
+#define		I5100_FERR_NF_MEM_M11ERR_MASK	(1 << 11)
+#define		I5100_FERR_NF_MEM_M10ERR_MASK	(1 << 10)
+#define		I5100_FERR_NF_MEM_M6ERR_MASK	(1 << 6)
+#define		I5100_FERR_NF_MEM_M5ERR_MASK	(1 << 5)
+#define		I5100_FERR_NF_MEM_M4ERR_MASK	(1 << 4)
+#define		I5100_FERR_NF_MEM_M1ERR_MASK	1
 #define		I5100_FERR_NF_MEM_ANY_MASK	\
 			(I5100_FERR_NF_MEM_M16ERR_MASK | \
 			I5100_FERR_NF_MEM_M15ERR_MASK | \
-			I5100_FERR_NF_MEM_M14ERR_MASK)
+			I5100_FERR_NF_MEM_M14ERR_MASK | \
+			I5100_FERR_NF_MEM_M12ERR_MASK | \
+			I5100_FERR_NF_MEM_M11ERR_MASK | \
+			I5100_FERR_NF_MEM_M10ERR_MASK | \
+			I5100_FERR_NF_MEM_M6ERR_MASK | \
+			I5100_FERR_NF_MEM_M5ERR_MASK | \
+			I5100_FERR_NF_MEM_M4ERR_MASK | \
+			I5100_FERR_NF_MEM_M1ERR_MASK)
 #define		I5100_FERR_NF_MEM_ANY(a)  ((a) & I5100_FERR_NF_MEM_ANY_MASK)
 #define	I5100_NERR_NF_MEM	0xa4	/* MC Next Non-Fatal Errors */
 #define		I5100_NERR_NF_MEM_ANY(a)  I5100_FERR_NF_MEM_ANY(a)

From 43920a598f9358a12eb59eeddc4cd950f03aea8c Mon Sep 17 00:00:00 2001
From: Arthur Jones <ajones@riverbed.com>
Date: Fri, 25 Jul 2008 01:49:06 -0700
Subject: [PATCH 0937/1171] edac: i5100 fix enable ecc hardware

It is possible that the BIOS did not enable ECC at boot time.  We check
for that case and fail to load if it is true.

Signed-off-by: Arthur Jones <ajones@riverbed.com>
Signed-off-by: Doug Thompson <dougthompson@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/edac/i5100_edac.c | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/edac/i5100_edac.c b/drivers/edac/i5100_edac.c
index a8767a6c1481..509eec860c37 100644
--- a/drivers/edac/i5100_edac.c
+++ b/drivers/edac/i5100_edac.c
@@ -24,6 +24,8 @@
 /* register addresses and bit field accessors... */
 
 /* device 16, func 1 */
+#define I5100_MC		0x40	/* Memory Control Register */
+#define		I5100_MC_ERRDETEN(a)	((a) >> 5 & 1)
 #define I5100_MS		0x44	/* Memory Status Register */
 #define I5100_SPDDATA		0x48	/* Serial Presence Detect Status Reg */
 #define		I5100_SPDDATA_RDO(a)	((a) >> 15 & 1)
@@ -688,6 +690,14 @@ static int __devinit i5100_init_one(struct pci_dev *pdev,
 		goto bail;
 	}
 
+	/* ECC enabled? */
+	pci_read_config_dword(pdev, I5100_MC, &dw);
+	if (!I5100_MC_ERRDETEN(dw)) {
+		printk(KERN_INFO "i5100_edac: ECC not enabled.\n");
+		ret = -ENODEV;
+		goto bail;
+	}
+
 	/* figure out how many ranks, from strapped state of 48GB_Mode input */
 	pci_read_config_dword(pdev, I5100_MS, &dw);
 	ranksperch = !!(dw & (1 << 8)) * 2 + 4;

From 178d5a742291976d13bff55fa2b130879d4510de Mon Sep 17 00:00:00 2001
From: Arthur Jones <ajones@riverbed.com>
Date: Fri, 25 Jul 2008 01:49:06 -0700
Subject: [PATCH 0938/1171] edac: i5100 fix unmask ecc bits

Explicitly unmask ECC errors we are interested in reporting.

Signed-off-by: Arthur Jones <ajones@riverbed.com>
Signed-off-by: Doug Thompson <dougthompson@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/edac/i5100_edac.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/edac/i5100_edac.c b/drivers/edac/i5100_edac.c
index 509eec860c37..d85e7992eb6a 100644
--- a/drivers/edac/i5100_edac.c
+++ b/drivers/edac/i5100_edac.c
@@ -75,6 +75,7 @@
 #define		I5100_FERR_NF_MEM_ANY(a)  ((a) & I5100_FERR_NF_MEM_ANY_MASK)
 #define	I5100_NERR_NF_MEM	0xa4	/* MC Next Non-Fatal Errors */
 #define		I5100_NERR_NF_MEM_ANY(a)  I5100_FERR_NF_MEM_ANY(a)
+#define I5100_EMASK_MEM		0xa8	/* MC Error Mask Register */
 
 /* device 21 and 22, func 0 */
 #define I5100_MTR_0	0x154	/* Memory Technology Registers 0-3 */
@@ -709,6 +710,11 @@ static int __devinit i5100_init_one(struct pci_dev *pdev,
 		goto bail;
 	}
 
+	/* enable error reporting... */
+	pci_read_config_dword(pdev, I5100_EMASK_MEM, &dw);
+	dw &= ~I5100_FERR_NF_MEM_ANY_MASK;
+	pci_write_config_dword(pdev, I5100_EMASK_MEM, dw);
+
 	/* device 21, func 0, Channel 0 Memory Map, Error Flag/Mask, etc... */
 	ch0mm = pci_get_device_func(PCI_VENDOR_ID_INTEL,
 				    PCI_DEVICE_ID_INTEL_5100_21, 0);

From b238e57723a6fb2c365fc35de5d7c48ccf9300cd Mon Sep 17 00:00:00 2001
From: Arthur Jones <ajones@riverbed.com>
Date: Fri, 25 Jul 2008 01:49:08 -0700
Subject: [PATCH 0939/1171] edac: i5100: cleanup

Some code cleanliness issues found by Andrew Morton (thanks!) which should
not affect functionality, but which should help make the code more
maintainable.

In particular, we now:

* convert all #define's w/ a parameter to static inlines
* use 1UL rather than 1ULL when calculating an unsigned long
* use pci_disable_device

The resulting code is tested and seems to work fine...

Signed-off-by: Arthur Jones <ajones@riverbed.com>
Cc: Doug Thompson <dougthompson@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/edac/i5100_edac.c | 396 +++++++++++++++++++++++++-------------
 1 file changed, 261 insertions(+), 135 deletions(-)

diff --git a/drivers/edac/i5100_edac.c b/drivers/edac/i5100_edac.c
index d85e7992eb6a..22db05a67bfb 100644
--- a/drivers/edac/i5100_edac.c
+++ b/drivers/edac/i5100_edac.c
@@ -21,36 +21,19 @@
 
 #include "edac_core.h"
 
-/* register addresses and bit field accessors... */
+/* register addresses */
 
 /* device 16, func 1 */
 #define I5100_MC		0x40	/* Memory Control Register */
-#define		I5100_MC_ERRDETEN(a)	((a) >> 5 & 1)
 #define I5100_MS		0x44	/* Memory Status Register */
 #define I5100_SPDDATA		0x48	/* Serial Presence Detect Status Reg */
-#define		I5100_SPDDATA_RDO(a)	((a) >> 15 & 1)
-#define		I5100_SPDDATA_SBE(a)	((a) >> 13 & 1)
-#define		I5100_SPDDATA_BUSY(a)	((a) >> 12 & 1)
-#define		I5100_SPDDATA_DATA(a)	((a)       & ((1 << 8) - 1))
 #define I5100_SPDCMD		0x4c	/* Serial Presence Detect Command Reg */
-#define		I5100_SPDCMD_DTI(a)	(((a) & ((1 << 4) - 1)) << 28)
-#define		I5100_SPDCMD_CKOVRD(a)	(((a) & 1)              << 27)
-#define		I5100_SPDCMD_SA(a)	(((a) & ((1 << 3) - 1)) << 24)
-#define		I5100_SPDCMD_BA(a)	(((a) & ((1 << 8) - 1)) << 16)
-#define		I5100_SPDCMD_DATA(a)	(((a) & ((1 << 8) - 1)) <<  8)
-#define		I5100_SPDCMD_CMD(a)	((a) & 1)
 #define I5100_TOLM		0x6c	/* Top of Low Memory */
-#define		I5100_TOLM_TOLM(a)	((a) >> 12 & ((1 << 4) - 1))
 #define I5100_MIR0		0x80	/* Memory Interleave Range 0 */
 #define I5100_MIR1		0x84	/* Memory Interleave Range 1 */
 #define I5100_AMIR_0		0x8c	/* Adjusted Memory Interleave Range 0 */
 #define I5100_AMIR_1		0x90	/* Adjusted Memory Interleave Range 1 */
-#define		I5100_MIR_LIMIT(a)	((a) >> 4 & ((1 << 12) - 1))
-#define		I5100_MIR_WAY1(a)	((a) >> 1 & 1)
-#define		I5100_MIR_WAY0(a)	((a)      & 1)
 #define I5100_FERR_NF_MEM	0xa0	/* MC First Non Fatal Errors */
-#define		I5100_FERR_NF_MEM_CHAN_INDX(a)	((a) >> 28 & 1)
-#define		I5100_FERR_NF_MEM_SPD_MASK	(1 << 18)
 #define		I5100_FERR_NF_MEM_M16ERR_MASK	(1 << 16)
 #define		I5100_FERR_NF_MEM_M15ERR_MASK	(1 << 15)
 #define		I5100_FERR_NF_MEM_M14ERR_MASK	(1 << 14)
@@ -72,47 +55,214 @@
 			I5100_FERR_NF_MEM_M5ERR_MASK | \
 			I5100_FERR_NF_MEM_M4ERR_MASK | \
 			I5100_FERR_NF_MEM_M1ERR_MASK)
-#define		I5100_FERR_NF_MEM_ANY(a)  ((a) & I5100_FERR_NF_MEM_ANY_MASK)
 #define	I5100_NERR_NF_MEM	0xa4	/* MC Next Non-Fatal Errors */
-#define		I5100_NERR_NF_MEM_ANY(a)  I5100_FERR_NF_MEM_ANY(a)
 #define I5100_EMASK_MEM		0xa8	/* MC Error Mask Register */
 
 /* device 21 and 22, func 0 */
 #define I5100_MTR_0	0x154	/* Memory Technology Registers 0-3 */
 #define I5100_DMIR	0x15c	/* DIMM Interleave Range */
-#define		I5100_DMIR_LIMIT(a)	((a) >> 16 & ((1 << 11) - 1))
-#define		I5100_DMIR_RANK(a, i)	((a) >> (4 * i) & ((1 <<  2) - 1))
-#define I5100_MTR_4	0x1b0	/* Memory Technology Registers 4,5 */
-#define		I5100_MTR_PRESENT(a)	((a) >> 10 & 1)
-#define		I5100_MTR_ETHROTTLE(a)	((a) >>  9 & 1)
-#define		I5100_MTR_WIDTH(a)	((a) >>  8 & 1)
-#define		I5100_MTR_NUMBANK(a)	((a) >>  6 & 1)
-#define		I5100_MTR_NUMROW(a)	((a) >>  2 & ((1 << 2) - 1))
-#define		I5100_MTR_NUMCOL(a)	((a)       & ((1 << 2) - 1))
 #define	I5100_VALIDLOG	0x18c	/* Valid Log Markers */
-#define		I5100_VALIDLOG_REDMEMVALID(a)	((a) >> 2 & 1)
-#define		I5100_VALIDLOG_RECMEMVALID(a)	((a) >> 1 & 1)
-#define		I5100_VALIDLOG_NRECMEMVALID(a)	((a)      & 1)
 #define	I5100_NRECMEMA	0x190	/* Non-Recoverable Memory Error Log Reg A */
-#define		I5100_NRECMEMA_MERR(a)		((a) >> 15 & ((1 << 5) - 1))
-#define		I5100_NRECMEMA_BANK(a)		((a) >> 12 & ((1 << 3) - 1))
-#define		I5100_NRECMEMA_RANK(a)		((a) >>  8 & ((1 << 3) - 1))
-#define		I5100_NRECMEMA_DM_BUF_ID(a)	((a)       & ((1 << 8) - 1))
 #define	I5100_NRECMEMB	0x194	/* Non-Recoverable Memory Error Log Reg B */
-#define		I5100_NRECMEMB_CAS(a)		((a) >> 16 & ((1 << 13) - 1))
-#define		I5100_NRECMEMB_RAS(a)		((a)       & ((1 << 16) - 1))
 #define	I5100_REDMEMA	0x198	/* Recoverable Memory Data Error Log Reg A */
-#define		I5100_REDMEMA_SYNDROME(a)	(a)
 #define	I5100_REDMEMB	0x19c	/* Recoverable Memory Data Error Log Reg B */
-#define		I5100_REDMEMB_ECC_LOCATOR(a)	((a) & ((1 << 18) - 1))
 #define	I5100_RECMEMA	0x1a0	/* Recoverable Memory Error Log Reg A */
-#define		I5100_RECMEMA_MERR(a)		I5100_NRECMEMA_MERR(a)
-#define		I5100_RECMEMA_BANK(a)		I5100_NRECMEMA_BANK(a)
-#define		I5100_RECMEMA_RANK(a)		I5100_NRECMEMA_RANK(a)
-#define		I5100_RECMEMA_DM_BUF_ID(a)	I5100_NRECMEMA_DM_BUF_ID(a)
 #define	I5100_RECMEMB	0x1a4	/* Recoverable Memory Error Log Reg B */
-#define		I5100_RECMEMB_CAS(a)		I5100_NRECMEMB_CAS(a)
-#define		I5100_RECMEMB_RAS(a)		I5100_NRECMEMB_RAS(a)
+#define I5100_MTR_4	0x1b0	/* Memory Technology Registers 4,5 */
+
+/* bit field accessors */
+
+static inline u32 i5100_mc_errdeten(u32 mc)
+{
+	return mc >> 5 & 1;
+}
+
+static inline u16 i5100_spddata_rdo(u16 a)
+{
+	return a >> 15 & 1;
+}
+
+static inline u16 i5100_spddata_sbe(u16 a)
+{
+	return a >> 13 & 1;
+}
+
+static inline u16 i5100_spddata_busy(u16 a)
+{
+	return a >> 12 & 1;
+}
+
+static inline u16 i5100_spddata_data(u16 a)
+{
+	return a & ((1 << 8) - 1);
+}
+
+static inline u32 i5100_spdcmd_create(u32 dti, u32 ckovrd, u32 sa, u32 ba,
+				      u32 data, u32 cmd)
+{
+	return	((dti & ((1 << 4) - 1))  << 28) |
+		((ckovrd & 1)            << 27) |
+		((sa & ((1 << 3) - 1))   << 24) |
+		((ba & ((1 << 8) - 1))   << 16) |
+		((data & ((1 << 8) - 1)) <<  8) |
+		(cmd & 1);
+}
+
+static inline u16 i5100_tolm_tolm(u16 a)
+{
+	return a >> 12 & ((1 << 4) - 1);
+}
+
+static inline u16 i5100_mir_limit(u16 a)
+{
+	return a >> 4 & ((1 << 12) - 1);
+}
+
+static inline u16 i5100_mir_way1(u16 a)
+{
+	return a >> 1 & 1;
+}
+
+static inline u16 i5100_mir_way0(u16 a)
+{
+	return a & 1;
+}
+
+static inline u32 i5100_ferr_nf_mem_chan_indx(u32 a)
+{
+	return a >> 28 & 1;
+}
+
+static inline u32 i5100_ferr_nf_mem_any(u32 a)
+{
+	return a & I5100_FERR_NF_MEM_ANY_MASK;
+}
+
+static inline u32 i5100_nerr_nf_mem_any(u32 a)
+{
+	return i5100_ferr_nf_mem_any(a);
+}
+
+static inline u32 i5100_dmir_limit(u32 a)
+{
+	return a >> 16 & ((1 << 11) - 1);
+}
+
+static inline u32 i5100_dmir_rank(u32 a, u32 i)
+{
+	return a >> (4 * i) & ((1 << 2) - 1);
+}
+
+static inline u16 i5100_mtr_present(u16 a)
+{
+	return a >> 10 & 1;
+}
+
+static inline u16 i5100_mtr_ethrottle(u16 a)
+{
+	return a >> 9 & 1;
+}
+
+static inline u16 i5100_mtr_width(u16 a)
+{
+	return a >> 8 & 1;
+}
+
+static inline u16 i5100_mtr_numbank(u16 a)
+{
+	return a >> 6 & 1;
+}
+
+static inline u16 i5100_mtr_numrow(u16 a)
+{
+	return a >> 2 & ((1 << 2) - 1);
+}
+
+static inline u16 i5100_mtr_numcol(u16 a)
+{
+	return a & ((1 << 2) - 1);
+}
+
+
+static inline u32 i5100_validlog_redmemvalid(u32 a)
+{
+	return a >> 2 & 1;
+}
+
+static inline u32 i5100_validlog_recmemvalid(u32 a)
+{
+	return a >> 1 & 1;
+}
+
+static inline u32 i5100_validlog_nrecmemvalid(u32 a)
+{
+	return a & 1;
+}
+
+static inline u32 i5100_nrecmema_merr(u32 a)
+{
+	return a >> 15 & ((1 << 5) - 1);
+}
+
+static inline u32 i5100_nrecmema_bank(u32 a)
+{
+	return a >> 12 & ((1 << 3) - 1);
+}
+
+static inline u32 i5100_nrecmema_rank(u32 a)
+{
+	return a >>  8 & ((1 << 3) - 1);
+}
+
+static inline u32 i5100_nrecmema_dm_buf_id(u32 a)
+{
+	return a & ((1 << 8) - 1);
+}
+
+static inline u32 i5100_nrecmemb_cas(u32 a)
+{
+	return a >> 16 & ((1 << 13) - 1);
+}
+
+static inline u32 i5100_nrecmemb_ras(u32 a)
+{
+	return a & ((1 << 16) - 1);
+}
+
+static inline u32 i5100_redmemb_ecc_locator(u32 a)
+{
+	return a & ((1 << 18) - 1);
+}
+
+static inline u32 i5100_recmema_merr(u32 a)
+{
+	return i5100_nrecmema_merr(a);
+}
+
+static inline u32 i5100_recmema_bank(u32 a)
+{
+	return i5100_nrecmema_bank(a);
+}
+
+static inline u32 i5100_recmema_rank(u32 a)
+{
+	return i5100_nrecmema_rank(a);
+}
+
+static inline u32 i5100_recmema_dm_buf_id(u32 a)
+{
+	return i5100_nrecmema_dm_buf_id(a);
+}
+
+static inline u32 i5100_recmemb_cas(u32 a)
+{
+	return i5100_nrecmemb_cas(a);
+}
+
+static inline u32 i5100_recmemb_ras(u32 a)
+{
+	return i5100_nrecmemb_ras(a);
+}
 
 /* some generic limits */
 #define I5100_MAX_RANKS_PER_CTLR	6
@@ -189,42 +339,9 @@ static int i5100_rank_to_slot(const struct mem_ctl_info *mci,
 	return -1;
 }
 
-/*
- * The processor bus memory addresses are broken into three
- * pieces, whereas the controller addresses are contiguous.
- *
- * here we map from the controller address space to the
- * processor address space:
- *
- *    Processor Address Space
- * +-----------------------------+
- * |                             |
- * |  "high" memory addresses    |
- * |                             |
- * +-----------------------------+ <- 4GB on the i5100
- * |                             |
- * |  other non-memory addresses |
- * |                             |
- * +-----------------------------+ <- top of low memory
- * |                             |
- * | "low" memory addresses      |
- * |                             |
- * +-----------------------------+
- */
-static unsigned long i5100_ctl_page_to_phys(struct mem_ctl_info *mci,
-					    unsigned long cntlr_addr)
-{
-	const struct i5100_priv *priv = mci->pvt_info;
-
-	if (cntlr_addr < priv->tolm)
-		return cntlr_addr;
-
-	return (1ULL << 32) + (cntlr_addr - priv->tolm);
-}
-
 static const char *i5100_err_msg(unsigned err)
 {
-	const char *merrs[] = {
+	static const char *merrs[] = {
 		"unknown", /* 0 */
 		"uncorrectable data ECC on replay", /* 1 */
 		"unknown", /* 2 */
@@ -341,24 +458,24 @@ static void i5100_read_log(struct mem_ctl_info *mci, int ctlr,
 
 	pci_read_config_dword(pdev, I5100_VALIDLOG, &dw);
 
-	if (I5100_VALIDLOG_REDMEMVALID(dw)) {
+	if (i5100_validlog_redmemvalid(dw)) {
 		pci_read_config_dword(pdev, I5100_REDMEMA, &dw2);
-		syndrome = I5100_REDMEMA_SYNDROME(dw2);
+		syndrome = dw2;
 		pci_read_config_dword(pdev, I5100_REDMEMB, &dw2);
-		ecc_loc = I5100_REDMEMB_ECC_LOCATOR(dw2);
+		ecc_loc = i5100_redmemb_ecc_locator(dw2);
 	}
 
-	if (I5100_VALIDLOG_RECMEMVALID(dw)) {
+	if (i5100_validlog_recmemvalid(dw)) {
 		const char *msg;
 
 		pci_read_config_dword(pdev, I5100_RECMEMA, &dw2);
-		merr = I5100_RECMEMA_MERR(dw2);
-		bank = I5100_RECMEMA_BANK(dw2);
-		rank = I5100_RECMEMA_RANK(dw2);
+		merr = i5100_recmema_merr(dw2);
+		bank = i5100_recmema_bank(dw2);
+		rank = i5100_recmema_rank(dw2);
 
 		pci_read_config_dword(pdev, I5100_RECMEMB, &dw2);
-		cas = I5100_RECMEMB_CAS(dw2);
-		ras = I5100_RECMEMB_RAS(dw2);
+		cas = i5100_recmemb_cas(dw2);
+		ras = i5100_recmemb_ras(dw2);
 
 		/* FIXME:  not really sure if this is what merr is...
 		 */
@@ -370,17 +487,17 @@ static void i5100_read_log(struct mem_ctl_info *mci, int ctlr,
 		i5100_handle_ce(mci, ctlr, bank, rank, syndrome, cas, ras, msg);
 	}
 
-	if (I5100_VALIDLOG_NRECMEMVALID(dw)) {
+	if (i5100_validlog_nrecmemvalid(dw)) {
 		const char *msg;
 
 		pci_read_config_dword(pdev, I5100_NRECMEMA, &dw2);
-		merr = I5100_NRECMEMA_MERR(dw2);
-		bank = I5100_NRECMEMA_BANK(dw2);
-		rank = I5100_NRECMEMA_RANK(dw2);
+		merr = i5100_nrecmema_merr(dw2);
+		bank = i5100_nrecmema_bank(dw2);
+		rank = i5100_nrecmema_rank(dw2);
 
 		pci_read_config_dword(pdev, I5100_NRECMEMB, &dw2);
-		cas = I5100_NRECMEMB_CAS(dw2);
-		ras = I5100_NRECMEMB_RAS(dw2);
+		cas = i5100_nrecmemb_cas(dw2);
+		ras = i5100_nrecmemb_ras(dw2);
 
 		/* FIXME:  not really sure if this is what merr is...
 		 */
@@ -402,7 +519,7 @@ static void i5100_check_error(struct mem_ctl_info *mci)
 
 
 	pci_read_config_dword(priv->mc, I5100_FERR_NF_MEM, &dw);
-	if (I5100_FERR_NF_MEM_ANY(dw)) {
+	if (i5100_ferr_nf_mem_any(dw)) {
 		u32 dw2;
 
 		pci_read_config_dword(priv->mc, I5100_NERR_NF_MEM, &dw2);
@@ -411,9 +528,9 @@ static void i5100_check_error(struct mem_ctl_info *mci)
 					       dw2);
 		pci_write_config_dword(priv->mc, I5100_FERR_NF_MEM, dw);
 
-		i5100_read_log(mci, I5100_FERR_NF_MEM_CHAN_INDX(dw),
-			       I5100_FERR_NF_MEM_ANY(dw),
-			       I5100_NERR_NF_MEM_ANY(dw2));
+		i5100_read_log(mci, i5100_ferr_nf_mem_chan_indx(dw),
+			       i5100_ferr_nf_mem_any(dw),
+			       i5100_nerr_nf_mem_any(dw2));
 	}
 }
 
@@ -476,12 +593,12 @@ static void __devinit i5100_init_mtr(struct mem_ctl_info *mci)
 
 			pci_read_config_word(pdev, addr, &w);
 
-			priv->mtr[i][j].present = I5100_MTR_PRESENT(w);
-			priv->mtr[i][j].ethrottle = I5100_MTR_ETHROTTLE(w);
-			priv->mtr[i][j].width = 4 + 4 * I5100_MTR_WIDTH(w);
-			priv->mtr[i][j].numbank = 2 + I5100_MTR_NUMBANK(w);
-			priv->mtr[i][j].numrow = 13 + I5100_MTR_NUMROW(w);
-			priv->mtr[i][j].numcol = 10 + I5100_MTR_NUMCOL(w);
+			priv->mtr[i][j].present = i5100_mtr_present(w);
+			priv->mtr[i][j].ethrottle = i5100_mtr_ethrottle(w);
+			priv->mtr[i][j].width = 4 + 4 * i5100_mtr_width(w);
+			priv->mtr[i][j].numbank = 2 + i5100_mtr_numbank(w);
+			priv->mtr[i][j].numrow = 13 + i5100_mtr_numrow(w);
+			priv->mtr[i][j].numcol = 10 + i5100_mtr_numcol(w);
 		}
 	}
 }
@@ -495,35 +612,30 @@ static int i5100_read_spd_byte(const struct mem_ctl_info *mci,
 {
 	struct i5100_priv *priv = mci->pvt_info;
 	u16 w;
-	u32 dw;
 	unsigned long et;
 
 	pci_read_config_word(priv->mc, I5100_SPDDATA, &w);
-	if (I5100_SPDDATA_BUSY(w))
+	if (i5100_spddata_busy(w))
 		return -1;
 
-	dw =	I5100_SPDCMD_DTI(0xa) |
-		I5100_SPDCMD_CKOVRD(1) |
-		I5100_SPDCMD_SA(ch * 4 + slot) |
-		I5100_SPDCMD_BA(addr) |
-		I5100_SPDCMD_DATA(0) |
-		I5100_SPDCMD_CMD(0);
-	pci_write_config_dword(priv->mc, I5100_SPDCMD, dw);
+	pci_write_config_dword(priv->mc, I5100_SPDCMD,
+			       i5100_spdcmd_create(0xa, 1, ch * 4 + slot, addr,
+						   0, 0));
 
 	/* wait up to 100ms */
 	et = jiffies + HZ / 10;
 	udelay(100);
 	while (1) {
 		pci_read_config_word(priv->mc, I5100_SPDDATA, &w);
-		if (!I5100_SPDDATA_BUSY(w))
+		if (!i5100_spddata_busy(w))
 			break;
 		udelay(100);
 	}
 
-	if (!I5100_SPDDATA_RDO(w) || I5100_SPDDATA_SBE(w))
+	if (!i5100_spddata_rdo(w) || i5100_spddata_sbe(w))
 		return -1;
 
-	*byte = I5100_SPDDATA_DATA(w);
+	*byte = i5100_spddata_data(w);
 
 	return 0;
 }
@@ -591,17 +703,17 @@ static void __devinit i5100_init_interleaving(struct pci_dev *pdev,
 	int i;
 
 	pci_read_config_word(pdev, I5100_TOLM, &w);
-	priv->tolm = (u64) I5100_TOLM_TOLM(w) * 256 * 1024 * 1024;
+	priv->tolm = (u64) i5100_tolm_tolm(w) * 256 * 1024 * 1024;
 
 	pci_read_config_word(pdev, I5100_MIR0, &w);
-	priv->mir[0].limit = (u64) I5100_MIR_LIMIT(w) << 28;
-	priv->mir[0].way[1] = I5100_MIR_WAY1(w);
-	priv->mir[0].way[0] = I5100_MIR_WAY0(w);
+	priv->mir[0].limit = (u64) i5100_mir_limit(w) << 28;
+	priv->mir[0].way[1] = i5100_mir_way1(w);
+	priv->mir[0].way[0] = i5100_mir_way0(w);
 
 	pci_read_config_word(pdev, I5100_MIR1, &w);
-	priv->mir[1].limit = (u64) I5100_MIR_LIMIT(w) << 28;
-	priv->mir[1].way[1] = I5100_MIR_WAY1(w);
-	priv->mir[1].way[0] = I5100_MIR_WAY0(w);
+	priv->mir[1].limit = (u64) i5100_mir_limit(w) << 28;
+	priv->mir[1].way[1] = i5100_mir_way1(w);
+	priv->mir[1].way[0] = i5100_mir_way0(w);
 
 	pci_read_config_word(pdev, I5100_AMIR_0, &w);
 	priv->amir[0] = w;
@@ -617,10 +729,10 @@ static void __devinit i5100_init_interleaving(struct pci_dev *pdev,
 			pci_read_config_dword(mms[i], I5100_DMIR + j * 4, &dw);
 
 			priv->dmir[i][j].limit =
-				(u64) I5100_DMIR_LIMIT(dw) << 28;
+				(u64) i5100_dmir_limit(dw) << 28;
 			for (k = 0; k < I5100_MAX_RANKS_PER_DIMM; k++)
 				priv->dmir[i][j].rank[k] =
-					I5100_DMIR_RANK(dw, k);
+					i5100_dmir_rank(dw, k);
 		}
 	}
 
@@ -693,10 +805,10 @@ static int __devinit i5100_init_one(struct pci_dev *pdev,
 
 	/* ECC enabled? */
 	pci_read_config_dword(pdev, I5100_MC, &dw);
-	if (!I5100_MC_ERRDETEN(dw)) {
+	if (!i5100_mc_errdeten(dw)) {
 		printk(KERN_INFO "i5100_edac: ECC not enabled.\n");
 		ret = -ENODEV;
-		goto bail;
+		goto bail_pdev;
 	}
 
 	/* figure out how many ranks, from strapped state of 48GB_Mode input */
@@ -707,7 +819,7 @@ static int __devinit i5100_init_one(struct pci_dev *pdev,
 		/* FIXME: get 6 ranks / controller to work - need hw... */
 		printk(KERN_INFO "i5100_edac: unsupported configuration.\n");
 		ret = -ENODEV;
-		goto bail;
+		goto bail_pdev;
 	}
 
 	/* enable error reporting... */
@@ -718,8 +830,10 @@ static int __devinit i5100_init_one(struct pci_dev *pdev,
 	/* device 21, func 0, Channel 0 Memory Map, Error Flag/Mask, etc... */
 	ch0mm = pci_get_device_func(PCI_VENDOR_ID_INTEL,
 				    PCI_DEVICE_ID_INTEL_5100_21, 0);
-	if (!ch0mm)
-		return -ENODEV;
+	if (!ch0mm) {
+		ret = -ENODEV;
+		goto bail_pdev;
+	}
 
 	rc = pci_enable_device(ch0mm);
 	if (rc < 0) {
@@ -732,7 +846,7 @@ static int __devinit i5100_init_one(struct pci_dev *pdev,
 				    PCI_DEVICE_ID_INTEL_5100_22, 0);
 	if (!ch1mm) {
 		ret = -ENODEV;
-		goto bail_ch0;
+		goto bail_disable_ch0;
 	}
 
 	rc = pci_enable_device(ch1mm);
@@ -744,7 +858,7 @@ static int __devinit i5100_init_one(struct pci_dev *pdev,
 	mci = edac_mc_alloc(sizeof(*priv), ranksperch * 2, 1, 0);
 	if (!mci) {
 		ret = -ENOMEM;
-		goto bail_ch1;
+		goto bail_disable_ch1;
 	}
 
 	mci->dev = &pdev->dev;
@@ -765,7 +879,7 @@ static int __devinit i5100_init_one(struct pci_dev *pdev,
 	mci->mod_ver = "not versioned";
 	mci->ctl_name = "i5100";
 	mci->dev_name = pci_name(pdev);
-	mci->ctl_page_to_phys = i5100_ctl_page_to_phys;
+	mci->ctl_page_to_phys = NULL;
 
 	mci->edac_check = i5100_check_error;
 
@@ -786,17 +900,26 @@ static int __devinit i5100_init_one(struct pci_dev *pdev,
 		goto bail_mc;
 	}
 
-	goto bail;
+	return ret;
 
 bail_mc:
 	edac_mc_free(mci);
 
+bail_disable_ch1:
+	pci_disable_device(ch1mm);
+
 bail_ch1:
 	pci_dev_put(ch1mm);
 
+bail_disable_ch0:
+	pci_disable_device(ch0mm);
+
 bail_ch0:
 	pci_dev_put(ch0mm);
 
+bail_pdev:
+	pci_disable_device(pdev);
+
 bail:
 	return ret;
 }
@@ -812,6 +935,9 @@ static void __devexit i5100_remove_one(struct pci_dev *pdev)
 		return;
 
 	priv = mci->pvt_info;
+	pci_disable_device(pdev);
+	pci_disable_device(priv->ch0mm);
+	pci_disable_device(priv->ch1mm);
 	pci_dev_put(priv->ch0mm);
 	pci_dev_put(priv->ch1mm);
 

From 14cc571bb1d072d3f4be2875ea520ab03e093471 Mon Sep 17 00:00:00 2001
From: Arthur Jones <ajones@riverbed.com>
Date: Fri, 25 Jul 2008 01:49:08 -0700
Subject: [PATCH 0940/1171] edac: core fix to use dynamic kobject

Static kobjects are not supported in linux kernel.  Convert the
edac_pci_top_main_kobj from static to dynamic.  This avoids the double
free of the edac_pci_top_main_kobj.name that we see on module reload of
the e752x edac driver (and probably others as well).

In addition Greg KH <greg@kroah.com> has pointed out that this code may be
cleaned up significantly.  I will look at that as a follow-on patch, for
now, I just want the minimum fix to get this double-free oops bug
squashed...

Many thanks to Greg KH for his patience in showing me what the
Documentation/kobject.txt already said (oops)...

Signed-off-by: Arthur Jones <ajones@riverbed.com>
Signed-off-by: Doug Thompson <dougthompson@xmission.com>
Acked-by: Greg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/edac/edac_pci_sysfs.c | 30 +++++++++++++++++++++---------
 1 file changed, 21 insertions(+), 9 deletions(-)

diff --git a/drivers/edac/edac_pci_sysfs.c b/drivers/edac/edac_pci_sysfs.c
index 2c1fa1bb6df2..5c153dccc95e 100644
--- a/drivers/edac/edac_pci_sysfs.c
+++ b/drivers/edac/edac_pci_sysfs.c
@@ -28,7 +28,7 @@ static int edac_pci_poll_msec = 1000;	/* one second workq period */
 static atomic_t pci_parity_count = ATOMIC_INIT(0);
 static atomic_t pci_nonparity_count = ATOMIC_INIT(0);
 
-static struct kobject edac_pci_top_main_kobj;
+static struct kobject *edac_pci_top_main_kobj;
 static atomic_t edac_pci_sysfs_refcount = ATOMIC_INIT(0);
 
 /* getter functions for the data variables */
@@ -83,7 +83,7 @@ static void edac_pci_instance_release(struct kobject *kobj)
 	pci = to_instance(kobj);
 
 	/* decrement reference count on top main kobj */
-	kobject_put(&edac_pci_top_main_kobj);
+	kobject_put(edac_pci_top_main_kobj);
 
 	kfree(pci);	/* Free the control struct */
 }
@@ -166,7 +166,7 @@ static int edac_pci_create_instance_kobj(struct edac_pci_ctl_info *pci, int idx)
 	 * track the number of PCI instances we have, and thus nest
 	 * properly on keeping the module loaded
 	 */
-	main_kobj = kobject_get(&edac_pci_top_main_kobj);
+	main_kobj = kobject_get(edac_pci_top_main_kobj);
 	if (!main_kobj) {
 		err = -ENODEV;
 		goto error_out;
@@ -174,11 +174,11 @@ static int edac_pci_create_instance_kobj(struct edac_pci_ctl_info *pci, int idx)
 
 	/* And now register this new kobject under the main kobj */
 	err = kobject_init_and_add(&pci->kobj, &ktype_pci_instance,
-				   &edac_pci_top_main_kobj, "pci%d", idx);
+				   edac_pci_top_main_kobj, "pci%d", idx);
 	if (err != 0) {
 		debugf2("%s() failed to register instance pci%d\n",
 			__func__, idx);
-		kobject_put(&edac_pci_top_main_kobj);
+		kobject_put(edac_pci_top_main_kobj);
 		goto error_out;
 	}
 
@@ -316,9 +316,10 @@ static struct edac_pci_dev_attribute *edac_pci_attr[] = {
  */
 static void edac_pci_release_main_kobj(struct kobject *kobj)
 {
-
 	debugf0("%s() here to module_put(THIS_MODULE)\n", __func__);
 
+	kfree(kobj);
+
 	/* last reference to top EDAC PCI kobject has been removed,
 	 * NOW release our ref count on the core module
 	 */
@@ -369,8 +370,16 @@ static int edac_pci_main_kobj_setup(void)
 		goto decrement_count_fail;
 	}
 
+	edac_pci_top_main_kobj = kzalloc(sizeof(struct kobject), GFP_KERNEL);
+	if (!edac_pci_top_main_kobj) {
+		debugf1("Failed to allocate\n");
+		err = -ENOMEM;
+		goto kzalloc_fail;
+	}
+
 	/* Instanstiate the pci object */
-	err = kobject_init_and_add(&edac_pci_top_main_kobj, &ktype_edac_pci_main_kobj,
+	err = kobject_init_and_add(edac_pci_top_main_kobj,
+				   &ktype_edac_pci_main_kobj,
 				   &edac_class->kset.kobj, "pci");
 	if (err) {
 		debugf1("Failed to register '.../edac/pci'\n");
@@ -381,13 +390,16 @@ static int edac_pci_main_kobj_setup(void)
 	 * for EDAC PCI, then edac_pci_main_kobj_teardown()
 	 * must be used, for resources to be cleaned up properly
 	 */
-	kobject_uevent(&edac_pci_top_main_kobj, KOBJ_ADD);
+	kobject_uevent(edac_pci_top_main_kobj, KOBJ_ADD);
 	debugf1("Registered '.../edac/pci' kobject\n");
 
 	return 0;
 
 	/* Error unwind statck */
 kobject_init_and_add_fail:
+	kfree(edac_pci_top_main_kobj);
+
+kzalloc_fail:
 	module_put(THIS_MODULE);
 
 decrement_count_fail:
@@ -414,7 +426,7 @@ static void edac_pci_main_kobj_teardown(void)
 	if (atomic_dec_return(&edac_pci_sysfs_refcount) == 0) {
 		debugf0("%s() called kobject_put on main kobj\n",
 			__func__);
-		kobject_put(&edac_pci_top_main_kobj);
+		kobject_put(edac_pci_top_main_kobj);
 	}
 }
 

From 096846e2b0ef39cb7c348f837f06984ef6ba8aa7 Mon Sep 17 00:00:00 2001
From: Arthur Jones <ajones@riverbed.com>
Date: Fri, 25 Jul 2008 01:49:09 -0700
Subject: [PATCH 0941/1171] edac: core fix workq timer

When updating the edac_mc_poll_msec module parameter from the sysfs
/sys/module/edac_core/parameters/edac_mc_poll_msec file, we don't update
the workq timers.  So that, if we move from a big poll time to a small
one, the small one won't take effect until the big one has timed out.

Here we provide a new module parameter set method to call out to the
update routine.  This brings the /sys/module/edac_core/parameters
functionality up to that provided by the /sys/drivers/system/edac/mc sysfs
module parameter files so that we can remove them or at least link to the
/sys/module files...

Signed-off-by: Arthur Jones <ajones@riverbed.com>
Signed-off-by: Doug Thompson <dougthompson@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/edac/edac_mc_sysfs.c | 22 +++++++++++++++++++++-
 1 file changed, 21 insertions(+), 1 deletion(-)

diff --git a/drivers/edac/edac_mc_sysfs.c b/drivers/edac/edac_mc_sysfs.c
index 021d18795145..7bb9c1532b90 100644
--- a/drivers/edac/edac_mc_sysfs.c
+++ b/drivers/edac/edac_mc_sysfs.c
@@ -44,6 +44,25 @@ int edac_mc_get_poll_msec(void)
 	return edac_mc_poll_msec;
 }
 
+static int edac_set_poll_msec(const char *val, struct kernel_param *kp)
+{
+	long l;
+	int ret;
+
+	if (!val)
+		return -EINVAL;
+
+	ret = strict_strtol(val, 0, &l);
+	if (ret == -EINVAL || ((int)l != l))
+		return -EINVAL;
+	*((int *)kp->arg) = l;
+
+	/* notify edac_mc engine to reset the poll period */
+	edac_mc_reset_delay_period(l);
+
+	return 0;
+}
+
 /* Parameter declarations for above */
 module_param(edac_mc_panic_on_ue, int, 0644);
 MODULE_PARM_DESC(edac_mc_panic_on_ue, "Panic on uncorrected error: 0=off 1=on");
@@ -53,7 +72,8 @@ MODULE_PARM_DESC(edac_mc_log_ue,
 module_param(edac_mc_log_ce, int, 0644);
 MODULE_PARM_DESC(edac_mc_log_ce,
 		 "Log correctable error to console: 0=off 1=on");
-module_param(edac_mc_poll_msec, int, 0644);
+module_param_call(edac_mc_poll_msec, edac_set_poll_msec, param_get_int,
+		  &edac_mc_poll_msec, 0644);
 MODULE_PARM_DESC(edac_mc_poll_msec, "Polling period in milliseconds");
 
 /*

From 327dafb1c61c9da7b95ac6cc7634a2340cc9509c Mon Sep 17 00:00:00 2001
From: Arthur Jones <ajones@riverbed.com>
Date: Fri, 25 Jul 2008 01:49:10 -0700
Subject: [PATCH 0942/1171] edac: core fix redundant sysfs controls to
 parameters

/sys/devices/system/edac/mc has a few files which are duplicated in
/sys/module/edac_core/parameters.  Now that all the functionality is
duplicated between these two locations, we remove the former kobject
attributes and update the documentation.

Signed-off-by: Arthur Jones <ajones@riverbed.com>
Signed-off-by: Doug Thompson <dougthompson@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/edac.txt       | 151 +++++++++++++++++------------------
 drivers/edac/edac_mc_sysfs.c | 117 +--------------------------
 2 files changed, 73 insertions(+), 195 deletions(-)

diff --git a/Documentation/edac.txt b/Documentation/edac.txt
index a5c36842ecef..ced527388001 100644
--- a/Documentation/edac.txt
+++ b/Documentation/edac.txt
@@ -222,74 +222,9 @@ both csrow2 and csrow3 are populated, this indicates a dual ranked
 set of DIMMs for channels 0 and 1.
 
 
-Within each of the 'mc','mcX' and 'csrowX' directories are several
+Within each of the 'mcX' and 'csrowX' directories are several
 EDAC control and attribute files.
 
-
-============================================================================
-DIRECTORY 'mc'
-
-In directory 'mc' are EDAC system overall control and attribute files:
-
-
-Panic on UE control file:
-
-	'edac_mc_panic_on_ue'
-
-	An uncorrectable error will cause a machine panic.  This is usually
-	desirable.  It is a bad idea to continue when an uncorrectable error
-	occurs - it is indeterminate what was uncorrected and the operating
-	system context might be so mangled that continuing will lead to further
-	corruption. If the kernel has MCE configured, then EDAC will never
-	notice the UE.
-
-	LOAD TIME: module/kernel parameter: panic_on_ue=[0|1]
-
-	RUN TIME:  echo "1" >/sys/devices/system/edac/mc/edac_mc_panic_on_ue
-
-
-Log UE control file:
-
-	'edac_mc_log_ue'
-
-	Generate kernel messages describing uncorrectable errors.  These errors
-	are reported through the system message log system.  UE statistics
-	will be accumulated even when UE logging is disabled.
-
-	LOAD TIME: module/kernel parameter: log_ue=[0|1]
-
-	RUN TIME: echo "1" >/sys/devices/system/edac/mc/edac_mc_log_ue
-
-
-Log CE control file:
-
-	'edac_mc_log_ce'
-
-	Generate kernel messages describing correctable errors.  These
-	errors are reported through the system message log system.
-	CE statistics will be accumulated even when CE logging is disabled.
-
-	LOAD TIME: module/kernel parameter: log_ce=[0|1]
-
-	RUN TIME: echo "1" >/sys/devices/system/edac/mc/edac_mc_log_ce
-
-
-Polling period control file:
-
-	'edac_mc_poll_msec'
-
-	The time period, in milliseconds, for polling for error information.
-	Too small a value wastes resources.  Too large a value might delay
-	necessary handling of errors and might loose valuable information for
-	locating the error.  1000 milliseconds (once each second) is the current
-	default. Systems which require all the bandwidth they can get, may
-	increase this.
-
-	LOAD TIME: module/kernel parameter: poll_msec=[0|1]
-
-	RUN TIME: echo "1000" >/sys/devices/system/edac/mc/edac_mc_poll_msec
-
-
 ============================================================================
 'mcX' DIRECTORIES
 
@@ -537,7 +472,6 @@ Channel 1 DIMM Label control file:
 	motherboard specific and determination of this information
 	must occur in userland at this time.
 
-
 ============================================================================
 SYSTEM LOGGING
 
@@ -570,7 +504,6 @@ error type, a notice of "no info" and then an optional,
 driver-specific error message.
 
 
-
 ============================================================================
 PCI Bus Parity Detection
 
@@ -604,6 +537,74 @@ Enable/Disable PCI Parity checking control file:
 	echo "0" >/sys/devices/system/edac/pci/check_pci_parity
 
 
+Parity Count:
+
+	'pci_parity_count'
+
+	This attribute file will display the number of parity errors that
+	have been detected.
+
+
+============================================================================
+MODULE PARAMETERS
+
+Panic on UE control file:
+
+	'edac_mc_panic_on_ue'
+
+	An uncorrectable error will cause a machine panic.  This is usually
+	desirable.  It is a bad idea to continue when an uncorrectable error
+	occurs - it is indeterminate what was uncorrected and the operating
+	system context might be so mangled that continuing will lead to further
+	corruption. If the kernel has MCE configured, then EDAC will never
+	notice the UE.
+
+	LOAD TIME: module/kernel parameter: edac_mc_panic_on_ue=[0|1]
+
+	RUN TIME:  echo "1" > /sys/module/edac_core/parameters/edac_mc_panic_on_ue
+
+
+Log UE control file:
+
+	'edac_mc_log_ue'
+
+	Generate kernel messages describing uncorrectable errors.  These errors
+	are reported through the system message log system.  UE statistics
+	will be accumulated even when UE logging is disabled.
+
+	LOAD TIME: module/kernel parameter: edac_mc_log_ue=[0|1]
+
+	RUN TIME: echo "1" > /sys/module/edac_core/parameters/edac_mc_log_ue
+
+
+Log CE control file:
+
+	'edac_mc_log_ce'
+
+	Generate kernel messages describing correctable errors.  These
+	errors are reported through the system message log system.
+	CE statistics will be accumulated even when CE logging is disabled.
+
+	LOAD TIME: module/kernel parameter: edac_mc_log_ce=[0|1]
+
+	RUN TIME: echo "1" > /sys/module/edac_core/parameters/edac_mc_log_ce
+
+
+Polling period control file:
+
+	'edac_mc_poll_msec'
+
+	The time period, in milliseconds, for polling for error information.
+	Too small a value wastes resources.  Too large a value might delay
+	necessary handling of errors and might loose valuable information for
+	locating the error.  1000 milliseconds (once each second) is the current
+	default. Systems which require all the bandwidth they can get, may
+	increase this.
+
+	LOAD TIME: module/kernel parameter: edac_mc_poll_msec=[0|1]
+
+	RUN TIME: echo "1000" > /sys/module/edac_core/parameters/edac_mc_poll_msec
+
 
 Panic on PCI PARITY Error:
 
@@ -614,21 +615,13 @@ Panic on PCI PARITY Error:
 	error has been detected.
 
 
-	module/kernel parameter: panic_on_pci_parity=[0|1]
+	module/kernel parameter: edac_panic_on_pci_pe=[0|1]
 
 	Enable:
-	echo "1" >/sys/devices/system/edac/pci/panic_on_pci_parity
+	echo "1" > /sys/module/edac_core/parameters/edac_panic_on_pci_pe
 
 	Disable:
-	echo "0" >/sys/devices/system/edac/pci/panic_on_pci_parity
-
-
-Parity Count:
-
-	'pci_parity_count'
-
-	This attribute file will display the number of parity errors that
-	have been detected.
+	echo "0" > /sys/module/edac_core/parameters/edac_panic_on_pci_pe
 
 
 
diff --git a/drivers/edac/edac_mc_sysfs.c b/drivers/edac/edac_mc_sysfs.c
index 7bb9c1532b90..cbe1a17e42f7 100644
--- a/drivers/edac/edac_mc_sysfs.c
+++ b/drivers/edac/edac_mc_sysfs.c
@@ -123,16 +123,6 @@ static const char *edac_caps[] = {
 
 
 
-/*
- * /sys/devices/system/edac/mc;
- *	data structures and methods
- */
-static ssize_t memctrl_int_show(void *ptr, char *buffer)
-{
-	int *value = (int *)ptr;
-	return sprintf(buffer, "%u\n", *value);
-}
-
 static ssize_t memctrl_int_store(void *ptr, const char *buffer, size_t count)
 {
 	int *value = (int *)ptr;
@@ -143,23 +133,6 @@ static ssize_t memctrl_int_store(void *ptr, const char *buffer, size_t count)
 	return count;
 }
 
-/*
- * mc poll_msec time value
- */
-static ssize_t poll_msec_int_store(void *ptr, const char *buffer, size_t count)
-{
-	int *value = (int *)ptr;
-
-	if (isdigit(*buffer)) {
-		*value = simple_strtoul(buffer, NULL, 0);
-
-		/* notify edac_mc engine to reset the poll period */
-		edac_mc_reset_delay_period(*value);
-	}
-
-	return count;
-}
-
 
 /* EDAC sysfs CSROW data structures and methods
  */
@@ -669,98 +642,10 @@ static struct kobj_type ktype_mci = {
 	.default_attrs = (struct attribute **)mci_attr,
 };
 
-/* show/store, tables, etc for the MC kset */
-
-
-struct memctrl_dev_attribute {
-	struct attribute attr;
-	void *value;
-	 ssize_t(*show) (void *, char *);
-	 ssize_t(*store) (void *, const char *, size_t);
-};
-
-/* Set of show/store abstract level functions for memory control object */
-static ssize_t memctrl_dev_show(struct kobject *kobj,
-				struct attribute *attr, char *buffer)
-{
-	struct memctrl_dev_attribute *memctrl_dev;
-	memctrl_dev = (struct memctrl_dev_attribute *)attr;
-
-	if (memctrl_dev->show)
-		return memctrl_dev->show(memctrl_dev->value, buffer);
-
-	return -EIO;
-}
-
-static ssize_t memctrl_dev_store(struct kobject *kobj, struct attribute *attr,
-				 const char *buffer, size_t count)
-{
-	struct memctrl_dev_attribute *memctrl_dev;
-	memctrl_dev = (struct memctrl_dev_attribute *)attr;
-
-	if (memctrl_dev->store)
-		return memctrl_dev->store(memctrl_dev->value, buffer, count);
-
-	return -EIO;
-}
-
-static struct sysfs_ops memctrlfs_ops = {
-	.show = memctrl_dev_show,
-	.store = memctrl_dev_store
-};
-
-#define MEMCTRL_ATTR(_name, _mode, _show, _store)			\
-static struct memctrl_dev_attribute attr_##_name = {			\
-	.attr = {.name = __stringify(_name), .mode = _mode },	\
-	.value  = &_name,					\
-	.show   = _show,					\
-	.store  = _store,					\
-};
-
-#define MEMCTRL_STRING_ATTR(_name, _data, _mode, _show, _store)	\
-static struct memctrl_dev_attribute attr_##_name = {			\
-	.attr = {.name = __stringify(_name), .mode = _mode },	\
-	.value  = _data,					\
-	.show   = _show,					\
-	.store  = _store,					\
-};
-
-/* csrow<id> control files */
-MEMCTRL_ATTR(edac_mc_panic_on_ue,
-	S_IRUGO | S_IWUSR, memctrl_int_show, memctrl_int_store);
-
-MEMCTRL_ATTR(edac_mc_log_ue,
-	S_IRUGO | S_IWUSR, memctrl_int_show, memctrl_int_store);
-
-MEMCTRL_ATTR(edac_mc_log_ce,
-	S_IRUGO | S_IWUSR, memctrl_int_show, memctrl_int_store);
-
-MEMCTRL_ATTR(edac_mc_poll_msec,
-	S_IRUGO | S_IWUSR, memctrl_int_show, poll_msec_int_store);
-
-/* Base Attributes of the memory ECC object */
-static struct memctrl_dev_attribute *memctrl_attr[] = {
-	&attr_edac_mc_panic_on_ue,
-	&attr_edac_mc_log_ue,
-	&attr_edac_mc_log_ce,
-	&attr_edac_mc_poll_msec,
-	NULL,
-};
-
-
-/* the ktype for the mc_kset internal kobj */
-static struct kobj_type ktype_mc_set_attribs = {
-	.sysfs_ops = &memctrlfs_ops,
-	.default_attrs = (struct attribute **)memctrl_attr,
-};
-
 /* EDAC memory controller sysfs kset:
  *	/sys/devices/system/edac/mc
  */
-static struct kset mc_kset = {
-	.kobj = {.ktype = &ktype_mc_set_attribs },
-};
-
+static struct kset mc_kset;
 
 /*
  * edac_mc_register_sysfs_main_kobj

From f9fc82adca43d38a1b79128d80750bd361e15abe Mon Sep 17 00:00:00 2001
From: Arthur Jones <ajones@riverbed.com>
Date: Fri, 25 Jul 2008 01:49:11 -0700
Subject: [PATCH 0943/1171] edac: core fix static to dynamic kset

Static kobjects and ksets are not supported in Linux kernel.  Convert the
mc_kset from static to dynamic.  This patch depends on my previous patch
to remove the module parameter attributes from mc...

Signed-off-by: Arthur Jones <ajones@riverbed.com>
Signed-off-by: Doug Thompson <dougthompson@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/edac/edac_mc_sysfs.c | 15 ++++++---------
 1 file changed, 6 insertions(+), 9 deletions(-)

diff --git a/drivers/edac/edac_mc_sysfs.c b/drivers/edac/edac_mc_sysfs.c
index cbe1a17e42f7..479492819dba 100644
--- a/drivers/edac/edac_mc_sysfs.c
+++ b/drivers/edac/edac_mc_sysfs.c
@@ -645,7 +645,7 @@ static struct kobj_type ktype_mci = {
 /* EDAC memory controller sysfs kset:
  *	/sys/devices/system/edac/mc
  */
-static struct kset mc_kset;
+static struct kset *mc_kset;
 
 /*
  * edac_mc_register_sysfs_main_kobj
@@ -676,7 +676,7 @@ int edac_mc_register_sysfs_main_kobj(struct mem_ctl_info *mci)
 	}
 
 	/* this instance become part of the mc_kset */
-	kobj_mci->kset = &mc_kset;
+	kobj_mci->kset = mc_kset;
 
 	/* register the mc<id> kobject to the mc_kset */
 	err = kobject_init_and_add(kobj_mci, &ktype_mci, NULL,
@@ -906,12 +906,9 @@ int edac_sysfs_setup_mc_kset(void)
 	}
 
 	/* Init the MC's kobject */
-	kobject_set_name(&mc_kset.kobj, "mc");
-	mc_kset.kobj.parent = &edac_class->kset.kobj;
-
-	/* register the mc_kset */
-	err = kset_register(&mc_kset);
-	if (err) {
+	mc_kset = kset_create_and_add("mc", NULL, &edac_class->kset.kobj);
+	if (!mc_kset) {
+		err = -ENOMEM;
 		debugf1("%s() Failed to register '.../edac/mc'\n", __func__);
 		goto fail_out;
 	}
@@ -933,6 +930,6 @@ fail_out:
  */
 void edac_sysfs_teardown_mc_kset(void)
 {
-	kset_unregister(&mc_kset);
+	kset_unregister(mc_kset);
 }
 

From 124682c78563e10ba8b2ecd21b0f1098903b7808 Mon Sep 17 00:00:00 2001
From: Arthur Jones <ajones@riverbed.com>
Date: Fri, 25 Jul 2008 01:49:12 -0700
Subject: [PATCH 0944/1171] edac: core fix added newline to sysfs dimm labels

The channel DIMM label does not seem to be used much in the edac code.
However, where it is used (in the core code), it is assumed to not have a
newline embedded.  This leaves the sysfs file newline free which looks
funny when cat'ing it.  Here we just add the trailing newline to the sysfs
chX_dimm_label output...

[Doug Thompson note: the DIMM label is one of the primary uses of EDAC.
User space daemon scripts, edac-utils@sourceforge, populate the DIMM label
fields, via /sys/devices/system/edac attributes, with the silk screen
labels of the motherboard in use.  dmidecode access BIOS tables, but BIOS
tables are well known to be incorrect and useless in these respects.
edac-utils will strip off any newlines before its use of the output, when
displaying DIMM slot silk screen labels.

Signed-off-by: Arthur Jones <ajones@riverbed.com>
Signed-off-by: Doug Thompson <dougthompson@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/edac/edac_mc_sysfs.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/edac/edac_mc_sysfs.c b/drivers/edac/edac_mc_sysfs.c
index 479492819dba..ad218fe4942d 100644
--- a/drivers/edac/edac_mc_sysfs.c
+++ b/drivers/edac/edac_mc_sysfs.c
@@ -178,7 +178,11 @@ static ssize_t csrow_edac_mode_show(struct csrow_info *csrow, char *data,
 static ssize_t channel_dimm_label_show(struct csrow_info *csrow,
 				char *data, int channel)
 {
-	return snprintf(data, EDAC_MC_LABEL_LEN, "%s",
+	/* if field has not been initialized, there is nothing to send */
+	if (!csrow->channels[channel].label[0])
+		return 0;
+
+	return snprintf(data, EDAC_MC_LABEL_LEN, "%s\n",
 			csrow->channels[channel].label);
 }
 

From 10d33e9c36827e5371479e55ef4089e000af2638 Mon Sep 17 00:00:00 2001
From: Doug Thompson <dougthompson@xmission.com>
Date: Fri, 25 Jul 2008 01:49:12 -0700
Subject: [PATCH 0945/1171] edac: e752x fix too loud on nonmemory errors

This module harvests more than just memory errors, it also harvests
various bus and dma errors that the Chipset detects.  Previously, it would
report all such errors, which would cause output to be TOO loud.

This patches therefore adds a parameter which is used to turn off
NON-MEMORY error reports by default.  Or the reporting can be enabled via
the parameter

Also did code style cleanup: less than 80 characters per line rule

Signed-off-by: Doug Thompson <dougthompson@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/edac/e752x_edac.c | 59 ++++++++++++++++++++++++++++-----------
 1 file changed, 42 insertions(+), 17 deletions(-)

diff --git a/drivers/edac/e752x_edac.c b/drivers/edac/e752x_edac.c
index c94a0eb492cb..facfdb1fa71c 100644
--- a/drivers/edac/e752x_edac.c
+++ b/drivers/edac/e752x_edac.c
@@ -28,6 +28,7 @@
 #define E752X_REVISION	" Ver: 2.0.2 " __DATE__
 #define EDAC_MOD_STR	"e752x_edac"
 
+static int report_non_memory_errors;
 static int force_function_unhide;
 static int sysbus_parity = -1;
 
@@ -117,7 +118,7 @@ static struct edac_pci_ctl_info *e752x_pci;
 #define E752X_BUF_FERR		0x70	/* Memory buffer first error reg (8b) */
 #define E752X_BUF_NERR		0x72	/* Memory buffer next error reg (8b) */
 #define E752X_BUF_ERRMASK	0x74	/* Memory buffer error mask reg (8b) */
-#define E752X_BUF_SMICMD	0x7A	/* Memory buffer SMI command reg (8b) */
+#define E752X_BUF_SMICMD	0x7A	/* Memory buffer SMI cmd reg (8b) */
 #define E752X_DRAM_FERR		0x80	/* DRAM first error register (16b) */
 #define E752X_DRAM_NERR		0x82	/* DRAM next error register (16b) */
 #define E752X_DRAM_ERRMASK	0x84	/* DRAM error mask register (8b) */
@@ -127,7 +128,7 @@ static struct edac_pci_ctl_info *e752x_pci;
 					/*     error address register (32b) */
 					/*
 					 * 31    Reserved
-					 * 30:2  CE address (64 byte block 34:6)
+					 * 30:2  CE address (64 byte block 34:6
 					 * 1     Reserved
 					 * 0     HiLoCS
 					 */
@@ -147,11 +148,11 @@ static struct edac_pci_ctl_info *e752x_pci;
 					 * 1     Reserved
 					 * 0     HiLoCS
 					 */
-#define E752X_DRAM_SCRB_ADD	0xA8	/* DRAM first uncorrectable scrub memory */
+#define E752X_DRAM_SCRB_ADD	0xA8	/* DRAM 1st uncorrectable scrub mem */
 					/*     error address register (32b) */
 					/*
 					 * 31    Reserved
-					 * 30:2  CE address (64 byte block 34:6)
+					 * 30:2  CE address (64 byte block 34:6
 					 * 1     Reserved
 					 * 0     HiLoCS
 					 */
@@ -394,9 +395,12 @@ static void do_process_ded_retry(struct mem_ctl_info *mci, u16 error,
 	struct e752x_pvt *pvt = (struct e752x_pvt *)mci->pvt_info;
 
 	error_1b = retry_add;
-	page = error_1b >> (PAGE_SHIFT - 4);	/* convert the addr to 4k page */
-	row = pvt->mc_symmetric ? ((page >> 1) & 3) :	/* chip select are bits 14 & 13 */
+	page = error_1b >> (PAGE_SHIFT - 4);  /* convert the addr to 4k page */
+
+	/* chip select are bits 14 & 13 */
+	row = pvt->mc_symmetric ? ((page >> 1) & 3) :
 		edac_mc_find_csrow_by_page(mci, page);
+
 	e752x_mc_printk(mci, KERN_WARNING,
 			"CE page 0x%lx, row %d : Memory read retry\n",
 			(long unsigned int)page, row);
@@ -422,12 +426,21 @@ static inline void process_threshold_ce(struct mem_ctl_info *mci, u16 error,
 }
 
 static char *global_message[11] = {
-	"PCI Express C1", "PCI Express C", "PCI Express B1",
-	"PCI Express B", "PCI Express A1", "PCI Express A",
-	"DMA Controler", "HUB or NS Interface", "System Bus",
-	"DRAM Controler", "Internal Buffer"
+	"PCI Express C1",
+	"PCI Express C",
+	"PCI Express B1",
+	"PCI Express B",
+	"PCI Express A1",
+	"PCI Express A",
+	"DMA Controller",
+	"HUB or NS Interface",
+	"System Bus",
+	"DRAM Controller",  /* 9th entry */
+	"Internal Buffer"
 };
 
+#define DRAM_ENTRY	9
+
 static char *fatal_message[2] = { "Non-Fatal ", "Fatal " };
 
 static void do_global_error(int fatal, u32 errors)
@@ -435,9 +448,16 @@ static void do_global_error(int fatal, u32 errors)
 	int i;
 
 	for (i = 0; i < 11; i++) {
-		if (errors & (1 << i))
-			e752x_printk(KERN_WARNING, "%sError %s\n",
-				fatal_message[fatal], global_message[i]);
+		if (errors & (1 << i)) {
+			/* If the error is from DRAM Controller OR
+			 * we are to report ALL errors, then
+			 * report the error
+			 */
+			if ((i == DRAM_ENTRY) || report_non_memory_errors)
+				e752x_printk(KERN_WARNING, "%sError %s\n",
+					fatal_message[fatal],
+					global_message[i]);
+		}
 	}
 }
 
@@ -1021,7 +1041,7 @@ static int e752x_get_devs(struct pci_dev *pdev, int dev_idx,
 	struct pci_dev *dev;
 
 	pvt->bridge_ck = pci_get_device(PCI_VENDOR_ID_INTEL,
-					pvt->dev_info->err_dev, pvt->bridge_ck);
+				pvt->dev_info->err_dev, pvt->bridge_ck);
 
 	if (pvt->bridge_ck == NULL)
 		pvt->bridge_ck = pci_scan_single_device(pdev->bus,
@@ -1034,8 +1054,9 @@ static int e752x_get_devs(struct pci_dev *pdev, int dev_idx,
 		return 1;
 	}
 
-	dev = pci_get_device(PCI_VENDOR_ID_INTEL, e752x_devs[dev_idx].ctl_dev,
-			NULL);
+	dev = pci_get_device(PCI_VENDOR_ID_INTEL,
+				e752x_devs[dev_idx].ctl_dev,
+				NULL);
 
 	if (dev == NULL)
 		goto fail;
@@ -1316,7 +1337,8 @@ MODULE_DESCRIPTION("MC support for Intel e752x/3100 memory controllers");
 
 module_param(force_function_unhide, int, 0444);
 MODULE_PARM_DESC(force_function_unhide, "if BIOS sets Dev0:Fun1 up as hidden:"
-		 " 1=force unhide and hope BIOS doesn't fight driver for Dev0:Fun1 access");
+		 " 1=force unhide and hope BIOS doesn't fight driver for "
+		"Dev0:Fun1 access");
 
 module_param(edac_op_state, int, 0444);
 MODULE_PARM_DESC(edac_op_state, "EDAC Error Reporting state: 0=Poll,1=NMI");
@@ -1324,3 +1346,6 @@ MODULE_PARM_DESC(edac_op_state, "EDAC Error Reporting state: 0=Poll,1=NMI");
 module_param(sysbus_parity, int, 0444);
 MODULE_PARM_DESC(sysbus_parity, "0=disable system bus parity checking,"
 		" 1=enable system bus parity checking, default=auto-detect");
+module_param(report_non_memory_errors, int, 0644);
+MODULE_PARM_DESC(report_non_memory_errors, "0=disable non-memory error "
+		"reporting, 1=enable non-memory error reporting");

From 596d3941035d4d4b484c820f10f57fd4816c6615 Mon Sep 17 00:00:00 2001
From: Dave Jiang <djiang@mvista.com>
Date: Fri, 25 Jul 2008 01:49:13 -0700
Subject: [PATCH 0946/1171] edac: mv64x60 fix get_property

Update get_property() call to use of_get_property() in order to fix compile

Signed-off-by: Dave Jiang <djiang@mvista.com>
Signed-off-by: Doug Thompson <dougthompson.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/edac/mv64x60_edac.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/edac/mv64x60_edac.c b/drivers/edac/mv64x60_edac.c
index bf071f140a05..de69163ff5b3 100644
--- a/drivers/edac/mv64x60_edac.c
+++ b/drivers/edac/mv64x60_edac.c
@@ -612,7 +612,7 @@ static void get_total_mem(struct mv64x60_mc_pdata *pdata)
 	if (!np)
 		return;
 
-	reg = get_property(np, "reg", NULL);
+	reg = of_get_property(np, "reg", NULL);
 
 	pdata->total_mem = reg[1];
 }

From fcb19171d196172a4f57e056f7a60e6d1e2e8c85 Mon Sep 17 00:00:00 2001
From: Dave Jiang <djiang@mvista.com>
Date: Fri, 25 Jul 2008 01:49:14 -0700
Subject: [PATCH 0947/1171] edac: mv64x60 add pci fixup

Fixup of missing bit 0 on 64360 PCIx_ERR_MASK and errata FEr-#11 and
FEr-#16 for the 64460.  Bit 0 must remain 0.

Signed-off-by: Dave Jiang <djiang@mvista.com>
Signed-off-by: Doug Thompson <dougthompson.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/edac/mv64x60_edac.c | 35 +++++++++++++++++++++++++++++++++++
 1 file changed, 35 insertions(+)

diff --git a/drivers/edac/mv64x60_edac.c b/drivers/edac/mv64x60_edac.c
index de69163ff5b3..083ce8d0c63d 100644
--- a/drivers/edac/mv64x60_edac.c
+++ b/drivers/edac/mv64x60_edac.c
@@ -71,6 +71,35 @@ static irqreturn_t mv64x60_pci_isr(int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
+/*
+ * Bit 0 of MV64x60_PCIx_ERR_MASK does not exist on the 64360 and because of
+ * errata FEr-#11 and FEr-##16 for the 64460, it should be 0 on that chip as
+ * well.  IOW, don't set bit 0.
+ */
+
+/* Erratum FEr PCI-#16: clear bit 0 of PCI SERRn Mask reg. */
+static int __init mv64x60_pci_fixup(struct platform_device *pdev)
+{
+	struct resource *r;
+	void __iomem *pci_serr;
+
+	r = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+	if (!r) {
+		printk(KERN_ERR "%s: Unable to get resource for "
+		       "PCI err regs\n", __func__);
+		return -ENOENT;
+	}
+
+	pci_serr = ioremap(r->start, r->end - r->start + 1);
+	if (!pci_serr)
+		return -ENOMEM;
+
+	out_le32(pci_serr, in_le32(pci_serr) & ~0x1);
+	iounmap(pci_serr);
+
+	return 0;
+}
+
 static int __devinit mv64x60_pci_err_probe(struct platform_device *pdev)
 {
 	struct edac_pci_ctl_info *pci;
@@ -128,6 +157,12 @@ static int __devinit mv64x60_pci_err_probe(struct platform_device *pdev)
 		goto err;
 	}
 
+	res = mv64x60_pci_fixup(pdev);
+	if (res < 0) {
+		printk(KERN_ERR "%s: PCI fixup failed\n", __func__);
+		goto err;
+	}
+
 	out_le32(pdata->pci_vbase + MV64X60_PCI_ERROR_CAUSE, 0);
 	out_le32(pdata->pci_vbase + MV64X60_PCI_ERROR_MASK, 0);
 	out_le32(pdata->pci_vbase + MV64X60_PCI_ERROR_MASK,

From f87bd330edf06fd49b3fbc368d90fb180375f2a2 Mon Sep 17 00:00:00 2001
From: Dave Jiang <djiang@mvista.com>
Date: Fri, 25 Jul 2008 01:49:14 -0700
Subject: [PATCH 0948/1171] edac: mpc85xx fix pci ofdev 2nd pass

Convert PCI err device from platform to open firmware of_dev to comply
with powerpc schemes.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Dave Jiang <djiang@mvista.com>
Signed-off-by: Doug Thompson <dougthompson@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/edac/mpc85xx_edac.c | 67 ++++++++++++++++++++++++-------------
 1 file changed, 43 insertions(+), 24 deletions(-)

diff --git a/drivers/edac/mpc85xx_edac.c b/drivers/edac/mpc85xx_edac.c
index d49361bfe670..2265d9ca1535 100644
--- a/drivers/edac/mpc85xx_edac.c
+++ b/drivers/edac/mpc85xx_edac.c
@@ -195,14 +195,15 @@ static irqreturn_t mpc85xx_pci_isr(int irq, void *dev_id)
 	return IRQ_HANDLED;
 }
 
-static int __devinit mpc85xx_pci_err_probe(struct platform_device *pdev)
+static int __devinit mpc85xx_pci_err_probe(struct of_device *op,
+					   const struct of_device_id *match)
 {
 	struct edac_pci_ctl_info *pci;
 	struct mpc85xx_pci_pdata *pdata;
-	struct resource *r;
+	struct resource r;
 	int res = 0;
 
-	if (!devres_open_group(&pdev->dev, mpc85xx_pci_err_probe, GFP_KERNEL))
+	if (!devres_open_group(&op->dev, mpc85xx_pci_err_probe, GFP_KERNEL))
 		return -ENOMEM;
 
 	pci = edac_pci_alloc_ctl_info(sizeof(*pdata), "mpc85xx_pci_err");
@@ -212,34 +213,37 @@ static int __devinit mpc85xx_pci_err_probe(struct platform_device *pdev)
 	pdata = pci->pvt_info;
 	pdata->name = "mpc85xx_pci_err";
 	pdata->irq = NO_IRQ;
-	platform_set_drvdata(pdev, pci);
-	pci->dev = &pdev->dev;
+	dev_set_drvdata(&op->dev, pci);
+	pci->dev = &op->dev;
 	pci->mod_name = EDAC_MOD_STR;
 	pci->ctl_name = pdata->name;
-	pci->dev_name = pdev->dev.bus_id;
+	pci->dev_name = op->dev.bus_id;
 
 	if (edac_op_state == EDAC_OPSTATE_POLL)
 		pci->edac_check = mpc85xx_pci_check;
 
 	pdata->edac_idx = edac_pci_idx++;
 
-	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (!r) {
+	res = of_address_to_resource(op->node, 0, &r);
+	if (res) {
 		printk(KERN_ERR "%s: Unable to get resource for "
 		       "PCI err regs\n", __func__);
 		goto err;
 	}
 
-	if (!devm_request_mem_region(&pdev->dev, r->start,
-				     r->end - r->start + 1, pdata->name)) {
+	/* we only need the error registers */
+	r.start += 0xe00;
+
+	if (!devm_request_mem_region(&op->dev, r.start,
+					r.end - r.start + 1, pdata->name)) {
 		printk(KERN_ERR "%s: Error while requesting mem region\n",
 		       __func__);
 		res = -EBUSY;
 		goto err;
 	}
 
-	pdata->pci_vbase = devm_ioremap(&pdev->dev, r->start,
-					r->end - r->start + 1);
+	pdata->pci_vbase = devm_ioremap(&op->dev, r.start,
+					r.end - r.start + 1);
 	if (!pdata->pci_vbase) {
 		printk(KERN_ERR "%s: Unable to setup PCI err regs\n", __func__);
 		res = -ENOMEM;
@@ -266,14 +270,15 @@ static int __devinit mpc85xx_pci_err_probe(struct platform_device *pdev)
 	}
 
 	if (edac_op_state == EDAC_OPSTATE_INT) {
-		pdata->irq = platform_get_irq(pdev, 0);
-		res = devm_request_irq(&pdev->dev, pdata->irq,
+		pdata->irq = irq_of_parse_and_map(op->node, 0);
+		res = devm_request_irq(&op->dev, pdata->irq,
 				       mpc85xx_pci_isr, IRQF_DISABLED,
 				       "[EDAC] PCI err", pci);
 		if (res < 0) {
 			printk(KERN_ERR
 			       "%s: Unable to requiest irq %d for "
 			       "MPC85xx PCI err\n", __func__, pdata->irq);
+			irq_dispose_mapping(pdata->irq);
 			res = -ENODEV;
 			goto err2;
 		}
@@ -282,23 +287,23 @@ static int __devinit mpc85xx_pci_err_probe(struct platform_device *pdev)
 		       pdata->irq);
 	}
 
-	devres_remove_group(&pdev->dev, mpc85xx_pci_err_probe);
+	devres_remove_group(&op->dev, mpc85xx_pci_err_probe);
 	debugf3("%s(): success\n", __func__);
 	printk(KERN_INFO EDAC_MOD_STR " PCI err registered\n");
 
 	return 0;
 
 err2:
-	edac_pci_del_device(&pdev->dev);
+	edac_pci_del_device(&op->dev);
 err:
 	edac_pci_free_ctl_info(pci);
-	devres_release_group(&pdev->dev, mpc85xx_pci_err_probe);
+	devres_release_group(&op->dev, mpc85xx_pci_err_probe);
 	return res;
 }
 
-static int mpc85xx_pci_err_remove(struct platform_device *pdev)
+static int mpc85xx_pci_err_remove(struct of_device *op)
 {
-	struct edac_pci_ctl_info *pci = platform_get_drvdata(pdev);
+	struct edac_pci_ctl_info *pci = dev_get_drvdata(&op->dev);
 	struct mpc85xx_pci_pdata *pdata = pci->pvt_info;
 
 	debugf0("%s()\n", __func__);
@@ -318,12 +323,26 @@ static int mpc85xx_pci_err_remove(struct platform_device *pdev)
 	return 0;
 }
 
-static struct platform_driver mpc85xx_pci_err_driver = {
+static struct of_device_id mpc85xx_pci_err_of_match[] = {
+	{
+	 .compatible = "fsl,mpc8540-pcix",
+	 },
+	{
+	 .compatible = "fsl,mpc8540-pci",
+	},
+	{},
+};
+
+static struct of_platform_driver mpc85xx_pci_err_driver = {
+	.owner = THIS_MODULE,
+	.name = "mpc85xx_pci_err",
+	.match_table = mpc85xx_pci_err_of_match,
 	.probe = mpc85xx_pci_err_probe,
 	.remove = __devexit_p(mpc85xx_pci_err_remove),
 	.driver = {
-		.name = "mpc85xx_pci_err",
-	}
+		   .name = "mpc85xx_pci_err",
+		   .owner = THIS_MODULE,
+		   },
 };
 
 #endif				/* CONFIG_PCI */
@@ -1002,7 +1021,7 @@ static int __init mpc85xx_mc_init(void)
 		printk(KERN_WARNING EDAC_MOD_STR "L2 fails to register\n");
 
 #ifdef CONFIG_PCI
-	res = platform_driver_register(&mpc85xx_pci_err_driver);
+	res = of_register_platform_driver(&mpc85xx_pci_err_driver);
 	if (res)
 		printk(KERN_WARNING EDAC_MOD_STR "PCI fails to register\n");
 #endif
@@ -1025,7 +1044,7 @@ static void __exit mpc85xx_mc_exit(void)
 {
 	mtspr(SPRN_HID1, orig_hid1);
 #ifdef CONFIG_PCI
-	platform_driver_unregister(&mpc85xx_pci_err_driver);
+	of_unregister_platform_driver(&mpc85xx_pci_err_driver);
 #endif
 	of_unregister_platform_driver(&mpc85xx_l2_err_driver);
 	of_unregister_platform_driver(&mpc85xx_mc_err_driver);

From 93082f0b15841b8926c38ef224d0e6f720000635 Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Fri, 25 Jul 2008 10:56:36 -0700
Subject: [PATCH 0949/1171] Fix ahci driver 'flags' type

The new type checking of the flags arguments to irqsave and friends
(commit 3f307891ce0e7b0438c432af1aacd656a092ff45) pointed out this thing
with a big nice warning.

Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/ata/ahci.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index dc7596f028b6..ef3e5522e1a4 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -1273,7 +1273,7 @@ static ssize_t ahci_transmit_led_message(struct ata_port *ap, u32 state,
 	void __iomem *mmio = ap->host->iomap[AHCI_PCI_BAR];
 	u32 em_ctl;
 	u32 message[] = {0, 0};
-	unsigned int flags;
+	unsigned long flags;
 	int pmp;
 	struct ahci_em_priv *emp;
 

From 44463f7dd6c8039904333e4374e5c6e9ad83006f Mon Sep 17 00:00:00 2001
From: David Woodhouse <David.Woodhouse@intel.com>
Date: Fri, 25 Jul 2008 14:17:19 -0400
Subject: [PATCH 0950/1171] firmware: create firmware binaries during 'make
 modules'.

This means that we no longer need write access to the source tree while
doing 'make modules_install'.

Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 Makefile                |  1 +
 scripts/Makefile.fwinst | 10 ++++++----
 2 files changed, 7 insertions(+), 4 deletions(-)

diff --git a/Makefile b/Makefile
index 4bcd1cf90cb1..3cad7db5eba7 100644
--- a/Makefile
+++ b/Makefile
@@ -1061,6 +1061,7 @@ modules: $(vmlinux-dirs) $(if $(KBUILD_BUILTIN),vmlinux)
 	$(Q)$(AWK) '!x[$$0]++' $(vmlinux-dirs:%=$(objtree)/%/modules.order) > $(objtree)/modules.order
 	@echo '  Building modules, stage 2.';
 	$(Q)$(MAKE) -f $(srctree)/scripts/Makefile.modpost
+	$(Q)$(MAKE) -f $(srctree)/scripts/Makefile.fwinst obj=firmware __fw_modbuild
 
 
 # Target to prepare building external modules
diff --git a/scripts/Makefile.fwinst b/scripts/Makefile.fwinst
index c972c0f54ce0..f63a663de158 100644
--- a/scripts/Makefile.fwinst
+++ b/scripts/Makefile.fwinst
@@ -17,14 +17,15 @@ include $(srctree)/$(obj)/Makefile
 
 include scripts/Makefile.host
 
-mod-fw := $(addprefix $(INSTALL_FW_PATH)/,$(fw-shipped-m))
-
+mod-fw := $(fw-shipped-m)
 # If CONFIG_FIRMWARE_IN_KERNEL isn't set, then install the 
 # firmware for in-kernel drivers too.
 ifndef CONFIG_FIRMWARE_IN_KERNEL
-mod-fw += $(addprefix $(INSTALL_FW_PATH)/,$(fw-shipped-y))
+mod-fw += $(fw-shipped-y)
 endif
 
+installed-mod-fw := $(addprefix $(INSTALL_FW_PATH)/,$(mod-fw))
+
 installed-fw := $(addprefix $(INSTALL_FW_PATH)/,$(fw-shipped-all))
 installed-fw-dirs := $(sort $(dir $(installed-fw))) $(INSTALL_FW_PATH)/.
 
@@ -49,7 +50,8 @@ PHONY +=  __fw_install __fw_modinst FORCE
 .PHONY: $(PHONY)
 
 __fw_install: $(installed-fw)
-__fw_modinst: $(mod-fw)
+__fw_modinst: $(installed-mod-fw)
+__fw_modbuild: $(addprefix $(obj)/,$(mod-fw))
 
 FORCE:
 

From e44d1b2998d62a1f2f4d7eb17b56ba396535509f Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 25 Jul 2008 12:57:41 +0200
Subject: [PATCH 0951/1171] mm/hugetlb.c: fix build failure with !CONFIG_SYSCTL

on !CONFIG_SYSCTL on x86 with latest -git i get:

     mm/hugetlb.c: In function 'decrement_hugepage_resv_vma':
     mm/hugetlb.c:83: error: 'reserve' undeclared (first use in this function)
     mm/hugetlb.c:83: error: (Each undeclared identifier is reported only once
     mm/hugetlb.c:83: error: for each function it appears in.)

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/hugetlb.c | 22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 41341c414194..a8bf4ab01f86 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -1026,6 +1026,17 @@ static void __init report_hugepages(void)
 	}
 }
 
+static unsigned int cpuset_mems_nr(unsigned int *array)
+{
+	int node;
+	unsigned int nr = 0;
+
+	for_each_node_mask(node, cpuset_current_mems_allowed)
+		nr += array[node];
+
+	return nr;
+}
+
 #ifdef CONFIG_SYSCTL
 #ifdef CONFIG_HIGHMEM
 static void try_to_free_low(struct hstate *h, unsigned long count)
@@ -1375,17 +1386,6 @@ static int __init hugetlb_default_setup(char *s)
 }
 __setup("default_hugepagesz=", hugetlb_default_setup);
 
-static unsigned int cpuset_mems_nr(unsigned int *array)
-{
-	int node;
-	unsigned int nr = 0;
-
-	for_each_node_mask(node, cpuset_current_mems_allowed)
-		nr += array[node];
-
-	return nr;
-}
-
 int hugetlb_sysctl_handler(struct ctl_table *table, int write,
 			   struct file *file, void __user *buffer,
 			   size_t *length, loff_t *ppos)

From 9b81361631bbb1d85c99ddec677d42afe516737b Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Fri, 25 Jul 2008 13:02:37 +0200
Subject: [PATCH 0952/1171] signalfd: fix undefined reference to
 `compat_sys_signalfd4' when !CONFIG_SIGNALFD

fix:

arch/x86/ia32/built-in.o: In function `ia32_sys_call_table':
(.rodata+0xa38): undefined reference to `compat_sys_signalfd4'

on !CONFIG_SIGNALFD.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/sys_ni.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c
index 55eca1594da9..08d6e1bb99ac 100644
--- a/kernel/sys_ni.c
+++ b/kernel/sys_ni.c
@@ -160,6 +160,7 @@ cond_syscall(sys_ioprio_get);
 cond_syscall(sys_signalfd);
 cond_syscall(sys_signalfd4);
 cond_syscall(compat_sys_signalfd);
+cond_syscall(compat_sys_signalfd4);
 cond_syscall(sys_timerfd_create);
 cond_syscall(sys_timerfd_settime);
 cond_syscall(sys_timerfd_gettime);

From 7dcf2a9fced59e58e4694cdcf15850c01fdba89b Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Tue, 1 Jul 2008 19:27:16 +0300
Subject: [PATCH 0953/1171] remove dummy asm/kvm.h files

This patch removes the dummy asm/kvm.h files on architectures not (yet)
supporting KVM and uses the same conditional headers installation as
already used for a.out.h .

Also removed are superfluous install rules in the s390 and x86 Kbuild
files (they are already in Kbuild.asm).

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
---
 include/asm-alpha/kvm.h        | 6 ------
 include/asm-arm/kvm.h          | 6 ------
 include/asm-avr32/kvm.h        | 6 ------
 include/asm-blackfin/kvm.h     | 6 ------
 include/asm-cris/kvm.h         | 6 ------
 include/asm-frv/kvm.h          | 6 ------
 include/asm-generic/Kbuild.asm | 2 ++
 include/asm-h8300/kvm.h        | 6 ------
 include/asm-m32r/kvm.h         | 6 ------
 include/asm-m68k/kvm.h         | 6 ------
 include/asm-m68knommu/kvm.h    | 6 ------
 include/asm-mips/kvm.h         | 6 ------
 include/asm-mn10300/kvm.h      | 6 ------
 include/asm-parisc/kvm.h       | 6 ------
 include/asm-s390/Kbuild        | 1 -
 include/asm-sh/kvm.h           | 6 ------
 include/asm-sparc/kvm.h        | 6 ------
 include/asm-sparc64/kvm.h      | 1 -
 include/asm-um/kvm.h           | 6 ------
 include/asm-x86/Kbuild         | 1 -
 include/asm-xtensa/kvm.h       | 6 ------
 include/linux/Kbuild           | 2 ++
 22 files changed, 4 insertions(+), 105 deletions(-)
 delete mode 100644 include/asm-alpha/kvm.h
 delete mode 100644 include/asm-arm/kvm.h
 delete mode 100644 include/asm-avr32/kvm.h
 delete mode 100644 include/asm-blackfin/kvm.h
 delete mode 100644 include/asm-cris/kvm.h
 delete mode 100644 include/asm-frv/kvm.h
 delete mode 100644 include/asm-h8300/kvm.h
 delete mode 100644 include/asm-m32r/kvm.h
 delete mode 100644 include/asm-m68k/kvm.h
 delete mode 100644 include/asm-m68knommu/kvm.h
 delete mode 100644 include/asm-mips/kvm.h
 delete mode 100644 include/asm-mn10300/kvm.h
 delete mode 100644 include/asm-parisc/kvm.h
 delete mode 100644 include/asm-sh/kvm.h
 delete mode 100644 include/asm-sparc/kvm.h
 delete mode 100644 include/asm-sparc64/kvm.h
 delete mode 100644 include/asm-um/kvm.h
 delete mode 100644 include/asm-xtensa/kvm.h

diff --git a/include/asm-alpha/kvm.h b/include/asm-alpha/kvm.h
deleted file mode 100644
index b9daec429689..000000000000
--- a/include/asm-alpha/kvm.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __LINUX_KVM_ALPHA_H
-#define __LINUX_KVM_ALPHA_H
-
-/* alpha does not support KVM */
-
-#endif
diff --git a/include/asm-arm/kvm.h b/include/asm-arm/kvm.h
deleted file mode 100644
index cb3c08cbcb9e..000000000000
--- a/include/asm-arm/kvm.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __LINUX_KVM_ARM_H
-#define __LINUX_KVM_ARM_H
-
-/* arm does not support KVM */
-
-#endif
diff --git a/include/asm-avr32/kvm.h b/include/asm-avr32/kvm.h
deleted file mode 100644
index 8c5777020e2c..000000000000
--- a/include/asm-avr32/kvm.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __LINUX_KVM_AVR32_H
-#define __LINUX_KVM_AVR32_H
-
-/* avr32 does not support KVM */
-
-#endif
diff --git a/include/asm-blackfin/kvm.h b/include/asm-blackfin/kvm.h
deleted file mode 100644
index e3477d77c014..000000000000
--- a/include/asm-blackfin/kvm.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __LINUX_KVM_BLACKFIN_H
-#define __LINUX_KVM_BLACKFIN_H
-
-/* blackfin does not support KVM */
-
-#endif
diff --git a/include/asm-cris/kvm.h b/include/asm-cris/kvm.h
deleted file mode 100644
index c860f51149f0..000000000000
--- a/include/asm-cris/kvm.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __LINUX_KVM_CRIS_H
-#define __LINUX_KVM_CRIS_H
-
-/* cris does not support KVM */
-
-#endif
diff --git a/include/asm-frv/kvm.h b/include/asm-frv/kvm.h
deleted file mode 100644
index 9c8a4f08d0a9..000000000000
--- a/include/asm-frv/kvm.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __LINUX_KVM_FRV_H
-#define __LINUX_KVM_FRV_H
-
-/* frv does not support KVM */
-
-#endif
diff --git a/include/asm-generic/Kbuild.asm b/include/asm-generic/Kbuild.asm
index 7cd25b8e7c9a..1170dc60e638 100644
--- a/include/asm-generic/Kbuild.asm
+++ b/include/asm-generic/Kbuild.asm
@@ -1,4 +1,6 @@
+ifneq ($(wildcard $(srctree)/include/asm-$(SRCARCH)/kvm.h),)
 header-y  += kvm.h
+endif
 
 ifneq ($(wildcard $(srctree)/include/asm-$(SRCARCH)/a.out.h),)
 unifdef-y += a.out.h
diff --git a/include/asm-h8300/kvm.h b/include/asm-h8300/kvm.h
deleted file mode 100644
index bdbed7b987e1..000000000000
--- a/include/asm-h8300/kvm.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __LINUX_KVM_H8300_H
-#define __LINUX_KVM_H8300_H
-
-/* h8300 does not support KVM */
-
-#endif
diff --git a/include/asm-m32r/kvm.h b/include/asm-m32r/kvm.h
deleted file mode 100644
index 99a40515b77e..000000000000
--- a/include/asm-m32r/kvm.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __LINUX_KVM_M32R_H
-#define __LINUX_KVM_M32R_H
-
-/* m32r does not support KVM */
-
-#endif
diff --git a/include/asm-m68k/kvm.h b/include/asm-m68k/kvm.h
deleted file mode 100644
index 7ed27fce5240..000000000000
--- a/include/asm-m68k/kvm.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __LINUX_KVM_M68K_H
-#define __LINUX_KVM_M68K_H
-
-/* m68k does not support KVM */
-
-#endif
diff --git a/include/asm-m68knommu/kvm.h b/include/asm-m68knommu/kvm.h
deleted file mode 100644
index b49d4258dabb..000000000000
--- a/include/asm-m68knommu/kvm.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __LINUX_KVM_M68KNOMMU_H
-#define __LINUX_KVM_M68KNOMMU_H
-
-/* m68knommu does not support KVM */
-
-#endif
diff --git a/include/asm-mips/kvm.h b/include/asm-mips/kvm.h
deleted file mode 100644
index 093a5b7f796b..000000000000
--- a/include/asm-mips/kvm.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __LINUX_KVM_MIPS_H
-#define __LINUX_KVM_MIPS_H
-
-/* mips does not support KVM */
-
-#endif
diff --git a/include/asm-mn10300/kvm.h b/include/asm-mn10300/kvm.h
deleted file mode 100644
index f6b609ff4a57..000000000000
--- a/include/asm-mn10300/kvm.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __LINUX_KVM_MN10300_H
-#define __LINUX_KVM_MN10300_H
-
-/* mn10300 does not support KVM */
-
-#endif
diff --git a/include/asm-parisc/kvm.h b/include/asm-parisc/kvm.h
deleted file mode 100644
index 00cc45812547..000000000000
--- a/include/asm-parisc/kvm.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __LINUX_KVM_PARISC_H
-#define __LINUX_KVM_PARISC_H
-
-/* parisc does not support KVM */
-
-#endif
diff --git a/include/asm-s390/Kbuild b/include/asm-s390/Kbuild
index bb5e9edb9825..63a23415fba6 100644
--- a/include/asm-s390/Kbuild
+++ b/include/asm-s390/Kbuild
@@ -7,7 +7,6 @@ header-y += tape390.h
 header-y += ucontext.h
 header-y += vtoc.h
 header-y += zcrypt.h
-header-y += kvm.h
 header-y += chsc.h
 
 unifdef-y += cmb.h
diff --git a/include/asm-sh/kvm.h b/include/asm-sh/kvm.h
deleted file mode 100644
index 6af51dbab2d0..000000000000
--- a/include/asm-sh/kvm.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __LINUX_KVM_SH_H
-#define __LINUX_KVM_SH_H
-
-/* sh does not support KVM */
-
-#endif
diff --git a/include/asm-sparc/kvm.h b/include/asm-sparc/kvm.h
deleted file mode 100644
index 2e5478da3819..000000000000
--- a/include/asm-sparc/kvm.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __LINUX_KVM_SPARC_H
-#define __LINUX_KVM_SPARC_H
-
-/* sparc does not support KVM */
-
-#endif
diff --git a/include/asm-sparc64/kvm.h b/include/asm-sparc64/kvm.h
deleted file mode 100644
index 53564ad86b15..000000000000
--- a/include/asm-sparc64/kvm.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-sparc/kvm.h>
diff --git a/include/asm-um/kvm.h b/include/asm-um/kvm.h
deleted file mode 100644
index 66aa77094551..000000000000
--- a/include/asm-um/kvm.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __LINUX_KVM_UM_H
-#define __LINUX_KVM_UM_H
-
-/* um does not support KVM */
-
-#endif
diff --git a/include/asm-x86/Kbuild b/include/asm-x86/Kbuild
index 1e3554596f72..811e9828ccb3 100644
--- a/include/asm-x86/Kbuild
+++ b/include/asm-x86/Kbuild
@@ -3,7 +3,6 @@ include include/asm-generic/Kbuild.asm
 header-y += boot.h
 header-y += bootparam.h
 header-y += debugreg.h
-header-y += kvm.h
 header-y += ldt.h
 header-y += msr-index.h
 header-y += prctl.h
diff --git a/include/asm-xtensa/kvm.h b/include/asm-xtensa/kvm.h
deleted file mode 100644
index bda4e331e98c..000000000000
--- a/include/asm-xtensa/kvm.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __LINUX_KVM_XTENSA_H
-#define __LINUX_KVM_XTENSA_H
-
-/* xtensa does not support KVM */
-
-#endif
diff --git a/include/linux/Kbuild b/include/linux/Kbuild
index 71d70d1fbce2..402c8f55d713 100644
--- a/include/linux/Kbuild
+++ b/include/linux/Kbuild
@@ -256,7 +256,9 @@ unifdef-y += kd.h
 unifdef-y += kernelcapi.h
 unifdef-y += kernel.h
 unifdef-y += keyboard.h
+ifneq ($(wildcard $(srctree)/include/asm-$(SRCARCH)/kvm.h),)
 unifdef-y += kvm.h
+endif
 unifdef-y += llc.h
 unifdef-y += loop.h
 unifdef-y += lp.h

From c6af5e9f8a57467df2e55e428316a43480174521 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Fri, 25 Jul 2008 15:48:04 +0200
Subject: [PATCH 0954/1171] bootmem: Move node allocation macros back to
 !HAVE_ARCH_BOOTMEM_NODE

These got unintentionally moved, put them back as x86 provides its own
versions.

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/bootmem.h | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/include/linux/bootmem.h b/include/linux/bootmem.h
index 4ddf2922fc8d..652470b687c9 100644
--- a/include/linux/bootmem.h
+++ b/include/linux/bootmem.h
@@ -103,17 +103,16 @@ extern void *__alloc_bootmem_low_node(pg_data_t *pgdat,
 	__alloc_bootmem(x, PAGE_SIZE, __pa(MAX_DMA_ADDRESS))
 #define alloc_bootmem_low_pages(x) \
 	__alloc_bootmem_low(x, PAGE_SIZE, 0)
-#endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
-
-extern int reserve_bootmem_generic(unsigned long addr, unsigned long size,
-				   int flags);
-
 #define alloc_bootmem_node(pgdat, x) \
 	__alloc_bootmem_node(pgdat, x, SMP_CACHE_BYTES, __pa(MAX_DMA_ADDRESS))
 #define alloc_bootmem_pages_node(pgdat, x) \
 	__alloc_bootmem_node(pgdat, x, PAGE_SIZE, __pa(MAX_DMA_ADDRESS))
 #define alloc_bootmem_low_pages_node(pgdat, x) \
 	__alloc_bootmem_low_node(pgdat, x, PAGE_SIZE, 0)
+#endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
+
+extern int reserve_bootmem_generic(unsigned long addr, unsigned long size,
+				   int flags);
 
 extern void *alloc_bootmem_section(unsigned long size,
 				   unsigned long section_nr);

From f1373da87be917e5b2356af44764620487376a07 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Fri, 25 Jul 2008 15:18:31 -0700
Subject: [PATCH 0955/1171] sparc: Wire up new system calls.

This wires up the recently added Wire up signalfd4, eventfd2,
epoll_create1, dup3, pipe2, and inotify_init1 system calls.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 arch/sparc/kernel/systbls.S   | 3 ++-
 arch/sparc64/kernel/systbls.S | 6 ++++--
 include/asm-sparc/unistd_32.h | 8 +++++++-
 include/asm-sparc/unistd_64.h | 8 +++++++-
 4 files changed, 20 insertions(+), 5 deletions(-)

diff --git a/arch/sparc/kernel/systbls.S b/arch/sparc/kernel/systbls.S
index 5a7c4c8345c3..e1b9233b90ab 100644
--- a/arch/sparc/kernel/systbls.S
+++ b/arch/sparc/kernel/systbls.S
@@ -80,4 +80,5 @@ sys_call_table:
 /*300*/	.long sys_set_robust_list, sys_get_robust_list, sys_migrate_pages, sys_mbind, sys_get_mempolicy
 /*305*/	.long sys_set_mempolicy, sys_kexec_load, sys_move_pages, sys_getcpu, sys_epoll_pwait
 /*310*/	.long sys_utimensat, sys_signalfd, sys_timerfd_create, sys_eventfd, sys_fallocate
-/*315*/	.long sys_timerfd_settime, sys_timerfd_gettime
+/*315*/	.long sys_timerfd_settime, sys_timerfd_gettime, sys_signalfd4, sys_eventfd2, sys_epoll_create1
+/*320*/	.long sys_dup3, sys_pipe2, sys_inotify_init1
diff --git a/arch/sparc64/kernel/systbls.S b/arch/sparc64/kernel/systbls.S
index 8b5282d433c4..1095bf4c5100 100644
--- a/arch/sparc64/kernel/systbls.S
+++ b/arch/sparc64/kernel/systbls.S
@@ -81,7 +81,8 @@ sys_call_table32:
 /*300*/	.word compat_sys_set_robust_list, compat_sys_get_robust_list, compat_sys_migrate_pages, compat_sys_mbind, compat_sys_get_mempolicy
 	.word compat_sys_set_mempolicy, compat_sys_kexec_load, compat_sys_move_pages, sys_getcpu, compat_sys_epoll_pwait
 /*310*/	.word compat_sys_utimensat, compat_sys_signalfd, sys_timerfd_create, sys_eventfd, compat_sys_fallocate
-	.word compat_sys_timerfd_settime, compat_sys_timerfd_gettime
+	.word compat_sys_timerfd_settime, compat_sys_timerfd_gettime, compat_sys_signalfd4, sys_eventfd2, sys_epoll_create1
+/*320*/	.word sys_dup3, sys_pipe2, sys_inotify_init1
 
 #endif /* CONFIG_COMPAT */
 
@@ -154,4 +155,5 @@ sys_call_table:
 /*300*/	.word sys_set_robust_list, sys_get_robust_list, sys_migrate_pages, sys_mbind, sys_get_mempolicy
 	.word sys_set_mempolicy, sys_kexec_load, sys_move_pages, sys_getcpu, sys_epoll_pwait
 /*310*/	.word sys_utimensat, sys_signalfd, sys_timerfd_create, sys_eventfd, sys_fallocate
-	.word sys_timerfd_settime, sys_timerfd_gettime
+	.word sys_timerfd_settime, sys_timerfd_gettime, sys_signalfd4, sys_eventfd2, sys_epoll_create1
+/*320*/	.word sys_dup3, sys_pipe2, sys_inotify_init1
diff --git a/include/asm-sparc/unistd_32.h b/include/asm-sparc/unistd_32.h
index 2338a0276377..648643a9f139 100644
--- a/include/asm-sparc/unistd_32.h
+++ b/include/asm-sparc/unistd_32.h
@@ -332,8 +332,14 @@
 #define __NR_fallocate		314
 #define __NR_timerfd_settime	315
 #define __NR_timerfd_gettime	316
+#define __NR_signalfd4		317
+#define __NR_eventfd2		318
+#define __NR_epoll_create1	319
+#define __NR_dup3		320
+#define __NR_pipe2		321
+#define __NR_inotify_init1	322
 
-#define NR_SYSCALLS		317
+#define NR_SYSCALLS		323
 
 /* Sparc 32-bit only has the "setresuid32", "getresuid32" variants,
  * it never had the plain ones and there is no value to adding those
diff --git a/include/asm-sparc/unistd_64.h b/include/asm-sparc/unistd_64.h
index 13be4453a1f0..c5cc0e052321 100644
--- a/include/asm-sparc/unistd_64.h
+++ b/include/asm-sparc/unistd_64.h
@@ -334,8 +334,14 @@
 #define __NR_fallocate		314
 #define __NR_timerfd_settime	315
 #define __NR_timerfd_gettime	316
+#define __NR_signalfd4		317
+#define __NR_eventfd2		318
+#define __NR_epoll_create1	319
+#define __NR_dup3		320
+#define __NR_pipe2		321
+#define __NR_inotify_init1	322
 
-#define NR_SYSCALLS		317
+#define NR_SYSCALLS		323
 
 #ifdef __KERNEL__
 #define __ARCH_WANT_IPC_PARSE_VERSION

From fc532f810832beb3306b42526a78f411972281c7 Mon Sep 17 00:00:00 2001
From: Nathan Lynch <ntl@pobox.com>
Date: Fri, 25 Jul 2008 17:50:30 -0500
Subject: [PATCH 0956/1171] powerpc: Fix boot problem due to AT_BASE_PLATFORM
 change

Commit 9115d13453dee22473a1e8cacc90a8d64a9c4bc9 ("powerpc: Enable
AT_BASE_PLATFORM aux vector") broke boot on 32-bit powerpc systems; we
have to use PTRRELOC to initialize powerpc_base_platform this early in
boot.

Bug reported by Jon Smirl.

Signed-off-by: Nathan Lynch <ntl@pobox.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
 arch/powerpc/kernel/cputable.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/powerpc/kernel/cputable.c b/arch/powerpc/kernel/cputable.c
index 25a052c16754..25c273c761d1 100644
--- a/arch/powerpc/kernel/cputable.c
+++ b/arch/powerpc/kernel/cputable.c
@@ -1660,8 +1660,8 @@ struct cpu_spec * __init identify_cpu(unsigned long offset, unsigned int pvr)
 			 * Set the base platform string once; assumes
 			 * we're called with real pvr first.
 			 */
-			if (powerpc_base_platform == NULL)
-				powerpc_base_platform = t->platform;
+			if (*PTRRELOC(&powerpc_base_platform) == NULL)
+				*PTRRELOC(&powerpc_base_platform) = t->platform;
 
 #if defined(CONFIG_PPC64) || defined(CONFIG_BOOKE)
 			/* ppc64 and booke expect identify_cpu to also call

From b4615e69b6c6353878b734a8202b65efbc554df4 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Fri, 25 Jul 2008 13:19:22 -0700
Subject: [PATCH 0957/1171] sys_paccept definition missing __user annotation

Introduced by commit aaca0bdca573f3f51ea03139f9c7289541e7bca3 ("flag
parameters: paccept"):

  net/socket.c:1515:17: error: symbol 'sys_paccept' redeclared with different type (originally declared at include/linux/syscalls.h:413) - incompatible argument 4 (different address spaces)

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/syscalls.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 06f2bf76c030..d6ff145919ca 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -411,7 +411,7 @@ asmlinkage long sys_bind(int, struct sockaddr __user *, int);
 asmlinkage long sys_connect(int, struct sockaddr __user *, int);
 asmlinkage long sys_accept(int, struct sockaddr __user *, int __user *);
 asmlinkage long sys_paccept(int, struct sockaddr __user *, int __user *,
-			    const sigset_t *, size_t, int);
+			    const __user sigset_t *, size_t, int);
 asmlinkage long sys_getsockname(int, struct sockaddr __user *, int __user *);
 asmlinkage long sys_getpeername(int, struct sockaddr __user *, int __user *);
 asmlinkage long sys_send(int, void __user *, size_t, unsigned);

From 8d25b36b77fe32c296ece83e94ca6ae4d17f3e25 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Sat, 26 Jul 2008 02:38:00 +0300
Subject: [PATCH 0958/1171] MFD_TC6393XB is ARM-only
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Compile error on other architectures:

    CC      drivers/mfd/tc6393xb.o
  /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/mfd/tc6393xb.c: In function ‘tc6393xb_attach_irq’:
  /home/bunk/linux/kernel-2.6/git/linux-2.6/drivers/mfd/tc6393xb.c:324: error: implicit declaration of function ‘set_irq_flags’
  ...

Reported-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/mfd/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mfd/Kconfig b/drivers/mfd/Kconfig
index 1f57a99fd968..883e7ea31de2 100644
--- a/drivers/mfd/Kconfig
+++ b/drivers/mfd/Kconfig
@@ -52,7 +52,7 @@ config HTC_PASIC3
 
 config MFD_TC6393XB
 	bool "Support Toshiba TC6393XB"
-	depends on GPIOLIB
+	depends on GPIOLIB && ARM
 	select MFD_CORE
 	help
 	  Support for Toshiba Mobile IO Controller TC6393XB

From 024e8ac04453b3525448c31ef39848cf675ba6db Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 20:00:10 -0700
Subject: [PATCH 0959/1171] x86_64: fix ia32 AMD syscall audit fast-path

The new code in commit 5cbf1565f29eb57a86a305b08836613508e294d7
has a bug in the version supporting the AMD 'syscall' instruction.
It clobbers the user's %ecx register value (with the %ebp value).

This change fixes it.

Signed-off-by: Roland McGrath <roland@redhat.com>
---
 arch/x86/ia32/ia32entry.S | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S
index e4bd1793a5e4..ffc1bb4fed7d 100644
--- a/arch/x86/ia32/ia32entry.S
+++ b/arch/x86/ia32/ia32entry.S
@@ -201,7 +201,7 @@ sysexit_from_sys_call:
 	movl RDI-ARGOFFSET(%rsp),%r8d	/* reload 5th syscall arg */
 	.endm
 
-	.macro auditsys_exit exit
+	.macro auditsys_exit exit,ebpsave=RBP
 	testl $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT),TI_flags(%r10)
 	jnz int_ret_from_sys_call
 	TRACE_IRQS_ON
@@ -214,7 +214,7 @@ sysexit_from_sys_call:
 	call audit_syscall_exit
 	GET_THREAD_INFO(%r10)
 	movl RAX-ARGOFFSET(%rsp),%eax	/* reload syscall return value */
-	movl RBP-ARGOFFSET(%rsp),%ebp	/* reload user register value */
+	movl \ebpsave-ARGOFFSET(%rsp),%ebp /* reload user register value */
 	movl $(_TIF_ALLWORK_MASK & ~_TIF_SYSCALL_AUDIT),%edi
 	cli
 	TRACE_IRQS_OFF
@@ -347,7 +347,7 @@ cstar_auditsys:
 	jmp cstar_dispatch
 
 sysretl_audit:
-	auditsys_exit sysretl_from_sys_call
+	auditsys_exit sysretl_from_sys_call, RCX /* user %ebp in RCX slot */
 #endif
 
 cstar_tracesys:

From 53e5e96ec18da6f65e89f05674711e1c93d8df67 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ilpo=20J=C3=A4rvinen?= <ilpo.jarvinen@helsinki.fi>
Date: Fri, 25 Jul 2008 21:40:45 -0700
Subject: [PATCH 0960/1171] drivers/net: convert BUG_TRAP to generic WARN_ON
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Removes legacy reinvent-the-wheel type thing. The generic
machinery integrates much better to automated debugging aids
such as kerneloops.org (and others), and is unambiguous due to
better naming. Non-intuively BUG_TRAP() is actually equal to
WARN_ON() rather than BUG_ON().

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/net/bnx2x_main.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/net/bnx2x_main.c b/drivers/net/bnx2x_main.c
index 0263bef9cc6d..93c95037c193 100644
--- a/drivers/net/bnx2x_main.c
+++ b/drivers/net/bnx2x_main.c
@@ -814,7 +814,7 @@ static u16 bnx2x_free_tx_pkt(struct bnx2x *bp, struct bnx2x_fastpath *fp,
 	}
 
 	/* release skb */
-	BUG_TRAP(skb);
+	WARN_ON(!skb);
 	dev_kfree_skb(skb);
 	tx_buf->first_bd = 0;
 	tx_buf->skb = NULL;
@@ -837,9 +837,9 @@ static inline u16 bnx2x_tx_avail(struct bnx2x_fastpath *fp)
 	used = SUB_S16(prod, cons) + (s16)NUM_TX_RINGS;
 
 #ifdef BNX2X_STOP_ON_ERROR
-	BUG_TRAP(used >= 0);
-	BUG_TRAP(used <= fp->bp->tx_ring_size);
-	BUG_TRAP((fp->bp->tx_ring_size - used) <= MAX_TX_AVAIL);
+	WARN_ON(used < 0);
+	WARN_ON(used > fp->bp->tx_ring_size);
+	WARN_ON((fp->bp->tx_ring_size - used) > MAX_TX_AVAIL);
 #endif
 
 	return (s16)(fp->bp->tx_ring_size) - used;
@@ -4374,7 +4374,7 @@ static void bnx2x_init_rx_rings(struct bnx2x *bp)
 			}
 			ring_prod = NEXT_RX_IDX(ring_prod);
 			cqe_ring_prod = NEXT_RCQ_IDX(cqe_ring_prod);
-			BUG_TRAP(ring_prod > i);
+			WARN_ON(ring_prod <= i);
 		}
 
 		fp->rx_bd_prod = ring_prod;

From 547b792cac0a038b9dbf958d3c120df3740b5572 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ilpo=20J=C3=A4rvinen?= <ilpo.jarvinen@helsinki.fi>
Date: Fri, 25 Jul 2008 21:43:18 -0700
Subject: [PATCH 0961/1171] net: convert BUG_TRAP to generic WARN_ON
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Removes legacy reinvent-the-wheel type thing. The generic
machinery integrates much better to automated debugging aids
such as kerneloops.org (and others), and is unambiguous due to
better naming. Non-intuively BUG_TRAP() is actually equal to
WARN_ON() rather than BUG_ON() though some might actually be
promoted to BUG_ON() but I left that to future.

I could make at least one BUILD_BUG_ON conversion.

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/net/request_sock.h              |  5 ++--
 net/appletalk/ddp.c                     |  4 ++--
 net/core/datagram.c                     |  8 +++----
 net/core/dev.c                          | 10 ++++----
 net/core/request_sock.c                 |  2 +-
 net/core/skbuff.c                       | 20 ++++++++--------
 net/core/stream.c                       |  6 ++---
 net/core/user_dma.c                     |  5 ++--
 net/dccp/dccp.h                         |  2 +-
 net/dccp/input.c                        |  2 +-
 net/dccp/ipv4.c                         |  2 +-
 net/dccp/ipv6.c                         |  2 +-
 net/dccp/proto.c                        |  4 ++--
 net/dccp/timer.c                        |  2 +-
 net/ipv4/af_inet.c                      | 14 +++++------
 net/ipv4/devinet.c                      |  6 ++---
 net/ipv4/inet_connection_sock.c         | 18 +++++++-------
 net/ipv4/inet_fragment.c                |  4 ++--
 net/ipv4/inet_hashtables.c              |  8 +++----
 net/ipv4/inet_timewait_sock.c           |  2 +-
 net/ipv4/ip_fragment.c                  |  4 ++--
 net/ipv4/ip_output.c                    |  2 +-
 net/ipv4/tcp.c                          | 12 +++++-----
 net/ipv4/tcp_input.c                    | 20 ++++++++--------
 net/ipv4/tcp_ipv4.c                     |  2 +-
 net/ipv4/tcp_timer.c                    |  2 +-
 net/ipv6/addrconf.c                     | 11 +++++----
 net/ipv6/af_inet6.c                     |  2 +-
 net/ipv6/inet6_connection_sock.c        |  2 +-
 net/ipv6/inet6_hashtables.c             |  4 ++--
 net/ipv6/ip6_fib.c                      | 31 +++++++++++++------------
 net/ipv6/ip6_output.c                   |  2 +-
 net/ipv6/mip6.c                         |  8 +++----
 net/ipv6/netfilter/nf_conntrack_reasm.c |  4 ++--
 net/ipv6/reassembly.c                   |  4 ++--
 net/ipv6/tcp_ipv6.c                     |  2 +-
 net/key/af_key.c                        |  4 ++--
 net/netlink/af_netlink.c                |  7 +++---
 net/packet/af_packet.c                  |  4 ++--
 net/rxrpc/af_rxrpc.c                    |  6 ++---
 net/sched/act_api.c                     |  2 +-
 net/sched/act_police.c                  |  2 +-
 net/sched/cls_u32.c                     | 10 ++++----
 net/sched/sch_cbq.c                     |  4 ++--
 net/sched/sch_generic.c                 |  2 +-
 net/sched/sch_htb.c                     | 16 ++++++-------
 net/sctp/associola.c                    |  2 +-
 net/unix/af_unix.c                      |  8 +++----
 net/xfrm/xfrm_algo.c                    |  4 ++--
 net/xfrm/xfrm_ipcomp.c                  |  3 +--
 net/xfrm/xfrm_state.c                   |  2 +-
 51 files changed, 159 insertions(+), 155 deletions(-)

diff --git a/include/net/request_sock.h b/include/net/request_sock.h
index 0c96e7bed5db..8d6e991ef4df 100644
--- a/include/net/request_sock.h
+++ b/include/net/request_sock.h
@@ -18,6 +18,7 @@
 #include <linux/slab.h>
 #include <linux/spinlock.h>
 #include <linux/types.h>
+#include <linux/bug.h>
 
 #include <net/sock.h>
 
@@ -170,7 +171,7 @@ static inline struct request_sock *reqsk_queue_remove(struct request_sock_queue
 {
 	struct request_sock *req = queue->rskq_accept_head;
 
-	BUG_TRAP(req != NULL);
+	WARN_ON(req == NULL);
 
 	queue->rskq_accept_head = req->dl_next;
 	if (queue->rskq_accept_head == NULL)
@@ -185,7 +186,7 @@ static inline struct sock *reqsk_queue_get_child(struct request_sock_queue *queu
 	struct request_sock *req = reqsk_queue_remove(queue);
 	struct sock *child = req->sk;
 
-	BUG_TRAP(child != NULL);
+	WARN_ON(child == NULL);
 
 	sk_acceptq_removed(parent);
 	__reqsk_free(req);
diff --git a/net/appletalk/ddp.c b/net/appletalk/ddp.c
index 07b5b82c5eab..0c850427a85b 100644
--- a/net/appletalk/ddp.c
+++ b/net/appletalk/ddp.c
@@ -959,7 +959,7 @@ static unsigned long atalk_sum_skb(const struct sk_buff *skb, int offset,
 	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
 		int end;
 
-		BUG_TRAP(start <= offset + len);
+		WARN_ON(start > offset + len);
 
 		end = start + skb_shinfo(skb)->frags[i].size;
 		if ((copy = end - offset) > 0) {
@@ -986,7 +986,7 @@ static unsigned long atalk_sum_skb(const struct sk_buff *skb, int offset,
 		for (; list; list = list->next) {
 			int end;
 
-			BUG_TRAP(start <= offset + len);
+			WARN_ON(start > offset + len);
 
 			end = start + list->len;
 			if ((copy = end - offset) > 0) {
diff --git a/net/core/datagram.c b/net/core/datagram.c
index 8a28fc93b724..dd61dcad6019 100644
--- a/net/core/datagram.c
+++ b/net/core/datagram.c
@@ -285,7 +285,7 @@ int skb_copy_datagram_iovec(const struct sk_buff *skb, int offset,
 	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
 		int end;
 
-		BUG_TRAP(start <= offset + len);
+		WARN_ON(start > offset + len);
 
 		end = start + skb_shinfo(skb)->frags[i].size;
 		if ((copy = end - offset) > 0) {
@@ -315,7 +315,7 @@ int skb_copy_datagram_iovec(const struct sk_buff *skb, int offset,
 		for (; list; list = list->next) {
 			int end;
 
-			BUG_TRAP(start <= offset + len);
+			WARN_ON(start > offset + len);
 
 			end = start + list->len;
 			if ((copy = end - offset) > 0) {
@@ -366,7 +366,7 @@ static int skb_copy_and_csum_datagram(const struct sk_buff *skb, int offset,
 	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
 		int end;
 
-		BUG_TRAP(start <= offset + len);
+		WARN_ON(start > offset + len);
 
 		end = start + skb_shinfo(skb)->frags[i].size;
 		if ((copy = end - offset) > 0) {
@@ -402,7 +402,7 @@ static int skb_copy_and_csum_datagram(const struct sk_buff *skb, int offset,
 		for (; list; list=list->next) {
 			int end;
 
-			BUG_TRAP(start <= offset + len);
+			WARN_ON(start > offset + len);
 
 			end = start + list->len;
 			if ((copy = end - offset) > 0) {
diff --git a/net/core/dev.c b/net/core/dev.c
index ccf97f9f37eb..c6f9c83745e6 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1973,7 +1973,7 @@ static void net_tx_action(struct softirq_action *h)
 			struct sk_buff *skb = clist;
 			clist = clist->next;
 
-			BUG_TRAP(!atomic_read(&skb->users));
+			WARN_ON(atomic_read(&skb->users));
 			__kfree_skb(skb);
 		}
 	}
@@ -3847,7 +3847,7 @@ static void rollback_registered(struct net_device *dev)
 		dev->uninit(dev);
 
 	/* Notifier chain MUST detach us from master device. */
-	BUG_TRAP(!dev->master);
+	WARN_ON(dev->master);
 
 	/* Remove entries from kobject tree */
 	netdev_unregister_kobject(dev);
@@ -4169,9 +4169,9 @@ void netdev_run_todo(void)
 
 		/* paranoia */
 		BUG_ON(atomic_read(&dev->refcnt));
-		BUG_TRAP(!dev->ip_ptr);
-		BUG_TRAP(!dev->ip6_ptr);
-		BUG_TRAP(!dev->dn_ptr);
+		WARN_ON(dev->ip_ptr);
+		WARN_ON(dev->ip6_ptr);
+		WARN_ON(dev->dn_ptr);
 
 		if (dev->destructor)
 			dev->destructor(dev);
diff --git a/net/core/request_sock.c b/net/core/request_sock.c
index 2d3035d3abd7..7552495aff7a 100644
--- a/net/core/request_sock.c
+++ b/net/core/request_sock.c
@@ -123,7 +123,7 @@ void reqsk_queue_destroy(struct request_sock_queue *queue)
 		}
 	}
 
-	BUG_TRAP(lopt->qlen == 0);
+	WARN_ON(lopt->qlen != 0);
 	if (lopt_size > PAGE_SIZE)
 		vfree(lopt);
 	else
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index e4115672b6cf..4e0c92274189 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -1200,7 +1200,7 @@ int skb_copy_bits(const struct sk_buff *skb, int offset, void *to, int len)
 	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
 		int end;
 
-		BUG_TRAP(start <= offset + len);
+		WARN_ON(start > offset + len);
 
 		end = start + skb_shinfo(skb)->frags[i].size;
 		if ((copy = end - offset) > 0) {
@@ -1229,7 +1229,7 @@ int skb_copy_bits(const struct sk_buff *skb, int offset, void *to, int len)
 		for (; list; list = list->next) {
 			int end;
 
-			BUG_TRAP(start <= offset + len);
+			WARN_ON(start > offset + len);
 
 			end = start + list->len;
 			if ((copy = end - offset) > 0) {
@@ -1475,7 +1475,7 @@ int skb_store_bits(struct sk_buff *skb, int offset, const void *from, int len)
 		skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
 		int end;
 
-		BUG_TRAP(start <= offset + len);
+		WARN_ON(start > offset + len);
 
 		end = start + frag->size;
 		if ((copy = end - offset) > 0) {
@@ -1503,7 +1503,7 @@ int skb_store_bits(struct sk_buff *skb, int offset, const void *from, int len)
 		for (; list; list = list->next) {
 			int end;
 
-			BUG_TRAP(start <= offset + len);
+			WARN_ON(start > offset + len);
 
 			end = start + list->len;
 			if ((copy = end - offset) > 0) {
@@ -1552,7 +1552,7 @@ __wsum skb_checksum(const struct sk_buff *skb, int offset,
 	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
 		int end;
 
-		BUG_TRAP(start <= offset + len);
+		WARN_ON(start > offset + len);
 
 		end = start + skb_shinfo(skb)->frags[i].size;
 		if ((copy = end - offset) > 0) {
@@ -1581,7 +1581,7 @@ __wsum skb_checksum(const struct sk_buff *skb, int offset,
 		for (; list; list = list->next) {
 			int end;
 
-			BUG_TRAP(start <= offset + len);
+			WARN_ON(start > offset + len);
 
 			end = start + list->len;
 			if ((copy = end - offset) > 0) {
@@ -1629,7 +1629,7 @@ __wsum skb_copy_and_csum_bits(const struct sk_buff *skb, int offset,
 	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
 		int end;
 
-		BUG_TRAP(start <= offset + len);
+		WARN_ON(start > offset + len);
 
 		end = start + skb_shinfo(skb)->frags[i].size;
 		if ((copy = end - offset) > 0) {
@@ -1662,7 +1662,7 @@ __wsum skb_copy_and_csum_bits(const struct sk_buff *skb, int offset,
 			__wsum csum2;
 			int end;
 
-			BUG_TRAP(start <= offset + len);
+			WARN_ON(start > offset + len);
 
 			end = start + list->len;
 			if ((copy = end - offset) > 0) {
@@ -2373,7 +2373,7 @@ __skb_to_sgvec(struct sk_buff *skb, struct scatterlist *sg, int offset, int len)
 	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
 		int end;
 
-		BUG_TRAP(start <= offset + len);
+		WARN_ON(start > offset + len);
 
 		end = start + skb_shinfo(skb)->frags[i].size;
 		if ((copy = end - offset) > 0) {
@@ -2397,7 +2397,7 @@ __skb_to_sgvec(struct sk_buff *skb, struct scatterlist *sg, int offset, int len)
 		for (; list; list = list->next) {
 			int end;
 
-			BUG_TRAP(start <= offset + len);
+			WARN_ON(start > offset + len);
 
 			end = start + list->len;
 			if ((copy = end - offset) > 0) {
diff --git a/net/core/stream.c b/net/core/stream.c
index 4a0ad152c9c4..a6b3437ff082 100644
--- a/net/core/stream.c
+++ b/net/core/stream.c
@@ -192,13 +192,13 @@ void sk_stream_kill_queues(struct sock *sk)
 	__skb_queue_purge(&sk->sk_error_queue);
 
 	/* Next, the write queue. */
-	BUG_TRAP(skb_queue_empty(&sk->sk_write_queue));
+	WARN_ON(!skb_queue_empty(&sk->sk_write_queue));
 
 	/* Account for returned memory. */
 	sk_mem_reclaim(sk);
 
-	BUG_TRAP(!sk->sk_wmem_queued);
-	BUG_TRAP(!sk->sk_forward_alloc);
+	WARN_ON(sk->sk_wmem_queued);
+	WARN_ON(sk->sk_forward_alloc);
 
 	/* It is _impossible_ for the backlog to contain anything
 	 * when we get here.  All user references to this socket
diff --git a/net/core/user_dma.c b/net/core/user_dma.c
index c77aff9c6eb3..53c6b67b2877 100644
--- a/net/core/user_dma.c
+++ b/net/core/user_dma.c
@@ -27,7 +27,6 @@
 
 #include <linux/dmaengine.h>
 #include <linux/socket.h>
-#include <linux/rtnetlink.h> /* for BUG_TRAP */
 #include <net/tcp.h>
 #include <net/netdma.h>
 
@@ -71,7 +70,7 @@ int dma_skb_copy_datagram_iovec(struct dma_chan *chan,
 	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
 		int end;
 
-		BUG_TRAP(start <= offset + len);
+		WARN_ON(start > offset + len);
 
 		end = start + skb_shinfo(skb)->frags[i].size;
 		copy = end - offset;
@@ -100,7 +99,7 @@ int dma_skb_copy_datagram_iovec(struct dma_chan *chan,
 		for (; list; list = list->next) {
 			int end;
 
-			BUG_TRAP(start <= offset + len);
+			WARN_ON(start > offset + len);
 
 			end = start + list->len;
 			copy = end - offset;
diff --git a/net/dccp/dccp.h b/net/dccp/dccp.h
index 32617e0576cb..743d85fcd651 100644
--- a/net/dccp/dccp.h
+++ b/net/dccp/dccp.h
@@ -164,7 +164,7 @@ static inline bool dccp_loss_free(const u64 s1, const u64 s2, const u64 ndp)
 {
 	s64 delta = dccp_delta_seqno(s1, s2);
 
-	BUG_TRAP(delta >= 0);
+	WARN_ON(delta < 0);
 	return (u64)delta <= ndp + 1;
 }
 
diff --git a/net/dccp/input.c b/net/dccp/input.c
index 08392ed86c25..df2f110df94a 100644
--- a/net/dccp/input.c
+++ b/net/dccp/input.c
@@ -413,7 +413,7 @@ static int dccp_rcv_request_sent_state_process(struct sock *sk,
 
 		/* Stop the REQUEST timer */
 		inet_csk_clear_xmit_timer(sk, ICSK_TIME_RETRANS);
-		BUG_TRAP(sk->sk_send_head != NULL);
+		WARN_ON(sk->sk_send_head == NULL);
 		__kfree_skb(sk->sk_send_head);
 		sk->sk_send_head = NULL;
 
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 2622ace17c46..a835b88237cb 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -283,7 +283,7 @@ static void dccp_v4_err(struct sk_buff *skb, u32 info)
 		 * ICMPs are not backlogged, hence we cannot get an established
 		 * socket here.
 		 */
-		BUG_TRAP(!req->sk);
+		WARN_ON(req->sk);
 
 		if (seq != dccp_rsk(req)->dreq_iss) {
 			NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index b74e8b2cbe55..da509127e00c 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -186,7 +186,7 @@ static void dccp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 		 * ICMPs are not backlogged, hence we cannot get an established
 		 * socket here.
 		 */
-		BUG_TRAP(req->sk == NULL);
+		WARN_ON(req->sk != NULL);
 
 		if (seq != dccp_rsk(req)->dreq_iss) {
 			NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
diff --git a/net/dccp/proto.c b/net/dccp/proto.c
index a0b56009611f..b622d9744856 100644
--- a/net/dccp/proto.c
+++ b/net/dccp/proto.c
@@ -327,7 +327,7 @@ int dccp_disconnect(struct sock *sk, int flags)
 	inet_csk_delack_init(sk);
 	__sk_dst_reset(sk);
 
-	BUG_TRAP(!inet->num || icsk->icsk_bind_hash);
+	WARN_ON(inet->num && !icsk->icsk_bind_hash);
 
 	sk->sk_error_report(sk);
 	return err;
@@ -981,7 +981,7 @@ adjudge_to_death:
 	 */
 	local_bh_disable();
 	bh_lock_sock(sk);
-	BUG_TRAP(!sock_owned_by_user(sk));
+	WARN_ON(sock_owned_by_user(sk));
 
 	/* Have we already been destroyed by a softirq or backlog? */
 	if (state != DCCP_CLOSED && sk->sk_state == DCCP_CLOSED)
diff --git a/net/dccp/timer.c b/net/dccp/timer.c
index 3608d5342ca2..6a5b961b6f5c 100644
--- a/net/dccp/timer.c
+++ b/net/dccp/timer.c
@@ -106,7 +106,7 @@ static void dccp_retransmit_timer(struct sock *sk)
 	 *	-- Acks     in client-PARTOPEN state (sec. 8.1.5)
 	 *	-- CloseReq in server-CLOSEREQ state (sec. 8.3)
 	 *	-- Close    in   node-CLOSING  state (sec. 8.3)                */
-	BUG_TRAP(sk->sk_send_head != NULL);
+	WARN_ON(sk->sk_send_head == NULL);
 
 	/*
 	 * More than than 4MSL (8 minutes) has passed, a RESET(aborted) was
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index dd919d84285f..a107f49eea41 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -148,10 +148,10 @@ void inet_sock_destruct(struct sock *sk)
 		return;
 	}
 
-	BUG_TRAP(!atomic_read(&sk->sk_rmem_alloc));
-	BUG_TRAP(!atomic_read(&sk->sk_wmem_alloc));
-	BUG_TRAP(!sk->sk_wmem_queued);
-	BUG_TRAP(!sk->sk_forward_alloc);
+	WARN_ON(atomic_read(&sk->sk_rmem_alloc));
+	WARN_ON(atomic_read(&sk->sk_wmem_alloc));
+	WARN_ON(sk->sk_wmem_queued);
+	WARN_ON(sk->sk_forward_alloc);
 
 	kfree(inet->opt);
 	dst_release(sk->sk_dst_cache);
@@ -341,7 +341,7 @@ lookup_protocol:
 	answer_flags = answer->flags;
 	rcu_read_unlock();
 
-	BUG_TRAP(answer_prot->slab != NULL);
+	WARN_ON(answer_prot->slab == NULL);
 
 	err = -ENOBUFS;
 	sk = sk_alloc(net, PF_INET, GFP_KERNEL, answer_prot);
@@ -661,8 +661,8 @@ int inet_accept(struct socket *sock, struct socket *newsock, int flags)
 
 	lock_sock(sk2);
 
-	BUG_TRAP((1 << sk2->sk_state) &
-		 (TCPF_ESTABLISHED | TCPF_CLOSE_WAIT | TCPF_CLOSE));
+	WARN_ON(!((1 << sk2->sk_state) &
+		  (TCPF_ESTABLISHED | TCPF_CLOSE_WAIT | TCPF_CLOSE)));
 
 	sock_graft(sk2, newsock);
 
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index 2e667e2f90df..91d3d96805d0 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -138,8 +138,8 @@ void in_dev_finish_destroy(struct in_device *idev)
 {
 	struct net_device *dev = idev->dev;
 
-	BUG_TRAP(!idev->ifa_list);
-	BUG_TRAP(!idev->mc_list);
+	WARN_ON(idev->ifa_list);
+	WARN_ON(idev->mc_list);
 #ifdef NET_REFCNT_DEBUG
 	printk(KERN_DEBUG "in_dev_finish_destroy: %p=%s\n",
 	       idev, dev ? dev->name : "NIL");
@@ -399,7 +399,7 @@ static int inet_set_ifa(struct net_device *dev, struct in_ifaddr *ifa)
 	}
 	ipv4_devconf_setall(in_dev);
 	if (ifa->ifa_dev != in_dev) {
-		BUG_TRAP(!ifa->ifa_dev);
+		WARN_ON(ifa->ifa_dev);
 		in_dev_hold(in_dev);
 		ifa->ifa_dev = in_dev;
 	}
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index bb81c958b744..0c1ae68ee84b 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -167,7 +167,7 @@ tb_not_found:
 success:
 	if (!inet_csk(sk)->icsk_bind_hash)
 		inet_bind_hash(sk, tb, snum);
-	BUG_TRAP(inet_csk(sk)->icsk_bind_hash == tb);
+	WARN_ON(inet_csk(sk)->icsk_bind_hash != tb);
 	ret = 0;
 
 fail_unlock:
@@ -260,7 +260,7 @@ struct sock *inet_csk_accept(struct sock *sk, int flags, int *err)
 	}
 
 	newsk = reqsk_queue_get_child(&icsk->icsk_accept_queue, sk);
-	BUG_TRAP(newsk->sk_state != TCP_SYN_RECV);
+	WARN_ON(newsk->sk_state == TCP_SYN_RECV);
 out:
 	release_sock(sk);
 	return newsk;
@@ -386,7 +386,7 @@ struct request_sock *inet_csk_search_req(const struct sock *sk,
 		    ireq->rmt_addr == raddr &&
 		    ireq->loc_addr == laddr &&
 		    AF_INET_FAMILY(req->rsk_ops->family)) {
-			BUG_TRAP(!req->sk);
+			WARN_ON(req->sk);
 			*prevp = prev;
 			break;
 		}
@@ -539,14 +539,14 @@ EXPORT_SYMBOL_GPL(inet_csk_clone);
  */
 void inet_csk_destroy_sock(struct sock *sk)
 {
-	BUG_TRAP(sk->sk_state == TCP_CLOSE);
-	BUG_TRAP(sock_flag(sk, SOCK_DEAD));
+	WARN_ON(sk->sk_state != TCP_CLOSE);
+	WARN_ON(!sock_flag(sk, SOCK_DEAD));
 
 	/* It cannot be in hash table! */
-	BUG_TRAP(sk_unhashed(sk));
+	WARN_ON(!sk_unhashed(sk));
 
 	/* If it has not 0 inet_sk(sk)->num, it must be bound */
-	BUG_TRAP(!inet_sk(sk)->num || inet_csk(sk)->icsk_bind_hash);
+	WARN_ON(inet_sk(sk)->num && !inet_csk(sk)->icsk_bind_hash);
 
 	sk->sk_prot->destroy(sk);
 
@@ -629,7 +629,7 @@ void inet_csk_listen_stop(struct sock *sk)
 
 		local_bh_disable();
 		bh_lock_sock(child);
-		BUG_TRAP(!sock_owned_by_user(child));
+		WARN_ON(sock_owned_by_user(child));
 		sock_hold(child);
 
 		sk->sk_prot->disconnect(child, O_NONBLOCK);
@@ -647,7 +647,7 @@ void inet_csk_listen_stop(struct sock *sk)
 		sk_acceptq_removed(sk);
 		__reqsk_free(req);
 	}
-	BUG_TRAP(!sk->sk_ack_backlog);
+	WARN_ON(sk->sk_ack_backlog);
 }
 
 EXPORT_SYMBOL_GPL(inet_csk_listen_stop);
diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c
index 0546a0bc97ea..6c52e08f786e 100644
--- a/net/ipv4/inet_fragment.c
+++ b/net/ipv4/inet_fragment.c
@@ -134,8 +134,8 @@ void inet_frag_destroy(struct inet_frag_queue *q, struct inet_frags *f,
 	struct sk_buff *fp;
 	struct netns_frags *nf;
 
-	BUG_TRAP(q->last_in & INET_FRAG_COMPLETE);
-	BUG_TRAP(del_timer(&q->timer) == 0);
+	WARN_ON(!(q->last_in & INET_FRAG_COMPLETE));
+	WARN_ON(del_timer(&q->timer) != 0);
 
 	/* Release all fragment data. */
 	fp = q->fragments;
diff --git a/net/ipv4/inet_hashtables.c b/net/ipv4/inet_hashtables.c
index 115f53722d20..44981906fb91 100644
--- a/net/ipv4/inet_hashtables.c
+++ b/net/ipv4/inet_hashtables.c
@@ -305,7 +305,7 @@ unique:
 	inet->num = lport;
 	inet->sport = htons(lport);
 	sk->sk_hash = hash;
-	BUG_TRAP(sk_unhashed(sk));
+	WARN_ON(!sk_unhashed(sk));
 	__sk_add_node(sk, &head->chain);
 	sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
 	write_unlock(lock);
@@ -342,7 +342,7 @@ void __inet_hash_nolisten(struct sock *sk)
 	rwlock_t *lock;
 	struct inet_ehash_bucket *head;
 
-	BUG_TRAP(sk_unhashed(sk));
+	WARN_ON(!sk_unhashed(sk));
 
 	sk->sk_hash = inet_sk_ehashfn(sk);
 	head = inet_ehash_bucket(hashinfo, sk->sk_hash);
@@ -367,7 +367,7 @@ static void __inet_hash(struct sock *sk)
 		return;
 	}
 
-	BUG_TRAP(sk_unhashed(sk));
+	WARN_ON(!sk_unhashed(sk));
 	list = &hashinfo->listening_hash[inet_sk_listen_hashfn(sk)];
 	lock = &hashinfo->lhash_lock;
 
@@ -450,7 +450,7 @@ int __inet_hash_connect(struct inet_timewait_death_row *death_row,
 			 */
 			inet_bind_bucket_for_each(tb, node, &head->chain) {
 				if (tb->ib_net == net && tb->port == port) {
-					BUG_TRAP(!hlist_empty(&tb->owners));
+					WARN_ON(hlist_empty(&tb->owners));
 					if (tb->fastreuse >= 0)
 						goto next_port;
 					if (!check_established(death_row, sk,
diff --git a/net/ipv4/inet_timewait_sock.c b/net/ipv4/inet_timewait_sock.c
index 75c2def8f9a0..d985bd613d25 100644
--- a/net/ipv4/inet_timewait_sock.c
+++ b/net/ipv4/inet_timewait_sock.c
@@ -86,7 +86,7 @@ void __inet_twsk_hashdance(struct inet_timewait_sock *tw, struct sock *sk,
 			hashinfo->bhash_size)];
 	spin_lock(&bhead->lock);
 	tw->tw_tb = icsk->icsk_bind_hash;
-	BUG_TRAP(icsk->icsk_bind_hash);
+	WARN_ON(!icsk->icsk_bind_hash);
 	inet_twsk_add_bind_node(tw, &tw->tw_tb->owners);
 	spin_unlock(&bhead->lock);
 
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index 38d38f058018..2152d222b954 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -488,8 +488,8 @@ static int ip_frag_reasm(struct ipq *qp, struct sk_buff *prev,
 		qp->q.fragments = head;
 	}
 
-	BUG_TRAP(head != NULL);
-	BUG_TRAP(FRAG_CB(head)->offset == 0);
+	WARN_ON(head == NULL);
+	WARN_ON(FRAG_CB(head)->offset != 0);
 
 	/* Allocate a new buffer for the datagram. */
 	ihlen = ip_hdrlen(head);
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 465544f6281a..d533a89e08de 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -118,7 +118,7 @@ static int ip_dev_loopback_xmit(struct sk_buff *newskb)
 	__skb_pull(newskb, skb_network_offset(newskb));
 	newskb->pkt_type = PACKET_LOOPBACK;
 	newskb->ip_summed = CHECKSUM_UNNECESSARY;
-	BUG_TRAP(newskb->dst);
+	WARN_ON(!newskb->dst);
 	netif_rx(newskb);
 	return 0;
 }
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 0b491bf03db4..1ab341e5d3e0 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -1096,7 +1096,7 @@ void tcp_cleanup_rbuf(struct sock *sk, int copied)
 #if TCP_DEBUG
 	struct sk_buff *skb = skb_peek(&sk->sk_receive_queue);
 
-	BUG_TRAP(!skb || before(tp->copied_seq, TCP_SKB_CB(skb)->end_seq));
+	WARN_ON(skb && !before(tp->copied_seq, TCP_SKB_CB(skb)->end_seq));
 #endif
 
 	if (inet_csk_ack_scheduled(sk)) {
@@ -1358,7 +1358,7 @@ int tcp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 				goto found_ok_skb;
 			if (tcp_hdr(skb)->fin)
 				goto found_fin_ok;
-			BUG_TRAP(flags & MSG_PEEK);
+			WARN_ON(!(flags & MSG_PEEK));
 			skb = skb->next;
 		} while (skb != (struct sk_buff *)&sk->sk_receive_queue);
 
@@ -1421,8 +1421,8 @@ int tcp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
 
 			tp->ucopy.len = len;
 
-			BUG_TRAP(tp->copied_seq == tp->rcv_nxt ||
-				 (flags & (MSG_PEEK | MSG_TRUNC)));
+			WARN_ON(tp->copied_seq != tp->rcv_nxt &&
+				!(flags & (MSG_PEEK | MSG_TRUNC)));
 
 			/* Ugly... If prequeue is not empty, we have to
 			 * process it before releasing socket, otherwise
@@ -1844,7 +1844,7 @@ adjudge_to_death:
 	 */
 	local_bh_disable();
 	bh_lock_sock(sk);
-	BUG_TRAP(!sock_owned_by_user(sk));
+	WARN_ON(sock_owned_by_user(sk));
 
 	/* Have we already been destroyed by a softirq or backlog? */
 	if (state != TCP_CLOSE && sk->sk_state == TCP_CLOSE)
@@ -1973,7 +1973,7 @@ int tcp_disconnect(struct sock *sk, int flags)
 	memset(&tp->rx_opt, 0, sizeof(tp->rx_opt));
 	__sk_dst_reset(sk);
 
-	BUG_TRAP(!inet->num || icsk->icsk_bind_hash);
+	WARN_ON(inet->num && !icsk->icsk_bind_hash);
 
 	sk->sk_error_report(sk);
 	return err;
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 75efd244f2af..67ccce2a96bd 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -1629,10 +1629,10 @@ advance_sp:
 out:
 
 #if FASTRETRANS_DEBUG > 0
-	BUG_TRAP((int)tp->sacked_out >= 0);
-	BUG_TRAP((int)tp->lost_out >= 0);
-	BUG_TRAP((int)tp->retrans_out >= 0);
-	BUG_TRAP((int)tcp_packets_in_flight(tp) >= 0);
+	WARN_ON((int)tp->sacked_out < 0);
+	WARN_ON((int)tp->lost_out < 0);
+	WARN_ON((int)tp->retrans_out < 0);
+	WARN_ON((int)tcp_packets_in_flight(tp) < 0);
 #endif
 	return flag;
 }
@@ -2181,7 +2181,7 @@ static void tcp_mark_head_lost(struct sock *sk, int packets)
 	int err;
 	unsigned int mss;
 
-	BUG_TRAP(packets <= tp->packets_out);
+	WARN_ON(packets > tp->packets_out);
 	if (tp->lost_skb_hint) {
 		skb = tp->lost_skb_hint;
 		cnt = tp->lost_cnt_hint;
@@ -2610,7 +2610,7 @@ static void tcp_fastretrans_alert(struct sock *sk, int pkts_acked, int flag)
 	/* E. Check state exit conditions. State can be terminated
 	 *    when high_seq is ACKed. */
 	if (icsk->icsk_ca_state == TCP_CA_Open) {
-		BUG_TRAP(tp->retrans_out == 0);
+		WARN_ON(tp->retrans_out != 0);
 		tp->retrans_stamp = 0;
 	} else if (!before(tp->snd_una, tp->high_seq)) {
 		switch (icsk->icsk_ca_state) {
@@ -2972,9 +2972,9 @@ static int tcp_clean_rtx_queue(struct sock *sk, int prior_fackets)
 	}
 
 #if FASTRETRANS_DEBUG > 0
-	BUG_TRAP((int)tp->sacked_out >= 0);
-	BUG_TRAP((int)tp->lost_out >= 0);
-	BUG_TRAP((int)tp->retrans_out >= 0);
+	WARN_ON((int)tp->sacked_out < 0);
+	WARN_ON((int)tp->lost_out < 0);
+	WARN_ON((int)tp->retrans_out < 0);
 	if (!tp->packets_out && tcp_is_sack(tp)) {
 		icsk = inet_csk(sk);
 		if (tp->lost_out) {
@@ -3877,7 +3877,7 @@ static void tcp_sack_remove(struct tcp_sock *tp)
 			int i;
 
 			/* RCV.NXT must cover all the block! */
-			BUG_TRAP(!before(tp->rcv_nxt, sp->end_seq));
+			WARN_ON(before(tp->rcv_nxt, sp->end_seq));
 
 			/* Zap this SACK, by moving forward any other SACKS. */
 			for (i=this_sack+1; i < num_sacks; i++)
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index a82df6307567..a2b06d0cc26b 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -418,7 +418,7 @@ void tcp_v4_err(struct sk_buff *skb, u32 info)
 		/* ICMPs are not backlogged, hence we cannot get
 		   an established socket here.
 		 */
-		BUG_TRAP(!req->sk);
+		WARN_ON(req->sk);
 
 		if (seq != tcp_rsk(req)->snt_isn) {
 			NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index 328e0cf42b3c..5ab6ba19c3ce 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -287,7 +287,7 @@ static void tcp_retransmit_timer(struct sock *sk)
 	if (!tp->packets_out)
 		goto out;
 
-	BUG_TRAP(!tcp_write_queue_empty(sk));
+	WARN_ON(tcp_write_queue_empty(sk));
 
 	if (!tp->snd_wnd && !sock_flag(sk, SOCK_DEAD) &&
 	    !((1 << sk->sk_state) & (TCPF_SYN_SENT | TCPF_SYN_RECV))) {
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 74d543d504a1..a7842c54f58a 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -313,8 +313,10 @@ static void in6_dev_finish_destroy_rcu(struct rcu_head *head)
 void in6_dev_finish_destroy(struct inet6_dev *idev)
 {
 	struct net_device *dev = idev->dev;
-	BUG_TRAP(idev->addr_list==NULL);
-	BUG_TRAP(idev->mc_list==NULL);
+
+	WARN_ON(idev->addr_list != NULL);
+	WARN_ON(idev->mc_list != NULL);
+
 #ifdef NET_REFCNT_DEBUG
 	printk(KERN_DEBUG "in6_dev_finish_destroy: %s\n", dev ? dev->name : "NIL");
 #endif
@@ -517,8 +519,9 @@ static void addrconf_fixup_forwarding(struct ctl_table *table, int *p, int old)
 
 void inet6_ifa_finish_destroy(struct inet6_ifaddr *ifp)
 {
-	BUG_TRAP(ifp->if_next==NULL);
-	BUG_TRAP(ifp->lst_next==NULL);
+	WARN_ON(ifp->if_next != NULL);
+	WARN_ON(ifp->lst_next != NULL);
+
 #ifdef NET_REFCNT_DEBUG
 	printk(KERN_DEBUG "inet6_ifa_finish_destroy\n");
 #endif
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 3d828bc4b1cf..0843c4d6218c 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -153,7 +153,7 @@ lookup_protocol:
 	answer_flags = answer->flags;
 	rcu_read_unlock();
 
-	BUG_TRAP(answer_prot->slab != NULL);
+	WARN_ON(answer_prot->slab == NULL);
 
 	err = -ENOBUFS;
 	sk = sk_alloc(net, PF_INET6, GFP_KERNEL, answer_prot);
diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c
index 87801cc1b2f8..16d43f20b32f 100644
--- a/net/ipv6/inet6_connection_sock.c
+++ b/net/ipv6/inet6_connection_sock.c
@@ -98,7 +98,7 @@ struct request_sock *inet6_csk_search_req(const struct sock *sk,
 		    ipv6_addr_equal(&treq->rmt_addr, raddr) &&
 		    ipv6_addr_equal(&treq->loc_addr, laddr) &&
 		    (!treq->iif || treq->iif == iif)) {
-			BUG_TRAP(req->sk == NULL);
+			WARN_ON(req->sk != NULL);
 			*prevp = prev;
 			return req;
 		}
diff --git a/net/ipv6/inet6_hashtables.c b/net/ipv6/inet6_hashtables.c
index 00a8a5f9380c..1646a5658255 100644
--- a/net/ipv6/inet6_hashtables.c
+++ b/net/ipv6/inet6_hashtables.c
@@ -28,7 +28,7 @@ void __inet6_hash(struct sock *sk)
 	struct hlist_head *list;
 	rwlock_t *lock;
 
-	BUG_TRAP(sk_unhashed(sk));
+	WARN_ON(!sk_unhashed(sk));
 
 	if (sk->sk_state == TCP_LISTEN) {
 		list = &hashinfo->listening_hash[inet_sk_listen_hashfn(sk)];
@@ -202,7 +202,7 @@ unique:
 	 * in hash table socket with a funny identity. */
 	inet->num = lport;
 	inet->sport = htons(lport);
-	BUG_TRAP(sk_unhashed(sk));
+	WARN_ON(!sk_unhashed(sk));
 	__sk_add_node(sk, &head->chain);
 	sk->sk_hash = hash;
 	sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 08ea2de28d63..52dddc25d3e6 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -287,7 +287,7 @@ static int fib6_dump_node(struct fib6_walker_t *w)
 			w->leaf = rt;
 			return 1;
 		}
-		BUG_TRAP(res!=0);
+		WARN_ON(res == 0);
 	}
 	w->leaf = NULL;
 	return 0;
@@ -778,7 +778,7 @@ out:
 			pn->leaf = fib6_find_prefix(info->nl_net, pn);
 #if RT6_DEBUG >= 2
 			if (!pn->leaf) {
-				BUG_TRAP(pn->leaf != NULL);
+				WARN_ON(pn->leaf == NULL);
 				pn->leaf = info->nl_net->ipv6.ip6_null_entry;
 			}
 #endif
@@ -942,7 +942,7 @@ struct fib6_node * fib6_locate(struct fib6_node *root,
 
 #ifdef CONFIG_IPV6_SUBTREES
 	if (src_len) {
-		BUG_TRAP(saddr!=NULL);
+		WARN_ON(saddr == NULL);
 		if (fn && fn->subtree)
 			fn = fib6_locate_1(fn->subtree, saddr, src_len,
 					   offsetof(struct rt6_info, rt6i_src));
@@ -996,9 +996,9 @@ static struct fib6_node *fib6_repair_tree(struct net *net,
 		RT6_TRACE("fixing tree: plen=%d iter=%d\n", fn->fn_bit, iter);
 		iter++;
 
-		BUG_TRAP(!(fn->fn_flags&RTN_RTINFO));
-		BUG_TRAP(!(fn->fn_flags&RTN_TL_ROOT));
-		BUG_TRAP(fn->leaf==NULL);
+		WARN_ON(fn->fn_flags & RTN_RTINFO);
+		WARN_ON(fn->fn_flags & RTN_TL_ROOT);
+		WARN_ON(fn->leaf != NULL);
 
 		children = 0;
 		child = NULL;
@@ -1014,7 +1014,7 @@ static struct fib6_node *fib6_repair_tree(struct net *net,
 			fn->leaf = fib6_find_prefix(net, fn);
 #if RT6_DEBUG >= 2
 			if (fn->leaf==NULL) {
-				BUG_TRAP(fn->leaf);
+				WARN_ON(!fn->leaf);
 				fn->leaf = net->ipv6.ip6_null_entry;
 			}
 #endif
@@ -1025,16 +1025,17 @@ static struct fib6_node *fib6_repair_tree(struct net *net,
 		pn = fn->parent;
 #ifdef CONFIG_IPV6_SUBTREES
 		if (FIB6_SUBTREE(pn) == fn) {
-			BUG_TRAP(fn->fn_flags&RTN_ROOT);
+			WARN_ON(!(fn->fn_flags & RTN_ROOT));
 			FIB6_SUBTREE(pn) = NULL;
 			nstate = FWS_L;
 		} else {
-			BUG_TRAP(!(fn->fn_flags&RTN_ROOT));
+			WARN_ON(fn->fn_flags & RTN_ROOT);
 #endif
 			if (pn->right == fn) pn->right = child;
 			else if (pn->left == fn) pn->left = child;
 #if RT6_DEBUG >= 2
-			else BUG_TRAP(0);
+			else
+				WARN_ON(1);
 #endif
 			if (child)
 				child->parent = pn;
@@ -1154,14 +1155,14 @@ int fib6_del(struct rt6_info *rt, struct nl_info *info)
 
 #if RT6_DEBUG >= 2
 	if (rt->u.dst.obsolete>0) {
-		BUG_TRAP(fn==NULL);
+		WARN_ON(fn != NULL);
 		return -ENOENT;
 	}
 #endif
 	if (fn == NULL || rt == net->ipv6.ip6_null_entry)
 		return -ENOENT;
 
-	BUG_TRAP(fn->fn_flags&RTN_RTINFO);
+	WARN_ON(!(fn->fn_flags & RTN_RTINFO));
 
 	if (!(rt->rt6i_flags&RTF_CACHE)) {
 		struct fib6_node *pn = fn;
@@ -1266,7 +1267,7 @@ static int fib6_walk_continue(struct fib6_walker_t *w)
 			w->node = pn;
 #ifdef CONFIG_IPV6_SUBTREES
 			if (FIB6_SUBTREE(pn) == fn) {
-				BUG_TRAP(fn->fn_flags&RTN_ROOT);
+				WARN_ON(!(fn->fn_flags & RTN_ROOT));
 				w->state = FWS_L;
 				continue;
 			}
@@ -1281,7 +1282,7 @@ static int fib6_walk_continue(struct fib6_walker_t *w)
 				continue;
 			}
 #if RT6_DEBUG >= 2
-			BUG_TRAP(0);
+			WARN_ON(1);
 #endif
 		}
 	}
@@ -1323,7 +1324,7 @@ static int fib6_clean_node(struct fib6_walker_t *w)
 			}
 			return 0;
 		}
-		BUG_TRAP(res==0);
+		WARN_ON(res != 0);
 	}
 	w->leaf = rt;
 	return 0;
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 6407c64ea4a5..6811901e6b1e 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -116,7 +116,7 @@ static int ip6_dev_loopback_xmit(struct sk_buff *newskb)
 	__skb_pull(newskb, skb_network_offset(newskb));
 	newskb->pkt_type = PACKET_LOOPBACK;
 	newskb->ip_summed = CHECKSUM_UNNECESSARY;
-	BUG_TRAP(newskb->dst);
+	WARN_ON(!newskb->dst);
 
 	netif_rx(newskb);
 	return 0;
diff --git a/net/ipv6/mip6.c b/net/ipv6/mip6.c
index ad1cc5bbf977..31295c8f6196 100644
--- a/net/ipv6/mip6.c
+++ b/net/ipv6/mip6.c
@@ -164,8 +164,8 @@ static int mip6_destopt_output(struct xfrm_state *x, struct sk_buff *skb)
 			calc_padlen(sizeof(*dstopt), 6));
 
 	hao->type = IPV6_TLV_HAO;
+	BUILD_BUG_ON(sizeof(*hao) != 18);
 	hao->length = sizeof(*hao) - 2;
-	BUG_TRAP(hao->length == 16);
 
 	len = ((char *)hao - (char *)dstopt) + sizeof(*hao);
 
@@ -174,7 +174,7 @@ static int mip6_destopt_output(struct xfrm_state *x, struct sk_buff *skb)
 	memcpy(&iph->saddr, x->coaddr, sizeof(iph->saddr));
 	spin_unlock_bh(&x->lock);
 
-	BUG_TRAP(len == x->props.header_len);
+	WARN_ON(len != x->props.header_len);
 	dstopt->hdrlen = (x->props.header_len >> 3) - 1;
 
 	return 0;
@@ -317,7 +317,7 @@ static int mip6_destopt_init_state(struct xfrm_state *x)
 	x->props.header_len = sizeof(struct ipv6_destopt_hdr) +
 		calc_padlen(sizeof(struct ipv6_destopt_hdr), 6) +
 		sizeof(struct ipv6_destopt_hao);
-	BUG_TRAP(x->props.header_len == 24);
+	WARN_ON(x->props.header_len != 24);
 
 	return 0;
 }
@@ -380,7 +380,7 @@ static int mip6_rthdr_output(struct xfrm_state *x, struct sk_buff *skb)
 	rt2->rt_hdr.segments_left = 1;
 	memset(&rt2->reserved, 0, sizeof(rt2->reserved));
 
-	BUG_TRAP(rt2->rt_hdr.hdrlen == 2);
+	WARN_ON(rt2->rt_hdr.hdrlen != 2);
 
 	memcpy(&rt2->addr, &iph->daddr, sizeof(rt2->addr));
 	spin_lock_bh(&x->lock);
diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c
index cf20bc4fd60d..52d06dd4b817 100644
--- a/net/ipv6/netfilter/nf_conntrack_reasm.c
+++ b/net/ipv6/netfilter/nf_conntrack_reasm.c
@@ -416,8 +416,8 @@ nf_ct_frag6_reasm(struct nf_ct_frag6_queue *fq, struct net_device *dev)
 
 	fq_kill(fq);
 
-	BUG_TRAP(head != NULL);
-	BUG_TRAP(NFCT_FRAG6_CB(head)->offset == 0);
+	WARN_ON(head == NULL);
+	WARN_ON(NFCT_FRAG6_CB(head)->offset != 0);
 
 	/* Unfragmented part is taken from the first segment. */
 	payload_len = ((head->data - skb_network_header(head)) -
diff --git a/net/ipv6/reassembly.c b/net/ipv6/reassembly.c
index 6ab957ec2dd6..89184b576e23 100644
--- a/net/ipv6/reassembly.c
+++ b/net/ipv6/reassembly.c
@@ -473,8 +473,8 @@ static int ip6_frag_reasm(struct frag_queue *fq, struct sk_buff *prev,
 		fq->q.fragments = head;
 	}
 
-	BUG_TRAP(head != NULL);
-	BUG_TRAP(FRAG6_CB(head)->offset == 0);
+	WARN_ON(head == NULL);
+	WARN_ON(FRAG6_CB(head)->offset != 0);
 
 	/* Unfragmented part is taken from the first segment. */
 	payload_len = ((head->data - skb_network_header(head)) -
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index ae45f9835014..cff778b23a7f 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -421,7 +421,7 @@ static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 		/* ICMPs are not backlogged, hence we cannot get
 		 * an established socket here.
 		 */
-		BUG_TRAP(req->sk == NULL);
+		WARN_ON(req->sk != NULL);
 
 		if (seq != tcp_rsk(req)->snt_isn) {
 			NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
diff --git a/net/key/af_key.c b/net/key/af_key.c
index f0fc46c8038d..d628df97e02e 100644
--- a/net/key/af_key.c
+++ b/net/key/af_key.c
@@ -96,8 +96,8 @@ static void pfkey_sock_destruct(struct sock *sk)
 		return;
 	}
 
-	BUG_TRAP(!atomic_read(&sk->sk_rmem_alloc));
-	BUG_TRAP(!atomic_read(&sk->sk_wmem_alloc));
+	WARN_ON(atomic_read(&sk->sk_rmem_alloc));
+	WARN_ON(atomic_read(&sk->sk_wmem_alloc));
 
 	atomic_dec(&pfkey_socks_nr);
 }
diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c
index 98bfe277eab2..b0eacc0007cc 100644
--- a/net/netlink/af_netlink.c
+++ b/net/netlink/af_netlink.c
@@ -158,9 +158,10 @@ static void netlink_sock_destruct(struct sock *sk)
 		printk(KERN_ERR "Freeing alive netlink socket %p\n", sk);
 		return;
 	}
-	BUG_TRAP(!atomic_read(&sk->sk_rmem_alloc));
-	BUG_TRAP(!atomic_read(&sk->sk_wmem_alloc));
-	BUG_TRAP(!nlk_sk(sk)->groups);
+
+	WARN_ON(atomic_read(&sk->sk_rmem_alloc));
+	WARN_ON(atomic_read(&sk->sk_wmem_alloc));
+	WARN_ON(nlk_sk(sk)->groups);
 }
 
 /* This lock without WQ_FLAG_EXCLUSIVE is good on UP and it is _very_ bad on
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index d56cae112dc8..c718e7e3f7de 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -260,8 +260,8 @@ static inline struct packet_sock *pkt_sk(struct sock *sk)
 
 static void packet_sock_destruct(struct sock *sk)
 {
-	BUG_TRAP(!atomic_read(&sk->sk_rmem_alloc));
-	BUG_TRAP(!atomic_read(&sk->sk_wmem_alloc));
+	WARN_ON(atomic_read(&sk->sk_rmem_alloc));
+	WARN_ON(atomic_read(&sk->sk_wmem_alloc));
 
 	if (!sock_flag(sk, SOCK_DEAD)) {
 		printk("Attempt to release alive packet socket: %p\n", sk);
diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c
index 4b2682feeedc..32e489118beb 100644
--- a/net/rxrpc/af_rxrpc.c
+++ b/net/rxrpc/af_rxrpc.c
@@ -660,9 +660,9 @@ static void rxrpc_sock_destructor(struct sock *sk)
 
 	rxrpc_purge_queue(&sk->sk_receive_queue);
 
-	BUG_TRAP(!atomic_read(&sk->sk_wmem_alloc));
-	BUG_TRAP(sk_unhashed(sk));
-	BUG_TRAP(!sk->sk_socket);
+	WARN_ON(atomic_read(&sk->sk_wmem_alloc));
+	WARN_ON(!sk_unhashed(sk));
+	WARN_ON(sk->sk_socket);
 
 	if (!sock_flag(sk, SOCK_DEAD)) {
 		printk("Attempt to release alive rxrpc socket: %p\n", sk);
diff --git a/net/sched/act_api.c b/net/sched/act_api.c
index 74e662cbb2c5..d308c19aa3f9 100644
--- a/net/sched/act_api.c
+++ b/net/sched/act_api.c
@@ -41,7 +41,7 @@ void tcf_hash_destroy(struct tcf_common *p, struct tcf_hashinfo *hinfo)
 			return;
 		}
 	}
-	BUG_TRAP(0);
+	WARN_ON(1);
 }
 EXPORT_SYMBOL(tcf_hash_destroy);
 
diff --git a/net/sched/act_police.c b/net/sched/act_police.c
index 32c3f9d9fb7a..38015b493947 100644
--- a/net/sched/act_police.c
+++ b/net/sched/act_police.c
@@ -116,7 +116,7 @@ static void tcf_police_destroy(struct tcf_police *p)
 			return;
 		}
 	}
-	BUG_TRAP(0);
+	WARN_ON(1);
 }
 
 static const struct nla_policy police_policy[TCA_POLICE_MAX + 1] = {
diff --git a/net/sched/cls_u32.c b/net/sched/cls_u32.c
index 527db2559dd2..246f9065ce34 100644
--- a/net/sched/cls_u32.c
+++ b/net/sched/cls_u32.c
@@ -345,7 +345,7 @@ static int u32_delete_key(struct tcf_proto *tp, struct tc_u_knode* key)
 			}
 		}
 	}
-	BUG_TRAP(0);
+	WARN_ON(1);
 	return 0;
 }
 
@@ -368,7 +368,7 @@ static int u32_destroy_hnode(struct tcf_proto *tp, struct tc_u_hnode *ht)
 	struct tc_u_common *tp_c = tp->data;
 	struct tc_u_hnode **hn;
 
-	BUG_TRAP(!ht->refcnt);
+	WARN_ON(ht->refcnt);
 
 	u32_clear_hnode(tp, ht);
 
@@ -380,7 +380,7 @@ static int u32_destroy_hnode(struct tcf_proto *tp, struct tc_u_hnode *ht)
 		}
 	}
 
-	BUG_TRAP(0);
+	WARN_ON(1);
 	return -ENOENT;
 }
 
@@ -389,7 +389,7 @@ static void u32_destroy(struct tcf_proto *tp)
 	struct tc_u_common *tp_c = tp->data;
 	struct tc_u_hnode *root_ht = xchg(&tp->root, NULL);
 
-	BUG_TRAP(root_ht != NULL);
+	WARN_ON(root_ht == NULL);
 
 	if (root_ht && --root_ht->refcnt == 0)
 		u32_destroy_hnode(tp, root_ht);
@@ -407,7 +407,7 @@ static void u32_destroy(struct tcf_proto *tp)
 		while ((ht = tp_c->hlist) != NULL) {
 			tp_c->hlist = ht->next;
 
-			BUG_TRAP(ht->refcnt == 0);
+			WARN_ON(ht->refcnt != 0);
 
 			kfree(ht);
 		}
diff --git a/net/sched/sch_cbq.c b/net/sched/sch_cbq.c
index f1d2f8ec8b4c..14954bf4a683 100644
--- a/net/sched/sch_cbq.c
+++ b/net/sched/sch_cbq.c
@@ -1175,7 +1175,7 @@ static void cbq_unlink_class(struct cbq_class *this)
 				this->tparent->children = NULL;
 		}
 	} else {
-		BUG_TRAP(this->sibling == this);
+		WARN_ON(this->sibling != this);
 	}
 }
 
@@ -1699,7 +1699,7 @@ static void cbq_destroy_class(struct Qdisc *sch, struct cbq_class *cl)
 {
 	struct cbq_sched_data *q = qdisc_priv(sch);
 
-	BUG_TRAP(!cl->filters);
+	WARN_ON(cl->filters);
 
 	tcf_destroy_chain(&cl->filter_list);
 	qdisc_destroy(cl->q);
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index 43abd4d27ea6..fd2a6cadb115 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -746,5 +746,5 @@ void dev_shutdown(struct net_device *dev)
 {
 	netdev_for_each_tx_queue(dev, shutdown_scheduler_queue, &noop_qdisc);
 	shutdown_scheduler_queue(dev, &dev->rx_queue, NULL);
-	BUG_TRAP(!timer_pending(&dev->watchdog_timer));
+	WARN_ON(timer_pending(&dev->watchdog_timer));
 }
diff --git a/net/sched/sch_htb.c b/net/sched/sch_htb.c
index 30c999c61b01..75a40951c4f2 100644
--- a/net/sched/sch_htb.c
+++ b/net/sched/sch_htb.c
@@ -524,7 +524,7 @@ htb_change_class_mode(struct htb_sched *q, struct htb_class *cl, long *diff)
  */
 static inline void htb_activate(struct htb_sched *q, struct htb_class *cl)
 {
-	BUG_TRAP(!cl->level && cl->un.leaf.q && cl->un.leaf.q->q.qlen);
+	WARN_ON(cl->level || !cl->un.leaf.q || !cl->un.leaf.q->q.qlen);
 
 	if (!cl->prio_activity) {
 		cl->prio_activity = 1 << (cl->un.leaf.aprio = cl->un.leaf.prio);
@@ -542,7 +542,7 @@ static inline void htb_activate(struct htb_sched *q, struct htb_class *cl)
  */
 static inline void htb_deactivate(struct htb_sched *q, struct htb_class *cl)
 {
-	BUG_TRAP(cl->prio_activity);
+	WARN_ON(!cl->prio_activity);
 
 	htb_deactivate_prios(q, cl);
 	cl->prio_activity = 0;
@@ -757,7 +757,7 @@ static struct htb_class *htb_lookup_leaf(struct rb_root *tree, int prio,
 		u32 *pid;
 	} stk[TC_HTB_MAXDEPTH], *sp = stk;
 
-	BUG_TRAP(tree->rb_node);
+	WARN_ON(!tree->rb_node);
 	sp->root = tree->rb_node;
 	sp->pptr = pptr;
 	sp->pid = pid;
@@ -777,7 +777,7 @@ static struct htb_class *htb_lookup_leaf(struct rb_root *tree, int prio,
 				*sp->pptr = (*sp->pptr)->rb_left;
 			if (sp > stk) {
 				sp--;
-				BUG_TRAP(*sp->pptr);
+				WARN_ON(!*sp->pptr);
 				if (!*sp->pptr)
 					return NULL;
 				htb_next_rb_node(sp->pptr);
@@ -792,7 +792,7 @@ static struct htb_class *htb_lookup_leaf(struct rb_root *tree, int prio,
 			sp->pid = cl->un.inner.last_ptr_id + prio;
 		}
 	}
-	BUG_TRAP(0);
+	WARN_ON(1);
 	return NULL;
 }
 
@@ -810,7 +810,7 @@ static struct sk_buff *htb_dequeue_tree(struct htb_sched *q, int prio,
 
 	do {
 next:
-		BUG_TRAP(cl);
+		WARN_ON(!cl);
 		if (!cl)
 			return NULL;
 
@@ -1185,7 +1185,7 @@ static void htb_parent_to_leaf(struct htb_sched *q, struct htb_class *cl,
 {
 	struct htb_class *parent = cl->parent;
 
-	BUG_TRAP(!cl->level && cl->un.leaf.q && !cl->prio_activity);
+	WARN_ON(cl->level || !cl->un.leaf.q || cl->prio_activity);
 
 	if (parent->cmode != HTB_CAN_SEND)
 		htb_safe_rb_erase(&parent->pq_node, q->wait_pq + parent->level);
@@ -1205,7 +1205,7 @@ static void htb_parent_to_leaf(struct htb_sched *q, struct htb_class *cl,
 static void htb_destroy_class(struct Qdisc *sch, struct htb_class *cl)
 {
 	if (!cl->level) {
-		BUG_TRAP(cl->un.leaf.q);
+		WARN_ON(!cl->un.leaf.q);
 		qdisc_destroy(cl->un.leaf.q);
 	}
 	gen_kill_estimator(&cl->bstats, &cl->rate_est);
diff --git a/net/sctp/associola.c b/net/sctp/associola.c
index ec2a0a33fd78..8472b8b349c4 100644
--- a/net/sctp/associola.c
+++ b/net/sctp/associola.c
@@ -464,7 +464,7 @@ static void sctp_association_destroy(struct sctp_association *asoc)
 		spin_unlock_bh(&sctp_assocs_id_lock);
 	}
 
-	BUG_TRAP(!atomic_read(&asoc->rmem_alloc));
+	WARN_ON(atomic_read(&asoc->rmem_alloc));
 
 	if (asoc->base.malloced) {
 		kfree(asoc);
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 70ceb1604ad8..24eb214581d5 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -227,7 +227,7 @@ static void __unix_remove_socket(struct sock *sk)
 
 static void __unix_insert_socket(struct hlist_head *list, struct sock *sk)
 {
-	BUG_TRAP(sk_unhashed(sk));
+	WARN_ON(!sk_unhashed(sk));
 	sk_add_node(sk, list);
 }
 
@@ -350,9 +350,9 @@ static void unix_sock_destructor(struct sock *sk)
 
 	skb_queue_purge(&sk->sk_receive_queue);
 
-	BUG_TRAP(!atomic_read(&sk->sk_wmem_alloc));
-	BUG_TRAP(sk_unhashed(sk));
-	BUG_TRAP(!sk->sk_socket);
+	WARN_ON(atomic_read(&sk->sk_wmem_alloc));
+	WARN_ON(!sk_unhashed(sk));
+	WARN_ON(sk->sk_socket);
 	if (!sock_flag(sk, SOCK_DEAD)) {
 		printk("Attempt to release alive unix socket: %p\n", sk);
 		return;
diff --git a/net/xfrm/xfrm_algo.c b/net/xfrm/xfrm_algo.c
index 23a2cc04b8cd..96036cf2216d 100644
--- a/net/xfrm/xfrm_algo.c
+++ b/net/xfrm/xfrm_algo.c
@@ -718,7 +718,7 @@ int skb_icv_walk(const struct sk_buff *skb, struct hash_desc *desc,
 	for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) {
 		int end;
 
-		BUG_TRAP(start <= offset + len);
+		WARN_ON(start > offset + len);
 
 		end = start + skb_shinfo(skb)->frags[i].size;
 		if ((copy = end - offset) > 0) {
@@ -748,7 +748,7 @@ int skb_icv_walk(const struct sk_buff *skb, struct hash_desc *desc,
 		for (; list; list = list->next) {
 			int end;
 
-			BUG_TRAP(start <= offset + len);
+			WARN_ON(start > offset + len);
 
 			end = start + list->len;
 			if ((copy = end - offset) > 0) {
diff --git a/net/xfrm/xfrm_ipcomp.c b/net/xfrm/xfrm_ipcomp.c
index 800f669083fb..c609a4b98e15 100644
--- a/net/xfrm/xfrm_ipcomp.c
+++ b/net/xfrm/xfrm_ipcomp.c
@@ -22,7 +22,6 @@
 #include <linux/module.h>
 #include <linux/mutex.h>
 #include <linux/percpu.h>
-#include <linux/rtnetlink.h>
 #include <linux/smp.h>
 #include <linux/vmalloc.h>
 #include <net/ip.h>
@@ -251,7 +250,7 @@ static void ipcomp_free_tfms(struct crypto_comp **tfms)
 			break;
 	}
 
-	BUG_TRAP(pos);
+	WARN_ON(!pos);
 
 	if (--pos->users)
 		return;
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 72fddafd891a..4c6914ef7d92 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -538,7 +538,7 @@ EXPORT_SYMBOL(xfrm_state_alloc);
 
 void __xfrm_state_destroy(struct xfrm_state *x)
 {
-	BUG_TRAP(x->km.state == XFRM_STATE_DEAD);
+	WARN_ON(x->km.state != XFRM_STATE_DEAD);
 
 	spin_lock_bh(&xfrm_state_lock);
 	list_del(&x->all);

From ec34c702ca8b7d6f0aa54379c3b0d0ec10b8ff23 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ilpo=20J=C3=A4rvinen?= <ilpo.jarvinen@helsinki.fi>
Date: Fri, 25 Jul 2008 21:45:49 -0700
Subject: [PATCH 0962/1171] net: drop unused BUG_TRAP()
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/rtnetlink.h | 7 -------
 1 file changed, 7 deletions(-)

diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index f4d386c191f5..ca643b13b026 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -755,13 +755,6 @@ extern void __rtnl_unlock(void);
 	} \
 } while(0)
 
-#define BUG_TRAP(x) do { \
-	if (unlikely(!(x))) { \
-		printk(KERN_ERR "KERNEL: assertion (%s) failed at %s (%d)\n", \
-			#x,  __FILE__ , __LINE__); \
-	} \
-} while(0)
-
 static inline u32 rtm_get_table(struct rtattr **rta, u8 table)
 {
 	return RTA_GET_U32(rta[RTA_TABLE-1]);

From 054a3fd824705543322d787893de9f3755151517 Mon Sep 17 00:00:00 2001
From: Wang Chen <wangchen@cn.fujitsu.com>
Date: Fri, 25 Jul 2008 15:44:33 +0800
Subject: [PATCH 0963/1171] flag parameters: fix compile error of
 sys_epoll_create1

  GEN     .version
  CHK     include/linux/compile.h
  UPD     include/linux/compile.h
  CC      init/version.o
  LD      init/built-in.o
  LD      vmlinux
arch/x86/kernel/built-in.o: In function `sys_call_table':
(.rodata+0x8a4): undefined reference to `sys_epoll_create1'
make: *** [vmlinux] Error 1

Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Davide Libenzi <davidel@xmailserver.org>
Cc: Ulrich Drepper <drepper@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/sys_ni.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c
index 0fea0ee12da9..0be4fc1409a2 100644
--- a/kernel/sys_ni.c
+++ b/kernel/sys_ni.c
@@ -56,6 +56,7 @@ cond_syscall(compat_sys_set_robust_list);
 cond_syscall(sys_get_robust_list);
 cond_syscall(compat_sys_get_robust_list);
 cond_syscall(sys_epoll_create);
+cond_syscall(sys_epoll_create1);
 cond_syscall(sys_epoll_ctl);
 cond_syscall(sys_epoll_wait);
 cond_syscall(sys_epoll_pwait);

From 16df845f4566bc252f3e09db12f5c2f22cb44226 Mon Sep 17 00:00:00 2001
From: Florian Westphal <fw@strlen.de>
Date: Sat, 26 Jul 2008 02:21:54 -0700
Subject: [PATCH 0964/1171] syncookies: Make sure ECN is disabled

ecn_ok is not initialized when a connection is established by cookies.
The cookie syn-ack never sets ECN, so ecn_ok must be set to 0.

Spotted using ns-3/network simulation cradle simulator and valgrind.

Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/syncookies.c | 1 +
 net/ipv6/syncookies.c | 1 +
 2 files changed, 2 insertions(+)

diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index 51bc24d3b8a7..9d38005abbac 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -299,6 +299,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb,
 	ireq->rmt_port		= th->source;
 	ireq->loc_addr		= ip_hdr(skb)->daddr;
 	ireq->rmt_addr		= ip_hdr(skb)->saddr;
+	ireq->ecn_ok		= 0;
 	ireq->snd_wscale	= tcp_opt.snd_wscale;
 	ireq->rcv_wscale	= tcp_opt.rcv_wscale;
 	ireq->sack_ok		= tcp_opt.sack_ok;
diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c
index 6a68eeb7bbf8..a46badd1082d 100644
--- a/net/ipv6/syncookies.c
+++ b/net/ipv6/syncookies.c
@@ -223,6 +223,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 
 	req->expires = 0UL;
 	req->retrans = 0;
+	ireq->ecn_ok		= 0;
 	ireq->snd_wscale	= tcp_opt.snd_wscale;
 	ireq->rcv_wscale	= tcp_opt.rcv_wscale;
 	ireq->sack_ok		= tcp_opt.sack_ok;

From 509e2562adfd63964aa30c1ddd9ddf4e57949351 Mon Sep 17 00:00:00 2001
From: Heiko Carstens <heiko.carstens@de.ibm.com>
Date: Sat, 26 Jul 2008 02:24:10 -0700
Subject: [PATCH 0965/1171] qeth: use dev->ml_priv instead of dev->priv

From: Heiko Carstens <heiko.carstens@de.ibm.com>

This makes qeth working again after git commit
e3c50d5d25ac09efd9acbe2b2a3e365466de84ed
"netdev: netdev_priv() can now be sane again.".

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 drivers/s390/net/qeth_core_main.c | 14 +++++++-------
 drivers/s390/net/qeth_l2_main.c   | 26 +++++++++++++-------------
 drivers/s390/net/qeth_l3_main.c   | 30 +++++++++++++++---------------
 3 files changed, 35 insertions(+), 35 deletions(-)

diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index c3ad89e302bd..cebb25e36e82 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -3321,7 +3321,7 @@ int qeth_change_mtu(struct net_device *dev, int new_mtu)
 	struct qeth_card *card;
 	char dbf_text[15];
 
-	card = netdev_priv(dev);
+	card = dev->ml_priv;
 
 	QETH_DBF_TEXT(TRACE, 4, "chgmtu");
 	sprintf(dbf_text, "%8x", new_mtu);
@@ -3343,7 +3343,7 @@ struct net_device_stats *qeth_get_stats(struct net_device *dev)
 {
 	struct qeth_card *card;
 
-	card = netdev_priv(dev);
+	card = dev->ml_priv;
 
 	QETH_DBF_TEXT(TRACE, 5, "getstat");
 
@@ -3395,7 +3395,7 @@ void qeth_tx_timeout(struct net_device *dev)
 {
 	struct qeth_card *card;
 
-	card = netdev_priv(dev);
+	card = dev->ml_priv;
 	card->stats.tx_errors++;
 	qeth_schedule_recovery(card);
 }
@@ -3403,7 +3403,7 @@ EXPORT_SYMBOL_GPL(qeth_tx_timeout);
 
 int qeth_mdio_read(struct net_device *dev, int phy_id, int regnum)
 {
-	struct qeth_card *card = netdev_priv(dev);
+	struct qeth_card *card = dev->ml_priv;
 	int rc = 0;
 
 	switch (regnum) {
@@ -4253,7 +4253,7 @@ EXPORT_SYMBOL_GPL(qeth_core_get_stats_count);
 void qeth_core_get_ethtool_stats(struct net_device *dev,
 		struct ethtool_stats *stats, u64 *data)
 {
-	struct qeth_card *card = netdev_priv(dev);
+	struct qeth_card *card = dev->ml_priv;
 	data[0] = card->stats.rx_packets -
 				card->perf_stats.initial_rx_packets;
 	data[1] = card->perf_stats.bufs_rec;
@@ -4313,7 +4313,7 @@ EXPORT_SYMBOL_GPL(qeth_core_get_strings);
 void qeth_core_get_drvinfo(struct net_device *dev,
 		struct ethtool_drvinfo *info)
 {
-	struct qeth_card *card = netdev_priv(dev);
+	struct qeth_card *card = dev->ml_priv;
 	if (card->options.layer2)
 		strcpy(info->driver, "qeth_l2");
 	else
@@ -4331,7 +4331,7 @@ EXPORT_SYMBOL_GPL(qeth_core_get_drvinfo);
 int qeth_core_ethtool_get_settings(struct net_device *netdev,
 					struct ethtool_cmd *ecmd)
 {
-	struct qeth_card *card = netdev_priv(netdev);
+	struct qeth_card *card = netdev->ml_priv;
 	enum qeth_link_types link_type;
 
 	if ((card->info.type == QETH_CARD_TYPE_IQD) || (card->info.guestlan))
diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c
index 3fbc3bdec0c5..a8b069cd9a4c 100644
--- a/drivers/s390/net/qeth_l2_main.c
+++ b/drivers/s390/net/qeth_l2_main.c
@@ -35,7 +35,7 @@ static int qeth_l2_recover(void *);
 
 static int qeth_l2_do_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
 {
-	struct qeth_card *card = netdev_priv(dev);
+	struct qeth_card *card = dev->ml_priv;
 	struct mii_ioctl_data *mii_data;
 	int rc = 0;
 
@@ -317,7 +317,7 @@ static void qeth_l2_process_vlans(struct qeth_card *card, int clear)
 
 static void qeth_l2_vlan_rx_add_vid(struct net_device *dev, unsigned short vid)
 {
-	struct qeth_card *card = netdev_priv(dev);
+	struct qeth_card *card = dev->ml_priv;
 	struct qeth_vlan_vid *id;
 
 	QETH_DBF_TEXT_(TRACE, 4, "aid:%d", vid);
@@ -334,7 +334,7 @@ static void qeth_l2_vlan_rx_add_vid(struct net_device *dev, unsigned short vid)
 static void qeth_l2_vlan_rx_kill_vid(struct net_device *dev, unsigned short vid)
 {
 	struct qeth_vlan_vid *id, *tmpid = NULL;
-	struct qeth_card *card = netdev_priv(dev);
+	struct qeth_card *card = dev->ml_priv;
 
 	QETH_DBF_TEXT_(TRACE, 4, "kid:%d", vid);
 	spin_lock_bh(&card->vlanlock);
@@ -566,7 +566,7 @@ static int qeth_l2_request_initial_mac(struct qeth_card *card)
 static int qeth_l2_set_mac_address(struct net_device *dev, void *p)
 {
 	struct sockaddr *addr = p;
-	struct qeth_card *card = netdev_priv(dev);
+	struct qeth_card *card = dev->ml_priv;
 	int rc = 0;
 
 	QETH_DBF_TEXT(TRACE, 3, "setmac");
@@ -590,7 +590,7 @@ static int qeth_l2_set_mac_address(struct net_device *dev, void *p)
 
 static void qeth_l2_set_multicast_list(struct net_device *dev)
 {
-	struct qeth_card *card = netdev_priv(dev);
+	struct qeth_card *card = dev->ml_priv;
 	struct dev_mc_list *dm;
 
 	if (card->info.type == QETH_CARD_TYPE_OSN)
@@ -612,7 +612,7 @@ static int qeth_l2_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	int rc;
 	struct qeth_hdr *hdr = NULL;
 	int elements = 0;
-	struct qeth_card *card = netdev_priv(dev);
+	struct qeth_card *card = dev->ml_priv;
 	struct sk_buff *new_skb = skb;
 	int ipv = qeth_get_ip_version(skb);
 	int cast_type = qeth_get_cast_type(card, skb);
@@ -767,7 +767,7 @@ static void qeth_l2_qdio_input_handler(struct ccw_device *ccwdev,
 
 static int qeth_l2_open(struct net_device *dev)
 {
-	struct qeth_card *card = netdev_priv(dev);
+	struct qeth_card *card = dev->ml_priv;
 
 	QETH_DBF_TEXT(TRACE, 4, "qethopen");
 	if (card->state != CARD_STATE_SOFTSETUP)
@@ -791,7 +791,7 @@ static int qeth_l2_open(struct net_device *dev)
 
 static int qeth_l2_stop(struct net_device *dev)
 {
-	struct qeth_card *card = netdev_priv(dev);
+	struct qeth_card *card = dev->ml_priv;
 
 	QETH_DBF_TEXT(TRACE, 4, "qethstop");
 	netif_tx_disable(dev);
@@ -838,7 +838,7 @@ static void qeth_l2_remove_device(struct ccwgroup_device *cgdev)
 
 static int qeth_l2_ethtool_set_tso(struct net_device *dev, u32 data)
 {
-	struct qeth_card *card = netdev_priv(dev);
+	struct qeth_card *card = dev->ml_priv;
 
 	if (data) {
 		if (card->options.large_send == QETH_LARGE_SEND_NO) {
@@ -894,7 +894,7 @@ static int qeth_l2_setup_netdev(struct qeth_card *card)
 	if (!card->dev)
 		return -ENODEV;
 
-	card->dev->priv = card;
+	card->dev->ml_priv = card;
 	card->dev->tx_timeout = &qeth_tx_timeout;
 	card->dev->watchdog_timeo = QETH_TX_TIMEOUT;
 	card->dev->open = qeth_l2_open;
@@ -1178,7 +1178,7 @@ int qeth_osn_assist(struct net_device *dev, void *data, int data_len)
 	QETH_DBF_TEXT(TRACE, 2, "osnsdmc");
 	if (!dev)
 		return -ENODEV;
-	card = netdev_priv(dev);
+	card = dev->ml_priv;
 	if (!card)
 		return -ENODEV;
 	if ((card->state != CARD_STATE_UP) &&
@@ -1201,7 +1201,7 @@ int qeth_osn_register(unsigned char *read_dev_no, struct net_device **dev,
 	*dev = qeth_l2_netdev_by_devno(read_dev_no);
 	if (*dev == NULL)
 		return -ENODEV;
-	card = netdev_priv(*dev);
+	card = (*dev)->ml_priv;
 	if (!card)
 		return -ENODEV;
 	if ((assist_cb == NULL) || (data_cb == NULL))
@@ -1219,7 +1219,7 @@ void qeth_osn_deregister(struct net_device *dev)
 	QETH_DBF_TEXT(TRACE, 2, "osndereg");
 	if (!dev)
 		return;
-	card = netdev_priv(dev);
+	card = dev->ml_priv;
 	if (!card)
 		return;
 	card->osn_info.assist_cb = NULL;
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index 38de31b55708..3e1d13857350 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -1813,7 +1813,7 @@ static void qeth_l3_free_vlan_addresses(struct qeth_card *card,
 static void qeth_l3_vlan_rx_register(struct net_device *dev,
 			struct vlan_group *grp)
 {
-	struct qeth_card *card = netdev_priv(dev);
+	struct qeth_card *card = dev->ml_priv;
 	unsigned long flags;
 
 	QETH_DBF_TEXT(TRACE, 4, "vlanreg");
@@ -1825,7 +1825,7 @@ static void qeth_l3_vlan_rx_register(struct net_device *dev,
 static void qeth_l3_vlan_rx_add_vid(struct net_device *dev, unsigned short vid)
 {
 	struct net_device *vlandev;
-	struct qeth_card *card = (struct qeth_card *) dev->priv;
+	struct qeth_card *card = dev->ml_priv;
 	struct in_device *in_dev;
 
 	if (card->info.type == QETH_CARD_TYPE_IQD)
@@ -1851,7 +1851,7 @@ static void qeth_l3_vlan_rx_add_vid(struct net_device *dev, unsigned short vid)
 
 static void qeth_l3_vlan_rx_kill_vid(struct net_device *dev, unsigned short vid)
 {
-	struct qeth_card *card = netdev_priv(dev);
+	struct qeth_card *card = dev->ml_priv;
 	unsigned long flags;
 
 	QETH_DBF_TEXT_(TRACE, 4, "kid:%d", vid);
@@ -2013,7 +2013,7 @@ static int qeth_l3_verify_vlan_dev(struct net_device *dev,
 		}
 	}
 
-	if (rc && !(netdev_priv(vlan_dev_real_dev(dev)) == (void *)card))
+	if (rc && !(vlan_dev_real_dev(dev)->ml_priv == (void *)card))
 		return 0;
 
 	return rc;
@@ -2047,9 +2047,9 @@ static struct qeth_card *qeth_l3_get_card_from_dev(struct net_device *dev)
 
 	rc = qeth_l3_verify_dev(dev);
 	if (rc == QETH_REAL_CARD)
-		card = netdev_priv(dev);
+		card = dev->ml_priv;
 	else if (rc == QETH_VLAN_CARD)
-		card = netdev_priv(vlan_dev_real_dev(dev));
+		card = vlan_dev_real_dev(dev)->ml_priv;
 	if (card && card->options.layer2)
 		card = NULL;
 	QETH_DBF_TEXT_(TRACE, 4, "%d", rc);
@@ -2110,7 +2110,7 @@ static int qeth_l3_stop_card(struct qeth_card *card, int recovery_mode)
 
 static void qeth_l3_set_multicast_list(struct net_device *dev)
 {
-	struct qeth_card *card = netdev_priv(dev);
+	struct qeth_card *card = dev->ml_priv;
 
 	QETH_DBF_TEXT(TRACE, 3, "setmulti");
 	qeth_l3_delete_mc_addresses(card);
@@ -2438,7 +2438,7 @@ static int qeth_l3_arp_flush_cache(struct qeth_card *card)
 
 static int qeth_l3_do_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
 {
-	struct qeth_card *card = netdev_priv(dev);
+	struct qeth_card *card = dev->ml_priv;
 	struct qeth_arp_cache_entry arp_entry;
 	struct mii_ioctl_data *mii_data;
 	int rc = 0;
@@ -2595,7 +2595,7 @@ static int qeth_l3_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	u16 *tag;
 	struct qeth_hdr *hdr = NULL;
 	int elements_needed = 0;
-	struct qeth_card *card = netdev_priv(dev);
+	struct qeth_card *card = dev->ml_priv;
 	struct sk_buff *new_skb = NULL;
 	int ipv = qeth_get_ip_version(skb);
 	int cast_type = qeth_get_cast_type(card, skb);
@@ -2763,7 +2763,7 @@ tx_drop:
 
 static int qeth_l3_open(struct net_device *dev)
 {
-	struct qeth_card *card = netdev_priv(dev);
+	struct qeth_card *card = dev->ml_priv;
 
 	QETH_DBF_TEXT(TRACE, 4, "qethopen");
 	if (card->state != CARD_STATE_SOFTSETUP)
@@ -2780,7 +2780,7 @@ static int qeth_l3_open(struct net_device *dev)
 
 static int qeth_l3_stop(struct net_device *dev)
 {
-	struct qeth_card *card = netdev_priv(dev);
+	struct qeth_card *card = dev->ml_priv;
 
 	QETH_DBF_TEXT(TRACE, 4, "qethstop");
 	netif_tx_disable(dev);
@@ -2792,14 +2792,14 @@ static int qeth_l3_stop(struct net_device *dev)
 
 static u32 qeth_l3_ethtool_get_rx_csum(struct net_device *dev)
 {
-	struct qeth_card *card = netdev_priv(dev);
+	struct qeth_card *card = dev->ml_priv;
 
 	return (card->options.checksum_type == HW_CHECKSUMMING);
 }
 
 static int qeth_l3_ethtool_set_rx_csum(struct net_device *dev, u32 data)
 {
-	struct qeth_card *card = netdev_priv(dev);
+	struct qeth_card *card = dev->ml_priv;
 	enum qeth_card_states old_state;
 	enum qeth_checksum_types csum_type;
 
@@ -2825,7 +2825,7 @@ static int qeth_l3_ethtool_set_rx_csum(struct net_device *dev, u32 data)
 
 static int qeth_l3_ethtool_set_tso(struct net_device *dev, u32 data)
 {
-	struct qeth_card *card = netdev_priv(dev);
+	struct qeth_card *card = dev->ml_priv;
 
 	if (data) {
 		if (card->options.large_send == QETH_LARGE_SEND_NO) {
@@ -2915,7 +2915,7 @@ static int qeth_l3_setup_netdev(struct qeth_card *card)
 		return -ENODEV;
 
 	card->dev->hard_start_xmit = qeth_l3_hard_start_xmit;
-	card->dev->priv = card;
+	card->dev->ml_priv = card;
 	card->dev->tx_timeout = &qeth_tx_timeout;
 	card->dev->watchdog_timeo = QETH_TX_TIMEOUT;
 	card->dev->open = qeth_l3_open;

From 36ac26171afa8dbf29226199699fe955d4a0b6f6 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Sat, 26 Jul 2008 11:22:33 +0200
Subject: [PATCH 0966/1171] crashdump: fix undefined reference to
 `elfcorehdr_addr'

fix build bug introduced by 95b68dec0d5 "calgary iommu: use the first
kernels TCE tables in kdump":

arch/x86/kernel/built-in.o: In function `calgary_iommu_init':
(.init.text+0x8399): undefined reference to `elfcorehdr_addr'
arch/x86/kernel/built-in.o: In function `calgary_iommu_init':
(.init.text+0x856c): undefined reference to `elfcorehdr_addr'
arch/x86/kernel/built-in.o: In function `detect_calgary':
(.init.text+0x8c68): undefined reference to `elfcorehdr_addr'
arch/x86/kernel/built-in.o: In function `detect_calgary':
(.init.text+0x8d0c): undefined reference to `elfcorehdr_addr'

make elfcorehdr_addr a generally available symbol.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/linux/crash_dump.h | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/include/linux/crash_dump.h b/include/linux/crash_dump.h
index 6cd39a927e1f..025e4f575103 100644
--- a/include/linux/crash_dump.h
+++ b/include/linux/crash_dump.h
@@ -8,7 +8,13 @@
 #include <linux/proc_fs.h>
 
 #define ELFCORE_ADDR_MAX	(-1ULL)
+
+#ifdef CONFIG_PROC_VMCORE
 extern unsigned long long elfcorehdr_addr;
+#else
+static const unsigned long long elfcorehdr_addr = ELFCORE_ADDR_MAX;
+#endif
+
 extern ssize_t copy_oldmem_page(unsigned long, char *, size_t,
 						unsigned long, int);
 extern const struct file_operations proc_vmcore_operations;

From cdec7e50a4896c5197d5575d9ca635eea6825149 Mon Sep 17 00:00:00 2001
From: "David S. Miller" <davem@davemloft.net>
Date: Sat, 26 Jul 2008 02:28:09 -0700
Subject: [PATCH 0967/1171] Revert "pkt_sched: sch_sfq: dump a real number of
 flows"

This reverts commit f867e6af94239a04ec23aeec2fcda5aa58e41db7.

Based upon discussions between Jarek and Patrick McHardy
this is field being set is more a config parameter than a
statistic.  And we should add a true statistic to provide
this information if we really want it.

Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/sched/sch_sfq.c | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

diff --git a/net/sched/sch_sfq.c b/net/sched/sch_sfq.c
index 73f53844ce97..8589da666568 100644
--- a/net/sched/sch_sfq.c
+++ b/net/sched/sch_sfq.c
@@ -536,14 +536,7 @@ static int sfq_dump(struct Qdisc *sch, struct sk_buff *skb)
 
 	opt.limit = q->limit;
 	opt.divisor = SFQ_HASH_DIVISOR;
-	opt.flows = 0;
-	if (q->tail != SFQ_DEPTH) {
-		unsigned int i;
-
-		for (i = 0; i < SFQ_HASH_DIVISOR; i++)
-			if (q->ht[i] != SFQ_DEPTH)
-				opt.flows++;
-	}
+	opt.flows = q->limit;
 
 	NLA_PUT(skb, TCA_OPTIONS, sizeof(opt), &opt);
 

From b1b154e503a0e590eb9e189586783dc8750f910e Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier.adi@gmail.com>
Date: Sat, 26 Jul 2008 18:02:05 +0800
Subject: [PATCH 0968/1171] Blackfin arch: check the EXTBANKS field of the
 DDRCTL1 register to see if we are using both memory banks

Signed-off-by: Mike Frysinger <vapier.adi@gmail.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/kernel/setup.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/arch/blackfin/kernel/setup.c b/arch/blackfin/kernel/setup.c
index 8671d1db1f99..102a9db98559 100644
--- a/arch/blackfin/kernel/setup.c
+++ b/arch/blackfin/kernel/setup.c
@@ -700,6 +700,8 @@ static inline int __init get_mem_size(void)
 		case DEVWD_8:  ret *= 2;
 		case DEVWD_16: break;
 	}
+	if ((ddrctl & 0xc000) == 0x4000)
+		ret *= 2;
 	return ret;
 #endif
 	BUG();

From 377d43e7d07fc9a64eb162c6f70e6d1961604629 Mon Sep 17 00:00:00 2001
From: Michael Hennerich <michael.hennerich@analog.com>
Date: Sat, 26 Jul 2008 18:28:03 +0800
Subject: [PATCH 0969/1171] Blackfin arch: fix bug - IMDMA is not type struct
 dma_register

TEMP Workaround - avoid access to PERIPHERAL_MAP

Signed-off-by: Michael Hennerich <michael.hennerich@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/kernel/bfin_dma_5xx.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/arch/blackfin/kernel/bfin_dma_5xx.c b/arch/blackfin/kernel/bfin_dma_5xx.c
index ad0e75845ac2..93229b3d6e3e 100644
--- a/arch/blackfin/kernel/bfin_dma_5xx.c
+++ b/arch/blackfin/kernel/bfin_dma_5xx.c
@@ -477,7 +477,11 @@ int blackfin_dma_suspend(void)
 {
 	int i;
 
+#ifdef CONFIG_BF561	/* IMDMA channels doesn't have a PERIPHERAL_MAP */
+	for (i = 0; i <= CH_MEM_STREAM3_SRC; i++) {
+#else
 	for (i = 0; i < MAX_BLACKFIN_DMA_CHANNEL; i++) {
+#endif
 		if (dma_ch[i].chan_status == DMA_CHANNEL_ENABLED) {
 			printk(KERN_ERR "DMA Channel %d failed to suspend\n", i);
 			return -EBUSY;
@@ -493,7 +497,11 @@ void blackfin_dma_resume(void)
 {
 	int i;
 
+#ifdef CONFIG_BF561	/* IMDMA channels doesn't have a PERIPHERAL_MAP */
+	for (i = 0; i <= CH_MEM_STREAM3_SRC; i++)
+#else
 	for (i = 0; i < MAX_BLACKFIN_DMA_CHANNEL; i++)
+#endif
 		dma_ch[i].regs->peripheral_map = dma_ch[i].saved_peripheral_map;
 }
 #endif

From 3a2521fa75359450f5ec7e6b76847f933b7ee680 Mon Sep 17 00:00:00 2001
From: Mike Frysinger <vapier.adi@gmail.com>
Date: Sat, 26 Jul 2008 18:52:56 +0800
Subject: [PATCH 0970/1171] Blackfin arch: cache the values of vco/sclk/cclk as
 the overhead of doing so (~24 bytes) is worth avoiding the software mult/div
 routines

Signed-off-by: Mike Frysinger <vapier.adi@gmail.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/kernel/setup.c | 44 ++++++++++++++++++++++++++++--------
 1 file changed, 34 insertions(+), 10 deletions(-)

diff --git a/arch/blackfin/kernel/setup.c b/arch/blackfin/kernel/setup.c
index 102a9db98559..5a455b6bb542 100644
--- a/arch/blackfin/kernel/setup.c
+++ b/arch/blackfin/kernel/setup.c
@@ -850,38 +850,55 @@ static int __init topology_init(void)
 
 subsys_initcall(topology_init);
 
+/* Get the voltage input multiplier */
+static u_long cached_vco_pll_ctl, cached_vco;
 static u_long get_vco(void)
 {
 	u_long msel;
-	u_long vco;
 
-	msel = (bfin_read_PLL_CTL() >> 9) & 0x3F;
+	u_long pll_ctl = bfin_read_PLL_CTL();
+	if (pll_ctl == cached_vco_pll_ctl)
+		return cached_vco;
+	else
+		cached_vco_pll_ctl = pll_ctl;
+
+	msel = (pll_ctl >> 9) & 0x3F;
 	if (0 == msel)
 		msel = 64;
 
-	vco = CONFIG_CLKIN_HZ;
-	vco >>= (1 & bfin_read_PLL_CTL());	/* DF bit */
-	vco = msel * vco;
-	return vco;
+	cached_vco = CONFIG_CLKIN_HZ;
+	cached_vco >>= (1 & pll_ctl);	/* DF bit */
+	cached_vco *= msel;
+	return cached_vco;
 }
 
 /* Get the Core clock */
+static u_long cached_cclk_pll_div, cached_cclk;
 u_long get_cclk(void)
 {
 	u_long csel, ssel;
+
 	if (bfin_read_PLL_STAT() & 0x1)
 		return CONFIG_CLKIN_HZ;
 
 	ssel = bfin_read_PLL_DIV();
+	if (ssel == cached_cclk_pll_div)
+		return cached_cclk;
+	else
+		cached_cclk_pll_div = ssel;
+
 	csel = ((ssel >> 4) & 0x03);
 	ssel &= 0xf;
 	if (ssel && ssel < (1 << csel))	/* SCLK > CCLK */
-		return get_vco() / ssel;
-	return get_vco() >> csel;
+		cached_cclk = get_vco() / ssel;
+	else
+		cached_cclk = get_vco() >> csel;
+	return cached_cclk;
 }
 EXPORT_SYMBOL(get_cclk);
 
 /* Get the System clock */
+static u_long cached_sclk_pll_div, cached_sclk;
 u_long get_sclk(void)
 {
 	u_long ssel;
@@ -889,13 +906,20 @@ u_long get_sclk(void)
 	if (bfin_read_PLL_STAT() & 0x1)
 		return CONFIG_CLKIN_HZ;
 
-	ssel = (bfin_read_PLL_DIV() & 0xf);
+	ssel = bfin_read_PLL_DIV();
+	if (ssel == cached_sclk_pll_div)
+		return cached_sclk;
+	else
+		cached_sclk_pll_div = ssel;
+
+	ssel &= 0xf;
 	if (0 == ssel) {
 		printk(KERN_WARNING "Invalid System Clock\n");
 		ssel = 1;
 	}
 
-	return get_vco() / ssel;
+	cached_sclk = get_vco() / ssel;
+	return cached_sclk;
 }
 EXPORT_SYMBOL(get_sclk);
 

From 0d1cdd7ab6e0e7ccaf9f3b1d2afa0ddeead23ccc Mon Sep 17 00:00:00 2001
From: Sonic Zhang <sonic.zhang@analog.com>
Date: Sat, 26 Jul 2008 18:54:38 +0800
Subject: [PATCH 0971/1171] Blackfin arch: Fix bug - skip single step in high
 priority interrupt handler instead of disabling all interrupts in single step
 debugging.

Skip single step if event priority of current instruction is higher than
that of the first instruction, from which gdb starts single step.

Signed-off-by: Sonic Zhang <sonic.zhang@analog.com>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/kernel/kgdb.c       | 16 +++++++++-
 arch/blackfin/mach-common/entry.S | 50 +++++++++++++++++++++++--------
 2 files changed, 53 insertions(+), 13 deletions(-)

diff --git a/arch/blackfin/kernel/kgdb.c b/arch/blackfin/kernel/kgdb.c
index a9c15515bfd7..a1f9641a6425 100644
--- a/arch/blackfin/kernel/kgdb.c
+++ b/arch/blackfin/kernel/kgdb.c
@@ -203,6 +203,8 @@ struct hw_breakpoint {
 
 int kgdb_arch_init(void)
 {
+	debugger_step = 0;
+
 	kgdb_remove_all_hw_break();
 	return 0;
 }
@@ -368,6 +370,7 @@ int kgdb_arch_handle_exception(int exceptionVector, int signo,
 	char *ptr;
 	int newPC;
 	int wp_status;
+	int i;
 
 	switch (remcom_in_buffer[0]) {
 	case 'c':
@@ -392,7 +395,18 @@ int kgdb_arch_handle_exception(int exceptionVector, int signo,
 		/* set the trace bit if we're stepping */
 		if (remcom_in_buffer[0] == 's') {
 			linux_regs->syscfg |= 0x1;
-			debugger_step = 1;
+			debugger_step = linux_regs->ipend;
+			debugger_step >>= 6;
+			for (i = 10; i > 0; i--, debugger_step >>= 1)
+				if (debugger_step & 1)
+					break;
+			/* i indicate event priority of current stopped instruction
+			 * user space instruction is 0, IVG15 is 1, IVTMR is 10.
+			 * debugger_step > 0 means in single step mode
+			 */
+			debugger_step = i + 1;
+		} else {
+			debugger_step = 0;
 		}
 
 		wp_status = bfin_read_WPSTAT();
diff --git a/arch/blackfin/mach-common/entry.S b/arch/blackfin/mach-common/entry.S
index 038f70e0be65..eceb484d90f9 100644
--- a/arch/blackfin/mach-common/entry.S
+++ b/arch/blackfin/mach-common/entry.S
@@ -158,23 +158,45 @@ ENTRY(_ex_single_step)
 	cc = r7 == r6;
 	if cc jump _bfin_return_from_exception;
 
+	/* Don't do single step in hardware exception handler */
+        p5.l = lo(IPEND);
+        p5.h = hi(IPEND);
+	r6 = [p5];
+	cc = bittst(r6, 5);
+	if cc jump _bfin_return_from_exception;
+
+#ifdef CONFIG_KGDB
+	/* skip single step if current interrupt priority is higher than
+	 * that of the first instruction, from which gdb starts single step */
+	r6 >>= 6;
+	r7 = 10;
+.Lfind_priority_start:
+	cc = bittst(r6, 0);
+	if cc jump .Lfind_priority_done;
+	r6 >>= 1;
+	r7 += -1;
+	cc = r7 == 0;
+	if cc jump .Lfind_priority_done;
+	jump.s .Lfind_priority_start;
+.Lfind_priority_done:
+	p4.l = _debugger_step;
+	p4.h = _debugger_step;
+	r6 = [p4];
+	cc = r6 == 0;
+	if cc jump .Ldo_single_step;
+	r6 += -1;
+	cc = r6 < r7;
+	if cc jump _bfin_return_from_exception;
+.Ldo_single_step:
+#endif
+
 	/* If we were in user mode, do the single step normally.  */
-	p5.l = lo(IPEND);
-	p5.h = hi(IPEND);
 	r6 = [p5];
 	r7 = 0xffe0 (z);
 	r7 = r7 & r6;
 	cc = r7 == 0;
-	if !cc jump 1f;
+	if cc jump 1f;
 
-	/* Single stepping only a single instruction, so clear the trace
-	 * bit here.  */
-	r7 = syscfg;
-	bitclr (r7, 0);
-	syscfg = R7;
-	jump _ex_trap_c;
-
-1:
 	/*
 	 * We were in an interrupt handler.  By convention, all of them save
 	 * SYSCFG with their first instruction, so by checking whether our
@@ -202,11 +224,15 @@ ENTRY(_ex_single_step)
 	cc = R7 == R6;
 	if !cc jump _bfin_return_from_exception;
 
+1:
+	/* Single stepping only a single instruction, so clear the trace
+	 * bit here.  */
 	r7 = syscfg;
 	bitclr (r7, 0);
 	syscfg = R7;
 
-	/* Fall through to _bfin_return_from_exception.  */
+	jump _ex_trap_c;
+
 ENDPROC(_ex_single_step)
 
 ENTRY(_bfin_return_from_exception)

From 9de3a0b6979a4839d67ca840e386ea06acaabe39 Mon Sep 17 00:00:00 2001
From: Robin Getz <rgetz@blackfin.uclinux.org>
Date: Sat, 26 Jul 2008 19:39:19 +0800
Subject: [PATCH 0972/1171] Blackfin arch: When icache is off, make sure people
 know it

Signed-off-by: Robin Getz <rgetz@blackfin.uclinux.org>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/kernel/setup.c | 17 ++++++++++++-----
 1 file changed, 12 insertions(+), 5 deletions(-)

diff --git a/arch/blackfin/kernel/setup.c b/arch/blackfin/kernel/setup.c
index 5a455b6bb542..c2248fe80588 100644
--- a/arch/blackfin/kernel/setup.c
+++ b/arch/blackfin/kernel/setup.c
@@ -948,7 +948,7 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 	uint32_t revid;
 
 	u_long cclk = 0, sclk = 0;
-	u_int dcache_size = 0, dsup_banks = 0;
+	u_int icache_size = BFIN_ICACHESIZE / 1024, dcache_size = 0, dsup_banks = 0;
 
 	cpu = CPU;
 	mmu = "none";
@@ -1017,12 +1017,15 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 	}
 
 	/* Is it turned on? */
-	if (!((bfin_read_DMEM_CONTROL()) & (ENDCPLB | DMC_ENABLE)))
+	if (bfin_read_DMEM_CONTROL() & (ENDCPLB | DMC_ENABLE) != (ENDCPLB | DMC_ENABLE))
 		dcache_size = 0;
 
+	if (bfin_read_IMEM_CONTROL() & (IMC | ENICPLB) == (IMC | ENICPLB))
+		icache_size = 0;
+
 	seq_printf(m, "cache size\t: %d KB(L1 icache) "
 		"%d KB(L1 dcache-%s) %d KB(L2 cache)\n",
-		BFIN_ICACHESIZE / 1024, dcache_size,
+		icache_size, dcache_size,
 #if defined CONFIG_BFIN_WB
 		"wb"
 #elif defined CONFIG_BFIN_WT
@@ -1032,8 +1035,12 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 
 	seq_printf(m, "%s\n", cache);
 
-	seq_printf(m, "icache setup\t: %d Sub-banks/%d Ways, %d Lines/Way\n",
-		   BFIN_ISUBBANKS, BFIN_IWAYS, BFIN_ILINES);
+	if (icache_size)
+		seq_printf(m, "icache setup\t: %d Sub-banks/%d Ways, %d Lines/Way\n",
+			   BFIN_ISUBBANKS, BFIN_IWAYS, BFIN_ILINES);
+	else
+		seq_printf(m, "icache setup\t: off\n");
+
 	seq_printf(m,
 		   "dcache setup\t: %d Super-banks/%d Sub-banks/%d Ways, %d Lines/Way\n",
 		   dsup_banks, BFIN_DSUBBANKS, BFIN_DWAYS,

From 2d2009806dd843f3adc0cbbb5d2204980f28111a Mon Sep 17 00:00:00 2001
From: Robin Getz <rgetz@blackfin.uclinux.org>
Date: Sat, 26 Jul 2008 19:41:40 +0800
Subject: [PATCH 0973/1171] Blackfin arch: If we double fault, rather than hang
 forever, reset

Signed-off-by: Robin Getz <rgetz@blackfin.uclinux.org>
Signed-off-by: Bryan Wu <cooloney@kernel.org>
---
 arch/blackfin/kernel/setup.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/arch/blackfin/kernel/setup.c b/arch/blackfin/kernel/setup.c
index c2248fe80588..23e637eb78da 100644
--- a/arch/blackfin/kernel/setup.c
+++ b/arch/blackfin/kernel/setup.c
@@ -771,6 +771,9 @@ void __init setup_arch(char **cmdline_p)
 
 	_bfin_swrst = bfin_read_SWRST();
 
+	/* If we double fault, reset the system - otherwise we hang forever */
+	bfin_write_SWRST(DOUBLE_FAULT);
+
 	if (_bfin_swrst & RESET_DOUBLE)
 		printk(KERN_INFO "Recovering from Double Fault event\n");
 	else if (_bfin_swrst & RESET_WDOG)
@@ -1017,10 +1020,10 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 	}
 
 	/* Is it turned on? */
-	if (bfin_read_DMEM_CONTROL() & (ENDCPLB | DMC_ENABLE) != (ENDCPLB | DMC_ENABLE))
+	if ((bfin_read_DMEM_CONTROL() & (ENDCPLB | DMC_ENABLE)) != (ENDCPLB | DMC_ENABLE))
 		dcache_size = 0;
 
-	if (bfin_read_IMEM_CONTROL() & (IMC | ENICPLB) == (IMC | ENICPLB))
+	if ((bfin_read_IMEM_CONTROL() & (IMC | ENICPLB)) == (IMC | ENICPLB))
 		icache_size = 0;
 
 	seq_printf(m, "cache size\t: %d KB(L1 icache) "

From 1f972768a1df1518f45adb6b8ffbf04fa1c99737 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Sat, 26 Jul 2008 13:52:50 +0200
Subject: [PATCH 0974/1171] x86, RDC321x: add to mach-default

first step to add RDC321x support to the default PC architecture.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/Kconfig  | 26 +++++++++++---------------
 arch/x86/Makefile |  5 -----
 2 files changed, 11 insertions(+), 20 deletions(-)

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index e3cba0b45600..39ae67985950 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -23,7 +23,7 @@ config X86
 	select HAVE_OPROFILE
 	select HAVE_IOREMAP_PROT
 	select HAVE_KPROBES
-	select ARCH_WANT_OPTIONAL_GPIOLIB if !X86_RDC321X
+	select ARCH_WANT_OPTIONAL_GPIOLIB
 	select HAVE_KRETPROBES
 	select HAVE_DYNAMIC_FTRACE
 	select HAVE_FTRACE
@@ -332,20 +332,6 @@ config X86_BIGSMP
 
 endif
 
-config X86_RDC321X
-	bool "RDC R-321x SoC"
-	depends on X86_32
-	select M486
-	select X86_REBOOTFIXUPS
-	select GENERIC_GPIO
-	select LEDS_CLASS
-	select LEDS_GPIO
-	select NEW_LEDS
-	help
-	  This option is needed for RDC R-321x system-on-chip, also known
-	  as R-8610-(G).
-	  If you don't have one of these chips, you should say N here.
-
 config X86_VSMP
 	bool "Support for ScaleMP vSMP"
 	select PARAVIRT
@@ -369,6 +355,16 @@ config X86_VISWS
 	  A kernel compiled for the Visual Workstation will run on general
 	  PCs as well. See <file:Documentation/sgi-visws.txt> for details.
 
+config X86_RDC321X
+	bool "RDC R-321x SoC"
+	depends on X86_32
+	select M486
+	select X86_REBOOTFIXUPS
+	help
+	  This option is needed for RDC R-321x system-on-chip, also known
+	  as R-8610-(G).
+	  If you don't have one of these chips, you should say N here.
+
 config SCHED_NO_NO_OMIT_FRAME_POINTER
 	def_bool y
 	prompt "Single-depth WCHAN output"
diff --git a/arch/x86/Makefile b/arch/x86/Makefile
index 919ce21ea654..f5631da585b6 100644
--- a/arch/x86/Makefile
+++ b/arch/x86/Makefile
@@ -118,11 +118,6 @@ mflags-$(CONFIG_X86_GENERICARCH):= -Iinclude/asm-x86/mach-generic
 fcore-$(CONFIG_X86_GENERICARCH)	+= arch/x86/mach-generic/
 mcore-$(CONFIG_X86_GENERICARCH)	:= arch/x86/mach-default/
 
-# RDC R-321x subarch support
-mflags-$(CONFIG_X86_RDC321X)	:= -Iinclude/asm-x86/mach-rdc321x
-mcore-$(CONFIG_X86_RDC321X)	:= arch/x86/mach-default/
-core-$(CONFIG_X86_RDC321X)	+= arch/x86/mach-rdc321x/
-
 # default subarch .h files
 mflags-y += -Iinclude/asm-x86/mach-default
 

From dcf309974555d17019c6a8e1a425238f17990b71 Mon Sep 17 00:00:00 2001
From: Steven Rostedt <rostedt@goodmis.org>
Date: Fri, 25 Jul 2008 18:00:42 -0400
Subject: [PATCH 0975/1171] ftrace: disable tracing on acpi idle calls

The acpi idle waits calls local_irq_save and then uses mwait to go into
idle. The tracer gets reenabled at local_irq_save but does not detect that
the idle allows for wake ups.

This patch adds code to disable the tracing when acpi puts the CPU to idle.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 drivers/acpi/processor_idle.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c
index d592dbb1d12a..b7f2963693a7 100644
--- a/drivers/acpi/processor_idle.c
+++ b/drivers/acpi/processor_idle.c
@@ -272,6 +272,8 @@ static atomic_t c3_cpu_count;
 /* Common C-state entry for C2, C3, .. */
 static void acpi_cstate_enter(struct acpi_processor_cx *cstate)
 {
+	/* Don't trace irqs off for idle */
+	stop_critical_timings();
 	if (cstate->entry_method == ACPI_CSTATE_FFH) {
 		/* Call into architectural FFH based C-state */
 		acpi_processor_ffh_cstate_enter(cstate);
@@ -284,6 +286,7 @@ static void acpi_cstate_enter(struct acpi_processor_cx *cstate)
 		   gets asserted in time to freeze execution properly. */
 		unused = inl(acpi_gbl_FADT.xpm_timer_block.address);
 	}
+	start_critical_timings();
 }
 #endif /* !CONFIG_CPU_IDLE */
 
@@ -1418,6 +1421,8 @@ static inline void acpi_idle_update_bm_rld(struct acpi_processor *pr,
  */
 static inline void acpi_idle_do_entry(struct acpi_processor_cx *cx)
 {
+	/* Don't trace irqs off for idle */
+	stop_critical_timings();
 	if (cx->entry_method == ACPI_CSTATE_FFH) {
 		/* Call into architectural FFH based C-state */
 		acpi_processor_ffh_cstate_enter(cx);
@@ -1432,6 +1437,7 @@ static inline void acpi_idle_do_entry(struct acpi_processor_cx *cx)
 		   gets asserted in time to freeze execution properly. */
 		unused = inl(acpi_gbl_FADT.xpm_timer_block.address);
 	}
+	start_critical_timings();
 }
 
 /**

From 071375bc76ee86e58592f4682030c81d410ddfd9 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Sat, 26 Jul 2008 14:52:26 +0200
Subject: [PATCH 0976/1171] x86, RDC321x: remove gpio.h complications

Remove the include/asm-x86/gpio.h specials, just use the generic
version.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/gpio.h | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/include/asm-x86/gpio.h b/include/asm-x86/gpio.h
index 116e9147fe66..c4c91b37c104 100644
--- a/include/asm-x86/gpio.h
+++ b/include/asm-x86/gpio.h
@@ -16,10 +16,6 @@
 #ifndef _ASM_I386_GPIO_H
 #define _ASM_I386_GPIO_H
 
-#ifdef CONFIG_X86_RDC321X
-#include <gpio.h>
-#else /* CONFIG_X86_RDC321X */
-
 #include <asm-generic/gpio.h>
 
 #ifdef CONFIG_GPIOLIB
@@ -57,6 +53,4 @@ static inline int irq_to_gpio(unsigned int irq)
 
 #endif /* CONFIG_GPIOLIB */
 
-#endif /* CONFIG_X86_RDC321X */
-
 #endif /* _ASM_I386_GPIO_H */

From 1ca9fda4b2f3413e4bc748b983f5585629ca0560 Mon Sep 17 00:00:00 2001
From: Chris McDermott <lcm@linux.vnet.ibm.com>
Date: Thu, 24 Jul 2008 19:06:09 -0700
Subject: [PATCH 0977/1171] x86: fix IBM Summit based systems'
 phys_cpu_present_map on 32-bit kernels

x86 kernels on IBM Summit based systems will only online 1 CPU because the
phys_cpu_present_map is not set up correctly. Patch below applied to
2.6.26-git10.

Signed-off-by: Chris McDermott <lcm@linux.vnet.ibm.com>
Tested-by: Tim Pepper <lnxninga@linux.vnet.ibm.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 include/asm-x86/mach-summit/mach_apic.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/asm-x86/mach-summit/mach_apic.h b/include/asm-x86/mach-summit/mach_apic.h
index 75d2c95005d7..c47e2ab5c5ca 100644
--- a/include/asm-x86/mach-summit/mach_apic.h
+++ b/include/asm-x86/mach-summit/mach_apic.h
@@ -122,7 +122,7 @@ static inline physid_mask_t ioapic_phys_id_map(physid_mask_t phys_id_map)
 
 static inline physid_mask_t apicid_to_cpu_present(int apicid)
 {
-	return physid_mask_of_physid(0);
+	return physid_mask_of_physid(apicid);
 }
 
 static inline void setup_portio_remap(void)

From 1fe371044b21b226b96a9dd959e971b50b28c78e Mon Sep 17 00:00:00 2001
From: Ingo Molnar <mingo@elte.hu>
Date: Sat, 26 Jul 2008 15:09:47 +0200
Subject: [PATCH 0978/1171] ftrace: fix modular build

fix:

 ERROR: "start_critical_timings" [drivers/acpi/processor.ko] undefined!
 ERROR: "stop_critical_timings" [drivers/acpi/processor.ko] undefined!

Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 kernel/trace/trace_irqsoff.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c
index b1e4a89b08eb..ece6cfb649fa 100644
--- a/kernel/trace/trace_irqsoff.c
+++ b/kernel/trace/trace_irqsoff.c
@@ -253,12 +253,14 @@ void start_critical_timings(void)
 	if (preempt_trace() || irq_trace())
 		start_critical_timing(CALLER_ADDR0, CALLER_ADDR1);
 }
+EXPORT_SYMBOL_GPL(start_critical_timings);
 
 void stop_critical_timings(void)
 {
 	if (preempt_trace() || irq_trace())
 		stop_critical_timing(CALLER_ADDR0, CALLER_ADDR1);
 }
+EXPORT_SYMBOL_GPL(stop_critical_timings);
 
 #ifdef CONFIG_IRQSOFF_TRACER
 #ifdef CONFIG_PROVE_LOCKING

From 3a61ec387c9092dfc91a5959145d36835a72fc4c Mon Sep 17 00:00:00 2001
From: Joerg Roedel <joerg.roedel@amd.com>
Date: Fri, 25 Jul 2008 13:07:50 +0200
Subject: [PATCH 0979/1171] x86, AMD IOMMU: include amd_iommu_last_bdf in
 device initialization

All the values read while searching for amd_iommu_last_bdf are defined as
inclusive. Let the code handle this value as such. Found by Wei Wang. Thanks
Wei.

Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
Cc: iommu@lists.linux-foundation.org
Cc: bhavna.sarathy@amd.com
Cc: robert.richter@amd.com
Cc: Wei Wang <wei.wang2@amd.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
 arch/x86/kernel/amd_iommu.c      | 4 ++--
 arch/x86/kernel/amd_iommu_init.c | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c
index c25210e6ac88..74697408576f 100644
--- a/arch/x86/kernel/amd_iommu.c
+++ b/arch/x86/kernel/amd_iommu.c
@@ -667,7 +667,7 @@ static int get_device_resources(struct device *dev,
 	_bdf = calc_devid(pcidev->bus->number, pcidev->devfn);
 
 	/* device not translated by any IOMMU in the system? */
-	if (_bdf >= amd_iommu_last_bdf) {
+	if (_bdf > amd_iommu_last_bdf) {
 		*iommu = NULL;
 		*domain = NULL;
 		*bdf = 0xffff;
@@ -1085,7 +1085,7 @@ void prealloc_protection_domains(void)
 
 	while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) {
 		devid = (dev->bus->number << 8) | dev->devfn;
-		if (devid >= amd_iommu_last_bdf)
+		if (devid > amd_iommu_last_bdf)
 			continue;
 		devid = amd_iommu_alias_table[devid];
 		if (domain_for_device(devid))
diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c
index c9d8ff2eb130..d9a9da597e79 100644
--- a/arch/x86/kernel/amd_iommu_init.c
+++ b/arch/x86/kernel/amd_iommu_init.c
@@ -732,7 +732,7 @@ static int __init init_exclusion_range(struct ivmd_header *m)
 		set_device_exclusion_range(m->devid, m);
 		break;
 	case ACPI_IVMD_TYPE_ALL:
-		for (i = 0; i < amd_iommu_last_bdf; ++i)
+		for (i = 0; i <= amd_iommu_last_bdf; ++i)
 			set_device_exclusion_range(i, m);
 		break;
 	case ACPI_IVMD_TYPE_RANGE:
@@ -934,7 +934,7 @@ int __init amd_iommu_init(void)
 	/*
 	 * let all alias entries point to itself
 	 */
-	for (i = 0; i < amd_iommu_last_bdf; ++i)
+	for (i = 0; i <= amd_iommu_last_bdf; ++i)
 		amd_iommu_alias_table[i] = i;
 
 	/*

From 0c7ad106e779549792deb307242dece6f3499bb9 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@linux-foundation.org>
Date: Fri, 25 Jul 2008 19:44:35 -0700
Subject: [PATCH 0980/1171] drivers/mmc/host/sdhci.h needs scatterlist.h

alpha:

drivers/mmc/host/sdhci.h:242: error: field 'sg_miter' has incomplete type

Cc: Pierre Ossman <drzeus@drzeus.cx>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/mmc/host/sdhci.h | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h
index a06bf8b89343..e354faee5df0 100644
--- a/drivers/mmc/host/sdhci.h
+++ b/drivers/mmc/host/sdhci.h
@@ -9,6 +9,8 @@
  * your option) any later version.
  */
 
+#include <linux/scatterlist.h>
+
 /*
  * Controller registers
  */

From 16d69265b930f7e2fa9eea381715696f780718f4 Mon Sep 17 00:00:00 2001
From: Andrew Morton <akpm@linux-foundation.org>
Date: Fri, 25 Jul 2008 19:44:36 -0700
Subject: [PATCH 0981/1171] uninline arch_pick_mmap_layout()

Fix this, on avr32:

  include/linux/utsname.h:35,
                   from init/main.c:20:
  include/linux/sched.h: In function 'arch_pick_mmap_layout':
  include/linux/sched.h:2149: error: implicit declaration of function 'PAGE_ALIGN'

Reported-by: Adrian Bunk <bunk@kernel.org>
Cc: Haavard Skinnemoen <hskinnemoen@atmel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/sched.h |  9 ---------
 mm/util.c             | 10 ++++++++++
 2 files changed, 10 insertions(+), 9 deletions(-)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 42036ffe6b00..3260a5c42b91 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -2139,16 +2139,7 @@ static inline void set_task_cpu(struct task_struct *p, unsigned int cpu)
 
 #endif /* CONFIG_SMP */
 
-#ifdef HAVE_ARCH_PICK_MMAP_LAYOUT
 extern void arch_pick_mmap_layout(struct mm_struct *mm);
-#else
-static inline void arch_pick_mmap_layout(struct mm_struct *mm)
-{
-	mm->mmap_base = TASK_UNMAPPED_BASE;
-	mm->get_unmapped_area = arch_get_unmapped_area;
-	mm->unmap_area = arch_unmap_area;
-}
-#endif
 
 #ifdef CONFIG_TRACING
 extern void
diff --git a/mm/util.c b/mm/util.c
index 8f18683825bc..0efd83097ecf 100644
--- a/mm/util.c
+++ b/mm/util.c
@@ -1,3 +1,4 @@
+#include <linux/mm.h>
 #include <linux/slab.h>
 #include <linux/string.h>
 #include <linux/module.h>
@@ -136,3 +137,12 @@ char *strndup_user(const char __user *s, long n)
 	return p;
 }
 EXPORT_SYMBOL(strndup_user);
+
+#ifndef HAVE_ARCH_PICK_MMAP_LAYOUT
+void arch_pick_mmap_layout(struct mm_struct *mm)
+{
+	mm->mmap_base = TASK_UNMAPPED_BASE;
+	mm->get_unmapped_area = arch_get_unmapped_area;
+	mm->unmap_area = arch_unmap_area;
+}
+#endif

From 8a21346058ad946134b6ddfeb5de975c3cfcf5da Mon Sep 17 00:00:00 2001
From: Nishanth Aravamudan <nacc@us.ibm.com>
Date: Fri, 25 Jul 2008 19:44:37 -0700
Subject: [PATCH 0982/1171] hugetlb: fix CONFIG_SYSCTL=n build

Fixes a build failure reported by Alan Cox:

mm/hugetlb.c: In function `hugetlb_acct_memory': mm/hugetlb.c:1507:
error: implicit declaration of function `cpuset_mems_nr'

Also reverts Ingo's

    commit e44d1b2998d62a1f2f4d7eb17b56ba396535509f
    Author: Ingo Molnar <mingo@elte.hu>
    Date:   Fri Jul 25 12:57:41 2008 +0200

        mm/hugetlb.c: fix build failure with !CONFIG_SYSCTL

which fixed the build error but added some unused-static-function warnings.

Signed-off-by: Nishanth Aravamudan <nacc@us.ibm.com>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/hugetlb.c | 24 ++++++++++++------------
 1 file changed, 12 insertions(+), 12 deletions(-)

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index a8bf4ab01f86..3be79dc18c5c 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -1026,18 +1026,6 @@ static void __init report_hugepages(void)
 	}
 }
 
-static unsigned int cpuset_mems_nr(unsigned int *array)
-{
-	int node;
-	unsigned int nr = 0;
-
-	for_each_node_mask(node, cpuset_current_mems_allowed)
-		nr += array[node];
-
-	return nr;
-}
-
-#ifdef CONFIG_SYSCTL
 #ifdef CONFIG_HIGHMEM
 static void try_to_free_low(struct hstate *h, unsigned long count)
 {
@@ -1386,6 +1374,18 @@ static int __init hugetlb_default_setup(char *s)
 }
 __setup("default_hugepagesz=", hugetlb_default_setup);
 
+static unsigned int cpuset_mems_nr(unsigned int *array)
+{
+	int node;
+	unsigned int nr = 0;
+
+	for_each_node_mask(node, cpuset_current_mems_allowed)
+		nr += array[node];
+
+	return nr;
+}
+
+#ifdef CONFIG_SYSCTL
 int hugetlb_sysctl_handler(struct ctl_table *table, int write,
 			   struct file *file, void __user *buffer,
 			   size_t *length, loff_t *ppos)

From 44ccac13c7f4728cf2992d49384671a176db74dd Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 19:44:38 -0700
Subject: [PATCH 0983/1171] include/video/atmel_lcdc.h must #include
 <linux/workqueue.h>

This patch fixes the following compile error caused by commit
d22579b837358cbef12ccca5adaf7e93ae09ab7a ("atmel_lcdfb: FIFO underflow
management"):

  In file included from arch/avr32/boards/atstk1000/atstk1004.c:21:
  include/video/atmel_lcdc.h:40: error: field 'task' has incomplete type

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/video/atmel_lcdc.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/include/video/atmel_lcdc.h b/include/video/atmel_lcdc.h
index 1ccf462b433a..613173b5db69 100644
--- a/include/video/atmel_lcdc.h
+++ b/include/video/atmel_lcdc.h
@@ -22,6 +22,7 @@
 #ifndef __ATMEL_LCDC_H__
 #define __ATMEL_LCDC_H__
 
+#include <linux/workqueue.h>
 
 /* Way LCD wires are connected to the chip:
  * Some Atmel chips use BGR color mode (instead of standard RGB)

From a2e2e3577c3ef2b5dbb866e97e612aae4adfa32f Mon Sep 17 00:00:00 2001
From: David Brownell <dbrownell@users.sourceforge.net>
Date: Fri, 25 Jul 2008 19:44:38 -0700
Subject: [PATCH 0984/1171] pm selftest: rtc paranoia

Cope with a quirk of some RTCs (notably ACPI ones) which aren't guaranteed
to implement oneshot behavior when they woke the system from sleeep:
forcibly disable the alarm, just in case.

Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/power/main.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/kernel/power/main.c b/kernel/power/main.c
index 95bff23ecdaa..0b7476f5d2a6 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -635,6 +635,13 @@ static void __init test_wakealarm(struct rtc_device *rtc, suspend_state_t state)
 	}
 	if (status < 0)
 		printk(err_suspend, status);
+
+	/* Some platforms can't detect that the alarm triggered the
+	 * wakeup, or (accordingly) disable it after it afterwards.
+	 * It's supposed to give oneshot behavior; cope.
+	 */
+	alm.enabled = false;
+	rtc_set_alarm(rtc, &alm);
 }
 
 static int __init has_wakealarm(struct device *dev, void *name_ptr)

From a76eef9573c93f8f324ebacfd090a3e319a64d59 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Fri, 25 Jul 2008 19:44:39 -0700
Subject: [PATCH 0985/1171] block/blk-map.c: use the new object_is_on_stack()
 helper

Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
Cc: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
Cc: Tejun Heo <htejun@gmail.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 block/blk-map.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/block/blk-map.c b/block/blk-map.c
index ddd96fb11a7d..af37e4ae62f5 100644
--- a/block/blk-map.c
+++ b/block/blk-map.c
@@ -269,7 +269,6 @@ int blk_rq_map_kern(struct request_queue *q, struct request *rq, void *kbuf,
 	int reading = rq_data_dir(rq) == READ;
 	int do_copy = 0;
 	struct bio *bio;
-	unsigned long stack_mask = ~(THREAD_SIZE - 1);
 
 	if (len > (q->max_hw_sectors << 9))
 		return -EINVAL;
@@ -278,11 +277,8 @@ int blk_rq_map_kern(struct request_queue *q, struct request *rq, void *kbuf,
 
 	kaddr = (unsigned long)kbuf;
 	alignment = queue_dma_alignment(q) | q->dma_pad_mask;
-	do_copy = ((kaddr & alignment) || (len & alignment));
-
-	if (!((kaddr & stack_mask) ^
-	      ((unsigned long)current->stack & stack_mask)))
-		do_copy = 1;
+	do_copy = ((kaddr & alignment) || (len & alignment) ||
+		   object_is_on_stack(kbuf));
 
 	if (do_copy)
 		bio = bio_copy_kern(q, kbuf, len, gfp_mask, reading);

From f3c6ba986ab4527b6dfacf9f3b9e40f72466a8b2 Mon Sep 17 00:00:00 2001
From: Kentaro Makita <k-makita@np.css.fujitsu.com>
Date: Fri, 25 Jul 2008 19:44:40 -0700
Subject: [PATCH 0986/1171] vfs: add cond_resched_lock while scanning dentry
 LRU lists

Add cond_resched_lock(&dcache_lock) while scanning LRU lists on
superblocks in __shrink_dcache_sb()

Signed-off-by: Kentaro Makita <k-makita@np.css.fujitsu.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/dcache.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/fs/dcache.c b/fs/dcache.c
index 3818d6ab76ca..f2584d22cb45 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -487,6 +487,7 @@ restart:
 				if (!cnt)
 					break;
 			}
+			cond_resched_lock(&dcache_lock);
 		}
 	}
 	while (!list_empty(&tmp)) {

From c491b2ffae3fad5e6e3cb2320b46bb8ea8729d49 Mon Sep 17 00:00:00 2001
From: Roel Kluin <12o3l@tiscali.nl>
Date: Fri, 25 Jul 2008 19:44:41 -0700
Subject: [PATCH 0987/1171] asic3: platform_get_irq() may return signed
 unnoticed

asic->irq_nr is unsigned. platform_get_irq() may return signed unnoticed

Signed-off-by: Roel Kluin <12o3l@tiscali.nl>
Cc: Joe Perches <joe@perches.com>
Acked-by: Samuel Ortiz <sameo@openedhand.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/mfd/asic3.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/mfd/asic3.c b/drivers/mfd/asic3.c
index 3b870e7fb3e1..eabf0bfccab4 100644
--- a/drivers/mfd/asic3.c
+++ b/drivers/mfd/asic3.c
@@ -314,10 +314,12 @@ static int __init asic3_irq_probe(struct platform_device *pdev)
 	unsigned long clksel = 0;
 	unsigned int irq, irq_base;
 	int map_size;
+	int ret;
 
-	asic->irq_nr = platform_get_irq(pdev, 0);
-	if (asic->irq_nr < 0)
-		return asic->irq_nr;
+	ret = platform_get_irq(pdev, 0);
+	if (ret < 0)
+		return ret;
+	asic->irq_nr = ret;
 
 	/* turn on clock to IRQ controller */
 	clksel |= CLOCK_SEL_CX;

From e86b19ce64a25d39bb0e10e0e695213fc5993dfb Mon Sep 17 00:00:00 2001
From: Rene Herman <rene.herman@keyaccess.nl>
Date: Fri, 25 Jul 2008 19:44:42 -0700
Subject: [PATCH 0988/1171] pnp: set the pnp_card dma_mask for use by ISAPnP
 cards

dma_alloc_coherent() on x86 currently takes a passed in NULL device
pointer to mean that it should allocate an ISA compatible (24-bit) buffer
which is a bit of a hack.

The ALSA ISA drivers are the main consumers of this but have a struct
device in fact readily available.

For the PnP drivers, the specific pnp_dev->dev device pointer is not
always available at the right time so for now we want to pass the
pnp_card->dev instead which is always available.  Set its dma_mask in
preparation for doing so.

This does not fix a current bug -- 2.6.26-rc1 stumbled over the NULL hack
in dma_alloc_coherent() but this has already been fixed in commit
4a367f3a9dbf2e7ffcee4702203479809236ee6e by Takashi Iwai.

Signed-off-by: Rene Herman <rene.herman@gmail.com>
Acked-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Acked-by: Takashi Iwai <tiwai@suse.de>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/pnp/card.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/pnp/card.c b/drivers/pnp/card.c
index a762a4176736..b00ef1030e45 100644
--- a/drivers/pnp/card.c
+++ b/drivers/pnp/card.c
@@ -8,6 +8,7 @@
 #include <linux/ctype.h>
 #include <linux/slab.h>
 #include <linux/pnp.h>
+#include <linux/dma-mapping.h>
 #include "base.h"
 
 LIST_HEAD(pnp_cards);
@@ -167,6 +168,9 @@ struct pnp_card *pnp_alloc_card(struct pnp_protocol *protocol, int id, char *pnp
 	sprintf(card->dev.bus_id, "%02x:%02x", card->protocol->number,
 		card->number);
 
+	card->dev.coherent_dma_mask = DMA_24BIT_MASK;
+	card->dev.dma_mask = &card->dev.coherent_dma_mask;
+
 	dev_id = pnp_add_card_id(card, pnpid);
 	if (!dev_id) {
 		kfree(card);

From 00412be1d7bdf451653c7dafeb09f4f83398d756 Mon Sep 17 00:00:00 2001
From: Rene Herman <rene.herman@keyaccess.nl>
Date: Fri, 25 Jul 2008 19:44:45 -0700
Subject: [PATCH 0989/1171] isa: set 24-bit dma_mask for ISA devices

dma_alloc_coherent() on x86 currently takes a passed in NULL device
pointer to mean that it should allocate an ISA compatible (24-bit) buffer
which is a bit of a hack.

The ALSA ISA drivers are the main consumers of this but have a struct
device in fact readily available.

For the legacy drivers, this sets the device dma_mask in preparation for
using the actual device with the DMA API so as to eventually not need the
NULL hack in dma_alloc_coherent().

This does not fix a current bug -- 2.6.26-rc1 stumbled over the NULL hack
in dma_alloc_coherent() but this has already been fixed in commit
4a367f3a9dbf2e7ffcee4702203479809236ee6e by Takashi Iwai.

Signed-off-by: Rene Herman <rene.herman@gmail.com>
Cc: Bjorn Helgaas <bjorn.helgaas@hp.com>
Acked-by: Takashi Iwai <tiwai@suse.de>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/base/isa.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/base/isa.c b/drivers/base/isa.c
index d2222397a401..efd577574948 100644
--- a/drivers/base/isa.c
+++ b/drivers/base/isa.c
@@ -7,6 +7,7 @@
 #include <linux/slab.h>
 #include <linux/module.h>
 #include <linux/init.h>
+#include <linux/dma-mapping.h>
 #include <linux/isa.h>
 
 static struct device isa_bus = {
@@ -141,6 +142,9 @@ int isa_register_driver(struct isa_driver *isa_driver, unsigned int ndev)
 		isa_dev->dev.release		= isa_dev_release;
 		isa_dev->id			= id;
 
+		isa_dev->dev.coherent_dma_mask = DMA_24BIT_MASK;
+		isa_dev->dev.dma_mask = &isa_dev->dev.coherent_dma_mask;
+
 		error = device_register(&isa_dev->dev);
 		if (error) {
 			put_device(&isa_dev->dev);

From 999ed65ad12e374d7445fbc13f5a1d146ae4b0da Mon Sep 17 00:00:00 2001
From: Rene Herman <rene.herman@gmail.com>
Date: Fri, 25 Jul 2008 19:44:47 -0700
Subject: [PATCH 0990/1171] pnp: have quirk_system_pci_resources() include io
 resources

quirk_system_pci_resources() disables a PnP mem resource that overlaps a
PCI BAR so as to not keep the PCI driver from claiming the resource.  Have
it do the same for io resources.

Here, ACPI claims ports that overlap with my soundcard causing the
soundcard driver to fail to load.  It's unknown why my ACPI BIOS claims
those ports; it did not use to but this is not a (kernel) regression.
Some odd BIOS reconfig triggered by temporarily removing the card seems to
have brought this on.

Signed-off-by: Rene Herman <rene.herman@gmail.com>
Acked-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/pnp/quirks.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/drivers/pnp/quirks.c b/drivers/pnp/quirks.c
index 55f55ed72dc7..0bdf9b8a5e58 100644
--- a/drivers/pnp/quirks.c
+++ b/drivers/pnp/quirks.c
@@ -245,15 +245,17 @@ static void quirk_system_pci_resources(struct pnp_dev *dev)
 	 */
 	for_each_pci_dev(pdev) {
 		for (i = 0; i < DEVICE_COUNT_RESOURCE; i++) {
-			if (!(pci_resource_flags(pdev, i) & IORESOURCE_MEM) ||
-			    pci_resource_len(pdev, i) == 0)
+			unsigned int type;
+
+			type = pci_resource_flags(pdev, i) &
+					(IORESOURCE_IO | IORESOURCE_MEM);
+			if (!type || pci_resource_len(pdev, i) == 0)
 				continue;
 
 			pci_start = pci_resource_start(pdev, i);
 			pci_end = pci_resource_end(pdev, i);
 			for (j = 0;
-			     (res = pnp_get_resource(dev, IORESOURCE_MEM, j));
-			     j++) {
+			     (res = pnp_get_resource(dev, type, j)); j++) {
 				if (res->start == 0 && res->end == 0)
 					continue;
 
@@ -283,9 +285,10 @@ static void quirk_system_pci_resources(struct pnp_dev *dev)
 				 * the PCI region, and that might prevent a PCI
 				 * driver from requesting its resources.
 				 */
-				dev_warn(&dev->dev, "mem resource "
+				dev_warn(&dev->dev, "%s resource "
 					"(0x%llx-0x%llx) overlaps %s BAR %d "
 					"(0x%llx-0x%llx), disabling\n",
+					pnp_resource_type_name(res),
 					(unsigned long long) pnp_start,
 					(unsigned long long) pnp_end,
 					pci_name(pdev), i,

From c485b465a031b6f9b9a51300e0ee1f86efc6db87 Mon Sep 17 00:00:00 2001
From: David Howells <dhowells@redhat.com>
Date: Fri, 25 Jul 2008 19:44:47 -0700
Subject: [PATCH 0991/1171] pnp: fix the fcpnp_driver declaration to only exist
 if CONFIG_PNP=y

Fix the fcpnp_driver declaration to only exist if CONFIG_PNP=y as it's
only accessed in that case.

The PNP=n variant was added by 30d55e71a81b1f5a8136f191dc9f4c21f18e77e6
("hisax: depend on CONFIG_PNP, not __ISAPNP__")

Fixes an unused variable warning.

Signed-off-by: David Howells <dhowells@redhat.com>
Acked-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Cc: Karsten Keil <kkeil@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/isdn/hisax/hisax_fcpcipnp.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/isdn/hisax/hisax_fcpcipnp.c b/drivers/isdn/hisax/hisax_fcpcipnp.c
index c0b4db2f8364..1925118122f8 100644
--- a/drivers/isdn/hisax/hisax_fcpcipnp.c
+++ b/drivers/isdn/hisax/hisax_fcpcipnp.c
@@ -974,8 +974,6 @@ static struct pnp_driver fcpnp_driver = {
 	.remove		= __devexit_p(fcpnp_remove),
 	.id_table	= fcpnp_ids,
 };
-#else
-static struct pnp_driver fcpnp_driver;
 #endif
 
 static void __devexit fcpci_remove(struct pci_dev *pdev)

From 8d8bb39b9eba32dd70e87fd5ad5c5dd4ba118e06 Mon Sep 17 00:00:00 2001
From: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Date: Fri, 25 Jul 2008 19:44:49 -0700
Subject: [PATCH 0992/1171] dma-mapping: add the device argument to
 dma_mapping_error()

Add per-device dma_mapping_ops support for CONFIG_X86_64 as POWER
architecture does:

This enables us to cleanly fix the Calgary IOMMU issue that some devices
are not behind the IOMMU (http://lkml.org/lkml/2008/5/8/423).

I think that per-device dma_mapping_ops support would be also helpful for
KVM people to support PCI passthrough but Andi thinks that this makes it
difficult to support the PCI passthrough (see the above thread).  So I
CC'ed this to KVM camp.  Comments are appreciated.

A pointer to dma_mapping_ops to struct dev_archdata is added.  If the
pointer is non NULL, DMA operations in asm/dma-mapping.h use it.  If it's
NULL, the system-wide dma_ops pointer is used as before.

If it's useful for KVM people, I plan to implement a mechanism to register
a hook called when a new pci (or dma capable) device is created (it works
with hot plugging).  It enables IOMMUs to set up an appropriate
dma_mapping_ops per device.

The major obstacle is that dma_mapping_error doesn't take a pointer to the
device unlike other DMA operations.  So x86 can't have dma_mapping_ops per
device.  Note all the POWER IOMMUs use the same dma_mapping_error function
so this is not a problem for POWER but x86 IOMMUs use different
dma_mapping_error functions.

The first patch adds the device argument to dma_mapping_error.  The patch
is trivial but large since it touches lots of drivers and dma-mapping.h in
all the architecture.

This patch:

dma_mapping_error() doesn't take a pointer to the device unlike other DMA
operations.  So we can't have dma_mapping_ops per device.

Note that POWER already has dma_mapping_ops per device but all the POWER
IOMMUs use the same dma_mapping_error function.  x86 IOMMUs use device
argument.

[akpm@linux-foundation.org: fix sge]
[akpm@linux-foundation.org: fix svc_rdma]
[akpm@linux-foundation.org: build fix]
[akpm@linux-foundation.org: fix bnx2x]
[akpm@linux-foundation.org: fix s2io]
[akpm@linux-foundation.org: fix pasemi_mac]
[akpm@linux-foundation.org: fix sdhci]
[akpm@linux-foundation.org: build fix]
[akpm@linux-foundation.org: fix sparc]
[akpm@linux-foundation.org: fix ibmvscsi]
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Muli Ben-Yehuda <muli@il.ibm.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Avi Kivity <avi@qumranet.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/DMA-API.txt                     |  4 +-
 arch/arm/common/dmabounce.c                   |  2 +-
 arch/ia64/hp/common/hwsw_iommu.c              |  5 +-
 arch/ia64/hp/common/sba_iommu.c               |  2 +-
 arch/ia64/sn/pci/pci_dma.c                    |  2 +-
 arch/mips/mm/dma-default.c                    |  2 +-
 .../powerpc/platforms/cell/celleb_scc_pciex.c |  2 +-
 arch/powerpc/platforms/cell/spider-pci.c      |  2 +-
 arch/powerpc/platforms/iseries/mf.c           |  2 +-
 arch/x86/kernel/pci-calgary_64.c              |  2 +-
 arch/x86/kernel/pci-dma.c                     | 27 ++---
 arch/x86/kernel/pci-gart_64.c                 |  3 +-
 arch/x86/kernel/pci-nommu.c                   | 14 +--
 arch/x86/kernel/pci-swiotlb_64.c              |  2 +-
 drivers/firewire/fw-iso.c                     |  2 +-
 drivers/firewire/fw-ohci.c                    |  2 +-
 drivers/firewire/fw-sbp2.c                    |  8 +-
 drivers/infiniband/hw/ipath/ipath_sdma.c      |  2 +-
 drivers/infiniband/hw/ipath/ipath_user_sdma.c |  6 +-
 drivers/infiniband/hw/mthca/mthca_eq.c        |  2 +-
 drivers/media/dvb/pluto2/pluto2.c             |  2 +-
 drivers/mmc/host/sdhci.c                      |  4 +-
 drivers/net/arm/ep93xx_eth.c                  |  4 +-
 drivers/net/bnx2x_main.c                      |  4 +-
 drivers/net/cxgb3/sge.c                       |  2 +-
 drivers/net/e100.c                            |  2 +-
 drivers/net/e1000e/ethtool.c                  |  4 +-
 drivers/net/e1000e/netdev.c                   | 11 ++-
 drivers/net/ibmveth.c                         | 38 +++----
 drivers/net/iseries_veth.c                    |  4 +-
 drivers/net/mlx4/eq.c                         |  2 +-
 drivers/net/pasemi_mac.c                      |  6 +-
 drivers/net/qla3xxx.c                         | 12 +--
 drivers/net/s2io.c                            | 48 +++++----
 drivers/net/sfc/rx.c                          |  4 +-
 drivers/net/sfc/tx.c                          |  7 +-
 drivers/net/spider_net.c                      |  4 +-
 drivers/net/tc35815.c                         |  4 +-
 drivers/net/wireless/ath5k/base.c             |  4 +-
 drivers/scsi/ibmvscsi/ibmvfc.c                |  4 +-
 drivers/scsi/ibmvscsi/ibmvscsi.c              |  4 +-
 drivers/scsi/ibmvscsi/ibmvstgt.c              |  2 +-
 drivers/scsi/ibmvscsi/rpa_vscsi.c             |  2 +-
 drivers/spi/atmel_spi.c                       |  4 +-
 drivers/spi/au1550_spi.c                      |  6 +-
 drivers/spi/omap2_mcspi.c                     |  4 +-
 drivers/spi/pxa2xx_spi.c                      |  4 +-
 drivers/spi/spi_imx.c                         |  6 +-
 include/asm-alpha/dma-mapping.h               |  6 +-
 include/asm-alpha/pci.h                       |  2 +-
 include/asm-arm/dma-mapping.h                 |  2 +-
 include/asm-avr32/dma-mapping.h               |  2 +-
 include/asm-cris/dma-mapping.h                |  2 +-
 include/asm-frv/dma-mapping.h                 |  2 +-
 include/asm-generic/dma-mapping-broken.h      |  2 +-
 include/asm-generic/dma-mapping.h             |  4 +-
 include/asm-generic/pci-dma-compat.h          |  4 +-
 include/asm-ia64/machvec.h                    |  2 +-
 include/asm-m68k/dma-mapping.h                |  2 +-
 include/asm-mips/dma-mapping.h                |  2 +-
 include/asm-mn10300/dma-mapping.h             |  2 +-
 include/asm-parisc/dma-mapping.h              |  2 +-
 include/asm-powerpc/dma-mapping.h             |  2 +-
 include/asm-sh/dma-mapping.h                  |  2 +-
 include/asm-sparc/dma-mapping_64.h            |  2 +-
 include/asm-sparc/pci_32.h                    |  3 +-
 include/asm-sparc/pci_64.h                    |  5 +-
 include/asm-x86/device.h                      |  3 +
 include/asm-x86/dma-mapping.h                 | 99 +++++++++++++------
 include/asm-x86/swiotlb.h                     |  2 +-
 include/asm-xtensa/dma-mapping.h              |  2 +-
 include/linux/i2o.h                           |  2 +-
 include/linux/ssb/ssb.h                       |  4 +-
 include/rdma/ib_verbs.h                       |  2 +-
 lib/swiotlb.c                                 |  4 +-
 net/sunrpc/xprtrdma/svc_rdma_sendto.c         |  3 +-
 76 files changed, 256 insertions(+), 210 deletions(-)

diff --git a/Documentation/DMA-API.txt b/Documentation/DMA-API.txt
index 80d150458c80..d8b63d164e41 100644
--- a/Documentation/DMA-API.txt
+++ b/Documentation/DMA-API.txt
@@ -298,10 +298,10 @@ recommended that you never use these unless you really know what the
 cache width is.
 
 int
-dma_mapping_error(dma_addr_t dma_addr)
+dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
 
 int
-pci_dma_mapping_error(dma_addr_t dma_addr)
+pci_dma_mapping_error(struct pci_dev *hwdev, dma_addr_t dma_addr)
 
 In some circumstances dma_map_single and dma_map_page will fail to create
 a mapping. A driver can check for these errors by testing the returned
diff --git a/arch/arm/common/dmabounce.c b/arch/arm/common/dmabounce.c
index dd2947342604..69130f365904 100644
--- a/arch/arm/common/dmabounce.c
+++ b/arch/arm/common/dmabounce.c
@@ -280,7 +280,7 @@ unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
 	/*
 	 * Trying to unmap an invalid mapping
 	 */
-	if (dma_mapping_error(dma_addr)) {
+	if (dma_mapping_error(dev, dma_addr)) {
 		dev_err(dev, "Trying to unmap invalid mapping\n");
 		return;
 	}
diff --git a/arch/ia64/hp/common/hwsw_iommu.c b/arch/ia64/hp/common/hwsw_iommu.c
index 1c44ec2a1d58..88b6e6f3fd88 100644
--- a/arch/ia64/hp/common/hwsw_iommu.c
+++ b/arch/ia64/hp/common/hwsw_iommu.c
@@ -186,9 +186,10 @@ hwsw_dma_supported (struct device *dev, u64 mask)
 }
 
 int
-hwsw_dma_mapping_error (dma_addr_t dma_addr)
+hwsw_dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
 {
-	return hwiommu_dma_mapping_error (dma_addr) || swiotlb_dma_mapping_error(dma_addr);
+	return hwiommu_dma_mapping_error(dev, dma_addr) ||
+		swiotlb_dma_mapping_error(dev, dma_addr);
 }
 
 EXPORT_SYMBOL(hwsw_dma_mapping_error);
diff --git a/arch/ia64/hp/common/sba_iommu.c b/arch/ia64/hp/common/sba_iommu.c
index 34421aed1e2a..4956be40d7b5 100644
--- a/arch/ia64/hp/common/sba_iommu.c
+++ b/arch/ia64/hp/common/sba_iommu.c
@@ -2147,7 +2147,7 @@ sba_dma_supported (struct device *dev, u64 mask)
 }
 
 int
-sba_dma_mapping_error (dma_addr_t dma_addr)
+sba_dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
 {
 	return 0;
 }
diff --git a/arch/ia64/sn/pci/pci_dma.c b/arch/ia64/sn/pci/pci_dma.c
index 52175af299a0..53ebb6484495 100644
--- a/arch/ia64/sn/pci/pci_dma.c
+++ b/arch/ia64/sn/pci/pci_dma.c
@@ -350,7 +350,7 @@ void sn_dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg,
 }
 EXPORT_SYMBOL(sn_dma_sync_sg_for_device);
 
-int sn_dma_mapping_error(dma_addr_t dma_addr)
+int sn_dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
 {
 	return 0;
 }
diff --git a/arch/mips/mm/dma-default.c b/arch/mips/mm/dma-default.c
index ae39dd88b9aa..891312f8e5a6 100644
--- a/arch/mips/mm/dma-default.c
+++ b/arch/mips/mm/dma-default.c
@@ -348,7 +348,7 @@ void dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nele
 
 EXPORT_SYMBOL(dma_sync_sg_for_device);
 
-int dma_mapping_error(dma_addr_t dma_addr)
+int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
 {
 	return 0;
 }
diff --git a/arch/powerpc/platforms/cell/celleb_scc_pciex.c b/arch/powerpc/platforms/cell/celleb_scc_pciex.c
index 0e04f8fb152a..3e7e0f1568ef 100644
--- a/arch/powerpc/platforms/cell/celleb_scc_pciex.c
+++ b/arch/powerpc/platforms/cell/celleb_scc_pciex.c
@@ -281,7 +281,7 @@ static int __init scc_pciex_iowa_init(struct iowa_bus *bus, void *data)
 
 	dummy_page_da = dma_map_single(bus->phb->parent, dummy_page_va,
 				       PAGE_SIZE, DMA_FROM_DEVICE);
-	if (dma_mapping_error(dummy_page_da)) {
+	if (dma_mapping_error(bus->phb->parent, dummy_page_da)) {
 		pr_err("PCIEX:Map dummy page failed.\n");
 		kfree(dummy_page_va);
 		return -1;
diff --git a/arch/powerpc/platforms/cell/spider-pci.c b/arch/powerpc/platforms/cell/spider-pci.c
index 418b605ac35a..5122ec145271 100644
--- a/arch/powerpc/platforms/cell/spider-pci.c
+++ b/arch/powerpc/platforms/cell/spider-pci.c
@@ -111,7 +111,7 @@ static int __init spiderpci_pci_setup_chip(struct pci_controller *phb,
 
 	dummy_page_da = dma_map_single(phb->parent, dummy_page_va,
 				       PAGE_SIZE, DMA_FROM_DEVICE);
-	if (dma_mapping_error(dummy_page_da)) {
+	if (dma_mapping_error(phb->parent, dummy_page_da)) {
 		pr_err("SPIDER-IOWA:Map dummy page filed.\n");
 		kfree(dummy_page_va);
 		return -1;
diff --git a/arch/powerpc/platforms/iseries/mf.c b/arch/powerpc/platforms/iseries/mf.c
index 1dc7295746da..731d7b157749 100644
--- a/arch/powerpc/platforms/iseries/mf.c
+++ b/arch/powerpc/platforms/iseries/mf.c
@@ -871,7 +871,7 @@ static int proc_mf_dump_cmdline(char *page, char **start, off_t off,
 		count = 256 - off;
 
 	dma_addr = iseries_hv_map(page, off + count, DMA_FROM_DEVICE);
-	if (dma_mapping_error(dma_addr))
+	if (dma_mapping_error(NULL, dma_addr))
 		return -ENOMEM;
 	memset(page, 0, off + count);
 	memset(&vsp_cmd, 0, sizeof(vsp_cmd));
diff --git a/arch/x86/kernel/pci-calgary_64.c b/arch/x86/kernel/pci-calgary_64.c
index 19e7fc7c2c4f..1eb86be93d7a 100644
--- a/arch/x86/kernel/pci-calgary_64.c
+++ b/arch/x86/kernel/pci-calgary_64.c
@@ -544,7 +544,7 @@ error:
 	return ret;
 }
 
-static const struct dma_mapping_ops calgary_dma_ops = {
+static struct dma_mapping_ops calgary_dma_ops = {
 	.alloc_coherent = calgary_alloc_coherent,
 	.map_single = calgary_map_single,
 	.unmap_single = calgary_unmap_single,
diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c
index cbecb05551bb..37544123896d 100644
--- a/arch/x86/kernel/pci-dma.c
+++ b/arch/x86/kernel/pci-dma.c
@@ -11,7 +11,7 @@
 
 static int forbid_dac __read_mostly;
 
-const struct dma_mapping_ops *dma_ops;
+struct dma_mapping_ops *dma_ops;
 EXPORT_SYMBOL(dma_ops);
 
 static int iommu_sac_force __read_mostly;
@@ -312,6 +312,8 @@ static int dma_release_coherent(struct device *dev, int order, void *vaddr)
 
 int dma_supported(struct device *dev, u64 mask)
 {
+	struct dma_mapping_ops *ops = get_dma_ops(dev);
+
 #ifdef CONFIG_PCI
 	if (mask > 0xffffffff && forbid_dac > 0) {
 		dev_info(dev, "PCI: Disallowing DAC for device\n");
@@ -319,8 +321,8 @@ int dma_supported(struct device *dev, u64 mask)
 	}
 #endif
 
-	if (dma_ops->dma_supported)
-		return dma_ops->dma_supported(dev, mask);
+	if (ops->dma_supported)
+		return ops->dma_supported(dev, mask);
 
 	/* Copied from i386. Doesn't make much sense, because it will
 	   only work for pci_alloc_coherent.
@@ -367,6 +369,7 @@ void *
 dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle,
 		   gfp_t gfp)
 {
+	struct dma_mapping_ops *ops = get_dma_ops(dev);
 	void *memory = NULL;
 	struct page *page;
 	unsigned long dma_mask = 0;
@@ -435,8 +438,8 @@ dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle,
 			/* Let low level make its own zone decisions */
 			gfp &= ~(GFP_DMA32|GFP_DMA);
 
-			if (dma_ops->alloc_coherent)
-				return dma_ops->alloc_coherent(dev, size,
+			if (ops->alloc_coherent)
+				return ops->alloc_coherent(dev, size,
 							   dma_handle, gfp);
 			return NULL;
 		}
@@ -448,14 +451,14 @@ dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle,
 		}
 	}
 
-	if (dma_ops->alloc_coherent) {
+	if (ops->alloc_coherent) {
 		free_pages((unsigned long)memory, get_order(size));
 		gfp &= ~(GFP_DMA|GFP_DMA32);
-		return dma_ops->alloc_coherent(dev, size, dma_handle, gfp);
+		return ops->alloc_coherent(dev, size, dma_handle, gfp);
 	}
 
-	if (dma_ops->map_simple) {
-		*dma_handle = dma_ops->map_simple(dev, virt_to_phys(memory),
+	if (ops->map_simple) {
+		*dma_handle = ops->map_simple(dev, virt_to_phys(memory),
 					      size,
 					      PCI_DMA_BIDIRECTIONAL);
 		if (*dma_handle != bad_dma_address)
@@ -477,12 +480,14 @@ EXPORT_SYMBOL(dma_alloc_coherent);
 void dma_free_coherent(struct device *dev, size_t size,
 			 void *vaddr, dma_addr_t bus)
 {
+	struct dma_mapping_ops *ops = get_dma_ops(dev);
+
 	int order = get_order(size);
 	WARN_ON(irqs_disabled());	/* for portability */
 	if (dma_release_coherent(dev, order, vaddr))
 		return;
-	if (dma_ops->unmap_single)
-		dma_ops->unmap_single(dev, bus, size, 0);
+	if (ops->unmap_single)
+		ops->unmap_single(dev, bus, size, 0);
 	free_pages((unsigned long)vaddr, order);
 }
 EXPORT_SYMBOL(dma_free_coherent);
diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c
index df5f142657d2..744126e64950 100644
--- a/arch/x86/kernel/pci-gart_64.c
+++ b/arch/x86/kernel/pci-gart_64.c
@@ -692,8 +692,7 @@ static __init int init_k8_gatt(struct agp_kern_info *info)
 
 extern int agp_amd64_init(void);
 
-static const struct dma_mapping_ops gart_dma_ops = {
-	.mapping_error			= NULL,
+static struct dma_mapping_ops gart_dma_ops = {
 	.map_single			= gart_map_single,
 	.map_simple			= gart_map_simple,
 	.unmap_single			= gart_unmap_single,
diff --git a/arch/x86/kernel/pci-nommu.c b/arch/x86/kernel/pci-nommu.c
index 792b9179eff3..3f91f71cdc3e 100644
--- a/arch/x86/kernel/pci-nommu.c
+++ b/arch/x86/kernel/pci-nommu.c
@@ -72,21 +72,9 @@ static int nommu_map_sg(struct device *hwdev, struct scatterlist *sg,
 	return nents;
 }
 
-/* Make sure we keep the same behaviour */
-static int nommu_mapping_error(dma_addr_t dma_addr)
-{
-#ifdef CONFIG_X86_32
-	return 0;
-#else
-	return (dma_addr == bad_dma_address);
-#endif
-}
-
-
-const struct dma_mapping_ops nommu_dma_ops = {
+struct dma_mapping_ops nommu_dma_ops = {
 	.map_single = nommu_map_single,
 	.map_sg = nommu_map_sg,
-	.mapping_error = nommu_mapping_error,
 	.is_phys = 1,
 };
 
diff --git a/arch/x86/kernel/pci-swiotlb_64.c b/arch/x86/kernel/pci-swiotlb_64.c
index 20df839b9c20..c4ce0332759e 100644
--- a/arch/x86/kernel/pci-swiotlb_64.c
+++ b/arch/x86/kernel/pci-swiotlb_64.c
@@ -18,7 +18,7 @@ swiotlb_map_single_phys(struct device *hwdev, phys_addr_t paddr, size_t size,
 	return swiotlb_map_single(hwdev, phys_to_virt(paddr), size, direction);
 }
 
-const struct dma_mapping_ops swiotlb_dma_ops = {
+struct dma_mapping_ops swiotlb_dma_ops = {
 	.mapping_error = swiotlb_dma_mapping_error,
 	.alloc_coherent = swiotlb_alloc_coherent,
 	.free_coherent = swiotlb_free_coherent,
diff --git a/drivers/firewire/fw-iso.c b/drivers/firewire/fw-iso.c
index bcbe794a3ea5..e14c03dc0065 100644
--- a/drivers/firewire/fw-iso.c
+++ b/drivers/firewire/fw-iso.c
@@ -50,7 +50,7 @@ fw_iso_buffer_init(struct fw_iso_buffer *buffer, struct fw_card *card,
 
 		address = dma_map_page(card->device, buffer->pages[i],
 				       0, PAGE_SIZE, direction);
-		if (dma_mapping_error(address)) {
+		if (dma_mapping_error(card->device, address)) {
 			__free_page(buffer->pages[i]);
 			goto out_pages;
 		}
diff --git a/drivers/firewire/fw-ohci.c b/drivers/firewire/fw-ohci.c
index 333b12544dd1..566672e0bcff 100644
--- a/drivers/firewire/fw-ohci.c
+++ b/drivers/firewire/fw-ohci.c
@@ -953,7 +953,7 @@ at_context_queue_packet(struct context *ctx, struct fw_packet *packet)
 		payload_bus =
 			dma_map_single(ohci->card.device, packet->payload,
 				       packet->payload_length, DMA_TO_DEVICE);
-		if (dma_mapping_error(payload_bus)) {
+		if (dma_mapping_error(ohci->card.device, payload_bus)) {
 			packet->ack = RCODE_SEND_ERROR;
 			return -1;
 		}
diff --git a/drivers/firewire/fw-sbp2.c b/drivers/firewire/fw-sbp2.c
index 53fc5a641e6d..aaff50ebba1d 100644
--- a/drivers/firewire/fw-sbp2.c
+++ b/drivers/firewire/fw-sbp2.c
@@ -543,7 +543,7 @@ sbp2_send_management_orb(struct sbp2_logical_unit *lu, int node_id,
 	orb->response_bus =
 		dma_map_single(device->card->device, &orb->response,
 			       sizeof(orb->response), DMA_FROM_DEVICE);
-	if (dma_mapping_error(orb->response_bus))
+	if (dma_mapping_error(device->card->device, orb->response_bus))
 		goto fail_mapping_response;
 
 	orb->request.response.high = 0;
@@ -577,7 +577,7 @@ sbp2_send_management_orb(struct sbp2_logical_unit *lu, int node_id,
 	orb->base.request_bus =
 		dma_map_single(device->card->device, &orb->request,
 			       sizeof(orb->request), DMA_TO_DEVICE);
-	if (dma_mapping_error(orb->base.request_bus))
+	if (dma_mapping_error(device->card->device, orb->base.request_bus))
 		goto fail_mapping_request;
 
 	sbp2_send_orb(&orb->base, lu, node_id, generation,
@@ -1424,7 +1424,7 @@ sbp2_map_scatterlist(struct sbp2_command_orb *orb, struct fw_device *device,
 	orb->page_table_bus =
 		dma_map_single(device->card->device, orb->page_table,
 			       sizeof(orb->page_table), DMA_TO_DEVICE);
-	if (dma_mapping_error(orb->page_table_bus))
+	if (dma_mapping_error(device->card->device, orb->page_table_bus))
 		goto fail_page_table;
 
 	/*
@@ -1509,7 +1509,7 @@ static int sbp2_scsi_queuecommand(struct scsi_cmnd *cmd, scsi_done_fn_t done)
 	orb->base.request_bus =
 		dma_map_single(device->card->device, &orb->request,
 			       sizeof(orb->request), DMA_TO_DEVICE);
-	if (dma_mapping_error(orb->base.request_bus))
+	if (dma_mapping_error(device->card->device, orb->base.request_bus))
 		goto out;
 
 	sbp2_send_orb(&orb->base, lu, lu->tgt->node_id, lu->generation,
diff --git a/drivers/infiniband/hw/ipath/ipath_sdma.c b/drivers/infiniband/hw/ipath/ipath_sdma.c
index eaba03273e4f..284c9bca517e 100644
--- a/drivers/infiniband/hw/ipath/ipath_sdma.c
+++ b/drivers/infiniband/hw/ipath/ipath_sdma.c
@@ -698,7 +698,7 @@ retry:
 
 	addr = dma_map_single(&dd->pcidev->dev, tx->txreq.map_addr,
 			      tx->map_len, DMA_TO_DEVICE);
-	if (dma_mapping_error(addr)) {
+	if (dma_mapping_error(&dd->pcidev->dev, addr)) {
 		ret = -EIO;
 		goto unlock;
 	}
diff --git a/drivers/infiniband/hw/ipath/ipath_user_sdma.c b/drivers/infiniband/hw/ipath/ipath_user_sdma.c
index 86e016916cd1..82d9a0b5ca2f 100644
--- a/drivers/infiniband/hw/ipath/ipath_user_sdma.c
+++ b/drivers/infiniband/hw/ipath/ipath_user_sdma.c
@@ -206,7 +206,7 @@ static int ipath_user_sdma_coalesce(const struct ipath_devdata *dd,
 
 	dma_addr = dma_map_page(&dd->pcidev->dev, page, 0, len,
 				DMA_TO_DEVICE);
-	if (dma_mapping_error(dma_addr)) {
+	if (dma_mapping_error(&dd->pcidev->dev, dma_addr)) {
 		ret = -ENOMEM;
 		goto free_unmap;
 	}
@@ -301,7 +301,7 @@ static int ipath_user_sdma_pin_pages(const struct ipath_devdata *dd,
 				     pages[j], 0, flen, DMA_TO_DEVICE);
 		unsigned long fofs = addr & ~PAGE_MASK;
 
-		if (dma_mapping_error(dma_addr)) {
+		if (dma_mapping_error(&dd->pcidev->dev, dma_addr)) {
 			ret = -ENOMEM;
 			goto done;
 		}
@@ -508,7 +508,7 @@ static int ipath_user_sdma_queue_pkts(const struct ipath_devdata *dd,
 		if (page) {
 			dma_addr = dma_map_page(&dd->pcidev->dev,
 						page, 0, len, DMA_TO_DEVICE);
-			if (dma_mapping_error(dma_addr)) {
+			if (dma_mapping_error(&dd->pcidev->dev, dma_addr)) {
 				ret = -ENOMEM;
 				goto free_pbc;
 			}
diff --git a/drivers/infiniband/hw/mthca/mthca_eq.c b/drivers/infiniband/hw/mthca/mthca_eq.c
index 4e36aa7cb3d2..cc6858f0b65b 100644
--- a/drivers/infiniband/hw/mthca/mthca_eq.c
+++ b/drivers/infiniband/hw/mthca/mthca_eq.c
@@ -780,7 +780,7 @@ int mthca_map_eq_icm(struct mthca_dev *dev, u64 icm_virt)
 		return -ENOMEM;
 	dev->eq_table.icm_dma  = pci_map_page(dev->pdev, dev->eq_table.icm_page, 0,
 					      PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
-	if (pci_dma_mapping_error(dev->eq_table.icm_dma)) {
+	if (pci_dma_mapping_error(dev->pdev, dev->eq_table.icm_dma)) {
 		__free_page(dev->eq_table.icm_page);
 		return -ENOMEM;
 	}
diff --git a/drivers/media/dvb/pluto2/pluto2.c b/drivers/media/dvb/pluto2/pluto2.c
index 1360403b88b6..a9653c63f4db 100644
--- a/drivers/media/dvb/pluto2/pluto2.c
+++ b/drivers/media/dvb/pluto2/pluto2.c
@@ -242,7 +242,7 @@ static int __devinit pluto_dma_map(struct pluto *pluto)
 	pluto->dma_addr = pci_map_single(pluto->pdev, pluto->dma_buf,
 			TS_DMA_BYTES, PCI_DMA_FROMDEVICE);
 
-	return pci_dma_mapping_error(pluto->dma_addr);
+	return pci_dma_mapping_error(pluto->pdev, pluto->dma_addr);
 }
 
 static void pluto_dma_unmap(struct pluto *pluto)
diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
index c3a5db72ddd7..5f95e10229b5 100644
--- a/drivers/mmc/host/sdhci.c
+++ b/drivers/mmc/host/sdhci.c
@@ -337,7 +337,7 @@ static int sdhci_adma_table_pre(struct sdhci_host *host,
 
 	host->align_addr = dma_map_single(mmc_dev(host->mmc),
 		host->align_buffer, 128 * 4, direction);
-	if (dma_mapping_error(host->align_addr))
+	if (dma_mapping_error(mmc_dev(host->mmc), host->align_addr))
 		goto fail;
 	BUG_ON(host->align_addr & 0x3);
 
@@ -439,7 +439,7 @@ static int sdhci_adma_table_pre(struct sdhci_host *host,
 
 	host->adma_addr = dma_map_single(mmc_dev(host->mmc),
 		host->adma_desc, (128 * 2 + 1) * 4, DMA_TO_DEVICE);
-	if (dma_mapping_error(host->align_addr))
+	if (dma_mapping_error(mmc_dev(host->mmc), host->align_addr))
 		goto unmap_entries;
 	BUG_ON(host->adma_addr & 0x3);
 
diff --git a/drivers/net/arm/ep93xx_eth.c b/drivers/net/arm/ep93xx_eth.c
index 7a14980f3472..18d3eeb7eab2 100644
--- a/drivers/net/arm/ep93xx_eth.c
+++ b/drivers/net/arm/ep93xx_eth.c
@@ -482,7 +482,7 @@ static int ep93xx_alloc_buffers(struct ep93xx_priv *ep)
 			goto err;
 
 		d = dma_map_single(NULL, page, PAGE_SIZE, DMA_FROM_DEVICE);
-		if (dma_mapping_error(d)) {
+		if (dma_mapping_error(NULL, d)) {
 			free_page((unsigned long)page);
 			goto err;
 		}
@@ -505,7 +505,7 @@ static int ep93xx_alloc_buffers(struct ep93xx_priv *ep)
 			goto err;
 
 		d = dma_map_single(NULL, page, PAGE_SIZE, DMA_TO_DEVICE);
-		if (dma_mapping_error(d)) {
+		if (dma_mapping_error(NULL, d)) {
 			free_page((unsigned long)page);
 			goto err;
 		}
diff --git a/drivers/net/bnx2x_main.c b/drivers/net/bnx2x_main.c
index 0263bef9cc6d..c7cc760a1777 100644
--- a/drivers/net/bnx2x_main.c
+++ b/drivers/net/bnx2x_main.c
@@ -1020,7 +1020,7 @@ static inline int bnx2x_alloc_rx_sge(struct bnx2x *bp,
 
 	mapping = pci_map_page(bp->pdev, page, 0, BCM_PAGE_SIZE*PAGES_PER_SGE,
 			       PCI_DMA_FROMDEVICE);
-	if (unlikely(dma_mapping_error(mapping))) {
+	if (unlikely(dma_mapping_error(&bp->pdev->dev, mapping))) {
 		__free_pages(page, PAGES_PER_SGE_SHIFT);
 		return -ENOMEM;
 	}
@@ -1048,7 +1048,7 @@ static inline int bnx2x_alloc_rx_skb(struct bnx2x *bp,
 
 	mapping = pci_map_single(bp->pdev, skb->data, bp->rx_buf_use_size,
 				 PCI_DMA_FROMDEVICE);
-	if (unlikely(dma_mapping_error(mapping))) {
+	if (unlikely(dma_mapping_error(&bp->pdev->dev, mapping))) {
 		dev_kfree_skb(skb);
 		return -ENOMEM;
 	}
diff --git a/drivers/net/cxgb3/sge.c b/drivers/net/cxgb3/sge.c
index a96331c875e6..1b0861d73ab7 100644
--- a/drivers/net/cxgb3/sge.c
+++ b/drivers/net/cxgb3/sge.c
@@ -386,7 +386,7 @@ static inline int add_one_rx_buf(void *va, unsigned int len,
 	dma_addr_t mapping;
 
 	mapping = pci_map_single(pdev, va, len, PCI_DMA_FROMDEVICE);
-	if (unlikely(pci_dma_mapping_error(mapping)))
+	if (unlikely(pci_dma_mapping_error(pdev, mapping)))
 		return -ENOMEM;
 
 	pci_unmap_addr_set(sd, dma_addr, mapping);
diff --git a/drivers/net/e100.c b/drivers/net/e100.c
index 1037b1332312..19d32a227be1 100644
--- a/drivers/net/e100.c
+++ b/drivers/net/e100.c
@@ -1790,7 +1790,7 @@ static int e100_rx_alloc_skb(struct nic *nic, struct rx *rx)
 	rx->dma_addr = pci_map_single(nic->pdev, rx->skb->data,
 		RFD_BUF_LEN, PCI_DMA_BIDIRECTIONAL);
 
-	if (pci_dma_mapping_error(rx->dma_addr)) {
+	if (pci_dma_mapping_error(nic->pdev, rx->dma_addr)) {
 		dev_kfree_skb_any(rx->skb);
 		rx->skb = NULL;
 		rx->dma_addr = 0;
diff --git a/drivers/net/e1000e/ethtool.c b/drivers/net/e1000e/ethtool.c
index a14561f40db0..9350564065e7 100644
--- a/drivers/net/e1000e/ethtool.c
+++ b/drivers/net/e1000e/ethtool.c
@@ -1090,7 +1090,7 @@ static int e1000_setup_desc_rings(struct e1000_adapter *adapter)
 		tx_ring->buffer_info[i].dma =
 			pci_map_single(pdev, skb->data, skb->len,
 				       PCI_DMA_TODEVICE);
-		if (pci_dma_mapping_error(tx_ring->buffer_info[i].dma)) {
+		if (pci_dma_mapping_error(pdev, tx_ring->buffer_info[i].dma)) {
 			ret_val = 4;
 			goto err_nomem;
 		}
@@ -1153,7 +1153,7 @@ static int e1000_setup_desc_rings(struct e1000_adapter *adapter)
 		rx_ring->buffer_info[i].dma =
 			pci_map_single(pdev, skb->data, 2048,
 				       PCI_DMA_FROMDEVICE);
-		if (pci_dma_mapping_error(rx_ring->buffer_info[i].dma)) {
+		if (pci_dma_mapping_error(pdev, rx_ring->buffer_info[i].dma)) {
 			ret_val = 8;
 			goto err_nomem;
 		}
diff --git a/drivers/net/e1000e/netdev.c b/drivers/net/e1000e/netdev.c
index 9c0f56b3c518..d13677899767 100644
--- a/drivers/net/e1000e/netdev.c
+++ b/drivers/net/e1000e/netdev.c
@@ -195,7 +195,7 @@ map_skb:
 		buffer_info->dma = pci_map_single(pdev, skb->data,
 						  adapter->rx_buffer_len,
 						  PCI_DMA_FROMDEVICE);
-		if (pci_dma_mapping_error(buffer_info->dma)) {
+		if (pci_dma_mapping_error(pdev, buffer_info->dma)) {
 			dev_err(&pdev->dev, "RX DMA map failed\n");
 			adapter->rx_dma_failed++;
 			break;
@@ -265,7 +265,7 @@ static void e1000_alloc_rx_buffers_ps(struct e1000_adapter *adapter,
 						   ps_page->page,
 						   0, PAGE_SIZE,
 						   PCI_DMA_FROMDEVICE);
-				if (pci_dma_mapping_error(ps_page->dma)) {
+				if (pci_dma_mapping_error(pdev, ps_page->dma)) {
 					dev_err(&adapter->pdev->dev,
 					  "RX DMA page map failed\n");
 					adapter->rx_dma_failed++;
@@ -300,7 +300,7 @@ static void e1000_alloc_rx_buffers_ps(struct e1000_adapter *adapter,
 		buffer_info->dma = pci_map_single(pdev, skb->data,
 						  adapter->rx_ps_bsize0,
 						  PCI_DMA_FROMDEVICE);
-		if (pci_dma_mapping_error(buffer_info->dma)) {
+		if (pci_dma_mapping_error(pdev, buffer_info->dma)) {
 			dev_err(&pdev->dev, "RX DMA map failed\n");
 			adapter->rx_dma_failed++;
 			/* cleanup skb */
@@ -3344,7 +3344,7 @@ static int e1000_tx_map(struct e1000_adapter *adapter,
 				skb->data + offset,
 				size,
 				PCI_DMA_TODEVICE);
-		if (pci_dma_mapping_error(buffer_info->dma)) {
+		if (pci_dma_mapping_error(adapter->pdev, buffer_info->dma)) {
 			dev_err(&adapter->pdev->dev, "TX DMA map failed\n");
 			adapter->tx_dma_failed++;
 			return -1;
@@ -3382,7 +3382,8 @@ static int e1000_tx_map(struct e1000_adapter *adapter,
 					offset,
 					size,
 					PCI_DMA_TODEVICE);
-			if (pci_dma_mapping_error(buffer_info->dma)) {
+			if (pci_dma_mapping_error(adapter->pdev,
+						  buffer_info->dma)) {
 				dev_err(&adapter->pdev->dev,
 					"TX DMA page map failed\n");
 				adapter->tx_dma_failed++;
diff --git a/drivers/net/ibmveth.c b/drivers/net/ibmveth.c
index e5a6e2e84540..91ec9fdc7184 100644
--- a/drivers/net/ibmveth.c
+++ b/drivers/net/ibmveth.c
@@ -260,7 +260,7 @@ static void ibmveth_replenish_buffer_pool(struct ibmveth_adapter *adapter, struc
 		dma_addr = dma_map_single(&adapter->vdev->dev, skb->data,
 				pool->buff_size, DMA_FROM_DEVICE);
 
-		if (dma_mapping_error(dma_addr))
+		if (dma_mapping_error((&adapter->vdev->dev, dma_addr))
 			goto failure;
 
 		pool->free_map[free_index] = IBM_VETH_INVALID_MAP;
@@ -294,7 +294,7 @@ failure:
 		pool->consumer_index = pool->size - 1;
 	else
 		pool->consumer_index--;
-	if (!dma_mapping_error(dma_addr))
+	if (!dma_mapping_error((&adapter->vdev->dev, dma_addr))
 		dma_unmap_single(&adapter->vdev->dev,
 		                 pool->dma_addr[index], pool->buff_size,
 		                 DMA_FROM_DEVICE);
@@ -448,11 +448,11 @@ static void ibmveth_rxq_harvest_buffer(struct ibmveth_adapter *adapter)
 static void ibmveth_cleanup(struct ibmveth_adapter *adapter)
 {
 	int i;
+	struct device *dev = &adapter->vdev->dev;
 
 	if(adapter->buffer_list_addr != NULL) {
-		if(!dma_mapping_error(adapter->buffer_list_dma)) {
-			dma_unmap_single(&adapter->vdev->dev,
-					adapter->buffer_list_dma, 4096,
+		if (!dma_mapping_error(dev, adapter->buffer_list_dma)) {
+			dma_unmap_single(dev, adapter->buffer_list_dma, 4096,
 					DMA_BIDIRECTIONAL);
 			adapter->buffer_list_dma = DMA_ERROR_CODE;
 		}
@@ -461,9 +461,8 @@ static void ibmveth_cleanup(struct ibmveth_adapter *adapter)
 	}
 
 	if(adapter->filter_list_addr != NULL) {
-		if(!dma_mapping_error(adapter->filter_list_dma)) {
-			dma_unmap_single(&adapter->vdev->dev,
-					adapter->filter_list_dma, 4096,
+		if (!dma_mapping_error(dev, adapter->filter_list_dma)) {
+			dma_unmap_single(dev, adapter->filter_list_dma, 4096,
 					DMA_BIDIRECTIONAL);
 			adapter->filter_list_dma = DMA_ERROR_CODE;
 		}
@@ -472,8 +471,8 @@ static void ibmveth_cleanup(struct ibmveth_adapter *adapter)
 	}
 
 	if(adapter->rx_queue.queue_addr != NULL) {
-		if(!dma_mapping_error(adapter->rx_queue.queue_dma)) {
-			dma_unmap_single(&adapter->vdev->dev,
+		if (!dma_mapping_error(dev, adapter->rx_queue.queue_dma)) {
+			dma_unmap_single(dev,
 					adapter->rx_queue.queue_dma,
 					adapter->rx_queue.queue_len,
 					DMA_BIDIRECTIONAL);
@@ -535,6 +534,7 @@ static int ibmveth_open(struct net_device *netdev)
 	int rc;
 	union ibmveth_buf_desc rxq_desc;
 	int i;
+	struct device *dev;
 
 	ibmveth_debug_printk("open starting\n");
 
@@ -563,17 +563,19 @@ static int ibmveth_open(struct net_device *netdev)
 		return -ENOMEM;
 	}
 
-	adapter->buffer_list_dma = dma_map_single(&adapter->vdev->dev,
+	dev = &adapter->vdev->dev;
+
+	adapter->buffer_list_dma = dma_map_single(dev,
 			adapter->buffer_list_addr, 4096, DMA_BIDIRECTIONAL);
-	adapter->filter_list_dma = dma_map_single(&adapter->vdev->dev,
+	adapter->filter_list_dma = dma_map_single(dev,
 			adapter->filter_list_addr, 4096, DMA_BIDIRECTIONAL);
-	adapter->rx_queue.queue_dma = dma_map_single(&adapter->vdev->dev,
+	adapter->rx_queue.queue_dma = dma_map_single(dev,
 			adapter->rx_queue.queue_addr,
 			adapter->rx_queue.queue_len, DMA_BIDIRECTIONAL);
 
-	if((dma_mapping_error(adapter->buffer_list_dma) ) ||
-	   (dma_mapping_error(adapter->filter_list_dma)) ||
-	   (dma_mapping_error(adapter->rx_queue.queue_dma))) {
+	if ((dma_mapping_error(dev, adapter->buffer_list_dma)) ||
+	    (dma_mapping_error(dev, adapter->filter_list_dma)) ||
+	    (dma_mapping_error(dev, adapter->rx_queue.queue_dma))) {
 		ibmveth_error_printk("unable to map filter or buffer list pages\n");
 		ibmveth_cleanup(adapter);
 		napi_disable(&adapter->napi);
@@ -645,7 +647,7 @@ static int ibmveth_open(struct net_device *netdev)
 	adapter->bounce_buffer_dma =
 	    dma_map_single(&adapter->vdev->dev, adapter->bounce_buffer,
 			   netdev->mtu + IBMVETH_BUFF_OH, DMA_BIDIRECTIONAL);
-	if (dma_mapping_error(adapter->bounce_buffer_dma)) {
+	if (dma_mapping_error(dev, adapter->bounce_buffer_dma)) {
 		ibmveth_error_printk("unable to map bounce buffer\n");
 		ibmveth_cleanup(adapter);
 		napi_disable(&adapter->napi);
@@ -922,7 +924,7 @@ static int ibmveth_start_xmit(struct sk_buff *skb, struct net_device *netdev)
 		buf[1] = 0;
 	}
 
-	if (dma_mapping_error(data_dma_addr)) {
+	if (dma_mapping_error((&adapter->vdev->dev, data_dma_addr)) {
 		if (!firmware_has_feature(FW_FEATURE_CMO))
 			ibmveth_error_printk("tx: unable to map xmit buffer\n");
 		skb_copy_from_linear_data(skb, adapter->bounce_buffer,
diff --git a/drivers/net/iseries_veth.c b/drivers/net/iseries_veth.c
index b8d0639c1cdf..c46864d626b2 100644
--- a/drivers/net/iseries_veth.c
+++ b/drivers/net/iseries_veth.c
@@ -1128,7 +1128,7 @@ static int veth_transmit_to_one(struct sk_buff *skb, HvLpIndex rlp,
 	msg->data.addr[0] = dma_map_single(port->dev, skb->data,
 				skb->len, DMA_TO_DEVICE);
 
-	if (dma_mapping_error(msg->data.addr[0]))
+	if (dma_mapping_error(port->dev, msg->data.addr[0]))
 		goto recycle_and_drop;
 
 	msg->dev = port->dev;
@@ -1226,7 +1226,7 @@ static void veth_recycle_msg(struct veth_lpar_connection *cnx,
 		dma_address = msg->data.addr[0];
 		dma_length = msg->data.len[0];
 
-		if (!dma_mapping_error(dma_address))
+		if (!dma_mapping_error(msg->dev, dma_address))
 			dma_unmap_single(msg->dev, dma_address, dma_length,
 					DMA_TO_DEVICE);
 
diff --git a/drivers/net/mlx4/eq.c b/drivers/net/mlx4/eq.c
index ea3a09aaa844..7df928d3a3d8 100644
--- a/drivers/net/mlx4/eq.c
+++ b/drivers/net/mlx4/eq.c
@@ -526,7 +526,7 @@ int mlx4_map_eq_icm(struct mlx4_dev *dev, u64 icm_virt)
 		return -ENOMEM;
 	priv->eq_table.icm_dma  = pci_map_page(dev->pdev, priv->eq_table.icm_page, 0,
 					       PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
-	if (pci_dma_mapping_error(priv->eq_table.icm_dma)) {
+	if (pci_dma_mapping_error(dev->pdev, priv->eq_table.icm_dma)) {
 		__free_page(priv->eq_table.icm_page);
 		return -ENOMEM;
 	}
diff --git a/drivers/net/pasemi_mac.c b/drivers/net/pasemi_mac.c
index 993d87c9296f..edc0fd588985 100644
--- a/drivers/net/pasemi_mac.c
+++ b/drivers/net/pasemi_mac.c
@@ -650,7 +650,7 @@ static void pasemi_mac_replenish_rx_ring(const struct net_device *dev,
 				     mac->bufsz - LOCAL_SKB_ALIGN,
 				     PCI_DMA_FROMDEVICE);
 
-		if (unlikely(dma_mapping_error(dma))) {
+		if (unlikely(pci_dma_mapping_error(mac->dma_pdev, dma))) {
 			dev_kfree_skb_irq(info->skb);
 			break;
 		}
@@ -1519,7 +1519,7 @@ static int pasemi_mac_start_tx(struct sk_buff *skb, struct net_device *dev)
 	map[0] = pci_map_single(mac->dma_pdev, skb->data, skb_headlen(skb),
 				PCI_DMA_TODEVICE);
 	map_size[0] = skb_headlen(skb);
-	if (dma_mapping_error(map[0]))
+	if (pci_dma_mapping_error(mac->dma_pdev, map[0]))
 		goto out_err_nolock;
 
 	for (i = 0; i < nfrags; i++) {
@@ -1529,7 +1529,7 @@ static int pasemi_mac_start_tx(struct sk_buff *skb, struct net_device *dev)
 					frag->page_offset, frag->size,
 					PCI_DMA_TODEVICE);
 		map_size[i+1] = frag->size;
-		if (dma_mapping_error(map[i+1])) {
+		if (pci_dma_mapping_error(mac->dma_pdev, map[i+1])) {
 			nfrags = i;
 			goto out_err_nolock;
 		}
diff --git a/drivers/net/qla3xxx.c b/drivers/net/qla3xxx.c
index e7d48a352beb..e82b37bbd6c3 100644
--- a/drivers/net/qla3xxx.c
+++ b/drivers/net/qla3xxx.c
@@ -328,7 +328,7 @@ static void ql_release_to_lrg_buf_free_list(struct ql3_adapter *qdev,
 					     qdev->lrg_buffer_len -
 					     QL_HEADER_SPACE,
 					     PCI_DMA_FROMDEVICE);
-			err = pci_dma_mapping_error(map);
+			err = pci_dma_mapping_error(qdev->pdev, map);
 			if(err) {
 				printk(KERN_ERR "%s: PCI mapping failed with error: %d\n",
 				       qdev->ndev->name, err);
@@ -1919,7 +1919,7 @@ static int ql_populate_free_queue(struct ql3_adapter *qdev)
 						     QL_HEADER_SPACE,
 						     PCI_DMA_FROMDEVICE);
 
-				err = pci_dma_mapping_error(map);
+				err = pci_dma_mapping_error(qdev->pdev, map);
 				if(err) {
 					printk(KERN_ERR "%s: PCI mapping failed with error: %d\n",
 					       qdev->ndev->name, err);
@@ -2454,7 +2454,7 @@ static int ql_send_map(struct ql3_adapter *qdev,
 	 */
 	map = pci_map_single(qdev->pdev, skb->data, len, PCI_DMA_TODEVICE);
 
-	err = pci_dma_mapping_error(map);
+	err = pci_dma_mapping_error(qdev->pdev, map);
 	if(err) {
 		printk(KERN_ERR "%s: PCI mapping failed with error: %d\n",
 		       qdev->ndev->name, err);
@@ -2487,7 +2487,7 @@ static int ql_send_map(struct ql3_adapter *qdev,
 						     sizeof(struct oal),
 						     PCI_DMA_TODEVICE);
 
-				err = pci_dma_mapping_error(map);
+				err = pci_dma_mapping_error(qdev->pdev, map);
 				if(err) {
 
 					printk(KERN_ERR "%s: PCI mapping outbound address list with error: %d\n",
@@ -2514,7 +2514,7 @@ static int ql_send_map(struct ql3_adapter *qdev,
 					 frag->page_offset, frag->size,
 					 PCI_DMA_TODEVICE);
 
-			err = pci_dma_mapping_error(map);
+			err = pci_dma_mapping_error(qdev->pdev, map);
 			if(err) {
 				printk(KERN_ERR "%s: PCI mapping frags failed with error: %d\n",
 				       qdev->ndev->name, err);
@@ -2916,7 +2916,7 @@ static int ql_alloc_large_buffers(struct ql3_adapter *qdev)
 					     QL_HEADER_SPACE,
 					     PCI_DMA_FROMDEVICE);
 
-			err = pci_dma_mapping_error(map);
+			err = pci_dma_mapping_error(qdev->pdev, map);
 			if(err) {
 				printk(KERN_ERR "%s: PCI mapping failed with error: %d\n",
 				       qdev->ndev->name, err);
diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c
index 9dae40ccf048..86d77d05190a 100644
--- a/drivers/net/s2io.c
+++ b/drivers/net/s2io.c
@@ -2512,8 +2512,8 @@ static void stop_nic(struct s2io_nic *nic)
  *   Return Value:
  *  SUCCESS on success or an appropriate -ve value on failure.
  */
-
-static int fill_rx_buffers(struct ring_info *ring, int from_card_up)
+static int fill_rx_buffers(struct s2io_nic *nic, struct ring_info *ring,
+				int from_card_up)
 {
 	struct sk_buff *skb;
 	struct RxD_t *rxdp;
@@ -2602,7 +2602,8 @@ static int fill_rx_buffers(struct ring_info *ring, int from_card_up)
 			rxdp1->Buffer0_ptr = pci_map_single
 			    (ring->pdev, skb->data, size - NET_IP_ALIGN,
 				PCI_DMA_FROMDEVICE);
-			if(pci_dma_mapping_error(rxdp1->Buffer0_ptr))
+			if (pci_dma_mapping_error(nic->pdev,
+						rxdp1->Buffer0_ptr))
 				goto pci_map_failed;
 
 			rxdp->Control_2 =
@@ -2636,7 +2637,8 @@ static int fill_rx_buffers(struct ring_info *ring, int from_card_up)
 				rxdp3->Buffer0_ptr =
 				   pci_map_single(ring->pdev, ba->ba_0,
 					BUF0_LEN, PCI_DMA_FROMDEVICE);
-				if (pci_dma_mapping_error(rxdp3->Buffer0_ptr))
+			if (pci_dma_mapping_error(nic->pdev,
+						rxdp3->Buffer0_ptr))
 					goto pci_map_failed;
 			} else
 				pci_dma_sync_single_for_device(ring->pdev,
@@ -2655,7 +2657,8 @@ static int fill_rx_buffers(struct ring_info *ring, int from_card_up)
 				(ring->pdev, skb->data, ring->mtu + 4,
 						PCI_DMA_FROMDEVICE);
 
-				if (pci_dma_mapping_error(rxdp3->Buffer2_ptr))
+				if (pci_dma_mapping_error(nic->pdev,
+							rxdp3->Buffer2_ptr))
 					goto pci_map_failed;
 
 				if (from_card_up) {
@@ -2664,8 +2667,8 @@ static int fill_rx_buffers(struct ring_info *ring, int from_card_up)
 						ba->ba_1, BUF1_LEN,
 						PCI_DMA_FROMDEVICE);
 
-					if (pci_dma_mapping_error
-						(rxdp3->Buffer1_ptr)) {
+					if (pci_dma_mapping_error(nic->pdev,
+						rxdp3->Buffer1_ptr)) {
 						pci_unmap_single
 							(ring->pdev,
 						    (dma_addr_t)(unsigned long)
@@ -2806,9 +2809,9 @@ static void free_rx_buffers(struct s2io_nic *sp)
 	}
 }
 
-static int s2io_chk_rx_buffers(struct ring_info *ring)
+static int s2io_chk_rx_buffers(struct s2io_nic *nic, struct ring_info *ring)
 {
-	if (fill_rx_buffers(ring, 0) == -ENOMEM) {
+	if (fill_rx_buffers(nic, ring, 0) == -ENOMEM) {
 		DBG_PRINT(INFO_DBG, "%s:Out of memory", ring->dev->name);
 		DBG_PRINT(INFO_DBG, " in Rx Intr!!\n");
 	}
@@ -2848,7 +2851,7 @@ static int s2io_poll_msix(struct napi_struct *napi, int budget)
 		return 0;
 
 	pkts_processed = rx_intr_handler(ring, budget);
-	s2io_chk_rx_buffers(ring);
+	s2io_chk_rx_buffers(nic, ring);
 
 	if (pkts_processed < budget_org) {
 		netif_rx_complete(dev, napi);
@@ -2882,7 +2885,7 @@ static int s2io_poll_inta(struct napi_struct *napi, int budget)
 	for (i = 0; i < config->rx_ring_num; i++) {
 		ring = &mac_control->rings[i];
 		ring_pkts_processed = rx_intr_handler(ring, budget);
-		s2io_chk_rx_buffers(ring);
+		s2io_chk_rx_buffers(nic, ring);
 		pkts_processed += ring_pkts_processed;
 		budget -= ring_pkts_processed;
 		if (budget <= 0)
@@ -2939,7 +2942,8 @@ static void s2io_netpoll(struct net_device *dev)
 		rx_intr_handler(&mac_control->rings[i], 0);
 
 	for (i = 0; i < config->rx_ring_num; i++) {
-		if (fill_rx_buffers(&mac_control->rings[i], 0) == -ENOMEM) {
+		if (fill_rx_buffers(nic, &mac_control->rings[i], 0) ==
+				-ENOMEM) {
 			DBG_PRINT(INFO_DBG, "%s:Out of memory", dev->name);
 			DBG_PRINT(INFO_DBG, " in Rx Netpoll!!\n");
 			break;
@@ -4235,14 +4239,14 @@ static int s2io_xmit(struct sk_buff *skb, struct net_device *dev)
 		txdp->Buffer_Pointer = pci_map_single(sp->pdev,
 					fifo->ufo_in_band_v,
 					sizeof(u64), PCI_DMA_TODEVICE);
-		if (pci_dma_mapping_error(txdp->Buffer_Pointer))
+		if (pci_dma_mapping_error(sp->pdev, txdp->Buffer_Pointer))
 			goto pci_map_failed;
 		txdp++;
 	}
 
 	txdp->Buffer_Pointer = pci_map_single
 	    (sp->pdev, skb->data, frg_len, PCI_DMA_TODEVICE);
-	if (pci_dma_mapping_error(txdp->Buffer_Pointer))
+	if (pci_dma_mapping_error(sp->pdev, txdp->Buffer_Pointer))
 		goto pci_map_failed;
 
 	txdp->Host_Control = (unsigned long) skb;
@@ -4345,7 +4349,7 @@ static irqreturn_t s2io_msix_ring_handle(int irq, void *dev_id)
 		netif_rx_schedule(dev, &ring->napi);
 	} else {
 		rx_intr_handler(ring, 0);
-		s2io_chk_rx_buffers(ring);
+		s2io_chk_rx_buffers(sp, ring);
 	}
 
 	return IRQ_HANDLED;
@@ -4826,7 +4830,7 @@ static irqreturn_t s2io_isr(int irq, void *dev_id)
 		 */
 		if (!config->napi) {
 			for (i = 0; i < config->rx_ring_num; i++)
-				s2io_chk_rx_buffers(&mac_control->rings[i]);
+				s2io_chk_rx_buffers(sp, &mac_control->rings[i]);
 		}
 		writeq(sp->general_int_mask, &bar0->general_int_mask);
 		readl(&bar0->general_int_status);
@@ -6859,7 +6863,7 @@ static int set_rxd_buffer_pointer(struct s2io_nic *sp, struct RxD_t *rxdp,
 				pci_map_single( sp->pdev, (*skb)->data,
 					size - NET_IP_ALIGN,
 					PCI_DMA_FROMDEVICE);
-			if (pci_dma_mapping_error(rxdp1->Buffer0_ptr))
+			if (pci_dma_mapping_error(sp->pdev, rxdp1->Buffer0_ptr))
 				goto memalloc_failed;
 			rxdp->Host_Control = (unsigned long) (*skb);
 		}
@@ -6886,12 +6890,13 @@ static int set_rxd_buffer_pointer(struct s2io_nic *sp, struct RxD_t *rxdp,
 				pci_map_single(sp->pdev, (*skb)->data,
 					       dev->mtu + 4,
 					       PCI_DMA_FROMDEVICE);
-			if (pci_dma_mapping_error(rxdp3->Buffer2_ptr))
+			if (pci_dma_mapping_error(sp->pdev, rxdp3->Buffer2_ptr))
 				goto memalloc_failed;
 			rxdp3->Buffer0_ptr = *temp0 =
 				pci_map_single( sp->pdev, ba->ba_0, BUF0_LEN,
 						PCI_DMA_FROMDEVICE);
-			if (pci_dma_mapping_error(rxdp3->Buffer0_ptr)) {
+			if (pci_dma_mapping_error(sp->pdev,
+						rxdp3->Buffer0_ptr)) {
 				pci_unmap_single (sp->pdev,
 					(dma_addr_t)rxdp3->Buffer2_ptr,
 					dev->mtu + 4, PCI_DMA_FROMDEVICE);
@@ -6903,7 +6908,8 @@ static int set_rxd_buffer_pointer(struct s2io_nic *sp, struct RxD_t *rxdp,
 			rxdp3->Buffer1_ptr = *temp1 =
 				pci_map_single(sp->pdev, ba->ba_1, BUF1_LEN,
 						PCI_DMA_FROMDEVICE);
-			if (pci_dma_mapping_error(rxdp3->Buffer1_ptr)) {
+			if (pci_dma_mapping_error(sp->pdev,
+						rxdp3->Buffer1_ptr)) {
 				pci_unmap_single (sp->pdev,
 					(dma_addr_t)rxdp3->Buffer0_ptr,
 					BUF0_LEN, PCI_DMA_FROMDEVICE);
@@ -7187,7 +7193,7 @@ static int s2io_card_up(struct s2io_nic * sp)
 
 	for (i = 0; i < config->rx_ring_num; i++) {
 		mac_control->rings[i].mtu = dev->mtu;
-		ret = fill_rx_buffers(&mac_control->rings[i], 1);
+		ret = fill_rx_buffers(sp, &mac_control->rings[i], 1);
 		if (ret) {
 			DBG_PRINT(ERR_DBG, "%s: Out of memory in Open\n",
 				  dev->name);
diff --git a/drivers/net/sfc/rx.c b/drivers/net/sfc/rx.c
index 601b001437c0..0d27dd39bc09 100644
--- a/drivers/net/sfc/rx.c
+++ b/drivers/net/sfc/rx.c
@@ -233,7 +233,7 @@ static inline int efx_init_rx_buffer_skb(struct efx_rx_queue *rx_queue,
 					  rx_buf->data, rx_buf->len,
 					  PCI_DMA_FROMDEVICE);
 
-	if (unlikely(pci_dma_mapping_error(rx_buf->dma_addr))) {
+	if (unlikely(pci_dma_mapping_error(efx->pci_dev, rx_buf->dma_addr))) {
 		dev_kfree_skb_any(rx_buf->skb);
 		rx_buf->skb = NULL;
 		return -EIO;
@@ -275,7 +275,7 @@ static inline int efx_init_rx_buffer_page(struct efx_rx_queue *rx_queue,
 					0, efx_rx_buf_size(efx),
 					PCI_DMA_FROMDEVICE);
 
-		if (unlikely(pci_dma_mapping_error(dma_addr))) {
+		if (unlikely(pci_dma_mapping_error(efx->pci_dev, dma_addr))) {
 			__free_pages(rx_buf->page, efx->rx_buffer_order);
 			rx_buf->page = NULL;
 			return -EIO;
diff --git a/drivers/net/sfc/tx.c b/drivers/net/sfc/tx.c
index 5cdd082ab8f6..5e8374ab28ee 100644
--- a/drivers/net/sfc/tx.c
+++ b/drivers/net/sfc/tx.c
@@ -172,7 +172,7 @@ static inline int efx_enqueue_skb(struct efx_tx_queue *tx_queue,
 
 	/* Process all fragments */
 	while (1) {
-		if (unlikely(pci_dma_mapping_error(dma_addr)))
+		if (unlikely(pci_dma_mapping_error(pci_dev, dma_addr)))
 			goto pci_err;
 
 		/* Store fields for marking in the per-fragment final
@@ -661,7 +661,8 @@ efx_tsoh_heap_alloc(struct efx_tx_queue *tx_queue, size_t header_len)
 	tsoh->dma_addr = pci_map_single(tx_queue->efx->pci_dev,
 					TSOH_BUFFER(tsoh), header_len,
 					PCI_DMA_TODEVICE);
-	if (unlikely(pci_dma_mapping_error(tsoh->dma_addr))) {
+	if (unlikely(pci_dma_mapping_error(tx_queue->efx->pci_dev,
+					   tsoh->dma_addr))) {
 		kfree(tsoh);
 		return NULL;
 	}
@@ -863,7 +864,7 @@ static inline int tso_get_fragment(struct tso_state *st, struct efx_nic *efx,
 
 	st->ifc.unmap_addr = pci_map_page(efx->pci_dev, page, page_off,
 					  len, PCI_DMA_TODEVICE);
-	if (likely(!pci_dma_mapping_error(st->ifc.unmap_addr))) {
+	if (likely(!pci_dma_mapping_error(efx->pci_dev, st->ifc.unmap_addr))) {
 		st->ifc.unmap_len = len;
 		st->ifc.len = len;
 		st->ifc.dma_addr = st->ifc.unmap_addr;
diff --git a/drivers/net/spider_net.c b/drivers/net/spider_net.c
index 00aa0b108cb9..b6435d0d71f9 100644
--- a/drivers/net/spider_net.c
+++ b/drivers/net/spider_net.c
@@ -452,7 +452,7 @@ spider_net_prepare_rx_descr(struct spider_net_card *card,
 	/* iommu-map the skb */
 	buf = pci_map_single(card->pdev, descr->skb->data,
 			SPIDER_NET_MAX_FRAME, PCI_DMA_FROMDEVICE);
-	if (pci_dma_mapping_error(buf)) {
+	if (pci_dma_mapping_error(card->pdev, buf)) {
 		dev_kfree_skb_any(descr->skb);
 		descr->skb = NULL;
 		if (netif_msg_rx_err(card) && net_ratelimit())
@@ -691,7 +691,7 @@ spider_net_prepare_tx_descr(struct spider_net_card *card,
 	unsigned long flags;
 
 	buf = pci_map_single(card->pdev, skb->data, skb->len, PCI_DMA_TODEVICE);
-	if (pci_dma_mapping_error(buf)) {
+	if (pci_dma_mapping_error(card->pdev, buf)) {
 		if (netif_msg_tx_err(card) && net_ratelimit())
 			dev_err(&card->netdev->dev, "could not iommu-map packet (%p, %i). "
 				  "Dropping packet\n", skb->data, skb->len);
diff --git a/drivers/net/tc35815.c b/drivers/net/tc35815.c
index a645e5028c14..8487ace9d2e3 100644
--- a/drivers/net/tc35815.c
+++ b/drivers/net/tc35815.c
@@ -506,7 +506,7 @@ static void *alloc_rxbuf_page(struct pci_dev *hwdev, dma_addr_t *dma_handle)
 		return NULL;
 	*dma_handle = pci_map_single(hwdev, buf, PAGE_SIZE,
 				     PCI_DMA_FROMDEVICE);
-	if (pci_dma_mapping_error(*dma_handle)) {
+	if (pci_dma_mapping_error(hwdev, *dma_handle)) {
 		free_page((unsigned long)buf);
 		return NULL;
 	}
@@ -536,7 +536,7 @@ static struct sk_buff *alloc_rxbuf_skb(struct net_device *dev,
 		return NULL;
 	*dma_handle = pci_map_single(hwdev, skb->data, RX_BUF_SIZE,
 				     PCI_DMA_FROMDEVICE);
-	if (pci_dma_mapping_error(*dma_handle)) {
+	if (pci_dma_mapping_error(hwdev, *dma_handle)) {
 		dev_kfree_skb_any(skb);
 		return NULL;
 	}
diff --git a/drivers/net/wireless/ath5k/base.c b/drivers/net/wireless/ath5k/base.c
index 217d506527a9..d9769c527346 100644
--- a/drivers/net/wireless/ath5k/base.c
+++ b/drivers/net/wireless/ath5k/base.c
@@ -1166,7 +1166,7 @@ ath5k_rxbuf_setup(struct ath5k_softc *sc, struct ath5k_buf *bf)
 		bf->skb = skb;
 		bf->skbaddr = pci_map_single(sc->pdev,
 			skb->data, sc->rxbufsize, PCI_DMA_FROMDEVICE);
-		if (unlikely(pci_dma_mapping_error(bf->skbaddr))) {
+		if (unlikely(pci_dma_mapping_error(sc->pdev, bf->skbaddr))) {
 			ATH5K_ERR(sc, "%s: DMA mapping failed\n", __func__);
 			dev_kfree_skb(skb);
 			bf->skb = NULL;
@@ -1918,7 +1918,7 @@ ath5k_beacon_setup(struct ath5k_softc *sc, struct ath5k_buf *bf)
 	ATH5K_DBG(sc, ATH5K_DEBUG_BEACON, "skb %p [data %p len %u] "
 			"skbaddr %llx\n", skb, skb->data, skb->len,
 			(unsigned long long)bf->skbaddr);
-	if (pci_dma_mapping_error(bf->skbaddr)) {
+	if (pci_dma_mapping_error(sc->pdev, bf->skbaddr)) {
 		ATH5K_ERR(sc, "beacon DMA mapping failed\n");
 		return -EIO;
 	}
diff --git a/drivers/scsi/ibmvscsi/ibmvfc.c b/drivers/scsi/ibmvscsi/ibmvfc.c
index c4a7c06793c5..61f8fdea2d96 100644
--- a/drivers/scsi/ibmvscsi/ibmvfc.c
+++ b/drivers/scsi/ibmvscsi/ibmvfc.c
@@ -3525,7 +3525,7 @@ static int ibmvfc_init_crq(struct ibmvfc_host *vhost)
 	crq->msg_token = dma_map_single(dev, crq->msgs,
 					PAGE_SIZE, DMA_BIDIRECTIONAL);
 
-	if (dma_mapping_error(crq->msg_token))
+	if (dma_mapping_error(dev, crq->msg_token))
 		goto map_failed;
 
 	retrc = rc = plpar_hcall_norets(H_REG_CRQ, vdev->unit_address,
@@ -3618,7 +3618,7 @@ static int ibmvfc_alloc_mem(struct ibmvfc_host *vhost)
 					    async_q->size * sizeof(*async_q->msgs),
 					    DMA_BIDIRECTIONAL);
 
-	if (dma_mapping_error(async_q->msg_token)) {
+	if (dma_mapping_error(dev, async_q->msg_token)) {
 		dev_err(dev, "Failed to map async queue\n");
 		goto free_async_crq;
 	}
diff --git a/drivers/scsi/ibmvscsi/ibmvscsi.c b/drivers/scsi/ibmvscsi/ibmvscsi.c
index 20000ec79b04..6b24b9cdb04c 100644
--- a/drivers/scsi/ibmvscsi/ibmvscsi.c
+++ b/drivers/scsi/ibmvscsi/ibmvscsi.c
@@ -859,7 +859,7 @@ static void send_mad_adapter_info(struct ibmvscsi_host_data *hostdata)
 					    sizeof(hostdata->madapter_info),
 					    DMA_BIDIRECTIONAL);
 
-	if (dma_mapping_error(req->buffer)) {
+	if (dma_mapping_error(hostdata->dev, req->buffer)) {
 		if (!firmware_has_feature(FW_FEATURE_CMO))
 			dev_err(hostdata->dev,
 			        "Unable to map request_buffer for "
@@ -1407,7 +1407,7 @@ static int ibmvscsi_do_host_config(struct ibmvscsi_host_data *hostdata,
 						    length,
 						    DMA_BIDIRECTIONAL);
 
-	if (dma_mapping_error(host_config->buffer)) {
+	if (dma_mapping_error(hostdata->dev, host_config->buffer)) {
 		if (!firmware_has_feature(FW_FEATURE_CMO))
 			dev_err(hostdata->dev,
 			        "dma_mapping error getting host config\n");
diff --git a/drivers/scsi/ibmvscsi/ibmvstgt.c b/drivers/scsi/ibmvscsi/ibmvstgt.c
index 3b9514c8f1f1..2e13ec00172a 100644
--- a/drivers/scsi/ibmvscsi/ibmvstgt.c
+++ b/drivers/scsi/ibmvscsi/ibmvstgt.c
@@ -564,7 +564,7 @@ static int crq_queue_create(struct crq_queue *queue, struct srp_target *target)
 					  queue->size * sizeof(*queue->msgs),
 					  DMA_BIDIRECTIONAL);
 
-	if (dma_mapping_error(queue->msg_token))
+	if (dma_mapping_error(target->dev, queue->msg_token))
 		goto map_failed;
 
 	err = h_reg_crq(vport->dma_dev->unit_address, queue->msg_token,
diff --git a/drivers/scsi/ibmvscsi/rpa_vscsi.c b/drivers/scsi/ibmvscsi/rpa_vscsi.c
index 182146100dc1..462a8574dad9 100644
--- a/drivers/scsi/ibmvscsi/rpa_vscsi.c
+++ b/drivers/scsi/ibmvscsi/rpa_vscsi.c
@@ -253,7 +253,7 @@ static int rpavscsi_init_crq_queue(struct crq_queue *queue,
 					  queue->size * sizeof(*queue->msgs),
 					  DMA_BIDIRECTIONAL);
 
-	if (dma_mapping_error(queue->msg_token))
+	if (dma_mapping_error(hostdata->dev, queue->msg_token))
 		goto map_failed;
 
 	gather_partition_info();
diff --git a/drivers/spi/atmel_spi.c b/drivers/spi/atmel_spi.c
index e81d59d78910..0c7165660853 100644
--- a/drivers/spi/atmel_spi.c
+++ b/drivers/spi/atmel_spi.c
@@ -313,14 +313,14 @@ atmel_spi_dma_map_xfer(struct atmel_spi *as, struct spi_transfer *xfer)
 		xfer->tx_dma = dma_map_single(dev,
 				(void *) xfer->tx_buf, xfer->len,
 				DMA_TO_DEVICE);
-		if (dma_mapping_error(xfer->tx_dma))
+		if (dma_mapping_error(dev, xfer->tx_dma))
 			return -ENOMEM;
 	}
 	if (xfer->rx_buf) {
 		xfer->rx_dma = dma_map_single(dev,
 				xfer->rx_buf, xfer->len,
 				DMA_FROM_DEVICE);
-		if (dma_mapping_error(xfer->rx_dma)) {
+		if (dma_mapping_error(dev, xfer->rx_dma)) {
 			if (xfer->tx_buf)
 				dma_unmap_single(dev,
 						xfer->tx_dma, xfer->len,
diff --git a/drivers/spi/au1550_spi.c b/drivers/spi/au1550_spi.c
index 9149689c79d9..87b73e0169c5 100644
--- a/drivers/spi/au1550_spi.c
+++ b/drivers/spi/au1550_spi.c
@@ -334,7 +334,7 @@ static int au1550_spi_dma_rxtmp_alloc(struct au1550_spi *hw, unsigned size)
 	hw->dma_rx_tmpbuf_size = size;
 	hw->dma_rx_tmpbuf_addr = dma_map_single(hw->dev, hw->dma_rx_tmpbuf,
 			size, DMA_FROM_DEVICE);
-	if (dma_mapping_error(hw->dma_rx_tmpbuf_addr)) {
+	if (dma_mapping_error(hw->dev, hw->dma_rx_tmpbuf_addr)) {
 		kfree(hw->dma_rx_tmpbuf);
 		hw->dma_rx_tmpbuf = 0;
 		hw->dma_rx_tmpbuf_size = 0;
@@ -378,7 +378,7 @@ static int au1550_spi_dma_txrxb(struct spi_device *spi, struct spi_transfer *t)
 			dma_rx_addr = dma_map_single(hw->dev,
 					(void *)t->rx_buf,
 					t->len, DMA_FROM_DEVICE);
-			if (dma_mapping_error(dma_rx_addr))
+			if (dma_mapping_error(hw->dev, dma_rx_addr))
 				dev_err(hw->dev, "rx dma map error\n");
 		}
 	} else {
@@ -401,7 +401,7 @@ static int au1550_spi_dma_txrxb(struct spi_device *spi, struct spi_transfer *t)
 			dma_tx_addr = dma_map_single(hw->dev,
 					(void *)t->tx_buf,
 					t->len, DMA_TO_DEVICE);
-			if (dma_mapping_error(dma_tx_addr))
+			if (dma_mapping_error(hw->dev, dma_tx_addr))
 				dev_err(hw->dev, "tx dma map error\n");
 		}
 	} else {
diff --git a/drivers/spi/omap2_mcspi.c b/drivers/spi/omap2_mcspi.c
index b1cc148036c1..f6f987bb71ca 100644
--- a/drivers/spi/omap2_mcspi.c
+++ b/drivers/spi/omap2_mcspi.c
@@ -836,7 +836,7 @@ static int omap2_mcspi_transfer(struct spi_device *spi, struct spi_message *m)
 		if (tx_buf != NULL) {
 			t->tx_dma = dma_map_single(&spi->dev, (void *) tx_buf,
 					len, DMA_TO_DEVICE);
-			if (dma_mapping_error(t->tx_dma)) {
+			if (dma_mapping_error(&spi->dev, t->tx_dma)) {
 				dev_dbg(&spi->dev, "dma %cX %d bytes error\n",
 						'T', len);
 				return -EINVAL;
@@ -845,7 +845,7 @@ static int omap2_mcspi_transfer(struct spi_device *spi, struct spi_message *m)
 		if (rx_buf != NULL) {
 			t->rx_dma = dma_map_single(&spi->dev, rx_buf, t->len,
 					DMA_FROM_DEVICE);
-			if (dma_mapping_error(t->rx_dma)) {
+			if (dma_mapping_error(&spi->dev, t->rx_dma)) {
 				dev_dbg(&spi->dev, "dma %cX %d bytes error\n",
 						'R', len);
 				if (tx_buf != NULL)
diff --git a/drivers/spi/pxa2xx_spi.c b/drivers/spi/pxa2xx_spi.c
index 0c452c46ab07..067299d6d192 100644
--- a/drivers/spi/pxa2xx_spi.c
+++ b/drivers/spi/pxa2xx_spi.c
@@ -353,7 +353,7 @@ static int map_dma_buffers(struct driver_data *drv_data)
 	drv_data->rx_dma = dma_map_single(dev, drv_data->rx,
 						drv_data->rx_map_len,
 						DMA_FROM_DEVICE);
-	if (dma_mapping_error(drv_data->rx_dma))
+	if (dma_mapping_error(dev, drv_data->rx_dma))
 		return 0;
 
 	/* Stream map the tx buffer */
@@ -361,7 +361,7 @@ static int map_dma_buffers(struct driver_data *drv_data)
 						drv_data->tx_map_len,
 						DMA_TO_DEVICE);
 
-	if (dma_mapping_error(drv_data->tx_dma)) {
+	if (dma_mapping_error(dev, drv_data->tx_dma)) {
 		dma_unmap_single(dev, drv_data->rx_dma,
 					drv_data->rx_map_len, DMA_FROM_DEVICE);
 		return 0;
diff --git a/drivers/spi/spi_imx.c b/drivers/spi/spi_imx.c
index 54ac7bea5f8c..6fb77fcc4971 100644
--- a/drivers/spi/spi_imx.c
+++ b/drivers/spi/spi_imx.c
@@ -491,7 +491,7 @@ static int map_dma_buffers(struct driver_data *drv_data)
 							buf,
 							drv_data->tx_map_len,
 							DMA_TO_DEVICE);
-			if (dma_mapping_error(drv_data->tx_dma))
+			if (dma_mapping_error(dev, drv_data->tx_dma))
 				return -1;
 
 			drv_data->tx_dma_needs_unmap = 1;
@@ -516,7 +516,7 @@ static int map_dma_buffers(struct driver_data *drv_data)
 					buf,
 					drv_data->len,
 					DMA_FROM_DEVICE);
-		if (dma_mapping_error(drv_data->rx_dma))
+		if (dma_mapping_error(dev, drv_data->rx_dma))
 			return -1;
 		drv_data->rx_dma_needs_unmap = 1;
 	}
@@ -534,7 +534,7 @@ static int map_dma_buffers(struct driver_data *drv_data)
 					buf,
 					drv_data->tx_map_len,
 					DMA_TO_DEVICE);
-	if (dma_mapping_error(drv_data->tx_dma)) {
+	if (dma_mapping_error(dev, drv_data->tx_dma)) {
 		if (drv_data->rx_dma) {
 			dma_unmap_single(dev,
 					drv_data->rx_dma,
diff --git a/include/asm-alpha/dma-mapping.h b/include/asm-alpha/dma-mapping.h
index db351d1296f4..a5801ae02e4b 100644
--- a/include/asm-alpha/dma-mapping.h
+++ b/include/asm-alpha/dma-mapping.h
@@ -24,8 +24,8 @@
 		pci_unmap_sg(alpha_gendev_to_pci(dev), sg, nents, dir)
 #define dma_supported(dev, mask)			\
 		pci_dma_supported(alpha_gendev_to_pci(dev), mask)
-#define dma_mapping_error(addr)				\
-		pci_dma_mapping_error(addr)
+#define dma_mapping_error(dev, addr)				\
+		pci_dma_mapping_error(alpha_gendev_to_pci(dev), addr)
 
 #else	/* no PCI - no IOMMU. */
 
@@ -45,7 +45,7 @@ int dma_map_sg(struct device *dev, struct scatterlist *sg, int nents,
 #define dma_unmap_page(dev, addr, size, dir)	((void)0)
 #define dma_unmap_sg(dev, sg, nents, dir)	((void)0)
 
-#define dma_mapping_error(addr)  (0)
+#define dma_mapping_error(dev, addr)  (0)
 
 #endif	/* !CONFIG_PCI */
 
diff --git a/include/asm-alpha/pci.h b/include/asm-alpha/pci.h
index d31fd49ff79a..2a14302c17a3 100644
--- a/include/asm-alpha/pci.h
+++ b/include/asm-alpha/pci.h
@@ -106,7 +106,7 @@ extern dma_addr_t pci_map_page(struct pci_dev *, struct page *,
 /* Test for pci_map_single or pci_map_page having generated an error.  */
 
 static inline int
-pci_dma_mapping_error(dma_addr_t dma_addr)
+pci_dma_mapping_error(struct pci_dev *pdev, dma_addr_t dma_addr)
 {
 	return dma_addr == 0;
 }
diff --git a/include/asm-arm/dma-mapping.h b/include/asm-arm/dma-mapping.h
index e99406a7bece..f41335ba6337 100644
--- a/include/asm-arm/dma-mapping.h
+++ b/include/asm-arm/dma-mapping.h
@@ -56,7 +56,7 @@ static inline int dma_is_consistent(struct device *dev, dma_addr_t handle)
 /*
  * DMA errors are defined by all-bits-set in the DMA address.
  */
-static inline int dma_mapping_error(dma_addr_t dma_addr)
+static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
 {
 	return dma_addr == ~0;
 }
diff --git a/include/asm-avr32/dma-mapping.h b/include/asm-avr32/dma-mapping.h
index 57dc672bab8e..0399359ab5d8 100644
--- a/include/asm-avr32/dma-mapping.h
+++ b/include/asm-avr32/dma-mapping.h
@@ -35,7 +35,7 @@ static inline int dma_set_mask(struct device *dev, u64 dma_mask)
 /*
  * dma_map_single can't fail as it is implemented now.
  */
-static inline int dma_mapping_error(dma_addr_t addr)
+static inline int dma_mapping_error(struct device *dev, dma_addr_t addr)
 {
 	return 0;
 }
diff --git a/include/asm-cris/dma-mapping.h b/include/asm-cris/dma-mapping.h
index edc8d1bfaae2..cb2fb25ff8d9 100644
--- a/include/asm-cris/dma-mapping.h
+++ b/include/asm-cris/dma-mapping.h
@@ -120,7 +120,7 @@ dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nelems,
 }
 
 static inline int
-dma_mapping_error(dma_addr_t dma_addr)
+dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
 {
 	return 0;
 }
diff --git a/include/asm-frv/dma-mapping.h b/include/asm-frv/dma-mapping.h
index 2e8966ca030d..b2898877c07b 100644
--- a/include/asm-frv/dma-mapping.h
+++ b/include/asm-frv/dma-mapping.h
@@ -126,7 +126,7 @@ void dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nele
 }
 
 static inline
-int dma_mapping_error(dma_addr_t dma_addr)
+int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
 {
 	return 0;
 }
diff --git a/include/asm-generic/dma-mapping-broken.h b/include/asm-generic/dma-mapping-broken.h
index e2468f894d2a..82cd0cb1c3fe 100644
--- a/include/asm-generic/dma-mapping-broken.h
+++ b/include/asm-generic/dma-mapping-broken.h
@@ -61,7 +61,7 @@ dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nelems,
 #define dma_sync_sg_for_device dma_sync_sg_for_cpu
 
 extern int
-dma_mapping_error(dma_addr_t dma_addr);
+dma_mapping_error(struct device *dev, dma_addr_t dma_addr);
 
 extern int
 dma_supported(struct device *dev, u64 mask);
diff --git a/include/asm-generic/dma-mapping.h b/include/asm-generic/dma-mapping.h
index 783ab9944d70..189486c3f92e 100644
--- a/include/asm-generic/dma-mapping.h
+++ b/include/asm-generic/dma-mapping.h
@@ -144,9 +144,9 @@ dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nelems,
 }
 
 static inline int
-dma_mapping_error(dma_addr_t dma_addr)
+dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
 {
-	return pci_dma_mapping_error(dma_addr);
+	return pci_dma_mapping_error(to_pci_dev(dev), dma_addr);
 }
 
 
diff --git a/include/asm-generic/pci-dma-compat.h b/include/asm-generic/pci-dma-compat.h
index 25c10e96b2b7..37b3706226e7 100644
--- a/include/asm-generic/pci-dma-compat.h
+++ b/include/asm-generic/pci-dma-compat.h
@@ -99,9 +99,9 @@ pci_dma_sync_sg_for_device(struct pci_dev *hwdev, struct scatterlist *sg,
 }
 
 static inline int
-pci_dma_mapping_error(dma_addr_t dma_addr)
+pci_dma_mapping_error(struct pci_dev *pdev, dma_addr_t dma_addr)
 {
-	return dma_mapping_error(dma_addr);
+	return dma_mapping_error(&pdev->dev, dma_addr);
 }
 
 #endif
diff --git a/include/asm-ia64/machvec.h b/include/asm-ia64/machvec.h
index 0721a5e8271e..a6d50c77b6bf 100644
--- a/include/asm-ia64/machvec.h
+++ b/include/asm-ia64/machvec.h
@@ -54,7 +54,7 @@ typedef void ia64_mv_dma_sync_single_for_cpu (struct device *, dma_addr_t, size_
 typedef void ia64_mv_dma_sync_sg_for_cpu (struct device *, struct scatterlist *, int, int);
 typedef void ia64_mv_dma_sync_single_for_device (struct device *, dma_addr_t, size_t, int);
 typedef void ia64_mv_dma_sync_sg_for_device (struct device *, struct scatterlist *, int, int);
-typedef int ia64_mv_dma_mapping_error (dma_addr_t dma_addr);
+typedef int ia64_mv_dma_mapping_error(struct device *, dma_addr_t dma_addr);
 typedef int ia64_mv_dma_supported (struct device *, u64);
 
 typedef dma_addr_t ia64_mv_dma_map_single_attrs (struct device *, void *, size_t, int, struct dma_attrs *);
diff --git a/include/asm-m68k/dma-mapping.h b/include/asm-m68k/dma-mapping.h
index a26cdeb46a57..91f7944333d4 100644
--- a/include/asm-m68k/dma-mapping.h
+++ b/include/asm-m68k/dma-mapping.h
@@ -84,7 +84,7 @@ static inline void dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *s
 {
 }
 
-static inline int dma_mapping_error(dma_addr_t handle)
+static inline int dma_mapping_error(struct device *dev, dma_addr_t handle)
 {
 	return 0;
 }
diff --git a/include/asm-mips/dma-mapping.h b/include/asm-mips/dma-mapping.h
index 230b3f1b69b1..c64afb40cd06 100644
--- a/include/asm-mips/dma-mapping.h
+++ b/include/asm-mips/dma-mapping.h
@@ -42,7 +42,7 @@ extern void dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg,
 	int nelems, enum dma_data_direction direction);
 extern void dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg,
 	int nelems, enum dma_data_direction direction);
-extern int dma_mapping_error(dma_addr_t dma_addr);
+extern int dma_mapping_error(struct device *dev, dma_addr_t dma_addr);
 extern int dma_supported(struct device *dev, u64 mask);
 
 static inline int
diff --git a/include/asm-mn10300/dma-mapping.h b/include/asm-mn10300/dma-mapping.h
index 7c882fca9ec8..ccae8f6c6326 100644
--- a/include/asm-mn10300/dma-mapping.h
+++ b/include/asm-mn10300/dma-mapping.h
@@ -182,7 +182,7 @@ void dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg,
 }
 
 static inline
-int dma_mapping_error(dma_addr_t dma_addr)
+int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
 {
 	return 0;
 }
diff --git a/include/asm-parisc/dma-mapping.h b/include/asm-parisc/dma-mapping.h
index c6c0e9ff6bde..53af696f23d2 100644
--- a/include/asm-parisc/dma-mapping.h
+++ b/include/asm-parisc/dma-mapping.h
@@ -248,6 +248,6 @@ void * sba_get_iommu(struct parisc_device *dev);
 #endif
 
 /* At the moment, we panic on error for IOMMU resource exaustion */
-#define dma_mapping_error(x)	0
+#define dma_mapping_error(dev, x)	0
 
 #endif
diff --git a/include/asm-powerpc/dma-mapping.h b/include/asm-powerpc/dma-mapping.h
index 74c549780987..c7ca45f97dd2 100644
--- a/include/asm-powerpc/dma-mapping.h
+++ b/include/asm-powerpc/dma-mapping.h
@@ -415,7 +415,7 @@ static inline void dma_sync_sg_for_device(struct device *dev,
 		__dma_sync_page(sg_page(sg), sg->offset, sg->length, direction);
 }
 
-static inline int dma_mapping_error(dma_addr_t dma_addr)
+static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
 {
 #ifdef CONFIG_PPC64
 	return (dma_addr == DMA_ERROR_CODE);
diff --git a/include/asm-sh/dma-mapping.h b/include/asm-sh/dma-mapping.h
index 22cc419389fe..6c0b8a2de143 100644
--- a/include/asm-sh/dma-mapping.h
+++ b/include/asm-sh/dma-mapping.h
@@ -171,7 +171,7 @@ static inline int dma_get_cache_alignment(void)
 	return L1_CACHE_BYTES;
 }
 
-static inline int dma_mapping_error(dma_addr_t dma_addr)
+static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
 {
 	return dma_addr == 0;
 }
diff --git a/include/asm-sparc/dma-mapping_64.h b/include/asm-sparc/dma-mapping_64.h
index 38cbec76a33f..bfa64f9702d5 100644
--- a/include/asm-sparc/dma-mapping_64.h
+++ b/include/asm-sparc/dma-mapping_64.h
@@ -135,7 +135,7 @@ static inline void dma_sync_sg_for_device(struct device *dev,
 	/* No flushing needed to sync cpu writes to the device.  */
 }
 
-static inline int dma_mapping_error(dma_addr_t dma_addr)
+static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
 {
 	return (dma_addr == DMA_ERROR_CODE);
 }
diff --git a/include/asm-sparc/pci_32.h b/include/asm-sparc/pci_32.h
index b93b6c79e08f..0ee949d220c0 100644
--- a/include/asm-sparc/pci_32.h
+++ b/include/asm-sparc/pci_32.h
@@ -154,7 +154,8 @@ static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 
 #define PCI_DMA_ERROR_CODE      (~(dma_addr_t)0x0)
 
-static inline int pci_dma_mapping_error(dma_addr_t dma_addr)
+static inline int pci_dma_mapping_error(struct pci_dev *pdev,
+					dma_addr_t dma_addr)
 {
         return (dma_addr == PCI_DMA_ERROR_CODE);
 }
diff --git a/include/asm-sparc/pci_64.h b/include/asm-sparc/pci_64.h
index f59f2571295b..4f79a54948f6 100644
--- a/include/asm-sparc/pci_64.h
+++ b/include/asm-sparc/pci_64.h
@@ -140,9 +140,10 @@ extern int pci_dma_supported(struct pci_dev *hwdev, u64 mask);
 #define PCI64_REQUIRED_MASK	(~(dma64_addr_t)0)
 #define PCI64_ADDR_BASE		0xfffc000000000000UL
 
-static inline int pci_dma_mapping_error(dma_addr_t dma_addr)
+static inline int pci_dma_mapping_error(struct pci_dev *pdev,
+					dma_addr_t dma_addr)
 {
-	return dma_mapping_error(dma_addr);
+	return dma_mapping_error(&pdev->dev, dma_addr);
 }
 
 #ifdef CONFIG_PCI
diff --git a/include/asm-x86/device.h b/include/asm-x86/device.h
index 87a715367a1b..3c034f48fdb0 100644
--- a/include/asm-x86/device.h
+++ b/include/asm-x86/device.h
@@ -5,6 +5,9 @@ struct dev_archdata {
 #ifdef CONFIG_ACPI
 	void	*acpi_handle;
 #endif
+#ifdef CONFIG_X86_64
+struct dma_mapping_ops *dma_ops;
+#endif
 #ifdef CONFIG_DMAR
 	void *iommu; /* hook for IOMMU specific extension */
 #endif
diff --git a/include/asm-x86/dma-mapping.h b/include/asm-x86/dma-mapping.h
index c2ddd3d1b883..0eaa9bf6011f 100644
--- a/include/asm-x86/dma-mapping.h
+++ b/include/asm-x86/dma-mapping.h
@@ -17,7 +17,8 @@ extern int panic_on_overflow;
 extern int force_iommu;
 
 struct dma_mapping_ops {
-	int             (*mapping_error)(dma_addr_t dma_addr);
+	int             (*mapping_error)(struct device *dev,
+					 dma_addr_t dma_addr);
 	void*           (*alloc_coherent)(struct device *dev, size_t size,
 				dma_addr_t *dma_handle, gfp_t gfp);
 	void            (*free_coherent)(struct device *dev, size_t size,
@@ -56,14 +57,32 @@ struct dma_mapping_ops {
 	int		is_phys;
 };
 
-extern const struct dma_mapping_ops *dma_ops;
+extern struct dma_mapping_ops *dma_ops;
 
-static inline int dma_mapping_error(dma_addr_t dma_addr)
+static inline struct dma_mapping_ops *get_dma_ops(struct device *dev)
 {
-	if (dma_ops->mapping_error)
-		return dma_ops->mapping_error(dma_addr);
+#ifdef CONFIG_X86_32
+	return dma_ops;
+#else
+	if (unlikely(!dev) || !dev->archdata.dma_ops)
+		return dma_ops;
+	else
+		return dev->archdata.dma_ops;
+#endif
+}
+
+/* Make sure we keep the same behaviour */
+static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
+{
+#ifdef CONFIG_X86_32
+	return 0;
+#else
+	struct dma_mapping_ops *ops = get_dma_ops(dev);
+	if (ops->mapping_error)
+		return ops->mapping_error(dev, dma_addr);
 
 	return (dma_addr == bad_dma_address);
+#endif
 }
 
 #define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
@@ -83,44 +102,53 @@ static inline dma_addr_t
 dma_map_single(struct device *hwdev, void *ptr, size_t size,
 	       int direction)
 {
+	struct dma_mapping_ops *ops = get_dma_ops(hwdev);
+
 	BUG_ON(!valid_dma_direction(direction));
-	return dma_ops->map_single(hwdev, virt_to_phys(ptr), size, direction);
+	return ops->map_single(hwdev, virt_to_phys(ptr), size, direction);
 }
 
 static inline void
 dma_unmap_single(struct device *dev, dma_addr_t addr, size_t size,
 		 int direction)
 {
+	struct dma_mapping_ops *ops = get_dma_ops(dev);
+
 	BUG_ON(!valid_dma_direction(direction));
-	if (dma_ops->unmap_single)
-		dma_ops->unmap_single(dev, addr, size, direction);
+	if (ops->unmap_single)
+		ops->unmap_single(dev, addr, size, direction);
 }
 
 static inline int
 dma_map_sg(struct device *hwdev, struct scatterlist *sg,
 	   int nents, int direction)
 {
+	struct dma_mapping_ops *ops = get_dma_ops(hwdev);
+
 	BUG_ON(!valid_dma_direction(direction));
-	return dma_ops->map_sg(hwdev, sg, nents, direction);
+	return ops->map_sg(hwdev, sg, nents, direction);
 }
 
 static inline void
 dma_unmap_sg(struct device *hwdev, struct scatterlist *sg, int nents,
 	     int direction)
 {
+	struct dma_mapping_ops *ops = get_dma_ops(hwdev);
+
 	BUG_ON(!valid_dma_direction(direction));
-	if (dma_ops->unmap_sg)
-		dma_ops->unmap_sg(hwdev, sg, nents, direction);
+	if (ops->unmap_sg)
+		ops->unmap_sg(hwdev, sg, nents, direction);
 }
 
 static inline void
 dma_sync_single_for_cpu(struct device *hwdev, dma_addr_t dma_handle,
 			size_t size, int direction)
 {
+	struct dma_mapping_ops *ops = get_dma_ops(hwdev);
+
 	BUG_ON(!valid_dma_direction(direction));
-	if (dma_ops->sync_single_for_cpu)
-		dma_ops->sync_single_for_cpu(hwdev, dma_handle, size,
-					     direction);
+	if (ops->sync_single_for_cpu)
+		ops->sync_single_for_cpu(hwdev, dma_handle, size, direction);
 	flush_write_buffers();
 }
 
@@ -128,10 +156,11 @@ static inline void
 dma_sync_single_for_device(struct device *hwdev, dma_addr_t dma_handle,
 			   size_t size, int direction)
 {
+	struct dma_mapping_ops *ops = get_dma_ops(hwdev);
+
 	BUG_ON(!valid_dma_direction(direction));
-	if (dma_ops->sync_single_for_device)
-		dma_ops->sync_single_for_device(hwdev, dma_handle, size,
-						direction);
+	if (ops->sync_single_for_device)
+		ops->sync_single_for_device(hwdev, dma_handle, size, direction);
 	flush_write_buffers();
 }
 
@@ -139,11 +168,12 @@ static inline void
 dma_sync_single_range_for_cpu(struct device *hwdev, dma_addr_t dma_handle,
 			      unsigned long offset, size_t size, int direction)
 {
-	BUG_ON(!valid_dma_direction(direction));
-	if (dma_ops->sync_single_range_for_cpu)
-		dma_ops->sync_single_range_for_cpu(hwdev, dma_handle, offset,
-						   size, direction);
+	struct dma_mapping_ops *ops = get_dma_ops(hwdev);
 
+	BUG_ON(!valid_dma_direction(direction));
+	if (ops->sync_single_range_for_cpu)
+		ops->sync_single_range_for_cpu(hwdev, dma_handle, offset,
+					       size, direction);
 	flush_write_buffers();
 }
 
@@ -152,11 +182,12 @@ dma_sync_single_range_for_device(struct device *hwdev, dma_addr_t dma_handle,
 				 unsigned long offset, size_t size,
 				 int direction)
 {
-	BUG_ON(!valid_dma_direction(direction));
-	if (dma_ops->sync_single_range_for_device)
-		dma_ops->sync_single_range_for_device(hwdev, dma_handle,
-						      offset, size, direction);
+	struct dma_mapping_ops *ops = get_dma_ops(hwdev);
 
+	BUG_ON(!valid_dma_direction(direction));
+	if (ops->sync_single_range_for_device)
+		ops->sync_single_range_for_device(hwdev, dma_handle,
+						  offset, size, direction);
 	flush_write_buffers();
 }
 
@@ -164,9 +195,11 @@ static inline void
 dma_sync_sg_for_cpu(struct device *hwdev, struct scatterlist *sg,
 		    int nelems, int direction)
 {
+	struct dma_mapping_ops *ops = get_dma_ops(hwdev);
+
 	BUG_ON(!valid_dma_direction(direction));
-	if (dma_ops->sync_sg_for_cpu)
-		dma_ops->sync_sg_for_cpu(hwdev, sg, nelems, direction);
+	if (ops->sync_sg_for_cpu)
+		ops->sync_sg_for_cpu(hwdev, sg, nelems, direction);
 	flush_write_buffers();
 }
 
@@ -174,9 +207,11 @@ static inline void
 dma_sync_sg_for_device(struct device *hwdev, struct scatterlist *sg,
 		       int nelems, int direction)
 {
+	struct dma_mapping_ops *ops = get_dma_ops(hwdev);
+
 	BUG_ON(!valid_dma_direction(direction));
-	if (dma_ops->sync_sg_for_device)
-		dma_ops->sync_sg_for_device(hwdev, sg, nelems, direction);
+	if (ops->sync_sg_for_device)
+		ops->sync_sg_for_device(hwdev, sg, nelems, direction);
 
 	flush_write_buffers();
 }
@@ -185,9 +220,11 @@ static inline dma_addr_t dma_map_page(struct device *dev, struct page *page,
 				      size_t offset, size_t size,
 				      int direction)
 {
+	struct dma_mapping_ops *ops = get_dma_ops(dev);
+
 	BUG_ON(!valid_dma_direction(direction));
-	return dma_ops->map_single(dev, page_to_phys(page)+offset,
-				   size, direction);
+	return ops->map_single(dev, page_to_phys(page) + offset,
+			       size, direction);
 }
 
 static inline void dma_unmap_page(struct device *dev, dma_addr_t addr,
diff --git a/include/asm-x86/swiotlb.h b/include/asm-x86/swiotlb.h
index c706a7442633..2730b351afcf 100644
--- a/include/asm-x86/swiotlb.h
+++ b/include/asm-x86/swiotlb.h
@@ -35,7 +35,7 @@ extern int swiotlb_map_sg(struct device *hwdev, struct scatterlist *sg,
 			  int nents, int direction);
 extern void swiotlb_unmap_sg(struct device *hwdev, struct scatterlist *sg,
 			     int nents, int direction);
-extern int swiotlb_dma_mapping_error(dma_addr_t dma_addr);
+extern int swiotlb_dma_mapping_error(struct device *hwdev, dma_addr_t dma_addr);
 extern void swiotlb_free_coherent(struct device *hwdev, size_t size,
 				  void *vaddr, dma_addr_t dma_handle);
 extern int swiotlb_dma_supported(struct device *hwdev, u64 mask);
diff --git a/include/asm-xtensa/dma-mapping.h b/include/asm-xtensa/dma-mapping.h
index 3c7d537dd15d..51882ae3db4d 100644
--- a/include/asm-xtensa/dma-mapping.h
+++ b/include/asm-xtensa/dma-mapping.h
@@ -139,7 +139,7 @@ dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nelems,
 		consistent_sync(sg_virt(sg), sg->length, dir);
 }
 static inline int
-dma_mapping_error(dma_addr_t dma_addr)
+dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
 {
 	return 0;
 }
diff --git a/include/linux/i2o.h b/include/linux/i2o.h
index 7d51cbca49ab..75ae6d8aba4f 100644
--- a/include/linux/i2o.h
+++ b/include/linux/i2o.h
@@ -758,7 +758,7 @@ static inline dma_addr_t i2o_dma_map_single(struct i2o_controller *c, void *ptr,
 	}
 
 	dma_addr = dma_map_single(&c->pdev->dev, ptr, size, direction);
-	if (!dma_mapping_error(dma_addr)) {
+	if (!dma_mapping_error(&c->pdev->dev, dma_addr)) {
 #ifdef CONFIG_I2O_EXT_ADAPTEC_DMA64
 		if ((sizeof(dma_addr_t) > 4) && c->pae_support) {
 			*mptr++ = cpu_to_le32(0x7C020002);
diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h
index 4bf8cade9dbc..e530026eedf7 100644
--- a/include/linux/ssb/ssb.h
+++ b/include/linux/ssb/ssb.h
@@ -427,9 +427,9 @@ static inline int ssb_dma_mapping_error(struct ssb_device *dev, dma_addr_t addr)
 {
 	switch (dev->bus->bustype) {
 	case SSB_BUSTYPE_PCI:
-		return pci_dma_mapping_error(addr);
+		return pci_dma_mapping_error(dev->bus->host_pci, addr);
 	case SSB_BUSTYPE_SSB:
-		return dma_mapping_error(addr);
+		return dma_mapping_error(dev->dev, addr);
 	default:
 		__ssb_dma_not_implemented(dev);
 	}
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
index 90b529f7a154..936e333e7ce5 100644
--- a/include/rdma/ib_verbs.h
+++ b/include/rdma/ib_verbs.h
@@ -1590,7 +1590,7 @@ static inline int ib_dma_mapping_error(struct ib_device *dev, u64 dma_addr)
 {
 	if (dev->dma_ops)
 		return dev->dma_ops->mapping_error(dev, dma_addr);
-	return dma_mapping_error(dma_addr);
+	return dma_mapping_error(dev->dma_device, dma_addr);
 }
 
 /**
diff --git a/lib/swiotlb.c b/lib/swiotlb.c
index d568894df8cc..977edbdbc1de 100644
--- a/lib/swiotlb.c
+++ b/lib/swiotlb.c
@@ -492,7 +492,7 @@ swiotlb_alloc_coherent(struct device *hwdev, size_t size,
 		 */
 		dma_addr_t handle;
 		handle = swiotlb_map_single(NULL, NULL, size, DMA_FROM_DEVICE);
-		if (swiotlb_dma_mapping_error(handle))
+		if (swiotlb_dma_mapping_error(hwdev, handle))
 			return NULL;
 
 		ret = bus_to_virt(handle);
@@ -824,7 +824,7 @@ swiotlb_sync_sg_for_device(struct device *hwdev, struct scatterlist *sg,
 }
 
 int
-swiotlb_dma_mapping_error(dma_addr_t dma_addr)
+swiotlb_dma_mapping_error(struct device *hwdev, dma_addr_t dma_addr)
 {
 	return (dma_addr == virt_to_bus(io_tlb_overflow_buffer));
 }
diff --git a/net/sunrpc/xprtrdma/svc_rdma_sendto.c b/net/sunrpc/xprtrdma/svc_rdma_sendto.c
index a19b22b452a3..84d328329d98 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_sendto.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_sendto.c
@@ -169,7 +169,8 @@ static int send_write(struct svcxprt_rdma *xprt, struct svc_rqst *rqstp,
 					  (void *)
 					  vec->sge[xdr_sge_no].iov_base + sge_off,
 					  sge_bytes, DMA_TO_DEVICE);
-		if (dma_mapping_error(sge[sge_no].addr))
+		if (dma_mapping_error(xprt->sc_cm_id->device->dma_device,
+					sge[sge_no].addr))
 			goto err;
 		sge_off = 0;
 		sge_no++;

From 1956a96de488feb05e95c08c9d5e80f63a4be2b1 Mon Sep 17 00:00:00 2001
From: Alexis Bruemmer <alexisb@us.ibm.com>
Date: Fri, 25 Jul 2008 19:44:51 -0700
Subject: [PATCH 0993/1171] x86 calgary: fix handling of devices that aren't
 behind the Calgary

The calgary code can give drivers addresses above 4GB which is very bad
for hardware that is only 32bit DMA addressable.

With this patch, the calgary code sets the global dma_ops to swiotlb or
nommu properly, and the dma_ops of devices behind the Calgary/CalIOC2
to calgary_dma_ops.  So the calgary code can handle devices safely that
aren't behind the Calgary/CalIOC2.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Alexis Bruemmer <alexisb@us.ibm.com>
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Cc: Muli Ben-Yehuda <muli@il.ibm.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/kernel/pci-calgary_64.c | 71 ++++++++++++--------------------
 include/asm-x86/iommu.h          |  1 +
 2 files changed, 27 insertions(+), 45 deletions(-)

diff --git a/arch/x86/kernel/pci-calgary_64.c b/arch/x86/kernel/pci-calgary_64.c
index 1eb86be93d7a..b67a4b1d4eae 100644
--- a/arch/x86/kernel/pci-calgary_64.c
+++ b/arch/x86/kernel/pci-calgary_64.c
@@ -37,6 +37,7 @@
 #include <linux/delay.h>
 #include <linux/scatterlist.h>
 #include <linux/iommu-helper.h>
+
 #include <asm/iommu.h>
 #include <asm/calgary.h>
 #include <asm/tce.h>
@@ -413,22 +414,6 @@ static void calgary_unmap_sg(struct device *dev,
 	}
 }
 
-static int calgary_nontranslate_map_sg(struct device* dev,
-	struct scatterlist *sg, int nelems, int direction)
-{
-	struct scatterlist *s;
-	int i;
-
-	for_each_sg(sg, s, nelems, i) {
-		struct page *p = sg_page(s);
-
-		BUG_ON(!p);
-		s->dma_address = virt_to_bus(sg_virt(s));
-		s->dma_length = s->length;
-	}
-	return nelems;
-}
-
 static int calgary_map_sg(struct device *dev, struct scatterlist *sg,
 	int nelems, int direction)
 {
@@ -439,9 +424,6 @@ static int calgary_map_sg(struct device *dev, struct scatterlist *sg,
 	unsigned long entry;
 	int i;
 
-	if (!translation_enabled(tbl))
-		return calgary_nontranslate_map_sg(dev, sg, nelems, direction);
-
 	for_each_sg(sg, s, nelems, i) {
 		BUG_ON(!sg_page(s));
 
@@ -477,7 +459,6 @@ error:
 static dma_addr_t calgary_map_single(struct device *dev, phys_addr_t paddr,
 	size_t size, int direction)
 {
-	dma_addr_t dma_handle = bad_dma_address;
 	void *vaddr = phys_to_virt(paddr);
 	unsigned long uaddr;
 	unsigned int npages;
@@ -486,12 +467,7 @@ static dma_addr_t calgary_map_single(struct device *dev, phys_addr_t paddr,
 	uaddr = (unsigned long)vaddr;
 	npages = num_dma_pages(uaddr, size);
 
-	if (translation_enabled(tbl))
-		dma_handle = iommu_alloc(dev, tbl, vaddr, npages, direction);
-	else
-		dma_handle = virt_to_bus(vaddr);
-
-	return dma_handle;
+	return iommu_alloc(dev, tbl, vaddr, npages, direction);
 }
 
 static void calgary_unmap_single(struct device *dev, dma_addr_t dma_handle,
@@ -500,9 +476,6 @@ static void calgary_unmap_single(struct device *dev, dma_addr_t dma_handle,
 	struct iommu_table *tbl = find_iommu_table(dev);
 	unsigned int npages;
 
-	if (!translation_enabled(tbl))
-		return;
-
 	npages = num_dma_pages(dma_handle, size);
 	iommu_free(tbl, dma_handle, npages);
 }
@@ -525,18 +498,12 @@ static void* calgary_alloc_coherent(struct device *dev, size_t size,
 		goto error;
 	memset(ret, 0, size);
 
-	if (translation_enabled(tbl)) {
-		/* set up tces to cover the allocated range */
-		mapping = iommu_alloc(dev, tbl, ret, npages, DMA_BIDIRECTIONAL);
-		if (mapping == bad_dma_address)
-			goto free;
-
-		*dma_handle = mapping;
-	} else /* non translated slot */
-		*dma_handle = virt_to_bus(ret);
-
+	/* set up tces to cover the allocated range */
+	mapping = iommu_alloc(dev, tbl, ret, npages, DMA_BIDIRECTIONAL);
+	if (mapping == bad_dma_address)
+		goto free;
+	*dma_handle = mapping;
 	return ret;
-
 free:
 	free_pages((unsigned long)ret, get_order(size));
 	ret = NULL;
@@ -1241,6 +1208,16 @@ static int __init calgary_init(void)
 			goto error;
 	} while (1);
 
+	dev = NULL;
+	for_each_pci_dev(dev) {
+		struct iommu_table *tbl;
+
+		tbl = find_iommu_table(&dev->dev);
+
+		if (translation_enabled(tbl))
+			dev->dev.archdata.dma_ops = &calgary_dma_ops;
+	}
+
 	return ret;
 
 error:
@@ -1262,6 +1239,7 @@ error:
 		calgary_disable_translation(dev);
 		calgary_free_bus(dev);
 		pci_dev_put(dev); /* Undo calgary_init_one()'s pci_dev_get() */
+		dev->dev.archdata.dma_ops = NULL;
 	} while (1);
 
 	return ret;
@@ -1503,6 +1481,10 @@ void __init detect_calgary(void)
 		printk(KERN_INFO "PCI-DMA: Calgary TCE table spec is %d, "
 		       "CONFIG_IOMMU_DEBUG is %s.\n", specified_table_size,
 		       debugging ? "enabled" : "disabled");
+
+		/* swiotlb for devices that aren't behind the Calgary. */
+		if (max_pfn > MAX_DMA32_PFN)
+			swiotlb = 1;
 	}
 	return;
 
@@ -1519,7 +1501,7 @@ int __init calgary_iommu_init(void)
 {
 	int ret;
 
-	if (no_iommu || swiotlb)
+	if (no_iommu || (swiotlb && !calgary_detected))
 		return -ENODEV;
 
 	if (!calgary_detected)
@@ -1532,15 +1514,14 @@ int __init calgary_iommu_init(void)
 	if (ret) {
 		printk(KERN_ERR "PCI-DMA: Calgary init failed %d, "
 		       "falling back to no_iommu\n", ret);
-		if (max_pfn > MAX_DMA32_PFN)
-			printk(KERN_ERR "WARNING more than 4GB of memory, "
-					"32bit PCI may malfunction.\n");
 		return ret;
 	}
 
 	force_iommu = 1;
 	bad_dma_address = 0x0;
-	dma_ops = &calgary_dma_ops;
+	/* dma_ops is set to swiotlb or nommu */
+	if (!dma_ops)
+		dma_ops = &nommu_dma_ops;
 
 	return 0;
 }
diff --git a/include/asm-x86/iommu.h b/include/asm-x86/iommu.h
index d63166fb3ab7..ecc8061904a9 100644
--- a/include/asm-x86/iommu.h
+++ b/include/asm-x86/iommu.h
@@ -3,6 +3,7 @@
 
 extern void pci_iommu_shutdown(void);
 extern void no_iommu_init(void);
+extern struct dma_mapping_ops nommu_dma_ops;
 extern int force_iommu, no_iommu;
 extern int iommu_detected;
 

From 7d135a5d50a08bbc53b189d79d8bdb03136f5303 Mon Sep 17 00:00:00 2001
From: Matthias Kaehlcke <matthias@kaehlcke.net>
Date: Fri, 25 Jul 2008 19:44:51 -0700
Subject: [PATCH 0994/1171] affs: convert s_bmlock into a mutex

The semaphore s_bmlock is used as a mutex.  Convert it to the mutex API.

Signed-off-by: Matthias Kaehlcke <matthias@kaehlcke.net>
Cc: Roman Zippel <zippel@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/affs/affs.h   |  3 ++-
 fs/affs/bitmap.c | 18 +++++++++---------
 fs/affs/super.c  |  2 +-
 3 files changed, 12 insertions(+), 11 deletions(-)

diff --git a/fs/affs/affs.h b/fs/affs/affs.h
index 223b1917093e..e9ec915f7553 100644
--- a/fs/affs/affs.h
+++ b/fs/affs/affs.h
@@ -2,6 +2,7 @@
 #include <linux/fs.h>
 #include <linux/buffer_head.h>
 #include <linux/amigaffs.h>
+#include <linux/mutex.h>
 
 /* AmigaOS allows file names with up to 30 characters length.
  * Names longer than that will be silently truncated. If you
@@ -98,7 +99,7 @@ struct affs_sb_info {
 	gid_t s_gid;			/* gid to override */
 	umode_t s_mode;			/* mode to override */
 	struct buffer_head *s_root_bh;	/* Cached root block. */
-	struct semaphore s_bmlock;	/* Protects bitmap access. */
+	struct mutex s_bmlock;		/* Protects bitmap access. */
 	struct affs_bm_info *s_bitmap;	/* Bitmap infos. */
 	u32 s_bmap_count;		/* # of bitmap blocks. */
 	u32 s_bmap_bits;		/* # of bits in one bitmap blocks */
diff --git a/fs/affs/bitmap.c b/fs/affs/bitmap.c
index c4a5ad09ddf2..dc5ef14bdc1c 100644
--- a/fs/affs/bitmap.c
+++ b/fs/affs/bitmap.c
@@ -45,14 +45,14 @@ affs_count_free_blocks(struct super_block *sb)
 	if (sb->s_flags & MS_RDONLY)
 		return 0;
 
-	down(&AFFS_SB(sb)->s_bmlock);
+	mutex_lock(&AFFS_SB(sb)->s_bmlock);
 
 	bm = AFFS_SB(sb)->s_bitmap;
 	free = 0;
 	for (i = AFFS_SB(sb)->s_bmap_count; i > 0; bm++, i--)
 		free += bm->bm_free;
 
-	up(&AFFS_SB(sb)->s_bmlock);
+	mutex_unlock(&AFFS_SB(sb)->s_bmlock);
 
 	return free;
 }
@@ -76,7 +76,7 @@ affs_free_block(struct super_block *sb, u32 block)
 	bit     = blk % sbi->s_bmap_bits;
 	bm      = &sbi->s_bitmap[bmap];
 
-	down(&sbi->s_bmlock);
+	mutex_lock(&sbi->s_bmlock);
 
 	bh = sbi->s_bmap_bh;
 	if (sbi->s_last_bmap != bmap) {
@@ -105,19 +105,19 @@ affs_free_block(struct super_block *sb, u32 block)
 	sb->s_dirt = 1;
 	bm->bm_free++;
 
-	up(&sbi->s_bmlock);
+	mutex_unlock(&sbi->s_bmlock);
 	return;
 
 err_free:
 	affs_warning(sb,"affs_free_block","Trying to free block %u which is already free", block);
-	up(&sbi->s_bmlock);
+	mutex_unlock(&sbi->s_bmlock);
 	return;
 
 err_bh_read:
 	affs_error(sb,"affs_free_block","Cannot read bitmap block %u", bm->bm_key);
 	sbi->s_bmap_bh = NULL;
 	sbi->s_last_bmap = ~0;
-	up(&sbi->s_bmlock);
+	mutex_unlock(&sbi->s_bmlock);
 	return;
 
 err_range:
@@ -168,7 +168,7 @@ affs_alloc_block(struct inode *inode, u32 goal)
 	bmap = blk / sbi->s_bmap_bits;
 	bm = &sbi->s_bitmap[bmap];
 
-	down(&sbi->s_bmlock);
+	mutex_lock(&sbi->s_bmlock);
 
 	if (bm->bm_free)
 		goto find_bmap_bit;
@@ -249,7 +249,7 @@ find_bit:
 	mark_buffer_dirty(bh);
 	sb->s_dirt = 1;
 
-	up(&sbi->s_bmlock);
+	mutex_unlock(&sbi->s_bmlock);
 
 	pr_debug("%d\n", blk);
 	return blk;
@@ -259,7 +259,7 @@ err_bh_read:
 	sbi->s_bmap_bh = NULL;
 	sbi->s_last_bmap = ~0;
 err_full:
-	up(&sbi->s_bmlock);
+	mutex_unlock(&sbi->s_bmlock);
 	pr_debug("failed\n");
 	return 0;
 }
diff --git a/fs/affs/super.c b/fs/affs/super.c
index d214837d5e42..4e0309566406 100644
--- a/fs/affs/super.c
+++ b/fs/affs/super.c
@@ -290,7 +290,7 @@ static int affs_fill_super(struct super_block *sb, void *data, int silent)
 	if (!sbi)
 		return -ENOMEM;
 	sb->s_fs_info = sbi;
-	init_MUTEX(&sbi->s_bmlock);
+	mutex_init(&sbi->s_bmlock);
 
 	if (!parse_options(data,&uid,&gid,&i,&reserved,&root_block,
 				&blocksize,&sbi->s_prefix,

From 75b25b4cabb7ce956c36442bf8225659b1864866 Mon Sep 17 00:00:00 2001
From: Dmitri Vorobiev <dmitri.vorobiev@movial.fi>
Date: Fri, 25 Jul 2008 19:44:52 -0700
Subject: [PATCH 0995/1171] bfs: assorted cleanups

This patch makes the following cleanups:

	o removing an unused variable from bfs_fill_super();
	o removing unneeded blank spaces from pointer
	  definitions.

Signed-off-by: Dmitri Vorobiev <dmitri.vorobiev@movial.fi>
Cc: Tigran Aivazian <tigran_aivazian@symantec.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/bfs/bfs.h   | 4 ++--
 fs/bfs/inode.c | 3 +--
 2 files changed, 3 insertions(+), 4 deletions(-)

diff --git a/fs/bfs/bfs.h b/fs/bfs/bfs.h
index 70f5d3a8eede..d1a6cd10afce 100644
--- a/fs/bfs/bfs.h
+++ b/fs/bfs/bfs.h
@@ -16,8 +16,8 @@ struct bfs_sb_info {
 	unsigned long si_freei;
 	unsigned long si_lf_eblk;
 	unsigned long si_lasti;
-	unsigned long * si_imap;
-	struct buffer_head * si_sbh;		/* buffer header w/superblock */
+	unsigned long *si_imap;
+	struct buffer_head *si_sbh;		/* buffer header w/superblock */
 };
 
 /*
diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c
index 8db623838b50..c8b3982e11ca 100644
--- a/fs/bfs/inode.c
+++ b/fs/bfs/inode.c
@@ -380,7 +380,7 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent)
 		struct bfs_inode *di;
 		int block = (i - BFS_ROOT_INO) / BFS_INODES_PER_BLOCK + 1;
 		int off = (i - BFS_ROOT_INO) % BFS_INODES_PER_BLOCK;
-		unsigned long sblock, eblock;
+		unsigned long eblock;
 
 		if (!off) {
 			brelse(bh);
@@ -399,7 +399,6 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent)
 		set_bit(i, info->si_imap);
 		info->si_freeb -= BFS_FILEBLOCKS(di);
 
-		sblock =  le32_to_cpu(di->i_sblock);
 		eblock =  le32_to_cpu(di->i_eblock);
 		if (eblock > info->si_lf_eblk)
 			info->si_lf_eblk = eblock;

From 3f165e4cf2af042af7d2440d688299c0d2a48b1f Mon Sep 17 00:00:00 2001
From: Dmitri Vorobiev <dmitri.vorobiev@movial.fi>
Date: Fri, 25 Jul 2008 19:44:54 -0700
Subject: [PATCH 0996/1171] bfs: kill BKL

Replace the BKL-based locking scheme used in the bfs driver by a private
filesystem-wide mutex.

Signed-off-by: Dmitri Vorobiev <dmitri.vorobiev@movial.fi>
Cc: Tigran Aivazian <tigran_aivazian@symantec.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/bfs/bfs.h   |  1 +
 fs/bfs/dir.c   | 46 ++++++++++++++++++++++++++--------------------
 fs/bfs/file.c  |  4 ++--
 fs/bfs/inode.c | 24 +++++++++++++++---------
 4 files changed, 44 insertions(+), 31 deletions(-)

diff --git a/fs/bfs/bfs.h b/fs/bfs/bfs.h
index d1a6cd10afce..7109e451abf7 100644
--- a/fs/bfs/bfs.h
+++ b/fs/bfs/bfs.h
@@ -18,6 +18,7 @@ struct bfs_sb_info {
 	unsigned long si_lasti;
 	unsigned long *si_imap;
 	struct buffer_head *si_sbh;		/* buffer header w/superblock */
+	struct mutex bfs_lock;
 };
 
 /*
diff --git a/fs/bfs/dir.c b/fs/bfs/dir.c
index 034950cb3cbe..87ee5ccee348 100644
--- a/fs/bfs/dir.c
+++ b/fs/bfs/dir.c
@@ -32,16 +32,17 @@ static int bfs_readdir(struct file *f, void *dirent, filldir_t filldir)
 	struct inode *dir = f->f_path.dentry->d_inode;
 	struct buffer_head *bh;
 	struct bfs_dirent *de;
+	struct bfs_sb_info *info = BFS_SB(dir->i_sb);
 	unsigned int offset;
 	int block;
 
-	lock_kernel();
+	mutex_lock(&info->bfs_lock);
 
 	if (f->f_pos & (BFS_DIRENT_SIZE - 1)) {
 		printf("Bad f_pos=%08lx for %s:%08lx\n",
 					(unsigned long)f->f_pos,
 					dir->i_sb->s_id, dir->i_ino);
-		unlock_kernel();
+		mutex_unlock(&info->bfs_lock);
 		return -EBADF;
 	}
 
@@ -61,7 +62,7 @@ static int bfs_readdir(struct file *f, void *dirent, filldir_t filldir)
 						le16_to_cpu(de->ino),
 						DT_UNKNOWN) < 0) {
 					brelse(bh);
-					unlock_kernel();
+					mutex_unlock(&info->bfs_lock);
 					return 0;
 				}
 			}
@@ -71,7 +72,7 @@ static int bfs_readdir(struct file *f, void *dirent, filldir_t filldir)
 		brelse(bh);
 	}
 
-	unlock_kernel();
+	mutex_unlock(&info->bfs_lock);
 	return 0;	
 }
 
@@ -95,10 +96,10 @@ static int bfs_create(struct inode *dir, struct dentry *dentry, int mode,
 	inode = new_inode(s);
 	if (!inode)
 		return -ENOSPC;
-	lock_kernel();
+	mutex_lock(&info->bfs_lock);
 	ino = find_first_zero_bit(info->si_imap, info->si_lasti);
 	if (ino > info->si_lasti) {
-		unlock_kernel();
+		mutex_unlock(&info->bfs_lock);
 		iput(inode);
 		return -ENOSPC;
 	}
@@ -125,10 +126,10 @@ static int bfs_create(struct inode *dir, struct dentry *dentry, int mode,
 	if (err) {
 		inode_dec_link_count(inode);
 		iput(inode);
-		unlock_kernel();
+		mutex_unlock(&info->bfs_lock);
 		return err;
 	}
-	unlock_kernel();
+	mutex_unlock(&info->bfs_lock);
 	d_instantiate(dentry, inode);
 	return 0;
 }
@@ -139,22 +140,23 @@ static struct dentry *bfs_lookup(struct inode *dir, struct dentry *dentry,
 	struct inode *inode = NULL;
 	struct buffer_head *bh;
 	struct bfs_dirent *de;
+	struct bfs_sb_info *info = BFS_SB(dir->i_sb);
 
 	if (dentry->d_name.len > BFS_NAMELEN)
 		return ERR_PTR(-ENAMETOOLONG);
 
-	lock_kernel();
+	mutex_lock(&info->bfs_lock);
 	bh = bfs_find_entry(dir, dentry->d_name.name, dentry->d_name.len, &de);
 	if (bh) {
 		unsigned long ino = (unsigned long)le16_to_cpu(de->ino);
 		brelse(bh);
 		inode = bfs_iget(dir->i_sb, ino);
 		if (IS_ERR(inode)) {
-			unlock_kernel();
+			mutex_unlock(&info->bfs_lock);
 			return ERR_CAST(inode);
 		}
 	}
-	unlock_kernel();
+	mutex_unlock(&info->bfs_lock);
 	d_add(dentry, inode);
 	return NULL;
 }
@@ -163,13 +165,14 @@ static int bfs_link(struct dentry *old, struct inode *dir,
 						struct dentry *new)
 {
 	struct inode *inode = old->d_inode;
+	struct bfs_sb_info *info = BFS_SB(inode->i_sb);
 	int err;
 
-	lock_kernel();
+	mutex_lock(&info->bfs_lock);
 	err = bfs_add_entry(dir, new->d_name.name, new->d_name.len,
 							inode->i_ino);
 	if (err) {
-		unlock_kernel();
+		mutex_unlock(&info->bfs_lock);
 		return err;
 	}
 	inc_nlink(inode);
@@ -177,19 +180,19 @@ static int bfs_link(struct dentry *old, struct inode *dir,
 	mark_inode_dirty(inode);
 	atomic_inc(&inode->i_count);
 	d_instantiate(new, inode);
-	unlock_kernel();
+	mutex_unlock(&info->bfs_lock);
 	return 0;
 }
 
 static int bfs_unlink(struct inode *dir, struct dentry *dentry)
 {
 	int error = -ENOENT;
-	struct inode *inode;
+	struct inode *inode = dentry->d_inode;
 	struct buffer_head *bh;
 	struct bfs_dirent *de;
+	struct bfs_sb_info *info = BFS_SB(inode->i_sb);
 
-	inode = dentry->d_inode;
-	lock_kernel();
+	mutex_lock(&info->bfs_lock);
 	bh = bfs_find_entry(dir, dentry->d_name.name, dentry->d_name.len, &de);
 	if (!bh || (le16_to_cpu(de->ino) != inode->i_ino))
 		goto out_brelse;
@@ -210,7 +213,7 @@ static int bfs_unlink(struct inode *dir, struct dentry *dentry)
 
 out_brelse:
 	brelse(bh);
-	unlock_kernel();
+	mutex_unlock(&info->bfs_lock);
 	return error;
 }
 
@@ -220,6 +223,7 @@ static int bfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	struct inode *old_inode, *new_inode;
 	struct buffer_head *old_bh, *new_bh;
 	struct bfs_dirent *old_de, *new_de;
+	struct bfs_sb_info *info;
 	int error = -ENOENT;
 
 	old_bh = new_bh = NULL;
@@ -227,7 +231,9 @@ static int bfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	if (S_ISDIR(old_inode->i_mode))
 		return -EINVAL;
 
-	lock_kernel();
+	info = BFS_SB(old_inode->i_sb);
+
+	mutex_lock(&info->bfs_lock);
 	old_bh = bfs_find_entry(old_dir, 
 				old_dentry->d_name.name, 
 				old_dentry->d_name.len, &old_de);
@@ -264,7 +270,7 @@ static int bfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	error = 0;
 
 end_rename:
-	unlock_kernel();
+	mutex_unlock(&info->bfs_lock);
 	brelse(old_bh);
 	brelse(new_bh);
 	return error;
diff --git a/fs/bfs/file.c b/fs/bfs/file.c
index b11e63e8fbcd..6a021265f018 100644
--- a/fs/bfs/file.c
+++ b/fs/bfs/file.c
@@ -99,7 +99,7 @@ static int bfs_get_block(struct inode *inode, sector_t block,
 		return -ENOSPC;
 
 	/* The rest has to be protected against itself. */
-	lock_kernel();
+	mutex_lock(&info->bfs_lock);
 
 	/*
 	 * If the last data block for this file is the last allocated
@@ -151,7 +151,7 @@ static int bfs_get_block(struct inode *inode, sector_t block,
 	mark_buffer_dirty(sbh);
 	map_bh(bh_result, sb, phys);
 out:
-	unlock_kernel();
+	mutex_unlock(&info->bfs_lock);
 	return err;
 }
 
diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c
index c8b3982e11ca..053e690ec9ed 100644
--- a/fs/bfs/inode.c
+++ b/fs/bfs/inode.c
@@ -104,6 +104,7 @@ static int bfs_write_inode(struct inode *inode, int unused)
 	struct bfs_inode *di;
 	struct buffer_head *bh;
 	int block, off;
+	struct bfs_sb_info *info = BFS_SB(inode->i_sb);
 
         dprintf("ino=%08x\n", ino);
 
@@ -112,13 +113,13 @@ static int bfs_write_inode(struct inode *inode, int unused)
 		return -EIO;
 	}
 
-	lock_kernel();
+	mutex_lock(&info->bfs_lock);
 	block = (ino - BFS_ROOT_INO) / BFS_INODES_PER_BLOCK + 1;
 	bh = sb_bread(inode->i_sb, block);
 	if (!bh) {
 		printf("Unable to read inode %s:%08x\n",
 				inode->i_sb->s_id, ino);
-		unlock_kernel();
+		mutex_unlock(&info->bfs_lock);
 		return -EIO;
 	}
 
@@ -145,7 +146,7 @@ static int bfs_write_inode(struct inode *inode, int unused)
 
 	mark_buffer_dirty(bh);
 	brelse(bh);
-	unlock_kernel();
+	mutex_unlock(&info->bfs_lock);
 	return 0;
 }
 
@@ -170,7 +171,7 @@ static void bfs_delete_inode(struct inode *inode)
 	
 	inode->i_size = 0;
 	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME_SEC;
-	lock_kernel();
+	mutex_lock(&info->bfs_lock);
 	mark_inode_dirty(inode);
 
 	block = (ino - BFS_ROOT_INO) / BFS_INODES_PER_BLOCK + 1;
@@ -178,7 +179,7 @@ static void bfs_delete_inode(struct inode *inode)
 	if (!bh) {
 		printf("Unable to read inode %s:%08lx\n",
 					inode->i_sb->s_id, ino);
-		unlock_kernel();
+		mutex_unlock(&info->bfs_lock);
 		return;
 	}
 	off = (ino - BFS_ROOT_INO) % BFS_INODES_PER_BLOCK;
@@ -204,14 +205,16 @@ static void bfs_delete_inode(struct inode *inode)
 		info->si_lf_eblk = bi->i_sblock - 1;
 		mark_buffer_dirty(info->si_sbh);
 	}
-	unlock_kernel();
+	mutex_unlock(&info->bfs_lock);
 	clear_inode(inode);
 }
 
 static void bfs_put_super(struct super_block *s)
 {
 	struct bfs_sb_info *info = BFS_SB(s);
+
 	brelse(info->si_sbh);
+	mutex_destroy(&info->bfs_lock);
 	kfree(info->si_imap);
 	kfree(info);
 	s->s_fs_info = NULL;
@@ -236,11 +239,13 @@ static int bfs_statfs(struct dentry *dentry, struct kstatfs *buf)
 
 static void bfs_write_super(struct super_block *s)
 {
-	lock_kernel();
+	struct bfs_sb_info *info = BFS_SB(s);
+
+	mutex_lock(&info->bfs_lock);
 	if (!(s->s_flags & MS_RDONLY))
-		mark_buffer_dirty(BFS_SB(s)->si_sbh);
+		mark_buffer_dirty(info->si_sbh);
 	s->s_dirt = 0;
-	unlock_kernel();
+	mutex_unlock(&info->bfs_lock);
 }
 
 static struct kmem_cache *bfs_inode_cachep;
@@ -409,6 +414,7 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent)
 		s->s_dirt = 1;
 	} 
 	dump_imap("read_super", s);
+	mutex_init(&info->bfs_lock);
 	return 0;
 
 out:

From 929dfb24fbcd60e2544b2de7bfb4a68da4dfc747 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 19:44:54 -0700
Subject: [PATCH 0997/1171] parport/share.c: proper externs

This patch adds proper externs for parport_default_timeslice and
parport_default_spintime in include/linux/parport.h

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/parport/procfs.c | 3 ---
 include/linux/parport.h  | 3 +++
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/parport/procfs.c b/drivers/parport/procfs.c
index d950fc34320a..554e11f9e1ce 100644
--- a/drivers/parport/procfs.c
+++ b/drivers/parport/procfs.c
@@ -429,9 +429,6 @@ struct parport_default_sysctl_table
 	ctl_table dev_dir[2];
 };
 
-extern unsigned long parport_default_timeslice;
-extern int parport_default_spintime;
-
 static struct parport_default_sysctl_table
 parport_default_sysctl_table = {
 	.sysctl_header	= NULL,
diff --git a/include/linux/parport.h b/include/linux/parport.h
index dcb9e01a69ca..6a0d7cdb5774 100644
--- a/include/linux/parport.h
+++ b/include/linux/parport.h
@@ -560,5 +560,8 @@ extern int parport_device_proc_unregister(struct pardevice *device);
 
 #endif /*  !CONFIG_PARPORT_NOT_PC  */
 
+extern unsigned long parport_default_timeslice;
+extern int parport_default_spintime;
+
 #endif /* __KERNEL__ */
 #endif /* _PARPORT_H_ */

From d99a0344aefbfe991147472d46a6ee1c1a0043de Mon Sep 17 00:00:00 2001
From: Andre Haupt <andre@bitwigglers.org>
Date: Fri, 25 Jul 2008 19:44:55 -0700
Subject: [PATCH 0998/1171] parport: remove superfluous local variable

Signed-off-by: Andre Haupt <andre@bitwigglers.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/parport/ieee1284.c | 2 --
 1 file changed, 2 deletions(-)

diff --git a/drivers/parport/ieee1284.c b/drivers/parport/ieee1284.c
index 0338b0912674..e97059415ab4 100644
--- a/drivers/parport/ieee1284.c
+++ b/drivers/parport/ieee1284.c
@@ -199,8 +199,6 @@ int parport_wait_peripheral(struct parport *port,
 	/* 40ms of slow polling. */
 	deadline = jiffies + msecs_to_jiffies(40);
 	while (time_before (jiffies, deadline)) {
-		int ret;
-
 		if (signal_pending (current))
 			return -EINTR;
 

From adbd321a17ccdd26752b57e68ab0a97a4aebc299 Mon Sep 17 00:00:00 2001
From: Bernhard Walle <bwalle@suse.de>
Date: Fri, 25 Jul 2008 19:44:56 -0700
Subject: [PATCH 0999/1171] parport_pc: add base_hi BAR for oxsemi_840

Use the 2nd BAR for the oxsemi_840 chip as BAR for base_hi.  Tested with:

Parallel controller [0701]: Oxford Semiconductor Ltd VScom 011H-EP1
        1 port parallel adaptor [1415:8403] (prog-if 03 [IEEE1284])

This patch is needed to make 'TRISTATE' work with that adaptor.

Signed-off-by: Bernhard Walle <bwalle@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/parport/parport_pc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/parport/parport_pc.c b/drivers/parport/parport_pc.c
index e0c2a4584ec6..8a846adf1dcf 100644
--- a/drivers/parport/parport_pc.c
+++ b/drivers/parport/parport_pc.c
@@ -2867,7 +2867,7 @@ static struct parport_pc_pci {
 	 * and 840 locks up if you write 1 to bit 2! */
 	/* oxsemi_952 */		{ 1, { { 0, 1 }, } },
 	/* oxsemi_954 */		{ 1, { { 0, -1 }, } },
-	/* oxsemi_840 */		{ 1, { { 0, -1 }, } },
+	/* oxsemi_840 */		{ 1, { { 0, 1 }, } },
 	/* aks_0100 */                  { 1, { { 0, -1 }, } },
 	/* mobility_pp */		{ 1, { { 0, 1 }, } },
 	/* netmos_9705 */               { 1, { { 0, -1 }, } }, /* untested */

From 061991ec6edceda48d60f7a53e17b8d3416266ae Mon Sep 17 00:00:00 2001
From: LE DISEZ Erwan <eledisez@grounation.org>
Date: Fri, 25 Jul 2008 19:44:56 -0700
Subject: [PATCH 1000/1171] tpm: add support for Broadcom TPM TIS device HID

Signed-off-by: Rajiv Andrade <srajiv@linux.vnet.ibm.com>
Cc: Marcel Selhorst <tpm@selhorst.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/tpm/tpm_tis.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/char/tpm/tpm_tis.c b/drivers/char/tpm/tpm_tis.c
index c7a977bc03e8..ed1879c0dd8d 100644
--- a/drivers/char/tpm/tpm_tis.c
+++ b/drivers/char/tpm/tpm_tis.c
@@ -622,6 +622,7 @@ static struct pnp_device_id tpm_pnp_tbl[] __devinitdata = {
 	{"ATM1200", 0},		/* Atmel */
 	{"IFX0102", 0},		/* Infineon */
 	{"BCM0101", 0},		/* Broadcom */
+	{"BCM0102", 0},		/* Broadcom */
 	{"NSC1200", 0},		/* National */
 	{"ICO0102", 0},		/* Intel */
 	/* Add new here */

From 3bd60464e3224820bc413c45ea2cc371edc63e9d Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 19:44:58 -0700
Subject: [PATCH 1001/1171] tpm_bios.c: make 2 structs static

This patch makes two needlessly global structs static.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Acked-by: Marcel Selhorst <tpm@selhorst.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/tpm/tpm_bios.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/char/tpm/tpm_bios.c b/drivers/char/tpm/tpm_bios.c
index 60a2d2630e36..68f052b42ed7 100644
--- a/drivers/char/tpm/tpm_bios.c
+++ b/drivers/char/tpm/tpm_bios.c
@@ -448,7 +448,7 @@ out_free:
 	goto out;
 }
 
-const struct file_operations tpm_ascii_bios_measurements_ops = {
+static const struct file_operations tpm_ascii_bios_measurements_ops = {
 	.open = tpm_ascii_bios_measurements_open,
 	.read = seq_read,
 	.llseek = seq_lseek,
@@ -486,7 +486,7 @@ out_free:
 	goto out;
 }
 
-const struct file_operations tpm_binary_bios_measurements_ops = {
+static const struct file_operations tpm_binary_bios_measurements_ops = {
 	.open = tpm_binary_bios_measurements_open,
 	.read = seq_read,
 	.llseek = seq_lseek,

From ec288bd37e1925f513db40871bc46115cf7fb733 Mon Sep 17 00:00:00 2001
From: Marcin Obara <marcin_obara@users.sourceforge.net>
Date: Fri, 25 Jul 2008 19:44:59 -0700
Subject: [PATCH 1002/1171] tpm: increase size of internal TPM response buffers

This patch increases size of driver internal response buffers.  Some TPM
responses defined in TCG TPM Specification Version 1.2 Revision 103 have
increased size and do not fit previously defined buffers.  Some TPM
responses do not have fixed size, so bigger response buffers have to be
allocated.  200B buffers should be enough.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Marcin Obara <marcin_obara@users.sourceforge.net>
Cc: Marcel Selhorst <tpm@selhorst.net>
Cc: Kylene Jo Hall <kjhall@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/tpm/tpm.c | 124 ++++++++++++++++++++++++++++++-----------
 1 file changed, 93 insertions(+), 31 deletions(-)

diff --git a/drivers/char/tpm/tpm.c b/drivers/char/tpm/tpm.c
index e1fc193d9396..f354d720b777 100644
--- a/drivers/char/tpm/tpm.c
+++ b/drivers/char/tpm/tpm.c
@@ -580,91 +580,133 @@ void tpm_continue_selftest(struct tpm_chip *chip)
 }
 EXPORT_SYMBOL_GPL(tpm_continue_selftest);
 
+#define  TPM_INTERNAL_RESULT_SIZE 200
+
 ssize_t tpm_show_enabled(struct device * dev, struct device_attribute * attr,
 			char *buf)
 {
-	u8 data[max_t(int, ARRAY_SIZE(tpm_cap), 35)];
+	u8 *data;
 	ssize_t rc;
 
 	struct tpm_chip *chip = dev_get_drvdata(dev);
 	if (chip == NULL)
 		return -ENODEV;
 
+	data = kzalloc(TPM_INTERNAL_RESULT_SIZE, GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
+
 	memcpy(data, tpm_cap, sizeof(tpm_cap));
 	data[TPM_CAP_IDX] = TPM_CAP_FLAG;
 	data[TPM_CAP_SUBCAP_IDX] = TPM_CAP_FLAG_PERM;
 
-	rc = transmit_cmd(chip, data, sizeof(data),
-			"attemtping to determine the permanent state");
-	if (rc)
+	rc = transmit_cmd(chip, data, TPM_INTERNAL_RESULT_SIZE,
+			"attemtping to determine the permanent enabled state");
+	if (rc) {
+		kfree(data);
 		return 0;
-	return sprintf(buf, "%d\n", !data[TPM_GET_CAP_PERM_DISABLE_IDX]);
+	}
+
+	rc = sprintf(buf, "%d\n", !data[TPM_GET_CAP_PERM_DISABLE_IDX]);
+
+	kfree(data);
+	return rc;
 }
 EXPORT_SYMBOL_GPL(tpm_show_enabled);
 
 ssize_t tpm_show_active(struct device * dev, struct device_attribute * attr,
 			char *buf)
 {
-	u8 data[max_t(int, ARRAY_SIZE(tpm_cap), 35)];
+	u8 *data;
 	ssize_t rc;
 
 	struct tpm_chip *chip = dev_get_drvdata(dev);
 	if (chip == NULL)
 		return -ENODEV;
 
+	data = kzalloc(TPM_INTERNAL_RESULT_SIZE, GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
+
 	memcpy(data, tpm_cap, sizeof(tpm_cap));
 	data[TPM_CAP_IDX] = TPM_CAP_FLAG;
 	data[TPM_CAP_SUBCAP_IDX] = TPM_CAP_FLAG_PERM;
 
-	rc = transmit_cmd(chip, data, sizeof(data),
-			"attemtping to determine the permanent state");
-	if (rc)
+	rc = transmit_cmd(chip, data, TPM_INTERNAL_RESULT_SIZE,
+			"attemtping to determine the permanent active state");
+	if (rc) {
+		kfree(data);
 		return 0;
-	return sprintf(buf, "%d\n", !data[TPM_GET_CAP_PERM_INACTIVE_IDX]);
+	}
+
+	rc = sprintf(buf, "%d\n", !data[TPM_GET_CAP_PERM_INACTIVE_IDX]);
+
+	kfree(data);
+	return rc;
 }
 EXPORT_SYMBOL_GPL(tpm_show_active);
 
 ssize_t tpm_show_owned(struct device * dev, struct device_attribute * attr,
 			char *buf)
 {
-	u8 data[sizeof(tpm_cap)];
+	u8 *data;
 	ssize_t rc;
 
 	struct tpm_chip *chip = dev_get_drvdata(dev);
 	if (chip == NULL)
 		return -ENODEV;
 
+	data = kzalloc(TPM_INTERNAL_RESULT_SIZE, GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
+
 	memcpy(data, tpm_cap, sizeof(tpm_cap));
 	data[TPM_CAP_IDX] = TPM_CAP_PROP;
 	data[TPM_CAP_SUBCAP_IDX] = TPM_CAP_PROP_OWNER;
 
-	rc = transmit_cmd(chip, data, sizeof(data),
+	rc = transmit_cmd(chip, data, TPM_INTERNAL_RESULT_SIZE,
 			"attempting to determine the owner state");
-	if (rc)
+	if (rc) {
+		kfree(data);
 		return 0;
-	return sprintf(buf, "%d\n", data[TPM_GET_CAP_RET_BOOL_1_IDX]);
+	}
+
+	rc = sprintf(buf, "%d\n", data[TPM_GET_CAP_RET_BOOL_1_IDX]);
+
+	kfree(data);
+	return rc;
 }
 EXPORT_SYMBOL_GPL(tpm_show_owned);
 
 ssize_t tpm_show_temp_deactivated(struct device * dev,
 				struct device_attribute * attr, char *buf)
 {
-	u8 data[sizeof(tpm_cap)];
+	u8 *data;
 	ssize_t rc;
 
 	struct tpm_chip *chip = dev_get_drvdata(dev);
 	if (chip == NULL)
 		return -ENODEV;
 
+	data = kzalloc(TPM_INTERNAL_RESULT_SIZE, GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
+
 	memcpy(data, tpm_cap, sizeof(tpm_cap));
 	data[TPM_CAP_IDX] = TPM_CAP_FLAG;
 	data[TPM_CAP_SUBCAP_IDX] = TPM_CAP_FLAG_VOL;
 
-	rc = transmit_cmd(chip, data, sizeof(data),
+	rc = transmit_cmd(chip, data, TPM_INTERNAL_RESULT_SIZE,
 			"attempting to determine the temporary state");
-	if (rc)
+	if (rc) {
+		kfree(data);
 		return 0;
-	return sprintf(buf, "%d\n", data[TPM_GET_CAP_TEMP_INACTIVE_IDX]);
+	}
+
+	rc = sprintf(buf, "%d\n", data[TPM_GET_CAP_TEMP_INACTIVE_IDX]);
+
+	kfree(data);
+	return rc;
 }
 EXPORT_SYMBOL_GPL(tpm_show_temp_deactivated);
 
@@ -678,7 +720,7 @@ static const u8 pcrread[] = {
 ssize_t tpm_show_pcrs(struct device *dev, struct device_attribute *attr,
 		      char *buf)
 {
-	u8 data[max_t(int, max(ARRAY_SIZE(tpm_cap), ARRAY_SIZE(pcrread)), 30)];
+	u8 *data;
 	ssize_t rc;
 	int i, j, num_pcrs;
 	__be32 index;
@@ -688,21 +730,27 @@ ssize_t tpm_show_pcrs(struct device *dev, struct device_attribute *attr,
 	if (chip == NULL)
 		return -ENODEV;
 
+	data = kzalloc(TPM_INTERNAL_RESULT_SIZE, GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
+
 	memcpy(data, tpm_cap, sizeof(tpm_cap));
 	data[TPM_CAP_IDX] = TPM_CAP_PROP;
 	data[TPM_CAP_SUBCAP_IDX] = TPM_CAP_PROP_PCR;
 
-	rc = transmit_cmd(chip, data, sizeof(data),
+	rc = transmit_cmd(chip, data, TPM_INTERNAL_RESULT_SIZE,
 			"attempting to determine the number of PCRS");
-	if (rc)
+	if (rc) {
+		kfree(data);
 		return 0;
+	}
 
 	num_pcrs = be32_to_cpu(*((__be32 *) (data + 14)));
 	for (i = 0; i < num_pcrs; i++) {
 		memcpy(data, pcrread, sizeof(pcrread));
 		index = cpu_to_be32(i);
 		memcpy(data + 10, &index, 4);
-		rc = transmit_cmd(chip, data, sizeof(data),
+		rc = transmit_cmd(chip, data, TPM_INTERNAL_RESULT_SIZE,
 				"attempting to read a PCR");
 		if (rc)
 			goto out;
@@ -712,6 +760,7 @@ ssize_t tpm_show_pcrs(struct device *dev, struct device_attribute *attr,
 		str += sprintf(str, "\n");
 	}
 out:
+	kfree(data);
 	return str - buf;
 }
 EXPORT_SYMBOL_GPL(tpm_show_pcrs);
@@ -795,7 +844,7 @@ static const u8 cap_version[] = {
 ssize_t tpm_show_caps(struct device *dev, struct device_attribute *attr,
 		      char *buf)
 {
-	u8 data[max_t(int, max(ARRAY_SIZE(tpm_cap), ARRAY_SIZE(cap_version)), 30)];
+	u8 *data;
 	ssize_t rc;
 	char *str = buf;
 
@@ -803,21 +852,27 @@ ssize_t tpm_show_caps(struct device *dev, struct device_attribute *attr,
 	if (chip == NULL)
 		return -ENODEV;
 
+	data = kzalloc(TPM_INTERNAL_RESULT_SIZE, GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
+
 	memcpy(data, tpm_cap, sizeof(tpm_cap));
 	data[TPM_CAP_IDX] = TPM_CAP_PROP;
 	data[TPM_CAP_SUBCAP_IDX] = TPM_CAP_PROP_MANUFACTURER;
 
-	rc = transmit_cmd(chip, data, sizeof(data),
+	rc = transmit_cmd(chip, data, TPM_INTERNAL_RESULT_SIZE,
 			"attempting to determine the manufacturer");
-	if (rc)
+	if (rc) {
+		kfree(data);
 		return 0;
+	}
 
 	str += sprintf(str, "Manufacturer: 0x%x\n",
 		       be32_to_cpu(*((__be32 *) (data + TPM_GET_CAP_RET_UINT32_1_IDX))));
 
 	memcpy(data, cap_version, sizeof(cap_version));
 	data[CAP_VERSION_IDX] = CAP_VERSION_1_1;
-	rc = transmit_cmd(chip, data, sizeof(data),
+	rc = transmit_cmd(chip, data, TPM_INTERNAL_RESULT_SIZE,
 			"attempting to determine the 1.1 version");
 	if (rc)
 		goto out;
@@ -828,6 +883,7 @@ ssize_t tpm_show_caps(struct device *dev, struct device_attribute *attr,
 		       (int) data[17]);
 
 out:
+	kfree(data);
 	return str - buf;
 }
 EXPORT_SYMBOL_GPL(tpm_show_caps);
@@ -835,7 +891,7 @@ EXPORT_SYMBOL_GPL(tpm_show_caps);
 ssize_t tpm_show_caps_1_2(struct device * dev,
 			  struct device_attribute * attr, char *buf)
 {
-	u8 data[max_t(int, max(ARRAY_SIZE(tpm_cap), ARRAY_SIZE(cap_version)), 30)];
+	u8 *data;
 	ssize_t len;
 	char *str = buf;
 
@@ -843,15 +899,20 @@ ssize_t tpm_show_caps_1_2(struct device * dev,
 	if (chip == NULL)
 		return -ENODEV;
 
+	data = kzalloc(TPM_INTERNAL_RESULT_SIZE, GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
+
 	memcpy(data, tpm_cap, sizeof(tpm_cap));
 	data[TPM_CAP_IDX] = TPM_CAP_PROP;
 	data[TPM_CAP_SUBCAP_IDX] = TPM_CAP_PROP_MANUFACTURER;
 
-	if ((len = tpm_transmit(chip, data, sizeof(data))) <=
-	    TPM_ERROR_SIZE) {
+	len = tpm_transmit(chip, data, TPM_INTERNAL_RESULT_SIZE);
+	if (len <= TPM_ERROR_SIZE) {
 		dev_dbg(chip->dev, "A TPM error (%d) occurred "
 			"attempting to determine the manufacturer\n",
 			be32_to_cpu(*((__be32 *) (data + TPM_RET_CODE_IDX))));
+		kfree(data);
 		return 0;
 	}
 
@@ -861,8 +922,8 @@ ssize_t tpm_show_caps_1_2(struct device * dev,
 	memcpy(data, cap_version, sizeof(cap_version));
 	data[CAP_VERSION_IDX] = CAP_VERSION_1_2;
 
-	if ((len = tpm_transmit(chip, data, sizeof(data))) <=
-	    TPM_ERROR_SIZE) {
+	len = tpm_transmit(chip, data, TPM_INTERNAL_RESULT_SIZE);
+	if (len <= TPM_ERROR_SIZE) {
 		dev_err(chip->dev, "A TPM error (%d) occurred "
 			"attempting to determine the 1.2 version\n",
 			be32_to_cpu(*((__be32 *) (data + TPM_RET_CODE_IDX))));
@@ -874,6 +935,7 @@ ssize_t tpm_show_caps_1_2(struct device * dev,
 		       (int) data[19]);
 
 out:
+	kfree(data);
 	return str - buf;
 }
 EXPORT_SYMBOL_GPL(tpm_show_caps_1_2);

From 0147600172b4a5d261165d1aa5ef818d84da1557 Mon Sep 17 00:00:00 2001
From: Michael Halcrow <mhalcrow@us.ibm.com>
Date: Fri, 25 Jul 2008 19:45:00 -0700
Subject: [PATCH 1003/1171] tpm: Use correct data types for sizes in
 tpm_write() and tpm_read()

Use the correct data types for the size parameters in tpm_write() and
tpm_read(). Note that rw_verify_area() makes sure that this bug cannot
be exploited to produce a buffer overrun.

Signed-off-by: Michael Halcrow <mhalcrow@us.ibm.com>
Cc: Marcel Selhorst <tpm@selhorst.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/tpm/tpm.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/char/tpm/tpm.c b/drivers/char/tpm/tpm.c
index f354d720b777..ae766d868454 100644
--- a/drivers/char/tpm/tpm.c
+++ b/drivers/char/tpm/tpm.c
@@ -1028,7 +1028,7 @@ ssize_t tpm_write(struct file *file, const char __user *buf,
 		  size_t size, loff_t *off)
 {
 	struct tpm_chip *chip = file->private_data;
-	int in_size = size, out_size;
+	size_t in_size = size, out_size;
 
 	/* cannot perform a write until the read has cleared
 	   either via tpm_read or a user_read_timer timeout */
@@ -1063,7 +1063,7 @@ ssize_t tpm_read(struct file *file, char __user *buf,
 		 size_t size, loff_t *off)
 {
 	struct tpm_chip *chip = file->private_data;
-	int ret_size;
+	ssize_t ret_size;
 
 	del_singleshot_timer_sync(&chip->user_read_timer);
 	flush_scheduled_work();

From b77899985bdfd85a8e5a6e485033a9b4713d2471 Mon Sep 17 00:00:00 2001
From: Alex Dubov <oakad@yahoo.com>
Date: Fri, 25 Jul 2008 19:45:00 -0700
Subject: [PATCH 1004/1171] memstick: allow "set_param" method to return an
 error code

Some controllers (Jmicron, for instance) can report temporal failure
condition during power-on.  It is desirable to account for this using a
return value of "set_param" device method.  The return value can also be
handy to distinguish between supported and unsupported device parameters
in run time.

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Alex Dubov <oakad@yahoo.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/memstick/core/memstick.c  | 18 ++++++---
 drivers/memstick/host/jmb38x_ms.c | 67 ++++++++++++++++++++++---------
 drivers/memstick/host/tifm_ms.c   | 17 ++++----
 include/linux/memstick.h          |  2 +-
 4 files changed, 71 insertions(+), 33 deletions(-)

diff --git a/drivers/memstick/core/memstick.c b/drivers/memstick/core/memstick.c
index 61b98c333cb0..3c7d9a79c1ea 100644
--- a/drivers/memstick/core/memstick.c
+++ b/drivers/memstick/core/memstick.c
@@ -415,10 +415,14 @@ err_out:
 	return NULL;
 }
 
-static void memstick_power_on(struct memstick_host *host)
+static int memstick_power_on(struct memstick_host *host)
 {
-	host->set_param(host, MEMSTICK_POWER, MEMSTICK_POWER_ON);
-	host->set_param(host, MEMSTICK_INTERFACE, MEMSTICK_SERIAL);
+	int rc = host->set_param(host, MEMSTICK_POWER, MEMSTICK_POWER_ON);
+
+	if (!rc)
+		rc = host->set_param(host, MEMSTICK_INTERFACE, MEMSTICK_SERIAL);
+
+	return rc;
 }
 
 static void memstick_check(struct work_struct *work)
@@ -573,11 +577,15 @@ EXPORT_SYMBOL(memstick_suspend_host);
  */
 void memstick_resume_host(struct memstick_host *host)
 {
+	int rc = 0;
+
 	mutex_lock(&host->lock);
 	if (host->card)
-		memstick_power_on(host);
+		rc = memstick_power_on(host);
 	mutex_unlock(&host->lock);
-	memstick_detect_change(host);
+
+	if (!rc)
+		memstick_detect_change(host);
 }
 EXPORT_SYMBOL(memstick_resume_host);
 
diff --git a/drivers/memstick/host/jmb38x_ms.c b/drivers/memstick/host/jmb38x_ms.c
index 4e3bfbcdf155..9d82e67737db 100644
--- a/drivers/memstick/host/jmb38x_ms.c
+++ b/drivers/memstick/host/jmb38x_ms.c
@@ -609,36 +609,68 @@ static void jmb38x_ms_request(struct memstick_host *msh)
 	spin_unlock_irqrestore(&host->lock, flags);
 }
 
-static void jmb38x_ms_reset(struct jmb38x_ms_host *host)
+static int jmb38x_ms_reset(struct jmb38x_ms_host *host)
 {
-	unsigned int host_ctl = readl(host->addr + HOST_CONTROL);
+	int cnt;
 
-	writel(HOST_CONTROL_RESET_REQ, host->addr + HOST_CONTROL);
+	writel(HOST_CONTROL_RESET_REQ | HOST_CONTROL_CLOCK_EN
+	       | readl(host->addr + HOST_CONTROL),
+	       host->addr + HOST_CONTROL);
+	mmiowb();
+
+	for (cnt = 0; cnt < 20; ++cnt) {
+		if (!(HOST_CONTROL_RESET_REQ
+		      & readl(host->addr + HOST_CONTROL)))
+			goto reset_next;
 
-	while (HOST_CONTROL_RESET_REQ
-	       & (host_ctl = readl(host->addr + HOST_CONTROL))) {
 		ndelay(20);
-		dev_dbg(&host->chip->pdev->dev, "reset %08x\n", host_ctl);
 	}
+	dev_dbg(&host->chip->pdev->dev, "reset_req timeout\n");
+	return -EIO;
 
-	writel(HOST_CONTROL_RESET, host->addr + HOST_CONTROL);
+reset_next:
+	writel(HOST_CONTROL_RESET | HOST_CONTROL_CLOCK_EN
+	       | readl(host->addr + HOST_CONTROL),
+	       host->addr + HOST_CONTROL);
+	mmiowb();
+
+	for (cnt = 0; cnt < 20; ++cnt) {
+		if (!(HOST_CONTROL_RESET
+		      & readl(host->addr + HOST_CONTROL)))
+			goto reset_ok;
+
+		ndelay(20);
+	}
+	dev_dbg(&host->chip->pdev->dev, "reset timeout\n");
+	return -EIO;
+
+reset_ok:
 	mmiowb();
 	writel(INT_STATUS_ALL, host->addr + INT_SIGNAL_ENABLE);
 	writel(INT_STATUS_ALL, host->addr + INT_STATUS_ENABLE);
+	return 0;
 }
 
-static void jmb38x_ms_set_param(struct memstick_host *msh,
-				enum memstick_param param,
-				int value)
+static int jmb38x_ms_set_param(struct memstick_host *msh,
+			       enum memstick_param param,
+			       int value)
 {
 	struct jmb38x_ms_host *host = memstick_priv(msh);
 	unsigned int host_ctl = readl(host->addr + HOST_CONTROL);
 	unsigned int clock_ctl = CLOCK_CONTROL_40MHZ, clock_delay = 0;
+	int rc = 0;
 
 	switch (param) {
 	case MEMSTICK_POWER:
 		if (value == MEMSTICK_POWER_ON) {
-			jmb38x_ms_reset(host);
+			rc = jmb38x_ms_reset(host);
+			if (rc)
+				return rc;
+
+			host_ctl = 7;
+			host_ctl |= HOST_CONTROL_POWER_EN
+				 | HOST_CONTROL_CLOCK_EN;
+			writel(host_ctl, host->addr + HOST_CONTROL);
 
 			writel(host->id ? PAD_PU_PD_ON_MS_SOCK1
 					: PAD_PU_PD_ON_MS_SOCK0,
@@ -647,11 +679,7 @@ static void jmb38x_ms_set_param(struct memstick_host *msh,
 			writel(PAD_OUTPUT_ENABLE_MS,
 			       host->addr + PAD_OUTPUT_ENABLE);
 
-			host_ctl = 7;
-			host_ctl |= HOST_CONTROL_POWER_EN
-				 | HOST_CONTROL_CLOCK_EN;
-			writel(host_ctl, host->addr + HOST_CONTROL);
-
+			msleep(10);
 			dev_dbg(&host->chip->pdev->dev, "power on\n");
 		} else if (value == MEMSTICK_POWER_OFF) {
 			host_ctl &= ~(HOST_CONTROL_POWER_EN
@@ -660,7 +688,8 @@ static void jmb38x_ms_set_param(struct memstick_host *msh,
 			writel(0, host->addr + PAD_OUTPUT_ENABLE);
 			writel(PAD_PU_PD_OFF, host->addr + PAD_PU_PD);
 			dev_dbg(&host->chip->pdev->dev, "power off\n");
-		}
+		} else
+			return -EINVAL;
 		break;
 	case MEMSTICK_INTERFACE:
 		host_ctl &= ~(3 << HOST_CONTROL_IF_SHIFT);
@@ -686,12 +715,14 @@ static void jmb38x_ms_set_param(struct memstick_host *msh,
 			host_ctl &= ~HOST_CONTROL_REI;
 			clock_ctl = CLOCK_CONTROL_60MHZ;
 			clock_delay = 0;
-		}
+		} else
+			return -EINVAL;
 		writel(host_ctl, host->addr + HOST_CONTROL);
 		writel(clock_ctl, host->addr + CLOCK_CONTROL);
 		writel(clock_delay, host->addr + CLOCK_DELAY);
 		break;
 	};
+	return 0;
 }
 
 #ifdef CONFIG_PM
diff --git a/drivers/memstick/host/tifm_ms.c b/drivers/memstick/host/tifm_ms.c
index 8577de4ebb0e..14458764588c 100644
--- a/drivers/memstick/host/tifm_ms.c
+++ b/drivers/memstick/host/tifm_ms.c
@@ -489,15 +489,12 @@ static void tifm_ms_request(struct memstick_host *msh)
 	return;
 }
 
-static void tifm_ms_set_param(struct memstick_host *msh,
-			      enum memstick_param param,
-			      int value)
+static int tifm_ms_set_param(struct memstick_host *msh,
+			     enum memstick_param param,
+			     int value)
 {
 	struct tifm_ms *host = memstick_priv(msh);
 	struct tifm_dev *sock = host->dev;
-	unsigned long flags;
-
-	spin_lock_irqsave(&sock->lock, flags);
 
 	switch (param) {
 	case MEMSTICK_POWER:
@@ -512,7 +509,8 @@ static void tifm_ms_set_param(struct memstick_host *msh,
 			writel(TIFM_MS_SYS_FCLR | TIFM_MS_SYS_INTCLR,
 			       sock->addr + SOCK_MS_SYSTEM);
 			writel(0xffffffff, sock->addr + SOCK_MS_STATUS);
-		}
+		} else
+			return -EINVAL;
 		break;
 	case MEMSTICK_INTERFACE:
 		if (value == MEMSTICK_SERIAL) {
@@ -525,11 +523,12 @@ static void tifm_ms_set_param(struct memstick_host *msh,
 			writel(TIFM_CTRL_FAST_CLK
 			       | readl(sock->addr + SOCK_CONTROL),
 			       sock->addr + SOCK_CONTROL);
-		}
+		} else
+			return -EINVAL;
 		break;
 	};
 
-	spin_unlock_irqrestore(&sock->lock, flags);
+	return 0;
 }
 
 static void tifm_ms_abort(unsigned long data)
diff --git a/include/linux/memstick.h b/include/linux/memstick.h
index 37a5cdb03918..2fe599c66d52 100644
--- a/include/linux/memstick.h
+++ b/include/linux/memstick.h
@@ -284,7 +284,7 @@ struct memstick_host {
 	/* Notify the host that some requests are pending. */
 	void                (*request)(struct memstick_host *host);
 	/* Set host IO parameters (power, clock, etc).     */
-	void                (*set_param)(struct memstick_host *host,
+	int                 (*set_param)(struct memstick_host *host,
 					 enum memstick_param param,
 					 int value);
 	unsigned long       private[0] ____cacheline_aligned;

From 17017d8d2c005734d7088d8281ce2daab8fcb097 Mon Sep 17 00:00:00 2001
From: Alex Dubov <oakad@yahoo.com>
Date: Fri, 25 Jul 2008 19:45:01 -0700
Subject: [PATCH 1005/1171] memstick: add "start" and "stop" methods to
 memstick device

In some cases it may be desirable to ensure that associated driver is not
going to access the media in some period of time.  "start" and "stop"
methods are provided therefore to allow it.

Signed-off-by: Alex Dubov <oakad@yahoo.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/memstick/core/memstick.c    | 11 +++++++---
 drivers/memstick/core/mspro_block.c | 33 +++++++++++++++++++++++++++++
 include/linux/memstick.h            |  4 ++++
 3 files changed, 45 insertions(+), 3 deletions(-)

diff --git a/drivers/memstick/core/memstick.c b/drivers/memstick/core/memstick.c
index 3c7d9a79c1ea..7162f772bbfb 100644
--- a/drivers/memstick/core/memstick.c
+++ b/drivers/memstick/core/memstick.c
@@ -433,8 +433,11 @@ static void memstick_check(struct work_struct *work)
 
 	dev_dbg(&host->dev, "memstick_check started\n");
 	mutex_lock(&host->lock);
-	if (!host->card)
-		memstick_power_on(host);
+	if (!host->card) {
+		if (memstick_power_on(host))
+			goto out_power_off;
+	} else
+		host->card->stop(host->card);
 
 	card = memstick_alloc_card(host);
 
@@ -452,7 +455,8 @@ static void memstick_check(struct work_struct *work)
 			    || !(host->card->check(host->card))) {
 				device_unregister(&host->card->dev);
 				host->card = NULL;
-			}
+			} else
+				host->card->start(host->card);
 		}
 
 		if (!host->card) {
@@ -465,6 +469,7 @@ static void memstick_check(struct work_struct *work)
 			kfree(card);
 	}
 
+out_power_off:
 	if (!host->card)
 		host->set_param(host, MEMSTICK_POWER, MEMSTICK_POWER_OFF);
 
diff --git a/drivers/memstick/core/mspro_block.c b/drivers/memstick/core/mspro_block.c
index 477d0fb6e588..004ac4d176d9 100644
--- a/drivers/memstick/core/mspro_block.c
+++ b/drivers/memstick/core/mspro_block.c
@@ -752,6 +752,37 @@ static int mspro_block_has_request(struct mspro_block_data *msb)
 	return rc;
 }
 
+static void mspro_block_stop(struct memstick_dev *card)
+{
+	struct mspro_block_data *msb = memstick_get_drvdata(card);
+	int rc = 0;
+	unsigned long flags;
+
+	while (1) {
+		spin_lock_irqsave(&msb->q_lock, flags);
+		if (!msb->has_request) {
+			blk_stop_queue(msb->queue);
+			rc = 1;
+		}
+		spin_unlock_irqrestore(&msb->q_lock, flags);
+
+		if (rc)
+			break;
+
+		wait_for_completion(&card->mrq_complete);
+	}
+}
+
+static void mspro_block_start(struct memstick_dev *card)
+{
+	struct mspro_block_data *msb = memstick_get_drvdata(card);
+	unsigned long flags;
+
+	spin_lock_irqsave(&msb->q_lock, flags);
+	blk_start_queue(msb->queue);
+	spin_unlock_irqrestore(&msb->q_lock, flags);
+}
+
 static int mspro_block_queue_thread(void *data)
 {
 	struct memstick_dev *card = data;
@@ -1272,6 +1303,8 @@ static int mspro_block_probe(struct memstick_dev *card)
 	rc = mspro_block_init_disk(card);
 	if (!rc) {
 		card->check = mspro_block_check_card;
+		card->stop = mspro_block_stop;
+		card->start = mspro_block_start;
 		return 0;
 	}
 
diff --git a/include/linux/memstick.h b/include/linux/memstick.h
index 2fe599c66d52..a9f998a3f48b 100644
--- a/include/linux/memstick.h
+++ b/include/linux/memstick.h
@@ -263,6 +263,10 @@ struct memstick_dev {
 	/* Get next request from the media driver.                         */
 	int                      (*next_request)(struct memstick_dev *card,
 						 struct memstick_request **mrq);
+	/* Tell the media driver to stop doing things                      */
+	void                     (*stop)(struct memstick_dev *card);
+	/* Allow the media driver to continue                              */
+	void                     (*start)(struct memstick_dev *card);
 
 	struct device            dev;
 };

From f1d82698029b92a88f5500b99f66514b6dee2bc3 Mon Sep 17 00:00:00 2001
From: Alex Dubov <oakad@yahoo.com>
Date: Fri, 25 Jul 2008 19:45:02 -0700
Subject: [PATCH 1006/1171] memstick: use fully asynchronous request processing

Instead of using a separate thread to pump requests from block layer queue
to memstick, do so inline, utilizing the callback design of the memstick.

[akpm@linux-foundation.org: fix warnings]
Signed-off-by: Alex Dubov <oakad@yahoo.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/memstick/core/memstick.c    |   7 +-
 drivers/memstick/core/mspro_block.c | 336 +++++++++++++---------------
 drivers/memstick/host/jmb38x_ms.c   |  35 ++-
 drivers/memstick/host/tifm_ms.c     |  49 ++--
 4 files changed, 214 insertions(+), 213 deletions(-)

diff --git a/drivers/memstick/core/memstick.c b/drivers/memstick/core/memstick.c
index 7162f772bbfb..a38005008a20 100644
--- a/drivers/memstick/core/memstick.c
+++ b/drivers/memstick/core/memstick.c
@@ -249,8 +249,11 @@ EXPORT_SYMBOL(memstick_next_req);
  */
 void memstick_new_req(struct memstick_host *host)
 {
-	host->retries = cmd_retries;
-	host->request(host);
+	if (host->card) {
+		host->retries = cmd_retries;
+		INIT_COMPLETION(host->card->mrq_complete);
+		host->request(host);
+	}
 }
 EXPORT_SYMBOL(memstick_new_req);
 
diff --git a/drivers/memstick/core/mspro_block.c b/drivers/memstick/core/mspro_block.c
index 004ac4d176d9..44b1817f2f2f 100644
--- a/drivers/memstick/core/mspro_block.c
+++ b/drivers/memstick/core/mspro_block.c
@@ -136,9 +136,8 @@ struct mspro_block_data {
 	unsigned int          caps;
 	struct gendisk        *disk;
 	struct request_queue  *queue;
+	struct request        *block_req;
 	spinlock_t            q_lock;
-	wait_queue_head_t     q_wait;
-	struct task_struct    *q_thread;
 
 	unsigned short        page_size;
 	unsigned short        cylinders;
@@ -147,9 +146,10 @@ struct mspro_block_data {
 
 	unsigned char         system;
 	unsigned char         read_only:1,
-			      active:1,
+			      eject:1,
 			      has_request:1,
-			      data_dir:1;
+			      data_dir:1,
+			      active:1;
 	unsigned char         transfer_cmd;
 
 	int                   (*mrq_handler)(struct memstick_dev *card,
@@ -160,12 +160,14 @@ struct mspro_block_data {
 	struct scatterlist    req_sg[MSPRO_BLOCK_MAX_SEGS];
 	unsigned int          seg_count;
 	unsigned int          current_seg;
-	unsigned short        current_page;
+	unsigned int          current_page;
 };
 
 static DEFINE_IDR(mspro_block_disk_idr);
 static DEFINE_MUTEX(mspro_block_disk_lock);
 
+static int mspro_block_complete_req(struct memstick_dev *card, int error);
+
 /*** Block device ***/
 
 static int mspro_block_bd_open(struct inode *inode, struct file *filp)
@@ -197,8 +199,10 @@ static int mspro_block_disk_release(struct gendisk *disk)
 
 	mutex_lock(&mspro_block_disk_lock);
 
-	if (msb->usage_count) {
-		msb->usage_count--;
+	if (msb) {
+		if (msb->usage_count)
+			msb->usage_count--;
+
 		if (!msb->usage_count) {
 			kfree(msb);
 			disk->private_data = NULL;
@@ -523,11 +527,13 @@ static int h_mspro_block_req_init(struct memstick_dev *card,
 static int h_mspro_block_default(struct memstick_dev *card,
 				 struct memstick_request **mrq)
 {
-	complete(&card->mrq_complete);
-	if (!(*mrq)->error)
-		return -EAGAIN;
-	else
-		return (*mrq)->error;
+	return mspro_block_complete_req(card, (*mrq)->error);
+}
+
+static int h_mspro_block_default_bad(struct memstick_dev *card,
+				     struct memstick_request **mrq)
+{
+	return -ENXIO;
 }
 
 static int h_mspro_block_get_ro(struct memstick_dev *card,
@@ -535,44 +541,30 @@ static int h_mspro_block_get_ro(struct memstick_dev *card,
 {
 	struct mspro_block_data *msb = memstick_get_drvdata(card);
 
-	if ((*mrq)->error) {
-		complete(&card->mrq_complete);
-		return (*mrq)->error;
+	if (!(*mrq)->error) {
+		if ((*mrq)->data[offsetof(struct ms_status_register, status0)]
+		    & MEMSTICK_STATUS0_WP)
+			msb->read_only = 1;
+		else
+			msb->read_only = 0;
 	}
 
-	if ((*mrq)->data[offsetof(struct ms_status_register, status0)]
-	    & MEMSTICK_STATUS0_WP)
-		msb->read_only = 1;
-	else
-		msb->read_only = 0;
-
-	complete(&card->mrq_complete);
-	return -EAGAIN;
+	return mspro_block_complete_req(card, (*mrq)->error);
 }
 
 static int h_mspro_block_wait_for_ced(struct memstick_dev *card,
 				      struct memstick_request **mrq)
 {
-	if ((*mrq)->error) {
-		complete(&card->mrq_complete);
-		return (*mrq)->error;
-	}
-
 	dev_dbg(&card->dev, "wait for ced: value %x\n", (*mrq)->data[0]);
 
-	if ((*mrq)->data[0] & (MEMSTICK_INT_CMDNAK | MEMSTICK_INT_ERR)) {
-		card->current_mrq.error = -EFAULT;
-		complete(&card->mrq_complete);
-		return card->current_mrq.error;
+	if (!(*mrq)->error) {
+		if ((*mrq)->data[0] & (MEMSTICK_INT_CMDNAK | MEMSTICK_INT_ERR))
+			(*mrq)->error = -EFAULT;
+		else if (!((*mrq)->data[0] & MEMSTICK_INT_CED))
+			return 0;
 	}
 
-	if (!((*mrq)->data[0] & MEMSTICK_INT_CED))
-		return 0;
-	else {
-		card->current_mrq.error = 0;
-		complete(&card->mrq_complete);
-		return -EAGAIN;
-	}
+	return mspro_block_complete_req(card, (*mrq)->error);
 }
 
 static int h_mspro_block_transfer_data(struct memstick_dev *card,
@@ -583,10 +575,8 @@ static int h_mspro_block_transfer_data(struct memstick_dev *card,
 	struct scatterlist t_sg = { 0 };
 	size_t t_offset;
 
-	if ((*mrq)->error) {
-		complete(&card->mrq_complete);
-		return (*mrq)->error;
-	}
+	if ((*mrq)->error)
+		return mspro_block_complete_req(card, (*mrq)->error);
 
 	switch ((*mrq)->tpc) {
 	case MS_TPC_WRITE_REG:
@@ -617,8 +607,8 @@ has_int_reg:
 
 			if (msb->current_seg == msb->seg_count) {
 				if (t_val & MEMSTICK_INT_CED) {
-					complete(&card->mrq_complete);
-					return -EAGAIN;
+					return mspro_block_complete_req(card,
+									0);
 				} else {
 					card->next_request
 						= h_mspro_block_wait_for_ced;
@@ -666,90 +656,120 @@ has_int_reg:
 
 /*** Data transfer ***/
 
-static void mspro_block_process_request(struct memstick_dev *card,
-					struct request *req)
+static int mspro_block_issue_req(struct memstick_dev *card, int chunk)
 {
 	struct mspro_block_data *msb = memstick_get_drvdata(card);
-	struct mspro_param_register param;
-	int rc, chunk, cnt;
-	unsigned short page_count;
 	sector_t t_sec;
-	unsigned long flags;
+	unsigned int count;
+	struct mspro_param_register param;
 
-	do {
-		page_count = 0;
+try_again:
+	while (chunk) {
+		msb->current_page = 0;
 		msb->current_seg = 0;
-		msb->seg_count = blk_rq_map_sg(req->q, req, msb->req_sg);
+		msb->seg_count = blk_rq_map_sg(msb->block_req->q,
+					       msb->block_req,
+					       msb->req_sg);
 
-		if (msb->seg_count) {
-			msb->current_page = 0;
-			for (rc = 0; rc < msb->seg_count; rc++)
-				page_count += msb->req_sg[rc].length
-					      / msb->page_size;
+		if (!msb->seg_count) {
+			chunk = __blk_end_request(msb->block_req, -ENOMEM,
+					blk_rq_cur_bytes(msb->block_req));
+			continue;
+		}
 
-			t_sec = req->sector;
-			sector_div(t_sec, msb->page_size >> 9);
-			param.system = msb->system;
-			param.data_count = cpu_to_be16(page_count);
-			param.data_address = cpu_to_be32((uint32_t)t_sec);
-			param.tpc_param = 0;
+		t_sec = msb->block_req->sector << 9;
+		sector_div(t_sec, msb->page_size);
 
-			msb->data_dir = rq_data_dir(req);
-			msb->transfer_cmd = msb->data_dir == READ
-					    ? MSPRO_CMD_READ_DATA
-					    : MSPRO_CMD_WRITE_DATA;
+		count = msb->block_req->nr_sectors << 9;
+		count /= msb->page_size;
 
-			dev_dbg(&card->dev, "data transfer: cmd %x, "
-				"lba %x, count %x\n", msb->transfer_cmd,
-				be32_to_cpu(param.data_address),
-				page_count);
+		param.system = msb->system;
+		param.data_count = cpu_to_be16(count);
+		param.data_address = cpu_to_be32((uint32_t)t_sec);
+		param.tpc_param = 0;
 
-			card->next_request = h_mspro_block_req_init;
-			msb->mrq_handler = h_mspro_block_transfer_data;
-			memstick_init_req(&card->current_mrq, MS_TPC_WRITE_REG,
-					  &param, sizeof(param));
-			memstick_new_req(card->host);
-			wait_for_completion(&card->mrq_complete);
-			rc = card->current_mrq.error;
+		msb->data_dir = rq_data_dir(msb->block_req);
+		msb->transfer_cmd = msb->data_dir == READ
+				    ? MSPRO_CMD_READ_DATA
+				    : MSPRO_CMD_WRITE_DATA;
 
-			if (rc || (card->current_mrq.tpc == MSPRO_CMD_STOP)) {
-				for (cnt = 0; cnt < msb->current_seg; cnt++)
-					page_count += msb->req_sg[cnt].length
-						      / msb->page_size;
+		dev_dbg(&card->dev, "data transfer: cmd %x, "
+			"lba %x, count %x\n", msb->transfer_cmd,
+			be32_to_cpu(param.data_address), count);
 
-				if (msb->current_page)
-					page_count += msb->current_page - 1;
+		card->next_request = h_mspro_block_req_init;
+		msb->mrq_handler = h_mspro_block_transfer_data;
+		memstick_init_req(&card->current_mrq, MS_TPC_WRITE_REG,
+				  &param, sizeof(param));
+		memstick_new_req(card->host);
+		return 0;
+	}
 
-				if (page_count && (msb->data_dir == READ))
-					rc = msb->page_size * page_count;
-				else
-					rc = -EIO;
-			} else
-				rc = msb->page_size * page_count;
-		} else
-			rc = -EFAULT;
+	dev_dbg(&card->dev, "elv_next\n");
+	msb->block_req = elv_next_request(msb->queue);
+	if (!msb->block_req) {
+		dev_dbg(&card->dev, "issue end\n");
+		return -EAGAIN;
+	}
 
-		spin_lock_irqsave(&msb->q_lock, flags);
-		if (rc >= 0)
-			chunk = __blk_end_request(req, 0, rc);
-		else
-			chunk = __blk_end_request(req, rc, 0);
-
-		dev_dbg(&card->dev, "end chunk %d, %d\n", rc, chunk);
-		spin_unlock_irqrestore(&msb->q_lock, flags);
-	} while (chunk);
+	dev_dbg(&card->dev, "trying again\n");
+	chunk = 1;
+	goto try_again;
 }
 
-static int mspro_block_has_request(struct mspro_block_data *msb)
+static int mspro_block_complete_req(struct memstick_dev *card, int error)
 {
-	int rc = 0;
+	struct mspro_block_data *msb = memstick_get_drvdata(card);
+	int chunk, cnt;
+	unsigned int t_len = 0;
 	unsigned long flags;
 
 	spin_lock_irqsave(&msb->q_lock, flags);
-	if (kthread_should_stop() || msb->has_request)
-		rc = 1;
+	dev_dbg(&card->dev, "complete %d, %d\n", msb->has_request ? 1 : 0,
+		error);
+
+	if (msb->has_request) {
+		/* Nothing to do - not really an error */
+		if (error == -EAGAIN)
+			error = 0;
+
+		if (error || (card->current_mrq.tpc == MSPRO_CMD_STOP)) {
+			if (msb->data_dir == READ) {
+				for (cnt = 0; cnt < msb->current_seg; cnt++)
+					t_len += msb->req_sg[cnt].length
+						 / msb->page_size;
+
+					if (msb->current_page)
+						t_len += msb->current_page - 1;
+
+					t_len *= msb->page_size;
+			}
+		} else
+			t_len = msb->block_req->nr_sectors << 9;
+
+		dev_dbg(&card->dev, "transferred %x (%d)\n", t_len, error);
+
+		if (error && !t_len)
+			t_len = blk_rq_cur_bytes(msb->block_req);
+
+		chunk = __blk_end_request(msb->block_req, error, t_len);
+
+		error = mspro_block_issue_req(card, chunk);
+
+		if (!error)
+			goto out;
+		else
+			msb->has_request = 0;
+	} else {
+		if (!error)
+			error = -EAGAIN;
+	}
+
+	card->next_request = h_mspro_block_default_bad;
+	complete_all(&card->mrq_complete);
+out:
 	spin_unlock_irqrestore(&msb->q_lock, flags);
-	return rc;
+	return error;
 }
 
 static void mspro_block_stop(struct memstick_dev *card)
@@ -783,54 +803,37 @@ static void mspro_block_start(struct memstick_dev *card)
 	spin_unlock_irqrestore(&msb->q_lock, flags);
 }
 
-static int mspro_block_queue_thread(void *data)
+static int mspro_block_prepare_req(struct request_queue *q, struct request *req)
 {
-	struct memstick_dev *card = data;
-	struct memstick_host *host = card->host;
-	struct mspro_block_data *msb = memstick_get_drvdata(card);
-	struct request *req;
-	unsigned long flags;
-
-	while (1) {
-		wait_event(msb->q_wait, mspro_block_has_request(msb));
-		dev_dbg(&card->dev, "thread iter\n");
-
-		spin_lock_irqsave(&msb->q_lock, flags);
-		req = elv_next_request(msb->queue);
-		dev_dbg(&card->dev, "next req %p\n", req);
-		if (!req) {
-			msb->has_request = 0;
-			if (kthread_should_stop()) {
-				spin_unlock_irqrestore(&msb->q_lock, flags);
-				break;
-			}
-		} else
-			msb->has_request = 1;
-		spin_unlock_irqrestore(&msb->q_lock, flags);
-
-		if (req) {
-			mutex_lock(&host->lock);
-			mspro_block_process_request(card, req);
-			mutex_unlock(&host->lock);
-		}
+	if (!blk_fs_request(req) && !blk_pc_request(req)) {
+		blk_dump_rq_flags(req, "MSPro unsupported request");
+		return BLKPREP_KILL;
 	}
-	dev_dbg(&card->dev, "thread finished\n");
-	return 0;
+
+	req->cmd_flags |= REQ_DONTPREP;
+
+	return BLKPREP_OK;
 }
 
-static void mspro_block_request(struct request_queue *q)
+static void mspro_block_submit_req(struct request_queue *q)
 {
 	struct memstick_dev *card = q->queuedata;
 	struct mspro_block_data *msb = memstick_get_drvdata(card);
 	struct request *req = NULL;
 
-	if (msb->q_thread) {
-		msb->has_request = 1;
-		wake_up_all(&msb->q_wait);
-	} else {
+	if (msb->has_request)
+		return;
+
+	if (msb->eject) {
 		while ((req = elv_next_request(q)) != NULL)
 			end_queued_request(req, -ENODEV);
+
+		return;
 	}
+
+	msb->has_request = 1;
+	if (mspro_block_issue_req(card, 0))
+		msb->has_request = 0;
 }
 
 /*** Initialization ***/
@@ -1200,16 +1203,14 @@ static int mspro_block_init_disk(struct memstick_dev *card)
 		goto out_release_id;
 	}
 
-	spin_lock_init(&msb->q_lock);
-	init_waitqueue_head(&msb->q_wait);
-
-	msb->queue = blk_init_queue(mspro_block_request, &msb->q_lock);
+	msb->queue = blk_init_queue(mspro_block_submit_req, &msb->q_lock);
 	if (!msb->queue) {
 		rc = -ENOMEM;
 		goto out_put_disk;
 	}
 
 	msb->queue->queuedata = card;
+	blk_queue_prep_rq(msb->queue, mspro_block_prepare_req);
 
 	blk_queue_bounce_limit(msb->queue, limit);
 	blk_queue_max_sectors(msb->queue, MSPRO_BLOCK_MAX_PAGES);
@@ -1235,14 +1236,8 @@ static int mspro_block_init_disk(struct memstick_dev *card)
 	capacity *= msb->page_size >> 9;
 	set_capacity(msb->disk, capacity);
 	dev_dbg(&card->dev, "capacity set %ld\n", capacity);
-	msb->q_thread = kthread_run(mspro_block_queue_thread, card,
-				    DRIVER_NAME"d");
-	if (IS_ERR(msb->q_thread))
-		goto out_put_disk;
 
-	mutex_unlock(&host->lock);
 	add_disk(msb->disk);
-	mutex_lock(&host->lock);
 	msb->active = 1;
 	return 0;
 
@@ -1290,6 +1285,7 @@ static int mspro_block_probe(struct memstick_dev *card)
 		return -ENOMEM;
 	memstick_set_drvdata(card, msb);
 	msb->card = card;
+	spin_lock_init(&msb->q_lock);
 
 	rc = mspro_block_init_card(card);
 
@@ -1319,26 +1315,17 @@ out_free:
 static void mspro_block_remove(struct memstick_dev *card)
 {
 	struct mspro_block_data *msb = memstick_get_drvdata(card);
-	struct task_struct *q_thread = NULL;
 	unsigned long flags;
 
 	del_gendisk(msb->disk);
 	dev_dbg(&card->dev, "mspro block remove\n");
 	spin_lock_irqsave(&msb->q_lock, flags);
-	q_thread = msb->q_thread;
-	msb->q_thread = NULL;
-	msb->active = 0;
+	msb->eject = 1;
+	blk_start_queue(msb->queue);
 	spin_unlock_irqrestore(&msb->q_lock, flags);
 
-	if (q_thread) {
-		mutex_unlock(&card->host->lock);
-		kthread_stop(q_thread);
-		mutex_lock(&card->host->lock);
-	}
-
-	dev_dbg(&card->dev, "queue thread stopped\n");
-
 	blk_cleanup_queue(msb->queue);
+	msb->queue = NULL;
 
 	sysfs_remove_group(&card->dev.kobj, &msb->attr_group);
 
@@ -1355,19 +1342,13 @@ static void mspro_block_remove(struct memstick_dev *card)
 static int mspro_block_suspend(struct memstick_dev *card, pm_message_t state)
 {
 	struct mspro_block_data *msb = memstick_get_drvdata(card);
-	struct task_struct *q_thread = NULL;
 	unsigned long flags;
 
 	spin_lock_irqsave(&msb->q_lock, flags);
-	q_thread = msb->q_thread;
-	msb->q_thread = NULL;
-	msb->active = 0;
 	blk_stop_queue(msb->queue);
+	msb->active = 0;
 	spin_unlock_irqrestore(&msb->q_lock, flags);
 
-	if (q_thread)
-		kthread_stop(q_thread);
-
 	return 0;
 }
 
@@ -1406,14 +1387,7 @@ static int mspro_block_resume(struct memstick_dev *card)
 			if (memcmp(s_attr->data, r_attr->data, s_attr->size))
 				break;
 
-			memstick_set_drvdata(card, msb);
-			msb->q_thread = kthread_run(mspro_block_queue_thread,
-						    card, DRIVER_NAME"d");
-			if (IS_ERR(msb->q_thread))
-				msb->q_thread = NULL;
-			else
-				msb->active = 1;
-
+			msb->active = 1;
 			break;
 		}
 	}
diff --git a/drivers/memstick/host/jmb38x_ms.c b/drivers/memstick/host/jmb38x_ms.c
index 9d82e67737db..3485c63d20b0 100644
--- a/drivers/memstick/host/jmb38x_ms.c
+++ b/drivers/memstick/host/jmb38x_ms.c
@@ -50,6 +50,7 @@ struct jmb38x_ms_host {
 	struct jmb38x_ms        *chip;
 	void __iomem            *addr;
 	spinlock_t              lock;
+	struct tasklet_struct   notify;
 	int                     id;
 	char                    host_id[32];
 	int                     irq;
@@ -590,25 +591,35 @@ static void jmb38x_ms_abort(unsigned long data)
 	spin_unlock_irqrestore(&host->lock, flags);
 }
 
-static void jmb38x_ms_request(struct memstick_host *msh)
+static void jmb38x_ms_req_tasklet(unsigned long data)
 {
+	struct memstick_host *msh = (struct memstick_host *)data;
 	struct jmb38x_ms_host *host = memstick_priv(msh);
 	unsigned long flags;
 	int rc;
 
 	spin_lock_irqsave(&host->lock, flags);
-	if (host->req) {
-		spin_unlock_irqrestore(&host->lock, flags);
-		BUG();
-		return;
+	if (!host->req) {
+		do {
+			rc = memstick_next_req(msh, &host->req);
+			dev_dbg(&host->chip->pdev->dev, "tasklet req %d\n", rc);
+		} while (!rc && jmb38x_ms_issue_cmd(msh));
 	}
-
-	do {
-		rc = memstick_next_req(msh, &host->req);
-	} while (!rc && jmb38x_ms_issue_cmd(msh));
 	spin_unlock_irqrestore(&host->lock, flags);
 }
 
+static void jmb38x_ms_dummy_submit(struct memstick_host *msh)
+{
+	return;
+}
+
+static void jmb38x_ms_submit_req(struct memstick_host *msh)
+{
+	struct jmb38x_ms_host *host = memstick_priv(msh);
+
+	tasklet_schedule(&host->notify);
+}
+
 static int jmb38x_ms_reset(struct jmb38x_ms_host *host)
 {
 	int cnt;
@@ -816,7 +827,9 @@ static struct memstick_host *jmb38x_ms_alloc_host(struct jmb38x_ms *jm, int cnt)
 		 host->id);
 	host->irq = jm->pdev->irq;
 	host->timeout_jiffies = msecs_to_jiffies(1000);
-	msh->request = jmb38x_ms_request;
+
+	tasklet_init(&host->notify, jmb38x_ms_req_tasklet, (unsigned long)msh);
+	msh->request = jmb38x_ms_submit_req;
 	msh->set_param = jmb38x_ms_set_param;
 
 	msh->caps = MEMSTICK_CAP_PAR4 | MEMSTICK_CAP_PAR8;
@@ -928,6 +941,8 @@ static void jmb38x_ms_remove(struct pci_dev *dev)
 
 		host = memstick_priv(jm->hosts[cnt]);
 
+		jm->hosts[cnt]->request = jmb38x_ms_dummy_submit;
+		tasklet_kill(&host->notify);
 		writel(0, host->addr + INT_SIGNAL_ENABLE);
 		writel(0, host->addr + INT_STATUS_ENABLE);
 		mmiowb();
diff --git a/drivers/memstick/host/tifm_ms.c b/drivers/memstick/host/tifm_ms.c
index 14458764588c..d32d6ad8f3fc 100644
--- a/drivers/memstick/host/tifm_ms.c
+++ b/drivers/memstick/host/tifm_ms.c
@@ -71,6 +71,7 @@ struct tifm_ms {
 	struct tifm_dev         *dev;
 	struct timer_list       timer;
 	struct memstick_request *req;
+	struct tasklet_struct   notify;
 	unsigned int            mode_mask;
 	unsigned int            block_pos;
 	unsigned long           timeout_jiffies;
@@ -455,40 +456,45 @@ static void tifm_ms_card_event(struct tifm_dev *sock)
 	return;
 }
 
-static void tifm_ms_request(struct memstick_host *msh)
+static void tifm_ms_req_tasklet(unsigned long data)
 {
+	struct memstick_host *msh = (struct memstick_host *)data;
 	struct tifm_ms *host = memstick_priv(msh);
 	struct tifm_dev *sock = host->dev;
 	unsigned long flags;
 	int rc;
 
 	spin_lock_irqsave(&sock->lock, flags);
-	if (host->req) {
-		printk(KERN_ERR "%s : unfinished request detected\n",
-		       sock->dev.bus_id);
-		spin_unlock_irqrestore(&sock->lock, flags);
-		tifm_eject(host->dev);
-		return;
-	}
+	if (!host->req) {
+		if (host->eject) {
+			do {
+				rc = memstick_next_req(msh, &host->req);
+				if (!rc)
+					host->req->error = -ETIME;
+			} while (!rc);
+			spin_unlock_irqrestore(&sock->lock, flags);
+			return;
+		}
 
-	if (host->eject) {
 		do {
 			rc = memstick_next_req(msh, &host->req);
-			if (!rc)
-				host->req->error = -ETIME;
-		} while (!rc);
-		spin_unlock_irqrestore(&sock->lock, flags);
-		return;
+		} while (!rc && tifm_ms_issue_cmd(host));
 	}
-
-	do {
-		rc = memstick_next_req(msh, &host->req);
-	} while (!rc && tifm_ms_issue_cmd(host));
-
 	spin_unlock_irqrestore(&sock->lock, flags);
+}
+
+static void tifm_ms_dummy_submit(struct memstick_host *msh)
+{
 	return;
 }
 
+static void tifm_ms_submit_req(struct memstick_host *msh)
+{
+	struct tifm_ms *host = memstick_priv(msh);
+
+	tasklet_schedule(&host->notify);
+}
+
 static int tifm_ms_set_param(struct memstick_host *msh,
 			     enum memstick_param param,
 			     int value)
@@ -569,8 +575,9 @@ static int tifm_ms_probe(struct tifm_dev *sock)
 	host->timeout_jiffies = msecs_to_jiffies(1000);
 
 	setup_timer(&host->timer, tifm_ms_abort, (unsigned long)host);
+	tasklet_init(&host->notify, tifm_ms_req_tasklet, (unsigned long)msh);
 
-	msh->request = tifm_ms_request;
+	msh->request = tifm_ms_submit_req;
 	msh->set_param = tifm_ms_set_param;
 	sock->card_event = tifm_ms_card_event;
 	sock->data_event = tifm_ms_data_event;
@@ -592,6 +599,8 @@ static void tifm_ms_remove(struct tifm_dev *sock)
 	int rc = 0;
 	unsigned long flags;
 
+	msh->request = tifm_ms_dummy_submit;
+	tasklet_kill(&host->notify);
 	spin_lock_irqsave(&sock->lock, flags);
 	host->eject = 1;
 	if (host->req) {

From 7fccf0326536c1b245b98740d489abb9aab69a12 Mon Sep 17 00:00:00 2001
From: WANG Cong <xiyou.wangcong@gmail.com>
Date: Fri, 25 Jul 2008 19:45:02 -0700
Subject: [PATCH 1007/1171] kernel/kexec.c: make 'kimage_terminate' void

Since kimage_terminate() always returns 0, make it void.

Signed-off-by: WANG Cong <wangcong@zeuux.org>
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/kexec.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/kernel/kexec.c b/kernel/kexec.c
index 1c5fcacbcf33..6db42ff8d520 100644
--- a/kernel/kexec.c
+++ b/kernel/kexec.c
@@ -589,14 +589,12 @@ static void kimage_free_extra_pages(struct kimage *image)
 	kimage_free_page_list(&image->unuseable_pages);
 
 }
-static int kimage_terminate(struct kimage *image)
+static void kimage_terminate(struct kimage *image)
 {
 	if (*image->entry != 0)
 		image->entry++;
 
 	*image->entry = IND_DONE;
-
-	return 0;
 }
 
 #define for_each_kimage_entry(image, ptr, entry) \
@@ -997,9 +995,7 @@ asmlinkage long sys_kexec_load(unsigned long entry, unsigned long nr_segments,
 			if (result)
 				goto out;
 		}
-		result = kimage_terminate(image);
-		if (result)
-			goto out;
+		kimage_terminate(image);
 	}
 	/* Install the new kernel, and  Uninstall the old */
 	image = xchg(dest_image, image);

From 3ab83521378268044a448113c6aa9a9e245f4d2f Mon Sep 17 00:00:00 2001
From: Huang Ying <ying.huang@intel.com>
Date: Fri, 25 Jul 2008 19:45:07 -0700
Subject: [PATCH 1008/1171] kexec jump

This patch provides an enhancement to kexec/kdump.  It implements the
following features:

- Backup/restore memory used by the original kernel before/after
  kexec.

- Save/restore CPU state before/after kexec.

The features of this patch can be used as a general method to call program in
physical mode (paging turning off).  This can be used to call BIOS code under
Linux.

kexec-tools needs to be patched to support kexec jump. The patches and
the precompiled kexec can be download from the following URL:

       source: http://khibernation.sourceforge.net/download/release_v10/kexec-tools/kexec-tools-src_git_kh10.tar.bz2
       patches: http://khibernation.sourceforge.net/download/release_v10/kexec-tools/kexec-tools-patches_git_kh10.tar.bz2
       binary: http://khibernation.sourceforge.net/download/release_v10/kexec-tools/kexec_git_kh10

Usage example of calling some physical mode code and return:

1. Compile and install patched kernel with following options selected:

CONFIG_X86_32=y
CONFIG_KEXEC=y
CONFIG_PM=y
CONFIG_KEXEC_JUMP=y

2. Build patched kexec-tool or download the pre-built one.

3. Build some physical mode executable named such as "phy_mode"

4. Boot kernel compiled in step 1.

5. Load physical mode executable with /sbin/kexec. The shell command
   line can be as follow:

   /sbin/kexec --load-preserve-context --args-none phy_mode

6. Call physical mode executable with following shell command line:

   /sbin/kexec -e

Implementation point:

To support jumping without reserving memory.  One shadow backup page (source
page) is allocated for each page used by kexeced code image (destination
page).  When do kexec_load, the image of kexeced code is loaded into source
pages, and before executing, the destination pages and the source pages are
swapped, so the contents of destination pages are backupped.  Before jumping
to the kexeced code image and after jumping back to the original kernel, the
destination pages and the source pages are swapped too.

C ABI (calling convention) is used as communication protocol between
kernel and called code.

A flag named KEXEC_PRESERVE_CONTEXT for sys_kexec_load is added to
indicate that the loaded kernel image is used for jumping back.

Now, only the i386 architecture is supported.

Signed-off-by: Huang Ying <ying.huang@intel.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Pavel Machek <pavel@ucw.cz>
Cc: Nigel Cunningham <nigel@nigel.suspend2.net>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/powerpc/kernel/machine_kexec.c  |   2 +-
 arch/sh/kernel/machine_kexec.c       |   2 +-
 arch/x86/Kconfig                     |   7 ++
 arch/x86/kernel/machine_kexec_32.c   |  27 +++-
 arch/x86/kernel/machine_kexec_64.c   |   2 +-
 arch/x86/kernel/relocate_kernel_32.S | 176 +++++++++++++++++++++++----
 include/asm-x86/kexec.h              |  18 +--
 include/linux/kexec.h                |  17 ++-
 kernel/kexec.c                       |  57 +++++++++
 kernel/sys.c                         |  31 ++---
 10 files changed, 270 insertions(+), 69 deletions(-)

diff --git a/arch/powerpc/kernel/machine_kexec.c b/arch/powerpc/kernel/machine_kexec.c
index 29a0e039d436..aab76887a842 100644
--- a/arch/powerpc/kernel/machine_kexec.c
+++ b/arch/powerpc/kernel/machine_kexec.c
@@ -48,7 +48,7 @@ void machine_kexec_cleanup(struct kimage *image)
  * Do not allocate memory (or fail in any way) in machine_kexec().
  * We are past the point of no return, committed to rebooting now.
  */
-NORET_TYPE void machine_kexec(struct kimage *image)
+void machine_kexec(struct kimage *image)
 {
 	if (ppc_md.machine_kexec)
 		ppc_md.machine_kexec(image);
diff --git a/arch/sh/kernel/machine_kexec.c b/arch/sh/kernel/machine_kexec.c
index 5c17de51987e..ec1eadce4aaa 100644
--- a/arch/sh/kernel/machine_kexec.c
+++ b/arch/sh/kernel/machine_kexec.c
@@ -70,7 +70,7 @@ static void kexec_info(struct kimage *image)
  * Do not allocate memory (or fail in any way) in machine_kexec().
  * We are past the point of no return, committed to rebooting now.
  */
-NORET_TYPE void machine_kexec(struct kimage *image)
+void machine_kexec(struct kimage *image)
 {
 
 	unsigned long page_list;
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index e3cba0b45600..7ecb679f0130 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -1279,6 +1279,13 @@ config CRASH_DUMP
 	  (CONFIG_RELOCATABLE=y).
 	  For more details see Documentation/kdump/kdump.txt
 
+config KEXEC_JUMP
+	bool "kexec jump (EXPERIMENTAL)"
+	depends on EXPERIMENTAL
+	depends on KEXEC && PM_SLEEP && X86_32
+	help
+	  Invoke code in physical address mode via KEXEC
+
 config PHYSICAL_START
 	hex "Physical address where the kernel is loaded" if (EMBEDDED || CRASH_DUMP)
 	default "0x1000000" if X86_NUMAQ
diff --git a/arch/x86/kernel/machine_kexec_32.c b/arch/x86/kernel/machine_kexec_32.c
index 8864230d55af..2b67609d0a1c 100644
--- a/arch/x86/kernel/machine_kexec_32.c
+++ b/arch/x86/kernel/machine_kexec_32.c
@@ -22,6 +22,7 @@
 #include <asm/cpufeature.h>
 #include <asm/desc.h>
 #include <asm/system.h>
+#include <asm/cacheflush.h>
 
 #define PAGE_ALIGNED __attribute__ ((__aligned__(PAGE_SIZE)))
 static u32 kexec_pgd[1024] PAGE_ALIGNED;
@@ -85,10 +86,12 @@ static void load_segments(void)
  * reboot code buffer to allow us to avoid allocations
  * later.
  *
- * Currently nothing.
+ * Make control page executable.
  */
 int machine_kexec_prepare(struct kimage *image)
 {
+	if (nx_enabled)
+		set_pages_x(image->control_code_page, 1);
 	return 0;
 }
 
@@ -98,16 +101,24 @@ int machine_kexec_prepare(struct kimage *image)
  */
 void machine_kexec_cleanup(struct kimage *image)
 {
+	if (nx_enabled)
+		set_pages_nx(image->control_code_page, 1);
 }
 
 /*
  * Do not allocate memory (or fail in any way) in machine_kexec().
  * We are past the point of no return, committed to rebooting now.
  */
-NORET_TYPE void machine_kexec(struct kimage *image)
+void machine_kexec(struct kimage *image)
 {
 	unsigned long page_list[PAGES_NR];
 	void *control_page;
+	asmlinkage unsigned long
+		(*relocate_kernel_ptr)(unsigned long indirection_page,
+				       unsigned long control_page,
+				       unsigned long start_address,
+				       unsigned int has_pae,
+				       unsigned int preserve_context);
 
 	tracer_disable();
 
@@ -115,10 +126,11 @@ NORET_TYPE void machine_kexec(struct kimage *image)
 	local_irq_disable();
 
 	control_page = page_address(image->control_code_page);
-	memcpy(control_page, relocate_kernel, PAGE_SIZE);
+	memcpy(control_page, relocate_kernel, PAGE_SIZE/2);
 
+	relocate_kernel_ptr = control_page;
 	page_list[PA_CONTROL_PAGE] = __pa(control_page);
-	page_list[VA_CONTROL_PAGE] = (unsigned long)relocate_kernel;
+	page_list[VA_CONTROL_PAGE] = (unsigned long)control_page;
 	page_list[PA_PGD] = __pa(kexec_pgd);
 	page_list[VA_PGD] = (unsigned long)kexec_pgd;
 #ifdef CONFIG_X86_PAE
@@ -131,6 +143,7 @@ NORET_TYPE void machine_kexec(struct kimage *image)
 	page_list[VA_PTE_0] = (unsigned long)kexec_pte0;
 	page_list[PA_PTE_1] = __pa(kexec_pte1);
 	page_list[VA_PTE_1] = (unsigned long)kexec_pte1;
+	page_list[PA_SWAP_PAGE] = (page_to_pfn(image->swap_page) << PAGE_SHIFT);
 
 	/* The segment registers are funny things, they have both a
 	 * visible and an invisible part.  Whenever the visible part is
@@ -149,8 +162,10 @@ NORET_TYPE void machine_kexec(struct kimage *image)
 	set_idt(phys_to_virt(0),0);
 
 	/* now call it */
-	relocate_kernel((unsigned long)image->head, (unsigned long)page_list,
-			image->start, cpu_has_pae);
+	image->start = relocate_kernel_ptr((unsigned long)image->head,
+					   (unsigned long)page_list,
+					   image->start, cpu_has_pae,
+					   image->preserve_context);
 }
 
 void arch_crash_save_vmcoreinfo(void)
diff --git a/arch/x86/kernel/machine_kexec_64.c b/arch/x86/kernel/machine_kexec_64.c
index 9dd9262693a3..c43caa3a91f3 100644
--- a/arch/x86/kernel/machine_kexec_64.c
+++ b/arch/x86/kernel/machine_kexec_64.c
@@ -181,7 +181,7 @@ void machine_kexec_cleanup(struct kimage *image)
  * Do not allocate memory (or fail in any way) in machine_kexec().
  * We are past the point of no return, committed to rebooting now.
  */
-NORET_TYPE void machine_kexec(struct kimage *image)
+void machine_kexec(struct kimage *image)
 {
 	unsigned long page_list[PAGES_NR];
 	void *control_page;
diff --git a/arch/x86/kernel/relocate_kernel_32.S b/arch/x86/kernel/relocate_kernel_32.S
index c30fe25d470d..703310a99023 100644
--- a/arch/x86/kernel/relocate_kernel_32.S
+++ b/arch/x86/kernel/relocate_kernel_32.S
@@ -20,11 +20,44 @@
 #define PAGE_ATTR (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
 #define PAE_PGD_ATTR (_PAGE_PRESENT)
 
+/* control_page + PAGE_SIZE/2 ~ control_page + PAGE_SIZE * 3/4 are
+ * used to save some data for jumping back
+ */
+#define DATA(offset)		(PAGE_SIZE/2+(offset))
+
+/* Minimal CPU state */
+#define ESP			DATA(0x0)
+#define CR0			DATA(0x4)
+#define CR3			DATA(0x8)
+#define CR4			DATA(0xc)
+
+/* other data */
+#define CP_VA_CONTROL_PAGE	DATA(0x10)
+#define CP_PA_PGD		DATA(0x14)
+#define CP_PA_SWAP_PAGE		DATA(0x18)
+#define CP_PA_BACKUP_PAGES_MAP	DATA(0x1c)
+
 	.text
 	.align PAGE_SIZE
 	.globl relocate_kernel
 relocate_kernel:
-	movl	8(%esp), %ebp /* list of pages */
+	/* Save the CPU context, used for jumping back */
+
+	pushl	%ebx
+	pushl	%esi
+	pushl	%edi
+	pushl	%ebp
+	pushf
+
+	movl	20+8(%esp), %ebp /* list of pages */
+	movl	PTR(VA_CONTROL_PAGE)(%ebp), %edi
+	movl	%esp, ESP(%edi)
+	movl	%cr0, %eax
+	movl	%eax, CR0(%edi)
+	movl	%cr3, %eax
+	movl	%eax, CR3(%edi)
+	movl	%cr4, %eax
+	movl	%eax, CR4(%edi)
 
 #ifdef CONFIG_X86_PAE
 	/* map the control page at its virtual address */
@@ -138,15 +171,25 @@ relocate_kernel:
 
 relocate_new_kernel:
 	/* read the arguments and say goodbye to the stack */
-	movl  4(%esp), %ebx /* page_list */
-	movl  8(%esp), %ebp /* list of pages */
-	movl  12(%esp), %edx /* start address */
-	movl  16(%esp), %ecx /* cpu_has_pae */
+	movl  20+4(%esp), %ebx /* page_list */
+	movl  20+8(%esp), %ebp /* list of pages */
+	movl  20+12(%esp), %edx /* start address */
+	movl  20+16(%esp), %ecx /* cpu_has_pae */
+	movl  20+20(%esp), %esi /* preserve_context */
 
 	/* zero out flags, and disable interrupts */
 	pushl $0
 	popfl
 
+	/* save some information for jumping back */
+	movl	PTR(VA_CONTROL_PAGE)(%ebp), %edi
+	movl	%edi, CP_VA_CONTROL_PAGE(%edi)
+	movl	PTR(PA_PGD)(%ebp), %eax
+	movl	%eax, CP_PA_PGD(%edi)
+	movl	PTR(PA_SWAP_PAGE)(%ebp), %eax
+	movl	%eax, CP_PA_SWAP_PAGE(%edi)
+	movl	%ebx, CP_PA_BACKUP_PAGES_MAP(%edi)
+
 	/* get physical address of control page now */
 	/* this is impossible after page table switch */
 	movl	PTR(PA_CONTROL_PAGE)(%ebp), %edi
@@ -197,8 +240,90 @@ identity_mapped:
 	xorl	%eax, %eax
 	movl	%eax, %cr3
 
+	movl	CP_PA_SWAP_PAGE(%edi), %eax
+	pushl	%eax
+	pushl	%ebx
+	call	swap_pages
+	addl	$8, %esp
+
+	/* To be certain of avoiding problems with self-modifying code
+	 * I need to execute a serializing instruction here.
+	 * So I flush the TLB, it's handy, and not processor dependent.
+	 */
+	xorl	%eax, %eax
+	movl	%eax, %cr3
+
+	/* set all of the registers to known values */
+	/* leave %esp alone */
+
+	testl	%esi, %esi
+	jnz 1f
+	xorl	%edi, %edi
+	xorl	%eax, %eax
+	xorl	%ebx, %ebx
+	xorl    %ecx, %ecx
+	xorl    %edx, %edx
+	xorl    %esi, %esi
+	xorl    %ebp, %ebp
+	ret
+1:
+	popl	%edx
+	movl	CP_PA_SWAP_PAGE(%edi), %esp
+	addl	$PAGE_SIZE, %esp
+2:
+	call	*%edx
+
+	/* get the re-entry point of the peer system */
+	movl	0(%esp), %ebp
+	call	1f
+1:
+	popl	%ebx
+	subl	$(1b - relocate_kernel), %ebx
+	movl	CP_VA_CONTROL_PAGE(%ebx), %edi
+	lea	PAGE_SIZE(%ebx), %esp
+	movl	CP_PA_SWAP_PAGE(%ebx), %eax
+	movl	CP_PA_BACKUP_PAGES_MAP(%ebx), %edx
+	pushl	%eax
+	pushl	%edx
+	call	swap_pages
+	addl	$8, %esp
+	movl	CP_PA_PGD(%ebx), %eax
+	movl	%eax, %cr3
+	movl	%cr0, %eax
+	orl	$(1<<31), %eax
+	movl	%eax, %cr0
+	lea	PAGE_SIZE(%edi), %esp
+	movl	%edi, %eax
+	addl	$(virtual_mapped - relocate_kernel), %eax
+	pushl	%eax
+	ret
+
+virtual_mapped:
+	movl	CR4(%edi), %eax
+	movl	%eax, %cr4
+	movl	CR3(%edi), %eax
+	movl	%eax, %cr3
+	movl	CR0(%edi), %eax
+	movl	%eax, %cr0
+	movl	ESP(%edi), %esp
+	movl	%ebp, %eax
+
+	popf
+	popl	%ebp
+	popl	%edi
+	popl	%esi
+	popl	%ebx
+	ret
+
 	/* Do the copies */
-	movl	%ebx, %ecx
+swap_pages:
+	movl	8(%esp), %edx
+	movl	4(%esp), %ecx
+	pushl	%ebp
+	pushl	%ebx
+	pushl	%edi
+	pushl	%esi
+	movl	%ecx, %ebx
 	jmp	1f
 
 0:	/* top, read another word from the indirection page */
@@ -226,27 +351,28 @@ identity_mapped:
 	movl    %ecx,   %esi /* For every source page do a copy */
 	andl    $0xfffff000, %esi
 
+	movl	%edi, %eax
+	movl	%esi, %ebp
+
+	movl	%edx, %edi
 	movl    $1024, %ecx
 	rep ; movsl
+
+	movl	%ebp, %edi
+	movl	%eax, %esi
+	movl	$1024, %ecx
+	rep ; movsl
+
+	movl	%eax, %edi
+	movl	%edx, %esi
+	movl	$1024, %ecx
+	rep ; movsl
+
+	lea	PAGE_SIZE(%ebp), %esi
 	jmp     0b
-
 3:
-
-	/* To be certain of avoiding problems with self-modifying code
-	 * I need to execute a serializing instruction here.
-	 * So I flush the TLB, it's handy, and not processor dependent.
-	 */
-	xorl	%eax, %eax
-	movl	%eax, %cr3
-
-	/* set all of the registers to known values */
-	/* leave %esp alone */
-
-	xorl	%eax, %eax
-	xorl	%ebx, %ebx
-	xorl    %ecx, %ecx
-	xorl    %edx, %edx
-	xorl    %esi, %esi
-	xorl    %edi, %edi
-	xorl    %ebp, %ebp
+	popl	%esi
+	popl	%edi
+	popl	%ebx
+	popl	%ebp
 	ret
diff --git a/include/asm-x86/kexec.h b/include/asm-x86/kexec.h
index 8f855a15f64d..c0e52a14fd4d 100644
--- a/include/asm-x86/kexec.h
+++ b/include/asm-x86/kexec.h
@@ -10,14 +10,15 @@
 # define VA_PTE_0		5
 # define PA_PTE_1		6
 # define VA_PTE_1		7
+# define PA_SWAP_PAGE		8
 # ifdef CONFIG_X86_PAE
-#  define PA_PMD_0		8
-#  define VA_PMD_0		9
-#  define PA_PMD_1		10
-#  define VA_PMD_1		11
-#  define PAGES_NR		12
+#  define PA_PMD_0		9
+#  define VA_PMD_0		10
+#  define PA_PMD_1		11
+#  define VA_PMD_1		12
+#  define PAGES_NR		13
 # else
-#  define PAGES_NR		8
+#  define PAGES_NR		9
 # endif
 #else
 # define PA_CONTROL_PAGE	0
@@ -152,11 +153,12 @@ static inline void crash_setup_regs(struct pt_regs *newregs,
 }
 
 #ifdef CONFIG_X86_32
-asmlinkage NORET_TYPE void
+asmlinkage unsigned long
 relocate_kernel(unsigned long indirection_page,
 		unsigned long control_page,
 		unsigned long start_address,
-		unsigned int has_pae) ATTRIB_NORET;
+		unsigned int has_pae,
+		unsigned int preserve_context);
 #else
 NORET_TYPE void
 relocate_kernel(unsigned long indirection_page,
diff --git a/include/linux/kexec.h b/include/linux/kexec.h
index 3265968cd2cd..82f88a8a827b 100644
--- a/include/linux/kexec.h
+++ b/include/linux/kexec.h
@@ -83,6 +83,7 @@ struct kimage {
 
 	unsigned long start;
 	struct page *control_code_page;
+	struct page *swap_page;
 
 	unsigned long nr_segments;
 	struct kexec_segment segment[KEXEC_SEGMENT_MAX];
@@ -98,18 +99,20 @@ struct kimage {
 	unsigned int type : 1;
 #define KEXEC_TYPE_DEFAULT 0
 #define KEXEC_TYPE_CRASH   1
+	unsigned int preserve_context : 1;
 };
 
 
 
 /* kexec interface functions */
-extern NORET_TYPE void machine_kexec(struct kimage *image) ATTRIB_NORET;
+extern void machine_kexec(struct kimage *image);
 extern int machine_kexec_prepare(struct kimage *image);
 extern void machine_kexec_cleanup(struct kimage *image);
 extern asmlinkage long sys_kexec_load(unsigned long entry,
 					unsigned long nr_segments,
 					struct kexec_segment __user *segments,
 					unsigned long flags);
+extern int kernel_kexec(void);
 #ifdef CONFIG_COMPAT
 extern asmlinkage long compat_sys_kexec_load(unsigned long entry,
 				unsigned long nr_segments,
@@ -156,8 +159,9 @@ extern struct kimage *kexec_crash_image;
 #define kexec_flush_icache_page(page)
 #endif
 
-#define KEXEC_ON_CRASH  0x00000001
-#define KEXEC_ARCH_MASK 0xffff0000
+#define KEXEC_ON_CRASH		0x00000001
+#define KEXEC_PRESERVE_CONTEXT	0x00000002
+#define KEXEC_ARCH_MASK		0xffff0000
 
 /* These values match the ELF architecture values.
  * Unless there is a good reason that should continue to be the case.
@@ -174,7 +178,12 @@ extern struct kimage *kexec_crash_image;
 #define KEXEC_ARCH_MIPS_LE (10 << 16)
 #define KEXEC_ARCH_MIPS    ( 8 << 16)
 
-#define KEXEC_FLAGS    (KEXEC_ON_CRASH)  /* List of defined/legal kexec flags */
+/* List of defined/legal kexec flags */
+#ifndef CONFIG_KEXEC_JUMP
+#define KEXEC_FLAGS    KEXEC_ON_CRASH
+#else
+#define KEXEC_FLAGS    (KEXEC_ON_CRASH | KEXEC_PRESERVE_CONTEXT)
+#endif
 
 #define VMCOREINFO_BYTES           (4096)
 #define VMCOREINFO_NOTE_NAME       "VMCOREINFO"
diff --git a/kernel/kexec.c b/kernel/kexec.c
index 6db42ff8d520..a0d920915b38 100644
--- a/kernel/kexec.c
+++ b/kernel/kexec.c
@@ -24,6 +24,8 @@
 #include <linux/utsrelease.h>
 #include <linux/utsname.h>
 #include <linux/numa.h>
+#include <linux/suspend.h>
+#include <linux/device.h>
 
 #include <asm/page.h>
 #include <asm/uaccess.h>
@@ -242,6 +244,12 @@ static int kimage_normal_alloc(struct kimage **rimage, unsigned long entry,
 		goto out;
 	}
 
+	image->swap_page = kimage_alloc_control_pages(image, 0);
+	if (!image->swap_page) {
+		printk(KERN_ERR "Could not allocate swap buffer\n");
+		goto out;
+	}
+
 	result = 0;
  out:
 	if (result == 0)
@@ -986,6 +994,8 @@ asmlinkage long sys_kexec_load(unsigned long entry, unsigned long nr_segments,
 		if (result)
 			goto out;
 
+		if (flags & KEXEC_PRESERVE_CONTEXT)
+			image->preserve_context = 1;
 		result = machine_kexec_prepare(image);
 		if (result)
 			goto out;
@@ -1411,3 +1421,50 @@ static int __init crash_save_vmcoreinfo_init(void)
 }
 
 module_init(crash_save_vmcoreinfo_init)
+
+/**
+ *	kernel_kexec - reboot the system
+ *
+ *	Move into place and start executing a preloaded standalone
+ *	executable.  If nothing was preloaded return an error.
+ */
+int kernel_kexec(void)
+{
+	int error = 0;
+
+	if (xchg(&kexec_lock, 1))
+		return -EBUSY;
+	if (!kexec_image) {
+		error = -EINVAL;
+		goto Unlock;
+	}
+
+	if (kexec_image->preserve_context) {
+#ifdef CONFIG_KEXEC_JUMP
+		local_irq_disable();
+		save_processor_state();
+#endif
+	} else {
+		blocking_notifier_call_chain(&reboot_notifier_list,
+					     SYS_RESTART, NULL);
+		system_state = SYSTEM_RESTART;
+		device_shutdown();
+		sysdev_shutdown();
+		printk(KERN_EMERG "Starting new kernel\n");
+		machine_shutdown();
+	}
+
+	machine_kexec(kexec_image);
+
+	if (kexec_image->preserve_context) {
+#ifdef CONFIG_KEXEC_JUMP
+		restore_processor_state();
+		local_irq_enable();
+#endif
+	}
+
+ Unlock:
+	xchg(&kexec_lock, 0);
+
+	return error;
+}
diff --git a/kernel/sys.c b/kernel/sys.c
index 0c9d3fa1f5ff..c01858090a98 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -301,26 +301,6 @@ void kernel_restart(char *cmd)
 }
 EXPORT_SYMBOL_GPL(kernel_restart);
 
-/**
- *	kernel_kexec - reboot the system
- *
- *	Move into place and start executing a preloaded standalone
- *	executable.  If nothing was preloaded return an error.
- */
-static void kernel_kexec(void)
-{
-#ifdef CONFIG_KEXEC
-	struct kimage *image;
-	image = xchg(&kexec_image, NULL);
-	if (!image)
-		return;
-	kernel_restart_prepare(NULL);
-	printk(KERN_EMERG "Starting new kernel\n");
-	machine_shutdown();
-	machine_kexec(image);
-#endif
-}
-
 static void kernel_shutdown_prepare(enum system_states state)
 {
 	blocking_notifier_call_chain(&reboot_notifier_list,
@@ -425,10 +405,15 @@ asmlinkage long sys_reboot(int magic1, int magic2, unsigned int cmd, void __user
 		kernel_restart(buffer);
 		break;
 
+#ifdef CONFIG_KEXEC
 	case LINUX_REBOOT_CMD_KEXEC:
-		kernel_kexec();
-		unlock_kernel();
-		return -EINVAL;
+		{
+			int ret;
+			ret = kernel_kexec();
+			unlock_kernel();
+			return ret;
+		}
+#endif
 
 #ifdef CONFIG_HIBERNATION
 	case LINUX_REBOOT_CMD_SW_SUSPEND:

From 89081d17f7bb81d89fa1aa9b70f821c5cf4d39e9 Mon Sep 17 00:00:00 2001
From: Huang Ying <ying.huang@intel.com>
Date: Fri, 25 Jul 2008 19:45:10 -0700
Subject: [PATCH 1009/1171] kexec jump: save/restore device state

This patch implements devices state save/restore before after kexec.

This patch together with features in kexec_jump patch can be used for
following:

- A simple hibernation implementation without ACPI support.  You can kexec a
  hibernating kernel, save the memory image of original system and shutdown
  the system.  When resuming, you restore the memory image of original system
  via ordinary kexec load then jump back.

- Kernel/system debug through making system snapshot.  You can make system
  snapshot, jump back, do some thing and make another system snapshot.

- Cooperative multi-kernel/system.  With kexec jump, you can switch between
  several kernels/systems quickly without boot process except the first time.
  This appears like swap a whole kernel/system out/in.

- A general method to call program in physical mode (paging turning
  off). This can be used to invoke BIOS code under Linux.

The following user-space tools can be used with kexec jump:

- kexec-tools needs to be patched to support kexec jump. The patches
  and the precompiled kexec can be download from the following URL:
       source: http://khibernation.sourceforge.net/download/release_v10/kexec-tools/kexec-tools-src_git_kh10.tar.bz2
       patches: http://khibernation.sourceforge.net/download/release_v10/kexec-tools/kexec-tools-patches_git_kh10.tar.bz2
       binary: http://khibernation.sourceforge.net/download/release_v10/kexec-tools/kexec_git_kh10

- makedumpfile with patches are used as memory image saving tool, it
  can exclude free pages from original kernel memory image file. The
  patches and the precompiled makedumpfile can be download from the
  following URL:
       source: http://khibernation.sourceforge.net/download/release_v10/makedumpfile/makedumpfile-src_cvs_kh10.tar.bz2
       patches: http://khibernation.sourceforge.net/download/release_v10/makedumpfile/makedumpfile-patches_cvs_kh10.tar.bz2
       binary: http://khibernation.sourceforge.net/download/release_v10/makedumpfile/makedumpfile_cvs_kh10

- An initramfs image can be used as the root file system of kexeced
  kernel. An initramfs image built with "BuildRoot" can be downloaded
  from the following URL:
       initramfs image: http://khibernation.sourceforge.net/download/release_v10/initramfs/rootfs_cvs_kh10.gz
  All user space tools above are included in the initramfs image.

Usage example of simple hibernation:

1. Compile and install patched kernel with following options selected:

CONFIG_X86_32=y
CONFIG_RELOCATABLE=y
CONFIG_KEXEC=y
CONFIG_CRASH_DUMP=y
CONFIG_PM=y
CONFIG_HIBERNATION=y
CONFIG_KEXEC_JUMP=y

2. Build an initramfs image contains kexec-tool and makedumpfile, or
   download the pre-built initramfs image, called rootfs.gz in
   following text.

3. Prepare a partition to save memory image of original kernel, called
   hibernating partition in following text.

4. Boot kernel compiled in step 1 (kernel A).

5. In the kernel A, load kernel compiled in step 1 (kernel B) with
   /sbin/kexec. The shell command line can be as follow:

   /sbin/kexec --load-preserve-context /boot/bzImage --mem-min=0x100000
     --mem-max=0xffffff --initrd=rootfs.gz

6. Boot the kernel B with following shell command line:

   /sbin/kexec -e

7. The kernel B will boot as normal kexec. In kernel B the memory
   image of kernel A can be saved into hibernating partition as
   follow:

   jump_back_entry=`cat /proc/cmdline | tr ' ' '\n' | grep kexec_jump_back_entry | cut -d '='`
   echo $jump_back_entry > kexec_jump_back_entry
   cp /proc/vmcore dump.elf

   Then you can shutdown the machine as normal.

8. Boot kernel compiled in step 1 (kernel C). Use the rootfs.gz as
   root file system.

9. In kernel C, load the memory image of kernel A as follow:

   /sbin/kexec -l --args-none --entry=`cat kexec_jump_back_entry` dump.elf

10. Jump back to the kernel A as follow:

   /sbin/kexec -e

   Then, kernel A is resumed.

Implementation point:

To support jumping between two kernels, before jumping to (executing)
the new kernel and jumping back to the original kernel, the devices
are put into quiescent state, and the state of devices and CPU is
saved. After jumping back from kexeced kernel and jumping to the new
kernel, the state of devices and CPU are restored accordingly. The
devices/CPU state save/restore code of software suspend is called to
implement corresponding function.

Known issues:

- Because the segment number supported by sys_kexec_load is limited,
  hibernation image with many segments may not be load. This is
  planned to be eliminated by adding a new flag to sys_kexec_load to
  make a image can be loaded with multiple sys_kexec_load invoking.

Now, only the i386 architecture is supported.

Signed-off-by: Huang Ying <ying.huang@intel.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Pavel Machek <pavel@ucw.cz>
Cc: Nigel Cunningham <nigel@nigel.suspend2.net>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/Kconfig                   |  5 ++--
 arch/x86/kernel/machine_kexec_32.c | 12 +++++++++
 include/linux/suspend.h            |  2 ++
 kernel/kexec.c                     | 39 ++++++++++++++++++++++++++++++
 kernel/power/power.h               |  2 --
 5 files changed, 56 insertions(+), 4 deletions(-)

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 7ecb679f0130..6b2debfabddc 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -1282,9 +1282,10 @@ config CRASH_DUMP
 config KEXEC_JUMP
 	bool "kexec jump (EXPERIMENTAL)"
 	depends on EXPERIMENTAL
-	depends on KEXEC && PM_SLEEP && X86_32
+	depends on KEXEC && HIBERNATION && X86_32
 	help
-	  Invoke code in physical address mode via KEXEC
+	  Jump between original kernel and kexeced kernel and invoke
+	  code in physical address mode via KEXEC
 
 config PHYSICAL_START
 	hex "Physical address where the kernel is loaded" if (EMBEDDED || CRASH_DUMP)
diff --git a/arch/x86/kernel/machine_kexec_32.c b/arch/x86/kernel/machine_kexec_32.c
index 2b67609d0a1c..9fe478d98406 100644
--- a/arch/x86/kernel/machine_kexec_32.c
+++ b/arch/x86/kernel/machine_kexec_32.c
@@ -125,6 +125,18 @@ void machine_kexec(struct kimage *image)
 	/* Interrupts aren't acceptable while we reboot */
 	local_irq_disable();
 
+	if (image->preserve_context) {
+#ifdef CONFIG_X86_IO_APIC
+		/* We need to put APICs in legacy mode so that we can
+		 * get timer interrupts in second kernel. kexec/kdump
+		 * paths already have calls to disable_IO_APIC() in
+		 * one form or other. kexec jump path also need
+		 * one.
+		 */
+		disable_IO_APIC();
+#endif
+	}
+
 	control_page = page_address(image->control_code_page);
 	memcpy(control_page, relocate_kernel, PAGE_SIZE/2);
 
diff --git a/include/linux/suspend.h b/include/linux/suspend.h
index e8e69159af71..c63435095970 100644
--- a/include/linux/suspend.h
+++ b/include/linux/suspend.h
@@ -278,4 +278,6 @@ static inline void register_nosave_region_late(unsigned long b, unsigned long e)
 }
 #endif
 
+extern struct mutex pm_mutex;
+
 #endif /* _LINUX_SUSPEND_H */
diff --git a/kernel/kexec.c b/kernel/kexec.c
index a0d920915b38..c8a4370e2a34 100644
--- a/kernel/kexec.c
+++ b/kernel/kexec.c
@@ -26,6 +26,10 @@
 #include <linux/numa.h>
 #include <linux/suspend.h>
 #include <linux/device.h>
+#include <linux/freezer.h>
+#include <linux/pm.h>
+#include <linux/cpu.h>
+#include <linux/console.h>
 
 #include <asm/page.h>
 #include <asm/uaccess.h>
@@ -1441,7 +1445,31 @@ int kernel_kexec(void)
 
 	if (kexec_image->preserve_context) {
 #ifdef CONFIG_KEXEC_JUMP
+		mutex_lock(&pm_mutex);
+		pm_prepare_console();
+		error = freeze_processes();
+		if (error) {
+			error = -EBUSY;
+			goto Restore_console;
+		}
+		suspend_console();
+		error = device_suspend(PMSG_FREEZE);
+		if (error)
+			goto Resume_console;
+		error = disable_nonboot_cpus();
+		if (error)
+			goto Resume_devices;
 		local_irq_disable();
+		/* At this point, device_suspend() has been called,
+		 * but *not* device_power_down(). We *must*
+		 * device_power_down() now.  Otherwise, drivers for
+		 * some devices (e.g. interrupt controllers) become
+		 * desynchronized with the actual state of the
+		 * hardware at resume time, and evil weirdness ensues.
+		 */
+		error = device_power_down(PMSG_FREEZE);
+		if (error)
+			goto Enable_irqs;
 		save_processor_state();
 #endif
 	} else {
@@ -1459,7 +1487,18 @@ int kernel_kexec(void)
 	if (kexec_image->preserve_context) {
 #ifdef CONFIG_KEXEC_JUMP
 		restore_processor_state();
+		device_power_up(PMSG_RESTORE);
+ Enable_irqs:
 		local_irq_enable();
+		enable_nonboot_cpus();
+ Resume_devices:
+		device_resume(PMSG_RESTORE);
+ Resume_console:
+		resume_console();
+		thaw_processes();
+ Restore_console:
+		pm_restore_console();
+		mutex_unlock(&pm_mutex);
 #endif
 	}
 
diff --git a/kernel/power/power.h b/kernel/power/power.h
index 700f44ec8406..acc0c101dbd5 100644
--- a/kernel/power/power.h
+++ b/kernel/power/power.h
@@ -53,8 +53,6 @@ extern int hibernation_platform_enter(void);
 
 extern int pfn_is_nosave(unsigned long);
 
-extern struct mutex pm_mutex;
-
 #define power_attr(_name) \
 static struct kobj_attribute _name##_attr = {	\
 	.attr	= {				\

From c2147a5092cfe13dbf3210e54e8a622015edeecc Mon Sep 17 00:00:00 2001
From: Eduard - Gabriel Munteanu <eduard.munteanu@linux360.ro>
Date: Fri, 25 Jul 2008 19:45:11 -0700
Subject: [PATCH 1010/1171] Better interface for hooking early initcalls

Added early initcall (pre-SMP) support, using an identical interface to
that of regular initcalls.  Functions called from do_pre_smp_initcalls()
could be converted to use this cleaner interface.

This is required by CPU hotplug, because early users have to register
notifiers before going SMP.  One such CPU hotplug user is the relay
interface with buffer-only channels, which needs to register such a
notifier, to be usable in early code.  This in turn is used by kmemtrace.

Signed-off-by: Eduard - Gabriel Munteanu <eduard.munteanu@linux360.ro>
Cc: Tom Zanussi <tzanussi@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/asm-generic/vmlinux.lds.h |  2 ++
 include/linux/init.h              |  7 +++++++
 init/main.c                       | 13 +++++++++++--
 3 files changed, 20 insertions(+), 2 deletions(-)

diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h
index 729f6b0a60e9..9cd44b162ba1 100644
--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -359,6 +359,8 @@
 	}
 
 #define INITCALLS							\
+	*(.initcallearly.init)						\
+	__early_initcall_end = .;					\
   	*(.initcall0.init)						\
   	*(.initcall0s.init)						\
   	*(.initcall1.init)						\
diff --git a/include/linux/init.h b/include/linux/init.h
index 42ae95411a93..11b84e106053 100644
--- a/include/linux/init.h
+++ b/include/linux/init.h
@@ -169,6 +169,13 @@ extern void (*late_time_init)(void);
 	static initcall_t __initcall_##fn##id __used \
 	__attribute__((__section__(".initcall" level ".init"))) = fn
 
+/*
+ * Early initcalls run before initializing SMP.
+ *
+ * Only for built-in code, not modules.
+ */
+#define early_initcall(fn)		__define_initcall("early",fn,early)
+
 /*
  * A "pure" initcall has no dependencies on anything else, and purely
  * initializes variables that couldn't be statically initialized.
diff --git a/init/main.c b/init/main.c
index 0604cbcaf1e4..b6fec08dbbef 100644
--- a/init/main.c
+++ b/init/main.c
@@ -743,13 +743,13 @@ static void __init do_one_initcall(initcall_t fn)
 }
 
 
-extern initcall_t __initcall_start[], __initcall_end[];
+extern initcall_t __initcall_start[], __initcall_end[], __early_initcall_end[];
 
 static void __init do_initcalls(void)
 {
 	initcall_t *call;
 
-	for (call = __initcall_start; call < __initcall_end; call++)
+	for (call = __early_initcall_end; call < __initcall_end; call++)
 		do_one_initcall(*call);
 
 	/* Make sure there is no pending stuff from the initcall sequence */
@@ -783,6 +783,14 @@ static int __init nosoftlockup_setup(char *str)
 }
 __setup("nosoftlockup", nosoftlockup_setup);
 
+static void __init __do_pre_smp_initcalls(void)
+{
+	initcall_t *call;
+
+	for (call = __initcall_start; call < __early_initcall_end; call++)
+		do_one_initcall(*call);
+}
+
 static void __init do_pre_smp_initcalls(void)
 {
 	extern int spawn_ksoftirqd(void);
@@ -865,6 +873,7 @@ static int __init kernel_init(void * unused)
 
 	smp_prepare_cpus(setup_max_cpus);
 
+	__do_pre_smp_initcalls();
 	do_pre_smp_initcalls();
 
 	smp_init();

From 7babe8db99d305340cf4828ce1f5a1481d5622ef Mon Sep 17 00:00:00 2001
From: Eduard - Gabriel Munteanu <eduard.munteanu@linux360.ro>
Date: Fri, 25 Jul 2008 19:45:11 -0700
Subject: [PATCH 1011/1171] Full conversion to early_initcall() interface,
 remove old interface

A previous patch added the early_initcall(), to allow a cleaner hooking of
pre-SMP initcalls.  Now we remove the older interface, converting all
existing users to the new one.

[akpm@linux-foundation.org: cleanups]
[akpm@linux-foundation.org: build fix]
[kosaki.motohiro@jp.fujitsu.com: warning fix]
[kosaki.motohiro@jp.fujitsu.com: warning fix]
Signed-off-by: Eduard - Gabriel Munteanu <eduard.munteanu@linux360.ro>
Cc: Tom Zanussi <tzanussi@gmail.com>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/sched.h |  9 ---------
 include/linux/smp.h   |  5 -----
 init/main.c           | 23 +----------------------
 kernel/sched.c        |  5 ++++-
 kernel/smp.c          |  4 +++-
 kernel/softirq.c      |  3 ++-
 kernel/softlockup.c   | 25 ++++++++++++++++++++++---
 7 files changed, 32 insertions(+), 42 deletions(-)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 3260a5c42b91..adb8077dc463 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -292,7 +292,6 @@ extern void sched_show_task(struct task_struct *p);
 
 #ifdef CONFIG_DETECT_SOFTLOCKUP
 extern void softlockup_tick(void);
-extern void spawn_softlockup_task(void);
 extern void touch_softlockup_watchdog(void);
 extern void touch_all_softlockup_watchdogs(void);
 extern unsigned int  softlockup_panic;
@@ -2222,14 +2221,6 @@ static inline void inc_syscw(struct task_struct *tsk)
 }
 #endif
 
-#ifdef CONFIG_SMP
-void migration_init(void);
-#else
-static inline void migration_init(void)
-{
-}
-#endif
-
 #ifndef TASK_SIZE_OF
 #define TASK_SIZE_OF(tsk)	TASK_SIZE
 #endif
diff --git a/include/linux/smp.h b/include/linux/smp.h
index 48262f86c969..66484d4a8459 100644
--- a/include/linux/smp.h
+++ b/include/linux/smp.h
@@ -74,15 +74,10 @@ void __smp_call_function_single(int cpuid, struct call_single_data *data);
 #ifdef CONFIG_USE_GENERIC_SMP_HELPERS
 void generic_smp_call_function_single_interrupt(void);
 void generic_smp_call_function_interrupt(void);
-void init_call_single_data(void);
 void ipi_call_lock(void);
 void ipi_call_unlock(void);
 void ipi_call_lock_irq(void);
 void ipi_call_unlock_irq(void);
-#else
-static inline void init_call_single_data(void)
-{
-}
 #endif
 
 /*
diff --git a/init/main.c b/init/main.c
index b6fec08dbbef..20fdc9884b77 100644
--- a/init/main.c
+++ b/init/main.c
@@ -774,16 +774,7 @@ static void __init do_basic_setup(void)
 	do_initcalls();
 }
 
-static int __initdata nosoftlockup;
-
-static int __init nosoftlockup_setup(char *str)
-{
-	nosoftlockup = 1;
-	return 1;
-}
-__setup("nosoftlockup", nosoftlockup_setup);
-
-static void __init __do_pre_smp_initcalls(void)
+static void __init do_pre_smp_initcalls(void)
 {
 	initcall_t *call;
 
@@ -791,17 +782,6 @@ static void __init __do_pre_smp_initcalls(void)
 		do_one_initcall(*call);
 }
 
-static void __init do_pre_smp_initcalls(void)
-{
-	extern int spawn_ksoftirqd(void);
-
-	init_call_single_data();
-	migration_init();
-	spawn_ksoftirqd();
-	if (!nosoftlockup)
-		spawn_softlockup_task();
-}
-
 static void run_init_process(char *init_filename)
 {
 	argv_init[0] = init_filename;
@@ -873,7 +853,6 @@ static int __init kernel_init(void * unused)
 
 	smp_prepare_cpus(setup_max_cpus);
 
-	__do_pre_smp_initcalls();
 	do_pre_smp_initcalls();
 
 	smp_init();
diff --git a/kernel/sched.c b/kernel/sched.c
index 0047bd9b96aa..fde1a1026359 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -6389,7 +6389,7 @@ static struct notifier_block __cpuinitdata migration_notifier = {
 	.priority = 10
 };
 
-void __init migration_init(void)
+static int __init migration_init(void)
 {
 	void *cpu = (void *)(long)smp_processor_id();
 	int err;
@@ -6399,7 +6399,10 @@ void __init migration_init(void)
 	BUG_ON(err == NOTIFY_BAD);
 	migration_call(&migration_notifier, CPU_ONLINE, cpu);
 	register_cpu_notifier(&migration_notifier);
+
+	return err;
 }
+early_initcall(migration_init);
 #endif
 
 #ifdef CONFIG_SMP
diff --git a/kernel/smp.c b/kernel/smp.c
index 462c785ca1ee..96fc7c0edc59 100644
--- a/kernel/smp.c
+++ b/kernel/smp.c
@@ -33,7 +33,7 @@ struct call_single_queue {
 	spinlock_t lock;
 };
 
-void __cpuinit init_call_single_data(void)
+static int __cpuinit init_call_single_data(void)
 {
 	int i;
 
@@ -43,7 +43,9 @@ void __cpuinit init_call_single_data(void)
 		spin_lock_init(&q->lock);
 		INIT_LIST_HEAD(&q->list);
 	}
+	return 0;
 }
+early_initcall(init_call_single_data);
 
 static void csd_flag_wait(struct call_single_data *data)
 {
diff --git a/kernel/softirq.c b/kernel/softirq.c
index f6b03d56c2bf..c506f266a6b9 100644
--- a/kernel/softirq.c
+++ b/kernel/softirq.c
@@ -630,7 +630,7 @@ static struct notifier_block __cpuinitdata cpu_nfb = {
 	.notifier_call = cpu_callback
 };
 
-__init int spawn_ksoftirqd(void)
+static __init int spawn_ksoftirqd(void)
 {
 	void *cpu = (void *)(long)smp_processor_id();
 	int err = cpu_callback(&cpu_nfb, CPU_UP_PREPARE, cpu);
@@ -640,6 +640,7 @@ __init int spawn_ksoftirqd(void)
 	register_cpu_notifier(&cpu_nfb);
 	return 0;
 }
+early_initcall(spawn_ksoftirqd);
 
 #ifdef CONFIG_SMP
 /*
diff --git a/kernel/softlockup.c b/kernel/softlockup.c
index 7bd8d1aadd5d..b75b492fbfcf 100644
--- a/kernel/softlockup.c
+++ b/kernel/softlockup.c
@@ -338,14 +338,33 @@ static struct notifier_block __cpuinitdata cpu_nfb = {
 	.notifier_call = cpu_callback
 };
 
-__init void spawn_softlockup_task(void)
+static int __initdata nosoftlockup;
+
+static int __init nosoftlockup_setup(char *str)
+{
+	nosoftlockup = 1;
+	return 1;
+}
+__setup("nosoftlockup", nosoftlockup_setup);
+
+static int __init spawn_softlockup_task(void)
 {
 	void *cpu = (void *)(long)smp_processor_id();
-	int err = cpu_callback(&cpu_nfb, CPU_UP_PREPARE, cpu);
+	int err;
 
-	BUG_ON(err == NOTIFY_BAD);
+	if (nosoftlockup)
+		return 0;
+
+	err = cpu_callback(&cpu_nfb, CPU_UP_PREPARE, cpu);
+	if (err == NOTIFY_BAD) {
+		BUG();
+		return 1;
+	}
 	cpu_callback(&cpu_nfb, CPU_ONLINE, cpu);
 	register_cpu_notifier(&cpu_nfb);
 
 	atomic_notifier_chain_register(&panic_notifier_list, &panic_block);
+
+	return 0;
 }
+early_initcall(spawn_softlockup_task);

From 20d8b67c06fa5e74f44e80b0a0fd68c8327f7c6a Mon Sep 17 00:00:00 2001
From: Eduard - Gabriel Munteanu <eduard.munteanu@linux360.ro>
Date: Fri, 25 Jul 2008 19:45:12 -0700
Subject: [PATCH 1012/1171] relay: add buffer-only channels; useful for early
 logging

Allows one to create and use a channel with no associated files.  Files
can be initialized later.  This is useful in scenarios such as logging in
early code, before VFS is up.  Therefore, such channels can be created and
used as soon as kmem_cache_init() completed.

This is needed by kmemtrace to do tracing in early kernel code.

[kosaki.motohiro@jp.fujitsu.com: build fix]
Signed-off-by: Eduard - Gabriel Munteanu <eduard.munteanu@linux360.ro>
Cc: Tom Zanussi <tzanussi@gmail.com>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/filesystems/relay.txt |  10 ++
 include/linux/relay.h               |   5 +
 kernel/relay.c                      | 170 +++++++++++++++++++++++-----
 3 files changed, 156 insertions(+), 29 deletions(-)

diff --git a/Documentation/filesystems/relay.txt b/Documentation/filesystems/relay.txt
index 094f2d2f38b1..510b722667ac 100644
--- a/Documentation/filesystems/relay.txt
+++ b/Documentation/filesystems/relay.txt
@@ -294,6 +294,16 @@ user-defined data with a channel, and is immediately available
 (including in create_buf_file()) via chan->private_data or
 buf->chan->private_data.
 
+Buffer-only channels
+--------------------
+
+These channels have no files associated and can be created with
+relay_open(NULL, NULL, ...). Such channels are useful in scenarios such
+as when doing early tracing in the kernel, before the VFS is up. In these
+cases, one may open a buffer-only channel and then call
+relay_late_setup_files() when the kernel is ready to handle files,
+to expose the buffered data to the userspace.
+
 Channel 'modes'
 ---------------
 
diff --git a/include/linux/relay.h b/include/linux/relay.h
index 6cd8c4425fc7..953fc055e875 100644
--- a/include/linux/relay.h
+++ b/include/linux/relay.h
@@ -48,6 +48,7 @@ struct rchan_buf
 	size_t *padding;		/* padding counts per sub-buffer */
 	size_t prev_padding;		/* temporary variable */
 	size_t bytes_consumed;		/* bytes consumed in cur read subbuf */
+	size_t early_bytes;		/* bytes consumed before VFS inited */
 	unsigned int cpu;		/* this buf's cpu */
 } ____cacheline_aligned;
 
@@ -68,6 +69,7 @@ struct rchan
 	int is_global;			/* One global buffer ? */
 	struct list_head list;		/* for channel list */
 	struct dentry *parent;		/* parent dentry passed to open */
+	int has_base_filename;		/* has a filename associated? */
 	char base_filename[NAME_MAX];	/* saved base filename */
 };
 
@@ -169,6 +171,9 @@ struct rchan *relay_open(const char *base_filename,
 			 size_t n_subbufs,
 			 struct rchan_callbacks *cb,
 			 void *private_data);
+extern int relay_late_setup_files(struct rchan *chan,
+				  const char *base_filename,
+				  struct dentry *parent);
 extern void relay_close(struct rchan *chan);
 extern void relay_flush(struct rchan *chan);
 extern void relay_subbufs_consumed(struct rchan *chan,
diff --git a/kernel/relay.c b/kernel/relay.c
index 7de644cdec43..04006ef970b8 100644
--- a/kernel/relay.c
+++ b/kernel/relay.c
@@ -407,6 +407,35 @@ void relay_reset(struct rchan *chan)
 }
 EXPORT_SYMBOL_GPL(relay_reset);
 
+static inline void relay_set_buf_dentry(struct rchan_buf *buf,
+					struct dentry *dentry)
+{
+	buf->dentry = dentry;
+	buf->dentry->d_inode->i_size = buf->early_bytes;
+}
+
+static struct dentry *relay_create_buf_file(struct rchan *chan,
+					    struct rchan_buf *buf,
+					    unsigned int cpu)
+{
+	struct dentry *dentry;
+	char *tmpname;
+
+	tmpname = kzalloc(NAME_MAX + 1, GFP_KERNEL);
+	if (!tmpname)
+		return NULL;
+	snprintf(tmpname, NAME_MAX, "%s%d", chan->base_filename, cpu);
+
+	/* Create file in fs */
+	dentry = chan->cb->create_buf_file(tmpname, chan->parent,
+					   S_IRUSR, buf,
+					   &chan->is_global);
+
+	kfree(tmpname);
+
+	return dentry;
+}
+
 /*
  *	relay_open_buf - create a new relay channel buffer
  *
@@ -416,45 +445,34 @@ static struct rchan_buf *relay_open_buf(struct rchan *chan, unsigned int cpu)
 {
  	struct rchan_buf *buf = NULL;
 	struct dentry *dentry;
- 	char *tmpname;
 
  	if (chan->is_global)
 		return chan->buf[0];
 
-	tmpname = kzalloc(NAME_MAX + 1, GFP_KERNEL);
- 	if (!tmpname)
- 		goto end;
- 	snprintf(tmpname, NAME_MAX, "%s%d", chan->base_filename, cpu);
-
 	buf = relay_create_buf(chan);
 	if (!buf)
- 		goto free_name;
+		return NULL;
+
+	if (chan->has_base_filename) {
+		dentry = relay_create_buf_file(chan, buf, cpu);
+		if (!dentry)
+			goto free_buf;
+		relay_set_buf_dentry(buf, dentry);
+	}
 
  	buf->cpu = cpu;
  	__relay_reset(buf, 1);
 
-	/* Create file in fs */
- 	dentry = chan->cb->create_buf_file(tmpname, chan->parent, S_IRUSR,
- 					   buf, &chan->is_global);
- 	if (!dentry)
- 		goto free_buf;
-
-	buf->dentry = dentry;
-
  	if(chan->is_global) {
  		chan->buf[0] = buf;
  		buf->cpu = 0;
   	}
 
- 	goto free_name;
+	return buf;
 
 free_buf:
  	relay_destroy_buf(buf);
- 	buf = NULL;
-free_name:
- 	kfree(tmpname);
-end:
-	return buf;
+	return NULL;
 }
 
 /**
@@ -537,8 +555,8 @@ static int __cpuinit relay_hotcpu_callback(struct notifier_block *nb,
 
 /**
  *	relay_open - create a new relay channel
- *	@base_filename: base name of files to create
- *	@parent: dentry of parent directory, %NULL for root directory
+ *	@base_filename: base name of files to create, %NULL for buffering only
+ *	@parent: dentry of parent directory, %NULL for root directory or buffer
  *	@subbuf_size: size of sub-buffers
  *	@n_subbufs: number of sub-buffers
  *	@cb: client callback functions
@@ -560,8 +578,6 @@ struct rchan *relay_open(const char *base_filename,
 {
 	unsigned int i;
 	struct rchan *chan;
-	if (!base_filename)
-		return NULL;
 
 	if (!(subbuf_size && n_subbufs))
 		return NULL;
@@ -576,7 +592,10 @@ struct rchan *relay_open(const char *base_filename,
 	chan->alloc_size = FIX_SIZE(subbuf_size * n_subbufs);
 	chan->parent = parent;
 	chan->private_data = private_data;
-	strlcpy(chan->base_filename, base_filename, NAME_MAX);
+	if (base_filename) {
+		chan->has_base_filename = 1;
+		strlcpy(chan->base_filename, base_filename, NAME_MAX);
+	}
 	setup_callbacks(chan, cb);
 	kref_init(&chan->kref);
 
@@ -604,6 +623,94 @@ free_bufs:
 }
 EXPORT_SYMBOL_GPL(relay_open);
 
+struct rchan_percpu_buf_dispatcher {
+	struct rchan_buf *buf;
+	struct dentry *dentry;
+};
+
+/* Called in atomic context. */
+static void __relay_set_buf_dentry(void *info)
+{
+	struct rchan_percpu_buf_dispatcher *p = info;
+
+	relay_set_buf_dentry(p->buf, p->dentry);
+}
+
+/**
+ *	relay_late_setup_files - triggers file creation
+ *	@chan: channel to operate on
+ *	@base_filename: base name of files to create
+ *	@parent: dentry of parent directory, %NULL for root directory
+ *
+ *	Returns 0 if successful, non-zero otherwise.
+ *
+ *	Use to setup files for a previously buffer-only channel.
+ *	Useful to do early tracing in kernel, before VFS is up, for example.
+ */
+int relay_late_setup_files(struct rchan *chan,
+			   const char *base_filename,
+			   struct dentry *parent)
+{
+	int err = 0;
+	unsigned int i, curr_cpu;
+	unsigned long flags;
+	struct dentry *dentry;
+	struct rchan_percpu_buf_dispatcher disp;
+
+	if (!chan || !base_filename)
+		return -EINVAL;
+
+	strlcpy(chan->base_filename, base_filename, NAME_MAX);
+
+	mutex_lock(&relay_channels_mutex);
+	/* Is chan already set up? */
+	if (unlikely(chan->has_base_filename))
+		return -EEXIST;
+	chan->has_base_filename = 1;
+	chan->parent = parent;
+	curr_cpu = get_cpu();
+	/*
+	 * The CPU hotplug notifier ran before us and created buffers with
+	 * no files associated. So it's safe to call relay_setup_buf_file()
+	 * on all currently online CPUs.
+	 */
+	for_each_online_cpu(i) {
+		if (unlikely(!chan->buf[i])) {
+			printk(KERN_ERR "relay_late_setup_files: CPU %u "
+					"has no buffer, it must have!\n", i);
+			BUG();
+			err = -EINVAL;
+			break;
+		}
+
+		dentry = relay_create_buf_file(chan, chan->buf[i], i);
+		if (unlikely(!dentry)) {
+			err = -EINVAL;
+			break;
+		}
+
+		if (curr_cpu == i) {
+			local_irq_save(flags);
+			relay_set_buf_dentry(chan->buf[i], dentry);
+			local_irq_restore(flags);
+		} else {
+			disp.buf = chan->buf[i];
+			disp.dentry = dentry;
+			smp_mb();
+			/* relay_channels_mutex must be held, so wait. */
+			err = smp_call_function_single(i,
+						       __relay_set_buf_dentry,
+						       &disp, 1);
+		}
+		if (unlikely(err))
+			break;
+	}
+	put_cpu();
+	mutex_unlock(&relay_channels_mutex);
+
+	return err;
+}
+
 /**
  *	relay_switch_subbuf - switch to a new sub-buffer
  *	@buf: channel buffer
@@ -627,8 +734,13 @@ size_t relay_switch_subbuf(struct rchan_buf *buf, size_t length)
 		old_subbuf = buf->subbufs_produced % buf->chan->n_subbufs;
 		buf->padding[old_subbuf] = buf->prev_padding;
 		buf->subbufs_produced++;
-		buf->dentry->d_inode->i_size += buf->chan->subbuf_size -
-			buf->padding[old_subbuf];
+		if (buf->dentry)
+			buf->dentry->d_inode->i_size +=
+				buf->chan->subbuf_size -
+				buf->padding[old_subbuf];
+		else
+			buf->early_bytes += buf->chan->subbuf_size -
+					    buf->padding[old_subbuf];
 		smp_mb();
 		if (waitqueue_active(&buf->read_wait))
 			/*
@@ -1237,4 +1349,4 @@ static __init int relay_init(void)
 	return 0;
 }
 
-module_init(relay_init);
+early_initcall(relay_init);

From 080ccd4573607a930367c2128fc709814b2ade5d Mon Sep 17 00:00:00 2001
From: Huang Weiyi <weiyi.huang@gmail.com>
Date: Fri, 25 Jul 2008 19:45:13 -0700
Subject: [PATCH 1013/1171] include/linux/aio.h: removed duplicated include

Removed duplicated include <linux/uio.h> in include/linux/aio.h

Signed-off-by: Huang Weiyi <weiyi.huang@gmail.com>
Signed-off-by: Benjamin LaHaise <bcrl@kvack.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/aio.h | 1 -
 1 file changed, 1 deletion(-)

diff --git a/include/linux/aio.h b/include/linux/aio.h
index b51ddd28444e..09b276c35227 100644
--- a/include/linux/aio.h
+++ b/include/linux/aio.h
@@ -7,7 +7,6 @@
 #include <linux/uio.h>
 
 #include <asm/atomic.h>
-#include <linux/uio.h>
 
 #define AIO_MAXSEGS		4
 #define AIO_KIOGRP_NR_ATOMIC	8

From a14e4b572b0ee5c6dbe4aceb83d00b2c969324e9 Mon Sep 17 00:00:00 2001
From: Bob Copeland <me@bobcopeland.com>
Date: Fri, 25 Jul 2008 19:45:14 -0700
Subject: [PATCH 1014/1171] omfs: add filesystem documentation

These patches add the Optimized MPEG Filesystem, a proprietary filesystem used
by the embedded devices Rio Karma and ReplayTV, which are no longer
manufactured.  This filesystem module enables people to access files on these
devices.

This patch:

OMFS is a proprietary filesystem created for the ReplayTV and also used by the
Rio Karma.  It uses hash tables with unordered, unbounded lists in each bucket
for directories, extents for data blocks, 64-bit addressing for blocks, with
up to 8K blocks (only 2K of a given block is ever used for metadata, so the FS
still works with 4K pages).

Document the filesystem usage and structures.

Signed-off-by: Bob Copeland <me@bobcopeland.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/filesystems/omfs.txt | 106 +++++++++++++++++++++++++++++
 1 file changed, 106 insertions(+)
 create mode 100644 Documentation/filesystems/omfs.txt

diff --git a/Documentation/filesystems/omfs.txt b/Documentation/filesystems/omfs.txt
new file mode 100644
index 000000000000..1d0d41ff5c65
--- /dev/null
+++ b/Documentation/filesystems/omfs.txt
@@ -0,0 +1,106 @@
+Optimized MPEG Filesystem (OMFS)
+
+Overview
+========
+
+OMFS is a filesystem created by SonicBlue for use in the ReplayTV DVR
+and Rio Karma MP3 player.  The filesystem is extent-based, utilizing
+block sizes from 2k to 8k, with hash-based directories.  This
+filesystem driver may be used to read and write disks from these
+devices.
+
+Note, it is not recommended that this FS be used in place of a general
+filesystem for your own streaming media device.  Native Linux filesystems
+will likely perform better.
+
+More information is available at:
+
+    http://linux-karma.sf.net/
+
+Various utilities, including mkomfs and omfsck, are included with
+omfsprogs, available at:
+
+    http://bobcopeland.com/karma/
+
+Instructions are included in its README.
+
+Options
+=======
+
+OMFS supports the following mount-time options:
+
+    uid=n        - make all files owned by specified user
+    gid=n        - make all files owned by specified group
+    umask=xxx    - set permission umask to xxx
+    fmask=xxx    - set umask to xxx for files
+    dmask=xxx    - set umask to xxx for directories
+
+Disk format
+===========
+
+OMFS discriminates between "sysblocks" and normal data blocks.  The sysblock
+group consists of super block information, file metadata, directory structures,
+and extents.  Each sysblock has a header containing CRCs of the entire
+sysblock, and may be mirrored in successive blocks on the disk.  A sysblock may
+have a smaller size than a data block, but since they are both addressed by the
+same 64-bit block number, any remaining space in the smaller sysblock is
+unused.
+
+Sysblock header information:
+
+struct omfs_header {
+        __be64 h_self;                  /* FS block where this is located */
+        __be32 h_body_size;             /* size of useful data after header */
+        __be16 h_crc;                   /* crc-ccitt of body_size bytes */
+        char h_fill1[2];
+        u8 h_version;                   /* version, always 1 */
+        char h_type;                    /* OMFS_INODE_X */
+        u8 h_magic;                     /* OMFS_IMAGIC */
+        u8 h_check_xor;                 /* XOR of header bytes before this */
+        __be32 h_fill2;
+};
+
+Files and directories are both represented by omfs_inode:
+
+struct omfs_inode {
+        struct omfs_header i_head;      /* header */
+        __be64 i_parent;                /* parent containing this inode */
+        __be64 i_sibling;               /* next inode in hash bucket */
+        __be64 i_ctime;                 /* ctime, in milliseconds */
+        char i_fill1[35];
+        char i_type;                    /* OMFS_[DIR,FILE] */
+        __be32 i_fill2;
+        char i_fill3[64];
+        char i_name[OMFS_NAMELEN];      /* filename */
+        __be64 i_size;                  /* size of file, in bytes */
+};
+
+Directories in OMFS are implemented as a large hash table.  Filenames are
+hashed then prepended into the bucket list beginning at OMFS_DIR_START.
+Lookup requires hashing the filename, then seeking across i_sibling pointers
+until a match is found on i_name.  Empty buckets are represented by block
+pointers with all-1s (~0).
+
+A file is an omfs_inode structure followed by an extent table beginning at
+OMFS_EXTENT_START:
+
+struct omfs_extent_entry {
+        __be64 e_cluster;               /* start location of a set of blocks */
+        __be64 e_blocks;                /* number of blocks after e_cluster */
+};
+
+struct omfs_extent {
+        __be64 e_next;                  /* next extent table location */
+        __be32 e_extent_count;          /* total # extents in this table */
+        __be32 e_fill;
+        struct omfs_extent_entry e_entry;       /* start of extent entries */
+};
+
+Each extent holds the block offset followed by number of blocks allocated to
+the extent.  The final extent in each table is a terminator with e_cluster
+being ~0 and e_blocks being ones'-complement of the total number of blocks
+in the table.
+
+If this table overflows, a continuation inode is written and pointed to by
+e_next.  These have a header but lack the rest of the inode structure.
+

From 1b002d7b173ae7cc15ed90d3c07f6d106babc510 Mon Sep 17 00:00:00 2001
From: Bob Copeland <me@bobcopeland.com>
Date: Fri, 25 Jul 2008 19:45:15 -0700
Subject: [PATCH 1015/1171] omfs: define filesystem structures

Add header files containing OMFS on-disk and memory structures.

Signed-off-by: Bob Copeland <me@bobcopeland.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/omfs/omfs.h    | 67 +++++++++++++++++++++++++++++++++++++++
 fs/omfs/omfs_fs.h | 80 +++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 147 insertions(+)
 create mode 100644 fs/omfs/omfs.h
 create mode 100644 fs/omfs/omfs_fs.h

diff --git a/fs/omfs/omfs.h b/fs/omfs/omfs.h
new file mode 100644
index 000000000000..2bc0f0670406
--- /dev/null
+++ b/fs/omfs/omfs.h
@@ -0,0 +1,67 @@
+#ifndef _OMFS_H
+#define _OMFS_H
+
+#include <linux/module.h>
+#include <linux/fs.h>
+
+#include "omfs_fs.h"
+
+/* In-memory structures */
+struct omfs_sb_info {
+	u64 s_num_blocks;
+	u64 s_bitmap_ino;
+	u64 s_root_ino;
+	u32 s_blocksize;
+	u32 s_mirrors;
+	u32 s_sys_blocksize;
+	u32 s_clustersize;
+	int s_block_shift;
+	unsigned long **s_imap;
+	int s_imap_size;
+	struct mutex s_bitmap_lock;
+	int s_uid;
+	int s_gid;
+	int s_dmask;
+	int s_fmask;
+};
+
+/* convert a cluster number to a scaled block number */
+static inline sector_t clus_to_blk(struct omfs_sb_info *sbi, sector_t block)
+{
+	return block << sbi->s_block_shift;
+}
+
+static inline struct omfs_sb_info *OMFS_SB(struct super_block *sb)
+{
+	return sb->s_fs_info;
+}
+
+/* bitmap.c */
+extern unsigned long omfs_count_free(struct super_block *sb);
+extern int omfs_allocate_block(struct super_block *sb, u64 block);
+extern int omfs_allocate_range(struct super_block *sb, int min_request,
+			int max_request, u64 *return_block, int *return_size);
+extern int omfs_clear_range(struct super_block *sb, u64 block, int count);
+
+/* dir.c */
+extern struct file_operations omfs_dir_operations;
+extern struct inode_operations omfs_dir_inops;
+extern int omfs_make_empty(struct inode *inode, struct super_block *sb);
+extern int omfs_is_bad(struct omfs_sb_info *sbi, struct omfs_header *header,
+			u64 fsblock);
+
+/* file.c */
+extern struct file_operations omfs_file_operations;
+extern struct inode_operations omfs_file_inops;
+extern struct address_space_operations omfs_aops;
+extern void omfs_make_empty_table(struct buffer_head *bh, int offset);
+extern int omfs_shrink_inode(struct inode *inode);
+
+/* inode.c */
+extern struct inode *omfs_iget(struct super_block *sb, ino_t inode);
+extern struct inode *omfs_new_inode(struct inode *dir, int mode);
+extern int omfs_reserve_block(struct super_block *sb, sector_t block);
+extern int omfs_find_empty_block(struct super_block *sb, int mode, ino_t *ino);
+extern int omfs_sync_inode(struct inode *inode);
+
+#endif
diff --git a/fs/omfs/omfs_fs.h b/fs/omfs/omfs_fs.h
new file mode 100644
index 000000000000..12cca245d6e8
--- /dev/null
+++ b/fs/omfs/omfs_fs.h
@@ -0,0 +1,80 @@
+#ifndef _OMFS_FS_H
+#define _OMFS_FS_H
+
+/* OMFS On-disk structures */
+
+#define OMFS_MAGIC 0xC2993D87
+#define OMFS_IMAGIC 0xD2
+
+#define OMFS_DIR 'D'
+#define OMFS_FILE 'F'
+#define OMFS_INODE_NORMAL 'e'
+#define OMFS_INODE_CONTINUATION 'c'
+#define OMFS_INODE_SYSTEM 's'
+#define OMFS_NAMELEN 256
+#define OMFS_DIR_START 0x1b8
+#define OMFS_EXTENT_START 0x1d0
+#define OMFS_EXTENT_CONT 0x40
+#define OMFS_XOR_COUNT 19
+#define OMFS_MAX_BLOCK_SIZE 8192
+
+struct omfs_super_block {
+	char s_fill1[256];
+	__be64 s_root_block;		/* block number of omfs_root_block */
+	__be64 s_num_blocks;		/* total number of FS blocks */
+	__be32 s_magic;			/* OMFS_MAGIC */
+	__be32 s_blocksize;		/* size of a block */
+	__be32 s_mirrors;		/* # of mirrors of system blocks */
+	__be32 s_sys_blocksize;		/* size of non-data blocks */
+};
+
+struct omfs_header {
+	__be64 h_self;			/* FS block where this is located */
+	__be32 h_body_size;		/* size of useful data after header */
+	__be16 h_crc;			/* crc-ccitt of body_size bytes */
+	char h_fill1[2];
+	u8 h_version;			/* version, always 1 */
+	char h_type;			/* OMFS_INODE_X */
+	u8 h_magic;			/* OMFS_IMAGIC */
+	u8 h_check_xor;			/* XOR of header bytes before this */
+	__be32 h_fill2;
+};
+
+struct omfs_root_block {
+	struct omfs_header r_head;	/* header */
+	__be64 r_fill1;
+	__be64 r_num_blocks;		/* total number of FS blocks */
+	__be64 r_root_dir;		/* block # of root directory */
+	__be64 r_bitmap;		/* block # of free space bitmap */
+	__be32 r_blocksize;		/* size of a block */
+	__be32 r_clustersize;		/* size allocated for data blocks */
+	__be64 r_mirrors;		/* # of mirrors of system blocks */
+	char r_name[OMFS_NAMELEN];	/* partition label */
+};
+
+struct omfs_inode {
+	struct omfs_header i_head;	/* header */
+	__be64 i_parent;		/* parent containing this inode */
+	__be64 i_sibling;		/* next inode in hash bucket */
+	__be64 i_ctime;			/* ctime, in milliseconds */
+	char i_fill1[35];
+	char i_type;			/* OMFS_[DIR,FILE] */
+	__be32 i_fill2;
+	char i_fill3[64];
+	char i_name[OMFS_NAMELEN];	/* filename */
+	__be64 i_size;			/* size of file, in bytes */
+};
+
+struct omfs_extent_entry {
+	__be64 e_cluster;		/* start location of a set of blocks */
+	__be64 e_blocks;		/* number of blocks after e_cluster */
+};
+
+struct omfs_extent {
+	__be64 e_next;			/* next extent table location */
+	__be32 e_extent_count;		/* total # extents in this table */
+	__be32 e_fill;
+	struct omfs_extent_entry e_entry;	/* start of extent entries */
+};
+
+#endif

From 555e3775ced1d05203934fc6529bbf0560dd8733 Mon Sep 17 00:00:00 2001
From: Bob Copeland <me@bobcopeland.com>
Date: Fri, 25 Jul 2008 19:45:15 -0700
Subject: [PATCH 1016/1171] omfs: add inode routines

Add basic superblock and inode handling routines for OMFS

Signed-off-by: Bob Copeland <me@bobcopeland.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/omfs/inode.c | 553 ++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 553 insertions(+)
 create mode 100644 fs/omfs/inode.c

diff --git a/fs/omfs/inode.c b/fs/omfs/inode.c
new file mode 100644
index 000000000000..d865f5535436
--- /dev/null
+++ b/fs/omfs/inode.c
@@ -0,0 +1,553 @@
+/*
+ * Optimized MPEG FS - inode and super operations.
+ * Copyright (C) 2006 Bob Copeland <me@bobcopeland.com>
+ * Released under GPL v2.
+ */
+#include <linux/version.h>
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/fs.h>
+#include <linux/vfs.h>
+#include <linux/parser.h>
+#include <linux/buffer_head.h>
+#include <linux/vmalloc.h>
+#include <linux/crc-itu-t.h>
+#include "omfs.h"
+
+MODULE_AUTHOR("Bob Copeland <me@bobcopeland.com>");
+MODULE_DESCRIPTION("OMFS (ReplayTV/Karma) Filesystem for Linux");
+MODULE_LICENSE("GPL");
+
+struct inode *omfs_new_inode(struct inode *dir, int mode)
+{
+	struct inode *inode;
+	u64 new_block;
+	int err;
+	int len;
+	struct omfs_sb_info *sbi = OMFS_SB(dir->i_sb);
+
+	inode = new_inode(dir->i_sb);
+	if (!inode)
+		return ERR_PTR(-ENOMEM);
+
+	err = omfs_allocate_range(dir->i_sb, sbi->s_mirrors, sbi->s_mirrors,
+			&new_block, &len);
+	if (err)
+		goto fail;
+
+	inode->i_ino = new_block;
+	inode->i_mode = mode;
+	inode->i_uid = current->fsuid;
+	inode->i_gid = current->fsgid;
+	inode->i_blocks = 0;
+	inode->i_mapping->a_ops = &omfs_aops;
+
+	inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
+	switch (mode & S_IFMT) {
+	case S_IFDIR:
+		inode->i_op = &omfs_dir_inops;
+		inode->i_fop = &omfs_dir_operations;
+		inode->i_size = sbi->s_sys_blocksize;
+		inc_nlink(inode);
+		break;
+	case S_IFREG:
+		inode->i_op = &omfs_file_inops;
+		inode->i_fop = &omfs_file_operations;
+		inode->i_size = 0;
+		break;
+	}
+
+	insert_inode_hash(inode);
+	mark_inode_dirty(inode);
+	return inode;
+fail:
+	make_bad_inode(inode);
+	iput(inode);
+	return ERR_PTR(err);
+}
+
+/*
+ * Update the header checksums for a dirty inode based on its contents.
+ * Caller is expected to hold the buffer head underlying oi and mark it
+ * dirty.
+ */
+static void omfs_update_checksums(struct omfs_inode *oi)
+{
+	int xor, i, ofs = 0, count;
+	u16 crc = 0;
+	unsigned char *ptr = (unsigned char *) oi;
+
+	count = be32_to_cpu(oi->i_head.h_body_size);
+	ofs = sizeof(struct omfs_header);
+
+	crc = crc_itu_t(crc, ptr + ofs, count);
+	oi->i_head.h_crc = cpu_to_be16(crc);
+
+	xor = ptr[0];
+	for (i = 1; i < OMFS_XOR_COUNT; i++)
+		xor ^= ptr[i];
+
+	oi->i_head.h_check_xor = xor;
+}
+
+static int omfs_write_inode(struct inode *inode, int wait)
+{
+	struct omfs_inode *oi;
+	struct omfs_sb_info *sbi = OMFS_SB(inode->i_sb);
+	struct buffer_head *bh, *bh2;
+	unsigned int block;
+	u64 ctime;
+	int i;
+	int ret = -EIO;
+	int sync_failed = 0;
+
+	/* get current inode since we may have written sibling ptrs etc. */
+	block = clus_to_blk(sbi, inode->i_ino);
+	bh = sb_bread(inode->i_sb, block);
+	if (!bh)
+		goto out;
+
+	oi = (struct omfs_inode *) bh->b_data;
+
+	oi->i_head.h_self = cpu_to_be64(inode->i_ino);
+	if (S_ISDIR(inode->i_mode))
+		oi->i_type = OMFS_DIR;
+	else if (S_ISREG(inode->i_mode))
+		oi->i_type = OMFS_FILE;
+	else {
+		printk(KERN_WARNING "omfs: unknown file type: %d\n",
+			inode->i_mode);
+		goto out_brelse;
+	}
+
+	oi->i_head.h_body_size = cpu_to_be32(sbi->s_sys_blocksize -
+		sizeof(struct omfs_header));
+	oi->i_head.h_version = 1;
+	oi->i_head.h_type = OMFS_INODE_NORMAL;
+	oi->i_head.h_magic = OMFS_IMAGIC;
+	oi->i_size = cpu_to_be64(inode->i_size);
+
+	ctime = inode->i_ctime.tv_sec * 1000LL +
+		((inode->i_ctime.tv_nsec + 999)/1000);
+	oi->i_ctime = cpu_to_be64(ctime);
+
+	omfs_update_checksums(oi);
+
+	mark_buffer_dirty(bh);
+	if (wait) {
+		sync_dirty_buffer(bh);
+		if (buffer_req(bh) && !buffer_uptodate(bh))
+			sync_failed = 1;
+	}
+
+	/* if mirroring writes, copy to next fsblock */
+	for (i = 1; i < sbi->s_mirrors; i++) {
+		bh2 = sb_bread(inode->i_sb, block + i *
+			(sbi->s_blocksize / sbi->s_sys_blocksize));
+		if (!bh2)
+			goto out_brelse;
+
+		memcpy(bh2->b_data, bh->b_data, bh->b_size);
+		mark_buffer_dirty(bh2);
+		if (wait) {
+			sync_dirty_buffer(bh2);
+			if (buffer_req(bh2) && !buffer_uptodate(bh2))
+				sync_failed = 1;
+		}
+		brelse(bh2);
+	}
+	ret = (sync_failed) ? -EIO : 0;
+out_brelse:
+	brelse(bh);
+out:
+	return ret;
+}
+
+int omfs_sync_inode(struct inode *inode)
+{
+	return omfs_write_inode(inode, 1);
+}
+
+/*
+ * called when an entry is deleted, need to clear the bits in the
+ * bitmaps.
+ */
+static void omfs_delete_inode(struct inode *inode)
+{
+	truncate_inode_pages(&inode->i_data, 0);
+
+	if (S_ISREG(inode->i_mode)) {
+		inode->i_size = 0;
+		omfs_shrink_inode(inode);
+	}
+
+	omfs_clear_range(inode->i_sb, inode->i_ino, 2);
+	clear_inode(inode);
+}
+
+struct inode *omfs_iget(struct super_block *sb, ino_t ino)
+{
+	struct omfs_sb_info *sbi = OMFS_SB(sb);
+	struct omfs_inode *oi;
+	struct buffer_head *bh;
+	unsigned int block;
+	u64 ctime;
+	unsigned long nsecs;
+	struct inode *inode;
+
+	inode = iget_locked(sb, ino);
+	if (!inode)
+		return ERR_PTR(-ENOMEM);
+	if (!(inode->i_state & I_NEW))
+		return inode;
+
+	block = clus_to_blk(sbi, ino);
+	bh = sb_bread(inode->i_sb, block);
+	if (!bh)
+		goto iget_failed;
+
+	oi = (struct omfs_inode *)bh->b_data;
+
+	/* check self */
+	if (ino != be64_to_cpu(oi->i_head.h_self))
+		goto fail_bh;
+
+	inode->i_uid = sbi->s_uid;
+	inode->i_gid = sbi->s_gid;
+
+	ctime = be64_to_cpu(oi->i_ctime);
+	nsecs = do_div(ctime, 1000) * 1000L;
+
+	inode->i_atime.tv_sec = ctime;
+	inode->i_mtime.tv_sec = ctime;
+	inode->i_ctime.tv_sec = ctime;
+	inode->i_atime.tv_nsec = nsecs;
+	inode->i_mtime.tv_nsec = nsecs;
+	inode->i_ctime.tv_nsec = nsecs;
+
+	inode->i_mapping->a_ops = &omfs_aops;
+
+	switch (oi->i_type) {
+	case OMFS_DIR:
+		inode->i_mode = S_IFDIR | (S_IRWXUGO & ~sbi->s_dmask);
+		inode->i_op = &omfs_dir_inops;
+		inode->i_fop = &omfs_dir_operations;
+		inode->i_size = be32_to_cpu(oi->i_head.h_body_size) +
+			sizeof(struct omfs_header);
+		inc_nlink(inode);
+		break;
+	case OMFS_FILE:
+		inode->i_mode = S_IFREG | (S_IRWXUGO & ~sbi->s_fmask);
+		inode->i_fop = &omfs_file_operations;
+		inode->i_size = be64_to_cpu(oi->i_size);
+		break;
+	}
+	brelse(bh);
+	unlock_new_inode(inode);
+	return inode;
+fail_bh:
+	brelse(bh);
+iget_failed:
+	iget_failed(inode);
+	return ERR_PTR(-EIO);
+}
+
+static void omfs_put_super(struct super_block *sb)
+{
+	struct omfs_sb_info *sbi = OMFS_SB(sb);
+	kfree(sbi->s_imap);
+	kfree(sbi);
+	sb->s_fs_info = NULL;
+}
+
+static int omfs_statfs(struct dentry *dentry, struct kstatfs *buf)
+{
+	struct super_block *s = dentry->d_sb;
+	struct omfs_sb_info *sbi = OMFS_SB(s);
+	buf->f_type = OMFS_MAGIC;
+	buf->f_bsize = sbi->s_blocksize;
+	buf->f_blocks = sbi->s_num_blocks;
+	buf->f_files = sbi->s_num_blocks;
+	buf->f_namelen = OMFS_NAMELEN;
+
+	buf->f_bfree = buf->f_bavail = buf->f_ffree =
+		omfs_count_free(s);
+	return 0;
+}
+
+static struct super_operations omfs_sops = {
+	.write_inode	= omfs_write_inode,
+	.delete_inode	= omfs_delete_inode,
+	.put_super	= omfs_put_super,
+	.statfs		= omfs_statfs,
+	.show_options	= generic_show_options,
+};
+
+/*
+ * For Rio Karma, there is an on-disk free bitmap whose location is
+ * stored in the root block.  For ReplayTV, there is no such free bitmap
+ * so we have to walk the tree.  Both inodes and file data are allocated
+ * from the same map.  This array can be big (300k) so we allocate
+ * in units of the blocksize.
+ */
+static int omfs_get_imap(struct super_block *sb)
+{
+	int bitmap_size;
+	int array_size;
+	int count;
+	struct omfs_sb_info *sbi = OMFS_SB(sb);
+	struct buffer_head *bh;
+	unsigned long **ptr;
+	sector_t block;
+
+	bitmap_size = DIV_ROUND_UP(sbi->s_num_blocks, 8);
+	array_size = DIV_ROUND_UP(bitmap_size, sb->s_blocksize);
+
+	if (sbi->s_bitmap_ino == ~0ULL)
+		goto out;
+
+	sbi->s_imap_size = array_size;
+	sbi->s_imap = kzalloc(array_size * sizeof(unsigned long *), GFP_KERNEL);
+	if (!sbi->s_imap)
+		goto nomem;
+
+	block = clus_to_blk(sbi, sbi->s_bitmap_ino);
+	ptr = sbi->s_imap;
+	for (count = bitmap_size; count > 0; count -= sb->s_blocksize) {
+		bh = sb_bread(sb, block++);
+		if (!bh)
+			goto nomem_free;
+		*ptr = kmalloc(sb->s_blocksize, GFP_KERNEL);
+		if (!*ptr) {
+			brelse(bh);
+			goto nomem_free;
+		}
+		memcpy(*ptr, bh->b_data, sb->s_blocksize);
+		if (count < sb->s_blocksize)
+			memset((void *)*ptr + count, 0xff,
+				sb->s_blocksize - count);
+		brelse(bh);
+		ptr++;
+	}
+out:
+	return 0;
+
+nomem_free:
+	for (count = 0; count < array_size; count++)
+		kfree(sbi->s_imap[count]);
+
+	kfree(sbi->s_imap);
+nomem:
+	sbi->s_imap = NULL;
+	sbi->s_imap_size = 0;
+	return -ENOMEM;
+}
+
+enum {
+	Opt_uid, Opt_gid, Opt_umask, Opt_dmask, Opt_fmask
+};
+
+static match_table_t tokens = {
+	{Opt_uid, "uid=%u"},
+	{Opt_gid, "gid=%u"},
+	{Opt_umask, "umask=%o"},
+	{Opt_dmask, "dmask=%o"},
+	{Opt_fmask, "fmask=%o"},
+};
+
+static int parse_options(char *options, struct omfs_sb_info *sbi)
+{
+	char *p;
+	substring_t args[MAX_OPT_ARGS];
+	int option;
+
+	if (!options)
+		return 1;
+
+	while ((p = strsep(&options, ",")) != NULL) {
+		int token;
+		if (!*p)
+			continue;
+
+		token = match_token(p, tokens, args);
+		switch (token) {
+		case Opt_uid:
+			if (match_int(&args[0], &option))
+				return 0;
+			sbi->s_uid = option;
+			break;
+		case Opt_gid:
+			if (match_int(&args[0], &option))
+				return 0;
+			sbi->s_gid = option;
+			break;
+		case Opt_umask:
+			if (match_octal(&args[0], &option))
+				return 0;
+			sbi->s_fmask = sbi->s_dmask = option;
+			break;
+		case Opt_dmask:
+			if (match_octal(&args[0], &option))
+				return 0;
+			sbi->s_dmask = option;
+			break;
+		case Opt_fmask:
+			if (match_octal(&args[0], &option))
+				return 0;
+			sbi->s_fmask = option;
+			break;
+		default:
+			return 0;
+		}
+	}
+	return 1;
+}
+
+static int omfs_fill_super(struct super_block *sb, void *data, int silent)
+{
+	struct buffer_head *bh, *bh2;
+	struct omfs_super_block *omfs_sb;
+	struct omfs_root_block *omfs_rb;
+	struct omfs_sb_info *sbi;
+	struct inode *root;
+	sector_t start;
+	int ret = -EINVAL;
+
+	save_mount_options(sb, (char *) data);
+
+	sbi = kzalloc(sizeof(struct omfs_sb_info), GFP_KERNEL);
+	if (!sbi)
+		return -ENOMEM;
+
+	sb->s_fs_info = sbi;
+
+	sbi->s_uid = current->uid;
+	sbi->s_gid = current->gid;
+	sbi->s_dmask = sbi->s_fmask = current->fs->umask;
+
+	if (!parse_options((char *) data, sbi))
+		goto end;
+
+	sb->s_maxbytes = 0xffffffff;
+
+	sb_set_blocksize(sb, 0x200);
+
+	bh = sb_bread(sb, 0);
+	if (!bh)
+		goto end;
+
+	omfs_sb = (struct omfs_super_block *)bh->b_data;
+
+	if (omfs_sb->s_magic != cpu_to_be32(OMFS_MAGIC)) {
+		if (!silent)
+			printk(KERN_ERR "omfs: Invalid superblock (%x)\n",
+				   omfs_sb->s_magic);
+		goto out_brelse_bh;
+	}
+	sb->s_magic = OMFS_MAGIC;
+
+	sbi->s_num_blocks = be64_to_cpu(omfs_sb->s_num_blocks);
+	sbi->s_blocksize = be32_to_cpu(omfs_sb->s_blocksize);
+	sbi->s_mirrors = be32_to_cpu(omfs_sb->s_mirrors);
+	sbi->s_root_ino = be64_to_cpu(omfs_sb->s_root_block);
+	sbi->s_sys_blocksize = be32_to_cpu(omfs_sb->s_sys_blocksize);
+	mutex_init(&sbi->s_bitmap_lock);
+
+	if (sbi->s_sys_blocksize > PAGE_SIZE) {
+		printk(KERN_ERR "omfs: sysblock size (%d) is out of range\n",
+			sbi->s_sys_blocksize);
+		goto out_brelse_bh;
+	}
+
+	if (sbi->s_blocksize < sbi->s_sys_blocksize ||
+	    sbi->s_blocksize > OMFS_MAX_BLOCK_SIZE) {
+		printk(KERN_ERR "omfs: block size (%d) is out of range\n",
+			sbi->s_blocksize);
+		goto out_brelse_bh;
+	}
+
+	/*
+	 * Use sys_blocksize as the fs block since it is smaller than a
+	 * page while the fs blocksize can be larger.
+	 */
+	sb_set_blocksize(sb, sbi->s_sys_blocksize);
+
+	/*
+	 * ...and the difference goes into a shift.  sys_blocksize is always
+	 * a power of two factor of blocksize.
+	 */
+	sbi->s_block_shift = get_bitmask_order(sbi->s_blocksize) -
+		get_bitmask_order(sbi->s_sys_blocksize);
+
+	start = clus_to_blk(sbi, be64_to_cpu(omfs_sb->s_root_block));
+	bh2 = sb_bread(sb, start);
+	if (!bh2)
+		goto out_brelse_bh;
+
+	omfs_rb = (struct omfs_root_block *)bh2->b_data;
+
+	sbi->s_bitmap_ino = be64_to_cpu(omfs_rb->r_bitmap);
+	sbi->s_clustersize = be32_to_cpu(omfs_rb->r_clustersize);
+
+	if (sbi->s_num_blocks != be64_to_cpu(omfs_rb->r_num_blocks)) {
+		printk(KERN_ERR "omfs: block count discrepancy between "
+			"super and root blocks (%llx, %llx)\n",
+			sbi->s_num_blocks, be64_to_cpu(omfs_rb->r_num_blocks));
+		goto out_brelse_bh2;
+	}
+
+	ret = omfs_get_imap(sb);
+	if (ret)
+		goto out_brelse_bh2;
+
+	sb->s_op = &omfs_sops;
+
+	root = omfs_iget(sb, be64_to_cpu(omfs_rb->r_root_dir));
+	if (IS_ERR(root)) {
+		ret = PTR_ERR(root);
+		goto out_brelse_bh2;
+	}
+
+	sb->s_root = d_alloc_root(root);
+	if (!sb->s_root) {
+		iput(root);
+		goto out_brelse_bh2;
+	}
+	printk(KERN_DEBUG "omfs: Mounted volume %s\n", omfs_rb->r_name);
+
+	ret = 0;
+out_brelse_bh2:
+	brelse(bh2);
+out_brelse_bh:
+	brelse(bh);
+end:
+	return ret;
+}
+
+static int omfs_get_sb(struct file_system_type *fs_type,
+			int flags, const char *dev_name,
+			void *data, struct vfsmount *m)
+{
+	return get_sb_bdev(fs_type, flags, dev_name, data, omfs_fill_super, m);
+}
+
+static struct file_system_type omfs_fs_type = {
+	.owner = THIS_MODULE,
+	.name = "omfs",
+	.get_sb = omfs_get_sb,
+	.kill_sb = kill_block_super,
+	.fs_flags = FS_REQUIRES_DEV,
+};
+
+static int __init init_omfs_fs(void)
+{
+	return register_filesystem(&omfs_fs_type);
+}
+
+static void __exit exit_omfs_fs(void)
+{
+	unregister_filesystem(&omfs_fs_type);
+}
+
+module_init(init_omfs_fs);
+module_exit(exit_omfs_fs);

From a3ab7155ea21aadc8a4d5687e91b3d876973185e Mon Sep 17 00:00:00 2001
From: Bob Copeland <me@bobcopeland.com>
Date: Fri, 25 Jul 2008 19:45:16 -0700
Subject: [PATCH 1017/1171] omfs: add directory routines

Add lookup and directory management routines for OMFS.  The filesystem uses
hashing based on the filename and stores collisions, unordered, in siblings
of files' inode structures.  To support telldir, the current position in
the hash table is encoded in fpos.

Signed-off-by: Bob Copeland <me@bobcopeland.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/omfs/dir.c | 504 ++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 504 insertions(+)
 create mode 100644 fs/omfs/dir.c

diff --git a/fs/omfs/dir.c b/fs/omfs/dir.c
new file mode 100644
index 000000000000..05a5bc31e4bd
--- /dev/null
+++ b/fs/omfs/dir.c
@@ -0,0 +1,504 @@
+/*
+ * OMFS (as used by RIO Karma) directory operations.
+ * Copyright (C) 2005 Bob Copeland <me@bobcopeland.com>
+ * Released under GPL v2.
+ */
+
+#include <linux/fs.h>
+#include <linux/ctype.h>
+#include <linux/buffer_head.h>
+#include "omfs.h"
+
+static int omfs_hash(const char *name, int namelen, int mod)
+{
+	int i, hash = 0;
+	for (i = 0; i < namelen; i++)
+		hash ^= tolower(name[i]) << (i % 24);
+	return hash % mod;
+}
+
+/*
+ * Finds the bucket for a given name and reads the containing block;
+ * *ofs is set to the offset of the first list entry.
+ */
+static struct buffer_head *omfs_get_bucket(struct inode *dir,
+		const char *name, int namelen, int *ofs)
+{
+	int nbuckets = (dir->i_size - OMFS_DIR_START)/8;
+	int block = clus_to_blk(OMFS_SB(dir->i_sb), dir->i_ino);
+	int bucket = omfs_hash(name, namelen, nbuckets);
+
+	*ofs = OMFS_DIR_START + bucket * 8;
+	return sb_bread(dir->i_sb, block);
+}
+
+static struct buffer_head *omfs_scan_list(struct inode *dir, u64 block,
+				const char *name, int namelen,
+				u64 *prev_block)
+{
+	struct buffer_head *bh;
+	struct omfs_inode *oi;
+	int err = -ENOENT;
+	*prev_block = ~0;
+
+	while (block != ~0) {
+		bh = sb_bread(dir->i_sb,
+			clus_to_blk(OMFS_SB(dir->i_sb), block));
+		if (!bh) {
+			err = -EIO;
+			goto err;
+		}
+
+		oi = (struct omfs_inode *) bh->b_data;
+		if (omfs_is_bad(OMFS_SB(dir->i_sb), &oi->i_head, block)) {
+			brelse(bh);
+			goto err;
+		}
+
+		if (strncmp(oi->i_name, name, namelen) == 0)
+			return bh;
+
+		*prev_block = block;
+		block = be64_to_cpu(oi->i_sibling);
+		brelse(bh);
+	}
+err:
+	return ERR_PTR(err);
+}
+
+static struct buffer_head *omfs_find_entry(struct inode *dir,
+					   const char *name, int namelen)
+{
+	struct buffer_head *bh;
+	int ofs;
+	u64 block, dummy;
+
+	bh = omfs_get_bucket(dir, name, namelen, &ofs);
+	if (!bh)
+		return ERR_PTR(-EIO);
+
+	block = be64_to_cpu(*((__be64 *) &bh->b_data[ofs]));
+	brelse(bh);
+
+	return omfs_scan_list(dir, block, name, namelen, &dummy);
+}
+
+int omfs_make_empty(struct inode *inode, struct super_block *sb)
+{
+	struct omfs_sb_info *sbi = OMFS_SB(sb);
+	int block = clus_to_blk(sbi, inode->i_ino);
+	struct buffer_head *bh;
+	struct omfs_inode *oi;
+
+	bh = sb_bread(sb, block);
+	if (!bh)
+		return -ENOMEM;
+
+	memset(bh->b_data, 0, sizeof(struct omfs_inode));
+
+	if (inode->i_mode & S_IFDIR) {
+		memset(&bh->b_data[OMFS_DIR_START], 0xff,
+			sbi->s_sys_blocksize - OMFS_DIR_START);
+	} else
+		omfs_make_empty_table(bh, OMFS_EXTENT_START);
+
+	oi = (struct omfs_inode *) bh->b_data;
+	oi->i_head.h_self = cpu_to_be64(inode->i_ino);
+	oi->i_sibling = ~0ULL;
+
+	mark_buffer_dirty(bh);
+	brelse(bh);
+	return 0;
+}
+
+static int omfs_add_link(struct dentry *dentry, struct inode *inode)
+{
+	struct inode *dir = dentry->d_parent->d_inode;
+	const char *name = dentry->d_name.name;
+	int namelen = dentry->d_name.len;
+	struct omfs_inode *oi;
+	struct buffer_head *bh;
+	u64 block;
+	__be64 *entry;
+	int ofs;
+
+	/* just prepend to head of queue in proper bucket */
+	bh = omfs_get_bucket(dir, name, namelen, &ofs);
+	if (!bh)
+		goto out;
+
+	entry = (__be64 *) &bh->b_data[ofs];
+	block = be64_to_cpu(*entry);
+	*entry = cpu_to_be64(inode->i_ino);
+	mark_buffer_dirty(bh);
+	brelse(bh);
+
+	/* now set the sibling and parent pointers on the new inode */
+	bh = sb_bread(dir->i_sb, clus_to_blk(OMFS_SB(dir->i_sb), inode->i_ino));
+	if (!bh)
+		goto out;
+
+	oi = (struct omfs_inode *) bh->b_data;
+	memcpy(oi->i_name, name, namelen);
+	memset(oi->i_name + namelen, 0, OMFS_NAMELEN - namelen);
+	oi->i_sibling = cpu_to_be64(block);
+	oi->i_parent = cpu_to_be64(dir->i_ino);
+	mark_buffer_dirty(bh);
+	brelse(bh);
+
+	dir->i_ctime = CURRENT_TIME_SEC;
+
+	/* mark affected inodes dirty to rebuild checksums */
+	mark_inode_dirty(dir);
+	mark_inode_dirty(inode);
+	return 0;
+out:
+	return -ENOMEM;
+}
+
+static int omfs_delete_entry(struct dentry *dentry)
+{
+	struct inode *dir = dentry->d_parent->d_inode;
+	struct inode *dirty;
+	const char *name = dentry->d_name.name;
+	int namelen = dentry->d_name.len;
+	struct omfs_inode *oi;
+	struct buffer_head *bh, *bh2;
+	__be64 *entry, next;
+	u64 block, prev;
+	int ofs;
+	int err = -ENOMEM;
+
+	/* delete the proper node in the bucket's linked list */
+	bh = omfs_get_bucket(dir, name, namelen, &ofs);
+	if (!bh)
+		goto out;
+
+	entry = (__be64 *) &bh->b_data[ofs];
+	block = be64_to_cpu(*entry);
+
+	bh2 = omfs_scan_list(dir, block, name, namelen, &prev);
+	if (IS_ERR(bh2)) {
+		err = PTR_ERR(bh2);
+		goto out_free_bh;
+	}
+
+	oi = (struct omfs_inode *) bh2->b_data;
+	next = oi->i_sibling;
+	brelse(bh2);
+
+	if (prev != ~0) {
+		/* found in middle of list, get list ptr */
+		brelse(bh);
+		bh = sb_bread(dir->i_sb,
+			clus_to_blk(OMFS_SB(dir->i_sb), prev));
+		if (!bh)
+			goto out;
+
+		oi = (struct omfs_inode *) bh->b_data;
+		entry = &oi->i_sibling;
+	}
+
+	*entry = next;
+	mark_buffer_dirty(bh);
+
+	if (prev != ~0) {
+		dirty = omfs_iget(dir->i_sb, prev);
+		if (!IS_ERR(dirty)) {
+			mark_inode_dirty(dirty);
+			iput(dirty);
+		}
+	}
+
+	err = 0;
+out_free_bh:
+	brelse(bh);
+out:
+	return err;
+}
+
+static int omfs_dir_is_empty(struct inode *inode)
+{
+	int nbuckets = (inode->i_size - OMFS_DIR_START) / 8;
+	struct buffer_head *bh;
+	u64 *ptr;
+	int i;
+
+	bh = sb_bread(inode->i_sb, clus_to_blk(OMFS_SB(inode->i_sb),
+			inode->i_ino));
+
+	if (!bh)
+		return 0;
+
+	ptr = (u64 *) &bh->b_data[OMFS_DIR_START];
+
+	for (i = 0; i < nbuckets; i++, ptr++)
+		if (*ptr != ~0)
+			break;
+
+	brelse(bh);
+	return *ptr != ~0;
+}
+
+static int omfs_unlink(struct inode *dir, struct dentry *dentry)
+{
+	int ret;
+	struct inode *inode = dentry->d_inode;
+
+	ret = omfs_delete_entry(dentry);
+	if (ret)
+		goto end_unlink;
+
+	inode_dec_link_count(inode);
+	mark_inode_dirty(dir);
+
+end_unlink:
+	return ret;
+}
+
+static int omfs_rmdir(struct inode *dir, struct dentry *dentry)
+{
+	int err = -ENOTEMPTY;
+	struct inode *inode = dentry->d_inode;
+
+	if (omfs_dir_is_empty(inode)) {
+		err = omfs_unlink(dir, dentry);
+		if (!err)
+			inode_dec_link_count(inode);
+	}
+	return err;
+}
+
+static int omfs_add_node(struct inode *dir, struct dentry *dentry, int mode)
+{
+	int err;
+	struct inode *inode = omfs_new_inode(dir, mode);
+
+	if (IS_ERR(inode))
+		return PTR_ERR(inode);
+
+	err = omfs_make_empty(inode, dir->i_sb);
+	if (err)
+		goto out_free_inode;
+
+	err = omfs_add_link(dentry, inode);
+	if (err)
+		goto out_free_inode;
+
+	d_instantiate(dentry, inode);
+	return 0;
+
+out_free_inode:
+	iput(inode);
+	return err;
+}
+
+static int omfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
+{
+	return omfs_add_node(dir, dentry, mode | S_IFDIR);
+}
+
+static int omfs_create(struct inode *dir, struct dentry *dentry, int mode,
+		struct nameidata *nd)
+{
+	return omfs_add_node(dir, dentry, mode | S_IFREG);
+}
+
+static struct dentry *omfs_lookup(struct inode *dir, struct dentry *dentry,
+				  struct nameidata *nd)
+{
+	struct buffer_head *bh;
+	struct inode *inode = NULL;
+
+	if (dentry->d_name.len > OMFS_NAMELEN)
+		return ERR_PTR(-ENAMETOOLONG);
+
+	bh = omfs_find_entry(dir, dentry->d_name.name, dentry->d_name.len);
+	if (!IS_ERR(bh)) {
+		struct omfs_inode *oi = (struct omfs_inode *)bh->b_data;
+		ino_t ino = be64_to_cpu(oi->i_head.h_self);
+		brelse(bh);
+		inode = omfs_iget(dir->i_sb, ino);
+		if (IS_ERR(inode))
+			return ERR_CAST(inode);
+	}
+	d_add(dentry, inode);
+	return NULL;
+}
+
+/* sanity check block's self pointer */
+int omfs_is_bad(struct omfs_sb_info *sbi, struct omfs_header *header,
+	u64 fsblock)
+{
+	int is_bad;
+	u64 ino = be64_to_cpu(header->h_self);
+	is_bad = ((ino != fsblock) || (ino < sbi->s_root_ino) ||
+		(ino > sbi->s_num_blocks));
+
+	if (is_bad)
+		printk(KERN_WARNING "omfs: bad hash chain detected\n");
+
+	return is_bad;
+}
+
+static int omfs_fill_chain(struct file *filp, void *dirent, filldir_t filldir,
+		u64 fsblock, int hindex)
+{
+	struct inode *dir = filp->f_dentry->d_inode;
+	struct buffer_head *bh;
+	struct omfs_inode *oi;
+	u64 self;
+	int res = 0;
+	unsigned char d_type;
+
+	/* follow chain in this bucket */
+	while (fsblock != ~0) {
+		bh = sb_bread(dir->i_sb, clus_to_blk(OMFS_SB(dir->i_sb),
+				fsblock));
+		if (!bh)
+			goto out;
+
+		oi = (struct omfs_inode *) bh->b_data;
+		if (omfs_is_bad(OMFS_SB(dir->i_sb), &oi->i_head, fsblock)) {
+			brelse(bh);
+			goto out;
+		}
+
+		self = fsblock;
+		fsblock = be64_to_cpu(oi->i_sibling);
+
+		/* skip visited nodes */
+		if (hindex) {
+			hindex--;
+			brelse(bh);
+			continue;
+		}
+
+		d_type = (oi->i_type == OMFS_DIR) ? DT_DIR : DT_REG;
+
+		res = filldir(dirent, oi->i_name, strnlen(oi->i_name,
+			OMFS_NAMELEN), filp->f_pos, self, d_type);
+		if (res == 0)
+			filp->f_pos++;
+		brelse(bh);
+	}
+out:
+	return res;
+}
+
+static int omfs_rename(struct inode *old_dir, struct dentry *old_dentry,
+		struct inode *new_dir, struct dentry *new_dentry)
+{
+	struct inode *new_inode = new_dentry->d_inode;
+	struct inode *old_inode = old_dentry->d_inode;
+	struct buffer_head *bh;
+	int is_dir;
+	int err;
+
+	is_dir = S_ISDIR(old_inode->i_mode);
+
+	if (new_inode) {
+		/* overwriting existing file/dir */
+		err = -ENOTEMPTY;
+		if (is_dir && !omfs_dir_is_empty(new_inode))
+			goto out;
+
+		err = -ENOENT;
+		bh = omfs_find_entry(new_dir, new_dentry->d_name.name,
+			new_dentry->d_name.len);
+		if (IS_ERR(bh))
+			goto out;
+		brelse(bh);
+
+		err = omfs_unlink(new_dir, new_dentry);
+		if (err)
+			goto out;
+	}
+
+	/* since omfs locates files by name, we need to unlink _before_
+	 * adding the new link or we won't find the old one */
+	inode_inc_link_count(old_inode);
+	err = omfs_unlink(old_dir, old_dentry);
+	if (err) {
+		inode_dec_link_count(old_inode);
+		goto out;
+	}
+
+	err = omfs_add_link(new_dentry, old_inode);
+	if (err)
+		goto out;
+
+	old_inode->i_ctime = CURRENT_TIME_SEC;
+out:
+	return err;
+}
+
+static int omfs_readdir(struct file *filp, void *dirent, filldir_t filldir)
+{
+	struct inode *dir = filp->f_dentry->d_inode;
+	struct buffer_head *bh;
+	loff_t offset, res;
+	unsigned int hchain, hindex;
+	int nbuckets;
+	u64 fsblock;
+	int ret = -EINVAL;
+
+	if (filp->f_pos >> 32)
+		goto success;
+
+	switch ((unsigned long) filp->f_pos) {
+	case 0:
+		if (filldir(dirent, ".", 1, 0, dir->i_ino, DT_DIR) < 0)
+			goto success;
+		filp->f_pos++;
+		/* fall through */
+	case 1:
+		if (filldir(dirent, "..", 2, 1,
+		    parent_ino(filp->f_dentry), DT_DIR) < 0)
+			goto success;
+		filp->f_pos = 1 << 20;
+		/* fall through */
+	}
+
+	nbuckets = (dir->i_size - OMFS_DIR_START) / 8;
+
+	/* high 12 bits store bucket + 1 and low 20 bits store hash index */
+	hchain = (filp->f_pos >> 20) - 1;
+	hindex = filp->f_pos & 0xfffff;
+
+	bh = sb_bread(dir->i_sb, clus_to_blk(OMFS_SB(dir->i_sb), dir->i_ino));
+	if (!bh)
+		goto out;
+
+	offset = OMFS_DIR_START + hchain * 8;
+
+	for (; hchain < nbuckets; hchain++, offset += 8) {
+		fsblock = be64_to_cpu(*((__be64 *) &bh->b_data[offset]));
+
+		res = omfs_fill_chain(filp, dirent, filldir, fsblock, hindex);
+		hindex = 0;
+		if (res < 0)
+			break;
+
+		filp->f_pos = (hchain+2) << 20;
+	}
+	brelse(bh);
+success:
+	ret = 0;
+out:
+	return ret;
+}
+
+struct inode_operations omfs_dir_inops = {
+	.lookup = omfs_lookup,
+	.mkdir = omfs_mkdir,
+	.rename = omfs_rename,
+	.create = omfs_create,
+	.unlink = omfs_unlink,
+	.rmdir = omfs_rmdir,
+};
+
+struct file_operations omfs_dir_operations = {
+	.read = generic_read_dir,
+	.readdir = omfs_readdir,
+};

From 8f09e98768c17287df076580c4cc72ac358312c6 Mon Sep 17 00:00:00 2001
From: Bob Copeland <me@bobcopeland.com>
Date: Fri, 25 Jul 2008 19:45:16 -0700
Subject: [PATCH 1018/1171] omfs: add file routines

Add functions for reading and manipulating the storage of file data in
the extent-based OMFS.

Signed-off-by: Bob Copeland <me@bobcopeland.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/omfs/file.c | 346 +++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 346 insertions(+)
 create mode 100644 fs/omfs/file.c

diff --git a/fs/omfs/file.c b/fs/omfs/file.c
new file mode 100644
index 000000000000..66e01fae4384
--- /dev/null
+++ b/fs/omfs/file.c
@@ -0,0 +1,346 @@
+/*
+ * OMFS (as used by RIO Karma) file operations.
+ * Copyright (C) 2005 Bob Copeland <me@bobcopeland.com>
+ * Released under GPL v2.
+ */
+
+#include <linux/version.h>
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/buffer_head.h>
+#include <linux/mpage.h>
+#include "omfs.h"
+
+static int omfs_sync_file(struct file *file, struct dentry *dentry,
+		int datasync)
+{
+	struct inode *inode = dentry->d_inode;
+	int err;
+
+	err = sync_mapping_buffers(inode->i_mapping);
+	if (!(inode->i_state & I_DIRTY))
+		return err;
+	if (datasync && !(inode->i_state & I_DIRTY_DATASYNC))
+		return err;
+	err |= omfs_sync_inode(inode);
+	return err ? -EIO : 0;
+}
+
+void omfs_make_empty_table(struct buffer_head *bh, int offset)
+{
+	struct omfs_extent *oe = (struct omfs_extent *) &bh->b_data[offset];
+
+	oe->e_next = ~0ULL;
+	oe->e_extent_count = cpu_to_be32(1),
+	oe->e_fill = cpu_to_be32(0x22),
+	oe->e_entry.e_cluster = ~0ULL;
+	oe->e_entry.e_blocks = ~0ULL;
+}
+
+int omfs_shrink_inode(struct inode *inode)
+{
+	struct omfs_sb_info *sbi = OMFS_SB(inode->i_sb);
+	struct omfs_extent *oe;
+	struct omfs_extent_entry *entry;
+	struct buffer_head *bh;
+	u64 next, last;
+	u32 extent_count;
+	int ret;
+
+	/* traverse extent table, freeing each entry that is greater
+	 * than inode->i_size;
+	 */
+	next = inode->i_ino;
+
+	/* only support truncate -> 0 for now */
+	ret = -EIO;
+	if (inode->i_size != 0)
+		goto out;
+
+	bh = sb_bread(inode->i_sb, clus_to_blk(sbi, next));
+	if (!bh)
+		goto out;
+
+	oe = (struct omfs_extent *)(&bh->b_data[OMFS_EXTENT_START]);
+
+	for (;;) {
+
+		if (omfs_is_bad(sbi, (struct omfs_header *) bh->b_data, next)) {
+			brelse(bh);
+			goto out;
+		}
+
+		extent_count = be32_to_cpu(oe->e_extent_count);
+		last = next;
+		next = be64_to_cpu(oe->e_next);
+		entry = &oe->e_entry;
+
+		/* ignore last entry as it is the terminator */
+		for (; extent_count > 1; extent_count--) {
+			u64 start, count;
+			start = be64_to_cpu(entry->e_cluster);
+			count = be64_to_cpu(entry->e_blocks);
+
+			omfs_clear_range(inode->i_sb, start, (int) count);
+			entry++;
+		}
+		omfs_make_empty_table(bh, (char *) oe - bh->b_data);
+		mark_buffer_dirty(bh);
+		brelse(bh);
+
+		if (last != inode->i_ino)
+			omfs_clear_range(inode->i_sb, last, sbi->s_mirrors);
+
+		if (next == ~0)
+			break;
+
+		bh = sb_bread(inode->i_sb, clus_to_blk(sbi, next));
+		if (!bh)
+			goto out;
+		oe = (struct omfs_extent *) (&bh->b_data[OMFS_EXTENT_CONT]);
+	}
+	ret = 0;
+out:
+	return ret;
+}
+
+static void omfs_truncate(struct inode *inode)
+{
+	omfs_shrink_inode(inode);
+	mark_inode_dirty(inode);
+}
+
+/*
+ * Add new blocks to the current extent, or create new entries/continuations
+ * as necessary.
+ */
+static int omfs_grow_extent(struct inode *inode, struct omfs_extent *oe,
+			u64 *ret_block)
+{
+	struct omfs_extent_entry *terminator;
+	struct omfs_extent_entry *entry = &oe->e_entry;
+	struct omfs_sb_info *sbi = OMFS_SB(inode->i_sb);
+	u32 extent_count = be32_to_cpu(oe->e_extent_count);
+	u64 new_block = 0;
+	u32 max_count;
+	int new_count;
+	int ret = 0;
+
+	/* reached the end of the extent table with no blocks mapped.
+	 * there are three possibilities for adding: grow last extent,
+	 * add a new extent to the current extent table, and add a
+	 * continuation inode.  in last two cases need an allocator for
+	 * sbi->s_cluster_size
+	 */
+
+	/* TODO: handle holes */
+
+	/* should always have a terminator */
+	if (extent_count < 1)
+		return -EIO;
+
+	/* trivially grow current extent, if next block is not taken */
+	terminator = entry + extent_count - 1;
+	if (extent_count > 1) {
+		entry = terminator-1;
+		new_block = be64_to_cpu(entry->e_cluster) +
+			be64_to_cpu(entry->e_blocks);
+
+		if (omfs_allocate_block(inode->i_sb, new_block)) {
+			entry->e_blocks =
+				cpu_to_be64(be64_to_cpu(entry->e_blocks) + 1);
+			terminator->e_blocks = ~(cpu_to_be64(
+				be64_to_cpu(~terminator->e_blocks) + 1));
+			goto out;
+		}
+	}
+	max_count = (sbi->s_sys_blocksize - OMFS_EXTENT_START -
+		sizeof(struct omfs_extent)) /
+		sizeof(struct omfs_extent_entry) + 1;
+
+	/* TODO: add a continuation block here */
+	if (be32_to_cpu(oe->e_extent_count) > max_count-1)
+		return -EIO;
+
+	/* try to allocate a new cluster */
+	ret = omfs_allocate_range(inode->i_sb, 1, sbi->s_clustersize,
+		&new_block, &new_count);
+	if (ret)
+		goto out_fail;
+
+	/* copy terminator down an entry */
+	entry = terminator;
+	terminator++;
+	memcpy(terminator, entry, sizeof(struct omfs_extent_entry));
+
+	entry->e_cluster = cpu_to_be64(new_block);
+	entry->e_blocks = cpu_to_be64((u64) new_count);
+
+	terminator->e_blocks = ~(cpu_to_be64(
+		be64_to_cpu(~terminator->e_blocks) + (u64) new_count));
+
+	/* write in new entry */
+	oe->e_extent_count = cpu_to_be32(1 + be32_to_cpu(oe->e_extent_count));
+
+out:
+	*ret_block = new_block;
+out_fail:
+	return ret;
+}
+
+/*
+ * Scans across the directory table for a given file block number.
+ * If block not found, return 0.
+ */
+static sector_t find_block(struct inode *inode, struct omfs_extent_entry *ent,
+			sector_t block, int count, int *left)
+{
+	/* count > 1 because of terminator */
+	sector_t searched = 0;
+	for (; count > 1; count--) {
+		int numblocks = clus_to_blk(OMFS_SB(inode->i_sb),
+			be64_to_cpu(ent->e_blocks));
+
+		if (block >= searched  &&
+		    block < searched + numblocks) {
+			/*
+			 * found it at cluster + (block - searched)
+			 * numblocks - (block - searched) is remainder
+			 */
+			*left = numblocks - (block - searched);
+			return clus_to_blk(OMFS_SB(inode->i_sb),
+				be64_to_cpu(ent->e_cluster)) +
+				block - searched;
+		}
+		searched += numblocks;
+		ent++;
+	}
+	return 0;
+}
+
+static int omfs_get_block(struct inode *inode, sector_t block,
+			  struct buffer_head *bh_result, int create)
+{
+	struct buffer_head *bh;
+	sector_t next, offset;
+	int ret;
+	u64 new_block;
+	int extent_count;
+	struct omfs_extent *oe;
+	struct omfs_extent_entry *entry;
+	struct omfs_sb_info *sbi = OMFS_SB(inode->i_sb);
+	int max_blocks = bh_result->b_size >> inode->i_blkbits;
+	int remain;
+
+	ret = -EIO;
+	bh = sb_bread(inode->i_sb, clus_to_blk(sbi, inode->i_ino));
+	if (!bh)
+		goto out;
+
+	oe = (struct omfs_extent *)(&bh->b_data[OMFS_EXTENT_START]);
+	next = inode->i_ino;
+
+	for (;;) {
+
+		if (omfs_is_bad(sbi, (struct omfs_header *) bh->b_data, next))
+			goto out_brelse;
+
+		extent_count = be32_to_cpu(oe->e_extent_count);
+		next = be64_to_cpu(oe->e_next);
+		entry = &oe->e_entry;
+
+		offset = find_block(inode, entry, block, extent_count, &remain);
+		if (offset > 0) {
+			ret = 0;
+			map_bh(bh_result, inode->i_sb, offset);
+			if (remain > max_blocks)
+				remain = max_blocks;
+			bh_result->b_size = (remain << inode->i_blkbits);
+			goto out_brelse;
+		}
+		if (next == ~0)
+			break;
+
+		brelse(bh);
+		bh = sb_bread(inode->i_sb, clus_to_blk(sbi, next));
+		if (!bh)
+			goto out;
+		oe = (struct omfs_extent *) (&bh->b_data[OMFS_EXTENT_CONT]);
+	}
+	if (create) {
+		ret = omfs_grow_extent(inode, oe, &new_block);
+		if (ret == 0) {
+			mark_buffer_dirty(bh);
+			mark_inode_dirty(inode);
+			map_bh(bh_result, inode->i_sb,
+					clus_to_blk(sbi, new_block));
+		}
+	}
+out_brelse:
+	brelse(bh);
+out:
+	return ret;
+}
+
+static int omfs_readpage(struct file *file, struct page *page)
+{
+	return block_read_full_page(page, omfs_get_block);
+}
+
+static int omfs_readpages(struct file *file, struct address_space *mapping,
+		struct list_head *pages, unsigned nr_pages)
+{
+	return mpage_readpages(mapping, pages, nr_pages, omfs_get_block);
+}
+
+static int omfs_writepage(struct page *page, struct writeback_control *wbc)
+{
+	return block_write_full_page(page, omfs_get_block, wbc);
+}
+
+static int
+omfs_writepages(struct address_space *mapping, struct writeback_control *wbc)
+{
+	return mpage_writepages(mapping, wbc, omfs_get_block);
+}
+
+static int omfs_write_begin(struct file *file, struct address_space *mapping,
+			loff_t pos, unsigned len, unsigned flags,
+			struct page **pagep, void **fsdata)
+{
+	*pagep = NULL;
+	return block_write_begin(file, mapping, pos, len, flags,
+				pagep, fsdata, omfs_get_block);
+}
+
+static sector_t omfs_bmap(struct address_space *mapping, sector_t block)
+{
+	return generic_block_bmap(mapping, block, omfs_get_block);
+}
+
+struct file_operations omfs_file_operations = {
+	.llseek = generic_file_llseek,
+	.read = do_sync_read,
+	.write = do_sync_write,
+	.aio_read = generic_file_aio_read,
+	.aio_write = generic_file_aio_write,
+	.mmap = generic_file_mmap,
+	.fsync = omfs_sync_file,
+	.splice_read = generic_file_splice_read,
+};
+
+struct inode_operations omfs_file_inops = {
+	.truncate = omfs_truncate
+};
+
+struct address_space_operations omfs_aops = {
+	.readpage = omfs_readpage,
+	.readpages = omfs_readpages,
+	.writepage = omfs_writepage,
+	.writepages = omfs_writepages,
+	.sync_page = block_sync_page,
+	.write_begin = omfs_write_begin,
+	.write_end = generic_write_end,
+	.bmap = omfs_bmap,
+};
+

From 36cc410a6799a205bfc6ccc38abd9d52f2afba64 Mon Sep 17 00:00:00 2001
From: Bob Copeland <me@bobcopeland.com>
Date: Fri, 25 Jul 2008 19:45:17 -0700
Subject: [PATCH 1019/1171] omfs: add bitmap routines

Add block allocation and block bitmap management routines for OMFS.

Signed-off-by: Bob Copeland <me@bobcopeland.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/omfs/bitmap.c | 192 +++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 192 insertions(+)
 create mode 100644 fs/omfs/bitmap.c

diff --git a/fs/omfs/bitmap.c b/fs/omfs/bitmap.c
new file mode 100644
index 000000000000..dc75f22be3f2
--- /dev/null
+++ b/fs/omfs/bitmap.c
@@ -0,0 +1,192 @@
+#include <linux/kernel.h>
+#include <linux/fs.h>
+#include <linux/buffer_head.h>
+#include <asm/div64.h>
+#include "omfs.h"
+
+unsigned long omfs_count_free(struct super_block *sb)
+{
+	unsigned int i;
+	unsigned long sum = 0;
+	struct omfs_sb_info *sbi = OMFS_SB(sb);
+	int nbits = sb->s_blocksize * 8;
+
+	for (i = 0; i < sbi->s_imap_size; i++)
+		sum += nbits - bitmap_weight(sbi->s_imap[i], nbits);
+
+	return sum;
+}
+
+/*
+ *  Counts the run of zero bits starting at bit up to max.
+ *  It handles the case where a run might spill over a buffer.
+ *  Called with bitmap lock.
+ */
+static int count_run(unsigned long **addr, int nbits,
+		int addrlen, int bit, int max)
+{
+	int count = 0;
+	int x;
+
+	for (; addrlen > 0; addrlen--, addr++) {
+		x = find_next_bit(*addr, nbits, bit);
+		count += x - bit;
+
+		if (x < nbits || count > max)
+			return min(count, max);
+
+		bit = 0;
+	}
+	return min(count, max);
+}
+
+/*
+ * Sets or clears the run of count bits starting with bit.
+ * Called with bitmap lock.
+ */
+static int set_run(struct super_block *sb, int map,
+		int nbits, int bit, int count, int set)
+{
+	int i;
+	int err;
+	struct buffer_head *bh;
+	struct omfs_sb_info *sbi = OMFS_SB(sb);
+
+ 	err = -ENOMEM;
+	bh = sb_bread(sb, clus_to_blk(sbi, sbi->s_bitmap_ino) + map);
+	if (!bh)
+		goto out;
+
+	for (i = 0; i < count; i++, bit++) {
+		if (bit >= nbits) {
+			bit = 0;
+			map++;
+
+			mark_buffer_dirty(bh);
+			brelse(bh);
+			bh = sb_bread(sb,
+				clus_to_blk(sbi, sbi->s_bitmap_ino) + map);
+			if (!bh)
+				goto out;
+		}
+		if (set) {
+			set_bit(bit, sbi->s_imap[map]);
+			set_bit(bit, (long *) bh->b_data);
+		} else {
+			clear_bit(bit, sbi->s_imap[map]);
+			clear_bit(bit, (long *) bh->b_data);
+		}
+	}
+	mark_buffer_dirty(bh);
+	brelse(bh);
+	err = 0;
+out:
+	return err;
+}
+
+/*
+ * Tries to allocate exactly one block.  Returns true if sucessful.
+ */
+int omfs_allocate_block(struct super_block *sb, u64 block)
+{
+	struct buffer_head *bh;
+	struct omfs_sb_info *sbi = OMFS_SB(sb);
+	int bits_per_entry = 8 * sb->s_blocksize;
+	int map, bit;
+	int ret = 0;
+	u64 tmp;
+
+	tmp = block;
+	bit = do_div(tmp, bits_per_entry);
+	map = tmp;
+
+	mutex_lock(&sbi->s_bitmap_lock);
+	if (map >= sbi->s_imap_size || test_and_set_bit(bit, sbi->s_imap[map]))
+		goto out;
+
+	if (sbi->s_bitmap_ino > 0) {
+		bh = sb_bread(sb, clus_to_blk(sbi, sbi->s_bitmap_ino) + map);
+		if (!bh)
+			goto out;
+
+		set_bit(bit, (long *) bh->b_data);
+		mark_buffer_dirty(bh);
+		brelse(bh);
+	}
+	ret = 1;
+out:
+	mutex_unlock(&sbi->s_bitmap_lock);
+	return ret;
+}
+
+
+/*
+ *  Tries to allocate a set of blocks.	The request size depends on the
+ *  type: for inodes, we must allocate sbi->s_mirrors blocks, and for file
+ *  blocks, we try to allocate sbi->s_clustersize, but can always get away
+ *  with just one block.
+ */
+int omfs_allocate_range(struct super_block *sb,
+			int min_request,
+			int max_request,
+			u64 *return_block,
+			int *return_size)
+{
+	struct omfs_sb_info *sbi = OMFS_SB(sb);
+	int bits_per_entry = 8 * sb->s_blocksize;
+	int ret = 0;
+	int i, run, bit;
+
+	mutex_lock(&sbi->s_bitmap_lock);
+	for (i = 0; i < sbi->s_imap_size; i++) {
+		bit = 0;
+		while (bit < bits_per_entry) {
+			bit = find_next_zero_bit(sbi->s_imap[i], bits_per_entry,
+				bit);
+
+			if (bit == bits_per_entry)
+				break;
+
+			run = count_run(&sbi->s_imap[i], bits_per_entry,
+				sbi->s_imap_size-i, bit, max_request);
+
+			if (run >= min_request)
+				goto found;
+			bit += run;
+		}
+	}
+	ret = -ENOSPC;
+	goto out;
+
+found:
+	*return_block = i * bits_per_entry + bit;
+	*return_size = run;
+	ret = set_run(sb, i, bits_per_entry, bit, run, 1);
+
+out:
+	mutex_unlock(&sbi->s_bitmap_lock);
+	return ret;
+}
+
+/*
+ * Clears count bits starting at a given block.
+ */
+int omfs_clear_range(struct super_block *sb, u64 block, int count)
+{
+	struct omfs_sb_info *sbi = OMFS_SB(sb);
+	int bits_per_entry = 8 * sb->s_blocksize;
+	u64 tmp;
+	int map, bit, ret;
+
+	tmp = block;
+	bit = do_div(tmp, bits_per_entry);
+	map = tmp;
+
+	if (map >= sbi->s_imap_size)
+		return 0;
+
+	mutex_lock(&sbi->s_bitmap_lock);
+	ret = set_run(sb, map, bits_per_entry, bit, count, 0);
+	mutex_unlock(&sbi->s_bitmap_lock);
+	return ret;
+}

From 63ca8ce2a2641f9cb5f0add33ced4591681d1cd7 Mon Sep 17 00:00:00 2001
From: Bob Copeland <me@bobcopeland.com>
Date: Fri, 25 Jul 2008 19:45:17 -0700
Subject: [PATCH 1020/1171] omfs: update kbuild to include OMFS

Adds OMFS to the fs Kconfig and Makefile

Signed-off-by: Bob Copeland <me@bobcopeland.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/Kconfig       | 13 +++++++++++++
 fs/Makefile      |  1 +
 fs/omfs/Makefile |  4 ++++
 3 files changed, 18 insertions(+)
 create mode 100644 fs/omfs/Makefile

diff --git a/fs/Kconfig b/fs/Kconfig
index 97e3bdedb1e6..d3873583360b 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -1383,6 +1383,19 @@ config MINIX_FS
 	  partition (the one containing the directory /) cannot be compiled as
 	  a module.
 
+config OMFS_FS
+	tristate "SonicBlue Optimized MPEG File System support"
+	depends on BLOCK
+	select CRC_ITU_T
+	help
+	  This is the proprietary file system used by the Rio Karma music
+	  player and ReplayTV DVR.  Despite the name, this filesystem is not
+	  more efficient than a standard FS for MPEG files, in fact likely
+	  the opposite is true.  Say Y if you have either of these devices
+	  and wish to mount its disk.
+
+	  To compile this file system support as a module, choose M here: the
+	  module will be called omfs.  If unsure, say N.
 
 config HPFS_FS
 	tristate "OS/2 HPFS file system support"
diff --git a/fs/Makefile b/fs/Makefile
index 3b2178b4bb66..a1482a5eff15 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -111,6 +111,7 @@ obj-$(CONFIG_ADFS_FS)		+= adfs/
 obj-$(CONFIG_FUSE_FS)		+= fuse/
 obj-$(CONFIG_UDF_FS)		+= udf/
 obj-$(CONFIG_SUN_OPENPROMFS)	+= openpromfs/
+obj-$(CONFIG_OMFS_FS)		+= omfs/
 obj-$(CONFIG_JFS_FS)		+= jfs/
 obj-$(CONFIG_XFS_FS)		+= xfs/
 obj-$(CONFIG_9P_FS)		+= 9p/
diff --git a/fs/omfs/Makefile b/fs/omfs/Makefile
new file mode 100644
index 000000000000..8b82b63f1129
--- /dev/null
+++ b/fs/omfs/Makefile
@@ -0,0 +1,4 @@
+
+obj-$(CONFIG_OMFS_FS) += omfs.o
+
+omfs-y := bitmap.o dir.o file.o inode.o

From 0ad122d901977890de554fdd3ff65474efc1a9bf Mon Sep 17 00:00:00 2001
From: Bob Copeland <me@bobcopeland.com>
Date: Fri, 25 Jul 2008 19:45:18 -0700
Subject: [PATCH 1021/1171] omfs: add MAINTAINERS entry

Add the MAINTAINERS entry for OMFS.

Signed-off-by: Bob Copeland <me@bobcopeland.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 MAINTAINERS | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 4cbf6016a9b9..03c5d6ccb9f8 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -3123,6 +3123,12 @@ W:	http://oss.oracle.com/projects/ocfs2/
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/mfasheh/ocfs2.git
 S:	Supported
 
+OMFS FILESYSTEM
+P:	Bob Copeland
+M:	me@bobcopeland.com
+L:	linux-karma-devel@lists.sourceforge.net
+S:	Maintained
+
 OMNIKEY CARDMAN 4000 DRIVER
 P:	Harald Welte
 M:	laforge@gnumonks.org

From 42a9a58361d3b65f4a7c2ad280cc1ec9b657e7d2 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Fri, 25 Jul 2008 19:45:18 -0700
Subject: [PATCH 1022/1171] cris: use the common ascii hex helpers

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Cc: Mikael Starvik <starvik@axis.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/cris/arch-v10/kernel/kgdb.c | 79 ++++++++++----------------------
 arch/cris/arch-v32/kernel/kgdb.c | 60 ++++++------------------
 2 files changed, 37 insertions(+), 102 deletions(-)

diff --git a/arch/cris/arch-v10/kernel/kgdb.c b/arch/cris/arch-v10/kernel/kgdb.c
index a3ca55150745..6fea45f2e40c 100644
--- a/arch/cris/arch-v10/kernel/kgdb.c
+++ b/arch/cris/arch-v10/kernel/kgdb.c
@@ -278,14 +278,6 @@ void putDebugChar (int val);
 
 void enableDebugIRQ (void);
 
-/* Returns the character equivalent of a nibble, bit 7, 6, 5, and 4 of a byte,
-   represented by int x. */
-static char highhex (int x);
-
-/* Returns the character equivalent of a nibble, bit 3, 2, 1, and 0 of a byte,
-   represented by int x. */
-static char lowhex (int x);
-
 /* Returns the integer equivalent of a hexadecimal character. */
 static int hex (char ch);
 
@@ -356,9 +348,6 @@ extern unsigned char executing_task;
 /* Run-length encoding maximum length. Send 64 at most. */
 #define RUNLENMAX 64
 
-/* Definition of all valid hexadecimal characters */
-static const char hexchars[] = "0123456789abcdef";
-
 /* The inbound/outbound buffers used in packet I/O */
 static char remcomInBuffer[BUFMAX];
 static char remcomOutBuffer[BUFMAX];
@@ -499,8 +488,8 @@ gdb_cris_strtol (const char *s, char **endptr, int base)
 	char *sd;
 	int x = 0;
 	
-	for (s1 = (char*)s; (sd = gdb_cris_memchr(hexchars, *s1, base)) != NULL; ++s1)
-		x = x * base + (sd - hexchars);
+	for (s1 = (char*)s; (sd = gdb_cris_memchr(hex_asc, *s1, base)) != NULL; ++s1)
+		x = x * base + (sd - hex_asc);
         
         if (endptr)
         {
@@ -670,22 +659,6 @@ read_register (char regno, unsigned int *valptr)
 }
 
 /********************************** Packet I/O ******************************/
-/* Returns the character equivalent of a nibble, bit 7, 6, 5, and 4 of a byte,
-   represented by int x. */
-static inline char
-highhex(int x)
-{
-	return hexchars[(x >> 4) & 0xf];
-}
-
-/* Returns the character equivalent of a nibble, bit 3, 2, 1, and 0 of a byte,
-   represented by int x. */
-static inline char
-lowhex(int x)
-{
-	return hexchars[x & 0xf];
-}
-
 /* Returns the integer equivalent of a hexadecimal character. */
 static int
 hex (char ch)
@@ -721,8 +694,7 @@ mem2hex(char *buf, unsigned char *mem, int count)
                 /* Valid mem address. */
                 for (i = 0; i < count; i++) {
                         ch = *mem++;
-                        *buf++ = highhex (ch);
-                        *buf++ = lowhex (ch);
+			buf = pack_hex_byte(buf, ch);
                 }
         }
         
@@ -857,9 +829,9 @@ putpacket(char *buffer)
 				src++;
 			}
 		}
-		putDebugChar ('#');
-		putDebugChar (highhex (checksum));
-		putDebugChar (lowhex (checksum));
+		putDebugChar('#');
+		putDebugChar(hex_asc_hi(checksum));
+		putDebugChar(hex_asc_lo(checksum));
 	} while(kgdb_started && (getDebugChar() != '+'));
 }
 
@@ -895,9 +867,8 @@ stub_is_stopped(int sigval)
         
 	/* Send trap type (converted to signal) */
 
-	*ptr++ = 'T';	
-	*ptr++ = highhex (sigval);
-	*ptr++ = lowhex (sigval);
+	*ptr++ = 'T';
+	ptr = pack_hex_byte(ptr, sigval);
 
 	/* Send register contents. We probably only need to send the
 	 * PC, frame pointer and stack pointer here. Other registers will be
@@ -910,9 +881,7 @@ stub_is_stopped(int sigval)
                 status = read_register (regno, &reg_cont);
                 
 		if (status == SUCCESS) {
-                        
-                        *ptr++ = highhex (regno);
-                        *ptr++ = lowhex (regno);
+			ptr = pack_hex_byte(ptr, regno);
                         *ptr++ = ':';
 
                         ptr = mem2hex(ptr, (unsigned char *)&reg_cont,
@@ -937,8 +906,8 @@ stub_is_stopped(int sigval)
 	/* Store thread:r...; with the executing task TID. */
 	gdb_cris_strcpy (&remcomOutBuffer[pos], "thread:");
 	pos += gdb_cris_strlen ("thread:");
-	remcomOutBuffer[pos++] = highhex (executing_task);
-	remcomOutBuffer[pos++] = lowhex (executing_task);
+	remcomOutBuffer[pos++] = hex_asc_hi(executing_task);
+	remcomOutBuffer[pos++] = hex_asc_lo(executing_task);
 	gdb_cris_strcpy (&remcomOutBuffer[pos], ";");
 #endif
 
@@ -1126,8 +1095,8 @@ handle_exception (int sigval)
 				   Success: SAA, where AA is the signal number.
 				   Failure: void. */
 				remcomOutBuffer[0] = 'S';
-				remcomOutBuffer[1] = highhex (sigval);
-				remcomOutBuffer[2] = lowhex (sigval);
+				remcomOutBuffer[1] = hex_asc_hi(sigval);
+				remcomOutBuffer[2] = hex_asc_lo(sigval);
 				remcomOutBuffer[3] = 0;
 				break;
 				
@@ -1224,23 +1193,23 @@ handle_exception (int sigval)
 						case 'C':
 							/* Identify the remote current thread. */
 							gdb_cris_strcpy (&remcomOutBuffer[0], "QC");
-							remcomOutBuffer[2] = highhex (current_thread_c);
-							remcomOutBuffer[3] = lowhex (current_thread_c);
+							remcomOutBuffer[2] = hex_asc_hi(current_thread_c);
+							remcomOutBuffer[3] = hex_asc_lo(current_thread_c);
 							remcomOutBuffer[4] = '\0';
 							break;
 						case 'L':
 							gdb_cris_strcpy (&remcomOutBuffer[0], "QM");
 							/* Reply with number of threads. */
 							if (os_is_started()) {
-								remcomOutBuffer[2] = highhex (number_of_tasks);
-								remcomOutBuffer[3] = lowhex (number_of_tasks);
+								remcomOutBuffer[2] = hex_asc_hi(number_of_tasks);
+								remcomOutBuffer[3] = hex_asc_lo(number_of_tasks);
 							}
 							else {
-								remcomOutBuffer[2] = highhex (0);
-								remcomOutBuffer[3] = lowhex (1);
+								remcomOutBuffer[2] = hex_asc_hi(0);
+								remcomOutBuffer[3] = hex_asc_lo(1);
 							}
 							/* Done with the reply. */
-							remcomOutBuffer[4] = lowhex (1);
+							remcomOutBuffer[4] = hex_asc_lo(1);
 							pos = 5;
 							/* Expects the argument thread id. */
 							for (; pos < (5 + HEXCHARS_IN_THREAD_ID); pos++)
@@ -1251,16 +1220,16 @@ handle_exception (int sigval)
 								for (thread_id = 0; thread_id < number_of_tasks; thread_id++) {
 									nextpos = pos + HEXCHARS_IN_THREAD_ID - 1;
 									for (; pos < nextpos; pos ++)
-										remcomOutBuffer[pos] = lowhex (0);
-									remcomOutBuffer[pos++] = lowhex (thread_id);
+										remcomOutBuffer[pos] = hex_asc_lo(0);
+									remcomOutBuffer[pos++] = hex_asc_lo(thread_id);
 								}
 							}
 							else {
 								/* Store the thread identifier of the boot task. */
 								nextpos = pos + HEXCHARS_IN_THREAD_ID - 1;
 								for (; pos < nextpos; pos ++)
-									remcomOutBuffer[pos] = lowhex (0);
-								remcomOutBuffer[pos++] = lowhex (current_thread_c);
+									remcomOutBuffer[pos] = hex_asc_lo(0);
+								remcomOutBuffer[pos++] = hex_asc_lo(current_thread_c);
 							}
 							remcomOutBuffer[pos] = '\0';
 							break;
diff --git a/arch/cris/arch-v32/kernel/kgdb.c b/arch/cris/arch-v32/kernel/kgdb.c
index 4e2e2e271efb..8bd5a5bc0dc7 100644
--- a/arch/cris/arch-v32/kernel/kgdb.c
+++ b/arch/cris/arch-v32/kernel/kgdb.c
@@ -398,14 +398,6 @@ void putDebugChar(int val)
 }
 #endif
 
-/* Returns the character equivalent of a nibble, bit 7, 6, 5, and 4 of a byte,
-   represented by int x. */
-static char highhex(int x);
-
-/* Returns the character equivalent of a nibble, bit 3, 2, 1, and 0 of a byte,
-   represented by int x. */
-static char lowhex(int x);
-
 /* Returns the integer equivalent of a hexadecimal character. */
 static int hex(char ch);
 
@@ -464,9 +456,6 @@ void breakpoint(void);
 /* Run-length encoding maximum length. Send 64 at most. */
 #define RUNLENMAX 64
 
-/* Definition of all valid hexadecimal characters */
-static const char hexchars[] = "0123456789abcdef";
-
 /* The inbound/outbound buffers used in packet I/O */
 static char input_buffer[BUFMAX];
 static char output_buffer[BUFMAX];
@@ -550,8 +539,8 @@ gdb_cris_strtol(const char *s, char **endptr, int base)
 	char *sd;
 	int x = 0;
 
-	for (s1 = (char*)s; (sd = gdb_cris_memchr(hexchars, *s1, base)) != NULL; ++s1)
-		x = x * base + (sd - hexchars);
+	for (s1 = (char*)s; (sd = gdb_cris_memchr(hex_asc, *s1, base)) != NULL; ++s1)
+		x = x * base + (sd - hex_asc);
 
         if (endptr) {
                 /* Unconverted suffix is stored in endptr unless endptr is NULL. */
@@ -655,22 +644,6 @@ read_register(char regno, unsigned int *valptr)
 }
 
 /********************************** Packet I/O ******************************/
-/* Returns the character equivalent of a nibble, bit 7, 6, 5, and 4 of a byte,
-   represented by int x. */
-static inline char
-highhex(int x)
-{
-	return hexchars[(x >> 4) & 0xf];
-}
-
-/* Returns the character equivalent of a nibble, bit 3, 2, 1, and 0 of a byte,
-   represented by int x. */
-static inline char
-lowhex(int x)
-{
-	return hexchars[x & 0xf];
-}
-
 /* Returns the integer equivalent of a hexadecimal character. */
 static int
 hex(char ch)
@@ -704,8 +677,7 @@ mem2hex(char *buf, unsigned char *mem, int count)
                 /* Valid mem address. */
 		for (i = 0; i < count; i++) {
 			ch = *mem++;
-			*buf++ = highhex (ch);
-			*buf++ = lowhex (ch);
+			buf = pack_hex_byte(buf, ch);
 		}
         }
         /* Terminate properly. */
@@ -723,8 +695,7 @@ mem2hex_nbo(char *buf, unsigned char *mem, int count)
 	mem += count - 1;
 	for (i = 0; i < count; i++) {
 		ch = *mem--;
-		*buf++ = highhex (ch);
-		*buf++ = lowhex (ch);
+		buf = pack_hex_byte(buf, ch);
         }
 
         /* Terminate properly. */
@@ -862,8 +833,8 @@ putpacket(char *buffer)
 			}
 		}
 		putDebugChar('#');
-		putDebugChar(highhex (checksum));
-		putDebugChar(lowhex (checksum));
+		putDebugChar(hex_asc_hi(checksum));
+		putDebugChar(hex_asc_lo(checksum));
 	} while(kgdb_started && (getDebugChar() != '+'));
 }
 
@@ -909,8 +880,7 @@ stub_is_stopped(int sigval)
 	/* Send trap type (converted to signal) */
 
 	*ptr++ = 'T';
-	*ptr++ = highhex(sigval);
-	*ptr++ = lowhex(sigval);
+	ptr = pack_hex_byte(ptr, sigval);
 
 	if (((reg.exs & 0xff00) >> 8) == 0xc) {
 
@@ -1018,30 +988,26 @@ stub_is_stopped(int sigval)
 	}
 	/* Only send PC, frame and stack pointer. */
 	read_register(PC, &reg_cont);
-	*ptr++ = highhex(PC);
-	*ptr++ = lowhex(PC);
+	ptr = pack_hex_byte(PC);
 	*ptr++ = ':';
 	ptr = mem2hex(ptr, (unsigned char *)&reg_cont, register_size[PC]);
 	*ptr++ = ';';
 
 	read_register(R8, &reg_cont);
-	*ptr++ = highhex(R8);
-	*ptr++ = lowhex(R8);
+	ptr = pack_hex_byte(R8);
 	*ptr++ = ':';
 	ptr = mem2hex(ptr, (unsigned char *)&reg_cont, register_size[R8]);
 	*ptr++ = ';';
 
 	read_register(SP, &reg_cont);
-	*ptr++ = highhex(SP);
-	*ptr++ = lowhex(SP);
+	ptr = pack_hex_byte(SP);
 	*ptr++ = ':';
 	ptr = mem2hex(ptr, (unsigned char *)&reg_cont, register_size[SP]);
 	*ptr++ = ';';
 
 	/* Send ERP as well; this will save us an entire register fetch in some cases. */
         read_register(ERP, &reg_cont);
-        *ptr++ = highhex(ERP);
-        *ptr++ = lowhex(ERP);
+	ptr = pack_hex_byte(ERP);
         *ptr++ = ':';
         ptr = mem2hex(ptr, (unsigned char *)&reg_cont, register_size[ERP]);
         *ptr++ = ';';
@@ -1533,8 +1499,8 @@ handle_exception(int sigval)
 				   Success: SAA, where AA is the signal number.
 				   Failure: void. */
 				output_buffer[0] = 'S';
-				output_buffer[1] = highhex(sigval);
-				output_buffer[2] = lowhex(sigval);
+				output_buffer[1] = hex_asc_hi(sigval);
+				output_buffer[2] = hex_asc_lo(sigval);
 				output_buffer[3] = 0;
 				break;
 

From 19caeed6339aec02901e2f4c49d8e1d3d6090559 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Fri, 25 Jul 2008 19:45:19 -0700
Subject: [PATCH 1023/1171] frv: use the common ascii hex helpers

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Acked-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/frv/kernel/gdb-stub.c | 88 ++++++++++++++++----------------------
 1 file changed, 36 insertions(+), 52 deletions(-)

diff --git a/arch/frv/kernel/gdb-stub.c b/arch/frv/kernel/gdb-stub.c
index 48a0393e7cee..7ca8a6b19ac9 100644
--- a/arch/frv/kernel/gdb-stub.c
+++ b/arch/frv/kernel/gdb-stub.c
@@ -182,8 +182,6 @@ extern volatile u32 __attribute__((section(".bss"))) gdbstub_trace_through_excep
 static char	input_buffer[BUFMAX];
 static char	output_buffer[BUFMAX];
 
-static const char hexchars[] = "0123456789abcdef";
-
 static const char *regnames[] = {
 	"PSR ", "ISR ", "CCR ", "CCCR",
 	"LR  ", "LCR ", "PC  ", "_stt",
@@ -383,8 +381,8 @@ static int gdbstub_send_packet(char *buffer)
 		}
 
 		gdbstub_tx_char('#');
-		gdbstub_tx_char(hexchars[checksum >> 4]);
-		gdbstub_tx_char(hexchars[checksum & 0xf]);
+		gdbstub_tx_char(hex_asc_hi(checksum));
+		gdbstub_tx_char(hex_asc_lo(checksum));
 
 	} while (gdbstub_rx_char(&ch,0),
 #ifdef GDBSTUB_DEBUG_PROTOCOL
@@ -674,8 +672,7 @@ static unsigned char *mem2hex(const void *_mem, char *buf, int count, int may_fa
 	if ((uint32_t)mem&1 && count>=1) {
 		if (!gdbstub_read_byte(mem,ch))
 			return NULL;
-		*buf++ = hexchars[ch[0] >> 4];
-		*buf++ = hexchars[ch[0] & 0xf];
+		buf = pack_hex_byte(buf, ch[0]);
 		mem++;
 		count--;
 	}
@@ -683,10 +680,8 @@ static unsigned char *mem2hex(const void *_mem, char *buf, int count, int may_fa
 	if ((uint32_t)mem&3 && count>=2) {
 		if (!gdbstub_read_word(mem,(uint16_t *)ch))
 			return NULL;
-		*buf++ = hexchars[ch[0] >> 4];
-		*buf++ = hexchars[ch[0] & 0xf];
-		*buf++ = hexchars[ch[1] >> 4];
-		*buf++ = hexchars[ch[1] & 0xf];
+		buf = pack_hex_byte(buf, ch[0]);
+		buf = pack_hex_byte(buf, ch[1]);
 		mem += 2;
 		count -= 2;
 	}
@@ -694,14 +689,10 @@ static unsigned char *mem2hex(const void *_mem, char *buf, int count, int may_fa
 	while (count>=4) {
 		if (!gdbstub_read_dword(mem,(uint32_t *)ch))
 			return NULL;
-		*buf++ = hexchars[ch[0] >> 4];
-		*buf++ = hexchars[ch[0] & 0xf];
-		*buf++ = hexchars[ch[1] >> 4];
-		*buf++ = hexchars[ch[1] & 0xf];
-		*buf++ = hexchars[ch[2] >> 4];
-		*buf++ = hexchars[ch[2] & 0xf];
-		*buf++ = hexchars[ch[3] >> 4];
-		*buf++ = hexchars[ch[3] & 0xf];
+		buf = pack_hex_byte(buf, ch[0]);
+		buf = pack_hex_byte(buf, ch[1]);
+		buf = pack_hex_byte(buf, ch[2]);
+		buf = pack_hex_byte(buf, ch[3]);
 		mem += 4;
 		count -= 4;
 	}
@@ -709,10 +700,8 @@ static unsigned char *mem2hex(const void *_mem, char *buf, int count, int may_fa
 	if (count>=2) {
 		if (!gdbstub_read_word(mem,(uint16_t *)ch))
 			return NULL;
-		*buf++ = hexchars[ch[0] >> 4];
-		*buf++ = hexchars[ch[0] & 0xf];
-		*buf++ = hexchars[ch[1] >> 4];
-		*buf++ = hexchars[ch[1] & 0xf];
+		buf = pack_hex_byte(buf, ch[0]);
+		buf = pack_hex_byte(buf, ch[1]);
 		mem += 2;
 		count -= 2;
 	}
@@ -720,8 +709,7 @@ static unsigned char *mem2hex(const void *_mem, char *buf, int count, int may_fa
 	if (count>=1) {
 		if (!gdbstub_read_byte(mem,ch))
 			return NULL;
-		*buf++ = hexchars[ch[0] >> 4];
-		*buf++ = hexchars[ch[0] & 0xf];
+		buf = pack_hex_byte(buf, ch[0]);
 	}
 
 	*buf = 0;
@@ -1471,22 +1459,22 @@ void gdbstub(int sigval)
 		*ptr++ = 'O';
 		ptr = mem2hex(title, ptr, sizeof(title) - 1,0);
 
-		hx = hexchars[(brr & 0xf0000000) >> 28];
-		*ptr++ = hexchars[hx >> 4];	*ptr++ = hexchars[hx & 0xf];
-		hx = hexchars[(brr & 0x0f000000) >> 24];
-		*ptr++ = hexchars[hx >> 4];	*ptr++ = hexchars[hx & 0xf];
-		hx = hexchars[(brr & 0x00f00000) >> 20];
-		*ptr++ = hexchars[hx >> 4];	*ptr++ = hexchars[hx & 0xf];
-		hx = hexchars[(brr & 0x000f0000) >> 16];
-		*ptr++ = hexchars[hx >> 4];	*ptr++ = hexchars[hx & 0xf];
-		hx = hexchars[(brr & 0x0000f000) >> 12];
-		*ptr++ = hexchars[hx >> 4];	*ptr++ = hexchars[hx & 0xf];
-		hx = hexchars[(brr & 0x00000f00) >> 8];
-		*ptr++ = hexchars[hx >> 4];	*ptr++ = hexchars[hx & 0xf];
-		hx = hexchars[(brr & 0x000000f0) >> 4];
-		*ptr++ = hexchars[hx >> 4];	*ptr++ = hexchars[hx & 0xf];
-		hx = hexchars[(brr & 0x0000000f)];
-		*ptr++ = hexchars[hx >> 4];	*ptr++ = hexchars[hx & 0xf];
+		hx = hex_asc_hi(brr >> 24);
+		ptr = pack_hex_byte(ptr, hx);
+		hx = hex_asc_lo(brr >> 24);
+		ptr = pack_hex_byte(ptr, hx);
+		hx = hex_asc_hi(brr >> 16);
+		ptr = pack_hex_byte(ptr, hx);
+		hx = hex_asc_lo(brr >> 16);
+		ptr = pack_hex_byte(ptr, hx);
+		hx = hex_asc_hi(brr >> 8);
+		ptr = pack_hex_byte(ptr, hx);
+		hx = hex_asc_lo(brr >> 8);
+		ptr = pack_hex_byte(ptr, hx);
+		hx = hex_asc_hi(brr);
+		ptr = pack_hex_byte(ptr, hx);
+		hx = hex_asc_lo(brr);
+		ptr = pack_hex_byte(ptr, hx);
 
 		ptr = mem2hex(crlf, ptr, sizeof(crlf) - 1, 0);
 		*ptr = 0;
@@ -1500,12 +1488,10 @@ void gdbstub(int sigval)
 
 	/* Send trap type (converted to signal) */
 	*ptr++ = 'T';
-	*ptr++ = hexchars[sigval >> 4];
-	*ptr++ = hexchars[sigval & 0xf];
+	ptr = pack_hex_byte(ptr, sigval);
 
 	/* Send Error PC */
-	*ptr++ = hexchars[GDB_REG_PC >> 4];
-	*ptr++ = hexchars[GDB_REG_PC & 0xf];
+	ptr = pack_hex_byte(ptr, GDB_REG_PC);
 	*ptr++ = ':';
 	ptr = mem2hex(&__debug_frame->pc, ptr, 4, 0);
 	*ptr++ = ';';
@@ -1513,8 +1499,7 @@ void gdbstub(int sigval)
 	/*
 	 * Send frame pointer
 	 */
-	*ptr++ = hexchars[GDB_REG_FP >> 4];
-	*ptr++ = hexchars[GDB_REG_FP & 0xf];
+	ptr = pack_hex_byte(ptr, GDB_REG_FP);
 	*ptr++ = ':';
 	ptr = mem2hex(&__debug_frame->fp, ptr, 4, 0);
 	*ptr++ = ';';
@@ -1522,8 +1507,7 @@ void gdbstub(int sigval)
 	/*
 	 * Send stack pointer
 	 */
-	*ptr++ = hexchars[GDB_REG_SP >> 4];
-	*ptr++ = hexchars[GDB_REG_SP & 0xf];
+	ptr = pack_hex_byte(ptr, GDB_REG_SP);
 	*ptr++ = ':';
 	ptr = mem2hex(&__debug_frame->sp, ptr, 4, 0);
 	*ptr++ = ';';
@@ -1548,8 +1532,8 @@ void gdbstub(int sigval)
 			/* request repeat of last signal number */
 		case '?':
 			output_buffer[0] = 'S';
-			output_buffer[1] = hexchars[sigval >> 4];
-			output_buffer[2] = hexchars[sigval & 0xf];
+			output_buffer[1] = hex_asc_hi(sigval);
+			output_buffer[2] = hex_asc_lo(sigval);
 			output_buffer[3] = 0;
 			break;
 
@@ -2059,8 +2043,8 @@ void gdbstub_exit(int status)
 	}
 
 	gdbstub_tx_char('#');
-	gdbstub_tx_char(hexchars[checksum >> 4]);
-	gdbstub_tx_char(hexchars[checksum & 0xf]);
+	gdbstub_tx_char(hex_asc_hi(checksum));
+	gdbstub_tx_char(hex_asc_lo(checksum));
 
 	/* make sure the output is flushed, or else RedBoot might clobber it */
 	gdbstub_tx_char('-');

From 2682497245e7d22160ae63032c378745a7f2cfe5 Mon Sep 17 00:00:00 2001
From: Harvey Harrison <harvey.harrison@gmail.com>
Date: Fri, 25 Jul 2008 19:45:20 -0700
Subject: [PATCH 1024/1171] mn10300: use the common ascii hex helpers

Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com>
Acked-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/mn10300/kernel/gdb-stub.c | 108 ++++++++++++++-------------------
 1 file changed, 46 insertions(+), 62 deletions(-)

diff --git a/arch/mn10300/kernel/gdb-stub.c b/arch/mn10300/kernel/gdb-stub.c
index 21891c71d549..54be6afb5555 100644
--- a/arch/mn10300/kernel/gdb-stub.c
+++ b/arch/mn10300/kernel/gdb-stub.c
@@ -163,8 +163,6 @@ static char	input_buffer[BUFMAX];
 static char	output_buffer[BUFMAX];
 static char	trans_buffer[BUFMAX];
 
-static const char hexchars[] = "0123456789abcdef";
-
 struct gdbstub_bkpt {
 	u8	*addr;		/* address of breakpoint */
 	u8	len;		/* size of breakpoint */
@@ -363,8 +361,8 @@ static int putpacket(char *buffer)
 		}
 
 		gdbstub_io_tx_char('#');
-		gdbstub_io_tx_char(hexchars[checksum >> 4]);
-		gdbstub_io_tx_char(hexchars[checksum & 0xf]);
+		gdbstub_io_tx_char(hex_asc_hi(checksum));
+		gdbstub_io_tx_char(hex_asc_lo(checksum));
 
 	} while (gdbstub_io_rx_char(&ch, 0),
 		 ch == '-' && (gdbstub_io("### GDB Rx NAK\n"), 0),
@@ -822,8 +820,7 @@ unsigned char *mem2hex(const void *_mem, char *buf, int count, int may_fault)
 	if ((u32) mem & 1 && count >= 1) {
 		if (gdbstub_read_byte(mem, ch) != 0)
 			return 0;
-		*buf++ = hexchars[ch[0] >> 4];
-		*buf++ = hexchars[ch[0] & 0xf];
+		buf = pack_hex_byte(buf, ch[0]);
 		mem++;
 		count--;
 	}
@@ -831,10 +828,8 @@ unsigned char *mem2hex(const void *_mem, char *buf, int count, int may_fault)
 	if ((u32) mem & 3 && count >= 2) {
 		if (gdbstub_read_word(mem, ch) != 0)
 			return 0;
-		*buf++ = hexchars[ch[0] >> 4];
-		*buf++ = hexchars[ch[0] & 0xf];
-		*buf++ = hexchars[ch[1] >> 4];
-		*buf++ = hexchars[ch[1] & 0xf];
+		buf = pack_hex_byte(buf, ch[0]);
+		buf = pack_hex_byte(buf, ch[1]);
 		mem += 2;
 		count -= 2;
 	}
@@ -842,14 +837,10 @@ unsigned char *mem2hex(const void *_mem, char *buf, int count, int may_fault)
 	while (count >= 4) {
 		if (gdbstub_read_dword(mem, ch) != 0)
 			return 0;
-		*buf++ = hexchars[ch[0] >> 4];
-		*buf++ = hexchars[ch[0] & 0xf];
-		*buf++ = hexchars[ch[1] >> 4];
-		*buf++ = hexchars[ch[1] & 0xf];
-		*buf++ = hexchars[ch[2] >> 4];
-		*buf++ = hexchars[ch[2] & 0xf];
-		*buf++ = hexchars[ch[3] >> 4];
-		*buf++ = hexchars[ch[3] & 0xf];
+		buf = pack_hex_byte(buf, ch[0]);
+		buf = pack_hex_byte(buf, ch[1]);
+		buf = pack_hex_byte(buf, ch[2]);
+		buf = pack_hex_byte(buf, ch[3]);
 		mem += 4;
 		count -= 4;
 	}
@@ -857,10 +848,8 @@ unsigned char *mem2hex(const void *_mem, char *buf, int count, int may_fault)
 	if (count >= 2) {
 		if (gdbstub_read_word(mem, ch) != 0)
 			return 0;
-		*buf++ = hexchars[ch[0] >> 4];
-		*buf++ = hexchars[ch[0] & 0xf];
-		*buf++ = hexchars[ch[1] >> 4];
-		*buf++ = hexchars[ch[1] & 0xf];
+		buf = pack_hex_byte(buf, ch[0]);
+		buf = pack_hex_byte(buf, ch[1]);
 		mem += 2;
 		count -= 2;
 	}
@@ -868,8 +857,7 @@ unsigned char *mem2hex(const void *_mem, char *buf, int count, int may_fault)
 	if (count >= 1) {
 		if (gdbstub_read_byte(mem, ch) != 0)
 			return 0;
-		*buf++ = hexchars[ch[0] >> 4];
-		*buf++ = hexchars[ch[0] & 0xf];
+		buf = pack_hex_byte(buf, ch[0]);
 	}
 
 	*buf = 0;
@@ -1304,14 +1292,14 @@ static int gdbstub(struct pt_regs *regs, enum exception_code excep)
 		*ptr++ = 'O';
 		ptr = mem2hex(title, ptr, sizeof(title) - 1, 0);
 
-		hx = hexchars[(excep & 0xf000) >> 12];
-		*ptr++ = hexchars[hx >> 4];	*ptr++ = hexchars[hx & 0xf];
-		hx = hexchars[(excep & 0x0f00) >> 8];
-		*ptr++ = hexchars[hx >> 4];	*ptr++ = hexchars[hx & 0xf];
-		hx = hexchars[(excep & 0x00f0) >> 4];
-		*ptr++ = hexchars[hx >> 4];	*ptr++ = hexchars[hx & 0xf];
-		hx = hexchars[(excep & 0x000f)];
-		*ptr++ = hexchars[hx >> 4];	*ptr++ = hexchars[hx & 0xf];
+		hx = hex_asc_hi(excep >> 8);
+		ptr = pack_hex_byte(ptr, hx);
+		hx = hex_asc_lo(excep >> 8);
+		ptr = pack_hex_byte(ptr, hx);
+		hx = hex_asc_hi(excep);
+		ptr = pack_hex_byte(ptr, hx);
+		hx = hex_asc_lo(excep);
+		ptr = pack_hex_byte(ptr, hx);
 
 		ptr = mem2hex(crlf, ptr, sizeof(crlf) - 1, 0);
 		*ptr = 0;
@@ -1322,22 +1310,22 @@ static int gdbstub(struct pt_regs *regs, enum exception_code excep)
 		*ptr++ = 'O';
 		ptr = mem2hex(tbcberr, ptr, sizeof(tbcberr) - 1, 0);
 
-		hx = hexchars[(bcberr & 0xf0000000) >> 28];
-		*ptr++ = hexchars[hx >> 4];	*ptr++ = hexchars[hx & 0xf];
-		hx = hexchars[(bcberr & 0x0f000000) >> 24];
-		*ptr++ = hexchars[hx >> 4];	*ptr++ = hexchars[hx & 0xf];
-		hx = hexchars[(bcberr & 0x00f00000) >> 20];
-		*ptr++ = hexchars[hx >> 4];	*ptr++ = hexchars[hx & 0xf];
-		hx = hexchars[(bcberr & 0x000f0000) >> 16];
-		*ptr++ = hexchars[hx >> 4];	*ptr++ = hexchars[hx & 0xf];
-		hx = hexchars[(bcberr & 0x0000f000) >> 12];
-		*ptr++ = hexchars[hx >> 4];	*ptr++ = hexchars[hx & 0xf];
-		hx = hexchars[(bcberr & 0x00000f00) >> 8];
-		*ptr++ = hexchars[hx >> 4];	*ptr++ = hexchars[hx & 0xf];
-		hx = hexchars[(bcberr & 0x000000f0) >> 4];
-		*ptr++ = hexchars[hx >> 4];	*ptr++ = hexchars[hx & 0xf];
-		hx = hexchars[(bcberr & 0x0000000f)];
-		*ptr++ = hexchars[hx >> 4];	*ptr++ = hexchars[hx & 0xf];
+		hx = hex_asc_hi(bcberr >> 24);
+		ptr = pack_hex_byte(ptr, hx);
+		hx = hex_asc_lo(bcberr >> 24);
+		ptr = pack_hex_byte(ptr, hx);
+		hx = hex_asc_hi(bcberr >> 16);
+		ptr = pack_hex_byte(ptr, hx);
+		hx = hex_asc_lo(bcberr >> 16);
+		ptr = pack_hex_byte(ptr, hx);
+		hx = hex_asc_hi(bcberr >> 8);
+		ptr = pack_hex_byte(ptr, hx);
+		hx = hex_asc_lo(bcberr >> 8);
+		ptr = pack_hex_byte(ptr, hx);
+		hx = hex_asc_hi(bcberr);
+		ptr = pack_hex_byte(ptr, hx);
+		hx = hex_asc_lo(bcberr);
+		ptr = pack_hex_byte(ptr, hx);
 
 		ptr = mem2hex(crlf, ptr, sizeof(crlf) - 1, 0);
 		*ptr = 0;
@@ -1353,14 +1341,12 @@ static int gdbstub(struct pt_regs *regs, enum exception_code excep)
 	 * Send trap type (converted to signal)
 	 */
 	*ptr++ = 'T';
-	*ptr++ = hexchars[sigval >> 4];
-	*ptr++ = hexchars[sigval & 0xf];
+	ptr = pack_hex_byte(ptr, sigval);
 
 	/*
 	 * Send Error PC
 	 */
-	*ptr++ = hexchars[GDB_REGID_PC >> 4];
-	*ptr++ = hexchars[GDB_REGID_PC & 0xf];
+	ptr = pack_hex_byte(ptr, GDB_REGID_PC);
 	*ptr++ = ':';
 	ptr = mem2hex(&regs->pc, ptr, 4, 0);
 	*ptr++ = ';';
@@ -1368,8 +1354,7 @@ static int gdbstub(struct pt_regs *regs, enum exception_code excep)
 	/*
 	 * Send frame pointer
 	 */
-	*ptr++ = hexchars[GDB_REGID_FP >> 4];
-	*ptr++ = hexchars[GDB_REGID_FP & 0xf];
+	ptr = pack_hex_byte(ptr, GDB_REGID_FP);
 	*ptr++ = ':';
 	ptr = mem2hex(&regs->a3, ptr, 4, 0);
 	*ptr++ = ';';
@@ -1378,8 +1363,7 @@ static int gdbstub(struct pt_regs *regs, enum exception_code excep)
 	 * Send stack pointer
 	 */
 	ssp = (unsigned long) (regs + 1);
-	*ptr++ = hexchars[GDB_REGID_SP >> 4];
-	*ptr++ = hexchars[GDB_REGID_SP & 0xf];
+	ptr = pack_hex_byte(ptr, GDB_REGID_SP);
 	*ptr++ = ':';
 	ptr = mem2hex(&ssp, ptr, 4, 0);
 	*ptr++ = ';';
@@ -1399,8 +1383,8 @@ packet_waiting:
 			/* request repeat of last signal number */
 		case '?':
 			output_buffer[0] = 'S';
-			output_buffer[1] = hexchars[sigval >> 4];
-			output_buffer[2] = hexchars[sigval & 0xf];
+			output_buffer[1] = hex_asc_hi(sigval);
+			output_buffer[2] = hex_asc_lo(sigval);
 			output_buffer[3] = 0;
 			break;
 
@@ -1838,8 +1822,8 @@ void gdbstub_exit(int status)
 
 	gdbstub_busy = 1;
 	output_buffer[0] = 'W';
-	output_buffer[1] = hexchars[(status >> 4) & 0x0F];
-	output_buffer[2] = hexchars[status & 0x0F];
+	output_buffer[1] = hex_asc_hi(status);
+	output_buffer[2] = hex_asc_lo(status);
 	output_buffer[3] = 0;
 
 	gdbstub_io_tx_char('$');
@@ -1853,8 +1837,8 @@ void gdbstub_exit(int status)
 	}
 
 	gdbstub_io_tx_char('#');
-	gdbstub_io_tx_char(hexchars[checksum >> 4]);
-	gdbstub_io_tx_char(hexchars[checksum & 0xf]);
+	gdbstub_io_tx_char(hex_asc_hi(checksum));
+	gdbstub_io_tx_char(hex_asc_lo(checksum));
 
 	/* make sure the output is flushed, or else RedBoot might clobber it */
 	gdbstub_io_tx_flush();

From a0a8f5364a5ad248aec6cb705e0092ff563edc2f Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Fri, 25 Jul 2008 19:45:20 -0700
Subject: [PATCH 1025/1171] x86: implement pte_special

Implement the pte_special bit for x86.  This is required to support
lockless get_user_pages, because we need to know whether or not we can
refcount a particular page given only its pte (and no vma).

[hugh@veritas.com: fix a BUG]
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Dave Kleikamp <shaggy@austin.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Dave Kleikamp <shaggy@austin.ibm.com>
Cc: Badari Pulavarty <pbadari@us.ibm.com>
Cc: Zach Brown <zach.brown@oracle.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/asm-x86/pgtable.h | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/include/asm-x86/pgtable.h b/include/asm-x86/pgtable.h
index 3e5dbc4195f4..04caa2f544df 100644
--- a/include/asm-x86/pgtable.h
+++ b/include/asm-x86/pgtable.h
@@ -18,6 +18,7 @@
 #define _PAGE_BIT_UNUSED2	10
 #define _PAGE_BIT_UNUSED3	11
 #define _PAGE_BIT_PAT_LARGE	12	/* On 2MB or 1GB pages */
+#define _PAGE_BIT_SPECIAL	_PAGE_BIT_UNUSED1
 #define _PAGE_BIT_NX           63       /* No execute: only valid after cpuid check */
 
 #define _PAGE_PRESENT	(_AT(pteval_t, 1) << _PAGE_BIT_PRESENT)
@@ -34,6 +35,8 @@
 #define _PAGE_UNUSED3	(_AT(pteval_t, 1) << _PAGE_BIT_UNUSED3)
 #define _PAGE_PAT	(_AT(pteval_t, 1) << _PAGE_BIT_PAT)
 #define _PAGE_PAT_LARGE (_AT(pteval_t, 1) << _PAGE_BIT_PAT_LARGE)
+#define _PAGE_SPECIAL	(_AT(pteval_t, 1) << _PAGE_BIT_SPECIAL)
+#define __HAVE_ARCH_PTE_SPECIAL
 
 #if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE)
 #define _PAGE_NX	(_AT(pteval_t, 1) << _PAGE_BIT_NX)
@@ -54,7 +57,7 @@
 
 /* Set of bits not changed in pte_modify */
 #define _PAGE_CHG_MASK	(PTE_PFN_MASK | _PAGE_PCD | _PAGE_PWT |		\
-			 _PAGE_ACCESSED | _PAGE_DIRTY)
+			 _PAGE_SPECIAL | _PAGE_ACCESSED | _PAGE_DIRTY)
 
 #define _PAGE_CACHE_MASK	(_PAGE_PCD | _PAGE_PWT)
 #define _PAGE_CACHE_WB		(0)
@@ -180,7 +183,7 @@ static inline int pte_exec(pte_t pte)
 
 static inline int pte_special(pte_t pte)
 {
-	return 0;
+	return pte_val(pte) & _PAGE_SPECIAL;
 }
 
 static inline int pmd_large(pmd_t pte)
@@ -246,7 +249,7 @@ static inline pte_t pte_clrglobal(pte_t pte)
 
 static inline pte_t pte_mkspecial(pte_t pte)
 {
-	return pte;
+	return __pte(pte_val(pte) | _PAGE_SPECIAL);
 }
 
 extern pteval_t __supported_pte_mask;

From 21cc199baa815d7b3f1ace4be20b9558cbddc00f Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Fri, 25 Jul 2008 19:45:22 -0700
Subject: [PATCH 1026/1171] mm: introduce get_user_pages_fast

Introduce a new get_user_pages_fast mm API, which is basically a
get_user_pages with a less general API (but still tends to be suited to
the common case):

- task and mm are always current and current->mm
- force is always 0
- pages is always non-NULL
- don't pass back vmas

This restricted API can be implemented in a much more scalable way on many
architectures when the ptes are present, by walking the page tables
locklessly (no mmap_sem or page table locks).  When the ptes are not
populated, get_user_pages_fast() could be slower.

This is implemented locklessly on x86, and used in some key direct IO call
sites, in later patches, which provides nearly 10% performance improvement
on a threaded database workload.

Lots of other code could use this too, depending on use cases (eg.  grep
drivers/).  And it might inspire some new and clever ways to use it.

[akpm@linux-foundation.org: build fix]
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Dave Kleikamp <shaggy@austin.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Dave Kleikamp <shaggy@austin.ibm.com>
Cc: Badari Pulavarty <pbadari@us.ibm.com>
Cc: Zach Brown <zach.brown@oracle.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/mm.h | 33 +++++++++++++++++++++++++++++++++
 1 file changed, 33 insertions(+)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index d87a5a5fe87d..f3fd70d6029f 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -833,6 +833,39 @@ extern int mprotect_fixup(struct vm_area_struct *vma,
 			  struct vm_area_struct **pprev, unsigned long start,
 			  unsigned long end, unsigned long newflags);
 
+#ifdef CONFIG_HAVE_GET_USER_PAGES_FAST
+/*
+ * get_user_pages_fast provides equivalent functionality to get_user_pages,
+ * operating on current and current->mm (force=0 and doesn't return any vmas).
+ *
+ * get_user_pages_fast may take mmap_sem and page tables, so no assumptions
+ * can be made about locking. get_user_pages_fast is to be implemented in a
+ * way that is advantageous (vs get_user_pages()) when the user memory area is
+ * already faulted in and present in ptes. However if the pages have to be
+ * faulted in, it may turn out to be slightly slower).
+ */
+int get_user_pages_fast(unsigned long start, int nr_pages, int write,
+			struct page **pages);
+
+#else
+/*
+ * Should probably be moved to asm-generic, and architectures can include it if
+ * they don't implement their own get_user_pages_fast.
+ */
+#define get_user_pages_fast(start, nr_pages, write, pages)	\
+({								\
+	struct mm_struct *mm = current->mm;			\
+	int ret;						\
+								\
+	down_read(&mm->mmap_sem);				\
+	ret = get_user_pages(current, mm, start, nr_pages,	\
+					write, 0, pages, NULL);	\
+	up_read(&mm->mmap_sem);					\
+								\
+	ret;							\
+})
+#endif
+
 /*
  * A callback you can register to apply pressure to ageable caches.
  *

From 8174c430e445a93016ef18f717fe570214fa38bf Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Fri, 25 Jul 2008 19:45:24 -0700
Subject: [PATCH 1027/1171] x86: lockless get_user_pages_fast()

Implement get_user_pages_fast without locking in the fastpath on x86.

Do an optimistic lockless pagetable walk, without taking mmap_sem or any
page table locks or even mmap_sem.  Page table existence is guaranteed by
turning interrupts off (combined with the fact that we're always looking
up the current mm, means we can do the lockless page table walk within the
constraints of the TLB shootdown design).  Basically we can do this
lockless pagetable walk in a similar manner to the way the CPU's pagetable
walker does not have to take any locks to find present ptes.

This patch (combined with the subsequent ones to convert direct IO to use
it) was found to give about 10% performance improvement on a 2 socket 8
core Intel Xeon system running an OLTP workload on DB2 v9.5

 "To test the effects of the patch, an OLTP workload was run on an IBM
  x3850 M2 server with 2 processors (quad-core Intel Xeon processors at
  2.93 GHz) using IBM DB2 v9.5 running Linux 2.6.24rc7 kernel.  Comparing
  runs with and without the patch resulted in an overall performance
  benefit of ~9.8%.  Correspondingly, oprofiles showed that samples from
  __up_read and __down_read routines that is seen during thread contention
  for system resources was reduced from 2.8% down to .05%.  Monitoring the
  /proc/vmstat output from the patched run showed that the counter for
  fast_gup contained a very high number while the fast_gup_slow value was
  zero."

(fast_gup is the old name for get_user_pages_fast, fast_gup_slow is a
counter we had for the number of times the slowpath was invoked).

The main reason for the improvement is that DB2 has multiple threads each
issuing direct-IO.  Direct-IO uses get_user_pages, and thus the threads
contend the mmap_sem cacheline, and can also contend on page table locks.

I would anticipate larger performance gains on larger systems, however I
think DB2 uses an adaptive mix of threads and processes, so it could be
that thread contention remains pretty constant as machine size increases.
In which case, we stuck with "only" a 10% gain.

The downside of using get_user_pages_fast is that if there is not a pte
with the correct permissions for the access, we end up falling back to
get_user_pages and so the get_user_pages_fast is a bit of extra work.
However this should not be the common case in most performance critical
code.

[akpm@linux-foundation.org: coding-style fixes]
[akpm@linux-foundation.org: build fix]
[akpm@linux-foundation.org: Kconfig fix]
[akpm@linux-foundation.org: Makefile fix/cleanup]
[akpm@linux-foundation.org: warning fix]
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Dave Kleikamp <shaggy@austin.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Dave Kleikamp <shaggy@austin.ibm.com>
Cc: Badari Pulavarty <pbadari@us.ibm.com>
Cc: Zach Brown <zach.brown@oracle.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/Kconfig          |   1 +
 arch/x86/mm/Makefile      |   1 +
 arch/x86/mm/gup.c         | 258 ++++++++++++++++++++++++++++++++++++++
 include/asm-x86/uaccess.h |   1 +
 mm/Kconfig                |   3 +
 5 files changed, 264 insertions(+)
 create mode 100644 arch/x86/mm/gup.c

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 6b2debfabddc..6bdde845818e 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -22,6 +22,7 @@ config X86
 	select HAVE_IDE
 	select HAVE_OPROFILE
 	select HAVE_IOREMAP_PROT
+	select HAVE_GET_USER_PAGES_FAST
 	select HAVE_KPROBES
 	select ARCH_WANT_OPTIONAL_GPIOLIB if !X86_RDC321X
 	select HAVE_KRETPROBES
diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile
index 1fbb844c3d7a..2977ea37791f 100644
--- a/arch/x86/mm/Makefile
+++ b/arch/x86/mm/Makefile
@@ -1,6 +1,7 @@
 obj-y	:=  init_$(BITS).o fault.o ioremap.o extable.o pageattr.o mmap.o \
 	    pat.o pgtable.o
 
+obj-$(CONFIG_HAVE_GET_USER_PAGES_FAST) += gup.o
 obj-$(CONFIG_X86_32)		+= pgtable_32.o
 
 obj-$(CONFIG_HUGETLB_PAGE)	+= hugetlbpage.o
diff --git a/arch/x86/mm/gup.c b/arch/x86/mm/gup.c
new file mode 100644
index 000000000000..6f733121f32e
--- /dev/null
+++ b/arch/x86/mm/gup.c
@@ -0,0 +1,258 @@
+/*
+ * Lockless get_user_pages_fast for x86
+ *
+ * Copyright (C) 2008 Nick Piggin
+ * Copyright (C) 2008 Novell Inc.
+ */
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/vmstat.h>
+#include <linux/highmem.h>
+
+#include <asm/pgtable.h>
+
+static inline pte_t gup_get_pte(pte_t *ptep)
+{
+#ifndef CONFIG_X86_PAE
+	return *ptep;
+#else
+	/*
+	 * With get_user_pages_fast, we walk down the pagetables without taking
+	 * any locks.  For this we would like to load the pointers atoimcally,
+	 * but that is not possible (without expensive cmpxchg8b) on PAE.  What
+	 * we do have is the guarantee that a pte will only either go from not
+	 * present to present, or present to not present or both -- it will not
+	 * switch to a completely different present page without a TLB flush in
+	 * between; something that we are blocking by holding interrupts off.
+	 *
+	 * Setting ptes from not present to present goes:
+	 * ptep->pte_high = h;
+	 * smp_wmb();
+	 * ptep->pte_low = l;
+	 *
+	 * And present to not present goes:
+	 * ptep->pte_low = 0;
+	 * smp_wmb();
+	 * ptep->pte_high = 0;
+	 *
+	 * We must ensure here that the load of pte_low sees l iff pte_high
+	 * sees h. We load pte_high *after* loading pte_low, which ensures we
+	 * don't see an older value of pte_high.  *Then* we recheck pte_low,
+	 * which ensures that we haven't picked up a changed pte high. We might
+	 * have got rubbish values from pte_low and pte_high, but we are
+	 * guaranteed that pte_low will not have the present bit set *unless*
+	 * it is 'l'. And get_user_pages_fast only operates on present ptes, so
+	 * we're safe.
+	 *
+	 * gup_get_pte should not be used or copied outside gup.c without being
+	 * very careful -- it does not atomically load the pte or anything that
+	 * is likely to be useful for you.
+	 */
+	pte_t pte;
+
+retry:
+	pte.pte_low = ptep->pte_low;
+	smp_rmb();
+	pte.pte_high = ptep->pte_high;
+	smp_rmb();
+	if (unlikely(pte.pte_low != ptep->pte_low))
+		goto retry;
+
+	return pte;
+#endif
+}
+
+/*
+ * The performance critical leaf functions are made noinline otherwise gcc
+ * inlines everything into a single function which results in too much
+ * register pressure.
+ */
+static noinline int gup_pte_range(pmd_t pmd, unsigned long addr,
+		unsigned long end, int write, struct page **pages, int *nr)
+{
+	unsigned long mask;
+	pte_t *ptep;
+
+	mask = _PAGE_PRESENT|_PAGE_USER;
+	if (write)
+		mask |= _PAGE_RW;
+
+	ptep = pte_offset_map(&pmd, addr);
+	do {
+		pte_t pte = gup_get_pte(ptep);
+		struct page *page;
+
+		if ((pte_val(pte) & (mask | _PAGE_SPECIAL)) != mask) {
+			pte_unmap(ptep);
+			return 0;
+		}
+		VM_BUG_ON(!pfn_valid(pte_pfn(pte)));
+		page = pte_page(pte);
+		get_page(page);
+		pages[*nr] = page;
+		(*nr)++;
+
+	} while (ptep++, addr += PAGE_SIZE, addr != end);
+	pte_unmap(ptep - 1);
+
+	return 1;
+}
+
+static inline void get_head_page_multiple(struct page *page, int nr)
+{
+	VM_BUG_ON(page != compound_head(page));
+	VM_BUG_ON(page_count(page) == 0);
+	atomic_add(nr, &page->_count);
+}
+
+static noinline int gup_huge_pmd(pmd_t pmd, unsigned long addr,
+		unsigned long end, int write, struct page **pages, int *nr)
+{
+	unsigned long mask;
+	pte_t pte = *(pte_t *)&pmd;
+	struct page *head, *page;
+	int refs;
+
+	mask = _PAGE_PRESENT|_PAGE_USER;
+	if (write)
+		mask |= _PAGE_RW;
+	if ((pte_val(pte) & mask) != mask)
+		return 0;
+	/* hugepages are never "special" */
+	VM_BUG_ON(pte_val(pte) & _PAGE_SPECIAL);
+	VM_BUG_ON(!pfn_valid(pte_pfn(pte)));
+
+	refs = 0;
+	head = pte_page(pte);
+	page = head + ((addr & ~HPAGE_MASK) >> PAGE_SHIFT);
+	do {
+		VM_BUG_ON(compound_head(page) != head);
+		pages[*nr] = page;
+		(*nr)++;
+		page++;
+		refs++;
+	} while (addr += PAGE_SIZE, addr != end);
+	get_head_page_multiple(head, refs);
+
+	return 1;
+}
+
+static int gup_pmd_range(pud_t pud, unsigned long addr, unsigned long end,
+		int write, struct page **pages, int *nr)
+{
+	unsigned long next;
+	pmd_t *pmdp;
+
+	pmdp = pmd_offset(&pud, addr);
+	do {
+		pmd_t pmd = *pmdp;
+
+		next = pmd_addr_end(addr, end);
+		if (pmd_none(pmd))
+			return 0;
+		if (unlikely(pmd_large(pmd))) {
+			if (!gup_huge_pmd(pmd, addr, next, write, pages, nr))
+				return 0;
+		} else {
+			if (!gup_pte_range(pmd, addr, next, write, pages, nr))
+				return 0;
+		}
+	} while (pmdp++, addr = next, addr != end);
+
+	return 1;
+}
+
+static int gup_pud_range(pgd_t pgd, unsigned long addr, unsigned long end,
+			int write, struct page **pages, int *nr)
+{
+	unsigned long next;
+	pud_t *pudp;
+
+	pudp = pud_offset(&pgd, addr);
+	do {
+		pud_t pud = *pudp;
+
+		next = pud_addr_end(addr, end);
+		if (pud_none(pud))
+			return 0;
+		if (!gup_pmd_range(pud, addr, next, write, pages, nr))
+			return 0;
+	} while (pudp++, addr = next, addr != end);
+
+	return 1;
+}
+
+int get_user_pages_fast(unsigned long start, int nr_pages, int write,
+			struct page **pages)
+{
+	struct mm_struct *mm = current->mm;
+	unsigned long end = start + (nr_pages << PAGE_SHIFT);
+	unsigned long addr = start;
+	unsigned long next;
+	pgd_t *pgdp;
+	int nr = 0;
+
+	if (unlikely(!access_ok(write ? VERIFY_WRITE : VERIFY_READ,
+					start, nr_pages*PAGE_SIZE)))
+		goto slow_irqon;
+
+	/*
+	 * XXX: batch / limit 'nr', to avoid large irq off latency
+	 * needs some instrumenting to determine the common sizes used by
+	 * important workloads (eg. DB2), and whether limiting the batch size
+	 * will decrease performance.
+	 *
+	 * It seems like we're in the clear for the moment. Direct-IO is
+	 * the main guy that batches up lots of get_user_pages, and even
+	 * they are limited to 64-at-a-time which is not so many.
+	 */
+	/*
+	 * This doesn't prevent pagetable teardown, but does prevent
+	 * the pagetables and pages from being freed on x86.
+	 *
+	 * So long as we atomically load page table pointers versus teardown
+	 * (which we do on x86, with the above PAE exception), we can follow the
+	 * address down to the the page and take a ref on it.
+	 */
+	local_irq_disable();
+	pgdp = pgd_offset(mm, addr);
+	do {
+		pgd_t pgd = *pgdp;
+
+		next = pgd_addr_end(addr, end);
+		if (pgd_none(pgd))
+			goto slow;
+		if (!gup_pud_range(pgd, addr, next, write, pages, &nr))
+			goto slow;
+	} while (pgdp++, addr = next, addr != end);
+	local_irq_enable();
+
+	VM_BUG_ON(nr != (end - start) >> PAGE_SHIFT);
+	return nr;
+
+	{
+		int ret;
+
+slow:
+		local_irq_enable();
+slow_irqon:
+		/* Try to get the remaining pages with get_user_pages */
+		start += nr << PAGE_SHIFT;
+		pages += nr;
+
+		down_read(&mm->mmap_sem);
+		ret = get_user_pages(current, mm, start,
+			(end - start) >> PAGE_SHIFT, write, 0, pages, NULL);
+		up_read(&mm->mmap_sem);
+
+		/* Have to be a bit careful with return values */
+		if (nr > 0) {
+			if (ret < 0)
+				ret = nr;
+			else
+				ret += nr;
+		}
+
+		return ret;
+	}
+}
diff --git a/include/asm-x86/uaccess.h b/include/asm-x86/uaccess.h
index f6fa4d841bbc..5f702d1d5218 100644
--- a/include/asm-x86/uaccess.h
+++ b/include/asm-x86/uaccess.h
@@ -451,3 +451,4 @@ extern struct movsl_mask {
 #endif
 
 #endif
+
diff --git a/mm/Kconfig b/mm/Kconfig
index aa799007a11b..efee5d379df4 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -77,6 +77,9 @@ config FLAT_NODE_MEM_MAP
 	def_bool y
 	depends on !SPARSEMEM
 
+config HAVE_GET_USER_PAGES_FAST
+	bool
+
 #
 # Both the NUMA code and DISCONTIGMEM use arrays of pg_data_t's
 # to represent different areas of memory.  This variable allows

From f5dd33c494a427b1d1a3b574de5c9e511c888864 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Fri, 25 Jul 2008 19:45:25 -0700
Subject: [PATCH 1028/1171] dio: use get_user_pages_fast

Use get_user_pages_fast in the common/generic block and fs direct IO paths.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Dave Kleikamp <shaggy@austin.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Dave Kleikamp <shaggy@austin.ibm.com>
Cc: Badari Pulavarty <pbadari@us.ibm.com>
Cc: Zach Brown <zach.brown@oracle.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/bio.c       |  8 ++------
 fs/direct-io.c | 10 ++--------
 2 files changed, 4 insertions(+), 14 deletions(-)

diff --git a/fs/bio.c b/fs/bio.c
index 88322b066acb..25f1af0d81e5 100644
--- a/fs/bio.c
+++ b/fs/bio.c
@@ -721,12 +721,8 @@ static struct bio *__bio_map_user_iov(struct request_queue *q,
 		const int local_nr_pages = end - start;
 		const int page_limit = cur_page + local_nr_pages;
 		
-		down_read(&current->mm->mmap_sem);
-		ret = get_user_pages(current, current->mm, uaddr,
-				     local_nr_pages,
-				     write_to_vm, 0, &pages[cur_page], NULL);
-		up_read(&current->mm->mmap_sem);
-
+		ret = get_user_pages_fast(uaddr, local_nr_pages,
+				write_to_vm, &pages[cur_page]);
 		if (ret < local_nr_pages) {
 			ret = -EFAULT;
 			goto out_unmap;
diff --git a/fs/direct-io.c b/fs/direct-io.c
index 9e81addbd6ea..9606ee848fd8 100644
--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -150,17 +150,11 @@ static int dio_refill_pages(struct dio *dio)
 	int nr_pages;
 
 	nr_pages = min(dio->total_pages - dio->curr_page, DIO_PAGES);
-	down_read(&current->mm->mmap_sem);
-	ret = get_user_pages(
-		current,			/* Task for fault acounting */
-		current->mm,			/* whose pages? */
+	ret = get_user_pages_fast(
 		dio->curr_user_address,		/* Where from? */
 		nr_pages,			/* How many pages? */
 		dio->rw == READ,		/* Write to memory? */
-		0,				/* force (?) */
-		&dio->pages[0],
-		NULL);				/* vmas */
-	up_read(&current->mm->mmap_sem);
+		&dio->pages[0]);		/* Put results here */
 
 	if (ret < 0 && dio->blocks_available && (dio->rw & WRITE)) {
 		struct page *page = ZERO_PAGE(0);

From bc40d73c950146725e9e768e856a416ec8949065 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Fri, 25 Jul 2008 19:45:26 -0700
Subject: [PATCH 1029/1171] splice: use get_user_pages_fast

Use get_user_pages_fast in splice.  This reverts some mmap_sem batching
there, however the biggest problem with mmap_sem tends to be hold times
blocking out other threads rather than cacheline bouncing.  Further: on
architectures that implement get_user_pages_fast without locks, mmap_sem
can be avoided completely anyway.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Dave Kleikamp <shaggy@austin.ibm.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Dave Kleikamp <shaggy@austin.ibm.com>
Cc: Badari Pulavarty <pbadari@us.ibm.com>
Cc: Zach Brown <zach.brown@oracle.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/splice.c | 41 +++--------------------------------------
 1 file changed, 3 insertions(+), 38 deletions(-)

diff --git a/fs/splice.c b/fs/splice.c
index 399442179d89..47dc1a445d1f 100644
--- a/fs/splice.c
+++ b/fs/splice.c
@@ -1160,36 +1160,6 @@ static long do_splice(struct file *in, loff_t __user *off_in,
 	return -EINVAL;
 }
 
-/*
- * Do a copy-from-user while holding the mmap_semaphore for reading, in a
- * manner safe from deadlocking with simultaneous mmap() (grabbing mmap_sem
- * for writing) and page faulting on the user memory pointed to by src.
- * This assumes that we will very rarely hit the partial != 0 path, or this
- * will not be a win.
- */
-static int copy_from_user_mmap_sem(void *dst, const void __user *src, size_t n)
-{
-	int partial;
-
-	if (!access_ok(VERIFY_READ, src, n))
-		return -EFAULT;
-
-	pagefault_disable();
-	partial = __copy_from_user_inatomic(dst, src, n);
-	pagefault_enable();
-
-	/*
-	 * Didn't copy everything, drop the mmap_sem and do a faulting copy
-	 */
-	if (unlikely(partial)) {
-		up_read(&current->mm->mmap_sem);
-		partial = copy_from_user(dst, src, n);
-		down_read(&current->mm->mmap_sem);
-	}
-
-	return partial;
-}
-
 /*
  * Map an iov into an array of pages and offset/length tupples. With the
  * partial_page structure, we can map several non-contiguous ranges into
@@ -1203,8 +1173,6 @@ static int get_iovec_page_array(const struct iovec __user *iov,
 {
 	int buffers = 0, error = 0;
 
-	down_read(&current->mm->mmap_sem);
-
 	while (nr_vecs) {
 		unsigned long off, npages;
 		struct iovec entry;
@@ -1213,7 +1181,7 @@ static int get_iovec_page_array(const struct iovec __user *iov,
 		int i;
 
 		error = -EFAULT;
-		if (copy_from_user_mmap_sem(&entry, iov, sizeof(entry)))
+		if (copy_from_user(&entry, iov, sizeof(entry)))
 			break;
 
 		base = entry.iov_base;
@@ -1247,9 +1215,8 @@ static int get_iovec_page_array(const struct iovec __user *iov,
 		if (npages > PIPE_BUFFERS - buffers)
 			npages = PIPE_BUFFERS - buffers;
 
-		error = get_user_pages(current, current->mm,
-				       (unsigned long) base, npages, 0, 0,
-				       &pages[buffers], NULL);
+		error = get_user_pages_fast((unsigned long)base, npages,
+					0, &pages[buffers]);
 
 		if (unlikely(error <= 0))
 			break;
@@ -1288,8 +1255,6 @@ static int get_iovec_page_array(const struct iovec __user *iov,
 		iov++;
 	}
 
-	up_read(&current->mm->mmap_sem);
-
 	if (buffers)
 		return buffers;
 

From 652ea695364142b2464744746beac206d050ef19 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Fri, 25 Jul 2008 19:45:27 -0700
Subject: [PATCH 1030/1171] x86: support 1GB hugepages with
 get_user_pages_lockless()

Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Andi Kleen <andi@firstfloor.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/mm/gup.c | 43 ++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 40 insertions(+), 3 deletions(-)

diff --git a/arch/x86/mm/gup.c b/arch/x86/mm/gup.c
index 6f733121f32e..3085f25b4355 100644
--- a/arch/x86/mm/gup.c
+++ b/arch/x86/mm/gup.c
@@ -124,7 +124,7 @@ static noinline int gup_huge_pmd(pmd_t pmd, unsigned long addr,
 
 	refs = 0;
 	head = pte_page(pte);
-	page = head + ((addr & ~HPAGE_MASK) >> PAGE_SHIFT);
+	page = head + ((addr & ~PMD_MASK) >> PAGE_SHIFT);
 	do {
 		VM_BUG_ON(compound_head(page) != head);
 		pages[*nr] = page;
@@ -162,6 +162,38 @@ static int gup_pmd_range(pud_t pud, unsigned long addr, unsigned long end,
 	return 1;
 }
 
+static noinline int gup_huge_pud(pud_t pud, unsigned long addr,
+		unsigned long end, int write, struct page **pages, int *nr)
+{
+	unsigned long mask;
+	pte_t pte = *(pte_t *)&pud;
+	struct page *head, *page;
+	int refs;
+
+	mask = _PAGE_PRESENT|_PAGE_USER;
+	if (write)
+		mask |= _PAGE_RW;
+	if ((pte_val(pte) & mask) != mask)
+		return 0;
+	/* hugepages are never "special" */
+	VM_BUG_ON(pte_val(pte) & _PAGE_SPECIAL);
+	VM_BUG_ON(!pfn_valid(pte_pfn(pte)));
+
+	refs = 0;
+	head = pte_page(pte);
+	page = head + ((addr & ~PUD_MASK) >> PAGE_SHIFT);
+	do {
+		VM_BUG_ON(compound_head(page) != head);
+		pages[*nr] = page;
+		(*nr)++;
+		page++;
+		refs++;
+	} while (addr += PAGE_SIZE, addr != end);
+	get_head_page_multiple(head, refs);
+
+	return 1;
+}
+
 static int gup_pud_range(pgd_t pgd, unsigned long addr, unsigned long end,
 			int write, struct page **pages, int *nr)
 {
@@ -175,8 +207,13 @@ static int gup_pud_range(pgd_t pgd, unsigned long addr, unsigned long end,
 		next = pud_addr_end(addr, end);
 		if (pud_none(pud))
 			return 0;
-		if (!gup_pmd_range(pud, addr, next, write, pages, nr))
-			return 0;
+		if (unlikely(pud_large(pud))) {
+			if (!gup_huge_pud(pud, addr, next, write, pages, nr))
+				return 0;
+		} else {
+			if (!gup_pmd_range(pud, addr, next, write, pages, nr))
+				return 0;
+		}
 	} while (pudp++, addr = next, addr != end);
 
 	return 1;

From 30002ed2e41830ec03ec3e577ad83ac6b188f96e Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Fri, 25 Jul 2008 19:45:28 -0700
Subject: [PATCH 1031/1171] mm: readahead scan lockless

radix_tree_next_hole() is implemented as a series of radix_tree_lookup()s.
So it can be called locklessly, under rcu_read_lock().

Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/readahead.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/mm/readahead.c b/mm/readahead.c
index d8723a5f6496..77e8ddf945e9 100644
--- a/mm/readahead.c
+++ b/mm/readahead.c
@@ -382,9 +382,9 @@ ondemand_readahead(struct address_space *mapping,
 	if (hit_readahead_marker) {
 		pgoff_t start;
 
-		read_lock_irq(&mapping->tree_lock);
-		start = radix_tree_next_hole(&mapping->page_tree, offset, max+1);
-		read_unlock_irq(&mapping->tree_lock);
+		rcu_read_lock();
+		start = radix_tree_next_hole(&mapping->page_tree, offset,max+1);
+		rcu_read_unlock();
 
 		if (!start || start - offset > max)
 			return 0;

From 47feff2c8eefe85099f87c43d3096855f0085ca0 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Fri, 25 Jul 2008 19:45:29 -0700
Subject: [PATCH 1032/1171] radix-tree: add gang_lookup_slot,
 gang_lookup_slot_tag

Introduce gang_lookup_slot() and gang_lookup_slot_tag() functions, which
are used by lockless pagecache.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/radix-tree.h |  12 ++-
 lib/radix-tree.c           | 178 ++++++++++++++++++++++++++++++++-----
 2 files changed, 166 insertions(+), 24 deletions(-)

diff --git a/include/linux/radix-tree.h b/include/linux/radix-tree.h
index b8ce2b444bb5..a916c6660dfa 100644
--- a/include/linux/radix-tree.h
+++ b/include/linux/radix-tree.h
@@ -99,12 +99,15 @@ do {									\
  *
  * The notable exceptions to this rule are the following functions:
  * radix_tree_lookup
+ * radix_tree_lookup_slot
  * radix_tree_tag_get
  * radix_tree_gang_lookup
+ * radix_tree_gang_lookup_slot
  * radix_tree_gang_lookup_tag
+ * radix_tree_gang_lookup_tag_slot
  * radix_tree_tagged
  *
- * The first 4 functions are able to be called locklessly, using RCU. The
+ * The first 7 functions are able to be called locklessly, using RCU. The
  * caller must ensure calls to these functions are made within rcu_read_lock()
  * regions. Other readers (lock-free or otherwise) and modifications may be
  * running concurrently.
@@ -159,6 +162,9 @@ void *radix_tree_delete(struct radix_tree_root *, unsigned long);
 unsigned int
 radix_tree_gang_lookup(struct radix_tree_root *root, void **results,
 			unsigned long first_index, unsigned int max_items);
+unsigned int
+radix_tree_gang_lookup_slot(struct radix_tree_root *root, void ***results,
+			unsigned long first_index, unsigned int max_items);
 unsigned long radix_tree_next_hole(struct radix_tree_root *root,
 				unsigned long index, unsigned long max_scan);
 int radix_tree_preload(gfp_t gfp_mask);
@@ -173,6 +179,10 @@ unsigned int
 radix_tree_gang_lookup_tag(struct radix_tree_root *root, void **results,
 		unsigned long first_index, unsigned int max_items,
 		unsigned int tag);
+unsigned int
+radix_tree_gang_lookup_tag_slot(struct radix_tree_root *root, void ***results,
+		unsigned long first_index, unsigned int max_items,
+		unsigned int tag);
 int radix_tree_tagged(struct radix_tree_root *root, unsigned int tag);
 
 static inline void radix_tree_preload_end(void)
diff --git a/lib/radix-tree.c b/lib/radix-tree.c
index 56ec21a7f73d..9c4f1ffa2864 100644
--- a/lib/radix-tree.c
+++ b/lib/radix-tree.c
@@ -359,18 +359,17 @@ EXPORT_SYMBOL(radix_tree_insert);
  *	Returns:  the slot corresponding to the position @index in the
  *	radix tree @root. This is useful for update-if-exists operations.
  *
- *	This function cannot be called under rcu_read_lock, it must be
- *	excluded from writers, as must the returned slot for subsequent
- *	use by radix_tree_deref_slot() and radix_tree_replace slot.
- *	Caller must hold tree write locked across slot lookup and
- *	replace.
+ *	This function can be called under rcu_read_lock iff the slot is not
+ *	modified by radix_tree_replace_slot, otherwise it must be called
+ *	exclusive from other writers. Any dereference of the slot must be done
+ *	using radix_tree_deref_slot.
  */
 void **radix_tree_lookup_slot(struct radix_tree_root *root, unsigned long index)
 {
 	unsigned int height, shift;
 	struct radix_tree_node *node, **slot;
 
-	node = root->rnode;
+	node = rcu_dereference(root->rnode);
 	if (node == NULL)
 		return NULL;
 
@@ -390,7 +389,7 @@ void **radix_tree_lookup_slot(struct radix_tree_root *root, unsigned long index)
 	do {
 		slot = (struct radix_tree_node **)
 			(node->slots + ((index>>shift) & RADIX_TREE_MAP_MASK));
-		node = *slot;
+		node = rcu_dereference(*slot);
 		if (node == NULL)
 			return NULL;
 
@@ -667,7 +666,7 @@ unsigned long radix_tree_next_hole(struct radix_tree_root *root,
 EXPORT_SYMBOL(radix_tree_next_hole);
 
 static unsigned int
-__lookup(struct radix_tree_node *slot, void **results, unsigned long index,
+__lookup(struct radix_tree_node *slot, void ***results, unsigned long index,
 	unsigned int max_items, unsigned long *next_index)
 {
 	unsigned int nr_found = 0;
@@ -701,11 +700,9 @@ __lookup(struct radix_tree_node *slot, void **results, unsigned long index,
 
 	/* Bottom level: grab some items */
 	for (i = index & RADIX_TREE_MAP_MASK; i < RADIX_TREE_MAP_SIZE; i++) {
-		struct radix_tree_node *node;
 		index++;
-		node = slot->slots[i];
-		if (node) {
-			results[nr_found++] = rcu_dereference(node);
+		if (slot->slots[i]) {
+			results[nr_found++] = &(slot->slots[i]);
 			if (nr_found == max_items)
 				goto out;
 		}
@@ -759,13 +756,22 @@ radix_tree_gang_lookup(struct radix_tree_root *root, void **results,
 
 	ret = 0;
 	while (ret < max_items) {
-		unsigned int nr_found;
+		unsigned int nr_found, slots_found, i;
 		unsigned long next_index;	/* Index of next search */
 
 		if (cur_index > max_index)
 			break;
-		nr_found = __lookup(node, results + ret, cur_index,
+		slots_found = __lookup(node, (void ***)results + ret, cur_index,
 					max_items - ret, &next_index);
+		nr_found = 0;
+		for (i = 0; i < slots_found; i++) {
+			struct radix_tree_node *slot;
+			slot = *(((void ***)results)[ret + i]);
+			if (!slot)
+				continue;
+			results[ret + nr_found] = rcu_dereference(slot);
+			nr_found++;
+		}
 		ret += nr_found;
 		if (next_index == 0)
 			break;
@@ -776,12 +782,71 @@ radix_tree_gang_lookup(struct radix_tree_root *root, void **results,
 }
 EXPORT_SYMBOL(radix_tree_gang_lookup);
 
+/**
+ *	radix_tree_gang_lookup_slot - perform multiple slot lookup on radix tree
+ *	@root:		radix tree root
+ *	@results:	where the results of the lookup are placed
+ *	@first_index:	start the lookup from this key
+ *	@max_items:	place up to this many items at *results
+ *
+ *	Performs an index-ascending scan of the tree for present items.  Places
+ *	their slots at *@results and returns the number of items which were
+ *	placed at *@results.
+ *
+ *	The implementation is naive.
+ *
+ *	Like radix_tree_gang_lookup as far as RCU and locking goes. Slots must
+ *	be dereferenced with radix_tree_deref_slot, and if using only RCU
+ *	protection, radix_tree_deref_slot may fail requiring a retry.
+ */
+unsigned int
+radix_tree_gang_lookup_slot(struct radix_tree_root *root, void ***results,
+			unsigned long first_index, unsigned int max_items)
+{
+	unsigned long max_index;
+	struct radix_tree_node *node;
+	unsigned long cur_index = first_index;
+	unsigned int ret;
+
+	node = rcu_dereference(root->rnode);
+	if (!node)
+		return 0;
+
+	if (!radix_tree_is_indirect_ptr(node)) {
+		if (first_index > 0)
+			return 0;
+		results[0] = (void **)&root->rnode;
+		return 1;
+	}
+	node = radix_tree_indirect_to_ptr(node);
+
+	max_index = radix_tree_maxindex(node->height);
+
+	ret = 0;
+	while (ret < max_items) {
+		unsigned int slots_found;
+		unsigned long next_index;	/* Index of next search */
+
+		if (cur_index > max_index)
+			break;
+		slots_found = __lookup(node, results + ret, cur_index,
+					max_items - ret, &next_index);
+		ret += slots_found;
+		if (next_index == 0)
+			break;
+		cur_index = next_index;
+	}
+
+	return ret;
+}
+EXPORT_SYMBOL(radix_tree_gang_lookup_slot);
+
 /*
  * FIXME: the two tag_get()s here should use find_next_bit() instead of
  * open-coding the search.
  */
 static unsigned int
-__lookup_tag(struct radix_tree_node *slot, void **results, unsigned long index,
+__lookup_tag(struct radix_tree_node *slot, void ***results, unsigned long index,
 	unsigned int max_items, unsigned long *next_index, unsigned int tag)
 {
 	unsigned int nr_found = 0;
@@ -811,11 +876,9 @@ __lookup_tag(struct radix_tree_node *slot, void **results, unsigned long index,
 			unsigned long j = index & RADIX_TREE_MAP_MASK;
 
 			for ( ; j < RADIX_TREE_MAP_SIZE; j++) {
-				struct radix_tree_node *node;
 				index++;
 				if (!tag_get(slot, tag, j))
 					continue;
-				node = slot->slots[j];
 				/*
 				 * Even though the tag was found set, we need to
 				 * recheck that we have a non-NULL node, because
@@ -826,9 +889,8 @@ __lookup_tag(struct radix_tree_node *slot, void **results, unsigned long index,
 				 * lookup ->slots[x] without a lock (ie. can't
 				 * rely on its value remaining the same).
 				 */
-				if (node) {
-					node = rcu_dereference(node);
-					results[nr_found++] = node;
+				if (slot->slots[j]) {
+					results[nr_found++] = &(slot->slots[j]);
 					if (nr_found == max_items)
 						goto out;
 				}
@@ -887,13 +949,22 @@ radix_tree_gang_lookup_tag(struct radix_tree_root *root, void **results,
 
 	ret = 0;
 	while (ret < max_items) {
-		unsigned int nr_found;
+		unsigned int nr_found, slots_found, i;
 		unsigned long next_index;	/* Index of next search */
 
 		if (cur_index > max_index)
 			break;
-		nr_found = __lookup_tag(node, results + ret, cur_index,
-					max_items - ret, &next_index, tag);
+		slots_found = __lookup_tag(node, (void ***)results + ret,
+				cur_index, max_items - ret, &next_index, tag);
+		nr_found = 0;
+		for (i = 0; i < slots_found; i++) {
+			struct radix_tree_node *slot;
+			slot = *(((void ***)results)[ret + i]);
+			if (!slot)
+				continue;
+			results[ret + nr_found] = rcu_dereference(slot);
+			nr_found++;
+		}
 		ret += nr_found;
 		if (next_index == 0)
 			break;
@@ -904,6 +975,67 @@ radix_tree_gang_lookup_tag(struct radix_tree_root *root, void **results,
 }
 EXPORT_SYMBOL(radix_tree_gang_lookup_tag);
 
+/**
+ *	radix_tree_gang_lookup_tag_slot - perform multiple slot lookup on a
+ *					  radix tree based on a tag
+ *	@root:		radix tree root
+ *	@results:	where the results of the lookup are placed
+ *	@first_index:	start the lookup from this key
+ *	@max_items:	place up to this many items at *results
+ *	@tag:		the tag index (< RADIX_TREE_MAX_TAGS)
+ *
+ *	Performs an index-ascending scan of the tree for present items which
+ *	have the tag indexed by @tag set.  Places the slots at *@results and
+ *	returns the number of slots which were placed at *@results.
+ */
+unsigned int
+radix_tree_gang_lookup_tag_slot(struct radix_tree_root *root, void ***results,
+		unsigned long first_index, unsigned int max_items,
+		unsigned int tag)
+{
+	struct radix_tree_node *node;
+	unsigned long max_index;
+	unsigned long cur_index = first_index;
+	unsigned int ret;
+
+	/* check the root's tag bit */
+	if (!root_tag_get(root, tag))
+		return 0;
+
+	node = rcu_dereference(root->rnode);
+	if (!node)
+		return 0;
+
+	if (!radix_tree_is_indirect_ptr(node)) {
+		if (first_index > 0)
+			return 0;
+		results[0] = (void **)&root->rnode;
+		return 1;
+	}
+	node = radix_tree_indirect_to_ptr(node);
+
+	max_index = radix_tree_maxindex(node->height);
+
+	ret = 0;
+	while (ret < max_items) {
+		unsigned int slots_found;
+		unsigned long next_index;	/* Index of next search */
+
+		if (cur_index > max_index)
+			break;
+		slots_found = __lookup_tag(node, results + ret,
+				cur_index, max_items - ret, &next_index, tag);
+		ret += slots_found;
+		if (next_index == 0)
+			break;
+		cur_index = next_index;
+	}
+
+	return ret;
+}
+EXPORT_SYMBOL(radix_tree_gang_lookup_tag_slot);
+
+
 /**
  *	radix_tree_shrink    -    shrink height of a radix tree to minimal
  *	@root		radix tree root

From e286781d5f2e9c846e012a39653a166e9d31777d Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Fri, 25 Jul 2008 19:45:30 -0700
Subject: [PATCH 1033/1171] mm: speculative page references

If we can be sure that elevating the page_count on a pagecache page will
pin it, we can speculatively run this operation, and subsequently check to
see if we hit the right page rather than relying on holding a lock or
otherwise pinning a reference to the page.

This can be done if get_page/put_page behaves consistently throughout the
whole tree (ie.  if we "get" the page after it has been used for something
else, we must be able to free it with a put_page).

Actually, there is a period where the count behaves differently: when the
page is free or if it is a constituent page of a compound page.  We need
an atomic_inc_not_zero operation to ensure we don't try to grab the page
in either case.

This patch introduces the core locking protocol to the pagecache (ie.
adds page_cache_get_speculative, and tweaks some update-side code to make
it work).

Thanks to Hugh for pointing out an improvement to the algorithm setting
page_count to zero when we have control of all references, in order to
hold off speculative getters.

[kamezawa.hiroyu@jp.fujitsu.com: fix migration_entry_wait()]
[hugh@veritas.com: fix add_to_page_cache]
[akpm@linux-foundation.org: repair a comment]
Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Jeff Garzik <jeff@garzik.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Acked-by: Nick Piggin <npiggin@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/net/cassini.c   |  12 +++++
 include/linux/pagemap.h | 111 +++++++++++++++++++++++++++++++++++++++-
 mm/filemap.c            |  32 +++++++-----
 mm/migrate.c            |  20 +++++++-
 mm/shmem.c              |   6 +--
 mm/swap_state.c         |  17 ++++--
 mm/vmscan.c             |  74 +++++++++++++++++++--------
 7 files changed, 227 insertions(+), 45 deletions(-)

diff --git a/drivers/net/cassini.c b/drivers/net/cassini.c
index 83768df27806..f1936d51b458 100644
--- a/drivers/net/cassini.c
+++ b/drivers/net/cassini.c
@@ -576,6 +576,18 @@ static void cas_spare_recover(struct cas *cp, const gfp_t flags)
 	list_for_each_safe(elem, tmp, &list) {
 		cas_page_t *page = list_entry(elem, cas_page_t, list);
 
+		/*
+		 * With the lockless pagecache, cassini buffering scheme gets
+		 * slightly less accurate: we might find that a page has an
+		 * elevated reference count here, due to a speculative ref,
+		 * and skip it as in-use. Ideally we would be able to reclaim
+		 * it. However this would be such a rare case, it doesn't
+		 * matter too much as we should pick it up the next time round.
+		 *
+		 * Importantly, if we find that the page has a refcount of 1
+		 * here (our refcount), then we know it is definitely not inuse
+		 * so we can reuse it.
+		 */
 		if (page_count(page->buffer) > 1)
 			continue;
 
diff --git a/include/linux/pagemap.h b/include/linux/pagemap.h
index ee1ec2c7723c..a81d81890422 100644
--- a/include/linux/pagemap.h
+++ b/include/linux/pagemap.h
@@ -12,6 +12,7 @@
 #include <asm/uaccess.h>
 #include <linux/gfp.h>
 #include <linux/bitops.h>
+#include <linux/hardirq.h> /* for in_interrupt() */
 
 /*
  * Bits in mapping->flags.  The lower __GFP_BITS_SHIFT bits are the page
@@ -62,6 +63,98 @@ static inline void mapping_set_gfp_mask(struct address_space *m, gfp_t mask)
 #define page_cache_release(page)	put_page(page)
 void release_pages(struct page **pages, int nr, int cold);
 
+/*
+ * speculatively take a reference to a page.
+ * If the page is free (_count == 0), then _count is untouched, and 0
+ * is returned. Otherwise, _count is incremented by 1 and 1 is returned.
+ *
+ * This function must be called inside the same rcu_read_lock() section as has
+ * been used to lookup the page in the pagecache radix-tree (or page table):
+ * this allows allocators to use a synchronize_rcu() to stabilize _count.
+ *
+ * Unless an RCU grace period has passed, the count of all pages coming out
+ * of the allocator must be considered unstable. page_count may return higher
+ * than expected, and put_page must be able to do the right thing when the
+ * page has been finished with, no matter what it is subsequently allocated
+ * for (because put_page is what is used here to drop an invalid speculative
+ * reference).
+ *
+ * This is the interesting part of the lockless pagecache (and lockless
+ * get_user_pages) locking protocol, where the lookup-side (eg. find_get_page)
+ * has the following pattern:
+ * 1. find page in radix tree
+ * 2. conditionally increment refcount
+ * 3. check the page is still in pagecache (if no, goto 1)
+ *
+ * Remove-side that cares about stability of _count (eg. reclaim) has the
+ * following (with tree_lock held for write):
+ * A. atomically check refcount is correct and set it to 0 (atomic_cmpxchg)
+ * B. remove page from pagecache
+ * C. free the page
+ *
+ * There are 2 critical interleavings that matter:
+ * - 2 runs before A: in this case, A sees elevated refcount and bails out
+ * - A runs before 2: in this case, 2 sees zero refcount and retries;
+ *   subsequently, B will complete and 1 will find no page, causing the
+ *   lookup to return NULL.
+ *
+ * It is possible that between 1 and 2, the page is removed then the exact same
+ * page is inserted into the same position in pagecache. That's OK: the
+ * old find_get_page using tree_lock could equally have run before or after
+ * such a re-insertion, depending on order that locks are granted.
+ *
+ * Lookups racing against pagecache insertion isn't a big problem: either 1
+ * will find the page or it will not. Likewise, the old find_get_page could run
+ * either before the insertion or afterwards, depending on timing.
+ */
+static inline int page_cache_get_speculative(struct page *page)
+{
+	VM_BUG_ON(in_interrupt());
+
+#if !defined(CONFIG_SMP) && defined(CONFIG_CLASSIC_RCU)
+# ifdef CONFIG_PREEMPT
+	VM_BUG_ON(!in_atomic());
+# endif
+	/*
+	 * Preempt must be disabled here - we rely on rcu_read_lock doing
+	 * this for us.
+	 *
+	 * Pagecache won't be truncated from interrupt context, so if we have
+	 * found a page in the radix tree here, we have pinned its refcount by
+	 * disabling preempt, and hence no need for the "speculative get" that
+	 * SMP requires.
+	 */
+	VM_BUG_ON(page_count(page) == 0);
+	atomic_inc(&page->_count);
+
+#else
+	if (unlikely(!get_page_unless_zero(page))) {
+		/*
+		 * Either the page has been freed, or will be freed.
+		 * In either case, retry here and the caller should
+		 * do the right thing (see comments above).
+		 */
+		return 0;
+	}
+#endif
+	VM_BUG_ON(PageTail(page));
+
+	return 1;
+}
+
+static inline int page_freeze_refs(struct page *page, int count)
+{
+	return likely(atomic_cmpxchg(&page->_count, count, 0) == count);
+}
+
+static inline void page_unfreeze_refs(struct page *page, int count)
+{
+	VM_BUG_ON(page_count(page) != 0);
+	VM_BUG_ON(count == 0);
+
+	atomic_set(&page->_count, count);
+}
+
 #ifdef CONFIG_NUMA
 extern struct page *__page_cache_alloc(gfp_t gfp);
 #else
@@ -133,13 +226,29 @@ static inline struct page *read_mapping_page(struct address_space *mapping,
 	return read_cache_page(mapping, index, filler, data);
 }
 
-int add_to_page_cache(struct page *page, struct address_space *mapping,
+int add_to_page_cache_locked(struct page *page, struct address_space *mapping,
 				pgoff_t index, gfp_t gfp_mask);
 int add_to_page_cache_lru(struct page *page, struct address_space *mapping,
 				pgoff_t index, gfp_t gfp_mask);
 extern void remove_from_page_cache(struct page *page);
 extern void __remove_from_page_cache(struct page *page);
 
+/*
+ * Like add_to_page_cache_locked, but used to add newly allocated pages:
+ * the page is new, so we can just run SetPageLocked() against it.
+ */
+static inline int add_to_page_cache(struct page *page,
+		struct address_space *mapping, pgoff_t offset, gfp_t gfp_mask)
+{
+	int error;
+
+	SetPageLocked(page);
+	error = add_to_page_cache_locked(page, mapping, offset, gfp_mask);
+	if (unlikely(error))
+		ClearPageLocked(page);
+	return error;
+}
+
 /*
  * Return byte-offset into filesystem object for page.
  */
diff --git a/mm/filemap.c b/mm/filemap.c
index 2d3ec1ffc66e..4e182a9a14c0 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -442,39 +442,43 @@ int filemap_write_and_wait_range(struct address_space *mapping,
 }
 
 /**
- * add_to_page_cache - add newly allocated pagecache pages
+ * add_to_page_cache_locked - add a locked page to the pagecache
  * @page:	page to add
  * @mapping:	the page's address_space
  * @offset:	page index
  * @gfp_mask:	page allocation mode
  *
- * This function is used to add newly allocated pagecache pages;
- * the page is new, so we can just run SetPageLocked() against it.
- * The other page state flags were set by rmqueue().
- *
+ * This function is used to add a page to the pagecache. It must be locked.
  * This function does not add the page to the LRU.  The caller must do that.
  */
-int add_to_page_cache(struct page *page, struct address_space *mapping,
+int add_to_page_cache_locked(struct page *page, struct address_space *mapping,
 		pgoff_t offset, gfp_t gfp_mask)
 {
-	int error = mem_cgroup_cache_charge(page, current->mm,
+	int error;
+
+	VM_BUG_ON(!PageLocked(page));
+
+	error = mem_cgroup_cache_charge(page, current->mm,
 					gfp_mask & ~__GFP_HIGHMEM);
 	if (error)
 		goto out;
 
 	error = radix_tree_preload(gfp_mask & ~__GFP_HIGHMEM);
 	if (error == 0) {
+		page_cache_get(page);
+		page->mapping = mapping;
+		page->index = offset;
+
 		write_lock_irq(&mapping->tree_lock);
 		error = radix_tree_insert(&mapping->page_tree, offset, page);
-		if (!error) {
-			page_cache_get(page);
-			SetPageLocked(page);
-			page->mapping = mapping;
-			page->index = offset;
+		if (likely(!error)) {
 			mapping->nrpages++;
 			__inc_zone_page_state(page, NR_FILE_PAGES);
-		} else
+		} else {
+			page->mapping = NULL;
 			mem_cgroup_uncharge_cache_page(page);
+			page_cache_release(page);
+		}
 
 		write_unlock_irq(&mapping->tree_lock);
 		radix_tree_preload_end();
@@ -483,7 +487,7 @@ int add_to_page_cache(struct page *page, struct address_space *mapping,
 out:
 	return error;
 }
-EXPORT_SYMBOL(add_to_page_cache);
+EXPORT_SYMBOL(add_to_page_cache_locked);
 
 int add_to_page_cache_lru(struct page *page, struct address_space *mapping,
 				pgoff_t offset, gfp_t gfp_mask)
diff --git a/mm/migrate.c b/mm/migrate.c
index d8c65a65c61d..3ca6392e82cc 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -285,7 +285,15 @@ void migration_entry_wait(struct mm_struct *mm, pmd_t *pmd,
 
 	page = migration_entry_to_page(entry);
 
-	get_page(page);
+	/*
+	 * Once radix-tree replacement of page migration started, page_count
+	 * *must* be zero. And, we don't want to call wait_on_page_locked()
+	 * against a page without get_page().
+	 * So, we use get_page_unless_zero(), here. Even failed, page fault
+	 * will occur again.
+	 */
+	if (!get_page_unless_zero(page))
+		goto out;
 	pte_unmap_unlock(ptep, ptl);
 	wait_on_page_locked(page);
 	put_page(page);
@@ -305,6 +313,7 @@ out:
 static int migrate_page_move_mapping(struct address_space *mapping,
 		struct page *newpage, struct page *page)
 {
+	int expected_count;
 	void **pslot;
 
 	if (!mapping) {
@@ -319,12 +328,18 @@ static int migrate_page_move_mapping(struct address_space *mapping,
 	pslot = radix_tree_lookup_slot(&mapping->page_tree,
  					page_index(page));
 
-	if (page_count(page) != 2 + !!PagePrivate(page) ||
+	expected_count = 2 + !!PagePrivate(page);
+	if (page_count(page) != expected_count ||
 			(struct page *)radix_tree_deref_slot(pslot) != page) {
 		write_unlock_irq(&mapping->tree_lock);
 		return -EAGAIN;
 	}
 
+	if (!page_freeze_refs(page, expected_count)) {
+		write_unlock_irq(&mapping->tree_lock);
+		return -EAGAIN;
+	}
+
 	/*
 	 * Now we know that no one else is looking at the page.
 	 */
@@ -338,6 +353,7 @@ static int migrate_page_move_mapping(struct address_space *mapping,
 
 	radix_tree_replace_slot(pslot, newpage);
 
+	page_unfreeze_refs(page, expected_count);
 	/*
 	 * Drop cache reference from old page.
 	 * We know this isn't the last reference.
diff --git a/mm/shmem.c b/mm/shmem.c
index f92fea94d037..1089092aecaf 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -936,7 +936,7 @@ found:
 	spin_lock(&info->lock);
 	ptr = shmem_swp_entry(info, idx, NULL);
 	if (ptr && ptr->val == entry.val) {
-		error = add_to_page_cache(page, inode->i_mapping,
+		error = add_to_page_cache_locked(page, inode->i_mapping,
 						idx, GFP_NOWAIT);
 		/* does mem_cgroup_uncharge_cache_page on error */
 	} else	/* we must compensate for our precharge above */
@@ -1301,8 +1301,8 @@ repeat:
 			SetPageUptodate(filepage);
 			set_page_dirty(filepage);
 			swap_free(swap);
-		} else if (!(error = add_to_page_cache(
-				swappage, mapping, idx, GFP_NOWAIT))) {
+		} else if (!(error = add_to_page_cache_locked(swappage, mapping,
+					idx, GFP_NOWAIT))) {
 			info->flags |= SHMEM_PAGEIN;
 			shmem_swp_set(info, entry, 0);
 			shmem_swp_unmap(entry);
diff --git a/mm/swap_state.c b/mm/swap_state.c
index d8aadaf2a0ba..3e3381d6c7ee 100644
--- a/mm/swap_state.c
+++ b/mm/swap_state.c
@@ -64,7 +64,7 @@ void show_swap_cache_info(void)
 }
 
 /*
- * add_to_swap_cache resembles add_to_page_cache on swapper_space,
+ * add_to_swap_cache resembles add_to_page_cache_locked on swapper_space,
  * but sets SwapCache flag and private instead of mapping and index.
  */
 int add_to_swap_cache(struct page *page, swp_entry_t entry, gfp_t gfp_mask)
@@ -76,19 +76,26 @@ int add_to_swap_cache(struct page *page, swp_entry_t entry, gfp_t gfp_mask)
 	BUG_ON(PagePrivate(page));
 	error = radix_tree_preload(gfp_mask);
 	if (!error) {
+		page_cache_get(page);
+		SetPageSwapCache(page);
+		set_page_private(page, entry.val);
+
 		write_lock_irq(&swapper_space.tree_lock);
 		error = radix_tree_insert(&swapper_space.page_tree,
 						entry.val, page);
-		if (!error) {
-			page_cache_get(page);
-			SetPageSwapCache(page);
-			set_page_private(page, entry.val);
+		if (likely(!error)) {
 			total_swapcache_pages++;
 			__inc_zone_page_state(page, NR_FILE_PAGES);
 			INC_CACHE_INFO(add_total);
 		}
 		write_unlock_irq(&swapper_space.tree_lock);
 		radix_tree_preload_end();
+
+		if (unlikely(error)) {
+			set_page_private(page, 0UL);
+			ClearPageSwapCache(page);
+			page_cache_release(page);
+		}
 	}
 	return error;
 }
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 26672c6cd3ce..0075eac1cd04 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -391,12 +391,10 @@ static pageout_t pageout(struct page *page, struct address_space *mapping,
 }
 
 /*
- * Attempt to detach a locked page from its ->mapping.  If it is dirty or if
- * someone else has a ref on the page, abort and return 0.  If it was
- * successfully detached, return 1.  Assumes the caller has a single ref on
- * this page.
+ * Same as remove_mapping, but if the page is removed from the mapping, it
+ * gets returned with a refcount of 0.
  */
-int remove_mapping(struct address_space *mapping, struct page *page)
+static int __remove_mapping(struct address_space *mapping, struct page *page)
 {
 	BUG_ON(!PageLocked(page));
 	BUG_ON(mapping != page_mapping(page));
@@ -427,24 +425,24 @@ int remove_mapping(struct address_space *mapping, struct page *page)
 	 * Note that if SetPageDirty is always performed via set_page_dirty,
 	 * and thus under tree_lock, then this ordering is not required.
 	 */
-	if (unlikely(page_count(page) != 2))
+	if (!page_freeze_refs(page, 2))
 		goto cannot_free;
-	smp_rmb();
-	if (unlikely(PageDirty(page)))
+	/* note: atomic_cmpxchg in page_freeze_refs provides the smp_rmb */
+	if (unlikely(PageDirty(page))) {
+		page_unfreeze_refs(page, 2);
 		goto cannot_free;
+	}
 
 	if (PageSwapCache(page)) {
 		swp_entry_t swap = { .val = page_private(page) };
 		__delete_from_swap_cache(page);
 		write_unlock_irq(&mapping->tree_lock);
 		swap_free(swap);
-		__put_page(page);	/* The pagecache ref */
-		return 1;
+	} else {
+		__remove_from_page_cache(page);
+		write_unlock_irq(&mapping->tree_lock);
 	}
 
-	__remove_from_page_cache(page);
-	write_unlock_irq(&mapping->tree_lock);
-	__put_page(page);
 	return 1;
 
 cannot_free:
@@ -452,6 +450,26 @@ cannot_free:
 	return 0;
 }
 
+/*
+ * Attempt to detach a locked page from its ->mapping.  If it is dirty or if
+ * someone else has a ref on the page, abort and return 0.  If it was
+ * successfully detached, return 1.  Assumes the caller has a single ref on
+ * this page.
+ */
+int remove_mapping(struct address_space *mapping, struct page *page)
+{
+	if (__remove_mapping(mapping, page)) {
+		/*
+		 * Unfreezing the refcount with 1 rather than 2 effectively
+		 * drops the pagecache ref for us without requiring another
+		 * atomic operation.
+		 */
+		page_unfreeze_refs(page, 1);
+		return 1;
+	}
+	return 0;
+}
+
 /*
  * shrink_page_list() returns the number of reclaimed pages
  */
@@ -598,18 +616,34 @@ static unsigned long shrink_page_list(struct list_head *page_list,
 		if (PagePrivate(page)) {
 			if (!try_to_release_page(page, sc->gfp_mask))
 				goto activate_locked;
-			if (!mapping && page_count(page) == 1)
-				goto free_it;
+			if (!mapping && page_count(page) == 1) {
+				unlock_page(page);
+				if (put_page_testzero(page))
+					goto free_it;
+				else {
+					/*
+					 * rare race with speculative reference.
+					 * the speculative reference will free
+					 * this page shortly, so we may
+					 * increment nr_reclaimed here (and
+					 * leave it off the LRU).
+					 */
+					nr_reclaimed++;
+					continue;
+				}
+			}
 		}
 
-		if (!mapping || !remove_mapping(mapping, page))
+		if (!mapping || !__remove_mapping(mapping, page))
 			goto keep_locked;
 
-free_it:
 		unlock_page(page);
+free_it:
 		nr_reclaimed++;
-		if (!pagevec_add(&freed_pvec, page))
-			__pagevec_release_nonlru(&freed_pvec);
+		if (!pagevec_add(&freed_pvec, page)) {
+			__pagevec_free(&freed_pvec);
+			pagevec_reinit(&freed_pvec);
+		}
 		continue;
 
 activate_locked:
@@ -623,7 +657,7 @@ keep:
 	}
 	list_splice(&ret_pages, page_list);
 	if (pagevec_count(&freed_pvec))
-		__pagevec_release_nonlru(&freed_pvec);
+		__pagevec_free(&freed_pvec);
 	count_vm_events(PGACTIVATE, pgactivate);
 	return nr_reclaimed;
 }

From a60637c85893e7191faaafa6a72e197c24386727 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Fri, 25 Jul 2008 19:45:31 -0700
Subject: [PATCH 1034/1171] mm: lockless pagecache

Combine page_cache_get_speculative with lockless radix tree lookups to
introduce lockless page cache lookups (ie.  no mapping->tree_lock on the
read-side).

The only atomicity changes this introduces is that the gang pagecache
lookup functions now behave as if they are implemented with multiple
find_get_page calls, rather than operating on a snapshot of the pages.  In
practice, this atomicity guarantee is not used anyway, and it is to
replace individual lookups, so these semantics are natural.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/filemap.c | 179 ++++++++++++++++++++++++++++++++++++++-------------
 1 file changed, 134 insertions(+), 45 deletions(-)

diff --git a/mm/filemap.c b/mm/filemap.c
index 4e182a9a14c0..feb8448d8618 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -637,15 +637,35 @@ void __lock_page_nosync(struct page *page)
  * Is there a pagecache struct page at the given (mapping, offset) tuple?
  * If yes, increment its refcount and return it; if no, return NULL.
  */
-struct page * find_get_page(struct address_space *mapping, pgoff_t offset)
+struct page *find_get_page(struct address_space *mapping, pgoff_t offset)
 {
+	void **pagep;
 	struct page *page;
 
-	read_lock_irq(&mapping->tree_lock);
-	page = radix_tree_lookup(&mapping->page_tree, offset);
-	if (page)
-		page_cache_get(page);
-	read_unlock_irq(&mapping->tree_lock);
+	rcu_read_lock();
+repeat:
+	page = NULL;
+	pagep = radix_tree_lookup_slot(&mapping->page_tree, offset);
+	if (pagep) {
+		page = radix_tree_deref_slot(pagep);
+		if (unlikely(!page || page == RADIX_TREE_RETRY))
+			goto repeat;
+
+		if (!page_cache_get_speculative(page))
+			goto repeat;
+
+		/*
+		 * Has the page moved?
+		 * This is part of the lockless pagecache protocol. See
+		 * include/linux/pagemap.h for details.
+		 */
+		if (unlikely(page != *pagep)) {
+			page_cache_release(page);
+			goto repeat;
+		}
+	}
+	rcu_read_unlock();
+
 	return page;
 }
 EXPORT_SYMBOL(find_get_page);
@@ -660,32 +680,22 @@ EXPORT_SYMBOL(find_get_page);
  *
  * Returns zero if the page was not present. find_lock_page() may sleep.
  */
-struct page *find_lock_page(struct address_space *mapping,
-				pgoff_t offset)
+struct page *find_lock_page(struct address_space *mapping, pgoff_t offset)
 {
 	struct page *page;
 
 repeat:
-	read_lock_irq(&mapping->tree_lock);
-	page = radix_tree_lookup(&mapping->page_tree, offset);
+	page = find_get_page(mapping, offset);
 	if (page) {
-		page_cache_get(page);
-		if (TestSetPageLocked(page)) {
-			read_unlock_irq(&mapping->tree_lock);
-			__lock_page(page);
-
-			/* Has the page been truncated while we slept? */
-			if (unlikely(page->mapping != mapping)) {
-				unlock_page(page);
-				page_cache_release(page);
-				goto repeat;
-			}
-			VM_BUG_ON(page->index != offset);
-			goto out;
+		lock_page(page);
+		/* Has the page been truncated? */
+		if (unlikely(page->mapping != mapping)) {
+			unlock_page(page);
+			page_cache_release(page);
+			goto repeat;
 		}
+		VM_BUG_ON(page->index != offset);
 	}
-	read_unlock_irq(&mapping->tree_lock);
-out:
 	return page;
 }
 EXPORT_SYMBOL(find_lock_page);
@@ -751,13 +761,39 @@ unsigned find_get_pages(struct address_space *mapping, pgoff_t start,
 {
 	unsigned int i;
 	unsigned int ret;
+	unsigned int nr_found;
 
-	read_lock_irq(&mapping->tree_lock);
-	ret = radix_tree_gang_lookup(&mapping->page_tree,
-				(void **)pages, start, nr_pages);
-	for (i = 0; i < ret; i++)
-		page_cache_get(pages[i]);
-	read_unlock_irq(&mapping->tree_lock);
+	rcu_read_lock();
+restart:
+	nr_found = radix_tree_gang_lookup_slot(&mapping->page_tree,
+				(void ***)pages, start, nr_pages);
+	ret = 0;
+	for (i = 0; i < nr_found; i++) {
+		struct page *page;
+repeat:
+		page = radix_tree_deref_slot((void **)pages[i]);
+		if (unlikely(!page))
+			continue;
+		/*
+		 * this can only trigger if nr_found == 1, making livelock
+		 * a non issue.
+		 */
+		if (unlikely(page == RADIX_TREE_RETRY))
+			goto restart;
+
+		if (!page_cache_get_speculative(page))
+			goto repeat;
+
+		/* Has the page moved? */
+		if (unlikely(page != *((void **)pages[i]))) {
+			page_cache_release(page);
+			goto repeat;
+		}
+
+		pages[ret] = page;
+		ret++;
+	}
+	rcu_read_unlock();
 	return ret;
 }
 
@@ -778,19 +814,44 @@ unsigned find_get_pages_contig(struct address_space *mapping, pgoff_t index,
 {
 	unsigned int i;
 	unsigned int ret;
+	unsigned int nr_found;
 
-	read_lock_irq(&mapping->tree_lock);
-	ret = radix_tree_gang_lookup(&mapping->page_tree,
-				(void **)pages, index, nr_pages);
-	for (i = 0; i < ret; i++) {
-		if (pages[i]->mapping == NULL || pages[i]->index != index)
+	rcu_read_lock();
+restart:
+	nr_found = radix_tree_gang_lookup_slot(&mapping->page_tree,
+				(void ***)pages, index, nr_pages);
+	ret = 0;
+	for (i = 0; i < nr_found; i++) {
+		struct page *page;
+repeat:
+		page = radix_tree_deref_slot((void **)pages[i]);
+		if (unlikely(!page))
+			continue;
+		/*
+		 * this can only trigger if nr_found == 1, making livelock
+		 * a non issue.
+		 */
+		if (unlikely(page == RADIX_TREE_RETRY))
+			goto restart;
+
+		if (page->mapping == NULL || page->index != index)
 			break;
 
-		page_cache_get(pages[i]);
+		if (!page_cache_get_speculative(page))
+			goto repeat;
+
+		/* Has the page moved? */
+		if (unlikely(page != *((void **)pages[i]))) {
+			page_cache_release(page);
+			goto repeat;
+		}
+
+		pages[ret] = page;
+		ret++;
 		index++;
 	}
-	read_unlock_irq(&mapping->tree_lock);
-	return i;
+	rcu_read_unlock();
+	return ret;
 }
 EXPORT_SYMBOL(find_get_pages_contig);
 
@@ -810,15 +871,43 @@ unsigned find_get_pages_tag(struct address_space *mapping, pgoff_t *index,
 {
 	unsigned int i;
 	unsigned int ret;
+	unsigned int nr_found;
+
+	rcu_read_lock();
+restart:
+	nr_found = radix_tree_gang_lookup_tag_slot(&mapping->page_tree,
+				(void ***)pages, *index, nr_pages, tag);
+	ret = 0;
+	for (i = 0; i < nr_found; i++) {
+		struct page *page;
+repeat:
+		page = radix_tree_deref_slot((void **)pages[i]);
+		if (unlikely(!page))
+			continue;
+		/*
+		 * this can only trigger if nr_found == 1, making livelock
+		 * a non issue.
+		 */
+		if (unlikely(page == RADIX_TREE_RETRY))
+			goto restart;
+
+		if (!page_cache_get_speculative(page))
+			goto repeat;
+
+		/* Has the page moved? */
+		if (unlikely(page != *((void **)pages[i]))) {
+			page_cache_release(page);
+			goto repeat;
+		}
+
+		pages[ret] = page;
+		ret++;
+	}
+	rcu_read_unlock();
 
-	read_lock_irq(&mapping->tree_lock);
-	ret = radix_tree_gang_lookup_tag(&mapping->page_tree,
-				(void **)pages, *index, nr_pages, tag);
-	for (i = 0; i < ret; i++)
-		page_cache_get(pages[i]);
 	if (ret)
 		*index = pages[ret - 1]->index + 1;
-	read_unlock_irq(&mapping->tree_lock);
+
 	return ret;
 }
 EXPORT_SYMBOL(find_get_pages_tag);

From 19fd6231279be3c3bdd02ed99f9b0eb195978064 Mon Sep 17 00:00:00 2001
From: Nick Piggin <npiggin@suse.de>
Date: Fri, 25 Jul 2008 19:45:32 -0700
Subject: [PATCH 1035/1171] mm: spinlock tree_lock

mapping->tree_lock has no read lockers.  convert the lock from an rwlock
to a spinlock.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: "Paul E. McKenney" <paulmck@us.ibm.com>
Reviewed-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/buffer.c                     |  4 ++--
 fs/inode.c                      |  2 +-
 include/asm-arm/cacheflush.h    |  4 ++--
 include/asm-parisc/cacheflush.h |  4 ++--
 include/linux/fs.h              |  2 +-
 mm/filemap.c                    | 10 +++++-----
 mm/migrate.c                    | 11 +++++------
 mm/page-writeback.c             | 12 ++++++------
 mm/swap_state.c                 | 10 +++++-----
 mm/swapfile.c                   |  4 ++--
 mm/truncate.c                   |  6 +++---
 mm/vmscan.c                     |  8 ++++----
 12 files changed, 38 insertions(+), 39 deletions(-)

diff --git a/fs/buffer.c b/fs/buffer.c
index d48caee12e2a..109b261192d9 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -706,7 +706,7 @@ static int __set_page_dirty(struct page *page,
 	if (TestSetPageDirty(page))
 		return 0;
 
-	write_lock_irq(&mapping->tree_lock);
+	spin_lock_irq(&mapping->tree_lock);
 	if (page->mapping) {	/* Race with truncate? */
 		WARN_ON_ONCE(warn && !PageUptodate(page));
 
@@ -719,7 +719,7 @@ static int __set_page_dirty(struct page *page,
 		radix_tree_tag_set(&mapping->page_tree,
 				page_index(page), PAGECACHE_TAG_DIRTY);
 	}
-	write_unlock_irq(&mapping->tree_lock);
+	spin_unlock_irq(&mapping->tree_lock);
 	__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
 
 	return 1;
diff --git a/fs/inode.c b/fs/inode.c
index c36d9480335c..35b6414522ea 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -209,7 +209,7 @@ void inode_init_once(struct inode *inode)
 	INIT_LIST_HEAD(&inode->i_dentry);
 	INIT_LIST_HEAD(&inode->i_devices);
 	INIT_RADIX_TREE(&inode->i_data.page_tree, GFP_ATOMIC);
-	rwlock_init(&inode->i_data.tree_lock);
+	spin_lock_init(&inode->i_data.tree_lock);
 	spin_lock_init(&inode->i_data.i_mmap_lock);
 	INIT_LIST_HEAD(&inode->i_data.private_list);
 	spin_lock_init(&inode->i_data.private_lock);
diff --git a/include/asm-arm/cacheflush.h b/include/asm-arm/cacheflush.h
index 70b0fe724b62..03cf1ee977b7 100644
--- a/include/asm-arm/cacheflush.h
+++ b/include/asm-arm/cacheflush.h
@@ -424,9 +424,9 @@ static inline void flush_anon_page(struct vm_area_struct *vma,
 }
 
 #define flush_dcache_mmap_lock(mapping) \
-	write_lock_irq(&(mapping)->tree_lock)
+	spin_lock_irq(&(mapping)->tree_lock)
 #define flush_dcache_mmap_unlock(mapping) \
-	write_unlock_irq(&(mapping)->tree_lock)
+	spin_unlock_irq(&(mapping)->tree_lock)
 
 #define flush_icache_user_range(vma,page,addr,len) \
 	flush_dcache_page(page)
diff --git a/include/asm-parisc/cacheflush.h b/include/asm-parisc/cacheflush.h
index 2f1e1b05440a..b7ca6dc7fddc 100644
--- a/include/asm-parisc/cacheflush.h
+++ b/include/asm-parisc/cacheflush.h
@@ -45,9 +45,9 @@ void flush_cache_mm(struct mm_struct *mm);
 extern void flush_dcache_page(struct page *page);
 
 #define flush_dcache_mmap_lock(mapping) \
-	write_lock_irq(&(mapping)->tree_lock)
+	spin_lock_irq(&(mapping)->tree_lock)
 #define flush_dcache_mmap_unlock(mapping) \
-	write_unlock_irq(&(mapping)->tree_lock)
+	spin_unlock_irq(&(mapping)->tree_lock)
 
 #define flush_icache_page(vma,page)	do { 		\
 	flush_kernel_dcache_page(page);			\
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 49d8eb7a71be..53d2edb709b3 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -499,7 +499,7 @@ struct backing_dev_info;
 struct address_space {
 	struct inode		*host;		/* owner: inode, block_device */
 	struct radix_tree_root	page_tree;	/* radix tree of all pages */
-	rwlock_t		tree_lock;	/* and rwlock protecting it */
+	spinlock_t		tree_lock;	/* and lock protecting it */
 	unsigned int		i_mmap_writable;/* count VM_SHARED mappings */
 	struct prio_tree_root	i_mmap;		/* tree of private and shared mappings */
 	struct list_head	i_mmap_nonlinear;/*list VM_NONLINEAR mappings */
diff --git a/mm/filemap.c b/mm/filemap.c
index feb8448d8618..2ed8b0389c51 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -109,7 +109,7 @@
 /*
  * Remove a page from the page cache and free it. Caller has to make
  * sure the page is locked and that nobody else uses it - or that usage
- * is safe.  The caller must hold a write_lock on the mapping's tree_lock.
+ * is safe.  The caller must hold the mapping's tree_lock.
  */
 void __remove_from_page_cache(struct page *page)
 {
@@ -141,9 +141,9 @@ void remove_from_page_cache(struct page *page)
 
 	BUG_ON(!PageLocked(page));
 
-	write_lock_irq(&mapping->tree_lock);
+	spin_lock_irq(&mapping->tree_lock);
 	__remove_from_page_cache(page);
-	write_unlock_irq(&mapping->tree_lock);
+	spin_unlock_irq(&mapping->tree_lock);
 }
 
 static int sync_page(void *word)
@@ -469,7 +469,7 @@ int add_to_page_cache_locked(struct page *page, struct address_space *mapping,
 		page->mapping = mapping;
 		page->index = offset;
 
-		write_lock_irq(&mapping->tree_lock);
+		spin_lock_irq(&mapping->tree_lock);
 		error = radix_tree_insert(&mapping->page_tree, offset, page);
 		if (likely(!error)) {
 			mapping->nrpages++;
@@ -480,7 +480,7 @@ int add_to_page_cache_locked(struct page *page, struct address_space *mapping,
 			page_cache_release(page);
 		}
 
-		write_unlock_irq(&mapping->tree_lock);
+		spin_unlock_irq(&mapping->tree_lock);
 		radix_tree_preload_end();
 	} else
 		mem_cgroup_uncharge_cache_page(page);
diff --git a/mm/migrate.c b/mm/migrate.c
index 3ca6392e82cc..153572fb60b8 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -323,7 +323,7 @@ static int migrate_page_move_mapping(struct address_space *mapping,
 		return 0;
 	}
 
-	write_lock_irq(&mapping->tree_lock);
+	spin_lock_irq(&mapping->tree_lock);
 
 	pslot = radix_tree_lookup_slot(&mapping->page_tree,
  					page_index(page));
@@ -331,12 +331,12 @@ static int migrate_page_move_mapping(struct address_space *mapping,
 	expected_count = 2 + !!PagePrivate(page);
 	if (page_count(page) != expected_count ||
 			(struct page *)radix_tree_deref_slot(pslot) != page) {
-		write_unlock_irq(&mapping->tree_lock);
+		spin_unlock_irq(&mapping->tree_lock);
 		return -EAGAIN;
 	}
 
 	if (!page_freeze_refs(page, expected_count)) {
-		write_unlock_irq(&mapping->tree_lock);
+		spin_unlock_irq(&mapping->tree_lock);
 		return -EAGAIN;
 	}
 
@@ -373,10 +373,9 @@ static int migrate_page_move_mapping(struct address_space *mapping,
 	__dec_zone_page_state(page, NR_FILE_PAGES);
 	__inc_zone_page_state(newpage, NR_FILE_PAGES);
 
-	write_unlock_irq(&mapping->tree_lock);
-	if (!PageSwapCache(newpage)) {
+	spin_unlock_irq(&mapping->tree_lock);
+	if (!PageSwapCache(newpage))
 		mem_cgroup_uncharge_cache_page(page);
-	}
 
 	return 0;
 }
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 94c6d8988ab3..24de8b65fdbd 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -1088,7 +1088,7 @@ int __set_page_dirty_nobuffers(struct page *page)
 		if (!mapping)
 			return 1;
 
-		write_lock_irq(&mapping->tree_lock);
+		spin_lock_irq(&mapping->tree_lock);
 		mapping2 = page_mapping(page);
 		if (mapping2) { /* Race with truncate? */
 			BUG_ON(mapping2 != mapping);
@@ -1102,7 +1102,7 @@ int __set_page_dirty_nobuffers(struct page *page)
 			radix_tree_tag_set(&mapping->page_tree,
 				page_index(page), PAGECACHE_TAG_DIRTY);
 		}
-		write_unlock_irq(&mapping->tree_lock);
+		spin_unlock_irq(&mapping->tree_lock);
 		if (mapping->host) {
 			/* !PageAnon && !swapper_space */
 			__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
@@ -1258,7 +1258,7 @@ int test_clear_page_writeback(struct page *page)
 		struct backing_dev_info *bdi = mapping->backing_dev_info;
 		unsigned long flags;
 
-		write_lock_irqsave(&mapping->tree_lock, flags);
+		spin_lock_irqsave(&mapping->tree_lock, flags);
 		ret = TestClearPageWriteback(page);
 		if (ret) {
 			radix_tree_tag_clear(&mapping->page_tree,
@@ -1269,7 +1269,7 @@ int test_clear_page_writeback(struct page *page)
 				__bdi_writeout_inc(bdi);
 			}
 		}
-		write_unlock_irqrestore(&mapping->tree_lock, flags);
+		spin_unlock_irqrestore(&mapping->tree_lock, flags);
 	} else {
 		ret = TestClearPageWriteback(page);
 	}
@@ -1287,7 +1287,7 @@ int test_set_page_writeback(struct page *page)
 		struct backing_dev_info *bdi = mapping->backing_dev_info;
 		unsigned long flags;
 
-		write_lock_irqsave(&mapping->tree_lock, flags);
+		spin_lock_irqsave(&mapping->tree_lock, flags);
 		ret = TestSetPageWriteback(page);
 		if (!ret) {
 			radix_tree_tag_set(&mapping->page_tree,
@@ -1300,7 +1300,7 @@ int test_set_page_writeback(struct page *page)
 			radix_tree_tag_clear(&mapping->page_tree,
 						page_index(page),
 						PAGECACHE_TAG_DIRTY);
-		write_unlock_irqrestore(&mapping->tree_lock, flags);
+		spin_unlock_irqrestore(&mapping->tree_lock, flags);
 	} else {
 		ret = TestSetPageWriteback(page);
 	}
diff --git a/mm/swap_state.c b/mm/swap_state.c
index 3e3381d6c7ee..2c217e33d497 100644
--- a/mm/swap_state.c
+++ b/mm/swap_state.c
@@ -39,7 +39,7 @@ static struct backing_dev_info swap_backing_dev_info = {
 
 struct address_space swapper_space = {
 	.page_tree	= RADIX_TREE_INIT(GFP_ATOMIC|__GFP_NOWARN),
-	.tree_lock	= __RW_LOCK_UNLOCKED(swapper_space.tree_lock),
+	.tree_lock	= __SPIN_LOCK_UNLOCKED(swapper_space.tree_lock),
 	.a_ops		= &swap_aops,
 	.i_mmap_nonlinear = LIST_HEAD_INIT(swapper_space.i_mmap_nonlinear),
 	.backing_dev_info = &swap_backing_dev_info,
@@ -80,7 +80,7 @@ int add_to_swap_cache(struct page *page, swp_entry_t entry, gfp_t gfp_mask)
 		SetPageSwapCache(page);
 		set_page_private(page, entry.val);
 
-		write_lock_irq(&swapper_space.tree_lock);
+		spin_lock_irq(&swapper_space.tree_lock);
 		error = radix_tree_insert(&swapper_space.page_tree,
 						entry.val, page);
 		if (likely(!error)) {
@@ -88,7 +88,7 @@ int add_to_swap_cache(struct page *page, swp_entry_t entry, gfp_t gfp_mask)
 			__inc_zone_page_state(page, NR_FILE_PAGES);
 			INC_CACHE_INFO(add_total);
 		}
-		write_unlock_irq(&swapper_space.tree_lock);
+		spin_unlock_irq(&swapper_space.tree_lock);
 		radix_tree_preload_end();
 
 		if (unlikely(error)) {
@@ -182,9 +182,9 @@ void delete_from_swap_cache(struct page *page)
 
 	entry.val = page_private(page);
 
-	write_lock_irq(&swapper_space.tree_lock);
+	spin_lock_irq(&swapper_space.tree_lock);
 	__delete_from_swap_cache(page);
-	write_unlock_irq(&swapper_space.tree_lock);
+	spin_unlock_irq(&swapper_space.tree_lock);
 
 	swap_free(entry);
 	page_cache_release(page);
diff --git a/mm/swapfile.c b/mm/swapfile.c
index 2f33edb8bee9..af283933c14e 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -369,13 +369,13 @@ int remove_exclusive_swap_page(struct page *page)
 	retval = 0;
 	if (p->swap_map[swp_offset(entry)] == 1) {
 		/* Recheck the page count with the swapcache lock held.. */
-		write_lock_irq(&swapper_space.tree_lock);
+		spin_lock_irq(&swapper_space.tree_lock);
 		if ((page_count(page) == 2) && !PageWriteback(page)) {
 			__delete_from_swap_cache(page);
 			SetPageDirty(page);
 			retval = 1;
 		}
-		write_unlock_irq(&swapper_space.tree_lock);
+		spin_unlock_irq(&swapper_space.tree_lock);
 	}
 	spin_unlock(&swap_lock);
 
diff --git a/mm/truncate.c b/mm/truncate.c
index b8961cb63414..e68443d74567 100644
--- a/mm/truncate.c
+++ b/mm/truncate.c
@@ -349,18 +349,18 @@ invalidate_complete_page2(struct address_space *mapping, struct page *page)
 	if (PagePrivate(page) && !try_to_release_page(page, GFP_KERNEL))
 		return 0;
 
-	write_lock_irq(&mapping->tree_lock);
+	spin_lock_irq(&mapping->tree_lock);
 	if (PageDirty(page))
 		goto failed;
 
 	BUG_ON(PagePrivate(page));
 	__remove_from_page_cache(page);
-	write_unlock_irq(&mapping->tree_lock);
+	spin_unlock_irq(&mapping->tree_lock);
 	ClearPageUptodate(page);
 	page_cache_release(page);	/* pagecache ref */
 	return 1;
 failed:
-	write_unlock_irq(&mapping->tree_lock);
+	spin_unlock_irq(&mapping->tree_lock);
 	return 0;
 }
 
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 0075eac1cd04..8f71761bc4b7 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -399,7 +399,7 @@ static int __remove_mapping(struct address_space *mapping, struct page *page)
 	BUG_ON(!PageLocked(page));
 	BUG_ON(mapping != page_mapping(page));
 
-	write_lock_irq(&mapping->tree_lock);
+	spin_lock_irq(&mapping->tree_lock);
 	/*
 	 * The non racy check for a busy page.
 	 *
@@ -436,17 +436,17 @@ static int __remove_mapping(struct address_space *mapping, struct page *page)
 	if (PageSwapCache(page)) {
 		swp_entry_t swap = { .val = page_private(page) };
 		__delete_from_swap_cache(page);
-		write_unlock_irq(&mapping->tree_lock);
+		spin_unlock_irq(&mapping->tree_lock);
 		swap_free(swap);
 	} else {
 		__remove_from_page_cache(page);
-		write_unlock_irq(&mapping->tree_lock);
+		spin_unlock_irq(&mapping->tree_lock);
 	}
 
 	return 1;
 
 cannot_free:
-	write_unlock_irq(&mapping->tree_lock);
+	spin_unlock_irq(&mapping->tree_lock);
 	return 0;
 }
 

From d91958815d214ea365b98cbff6215383897edcb6 Mon Sep 17 00:00:00 2001
From: Matt LaPlante <kernel1@cyberdogtech.com>
Date: Fri, 25 Jul 2008 19:45:33 -0700
Subject: [PATCH 1036/1171] Documentation cleanup: trivial misspelling,
 punctuation, and grammar corrections.

Cc: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 Documentation/Intel-IOMMU.txt                     |  4 ++--
 Documentation/accounting/taskstats-struct.txt     |  2 +-
 Documentation/cpu-freq/governors.txt              |  2 +-
 Documentation/edac.txt                            |  2 +-
 Documentation/filesystems/proc.txt                |  4 ++--
 Documentation/filesystems/vfs.txt                 |  6 +++---
 Documentation/ia64/kvm.txt                        |  8 ++++----
 Documentation/input/cs461x.txt                    |  2 +-
 Documentation/ioctl/ioctl-decoding.txt            |  4 ++--
 Documentation/iostats.txt                         |  2 +-
 Documentation/keys.txt                            |  2 +-
 Documentation/leds-class.txt                      |  2 +-
 Documentation/local_ops.txt                       |  2 +-
 Documentation/networking/bonding.txt              |  2 +-
 Documentation/networking/can.txt                  |  4 ++--
 Documentation/networking/packet_mmap.txt          |  2 +-
 Documentation/networking/tc-actions-env-rules.txt | 15 ++++++++-------
 Documentation/powerpc/booting-without-of.txt      |  8 ++++----
 Documentation/powerpc/qe_firmware.txt             |  2 +-
 Documentation/s390/driver-model.txt               |  2 +-
 Documentation/scsi/ibmmca.txt                     |  6 +++---
 Documentation/scsi/lpfc.txt                       |  2 +-
 Documentation/scsi/scsi_fc_transport.txt          |  6 +++---
 Documentation/sh/clk.txt                          |  2 +-
 Documentation/sound/alsa/Audiophile-Usb.txt       | 10 +++++-----
 Documentation/sound/alsa/hda_codec.txt            |  2 +-
 Documentation/sound/alsa/soc/dapm.txt             |  2 +-
 Documentation/sysctl/vm.txt                       |  2 +-
 Documentation/timers/highres.txt                  |  2 +-
 Documentation/usb/authorization.txt               |  2 +-
 Documentation/video4linux/sn9c102.txt             |  2 +-
 Documentation/vm/hugetlbpage.txt                  |  2 +-
 Documentation/vm/numa_memory_policy.txt           |  4 ++--
 Documentation/volatile-considered-harmful.txt     |  2 +-
 drivers/message/fusion/lsi/mpi_history.txt        |  6 +++---
 35 files changed, 65 insertions(+), 64 deletions(-)

diff --git a/Documentation/Intel-IOMMU.txt b/Documentation/Intel-IOMMU.txt
index c2321903aa09..21bc416d887e 100644
--- a/Documentation/Intel-IOMMU.txt
+++ b/Documentation/Intel-IOMMU.txt
@@ -48,7 +48,7 @@ IOVA generation is pretty generic. We used the same technique as vmalloc()
 but these are not global address spaces, but separate for each domain.
 Different DMA engines may support different number of domains.
 
-We also allocate gaurd pages with each mapping, so we can attempt to catch
+We also allocate guard pages with each mapping, so we can attempt to catch
 any overflow that might happen.
 
 
@@ -112,4 +112,4 @@ TBD
 
 - For compatibility testing, could use unity map domain for all devices, just
   provide a 1-1 for all useful memory under a single domain for all devices.
-- API for paravirt ops for abstracting functionlity for VMM folks.
+- API for paravirt ops for abstracting functionality for VMM folks.
diff --git a/Documentation/accounting/taskstats-struct.txt b/Documentation/accounting/taskstats-struct.txt
index b988d110db59..e7512c061c15 100644
--- a/Documentation/accounting/taskstats-struct.txt
+++ b/Documentation/accounting/taskstats-struct.txt
@@ -6,7 +6,7 @@ This document contains an explanation of the struct taskstats fields.
 There are three different groups of fields in the struct taskstats:
 
 1) Common and basic accounting fields
-    If CONFIG_TASKSTATS is set, the taskstats inteface is enabled and
+    If CONFIG_TASKSTATS is set, the taskstats interface is enabled and
     the common fields and basic accounting fields are collected for
     delivery at do_exit() of a task.
 2) Delay accounting fields
diff --git a/Documentation/cpu-freq/governors.txt b/Documentation/cpu-freq/governors.txt
index dcec0564d040..5b0cfa67aff9 100644
--- a/Documentation/cpu-freq/governors.txt
+++ b/Documentation/cpu-freq/governors.txt
@@ -122,7 +122,7 @@ around '10000' or more.
 show_sampling_rate_(min|max): the minimum and maximum sampling rates
 available that you may set 'sampling_rate' to.
 
-up_threshold: defines what the average CPU usaged between the samplings
+up_threshold: defines what the average CPU usage between the samplings
 of 'sampling_rate' needs to be for the kernel to make a decision on
 whether it should increase the frequency.  For example when it is set
 to its default value of '80' it means that between the checking
diff --git a/Documentation/edac.txt b/Documentation/edac.txt
index ced527388001..8eda3fb66416 100644
--- a/Documentation/edac.txt
+++ b/Documentation/edac.txt
@@ -327,7 +327,7 @@ Sdram memory scrubbing rate:
 	'sdram_scrub_rate'
 
 	Read/Write attribute file that controls memory scrubbing. The scrubbing
-	rate is set by writing a minimum bandwith in bytes/sec to the attribute
+	rate is set by writing a minimum bandwidth in bytes/sec to the attribute
 	file. The rate will be translated to an internal value that gives at
 	least the specified rate.
 
diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt
index 8c6384bdfed4..64557821ee59 100644
--- a/Documentation/filesystems/proc.txt
+++ b/Documentation/filesystems/proc.txt
@@ -931,7 +931,7 @@ group_prealloc  max_to_scan  mb_groups  mb_history  min_to_scan  order2_req
 stats  stream_req
 
 mb_groups:
-This file gives the details of mutiblock allocator buddy cache of free blocks
+This file gives the details of multiblock allocator buddy cache of free blocks
 
 mb_history:
 Multiblock allocation history.
@@ -1474,7 +1474,7 @@ used because pages_free(1355) is smaller than watermark + protection[2]
 normal page requirement. If requirement is DMA zone(index=0), protection[0]
 (=0) is used.
 
-zone[i]'s protection[j] is calculated by following exprssion.
+zone[i]'s protection[j] is calculated by following expression.
 
 (i < j):
   zone[i]->protection[j]
diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt
index b7522c6cbae3..c4d348dabe94 100644
--- a/Documentation/filesystems/vfs.txt
+++ b/Documentation/filesystems/vfs.txt
@@ -143,7 +143,7 @@ struct file_system_type {
 
 The get_sb() method has the following arguments:
 
-  struct file_system_type *fs_type: decribes the filesystem, partly initialized
+  struct file_system_type *fs_type: describes the filesystem, partly initialized
   	by the specific filesystem code
 
   int flags: mount flags
@@ -895,9 +895,9 @@ struct dentry_operations {
 	iput() yourself
 
   d_dname: called when the pathname of a dentry should be generated.
-	Usefull for some pseudo filesystems (sockfs, pipefs, ...) to delay
+	Useful for some pseudo filesystems (sockfs, pipefs, ...) to delay
 	pathname generation. (Instead of doing it when dentry is created,
-	its done only when the path is needed.). Real filesystems probably
+	it's done only when the path is needed.). Real filesystems probably
 	dont want to use it, because their dentries are present in global
 	dcache hash, so their hash should be an invariant. As no lock is
 	held, d_dname() should not try to modify the dentry itself, unless
diff --git a/Documentation/ia64/kvm.txt b/Documentation/ia64/kvm.txt
index bec9d815da33..914d07f49268 100644
--- a/Documentation/ia64/kvm.txt
+++ b/Documentation/ia64/kvm.txt
@@ -50,9 +50,9 @@ Note: For step 2, please make sure that host page size == TARGET_PAGE_SIZE of qe
 		/usr/local/bin/qemu-system-ia64 -smp xx -m 512 -hda $your_image
 		(xx is the number of virtual processors for the guest, now the maximum value is 4)
 
-5. Known possibile issue on some platforms with old Firmware.
+5. Known possible issue on some platforms with old Firmware.
 
-If meet strange host crashe issues, try to solve it through either of the following ways:
+In the event of strange host crash issues, try to solve it through either of the following ways:
 
 (1): Upgrade your Firmware to the latest one.
 
@@ -65,8 +65,8 @@ index 0b53344..f02b0f7 100644
 	mov ar.pfs = loc1
 	mov rp = loc0
 	;;
--	srlz.d				// seralize restoration of psr.l
-+	srlz.i			// seralize restoration of psr.l
+-	srlz.d				// serialize restoration of psr.l
++	srlz.i			// serialize restoration of psr.l
 +	;;
 	br.ret.sptk.many b0
  END(ia64_pal_call_static)
diff --git a/Documentation/input/cs461x.txt b/Documentation/input/cs461x.txt
index afe0d6543e09..202e9dbacec3 100644
--- a/Documentation/input/cs461x.txt
+++ b/Documentation/input/cs461x.txt
@@ -31,7 +31,7 @@ The driver works with ALSA drivers simultaneously. For example, the xracer
 uses joystick as input device and PCM device as sound output in one time.
 There are no sound or input collisions detected. The source code have
 comments about them; but I've found the joystick can be initialized 
-separately of ALSA modules. So, you canm use only one joystick driver
+separately of ALSA modules. So, you can use only one joystick driver
 without ALSA drivers. The ALSA drivers are not needed to compile or
 run this driver.
 
diff --git a/Documentation/ioctl/ioctl-decoding.txt b/Documentation/ioctl/ioctl-decoding.txt
index bfdf7f3ee4f0..e35efb0cec2e 100644
--- a/Documentation/ioctl/ioctl-decoding.txt
+++ b/Documentation/ioctl/ioctl-decoding.txt
@@ -1,6 +1,6 @@
 To decode a hex IOCTL code:
 
-Most architecures use this generic format, but check
+Most architectures use this generic format, but check
 include/ARCH/ioctl.h for specifics, e.g. powerpc
 uses 3 bits to encode read/write and 13 bits for size.
 
@@ -18,7 +18,7 @@ uses 3 bits to encode read/write and 13 bits for size.
  7-0	function #
 
 
- So for example 0x82187201 is a read with arg length of 0x218,
+So for example 0x82187201 is a read with arg length of 0x218,
 character 'r' function 1. Grepping the source reveals this is:
 
 #define VFAT_IOCTL_READDIR_BOTH         _IOR('r', 1, struct dirent [2])
diff --git a/Documentation/iostats.txt b/Documentation/iostats.txt
index 5925c3cd030d..59a69ec67c40 100644
--- a/Documentation/iostats.txt
+++ b/Documentation/iostats.txt
@@ -143,7 +143,7 @@ disk and partition statistics are consistent again. Since we still don't
 keep record of the partition-relative address, an operation is attributed to
 the partition which contains the first sector of the request after the
 eventual merges. As requests can be merged across partition, this could lead
-to some (probably insignificant) innacuracy.
+to some (probably insignificant) inaccuracy.
 
 Additional notes
 ----------------
diff --git a/Documentation/keys.txt b/Documentation/keys.txt
index d5c7a57d1700..b56aacc1fff8 100644
--- a/Documentation/keys.txt
+++ b/Documentation/keys.txt
@@ -864,7 +864,7 @@ payload contents" for more information.
     request_key_with_auxdata() respectively.
 
     These two functions return with the key potentially still under
-    construction.  To wait for contruction completion, the following should be
+    construction.  To wait for construction completion, the following should be
     called:
 
 	int wait_for_key_construction(struct key *key, bool intr);
diff --git a/Documentation/leds-class.txt b/Documentation/leds-class.txt
index 18860ad9935a..6399557cdab3 100644
--- a/Documentation/leds-class.txt
+++ b/Documentation/leds-class.txt
@@ -59,7 +59,7 @@ Hardware accelerated blink of LEDs
 
 Some LEDs can be programmed to blink without any CPU interaction. To
 support this feature, a LED driver can optionally implement the
-blink_set() function (see <linux/leds.h>). If implemeted, triggers can
+blink_set() function (see <linux/leds.h>). If implemented, triggers can
 attempt to use it before falling back to software timers. The blink_set()
 function should return 0 if the blink setting is supported, or -EINVAL
 otherwise, which means that LED blinking will be handled by software.
diff --git a/Documentation/local_ops.txt b/Documentation/local_ops.txt
index 4269a1105b37..f4f8b1c6c8ba 100644
--- a/Documentation/local_ops.txt
+++ b/Documentation/local_ops.txt
@@ -36,7 +36,7 @@ It can be done by slightly modifying the standard atomic operations : only
 their UP variant must be kept. It typically means removing LOCK prefix (on
 i386 and x86_64) and any SMP sychronization barrier. If the architecture does
 not have a different behavior between SMP and UP, including asm-generic/local.h
-in your archtecture's local.h is sufficient.
+in your architecture's local.h is sufficient.
 
 The local_t type is defined as an opaque signed long by embedding an
 atomic_long_t inside a structure. This is made so a cast from this type to a
diff --git a/Documentation/networking/bonding.txt b/Documentation/networking/bonding.txt
index 7fa7fe71d7a8..688dfe1e6b70 100644
--- a/Documentation/networking/bonding.txt
+++ b/Documentation/networking/bonding.txt
@@ -631,7 +631,7 @@ xmit_hash_policy
 		in environments where a layer3 gateway device is
 		required to reach most destinations.
 
-		This algorithm is 802.3ad complient.
+		This algorithm is 802.3ad compliant.
 
 	layer3+4
 
diff --git a/Documentation/networking/can.txt b/Documentation/networking/can.txt
index 641d2afacffa..297ba7b1ccaf 100644
--- a/Documentation/networking/can.txt
+++ b/Documentation/networking/can.txt
@@ -186,7 +186,7 @@ solution for a couple of reasons:
 
   The Linux network devices (by default) just can handle the
   transmission and reception of media dependent frames. Due to the
-  arbritration on the CAN bus the transmission of a low prio CAN-ID
+  arbitration on the CAN bus the transmission of a low prio CAN-ID
   may be delayed by the reception of a high prio CAN frame. To
   reflect the correct* traffic on the node the loopback of the sent
   data has to be performed right after a successful transmission. If
@@ -481,7 +481,7 @@ solution for a couple of reasons:
   - stats_timer: To calculate the Socket CAN core statistics
     (e.g. current/maximum frames per second) this 1 second timer is
     invoked at can.ko module start time by default. This timer can be
-    disabled by using stattimer=0 on the module comandline.
+    disabled by using stattimer=0 on the module commandline.
 
   - debug: (removed since SocketCAN SVN r546)
 
diff --git a/Documentation/networking/packet_mmap.txt b/Documentation/networking/packet_mmap.txt
index db0cd5169581..07c53d596035 100644
--- a/Documentation/networking/packet_mmap.txt
+++ b/Documentation/networking/packet_mmap.txt
@@ -326,7 +326,7 @@ just one call to mmap is needed:
     mmap(0, size, PROT_READ|PROT_WRITE, MAP_SHARED, fd, 0);
 
 If tp_frame_size is a divisor of tp_block_size frames will be 
-contiguosly spaced by tp_frame_size bytes. If not, each 
+contiguously spaced by tp_frame_size bytes. If not, each
 tp_block_size/tp_frame_size frames there will be a gap between 
 the frames. This is because a frame cannot be spawn across two
 blocks. 
diff --git a/Documentation/networking/tc-actions-env-rules.txt b/Documentation/networking/tc-actions-env-rules.txt
index 01e716d185f4..dcadf6f88e34 100644
--- a/Documentation/networking/tc-actions-env-rules.txt
+++ b/Documentation/networking/tc-actions-env-rules.txt
@@ -4,26 +4,27 @@ The "enviromental" rules for authors of any new tc actions are:
 1) If you stealeth or borroweth any packet thou shalt be branching
 from the righteous path and thou shalt cloneth.
 
-For example if your action queues a packet to be processed later
-or intentionaly branches by redirecting a packet then you need to
+For example if your action queues a packet to be processed later,
+or intentionally branches by redirecting a packet, then you need to
 clone the packet.
+
 There are certain fields in the skb tc_verd that need to be reset so we
-avoid loops etc. A few are generic enough so much so that skb_act_clone()
-resets them for you. So invoke skb_act_clone() rather than skb_clone()
+avoid loops, etc.  A few are generic enough that skb_act_clone()
+resets them for you, so invoke skb_act_clone() rather than skb_clone().
 
 2) If you munge any packet thou shalt call pskb_expand_head in the case
 someone else is referencing the skb. After that you "own" the skb.
 You must also tell us if it is ok to munge the packet (TC_OK2MUNGE),
 this way any action downstream can stomp on the packet.
 
-3) dropping packets you dont own is a nono. You simply return
+3) Dropping packets you don't own is a no-no. You simply return
 TC_ACT_SHOT to the caller and they will drop it.
 
 The "enviromental" rules for callers of actions (qdiscs etc) are:
 
-*) thou art responsible for freeing anything returned as being
+*) Thou art responsible for freeing anything returned as being
 TC_ACT_SHOT/STOLEN/QUEUED. If none of TC_ACT_SHOT/STOLEN/QUEUED is
-returned then all is great and you dont need to do anything.
+returned, then all is great and you don't need to do anything.
 
 Post on netdev if something is unclear.
 
diff --git a/Documentation/powerpc/booting-without-of.txt b/Documentation/powerpc/booting-without-of.txt
index 99514ced82c5..928a79ceb7aa 100644
--- a/Documentation/powerpc/booting-without-of.txt
+++ b/Documentation/powerpc/booting-without-of.txt
@@ -708,7 +708,7 @@ device or bus to be described by the device tree.
 In general, the format of an address for a device is defined by the
 parent bus type, based on the #address-cells and #size-cells
 properties.  Note that the parent's parent definitions of #address-cells
-and #size-cells are not inhereted so every node with children must specify
+and #size-cells are not inherited so every node with children must specify
 them.  The kernel requires the root node to have those properties defining
 addresses format for devices directly mapped on the processor bus.
 
@@ -1777,7 +1777,7 @@ platforms are moved over to use the flattened-device-tree model.
 
       Xilinx uartlite devices are simple fixed speed serial ports.
 
-      Requred properties:
+      Required properties:
        - current-speed : Baud rate of uartlite
 
       v) Xilinx hwicap
@@ -1799,7 +1799,7 @@ platforms are moved over to use the flattened-device-tree model.
       Xilinx UART 16550 devices are very similar to the NS16550 but with
       different register spacing and an offset from the base address.
 
-      Requred properties:
+      Required properties:
        - clock-frequency : Frequency of the clock input
        - reg-offset : A value of 3 is required
        - reg-shift : A value of 2 is required
@@ -1953,7 +1953,7 @@ prefixed with the string "marvell,", for Marvell Technology Group Ltd.
 1) The /system-controller node
 
   This node is used to represent the system-controller and must be
-  present when the system uses a system contller chip. The top-level
+  present when the system uses a system controller chip. The top-level
   system-controller node contains information that is global to all
   devices within the system controller chip. The node name begins
   with "system-controller" followed by the unit address, which is
diff --git a/Documentation/powerpc/qe_firmware.txt b/Documentation/powerpc/qe_firmware.txt
index 896266432d33..06da4d4b44f9 100644
--- a/Documentation/powerpc/qe_firmware.txt
+++ b/Documentation/powerpc/qe_firmware.txt
@@ -217,7 +217,7 @@ Although it is not recommended, you can specify '0' in the soc.model
 field to skip matching SOCs altogether.
 
 The 'model' field is a 16-bit number that matches the actual SOC. The
-'major' and 'minor' fields are the major and minor revision numbrs,
+'major' and 'minor' fields are the major and minor revision numbers,
 respectively, of the SOC.
 
 For example, to match the 8323, revision 1.0:
diff --git a/Documentation/s390/driver-model.txt b/Documentation/s390/driver-model.txt
index e938c442277d..bde473df748d 100644
--- a/Documentation/s390/driver-model.txt
+++ b/Documentation/s390/driver-model.txt
@@ -25,7 +25,7 @@ device 4711 via subchannel 1 in subchannel set 0, and subchannel 2 is a non-I/O
 subchannel. Device 1234 is accessed via subchannel 0 in subchannel set 1.
 
 The subchannel named 'defunct' does not represent any real subchannel on the
-system; it is a pseudo subchannel where disconnnected ccw devices are moved to
+system; it is a pseudo subchannel where disconnected ccw devices are moved to
 if they are displaced by another ccw device becoming operational on their
 former subchannel. The ccw devices will be moved again to a proper subchannel
 if they become operational again on that subchannel.
diff --git a/Documentation/scsi/ibmmca.txt b/Documentation/scsi/ibmmca.txt
index a810421f1fb3..3920f28710c4 100644
--- a/Documentation/scsi/ibmmca.txt
+++ b/Documentation/scsi/ibmmca.txt
@@ -524,7 +524,7 @@
    - Michael Lang
  
    June 25 1997: (v1.8b)
-   1) Some cosmetical changes for the handling of SCSI-device-types.
+   1) Some cosmetic changes for the handling of SCSI-device-types.
       Now, also CD-Burners / WORMs and SCSI-scanners should work. For
       MO-drives I have no experience, therefore not yet supported.
       In logical_devices I changed from different type-variables to one
@@ -914,7 +914,7 @@
       in version 4.0. This was never really necessary, as all troubles were
       based on non-command related reasons up to now, so bypassing commands
       did not help to avoid any bugs. It is kept in 3.2X for debugging reasons.
-   5) Dynamical reassignment of ldns was again verified and analyzed to be
+   5) Dynamic reassignment of ldns was again verified and analyzed to be
       completely inoperational. This is corrected and should work now.
    6) All commands that get sent to the SCSI adapter were verified and
       completed in such a way, that they are now completely conform to the
@@ -1386,7 +1386,7 @@
    concerning the Linux-kernel in special, this SCSI-driver comes without any
    warranty. Its functionality is tested as good as possible on certain 
    machines and combinations of computer hardware, which does not exclude,
-   that dataloss or severe damage of hardware is possible while using this
+   that data loss or severe damage of hardware is possible while using this
    part of software on some arbitrary computer hardware or in combination 
    with other software packages. It is highly recommended to make backup
    copies of your data before using this software. Furthermore, personal
diff --git a/Documentation/scsi/lpfc.txt b/Documentation/scsi/lpfc.txt
index 4dbe41370a6d..5741ea8aa88a 100644
--- a/Documentation/scsi/lpfc.txt
+++ b/Documentation/scsi/lpfc.txt
@@ -36,7 +36,7 @@ Cable pull and temporary device Loss:
   being removed, a switch rebooting, or a device reboot), the driver could
   hide the disappearance of the device from the midlayer. I/O's issued to
   the LLDD would simply be queued for a short duration, allowing the device
-  to reappear or link come back alive, with no inadvertant side effects
+  to reappear or link come back alive, with no inadvertent side effects
   to the system. If the driver did not hide these conditions, i/o would be
   errored by the driver, the mid-layer would exhaust its retries, and the
   device would be taken offline. Manual intervention would be required to
diff --git a/Documentation/scsi/scsi_fc_transport.txt b/Documentation/scsi/scsi_fc_transport.txt
index d403e46d8463..75143f0c23b6 100644
--- a/Documentation/scsi/scsi_fc_transport.txt
+++ b/Documentation/scsi/scsi_fc_transport.txt
@@ -65,7 +65,7 @@ Overview:
     discussion will concentrate on NPIV.
 
   Note: World Wide Name assignment (and uniqueness guarantees) are left
-    up to an administrative entity controling the vport. For example,
+    up to an administrative entity controlling the vport. For example,
     if vports are to be associated with virtual machines, a XEN mgmt
     utility would be responsible for creating wwpn/wwnn's for the vport,
     using it's own naming authority and OUI. (Note: it already does this
@@ -91,7 +91,7 @@ Device Trees and Vport Objects:
   Here's what to expect in the device tree :
    The typical Physical Port's Scsi_Host:
      /sys/devices/.../host17/
-   and it has the typical decendent tree:
+   and it has the typical descendant tree:
      /sys/devices/.../host17/rport-17:0-0/target17:0:0/17:0:0:0:
    and then the vport is created on the Physical Port:
      /sys/devices/.../host17/vport-17:0-0
@@ -192,7 +192,7 @@ Vport States:
       independent of the adapter's link state.
     - Instantiation of the vport on the FC link via ELS traffic, etc.
       This is equivalent to a "link up" and successfull link initialization.
-  Futher information can be found in the interfaces section below for
+  Further information can be found in the interfaces section below for
   Vport Creation.
 
   Once a vport has been instantiated with the kernel/LLDD, a vport state
diff --git a/Documentation/sh/clk.txt b/Documentation/sh/clk.txt
index 9aef710e9a4b..114b595cfa97 100644
--- a/Documentation/sh/clk.txt
+++ b/Documentation/sh/clk.txt
@@ -12,7 +12,7 @@ means no changes to adjanced clock
 Internally, the clk_set_rate_ex forwards request to clk->ops->set_rate method,
 if it is present in ops structure. The method should set the clock rate and adjust
 all needed clocks according to the passed algo_id.
-Exact values for algo_id are machine-dependend. For the sh7722, the following
+Exact values for algo_id are machine-dependent. For the sh7722, the following
 values are defined:
 
 	NO_CHANGE	= 0,
diff --git a/Documentation/sound/alsa/Audiophile-Usb.txt b/Documentation/sound/alsa/Audiophile-Usb.txt
index 2ad5e6306c44..a4c53d8961e1 100644
--- a/Documentation/sound/alsa/Audiophile-Usb.txt
+++ b/Documentation/sound/alsa/Audiophile-Usb.txt
@@ -236,15 +236,15 @@ The parameter can be given:
        alias snd-card-1 snd-usb-audio
        options snd-usb-audio index=1 device_setup=0x09
 
-CAUTION when initializaing the device
+CAUTION when initializing the device
 -------------------------------------
 
  * Correct initialization on the device requires that device_setup is given to
    the module BEFORE the device is turned on. So, if you use the "manual probing"
    method described above, take care to power-on the device AFTER this initialization.
 
- * Failing to respect this will lead in a misconfiguration of the device. In this case
-   turn off the device, unproble the snd-usb-audio module, then probe it again with 
+ * Failing to respect this will lead to a misconfiguration of the device. In this case
+   turn off the device, unprobe the snd-usb-audio module, then probe it again with
    correct device_setup parameter and then (and only then) turn on the device again.
 
  * If you've correctly initialized the device in a valid mode and then want to switch
@@ -388,9 +388,9 @@ There are 2 main potential issues when using Jackd with the device:
 
 Jack supports big endian devices only in recent versions (thanks to
 Andreas Steinmetz for his first big-endian patch). I can't remember 
-extacly when this support was released into jackd, let's just say that 
+exactly when this support was released into jackd, let's just say that
 with jackd version 0.103.0 it's almost ok (just a small bug is affecting 
-16bits Big-Endian devices, but since you've read  carefully the above 
+16bits Big-Endian devices, but since you've read carefully the above
 paragraphs, you're now using kernel >= 2.6.23 and your 16bits devices 
 are now Little Endians ;-) ).
 
diff --git a/Documentation/sound/alsa/hda_codec.txt b/Documentation/sound/alsa/hda_codec.txt
index 8e1b02526698..34e87ec1379c 100644
--- a/Documentation/sound/alsa/hda_codec.txt
+++ b/Documentation/sound/alsa/hda_codec.txt
@@ -67,7 +67,7 @@ CONFIG_SND_HDA_POWER_SAVE kconfig.  It's called when the codec needs
 to power up or may power down.  The controller should check the all
 belonging codecs on the bus whether they are actually powered off
 (check codec->power_on), and optionally the driver may power down the
-contoller side, too.
+controller side, too.
 
 The bus instance is created via snd_hda_bus_new().  You need to pass
 the card instance, the template, and the pointer to store the
diff --git a/Documentation/sound/alsa/soc/dapm.txt b/Documentation/sound/alsa/soc/dapm.txt
index c784a18b94dc..b2ed6983f40d 100644
--- a/Documentation/sound/alsa/soc/dapm.txt
+++ b/Documentation/sound/alsa/soc/dapm.txt
@@ -68,7 +68,7 @@ Audio DAPM widgets fall into a number of types:-
 (Widgets are defined in include/sound/soc-dapm.h)
 
 Widgets are usually added in the codec driver and the machine driver. There are
-convience macros defined in soc-dapm.h that can be used to quickly build a
+convenience macros defined in soc-dapm.h that can be used to quickly build a
 list of widgets of the codecs and machines DAPM widgets.
 
 Most widgets have a name, register, shift and invert. Some widgets have extra
diff --git a/Documentation/sysctl/vm.txt b/Documentation/sysctl/vm.txt
index 8a4863c4edd4..d79eeda7a699 100644
--- a/Documentation/sysctl/vm.txt
+++ b/Documentation/sysctl/vm.txt
@@ -116,7 +116,7 @@ of kilobytes free.  The VM uses this number to compute a pages_min
 value for each lowmem zone in the system.  Each lowmem zone gets 
 a number of reserved free pages based proportionally on its size.
 
-Some minimal ammount of memory is needed to satisfy PF_MEMALLOC
+Some minimal amount of memory is needed to satisfy PF_MEMALLOC
 allocations; if you set this to lower than 1024KB, your system will
 become subtly broken, and prone to deadlock under high loads.
 
diff --git a/Documentation/timers/highres.txt b/Documentation/timers/highres.txt
index a73ecf5b4bdb..21332233cef1 100644
--- a/Documentation/timers/highres.txt
+++ b/Documentation/timers/highres.txt
@@ -125,7 +125,7 @@ increase of flexibility and the avoidance of duplicated code across
 architectures justifies the slight increase of the binary size.
 
 The conversion of an architecture has no functional impact, but allows to
-utilize the high resolution and dynamic tick functionalites without any change
+utilize the high resolution and dynamic tick functionalities without any change
 to the clock event device and timer interrupt code. After the conversion the
 enabling of high resolution timers and dynamic ticks is simply provided by
 adding the kernel/time/Kconfig file to the architecture specific Kconfig and
diff --git a/Documentation/usb/authorization.txt b/Documentation/usb/authorization.txt
index 2af400609498..381b22ee7834 100644
--- a/Documentation/usb/authorization.txt
+++ b/Documentation/usb/authorization.txt
@@ -8,7 +8,7 @@ not) in a system. This feature will allow you to implement a lock-down
 of USB devices, fully controlled by user space.
 
 As of now, when a USB device is connected it is configured and
-it's interfaces inmediately made available to the users. With this
+its interfaces are immediately made available to the users.  With this
 modification, only if root authorizes the device to be configured will
 then it be possible to use it.
 
diff --git a/Documentation/video4linux/sn9c102.txt b/Documentation/video4linux/sn9c102.txt
index b26f5195af51..73de4050d637 100644
--- a/Documentation/video4linux/sn9c102.txt
+++ b/Documentation/video4linux/sn9c102.txt
@@ -157,7 +157,7 @@ Loading can be done as shown below:
 
 	[root@localhost home]# modprobe sn9c102
 
-Note that the module is called "sn9c102" for historic reasons, althought it
+Note that the module is called "sn9c102" for historic reasons, although it
 does not just support the SN9C102.
 
 At this point all the devices supported by the driver and connected to the USB
diff --git a/Documentation/vm/hugetlbpage.txt b/Documentation/vm/hugetlbpage.txt
index 8a5b5763f0fe..ea8714fcc3ad 100644
--- a/Documentation/vm/hugetlbpage.txt
+++ b/Documentation/vm/hugetlbpage.txt
@@ -77,7 +77,7 @@ memory that is preset in system at this time.  System administrators may want
 to put this command in one of the local rc init files.  This will enable the
 kernel to request huge pages early in the boot process (when the possibility
 of getting physical contiguous pages is still very high). In either
-case, adminstrators will want to verify the number of hugepages actually
+case, administrators will want to verify the number of hugepages actually
 allocated by checking the sysctl or meminfo.
 
 /proc/sys/vm/nr_overcommit_hugepages indicates how large the pool of
diff --git a/Documentation/vm/numa_memory_policy.txt b/Documentation/vm/numa_memory_policy.txt
index bad16d3f6a47..6aaaeb38730c 100644
--- a/Documentation/vm/numa_memory_policy.txt
+++ b/Documentation/vm/numa_memory_policy.txt
@@ -58,7 +58,7 @@ most general to most specific:
 	the policy at the time they were allocated.
 
     VMA Policy:  A "VMA" or "Virtual Memory Area" refers to a range of a task's
-    virtual adddress space.  A task may define a specific policy for a range
+    virtual address space.  A task may define a specific policy for a range
     of its virtual address space.   See the MEMORY POLICIES APIS section,
     below, for an overview of the mbind() system call used to set a VMA
     policy.
@@ -353,7 +353,7 @@ follows:
 
    Because of this extra reference counting, and because we must lookup
    shared policies in a tree structure under spinlock, shared policies are
-   more expensive to use in the page allocation path.  This is expecially
+   more expensive to use in the page allocation path.  This is especially
    true for shared policies on shared memory regions shared by tasks running
    on different NUMA nodes.  This extra overhead can be avoided by always
    falling back to task or system default policy for shared memory regions,
diff --git a/Documentation/volatile-considered-harmful.txt b/Documentation/volatile-considered-harmful.txt
index 10c2e411cca8..991c26a6ef64 100644
--- a/Documentation/volatile-considered-harmful.txt
+++ b/Documentation/volatile-considered-harmful.txt
@@ -114,6 +114,6 @@ CREDITS
 
 Original impetus and research by Randy Dunlap
 Written by Jonathan Corbet
-Improvements via coments from Satyam Sharma, Johannes Stezenbach, Jesper
+Improvements via comments from Satyam Sharma, Johannes Stezenbach, Jesper
 	Juhl, Heikki Orsila, H. Peter Anvin, Philipp Hahn, and Stefan
 	Richter.
diff --git a/drivers/message/fusion/lsi/mpi_history.txt b/drivers/message/fusion/lsi/mpi_history.txt
index 241592ab13ad..3f15fcfe4a2e 100644
--- a/drivers/message/fusion/lsi/mpi_history.txt
+++ b/drivers/message/fusion/lsi/mpi_history.txt
@@ -127,7 +127,7 @@ mpi_ioc.h
  *  08-08-01  01.02.01  Original release for v1.2 work.
  *                      New format for FWVersion and ProductId in
  *                      MSG_IOC_FACTS_REPLY and MPI_FW_HEADER.
- *  08-31-01  01.02.02  Addded event MPI_EVENT_SCSI_DEVICE_STATUS_CHANGE and
+ *  08-31-01  01.02.02  Added event MPI_EVENT_SCSI_DEVICE_STATUS_CHANGE and
  *                      related structure and defines.
  *                      Added event MPI_EVENT_ON_BUS_TIMER_EXPIRED.
  *                      Added MPI_IOCINIT_FLAGS_DISCARD_FW_IMAGE.
@@ -187,7 +187,7 @@ mpi_ioc.h
  *  10-11-06  01.05.12  Added MPI_IOCFACTS_EXCEPT_METADATA_UNSUPPORTED.
  *                      Added MaxInitiators field to PortFacts reply.
  *                      Added SAS Device Status Change ReasonCode for
- *                      asynchronous notificaiton.
+ *                      asynchronous notification.
  *                      Added MPI_EVENT_SAS_EXPANDER_STATUS_CHANGE and event
  *                      data structure.
  *                      Added new ImageType values for FWDownload and FWUpload
@@ -213,7 +213,7 @@ mpi_cnfg.h
  *                      Added _RESPONSE_ID_MASK definition to SCSI_PORT_1
  *                      page and updated the page version.
  *                      Added Information field and _INFO_PARAMS_NEGOTIATED
- *                      definitionto SCSI_DEVICE_0 page.
+ *                      definition to SCSI_DEVICE_0 page.
  *  06-22-00  01.00.03  Removed batch controls from LAN_0 page and updated the
  *                      page version.
  *                      Added BucketsRemaining to LAN_1 page, redefined the

From 51cc50685a4275c6a02653670af9f108a64e01cf Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Fri, 25 Jul 2008 19:45:34 -0700
Subject: [PATCH 1037/1171] SL*B: drop kmem cache argument from constructor

Kmem cache passed to constructor is only needed for constructors that are
themselves multiplexeres.  Nobody uses this "feature", nor does anybody uses
passed kmem cache in non-trivial way, so pass only pointer to object.

Non-trivial places are:
	arch/powerpc/mm/init_64.c
	arch/powerpc/mm/hugetlbpage.c

This is flag day, yes.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Acked-by: Pekka Enberg <penberg@cs.helsinki.fi>
Acked-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Jon Tollefson <kniht@linux.vnet.ibm.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Matt Mackall <mpm@selenic.com>
[akpm@linux-foundation.org: fix arch/powerpc/mm/hugetlbpage.c]
[akpm@linux-foundation.org: fix mm/slab.c]
[akpm@linux-foundation.org: fix ubifs]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/arm/plat-s3c24xx/dma.c               |  2 +-
 arch/powerpc/kernel/rtas_flash.c          |  2 +-
 arch/powerpc/mm/hugetlbpage.c             |  9 ++-------
 arch/powerpc/mm/init_64.c                 | 24 +++++++++--------------
 arch/powerpc/platforms/cell/spufs/inode.c |  2 +-
 arch/sh/mm/pmb.c                          |  2 +-
 arch/xtensa/mm/init.c                     |  2 +-
 drivers/usb/mon/mon_text.c                |  4 ++--
 fs/adfs/super.c                           |  2 +-
 fs/affs/super.c                           |  2 +-
 fs/afs/super.c                            |  4 ++--
 fs/befs/linuxvfs.c                        |  2 +-
 fs/bfs/inode.c                            |  2 +-
 fs/block_dev.c                            |  2 +-
 fs/buffer.c                               |  2 +-
 fs/cifs/cifsfs.c                          |  2 +-
 fs/coda/inode.c                           |  2 +-
 fs/ecryptfs/main.c                        |  4 ++--
 fs/efs/super.c                            |  2 +-
 fs/ext2/super.c                           |  2 +-
 fs/ext3/super.c                           |  2 +-
 fs/ext4/super.c                           |  2 +-
 fs/fat/cache.c                            |  2 +-
 fs/fat/inode.c                            |  2 +-
 fs/fuse/inode.c                           |  2 +-
 fs/gfs2/main.c                            |  4 ++--
 fs/hfs/super.c                            |  2 +-
 fs/hfsplus/super.c                        |  2 +-
 fs/hpfs/super.c                           |  2 +-
 fs/hugetlbfs/inode.c                      |  2 +-
 fs/inode.c                                |  2 +-
 fs/isofs/inode.c                          |  2 +-
 fs/jffs2/super.c                          |  2 +-
 fs/jfs/jfs_metapage.c                     |  2 +-
 fs/jfs/super.c                            |  2 +-
 fs/locks.c                                |  2 +-
 fs/minix/inode.c                          |  2 +-
 fs/ncpfs/inode.c                          |  2 +-
 fs/nfs/inode.c                            |  2 +-
 fs/ntfs/super.c                           |  2 +-
 fs/ocfs2/dlm/dlmfs.c                      |  3 +--
 fs/ocfs2/super.c                          |  2 +-
 fs/openpromfs/inode.c                     |  2 +-
 fs/proc/inode.c                           |  2 +-
 fs/qnx4/inode.c                           |  2 +-
 fs/reiserfs/super.c                       |  2 +-
 fs/romfs/inode.c                          |  2 +-
 fs/smbfs/inode.c                          |  2 +-
 fs/sysv/inode.c                           |  2 +-
 fs/ubifs/super.c                          |  2 +-
 fs/udf/super.c                            |  2 +-
 fs/ufs/super.c                            |  2 +-
 fs/xfs/linux-2.6/kmem.h                   |  2 +-
 fs/xfs/linux-2.6/xfs_super.c              |  1 -
 include/linux/slab.h                      |  2 +-
 include/linux/slub_def.h                  |  2 +-
 ipc/mqueue.c                              |  2 +-
 kernel/fork.c                             |  2 +-
 lib/idr.c                                 |  2 +-
 lib/radix-tree.c                          |  2 +-
 mm/rmap.c                                 |  2 +-
 mm/shmem.c                                |  2 +-
 mm/slab.c                                 | 11 +++++------
 mm/slob.c                                 |  7 +++----
 mm/slub.c                                 | 13 ++++++------
 net/socket.c                              |  2 +-
 net/sunrpc/rpc_pipe.c                     |  2 +-
 67 files changed, 90 insertions(+), 106 deletions(-)

diff --git a/arch/arm/plat-s3c24xx/dma.c b/arch/arm/plat-s3c24xx/dma.c
index 60f162dc4fad..8c5e656d5d8c 100644
--- a/arch/arm/plat-s3c24xx/dma.c
+++ b/arch/arm/plat-s3c24xx/dma.c
@@ -1304,7 +1304,7 @@ struct sysdev_class dma_sysclass = {
 
 /* kmem cache implementation */
 
-static void s3c2410_dma_cache_ctor(struct kmem_cache *c, void *p)
+static void s3c2410_dma_cache_ctor(void *p)
 {
 	memset(p, 0, sizeof(struct s3c2410_dma_buf));
 }
diff --git a/arch/powerpc/kernel/rtas_flash.c b/arch/powerpc/kernel/rtas_flash.c
index 09ded5c424a9..149cb112cd1a 100644
--- a/arch/powerpc/kernel/rtas_flash.c
+++ b/arch/powerpc/kernel/rtas_flash.c
@@ -286,7 +286,7 @@ static ssize_t rtas_flash_read(struct file *file, char __user *buf,
 }
 
 /* constructor for flash_block_cache */
-void rtas_block_ctor(struct kmem_cache *cache, void *ptr)
+void rtas_block_ctor(void *ptr)
 {
 	memset(ptr, 0, RTAS_BLK_SIZE);
 }
diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c
index fb42c4dd3217..ed0aab0208a6 100644
--- a/arch/powerpc/mm/hugetlbpage.c
+++ b/arch/powerpc/mm/hugetlbpage.c
@@ -113,7 +113,7 @@ static inline pte_t *hugepte_offset(hugepd_t *hpdp, unsigned long addr,
 static int __hugepte_alloc(struct mm_struct *mm, hugepd_t *hpdp,
 			   unsigned long address, unsigned int psize)
 {
-	pte_t *new = kmem_cache_alloc(huge_pgtable_cache(psize),
+	pte_t *new = kmem_cache_zalloc(huge_pgtable_cache(psize),
 				      GFP_KERNEL|__GFP_REPEAT);
 
 	if (! new)
@@ -730,11 +730,6 @@ static int __init hugepage_setup_sz(char *str)
 }
 __setup("hugepagesz=", hugepage_setup_sz);
 
-static void zero_ctor(struct kmem_cache *cache, void *addr)
-{
-	memset(addr, 0, kmem_cache_size(cache));
-}
-
 static int __init hugetlbpage_init(void)
 {
 	unsigned int psize;
@@ -756,7 +751,7 @@ static int __init hugetlbpage_init(void)
 						HUGEPTE_TABLE_SIZE(psize),
 						HUGEPTE_TABLE_SIZE(psize),
 						0,
-						zero_ctor);
+						NULL);
 			if (!huge_pgtable_cache(psize))
 				panic("hugetlbpage_init(): could not create %s"\
 				      "\n", HUGEPTE_CACHE_NAME(psize));
diff --git a/arch/powerpc/mm/init_64.c b/arch/powerpc/mm/init_64.c
index a41bc5aa2043..4f7df85129d8 100644
--- a/arch/powerpc/mm/init_64.c
+++ b/arch/powerpc/mm/init_64.c
@@ -136,9 +136,14 @@ static int __init setup_kcore(void)
 module_init(setup_kcore);
 #endif
 
-static void zero_ctor(struct kmem_cache *cache, void *addr)
+static void pgd_ctor(void *addr)
 {
-	memset(addr, 0, kmem_cache_size(cache));
+	memset(addr, 0, PGD_TABLE_SIZE);
+}
+
+static void pmd_ctor(void *addr)
+{
+	memset(addr, 0, PMD_TABLE_SIZE);
 }
 
 static const unsigned int pgtable_cache_size[2] = {
@@ -163,19 +168,8 @@ struct kmem_cache *pgtable_cache[ARRAY_SIZE(pgtable_cache_size)];
 
 void pgtable_cache_init(void)
 {
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(pgtable_cache_size); i++) {
-		int size = pgtable_cache_size[i];
-		const char *name = pgtable_cache_name[i];
-
-		pr_debug("Allocating page table cache %s (#%d) "
-			"for size: %08x...\n", name, i, size);
-		pgtable_cache[i] = kmem_cache_create(name,
-						     size, size,
-						     SLAB_PANIC,
-						     zero_ctor);
-	}
+	pgtable_cache[0] = kmem_cache_create(pgtable_cache_name[0], PGD_TABLE_SIZE, PGD_TABLE_SIZE, SLAB_PANIC, pgd_ctor);
+	pgtable_cache[1] = kmem_cache_create(pgtable_cache_name[1], PMD_TABLE_SIZE, PMD_TABLE_SIZE, SLAB_PANIC, pmd_ctor);
 }
 
 #ifdef CONFIG_SPARSEMEM_VMEMMAP
diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c
index 7123472801d9..690ca7b0dcf6 100644
--- a/arch/powerpc/platforms/cell/spufs/inode.c
+++ b/arch/powerpc/platforms/cell/spufs/inode.c
@@ -78,7 +78,7 @@ spufs_destroy_inode(struct inode *inode)
 }
 
 static void
-spufs_init_once(struct kmem_cache *cachep, void *p)
+spufs_init_once(void *p)
 {
 	struct spufs_inode_info *ei = p;
 
diff --git a/arch/sh/mm/pmb.c b/arch/sh/mm/pmb.c
index 0b0ec6e04753..46911bcbf17b 100644
--- a/arch/sh/mm/pmb.c
+++ b/arch/sh/mm/pmb.c
@@ -293,7 +293,7 @@ void pmb_unmap(unsigned long addr)
 	} while (pmbe);
 }
 
-static void pmb_cache_ctor(struct kmem_cache *cachep, void *pmb)
+static void pmb_cache_ctor(void *pmb)
 {
 	struct pmb_entry *pmbe = pmb;
 
diff --git a/arch/xtensa/mm/init.c b/arch/xtensa/mm/init.c
index 81d0560eaea2..ee261005b363 100644
--- a/arch/xtensa/mm/init.c
+++ b/arch/xtensa/mm/init.c
@@ -309,7 +309,7 @@ void show_mem(void)
 
 struct kmem_cache *pgtable_cache __read_mostly;
 
-static void pgd_ctor(struct kmem_cache *cache, void* addr)
+static void pgd_ctor(void* addr)
 {
 	pte_t* ptep = (pte_t*)addr;
 	int i;
diff --git a/drivers/usb/mon/mon_text.c b/drivers/usb/mon/mon_text.c
index 5e3e4e9b6c77..1f715436d6d3 100644
--- a/drivers/usb/mon/mon_text.c
+++ b/drivers/usb/mon/mon_text.c
@@ -87,7 +87,7 @@ struct mon_reader_text {
 
 static struct dentry *mon_dir;		/* Usually /sys/kernel/debug/usbmon */
 
-static void mon_text_ctor(struct kmem_cache *, void *);
+static void mon_text_ctor(void *);
 
 struct mon_text_ptr {
 	int cnt, limit;
@@ -720,7 +720,7 @@ void mon_text_del(struct mon_bus *mbus)
 /*
  * Slab interface: constructor.
  */
-static void mon_text_ctor(struct kmem_cache *slab, void *mem)
+static void mon_text_ctor(void *mem)
 {
 	/*
 	 * Nothing to initialize. No, really!
diff --git a/fs/adfs/super.c b/fs/adfs/super.c
index 9e421eeb672b..26f3b43726bb 100644
--- a/fs/adfs/super.c
+++ b/fs/adfs/super.c
@@ -249,7 +249,7 @@ static void adfs_destroy_inode(struct inode *inode)
 	kmem_cache_free(adfs_inode_cachep, ADFS_I(inode));
 }
 
-static void init_once(struct kmem_cache *cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct adfs_inode_info *ei = (struct adfs_inode_info *) foo;
 
diff --git a/fs/affs/super.c b/fs/affs/super.c
index 4e0309566406..3a89094f93d0 100644
--- a/fs/affs/super.c
+++ b/fs/affs/super.c
@@ -90,7 +90,7 @@ static void affs_destroy_inode(struct inode *inode)
 	kmem_cache_free(affs_inode_cachep, AFFS_I(inode));
 }
 
-static void init_once(struct kmem_cache *cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct affs_inode_info *ei = (struct affs_inode_info *) foo;
 
diff --git a/fs/afs/super.c b/fs/afs/super.c
index 7e3faeef6818..250d8c4d66e4 100644
--- a/fs/afs/super.c
+++ b/fs/afs/super.c
@@ -27,7 +27,7 @@
 
 #define AFS_FS_MAGIC 0x6B414653 /* 'kAFS' */
 
-static void afs_i_init_once(struct kmem_cache *cachep, void *foo);
+static void afs_i_init_once(void *foo);
 static int afs_get_sb(struct file_system_type *fs_type,
 		      int flags, const char *dev_name,
 		      void *data, struct vfsmount *mnt);
@@ -449,7 +449,7 @@ static void afs_put_super(struct super_block *sb)
 /*
  * initialise an inode cache slab element prior to any use
  */
-static void afs_i_init_once(struct kmem_cache *cachep, void *_vnode)
+static void afs_i_init_once(void *_vnode)
 {
 	struct afs_vnode *vnode = _vnode;
 
diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c
index e8717de3bab3..02c6e62b72f8 100644
--- a/fs/befs/linuxvfs.c
+++ b/fs/befs/linuxvfs.c
@@ -289,7 +289,7 @@ befs_destroy_inode(struct inode *inode)
         kmem_cache_free(befs_inode_cachep, BEFS_I(inode));
 }
 
-static void init_once(struct kmem_cache *cachep, void *foo)
+static void init_once(void *foo)
 {
         struct befs_inode_info *bi = (struct befs_inode_info *) foo;
 
diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c
index 053e690ec9ed..0ed57b5ee012 100644
--- a/fs/bfs/inode.c
+++ b/fs/bfs/inode.c
@@ -264,7 +264,7 @@ static void bfs_destroy_inode(struct inode *inode)
 	kmem_cache_free(bfs_inode_cachep, BFS_I(inode));
 }
 
-static void init_once(struct kmem_cache *cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct bfs_inode_info *bi = foo;
 
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 10d8a0aa871a..dcf37cada369 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -271,7 +271,7 @@ static void bdev_destroy_inode(struct inode *inode)
 	kmem_cache_free(bdev_cachep, bdi);
 }
 
-static void init_once(struct kmem_cache * cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct bdev_inode *ei = (struct bdev_inode *) foo;
 	struct block_device *bdev = &ei->bdev;
diff --git a/fs/buffer.c b/fs/buffer.c
index 109b261192d9..5fd497cdd6f3 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -3272,7 +3272,7 @@ int bh_submit_read(struct buffer_head *bh)
 EXPORT_SYMBOL(bh_submit_read);
 
 static void
-init_buffer_head(struct kmem_cache *cachep, void *data)
+init_buffer_head(void *data)
 {
 	struct buffer_head *bh = data;
 
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 22857c639df5..fe5f6809cba6 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -766,7 +766,7 @@ const struct file_operations cifs_dir_ops = {
 };
 
 static void
-cifs_init_once(struct kmem_cache *cachep, void *inode)
+cifs_init_once(void *inode)
 {
 	struct cifsInodeInfo *cifsi = inode;
 
diff --git a/fs/coda/inode.c b/fs/coda/inode.c
index 2f58dfc70083..830f51abb971 100644
--- a/fs/coda/inode.c
+++ b/fs/coda/inode.c
@@ -58,7 +58,7 @@ static void coda_destroy_inode(struct inode *inode)
 	kmem_cache_free(coda_inode_cachep, ITOC(inode));
 }
 
-static void init_once(struct kmem_cache * cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct coda_inode_info *ei = (struct coda_inode_info *) foo;
 
diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c
index 6f403cfba14f..448dfd597b5f 100644
--- a/fs/ecryptfs/main.c
+++ b/fs/ecryptfs/main.c
@@ -578,7 +578,7 @@ static struct file_system_type ecryptfs_fs_type = {
  * Initializes the ecryptfs_inode_info_cache when it is created
  */
 static void
-inode_info_init_once(struct kmem_cache *cachep, void *vptr)
+inode_info_init_once(void *vptr)
 {
 	struct ecryptfs_inode_info *ei = (struct ecryptfs_inode_info *)vptr;
 
@@ -589,7 +589,7 @@ static struct ecryptfs_cache_info {
 	struct kmem_cache **cache;
 	const char *name;
 	size_t size;
-	void (*ctor)(struct kmem_cache *cache, void *obj);
+	void (*ctor)(void *obj);
 } ecryptfs_cache_infos[] = {
 	{
 		.cache = &ecryptfs_auth_tok_list_item_cache,
diff --git a/fs/efs/super.c b/fs/efs/super.c
index d733531b55e2..567b134fa1f1 100644
--- a/fs/efs/super.c
+++ b/fs/efs/super.c
@@ -70,7 +70,7 @@ static void efs_destroy_inode(struct inode *inode)
 	kmem_cache_free(efs_inode_cachep, INODE_INFO(inode));
 }
 
-static void init_once(struct kmem_cache *cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct efs_inode_info *ei = (struct efs_inode_info *) foo;
 
diff --git a/fs/ext2/super.c b/fs/ext2/super.c
index 31308a3b0b8b..fd88c7b43e66 100644
--- a/fs/ext2/super.c
+++ b/fs/ext2/super.c
@@ -159,7 +159,7 @@ static void ext2_destroy_inode(struct inode *inode)
 	kmem_cache_free(ext2_inode_cachep, EXT2_I(inode));
 }
 
-static void init_once(struct kmem_cache * cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct ext2_inode_info *ei = (struct ext2_inode_info *) foo;
 
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 615788c6843a..8ddced384674 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -472,7 +472,7 @@ static void ext3_destroy_inode(struct inode *inode)
 	kmem_cache_free(ext3_inode_cachep, EXT3_I(inode));
 }
 
-static void init_once(struct kmem_cache * cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct ext3_inode_info *ei = (struct ext3_inode_info *) foo;
 
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 1cb371dcd609..b5479b1dff14 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -595,7 +595,7 @@ static void ext4_destroy_inode(struct inode *inode)
 	kmem_cache_free(ext4_inode_cachep, EXT4_I(inode));
 }
 
-static void init_once(struct kmem_cache *cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct ext4_inode_info *ei = (struct ext4_inode_info *) foo;
 
diff --git a/fs/fat/cache.c b/fs/fat/cache.c
index 3a9ecac8d61f..3222f51c41cf 100644
--- a/fs/fat/cache.c
+++ b/fs/fat/cache.c
@@ -36,7 +36,7 @@ static inline int fat_max_cache(struct inode *inode)
 
 static struct kmem_cache *fat_cache_cachep;
 
-static void init_once(struct kmem_cache *cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct fat_cache *cache = (struct fat_cache *)foo;
 
diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index 23676f9d79ce..6d266d793e2c 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -498,7 +498,7 @@ static void fat_destroy_inode(struct inode *inode)
 	kmem_cache_free(fat_inode_cachep, MSDOS_I(inode));
 }
 
-static void init_once(struct kmem_cache *cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct msdos_inode_info *ei = (struct msdos_inode_info *)foo;
 
diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index 7d2f7d6e22e2..d2249f174e20 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -956,7 +956,7 @@ static inline void unregister_fuseblk(void)
 }
 #endif
 
-static void fuse_inode_init_once(struct kmem_cache *cachep, void *foo)
+static void fuse_inode_init_once(void *foo)
 {
 	struct inode * inode = foo;
 
diff --git a/fs/gfs2/main.c b/fs/gfs2/main.c
index bcc668d0fadd..bb2cc303ac29 100644
--- a/fs/gfs2/main.c
+++ b/fs/gfs2/main.c
@@ -24,7 +24,7 @@
 #include "util.h"
 #include "glock.h"
 
-static void gfs2_init_inode_once(struct kmem_cache *cachep, void *foo)
+static void gfs2_init_inode_once(void *foo)
 {
 	struct gfs2_inode *ip = foo;
 
@@ -33,7 +33,7 @@ static void gfs2_init_inode_once(struct kmem_cache *cachep, void *foo)
 	ip->i_alloc = NULL;
 }
 
-static void gfs2_init_glock_once(struct kmem_cache *cachep, void *foo)
+static void gfs2_init_glock_once(void *foo)
 {
 	struct gfs2_glock *gl = foo;
 
diff --git a/fs/hfs/super.c b/fs/hfs/super.c
index ac2ec5ef66e4..4abb1047c689 100644
--- a/fs/hfs/super.c
+++ b/fs/hfs/super.c
@@ -432,7 +432,7 @@ static struct file_system_type hfs_fs_type = {
 	.fs_flags	= FS_REQUIRES_DEV,
 };
 
-static void hfs_init_once(struct kmem_cache *cachep, void *p)
+static void hfs_init_once(void *p)
 {
 	struct hfs_inode_info *i = p;
 
diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c
index 3859118531c7..e834e578c93f 100644
--- a/fs/hfsplus/super.c
+++ b/fs/hfsplus/super.c
@@ -485,7 +485,7 @@ static struct file_system_type hfsplus_fs_type = {
 	.fs_flags	= FS_REQUIRES_DEV,
 };
 
-static void hfsplus_init_once(struct kmem_cache *cachep, void *p)
+static void hfsplus_init_once(void *p)
 {
 	struct hfsplus_inode_info *i = p;
 
diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c
index f63a699ec659..b8ae9c90ada0 100644
--- a/fs/hpfs/super.c
+++ b/fs/hpfs/super.c
@@ -173,7 +173,7 @@ static void hpfs_destroy_inode(struct inode *inode)
 	kmem_cache_free(hpfs_inode_cachep, hpfs_i(inode));
 }
 
-static void init_once(struct kmem_cache *cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct hpfs_inode_info *ei = (struct hpfs_inode_info *) foo;
 
diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index dbd01d262ca4..3f58923fb39b 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -705,7 +705,7 @@ static const struct address_space_operations hugetlbfs_aops = {
 };
 
 
-static void init_once(struct kmem_cache *cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct hugetlbfs_inode_info *ei = (struct hugetlbfs_inode_info *)foo;
 
diff --git a/fs/inode.c b/fs/inode.c
index 35b6414522ea..b6726f644530 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -224,7 +224,7 @@ void inode_init_once(struct inode *inode)
 
 EXPORT_SYMBOL(inode_init_once);
 
-static void init_once(struct kmem_cache * cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct inode * inode = (struct inode *) foo;
 
diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c
index 044a254d526b..26948a6033b6 100644
--- a/fs/isofs/inode.c
+++ b/fs/isofs/inode.c
@@ -73,7 +73,7 @@ static void isofs_destroy_inode(struct inode *inode)
 	kmem_cache_free(isofs_inode_cachep, ISOFS_I(inode));
 }
 
-static void init_once(struct kmem_cache *cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct iso_inode_info *ei = foo;
 
diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c
index 7da69eae49e4..efd401257ed9 100644
--- a/fs/jffs2/super.c
+++ b/fs/jffs2/super.c
@@ -44,7 +44,7 @@ static void jffs2_destroy_inode(struct inode *inode)
 	kmem_cache_free(jffs2_inode_cachep, JFFS2_INODE_INFO(inode));
 }
 
-static void jffs2_i_init_once(struct kmem_cache *cachep, void *foo)
+static void jffs2_i_init_once(void *foo)
 {
 	struct jffs2_inode_info *f = foo;
 
diff --git a/fs/jfs/jfs_metapage.c b/fs/jfs/jfs_metapage.c
index 854ff0ec574f..c350057087dd 100644
--- a/fs/jfs/jfs_metapage.c
+++ b/fs/jfs/jfs_metapage.c
@@ -182,7 +182,7 @@ static inline void remove_metapage(struct page *page, struct metapage *mp)
 
 #endif
 
-static void init_once(struct kmem_cache *cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct metapage *mp = (struct metapage *)foo;
 
diff --git a/fs/jfs/super.c b/fs/jfs/super.c
index 359c091d8965..3630718be395 100644
--- a/fs/jfs/super.c
+++ b/fs/jfs/super.c
@@ -760,7 +760,7 @@ static struct file_system_type jfs_fs_type = {
 	.fs_flags	= FS_REQUIRES_DEV,
 };
 
-static void init_once(struct kmem_cache *cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct jfs_inode_info *jfs_ip = (struct jfs_inode_info *) foo;
 
diff --git a/fs/locks.c b/fs/locks.c
index 01490300f7cb..5eb259e3cd38 100644
--- a/fs/locks.c
+++ b/fs/locks.c
@@ -201,7 +201,7 @@ EXPORT_SYMBOL(locks_init_lock);
  * Initialises the fields of the file lock which are invariant for
  * free file_locks.
  */
-static void init_once(struct kmem_cache *cache, void *foo)
+static void init_once(void *foo)
 {
 	struct file_lock *lock = (struct file_lock *) foo;
 
diff --git a/fs/minix/inode.c b/fs/minix/inode.c
index 523d73713418..d1d1eb84679d 100644
--- a/fs/minix/inode.c
+++ b/fs/minix/inode.c
@@ -68,7 +68,7 @@ static void minix_destroy_inode(struct inode *inode)
 	kmem_cache_free(minix_inode_cachep, minix_i(inode));
 }
 
-static void init_once(struct kmem_cache * cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct minix_inode_info *ei = (struct minix_inode_info *) foo;
 
diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c
index 2e5ab1204dec..d642f0e5b365 100644
--- a/fs/ncpfs/inode.c
+++ b/fs/ncpfs/inode.c
@@ -64,7 +64,7 @@ static void ncp_destroy_inode(struct inode *inode)
 	kmem_cache_free(ncp_inode_cachep, NCP_FINFO(inode));
 }
 
-static void init_once(struct kmem_cache *cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct ncp_inode_info *ei = (struct ncp_inode_info *) foo;
 
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index df23f987da6b..52daefa2f521 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -1242,7 +1242,7 @@ static inline void nfs4_init_once(struct nfs_inode *nfsi)
 #endif
 }
 
-static void init_once(struct kmem_cache * cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct nfs_inode *nfsi = (struct nfs_inode *) foo;
 
diff --git a/fs/ntfs/super.c b/fs/ntfs/super.c
index 3e76f3b216bc..4a46743b5077 100644
--- a/fs/ntfs/super.c
+++ b/fs/ntfs/super.c
@@ -3080,7 +3080,7 @@ struct kmem_cache *ntfs_inode_cache;
 struct kmem_cache *ntfs_big_inode_cache;
 
 /* Init once constructor for the inode slab cache. */
-static void ntfs_big_inode_init_once(struct kmem_cache *cachep, void *foo)
+static void ntfs_big_inode_init_once(void *foo)
 {
 	ntfs_inode *ni = (ntfs_inode *)foo;
 
diff --git a/fs/ocfs2/dlm/dlmfs.c b/fs/ocfs2/dlm/dlmfs.c
index e48aba698b77..533a789c3ef8 100644
--- a/fs/ocfs2/dlm/dlmfs.c
+++ b/fs/ocfs2/dlm/dlmfs.c
@@ -267,8 +267,7 @@ static ssize_t dlmfs_file_write(struct file *filp,
 	return writelen;
 }
 
-static void dlmfs_init_once(struct kmem_cache *cachep,
-			    void *foo)
+static void dlmfs_init_once(void *foo)
 {
 	struct dlmfs_inode_private *ip =
 		(struct dlmfs_inode_private *) foo;
diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c
index ccecfe5094fa..2560b33889aa 100644
--- a/fs/ocfs2/super.c
+++ b/fs/ocfs2/super.c
@@ -1118,7 +1118,7 @@ bail:
 	return status;
 }
 
-static void ocfs2_inode_init_once(struct kmem_cache *cachep, void *data)
+static void ocfs2_inode_init_once(void *data)
 {
 	struct ocfs2_inode_info *oi = data;
 
diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c
index d17b4fd204e1..9f5b054f06b9 100644
--- a/fs/openpromfs/inode.c
+++ b/fs/openpromfs/inode.c
@@ -430,7 +430,7 @@ static struct file_system_type openprom_fs_type = {
 	.kill_sb	= kill_anon_super,
 };
 
-static void op_inode_init_once(struct kmem_cache * cachep, void *data)
+static void op_inode_init_once(void *data)
 {
 	struct op_inode_info *oi = (struct op_inode_info *) data;
 
diff --git a/fs/proc/inode.c b/fs/proc/inode.c
index 02eca2ed9dd7..b37f25dc45a5 100644
--- a/fs/proc/inode.c
+++ b/fs/proc/inode.c
@@ -94,7 +94,7 @@ static void proc_destroy_inode(struct inode *inode)
 	kmem_cache_free(proc_inode_cachep, PROC_I(inode));
 }
 
-static void init_once(struct kmem_cache * cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct proc_inode *ei = (struct proc_inode *) foo;
 
diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c
index b31ab78052b3..2aad1044b84c 100644
--- a/fs/qnx4/inode.c
+++ b/fs/qnx4/inode.c
@@ -553,7 +553,7 @@ static void qnx4_destroy_inode(struct inode *inode)
 	kmem_cache_free(qnx4_inode_cachep, qnx4_i(inode));
 }
 
-static void init_once(struct kmem_cache *cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct qnx4_inode_info *ei = (struct qnx4_inode_info *) foo;
 
diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c
index 2ec748ba0bd3..879e54d35c2d 100644
--- a/fs/reiserfs/super.c
+++ b/fs/reiserfs/super.c
@@ -521,7 +521,7 @@ static void reiserfs_destroy_inode(struct inode *inode)
 	kmem_cache_free(reiserfs_inode_cachep, REISERFS_I(inode));
 }
 
-static void init_once(struct kmem_cache * cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct reiserfs_inode_info *ei = (struct reiserfs_inode_info *)foo;
 
diff --git a/fs/romfs/inode.c b/fs/romfs/inode.c
index 3f13d491c7c7..8e51a2aaa977 100644
--- a/fs/romfs/inode.c
+++ b/fs/romfs/inode.c
@@ -577,7 +577,7 @@ static void romfs_destroy_inode(struct inode *inode)
 	kmem_cache_free(romfs_inode_cachep, ROMFS_I(inode));
 }
 
-static void init_once(struct kmem_cache *cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct romfs_inode_info *ei = foo;
 
diff --git a/fs/smbfs/inode.c b/fs/smbfs/inode.c
index 376ef3ee6ed7..3528f40ffb0f 100644
--- a/fs/smbfs/inode.c
+++ b/fs/smbfs/inode.c
@@ -67,7 +67,7 @@ static void smb_destroy_inode(struct inode *inode)
 	kmem_cache_free(smb_inode_cachep, SMB_I(inode));
 }
 
-static void init_once(struct kmem_cache *cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct smb_inode_info *ei = (struct smb_inode_info *) foo;
 
diff --git a/fs/sysv/inode.c b/fs/sysv/inode.c
index c5d60de0658f..df0d435baa48 100644
--- a/fs/sysv/inode.c
+++ b/fs/sysv/inode.c
@@ -326,7 +326,7 @@ static void sysv_destroy_inode(struct inode *inode)
 	kmem_cache_free(sysv_inode_cachep, SYSV_I(inode));
 }
 
-static void init_once(struct kmem_cache *cachep, void *p)
+static void init_once(void *p)
 {
 	struct sysv_inode_info *si = (struct sysv_inode_info *)p;
 
diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c
index 00eb9c68ad03..ca1e2d4e03cc 100644
--- a/fs/ubifs/super.c
+++ b/fs/ubifs/super.c
@@ -1841,7 +1841,7 @@ static struct file_system_type ubifs_fs_type = {
 /*
  * Inode slab cache constructor.
  */
-static void inode_slab_ctor(struct kmem_cache *cachep, void *obj)
+static void inode_slab_ctor(void *obj)
 {
 	struct ubifs_inode *ui = obj;
 	inode_init_once(&ui->vfs_inode);
diff --git a/fs/udf/super.c b/fs/udf/super.c
index 44cc702f96cc..5698bbf83bbf 100644
--- a/fs/udf/super.c
+++ b/fs/udf/super.c
@@ -148,7 +148,7 @@ static void udf_destroy_inode(struct inode *inode)
 	kmem_cache_free(udf_inode_cachep, UDF_I(inode));
 }
 
-static void init_once(struct kmem_cache *cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct udf_inode_info *ei = (struct udf_inode_info *)foo;
 
diff --git a/fs/ufs/super.c b/fs/ufs/super.c
index 227c9d700040..3e30e40aa24d 100644
--- a/fs/ufs/super.c
+++ b/fs/ufs/super.c
@@ -1302,7 +1302,7 @@ static void ufs_destroy_inode(struct inode *inode)
 	kmem_cache_free(ufs_inode_cachep, UFS_I(inode));
 }
 
-static void init_once(struct kmem_cache * cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct ufs_inode_info *ei = (struct ufs_inode_info *) foo;
 
diff --git a/fs/xfs/linux-2.6/kmem.h b/fs/xfs/linux-2.6/kmem.h
index 5e9564902976..a20683cf74dd 100644
--- a/fs/xfs/linux-2.6/kmem.h
+++ b/fs/xfs/linux-2.6/kmem.h
@@ -79,7 +79,7 @@ kmem_zone_init(int size, char *zone_name)
 
 static inline kmem_zone_t *
 kmem_zone_init_flags(int size, char *zone_name, unsigned long flags,
-		     void (*construct)(kmem_zone_t *, void *))
+		     void (*construct)(void *))
 {
 	return kmem_cache_create(zone_name, size, 0, flags, construct);
 }
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index 742b2c7852c1..943381284e2e 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -843,7 +843,6 @@ xfs_fs_destroy_inode(
 
 STATIC void
 xfs_fs_inode_init_once(
-	kmem_zone_t		*zonep,
 	void			*vnode)
 {
 	inode_init_once(vn_to_inode((bhv_vnode_t *)vnode));
diff --git a/include/linux/slab.h b/include/linux/slab.h
index 41103910f8a2..9ff8e8499403 100644
--- a/include/linux/slab.h
+++ b/include/linux/slab.h
@@ -58,7 +58,7 @@ int slab_is_available(void);
 
 struct kmem_cache *kmem_cache_create(const char *, size_t, size_t,
 			unsigned long,
-			void (*)(struct kmem_cache *, void *));
+			void (*)(void *));
 void kmem_cache_destroy(struct kmem_cache *);
 int kmem_cache_shrink(struct kmem_cache *);
 void kmem_cache_free(struct kmem_cache *, void *);
diff --git a/include/linux/slub_def.h b/include/linux/slub_def.h
index d117ea2825a9..5bad61a93f65 100644
--- a/include/linux/slub_def.h
+++ b/include/linux/slub_def.h
@@ -85,7 +85,7 @@ struct kmem_cache {
 	struct kmem_cache_order_objects min;
 	gfp_t allocflags;	/* gfp flags to use on each alloc */
 	int refcount;		/* Refcount for slab cache destroy */
-	void (*ctor)(struct kmem_cache *, void *);
+	void (*ctor)(void *);
 	int inuse;		/* Offset to metadata */
 	int align;		/* Alignment */
 	const char *name;	/* Name (only for display!) */
diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index 1fdc2eb2f6d8..474984f9e032 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -207,7 +207,7 @@ static int mqueue_get_sb(struct file_system_type *fs_type,
 	return get_sb_single(fs_type, flags, data, mqueue_fill_super, mnt);
 }
 
-static void init_once(struct kmem_cache *cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct mqueue_inode_info *p = (struct mqueue_inode_info *) foo;
 
diff --git a/kernel/fork.c b/kernel/fork.c
index b99d73e971a4..80e83e459b17 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1442,7 +1442,7 @@ long do_fork(unsigned long clone_flags,
 #define ARCH_MIN_MMSTRUCT_ALIGN 0
 #endif
 
-static void sighand_ctor(struct kmem_cache *cachep, void *data)
+static void sighand_ctor(void *data)
 {
 	struct sighand_struct *sighand = data;
 
diff --git a/lib/idr.c b/lib/idr.c
index 3476f8203e97..e728c7fccc4d 100644
--- a/lib/idr.c
+++ b/lib/idr.c
@@ -607,7 +607,7 @@ void *idr_replace(struct idr *idp, void *ptr, int id)
 }
 EXPORT_SYMBOL(idr_replace);
 
-static void idr_cache_ctor(struct kmem_cache *idr_layer_cache, void *idr_layer)
+static void idr_cache_ctor(void *idr_layer)
 {
 	memset(idr_layer, 0, sizeof(struct idr_layer));
 }
diff --git a/lib/radix-tree.c b/lib/radix-tree.c
index 9c4f1ffa2864..be86b32bc874 100644
--- a/lib/radix-tree.c
+++ b/lib/radix-tree.c
@@ -1183,7 +1183,7 @@ int radix_tree_tagged(struct radix_tree_root *root, unsigned int tag)
 EXPORT_SYMBOL(radix_tree_tagged);
 
 static void
-radix_tree_node_ctor(struct kmem_cache *cachep, void *node)
+radix_tree_node_ctor(void *node)
 {
 	memset(node, 0, sizeof(struct radix_tree_node));
 }
diff --git a/mm/rmap.c b/mm/rmap.c
index abbd29f7c43f..39ae5a9bf382 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -138,7 +138,7 @@ void anon_vma_unlink(struct vm_area_struct *vma)
 		anon_vma_free(anon_vma);
 }
 
-static void anon_vma_ctor(struct kmem_cache *cachep, void *data)
+static void anon_vma_ctor(void *data)
 {
 	struct anon_vma *anon_vma = data;
 
diff --git a/mm/shmem.c b/mm/shmem.c
index 1089092aecaf..952d361774bb 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -2352,7 +2352,7 @@ static void shmem_destroy_inode(struct inode *inode)
 	kmem_cache_free(shmem_inode_cachep, SHMEM_I(inode));
 }
 
-static void init_once(struct kmem_cache *cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct shmem_inode_info *p = (struct shmem_inode_info *) foo;
 
diff --git a/mm/slab.c b/mm/slab.c
index 052e7d64537e..918f04f7fef1 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -406,7 +406,7 @@ struct kmem_cache {
 	unsigned int dflags;		/* dynamic flags */
 
 	/* constructor func */
-	void (*ctor)(struct kmem_cache *, void *);
+	void (*ctor)(void *obj);
 
 /* 5) cache creation/removal */
 	const char *name;
@@ -2137,8 +2137,7 @@ static int __init_refok setup_cpu_cache(struct kmem_cache *cachep)
  */
 struct kmem_cache *
 kmem_cache_create (const char *name, size_t size, size_t align,
-	unsigned long flags,
-	void (*ctor)(struct kmem_cache *, void *))
+	unsigned long flags, void (*ctor)(void *))
 {
 	size_t left_over, slab_size, ralign;
 	struct kmem_cache *cachep = NULL, *pc;
@@ -2653,7 +2652,7 @@ static void cache_init_objs(struct kmem_cache *cachep,
 		 * They must also be threaded.
 		 */
 		if (cachep->ctor && !(cachep->flags & SLAB_POISON))
-			cachep->ctor(cachep, objp + obj_offset(cachep));
+			cachep->ctor(objp + obj_offset(cachep));
 
 		if (cachep->flags & SLAB_RED_ZONE) {
 			if (*dbg_redzone2(cachep, objp) != RED_INACTIVE)
@@ -2669,7 +2668,7 @@ static void cache_init_objs(struct kmem_cache *cachep,
 					 cachep->buffer_size / PAGE_SIZE, 0);
 #else
 		if (cachep->ctor)
-			cachep->ctor(cachep, objp);
+			cachep->ctor(objp);
 #endif
 		slab_bufctl(slabp)[i] = i + 1;
 	}
@@ -3093,7 +3092,7 @@ static void *cache_alloc_debugcheck_after(struct kmem_cache *cachep,
 #endif
 	objp += obj_offset(cachep);
 	if (cachep->ctor && cachep->flags & SLAB_POISON)
-		cachep->ctor(cachep, objp);
+		cachep->ctor(objp);
 #if ARCH_SLAB_MINALIGN
 	if ((u32)objp & (ARCH_SLAB_MINALIGN-1)) {
 		printk(KERN_ERR "0x%p: not aligned to ARCH_SLAB_MINALIGN=%d\n",
diff --git a/mm/slob.c b/mm/slob.c
index de268eb7ac70..d8fbd4d1bfa7 100644
--- a/mm/slob.c
+++ b/mm/slob.c
@@ -525,12 +525,11 @@ struct kmem_cache {
 	unsigned int size, align;
 	unsigned long flags;
 	const char *name;
-	void (*ctor)(struct kmem_cache *, void *);
+	void (*ctor)(void *);
 };
 
 struct kmem_cache *kmem_cache_create(const char *name, size_t size,
-	size_t align, unsigned long flags,
-	void (*ctor)(struct kmem_cache *, void *))
+	size_t align, unsigned long flags, void (*ctor)(void *))
 {
 	struct kmem_cache *c;
 
@@ -575,7 +574,7 @@ void *kmem_cache_alloc_node(struct kmem_cache *c, gfp_t flags, int node)
 		b = slob_new_page(flags, get_order(c->size), node);
 
 	if (c->ctor)
-		c->ctor(c, b);
+		c->ctor(b);
 
 	return b;
 }
diff --git a/mm/slub.c b/mm/slub.c
index 77c21cf53ff9..b7e2cd5d82db 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -1012,7 +1012,7 @@ __setup("slub_debug", setup_slub_debug);
 
 static unsigned long kmem_cache_flags(unsigned long objsize,
 	unsigned long flags, const char *name,
-	void (*ctor)(struct kmem_cache *, void *))
+	void (*ctor)(void *))
 {
 	/*
 	 * Enable debugging if selected on the kernel commandline.
@@ -1040,7 +1040,7 @@ static inline int check_object(struct kmem_cache *s, struct page *page,
 static inline void add_full(struct kmem_cache_node *n, struct page *page) {}
 static inline unsigned long kmem_cache_flags(unsigned long objsize,
 	unsigned long flags, const char *name,
-	void (*ctor)(struct kmem_cache *, void *))
+	void (*ctor)(void *))
 {
 	return flags;
 }
@@ -1103,7 +1103,7 @@ static void setup_object(struct kmem_cache *s, struct page *page,
 {
 	setup_object_debug(s, page, object);
 	if (unlikely(s->ctor))
-		s->ctor(s, object);
+		s->ctor(object);
 }
 
 static struct page *new_slab(struct kmem_cache *s, gfp_t flags, int node)
@@ -2286,7 +2286,7 @@ static int calculate_sizes(struct kmem_cache *s, int forced_order)
 static int kmem_cache_open(struct kmem_cache *s, gfp_t gfpflags,
 		const char *name, size_t size,
 		size_t align, unsigned long flags,
-		void (*ctor)(struct kmem_cache *, void *))
+		void (*ctor)(void *))
 {
 	memset(s, 0, kmem_size);
 	s->name = name;
@@ -3042,7 +3042,7 @@ static int slab_unmergeable(struct kmem_cache *s)
 
 static struct kmem_cache *find_mergeable(size_t size,
 		size_t align, unsigned long flags, const char *name,
-		void (*ctor)(struct kmem_cache *, void *))
+		void (*ctor)(void *))
 {
 	struct kmem_cache *s;
 
@@ -3082,8 +3082,7 @@ static struct kmem_cache *find_mergeable(size_t size,
 }
 
 struct kmem_cache *kmem_cache_create(const char *name, size_t size,
-		size_t align, unsigned long flags,
-		void (*ctor)(struct kmem_cache *, void *))
+		size_t align, unsigned long flags, void (*ctor)(void *))
 {
 	struct kmem_cache *s;
 
diff --git a/net/socket.c b/net/socket.c
index 1310a82cbba7..8ef8ba81b9e2 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -265,7 +265,7 @@ static void sock_destroy_inode(struct inode *inode)
 			container_of(inode, struct socket_alloc, vfs_inode));
 }
 
-static void init_once(struct kmem_cache *cachep, void *foo)
+static void init_once(void *foo)
 {
 	struct socket_alloc *ei = (struct socket_alloc *)foo;
 
diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c
index 5a9b0e7828cd..23a2b8f6dc49 100644
--- a/net/sunrpc/rpc_pipe.c
+++ b/net/sunrpc/rpc_pipe.c
@@ -897,7 +897,7 @@ static struct file_system_type rpc_pipe_fs_type = {
 };
 
 static void
-init_once(struct kmem_cache * cachep, void *foo)
+init_once(void *foo)
 {
 	struct rpc_inode *rpci = (struct rpc_inode *) foo;
 

From b8c512f6190e313df69060bae4a161c5c044e272 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Fri, 25 Jul 2008 19:45:36 -0700
Subject: [PATCH 1038/1171] Use WARN() in kernel/irq/manage.c

Replace a printk+WARN_ON() by a WARN(); this increases the chance of the
string making it into the bugreport (ie: it goes inside the
---[ cut here ]--- section)

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/irq/manage.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
index f8914b92b664..152abfd3589f 100644
--- a/kernel/irq/manage.c
+++ b/kernel/irq/manage.c
@@ -177,8 +177,7 @@ static void __enable_irq(struct irq_desc *desc, unsigned int irq)
 {
 	switch (desc->depth) {
 	case 0:
-		printk(KERN_WARNING "Unbalanced enable for IRQ %d\n", irq);
-		WARN_ON(1);
+		WARN(1, KERN_WARNING "Unbalanced enable for IRQ %d\n", irq);
 		break;
 	case 1: {
 		unsigned int status = desc->status & ~IRQ_DISABLED;

From 4c8573e25f27b60b495aaa23089032f685ffd5ba Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Fri, 25 Jul 2008 19:45:37 -0700
Subject: [PATCH 1039/1171] Use WARN() in mm/vmalloc.c

Use WARN() instead of a printk+WARN_ON() pair; this way the message becomes
part of the warning section for better reporting/collection.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/vmalloc.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index 35f293816294..85b9a0d2c877 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -381,16 +381,14 @@ static void __vunmap(const void *addr, int deallocate_pages)
 		return;
 
 	if ((PAGE_SIZE-1) & (unsigned long)addr) {
-		printk(KERN_ERR "Trying to vfree() bad address (%p)\n", addr);
-		WARN_ON(1);
+		WARN(1, KERN_ERR "Trying to vfree() bad address (%p)\n", addr);
 		return;
 	}
 
 	area = remove_vm_area(addr);
 	if (unlikely(!area)) {
-		printk(KERN_ERR "Trying to vfree() nonexistent vm area (%p)\n",
+		WARN(1, KERN_ERR "Trying to vfree() nonexistent vm area (%p)\n",
 				addr);
-		WARN_ON(1);
 		return;
 	}
 

From 261c40c1191ad8d7a2e49fa2bb5f6a84e3d44b10 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Fri, 25 Jul 2008 19:45:37 -0700
Subject: [PATCH 1040/1171] use WARN() in kernel/irq/chip.c

Use WARN() instead of a printk+WARN_ON() pair; this way the message
becomes part of the warning section for better reporting/collection.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/irq/chip.c | 12 ++++--------
 1 file changed, 4 insertions(+), 8 deletions(-)

diff --git a/kernel/irq/chip.c b/kernel/irq/chip.c
index 964964baefa2..3cd441ebf5d2 100644
--- a/kernel/irq/chip.c
+++ b/kernel/irq/chip.c
@@ -28,8 +28,7 @@ void dynamic_irq_init(unsigned int irq)
 	unsigned long flags;
 
 	if (irq >= NR_IRQS) {
-		printk(KERN_ERR "Trying to initialize invalid IRQ%d\n", irq);
-		WARN_ON(1);
+		WARN(1, KERN_ERR "Trying to initialize invalid IRQ%d\n", irq);
 		return;
 	}
 
@@ -62,8 +61,7 @@ void dynamic_irq_cleanup(unsigned int irq)
 	unsigned long flags;
 
 	if (irq >= NR_IRQS) {
-		printk(KERN_ERR "Trying to cleanup invalid IRQ%d\n", irq);
-		WARN_ON(1);
+		WARN(1, KERN_ERR "Trying to cleanup invalid IRQ%d\n", irq);
 		return;
 	}
 
@@ -71,9 +69,8 @@ void dynamic_irq_cleanup(unsigned int irq)
 	spin_lock_irqsave(&desc->lock, flags);
 	if (desc->action) {
 		spin_unlock_irqrestore(&desc->lock, flags);
-		printk(KERN_ERR "Destroying IRQ%d without calling free_irq\n",
+		WARN(1, KERN_ERR "Destroying IRQ%d without calling free_irq\n",
 			irq);
-		WARN_ON(1);
 		return;
 	}
 	desc->msi_desc = NULL;
@@ -96,8 +93,7 @@ int set_irq_chip(unsigned int irq, struct irq_chip *chip)
 	unsigned long flags;
 
 	if (irq >= NR_IRQS) {
-		printk(KERN_ERR "Trying to install chip for IRQ%d\n", irq);
-		WARN_ON(1);
+		WARN(1, KERN_ERR "Trying to install chip for IRQ%d\n", irq);
 		return -EINVAL;
 	}
 

From 12e0036818eed243c8ed6583ebf98261a2554e12 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Fri, 25 Jul 2008 19:45:38 -0700
Subject: [PATCH 1041/1171] Use WARN() in block/

Use WARN() instead of a printk+WARN_ON() pair; this way the message
becomes part of the warning section for better reporting/collection.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Cc: Jens Axboe <jens.axboe@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 block/as-iosched.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/block/as-iosched.c b/block/as-iosched.c
index 9735acb5b4f5..cf4eb0eefbbf 100644
--- a/block/as-iosched.c
+++ b/block/as-iosched.c
@@ -837,8 +837,7 @@ static void as_completed_request(struct request_queue *q, struct request *rq)
 	WARN_ON(!list_empty(&rq->queuelist));
 
 	if (RQ_STATE(rq) != AS_RQ_REMOVED) {
-		printk("rq->state %d\n", RQ_STATE(rq));
-		WARN_ON(1);
+		WARN(1, "rq->state %d\n", RQ_STATE(rq));
 		goto out;
 	}
 

From f810a5cf28a818db96333cd23646f0227ec015b4 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Fri, 25 Jul 2008 19:45:39 -0700
Subject: [PATCH 1042/1171] Use WARN() in drivers/base/

Use WARN() instead of a printk+WARN_ON() pair; this way the message
becomes part of the warning section for better reporting/collection.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Cc: Greg KH <greg@kroah.com>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/base/core.c   |  6 ++----
 drivers/base/memory.c |  3 +--
 drivers/base/sys.c    | 12 ++++--------
 3 files changed, 7 insertions(+), 14 deletions(-)

diff --git a/drivers/base/core.c b/drivers/base/core.c
index 7d5c63c81a59..068aa1c9538c 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -116,12 +116,10 @@ static void device_release(struct kobject *kobj)
 		dev->type->release(dev);
 	else if (dev->class && dev->class->dev_release)
 		dev->class->dev_release(dev);
-	else {
-		printk(KERN_ERR "Device '%s' does not have a release() "
+	else
+		WARN(1, KERN_ERR "Device '%s' does not have a release() "
 			"function, it is broken and must be fixed.\n",
 			dev->bus_id);
-		WARN_ON(1);
-	}
 }
 
 static struct kobj_type device_ktype = {
diff --git a/drivers/base/memory.c b/drivers/base/memory.c
index 855ed1a9f97b..3ad49a00029f 100644
--- a/drivers/base/memory.c
+++ b/drivers/base/memory.c
@@ -204,9 +204,8 @@ memory_block_action(struct memory_block *mem, unsigned long action)
 			}
 			break;
 		default:
-			printk(KERN_WARNING "%s(%p, %ld) unknown action: %ld\n",
+			WARN(1, KERN_WARNING "%s(%p, %ld) unknown action: %ld\n",
 					__func__, mem, action, action);
-			WARN_ON(1);
 			ret = -EINVAL;
 	}
 
diff --git a/drivers/base/sys.c b/drivers/base/sys.c
index 40fc14f03540..75dd6e22faff 100644
--- a/drivers/base/sys.c
+++ b/drivers/base/sys.c
@@ -168,19 +168,16 @@ int sysdev_driver_register(struct sysdev_class *cls, struct sysdev_driver *drv)
 	int err = 0;
 
 	if (!cls) {
-		printk(KERN_WARNING "sysdev: invalid class passed to "
+		WARN(1, KERN_WARNING "sysdev: invalid class passed to "
 			"sysdev_driver_register!\n");
-		WARN_ON(1);
 		return -EINVAL;
 	}
 
 	/* Check whether this driver has already been added to a class. */
-	if (drv->entry.next && !list_empty(&drv->entry)) {
-		printk(KERN_WARNING "sysdev: class %s: driver (%p) has already"
+	if (drv->entry.next && !list_empty(&drv->entry))
+		WARN(1, KERN_WARNING "sysdev: class %s: driver (%p) has already"
 			" been registered to a class, something is wrong, but "
 			"will forge on!\n", cls->name, drv);
-		WARN_ON(1);
-	}
 
 	mutex_lock(&sysdev_drivers_lock);
 	if (cls && kset_get(&cls->kset)) {
@@ -194,8 +191,7 @@ int sysdev_driver_register(struct sysdev_class *cls, struct sysdev_driver *drv)
 		}
 	} else {
 		err = -EINVAL;
-		printk(KERN_ERR "%s: invalid device class\n", __func__);
-		WARN_ON(1);
+		WARN(1, KERN_ERR "%s: invalid device class\n", __func__);
 	}
 	mutex_unlock(&sysdev_drivers_lock);
 	return err;

From 5cd2b459d326a424671dcd95f038649f7bf7cb96 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Fri, 25 Jul 2008 19:45:39 -0700
Subject: [PATCH 1043/1171] Use WARN() in lib/

Use WARN() instead of a printk+WARN_ON() pair; this way the message becomes
part of the warning section for better reporting/collection.  In addition, one
of the if() clauses collapes into the WARN() entirely now.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 lib/debugobjects.c   | 15 +++++----------
 lib/iomap.c          |  3 +--
 lib/kobject_uevent.c |  6 ++----
 lib/plist.c          | 13 +++++++------
 4 files changed, 15 insertions(+), 22 deletions(-)

diff --git a/lib/debugobjects.c b/lib/debugobjects.c
index f86196390cfd..45a6bde762d1 100644
--- a/lib/debugobjects.c
+++ b/lib/debugobjects.c
@@ -205,9 +205,8 @@ static void debug_print_object(struct debug_obj *obj, char *msg)
 
 	if (limit < 5 && obj->descr != descr_test) {
 		limit++;
-		printk(KERN_ERR "ODEBUG: %s %s object type: %s\n", msg,
+		WARN(1, KERN_ERR "ODEBUG: %s %s object type: %s\n", msg,
 		       obj_states[obj->state], obj->descr->name);
-		WARN_ON(1);
 	}
 	debug_objects_warnings++;
 }
@@ -733,26 +732,22 @@ check_results(void *addr, enum debug_obj_state state, int fixups, int warnings)
 
 	obj = lookup_object(addr, db);
 	if (!obj && state != ODEBUG_STATE_NONE) {
-		printk(KERN_ERR "ODEBUG: selftest object not found\n");
-		WARN_ON(1);
+		WARN(1, KERN_ERR "ODEBUG: selftest object not found\n");
 		goto out;
 	}
 	if (obj && obj->state != state) {
-		printk(KERN_ERR "ODEBUG: selftest wrong state: %d != %d\n",
+		WARN(1, KERN_ERR "ODEBUG: selftest wrong state: %d != %d\n",
 		       obj->state, state);
-		WARN_ON(1);
 		goto out;
 	}
 	if (fixups != debug_objects_fixups) {
-		printk(KERN_ERR "ODEBUG: selftest fixups failed %d != %d\n",
+		WARN(1, KERN_ERR "ODEBUG: selftest fixups failed %d != %d\n",
 		       fixups, debug_objects_fixups);
-		WARN_ON(1);
 		goto out;
 	}
 	if (warnings != debug_objects_warnings) {
-		printk(KERN_ERR "ODEBUG: selftest warnings failed %d != %d\n",
+		WARN(1, KERN_ERR "ODEBUG: selftest warnings failed %d != %d\n",
 		       warnings, debug_objects_warnings);
-		WARN_ON(1);
 		goto out;
 	}
 	res = 0;
diff --git a/lib/iomap.c b/lib/iomap.c
index 37a3ea4cac9f..d32229385151 100644
--- a/lib/iomap.c
+++ b/lib/iomap.c
@@ -40,8 +40,7 @@ static void bad_io_access(unsigned long port, const char *access)
 	static int count = 10;
 	if (count) {
 		count--;
-		printk(KERN_ERR "Bad IO access at port %#lx (%s)\n", port, access);
-		WARN_ON(1);
+		WARN(1, KERN_ERR "Bad IO access at port %#lx (%s)\n", port, access);
 	}
 }
 
diff --git a/lib/kobject_uevent.c b/lib/kobject_uevent.c
index 9f8d599459d1..3f914725bda8 100644
--- a/lib/kobject_uevent.c
+++ b/lib/kobject_uevent.c
@@ -285,8 +285,7 @@ int add_uevent_var(struct kobj_uevent_env *env, const char *format, ...)
 	int len;
 
 	if (env->envp_idx >= ARRAY_SIZE(env->envp)) {
-		printk(KERN_ERR "add_uevent_var: too many keys\n");
-		WARN_ON(1);
+		WARN(1, KERN_ERR "add_uevent_var: too many keys\n");
 		return -ENOMEM;
 	}
 
@@ -297,8 +296,7 @@ int add_uevent_var(struct kobj_uevent_env *env, const char *format, ...)
 	va_end(args);
 
 	if (len >= (sizeof(env->buf) - env->buflen)) {
-		printk(KERN_ERR "add_uevent_var: buffer size too small\n");
-		WARN_ON(1);
+		WARN(1, KERN_ERR "add_uevent_var: buffer size too small\n");
 		return -ENOMEM;
 	}
 
diff --git a/lib/plist.c b/lib/plist.c
index 3074a02272f3..d6c64a824e1d 100644
--- a/lib/plist.c
+++ b/lib/plist.c
@@ -31,12 +31,13 @@
 static void plist_check_prev_next(struct list_head *t, struct list_head *p,
 				  struct list_head *n)
 {
-	if (n->prev != p || p->next != n) {
-		printk("top: %p, n: %p, p: %p\n", t, t->next, t->prev);
-		printk("prev: %p, n: %p, p: %p\n", p, p->next, p->prev);
-		printk("next: %p, n: %p, p: %p\n", n, n->next, n->prev);
-		WARN_ON(1);
-	}
+	WARN(n->prev != p || p->next != n,
+			"top: %p, n: %p, p: %p\n"
+			"prev: %p, n: %p, p: %p\n"
+			"next: %p, n: %p, p: %p\n",
+			 t, t->next, t->prev,
+			p, p->next, p->prev,
+			n, n->next, n->prev);
 }
 
 static void plist_check_list(struct list_head *top)

From 5c752ad9f35910ff1912b3f3ae82878178ddc432 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Fri, 25 Jul 2008 19:45:40 -0700
Subject: [PATCH 1044/1171] Use WARN() in fs/

Use WARN() instead of a printk+WARN_ON() pair; this way the message
becomes part of the warning section for better reporting/collection.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/buffer.c    | 3 +--
 fs/namespace.c | 3 +--
 2 files changed, 2 insertions(+), 4 deletions(-)

diff --git a/fs/buffer.c b/fs/buffer.c
index 5fd497cdd6f3..f95805019639 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -1214,8 +1214,7 @@ void __brelse(struct buffer_head * buf)
 		put_bh(buf);
 		return;
 	}
-	printk(KERN_ERR "VFS: brelse: Trying to free free buffer\n");
-	WARN_ON(1);
+	WARN(1, KERN_ERR "VFS: brelse: Trying to free free buffer\n");
 }
 
 /*
diff --git a/fs/namespace.c b/fs/namespace.c
index 4f6f7635b59c..f30b11e2240e 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -309,10 +309,9 @@ static void handle_write_count_underflow(struct vfsmount *mnt)
 	 */
 	if ((atomic_read(&mnt->__mnt_writers) < 0) &&
 	    !(mnt->mnt_flags & MNT_IMBALANCED_WRITE_COUNT)) {
-		printk(KERN_DEBUG "leak detected on mount(%p) writers "
+		WARN(1, KERN_DEBUG "leak detected on mount(%p) writers "
 				"count: %d\n",
 			mnt, atomic_read(&mnt->__mnt_writers));
-		WARN_ON(1);
 		/* use the flag to keep the dmesg spam down */
 		mnt->mnt_flags |= MNT_IMBALANCED_WRITE_COUNT;
 	}

From 99fcd77d15357e8ba51005c25cc750b9c28b2688 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Fri, 25 Jul 2008 19:45:41 -0700
Subject: [PATCH 1045/1171] Use WARN() in fs/sysfs

Use WARN() instead of a printk+WARN_ON() pair; this way the message becomes
part of the warning section for better reporting/collection.  Also, with this,
one fo the if() sections collapses entirely into the WARN().

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/sysfs/dir.c   | 5 +----
 fs/sysfs/file.c  | 3 +--
 fs/sysfs/group.c | 3 +--
 3 files changed, 3 insertions(+), 8 deletions(-)

diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c
index c1a7efb310bf..aedaeba82ae5 100644
--- a/fs/sysfs/dir.c
+++ b/fs/sysfs/dir.c
@@ -459,11 +459,8 @@ int sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd)
 	int ret;
 
 	ret = __sysfs_add_one(acxt, sd);
-	if (ret == -EEXIST) {
-		printk(KERN_WARNING "sysfs: duplicate filename '%s' "
+	WARN(ret == -EEXIST, KERN_WARNING "sysfs: duplicate filename '%s' "
 		       "can not be created\n", sd->s_name);
-		WARN_ON(1);
-	}
 	return ret;
 }
 
diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c
index 3f07893ff896..c9e4e5091da1 100644
--- a/fs/sysfs/file.c
+++ b/fs/sysfs/file.c
@@ -337,9 +337,8 @@ static int sysfs_open_file(struct inode *inode, struct file *file)
 	if (kobj->ktype && kobj->ktype->sysfs_ops)
 		ops = kobj->ktype->sysfs_ops;
 	else {
-		printk(KERN_ERR "missing sysfs attribute operations for "
+		WARN(1, KERN_ERR "missing sysfs attribute operations for "
 		       "kobject: %s\n", kobject_name(kobj));
-		WARN_ON(1);
 		goto err_out;
 	}
 
diff --git a/fs/sysfs/group.c b/fs/sysfs/group.c
index eeba38417b1d..fe611949a7f7 100644
--- a/fs/sysfs/group.c
+++ b/fs/sysfs/group.c
@@ -134,9 +134,8 @@ void sysfs_remove_group(struct kobject * kobj,
 	if (grp->name) {
 		sd = sysfs_get_dirent(dir_sd, grp->name);
 		if (!sd) {
-			printk(KERN_WARNING "sysfs group %p not found for "
+			WARN(!sd, KERN_WARNING "sysfs group %p not found for "
 				"kobject '%s'\n", grp, kobject_name(kobj));
-			WARN_ON(!sd);
 			return;
 		}
 	} else

From 267e2a9c71b8e088ac307f9549f71468e86e26c1 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Fri, 25 Jul 2008 19:45:41 -0700
Subject: [PATCH 1046/1171] Use WARN() in fs/proc/

Use WARN() instead of a printk+WARN_ON() pair; this way the message
becomes part of the warning section for better reporting/collection.
This way, the entire if() {} section can collapse into the WARN() as well.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Acked-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/proc/generic.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/fs/proc/generic.c b/fs/proc/generic.c
index bc0a0dd2d844..cb4096cc3fb7 100644
--- a/fs/proc/generic.c
+++ b/fs/proc/generic.c
@@ -806,12 +806,9 @@ continue_removing:
 	if (S_ISDIR(de->mode))
 		parent->nlink--;
 	de->nlink = 0;
-	if (de->subdir) {
-		printk(KERN_WARNING "%s: removing non-empty directory "
+	WARN(de->subdir, KERN_WARNING "%s: removing non-empty directory "
 			"'%s/%s', leaking at least '%s'\n", __func__,
 			de->parent->name, de->name, de->subdir->name);
-		WARN_ON(1);
-	}
 	if (atomic_dec_and_test(&de->count))
 		free_proc_entry(de);
 }

From 88ac2921a71f788ed693bcd44731dd6bc1994640 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:43 -0700
Subject: [PATCH 1047/1171] tracehook: add linux/tracehook.h

This patch series introduces the "tracehook" interface layer of inlines in
<linux/tracehook.h>.  There are more details in the log entry for patch
01/23 and in the header file comments inside that patch.  Most of these
changes move code around with little or no change, and they should not
break anything or change any behavior.

This sets a new standard for uniform arch support to enable clean
arch-independent implementations of new debugging and tracing stuff,
denoted by CONFIG_HAVE_ARCH_TRACEHOOK.  Patch 20/23 adds that symbol to
arch/Kconfig, with comments listing everything an arch has to do before
setting "select HAVE_ARCH_TRACEHOOK".  These are elaborted a bit at:

	http://sourceware.org/systemtap/wiki/utrace/arch/HowTo

The new inlines that arch code must define or call have detailed kerneldoc
comments in the generic header files that say what is required.

No arch is obligated to do any work, and no arch's build should be broken
by these changes.  There are several steps that each arch should take so
it can set HAVE_ARCH_TRACEHOOK.  Most of these are simple.  Providing this
support will let new things people add for doing debugging and tracing of
user-level threads "just work" for your arch in the future.  For an arch
that does not provide HAVE_ARCH_TRACEHOOK, some new options for such
features will not be available for config.

I have done some arch work and will submit this to the arch maintainers
after the generic tracehook series settles in.  For now, that work is
available in my GIT repositories, and in patch and mbox-of-patches form at
http://people.redhat.com/roland/utrace/2.6-current/

This paves the way for my "utrace" work, to be submitted later.  But it is
not innately tied to that.  I hope that the tracehook series can go in
soon regardless of what eventually does or doesn't go on top of it.  For
anyone implementing any kind of new tracing/debugging plan, or just
understanding all the context of the existing ptrace implementation,
having tracehook.h makes things much easier to find and understand.

This patch:

This adds the new kernel-internal header file <linux/tracehook.h>.  This
is not yet used at all.  The comments in the header introduce what the
following series of patches is about.

The aim is to formalize and consolidate all the places that the core
kernel code and the arch code now ties into the ptrace implementation.

These patches mostly don't cause any functional change.  They just move
the details of ptrace logic out of core code into tracehook.h inlines,
where they are mostly compiled away to the same as before.  All that
changes is that everything is thoroughly documented and any future
reworking of ptrace, or addition of something new, would not have to touch
core code all over, just change the tracehook.h inlines.

The new linux/ptrace.h inlines are used by the following patches in the
new tracehook_*() inlines.  Using these helpers for the ptrace event stops
makes it simple to change or disable the old ptrace implementation of
these stops conditionally later.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/ptrace.h    | 33 +++++++++++++++++++++++++
 include/linux/tracehook.h | 52 +++++++++++++++++++++++++++++++++++++++
 2 files changed, 85 insertions(+)
 create mode 100644 include/linux/tracehook.h

diff --git a/include/linux/ptrace.h b/include/linux/ptrace.h
index c6f5f9dd0cee..c74abfc4c7e8 100644
--- a/include/linux/ptrace.h
+++ b/include/linux/ptrace.h
@@ -121,6 +121,39 @@ static inline void ptrace_unlink(struct task_struct *child)
 int generic_ptrace_peekdata(struct task_struct *tsk, long addr, long data);
 int generic_ptrace_pokedata(struct task_struct *tsk, long addr, long data);
 
+/**
+ * task_ptrace - return %PT_* flags that apply to a task
+ * @task:	pointer to &task_struct in question
+ *
+ * Returns the %PT_* flags that apply to @task.
+ */
+static inline int task_ptrace(struct task_struct *task)
+{
+	return task->ptrace;
+}
+
+/**
+ * ptrace_event - possibly stop for a ptrace event notification
+ * @mask:	%PT_* bit to check in @current->ptrace
+ * @event:	%PTRACE_EVENT_* value to report if @mask is set
+ * @message:	value for %PTRACE_GETEVENTMSG to return
+ *
+ * This checks the @mask bit to see if ptrace wants stops for this event.
+ * If so we stop, reporting @event and @message to the ptrace parent.
+ *
+ * Returns nonzero if we did a ptrace notification, zero if not.
+ *
+ * Called without locks.
+ */
+static inline int ptrace_event(int mask, int event, unsigned long message)
+{
+	if (mask && likely(!(current->ptrace & mask)))
+		return 0;
+	current->ptrace_message = message;
+	ptrace_notify((event << 8) | SIGTRAP);
+	return 1;
+}
+
 #ifndef force_successful_syscall_return
 /*
  * System call handlers that, upon successful completion, need to return a
diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
new file mode 100644
index 000000000000..bea0f3eeff54
--- /dev/null
+++ b/include/linux/tracehook.h
@@ -0,0 +1,52 @@
+/*
+ * Tracing hooks
+ *
+ * Copyright (C) 2008 Red Hat, Inc.  All rights reserved.
+ *
+ * This copyrighted material is made available to anyone wishing to use,
+ * modify, copy, or redistribute it subject to the terms and conditions
+ * of the GNU General Public License v.2.
+ *
+ * This file defines hook entry points called by core code where
+ * user tracing/debugging support might need to do something.  These
+ * entry points are called tracehook_*().  Each hook declared below
+ * has a detailed kerneldoc comment giving the context (locking et
+ * al) from which it is called, and the meaning of its return value.
+ *
+ * Each function here typically has only one call site, so it is ok
+ * to have some nontrivial tracehook_*() inlines.  In all cases, the
+ * fast path when no tracing is enabled should be very short.
+ *
+ * The purpose of this file and the tracehook_* layer is to consolidate
+ * the interface that the kernel core and arch code uses to enable any
+ * user debugging or tracing facility (such as ptrace).  The interfaces
+ * here are carefully documented so that maintainers of core and arch
+ * code do not need to think about the implementation details of the
+ * tracing facilities.  Likewise, maintainers of the tracing code do not
+ * need to understand all the calling core or arch code in detail, just
+ * documented circumstances of each call, such as locking conditions.
+ *
+ * If the calling core code changes so that locking is different, then
+ * it is ok to change the interface documented here.  The maintainer of
+ * core code changing should notify the maintainers of the tracing code
+ * that they need to work out the change.
+ *
+ * Some tracehook_*() inlines take arguments that the current tracing
+ * implementations might not necessarily use.  These function signatures
+ * are chosen to pass in all the information that is on hand in the
+ * caller and might conceivably be relevant to a tracer, so that the
+ * core code won't have to be updated when tracing adds more features.
+ * If a call site changes so that some of those parameters are no longer
+ * already on hand without extra work, then the tracehook_* interface
+ * can change so there is no make-work burden on the core code.  The
+ * maintainer of core code changing should notify the maintainers of the
+ * tracing code that they need to work out the change.
+ */
+
+#ifndef _LINUX_TRACEHOOK_H
+#define _LINUX_TRACEHOOK_H	1
+
+#include <linux/sched.h>
+#include <linux/ptrace.h>
+
+#endif	/* <linux/tracehook.h> */

From 6341c393fcc37d58727865f1ee2f65e632e9d4f0 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:44 -0700
Subject: [PATCH 1048/1171] tracehook: exec

This moves all the ptrace hooks related to exec into tracehook.h inlines.

This also lifts the calls for tracing out of the binfmt load_binary hooks
into search_binary_handler() after it calls into the binfmt module.  This
change has no effect, since all the binfmt modules' load_binary functions
did the call at the end on success, and now search_binary_handler() does
it immediately after return if successful.  We consolidate the repeated
code, and binfmt modules no longer need to import ptrace_notify().

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/ia32/ia32_aout.c |  6 -----
 fs/binfmt_aout.c          |  6 -----
 fs/binfmt_elf.c           |  6 -----
 fs/binfmt_elf_fdpic.c     |  7 ------
 fs/binfmt_flat.c          |  3 ---
 fs/binfmt_som.c           |  2 --
 fs/exec.c                 | 12 ++++------
 include/linux/tracehook.h | 46 +++++++++++++++++++++++++++++++++++++++
 8 files changed, 50 insertions(+), 38 deletions(-)

diff --git a/arch/x86/ia32/ia32_aout.c b/arch/x86/ia32/ia32_aout.c
index 58cccb6483b0..a0e1dbe67dc1 100644
--- a/arch/x86/ia32/ia32_aout.c
+++ b/arch/x86/ia32/ia32_aout.c
@@ -441,12 +441,6 @@ beyond_if:
 	regs->r8 = regs->r9 = regs->r10 = regs->r11 =
 	regs->r12 = regs->r13 = regs->r14 = regs->r15 = 0;
 	set_fs(USER_DS);
-	if (unlikely(current->ptrace & PT_PTRACED)) {
-		if (current->ptrace & PT_TRACE_EXEC)
-			ptrace_notify((PTRACE_EVENT_EXEC << 8) | SIGTRAP);
-		else
-			send_sig(SIGTRAP, current, 0);
-	}
 	return 0;
 }
 
diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c
index ba4cddb92f1d..204cfd1d7676 100644
--- a/fs/binfmt_aout.c
+++ b/fs/binfmt_aout.c
@@ -444,12 +444,6 @@ beyond_if:
 	regs->gp = ex.a_gpvalue;
 #endif
 	start_thread(regs, ex.a_entry, current->mm->start_stack);
-	if (unlikely(current->ptrace & PT_PTRACED)) {
-		if (current->ptrace & PT_TRACE_EXEC)
-			ptrace_notify ((PTRACE_EVENT_EXEC << 8) | SIGTRAP);
-		else
-			send_sig(SIGTRAP, current, 0);
-	}
 	return 0;
 }
 
diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index 3b6ff854d983..655ed8d30a86 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -1003,12 +1003,6 @@ static int load_elf_binary(struct linux_binprm *bprm, struct pt_regs *regs)
 #endif
 
 	start_thread(regs, elf_entry, bprm->p);
-	if (unlikely(current->ptrace & PT_PTRACED)) {
-		if (current->ptrace & PT_TRACE_EXEC)
-			ptrace_notify ((PTRACE_EVENT_EXEC << 8) | SIGTRAP);
-		else
-			send_sig(SIGTRAP, current, 0);
-	}
 	retval = 0;
 out:
 	kfree(loc);
diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c
index 1b59b1edf26d..fdeadab2f18b 100644
--- a/fs/binfmt_elf_fdpic.c
+++ b/fs/binfmt_elf_fdpic.c
@@ -433,13 +433,6 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm,
 	entryaddr = interp_params.entry_addr ?: exec_params.entry_addr;
 	start_thread(regs, entryaddr, current->mm->start_stack);
 
-	if (unlikely(current->ptrace & PT_PTRACED)) {
-		if (current->ptrace & PT_TRACE_EXEC)
-			ptrace_notify((PTRACE_EVENT_EXEC << 8) | SIGTRAP);
-		else
-			send_sig(SIGTRAP, current, 0);
-	}
-
 	retval = 0;
 
 error:
diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c
index 2cb1acda3a82..56372ecf1690 100644
--- a/fs/binfmt_flat.c
+++ b/fs/binfmt_flat.c
@@ -920,9 +920,6 @@ static int load_flat_binary(struct linux_binprm * bprm, struct pt_regs * regs)
 	
 	start_thread(regs, start_addr, current->mm->start_stack);
 
-	if (current->ptrace & PT_PTRACED)
-		send_sig(SIGTRAP, current, 0);
-
 	return 0;
 }
 
diff --git a/fs/binfmt_som.c b/fs/binfmt_som.c
index fdc36bfd6a7b..68be580ba289 100644
--- a/fs/binfmt_som.c
+++ b/fs/binfmt_som.c
@@ -274,8 +274,6 @@ load_som_binary(struct linux_binprm * bprm, struct pt_regs * regs)
 	map_hpux_gateway_page(current,current->mm);
 
 	start_thread_som(regs, som_entry, bprm->p);
-	if (current->ptrace & PT_PTRACED)
-		send_sig(SIGTRAP, current, 0);
 	return 0;
 
 	/* error cleanup */
diff --git a/fs/exec.c b/fs/exec.c
index 5e559013e303..b8792a131533 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -42,13 +42,13 @@
 #include <linux/module.h>
 #include <linux/namei.h>
 #include <linux/proc_fs.h>
-#include <linux/ptrace.h>
 #include <linux/mount.h>
 #include <linux/security.h>
 #include <linux/syscalls.h>
 #include <linux/tsacct_kern.h>
 #include <linux/cn_proc.h>
 #include <linux/audit.h>
+#include <linux/tracehook.h>
 
 #include <asm/uaccess.h>
 #include <asm/mmu_context.h>
@@ -1071,13 +1071,8 @@ EXPORT_SYMBOL(prepare_binprm);
 
 static int unsafe_exec(struct task_struct *p)
 {
-	int unsafe = 0;
-	if (p->ptrace & PT_PTRACED) {
-		if (p->ptrace & PT_PTRACE_CAP)
-			unsafe |= LSM_UNSAFE_PTRACE_CAP;
-		else
-			unsafe |= LSM_UNSAFE_PTRACE;
-	}
+	int unsafe = tracehook_unsafe_exec(p);
+
 	if (atomic_read(&p->fs->count) > 1 ||
 	    atomic_read(&p->files->count) > 1 ||
 	    atomic_read(&p->sighand->count) > 1)
@@ -1214,6 +1209,7 @@ int search_binary_handler(struct linux_binprm *bprm,struct pt_regs *regs)
 			read_unlock(&binfmt_lock);
 			retval = fn(bprm, regs);
 			if (retval >= 0) {
+				tracehook_report_exec(fmt, bprm, regs);
 				put_binfmt(fmt);
 				allow_write_access(bprm->file);
 				if (bprm->file)
diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
index bea0f3eeff54..6276353709c1 100644
--- a/include/linux/tracehook.h
+++ b/include/linux/tracehook.h
@@ -48,5 +48,51 @@
 
 #include <linux/sched.h>
 #include <linux/ptrace.h>
+#include <linux/security.h>
+struct linux_binprm;
+
+/**
+ * tracehook_unsafe_exec - check for exec declared unsafe due to tracing
+ * @task:		current task doing exec
+ *
+ * Return %LSM_UNSAFE_* bits applied to an exec because of tracing.
+ *
+ * Called with task_lock() held on @task.
+ */
+static inline int tracehook_unsafe_exec(struct task_struct *task)
+{
+	int unsafe = 0;
+	int ptrace = task_ptrace(task);
+	if (ptrace & PT_PTRACED) {
+		if (ptrace & PT_PTRACE_CAP)
+			unsafe |= LSM_UNSAFE_PTRACE_CAP;
+		else
+			unsafe |= LSM_UNSAFE_PTRACE;
+	}
+	return unsafe;
+}
+
+/**
+ * tracehook_report_exec - a successful exec was completed
+ * @fmt:		&struct linux_binfmt that performed the exec
+ * @bprm:		&struct linux_binprm containing exec details
+ * @regs:		user-mode register state
+ *
+ * An exec just completed, we are shortly going to return to user mode.
+ * The freshly initialized register state can be seen and changed in @regs.
+ * The name, file and other pointers in @bprm are still on hand to be
+ * inspected, but will be freed as soon as this returns.
+ *
+ * Called with no locks, but with some kernel resources held live
+ * and a reference on @fmt->module.
+ */
+static inline void tracehook_report_exec(struct linux_binfmt *fmt,
+					 struct linux_binprm *bprm,
+					 struct pt_regs *regs)
+{
+	if (!ptrace_event(PT_TRACE_EXEC, PTRACE_EVENT_EXEC, 0) &&
+	    unlikely(task_ptrace(current) & PT_PTRACED))
+		send_sig(SIGTRAP, current, 0);
+}
 
 #endif	/* <linux/tracehook.h> */

From ff1188646c6870f336e910fb894eeed74f50471f Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:45 -0700
Subject: [PATCH 1049/1171] tracehook: unexport ptrace_notify

The ptrace_notify() function should not be called by any modules.  It was
only ever exported to be called by binfmt exec functions.  But that is no
longer necessary since fs/exec.c deals with that generically now.  There
should be no calls to ptrace_notify() from outside the core kernel.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/signal.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/kernel/signal.c b/kernel/signal.c
index 82c3545596c5..8715c18b27b9 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -1895,7 +1895,6 @@ EXPORT_SYMBOL(recalc_sigpending);
 EXPORT_SYMBOL_GPL(dequeue_signal);
 EXPORT_SYMBOL(flush_signals);
 EXPORT_SYMBOL(force_sig);
-EXPORT_SYMBOL(ptrace_notify);
 EXPORT_SYMBOL(send_sig);
 EXPORT_SYMBOL(send_sig_info);
 EXPORT_SYMBOL(sigprocmask);

From 30199f5a46aee204bf437a4f5b0740f3efe448b7 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:46 -0700
Subject: [PATCH 1050/1171] tracehook: exit

This moves the PTRACE_EVENT_EXIT tracing into a tracehook.h inline,
tracehook_report_exec().  The change has no effect, just clean-up.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/tracehook.h | 15 +++++++++++++++
 kernel/exit.c             |  6 ++----
 2 files changed, 17 insertions(+), 4 deletions(-)

diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
index 6276353709c1..967ab473afbc 100644
--- a/include/linux/tracehook.h
+++ b/include/linux/tracehook.h
@@ -95,4 +95,19 @@ static inline void tracehook_report_exec(struct linux_binfmt *fmt,
 		send_sig(SIGTRAP, current, 0);
 }
 
+/**
+ * tracehook_report_exit - task has begun to exit
+ * @exit_code:		pointer to value destined for @current->exit_code
+ *
+ * @exit_code points to the value passed to do_exit(), which tracing
+ * might change here.  This is almost the first thing in do_exit(),
+ * before freeing any resources or setting the %PF_EXITING flag.
+ *
+ * Called with no locks held.
+ */
+static inline void tracehook_report_exit(long *exit_code)
+{
+	ptrace_event(PT_TRACE_EXIT, PTRACE_EVENT_EXIT, *exit_code);
+}
+
 #endif	/* <linux/tracehook.h> */
diff --git a/kernel/exit.c b/kernel/exit.c
index ad933bb29ec7..c3691cbc220a 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -46,6 +46,7 @@
 #include <linux/resource.h>
 #include <linux/blkdev.h>
 #include <linux/task_io_accounting_ops.h>
+#include <linux/tracehook.h>
 
 #include <asm/uaccess.h>
 #include <asm/unistd.h>
@@ -1029,10 +1030,7 @@ NORET_TYPE void do_exit(long code)
 	if (unlikely(!tsk->pid))
 		panic("Attempted to kill the idle task!");
 
-	if (unlikely(current->ptrace & PT_TRACE_EXIT)) {
-		current->ptrace_message = code;
-		ptrace_notify((PTRACE_EVENT_EXIT << 8) | SIGTRAP);
-	}
+	tracehook_report_exit(&code);
 
 	/*
 	 * We're taking recursive faults here in do_exit. Safest is to just

From 09a05394fe2448a4139b014936330af23fa7ec83 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:47 -0700
Subject: [PATCH 1051/1171] tracehook: clone

This moves all the ptrace initialization and tracing logic for task
creation into tracehook.h and ptrace.h inlines.  It reorganizes the code
slightly, but should not change any behavior.

There are four tracehook entry points, at each important stage of task
creation.  This keeps the interface from the core fork.c code fairly
clean, while supporting the complex setup required for ptrace or something
like it.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/ptrace.h    |  22 +++++++++
 include/linux/tracehook.h | 100 ++++++++++++++++++++++++++++++++++++++
 kernel/fork.c             |  69 +++++++++++---------------
 3 files changed, 150 insertions(+), 41 deletions(-)

diff --git a/include/linux/ptrace.h b/include/linux/ptrace.h
index c74abfc4c7e8..dae6d85520fb 100644
--- a/include/linux/ptrace.h
+++ b/include/linux/ptrace.h
@@ -154,6 +154,28 @@ static inline int ptrace_event(int mask, int event, unsigned long message)
 	return 1;
 }
 
+/**
+ * ptrace_init_task - initialize ptrace state for a new child
+ * @child:		new child task
+ * @ptrace:		true if child should be ptrace'd by parent's tracer
+ *
+ * This is called immediately after adding @child to its parent's children
+ * list.  @ptrace is false in the normal case, and true to ptrace @child.
+ *
+ * Called with current's siglock and write_lock_irq(&tasklist_lock) held.
+ */
+static inline void ptrace_init_task(struct task_struct *child, bool ptrace)
+{
+	INIT_LIST_HEAD(&child->ptrace_entry);
+	INIT_LIST_HEAD(&child->ptraced);
+	child->parent = child->real_parent;
+	child->ptrace = 0;
+	if (unlikely(ptrace)) {
+		child->ptrace = current->ptrace;
+		__ptrace_link(child, current->parent);
+	}
+}
+
 #ifndef force_successful_syscall_return
 /*
  * System call handlers that, upon successful completion, need to return a
diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
index 967ab473afbc..3ebc58b59766 100644
--- a/include/linux/tracehook.h
+++ b/include/linux/tracehook.h
@@ -110,4 +110,104 @@ static inline void tracehook_report_exit(long *exit_code)
 	ptrace_event(PT_TRACE_EXIT, PTRACE_EVENT_EXIT, *exit_code);
 }
 
+/**
+ * tracehook_prepare_clone - prepare for new child to be cloned
+ * @clone_flags:	%CLONE_* flags from clone/fork/vfork system call
+ *
+ * This is called before a new user task is to be cloned.
+ * Its return value will be passed to tracehook_finish_clone().
+ *
+ * Called with no locks held.
+ */
+static inline int tracehook_prepare_clone(unsigned clone_flags)
+{
+	if (clone_flags & CLONE_UNTRACED)
+		return 0;
+
+	if (clone_flags & CLONE_VFORK) {
+		if (current->ptrace & PT_TRACE_VFORK)
+			return PTRACE_EVENT_VFORK;
+	} else if ((clone_flags & CSIGNAL) != SIGCHLD) {
+		if (current->ptrace & PT_TRACE_CLONE)
+			return PTRACE_EVENT_CLONE;
+	} else if (current->ptrace & PT_TRACE_FORK)
+		return PTRACE_EVENT_FORK;
+
+	return 0;
+}
+
+/**
+ * tracehook_finish_clone - new child created and being attached
+ * @child:		new child task
+ * @clone_flags:	%CLONE_* flags from clone/fork/vfork system call
+ * @trace:		return value from tracehook_clone_prepare()
+ *
+ * This is called immediately after adding @child to its parent's children list.
+ * The @trace value is that returned by tracehook_prepare_clone().
+ *
+ * Called with current's siglock and write_lock_irq(&tasklist_lock) held.
+ */
+static inline void tracehook_finish_clone(struct task_struct *child,
+					  unsigned long clone_flags, int trace)
+{
+	ptrace_init_task(child, (clone_flags & CLONE_PTRACE) || trace);
+}
+
+/**
+ * tracehook_report_clone - in parent, new child is about to start running
+ * @trace:		return value from tracehook_clone_prepare()
+ * @regs:		parent's user register state
+ * @clone_flags:	flags from parent's system call
+ * @pid:		new child's PID in the parent's namespace
+ * @child:		new child task
+ *
+ * Called after a child is set up, but before it has been started running.
+ * The @trace value is that returned by tracehook_clone_prepare().
+ * This is not a good place to block, because the child has not started yet.
+ * Suspend the child here if desired, and block in tracehook_clone_complete().
+ * This must prevent the child from self-reaping if tracehook_clone_complete()
+ * uses the @child pointer; otherwise it might have died and been released by
+ * the time tracehook_report_clone_complete() is called.
+ *
+ * Called with no locks held, but the child cannot run until this returns.
+ */
+static inline void tracehook_report_clone(int trace, struct pt_regs *regs,
+					  unsigned long clone_flags,
+					  pid_t pid, struct task_struct *child)
+{
+	if (unlikely(trace)) {
+		/*
+		 * The child starts up with an immediate SIGSTOP.
+		 */
+		sigaddset(&child->pending.signal, SIGSTOP);
+		set_tsk_thread_flag(child, TIF_SIGPENDING);
+	}
+}
+
+/**
+ * tracehook_report_clone_complete - new child is running
+ * @trace:		return value from tracehook_clone_prepare()
+ * @regs:		parent's user register state
+ * @clone_flags:	flags from parent's system call
+ * @pid:		new child's PID in the parent's namespace
+ * @child:		child task, already running
+ *
+ * This is called just after the child has started running.  This is
+ * just before the clone/fork syscall returns, or blocks for vfork
+ * child completion if @clone_flags has the %CLONE_VFORK bit set.
+ * The @child pointer may be invalid if a self-reaping child died and
+ * tracehook_report_clone() took no action to prevent it from self-reaping.
+ *
+ * Called with no locks held.
+ */
+static inline void tracehook_report_clone_complete(int trace,
+						   struct pt_regs *regs,
+						   unsigned long clone_flags,
+						   pid_t pid,
+						   struct task_struct *child)
+{
+	if (unlikely(trace))
+		ptrace_event(0, trace, pid);
+}
+
 #endif	/* <linux/tracehook.h> */
diff --git a/kernel/fork.c b/kernel/fork.c
index 80e83e459b17..b42f8ed23611 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -37,6 +37,7 @@
 #include <linux/swap.h>
 #include <linux/syscalls.h>
 #include <linux/jiffies.h>
+#include <linux/tracehook.h>
 #include <linux/futex.h>
 #include <linux/task_io_accounting_ops.h>
 #include <linux/rcupdate.h>
@@ -865,8 +866,7 @@ static void copy_flags(unsigned long clone_flags, struct task_struct *p)
 
 	new_flags &= ~PF_SUPERPRIV;
 	new_flags |= PF_FORKNOEXEC;
-	if (!(clone_flags & CLONE_PTRACE))
-		p->ptrace = 0;
+	new_flags |= PF_STARTING;
 	p->flags = new_flags;
 	clear_freeze_flag(p);
 }
@@ -907,7 +907,8 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 					struct pt_regs *regs,
 					unsigned long stack_size,
 					int __user *child_tidptr,
-					struct pid *pid)
+					struct pid *pid,
+					int trace)
 {
 	int retval;
 	struct task_struct *p;
@@ -1163,8 +1164,6 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 	 */
 	p->group_leader = p;
 	INIT_LIST_HEAD(&p->thread_group);
-	INIT_LIST_HEAD(&p->ptrace_entry);
-	INIT_LIST_HEAD(&p->ptraced);
 
 	/* Now that the task is set up, run cgroup callbacks if
 	 * necessary. We need to run them before the task is visible
@@ -1195,7 +1194,6 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 		p->real_parent = current->real_parent;
 	else
 		p->real_parent = current;
-	p->parent = p->real_parent;
 
 	spin_lock(&current->sighand->siglock);
 
@@ -1237,8 +1235,7 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 
 	if (likely(p->pid)) {
 		list_add_tail(&p->sibling, &p->real_parent->children);
-		if (unlikely(p->ptrace & PT_PTRACED))
-			__ptrace_link(p, current->parent);
+		tracehook_finish_clone(p, clone_flags, trace);
 
 		if (thread_group_leader(p)) {
 			if (clone_flags & CLONE_NEWPID)
@@ -1323,29 +1320,13 @@ struct task_struct * __cpuinit fork_idle(int cpu)
 	struct pt_regs regs;
 
 	task = copy_process(CLONE_VM, 0, idle_regs(&regs), 0, NULL,
-				&init_struct_pid);
+			    &init_struct_pid, 0);
 	if (!IS_ERR(task))
 		init_idle(task, cpu);
 
 	return task;
 }
 
-static int fork_traceflag(unsigned clone_flags)
-{
-	if (clone_flags & CLONE_UNTRACED)
-		return 0;
-	else if (clone_flags & CLONE_VFORK) {
-		if (current->ptrace & PT_TRACE_VFORK)
-			return PTRACE_EVENT_VFORK;
-	} else if ((clone_flags & CSIGNAL) != SIGCHLD) {
-		if (current->ptrace & PT_TRACE_CLONE)
-			return PTRACE_EVENT_CLONE;
-	} else if (current->ptrace & PT_TRACE_FORK)
-		return PTRACE_EVENT_FORK;
-
-	return 0;
-}
-
 /*
  *  Ok, this is the main fork-routine.
  *
@@ -1380,14 +1361,14 @@ long do_fork(unsigned long clone_flags,
 		}
 	}
 
-	if (unlikely(current->ptrace)) {
-		trace = fork_traceflag (clone_flags);
-		if (trace)
-			clone_flags |= CLONE_PTRACE;
-	}
+	/*
+	 * When called from kernel_thread, don't do user tracing stuff.
+	 */
+	if (likely(user_mode(regs)))
+		trace = tracehook_prepare_clone(clone_flags);
 
 	p = copy_process(clone_flags, stack_start, regs, stack_size,
-			child_tidptr, NULL);
+			 child_tidptr, NULL, trace);
 	/*
 	 * Do this prior waking up the new thread - the thread pointer
 	 * might get invalid after that point, if the thread exits quickly.
@@ -1405,24 +1386,30 @@ long do_fork(unsigned long clone_flags,
 			init_completion(&vfork);
 		}
 
-		if ((p->ptrace & PT_PTRACED) || (clone_flags & CLONE_STOPPED)) {
+		tracehook_report_clone(trace, regs, clone_flags, nr, p);
+
+		/*
+		 * We set PF_STARTING at creation in case tracing wants to
+		 * use this to distinguish a fully live task from one that
+		 * hasn't gotten to tracehook_report_clone() yet.  Now we
+		 * clear it and set the child going.
+		 */
+		p->flags &= ~PF_STARTING;
+
+		if (unlikely(clone_flags & CLONE_STOPPED)) {
 			/*
 			 * We'll start up with an immediate SIGSTOP.
 			 */
 			sigaddset(&p->pending.signal, SIGSTOP);
 			set_tsk_thread_flag(p, TIF_SIGPENDING);
-		}
-
-		if (!(clone_flags & CLONE_STOPPED))
-			wake_up_new_task(p, clone_flags);
-		else
 			__set_task_state(p, TASK_STOPPED);
-
-		if (unlikely (trace)) {
-			current->ptrace_message = nr;
-			ptrace_notify ((trace << 8) | SIGTRAP);
+		} else {
+			wake_up_new_task(p, clone_flags);
 		}
 
+		tracehook_report_clone_complete(trace, regs,
+						clone_flags, nr, p);
+
 		if (clone_flags & CLONE_VFORK) {
 			freezer_do_not_count();
 			wait_for_completion(&vfork);

From daded34be96b1975ff8539ff62ad8b158ce7d842 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:47 -0700
Subject: [PATCH 1052/1171] tracehook: vfork-done

This moves the PTRACE_EVENT_VFORK_DONE tracing into a tracehook.h inline,
tracehook_report_vfork_done().  The change has no effect, just clean-up.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/tracehook.h | 18 ++++++++++++++++++
 kernel/fork.c             |  5 +----
 2 files changed, 19 insertions(+), 4 deletions(-)

diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
index 3ebc58b59766..830e6e16097d 100644
--- a/include/linux/tracehook.h
+++ b/include/linux/tracehook.h
@@ -210,4 +210,22 @@ static inline void tracehook_report_clone_complete(int trace,
 		ptrace_event(0, trace, pid);
 }
 
+/**
+ * tracehook_report_vfork_done - vfork parent's child has exited or exec'd
+ * @child:		child task, already running
+ * @pid:		new child's PID in the parent's namespace
+ *
+ * Called after a %CLONE_VFORK parent has waited for the child to complete.
+ * The clone/vfork system call will return immediately after this.
+ * The @child pointer may be invalid if a self-reaping child died and
+ * tracehook_report_clone() took no action to prevent it from self-reaping.
+ *
+ * Called with no locks held.
+ */
+static inline void tracehook_report_vfork_done(struct task_struct *child,
+					       pid_t pid)
+{
+	ptrace_event(PT_TRACE_VFORK_DONE, PTRACE_EVENT_VFORK_DONE, pid);
+}
+
 #endif	/* <linux/tracehook.h> */
diff --git a/kernel/fork.c b/kernel/fork.c
index b42f8ed23611..abb3ed6298f6 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -1414,10 +1414,7 @@ long do_fork(unsigned long clone_flags,
 			freezer_do_not_count();
 			wait_for_completion(&vfork);
 			freezer_count();
-			if (unlikely (current->ptrace & PT_TRACE_VFORK_DONE)) {
-				current->ptrace_message = nr;
-				ptrace_notify ((PTRACE_EVENT_VFORK_DONE << 8) | SIGTRAP);
-			}
+			tracehook_report_vfork_done(p, nr);
 		}
 	} else {
 		nr = PTR_ERR(p);

From dae33574dcf5211e1f43c7e45fa29f73ba3e00cb Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:48 -0700
Subject: [PATCH 1053/1171] tracehook: release_task

This moves the ptrace-related logic from release_task into tracehook.h and
ptrace.h inlines.  It provides clean hooks both before and after locking
tasklist_lock, for future tracing logic to do more cleanup without the
lock.

This also changes release_task() itself in the rare "zap_leader" case to
set the leader to EXIT_DEAD before iterating.  This maintains the
invariant that release_task() only ever handles a task in EXIT_DEAD.  This
is a common-sense invariant that is already always true except in this one
arcane case of zombie leader whose parent ignores SIGCHLD.

This change is harmless and only costs one store in this one rare case.
It keeps the expected state more consisently sane, which is nicer when
debugging weirdness in release_task().  It also lets some future code in
the tracehook entry points rely on this invariant for bookkeeping.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/ptrace.h    | 13 +++++++++++++
 include/linux/tracehook.h | 28 ++++++++++++++++++++++++++++
 kernel/exit.c             | 21 +++++++++------------
 3 files changed, 50 insertions(+), 12 deletions(-)

diff --git a/include/linux/ptrace.h b/include/linux/ptrace.h
index dae6d85520fb..ed69c03692d9 100644
--- a/include/linux/ptrace.h
+++ b/include/linux/ptrace.h
@@ -176,6 +176,19 @@ static inline void ptrace_init_task(struct task_struct *child, bool ptrace)
 	}
 }
 
+/**
+ * ptrace_release_task - final ptrace-related cleanup of a zombie being reaped
+ * @task:	task in %EXIT_DEAD state
+ *
+ * Called with write_lock(&tasklist_lock) held.
+ */
+static inline void ptrace_release_task(struct task_struct *task)
+{
+	BUG_ON(!list_empty(&task->ptraced));
+	ptrace_unlink(task);
+	BUG_ON(!list_empty(&task->ptrace_entry));
+}
+
 #ifndef force_successful_syscall_return
 /*
  * System call handlers that, upon successful completion, need to return a
diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
index 830e6e16097d..9a5b3be2503a 100644
--- a/include/linux/tracehook.h
+++ b/include/linux/tracehook.h
@@ -228,4 +228,32 @@ static inline void tracehook_report_vfork_done(struct task_struct *child,
 	ptrace_event(PT_TRACE_VFORK_DONE, PTRACE_EVENT_VFORK_DONE, pid);
 }
 
+/**
+ * tracehook_prepare_release_task - task is being reaped, clean up tracing
+ * @task:		task in %EXIT_DEAD state
+ *
+ * This is called in release_task() just before @task gets finally reaped
+ * and freed.  This would be the ideal place to remove and clean up any
+ * tracing-related state for @task.
+ *
+ * Called with no locks held.
+ */
+static inline void tracehook_prepare_release_task(struct task_struct *task)
+{
+}
+
+/**
+ * tracehook_finish_release_task - task is being reaped, clean up tracing
+ * @task:		task in %EXIT_DEAD state
+ *
+ * This is called in release_task() when @task is being in the middle of
+ * being reaped.  After this, there must be no tracing entanglements.
+ *
+ * Called with write_lock_irq(&tasklist_lock) held.
+ */
+static inline void tracehook_finish_release_task(struct task_struct *task)
+{
+	ptrace_release_task(task);
+}
+
 #endif	/* <linux/tracehook.h> */
diff --git a/kernel/exit.c b/kernel/exit.c
index c3691cbc220a..da28745f7c38 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -163,27 +163,17 @@ static void delayed_put_task_struct(struct rcu_head *rhp)
 	put_task_struct(container_of(rhp, struct task_struct, rcu));
 }
 
-/*
- * Do final ptrace-related cleanup of a zombie being reaped.
- *
- * Called with write_lock(&tasklist_lock) held.
- */
-static void ptrace_release_task(struct task_struct *p)
-{
-	BUG_ON(!list_empty(&p->ptraced));
-	ptrace_unlink(p);
-	BUG_ON(!list_empty(&p->ptrace_entry));
-}
 
 void release_task(struct task_struct * p)
 {
 	struct task_struct *leader;
 	int zap_leader;
 repeat:
+	tracehook_prepare_release_task(p);
 	atomic_dec(&p->user->processes);
 	proc_flush_task(p);
 	write_lock_irq(&tasklist_lock);
-	ptrace_release_task(p);
+	tracehook_finish_release_task(p);
 	__exit_signal(p);
 
 	/*
@@ -205,6 +195,13 @@ repeat:
 		 * that case.
 		 */
 		zap_leader = task_detached(leader);
+
+		/*
+		 * This maintains the invariant that release_task()
+		 * only runs on a task in EXIT_DEAD, just for sanity.
+		 */
+		if (zap_leader)
+			leader->exit_state = EXIT_DEAD;
 	}
 
 	write_unlock_irq(&tasklist_lock);

From 0d094efeb1e98010c6b99923f1eb7e17bf1e3a74 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:49 -0700
Subject: [PATCH 1054/1171] tracehook: tracehook_tracer_task

This adds the tracehook_tracer_task() hook to consolidate all forms of
"Who is using ptrace on me?" logic.  This is used for "TracerPid:" in
/proc and for permission checks.  We also clean up the selinux code the
called an identical accessor.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/proc/array.c           |  9 +++++++--
 fs/proc/base.c            | 13 +++++++++----
 include/linux/tracehook.h | 18 ++++++++++++++++++
 security/selinux/hooks.c  | 22 +++-------------------
 4 files changed, 37 insertions(+), 25 deletions(-)

diff --git a/fs/proc/array.c b/fs/proc/array.c
index 797d775e0354..0d6eb33597c6 100644
--- a/fs/proc/array.c
+++ b/fs/proc/array.c
@@ -80,6 +80,7 @@
 #include <linux/delayacct.h>
 #include <linux/seq_file.h>
 #include <linux/pid_namespace.h>
+#include <linux/tracehook.h>
 
 #include <asm/pgtable.h>
 #include <asm/processor.h>
@@ -168,8 +169,12 @@ static inline void task_state(struct seq_file *m, struct pid_namespace *ns,
 	rcu_read_lock();
 	ppid = pid_alive(p) ?
 		task_tgid_nr_ns(rcu_dereference(p->real_parent), ns) : 0;
-	tpid = pid_alive(p) && p->ptrace ?
-		task_pid_nr_ns(rcu_dereference(p->parent), ns) : 0;
+	tpid = 0;
+	if (pid_alive(p)) {
+		struct task_struct *tracer = tracehook_tracer_task(p);
+		if (tracer)
+			tpid = task_pid_nr_ns(tracer, ns);
+	}
 	seq_printf(m,
 		"State:\t%s\n"
 		"Tgid:\t%d\n"
diff --git a/fs/proc/base.c b/fs/proc/base.c
index a891fe4cb43b..4b74dba69a6d 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -69,6 +69,7 @@
 #include <linux/mount.h>
 #include <linux/security.h>
 #include <linux/ptrace.h>
+#include <linux/tracehook.h>
 #include <linux/cgroup.h>
 #include <linux/cpuset.h>
 #include <linux/audit.h>
@@ -231,10 +232,14 @@ static int check_mem_permission(struct task_struct *task)
 	 * If current is actively ptrace'ing, and would also be
 	 * permitted to freshly attach with ptrace now, permit it.
 	 */
-	if (task->parent == current && (task->ptrace & PT_PTRACED) &&
-	    task_is_stopped_or_traced(task) &&
-	    ptrace_may_access(task, PTRACE_MODE_ATTACH))
-		return 0;
+	if (task_is_stopped_or_traced(task)) {
+		int match;
+		rcu_read_lock();
+		match = (tracehook_tracer_task(task) == current);
+		rcu_read_unlock();
+		if (match && ptrace_may_access(task, PTRACE_MODE_ATTACH))
+			return 0;
+	}
 
 	/*
 	 * Noone else is allowed.
diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
index 9a5b3be2503a..6468ca0fe69b 100644
--- a/include/linux/tracehook.h
+++ b/include/linux/tracehook.h
@@ -72,6 +72,24 @@ static inline int tracehook_unsafe_exec(struct task_struct *task)
 	return unsafe;
 }
 
+/**
+ * tracehook_tracer_task - return the task that is tracing the given task
+ * @tsk:		task to consider
+ *
+ * Returns NULL if noone is tracing @task, or the &struct task_struct
+ * pointer to its tracer.
+ *
+ * Must called under rcu_read_lock().  The pointer returned might be kept
+ * live only by RCU.  During exec, this may be called with task_lock()
+ * held on @task, still held from when tracehook_unsafe_exec() was called.
+ */
+static inline struct task_struct *tracehook_tracer_task(struct task_struct *tsk)
+{
+	if (task_ptrace(tsk) & PT_PTRACED)
+		return rcu_dereference(tsk->parent);
+	return NULL;
+}
+
 /**
  * tracehook_report_exec - a successful exec was completed
  * @fmt:		&struct linux_binfmt that performed the exec
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 63f131fc42e4..3481cde5bf15 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -25,7 +25,7 @@
 
 #include <linux/init.h>
 #include <linux/kernel.h>
-#include <linux/ptrace.h>
+#include <linux/tracehook.h>
 #include <linux/errno.h>
 #include <linux/sched.h>
 #include <linux/security.h>
@@ -1971,22 +1971,6 @@ static int selinux_vm_enough_memory(struct mm_struct *mm, long pages)
 	return __vm_enough_memory(mm, pages, cap_sys_admin);
 }
 
-/**
- * task_tracer_task - return the task that is tracing the given task
- * @task:		task to consider
- *
- * Returns NULL if noone is tracing @task, or the &struct task_struct
- * pointer to its tracer.
- *
- * Must be called under rcu_read_lock().
- */
-static struct task_struct *task_tracer_task(struct task_struct *task)
-{
-	if (task->ptrace & PT_PTRACED)
-		return rcu_dereference(task->parent);
-	return NULL;
-}
-
 /* binprm security operations */
 
 static int selinux_bprm_alloc_security(struct linux_binprm *bprm)
@@ -2238,7 +2222,7 @@ static void selinux_bprm_apply_creds(struct linux_binprm *bprm, int unsafe)
 			u32 ptsid = 0;
 
 			rcu_read_lock();
-			tracer = task_tracer_task(current);
+			tracer = tracehook_tracer_task(current);
 			if (likely(tracer != NULL)) {
 				sec = tracer->security;
 				ptsid = sec->sid;
@@ -5247,7 +5231,7 @@ static int selinux_setprocattr(struct task_struct *p,
 		   Otherwise, leave SID unchanged and fail. */
 		task_lock(p);
 		rcu_read_lock();
-		tracer = task_tracer_task(p);
+		tracer = tracehook_tracer_task(p);
 		if (tracer != NULL) {
 			struct task_security_struct *ptsec = tracer->security;
 			u32 ptsid = ptsec->sid;

From fa8e26ccd485216fc45c8c2dd1ec3b7ef1a0a2f8 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:50 -0700
Subject: [PATCH 1055/1171] tracehook: tracehook_expect_breakpoints

This adds tracehook_expect_breakpoints() as a formal hook for the nommu
code to use for its, "Is text-poking likely?" check at mmap time.  This
names the actual semantics the code means to test, and documents it.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/tracehook.h | 15 +++++++++++++++
 mm/nommu.c                |  4 ++--
 2 files changed, 17 insertions(+), 2 deletions(-)

diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
index 6468ca0fe69b..e113e09b0341 100644
--- a/include/linux/tracehook.h
+++ b/include/linux/tracehook.h
@@ -51,6 +51,21 @@
 #include <linux/security.h>
 struct linux_binprm;
 
+/**
+ * tracehook_expect_breakpoints - guess if task memory might be touched
+ * @task:		current task, making a new mapping
+ *
+ * Return nonzero if @task is expected to want breakpoint insertion in
+ * its memory at some point.  A zero return is no guarantee it won't
+ * be done, but this is a hint that it's known to be likely.
+ *
+ * May be called with @task->mm->mmap_sem held for writing.
+ */
+static inline int tracehook_expect_breakpoints(struct task_struct *task)
+{
+	return (task_ptrace(task) & PT_PTRACED) != 0;
+}
+
 /**
  * tracehook_unsafe_exec - check for exec declared unsafe due to tracing
  * @task:		current task doing exec
diff --git a/mm/nommu.c b/mm/nommu.c
index 4462b6a3fcb9..5edccd9c9218 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -22,7 +22,7 @@
 #include <linux/pagemap.h>
 #include <linux/slab.h>
 #include <linux/vmalloc.h>
-#include <linux/ptrace.h>
+#include <linux/tracehook.h>
 #include <linux/blkdev.h>
 #include <linux/backing-dev.h>
 #include <linux/mount.h>
@@ -745,7 +745,7 @@ static unsigned long determine_vm_flags(struct file *file,
 	 * it's being traced - otherwise breakpoints set in it may interfere
 	 * with another untraced process
 	 */
-	if ((flags & MAP_PRIVATE) && (current->ptrace & PT_PTRACED))
+	if ((flags & MAP_PRIVATE) && tracehook_expect_breakpoints(current))
 		vm_flags &= ~VM_MAYSHARE;
 
 	return vm_flags;

From c45aea27617d6a1e0aacddc3b0233f704222fcbd Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:50 -0700
Subject: [PATCH 1056/1171] tracehook: tracehook_signal_handler

This defines tracehook_signal_handler() as a hook for the arch signal
handling code to call.  It gives ptrace the opportunity to stop for a
pseudo-single-step trap immediately after signal handler setup is done.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/tracehook.h | 23 +++++++++++++++++++++++
 1 file changed, 23 insertions(+)

diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
index e113e09b0341..2d1426f8e33b 100644
--- a/include/linux/tracehook.h
+++ b/include/linux/tracehook.h
@@ -289,4 +289,27 @@ static inline void tracehook_finish_release_task(struct task_struct *task)
 	ptrace_release_task(task);
 }
 
+/**
+ * tracehook_signal_handler - signal handler setup is complete
+ * @sig:		number of signal being delivered
+ * @info:		siginfo_t of signal being delivered
+ * @ka:			sigaction setting that chose the handler
+ * @regs:		user register state
+ * @stepping:		nonzero if debugger single-step or block-step in use
+ *
+ * Called by the arch code after a signal handler has been set up.
+ * Register and stack state reflects the user handler about to run.
+ * Signal mask changes have already been made.
+ *
+ * Called without locks, shortly before returning to user mode
+ * (or handling more signals).
+ */
+static inline void tracehook_signal_handler(int sig, siginfo_t *info,
+					    const struct k_sigaction *ka,
+					    struct pt_regs *regs, int stepping)
+{
+	if (stepping)
+		ptrace_notify(SIGTRAP);
+}
+
 #endif	/* <linux/tracehook.h> */

From 35de254dc60f91004b3b5ebb1fc7b2c3093d6032 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:51 -0700
Subject: [PATCH 1057/1171] tracehook: tracehook_consider_ignored_signal

This defines tracehook_consider_ignored_signal() has a fine-grained hook
for deciding to prevent the normal short-circuit of sending an ignored
signal, as ptrace does.  There is no change, only cleanup.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/tracehook.h | 19 +++++++++++++++++++
 kernel/signal.c           | 27 ++++++++++++++++-----------
 2 files changed, 35 insertions(+), 11 deletions(-)

diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
index 2d1426f8e33b..8cffd34f88d5 100644
--- a/include/linux/tracehook.h
+++ b/include/linux/tracehook.h
@@ -312,4 +312,23 @@ static inline void tracehook_signal_handler(int sig, siginfo_t *info,
 		ptrace_notify(SIGTRAP);
 }
 
+/**
+ * tracehook_consider_ignored_signal - suppress short-circuit of ignored signal
+ * @task:		task receiving the signal
+ * @sig:		signal number being sent
+ * @handler:		%SIG_IGN or %SIG_DFL
+ *
+ * Return zero iff tracing doesn't care to examine this ignored signal,
+ * so it can short-circuit normal delivery and never even get queued.
+ * Either @handler is %SIG_DFL and @sig's default is ignore, or it's %SIG_IGN.
+ *
+ * Called with @task->sighand->siglock held.
+ */
+static inline int tracehook_consider_ignored_signal(struct task_struct *task,
+						    int sig,
+						    void __user *handler)
+{
+	return (task_ptrace(task) & PT_PTRACED) != 0;
+}
+
 #endif	/* <linux/tracehook.h> */
diff --git a/kernel/signal.c b/kernel/signal.c
index 8715c18b27b9..9efd1cee6d0b 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -22,6 +22,7 @@
 #include <linux/ptrace.h>
 #include <linux/signal.h>
 #include <linux/signalfd.h>
+#include <linux/tracehook.h>
 #include <linux/capability.h>
 #include <linux/freezer.h>
 #include <linux/pid_namespace.h>
@@ -39,24 +40,21 @@
 
 static struct kmem_cache *sigqueue_cachep;
 
-static int __sig_ignored(struct task_struct *t, int sig)
+static void __user *sig_handler(struct task_struct *t, int sig)
 {
-	void __user *handler;
+	return t->sighand->action[sig - 1].sa.sa_handler;
+}
 
+static int sig_handler_ignored(void __user *handler, int sig)
+{
 	/* Is it explicitly or implicitly ignored? */
-
-	handler = t->sighand->action[sig - 1].sa.sa_handler;
 	return handler == SIG_IGN ||
 		(handler == SIG_DFL && sig_kernel_ignore(sig));
 }
 
 static int sig_ignored(struct task_struct *t, int sig)
 {
-	/*
-	 * Tracers always want to know about signals..
-	 */
-	if (t->ptrace & PT_PTRACED)
-		return 0;
+	void __user *handler;
 
 	/*
 	 * Blocked signals are never ignored, since the
@@ -66,7 +64,14 @@ static int sig_ignored(struct task_struct *t, int sig)
 	if (sigismember(&t->blocked, sig) || sigismember(&t->real_blocked, sig))
 		return 0;
 
-	return __sig_ignored(t, sig);
+	handler = sig_handler(t, sig);
+	if (!sig_handler_ignored(handler, sig))
+		return 0;
+
+	/*
+	 * Tracers may want to know about even ignored signals.
+	 */
+	return !tracehook_consider_ignored_signal(t, sig, handler);
 }
 
 /*
@@ -2298,7 +2303,7 @@ int do_sigaction(int sig, struct k_sigaction *act, struct k_sigaction *oact)
 		 *   (for example, SIGCHLD), shall cause the pending signal to
 		 *   be discarded, whether or not it is blocked"
 		 */
-		if (__sig_ignored(t, sig)) {
+		if (sig_handler_ignored(sig_handler(t, sig), sig)) {
 			sigemptyset(&mask);
 			sigaddset(&mask, sig);
 			rm_from_queue_full(&mask, &t->signal->shared_pending);

From 445a91d2fe3667fb8fc251433645f686933cf56a Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:52 -0700
Subject: [PATCH 1058/1171] tracehook: tracehook_consider_fatal_signal

This defines tracehook_consider_fatal_signal() has a fine-grained hook for
deciding to skip the special cases for a fatal signal, as ptrace does.
There is no change, only cleanup.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/tracehook.h | 21 +++++++++++++++++++++
 kernel/signal.c           |  9 +++++----
 2 files changed, 26 insertions(+), 4 deletions(-)

diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
index 8cffd34f88d5..8b4c15e208fe 100644
--- a/include/linux/tracehook.h
+++ b/include/linux/tracehook.h
@@ -331,4 +331,25 @@ static inline int tracehook_consider_ignored_signal(struct task_struct *task,
 	return (task_ptrace(task) & PT_PTRACED) != 0;
 }
 
+/**
+ * tracehook_consider_fatal_signal - suppress special handling of fatal signal
+ * @task:		task receiving the signal
+ * @sig:		signal number being sent
+ * @handler:		%SIG_DFL or %SIG_IGN
+ *
+ * Return nonzero to prevent special handling of this termination signal.
+ * Normally @handler is %SIG_DFL.  It can be %SIG_IGN if @sig is ignored,
+ * in which case force_sig() is about to reset it to %SIG_DFL.
+ * When this returns zero, this signal might cause a quick termination
+ * that does not give the debugger a chance to intercept the signal.
+ *
+ * Called with or without @task->sighand->siglock held.
+ */
+static inline int tracehook_consider_fatal_signal(struct task_struct *task,
+						  int sig,
+						  void __user *handler)
+{
+	return (task_ptrace(task) & PT_PTRACED) != 0;
+}
+
 #endif	/* <linux/tracehook.h> */
diff --git a/kernel/signal.c b/kernel/signal.c
index 9efd1cee6d0b..1a942ce32ba0 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -300,12 +300,12 @@ flush_signal_handlers(struct task_struct *t, int force_default)
 
 int unhandled_signal(struct task_struct *tsk, int sig)
 {
+	void __user *handler = tsk->sighand->action[sig-1].sa.sa_handler;
 	if (is_global_init(tsk))
 		return 1;
-	if (tsk->ptrace & PT_PTRACED)
+	if (handler != SIG_IGN && handler != SIG_DFL)
 		return 0;
-	return (tsk->sighand->action[sig-1].sa.sa_handler == SIG_IGN) ||
-		(tsk->sighand->action[sig-1].sa.sa_handler == SIG_DFL);
+	return !tracehook_consider_fatal_signal(tsk, sig, handler);
 }
 
 
@@ -761,7 +761,8 @@ static void complete_signal(int sig, struct task_struct *p, int group)
 	if (sig_fatal(p, sig) &&
 	    !(signal->flags & (SIGNAL_UNKILLABLE | SIGNAL_GROUP_EXIT)) &&
 	    !sigismember(&t->real_blocked, sig) &&
-	    (sig == SIGKILL || !(t->ptrace & PT_PTRACED))) {
+	    (sig == SIGKILL ||
+	     !tracehook_consider_fatal_signal(t, sig, SIG_DFL))) {
 		/*
 		 * This signal will be fatal to the whole group.
 		 */

From 283d7559e7712f95a05331eb0a85394c6368101b Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:52 -0700
Subject: [PATCH 1059/1171] tracehook: syscall

This adds standard tracehook.h inlines for arch code to call when
TIF_SYSCALL_TRACE has been set.  This replaces having each arch implement
the ptrace guts for its syscall tracing support.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/tracehook.h | 70 +++++++++++++++++++++++++++++++++++++++
 1 file changed, 70 insertions(+)

diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
index 8b4c15e208fe..3548694a24db 100644
--- a/include/linux/tracehook.h
+++ b/include/linux/tracehook.h
@@ -66,6 +66,76 @@ static inline int tracehook_expect_breakpoints(struct task_struct *task)
 	return (task_ptrace(task) & PT_PTRACED) != 0;
 }
 
+/*
+ * ptrace report for syscall entry and exit looks identical.
+ */
+static inline void ptrace_report_syscall(struct pt_regs *regs)
+{
+	int ptrace = task_ptrace(current);
+
+	if (!(ptrace & PT_PTRACED))
+		return;
+
+	ptrace_notify(SIGTRAP | ((ptrace & PT_TRACESYSGOOD) ? 0x80 : 0));
+
+	/*
+	 * this isn't the same as continuing with a signal, but it will do
+	 * for normal use.  strace only continues with a signal if the
+	 * stopping signal is not SIGTRAP.  -brl
+	 */
+	if (current->exit_code) {
+		send_sig(current->exit_code, current, 1);
+		current->exit_code = 0;
+	}
+}
+
+/**
+ * tracehook_report_syscall_entry - task is about to attempt a system call
+ * @regs:		user register state of current task
+ *
+ * This will be called if %TIF_SYSCALL_TRACE has been set, when the
+ * current task has just entered the kernel for a system call.
+ * Full user register state is available here.  Changing the values
+ * in @regs can affect the system call number and arguments to be tried.
+ * It is safe to block here, preventing the system call from beginning.
+ *
+ * Returns zero normally, or nonzero if the calling arch code should abort
+ * the system call.  That must prevent normal entry so no system call is
+ * made.  If @task ever returns to user mode after this, its register state
+ * is unspecified, but should be something harmless like an %ENOSYS error
+ * return.
+ *
+ * Called without locks, just after entering kernel mode.
+ */
+static inline __must_check int tracehook_report_syscall_entry(
+	struct pt_regs *regs)
+{
+	ptrace_report_syscall(regs);
+	return 0;
+}
+
+/**
+ * tracehook_report_syscall_exit - task has just finished a system call
+ * @regs:		user register state of current task
+ * @step:		nonzero if simulating single-step or block-step
+ *
+ * This will be called if %TIF_SYSCALL_TRACE has been set, when the
+ * current task has just finished an attempted system call.  Full
+ * user register state is available here.  It is safe to block here,
+ * preventing signals from being processed.
+ *
+ * If @step is nonzero, this report is also in lieu of the normal
+ * trap that would follow the system call instruction because
+ * user_enable_block_step() or user_enable_single_step() was used.
+ * In this case, %TIF_SYSCALL_TRACE might not be set.
+ *
+ * Called without locks, just before checking for pending signals.
+ */
+static inline void tracehook_report_syscall_exit(struct pt_regs *regs, int step)
+{
+	ptrace_report_syscall(regs);
+}
+
 /**
  * tracehook_unsafe_exec - check for exec declared unsafe due to tracing
  * @task:		current task doing exec

From 7bcf6a2ca5f639b038c48711ebe6c4eca2036641 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:53 -0700
Subject: [PATCH 1060/1171] tracehook: get_signal_to_deliver

This defines the tracehook_get_signal() hook to allow tracing code to slip
in before normal signal dequeuing.  This lays the groundwork for new
tracing features that can inject synthetic signals outside the normal
queue or control the disposition of delivered signals.  The calling
convention lets tracehook_get_signal() decide both exactly what will
happen and what signal number to report in the handler/exit.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/tracehook.h | 29 +++++++++++++++++++++++++++++
 kernel/signal.c           | 38 +++++++++++++++++++++++++++-----------
 2 files changed, 56 insertions(+), 11 deletions(-)

diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
index 3548694a24db..42a0d7b11959 100644
--- a/include/linux/tracehook.h
+++ b/include/linux/tracehook.h
@@ -422,4 +422,33 @@ static inline int tracehook_consider_fatal_signal(struct task_struct *task,
 	return (task_ptrace(task) & PT_PTRACED) != 0;
 }
 
+/**
+ * tracehook_get_signal - deliver synthetic signal to traced task
+ * @task:		@current
+ * @regs:		task_pt_regs(@current)
+ * @info:		details of synthetic signal
+ * @return_ka:		sigaction for synthetic signal
+ *
+ * Return zero to check for a real pending signal normally.
+ * Return -1 after releasing the siglock to repeat the check.
+ * Return a signal number to induce an artifical signal delivery,
+ * setting *@info and *@return_ka to specify its details and behavior.
+ *
+ * The @return_ka->sa_handler value controls the disposition of the
+ * signal, no matter the signal number.  For %SIG_DFL, the return value
+ * is a representative signal to indicate the behavior (e.g. %SIGTERM
+ * for death, %SIGQUIT for core dump, %SIGSTOP for job control stop,
+ * %SIGTSTP for stop unless in an orphaned pgrp), but the signal number
+ * reported will be @info->si_signo instead.
+ *
+ * Called with @task->sighand->siglock held, before dequeuing pending signals.
+ */
+static inline int tracehook_get_signal(struct task_struct *task,
+				       struct pt_regs *regs,
+				       siginfo_t *info,
+				       struct k_sigaction *return_ka)
+{
+	return 0;
+}
+
 #endif	/* <linux/tracehook.h> */
diff --git a/kernel/signal.c b/kernel/signal.c
index 1a942ce32ba0..10b31ecdd9c8 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -1754,17 +1754,33 @@ relock:
 		    do_signal_stop(0))
 			goto relock;
 
-		signr = dequeue_signal(current, &current->blocked, info);
-		if (!signr)
-			break; /* will return 0 */
+		/*
+		 * Tracing can induce an artifical signal and choose sigaction.
+		 * The return value in @signr determines the default action,
+		 * but @info->si_signo is the signal number we will report.
+		 */
+		signr = tracehook_get_signal(current, regs, info, return_ka);
+		if (unlikely(signr < 0))
+			goto relock;
+		if (unlikely(signr != 0))
+			ka = return_ka;
+		else {
+			signr = dequeue_signal(current, &current->blocked,
+					       info);
 
-		if (signr != SIGKILL) {
-			signr = ptrace_signal(signr, info, regs, cookie);
 			if (!signr)
-				continue;
+				break; /* will return 0 */
+
+			if (signr != SIGKILL) {
+				signr = ptrace_signal(signr, info,
+						      regs, cookie);
+				if (!signr)
+					continue;
+			}
+
+			ka = &sighand->action[signr-1];
 		}
 
-		ka = &sighand->action[signr-1];
 		if (ka->sa.sa_handler == SIG_IGN) /* Do nothing.  */
 			continue;
 		if (ka->sa.sa_handler != SIG_DFL) {
@@ -1812,7 +1828,7 @@ relock:
 				spin_lock_irq(&sighand->siglock);
 			}
 
-			if (likely(do_signal_stop(signr))) {
+			if (likely(do_signal_stop(info->si_signo))) {
 				/* It released the siglock.  */
 				goto relock;
 			}
@@ -1833,7 +1849,7 @@ relock:
 
 		if (sig_kernel_coredump(signr)) {
 			if (print_fatal_signals)
-				print_fatal_signal(regs, signr);
+				print_fatal_signal(regs, info->si_signo);
 			/*
 			 * If it was able to dump core, this kills all
 			 * other threads in the group and synchronizes with
@@ -1842,13 +1858,13 @@ relock:
 			 * first and our do_group_exit call below will use
 			 * that value and ignore the one we pass it.
 			 */
-			do_coredump((long)signr, signr, regs);
+			do_coredump(info->si_signo, info->si_signo, regs);
 		}
 
 		/*
 		 * Death signals, no core dump.
 		 */
-		do_group_exit(signr);
+		do_group_exit(info->si_signo);
 		/* NOTREACHED */
 	}
 	spin_unlock_irq(&sighand->siglock);

From fa00b80b3c41a845b3d56f866fb40a2e98754c51 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:54 -0700
Subject: [PATCH 1061/1171] tracehook: job control

This defines the tracehook_notify_jctl() hook to formalize the ptrace
effects on the job control notifications.  There is no change, only
cleanup.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/tracehook.h | 20 ++++++++++++++++++++
 kernel/signal.c           | 10 +++++-----
 2 files changed, 25 insertions(+), 5 deletions(-)

diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
index 42a0d7b11959..6dc428dd2f38 100644
--- a/include/linux/tracehook.h
+++ b/include/linux/tracehook.h
@@ -451,4 +451,24 @@ static inline int tracehook_get_signal(struct task_struct *task,
 	return 0;
 }
 
+/**
+ * tracehook_notify_jctl - report about job control stop/continue
+ * @notify:		nonzero if this is the last thread in the group to stop
+ * @why:		%CLD_STOPPED or %CLD_CONTINUED
+ *
+ * This is called when we might call do_notify_parent_cldstop().
+ * It's called when about to stop for job control; we are already in
+ * %TASK_STOPPED state, about to call schedule().  It's also called when
+ * a delayed %CLD_STOPPED or %CLD_CONTINUED report is ready to be made.
+ *
+ * Return nonzero to generate a %SIGCHLD with @why, which is
+ * normal if @notify is nonzero.
+ *
+ * Called with no locks held.
+ */
+static inline int tracehook_notify_jctl(int notify, int why)
+{
+	return notify || (current->ptrace & PT_PTRACED);
+}
+
 #endif	/* <linux/tracehook.h> */
diff --git a/kernel/signal.c b/kernel/signal.c
index 10b31ecdd9c8..e9e699f4b1bd 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -596,9 +596,6 @@ static int check_kill_permission(int sig, struct siginfo *info,
 	return security_task_kill(t, info, sig, 0);
 }
 
-/* forward decl */
-static void do_notify_parent_cldstop(struct task_struct *tsk, int why);
-
 /*
  * Handle magic process-wide effects of stop/continue signals. Unlike
  * the signal actions, these happen immediately at signal-generation
@@ -1605,7 +1602,7 @@ finish_stop(int stop_count)
 	 * a group stop in progress and we are the last to stop,
 	 * report to the parent.  When ptraced, every thread reports itself.
 	 */
-	if (stop_count == 0 || (current->ptrace & PT_PTRACED)) {
+	if (tracehook_notify_jctl(stop_count == 0, CLD_STOPPED)) {
 		read_lock(&tasklist_lock);
 		do_notify_parent_cldstop(current, CLD_STOPPED);
 		read_unlock(&tasklist_lock);
@@ -1741,6 +1738,9 @@ relock:
 		signal->flags &= ~SIGNAL_CLD_MASK;
 		spin_unlock_irq(&sighand->siglock);
 
+		if (unlikely(!tracehook_notify_jctl(1, why)))
+			goto relock;
+
 		read_lock(&tasklist_lock);
 		do_notify_parent_cldstop(current->group_leader, why);
 		read_unlock(&tasklist_lock);
@@ -1906,7 +1906,7 @@ void exit_signals(struct task_struct *tsk)
 out:
 	spin_unlock_irq(&tsk->sighand->siglock);
 
-	if (unlikely(group_stop)) {
+	if (unlikely(group_stop) && tracehook_notify_jctl(1, CLD_STOPPED)) {
 		read_lock(&tasklist_lock);
 		do_notify_parent_cldstop(tsk, CLD_STOPPED);
 		read_unlock(&tasklist_lock);

From 2b2a1ff64afbadac842bbc58c5166962cf4f7664 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:54 -0700
Subject: [PATCH 1062/1171] tracehook: death

This moves the ptrace logic in task death (exit_notify) into tracehook.h
inlines.  Some code is rearranged slightly to make things nicer.  There is
no change, only cleanup.

There is one hook called with the tasklist_lock write-locked, as ptrace
needs.  There is also a new hook called after exit_state changes and
without locks.  This is a better place for tracing work to be in the
future, since it doesn't delay the whole system with locking.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/sched.h     |  2 +-
 include/linux/tracehook.h | 52 +++++++++++++++++++++++++++++++++++++++
 kernel/exit.c             | 26 +++++++-------------
 kernel/signal.c           | 10 +++++---
 4 files changed, 69 insertions(+), 21 deletions(-)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index adb8077dc463..a95d84d0da95 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1796,7 +1796,7 @@ extern int kill_pid_info_as_uid(int, struct siginfo *, struct pid *, uid_t, uid_
 extern int kill_pgrp(struct pid *pid, int sig, int priv);
 extern int kill_pid(struct pid *pid, int sig, int priv);
 extern int kill_proc_info(int, struct siginfo *, pid_t);
-extern void do_notify_parent(struct task_struct *, int);
+extern int do_notify_parent(struct task_struct *, int);
 extern void force_sig(int, struct task_struct *);
 extern void force_sig_specific(int, struct task_struct *);
 extern int send_sig(int, struct task_struct *, int);
diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
index 6dc428dd2f38..4c50e1b57349 100644
--- a/include/linux/tracehook.h
+++ b/include/linux/tracehook.h
@@ -471,4 +471,56 @@ static inline int tracehook_notify_jctl(int notify, int why)
 	return notify || (current->ptrace & PT_PTRACED);
 }
 
+/**
+ * tracehook_notify_death - task is dead, ready to notify parent
+ * @task:		@current task now exiting
+ * @death_cookie:	value to pass to tracehook_report_death()
+ * @group_dead:		nonzero if this was the last thread in the group to die
+ *
+ * Return the signal number to send our parent with do_notify_parent(), or
+ * zero to send no signal and leave a zombie, or -1 to self-reap right now.
+ *
+ * Called with write_lock_irq(&tasklist_lock) held.
+ */
+static inline int tracehook_notify_death(struct task_struct *task,
+					 void **death_cookie, int group_dead)
+{
+	if (task->exit_signal == -1)
+		return task->ptrace ? SIGCHLD : -1;
+
+	/*
+	 * If something other than our normal parent is ptracing us, then
+	 * send it a SIGCHLD instead of honoring exit_signal.  exit_signal
+	 * only has special meaning to our real parent.
+	 */
+	if (thread_group_empty(task) && !ptrace_reparented(task))
+		return task->exit_signal;
+
+	return task->ptrace ? SIGCHLD : 0;
+}
+
+/**
+ * tracehook_report_death - task is dead and ready to be reaped
+ * @task:		@current task now exiting
+ * @signal:		signal number sent to parent, or 0 or -1
+ * @death_cookie:	value passed back from tracehook_notify_death()
+ * @group_dead:		nonzero if this was the last thread in the group to die
+ *
+ * Thread has just become a zombie or is about to self-reap.  If positive,
+ * @signal is the signal number just sent to the parent (usually %SIGCHLD).
+ * If @signal is -1, this thread will self-reap.  If @signal is 0, this is
+ * a delayed_group_leader() zombie.  The @death_cookie was passed back by
+ * tracehook_notify_death().
+ *
+ * If normal reaping is not inhibited, @task->exit_state might be changing
+ * in parallel.
+ *
+ * Called without locks.
+ */
+static inline void tracehook_report_death(struct task_struct *task,
+					  int signal, void *death_cookie,
+					  int group_dead)
+{
+}
+
 #endif	/* <linux/tracehook.h> */
diff --git a/kernel/exit.c b/kernel/exit.c
index da28745f7c38..6cdf60712bd2 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -885,7 +885,8 @@ static void forget_original_parent(struct task_struct *father)
  */
 static void exit_notify(struct task_struct *tsk, int group_dead)
 {
-	int state;
+	int signal;
+	void *cookie;
 
 	/*
 	 * This does two things:
@@ -922,22 +923,11 @@ static void exit_notify(struct task_struct *tsk, int group_dead)
 	    !capable(CAP_KILL))
 		tsk->exit_signal = SIGCHLD;
 
-	/* If something other than our normal parent is ptracing us, then
-	 * send it a SIGCHLD instead of honoring exit_signal.  exit_signal
-	 * only has special meaning to our real parent.
-	 */
-	if (!task_detached(tsk) && thread_group_empty(tsk)) {
-		int signal = ptrace_reparented(tsk) ?
-				SIGCHLD : tsk->exit_signal;
-		do_notify_parent(tsk, signal);
-	} else if (tsk->ptrace) {
-		do_notify_parent(tsk, SIGCHLD);
-	}
+	signal = tracehook_notify_death(tsk, &cookie, group_dead);
+	if (signal > 0)
+		signal = do_notify_parent(tsk, signal);
 
-	state = EXIT_ZOMBIE;
-	if (task_detached(tsk) && likely(!tsk->ptrace))
-		state = EXIT_DEAD;
-	tsk->exit_state = state;
+	tsk->exit_state = signal < 0 ? EXIT_DEAD : EXIT_ZOMBIE;
 
 	/* mt-exec, de_thread() is waiting for us */
 	if (thread_group_leader(tsk) &&
@@ -947,8 +937,10 @@ static void exit_notify(struct task_struct *tsk, int group_dead)
 
 	write_unlock_irq(&tasklist_lock);
 
+	tracehook_report_death(tsk, signal, cookie, group_dead);
+
 	/* If the process is dead, release it - nobody will wait for it */
-	if (state == EXIT_DEAD)
+	if (signal < 0)
 		release_task(tsk);
 }
 
diff --git a/kernel/signal.c b/kernel/signal.c
index e9e699f4b1bd..0e862d3130ff 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -1326,9 +1326,11 @@ static inline void __wake_up_parent(struct task_struct *p,
 /*
  * Let a parent know about the death of a child.
  * For a stopped/continued status change, use do_notify_parent_cldstop instead.
+ *
+ * Returns -1 if our parent ignored us and so we've switched to
+ * self-reaping, or else @sig.
  */
-
-void do_notify_parent(struct task_struct *tsk, int sig)
+int do_notify_parent(struct task_struct *tsk, int sig)
 {
 	struct siginfo info;
 	unsigned long flags;
@@ -1399,12 +1401,14 @@ void do_notify_parent(struct task_struct *tsk, int sig)
 		 */
 		tsk->exit_signal = -1;
 		if (psig->action[SIGCHLD-1].sa.sa_handler == SIG_IGN)
-			sig = 0;
+			sig = -1;
 	}
 	if (valid_signal(sig) && sig > 0)
 		__group_send_sig_info(sig, &info, tsk->parent);
 	__wake_up_parent(tsk, tsk->parent);
 	spin_unlock_irqrestore(&psig->siglock, flags);
+
+	return sig;
 }
 
 static void do_notify_parent_cldstop(struct task_struct *tsk, int why)

From b787f7ba677840da16a2228c16571ce8a1fcb799 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:55 -0700
Subject: [PATCH 1063/1171] tracehook: force signal_pending()

This defines a new hook tracehook_force_sigpending() that lets tracing
code decide to force TIF_SIGPENDING on in recalc_sigpending().

This is not used yet, so it compiles away to nothing for now.  It lays the
groundwork for new tracing code that can interrupt a task synthetically
without actually sending a signal.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/tracehook.h | 14 ++++++++++++++
 kernel/signal.c           |  4 +++-
 2 files changed, 17 insertions(+), 1 deletion(-)

diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
index 4c50e1b57349..43bc51b6bd33 100644
--- a/include/linux/tracehook.h
+++ b/include/linux/tracehook.h
@@ -422,6 +422,20 @@ static inline int tracehook_consider_fatal_signal(struct task_struct *task,
 	return (task_ptrace(task) & PT_PTRACED) != 0;
 }
 
+/**
+ * tracehook_force_sigpending - let tracing force signal_pending(current) on
+ *
+ * Called when recomputing our signal_pending() flag.  Return nonzero
+ * to force the signal_pending() flag on, so that tracehook_get_signal()
+ * will be called before the next return to user mode.
+ *
+ * Called with @current->sighand->siglock held.
+ */
+static inline int tracehook_force_sigpending(void)
+{
+	return 0;
+}
+
 /**
  * tracehook_get_signal - deliver synthetic signal to traced task
  * @task:		@current
diff --git a/kernel/signal.c b/kernel/signal.c
index 0e862d3130ff..954f77d7e3bc 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -134,7 +134,9 @@ void recalc_sigpending_and_wake(struct task_struct *t)
 
 void recalc_sigpending(void)
 {
-	if (!recalc_sigpending_tsk(current) && !freezing(current))
+	if (unlikely(tracehook_force_sigpending()))
+		set_thread_flag(TIF_SIGPENDING);
+	else if (!recalc_sigpending_tsk(current) && !freezing(current))
 		clear_thread_flag(TIF_SIGPENDING);
 
 }

From 64b1208d5b0ef8859fd52ea7ae286a3eb994669b Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:56 -0700
Subject: [PATCH 1064/1171] tracehook: TIF_NOTIFY_RESUME

This adds tracehook.h inlines to enable a new arch feature in support of
user debugging/tracing.  This is not used yet, but it lays the groundwork
for a debugger to be able to wrangle a task that's possibly running,
without interrupting its syscalls in progress.

Each arch should define TIF_NOTIFY_RESUME, and in their entry.S code treat
it much like TIF_SIGPENDING.  That is, it causes you to take the slow path
when returning to user mode, where you get the full user-mode state
accessible as for signal handling or ptrace.  The arch code should check
TIF_NOTIFY_RESUME after handling TIF_SIGPENDING.  When it's set, clear it
and then call tracehook_notify_resume().

In future, tracing code will call set_notify_resume() when it wants to get
a callback in tracehook_notify_resume().

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/tracehook.h | 34 ++++++++++++++++++++++++++++++++++
 1 file changed, 34 insertions(+)

diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
index 43bc51b6bd33..32867ab86c70 100644
--- a/include/linux/tracehook.h
+++ b/include/linux/tracehook.h
@@ -537,4 +537,38 @@ static inline void tracehook_report_death(struct task_struct *task,
 {
 }
 
+#ifdef TIF_NOTIFY_RESUME
+/**
+ * set_notify_resume - cause tracehook_notify_resume() to be called
+ * @task:		task that will call tracehook_notify_resume()
+ *
+ * Calling this arranges that @task will call tracehook_notify_resume()
+ * before returning to user mode.  If it's already running in user mode,
+ * it will enter the kernel and call tracehook_notify_resume() soon.
+ * If it's blocked, it will not be woken.
+ */
+static inline void set_notify_resume(struct task_struct *task)
+{
+	if (!test_and_set_tsk_thread_flag(task, TIF_NOTIFY_RESUME))
+		kick_process(task);
+}
+
+/**
+ * tracehook_notify_resume - report when about to return to user mode
+ * @regs:		user-mode registers of @current task
+ *
+ * This is called when %TIF_NOTIFY_RESUME has been set.  Now we are
+ * about to return to user mode, and the user state in @regs can be
+ * inspected or adjusted.  The caller in arch code has cleared
+ * %TIF_NOTIFY_RESUME before the call.  If the flag gets set again
+ * asynchronously, this will be called again before we return to
+ * user mode.
+ *
+ * Called without locks.
+ */
+static inline void tracehook_notify_resume(struct pt_regs *regs)
+{
+}
+#endif	/* TIF_NOTIFY_RESUME */
+
 #endif	/* <linux/tracehook.h> */

From 828c365cc8b8d38c346fccb19fa80d28f2240831 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:57 -0700
Subject: [PATCH 1065/1171] tracehook: asm/syscall.h

This adds asm-generic/syscall.h, which documents what a real
asm-ARCH/syscall.h file should define.  This is not used yet, but will
provide all the machine-dependent details of examining a user system call
about to begin, in progress, or just ended.

Each arch should add an asm-ARCH/syscall.h that defines all the entry
points documented in asm-generic/syscall.h, as short inlines if possible.
This lets us write new tracing code that understands user system call
registers, without any new arch-specific work.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/asm-generic/syscall.h | 141 ++++++++++++++++++++++++++++++++++
 include/linux/tracehook.h     |   3 +-
 2 files changed, 143 insertions(+), 1 deletion(-)
 create mode 100644 include/asm-generic/syscall.h

diff --git a/include/asm-generic/syscall.h b/include/asm-generic/syscall.h
new file mode 100644
index 000000000000..abcf34c2fdc7
--- /dev/null
+++ b/include/asm-generic/syscall.h
@@ -0,0 +1,141 @@
+/*
+ * Access to user system call parameters and results
+ *
+ * Copyright (C) 2008 Red Hat, Inc.  All rights reserved.
+ *
+ * This copyrighted material is made available to anyone wishing to use,
+ * modify, copy, or redistribute it subject to the terms and conditions
+ * of the GNU General Public License v.2.
+ *
+ * This file is a stub providing documentation for what functions
+ * asm-ARCH/syscall.h files need to define.  Most arch definitions
+ * will be simple inlines.
+ *
+ * All of these functions expect to be called with no locks,
+ * and only when the caller is sure that the task of interest
+ * cannot return to user mode while we are looking at it.
+ */
+
+#ifndef _ASM_SYSCALL_H
+#define _ASM_SYSCALL_H	1
+
+struct task_struct;
+struct pt_regs;
+
+/**
+ * syscall_get_nr - find what system call a task is executing
+ * @task:	task of interest, must be blocked
+ * @regs:	task_pt_regs() of @task
+ *
+ * If @task is executing a system call or is at system call
+ * tracing about to attempt one, returns the system call number.
+ * If @task is not executing a system call, i.e. it's blocked
+ * inside the kernel for a fault or signal, returns -1.
+ *
+ * It's only valid to call this when @task is known to be blocked.
+ */
+long syscall_get_nr(struct task_struct *task, struct pt_regs *regs);
+
+/**
+ * syscall_rollback - roll back registers after an aborted system call
+ * @task:	task of interest, must be in system call exit tracing
+ * @regs:	task_pt_regs() of @task
+ *
+ * It's only valid to call this when @task is stopped for system
+ * call exit tracing (due to TIF_SYSCALL_TRACE or TIF_SYSCALL_AUDIT),
+ * after tracehook_report_syscall_entry() returned nonzero to prevent
+ * the system call from taking place.
+ *
+ * This rolls back the register state in @regs so it's as if the
+ * system call instruction was a no-op.  The registers containing
+ * the system call number and arguments are as they were before the
+ * system call instruction.  This may not be the same as what the
+ * register state looked like at system call entry tracing.
+ */
+void syscall_rollback(struct task_struct *task, struct pt_regs *regs);
+
+/**
+ * syscall_get_error - check result of traced system call
+ * @task:	task of interest, must be blocked
+ * @regs:	task_pt_regs() of @task
+ *
+ * Returns 0 if the system call succeeded, or -ERRORCODE if it failed.
+ *
+ * It's only valid to call this when @task is stopped for tracing on exit
+ * from a system call, due to %TIF_SYSCALL_TRACE or %TIF_SYSCALL_AUDIT.
+ */
+long syscall_get_error(struct task_struct *task, struct pt_regs *regs);
+
+/**
+ * syscall_get_return_value - get the return value of a traced system call
+ * @task:	task of interest, must be blocked
+ * @regs:	task_pt_regs() of @task
+ *
+ * Returns the return value of the successful system call.
+ * This value is meaningless if syscall_get_error() returned nonzero.
+ *
+ * It's only valid to call this when @task is stopped for tracing on exit
+ * from a system call, due to %TIF_SYSCALL_TRACE or %TIF_SYSCALL_AUDIT.
+ */
+long syscall_get_return_value(struct task_struct *task, struct pt_regs *regs);
+
+/**
+ * syscall_set_return_value - change the return value of a traced system call
+ * @task:	task of interest, must be blocked
+ * @regs:	task_pt_regs() of @task
+ * @error:	negative error code, or zero to indicate success
+ * @val:	user return value if @error is zero
+ *
+ * This changes the results of the system call that user mode will see.
+ * If @error is zero, the user sees a successful system call with a
+ * return value of @val.  If @error is nonzero, it's a negated errno
+ * code; the user sees a failed system call with this errno code.
+ *
+ * It's only valid to call this when @task is stopped for tracing on exit
+ * from a system call, due to %TIF_SYSCALL_TRACE or %TIF_SYSCALL_AUDIT.
+ */
+void syscall_set_return_value(struct task_struct *task, struct pt_regs *regs,
+			      int error, long val);
+
+/**
+ * syscall_get_arguments - extract system call parameter values
+ * @task:	task of interest, must be blocked
+ * @regs:	task_pt_regs() of @task
+ * @i:		argument index [0,5]
+ * @n:		number of arguments; n+i must be [1,6].
+ * @args:	array filled with argument values
+ *
+ * Fetches @n arguments to the system call starting with the @i'th argument
+ * (from 0 through 5).  Argument @i is stored in @args[0], and so on.
+ * An arch inline version is probably optimal when @i and @n are constants.
+ *
+ * It's only valid to call this when @task is stopped for tracing on
+ * entry to a system call, due to %TIF_SYSCALL_TRACE or %TIF_SYSCALL_AUDIT.
+ * It's invalid to call this with @i + @n > 6; we only support system calls
+ * taking up to 6 arguments.
+ */
+void syscall_get_arguments(struct task_struct *task, struct pt_regs *regs,
+			   unsigned int i, unsigned int n, unsigned long *args);
+
+/**
+ * syscall_set_arguments - change system call parameter value
+ * @task:	task of interest, must be in system call entry tracing
+ * @regs:	task_pt_regs() of @task
+ * @i:		argument index [0,5]
+ * @n:		number of arguments; n+i must be [1,6].
+ * @args:	array of argument values to store
+ *
+ * Changes @n arguments to the system call starting with the @i'th argument.
+ * @n'th argument to @val.  Argument @i gets value @args[0], and so on.
+ * An arch inline version is probably optimal when @i and @n are constants.
+ *
+ * It's only valid to call this when @task is stopped for tracing on
+ * entry to a system call, due to %TIF_SYSCALL_TRACE or %TIF_SYSCALL_AUDIT.
+ * It's invalid to call this with @i + @n > 6; we only support system calls
+ * taking up to 6 arguments.
+ */
+void syscall_set_arguments(struct task_struct *task, struct pt_regs *regs,
+			   unsigned int i, unsigned int n,
+			   const unsigned long *args);
+
+#endif	/* _ASM_SYSCALL_H */
diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
index 32867ab86c70..589f429619c9 100644
--- a/include/linux/tracehook.h
+++ b/include/linux/tracehook.h
@@ -103,7 +103,8 @@ static inline void ptrace_report_syscall(struct pt_regs *regs)
  * the system call.  That must prevent normal entry so no system call is
  * made.  If @task ever returns to user mode after this, its register state
  * is unspecified, but should be something harmless like an %ENOSYS error
- * return.
+ * return.  It should preserve enough information so that syscall_rollback()
+ * can work (see asm-generic/syscall.h).
  *
  * Called without locks, just after entering kernel mode.
  */

From 1f5a4ad97a0e7b663d527ecc02aabe203d000d91 Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:57 -0700
Subject: [PATCH 1066/1171] tracehook: CONFIG_HAVE_ARCH_TRACEHOOK

This adds the generic HAVE_ARCH_TRACEHOOK kconfig item.  Each arch should
add to some Kconfig file:

	select HAVE_ARCH_TRACEHOOK

if the arch code uses all the latest hooks to enable newfangled tracing
and debugging code.  The comment in arch/Kconfig lists all the
prerequisite arch support.  When all these are available, setting
HAVE_ARCH_TRACEHOOK will allow enabling any new features that depend on
the modern arch interfaces.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/Kconfig | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/arch/Kconfig b/arch/Kconfig
index b0fabfa864ff..364c6dadde0a 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -59,6 +59,24 @@ config HAVE_KPROBES
 config HAVE_KRETPROBES
 	def_bool n
 
+#
+# An arch should select this if it provides all these things:
+#
+#	task_pt_regs()		in asm/processor.h or asm/ptrace.h
+#	arch_has_single_step()	if there is hardware single-step support
+#	arch_has_block_step()	if there is hardware block-step support
+#	arch_ptrace()		and not #define __ARCH_SYS_PTRACE
+#	compat_arch_ptrace()	and #define __ARCH_WANT_COMPAT_SYS_PTRACE
+#	asm/syscall.h		supplying asm-generic/syscall.h interface
+#	linux/regset.h		user_regset interfaces
+#	CORE_DUMP_USE_REGSET	#define'd in linux/elf.h
+#	TIF_SYSCALL_TRACE	calls tracehook_report_syscall_{entry,exit}
+#	TIF_NOTIFY_RESUME	calls tracehook_notify_resume()
+#	signal delivery		calls tracehook_signal_handler()
+#
+config HAVE_ARCH_TRACEHOOK
+	def_bool n
+
 config HAVE_DMA_ATTRS
 	def_bool n
 

From 85ba2d862e521375a8ee01526c5c46b1f24bb4af Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:58 -0700
Subject: [PATCH 1067/1171] tracehook: wait_task_inactive

This extends wait_task_inactive() with a new argument so it can be used in
a "soft" mode where it will check for the task changing state unexpectedly
and back off.  There is no change to existing callers.  This lays the
groundwork to allow robust, noninvasive tracing that can try to sample a
blocked thread but back off safely if it wakes up.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/ia64/kernel/perfmon.c |  4 ++--
 include/linux/sched.h      |  8 ++++++--
 kernel/kthread.c           |  2 +-
 kernel/ptrace.c            |  2 +-
 kernel/sched.c             | 29 +++++++++++++++++++++++++++--
 5 files changed, 37 insertions(+), 8 deletions(-)

diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c
index 19d4493c6193..fc8f3509df27 100644
--- a/arch/ia64/kernel/perfmon.c
+++ b/arch/ia64/kernel/perfmon.c
@@ -2626,7 +2626,7 @@ pfm_task_incompatible(pfm_context_t *ctx, struct task_struct *task)
 	/*
 	 * make sure the task is off any CPU
 	 */
-	wait_task_inactive(task);
+	wait_task_inactive(task, 0);
 
 	/* more to come... */
 
@@ -4774,7 +4774,7 @@ recheck:
 
 		UNPROTECT_CTX(ctx, flags);
 
-		wait_task_inactive(task);
+		wait_task_inactive(task, 0);
 
 		PROTECT_CTX(ctx, flags);
 
diff --git a/include/linux/sched.h b/include/linux/sched.h
index a95d84d0da95..f59318a0099b 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1882,9 +1882,13 @@ extern void set_task_comm(struct task_struct *tsk, char *from);
 extern char *get_task_comm(char *to, struct task_struct *tsk);
 
 #ifdef CONFIG_SMP
-extern void wait_task_inactive(struct task_struct * p);
+extern unsigned long wait_task_inactive(struct task_struct *, long match_state);
 #else
-#define wait_task_inactive(p)	do { } while (0)
+static inline unsigned long wait_task_inactive(struct task_struct *p,
+					       long match_state)
+{
+	return 1;
+}
 #endif
 
 #define next_task(p)	list_entry(rcu_dereference((p)->tasks.next), struct task_struct, tasks)
diff --git a/kernel/kthread.c b/kernel/kthread.c
index 6111c27491b1..96cff2f8710b 100644
--- a/kernel/kthread.c
+++ b/kernel/kthread.c
@@ -176,7 +176,7 @@ void kthread_bind(struct task_struct *k, unsigned int cpu)
 		return;
 	}
 	/* Must have done schedule() in kthread() before we set_task_cpu */
-	wait_task_inactive(k);
+	wait_task_inactive(k, 0);
 	set_task_cpu(k, cpu);
 	k->cpus_allowed = cpumask_of_cpu(cpu);
 	k->rt.nr_cpus_allowed = 1;
diff --git a/kernel/ptrace.c b/kernel/ptrace.c
index 8392a9da6450..082b3fcb32a0 100644
--- a/kernel/ptrace.c
+++ b/kernel/ptrace.c
@@ -107,7 +107,7 @@ int ptrace_check_attach(struct task_struct *child, int kill)
 	read_unlock(&tasklist_lock);
 
 	if (!ret && !kill)
-		wait_task_inactive(child);
+		ret = wait_task_inactive(child, TASK_TRACED) ? 0 : -ESRCH;
 
 	/* All systems go.. */
 	return ret;
diff --git a/kernel/sched.c b/kernel/sched.c
index fde1a1026359..0236958addcb 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -1867,16 +1867,24 @@ migrate_task(struct task_struct *p, int dest_cpu, struct migration_req *req)
 /*
  * wait_task_inactive - wait for a thread to unschedule.
  *
+ * If @match_state is nonzero, it's the @p->state value just checked and
+ * not expected to change.  If it changes, i.e. @p might have woken up,
+ * then return zero.  When we succeed in waiting for @p to be off its CPU,
+ * we return a positive number (its total switch count).  If a second call
+ * a short while later returns the same number, the caller can be sure that
+ * @p has remained unscheduled the whole time.
+ *
  * The caller must ensure that the task *will* unschedule sometime soon,
  * else this function might spin for a *long* time. This function can't
  * be called with interrupts off, or it may introduce deadlock with
  * smp_call_function() if an IPI is sent by the same process we are
  * waiting to become inactive.
  */
-void wait_task_inactive(struct task_struct *p)
+unsigned long wait_task_inactive(struct task_struct *p, long match_state)
 {
 	unsigned long flags;
 	int running, on_rq;
+	unsigned long ncsw;
 	struct rq *rq;
 
 	for (;;) {
@@ -1899,8 +1907,11 @@ void wait_task_inactive(struct task_struct *p)
 		 * return false if the runqueue has changed and p
 		 * is actually now running somewhere else!
 		 */
-		while (task_running(rq, p))
+		while (task_running(rq, p)) {
+			if (match_state && unlikely(p->state != match_state))
+				return 0;
 			cpu_relax();
+		}
 
 		/*
 		 * Ok, time to look more closely! We need the rq
@@ -1910,8 +1921,20 @@ void wait_task_inactive(struct task_struct *p)
 		rq = task_rq_lock(p, &flags);
 		running = task_running(rq, p);
 		on_rq = p->se.on_rq;
+		ncsw = 0;
+		if (!match_state || p->state == match_state) {
+			ncsw = p->nivcsw + p->nvcsw;
+			if (unlikely(!ncsw))
+				ncsw = 1;
+		}
 		task_rq_unlock(rq, &flags);
 
+		/*
+		 * If it changed from the expected state, bail out now.
+		 */
+		if (unlikely(!ncsw))
+			break;
+
 		/*
 		 * Was it really running after all now that we
 		 * checked with the proper locks actually held?
@@ -1944,6 +1967,8 @@ void wait_task_inactive(struct task_struct *p)
 		 */
 		break;
 	}
+
+	return ncsw;
 }
 
 /***

From bbc698636ed48b6fcd323964e0f847a6a796325d Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:45:59 -0700
Subject: [PATCH 1068/1171] task_current_syscall

This adds the new function task_current_syscall() on machines where the
asm/syscall.h interface is supported (CONFIG_HAVE_ARCH_TRACEHOOK).  It's
exported for modules to use in the future.  This function safely samples
the state of a blocked thread to collect what system call it is blocked
in, and the six system call argument registers.

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/ptrace.h |  4 +++
 lib/Makefile           |  2 ++
 lib/syscall.c          | 75 ++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 81 insertions(+)
 create mode 100644 lib/syscall.c

diff --git a/include/linux/ptrace.h b/include/linux/ptrace.h
index ed69c03692d9..fd31756e1a00 100644
--- a/include/linux/ptrace.h
+++ b/include/linux/ptrace.h
@@ -314,6 +314,10 @@ static inline void user_enable_block_step(struct task_struct *task)
 #define arch_ptrace_stop(code, info)		do { } while (0)
 #endif
 
+extern int task_current_syscall(struct task_struct *target, long *callno,
+				unsigned long args[6], unsigned int maxargs,
+				unsigned long *sp, unsigned long *pc);
+
 #endif
 
 #endif
diff --git a/lib/Makefile b/lib/Makefile
index 9085ad6fa53d..942c7250f603 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -78,6 +78,8 @@ lib-$(CONFIG_GENERIC_BUG) += bug.o
 
 obj-$(CONFIG_HAVE_LMB) += lmb.o
 
+obj-$(CONFIG_HAVE_ARCH_TRACEHOOK) += syscall.o
+
 hostprogs-y	:= gen_crc32table
 clean-files	:= crc32table.h
 
diff --git a/lib/syscall.c b/lib/syscall.c
new file mode 100644
index 000000000000..a4f7067f72fa
--- /dev/null
+++ b/lib/syscall.c
@@ -0,0 +1,75 @@
+#include <linux/ptrace.h>
+#include <linux/sched.h>
+#include <linux/module.h>
+#include <asm/syscall.h>
+
+static int collect_syscall(struct task_struct *target, long *callno,
+			   unsigned long args[6], unsigned int maxargs,
+			   unsigned long *sp, unsigned long *pc)
+{
+	struct pt_regs *regs = task_pt_regs(target);
+	if (unlikely(!regs))
+		return -EAGAIN;
+
+	*sp = user_stack_pointer(regs);
+	*pc = instruction_pointer(regs);
+
+	*callno = syscall_get_nr(target, regs);
+	if (*callno != -1L && maxargs > 0)
+		syscall_get_arguments(target, regs, 0, maxargs, args);
+
+	return 0;
+}
+
+/**
+ * task_current_syscall - Discover what a blocked task is doing.
+ * @target:		thread to examine
+ * @callno:		filled with system call number or -1
+ * @args:		filled with @maxargs system call arguments
+ * @maxargs:		number of elements in @args to fill
+ * @sp:			filled with user stack pointer
+ * @pc:			filled with user PC
+ *
+ * If @target is blocked in a system call, returns zero with *@callno
+ * set to the the call's number and @args filled in with its arguments.
+ * Registers not used for system call arguments may not be available and
+ * it is not kosher to use &struct user_regset calls while the system
+ * call is still in progress.  Note we may get this result if @target
+ * has finished its system call but not yet returned to user mode, such
+ * as when it's stopped for signal handling or syscall exit tracing.
+ *
+ * If @target is blocked in the kernel during a fault or exception,
+ * returns zero with *@callno set to -1 and does not fill in @args.
+ * If so, it's now safe to examine @target using &struct user_regset
+ * get() calls as long as we're sure @target won't return to user mode.
+ *
+ * Returns -%EAGAIN if @target does not remain blocked.
+ *
+ * Returns -%EINVAL if @maxargs is too large (maximum is six).
+ */
+int task_current_syscall(struct task_struct *target, long *callno,
+			 unsigned long args[6], unsigned int maxargs,
+			 unsigned long *sp, unsigned long *pc)
+{
+	long state;
+	unsigned long ncsw;
+
+	if (unlikely(maxargs > 6))
+		return -EINVAL;
+
+	if (target == current)
+		return collect_syscall(target, callno, args, maxargs, sp, pc);
+
+	state = target->state;
+	if (unlikely(!state))
+		return -EAGAIN;
+
+	ncsw = wait_task_inactive(target, state);
+	if (unlikely(!ncsw) ||
+	    unlikely(collect_syscall(target, callno, args, maxargs, sp, pc)) ||
+	    unlikely(wait_task_inactive(target, state) != ncsw))
+		return -EAGAIN;
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(task_current_syscall);

From ebcb67341fee34061430f3367f2e507e52ee051b Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Fri, 25 Jul 2008 19:46:00 -0700
Subject: [PATCH 1069/1171] /proc/PID/syscall

This adds /proc/PID/syscall and /proc/PID/task/TID/syscall magic files.
These use task_current_syscall() to show the task's current system call
number and argument registers, stack pointer and PC.  For a task blocked
but not in a syscall, the file shows "-1" in place of the syscall number,
followed by only the SP and PC.  For a task that's not blocked, it shows
"running".

Signed-off-by: Roland McGrath <roland@redhat.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/proc/base.c | 26 ++++++++++++++++++++++++++
 1 file changed, 26 insertions(+)

diff --git a/fs/proc/base.c b/fs/proc/base.c
index 4b74dba69a6d..81bce6791bfc 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -509,6 +509,26 @@ static int proc_pid_limits(struct task_struct *task, char *buffer)
 	return count;
 }
 
+#ifdef CONFIG_HAVE_ARCH_TRACEHOOK
+static int proc_pid_syscall(struct task_struct *task, char *buffer)
+{
+	long nr;
+	unsigned long args[6], sp, pc;
+
+	if (task_current_syscall(task, &nr, args, 6, &sp, &pc))
+		return sprintf(buffer, "running\n");
+
+	if (nr < 0)
+		return sprintf(buffer, "%ld 0x%lx 0x%lx\n", nr, sp, pc);
+
+	return sprintf(buffer,
+		       "%ld 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx 0x%lx\n",
+		       nr,
+		       args[0], args[1], args[2], args[3], args[4], args[5],
+		       sp, pc);
+}
+#endif /* CONFIG_HAVE_ARCH_TRACEHOOK */
+
 /************************************************************************/
 /*                       Here the fs part begins                        */
 /************************************************************************/
@@ -2477,6 +2497,9 @@ static const struct pid_entry tgid_base_stuff[] = {
 	INF("limits",	  S_IRUSR, pid_limits),
 #ifdef CONFIG_SCHED_DEBUG
 	REG("sched",      S_IRUGO|S_IWUSR, pid_sched),
+#endif
+#ifdef CONFIG_HAVE_ARCH_TRACEHOOK
+	INF("syscall",    S_IRUSR, pid_syscall),
 #endif
 	INF("cmdline",    S_IRUGO, pid_cmdline),
 	ONE("stat",       S_IRUGO, tgid_stat),
@@ -2809,6 +2832,9 @@ static const struct pid_entry tid_base_stuff[] = {
 	INF("limits",	 S_IRUSR, pid_limits),
 #ifdef CONFIG_SCHED_DEBUG
 	REG("sched",     S_IRUGO|S_IWUSR, pid_sched),
+#endif
+#ifdef CONFIG_HAVE_ARCH_TRACEHOOK
+	INF("syscall",   S_IRUSR, pid_syscall),
 #endif
 	INF("cmdline",   S_IRUGO, pid_cmdline),
 	ONE("stat",      S_IRUGO, tid_stat),

From 2c97b7fc0d8c8661981beb9517da342ced3b3bc7 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Fri, 25 Jul 2008 19:46:01 -0700
Subject: [PATCH 1070/1171] mm: print swapcache page count in
 show_swap_cache_info()

Every arch implements its own show_mem() function.  Most of them share
quite some code, some of them are completely identical.

This series implements a generic version of this function and migrates
almost all architectures to it.

This patch:

Most show_mem() implementations calculate the amount of pages within
the swapcache every time.  Move the output to a more appropriate place
and use the anyway available total_swapcache_pages variable.

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Haavard Skinnemoen <hskinnemoen@atmel.com>
Cc: Bryan Wu <cooloney@kernel.org>
Cc: Chris Zankel <chris@zankel.net>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Greg Ungerer <gerg@uclinux.org>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Roman Zippel <zippel@linux-m68k.org>
Cc: Hirokazu Takata <takata@linux-m32r.org>
Cc: Mikael Starvik <starvik@axis.com>
Cc: Hugh Dickins <hugh@veritas.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/swap_state.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/mm/swap_state.c b/mm/swap_state.c
index 2c217e33d497..b8035b055129 100644
--- a/mm/swap_state.c
+++ b/mm/swap_state.c
@@ -56,7 +56,8 @@ static struct {
 
 void show_swap_cache_info(void)
 {
-	printk("Swap cache: add %lu, delete %lu, find %lu/%lu\n",
+	printk("%lu pages in swap cache\n", total_swapcache_pages);
+	printk("Swap cache stats: add %lu, delete %lu, find %lu/%lu\n",
 		swap_cache_info.add_total, swap_cache_info.del_total,
 		swap_cache_info.find_success, swap_cache_info.find_total);
 	printk("Free swap  = %lukB\n", nr_swap_pages << (PAGE_SHIFT - 10));

From 454c63b02e530f10b4345343f63596dd705888d0 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Fri, 25 Jul 2008 19:46:07 -0700
Subject: [PATCH 1071/1171] lib: generic show_mem()

This implements a platform-independent version of show_mem().

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Haavard Skinnemoen <hskinnemoen@atmel.com>
Cc: Bryan Wu <cooloney@kernel.org>
Cc: Chris Zankel <chris@zankel.net>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jeff Dike <jdike@addtoit.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Paul Mundt <lethal@linux-sh.org>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Greg Ungerer <gerg@uclinux.org>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Roman Zippel <zippel@linux-m68k.org>
Cc: Hirokazu Takata <takata@linux-m32r.org>
Cc: Mikael Starvik <starvik@axis.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 lib/Makefile   |  2 +-
 lib/show_mem.c | 63 ++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 64 insertions(+), 1 deletion(-)
 create mode 100644 lib/show_mem.c

diff --git a/lib/Makefile b/lib/Makefile
index 942c7250f603..3b1f94bbe9de 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -11,7 +11,7 @@ lib-y := ctype.o string.o vsprintf.o cmdline.o \
 	 rbtree.o radix-tree.o dump_stack.o \
 	 idr.o int_sqrt.o extable.o prio_tree.o \
 	 sha1.o irq_regs.o reciprocal_div.o argv_split.o \
-	 proportions.o prio_heap.o ratelimit.o
+	 proportions.o prio_heap.o ratelimit.o show_mem.o
 
 lib-$(CONFIG_MMU) += ioremap.o
 lib-$(CONFIG_SMP) += cpumask.o
diff --git a/lib/show_mem.c b/lib/show_mem.c
new file mode 100644
index 000000000000..238e72a18ce1
--- /dev/null
+++ b/lib/show_mem.c
@@ -0,0 +1,63 @@
+/*
+ * Generic show_mem() implementation
+ *
+ * Copyright (C) 2008 Johannes Weiner <hannes@saeurebad.de>
+ * All code subject to the GPL version 2.
+ */
+
+#include <linux/mm.h>
+#include <linux/nmi.h>
+#include <linux/quicklist.h>
+
+void show_mem(void)
+{
+	pg_data_t *pgdat;
+	unsigned long total = 0, reserved = 0, shared = 0,
+		nonshared = 0, highmem = 0;
+
+	printk(KERN_INFO "Mem-Info:\n");
+	show_free_areas();
+
+	for_each_online_pgdat(pgdat) {
+		unsigned long i, flags;
+
+		pgdat_resize_lock(pgdat, &flags);
+		for (i = 0; i < pgdat->node_spanned_pages; i++) {
+			struct page *page;
+			unsigned long pfn = pgdat->node_start_pfn + i;
+
+			if (unlikely(!(i % MAX_ORDER_NR_PAGES)))
+				touch_nmi_watchdog();
+
+			if (!pfn_valid(pfn))
+				continue;
+
+			page = pfn_to_page(pfn);
+
+			if (PageHighMem(page))
+				highmem++;
+
+			if (PageReserved(page))
+				reserved++;
+			else if (page_count(page) == 1)
+				nonshared++;
+			else if (page_count(page) > 1)
+				shared += page_count(page) - 1;
+
+			total++;
+		}
+		pgdat_resize_unlock(pgdat, &flags);
+	}
+
+	printk(KERN_INFO "%lu pages RAM\n", total);
+#ifdef CONFIG_HIGHMEM
+	printk(KERN_INFO "%lu pages HighMem\n", highmem);
+#endif
+	printk(KERN_INFO "%lu pages reserved\n", reserved);
+	printk(KERN_INFO "%lu pages shared\n", shared);
+	printk(KERN_INFO "%lu pages non-shared\n", nonshared);
+#ifdef CONFIG_QUICKLIST
+	printk(KERN_INFO "%lu pages in pagetable cache\n",
+		quicklist_total_size());
+#endif
+}

From 5f834b9f3bd3769162fc169efd7ecae0712a2d1d Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Fri, 25 Jul 2008 19:46:09 -0700
Subject: [PATCH 1072/1171] alpha: use generic show_mem()

Remove arch-specific show_mem() in favor of the generic version.

This also removes the following redundant information display:

	- free pages, printed by show_free_areas()
	- free swap pages, printed by show_swap_cache_info()
	- pages in swapcache, printed by show_swap_cache_info()

where show_mem() calls show_free_areas(), which calls
show_swap_cache_info().

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/alpha/mm/init.c | 30 ------------------------------
 arch/alpha/mm/numa.c | 35 -----------------------------------
 2 files changed, 65 deletions(-)

diff --git a/arch/alpha/mm/init.c b/arch/alpha/mm/init.c
index 40c15e7301de..234e42b8ee74 100644
--- a/arch/alpha/mm/init.c
+++ b/arch/alpha/mm/init.c
@@ -94,36 +94,6 @@ __bad_page(void)
 	return pte_mkdirty(mk_pte(virt_to_page(EMPTY_PGE), PAGE_SHARED));
 }
 
-#ifndef CONFIG_DISCONTIGMEM
-void
-show_mem(void)
-{
-	long i,free = 0,total = 0,reserved = 0;
-	long shared = 0, cached = 0;
-
-	printk("\nMem-info:\n");
-	show_free_areas();
-	printk("Free swap:       %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
-	i = max_mapnr;
-	while (i-- > 0) {
-		total++;
-		if (PageReserved(mem_map+i))
-			reserved++;
-		else if (PageSwapCache(mem_map+i))
-			cached++;
-		else if (!page_count(mem_map+i))
-			free++;
-		else
-			shared += page_count(mem_map + i) - 1;
-	}
-	printk("%ld pages of RAM\n",total);
-	printk("%ld free pages\n",free);
-	printk("%ld reserved pages\n",reserved);
-	printk("%ld pages shared\n",shared);
-	printk("%ld pages swap cached\n",cached);
-}
-#endif
-
 static inline unsigned long
 load_PCB(struct pcb_struct *pcb)
 {
diff --git a/arch/alpha/mm/numa.c b/arch/alpha/mm/numa.c
index d8c4ceaf00b9..a13de49d1265 100644
--- a/arch/alpha/mm/numa.c
+++ b/arch/alpha/mm/numa.c
@@ -359,38 +359,3 @@ void __init mem_init(void)
 	mem_stress();
 #endif
 }
-
-void
-show_mem(void)
-{
-	long i,free = 0,total = 0,reserved = 0;
-	long shared = 0, cached = 0;
-	int nid;
-
-	printk("\nMem-info:\n");
-	show_free_areas();
-	printk("Free swap:       %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
-	for_each_online_node(nid) {
-		unsigned long flags;
-		pgdat_resize_lock(NODE_DATA(nid), &flags);
-		i = node_spanned_pages(nid);
-		while (i-- > 0) {
-			struct page *page = nid_page_nr(nid, i);
-			total++;
-			if (PageReserved(page))
-				reserved++;
-			else if (PageSwapCache(page))
-				cached++;
-			else if (!page_count(page))
-				free++;
-			else
-				shared += page_count(page) - 1;
-		}
-		pgdat_resize_unlock(NODE_DATA(nid), &flags);
-	}
-	printk("%ld pages of RAM\n",total);
-	printk("%ld free pages\n",free);
-	printk("%ld reserved pages\n",reserved);
-	printk("%ld pages shared\n",shared);
-	printk("%ld pages swap cached\n",cached);
-}

From 46244a2b3c1dfd83041bcf0428c584d3a66b30cc Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Fri, 25 Jul 2008 19:46:09 -0700
Subject: [PATCH 1073/1171] avr32: use generic show_mem()

Remove arch-specific show_mem() in favor of the generic version.

This also removes the following redundant information display:

	- free pages, printed by show_free_areas()
	- pages in slabs, printed by show_free_areas()
	- pages in swapcache, printed by show_swap_cache_info()

where show_mem() calls show_free_areas(), which calls
show_swap_cache_info().

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Acked-by: Haavard Skinnemoen <hskinnemoen@atmel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/avr32/mm/init.c | 39 ---------------------------------------
 1 file changed, 39 deletions(-)

diff --git a/arch/avr32/mm/init.c b/arch/avr32/mm/init.c
index 3c85fdaa9487..fa92ff6d95f7 100644
--- a/arch/avr32/mm/init.c
+++ b/arch/avr32/mm/init.c
@@ -38,45 +38,6 @@ EXPORT_SYMBOL(empty_zero_page);
  */
 unsigned long mmu_context_cache = NO_CONTEXT;
 
-void show_mem(void)
-{
-	int total = 0, reserved = 0, cached = 0;
-	int slab = 0, free = 0, shared = 0;
-	pg_data_t *pgdat;
-
-	printk("Mem-info:\n");
-	show_free_areas();
-
-	for_each_online_pgdat(pgdat) {
-		struct page *page, *end;
-
-		page = pgdat->node_mem_map;
-		end = page + pgdat->node_spanned_pages;
-
-		do {
-			total++;
-			if (PageReserved(page))
-				reserved++;
-			else if (PageSwapCache(page))
-				cached++;
-			else if (PageSlab(page))
-				slab++;
-			else if (!page_count(page))
-				free++;
-			else
-				shared += page_count(page) - 1;
-			page++;
-		} while (page < end);
-	}
-
-	printk ("%d pages of RAM\n", total);
-	printk ("%d free pages\n", free);
-	printk ("%d reserved pages\n", reserved);
-	printk ("%d slab pages\n", slab);
-	printk ("%d pages shared\n", shared);
-	printk ("%d pages swap cached\n", cached);
-}
-
 /*
  * paging_init() sets up the page tables
  *

From a2e9c44337b53aef37f3382920a478cac3777ba9 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Fri, 25 Jul 2008 19:46:10 -0700
Subject: [PATCH 1074/1171] blackfin: use generic show_mem()

Remove arch-specific show_mem() in favor of the generic version.

This also removes the following redundant information display:

	- free pages, printed by show_free_areas()
	- pages in swapcache, printed by show_swap_cache_info()

where show_mem() calls show_free_areas(), which calls
show_swap_cache_info().

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Acked-by: Bryan Wu <cooloney@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/blackfin/mm/init.c | 27 ---------------------------
 1 file changed, 27 deletions(-)

diff --git a/arch/blackfin/mm/init.c b/arch/blackfin/mm/init.c
index ec3141fefd20..4d5326ee9a85 100644
--- a/arch/blackfin/mm/init.c
+++ b/arch/blackfin/mm/init.c
@@ -53,33 +53,6 @@ static unsigned long empty_bad_page;
 
 unsigned long empty_zero_page;
 
-void show_mem(void)
-{
-	unsigned long i;
-	int free = 0, total = 0, reserved = 0, shared = 0;
-
-	int cached = 0;
-	printk(KERN_INFO "Mem-info:\n");
-	show_free_areas();
-	i = max_mapnr;
-	while (i-- > 0) {
-		total++;
-		if (PageReserved(mem_map + i))
-			reserved++;
-		else if (PageSwapCache(mem_map + i))
-			cached++;
-		else if (!page_count(mem_map + i))
-			free++;
-		else
-			shared += page_count(mem_map + i) - 1;
-	}
-	printk(KERN_INFO "%d pages of RAM\n", total);
-	printk(KERN_INFO "%d free pages\n", free);
-	printk(KERN_INFO "%d reserved pages\n", reserved);
-	printk(KERN_INFO "%d pages shared\n", shared);
-	printk(KERN_INFO "%d pages swap cached\n", cached);
-}
-
 /*
  * paging_init() continues the virtual memory environment setup which
  * was begun by the code in arch/head.S.

From 47221222a59a565e11954c078a2cf6a07a7e690e Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Fri, 25 Jul 2008 19:46:10 -0700
Subject: [PATCH 1075/1171] xtensa: use generic show_mem()

Remove arch-specific show_mem() in favor of the generic version.

This also removes the following redundant information display:

	- free pages, printed by show_free_areas()
	- pages in swapcache, printed by show_swap_cache_info()

where show_mem() calls show_free_areas(), which calls
show_swap_cache_info().

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Cc: Chris Zankel <chris@zankel.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/xtensa/mm/init.c | 27 ---------------------------
 1 file changed, 27 deletions(-)

diff --git a/arch/xtensa/mm/init.c b/arch/xtensa/mm/init.c
index ee261005b363..34163cfaaffc 100644
--- a/arch/xtensa/mm/init.c
+++ b/arch/xtensa/mm/init.c
@@ -280,33 +280,6 @@ void free_initmem(void)
 	       (&__init_end - &__init_begin) >> 10);
 }
 
-void show_mem(void)
-{
-	int i, free = 0, total = 0, reserved = 0;
-	int shared = 0, cached = 0;
-
-	printk("Mem-info:\n");
-	show_free_areas();
-	printk("Free swap:       %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
-	i = max_mapnr;
-	while (i-- > 0) {
-		total++;
-		if (PageReserved(mem_map+i))
-			reserved++;
-		else if (PageSwapCache(mem_map+i))
-			cached++;
-		else if (!page_count(mem_map + i))
-			free++;
-		else
-			shared += page_count(mem_map + i) - 1;
-	}
-	printk("%d pages of RAM\n", total);
-	printk("%d reserved pages\n", reserved);
-	printk("%d pages shared\n", shared);
-	printk("%d pages swap cached\n",cached);
-	printk("%d free pages\n", free);
-}
-
 struct kmem_cache *pgtable_cache __read_mostly;
 
 static void pgd_ctor(void* addr)

From 8dad322f5449010c14990dd6934878f576b2ee60 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Fri, 25 Jul 2008 19:46:11 -0700
Subject: [PATCH 1076/1171] x86: use generic show_mem()

Remove arch-specific show_mem() in favor of the generic version.

This also removes the following redundant information display:

	- pages in swapcache, printed by show_swap_cache_info()
	- dirty pages, writeback pages, mapped pages, slab pages,
	  pagetable pages, printed by show_free_areas()

where show_mem() calls show_free_areas(), which calls
show_swap_cache_info().

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Acked-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/x86/mm/init_64.c    | 37 -------------------------------
 arch/x86/mm/pgtable_32.c | 47 ----------------------------------------
 2 files changed, 84 deletions(-)

diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index ec37121f6709..129618ca0ea2 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -86,43 +86,6 @@ early_param("gbpages", parse_direct_gbpages_on);
  * around without checking the pgd every time.
  */
 
-void show_mem(void)
-{
-	long i, total = 0, reserved = 0;
-	long shared = 0, cached = 0;
-	struct page *page;
-	pg_data_t *pgdat;
-
-	printk(KERN_INFO "Mem-info:\n");
-	show_free_areas();
-	for_each_online_pgdat(pgdat) {
-		for (i = 0; i < pgdat->node_spanned_pages; ++i) {
-			/*
-			 * This loop can take a while with 256 GB and
-			 * 4k pages so defer the NMI watchdog:
-			 */
-			if (unlikely(i % MAX_ORDER_NR_PAGES == 0))
-				touch_nmi_watchdog();
-
-			if (!pfn_valid(pgdat->node_start_pfn + i))
-				continue;
-
-			page = pfn_to_page(pgdat->node_start_pfn + i);
-			total++;
-			if (PageReserved(page))
-				reserved++;
-			else if (PageSwapCache(page))
-				cached++;
-			else if (page_count(page))
-				shared += page_count(page) - 1;
-		}
-	}
-	printk(KERN_INFO "%lu pages of RAM\n",		total);
-	printk(KERN_INFO "%lu reserved pages\n",	reserved);
-	printk(KERN_INFO "%lu pages shared\n",		shared);
-	printk(KERN_INFO "%lu pages swap cached\n",	cached);
-}
-
 int after_bootmem;
 
 static __init void *spp_getpage(void)
diff --git a/arch/x86/mm/pgtable_32.c b/arch/x86/mm/pgtable_32.c
index b4becbf8c570..cab0abbd1ebe 100644
--- a/arch/x86/mm/pgtable_32.c
+++ b/arch/x86/mm/pgtable_32.c
@@ -20,53 +20,6 @@
 #include <asm/tlb.h>
 #include <asm/tlbflush.h>
 
-void show_mem(void)
-{
-	int total = 0, reserved = 0;
-	int shared = 0, cached = 0;
-	int highmem = 0;
-	struct page *page;
-	pg_data_t *pgdat;
-	unsigned long i;
-	unsigned long flags;
-
-	printk(KERN_INFO "Mem-info:\n");
-	show_free_areas();
-	for_each_online_pgdat(pgdat) {
-		pgdat_resize_lock(pgdat, &flags);
-		for (i = 0; i < pgdat->node_spanned_pages; ++i) {
-			if (unlikely(i % MAX_ORDER_NR_PAGES == 0))
-				touch_nmi_watchdog();
-			page = pgdat_page_nr(pgdat, i);
-			total++;
-			if (PageHighMem(page))
-				highmem++;
-			if (PageReserved(page))
-				reserved++;
-			else if (PageSwapCache(page))
-				cached++;
-			else if (page_count(page))
-				shared += page_count(page) - 1;
-		}
-		pgdat_resize_unlock(pgdat, &flags);
-	}
-	printk(KERN_INFO "%d pages of RAM\n", total);
-	printk(KERN_INFO "%d pages of HIGHMEM\n", highmem);
-	printk(KERN_INFO "%d reserved pages\n", reserved);
-	printk(KERN_INFO "%d pages shared\n", shared);
-	printk(KERN_INFO "%d pages swap cached\n", cached);
-
-	printk(KERN_INFO "%lu pages dirty\n", global_page_state(NR_FILE_DIRTY));
-	printk(KERN_INFO "%lu pages writeback\n",
-					global_page_state(NR_WRITEBACK));
-	printk(KERN_INFO "%lu pages mapped\n", global_page_state(NR_FILE_MAPPED));
-	printk(KERN_INFO "%lu pages slab\n",
-		global_page_state(NR_SLAB_RECLAIMABLE) +
-		global_page_state(NR_SLAB_UNRECLAIMABLE));
-	printk(KERN_INFO "%lu pages pagetables\n",
-					global_page_state(NR_PAGETABLE));
-}
-
 /*
  * Associate a virtual page frame with a given physical page frame 
  * and protection flags for that frame.

From 79b0cbd113a9de1eaa3322528ccaeb97bd9189cc Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Fri, 25 Jul 2008 19:46:12 -0700
Subject: [PATCH 1077/1171] um: use generic show_mem()

Remove arch-specific show_mem() in favor of the generic version.

This also removes the following redundant information display:

	- free swap pages, printed by show_swap_cache_info()
	- pages in swapcache, printed by show_swap_cache_info()

where show_mem() calls show_free_areas(), which calls
show_swap_cache_info().

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Acked-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/um/kernel/mem.c | 31 -------------------------------
 1 file changed, 31 deletions(-)

diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c
index e2274ef3155d..61d7e6138ff5 100644
--- a/arch/um/kernel/mem.c
+++ b/arch/um/kernel/mem.c
@@ -264,37 +264,6 @@ void free_initrd_mem(unsigned long start, unsigned long end)
 }
 #endif
 
-void show_mem(void)
-{
-	int pfn, total = 0, reserved = 0;
-	int shared = 0, cached = 0;
-	int high_mem = 0;
-	struct page *page;
-
-	printk(KERN_INFO "Mem-info:\n");
-	show_free_areas();
-	printk(KERN_INFO "Free swap:       %6ldkB\n",
-	       nr_swap_pages<<(PAGE_SHIFT-10));
-	pfn = max_mapnr;
-	while (pfn-- > 0) {
-		page = pfn_to_page(pfn);
-		total++;
-		if (PageHighMem(page))
-			high_mem++;
-		if (PageReserved(page))
-			reserved++;
-		else if (PageSwapCache(page))
-			cached++;
-		else if (page_count(page))
-			shared += page_count(page) - 1;
-	}
-	printk(KERN_INFO "%d pages of RAM\n", total);
-	printk(KERN_INFO "%d pages of HIGHMEM\n", high_mem);
-	printk(KERN_INFO "%d reserved pages\n", reserved);
-	printk(KERN_INFO "%d pages shared\n", shared);
-	printk(KERN_INFO "%d pages swap cached\n", cached);
-}
-
 /* Allocate and free page tables. */
 
 pgd_t *pgd_alloc(struct mm_struct *mm)

From 471a54239a1c8a09cd12733d4e28db2aeebedb59 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Fri, 25 Jul 2008 19:46:12 -0700
Subject: [PATCH 1078/1171] sparc64: use generic show_mem()

Remove arch-specific show_mem() in favor of the generic version.

This also removes the following redundant information display:

	- free swap pages, printed by show_swap_cache_info()
	- pages in swapcache, printed by show_swap_cache_info()
	- dirty pages, writeback pages, mapped pages, slab pages,
	  pagetables pages, printed by show_free_areas()

where show_mem() calls show_free_areas(), which calls
show_swap_cache_info().

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/sparc64/mm/init.c | 45 ------------------------------------------
 1 file changed, 45 deletions(-)

diff --git a/arch/sparc64/mm/init.c b/arch/sparc64/mm/init.c
index 713297473951..4e821b3ecb03 100644
--- a/arch/sparc64/mm/init.c
+++ b/arch/sparc64/mm/init.c
@@ -392,51 +392,6 @@ void __kprobes flush_icache_range(unsigned long start, unsigned long end)
 	}
 }
 
-void show_mem(void)
-{
-	unsigned long total = 0, reserved = 0;
-	unsigned long shared = 0, cached = 0;
-	pg_data_t *pgdat;
-
-	printk(KERN_INFO "Mem-info:\n");
-	show_free_areas();
-	printk(KERN_INFO "Free swap:       %6ldkB\n",
-	       nr_swap_pages << (PAGE_SHIFT-10));
-	for_each_online_pgdat(pgdat) {
-		unsigned long i, flags;
-
-		pgdat_resize_lock(pgdat, &flags);
-		for (i = 0; i < pgdat->node_spanned_pages; i++) {
-			struct page *page = pgdat_page_nr(pgdat, i);
-			total++;
-			if (PageReserved(page))
-				reserved++;
-			else if (PageSwapCache(page))
-				cached++;
-			else if (page_count(page))
-				shared += page_count(page) - 1;
-		}
-		pgdat_resize_unlock(pgdat, &flags);
-	}
-
-	printk(KERN_INFO "%lu pages of RAM\n", total);
-	printk(KERN_INFO "%lu reserved pages\n", reserved);
-	printk(KERN_INFO "%lu pages shared\n", shared);
-	printk(KERN_INFO "%lu pages swap cached\n", cached);
-
-	printk(KERN_INFO "%lu pages dirty\n",
-	       global_page_state(NR_FILE_DIRTY));
-	printk(KERN_INFO "%lu pages writeback\n",
-	       global_page_state(NR_WRITEBACK));
-	printk(KERN_INFO "%lu pages mapped\n",
-	       global_page_state(NR_FILE_MAPPED));
-	printk(KERN_INFO "%lu pages slab\n",
-		global_page_state(NR_SLAB_RECLAIMABLE) +
-		global_page_state(NR_SLAB_UNRECLAIMABLE));
-	printk(KERN_INFO "%lu pages pagetables\n",
-	       global_page_state(NR_PAGETABLE));
-}
-
 void mmu_info(struct seq_file *m)
 {
 	if (tlb_type == cheetah)

From 03da6bfb5b40d454f5439ea905a68441aab23637 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Fri, 25 Jul 2008 19:46:13 -0700
Subject: [PATCH 1079/1171] sh: use generic show_mem()

Remove arch-specific show_mem() in favor of the generic version.

This also removes the following redundant information display:

	- free pages, printed by show_free_areas()
	- pages in slab, printed by show_free_areas()
	- free swap pages, printed by show_swap_cache_info()
	- pages in swapcache, printed by show_swap_cache_info()

where show_mem() calls show_free_areas(), which calls
show_swap_cache_info().

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Acked-by: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/sh/mm/init.c | 41 -----------------------------------------
 1 file changed, 41 deletions(-)

diff --git a/arch/sh/mm/init.c b/arch/sh/mm/init.c
index d652d375eb1e..b75a7acd62fb 100644
--- a/arch/sh/mm/init.c
+++ b/arch/sh/mm/init.c
@@ -25,47 +25,6 @@ DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
 pgd_t swapper_pg_dir[PTRS_PER_PGD];
 unsigned long cached_to_uncached = 0;
 
-void show_mem(void)
-{
-	int total = 0, reserved = 0, free = 0;
-	int shared = 0, cached = 0, slab = 0;
-	pg_data_t *pgdat;
-
-	printk("Mem-info:\n");
-	show_free_areas();
-
-	for_each_online_pgdat(pgdat) {
-		unsigned long flags, i;
-
-		pgdat_resize_lock(pgdat, &flags);
-		for (i = 0; i < pgdat->node_spanned_pages; i++) {
-			struct page *page = pgdat_page_nr(pgdat, i);
-			total++;
-			if (PageReserved(page))
-				reserved++;
-			else if (PageSwapCache(page))
-				cached++;
-			else if (PageSlab(page))
-				slab++;
-			else if (!page_count(page))
-				free++;
-			else
-				shared += page_count(page) - 1;
-		}
-		pgdat_resize_unlock(pgdat, &flags);
-	}
-
-	printk("Free swap:       %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
-	printk("%d pages of RAM\n", total);
-	printk("%d free pages\n", free);
-	printk("%d reserved pages\n", reserved);
-	printk("%d slab pages\n", slab);
-	printk("%d pages shared\n", shared);
-	printk("%d pages swap cached\n", cached);
-	printk(KERN_INFO "Total of %ld pages in page table cache\n",
-	       quicklist_total_size());
-}
-
 #ifdef CONFIG_MMU
 static void set_pte_phys(unsigned long addr, unsigned long phys, pgprot_t prot)
 {

From c55281dee09a843dd6bf5070324b86b84847e6ea Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Fri, 25 Jul 2008 19:46:14 -0700
Subject: [PATCH 1080/1171] s390: use generic show_mem()

Remove arch-specific show_mem() in favor of the generic version.

This also removes the following redundant information display:

	- pages in swapcache, printed by show_swap_cache_info()

where show_mem() calls show_free_areas(), which calls
show_swap_cache_info().

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Acked-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/s390/mm/init.c | 32 --------------------------------
 1 file changed, 32 deletions(-)

diff --git a/arch/s390/mm/init.c b/arch/s390/mm/init.c
index 388cc7420055..4993b0f594eb 100644
--- a/arch/s390/mm/init.c
+++ b/arch/s390/mm/init.c
@@ -42,38 +42,6 @@ DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
 pgd_t swapper_pg_dir[PTRS_PER_PGD] __attribute__((__aligned__(PAGE_SIZE)));
 char  empty_zero_page[PAGE_SIZE] __attribute__((__aligned__(PAGE_SIZE)));
 
-void show_mem(void)
-{
-	unsigned long i, total = 0, reserved = 0;
-	unsigned long shared = 0, cached = 0;
-	unsigned long flags;
-	struct page *page;
-	pg_data_t *pgdat;
-
-	printk("Mem-info:\n");
-	show_free_areas();
-	for_each_online_pgdat(pgdat) {
-		pgdat_resize_lock(pgdat, &flags);
-		for (i = 0; i < pgdat->node_spanned_pages; i++) {
-			if (!pfn_valid(pgdat->node_start_pfn + i))
-				continue;
-			page = pfn_to_page(pgdat->node_start_pfn + i);
-			total++;
-			if (PageReserved(page))
-				reserved++;
-			else if (PageSwapCache(page))
-				cached++;
-			else if (page_count(page))
-				shared += page_count(page) - 1;
-		}
-		pgdat_resize_unlock(pgdat, &flags);
-	}
-	printk("%ld pages of RAM\n", total);
-	printk("%ld reserved pages\n", reserved);
-	printk("%ld pages shared\n", shared);
-	printk("%ld pages swap cached\n", cached);
-}
-
 /*
  * paging_init() sets up the page tables
  */

From bda2fa535564ace56a395d5b65c6dc81305401fa Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Fri, 25 Jul 2008 19:46:14 -0700
Subject: [PATCH 1081/1171] powerpc: use generic show_mem()

Remove arch-specific show_mem() in favor of the generic version.

This also removes the following redundant information display:

	- pages in swapcache, printed by show_swap_cache_info()

where show_mem() calls show_free_areas(), which calls
show_swap_cache_info().

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Paul Mackerras <paulus@samba.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/powerpc/mm/mem.c | 39 ---------------------------------------
 1 file changed, 39 deletions(-)

diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c
index 1ca2235f0965..702691cb9e82 100644
--- a/arch/powerpc/mm/mem.c
+++ b/arch/powerpc/mm/mem.c
@@ -186,45 +186,6 @@ walk_memory_resource(unsigned long start_pfn, unsigned long nr_pages, void *arg,
 }
 EXPORT_SYMBOL_GPL(walk_memory_resource);
 
-void show_mem(void)
-{
-	unsigned long total = 0, reserved = 0;
-	unsigned long shared = 0, cached = 0;
-	unsigned long highmem = 0;
-	struct page *page;
-	pg_data_t *pgdat;
-	unsigned long i;
-
-	printk("Mem-info:\n");
-	show_free_areas();
-	for_each_online_pgdat(pgdat) {
-		unsigned long flags;
-		pgdat_resize_lock(pgdat, &flags);
-		for (i = 0; i < pgdat->node_spanned_pages; i++) {
-			if (!pfn_valid(pgdat->node_start_pfn + i))
-				continue;
-			page = pgdat_page_nr(pgdat, i);
-			total++;
-			if (PageHighMem(page))
-				highmem++;
-			if (PageReserved(page))
-				reserved++;
-			else if (PageSwapCache(page))
-				cached++;
-			else if (page_count(page))
-				shared += page_count(page) - 1;
-		}
-		pgdat_resize_unlock(pgdat, &flags);
-	}
-	printk("%ld pages of RAM\n", total);
-#ifdef CONFIG_HIGHMEM
-	printk("%ld pages of HIGHMEM\n", highmem);
-#endif
-	printk("%ld reserved pages\n", reserved);
-	printk("%ld pages shared\n", shared);
-	printk("%ld pages swap cached\n", cached);
-}
-
 /*
  * Initialize the bootmem system and give it all the memory we
  * have available.  If we are using highmem, we only put the

From 4984d2d88805609e7702e8ee72b5e148c992ae7b Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Fri, 25 Jul 2008 19:46:15 -0700
Subject: [PATCH 1082/1171] mn10300: use generic show_mem()

Remove arch-specific show_mem() in favor of the generic version.

This also removes the following redundant information display:

	- free pages, printed by show_free_areas()
	- pages in swapcache, printed by show_swap_cache_info()

where show_mem() calls show_free_areas(), which calls
show_swap_cache_info().

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Acked-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/mn10300/mm/pgtable.c | 27 ---------------------------
 1 file changed, 27 deletions(-)

diff --git a/arch/mn10300/mm/pgtable.c b/arch/mn10300/mm/pgtable.c
index a477038752ba..baffc581e031 100644
--- a/arch/mn10300/mm/pgtable.c
+++ b/arch/mn10300/mm/pgtable.c
@@ -27,33 +27,6 @@
 #include <asm/tlb.h>
 #include <asm/tlbflush.h>
 
-void show_mem(void)
-{
-	unsigned long i;
-	int free = 0, total = 0, reserved = 0, shared = 0;
-
-	int cached = 0;
-	printk(KERN_INFO "Mem-info:\n");
-	show_free_areas();
-	i = max_mapnr;
-	while (i-- > 0) {
-		total++;
-		if (PageReserved(mem_map + i))
-			reserved++;
-		else if (PageSwapCache(mem_map + i))
-			cached++;
-		else if (!page_count(mem_map + i))
-			free++;
-		else
-			shared += page_count(mem_map + i) - 1;
-	}
-	printk(KERN_INFO "%d pages of RAM\n", total);
-	printk(KERN_INFO "%d free pages\n", free);
-	printk(KERN_INFO "%d reserved pages\n", reserved);
-	printk(KERN_INFO "%d pages shared\n", shared);
-	printk(KERN_INFO "%d pages swap cached\n", cached);
-}
-
 /*
  * Associate a large virtual page frame with a given physical page frame
  * and protection flags for that frame. pfn is for the base of the page,

From 4bad9ea20146860d726ac14a53206da6be132b98 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Fri, 25 Jul 2008 19:46:16 -0700
Subject: [PATCH 1083/1171] h8300: use generic show_mem()

Remove arch-specific show_mem() in favor of the generic version.

This also removes the following redundant information display:

	- free pages, printed by show_free_areas()
	- pages in swapcache, printed by show_swap_cache_info()

where show_mem() calls show_free_areas(), which calls
show_swap_cache_info().

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>

Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/h8300/mm/init.c | 27 ---------------------------
 1 file changed, 27 deletions(-)

diff --git a/arch/h8300/mm/init.c b/arch/h8300/mm/init.c
index e4f4199f97ab..a1d228f5e4e6 100644
--- a/arch/h8300/mm/init.c
+++ b/arch/h8300/mm/init.c
@@ -64,33 +64,6 @@ unsigned long empty_zero_page;
 
 extern unsigned long rom_length;
 
-void show_mem(void)
-{
-    unsigned long i;
-    int free = 0, total = 0, reserved = 0, shared = 0;
-    int cached = 0;
-
-    printk("\nMem-info:\n");
-    show_free_areas();
-    i = max_mapnr;
-    while (i-- > 0) {
-	total++;
-	if (PageReserved(mem_map+i))
-	    reserved++;
-	else if (PageSwapCache(mem_map+i))
-	    cached++;
-	else if (!page_count(mem_map+i))
-	    free++;
-	else
-	    shared += page_count(mem_map+i) - 1;
-    }
-    printk("%d pages of RAM\n",total);
-    printk("%d free pages\n",free);
-    printk("%d reserved pages\n",reserved);
-    printk("%d pages shared\n",shared);
-    printk("%d pages swap cached\n",cached);
-}
-
 extern unsigned long memory_start;
 extern unsigned long memory_end;
 

From e66ddf1a4593d9bec8623782b94d7b50cc58cec5 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Fri, 25 Jul 2008 19:46:16 -0700
Subject: [PATCH 1084/1171] mips: use generic show_mem()

Remove arch-specific show_mem() in favor of the generic version.

This also removes the following redundant information display:

	- pages in swapcache, printed by show_swap_cache_info()

where show_mem() calls show_free_areas(), which calls
show_swap_cache_info().

And show_mem() does now actually print something on configurations
with multiple nodes.

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Acked-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/mips/mm/Makefile  |  3 +--
 arch/mips/mm/pgtable.c | 36 ------------------------------------
 2 files changed, 1 insertion(+), 38 deletions(-)
 delete mode 100644 arch/mips/mm/pgtable.c

diff --git a/arch/mips/mm/Makefile b/arch/mips/mm/Makefile
index 48731020ca0e..44e8dd8106bf 100644
--- a/arch/mips/mm/Makefile
+++ b/arch/mips/mm/Makefile
@@ -3,8 +3,7 @@
 #
 
 obj-y				+= cache.o dma-default.o extable.o fault.o \
-				   init.o pgtable.o tlbex.o tlbex-fault.o \
-				   uasm.o page.o
+				   init.o tlbex.o tlbex-fault.o uasm.o page.o
 
 obj-$(CONFIG_32BIT)		+= ioremap.o pgtable-32.o
 obj-$(CONFIG_64BIT)		+= pgtable-64.o
diff --git a/arch/mips/mm/pgtable.c b/arch/mips/mm/pgtable.c
deleted file mode 100644
index 7dfa579ab24c..000000000000
--- a/arch/mips/mm/pgtable.c
+++ /dev/null
@@ -1,36 +0,0 @@
-#include <linux/kernel.h>
-#include <linux/mm.h>
-#include <linux/swap.h>
-
-void show_mem(void)
-{
-#ifndef CONFIG_NEED_MULTIPLE_NODES  /* XXX(hch): later.. */
-	int pfn, total = 0, reserved = 0;
-	int shared = 0, cached = 0;
-	int highmem = 0;
-	struct page *page;
-
-	printk("Mem-info:\n");
-	show_free_areas();
-	pfn = max_mapnr;
-	while (pfn-- > 0) {
-		if (!pfn_valid(pfn))
-			continue;
-		page = pfn_to_page(pfn);
-		total++;
-		if (PageHighMem(page))
-			highmem++;
-		if (PageReserved(page))
-			reserved++;
-		else if (PageSwapCache(page))
-			cached++;
-		else if (page_count(page))
-			shared += page_count(page) - 1;
-	}
-	printk("%d pages of RAM\n", total);
-	printk("%d pages of HIGHMEM\n", highmem);
-	printk("%d reserved pages\n", reserved);
-	printk("%d pages shared\n", shared);
-	printk("%d pages swap cached\n", cached);
-#endif
-}

From 0ec5e0edd155385cb59cb9857de3176524ba4ff8 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Fri, 25 Jul 2008 19:46:17 -0700
Subject: [PATCH 1085/1171] m68knommu: use generic show_mem()

Remove arch-specific show_mem() in favor of the generic version.

This also removes the following redundant information display:

	- free pages, printed by show_free_areas()
	- pages in swapcache, printed by show_swap_cache_info()

where show_mem() calls show_free_areas(), which calls
show_swap_cache_info().

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Acked-by: Greg Ungerer <gerg@uclinux.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/m68knommu/mm/init.c | 27 ---------------------------
 1 file changed, 27 deletions(-)

diff --git a/arch/m68knommu/mm/init.c b/arch/m68knommu/mm/init.c
index 22e2a0d02b81..3bf249c53e41 100644
--- a/arch/m68knommu/mm/init.c
+++ b/arch/m68knommu/mm/init.c
@@ -62,33 +62,6 @@ static unsigned long empty_bad_page;
 
 unsigned long empty_zero_page;
 
-void show_mem(void)
-{
-    unsigned long i;
-    int free = 0, total = 0, reserved = 0, shared = 0;
-    int cached = 0;
-
-    printk(KERN_INFO "\nMem-info:\n");
-    show_free_areas();
-    i = max_mapnr;
-    while (i-- > 0) {
-	total++;
-	if (PageReserved(mem_map+i))
-	    reserved++;
-	else if (PageSwapCache(mem_map+i))
-	    cached++;
-	else if (!page_count(mem_map+i))
-	    free++;
-	else
-	    shared += page_count(mem_map+i) - 1;
-    }
-    printk(KERN_INFO "%d pages of RAM\n",total);
-    printk(KERN_INFO "%d free pages\n",free);
-    printk(KERN_INFO "%d reserved pages\n",reserved);
-    printk(KERN_INFO "%d pages shared\n",shared);
-    printk(KERN_INFO "%d pages swap cached\n",cached);
-}
-
 extern unsigned long memory_start;
 extern unsigned long memory_end;
 

From 969eefb516b71f8abadf5bc4861a672694f48c38 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Fri, 25 Jul 2008 19:46:18 -0700
Subject: [PATCH 1086/1171] m68k: use generic show_mem()

Remove arch-specific show_mem() in favor of the generic version.

This also removes the following redundant information display:

	- free pages, printed by show_free_areas()
	- pages in swapcache, printed by show_swap_cache_info()

where show_mem() calls show_free_areas(), which calls
show_swap_cache_info().

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Roman Zippel <zippel@linux-m68k.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/m68k/mm/init.c | 30 ------------------------------
 1 file changed, 30 deletions(-)

diff --git a/arch/m68k/mm/init.c b/arch/m68k/mm/init.c
index 79f5f94d4800..81bb08ceec18 100644
--- a/arch/m68k/mm/init.c
+++ b/arch/m68k/mm/init.c
@@ -69,36 +69,6 @@ void __init m68k_setup_node(int node)
 void *empty_zero_page;
 EXPORT_SYMBOL(empty_zero_page);
 
-void show_mem(void)
-{
-	pg_data_t *pgdat;
-	int free = 0, total = 0, reserved = 0, shared = 0;
-	int cached = 0;
-	int i;
-
-	printk("\nMem-info:\n");
-	show_free_areas();
-	for_each_online_pgdat(pgdat) {
-		for (i = 0; i < pgdat->node_spanned_pages; i++) {
-			struct page *page = pgdat->node_mem_map + i;
-			total++;
-			if (PageReserved(page))
-				reserved++;
-			else if (PageSwapCache(page))
-				cached++;
-			else if (!page_count(page))
-				free++;
-			else
-				shared += page_count(page) - 1;
-		}
-	}
-	printk("%d pages of RAM\n",total);
-	printk("%d free pages\n",free);
-	printk("%d reserved pages\n",reserved);
-	printk("%d pages shared\n",shared);
-	printk("%d pages swap cached\n",cached);
-}
-
 extern void init_pointer_table(unsigned long ptable);
 
 /* References to section boundaries */

From aca532016f49eca0341ed7be5885d6c63ca69e63 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Fri, 25 Jul 2008 19:46:18 -0700
Subject: [PATCH 1087/1171] m32r: use generic show_mem()

Remove arch-specific show_mem() in favor of the generic version.

This also removes the following redundant information display:

	- free swap pages, printed by show_swap_cache_info()
	- pages in swapcache, printed by show_swap_cache_info()

where show_mem() calls show_free_areas(), which calls
show_swap_cache_info().

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Cc: Hirokazu Takata <takata@linux-m32r.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/m32r/mm/init.c | 36 ------------------------------------
 1 file changed, 36 deletions(-)

diff --git a/arch/m32r/mm/init.c b/arch/m32r/mm/init.c
index 2554eb59cfef..24d429f9358a 100644
--- a/arch/m32r/mm/init.c
+++ b/arch/m32r/mm/init.c
@@ -36,42 +36,6 @@ pgd_t swapper_pg_dir[1024];
 
 DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
 
-void show_mem(void)
-{
-	int total = 0, reserved = 0;
-	int shared = 0, cached = 0;
-	int highmem = 0;
-	struct page *page;
-	pg_data_t *pgdat;
-	unsigned long i;
-
-	printk("Mem-info:\n");
-	show_free_areas();
-	printk("Free swap:       %6ldkB\n",nr_swap_pages<<(PAGE_SHIFT-10));
-	for_each_online_pgdat(pgdat) {
-		unsigned long flags;
-		pgdat_resize_lock(pgdat, &flags);
-		for (i = 0; i < pgdat->node_spanned_pages; ++i) {
-			page = pgdat_page_nr(pgdat, i);
-			total++;
-			if (PageHighMem(page))
-				highmem++;
-			if (PageReserved(page))
-				reserved++;
-			else if (PageSwapCache(page))
-				cached++;
-			else if (page_count(page))
-				shared += page_count(page) - 1;
-		}
-		pgdat_resize_unlock(pgdat, &flags);
-	}
-	printk("%d pages of RAM\n", total);
-	printk("%d pages of HIGHMEM\n",highmem);
-	printk("%d reserved pages\n",reserved);
-	printk("%d pages shared\n",shared);
-	printk("%d pages swap cached\n",cached);
-}
-
 /*
  * Cache of MMU context last used.
  */

From e275e0a687e83add62b10e6d36b67f0964455469 Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Fri, 25 Jul 2008 19:46:19 -0700
Subject: [PATCH 1088/1171] frv: use generic show_mem()

Remove arch-specific show_mem() in favor of the generic version.

This also removes the following redundant information display:

	- free pages, printed by show_free_areas()

where show_mem() calls show_free_areas().

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Acked-by: David Howells <dhowells@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/frv/mm/init.c | 31 -------------------------------
 1 file changed, 31 deletions(-)

diff --git a/arch/frv/mm/init.c b/arch/frv/mm/init.c
index 9af7740f32fb..1b851db34186 100644
--- a/arch/frv/mm/init.c
+++ b/arch/frv/mm/init.c
@@ -61,37 +61,6 @@ static unsigned long empty_bad_page;
 unsigned long empty_zero_page;
 EXPORT_SYMBOL(empty_zero_page);
 
-/*****************************************************************************/
-/*
- *
- */
-void show_mem(void)
-{
-	unsigned long i;
-	int free = 0, total = 0, reserved = 0, shared = 0;
-
-	printk("\nMem-info:\n");
-	show_free_areas();
-	i = max_mapnr;
-	while (i-- > 0) {
-		struct page *page = &mem_map[i];
-
-		total++;
-		if (PageReserved(page))
-			reserved++;
-		else if (!page_count(page))
-			free++;
-		else
-			shared += page_count(page) - 1;
-	}
-
-	printk("%d pages of RAM\n",total);
-	printk("%d free pages\n",free);
-	printk("%d reserved pages\n",reserved);
-	printk("%d pages shared\n",shared);
-
-} /* end show_mem() */
-
 /*****************************************************************************/
 /*
  * paging_init() continues the virtual memory environment setup which

From a4074d93dd4e8f46f727840baa2be02d544fd17f Mon Sep 17 00:00:00 2001
From: Johannes Weiner <hannes@saeurebad.de>
Date: Fri, 25 Jul 2008 19:46:20 -0700
Subject: [PATCH 1089/1171] cris: use generic show_mem()

Remove arch-specific show_mem() in favor of the generic version.

This also removes the following redundant information display:

	- free pages, printed by show_free_areas()
	- pages in swapcache, printed by show_swap_cache_info()

where show_mem() calls show_free_areas(), which calls
show_swap_cache_info().

Signed-off-by: Johannes Weiner <hannes@saeurebad.de>
Acked-by: Mikael Starvik <starvik@axis.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 arch/cris/mm/init.c | 30 ------------------------------
 1 file changed, 30 deletions(-)

diff --git a/arch/cris/mm/init.c b/arch/cris/mm/init.c
index 5b06ffa15e34..2fdd212eb250 100644
--- a/arch/cris/mm/init.c
+++ b/arch/cris/mm/init.c
@@ -19,36 +19,6 @@ unsigned long empty_zero_page;
 extern char _stext, _edata, _etext; /* From linkerscript */
 extern char __init_begin, __init_end;
 
-void 
-show_mem(void)
-{
-	int i,free = 0,total = 0,cached = 0, reserved = 0, nonshared = 0;
-	int shared = 0;
-
-	printk("\nMem-info:\n");
-	show_free_areas();
-	i = max_mapnr;
-	while (i-- > 0) {
-		total++;
-		if (PageReserved(mem_map+i))
-			reserved++;
-		else if (PageSwapCache(mem_map+i))
-			cached++;
-		else if (!page_count(mem_map+i))
-			free++;
-		else if (page_count(mem_map+i) == 1)
-			nonshared++;
-		else
-			shared += page_count(mem_map+i) - 1;
-	}
-	printk("%d pages of RAM\n",total);
-	printk("%d free pages\n",free);
-	printk("%d reserved pages\n",reserved);
-	printk("%d pages nonshared\n",nonshared);
-	printk("%d pages shared\n",shared);
-	printk("%d pages swap cached\n",cached);
-}
-
 void __init
 mem_init(void)
 {

From 0e1451da4f928ae1c9d5ca617faebde9f02985db Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 19:46:20 -0700
Subject: [PATCH 1090/1171] drm: make drm_minors_cleanup() static

Make the needlessly global drm_minors_cleanup() static.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/gpu/drm/drm_drv.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index 564138714bb5..452c2d866ec5 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -318,7 +318,7 @@ static void drm_cleanup(struct drm_device * dev)
 		DRM_ERROR("Cannot unload module\n");
 }
 
-int drm_minors_cleanup(int id, void *ptr, void *data)
+static int drm_minors_cleanup(int id, void *ptr, void *data)
 {
 	struct drm_minor *minor = ptr;
 	struct drm_device *dev;

From 96930a6365c99c160138a395566e360b27348b8f Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 19:46:21 -0700
Subject: [PATCH 1091/1171] make cgroup_seqfile_release() static

cgroup_seqfile_release() can become static.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Acked-by: Paul Menage <menage@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 kernel/cgroup.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index 66ec9fd21e0c..89bd6fb7894f 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -1529,7 +1529,7 @@ static int cgroup_seqfile_show(struct seq_file *m, void *arg)
 	return cft->read_seq_string(state->cgroup, cft, m);
 }
 
-int cgroup_seqfile_release(struct inode *inode, struct file *file)
+static int cgroup_seqfile_release(struct inode *inode, struct file *file)
 {
 	struct seq_file *seq = file->private_data;
 	kfree(seq->private);

From 25cdcd0086d97a011fcd0c1ff572e30da24790ec Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 19:46:21 -0700
Subject: [PATCH 1092/1171] make pnp_add_card_id() static

pnp_add_card_id() can now become static.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Cc: Bjorn Helgaas <bjorn.helgaas@hp.com>
Cc: Rene Herman <rene.herman@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/pnp/base.h | 1 -
 drivers/pnp/card.c | 2 +-
 2 files changed, 1 insertion(+), 2 deletions(-)

diff --git a/drivers/pnp/base.h b/drivers/pnp/base.h
index e3fa9a2d9a3d..9fd7bb9b7dce 100644
--- a/drivers/pnp/base.h
+++ b/drivers/pnp/base.h
@@ -19,7 +19,6 @@ struct pnp_id *pnp_add_id(struct pnp_dev *dev, char *id);
 int pnp_interface_attach_device(struct pnp_dev *dev);
 
 int pnp_add_card(struct pnp_card *card);
-struct pnp_id *pnp_add_card_id(struct pnp_card *card, char *id);
 void pnp_remove_card(struct pnp_card *card);
 int pnp_add_card_device(struct pnp_card *card, struct pnp_dev *dev);
 void pnp_remove_card_device(struct pnp_dev *dev);
diff --git a/drivers/pnp/card.c b/drivers/pnp/card.c
index b00ef1030e45..e75b060daa95 100644
--- a/drivers/pnp/card.c
+++ b/drivers/pnp/card.c
@@ -102,7 +102,7 @@ static int card_probe(struct pnp_card *card, struct pnp_card_driver *drv)
  * @id: pointer to a pnp_id structure
  * @card: pointer to the desired card
  */
-struct pnp_id *pnp_add_card_id(struct pnp_card *card, char *id)
+static struct pnp_id *pnp_add_card_id(struct pnp_card *card, char *id)
 {
 	struct pnp_id *dev_id, *ptr;
 

From 23d5f96ce6571da51c0f6bfa7361e5f91f314b2b Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 19:46:22 -0700
Subject: [PATCH 1093/1171] make parport_cs_release() static

This patch makes the needlessly global parport_cs_release() static.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/parport/parport_cs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/parport/parport_cs.c b/drivers/parport/parport_cs.c
index 802a81d47367..00e1d9620f7c 100644
--- a/drivers/parport/parport_cs.c
+++ b/drivers/parport/parport_cs.c
@@ -235,7 +235,7 @@ failed:
     
 ======================================================================*/
 
-void parport_cs_release(struct pcmcia_device *link)
+static void parport_cs_release(struct pcmcia_device *link)
 {
 	parport_info_t *info = link->priv;
 

From 9e5c6da71e89fa25ced6e88182225a99941bec90 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 19:46:22 -0700
Subject: [PATCH 1094/1171] make mm/sparse.c: make a function static

This patch makes the needlessly global sparse_early_mem_map_alloc()
static.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/sparse.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/sparse.c b/mm/sparse.c
index 8ffc08990008..5d9dbbb9d39e 100644
--- a/mm/sparse.c
+++ b/mm/sparse.c
@@ -377,7 +377,7 @@ struct page __init *sparse_mem_map_populate(unsigned long pnum, int nid)
 }
 #endif /* !CONFIG_SPARSEMEM_VMEMMAP */
 
-struct page __init *sparse_early_mem_map_alloc(unsigned long pnum)
+static struct page __init *sparse_early_mem_map_alloc(unsigned long pnum)
 {
 	struct page *map;
 	struct mem_section *ms = __nr_to_section(pnum);

From 9d8fddfb17aaee4ffc5e3d0560620d0fa8b50a42 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 19:46:23 -0700
Subject: [PATCH 1095/1171] mm/allocpercpu.c: make 4 functions static

This patch makes the following needlessly global functions static:
 - percpu_depopulate()
 - __percpu_depopulate_mask()
 - percpu_populate()
 - __percpu_populate_mask()

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Acked-by: Christoph Lameter <cl@linux-foundation.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/percpu.h | 29 -----------------------------
 mm/allocpercpu.c       | 20 +++++++++++---------
 2 files changed, 11 insertions(+), 38 deletions(-)

diff --git a/include/linux/percpu.h b/include/linux/percpu.h
index 4cdd393e71e1..fac3337547eb 100644
--- a/include/linux/percpu.h
+++ b/include/linux/percpu.h
@@ -74,11 +74,6 @@ struct percpu_data {
         (__typeof__(ptr))__p->ptrs[(cpu)];	          \
 })
 
-extern void *percpu_populate(void *__pdata, size_t size, gfp_t gfp, int cpu);
-extern void percpu_depopulate(void *__pdata, int cpu);
-extern int __percpu_populate_mask(void *__pdata, size_t size, gfp_t gfp,
-				  cpumask_t *mask);
-extern void __percpu_depopulate_mask(void *__pdata, cpumask_t *mask);
 extern void *__percpu_alloc_mask(size_t size, gfp_t gfp, cpumask_t *mask);
 extern void percpu_free(void *__pdata);
 
@@ -86,26 +81,6 @@ extern void percpu_free(void *__pdata);
 
 #define percpu_ptr(ptr, cpu) ({ (void)(cpu); (ptr); })
 
-static inline void percpu_depopulate(void *__pdata, int cpu)
-{
-}
-
-static inline void __percpu_depopulate_mask(void *__pdata, cpumask_t *mask)
-{
-}
-
-static inline void *percpu_populate(void *__pdata, size_t size, gfp_t gfp,
-				    int cpu)
-{
-	return percpu_ptr(__pdata, cpu);
-}
-
-static inline int __percpu_populate_mask(void *__pdata, size_t size, gfp_t gfp,
-					 cpumask_t *mask)
-{
-	return 0;
-}
-
 static __always_inline void *__percpu_alloc_mask(size_t size, gfp_t gfp, cpumask_t *mask)
 {
 	return kzalloc(size, gfp);
@@ -118,10 +93,6 @@ static inline void percpu_free(void *__pdata)
 
 #endif /* CONFIG_SMP */
 
-#define percpu_populate_mask(__pdata, size, gfp, mask) \
-	__percpu_populate_mask((__pdata), (size), (gfp), &(mask))
-#define percpu_depopulate_mask(__pdata, mask) \
-	__percpu_depopulate_mask((__pdata), &(mask))
 #define percpu_alloc_mask(size, gfp, mask) \
 	__percpu_alloc_mask((size), (gfp), &(mask))
 
diff --git a/mm/allocpercpu.c b/mm/allocpercpu.c
index 843364594e23..4297bc41bfd2 100644
--- a/mm/allocpercpu.c
+++ b/mm/allocpercpu.c
@@ -18,27 +18,28 @@
  * Depopulating per-cpu data for a cpu going offline would be a typical
  * use case. You need to register a cpu hotplug handler for that purpose.
  */
-void percpu_depopulate(void *__pdata, int cpu)
+static void percpu_depopulate(void *__pdata, int cpu)
 {
 	struct percpu_data *pdata = __percpu_disguise(__pdata);
 
 	kfree(pdata->ptrs[cpu]);
 	pdata->ptrs[cpu] = NULL;
 }
-EXPORT_SYMBOL_GPL(percpu_depopulate);
 
 /**
  * percpu_depopulate_mask - depopulate per-cpu data for some cpu's
  * @__pdata: per-cpu data to depopulate
  * @mask: depopulate per-cpu data for cpu's selected through mask bits
  */
-void __percpu_depopulate_mask(void *__pdata, cpumask_t *mask)
+static void __percpu_depopulate_mask(void *__pdata, cpumask_t *mask)
 {
 	int cpu;
 	for_each_cpu_mask_nr(cpu, *mask)
 		percpu_depopulate(__pdata, cpu);
 }
-EXPORT_SYMBOL_GPL(__percpu_depopulate_mask);
+
+#define percpu_depopulate_mask(__pdata, mask) \
+	__percpu_depopulate_mask((__pdata), &(mask))
 
 /**
  * percpu_populate - populate per-cpu data for given cpu
@@ -51,7 +52,7 @@ EXPORT_SYMBOL_GPL(__percpu_depopulate_mask);
  * use case. You need to register a cpu hotplug handler for that purpose.
  * Per-cpu object is populated with zeroed buffer.
  */
-void *percpu_populate(void *__pdata, size_t size, gfp_t gfp, int cpu)
+static void *percpu_populate(void *__pdata, size_t size, gfp_t gfp, int cpu)
 {
 	struct percpu_data *pdata = __percpu_disguise(__pdata);
 	int node = cpu_to_node(cpu);
@@ -68,7 +69,6 @@ void *percpu_populate(void *__pdata, size_t size, gfp_t gfp, int cpu)
 		pdata->ptrs[cpu] = kzalloc(size, gfp);
 	return pdata->ptrs[cpu];
 }
-EXPORT_SYMBOL_GPL(percpu_populate);
 
 /**
  * percpu_populate_mask - populate per-cpu data for more cpu's
@@ -79,8 +79,8 @@ EXPORT_SYMBOL_GPL(percpu_populate);
  *
  * Per-cpu objects are populated with zeroed buffers.
  */
-int __percpu_populate_mask(void *__pdata, size_t size, gfp_t gfp,
-			   cpumask_t *mask)
+static int __percpu_populate_mask(void *__pdata, size_t size, gfp_t gfp,
+				  cpumask_t *mask)
 {
 	cpumask_t populated;
 	int cpu;
@@ -94,7 +94,9 @@ int __percpu_populate_mask(void *__pdata, size_t size, gfp_t gfp,
 			cpu_set(cpu, populated);
 	return 0;
 }
-EXPORT_SYMBOL_GPL(__percpu_populate_mask);
+
+#define percpu_populate_mask(__pdata, size, gfp, mask) \
+	__percpu_populate_mask((__pdata), (size), (gfp), &(mask))
 
 /**
  * percpu_alloc_mask - initial setup of per-cpu data

From 15f59adae001766a2c7f7fe4f196387bb04bcff5 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 19:46:23 -0700
Subject: [PATCH 1096/1171] make mm/memory.c:print_bad_pte() static

This patch makes the needlessly global print_bad_pte() static.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/mm.h | 1 -
 mm/memory.c        | 3 ++-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/include/linux/mm.h b/include/linux/mm.h
index f3fd70d6029f..6e695eaab4ce 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -810,7 +810,6 @@ extern int access_process_vm(struct task_struct *tsk, unsigned long addr, void *
 
 int get_user_pages(struct task_struct *tsk, struct mm_struct *mm, unsigned long start,
 		int len, int write, int force, struct page **pages, struct vm_area_struct **vmas);
-void print_bad_pte(struct vm_area_struct *, pte_t, unsigned long);
 
 extern int try_to_release_page(struct page * page, gfp_t gfp_mask);
 extern void do_invalidatepage(struct page *page, unsigned long offset);
diff --git a/mm/memory.c b/mm/memory.c
index 262e3eb6601a..a8ca04faaea6 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -374,7 +374,8 @@ static inline void add_mm_rss(struct mm_struct *mm, int file_rss, int anon_rss)
  *
  * The calling function must still handle the error.
  */
-void print_bad_pte(struct vm_area_struct *vma, pte_t pte, unsigned long vaddr)
+static void print_bad_pte(struct vm_area_struct *vma, pte_t pte,
+			  unsigned long vaddr)
 {
 	printk(KERN_ERR "Bad pte = %08llx, process = %s, "
 			"vm_flags = %lx, vaddr = %lx\n",

From 7c363b8c6536f26934172d3c46f0bbec01a97c61 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 19:46:24 -0700
Subject: [PATCH 1097/1171] mm/swapfile.c: make code static

This patch makes the following needlessly global code static:
 - swap_lock
 - nr_swapfiles
 - struct swap_list

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/swap.h | 3 ---
 mm/swapfile.c        | 6 +++---
 2 files changed, 3 insertions(+), 6 deletions(-)

diff --git a/include/linux/swap.h b/include/linux/swap.h
index 0b3377650c85..de40f169a4e4 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -237,7 +237,6 @@ extern struct page *swapin_readahead(swp_entry_t, gfp_t,
 
 /* linux/mm/swapfile.c */
 extern long total_swap_pages;
-extern unsigned int nr_swapfiles;
 extern void si_swapinfo(struct sysinfo *);
 extern swp_entry_t get_swap_page(void);
 extern swp_entry_t get_swap_page_of_type(int);
@@ -254,8 +253,6 @@ extern int can_share_swap_page(struct page *);
 extern int remove_exclusive_swap_page(struct page *);
 struct backing_dev_info;
 
-extern spinlock_t swap_lock;
-
 /* linux/mm/thrash.c */
 extern struct mm_struct * swap_token_mm;
 extern void grab_swap_token(void);
diff --git a/mm/swapfile.c b/mm/swapfile.c
index af283933c14e..6beb6251e99d 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -33,8 +33,8 @@
 #include <asm/tlbflush.h>
 #include <linux/swapops.h>
 
-DEFINE_SPINLOCK(swap_lock);
-unsigned int nr_swapfiles;
+static DEFINE_SPINLOCK(swap_lock);
+static unsigned int nr_swapfiles;
 long total_swap_pages;
 static int swap_overflow;
 static int least_priority;
@@ -44,7 +44,7 @@ static const char Unused_file[] = "Unused swap file entry ";
 static const char Bad_offset[] = "Bad swap offset entry ";
 static const char Unused_offset[] = "Unused swap offset entry ";
 
-struct swap_list_t swap_list = {-1, -1};
+static struct swap_list_t swap_list = {-1, -1};
 
 static struct swap_info_struct swap_info[MAX_SWAPFILES];
 

From 9580d85f9cdb076c4bfb467bc6c0d3c5e499957a Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 19:46:25 -0700
Subject: [PATCH 1098/1171] drivers/char/rtc.c: make 2 functions static

The following functions can now become static:
 - rtc_interrupt()
 - rtc_get_rtc_time()

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Acked-by: Bernhard Walle <bwalle@suse.de>
Acked-by: Paul Gortmaker <p_gortmaker@yahoo.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/rtc.c  | 5 +++--
 include/linux/rtc.h | 2 --
 2 files changed, 3 insertions(+), 4 deletions(-)

diff --git a/drivers/char/rtc.c b/drivers/char/rtc.c
index dbefbb30ed44..d9799e2bcfbf 100644
--- a/drivers/char/rtc.c
+++ b/drivers/char/rtc.c
@@ -144,6 +144,7 @@ static ssize_t rtc_read(struct file *file, char __user *buf,
 			size_t count, loff_t *ppos);
 
 static long rtc_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
+static void rtc_get_rtc_time(struct rtc_time *rtc_tm);
 
 #ifdef RTC_IRQ
 static unsigned int rtc_poll(struct file *file, poll_table *wait);
@@ -235,7 +236,7 @@ static inline unsigned char rtc_is_updating(void)
  *	(See ./arch/XXXX/kernel/time.c for the set_rtc_mmss() function.)
  */
 
-irqreturn_t rtc_interrupt(int irq, void *dev_id)
+static irqreturn_t rtc_interrupt(int irq, void *dev_id)
 {
 	/*
 	 *	Can be an alarm interrupt, update complete interrupt,
@@ -1303,7 +1304,7 @@ static int rtc_proc_open(struct inode *inode, struct file *file)
 }
 #endif
 
-void rtc_get_rtc_time(struct rtc_time *rtc_tm)
+static void rtc_get_rtc_time(struct rtc_time *rtc_tm)
 {
 	unsigned long uip_watchdog = jiffies, flags;
 	unsigned char ctrl;
diff --git a/include/linux/rtc.h b/include/linux/rtc.h
index b01fe004cb5e..91f597ad6acc 100644
--- a/include/linux/rtc.h
+++ b/include/linux/rtc.h
@@ -225,8 +225,6 @@ typedef struct rtc_task {
 int rtc_register(rtc_task_t *task);
 int rtc_unregister(rtc_task_t *task);
 int rtc_control(rtc_task_t *t, unsigned int cmd, unsigned long arg);
-void rtc_get_rtc_time(struct rtc_time *rtc_tm);
-irqreturn_t rtc_interrupt(int irq, void *dev_id);
 
 #endif /* __KERNEL__ */
 

From f56f6d30c75984d0aba2656eaeb35f3806395191 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 19:46:25 -0700
Subject: [PATCH 1099/1171] make init/do_mounts.c:root_device_name static

This patch makes the needlessly global root_device_name static.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 init/do_mounts.c | 2 +-
 init/do_mounts.h | 1 -
 2 files changed, 1 insertion(+), 2 deletions(-)

diff --git a/init/do_mounts.c b/init/do_mounts.c
index f769fac4f4c0..3715feb8446d 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -23,7 +23,7 @@
 int __initdata rd_doload;	/* 1 = load RAM disk, 0 = don't load */
 
 int root_mountflags = MS_RDONLY | MS_SILENT;
-char * __initdata root_device_name;
+static char * __initdata root_device_name;
 static char __initdata saved_root_name[64];
 static int __initdata root_wait;
 
diff --git a/init/do_mounts.h b/init/do_mounts.h
index 735705d137ff..9aa968d54329 100644
--- a/init/do_mounts.h
+++ b/init/do_mounts.h
@@ -11,7 +11,6 @@ void  change_floppy(char *fmt, ...);
 void  mount_block_root(char *name, int flags);
 void  mount_root(void);
 extern int root_mountflags;
-extern char *root_device_name;
 
 static inline int create_dev(char *name, dev_t dev)
 {

From 511e7483abe1ab433d8ab7a7998f799042b52941 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 19:46:26 -0700
Subject: [PATCH 1100/1171] make macfb_setup() static

This patch makes the needlessly global macfb_setup() static.

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/macfb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/video/macfb.c b/drivers/video/macfb.c
index aa8c714d6245..b790ddff76f9 100644
--- a/drivers/video/macfb.c
+++ b/drivers/video/macfb.c
@@ -596,7 +596,7 @@ static struct fb_ops macfb_ops = {
 	.fb_imageblit	= cfb_imageblit,
 };
 
-void __init macfb_setup(char *options)
+static void __init macfb_setup(char *options)
 {
 	char *this_opt;
 	

From 3d1e412ac5570a669e1b1fc5fd0f6859250c3d76 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 19:46:26 -0700
Subject: [PATCH 1101/1171] video/console/stico{n,re}.c: make code static

This patch makes the following needlessly global code static:
 - sticon.c: sticonsole_init()
 - sticore.c: struct default_sti
 - sticore.c: sti_init_graph()
 - sticore.c: sti_inq_conf()
 - sticore.c: sti_rom_copy()
 - sticore.c: sti_select_fbfont()
 - sticore.c: sti_select_font()
 - sticore.c: sti_get_wmode_rom()
 - sticore.c: sti_read_rom()

[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/console/sticon.c  |  2 +-
 drivers/video/console/sticore.c | 33 +++++++++++++++------------------
 drivers/video/sticore.h         |  2 --
 3 files changed, 16 insertions(+), 21 deletions(-)

diff --git a/drivers/video/console/sticon.c b/drivers/video/console/sticon.c
index a11cc2fdd4cd..4055dbdd1b42 100644
--- a/drivers/video/console/sticon.c
+++ b/drivers/video/console/sticon.c
@@ -370,7 +370,7 @@ static const struct consw sti_con = {
 
 
 
-int __init sticonsole_init(void)
+static int __init sticonsole_init(void)
 {
     /* already initialized ? */
     if (sticon_sti)
diff --git a/drivers/video/console/sticore.c b/drivers/video/console/sticore.c
index e9ab657f0bb7..d7822af0e00a 100644
--- a/drivers/video/console/sticore.c
+++ b/drivers/video/console/sticore.c
@@ -29,7 +29,7 @@
 
 #define STI_DRIVERVERSION "Version 0.9a"
 
-struct sti_struct *default_sti __read_mostly;
+static struct sti_struct *default_sti __read_mostly;
 
 /* number of STI ROMS found and their ptrs to each struct */
 static int num_sti_roms __read_mostly;
@@ -68,8 +68,7 @@ static const struct sti_init_flags default_init_flags = {
 	.init_cmap_tx = 1,
 };
 
-int
-sti_init_graph(struct sti_struct *sti) 
+static int sti_init_graph(struct sti_struct *sti)
 {
 	struct sti_init_inptr_ext inptr_ext = { 0, };
 	struct sti_init_inptr inptr = {
@@ -100,8 +99,7 @@ static const struct sti_conf_flags default_conf_flags = {
 	.wait	= STI_WAIT,
 };
 
-void
-sti_inq_conf(struct sti_struct *sti)
+static void sti_inq_conf(struct sti_struct *sti)
 {
 	struct sti_conf_inptr inptr = { 0, };
 	unsigned long flags;
@@ -237,8 +235,8 @@ static void sti_flush(unsigned long start, unsigned long end)
 	flush_icache_range(start, end);
 }
 
-void __devinit
-sti_rom_copy(unsigned long base, unsigned long count, void *dest)
+static void __devinit sti_rom_copy(unsigned long base, unsigned long count,
+				   void *dest)
 {
 	unsigned long dest_start = (unsigned long) dest;
 
@@ -478,8 +476,8 @@ sti_init_glob_cfg(struct sti_struct *sti,
 }
 
 #ifdef CONFIG_FB
-struct sti_cooked_font * __devinit
-sti_select_fbfont(struct sti_cooked_rom *cooked_rom, const char *fbfont_name)
+static struct sti_cooked_font __devinit
+*sti_select_fbfont(struct sti_cooked_rom *cooked_rom, const char *fbfont_name)
 {
 	const struct font_desc *fbfont;
 	unsigned int size, bpc;
@@ -534,16 +532,16 @@ sti_select_fbfont(struct sti_cooked_rom *cooked_rom, const char *fbfont_name)
 	return cooked_font;
 }
 #else
-struct sti_cooked_font * __devinit
-sti_select_fbfont(struct sti_cooked_rom *cooked_rom, const char *fbfont_name)
+static struct sti_cooked_font __devinit
+*sti_select_fbfont(struct sti_cooked_rom *cooked_rom, const char *fbfont_name)
 {
 	return NULL;
 }
 #endif
 
-struct sti_cooked_font * __devinit
-sti_select_font(struct sti_cooked_rom *rom,
-	    int (*search_font_fnc) (struct sti_cooked_rom *,int,int) )
+static struct sti_cooked_font __devinit
+*sti_select_font(struct sti_cooked_rom *rom,
+		 int (*search_font_fnc)(struct sti_cooked_rom *, int, int))
 {
 	struct sti_cooked_font *font;
 	int i;
@@ -707,8 +705,7 @@ sti_get_bmode_rom (unsigned long address)
 	return raw;
 }
 
-struct sti_rom * __devinit
-sti_get_wmode_rom (unsigned long address)
+static struct sti_rom __devinit *sti_get_wmode_rom(unsigned long address)
 {
 	struct sti_rom *raw;
 	unsigned long size;
@@ -723,8 +720,8 @@ sti_get_wmode_rom (unsigned long address)
 	return raw;
 }
 
-int __devinit
-sti_read_rom(int wordmode, struct sti_struct *sti, unsigned long address)
+static int __devinit sti_read_rom(int wordmode, struct sti_struct *sti,
+				  unsigned long address)
 {
 	struct sti_cooked_rom *cooked;
 	struct sti_rom *raw = NULL;
diff --git a/drivers/video/sticore.h b/drivers/video/sticore.h
index 1a9a60c74be3..7fe5be4bc70e 100644
--- a/drivers/video/sticore.h
+++ b/drivers/video/sticore.h
@@ -352,8 +352,6 @@ struct sti_struct *sti_get_rom(unsigned int index); /* 0: default sti */
 
 /* functions to call the STI ROM directly */
 
-int  sti_init_graph(struct sti_struct *sti);
-void sti_inq_conf(struct sti_struct *sti);
 void sti_putc(struct sti_struct *sti, int c, int y, int x);
 void sti_set(struct sti_struct *sti, int src_y, int src_x,
 	     int height, int width, u8 color);

From 6ca813c4e515d9b868cd71703ef15f4af3aebb21 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Fri, 25 Jul 2008 19:46:27 -0700
Subject: [PATCH 1102/1171] video/stifb.c: make 2 functions static

This patch makes the following needlessly global functions static:
 - stifb_init_fb()
 - stifb_init()

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/stifb.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/drivers/video/stifb.c b/drivers/video/stifb.c
index 598d35eff935..166481402412 100644
--- a/drivers/video/stifb.c
+++ b/drivers/video/stifb.c
@@ -1078,8 +1078,7 @@ static struct fb_ops stifb_ops = {
  *  Initialization
  */
 
-int __init
-stifb_init_fb(struct sti_struct *sti, int bpp_pref)
+static int __init stifb_init_fb(struct sti_struct *sti, int bpp_pref)
 {
 	struct fb_fix_screeninfo *fix;
 	struct fb_var_screeninfo *var;
@@ -1315,8 +1314,7 @@ static int stifb_disabled __initdata;
 int __init
 stifb_setup(char *options);
 
-int __init
-stifb_init(void)
+static int __init stifb_init(void)
 {
 	struct sti_struct *sti;
 	struct sti_struct *def_sti;

From 37e67b75804b84e092ae9f1d7a19dc3522ef78ab Mon Sep 17 00:00:00 2001
From: Huang Weiyi <weiyi.huang@gmail.com>
Date: Sat, 26 Jul 2008 19:37:57 +0800
Subject: [PATCH 1103/1171] drivers/video/fbmem.c: removed duplicated include

Removed duplicated include <linux/major.h>
in drivers/video/fbmem.c

Signed-off-by: Huang Weiyi <weiyi.huang@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/video/fbmem.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/drivers/video/fbmem.c b/drivers/video/fbmem.c
index 5d84b3431098..6b487801eeae 100644
--- a/drivers/video/fbmem.c
+++ b/drivers/video/fbmem.c
@@ -35,7 +35,6 @@
 #include <linux/device.h>
 #include <linux/efi.h>
 #include <linux/fb.h>
-#include <linux/major.h>
 
 #include <asm/fb.h>
 

From 689796a141cea79d745a4689c65dd01c39e5e100 Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert@linux-m68k.org>
Date: Sat, 26 Jul 2008 17:15:05 +0200
Subject: [PATCH 1104/1171] dsp56k: Fix BKL pushdown

commit 236b8756a2b6f90498d45b2c36d43e5372f2d4b8 ("dsp56k: BKL pushdown")
removed the `struct inode *inode' parameter from dsp56k_ioctl(), but
forgot to update the use of `inode' in the first line of the function.

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/dsp56k.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/char/dsp56k.c b/drivers/char/dsp56k.c
index 19b88504e960..ca7c72a486b2 100644
--- a/drivers/char/dsp56k.c
+++ b/drivers/char/dsp56k.c
@@ -304,9 +304,9 @@ static ssize_t dsp56k_write(struct file *file, const char __user *buf, size_t co
 }
 
 static long dsp56k_ioctl(struct file *file, unsigned int cmd,
-							unsigned long arg)
+			 unsigned long arg)
 {
-	int dev = iminor(inode) & 0x0f;
+	int dev = iminor(file->f_path.dentry->d_inode) & 0x0f;
 	void __user *argp = (void __user *)arg;
 
 	switch(dev)

From a9906a19193db69ad0158f289f839edf8aaf103f Mon Sep 17 00:00:00 2001
From: Roland McGrath <roland@redhat.com>
Date: Sat, 26 Jul 2008 14:41:26 -0700
Subject: [PATCH 1105/1171] tracehook: comment fixes

This fixes some typos and errors in <linux/tracehook.h> comments.
No code changes.

Signed-off-by: Roland McGrath <roland@redhat.com>
---
 include/linux/tracehook.h | 23 ++++++++++++-----------
 1 file changed, 12 insertions(+), 11 deletions(-)

diff --git a/include/linux/tracehook.h b/include/linux/tracehook.h
index 589f429619c9..b1875582c1a1 100644
--- a/include/linux/tracehook.h
+++ b/include/linux/tracehook.h
@@ -244,7 +244,7 @@ static inline int tracehook_prepare_clone(unsigned clone_flags)
  * tracehook_finish_clone - new child created and being attached
  * @child:		new child task
  * @clone_flags:	%CLONE_* flags from clone/fork/vfork system call
- * @trace:		return value from tracehook_clone_prepare()
+ * @trace:		return value from tracehook_prepare_clone()
  *
  * This is called immediately after adding @child to its parent's children list.
  * The @trace value is that returned by tracehook_prepare_clone().
@@ -259,19 +259,20 @@ static inline void tracehook_finish_clone(struct task_struct *child,
 
 /**
  * tracehook_report_clone - in parent, new child is about to start running
- * @trace:		return value from tracehook_clone_prepare()
+ * @trace:		return value from tracehook_prepare_clone()
  * @regs:		parent's user register state
  * @clone_flags:	flags from parent's system call
  * @pid:		new child's PID in the parent's namespace
  * @child:		new child task
  *
- * Called after a child is set up, but before it has been started running.
- * The @trace value is that returned by tracehook_clone_prepare().
- * This is not a good place to block, because the child has not started yet.
- * Suspend the child here if desired, and block in tracehook_clone_complete().
- * This must prevent the child from self-reaping if tracehook_clone_complete()
- * uses the @child pointer; otherwise it might have died and been released by
- * the time tracehook_report_clone_complete() is called.
+ * Called after a child is set up, but before it has been started
+ * running.  @trace is the value returned by tracehook_prepare_clone().
+ * This is not a good place to block, because the child has not started
+ * yet.  Suspend the child here if desired, and then block in
+ * tracehook_report_clone_complete().  This must prevent the child from
+ * self-reaping if tracehook_report_clone_complete() uses the @child
+ * pointer; otherwise it might have died and been released by the time
+ * tracehook_report_report_clone_complete() is called.
  *
  * Called with no locks held, but the child cannot run until this returns.
  */
@@ -290,7 +291,7 @@ static inline void tracehook_report_clone(int trace, struct pt_regs *regs,
 
 /**
  * tracehook_report_clone_complete - new child is running
- * @trace:		return value from tracehook_clone_prepare()
+ * @trace:		return value from tracehook_prepare_clone()
  * @regs:		parent's user register state
  * @clone_flags:	flags from parent's system call
  * @pid:		new child's PID in the parent's namespace
@@ -347,7 +348,7 @@ static inline void tracehook_prepare_release_task(struct task_struct *task)
 }
 
 /**
- * tracehook_finish_release_task - task is being reaped, clean up tracing
+ * tracehook_finish_release_task - final tracing clean-up
  * @task:		task in %EXIT_DEAD state
  *
  * This is called in release_task() when @task is being in the middle of

From 322069c9df1fd4da269b2a57d78f753684962523 Mon Sep 17 00:00:00 2001
From: Paulius Zaleckas <paulius.zaleckas@teltonika.lt>
Date: Wed, 9 Jul 2008 16:03:13 +0300
Subject: [PATCH 1106/1171] imxmmc: remove DEBUG definition

Removed DEBUG #define #undef, because module is automaticaly
compiled with -DDEBUG when CONFIG_MMC_DEBUG is defined.
Currently it just generates compiler warning about redefinition.

Signed-off-by: Paulius Zaleckas <paulius.zaleckas@teltonika.lt>
Acked-by: Pavel Pisa <pisa@cmp.felk.cvut.cz>
Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
---
 drivers/mmc/host/imxmmc.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/drivers/mmc/host/imxmmc.c b/drivers/mmc/host/imxmmc.c
index 5e880c0f1349..77493339303b 100644
--- a/drivers/mmc/host/imxmmc.c
+++ b/drivers/mmc/host/imxmmc.c
@@ -26,12 +26,6 @@
  *
  */
 
-#ifdef CONFIG_MMC_DEBUG
-#define DEBUG
-#else
-#undef  DEBUG
-#endif
-
 #include <linux/module.h>
 #include <linux/init.h>
 #include <linux/ioport.h>

From 5fc63dfba8a016caf832572aeaa90abef82f0ba0 Mon Sep 17 00:00:00 2001
From: Paulius Zaleckas <paulius.zaleckas@teltonika.lt>
Date: Wed, 9 Jul 2008 16:03:17 +0300
Subject: [PATCH 1107/1171] imxmmc: fix platform resources

Fixup platform resources handling.

Signed-off-by: Paulius Zaleckas <paulius.zaleckas@teltonika.lt>
Acked-by: Pavel Pisa <pisa@cmp.felk.cvut.cz>
Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
---
 drivers/mmc/host/imxmmc.c | 33 ++++++---------------------------
 1 file changed, 6 insertions(+), 27 deletions(-)

diff --git a/drivers/mmc/host/imxmmc.c b/drivers/mmc/host/imxmmc.c
index 77493339303b..9cda12fba9aa 100644
--- a/drivers/mmc/host/imxmmc.c
+++ b/drivers/mmc/host/imxmmc.c
@@ -901,26 +901,6 @@ static const struct mmc_host_ops imxmci_ops = {
 	.get_ro		= imxmci_get_ro,
 };
 
-static struct resource *platform_device_resource(struct platform_device *dev, unsigned int mask, int nr)
-{
-	int i;
-
-	for (i = 0; i < dev->num_resources; i++)
-		if (dev->resource[i].flags == mask && nr-- == 0)
-			return &dev->resource[i];
-	return NULL;
-}
-
-static int platform_device_irq(struct platform_device *dev, int nr)
-{
-	int i;
-
-	for (i = 0; i < dev->num_resources; i++)
-		if (dev->resource[i].flags == IORESOURCE_IRQ && nr-- == 0)
-			return dev->resource[i].start;
-	return NO_IRQ;
-}
-
 static void imxmci_check_status(unsigned long data)
 {
 	struct imxmci_host *host = (struct imxmci_host *)data;
@@ -956,13 +936,12 @@ static int imxmci_probe(struct platform_device *pdev)
 
 	printk(KERN_INFO "i.MX mmc driver\n");
 
-	r = platform_device_resource(pdev, IORESOURCE_MEM, 0);
-	irq = platform_device_irq(pdev, 0);
-	if (!r || irq == NO_IRQ)
+	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	irq = platform_get_irq(pdev, 0);
+	if (!r || irq < 0)
 		return -ENXIO;
 
-	r = request_mem_region(r->start, 0x100, "IMXMCI");
-	if (!r)
+	if (!request_mem_region(r->start, 0x100, pdev->name))
 		return -EBUSY;
 
 	mmc = mmc_alloc_host(sizeof(struct imxmci_host), &pdev->dev);
@@ -1067,7 +1046,7 @@ out:
 	}
 	if (mmc)
 		mmc_free_host(mmc);
-	release_resource(r);
+	release_mem_region(r->start, 0x100);
 	return ret;
 }
 
@@ -1096,7 +1075,7 @@ static int imxmci_remove(struct platform_device *pdev)
 		clk_disable(host->clk);
 		clk_put(host->clk);
 
-		release_resource(host->res);
+		release_mem_region(host->res->start, 0x100);
 
 		mmc_free_host(mmc);
 	}

From c5d5e9c40fc6cabedd5fdc7441e6e9d37f5c9bba Mon Sep 17 00:00:00 2001
From: Paulius Zaleckas <paulius.zaleckas@teltonika.lt>
Date: Wed, 9 Jul 2008 16:03:20 +0300
Subject: [PATCH 1108/1171] imxmmc: fix crash when no platform data is provided

Don't crash if no platform data is provided.
In this case assume that card is present.

Signed-off-by: Paulius Zaleckas <paulius.zaleckas@teltonika.lt>
Acked-by: Pavel Pisa <pisa@cmp.felk.cvut.cz>
Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
---
 drivers/mmc/host/imxmmc.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/drivers/mmc/host/imxmmc.c b/drivers/mmc/host/imxmmc.c
index 9cda12fba9aa..f61406da65d2 100644
--- a/drivers/mmc/host/imxmmc.c
+++ b/drivers/mmc/host/imxmmc.c
@@ -905,7 +905,8 @@ static void imxmci_check_status(unsigned long data)
 {
 	struct imxmci_host *host = (struct imxmci_host *)data;
 
-	if( host->pdata->card_present(mmc_dev(host->mmc)) != host->present ) {
+	if (host->pdata && host->pdata->card_present &&
+	    host->pdata->card_present(mmc_dev(host->mmc)) != host->present) {
 		host->present ^= 1;
 		dev_info(mmc_dev(host->mmc), "card %s\n",
 		      host->present ? "inserted" : "removed");
@@ -968,6 +969,8 @@ static int imxmci_probe(struct platform_device *pdev)
 	host->mmc = mmc;
 	host->dma_allocated = 0;
 	host->pdata = pdev->dev.platform_data;
+	if (!host->pdata)
+		dev_warn(&pdev->dev, "No platform data provided!\n");
 
 	spin_lock_init(&host->lock);
 	host->res = r;
@@ -1020,7 +1023,11 @@ static int imxmci_probe(struct platform_device *pdev)
 	if (ret)
 		goto out;
 
-	host->present = host->pdata->card_present(mmc_dev(mmc));
+	if (host->pdata && host->pdata->card_present)
+		host->present = host->pdata->card_present(mmc_dev(mmc));
+	else	/* if there is no way to detect assume that card is present */
+		host->present = 1;
+
 	init_timer(&host->timer);
 	host->timer.data = (unsigned long)host;
 	host->timer.function = imxmci_check_status;

From 6edd8ee60ac9b974bd6ec3b1bcb2aab02762fa8c Mon Sep 17 00:00:00 2001
From: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Date: Thu, 24 Jul 2008 14:18:57 +0200
Subject: [PATCH 1109/1171] mmc: Export internal host state through debugfs

When CONFIG_DEBUG_FS is set, create a few files under /sys/kernel/debug
containing information about an mmc host's internal state. Currently,
just a single file is created, "ios", which contains information about
the current operating parameters for the bus (clock speed, bus width,
etc.)

Host drivers can add additional files and directories under the host's
root directory by passing the debugfs_root field in struct mmc_host as
the 'parent' parameter to debugfs_create_*.

Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
---
 drivers/mmc/core/Makefile  |   1 +
 drivers/mmc/core/core.h    |   4 +
 drivers/mmc/core/debugfs.c | 164 +++++++++++++++++++++++++++++++++++++
 drivers/mmc/core/host.c    |   8 ++
 include/linux/mmc/host.h   |   2 +
 5 files changed, 179 insertions(+)
 create mode 100644 drivers/mmc/core/debugfs.c

diff --git a/drivers/mmc/core/Makefile b/drivers/mmc/core/Makefile
index 19a1a254a0c5..889e5f898f6f 100644
--- a/drivers/mmc/core/Makefile
+++ b/drivers/mmc/core/Makefile
@@ -12,3 +12,4 @@ mmc_core-y			:= core.o bus.o host.o \
 				   sdio.o sdio_ops.o sdio_bus.o \
 				   sdio_cis.o sdio_io.o sdio_irq.o
 
+mmc_core-$(CONFIG_DEBUG_FS)	+= debugfs.o
diff --git a/drivers/mmc/core/core.h b/drivers/mmc/core/core.h
index cdb332b7dedc..745da9881aa7 100644
--- a/drivers/mmc/core/core.h
+++ b/drivers/mmc/core/core.h
@@ -52,5 +52,9 @@ int mmc_attach_sdio(struct mmc_host *host, u32 ocr);
 
 extern int use_spi_crc;
 
+/* Debugfs information for hosts and cards */
+void mmc_add_host_debugfs(struct mmc_host *host);
+void mmc_remove_host_debugfs(struct mmc_host *host);
+
 #endif
 
diff --git a/drivers/mmc/core/debugfs.c b/drivers/mmc/core/debugfs.c
new file mode 100644
index 000000000000..133c6e51f26b
--- /dev/null
+++ b/drivers/mmc/core/debugfs.c
@@ -0,0 +1,164 @@
+/*
+ * Debugfs support for hosts and cards
+ *
+ * Copyright (C) 2008 Atmel Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/debugfs.h>
+#include <linux/fs.h>
+#include <linux/seq_file.h>
+#include <linux/stat.h>
+
+#include <linux/mmc/host.h>
+
+#include "core.h"
+
+/* The debugfs functions are optimized away when CONFIG_DEBUG_FS isn't set. */
+static int mmc_ios_show(struct seq_file *s, void *data)
+{
+	static const char *vdd_str[] = {
+		[8]	= "2.0",
+		[9]	= "2.1",
+		[10]	= "2.2",
+		[11]	= "2.3",
+		[12]	= "2.4",
+		[13]	= "2.5",
+		[14]	= "2.6",
+		[15]	= "2.7",
+		[16]	= "2.8",
+		[17]	= "2.9",
+		[18]	= "3.0",
+		[19]	= "3.1",
+		[20]	= "3.2",
+		[21]	= "3.3",
+		[22]	= "3.4",
+		[23]	= "3.5",
+		[24]	= "3.6",
+	};
+	struct mmc_host	*host = s->private;
+	struct mmc_ios	*ios = &host->ios;
+	const char *str;
+
+	seq_printf(s, "clock:\t\t%u Hz\n", ios->clock);
+	seq_printf(s, "vdd:\t\t%u ", ios->vdd);
+	if ((1 << ios->vdd) & MMC_VDD_165_195)
+		seq_printf(s, "(1.65 - 1.95 V)\n");
+	else if (ios->vdd < (ARRAY_SIZE(vdd_str) - 1)
+			&& vdd_str[ios->vdd] && vdd_str[ios->vdd + 1])
+		seq_printf(s, "(%s ~ %s V)\n", vdd_str[ios->vdd],
+				vdd_str[ios->vdd + 1]);
+	else
+		seq_printf(s, "(invalid)\n");
+
+	switch (ios->bus_mode) {
+	case MMC_BUSMODE_OPENDRAIN:
+		str = "open drain";
+		break;
+	case MMC_BUSMODE_PUSHPULL:
+		str = "push-pull";
+		break;
+	default:
+		str = "invalid";
+		break;
+	}
+	seq_printf(s, "bus mode:\t%u (%s)\n", ios->bus_mode, str);
+
+	switch (ios->chip_select) {
+	case MMC_CS_DONTCARE:
+		str = "don't care";
+		break;
+	case MMC_CS_HIGH:
+		str = "active high";
+		break;
+	case MMC_CS_LOW:
+		str = "active low";
+		break;
+	default:
+		str = "invalid";
+		break;
+	}
+	seq_printf(s, "chip select:\t%u (%s)\n", ios->chip_select, str);
+
+	switch (ios->power_mode) {
+	case MMC_POWER_OFF:
+		str = "off";
+		break;
+	case MMC_POWER_UP:
+		str = "up";
+		break;
+	case MMC_POWER_ON:
+		str = "on";
+		break;
+	default:
+		str = "invalid";
+		break;
+	}
+	seq_printf(s, "power mode:\t%u (%s)\n", ios->power_mode, str);
+	seq_printf(s, "bus width:\t%u (%u bits)\n",
+			ios->bus_width, 1 << ios->bus_width);
+
+	switch (ios->timing) {
+	case MMC_TIMING_LEGACY:
+		str = "legacy";
+		break;
+	case MMC_TIMING_MMC_HS:
+		str = "mmc high-speed";
+		break;
+	case MMC_TIMING_SD_HS:
+		str = "sd high-speed";
+		break;
+	default:
+		str = "invalid";
+		break;
+	}
+	seq_printf(s, "timing spec:\t%u (%s)\n", ios->timing, str);
+
+	return 0;
+}
+
+static int mmc_ios_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, mmc_ios_show, inode->i_private);
+}
+
+static const struct file_operations mmc_ios_fops = {
+	.open		= mmc_ios_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
+void mmc_add_host_debugfs(struct mmc_host *host)
+{
+	struct dentry *root;
+
+	root = debugfs_create_dir(mmc_hostname(host), NULL);
+	if (IS_ERR(root))
+		/* Don't complain -- debugfs just isn't enabled */
+		return;
+	if (!root)
+		/* Complain -- debugfs is enabled, but it failed to
+		 * create the directory. */
+		goto err_root;
+
+	host->debugfs_root = root;
+
+	if (!debugfs_create_file("ios", S_IRUSR, root, host, &mmc_ios_fops))
+		goto err_ios;
+
+	return;
+
+err_ios:
+	debugfs_remove_recursive(root);
+	host->debugfs_root = NULL;
+err_root:
+	dev_err(&host->class_dev, "failed to initialize debugfs\n");
+}
+
+void mmc_remove_host_debugfs(struct mmc_host *host)
+{
+	debugfs_remove_recursive(host->debugfs_root);
+}
diff --git a/drivers/mmc/core/host.c b/drivers/mmc/core/host.c
index 1d795c5379b5..6da80fd4d974 100644
--- a/drivers/mmc/core/host.c
+++ b/drivers/mmc/core/host.c
@@ -127,6 +127,10 @@ int mmc_add_host(struct mmc_host *host)
 	if (err)
 		return err;
 
+#ifdef CONFIG_DEBUG_FS
+	mmc_add_host_debugfs(host);
+#endif
+
 	mmc_start_host(host);
 
 	return 0;
@@ -146,6 +150,10 @@ void mmc_remove_host(struct mmc_host *host)
 {
 	mmc_stop_host(host);
 
+#ifdef CONFIG_DEBUG_FS
+	mmc_remove_host_debugfs(host);
+#endif
+
 	device_del(&host->class_dev);
 
 	led_trigger_unregister_simple(host->led);
diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h
index 10a2080086ca..9c288c909878 100644
--- a/include/linux/mmc/host.h
+++ b/include/linux/mmc/host.h
@@ -157,6 +157,8 @@ struct mmc_host {
 	struct led_trigger	*led;		/* activity led */
 #endif
 
+	struct dentry		*debugfs_root;
+
 	unsigned long		private[0] ____cacheline_aligned;
 };
 

From f4b7f927b531ca350cfc4ca1bdc3377dac7f9a32 Mon Sep 17 00:00:00 2001
From: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Date: Thu, 24 Jul 2008 14:18:58 +0200
Subject: [PATCH 1110/1171] mmc: Add per-card debugfs support

For each card successfully added to the bus, create a subdirectory under
the host's debugfs root with information about the card.

At the moment, only a single file is added to the card directory for
all cards: "state". It reflects the "state" field in struct mmc_card,
indicating whether the card is present, readonly, etc.

For MMC and SD cards (not SDIO), another file is added: "status".
Reading this file will ask the card about its current status and
return it. This can be useful if the card just refuses to respond to
any commands, which might indicate that the card state is not what the
MMC core thinks it is (due to a missing stop command, for example.)

Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
---
 drivers/mmc/core/bus.c     |  8 +++++
 drivers/mmc/core/core.h    |  3 ++
 drivers/mmc/core/debugfs.c | 61 ++++++++++++++++++++++++++++++++++++++
 include/linux/mmc/card.h   |  2 ++
 4 files changed, 74 insertions(+)

diff --git a/drivers/mmc/core/bus.c b/drivers/mmc/core/bus.c
index fd95b18e988b..0d9b2d6f9ebf 100644
--- a/drivers/mmc/core/bus.c
+++ b/drivers/mmc/core/bus.c
@@ -252,6 +252,10 @@ int mmc_add_card(struct mmc_card *card)
 	if (ret)
 		return ret;
 
+#ifdef CONFIG_DEBUG_FS
+	mmc_add_card_debugfs(card);
+#endif
+
 	mmc_card_set_present(card);
 
 	return 0;
@@ -263,6 +267,10 @@ int mmc_add_card(struct mmc_card *card)
  */
 void mmc_remove_card(struct mmc_card *card)
 {
+#ifdef CONFIG_DEBUG_FS
+	mmc_remove_card_debugfs(card);
+#endif
+
 	if (mmc_card_present(card)) {
 		if (mmc_host_is_spi(card->host)) {
 			printk(KERN_INFO "%s: SPI card removed\n",
diff --git a/drivers/mmc/core/core.h b/drivers/mmc/core/core.h
index 745da9881aa7..c819effa1032 100644
--- a/drivers/mmc/core/core.h
+++ b/drivers/mmc/core/core.h
@@ -56,5 +56,8 @@ extern int use_spi_crc;
 void mmc_add_host_debugfs(struct mmc_host *host);
 void mmc_remove_host_debugfs(struct mmc_host *host);
 
+void mmc_add_card_debugfs(struct mmc_card *card);
+void mmc_remove_card_debugfs(struct mmc_card *card);
+
 #endif
 
diff --git a/drivers/mmc/core/debugfs.c b/drivers/mmc/core/debugfs.c
index 133c6e51f26b..1237bb4c722b 100644
--- a/drivers/mmc/core/debugfs.c
+++ b/drivers/mmc/core/debugfs.c
@@ -12,9 +12,11 @@
 #include <linux/seq_file.h>
 #include <linux/stat.h>
 
+#include <linux/mmc/card.h>
 #include <linux/mmc/host.h>
 
 #include "core.h"
+#include "mmc_ops.h"
 
 /* The debugfs functions are optimized away when CONFIG_DEBUG_FS isn't set. */
 static int mmc_ios_show(struct seq_file *s, void *data)
@@ -162,3 +164,62 @@ void mmc_remove_host_debugfs(struct mmc_host *host)
 {
 	debugfs_remove_recursive(host->debugfs_root);
 }
+
+static int mmc_dbg_card_status_get(void *data, u64 *val)
+{
+	struct mmc_card	*card = data;
+	u32		status;
+	int		ret;
+
+	mmc_claim_host(card->host);
+
+	ret = mmc_send_status(data, &status);
+	if (!ret)
+		*val = status;
+
+	mmc_release_host(card->host);
+
+	return ret;
+}
+DEFINE_SIMPLE_ATTRIBUTE(mmc_dbg_card_status_fops, mmc_dbg_card_status_get,
+		NULL, "%08llx\n");
+
+void mmc_add_card_debugfs(struct mmc_card *card)
+{
+	struct mmc_host	*host = card->host;
+	struct dentry	*root;
+
+	if (!host->debugfs_root)
+		return;
+
+	root = debugfs_create_dir(mmc_card_id(card), host->debugfs_root);
+	if (IS_ERR(root))
+		/* Don't complain -- debugfs just isn't enabled */
+		return;
+	if (!root)
+		/* Complain -- debugfs is enabled, but it failed to
+		 * create the directory. */
+		goto err;
+
+	card->debugfs_root = root;
+
+	if (!debugfs_create_x32("state", S_IRUSR, root, &card->state))
+		goto err;
+
+	if (mmc_card_mmc(card) || mmc_card_sd(card))
+		if (!debugfs_create_file("status", S_IRUSR, root, card,
+					&mmc_dbg_card_status_fops))
+			goto err;
+
+	return;
+
+err:
+	debugfs_remove_recursive(root);
+	card->debugfs_root = NULL;
+	dev_err(&card->dev, "failed to initialize debugfs\n");
+}
+
+void mmc_remove_card_debugfs(struct mmc_card *card)
+{
+	debugfs_remove_recursive(card->debugfs_root);
+}
diff --git a/include/linux/mmc/card.h b/include/linux/mmc/card.h
index 0d508ac17d64..ee6e822d5994 100644
--- a/include/linux/mmc/card.h
+++ b/include/linux/mmc/card.h
@@ -111,6 +111,8 @@ struct mmc_card {
 	unsigned		num_info;	/* number of info strings */
 	const char		**info;		/* info strings */
 	struct sdio_func_tuple	*tuples;	/* unknown common tuples */
+
+	struct dentry		*debugfs_root;
 };
 
 #define mmc_card_mmc(c)		((c)->type == MMC_TYPE_MMC)

From deec9ae31e6079551ce9260d29a4cf83e5b19a83 Mon Sep 17 00:00:00 2001
From: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Date: Thu, 24 Jul 2008 14:18:59 +0200
Subject: [PATCH 1111/1171] atmel-mci: debugfs support

Create additional files under the host's debugfs directory containing
additional host-specific debug information.

Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
---
 drivers/mmc/host/atmel-mci-regs.h |   2 +
 drivers/mmc/host/atmel-mci.c      | 189 ++++++++++++++++++++++++++++++
 2 files changed, 191 insertions(+)

diff --git a/drivers/mmc/host/atmel-mci-regs.h b/drivers/mmc/host/atmel-mci-regs.h
index a9a5657706c6..26bd80e65031 100644
--- a/drivers/mmc/host/atmel-mci-regs.h
+++ b/drivers/mmc/host/atmel-mci-regs.h
@@ -82,6 +82,8 @@
 # define MCI_OVRE		(  1 <<  30)	/* RX Overrun Error */
 # define MCI_UNRE		(  1 <<  31)	/* TX Underrun Error */
 
+#define MCI_REGS_SIZE		0x100
+
 /* Register access macros */
 #define mci_readl(port,reg)				\
 	__raw_readl((port)->regs + MCI_##reg)
diff --git a/drivers/mmc/host/atmel-mci.c b/drivers/mmc/host/atmel-mci.c
index cce873c5a149..b68381f7bfdd 100644
--- a/drivers/mmc/host/atmel-mci.c
+++ b/drivers/mmc/host/atmel-mci.c
@@ -9,6 +9,7 @@
  */
 #include <linux/blkdev.h>
 #include <linux/clk.h>
+#include <linux/debugfs.h>
 #include <linux/device.h>
 #include <linux/init.h>
 #include <linux/interrupt.h>
@@ -16,6 +17,8 @@
 #include <linux/module.h>
 #include <linux/platform_device.h>
 #include <linux/scatterlist.h>
+#include <linux/seq_file.h>
+#include <linux/stat.h>
 
 #include <linux/mmc/host.h>
 
@@ -88,6 +91,188 @@ struct atmel_mci {
 #define atmci_clear_pending(host, event)			\
 	clear_bit(event, &host->pending_events)
 
+/*
+ * The debugfs stuff below is mostly optimized away when
+ * CONFIG_DEBUG_FS is not set.
+ */
+static int atmci_req_show(struct seq_file *s, void *v)
+{
+	struct atmel_mci	*host = s->private;
+	struct mmc_request	*mrq = host->mrq;
+	struct mmc_command	*cmd;
+	struct mmc_command	*stop;
+	struct mmc_data		*data;
+
+	/* Make sure we get a consistent snapshot */
+	spin_lock_irq(&host->mmc->lock);
+
+	if (mrq) {
+		cmd = mrq->cmd;
+		data = mrq->data;
+		stop = mrq->stop;
+
+		if (cmd)
+			seq_printf(s,
+				"CMD%u(0x%x) flg %x rsp %x %x %x %x err %d\n",
+				cmd->opcode, cmd->arg, cmd->flags,
+				cmd->resp[0], cmd->resp[1], cmd->resp[2],
+				cmd->resp[2], cmd->error);
+		if (data)
+			seq_printf(s, "DATA %u / %u * %u flg %x err %d\n",
+				data->bytes_xfered, data->blocks,
+				data->blksz, data->flags, data->error);
+		if (stop)
+			seq_printf(s,
+				"CMD%u(0x%x) flg %x rsp %x %x %x %x err %d\n",
+				stop->opcode, stop->arg, stop->flags,
+				stop->resp[0], stop->resp[1], stop->resp[2],
+				stop->resp[2], stop->error);
+	}
+
+	spin_unlock_irq(&host->mmc->lock);
+
+	return 0;
+}
+
+static int atmci_req_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, atmci_req_show, inode->i_private);
+}
+
+static const struct file_operations atmci_req_fops = {
+	.owner		= THIS_MODULE,
+	.open		= atmci_req_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
+static void atmci_show_status_reg(struct seq_file *s,
+		const char *regname, u32 value)
+{
+	static const char	*sr_bit[] = {
+		[0]	= "CMDRDY",
+		[1]	= "RXRDY",
+		[2]	= "TXRDY",
+		[3]	= "BLKE",
+		[4]	= "DTIP",
+		[5]	= "NOTBUSY",
+		[8]	= "SDIOIRQA",
+		[9]	= "SDIOIRQB",
+		[16]	= "RINDE",
+		[17]	= "RDIRE",
+		[18]	= "RCRCE",
+		[19]	= "RENDE",
+		[20]	= "RTOE",
+		[21]	= "DCRCE",
+		[22]	= "DTOE",
+		[30]	= "OVRE",
+		[31]	= "UNRE",
+	};
+	unsigned int		i;
+
+	seq_printf(s, "%s:\t0x%08x", regname, value);
+	for (i = 0; i < ARRAY_SIZE(sr_bit); i++) {
+		if (value & (1 << i)) {
+			if (sr_bit[i])
+				seq_printf(s, " %s", sr_bit[i]);
+			else
+				seq_puts(s, " UNKNOWN");
+		}
+	}
+	seq_putc(s, '\n');
+}
+
+static int atmci_regs_show(struct seq_file *s, void *v)
+{
+	struct atmel_mci	*host = s->private;
+	u32			*buf;
+
+	buf = kmalloc(MCI_REGS_SIZE, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	/* Grab a more or less consistent snapshot */
+	spin_lock_irq(&host->mmc->lock);
+	memcpy_fromio(buf, host->regs, MCI_REGS_SIZE);
+	spin_unlock_irq(&host->mmc->lock);
+
+	seq_printf(s, "MR:\t0x%08x%s%s CLKDIV=%u\n",
+			buf[MCI_MR / 4],
+			buf[MCI_MR / 4] & MCI_MR_RDPROOF ? " RDPROOF" : "",
+			buf[MCI_MR / 4] & MCI_MR_WRPROOF ? " WRPROOF" : "",
+			buf[MCI_MR / 4] & 0xff);
+	seq_printf(s, "DTOR:\t0x%08x\n", buf[MCI_DTOR / 4]);
+	seq_printf(s, "SDCR:\t0x%08x\n", buf[MCI_SDCR / 4]);
+	seq_printf(s, "ARGR:\t0x%08x\n", buf[MCI_ARGR / 4]);
+	seq_printf(s, "BLKR:\t0x%08x BCNT=%u BLKLEN=%u\n",
+			buf[MCI_BLKR / 4],
+			buf[MCI_BLKR / 4] & 0xffff,
+			(buf[MCI_BLKR / 4] >> 16) & 0xffff);
+
+	/* Don't read RSPR and RDR; it will consume the data there */
+
+	atmci_show_status_reg(s, "SR", buf[MCI_SR / 4]);
+	atmci_show_status_reg(s, "IMR", buf[MCI_IMR / 4]);
+
+	return 0;
+}
+
+static int atmci_regs_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, atmci_regs_show, inode->i_private);
+}
+
+static const struct file_operations atmci_regs_fops = {
+	.owner		= THIS_MODULE,
+	.open		= atmci_regs_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
+static void atmci_init_debugfs(struct atmel_mci *host)
+{
+	struct mmc_host	*mmc;
+	struct dentry	*root;
+	struct dentry	*node;
+	struct resource	*res;
+
+	mmc = host->mmc;
+	root = mmc->debugfs_root;
+	if (!root)
+		return;
+
+	node = debugfs_create_file("regs", S_IRUSR, root, host,
+			&atmci_regs_fops);
+	if (IS_ERR(node))
+		return;
+	if (!node)
+		goto err;
+
+	res = platform_get_resource(host->pdev, IORESOURCE_MEM, 0);
+	node->d_inode->i_size = res->end - res->start + 1;
+
+	node = debugfs_create_file("req", S_IRUSR, root, host, &atmci_req_fops);
+	if (!node)
+		goto err;
+
+	node = debugfs_create_x32("pending_events", S_IRUSR, root,
+				     (u32 *)&host->pending_events);
+	if (!node)
+		goto err;
+
+	node = debugfs_create_x32("completed_events", S_IRUSR, root,
+				     (u32 *)&host->completed_events);
+	if (!node)
+		goto err;
+
+	return;
+
+err:
+	dev_err(&host->pdev->dev,
+		"failed to initialize debugfs for controller\n");
+}
 
 static void atmci_enable(struct atmel_mci *host)
 {
@@ -905,6 +1090,8 @@ static int __init atmci_probe(struct platform_device *pdev)
 			"Atmel MCI controller at 0x%08lx irq %d\n",
 			host->mapbase, irq);
 
+	atmci_init_debugfs(host);
+
 	return 0;
 
 err_request_irq:
@@ -923,6 +1110,8 @@ static int __exit atmci_remove(struct platform_device *pdev)
 	platform_set_drvdata(pdev, NULL);
 
 	if (host) {
+		/* Debugfs stuff is cleaned up by mmc core */
+
 		if (host->detect_pin >= 0) {
 			int pin = host->detect_pin;
 

From e4ac9bc1f6686dcb8c34e2756aa93cc9546fa6ae Mon Sep 17 00:00:00 2001
From: Karsten Keil <kkeil@suse.de>
Date: Sat, 26 Jul 2008 18:52:11 +0200
Subject: [PATCH 1112/1171] Add mISDN driver

mISDN is a new modular ISDN driver, in the long term it should replace
the old I4L driver architecture for passiv ISDN cards.

Signed-off-by: Karsten Keil <kkeil@suse.de>
---
 Documentation/isdn/README.mISDN | 6 ++++++
 drivers/isdn/Kconfig            | 4 +++-
 drivers/isdn/Makefile           | 1 +
 drivers/isdn/hardware/Makefile  | 1 +
 4 files changed, 11 insertions(+), 1 deletion(-)
 create mode 100644 Documentation/isdn/README.mISDN

diff --git a/Documentation/isdn/README.mISDN b/Documentation/isdn/README.mISDN
new file mode 100644
index 000000000000..cd8bf920e77b
--- /dev/null
+++ b/Documentation/isdn/README.mISDN
@@ -0,0 +1,6 @@
+mISDN is a new modular ISDN driver, in the long term it should replace
+the old I4L driver architecture for passiv ISDN cards.
+It was designed to allow a broad range of applications and interfaces
+but only have the basic function in kernel, the interface to the user
+space is based on sockets with a own address family AF_ISDN.
+
diff --git a/drivers/isdn/Kconfig b/drivers/isdn/Kconfig
index 66f946aa30b3..3d113c6e4a70 100644
--- a/drivers/isdn/Kconfig
+++ b/drivers/isdn/Kconfig
@@ -3,7 +3,7 @@
 #
 
 menuconfig ISDN
-	tristate "ISDN support"
+	bool "ISDN support"
 	depends on NET
 	depends on !S390
 	---help---
@@ -21,6 +21,8 @@ menuconfig ISDN
 
 if ISDN
 
+source "drivers/isdn/mISDN/Kconfig"
+
 menuconfig ISDN_I4L
 	tristate "Old ISDN4Linux (deprecated)"
 	---help---
diff --git a/drivers/isdn/Makefile b/drivers/isdn/Makefile
index 988142c30a6d..8380a4568d11 100644
--- a/drivers/isdn/Makefile
+++ b/drivers/isdn/Makefile
@@ -4,6 +4,7 @@
 
 obj-$(CONFIG_ISDN_I4L)			+= i4l/
 obj-$(CONFIG_ISDN_CAPI)			+= capi/
+obj-$(CONFIG_MISDN)			+= mISDN/
 obj-$(CONFIG_ISDN_CAPI)			+= hardware/
 obj-$(CONFIG_ISDN_DIVERSION)		+= divert/
 obj-$(CONFIG_ISDN_DRV_HISAX)		+= hisax/
diff --git a/drivers/isdn/hardware/Makefile b/drivers/isdn/hardware/Makefile
index 11c8a183948c..a5d8fce4c4c4 100644
--- a/drivers/isdn/hardware/Makefile
+++ b/drivers/isdn/hardware/Makefile
@@ -4,3 +4,4 @@
 
 obj-$(CONFIG_CAPI_AVM)		+= avm/
 obj-$(CONFIG_CAPI_EICON)	+= eicon/
+obj-$(CONFIG_MISDN)		+= mISDN/

From 04578dd330f1ec6bc9c4233833bee0d0ca73ff09 Mon Sep 17 00:00:00 2001
From: Karsten Keil <kkeil@suse.de>
Date: Sat, 26 Jul 2008 18:52:34 +0200
Subject: [PATCH 1113/1171] Define AF_ISDN and PF_ISDN

Define the address and protocol family value for mISDN.

Signed-off-by: Karsten Keil <kkeil@suse.de>
---
 include/linux/socket.h | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/include/linux/socket.h b/include/linux/socket.h
index 950af631e7fb..dc5086fe7736 100644
--- a/include/linux/socket.h
+++ b/include/linux/socket.h
@@ -189,7 +189,8 @@ struct ucred {
 #define AF_BLUETOOTH	31	/* Bluetooth sockets 		*/
 #define AF_IUCV		32	/* IUCV sockets			*/
 #define AF_RXRPC	33	/* RxRPC sockets 		*/
-#define AF_MAX		34	/* For now.. */
+#define AF_ISDN		34	/* mISDN sockets 		*/
+#define AF_MAX		35	/* For now.. */
 
 /* Protocol families, same as address families. */
 #define PF_UNSPEC	AF_UNSPEC
@@ -225,6 +226,7 @@ struct ucred {
 #define PF_BLUETOOTH	AF_BLUETOOTH
 #define PF_IUCV		AF_IUCV
 #define PF_RXRPC	AF_RXRPC
+#define PF_ISDN		AF_ISDN
 #define PF_MAX		AF_MAX
 
 /* Maximum queue length specifiable by listen.  */

From 1b2b03f8e514e4f68e293846ba511a948b80243c Mon Sep 17 00:00:00 2001
From: Karsten Keil <kkeil@suse.de>
Date: Sun, 27 Jul 2008 01:54:58 +0200
Subject: [PATCH 1114/1171] Add mISDN core files

Add mISDN core files

Signed-off-by: Karsten Keil <kkeil@suse.de>
---
 drivers/isdn/mISDN/Kconfig     |    9 +
 drivers/isdn/mISDN/Makefile    |    9 +
 drivers/isdn/mISDN/core.c      |  244 ++++
 drivers/isdn/mISDN/core.h      |   77 ++
 drivers/isdn/mISDN/fsm.c       |  183 +++
 drivers/isdn/mISDN/fsm.h       |   67 +
 drivers/isdn/mISDN/hwchannel.c |  365 ++++++
 drivers/isdn/mISDN/layer1.c    |  403 ++++++
 drivers/isdn/mISDN/layer1.h    |   26 +
 drivers/isdn/mISDN/layer2.c    | 2216 ++++++++++++++++++++++++++++++++
 drivers/isdn/mISDN/layer2.h    |  140 ++
 drivers/isdn/mISDN/socket.c    |  781 +++++++++++
 drivers/isdn/mISDN/stack.c     |  674 ++++++++++
 drivers/isdn/mISDN/tei.c       | 1340 +++++++++++++++++++
 drivers/isdn/mISDN/timerdev.c  |  301 +++++
 include/linux/mISDNhw.h        |  193 +++
 include/linux/mISDNif.h        |  487 +++++++
 17 files changed, 7515 insertions(+)
 create mode 100644 drivers/isdn/mISDN/Kconfig
 create mode 100644 drivers/isdn/mISDN/Makefile
 create mode 100644 drivers/isdn/mISDN/core.c
 create mode 100644 drivers/isdn/mISDN/core.h
 create mode 100644 drivers/isdn/mISDN/fsm.c
 create mode 100644 drivers/isdn/mISDN/fsm.h
 create mode 100644 drivers/isdn/mISDN/hwchannel.c
 create mode 100644 drivers/isdn/mISDN/layer1.c
 create mode 100644 drivers/isdn/mISDN/layer1.h
 create mode 100644 drivers/isdn/mISDN/layer2.c
 create mode 100644 drivers/isdn/mISDN/layer2.h
 create mode 100644 drivers/isdn/mISDN/socket.c
 create mode 100644 drivers/isdn/mISDN/stack.c
 create mode 100644 drivers/isdn/mISDN/tei.c
 create mode 100644 drivers/isdn/mISDN/timerdev.c
 create mode 100644 include/linux/mISDNhw.h
 create mode 100644 include/linux/mISDNif.h

diff --git a/drivers/isdn/mISDN/Kconfig b/drivers/isdn/mISDN/Kconfig
new file mode 100644
index 000000000000..231bd0d08316
--- /dev/null
+++ b/drivers/isdn/mISDN/Kconfig
@@ -0,0 +1,9 @@
+#
+# modularer ISDN driver
+#
+
+menuconfig MISDN
+	tristate "Modular ISDN driver"
+	help
+	  Enable support for the modular ISDN driver.
+
diff --git a/drivers/isdn/mISDN/Makefile b/drivers/isdn/mISDN/Makefile
new file mode 100644
index 000000000000..87c563d33612
--- /dev/null
+++ b/drivers/isdn/mISDN/Makefile
@@ -0,0 +1,9 @@
+#
+# Makefile for the modular ISDN driver
+#
+
+obj-$(CONFIG_MISDN) += mISDN_core.o
+
+# multi objects
+
+mISDN_core-objs := core.o fsm.o socket.o hwchannel.o stack.o layer1.o layer2.o tei.o timerdev.o
diff --git a/drivers/isdn/mISDN/core.c b/drivers/isdn/mISDN/core.c
new file mode 100644
index 000000000000..33068177b7c9
--- /dev/null
+++ b/drivers/isdn/mISDN/core.c
@@ -0,0 +1,244 @@
+/*
+ * Copyright 2008  by Karsten Keil <kkeil@novell.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/types.h>
+#include <linux/stddef.h>
+#include <linux/module.h>
+#include <linux/spinlock.h>
+#include <linux/mISDNif.h>
+#include "core.h"
+
+static u_int debug;
+
+MODULE_AUTHOR("Karsten Keil");
+MODULE_LICENSE("GPL");
+module_param(debug, uint, S_IRUGO | S_IWUSR);
+
+static LIST_HEAD(devices);
+DEFINE_RWLOCK(device_lock);
+static u64		device_ids;
+#define MAX_DEVICE_ID	63
+
+static LIST_HEAD(Bprotocols);
+DEFINE_RWLOCK(bp_lock);
+
+struct mISDNdevice
+*get_mdevice(u_int id)
+{
+	struct mISDNdevice	*dev;
+
+	read_lock(&device_lock);
+	list_for_each_entry(dev, &devices, D.list)
+		if (dev->id == id) {
+			read_unlock(&device_lock);
+			return dev;
+		}
+	read_unlock(&device_lock);
+	return NULL;
+}
+
+int
+get_mdevice_count(void)
+{
+	struct mISDNdevice	*dev;
+	int			cnt = 0;
+
+	read_lock(&device_lock);
+	list_for_each_entry(dev, &devices, D.list)
+		cnt++;
+	read_unlock(&device_lock);
+	return cnt;
+}
+
+static int
+get_free_devid(void)
+{
+	u_int	i;
+
+	for (i = 0; i <= MAX_DEVICE_ID; i++)
+		if (!test_and_set_bit(i, (u_long *)&device_ids))
+			return i;
+	return -1;
+}
+
+int
+mISDN_register_device(struct mISDNdevice *dev, char *name)
+{
+	u_long	flags;
+	int	err;
+
+	dev->id = get_free_devid();
+	if (dev->id < 0)
+		return -EBUSY;
+	if (name && name[0])
+		strcpy(dev->name, name);
+	else
+		sprintf(dev->name, "mISDN%d", dev->id);
+	if (debug & DEBUG_CORE)
+		printk(KERN_DEBUG "mISDN_register %s %d\n",
+			dev->name, dev->id);
+	err = create_stack(dev);
+	if (err)
+		return err;
+	write_lock_irqsave(&device_lock, flags);
+	list_add_tail(&dev->D.list, &devices);
+	write_unlock_irqrestore(&device_lock, flags);
+	return 0;
+}
+EXPORT_SYMBOL(mISDN_register_device);
+
+void
+mISDN_unregister_device(struct mISDNdevice *dev) {
+	u_long	flags;
+
+	if (debug & DEBUG_CORE)
+		printk(KERN_DEBUG "mISDN_unregister %s %d\n",
+			dev->name, dev->id);
+	write_lock_irqsave(&device_lock, flags);
+	list_del(&dev->D.list);
+	write_unlock_irqrestore(&device_lock, flags);
+	test_and_clear_bit(dev->id, (u_long *)&device_ids);
+	delete_stack(dev);
+}
+EXPORT_SYMBOL(mISDN_unregister_device);
+
+u_int
+get_all_Bprotocols(void)
+{
+	struct Bprotocol	*bp;
+	u_int	m = 0;
+
+	read_lock(&bp_lock);
+	list_for_each_entry(bp, &Bprotocols, list)
+		m |= bp->Bprotocols;
+	read_unlock(&bp_lock);
+	return m;
+}
+
+struct Bprotocol *
+get_Bprotocol4mask(u_int m)
+{
+	struct Bprotocol	*bp;
+
+	read_lock(&bp_lock);
+	list_for_each_entry(bp, &Bprotocols, list)
+		if (bp->Bprotocols & m) {
+			read_unlock(&bp_lock);
+			return bp;
+		}
+	read_unlock(&bp_lock);
+	return NULL;
+}
+
+struct Bprotocol *
+get_Bprotocol4id(u_int id)
+{
+	u_int	m;
+
+	if (id < ISDN_P_B_START || id > 63) {
+		printk(KERN_WARNING "%s id not in range  %d\n",
+		    __func__, id);
+		return NULL;
+	}
+	m = 1 << (id & ISDN_P_B_MASK);
+	return get_Bprotocol4mask(m);
+}
+
+int
+mISDN_register_Bprotocol(struct Bprotocol *bp)
+{
+	u_long			flags;
+	struct Bprotocol	*old;
+
+	if (debug & DEBUG_CORE)
+		printk(KERN_DEBUG "%s: %s/%x\n", __func__,
+		    bp->name, bp->Bprotocols);
+	old = get_Bprotocol4mask(bp->Bprotocols);
+	if (old) {
+		printk(KERN_WARNING
+		    "register duplicate protocol old %s/%x new %s/%x\n",
+		    old->name, old->Bprotocols, bp->name, bp->Bprotocols);
+		return -EBUSY;
+	}
+	write_lock_irqsave(&bp_lock, flags);
+	list_add_tail(&bp->list, &Bprotocols);
+	write_unlock_irqrestore(&bp_lock, flags);
+	return 0;
+}
+EXPORT_SYMBOL(mISDN_register_Bprotocol);
+
+void
+mISDN_unregister_Bprotocol(struct Bprotocol *bp)
+{
+	u_long	flags;
+
+	if (debug & DEBUG_CORE)
+		printk(KERN_DEBUG "%s: %s/%x\n", __func__, bp->name,
+			bp->Bprotocols);
+	write_lock_irqsave(&bp_lock, flags);
+	list_del(&bp->list);
+	write_unlock_irqrestore(&bp_lock, flags);
+}
+EXPORT_SYMBOL(mISDN_unregister_Bprotocol);
+
+int
+mISDNInit(void)
+{
+	int	err;
+
+	printk(KERN_INFO "Modular ISDN core version %d.%d.%d\n",
+		MISDN_MAJOR_VERSION, MISDN_MINOR_VERSION, MISDN_RELEASE);
+	mISDN_initstack(&debug);
+	err = mISDN_inittimer(&debug);
+	if (err)
+		goto error;
+	err = l1_init(&debug);
+	if (err) {
+		mISDN_timer_cleanup();
+		goto error;
+	}
+	err = Isdnl2_Init(&debug);
+	if (err) {
+		mISDN_timer_cleanup();
+		l1_cleanup();
+		goto error;
+	}
+	err = misdn_sock_init(&debug);
+	if (err) {
+		mISDN_timer_cleanup();
+		l1_cleanup();
+		Isdnl2_cleanup();
+	}
+error:
+	return err;
+}
+
+void mISDN_cleanup(void)
+{
+	misdn_sock_cleanup();
+	mISDN_timer_cleanup();
+	l1_cleanup();
+	Isdnl2_cleanup();
+
+	if (!list_empty(&devices))
+		printk(KERN_ERR "%s devices still registered\n", __func__);
+
+	if (!list_empty(&Bprotocols))
+		printk(KERN_ERR "%s Bprotocols still registered\n", __func__);
+	printk(KERN_DEBUG "mISDNcore unloaded\n");
+}
+
+module_init(mISDNInit);
+module_exit(mISDN_cleanup);
+
diff --git a/drivers/isdn/mISDN/core.h b/drivers/isdn/mISDN/core.h
new file mode 100644
index 000000000000..7da7233b4c1a
--- /dev/null
+++ b/drivers/isdn/mISDN/core.h
@@ -0,0 +1,77 @@
+/*
+ * Copyright 2008  by Karsten Keil <kkeil@novell.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef mISDN_CORE_H
+#define mISDN_CORE_H
+
+extern struct mISDNdevice	*get_mdevice(u_int);
+extern int			get_mdevice_count(void);
+
+/* stack status flag */
+#define mISDN_STACK_ACTION_MASK		0x0000ffff
+#define mISDN_STACK_COMMAND_MASK	0x000f0000
+#define mISDN_STACK_STATUS_MASK		0xfff00000
+/* action bits 0-15 */
+#define mISDN_STACK_WORK	0
+#define mISDN_STACK_SETUP	1
+#define mISDN_STACK_CLEARING	2
+#define mISDN_STACK_RESTART	3
+#define mISDN_STACK_WAKEUP	4
+#define mISDN_STACK_ABORT	15
+/* command bits 16-19 */
+#define mISDN_STACK_STOPPED	16
+#define mISDN_STACK_INIT	17
+#define mISDN_STACK_THREADSTART	18
+/* status bits 20-31 */
+#define mISDN_STACK_BCHANNEL	20
+#define mISDN_STACK_ACTIVE      29
+#define mISDN_STACK_RUNNING     30
+#define mISDN_STACK_KILLED      31
+
+
+/* manager options */
+#define MGR_OPT_USER		24
+#define MGR_OPT_NETWORK		25
+
+extern int	connect_Bstack(struct mISDNdevice *, struct mISDNchannel *,
+			u_int, struct sockaddr_mISDN *);
+extern int	connect_layer1(struct mISDNdevice *, struct mISDNchannel *,
+			u_int, struct sockaddr_mISDN *);
+extern int	create_l2entity(struct mISDNdevice *, struct mISDNchannel *,
+			u_int, struct sockaddr_mISDN *);
+
+extern int	create_stack(struct mISDNdevice *);
+extern int	create_teimanager(struct mISDNdevice *);
+extern void	delete_teimanager(struct mISDNchannel *);
+extern void	delete_channel(struct mISDNchannel *);
+extern void	delete_stack(struct mISDNdevice *);
+extern void	mISDN_initstack(u_int *);
+extern int      misdn_sock_init(u_int *);
+extern void     misdn_sock_cleanup(void);
+extern void	add_layer2(struct mISDNchannel *, struct mISDNstack *);
+extern void	__add_layer2(struct mISDNchannel *, struct mISDNstack *);
+
+extern u_int		get_all_Bprotocols(void);
+struct Bprotocol	*get_Bprotocol4mask(u_int);
+struct Bprotocol	*get_Bprotocol4id(u_int);
+
+extern int	mISDN_inittimer(u_int *);
+extern void	mISDN_timer_cleanup(void);
+
+extern int	l1_init(u_int *);
+extern void	l1_cleanup(void);
+extern int 	Isdnl2_Init(u_int *);
+extern void	Isdnl2_cleanup(void);
+
+#endif
diff --git a/drivers/isdn/mISDN/fsm.c b/drivers/isdn/mISDN/fsm.c
new file mode 100644
index 000000000000..b5d6553f2dc8
--- /dev/null
+++ b/drivers/isdn/mISDN/fsm.c
@@ -0,0 +1,183 @@
+/*
+ * finite state machine implementation
+ *
+ * Author       Karsten Keil <kkeil@novell.com>
+ *
+ * Thanks to    Jan den Ouden
+ *              Fritz Elfert
+ * Copyright 2008  by Karsten Keil <kkeil@novell.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/module.h>
+#include <linux/string.h>
+#include "fsm.h"
+
+#define FSM_TIMER_DEBUG 0
+
+void
+mISDN_FsmNew(struct Fsm *fsm,
+       struct FsmNode *fnlist, int fncount)
+{
+	int i;
+
+	fsm->jumpmatrix = kzalloc(sizeof(FSMFNPTR) * fsm->state_count *
+		fsm->event_count, GFP_KERNEL);
+
+	for (i = 0; i < fncount; i++)
+		if ((fnlist[i].state >= fsm->state_count) ||
+		    (fnlist[i].event >= fsm->event_count)) {
+			printk(KERN_ERR
+			    "mISDN_FsmNew Error: %d st(%ld/%ld) ev(%ld/%ld)\n",
+			    i, (long)fnlist[i].state, (long)fsm->state_count,
+			    (long)fnlist[i].event, (long)fsm->event_count);
+		} else
+			fsm->jumpmatrix[fsm->state_count * fnlist[i].event +
+			    fnlist[i].state] = (FSMFNPTR) fnlist[i].routine;
+}
+EXPORT_SYMBOL(mISDN_FsmNew);
+
+void
+mISDN_FsmFree(struct Fsm *fsm)
+{
+	kfree((void *) fsm->jumpmatrix);
+}
+EXPORT_SYMBOL(mISDN_FsmFree);
+
+int
+mISDN_FsmEvent(struct FsmInst *fi, int event, void *arg)
+{
+	FSMFNPTR r;
+
+	if ((fi->state >= fi->fsm->state_count) ||
+	    (event >= fi->fsm->event_count)) {
+		printk(KERN_ERR
+		    "mISDN_FsmEvent Error st(%ld/%ld) ev(%d/%ld)\n",
+		    (long)fi->state, (long)fi->fsm->state_count, event,
+		    (long)fi->fsm->event_count);
+		return 1;
+	}
+	r = fi->fsm->jumpmatrix[fi->fsm->state_count * event + fi->state];
+	if (r) {
+		if (fi->debug)
+			fi->printdebug(fi, "State %s Event %s",
+				fi->fsm->strState[fi->state],
+				fi->fsm->strEvent[event]);
+		r(fi, event, arg);
+		return 0;
+	} else {
+		if (fi->debug)
+			fi->printdebug(fi, "State %s Event %s no action",
+				fi->fsm->strState[fi->state],
+				fi->fsm->strEvent[event]);
+		return 1;
+	}
+}
+EXPORT_SYMBOL(mISDN_FsmEvent);
+
+void
+mISDN_FsmChangeState(struct FsmInst *fi, int newstate)
+{
+	fi->state = newstate;
+	if (fi->debug)
+		fi->printdebug(fi, "ChangeState %s",
+			fi->fsm->strState[newstate]);
+}
+EXPORT_SYMBOL(mISDN_FsmChangeState);
+
+static void
+FsmExpireTimer(struct FsmTimer *ft)
+{
+#if FSM_TIMER_DEBUG
+	if (ft->fi->debug)
+		ft->fi->printdebug(ft->fi, "FsmExpireTimer %lx", (long) ft);
+#endif
+	mISDN_FsmEvent(ft->fi, ft->event, ft->arg);
+}
+
+void
+mISDN_FsmInitTimer(struct FsmInst *fi, struct FsmTimer *ft)
+{
+	ft->fi = fi;
+	ft->tl.function = (void *) FsmExpireTimer;
+	ft->tl.data = (long) ft;
+#if FSM_TIMER_DEBUG
+	if (ft->fi->debug)
+		ft->fi->printdebug(ft->fi, "mISDN_FsmInitTimer %lx", (long) ft);
+#endif
+	init_timer(&ft->tl);
+}
+EXPORT_SYMBOL(mISDN_FsmInitTimer);
+
+void
+mISDN_FsmDelTimer(struct FsmTimer *ft, int where)
+{
+#if FSM_TIMER_DEBUG
+	if (ft->fi->debug)
+		ft->fi->printdebug(ft->fi, "mISDN_FsmDelTimer %lx %d",
+			(long) ft, where);
+#endif
+	del_timer(&ft->tl);
+}
+EXPORT_SYMBOL(mISDN_FsmDelTimer);
+
+int
+mISDN_FsmAddTimer(struct FsmTimer *ft,
+	    int millisec, int event, void *arg, int where)
+{
+
+#if FSM_TIMER_DEBUG
+	if (ft->fi->debug)
+		ft->fi->printdebug(ft->fi, "mISDN_FsmAddTimer %lx %d %d",
+			(long) ft, millisec, where);
+#endif
+
+	if (timer_pending(&ft->tl)) {
+		if (ft->fi->debug) {
+			printk(KERN_WARNING
+				"mISDN_FsmAddTimer: timer already active!\n");
+			ft->fi->printdebug(ft->fi,
+				"mISDN_FsmAddTimer already active!");
+		}
+		return -1;
+	}
+	init_timer(&ft->tl);
+	ft->event = event;
+	ft->arg = arg;
+	ft->tl.expires = jiffies + (millisec * HZ) / 1000;
+	add_timer(&ft->tl);
+	return 0;
+}
+EXPORT_SYMBOL(mISDN_FsmAddTimer);
+
+void
+mISDN_FsmRestartTimer(struct FsmTimer *ft,
+	    int millisec, int event, void *arg, int where)
+{
+
+#if FSM_TIMER_DEBUG
+	if (ft->fi->debug)
+		ft->fi->printdebug(ft->fi, "mISDN_FsmRestartTimer %lx %d %d",
+			(long) ft, millisec, where);
+#endif
+
+	if (timer_pending(&ft->tl))
+		del_timer(&ft->tl);
+	init_timer(&ft->tl);
+	ft->event = event;
+	ft->arg = arg;
+	ft->tl.expires = jiffies + (millisec * HZ) / 1000;
+	add_timer(&ft->tl);
+}
+EXPORT_SYMBOL(mISDN_FsmRestartTimer);
diff --git a/drivers/isdn/mISDN/fsm.h b/drivers/isdn/mISDN/fsm.h
new file mode 100644
index 000000000000..928f5be192c1
--- /dev/null
+++ b/drivers/isdn/mISDN/fsm.h
@@ -0,0 +1,67 @@
+/*
+ *
+ * Author       Karsten Keil <kkeil@novell.com>
+ *
+ * Thanks to    Jan den Ouden
+ *              Fritz Elfert
+ * Copyright 2008  by Karsten Keil <kkeil@novell.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef _MISDN_FSM_H
+#define _MISDN_FSM_H
+
+#include <linux/timer.h>
+
+/* Statemachine */
+
+struct FsmInst;
+
+typedef void (*FSMFNPTR)(struct FsmInst *, int, void *);
+
+struct Fsm {
+	FSMFNPTR *jumpmatrix;
+	int state_count, event_count;
+	char **strEvent, **strState;
+};
+
+struct FsmInst {
+	struct Fsm *fsm;
+	int state;
+	int debug;
+	void *userdata;
+	int userint;
+	void (*printdebug) (struct FsmInst *, char *, ...);
+};
+
+struct FsmNode {
+	int state, event;
+	void (*routine) (struct FsmInst *, int, void *);
+};
+
+struct FsmTimer {
+	struct FsmInst *fi;
+	struct timer_list tl;
+	int event;
+	void *arg;
+};
+
+extern void mISDN_FsmNew(struct Fsm *, struct FsmNode *, int);
+extern void mISDN_FsmFree(struct Fsm *);
+extern int mISDN_FsmEvent(struct FsmInst *, int , void *);
+extern void mISDN_FsmChangeState(struct FsmInst *, int);
+extern void mISDN_FsmInitTimer(struct FsmInst *, struct FsmTimer *);
+extern int mISDN_FsmAddTimer(struct FsmTimer *, int, int, void *, int);
+extern void mISDN_FsmRestartTimer(struct FsmTimer *, int, int, void *, int);
+extern void mISDN_FsmDelTimer(struct FsmTimer *, int);
+
+#endif
diff --git a/drivers/isdn/mISDN/hwchannel.c b/drivers/isdn/mISDN/hwchannel.c
new file mode 100644
index 000000000000..2596fba4e614
--- /dev/null
+++ b/drivers/isdn/mISDN/hwchannel.c
@@ -0,0 +1,365 @@
+/*
+ *
+ * Author	Karsten Keil <kkeil@novell.com>
+ *
+ * Copyright 2008  by Karsten Keil <kkeil@novell.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/mISDNhw.h>
+
+static void
+dchannel_bh(struct work_struct *ws)
+{
+	struct dchannel	*dch  = container_of(ws, struct dchannel, workq);
+	struct sk_buff	*skb;
+	int		err;
+
+	if (test_and_clear_bit(FLG_RECVQUEUE, &dch->Flags)) {
+		while ((skb = skb_dequeue(&dch->rqueue))) {
+			if (likely(dch->dev.D.peer)) {
+				err = dch->dev.D.recv(dch->dev.D.peer, skb);
+				if (err)
+					dev_kfree_skb(skb);
+			} else
+				dev_kfree_skb(skb);
+		}
+	}
+	if (test_and_clear_bit(FLG_PHCHANGE, &dch->Flags)) {
+		if (dch->phfunc)
+			dch->phfunc(dch);
+	}
+}
+
+static void
+bchannel_bh(struct work_struct *ws)
+{
+	struct bchannel	*bch  = container_of(ws, struct bchannel, workq);
+	struct sk_buff	*skb;
+	int		err;
+
+	if (test_and_clear_bit(FLG_RECVQUEUE, &bch->Flags)) {
+		while ((skb = skb_dequeue(&bch->rqueue))) {
+			if (bch->rcount >= 64)
+				printk(KERN_WARNING "B-channel %p receive "
+					"queue if full, but empties...\n", bch);
+			bch->rcount--;
+			if (likely(bch->ch.peer)) {
+				err = bch->ch.recv(bch->ch.peer, skb);
+				if (err)
+					dev_kfree_skb(skb);
+			} else
+				dev_kfree_skb(skb);
+		}
+	}
+}
+
+int
+mISDN_initdchannel(struct dchannel *ch, int maxlen, void *phf)
+{
+	test_and_set_bit(FLG_HDLC, &ch->Flags);
+	ch->maxlen = maxlen;
+	ch->hw = NULL;
+	ch->rx_skb = NULL;
+	ch->tx_skb = NULL;
+	ch->tx_idx = 0;
+	ch->phfunc = phf;
+	skb_queue_head_init(&ch->squeue);
+	skb_queue_head_init(&ch->rqueue);
+	INIT_LIST_HEAD(&ch->dev.bchannels);
+	INIT_WORK(&ch->workq, dchannel_bh);
+	return 0;
+}
+EXPORT_SYMBOL(mISDN_initdchannel);
+
+int
+mISDN_initbchannel(struct bchannel *ch, int maxlen)
+{
+	ch->Flags = 0;
+	ch->maxlen = maxlen;
+	ch->hw = NULL;
+	ch->rx_skb = NULL;
+	ch->tx_skb = NULL;
+	ch->tx_idx = 0;
+	skb_queue_head_init(&ch->rqueue);
+	ch->rcount = 0;
+	ch->next_skb = NULL;
+	INIT_WORK(&ch->workq, bchannel_bh);
+	return 0;
+}
+EXPORT_SYMBOL(mISDN_initbchannel);
+
+int
+mISDN_freedchannel(struct dchannel *ch)
+{
+	if (ch->tx_skb) {
+		dev_kfree_skb(ch->tx_skb);
+		ch->tx_skb = NULL;
+	}
+	if (ch->rx_skb) {
+		dev_kfree_skb(ch->rx_skb);
+		ch->rx_skb = NULL;
+	}
+	skb_queue_purge(&ch->squeue);
+	skb_queue_purge(&ch->rqueue);
+	flush_scheduled_work();
+	return 0;
+}
+EXPORT_SYMBOL(mISDN_freedchannel);
+
+int
+mISDN_freebchannel(struct bchannel *ch)
+{
+	if (ch->tx_skb) {
+		dev_kfree_skb(ch->tx_skb);
+		ch->tx_skb = NULL;
+	}
+	if (ch->rx_skb) {
+		dev_kfree_skb(ch->rx_skb);
+		ch->rx_skb = NULL;
+	}
+	if (ch->next_skb) {
+		dev_kfree_skb(ch->next_skb);
+		ch->next_skb = NULL;
+	}
+	skb_queue_purge(&ch->rqueue);
+	ch->rcount = 0;
+	flush_scheduled_work();
+	return 0;
+}
+EXPORT_SYMBOL(mISDN_freebchannel);
+
+static inline u_int
+get_sapi_tei(u_char *p)
+{
+	u_int	sapi, tei;
+
+	sapi = *p >> 2;
+	tei = p[1] >> 1;
+	return sapi | (tei << 8);
+}
+
+void
+recv_Dchannel(struct dchannel *dch)
+{
+	struct mISDNhead *hh;
+
+	if (dch->rx_skb->len < 2) { /* at least 2 for sapi / tei */
+		dev_kfree_skb(dch->rx_skb);
+		dch->rx_skb = NULL;
+		return;
+	}
+	hh = mISDN_HEAD_P(dch->rx_skb);
+	hh->prim = PH_DATA_IND;
+	hh->id = get_sapi_tei(dch->rx_skb->data);
+	skb_queue_tail(&dch->rqueue, dch->rx_skb);
+	dch->rx_skb = NULL;
+	schedule_event(dch, FLG_RECVQUEUE);
+}
+EXPORT_SYMBOL(recv_Dchannel);
+
+void
+recv_Bchannel(struct bchannel *bch)
+{
+	struct mISDNhead *hh;
+
+	hh = mISDN_HEAD_P(bch->rx_skb);
+	hh->prim = PH_DATA_IND;
+	hh->id = MISDN_ID_ANY;
+	if (bch->rcount >= 64) {
+		dev_kfree_skb(bch->rx_skb);
+		bch->rx_skb = NULL;
+		return;
+	}
+	bch->rcount++;
+	skb_queue_tail(&bch->rqueue, bch->rx_skb);
+	bch->rx_skb = NULL;
+	schedule_event(bch, FLG_RECVQUEUE);
+}
+EXPORT_SYMBOL(recv_Bchannel);
+
+void
+recv_Dchannel_skb(struct dchannel *dch, struct sk_buff *skb)
+{
+	skb_queue_tail(&dch->rqueue, skb);
+	schedule_event(dch, FLG_RECVQUEUE);
+}
+EXPORT_SYMBOL(recv_Dchannel_skb);
+
+void
+recv_Bchannel_skb(struct bchannel *bch, struct sk_buff *skb)
+{
+	if (bch->rcount >= 64) {
+		dev_kfree_skb(skb);
+		return;
+	}
+	bch->rcount++;
+	skb_queue_tail(&bch->rqueue, skb);
+	schedule_event(bch, FLG_RECVQUEUE);
+}
+EXPORT_SYMBOL(recv_Bchannel_skb);
+
+static void
+confirm_Dsend(struct dchannel *dch)
+{
+	struct sk_buff	*skb;
+
+	skb = _alloc_mISDN_skb(PH_DATA_CNF, mISDN_HEAD_ID(dch->tx_skb),
+	    0, NULL, GFP_ATOMIC);
+	if (!skb) {
+		printk(KERN_ERR "%s: no skb id %x\n", __func__,
+		    mISDN_HEAD_ID(dch->tx_skb));
+		return;
+	}
+	skb_queue_tail(&dch->rqueue, skb);
+	schedule_event(dch, FLG_RECVQUEUE);
+}
+
+int
+get_next_dframe(struct dchannel *dch)
+{
+	dch->tx_idx = 0;
+	dch->tx_skb = skb_dequeue(&dch->squeue);
+	if (dch->tx_skb) {
+		confirm_Dsend(dch);
+		return 1;
+	}
+	dch->tx_skb = NULL;
+	test_and_clear_bit(FLG_TX_BUSY, &dch->Flags);
+	return 0;
+}
+EXPORT_SYMBOL(get_next_dframe);
+
+void
+confirm_Bsend(struct bchannel *bch)
+{
+	struct sk_buff	*skb;
+
+	if (bch->rcount >= 64)
+		return;
+	skb = _alloc_mISDN_skb(PH_DATA_CNF, mISDN_HEAD_ID(bch->tx_skb),
+	    0, NULL, GFP_ATOMIC);
+	if (!skb) {
+		printk(KERN_ERR "%s: no skb id %x\n", __func__,
+		    mISDN_HEAD_ID(bch->tx_skb));
+		return;
+	}
+	bch->rcount++;
+	skb_queue_tail(&bch->rqueue, skb);
+	schedule_event(bch, FLG_RECVQUEUE);
+}
+EXPORT_SYMBOL(confirm_Bsend);
+
+int
+get_next_bframe(struct bchannel *bch)
+{
+	bch->tx_idx = 0;
+	if (test_bit(FLG_TX_NEXT, &bch->Flags)) {
+		bch->tx_skb = bch->next_skb;
+		if (bch->tx_skb) {
+			bch->next_skb = NULL;
+			test_and_clear_bit(FLG_TX_NEXT, &bch->Flags);
+			if (!test_bit(FLG_TRANSPARENT, &bch->Flags))
+				confirm_Bsend(bch); /* not for transparent */
+			return 1;
+		} else {
+			test_and_clear_bit(FLG_TX_NEXT, &bch->Flags);
+			printk(KERN_WARNING "B TX_NEXT without skb\n");
+		}
+	}
+	bch->tx_skb = NULL;
+	test_and_clear_bit(FLG_TX_BUSY, &bch->Flags);
+	return 0;
+}
+EXPORT_SYMBOL(get_next_bframe);
+
+void
+queue_ch_frame(struct mISDNchannel *ch, u_int pr, int id, struct sk_buff *skb)
+{
+	struct mISDNhead *hh;
+
+	if (!skb) {
+		_queue_data(ch, pr, id, 0, NULL, GFP_ATOMIC);
+	} else {
+		if (ch->peer) {
+			hh = mISDN_HEAD_P(skb);
+			hh->prim = pr;
+			hh->id = id;
+			if (!ch->recv(ch->peer, skb))
+				return;
+		}
+		dev_kfree_skb(skb);
+	}
+}
+EXPORT_SYMBOL(queue_ch_frame);
+
+int
+dchannel_senddata(struct dchannel *ch, struct sk_buff *skb)
+{
+	/* check oversize */
+	if (skb->len <= 0) {
+		printk(KERN_WARNING "%s: skb too small\n", __func__);
+		return -EINVAL;
+	}
+	if (skb->len > ch->maxlen) {
+		printk(KERN_WARNING "%s: skb too large(%d/%d)\n",
+			__func__, skb->len, ch->maxlen);
+		return -EINVAL;
+	}
+	/* HW lock must be obtained */
+	if (test_and_set_bit(FLG_TX_BUSY, &ch->Flags)) {
+		skb_queue_tail(&ch->squeue, skb);
+		return 0;
+	} else {
+		/* write to fifo */
+		ch->tx_skb = skb;
+		ch->tx_idx = 0;
+		return 1;
+	}
+}
+EXPORT_SYMBOL(dchannel_senddata);
+
+int
+bchannel_senddata(struct bchannel *ch, struct sk_buff *skb)
+{
+
+	/* check oversize */
+	if (skb->len <= 0) {
+		printk(KERN_WARNING "%s: skb too small\n", __func__);
+		return -EINVAL;
+	}
+	if (skb->len > ch->maxlen) {
+		printk(KERN_WARNING "%s: skb too large(%d/%d)\n",
+			__func__, skb->len, ch->maxlen);
+		return -EINVAL;
+	}
+	/* HW lock must be obtained */
+	/* check for pending next_skb */
+	if (ch->next_skb) {
+		printk(KERN_WARNING
+		    "%s: next_skb exist ERROR (skb->len=%d next_skb->len=%d)\n",
+		    __func__, skb->len, ch->next_skb->len);
+		return -EBUSY;
+	}
+	if (test_and_set_bit(FLG_TX_BUSY, &ch->Flags)) {
+		test_and_set_bit(FLG_TX_NEXT, &ch->Flags);
+		ch->next_skb = skb;
+		return 0;
+	} else {
+		/* write to fifo */
+		ch->tx_skb = skb;
+		ch->tx_idx = 0;
+		return 1;
+	}
+}
+EXPORT_SYMBOL(bchannel_senddata);
diff --git a/drivers/isdn/mISDN/layer1.c b/drivers/isdn/mISDN/layer1.c
new file mode 100644
index 000000000000..fced1a2755f8
--- /dev/null
+++ b/drivers/isdn/mISDN/layer1.c
@@ -0,0 +1,403 @@
+/*
+ *
+ * Author	Karsten Keil <kkeil@novell.com>
+ *
+ * Copyright 2008  by Karsten Keil <kkeil@novell.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+
+#include <linux/module.h>
+#include <linux/mISDNhw.h>
+#include "layer1.h"
+#include "fsm.h"
+
+static int *debug;
+
+struct layer1 {
+	u_long			Flags;
+	struct FsmInst		l1m;
+	struct FsmTimer 	timer;
+	int			delay;
+	struct dchannel		*dch;
+	dchannel_l1callback	*dcb;
+};
+
+#define TIMER3_VALUE 7000
+
+static
+struct Fsm l1fsm_s = {NULL, 0, 0, NULL, NULL};
+
+enum {
+	ST_L1_F2,
+	ST_L1_F3,
+	ST_L1_F4,
+	ST_L1_F5,
+	ST_L1_F6,
+	ST_L1_F7,
+	ST_L1_F8,
+};
+
+#define L1S_STATE_COUNT (ST_L1_F8+1)
+
+static char *strL1SState[] =
+{
+	"ST_L1_F2",
+	"ST_L1_F3",
+	"ST_L1_F4",
+	"ST_L1_F5",
+	"ST_L1_F6",
+	"ST_L1_F7",
+	"ST_L1_F8",
+};
+
+enum {
+	EV_PH_ACTIVATE,
+	EV_PH_DEACTIVATE,
+	EV_RESET_IND,
+	EV_DEACT_CNF,
+	EV_DEACT_IND,
+	EV_POWER_UP,
+	EV_ANYSIG_IND,
+	EV_INFO2_IND,
+	EV_INFO4_IND,
+	EV_TIMER_DEACT,
+	EV_TIMER_ACT,
+	EV_TIMER3,
+};
+
+#define L1_EVENT_COUNT (EV_TIMER3 + 1)
+
+static char *strL1Event[] =
+{
+	"EV_PH_ACTIVATE",
+	"EV_PH_DEACTIVATE",
+	"EV_RESET_IND",
+	"EV_DEACT_CNF",
+	"EV_DEACT_IND",
+	"EV_POWER_UP",
+	"EV_ANYSIG_IND",
+	"EV_INFO2_IND",
+	"EV_INFO4_IND",
+	"EV_TIMER_DEACT",
+	"EV_TIMER_ACT",
+	"EV_TIMER3",
+};
+
+static void
+l1m_debug(struct FsmInst *fi, char *fmt, ...)
+{
+	struct layer1 *l1 = fi->userdata;
+	va_list va;
+
+	va_start(va, fmt);
+	printk(KERN_DEBUG "%s: ", l1->dch->dev.name);
+	vprintk(fmt, va);
+	printk("\n");
+	va_end(va);
+}
+
+static void
+l1_reset(struct FsmInst *fi, int event, void *arg)
+{
+	mISDN_FsmChangeState(fi, ST_L1_F3);
+}
+
+static void
+l1_deact_cnf(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer1 *l1 = fi->userdata;
+
+	mISDN_FsmChangeState(fi, ST_L1_F3);
+	if (test_bit(FLG_L1_ACTIVATING, &l1->Flags))
+		l1->dcb(l1->dch, HW_POWERUP_REQ);
+}
+
+static void
+l1_deact_req_s(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer1 *l1 = fi->userdata;
+
+	mISDN_FsmChangeState(fi, ST_L1_F3);
+	mISDN_FsmRestartTimer(&l1->timer, 550, EV_TIMER_DEACT, NULL, 2);
+	test_and_set_bit(FLG_L1_DEACTTIMER, &l1->Flags);
+}
+
+static void
+l1_power_up_s(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer1 *l1 = fi->userdata;
+
+	if (test_bit(FLG_L1_ACTIVATING, &l1->Flags)) {
+		mISDN_FsmChangeState(fi, ST_L1_F4);
+		l1->dcb(l1->dch, INFO3_P8);
+	} else
+		mISDN_FsmChangeState(fi, ST_L1_F3);
+}
+
+static void
+l1_go_F5(struct FsmInst *fi, int event, void *arg)
+{
+	mISDN_FsmChangeState(fi, ST_L1_F5);
+}
+
+static void
+l1_go_F8(struct FsmInst *fi, int event, void *arg)
+{
+	mISDN_FsmChangeState(fi, ST_L1_F8);
+}
+
+static void
+l1_info2_ind(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer1 *l1 = fi->userdata;
+
+	mISDN_FsmChangeState(fi, ST_L1_F6);
+	l1->dcb(l1->dch, INFO3_P8);
+}
+
+static void
+l1_info4_ind(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer1 *l1 = fi->userdata;
+
+	mISDN_FsmChangeState(fi, ST_L1_F7);
+	l1->dcb(l1->dch, INFO3_P8);
+	if (test_and_clear_bit(FLG_L1_DEACTTIMER, &l1->Flags))
+		mISDN_FsmDelTimer(&l1->timer, 4);
+	if (!test_bit(FLG_L1_ACTIVATED, &l1->Flags)) {
+		if (test_and_clear_bit(FLG_L1_T3RUN, &l1->Flags))
+			mISDN_FsmDelTimer(&l1->timer, 3);
+		mISDN_FsmRestartTimer(&l1->timer, 110, EV_TIMER_ACT, NULL, 2);
+		test_and_set_bit(FLG_L1_ACTTIMER, &l1->Flags);
+	}
+}
+
+static void
+l1_timer3(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer1 *l1 = fi->userdata;
+
+	test_and_clear_bit(FLG_L1_T3RUN, &l1->Flags);
+	if (test_and_clear_bit(FLG_L1_ACTIVATING, &l1->Flags)) {
+		if (test_and_clear_bit(FLG_L1_DBLOCKED, &l1->Flags))
+			l1->dcb(l1->dch, HW_D_NOBLOCKED);
+		l1->dcb(l1->dch, PH_DEACTIVATE_IND);
+	}
+	if (l1->l1m.state != ST_L1_F6) {
+		mISDN_FsmChangeState(fi, ST_L1_F3);
+		l1->dcb(l1->dch, HW_POWERUP_REQ);
+	}
+}
+
+static void
+l1_timer_act(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer1 *l1 = fi->userdata;
+
+	test_and_clear_bit(FLG_L1_ACTTIMER, &l1->Flags);
+	test_and_set_bit(FLG_L1_ACTIVATED, &l1->Flags);
+	l1->dcb(l1->dch, PH_ACTIVATE_IND);
+}
+
+static void
+l1_timer_deact(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer1 *l1 = fi->userdata;
+
+	test_and_clear_bit(FLG_L1_DEACTTIMER, &l1->Flags);
+	test_and_clear_bit(FLG_L1_ACTIVATED, &l1->Flags);
+	if (test_and_clear_bit(FLG_L1_DBLOCKED, &l1->Flags))
+		l1->dcb(l1->dch, HW_D_NOBLOCKED);
+	l1->dcb(l1->dch, PH_DEACTIVATE_IND);
+	l1->dcb(l1->dch, HW_DEACT_REQ);
+}
+
+static void
+l1_activate_s(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer1 *l1 = fi->userdata;
+
+	mISDN_FsmRestartTimer(&l1->timer, TIMER3_VALUE, EV_TIMER3, NULL, 2);
+	test_and_set_bit(FLG_L1_T3RUN, &l1->Flags);
+	l1->dcb(l1->dch, HW_RESET_REQ);
+}
+
+static void
+l1_activate_no(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer1 *l1 = fi->userdata;
+
+	if ((!test_bit(FLG_L1_DEACTTIMER, &l1->Flags)) &&
+	    (!test_bit(FLG_L1_T3RUN, &l1->Flags))) {
+		test_and_clear_bit(FLG_L1_ACTIVATING, &l1->Flags);
+		if (test_and_clear_bit(FLG_L1_DBLOCKED, &l1->Flags))
+			l1->dcb(l1->dch, HW_D_NOBLOCKED);
+		l1->dcb(l1->dch, PH_DEACTIVATE_IND);
+	}
+}
+
+static struct FsmNode L1SFnList[] =
+{
+	{ST_L1_F3, EV_PH_ACTIVATE, l1_activate_s},
+	{ST_L1_F6, EV_PH_ACTIVATE, l1_activate_no},
+	{ST_L1_F8, EV_PH_ACTIVATE, l1_activate_no},
+	{ST_L1_F3, EV_RESET_IND, l1_reset},
+	{ST_L1_F4, EV_RESET_IND, l1_reset},
+	{ST_L1_F5, EV_RESET_IND, l1_reset},
+	{ST_L1_F6, EV_RESET_IND, l1_reset},
+	{ST_L1_F7, EV_RESET_IND, l1_reset},
+	{ST_L1_F8, EV_RESET_IND, l1_reset},
+	{ST_L1_F3, EV_DEACT_CNF, l1_deact_cnf},
+	{ST_L1_F4, EV_DEACT_CNF, l1_deact_cnf},
+	{ST_L1_F5, EV_DEACT_CNF, l1_deact_cnf},
+	{ST_L1_F6, EV_DEACT_CNF, l1_deact_cnf},
+	{ST_L1_F7, EV_DEACT_CNF, l1_deact_cnf},
+	{ST_L1_F8, EV_DEACT_CNF, l1_deact_cnf},
+	{ST_L1_F6, EV_DEACT_IND, l1_deact_req_s},
+	{ST_L1_F7, EV_DEACT_IND, l1_deact_req_s},
+	{ST_L1_F8, EV_DEACT_IND, l1_deact_req_s},
+	{ST_L1_F3, EV_POWER_UP,  l1_power_up_s},
+	{ST_L1_F4, EV_ANYSIG_IND, l1_go_F5},
+	{ST_L1_F6, EV_ANYSIG_IND, l1_go_F8},
+	{ST_L1_F7, EV_ANYSIG_IND, l1_go_F8},
+	{ST_L1_F3, EV_INFO2_IND, l1_info2_ind},
+	{ST_L1_F4, EV_INFO2_IND, l1_info2_ind},
+	{ST_L1_F5, EV_INFO2_IND, l1_info2_ind},
+	{ST_L1_F7, EV_INFO2_IND, l1_info2_ind},
+	{ST_L1_F8, EV_INFO2_IND, l1_info2_ind},
+	{ST_L1_F3, EV_INFO4_IND, l1_info4_ind},
+	{ST_L1_F4, EV_INFO4_IND, l1_info4_ind},
+	{ST_L1_F5, EV_INFO4_IND, l1_info4_ind},
+	{ST_L1_F6, EV_INFO4_IND, l1_info4_ind},
+	{ST_L1_F8, EV_INFO4_IND, l1_info4_ind},
+	{ST_L1_F3, EV_TIMER3, l1_timer3},
+	{ST_L1_F4, EV_TIMER3, l1_timer3},
+	{ST_L1_F5, EV_TIMER3, l1_timer3},
+	{ST_L1_F6, EV_TIMER3, l1_timer3},
+	{ST_L1_F8, EV_TIMER3, l1_timer3},
+	{ST_L1_F7, EV_TIMER_ACT, l1_timer_act},
+	{ST_L1_F3, EV_TIMER_DEACT, l1_timer_deact},
+	{ST_L1_F4, EV_TIMER_DEACT, l1_timer_deact},
+	{ST_L1_F5, EV_TIMER_DEACT, l1_timer_deact},
+	{ST_L1_F6, EV_TIMER_DEACT, l1_timer_deact},
+	{ST_L1_F7, EV_TIMER_DEACT, l1_timer_deact},
+	{ST_L1_F8, EV_TIMER_DEACT, l1_timer_deact},
+};
+
+static void
+release_l1(struct layer1 *l1) {
+	mISDN_FsmDelTimer(&l1->timer, 0);
+	if (l1->dch)
+		l1->dch->l1 = NULL;
+	module_put(THIS_MODULE);
+	kfree(l1);
+}
+
+int
+l1_event(struct layer1 *l1, u_int event)
+{
+	int		err = 0;
+
+	if (!l1)
+		return -EINVAL;
+	switch (event) {
+	case HW_RESET_IND:
+		mISDN_FsmEvent(&l1->l1m, EV_RESET_IND, NULL);
+		break;
+	case HW_DEACT_IND:
+		mISDN_FsmEvent(&l1->l1m, EV_DEACT_IND, NULL);
+		break;
+	case HW_POWERUP_IND:
+		mISDN_FsmEvent(&l1->l1m, EV_POWER_UP, NULL);
+		break;
+	case HW_DEACT_CNF:
+		mISDN_FsmEvent(&l1->l1m, EV_DEACT_CNF, NULL);
+		break;
+	case ANYSIGNAL:
+		mISDN_FsmEvent(&l1->l1m, EV_ANYSIG_IND, NULL);
+		break;
+	case LOSTFRAMING:
+		mISDN_FsmEvent(&l1->l1m, EV_ANYSIG_IND, NULL);
+		break;
+	case INFO2:
+		mISDN_FsmEvent(&l1->l1m, EV_INFO2_IND, NULL);
+		break;
+	case INFO4_P8:
+		mISDN_FsmEvent(&l1->l1m, EV_INFO4_IND, NULL);
+		break;
+	case INFO4_P10:
+		mISDN_FsmEvent(&l1->l1m, EV_INFO4_IND, NULL);
+		break;
+	case PH_ACTIVATE_REQ:
+		if (test_bit(FLG_L1_ACTIVATED, &l1->Flags))
+			l1->dcb(l1->dch, PH_ACTIVATE_IND);
+		else {
+			test_and_set_bit(FLG_L1_ACTIVATING, &l1->Flags);
+			mISDN_FsmEvent(&l1->l1m, EV_PH_ACTIVATE, NULL);
+		}
+		break;
+	case CLOSE_CHANNEL:
+		release_l1(l1);
+		break;
+	default:
+		if (*debug & DEBUG_L1)
+			printk(KERN_DEBUG "%s %x unhandled\n",
+			    __func__, event);
+		err = -EINVAL;
+	}
+	return err;
+}
+EXPORT_SYMBOL(l1_event);
+
+int
+create_l1(struct dchannel *dch, dchannel_l1callback *dcb) {
+	struct layer1	*nl1;
+
+	nl1 = kzalloc(sizeof(struct layer1), GFP_ATOMIC);
+	if (!nl1) {
+		printk(KERN_ERR "kmalloc struct layer1 failed\n");
+		return -ENOMEM;
+	}
+	nl1->l1m.fsm = &l1fsm_s;
+	nl1->l1m.state = ST_L1_F3;
+	nl1->Flags = 0;
+	nl1->l1m.debug = *debug & DEBUG_L1_FSM;
+	nl1->l1m.userdata = nl1;
+	nl1->l1m.userint = 0;
+	nl1->l1m.printdebug = l1m_debug;
+	nl1->dch = dch;
+	nl1->dcb = dcb;
+	mISDN_FsmInitTimer(&nl1->l1m, &nl1->timer);
+	__module_get(THIS_MODULE);
+	dch->l1 = nl1;
+	return 0;
+}
+EXPORT_SYMBOL(create_l1);
+
+int
+l1_init(u_int *deb)
+{
+	debug = deb;
+	l1fsm_s.state_count = L1S_STATE_COUNT;
+	l1fsm_s.event_count = L1_EVENT_COUNT;
+	l1fsm_s.strEvent = strL1Event;
+	l1fsm_s.strState = strL1SState;
+	mISDN_FsmNew(&l1fsm_s, L1SFnList, ARRAY_SIZE(L1SFnList));
+	return 0;
+}
+
+void
+l1_cleanup(void)
+{
+	mISDN_FsmFree(&l1fsm_s);
+}
diff --git a/drivers/isdn/mISDN/layer1.h b/drivers/isdn/mISDN/layer1.h
new file mode 100644
index 000000000000..9c8125fd89af
--- /dev/null
+++ b/drivers/isdn/mISDN/layer1.h
@@ -0,0 +1,26 @@
+/*
+ *
+ * Layer 1 defines
+ *
+ * Copyright 2008  by Karsten Keil <kkeil@novell.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#define FLG_L1_ACTIVATING	1
+#define FLG_L1_ACTIVATED	2
+#define FLG_L1_DEACTTIMER	3
+#define FLG_L1_ACTTIMER		4
+#define FLG_L1_T3RUN		5
+#define FLG_L1_PULL_REQ		6
+#define FLG_L1_UINT		7
+#define FLG_L1_DBLOCKED		8
+
diff --git a/drivers/isdn/mISDN/layer2.c b/drivers/isdn/mISDN/layer2.c
new file mode 100644
index 000000000000..f5ad888ee71e
--- /dev/null
+++ b/drivers/isdn/mISDN/layer2.c
@@ -0,0 +1,2216 @@
+/*
+ *
+ * Author	Karsten Keil <kkeil@novell.com>
+ *
+ * Copyright 2008  by Karsten Keil <kkeil@novell.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include "fsm.h"
+#include "layer2.h"
+
+static int *debug;
+
+static
+struct Fsm l2fsm = {NULL, 0, 0, NULL, NULL};
+
+static char *strL2State[] =
+{
+	"ST_L2_1",
+	"ST_L2_2",
+	"ST_L2_3",
+	"ST_L2_4",
+	"ST_L2_5",
+	"ST_L2_6",
+	"ST_L2_7",
+	"ST_L2_8",
+};
+
+enum {
+	EV_L2_UI,
+	EV_L2_SABME,
+	EV_L2_DISC,
+	EV_L2_DM,
+	EV_L2_UA,
+	EV_L2_FRMR,
+	EV_L2_SUPER,
+	EV_L2_I,
+	EV_L2_DL_DATA,
+	EV_L2_ACK_PULL,
+	EV_L2_DL_UNITDATA,
+	EV_L2_DL_ESTABLISH_REQ,
+	EV_L2_DL_RELEASE_REQ,
+	EV_L2_MDL_ASSIGN,
+	EV_L2_MDL_REMOVE,
+	EV_L2_MDL_ERROR,
+	EV_L1_DEACTIVATE,
+	EV_L2_T200,
+	EV_L2_T203,
+	EV_L2_SET_OWN_BUSY,
+	EV_L2_CLEAR_OWN_BUSY,
+	EV_L2_FRAME_ERROR,
+};
+
+#define L2_EVENT_COUNT (EV_L2_FRAME_ERROR+1)
+
+static char *strL2Event[] =
+{
+	"EV_L2_UI",
+	"EV_L2_SABME",
+	"EV_L2_DISC",
+	"EV_L2_DM",
+	"EV_L2_UA",
+	"EV_L2_FRMR",
+	"EV_L2_SUPER",
+	"EV_L2_I",
+	"EV_L2_DL_DATA",
+	"EV_L2_ACK_PULL",
+	"EV_L2_DL_UNITDATA",
+	"EV_L2_DL_ESTABLISH_REQ",
+	"EV_L2_DL_RELEASE_REQ",
+	"EV_L2_MDL_ASSIGN",
+	"EV_L2_MDL_REMOVE",
+	"EV_L2_MDL_ERROR",
+	"EV_L1_DEACTIVATE",
+	"EV_L2_T200",
+	"EV_L2_T203",
+	"EV_L2_SET_OWN_BUSY",
+	"EV_L2_CLEAR_OWN_BUSY",
+	"EV_L2_FRAME_ERROR",
+};
+
+static void
+l2m_debug(struct FsmInst *fi, char *fmt, ...)
+{
+	struct layer2 *l2 = fi->userdata;
+	va_list va;
+
+	if (!(*debug & DEBUG_L2_FSM))
+		return;
+	va_start(va, fmt);
+	printk(KERN_DEBUG "l2 (tei %d): ", l2->tei);
+	vprintk(fmt, va);
+	printk("\n");
+	va_end(va);
+}
+
+inline u_int
+l2headersize(struct layer2 *l2, int ui)
+{
+	return ((test_bit(FLG_MOD128, &l2->flag) && (!ui)) ? 2 : 1) +
+		(test_bit(FLG_LAPD, &l2->flag) ? 2 : 1);
+}
+
+inline u_int
+l2addrsize(struct layer2 *l2)
+{
+	return test_bit(FLG_LAPD, &l2->flag) ? 2 : 1;
+}
+
+static u_int
+l2_newid(struct layer2 *l2)
+{
+	u_int	id;
+
+	id = l2->next_id++;
+	if (id == 0x7fff)
+		l2->next_id = 1;
+	id <<= 16;
+	id |= l2->tei << 8;
+	id |= l2->sapi;
+	return id;
+}
+
+static void
+l2up(struct layer2 *l2, u_int prim, struct sk_buff *skb)
+{
+	int	err;
+
+	if (!l2->up)
+		return;
+	mISDN_HEAD_PRIM(skb) = prim;
+	mISDN_HEAD_ID(skb) = (l2->ch.nr << 16) | l2->ch.addr;
+	err = l2->up->send(l2->up, skb);
+	if (err) {
+		printk(KERN_WARNING "%s: err=%d\n", __func__, err);
+		dev_kfree_skb(skb);
+	}
+}
+
+static void
+l2up_create(struct layer2 *l2, u_int prim, int len, void *arg)
+{
+	struct sk_buff	*skb;
+	struct mISDNhead *hh;
+	int		err;
+
+	if (!l2->up)
+		return;
+	skb = mI_alloc_skb(len, GFP_ATOMIC);
+	if (!skb)
+		return;
+	hh = mISDN_HEAD_P(skb);
+	hh->prim = prim;
+	hh->id = (l2->ch.nr << 16) | l2->ch.addr;
+	if (len)
+		memcpy(skb_put(skb, len), arg, len);
+	err = l2->up->send(l2->up, skb);
+	if (err) {
+		printk(KERN_WARNING "%s: err=%d\n", __func__, err);
+		dev_kfree_skb(skb);
+	}
+}
+
+static int
+l2down_skb(struct layer2 *l2, struct sk_buff *skb) {
+	int ret;
+
+	ret = l2->ch.recv(l2->ch.peer, skb);
+	if (ret && (*debug & DEBUG_L2_RECV))
+		printk(KERN_DEBUG "l2down_skb: ret(%d)\n", ret);
+	return ret;
+}
+
+static int
+l2down_raw(struct layer2 *l2, struct sk_buff *skb)
+{
+	struct mISDNhead *hh = mISDN_HEAD_P(skb);
+
+	if (hh->prim == PH_DATA_REQ) {
+		if (test_and_set_bit(FLG_L1_NOTREADY, &l2->flag)) {
+			skb_queue_tail(&l2->down_queue, skb);
+			return 0;
+		}
+		l2->down_id = mISDN_HEAD_ID(skb);
+	}
+	return l2down_skb(l2, skb);
+}
+
+static int
+l2down(struct layer2 *l2, u_int prim, u_int id, struct sk_buff *skb)
+{
+	struct mISDNhead *hh = mISDN_HEAD_P(skb);
+
+	hh->prim = prim;
+	hh->id = id;
+	return l2down_raw(l2, skb);
+}
+
+static int
+l2down_create(struct layer2 *l2, u_int prim, u_int id, int len, void *arg)
+{
+	struct sk_buff	*skb;
+	int		err;
+	struct mISDNhead *hh;
+
+	skb = mI_alloc_skb(len, GFP_ATOMIC);
+	if (!skb)
+		return -ENOMEM;
+	hh = mISDN_HEAD_P(skb);
+	hh->prim = prim;
+	hh->id = id;
+	if (len)
+		memcpy(skb_put(skb, len), arg, len);
+	err = l2down_raw(l2, skb);
+	if (err)
+		dev_kfree_skb(skb);
+	return err;
+}
+
+static int
+ph_data_confirm(struct layer2 *l2, struct mISDNhead *hh, struct sk_buff *skb) {
+	struct sk_buff *nskb = skb;
+	int ret = -EAGAIN;
+
+	if (test_bit(FLG_L1_NOTREADY, &l2->flag)) {
+		if (hh->id == l2->down_id) {
+			nskb = skb_dequeue(&l2->down_queue);
+			if (nskb) {
+				l2->down_id = mISDN_HEAD_ID(nskb);
+				if (l2down_skb(l2, nskb)) {
+					dev_kfree_skb(nskb);
+					l2->down_id = MISDN_ID_NONE;
+				}
+			} else
+				l2->down_id = MISDN_ID_NONE;
+			if (ret) {
+				dev_kfree_skb(skb);
+				ret = 0;
+			}
+			if (l2->down_id == MISDN_ID_NONE) {
+				test_and_clear_bit(FLG_L1_NOTREADY, &l2->flag);
+				mISDN_FsmEvent(&l2->l2m, EV_L2_ACK_PULL, NULL);
+			}
+		}
+	}
+	if (!test_and_set_bit(FLG_L1_NOTREADY, &l2->flag)) {
+		nskb = skb_dequeue(&l2->down_queue);
+		if (nskb) {
+			l2->down_id = mISDN_HEAD_ID(nskb);
+			if (l2down_skb(l2, nskb)) {
+				dev_kfree_skb(nskb);
+				l2->down_id = MISDN_ID_NONE;
+				test_and_clear_bit(FLG_L1_NOTREADY, &l2->flag);
+			}
+		} else
+			test_and_clear_bit(FLG_L1_NOTREADY, &l2->flag);
+	}
+	return ret;
+}
+
+static int
+l2mgr(struct layer2 *l2, u_int prim, void *arg) {
+	long c = (long)arg;
+
+	printk(KERN_WARNING
+	    "l2mgr: addr:%x prim %x %c\n", l2->id, prim, (char)c);
+	if (test_bit(FLG_LAPD, &l2->flag) &&
+		!test_bit(FLG_FIXED_TEI, &l2->flag)) {
+		switch (c) {
+		case 'C':
+		case 'D':
+		case 'G':
+		case 'H':
+			l2_tei(l2, prim, (u_long)arg);
+			break;
+		}
+	}
+	return 0;
+}
+
+static void
+set_peer_busy(struct layer2 *l2) {
+	test_and_set_bit(FLG_PEER_BUSY, &l2->flag);
+	if (skb_queue_len(&l2->i_queue) || skb_queue_len(&l2->ui_queue))
+		test_and_set_bit(FLG_L2BLOCK, &l2->flag);
+}
+
+static void
+clear_peer_busy(struct layer2 *l2) {
+	if (test_and_clear_bit(FLG_PEER_BUSY, &l2->flag))
+		test_and_clear_bit(FLG_L2BLOCK, &l2->flag);
+}
+
+static void
+InitWin(struct layer2 *l2)
+{
+	int i;
+
+	for (i = 0; i < MAX_WINDOW; i++)
+		l2->windowar[i] = NULL;
+}
+
+static int
+freewin(struct layer2 *l2)
+{
+	int i, cnt = 0;
+
+	for (i = 0; i < MAX_WINDOW; i++) {
+		if (l2->windowar[i]) {
+			cnt++;
+			dev_kfree_skb(l2->windowar[i]);
+			l2->windowar[i] = NULL;
+		}
+	}
+	return cnt;
+}
+
+static void
+ReleaseWin(struct layer2 *l2)
+{
+	int cnt = freewin(l2);
+
+	if (cnt)
+		printk(KERN_WARNING
+		    "isdnl2 freed %d skbuffs in release\n", cnt);
+}
+
+inline unsigned int
+cansend(struct layer2 *l2)
+{
+	unsigned int p1;
+
+	if (test_bit(FLG_MOD128, &l2->flag))
+		p1 = (l2->vs - l2->va) % 128;
+	else
+		p1 = (l2->vs - l2->va) % 8;
+	return (p1 < l2->window) && !test_bit(FLG_PEER_BUSY, &l2->flag);
+}
+
+inline void
+clear_exception(struct layer2 *l2)
+{
+	test_and_clear_bit(FLG_ACK_PEND, &l2->flag);
+	test_and_clear_bit(FLG_REJEXC, &l2->flag);
+	test_and_clear_bit(FLG_OWN_BUSY, &l2->flag);
+	clear_peer_busy(l2);
+}
+
+static int
+sethdraddr(struct layer2 *l2, u_char *header, int rsp)
+{
+	u_char *ptr = header;
+	int crbit = rsp;
+
+	if (test_bit(FLG_LAPD, &l2->flag)) {
+		if (test_bit(FLG_LAPD_NET, &l2->flag))
+			crbit = !crbit;
+		*ptr++ = (l2->sapi << 2) | (crbit ? 2 : 0);
+		*ptr++ = (l2->tei << 1) | 1;
+		return 2;
+	} else {
+		if (test_bit(FLG_ORIG, &l2->flag))
+			crbit = !crbit;
+		if (crbit)
+			*ptr++ = l2->addr.B;
+		else
+			*ptr++ = l2->addr.A;
+		return 1;
+	}
+}
+
+static inline void
+enqueue_super(struct layer2 *l2, struct sk_buff *skb)
+{
+	if (l2down(l2, PH_DATA_REQ, l2_newid(l2), skb))
+		dev_kfree_skb(skb);
+}
+
+static inline void
+enqueue_ui(struct layer2 *l2, struct sk_buff *skb)
+{
+	if (l2->tm)
+		l2_tei(l2, MDL_STATUS_UI_IND, 0);
+	if (l2down(l2, PH_DATA_REQ, l2_newid(l2), skb))
+		dev_kfree_skb(skb);
+}
+
+inline int
+IsUI(u_char *data)
+{
+	return (data[0] & 0xef) == UI;
+}
+
+inline int
+IsUA(u_char *data)
+{
+	return (data[0] & 0xef) == UA;
+}
+
+inline int
+IsDM(u_char *data)
+{
+	return (data[0] & 0xef) == DM;
+}
+
+inline int
+IsDISC(u_char *data)
+{
+	return (data[0] & 0xef) == DISC;
+}
+
+inline int
+IsRR(u_char *data, struct layer2 *l2)
+{
+	if (test_bit(FLG_MOD128, &l2->flag))
+		return data[0] == RR;
+	else
+		return (data[0] & 0xf) == 1;
+}
+
+inline int
+IsSFrame(u_char *data, struct layer2 *l2)
+{
+	register u_char d = *data;
+
+	if (!test_bit(FLG_MOD128, &l2->flag))
+		d &= 0xf;
+	return ((d & 0xf3) == 1) && ((d & 0x0c) != 0x0c);
+}
+
+inline int
+IsSABME(u_char *data, struct layer2 *l2)
+{
+	u_char d = data[0] & ~0x10;
+
+	return test_bit(FLG_MOD128, &l2->flag) ? d == SABME : d == SABM;
+}
+
+inline int
+IsREJ(u_char *data, struct layer2 *l2)
+{
+	return test_bit(FLG_MOD128, &l2->flag) ?
+		data[0] == REJ : (data[0] & 0xf) == REJ;
+}
+
+inline int
+IsFRMR(u_char *data)
+{
+	return (data[0] & 0xef) == FRMR;
+}
+
+inline int
+IsRNR(u_char *data, struct layer2 *l2)
+{
+	return test_bit(FLG_MOD128, &l2->flag) ?
+	    data[0] == RNR : (data[0] & 0xf) == RNR;
+}
+
+int
+iframe_error(struct layer2 *l2, struct sk_buff *skb)
+{
+	u_int	i;
+	int	rsp = *skb->data & 0x2;
+
+	i = l2addrsize(l2) + (test_bit(FLG_MOD128, &l2->flag) ? 2 : 1);
+	if (test_bit(FLG_ORIG, &l2->flag))
+		rsp = !rsp;
+	if (rsp)
+		return 'L';
+	if (skb->len < i)
+		return 'N';
+	if ((skb->len - i) > l2->maxlen)
+		return 'O';
+	return 0;
+}
+
+int
+super_error(struct layer2 *l2, struct sk_buff *skb)
+{
+	if (skb->len != l2addrsize(l2) +
+	    (test_bit(FLG_MOD128, &l2->flag) ? 2 : 1))
+		return 'N';
+	return 0;
+}
+
+int
+unnum_error(struct layer2 *l2, struct sk_buff *skb, int wantrsp)
+{
+	int rsp = (*skb->data & 0x2) >> 1;
+	if (test_bit(FLG_ORIG, &l2->flag))
+		rsp = !rsp;
+	if (rsp != wantrsp)
+		return 'L';
+	if (skb->len != l2addrsize(l2) + 1)
+		return 'N';
+	return 0;
+}
+
+int
+UI_error(struct layer2 *l2, struct sk_buff *skb)
+{
+	int rsp = *skb->data & 0x2;
+	if (test_bit(FLG_ORIG, &l2->flag))
+		rsp = !rsp;
+	if (rsp)
+		return 'L';
+	if (skb->len > l2->maxlen + l2addrsize(l2) + 1)
+		return 'O';
+	return 0;
+}
+
+int
+FRMR_error(struct layer2 *l2, struct sk_buff *skb)
+{
+	u_int	headers = l2addrsize(l2) + 1;
+	u_char	*datap = skb->data + headers;
+	int	rsp = *skb->data & 0x2;
+
+	if (test_bit(FLG_ORIG, &l2->flag))
+		rsp = !rsp;
+	if (!rsp)
+		return 'L';
+	if (test_bit(FLG_MOD128, &l2->flag)) {
+		if (skb->len < headers + 5)
+			return 'N';
+		else if (*debug & DEBUG_L2)
+			l2m_debug(&l2->l2m,
+			    "FRMR information %2x %2x %2x %2x %2x",
+			    datap[0], datap[1], datap[2], datap[3], datap[4]);
+	} else {
+		if (skb->len < headers + 3)
+			return 'N';
+		else if (*debug & DEBUG_L2)
+			l2m_debug(&l2->l2m,
+			    "FRMR information %2x %2x %2x",
+			    datap[0], datap[1], datap[2]);
+	}
+	return 0;
+}
+
+static unsigned int
+legalnr(struct layer2 *l2, unsigned int nr)
+{
+	if (test_bit(FLG_MOD128, &l2->flag))
+		return ((nr - l2->va) % 128) <= ((l2->vs - l2->va) % 128);
+	else
+		return ((nr - l2->va) % 8) <= ((l2->vs - l2->va) % 8);
+}
+
+static void
+setva(struct layer2 *l2, unsigned int nr)
+{
+	struct sk_buff	*skb;
+
+	while (l2->va != nr) {
+		l2->va++;
+		if (test_bit(FLG_MOD128, &l2->flag))
+			l2->va %= 128;
+		else
+			l2->va %= 8;
+		if (l2->windowar[l2->sow]) {
+			skb_trim(l2->windowar[l2->sow], 0);
+			skb_queue_tail(&l2->tmp_queue, l2->windowar[l2->sow]);
+			l2->windowar[l2->sow] = NULL;
+		}
+		l2->sow = (l2->sow + 1) % l2->window;
+	}
+	skb = skb_dequeue(&l2->tmp_queue);
+	while (skb) {
+		dev_kfree_skb(skb);
+		skb = skb_dequeue(&l2->tmp_queue);
+	}
+}
+
+static void
+send_uframe(struct layer2 *l2, struct sk_buff *skb, u_char cmd, u_char cr)
+{
+	u_char tmp[MAX_L2HEADER_LEN];
+	int i;
+
+	i = sethdraddr(l2, tmp, cr);
+	tmp[i++] = cmd;
+	if (skb)
+		skb_trim(skb, 0);
+	else {
+		skb = mI_alloc_skb(i, GFP_ATOMIC);
+		if (!skb) {
+			printk(KERN_WARNING "%s: can't alloc skbuff\n",
+				__func__);
+			return;
+		}
+	}
+	memcpy(skb_put(skb, i), tmp, i);
+	enqueue_super(l2, skb);
+}
+
+
+inline u_char
+get_PollFlag(struct layer2 *l2, struct sk_buff *skb)
+{
+	return skb->data[l2addrsize(l2)] & 0x10;
+}
+
+inline u_char
+get_PollFlagFree(struct layer2 *l2, struct sk_buff *skb)
+{
+	u_char PF;
+
+	PF = get_PollFlag(l2, skb);
+	dev_kfree_skb(skb);
+	return PF;
+}
+
+inline void
+start_t200(struct layer2 *l2, int i)
+{
+	mISDN_FsmAddTimer(&l2->t200, l2->T200, EV_L2_T200, NULL, i);
+	test_and_set_bit(FLG_T200_RUN, &l2->flag);
+}
+
+inline void
+restart_t200(struct layer2 *l2, int i)
+{
+	mISDN_FsmRestartTimer(&l2->t200, l2->T200, EV_L2_T200, NULL, i);
+	test_and_set_bit(FLG_T200_RUN, &l2->flag);
+}
+
+inline void
+stop_t200(struct layer2 *l2, int i)
+{
+	if (test_and_clear_bit(FLG_T200_RUN, &l2->flag))
+		mISDN_FsmDelTimer(&l2->t200, i);
+}
+
+inline void
+st5_dl_release_l2l3(struct layer2 *l2)
+{
+	int pr;
+
+	if (test_and_clear_bit(FLG_PEND_REL, &l2->flag))
+		pr = DL_RELEASE_CNF;
+	else
+		pr = DL_RELEASE_IND;
+	l2up_create(l2, pr, 0, NULL);
+}
+
+inline void
+lapb_dl_release_l2l3(struct layer2 *l2, int f)
+{
+	if (test_bit(FLG_LAPB, &l2->flag))
+		l2down_create(l2, PH_DEACTIVATE_REQ, l2_newid(l2), 0, NULL);
+	l2up_create(l2, f, 0, NULL);
+}
+
+static void
+establishlink(struct FsmInst *fi)
+{
+	struct layer2 *l2 = fi->userdata;
+	u_char cmd;
+
+	clear_exception(l2);
+	l2->rc = 0;
+	cmd = (test_bit(FLG_MOD128, &l2->flag) ? SABME : SABM) | 0x10;
+	send_uframe(l2, NULL, cmd, CMD);
+	mISDN_FsmDelTimer(&l2->t203, 1);
+	restart_t200(l2, 1);
+	test_and_clear_bit(FLG_PEND_REL, &l2->flag);
+	freewin(l2);
+	mISDN_FsmChangeState(fi, ST_L2_5);
+}
+
+static void
+l2_mdl_error_ua(struct FsmInst *fi, int event, void *arg)
+{
+	struct sk_buff *skb = arg;
+	struct layer2 *l2 = fi->userdata;
+
+	if (get_PollFlagFree(l2, skb))
+		l2mgr(l2, MDL_ERROR_IND, (void *) 'C');
+	else
+		l2mgr(l2, MDL_ERROR_IND, (void *) 'D');
+
+}
+
+static void
+l2_mdl_error_dm(struct FsmInst *fi, int event, void *arg)
+{
+	struct sk_buff *skb = arg;
+	struct layer2 *l2 = fi->userdata;
+
+	if (get_PollFlagFree(l2, skb))
+		l2mgr(l2, MDL_ERROR_IND, (void *) 'B');
+	else {
+		l2mgr(l2, MDL_ERROR_IND, (void *) 'E');
+		establishlink(fi);
+		test_and_clear_bit(FLG_L3_INIT, &l2->flag);
+	}
+}
+
+static void
+l2_st8_mdl_error_dm(struct FsmInst *fi, int event, void *arg)
+{
+	struct sk_buff *skb = arg;
+	struct layer2 *l2 = fi->userdata;
+
+	if (get_PollFlagFree(l2, skb))
+		l2mgr(l2, MDL_ERROR_IND, (void *) 'B');
+	else
+		l2mgr(l2, MDL_ERROR_IND, (void *) 'E');
+	establishlink(fi);
+	test_and_clear_bit(FLG_L3_INIT, &l2->flag);
+}
+
+static void
+l2_go_st3(struct FsmInst *fi, int event, void *arg)
+{
+	dev_kfree_skb((struct sk_buff *)arg);
+	mISDN_FsmChangeState(fi, ST_L2_3);
+}
+
+static void
+l2_mdl_assign(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2	*l2 = fi->userdata;
+
+	mISDN_FsmChangeState(fi, ST_L2_3);
+	dev_kfree_skb((struct sk_buff *)arg);
+	l2_tei(l2, MDL_ASSIGN_IND, 0);
+}
+
+static void
+l2_queue_ui_assign(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+
+	skb_queue_tail(&l2->ui_queue, skb);
+	mISDN_FsmChangeState(fi, ST_L2_2);
+	l2_tei(l2, MDL_ASSIGN_IND, 0);
+}
+
+static void
+l2_queue_ui(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+
+	skb_queue_tail(&l2->ui_queue, skb);
+}
+
+static void
+tx_ui(struct layer2 *l2)
+{
+	struct sk_buff *skb;
+	u_char header[MAX_L2HEADER_LEN];
+	int i;
+
+	i = sethdraddr(l2, header, CMD);
+	if (test_bit(FLG_LAPD_NET, &l2->flag))
+		header[1] = 0xff; /* tei 127 */
+	header[i++] = UI;
+	while ((skb = skb_dequeue(&l2->ui_queue))) {
+		memcpy(skb_push(skb, i), header, i);
+		enqueue_ui(l2, skb);
+	}
+}
+
+static void
+l2_send_ui(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+
+	skb_queue_tail(&l2->ui_queue, skb);
+	tx_ui(l2);
+}
+
+static void
+l2_got_ui(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+
+	skb_pull(skb, l2headersize(l2, 1));
+/*
+ *		in states 1-3 for broadcast
+ */
+
+	if (l2->tm)
+		l2_tei(l2, MDL_STATUS_UI_IND, 0);
+	l2up(l2, DL_UNITDATA_IND, skb);
+}
+
+static void
+l2_establish(struct FsmInst *fi, int event, void *arg)
+{
+	struct sk_buff *skb = arg;
+	struct layer2 *l2 = fi->userdata;
+
+	establishlink(fi);
+	test_and_set_bit(FLG_L3_INIT, &l2->flag);
+	dev_kfree_skb(skb);
+}
+
+static void
+l2_discard_i_setl3(struct FsmInst *fi, int event, void *arg)
+{
+	struct sk_buff *skb = arg;
+	struct layer2 *l2 = fi->userdata;
+
+	skb_queue_purge(&l2->i_queue);
+	test_and_set_bit(FLG_L3_INIT, &l2->flag);
+	test_and_clear_bit(FLG_PEND_REL, &l2->flag);
+	dev_kfree_skb(skb);
+}
+
+static void
+l2_l3_reestablish(struct FsmInst *fi, int event, void *arg)
+{
+	struct sk_buff *skb = arg;
+	struct layer2 *l2 = fi->userdata;
+
+	skb_queue_purge(&l2->i_queue);
+	establishlink(fi);
+	test_and_set_bit(FLG_L3_INIT, &l2->flag);
+	dev_kfree_skb(skb);
+}
+
+static void
+l2_release(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+
+	skb_trim(skb, 0);
+	l2up(l2, DL_RELEASE_CNF, skb);
+}
+
+static void
+l2_pend_rel(struct FsmInst *fi, int event, void *arg)
+{
+	struct sk_buff *skb = arg;
+	struct layer2 *l2 = fi->userdata;
+
+	test_and_set_bit(FLG_PEND_REL, &l2->flag);
+	dev_kfree_skb(skb);
+}
+
+static void
+l2_disconnect(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+
+	skb_queue_purge(&l2->i_queue);
+	freewin(l2);
+	mISDN_FsmChangeState(fi, ST_L2_6);
+	l2->rc = 0;
+	send_uframe(l2, NULL, DISC | 0x10, CMD);
+	mISDN_FsmDelTimer(&l2->t203, 1);
+	restart_t200(l2, 2);
+	if (skb)
+		dev_kfree_skb(skb);
+}
+
+static void
+l2_start_multi(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2	*l2 = fi->userdata;
+	struct sk_buff	*skb = arg;
+
+	l2->vs = 0;
+	l2->va = 0;
+	l2->vr = 0;
+	l2->sow = 0;
+	clear_exception(l2);
+	send_uframe(l2, NULL, UA | get_PollFlag(l2, skb), RSP);
+	mISDN_FsmChangeState(fi, ST_L2_7);
+	mISDN_FsmAddTimer(&l2->t203, l2->T203, EV_L2_T203, NULL, 3);
+	skb_trim(skb, 0);
+	l2up(l2, DL_ESTABLISH_IND, skb);
+	if (l2->tm)
+		l2_tei(l2, MDL_STATUS_UP_IND, 0);
+}
+
+static void
+l2_send_UA(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+
+	send_uframe(l2, skb, UA | get_PollFlag(l2, skb), RSP);
+}
+
+static void
+l2_send_DM(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+
+	send_uframe(l2, skb, DM | get_PollFlag(l2, skb), RSP);
+}
+
+static void
+l2_restart_multi(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2	*l2 = fi->userdata;
+	struct sk_buff	*skb = arg;
+	int		est = 0;
+
+	send_uframe(l2, skb, UA | get_PollFlag(l2, skb), RSP);
+
+	l2mgr(l2, MDL_ERROR_IND, (void *) 'F');
+
+	if (l2->vs != l2->va) {
+		skb_queue_purge(&l2->i_queue);
+		est = 1;
+	}
+
+	clear_exception(l2);
+	l2->vs = 0;
+	l2->va = 0;
+	l2->vr = 0;
+	l2->sow = 0;
+	mISDN_FsmChangeState(fi, ST_L2_7);
+	stop_t200(l2, 3);
+	mISDN_FsmRestartTimer(&l2->t203, l2->T203, EV_L2_T203, NULL, 3);
+
+	if (est)
+		l2up_create(l2, DL_ESTABLISH_IND, 0, NULL);
+/*		mISDN_queue_data(&l2->inst, l2->inst.id | MSG_BROADCAST,
+ *		    MGR_SHORTSTATUS | INDICATION, SSTATUS_L2_ESTABLISHED,
+ *		    0, NULL, 0);
+ */
+	if (skb_queue_len(&l2->i_queue) && cansend(l2))
+		mISDN_FsmEvent(fi, EV_L2_ACK_PULL, NULL);
+}
+
+static void
+l2_stop_multi(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2	*l2 = fi->userdata;
+	struct sk_buff	*skb = arg;
+
+	mISDN_FsmChangeState(fi, ST_L2_4);
+	mISDN_FsmDelTimer(&l2->t203, 3);
+	stop_t200(l2, 4);
+
+	send_uframe(l2, skb, UA | get_PollFlag(l2, skb), RSP);
+	skb_queue_purge(&l2->i_queue);
+	freewin(l2);
+	lapb_dl_release_l2l3(l2, DL_RELEASE_IND);
+	if (l2->tm)
+		l2_tei(l2, MDL_STATUS_DOWN_IND, 0);
+}
+
+static void
+l2_connected(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2	*l2 = fi->userdata;
+	struct sk_buff	*skb = arg;
+	int pr = -1;
+
+	if (!get_PollFlag(l2, skb)) {
+		l2_mdl_error_ua(fi, event, arg);
+		return;
+	}
+	dev_kfree_skb(skb);
+	if (test_and_clear_bit(FLG_PEND_REL, &l2->flag))
+		l2_disconnect(fi, event, NULL);
+	if (test_and_clear_bit(FLG_L3_INIT, &l2->flag)) {
+		pr = DL_ESTABLISH_CNF;
+	} else if (l2->vs != l2->va) {
+		skb_queue_purge(&l2->i_queue);
+		pr = DL_ESTABLISH_IND;
+	}
+	stop_t200(l2, 5);
+	l2->vr = 0;
+	l2->vs = 0;
+	l2->va = 0;
+	l2->sow = 0;
+	mISDN_FsmChangeState(fi, ST_L2_7);
+	mISDN_FsmAddTimer(&l2->t203, l2->T203, EV_L2_T203, NULL, 4);
+	if (pr != -1)
+		l2up_create(l2, pr, 0, NULL);
+
+	if (skb_queue_len(&l2->i_queue) && cansend(l2))
+		mISDN_FsmEvent(fi, EV_L2_ACK_PULL, NULL);
+
+	if (l2->tm)
+		l2_tei(l2, MDL_STATUS_UP_IND, 0);
+}
+
+static void
+l2_released(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+
+	if (!get_PollFlag(l2, skb)) {
+		l2_mdl_error_ua(fi, event, arg);
+		return;
+	}
+	dev_kfree_skb(skb);
+	stop_t200(l2, 6);
+	lapb_dl_release_l2l3(l2, DL_RELEASE_CNF);
+	mISDN_FsmChangeState(fi, ST_L2_4);
+	if (l2->tm)
+		l2_tei(l2, MDL_STATUS_DOWN_IND, 0);
+}
+
+static void
+l2_reestablish(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+
+	if (!get_PollFlagFree(l2, skb)) {
+		establishlink(fi);
+		test_and_set_bit(FLG_L3_INIT, &l2->flag);
+	}
+}
+
+static void
+l2_st5_dm_release(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+
+	if (get_PollFlagFree(l2, skb)) {
+		stop_t200(l2, 7);
+		if (!test_bit(FLG_L3_INIT, &l2->flag))
+			skb_queue_purge(&l2->i_queue);
+		if (test_bit(FLG_LAPB, &l2->flag))
+			l2down_create(l2, PH_DEACTIVATE_REQ,
+				l2_newid(l2), 0, NULL);
+		st5_dl_release_l2l3(l2);
+		mISDN_FsmChangeState(fi, ST_L2_4);
+		if (l2->tm)
+			l2_tei(l2, MDL_STATUS_DOWN_IND, 0);
+	}
+}
+
+static void
+l2_st6_dm_release(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+
+	if (get_PollFlagFree(l2, skb)) {
+		stop_t200(l2, 8);
+		lapb_dl_release_l2l3(l2, DL_RELEASE_CNF);
+		mISDN_FsmChangeState(fi, ST_L2_4);
+		if (l2->tm)
+			l2_tei(l2, MDL_STATUS_DOWN_IND, 0);
+	}
+}
+
+void
+enquiry_cr(struct layer2 *l2, u_char typ, u_char cr, u_char pf)
+{
+	struct sk_buff *skb;
+	u_char tmp[MAX_L2HEADER_LEN];
+	int i;
+
+	i = sethdraddr(l2, tmp, cr);
+	if (test_bit(FLG_MOD128, &l2->flag)) {
+		tmp[i++] = typ;
+		tmp[i++] = (l2->vr << 1) | (pf ? 1 : 0);
+	} else
+		tmp[i++] = (l2->vr << 5) | typ | (pf ? 0x10 : 0);
+	skb = mI_alloc_skb(i, GFP_ATOMIC);
+	if (!skb) {
+		printk(KERN_WARNING
+		    "isdnl2 can't alloc sbbuff for enquiry_cr\n");
+		return;
+	}
+	memcpy(skb_put(skb, i), tmp, i);
+	enqueue_super(l2, skb);
+}
+
+inline void
+enquiry_response(struct layer2 *l2)
+{
+	if (test_bit(FLG_OWN_BUSY, &l2->flag))
+		enquiry_cr(l2, RNR, RSP, 1);
+	else
+		enquiry_cr(l2, RR, RSP, 1);
+	test_and_clear_bit(FLG_ACK_PEND, &l2->flag);
+}
+
+inline void
+transmit_enquiry(struct layer2 *l2)
+{
+	if (test_bit(FLG_OWN_BUSY, &l2->flag))
+		enquiry_cr(l2, RNR, CMD, 1);
+	else
+		enquiry_cr(l2, RR, CMD, 1);
+	test_and_clear_bit(FLG_ACK_PEND, &l2->flag);
+	start_t200(l2, 9);
+}
+
+
+static void
+nrerrorrecovery(struct FsmInst *fi)
+{
+	struct layer2 *l2 = fi->userdata;
+
+	l2mgr(l2, MDL_ERROR_IND, (void *) 'J');
+	establishlink(fi);
+	test_and_clear_bit(FLG_L3_INIT, &l2->flag);
+}
+
+static void
+invoke_retransmission(struct layer2 *l2, unsigned int nr)
+{
+	u_int	p1;
+
+	if (l2->vs != nr) {
+		while (l2->vs != nr) {
+			(l2->vs)--;
+			if (test_bit(FLG_MOD128, &l2->flag)) {
+				l2->vs %= 128;
+				p1 = (l2->vs - l2->va) % 128;
+			} else {
+				l2->vs %= 8;
+				p1 = (l2->vs - l2->va) % 8;
+			}
+			p1 = (p1 + l2->sow) % l2->window;
+			if (l2->windowar[p1])
+				skb_queue_head(&l2->i_queue, l2->windowar[p1]);
+			else
+				printk(KERN_WARNING
+				    "%s: windowar[%d] is NULL\n",
+				    __func__, p1);
+			l2->windowar[p1] = NULL;
+		}
+		mISDN_FsmEvent(&l2->l2m, EV_L2_ACK_PULL, NULL);
+	}
+}
+
+static void
+l2_st7_got_super(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+	int PollFlag, rsp, typ = RR;
+	unsigned int nr;
+
+	rsp = *skb->data & 0x2;
+	if (test_bit(FLG_ORIG, &l2->flag))
+		rsp = !rsp;
+
+	skb_pull(skb, l2addrsize(l2));
+	if (IsRNR(skb->data, l2)) {
+		set_peer_busy(l2);
+		typ = RNR;
+	} else
+		clear_peer_busy(l2);
+	if (IsREJ(skb->data, l2))
+		typ = REJ;
+
+	if (test_bit(FLG_MOD128, &l2->flag)) {
+		PollFlag = (skb->data[1] & 0x1) == 0x1;
+		nr = skb->data[1] >> 1;
+	} else {
+		PollFlag = (skb->data[0] & 0x10);
+		nr = (skb->data[0] >> 5) & 0x7;
+	}
+	dev_kfree_skb(skb);
+
+	if (PollFlag) {
+		if (rsp)
+			l2mgr(l2, MDL_ERROR_IND, (void *) 'A');
+		else
+			enquiry_response(l2);
+	}
+	if (legalnr(l2, nr)) {
+		if (typ == REJ) {
+			setva(l2, nr);
+			invoke_retransmission(l2, nr);
+			stop_t200(l2, 10);
+			if (mISDN_FsmAddTimer(&l2->t203, l2->T203,
+					EV_L2_T203, NULL, 6))
+				l2m_debug(&l2->l2m, "Restart T203 ST7 REJ");
+		} else if ((nr == l2->vs) && (typ == RR)) {
+			setva(l2, nr);
+			stop_t200(l2, 11);
+			mISDN_FsmRestartTimer(&l2->t203, l2->T203,
+					EV_L2_T203, NULL, 7);
+		} else if ((l2->va != nr) || (typ == RNR)) {
+			setva(l2, nr);
+			if (typ != RR)
+				mISDN_FsmDelTimer(&l2->t203, 9);
+			restart_t200(l2, 12);
+		}
+		if (skb_queue_len(&l2->i_queue) && (typ == RR))
+			mISDN_FsmEvent(fi, EV_L2_ACK_PULL, NULL);
+	} else
+		nrerrorrecovery(fi);
+}
+
+static void
+l2_feed_i_if_reest(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+
+	if (!test_bit(FLG_L3_INIT, &l2->flag))
+		skb_queue_tail(&l2->i_queue, skb);
+	else
+		dev_kfree_skb(skb);
+}
+
+static void
+l2_feed_i_pull(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+
+	skb_queue_tail(&l2->i_queue, skb);
+	mISDN_FsmEvent(fi, EV_L2_ACK_PULL, NULL);
+}
+
+static void
+l2_feed_iqueue(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+
+	skb_queue_tail(&l2->i_queue, skb);
+}
+
+static void
+l2_got_iframe(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2	*l2 = fi->userdata;
+	struct sk_buff	*skb = arg;
+	int		PollFlag, i;
+	u_int		ns, nr;
+
+	i = l2addrsize(l2);
+	if (test_bit(FLG_MOD128, &l2->flag)) {
+		PollFlag = ((skb->data[i + 1] & 0x1) == 0x1);
+		ns = skb->data[i] >> 1;
+		nr = (skb->data[i + 1] >> 1) & 0x7f;
+	} else {
+		PollFlag = (skb->data[i] & 0x10);
+		ns = (skb->data[i] >> 1) & 0x7;
+		nr = (skb->data[i] >> 5) & 0x7;
+	}
+	if (test_bit(FLG_OWN_BUSY, &l2->flag)) {
+		dev_kfree_skb(skb);
+		if (PollFlag)
+			enquiry_response(l2);
+	} else {
+		if (l2->vr == ns) {
+			l2->vr++;
+			if (test_bit(FLG_MOD128, &l2->flag))
+				l2->vr %= 128;
+			else
+				l2->vr %= 8;
+			test_and_clear_bit(FLG_REJEXC, &l2->flag);
+			if (PollFlag)
+				enquiry_response(l2);
+			else
+				test_and_set_bit(FLG_ACK_PEND, &l2->flag);
+			skb_pull(skb, l2headersize(l2, 0));
+			l2up(l2, DL_DATA_IND, skb);
+		} else {
+			/* n(s)!=v(r) */
+			dev_kfree_skb(skb);
+			if (test_and_set_bit(FLG_REJEXC, &l2->flag)) {
+				if (PollFlag)
+					enquiry_response(l2);
+			} else {
+				enquiry_cr(l2, REJ, RSP, PollFlag);
+				test_and_clear_bit(FLG_ACK_PEND, &l2->flag);
+			}
+		}
+	}
+	if (legalnr(l2, nr)) {
+		if (!test_bit(FLG_PEER_BUSY, &l2->flag) &&
+		    (fi->state == ST_L2_7)) {
+			if (nr == l2->vs) {
+				stop_t200(l2, 13);
+				mISDN_FsmRestartTimer(&l2->t203, l2->T203,
+						EV_L2_T203, NULL, 7);
+			} else if (nr != l2->va)
+				restart_t200(l2, 14);
+		}
+		setva(l2, nr);
+	} else {
+		nrerrorrecovery(fi);
+		return;
+	}
+	if (skb_queue_len(&l2->i_queue) && (fi->state == ST_L2_7))
+		mISDN_FsmEvent(fi, EV_L2_ACK_PULL, NULL);
+	if (test_and_clear_bit(FLG_ACK_PEND, &l2->flag))
+		enquiry_cr(l2, RR, RSP, 0);
+}
+
+static void
+l2_got_tei(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2	*l2 = fi->userdata;
+	u_int		info;
+
+	l2->tei = (signed char)(long)arg;
+	set_channel_address(&l2->ch, l2->sapi, l2->tei);
+	info = DL_INFO_L2_CONNECT;
+	l2up_create(l2, DL_INFORMATION_IND, sizeof(info), &info);
+	if (fi->state == ST_L2_3) {
+		establishlink(fi);
+		test_and_set_bit(FLG_L3_INIT, &l2->flag);
+	} else
+		mISDN_FsmChangeState(fi, ST_L2_4);
+	if (skb_queue_len(&l2->ui_queue))
+		tx_ui(l2);
+}
+
+static void
+l2_st5_tout_200(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+
+	if (test_bit(FLG_LAPD, &l2->flag) &&
+		test_bit(FLG_DCHAN_BUSY, &l2->flag)) {
+		mISDN_FsmAddTimer(&l2->t200, l2->T200, EV_L2_T200, NULL, 9);
+	} else if (l2->rc == l2->N200) {
+		mISDN_FsmChangeState(fi, ST_L2_4);
+		test_and_clear_bit(FLG_T200_RUN, &l2->flag);
+		skb_queue_purge(&l2->i_queue);
+		l2mgr(l2, MDL_ERROR_IND, (void *) 'G');
+		if (test_bit(FLG_LAPB, &l2->flag))
+			l2down_create(l2, PH_DEACTIVATE_REQ,
+				l2_newid(l2), 0, NULL);
+		st5_dl_release_l2l3(l2);
+		if (l2->tm)
+			l2_tei(l2, MDL_STATUS_DOWN_IND, 0);
+	} else {
+		l2->rc++;
+		mISDN_FsmAddTimer(&l2->t200, l2->T200, EV_L2_T200, NULL, 9);
+		send_uframe(l2, NULL, (test_bit(FLG_MOD128, &l2->flag) ?
+			SABME : SABM) | 0x10, CMD);
+	}
+}
+
+static void
+l2_st6_tout_200(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+
+	if (test_bit(FLG_LAPD, &l2->flag) &&
+		test_bit(FLG_DCHAN_BUSY, &l2->flag)) {
+		mISDN_FsmAddTimer(&l2->t200, l2->T200, EV_L2_T200, NULL, 9);
+	} else if (l2->rc == l2->N200) {
+		mISDN_FsmChangeState(fi, ST_L2_4);
+		test_and_clear_bit(FLG_T200_RUN, &l2->flag);
+		l2mgr(l2, MDL_ERROR_IND, (void *) 'H');
+		lapb_dl_release_l2l3(l2, DL_RELEASE_CNF);
+		if (l2->tm)
+			l2_tei(l2, MDL_STATUS_DOWN_IND, 0);
+	} else {
+		l2->rc++;
+		mISDN_FsmAddTimer(&l2->t200, l2->T200, EV_L2_T200,
+			    NULL, 9);
+		send_uframe(l2, NULL, DISC | 0x10, CMD);
+	}
+}
+
+static void
+l2_st7_tout_200(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+
+	if (test_bit(FLG_LAPD, &l2->flag) &&
+		test_bit(FLG_DCHAN_BUSY, &l2->flag)) {
+		mISDN_FsmAddTimer(&l2->t200, l2->T200, EV_L2_T200, NULL, 9);
+		return;
+	}
+	test_and_clear_bit(FLG_T200_RUN, &l2->flag);
+	l2->rc = 0;
+	mISDN_FsmChangeState(fi, ST_L2_8);
+	transmit_enquiry(l2);
+	l2->rc++;
+}
+
+static void
+l2_st8_tout_200(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+
+	if (test_bit(FLG_LAPD, &l2->flag) &&
+		test_bit(FLG_DCHAN_BUSY, &l2->flag)) {
+		mISDN_FsmAddTimer(&l2->t200, l2->T200, EV_L2_T200, NULL, 9);
+		return;
+	}
+	test_and_clear_bit(FLG_T200_RUN, &l2->flag);
+	if (l2->rc == l2->N200) {
+		l2mgr(l2, MDL_ERROR_IND, (void *) 'I');
+		establishlink(fi);
+		test_and_clear_bit(FLG_L3_INIT, &l2->flag);
+	} else {
+		transmit_enquiry(l2);
+		l2->rc++;
+	}
+}
+
+static void
+l2_st7_tout_203(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+
+	if (test_bit(FLG_LAPD, &l2->flag) &&
+		test_bit(FLG_DCHAN_BUSY, &l2->flag)) {
+		mISDN_FsmAddTimer(&l2->t203, l2->T203, EV_L2_T203, NULL, 9);
+		return;
+	}
+	mISDN_FsmChangeState(fi, ST_L2_8);
+	transmit_enquiry(l2);
+	l2->rc = 0;
+}
+
+static void
+l2_pull_iqueue(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2	*l2 = fi->userdata;
+	struct sk_buff	*skb, *nskb, *oskb;
+	u_char		header[MAX_L2HEADER_LEN];
+	u_int		i, p1;
+
+	if (!cansend(l2))
+		return;
+
+	skb = skb_dequeue(&l2->i_queue);
+	if (!skb)
+		return;
+
+	if (test_bit(FLG_MOD128, &l2->flag))
+		p1 = (l2->vs - l2->va) % 128;
+	else
+		p1 = (l2->vs - l2->va) % 8;
+	p1 = (p1 + l2->sow) % l2->window;
+	if (l2->windowar[p1]) {
+		printk(KERN_WARNING "isdnl2 try overwrite ack queue entry %d\n",
+		    p1);
+		dev_kfree_skb(l2->windowar[p1]);
+	}
+	l2->windowar[p1] = skb;
+	i = sethdraddr(l2, header, CMD);
+	if (test_bit(FLG_MOD128, &l2->flag)) {
+		header[i++] = l2->vs << 1;
+		header[i++] = l2->vr << 1;
+		l2->vs = (l2->vs + 1) % 128;
+	} else {
+		header[i++] = (l2->vr << 5) | (l2->vs << 1);
+		l2->vs = (l2->vs + 1) % 8;
+	}
+
+	nskb = skb_clone(skb, GFP_ATOMIC);
+	p1 = skb_headroom(nskb);
+	if (p1 >= i)
+		memcpy(skb_push(nskb, i), header, i);
+	else {
+		printk(KERN_WARNING
+		    "isdnl2 pull_iqueue skb header(%d/%d) too short\n", i, p1);
+		oskb = nskb;
+		nskb = mI_alloc_skb(oskb->len + i, GFP_ATOMIC);
+		if (!nskb) {
+			dev_kfree_skb(oskb);
+			printk(KERN_WARNING "%s: no skb mem\n", __func__);
+			return;
+		}
+		memcpy(skb_put(nskb, i), header, i);
+		memcpy(skb_put(nskb, oskb->len), oskb->data, oskb->len);
+		dev_kfree_skb(oskb);
+	}
+	l2down(l2, PH_DATA_REQ, l2_newid(l2), nskb);
+	test_and_clear_bit(FLG_ACK_PEND, &l2->flag);
+	if (!test_and_set_bit(FLG_T200_RUN, &l2->flag)) {
+		mISDN_FsmDelTimer(&l2->t203, 13);
+		mISDN_FsmAddTimer(&l2->t200, l2->T200, EV_L2_T200, NULL, 11);
+	}
+}
+
+static void
+l2_st8_got_super(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+	int PollFlag, rsp, rnr = 0;
+	unsigned int nr;
+
+	rsp = *skb->data & 0x2;
+	if (test_bit(FLG_ORIG, &l2->flag))
+		rsp = !rsp;
+
+	skb_pull(skb, l2addrsize(l2));
+
+	if (IsRNR(skb->data, l2)) {
+		set_peer_busy(l2);
+		rnr = 1;
+	} else
+		clear_peer_busy(l2);
+
+	if (test_bit(FLG_MOD128, &l2->flag)) {
+		PollFlag = (skb->data[1] & 0x1) == 0x1;
+		nr = skb->data[1] >> 1;
+	} else {
+		PollFlag = (skb->data[0] & 0x10);
+		nr = (skb->data[0] >> 5) & 0x7;
+	}
+	dev_kfree_skb(skb);
+	if (rsp && PollFlag) {
+		if (legalnr(l2, nr)) {
+			if (rnr) {
+				restart_t200(l2, 15);
+			} else {
+				stop_t200(l2, 16);
+				mISDN_FsmAddTimer(&l2->t203, l2->T203,
+					    EV_L2_T203, NULL, 5);
+				setva(l2, nr);
+			}
+			invoke_retransmission(l2, nr);
+			mISDN_FsmChangeState(fi, ST_L2_7);
+			if (skb_queue_len(&l2->i_queue) && cansend(l2))
+				mISDN_FsmEvent(fi, EV_L2_ACK_PULL, NULL);
+		} else
+			nrerrorrecovery(fi);
+	} else {
+		if (!rsp && PollFlag)
+			enquiry_response(l2);
+		if (legalnr(l2, nr))
+			setva(l2, nr);
+		else
+			nrerrorrecovery(fi);
+	}
+}
+
+static void
+l2_got_FRMR(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+
+	skb_pull(skb, l2addrsize(l2) + 1);
+
+	if (!(skb->data[0] & 1) || ((skb->data[0] & 3) == 1) || /* I or S */
+	    (IsUA(skb->data) && (fi->state == ST_L2_7))) {
+		l2mgr(l2, MDL_ERROR_IND, (void *) 'K');
+		establishlink(fi);
+		test_and_clear_bit(FLG_L3_INIT, &l2->flag);
+	}
+	dev_kfree_skb(skb);
+}
+
+static void
+l2_st24_tei_remove(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+
+	skb_queue_purge(&l2->ui_queue);
+	l2->tei = GROUP_TEI;
+	mISDN_FsmChangeState(fi, ST_L2_1);
+}
+
+static void
+l2_st3_tei_remove(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+
+	skb_queue_purge(&l2->ui_queue);
+	l2->tei = GROUP_TEI;
+	l2up_create(l2, DL_RELEASE_IND, 0, NULL);
+	mISDN_FsmChangeState(fi, ST_L2_1);
+}
+
+static void
+l2_st5_tei_remove(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+
+	skb_queue_purge(&l2->i_queue);
+	skb_queue_purge(&l2->ui_queue);
+	freewin(l2);
+	l2->tei = GROUP_TEI;
+	stop_t200(l2, 17);
+	st5_dl_release_l2l3(l2);
+	mISDN_FsmChangeState(fi, ST_L2_1);
+}
+
+static void
+l2_st6_tei_remove(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+
+	skb_queue_purge(&l2->ui_queue);
+	l2->tei = GROUP_TEI;
+	stop_t200(l2, 18);
+	l2up_create(l2, DL_RELEASE_IND, 0, NULL);
+	mISDN_FsmChangeState(fi, ST_L2_1);
+}
+
+static void
+l2_tei_remove(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+
+	skb_queue_purge(&l2->i_queue);
+	skb_queue_purge(&l2->ui_queue);
+	freewin(l2);
+	l2->tei = GROUP_TEI;
+	stop_t200(l2, 17);
+	mISDN_FsmDelTimer(&l2->t203, 19);
+	l2up_create(l2, DL_RELEASE_IND, 0, NULL);
+/*	mISDN_queue_data(&l2->inst, l2->inst.id | MSG_BROADCAST,
+ *		MGR_SHORTSTATUS_IND, SSTATUS_L2_RELEASED,
+ *		0, NULL, 0);
+ */
+	mISDN_FsmChangeState(fi, ST_L2_1);
+}
+
+static void
+l2_st14_persistant_da(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+
+	skb_queue_purge(&l2->i_queue);
+	skb_queue_purge(&l2->ui_queue);
+	if (test_and_clear_bit(FLG_ESTAB_PEND, &l2->flag))
+		l2up(l2, DL_RELEASE_IND, skb);
+	else
+		dev_kfree_skb(skb);
+}
+
+static void
+l2_st5_persistant_da(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+
+	skb_queue_purge(&l2->i_queue);
+	skb_queue_purge(&l2->ui_queue);
+	freewin(l2);
+	stop_t200(l2, 19);
+	st5_dl_release_l2l3(l2);
+	mISDN_FsmChangeState(fi, ST_L2_4);
+	if (l2->tm)
+		l2_tei(l2, MDL_STATUS_DOWN_IND, 0);
+	dev_kfree_skb(skb);
+}
+
+static void
+l2_st6_persistant_da(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+
+	skb_queue_purge(&l2->ui_queue);
+	stop_t200(l2, 20);
+	l2up(l2, DL_RELEASE_CNF, skb);
+	mISDN_FsmChangeState(fi, ST_L2_4);
+	if (l2->tm)
+		l2_tei(l2, MDL_STATUS_DOWN_IND, 0);
+}
+
+static void
+l2_persistant_da(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+
+	skb_queue_purge(&l2->i_queue);
+	skb_queue_purge(&l2->ui_queue);
+	freewin(l2);
+	stop_t200(l2, 19);
+	mISDN_FsmDelTimer(&l2->t203, 19);
+	l2up(l2, DL_RELEASE_IND, skb);
+	mISDN_FsmChangeState(fi, ST_L2_4);
+	if (l2->tm)
+		l2_tei(l2, MDL_STATUS_DOWN_IND, 0);
+}
+
+static void
+l2_set_own_busy(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+
+	if (!test_and_set_bit(FLG_OWN_BUSY, &l2->flag)) {
+		enquiry_cr(l2, RNR, RSP, 0);
+		test_and_clear_bit(FLG_ACK_PEND, &l2->flag);
+	}
+	if (skb)
+		dev_kfree_skb(skb);
+}
+
+static void
+l2_clear_own_busy(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+	struct sk_buff *skb = arg;
+
+	if (!test_and_clear_bit(FLG_OWN_BUSY, &l2->flag)) {
+		enquiry_cr(l2, RR, RSP, 0);
+		test_and_clear_bit(FLG_ACK_PEND, &l2->flag);
+	}
+	if (skb)
+		dev_kfree_skb(skb);
+}
+
+static void
+l2_frame_error(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+
+	l2mgr(l2, MDL_ERROR_IND, arg);
+}
+
+static void
+l2_frame_error_reest(struct FsmInst *fi, int event, void *arg)
+{
+	struct layer2 *l2 = fi->userdata;
+
+	l2mgr(l2, MDL_ERROR_IND, arg);
+	establishlink(fi);
+	test_and_clear_bit(FLG_L3_INIT, &l2->flag);
+}
+
+static struct FsmNode L2FnList[] =
+{
+	{ST_L2_1, EV_L2_DL_ESTABLISH_REQ, l2_mdl_assign},
+	{ST_L2_2, EV_L2_DL_ESTABLISH_REQ, l2_go_st3},
+	{ST_L2_4, EV_L2_DL_ESTABLISH_REQ, l2_establish},
+	{ST_L2_5, EV_L2_DL_ESTABLISH_REQ, l2_discard_i_setl3},
+	{ST_L2_7, EV_L2_DL_ESTABLISH_REQ, l2_l3_reestablish},
+	{ST_L2_8, EV_L2_DL_ESTABLISH_REQ, l2_l3_reestablish},
+	{ST_L2_4, EV_L2_DL_RELEASE_REQ, l2_release},
+	{ST_L2_5, EV_L2_DL_RELEASE_REQ, l2_pend_rel},
+	{ST_L2_7, EV_L2_DL_RELEASE_REQ, l2_disconnect},
+	{ST_L2_8, EV_L2_DL_RELEASE_REQ, l2_disconnect},
+	{ST_L2_5, EV_L2_DL_DATA, l2_feed_i_if_reest},
+	{ST_L2_7, EV_L2_DL_DATA, l2_feed_i_pull},
+	{ST_L2_8, EV_L2_DL_DATA, l2_feed_iqueue},
+	{ST_L2_1, EV_L2_DL_UNITDATA, l2_queue_ui_assign},
+	{ST_L2_2, EV_L2_DL_UNITDATA, l2_queue_ui},
+	{ST_L2_3, EV_L2_DL_UNITDATA, l2_queue_ui},
+	{ST_L2_4, EV_L2_DL_UNITDATA, l2_send_ui},
+	{ST_L2_5, EV_L2_DL_UNITDATA, l2_send_ui},
+	{ST_L2_6, EV_L2_DL_UNITDATA, l2_send_ui},
+	{ST_L2_7, EV_L2_DL_UNITDATA, l2_send_ui},
+	{ST_L2_8, EV_L2_DL_UNITDATA, l2_send_ui},
+	{ST_L2_1, EV_L2_MDL_ASSIGN, l2_got_tei},
+	{ST_L2_2, EV_L2_MDL_ASSIGN, l2_got_tei},
+	{ST_L2_3, EV_L2_MDL_ASSIGN, l2_got_tei},
+	{ST_L2_2, EV_L2_MDL_ERROR, l2_st24_tei_remove},
+	{ST_L2_3, EV_L2_MDL_ERROR, l2_st3_tei_remove},
+	{ST_L2_4, EV_L2_MDL_REMOVE, l2_st24_tei_remove},
+	{ST_L2_5, EV_L2_MDL_REMOVE, l2_st5_tei_remove},
+	{ST_L2_6, EV_L2_MDL_REMOVE, l2_st6_tei_remove},
+	{ST_L2_7, EV_L2_MDL_REMOVE, l2_tei_remove},
+	{ST_L2_8, EV_L2_MDL_REMOVE, l2_tei_remove},
+	{ST_L2_4, EV_L2_SABME, l2_start_multi},
+	{ST_L2_5, EV_L2_SABME, l2_send_UA},
+	{ST_L2_6, EV_L2_SABME, l2_send_DM},
+	{ST_L2_7, EV_L2_SABME, l2_restart_multi},
+	{ST_L2_8, EV_L2_SABME, l2_restart_multi},
+	{ST_L2_4, EV_L2_DISC, l2_send_DM},
+	{ST_L2_5, EV_L2_DISC, l2_send_DM},
+	{ST_L2_6, EV_L2_DISC, l2_send_UA},
+	{ST_L2_7, EV_L2_DISC, l2_stop_multi},
+	{ST_L2_8, EV_L2_DISC, l2_stop_multi},
+	{ST_L2_4, EV_L2_UA, l2_mdl_error_ua},
+	{ST_L2_5, EV_L2_UA, l2_connected},
+	{ST_L2_6, EV_L2_UA, l2_released},
+	{ST_L2_7, EV_L2_UA, l2_mdl_error_ua},
+	{ST_L2_8, EV_L2_UA, l2_mdl_error_ua},
+	{ST_L2_4, EV_L2_DM, l2_reestablish},
+	{ST_L2_5, EV_L2_DM, l2_st5_dm_release},
+	{ST_L2_6, EV_L2_DM, l2_st6_dm_release},
+	{ST_L2_7, EV_L2_DM, l2_mdl_error_dm},
+	{ST_L2_8, EV_L2_DM, l2_st8_mdl_error_dm},
+	{ST_L2_1, EV_L2_UI, l2_got_ui},
+	{ST_L2_2, EV_L2_UI, l2_got_ui},
+	{ST_L2_3, EV_L2_UI, l2_got_ui},
+	{ST_L2_4, EV_L2_UI, l2_got_ui},
+	{ST_L2_5, EV_L2_UI, l2_got_ui},
+	{ST_L2_6, EV_L2_UI, l2_got_ui},
+	{ST_L2_7, EV_L2_UI, l2_got_ui},
+	{ST_L2_8, EV_L2_UI, l2_got_ui},
+	{ST_L2_7, EV_L2_FRMR, l2_got_FRMR},
+	{ST_L2_8, EV_L2_FRMR, l2_got_FRMR},
+	{ST_L2_7, EV_L2_SUPER, l2_st7_got_super},
+	{ST_L2_8, EV_L2_SUPER, l2_st8_got_super},
+	{ST_L2_7, EV_L2_I, l2_got_iframe},
+	{ST_L2_8, EV_L2_I, l2_got_iframe},
+	{ST_L2_5, EV_L2_T200, l2_st5_tout_200},
+	{ST_L2_6, EV_L2_T200, l2_st6_tout_200},
+	{ST_L2_7, EV_L2_T200, l2_st7_tout_200},
+	{ST_L2_8, EV_L2_T200, l2_st8_tout_200},
+	{ST_L2_7, EV_L2_T203, l2_st7_tout_203},
+	{ST_L2_7, EV_L2_ACK_PULL, l2_pull_iqueue},
+	{ST_L2_7, EV_L2_SET_OWN_BUSY, l2_set_own_busy},
+	{ST_L2_8, EV_L2_SET_OWN_BUSY, l2_set_own_busy},
+	{ST_L2_7, EV_L2_CLEAR_OWN_BUSY, l2_clear_own_busy},
+	{ST_L2_8, EV_L2_CLEAR_OWN_BUSY, l2_clear_own_busy},
+	{ST_L2_4, EV_L2_FRAME_ERROR, l2_frame_error},
+	{ST_L2_5, EV_L2_FRAME_ERROR, l2_frame_error},
+	{ST_L2_6, EV_L2_FRAME_ERROR, l2_frame_error},
+	{ST_L2_7, EV_L2_FRAME_ERROR, l2_frame_error_reest},
+	{ST_L2_8, EV_L2_FRAME_ERROR, l2_frame_error_reest},
+	{ST_L2_1, EV_L1_DEACTIVATE, l2_st14_persistant_da},
+	{ST_L2_2, EV_L1_DEACTIVATE, l2_st24_tei_remove},
+	{ST_L2_3, EV_L1_DEACTIVATE, l2_st3_tei_remove},
+	{ST_L2_4, EV_L1_DEACTIVATE, l2_st14_persistant_da},
+	{ST_L2_5, EV_L1_DEACTIVATE, l2_st5_persistant_da},
+	{ST_L2_6, EV_L1_DEACTIVATE, l2_st6_persistant_da},
+	{ST_L2_7, EV_L1_DEACTIVATE, l2_persistant_da},
+	{ST_L2_8, EV_L1_DEACTIVATE, l2_persistant_da},
+};
+
+#define L2_FN_COUNT (sizeof(L2FnList)/sizeof(struct FsmNode))
+
+static int
+ph_data_indication(struct layer2 *l2, struct mISDNhead *hh, struct sk_buff *skb)
+{
+	u_char	*datap = skb->data;
+	int	ret = -EINVAL;
+	int	psapi, ptei;
+	u_int	l;
+	int	c = 0;
+
+	l = l2addrsize(l2);
+	if (skb->len <= l) {
+		mISDN_FsmEvent(&l2->l2m, EV_L2_FRAME_ERROR, (void *) 'N');
+		return ret;
+	}
+	if (test_bit(FLG_LAPD, &l2->flag)) { /* Maybe not needed */
+		psapi = *datap++;
+		ptei = *datap++;
+		if ((psapi & 1) || !(ptei & 1)) {
+			printk(KERN_WARNING
+			    "l2 D-channel frame wrong EA0/EA1\n");
+			return ret;
+		}
+		psapi >>= 2;
+		ptei >>= 1;
+		if (psapi != l2->sapi) {
+			/* not our bussiness
+			 * printk(KERN_DEBUG "%s: sapi %d/%d sapi mismatch\n",
+			 *  __func__,
+			 *	psapi, l2->sapi);
+			 */
+			dev_kfree_skb(skb);
+			return 0;
+		}
+		if ((ptei != l2->tei) && (ptei != GROUP_TEI)) {
+			/* not our bussiness
+			 * printk(KERN_DEBUG "%s: tei %d/%d sapi %d mismatch\n",
+			 *  __func__,
+			 *	ptei, l2->tei, psapi);
+			 */
+			dev_kfree_skb(skb);
+			return 0;
+		}
+	} else
+		datap += l;
+	if (!(*datap & 1)) {	/* I-Frame */
+		c = iframe_error(l2, skb);
+		if (!c)
+			ret = mISDN_FsmEvent(&l2->l2m, EV_L2_I, skb);
+	} else if (IsSFrame(datap, l2)) {	/* S-Frame */
+		c = super_error(l2, skb);
+		if (!c)
+			ret = mISDN_FsmEvent(&l2->l2m, EV_L2_SUPER, skb);
+	} else if (IsUI(datap)) {
+		c = UI_error(l2, skb);
+		if (!c)
+			ret = mISDN_FsmEvent(&l2->l2m, EV_L2_UI, skb);
+	} else if (IsSABME(datap, l2)) {
+		c = unnum_error(l2, skb, CMD);
+		if (!c)
+			ret = mISDN_FsmEvent(&l2->l2m, EV_L2_SABME, skb);
+	} else if (IsUA(datap)) {
+		c = unnum_error(l2, skb, RSP);
+		if (!c)
+			ret = mISDN_FsmEvent(&l2->l2m, EV_L2_UA, skb);
+	} else if (IsDISC(datap)) {
+		c = unnum_error(l2, skb, CMD);
+		if (!c)
+			ret = mISDN_FsmEvent(&l2->l2m, EV_L2_DISC, skb);
+	} else if (IsDM(datap)) {
+		c = unnum_error(l2, skb, RSP);
+		if (!c)
+			ret = mISDN_FsmEvent(&l2->l2m, EV_L2_DM, skb);
+	} else if (IsFRMR(datap)) {
+		c = FRMR_error(l2, skb);
+		if (!c)
+			ret = mISDN_FsmEvent(&l2->l2m, EV_L2_FRMR, skb);
+	} else
+		c = 'L';
+	if (c) {
+		printk(KERN_WARNING "l2 D-channel frame error %c\n", c);
+		mISDN_FsmEvent(&l2->l2m, EV_L2_FRAME_ERROR, (void *)(long)c);
+	}
+	return ret;
+}
+
+static int
+l2_send(struct mISDNchannel *ch, struct sk_buff *skb)
+{
+	struct layer2		*l2 = container_of(ch, struct layer2, ch);
+	struct mISDNhead	*hh =  mISDN_HEAD_P(skb);
+	int 			ret = -EINVAL;
+
+	if (*debug & DEBUG_L2_RECV)
+		printk(KERN_DEBUG "%s: prim(%x) id(%x) tei(%d)\n",
+		    __func__, hh->prim, hh->id, l2->tei);
+	switch (hh->prim) {
+	case PH_DATA_IND:
+		ret = ph_data_indication(l2, hh, skb);
+		break;
+	case PH_DATA_CNF:
+		ret = ph_data_confirm(l2, hh, skb);
+		break;
+	case PH_ACTIVATE_IND:
+		test_and_set_bit(FLG_L1_ACTIV, &l2->flag);
+		l2up_create(l2, MPH_ACTIVATE_IND, 0, NULL);
+		if (test_and_clear_bit(FLG_ESTAB_PEND, &l2->flag))
+			ret = mISDN_FsmEvent(&l2->l2m,
+				EV_L2_DL_ESTABLISH_REQ, skb);
+		break;
+	case PH_DEACTIVATE_IND:
+		test_and_clear_bit(FLG_L1_ACTIV, &l2->flag);
+		l2up_create(l2, MPH_DEACTIVATE_IND, 0, NULL);
+		ret = mISDN_FsmEvent(&l2->l2m, EV_L1_DEACTIVATE, skb);
+		break;
+	case MPH_INFORMATION_IND:
+		if (!l2->up)
+			break;
+		ret = l2->up->send(l2->up, skb);
+		break;
+	case DL_DATA_REQ:
+		ret = mISDN_FsmEvent(&l2->l2m, EV_L2_DL_DATA, skb);
+		break;
+	case DL_UNITDATA_REQ:
+		ret = mISDN_FsmEvent(&l2->l2m, EV_L2_DL_UNITDATA, skb);
+		break;
+	case DL_ESTABLISH_REQ:
+		if (test_bit(FLG_LAPB, &l2->flag))
+			test_and_set_bit(FLG_ORIG, &l2->flag);
+		if (test_bit(FLG_L1_ACTIV, &l2->flag)) {
+			if (test_bit(FLG_LAPD, &l2->flag) ||
+				test_bit(FLG_ORIG, &l2->flag))
+				ret = mISDN_FsmEvent(&l2->l2m,
+					EV_L2_DL_ESTABLISH_REQ, skb);
+		} else {
+			if (test_bit(FLG_LAPD, &l2->flag) ||
+				test_bit(FLG_ORIG, &l2->flag)) {
+				test_and_set_bit(FLG_ESTAB_PEND,
+					&l2->flag);
+			}
+			ret = l2down(l2, PH_ACTIVATE_REQ, l2_newid(l2),
+			    skb);
+		}
+		break;
+	case DL_RELEASE_REQ:
+		if (test_bit(FLG_LAPB, &l2->flag))
+			l2down_create(l2, PH_DEACTIVATE_REQ,
+				l2_newid(l2), 0, NULL);
+		ret = mISDN_FsmEvent(&l2->l2m, EV_L2_DL_RELEASE_REQ,
+		    skb);
+		break;
+	default:
+		if (*debug & DEBUG_L2)
+			l2m_debug(&l2->l2m, "l2 unknown pr %04x",
+			    hh->prim);
+	}
+	if (ret) {
+		dev_kfree_skb(skb);
+		ret = 0;
+	}
+	return ret;
+}
+
+int
+tei_l2(struct layer2 *l2, u_int cmd, u_long arg)
+{
+	int		ret = -EINVAL;
+
+	if (*debug & DEBUG_L2_TEI)
+		printk(KERN_DEBUG "%s: cmd(%x)\n", __func__, cmd);
+	switch (cmd) {
+	case (MDL_ASSIGN_REQ):
+		ret = mISDN_FsmEvent(&l2->l2m, EV_L2_MDL_ASSIGN, (void *)arg);
+		break;
+	case (MDL_REMOVE_REQ):
+		ret = mISDN_FsmEvent(&l2->l2m, EV_L2_MDL_REMOVE, NULL);
+		break;
+	case (MDL_ERROR_IND):
+		ret = mISDN_FsmEvent(&l2->l2m, EV_L2_MDL_ERROR, NULL);
+		break;
+	case (MDL_ERROR_RSP):
+		/* ETS 300-125 5.3.2.1 Test: TC13010 */
+		printk(KERN_NOTICE "MDL_ERROR|REQ (tei_l2)\n");
+		ret = mISDN_FsmEvent(&l2->l2m, EV_L2_MDL_ERROR, NULL);
+		break;
+	}
+	return ret;
+}
+
+static void
+release_l2(struct layer2 *l2)
+{
+	mISDN_FsmDelTimer(&l2->t200, 21);
+	mISDN_FsmDelTimer(&l2->t203, 16);
+	skb_queue_purge(&l2->i_queue);
+	skb_queue_purge(&l2->ui_queue);
+	skb_queue_purge(&l2->down_queue);
+	ReleaseWin(l2);
+	if (test_bit(FLG_LAPD, &l2->flag)) {
+		release_tei(l2);
+		if (l2->ch.st)
+			l2->ch.st->dev->D.ctrl(&l2->ch.st->dev->D,
+			    CLOSE_CHANNEL, NULL);
+	}
+	kfree(l2);
+}
+
+static int
+l2_ctrl(struct mISDNchannel *ch, u_int cmd, void *arg)
+{
+	struct layer2		*l2 = container_of(ch, struct layer2, ch);
+	u_int			info;
+
+	if (*debug & DEBUG_L2_CTRL)
+		printk(KERN_DEBUG "%s:(%x)\n", __func__, cmd);
+
+	switch (cmd) {
+	case OPEN_CHANNEL:
+		if (test_bit(FLG_LAPD, &l2->flag)) {
+			set_channel_address(&l2->ch, l2->sapi, l2->tei);
+			info = DL_INFO_L2_CONNECT;
+			l2up_create(l2, DL_INFORMATION_IND,
+			    sizeof(info), &info);
+		}
+		break;
+	case CLOSE_CHANNEL:
+		if (l2->ch.peer)
+			l2->ch.peer->ctrl(l2->ch.peer, CLOSE_CHANNEL, NULL);
+		release_l2(l2);
+		break;
+	}
+	return 0;
+}
+
+struct layer2 *
+create_l2(struct mISDNchannel *ch, u_int protocol, u_long options, u_long arg)
+{
+	struct layer2		*l2;
+	struct channel_req	rq;
+
+	l2 = kzalloc(sizeof(struct layer2), GFP_KERNEL);
+	if (!l2) {
+		printk(KERN_ERR "kzalloc layer2 failed\n");
+		return NULL;
+	}
+	l2->next_id = 1;
+	l2->down_id = MISDN_ID_NONE;
+	l2->up = ch;
+	l2->ch.st = ch->st;
+	l2->ch.send = l2_send;
+	l2->ch.ctrl = l2_ctrl;
+	switch (protocol) {
+	case ISDN_P_LAPD_NT:
+		test_and_set_bit(FLG_LAPD, &l2->flag);
+		test_and_set_bit(FLG_LAPD_NET, &l2->flag);
+		test_and_set_bit(FLG_MOD128, &l2->flag);
+		l2->sapi = 0;
+		l2->maxlen = MAX_DFRAME_LEN;
+		if (test_bit(OPTION_L2_PMX, &options))
+			l2->window = 7;
+		else
+			l2->window = 1;
+		if (test_bit(OPTION_L2_PTP, &options))
+			test_and_set_bit(FLG_PTP, &l2->flag);
+		if (test_bit(OPTION_L2_FIXEDTEI, &options))
+			test_and_set_bit(FLG_FIXED_TEI, &l2->flag);
+		l2->tei = (u_int)arg;
+		l2->T200 = 1000;
+		l2->N200 = 3;
+		l2->T203 = 10000;
+		if (test_bit(OPTION_L2_PMX, &options))
+			rq.protocol = ISDN_P_NT_E1;
+		else
+			rq.protocol = ISDN_P_NT_S0;
+		rq.adr.channel = 0;
+		l2->ch.st->dev->D.ctrl(&l2->ch.st->dev->D, OPEN_CHANNEL, &rq);
+		break;
+	case ISDN_P_LAPD_TE:
+		test_and_set_bit(FLG_LAPD, &l2->flag);
+		test_and_set_bit(FLG_MOD128, &l2->flag);
+		test_and_set_bit(FLG_ORIG, &l2->flag);
+		l2->sapi = 0;
+		l2->maxlen = MAX_DFRAME_LEN;
+		if (test_bit(OPTION_L2_PMX, &options))
+			l2->window = 7;
+		else
+			l2->window = 1;
+		if (test_bit(OPTION_L2_PTP, &options))
+			test_and_set_bit(FLG_PTP, &l2->flag);
+		if (test_bit(OPTION_L2_FIXEDTEI, &options))
+			test_and_set_bit(FLG_FIXED_TEI, &l2->flag);
+		l2->tei = (u_int)arg;
+		l2->T200 = 1000;
+		l2->N200 = 3;
+		l2->T203 = 10000;
+		if (test_bit(OPTION_L2_PMX, &options))
+			rq.protocol = ISDN_P_TE_E1;
+		else
+			rq.protocol = ISDN_P_TE_S0;
+		rq.adr.channel = 0;
+		l2->ch.st->dev->D.ctrl(&l2->ch.st->dev->D, OPEN_CHANNEL, &rq);
+		break;
+	case ISDN_P_B_X75SLP:
+		test_and_set_bit(FLG_LAPB, &l2->flag);
+		l2->window = 7;
+		l2->maxlen = MAX_DATA_SIZE;
+		l2->T200 = 1000;
+		l2->N200 = 4;
+		l2->T203 = 5000;
+		l2->addr.A = 3;
+		l2->addr.B = 1;
+		break;
+	default:
+		printk(KERN_ERR "layer2 create failed prt %x\n",
+			protocol);
+		kfree(l2);
+		return NULL;
+	}
+	skb_queue_head_init(&l2->i_queue);
+	skb_queue_head_init(&l2->ui_queue);
+	skb_queue_head_init(&l2->down_queue);
+	skb_queue_head_init(&l2->tmp_queue);
+	InitWin(l2);
+	l2->l2m.fsm = &l2fsm;
+	if (test_bit(FLG_LAPB, &l2->flag) ||
+		test_bit(FLG_PTP, &l2->flag) ||
+		test_bit(FLG_LAPD_NET, &l2->flag))
+		l2->l2m.state = ST_L2_4;
+	else
+		l2->l2m.state = ST_L2_1;
+	l2->l2m.debug = *debug;
+	l2->l2m.userdata = l2;
+	l2->l2m.userint = 0;
+	l2->l2m.printdebug = l2m_debug;
+
+	mISDN_FsmInitTimer(&l2->l2m, &l2->t200);
+	mISDN_FsmInitTimer(&l2->l2m, &l2->t203);
+	return l2;
+}
+
+static int
+x75create(struct channel_req *crq)
+{
+	struct layer2	*l2;
+
+	if (crq->protocol != ISDN_P_B_X75SLP)
+		return -EPROTONOSUPPORT;
+	l2 = create_l2(crq->ch, crq->protocol, 0, 0);
+	if (!l2)
+		return -ENOMEM;
+	crq->ch = &l2->ch;
+	crq->protocol = ISDN_P_B_HDLC;
+	return 0;
+}
+
+static struct Bprotocol X75SLP = {
+	.Bprotocols = (1 << (ISDN_P_B_X75SLP & ISDN_P_B_MASK)),
+	.name = "X75SLP",
+	.create = x75create
+};
+
+int
+Isdnl2_Init(u_int *deb)
+{
+	debug = deb;
+	mISDN_register_Bprotocol(&X75SLP);
+	l2fsm.state_count = L2_STATE_COUNT;
+	l2fsm.event_count = L2_EVENT_COUNT;
+	l2fsm.strEvent = strL2Event;
+	l2fsm.strState = strL2State;
+	mISDN_FsmNew(&l2fsm, L2FnList, ARRAY_SIZE(L2FnList));
+	TEIInit(deb);
+	return 0;
+}
+
+void
+Isdnl2_cleanup(void)
+{
+	mISDN_unregister_Bprotocol(&X75SLP);
+	TEIFree();
+	mISDN_FsmFree(&l2fsm);
+}
+
diff --git a/drivers/isdn/mISDN/layer2.h b/drivers/isdn/mISDN/layer2.h
new file mode 100644
index 000000000000..de2dd02056a3
--- /dev/null
+++ b/drivers/isdn/mISDN/layer2.h
@@ -0,0 +1,140 @@
+/*
+ * Layer 2 defines
+ *
+ * Copyright 2008  by Karsten Keil <kkeil@novell.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/mISDNif.h>
+#include <linux/skbuff.h>
+#include "fsm.h"
+
+#define MAX_WINDOW	8
+
+struct manager {
+	struct mISDNchannel	ch;
+	struct mISDNchannel	bcast;
+	u_long			options;
+	struct list_head	layer2;
+	rwlock_t		lock;
+	struct FsmInst		deact;
+	struct FsmTimer		datimer;
+	struct sk_buff_head	sendq;
+	struct mISDNchannel	*up;
+	u_int			nextid;
+	u_int			lastid;
+};
+
+struct teimgr {
+	int			ri;
+	int			rcnt;
+	struct FsmInst		tei_m;
+	struct FsmTimer		timer;
+	int			tval, nval;
+	struct layer2		*l2;
+	struct manager		*mgr;
+};
+
+struct laddr {
+	u_char	A;
+	u_char	B;
+};
+
+struct layer2 {
+	struct list_head	list;
+	struct mISDNchannel	ch;
+	u_long			flag;
+	int			id;
+	struct mISDNchannel	*up;
+	signed char		sapi;
+	signed char		tei;
+	struct laddr		addr;
+	u_int			maxlen;
+	struct teimgr		*tm;
+	u_int			vs, va, vr;
+	int			rc;
+	u_int			window;
+	u_int			sow;
+	struct FsmInst		l2m;
+	struct FsmTimer		t200, t203;
+	int			T200, N200, T203;
+	u_int			next_id;
+	u_int			down_id;
+	struct sk_buff		*windowar[MAX_WINDOW];
+	struct sk_buff_head	i_queue;
+	struct sk_buff_head	ui_queue;
+	struct sk_buff_head	down_queue;
+	struct sk_buff_head	tmp_queue;
+};
+
+enum {
+	ST_L2_1,
+	ST_L2_2,
+	ST_L2_3,
+	ST_L2_4,
+	ST_L2_5,
+	ST_L2_6,
+	ST_L2_7,
+	ST_L2_8,
+};
+
+#define L2_STATE_COUNT (ST_L2_8+1)
+
+extern struct layer2	*create_l2(struct mISDNchannel *, u_int,
+				u_long, u_long);
+extern int		tei_l2(struct layer2 *, u_int, u_long arg);
+
+
+/* from tei.c */
+extern int 		l2_tei(struct layer2 *, u_int, u_long arg);
+extern void 		release_tei(struct layer2 *);
+extern int 		TEIInit(u_int *);
+extern void 		TEIFree(void);
+
+#define MAX_L2HEADER_LEN 4
+
+#define RR	0x01
+#define RNR	0x05
+#define REJ	0x09
+#define SABME	0x6f
+#define SABM	0x2f
+#define DM	0x0f
+#define UI	0x03
+#define DISC	0x43
+#define UA	0x63
+#define FRMR	0x87
+#define XID	0xaf
+
+#define CMD	0
+#define RSP	1
+
+#define LC_FLUSH_WAIT 1
+
+#define FLG_LAPB	0
+#define FLG_LAPD	1
+#define FLG_ORIG	2
+#define FLG_MOD128	3
+#define FLG_PEND_REL	4
+#define FLG_L3_INIT	5
+#define FLG_T200_RUN	6
+#define FLG_ACK_PEND	7
+#define FLG_REJEXC	8
+#define FLG_OWN_BUSY	9
+#define FLG_PEER_BUSY	10
+#define FLG_DCHAN_BUSY	11
+#define FLG_L1_ACTIV	12
+#define FLG_ESTAB_PEND	13
+#define FLG_PTP		14
+#define FLG_FIXED_TEI	15
+#define FLG_L2BLOCK	16
+#define FLG_L1_NOTREADY	17
+#define FLG_LAPD_NET	18
diff --git a/drivers/isdn/mISDN/socket.c b/drivers/isdn/mISDN/socket.c
new file mode 100644
index 000000000000..4ba4cc364c9e
--- /dev/null
+++ b/drivers/isdn/mISDN/socket.c
@@ -0,0 +1,781 @@
+/*
+ *
+ * Author	Karsten Keil <kkeil@novell.com>
+ *
+ * Copyright 2008  by Karsten Keil <kkeil@novell.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/mISDNif.h>
+#include "core.h"
+
+static int	*debug;
+
+static struct proto mISDN_proto = {
+	.name		= "misdn",
+	.owner		= THIS_MODULE,
+	.obj_size	= sizeof(struct mISDN_sock)
+};
+
+#define _pms(sk)	((struct mISDN_sock *)sk)
+
+static struct mISDN_sock_list	data_sockets = {
+	.lock = __RW_LOCK_UNLOCKED(data_sockets.lock)
+};
+
+static struct mISDN_sock_list	base_sockets = {
+	.lock = __RW_LOCK_UNLOCKED(base_sockets.lock)
+};
+
+#define L2_HEADER_LEN	4
+
+static inline struct sk_buff *
+_l2_alloc_skb(unsigned int len, gfp_t gfp_mask)
+{
+	struct sk_buff  *skb;
+
+	skb = alloc_skb(len + L2_HEADER_LEN, gfp_mask);
+	if (likely(skb))
+		skb_reserve(skb, L2_HEADER_LEN);
+	return skb;
+}
+
+static void
+mISDN_sock_link(struct mISDN_sock_list *l, struct sock *sk)
+{
+	write_lock_bh(&l->lock);
+	sk_add_node(sk, &l->head);
+	write_unlock_bh(&l->lock);
+}
+
+static void mISDN_sock_unlink(struct mISDN_sock_list *l, struct sock *sk)
+{
+	write_lock_bh(&l->lock);
+	sk_del_node_init(sk);
+	write_unlock_bh(&l->lock);
+}
+
+static int
+mISDN_send(struct mISDNchannel *ch, struct sk_buff *skb)
+{
+	struct mISDN_sock *msk;
+	int	err;
+
+	msk = container_of(ch, struct mISDN_sock, ch);
+	if (*debug & DEBUG_SOCKET)
+		printk(KERN_DEBUG "%s len %d %p\n", __func__, skb->len, skb);
+	if (msk->sk.sk_state == MISDN_CLOSED)
+		return -EUNATCH;
+	__net_timestamp(skb);
+	err = sock_queue_rcv_skb(&msk->sk, skb);
+	if (err)
+		printk(KERN_WARNING "%s: error %d\n", __func__, err);
+	return err;
+}
+
+static int
+mISDN_ctrl(struct mISDNchannel *ch, u_int cmd, void *arg)
+{
+	struct mISDN_sock *msk;
+
+	msk = container_of(ch, struct mISDN_sock, ch);
+	if (*debug & DEBUG_SOCKET)
+		printk(KERN_DEBUG "%s(%p, %x, %p)\n", __func__, ch, cmd, arg);
+	switch (cmd) {
+	case CLOSE_CHANNEL:
+		msk->sk.sk_state = MISDN_CLOSED;
+		break;
+	}
+	return 0;
+}
+
+static inline void
+mISDN_sock_cmsg(struct sock *sk, struct msghdr *msg, struct sk_buff *skb)
+{
+	struct timeval	tv;
+
+	if (_pms(sk)->cmask & MISDN_TIME_STAMP) {
+		skb_get_timestamp(skb, &tv);
+		put_cmsg(msg, SOL_MISDN, MISDN_TIME_STAMP, sizeof(tv), &tv);
+	}
+}
+
+static int
+mISDN_sock_recvmsg(struct kiocb *iocb, struct socket *sock,
+    struct msghdr *msg, size_t len, int flags)
+{
+	struct sk_buff		*skb;
+	struct sock		*sk = sock->sk;
+	struct sockaddr_mISDN	*maddr;
+
+	int		copied, err;
+
+	if (*debug & DEBUG_SOCKET)
+		printk(KERN_DEBUG "%s: len %d, flags %x ch.nr %d, proto %x\n",
+			__func__, (int)len, flags, _pms(sk)->ch.nr,
+			sk->sk_protocol);
+	if (flags & (MSG_OOB))
+		return -EOPNOTSUPP;
+
+	if (sk->sk_state == MISDN_CLOSED)
+		return 0;
+
+	skb = skb_recv_datagram(sk, flags, flags & MSG_DONTWAIT, &err);
+	if (!skb)
+		return err;
+
+	if (msg->msg_namelen >= sizeof(struct sockaddr_mISDN)) {
+		msg->msg_namelen = sizeof(struct sockaddr_mISDN);
+		maddr = (struct sockaddr_mISDN *)msg->msg_name;
+		maddr->family = AF_ISDN;
+		maddr->dev = _pms(sk)->dev->id;
+		if ((sk->sk_protocol == ISDN_P_LAPD_TE) ||
+		    (sk->sk_protocol == ISDN_P_LAPD_NT)) {
+			maddr->channel = (mISDN_HEAD_ID(skb) >> 16) & 0xff;
+			maddr->tei =  (mISDN_HEAD_ID(skb) >> 8) & 0xff;
+			maddr->sapi = mISDN_HEAD_ID(skb) & 0xff;
+		} else {
+			maddr->channel = _pms(sk)->ch.nr;
+			maddr->sapi = _pms(sk)->ch.addr & 0xFF;
+			maddr->tei =  (_pms(sk)->ch.addr >> 8) & 0xFF;
+		}
+	} else {
+		if (msg->msg_namelen)
+			printk(KERN_WARNING "%s: too small namelen %d\n",
+			    __func__, msg->msg_namelen);
+		msg->msg_namelen = 0;
+	}
+
+	copied = skb->len + MISDN_HEADER_LEN;
+	if (len < copied) {
+		if (flags & MSG_PEEK)
+			atomic_dec(&skb->users);
+		else
+			skb_queue_head(&sk->sk_receive_queue, skb);
+		return -ENOSPC;
+	}
+	memcpy(skb_push(skb, MISDN_HEADER_LEN), mISDN_HEAD_P(skb),
+	    MISDN_HEADER_LEN);
+
+	err = skb_copy_datagram_iovec(skb, 0, msg->msg_iov, copied);
+
+	mISDN_sock_cmsg(sk, msg, skb);
+
+	skb_free_datagram(sk, skb);
+
+	return err ? : copied;
+}
+
+static int
+mISDN_sock_sendmsg(struct kiocb *iocb, struct socket *sock,
+    struct msghdr *msg, size_t len)
+{
+	struct sock		*sk = sock->sk;
+	struct sk_buff		*skb;
+	int			err = -ENOMEM;
+	struct sockaddr_mISDN	*maddr;
+
+	if (*debug & DEBUG_SOCKET)
+		printk(KERN_DEBUG "%s: len %d flags %x ch %d proto %x\n",
+		     __func__, (int)len, msg->msg_flags, _pms(sk)->ch.nr,
+		     sk->sk_protocol);
+
+	if (msg->msg_flags & MSG_OOB)
+		return -EOPNOTSUPP;
+
+	if (msg->msg_flags & ~(MSG_DONTWAIT|MSG_NOSIGNAL|MSG_ERRQUEUE))
+		return -EINVAL;
+
+	if (len < MISDN_HEADER_LEN)
+		return -EINVAL;
+
+	if (sk->sk_state != MISDN_BOUND)
+		return -EBADFD;
+
+	lock_sock(sk);
+
+	skb = _l2_alloc_skb(len, GFP_KERNEL);
+	if (!skb)
+		goto done;
+
+	if (memcpy_fromiovec(skb_put(skb, len), msg->msg_iov, len)) {
+		err = -EFAULT;
+		goto drop;
+	}
+
+	memcpy(mISDN_HEAD_P(skb), skb->data, MISDN_HEADER_LEN);
+	skb_pull(skb, MISDN_HEADER_LEN);
+
+	if (msg->msg_namelen >= sizeof(struct sockaddr_mISDN)) {
+		/* if we have a address, we use it */
+		maddr = (struct sockaddr_mISDN *)msg->msg_name;
+		mISDN_HEAD_ID(skb) = maddr->channel;
+	} else { /* use default for L2 messages */
+		if ((sk->sk_protocol == ISDN_P_LAPD_TE) ||
+		    (sk->sk_protocol == ISDN_P_LAPD_NT))
+		    mISDN_HEAD_ID(skb) = _pms(sk)->ch.nr;
+	}
+
+	if (*debug & DEBUG_SOCKET)
+		printk(KERN_DEBUG "%s: ID:%x\n",
+		     __func__, mISDN_HEAD_ID(skb));
+
+	err = -ENODEV;
+	if (!_pms(sk)->ch.peer ||
+	    (err = _pms(sk)->ch.recv(_pms(sk)->ch.peer, skb)))
+		goto drop;
+
+	err = len;
+
+done:
+	release_sock(sk);
+	return err;
+
+drop:
+	kfree_skb(skb);
+	goto done;
+}
+
+static int
+data_sock_release(struct socket *sock)
+{
+	struct sock *sk = sock->sk;
+
+	if (*debug & DEBUG_SOCKET)
+		printk(KERN_DEBUG "%s(%p) sk=%p\n", __func__, sock, sk);
+	if (!sk)
+		return 0;
+	switch (sk->sk_protocol) {
+	case ISDN_P_TE_S0:
+	case ISDN_P_NT_S0:
+	case ISDN_P_TE_E1:
+	case ISDN_P_NT_E1:
+		if (sk->sk_state == MISDN_BOUND)
+			delete_channel(&_pms(sk)->ch);
+		else
+			mISDN_sock_unlink(&data_sockets, sk);
+		break;
+	case ISDN_P_LAPD_TE:
+	case ISDN_P_LAPD_NT:
+	case ISDN_P_B_RAW:
+	case ISDN_P_B_HDLC:
+	case ISDN_P_B_X75SLP:
+	case ISDN_P_B_L2DTMF:
+	case ISDN_P_B_L2DSP:
+	case ISDN_P_B_L2DSPHDLC:
+		delete_channel(&_pms(sk)->ch);
+		mISDN_sock_unlink(&data_sockets, sk);
+		break;
+	}
+
+	lock_sock(sk);
+
+	sock_orphan(sk);
+	skb_queue_purge(&sk->sk_receive_queue);
+
+	release_sock(sk);
+	sock_put(sk);
+
+	return 0;
+}
+
+static int
+data_sock_ioctl_bound(struct sock *sk, unsigned int cmd, void __user *p)
+{
+	struct mISDN_ctrl_req	cq;
+	int			err = -EINVAL, val;
+	struct mISDNchannel	*bchan, *next;
+
+	lock_sock(sk);
+	if (!_pms(sk)->dev) {
+		err = -ENODEV;
+		goto done;
+	}
+	switch (cmd) {
+	case IMCTRLREQ:
+		if (copy_from_user(&cq, p, sizeof(cq))) {
+			err = -EFAULT;
+			break;
+		}
+		if ((sk->sk_protocol & ~ISDN_P_B_MASK) == ISDN_P_B_START) {
+			list_for_each_entry_safe(bchan, next,
+				&_pms(sk)->dev->bchannels, list) {
+				if (bchan->nr == cq.channel) {
+					err = bchan->ctrl(bchan,
+						CONTROL_CHANNEL, &cq);
+					break;
+				}
+			}
+		} else
+			err = _pms(sk)->dev->D.ctrl(&_pms(sk)->dev->D,
+				CONTROL_CHANNEL, &cq);
+		if (err)
+			break;
+		if (copy_to_user(p, &cq, sizeof(cq)))
+			err = -EFAULT;
+		break;
+	case IMCLEAR_L2:
+		if (sk->sk_protocol != ISDN_P_LAPD_NT) {
+			err = -EINVAL;
+			break;
+		}
+		if (get_user(val, (int __user *)p)) {
+			err = -EFAULT;
+			break;
+		}
+		err = _pms(sk)->dev->teimgr->ctrl(_pms(sk)->dev->teimgr,
+		    CONTROL_CHANNEL, &val);
+		break;
+	default:
+		err = -EINVAL;
+		break;
+	}
+done:
+	release_sock(sk);
+	return err;
+}
+
+static int
+data_sock_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
+{
+	int 			err = 0, id;
+	struct sock		*sk = sock->sk;
+	struct mISDNdevice	*dev;
+	struct mISDNversion	ver;
+
+	switch (cmd) {
+	case IMGETVERSION:
+		ver.major = MISDN_MAJOR_VERSION;
+		ver.minor = MISDN_MINOR_VERSION;
+		ver.release = MISDN_RELEASE;
+		if (copy_to_user((void __user *)arg, &ver, sizeof(ver)))
+			err = -EFAULT;
+		break;
+	case IMGETCOUNT:
+		id = get_mdevice_count();
+		if (put_user(id, (int __user *)arg))
+			err = -EFAULT;
+		break;
+	case IMGETDEVINFO:
+		if (get_user(id, (int __user *)arg)) {
+			err = -EFAULT;
+			break;
+		}
+		dev = get_mdevice(id);
+		if (dev) {
+			struct mISDN_devinfo di;
+
+			di.id = dev->id;
+			di.Dprotocols = dev->Dprotocols;
+			di.Bprotocols = dev->Bprotocols | get_all_Bprotocols();
+			di.protocol = dev->D.protocol;
+			memcpy(di.channelmap, dev->channelmap,
+				MISDN_CHMAP_SIZE * 4);
+			di.nrbchan = dev->nrbchan;
+			strcpy(di.name, dev->name);
+			if (copy_to_user((void __user *)arg, &di, sizeof(di)))
+				err = -EFAULT;
+		} else
+			err = -ENODEV;
+		break;
+	default:
+		if (sk->sk_state == MISDN_BOUND)
+			err = data_sock_ioctl_bound(sk, cmd,
+				(void __user *)arg);
+		else
+			err = -ENOTCONN;
+	}
+	return err;
+}
+
+static int data_sock_setsockopt(struct socket *sock, int level, int optname,
+	char __user *optval, int len)
+{
+	struct sock *sk = sock->sk;
+	int err = 0, opt = 0;
+
+	if (*debug & DEBUG_SOCKET)
+		printk(KERN_DEBUG "%s(%p, %d, %x, %p, %d)\n", __func__, sock,
+		    level, optname, optval, len);
+
+	lock_sock(sk);
+
+	switch (optname) {
+	case MISDN_TIME_STAMP:
+		if (get_user(opt, (int __user *)optval)) {
+			err = -EFAULT;
+			break;
+		}
+
+		if (opt)
+			_pms(sk)->cmask |= MISDN_TIME_STAMP;
+		else
+			_pms(sk)->cmask &= ~MISDN_TIME_STAMP;
+		break;
+	default:
+		err = -ENOPROTOOPT;
+		break;
+	}
+	release_sock(sk);
+	return err;
+}
+
+static int data_sock_getsockopt(struct socket *sock, int level, int optname,
+	char __user *optval, int __user *optlen)
+{
+	struct sock *sk = sock->sk;
+	int len, opt;
+
+	if (get_user(len, optlen))
+		return -EFAULT;
+
+	switch (optname) {
+	case MISDN_TIME_STAMP:
+		if (_pms(sk)->cmask & MISDN_TIME_STAMP)
+			opt = 1;
+		else
+			opt = 0;
+
+		if (put_user(opt, optval))
+			return -EFAULT;
+		break;
+	default:
+		return -ENOPROTOOPT;
+	}
+
+	return 0;
+}
+
+static int
+data_sock_bind(struct socket *sock, struct sockaddr *addr, int addr_len)
+{
+	struct sockaddr_mISDN *maddr = (struct sockaddr_mISDN *) addr;
+	struct sock *sk = sock->sk;
+	int err = 0;
+
+	if (*debug & DEBUG_SOCKET)
+		printk(KERN_DEBUG "%s(%p) sk=%p\n", __func__, sock, sk);
+	if (addr_len != sizeof(struct sockaddr_mISDN))
+		return -EINVAL;
+	if (!maddr || maddr->family != AF_ISDN)
+		return -EINVAL;
+
+	lock_sock(sk);
+
+	if (_pms(sk)->dev) {
+		err = -EALREADY;
+		goto done;
+	}
+	_pms(sk)->dev = get_mdevice(maddr->dev);
+	if (!_pms(sk)->dev) {
+		err = -ENODEV;
+		goto done;
+	}
+	_pms(sk)->ch.send = mISDN_send;
+	_pms(sk)->ch.ctrl = mISDN_ctrl;
+
+	switch (sk->sk_protocol) {
+	case ISDN_P_TE_S0:
+	case ISDN_P_NT_S0:
+	case ISDN_P_TE_E1:
+	case ISDN_P_NT_E1:
+		mISDN_sock_unlink(&data_sockets, sk);
+		err = connect_layer1(_pms(sk)->dev, &_pms(sk)->ch,
+		    sk->sk_protocol, maddr);
+		if (err)
+			mISDN_sock_link(&data_sockets, sk);
+		break;
+	case ISDN_P_LAPD_TE:
+	case ISDN_P_LAPD_NT:
+		err = create_l2entity(_pms(sk)->dev, &_pms(sk)->ch,
+		    sk->sk_protocol, maddr);
+		break;
+	case ISDN_P_B_RAW:
+	case ISDN_P_B_HDLC:
+	case ISDN_P_B_X75SLP:
+	case ISDN_P_B_L2DTMF:
+	case ISDN_P_B_L2DSP:
+	case ISDN_P_B_L2DSPHDLC:
+		err = connect_Bstack(_pms(sk)->dev, &_pms(sk)->ch,
+		    sk->sk_protocol, maddr);
+		break;
+	default:
+		err = -EPROTONOSUPPORT;
+	}
+	if (err)
+		goto done;
+	sk->sk_state = MISDN_BOUND;
+	_pms(sk)->ch.protocol = sk->sk_protocol;
+
+done:
+	release_sock(sk);
+	return err;
+}
+
+static int
+data_sock_getname(struct socket *sock, struct sockaddr *addr,
+    int *addr_len, int peer)
+{
+	struct sockaddr_mISDN 	*maddr = (struct sockaddr_mISDN *) addr;
+	struct sock		*sk = sock->sk;
+
+	if (!_pms(sk)->dev)
+		return -EBADFD;
+
+	lock_sock(sk);
+
+	*addr_len = sizeof(*maddr);
+	maddr->dev = _pms(sk)->dev->id;
+	maddr->channel = _pms(sk)->ch.nr;
+	maddr->sapi = _pms(sk)->ch.addr & 0xff;
+	maddr->tei = (_pms(sk)->ch.addr >> 8) & 0xff;
+	release_sock(sk);
+	return 0;
+}
+
+static const struct proto_ops data_sock_ops = {
+	.family		= PF_ISDN,
+	.owner		= THIS_MODULE,
+	.release	= data_sock_release,
+	.ioctl		= data_sock_ioctl,
+	.bind		= data_sock_bind,
+	.getname	= data_sock_getname,
+	.sendmsg	= mISDN_sock_sendmsg,
+	.recvmsg	= mISDN_sock_recvmsg,
+	.poll		= datagram_poll,
+	.listen		= sock_no_listen,
+	.shutdown	= sock_no_shutdown,
+	.setsockopt	= data_sock_setsockopt,
+	.getsockopt	= data_sock_getsockopt,
+	.connect	= sock_no_connect,
+	.socketpair	= sock_no_socketpair,
+	.accept		= sock_no_accept,
+	.mmap		= sock_no_mmap
+};
+
+static int
+data_sock_create(struct net *net, struct socket *sock, int protocol)
+{
+	struct sock *sk;
+
+	if (sock->type != SOCK_DGRAM)
+		return -ESOCKTNOSUPPORT;
+
+	sk = sk_alloc(net, PF_ISDN, GFP_KERNEL, &mISDN_proto);
+	if (!sk)
+		return -ENOMEM;
+
+	sock_init_data(sock, sk);
+
+	sock->ops = &data_sock_ops;
+	sock->state = SS_UNCONNECTED;
+	sock_reset_flag(sk, SOCK_ZAPPED);
+
+	sk->sk_protocol = protocol;
+	sk->sk_state    = MISDN_OPEN;
+	mISDN_sock_link(&data_sockets, sk);
+
+	return 0;
+}
+
+static int
+base_sock_release(struct socket *sock)
+{
+	struct sock *sk = sock->sk;
+
+	printk(KERN_DEBUG "%s(%p) sk=%p\n", __func__, sock, sk);
+	if (!sk)
+		return 0;
+
+	mISDN_sock_unlink(&base_sockets, sk);
+	sock_orphan(sk);
+	sock_put(sk);
+
+	return 0;
+}
+
+static int
+base_sock_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
+{
+	int 			err = 0, id;
+	struct mISDNdevice	*dev;
+	struct mISDNversion	ver;
+
+	switch (cmd) {
+	case IMGETVERSION:
+		ver.major = MISDN_MAJOR_VERSION;
+		ver.minor = MISDN_MINOR_VERSION;
+		ver.release = MISDN_RELEASE;
+		if (copy_to_user((void __user *)arg, &ver, sizeof(ver)))
+			err = -EFAULT;
+		break;
+	case IMGETCOUNT:
+		id = get_mdevice_count();
+		if (put_user(id, (int __user *)arg))
+			err = -EFAULT;
+		break;
+	case IMGETDEVINFO:
+		if (get_user(id, (int __user *)arg)) {
+			err = -EFAULT;
+			break;
+		}
+		dev = get_mdevice(id);
+		if (dev) {
+			struct mISDN_devinfo di;
+
+			di.id = dev->id;
+			di.Dprotocols = dev->Dprotocols;
+			di.Bprotocols = dev->Bprotocols | get_all_Bprotocols();
+			di.protocol = dev->D.protocol;
+			memcpy(di.channelmap, dev->channelmap,
+				MISDN_CHMAP_SIZE * 4);
+			di.nrbchan = dev->nrbchan;
+			strcpy(di.name, dev->name);
+			if (copy_to_user((void __user *)arg, &di, sizeof(di)))
+				err = -EFAULT;
+		} else
+			err = -ENODEV;
+		break;
+	default:
+		err = -EINVAL;
+	}
+	return err;
+}
+
+static int
+base_sock_bind(struct socket *sock, struct sockaddr *addr, int addr_len)
+{
+	struct sockaddr_mISDN *maddr = (struct sockaddr_mISDN *) addr;
+	struct sock *sk = sock->sk;
+	int err = 0;
+
+	if (!maddr || maddr->family != AF_ISDN)
+		return -EINVAL;
+
+	lock_sock(sk);
+
+	if (_pms(sk)->dev) {
+		err = -EALREADY;
+		goto done;
+	}
+
+	_pms(sk)->dev = get_mdevice(maddr->dev);
+	if (!_pms(sk)->dev) {
+		err = -ENODEV;
+		goto done;
+	}
+	sk->sk_state = MISDN_BOUND;
+
+done:
+	release_sock(sk);
+	return err;
+}
+
+static const struct proto_ops base_sock_ops = {
+	.family		= PF_ISDN,
+	.owner		= THIS_MODULE,
+	.release	= base_sock_release,
+	.ioctl		= base_sock_ioctl,
+	.bind		= base_sock_bind,
+	.getname	= sock_no_getname,
+	.sendmsg	= sock_no_sendmsg,
+	.recvmsg	= sock_no_recvmsg,
+	.poll		= sock_no_poll,
+	.listen		= sock_no_listen,
+	.shutdown	= sock_no_shutdown,
+	.setsockopt	= sock_no_setsockopt,
+	.getsockopt	= sock_no_getsockopt,
+	.connect	= sock_no_connect,
+	.socketpair	= sock_no_socketpair,
+	.accept		= sock_no_accept,
+	.mmap		= sock_no_mmap
+};
+
+
+static int
+base_sock_create(struct net *net, struct socket *sock, int protocol)
+{
+	struct sock *sk;
+
+	if (sock->type != SOCK_RAW)
+		return -ESOCKTNOSUPPORT;
+
+	sk = sk_alloc(net, PF_ISDN, GFP_KERNEL, &mISDN_proto);
+	if (!sk)
+		return -ENOMEM;
+
+	sock_init_data(sock, sk);
+	sock->ops = &base_sock_ops;
+	sock->state = SS_UNCONNECTED;
+	sock_reset_flag(sk, SOCK_ZAPPED);
+	sk->sk_protocol = protocol;
+	sk->sk_state    = MISDN_OPEN;
+	mISDN_sock_link(&base_sockets, sk);
+
+	return 0;
+}
+
+static int
+mISDN_sock_create(struct net *net, struct socket *sock, int proto)
+{
+	int err = -EPROTONOSUPPORT;
+
+	switch	(proto) {
+	case ISDN_P_BASE:
+		err = base_sock_create(net, sock, proto);
+		break;
+	case ISDN_P_TE_S0:
+	case ISDN_P_NT_S0:
+	case ISDN_P_TE_E1:
+	case ISDN_P_NT_E1:
+	case ISDN_P_LAPD_TE:
+	case ISDN_P_LAPD_NT:
+	case ISDN_P_B_RAW:
+	case ISDN_P_B_HDLC:
+	case ISDN_P_B_X75SLP:
+	case ISDN_P_B_L2DTMF:
+	case ISDN_P_B_L2DSP:
+	case ISDN_P_B_L2DSPHDLC:
+		err = data_sock_create(net, sock, proto);
+		break;
+	default:
+		return err;
+	}
+
+	return err;
+}
+
+static struct
+net_proto_family mISDN_sock_family_ops = {
+	.owner  = THIS_MODULE,
+	.family = PF_ISDN,
+	.create = mISDN_sock_create,
+};
+
+int
+misdn_sock_init(u_int *deb)
+{
+	int err;
+
+	debug = deb;
+	err = sock_register(&mISDN_sock_family_ops);
+	if (err)
+		printk(KERN_ERR "%s: error(%d)\n", __func__, err);
+	return err;
+}
+
+void
+misdn_sock_cleanup(void)
+{
+	sock_unregister(PF_ISDN);
+}
+
diff --git a/drivers/isdn/mISDN/stack.c b/drivers/isdn/mISDN/stack.c
new file mode 100644
index 000000000000..54cfddcc4784
--- /dev/null
+++ b/drivers/isdn/mISDN/stack.c
@@ -0,0 +1,674 @@
+/*
+ *
+ * Author	Karsten Keil <kkeil@novell.com>
+ *
+ * Copyright 2008  by Karsten Keil <kkeil@novell.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/mISDNif.h>
+#include <linux/kthread.h>
+#include "core.h"
+
+static u_int	*debug;
+
+static inline void
+_queue_message(struct mISDNstack *st, struct sk_buff *skb)
+{
+	struct mISDNhead	*hh = mISDN_HEAD_P(skb);
+
+	if (*debug & DEBUG_QUEUE_FUNC)
+		printk(KERN_DEBUG "%s prim(%x) id(%x) %p\n",
+		    __func__, hh->prim, hh->id, skb);
+	skb_queue_tail(&st->msgq, skb);
+	if (likely(!test_bit(mISDN_STACK_STOPPED, &st->status))) {
+		test_and_set_bit(mISDN_STACK_WORK, &st->status);
+		wake_up_interruptible(&st->workq);
+	}
+}
+
+int
+mISDN_queue_message(struct mISDNchannel *ch, struct sk_buff *skb)
+{
+	_queue_message(ch->st, skb);
+	return 0;
+}
+
+static struct mISDNchannel *
+get_channel4id(struct mISDNstack *st, u_int id)
+{
+	struct mISDNchannel	*ch;
+
+	mutex_lock(&st->lmutex);
+	list_for_each_entry(ch, &st->layer2, list) {
+		if (id == ch->nr)
+			goto unlock;
+	}
+	ch = NULL;
+unlock:
+	mutex_unlock(&st->lmutex);
+	return ch;
+}
+
+static void
+send_socklist(struct mISDN_sock_list *sl, struct sk_buff *skb)
+{
+	struct hlist_node	*node;
+	struct sock		*sk;
+	struct sk_buff		*cskb = NULL;
+
+	read_lock(&sl->lock);
+	sk_for_each(sk, node, &sl->head) {
+		if (sk->sk_state != MISDN_BOUND)
+			continue;
+		if (!cskb)
+			cskb = skb_copy(skb, GFP_KERNEL);
+		if (!cskb) {
+			printk(KERN_WARNING "%s no skb\n", __func__);
+			break;
+		}
+		if (!sock_queue_rcv_skb(sk, cskb))
+			cskb = NULL;
+	}
+	read_unlock(&sl->lock);
+	if (cskb)
+		dev_kfree_skb(cskb);
+}
+
+static void
+send_layer2(struct mISDNstack *st, struct sk_buff *skb)
+{
+	struct sk_buff		*cskb;
+	struct mISDNhead	*hh = mISDN_HEAD_P(skb);
+	struct mISDNchannel	*ch;
+	int			ret;
+
+	if (!st)
+		return;
+	mutex_lock(&st->lmutex);
+	if ((hh->id & MISDN_ID_ADDR_MASK) == MISDN_ID_ANY) { /* L2 for all */
+		list_for_each_entry(ch, &st->layer2, list) {
+			if (list_is_last(&ch->list, &st->layer2)) {
+				cskb = skb;
+				skb = NULL;
+			} else {
+				cskb = skb_copy(skb, GFP_KERNEL);
+			}
+			if (cskb) {
+				ret = ch->send(ch, cskb);
+				if (ret) {
+					if (*debug & DEBUG_SEND_ERR)
+						printk(KERN_DEBUG
+						    "%s ch%d prim(%x) addr(%x)"
+						    " err %d\n",
+						    __func__, ch->nr,
+						    hh->prim, ch->addr, ret);
+					dev_kfree_skb(cskb);
+				}
+			} else {
+				printk(KERN_WARNING "%s ch%d addr %x no mem\n",
+				    __func__, ch->nr, ch->addr);
+				goto out;
+			}
+		}
+	} else {
+		list_for_each_entry(ch, &st->layer2, list) {
+			if ((hh->id & MISDN_ID_ADDR_MASK) == ch->addr) {
+				ret = ch->send(ch, skb);
+				if (!ret)
+					skb = NULL;
+				goto out;
+			}
+		}
+		ret = st->dev->teimgr->ctrl(st->dev->teimgr, CHECK_DATA, skb);
+		if (!ret)
+			skb = NULL;
+		else if (*debug & DEBUG_SEND_ERR)
+			printk(KERN_DEBUG
+			    "%s ch%d mgr prim(%x) addr(%x) err %d\n",
+			    __func__, ch->nr, hh->prim, ch->addr, ret);
+	}
+out:
+	mutex_unlock(&st->lmutex);
+	if (skb)
+		dev_kfree_skb(skb);
+}
+
+static inline int
+send_msg_to_layer(struct mISDNstack *st, struct sk_buff *skb)
+{
+	struct mISDNhead	*hh = mISDN_HEAD_P(skb);
+	struct mISDNchannel	*ch;
+	int	lm;
+
+	lm = hh->prim & MISDN_LAYERMASK;
+	if (*debug & DEBUG_QUEUE_FUNC)
+		printk(KERN_DEBUG "%s prim(%x) id(%x) %p\n",
+		    __func__, hh->prim, hh->id, skb);
+	if (lm == 0x1) {
+		if (!hlist_empty(&st->l1sock.head)) {
+			__net_timestamp(skb);
+			send_socklist(&st->l1sock, skb);
+		}
+		return st->layer1->send(st->layer1, skb);
+	} else if (lm == 0x2) {
+		if (!hlist_empty(&st->l1sock.head))
+			send_socklist(&st->l1sock, skb);
+		send_layer2(st, skb);
+		return 0;
+	} else if (lm == 0x4) {
+		ch = get_channel4id(st, hh->id);
+		if (ch)
+			return ch->send(ch, skb);
+		else
+			printk(KERN_WARNING
+			    "%s: dev(%s) prim(%x) id(%x) no channel\n",
+			    __func__, st->dev->name, hh->prim, hh->id);
+	} else if (lm == 0x8) {
+		WARN_ON(lm == 0x8);
+		ch = get_channel4id(st, hh->id);
+		if (ch)
+			return ch->send(ch, skb);
+		else
+			printk(KERN_WARNING
+			    "%s: dev(%s) prim(%x) id(%x) no channel\n",
+			    __func__, st->dev->name, hh->prim, hh->id);
+	} else {
+		/* broadcast not handled yet */
+		printk(KERN_WARNING "%s: dev(%s) prim %x not delivered\n",
+		    __func__, st->dev->name, hh->prim);
+	}
+	return -ESRCH;
+}
+
+static void
+do_clear_stack(struct mISDNstack *st)
+{
+}
+
+static int
+mISDNStackd(void *data)
+{
+	struct mISDNstack *st = data;
+	int err = 0;
+
+#ifdef CONFIG_SMP
+	lock_kernel();
+#endif
+	sigfillset(&current->blocked);
+#ifdef CONFIG_SMP
+	unlock_kernel();
+#endif
+	if (*debug & DEBUG_MSG_THREAD)
+		printk(KERN_DEBUG "mISDNStackd %s started\n", st->dev->name);
+
+	if (st->notify != NULL) {
+		complete(st->notify);
+		st->notify = NULL;
+	}
+
+	for (;;) {
+		struct sk_buff	*skb;
+
+		if (unlikely(test_bit(mISDN_STACK_STOPPED, &st->status))) {
+			test_and_clear_bit(mISDN_STACK_WORK, &st->status);
+			test_and_clear_bit(mISDN_STACK_RUNNING, &st->status);
+		} else
+			test_and_set_bit(mISDN_STACK_RUNNING, &st->status);
+		while (test_bit(mISDN_STACK_WORK, &st->status)) {
+			skb = skb_dequeue(&st->msgq);
+			if (!skb) {
+				test_and_clear_bit(mISDN_STACK_WORK,
+					&st->status);
+				/* test if a race happens */
+				skb = skb_dequeue(&st->msgq);
+				if (!skb)
+					continue;
+				test_and_set_bit(mISDN_STACK_WORK,
+				    &st->status);
+			}
+#ifdef MISDN_MSG_STATS
+			st->msg_cnt++;
+#endif
+			err = send_msg_to_layer(st, skb);
+			if (unlikely(err)) {
+				if (*debug & DEBUG_SEND_ERR)
+					printk(KERN_DEBUG
+					    "%s: %s prim(%x) id(%x) "
+					    "send call(%d)\n",
+					    __func__, st->dev->name,
+					    mISDN_HEAD_PRIM(skb),
+					    mISDN_HEAD_ID(skb), err);
+				dev_kfree_skb(skb);
+				continue;
+			}
+			if (unlikely(test_bit(mISDN_STACK_STOPPED,
+			    &st->status))) {
+				test_and_clear_bit(mISDN_STACK_WORK,
+				    &st->status);
+				test_and_clear_bit(mISDN_STACK_RUNNING,
+				    &st->status);
+				break;
+			}
+		}
+		if (test_bit(mISDN_STACK_CLEARING, &st->status)) {
+			test_and_set_bit(mISDN_STACK_STOPPED, &st->status);
+			test_and_clear_bit(mISDN_STACK_RUNNING, &st->status);
+			do_clear_stack(st);
+			test_and_clear_bit(mISDN_STACK_CLEARING, &st->status);
+			test_and_set_bit(mISDN_STACK_RESTART, &st->status);
+		}
+		if (test_and_clear_bit(mISDN_STACK_RESTART, &st->status)) {
+			test_and_clear_bit(mISDN_STACK_STOPPED, &st->status);
+			test_and_set_bit(mISDN_STACK_RUNNING, &st->status);
+			if (!skb_queue_empty(&st->msgq))
+				test_and_set_bit(mISDN_STACK_WORK,
+				    &st->status);
+		}
+		if (test_bit(mISDN_STACK_ABORT, &st->status))
+			break;
+		if (st->notify != NULL) {
+			complete(st->notify);
+			st->notify = NULL;
+		}
+#ifdef MISDN_MSG_STATS
+		st->sleep_cnt++;
+#endif
+		test_and_clear_bit(mISDN_STACK_ACTIVE, &st->status);
+		wait_event_interruptible(st->workq, (st->status &
+		    mISDN_STACK_ACTION_MASK));
+		if (*debug & DEBUG_MSG_THREAD)
+			printk(KERN_DEBUG "%s: %s wake status %08lx\n",
+			    __func__, st->dev->name, st->status);
+		test_and_set_bit(mISDN_STACK_ACTIVE, &st->status);
+
+		test_and_clear_bit(mISDN_STACK_WAKEUP, &st->status);
+
+		if (test_bit(mISDN_STACK_STOPPED, &st->status)) {
+			test_and_clear_bit(mISDN_STACK_RUNNING, &st->status);
+#ifdef MISDN_MSG_STATS
+			st->stopped_cnt++;
+#endif
+		}
+	}
+#ifdef MISDN_MSG_STATS
+	printk(KERN_DEBUG "mISDNStackd daemon for %s proceed %d "
+	    "msg %d sleep %d stopped\n",
+	    st->dev->name, st->msg_cnt, st->sleep_cnt, st->stopped_cnt);
+	printk(KERN_DEBUG
+	    "mISDNStackd daemon for %s utime(%ld) stime(%ld)\n",
+	    st->dev->name, st->thread->utime, st->thread->stime);
+	printk(KERN_DEBUG
+	    "mISDNStackd daemon for %s nvcsw(%ld) nivcsw(%ld)\n",
+	    st->dev->name, st->thread->nvcsw, st->thread->nivcsw);
+	printk(KERN_DEBUG "mISDNStackd daemon for %s killed now\n",
+	    st->dev->name);
+#endif
+	test_and_set_bit(mISDN_STACK_KILLED, &st->status);
+	test_and_clear_bit(mISDN_STACK_RUNNING, &st->status);
+	test_and_clear_bit(mISDN_STACK_ACTIVE, &st->status);
+	test_and_clear_bit(mISDN_STACK_ABORT, &st->status);
+	skb_queue_purge(&st->msgq);
+	st->thread = NULL;
+	if (st->notify != NULL) {
+		complete(st->notify);
+		st->notify = NULL;
+	}
+	return 0;
+}
+
+static int
+l1_receive(struct mISDNchannel *ch, struct sk_buff *skb)
+{
+	if (!ch->st)
+		return -ENODEV;
+	__net_timestamp(skb);
+	_queue_message(ch->st, skb);
+	return 0;
+}
+
+void
+set_channel_address(struct mISDNchannel *ch, u_int sapi, u_int tei)
+{
+	ch->addr = sapi | (tei << 8);
+}
+
+void
+__add_layer2(struct mISDNchannel *ch, struct mISDNstack *st)
+{
+	list_add_tail(&ch->list, &st->layer2);
+}
+
+void
+add_layer2(struct mISDNchannel *ch, struct mISDNstack *st)
+{
+	mutex_lock(&st->lmutex);
+	__add_layer2(ch, st);
+	mutex_unlock(&st->lmutex);
+}
+
+static int
+st_own_ctrl(struct mISDNchannel *ch, u_int cmd, void *arg)
+{
+	if (!ch->st || ch->st->layer1)
+		return -EINVAL;
+	return ch->st->layer1->ctrl(ch->st->layer1, cmd, arg);
+}
+
+int
+create_stack(struct mISDNdevice *dev)
+{
+	struct mISDNstack	*newst;
+	int			err;
+	DECLARE_COMPLETION_ONSTACK(done);
+
+	newst = kzalloc(sizeof(struct mISDNstack), GFP_KERNEL);
+	if (!newst) {
+		printk(KERN_ERR "kmalloc mISDN_stack failed\n");
+		return -ENOMEM;
+	}
+	newst->dev = dev;
+	INIT_LIST_HEAD(&newst->layer2);
+	INIT_HLIST_HEAD(&newst->l1sock.head);
+	rwlock_init(&newst->l1sock.lock);
+	init_waitqueue_head(&newst->workq);
+	skb_queue_head_init(&newst->msgq);
+	mutex_init(&newst->lmutex);
+	dev->D.st = newst;
+	err = create_teimanager(dev);
+	if (err) {
+		printk(KERN_ERR "kmalloc teimanager failed\n");
+		kfree(newst);
+		return err;
+	}
+	dev->teimgr->peer = &newst->own;
+	dev->teimgr->recv = mISDN_queue_message;
+	dev->teimgr->st = newst;
+	newst->layer1 = &dev->D;
+	dev->D.recv = l1_receive;
+	dev->D.peer = &newst->own;
+	newst->own.st = newst;
+	newst->own.ctrl = st_own_ctrl;
+	newst->own.send = mISDN_queue_message;
+	newst->own.recv = mISDN_queue_message;
+	if (*debug & DEBUG_CORE_FUNC)
+		printk(KERN_DEBUG "%s: st(%s)\n", __func__, newst->dev->name);
+	newst->notify = &done;
+	newst->thread = kthread_run(mISDNStackd, (void *)newst, "mISDN_%s",
+		newst->dev->name);
+	if (IS_ERR(newst->thread)) {
+		err = PTR_ERR(newst->thread);
+		printk(KERN_ERR
+			"mISDN:cannot create kernel thread for %s (%d)\n",
+			newst->dev->name, err);
+		delete_teimanager(dev->teimgr);
+		kfree(newst);
+	} else
+		wait_for_completion(&done);
+	return err;
+}
+
+int
+connect_layer1(struct mISDNdevice *dev, struct mISDNchannel *ch,
+		u_int protocol, struct sockaddr_mISDN *adr)
+{
+	struct mISDN_sock	*msk = container_of(ch, struct mISDN_sock, ch);
+	struct channel_req	rq;
+	int			err;
+
+
+	if (*debug &  DEBUG_CORE_FUNC)
+		printk(KERN_DEBUG "%s: %s proto(%x) adr(%d %d %d %d)\n",
+			__func__, dev->name, protocol, adr->dev, adr->channel,
+			 adr->sapi, adr->tei);
+	switch (protocol) {
+	case ISDN_P_NT_S0:
+	case ISDN_P_NT_E1:
+	case ISDN_P_TE_S0:
+	case ISDN_P_TE_E1:
+#ifdef PROTOCOL_CHECK
+		/* this should be enhanced */
+		if (!list_empty(&dev->D.st->layer2)
+			&& dev->D.protocol != protocol)
+			return -EBUSY;
+		if (!hlist_empty(&dev->D.st->l1sock.head)
+			&& dev->D.protocol != protocol)
+			return -EBUSY;
+#endif
+		ch->recv = mISDN_queue_message;
+		ch->peer = &dev->D.st->own;
+		ch->st = dev->D.st;
+		rq.protocol = protocol;
+		rq.adr.channel = 0;
+		err = dev->D.ctrl(&dev->D, OPEN_CHANNEL, &rq);
+		printk(KERN_DEBUG "%s: ret 1 %d\n", __func__, err);
+		if (err)
+			return err;
+		write_lock_bh(&dev->D.st->l1sock.lock);
+		sk_add_node(&msk->sk, &dev->D.st->l1sock.head);
+		write_unlock_bh(&dev->D.st->l1sock.lock);
+		break;
+	default:
+		return -ENOPROTOOPT;
+	}
+	return 0;
+}
+
+int
+connect_Bstack(struct mISDNdevice *dev, struct mISDNchannel *ch,
+    u_int protocol, struct sockaddr_mISDN *adr)
+{
+	struct channel_req	rq, rq2;
+	int			pmask, err;
+	struct Bprotocol	*bp;
+
+	if (*debug &  DEBUG_CORE_FUNC)
+		printk(KERN_DEBUG "%s: %s proto(%x) adr(%d %d %d %d)\n",
+			__func__, dev->name, protocol,
+			adr->dev, adr->channel, adr->sapi,
+			adr->tei);
+	ch->st = dev->D.st;
+	pmask = 1 << (protocol & ISDN_P_B_MASK);
+	if (pmask & dev->Bprotocols) {
+		rq.protocol = protocol;
+		rq.adr = *adr;
+		err = dev->D.ctrl(&dev->D, OPEN_CHANNEL, &rq);
+		if (err)
+			return err;
+		ch->recv = rq.ch->send;
+		ch->peer = rq.ch;
+		rq.ch->recv = ch->send;
+		rq.ch->peer = ch;
+		rq.ch->st = dev->D.st;
+	} else {
+		bp = get_Bprotocol4mask(pmask);
+		if (!bp)
+			return -ENOPROTOOPT;
+		rq2.protocol = protocol;
+		rq2.adr = *adr;
+		rq2.ch = ch;
+		err = bp->create(&rq2);
+		if (err)
+			return err;
+		ch->recv = rq2.ch->send;
+		ch->peer = rq2.ch;
+		rq2.ch->st = dev->D.st;
+		rq.protocol = rq2.protocol;
+		rq.adr = *adr;
+		err = dev->D.ctrl(&dev->D, OPEN_CHANNEL, &rq);
+		if (err) {
+			rq2.ch->ctrl(rq2.ch, CLOSE_CHANNEL, NULL);
+			return err;
+		}
+		rq2.ch->recv = rq.ch->send;
+		rq2.ch->peer = rq.ch;
+		rq.ch->recv = rq2.ch->send;
+		rq.ch->peer = rq2.ch;
+		rq.ch->st = dev->D.st;
+	}
+	ch->protocol = protocol;
+	ch->nr = rq.ch->nr;
+	return 0;
+}
+
+int
+create_l2entity(struct mISDNdevice *dev, struct mISDNchannel *ch,
+    u_int protocol, struct sockaddr_mISDN *adr)
+{
+	struct channel_req	rq;
+	int			err;
+
+	if (*debug &  DEBUG_CORE_FUNC)
+		printk(KERN_DEBUG "%s: %s proto(%x) adr(%d %d %d %d)\n",
+			__func__, dev->name, protocol,
+			adr->dev, adr->channel, adr->sapi,
+			adr->tei);
+	rq.protocol = ISDN_P_TE_S0;
+	if (dev->Dprotocols & (1 << ISDN_P_TE_E1))
+		rq.protocol = ISDN_P_TE_E1;
+	switch (protocol) {
+	case ISDN_P_LAPD_NT:
+		rq.protocol = ISDN_P_NT_S0;
+		if (dev->Dprotocols & (1 << ISDN_P_NT_E1))
+			rq.protocol = ISDN_P_NT_E1;
+	case ISDN_P_LAPD_TE:
+#ifdef PROTOCOL_CHECK
+		/* this should be enhanced */
+		if (!list_empty(&dev->D.st->layer2)
+			&& dev->D.protocol != protocol)
+			return -EBUSY;
+		if (!hlist_empty(&dev->D.st->l1sock.head)
+			&& dev->D.protocol != protocol)
+			return -EBUSY;
+#endif
+		ch->recv = mISDN_queue_message;
+		ch->peer = &dev->D.st->own;
+		ch->st = dev->D.st;
+		rq.adr.channel = 0;
+		err = dev->D.ctrl(&dev->D, OPEN_CHANNEL, &rq);
+		printk(KERN_DEBUG "%s: ret 1 %d\n", __func__, err);
+		if (err)
+			break;
+		rq.protocol = protocol;
+		rq.adr = *adr;
+		rq.ch = ch;
+		err = dev->teimgr->ctrl(dev->teimgr, OPEN_CHANNEL, &rq);
+		printk(KERN_DEBUG "%s: ret 2 %d\n", __func__, err);
+		if (!err) {
+			if ((protocol == ISDN_P_LAPD_NT) && !rq.ch)
+				break;
+			add_layer2(rq.ch, dev->D.st);
+			rq.ch->recv = mISDN_queue_message;
+			rq.ch->peer = &dev->D.st->own;
+			rq.ch->ctrl(rq.ch, OPEN_CHANNEL, NULL); /* can't fail */
+		}
+		break;
+	default:
+		err = -EPROTONOSUPPORT;
+	}
+	return err;
+}
+
+void
+delete_channel(struct mISDNchannel *ch)
+{
+	struct mISDN_sock	*msk = container_of(ch, struct mISDN_sock, ch);
+	struct mISDNchannel	*pch;
+
+	if (!ch->st) {
+		printk(KERN_WARNING "%s: no stack\n", __func__);
+		return;
+	}
+	if (*debug & DEBUG_CORE_FUNC)
+		printk(KERN_DEBUG "%s: st(%s) protocol(%x)\n", __func__,
+		    ch->st->dev->name, ch->protocol);
+	if (ch->protocol >= ISDN_P_B_START) {
+		if (ch->peer) {
+			ch->peer->ctrl(ch->peer, CLOSE_CHANNEL, NULL);
+			ch->peer = NULL;
+		}
+		return;
+	}
+	switch (ch->protocol) {
+	case ISDN_P_NT_S0:
+	case ISDN_P_TE_S0:
+	case ISDN_P_NT_E1:
+	case ISDN_P_TE_E1:
+		write_lock_bh(&ch->st->l1sock.lock);
+		sk_del_node_init(&msk->sk);
+		write_unlock_bh(&ch->st->l1sock.lock);
+		ch->st->dev->D.ctrl(&ch->st->dev->D, CLOSE_CHANNEL, NULL);
+		break;
+	case ISDN_P_LAPD_TE:
+		pch = get_channel4id(ch->st, ch->nr);
+		if (pch) {
+			mutex_lock(&ch->st->lmutex);
+			list_del(&pch->list);
+			mutex_unlock(&ch->st->lmutex);
+			pch->ctrl(pch, CLOSE_CHANNEL, NULL);
+			pch = ch->st->dev->teimgr;
+			pch->ctrl(pch, CLOSE_CHANNEL, NULL);
+		} else
+			printk(KERN_WARNING "%s: no l2 channel\n",
+			    __func__);
+		break;
+	case ISDN_P_LAPD_NT:
+		pch = ch->st->dev->teimgr;
+		if (pch) {
+			pch->ctrl(pch, CLOSE_CHANNEL, NULL);
+		} else
+			printk(KERN_WARNING "%s: no l2 channel\n",
+			    __func__);
+		break;
+	default:
+		break;
+	}
+	return;
+}
+
+void
+delete_stack(struct mISDNdevice *dev)
+{
+	struct mISDNstack	*st = dev->D.st;
+	DECLARE_COMPLETION_ONSTACK(done);
+
+	if (*debug & DEBUG_CORE_FUNC)
+		printk(KERN_DEBUG "%s: st(%s)\n", __func__,
+		    st->dev->name);
+	if (dev->teimgr)
+		delete_teimanager(dev->teimgr);
+	if (st->thread) {
+		if (st->notify) {
+			printk(KERN_WARNING "%s: notifier in use\n",
+			    __func__);
+				complete(st->notify);
+		}
+		st->notify = &done;
+		test_and_set_bit(mISDN_STACK_ABORT, &st->status);
+		test_and_set_bit(mISDN_STACK_WAKEUP, &st->status);
+		wake_up_interruptible(&st->workq);
+		wait_for_completion(&done);
+	}
+	if (!list_empty(&st->layer2))
+		printk(KERN_WARNING "%s: layer2 list not empty\n",
+		    __func__);
+	if (!hlist_empty(&st->l1sock.head))
+		printk(KERN_WARNING "%s: layer1 list not empty\n",
+		    __func__);
+	kfree(st);
+}
+
+void
+mISDN_initstack(u_int *dp)
+{
+	debug = dp;
+}
diff --git a/drivers/isdn/mISDN/tei.c b/drivers/isdn/mISDN/tei.c
new file mode 100644
index 000000000000..56a76a0ffddd
--- /dev/null
+++ b/drivers/isdn/mISDN/tei.c
@@ -0,0 +1,1340 @@
+/*
+ *
+ * Author	Karsten Keil <kkeil@novell.com>
+ *
+ * Copyright 2008  by Karsten Keil <kkeil@novell.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+#include "layer2.h"
+#include <linux/random.h>
+#include "core.h"
+
+#define ID_REQUEST	1
+#define ID_ASSIGNED	2
+#define ID_DENIED	3
+#define ID_CHK_REQ	4
+#define ID_CHK_RES	5
+#define ID_REMOVE	6
+#define ID_VERIFY	7
+
+#define TEI_ENTITY_ID	0xf
+
+#define MGR_PH_ACTIVE	16
+#define MGR_PH_NOTREADY	17
+
+#define DATIMER_VAL	10000
+
+static 	u_int	*debug;
+
+static struct Fsm deactfsm = {NULL, 0, 0, NULL, NULL};
+static struct Fsm teifsmu = {NULL, 0, 0, NULL, NULL};
+static struct Fsm teifsmn = {NULL, 0, 0, NULL, NULL};
+
+enum {
+	ST_L1_DEACT,
+	ST_L1_DEACT_PENDING,
+	ST_L1_ACTIV,
+};
+#define DEACT_STATE_COUNT (ST_L1_ACTIV+1)
+
+static char *strDeactState[] =
+{
+	"ST_L1_DEACT",
+	"ST_L1_DEACT_PENDING",
+	"ST_L1_ACTIV",
+};
+
+enum {
+	EV_ACTIVATE,
+	EV_ACTIVATE_IND,
+	EV_DEACTIVATE,
+	EV_DEACTIVATE_IND,
+	EV_UI,
+	EV_DATIMER,
+};
+
+#define DEACT_EVENT_COUNT (EV_DATIMER+1)
+
+static char *strDeactEvent[] =
+{
+	"EV_ACTIVATE",
+	"EV_ACTIVATE_IND",
+	"EV_DEACTIVATE",
+	"EV_DEACTIVATE_IND",
+	"EV_UI",
+	"EV_DATIMER",
+};
+
+static void
+da_debug(struct FsmInst *fi, char *fmt, ...)
+{
+	struct manager	*mgr = fi->userdata;
+	va_list va;
+
+	if (!(*debug & DEBUG_L2_TEIFSM))
+		return;
+	va_start(va, fmt);
+	printk(KERN_DEBUG "mgr(%d): ", mgr->ch.st->dev->id);
+	vprintk(fmt, va);
+	printk("\n");
+	va_end(va);
+}
+
+static void
+da_activate(struct FsmInst *fi, int event, void *arg)
+{
+	struct manager	*mgr = fi->userdata;
+
+	if (fi->state == ST_L1_DEACT_PENDING)
+		mISDN_FsmDelTimer(&mgr->datimer, 1);
+	mISDN_FsmChangeState(fi, ST_L1_ACTIV);
+}
+
+static void
+da_deactivate_ind(struct FsmInst *fi, int event, void *arg)
+{
+	mISDN_FsmChangeState(fi, ST_L1_DEACT);
+}
+
+static void
+da_deactivate(struct FsmInst *fi, int event, void *arg)
+{
+	struct manager	*mgr = fi->userdata;
+	struct layer2	*l2;
+	u_long		flags;
+
+	read_lock_irqsave(&mgr->lock, flags);
+	list_for_each_entry(l2, &mgr->layer2, list) {
+		if (l2->l2m.state > ST_L2_4) {
+			/* have still activ TEI */
+			read_unlock_irqrestore(&mgr->lock, flags);
+			return;
+		}
+	}
+	read_unlock_irqrestore(&mgr->lock, flags);
+	/* All TEI are inactiv */
+	mISDN_FsmAddTimer(&mgr->datimer, DATIMER_VAL, EV_DATIMER, NULL, 1);
+	mISDN_FsmChangeState(fi, ST_L1_DEACT_PENDING);
+}
+
+static void
+da_ui(struct FsmInst *fi, int event, void *arg)
+{
+	struct manager	*mgr = fi->userdata;
+
+	/* restart da timer */
+	mISDN_FsmDelTimer(&mgr->datimer, 2);
+	mISDN_FsmAddTimer(&mgr->datimer, DATIMER_VAL, EV_DATIMER, NULL, 2);
+
+}
+
+static void
+da_timer(struct FsmInst *fi, int event, void *arg)
+{
+	struct manager	*mgr = fi->userdata;
+	struct layer2	*l2;
+	u_long		flags;
+
+	/* check again */
+	read_lock_irqsave(&mgr->lock, flags);
+	list_for_each_entry(l2, &mgr->layer2, list) {
+		if (l2->l2m.state > ST_L2_4) {
+			/* have still activ TEI */
+			read_unlock_irqrestore(&mgr->lock, flags);
+			mISDN_FsmChangeState(fi, ST_L1_ACTIV);
+			return;
+		}
+	}
+	read_unlock_irqrestore(&mgr->lock, flags);
+	/* All TEI are inactiv */
+	mISDN_FsmChangeState(fi, ST_L1_DEACT);
+	_queue_data(&mgr->ch, PH_DEACTIVATE_REQ, MISDN_ID_ANY, 0, NULL,
+	    GFP_ATOMIC);
+}
+
+static struct FsmNode DeactFnList[] =
+{
+	{ST_L1_DEACT, EV_ACTIVATE_IND, da_activate},
+	{ST_L1_ACTIV, EV_DEACTIVATE_IND, da_deactivate_ind},
+	{ST_L1_ACTIV, EV_DEACTIVATE, da_deactivate},
+	{ST_L1_DEACT_PENDING, EV_ACTIVATE, da_activate},
+	{ST_L1_DEACT_PENDING, EV_UI, da_ui},
+	{ST_L1_DEACT_PENDING, EV_DATIMER, da_timer},
+};
+
+enum {
+	ST_TEI_NOP,
+	ST_TEI_IDREQ,
+	ST_TEI_IDVERIFY,
+};
+
+#define TEI_STATE_COUNT (ST_TEI_IDVERIFY+1)
+
+static char *strTeiState[] =
+{
+	"ST_TEI_NOP",
+	"ST_TEI_IDREQ",
+	"ST_TEI_IDVERIFY",
+};
+
+enum {
+	EV_IDREQ,
+	EV_ASSIGN,
+	EV_ASSIGN_REQ,
+	EV_DENIED,
+	EV_CHKREQ,
+	EV_CHKRESP,
+	EV_REMOVE,
+	EV_VERIFY,
+	EV_TIMER,
+};
+
+#define TEI_EVENT_COUNT (EV_TIMER+1)
+
+static char *strTeiEvent[] =
+{
+	"EV_IDREQ",
+	"EV_ASSIGN",
+	"EV_ASSIGN_REQ",
+	"EV_DENIED",
+	"EV_CHKREQ",
+	"EV_CHKRESP",
+	"EV_REMOVE",
+	"EV_VERIFY",
+	"EV_TIMER",
+};
+
+static void
+tei_debug(struct FsmInst *fi, char *fmt, ...)
+{
+	struct teimgr	*tm = fi->userdata;
+	va_list va;
+
+	if (!(*debug & DEBUG_L2_TEIFSM))
+		return;
+	va_start(va, fmt);
+	printk(KERN_DEBUG "tei(%d): ", tm->l2->tei);
+	vprintk(fmt, va);
+	printk("\n");
+	va_end(va);
+}
+
+
+
+static int
+get_free_id(struct manager *mgr)
+{
+	u64		ids = 0;
+	int		i;
+	struct layer2	*l2;
+
+	list_for_each_entry(l2, &mgr->layer2, list) {
+		if (l2->ch.nr > 63) {
+			printk(KERN_WARNING
+			    "%s: more as 63 layer2 for one device\n",
+			    __func__);
+			return -EBUSY;
+		}
+		test_and_set_bit(l2->ch.nr, (u_long *)&ids);
+	}
+	for (i = 1; i < 64; i++)
+		if (!test_bit(i, (u_long *)&ids))
+			return i;
+	printk(KERN_WARNING "%s: more as 63 layer2 for one device\n",
+	    __func__);
+	return -EBUSY;
+}
+
+static int
+get_free_tei(struct manager *mgr)
+{
+	u64		ids = 0;
+	int		i;
+	struct layer2	*l2;
+
+	list_for_each_entry(l2, &mgr->layer2, list) {
+		if (l2->ch.nr == 0)
+			continue;
+		if ((l2->ch.addr & 0xff) != 0)
+			continue;
+		i = l2->ch.addr >> 8;
+		if (i < 64)
+			continue;
+		i -= 64;
+
+		test_and_set_bit(i, (u_long *)&ids);
+	}
+	for (i = 0; i < 64; i++)
+		if (!test_bit(i, (u_long *)&ids))
+			return i + 64;
+	printk(KERN_WARNING "%s: more as 63 dynamic tei for one device\n",
+	    __func__);
+	return -1;
+}
+
+static void
+teiup_create(struct manager *mgr, u_int prim, int len, void *arg)
+{
+	struct sk_buff	*skb;
+	struct mISDNhead *hh;
+	int		err;
+
+	skb = mI_alloc_skb(len, GFP_ATOMIC);
+	if (!skb)
+		return;
+	hh = mISDN_HEAD_P(skb);
+	hh->prim = prim;
+	hh->id = (mgr->ch.nr << 16) | mgr->ch.addr;
+	if (len)
+		memcpy(skb_put(skb, len), arg, len);
+	err = mgr->up->send(mgr->up, skb);
+	if (err) {
+		printk(KERN_WARNING "%s: err=%d\n", __func__, err);
+		dev_kfree_skb(skb);
+	}
+}
+
+static u_int
+new_id(struct manager *mgr)
+{
+	u_int	id;
+
+	id = mgr->nextid++;
+	if (id == 0x7fff)
+		mgr->nextid = 1;
+	id <<= 16;
+	id |= GROUP_TEI << 8;
+	id |= TEI_SAPI;
+	return id;
+}
+
+static void
+do_send(struct manager *mgr)
+{
+	if (!test_bit(MGR_PH_ACTIVE, &mgr->options))
+		return;
+
+	if (!test_and_set_bit(MGR_PH_NOTREADY, &mgr->options)) {
+		struct sk_buff	*skb = skb_dequeue(&mgr->sendq);
+
+		if (!skb) {
+			test_and_clear_bit(MGR_PH_NOTREADY, &mgr->options);
+			return;
+		}
+		mgr->lastid = mISDN_HEAD_ID(skb);
+		mISDN_FsmEvent(&mgr->deact, EV_UI, NULL);
+		if (mgr->ch.recv(mgr->ch.peer, skb)) {
+			dev_kfree_skb(skb);
+			test_and_clear_bit(MGR_PH_NOTREADY, &mgr->options);
+			mgr->lastid = MISDN_ID_NONE;
+		}
+	}
+}
+
+static void
+do_ack(struct manager *mgr, u_int id)
+{
+	if (test_bit(MGR_PH_NOTREADY, &mgr->options)) {
+		if (id == mgr->lastid) {
+			if (test_bit(MGR_PH_ACTIVE, &mgr->options)) {
+				struct sk_buff	*skb;
+
+				skb = skb_dequeue(&mgr->sendq);
+				if (skb) {
+					mgr->lastid = mISDN_HEAD_ID(skb);
+					if (!mgr->ch.recv(mgr->ch.peer, skb))
+						return;
+					dev_kfree_skb(skb);
+				}
+			}
+			mgr->lastid = MISDN_ID_NONE;
+			test_and_clear_bit(MGR_PH_NOTREADY, &mgr->options);
+		}
+	}
+}
+
+static void
+mgr_send_down(struct manager *mgr, struct sk_buff *skb)
+{
+	skb_queue_tail(&mgr->sendq, skb);
+	if (!test_bit(MGR_PH_ACTIVE, &mgr->options)) {
+		_queue_data(&mgr->ch, PH_ACTIVATE_REQ, MISDN_ID_ANY, 0,
+		    NULL, GFP_KERNEL);
+	} else {
+		do_send(mgr);
+	}
+}
+
+static int
+dl_unit_data(struct manager *mgr, struct sk_buff *skb)
+{
+	if (!test_bit(MGR_OPT_NETWORK, &mgr->options)) /* only net send UI */
+		return -EINVAL;
+	if (!test_bit(MGR_PH_ACTIVE, &mgr->options))
+		_queue_data(&mgr->ch, PH_ACTIVATE_REQ, MISDN_ID_ANY, 0,
+		    NULL, GFP_KERNEL);
+	skb_push(skb, 3);
+	skb->data[0] = 0x02; /* SAPI 0 C/R = 1 */
+	skb->data[1] = 0xff; /* TEI 127 */
+	skb->data[2] = UI;   /* UI frame */
+	mISDN_HEAD_PRIM(skb) = PH_DATA_REQ;
+	mISDN_HEAD_ID(skb) = new_id(mgr);
+	skb_queue_tail(&mgr->sendq, skb);
+	do_send(mgr);
+	return 0;
+}
+
+unsigned int
+random_ri(void)
+{
+	u16 x;
+
+	get_random_bytes(&x, sizeof(x));
+	return x;
+}
+
+static struct layer2 *
+findtei(struct manager *mgr, int tei)
+{
+	struct layer2	*l2;
+	u_long		flags;
+
+	read_lock_irqsave(&mgr->lock, flags);
+	list_for_each_entry(l2, &mgr->layer2, list) {
+		if ((l2->sapi == 0) && (l2->tei > 0) &&
+		    (l2->tei != GROUP_TEI) && (l2->tei == tei))
+			goto done;
+	}
+	l2 = NULL;
+done:
+	read_unlock_irqrestore(&mgr->lock, flags);
+	return l2;
+}
+
+static void
+put_tei_msg(struct manager *mgr, u_char m_id, unsigned int ri, u_char tei)
+{
+	struct sk_buff *skb;
+	u_char bp[8];
+
+	bp[0] = (TEI_SAPI << 2);
+	if (test_bit(MGR_OPT_NETWORK, &mgr->options))
+		bp[0] |= 2; /* CR:=1 for net command */
+	bp[1] = (GROUP_TEI << 1) | 0x1;
+	bp[2] = UI;
+	bp[3] = TEI_ENTITY_ID;
+	bp[4] = ri >> 8;
+	bp[5] = ri & 0xff;
+	bp[6] = m_id;
+	bp[7] = (tei << 1) | 1;
+	skb = _alloc_mISDN_skb(PH_DATA_REQ, new_id(mgr),
+	    8, bp, GFP_ATOMIC);
+	if (!skb) {
+		printk(KERN_WARNING "%s: no skb for tei msg\n", __func__);
+		return;
+	}
+	mgr_send_down(mgr, skb);
+}
+
+static void
+tei_id_request(struct FsmInst *fi, int event, void *arg)
+{
+	struct teimgr *tm = fi->userdata;
+
+	if (tm->l2->tei != GROUP_TEI) {
+		tm->tei_m.printdebug(&tm->tei_m,
+			"assign request for allready assigned tei %d",
+			tm->l2->tei);
+		return;
+	}
+	tm->ri = random_ri();
+	if (*debug & DEBUG_L2_TEI)
+		tm->tei_m.printdebug(&tm->tei_m,
+			"assign request ri %d", tm->ri);
+	put_tei_msg(tm->mgr, ID_REQUEST, tm->ri, GROUP_TEI);
+	mISDN_FsmChangeState(fi, ST_TEI_IDREQ);
+	mISDN_FsmAddTimer(&tm->timer, tm->tval, EV_TIMER, NULL, 1);
+	tm->nval = 3;
+}
+
+static void
+tei_id_assign(struct FsmInst *fi, int event, void *arg)
+{
+	struct teimgr	*tm = fi->userdata;
+	struct layer2	*l2;
+	u_char *dp = arg;
+	int ri, tei;
+
+	ri = ((unsigned int) *dp++ << 8);
+	ri += *dp++;
+	dp++;
+	tei = *dp >> 1;
+	if (*debug & DEBUG_L2_TEI)
+		tm->tei_m.printdebug(fi, "identity assign ri %d tei %d",
+			ri, tei);
+	l2 = findtei(tm->mgr, tei);
+	if (l2) {	/* same tei is in use */
+		if (ri != l2->tm->ri) {
+			tm->tei_m.printdebug(fi,
+				"possible duplicate assignment tei %d", tei);
+			tei_l2(l2, MDL_ERROR_RSP, 0);
+		}
+	} else if (ri == tm->ri) {
+		mISDN_FsmDelTimer(&tm->timer, 1);
+		mISDN_FsmChangeState(fi, ST_TEI_NOP);
+		tei_l2(tm->l2, MDL_ASSIGN_REQ, tei);
+	}
+}
+
+static void
+tei_id_test_dup(struct FsmInst *fi, int event, void *arg)
+{
+	struct teimgr	*tm = fi->userdata;
+	struct layer2	*l2;
+	u_char *dp = arg;
+	int tei, ri;
+
+	ri = ((unsigned int) *dp++ << 8);
+	ri += *dp++;
+	dp++;
+	tei = *dp >> 1;
+	if (*debug & DEBUG_L2_TEI)
+		tm->tei_m.printdebug(fi, "foreign identity assign ri %d tei %d",
+			ri, tei);
+	l2 = findtei(tm->mgr, tei);
+	if (l2) {	/* same tei is in use */
+		if (ri != l2->tm->ri) {	/* and it wasn't our request */
+			tm->tei_m.printdebug(fi,
+				"possible duplicate assignment tei %d", tei);
+			mISDN_FsmEvent(&l2->tm->tei_m, EV_VERIFY, NULL);
+		}
+	}
+}
+
+static void
+tei_id_denied(struct FsmInst *fi, int event, void *arg)
+{
+	struct teimgr *tm = fi->userdata;
+	u_char *dp = arg;
+	int ri, tei;
+
+	ri = ((unsigned int) *dp++ << 8);
+	ri += *dp++;
+	dp++;
+	tei = *dp >> 1;
+	if (*debug & DEBUG_L2_TEI)
+		tm->tei_m.printdebug(fi, "identity denied ri %d tei %d",
+			ri, tei);
+}
+
+static void
+tei_id_chk_req(struct FsmInst *fi, int event, void *arg)
+{
+	struct teimgr *tm = fi->userdata;
+	u_char *dp = arg;
+	int tei;
+
+	tei = *(dp+3) >> 1;
+	if (*debug & DEBUG_L2_TEI)
+		tm->tei_m.printdebug(fi, "identity check req tei %d", tei);
+	if ((tm->l2->tei != GROUP_TEI) && ((tei == GROUP_TEI) ||
+	    (tei == tm->l2->tei))) {
+		mISDN_FsmDelTimer(&tm->timer, 4);
+		mISDN_FsmChangeState(&tm->tei_m, ST_TEI_NOP);
+		put_tei_msg(tm->mgr, ID_CHK_RES, random_ri(), tm->l2->tei);
+	}
+}
+
+static void
+tei_id_remove(struct FsmInst *fi, int event, void *arg)
+{
+	struct teimgr *tm = fi->userdata;
+	u_char *dp = arg;
+	int tei;
+
+	tei = *(dp+3) >> 1;
+	if (*debug & DEBUG_L2_TEI)
+		tm->tei_m.printdebug(fi, "identity remove tei %d", tei);
+	if ((tm->l2->tei != GROUP_TEI) &&
+	    ((tei == GROUP_TEI) || (tei == tm->l2->tei))) {
+		mISDN_FsmDelTimer(&tm->timer, 5);
+		mISDN_FsmChangeState(&tm->tei_m, ST_TEI_NOP);
+		tei_l2(tm->l2, MDL_REMOVE_REQ, 0);
+	}
+}
+
+static void
+tei_id_verify(struct FsmInst *fi, int event, void *arg)
+{
+	struct teimgr *tm = fi->userdata;
+
+	if (*debug & DEBUG_L2_TEI)
+		tm->tei_m.printdebug(fi, "id verify request for tei %d",
+			tm->l2->tei);
+	put_tei_msg(tm->mgr, ID_VERIFY, 0, tm->l2->tei);
+	mISDN_FsmChangeState(&tm->tei_m, ST_TEI_IDVERIFY);
+	mISDN_FsmAddTimer(&tm->timer, tm->tval, EV_TIMER, NULL, 2);
+	tm->nval = 2;
+}
+
+static void
+tei_id_req_tout(struct FsmInst *fi, int event, void *arg)
+{
+	struct teimgr *tm = fi->userdata;
+
+	if (--tm->nval) {
+		tm->ri = random_ri();
+		if (*debug & DEBUG_L2_TEI)
+			tm->tei_m.printdebug(fi, "assign req(%d) ri %d",
+				4 - tm->nval, tm->ri);
+		put_tei_msg(tm->mgr, ID_REQUEST, tm->ri, GROUP_TEI);
+		mISDN_FsmAddTimer(&tm->timer, tm->tval, EV_TIMER, NULL, 3);
+	} else {
+		tm->tei_m.printdebug(fi, "assign req failed");
+		tei_l2(tm->l2, MDL_ERROR_RSP, 0);
+		mISDN_FsmChangeState(fi, ST_TEI_NOP);
+	}
+}
+
+static void
+tei_id_ver_tout(struct FsmInst *fi, int event, void *arg)
+{
+	struct teimgr *tm = fi->userdata;
+
+	if (--tm->nval) {
+		if (*debug & DEBUG_L2_TEI)
+			tm->tei_m.printdebug(fi,
+				"id verify req(%d) for tei %d",
+				3 - tm->nval, tm->l2->tei);
+		put_tei_msg(tm->mgr, ID_VERIFY, 0, tm->l2->tei);
+		mISDN_FsmAddTimer(&tm->timer, tm->tval, EV_TIMER, NULL, 4);
+	} else {
+		tm->tei_m.printdebug(fi, "verify req for tei %d failed",
+			tm->l2->tei);
+		tei_l2(tm->l2, MDL_REMOVE_REQ, 0);
+		mISDN_FsmChangeState(fi, ST_TEI_NOP);
+	}
+}
+
+static struct FsmNode TeiFnListUser[] =
+{
+	{ST_TEI_NOP, EV_IDREQ, tei_id_request},
+	{ST_TEI_NOP, EV_ASSIGN, tei_id_test_dup},
+	{ST_TEI_NOP, EV_VERIFY, tei_id_verify},
+	{ST_TEI_NOP, EV_REMOVE, tei_id_remove},
+	{ST_TEI_NOP, EV_CHKREQ, tei_id_chk_req},
+	{ST_TEI_IDREQ, EV_TIMER, tei_id_req_tout},
+	{ST_TEI_IDREQ, EV_ASSIGN, tei_id_assign},
+	{ST_TEI_IDREQ, EV_DENIED, tei_id_denied},
+	{ST_TEI_IDVERIFY, EV_TIMER, tei_id_ver_tout},
+	{ST_TEI_IDVERIFY, EV_REMOVE, tei_id_remove},
+	{ST_TEI_IDVERIFY, EV_CHKREQ, tei_id_chk_req},
+};
+
+static void
+tei_l2remove(struct layer2 *l2)
+{
+	put_tei_msg(l2->tm->mgr, ID_REMOVE, 0, l2->tei);
+	tei_l2(l2, MDL_REMOVE_REQ, 0);
+	list_del(&l2->ch.list);
+	l2->ch.ctrl(&l2->ch, CLOSE_CHANNEL, NULL);
+}
+
+static void
+tei_assign_req(struct FsmInst *fi, int event, void *arg)
+{
+	struct teimgr *tm = fi->userdata;
+	u_char *dp = arg;
+
+	if (tm->l2->tei == GROUP_TEI) {
+		tm->tei_m.printdebug(&tm->tei_m,
+			"net tei assign request without tei");
+		return;
+	}
+	tm->ri = ((unsigned int) *dp++ << 8);
+	tm->ri += *dp++;
+	if (*debug & DEBUG_L2_TEI)
+		tm->tei_m.printdebug(&tm->tei_m,
+			"net assign request ri %d teim %d", tm->ri, *dp);
+	put_tei_msg(tm->mgr, ID_ASSIGNED, tm->ri, tm->l2->tei);
+	mISDN_FsmChangeState(fi, ST_TEI_NOP);
+}
+
+static void
+tei_id_chk_req_net(struct FsmInst *fi, int event, void *arg)
+{
+	struct teimgr	*tm = fi->userdata;
+
+	if (*debug & DEBUG_L2_TEI)
+		tm->tei_m.printdebug(fi, "id check request for tei %d",
+		    tm->l2->tei);
+	tm->rcnt = 0;
+	put_tei_msg(tm->mgr, ID_CHK_REQ, 0, tm->l2->tei);
+	mISDN_FsmChangeState(&tm->tei_m, ST_TEI_IDVERIFY);
+	mISDN_FsmAddTimer(&tm->timer, tm->tval, EV_TIMER, NULL, 2);
+	tm->nval = 2;
+}
+
+static void
+tei_id_chk_resp(struct FsmInst *fi, int event, void *arg)
+{
+	struct teimgr *tm = fi->userdata;
+	u_char *dp = arg;
+	int tei;
+
+	tei = dp[3] >> 1;
+	if (*debug & DEBUG_L2_TEI)
+		tm->tei_m.printdebug(fi, "identity check resp tei %d", tei);
+	if (tei == tm->l2->tei)
+		tm->rcnt++;
+}
+
+static void
+tei_id_verify_net(struct FsmInst *fi, int event, void *arg)
+{
+	struct teimgr *tm = fi->userdata;
+	u_char *dp = arg;
+	int tei;
+
+	tei = dp[3] >> 1;
+	if (*debug & DEBUG_L2_TEI)
+		tm->tei_m.printdebug(fi, "identity verify req tei %d/%d",
+		    tei, tm->l2->tei);
+	if (tei == tm->l2->tei)
+		tei_id_chk_req_net(fi, event, arg);
+}
+
+static void
+tei_id_ver_tout_net(struct FsmInst *fi, int event, void *arg)
+{
+	struct teimgr *tm = fi->userdata;
+
+	if (tm->rcnt == 1) {
+		if (*debug & DEBUG_L2_TEI)
+			tm->tei_m.printdebug(fi,
+			    "check req for tei %d sucessful\n", tm->l2->tei);
+		mISDN_FsmChangeState(fi, ST_TEI_NOP);
+	} else if (tm->rcnt > 1) {
+		/* duplicate assignment; remove */
+		tei_l2remove(tm->l2);
+	} else if (--tm->nval) {
+		if (*debug & DEBUG_L2_TEI)
+			tm->tei_m.printdebug(fi,
+				"id check req(%d) for tei %d",
+				3 - tm->nval, tm->l2->tei);
+		put_tei_msg(tm->mgr, ID_CHK_REQ, 0, tm->l2->tei);
+		mISDN_FsmAddTimer(&tm->timer, tm->tval, EV_TIMER, NULL, 4);
+	} else {
+		tm->tei_m.printdebug(fi, "check req for tei %d failed",
+			tm->l2->tei);
+		mISDN_FsmChangeState(fi, ST_TEI_NOP);
+		tei_l2remove(tm->l2);
+	}
+}
+
+static struct FsmNode TeiFnListNet[] =
+{
+	{ST_TEI_NOP, EV_ASSIGN_REQ, tei_assign_req},
+	{ST_TEI_NOP, EV_VERIFY, tei_id_verify_net},
+	{ST_TEI_NOP, EV_CHKREQ, tei_id_chk_req_net},
+	{ST_TEI_IDVERIFY, EV_TIMER, tei_id_ver_tout_net},
+	{ST_TEI_IDVERIFY, EV_CHKRESP, tei_id_chk_resp},
+};
+
+static void
+tei_ph_data_ind(struct teimgr *tm, u_int mt, u_char *dp, int len)
+{
+	if (test_bit(FLG_FIXED_TEI, &tm->l2->flag))
+		return;
+	if (*debug & DEBUG_L2_TEI)
+		tm->tei_m.printdebug(&tm->tei_m, "tei handler mt %x", mt);
+	if (mt == ID_ASSIGNED)
+		mISDN_FsmEvent(&tm->tei_m, EV_ASSIGN, dp);
+	else if (mt == ID_DENIED)
+		mISDN_FsmEvent(&tm->tei_m, EV_DENIED, dp);
+	else if (mt == ID_CHK_REQ)
+		mISDN_FsmEvent(&tm->tei_m, EV_CHKREQ, dp);
+	else if (mt == ID_REMOVE)
+		mISDN_FsmEvent(&tm->tei_m, EV_REMOVE, dp);
+	else if (mt == ID_VERIFY)
+		mISDN_FsmEvent(&tm->tei_m, EV_VERIFY, dp);
+	else if (mt == ID_CHK_RES)
+		mISDN_FsmEvent(&tm->tei_m, EV_CHKRESP, dp);
+}
+
+static struct layer2 *
+create_new_tei(struct manager *mgr, int tei)
+{
+	u_long		opt = 0;
+	u_long		flags;
+	int		id;
+	struct layer2	*l2;
+
+	if (!mgr->up)
+		return NULL;
+	if (tei < 64)
+		test_and_set_bit(OPTION_L2_FIXEDTEI, &opt);
+	if (mgr->ch.st->dev->Dprotocols
+	  & ((1 << ISDN_P_TE_E1) | (1 << ISDN_P_NT_E1)))
+		test_and_set_bit(OPTION_L2_PMX, &opt);
+	l2 = create_l2(mgr->up, ISDN_P_LAPD_NT, (u_int)opt, (u_long)tei);
+	if (!l2) {
+		printk(KERN_WARNING "%s:no memory for layer2\n", __func__);
+		return NULL;
+	}
+	l2->tm = kzalloc(sizeof(struct teimgr), GFP_KERNEL);
+	if (!l2->tm) {
+		kfree(l2);
+		printk(KERN_WARNING "%s:no memory for teimgr\n", __func__);
+		return NULL;
+	}
+	l2->tm->mgr = mgr;
+	l2->tm->l2 = l2;
+	l2->tm->tei_m.debug = *debug & DEBUG_L2_TEIFSM;
+	l2->tm->tei_m.userdata = l2->tm;
+	l2->tm->tei_m.printdebug = tei_debug;
+	l2->tm->tei_m.fsm = &teifsmn;
+	l2->tm->tei_m.state = ST_TEI_NOP;
+	l2->tm->tval = 2000; /* T202  2 sec */
+	mISDN_FsmInitTimer(&l2->tm->tei_m, &l2->tm->timer);
+	write_lock_irqsave(&mgr->lock, flags);
+	id = get_free_id(mgr);
+	list_add_tail(&l2->list, &mgr->layer2);
+	write_unlock_irqrestore(&mgr->lock, flags);
+	if (id < 0) {
+		l2->ch.ctrl(&l2->ch, CLOSE_CHANNEL, NULL);
+		printk(KERN_WARNING "%s:no free id\n", __func__);
+		return NULL;
+	} else {
+		l2->ch.nr = id;
+		__add_layer2(&l2->ch, mgr->ch.st);
+		l2->ch.recv = mgr->ch.recv;
+		l2->ch.peer = mgr->ch.peer;
+		l2->ch.ctrl(&l2->ch, OPEN_CHANNEL, NULL);
+	}
+	return l2;
+}
+
+static void
+new_tei_req(struct manager *mgr, u_char *dp)
+{
+	int		tei, ri;
+	struct layer2	*l2;
+
+	ri = dp[0] << 8;
+	ri += dp[1];
+	if (!mgr->up)
+		goto denied;
+	tei = get_free_tei(mgr);
+	if (tei < 0) {
+		printk(KERN_WARNING "%s:No free tei\n", __func__);
+		goto denied;
+	}
+	l2 = create_new_tei(mgr, tei);
+	if (!l2)
+		goto denied;
+	else
+		mISDN_FsmEvent(&l2->tm->tei_m, EV_ASSIGN_REQ, dp);
+	return;
+denied:
+	put_tei_msg(mgr, ID_DENIED, ri, GROUP_TEI);
+}
+
+static int
+ph_data_ind(struct manager *mgr, struct sk_buff *skb)
+{
+	int		ret = -EINVAL;
+	struct layer2	*l2;
+	u_long		flags;
+	u_char		mt;
+
+	if (skb->len < 8) {
+		if (*debug  & DEBUG_L2_TEI)
+			printk(KERN_DEBUG "%s: short mgr frame %d/8\n",
+			    __func__, skb->len);
+		goto done;
+	}
+	if (*debug  & DEBUG_L2_TEI)
+
+	if ((skb->data[0] >> 2) != TEI_SAPI) /* not for us */
+		goto done;
+	if (skb->data[0] & 1) /* EA0 formal error */
+		goto done;
+	if (!(skb->data[1] & 1)) /* EA1 formal error */
+		goto done;
+	if ((skb->data[1] >> 1) != GROUP_TEI) /* not for us */
+		goto done;
+	if ((skb->data[2] & 0xef) != UI) /* not UI */
+		goto done;
+	if (skb->data[3] != TEI_ENTITY_ID) /* not tei entity */
+		goto done;
+	mt = skb->data[6];
+	switch (mt) {
+	case ID_REQUEST:
+	case ID_CHK_RES:
+	case ID_VERIFY:
+		if (!test_bit(MGR_OPT_NETWORK, &mgr->options))
+			goto done;
+		break;
+	case ID_ASSIGNED:
+	case ID_DENIED:
+	case ID_CHK_REQ:
+	case ID_REMOVE:
+		if (test_bit(MGR_OPT_NETWORK, &mgr->options))
+			goto done;
+		break;
+	default:
+		goto done;
+	}
+	ret = 0;
+	if (mt == ID_REQUEST) {
+		new_tei_req(mgr, &skb->data[4]);
+		goto done;
+	}
+	read_lock_irqsave(&mgr->lock, flags);
+	list_for_each_entry(l2, &mgr->layer2, list) {
+		tei_ph_data_ind(l2->tm, mt, &skb->data[4], skb->len - 4);
+	}
+	read_unlock_irqrestore(&mgr->lock, flags);
+done:
+	return ret;
+}
+
+int
+l2_tei(struct layer2 *l2, u_int cmd, u_long arg)
+{
+	struct teimgr	*tm = l2->tm;
+
+	if (test_bit(FLG_FIXED_TEI, &l2->flag))
+		return 0;
+	if (*debug & DEBUG_L2_TEI)
+		printk(KERN_DEBUG "%s: cmd(%x)\n", __func__, cmd);
+	switch (cmd) {
+	case MDL_ASSIGN_IND:
+		mISDN_FsmEvent(&tm->tei_m, EV_IDREQ, NULL);
+		break;
+	case MDL_ERROR_IND:
+		if (test_bit(MGR_OPT_NETWORK, &tm->mgr->options))
+			mISDN_FsmEvent(&tm->tei_m, EV_CHKREQ, &l2->tei);
+		if (test_bit(MGR_OPT_USER, &tm->mgr->options))
+			mISDN_FsmEvent(&tm->tei_m, EV_VERIFY, NULL);
+		break;
+	case MDL_STATUS_UP_IND:
+		if (test_bit(MGR_OPT_NETWORK, &tm->mgr->options))
+			mISDN_FsmEvent(&tm->mgr->deact, EV_ACTIVATE, NULL);
+		break;
+	case MDL_STATUS_DOWN_IND:
+		if (test_bit(MGR_OPT_NETWORK, &tm->mgr->options))
+			mISDN_FsmEvent(&tm->mgr->deact, EV_DEACTIVATE, NULL);
+		break;
+	case MDL_STATUS_UI_IND:
+		if (test_bit(MGR_OPT_NETWORK, &tm->mgr->options))
+			mISDN_FsmEvent(&tm->mgr->deact, EV_UI, NULL);
+		break;
+	}
+	return 0;
+}
+
+void
+release_tei(struct layer2 *l2)
+{
+	struct teimgr	*tm = l2->tm;
+	u_long		flags;
+
+	mISDN_FsmDelTimer(&tm->timer, 1);
+	write_lock_irqsave(&tm->mgr->lock, flags);
+	list_del(&l2->list);
+	write_unlock_irqrestore(&tm->mgr->lock, flags);
+	l2->tm = NULL;
+	kfree(tm);
+}
+
+static int
+create_teimgr(struct manager *mgr, struct channel_req *crq)
+{
+	struct layer2	*l2;
+	u_long 		opt = 0;
+	u_long		flags;
+	int		id;
+
+	if (*debug & DEBUG_L2_TEI)
+		printk(KERN_DEBUG "%s: %s proto(%x) adr(%d %d %d %d)\n",
+			__func__, mgr->ch.st->dev->name, crq->protocol,
+			crq->adr.dev, crq->adr.channel, crq->adr.sapi,
+			crq->adr.tei);
+	if (crq->adr.sapi != 0) /* not supported yet */
+		return -EINVAL;
+	if (crq->adr.tei > GROUP_TEI)
+		return -EINVAL;
+	if (crq->adr.tei < 64)
+		test_and_set_bit(OPTION_L2_FIXEDTEI, &opt);
+	if (crq->adr.tei == 0)
+		test_and_set_bit(OPTION_L2_PTP, &opt);
+	if (test_bit(MGR_OPT_NETWORK, &mgr->options)) {
+		if (crq->protocol == ISDN_P_LAPD_TE)
+			return -EPROTONOSUPPORT;
+		if ((crq->adr.tei != 0) && (crq->adr.tei != 127))
+			return -EINVAL;
+		if (mgr->up) {
+			printk(KERN_WARNING
+			    "%s: only one network manager is allowed\n",
+			    __func__);
+			return -EBUSY;
+		}
+	} else if (test_bit(MGR_OPT_USER, &mgr->options)) {
+		if (crq->protocol == ISDN_P_LAPD_NT)
+			return -EPROTONOSUPPORT;
+		if ((crq->adr.tei >= 64) && (crq->adr.tei < GROUP_TEI))
+			return -EINVAL; /* dyn tei */
+	} else {
+		if (crq->protocol == ISDN_P_LAPD_NT)
+			test_and_set_bit(MGR_OPT_NETWORK, &mgr->options);
+		if (crq->protocol == ISDN_P_LAPD_TE)
+			test_and_set_bit(MGR_OPT_USER, &mgr->options);
+	}
+	if (mgr->ch.st->dev->Dprotocols
+	  & ((1 << ISDN_P_TE_E1) | (1 << ISDN_P_NT_E1)))
+		test_and_set_bit(OPTION_L2_PMX, &opt);
+	if ((crq->protocol == ISDN_P_LAPD_NT) && (crq->adr.tei == 127)) {
+		mgr->up = crq->ch;
+		id = DL_INFO_L2_CONNECT;
+		teiup_create(mgr, DL_INFORMATION_IND, sizeof(id), &id);
+		crq->ch = NULL;
+		if (!list_empty(&mgr->layer2)) {
+			read_lock_irqsave(&mgr->lock, flags);
+			list_for_each_entry(l2, &mgr->layer2, list) {
+				l2->up = mgr->up;
+				l2->ch.ctrl(&l2->ch, OPEN_CHANNEL, NULL);
+			}
+			read_unlock_irqrestore(&mgr->lock, flags);
+		}
+		return 0;
+	}
+	l2 = create_l2(crq->ch, crq->protocol, (u_int)opt,
+		(u_long)crq->adr.tei);
+	if (!l2)
+		return -ENOMEM;
+	l2->tm = kzalloc(sizeof(struct teimgr), GFP_KERNEL);
+	if (!l2->tm) {
+		kfree(l2);
+		printk(KERN_ERR "kmalloc teimgr failed\n");
+		return -ENOMEM;
+	}
+	l2->tm->mgr = mgr;
+	l2->tm->l2 = l2;
+	l2->tm->tei_m.debug = *debug & DEBUG_L2_TEIFSM;
+	l2->tm->tei_m.userdata = l2->tm;
+	l2->tm->tei_m.printdebug = tei_debug;
+	if (crq->protocol == ISDN_P_LAPD_TE) {
+		l2->tm->tei_m.fsm = &teifsmu;
+		l2->tm->tei_m.state = ST_TEI_NOP;
+		l2->tm->tval = 1000; /* T201  1 sec */
+	} else {
+		l2->tm->tei_m.fsm = &teifsmn;
+		l2->tm->tei_m.state = ST_TEI_NOP;
+		l2->tm->tval = 2000; /* T202  2 sec */
+	}
+	mISDN_FsmInitTimer(&l2->tm->tei_m, &l2->tm->timer);
+	write_lock_irqsave(&mgr->lock, flags);
+	id = get_free_id(mgr);
+	list_add_tail(&l2->list, &mgr->layer2);
+	write_unlock_irqrestore(&mgr->lock, flags);
+	if (id < 0) {
+		l2->ch.ctrl(&l2->ch, CLOSE_CHANNEL, NULL);
+	} else {
+		l2->ch.nr = id;
+		l2->up->nr = id;
+		crq->ch = &l2->ch;
+		id = 0;
+	}
+	return id;
+}
+
+static int
+mgr_send(struct mISDNchannel *ch, struct sk_buff *skb)
+{
+	struct manager	*mgr;
+	struct mISDNhead	*hh =  mISDN_HEAD_P(skb);
+	int			ret = -EINVAL;
+
+	mgr = container_of(ch, struct manager, ch);
+	if (*debug & DEBUG_L2_RECV)
+		printk(KERN_DEBUG "%s: prim(%x) id(%x)\n",
+		    __func__, hh->prim, hh->id);
+	switch (hh->prim) {
+	case PH_DATA_IND:
+		mISDN_FsmEvent(&mgr->deact, EV_UI, NULL);
+		ret = ph_data_ind(mgr, skb);
+		break;
+	case PH_DATA_CNF:
+		do_ack(mgr, hh->id);
+		ret = 0;
+		break;
+	case PH_ACTIVATE_IND:
+		test_and_set_bit(MGR_PH_ACTIVE, &mgr->options);
+		mISDN_FsmEvent(&mgr->deact, EV_ACTIVATE_IND, NULL);
+		do_send(mgr);
+		ret = 0;
+		break;
+	case PH_DEACTIVATE_IND:
+		test_and_clear_bit(MGR_PH_ACTIVE, &mgr->options);
+		mISDN_FsmEvent(&mgr->deact, EV_DEACTIVATE_IND, NULL);
+		ret = 0;
+		break;
+	case DL_UNITDATA_REQ:
+		return dl_unit_data(mgr, skb);
+	}
+	if (!ret)
+		dev_kfree_skb(skb);
+	return ret;
+}
+
+static int
+free_teimanager(struct manager *mgr)
+{
+	struct layer2	*l2, *nl2;
+
+	if (test_bit(MGR_OPT_NETWORK, &mgr->options)) {
+		/* not locked lock is taken in release tei */
+		mgr->up = NULL;
+		if (test_bit(OPTION_L2_CLEANUP, &mgr->options)) {
+			list_for_each_entry_safe(l2, nl2, &mgr->layer2, list) {
+				put_tei_msg(mgr, ID_REMOVE, 0, l2->tei);
+				mutex_lock(&mgr->ch.st->lmutex);
+				list_del(&l2->ch.list);
+				mutex_unlock(&mgr->ch.st->lmutex);
+				l2->ch.ctrl(&l2->ch, CLOSE_CHANNEL, NULL);
+			}
+			test_and_clear_bit(MGR_OPT_NETWORK, &mgr->options);
+		} else {
+			list_for_each_entry_safe(l2, nl2, &mgr->layer2, list) {
+				l2->up = NULL;
+			}
+		}
+	}
+	if (test_bit(MGR_OPT_USER, &mgr->options)) {
+		if (list_empty(&mgr->layer2))
+			test_and_clear_bit(MGR_OPT_USER, &mgr->options);
+	}
+	mgr->ch.st->dev->D.ctrl(&mgr->ch.st->dev->D, CLOSE_CHANNEL, NULL);
+	return 0;
+}
+
+static int
+ctrl_teimanager(struct manager *mgr, void *arg)
+{
+	/* currently we only have one option */
+	int	clean = *((int *)arg);
+
+	if (clean)
+		test_and_set_bit(OPTION_L2_CLEANUP, &mgr->options);
+	else
+		test_and_clear_bit(OPTION_L2_CLEANUP, &mgr->options);
+	return 0;
+}
+
+/* This function does create a L2 for fixed TEI in NT Mode */
+static int
+check_data(struct manager *mgr, struct sk_buff *skb)
+{
+	struct mISDNhead	*hh =  mISDN_HEAD_P(skb);
+	int			ret, tei;
+	struct layer2		*l2;
+
+	if (*debug & DEBUG_L2_CTRL)
+		printk(KERN_DEBUG "%s: prim(%x) id(%x)\n",
+		    __func__, hh->prim, hh->id);
+	if (test_bit(MGR_OPT_USER, &mgr->options))
+		return -ENOTCONN;
+	if (hh->prim != PH_DATA_IND)
+		return -ENOTCONN;
+	if (skb->len != 3)
+		return -ENOTCONN;
+	if (skb->data[0] != 0)
+		/* only SAPI 0 command */
+		return -ENOTCONN;
+	if (!(skb->data[1] & 1)) /* invalid EA1 */
+		return -EINVAL;
+	tei = skb->data[1] >> 0;
+	if (tei > 63) /* not a fixed tei */
+		return -ENOTCONN;
+	if ((skb->data[2] & ~0x10) != SABME)
+		return -ENOTCONN;
+	/* We got a SABME for a fixed TEI */
+	l2 = create_new_tei(mgr, tei);
+	if (!l2)
+		return -ENOMEM;
+	ret = l2->ch.send(&l2->ch, skb);
+	return ret;
+}
+
+void
+delete_teimanager(struct mISDNchannel *ch)
+{
+	struct manager	*mgr;
+	struct layer2	*l2, *nl2;
+
+	mgr = container_of(ch, struct manager, ch);
+	/* not locked lock is taken in release tei */
+	list_for_each_entry_safe(l2, nl2, &mgr->layer2, list) {
+		mutex_lock(&mgr->ch.st->lmutex);
+		list_del(&l2->ch.list);
+		mutex_unlock(&mgr->ch.st->lmutex);
+		l2->ch.ctrl(&l2->ch, CLOSE_CHANNEL, NULL);
+	}
+	list_del(&mgr->ch.list);
+	list_del(&mgr->bcast.list);
+	skb_queue_purge(&mgr->sendq);
+	kfree(mgr);
+}
+
+static int
+mgr_ctrl(struct mISDNchannel *ch, u_int cmd, void *arg)
+{
+	struct manager	*mgr;
+	int		ret = -EINVAL;
+
+	mgr = container_of(ch, struct manager, ch);
+	if (*debug & DEBUG_L2_CTRL)
+		printk(KERN_DEBUG "%s(%x, %p)\n", __func__, cmd, arg);
+	switch (cmd) {
+	case OPEN_CHANNEL:
+		ret = create_teimgr(mgr, arg);
+		break;
+	case CLOSE_CHANNEL:
+		ret = free_teimanager(mgr);
+		break;
+	case CONTROL_CHANNEL:
+		ret = ctrl_teimanager(mgr, arg);
+		break;
+	case CHECK_DATA:
+		ret = check_data(mgr, arg);
+		break;
+	}
+	return ret;
+}
+
+static int
+mgr_bcast(struct mISDNchannel *ch, struct sk_buff *skb)
+{
+	struct manager		*mgr = container_of(ch, struct manager, bcast);
+	struct mISDNhead	*hh = mISDN_HEAD_P(skb);
+	struct sk_buff		*cskb = NULL;
+	struct layer2		*l2;
+	u_long			flags;
+	int			ret;
+
+	read_lock_irqsave(&mgr->lock, flags);
+	list_for_each_entry(l2, &mgr->layer2, list) {
+		if ((hh->id & MISDN_ID_SAPI_MASK) ==
+		    (l2->ch.addr & MISDN_ID_SAPI_MASK)) {
+			if (list_is_last(&l2->list, &mgr->layer2)) {
+				cskb = skb;
+				skb = NULL;
+			} else {
+				if (!cskb)
+					cskb = skb_copy(skb, GFP_KERNEL);
+			}
+			if (cskb) {
+				ret = l2->ch.send(&l2->ch, cskb);
+				if (ret) {
+					if (*debug & DEBUG_SEND_ERR)
+						printk(KERN_DEBUG
+						    "%s ch%d prim(%x) addr(%x)"
+						    " err %d\n",
+						    __func__, l2->ch.nr,
+						    hh->prim, l2->ch.addr, ret);
+				} else
+					cskb = NULL;
+			} else {
+				printk(KERN_WARNING "%s ch%d addr %x no mem\n",
+				    __func__, ch->nr, ch->addr);
+				goto out;
+			}
+		}
+	}
+out:
+	read_unlock_irqrestore(&mgr->lock, flags);
+	if (cskb)
+		dev_kfree_skb(cskb);
+	if (skb)
+		dev_kfree_skb(skb);
+	return 0;
+}
+
+static int
+mgr_bcast_ctrl(struct mISDNchannel *ch, u_int cmd, void *arg)
+{
+
+	return -EINVAL;
+}
+
+int
+create_teimanager(struct mISDNdevice *dev)
+{
+	struct manager *mgr;
+
+	mgr = kzalloc(sizeof(struct manager), GFP_KERNEL);
+	if (!mgr)
+		return -ENOMEM;
+	INIT_LIST_HEAD(&mgr->layer2);
+	mgr->lock = __RW_LOCK_UNLOCKED(mgr->lock);
+	skb_queue_head_init(&mgr->sendq);
+	mgr->nextid = 1;
+	mgr->lastid = MISDN_ID_NONE;
+	mgr->ch.send = mgr_send;
+	mgr->ch.ctrl = mgr_ctrl;
+	mgr->ch.st = dev->D.st;
+	set_channel_address(&mgr->ch, TEI_SAPI, GROUP_TEI);
+	add_layer2(&mgr->ch, dev->D.st);
+	mgr->bcast.send = mgr_bcast;
+	mgr->bcast.ctrl = mgr_bcast_ctrl;
+	mgr->bcast.st = dev->D.st;
+	set_channel_address(&mgr->bcast, 0, GROUP_TEI);
+	add_layer2(&mgr->bcast, dev->D.st);
+	mgr->deact.debug = *debug & DEBUG_MANAGER;
+	mgr->deact.userdata = mgr;
+	mgr->deact.printdebug = da_debug;
+	mgr->deact.fsm = &deactfsm;
+	mgr->deact.state = ST_L1_DEACT;
+	mISDN_FsmInitTimer(&mgr->deact, &mgr->datimer);
+	dev->teimgr = &mgr->ch;
+	return 0;
+}
+
+int TEIInit(u_int *deb)
+{
+	debug = deb;
+	teifsmu.state_count = TEI_STATE_COUNT;
+	teifsmu.event_count = TEI_EVENT_COUNT;
+	teifsmu.strEvent = strTeiEvent;
+	teifsmu.strState = strTeiState;
+	mISDN_FsmNew(&teifsmu, TeiFnListUser, ARRAY_SIZE(TeiFnListUser));
+	teifsmn.state_count = TEI_STATE_COUNT;
+	teifsmn.event_count = TEI_EVENT_COUNT;
+	teifsmn.strEvent = strTeiEvent;
+	teifsmn.strState = strTeiState;
+	mISDN_FsmNew(&teifsmn, TeiFnListNet, ARRAY_SIZE(TeiFnListNet));
+	deactfsm.state_count =  DEACT_STATE_COUNT;
+	deactfsm.event_count = DEACT_EVENT_COUNT;
+	deactfsm.strEvent = strDeactEvent;
+	deactfsm.strState = strDeactState;
+	mISDN_FsmNew(&deactfsm, DeactFnList, ARRAY_SIZE(DeactFnList));
+	return 0;
+}
+
+void TEIFree(void)
+{
+	mISDN_FsmFree(&teifsmu);
+	mISDN_FsmFree(&teifsmn);
+	mISDN_FsmFree(&deactfsm);
+}
diff --git a/drivers/isdn/mISDN/timerdev.c b/drivers/isdn/mISDN/timerdev.c
new file mode 100644
index 000000000000..b5fabc7019d8
--- /dev/null
+++ b/drivers/isdn/mISDN/timerdev.c
@@ -0,0 +1,301 @@
+/*
+ *
+ * general timer device for using in ISDN stacks
+ *
+ * Author	Karsten Keil <kkeil@novell.com>
+ *
+ * Copyright 2008  by Karsten Keil <kkeil@novell.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/poll.h>
+#include <linux/vmalloc.h>
+#include <linux/timer.h>
+#include <linux/miscdevice.h>
+#include <linux/module.h>
+#include <linux/mISDNif.h>
+
+static int	*debug;
+
+
+struct mISDNtimerdev {
+	int			next_id;
+	struct list_head	pending;
+	struct list_head	expired;
+	wait_queue_head_t	wait;
+	u_int			work;
+	spinlock_t		lock; /* protect lists */
+};
+
+struct mISDNtimer {
+	struct list_head	list;
+	struct  mISDNtimerdev	*dev;
+	struct timer_list	tl;
+	int			id;
+};
+
+static int
+mISDN_open(struct inode *ino, struct file *filep)
+{
+	struct mISDNtimerdev	*dev;
+
+	if (*debug & DEBUG_TIMER)
+		printk(KERN_DEBUG "%s(%p,%p)\n", __func__, ino, filep);
+	dev = kmalloc(sizeof(struct mISDNtimerdev) , GFP_KERNEL);
+	if (!dev)
+		return -ENOMEM;
+	dev->next_id = 1;
+	INIT_LIST_HEAD(&dev->pending);
+	INIT_LIST_HEAD(&dev->expired);
+	spin_lock_init(&dev->lock);
+	dev->work = 0;
+	init_waitqueue_head(&dev->wait);
+	filep->private_data = dev;
+	__module_get(THIS_MODULE);
+	return 0;
+}
+
+static int
+mISDN_close(struct inode *ino, struct file *filep)
+{
+	struct mISDNtimerdev	*dev = filep->private_data;
+	struct mISDNtimer	*timer, *next;
+
+	if (*debug & DEBUG_TIMER)
+		printk(KERN_DEBUG "%s(%p,%p)\n", __func__, ino, filep);
+	list_for_each_entry_safe(timer, next, &dev->pending, list) {
+		del_timer(&timer->tl);
+		kfree(timer);
+	}
+	list_for_each_entry_safe(timer, next, &dev->expired, list) {
+		kfree(timer);
+	}
+	kfree(dev);
+	module_put(THIS_MODULE);
+	return 0;
+}
+
+static ssize_t
+mISDN_read(struct file *filep, char *buf, size_t count, loff_t *off)
+{
+	struct mISDNtimerdev	*dev = filep->private_data;
+	struct mISDNtimer	*timer;
+	u_long	flags;
+	int	ret = 0;
+
+	if (*debug & DEBUG_TIMER)
+		printk(KERN_DEBUG "%s(%p, %p, %d, %p)\n", __func__,
+			filep, buf, (int)count, off);
+	if (*off != filep->f_pos)
+		return -ESPIPE;
+
+	if (list_empty(&dev->expired) && (dev->work == 0)) {
+		if (filep->f_flags & O_NONBLOCK)
+			return -EAGAIN;
+		wait_event_interruptible(dev->wait, (dev->work ||
+		    !list_empty(&dev->expired)));
+		if (signal_pending(current))
+			return -ERESTARTSYS;
+	}
+	if (count < sizeof(int))
+		return -ENOSPC;
+	if (dev->work)
+		dev->work = 0;
+	if (!list_empty(&dev->expired)) {
+		spin_lock_irqsave(&dev->lock, flags);
+		timer = (struct mISDNtimer *)dev->expired.next;
+		list_del(&timer->list);
+		spin_unlock_irqrestore(&dev->lock, flags);
+		if (put_user(timer->id, (int *)buf))
+			ret = -EFAULT;
+		else
+			ret = sizeof(int);
+		kfree(timer);
+	}
+	return ret;
+}
+
+static loff_t
+mISDN_llseek(struct file *filep, loff_t offset, int orig)
+{
+	return -ESPIPE;
+}
+
+static ssize_t
+mISDN_write(struct file *filep, const char *buf, size_t count, loff_t *off)
+{
+	return -EOPNOTSUPP;
+}
+
+static unsigned int
+mISDN_poll(struct file *filep, poll_table *wait)
+{
+	struct mISDNtimerdev	*dev = filep->private_data;
+	unsigned int		mask = POLLERR;
+
+	if (*debug & DEBUG_TIMER)
+		printk(KERN_DEBUG "%s(%p, %p)\n", __func__, filep, wait);
+	if (dev) {
+		poll_wait(filep, &dev->wait, wait);
+		mask = 0;
+		if (dev->work || !list_empty(&dev->expired))
+			mask |= (POLLIN | POLLRDNORM);
+		if (*debug & DEBUG_TIMER)
+			printk(KERN_DEBUG "%s work(%d) empty(%d)\n", __func__,
+				dev->work, list_empty(&dev->expired));
+	}
+	return mask;
+}
+
+static void
+dev_expire_timer(struct mISDNtimer *timer)
+{
+	u_long			flags;
+
+	spin_lock_irqsave(&timer->dev->lock, flags);
+	list_del(&timer->list);
+	list_add_tail(&timer->list, &timer->dev->expired);
+	spin_unlock_irqrestore(&timer->dev->lock, flags);
+	wake_up_interruptible(&timer->dev->wait);
+}
+
+static int
+misdn_add_timer(struct mISDNtimerdev *dev, int timeout)
+{
+	int 			id;
+	u_long			flags;
+	struct mISDNtimer	*timer;
+
+	if (!timeout) {
+		dev->work = 1;
+		wake_up_interruptible(&dev->wait);
+		id = 0;
+	} else {
+		timer = kzalloc(sizeof(struct mISDNtimer), GFP_KERNEL);
+		if (!timer)
+			return -ENOMEM;
+		spin_lock_irqsave(&dev->lock, flags);
+		timer->id = dev->next_id++;
+		if (dev->next_id < 0)
+			dev->next_id = 1;
+		list_add_tail(&timer->list, &dev->pending);
+		spin_unlock_irqrestore(&dev->lock, flags);
+		timer->dev = dev;
+		timer->tl.data = (long)timer;
+		timer->tl.function = (void *) dev_expire_timer;
+		init_timer(&timer->tl);
+		timer->tl.expires = jiffies + ((HZ * (u_long)timeout) / 1000);
+		add_timer(&timer->tl);
+		id = timer->id;
+	}
+	return id;
+}
+
+static int
+misdn_del_timer(struct mISDNtimerdev *dev, int id)
+{
+	u_long			flags;
+	struct mISDNtimer	*timer;
+	int			ret = 0;
+
+	spin_lock_irqsave(&dev->lock, flags);
+	list_for_each_entry(timer, &dev->pending, list) {
+		if (timer->id == id) {
+			list_del_init(&timer->list);
+			del_timer(&timer->tl);
+			ret = timer->id;
+			kfree(timer);
+			goto unlock;
+		}
+	}
+unlock:
+	spin_unlock_irqrestore(&dev->lock, flags);
+	return ret;
+}
+
+static int
+mISDN_ioctl(struct inode *inode, struct file *filep, unsigned int cmd,
+    unsigned long arg)
+{
+	struct mISDNtimerdev	*dev = filep->private_data;
+	int			id, tout, ret = 0;
+
+
+	if (*debug & DEBUG_TIMER)
+		printk(KERN_DEBUG "%s(%p, %x, %lx)\n", __func__,
+		    filep, cmd, arg);
+	switch (cmd) {
+	case IMADDTIMER:
+		if (get_user(tout, (int __user *)arg)) {
+			ret = -EFAULT;
+			break;
+		}
+		id = misdn_add_timer(dev, tout);
+		if (*debug & DEBUG_TIMER)
+			printk(KERN_DEBUG "%s add %d id %d\n", __func__,
+			    tout, id);
+		if (id < 0) {
+			ret = id;
+			break;
+		}
+		if (put_user(id, (int __user *)arg))
+			ret = -EFAULT;
+		break;
+	case IMDELTIMER:
+		if (get_user(id, (int __user *)arg)) {
+			ret = -EFAULT;
+			break;
+		}
+		if (*debug & DEBUG_TIMER)
+			printk(KERN_DEBUG "%s del id %d\n", __func__, id);
+		id = misdn_del_timer(dev, id);
+		if (put_user(id, (int __user *)arg))
+			ret = -EFAULT;
+		break;
+	default:
+		ret = -EINVAL;
+	}
+	return ret;
+}
+
+static struct file_operations mISDN_fops = {
+	.llseek		= mISDN_llseek,
+	.read		= mISDN_read,
+	.write		= mISDN_write,
+	.poll		= mISDN_poll,
+	.ioctl		= mISDN_ioctl,
+	.open		= mISDN_open,
+	.release	= mISDN_close,
+};
+
+static struct miscdevice mISDNtimer = {
+	.minor	= MISC_DYNAMIC_MINOR,
+	.name	= "mISDNtimer",
+	.fops	= &mISDN_fops,
+};
+
+int
+mISDN_inittimer(int *deb)
+{
+	int	err;
+
+	debug = deb;
+	err = misc_register(&mISDNtimer);
+	if (err)
+		printk(KERN_WARNING "mISDN: Could not register timer device\n");
+	return err;
+}
+
+void mISDN_timer_cleanup(void)
+{
+	misc_deregister(&mISDNtimer);
+}
diff --git a/include/linux/mISDNhw.h b/include/linux/mISDNhw.h
new file mode 100644
index 000000000000..e794dfb87504
--- /dev/null
+++ b/include/linux/mISDNhw.h
@@ -0,0 +1,193 @@
+/*
+ *
+ * Author	Karsten Keil <kkeil@novell.com>
+ *
+ *   Basic declarations for the mISDN HW channels
+ *
+ * Copyright 2008  by Karsten Keil <kkeil@novell.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef MISDNHW_H
+#define MISDNHW_H
+#include <linux/mISDNif.h>
+#include <linux/timer.h>
+
+/*
+ * HW DEBUG 0xHHHHGGGG
+ * H - hardware driver specific bits
+ * G - for all drivers
+ */
+
+#define DEBUG_HW		0x00000001
+#define DEBUG_HW_OPEN		0x00000002
+#define DEBUG_HW_DCHANNEL	0x00000100
+#define DEBUG_HW_DFIFO		0x00000200
+#define DEBUG_HW_BCHANNEL	0x00001000
+#define DEBUG_HW_BFIFO		0x00002000
+
+#define MAX_DFRAME_LEN_L1	300
+#define MAX_MON_FRAME		32
+#define MAX_LOG_SPACE		2048
+#define MISDN_COPY_SIZE		32
+
+/* channel->Flags bit field */
+#define FLG_TX_BUSY		0	/* tx_buf in use */
+#define FLG_TX_NEXT		1	/* next_skb in use */
+#define FLG_L1_BUSY		2	/* L1 is permanent busy */
+#define FLG_L2_ACTIVATED	3	/* activated from L2 */
+#define FLG_OPEN		5	/* channel is in use */
+#define FLG_ACTIVE		6	/* channel is activated */
+#define FLG_BUSY_TIMER		7
+/* channel type */
+#define FLG_DCHANNEL		8	/* channel is D-channel */
+#define FLG_BCHANNEL		9	/* channel is B-channel */
+#define FLG_ECHANNEL		10	/* channel is E-channel */
+#define FLG_TRANSPARENT		12	/* channel use transparent data */
+#define FLG_HDLC		13	/* channel use hdlc data */
+#define FLG_L2DATA		14	/* channel use L2 DATA primitivs */
+#define FLG_ORIGIN		15	/* channel is on origin site */
+/* channel specific stuff */
+/* arcofi specific */
+#define FLG_ARCOFI_TIMER	16
+#define FLG_ARCOFI_ERROR	17
+/* isar specific */
+#define FLG_INITIALIZED		16
+#define FLG_DLEETX		17
+#define FLG_LASTDLE		18
+#define FLG_FIRST		19
+#define FLG_LASTDATA		20
+#define FLG_NMD_DATA		21
+#define FLG_FTI_RUN		22
+#define FLG_LL_OK		23
+#define FLG_LL_CONN		24
+#define FLG_DTMFSEND		25
+
+/* workq events */
+#define FLG_RECVQUEUE		30
+#define	FLG_PHCHANGE		31
+
+#define schedule_event(s, ev)	do { \
+					test_and_set_bit(ev, &((s)->Flags)); \
+					schedule_work(&((s)->workq)); \
+				} while (0)
+
+struct dchannel {
+	struct mISDNdevice	dev;
+	u_long			Flags;
+	struct work_struct	workq;
+	void			(*phfunc) (struct dchannel *);
+	u_int			state;
+	void			*l1;
+	/* HW access */
+	u_char			(*read_reg) (void *, u_char);
+	void			(*write_reg) (void *, u_char, u_char);
+	void			(*read_fifo) (void *, u_char *, int);
+	void			(*write_fifo) (void *, u_char *, int);
+	void			*hw;
+	int			slot;	/* multiport card channel slot */
+	struct timer_list	timer;
+	/* receive data */
+	struct sk_buff		*rx_skb;
+	int			maxlen;
+	/* send data */
+	struct sk_buff_head	squeue;
+	struct sk_buff_head	rqueue;
+	struct sk_buff		*tx_skb;
+	int			tx_idx;
+	int			debug;
+	/* statistics */
+	int			err_crc;
+	int			err_tx;
+	int			err_rx;
+};
+
+typedef int	(dchannel_l1callback)(struct dchannel *, u_int);
+extern int	create_l1(struct dchannel *, dchannel_l1callback *);
+
+/* private L1 commands */
+#define INFO0		0x8002
+#define INFO1		0x8102
+#define INFO2		0x8202
+#define INFO3_P8	0x8302
+#define INFO3_P10	0x8402
+#define INFO4_P8	0x8502
+#define INFO4_P10	0x8602
+#define LOSTFRAMING	0x8702
+#define ANYSIGNAL	0x8802
+#define HW_POWERDOWN	0x8902
+#define HW_RESET_REQ	0x8a02
+#define HW_POWERUP_REQ	0x8b02
+#define HW_DEACT_REQ	0x8c02
+#define HW_ACTIVATE_REQ	0x8e02
+#define HW_D_NOBLOCKED  0x8f02
+#define HW_RESET_IND	0x9002
+#define HW_POWERUP_IND	0x9102
+#define HW_DEACT_IND	0x9202
+#define HW_ACTIVATE_IND	0x9302
+#define HW_DEACT_CNF	0x9402
+#define HW_TESTLOOP	0x9502
+#define HW_TESTRX_RAW	0x9602
+#define HW_TESTRX_HDLC	0x9702
+#define HW_TESTRX_OFF	0x9802
+
+struct layer1;
+extern int	l1_event(struct layer1 *, u_int);
+
+
+struct bchannel {
+	struct mISDNchannel	ch;
+	int			nr;
+	u_long			Flags;
+	struct work_struct	workq;
+	u_int			state;
+	/* HW access */
+	u_char			(*read_reg) (void *, u_char);
+	void			(*write_reg) (void *, u_char, u_char);
+	void			(*read_fifo) (void *, u_char *, int);
+	void			(*write_fifo) (void *, u_char *, int);
+	void			*hw;
+	int			slot;	/* multiport card channel slot */
+	struct timer_list	timer;
+	/* receive data */
+	struct sk_buff		*rx_skb;
+	int			maxlen;
+	/* send data */
+	struct sk_buff		*next_skb;
+	struct sk_buff		*tx_skb;
+	struct sk_buff_head	rqueue;
+	int			rcount;
+	int			tx_idx;
+	int			debug;
+	/* statistics */
+	int			err_crc;
+	int			err_tx;
+	int			err_rx;
+};
+
+extern int	mISDN_initdchannel(struct dchannel *, int, void *);
+extern int	mISDN_initbchannel(struct bchannel *, int);
+extern int	mISDN_freedchannel(struct dchannel *);
+extern int	mISDN_freebchannel(struct bchannel *);
+extern void	queue_ch_frame(struct mISDNchannel *, u_int,
+			int, struct sk_buff *);
+extern int	dchannel_senddata(struct dchannel *, struct sk_buff *);
+extern int	bchannel_senddata(struct bchannel *, struct sk_buff *);
+extern void	recv_Dchannel(struct dchannel *);
+extern void	recv_Bchannel(struct bchannel *);
+extern void	recv_Dchannel_skb(struct dchannel *, struct sk_buff *);
+extern void	recv_Bchannel_skb(struct bchannel *, struct sk_buff *);
+extern void	confirm_Bsend(struct bchannel *bch);
+extern int	get_next_bframe(struct bchannel *);
+extern int	get_next_dframe(struct dchannel *);
+
+#endif
diff --git a/include/linux/mISDNif.h b/include/linux/mISDNif.h
new file mode 100644
index 000000000000..5c948f337817
--- /dev/null
+++ b/include/linux/mISDNif.h
@@ -0,0 +1,487 @@
+/*
+ *
+ * Author	Karsten Keil <kkeil@novell.com>
+ *
+ * Copyright 2008  by Karsten Keil <kkeil@novell.com>
+ *
+ * This code is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU LESSER GENERAL PUBLIC LICENSE
+ * version 2.1 as published by the Free Software Foundation.
+ *
+ * This code is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU LESSER GENERAL PUBLIC LICENSE for more details.
+ *
+ */
+
+#ifndef mISDNIF_H
+#define mISDNIF_H
+
+#include <stdarg.h>
+#include <linux/types.h>
+#include <linux/errno.h>
+#include <linux/socket.h>
+
+/*
+ * ABI Version 32 bit
+ *
+ * <8 bit> Major version
+ *		- changed if any interface become backwards incompatible
+ *
+ * <8 bit> Minor version
+ *              - changed if any interface is extended but backwards compatible
+ *
+ * <16 bit> Release number
+ *              - should be incremented on every checkin
+ */
+#define	MISDN_MAJOR_VERSION	1
+#define	MISDN_MINOR_VERSION	0
+#define MISDN_RELEASE		18
+
+/* primitives for information exchange
+ * generell format
+ * <16  bit  0 >
+ * <8  bit command>
+ *    BIT 8 = 1 LAYER private
+ *    BIT 7 = 1 answer
+ *    BIT 6 = 1 DATA
+ * <8  bit target layer mask>
+ *
+ * Layer = 00 is reserved for general commands
+   Layer = 01  L2 -> HW
+   Layer = 02  HW -> L2
+   Layer = 04  L3 -> L2
+   Layer = 08  L2 -> L3
+ * Layer = FF is reserved for broadcast commands
+ */
+
+#define MISDN_CMDMASK		0xff00
+#define MISDN_LAYERMASK		0x00ff
+
+/* generell commands */
+#define OPEN_CHANNEL		0x0100
+#define CLOSE_CHANNEL		0x0200
+#define CONTROL_CHANNEL		0x0300
+#define CHECK_DATA		0x0400
+
+/* layer 2 -> layer 1 */
+#define PH_ACTIVATE_REQ		0x0101
+#define PH_DEACTIVATE_REQ	0x0201
+#define PH_DATA_REQ		0x2001
+#define MPH_ACTIVATE_REQ	0x0501
+#define MPH_DEACTIVATE_REQ	0x0601
+#define MPH_INFORMATION_REQ	0x0701
+#define PH_CONTROL_REQ		0x0801
+
+/* layer 1 -> layer 2 */
+#define PH_ACTIVATE_IND		0x0102
+#define PH_ACTIVATE_CNF		0x4102
+#define PH_DEACTIVATE_IND	0x0202
+#define PH_DEACTIVATE_CNF	0x4202
+#define PH_DATA_IND		0x2002
+#define MPH_ACTIVATE_IND	0x0502
+#define MPH_DEACTIVATE_IND	0x0602
+#define MPH_INFORMATION_IND	0x0702
+#define PH_DATA_CNF		0x6002
+#define PH_CONTROL_IND		0x0802
+#define PH_CONTROL_CNF		0x4802
+
+/* layer 3 -> layer 2 */
+#define DL_ESTABLISH_REQ	0x1004
+#define DL_RELEASE_REQ		0x1104
+#define DL_DATA_REQ		0x3004
+#define DL_UNITDATA_REQ		0x3104
+#define DL_INFORMATION_REQ	0x0004
+
+/* layer 2 -> layer 3 */
+#define DL_ESTABLISH_IND	0x1008
+#define DL_ESTABLISH_CNF	0x5008
+#define DL_RELEASE_IND		0x1108
+#define DL_RELEASE_CNF		0x5108
+#define DL_DATA_IND		0x3008
+#define DL_UNITDATA_IND		0x3108
+#define DL_INFORMATION_IND	0x0008
+
+/* intern layer 2 managment */
+#define MDL_ASSIGN_REQ		0x1804
+#define MDL_ASSIGN_IND		0x1904
+#define MDL_REMOVE_REQ		0x1A04
+#define MDL_REMOVE_IND		0x1B04
+#define MDL_STATUS_UP_IND	0x1C04
+#define MDL_STATUS_DOWN_IND	0x1D04
+#define MDL_STATUS_UI_IND	0x1E04
+#define MDL_ERROR_IND		0x1F04
+#define MDL_ERROR_RSP		0x5F04
+
+/* DL_INFORMATION_IND types */
+#define DL_INFO_L2_CONNECT	0x0001
+#define DL_INFO_L2_REMOVED	0x0002
+
+/* PH_CONTROL types */
+/* TOUCH TONE IS 0x20XX  XX "0"..."9", "A","B","C","D","*","#" */
+#define DTMF_TONE_VAL		0x2000
+#define DTMF_TONE_MASK		0x007F
+#define DTMF_TONE_START		0x2100
+#define DTMF_TONE_STOP		0x2200
+#define DTMF_HFC_COEF		0x4000
+#define DSP_CONF_JOIN		0x2403
+#define DSP_CONF_SPLIT		0x2404
+#define DSP_RECEIVE_OFF		0x2405
+#define DSP_RECEIVE_ON		0x2406
+#define DSP_ECHO_ON		0x2407
+#define DSP_ECHO_OFF		0x2408
+#define DSP_MIX_ON		0x2409
+#define DSP_MIX_OFF		0x240a
+#define DSP_DELAY		0x240b
+#define DSP_JITTER		0x240c
+#define DSP_TXDATA_ON		0x240d
+#define DSP_TXDATA_OFF		0x240e
+#define DSP_TX_DEJITTER		0x240f
+#define DSP_TX_DEJ_OFF		0x2410
+#define DSP_TONE_PATT_ON	0x2411
+#define DSP_TONE_PATT_OFF	0x2412
+#define DSP_VOL_CHANGE_TX	0x2413
+#define DSP_VOL_CHANGE_RX	0x2414
+#define DSP_BF_ENABLE_KEY	0x2415
+#define DSP_BF_DISABLE		0x2416
+#define DSP_BF_ACCEPT		0x2416
+#define DSP_BF_REJECT		0x2417
+#define DSP_PIPELINE_CFG	0x2418
+#define HFC_VOL_CHANGE_TX	0x2601
+#define HFC_VOL_CHANGE_RX	0x2602
+#define HFC_SPL_LOOP_ON		0x2603
+#define HFC_SPL_LOOP_OFF	0x2604
+
+/* DSP_TONE_PATT_ON parameter */
+#define TONE_OFF			0x0000
+#define TONE_GERMAN_DIALTONE		0x0001
+#define TONE_GERMAN_OLDDIALTONE		0x0002
+#define TONE_AMERICAN_DIALTONE		0x0003
+#define TONE_GERMAN_DIALPBX		0x0004
+#define TONE_GERMAN_OLDDIALPBX		0x0005
+#define TONE_AMERICAN_DIALPBX		0x0006
+#define TONE_GERMAN_RINGING		0x0007
+#define TONE_GERMAN_OLDRINGING		0x0008
+#define TONE_AMERICAN_RINGPBX		0x000b
+#define TONE_GERMAN_RINGPBX		0x000c
+#define TONE_GERMAN_OLDRINGPBX		0x000d
+#define TONE_AMERICAN_RINGING		0x000e
+#define TONE_GERMAN_BUSY		0x000f
+#define TONE_GERMAN_OLDBUSY		0x0010
+#define TONE_AMERICAN_BUSY		0x0011
+#define TONE_GERMAN_HANGUP		0x0012
+#define TONE_GERMAN_OLDHANGUP		0x0013
+#define TONE_AMERICAN_HANGUP		0x0014
+#define TONE_SPECIAL_INFO		0x0015
+#define TONE_GERMAN_GASSENBESETZT	0x0016
+#define TONE_GERMAN_AUFSCHALTTON	0x0016
+
+/* MPH_INFORMATION_IND */
+#define L1_SIGNAL_LOS_OFF	0x0010
+#define L1_SIGNAL_LOS_ON	0x0011
+#define L1_SIGNAL_AIS_OFF	0x0012
+#define L1_SIGNAL_AIS_ON	0x0013
+#define L1_SIGNAL_RDI_OFF	0x0014
+#define L1_SIGNAL_RDI_ON	0x0015
+#define L1_SIGNAL_SLIP_RX	0x0020
+#define L1_SIGNAL_SLIP_TX	0x0021
+
+/*
+ * protocol ids
+ * D channel 1-31
+ * B channel 33 - 63
+ */
+
+#define ISDN_P_NONE		0
+#define ISDN_P_BASE		0
+#define ISDN_P_TE_S0		0x01
+#define ISDN_P_NT_S0  		0x02
+#define ISDN_P_TE_E1		0x03
+#define ISDN_P_NT_E1  		0x04
+#define ISDN_P_LAPD_TE		0x10
+#define	ISDN_P_LAPD_NT		0x11
+
+#define ISDN_P_B_MASK		0x1f
+#define ISDN_P_B_START		0x20
+
+#define ISDN_P_B_RAW		0x21
+#define ISDN_P_B_HDLC		0x22
+#define ISDN_P_B_X75SLP		0x23
+#define ISDN_P_B_L2DTMF		0x24
+#define ISDN_P_B_L2DSP		0x25
+#define ISDN_P_B_L2DSPHDLC	0x26
+
+#define OPTION_L2_PMX		1
+#define OPTION_L2_PTP		2
+#define OPTION_L2_FIXEDTEI	3
+#define OPTION_L2_CLEANUP	4
+
+/* should be in sync with linux/kobject.h:KOBJ_NAME_LEN */
+#define MISDN_MAX_IDLEN		20
+
+struct mISDNhead {
+	unsigned int	prim;
+	unsigned int	id;
+}  __attribute__((packed));
+
+#define MISDN_HEADER_LEN	sizeof(struct mISDNhead)
+#define MAX_DATA_SIZE		2048
+#define MAX_DATA_MEM		(MAX_DATA_SIZE + MISDN_HEADER_LEN)
+#define MAX_DFRAME_LEN		260
+
+#define MISDN_ID_ADDR_MASK	0xFFFF
+#define MISDN_ID_TEI_MASK	0xFF00
+#define MISDN_ID_SAPI_MASK	0x00FF
+#define MISDN_ID_TEI_ANY	0x7F00
+
+#define MISDN_ID_ANY		0xFFFF
+#define MISDN_ID_NONE		0xFFFE
+
+#define GROUP_TEI		127
+#define TEI_SAPI		63
+#define CTRL_SAPI		0
+
+#define MISDN_CHMAP_SIZE	4
+
+#define SOL_MISDN	0
+
+struct sockaddr_mISDN {
+	sa_family_t    family;
+	unsigned char	dev;
+	unsigned char	channel;
+	unsigned char	sapi;
+	unsigned char	tei;
+};
+
+/* timer device ioctl */
+#define IMADDTIMER	_IOR('I', 64, int)
+#define IMDELTIMER	_IOR('I', 65, int)
+/* socket ioctls */
+#define	IMGETVERSION	_IOR('I', 66, int)
+#define	IMGETCOUNT	_IOR('I', 67, int)
+#define IMGETDEVINFO	_IOR('I', 68, int)
+#define IMCTRLREQ	_IOR('I', 69, int)
+#define IMCLEAR_L2	_IOR('I', 70, int)
+
+struct mISDNversion {
+	unsigned char	major;
+	unsigned char	minor;
+	unsigned short	release;
+};
+
+struct mISDN_devinfo {
+	u_int			id;
+	u_int			Dprotocols;
+	u_int			Bprotocols;
+	u_int			protocol;
+	u_long			channelmap[MISDN_CHMAP_SIZE];
+	u_int			nrbchan;
+	char			name[MISDN_MAX_IDLEN];
+};
+
+/* CONTROL_CHANNEL parameters */
+#define MISDN_CTRL_GETOP		0x0000
+#define MISDN_CTRL_LOOP			0x0001
+#define MISDN_CTRL_CONNECT		0x0002
+#define MISDN_CTRL_DISCONNECT		0x0004
+#define MISDN_CTRL_PCMCONNECT		0x0010
+#define MISDN_CTRL_PCMDISCONNECT	0x0020
+#define MISDN_CTRL_SETPEER		0x0040
+#define MISDN_CTRL_UNSETPEER		0x0080
+#define MISDN_CTRL_RX_OFF		0x0100
+#define MISDN_CTRL_HW_FEATURES_OP	0x2000
+#define MISDN_CTRL_HW_FEATURES		0x2001
+#define MISDN_CTRL_HFC_OP		0x4000
+#define MISDN_CTRL_HFC_PCM_CONN		0x4001
+#define MISDN_CTRL_HFC_PCM_DISC		0x4002
+#define MISDN_CTRL_HFC_CONF_JOIN	0x4003
+#define MISDN_CTRL_HFC_CONF_SPLIT	0x4004
+#define MISDN_CTRL_HFC_RECEIVE_OFF	0x4005
+#define MISDN_CTRL_HFC_RECEIVE_ON	0x4006
+#define MISDN_CTRL_HFC_ECHOCAN_ON 	0x4007
+#define MISDN_CTRL_HFC_ECHOCAN_OFF 	0x4008
+
+
+/* socket options */
+#define MISDN_TIME_STAMP		0x0001
+
+struct mISDN_ctrl_req {
+	int		op;
+	int		channel;
+	int		p1;
+	int		p2;
+};
+
+/* muxer options */
+#define MISDN_OPT_ALL		1
+#define MISDN_OPT_TEIMGR	2
+
+#ifdef __KERNEL__
+#include <linux/list.h>
+#include <linux/skbuff.h>
+#include <linux/net.h>
+#include <net/sock.h>
+#include <linux/completion.h>
+
+#define DEBUG_CORE		0x000000ff
+#define DEBUG_CORE_FUNC		0x00000002
+#define DEBUG_SOCKET		0x00000004
+#define DEBUG_MANAGER		0x00000008
+#define DEBUG_SEND_ERR		0x00000010
+#define DEBUG_MSG_THREAD	0x00000020
+#define DEBUG_QUEUE_FUNC	0x00000040
+#define DEBUG_L1		0x0000ff00
+#define DEBUG_L1_FSM		0x00000200
+#define DEBUG_L2		0x00ff0000
+#define DEBUG_L2_FSM		0x00020000
+#define DEBUG_L2_CTRL		0x00040000
+#define DEBUG_L2_RECV		0x00080000
+#define DEBUG_L2_TEI		0x00100000
+#define DEBUG_L2_TEIFSM		0x00200000
+#define DEBUG_TIMER		0x01000000
+
+#define mISDN_HEAD_P(s)		((struct mISDNhead *)&s->cb[0])
+#define mISDN_HEAD_PRIM(s)	(((struct mISDNhead *)&s->cb[0])->prim)
+#define mISDN_HEAD_ID(s)	(((struct mISDNhead *)&s->cb[0])->id)
+
+/* socket states */
+#define MISDN_OPEN	1
+#define MISDN_BOUND	2
+#define MISDN_CLOSED	3
+
+struct mISDNchannel;
+struct mISDNdevice;
+struct mISDNstack;
+
+struct channel_req {
+	u_int			protocol;
+	struct sockaddr_mISDN	adr;
+	struct mISDNchannel	*ch;
+};
+
+typedef	int	(ctrl_func_t)(struct mISDNchannel *, u_int, void *);
+typedef	int	(send_func_t)(struct mISDNchannel *, struct sk_buff *);
+typedef int	(create_func_t)(struct channel_req *);
+
+struct Bprotocol {
+	struct list_head	list;
+	char			*name;
+	u_int			Bprotocols;
+	create_func_t		*create;
+};
+
+struct mISDNchannel {
+	struct list_head	list;
+	u_int			protocol;
+	u_int			nr;
+	u_long			opt;
+	u_int			addr;
+	struct mISDNstack	*st;
+	struct mISDNchannel	*peer;
+	send_func_t		*send;
+	send_func_t		*recv;
+	ctrl_func_t		*ctrl;
+};
+
+struct mISDN_sock_list {
+	struct hlist_head	head;
+	rwlock_t		lock;
+};
+
+struct mISDN_sock {
+	struct sock		sk;
+	struct mISDNchannel	ch;
+	u_int			cmask;
+	struct mISDNdevice	*dev;
+};
+
+
+
+struct mISDNdevice {
+	struct mISDNchannel	D;
+	u_int			id;
+	char			name[MISDN_MAX_IDLEN];
+	u_int			Dprotocols;
+	u_int			Bprotocols;
+	u_int			nrbchan;
+	u_long			channelmap[MISDN_CHMAP_SIZE];
+	struct list_head	bchannels;
+	struct mISDNchannel	*teimgr;
+	struct device		dev;
+};
+
+struct mISDNstack {
+	u_long			status;
+	struct mISDNdevice	*dev;
+	struct task_struct	*thread;
+	struct completion	*notify;
+	wait_queue_head_t	workq;
+	struct sk_buff_head	msgq;
+	struct list_head	layer2;
+	struct mISDNchannel	*layer1;
+	struct mISDNchannel	own;
+	struct mutex		lmutex; /* protect lists */
+	struct mISDN_sock_list	l1sock;
+#ifdef MISDN_MSG_STATS
+	u_int			msg_cnt;
+	u_int			sleep_cnt;
+	u_int			stopped_cnt;
+#endif
+};
+
+/* global alloc/queue dunctions */
+
+static inline struct sk_buff *
+mI_alloc_skb(unsigned int len, gfp_t gfp_mask)
+{
+	struct sk_buff	*skb;
+
+	skb = alloc_skb(len + MISDN_HEADER_LEN, gfp_mask);
+	if (likely(skb))
+		skb_reserve(skb, MISDN_HEADER_LEN);
+	return skb;
+}
+
+static inline struct sk_buff *
+_alloc_mISDN_skb(u_int prim, u_int id, u_int len, void *dp, gfp_t gfp_mask)
+{
+	struct sk_buff	*skb = mI_alloc_skb(len, gfp_mask);
+	struct mISDNhead *hh;
+
+	if (!skb)
+		return NULL;
+	if (len)
+		memcpy(skb_put(skb, len), dp, len);
+	hh = mISDN_HEAD_P(skb);
+	hh->prim = prim;
+	hh->id = id;
+	return skb;
+}
+
+static inline void
+_queue_data(struct mISDNchannel *ch, u_int prim,
+    u_int id, u_int len, void *dp, gfp_t gfp_mask)
+{
+	struct sk_buff		*skb;
+
+	if (!ch->peer)
+		return;
+	skb = _alloc_mISDN_skb(prim, id, len, dp, gfp_mask);
+	if (!skb)
+		return;
+	if (ch->recv(ch->peer, skb))
+		dev_kfree_skb(skb);
+}
+
+/* global register/unregister functions */
+
+extern int	mISDN_register_device(struct mISDNdevice *, char *name);
+extern void	mISDN_unregister_device(struct mISDNdevice *);
+extern int	mISDN_register_Bprotocol(struct Bprotocol *);
+extern void	mISDN_unregister_Bprotocol(struct Bprotocol *);
+
+extern void	set_channel_address(struct mISDNchannel *, u_int, u_int);
+
+#endif /* __KERNEL__ */
+#endif /* mISDNIF_H */

From 960366cf8dbb3359afaca30cf7fdbf69a6d6dda7 Mon Sep 17 00:00:00 2001
From: Karsten Keil <kkeil@suse.de>
Date: Sun, 27 Jul 2008 01:56:38 +0200
Subject: [PATCH 1115/1171] Add mISDN DSP

Enable support for digital audio processing capability.
This module may be used for special applications that require
cross connecting of bchannels, conferencing, dtmf decoding
echo cancelation, tone generation, and Blowfish encryption and
decryption.
It may use hardware features if available.

Signed-off-by: Karsten Keil <kkeil@suse.de>
---
 drivers/isdn/mISDN/Kconfig        |   18 +
 drivers/isdn/mISDN/Makefile       |    2 +
 drivers/isdn/mISDN/dsp.h          |  263 ++++
 drivers/isdn/mISDN/dsp_audio.c    |  434 +++++++
 drivers/isdn/mISDN/dsp_biquad.h   |   65 +
 drivers/isdn/mISDN/dsp_blowfish.c |  672 ++++++++++
 drivers/isdn/mISDN/dsp_cmx.c      | 1886 +++++++++++++++++++++++++++++
 drivers/isdn/mISDN/dsp_core.c     | 1191 ++++++++++++++++++
 drivers/isdn/mISDN/dsp_dtmf.c     |  303 +++++
 drivers/isdn/mISDN/dsp_ecdis.h    |  110 ++
 drivers/isdn/mISDN/dsp_hwec.c     |  138 +++
 drivers/isdn/mISDN/dsp_hwec.h     |   10 +
 drivers/isdn/mISDN/dsp_pipeline.c |  348 ++++++
 drivers/isdn/mISDN/dsp_tones.c    |  551 +++++++++
 include/linux/mISDNdsp.h          |   37 +
 15 files changed, 6028 insertions(+)
 create mode 100644 drivers/isdn/mISDN/dsp.h
 create mode 100644 drivers/isdn/mISDN/dsp_audio.c
 create mode 100644 drivers/isdn/mISDN/dsp_biquad.h
 create mode 100644 drivers/isdn/mISDN/dsp_blowfish.c
 create mode 100644 drivers/isdn/mISDN/dsp_cmx.c
 create mode 100644 drivers/isdn/mISDN/dsp_core.c
 create mode 100644 drivers/isdn/mISDN/dsp_dtmf.c
 create mode 100644 drivers/isdn/mISDN/dsp_ecdis.h
 create mode 100644 drivers/isdn/mISDN/dsp_hwec.c
 create mode 100644 drivers/isdn/mISDN/dsp_hwec.h
 create mode 100644 drivers/isdn/mISDN/dsp_pipeline.c
 create mode 100644 drivers/isdn/mISDN/dsp_tones.c
 create mode 100644 include/linux/mISDNdsp.h

diff --git a/drivers/isdn/mISDN/Kconfig b/drivers/isdn/mISDN/Kconfig
index 231bd0d08316..6a97e86e7f21 100644
--- a/drivers/isdn/mISDN/Kconfig
+++ b/drivers/isdn/mISDN/Kconfig
@@ -7,3 +7,21 @@ menuconfig MISDN
 	help
 	  Enable support for the modular ISDN driver.
 
+if MISDN != n
+
+config MISDN_DSP
+	tristate "Digital Audio Processing of transparent data"
+	depends on MISDN
+	help
+	  Enable support for digital audio processing capability.
+	  This module may be used for special applications that require
+	  cross connecting of bchannels, conferencing, dtmf decoding
+	  echo cancelation, tone generation, and Blowfish encryption and
+	  decryption.
+	  It may use hardware features if available.
+	  E.g. it is required for PBX4Linux. Go to http://isdn.eversberg.eu
+	  and get more informations about this module and it's usage.
+	  If unsure, say 'N'.
+
+	source "drivers/isdn/hardware/mISDN/Kconfig"
+endif #MISDN
diff --git a/drivers/isdn/mISDN/Makefile b/drivers/isdn/mISDN/Makefile
index 87c563d33612..7f1a21804208 100644
--- a/drivers/isdn/mISDN/Makefile
+++ b/drivers/isdn/mISDN/Makefile
@@ -3,7 +3,9 @@
 #
 
 obj-$(CONFIG_MISDN) += mISDN_core.o
+obj-$(CONFIG_MISDN_DSP) += mISDN_dsp.o
 
 # multi objects
 
 mISDN_core-objs := core.o fsm.o socket.o hwchannel.o stack.o layer1.o layer2.o tei.o timerdev.o
+mISDN_dsp-objs := dsp_core.o dsp_cmx.o dsp_tones.o dsp_dtmf.o dsp_audio.o dsp_blowfish.o dsp_pipeline.o dsp_hwec.o
diff --git a/drivers/isdn/mISDN/dsp.h b/drivers/isdn/mISDN/dsp.h
new file mode 100644
index 000000000000..6c3fed6b8d4f
--- /dev/null
+++ b/drivers/isdn/mISDN/dsp.h
@@ -0,0 +1,263 @@
+/*
+ * Audio support data for ISDN4Linux.
+ *
+ * Copyright 2002/2003 by Andreas Eversberg (jolly@eversberg.eu)
+ *
+ * This software may be used and distributed according to the terms
+ * of the GNU General Public License, incorporated herein by reference.
+ *
+ */
+
+#define DEBUG_DSP_CTRL		0x0001
+#define DEBUG_DSP_CORE		0x0002
+#define DEBUG_DSP_DTMF		0x0004
+#define DEBUG_DSP_CMX		0x0010
+#define DEBUG_DSP_TONE		0x0020
+#define DEBUG_DSP_BLOWFISH	0x0040
+#define DEBUG_DSP_DELAY		0x0100
+#define DEBUG_DSP_DTMFCOEFF	0x8000 /* heavy output */
+
+/* options may be:
+ *
+ * bit 0 = use ulaw instead of alaw
+ * bit 1 = enable hfc hardware accelleration for all channels
+ *
+ */
+#define DSP_OPT_ULAW		(1<<0)
+#define DSP_OPT_NOHARDWARE	(1<<1)
+
+#include <linux/timer.h>
+#include <linux/workqueue.h>
+
+#include "dsp_ecdis.h"
+
+extern int dsp_options;
+extern int dsp_debug;
+extern int dsp_poll;
+extern int dsp_tics;
+extern spinlock_t dsp_lock;
+extern struct work_struct dsp_workq;
+extern u32 dsp_poll_diff; /* calculated fix-comma corrected poll value */
+
+/***************
+ * audio stuff *
+ ***************/
+
+extern s32 dsp_audio_alaw_to_s32[256];
+extern s32 dsp_audio_ulaw_to_s32[256];
+extern s32 *dsp_audio_law_to_s32;
+extern u8 dsp_audio_s16_to_law[65536];
+extern u8 dsp_audio_alaw_to_ulaw[256];
+extern u8 dsp_audio_mix_law[65536];
+extern u8 dsp_audio_seven2law[128];
+extern u8 dsp_audio_law2seven[256];
+extern void dsp_audio_generate_law_tables(void);
+extern void dsp_audio_generate_s2law_table(void);
+extern void dsp_audio_generate_seven(void);
+extern void dsp_audio_generate_mix_table(void);
+extern void dsp_audio_generate_ulaw_samples(void);
+extern void dsp_audio_generate_volume_changes(void);
+extern u8 dsp_silence;
+
+
+/*************
+ * cmx stuff *
+ *************/
+
+#define MAX_POLL	256	/* maximum number of send-chunks */
+
+#define CMX_BUFF_SIZE	0x8000	/* must be 2**n (0x1000 about 1/2 second) */
+#define CMX_BUFF_HALF	0x4000	/* CMX_BUFF_SIZE / 2 */
+#define CMX_BUFF_MASK	0x7fff	/* CMX_BUFF_SIZE - 1 */
+
+/* how many seconds will we check the lowest delay until the jitter buffer
+   is reduced by that delay */
+#define MAX_SECONDS_JITTER_CHECK 5
+
+extern struct timer_list dsp_spl_tl;
+extern u32 dsp_spl_jiffies;
+
+/* the structure of conferences:
+ *
+ * each conference has a unique number, given by user space.
+ * the conferences are linked in a chain.
+ * each conference has members linked in a chain.
+ * each dsplayer points to a member, each member points to a dsplayer.
+ */
+
+/* all members within a conference (this is linked 1:1 with the dsp) */
+struct dsp;
+struct dsp_conf_member {
+	struct list_head	list;
+	struct dsp		*dsp;
+};
+
+/* the list of all conferences */
+struct dsp_conf {
+	struct list_head	list;
+	u32			id;
+				/* all cmx stacks with the same ID are
+				 connected */
+	struct list_head	mlist;
+	int			software; /* conf is processed by software */
+	int			hardware; /* conf is processed by hardware */
+				/* note: if both unset, has only one member */
+};
+
+
+/**************
+ * DTMF stuff *
+ **************/
+
+#define DSP_DTMF_NPOINTS 102
+
+#define ECHOCAN_BUFLEN (4*128)
+
+struct dsp_dtmf {
+	int		treshold; /* above this is dtmf (square of) */
+	int		software; /* dtmf uses software decoding */
+	int		hardware; /* dtmf uses hardware decoding */
+	int		size; /* number of bytes in buffer */
+	signed short	buffer[DSP_DTMF_NPOINTS];
+		/* buffers one full dtmf frame */
+	u8		lastwhat, lastdigit;
+	int		count;
+	u8		digits[16]; /* just the dtmf result */
+};
+
+
+/******************
+ * pipeline stuff *
+ ******************/
+struct dsp_pipeline {
+	rwlock_t  lock;
+	struct list_head list;
+	int inuse;
+};
+
+/***************
+ * tones stuff *
+ ***************/
+
+struct dsp_tone {
+	int		software; /* tones are generated by software */
+	int		hardware; /* tones are generated by hardware */
+	int		tone;
+	void		*pattern;
+	int		count;
+	int		index;
+	struct timer_list tl;
+};
+
+/*****************
+ * general stuff *
+ *****************/
+
+struct dsp {
+	struct list_head list;
+	struct mISDNchannel	ch;
+	struct mISDNchannel	*up;
+	unsigned char	name[64];
+	int		b_active;
+	int		echo; /* echo is enabled */
+	int		rx_disabled; /* what the user wants */
+	int		rx_is_off; /* what the card is */
+	int		tx_mix;
+	struct dsp_tone	tone;
+	struct dsp_dtmf	dtmf;
+	int		tx_volume, rx_volume;
+
+	/* queue for sending frames */
+	struct work_struct	workq;
+	struct sk_buff_head	sendq;
+	int		hdlc;	/* if mode is hdlc */
+	int		data_pending;	/* currently an unconfirmed frame */
+
+	/* conference stuff */
+	u32		conf_id;
+	struct dsp_conf	*conf;
+	struct dsp_conf_member
+			*member;
+
+	/* buffer stuff */
+	int		rx_W; /* current write pos for data without timestamp */
+	int		rx_R; /* current read pos for transmit clock */
+	int		rx_init; /* if set, pointers will be adjusted first */
+	int		tx_W; /* current write pos for transmit data */
+	int		tx_R; /* current read pos for transmit clock */
+	int		rx_delay[MAX_SECONDS_JITTER_CHECK];
+	int		tx_delay[MAX_SECONDS_JITTER_CHECK];
+	u8		tx_buff[CMX_BUFF_SIZE];
+	u8		rx_buff[CMX_BUFF_SIZE];
+	int		last_tx; /* if set, we transmitted last poll interval */
+	int		cmx_delay; /* initial delay of buffers,
+				or 0 for dynamic jitter buffer */
+	int		tx_dejitter; /* if set, dejitter tx buffer */
+	int		tx_data; /* enables tx-data of CMX to upper layer */
+
+	/* hardware stuff */
+	struct dsp_features features;
+	int		features_rx_off; /* set if rx_off is featured */
+	int		pcm_slot_rx; /* current PCM slot (or -1) */
+	int		pcm_bank_rx;
+	int		pcm_slot_tx;
+	int		pcm_bank_tx;
+	int		hfc_conf; /* unique id of current conference (or -1) */
+
+	/* encryption stuff */
+	int		bf_enable;
+	u32		bf_p[18];
+	u32		bf_s[1024];
+	int		bf_crypt_pos;
+	u8		bf_data_in[9];
+	u8		bf_crypt_out[9];
+	int		bf_decrypt_in_pos;
+	int		bf_decrypt_out_pos;
+	u8		bf_crypt_inring[16];
+	u8		bf_data_out[9];
+	int		bf_sync;
+
+	struct dsp_pipeline
+			pipeline;
+};
+
+/* functions */
+
+extern void dsp_change_volume(struct sk_buff *skb, int volume);
+
+extern struct list_head dsp_ilist;
+extern struct list_head conf_ilist;
+extern void dsp_cmx_debug(struct dsp *dsp);
+extern void dsp_cmx_hardware(struct dsp_conf *conf, struct dsp *dsp);
+extern int dsp_cmx_conf(struct dsp *dsp, u32 conf_id);
+extern void dsp_cmx_receive(struct dsp *dsp, struct sk_buff *skb);
+extern void dsp_cmx_hdlc(struct dsp *dsp, struct sk_buff *skb);
+extern void dsp_cmx_send(void *arg);
+extern void dsp_cmx_transmit(struct dsp *dsp, struct sk_buff *skb);
+extern int dsp_cmx_del_conf_member(struct dsp *dsp);
+extern int dsp_cmx_del_conf(struct dsp_conf *conf);
+
+extern void dsp_dtmf_goertzel_init(struct dsp *dsp);
+extern void dsp_dtmf_hardware(struct dsp *dsp);
+extern u8 *dsp_dtmf_goertzel_decode(struct dsp *dsp, u8 *data, int len,
+		int fmt);
+
+extern int dsp_tone(struct dsp *dsp, int tone);
+extern void dsp_tone_copy(struct dsp *dsp, u8 *data, int len);
+extern void dsp_tone_timeout(void *arg);
+
+extern void dsp_bf_encrypt(struct dsp *dsp, u8 *data, int len);
+extern void dsp_bf_decrypt(struct dsp *dsp, u8 *data, int len);
+extern int dsp_bf_init(struct dsp *dsp, const u8 *key, unsigned int keylen);
+extern void dsp_bf_cleanup(struct dsp *dsp);
+
+extern int  dsp_pipeline_module_init(void);
+extern void dsp_pipeline_module_exit(void);
+extern int  dsp_pipeline_init(struct dsp_pipeline *pipeline);
+extern void dsp_pipeline_destroy(struct dsp_pipeline *pipeline);
+extern int  dsp_pipeline_build(struct dsp_pipeline *pipeline, const char *cfg);
+extern void dsp_pipeline_process_tx(struct dsp_pipeline *pipeline, u8 *data,
+		int len);
+extern void dsp_pipeline_process_rx(struct dsp_pipeline *pipeline, u8 *data,
+		int len);
+
diff --git a/drivers/isdn/mISDN/dsp_audio.c b/drivers/isdn/mISDN/dsp_audio.c
new file mode 100644
index 000000000000..1c2dd5694773
--- /dev/null
+++ b/drivers/isdn/mISDN/dsp_audio.c
@@ -0,0 +1,434 @@
+/*
+ * Audio support data for mISDN_dsp.
+ *
+ * Copyright 2002/2003 by Andreas Eversberg (jolly@eversberg.eu)
+ * Rewritten by Peter
+ *
+ * This software may be used and distributed according to the terms
+ * of the GNU General Public License, incorporated herein by reference.
+ *
+ */
+
+#include <linux/delay.h>
+#include <linux/mISDNif.h>
+#include <linux/mISDNdsp.h>
+#include "core.h"
+#include "dsp.h"
+
+/* ulaw[unsigned char] -> signed 16-bit */
+s32 dsp_audio_ulaw_to_s32[256];
+/* alaw[unsigned char] -> signed 16-bit */
+s32 dsp_audio_alaw_to_s32[256];
+
+s32 *dsp_audio_law_to_s32;
+EXPORT_SYMBOL(dsp_audio_law_to_s32);
+
+/* signed 16-bit -> law */
+u8 dsp_audio_s16_to_law[65536];
+EXPORT_SYMBOL(dsp_audio_s16_to_law);
+
+/* alaw -> ulaw */
+u8 dsp_audio_alaw_to_ulaw[256];
+/* ulaw -> alaw */
+u8 dsp_audio_ulaw_to_alaw[256];
+u8 dsp_silence;
+
+
+/*****************************************************
+ * generate table for conversion of s16 to alaw/ulaw *
+ *****************************************************/
+
+#define AMI_MASK 0x55
+
+static inline unsigned char linear2alaw(short int linear)
+{
+	int mask;
+	int seg;
+	int pcm_val;
+	static int seg_end[8] = {
+		0xFF, 0x1FF, 0x3FF, 0x7FF, 0xFFF, 0x1FFF, 0x3FFF, 0x7FFF
+	};
+
+	pcm_val = linear;
+	if (pcm_val >= 0) {
+		/* Sign (7th) bit = 1 */
+		mask = AMI_MASK | 0x80;
+	} else {
+		/* Sign bit = 0 */
+		mask = AMI_MASK;
+		pcm_val = -pcm_val;
+	}
+
+	/* Convert the scaled magnitude to segment number. */
+	for (seg = 0;  seg < 8;  seg++) {
+		if (pcm_val <= seg_end[seg])
+			break;
+	}
+	/* Combine the sign, segment, and quantization bits. */
+	return  ((seg << 4) |
+		 ((pcm_val >> ((seg)  ?  (seg + 3)  :  4)) & 0x0F)) ^ mask;
+}
+
+
+static inline short int alaw2linear(unsigned char alaw)
+{
+	int i;
+	int seg;
+
+	alaw ^= AMI_MASK;
+	i = ((alaw & 0x0F) << 4) + 8 /* rounding error */;
+	seg = (((int) alaw & 0x70) >> 4);
+	if (seg)
+		i = (i + 0x100) << (seg - 1);
+	return (short int) ((alaw & 0x80)  ?  i  :  -i);
+}
+
+static inline short int ulaw2linear(unsigned char ulaw)
+{
+	short mu, e, f, y;
+	static short etab[] = {0, 132, 396, 924, 1980, 4092, 8316, 16764};
+
+	mu = 255 - ulaw;
+	e = (mu & 0x70) / 16;
+	f = mu & 0x0f;
+	y = f * (1 << (e + 3));
+	y += etab[e];
+	if (mu & 0x80)
+		y = -y;
+	return y;
+}
+
+#define BIAS 0x84   /*!< define the add-in bias for 16 bit samples */
+
+static unsigned char linear2ulaw(short sample)
+{
+	static int exp_lut[256] = {
+		0, 0, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 3, 3,
+		4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4,
+		5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5,
+		5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5,
+		6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6,
+		6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6,
+		6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6,
+		6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6,
+		7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
+		7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
+		7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
+		7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
+		7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
+		7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
+		7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7,
+		7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7};
+	int sign, exponent, mantissa;
+	unsigned char ulawbyte;
+
+	/* Get the sample into sign-magnitude. */
+	sign = (sample >> 8) & 0x80;	  /* set aside the sign */
+	if (sign != 0)
+		sample = -sample;	      /* get magnitude */
+
+	/* Convert from 16 bit linear to ulaw. */
+	sample = sample + BIAS;
+	exponent = exp_lut[(sample >> 7) & 0xFF];
+	mantissa = (sample >> (exponent + 3)) & 0x0F;
+	ulawbyte = ~(sign | (exponent << 4) | mantissa);
+
+	return ulawbyte;
+}
+
+static int reverse_bits(int i)
+{
+	int z, j;
+	z = 0;
+
+	for (j = 0; j < 8; j++) {
+		if ((i & (1 << j)) != 0)
+			z |= 1 << (7 - j);
+	}
+	return z;
+}
+
+
+void dsp_audio_generate_law_tables(void)
+{
+	int i;
+	for (i = 0; i < 256; i++)
+		dsp_audio_alaw_to_s32[i] = alaw2linear(reverse_bits(i));
+
+	for (i = 0; i < 256; i++)
+		dsp_audio_ulaw_to_s32[i] = ulaw2linear(reverse_bits(i));
+
+	for (i = 0; i < 256; i++) {
+		dsp_audio_alaw_to_ulaw[i] =
+			linear2ulaw(dsp_audio_alaw_to_s32[i]);
+		dsp_audio_ulaw_to_alaw[i] =
+			linear2alaw(dsp_audio_ulaw_to_s32[i]);
+	}
+}
+
+void
+dsp_audio_generate_s2law_table(void)
+{
+	int i;
+
+	if (dsp_options & DSP_OPT_ULAW) {
+		/* generating ulaw-table */
+		for (i = -32768; i < 32768; i++) {
+			dsp_audio_s16_to_law[i & 0xffff] =
+				reverse_bits(linear2ulaw(i));
+		}
+	} else {
+		/* generating alaw-table */
+		for (i = -32768; i < 32768; i++) {
+			dsp_audio_s16_to_law[i & 0xffff] =
+				reverse_bits(linear2alaw(i));
+		}
+	}
+}
+
+
+/*
+ * the seven bit sample is the number of every second alaw-sample ordered by
+ * aplitude. 0x00 is negative, 0x7f is positive amplitude.
+ */
+u8 dsp_audio_seven2law[128];
+u8 dsp_audio_law2seven[256];
+
+/********************************************************************
+ * generate table for conversion law from/to 7-bit alaw-like sample *
+ ********************************************************************/
+
+void
+dsp_audio_generate_seven(void)
+{
+	int i, j, k;
+	u8 spl;
+	u8 sorted_alaw[256];
+
+	/* generate alaw table, sorted by the linear value */
+	for (i = 0; i < 256; i++) {
+		j = 0;
+		for (k = 0; k < 256; k++) {
+			if (dsp_audio_alaw_to_s32[k]
+				< dsp_audio_alaw_to_s32[i]) {
+			j++;
+			}
+		}
+		sorted_alaw[j] = i;
+	}
+
+	/* generate tabels */
+	for (i = 0; i < 256; i++) {
+		/* spl is the source: the law-sample (converted to alaw) */
+		spl = i;
+		if (dsp_options & DSP_OPT_ULAW)
+			spl = dsp_audio_ulaw_to_alaw[i];
+		/* find the 7-bit-sample */
+		for (j = 0; j < 256; j++) {
+			if (sorted_alaw[j] == spl)
+				break;
+		}
+		/* write 7-bit audio value */
+		dsp_audio_law2seven[i] = j >> 1;
+	}
+	for (i = 0; i < 128; i++) {
+		spl = sorted_alaw[i << 1];
+		if (dsp_options & DSP_OPT_ULAW)
+			spl = dsp_audio_alaw_to_ulaw[spl];
+		dsp_audio_seven2law[i] = spl;
+	}
+}
+
+
+/* mix 2*law -> law */
+u8 dsp_audio_mix_law[65536];
+
+/******************************************************
+ * generate mix table to mix two law samples into one *
+ ******************************************************/
+
+void
+dsp_audio_generate_mix_table(void)
+{
+	int i, j;
+	s32 sample;
+
+	i = 0;
+	while (i < 256) {
+		j = 0;
+		while (j < 256) {
+			sample = dsp_audio_law_to_s32[i];
+			sample += dsp_audio_law_to_s32[j];
+			if (sample > 32767)
+				sample = 32767;
+			if (sample < -32768)
+				sample = -32768;
+			dsp_audio_mix_law[(i<<8)|j] =
+				dsp_audio_s16_to_law[sample & 0xffff];
+			j++;
+		}
+		i++;
+	}
+}
+
+
+/*************************************
+ * generate different volume changes *
+ *************************************/
+
+static u8 dsp_audio_reduce8[256];
+static u8 dsp_audio_reduce7[256];
+static u8 dsp_audio_reduce6[256];
+static u8 dsp_audio_reduce5[256];
+static u8 dsp_audio_reduce4[256];
+static u8 dsp_audio_reduce3[256];
+static u8 dsp_audio_reduce2[256];
+static u8 dsp_audio_reduce1[256];
+static u8 dsp_audio_increase1[256];
+static u8 dsp_audio_increase2[256];
+static u8 dsp_audio_increase3[256];
+static u8 dsp_audio_increase4[256];
+static u8 dsp_audio_increase5[256];
+static u8 dsp_audio_increase6[256];
+static u8 dsp_audio_increase7[256];
+static u8 dsp_audio_increase8[256];
+
+static u8 *dsp_audio_volume_change[16] = {
+	dsp_audio_reduce8,
+	dsp_audio_reduce7,
+	dsp_audio_reduce6,
+	dsp_audio_reduce5,
+	dsp_audio_reduce4,
+	dsp_audio_reduce3,
+	dsp_audio_reduce2,
+	dsp_audio_reduce1,
+	dsp_audio_increase1,
+	dsp_audio_increase2,
+	dsp_audio_increase3,
+	dsp_audio_increase4,
+	dsp_audio_increase5,
+	dsp_audio_increase6,
+	dsp_audio_increase7,
+	dsp_audio_increase8,
+};
+
+void
+dsp_audio_generate_volume_changes(void)
+{
+	register s32 sample;
+	int i;
+	int num[]   = { 110, 125, 150, 175, 200, 300, 400, 500 };
+	int denum[] = { 100, 100, 100, 100, 100, 100, 100, 100 };
+
+	i = 0;
+	while (i < 256) {
+		dsp_audio_reduce8[i] = dsp_audio_s16_to_law[
+			(dsp_audio_law_to_s32[i] * denum[7] / num[7]) & 0xffff];
+		dsp_audio_reduce7[i] = dsp_audio_s16_to_law[
+			(dsp_audio_law_to_s32[i] * denum[6] / num[6]) & 0xffff];
+		dsp_audio_reduce6[i] = dsp_audio_s16_to_law[
+			(dsp_audio_law_to_s32[i] * denum[5] / num[5]) & 0xffff];
+		dsp_audio_reduce5[i] = dsp_audio_s16_to_law[
+			(dsp_audio_law_to_s32[i] * denum[4] / num[4]) & 0xffff];
+		dsp_audio_reduce4[i] = dsp_audio_s16_to_law[
+			(dsp_audio_law_to_s32[i] * denum[3] / num[3]) & 0xffff];
+		dsp_audio_reduce3[i] = dsp_audio_s16_to_law[
+			(dsp_audio_law_to_s32[i] * denum[2] / num[2]) & 0xffff];
+		dsp_audio_reduce2[i] = dsp_audio_s16_to_law[
+			(dsp_audio_law_to_s32[i] * denum[1] / num[1]) & 0xffff];
+		dsp_audio_reduce1[i] = dsp_audio_s16_to_law[
+			(dsp_audio_law_to_s32[i] * denum[0] / num[0]) & 0xffff];
+		sample = dsp_audio_law_to_s32[i] * num[0] / denum[0];
+		if (sample < -32768)
+			sample = -32768;
+		else if (sample > 32767)
+			sample = 32767;
+		dsp_audio_increase1[i] = dsp_audio_s16_to_law[sample & 0xffff];
+		sample = dsp_audio_law_to_s32[i] * num[1] / denum[1];
+		if (sample < -32768)
+			sample = -32768;
+		else if (sample > 32767)
+			sample = 32767;
+		dsp_audio_increase2[i] = dsp_audio_s16_to_law[sample & 0xffff];
+		sample = dsp_audio_law_to_s32[i] * num[2] / denum[2];
+		if (sample < -32768)
+			sample = -32768;
+		else if (sample > 32767)
+			sample = 32767;
+		dsp_audio_increase3[i] = dsp_audio_s16_to_law[sample & 0xffff];
+		sample = dsp_audio_law_to_s32[i] * num[3] / denum[3];
+		if (sample < -32768)
+			sample = -32768;
+		else if (sample > 32767)
+			sample = 32767;
+		dsp_audio_increase4[i] = dsp_audio_s16_to_law[sample & 0xffff];
+		sample = dsp_audio_law_to_s32[i] * num[4] / denum[4];
+		if (sample < -32768)
+			sample = -32768;
+		else if (sample > 32767)
+			sample = 32767;
+		dsp_audio_increase5[i] = dsp_audio_s16_to_law[sample & 0xffff];
+		sample = dsp_audio_law_to_s32[i] * num[5] / denum[5];
+		if (sample < -32768)
+			sample = -32768;
+		else if (sample > 32767)
+			sample = 32767;
+		dsp_audio_increase6[i] = dsp_audio_s16_to_law[sample & 0xffff];
+		sample = dsp_audio_law_to_s32[i] * num[6] / denum[6];
+		if (sample < -32768)
+			sample = -32768;
+		else if (sample > 32767)
+			sample = 32767;
+		dsp_audio_increase7[i] = dsp_audio_s16_to_law[sample & 0xffff];
+		sample = dsp_audio_law_to_s32[i] * num[7] / denum[7];
+		if (sample < -32768)
+			sample = -32768;
+		else if (sample > 32767)
+			sample = 32767;
+		dsp_audio_increase8[i] = dsp_audio_s16_to_law[sample & 0xffff];
+
+		i++;
+	}
+}
+
+
+/**************************************
+ * change the volume of the given skb *
+ **************************************/
+
+/* this is a helper function for changing volume of skb. the range may be
+ * -8 to 8, which is a shift to the power of 2. 0 == no volume, 3 == volume*8
+ */
+void
+dsp_change_volume(struct sk_buff *skb, int volume)
+{
+	u8 *volume_change;
+	int i, ii;
+	u8 *p;
+	int shift;
+
+	if (volume == 0)
+		return;
+
+	/* get correct conversion table */
+	if (volume < 0) {
+		shift = volume + 8;
+		if (shift < 0)
+			shift = 0;
+	} else {
+		shift = volume + 7;
+		if (shift > 15)
+			shift = 15;
+	}
+	volume_change = dsp_audio_volume_change[shift];
+	i = 0;
+	ii = skb->len;
+	p = skb->data;
+	/* change volume */
+	while (i < ii) {
+		*p = volume_change[*p];
+		p++;
+		i++;
+	}
+}
+
diff --git a/drivers/isdn/mISDN/dsp_biquad.h b/drivers/isdn/mISDN/dsp_biquad.h
new file mode 100644
index 000000000000..038191bc45f5
--- /dev/null
+++ b/drivers/isdn/mISDN/dsp_biquad.h
@@ -0,0 +1,65 @@
+/*
+ * SpanDSP - a series of DSP components for telephony
+ *
+ * biquad.h - General telephony bi-quad section routines (currently this just
+ *            handles canonic/type 2 form)
+ *
+ * Written by Steve Underwood <steveu@coppice.org>
+ *
+ * Copyright (C) 2001 Steve Underwood
+ *
+ * All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ *
+ */
+
+struct biquad2_state {
+	int32_t gain;
+	int32_t a1;
+	int32_t a2;
+	int32_t b1;
+	int32_t b2;
+
+	int32_t z1;
+	int32_t z2;
+};
+
+static inline void biquad2_init(struct biquad2_state *bq,
+    int32_t gain, int32_t a1, int32_t a2, int32_t b1, int32_t b2)
+{
+	bq->gain = gain;
+	bq->a1 = a1;
+	bq->a2 = a2;
+	bq->b1 = b1;
+	bq->b2 = b2;
+
+	bq->z1 = 0;
+	bq->z2 = 0;
+}
+
+static inline int16_t biquad2(struct biquad2_state *bq, int16_t sample)
+{
+	int32_t y;
+	int32_t z0;
+
+	z0 = sample*bq->gain + bq->z1*bq->a1 + bq->z2*bq->a2;
+	y = z0 + bq->z1*bq->b1 + bq->z2*bq->b2;
+
+	bq->z2 = bq->z1;
+	bq->z1 = z0 >> 15;
+	y >>= 15;
+	return  y;
+}
diff --git a/drivers/isdn/mISDN/dsp_blowfish.c b/drivers/isdn/mISDN/dsp_blowfish.c
new file mode 100644
index 000000000000..18e411e95bba
--- /dev/null
+++ b/drivers/isdn/mISDN/dsp_blowfish.c
@@ -0,0 +1,672 @@
+/*
+ * Blowfish encryption/decryption for mISDN_dsp.
+ *
+ * Copyright Andreas Eversberg (jolly@eversberg.eu)
+ *
+ * This software may be used and distributed according to the terms
+ * of the GNU General Public License, incorporated herein by reference.
+ *
+ */
+
+#include <linux/mISDNif.h>
+#include <linux/mISDNdsp.h>
+#include "core.h"
+#include "dsp.h"
+
+/*
+ * how to encode a sample stream to 64-bit blocks that will be encryped
+ *
+ * first of all, data is collected until a block of 9 samples are received.
+ * of course, a packet may have much more than 9 sample, but is may have
+ * not excacly the multiple of 9 samples. if there is a rest, the next
+ * received data will complete the block.
+ *
+ * the block is then converted to 9 uLAW samples without the least sigificant
+ * bit. the result is a 7-bit encoded sample.
+ *
+ * the samples will be reoganised to form 8 bytes of data:
+ * (5(6) means: encoded sample no. 5, bit 6)
+ *
+ * 0(6) 0(5) 0(4) 0(3) 0(2) 0(1) 0(0) 1(6)
+ * 1(5) 1(4) 1(3) 1(2) 1(1) 1(0) 2(6) 2(5)
+ * 2(4) 2(3) 2(2) 2(1) 2(0) 3(6) 3(5) 3(4)
+ * 3(3) 3(2) 3(1) 3(0) 4(6) 4(5) 4(4) 4(3)
+ * 4(2) 4(1) 4(0) 5(6) 5(5) 5(4) 5(3) 5(2)
+ * 5(1) 5(0) 6(6) 6(5) 6(4) 6(3) 6(2) 6(1)
+ * 6(0) 7(6) 7(5) 7(4) 7(3) 7(2) 7(1) 7(0)
+ * 8(6) 8(5) 8(4) 8(3) 8(2) 8(1) 8(0)
+ *
+ * the missing bit 0 of the last byte is filled with some
+ * random noise, to fill all 8 bytes.
+ *
+ * the 8 bytes will be encrypted using blowfish.
+ *
+ * the result will be converted into 9 bytes. the bit 7 is used for
+ * checksumme (CS) for sync (0, 1) and for the last bit:
+ * (5(6) means: crypted byte 5, bit 6)
+ *
+ * 1    0(7) 0(6) 0(5) 0(4) 0(3) 0(2) 0(1)
+ * 0    0(0) 1(7) 1(6) 1(5) 1(4) 1(3) 1(2)
+ * 0    1(1) 1(0) 2(7) 2(6) 2(5) 2(4) 2(3)
+ * 0    2(2) 2(1) 2(0) 3(7) 3(6) 3(5) 3(4)
+ * 0    3(3) 3(2) 3(1) 3(0) 4(7) 4(6) 4(5)
+ * CS   4(4) 4(3) 4(2) 4(1) 4(0) 5(7) 5(6)
+ * CS   5(5) 5(4) 5(3) 5(2) 5(1) 5(0) 6(7)
+ * CS   6(6) 6(5) 6(4) 6(3) 6(2) 6(1) 6(0)
+ * 7(0) 7(6) 7(5) 7(4) 7(3) 7(2) 7(1) 7(0)
+ *
+ * the checksum is used to detect transmission errors and frame drops.
+ *
+ * synchronisation of received block is done by shifting the upper bit of each
+ * byte (bit 7) to a shift register. if the rigister has the first five bits
+ * (10000), this is used to find the sync. only if sync has been found, the
+ * current block of 9 received bytes are decrypted. before that the check
+ * sum is calculated. if it is incorrect the block is dropped.
+ * this will avoid loud noise due to corrupt encrypted data.
+ *
+ * if the last block is corrupt, the current decoded block is repeated
+ * until a valid block has been received.
+ */
+
+/*
+ *  some blowfish parts are taken from the
+ * crypto-api for faster implementation
+ */
+
+struct bf_ctx {
+	u32 p[18];
+	u32 s[1024];
+};
+
+static const u32 bf_pbox[16 + 2] = {
+	0x243f6a88, 0x85a308d3, 0x13198a2e, 0x03707344,
+	0xa4093822, 0x299f31d0, 0x082efa98, 0xec4e6c89,
+	0x452821e6, 0x38d01377, 0xbe5466cf, 0x34e90c6c,
+	0xc0ac29b7, 0xc97c50dd, 0x3f84d5b5, 0xb5470917,
+	0x9216d5d9, 0x8979fb1b,
+};
+
+static const u32 bf_sbox[256 * 4] = {
+	0xd1310ba6, 0x98dfb5ac, 0x2ffd72db, 0xd01adfb7,
+	0xb8e1afed, 0x6a267e96, 0xba7c9045, 0xf12c7f99,
+	0x24a19947, 0xb3916cf7, 0x0801f2e2, 0x858efc16,
+	0x636920d8, 0x71574e69, 0xa458fea3, 0xf4933d7e,
+	0x0d95748f, 0x728eb658, 0x718bcd58, 0x82154aee,
+	0x7b54a41d, 0xc25a59b5, 0x9c30d539, 0x2af26013,
+	0xc5d1b023, 0x286085f0, 0xca417918, 0xb8db38ef,
+	0x8e79dcb0, 0x603a180e, 0x6c9e0e8b, 0xb01e8a3e,
+	0xd71577c1, 0xbd314b27, 0x78af2fda, 0x55605c60,
+	0xe65525f3, 0xaa55ab94, 0x57489862, 0x63e81440,
+	0x55ca396a, 0x2aab10b6, 0xb4cc5c34, 0x1141e8ce,
+	0xa15486af, 0x7c72e993, 0xb3ee1411, 0x636fbc2a,
+	0x2ba9c55d, 0x741831f6, 0xce5c3e16, 0x9b87931e,
+	0xafd6ba33, 0x6c24cf5c, 0x7a325381, 0x28958677,
+	0x3b8f4898, 0x6b4bb9af, 0xc4bfe81b, 0x66282193,
+	0x61d809cc, 0xfb21a991, 0x487cac60, 0x5dec8032,
+	0xef845d5d, 0xe98575b1, 0xdc262302, 0xeb651b88,
+	0x23893e81, 0xd396acc5, 0x0f6d6ff3, 0x83f44239,
+	0x2e0b4482, 0xa4842004, 0x69c8f04a, 0x9e1f9b5e,
+	0x21c66842, 0xf6e96c9a, 0x670c9c61, 0xabd388f0,
+	0x6a51a0d2, 0xd8542f68, 0x960fa728, 0xab5133a3,
+	0x6eef0b6c, 0x137a3be4, 0xba3bf050, 0x7efb2a98,
+	0xa1f1651d, 0x39af0176, 0x66ca593e, 0x82430e88,
+	0x8cee8619, 0x456f9fb4, 0x7d84a5c3, 0x3b8b5ebe,
+	0xe06f75d8, 0x85c12073, 0x401a449f, 0x56c16aa6,
+	0x4ed3aa62, 0x363f7706, 0x1bfedf72, 0x429b023d,
+	0x37d0d724, 0xd00a1248, 0xdb0fead3, 0x49f1c09b,
+	0x075372c9, 0x80991b7b, 0x25d479d8, 0xf6e8def7,
+	0xe3fe501a, 0xb6794c3b, 0x976ce0bd, 0x04c006ba,
+	0xc1a94fb6, 0x409f60c4, 0x5e5c9ec2, 0x196a2463,
+	0x68fb6faf, 0x3e6c53b5, 0x1339b2eb, 0x3b52ec6f,
+	0x6dfc511f, 0x9b30952c, 0xcc814544, 0xaf5ebd09,
+	0xbee3d004, 0xde334afd, 0x660f2807, 0x192e4bb3,
+	0xc0cba857, 0x45c8740f, 0xd20b5f39, 0xb9d3fbdb,
+	0x5579c0bd, 0x1a60320a, 0xd6a100c6, 0x402c7279,
+	0x679f25fe, 0xfb1fa3cc, 0x8ea5e9f8, 0xdb3222f8,
+	0x3c7516df, 0xfd616b15, 0x2f501ec8, 0xad0552ab,
+	0x323db5fa, 0xfd238760, 0x53317b48, 0x3e00df82,
+	0x9e5c57bb, 0xca6f8ca0, 0x1a87562e, 0xdf1769db,
+	0xd542a8f6, 0x287effc3, 0xac6732c6, 0x8c4f5573,
+	0x695b27b0, 0xbbca58c8, 0xe1ffa35d, 0xb8f011a0,
+	0x10fa3d98, 0xfd2183b8, 0x4afcb56c, 0x2dd1d35b,
+	0x9a53e479, 0xb6f84565, 0xd28e49bc, 0x4bfb9790,
+	0xe1ddf2da, 0xa4cb7e33, 0x62fb1341, 0xcee4c6e8,
+	0xef20cada, 0x36774c01, 0xd07e9efe, 0x2bf11fb4,
+	0x95dbda4d, 0xae909198, 0xeaad8e71, 0x6b93d5a0,
+	0xd08ed1d0, 0xafc725e0, 0x8e3c5b2f, 0x8e7594b7,
+	0x8ff6e2fb, 0xf2122b64, 0x8888b812, 0x900df01c,
+	0x4fad5ea0, 0x688fc31c, 0xd1cff191, 0xb3a8c1ad,
+	0x2f2f2218, 0xbe0e1777, 0xea752dfe, 0x8b021fa1,
+	0xe5a0cc0f, 0xb56f74e8, 0x18acf3d6, 0xce89e299,
+	0xb4a84fe0, 0xfd13e0b7, 0x7cc43b81, 0xd2ada8d9,
+	0x165fa266, 0x80957705, 0x93cc7314, 0x211a1477,
+	0xe6ad2065, 0x77b5fa86, 0xc75442f5, 0xfb9d35cf,
+	0xebcdaf0c, 0x7b3e89a0, 0xd6411bd3, 0xae1e7e49,
+	0x00250e2d, 0x2071b35e, 0x226800bb, 0x57b8e0af,
+	0x2464369b, 0xf009b91e, 0x5563911d, 0x59dfa6aa,
+	0x78c14389, 0xd95a537f, 0x207d5ba2, 0x02e5b9c5,
+	0x83260376, 0x6295cfa9, 0x11c81968, 0x4e734a41,
+	0xb3472dca, 0x7b14a94a, 0x1b510052, 0x9a532915,
+	0xd60f573f, 0xbc9bc6e4, 0x2b60a476, 0x81e67400,
+	0x08ba6fb5, 0x571be91f, 0xf296ec6b, 0x2a0dd915,
+	0xb6636521, 0xe7b9f9b6, 0xff34052e, 0xc5855664,
+	0x53b02d5d, 0xa99f8fa1, 0x08ba4799, 0x6e85076a,
+	0x4b7a70e9, 0xb5b32944, 0xdb75092e, 0xc4192623,
+	0xad6ea6b0, 0x49a7df7d, 0x9cee60b8, 0x8fedb266,
+	0xecaa8c71, 0x699a17ff, 0x5664526c, 0xc2b19ee1,
+	0x193602a5, 0x75094c29, 0xa0591340, 0xe4183a3e,
+	0x3f54989a, 0x5b429d65, 0x6b8fe4d6, 0x99f73fd6,
+	0xa1d29c07, 0xefe830f5, 0x4d2d38e6, 0xf0255dc1,
+	0x4cdd2086, 0x8470eb26, 0x6382e9c6, 0x021ecc5e,
+	0x09686b3f, 0x3ebaefc9, 0x3c971814, 0x6b6a70a1,
+	0x687f3584, 0x52a0e286, 0xb79c5305, 0xaa500737,
+	0x3e07841c, 0x7fdeae5c, 0x8e7d44ec, 0x5716f2b8,
+	0xb03ada37, 0xf0500c0d, 0xf01c1f04, 0x0200b3ff,
+	0xae0cf51a, 0x3cb574b2, 0x25837a58, 0xdc0921bd,
+	0xd19113f9, 0x7ca92ff6, 0x94324773, 0x22f54701,
+	0x3ae5e581, 0x37c2dadc, 0xc8b57634, 0x9af3dda7,
+	0xa9446146, 0x0fd0030e, 0xecc8c73e, 0xa4751e41,
+	0xe238cd99, 0x3bea0e2f, 0x3280bba1, 0x183eb331,
+	0x4e548b38, 0x4f6db908, 0x6f420d03, 0xf60a04bf,
+	0x2cb81290, 0x24977c79, 0x5679b072, 0xbcaf89af,
+	0xde9a771f, 0xd9930810, 0xb38bae12, 0xdccf3f2e,
+	0x5512721f, 0x2e6b7124, 0x501adde6, 0x9f84cd87,
+	0x7a584718, 0x7408da17, 0xbc9f9abc, 0xe94b7d8c,
+	0xec7aec3a, 0xdb851dfa, 0x63094366, 0xc464c3d2,
+	0xef1c1847, 0x3215d908, 0xdd433b37, 0x24c2ba16,
+	0x12a14d43, 0x2a65c451, 0x50940002, 0x133ae4dd,
+	0x71dff89e, 0x10314e55, 0x81ac77d6, 0x5f11199b,
+	0x043556f1, 0xd7a3c76b, 0x3c11183b, 0x5924a509,
+	0xf28fe6ed, 0x97f1fbfa, 0x9ebabf2c, 0x1e153c6e,
+	0x86e34570, 0xeae96fb1, 0x860e5e0a, 0x5a3e2ab3,
+	0x771fe71c, 0x4e3d06fa, 0x2965dcb9, 0x99e71d0f,
+	0x803e89d6, 0x5266c825, 0x2e4cc978, 0x9c10b36a,
+	0xc6150eba, 0x94e2ea78, 0xa5fc3c53, 0x1e0a2df4,
+	0xf2f74ea7, 0x361d2b3d, 0x1939260f, 0x19c27960,
+	0x5223a708, 0xf71312b6, 0xebadfe6e, 0xeac31f66,
+	0xe3bc4595, 0xa67bc883, 0xb17f37d1, 0x018cff28,
+	0xc332ddef, 0xbe6c5aa5, 0x65582185, 0x68ab9802,
+	0xeecea50f, 0xdb2f953b, 0x2aef7dad, 0x5b6e2f84,
+	0x1521b628, 0x29076170, 0xecdd4775, 0x619f1510,
+	0x13cca830, 0xeb61bd96, 0x0334fe1e, 0xaa0363cf,
+	0xb5735c90, 0x4c70a239, 0xd59e9e0b, 0xcbaade14,
+	0xeecc86bc, 0x60622ca7, 0x9cab5cab, 0xb2f3846e,
+	0x648b1eaf, 0x19bdf0ca, 0xa02369b9, 0x655abb50,
+	0x40685a32, 0x3c2ab4b3, 0x319ee9d5, 0xc021b8f7,
+	0x9b540b19, 0x875fa099, 0x95f7997e, 0x623d7da8,
+	0xf837889a, 0x97e32d77, 0x11ed935f, 0x16681281,
+	0x0e358829, 0xc7e61fd6, 0x96dedfa1, 0x7858ba99,
+	0x57f584a5, 0x1b227263, 0x9b83c3ff, 0x1ac24696,
+	0xcdb30aeb, 0x532e3054, 0x8fd948e4, 0x6dbc3128,
+	0x58ebf2ef, 0x34c6ffea, 0xfe28ed61, 0xee7c3c73,
+	0x5d4a14d9, 0xe864b7e3, 0x42105d14, 0x203e13e0,
+	0x45eee2b6, 0xa3aaabea, 0xdb6c4f15, 0xfacb4fd0,
+	0xc742f442, 0xef6abbb5, 0x654f3b1d, 0x41cd2105,
+	0xd81e799e, 0x86854dc7, 0xe44b476a, 0x3d816250,
+	0xcf62a1f2, 0x5b8d2646, 0xfc8883a0, 0xc1c7b6a3,
+	0x7f1524c3, 0x69cb7492, 0x47848a0b, 0x5692b285,
+	0x095bbf00, 0xad19489d, 0x1462b174, 0x23820e00,
+	0x58428d2a, 0x0c55f5ea, 0x1dadf43e, 0x233f7061,
+	0x3372f092, 0x8d937e41, 0xd65fecf1, 0x6c223bdb,
+	0x7cde3759, 0xcbee7460, 0x4085f2a7, 0xce77326e,
+	0xa6078084, 0x19f8509e, 0xe8efd855, 0x61d99735,
+	0xa969a7aa, 0xc50c06c2, 0x5a04abfc, 0x800bcadc,
+	0x9e447a2e, 0xc3453484, 0xfdd56705, 0x0e1e9ec9,
+	0xdb73dbd3, 0x105588cd, 0x675fda79, 0xe3674340,
+	0xc5c43465, 0x713e38d8, 0x3d28f89e, 0xf16dff20,
+	0x153e21e7, 0x8fb03d4a, 0xe6e39f2b, 0xdb83adf7,
+	0xe93d5a68, 0x948140f7, 0xf64c261c, 0x94692934,
+	0x411520f7, 0x7602d4f7, 0xbcf46b2e, 0xd4a20068,
+	0xd4082471, 0x3320f46a, 0x43b7d4b7, 0x500061af,
+	0x1e39f62e, 0x97244546, 0x14214f74, 0xbf8b8840,
+	0x4d95fc1d, 0x96b591af, 0x70f4ddd3, 0x66a02f45,
+	0xbfbc09ec, 0x03bd9785, 0x7fac6dd0, 0x31cb8504,
+	0x96eb27b3, 0x55fd3941, 0xda2547e6, 0xabca0a9a,
+	0x28507825, 0x530429f4, 0x0a2c86da, 0xe9b66dfb,
+	0x68dc1462, 0xd7486900, 0x680ec0a4, 0x27a18dee,
+	0x4f3ffea2, 0xe887ad8c, 0xb58ce006, 0x7af4d6b6,
+	0xaace1e7c, 0xd3375fec, 0xce78a399, 0x406b2a42,
+	0x20fe9e35, 0xd9f385b9, 0xee39d7ab, 0x3b124e8b,
+	0x1dc9faf7, 0x4b6d1856, 0x26a36631, 0xeae397b2,
+	0x3a6efa74, 0xdd5b4332, 0x6841e7f7, 0xca7820fb,
+	0xfb0af54e, 0xd8feb397, 0x454056ac, 0xba489527,
+	0x55533a3a, 0x20838d87, 0xfe6ba9b7, 0xd096954b,
+	0x55a867bc, 0xa1159a58, 0xcca92963, 0x99e1db33,
+	0xa62a4a56, 0x3f3125f9, 0x5ef47e1c, 0x9029317c,
+	0xfdf8e802, 0x04272f70, 0x80bb155c, 0x05282ce3,
+	0x95c11548, 0xe4c66d22, 0x48c1133f, 0xc70f86dc,
+	0x07f9c9ee, 0x41041f0f, 0x404779a4, 0x5d886e17,
+	0x325f51eb, 0xd59bc0d1, 0xf2bcc18f, 0x41113564,
+	0x257b7834, 0x602a9c60, 0xdff8e8a3, 0x1f636c1b,
+	0x0e12b4c2, 0x02e1329e, 0xaf664fd1, 0xcad18115,
+	0x6b2395e0, 0x333e92e1, 0x3b240b62, 0xeebeb922,
+	0x85b2a20e, 0xe6ba0d99, 0xde720c8c, 0x2da2f728,
+	0xd0127845, 0x95b794fd, 0x647d0862, 0xe7ccf5f0,
+	0x5449a36f, 0x877d48fa, 0xc39dfd27, 0xf33e8d1e,
+	0x0a476341, 0x992eff74, 0x3a6f6eab, 0xf4f8fd37,
+	0xa812dc60, 0xa1ebddf8, 0x991be14c, 0xdb6e6b0d,
+	0xc67b5510, 0x6d672c37, 0x2765d43b, 0xdcd0e804,
+	0xf1290dc7, 0xcc00ffa3, 0xb5390f92, 0x690fed0b,
+	0x667b9ffb, 0xcedb7d9c, 0xa091cf0b, 0xd9155ea3,
+	0xbb132f88, 0x515bad24, 0x7b9479bf, 0x763bd6eb,
+	0x37392eb3, 0xcc115979, 0x8026e297, 0xf42e312d,
+	0x6842ada7, 0xc66a2b3b, 0x12754ccc, 0x782ef11c,
+	0x6a124237, 0xb79251e7, 0x06a1bbe6, 0x4bfb6350,
+	0x1a6b1018, 0x11caedfa, 0x3d25bdd8, 0xe2e1c3c9,
+	0x44421659, 0x0a121386, 0xd90cec6e, 0xd5abea2a,
+	0x64af674e, 0xda86a85f, 0xbebfe988, 0x64e4c3fe,
+	0x9dbc8057, 0xf0f7c086, 0x60787bf8, 0x6003604d,
+	0xd1fd8346, 0xf6381fb0, 0x7745ae04, 0xd736fccc,
+	0x83426b33, 0xf01eab71, 0xb0804187, 0x3c005e5f,
+	0x77a057be, 0xbde8ae24, 0x55464299, 0xbf582e61,
+	0x4e58f48f, 0xf2ddfda2, 0xf474ef38, 0x8789bdc2,
+	0x5366f9c3, 0xc8b38e74, 0xb475f255, 0x46fcd9b9,
+	0x7aeb2661, 0x8b1ddf84, 0x846a0e79, 0x915f95e2,
+	0x466e598e, 0x20b45770, 0x8cd55591, 0xc902de4c,
+	0xb90bace1, 0xbb8205d0, 0x11a86248, 0x7574a99e,
+	0xb77f19b6, 0xe0a9dc09, 0x662d09a1, 0xc4324633,
+	0xe85a1f02, 0x09f0be8c, 0x4a99a025, 0x1d6efe10,
+	0x1ab93d1d, 0x0ba5a4df, 0xa186f20f, 0x2868f169,
+	0xdcb7da83, 0x573906fe, 0xa1e2ce9b, 0x4fcd7f52,
+	0x50115e01, 0xa70683fa, 0xa002b5c4, 0x0de6d027,
+	0x9af88c27, 0x773f8641, 0xc3604c06, 0x61a806b5,
+	0xf0177a28, 0xc0f586e0, 0x006058aa, 0x30dc7d62,
+	0x11e69ed7, 0x2338ea63, 0x53c2dd94, 0xc2c21634,
+	0xbbcbee56, 0x90bcb6de, 0xebfc7da1, 0xce591d76,
+	0x6f05e409, 0x4b7c0188, 0x39720a3d, 0x7c927c24,
+	0x86e3725f, 0x724d9db9, 0x1ac15bb4, 0xd39eb8fc,
+	0xed545578, 0x08fca5b5, 0xd83d7cd3, 0x4dad0fc4,
+	0x1e50ef5e, 0xb161e6f8, 0xa28514d9, 0x6c51133c,
+	0x6fd5c7e7, 0x56e14ec4, 0x362abfce, 0xddc6c837,
+	0xd79a3234, 0x92638212, 0x670efa8e, 0x406000e0,
+	0x3a39ce37, 0xd3faf5cf, 0xabc27737, 0x5ac52d1b,
+	0x5cb0679e, 0x4fa33742, 0xd3822740, 0x99bc9bbe,
+	0xd5118e9d, 0xbf0f7315, 0xd62d1c7e, 0xc700c47b,
+	0xb78c1b6b, 0x21a19045, 0xb26eb1be, 0x6a366eb4,
+	0x5748ab2f, 0xbc946e79, 0xc6a376d2, 0x6549c2c8,
+	0x530ff8ee, 0x468dde7d, 0xd5730a1d, 0x4cd04dc6,
+	0x2939bbdb, 0xa9ba4650, 0xac9526e8, 0xbe5ee304,
+	0xa1fad5f0, 0x6a2d519a, 0x63ef8ce2, 0x9a86ee22,
+	0xc089c2b8, 0x43242ef6, 0xa51e03aa, 0x9cf2d0a4,
+	0x83c061ba, 0x9be96a4d, 0x8fe51550, 0xba645bd6,
+	0x2826a2f9, 0xa73a3ae1, 0x4ba99586, 0xef5562e9,
+	0xc72fefd3, 0xf752f7da, 0x3f046f69, 0x77fa0a59,
+	0x80e4a915, 0x87b08601, 0x9b09e6ad, 0x3b3ee593,
+	0xe990fd5a, 0x9e34d797, 0x2cf0b7d9, 0x022b8b51,
+	0x96d5ac3a, 0x017da67d, 0xd1cf3ed6, 0x7c7d2d28,
+	0x1f9f25cf, 0xadf2b89b, 0x5ad6b472, 0x5a88f54c,
+	0xe029ac71, 0xe019a5e6, 0x47b0acfd, 0xed93fa9b,
+	0xe8d3c48d, 0x283b57cc, 0xf8d56629, 0x79132e28,
+	0x785f0191, 0xed756055, 0xf7960e44, 0xe3d35e8c,
+	0x15056dd4, 0x88f46dba, 0x03a16125, 0x0564f0bd,
+	0xc3eb9e15, 0x3c9057a2, 0x97271aec, 0xa93a072a,
+	0x1b3f6d9b, 0x1e6321f5, 0xf59c66fb, 0x26dcf319,
+	0x7533d928, 0xb155fdf5, 0x03563482, 0x8aba3cbb,
+	0x28517711, 0xc20ad9f8, 0xabcc5167, 0xccad925f,
+	0x4de81751, 0x3830dc8e, 0x379d5862, 0x9320f991,
+	0xea7a90c2, 0xfb3e7bce, 0x5121ce64, 0x774fbe32,
+	0xa8b6e37e, 0xc3293d46, 0x48de5369, 0x6413e680,
+	0xa2ae0810, 0xdd6db224, 0x69852dfd, 0x09072166,
+	0xb39a460a, 0x6445c0dd, 0x586cdecf, 0x1c20c8ae,
+	0x5bbef7dd, 0x1b588d40, 0xccd2017f, 0x6bb4e3bb,
+	0xdda26a7e, 0x3a59ff45, 0x3e350a44, 0xbcb4cdd5,
+	0x72eacea8, 0xfa6484bb, 0x8d6612ae, 0xbf3c6f47,
+	0xd29be463, 0x542f5d9e, 0xaec2771b, 0xf64e6370,
+	0x740e0d8d, 0xe75b1357, 0xf8721671, 0xaf537d5d,
+	0x4040cb08, 0x4eb4e2cc, 0x34d2466a, 0x0115af84,
+	0xe1b00428, 0x95983a1d, 0x06b89fb4, 0xce6ea048,
+	0x6f3f3b82, 0x3520ab82, 0x011a1d4b, 0x277227f8,
+	0x611560b1, 0xe7933fdc, 0xbb3a792b, 0x344525bd,
+	0xa08839e1, 0x51ce794b, 0x2f32c9b7, 0xa01fbac9,
+	0xe01cc87e, 0xbcc7d1f6, 0xcf0111c3, 0xa1e8aac7,
+	0x1a908749, 0xd44fbd9a, 0xd0dadecb, 0xd50ada38,
+	0x0339c32a, 0xc6913667, 0x8df9317c, 0xe0b12b4f,
+	0xf79e59b7, 0x43f5bb3a, 0xf2d519ff, 0x27d9459c,
+	0xbf97222c, 0x15e6fc2a, 0x0f91fc71, 0x9b941525,
+	0xfae59361, 0xceb69ceb, 0xc2a86459, 0x12baa8d1,
+	0xb6c1075e, 0xe3056a0c, 0x10d25065, 0xcb03a442,
+	0xe0ec6e0e, 0x1698db3b, 0x4c98a0be, 0x3278e964,
+	0x9f1f9532, 0xe0d392df, 0xd3a0342b, 0x8971f21e,
+	0x1b0a7441, 0x4ba3348c, 0xc5be7120, 0xc37632d8,
+	0xdf359f8d, 0x9b992f2e, 0xe60b6f47, 0x0fe3f11d,
+	0xe54cda54, 0x1edad891, 0xce6279cf, 0xcd3e7e6f,
+	0x1618b166, 0xfd2c1d05, 0x848fd2c5, 0xf6fb2299,
+	0xf523f357, 0xa6327623, 0x93a83531, 0x56cccd02,
+	0xacf08162, 0x5a75ebb5, 0x6e163697, 0x88d273cc,
+	0xde966292, 0x81b949d0, 0x4c50901b, 0x71c65614,
+	0xe6c6c7bd, 0x327a140a, 0x45e1d006, 0xc3f27b9a,
+	0xc9aa53fd, 0x62a80f00, 0xbb25bfe2, 0x35bdd2f6,
+	0x71126905, 0xb2040222, 0xb6cbcf7c, 0xcd769c2b,
+	0x53113ec0, 0x1640e3d3, 0x38abbd60, 0x2547adf0,
+	0xba38209c, 0xf746ce76, 0x77afa1c5, 0x20756060,
+	0x85cbfe4e, 0x8ae88dd8, 0x7aaaf9b0, 0x4cf9aa7e,
+	0x1948c25c, 0x02fb8a8c, 0x01c36ae4, 0xd6ebe1f9,
+	0x90d4f869, 0xa65cdea0, 0x3f09252d, 0xc208e69f,
+	0xb74e6132, 0xce77e25b, 0x578fdfe3, 0x3ac372e6,
+};
+
+/*
+ * Round loop unrolling macros, S is a pointer to a S-Box array
+ * organized in 4 unsigned longs at a row.
+ */
+#define GET32_3(x) (((x) & 0xff))
+#define GET32_2(x) (((x) >> (8)) & (0xff))
+#define GET32_1(x) (((x) >> (16)) & (0xff))
+#define GET32_0(x) (((x) >> (24)) & (0xff))
+
+#define bf_F(x) (((S[GET32_0(x)] + S[256 + GET32_1(x)]) ^ \
+    S[512 + GET32_2(x)]) + S[768 + GET32_3(x)])
+
+#define EROUND(a, b, n)  do { b ^= P[n]; a ^= bf_F(b); } while (0)
+#define DROUND(a, b, n)  do { a ^= bf_F(b); b ^= P[n]; } while (0)
+
+
+/*
+ * encrypt isdn data frame
+ * every block with 9 samples is encrypted
+ */
+void
+dsp_bf_encrypt(struct dsp *dsp, u8 *data, int len)
+{
+	int i = 0, j = dsp->bf_crypt_pos;
+	u8 *bf_data_in = dsp->bf_data_in;
+	u8 *bf_crypt_out = dsp->bf_crypt_out;
+	u32 *P = dsp->bf_p;
+	u32 *S = dsp->bf_s;
+	u32 yl, yr;
+	u32 cs;
+	u8 nibble;
+
+	while (i < len) {
+		/* collect a block of 9 samples */
+		if (j < 9) {
+			bf_data_in[j] = *data;
+			*data++ = bf_crypt_out[j++];
+			i++;
+			continue;
+		}
+		j = 0;
+		/* transcode 9 samples xlaw to 8 bytes */
+		yl = dsp_audio_law2seven[bf_data_in[0]];
+		yl = (yl<<7) | dsp_audio_law2seven[bf_data_in[1]];
+		yl = (yl<<7) | dsp_audio_law2seven[bf_data_in[2]];
+		yl = (yl<<7) | dsp_audio_law2seven[bf_data_in[3]];
+		nibble = dsp_audio_law2seven[bf_data_in[4]];
+		yr = nibble;
+		yl = (yl<<4) | (nibble>>3);
+		yr = (yr<<7) | dsp_audio_law2seven[bf_data_in[5]];
+		yr = (yr<<7) | dsp_audio_law2seven[bf_data_in[6]];
+		yr = (yr<<7) | dsp_audio_law2seven[bf_data_in[7]];
+		yr = (yr<<7) | dsp_audio_law2seven[bf_data_in[8]];
+		yr = (yr<<1) | (bf_data_in[0] & 1);
+
+		/* fill unused bit with random noise of audio input */
+		/* encrypt */
+
+		EROUND(yr, yl, 0);
+		EROUND(yl, yr, 1);
+		EROUND(yr, yl, 2);
+		EROUND(yl, yr, 3);
+		EROUND(yr, yl, 4);
+		EROUND(yl, yr, 5);
+		EROUND(yr, yl, 6);
+		EROUND(yl, yr, 7);
+		EROUND(yr, yl, 8);
+		EROUND(yl, yr, 9);
+		EROUND(yr, yl, 10);
+		EROUND(yl, yr, 11);
+		EROUND(yr, yl, 12);
+		EROUND(yl, yr, 13);
+		EROUND(yr, yl, 14);
+		EROUND(yl, yr, 15);
+		yl ^= P[16];
+		yr ^= P[17];
+
+		/* calculate 3-bit checksumme */
+		cs = yl ^ (yl>>3) ^ (yl>>6) ^ (yl>>9) ^ (yl>>12) ^ (yl>>15)
+			^ (yl>>18) ^ (yl>>21) ^ (yl>>24) ^ (yl>>27) ^ (yl>>30)
+			^ (yr<<2) ^ (yr>>1) ^ (yr>>4) ^ (yr>>7) ^ (yr>>10)
+			^ (yr>>13) ^ (yr>>16) ^ (yr>>19) ^ (yr>>22) ^ (yr>>25)
+			^ (yr>>28) ^ (yr>>31);
+
+		/*
+		 * transcode 8 crypted bytes to 9 data bytes with sync
+		 * and checksum information
+		 */
+		bf_crypt_out[0] = (yl>>25) | 0x80;
+		bf_crypt_out[1] = (yl>>18) & 0x7f;
+		bf_crypt_out[2] = (yl>>11) & 0x7f;
+		bf_crypt_out[3] = (yl>>4) & 0x7f;
+		bf_crypt_out[4] = ((yl<<3) & 0x78) | ((yr>>29) & 0x07);
+		bf_crypt_out[5] = ((yr>>22) & 0x7f) | ((cs<<5) & 0x80);
+		bf_crypt_out[6] = ((yr>>15) & 0x7f) | ((cs<<6) & 0x80);
+		bf_crypt_out[7] = ((yr>>8) & 0x7f) | (cs<<7);
+		bf_crypt_out[8] = yr;
+	}
+
+	/* write current count */
+	dsp->bf_crypt_pos = j;
+
+}
+
+
+/*
+ * decrypt isdn data frame
+ * every block with 9 bytes is decrypted
+ */
+void
+dsp_bf_decrypt(struct dsp *dsp, u8 *data, int len)
+{
+	int i = 0;
+	u8 j = dsp->bf_decrypt_in_pos;
+	u8 k = dsp->bf_decrypt_out_pos;
+	u8 *bf_crypt_inring = dsp->bf_crypt_inring;
+	u8 *bf_data_out = dsp->bf_data_out;
+	u16 sync = dsp->bf_sync;
+	u32 *P = dsp->bf_p;
+	u32 *S = dsp->bf_s;
+	u32 yl, yr;
+	u8 nibble;
+	u8 cs, cs0, cs1, cs2;
+
+	while (i < len) {
+		/*
+		 * shift upper bit and rotate data to buffer ring
+		 * send current decrypted data
+		 */
+		sync = (sync<<1) | ((*data)>>7);
+		bf_crypt_inring[j++ & 15] = *data;
+		*data++ = bf_data_out[k++];
+		i++;
+		if (k == 9)
+			k = 0; /* repeat if no sync has been found */
+		/* check if not in sync */
+		if ((sync&0x1f0) != 0x100)
+			continue;
+		j -= 9;
+		/* transcode receive data to 64 bit block of encrypted data */
+		yl = bf_crypt_inring[j++ & 15];
+		yl = (yl<<7) | bf_crypt_inring[j++ & 15]; /* bit7 = 0 */
+		yl = (yl<<7) | bf_crypt_inring[j++ & 15]; /* bit7 = 0 */
+		yl = (yl<<7) | bf_crypt_inring[j++ & 15]; /* bit7 = 0 */
+		nibble = bf_crypt_inring[j++ & 15]; /* bit7 = 0 */
+		yr = nibble;
+		yl = (yl<<4) | (nibble>>3);
+		cs2 = bf_crypt_inring[j++ & 15];
+		yr = (yr<<7) | (cs2 & 0x7f);
+		cs1 = bf_crypt_inring[j++ & 15];
+		yr = (yr<<7) | (cs1 & 0x7f);
+		cs0 = bf_crypt_inring[j++ & 15];
+		yr = (yr<<7) | (cs0 & 0x7f);
+		yr = (yr<<8) | bf_crypt_inring[j++ & 15];
+
+		/* calculate 3-bit checksumme */
+		cs = yl ^ (yl>>3) ^ (yl>>6) ^ (yl>>9) ^ (yl>>12) ^ (yl>>15)
+			^ (yl>>18) ^ (yl>>21) ^ (yl>>24) ^ (yl>>27) ^ (yl>>30)
+			^ (yr<<2) ^ (yr>>1) ^ (yr>>4) ^ (yr>>7) ^ (yr>>10)
+			^ (yr>>13) ^ (yr>>16) ^ (yr>>19) ^ (yr>>22) ^ (yr>>25)
+			^ (yr>>28) ^ (yr>>31);
+
+		/* check if frame is valid */
+		if ((cs&0x7) != (((cs2>>5)&4) | ((cs1>>6)&2) | (cs0 >> 7))) {
+			if (dsp_debug & DEBUG_DSP_BLOWFISH)
+				printk(KERN_DEBUG
+				    "DSP BLOWFISH: received corrupt frame, "
+				    "checksumme is not correct\n");
+			continue;
+		}
+
+		/* decrypt */
+		yr ^= P[17];
+		yl ^= P[16];
+		DROUND(yl, yr, 15);
+		DROUND(yr, yl, 14);
+		DROUND(yl, yr, 13);
+		DROUND(yr, yl, 12);
+		DROUND(yl, yr, 11);
+		DROUND(yr, yl, 10);
+		DROUND(yl, yr, 9);
+		DROUND(yr, yl, 8);
+		DROUND(yl, yr, 7);
+		DROUND(yr, yl, 6);
+		DROUND(yl, yr, 5);
+		DROUND(yr, yl, 4);
+		DROUND(yl, yr, 3);
+		DROUND(yr, yl, 2);
+		DROUND(yl, yr, 1);
+		DROUND(yr, yl, 0);
+
+		/* transcode 8 crypted bytes to 9 sample bytes */
+		bf_data_out[0] = dsp_audio_seven2law[(yl>>25) & 0x7f];
+		bf_data_out[1] = dsp_audio_seven2law[(yl>>18) & 0x7f];
+		bf_data_out[2] = dsp_audio_seven2law[(yl>>11) & 0x7f];
+		bf_data_out[3] = dsp_audio_seven2law[(yl>>4) & 0x7f];
+		bf_data_out[4] = dsp_audio_seven2law[((yl<<3) & 0x78) |
+		    ((yr>>29) & 0x07)];
+
+		bf_data_out[5] = dsp_audio_seven2law[(yr>>22) & 0x7f];
+		bf_data_out[6] = dsp_audio_seven2law[(yr>>15) & 0x7f];
+		bf_data_out[7] = dsp_audio_seven2law[(yr>>8) & 0x7f];
+		bf_data_out[8] = dsp_audio_seven2law[(yr>>1) & 0x7f];
+		k = 0; /* start with new decoded frame */
+	}
+
+	/* write current count and sync */
+	dsp->bf_decrypt_in_pos = j;
+	dsp->bf_decrypt_out_pos = k;
+	dsp->bf_sync = sync;
+}
+
+
+/* used to encrypt S and P boxes */
+static inline void
+encrypt_block(const u32 *P, const u32 *S, u32 *dst, u32 *src)
+{
+	u32 yl = src[0];
+	u32 yr = src[1];
+
+	EROUND(yr, yl, 0);
+	EROUND(yl, yr, 1);
+	EROUND(yr, yl, 2);
+	EROUND(yl, yr, 3);
+	EROUND(yr, yl, 4);
+	EROUND(yl, yr, 5);
+	EROUND(yr, yl, 6);
+	EROUND(yl, yr, 7);
+	EROUND(yr, yl, 8);
+	EROUND(yl, yr, 9);
+	EROUND(yr, yl, 10);
+	EROUND(yl, yr, 11);
+	EROUND(yr, yl, 12);
+	EROUND(yl, yr, 13);
+	EROUND(yr, yl, 14);
+	EROUND(yl, yr, 15);
+
+	yl ^= P[16];
+	yr ^= P[17];
+
+	dst[0] = yr;
+	dst[1] = yl;
+}
+
+/*
+ * initialize the dsp for encryption and decryption using the same key
+ * Calculates the blowfish S and P boxes for encryption and decryption.
+ * The margin of keylen must be 4-56 bytes.
+ * returns 0 if ok.
+ */
+int
+dsp_bf_init(struct dsp *dsp, const u8 *key, uint keylen)
+{
+	short i, j, count;
+	u32 data[2], temp;
+	u32 *P = (u32 *)dsp->bf_p;
+	u32 *S = (u32 *)dsp->bf_s;
+
+	if (keylen < 4 || keylen > 56)
+		return 1;
+
+	/* Set dsp states */
+	i = 0;
+	while (i < 9) {
+		dsp->bf_crypt_out[i] = 0xff;
+		dsp->bf_data_out[i] = dsp_silence;
+		i++;
+	}
+	dsp->bf_crypt_pos = 0;
+	dsp->bf_decrypt_in_pos = 0;
+	dsp->bf_decrypt_out_pos = 0;
+	dsp->bf_sync = 0x1ff;
+	dsp->bf_enable = 1;
+
+	/* Copy the initialization s-boxes */
+	for (i = 0, count = 0; i < 256; i++)
+		for (j = 0; j < 4; j++, count++)
+			S[count] = bf_sbox[count];
+
+	/* Set the p-boxes */
+	for (i = 0; i < 16 + 2; i++)
+		P[i] = bf_pbox[i];
+
+	/* Actual subkey generation */
+	for (j = 0, i = 0; i < 16 + 2; i++) {
+		temp = (((u32)key[j] << 24) |
+		    ((u32)key[(j + 1) % keylen] << 16) |
+		    ((u32)key[(j + 2) % keylen] << 8) |
+		    ((u32)key[(j + 3) % keylen]));
+
+		P[i] = P[i] ^ temp;
+		j = (j + 4) % keylen;
+	}
+
+	data[0] = 0x00000000;
+	data[1] = 0x00000000;
+
+	for (i = 0; i < 16 + 2; i += 2) {
+		encrypt_block(P, S, data, data);
+
+		P[i] = data[0];
+		P[i + 1] = data[1];
+	}
+
+	for (i = 0; i < 4; i++) {
+		for (j = 0, count = i * 256; j < 256; j += 2, count += 2) {
+			encrypt_block(P, S, data, data);
+
+			S[count] = data[0];
+			S[count + 1] = data[1];
+		}
+	}
+
+	return 0;
+}
+
+
+/*
+ * turn encryption off
+ */
+void
+dsp_bf_cleanup(struct dsp *dsp)
+{
+	dsp->bf_enable = 0;
+}
diff --git a/drivers/isdn/mISDN/dsp_cmx.c b/drivers/isdn/mISDN/dsp_cmx.c
new file mode 100644
index 000000000000..e92b1ba4b45e
--- /dev/null
+++ b/drivers/isdn/mISDN/dsp_cmx.c
@@ -0,0 +1,1886 @@
+/*
+ * Audio crossconnecting/conferrencing (hardware level).
+ *
+ * Copyright 2002 by Andreas Eversberg (jolly@eversberg.eu)
+ *
+ * This software may be used and distributed according to the terms
+ * of the GNU General Public License, incorporated herein by reference.
+ *
+ */
+
+/*
+ * The process of adding and removing parties to/from a conference:
+ *
+ * There is a chain of struct dsp_conf which has one or more members in a chain
+ * of struct dsp_conf_member.
+ *
+ * After a party is added, the conference is checked for hardware capability.
+ * Also if a party is removed, the conference is checked again.
+ *
+ * There are 3 different solutions: -1 = software, 0 = hardware-crossconnect
+ * 1-n = hardware-conference. The n will give the conference number.
+ *
+ * Depending on the change after removal or insertion of a party, hardware
+ * commands are given.
+ *
+ * The current solution is stored within the struct dsp_conf entry.
+ */
+
+/*
+ * HOW THE CMX WORKS:
+ *
+ * There are 3 types of interaction: One member is alone, in this case only
+ * data flow from upper to lower layer is done.
+ * Two members will also exchange their data so they are crossconnected.
+ * Three or more members will be added in a conference and will hear each
+ * other but will not receive their own speech (echo) if not enabled.
+ *
+ * Features of CMX are:
+ *  - Crossconnecting or even conference, if more than two members are together.
+ *  - Force mixing of transmit data with other crossconnect/conference members.
+ *  - Echo generation to benchmark the delay of audio processing.
+ *  - Use hardware to minimize cpu load, disable FIFO load and minimize delay.
+ *  - Dejittering and clock generation.
+ *
+ * There are 2 buffers:
+ *
+ *
+ * RX-Buffer
+ *                 R             W
+ *                 |             |
+ * ----------------+-------------+-------------------
+ *
+ * The rx-buffer is a ring buffer used to store the received data for each
+ * individual member. This is only the case if data needs to be dejittered
+ * or in case of a conference where different clocks require reclocking.
+ * The transmit-clock (R) will read the buffer.
+ * If the clock overruns the write-pointer, we will have a buffer underrun.
+ * If the write pointer always has a certain distance from the transmit-
+ * clock, we will have a delay. The delay will dynamically be increased and
+ * reduced.
+ *
+ *
+ * TX-Buffer
+ *                  R        W
+ *                  |        |
+ * -----------------+--------+-----------------------
+ *
+ * The tx-buffer is a ring buffer to queue the transmit data from user space
+ * until it will be mixed or sent. There are two pointers, R and W. If the write
+ * pointer W would reach or overrun R, the buffer would overrun. In this case
+ * (some) data is dropped so that it will not overrun.
+ * Additionally a dynamic dejittering can be enabled. this allows data from
+ * user space that have jitter and different clock source.
+ *
+ *
+ * Clock:
+ *
+ * A Clock is not required, if the data source has exactly one clock. In this
+ * case the data source is forwarded to the destination.
+ *
+ * A Clock is required, because the data source
+ *  - has multiple clocks.
+ *  - has no usable clock due to jitter or packet loss (VoIP).
+ * In this case the system's clock is used. The clock resolution depends on
+ * the jiffie resolution.
+ *
+ * If a member joins a conference:
+ *
+ * - If a member joins, its rx_buff is set to silence and change read pointer
+ *   to transmit clock.
+ *
+ * The procedure of received data from card is explained in cmx_receive.
+ * The procedure of received data from user space is explained in cmx_transmit.
+ * The procedure of transmit data to card is cmx_send.
+ *
+ *
+ * Interaction with other features:
+ *
+ * DTMF:
+ * DTMF decoding is done before the data is crossconnected.
+ *
+ * Volume change:
+ * Changing rx-volume is done before the data is crossconnected. The tx-volume
+ * must be changed whenever data is transmitted to the card by the cmx.
+ *
+ * Tones:
+ * If a tone is enabled, it will be processed whenever data is transmitted to
+ * the card. It will replace the tx-data from the user space.
+ * If tones are generated by hardware, this conference member is removed for
+ * this time.
+ *
+ * Disable rx-data:
+ * If cmx is realized in hardware, rx data will be disabled if requested by
+ * the upper layer. If dtmf decoding is done by software and enabled, rx data
+ * will not be diabled but blocked to the upper layer.
+ *
+ * HFC conference engine:
+ * If it is possible to realize all features using hardware, hardware will be
+ * used if not forbidden by control command. Disabling rx-data provides
+ * absolutely traffic free audio processing. (except for the quick 1-frame
+ * upload of a tone loop, only once for a new tone)
+ *
+ */
+
+/* delay.h is required for hw_lock.h */
+
+#include <linux/delay.h>
+#include <linux/mISDNif.h>
+#include <linux/mISDNdsp.h>
+#include "core.h"
+#include "dsp.h"
+/*
+ * debugging of multi party conference,
+ * by using conference even with two members
+ */
+
+/* #define CMX_CONF_DEBUG */
+
+/*#define CMX_DEBUG * massive read/write pointer output */
+/*#define CMX_TX_DEBUG * massive read/write on tx-buffer with content */
+
+static inline int
+count_list_member(struct list_head *head)
+{
+	int			cnt = 0;
+	struct list_head	*m;
+
+	list_for_each(m, head)
+		cnt++;
+	return cnt;
+}
+
+/*
+ * debug cmx memory structure
+ */
+void
+dsp_cmx_debug(struct dsp *dsp)
+{
+	struct dsp_conf	*conf;
+	struct dsp_conf_member	*member;
+	struct dsp		*odsp;
+
+	printk(KERN_DEBUG "-----Current DSP\n");
+	list_for_each_entry(odsp, &dsp_ilist, list) {
+		printk(KERN_DEBUG "* %s echo=%d txmix=%d",
+		    odsp->name, odsp->echo, odsp->tx_mix);
+		if (odsp->conf)
+			printk(" (Conf %d)", odsp->conf->id);
+		if (dsp == odsp)
+			printk(" *this*");
+		printk("\n");
+	}
+	printk(KERN_DEBUG "-----Current Conf:\n");
+	list_for_each_entry(conf, &conf_ilist, list) {
+		printk(KERN_DEBUG "* Conf %d (%p)\n", conf->id, conf);
+		list_for_each_entry(member, &conf->mlist, list) {
+			printk(KERN_DEBUG
+			    "  - member = %s (slot_tx %d, bank_tx %d, "
+			    "slot_rx %d, bank_rx %d hfc_conf %d)%s\n",
+			    member->dsp->name, member->dsp->pcm_slot_tx,
+			    member->dsp->pcm_bank_tx, member->dsp->pcm_slot_rx,
+			    member->dsp->pcm_bank_rx, member->dsp->hfc_conf,
+			    (member->dsp == dsp) ? " *this*" : "");
+		}
+	}
+	printk(KERN_DEBUG "-----end\n");
+}
+
+/*
+ * search conference
+ */
+static struct dsp_conf *
+dsp_cmx_search_conf(u32 id)
+{
+	struct dsp_conf *conf;
+
+	if (!id) {
+		printk(KERN_WARNING "%s: conference ID is 0.\n", __func__);
+		return NULL;
+	}
+
+	/* search conference */
+	list_for_each_entry(conf, &conf_ilist, list)
+		if (conf->id == id)
+			return conf;
+
+	return NULL;
+}
+
+
+/*
+ * add member to conference
+ */
+static int
+dsp_cmx_add_conf_member(struct dsp *dsp, struct dsp_conf *conf)
+{
+	struct dsp_conf_member *member;
+
+	if (!conf || !dsp) {
+		printk(KERN_WARNING "%s: conf or dsp is 0.\n", __func__);
+		return -EINVAL;
+	}
+	if (dsp->member) {
+		printk(KERN_WARNING "%s: dsp is already member in a conf.\n",
+			__func__);
+		return -EINVAL;
+	}
+
+	if (dsp->conf) {
+		printk(KERN_WARNING "%s: dsp is already in a conf.\n",
+			__func__);
+		return -EINVAL;
+	}
+
+	member = kzalloc(sizeof(struct dsp_conf_member), GFP_ATOMIC);
+	if (!member) {
+		printk(KERN_ERR "kmalloc struct dsp_conf_member failed\n");
+		return -ENOMEM;
+	}
+	member->dsp = dsp;
+	/* clear rx buffer */
+	memset(dsp->rx_buff, dsp_silence, sizeof(dsp->rx_buff));
+	dsp->rx_init = 1; /* rx_W and rx_R will be adjusted on first frame */
+	dsp->rx_W = 0;
+	dsp->rx_R = 0;
+
+	list_add_tail(&member->list, &conf->mlist);
+
+	dsp->conf = conf;
+	dsp->member = member;
+
+	return 0;
+}
+
+
+/*
+ * del member from conference
+ */
+int
+dsp_cmx_del_conf_member(struct dsp *dsp)
+{
+	struct dsp_conf_member *member;
+
+	if (!dsp) {
+		printk(KERN_WARNING "%s: dsp is 0.\n",
+			__func__);
+		return -EINVAL;
+	}
+
+	if (!dsp->conf) {
+		printk(KERN_WARNING "%s: dsp is not in a conf.\n",
+			__func__);
+		return -EINVAL;
+	}
+
+	if (list_empty(&dsp->conf->mlist)) {
+		printk(KERN_WARNING "%s: dsp has linked an empty conf.\n",
+			__func__);
+		return -EINVAL;
+	}
+
+	/* find us in conf */
+	list_for_each_entry(member, &dsp->conf->mlist, list) {
+		if (member->dsp == dsp) {
+			list_del(&member->list);
+			dsp->conf = NULL;
+			dsp->member = NULL;
+			kfree(member);
+			return 0;
+		}
+	}
+	printk(KERN_WARNING
+	    "%s: dsp is not present in its own conf_meber list.\n",
+	    __func__);
+
+	return -EINVAL;
+}
+
+
+/*
+ * new conference
+ */
+static struct dsp_conf
+*dsp_cmx_new_conf(u32 id)
+{
+	struct dsp_conf *conf;
+
+	if (!id) {
+		printk(KERN_WARNING "%s: id is 0.\n",
+		    __func__);
+		return NULL;
+	}
+
+	conf = kzalloc(sizeof(struct dsp_conf), GFP_ATOMIC);
+	if (!conf) {
+		printk(KERN_ERR "kmalloc struct dsp_conf failed\n");
+		return NULL;
+	}
+	INIT_LIST_HEAD(&conf->mlist);
+	conf->id = id;
+
+	list_add_tail(&conf->list, &conf_ilist);
+
+	return conf;
+}
+
+
+/*
+ * del conference
+ */
+int
+dsp_cmx_del_conf(struct dsp_conf *conf)
+{
+	if (!conf) {
+		printk(KERN_WARNING "%s: conf is null.\n",
+		    __func__);
+		return -EINVAL;
+	}
+
+	if (!list_empty(&conf->mlist)) {
+		printk(KERN_WARNING "%s: conf not empty.\n",
+		    __func__);
+		return -EINVAL;
+	}
+	list_del(&conf->list);
+	kfree(conf);
+
+	return 0;
+}
+
+
+/*
+ * send HW message to hfc card
+ */
+static void
+dsp_cmx_hw_message(struct dsp *dsp, u32 message, u32 param1, u32 param2,
+    u32 param3, u32 param4)
+{
+	struct mISDN_ctrl_req cq;
+
+	memset(&cq, 0, sizeof(cq));
+	cq.op = message;
+	cq.p1 = param1 | (param2 << 8);
+	cq.p2 = param3 | (param4 << 8);
+	if (dsp->ch.peer)
+		dsp->ch.peer->ctrl(dsp->ch.peer, CONTROL_CHANNEL, &cq);
+}
+
+
+/*
+ * do hardware update and set the software/hardware flag
+ *
+ * either a conference or a dsp instance can be given
+ * if only dsp instance is given, the instance is not associated with a conf
+ * and therefore removed. if a conference is given, the dsp is expected to
+ * be member of that conference.
+ */
+void
+dsp_cmx_hardware(struct dsp_conf *conf, struct dsp *dsp)
+{
+	struct dsp_conf_member	*member, *nextm;
+	struct dsp		*finddsp;
+	int		memb = 0, i, ii, i1, i2;
+	int		freeunits[8];
+	u_char		freeslots[256];
+	int		same_hfc = -1, same_pcm = -1, current_conf = -1,
+	    all_conf = 1;
+
+	/* dsp gets updated (no conf) */
+	if (!conf) {
+		if (!dsp)
+			return;
+		if (dsp_debug & DEBUG_DSP_CMX)
+			printk(KERN_DEBUG "%s checking dsp %s\n",
+			    __func__, dsp->name);
+one_member:
+		/* remove HFC conference if enabled */
+		if (dsp->hfc_conf >= 0) {
+			if (dsp_debug & DEBUG_DSP_CMX)
+				printk(KERN_DEBUG
+				    "%s removing %s from HFC conf %d "
+				    "because dsp is split\n", __func__,
+				    dsp->name, dsp->hfc_conf);
+			dsp_cmx_hw_message(dsp, MISDN_CTRL_HFC_CONF_SPLIT,
+			    0, 0, 0, 0);
+			dsp->hfc_conf = -1;
+		}
+		/* process hw echo */
+		if (dsp->features.pcm_banks < 1)
+			return;
+		if (!dsp->echo) {
+			/* NO ECHO: remove PCM slot if assigned */
+			if (dsp->pcm_slot_tx >= 0 || dsp->pcm_slot_rx >= 0) {
+				if (dsp_debug & DEBUG_DSP_CMX)
+					printk(KERN_DEBUG "%s removing %s from"
+					    " PCM slot %d (TX) %d (RX) because"
+					    " dsp is split (no echo)\n",
+					    __func__, dsp->name,
+					    dsp->pcm_slot_tx, dsp->pcm_slot_rx);
+				dsp_cmx_hw_message(dsp, MISDN_CTRL_HFC_PCM_DISC,
+				    0, 0, 0, 0);
+				dsp->pcm_slot_tx = -1;
+				dsp->pcm_bank_tx = -1;
+				dsp->pcm_slot_rx = -1;
+				dsp->pcm_bank_rx = -1;
+			}
+			return;
+		}
+		/* ECHO: already echo */
+		if (dsp->pcm_slot_tx >= 0 && dsp->pcm_slot_rx < 0 &&
+		    dsp->pcm_bank_tx == 2 && dsp->pcm_bank_rx == 2)
+			return;
+		/* ECHO: if slot already assigned */
+		if (dsp->pcm_slot_tx >= 0) {
+			dsp->pcm_slot_rx = dsp->pcm_slot_tx;
+			dsp->pcm_bank_tx = 2; /* 2 means loop */
+			dsp->pcm_bank_rx = 2;
+			if (dsp_debug & DEBUG_DSP_CMX)
+				printk(KERN_DEBUG
+				    "%s refresh %s for echo using slot %d\n",
+				    __func__, dsp->name,
+				    dsp->pcm_slot_tx);
+			dsp_cmx_hw_message(dsp, MISDN_CTRL_HFC_PCM_CONN,
+			    dsp->pcm_slot_tx, 2, dsp->pcm_slot_rx, 2);
+			return;
+		}
+		/* ECHO: find slot */
+		dsp->pcm_slot_tx = -1;
+		dsp->pcm_slot_rx = -1;
+		memset(freeslots, 1, sizeof(freeslots));
+		list_for_each_entry(finddsp, &dsp_ilist, list) {
+			if (finddsp->features.pcm_id == dsp->features.pcm_id) {
+				if (finddsp->pcm_slot_rx >= 0 &&
+				    finddsp->pcm_slot_rx < sizeof(freeslots))
+					freeslots[finddsp->pcm_slot_tx] = 0;
+				if (finddsp->pcm_slot_tx >= 0 &&
+				    finddsp->pcm_slot_tx < sizeof(freeslots))
+					freeslots[finddsp->pcm_slot_rx] = 0;
+			}
+		}
+		i = 0;
+		ii = dsp->features.pcm_slots;
+		while (i < ii) {
+			if (freeslots[i])
+				break;
+			i++;
+		}
+		if (i == ii) {
+			if (dsp_debug & DEBUG_DSP_CMX)
+				printk(KERN_DEBUG
+				    "%s no slot available for echo\n",
+				    __func__);
+			/* no more slots available */
+			return;
+		}
+		/* assign free slot */
+		dsp->pcm_slot_tx = i;
+		dsp->pcm_slot_rx = i;
+		dsp->pcm_bank_tx = 2; /* loop */
+		dsp->pcm_bank_rx = 2;
+		if (dsp_debug & DEBUG_DSP_CMX)
+			printk(KERN_DEBUG
+			    "%s assign echo for %s using slot %d\n",
+			    __func__, dsp->name, dsp->pcm_slot_tx);
+		dsp_cmx_hw_message(dsp, MISDN_CTRL_HFC_PCM_CONN,
+		    dsp->pcm_slot_tx, 2, dsp->pcm_slot_rx, 2);
+		return;
+	}
+
+	/* conf gets updated (all members) */
+	if (dsp_debug & DEBUG_DSP_CMX)
+		printk(KERN_DEBUG "%s checking conference %d\n",
+		    __func__, conf->id);
+
+	if (list_empty(&conf->mlist)) {
+		printk(KERN_ERR "%s: conference whithout members\n",
+		    __func__);
+		return;
+	}
+	member = list_entry(conf->mlist.next, struct dsp_conf_member, list);
+	same_hfc = member->dsp->features.hfc_id;
+	same_pcm = member->dsp->features.pcm_id;
+	/* check all members in our conference */
+	list_for_each_entry(member, &conf->mlist, list) {
+		/* check if member uses mixing */
+		if (member->dsp->tx_mix) {
+			if (dsp_debug & DEBUG_DSP_CMX)
+				printk(KERN_DEBUG
+				    "%s dsp %s cannot form a conf, because "
+				    "tx_mix is turned on\n", __func__,
+				    member->dsp->name);
+conf_software:
+			list_for_each_entry(member, &conf->mlist, list) {
+				dsp = member->dsp;
+				/* remove HFC conference if enabled */
+				if (dsp->hfc_conf >= 0) {
+					if (dsp_debug & DEBUG_DSP_CMX)
+						printk(KERN_DEBUG
+						    "%s removing %s from HFC "
+						    "conf %d because not "
+						    "possible with hardware\n",
+						    __func__,
+						    dsp->name,
+						    dsp->hfc_conf);
+					dsp_cmx_hw_message(dsp,
+					    MISDN_CTRL_HFC_CONF_SPLIT,
+					    0, 0, 0, 0);
+					dsp->hfc_conf = -1;
+				}
+				/* remove PCM slot if assigned */
+				if (dsp->pcm_slot_tx >= 0 ||
+				    dsp->pcm_slot_rx >= 0) {
+					if (dsp_debug & DEBUG_DSP_CMX)
+						printk(KERN_DEBUG "%s removing "
+						    "%s from PCM slot %d (TX)"
+						    " slot %d (RX) because not"
+						    " possible with hardware\n",
+						    __func__,
+						    dsp->name,
+						    dsp->pcm_slot_tx,
+						    dsp->pcm_slot_rx);
+					dsp_cmx_hw_message(dsp,
+					    MISDN_CTRL_HFC_PCM_DISC,
+					    0, 0, 0, 0);
+					dsp->pcm_slot_tx = -1;
+					dsp->pcm_bank_tx = -1;
+					dsp->pcm_slot_rx = -1;
+					dsp->pcm_bank_rx = -1;
+				}
+			}
+			conf->hardware = 0;
+			conf->software = 1;
+			return;
+		}
+		/* check if member has echo turned on */
+		if (member->dsp->echo) {
+			if (dsp_debug & DEBUG_DSP_CMX)
+				printk(KERN_DEBUG
+				    "%s dsp %s cannot form a conf, because "
+				    "echo is turned on\n", __func__,
+				    member->dsp->name);
+			goto conf_software;
+		}
+		/* check if member has tx_mix turned on */
+		if (member->dsp->tx_mix) {
+			if (dsp_debug & DEBUG_DSP_CMX)
+				printk(KERN_DEBUG
+				    "%s dsp %s cannot form a conf, because "
+				    "tx_mix is turned on\n",
+				    __func__, member->dsp->name);
+			goto conf_software;
+		}
+		/* check if member changes volume at an not suppoted level */
+		if (member->dsp->tx_volume) {
+			if (dsp_debug & DEBUG_DSP_CMX)
+				printk(KERN_DEBUG
+				    "%s dsp %s cannot form a conf, because "
+				    "tx_volume is changed\n",
+				    __func__, member->dsp->name);
+			goto conf_software;
+		}
+		if (member->dsp->rx_volume) {
+			if (dsp_debug & DEBUG_DSP_CMX)
+				printk(KERN_DEBUG
+				    "%s dsp %s cannot form a conf, because "
+				    "rx_volume is changed\n",
+				    __func__, member->dsp->name);
+			goto conf_software;
+		}
+		/* check if tx-data turned on */
+		if (member->dsp->tx_data) {
+			if (dsp_debug & DEBUG_DSP_CMX)
+				printk(KERN_DEBUG
+				    "%s dsp %s cannot form a conf, because "
+				    "tx_data is turned on\n",
+				    __func__, member->dsp->name);
+			goto conf_software;
+		}
+		/* check if pipeline exists */
+		if (member->dsp->pipeline.inuse) {
+			if (dsp_debug & DEBUG_DSP_CMX)
+				printk(KERN_DEBUG
+				    "%s dsp %s cannot form a conf, because "
+				    "pipeline exists\n", __func__,
+				    member->dsp->name);
+			goto conf_software;
+		}
+		/* check if encryption is enabled */
+		if (member->dsp->bf_enable) {
+			if (dsp_debug & DEBUG_DSP_CMX)
+				printk(KERN_DEBUG "%s dsp %s cannot form a "
+				    "conf, because encryption is enabled\n",
+				    __func__, member->dsp->name);
+			goto conf_software;
+		}
+		/* check if member is on a card with PCM support */
+		if (member->dsp->features.pcm_id < 0) {
+			if (dsp_debug & DEBUG_DSP_CMX)
+				printk(KERN_DEBUG
+				    "%s dsp %s cannot form a conf, because "
+				    "dsp has no PCM bus\n",
+				    __func__, member->dsp->name);
+			goto conf_software;
+		}
+		/* check if relations are on the same PCM bus */
+		if (member->dsp->features.pcm_id != same_pcm) {
+			if (dsp_debug & DEBUG_DSP_CMX)
+				printk(KERN_DEBUG
+				    "%s dsp %s cannot form a conf, because "
+				    "dsp is on a different PCM bus than the "
+				    "first dsp\n",
+				    __func__, member->dsp->name);
+			goto conf_software;
+		}
+		/* determine if members are on the same hfc chip */
+		if (same_hfc != member->dsp->features.hfc_id)
+			same_hfc = -1;
+		/* if there are members already in a conference */
+		if (current_conf < 0 && member->dsp->hfc_conf >= 0)
+			current_conf = member->dsp->hfc_conf;
+		/* if any member is not in a conference */
+		if (member->dsp->hfc_conf < 0)
+			all_conf = 0;
+
+		memb++;
+	}
+
+	/* if no member, this is an error */
+	if (memb < 1)
+		return;
+
+	/* one member */
+	if (memb == 1) {
+		if (dsp_debug & DEBUG_DSP_CMX)
+			printk(KERN_DEBUG
+			    "%s conf %d cannot form a HW conference, "
+			    "because dsp is alone\n", __func__, conf->id);
+		conf->hardware = 0;
+		conf->software = 0;
+		member = list_entry(conf->mlist.next, struct dsp_conf_member,
+			list);
+		dsp = member->dsp;
+		goto one_member;
+	}
+
+	/*
+	 * ok, now we are sure that all members are on the same pcm.
+	 * now we will see if we have only two members, so we can do
+	 * crossconnections, which don't have any limitations.
+	 */
+
+	/* if we have only two members */
+	if (memb == 2) {
+		member = list_entry(conf->mlist.next, struct dsp_conf_member,
+			list);
+		nextm = list_entry(member->list.next, struct dsp_conf_member,
+			list);
+		/* remove HFC conference if enabled */
+		if (member->dsp->hfc_conf >= 0) {
+			if (dsp_debug & DEBUG_DSP_CMX)
+				printk(KERN_DEBUG
+				    "%s removing %s from HFC conf %d because "
+				    "two parties require only a PCM slot\n",
+				    __func__, member->dsp->name,
+				    member->dsp->hfc_conf);
+			dsp_cmx_hw_message(member->dsp,
+			    MISDN_CTRL_HFC_CONF_SPLIT, 0, 0, 0, 0);
+			member->dsp->hfc_conf = -1;
+		}
+		if (nextm->dsp->hfc_conf >= 0) {
+			if (dsp_debug & DEBUG_DSP_CMX)
+				printk(KERN_DEBUG
+				    "%s removing %s from HFC conf %d because "
+				    "two parties require only a PCM slot\n",
+				    __func__, nextm->dsp->name,
+				    nextm->dsp->hfc_conf);
+			dsp_cmx_hw_message(nextm->dsp,
+			    MISDN_CTRL_HFC_CONF_SPLIT, 0, 0, 0, 0);
+			nextm->dsp->hfc_conf = -1;
+		}
+		/* if members have two banks (and not on the same chip) */
+		if (member->dsp->features.pcm_banks > 1 &&
+		    nextm->dsp->features.pcm_banks > 1 &&
+		    member->dsp->features.hfc_id !=
+		    nextm->dsp->features.hfc_id) {
+			/* if both members have same slots with crossed banks */
+			if (member->dsp->pcm_slot_tx >= 0 &&
+			    member->dsp->pcm_slot_rx >= 0 &&
+			    nextm->dsp->pcm_slot_tx >= 0 &&
+			    nextm->dsp->pcm_slot_rx >= 0 &&
+			    nextm->dsp->pcm_slot_tx ==
+			    member->dsp->pcm_slot_rx &&
+			    nextm->dsp->pcm_slot_rx ==
+			    member->dsp->pcm_slot_tx &&
+			    nextm->dsp->pcm_slot_tx ==
+			    member->dsp->pcm_slot_tx &&
+			    member->dsp->pcm_bank_tx !=
+			    member->dsp->pcm_bank_rx &&
+			    nextm->dsp->pcm_bank_tx !=
+			    nextm->dsp->pcm_bank_rx) {
+				/* all members have same slot */
+				if (dsp_debug & DEBUG_DSP_CMX)
+					printk(KERN_DEBUG
+					    "%s dsp %s & %s stay joined on "
+					    "PCM slot %d bank %d (TX) bank %d "
+					    "(RX) (on different chips)\n",
+					    __func__,
+					    member->dsp->name,
+					    nextm->dsp->name,
+					    member->dsp->pcm_slot_tx,
+					    member->dsp->pcm_bank_tx,
+					    member->dsp->pcm_bank_rx);
+				conf->hardware = 0;
+				conf->software = 1;
+				return;
+			}
+			/* find a new slot */
+			memset(freeslots, 1, sizeof(freeslots));
+			list_for_each_entry(dsp, &dsp_ilist, list) {
+				if (dsp != member->dsp &&
+				    dsp != nextm->dsp &&
+				    member->dsp->features.pcm_id ==
+				    dsp->features.pcm_id) {
+					if (dsp->pcm_slot_rx >= 0 &&
+					    dsp->pcm_slot_rx <
+					    sizeof(freeslots))
+						freeslots[dsp->pcm_slot_tx] = 0;
+					if (dsp->pcm_slot_tx >= 0 &&
+					    dsp->pcm_slot_tx <
+					    sizeof(freeslots))
+						freeslots[dsp->pcm_slot_rx] = 0;
+				}
+			}
+			i = 0;
+			ii = member->dsp->features.pcm_slots;
+			while (i < ii) {
+				if (freeslots[i])
+					break;
+				i++;
+			}
+			if (i == ii) {
+				if (dsp_debug & DEBUG_DSP_CMX)
+					printk(KERN_DEBUG
+					    "%s no slot available for "
+					    "%s & %s\n", __func__,
+					    member->dsp->name,
+					    nextm->dsp->name);
+				/* no more slots available */
+				goto conf_software;
+			}
+			/* assign free slot */
+			member->dsp->pcm_slot_tx = i;
+			member->dsp->pcm_slot_rx = i;
+			nextm->dsp->pcm_slot_tx = i;
+			nextm->dsp->pcm_slot_rx = i;
+			member->dsp->pcm_bank_rx = 0;
+			member->dsp->pcm_bank_tx = 1;
+			nextm->dsp->pcm_bank_rx = 1;
+			nextm->dsp->pcm_bank_tx = 0;
+			if (dsp_debug & DEBUG_DSP_CMX)
+				printk(KERN_DEBUG
+				    "%s adding %s & %s to new PCM slot %d "
+				    "(TX and RX on different chips) because "
+				    "both members have not same slots\n",
+				    __func__,
+				    member->dsp->name,
+				    nextm->dsp->name,
+				    member->dsp->pcm_slot_tx);
+			dsp_cmx_hw_message(member->dsp, MISDN_CTRL_HFC_PCM_CONN,
+			    member->dsp->pcm_slot_tx, member->dsp->pcm_bank_tx,
+			    member->dsp->pcm_slot_rx, member->dsp->pcm_bank_rx);
+			dsp_cmx_hw_message(nextm->dsp, MISDN_CTRL_HFC_PCM_CONN,
+			    nextm->dsp->pcm_slot_tx, nextm->dsp->pcm_bank_tx,
+			    nextm->dsp->pcm_slot_rx, nextm->dsp->pcm_bank_rx);
+			conf->hardware = 1;
+			conf->software = 0;
+			return;
+		/* if members have one bank (or on the same chip) */
+		} else {
+			/* if both members have different crossed slots */
+			if (member->dsp->pcm_slot_tx >= 0 &&
+			    member->dsp->pcm_slot_rx >= 0 &&
+			    nextm->dsp->pcm_slot_tx >= 0 &&
+			    nextm->dsp->pcm_slot_rx >= 0 &&
+			    nextm->dsp->pcm_slot_tx ==
+			    member->dsp->pcm_slot_rx &&
+			    nextm->dsp->pcm_slot_rx ==
+			    member->dsp->pcm_slot_tx &&
+			    member->dsp->pcm_slot_tx !=
+			    member->dsp->pcm_slot_rx &&
+			    member->dsp->pcm_bank_tx == 0 &&
+			    member->dsp->pcm_bank_rx == 0 &&
+			    nextm->dsp->pcm_bank_tx == 0 &&
+			    nextm->dsp->pcm_bank_rx == 0) {
+				/* all members have same slot */
+				if (dsp_debug & DEBUG_DSP_CMX)
+					printk(KERN_DEBUG
+					    "%s dsp %s & %s stay joined on PCM "
+					    "slot %d (TX) %d (RX) on same chip "
+					    "or one bank PCM)\n", __func__,
+					    member->dsp->name,
+					    nextm->dsp->name,
+					    member->dsp->pcm_slot_tx,
+					    member->dsp->pcm_slot_rx);
+				conf->hardware = 0;
+				conf->software = 1;
+				return;
+			}
+			/* find two new slot */
+			memset(freeslots, 1, sizeof(freeslots));
+			list_for_each_entry(dsp, &dsp_ilist, list) {
+				if (dsp != member->dsp &&
+				    dsp != nextm->dsp &&
+				    member->dsp->features.pcm_id ==
+				    dsp->features.pcm_id) {
+					if (dsp->pcm_slot_rx >= 0 &&
+					    dsp->pcm_slot_rx <
+					    sizeof(freeslots))
+						freeslots[dsp->pcm_slot_tx] = 0;
+					if (dsp->pcm_slot_tx >= 0 &&
+					    dsp->pcm_slot_tx <
+					    sizeof(freeslots))
+						freeslots[dsp->pcm_slot_rx] = 0;
+				}
+			}
+			i1 = 0;
+			ii = member->dsp->features.pcm_slots;
+			while (i1 < ii) {
+				if (freeslots[i1])
+					break;
+				i1++;
+			}
+			if (i1 == ii) {
+				if (dsp_debug & DEBUG_DSP_CMX)
+					printk(KERN_DEBUG
+					    "%s no slot available "
+					    "for %s & %s\n", __func__,
+					    member->dsp->name,
+					    nextm->dsp->name);
+				/* no more slots available */
+				goto conf_software;
+			}
+			i2 = i1+1;
+			while (i2 < ii) {
+				if (freeslots[i2])
+					break;
+				i2++;
+			}
+			if (i2 == ii) {
+				if (dsp_debug & DEBUG_DSP_CMX)
+					printk(KERN_DEBUG
+					    "%s no slot available "
+					    "for %s & %s\n",
+					    __func__,
+					    member->dsp->name,
+					    nextm->dsp->name);
+				/* no more slots available */
+				goto conf_software;
+			}
+			/* assign free slots */
+			member->dsp->pcm_slot_tx = i1;
+			member->dsp->pcm_slot_rx = i2;
+			nextm->dsp->pcm_slot_tx = i2;
+			nextm->dsp->pcm_slot_rx = i1;
+			member->dsp->pcm_bank_rx = 0;
+			member->dsp->pcm_bank_tx = 0;
+			nextm->dsp->pcm_bank_rx = 0;
+			nextm->dsp->pcm_bank_tx = 0;
+			if (dsp_debug & DEBUG_DSP_CMX)
+				printk(KERN_DEBUG
+				    "%s adding %s & %s to new PCM slot %d "
+				    "(TX) %d (RX) on same chip or one bank "
+				    "PCM, because both members have not "
+				    "crossed slots\n", __func__,
+				    member->dsp->name,
+				    nextm->dsp->name,
+				    member->dsp->pcm_slot_tx,
+				    member->dsp->pcm_slot_rx);
+			dsp_cmx_hw_message(member->dsp, MISDN_CTRL_HFC_PCM_CONN,
+			    member->dsp->pcm_slot_tx, member->dsp->pcm_bank_tx,
+			    member->dsp->pcm_slot_rx, member->dsp->pcm_bank_rx);
+			dsp_cmx_hw_message(nextm->dsp, MISDN_CTRL_HFC_PCM_CONN,
+			    nextm->dsp->pcm_slot_tx, nextm->dsp->pcm_bank_tx,
+			    nextm->dsp->pcm_slot_rx, nextm->dsp->pcm_bank_rx);
+			conf->hardware = 1;
+			conf->software = 0;
+			return;
+		}
+	}
+
+	/*
+	 * if we have more than two, we may check if we have a conference
+	 * unit available on the chip. also all members must be on the same
+	 */
+
+	/* if not the same HFC chip */
+	if (same_hfc < 0) {
+		if (dsp_debug & DEBUG_DSP_CMX)
+			printk(KERN_DEBUG
+			    "%s conference %d cannot be formed, because "
+			    "members are on different chips or not "
+			    "on HFC chip\n",
+			    __func__, conf->id);
+		goto conf_software;
+	}
+
+	/* for more than two members.. */
+
+	/* in case of hdlc, we change to software */
+	if (dsp->hdlc)
+		goto conf_software;
+
+	/* if all members already have the same conference */
+	if (all_conf)
+		return;
+
+	/*
+	 * if there is an existing conference, but not all members have joined
+	 */
+	if (current_conf >= 0) {
+join_members:
+		list_for_each_entry(member, &conf->mlist, list) {
+			/* join to current conference */
+			if (member->dsp->hfc_conf == current_conf)
+				continue;
+			/* get a free timeslot first */
+			memset(freeslots, 1, sizeof(freeslots));
+			list_for_each_entry(dsp, &dsp_ilist, list) {
+				/*
+				 * not checking current member, because
+				 * slot will be overwritten.
+				 */
+				if (
+				    dsp != member->dsp &&
+				/* dsp must be on the same PCM */
+				    member->dsp->features.pcm_id ==
+				    dsp->features.pcm_id) {
+					/* dsp must be on a slot */
+					if (dsp->pcm_slot_tx >= 0 &&
+					    dsp->pcm_slot_tx <
+					    sizeof(freeslots))
+						freeslots[dsp->pcm_slot_tx] = 0;
+					if (dsp->pcm_slot_rx >= 0 &&
+					    dsp->pcm_slot_rx <
+					    sizeof(freeslots))
+						freeslots[dsp->pcm_slot_rx] = 0;
+				}
+			}
+			i = 0;
+			ii = member->dsp->features.pcm_slots;
+			while (i < ii) {
+				if (freeslots[i])
+					break;
+				i++;
+			}
+			if (i == ii) {
+				/* no more slots available */
+				if (dsp_debug & DEBUG_DSP_CMX)
+					printk(KERN_DEBUG
+					    "%s conference %d cannot be formed,"
+					    " because no slot free\n",
+					    __func__, conf->id);
+				goto conf_software;
+			}
+			if (dsp_debug & DEBUG_DSP_CMX)
+				printk(KERN_DEBUG
+				    "%s changing dsp %s to HW conference "
+				    "%d slot %d\n", __func__,
+				    member->dsp->name, current_conf, i);
+			/* assign free slot & set PCM & join conf */
+			member->dsp->pcm_slot_tx = i;
+			member->dsp->pcm_slot_rx = i;
+			member->dsp->pcm_bank_tx = 2; /* loop */
+			member->dsp->pcm_bank_rx = 2;
+			member->dsp->hfc_conf = current_conf;
+			dsp_cmx_hw_message(member->dsp, MISDN_CTRL_HFC_PCM_CONN,
+			    i, 2, i, 2);
+			dsp_cmx_hw_message(member->dsp,
+			    MISDN_CTRL_HFC_CONF_JOIN, current_conf, 0, 0, 0);
+		}
+		return;
+	}
+
+	/*
+	 * no member is in a conference yet, so we find a free one
+	 */
+	memset(freeunits, 1, sizeof(freeunits));
+	list_for_each_entry(dsp, &dsp_ilist, list) {
+		/* dsp must be on the same chip */
+		if (dsp->features.hfc_id == same_hfc &&
+		    /* dsp must have joined a HW conference */
+		    dsp->hfc_conf >= 0 &&
+		    /* slot must be within range */
+		    dsp->hfc_conf < 8)
+			freeunits[dsp->hfc_conf] = 0;
+	}
+	i = 0;
+	ii = 8;
+	while (i < ii) {
+		if (freeunits[i])
+			break;
+		i++;
+	}
+	if (i == ii) {
+		/* no more conferences available */
+		if (dsp_debug & DEBUG_DSP_CMX)
+			printk(KERN_DEBUG
+			    "%s conference %d cannot be formed, because "
+			    "no conference number free\n",
+			    __func__, conf->id);
+		goto conf_software;
+	}
+	/* join all members */
+	current_conf = i;
+	goto join_members;
+}
+
+
+/*
+ * conf_id != 0: join or change conference
+ * conf_id == 0: split from conference if not already
+ */
+int
+dsp_cmx_conf(struct dsp *dsp, u32 conf_id)
+{
+	int err;
+	struct dsp_conf *conf;
+	struct dsp_conf_member	*member;
+
+	/* if conference doesn't change */
+	if (dsp->conf_id == conf_id)
+		return 0;
+
+	/* first remove us from current conf */
+	if (dsp->conf_id) {
+		if (dsp_debug & DEBUG_DSP_CMX)
+			printk(KERN_DEBUG "removing us from conference %d\n",
+				dsp->conf->id);
+		/* remove us from conf */
+		conf = dsp->conf;
+		err = dsp_cmx_del_conf_member(dsp);
+		if (err)
+			return err;
+		dsp->conf_id = 0;
+
+		/* update hardware */
+		dsp_cmx_hardware(NULL, dsp);
+
+		/* conf now empty? */
+		if (list_empty(&conf->mlist)) {
+			if (dsp_debug & DEBUG_DSP_CMX)
+				printk(KERN_DEBUG
+				    "conference is empty, so we remove it.\n");
+			err = dsp_cmx_del_conf(conf);
+			if (err)
+				return err;
+		} else {
+			/* update members left on conf */
+			dsp_cmx_hardware(conf, NULL);
+		}
+	}
+
+	/* if split */
+	if (!conf_id)
+		return 0;
+
+	/* now add us to conf */
+	if (dsp_debug & DEBUG_DSP_CMX)
+		printk(KERN_DEBUG "searching conference %d\n",
+			conf_id);
+	conf = dsp_cmx_search_conf(conf_id);
+	if (!conf) {
+		if (dsp_debug & DEBUG_DSP_CMX)
+			printk(KERN_DEBUG
+			    "conference doesn't exist yet, creating.\n");
+		/* the conference doesn't exist, so we create */
+		conf = dsp_cmx_new_conf(conf_id);
+		if (!conf)
+			return -EINVAL;
+	} else if (!list_empty(&conf->mlist)) {
+		member = list_entry(conf->mlist.next, struct dsp_conf_member,
+			list);
+		if (dsp->hdlc && !member->dsp->hdlc) {
+			if (dsp_debug & DEBUG_DSP_CMX)
+				printk(KERN_DEBUG
+				    "cannot join transparent conference.\n");
+			return -EINVAL;
+		}
+		if (!dsp->hdlc && member->dsp->hdlc) {
+			if (dsp_debug & DEBUG_DSP_CMX)
+				printk(KERN_DEBUG
+				    "cannot join hdlc conference.\n");
+			return -EINVAL;
+		}
+	}
+	/* add conference member */
+	err = dsp_cmx_add_conf_member(dsp, conf);
+	if (err)
+		return err;
+	dsp->conf_id = conf_id;
+
+	/* if we are alone, we do nothing! */
+	if (list_empty(&conf->mlist)) {
+		if (dsp_debug & DEBUG_DSP_CMX)
+			printk(KERN_DEBUG
+			    "we are alone in this conference, so exit.\n");
+		/* update hardware */
+		dsp_cmx_hardware(NULL, dsp);
+		return 0;
+	}
+
+	/* update members on conf */
+	dsp_cmx_hardware(conf, NULL);
+
+	return 0;
+}
+
+
+/*
+ * audio data is received from card
+ */
+void
+dsp_cmx_receive(struct dsp *dsp, struct sk_buff *skb)
+{
+	u8 *d, *p;
+	int len = skb->len;
+	struct mISDNhead *hh = mISDN_HEAD_P(skb);
+	int w, i, ii;
+
+	/* check if we have sompen */
+	if (len < 1)
+		return;
+
+	/* half of the buffer should be larger than maximum packet size */
+	if (len >= CMX_BUFF_HALF) {
+		printk(KERN_ERR
+		    "%s line %d: packet from card is too large (%d bytes). "
+		    "please make card send smaller packets OR increase "
+		    "CMX_BUFF_SIZE\n", __FILE__, __LINE__, len);
+		return;
+	}
+
+	/*
+	 * initialize pointers if not already -
+	 * also add delay if requested by PH_SIGNAL
+	 */
+	if (dsp->rx_init) {
+		dsp->rx_init = 0;
+		if (dsp->features.unordered) {
+			dsp->rx_R = (hh->id & CMX_BUFF_MASK);
+			dsp->rx_W = (dsp->rx_R + dsp->cmx_delay)
+				& CMX_BUFF_MASK;
+		} else {
+			dsp->rx_R = 0;
+			dsp->rx_W = dsp->cmx_delay;
+		}
+	}
+	/* if frame contains time code, write directly */
+	if (dsp->features.unordered) {
+		dsp->rx_W = (hh->id & CMX_BUFF_MASK);
+		/* printk(KERN_DEBUG "%s %08x\n", dsp->name, hh->id); */
+	}
+	/*
+	 * if we underrun (or maybe overrun),
+	 * we set our new read pointer, and write silence to buffer
+	 */
+	if (((dsp->rx_W-dsp->rx_R) & CMX_BUFF_MASK) >= CMX_BUFF_HALF) {
+		if (dsp_debug & DEBUG_DSP_CMX)
+			printk(KERN_DEBUG
+			    "cmx_receive(dsp=%lx): UNDERRUN (or overrun the "
+			    "maximum delay), adjusting read pointer! "
+			    "(inst %s)\n", (u_long)dsp, dsp->name);
+		/* flush buffer */
+		if (dsp->features.unordered) {
+			dsp->rx_R = (hh->id & CMX_BUFF_MASK);
+			dsp->rx_W = (dsp->rx_R + dsp->cmx_delay)
+				& CMX_BUFF_MASK;
+		} else {
+			dsp->rx_R = 0;
+			dsp->rx_W = dsp->cmx_delay;
+		}
+		memset(dsp->rx_buff, dsp_silence, sizeof(dsp->rx_buff));
+	}
+	/* if we have reached double delay, jump back to middle */
+	if (dsp->cmx_delay)
+		if (((dsp->rx_W - dsp->rx_R) & CMX_BUFF_MASK) >=
+		    (dsp->cmx_delay << 1)) {
+			if (dsp_debug & DEBUG_DSP_CMX)
+				printk(KERN_DEBUG
+				    "cmx_receive(dsp=%lx): OVERRUN (because "
+				    "twice the delay is reached), adjusting "
+				    "read pointer! (inst %s)\n",
+				    (u_long)dsp, dsp->name);
+		/* flush buffer */
+		if (dsp->features.unordered) {
+			dsp->rx_R = (hh->id & CMX_BUFF_MASK);
+			dsp->rx_W = (dsp->rx_R + dsp->cmx_delay)
+				& CMX_BUFF_MASK;
+		} else {
+			dsp->rx_R = 0;
+			dsp->rx_W = dsp->cmx_delay;
+		}
+		memset(dsp->rx_buff, dsp_silence, sizeof(dsp->rx_buff));
+	}
+
+	/* show where to write */
+#ifdef CMX_DEBUG
+	printk(KERN_DEBUG
+	    "cmx_receive(dsp=%lx): rx_R(dsp)=%05x rx_W(dsp)=%05x len=%d %s\n",
+	    (u_long)dsp, dsp->rx_R, dsp->rx_W, len, dsp->name);
+#endif
+
+	/* write data into rx_buffer */
+	p = skb->data;
+	d = dsp->rx_buff;
+	w = dsp->rx_W;
+	i = 0;
+	ii = len;
+	while (i < ii) {
+		d[w++ & CMX_BUFF_MASK] = *p++;
+		i++;
+	}
+
+	/* increase write-pointer */
+	dsp->rx_W = ((dsp->rx_W+len) & CMX_BUFF_MASK);
+}
+
+
+/*
+ * send (mixed) audio data to card and control jitter
+ */
+static void
+dsp_cmx_send_member(struct dsp *dsp, int len, s32 *c, int members)
+{
+	struct dsp_conf *conf = dsp->conf;
+	struct dsp *member, *other;
+	register s32 sample;
+	u8 *d, *p, *q, *o_q;
+	struct sk_buff *nskb, *txskb;
+	int r, rr, t, tt, o_r, o_rr;
+	int preload = 0;
+	struct mISDNhead *hh, *thh;
+
+	/* don't process if: */
+	if (!dsp->b_active) { /* if not active */
+		dsp->last_tx = 0;
+		return;
+	}
+	if (dsp->pcm_slot_tx >= 0 && /* connected to pcm slot */
+	    dsp->tx_R == dsp->tx_W && /* AND no tx-data */
+	    !(dsp->tone.tone && dsp->tone.software)) { /* AND not soft tones */
+		dsp->last_tx = 0;
+		return;
+	}
+
+#ifdef CMX_DEBUG
+	printk(KERN_DEBUG
+	    "SEND members=%d dsp=%s, conf=%p, rx_R=%05x rx_W=%05x\n",
+	    members, dsp->name, conf, dsp->rx_R, dsp->rx_W);
+#endif
+
+	/* preload if we have delay set */
+	if (dsp->cmx_delay && !dsp->last_tx) {
+		preload = len;
+		if (preload < 128)
+			preload = 128;
+	}
+
+	/* PREPARE RESULT */
+	nskb = mI_alloc_skb(len + preload, GFP_ATOMIC);
+	if (!nskb) {
+		printk(KERN_ERR
+		    "FATAL ERROR in mISDN_dsp.o: cannot alloc %d bytes\n",
+		    len + preload);
+		return;
+	}
+	hh = mISDN_HEAD_P(nskb);
+	hh->prim = PH_DATA_REQ;
+	hh->id = 0;
+	dsp->last_tx = 1;
+
+	/* set pointers, indexes and stuff */
+	member = dsp;
+	p = dsp->tx_buff; /* transmit data */
+	q = dsp->rx_buff; /* received data */
+	d = skb_put(nskb, preload + len); /* result */
+	t = dsp->tx_R; /* tx-pointers */
+	tt = dsp->tx_W;
+	r = dsp->rx_R; /* rx-pointers */
+	rr = (r + len) & CMX_BUFF_MASK;
+
+	/* preload with silence, if required */
+	if (preload) {
+		memset(d, dsp_silence, preload);
+		d += preload;
+	}
+
+	/* PROCESS TONES/TX-DATA ONLY */
+	if (dsp->tone.tone && dsp->tone.software) {
+		/* -> copy tone */
+		dsp_tone_copy(dsp, d, len);
+		dsp->tx_R = 0; /* clear tx buffer */
+		dsp->tx_W = 0;
+		goto send_packet;
+	}
+	/* if we have tx-data but do not use mixing */
+	if (!dsp->tx_mix && t != tt) {
+		/* -> send tx-data and continue when not enough */
+#ifdef CMX_TX_DEBUG
+	sprintf(debugbuf, "TX sending (%04x-%04x)%p: ", t, tt, p);
+#endif
+		while (r != rr && t != tt) {
+#ifdef CMX_TX_DEBUG
+			if (strlen(debugbuf) < 48)
+			    sprintf(debugbuf+strlen(debugbuf), " %02x", p[t]);
+#endif
+			*d++ = p[t]; /* write tx_buff */
+			t = (t+1) & CMX_BUFF_MASK;
+			r = (r+1) & CMX_BUFF_MASK;
+		}
+		if (r == rr) {
+			dsp->tx_R = t;
+#ifdef CMX_TX_DEBUG
+	printk(KERN_DEBUG "%s\n", debugbuf);
+#endif
+			goto send_packet;
+		}
+	}
+#ifdef CMX_TX_DEBUG
+	printk(KERN_DEBUG "%s\n", debugbuf);
+#endif
+
+	/* PROCESS DATA (one member / no conf) */
+	if (!conf || members <= 1) {
+		/* -> if echo is NOT enabled */
+		if (!dsp->echo) {
+			/* -> send tx-data if available or use 0-volume */
+			while (r != rr && t != tt) {
+				*d++ = p[t]; /* write tx_buff */
+				t = (t+1) & CMX_BUFF_MASK;
+				r = (r+1) & CMX_BUFF_MASK;
+			}
+			if (r != rr)
+				memset(d, dsp_silence, (rr-r)&CMX_BUFF_MASK);
+		/* -> if echo is enabled */
+		} else {
+			/*
+			 * -> mix tx-data with echo if available,
+			 * or use echo only
+			 */
+			while (r != rr && t != tt) {
+				*d++ = dsp_audio_mix_law[(p[t]<<8)|q[r]];
+				t = (t+1) & CMX_BUFF_MASK;
+				r = (r+1) & CMX_BUFF_MASK;
+			}
+			while (r != rr) {
+				*d++ = q[r]; /* echo */
+				r = (r+1) & CMX_BUFF_MASK;
+			}
+		}
+		dsp->tx_R = t;
+		goto send_packet;
+	}
+	/* PROCESS DATA (two members) */
+#ifdef CMX_CONF_DEBUG
+	if (0) {
+#else
+	if (members == 2) {
+#endif
+		/* "other" becomes other party */
+		other = (list_entry(conf->mlist.next,
+		    struct dsp_conf_member, list))->dsp;
+		if (other == member)
+			other = (list_entry(conf->mlist.prev,
+			    struct dsp_conf_member, list))->dsp;
+		o_q = other->rx_buff; /* received data */
+		o_rr = (other->rx_R + len) & CMX_BUFF_MASK;
+			/* end of rx-pointer */
+		o_r = (o_rr - rr + r) & CMX_BUFF_MASK;
+			/* start rx-pointer at current read position*/
+		/* -> if echo is NOT enabled */
+		if (!dsp->echo) {
+			/*
+			 * -> copy other member's rx-data,
+			 * if tx-data is available, mix
+			 */
+			while (o_r != o_rr && t != tt) {
+				*d++ = dsp_audio_mix_law[(p[t]<<8)|o_q[o_r]];
+				t = (t+1) & CMX_BUFF_MASK;
+				o_r = (o_r+1) & CMX_BUFF_MASK;
+			}
+			while (o_r != o_rr) {
+				*d++ = o_q[o_r];
+				o_r = (o_r+1) & CMX_BUFF_MASK;
+			}
+		/* -> if echo is enabled */
+		} else {
+			/*
+			 * -> mix other member's rx-data with echo,
+			 * if tx-data is available, mix
+			 */
+			while (r != rr && t != tt) {
+				sample = dsp_audio_law_to_s32[p[t]] +
+				    dsp_audio_law_to_s32[q[r]] +
+				    dsp_audio_law_to_s32[o_q[o_r]];
+				if (sample < -32768)
+					sample = -32768;
+				else if (sample > 32767)
+					sample = 32767;
+				*d++ = dsp_audio_s16_to_law[sample & 0xffff];
+				    /* tx-data + rx_data + echo */
+				t = (t+1) & CMX_BUFF_MASK;
+				r = (r+1) & CMX_BUFF_MASK;
+				o_r = (o_r+1) & CMX_BUFF_MASK;
+			}
+			while (r != rr) {
+				*d++ = dsp_audio_mix_law[(q[r]<<8)|o_q[o_r]];
+				r = (r+1) & CMX_BUFF_MASK;
+				o_r = (o_r+1) & CMX_BUFF_MASK;
+			}
+		}
+		dsp->tx_R = t;
+		goto send_packet;
+	}
+#ifdef DSP_NEVER_DEFINED
+	}
+#endif
+	/* PROCESS DATA (three or more members) */
+	/* -> if echo is NOT enabled */
+	if (!dsp->echo) {
+		/*
+		 * -> substract rx-data from conf-data,
+		 * if tx-data is available, mix
+		 */
+		while (r != rr && t != tt) {
+			sample = dsp_audio_law_to_s32[p[t]] + *c++ -
+			    dsp_audio_law_to_s32[q[r]];
+			if (sample < -32768)
+				sample = -32768;
+			else if (sample > 32767)
+				sample = 32767;
+			*d++ = dsp_audio_s16_to_law[sample & 0xffff];
+			    /* conf-rx+tx */
+			r = (r+1) & CMX_BUFF_MASK;
+			t = (t+1) & CMX_BUFF_MASK;
+		}
+		while (r != rr) {
+			sample = *c++ - dsp_audio_law_to_s32[q[r]];
+			if (sample < -32768)
+				sample = -32768;
+			else if (sample > 32767)
+				sample = 32767;
+			*d++ = dsp_audio_s16_to_law[sample & 0xffff];
+			    /* conf-rx */
+			r = (r+1) & CMX_BUFF_MASK;
+		}
+	/* -> if echo is enabled */
+	} else {
+		/*
+		 * -> encode conf-data, if tx-data
+		 * is available, mix
+		 */
+		while (r != rr && t != tt) {
+			sample = dsp_audio_law_to_s32[p[t]] + *c++;
+			if (sample < -32768)
+				sample = -32768;
+			else if (sample > 32767)
+				sample = 32767;
+			*d++ = dsp_audio_s16_to_law[sample & 0xffff];
+			    /* conf(echo)+tx */
+			t = (t+1) & CMX_BUFF_MASK;
+			r = (r+1) & CMX_BUFF_MASK;
+		}
+		while (r != rr) {
+			sample = *c++;
+			if (sample < -32768)
+				sample = -32768;
+			else if (sample > 32767)
+				sample = 32767;
+			*d++ = dsp_audio_s16_to_law[sample & 0xffff];
+			    /* conf(echo) */
+			r = (r+1) & CMX_BUFF_MASK;
+		}
+	}
+	dsp->tx_R = t;
+	goto send_packet;
+
+send_packet:
+	/*
+	 * send tx-data if enabled - don't filter,
+	 * becuase we want what we send, not what we filtered
+	 */
+	if (dsp->tx_data) {
+		/* PREPARE RESULT */
+		txskb = mI_alloc_skb(len, GFP_ATOMIC);
+		if (!txskb) {
+			printk(KERN_ERR
+			    "FATAL ERROR in mISDN_dsp.o: "
+			    "cannot alloc %d bytes\n", len);
+		} else {
+			thh = mISDN_HEAD_P(txskb);
+			thh->prim = DL_DATA_REQ;
+			thh->id = 0;
+			memcpy(skb_put(txskb, len), nskb->data+preload, len);
+			/* queue (trigger later) */
+			skb_queue_tail(&dsp->sendq, txskb);
+		}
+	}
+	/* adjust volume */
+	if (dsp->tx_volume)
+		dsp_change_volume(nskb, dsp->tx_volume);
+	/* pipeline */
+	if (dsp->pipeline.inuse)
+		dsp_pipeline_process_tx(&dsp->pipeline, nskb->data, nskb->len);
+	/* crypt */
+	if (dsp->bf_enable)
+		dsp_bf_encrypt(dsp, nskb->data, nskb->len);
+	/* queue and trigger */
+	skb_queue_tail(&dsp->sendq, nskb);
+	schedule_work(&dsp->workq);
+}
+
+u32	samplecount;
+struct timer_list dsp_spl_tl;
+u32	dsp_spl_jiffies; /* calculate the next time to fire */
+u32	dsp_start_jiffies; /* jiffies at the time, the calculation begins */
+struct timeval dsp_start_tv; /* time at start of calculation */
+
+void
+dsp_cmx_send(void *arg)
+{
+	struct dsp_conf *conf;
+	struct dsp_conf_member *member;
+	struct dsp *dsp;
+	int mustmix, members;
+	s32 mixbuffer[MAX_POLL+100], *c;
+	u8 *p, *q;
+	int r, rr;
+	int jittercheck = 0, delay, i;
+	u_long flags;
+	struct timeval tv;
+	u32 elapsed;
+	s16 length;
+
+	/* lock */
+	spin_lock_irqsave(&dsp_lock, flags);
+
+	if (!dsp_start_tv.tv_sec) {
+		do_gettimeofday(&dsp_start_tv);
+		length = dsp_poll;
+	} else {
+		do_gettimeofday(&tv);
+		elapsed = ((tv.tv_sec - dsp_start_tv.tv_sec) * 8000)
+		    + ((s32)(tv.tv_usec / 125) - (dsp_start_tv.tv_usec / 125));
+		dsp_start_tv.tv_sec = tv.tv_sec;
+		dsp_start_tv.tv_usec = tv.tv_usec;
+		length = elapsed;
+	}
+	if (length > MAX_POLL + 100)
+		length = MAX_POLL + 100;
+/* printk(KERN_DEBUG "len=%d dsp_count=0x%x.%04x dsp_poll_diff=0x%x.%04x\n",
+ length, dsp_count >> 16, dsp_count & 0xffff, dsp_poll_diff >> 16,
+ dsp_poll_diff & 0xffff);
+ */
+
+	/*
+	 * check if jitter needs to be checked
+	 * (this is about every second = 8192 samples)
+	 */
+	samplecount += length;
+	if ((samplecount & 8191) < length)
+		jittercheck = 1;
+
+	/* loop all members that do not require conference mixing */
+	list_for_each_entry(dsp, &dsp_ilist, list) {
+		if (dsp->hdlc)
+			continue;
+		conf = dsp->conf;
+		mustmix = 0;
+		members = 0;
+		if (conf) {
+			members = count_list_member(&conf->mlist);
+#ifdef CMX_CONF_DEBUG
+			if (conf->software && members > 1)
+#else
+			if (conf->software && members > 2)
+#endif
+				mustmix = 1;
+		}
+
+		/* transmission required */
+		if (!mustmix) {
+			dsp_cmx_send_member(dsp, length, mixbuffer, members);
+
+			/*
+			 * unused mixbuffer is given to prevent a
+			 * potential null-pointer-bug
+			 */
+		}
+	}
+
+	/* loop all members that require conference mixing */
+	list_for_each_entry(conf, &conf_ilist, list) {
+		/* count members and check hardware */
+		members = count_list_member(&conf->mlist);
+#ifdef CMX_CONF_DEBUG
+		if (conf->software && members > 1) {
+#else
+		if (conf->software && members > 2) {
+#endif
+			/* check for hdlc conf */
+			member = list_entry(conf->mlist.next,
+				struct dsp_conf_member, list);
+			if (member->dsp->hdlc)
+				continue;
+			/* mix all data */
+			memset(mixbuffer, 0, length*sizeof(s32));
+			list_for_each_entry(member, &conf->mlist, list) {
+				dsp = member->dsp;
+				/* get range of data to mix */
+				c = mixbuffer;
+				q = dsp->rx_buff;
+				r = dsp->rx_R;
+				rr = (r + length) & CMX_BUFF_MASK;
+				/* add member's data */
+				while (r != rr) {
+					*c++ += dsp_audio_law_to_s32[q[r]];
+					r = (r+1) & CMX_BUFF_MASK;
+				}
+			}
+
+			/* process each member */
+			list_for_each_entry(member, &conf->mlist, list) {
+				/* transmission */
+				dsp_cmx_send_member(member->dsp, length,
+				    mixbuffer, members);
+			}
+		}
+	}
+
+	/* delete rx-data, increment buffers, change pointers */
+	list_for_each_entry(dsp, &dsp_ilist, list) {
+		if (dsp->hdlc)
+			continue;
+		p = dsp->rx_buff;
+		q = dsp->tx_buff;
+		r = dsp->rx_R;
+		/* move receive pointer when receiving */
+		if (!dsp->rx_is_off) {
+			rr = (r + length) & CMX_BUFF_MASK;
+			/* delete rx-data */
+			while (r != rr) {
+				p[r] = dsp_silence;
+				r = (r+1) & CMX_BUFF_MASK;
+			}
+			/* increment rx-buffer pointer */
+			dsp->rx_R = r; /* write incremented read pointer */
+		}
+
+		/* check current rx_delay */
+		delay = (dsp->rx_W-dsp->rx_R) & CMX_BUFF_MASK;
+		if (delay >= CMX_BUFF_HALF)
+			delay = 0; /* will be the delay before next write */
+		/* check for lower delay */
+		if (delay < dsp->rx_delay[0])
+			dsp->rx_delay[0] = delay;
+		/* check current tx_delay */
+		delay = (dsp->tx_W-dsp->tx_R) & CMX_BUFF_MASK;
+		if (delay >= CMX_BUFF_HALF)
+			delay = 0; /* will be the delay before next write */
+		/* check for lower delay */
+		if (delay < dsp->tx_delay[0])
+			dsp->tx_delay[0] = delay;
+		if (jittercheck) {
+			/* find the lowest of all rx_delays */
+			delay = dsp->rx_delay[0];
+			i = 1;
+			while (i < MAX_SECONDS_JITTER_CHECK) {
+				if (delay > dsp->rx_delay[i])
+					delay = dsp->rx_delay[i];
+				i++;
+			}
+			/*
+			 * remove rx_delay only if we have delay AND we
+			 * have not preset cmx_delay
+			 */
+			if (delay && !dsp->cmx_delay) {
+				if (dsp_debug & DEBUG_DSP_CMX)
+					printk(KERN_DEBUG
+					    "%s lowest rx_delay of %d bytes for"
+					    " dsp %s are now removed.\n",
+					    __func__, delay,
+					    dsp->name);
+				r = dsp->rx_R;
+				rr = (r + delay) & CMX_BUFF_MASK;
+				/* delete rx-data */
+				while (r != rr) {
+					p[r] = dsp_silence;
+					r = (r+1) & CMX_BUFF_MASK;
+				}
+				/* increment rx-buffer pointer */
+				dsp->rx_R = r;
+				    /* write incremented read pointer */
+			}
+			/* find the lowest of all tx_delays */
+			delay = dsp->tx_delay[0];
+			i = 1;
+			while (i < MAX_SECONDS_JITTER_CHECK) {
+				if (delay > dsp->tx_delay[i])
+					delay = dsp->tx_delay[i];
+				i++;
+			}
+			/*
+			 * remove delay only if we have delay AND we
+			 * have enabled tx_dejitter
+			 */
+			if (delay && dsp->tx_dejitter) {
+				if (dsp_debug & DEBUG_DSP_CMX)
+					printk(KERN_DEBUG
+					    "%s lowest tx_delay of %d bytes for"
+					    " dsp %s are now removed.\n",
+					    __func__, delay,
+					    dsp->name);
+				r = dsp->tx_R;
+				rr = (r + delay) & CMX_BUFF_MASK;
+				/* delete tx-data */
+				while (r != rr) {
+					q[r] = dsp_silence;
+					r = (r+1) & CMX_BUFF_MASK;
+				}
+				/* increment rx-buffer pointer */
+				dsp->tx_R = r;
+				    /* write incremented read pointer */
+			}
+			/* scroll up delays */
+			i = MAX_SECONDS_JITTER_CHECK - 1;
+			while (i) {
+				dsp->rx_delay[i] = dsp->rx_delay[i-1];
+				dsp->tx_delay[i] = dsp->tx_delay[i-1];
+				i--;
+			}
+			dsp->tx_delay[0] = CMX_BUFF_HALF; /* (infinite) delay */
+			dsp->rx_delay[0] = CMX_BUFF_HALF; /* (infinite) delay */
+		}
+	}
+
+	/* if next event would be in the past ... */
+	if ((s32)(dsp_spl_jiffies+dsp_tics-jiffies) <= 0)
+		dsp_spl_jiffies = jiffies + 1;
+	else
+		dsp_spl_jiffies += dsp_tics;
+
+	dsp_spl_tl.expires = dsp_spl_jiffies;
+	add_timer(&dsp_spl_tl);
+
+	/* unlock */
+	spin_unlock_irqrestore(&dsp_lock, flags);
+}
+
+/*
+ * audio data is transmitted from upper layer to the dsp
+ */
+void
+dsp_cmx_transmit(struct dsp *dsp, struct sk_buff *skb)
+{
+	u_int w, ww;
+	u8 *d, *p;
+	int space; /* todo: , l = skb->len; */
+#ifdef CMX_TX_DEBUG
+	char debugbuf[256] = "";
+#endif
+
+	/* check if there is enough space, and then copy */
+	w = dsp->tx_W;
+	ww = dsp->tx_R;
+	p = dsp->tx_buff;
+	d = skb->data;
+	space = ww-w;
+	if (space <= 0)
+		space += CMX_BUFF_SIZE;
+	/* write-pointer should not overrun nor reach read pointer */
+	if (space-1 < skb->len)
+		/* write to the space we have left */
+		ww = (ww - 1) & CMX_BUFF_MASK;
+	else
+		/* write until all byte are copied */
+		ww = (w + skb->len) & CMX_BUFF_MASK;
+	dsp->tx_W = ww;
+
+	/* show current buffer */
+#ifdef CMX_DEBUG
+	printk(KERN_DEBUG
+	    "cmx_transmit(dsp=%lx) %d bytes to 0x%x-0x%x. %s\n",
+	    (u_long)dsp, (ww-w)&CMX_BUFF_MASK, w, ww, dsp->name);
+#endif
+
+	/* copy transmit data to tx-buffer */
+#ifdef CMX_TX_DEBUG
+	sprintf(debugbuf, "TX getting (%04x-%04x)%p: ", w, ww, p);
+#endif
+	while (w != ww) {
+#ifdef CMX_TX_DEBUG
+		if (strlen(debugbuf) < 48)
+			sprintf(debugbuf+strlen(debugbuf), " %02x", *d);
+#endif
+		p[w] = *d++;
+		w = (w+1) & CMX_BUFF_MASK;
+	}
+#ifdef CMX_TX_DEBUG
+	printk(KERN_DEBUG "%s\n", debugbuf);
+#endif
+
+}
+
+/*
+ * hdlc data is received from card and sent to all members.
+ */
+void
+dsp_cmx_hdlc(struct dsp *dsp, struct sk_buff *skb)
+{
+	struct sk_buff *nskb = NULL;
+	struct dsp_conf_member *member;
+	struct mISDNhead *hh;
+
+	/* not if not active */
+	if (!dsp->b_active)
+		return;
+
+	/* check if we have sompen */
+	if (skb->len < 1)
+		return;
+
+	/* no conf */
+	if (!dsp->conf) {
+		/* in case of hardware (echo) */
+		if (dsp->pcm_slot_tx >= 0)
+			return;
+		if (dsp->echo)
+			nskb = skb_clone(skb, GFP_ATOMIC);
+			if (nskb) {
+				hh = mISDN_HEAD_P(nskb);
+				hh->prim = PH_DATA_REQ;
+				hh->id = 0;
+				skb_queue_tail(&dsp->sendq, nskb);
+				schedule_work(&dsp->workq);
+			}
+		return;
+	}
+	/* in case of hardware conference */
+	if (dsp->conf->hardware)
+		return;
+	list_for_each_entry(member, &dsp->conf->mlist, list) {
+		if (dsp->echo || member->dsp != dsp) {
+			nskb = skb_clone(skb, GFP_ATOMIC);
+			if (nskb) {
+				hh = mISDN_HEAD_P(nskb);
+				hh->prim = PH_DATA_REQ;
+				hh->id = 0;
+				skb_queue_tail(&member->dsp->sendq, nskb);
+				schedule_work(&member->dsp->workq);
+			}
+		}
+	}
+}
+
+
diff --git a/drivers/isdn/mISDN/dsp_core.c b/drivers/isdn/mISDN/dsp_core.c
new file mode 100644
index 000000000000..2f10ed82c0db
--- /dev/null
+++ b/drivers/isdn/mISDN/dsp_core.c
@@ -0,0 +1,1191 @@
+/*
+ * Author       Andreas Eversberg (jolly@eversberg.eu)
+ * Based on source code structure by
+ *		Karsten Keil (keil@isdn4linux.de)
+ *
+ *		This file is (c) under GNU PUBLIC LICENSE
+ *		For changes and modifications please read
+ *		../../../Documentation/isdn/mISDN.cert
+ *
+ * Thanks to    Karsten Keil (great drivers)
+ *              Cologne Chip (great chips)
+ *
+ * This module does:
+ *		Real-time tone generation
+ *		DTMF detection
+ *		Real-time cross-connection and conferrence
+ *		Compensate jitter due to system load and hardware fault.
+ *		All features are done in kernel space and will be realized
+ *		using hardware, if available and supported by chip set.
+ *		Blowfish encryption/decryption
+ */
+
+/* STRUCTURE:
+ *
+ * The dsp module provides layer 2 for b-channels (64kbit). It provides
+ * transparent audio forwarding with special digital signal processing:
+ *
+ * - (1) generation of tones
+ * - (2) detection of dtmf tones
+ * - (3) crossconnecting and conferences (clocking)
+ * - (4) echo generation for delay test
+ * - (5) volume control
+ * - (6) disable receive data
+ * - (7) pipeline
+ * - (8) encryption/decryption
+ *
+ * Look:
+ *             TX            RX
+ *         ------upper layer------
+ *             |             ^
+ *             |             |(6)
+ *             v             |
+ *       +-----+-------------+-----+
+ *       |(3)(4)                   |
+ *       |           CMX           |
+ *       |                         |
+ *       |           +-------------+
+ *       |           |       ^
+ *       |           |       |
+ *       |+---------+|  +----+----+
+ *       ||(1)      ||  |(2)      |
+ *       ||         ||  |         |
+ *       ||  Tones  ||  |  DTMF   |
+ *       ||         ||  |         |
+ *       ||         ||  |         |
+ *       |+----+----+|  +----+----+
+ *       +-----+-----+       ^
+ *             |             |
+ *             v             |
+ *        +----+----+   +----+----+
+ *        |(5)      |   |(5)      |
+ *        |         |   |         |
+ *        |TX Volume|   |RX Volume|
+ *        |         |   |         |
+ *        |         |   |         |
+ *        +----+----+   +----+----+
+ *             |             ^
+ *             |             |
+ *             v             |
+ *        +----+-------------+----+
+ *        |(7)                    |
+ *        |                       |
+ *        |  Pipeline Processing  |
+ *        |                       |
+ *        |                       |
+ *        +----+-------------+----+
+ *             |             ^
+ *             |             |
+ *             v             |
+ *        +----+----+   +----+----+
+ *        |(8)      |   |(8)      |
+ *        |         |   |         |
+ *        | Encrypt |   | Decrypt |
+ *        |         |   |         |
+ *        |         |   |         |
+ *        +----+----+   +----+----+
+ *             |             ^
+ *             |             |
+ *             v             |
+ *         ------card  layer------
+ *             TX            RX
+ *
+ * Above you can see the logical data flow. If software is used to do the
+ * process, it is actually the real data flow. If hardware is used, data
+ * may not flow, but hardware commands to the card, to provide the data flow
+ * as shown.
+ *
+ * NOTE: The channel must be activated in order to make dsp work, even if
+ * no data flow to the upper layer is intended. Activation can be done
+ * after and before controlling the setting using PH_CONTROL requests.
+ *
+ * DTMF: Will be detected by hardware if possible. It is done before CMX
+ * processing.
+ *
+ * Tones: Will be generated via software if endless looped audio fifos are
+ * not supported by hardware. Tones will override all data from CMX.
+ * It is not required to join a conference to use tones at any time.
+ *
+ * CMX: Is transparent when not used. When it is used, it will do
+ * crossconnections and conferences via software if not possible through
+ * hardware. If hardware capability is available, hardware is used.
+ *
+ * Echo: Is generated by CMX and is used to check performane of hard and
+ * software CMX.
+ *
+ * The CMX has special functions for conferences with one, two and more
+ * members. It will allow different types of data flow. Receive and transmit
+ * data to/form upper layer may be swithed on/off individually without loosing
+ * features of CMX, Tones and DTMF.
+ *
+ * Echo Cancellation: Sometimes we like to cancel echo from the interface.
+ * Note that a VoIP call may not have echo caused by the IP phone. The echo
+ * is generated by the telephone line connected to it. Because the delay
+ * is high, it becomes an echo. RESULT: Echo Cachelation is required if
+ * both echo AND delay is applied to an interface.
+ * Remember that software CMX always generates a more or less delay.
+ *
+ * If all used features can be realized in hardware, and if transmit and/or
+ * receive data ist disabled, the card may not send/receive any data at all.
+ * Not receiving is usefull if only announcements are played. Not sending is
+ * usefull if an answering machine records audio. Not sending and receiving is
+ * usefull during most states of the call. If supported by hardware, tones
+ * will be played without cpu load. Small PBXs and NT-Mode applications will
+ * not need expensive hardware when processing calls.
+ *
+ *
+ * LOCKING:
+ *
+ * When data is received from upper or lower layer (card), the complete dsp
+ * module is locked by a global lock.  This lock MUST lock irq, because it
+ * must lock timer events by DSP poll timer.
+ * When data is ready to be transmitted down, the data is queued and sent
+ * outside lock and timer event.
+ * PH_CONTROL must not change any settings, join or split conference members
+ * during process of data.
+ *
+ * HDLC:
+ *
+ * It works quite the same as transparent, except that HDLC data is forwarded
+ * to all other conference members if no hardware bridging is possible.
+ * Send data will be writte to sendq. Sendq will be sent if confirm is received.
+ * Conference cannot join, if one member is not hdlc.
+ *
+ */
+
+#include <linux/delay.h>
+#include <linux/mISDNif.h>
+#include <linux/mISDNdsp.h>
+#include <linux/module.h>
+#include <linux/vmalloc.h>
+#include "core.h"
+#include "dsp.h"
+
+const char *mISDN_dsp_revision = "2.0";
+
+static int debug;
+static int options;
+static int poll;
+static int dtmfthreshold = 100;
+
+MODULE_AUTHOR("Andreas Eversberg");
+module_param(debug, uint, S_IRUGO | S_IWUSR);
+module_param(options, uint, S_IRUGO | S_IWUSR);
+module_param(poll, uint, S_IRUGO | S_IWUSR);
+module_param(dtmfthreshold, uint, S_IRUGO | S_IWUSR);
+MODULE_LICENSE("GPL");
+
+/*int spinnest = 0;*/
+
+spinlock_t dsp_lock; /* global dsp lock */
+struct list_head dsp_ilist;
+struct list_head conf_ilist;
+int dsp_debug;
+int dsp_options;
+int dsp_poll, dsp_tics;
+
+/* check if rx may be turned off or must be turned on */
+static void
+dsp_rx_off_member(struct dsp *dsp)
+{
+	struct mISDN_ctrl_req	cq;
+	int rx_off = 1;
+
+	if (!dsp->features_rx_off)
+		return;
+
+	/* not disabled */
+	if (!dsp->rx_disabled)
+		rx_off = 0;
+	/* software dtmf */
+	else if (dsp->dtmf.software)
+		rx_off = 0;
+	/* echo in software */
+	else if (dsp->echo && dsp->pcm_slot_tx < 0)
+		rx_off = 0;
+	/* bridge in software */
+	else if (dsp->conf) {
+		if (dsp->conf->software)
+			rx_off = 0;
+	}
+
+	if (rx_off == dsp->rx_is_off)
+		return;
+
+	if (!dsp->ch.peer) {
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: no peer, no rx_off\n",
+				__func__);
+		return;
+	}
+	cq.op = MISDN_CTRL_RX_OFF;
+	cq.p1 = rx_off;
+	if (dsp->ch.peer->ctrl(dsp->ch.peer, CONTROL_CHANNEL, &cq)) {
+		printk(KERN_DEBUG "%s: 2nd CONTROL_CHANNEL failed\n",
+			__func__);
+		return;
+	}
+	dsp->rx_is_off = rx_off;
+	if (dsp_debug & DEBUG_DSP_CORE)
+		printk(KERN_DEBUG "%s: %s set rx_off = %d\n",
+			__func__, dsp->name, rx_off);
+}
+static void
+dsp_rx_off(struct dsp *dsp)
+{
+	struct dsp_conf_member	*member;
+
+	if (dsp_options & DSP_OPT_NOHARDWARE)
+		return;
+
+	/* no conf */
+	if (!dsp->conf) {
+		dsp_rx_off_member(dsp);
+		return;
+	}
+	/* check all members in conf */
+	list_for_each_entry(member, &dsp->conf->mlist, list) {
+		dsp_rx_off_member(member->dsp);
+	}
+}
+
+static int
+dsp_control_req(struct dsp *dsp, struct mISDNhead *hh, struct sk_buff *skb)
+{
+	struct		sk_buff *nskb;
+	int ret = 0;
+	int cont;
+	u8 *data;
+	int len;
+
+	if (skb->len < sizeof(int))
+		printk(KERN_ERR "%s: PH_CONTROL message too short\n", __func__);
+	cont = *((int *)skb->data);
+	len = skb->len - sizeof(int);
+	data = skb->data + sizeof(int);
+
+	switch (cont) {
+	case DTMF_TONE_START: /* turn on DTMF */
+		if (dsp->hdlc) {
+			ret = -EINVAL;
+			break;
+		}
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: start dtmf\n", __func__);
+		if (len == sizeof(int)) {
+			printk(KERN_NOTICE "changing DTMF Threshold "
+				"to %d\n", *((int *)data));
+			dsp->dtmf.treshold = (*(int *)data) * 10000;
+		}
+		/* init goertzel */
+		dsp_dtmf_goertzel_init(dsp);
+
+		/* check dtmf hardware */
+		dsp_dtmf_hardware(dsp);
+		break;
+	case DTMF_TONE_STOP: /* turn off DTMF */
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: stop dtmf\n", __func__);
+		dsp->dtmf.hardware = 0;
+		dsp->dtmf.software = 0;
+		break;
+	case DSP_CONF_JOIN: /* join / update conference */
+		if (len < sizeof(int)) {
+			ret = -EINVAL;
+			break;
+		}
+		if (*((u32 *)data) == 0)
+			goto conf_split;
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: join conference %d\n",
+				__func__, *((u32 *)data));
+		ret = dsp_cmx_conf(dsp, *((u32 *)data));
+			/* dsp_cmx_hardware will also be called here */
+		dsp_rx_off(dsp);
+		if (dsp_debug & DEBUG_DSP_CMX)
+			dsp_cmx_debug(dsp);
+		break;
+	case DSP_CONF_SPLIT: /* remove from conference */
+conf_split:
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: release conference\n", __func__);
+		ret = dsp_cmx_conf(dsp, 0);
+			/* dsp_cmx_hardware will also be called here */
+		if (dsp_debug & DEBUG_DSP_CMX)
+			dsp_cmx_debug(dsp);
+		dsp_rx_off(dsp);
+		break;
+	case DSP_TONE_PATT_ON: /* play tone */
+		if (dsp->hdlc) {
+			ret = -EINVAL;
+			break;
+		}
+		if (len < sizeof(int)) {
+			ret = -EINVAL;
+			break;
+		}
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: turn tone 0x%x on\n",
+				__func__, *((int *)skb->data));
+		ret = dsp_tone(dsp, *((int *)data));
+		if (!ret) {
+			dsp_cmx_hardware(dsp->conf, dsp);
+			dsp_rx_off(dsp);
+		}
+		if (!dsp->tone.tone)
+			goto tone_off;
+		break;
+	case DSP_TONE_PATT_OFF: /* stop tone */
+		if (dsp->hdlc) {
+			ret = -EINVAL;
+			break;
+		}
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: turn tone off\n", __func__);
+		dsp_tone(dsp, 0);
+		dsp_cmx_hardware(dsp->conf, dsp);
+		dsp_rx_off(dsp);
+		/* reset tx buffers (user space data) */
+tone_off:
+		dsp->rx_W = 0;
+		dsp->rx_R = 0;
+		break;
+	case DSP_VOL_CHANGE_TX: /* change volume */
+		if (dsp->hdlc) {
+			ret = -EINVAL;
+			break;
+		}
+		if (len < sizeof(int)) {
+			ret = -EINVAL;
+			break;
+		}
+		dsp->tx_volume = *((int *)data);
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: change tx vol to %d\n",
+				__func__, dsp->tx_volume);
+		dsp_cmx_hardware(dsp->conf, dsp);
+		dsp_dtmf_hardware(dsp);
+		dsp_rx_off(dsp);
+		break;
+	case DSP_VOL_CHANGE_RX: /* change volume */
+		if (dsp->hdlc) {
+			ret = -EINVAL;
+			break;
+		}
+		if (len < sizeof(int)) {
+			ret = -EINVAL;
+			break;
+		}
+		dsp->rx_volume = *((int *)data);
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: change rx vol to %d\n",
+				__func__, dsp->tx_volume);
+		dsp_cmx_hardware(dsp->conf, dsp);
+		dsp_dtmf_hardware(dsp);
+		dsp_rx_off(dsp);
+		break;
+	case DSP_ECHO_ON: /* enable echo */
+		dsp->echo = 1; /* soft echo */
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: enable cmx-echo\n", __func__);
+		dsp_cmx_hardware(dsp->conf, dsp);
+		dsp_rx_off(dsp);
+		if (dsp_debug & DEBUG_DSP_CMX)
+			dsp_cmx_debug(dsp);
+		break;
+	case DSP_ECHO_OFF: /* disable echo */
+		dsp->echo = 0;
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: disable cmx-echo\n", __func__);
+		dsp_cmx_hardware(dsp->conf, dsp);
+		dsp_rx_off(dsp);
+		if (dsp_debug & DEBUG_DSP_CMX)
+			dsp_cmx_debug(dsp);
+		break;
+	case DSP_RECEIVE_ON: /* enable receive to user space */
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: enable receive to user "
+				"space\n", __func__);
+		dsp->rx_disabled = 0;
+		dsp_rx_off(dsp);
+		break;
+	case DSP_RECEIVE_OFF: /* disable receive to user space */
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: disable receive to "
+				"user space\n", __func__);
+		dsp->rx_disabled = 1;
+		dsp_rx_off(dsp);
+		break;
+	case DSP_MIX_ON: /* enable mixing of tx data */
+		if (dsp->hdlc) {
+			ret = -EINVAL;
+			break;
+		}
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: enable mixing of "
+				"tx-data with conf mebers\n", __func__);
+		dsp->tx_mix = 1;
+		dsp_cmx_hardware(dsp->conf, dsp);
+		dsp_rx_off(dsp);
+		if (dsp_debug & DEBUG_DSP_CMX)
+			dsp_cmx_debug(dsp);
+		break;
+	case DSP_MIX_OFF: /* disable mixing of tx data */
+		if (dsp->hdlc) {
+			ret = -EINVAL;
+			break;
+		}
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: disable mixing of "
+				"tx-data with conf mebers\n", __func__);
+		dsp->tx_mix = 0;
+		dsp_cmx_hardware(dsp->conf, dsp);
+		dsp_rx_off(dsp);
+		if (dsp_debug & DEBUG_DSP_CMX)
+			dsp_cmx_debug(dsp);
+		break;
+	case DSP_TXDATA_ON: /* enable txdata */
+		dsp->tx_data = 1;
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: enable tx-data\n", __func__);
+		dsp_cmx_hardware(dsp->conf, dsp);
+		dsp_rx_off(dsp);
+		if (dsp_debug & DEBUG_DSP_CMX)
+			dsp_cmx_debug(dsp);
+		break;
+	case DSP_TXDATA_OFF: /* disable txdata */
+		dsp->tx_data = 0;
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: disable tx-data\n", __func__);
+		dsp_cmx_hardware(dsp->conf, dsp);
+		dsp_rx_off(dsp);
+		if (dsp_debug & DEBUG_DSP_CMX)
+			dsp_cmx_debug(dsp);
+		break;
+	case DSP_DELAY: /* use delay algorithm instead of dynamic
+			   jitter algorithm */
+		if (dsp->hdlc) {
+			ret = -EINVAL;
+			break;
+		}
+		if (len < sizeof(int)) {
+			ret = -EINVAL;
+			break;
+		}
+		dsp->cmx_delay = (*((int *)data)) << 3;
+			/* miliseconds to samples */
+		if (dsp->cmx_delay >= (CMX_BUFF_HALF>>1))
+			/* clip to half of maximum usable buffer
+			(half of half buffer) */
+			dsp->cmx_delay = (CMX_BUFF_HALF>>1) - 1;
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: use delay algorithm to "
+				"compensate jitter (%d samples)\n",
+				__func__, dsp->cmx_delay);
+		break;
+	case DSP_JITTER: /* use dynamic jitter algorithm instead of
+		    delay algorithm */
+		if (dsp->hdlc) {
+			ret = -EINVAL;
+			break;
+		}
+		dsp->cmx_delay = 0;
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: use jitter algorithm to "
+				"compensate jitter\n", __func__);
+		break;
+	case DSP_TX_DEJITTER: /* use dynamic jitter algorithm for tx-buffer */
+		if (dsp->hdlc) {
+			ret = -EINVAL;
+			break;
+		}
+		dsp->tx_dejitter = 1;
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: use dejitter on TX "
+				"buffer\n", __func__);
+		break;
+	case DSP_TX_DEJ_OFF: /* use tx-buffer without dejittering*/
+		if (dsp->hdlc) {
+			ret = -EINVAL;
+			break;
+		}
+		dsp->tx_dejitter = 0;
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: use TX buffer without "
+				"dejittering\n", __func__);
+		break;
+	case DSP_PIPELINE_CFG:
+		if (dsp->hdlc) {
+			ret = -EINVAL;
+			break;
+		}
+		if (len > 0 && ((char *)data)[len - 1]) {
+			printk(KERN_DEBUG "%s: pipeline config string "
+				"is not NULL terminated!\n", __func__);
+			ret = -EINVAL;
+		} else {
+			dsp->pipeline.inuse = 1;
+			dsp_cmx_hardware(dsp->conf, dsp);
+			ret = dsp_pipeline_build(&dsp->pipeline,
+				len > 0 ? (char *)data : NULL);
+			dsp_cmx_hardware(dsp->conf, dsp);
+			dsp_rx_off(dsp);
+		}
+		break;
+	case DSP_BF_ENABLE_KEY: /* turn blowfish on */
+		if (dsp->hdlc) {
+			ret = -EINVAL;
+			break;
+		}
+		if (len < 4 || len > 56) {
+			ret = -EINVAL;
+			break;
+		}
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: turn blowfish on (key "
+				"not shown)\n", __func__);
+		ret = dsp_bf_init(dsp, (u8 *)data, len);
+		/* set new cont */
+		if (!ret)
+			cont = DSP_BF_ACCEPT;
+		else
+			cont = DSP_BF_REJECT;
+		/* send indication if it worked to set it */
+		nskb = _alloc_mISDN_skb(PH_CONTROL_IND, MISDN_ID_ANY,
+			sizeof(int), &cont, GFP_ATOMIC);
+		if (nskb) {
+			if (dsp->up) {
+				if (dsp->up->send(dsp->up, nskb))
+					dev_kfree_skb(nskb);
+			} else
+				dev_kfree_skb(nskb);
+		}
+		if (!ret) {
+			dsp_cmx_hardware(dsp->conf, dsp);
+			dsp_dtmf_hardware(dsp);
+			dsp_rx_off(dsp);
+		}
+		break;
+	case DSP_BF_DISABLE: /* turn blowfish off */
+		if (dsp->hdlc) {
+			ret = -EINVAL;
+			break;
+		}
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: turn blowfish off\n", __func__);
+		dsp_bf_cleanup(dsp);
+		dsp_cmx_hardware(dsp->conf, dsp);
+		dsp_dtmf_hardware(dsp);
+		dsp_rx_off(dsp);
+		break;
+	default:
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: ctrl req %x unhandled\n",
+				__func__, cont);
+		ret = -EINVAL;
+	}
+	return ret;
+}
+
+static void
+get_features(struct mISDNchannel *ch)
+{
+	struct dsp		*dsp = container_of(ch, struct dsp, ch);
+	struct mISDN_ctrl_req	cq;
+
+	if (dsp_options & DSP_OPT_NOHARDWARE)
+		return;
+	if (!ch->peer) {
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: no peer, no features\n",
+				__func__);
+		return;
+	}
+	memset(&cq, 0, sizeof(cq));
+	cq.op = MISDN_CTRL_GETOP;
+	if (ch->peer->ctrl(ch->peer, CONTROL_CHANNEL, &cq) < 0) {
+		printk(KERN_DEBUG "%s: CONTROL_CHANNEL failed\n",
+			__func__);
+		return;
+	}
+	if (cq.op & MISDN_CTRL_RX_OFF)
+		dsp->features_rx_off = 1;
+	if ((cq.op & MISDN_CTRL_HW_FEATURES_OP)) {
+		cq.op = MISDN_CTRL_HW_FEATURES;
+		*((u_long *)&cq.p1) = (u_long)&dsp->features;
+		if (ch->peer->ctrl(ch->peer, CONTROL_CHANNEL, &cq)) {
+			printk(KERN_DEBUG "%s: 2nd CONTROL_CHANNEL failed\n",
+				__func__);
+		}
+	} else
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: features not supported for %s\n",
+				__func__, dsp->name);
+}
+
+static int
+dsp_function(struct mISDNchannel *ch,  struct sk_buff *skb)
+{
+	struct dsp			*dsp = container_of(ch, struct dsp, ch);
+	struct mISDNhead	*hh;
+	int			ret = 0;
+	u8			*digits;
+	int			cont;
+	struct			sk_buff *nskb;
+	u_long			flags;
+
+	hh = mISDN_HEAD_P(skb);
+	switch (hh->prim) {
+	/* FROM DOWN */
+	case (PH_DATA_CNF):
+		dsp->data_pending = 0;
+		/* trigger next hdlc frame, if any */
+		if (dsp->hdlc) {
+			spin_lock_irqsave(&dsp_lock, flags);
+			if (dsp->b_active)
+				schedule_work(&dsp->workq);
+			spin_unlock_irqrestore(&dsp_lock, flags);
+		}
+		break;
+	case (PH_DATA_IND):
+	case (DL_DATA_IND):
+		if (skb->len < 1) {
+			ret = -EINVAL;
+			break;
+		}
+		if (dsp->rx_is_off) {
+			if (dsp_debug & DEBUG_DSP_CORE)
+				printk(KERN_DEBUG "%s: rx-data during rx_off"
+					" for %s\n",
+				__func__, dsp->name);
+		}
+		if (dsp->hdlc) {
+			/* hdlc */
+			spin_lock_irqsave(&dsp_lock, flags);
+			dsp_cmx_hdlc(dsp, skb);
+			spin_unlock_irqrestore(&dsp_lock, flags);
+			if (dsp->rx_disabled) {
+				/* if receive is not allowed */
+				break;
+			}
+			hh->prim = DL_DATA_IND;
+			if (dsp->up)
+				return dsp->up->send(dsp->up, skb);
+			break;
+		}
+
+		/* decrypt if enabled */
+		if (dsp->bf_enable)
+			dsp_bf_decrypt(dsp, skb->data, skb->len);
+		/* pipeline */
+		if (dsp->pipeline.inuse)
+			dsp_pipeline_process_rx(&dsp->pipeline, skb->data,
+				skb->len);
+		/* change volume if requested */
+		if (dsp->rx_volume)
+			dsp_change_volume(skb, dsp->rx_volume);
+
+		/* check if dtmf soft decoding is turned on */
+		if (dsp->dtmf.software) {
+			digits = dsp_dtmf_goertzel_decode(dsp, skb->data,
+				skb->len, (dsp_options&DSP_OPT_ULAW)?1:0);
+			while (*digits) {
+				if (dsp_debug & DEBUG_DSP_DTMF)
+					printk(KERN_DEBUG "%s: digit"
+					    "(%c) to layer %s\n",
+					    __func__, *digits, dsp->name);
+				cont = DTMF_TONE_VAL | *digits;
+				nskb = _alloc_mISDN_skb(PH_CONTROL_IND,
+				    MISDN_ID_ANY, sizeof(int), &cont,
+				    GFP_ATOMIC);
+				if (nskb) {
+					if (dsp->up) {
+						if (dsp->up->send(
+						    dsp->up, nskb))
+						dev_kfree_skb(nskb);
+					} else
+						dev_kfree_skb(nskb);
+				}
+				digits++;
+			}
+		}
+		/* we need to process receive data if software */
+		spin_lock_irqsave(&dsp_lock, flags);
+		if (dsp->pcm_slot_tx < 0 && dsp->pcm_slot_rx < 0) {
+			/* process data from card at cmx */
+			dsp_cmx_receive(dsp, skb);
+		}
+		spin_unlock_irqrestore(&dsp_lock, flags);
+
+		if (dsp->rx_disabled) {
+			/* if receive is not allowed */
+			break;
+		}
+		hh->prim = DL_DATA_IND;
+		if (dsp->up)
+			return dsp->up->send(dsp->up, skb);
+		break;
+	case (PH_CONTROL_IND):
+		if (dsp_debug & DEBUG_DSP_DTMFCOEFF)
+			printk(KERN_DEBUG "%s: PH_CONTROL INDICATION "
+				"received: %x (len %d) %s\n", __func__,
+				hh->id, skb->len, dsp->name);
+		switch (hh->id) {
+		case (DTMF_HFC_COEF): /* getting coefficients */
+			if (!dsp->dtmf.hardware) {
+				if (dsp_debug & DEBUG_DSP_DTMFCOEFF)
+					printk(KERN_DEBUG "%s: ignoring DTMF "
+						"coefficients from HFC\n",
+						__func__);
+				break;
+			}
+			digits = dsp_dtmf_goertzel_decode(dsp, skb->data,
+				skb->len, 2);
+			while (*digits) {
+				int k;
+				struct sk_buff *nskb;
+				if (dsp_debug & DEBUG_DSP_DTMF)
+					printk(KERN_DEBUG "%s: digit"
+					    "(%c) to layer %s\n",
+					    __func__, *digits, dsp->name);
+				k = *digits | DTMF_TONE_VAL;
+				nskb = _alloc_mISDN_skb(PH_CONTROL_IND,
+					MISDN_ID_ANY, sizeof(int), &k,
+					GFP_ATOMIC);
+				if (nskb) {
+					if (dsp->up) {
+						if (dsp->up->send(
+						    dsp->up, nskb))
+						dev_kfree_skb(nskb);
+					} else
+						dev_kfree_skb(nskb);
+				}
+				digits++;
+			}
+			break;
+		case (HFC_VOL_CHANGE_TX): /* change volume */
+			if (skb->len != sizeof(int)) {
+				ret = -EINVAL;
+				break;
+			}
+			spin_lock_irqsave(&dsp_lock, flags);
+			dsp->tx_volume = *((int *)skb->data);
+			if (dsp_debug & DEBUG_DSP_CORE)
+				printk(KERN_DEBUG "%s: change tx volume to "
+					"%d\n", __func__, dsp->tx_volume);
+			dsp_cmx_hardware(dsp->conf, dsp);
+			dsp_dtmf_hardware(dsp);
+			dsp_rx_off(dsp);
+			spin_unlock_irqrestore(&dsp_lock, flags);
+			break;
+		default:
+			if (dsp_debug & DEBUG_DSP_CORE)
+				printk(KERN_DEBUG "%s: ctrl ind %x unhandled "
+					"%s\n", __func__, hh->id, dsp->name);
+			ret = -EINVAL;
+		}
+		break;
+	case (PH_ACTIVATE_IND):
+	case (PH_ACTIVATE_CNF):
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: b_channel is now active %s\n",
+				__func__, dsp->name);
+		/* bchannel now active */
+		spin_lock_irqsave(&dsp_lock, flags);
+		dsp->b_active = 1;
+		dsp->data_pending = 0;
+		dsp->rx_init = 1;
+			/* rx_W and rx_R will be adjusted on first frame */
+		dsp->rx_W = 0;
+		dsp->rx_R = 0;
+		memset(dsp->rx_buff, 0, sizeof(dsp->rx_buff));
+		dsp_cmx_hardware(dsp->conf, dsp);
+		dsp_dtmf_hardware(dsp);
+		dsp_rx_off(dsp);
+		spin_unlock_irqrestore(&dsp_lock, flags);
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: done with activation, sending "
+				"confirm to user space. %s\n", __func__,
+				dsp->name);
+		/* send activation to upper layer */
+		hh->prim = DL_ESTABLISH_CNF;
+		if (dsp->up)
+			return dsp->up->send(dsp->up, skb);
+		break;
+	case (PH_DEACTIVATE_IND):
+	case (PH_DEACTIVATE_CNF):
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: b_channel is now inactive %s\n",
+				__func__, dsp->name);
+		/* bchannel now inactive */
+		spin_lock_irqsave(&dsp_lock, flags);
+		dsp->b_active = 0;
+		dsp->data_pending = 0;
+		dsp_cmx_hardware(dsp->conf, dsp);
+		dsp_rx_off(dsp);
+		spin_unlock_irqrestore(&dsp_lock, flags);
+		hh->prim = DL_RELEASE_CNF;
+		if (dsp->up)
+			return dsp->up->send(dsp->up, skb);
+		break;
+	/* FROM UP */
+	case (DL_DATA_REQ):
+	case (PH_DATA_REQ):
+		if (skb->len < 1) {
+			ret = -EINVAL;
+			break;
+		}
+		if (dsp->hdlc) {
+			/* hdlc */
+			spin_lock_irqsave(&dsp_lock, flags);
+			if (dsp->b_active) {
+				skb_queue_tail(&dsp->sendq, skb);
+				schedule_work(&dsp->workq);
+			}
+			spin_unlock_irqrestore(&dsp_lock, flags);
+			return 0;
+		}
+		/* send data to tx-buffer (if no tone is played) */
+		if (!dsp->tone.tone) {
+			spin_lock_irqsave(&dsp_lock, flags);
+			dsp_cmx_transmit(dsp, skb);
+			spin_unlock_irqrestore(&dsp_lock, flags);
+		}
+		break;
+	case (PH_CONTROL_REQ):
+		spin_lock_irqsave(&dsp_lock, flags);
+		ret = dsp_control_req(dsp, hh, skb);
+		spin_unlock_irqrestore(&dsp_lock, flags);
+		break;
+	case (DL_ESTABLISH_REQ):
+	case (PH_ACTIVATE_REQ):
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: activating b_channel %s\n",
+				__func__, dsp->name);
+		if (dsp->dtmf.hardware || dsp->dtmf.software)
+			dsp_dtmf_goertzel_init(dsp);
+		get_features(ch);
+		/* send ph_activate */
+		hh->prim = PH_ACTIVATE_REQ;
+		if (ch->peer)
+			return ch->recv(ch->peer, skb);
+		break;
+	case (DL_RELEASE_REQ):
+	case (PH_DEACTIVATE_REQ):
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: releasing b_channel %s\n",
+				__func__, dsp->name);
+		spin_lock_irqsave(&dsp_lock, flags);
+		dsp->tone.tone = 0;
+		dsp->tone.hardware = 0;
+		dsp->tone.software = 0;
+		if (timer_pending(&dsp->tone.tl))
+			del_timer(&dsp->tone.tl);
+		if (dsp->conf)
+			dsp_cmx_conf(dsp, 0); /* dsp_cmx_hardware will also be
+						 called here */
+		skb_queue_purge(&dsp->sendq);
+		spin_unlock_irqrestore(&dsp_lock, flags);
+		hh->prim = PH_DEACTIVATE_REQ;
+		if (ch->peer)
+			return ch->recv(ch->peer, skb);
+		break;
+	default:
+		if (dsp_debug & DEBUG_DSP_CORE)
+			printk(KERN_DEBUG "%s: msg %x unhandled %s\n",
+				__func__, hh->prim, dsp->name);
+		ret = -EINVAL;
+	}
+	if (!ret)
+		dev_kfree_skb(skb);
+	return ret;
+}
+
+static int
+dsp_ctrl(struct mISDNchannel *ch, u_int cmd, void *arg)
+{
+	struct dsp		*dsp = container_of(ch, struct dsp, ch);
+	u_long		flags;
+	int		err = 0;
+
+	if (debug & DEBUG_DSP_CTRL)
+	printk(KERN_DEBUG "%s:(%x)\n", __func__, cmd);
+
+	switch (cmd) {
+	case OPEN_CHANNEL:
+		break;
+	case CLOSE_CHANNEL:
+		if (dsp->ch.peer)
+			dsp->ch.peer->ctrl(dsp->ch.peer, CLOSE_CHANNEL, NULL);
+
+		/* wait until workqueue has finished,
+		 * must lock here, or we may hit send-process currently
+		 * queueing. */
+		spin_lock_irqsave(&dsp_lock, flags);
+		dsp->b_active = 0;
+		spin_unlock_irqrestore(&dsp_lock, flags);
+		/* MUST not be locked, because it waits until queue is done. */
+		cancel_work_sync(&dsp->workq);
+		spin_lock_irqsave(&dsp_lock, flags);
+		if (timer_pending(&dsp->tone.tl))
+			del_timer(&dsp->tone.tl);
+		skb_queue_purge(&dsp->sendq);
+		if (dsp_debug & DEBUG_DSP_CTRL)
+			printk(KERN_DEBUG "%s: releasing member %s\n",
+				__func__, dsp->name);
+		dsp->b_active = 0;
+		dsp_cmx_conf(dsp, 0); /* dsp_cmx_hardware will also be called
+					 here */
+		dsp_pipeline_destroy(&dsp->pipeline);
+
+		if (dsp_debug & DEBUG_DSP_CTRL)
+			printk(KERN_DEBUG "%s: remove & destroy object %s\n",
+				__func__, dsp->name);
+		list_del(&dsp->list);
+		spin_unlock_irqrestore(&dsp_lock, flags);
+
+		if (dsp_debug & DEBUG_DSP_CTRL)
+			printk(KERN_DEBUG "%s: dsp instance released\n",
+				__func__);
+		vfree(dsp);
+		module_put(THIS_MODULE);
+		break;
+	}
+	return err;
+}
+
+static void
+dsp_send_bh(struct work_struct *work)
+{
+	struct dsp *dsp = container_of(work, struct dsp, workq);
+	struct sk_buff *skb;
+	struct mISDNhead	*hh;
+
+	if (dsp->hdlc && dsp->data_pending)
+		return; /* wait until data has been acknowledged */
+
+	/* send queued data */
+	while ((skb = skb_dequeue(&dsp->sendq))) {
+		/* in locked date, we must have still data in queue */
+		if (dsp->data_pending) {
+			if (dsp_debug & DEBUG_DSP_CORE)
+				printk(KERN_DEBUG "%s: fifo full %s, this is "
+					"no bug!\n", __func__, dsp->name);
+			/* flush transparent data, if not acked */
+			dev_kfree_skb(skb);
+			continue;
+		}
+		hh = mISDN_HEAD_P(skb);
+		if (hh->prim == DL_DATA_REQ) {
+			/* send packet up */
+			if (dsp->up) {
+				if (dsp->up->send(dsp->up, skb))
+					dev_kfree_skb(skb);
+			} else
+				dev_kfree_skb(skb);
+		} else {
+			/* send packet down */
+			if (dsp->ch.peer) {
+				dsp->data_pending = 1;
+				if (dsp->ch.recv(dsp->ch.peer, skb)) {
+					dev_kfree_skb(skb);
+					dsp->data_pending = 0;
+				}
+			} else
+				dev_kfree_skb(skb);
+		}
+	}
+}
+
+static int
+dspcreate(struct channel_req *crq)
+{
+	struct dsp		*ndsp;
+	u_long		flags;
+
+	if (crq->protocol != ISDN_P_B_L2DSP
+	 && crq->protocol != ISDN_P_B_L2DSPHDLC)
+		return -EPROTONOSUPPORT;
+	ndsp = vmalloc(sizeof(struct dsp));
+	if (!ndsp) {
+		printk(KERN_ERR "%s: vmalloc struct dsp failed\n", __func__);
+		return -ENOMEM;
+	}
+	memset(ndsp, 0, sizeof(struct dsp));
+	if (dsp_debug & DEBUG_DSP_CTRL)
+		printk(KERN_DEBUG "%s: creating new dsp instance\n", __func__);
+
+	/* default enabled */
+	INIT_WORK(&ndsp->workq, (void *)dsp_send_bh);
+	skb_queue_head_init(&ndsp->sendq);
+	ndsp->ch.send = dsp_function;
+	ndsp->ch.ctrl = dsp_ctrl;
+	ndsp->up = crq->ch;
+	crq->ch = &ndsp->ch;
+	if (crq->protocol == ISDN_P_B_L2DSP) {
+		crq->protocol = ISDN_P_B_RAW;
+		ndsp->hdlc = 0;
+	} else {
+		crq->protocol = ISDN_P_B_HDLC;
+		ndsp->hdlc = 1;
+	}
+	if (!try_module_get(THIS_MODULE))
+		printk(KERN_WARNING "%s:cannot get module\n",
+			__func__);
+
+	sprintf(ndsp->name, "DSP_C%x(0x%p)",
+		ndsp->up->st->dev->id + 1, ndsp);
+	/* set frame size to start */
+	ndsp->features.hfc_id = -1; /* current PCM id */
+	ndsp->features.pcm_id = -1; /* current PCM id */
+	ndsp->pcm_slot_rx = -1; /* current CPM slot */
+	ndsp->pcm_slot_tx = -1;
+	ndsp->pcm_bank_rx = -1;
+	ndsp->pcm_bank_tx = -1;
+	ndsp->hfc_conf = -1; /* current conference number */
+	/* set tone timer */
+	ndsp->tone.tl.function = (void *)dsp_tone_timeout;
+	ndsp->tone.tl.data = (long) ndsp;
+	init_timer(&ndsp->tone.tl);
+
+	if (dtmfthreshold < 20 || dtmfthreshold > 500)
+		dtmfthreshold = 200;
+	ndsp->dtmf.treshold = dtmfthreshold*10000;
+
+	/* init pipeline append to list */
+	spin_lock_irqsave(&dsp_lock, flags);
+	dsp_pipeline_init(&ndsp->pipeline);
+	list_add_tail(&ndsp->list, &dsp_ilist);
+	spin_unlock_irqrestore(&dsp_lock, flags);
+
+	return 0;
+}
+
+
+static struct Bprotocol DSP = {
+	.Bprotocols = (1 << (ISDN_P_B_L2DSP & ISDN_P_B_MASK))
+		| (1 << (ISDN_P_B_L2DSPHDLC & ISDN_P_B_MASK)),
+	.name = "dsp",
+	.create = dspcreate
+};
+
+static int dsp_init(void)
+{
+	int err;
+	int tics;
+
+	printk(KERN_INFO "DSP modul %s\n", mISDN_dsp_revision);
+
+	dsp_options = options;
+	dsp_debug = debug;
+
+	/* set packet size */
+	dsp_poll = poll;
+	if (dsp_poll) {
+		if (dsp_poll > MAX_POLL) {
+			printk(KERN_ERR "%s: Wrong poll value (%d), use %d "
+				"maximum.\n", __func__, poll, MAX_POLL);
+			err = -EINVAL;
+			return err;
+		}
+		if (dsp_poll < 8) {
+			printk(KERN_ERR "%s: Wrong poll value (%d), use 8 "
+				"minimum.\n", __func__, dsp_poll);
+			err = -EINVAL;
+			return err;
+		}
+		dsp_tics = poll * HZ / 8000;
+		if (dsp_tics * 8000 != poll * HZ) {
+			printk(KERN_INFO "mISDN_dsp: Cannot clock every %d "
+				"samples (0,125 ms). It is not a multiple of "
+				"%d HZ.\n", poll, HZ);
+			err = -EINVAL;
+			return err;
+		}
+	} else {
+		poll = 8;
+		while (poll <= MAX_POLL) {
+			tics = poll * HZ / 8000;
+			if (tics * 8000 == poll * HZ) {
+				dsp_tics = tics;
+				dsp_poll = poll;
+				if (poll >= 64)
+					break;
+			}
+			poll++;
+		}
+	}
+	if (dsp_poll == 0) {
+		printk(KERN_INFO "mISDN_dsp: There is no multiple of kernel "
+			"clock that equals exactly the duration of 8-256 "
+			"samples. (Choose kernel clock speed like 100, 250, "
+			"300, 1000)\n");
+		err = -EINVAL;
+		return err;
+	}
+	printk(KERN_INFO "mISDN_dsp: DSP clocks every %d samples. This equals "
+		"%d jiffies.\n", dsp_poll, dsp_tics);
+
+	spin_lock_init(&dsp_lock);
+	INIT_LIST_HEAD(&dsp_ilist);
+	INIT_LIST_HEAD(&conf_ilist);
+
+	/* init conversion tables */
+	dsp_audio_generate_law_tables();
+	dsp_silence = (dsp_options&DSP_OPT_ULAW)?0xff:0x2a;
+	dsp_audio_law_to_s32 = (dsp_options&DSP_OPT_ULAW)?dsp_audio_ulaw_to_s32:
+		dsp_audio_alaw_to_s32;
+	dsp_audio_generate_s2law_table();
+	dsp_audio_generate_seven();
+	dsp_audio_generate_mix_table();
+	if (dsp_options & DSP_OPT_ULAW)
+		dsp_audio_generate_ulaw_samples();
+	dsp_audio_generate_volume_changes();
+
+	err = dsp_pipeline_module_init();
+	if (err) {
+		printk(KERN_ERR "mISDN_dsp: Can't initialize pipeline, "
+			"error(%d)\n", err);
+		return err;
+	}
+
+	err = mISDN_register_Bprotocol(&DSP);
+	if (err) {
+		printk(KERN_ERR "Can't register %s error(%d)\n", DSP.name, err);
+		return err;
+	}
+
+	/* set sample timer */
+	dsp_spl_tl.function = (void *)dsp_cmx_send;
+	dsp_spl_tl.data = 0;
+	init_timer(&dsp_spl_tl);
+	dsp_spl_tl.expires = jiffies + dsp_tics;
+	dsp_spl_jiffies = dsp_spl_tl.expires;
+	add_timer(&dsp_spl_tl);
+
+	return 0;
+}
+
+
+static void dsp_cleanup(void)
+{
+	mISDN_unregister_Bprotocol(&DSP);
+
+	if (timer_pending(&dsp_spl_tl))
+		del_timer(&dsp_spl_tl);
+
+	if (!list_empty(&dsp_ilist)) {
+		printk(KERN_ERR "mISDN_dsp: Audio DSP object inst list not "
+			"empty.\n");
+	}
+	if (!list_empty(&conf_ilist)) {
+		printk(KERN_ERR "mISDN_dsp: Conference list not empty. Not "
+			"all memory freed.\n");
+	}
+
+	dsp_pipeline_module_exit();
+}
+
+module_init(dsp_init);
+module_exit(dsp_cleanup);
+
diff --git a/drivers/isdn/mISDN/dsp_dtmf.c b/drivers/isdn/mISDN/dsp_dtmf.c
new file mode 100644
index 000000000000..efc371c1f0dc
--- /dev/null
+++ b/drivers/isdn/mISDN/dsp_dtmf.c
@@ -0,0 +1,303 @@
+/*
+ * DTMF decoder.
+ *
+ * Copyright            by Andreas Eversberg (jolly@eversberg.eu)
+ *			based on different decoders such as ISDN4Linux
+ *
+ * This software may be used and distributed according to the terms
+ * of the GNU General Public License, incorporated herein by reference.
+ *
+ */
+
+#include <linux/mISDNif.h>
+#include <linux/mISDNdsp.h>
+#include "core.h"
+#include "dsp.h"
+
+#define NCOEFF            8     /* number of frequencies to be analyzed */
+
+/* For DTMF recognition:
+ * 2 * cos(2 * PI * k / N) precalculated for all k
+ */
+static u64 cos2pik[NCOEFF] =
+{
+	/* k << 15 (source: hfc-4s/8s documentation (www.colognechip.de)) */
+	55960, 53912, 51402, 48438, 38146, 32650, 26170, 18630
+};
+
+/* digit matrix */
+static char dtmf_matrix[4][4] =
+{
+	{'1', '2', '3', 'A'},
+	{'4', '5', '6', 'B'},
+	{'7', '8', '9', 'C'},
+	{'*', '0', '#', 'D'}
+};
+
+/* dtmf detection using goertzel algorithm
+ * init function
+ */
+void dsp_dtmf_goertzel_init(struct dsp *dsp)
+{
+	dsp->dtmf.size = 0;
+	dsp->dtmf.lastwhat = '\0';
+	dsp->dtmf.lastdigit = '\0';
+	dsp->dtmf.count = 0;
+}
+
+/* check for hardware or software features
+ */
+void dsp_dtmf_hardware(struct dsp *dsp)
+{
+	int hardware = 1;
+
+	if (!dsp->features.hfc_dtmf)
+		hardware = 0;
+
+	/* check for volume change */
+	if (dsp->tx_volume) {
+		if (dsp_debug & DEBUG_DSP_DTMF)
+			printk(KERN_DEBUG "%s dsp %s cannot do hardware DTMF, "
+				"because tx_volume is changed\n",
+				__func__, dsp->name);
+		hardware = 0;
+	}
+	if (dsp->rx_volume) {
+		if (dsp_debug & DEBUG_DSP_DTMF)
+			printk(KERN_DEBUG "%s dsp %s cannot do hardware DTMF, "
+				"because rx_volume is changed\n",
+				__func__, dsp->name);
+		hardware = 0;
+	}
+	/* check if encryption is enabled */
+	if (dsp->bf_enable) {
+		if (dsp_debug & DEBUG_DSP_DTMF)
+			printk(KERN_DEBUG "%s dsp %s cannot do hardware DTMF, "
+				"because encryption is enabled\n",
+				__func__, dsp->name);
+		hardware = 0;
+	}
+	/* check if pipeline exists */
+	if (dsp->pipeline.inuse) {
+		if (dsp_debug & DEBUG_DSP_DTMF)
+			printk(KERN_DEBUG "%s dsp %s cannot do hardware DTMF, "
+				"because pipeline exists.\n",
+				__func__, dsp->name);
+		hardware = 0;
+	}
+
+	dsp->dtmf.hardware = hardware;
+	dsp->dtmf.software = !hardware;
+}
+
+
+/*************************************************************
+ * calculate the coefficients of the given sample and decode *
+ *************************************************************/
+
+/* the given sample is decoded. if the sample is not long enough for a
+ * complete frame, the decoding is finished and continued with the next
+ * call of this function.
+ *
+ * the algorithm is very good for detection with a minimum of errors. i
+ * tested it allot. it even works with very short tones (40ms). the only
+ * disadvantage is, that it doesn't work good with different volumes of both
+ * tones. this will happen, if accoustically coupled dialers are used.
+ * it sometimes detects tones during speach, which is normal for decoders.
+ * use sequences to given commands during calls.
+ *
+ * dtmf - points to a structure of the current dtmf state
+ * spl and len - the sample
+ * fmt - 0 = alaw, 1 = ulaw, 2 = coefficients from HFC DTMF hw-decoder
+ */
+
+u8
+*dsp_dtmf_goertzel_decode(struct dsp *dsp, u8 *data, int len, int fmt)
+{
+	u8 what;
+	int size;
+	signed short *buf;
+	s32 sk, sk1, sk2;
+	int k, n, i;
+	s32 *hfccoeff;
+	s32 result[NCOEFF], tresh, treshl;
+	int lowgroup, highgroup;
+	s64 cos2pik_;
+
+	dsp->dtmf.digits[0] = '\0';
+
+	/* Note: The function will loop until the buffer has not enough samples
+	 * left to decode a full frame.
+	 */
+again:
+	/* convert samples */
+	size = dsp->dtmf.size;
+	buf = dsp->dtmf.buffer;
+	switch (fmt) {
+	case 0: /* alaw */
+	case 1: /* ulaw */
+		while (size < DSP_DTMF_NPOINTS && len) {
+			buf[size++] = dsp_audio_law_to_s32[*data++];
+			len--;
+		}
+		break;
+
+	case 2: /* HFC coefficients */
+	default:
+		if (len < 64) {
+			if (len > 0)
+				printk(KERN_ERR "%s: coefficients have invalid "
+					"size. (is=%d < must=%d)\n",
+					__func__, len, 64);
+			return dsp->dtmf.digits;
+		}
+		hfccoeff = (s32 *)data;
+		for (k = 0; k < NCOEFF; k++) {
+			sk2 = (*hfccoeff++)>>4;
+			sk = (*hfccoeff++)>>4;
+			if (sk > 32767 || sk < -32767 || sk2 > 32767
+			    || sk2 < -32767)
+				printk(KERN_WARNING
+					"DTMF-Detection overflow\n");
+			/* compute |X(k)|**2 */
+			result[k] =
+				 (sk * sk) -
+				 (((cos2pik[k] * sk) >> 15) * sk2) +
+				 (sk2 * sk2);
+		}
+		data += 64;
+		len -= 64;
+		goto coefficients;
+		break;
+	}
+	dsp->dtmf.size = size;
+
+	if (size < DSP_DTMF_NPOINTS)
+		return dsp->dtmf.digits;
+
+	dsp->dtmf.size = 0;
+
+	/* now we have a full buffer of signed long samples - we do goertzel */
+	for (k = 0; k < NCOEFF; k++) {
+		sk = 0;
+		sk1 = 0;
+		sk2 = 0;
+		buf = dsp->dtmf.buffer;
+		cos2pik_ = cos2pik[k];
+		for (n = 0; n < DSP_DTMF_NPOINTS; n++) {
+			sk = ((cos2pik_*sk1)>>15) - sk2 + (*buf++);
+			sk2 = sk1;
+			sk1 = sk;
+		}
+		sk >>= 8;
+		sk2 >>= 8;
+		if (sk > 32767 || sk < -32767 || sk2 > 32767 || sk2 < -32767)
+			printk(KERN_WARNING "DTMF-Detection overflow\n");
+		/* compute |X(k)|**2 */
+		result[k] =
+			(sk * sk) -
+			(((cos2pik[k] * sk) >> 15) * sk2) +
+			(sk2 * sk2);
+	}
+
+	/* our (squared) coefficients have been calculated, we need to process
+	 * them.
+	 */
+coefficients:
+	tresh = 0;
+	for (i = 0; i < NCOEFF; i++) {
+		if (result[i] < 0)
+			result[i] = 0;
+		if (result[i] > dsp->dtmf.treshold) {
+			if (result[i] > tresh)
+				tresh = result[i];
+		}
+	}
+
+	if (tresh == 0) {
+		what = 0;
+		goto storedigit;
+	}
+
+	if (dsp_debug & DEBUG_DSP_DTMFCOEFF)
+		printk(KERN_DEBUG "a %3d %3d %3d %3d %3d %3d %3d %3d"
+			" tr:%3d r %3d %3d %3d %3d %3d %3d %3d %3d\n",
+			result[0]/10000, result[1]/10000, result[2]/10000,
+			result[3]/10000, result[4]/10000, result[5]/10000,
+			result[6]/10000, result[7]/10000, tresh/10000,
+			result[0]/(tresh/100), result[1]/(tresh/100),
+			result[2]/(tresh/100), result[3]/(tresh/100),
+			result[4]/(tresh/100), result[5]/(tresh/100),
+			result[6]/(tresh/100), result[7]/(tresh/100));
+
+	/* calc digit (lowgroup/highgroup) */
+	lowgroup = -1;
+	highgroup = -1;
+	treshl = tresh >> 3;  /* tones which are not on, must be below 9 dB */
+	tresh = tresh >> 2;  /* touchtones must match within 6 dB */
+	for (i = 0; i < NCOEFF; i++) {
+		if (result[i] < treshl)
+			continue;  /* ignore */
+		if (result[i] < tresh) {
+			lowgroup = -1;
+			highgroup = -1;
+			break;  /* noise inbetween */
+		}
+		/* good level found. This is allowed only one time per group */
+		if (i < NCOEFF/2) {
+			/* lowgroup */
+			if (lowgroup >= 0) {
+				/* Bad. Another tone found. */
+				lowgroup = -1;
+				break;
+			} else
+				lowgroup = i;
+		} else {
+			/* higroup */
+			if (highgroup >= 0) {
+				/* Bad. Another tone found. */
+				highgroup = -1;
+				break;
+			} else
+				highgroup = i-(NCOEFF/2);
+		}
+	}
+
+	/* get digit or null */
+	what = 0;
+	if (lowgroup >= 0 && highgroup >= 0)
+		what = dtmf_matrix[lowgroup][highgroup];
+
+storedigit:
+	if (what && (dsp_debug & DEBUG_DSP_DTMF))
+		printk(KERN_DEBUG "DTMF what: %c\n", what);
+
+	if (dsp->dtmf.lastwhat != what)
+		dsp->dtmf.count = 0;
+
+	/* the tone (or no tone) must remain 3 times without change */
+	if (dsp->dtmf.count == 2) {
+		if (dsp->dtmf.lastdigit != what) {
+			dsp->dtmf.lastdigit = what;
+			if (what) {
+				if (dsp_debug & DEBUG_DSP_DTMF)
+					printk(KERN_DEBUG "DTMF digit: %c\n",
+						what);
+				if ((strlen(dsp->dtmf.digits)+1)
+					< sizeof(dsp->dtmf.digits)) {
+					dsp->dtmf.digits[strlen(
+						dsp->dtmf.digits)+1] = '\0';
+					dsp->dtmf.digits[strlen(
+						dsp->dtmf.digits)] = what;
+				}
+			}
+		}
+	} else
+		dsp->dtmf.count++;
+
+	dsp->dtmf.lastwhat = what;
+
+	goto again;
+}
+
+
diff --git a/drivers/isdn/mISDN/dsp_ecdis.h b/drivers/isdn/mISDN/dsp_ecdis.h
new file mode 100644
index 000000000000..8a20af43308b
--- /dev/null
+++ b/drivers/isdn/mISDN/dsp_ecdis.h
@@ -0,0 +1,110 @@
+/*
+ * SpanDSP - a series of DSP components for telephony
+ *
+ * ec_disable_detector.h - A detector which should eventually meet the
+ *                         G.164/G.165 requirements for detecting the
+ *                         2100Hz echo cancellor disable tone.
+ *
+ * Written by Steve Underwood <steveu@coppice.org>
+ *
+ * Copyright (C) 2001 Steve Underwood
+ *
+ * All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ *
+ */
+
+#include "dsp_biquad.h"
+
+struct ec_disable_detector_state {
+	struct biquad2_state notch;
+	int notch_level;
+	int channel_level;
+	int tone_present;
+	int tone_cycle_duration;
+	int good_cycles;
+	int hit;
+};
+
+
+#define FALSE 0
+#define TRUE (!FALSE)
+
+static inline void
+echo_can_disable_detector_init(struct ec_disable_detector_state *det)
+{
+    /* Elliptic notch */
+    /* This is actually centred at 2095Hz, but gets the balance we want, due
+       to the asymmetric walls of the notch */
+	biquad2_init(&det->notch,
+		(int32_t) (-0.7600000*32768.0),
+		(int32_t) (-0.1183852*32768.0),
+		(int32_t) (-0.5104039*32768.0),
+		(int32_t) (0.1567596*32768.0),
+		(int32_t) (1.0000000*32768.0));
+
+	det->channel_level = 0;
+	det->notch_level = 0;
+	det->tone_present = FALSE;
+	det->tone_cycle_duration = 0;
+	det->good_cycles = 0;
+	det->hit = 0;
+}
+/*- End of function --------------------------------------------------------*/
+
+static inline int
+echo_can_disable_detector_update(struct ec_disable_detector_state *det,
+int16_t amp)
+{
+	int16_t notched;
+
+	notched = biquad2(&det->notch, amp);
+	/* Estimate the overall energy in the channel, and the energy in
+	   the notch (i.e. overall channel energy - tone energy => noise).
+	   Use abs instead of multiply for speed (is it really faster?).
+	   Damp the overall energy a little more for a stable result.
+	   Damp the notch energy a little less, so we don't damp out the
+	   blip every time the phase reverses */
+	det->channel_level += ((abs(amp) - det->channel_level) >> 5);
+	det->notch_level += ((abs(notched) - det->notch_level) >> 4);
+	if (det->channel_level > 280) {
+		/* There is adequate energy in the channel.
+		 Is it mostly at 2100Hz? */
+		if (det->notch_level*6 < det->channel_level) {
+			/* The notch says yes, so we have the tone. */
+			if (!det->tone_present) {
+				/* Do we get a kick every 450+-25ms? */
+				if (det->tone_cycle_duration >= 425*8
+					&& det->tone_cycle_duration <= 475*8) {
+					det->good_cycles++;
+					if (det->good_cycles > 2)
+					det->hit = TRUE;
+				}
+				det->tone_cycle_duration = 0;
+			}
+			det->tone_present = TRUE;
+		} else
+			det->tone_present = FALSE;
+		det->tone_cycle_duration++;
+	} else {
+		det->tone_present = FALSE;
+		det->tone_cycle_duration = 0;
+		det->good_cycles = 0;
+	}
+	return det->hit;
+}
+/*- End of function --------------------------------------------------------*/
+/*- End of file ------------------------------------------------------------*/
diff --git a/drivers/isdn/mISDN/dsp_hwec.c b/drivers/isdn/mISDN/dsp_hwec.c
new file mode 100644
index 000000000000..eb892d9dd5c6
--- /dev/null
+++ b/drivers/isdn/mISDN/dsp_hwec.c
@@ -0,0 +1,138 @@
+/*
+ * dsp_hwec.c:
+ * builtin mISDN dsp pipeline element for enabling the hw echocanceller
+ *
+ * Copyright (C) 2007, Nadi Sarrar
+ *
+ * Nadi Sarrar <nadi@beronet.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc., 59
+ * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
+ *
+ * The full GNU General Public License is included in this distribution in the
+ * file called LICENSE.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/string.h>
+#include <linux/mISDNdsp.h>
+#include <linux/mISDNif.h>
+#include "core.h"
+#include "dsp.h"
+#include "dsp_hwec.h"
+
+static struct mISDN_dsp_element_arg args[] = {
+	{ "deftaps", "128", "Set the number of taps of cancellation." },
+};
+
+static struct mISDN_dsp_element dsp_hwec_p = {
+	.name = "hwec",
+	.new = NULL,
+	.free = NULL,
+	.process_tx = NULL,
+	.process_rx = NULL,
+	.num_args = sizeof(args) / sizeof(struct mISDN_dsp_element_arg),
+	.args = args,
+};
+struct mISDN_dsp_element *dsp_hwec = &dsp_hwec_p;
+
+void dsp_hwec_enable(struct dsp *dsp, const char *arg)
+{
+	int deftaps = 128,
+		len;
+	struct mISDN_ctrl_req	cq;
+
+	if (!dsp) {
+		printk(KERN_ERR "%s: failed to enable hwec: dsp is NULL\n",
+			__func__);
+		return;
+	}
+
+	if (!arg)
+		goto _do;
+
+	len = strlen(arg);
+	if (!len)
+		goto _do;
+
+	{
+		char _dup[len + 1];
+		char *dup, *tok, *name, *val;
+		int tmp;
+
+		strcpy(_dup, arg);
+		dup = _dup;
+
+		while ((tok = strsep(&dup, ","))) {
+			if (!strlen(tok))
+				continue;
+			name = strsep(&tok, "=");
+			val = tok;
+
+			if (!val)
+				continue;
+
+			if (!strcmp(name, "deftaps")) {
+				if (sscanf(val, "%d", &tmp) == 1)
+					deftaps = tmp;
+			}
+		}
+	}
+
+_do:
+	printk(KERN_DEBUG "%s: enabling hwec with deftaps=%d\n",
+		__func__, deftaps);
+	memset(&cq, 0, sizeof(cq));
+	cq.op = MISDN_CTRL_HFC_ECHOCAN_ON;
+	cq.p1 = deftaps;
+	if (!dsp->ch.peer->ctrl(&dsp->ch, CONTROL_CHANNEL, &cq)) {
+		printk(KERN_DEBUG "%s: CONTROL_CHANNEL failed\n",
+			__func__);
+		return;
+	}
+}
+
+void dsp_hwec_disable(struct dsp *dsp)
+{
+	struct mISDN_ctrl_req	cq;
+
+	if (!dsp) {
+		printk(KERN_ERR "%s: failed to disable hwec: dsp is NULL\n",
+			__func__);
+		return;
+	}
+
+	printk(KERN_DEBUG "%s: disabling hwec\n", __func__);
+	memset(&cq, 0, sizeof(cq));
+	cq.op = MISDN_CTRL_HFC_ECHOCAN_OFF;
+	if (!dsp->ch.peer->ctrl(&dsp->ch, CONTROL_CHANNEL, &cq)) {
+		printk(KERN_DEBUG "%s: CONTROL_CHANNEL failed\n",
+			__func__);
+		return;
+	}
+}
+
+int dsp_hwec_init(void)
+{
+	mISDN_dsp_element_register(dsp_hwec);
+
+	return 0;
+}
+
+void dsp_hwec_exit(void)
+{
+	mISDN_dsp_element_unregister(dsp_hwec);
+}
+
diff --git a/drivers/isdn/mISDN/dsp_hwec.h b/drivers/isdn/mISDN/dsp_hwec.h
new file mode 100644
index 000000000000..eebe80c3f713
--- /dev/null
+++ b/drivers/isdn/mISDN/dsp_hwec.h
@@ -0,0 +1,10 @@
+/*
+ * dsp_hwec.h
+ */
+
+extern struct mISDN_dsp_element *dsp_hwec;
+extern void dsp_hwec_enable(struct dsp *dsp, const char *arg);
+extern void dsp_hwec_disable(struct dsp *dsp);
+extern int  dsp_hwec_init(void);
+extern void dsp_hwec_exit(void);
+
diff --git a/drivers/isdn/mISDN/dsp_pipeline.c b/drivers/isdn/mISDN/dsp_pipeline.c
new file mode 100644
index 000000000000..850260ab57d0
--- /dev/null
+++ b/drivers/isdn/mISDN/dsp_pipeline.c
@@ -0,0 +1,348 @@
+/*
+ * dsp_pipeline.c: pipelined audio processing
+ *
+ * Copyright (C) 2007, Nadi Sarrar
+ *
+ * Nadi Sarrar <nadi@beronet.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc., 59
+ * Temple Place - Suite 330, Boston, MA  02111-1307, USA.
+ *
+ * The full GNU General Public License is included in this distribution in the
+ * file called LICENSE.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/list.h>
+#include <linux/string.h>
+#include <linux/mISDNif.h>
+#include <linux/mISDNdsp.h>
+#include "dsp.h"
+#include "dsp_hwec.h"
+
+/* uncomment for debugging */
+/*#define PIPELINE_DEBUG*/
+
+struct dsp_pipeline_entry {
+	struct mISDN_dsp_element *elem;
+	void                *p;
+	struct list_head     list;
+};
+struct dsp_element_entry {
+	struct mISDN_dsp_element *elem;
+	struct device	     dev;
+	struct list_head     list;
+};
+
+static LIST_HEAD(dsp_elements);
+
+/* sysfs */
+static struct class *elements_class;
+
+static ssize_t
+attr_show_args(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct mISDN_dsp_element *elem = dev_get_drvdata(dev);
+	ssize_t len = 0;
+	int i = 0;
+
+	*buf = 0;
+	for (; i < elem->num_args; ++i)
+		len = sprintf(buf, "%sName:        %s\n%s%s%sDescription: %s\n"
+			"\n", buf,
+			  elem->args[i].name,
+			  elem->args[i].def ? "Default:     " : "",
+			  elem->args[i].def ? elem->args[i].def : "",
+			  elem->args[i].def ? "\n" : "",
+			  elem->args[i].desc);
+
+	return len;
+}
+
+static struct device_attribute element_attributes[] = {
+	__ATTR(args, 0444, attr_show_args, NULL),
+};
+
+int mISDN_dsp_element_register(struct mISDN_dsp_element *elem)
+{
+	struct dsp_element_entry *entry;
+	int ret, i;
+
+	if (!elem)
+		return -EINVAL;
+
+	entry = kzalloc(sizeof(struct dsp_element_entry), GFP_KERNEL);
+	if (!entry)
+		return -ENOMEM;
+
+	entry->elem = elem;
+
+	entry->dev.class = elements_class;
+	dev_set_drvdata(&entry->dev, elem);
+	snprintf(entry->dev.bus_id, BUS_ID_SIZE, elem->name);
+	ret = device_register(&entry->dev);
+	if (ret) {
+		printk(KERN_ERR "%s: failed to register %s\n",
+			__func__, elem->name);
+		goto err1;
+	}
+
+	for (i = 0; i < (sizeof(element_attributes)
+		/ sizeof(struct device_attribute)); ++i)
+		ret = device_create_file(&entry->dev,
+				&element_attributes[i]);
+		if (ret) {
+			printk(KERN_ERR "%s: failed to create device file\n",
+				__func__);
+			goto err2;
+		}
+
+	list_add_tail(&entry->list, &dsp_elements);
+
+	printk(KERN_DEBUG "%s: %s registered\n", __func__, elem->name);
+
+	return 0;
+
+err2:
+	device_unregister(&entry->dev);
+err1:
+	kfree(entry);
+	return ret;
+}
+EXPORT_SYMBOL(mISDN_dsp_element_register);
+
+void mISDN_dsp_element_unregister(struct mISDN_dsp_element *elem)
+{
+	struct dsp_element_entry *entry, *n;
+
+	if (!elem)
+		return;
+
+	list_for_each_entry_safe(entry, n, &dsp_elements, list)
+		if (entry->elem == elem) {
+			list_del(&entry->list);
+			device_unregister(&entry->dev);
+			kfree(entry);
+			printk(KERN_DEBUG "%s: %s unregistered\n",
+				__func__, elem->name);
+			return;
+		}
+	printk(KERN_ERR "%s: element %s not in list.\n", __func__, elem->name);
+}
+EXPORT_SYMBOL(mISDN_dsp_element_unregister);
+
+int dsp_pipeline_module_init(void)
+{
+	elements_class = class_create(THIS_MODULE, "dsp_pipeline");
+	if (IS_ERR(elements_class))
+		return PTR_ERR(elements_class);
+
+#ifdef PIPELINE_DEBUG
+	printk(KERN_DEBUG "%s: dsp pipeline module initialized\n", __func__);
+#endif
+
+	dsp_hwec_init();
+
+	return 0;
+}
+
+void dsp_pipeline_module_exit(void)
+{
+	struct dsp_element_entry *entry, *n;
+
+	dsp_hwec_exit();
+
+	class_destroy(elements_class);
+
+	list_for_each_entry_safe(entry, n, &dsp_elements, list) {
+		list_del(&entry->list);
+		printk(KERN_WARNING "%s: element was still registered: %s\n",
+			__func__, entry->elem->name);
+		kfree(entry);
+	}
+
+	printk(KERN_DEBUG "%s: dsp pipeline module exited\n", __func__);
+}
+
+int dsp_pipeline_init(struct dsp_pipeline *pipeline)
+{
+	if (!pipeline)
+		return -EINVAL;
+
+	INIT_LIST_HEAD(&pipeline->list);
+
+#ifdef PIPELINE_DEBUG
+	printk(KERN_DEBUG "%s: dsp pipeline ready\n", __func__);
+#endif
+
+	return 0;
+}
+
+static inline void _dsp_pipeline_destroy(struct dsp_pipeline *pipeline)
+{
+	struct dsp_pipeline_entry *entry, *n;
+
+	list_for_each_entry_safe(entry, n, &pipeline->list, list) {
+		list_del(&entry->list);
+		if (entry->elem == dsp_hwec)
+			dsp_hwec_disable(container_of(pipeline, struct dsp,
+				pipeline));
+		else
+			entry->elem->free(entry->p);
+		kfree(entry);
+	}
+}
+
+void dsp_pipeline_destroy(struct dsp_pipeline *pipeline)
+{
+
+	if (!pipeline)
+		return;
+
+	_dsp_pipeline_destroy(pipeline);
+
+#ifdef PIPELINE_DEBUG
+	printk(KERN_DEBUG "%s: dsp pipeline destroyed\n", __func__);
+#endif
+}
+
+int dsp_pipeline_build(struct dsp_pipeline *pipeline, const char *cfg)
+{
+	int len, incomplete = 0, found = 0;
+	char *dup, *tok, *name, *args;
+	struct dsp_element_entry *entry, *n;
+	struct dsp_pipeline_entry *pipeline_entry;
+	struct mISDN_dsp_element *elem;
+
+	if (!pipeline)
+		return -EINVAL;
+
+	if (!list_empty(&pipeline->list))
+		_dsp_pipeline_destroy(pipeline);
+
+	if (!cfg)
+		return 0;
+
+	len = strlen(cfg);
+	if (!len)
+		return 0;
+
+	dup = kmalloc(len + 1, GFP_KERNEL);
+	if (!dup)
+		return 0;
+	strcpy(dup, cfg);
+	while ((tok = strsep(&dup, "|"))) {
+		if (!strlen(tok))
+			continue;
+		name = strsep(&tok, "(");
+		args = strsep(&tok, ")");
+		if (args && !*args)
+			args = 0;
+
+		list_for_each_entry_safe(entry, n, &dsp_elements, list)
+			if (!strcmp(entry->elem->name, name)) {
+				elem = entry->elem;
+
+				pipeline_entry = kmalloc(sizeof(struct
+					dsp_pipeline_entry), GFP_KERNEL);
+				if (!pipeline_entry) {
+					printk(KERN_DEBUG "%s: failed to add "
+					    "entry to pipeline: %s (out of "
+					    "memory)\n", __func__, elem->name);
+					incomplete = 1;
+					goto _out;
+				}
+				pipeline_entry->elem = elem;
+
+				if (elem == dsp_hwec) {
+					/* This is a hack to make the hwec
+					   available as a pipeline module */
+					dsp_hwec_enable(container_of(pipeline,
+						struct dsp, pipeline), args);
+					list_add_tail(&pipeline_entry->list,
+						&pipeline->list);
+				} else {
+					pipeline_entry->p = elem->new(args);
+					if (pipeline_entry->p) {
+						list_add_tail(&pipeline_entry->
+							list, &pipeline->list);
+#ifdef PIPELINE_DEBUG
+						printk(KERN_DEBUG "%s: created "
+						    "instance of %s%s%s\n",
+						    __func__, name, args ?
+						    " with args " : "", args ?
+						    args : "");
+#endif
+					} else {
+						printk(KERN_DEBUG "%s: failed "
+						  "to add entry to pipeline: "
+						  "%s (new() returned NULL)\n",
+						  __func__, elem->name);
+						kfree(pipeline_entry);
+						incomplete = 1;
+					}
+				}
+				found = 1;
+				break;
+			}
+
+		if (found)
+			found = 0;
+		else {
+			printk(KERN_DEBUG "%s: element not found, skipping: "
+				"%s\n", __func__, name);
+			incomplete = 1;
+		}
+	}
+
+_out:
+	if (!list_empty(&pipeline->list))
+		pipeline->inuse = 1;
+	else
+		pipeline->inuse = 0;
+
+#ifdef PIPELINE_DEBUG
+	printk(KERN_DEBUG "%s: dsp pipeline built%s: %s\n",
+		__func__, incomplete ? " incomplete" : "", cfg);
+#endif
+	kfree(dup);
+	return 0;
+}
+
+void dsp_pipeline_process_tx(struct dsp_pipeline *pipeline, u8 *data, int len)
+{
+	struct dsp_pipeline_entry *entry;
+
+	if (!pipeline)
+		return;
+
+	list_for_each_entry(entry, &pipeline->list, list)
+		if (entry->elem->process_tx)
+			entry->elem->process_tx(entry->p, data, len);
+}
+
+void dsp_pipeline_process_rx(struct dsp_pipeline *pipeline, u8 *data, int len)
+{
+	struct dsp_pipeline_entry *entry;
+
+	if (!pipeline)
+		return;
+
+	list_for_each_entry_reverse(entry, &pipeline->list, list)
+		if (entry->elem->process_rx)
+			entry->elem->process_rx(entry->p, data, len);
+}
+
+
diff --git a/drivers/isdn/mISDN/dsp_tones.c b/drivers/isdn/mISDN/dsp_tones.c
new file mode 100644
index 000000000000..23dd0dd21524
--- /dev/null
+++ b/drivers/isdn/mISDN/dsp_tones.c
@@ -0,0 +1,551 @@
+/*
+ * Audio support data for ISDN4Linux.
+ *
+ * Copyright Andreas Eversberg (jolly@eversberg.eu)
+ *
+ * This software may be used and distributed according to the terms
+ * of the GNU General Public License, incorporated herein by reference.
+ *
+ */
+
+#include <linux/mISDNif.h>
+#include <linux/mISDNdsp.h>
+#include "core.h"
+#include "dsp.h"
+
+
+#define DATA_S sample_silence
+#define SIZE_S (&sizeof_silence)
+#define DATA_GA sample_german_all
+#define SIZE_GA (&sizeof_german_all)
+#define DATA_GO sample_german_old
+#define SIZE_GO (&sizeof_german_old)
+#define DATA_DT sample_american_dialtone
+#define SIZE_DT (&sizeof_american_dialtone)
+#define DATA_RI sample_american_ringing
+#define SIZE_RI (&sizeof_american_ringing)
+#define DATA_BU sample_american_busy
+#define SIZE_BU (&sizeof_american_busy)
+#define DATA_S1 sample_special1
+#define SIZE_S1 (&sizeof_special1)
+#define DATA_S2 sample_special2
+#define SIZE_S2 (&sizeof_special2)
+#define DATA_S3 sample_special3
+#define SIZE_S3 (&sizeof_special3)
+
+/***************/
+/* tones loops */
+/***************/
+
+/* all tones are alaw encoded */
+/* the last sample+1 is in phase with the first sample. the error is low */
+
+static u8 sample_german_all[] = {
+	0x80, 0xab, 0x81, 0x6d, 0xfd, 0xdd, 0x5d, 0x9d,
+	0x4d, 0xd1, 0x89, 0x88, 0xd0, 0x4c, 0x9c, 0x5c,
+	0xdc, 0xfc, 0x6c,
+	0x80, 0xab, 0x81, 0x6d, 0xfd, 0xdd, 0x5d, 0x9d,
+	0x4d, 0xd1, 0x89, 0x88, 0xd0, 0x4c, 0x9c, 0x5c,
+	0xdc, 0xfc, 0x6c,
+	0x80, 0xab, 0x81, 0x6d, 0xfd, 0xdd, 0x5d, 0x9d,
+	0x4d, 0xd1, 0x89, 0x88, 0xd0, 0x4c, 0x9c, 0x5c,
+	0xdc, 0xfc, 0x6c,
+	0x80, 0xab, 0x81, 0x6d, 0xfd, 0xdd, 0x5d, 0x9d,
+	0x4d, 0xd1, 0x89, 0x88, 0xd0, 0x4c, 0x9c, 0x5c,
+	0xdc, 0xfc, 0x6c,
+};
+static u32 sizeof_german_all = sizeof(sample_german_all);
+
+static u8 sample_german_old[] = {
+	0xec, 0x68, 0xe1, 0x6d, 0x6d, 0x91, 0x51, 0xed,
+	0x6d, 0x01, 0x1e, 0x10, 0x0c, 0x90, 0x60, 0x70,
+	0x8c,
+	0xec, 0x68, 0xe1, 0x6d, 0x6d, 0x91, 0x51, 0xed,
+	0x6d, 0x01, 0x1e, 0x10, 0x0c, 0x90, 0x60, 0x70,
+	0x8c,
+	0xec, 0x68, 0xe1, 0x6d, 0x6d, 0x91, 0x51, 0xed,
+	0x6d, 0x01, 0x1e, 0x10, 0x0c, 0x90, 0x60, 0x70,
+	0x8c,
+	0xec, 0x68, 0xe1, 0x6d, 0x6d, 0x91, 0x51, 0xed,
+	0x6d, 0x01, 0x1e, 0x10, 0x0c, 0x90, 0x60, 0x70,
+	0x8c,
+};
+static u32 sizeof_german_old = sizeof(sample_german_old);
+
+static u8 sample_american_dialtone[] = {
+	0x2a, 0x18, 0x90, 0x6c, 0x4c, 0xbc, 0x4c, 0x6c,
+	0x10, 0x58, 0x32, 0xb9, 0x31, 0x2d, 0x8d, 0x0d,
+	0x8d, 0x2d, 0x31, 0x99, 0x0f, 0x28, 0x60, 0xf0,
+	0xd0, 0x50, 0xd0, 0x30, 0x60, 0x08, 0x8e, 0x67,
+	0x09, 0x19, 0x21, 0xe1, 0xd9, 0xb9, 0x29, 0x67,
+	0x83, 0x02, 0xce, 0xbe, 0xee, 0x1a, 0x1b, 0xef,
+	0xbf, 0xcf, 0x03, 0x82, 0x66, 0x28, 0xb8, 0xd8,
+	0xe0, 0x20, 0x18, 0x08, 0x66, 0x8f, 0x09, 0x61,
+	0x31, 0xd1, 0x51, 0xd1, 0xf1, 0x61, 0x29, 0x0e,
+	0x98, 0x30, 0x2c, 0x8c, 0x0c, 0x8c, 0x2c, 0x30,
+	0xb8, 0x33, 0x59, 0x11, 0x6d, 0x4d, 0xbd, 0x4d,
+	0x6d, 0x91, 0x19,
+};
+static u32 sizeof_american_dialtone = sizeof(sample_american_dialtone);
+
+static u8 sample_american_ringing[] = {
+	0x2a, 0xe0, 0xac, 0x0c, 0xbc, 0x4c, 0x8c, 0x90,
+	0x48, 0xc7, 0xc1, 0xed, 0xcd, 0x4d, 0xcd, 0xed,
+	0xc1, 0xb7, 0x08, 0x30, 0xec, 0xcc, 0xcc, 0x8c,
+	0x10, 0x58, 0x1a, 0x99, 0x71, 0xed, 0x8d, 0x8d,
+	0x2d, 0x41, 0x89, 0x9e, 0x20, 0x70, 0x2c, 0xec,
+	0x2c, 0x70, 0x20, 0x86, 0x77, 0xe1, 0x31, 0x11,
+	0xd1, 0xf1, 0x81, 0x09, 0xa3, 0x56, 0x58, 0x00,
+	0x40, 0xc0, 0x60, 0x38, 0x46, 0x43, 0x57, 0x39,
+	0xd9, 0x59, 0x99, 0xc9, 0x77, 0x2f, 0x2e, 0xc6,
+	0xd6, 0x28, 0xd6, 0x36, 0x26, 0x2e, 0x8a, 0xa3,
+	0x43, 0x63, 0x4b, 0x4a, 0x62, 0x42, 0xa2, 0x8b,
+	0x2f, 0x27, 0x37, 0xd7, 0x29, 0xd7, 0xc7, 0x2f,
+	0x2e, 0x76, 0xc8, 0x98, 0x58, 0xd8, 0x38, 0x56,
+	0x42, 0x47, 0x39, 0x61, 0xc1, 0x41, 0x01, 0x59,
+	0x57, 0xa2, 0x08, 0x80, 0xf0, 0xd0, 0x10, 0x30,
+	0xe0, 0x76, 0x87, 0x21, 0x71, 0x2d, 0xed, 0x2d,
+	0x71, 0x21, 0x9f, 0x88, 0x40, 0x2c, 0x8c, 0x8c,
+	0xec, 0x70, 0x98, 0x1b, 0x59, 0x11, 0x8d, 0xcd,
+	0xcd, 0xed, 0x31, 0x09, 0xb6, 0xc0, 0xec, 0xcc,
+	0x4c, 0xcc, 0xec, 0xc0, 0xc6, 0x49, 0x91, 0x8d,
+	0x4d, 0xbd, 0x0d, 0xad, 0xe1,
+};
+static u32 sizeof_american_ringing = sizeof(sample_american_ringing);
+
+static u8 sample_american_busy[] = {
+	0x2a, 0x00, 0x6c, 0x4c, 0x4c, 0x6c, 0xb0, 0x66,
+	0x99, 0x11, 0x6d, 0x8d, 0x2d, 0x41, 0xd7, 0x96,
+	0x60, 0xf0, 0x70, 0x40, 0x58, 0xf6, 0x53, 0x57,
+	0x09, 0x89, 0xd7, 0x5f, 0xe3, 0x2a, 0xe3, 0x5f,
+	0xd7, 0x89, 0x09, 0x57, 0x53, 0xf6, 0x58, 0x40,
+	0x70, 0xf0, 0x60, 0x96, 0xd7, 0x41, 0x2d, 0x8d,
+	0x6d, 0x11, 0x99, 0x66, 0xb0, 0x6c, 0x4c, 0x4c,
+	0x6c, 0x00, 0x2a, 0x01, 0x6d, 0x4d, 0x4d, 0x6d,
+	0xb1, 0x67, 0x98, 0x10, 0x6c, 0x8c, 0x2c, 0x40,
+	0xd6, 0x97, 0x61, 0xf1, 0x71, 0x41, 0x59, 0xf7,
+	0x52, 0x56, 0x08, 0x88, 0xd6, 0x5e, 0xe2, 0x2a,
+	0xe2, 0x5e, 0xd6, 0x88, 0x08, 0x56, 0x52, 0xf7,
+	0x59, 0x41, 0x71, 0xf1, 0x61, 0x97, 0xd6, 0x40,
+	0x2c, 0x8c, 0x6c, 0x10, 0x98, 0x67, 0xb1, 0x6d,
+	0x4d, 0x4d, 0x6d, 0x01,
+};
+static u32 sizeof_american_busy = sizeof(sample_american_busy);
+
+static u8 sample_special1[] = {
+	0x2a, 0x2c, 0xbc, 0x6c, 0xd6, 0x71, 0xbd, 0x0d,
+	0xd9, 0x80, 0xcc, 0x4c, 0x40, 0x39, 0x0d, 0xbd,
+	0x11, 0x86, 0xec, 0xbc, 0xec, 0x0e, 0x51, 0xbd,
+	0x8d, 0x89, 0x30, 0x4c, 0xcc, 0xe0, 0xe1, 0xcd,
+	0x4d, 0x31, 0x88, 0x8c, 0xbc, 0x50, 0x0f, 0xed,
+	0xbd, 0xed, 0x87, 0x10, 0xbc, 0x0c, 0x38, 0x41,
+	0x4d, 0xcd, 0x81, 0xd8, 0x0c, 0xbc, 0x70, 0xd7,
+	0x6d, 0xbd, 0x2d,
+};
+static u32 sizeof_special1 = sizeof(sample_special1);
+
+static u8 sample_special2[] = {
+	0x2a, 0xcc, 0x8c, 0xd7, 0x4d, 0x2d, 0x18, 0xbc,
+	0x10, 0xc1, 0xbd, 0xc1, 0x10, 0xbc, 0x18, 0x2d,
+	0x4d, 0xd7, 0x8c, 0xcc, 0x2a, 0xcd, 0x8d, 0xd6,
+	0x4c, 0x2c, 0x19, 0xbd, 0x11, 0xc0, 0xbc, 0xc0,
+	0x11, 0xbd, 0x19, 0x2c, 0x4c, 0xd6, 0x8d, 0xcd,
+	0x2a, 0xcc, 0x8c, 0xd7, 0x4d, 0x2d, 0x18, 0xbc,
+	0x10, 0xc1, 0xbd, 0xc1, 0x10, 0xbc, 0x18, 0x2d,
+	0x4d, 0xd7, 0x8c, 0xcc, 0x2a, 0xcd, 0x8d, 0xd6,
+	0x4c, 0x2c, 0x19, 0xbd, 0x11, 0xc0, 0xbc, 0xc0,
+	0x11, 0xbd, 0x19, 0x2c, 0x4c, 0xd6, 0x8d, 0xcd,
+};
+static u32 sizeof_special2 = sizeof(sample_special2);
+
+static u8 sample_special3[] = {
+	0x2a, 0xbc, 0x18, 0xcd, 0x11, 0x2c, 0x8c, 0xc1,
+	0x4d, 0xd6, 0xbc, 0xd6, 0x4d, 0xc1, 0x8c, 0x2c,
+	0x11, 0xcd, 0x18, 0xbc, 0x2a, 0xbd, 0x19, 0xcc,
+	0x10, 0x2d, 0x8d, 0xc0, 0x4c, 0xd7, 0xbd, 0xd7,
+	0x4c, 0xc0, 0x8d, 0x2d, 0x10, 0xcc, 0x19, 0xbd,
+	0x2a, 0xbc, 0x18, 0xcd, 0x11, 0x2c, 0x8c, 0xc1,
+	0x4d, 0xd6, 0xbc, 0xd6, 0x4d, 0xc1, 0x8c, 0x2c,
+	0x11, 0xcd, 0x18, 0xbc, 0x2a, 0xbd, 0x19, 0xcc,
+	0x10, 0x2d, 0x8d, 0xc0, 0x4c, 0xd7, 0xbd, 0xd7,
+	0x4c, 0xc0, 0x8d, 0x2d, 0x10, 0xcc, 0x19, 0xbd,
+};
+static u32 sizeof_special3 = sizeof(sample_special3);
+
+static u8 sample_silence[] = {
+	0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a,
+	0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a,
+	0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a,
+	0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a,
+	0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a,
+	0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a,
+	0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a,
+	0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a,
+	0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a,
+	0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a,
+	0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a,
+	0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a, 0x2a,
+};
+static u32 sizeof_silence = sizeof(sample_silence);
+
+struct tones_samples {
+	u32 *len;
+	u8 *data;
+};
+static struct
+tones_samples samples[] = {
+	{&sizeof_german_all, sample_german_all},
+	{&sizeof_german_old, sample_german_old},
+	{&sizeof_american_dialtone, sample_american_dialtone},
+	{&sizeof_american_ringing, sample_american_ringing},
+	{&sizeof_american_busy, sample_american_busy},
+	{&sizeof_special1, sample_special1},
+	{&sizeof_special2, sample_special2},
+	{&sizeof_special3, sample_special3},
+	{NULL, NULL},
+};
+
+/***********************************
+ * generate ulaw from alaw samples *
+ ***********************************/
+
+void
+dsp_audio_generate_ulaw_samples(void)
+{
+	int i, j;
+
+	i = 0;
+	while (samples[i].len) {
+		j = 0;
+		while (j < (*samples[i].len)) {
+			samples[i].data[j] =
+				dsp_audio_alaw_to_ulaw[samples[i].data[j]];
+			j++;
+		}
+		i++;
+	}
+}
+
+
+/****************************
+ * tone sequence definition *
+ ****************************/
+
+struct pattern {
+	int tone;
+	u8 *data[10];
+	u32 *siz[10];
+	u32 seq[10];
+} pattern[] = {
+	{TONE_GERMAN_DIALTONE,
+	{DATA_GA, 0, 0, 0, 0, 0, 0, 0, 0, 0},
+	{SIZE_GA, 0, 0, 0, 0, 0, 0, 0, 0, 0},
+	{1900, 0, 0, 0, 0, 0, 0, 0, 0, 0} },
+
+	{TONE_GERMAN_OLDDIALTONE,
+	{DATA_GO, 0, 0, 0, 0, 0, 0, 0, 0, 0},
+	{SIZE_GO, 0, 0, 0, 0, 0, 0, 0, 0, 0},
+	{1998, 0, 0, 0, 0, 0, 0, 0, 0, 0} },
+
+	{TONE_AMERICAN_DIALTONE,
+	{DATA_DT, 0, 0, 0, 0, 0, 0, 0, 0, 0},
+	{SIZE_DT, 0, 0, 0, 0, 0, 0, 0, 0, 0},
+	{8000, 0, 0, 0, 0, 0, 0, 0, 0, 0} },
+
+	{TONE_GERMAN_DIALPBX,
+	{DATA_GA, DATA_S, DATA_GA, DATA_S, DATA_GA, DATA_S, 0, 0, 0, 0},
+	{SIZE_GA, SIZE_S, SIZE_GA, SIZE_S, SIZE_GA, SIZE_S, 0, 0, 0, 0},
+	{2000, 2000, 2000, 2000, 2000, 12000, 0, 0, 0, 0} },
+
+	{TONE_GERMAN_OLDDIALPBX,
+	{DATA_GO, DATA_S, DATA_GO, DATA_S, DATA_GO, DATA_S, 0, 0, 0, 0},
+	{SIZE_GO, SIZE_S, SIZE_GO, SIZE_S, SIZE_GO, SIZE_S, 0, 0, 0, 0},
+	{2000, 2000, 2000, 2000, 2000, 12000, 0, 0, 0, 0} },
+
+	{TONE_AMERICAN_DIALPBX,
+	{DATA_DT, DATA_S, DATA_DT, DATA_S, DATA_DT, DATA_S, 0, 0, 0, 0},
+	{SIZE_DT, SIZE_S, SIZE_DT, SIZE_S, SIZE_DT, SIZE_S, 0, 0, 0, 0},
+	{2000, 2000, 2000, 2000, 2000, 12000, 0, 0, 0, 0} },
+
+	{TONE_GERMAN_RINGING,
+	{DATA_GA, DATA_S, 0, 0, 0, 0, 0, 0, 0, 0},
+	{SIZE_GA, SIZE_S, 0, 0, 0, 0, 0, 0, 0, 0},
+	{8000, 32000, 0, 0, 0, 0, 0, 0, 0, 0} },
+
+	{TONE_GERMAN_OLDRINGING,
+	{DATA_GO, DATA_S, 0, 0, 0, 0, 0, 0, 0, 0},
+	{SIZE_GO, SIZE_S, 0, 0, 0, 0, 0, 0, 0, 0},
+	{8000, 40000, 0, 0, 0, 0, 0, 0, 0, 0} },
+
+	{TONE_AMERICAN_RINGING,
+	{DATA_RI, DATA_S, 0, 0, 0, 0, 0, 0, 0, 0},
+	{SIZE_RI, SIZE_S, 0, 0, 0, 0, 0, 0, 0, 0},
+	{8000, 32000, 0, 0, 0, 0, 0, 0, 0, 0} },
+
+	{TONE_GERMAN_RINGPBX,
+	{DATA_GA, DATA_S, DATA_GA, DATA_S, 0, 0, 0, 0, 0, 0},
+	{SIZE_GA, SIZE_S, SIZE_GA, SIZE_S, 0, 0, 0, 0, 0, 0},
+	{4000, 4000, 4000, 28000, 0, 0, 0, 0, 0, 0} },
+
+	{TONE_GERMAN_OLDRINGPBX,
+	{DATA_GO, DATA_S, DATA_GO, DATA_S, 0, 0, 0, 0, 0, 0},
+	{SIZE_GO, SIZE_S, SIZE_GO, SIZE_S, 0, 0, 0, 0, 0, 0},
+	{4000, 4000, 4000, 28000, 0, 0, 0, 0, 0, 0} },
+
+	{TONE_AMERICAN_RINGPBX,
+	{DATA_RI, DATA_S, DATA_RI, DATA_S, 0, 0, 0, 0, 0, 0},
+	{SIZE_RI, SIZE_S, SIZE_RI, SIZE_S, 0, 0, 0, 0, 0, 0},
+	{4000, 4000, 4000, 28000, 0, 0, 0, 0, 0, 0} },
+
+	{TONE_GERMAN_BUSY,
+	{DATA_GA, DATA_S, 0, 0, 0, 0, 0, 0, 0, 0},
+	{SIZE_GA, SIZE_S, 0, 0, 0, 0, 0, 0, 0, 0},
+	{4000, 4000, 0, 0, 0, 0, 0, 0, 0, 0} },
+
+	{TONE_GERMAN_OLDBUSY,
+	{DATA_GO, DATA_S, 0, 0, 0, 0, 0, 0, 0, 0},
+	{SIZE_GO, SIZE_S, 0, 0, 0, 0, 0, 0, 0, 0},
+	{1000, 5000, 0, 0, 0, 0, 0, 0, 0, 0} },
+
+	{TONE_AMERICAN_BUSY,
+	{DATA_BU, DATA_S, 0, 0, 0, 0, 0, 0, 0, 0},
+	{SIZE_BU, SIZE_S, 0, 0, 0, 0, 0, 0, 0, 0},
+	{4000, 4000, 0, 0, 0, 0, 0, 0, 0, 0} },
+
+	{TONE_GERMAN_HANGUP,
+	{DATA_GA, DATA_S, 0, 0, 0, 0, 0, 0, 0, 0},
+	{SIZE_GA, SIZE_S, 0, 0, 0, 0, 0, 0, 0, 0},
+	{4000, 4000, 0, 0, 0, 0, 0, 0, 0, 0} },
+
+	{TONE_GERMAN_OLDHANGUP,
+	{DATA_GO, DATA_S, 0, 0, 0, 0, 0, 0, 0, 0},
+	{SIZE_GO, SIZE_S, 0, 0, 0, 0, 0, 0, 0, 0},
+	{1000, 5000, 0, 0, 0, 0, 0, 0, 0, 0} },
+
+	{TONE_AMERICAN_HANGUP,
+	{DATA_DT, 0, 0, 0, 0, 0, 0, 0, 0, 0},
+	{SIZE_DT, 0, 0, 0, 0, 0, 0, 0, 0, 0},
+	{8000, 0, 0, 0, 0, 0, 0, 0, 0, 0} },
+
+	{TONE_SPECIAL_INFO,
+	{DATA_S1, DATA_S2, DATA_S3, DATA_S, 0, 0, 0, 0, 0, 0},
+	{SIZE_S1, SIZE_S2, SIZE_S3, SIZE_S, 0, 0, 0, 0, 0, 0},
+	{2666, 2666, 2666, 8002, 0, 0, 0, 0, 0, 0} },
+
+	{TONE_GERMAN_GASSENBESETZT,
+	{DATA_GA, DATA_S, 0, 0, 0, 0, 0, 0, 0, 0},
+	{SIZE_GA, SIZE_S, 0, 0, 0, 0, 0, 0, 0, 0},
+	{2000, 2000, 0, 0, 0, 0, 0, 0, 0, 0} },
+
+	{TONE_GERMAN_AUFSCHALTTON,
+	{DATA_GO, DATA_S, DATA_GO, DATA_S, 0, 0, 0, 0, 0, 0},
+	{SIZE_GO, SIZE_S, SIZE_GO, SIZE_S, 0, 0, 0, 0, 0, 0},
+	{1000, 5000, 1000, 17000, 0, 0, 0, 0, 0, 0} },
+
+	{0,
+	{0, 0, 0, 0, 0, 0, 0, 0, 0, 0},
+	{0, 0, 0, 0, 0, 0, 0, 0, 0, 0},
+	{0, 0, 0, 0, 0, 0, 0, 0, 0, 0} },
+};
+
+/******************
+ * copy tone data *
+ ******************/
+
+/* an sk_buff is generated from the number of samples needed.
+ * the count will be changed and may begin from 0 each pattern period.
+ * the clue is to precalculate the pointers and legths to use only one
+ * memcpy per function call, or two memcpy if the tone sequence changes.
+ *
+ * pattern - the type of the pattern
+ * count - the sample from the beginning of the pattern (phase)
+ * len - the number of bytes
+ *
+ * return - the sk_buff with the sample
+ *
+ * if tones has finished (e.g. knocking tone), dsp->tones is turned off
+ */
+void dsp_tone_copy(struct dsp *dsp, u8 *data, int len)
+{
+	int index, count, start, num;
+	struct pattern *pat;
+	struct dsp_tone *tone = &dsp->tone;
+
+	/* if we have no tone, we copy silence */
+	if (!tone->tone) {
+		memset(data, dsp_silence, len);
+		return;
+	}
+
+	/* process pattern */
+	pat = (struct pattern *)tone->pattern;
+		/* points to the current pattern */
+	index = tone->index; /* gives current sequence index */
+	count = tone->count; /* gives current sample */
+
+	/* copy sample */
+	while (len) {
+		/* find sample to start with */
+		while (42) {
+			/* warp arround */
+			if (!pat->seq[index]) {
+				count = 0;
+				index = 0;
+			}
+			/* check if we are currently playing this tone */
+			if (count < pat->seq[index])
+				break;
+			if (dsp_debug & DEBUG_DSP_TONE)
+				printk(KERN_DEBUG "%s: reaching next sequence "
+					"(index=%d)\n", __func__, index);
+			count -= pat->seq[index];
+			index++;
+		}
+		/* calculate start and number of samples */
+		start = count % (*(pat->siz[index]));
+		num = len;
+		if (num+count > pat->seq[index])
+			num = pat->seq[index] - count;
+		if (num+start > (*(pat->siz[index])))
+			num = (*(pat->siz[index])) - start;
+		/* copy memory */
+		memcpy(data, pat->data[index]+start, num);
+		/* reduce length */
+		data += num;
+		count += num;
+		len -= num;
+	}
+	tone->index = index;
+	tone->count = count;
+
+	/* return sk_buff */
+	return;
+}
+
+
+/*******************************
+ * send HW message to hfc card *
+ *******************************/
+
+static void
+dsp_tone_hw_message(struct dsp *dsp, u8 *sample, int len)
+{
+	struct sk_buff *nskb;
+
+	/* unlocking is not required, because we don't expect a response */
+	nskb = _alloc_mISDN_skb(PH_CONTROL_REQ,
+		(len)?HFC_SPL_LOOP_ON:HFC_SPL_LOOP_OFF, len, sample,
+		GFP_ATOMIC);
+	if (nskb) {
+		if (dsp->ch.peer) {
+			if (dsp->ch.recv(dsp->ch.peer, nskb))
+				dev_kfree_skb(nskb);
+		} else
+			dev_kfree_skb(nskb);
+	}
+}
+
+
+/*****************
+ * timer expires *
+ *****************/
+void
+dsp_tone_timeout(void *arg)
+{
+	struct dsp *dsp = arg;
+	struct dsp_tone *tone = &dsp->tone;
+	struct pattern *pat = (struct pattern *)tone->pattern;
+	int index = tone->index;
+
+	if (!tone->tone)
+		return;
+
+	index++;
+	if (!pat->seq[index])
+		index = 0;
+	tone->index = index;
+
+	/* set next tone */
+	if (pat->data[index] == DATA_S)
+		dsp_tone_hw_message(dsp, 0, 0);
+	else
+		dsp_tone_hw_message(dsp, pat->data[index], *(pat->siz[index]));
+	/* set timer */
+	init_timer(&tone->tl);
+	tone->tl.expires = jiffies + (pat->seq[index] * HZ) / 8000;
+	add_timer(&tone->tl);
+}
+
+
+/********************
+ * set/release tone *
+ ********************/
+
+/*
+ * tones are relaized by streaming or by special loop commands if supported
+ * by hardware. when hardware is used, the patterns will be controlled by
+ * timers.
+ */
+int
+dsp_tone(struct dsp *dsp, int tone)
+{
+	struct pattern *pat;
+	int i;
+	struct dsp_tone *tonet = &dsp->tone;
+
+	tonet->software = 0;
+	tonet->hardware = 0;
+
+	/* we turn off the tone */
+	if (!tone) {
+		if (dsp->features.hfc_loops)
+		if (timer_pending(&tonet->tl))
+			del_timer(&tonet->tl);
+		if (dsp->features.hfc_loops)
+			dsp_tone_hw_message(dsp, NULL, 0);
+		tonet->tone = 0;
+		return 0;
+	}
+
+	pat = NULL;
+	i = 0;
+	while (pattern[i].tone) {
+		if (pattern[i].tone == tone) {
+			pat = &pattern[i];
+			break;
+		}
+		i++;
+	}
+	if (!pat) {
+		printk(KERN_WARNING "dsp: given tone 0x%x is invalid\n", tone);
+		return -EINVAL;
+	}
+	if (dsp_debug & DEBUG_DSP_TONE)
+		printk(KERN_DEBUG "%s: now starting tone %d (index=%d)\n",
+			__func__, tone, 0);
+	tonet->tone = tone;
+	tonet->pattern = pat;
+	tonet->index = 0;
+	tonet->count = 0;
+
+	if (dsp->features.hfc_loops) {
+		tonet->hardware = 1;
+		/* set first tone */
+		dsp_tone_hw_message(dsp, pat->data[0], *(pat->siz[0]));
+		/* set timer */
+		if (timer_pending(&tonet->tl))
+			del_timer(&tonet->tl);
+		init_timer(&tonet->tl);
+		tonet->tl.expires = jiffies + (pat->seq[0] * HZ) / 8000;
+		add_timer(&tonet->tl);
+	} else {
+		tonet->software = 1;
+	}
+
+	return 0;
+}
+
+
+
+
+
diff --git a/include/linux/mISDNdsp.h b/include/linux/mISDNdsp.h
new file mode 100644
index 000000000000..6b71d2dce508
--- /dev/null
+++ b/include/linux/mISDNdsp.h
@@ -0,0 +1,37 @@
+#ifndef __mISDNdsp_H__
+#define __mISDNdsp_H__
+
+struct mISDN_dsp_element_arg {
+	char	*name;
+	char	*def;
+	char	*desc;
+};
+
+struct mISDN_dsp_element {
+	char	*name;
+	void	*(*new)(const char *arg);
+	void	(*free)(void *p);
+	void	(*process_tx)(void *p, unsigned char *data, int len);
+	void	(*process_rx)(void *p, unsigned char *data, int len);
+	int	num_args;
+	struct mISDN_dsp_element_arg
+		*args;
+};
+
+extern int  mISDN_dsp_element_register(struct mISDN_dsp_element *elem);
+extern void mISDN_dsp_element_unregister(struct mISDN_dsp_element *elem);
+
+struct dsp_features {
+	int	hfc_id; /* unique id to identify the chip (or -1) */
+	int	hfc_dtmf; /* set if HFCmulti card supports dtmf */
+	int	hfc_loops; /* set if card supports tone loops */
+	int	hfc_echocanhw; /* set if card supports echocancelation*/
+	int	pcm_id; /* unique id to identify the pcm bus (or -1) */
+	int	pcm_slots; /* number of slots on the pcm bus */
+	int	pcm_banks; /* number of IO banks of pcm bus */
+	int	unclocked; /* data is not clocked (has jitter/loss) */
+	int	unordered; /* data is unordered (packets have index) */
+};
+
+#endif
+

From 1700fe1a10dc0eaac0ef60a8093eaeafa9bff9ae Mon Sep 17 00:00:00 2001
From: Karsten Keil <kkeil@suse.de>
Date: Sat, 26 Jul 2008 18:55:28 +0200
Subject: [PATCH 1116/1171] Add mISDN HFC PCI driver

Enable support for card with Cologne Chip AG's
HFC PCIbased cards

Signed-off-by: Karsten Keil <kkeil@suse.de>
---
 drivers/isdn/hardware/mISDN/Kconfig   |   13 +
 drivers/isdn/hardware/mISDN/Makefile  |    6 +
 drivers/isdn/hardware/mISDN/hfc_pci.h |  228 +++
 drivers/isdn/hardware/mISDN/hfcpci.c  | 2256 +++++++++++++++++++++++++
 4 files changed, 2503 insertions(+)
 create mode 100644 drivers/isdn/hardware/mISDN/Kconfig
 create mode 100644 drivers/isdn/hardware/mISDN/Makefile
 create mode 100644 drivers/isdn/hardware/mISDN/hfc_pci.h
 create mode 100644 drivers/isdn/hardware/mISDN/hfcpci.c

diff --git a/drivers/isdn/hardware/mISDN/Kconfig b/drivers/isdn/hardware/mISDN/Kconfig
new file mode 100644
index 000000000000..f62dc8752be9
--- /dev/null
+++ b/drivers/isdn/hardware/mISDN/Kconfig
@@ -0,0 +1,13 @@
+#
+# Hardware for mISDN
+#
+comment "mISDN hardware drivers"
+
+config MISDN_HFCPCI
+	tristate "Support for HFC PCI cards"
+	depends on MISDN
+	depends on PCI
+	help
+	  Enable support for cards with Cologne Chip AG's
+          HFC PCI chip.
+
diff --git a/drivers/isdn/hardware/mISDN/Makefile b/drivers/isdn/hardware/mISDN/Makefile
new file mode 100644
index 000000000000..6f20a40b9d54
--- /dev/null
+++ b/drivers/isdn/hardware/mISDN/Makefile
@@ -0,0 +1,6 @@
+#
+# Makefile for the modular ISDN hardware drivers
+#
+#
+
+obj-$(CONFIG_MISDN_HFCPCI) += hfcpci.o
diff --git a/drivers/isdn/hardware/mISDN/hfc_pci.h b/drivers/isdn/hardware/mISDN/hfc_pci.h
new file mode 100644
index 000000000000..fd2c9be6d849
--- /dev/null
+++ b/drivers/isdn/hardware/mISDN/hfc_pci.h
@@ -0,0 +1,228 @@
+/*
+ *  specific defines for CCD's HFC 2BDS0 PCI chips
+ *
+ * Author     Werner Cornelius (werner@isdn4linux.de)
+ *
+ * Copyright 1999  by Werner Cornelius (werner@isdn4linux.de)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2, or (at your option)
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ *
+ */
+
+/*
+ * thresholds for transparent B-channel mode
+ * change mask and threshold simultaneously
+ */
+#define HFCPCI_BTRANS_THRESHOLD 128
+#define HFCPCI_BTRANS_MAX	256
+#define HFCPCI_BTRANS_THRESMASK 0x00
+
+/* defines for PCI config */
+#define PCI_ENA_MEMIO		0x02
+#define PCI_ENA_MASTER		0x04
+
+/* GCI/IOM bus monitor registers */
+#define HCFPCI_C_I		0x08
+#define HFCPCI_TRxR		0x0C
+#define HFCPCI_MON1_D		0x28
+#define HFCPCI_MON2_D		0x2C
+
+/* GCI/IOM bus timeslot registers */
+#define HFCPCI_B1_SSL		0x80
+#define HFCPCI_B2_SSL		0x84
+#define HFCPCI_AUX1_SSL		0x88
+#define HFCPCI_AUX2_SSL		0x8C
+#define HFCPCI_B1_RSL		0x90
+#define HFCPCI_B2_RSL		0x94
+#define HFCPCI_AUX1_RSL		0x98
+#define HFCPCI_AUX2_RSL		0x9C
+
+/* GCI/IOM bus data registers */
+#define HFCPCI_B1_D		0xA0
+#define HFCPCI_B2_D		0xA4
+#define HFCPCI_AUX1_D		0xA8
+#define HFCPCI_AUX2_D		0xAC
+
+/* GCI/IOM bus configuration registers */
+#define HFCPCI_MST_EMOD		0xB4
+#define HFCPCI_MST_MODE		0xB8
+#define HFCPCI_CONNECT 		0xBC
+
+
+/* Interrupt and status registers */
+#define HFCPCI_FIFO_EN		0x44
+#define HFCPCI_TRM		0x48
+#define HFCPCI_B_MODE		0x4C
+#define HFCPCI_CHIP_ID		0x58
+#define HFCPCI_CIRM		0x60
+#define HFCPCI_CTMT		0x64
+#define HFCPCI_INT_M1		0x68
+#define HFCPCI_INT_M2		0x6C
+#define HFCPCI_INT_S1		0x78
+#define HFCPCI_INT_S2		0x7C
+#define HFCPCI_STATUS		0x70
+
+/* S/T section registers */
+#define HFCPCI_STATES		0xC0
+#define HFCPCI_SCTRL		0xC4
+#define HFCPCI_SCTRL_E		0xC8
+#define HFCPCI_SCTRL_R		0xCC
+#define HFCPCI_SQ		0xD0
+#define HFCPCI_CLKDEL		0xDC
+#define HFCPCI_B1_REC		0xF0
+#define HFCPCI_B1_SEND		0xF0
+#define HFCPCI_B2_REC		0xF4
+#define HFCPCI_B2_SEND		0xF4
+#define HFCPCI_D_REC		0xF8
+#define HFCPCI_D_SEND		0xF8
+#define HFCPCI_E_REC		0xFC
+
+
+/* bits in status register (READ) */
+#define HFCPCI_PCI_PROC		0x02
+#define HFCPCI_NBUSY		0x04
+#define HFCPCI_TIMER_ELAP	0x10
+#define HFCPCI_STATINT		0x20
+#define HFCPCI_FRAMEINT		0x40
+#define HFCPCI_ANYINT		0x80
+
+/* bits in CTMT (Write) */
+#define HFCPCI_CLTIMER		0x80
+#define HFCPCI_TIM3_125		0x04
+#define HFCPCI_TIM25		0x10
+#define HFCPCI_TIM50		0x14
+#define HFCPCI_TIM400		0x18
+#define HFCPCI_TIM800		0x1C
+#define HFCPCI_AUTO_TIMER	0x20
+#define HFCPCI_TRANSB2		0x02
+#define HFCPCI_TRANSB1		0x01
+
+/* bits in CIRM (Write) */
+#define HFCPCI_AUX_MSK		0x07
+#define HFCPCI_RESET		0x08
+#define HFCPCI_B1_REV		0x40
+#define HFCPCI_B2_REV		0x80
+
+/* bits in INT_M1 and INT_S1 */
+#define HFCPCI_INTS_B1TRANS	0x01
+#define HFCPCI_INTS_B2TRANS	0x02
+#define HFCPCI_INTS_DTRANS	0x04
+#define HFCPCI_INTS_B1REC	0x08
+#define HFCPCI_INTS_B2REC	0x10
+#define HFCPCI_INTS_DREC	0x20
+#define HFCPCI_INTS_L1STATE	0x40
+#define HFCPCI_INTS_TIMER	0x80
+
+/* bits in INT_M2 */
+#define HFCPCI_PROC_TRANS	0x01
+#define HFCPCI_GCI_I_CHG	0x02
+#define HFCPCI_GCI_MON_REC	0x04
+#define HFCPCI_IRQ_ENABLE	0x08
+#define HFCPCI_PMESEL		0x80
+
+/* bits in STATES */
+#define HFCPCI_STATE_MSK	0x0F
+#define HFCPCI_LOAD_STATE	0x10
+#define HFCPCI_ACTIVATE		0x20
+#define HFCPCI_DO_ACTION	0x40
+#define HFCPCI_NT_G2_G3		0x80
+
+/* bits in HFCD_MST_MODE */
+#define HFCPCI_MASTER		0x01
+#define HFCPCI_SLAVE		0x00
+#define HFCPCI_F0IO_POSITIV	0x02
+#define HFCPCI_F0_NEGATIV	0x04
+#define HFCPCI_F0_2C4		0x08
+/* remaining bits are for codecs control */
+
+/* bits in HFCD_SCTRL */
+#define SCTRL_B1_ENA		0x01
+#define SCTRL_B2_ENA		0x02
+#define SCTRL_MODE_TE		0x00
+#define SCTRL_MODE_NT		0x04
+#define SCTRL_LOW_PRIO		0x08
+#define SCTRL_SQ_ENA		0x10
+#define SCTRL_TEST		0x20
+#define SCTRL_NONE_CAP		0x40
+#define SCTRL_PWR_DOWN		0x80
+
+/* bits in SCTRL_E  */
+#define HFCPCI_AUTO_AWAKE	0x01
+#define HFCPCI_DBIT_1		0x04
+#define HFCPCI_IGNORE_COL	0x08
+#define HFCPCI_CHG_B1_B2	0x80
+
+/* bits in FIFO_EN register */
+#define HFCPCI_FIFOEN_B1	0x03
+#define HFCPCI_FIFOEN_B2	0x0C
+#define HFCPCI_FIFOEN_DTX	0x10
+#define HFCPCI_FIFOEN_B1TX	0x01
+#define HFCPCI_FIFOEN_B1RX	0x02
+#define HFCPCI_FIFOEN_B2TX	0x04
+#define HFCPCI_FIFOEN_B2RX	0x08
+
+
+/* definitions of fifo memory area */
+#define MAX_D_FRAMES 15
+#define MAX_B_FRAMES 31
+#define B_SUB_VAL    0x200
+#define B_FIFO_SIZE  (0x2000 - B_SUB_VAL)
+#define D_FIFO_SIZE  512
+#define D_FREG_MASK  0xF
+
+struct zt {
+	unsigned short z1;  /* Z1 pointer 16 Bit */
+	unsigned short z2;  /* Z2 pointer 16 Bit */
+};
+
+struct dfifo {
+	u_char data[D_FIFO_SIZE]; /* FIFO data space */
+	u_char fill1[0x20A0-D_FIFO_SIZE]; /* reserved, do not use */
+	u_char f1, f2; /* f pointers */
+	u_char fill2[0x20C0-0x20A2]; /* reserved, do not use */
+	/* mask index with D_FREG_MASK for access */
+	struct zt za[MAX_D_FRAMES+1];
+	u_char fill3[0x4000-0x2100]; /* align 16K */
+};
+
+struct bzfifo {
+	struct zt	za[MAX_B_FRAMES+1]; /* only range 0x0..0x1F allowed */
+	u_char		f1, f2; /* f pointers */
+	u_char		fill[0x2100-0x2082]; /* alignment */
+};
+
+
+union fifo_area {
+	struct {
+		struct dfifo d_tx; /* D-send channel */
+		struct dfifo d_rx; /* D-receive channel */
+	} d_chan;
+	struct {
+		u_char		fill1[0x200];
+		u_char		txdat_b1[B_FIFO_SIZE];
+		struct bzfifo	txbz_b1;
+		struct bzfifo	txbz_b2;
+		u_char		txdat_b2[B_FIFO_SIZE];
+		u_char		fill2[D_FIFO_SIZE];
+		u_char		rxdat_b1[B_FIFO_SIZE];
+		struct bzfifo	rxbz_b1;
+		struct bzfifo	rxbz_b2;
+		u_char rxdat_b2[B_FIFO_SIZE];
+	} b_chans;
+	u_char fill[32768];
+};
+
+#define Write_hfc(a, b, c) (writeb(c, (a->hw.pci_io)+b))
+#define Read_hfc(a, b) (readb((a->hw.pci_io)+b))
diff --git a/drivers/isdn/hardware/mISDN/hfcpci.c b/drivers/isdn/hardware/mISDN/hfcpci.c
new file mode 100644
index 000000000000..917968530e1e
--- /dev/null
+++ b/drivers/isdn/hardware/mISDN/hfcpci.c
@@ -0,0 +1,2256 @@
+/*
+ *
+ * hfcpci.c     low level driver for CCD's hfc-pci based cards
+ *
+ * Author     Werner Cornelius (werner@isdn4linux.de)
+ *            based on existing driver for CCD hfc ISA cards
+ *            type approval valid for HFC-S PCI A based card
+ *
+ * Copyright 1999  by Werner Cornelius (werner@isdn-development.de)
+ * Copyright 2008  by Karsten Keil <kkeil@novell.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2, or (at your option)
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include <linux/mISDNhw.h>
+
+#include "hfc_pci.h"
+
+static const char *hfcpci_revision = "2.0";
+
+#define MAX_CARDS	8
+static int HFC_cnt;
+static uint debug;
+
+MODULE_AUTHOR("Karsten Keil");
+MODULE_LICENSE("GPL");
+module_param(debug, uint, 0);
+
+static LIST_HEAD(HFClist);
+DEFINE_RWLOCK(HFClock);
+
+enum {
+	HFC_CCD_2BD0,
+	HFC_CCD_B000,
+	HFC_CCD_B006,
+	HFC_CCD_B007,
+	HFC_CCD_B008,
+	HFC_CCD_B009,
+	HFC_CCD_B00A,
+	HFC_CCD_B00B,
+	HFC_CCD_B00C,
+	HFC_CCD_B100,
+	HFC_CCD_B700,
+	HFC_CCD_B701,
+	HFC_ASUS_0675,
+	HFC_BERKOM_A1T,
+	HFC_BERKOM_TCONCEPT,
+	HFC_ANIGMA_MC145575,
+	HFC_ZOLTRIX_2BD0,
+	HFC_DIGI_DF_M_IOM2_E,
+	HFC_DIGI_DF_M_E,
+	HFC_DIGI_DF_M_IOM2_A,
+	HFC_DIGI_DF_M_A,
+	HFC_ABOCOM_2BD1,
+	HFC_SITECOM_DC105V2,
+};
+
+struct hfcPCI_hw {
+	unsigned char		cirm;
+	unsigned char		ctmt;
+	unsigned char		clkdel;
+	unsigned char		states;
+	unsigned char		conn;
+	unsigned char		mst_m;
+	unsigned char		int_m1;
+	unsigned char		int_m2;
+	unsigned char		sctrl;
+	unsigned char		sctrl_r;
+	unsigned char		sctrl_e;
+	unsigned char		trm;
+	unsigned char		fifo_en;
+	unsigned char		bswapped;
+	unsigned char		protocol;
+	int			nt_timer;
+	unsigned char		*pci_io; /* start of PCI IO memory */
+	dma_addr_t		dmahandle;
+	void			*fifos; /* FIFO memory */
+	int			last_bfifo_cnt[2];
+	    /* marker saving last b-fifo frame count */
+	struct timer_list	timer;
+};
+
+#define	HFC_CFG_MASTER		1
+#define HFC_CFG_SLAVE		2
+#define	HFC_CFG_PCM		3
+#define HFC_CFG_2HFC		4
+#define HFC_CFG_SLAVEHFC	5
+#define HFC_CFG_NEG_F0		6
+#define HFC_CFG_SW_DD_DU	7
+
+#define FLG_HFC_TIMER_T1	16
+#define FLG_HFC_TIMER_T3	17
+
+#define NT_T1_COUNT	1120	/* number of 3.125ms interrupts (3.5s) */
+#define NT_T3_COUNT	31	/* number of 3.125ms interrupts (97 ms) */
+#define CLKDEL_TE	0x0e	/* CLKDEL in TE mode */
+#define CLKDEL_NT	0x6c	/* CLKDEL in NT mode */
+
+
+struct hfc_pci {
+	struct list_head	list;
+	u_char			subtype;
+	u_char			chanlimit;
+	u_char			initdone;
+	u_long			cfg;
+	u_int			irq;
+	u_int			irqcnt;
+	struct pci_dev		*pdev;
+	struct hfcPCI_hw	hw;
+	spinlock_t		lock;	/* card lock */
+	struct dchannel		dch;
+	struct bchannel		bch[2];
+};
+
+/* Interface functions */
+static void
+enable_hwirq(struct hfc_pci *hc)
+{
+	hc->hw.int_m2 |= HFCPCI_IRQ_ENABLE;
+	Write_hfc(hc, HFCPCI_INT_M2, hc->hw.int_m2);
+}
+
+static void
+disable_hwirq(struct hfc_pci *hc)
+{
+	hc->hw.int_m2 &= ~((u_char)HFCPCI_IRQ_ENABLE);
+	Write_hfc(hc, HFCPCI_INT_M2, hc->hw.int_m2);
+}
+
+/*
+ * free hardware resources used by driver
+ */
+static void
+release_io_hfcpci(struct hfc_pci *hc)
+{
+	/* disable memory mapped ports + busmaster */
+	pci_write_config_word(hc->pdev, PCI_COMMAND, 0);
+	del_timer(&hc->hw.timer);
+	pci_free_consistent(hc->pdev, 0x8000, hc->hw.fifos, hc->hw.dmahandle);
+	iounmap((void *)hc->hw.pci_io);
+}
+
+/*
+ * set mode (NT or TE)
+ */
+static void
+hfcpci_setmode(struct hfc_pci *hc)
+{
+	if (hc->hw.protocol == ISDN_P_NT_S0) {
+		hc->hw.clkdel = CLKDEL_NT;	/* ST-Bit delay for NT-Mode */
+		hc->hw.sctrl |= SCTRL_MODE_NT;	/* NT-MODE */
+		hc->hw.states = 1;		/* G1 */
+	} else {
+		hc->hw.clkdel = CLKDEL_TE;	/* ST-Bit delay for TE-Mode */
+		hc->hw.sctrl &= ~SCTRL_MODE_NT;	/* TE-MODE */
+		hc->hw.states = 2;		/* F2 */
+	}
+	Write_hfc(hc, HFCPCI_CLKDEL, hc->hw.clkdel);
+	Write_hfc(hc, HFCPCI_STATES, HFCPCI_LOAD_STATE | hc->hw.states);
+	udelay(10);
+	Write_hfc(hc, HFCPCI_STATES, hc->hw.states | 0x40); /* Deactivate */
+	Write_hfc(hc, HFCPCI_SCTRL, hc->hw.sctrl);
+}
+
+/*
+ * function called to reset the HFC PCI chip. A complete software reset of chip
+ * and fifos is done.
+ */
+static void
+reset_hfcpci(struct hfc_pci *hc)
+{
+	u_char	val;
+	int	cnt = 0;
+
+	printk(KERN_DEBUG "reset_hfcpci: entered\n");
+	val = Read_hfc(hc, HFCPCI_CHIP_ID);
+	printk(KERN_INFO "HFC_PCI: resetting HFC ChipId(%x)\n", val);
+	/* enable memory mapped ports, disable busmaster */
+	pci_write_config_word(hc->pdev, PCI_COMMAND, PCI_ENA_MEMIO);
+	disable_hwirq(hc);
+	/* enable memory ports + busmaster */
+	pci_write_config_word(hc->pdev, PCI_COMMAND,
+	    PCI_ENA_MEMIO + PCI_ENA_MASTER);
+	val = Read_hfc(hc, HFCPCI_STATUS);
+	printk(KERN_DEBUG "HFC-PCI status(%x) before reset\n", val);
+	hc->hw.cirm = HFCPCI_RESET;	/* Reset On */
+	Write_hfc(hc, HFCPCI_CIRM, hc->hw.cirm);
+	set_current_state(TASK_UNINTERRUPTIBLE);
+	mdelay(10);			/* Timeout 10ms */
+	hc->hw.cirm = 0;		/* Reset Off */
+	Write_hfc(hc, HFCPCI_CIRM, hc->hw.cirm);
+	val = Read_hfc(hc, HFCPCI_STATUS);
+	printk(KERN_DEBUG "HFC-PCI status(%x) after reset\n", val);
+	while (cnt < 50000) { /* max 50000 us */
+		udelay(5);
+		cnt += 5;
+		val = Read_hfc(hc, HFCPCI_STATUS);
+		if (!(val & 2))
+			break;
+	}
+	printk(KERN_DEBUG "HFC-PCI status(%x) after %dus\n", val, cnt);
+
+	hc->hw.fifo_en = 0x30;	/* only D fifos enabled */
+
+	hc->hw.bswapped = 0;	/* no exchange */
+	hc->hw.ctmt = HFCPCI_TIM3_125 | HFCPCI_AUTO_TIMER;
+	hc->hw.trm = HFCPCI_BTRANS_THRESMASK; /* no echo connect , threshold */
+	hc->hw.sctrl = 0x40;	/* set tx_lo mode, error in datasheet ! */
+	hc->hw.sctrl_r = 0;
+	hc->hw.sctrl_e = HFCPCI_AUTO_AWAKE;	/* S/T Auto awake */
+	hc->hw.mst_m = 0;
+	if (test_bit(HFC_CFG_MASTER, &hc->cfg))
+		hc->hw.mst_m |= HFCPCI_MASTER;	/* HFC Master Mode */
+	if (test_bit(HFC_CFG_NEG_F0, &hc->cfg))
+		hc->hw.mst_m |= HFCPCI_F0_NEGATIV;
+	Write_hfc(hc, HFCPCI_FIFO_EN, hc->hw.fifo_en);
+	Write_hfc(hc, HFCPCI_TRM, hc->hw.trm);
+	Write_hfc(hc, HFCPCI_SCTRL_E, hc->hw.sctrl_e);
+	Write_hfc(hc, HFCPCI_CTMT, hc->hw.ctmt);
+
+	hc->hw.int_m1 = HFCPCI_INTS_DTRANS | HFCPCI_INTS_DREC |
+	    HFCPCI_INTS_L1STATE | HFCPCI_INTS_TIMER;
+	Write_hfc(hc, HFCPCI_INT_M1, hc->hw.int_m1);
+
+	/* Clear already pending ints */
+	if (Read_hfc(hc, HFCPCI_INT_S1));
+
+	/* set NT/TE mode */
+	hfcpci_setmode(hc);
+
+	Write_hfc(hc, HFCPCI_MST_MODE, hc->hw.mst_m);
+	Write_hfc(hc, HFCPCI_SCTRL_R, hc->hw.sctrl_r);
+
+	/*
+	 * Init GCI/IOM2 in master mode
+	 * Slots 0 and 1 are set for B-chan 1 and 2
+	 * D- and monitor/CI channel are not enabled
+	 * STIO1 is used as output for data, B1+B2 from ST->IOM+HFC
+	 * STIO2 is used as data input, B1+B2 from IOM->ST
+	 * ST B-channel send disabled -> continous 1s
+	 * The IOM slots are always enabled
+	 */
+	if (test_bit(HFC_CFG_PCM, &hc->cfg)) {
+		/* set data flow directions: connect B1,B2: HFC to/from PCM */
+		hc->hw.conn = 0x09;
+	} else {
+		hc->hw.conn = 0x36;	/* set data flow directions */
+		if (test_bit(HFC_CFG_SW_DD_DU, &hc->cfg)) {
+			Write_hfc(hc, HFCPCI_B1_SSL, 0xC0);
+			Write_hfc(hc, HFCPCI_B2_SSL, 0xC1);
+			Write_hfc(hc, HFCPCI_B1_RSL, 0xC0);
+			Write_hfc(hc, HFCPCI_B2_RSL, 0xC1);
+		} else {
+			Write_hfc(hc, HFCPCI_B1_SSL, 0x80);
+			Write_hfc(hc, HFCPCI_B2_SSL, 0x81);
+			Write_hfc(hc, HFCPCI_B1_RSL, 0x80);
+			Write_hfc(hc, HFCPCI_B2_RSL, 0x81);
+		}
+	}
+	Write_hfc(hc, HFCPCI_CONNECT, hc->hw.conn);
+	val = Read_hfc(hc, HFCPCI_INT_S2);
+}
+
+/*
+ * Timer function called when kernel timer expires
+ */
+static void
+hfcpci_Timer(struct hfc_pci *hc)
+{
+	hc->hw.timer.expires = jiffies + 75;
+	/* WD RESET */
+/*
+ *	WriteReg(hc, HFCD_DATA, HFCD_CTMT, hc->hw.ctmt | 0x80);
+ *	add_timer(&hc->hw.timer);
+ */
+}
+
+
+/*
+ * select a b-channel entry matching and active
+ */
+static struct bchannel *
+Sel_BCS(struct hfc_pci *hc, int channel)
+{
+	if (test_bit(FLG_ACTIVE, &hc->bch[0].Flags) &&
+		(hc->bch[0].nr & channel))
+		return &hc->bch[0];
+	else if (test_bit(FLG_ACTIVE, &hc->bch[1].Flags) &&
+		(hc->bch[1].nr & channel))
+		return &hc->bch[1];
+	else
+		return NULL;
+}
+
+/*
+ * clear the desired B-channel rx fifo
+ */
+static void
+hfcpci_clear_fifo_rx(struct hfc_pci *hc, int fifo)
+{
+	u_char		fifo_state;
+	struct bzfifo	*bzr;
+
+	if (fifo) {
+		bzr = &((union fifo_area *)(hc->hw.fifos))->b_chans.rxbz_b2;
+		fifo_state = hc->hw.fifo_en & HFCPCI_FIFOEN_B2RX;
+	} else {
+		bzr = &((union fifo_area *)(hc->hw.fifos))->b_chans.rxbz_b1;
+		fifo_state = hc->hw.fifo_en & HFCPCI_FIFOEN_B1RX;
+	}
+	if (fifo_state)
+		hc->hw.fifo_en ^= fifo_state;
+	Write_hfc(hc, HFCPCI_FIFO_EN, hc->hw.fifo_en);
+	hc->hw.last_bfifo_cnt[fifo] = 0;
+	bzr->f1 = MAX_B_FRAMES;
+	bzr->f2 = bzr->f1;	/* init F pointers to remain constant */
+	bzr->za[MAX_B_FRAMES].z1 = cpu_to_le16(B_FIFO_SIZE + B_SUB_VAL - 1);
+	bzr->za[MAX_B_FRAMES].z2 = cpu_to_le16(
+	    le16_to_cpu(bzr->za[MAX_B_FRAMES].z1));
+	if (fifo_state)
+		hc->hw.fifo_en |= fifo_state;
+	Write_hfc(hc, HFCPCI_FIFO_EN, hc->hw.fifo_en);
+}
+
+/*
+ * clear the desired B-channel tx fifo
+ */
+static void hfcpci_clear_fifo_tx(struct hfc_pci *hc, int fifo)
+{
+	u_char		fifo_state;
+	struct bzfifo	*bzt;
+
+	if (fifo) {
+		bzt = &((union fifo_area *)(hc->hw.fifos))->b_chans.txbz_b2;
+		fifo_state = hc->hw.fifo_en & HFCPCI_FIFOEN_B2TX;
+	} else {
+		bzt = &((union fifo_area *)(hc->hw.fifos))->b_chans.txbz_b1;
+		fifo_state = hc->hw.fifo_en & HFCPCI_FIFOEN_B1TX;
+	}
+	if (fifo_state)
+		hc->hw.fifo_en ^= fifo_state;
+	Write_hfc(hc, HFCPCI_FIFO_EN, hc->hw.fifo_en);
+	if (hc->bch[fifo].debug & DEBUG_HW_BCHANNEL)
+		printk(KERN_DEBUG "hfcpci_clear_fifo_tx%d f1(%x) f2(%x) "
+		    "z1(%x) z2(%x) state(%x)\n",
+		    fifo, bzt->f1, bzt->f2,
+		    le16_to_cpu(bzt->za[MAX_B_FRAMES].z1),
+		    le16_to_cpu(bzt->za[MAX_B_FRAMES].z2),
+		    fifo_state);
+	bzt->f2 = MAX_B_FRAMES;
+	bzt->f1 = bzt->f2;	/* init F pointers to remain constant */
+	bzt->za[MAX_B_FRAMES].z1 = cpu_to_le16(B_FIFO_SIZE + B_SUB_VAL - 1);
+	bzt->za[MAX_B_FRAMES].z2 = cpu_to_le16(
+	    le16_to_cpu(bzt->za[MAX_B_FRAMES].z1 - 1));
+	if (fifo_state)
+		hc->hw.fifo_en |= fifo_state;
+	Write_hfc(hc, HFCPCI_FIFO_EN, hc->hw.fifo_en);
+	if (hc->bch[fifo].debug & DEBUG_HW_BCHANNEL)
+		printk(KERN_DEBUG
+		    "hfcpci_clear_fifo_tx%d f1(%x) f2(%x) z1(%x) z2(%x)\n",
+		    fifo, bzt->f1, bzt->f2,
+		    le16_to_cpu(bzt->za[MAX_B_FRAMES].z1),
+		    le16_to_cpu(bzt->za[MAX_B_FRAMES].z2));
+}
+
+/*
+ * read a complete B-frame out of the buffer
+ */
+static void
+hfcpci_empty_bfifo(struct bchannel *bch, struct bzfifo *bz,
+    u_char *bdata, int count)
+{
+	u_char		*ptr, *ptr1, new_f2;
+	int		total, maxlen, new_z2;
+	struct zt	*zp;
+
+	if ((bch->debug & DEBUG_HW_BCHANNEL) && !(bch->debug & DEBUG_HW_BFIFO))
+		printk(KERN_DEBUG "hfcpci_empty_fifo\n");
+	zp = &bz->za[bz->f2];	/* point to Z-Regs */
+	new_z2 = le16_to_cpu(zp->z2) + count;	/* new position in fifo */
+	if (new_z2 >= (B_FIFO_SIZE + B_SUB_VAL))
+		new_z2 -= B_FIFO_SIZE;	/* buffer wrap */
+	new_f2 = (bz->f2 + 1) & MAX_B_FRAMES;
+	if ((count > MAX_DATA_SIZE + 3) || (count < 4) ||
+	    (*(bdata + (le16_to_cpu(zp->z1) - B_SUB_VAL)))) {
+		if (bch->debug & DEBUG_HW)
+			printk(KERN_DEBUG "hfcpci_empty_fifo: incoming packet "
+			    "invalid length %d or crc\n", count);
+#ifdef ERROR_STATISTIC
+		bch->err_inv++;
+#endif
+		bz->za[new_f2].z2 = cpu_to_le16(new_z2);
+		bz->f2 = new_f2;	/* next buffer */
+	} else {
+		bch->rx_skb = mI_alloc_skb(count - 3, GFP_ATOMIC);
+		if (!bch->rx_skb) {
+			printk(KERN_WARNING "HFCPCI: receive out of memory\n");
+			return;
+		}
+		total = count;
+		count -= 3;
+		ptr = skb_put(bch->rx_skb, count);
+
+		if (le16_to_cpu(zp->z2) + count <= B_FIFO_SIZE + B_SUB_VAL)
+			maxlen = count;		/* complete transfer */
+		else
+			maxlen = B_FIFO_SIZE + B_SUB_VAL -
+			    le16_to_cpu(zp->z2);	/* maximum */
+
+		ptr1 = bdata + (le16_to_cpu(zp->z2) - B_SUB_VAL);
+		    /* start of data */
+		memcpy(ptr, ptr1, maxlen);	/* copy data */
+		count -= maxlen;
+
+		if (count) {	/* rest remaining */
+			ptr += maxlen;
+			ptr1 = bdata;	/* start of buffer */
+			memcpy(ptr, ptr1, count);	/* rest */
+		}
+		bz->za[new_f2].z2 = cpu_to_le16(new_z2);
+		bz->f2 = new_f2;	/* next buffer */
+		recv_Bchannel(bch);
+	}
+}
+
+/*
+ * D-channel receive procedure
+ */
+static int
+receive_dmsg(struct hfc_pci *hc)
+{
+	struct dchannel	*dch = &hc->dch;
+	int		maxlen;
+	int		rcnt, total;
+	int		count = 5;
+	u_char		*ptr, *ptr1;
+	struct dfifo	*df;
+	struct zt	*zp;
+
+	df = &((union fifo_area *)(hc->hw.fifos))->d_chan.d_rx;
+	while (((df->f1 & D_FREG_MASK) != (df->f2 & D_FREG_MASK)) && count--) {
+		zp = &df->za[df->f2 & D_FREG_MASK];
+		rcnt = le16_to_cpu(zp->z1) - le16_to_cpu(zp->z2);
+		if (rcnt < 0)
+			rcnt += D_FIFO_SIZE;
+		rcnt++;
+		if (dch->debug & DEBUG_HW_DCHANNEL)
+			printk(KERN_DEBUG
+			    "hfcpci recd f1(%d) f2(%d) z1(%x) z2(%x) cnt(%d)\n",
+				df->f1, df->f2,
+				le16_to_cpu(zp->z1),
+				le16_to_cpu(zp->z2),
+				rcnt);
+
+		if ((rcnt > MAX_DFRAME_LEN + 3) || (rcnt < 4) ||
+		    (df->data[le16_to_cpu(zp->z1)])) {
+			if (dch->debug & DEBUG_HW)
+				printk(KERN_DEBUG
+				    "empty_fifo hfcpci paket inv. len "
+				    "%d or crc %d\n",
+				    rcnt,
+				    df->data[le16_to_cpu(zp->z1)]);
+#ifdef ERROR_STATISTIC
+			cs->err_rx++;
+#endif
+			df->f2 = ((df->f2 + 1) & MAX_D_FRAMES) |
+			    (MAX_D_FRAMES + 1);	/* next buffer */
+			df->za[df->f2 & D_FREG_MASK].z2 =
+			    cpu_to_le16((zp->z2 + rcnt) & (D_FIFO_SIZE - 1));
+		} else {
+			dch->rx_skb = mI_alloc_skb(rcnt - 3, GFP_ATOMIC);
+			if (!dch->rx_skb) {
+				printk(KERN_WARNING
+				    "HFC-PCI: D receive out of memory\n");
+				break;
+			}
+			total = rcnt;
+			rcnt -= 3;
+			ptr = skb_put(dch->rx_skb, rcnt);
+
+			if (le16_to_cpu(zp->z2) + rcnt <= D_FIFO_SIZE)
+				maxlen = rcnt;	/* complete transfer */
+			else
+				maxlen = D_FIFO_SIZE - le16_to_cpu(zp->z2);
+				    /* maximum */
+
+			ptr1 = df->data + le16_to_cpu(zp->z2);
+			    /* start of data */
+			memcpy(ptr, ptr1, maxlen);	/* copy data */
+			rcnt -= maxlen;
+
+			if (rcnt) {	/* rest remaining */
+				ptr += maxlen;
+				ptr1 = df->data;	/* start of buffer */
+				memcpy(ptr, ptr1, rcnt);	/* rest */
+			}
+			df->f2 = ((df->f2 + 1) & MAX_D_FRAMES) |
+			    (MAX_D_FRAMES + 1);	/* next buffer */
+			df->za[df->f2 & D_FREG_MASK].z2 = cpu_to_le16((
+			    le16_to_cpu(zp->z2) + total) & (D_FIFO_SIZE - 1));
+			recv_Dchannel(dch);
+		}
+	}
+	return 1;
+}
+
+/*
+ * check for transparent receive data and read max one threshold size if avail
+ */
+int
+hfcpci_empty_fifo_trans(struct bchannel *bch, struct bzfifo *bz, u_char *bdata)
+{
+	unsigned short	*z1r, *z2r;
+	int		new_z2, fcnt, maxlen;
+	u_char		*ptr, *ptr1;
+
+	z1r = &bz->za[MAX_B_FRAMES].z1;		/* pointer to z reg */
+	z2r = z1r + 1;
+
+	fcnt = le16_to_cpu(*z1r) - le16_to_cpu(*z2r);
+	if (!fcnt)
+		return 0;	/* no data avail */
+
+	if (fcnt <= 0)
+		fcnt += B_FIFO_SIZE;	/* bytes actually buffered */
+	if (fcnt > HFCPCI_BTRANS_THRESHOLD)
+		fcnt = HFCPCI_BTRANS_THRESHOLD;		/* limit size */
+
+	new_z2 = le16_to_cpu(*z2r) + fcnt;	/* new position in fifo */
+	if (new_z2 >= (B_FIFO_SIZE + B_SUB_VAL))
+		new_z2 -= B_FIFO_SIZE;	/* buffer wrap */
+
+	bch->rx_skb = mI_alloc_skb(fcnt, GFP_ATOMIC);
+	if (bch->rx_skb) {
+		ptr = skb_put(bch->rx_skb, fcnt);
+		if (le16_to_cpu(*z2r) + fcnt <= B_FIFO_SIZE + B_SUB_VAL)
+			maxlen = fcnt;	/* complete transfer */
+		else
+			maxlen = B_FIFO_SIZE + B_SUB_VAL - le16_to_cpu(*z2r);
+			    /* maximum */
+
+		ptr1 = bdata + (le16_to_cpu(*z2r) - B_SUB_VAL);
+		    /* start of data */
+		memcpy(ptr, ptr1, maxlen);	/* copy data */
+		fcnt -= maxlen;
+
+		if (fcnt) {	/* rest remaining */
+			ptr += maxlen;
+			ptr1 = bdata;	/* start of buffer */
+			memcpy(ptr, ptr1, fcnt);	/* rest */
+		}
+		recv_Bchannel(bch);
+	} else
+		printk(KERN_WARNING "HFCPCI: receive out of memory\n");
+
+	*z2r = cpu_to_le16(new_z2);		/* new position */
+	return 1;
+}
+
+/*
+ * B-channel main receive routine
+ */
+void
+main_rec_hfcpci(struct bchannel *bch)
+{
+	struct hfc_pci	*hc = bch->hw;
+	int		rcnt, real_fifo;
+	int		receive, count = 5;
+	struct bzfifo	*bz;
+	u_char		*bdata;
+	struct zt	*zp;
+
+
+	if ((bch->nr & 2) && (!hc->hw.bswapped)) {
+		bz = &((union fifo_area *)(hc->hw.fifos))->b_chans.rxbz_b2;
+		bdata = ((union fifo_area *)(hc->hw.fifos))->b_chans.rxdat_b2;
+		real_fifo = 1;
+	} else {
+		bz = &((union fifo_area *)(hc->hw.fifos))->b_chans.rxbz_b1;
+		bdata = ((union fifo_area *)(hc->hw.fifos))->b_chans.rxdat_b1;
+		real_fifo = 0;
+	}
+Begin:
+	count--;
+	if (bz->f1 != bz->f2) {
+		if (bch->debug & DEBUG_HW_BCHANNEL)
+			printk(KERN_DEBUG "hfcpci rec ch(%x) f1(%d) f2(%d)\n",
+			    bch->nr, bz->f1, bz->f2);
+		zp = &bz->za[bz->f2];
+
+		rcnt = le16_to_cpu(zp->z1) - le16_to_cpu(zp->z2);
+		if (rcnt < 0)
+			rcnt += B_FIFO_SIZE;
+		rcnt++;
+		if (bch->debug & DEBUG_HW_BCHANNEL)
+			printk(KERN_DEBUG
+			    "hfcpci rec ch(%x) z1(%x) z2(%x) cnt(%d)\n",
+			    bch->nr, le16_to_cpu(zp->z1),
+			    le16_to_cpu(zp->z2), rcnt);
+		hfcpci_empty_bfifo(bch, bz, bdata, rcnt);
+		rcnt = bz->f1 - bz->f2;
+		if (rcnt < 0)
+			rcnt += MAX_B_FRAMES + 1;
+		if (hc->hw.last_bfifo_cnt[real_fifo] > rcnt + 1) {
+			rcnt = 0;
+			hfcpci_clear_fifo_rx(hc, real_fifo);
+		}
+		hc->hw.last_bfifo_cnt[real_fifo] = rcnt;
+		if (rcnt > 1)
+			receive = 1;
+		else
+			receive = 0;
+	} else if (test_bit(FLG_TRANSPARENT, &bch->Flags))
+		receive = hfcpci_empty_fifo_trans(bch, bz, bdata);
+	else
+		receive = 0;
+	if (count && receive)
+		goto Begin;
+
+}
+
+/*
+ * D-channel send routine
+ */
+static void
+hfcpci_fill_dfifo(struct hfc_pci *hc)
+{
+	struct dchannel	*dch = &hc->dch;
+	int		fcnt;
+	int		count, new_z1, maxlen;
+	struct dfifo	*df;
+	u_char		*src, *dst, new_f1;
+
+	if ((dch->debug & DEBUG_HW_DCHANNEL) && !(dch->debug & DEBUG_HW_DFIFO))
+		printk(KERN_DEBUG "%s\n", __func__);
+
+	if (!dch->tx_skb)
+		return;
+	count = dch->tx_skb->len - dch->tx_idx;
+	if (count <= 0)
+		return;
+	df = &((union fifo_area *) (hc->hw.fifos))->d_chan.d_tx;
+
+	if (dch->debug & DEBUG_HW_DFIFO)
+		printk(KERN_DEBUG "%s:f1(%d) f2(%d) z1(f1)(%x)\n", __func__,
+		    df->f1, df->f2,
+		    le16_to_cpu(df->za[df->f1 & D_FREG_MASK].z1));
+	fcnt = df->f1 - df->f2;	/* frame count actually buffered */
+	if (fcnt < 0)
+		fcnt += (MAX_D_FRAMES + 1);	/* if wrap around */
+	if (fcnt > (MAX_D_FRAMES - 1)) {
+		if (dch->debug & DEBUG_HW_DCHANNEL)
+			printk(KERN_DEBUG
+			    "hfcpci_fill_Dfifo more as 14 frames\n");
+#ifdef ERROR_STATISTIC
+		cs->err_tx++;
+#endif
+		return;
+	}
+	/* now determine free bytes in FIFO buffer */
+	maxlen = le16_to_cpu(df->za[df->f2 & D_FREG_MASK].z2) -
+	    le16_to_cpu(df->za[df->f1 & D_FREG_MASK].z1) - 1;
+	if (maxlen <= 0)
+		maxlen += D_FIFO_SIZE;	/* count now contains available bytes */
+
+	if (dch->debug & DEBUG_HW_DCHANNEL)
+		printk(KERN_DEBUG "hfcpci_fill_Dfifo count(%d/%d)\n",
+			count, maxlen);
+	if (count > maxlen) {
+		if (dch->debug & DEBUG_HW_DCHANNEL)
+			printk(KERN_DEBUG "hfcpci_fill_Dfifo no fifo mem\n");
+		return;
+	}
+	new_z1 = (le16_to_cpu(df->za[df->f1 & D_FREG_MASK].z1) + count) &
+	    (D_FIFO_SIZE - 1);
+	new_f1 = ((df->f1 + 1) & D_FREG_MASK) | (D_FREG_MASK + 1);
+	src = dch->tx_skb->data + dch->tx_idx;	/* source pointer */
+	dst = df->data + le16_to_cpu(df->za[df->f1 & D_FREG_MASK].z1);
+	maxlen = D_FIFO_SIZE - le16_to_cpu(df->za[df->f1 & D_FREG_MASK].z1);
+	    /* end fifo */
+	if (maxlen > count)
+		maxlen = count;	/* limit size */
+	memcpy(dst, src, maxlen);	/* first copy */
+
+	count -= maxlen;	/* remaining bytes */
+	if (count) {
+		dst = df->data;	/* start of buffer */
+		src += maxlen;	/* new position */
+		memcpy(dst, src, count);
+	}
+	df->za[new_f1 & D_FREG_MASK].z1 = cpu_to_le16(new_z1);
+	    /* for next buffer */
+	df->za[df->f1 & D_FREG_MASK].z1 = cpu_to_le16(new_z1);
+	    /* new pos actual buffer */
+	df->f1 = new_f1;	/* next frame */
+	dch->tx_idx = dch->tx_skb->len;
+}
+
+/*
+ * B-channel send routine
+ */
+static void
+hfcpci_fill_fifo(struct bchannel *bch)
+{
+	struct hfc_pci 	*hc = bch->hw;
+	int		maxlen, fcnt;
+	int		count, new_z1;
+	struct bzfifo	*bz;
+	u_char		*bdata;
+	u_char		new_f1, *src, *dst;
+	unsigned short	*z1t, *z2t;
+
+	if ((bch->debug & DEBUG_HW_BCHANNEL) && !(bch->debug & DEBUG_HW_BFIFO))
+		printk(KERN_DEBUG "%s\n", __func__);
+	if ((!bch->tx_skb) || bch->tx_skb->len <= 0)
+		return;
+	count = bch->tx_skb->len - bch->tx_idx;
+	if ((bch->nr & 2) && (!hc->hw.bswapped)) {
+		bz = &((union fifo_area *)(hc->hw.fifos))->b_chans.txbz_b2;
+		bdata = ((union fifo_area *)(hc->hw.fifos))->b_chans.txdat_b2;
+	} else {
+		bz = &((union fifo_area *)(hc->hw.fifos))->b_chans.txbz_b1;
+		bdata = ((union fifo_area *)(hc->hw.fifos))->b_chans.txdat_b1;
+	}
+
+	if (test_bit(FLG_TRANSPARENT, &bch->Flags)) {
+		z1t = &bz->za[MAX_B_FRAMES].z1;
+		z2t = z1t + 1;
+		if (bch->debug & DEBUG_HW_BCHANNEL)
+			printk(KERN_DEBUG "hfcpci_fill_fifo_trans ch(%x) "
+			    "cnt(%d) z1(%x) z2(%x)\n", bch->nr, count,
+			    le16_to_cpu(*z1t), le16_to_cpu(*z2t));
+		fcnt = le16_to_cpu(*z2t) - le16_to_cpu(*z1t);
+		if (fcnt <= 0)
+			fcnt += B_FIFO_SIZE;
+			    /* fcnt contains available bytes in fifo */
+		fcnt = B_FIFO_SIZE - fcnt;
+		    /* remaining bytes to send (bytes in fifo) */
+next_t_frame:
+		count = bch->tx_skb->len - bch->tx_idx;
+		/* maximum fill shall be HFCPCI_BTRANS_MAX */
+		if (count > HFCPCI_BTRANS_MAX - fcnt)
+			count = HFCPCI_BTRANS_MAX - fcnt;
+		if (count <= 0)
+			return;
+		/* data is suitable for fifo */
+		new_z1 = le16_to_cpu(*z1t) + count;
+		    /* new buffer Position */
+		if (new_z1 >= (B_FIFO_SIZE + B_SUB_VAL))
+			new_z1 -= B_FIFO_SIZE;	/* buffer wrap */
+		src = bch->tx_skb->data + bch->tx_idx;
+		    /* source pointer */
+		dst = bdata + (le16_to_cpu(*z1t) - B_SUB_VAL);
+		maxlen = (B_FIFO_SIZE + B_SUB_VAL) - le16_to_cpu(*z1t);
+		    /* end of fifo */
+		if (bch->debug & DEBUG_HW_BFIFO)
+			printk(KERN_DEBUG "hfcpci_FFt fcnt(%d) "
+			    "maxl(%d) nz1(%x) dst(%p)\n",
+			    fcnt, maxlen, new_z1, dst);
+		fcnt += count;
+		bch->tx_idx += count;
+		if (maxlen > count)
+			maxlen = count;		/* limit size */
+		memcpy(dst, src, maxlen);	/* first copy */
+		count -= maxlen;	/* remaining bytes */
+		if (count) {
+			dst = bdata;	/* start of buffer */
+			src += maxlen;	/* new position */
+			memcpy(dst, src, count);
+		}
+		*z1t = cpu_to_le16(new_z1);	/* now send data */
+		if (bch->tx_idx < bch->tx_skb->len)
+			return;
+		/* send confirm, on trans, free on hdlc. */
+		if (test_bit(FLG_TRANSPARENT, &bch->Flags))
+			confirm_Bsend(bch);
+		dev_kfree_skb(bch->tx_skb);
+		if (get_next_bframe(bch))
+			goto next_t_frame;
+		return;
+	}
+	if (bch->debug & DEBUG_HW_BCHANNEL)
+		printk(KERN_DEBUG
+		    "%s: ch(%x) f1(%d) f2(%d) z1(f1)(%x)\n",
+		    __func__, bch->nr, bz->f1, bz->f2,
+		    bz->za[bz->f1].z1);
+	fcnt = bz->f1 - bz->f2;	/* frame count actually buffered */
+	if (fcnt < 0)
+		fcnt += (MAX_B_FRAMES + 1);	/* if wrap around */
+	if (fcnt > (MAX_B_FRAMES - 1)) {
+		if (bch->debug & DEBUG_HW_BCHANNEL)
+			printk(KERN_DEBUG
+			    "hfcpci_fill_Bfifo more as 14 frames\n");
+		return;
+	}
+	/* now determine free bytes in FIFO buffer */
+	maxlen = le16_to_cpu(bz->za[bz->f2].z2) -
+	    le16_to_cpu(bz->za[bz->f1].z1) - 1;
+	if (maxlen <= 0)
+		maxlen += B_FIFO_SIZE;	/* count now contains available bytes */
+
+	if (bch->debug & DEBUG_HW_BCHANNEL)
+		printk(KERN_DEBUG "hfcpci_fill_fifo ch(%x) count(%d/%d)\n",
+			bch->nr, count, maxlen);
+
+	if (maxlen < count) {
+		if (bch->debug & DEBUG_HW_BCHANNEL)
+			printk(KERN_DEBUG "hfcpci_fill_fifo no fifo mem\n");
+		return;
+	}
+	new_z1 = le16_to_cpu(bz->za[bz->f1].z1) + count;
+	    /* new buffer Position */
+	if (new_z1 >= (B_FIFO_SIZE + B_SUB_VAL))
+		new_z1 -= B_FIFO_SIZE;	/* buffer wrap */
+
+	new_f1 = ((bz->f1 + 1) & MAX_B_FRAMES);
+	src = bch->tx_skb->data + bch->tx_idx;	/* source pointer */
+	dst = bdata + (le16_to_cpu(bz->za[bz->f1].z1) - B_SUB_VAL);
+	maxlen = (B_FIFO_SIZE + B_SUB_VAL) - le16_to_cpu(bz->za[bz->f1].z1);
+	    /* end fifo */
+	if (maxlen > count)
+		maxlen = count;	/* limit size */
+	memcpy(dst, src, maxlen);	/* first copy */
+
+	count -= maxlen;	/* remaining bytes */
+	if (count) {
+		dst = bdata;	/* start of buffer */
+		src += maxlen;	/* new position */
+		memcpy(dst, src, count);
+	}
+	bz->za[new_f1].z1 = cpu_to_le16(new_z1);	/* for next buffer */
+	bz->f1 = new_f1;	/* next frame */
+	dev_kfree_skb(bch->tx_skb);
+	get_next_bframe(bch);
+}
+
+
+
+/*
+ * handle L1 state changes TE
+ */
+
+static void
+ph_state_te(struct dchannel *dch)
+{
+	if (dch->debug)
+		printk(KERN_DEBUG "%s: TE newstate %x\n",
+			__func__, dch->state);
+	switch (dch->state) {
+	case 0:
+		l1_event(dch->l1, HW_RESET_IND);
+		break;
+	case 3:
+		l1_event(dch->l1, HW_DEACT_IND);
+		break;
+	case 5:
+	case 8:
+		l1_event(dch->l1, ANYSIGNAL);
+		break;
+	case 6:
+		l1_event(dch->l1, INFO2);
+		break;
+	case 7:
+		l1_event(dch->l1, INFO4_P8);
+		break;
+	}
+}
+
+/*
+ * handle L1 state changes NT
+ */
+
+static void
+handle_nt_timer3(struct dchannel *dch) {
+	struct hfc_pci	*hc = dch->hw;
+
+	test_and_clear_bit(FLG_HFC_TIMER_T3, &dch->Flags);
+	hc->hw.int_m1 &= ~HFCPCI_INTS_TIMER;
+	Write_hfc(hc, HFCPCI_INT_M1, hc->hw.int_m1);
+	hc->hw.nt_timer = 0;
+	test_and_set_bit(FLG_ACTIVE, &dch->Flags);
+	if (test_bit(HFC_CFG_MASTER, &hc->cfg))
+		hc->hw.mst_m |= HFCPCI_MASTER;
+	Write_hfc(hc, HFCPCI_MST_MODE, hc->hw.mst_m);
+	_queue_data(&dch->dev.D, PH_ACTIVATE_IND,
+	    MISDN_ID_ANY, 0, NULL, GFP_ATOMIC);
+}
+
+static void
+ph_state_nt(struct dchannel *dch)
+{
+	struct hfc_pci	*hc = dch->hw;
+
+	if (dch->debug)
+		printk(KERN_DEBUG "%s: NT newstate %x\n",
+			__func__, dch->state);
+	switch (dch->state) {
+	case 2:
+		if (hc->hw.nt_timer < 0) {
+			hc->hw.nt_timer = 0;
+			test_and_clear_bit(FLG_HFC_TIMER_T3, &dch->Flags);
+			test_and_clear_bit(FLG_HFC_TIMER_T1, &dch->Flags);
+			hc->hw.int_m1 &= ~HFCPCI_INTS_TIMER;
+			Write_hfc(hc, HFCPCI_INT_M1, hc->hw.int_m1);
+			/* Clear already pending ints */
+			if (Read_hfc(hc, HFCPCI_INT_S1));
+			Write_hfc(hc, HFCPCI_STATES, 4 | HFCPCI_LOAD_STATE);
+			udelay(10);
+			Write_hfc(hc, HFCPCI_STATES, 4);
+			dch->state = 4;
+		} else if (hc->hw.nt_timer == 0) {
+			hc->hw.int_m1 |= HFCPCI_INTS_TIMER;
+			Write_hfc(hc, HFCPCI_INT_M1, hc->hw.int_m1);
+			hc->hw.nt_timer = NT_T1_COUNT;
+			hc->hw.ctmt &= ~HFCPCI_AUTO_TIMER;
+			hc->hw.ctmt |= HFCPCI_TIM3_125;
+			Write_hfc(hc, HFCPCI_CTMT, hc->hw.ctmt |
+				HFCPCI_CLTIMER);
+			test_and_clear_bit(FLG_HFC_TIMER_T3, &dch->Flags);
+			test_and_set_bit(FLG_HFC_TIMER_T1, &dch->Flags);
+			/* allow G2 -> G3 transition */
+			Write_hfc(hc, HFCPCI_STATES, 2 | HFCPCI_NT_G2_G3);
+		} else {
+			Write_hfc(hc, HFCPCI_STATES, 2 | HFCPCI_NT_G2_G3);
+		}
+		break;
+	case 1:
+		hc->hw.nt_timer = 0;
+		test_and_clear_bit(FLG_HFC_TIMER_T3, &dch->Flags);
+		test_and_clear_bit(FLG_HFC_TIMER_T1, &dch->Flags);
+		hc->hw.int_m1 &= ~HFCPCI_INTS_TIMER;
+		Write_hfc(hc, HFCPCI_INT_M1, hc->hw.int_m1);
+		test_and_clear_bit(FLG_ACTIVE, &dch->Flags);
+		hc->hw.mst_m &= ~HFCPCI_MASTER;
+		Write_hfc(hc, HFCPCI_MST_MODE, hc->hw.mst_m);
+		test_and_clear_bit(FLG_L2_ACTIVATED, &dch->Flags);
+		_queue_data(&dch->dev.D, PH_DEACTIVATE_IND,
+		    MISDN_ID_ANY, 0, NULL, GFP_ATOMIC);
+		break;
+	case 4:
+		hc->hw.nt_timer = 0;
+		test_and_clear_bit(FLG_HFC_TIMER_T3, &dch->Flags);
+		test_and_clear_bit(FLG_HFC_TIMER_T1, &dch->Flags);
+		hc->hw.int_m1 &= ~HFCPCI_INTS_TIMER;
+		Write_hfc(hc, HFCPCI_INT_M1, hc->hw.int_m1);
+		break;
+	case 3:
+		if (!test_and_set_bit(FLG_HFC_TIMER_T3, &dch->Flags)) {
+			if (!test_and_clear_bit(FLG_L2_ACTIVATED,
+			    &dch->Flags)) {
+				handle_nt_timer3(dch);
+				break;
+			}
+			test_and_clear_bit(FLG_HFC_TIMER_T1, &dch->Flags);
+			hc->hw.int_m1 |= HFCPCI_INTS_TIMER;
+			Write_hfc(hc, HFCPCI_INT_M1, hc->hw.int_m1);
+			hc->hw.nt_timer = NT_T3_COUNT;
+			hc->hw.ctmt &= ~HFCPCI_AUTO_TIMER;
+			hc->hw.ctmt |= HFCPCI_TIM3_125;
+			Write_hfc(hc, HFCPCI_CTMT, hc->hw.ctmt |
+				HFCPCI_CLTIMER);
+		}
+		break;
+	}
+}
+
+static void
+ph_state(struct dchannel *dch)
+{
+	struct hfc_pci	*hc = dch->hw;
+
+	if (hc->hw.protocol == ISDN_P_NT_S0) {
+		if (test_bit(FLG_HFC_TIMER_T3, &dch->Flags) &&
+		    hc->hw.nt_timer < 0)
+			handle_nt_timer3(dch);
+		else
+			ph_state_nt(dch);
+	} else
+		ph_state_te(dch);
+}
+
+/*
+ * Layer 1 callback function
+ */
+static int
+hfc_l1callback(struct dchannel *dch, u_int cmd)
+{
+	struct hfc_pci		*hc = dch->hw;
+
+	switch (cmd) {
+	case INFO3_P8:
+	case INFO3_P10:
+		if (test_bit(HFC_CFG_MASTER, &hc->cfg))
+			hc->hw.mst_m |= HFCPCI_MASTER;
+		Write_hfc(hc, HFCPCI_MST_MODE, hc->hw.mst_m);
+		break;
+	case HW_RESET_REQ:
+		Write_hfc(hc, HFCPCI_STATES, HFCPCI_LOAD_STATE | 3);
+		/* HFC ST 3 */
+		udelay(6);
+		Write_hfc(hc, HFCPCI_STATES, 3);	/* HFC ST 2 */
+		if (test_bit(HFC_CFG_MASTER, &hc->cfg))
+			hc->hw.mst_m |= HFCPCI_MASTER;
+		Write_hfc(hc, HFCPCI_MST_MODE, hc->hw.mst_m);
+		Write_hfc(hc, HFCPCI_STATES, HFCPCI_ACTIVATE |
+		   HFCPCI_DO_ACTION);
+		l1_event(dch->l1, HW_POWERUP_IND);
+		break;
+	case HW_DEACT_REQ:
+		hc->hw.mst_m &= ~HFCPCI_MASTER;
+		Write_hfc(hc, HFCPCI_MST_MODE, hc->hw.mst_m);
+		skb_queue_purge(&dch->squeue);
+		if (dch->tx_skb) {
+			dev_kfree_skb(dch->tx_skb);
+			dch->tx_skb = NULL;
+		}
+		dch->tx_idx = 0;
+		if (dch->rx_skb) {
+			dev_kfree_skb(dch->rx_skb);
+			dch->rx_skb = NULL;
+		}
+		test_and_clear_bit(FLG_TX_BUSY, &dch->Flags);
+		if (test_and_clear_bit(FLG_BUSY_TIMER, &dch->Flags))
+			del_timer(&dch->timer);
+		break;
+	case HW_POWERUP_REQ:
+		Write_hfc(hc, HFCPCI_STATES, HFCPCI_DO_ACTION);
+		break;
+	case PH_ACTIVATE_IND:
+		test_and_set_bit(FLG_ACTIVE, &dch->Flags);
+		_queue_data(&dch->dev.D, cmd, MISDN_ID_ANY, 0, NULL,
+			GFP_ATOMIC);
+		break;
+	case PH_DEACTIVATE_IND:
+		test_and_clear_bit(FLG_ACTIVE, &dch->Flags);
+		_queue_data(&dch->dev.D, cmd, MISDN_ID_ANY, 0, NULL,
+			GFP_ATOMIC);
+		break;
+	default:
+		if (dch->debug & DEBUG_HW)
+			printk(KERN_DEBUG "%s: unknown command %x\n",
+			    __func__, cmd);
+		return -1;
+	}
+	return 0;
+}
+
+/*
+ * Interrupt handler
+ */
+static inline void
+tx_birq(struct bchannel *bch)
+{
+	if (bch->tx_skb && bch->tx_idx < bch->tx_skb->len)
+		hfcpci_fill_fifo(bch);
+	else {
+		if (bch->tx_skb)
+			dev_kfree_skb(bch->tx_skb);
+		if (get_next_bframe(bch))
+			hfcpci_fill_fifo(bch);
+	}
+}
+
+static inline void
+tx_dirq(struct dchannel *dch)
+{
+	if (dch->tx_skb && dch->tx_idx < dch->tx_skb->len)
+		hfcpci_fill_dfifo(dch->hw);
+	else {
+		if (dch->tx_skb)
+			dev_kfree_skb(dch->tx_skb);
+		if (get_next_dframe(dch))
+			hfcpci_fill_dfifo(dch->hw);
+	}
+}
+
+static irqreturn_t
+hfcpci_int(int intno, void *dev_id)
+{
+	struct hfc_pci	*hc = dev_id;
+	u_char		exval;
+	struct bchannel	*bch;
+	u_char		val, stat;
+
+	spin_lock(&hc->lock);
+	if (!(hc->hw.int_m2 & 0x08)) {
+		spin_unlock(&hc->lock);
+		return IRQ_NONE; /* not initialised */
+	}
+	stat = Read_hfc(hc, HFCPCI_STATUS);
+	if (HFCPCI_ANYINT & stat) {
+		val = Read_hfc(hc, HFCPCI_INT_S1);
+		if (hc->dch.debug & DEBUG_HW_DCHANNEL)
+			printk(KERN_DEBUG
+			    "HFC-PCI: stat(%02x) s1(%02x)\n", stat, val);
+	} else {
+		/* shared */
+		spin_unlock(&hc->lock);
+		return IRQ_NONE;
+	}
+	hc->irqcnt++;
+
+	if (hc->dch.debug & DEBUG_HW_DCHANNEL)
+		printk(KERN_DEBUG "HFC-PCI irq %x\n", val);
+	val &= hc->hw.int_m1;
+	if (val & 0x40) {	/* state machine irq */
+		exval = Read_hfc(hc, HFCPCI_STATES) & 0xf;
+		if (hc->dch.debug & DEBUG_HW_DCHANNEL)
+			printk(KERN_DEBUG "ph_state chg %d->%d\n",
+				hc->dch.state, exval);
+		hc->dch.state = exval;
+		schedule_event(&hc->dch, FLG_PHCHANGE);
+		val &= ~0x40;
+	}
+	if (val & 0x80) {	/* timer irq */
+		if (hc->hw.protocol == ISDN_P_NT_S0) {
+			if ((--hc->hw.nt_timer) < 0)
+				schedule_event(&hc->dch, FLG_PHCHANGE);
+		}
+		val &= ~0x80;
+		Write_hfc(hc, HFCPCI_CTMT, hc->hw.ctmt | HFCPCI_CLTIMER);
+	}
+	if (val & 0x08) {
+		bch = Sel_BCS(hc, hc->hw.bswapped ? 2 : 1);
+		if (bch)
+			main_rec_hfcpci(bch);
+		else if (hc->dch.debug)
+			printk(KERN_DEBUG "hfcpci spurious 0x08 IRQ\n");
+	}
+	if (val & 0x10) {
+		bch = Sel_BCS(hc, 2);
+		if (bch)
+			main_rec_hfcpci(bch);
+		else if (hc->dch.debug)
+			printk(KERN_DEBUG "hfcpci spurious 0x10 IRQ\n");
+	}
+	if (val & 0x01) {
+		bch = Sel_BCS(hc, hc->hw.bswapped ? 2 : 1);
+		if (bch)
+			tx_birq(bch);
+		else if (hc->dch.debug)
+			printk(KERN_DEBUG "hfcpci spurious 0x01 IRQ\n");
+	}
+	if (val & 0x02) {
+		bch = Sel_BCS(hc, 2);
+		if (bch)
+			tx_birq(bch);
+		else if (hc->dch.debug)
+			printk(KERN_DEBUG "hfcpci spurious 0x02 IRQ\n");
+	}
+	if (val & 0x20)
+		receive_dmsg(hc);
+	if (val & 0x04) {	/* dframe transmitted */
+		if (test_and_clear_bit(FLG_BUSY_TIMER, &hc->dch.Flags))
+			del_timer(&hc->dch.timer);
+		tx_dirq(&hc->dch);
+	}
+	spin_unlock(&hc->lock);
+	return IRQ_HANDLED;
+}
+
+/*
+ * timer callback for D-chan busy resolution. Currently no function
+ */
+static void
+hfcpci_dbusy_timer(struct hfc_pci *hc)
+{
+}
+
+/*
+ * activate/deactivate hardware for selected channels and mode
+ */
+static int
+mode_hfcpci(struct bchannel *bch, int bc, int protocol)
+{
+	struct hfc_pci	*hc = bch->hw;
+	int		fifo2;
+	u_char		rx_slot = 0, tx_slot = 0, pcm_mode;
+
+	if (bch->debug & DEBUG_HW_BCHANNEL)
+		printk(KERN_DEBUG
+		    "HFCPCI bchannel protocol %x-->%x ch %x-->%x\n",
+		    bch->state, protocol, bch->nr, bc);
+
+	fifo2 = bc;
+	pcm_mode = (bc>>24) & 0xff;
+	if (pcm_mode) { /* PCM SLOT USE */
+		if (!test_bit(HFC_CFG_PCM, &hc->cfg))
+			printk(KERN_WARNING
+			    "%s: pcm channel id without HFC_CFG_PCM\n",
+			    __func__);
+		rx_slot = (bc>>8) & 0xff;
+		tx_slot = (bc>>16) & 0xff;
+		bc = bc & 0xff;
+	} else if (test_bit(HFC_CFG_PCM, &hc->cfg) &&
+	    (protocol > ISDN_P_NONE))
+		printk(KERN_WARNING "%s: no pcm channel id but HFC_CFG_PCM\n",
+		    __func__);
+	if (hc->chanlimit > 1) {
+		hc->hw.bswapped = 0;	/* B1 and B2 normal mode */
+		hc->hw.sctrl_e &= ~0x80;
+	} else {
+		if (bc & 2) {
+			if (protocol != ISDN_P_NONE) {
+				hc->hw.bswapped = 1; /* B1 and B2 exchanged */
+				hc->hw.sctrl_e |= 0x80;
+			} else {
+				hc->hw.bswapped = 0; /* B1 and B2 normal mode */
+				hc->hw.sctrl_e &= ~0x80;
+			}
+			fifo2 = 1;
+		} else {
+			hc->hw.bswapped = 0;	/* B1 and B2 normal mode */
+			hc->hw.sctrl_e &= ~0x80;
+		}
+	}
+	switch (protocol) {
+	case (-1): /* used for init */
+		bch->state = -1;
+		bch->nr = bc;
+	case (ISDN_P_NONE):
+		if (bch->state == ISDN_P_NONE)
+			return 0;
+		if (bc & 2) {
+			hc->hw.sctrl &= ~SCTRL_B2_ENA;
+			hc->hw.sctrl_r &= ~SCTRL_B2_ENA;
+		} else {
+			hc->hw.sctrl &= ~SCTRL_B1_ENA;
+			hc->hw.sctrl_r &= ~SCTRL_B1_ENA;
+		}
+		if (fifo2 & 2) {
+			hc->hw.fifo_en &= ~HFCPCI_FIFOEN_B2;
+			hc->hw.int_m1 &= ~(HFCPCI_INTS_B2TRANS +
+				HFCPCI_INTS_B2REC);
+		} else {
+			hc->hw.fifo_en &= ~HFCPCI_FIFOEN_B1;
+			hc->hw.int_m1 &= ~(HFCPCI_INTS_B1TRANS +
+				HFCPCI_INTS_B1REC);
+		}
+#ifdef REVERSE_BITORDER
+		if (bch->nr & 2)
+			hc->hw.cirm &= 0x7f;
+		else
+			hc->hw.cirm &= 0xbf;
+#endif
+		bch->state = ISDN_P_NONE;
+		bch->nr = bc;
+		test_and_clear_bit(FLG_HDLC, &bch->Flags);
+		test_and_clear_bit(FLG_TRANSPARENT, &bch->Flags);
+		break;
+	case (ISDN_P_B_RAW):
+		bch->state = protocol;
+		bch->nr = bc;
+		hfcpci_clear_fifo_rx(hc, (fifo2 & 2)?1:0);
+		hfcpci_clear_fifo_tx(hc, (fifo2 & 2)?1:0);
+		if (bc & 2) {
+			hc->hw.sctrl |= SCTRL_B2_ENA;
+			hc->hw.sctrl_r |= SCTRL_B2_ENA;
+#ifdef REVERSE_BITORDER
+			hc->hw.cirm |= 0x80;
+#endif
+		} else {
+			hc->hw.sctrl |= SCTRL_B1_ENA;
+			hc->hw.sctrl_r |= SCTRL_B1_ENA;
+#ifdef REVERSE_BITORDER
+			hc->hw.cirm |= 0x40;
+#endif
+		}
+		if (fifo2 & 2) {
+			hc->hw.fifo_en |= HFCPCI_FIFOEN_B2;
+			hc->hw.int_m1 |= (HFCPCI_INTS_B2TRANS +
+			    HFCPCI_INTS_B2REC);
+			hc->hw.ctmt |= 2;
+			hc->hw.conn &= ~0x18;
+		} else {
+			hc->hw.fifo_en |= HFCPCI_FIFOEN_B1;
+			hc->hw.int_m1 |= (HFCPCI_INTS_B1TRANS +
+			    HFCPCI_INTS_B1REC);
+			hc->hw.ctmt |= 1;
+			hc->hw.conn &= ~0x03;
+		}
+		test_and_set_bit(FLG_TRANSPARENT, &bch->Flags);
+		break;
+	case (ISDN_P_B_HDLC):
+		bch->state = protocol;
+		bch->nr = bc;
+		hfcpci_clear_fifo_rx(hc, (fifo2 & 2)?1:0);
+		hfcpci_clear_fifo_tx(hc, (fifo2 & 2)?1:0);
+		if (bc & 2) {
+			hc->hw.sctrl |= SCTRL_B2_ENA;
+			hc->hw.sctrl_r |= SCTRL_B2_ENA;
+		} else {
+			hc->hw.sctrl |= SCTRL_B1_ENA;
+			hc->hw.sctrl_r |= SCTRL_B1_ENA;
+		}
+		if (fifo2 & 2) {
+			hc->hw.last_bfifo_cnt[1] = 0;
+			hc->hw.fifo_en |= HFCPCI_FIFOEN_B2;
+			hc->hw.int_m1 |= (HFCPCI_INTS_B2TRANS +
+			    HFCPCI_INTS_B2REC);
+			hc->hw.ctmt &= ~2;
+			hc->hw.conn &= ~0x18;
+		} else {
+			hc->hw.last_bfifo_cnt[0] = 0;
+			hc->hw.fifo_en |= HFCPCI_FIFOEN_B1;
+			hc->hw.int_m1 |= (HFCPCI_INTS_B1TRANS +
+			    HFCPCI_INTS_B1REC);
+			hc->hw.ctmt &= ~1;
+			hc->hw.conn &= ~0x03;
+		}
+		test_and_set_bit(FLG_HDLC, &bch->Flags);
+		break;
+	default:
+		printk(KERN_DEBUG "prot not known %x\n", protocol);
+		return -ENOPROTOOPT;
+	}
+	if (test_bit(HFC_CFG_PCM, &hc->cfg)) {
+		if ((protocol == ISDN_P_NONE) ||
+			(protocol == -1)) {	/* init case */
+			rx_slot = 0;
+			tx_slot = 0;
+		} else {
+			if (test_bit(HFC_CFG_SW_DD_DU, &hc->cfg)) {
+				rx_slot |= 0xC0;
+				tx_slot |= 0xC0;
+			} else {
+				rx_slot |= 0x80;
+				tx_slot |= 0x80;
+			}
+		}
+		if (bc & 2) {
+			hc->hw.conn &= 0xc7;
+			hc->hw.conn |= 0x08;
+			printk(KERN_DEBUG "%s: Write_hfc: B2_SSL 0x%x\n",
+				__func__, tx_slot);
+			printk(KERN_DEBUG "%s: Write_hfc: B2_RSL 0x%x\n",
+				__func__, rx_slot);
+			Write_hfc(hc, HFCPCI_B2_SSL, tx_slot);
+			Write_hfc(hc, HFCPCI_B2_RSL, rx_slot);
+		} else {
+			hc->hw.conn &= 0xf8;
+			hc->hw.conn |= 0x01;
+			printk(KERN_DEBUG "%s: Write_hfc: B1_SSL 0x%x\n",
+				__func__, tx_slot);
+			printk(KERN_DEBUG "%s: Write_hfc: B1_RSL 0x%x\n",
+				__func__, rx_slot);
+			Write_hfc(hc, HFCPCI_B1_SSL, tx_slot);
+			Write_hfc(hc, HFCPCI_B1_RSL, rx_slot);
+		}
+	}
+	Write_hfc(hc, HFCPCI_SCTRL_E, hc->hw.sctrl_e);
+	Write_hfc(hc, HFCPCI_INT_M1, hc->hw.int_m1);
+	Write_hfc(hc, HFCPCI_FIFO_EN, hc->hw.fifo_en);
+	Write_hfc(hc, HFCPCI_SCTRL, hc->hw.sctrl);
+	Write_hfc(hc, HFCPCI_SCTRL_R, hc->hw.sctrl_r);
+	Write_hfc(hc, HFCPCI_CTMT, hc->hw.ctmt);
+	Write_hfc(hc, HFCPCI_CONNECT, hc->hw.conn);
+#ifdef REVERSE_BITORDER
+	Write_hfc(hc, HFCPCI_CIRM, hc->hw.cirm);
+#endif
+	return 0;
+}
+
+static int
+set_hfcpci_rxtest(struct bchannel *bch, int protocol, int chan)
+{
+	struct hfc_pci	*hc = bch->hw;
+
+	if (bch->debug & DEBUG_HW_BCHANNEL)
+		printk(KERN_DEBUG
+		    "HFCPCI bchannel test rx protocol %x-->%x ch %x-->%x\n",
+		    bch->state, protocol, bch->nr, chan);
+	if (bch->nr != chan) {
+		printk(KERN_DEBUG
+		    "HFCPCI rxtest wrong channel parameter %x/%x\n",
+		    bch->nr, chan);
+		return -EINVAL;
+	}
+	switch (protocol) {
+	case (ISDN_P_B_RAW):
+		bch->state = protocol;
+		hfcpci_clear_fifo_rx(hc, (chan & 2)?1:0);
+		if (chan & 2) {
+			hc->hw.sctrl_r |= SCTRL_B2_ENA;
+			hc->hw.fifo_en |= HFCPCI_FIFOEN_B2RX;
+			hc->hw.int_m1 |= HFCPCI_INTS_B2REC;
+			hc->hw.ctmt |= 2;
+			hc->hw.conn &= ~0x18;
+#ifdef REVERSE_BITORDER
+			hc->hw.cirm |= 0x80;
+#endif
+		} else {
+			hc->hw.sctrl_r |= SCTRL_B1_ENA;
+			hc->hw.fifo_en |= HFCPCI_FIFOEN_B1RX;
+			hc->hw.int_m1 |= HFCPCI_INTS_B1REC;
+			hc->hw.ctmt |= 1;
+			hc->hw.conn &= ~0x03;
+#ifdef REVERSE_BITORDER
+			hc->hw.cirm |= 0x40;
+#endif
+		}
+		break;
+	case (ISDN_P_B_HDLC):
+		bch->state = protocol;
+		hfcpci_clear_fifo_rx(hc, (chan & 2)?1:0);
+		if (chan & 2) {
+			hc->hw.sctrl_r |= SCTRL_B2_ENA;
+			hc->hw.last_bfifo_cnt[1] = 0;
+			hc->hw.fifo_en |= HFCPCI_FIFOEN_B2RX;
+			hc->hw.int_m1 |= HFCPCI_INTS_B2REC;
+			hc->hw.ctmt &= ~2;
+			hc->hw.conn &= ~0x18;
+		} else {
+			hc->hw.sctrl_r |= SCTRL_B1_ENA;
+			hc->hw.last_bfifo_cnt[0] = 0;
+			hc->hw.fifo_en |= HFCPCI_FIFOEN_B1RX;
+			hc->hw.int_m1 |= HFCPCI_INTS_B1REC;
+			hc->hw.ctmt &= ~1;
+			hc->hw.conn &= ~0x03;
+		}
+		break;
+	default:
+		printk(KERN_DEBUG "prot not known %x\n", protocol);
+		return -ENOPROTOOPT;
+	}
+	Write_hfc(hc, HFCPCI_INT_M1, hc->hw.int_m1);
+	Write_hfc(hc, HFCPCI_FIFO_EN, hc->hw.fifo_en);
+	Write_hfc(hc, HFCPCI_SCTRL_R, hc->hw.sctrl_r);
+	Write_hfc(hc, HFCPCI_CTMT, hc->hw.ctmt);
+	Write_hfc(hc, HFCPCI_CONNECT, hc->hw.conn);
+#ifdef REVERSE_BITORDER
+	Write_hfc(hc, HFCPCI_CIRM, hc->hw.cirm);
+#endif
+	return 0;
+}
+
+static void
+deactivate_bchannel(struct bchannel *bch)
+{
+	struct hfc_pci	*hc = bch->hw;
+	u_long		flags;
+
+	spin_lock_irqsave(&hc->lock, flags);
+	if (test_and_clear_bit(FLG_TX_NEXT, &bch->Flags)) {
+		dev_kfree_skb(bch->next_skb);
+		bch->next_skb = NULL;
+	}
+	if (bch->tx_skb) {
+		dev_kfree_skb(bch->tx_skb);
+		bch->tx_skb = NULL;
+	}
+	bch->tx_idx = 0;
+	if (bch->rx_skb) {
+		dev_kfree_skb(bch->rx_skb);
+		bch->rx_skb = NULL;
+	}
+	mode_hfcpci(bch, bch->nr, ISDN_P_NONE);
+	test_and_clear_bit(FLG_ACTIVE, &bch->Flags);
+	test_and_clear_bit(FLG_TX_BUSY, &bch->Flags);
+	spin_unlock_irqrestore(&hc->lock, flags);
+}
+
+/*
+ * Layer 1 B-channel hardware access
+ */
+static int
+channel_bctrl(struct bchannel *bch, struct mISDN_ctrl_req *cq)
+{
+	int			ret = 0;
+
+	switch (cq->op) {
+	case MISDN_CTRL_GETOP:
+		cq->op = 0;
+		break;
+	default:
+		printk(KERN_WARNING "%s: unknown Op %x\n", __func__, cq->op);
+		ret = -EINVAL;
+		break;
+	}
+	return ret;
+}
+static int
+hfc_bctrl(struct mISDNchannel *ch, u_int cmd, void *arg)
+{
+	struct bchannel	*bch = container_of(ch, struct bchannel, ch);
+	struct hfc_pci	*hc = bch->hw;
+	int		ret = -EINVAL;
+	u_long		flags;
+
+	if (bch->debug & DEBUG_HW)
+		printk(KERN_DEBUG "%s: cmd:%x %p\n", __func__, cmd, arg);
+	switch (cmd) {
+	case HW_TESTRX_RAW:
+		spin_lock_irqsave(&hc->lock, flags);
+		ret = set_hfcpci_rxtest(bch, ISDN_P_B_RAW, (int)(long)arg);
+		spin_unlock_irqrestore(&hc->lock, flags);
+		break;
+	case HW_TESTRX_HDLC:
+		spin_lock_irqsave(&hc->lock, flags);
+		ret = set_hfcpci_rxtest(bch, ISDN_P_B_HDLC, (int)(long)arg);
+		spin_unlock_irqrestore(&hc->lock, flags);
+		break;
+	case HW_TESTRX_OFF:
+		spin_lock_irqsave(&hc->lock, flags);
+		mode_hfcpci(bch, bch->nr, ISDN_P_NONE);
+		spin_unlock_irqrestore(&hc->lock, flags);
+		ret = 0;
+		break;
+	case CLOSE_CHANNEL:
+		test_and_clear_bit(FLG_OPEN, &bch->Flags);
+		if (test_bit(FLG_ACTIVE, &bch->Flags))
+			deactivate_bchannel(bch);
+		ch->protocol = ISDN_P_NONE;
+		ch->peer = NULL;
+		module_put(THIS_MODULE);
+		ret = 0;
+		break;
+	case CONTROL_CHANNEL:
+		ret = channel_bctrl(bch, arg);
+		break;
+	default:
+		printk(KERN_WARNING "%s: unknown prim(%x)\n",
+			__func__, cmd);
+	}
+	return ret;
+}
+
+/*
+ * Layer2 -> Layer 1 Dchannel data
+ */
+static int
+hfcpci_l2l1D(struct mISDNchannel *ch, struct sk_buff *skb)
+{
+	struct mISDNdevice	*dev = container_of(ch, struct mISDNdevice, D);
+	struct dchannel		*dch = container_of(dev, struct dchannel, dev);
+	struct hfc_pci		*hc = dch->hw;
+	int			ret = -EINVAL;
+	struct mISDNhead	*hh = mISDN_HEAD_P(skb);
+	unsigned int		id;
+	u_long			flags;
+
+	switch (hh->prim) {
+	case PH_DATA_REQ:
+		spin_lock_irqsave(&hc->lock, flags);
+		ret = dchannel_senddata(dch, skb);
+		if (ret > 0) { /* direct TX */
+			id = hh->id; /* skb can be freed */
+			hfcpci_fill_dfifo(dch->hw);
+			ret = 0;
+			spin_unlock_irqrestore(&hc->lock, flags);
+			queue_ch_frame(ch, PH_DATA_CNF, id, NULL);
+		} else
+			spin_unlock_irqrestore(&hc->lock, flags);
+		return ret;
+	case PH_ACTIVATE_REQ:
+		spin_lock_irqsave(&hc->lock, flags);
+		if (hc->hw.protocol == ISDN_P_NT_S0) {
+			ret = 0;
+			if (test_bit(HFC_CFG_MASTER, &hc->cfg))
+				hc->hw.mst_m |= HFCPCI_MASTER;
+			Write_hfc(hc, HFCPCI_MST_MODE, hc->hw.mst_m);
+			if (test_bit(FLG_ACTIVE, &dch->Flags)) {
+				spin_unlock_irqrestore(&hc->lock, flags);
+				_queue_data(&dch->dev.D, PH_ACTIVATE_IND,
+				    MISDN_ID_ANY, 0, NULL, GFP_ATOMIC);
+				break;
+			}
+			test_and_set_bit(FLG_L2_ACTIVATED, &dch->Flags);
+			Write_hfc(hc, HFCPCI_STATES, HFCPCI_ACTIVATE |
+			    HFCPCI_DO_ACTION | 1);
+		} else
+			ret = l1_event(dch->l1, hh->prim);
+		spin_unlock_irqrestore(&hc->lock, flags);
+		break;
+	case PH_DEACTIVATE_REQ:
+		test_and_clear_bit(FLG_L2_ACTIVATED, &dch->Flags);
+		spin_lock_irqsave(&hc->lock, flags);
+		if (hc->hw.protocol == ISDN_P_NT_S0) {
+			/* prepare deactivation */
+			Write_hfc(hc, HFCPCI_STATES, 0x40);
+			skb_queue_purge(&dch->squeue);
+			if (dch->tx_skb) {
+				dev_kfree_skb(dch->tx_skb);
+				dch->tx_skb = NULL;
+			}
+			dch->tx_idx = 0;
+			if (dch->rx_skb) {
+				dev_kfree_skb(dch->rx_skb);
+				dch->rx_skb = NULL;
+			}
+			test_and_clear_bit(FLG_TX_BUSY, &dch->Flags);
+			if (test_and_clear_bit(FLG_BUSY_TIMER, &dch->Flags))
+				del_timer(&dch->timer);
+#ifdef FIXME
+			if (test_and_clear_bit(FLG_L1_BUSY, &dch->Flags))
+				dchannel_sched_event(&hc->dch, D_CLEARBUSY);
+#endif
+			hc->hw.mst_m &= ~HFCPCI_MASTER;
+			Write_hfc(hc, HFCPCI_MST_MODE, hc->hw.mst_m);
+			ret = 0;
+		} else {
+			ret = l1_event(dch->l1, hh->prim);
+		}
+		spin_unlock_irqrestore(&hc->lock, flags);
+		break;
+	}
+	if (!ret)
+		dev_kfree_skb(skb);
+	return ret;
+}
+
+/*
+ * Layer2 -> Layer 1 Bchannel data
+ */
+static int
+hfcpci_l2l1B(struct mISDNchannel *ch, struct sk_buff *skb)
+{
+	struct bchannel		*bch = container_of(ch, struct bchannel, ch);
+	struct hfc_pci		*hc = bch->hw;
+	int			ret = -EINVAL;
+	struct mISDNhead	*hh = mISDN_HEAD_P(skb);
+	unsigned int		id;
+	u_long			flags;
+
+	switch (hh->prim) {
+	case PH_DATA_REQ:
+		spin_lock_irqsave(&hc->lock, flags);
+		ret = bchannel_senddata(bch, skb);
+		if (ret > 0) { /* direct TX */
+			id = hh->id; /* skb can be freed */
+			hfcpci_fill_fifo(bch);
+			ret = 0;
+			spin_unlock_irqrestore(&hc->lock, flags);
+			if (!test_bit(FLG_TRANSPARENT, &bch->Flags))
+				queue_ch_frame(ch, PH_DATA_CNF, id, NULL);
+		} else
+			spin_unlock_irqrestore(&hc->lock, flags);
+		return ret;
+	case PH_ACTIVATE_REQ:
+		spin_lock_irqsave(&hc->lock, flags);
+		if (!test_and_set_bit(FLG_ACTIVE, &bch->Flags))
+			ret = mode_hfcpci(bch, bch->nr, ch->protocol);
+		else
+			ret = 0;
+		spin_unlock_irqrestore(&hc->lock, flags);
+		if (!ret)
+			_queue_data(ch, PH_ACTIVATE_IND, MISDN_ID_ANY, 0,
+				NULL, GFP_KERNEL);
+		break;
+	case PH_DEACTIVATE_REQ:
+		deactivate_bchannel(bch);
+		_queue_data(ch, PH_DEACTIVATE_IND, MISDN_ID_ANY, 0,
+			NULL, GFP_KERNEL);
+		ret = 0;
+		break;
+	}
+	if (!ret)
+		dev_kfree_skb(skb);
+	return ret;
+}
+
+/*
+ * called for card init message
+ */
+
+void
+inithfcpci(struct hfc_pci *hc)
+{
+	printk(KERN_DEBUG "inithfcpci: entered\n");
+	hc->dch.timer.function = (void *) hfcpci_dbusy_timer;
+	hc->dch.timer.data = (long) &hc->dch;
+	init_timer(&hc->dch.timer);
+	hc->chanlimit = 2;
+	mode_hfcpci(&hc->bch[0], 1, -1);
+	mode_hfcpci(&hc->bch[1], 2, -1);
+}
+
+
+static int
+init_card(struct hfc_pci *hc)
+{
+	int	cnt = 3;
+	u_long	flags;
+
+	printk(KERN_DEBUG "init_card: entered\n");
+
+
+	spin_lock_irqsave(&hc->lock, flags);
+	disable_hwirq(hc);
+	spin_unlock_irqrestore(&hc->lock, flags);
+	if (request_irq(hc->irq, hfcpci_int, IRQF_SHARED, "HFC PCI", hc)) {
+		printk(KERN_WARNING
+		    "mISDN: couldn't get interrupt %d\n", hc->irq);
+		return -EIO;
+	}
+	spin_lock_irqsave(&hc->lock, flags);
+	reset_hfcpci(hc);
+	while (cnt) {
+		inithfcpci(hc);
+		/*
+		 * Finally enable IRQ output
+		 * this is only allowed, if an IRQ routine is allready
+		 * established for this HFC, so don't do that earlier
+		 */
+		enable_hwirq(hc);
+		spin_unlock_irqrestore(&hc->lock, flags);
+		/* Timeout 80ms */
+		current->state = TASK_UNINTERRUPTIBLE;
+		schedule_timeout((80*HZ)/1000);
+		printk(KERN_INFO "HFC PCI: IRQ %d count %d\n",
+			hc->irq, hc->irqcnt);
+		/* now switch timer interrupt off */
+		spin_lock_irqsave(&hc->lock, flags);
+		hc->hw.int_m1 &= ~HFCPCI_INTS_TIMER;
+		Write_hfc(hc, HFCPCI_INT_M1, hc->hw.int_m1);
+		/* reinit mode reg */
+		Write_hfc(hc, HFCPCI_MST_MODE, hc->hw.mst_m);
+		if (!hc->irqcnt) {
+			printk(KERN_WARNING
+			    "HFC PCI: IRQ(%d) getting no interrupts "
+			    "during init %d\n", hc->irq, 4 - cnt);
+			if (cnt == 1) {
+				spin_unlock_irqrestore(&hc->lock, flags);
+				return -EIO;
+			} else {
+				reset_hfcpci(hc);
+				cnt--;
+			}
+		} else {
+			spin_unlock_irqrestore(&hc->lock, flags);
+			hc->initdone = 1;
+			return 0;
+		}
+	}
+	disable_hwirq(hc);
+	spin_unlock_irqrestore(&hc->lock, flags);
+	free_irq(hc->irq, hc);
+	return -EIO;
+}
+
+static int
+channel_ctrl(struct hfc_pci *hc, struct mISDN_ctrl_req *cq)
+{
+	int	ret = 0;
+	u_char	slot;
+
+	switch (cq->op) {
+	case MISDN_CTRL_GETOP:
+		cq->op = MISDN_CTRL_LOOP | MISDN_CTRL_CONNECT |
+		    MISDN_CTRL_DISCONNECT;
+		break;
+	case MISDN_CTRL_LOOP:
+		/* channel 0 disabled loop */
+		if (cq->channel < 0 || cq->channel > 2) {
+			ret = -EINVAL;
+			break;
+		}
+		if (cq->channel & 1) {
+			if (test_bit(HFC_CFG_SW_DD_DU, &hc->cfg))
+				slot = 0xC0;
+			else
+				slot = 0x80;
+			printk(KERN_DEBUG "%s: Write_hfc: B1_SSL/RSL 0x%x\n",
+			    __func__, slot);
+			Write_hfc(hc, HFCPCI_B1_SSL, slot);
+			Write_hfc(hc, HFCPCI_B1_RSL, slot);
+			hc->hw.conn = (hc->hw.conn & ~7) | 6;
+			Write_hfc(hc, HFCPCI_CONNECT, hc->hw.conn);
+		}
+		if (cq->channel & 2) {
+			if (test_bit(HFC_CFG_SW_DD_DU, &hc->cfg))
+				slot = 0xC1;
+			else
+				slot = 0x81;
+			printk(KERN_DEBUG "%s: Write_hfc: B2_SSL/RSL 0x%x\n",
+			    __func__, slot);
+			Write_hfc(hc, HFCPCI_B2_SSL, slot);
+			Write_hfc(hc, HFCPCI_B2_RSL, slot);
+			hc->hw.conn = (hc->hw.conn & ~0x38) | 0x30;
+			Write_hfc(hc, HFCPCI_CONNECT, hc->hw.conn);
+		}
+		if (cq->channel & 3)
+			hc->hw.trm |= 0x80;	/* enable IOM-loop */
+		else {
+			hc->hw.conn = (hc->hw.conn & ~0x3f) | 0x09;
+			Write_hfc(hc, HFCPCI_CONNECT, hc->hw.conn);
+			hc->hw.trm &= 0x7f;	/* disable IOM-loop */
+		}
+		Write_hfc(hc, HFCPCI_TRM, hc->hw.trm);
+		break;
+	case MISDN_CTRL_CONNECT:
+		if (cq->channel == cq->p1) {
+			ret = -EINVAL;
+			break;
+		}
+		if (cq->channel < 1 || cq->channel > 2 ||
+		    cq->p1 < 1 || cq->p1 > 2) {
+			ret = -EINVAL;
+			break;
+		}
+		if (test_bit(HFC_CFG_SW_DD_DU, &hc->cfg))
+			slot = 0xC0;
+		else
+			slot = 0x80;
+		printk(KERN_DEBUG "%s: Write_hfc: B1_SSL/RSL 0x%x\n",
+		    __func__, slot);
+		Write_hfc(hc, HFCPCI_B1_SSL, slot);
+		Write_hfc(hc, HFCPCI_B2_RSL, slot);
+		if (test_bit(HFC_CFG_SW_DD_DU, &hc->cfg))
+			slot = 0xC1;
+		else
+			slot = 0x81;
+		printk(KERN_DEBUG "%s: Write_hfc: B2_SSL/RSL 0x%x\n",
+		    __func__, slot);
+		Write_hfc(hc, HFCPCI_B2_SSL, slot);
+		Write_hfc(hc, HFCPCI_B1_RSL, slot);
+		hc->hw.conn = (hc->hw.conn & ~0x3f) | 0x36;
+		Write_hfc(hc, HFCPCI_CONNECT, hc->hw.conn);
+		hc->hw.trm |= 0x80;
+		Write_hfc(hc, HFCPCI_TRM, hc->hw.trm);
+		break;
+	case MISDN_CTRL_DISCONNECT:
+		hc->hw.conn = (hc->hw.conn & ~0x3f) | 0x09;
+		Write_hfc(hc, HFCPCI_CONNECT, hc->hw.conn);
+		hc->hw.trm &= 0x7f;	/* disable IOM-loop */
+		break;
+	default:
+		printk(KERN_WARNING "%s: unknown Op %x\n",
+		    __func__, cq->op);
+		ret = -EINVAL;
+		break;
+	}
+	return ret;
+}
+
+static int
+open_dchannel(struct hfc_pci *hc, struct mISDNchannel *ch,
+    struct channel_req *rq)
+{
+	int err = 0;
+
+	if (debug & DEBUG_HW_OPEN)
+		printk(KERN_DEBUG "%s: dev(%d) open from %p\n", __func__,
+		    hc->dch.dev.id, __builtin_return_address(0));
+	if (rq->protocol == ISDN_P_NONE)
+		return -EINVAL;
+	if (!hc->initdone) {
+		if (rq->protocol == ISDN_P_TE_S0) {
+			err = create_l1(&hc->dch, hfc_l1callback);
+			if (err)
+				return err;
+		}
+		hc->hw.protocol = rq->protocol;
+		ch->protocol = rq->protocol;
+		err = init_card(hc);
+		if (err)
+			return err;
+	} else {
+		if (rq->protocol != ch->protocol) {
+			if (hc->hw.protocol == ISDN_P_TE_S0)
+				l1_event(hc->dch.l1, CLOSE_CHANNEL);
+			hc->hw.protocol = rq->protocol;
+			ch->protocol = rq->protocol;
+			hfcpci_setmode(hc);
+		}
+	}
+
+	if (((ch->protocol == ISDN_P_NT_S0) && (hc->dch.state == 3)) ||
+	    ((ch->protocol == ISDN_P_TE_S0) && (hc->dch.state == 7))) {
+		_queue_data(ch, PH_ACTIVATE_IND, MISDN_ID_ANY,
+		    0, NULL, GFP_KERNEL);
+	}
+	rq->ch = ch;
+	if (!try_module_get(THIS_MODULE))
+		printk(KERN_WARNING "%s:cannot get module\n", __func__);
+	return 0;
+}
+
+static int
+open_bchannel(struct hfc_pci *hc, struct channel_req *rq)
+{
+	struct bchannel		*bch;
+
+	if (rq->adr.channel > 2)
+		return -EINVAL;
+	if (rq->protocol == ISDN_P_NONE)
+		return -EINVAL;
+	bch = &hc->bch[rq->adr.channel - 1];
+	if (test_and_set_bit(FLG_OPEN, &bch->Flags))
+		return -EBUSY; /* b-channel can be only open once */
+	bch->ch.protocol = rq->protocol;
+	rq->ch = &bch->ch; /* TODO: E-channel */
+	if (!try_module_get(THIS_MODULE))
+		printk(KERN_WARNING "%s:cannot get module\n", __func__);
+	return 0;
+}
+
+/*
+ * device control function
+ */
+static int
+hfc_dctrl(struct mISDNchannel *ch, u_int cmd, void *arg)
+{
+	struct mISDNdevice	*dev = container_of(ch, struct mISDNdevice, D);
+	struct dchannel		*dch = container_of(dev, struct dchannel, dev);
+	struct hfc_pci		*hc = dch->hw;
+	struct channel_req	*rq;
+	int			err = 0;
+
+	if (dch->debug & DEBUG_HW)
+		printk(KERN_DEBUG "%s: cmd:%x %p\n",
+		    __func__, cmd, arg);
+	switch (cmd) {
+	case OPEN_CHANNEL:
+		rq = arg;
+		if (rq->adr.channel == 0)
+			err = open_dchannel(hc, ch, rq);
+		else
+			err = open_bchannel(hc, rq);
+		break;
+	case CLOSE_CHANNEL:
+		if (debug & DEBUG_HW_OPEN)
+			printk(KERN_DEBUG "%s: dev(%d) close from %p\n",
+			    __func__, hc->dch.dev.id,
+			    __builtin_return_address(0));
+		module_put(THIS_MODULE);
+		break;
+	case CONTROL_CHANNEL:
+		err = channel_ctrl(hc, arg);
+		break;
+	default:
+		if (dch->debug & DEBUG_HW)
+			printk(KERN_DEBUG "%s: unknown command %x\n",
+			    __func__, cmd);
+		return -EINVAL;
+	}
+	return err;
+}
+
+static int
+setup_hw(struct hfc_pci *hc)
+{
+	void	*buffer;
+
+	printk(KERN_INFO "mISDN: HFC-PCI driver %s\n", hfcpci_revision);
+	hc->hw.cirm = 0;
+	hc->dch.state = 0;
+	pci_set_master(hc->pdev);
+	if (!hc->irq) {
+		printk(KERN_WARNING "HFC-PCI: No IRQ for PCI card found\n");
+		return 1;
+	}
+	hc->hw.pci_io = (char *)(ulong)hc->pdev->resource[1].start;
+
+	if (!hc->hw.pci_io) {
+		printk(KERN_WARNING "HFC-PCI: No IO-Mem for PCI card found\n");
+		return 1;
+	}
+	/* Allocate memory for FIFOS */
+	/* the memory needs to be on a 32k boundary within the first 4G */
+	pci_set_dma_mask(hc->pdev, 0xFFFF8000);
+	buffer = pci_alloc_consistent(hc->pdev, 0x8000, &hc->hw.dmahandle);
+	/* We silently assume the address is okay if nonzero */
+	if (!buffer) {
+		printk(KERN_WARNING
+		    "HFC-PCI: Error allocating memory for FIFO!\n");
+		return 1;
+	}
+	hc->hw.fifos = buffer;
+	pci_write_config_dword(hc->pdev, 0x80, hc->hw.dmahandle);
+	hc->hw.pci_io = ioremap((ulong) hc->hw.pci_io, 256);
+	printk(KERN_INFO
+		"HFC-PCI: defined at mem %#lx fifo %#lx(%#lx) IRQ %d HZ %d\n",
+		(u_long) hc->hw.pci_io, (u_long) hc->hw.fifos,
+		(u_long) virt_to_bus(hc->hw.fifos),
+		hc->irq, HZ);
+	/* enable memory mapped ports, disable busmaster */
+	pci_write_config_word(hc->pdev, PCI_COMMAND, PCI_ENA_MEMIO);
+	hc->hw.int_m2 = 0;
+	disable_hwirq(hc);
+	hc->hw.int_m1 = 0;
+	Write_hfc(hc, HFCPCI_INT_M1, hc->hw.int_m1);
+	/* At this point the needed PCI config is done */
+	/* fifos are still not enabled */
+	hc->hw.timer.function = (void *) hfcpci_Timer;
+	hc->hw.timer.data = (long) hc;
+	init_timer(&hc->hw.timer);
+	/* default PCM master */
+	test_and_set_bit(HFC_CFG_MASTER, &hc->cfg);
+	return 0;
+}
+
+static void
+release_card(struct hfc_pci *hc) {
+	u_long	flags;
+
+	spin_lock_irqsave(&hc->lock, flags);
+	hc->hw.int_m2 = 0; /* interrupt output off ! */
+	disable_hwirq(hc);
+	mode_hfcpci(&hc->bch[0], 1, ISDN_P_NONE);
+	mode_hfcpci(&hc->bch[1], 2, ISDN_P_NONE);
+	if (hc->dch.timer.function != NULL) {
+		del_timer(&hc->dch.timer);
+		hc->dch.timer.function = NULL;
+	}
+	spin_unlock_irqrestore(&hc->lock, flags);
+	if (hc->hw.protocol == ISDN_P_TE_S0)
+		l1_event(hc->dch.l1, CLOSE_CHANNEL);
+	if (hc->initdone)
+		free_irq(hc->irq, hc);
+	release_io_hfcpci(hc); /* must release after free_irq! */
+	mISDN_unregister_device(&hc->dch.dev);
+	mISDN_freebchannel(&hc->bch[1]);
+	mISDN_freebchannel(&hc->bch[0]);
+	mISDN_freedchannel(&hc->dch);
+	list_del(&hc->list);
+	pci_set_drvdata(hc->pdev, NULL);
+	kfree(hc);
+}
+
+static int
+setup_card(struct hfc_pci *card)
+{
+	int		err = -EINVAL;
+	u_int		i;
+	u_long		flags;
+	char		name[MISDN_MAX_IDLEN];
+
+	if (HFC_cnt >= MAX_CARDS)
+		return -EINVAL; /* maybe better value */
+
+	card->dch.debug = debug;
+	spin_lock_init(&card->lock);
+	mISDN_initdchannel(&card->dch, MAX_DFRAME_LEN_L1, ph_state);
+	card->dch.hw = card;
+	card->dch.dev.Dprotocols = (1 << ISDN_P_TE_S0) | (1 << ISDN_P_NT_S0);
+	card->dch.dev.Bprotocols = (1 << (ISDN_P_B_RAW & ISDN_P_B_MASK)) |
+	    (1 << (ISDN_P_B_HDLC & ISDN_P_B_MASK));
+	card->dch.dev.D.send = hfcpci_l2l1D;
+	card->dch.dev.D.ctrl = hfc_dctrl;
+	card->dch.dev.nrbchan = 2;
+	for (i = 0; i < 2; i++) {
+		card->bch[i].nr = i + 1;
+		test_and_set_bit(i + 1, &card->dch.dev.channelmap[0]);
+		card->bch[i].debug = debug;
+		mISDN_initbchannel(&card->bch[i], MAX_DATA_MEM);
+		card->bch[i].hw = card;
+		card->bch[i].ch.send = hfcpci_l2l1B;
+		card->bch[i].ch.ctrl = hfc_bctrl;
+		card->bch[i].ch.nr = i + 1;
+		list_add(&card->bch[i].ch.list, &card->dch.dev.bchannels);
+	}
+	err = setup_hw(card);
+	if (err)
+		goto error;
+	snprintf(name, MISDN_MAX_IDLEN - 1, "hfc-pci.%d", HFC_cnt + 1);
+	err = mISDN_register_device(&card->dch.dev, name);
+	if (err)
+		goto error;
+	HFC_cnt++;
+	write_lock_irqsave(&HFClock, flags);
+	list_add_tail(&card->list, &HFClist);
+	write_unlock_irqrestore(&HFClock, flags);
+	printk(KERN_INFO "HFC %d cards installed\n", HFC_cnt);
+	return 0;
+error:
+	mISDN_freebchannel(&card->bch[1]);
+	mISDN_freebchannel(&card->bch[0]);
+	mISDN_freedchannel(&card->dch);
+	kfree(card);
+	return err;
+}
+
+/* private data in the PCI devices list */
+struct _hfc_map {
+	u_int	subtype;
+	u_int	flag;
+	char	*name;
+};
+
+static const struct _hfc_map hfc_map[] =
+{
+	{HFC_CCD_2BD0, 0, "CCD/Billion/Asuscom 2BD0"},
+	{HFC_CCD_B000, 0, "Billion B000"},
+	{HFC_CCD_B006, 0, "Billion B006"},
+	{HFC_CCD_B007, 0, "Billion B007"},
+	{HFC_CCD_B008, 0, "Billion B008"},
+	{HFC_CCD_B009, 0, "Billion B009"},
+	{HFC_CCD_B00A, 0, "Billion B00A"},
+	{HFC_CCD_B00B, 0, "Billion B00B"},
+	{HFC_CCD_B00C, 0, "Billion B00C"},
+	{HFC_CCD_B100, 0, "Seyeon B100"},
+	{HFC_CCD_B700, 0, "Primux II S0 B700"},
+	{HFC_CCD_B701, 0, "Primux II S0 NT B701"},
+	{HFC_ABOCOM_2BD1, 0, "Abocom/Magitek 2BD1"},
+	{HFC_ASUS_0675, 0, "Asuscom/Askey 675"},
+	{HFC_BERKOM_TCONCEPT, 0, "German telekom T-Concept"},
+	{HFC_BERKOM_A1T, 0, "German telekom A1T"},
+	{HFC_ANIGMA_MC145575, 0, "Motorola MC145575"},
+	{HFC_ZOLTRIX_2BD0, 0, "Zoltrix 2BD0"},
+	{HFC_DIGI_DF_M_IOM2_E, 0,
+	    "Digi International DataFire Micro V IOM2 (Europe)"},
+	{HFC_DIGI_DF_M_E, 0,
+	    "Digi International DataFire Micro V (Europe)"},
+	{HFC_DIGI_DF_M_IOM2_A, 0,
+	    "Digi International DataFire Micro V IOM2 (North America)"},
+	{HFC_DIGI_DF_M_A, 0,
+	    "Digi International DataFire Micro V (North America)"},
+	{HFC_SITECOM_DC105V2, 0, "Sitecom Connectivity DC-105 ISDN TA"},
+	{},
+};
+
+static struct pci_device_id hfc_ids[] =
+{
+	{PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_2BD0,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[0]},
+	{PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_B000,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[1]},
+	{PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_B006,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[2]},
+	{PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_B007,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[3]},
+	{PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_B008,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[4]},
+	{PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_B009,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[5]},
+	{PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_B00A,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[6]},
+	{PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_B00B,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[7]},
+	{PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_B00C,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[8]},
+	{PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_B100,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[9]},
+	{PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_B700,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[10]},
+	{PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_B701,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[11]},
+	{PCI_VENDOR_ID_ABOCOM, PCI_DEVICE_ID_ABOCOM_2BD1,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[12]},
+	{PCI_VENDOR_ID_ASUSTEK, PCI_DEVICE_ID_ASUSTEK_0675,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[13]},
+	{PCI_VENDOR_ID_BERKOM, PCI_DEVICE_ID_BERKOM_T_CONCEPT,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[14]},
+	{PCI_VENDOR_ID_BERKOM, PCI_DEVICE_ID_BERKOM_A1T,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[15]},
+	{PCI_VENDOR_ID_ANIGMA, PCI_DEVICE_ID_ANIGMA_MC145575,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[16]},
+	{PCI_VENDOR_ID_ZOLTRIX, PCI_DEVICE_ID_ZOLTRIX_2BD0,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[17]},
+	{PCI_VENDOR_ID_DIGI, PCI_DEVICE_ID_DIGI_DF_M_IOM2_E,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[18]},
+	{PCI_VENDOR_ID_DIGI, PCI_DEVICE_ID_DIGI_DF_M_E,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[19]},
+	{PCI_VENDOR_ID_DIGI, PCI_DEVICE_ID_DIGI_DF_M_IOM2_A,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[20]},
+	{PCI_VENDOR_ID_DIGI, PCI_DEVICE_ID_DIGI_DF_M_A,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[21]},
+	{PCI_VENDOR_ID_SITECOM, PCI_DEVICE_ID_SITECOM_DC105V2,
+		PCI_ANY_ID, PCI_ANY_ID, 0, 0, (unsigned long) &hfc_map[22]},
+	{},
+};
+
+static int __devinit
+hfc_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+	int		err = -ENOMEM;
+	struct hfc_pci	*card;
+	struct _hfc_map	*m = (struct _hfc_map *)ent->driver_data;
+
+	card = kzalloc(sizeof(struct hfc_pci), GFP_ATOMIC);
+	if (!card) {
+		printk(KERN_ERR "No kmem for HFC card\n");
+		return err;
+	}
+	card->pdev = pdev;
+	card->subtype = m->subtype;
+	err = pci_enable_device(pdev);
+	if (err) {
+		kfree(card);
+		return err;
+	}
+
+	printk(KERN_INFO "mISDN_hfcpci: found adapter %s at %s\n",
+	       m->name, pci_name(pdev));
+
+	card->irq = pdev->irq;
+	pci_set_drvdata(pdev, card);
+	err = setup_card(card);
+	if (err)
+		pci_set_drvdata(pdev, NULL);
+	return err;
+}
+
+static void __devexit
+hfc_remove_pci(struct pci_dev *pdev)
+{
+	struct hfc_pci	*card = pci_get_drvdata(pdev);
+	u_long		flags;
+
+	if (card) {
+		write_lock_irqsave(&HFClock, flags);
+		release_card(card);
+		write_unlock_irqrestore(&HFClock, flags);
+	} else
+		if (debug)
+			printk(KERN_WARNING "%s: drvdata allready removed\n",
+			    __func__);
+}
+
+
+static struct pci_driver hfc_driver = {
+	.name = "hfcpci",
+	.probe = hfc_probe,
+	.remove = __devexit_p(hfc_remove_pci),
+	.id_table = hfc_ids,
+};
+
+static int __init
+HFC_init(void)
+{
+	int		err;
+
+	err = pci_register_driver(&hfc_driver);
+	return err;
+}
+
+static void __exit
+HFC_cleanup(void)
+{
+	struct hfc_pci	*card, *next;
+
+	list_for_each_entry_safe(card, next, &HFClist, list) {
+		release_card(card);
+	}
+	pci_unregister_driver(&hfc_driver);
+}
+
+module_init(HFC_init);
+module_exit(HFC_cleanup);

From af69fb3a8ffa37e986db00ed93099dc44babeef4 Mon Sep 17 00:00:00 2001
From: Karsten Keil <kkeil@suse.de>
Date: Sun, 27 Jul 2008 02:00:43 +0200
Subject: [PATCH 1117/1171] Add mISDN HFC multiport driver

Enable support for cards with Cologne Chip AG's HFC multiport
chip.

Signed-off-by: Karsten Keil <kkeil@suse.de>
---
 drivers/isdn/hardware/mISDN/Kconfig     |   12 +
 drivers/isdn/hardware/mISDN/Makefile    |    1 +
 drivers/isdn/hardware/mISDN/hfc_multi.h | 1204 +++++
 drivers/isdn/hardware/mISDN/hfcmulti.c  | 5320 +++++++++++++++++++++++
 include/linux/pci_ids.h                 |   33 +
 5 files changed, 6570 insertions(+)
 create mode 100644 drivers/isdn/hardware/mISDN/hfc_multi.h
 create mode 100644 drivers/isdn/hardware/mISDN/hfcmulti.c

diff --git a/drivers/isdn/hardware/mISDN/Kconfig b/drivers/isdn/hardware/mISDN/Kconfig
index f62dc8752be9..14793480c453 100644
--- a/drivers/isdn/hardware/mISDN/Kconfig
+++ b/drivers/isdn/hardware/mISDN/Kconfig
@@ -11,3 +11,15 @@ config MISDN_HFCPCI
 	  Enable support for cards with Cologne Chip AG's
           HFC PCI chip.
 
+config MISDN_HFCMULTI
+	tristate "Support for HFC multiport cards (HFC-4S/8S/E1)"
+	depends on PCI
+	depends on MISDN
+	help
+	  Enable support for cards with Cologne Chip AG's HFC multiport
+	  chip. There are three types of chips that are quite similar,
+	  but the interface is different:
+	   * HFC-4S (4 S/T interfaces on one chip)
+	   * HFC-8S (8 S/T interfaces on one chip)
+	   * HFC-E1 (E1 interface for 2Mbit ISDN)
+
diff --git a/drivers/isdn/hardware/mISDN/Makefile b/drivers/isdn/hardware/mISDN/Makefile
index 6f20a40b9d54..1e7ca5332ad7 100644
--- a/drivers/isdn/hardware/mISDN/Makefile
+++ b/drivers/isdn/hardware/mISDN/Makefile
@@ -4,3 +4,4 @@
 #
 
 obj-$(CONFIG_MISDN_HFCPCI) += hfcpci.o
+obj-$(CONFIG_MISDN_HFCMULTI) += hfcmulti.o
diff --git a/drivers/isdn/hardware/mISDN/hfc_multi.h b/drivers/isdn/hardware/mISDN/hfc_multi.h
new file mode 100644
index 000000000000..a33d87afc843
--- /dev/null
+++ b/drivers/isdn/hardware/mISDN/hfc_multi.h
@@ -0,0 +1,1204 @@
+/*
+ * see notice in hfc_multi.c
+ */
+
+extern void ztdummy_extern_interrupt(void);
+extern void ztdummy_register_interrupt(void);
+extern int ztdummy_unregister_interrupt(void);
+
+#define DEBUG_HFCMULTI_FIFO	0x00010000
+#define	DEBUG_HFCMULTI_CRC	0x00020000
+#define	DEBUG_HFCMULTI_INIT	0x00040000
+#define	DEBUG_HFCMULTI_PLXSD	0x00080000
+#define	DEBUG_HFCMULTI_MODE	0x00100000
+#define	DEBUG_HFCMULTI_MSG	0x00200000
+#define	DEBUG_HFCMULTI_STATE	0x00400000
+#define	DEBUG_HFCMULTI_SYNC	0x01000000
+#define	DEBUG_HFCMULTI_DTMF	0x02000000
+#define	DEBUG_HFCMULTI_LOCK	0x80000000
+
+#define	PCI_ENA_REGIO	0x01
+#define	PCI_ENA_MEMIO	0x02
+
+/*
+ * NOTE: some registers are assigned multiple times due to different modes
+ *       also registers are assigned differen for HFC-4s/8s and HFC-E1
+ */
+
+/*
+#define MAX_FRAME_SIZE	2048
+*/
+
+struct hfc_chan {
+	struct dchannel	*dch;	/* link if channel is a D-channel */
+	struct bchannel	*bch;	/* link if channel is a B-channel */
+	int		port; 	/* the interface port this */
+				/* channel is associated with */
+	int		nt_timer; /* -1 if off, 0 if elapsed, >0 if running */
+	int		los, ais, slip_tx, slip_rx, rdi; /* current alarms */
+	int		jitter;
+	u_long		cfg;	/* port configuration */
+	int		sync;	/* sync state (used by E1) */
+	u_int		protocol; /* current protocol */
+	int		slot_tx; /* current pcm slot */
+	int		bank_tx; /* current pcm bank */
+	int		slot_rx;
+	int		bank_rx;
+	int		conf;	/* conference setting of TX slot */
+	int		txpending;	/* if there is currently data in */
+					/* the FIFO 0=no, 1=yes, 2=splloop */
+	int		rx_off; /* set to turn fifo receive off */
+	int		coeff_count; /* curren coeff block */
+	s32		*coeff; /* memory pointer to 8 coeff blocks */
+};
+
+
+struct hfcm_hw {
+	u_char	r_ctrl;
+	u_char	r_irq_ctrl;
+	u_char	r_cirm;
+	u_char	r_ram_sz;
+	u_char	r_pcm_md0;
+	u_char	r_irqmsk_misc;
+	u_char	r_dtmf;
+	u_char	r_st_sync;
+	u_char	r_sci_msk;
+	u_char	r_tx0, r_tx1;
+	u_char	a_st_ctrl0[8];
+	timer_t	timer;
+};
+
+
+/* for each stack these flags are used (cfg) */
+#define	HFC_CFG_NONCAP_TX	1 /* S/T TX interface has less capacity */
+#define	HFC_CFG_DIS_ECHANNEL	2 /* disable E-channel processing */
+#define	HFC_CFG_REG_ECHANNEL	3 /* register E-channel */
+#define	HFC_CFG_OPTICAL		4 /* the E1 interface is optical */
+#define	HFC_CFG_REPORT_LOS	5 /* the card should report loss of signal */
+#define	HFC_CFG_REPORT_AIS	6 /* the card should report alarm ind. sign. */
+#define	HFC_CFG_REPORT_SLIP	7 /* the card should report bit slips */
+#define	HFC_CFG_REPORT_RDI	8 /* the card should report remote alarm */
+#define	HFC_CFG_DTMF		9 /* enable DTMF-detection */
+#define	HFC_CFG_CRC4		10 /* disable CRC-4 Multiframe mode, */
+					/* use double frame instead. */
+
+#define	HFC_CHIP_EXRAM_128	0 /* external ram 128k */
+#define	HFC_CHIP_EXRAM_512	1 /* external ram 256k */
+#define	HFC_CHIP_REVISION0	2 /* old fifo handling */
+#define	HFC_CHIP_PCM_SLAVE	3 /* PCM is slave */
+#define	HFC_CHIP_PCM_MASTER	4 /* PCM is master */
+#define	HFC_CHIP_RX_SYNC	5 /* disable pll sync for pcm */
+#define	HFC_CHIP_DTMF		6 /* DTMF decoding is enabled */
+#define	HFC_CHIP_ULAW		7 /* ULAW mode */
+#define	HFC_CHIP_CLOCK2		8 /* double clock mode */
+#define	HFC_CHIP_E1CLOCK_GET	9 /* always get clock from E1 interface */
+#define	HFC_CHIP_E1CLOCK_PUT	10 /* always put clock from E1 interface */
+#define	HFC_CHIP_WATCHDOG	11 /* whether we should send signals */
+					/* to the watchdog */
+#define	HFC_CHIP_B410P		12 /* whether we have a b410p with echocan in */
+					/* hw */
+#define	HFC_CHIP_PLXSD		13 /* whether we have a Speech-Design PLX */
+
+#define HFC_IO_MODE_PCIMEM	0x00 /* normal memory mapped IO */
+#define HFC_IO_MODE_REGIO	0x01 /* PCI io access */
+#define HFC_IO_MODE_PLXSD	0x02 /* access HFC via PLX9030 */
+
+/* table entry in the PCI devices list */
+struct hm_map {
+	char *vendor_name;
+	char *card_name;
+	int type;
+	int ports;
+	int clock2;
+	int leds;
+	int opticalsupport;
+	int dip_type;
+	int io_mode;
+};
+
+struct hfc_multi {
+	struct list_head	list;
+	struct hm_map	*mtyp;
+	int		id;
+	int		pcm;	/* id of pcm bus */
+	int		type;
+	int		ports;
+
+	u_int		irq;	/* irq used by card */
+	u_int		irqcnt;
+	struct pci_dev	*pci_dev;
+	int		io_mode; /* selects mode */
+#ifdef HFC_REGISTER_DEBUG
+	void		(*HFC_outb)(struct hfc_multi *hc, u_char reg,
+				u_char val, const char *function, int line);
+	void		(*HFC_outb_nodebug)(struct hfc_multi *hc, u_char reg,
+				u_char val, const char *function, int line);
+	u_char		(*HFC_inb)(struct hfc_multi *hc, u_char reg,
+				const char *function, int line);
+	u_char		(*HFC_inb_nodebug)(struct hfc_multi *hc, u_char reg,
+				const char *function, int line);
+	u_short		(*HFC_inw)(struct hfc_multi *hc, u_char reg,
+				const char *function, int line);
+	u_short		(*HFC_inw_nodebug)(struct hfc_multi *hc, u_char reg,
+				const char *function, int line);
+	void		(*HFC_wait)(struct hfc_multi *hc,
+				const char *function, int line);
+	void		(*HFC_wait_nodebug)(struct hfc_multi *hc,
+				const char *function, int line);
+#else
+	void		(*HFC_outb)(struct hfc_multi *hc, u_char reg,
+				u_char val);
+	void		(*HFC_outb_nodebug)(struct hfc_multi *hc, u_char reg,
+				u_char val);
+	u_char		(*HFC_inb)(struct hfc_multi *hc, u_char reg);
+	u_char		(*HFC_inb_nodebug)(struct hfc_multi *hc, u_char reg);
+	u_short		(*HFC_inw)(struct hfc_multi *hc, u_char reg);
+	u_short		(*HFC_inw_nodebug)(struct hfc_multi *hc, u_char reg);
+	void		(*HFC_wait)(struct hfc_multi *hc);
+	void		(*HFC_wait_nodebug)(struct hfc_multi *hc);
+#endif
+	void		(*read_fifo)(struct hfc_multi *hc, u_char *data,
+				int len);
+	void		(*write_fifo)(struct hfc_multi *hc, u_char *data,
+				int len);
+	u_long		pci_origmembase, plx_origmembase, dsp_origmembase;
+	u_char		*pci_membase; /* PCI memory (MUST BE BYTE POINTER) */
+	u_char		*plx_membase; /* PLX memory */
+	u_char		*dsp_membase; /* DSP on PLX */
+	u_long		pci_iobase; /* PCI IO */
+	struct hfcm_hw	hw;	/* remember data of write-only-registers */
+
+	u_long		chip;	/* chip configuration */
+	int		masterclk; /* port that provides master clock -1=off */
+	int		dtmf;	/* flag that dtmf is currently in process */
+	int		Flen;	/* F-buffer size */
+	int		Zlen;	/* Z-buffer size (must be int for calculation)*/
+	int		max_trans; /* maximum transparent fifo fill */
+	int		Zmin;	/* Z-buffer offset */
+	int		DTMFbase; /* base address of DTMF coefficients */
+
+	u_int		slots;	/* number of PCM slots */
+	u_int		leds;	/* type of leds */
+	u_int		ledcount; /* used to animate leds */
+	u_long		ledstate; /* save last state of leds */
+	int		opticalsupport; /* has the e1 board */
+					/* an optical Interface */
+	int		dslot;	/* channel # of d-channel (E1) default 16 */
+
+	u_long		wdcount; 	/* every 500 ms we need to */
+					/* send the watchdog a signal */
+	u_char		wdbyte; /* watchdog toggle byte */
+	u_int		activity[8]; 	/* if there is any action on this */
+					/* port (will be cleared after */
+					/* showing led-states) */
+	int		e1_state; /* keep track of last state */
+	int		e1_getclock; /* if sync is retrieved from interface */
+	int		syncronized; /* keep track of existing sync interface */
+	int		e1_resync; /* resync jobs */
+
+	spinlock_t	lock;	/* the lock */
+
+	/*
+	 * the channel index is counted from 0, regardless where the channel
+	 * is located on the hfc-channel.
+	 * the bch->channel is equvalent to the hfc-channel
+	 */
+	struct hfc_chan	chan[32];
+	u_char		created[8]; /* what port is created */
+	signed char	slot_owner[256]; /* owner channel of slot */
+};
+
+/* PLX GPIOs */
+#define	PLX_GPIO4_DIR_BIT	13
+#define	PLX_GPIO4_BIT		14
+#define	PLX_GPIO5_DIR_BIT	16
+#define	PLX_GPIO5_BIT		17
+#define	PLX_GPIO6_DIR_BIT	19
+#define	PLX_GPIO6_BIT		20
+#define	PLX_GPIO7_DIR_BIT	22
+#define	PLX_GPIO7_BIT		23
+#define PLX_GPIO8_DIR_BIT	25
+#define PLX_GPIO8_BIT		26
+
+#define	PLX_GPIO4		(1 << PLX_GPIO4_BIT)
+#define	PLX_GPIO5		(1 << PLX_GPIO5_BIT)
+#define	PLX_GPIO6		(1 << PLX_GPIO6_BIT)
+#define	PLX_GPIO7		(1 << PLX_GPIO7_BIT)
+#define PLX_GPIO8		(1 << PLX_GPIO8_BIT)
+
+#define	PLX_GPIO4_DIR		(1 << PLX_GPIO4_DIR_BIT)
+#define	PLX_GPIO5_DIR		(1 << PLX_GPIO5_DIR_BIT)
+#define	PLX_GPIO6_DIR		(1 << PLX_GPIO6_DIR_BIT)
+#define	PLX_GPIO7_DIR		(1 << PLX_GPIO7_DIR_BIT)
+#define PLX_GPIO8_DIR		(1 << PLX_GPIO8_DIR_BIT)
+
+#define	PLX_TERM_ON			PLX_GPIO7
+#define	PLX_SLAVE_EN_N		PLX_GPIO5
+#define	PLX_MASTER_EN		PLX_GPIO6
+#define	PLX_SYNC_O_EN		PLX_GPIO4
+#define PLX_DSP_RES_N		PLX_GPIO8
+/* GPIO4..8 Enable & Set to OUT, SLAVE_EN_N = 1 */
+#define PLX_GPIOC_INIT		(PLX_GPIO4_DIR | PLX_GPIO5_DIR | PLX_GPIO6_DIR \
+			| PLX_GPIO7_DIR | PLX_GPIO8_DIR | PLX_SLAVE_EN_N)
+
+/* PLX Interrupt Control/STATUS */
+#define PLX_INTCSR_LINTI1_ENABLE 0x01
+#define PLX_INTCSR_LINTI1_STATUS 0x04
+#define PLX_INTCSR_LINTI2_ENABLE 0x08
+#define PLX_INTCSR_LINTI2_STATUS 0x20
+#define PLX_INTCSR_PCIINT_ENABLE 0x40
+
+/* PLX Registers */
+#define PLX_INTCSR 0x4c
+#define PLX_CNTRL  0x50
+#define PLX_GPIOC  0x54
+
+
+/*
+ * REGISTER SETTING FOR HFC-4S/8S AND HFC-E1
+ */
+
+/* write only registers */
+#define R_CIRM			0x00
+#define R_CTRL			0x01
+#define R_BRG_PCM_CFG 		0x02
+#define R_RAM_ADDR0		0x08
+#define R_RAM_ADDR1		0x09
+#define R_RAM_ADDR2		0x0A
+#define R_FIRST_FIFO		0x0B
+#define R_RAM_SZ		0x0C
+#define R_FIFO_MD		0x0D
+#define R_INC_RES_FIFO		0x0E
+#define R_FSM_IDX		0x0F
+#define R_FIFO			0x0F
+#define R_SLOT			0x10
+#define R_IRQMSK_MISC		0x11
+#define R_SCI_MSK		0x12
+#define R_IRQ_CTRL		0x13
+#define R_PCM_MD0		0x14
+#define R_PCM_MD1		0x15
+#define R_PCM_MD2		0x15
+#define R_SH0H			0x15
+#define R_SH1H			0x15
+#define R_SH0L			0x15
+#define R_SH1L			0x15
+#define R_SL_SEL0		0x15
+#define R_SL_SEL1		0x15
+#define R_SL_SEL2		0x15
+#define R_SL_SEL3		0x15
+#define R_SL_SEL4		0x15
+#define R_SL_SEL5		0x15
+#define R_SL_SEL6		0x15
+#define R_SL_SEL7		0x15
+#define R_ST_SEL		0x16
+#define R_ST_SYNC		0x17
+#define R_CONF_EN		0x18
+#define R_TI_WD			0x1A
+#define R_BERT_WD_MD		0x1B
+#define R_DTMF			0x1C
+#define R_DTMF_N		0x1D
+#define R_E1_WR_STA		0x20
+#define R_E1_RD_STA		0x20
+#define R_LOS0			0x22
+#define R_LOS1			0x23
+#define R_RX0			0x24
+#define R_RX_FR0		0x25
+#define R_RX_FR1		0x26
+#define R_TX0			0x28
+#define R_TX1			0x29
+#define R_TX_FR0		0x2C
+
+#define R_TX_FR1		0x2D
+#define R_TX_FR2		0x2E
+#define R_JATT_ATT		0x2F /* undocumented */
+#define A_ST_RD_STATE		0x30
+#define A_ST_WR_STATE		0x30
+#define R_RX_OFF		0x30
+#define A_ST_CTRL0		0x31
+#define R_SYNC_OUT		0x31
+#define A_ST_CTRL1		0x32
+#define A_ST_CTRL2		0x33
+#define A_ST_SQ_WR		0x34
+#define R_TX_OFF		0x34
+#define R_SYNC_CTRL		0x35
+#define A_ST_CLK_DLY		0x37
+#define R_PWM0			0x38
+#define R_PWM1			0x39
+#define A_ST_B1_TX		0x3C
+#define A_ST_B2_TX		0x3D
+#define A_ST_D_TX		0x3E
+#define R_GPIO_OUT0		0x40
+#define R_GPIO_OUT1		0x41
+#define R_GPIO_EN0		0x42
+#define R_GPIO_EN1		0x43
+#define R_GPIO_SEL		0x44
+#define R_BRG_CTRL		0x45
+#define R_PWM_MD		0x46
+#define R_BRG_MD		0x47
+#define R_BRG_TIM0		0x48
+#define R_BRG_TIM1		0x49
+#define R_BRG_TIM2		0x4A
+#define R_BRG_TIM3		0x4B
+#define R_BRG_TIM_SEL01		0x4C
+#define R_BRG_TIM_SEL23		0x4D
+#define R_BRG_TIM_SEL45		0x4E
+#define R_BRG_TIM_SEL67		0x4F
+#define A_SL_CFG		0xD0
+#define A_CONF			0xD1
+#define A_CH_MSK		0xF4
+#define A_CON_HDLC		0xFA
+#define A_SUBCH_CFG		0xFB
+#define A_CHANNEL		0xFC
+#define A_FIFO_SEQ		0xFD
+#define A_IRQ_MSK		0xFF
+
+/* read only registers */
+#define A_Z12			0x04
+#define A_Z1L			0x04
+#define A_Z1			0x04
+#define A_Z1H			0x05
+#define A_Z2L			0x06
+#define A_Z2			0x06
+#define A_Z2H			0x07
+#define A_F1			0x0C
+#define A_F12			0x0C
+#define A_F2			0x0D
+#define R_IRQ_OVIEW		0x10
+#define R_IRQ_MISC		0x11
+#define R_IRQ_STATECH		0x12
+#define R_CONF_OFLOW		0x14
+#define R_RAM_USE		0x15
+#define R_CHIP_ID		0x16
+#define R_BERT_STA		0x17
+#define R_F0_CNTL		0x18
+#define R_F0_CNTH		0x19
+#define R_BERT_EC		0x1A
+#define R_BERT_ECL		0x1A
+#define R_BERT_ECH		0x1B
+#define R_STATUS		0x1C
+#define R_CHIP_RV		0x1F
+#define R_STATE			0x20
+#define R_SYNC_STA		0x24
+#define R_RX_SL0_0		0x25
+#define R_RX_SL0_1		0x26
+#define R_RX_SL0_2		0x27
+#define R_JATT_DIR		0x2b /* undocumented */
+#define R_SLIP			0x2c
+#define A_ST_RD_STA		0x30
+#define R_FAS_EC		0x30
+#define R_FAS_ECL		0x30
+#define R_FAS_ECH		0x31
+#define R_VIO_EC		0x32
+#define R_VIO_ECL		0x32
+#define R_VIO_ECH		0x33
+#define A_ST_SQ_RD		0x34
+#define R_CRC_EC		0x34
+#define R_CRC_ECL		0x34
+#define R_CRC_ECH		0x35
+#define R_E_EC			0x36
+#define R_E_ECL			0x36
+#define R_E_ECH			0x37
+#define R_SA6_SA13_EC		0x38
+#define R_SA6_SA13_ECL		0x38
+#define R_SA6_SA13_ECH		0x39
+#define R_SA6_SA23_EC		0x3A
+#define R_SA6_SA23_ECL		0x3A
+#define R_SA6_SA23_ECH		0x3B
+#define A_ST_B1_RX		0x3C
+#define A_ST_B2_RX		0x3D
+#define A_ST_D_RX		0x3E
+#define A_ST_E_RX		0x3F
+#define R_GPIO_IN0		0x40
+#define R_GPIO_IN1		0x41
+#define R_GPI_IN0		0x44
+#define R_GPI_IN1		0x45
+#define R_GPI_IN2		0x46
+#define R_GPI_IN3		0x47
+#define R_INT_DATA		0x88
+#define R_IRQ_FIFO_BL0		0xC8
+#define R_IRQ_FIFO_BL1		0xC9
+#define R_IRQ_FIFO_BL2		0xCA
+#define R_IRQ_FIFO_BL3		0xCB
+#define R_IRQ_FIFO_BL4		0xCC
+#define R_IRQ_FIFO_BL5		0xCD
+#define R_IRQ_FIFO_BL6		0xCE
+#define R_IRQ_FIFO_BL7		0xCF
+
+/* read and write registers */
+#define A_FIFO_DATA0		0x80
+#define A_FIFO_DATA1		0x80
+#define A_FIFO_DATA2		0x80
+#define A_FIFO_DATA0_NOINC	0x84
+#define A_FIFO_DATA1_NOINC	0x84
+#define A_FIFO_DATA2_NOINC	0x84
+#define R_RAM_DATA		0xC0
+
+
+/*
+ * BIT SETTING FOR HFC-4S/8S AND HFC-E1
+ */
+
+/* chapter 2: universal bus interface */
+/* R_CIRM */
+#define V_IRQ_SEL		0x01
+#define V_SRES			0x08
+#define V_HFCRES		0x10
+#define V_PCMRES		0x20
+#define V_STRES			0x40
+#define V_ETRES			0x40
+#define V_RLD_EPR		0x80
+/* R_CTRL */
+#define V_FIFO_LPRIO		0x02
+#define V_SLOW_RD		0x04
+#define V_EXT_RAM		0x08
+#define V_CLK_OFF		0x20
+#define V_ST_CLK		0x40
+/* R_RAM_ADDR0 */
+#define V_RAM_ADDR2		0x01
+#define V_ADDR_RES		0x40
+#define V_ADDR_INC		0x80
+/* R_RAM_SZ */
+#define V_RAM_SZ		0x01
+#define V_PWM0_16KHZ		0x10
+#define V_PWM1_16KHZ		0x20
+#define V_FZ_MD			0x80
+/* R_CHIP_ID */
+#define V_PNP_IRQ		0x01
+#define V_CHIP_ID		0x10
+
+/* chapter 3: data flow */
+/* R_FIRST_FIFO */
+#define V_FIRST_FIRO_DIR	0x01
+#define V_FIRST_FIFO_NUM	0x02
+/* R_FIFO_MD */
+#define V_FIFO_MD		0x01
+#define V_CSM_MD		0x04
+#define V_FSM_MD		0x08
+#define V_FIFO_SZ		0x10
+/* R_FIFO */
+#define V_FIFO_DIR		0x01
+#define V_FIFO_NUM		0x02
+#define V_REV			0x80
+/* R_SLOT */
+#define V_SL_DIR		0x01
+#define V_SL_NUM		0x02
+/* A_SL_CFG */
+#define V_CH_DIR		0x01
+#define V_CH_SEL		0x02
+#define V_ROUTING		0x40
+/* A_CON_HDLC */
+#define V_IFF			0x01
+#define V_HDLC_TRP		0x02
+#define V_TRP_IRQ		0x04
+#define V_DATA_FLOW		0x20
+/* A_SUBCH_CFG */
+#define V_BIT_CNT		0x01
+#define V_START_BIT		0x08
+#define V_LOOP_FIFO		0x40
+#define V_INV_DATA		0x80
+/* A_CHANNEL */
+#define V_CH_DIR0		0x01
+#define V_CH_NUM0		0x02
+/* A_FIFO_SEQ */
+#define V_NEXT_FIFO_DIR		0x01
+#define V_NEXT_FIFO_NUM		0x02
+#define V_SEQ_END		0x40
+
+/* chapter 4: FIFO handling and HDLC controller */
+/* R_INC_RES_FIFO */
+#define V_INC_F			0x01
+#define V_RES_F			0x02
+#define V_RES_LOST		0x04
+
+/* chapter 5: S/T interface */
+/* R_SCI_MSK */
+#define V_SCI_MSK_ST0		0x01
+#define V_SCI_MSK_ST1		0x02
+#define V_SCI_MSK_ST2		0x04
+#define V_SCI_MSK_ST3		0x08
+#define V_SCI_MSK_ST4		0x10
+#define V_SCI_MSK_ST5		0x20
+#define V_SCI_MSK_ST6		0x40
+#define V_SCI_MSK_ST7		0x80
+/* R_ST_SEL */
+#define V_ST_SEL		0x01
+#define V_MULT_ST		0x08
+/* R_ST_SYNC */
+#define V_SYNC_SEL		0x01
+#define V_AUTO_SYNC		0x08
+/* A_ST_WR_STA */
+#define V_ST_SET_STA		0x01
+#define V_ST_LD_STA		0x10
+#define V_ST_ACT		0x20
+#define V_SET_G2_G3		0x80
+/* A_ST_CTRL0 */
+#define V_B1_EN			0x01
+#define V_B2_EN			0x02
+#define V_ST_MD			0x04
+#define V_D_PRIO		0x08
+#define V_SQ_EN			0x10
+#define V_96KHZ			0x20
+#define V_TX_LI			0x40
+#define V_ST_STOP		0x80
+/* A_ST_CTRL1 */
+#define V_G2_G3_EN		0x01
+#define V_D_HI			0x04
+#define V_E_IGNO		0x08
+#define V_E_LO			0x10
+#define V_B12_SWAP		0x80
+/* A_ST_CTRL2 */
+#define V_B1_RX_EN		0x01
+#define V_B2_RX_EN		0x02
+#define V_ST_TRIS		0x40
+/* A_ST_CLK_DLY */
+#define V_ST_CK_DLY		0x01
+#define V_ST_SMPL		0x10
+/* A_ST_D_TX */
+#define V_ST_D_TX		0x40
+/* R_IRQ_STATECH */
+#define V_SCI_ST0		0x01
+#define V_SCI_ST1		0x02
+#define V_SCI_ST2		0x04
+#define V_SCI_ST3		0x08
+#define V_SCI_ST4		0x10
+#define V_SCI_ST5		0x20
+#define V_SCI_ST6		0x40
+#define V_SCI_ST7		0x80
+/* A_ST_RD_STA */
+#define V_ST_STA		0x01
+#define V_FR_SYNC_ST		0x10
+#define V_TI2_EXP		0x20
+#define V_INFO0			0x40
+#define V_G2_G3			0x80
+/* A_ST_SQ_RD */
+#define V_ST_SQ			0x01
+#define V_MF_RX_RDY		0x10
+#define V_MF_TX_RDY		0x80
+/* A_ST_D_RX */
+#define V_ST_D_RX		0x40
+/* A_ST_E_RX */
+#define V_ST_E_RX		0x40
+
+/* chapter 5: E1 interface */
+/* R_E1_WR_STA */
+/* R_E1_RD_STA */
+#define V_E1_SET_STA		0x01
+#define V_E1_LD_STA		0x10
+/* R_RX0 */
+#define V_RX_CODE		0x01
+#define V_RX_FBAUD		0x04
+#define V_RX_CMI		0x08
+#define V_RX_INV_CMI		0x10
+#define V_RX_INV_CLK		0x20
+#define V_RX_INV_DATA		0x40
+#define V_AIS_ITU		0x80
+/* R_RX_FR0 */
+#define V_NO_INSYNC		0x01
+#define V_AUTO_RESYNC		0x02
+#define V_AUTO_RECO		0x04
+#define V_SWORD_COND		0x08
+#define V_SYNC_LOSS		0x10
+#define V_XCRC_SYNC		0x20
+#define V_MF_RESYNC		0x40
+#define V_RESYNC		0x80
+/* R_RX_FR1 */
+#define V_RX_MF			0x01
+#define V_RX_MF_SYNC		0x02
+#define V_RX_SL0_RAM		0x04
+#define V_ERR_SIM		0x20
+#define V_RES_NMF		0x40
+/* R_TX0 */
+#define V_TX_CODE		0x01
+#define V_TX_FBAUD		0x04
+#define V_TX_CMI_CODE		0x08
+#define V_TX_INV_CMI_CODE	0x10
+#define V_TX_INV_CLK		0x20
+#define V_TX_INV_DATA		0x40
+#define V_OUT_EN		0x80
+/* R_TX1 */
+#define V_INV_CLK		0x01
+#define V_EXCHG_DATA_LI		0x02
+#define V_AIS_OUT		0x04
+#define V_ATX			0x20
+#define V_NTRI			0x40
+#define V_AUTO_ERR_RES		0x80
+/* R_TX_FR0 */
+#define V_TRP_FAS		0x01
+#define V_TRP_NFAS		0x02
+#define V_TRP_RAL		0x04
+#define V_TRP_SA		0x08
+/* R_TX_FR1 */
+#define V_TX_FAS		0x01
+#define V_TX_NFAS		0x02
+#define V_TX_RAL		0x04
+#define V_TX_SA			0x08
+/* R_TX_FR2 */
+#define V_TX_MF			0x01
+#define V_TRP_SL0		0x02
+#define V_TX_SL0_RAM		0x04
+#define V_TX_E			0x10
+#define V_NEG_E			0x20
+#define V_XS12_ON		0x40
+#define V_XS15_ON		0x80
+/* R_RX_OFF */
+#define V_RX_SZ			0x01
+#define V_RX_INIT		0x04
+/* R_SYNC_OUT */
+#define V_SYNC_E1_RX		0x01
+#define V_IPATS0		0x20
+#define V_IPATS1		0x40
+#define V_IPATS2		0x80
+/* R_TX_OFF */
+#define V_TX_SZ			0x01
+#define V_TX_INIT		0x04
+/* R_SYNC_CTRL */
+#define V_EXT_CLK_SYNC		0x01
+#define V_SYNC_OFFS		0x02
+#define V_PCM_SYNC		0x04
+#define V_NEG_CLK		0x08
+#define V_HCLK			0x10
+/*
+#define V_JATT_AUTO_DEL		0x20
+#define V_JATT_AUTO		0x40
+*/
+#define V_JATT_OFF		0x80
+/* R_STATE */
+#define V_E1_STA		0x01
+#define V_ALT_FR_RX		0x40
+#define V_ALT_FR_TX		0x80
+/* R_SYNC_STA */
+#define V_RX_STA		0x01
+#define V_FR_SYNC_E1		0x04
+#define V_SIG_LOS		0x08
+#define V_MFA_STA		0x10
+#define V_AIS			0x40
+#define V_NO_MF_SYNC		0x80
+/* R_RX_SL0_0 */
+#define V_SI_FAS		0x01
+#define V_SI_NFAS		0x02
+#define V_A			0x04
+#define V_CRC_OK		0x08
+#define V_TX_E1			0x10
+#define V_TX_E2			0x20
+#define V_RX_E1			0x40
+#define V_RX_E2			0x80
+/* R_SLIP */
+#define V_SLIP_RX		0x01
+#define V_FOSLIP_RX		0x08
+#define V_SLIP_TX		0x10
+#define V_FOSLIP_TX		0x80
+
+/* chapter 6: PCM interface */
+/* R_PCM_MD0 */
+#define V_PCM_MD		0x01
+#define V_C4_POL		0x02
+#define V_F0_NEG		0x04
+#define V_F0_LEN		0x08
+#define V_PCM_ADDR		0x10
+/* R_SL_SEL0 */
+#define V_SL_SEL0		0x01
+#define V_SH_SEL0		0x80
+/* R_SL_SEL1 */
+#define V_SL_SEL1		0x01
+#define V_SH_SEL1		0x80
+/* R_SL_SEL2 */
+#define V_SL_SEL2		0x01
+#define V_SH_SEL2		0x80
+/* R_SL_SEL3 */
+#define V_SL_SEL3		0x01
+#define V_SH_SEL3		0x80
+/* R_SL_SEL4 */
+#define V_SL_SEL4		0x01
+#define V_SH_SEL4		0x80
+/* R_SL_SEL5 */
+#define V_SL_SEL5		0x01
+#define V_SH_SEL5		0x80
+/* R_SL_SEL6 */
+#define V_SL_SEL6		0x01
+#define V_SH_SEL6		0x80
+/* R_SL_SEL7 */
+#define V_SL_SEL7		0x01
+#define V_SH_SEL7		0x80
+/* R_PCM_MD1 */
+#define V_ODEC_CON		0x01
+#define V_PLL_ADJ		0x04
+#define V_PCM_DR		0x10
+#define V_PCM_LOOP		0x40
+/* R_PCM_MD2 */
+#define V_SYNC_PLL		0x02
+#define V_SYNC_SRC		0x04
+#define V_SYNC_OUT		0x08
+#define V_ICR_FR_TIME		0x40
+#define V_EN_PLL		0x80
+
+/* chapter 7: pulse width modulation */
+/* R_PWM_MD */
+#define V_EXT_IRQ_EN		0x08
+#define V_PWM0_MD		0x10
+#define V_PWM1_MD		0x40
+
+/* chapter 8: multiparty audio conferences */
+/* R_CONF_EN */
+#define V_CONF_EN		0x01
+#define V_ULAW			0x80
+/* A_CONF */
+#define V_CONF_NUM		0x01
+#define V_NOISE_SUPPR		0x08
+#define V_ATT_LEV		0x20
+#define V_CONF_SL		0x80
+/* R_CONF_OFLOW */
+#define V_CONF_OFLOW0		0x01
+#define V_CONF_OFLOW1		0x02
+#define V_CONF_OFLOW2		0x04
+#define V_CONF_OFLOW3		0x08
+#define V_CONF_OFLOW4		0x10
+#define V_CONF_OFLOW5		0x20
+#define V_CONF_OFLOW6		0x40
+#define V_CONF_OFLOW7		0x80
+
+/* chapter 9: DTMF contoller */
+/* R_DTMF0 */
+#define V_DTMF_EN		0x01
+#define V_HARM_SEL		0x02
+#define V_DTMF_RX_CH		0x04
+#define V_DTMF_STOP		0x08
+#define V_CHBL_SEL		0x10
+#define V_RST_DTMF		0x40
+#define V_ULAW_SEL		0x80
+
+/* chapter 10: BERT */
+/* R_BERT_WD_MD */
+#define V_PAT_SEQ		0x01
+#define V_BERT_ERR		0x08
+#define V_AUTO_WD_RES		0x20
+#define V_WD_RES		0x80
+/* R_BERT_STA */
+#define V_BERT_SYNC_SRC		0x01
+#define V_BERT_SYNC		0x10
+#define V_BERT_INV_DATA		0x20
+
+/* chapter 11: auxiliary interface */
+/* R_BRG_PCM_CFG */
+#define V_BRG_EN		0x01
+#define V_BRG_MD		0x02
+#define V_PCM_CLK		0x20
+#define V_ADDR_WRDLY		0x40
+/* R_BRG_CTRL */
+#define V_BRG_CS		0x01
+#define V_BRG_ADDR		0x08
+#define V_BRG_CS_SRC		0x80
+/* R_BRG_MD */
+#define V_BRG_MD0		0x01
+#define V_BRG_MD1		0x02
+#define V_BRG_MD2		0x04
+#define V_BRG_MD3		0x08
+#define V_BRG_MD4		0x10
+#define V_BRG_MD5		0x20
+#define V_BRG_MD6		0x40
+#define V_BRG_MD7		0x80
+/* R_BRG_TIM0 */
+#define V_BRG_TIM0_IDLE		0x01
+#define V_BRG_TIM0_CLK		0x10
+/* R_BRG_TIM1 */
+#define V_BRG_TIM1_IDLE		0x01
+#define V_BRG_TIM1_CLK		0x10
+/* R_BRG_TIM2 */
+#define V_BRG_TIM2_IDLE		0x01
+#define V_BRG_TIM2_CLK		0x10
+/* R_BRG_TIM3 */
+#define V_BRG_TIM3_IDLE		0x01
+#define V_BRG_TIM3_CLK		0x10
+/* R_BRG_TIM_SEL01 */
+#define V_BRG_WR_SEL0		0x01
+#define V_BRG_RD_SEL0		0x04
+#define V_BRG_WR_SEL1		0x10
+#define V_BRG_RD_SEL1		0x40
+/* R_BRG_TIM_SEL23 */
+#define V_BRG_WR_SEL2		0x01
+#define V_BRG_RD_SEL2		0x04
+#define V_BRG_WR_SEL3		0x10
+#define V_BRG_RD_SEL3		0x40
+/* R_BRG_TIM_SEL45 */
+#define V_BRG_WR_SEL4		0x01
+#define V_BRG_RD_SEL4		0x04
+#define V_BRG_WR_SEL5		0x10
+#define V_BRG_RD_SEL5		0x40
+/* R_BRG_TIM_SEL67 */
+#define V_BRG_WR_SEL6		0x01
+#define V_BRG_RD_SEL6		0x04
+#define V_BRG_WR_SEL7		0x10
+#define V_BRG_RD_SEL7		0x40
+
+/* chapter 12: clock, reset, interrupt, timer and watchdog */
+/* R_IRQMSK_MISC */
+#define V_STA_IRQMSK		0x01
+#define V_TI_IRQMSK		0x02
+#define V_PROC_IRQMSK		0x04
+#define V_DTMF_IRQMSK		0x08
+#define V_IRQ1S_MSK		0x10
+#define V_SA6_IRQMSK		0x20
+#define V_RX_EOMF_MSK		0x40
+#define V_TX_EOMF_MSK		0x80
+/* R_IRQ_CTRL */
+#define V_FIFO_IRQ		0x01
+#define V_GLOB_IRQ_EN		0x08
+#define V_IRQ_POL		0x10
+/* R_TI_WD */
+#define V_EV_TS			0x01
+#define V_WD_TS			0x10
+/* A_IRQ_MSK */
+#define V_IRQ			0x01
+#define V_BERT_EN		0x02
+#define V_MIX_IRQ		0x04
+/* R_IRQ_OVIEW */
+#define V_IRQ_FIFO_BL0		0x01
+#define V_IRQ_FIFO_BL1		0x02
+#define V_IRQ_FIFO_BL2		0x04
+#define V_IRQ_FIFO_BL3		0x08
+#define V_IRQ_FIFO_BL4		0x10
+#define V_IRQ_FIFO_BL5		0x20
+#define V_IRQ_FIFO_BL6		0x40
+#define V_IRQ_FIFO_BL7		0x80
+/* R_IRQ_MISC */
+#define V_STA_IRQ		0x01
+#define V_TI_IRQ		0x02
+#define V_IRQ_PROC		0x04
+#define V_DTMF_IRQ		0x08
+#define V_IRQ1S			0x10
+#define V_SA6_IRQ		0x20
+#define V_RX_EOMF		0x40
+#define V_TX_EOMF		0x80
+/* R_STATUS */
+#define V_BUSY			0x01
+#define V_PROC			0x02
+#define V_DTMF_STA		0x04
+#define V_LOST_STA		0x08
+#define V_SYNC_IN		0x10
+#define V_EXT_IRQSTA		0x20
+#define V_MISC_IRQSTA		0x40
+#define V_FR_IRQSTA		0x80
+/* R_IRQ_FIFO_BL0 */
+#define V_IRQ_FIFO0_TX		0x01
+#define V_IRQ_FIFO0_RX		0x02
+#define V_IRQ_FIFO1_TX		0x04
+#define V_IRQ_FIFO1_RX		0x08
+#define V_IRQ_FIFO2_TX		0x10
+#define V_IRQ_FIFO2_RX		0x20
+#define V_IRQ_FIFO3_TX		0x40
+#define V_IRQ_FIFO3_RX		0x80
+/* R_IRQ_FIFO_BL1 */
+#define V_IRQ_FIFO4_TX		0x01
+#define V_IRQ_FIFO4_RX		0x02
+#define V_IRQ_FIFO5_TX		0x04
+#define V_IRQ_FIFO5_RX		0x08
+#define V_IRQ_FIFO6_TX		0x10
+#define V_IRQ_FIFO6_RX		0x20
+#define V_IRQ_FIFO7_TX		0x40
+#define V_IRQ_FIFO7_RX		0x80
+/* R_IRQ_FIFO_BL2 */
+#define V_IRQ_FIFO8_TX		0x01
+#define V_IRQ_FIFO8_RX		0x02
+#define V_IRQ_FIFO9_TX		0x04
+#define V_IRQ_FIFO9_RX		0x08
+#define V_IRQ_FIFO10_TX		0x10
+#define V_IRQ_FIFO10_RX		0x20
+#define V_IRQ_FIFO11_TX		0x40
+#define V_IRQ_FIFO11_RX		0x80
+/* R_IRQ_FIFO_BL3 */
+#define V_IRQ_FIFO12_TX		0x01
+#define V_IRQ_FIFO12_RX		0x02
+#define V_IRQ_FIFO13_TX		0x04
+#define V_IRQ_FIFO13_RX		0x08
+#define V_IRQ_FIFO14_TX		0x10
+#define V_IRQ_FIFO14_RX		0x20
+#define V_IRQ_FIFO15_TX		0x40
+#define V_IRQ_FIFO15_RX		0x80
+/* R_IRQ_FIFO_BL4 */
+#define V_IRQ_FIFO16_TX		0x01
+#define V_IRQ_FIFO16_RX		0x02
+#define V_IRQ_FIFO17_TX		0x04
+#define V_IRQ_FIFO17_RX		0x08
+#define V_IRQ_FIFO18_TX		0x10
+#define V_IRQ_FIFO18_RX		0x20
+#define V_IRQ_FIFO19_TX		0x40
+#define V_IRQ_FIFO19_RX		0x80
+/* R_IRQ_FIFO_BL5 */
+#define V_IRQ_FIFO20_TX		0x01
+#define V_IRQ_FIFO20_RX		0x02
+#define V_IRQ_FIFO21_TX		0x04
+#define V_IRQ_FIFO21_RX		0x08
+#define V_IRQ_FIFO22_TX		0x10
+#define V_IRQ_FIFO22_RX		0x20
+#define V_IRQ_FIFO23_TX		0x40
+#define V_IRQ_FIFO23_RX		0x80
+/* R_IRQ_FIFO_BL6 */
+#define V_IRQ_FIFO24_TX		0x01
+#define V_IRQ_FIFO24_RX		0x02
+#define V_IRQ_FIFO25_TX		0x04
+#define V_IRQ_FIFO25_RX		0x08
+#define V_IRQ_FIFO26_TX		0x10
+#define V_IRQ_FIFO26_RX		0x20
+#define V_IRQ_FIFO27_TX		0x40
+#define V_IRQ_FIFO27_RX		0x80
+/* R_IRQ_FIFO_BL7 */
+#define V_IRQ_FIFO28_TX		0x01
+#define V_IRQ_FIFO28_RX		0x02
+#define V_IRQ_FIFO29_TX		0x04
+#define V_IRQ_FIFO29_RX		0x08
+#define V_IRQ_FIFO30_TX		0x10
+#define V_IRQ_FIFO30_RX		0x20
+#define V_IRQ_FIFO31_TX		0x40
+#define V_IRQ_FIFO31_RX		0x80
+
+/* chapter 13: general purpose I/O pins (GPIO) and input pins (GPI) */
+/* R_GPIO_OUT0 */
+#define V_GPIO_OUT0		0x01
+#define V_GPIO_OUT1		0x02
+#define V_GPIO_OUT2		0x04
+#define V_GPIO_OUT3		0x08
+#define V_GPIO_OUT4		0x10
+#define V_GPIO_OUT5		0x20
+#define V_GPIO_OUT6		0x40
+#define V_GPIO_OUT7		0x80
+/* R_GPIO_OUT1 */
+#define V_GPIO_OUT8		0x01
+#define V_GPIO_OUT9		0x02
+#define V_GPIO_OUT10		0x04
+#define V_GPIO_OUT11		0x08
+#define V_GPIO_OUT12		0x10
+#define V_GPIO_OUT13		0x20
+#define V_GPIO_OUT14		0x40
+#define V_GPIO_OUT15		0x80
+/* R_GPIO_EN0 */
+#define V_GPIO_EN0		0x01
+#define V_GPIO_EN1		0x02
+#define V_GPIO_EN2		0x04
+#define V_GPIO_EN3		0x08
+#define V_GPIO_EN4		0x10
+#define V_GPIO_EN5		0x20
+#define V_GPIO_EN6		0x40
+#define V_GPIO_EN7		0x80
+/* R_GPIO_EN1 */
+#define V_GPIO_EN8		0x01
+#define V_GPIO_EN9		0x02
+#define V_GPIO_EN10		0x04
+#define V_GPIO_EN11		0x08
+#define V_GPIO_EN12		0x10
+#define V_GPIO_EN13		0x20
+#define V_GPIO_EN14		0x40
+#define V_GPIO_EN15		0x80
+/* R_GPIO_SEL */
+#define V_GPIO_SEL0		0x01
+#define V_GPIO_SEL1		0x02
+#define V_GPIO_SEL2		0x04
+#define V_GPIO_SEL3		0x08
+#define V_GPIO_SEL4		0x10
+#define V_GPIO_SEL5		0x20
+#define V_GPIO_SEL6		0x40
+#define V_GPIO_SEL7		0x80
+/* R_GPIO_IN0 */
+#define V_GPIO_IN0		0x01
+#define V_GPIO_IN1		0x02
+#define V_GPIO_IN2		0x04
+#define V_GPIO_IN3		0x08
+#define V_GPIO_IN4		0x10
+#define V_GPIO_IN5		0x20
+#define V_GPIO_IN6		0x40
+#define V_GPIO_IN7		0x80
+/* R_GPIO_IN1 */
+#define V_GPIO_IN8		0x01
+#define V_GPIO_IN9		0x02
+#define V_GPIO_IN10		0x04
+#define V_GPIO_IN11		0x08
+#define V_GPIO_IN12		0x10
+#define V_GPIO_IN13		0x20
+#define V_GPIO_IN14		0x40
+#define V_GPIO_IN15		0x80
+/* R_GPI_IN0 */
+#define V_GPI_IN0		0x01
+#define V_GPI_IN1		0x02
+#define V_GPI_IN2		0x04
+#define V_GPI_IN3		0x08
+#define V_GPI_IN4		0x10
+#define V_GPI_IN5		0x20
+#define V_GPI_IN6		0x40
+#define V_GPI_IN7		0x80
+/* R_GPI_IN1 */
+#define V_GPI_IN8		0x01
+#define V_GPI_IN9		0x02
+#define V_GPI_IN10		0x04
+#define V_GPI_IN11		0x08
+#define V_GPI_IN12		0x10
+#define V_GPI_IN13		0x20
+#define V_GPI_IN14		0x40
+#define V_GPI_IN15		0x80
+/* R_GPI_IN2 */
+#define V_GPI_IN16		0x01
+#define V_GPI_IN17		0x02
+#define V_GPI_IN18		0x04
+#define V_GPI_IN19		0x08
+#define V_GPI_IN20		0x10
+#define V_GPI_IN21		0x20
+#define V_GPI_IN22		0x40
+#define V_GPI_IN23		0x80
+/* R_GPI_IN3 */
+#define V_GPI_IN24		0x01
+#define V_GPI_IN25		0x02
+#define V_GPI_IN26		0x04
+#define V_GPI_IN27		0x08
+#define V_GPI_IN28		0x10
+#define V_GPI_IN29		0x20
+#define V_GPI_IN30		0x40
+#define V_GPI_IN31		0x80
+
+/* map of all registers, used for debugging */
+
+#ifdef HFC_REGISTER_DEBUG
+struct hfc_register_names {
+	char *name;
+	u_char reg;
+} hfc_register_names[] = {
+	/* write registers */
+	{"R_CIRM",		0x00},
+	{"R_CTRL",		0x01},
+	{"R_BRG_PCM_CFG ",	0x02},
+	{"R_RAM_ADDR0",		0x08},
+	{"R_RAM_ADDR1",		0x09},
+	{"R_RAM_ADDR2",		0x0A},
+	{"R_FIRST_FIFO",	0x0B},
+	{"R_RAM_SZ",		0x0C},
+	{"R_FIFO_MD",		0x0D},
+	{"R_INC_RES_FIFO",	0x0E},
+	{"R_FIFO / R_FSM_IDX",	0x0F},
+	{"R_SLOT",		0x10},
+	{"R_IRQMSK_MISC",	0x11},
+	{"R_SCI_MSK",		0x12},
+	{"R_IRQ_CTRL",		0x13},
+	{"R_PCM_MD0",		0x14},
+	{"R_0x15",		0x15},
+	{"R_ST_SEL",		0x16},
+	{"R_ST_SYNC",		0x17},
+	{"R_CONF_EN",		0x18},
+	{"R_TI_WD",		0x1A},
+	{"R_BERT_WD_MD",	0x1B},
+	{"R_DTMF",		0x1C},
+	{"R_DTMF_N",		0x1D},
+	{"R_E1_XX_STA",		0x20},
+	{"R_LOS0",		0x22},
+	{"R_LOS1",		0x23},
+	{"R_RX0",		0x24},
+	{"R_RX_FR0",		0x25},
+	{"R_RX_FR1",		0x26},
+	{"R_TX0",		0x28},
+	{"R_TX1",		0x29},
+	{"R_TX_FR0",		0x2C},
+	{"R_TX_FR1",		0x2D},
+	{"R_TX_FR2",		0x2E},
+	{"R_JATT_ATT",		0x2F},
+	{"A_ST_xx_STA/R_RX_OFF", 0x30},
+	{"A_ST_CTRL0/R_SYNC_OUT", 0x31},
+	{"A_ST_CTRL1",		0x32},
+	{"A_ST_CTRL2",		0x33},
+	{"A_ST_SQ_WR",		0x34},
+	{"R_TX_OFF",		0x34},
+	{"R_SYNC_CTRL",		0x35},
+	{"A_ST_CLK_DLY",	0x37},
+	{"R_PWM0",		0x38},
+	{"R_PWM1",		0x39},
+	{"A_ST_B1_TX",		0x3C},
+	{"A_ST_B2_TX",		0x3D},
+	{"A_ST_D_TX",		0x3E},
+	{"R_GPIO_OUT0",		0x40},
+	{"R_GPIO_OUT1",		0x41},
+	{"R_GPIO_EN0",		0x42},
+	{"R_GPIO_EN1",		0x43},
+	{"R_GPIO_SEL",		0x44},
+	{"R_BRG_CTRL",		0x45},
+	{"R_PWM_MD",		0x46},
+	{"R_BRG_MD",		0x47},
+	{"R_BRG_TIM0",		0x48},
+	{"R_BRG_TIM1",		0x49},
+	{"R_BRG_TIM2",		0x4A},
+	{"R_BRG_TIM3",		0x4B},
+	{"R_BRG_TIM_SEL01",	0x4C},
+	{"R_BRG_TIM_SEL23",	0x4D},
+	{"R_BRG_TIM_SEL45",	0x4E},
+	{"R_BRG_TIM_SEL67",	0x4F},
+	{"A_FIFO_DATA0-2",	0x80},
+	{"A_FIFO_DATA0-2_NOINC", 0x84},
+	{"R_RAM_DATA",		0xC0},
+	{"A_SL_CFG",		0xD0},
+	{"A_CONF",		0xD1},
+	{"A_CH_MSK",		0xF4},
+	{"A_CON_HDLC",		0xFA},
+	{"A_SUBCH_CFG",		0xFB},
+	{"A_CHANNEL",		0xFC},
+	{"A_FIFO_SEQ",		0xFD},
+	{"A_IRQ_MSK",		0xFF},
+	{NULL, 0},
+
+	/* read registers */
+	{"A_Z1",		0x04},
+	{"A_Z1H",		0x05},
+	{"A_Z2",		0x06},
+	{"A_Z2H",		0x07},
+	{"A_F1",		0x0C},
+	{"A_F2",		0x0D},
+	{"R_IRQ_OVIEW",		0x10},
+	{"R_IRQ_MISC",		0x11},
+	{"R_IRQ_STATECH",	0x12},
+	{"R_CONF_OFLOW",	0x14},
+	{"R_RAM_USE",		0x15},
+	{"R_CHIP_ID",		0x16},
+	{"R_BERT_STA",		0x17},
+	{"R_F0_CNTL",		0x18},
+	{"R_F0_CNTH",		0x19},
+	{"R_BERT_ECL",		0x1A},
+	{"R_BERT_ECH",		0x1B},
+	{"R_STATUS",		0x1C},
+	{"R_CHIP_RV",		0x1F},
+	{"R_STATE",		0x20},
+	{"R_SYNC_STA",		0x24},
+	{"R_RX_SL0_0",		0x25},
+	{"R_RX_SL0_1",		0x26},
+	{"R_RX_SL0_2",		0x27},
+	{"R_JATT_DIR",		0x2b},
+	{"R_SLIP",		0x2c},
+	{"A_ST_RD_STA",		0x30},
+	{"R_FAS_ECL",		0x30},
+	{"R_FAS_ECH",		0x31},
+	{"R_VIO_ECL",		0x32},
+	{"R_VIO_ECH",		0x33},
+	{"R_CRC_ECL / A_ST_SQ_RD", 0x34},
+	{"R_CRC_ECH",		0x35},
+	{"R_E_ECL",		0x36},
+	{"R_E_ECH",		0x37},
+	{"R_SA6_SA13_ECL",	0x38},
+	{"R_SA6_SA13_ECH",	0x39},
+	{"R_SA6_SA23_ECL",	0x3A},
+	{"R_SA6_SA23_ECH",	0x3B},
+	{"A_ST_B1_RX",		0x3C},
+	{"A_ST_B2_RX",		0x3D},
+	{"A_ST_D_RX",		0x3E},
+	{"A_ST_E_RX",		0x3F},
+	{"R_GPIO_IN0",		0x40},
+	{"R_GPIO_IN1",		0x41},
+	{"R_GPI_IN0",		0x44},
+	{"R_GPI_IN1",		0x45},
+	{"R_GPI_IN2",		0x46},
+	{"R_GPI_IN3",		0x47},
+	{"A_FIFO_DATA0-2",	0x80},
+	{"A_FIFO_DATA0-2_NOINC", 0x84},
+	{"R_INT_DATA",		0x88},
+	{"R_RAM_DATA",		0xC0},
+	{"R_IRQ_FIFO_BL0",	0xC8},
+	{"R_IRQ_FIFO_BL1",	0xC9},
+	{"R_IRQ_FIFO_BL2",	0xCA},
+	{"R_IRQ_FIFO_BL3",	0xCB},
+	{"R_IRQ_FIFO_BL4",	0xCC},
+	{"R_IRQ_FIFO_BL5",	0xCD},
+	{"R_IRQ_FIFO_BL6",	0xCE},
+	{"R_IRQ_FIFO_BL7",	0xCF},
+};
+#endif /* HFC_REGISTER_DEBUG */
+
diff --git a/drivers/isdn/hardware/mISDN/hfcmulti.c b/drivers/isdn/hardware/mISDN/hfcmulti.c
new file mode 100644
index 000000000000..2649ea55a9e8
--- /dev/null
+++ b/drivers/isdn/hardware/mISDN/hfcmulti.c
@@ -0,0 +1,5320 @@
+/*
+ * hfcmulti.c  low level driver for hfc-4s/hfc-8s/hfc-e1 based cards
+ *
+ * Author	Andreas Eversberg (jolly@eversberg.eu)
+ * ported to mqueue mechanism:
+ *		Peter Sprenger (sprengermoving-bytes.de)
+ *
+ * inspired by existing hfc-pci driver:
+ * Copyright 1999  by Werner Cornelius (werner@isdn-development.de)
+ * Copyright 2008  by Karsten Keil (kkeil@suse.de)
+ * Copyright 2008  by Andreas Eversberg (jolly@eversberg.eu)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2, or (at your option)
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ *
+ *
+ * Thanks to Cologne Chip AG for this great controller!
+ */
+
+/*
+ * module parameters:
+ * type:
+ *	By default (0), the card is automatically detected.
+ *	Or use the following combinations:
+ *	Bit 0-7   = 0x00001 = HFC-E1 (1 port)
+ * or	Bit 0-7   = 0x00004 = HFC-4S (4 ports)
+ * or	Bit 0-7   = 0x00008 = HFC-8S (8 ports)
+ *	Bit 8     = 0x00100 = uLaw (instead of aLaw)
+ *	Bit 9     = 0x00200 = Disable DTMF detect on all B-channels via hardware
+ *	Bit 10    = spare
+ *	Bit 11    = 0x00800 = Force PCM bus into slave mode. (otherwhise auto)
+ * or   Bit 12    = 0x01000 = Force PCM bus into master mode. (otherwhise auto)
+ *	Bit 13	  = spare
+ *	Bit 14    = 0x04000 = Use external ram (128K)
+ *	Bit 15    = 0x08000 = Use external ram (512K)
+ *	Bit 16    = 0x10000 = Use 64 timeslots instead of 32
+ * or	Bit 17    = 0x20000 = Use 128 timeslots instead of anything else
+ *	Bit 18    = spare
+ *	Bit 19    = 0x80000 = Send the Watchdog a Signal (Dual E1 with Watchdog)
+ * (all other bits are reserved and shall be 0)
+ *	example: 0x20204 one HFC-4S with dtmf detection and 128 timeslots on PCM
+ *		 bus (PCM master)
+ *
+ * port: (optional or required for all ports on all installed cards)
+ *	HFC-4S/HFC-8S only bits:
+ *	Bit 0	  = 0x001 = Use master clock for this S/T interface
+ *			    (ony once per chip).
+ *	Bit 1     = 0x002 = transmitter line setup (non capacitive mode)
+ *			    Don't use this unless you know what you are doing!
+ *	Bit 2     = 0x004 = Disable E-channel. (No E-channel processing)
+ *	example: 0x0001,0x0000,0x0000,0x0000 one HFC-4S with master clock
+ *		 received from port 1
+ *
+ *	HFC-E1 only bits:
+ *	Bit 0     = 0x0001 = interface: 0=copper, 1=optical
+ *	Bit 1     = 0x0002 = reserved (later for 32 B-channels transparent mode)
+ *	Bit 2     = 0x0004 = Report LOS
+ *	Bit 3     = 0x0008 = Report AIS
+ *	Bit 4     = 0x0010 = Report SLIP
+ *	Bit 5     = 0x0020 = Report RDI
+ *	Bit 8     = 0x0100 = Turn off CRC-4 Multiframe Mode, use double frame
+ *			     mode instead.
+ *	Bit 9	  = 0x0200 = Force get clock from interface, even in NT mode.
+ * or	Bit 10	  = 0x0400 = Force put clock to interface, even in TE mode.
+ *	Bit 11    = 0x0800 = Use direct RX clock for PCM sync rather than PLL.
+ *			     (E1 only)
+ *	Bit 12-13 = 0xX000 = elastic jitter buffer (1-3), Set both bits to 0
+ *			     for default.
+ * (all other bits are reserved and shall be 0)
+ *
+ * debug:
+ *	NOTE: only one debug value must be given for all cards
+ *	enable debugging (see hfc_multi.h for debug options)
+ *
+ * poll:
+ *	NOTE: only one poll value must be given for all cards
+ *	Give the number of samples for each fifo process.
+ *	By default 128 is used. Decrease to reduce delay, increase to
+ *	reduce cpu load. If unsure, don't mess with it!
+ *	Valid is 8, 16, 32, 64, 128, 256.
+ *
+ * pcm:
+ *	NOTE: only one pcm value must be given for every card.
+ *	The PCM bus id tells the mISDNdsp module about the connected PCM bus.
+ *	By default (0), the PCM bus id is 100 for the card that is PCM master.
+ *	If multiple cards are PCM master (because they are not interconnected),
+ *	each card with PCM master will have increasing PCM id.
+ *	All PCM busses with the same ID are expected to be connected and have
+ *	common time slots slots.
+ *	Only one chip of the PCM bus must be master, the others slave.
+ *	-1 means no support of PCM bus not even.
+ *	Omit this value, if all cards are interconnected or none is connected.
+ *	If unsure, don't give this parameter.
+ *
+ * dslot:
+ *	NOTE: only one poll value must be given for every card.
+ *	Also this value must be given for non-E1 cards. If omitted, the E1
+ *	card has D-channel on time slot 16, which is default.
+ *	If 1..15 or 17..31, an alternate time slot is used for D-channel.
+ *	In this case, the application must be able to handle this.
+ *	If -1 is given, the D-channel is disabled and all 31 slots can be used
+ *	for B-channel. (only for specific applications)
+ *	If you don't know how to use it, you don't need it!
+ *
+ * iomode:
+ *	NOTE: only one mode value must be given for every card.
+ *	-> See hfc_multi.h for HFC_IO_MODE_* values
+ *	By default, the IO mode is pci memory IO (MEMIO).
+ *	Some cards requre specific IO mode, so it cannot be changed.
+ *	It may be usefull to set IO mode to register io (REGIO) to solve
+ *	PCI bridge problems.
+ *	If unsure, don't give this parameter.
+ *
+ * clockdelay_nt:
+ *	NOTE: only one clockdelay_nt value must be given once for all cards.
+ *	Give the value of the clock control register (A_ST_CLK_DLY)
+ *	of the S/T interfaces in NT mode.
+ *	This register is needed for the TBR3 certification, so don't change it.
+ *
+ * clockdelay_te:
+ *	NOTE: only one clockdelay_te value must be given once
+ *	Give the value of the clock control register (A_ST_CLK_DLY)
+ *	of the S/T interfaces in TE mode.
+ *	This register is needed for the TBR3 certification, so don't change it.
+ */
+
+/*
+ * debug register access (never use this, it will flood your system log)
+ * #define HFC_REGISTER_DEBUG
+ */
+
+static const char *hfcmulti_revision = "2.00";
+
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include <linux/mISDNhw.h>
+#include <linux/mISDNdsp.h>
+
+/*
+#define IRQCOUNT_DEBUG
+#define IRQ_DEBUG
+*/
+
+#include "hfc_multi.h"
+#ifdef ECHOPREP
+#include "gaintab.h"
+#endif
+
+#define	MAX_CARDS	8
+#define	MAX_PORTS	(8 * MAX_CARDS)
+
+static LIST_HEAD(HFClist);
+static spinlock_t HFClock; /* global hfc list lock */
+
+static void ph_state_change(struct dchannel *);
+static void (*hfc_interrupt)(void);
+static void (*register_interrupt)(void);
+static int (*unregister_interrupt)(void);
+static int interrupt_registered;
+
+static struct hfc_multi *syncmaster;
+int plxsd_master; /* if we have a master card (yet) */
+static spinlock_t plx_lock; /* may not acquire other lock inside */
+EXPORT_SYMBOL(plx_lock);
+
+#define	TYP_E1		1
+#define	TYP_4S		4
+#define TYP_8S		8
+
+static int poll_timer = 6;	/* default = 128 samples = 16ms */
+/* number of POLL_TIMER interrupts for G2 timeout (ca 1s) */
+static int nt_t1_count[] = { 3840, 1920, 960, 480, 240, 120, 60, 30  };
+#define	CLKDEL_TE	0x0f	/* CLKDEL in TE mode */
+#define	CLKDEL_NT	0x6c	/* CLKDEL in NT mode
+				   (0x60 MUST be included!) */
+static u_char silence =	0xff;	/* silence by LAW */
+
+#define	DIP_4S	0x1		/* DIP Switches for Beronet 1S/2S/4S cards */
+#define	DIP_8S	0x2		/* DIP Switches for Beronet 8S+ cards */
+#define	DIP_E1	0x3		/* DIP Switches for Beronet E1 cards */
+
+/*
+ * module stuff
+ */
+
+static uint	type[MAX_CARDS];
+static uint	pcm[MAX_CARDS];
+static uint	dslot[MAX_CARDS];
+static uint	iomode[MAX_CARDS];
+static uint	port[MAX_PORTS];
+static uint	debug;
+static uint	poll;
+static uint	timer;
+static uint	clockdelay_te = CLKDEL_TE;
+static uint	clockdelay_nt = CLKDEL_NT;
+
+static int	HFC_cnt, Port_cnt, PCM_cnt = 99;
+
+MODULE_AUTHOR("Andreas Eversberg");
+MODULE_LICENSE("GPL");
+module_param(debug, uint, S_IRUGO | S_IWUSR);
+module_param(poll, uint, S_IRUGO | S_IWUSR);
+module_param(timer, uint, S_IRUGO | S_IWUSR);
+module_param(clockdelay_te, uint, S_IRUGO | S_IWUSR);
+module_param(clockdelay_nt, uint, S_IRUGO | S_IWUSR);
+module_param_array(type, uint, NULL, S_IRUGO | S_IWUSR);
+module_param_array(pcm, uint, NULL, S_IRUGO | S_IWUSR);
+module_param_array(dslot, uint, NULL, S_IRUGO | S_IWUSR);
+module_param_array(iomode, uint, NULL, S_IRUGO | S_IWUSR);
+module_param_array(port, uint, NULL, S_IRUGO | S_IWUSR);
+
+#ifdef HFC_REGISTER_DEBUG
+#define HFC_outb(hc, reg, val) \
+	(hc->HFC_outb(hc, reg, val, __func__, __LINE__))
+#define HFC_outb_nodebug(hc, reg, val) \
+	(hc->HFC_outb_nodebug(hc, reg, val, __func__, __LINE__))
+#define HFC_inb(hc, reg) \
+	(hc->HFC_inb(hc, reg, __func__, __LINE__))
+#define HFC_inb_nodebug(hc, reg) \
+	(hc->HFC_inb_nodebug(hc, reg, __func__, __LINE__))
+#define HFC_inw(hc, reg) \
+	(hc->HFC_inw(hc, reg, __func__, __LINE__))
+#define HFC_inw_nodebug(hc, reg) \
+	(hc->HFC_inw_nodebug(hc, reg, __func__, __LINE__))
+#define HFC_wait(hc) \
+	(hc->HFC_wait(hc, __func__, __LINE__))
+#define HFC_wait_nodebug(hc) \
+	(hc->HFC_wait_nodebug(hc, __func__, __LINE__))
+#else
+#define HFC_outb(hc, reg, val)		(hc->HFC_outb(hc, reg, val))
+#define HFC_outb_nodebug(hc, reg, val)	(hc->HFC_outb_nodebug(hc, reg, val))
+#define HFC_inb(hc, reg)		(hc->HFC_inb(hc, reg))
+#define HFC_inb_nodebug(hc, reg)	(hc->HFC_inb_nodebug(hc, reg))
+#define HFC_inw(hc, reg)		(hc->HFC_inw(hc, reg))
+#define HFC_inw_nodebug(hc, reg)	(hc->HFC_inw_nodebug(hc, reg))
+#define HFC_wait(hc)			(hc->HFC_wait(hc))
+#define HFC_wait_nodebug(hc)		(hc->HFC_wait_nodebug(hc))
+#endif
+
+/* HFC_IO_MODE_PCIMEM */
+static void
+#ifdef HFC_REGISTER_DEBUG
+HFC_outb_pcimem(struct hfc_multi *hc, u_char reg, u_char val,
+		const char *function, int line)
+#else
+HFC_outb_pcimem(struct hfc_multi *hc, u_char reg, u_char val)
+#endif
+{
+	writeb(val, (hc->pci_membase)+reg);
+}
+static u_char
+#ifdef HFC_REGISTER_DEBUG
+HFC_inb_pcimem(struct hfc_multi *hc, u_char reg, const char *function, int line)
+#else
+HFC_inb_pcimem(struct hfc_multi *hc, u_char reg)
+#endif
+{
+	return readb((hc->pci_membase)+reg);
+}
+static u_short
+#ifdef HFC_REGISTER_DEBUG
+HFC_inw_pcimem(struct hfc_multi *hc, u_char reg, const char *function, int line)
+#else
+HFC_inw_pcimem(struct hfc_multi *hc, u_char reg)
+#endif
+{
+	return readw((hc->pci_membase)+reg);
+}
+static void
+#ifdef HFC_REGISTER_DEBUG
+HFC_wait_pcimem(struct hfc_multi *hc, const char *function, int line)
+#else
+HFC_wait_pcimem(struct hfc_multi *hc)
+#endif
+{
+	while (readb((hc->pci_membase)+R_STATUS) & V_BUSY);
+}
+
+/* HFC_IO_MODE_REGIO */
+static void
+#ifdef HFC_REGISTER_DEBUG
+HFC_outb_regio(struct hfc_multi *hc, u_char reg, u_char val,
+	const char *function, int line)
+#else
+HFC_outb_regio(struct hfc_multi *hc, u_char reg, u_char val)
+#endif
+{
+	outb(reg, (hc->pci_iobase)+4);
+	outb(val, hc->pci_iobase);
+}
+static u_char
+#ifdef HFC_REGISTER_DEBUG
+HFC_inb_regio(struct hfc_multi *hc, u_char reg, const char *function, int line)
+#else
+HFC_inb_regio(struct hfc_multi *hc, u_char reg)
+#endif
+{
+	outb(reg, (hc->pci_iobase)+4);
+	return inb(hc->pci_iobase);
+}
+static u_short
+#ifdef HFC_REGISTER_DEBUG
+HFC_inw_regio(struct hfc_multi *hc, u_char reg, const char *function, int line)
+#else
+HFC_inw_regio(struct hfc_multi *hc, u_char reg)
+#endif
+{
+	outb(reg, (hc->pci_iobase)+4);
+	return inw(hc->pci_iobase);
+}
+static void
+#ifdef HFC_REGISTER_DEBUG
+HFC_wait_regio(struct hfc_multi *hc, const char *function, int line)
+#else
+HFC_wait_regio(struct hfc_multi *hc)
+#endif
+{
+	outb(R_STATUS, (hc->pci_iobase)+4);
+	while (inb(hc->pci_iobase) & V_BUSY);
+}
+
+#ifdef HFC_REGISTER_DEBUG
+static void
+HFC_outb_debug(struct hfc_multi *hc, u_char reg, u_char val,
+		const char *function, int line)
+{
+	char regname[256] = "", bits[9] = "xxxxxxxx";
+	int i;
+
+	i = -1;
+	while (hfc_register_names[++i].name) {
+		if (hfc_register_names[i].reg == reg)
+			strcat(regname, hfc_register_names[i].name);
+	}
+	if (regname[0] == '\0')
+		strcpy(regname, "register");
+
+	bits[7] = '0'+(!!(val&1));
+	bits[6] = '0'+(!!(val&2));
+	bits[5] = '0'+(!!(val&4));
+	bits[4] = '0'+(!!(val&8));
+	bits[3] = '0'+(!!(val&16));
+	bits[2] = '0'+(!!(val&32));
+	bits[1] = '0'+(!!(val&64));
+	bits[0] = '0'+(!!(val&128));
+	printk(KERN_DEBUG
+	    "HFC_outb(chip %d, %02x=%s, 0x%02x=%s); in %s() line %d\n",
+	    hc->id, reg, regname, val, bits, function, line);
+	HFC_outb_nodebug(hc, reg, val);
+}
+static u_char
+HFC_inb_debug(struct hfc_multi *hc, u_char reg, const char *function, int line)
+{
+	char regname[256] = "", bits[9] = "xxxxxxxx";
+	u_char val = HFC_inb_nodebug(hc, reg);
+	int i;
+
+	i = 0;
+	while (hfc_register_names[i++].name)
+		;
+	while (hfc_register_names[++i].name) {
+		if (hfc_register_names[i].reg == reg)
+			strcat(regname, hfc_register_names[i].name);
+	}
+	if (regname[0] == '\0')
+		strcpy(regname, "register");
+
+	bits[7] = '0'+(!!(val&1));
+	bits[6] = '0'+(!!(val&2));
+	bits[5] = '0'+(!!(val&4));
+	bits[4] = '0'+(!!(val&8));
+	bits[3] = '0'+(!!(val&16));
+	bits[2] = '0'+(!!(val&32));
+	bits[1] = '0'+(!!(val&64));
+	bits[0] = '0'+(!!(val&128));
+	printk(KERN_DEBUG
+	    "HFC_inb(chip %d, %02x=%s) = 0x%02x=%s; in %s() line %d\n",
+	    hc->id, reg, regname, val, bits, function, line);
+	return val;
+}
+static u_short
+HFC_inw_debug(struct hfc_multi *hc, u_char reg, const char *function, int line)
+{
+	char regname[256] = "";
+	u_short val = HFC_inw_nodebug(hc, reg);
+	int i;
+
+	i = 0;
+	while (hfc_register_names[i++].name)
+		;
+	while (hfc_register_names[++i].name) {
+		if (hfc_register_names[i].reg == reg)
+			strcat(regname, hfc_register_names[i].name);
+	}
+	if (regname[0] == '\0')
+		strcpy(regname, "register");
+
+	printk(KERN_DEBUG
+	    "HFC_inw(chip %d, %02x=%s) = 0x%04x; in %s() line %d\n",
+	    hc->id, reg, regname, val, function, line);
+	return val;
+}
+static void
+HFC_wait_debug(struct hfc_multi *hc, const char *function, int line)
+{
+	printk(KERN_DEBUG "HFC_wait(chip %d); in %s() line %d\n",
+	    hc->id, function, line);
+	HFC_wait_nodebug(hc);
+}
+#endif
+
+/* write fifo data (REGIO) */
+void
+write_fifo_regio(struct hfc_multi *hc, u_char *data, int len)
+{
+	outb(A_FIFO_DATA0, (hc->pci_iobase)+4);
+	while (len>>2) {
+		outl(*(u32 *)data, hc->pci_iobase);
+		data += 4;
+		len -= 4;
+	}
+	while (len>>1) {
+		outw(*(u16 *)data, hc->pci_iobase);
+		data += 2;
+		len -= 2;
+	}
+	while (len) {
+		outb(*data, hc->pci_iobase);
+		data++;
+		len--;
+	}
+}
+/* write fifo data (PCIMEM) */
+void
+write_fifo_pcimem(struct hfc_multi *hc, u_char *data, int len)
+{
+	while (len>>2) {
+		writel(*(u32 *)data, (hc->pci_membase)+A_FIFO_DATA0);
+		data += 4;
+		len -= 4;
+	}
+	while (len>>1) {
+		writew(*(u16 *)data, (hc->pci_membase)+A_FIFO_DATA0);
+		data += 2;
+		len -= 2;
+	}
+	while (len) {
+		writeb(*data, (hc->pci_membase)+A_FIFO_DATA0);
+		data++;
+		len--;
+	}
+}
+/* read fifo data (REGIO) */
+void
+read_fifo_regio(struct hfc_multi *hc, u_char *data, int len)
+{
+	outb(A_FIFO_DATA0, (hc->pci_iobase)+4);
+	while (len>>2) {
+		*(u32 *)data = inl(hc->pci_iobase);
+		data += 4;
+		len -= 4;
+	}
+	while (len>>1) {
+		*(u16 *)data = inw(hc->pci_iobase);
+		data += 2;
+		len -= 2;
+	}
+	while (len) {
+		*data = inb(hc->pci_iobase);
+		data++;
+		len--;
+	}
+}
+
+/* read fifo data (PCIMEM) */
+void
+read_fifo_pcimem(struct hfc_multi *hc, u_char *data, int len)
+{
+	while (len>>2) {
+		*(u32 *)data =
+			readl((hc->pci_membase)+A_FIFO_DATA0);
+		data += 4;
+		len -= 4;
+	}
+	while (len>>1) {
+		*(u16 *)data =
+			readw((hc->pci_membase)+A_FIFO_DATA0);
+		data += 2;
+		len -= 2;
+	}
+	while (len) {
+		*data = readb((hc->pci_membase)+A_FIFO_DATA0);
+		data++;
+		len--;
+	}
+}
+
+
+static void
+enable_hwirq(struct hfc_multi *hc)
+{
+	hc->hw.r_irq_ctrl |= V_GLOB_IRQ_EN;
+	HFC_outb(hc, R_IRQ_CTRL, hc->hw.r_irq_ctrl);
+}
+
+static void
+disable_hwirq(struct hfc_multi *hc)
+{
+	hc->hw.r_irq_ctrl &= ~((u_char)V_GLOB_IRQ_EN);
+	HFC_outb(hc, R_IRQ_CTRL, hc->hw.r_irq_ctrl);
+}
+
+#define	NUM_EC 2
+#define	MAX_TDM_CHAN 32
+
+
+inline void
+enablepcibridge(struct hfc_multi *c)
+{
+	HFC_outb(c, R_BRG_PCM_CFG, (0x0 << 6) | 0x3); /* was _io before */
+}
+
+inline void
+disablepcibridge(struct hfc_multi *c)
+{
+	HFC_outb(c, R_BRG_PCM_CFG, (0x0 << 6) | 0x2); /* was _io before */
+}
+
+inline unsigned char
+readpcibridge(struct hfc_multi *hc, unsigned char address)
+{
+	unsigned short cipv;
+	unsigned char data;
+
+	if (!hc->pci_iobase)
+		return 0;
+
+	/* slow down a PCI read access by 1 PCI clock cycle */
+	HFC_outb(hc, R_CTRL, 0x4); /*was _io before*/
+
+	if (address == 0)
+		cipv = 0x4000;
+	else
+		cipv = 0x5800;
+
+	/* select local bridge port address by writing to CIP port */
+	/* data = HFC_inb(c, cipv); * was _io before */
+	outw(cipv, hc->pci_iobase + 4);
+	data = inb(hc->pci_iobase);
+
+	/* restore R_CTRL for normal PCI read cycle speed */
+	HFC_outb(hc, R_CTRL, 0x0); /* was _io before */
+
+	return data;
+}
+
+inline void
+writepcibridge(struct hfc_multi *hc, unsigned char address, unsigned char data)
+{
+	unsigned short cipv;
+	unsigned int datav;
+
+	if (!hc->pci_iobase)
+		return;
+
+	if (address == 0)
+		cipv = 0x4000;
+	else
+		cipv = 0x5800;
+
+	/* select local bridge port address by writing to CIP port */
+	outw(cipv, hc->pci_iobase + 4);
+	/* define a 32 bit dword with 4 identical bytes for write sequence */
+	datav = data | ((__u32) data << 8) | ((__u32) data << 16) |
+	    ((__u32) data << 24);
+
+	/*
+	 * write this 32 bit dword to the bridge data port
+	 * this will initiate a write sequence of up to 4 writes to the same
+	 * address on the local bus interface the number of write accesses
+	 * is undefined but >=1 and depends on the next PCI transaction
+	 * during write sequence on the local bus
+	 */
+	outl(datav, hc->pci_iobase);
+}
+
+inline void
+cpld_set_reg(struct hfc_multi *hc, unsigned char reg)
+{
+	/* Do data pin read low byte */
+	HFC_outb(hc, R_GPIO_OUT1, reg);
+}
+
+inline void
+cpld_write_reg(struct hfc_multi *hc, unsigned char reg, unsigned char val)
+{
+	cpld_set_reg(hc, reg);
+
+	enablepcibridge(hc);
+	writepcibridge(hc, 1, val);
+	disablepcibridge(hc);
+
+	return;
+}
+
+inline unsigned char
+cpld_read_reg(struct hfc_multi *hc, unsigned char reg)
+{
+	unsigned char bytein;
+
+	cpld_set_reg(hc, reg);
+
+	/* Do data pin read low byte */
+	HFC_outb(hc, R_GPIO_OUT1, reg);
+
+	enablepcibridge(hc);
+	bytein = readpcibridge(hc, 1);
+	disablepcibridge(hc);
+
+	return bytein;
+}
+
+inline void
+vpm_write_address(struct hfc_multi *hc, unsigned short addr)
+{
+	cpld_write_reg(hc, 0, 0xff & addr);
+	cpld_write_reg(hc, 1, 0x01 & (addr >> 8));
+}
+
+inline unsigned short
+vpm_read_address(struct hfc_multi *c)
+{
+	unsigned short addr;
+	unsigned short highbit;
+
+	addr = cpld_read_reg(c, 0);
+	highbit = cpld_read_reg(c, 1);
+
+	addr = addr | (highbit << 8);
+
+	return addr & 0x1ff;
+}
+
+inline unsigned char
+vpm_in(struct hfc_multi *c, int which, unsigned short addr)
+{
+	unsigned char res;
+
+	vpm_write_address(c, addr);
+
+	if (!which)
+		cpld_set_reg(c, 2);
+	else
+		cpld_set_reg(c, 3);
+
+	enablepcibridge(c);
+	res = readpcibridge(c, 1);
+	disablepcibridge(c);
+
+	cpld_set_reg(c, 0);
+
+	return res;
+}
+
+inline void
+vpm_out(struct hfc_multi *c, int which, unsigned short addr,
+    unsigned char data)
+{
+	vpm_write_address(c, addr);
+
+	enablepcibridge(c);
+
+	if (!which)
+		cpld_set_reg(c, 2);
+	else
+		cpld_set_reg(c, 3);
+
+	writepcibridge(c, 1, data);
+
+	cpld_set_reg(c, 0);
+
+	disablepcibridge(c);
+
+	{
+	unsigned char regin;
+	regin = vpm_in(c, which, addr);
+	if (regin != data)
+		printk(KERN_DEBUG "Wrote 0x%x to register 0x%x but got back "
+			"0x%x\n", data, addr, regin);
+	}
+
+}
+
+
+void
+vpm_init(struct hfc_multi *wc)
+{
+	unsigned char reg;
+	unsigned int mask;
+	unsigned int i, x, y;
+	unsigned int ver;
+
+	for (x = 0; x < NUM_EC; x++) {
+		/* Setup GPIO's */
+		if (!x) {
+			ver = vpm_in(wc, x, 0x1a0);
+			printk(KERN_DEBUG "VPM: Chip %d: ver %02x\n", x, ver);
+		}
+
+		for (y = 0; y < 4; y++) {
+			vpm_out(wc, x, 0x1a8 + y, 0x00); /* GPIO out */
+			vpm_out(wc, x, 0x1ac + y, 0x00); /* GPIO dir */
+			vpm_out(wc, x, 0x1b0 + y, 0x00); /* GPIO sel */
+		}
+
+		/* Setup TDM path - sets fsync and tdm_clk as inputs */
+		reg = vpm_in(wc, x, 0x1a3); /* misc_con */
+		vpm_out(wc, x, 0x1a3, reg & ~2);
+
+		/* Setup Echo length (256 taps) */
+		vpm_out(wc, x, 0x022, 1);
+		vpm_out(wc, x, 0x023, 0xff);
+
+		/* Setup timeslots */
+		vpm_out(wc, x, 0x02f, 0x00);
+		mask = 0x02020202 << (x * 4);
+
+		/* Setup the tdm channel masks for all chips */
+		for (i = 0; i < 4; i++)
+			vpm_out(wc, x, 0x33 - i, (mask >> (i << 3)) & 0xff);
+
+		/* Setup convergence rate */
+		printk(KERN_DEBUG "VPM: A-law mode\n");
+		reg = 0x00 | 0x10 | 0x01;
+		vpm_out(wc, x, 0x20, reg);
+		printk(KERN_DEBUG "VPM reg 0x20 is %x\n", reg);
+		/*vpm_out(wc, x, 0x20, (0x00 | 0x08 | 0x20 | 0x10)); */
+
+		vpm_out(wc, x, 0x24, 0x02);
+		reg = vpm_in(wc, x, 0x24);
+		printk(KERN_DEBUG "NLP Thresh is set to %d (0x%x)\n", reg, reg);
+
+		/* Initialize echo cans */
+		for (i = 0; i < MAX_TDM_CHAN; i++) {
+			if (mask & (0x00000001 << i))
+				vpm_out(wc, x, i, 0x00);
+		}
+
+		/*
+		 * ARM arch at least disallows a udelay of
+		 * more than 2ms... it gives a fake "__bad_udelay"
+		 * reference at link-time.
+		 * long delays in kernel code are pretty sucky anyway
+		 * for now work around it using 5 x 2ms instead of 1 x 10ms
+		 */
+
+		udelay(2000);
+		udelay(2000);
+		udelay(2000);
+		udelay(2000);
+		udelay(2000);
+
+		/* Put in bypass mode */
+		for (i = 0; i < MAX_TDM_CHAN; i++) {
+			if (mask & (0x00000001 << i))
+				vpm_out(wc, x, i, 0x01);
+		}
+
+		/* Enable bypass */
+		for (i = 0; i < MAX_TDM_CHAN; i++) {
+			if (mask & (0x00000001 << i))
+				vpm_out(wc, x, 0x78 + i, 0x01);
+		}
+
+	}
+}
+
+void
+vpm_check(struct hfc_multi *hctmp)
+{
+	unsigned char gpi2;
+
+	gpi2 = HFC_inb(hctmp, R_GPI_IN2);
+
+	if ((gpi2 & 0x3) != 0x3)
+		printk(KERN_DEBUG "Got interrupt 0x%x from VPM!\n", gpi2);
+}
+
+
+/*
+ * Interface to enable/disable the HW Echocan
+ *
+ * these functions are called within a spin_lock_irqsave on
+ * the channel instance lock, so we are not disturbed by irqs
+ *
+ * we can later easily change the interface to make  other
+ * things configurable, for now we configure the taps
+ *
+ */
+
+void
+vpm_echocan_on(struct hfc_multi *hc, int ch, int taps)
+{
+	unsigned int timeslot;
+	unsigned int unit;
+	struct bchannel *bch = hc->chan[ch].bch;
+#ifdef TXADJ
+	int txadj = -4;
+	struct sk_buff *skb;
+#endif
+	if (hc->chan[ch].protocol != ISDN_P_B_RAW)
+		return;
+
+	if (!bch)
+		return;
+
+#ifdef TXADJ
+	skb = _alloc_mISDN_skb(PH_CONTROL_IND, HFC_VOL_CHANGE_TX,
+		sizeof(int), &txadj, GFP_ATOMIC);
+	if (skb)
+		recv_Bchannel_skb(bch, skb);
+#endif
+
+	timeslot = ((ch/4)*8) + ((ch%4)*4) + 1;
+	unit = ch % 4;
+
+	printk(KERN_NOTICE "vpm_echocan_on called taps [%d] on timeslot %d\n",
+	    taps, timeslot);
+
+	vpm_out(hc, unit, timeslot, 0x7e);
+}
+
+void
+vpm_echocan_off(struct hfc_multi *hc, int ch)
+{
+	unsigned int timeslot;
+	unsigned int unit;
+	struct bchannel *bch = hc->chan[ch].bch;
+#ifdef TXADJ
+	int txadj = 0;
+	struct sk_buff *skb;
+#endif
+
+	if (hc->chan[ch].protocol != ISDN_P_B_RAW)
+		return;
+
+	if (!bch)
+		return;
+
+#ifdef TXADJ
+	skb = _alloc_mISDN_skb(PH_CONTROL_IND, HFC_VOL_CHANGE_TX,
+		sizeof(int), &txadj, GFP_ATOMIC);
+	if (skb)
+		recv_Bchannel_skb(bch, skb);
+#endif
+
+	timeslot = ((ch/4)*8) + ((ch%4)*4) + 1;
+	unit = ch % 4;
+
+	printk(KERN_NOTICE "vpm_echocan_off called on timeslot %d\n",
+	    timeslot);
+	/* FILLME */
+	vpm_out(hc, unit, timeslot, 0x01);
+}
+
+
+/*
+ * Speech Design resync feature
+ * NOTE: This is called sometimes outside interrupt handler.
+ * We must lock irqsave, so no other interrupt (other card) will occurr!
+ * Also multiple interrupts may nest, so must lock each access (lists, card)!
+ */
+static inline void
+hfcmulti_resync(struct hfc_multi *locked, struct hfc_multi *newmaster, int rm)
+{
+	struct hfc_multi *hc, *next, *pcmmaster = 0;
+	u_int *plx_acc_32, pv;
+	u_long flags;
+
+	spin_lock_irqsave(&HFClock, flags);
+	spin_lock(&plx_lock); /* must be locked inside other locks */
+
+	if (debug & DEBUG_HFCMULTI_PLXSD)
+		printk(KERN_DEBUG "%s: RESYNC(syncmaster=0x%p)\n",
+			__func__, syncmaster);
+
+	/* select new master */
+	if (newmaster) {
+		if (debug & DEBUG_HFCMULTI_PLXSD)
+			printk(KERN_DEBUG "using provided controller\n");
+	} else {
+		list_for_each_entry_safe(hc, next, &HFClist, list) {
+			if (test_bit(HFC_CHIP_PLXSD, &hc->chip)) {
+				if (hc->syncronized) {
+					newmaster = hc;
+					break;
+				}
+			}
+		}
+	}
+
+	/* Disable sync of all cards */
+	list_for_each_entry_safe(hc, next, &HFClist, list) {
+		if (test_bit(HFC_CHIP_PLXSD, &hc->chip)) {
+			plx_acc_32 = (u_int *)(hc->plx_membase+PLX_GPIOC);
+			pv = readl(plx_acc_32);
+			pv &= ~PLX_SYNC_O_EN;
+			writel(pv, plx_acc_32);
+			if (test_bit(HFC_CHIP_PCM_MASTER, &hc->chip)) {
+				pcmmaster = hc;
+				if (hc->type == 1) {
+					if (debug & DEBUG_HFCMULTI_PLXSD)
+						printk(KERN_DEBUG
+							"Schedule SYNC_I\n");
+					hc->e1_resync |= 1; /* get SYNC_I */
+				}
+			}
+		}
+	}
+
+	if (newmaster) {
+		hc = newmaster;
+		if (debug & DEBUG_HFCMULTI_PLXSD)
+			printk(KERN_DEBUG "id=%d (0x%p) = syncronized with "
+				"interface.\n", hc->id, hc);
+		/* Enable new sync master */
+		plx_acc_32 = (u_int *)(hc->plx_membase+PLX_GPIOC);
+		pv = readl(plx_acc_32);
+		pv |= PLX_SYNC_O_EN;
+		writel(pv, plx_acc_32);
+		/* switch to jatt PLL, if not disabled by RX_SYNC */
+		if (hc->type == 1 && !test_bit(HFC_CHIP_RX_SYNC, &hc->chip)) {
+			if (debug & DEBUG_HFCMULTI_PLXSD)
+				printk(KERN_DEBUG "Schedule jatt PLL\n");
+			hc->e1_resync |= 2; /* switch to jatt */
+		}
+	} else {
+		if (pcmmaster) {
+			hc = pcmmaster;
+			if (debug & DEBUG_HFCMULTI_PLXSD)
+				printk(KERN_DEBUG
+					"id=%d (0x%p) = PCM master syncronized "
+					"with QUARTZ\n", hc->id, hc);
+			if (hc->type == 1) {
+				/* Use the crystal clock for the PCM
+				   master card */
+				if (debug & DEBUG_HFCMULTI_PLXSD)
+					printk(KERN_DEBUG
+					    "Schedule QUARTZ for HFC-E1\n");
+				hc->e1_resync |= 4; /* switch quartz */
+			} else {
+				if (debug & DEBUG_HFCMULTI_PLXSD)
+					printk(KERN_DEBUG
+					    "QUARTZ is automatically "
+					    "enabled by HFC-%dS\n", hc->type);
+			}
+			plx_acc_32 = (u_int *)(hc->plx_membase+PLX_GPIOC);
+			pv = readl(plx_acc_32);
+			pv |= PLX_SYNC_O_EN;
+			writel(pv, plx_acc_32);
+		} else
+			if (!rm)
+				printk(KERN_ERR "%s no pcm master, this MUST "
+					"not happen!\n", __func__);
+	}
+	syncmaster = newmaster;
+
+	spin_unlock(&plx_lock);
+	spin_unlock_irqrestore(&HFClock, flags);
+}
+
+/* This must be called AND hc must be locked irqsave!!! */
+inline void
+plxsd_checksync(struct hfc_multi *hc, int rm)
+{
+	if (hc->syncronized) {
+		if (syncmaster == NULL) {
+			if (debug & DEBUG_HFCMULTI_PLXSD)
+				printk(KERN_WARNING "%s: GOT sync on card %d"
+					" (id=%d)\n", __func__, hc->id + 1,
+					hc->id);
+			hfcmulti_resync(hc, hc, rm);
+		}
+	} else {
+		if (syncmaster == hc) {
+			if (debug & DEBUG_HFCMULTI_PLXSD)
+				printk(KERN_WARNING "%s: LOST sync on card %d"
+					" (id=%d)\n", __func__, hc->id + 1,
+					hc->id);
+			hfcmulti_resync(hc, NULL, rm);
+		}
+	}
+}
+
+
+/*
+ * free hardware resources used by driver
+ */
+static void
+release_io_hfcmulti(struct hfc_multi *hc)
+{
+	u_int	*plx_acc_32, pv;
+	u_long	plx_flags;
+
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_DEBUG "%s: entered\n", __func__);
+
+	/* soft reset also masks all interrupts */
+	hc->hw.r_cirm |= V_SRES;
+	HFC_outb(hc, R_CIRM, hc->hw.r_cirm);
+	udelay(1000);
+	hc->hw.r_cirm &= ~V_SRES;
+	HFC_outb(hc, R_CIRM, hc->hw.r_cirm);
+	udelay(1000); /* instead of 'wait' that may cause locking */
+
+	/* release Speech Design card, if PLX was initialized */
+	if (test_bit(HFC_CHIP_PLXSD, &hc->chip) && hc->plx_membase) {
+		if (debug & DEBUG_HFCMULTI_PLXSD)
+			printk(KERN_DEBUG "%s: release PLXSD card %d\n",
+			    __func__, hc->id + 1);
+		spin_lock_irqsave(&plx_lock, plx_flags);
+		plx_acc_32 = (u_int *)(hc->plx_membase+PLX_GPIOC);
+		writel(PLX_GPIOC_INIT, plx_acc_32);
+		pv = readl(plx_acc_32);
+		/* Termination off */
+		pv &= ~PLX_TERM_ON;
+		/* Disconnect the PCM */
+		pv |= PLX_SLAVE_EN_N;
+		pv &= ~PLX_MASTER_EN;
+		pv &= ~PLX_SYNC_O_EN;
+		/* Put the DSP in Reset */
+		pv &= ~PLX_DSP_RES_N;
+		writel(pv, plx_acc_32);
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_WARNING "%s: PCM off: PLX_GPIO=%x\n",
+				__func__, pv);
+		spin_unlock_irqrestore(&plx_lock, plx_flags);
+	}
+
+	/* disable memory mapped ports / io ports */
+	test_and_clear_bit(HFC_CHIP_PLXSD, &hc->chip); /* prevent resync */
+	pci_write_config_word(hc->pci_dev, PCI_COMMAND, 0);
+	if (hc->pci_membase)
+		iounmap((void *)hc->pci_membase);
+	if (hc->plx_membase)
+		iounmap((void *)hc->plx_membase);
+	if (hc->pci_iobase)
+		release_region(hc->pci_iobase, 8);
+
+	if (hc->pci_dev) {
+		pci_disable_device(hc->pci_dev);
+		pci_set_drvdata(hc->pci_dev, NULL);
+	}
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_DEBUG "%s: done\n", __func__);
+}
+
+/*
+ * function called to reset the HFC chip. A complete software reset of chip
+ * and fifos is done. All configuration of the chip is done.
+ */
+
+static int
+init_chip(struct hfc_multi *hc)
+{
+	u_long			flags, val, val2 = 0, rev;
+	int			i, err = 0;
+	u_char			r_conf_en, rval;
+	u_int			*plx_acc_32, pv;
+	u_long			plx_flags, hfc_flags;
+	int			plx_count;
+	struct hfc_multi	*pos, *next, *plx_last_hc;
+
+	spin_lock_irqsave(&hc->lock, flags);
+	/* reset all registers */
+	memset(&hc->hw, 0, sizeof(struct hfcm_hw));
+
+	/* revision check */
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_DEBUG "%s: entered\n", __func__);
+	val = HFC_inb(hc, R_CHIP_ID)>>4;
+	if (val != 0x8 && val != 0xc && val != 0xe) {
+		printk(KERN_INFO "HFC_multi: unknown CHIP_ID:%x\n", (u_int)val);
+		err = -EIO;
+		goto out;
+	}
+	rev = HFC_inb(hc, R_CHIP_RV);
+	printk(KERN_INFO
+	    "HFC_multi: detected HFC with chip ID=0x%lx revision=%ld%s\n",
+	    val, rev, (rev == 0) ? " (old FIFO handling)" : "");
+	if (rev == 0) {
+		test_and_set_bit(HFC_CHIP_REVISION0, &hc->chip);
+		printk(KERN_WARNING
+		    "HFC_multi: NOTE: Your chip is revision 0, "
+		    "ask Cologne Chip for update. Newer chips "
+		    "have a better FIFO handling. Old chips "
+		    "still work but may have slightly lower "
+		    "HDLC transmit performance.\n");
+	}
+	if (rev > 1) {
+		printk(KERN_WARNING "HFC_multi: WARNING: This driver doesn't "
+		    "consider chip revision = %ld. The chip / "
+		    "bridge may not work.\n", rev);
+	}
+
+	/* set s-ram size */
+	hc->Flen = 0x10;
+	hc->Zmin = 0x80;
+	hc->Zlen = 384;
+	hc->DTMFbase = 0x1000;
+	if (test_bit(HFC_CHIP_EXRAM_128, &hc->chip)) {
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG "%s: changing to 128K extenal RAM\n",
+			    __func__);
+		hc->hw.r_ctrl |= V_EXT_RAM;
+		hc->hw.r_ram_sz = 1;
+		hc->Flen = 0x20;
+		hc->Zmin = 0xc0;
+		hc->Zlen = 1856;
+		hc->DTMFbase = 0x2000;
+	}
+	if (test_bit(HFC_CHIP_EXRAM_512, &hc->chip)) {
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG "%s: changing to 512K extenal RAM\n",
+			    __func__);
+		hc->hw.r_ctrl |= V_EXT_RAM;
+		hc->hw.r_ram_sz = 2;
+		hc->Flen = 0x20;
+		hc->Zmin = 0xc0;
+		hc->Zlen = 8000;
+		hc->DTMFbase = 0x2000;
+	}
+	hc->max_trans = poll << 1;
+	if (hc->max_trans > hc->Zlen)
+		hc->max_trans = hc->Zlen;
+
+	/* Speech Design PLX bridge */
+	if (test_bit(HFC_CHIP_PLXSD, &hc->chip)) {
+		if (debug & DEBUG_HFCMULTI_PLXSD)
+			printk(KERN_DEBUG "%s: initializing PLXSD card %d\n",
+			    __func__, hc->id + 1);
+		spin_lock_irqsave(&plx_lock, plx_flags);
+		plx_acc_32 = (u_int *)(hc->plx_membase+PLX_GPIOC);
+		writel(PLX_GPIOC_INIT, plx_acc_32);
+		pv = readl(plx_acc_32);
+		/* The first and the last cards are terminating the PCM bus */
+		pv |= PLX_TERM_ON; /* hc is currently the last */
+		/* Disconnect the PCM */
+		pv |= PLX_SLAVE_EN_N;
+		pv &= ~PLX_MASTER_EN;
+		pv &= ~PLX_SYNC_O_EN;
+		/* Put the DSP in Reset */
+		pv &= ~PLX_DSP_RES_N;
+		writel(pv, plx_acc_32);
+		spin_unlock_irqrestore(&plx_lock, plx_flags);
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_WARNING "%s: slave/term: PLX_GPIO=%x\n",
+				__func__, pv);
+		/*
+		 * If we are the 3rd PLXSD card or higher, we must turn
+		 * termination of last PLXSD card off.
+		 */
+		spin_lock_irqsave(&HFClock, hfc_flags);
+		plx_count = 0;
+		plx_last_hc = NULL;
+		list_for_each_entry_safe(pos, next, &HFClist, list) {
+			if (test_bit(HFC_CHIP_PLXSD, &pos->chip)) {
+				plx_count++;
+				if (pos != hc)
+					plx_last_hc = pos;
+			}
+		}
+		if (plx_count >= 3) {
+			if (debug & DEBUG_HFCMULTI_PLXSD)
+				printk(KERN_DEBUG "%s: card %d is between, so "
+					"we disable termination\n",
+				    __func__, plx_last_hc->id + 1);
+			spin_lock_irqsave(&plx_lock, plx_flags);
+			plx_acc_32 = (u_int *)(plx_last_hc->plx_membase
+					+ PLX_GPIOC);
+			pv = readl(plx_acc_32);
+			pv &= ~PLX_TERM_ON;
+			writel(pv, plx_acc_32);
+			spin_unlock_irqrestore(&plx_lock, plx_flags);
+			if (debug & DEBUG_HFCMULTI_INIT)
+			    printk(KERN_WARNING "%s: term off: PLX_GPIO=%x\n",
+					__func__, pv);
+		}
+		spin_unlock_irqrestore(&HFClock, hfc_flags);
+		hc->hw.r_pcm_md0 = V_F0_LEN; /* shift clock for DSP */
+	}
+
+	/* we only want the real Z2 read-pointer for revision > 0 */
+	if (!test_bit(HFC_CHIP_REVISION0, &hc->chip))
+		hc->hw.r_ram_sz |= V_FZ_MD;
+
+	/* select pcm mode */
+	if (test_bit(HFC_CHIP_PCM_SLAVE, &hc->chip)) {
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG "%s: setting PCM into slave mode\n",
+			    __func__);
+	} else
+	if (test_bit(HFC_CHIP_PCM_MASTER, &hc->chip) && !plxsd_master) {
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG "%s: setting PCM into master mode\n",
+			    __func__);
+		hc->hw.r_pcm_md0 |= V_PCM_MD;
+	} else {
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG "%s: performing PCM auto detect\n",
+			    __func__);
+	}
+
+	/* soft reset */
+	HFC_outb(hc, R_CTRL, hc->hw.r_ctrl);
+	HFC_outb(hc, R_RAM_SZ, hc->hw.r_ram_sz);
+	HFC_outb(hc, R_FIFO_MD, 0);
+	hc->hw.r_cirm = V_SRES | V_HFCRES | V_PCMRES | V_STRES | V_RLD_EPR;
+	HFC_outb(hc, R_CIRM, hc->hw.r_cirm);
+	udelay(100);
+	hc->hw.r_cirm = 0;
+	HFC_outb(hc, R_CIRM, hc->hw.r_cirm);
+	udelay(100);
+	HFC_outb(hc, R_RAM_SZ, hc->hw.r_ram_sz);
+
+	/* Speech Design PLX bridge pcm and sync mode */
+	if (test_bit(HFC_CHIP_PLXSD, &hc->chip)) {
+		spin_lock_irqsave(&plx_lock, plx_flags);
+		plx_acc_32 = (u_int *)(hc->plx_membase+PLX_GPIOC);
+		pv = readl(plx_acc_32);
+		/* Connect PCM */
+		if (hc->hw.r_pcm_md0 & V_PCM_MD) {
+			pv |= PLX_MASTER_EN | PLX_SLAVE_EN_N;
+			pv |= PLX_SYNC_O_EN;
+			if (debug & DEBUG_HFCMULTI_INIT)
+				printk(KERN_WARNING "%s: master: PLX_GPIO=%x\n",
+					__func__, pv);
+		} else {
+			pv &= ~(PLX_MASTER_EN | PLX_SLAVE_EN_N);
+			pv &= ~PLX_SYNC_O_EN;
+			if (debug & DEBUG_HFCMULTI_INIT)
+				printk(KERN_WARNING "%s: slave: PLX_GPIO=%x\n",
+					__func__, pv);
+		}
+		writel(pv, plx_acc_32);
+		spin_unlock_irqrestore(&plx_lock, plx_flags);
+	}
+
+	/* PCM setup */
+	HFC_outb(hc, R_PCM_MD0, hc->hw.r_pcm_md0 | 0x90);
+	if (hc->slots == 32)
+		HFC_outb(hc, R_PCM_MD1, 0x00);
+	if (hc->slots == 64)
+		HFC_outb(hc, R_PCM_MD1, 0x10);
+	if (hc->slots == 128)
+		HFC_outb(hc, R_PCM_MD1, 0x20);
+	HFC_outb(hc, R_PCM_MD0, hc->hw.r_pcm_md0 | 0xa0);
+	if (test_bit(HFC_CHIP_PLXSD, &hc->chip))
+		HFC_outb(hc, R_PCM_MD2, V_SYNC_SRC); /* sync via SYNC_I / O */
+	else
+		HFC_outb(hc, R_PCM_MD2, 0x00); /* sync from interface */
+	HFC_outb(hc, R_PCM_MD0, hc->hw.r_pcm_md0 | 0x00);
+	for (i = 0; i < 256; i++) {
+		HFC_outb_nodebug(hc, R_SLOT, i);
+		HFC_outb_nodebug(hc, A_SL_CFG, 0);
+		HFC_outb_nodebug(hc, A_CONF, 0);
+		hc->slot_owner[i] = -1;
+	}
+
+	/* set clock speed */
+	if (test_bit(HFC_CHIP_CLOCK2, &hc->chip)) {
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG
+			    "%s: setting double clock\n", __func__);
+		HFC_outb(hc, R_BRG_PCM_CFG, V_PCM_CLK);
+	}
+
+	/* B410P GPIO */
+	if (test_bit(HFC_CHIP_B410P, &hc->chip)) {
+		printk(KERN_NOTICE "Setting GPIOs\n");
+		HFC_outb(hc, R_GPIO_SEL, 0x30);
+		HFC_outb(hc, R_GPIO_EN1, 0x3);
+		udelay(1000);
+		printk(KERN_NOTICE "calling vpm_init\n");
+		vpm_init(hc);
+	}
+
+	/* check if R_F0_CNT counts (8 kHz frame count) */
+	val = HFC_inb(hc, R_F0_CNTL);
+	val += HFC_inb(hc, R_F0_CNTH) << 8;
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_DEBUG
+		    "HFC_multi F0_CNT %ld after reset\n", val);
+	spin_unlock_irqrestore(&hc->lock, flags);
+	set_current_state(TASK_UNINTERRUPTIBLE);
+	schedule_timeout((HZ/100)?:1); /* Timeout minimum 10ms */
+	spin_lock_irqsave(&hc->lock, flags);
+	val2 = HFC_inb(hc, R_F0_CNTL);
+	val2 += HFC_inb(hc, R_F0_CNTH) << 8;
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_DEBUG
+			"HFC_multi F0_CNT %ld after 10 ms (1st try)\n",
+		    val2);
+	if (val2 >= val+8) { /* 1 ms */
+		/* it counts, so we keep the pcm mode */
+		if (test_bit(HFC_CHIP_PCM_MASTER, &hc->chip))
+			printk(KERN_INFO "controller is PCM bus MASTER\n");
+		else
+		if (test_bit(HFC_CHIP_PCM_SLAVE, &hc->chip))
+			printk(KERN_INFO "controller is PCM bus SLAVE\n");
+		else {
+			test_and_set_bit(HFC_CHIP_PCM_SLAVE, &hc->chip);
+			printk(KERN_INFO "controller is PCM bus SLAVE "
+				"(auto detected)\n");
+		}
+	} else {
+		/* does not count */
+		if (test_bit(HFC_CHIP_PCM_MASTER, &hc->chip)) {
+controller_fail:
+			printk(KERN_ERR "HFC_multi ERROR, getting no 125us "
+			    "pulse. Seems that controller fails.\n");
+			err = -EIO;
+			goto out;
+		}
+		if (test_bit(HFC_CHIP_PCM_SLAVE, &hc->chip)) {
+			printk(KERN_INFO "controller is PCM bus SLAVE "
+				"(ignoring missing PCM clock)\n");
+		} else {
+			/* only one pcm master */
+			if (test_bit(HFC_CHIP_PLXSD, &hc->chip)
+				&& plxsd_master) {
+				printk(KERN_ERR "HFC_multi ERROR, no clock "
+				    "on another Speech Design card found. "
+				    "Please be sure to connect PCM cable.\n");
+				err = -EIO;
+				goto out;
+			}
+			/* retry with master clock */
+			if (test_bit(HFC_CHIP_PLXSD, &hc->chip)) {
+				spin_lock_irqsave(&plx_lock, plx_flags);
+				plx_acc_32 = (u_int *)(hc->plx_membase +
+					PLX_GPIOC);
+				pv = readl(plx_acc_32);
+				pv |= PLX_MASTER_EN | PLX_SLAVE_EN_N;
+				pv |= PLX_SYNC_O_EN;
+				writel(pv, plx_acc_32);
+				spin_unlock_irqrestore(&plx_lock, plx_flags);
+				if (debug & DEBUG_HFCMULTI_INIT)
+				    printk(KERN_WARNING "%s: master: PLX_GPIO"
+					"=%x\n", __func__, pv);
+			}
+			hc->hw.r_pcm_md0 |= V_PCM_MD;
+			HFC_outb(hc, R_PCM_MD0, hc->hw.r_pcm_md0 | 0x00);
+			spin_unlock_irqrestore(&hc->lock, flags);
+			set_current_state(TASK_UNINTERRUPTIBLE);
+			schedule_timeout((HZ/100)?:1); /* Timeout min. 10ms */
+			spin_lock_irqsave(&hc->lock, flags);
+			val2 = HFC_inb(hc, R_F0_CNTL);
+			val2 += HFC_inb(hc, R_F0_CNTH) << 8;
+			if (debug & DEBUG_HFCMULTI_INIT)
+				printk(KERN_DEBUG "HFC_multi F0_CNT %ld after "
+					"10 ms (2nd try)\n", val2);
+			if (val2 >= val+8) { /* 1 ms */
+				test_and_set_bit(HFC_CHIP_PCM_MASTER,
+					&hc->chip);
+				printk(KERN_INFO "controller is PCM bus MASTER "
+					"(auto detected)\n");
+			} else
+				goto controller_fail;
+		}
+	}
+
+	/* Release the DSP Reset */
+	if (test_bit(HFC_CHIP_PLXSD, &hc->chip)) {
+		if (test_bit(HFC_CHIP_PCM_MASTER, &hc->chip))
+			plxsd_master = 1;
+		spin_lock_irqsave(&plx_lock, plx_flags);
+		plx_acc_32 = (u_int *)(hc->plx_membase+PLX_GPIOC);
+		pv = readl(plx_acc_32);
+		pv |=  PLX_DSP_RES_N;
+		writel(pv, plx_acc_32);
+		spin_unlock_irqrestore(&plx_lock, plx_flags);
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_WARNING "%s: reset off: PLX_GPIO=%x\n",
+				__func__, pv);
+	}
+
+	/* pcm id */
+	if (hc->pcm)
+		printk(KERN_INFO "controller has given PCM BUS ID %d\n",
+			hc->pcm);
+	else {
+		if (test_bit(HFC_CHIP_PCM_MASTER, &hc->chip)
+		 || test_bit(HFC_CHIP_PLXSD, &hc->chip)) {
+			PCM_cnt++; /* SD has proprietary bridging */
+		}
+		hc->pcm = PCM_cnt;
+		printk(KERN_INFO "controller has PCM BUS ID %d "
+			"(auto selected)\n", hc->pcm);
+	}
+
+	/* set up timer */
+	HFC_outb(hc, R_TI_WD, poll_timer);
+	hc->hw.r_irqmsk_misc |= V_TI_IRQMSK;
+
+	/*
+	 * set up 125us interrupt, only if function pointer is available
+	 * and module parameter timer is set
+	 */
+	if (timer && hfc_interrupt && register_interrupt) {
+		/* only one chip should use this interrupt */
+		timer = 0;
+		interrupt_registered = 1;
+		hc->hw.r_irqmsk_misc |= V_PROC_IRQMSK;
+		/* deactivate other interrupts in ztdummy */
+		register_interrupt();
+	}
+
+	/* set E1 state machine IRQ */
+	if (hc->type == 1)
+		hc->hw.r_irqmsk_misc |= V_STA_IRQMSK;
+
+	/* set DTMF detection */
+	if (test_bit(HFC_CHIP_DTMF, &hc->chip)) {
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG "%s: enabling DTMF detection "
+			    "for all B-channel\n", __func__);
+		hc->hw.r_dtmf = V_DTMF_EN | V_DTMF_STOP;
+		if (test_bit(HFC_CHIP_ULAW, &hc->chip))
+			hc->hw.r_dtmf |= V_ULAW_SEL;
+		HFC_outb(hc, R_DTMF_N, 102 - 1);
+		hc->hw.r_irqmsk_misc |= V_DTMF_IRQMSK;
+	}
+
+	/* conference engine */
+	if (test_bit(HFC_CHIP_ULAW, &hc->chip))
+		r_conf_en = V_CONF_EN | V_ULAW;
+	else
+		r_conf_en = V_CONF_EN;
+	HFC_outb(hc, R_CONF_EN, r_conf_en);
+
+	/* setting leds */
+	switch (hc->leds) {
+	case 1: /* HFC-E1 OEM */
+		if (test_bit(HFC_CHIP_WATCHDOG, &hc->chip))
+			HFC_outb(hc, R_GPIO_SEL, 0x32);
+		else
+			HFC_outb(hc, R_GPIO_SEL, 0x30);
+
+		HFC_outb(hc, R_GPIO_EN1, 0x0f);
+		HFC_outb(hc, R_GPIO_OUT1, 0x00);
+
+		HFC_outb(hc, R_GPIO_EN0, V_GPIO_EN2 | V_GPIO_EN3);
+		break;
+
+	case 2: /* HFC-4S OEM */
+	case 3:
+		HFC_outb(hc, R_GPIO_SEL, 0xf0);
+		HFC_outb(hc, R_GPIO_EN1, 0xff);
+		HFC_outb(hc, R_GPIO_OUT1, 0x00);
+		break;
+	}
+
+	/* set master clock */
+	if (hc->masterclk >= 0) {
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG "%s: setting ST master clock "
+			    "to port %d (0..%d)\n",
+			    __func__, hc->masterclk, hc->ports-1);
+		hc->hw.r_st_sync = hc->masterclk | V_AUTO_SYNC;
+		HFC_outb(hc, R_ST_SYNC, hc->hw.r_st_sync);
+	}
+
+	/* setting misc irq */
+	HFC_outb(hc, R_IRQMSK_MISC, hc->hw.r_irqmsk_misc);
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_DEBUG "r_irqmsk_misc.2: 0x%x\n",
+		    hc->hw.r_irqmsk_misc);
+
+	/* RAM access test */
+	HFC_outb(hc, R_RAM_ADDR0, 0);
+	HFC_outb(hc, R_RAM_ADDR1, 0);
+	HFC_outb(hc, R_RAM_ADDR2, 0);
+	for (i = 0; i < 256; i++) {
+		HFC_outb_nodebug(hc, R_RAM_ADDR0, i);
+		HFC_outb_nodebug(hc, R_RAM_DATA, ((i*3)&0xff));
+	}
+	for (i = 0; i < 256; i++) {
+		HFC_outb_nodebug(hc, R_RAM_ADDR0, i);
+		HFC_inb_nodebug(hc, R_RAM_DATA);
+		rval = HFC_inb_nodebug(hc, R_INT_DATA);
+		if (rval != ((i * 3) & 0xff)) {
+			printk(KERN_DEBUG
+			    "addr:%x val:%x should:%x\n", i, rval,
+			    (i * 3) & 0xff);
+			err++;
+		}
+	}
+	if (err) {
+		printk(KERN_DEBUG "aborting - %d RAM access errors\n", err);
+		err = -EIO;
+		goto out;
+	}
+
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_DEBUG "%s: done\n", __func__);
+out:
+	spin_unlock_irqrestore(&hc->lock, flags);
+	return err;
+}
+
+
+/*
+ * control the watchdog
+ */
+static void
+hfcmulti_watchdog(struct hfc_multi *hc)
+{
+	hc->wdcount++;
+
+	if (hc->wdcount > 10) {
+		hc->wdcount = 0;
+		hc->wdbyte = hc->wdbyte == V_GPIO_OUT2 ?
+		    V_GPIO_OUT3 : V_GPIO_OUT2;
+
+	/* printk("Sending Watchdog Kill %x\n",hc->wdbyte); */
+		HFC_outb(hc, R_GPIO_EN0, V_GPIO_EN2 | V_GPIO_EN3);
+		HFC_outb(hc, R_GPIO_OUT0, hc->wdbyte);
+	}
+}
+
+
+
+/*
+ * output leds
+ */
+static void
+hfcmulti_leds(struct hfc_multi *hc)
+{
+	unsigned long lled;
+	unsigned long leddw;
+	int i, state, active, leds;
+	struct dchannel *dch;
+	int led[4];
+
+	hc->ledcount += poll;
+	if (hc->ledcount > 4096) {
+		hc->ledcount -= 4096;
+		hc->ledstate = 0xAFFEAFFE;
+	}
+
+	switch (hc->leds) {
+	case 1: /* HFC-E1 OEM */
+		/* 2 red blinking: NT mode deactivate
+		 * 2 red steady:   TE mode deactivate
+		 * left green:     L1 active
+		 * left red:       frame sync, but no L1
+		 * right green:    L2 active
+		 */
+		if (hc->chan[hc->dslot].sync != 2) { /* no frame sync */
+			if (hc->chan[hc->dslot].dch->dev.D.protocol
+				!= ISDN_P_NT_E1) {
+				led[0] = 1;
+				led[1] = 1;
+			} else if (hc->ledcount>>11) {
+				led[0] = 1;
+				led[1] = 1;
+			} else {
+				led[0] = 0;
+				led[1] = 0;
+			}
+			led[2] = 0;
+			led[3] = 0;
+		} else { /* with frame sync */
+			/* TODO make it work */
+			led[0] = 0;
+			led[1] = 0;
+			led[2] = 0;
+			led[3] = 1;
+		}
+		leds = (led[0] | (led[1]<<2) | (led[2]<<1) | (led[3]<<3))^0xF;
+			/* leds are inverted */
+		if (leds != (int)hc->ledstate) {
+			HFC_outb_nodebug(hc, R_GPIO_OUT1, leds);
+			hc->ledstate = leds;
+		}
+		break;
+
+	case 2: /* HFC-4S OEM */
+		/* red blinking = PH_DEACTIVATE NT Mode
+		 * red steady   = PH_DEACTIVATE TE Mode
+		 * green steady = PH_ACTIVATE
+		 */
+		for (i = 0; i < 4; i++) {
+			state = 0;
+			active = -1;
+			dch = hc->chan[(i << 2) | 2].dch;
+			if (dch) {
+				state = dch->state;
+				if (dch->dev.D.protocol == ISDN_P_NT_S0)
+					active = 3;
+				else
+					active = 7;
+			}
+			if (state) {
+				if (state == active) {
+					led[i] = 1; /* led green */
+				} else
+					if (dch->dev.D.protocol == ISDN_P_TE_S0)
+						/* TE mode: led red */
+						led[i] = 2;
+					else
+						if (hc->ledcount>>11)
+							/* led red */
+							led[i] = 2;
+						else
+							/* led off */
+							led[i] = 0;
+			} else
+				led[i] = 0; /* led off */
+		}
+		if (test_bit(HFC_CHIP_B410P, &hc->chip)) {
+			leds = 0;
+			for (i = 0; i < 4; i++) {
+				if (led[i] == 1) {
+					/*green*/
+					leds |= (0x2 << (i * 2));
+				} else if (led[i] == 2) {
+					/*red*/
+					leds |= (0x1 << (i * 2));
+				}
+			}
+			if (leds != (int)hc->ledstate) {
+				vpm_out(hc, 0, 0x1a8 + 3, leds);
+				hc->ledstate = leds;
+			}
+		} else {
+			leds = ((led[3] > 0) << 0) | ((led[1] > 0) << 1) |
+			    ((led[0] > 0) << 2) | ((led[2] > 0) << 3) |
+			    ((led[3] & 1) << 4) | ((led[1] & 1) << 5) |
+			    ((led[0] & 1) << 6) | ((led[2] & 1) << 7);
+			if (leds != (int)hc->ledstate) {
+				HFC_outb_nodebug(hc, R_GPIO_EN1, leds & 0x0F);
+				HFC_outb_nodebug(hc, R_GPIO_OUT1, leds >> 4);
+				hc->ledstate = leds;
+			}
+		}
+		break;
+
+	case 3: /* HFC 1S/2S Beronet */
+		/* red blinking = PH_DEACTIVATE NT Mode
+		 * red steady   = PH_DEACTIVATE TE Mode
+		 * green steady = PH_ACTIVATE
+		 */
+		for (i = 0; i < 2; i++) {
+			state = 0;
+			active = -1;
+			dch = hc->chan[(i << 2) | 2].dch;
+			if (dch) {
+				state = dch->state;
+				if (dch->dev.D.protocol == ISDN_P_NT_S0)
+					active = 3;
+				else
+					active = 7;
+			}
+			if (state) {
+				if (state == active) {
+					led[i] = 1; /* led green */
+				} else
+					if (dch->dev.D.protocol == ISDN_P_TE_S0)
+						/* TE mode: led red */
+						led[i] = 2;
+					else
+						if (hc->ledcount >> 11)
+							/* led red */
+							led[i] = 2;
+						else
+							/* led off */
+							led[i] = 0;
+			} else
+				led[i] = 0; /* led off */
+		}
+
+
+		leds = (led[0] > 0) | ((led[1] > 0)<<1) | ((led[0]&1)<<2)
+			| ((led[1]&1)<<3);
+		if (leds != (int)hc->ledstate) {
+			HFC_outb_nodebug(hc, R_GPIO_EN1,
+			    ((led[0] > 0) << 2) | ((led[1] > 0) << 3));
+			HFC_outb_nodebug(hc, R_GPIO_OUT1,
+			    ((led[0] & 1) << 2) | ((led[1] & 1) << 3));
+			hc->ledstate = leds;
+		}
+		break;
+	case 8: /* HFC 8S+ Beronet */
+		lled = 0;
+
+		for (i = 0; i < 8; i++) {
+			state = 0;
+			active = -1;
+			dch = hc->chan[(i << 2) | 2].dch;
+			if (dch) {
+				state = dch->state;
+				if (dch->dev.D.protocol == ISDN_P_NT_S0)
+					active = 3;
+				else
+					active = 7;
+			}
+			if (state) {
+				if (state == active) {
+					lled |= 0 << i;
+				} else
+					if (hc->ledcount >> 11)
+						lled |= 0 << i;
+					else
+						lled |= 1 << i;
+			} else
+				lled |= 1 << i;
+		}
+		leddw = lled << 24 | lled << 16 | lled << 8 | lled;
+		if (leddw != hc->ledstate) {
+			/* HFC_outb(hc, R_BRG_PCM_CFG, 1);
+			HFC_outb(c, R_BRG_PCM_CFG, (0x0 << 6) | 0x3); */
+			/* was _io before */
+			HFC_outb_nodebug(hc, R_BRG_PCM_CFG, 1 | V_PCM_CLK);
+			outw(0x4000, hc->pci_iobase + 4);
+			outl(leddw, hc->pci_iobase);
+			HFC_outb_nodebug(hc, R_BRG_PCM_CFG, V_PCM_CLK);
+			hc->ledstate = leddw;
+		}
+		break;
+	}
+}
+/*
+ * read dtmf coefficients
+ */
+
+static void
+hfcmulti_dtmf(struct hfc_multi *hc)
+{
+	s32		*coeff;
+	u_int		mantissa;
+	int		co, ch;
+	struct bchannel	*bch = NULL;
+	u8		exponent;
+	int		dtmf = 0;
+	int		addr;
+	u16		w_float;
+	struct sk_buff	*skb;
+	struct mISDNhead *hh;
+
+	if (debug & DEBUG_HFCMULTI_DTMF)
+		printk(KERN_DEBUG "%s: dtmf detection irq\n", __func__);
+	for (ch = 0; ch <= 31; ch++) {
+		/* only process enabled B-channels */
+		bch = hc->chan[ch].bch;
+		if (!bch)
+			continue;
+		if (!hc->created[hc->chan[ch].port])
+			continue;
+		if (!test_bit(FLG_TRANSPARENT, &bch->Flags))
+			continue;
+		if (debug & DEBUG_HFCMULTI_DTMF)
+			printk(KERN_DEBUG "%s: dtmf channel %d:",
+				__func__, ch);
+		coeff = &(hc->chan[ch].coeff[hc->chan[ch].coeff_count * 16]);
+		dtmf = 1;
+		for (co = 0; co < 8; co++) {
+			/* read W(n-1) coefficient */
+			addr = hc->DTMFbase + ((co<<7) | (ch<<2));
+			HFC_outb_nodebug(hc, R_RAM_ADDR0, addr);
+			HFC_outb_nodebug(hc, R_RAM_ADDR1, addr>>8);
+			HFC_outb_nodebug(hc, R_RAM_ADDR2, (addr>>16)
+				| V_ADDR_INC);
+			w_float = HFC_inb_nodebug(hc, R_RAM_DATA);
+			w_float |= (HFC_inb_nodebug(hc, R_RAM_DATA) << 8);
+			if (debug & DEBUG_HFCMULTI_DTMF)
+				printk(" %04x", w_float);
+
+			/* decode float (see chip doc) */
+			mantissa = w_float & 0x0fff;
+			if (w_float & 0x8000)
+				mantissa |= 0xfffff000;
+			exponent = (w_float>>12) & 0x7;
+			if (exponent) {
+				mantissa ^= 0x1000;
+				mantissa <<= (exponent-1);
+			}
+
+			/* store coefficient */
+			coeff[co<<1] = mantissa;
+
+			/* read W(n) coefficient */
+			w_float = HFC_inb_nodebug(hc, R_RAM_DATA);
+			w_float |= (HFC_inb_nodebug(hc, R_RAM_DATA) << 8);
+			if (debug & DEBUG_HFCMULTI_DTMF)
+				printk(" %04x", w_float);
+
+			/* decode float (see chip doc) */
+			mantissa = w_float & 0x0fff;
+			if (w_float & 0x8000)
+				mantissa |= 0xfffff000;
+			exponent = (w_float>>12) & 0x7;
+			if (exponent) {
+				mantissa ^= 0x1000;
+				mantissa <<= (exponent-1);
+			}
+
+			/* store coefficient */
+			coeff[(co<<1)|1] = mantissa;
+		}
+		if (debug & DEBUG_HFCMULTI_DTMF)
+			printk("%s: DTMF ready %08x %08x %08x %08x "
+			    "%08x %08x %08x %08x\n", __func__,
+			    coeff[0], coeff[1], coeff[2], coeff[3],
+			    coeff[4], coeff[5], coeff[6], coeff[7]);
+		hc->chan[ch].coeff_count++;
+		if (hc->chan[ch].coeff_count == 8) {
+			hc->chan[ch].coeff_count = 0;
+			skb = mI_alloc_skb(512, GFP_ATOMIC);
+			if (!skb) {
+				printk(KERN_WARNING "%s: No memory for skb\n",
+				    __func__);
+				continue;
+			}
+			hh = mISDN_HEAD_P(skb);
+			hh->prim = PH_CONTROL_IND;
+			hh->id = DTMF_HFC_COEF;
+			memcpy(skb_put(skb, 512), hc->chan[ch].coeff, 512);
+			recv_Bchannel_skb(bch, skb);
+		}
+	}
+
+	/* restart DTMF processing */
+	hc->dtmf = dtmf;
+	if (dtmf)
+		HFC_outb_nodebug(hc, R_DTMF, hc->hw.r_dtmf | V_RST_DTMF);
+}
+
+
+/*
+ * fill fifo as much as possible
+ */
+
+static void
+hfcmulti_tx(struct hfc_multi *hc, int ch)
+{
+	int i, ii, temp, len = 0;
+	int Zspace, z1, z2; /* must be int for calculation */
+	int Fspace, f1, f2;
+	u_char *d;
+	int *txpending, slot_tx;
+	struct	bchannel *bch;
+	struct  dchannel *dch;
+	struct  sk_buff **sp = NULL;
+	int *idxp;
+
+	bch = hc->chan[ch].bch;
+	dch = hc->chan[ch].dch;
+	if ((!dch) && (!bch))
+		return;
+
+	txpending = &hc->chan[ch].txpending;
+	slot_tx = hc->chan[ch].slot_tx;
+	if (dch) {
+		if (!test_bit(FLG_ACTIVE, &dch->Flags))
+			return;
+		sp = &dch->tx_skb;
+		idxp = &dch->tx_idx;
+	} else {
+		if (!test_bit(FLG_ACTIVE, &bch->Flags))
+			return;
+		sp = &bch->tx_skb;
+		idxp = &bch->tx_idx;
+	}
+	if (*sp)
+		len = (*sp)->len;
+
+	if ((!len) && *txpending != 1)
+		return; /* no data */
+
+	if (test_bit(HFC_CHIP_B410P, &hc->chip) &&
+	    (hc->chan[ch].protocol == ISDN_P_B_RAW) &&
+	    (hc->chan[ch].slot_rx < 0) &&
+	    (hc->chan[ch].slot_tx < 0))
+		HFC_outb_nodebug(hc, R_FIFO, 0x20 | (ch << 1));
+	else
+		HFC_outb_nodebug(hc, R_FIFO, ch << 1);
+	HFC_wait_nodebug(hc);
+
+	if (*txpending == 2) {
+		/* reset fifo */
+		HFC_outb_nodebug(hc, R_INC_RES_FIFO, V_RES_F);
+		HFC_wait_nodebug(hc);
+		HFC_outb(hc, A_SUBCH_CFG, 0);
+		*txpending = 1;
+	}
+next_frame:
+	if (dch || test_bit(FLG_HDLC, &bch->Flags)) {
+		f1 = HFC_inb_nodebug(hc, A_F1);
+		f2 = HFC_inb_nodebug(hc, A_F2);
+		while (f2 != (temp = HFC_inb_nodebug(hc, A_F2))) {
+			if (debug & DEBUG_HFCMULTI_FIFO)
+				printk(KERN_DEBUG
+				    "%s(card %d): reread f2 because %d!=%d\n",
+				    __func__, hc->id + 1, temp, f2);
+			f2 = temp; /* repeat until F2 is equal */
+		}
+		Fspace = f2 - f1 - 1;
+		if (Fspace < 0)
+			Fspace += hc->Flen;
+		/*
+		 * Old FIFO handling doesn't give us the current Z2 read
+		 * pointer, so we cannot send the next frame before the fifo
+		 * is empty. It makes no difference except for a slightly
+		 * lower performance.
+		 */
+		if (test_bit(HFC_CHIP_REVISION0, &hc->chip)) {
+			if (f1 != f2)
+				Fspace = 0;
+			else
+				Fspace = 1;
+		}
+		/* one frame only for ST D-channels, to allow resending */
+		if (hc->type != 1 && dch) {
+			if (f1 != f2)
+				Fspace = 0;
+		}
+		/* F-counter full condition */
+		if (Fspace == 0)
+			return;
+	}
+	z1 = HFC_inw_nodebug(hc, A_Z1) - hc->Zmin;
+	z2 = HFC_inw_nodebug(hc, A_Z2) - hc->Zmin;
+	while (z2 != (temp = (HFC_inw_nodebug(hc, A_Z2) - hc->Zmin))) {
+		if (debug & DEBUG_HFCMULTI_FIFO)
+			printk(KERN_DEBUG "%s(card %d): reread z2 because "
+				"%d!=%d\n", __func__, hc->id + 1, temp, z2);
+		z2 = temp; /* repeat unti Z2 is equal */
+	}
+	Zspace = z2 - z1;
+	if (Zspace <= 0)
+		Zspace += hc->Zlen;
+	Zspace -= 4; /* keep not too full, so pointers will not overrun */
+	/* fill transparent data only to maxinum transparent load (minus 4) */
+	if (bch && test_bit(FLG_TRANSPARENT, &bch->Flags))
+		Zspace = Zspace - hc->Zlen + hc->max_trans;
+	if (Zspace <= 0) /* no space of 4 bytes */
+		return;
+
+	/* if no data */
+	if (!len) {
+		if (z1 == z2) { /* empty */
+			/* if done with FIFO audio data during PCM connection */
+			if (bch && (!test_bit(FLG_HDLC, &bch->Flags)) &&
+			    *txpending && slot_tx >= 0) {
+				if (debug & DEBUG_HFCMULTI_MODE)
+					printk(KERN_DEBUG
+					    "%s: reconnecting PCM due to no "
+					    "more FIFO data: channel %d "
+					    "slot_tx %d\n",
+					    __func__, ch, slot_tx);
+				/* connect slot */
+				HFC_outb(hc, A_CON_HDLC, 0xc0 | 0x00 |
+				    V_HDLC_TRP | V_IFF);
+				HFC_outb_nodebug(hc, R_FIFO, ch<<1 | 1);
+				HFC_wait_nodebug(hc);
+				HFC_outb(hc, A_CON_HDLC, 0xc0 | 0x00 |
+				    V_HDLC_TRP | V_IFF);
+				HFC_outb_nodebug(hc, R_FIFO, ch<<1);
+				HFC_wait_nodebug(hc);
+			}
+			*txpending = 0;
+		}
+		return; /* no data */
+	}
+
+	/* if audio data and connected slot */
+	if (bch && (!test_bit(FLG_HDLC, &bch->Flags)) && (!*txpending)
+		&& slot_tx >= 0) {
+		if (debug & DEBUG_HFCMULTI_MODE)
+			printk(KERN_DEBUG "%s: disconnecting PCM due to "
+			    "FIFO data: channel %d slot_tx %d\n",
+			    __func__, ch, slot_tx);
+		/* disconnect slot */
+		HFC_outb(hc, A_CON_HDLC, 0x80 | 0x00 | V_HDLC_TRP | V_IFF);
+		HFC_outb_nodebug(hc, R_FIFO, ch<<1 | 1);
+		HFC_wait_nodebug(hc);
+		HFC_outb(hc, A_CON_HDLC, 0x80 | 0x00 | V_HDLC_TRP | V_IFF);
+		HFC_outb_nodebug(hc, R_FIFO, ch<<1);
+		HFC_wait_nodebug(hc);
+	}
+	*txpending = 1;
+
+	/* show activity */
+	hc->activity[hc->chan[ch].port] = 1;
+
+	/* fill fifo to what we have left */
+	ii = len;
+	if (dch || test_bit(FLG_HDLC, &bch->Flags))
+		temp = 1;
+	else
+		temp = 0;
+	i = *idxp;
+	d = (*sp)->data + i;
+	if (ii - i > Zspace)
+		ii = Zspace + i;
+	if (debug & DEBUG_HFCMULTI_FIFO)
+		printk(KERN_DEBUG "%s(card %d): fifo(%d) has %d bytes space "
+		    "left (z1=%04x, z2=%04x) sending %d of %d bytes %s\n",
+			__func__, hc->id + 1, ch, Zspace, z1, z2, ii-i, len-i,
+			temp ? "HDLC":"TRANS");
+
+
+	/* Have to prep the audio data */
+	hc->write_fifo(hc, d, ii - i);
+	*idxp = ii;
+
+	/* if not all data has been written */
+	if (ii != len) {
+		/* NOTE: fifo is started by the calling function */
+		return;
+	}
+
+	/* if all data has been written, terminate frame */
+	if (dch || test_bit(FLG_HDLC, &bch->Flags)) {
+		/* increment f-counter */
+		HFC_outb_nodebug(hc, R_INC_RES_FIFO, V_INC_F);
+		HFC_wait_nodebug(hc);
+	}
+
+	/* send confirm, since get_net_bframe will not do it with trans */
+	if (bch && test_bit(FLG_TRANSPARENT, &bch->Flags))
+		confirm_Bsend(bch);
+
+	/* check for next frame */
+	dev_kfree_skb(*sp);
+	if (bch && get_next_bframe(bch)) { /* hdlc is confirmed here */
+		len = (*sp)->len;
+		goto next_frame;
+	}
+	if (dch && get_next_dframe(dch)) {
+		len = (*sp)->len;
+		goto next_frame;
+	}
+
+	/*
+	 * now we have no more data, so in case of transparent,
+	 * we set the last byte in fifo to 'silence' in case we will get
+	 * no more data at all. this prevents sending an undefined value.
+	 */
+	if (bch && test_bit(FLG_TRANSPARENT, &bch->Flags))
+		HFC_outb_nodebug(hc, A_FIFO_DATA0_NOINC, silence);
+}
+
+
+/* NOTE: only called if E1 card is in active state */
+static void
+hfcmulti_rx(struct hfc_multi *hc, int ch)
+{
+	int temp;
+	int Zsize, z1, z2 = 0; /* = 0, to make GCC happy */
+	int f1 = 0, f2 = 0; /* = 0, to make GCC happy */
+	int again = 0;
+	struct	bchannel *bch;
+	struct  dchannel *dch;
+	struct sk_buff	*skb, **sp = NULL;
+	int	maxlen;
+
+	bch = hc->chan[ch].bch;
+	dch = hc->chan[ch].dch;
+	if ((!dch) && (!bch))
+		return;
+	if (dch) {
+		if (!test_bit(FLG_ACTIVE, &dch->Flags))
+			return;
+		sp = &dch->rx_skb;
+		maxlen = dch->maxlen;
+	} else {
+		if (!test_bit(FLG_ACTIVE, &bch->Flags))
+			return;
+		sp = &bch->rx_skb;
+		maxlen = bch->maxlen;
+	}
+next_frame:
+	/* on first AND before getting next valid frame, R_FIFO must be written
+	   to. */
+	if (test_bit(HFC_CHIP_B410P, &hc->chip) &&
+	    (hc->chan[ch].protocol == ISDN_P_B_RAW) &&
+	    (hc->chan[ch].slot_rx < 0) &&
+	    (hc->chan[ch].slot_tx < 0))
+		HFC_outb_nodebug(hc, R_FIFO, 0x20 | (ch<<1) | 1);
+	else
+		HFC_outb_nodebug(hc, R_FIFO, (ch<<1)|1);
+	HFC_wait_nodebug(hc);
+
+	/* ignore if rx is off BUT change fifo (above) to start pending TX */
+	if (hc->chan[ch].rx_off)
+		return;
+
+	if (dch || test_bit(FLG_HDLC, &bch->Flags)) {
+		f1 = HFC_inb_nodebug(hc, A_F1);
+		while (f1 != (temp = HFC_inb_nodebug(hc, A_F1))) {
+			if (debug & DEBUG_HFCMULTI_FIFO)
+				printk(KERN_DEBUG
+				    "%s(card %d): reread f1 because %d!=%d\n",
+				    __func__, hc->id + 1, temp, f1);
+			f1 = temp; /* repeat until F1 is equal */
+		}
+		f2 = HFC_inb_nodebug(hc, A_F2);
+	}
+	z1 = HFC_inw_nodebug(hc, A_Z1) - hc->Zmin;
+	while (z1 != (temp = (HFC_inw_nodebug(hc, A_Z1) - hc->Zmin))) {
+		if (debug & DEBUG_HFCMULTI_FIFO)
+			printk(KERN_DEBUG "%s(card %d): reread z2 because "
+				"%d!=%d\n", __func__, hc->id + 1, temp, z2);
+		z1 = temp; /* repeat until Z1 is equal */
+	}
+	z2 = HFC_inw_nodebug(hc, A_Z2) - hc->Zmin;
+	Zsize = z1 - z2;
+	if ((dch || test_bit(FLG_HDLC, &bch->Flags)) && f1 != f2)
+		/* complete hdlc frame */
+		Zsize++;
+	if (Zsize < 0)
+		Zsize += hc->Zlen;
+	/* if buffer is empty */
+	if (Zsize <= 0)
+		return;
+
+	if (*sp == NULL) {
+		*sp = mI_alloc_skb(maxlen + 3, GFP_ATOMIC);
+		if (*sp == NULL) {
+			printk(KERN_DEBUG "%s: No mem for rx_skb\n",
+			    __func__);
+			return;
+		}
+	}
+	/* show activity */
+	hc->activity[hc->chan[ch].port] = 1;
+
+	/* empty fifo with what we have */
+	if (dch || test_bit(FLG_HDLC, &bch->Flags)) {
+		if (debug & DEBUG_HFCMULTI_FIFO)
+			printk(KERN_DEBUG "%s(card %d): fifo(%d) reading %d "
+			    "bytes (z1=%04x, z2=%04x) HDLC %s (f1=%d, f2=%d) "
+			    "got=%d (again %d)\n", __func__, hc->id + 1, ch,
+			    Zsize, z1, z2, (f1 == f2) ? "fragment" : "COMPLETE",
+			    f1, f2, Zsize + (*sp)->len, again);
+		/* HDLC */
+		if ((Zsize + (*sp)->len) > (maxlen + 3)) {
+			if (debug & DEBUG_HFCMULTI_FIFO)
+				printk(KERN_DEBUG
+				    "%s(card %d): hdlc-frame too large.\n",
+				    __func__, hc->id + 1);
+			skb_trim(*sp, 0);
+			HFC_outb_nodebug(hc, R_INC_RES_FIFO, V_RES_F);
+			HFC_wait_nodebug(hc);
+			return;
+		}
+
+		hc->read_fifo(hc, skb_put(*sp, Zsize), Zsize);
+
+		if (f1 != f2) {
+			/* increment Z2,F2-counter */
+			HFC_outb_nodebug(hc, R_INC_RES_FIFO, V_INC_F);
+			HFC_wait_nodebug(hc);
+			/* check size */
+			if ((*sp)->len < 4) {
+				if (debug & DEBUG_HFCMULTI_FIFO)
+					printk(KERN_DEBUG
+					    "%s(card %d): Frame below minimum "
+					    "size\n", __func__, hc->id + 1);
+				skb_trim(*sp, 0);
+				goto next_frame;
+			}
+			/* there is at least one complete frame, check crc */
+			if ((*sp)->data[(*sp)->len - 1]) {
+				if (debug & DEBUG_HFCMULTI_CRC)
+					printk(KERN_DEBUG
+					    "%s: CRC-error\n", __func__);
+				skb_trim(*sp, 0);
+				goto next_frame;
+			}
+			skb_trim(*sp, (*sp)->len - 3);
+			if ((*sp)->len < MISDN_COPY_SIZE) {
+				skb = *sp;
+				*sp = mI_alloc_skb(skb->len, GFP_ATOMIC);
+				if (*sp) {
+					memcpy(skb_put(*sp, skb->len),
+					    skb->data, skb->len);
+					skb_trim(skb, 0);
+				} else {
+					printk(KERN_DEBUG "%s: No mem\n",
+					    __func__);
+					*sp = skb;
+					skb = NULL;
+				}
+			} else {
+				skb = NULL;
+			}
+			if (debug & DEBUG_HFCMULTI_FIFO) {
+				printk(KERN_DEBUG "%s(card %d):",
+					__func__, hc->id + 1);
+				temp = 0;
+				while (temp < (*sp)->len)
+					printk(" %02x", (*sp)->data[temp++]);
+				printk("\n");
+			}
+			if (dch)
+				recv_Dchannel(dch);
+			else
+				recv_Bchannel(bch);
+			*sp = skb;
+			again++;
+			goto next_frame;
+		}
+		/* there is an incomplete frame */
+	} else {
+		/* transparent */
+		if (Zsize > skb_tailroom(*sp))
+			Zsize = skb_tailroom(*sp);
+		hc->read_fifo(hc, skb_put(*sp, Zsize), Zsize);
+		if (((*sp)->len) < MISDN_COPY_SIZE) {
+			skb = *sp;
+			*sp = mI_alloc_skb(skb->len, GFP_ATOMIC);
+			if (*sp) {
+				memcpy(skb_put(*sp, skb->len),
+				    skb->data, skb->len);
+				skb_trim(skb, 0);
+			} else {
+				printk(KERN_DEBUG "%s: No mem\n", __func__);
+				*sp = skb;
+				skb = NULL;
+			}
+		} else {
+			skb = NULL;
+		}
+		if (debug & DEBUG_HFCMULTI_FIFO)
+			printk(KERN_DEBUG
+			    "%s(card %d): fifo(%d) reading %d bytes "
+			    "(z1=%04x, z2=%04x) TRANS\n",
+				__func__, hc->id + 1, ch, Zsize, z1, z2);
+		/* only bch is transparent */
+		recv_Bchannel(bch);
+		*sp = skb;
+	}
+}
+
+
+/*
+ * Interrupt handler
+ */
+static void
+signal_state_up(struct dchannel *dch, int info, char *msg)
+{
+	struct sk_buff	*skb;
+	int		id, data = info;
+
+	if (debug & DEBUG_HFCMULTI_STATE)
+		printk(KERN_DEBUG "%s: %s\n", __func__, msg);
+
+	id = TEI_SAPI | (GROUP_TEI << 8); /* manager address */
+
+	skb = _alloc_mISDN_skb(MPH_INFORMATION_IND, id, sizeof(data), &data,
+		GFP_ATOMIC);
+	if (!skb)
+		return;
+	recv_Dchannel_skb(dch, skb);
+}
+
+static inline void
+handle_timer_irq(struct hfc_multi *hc)
+{
+	int		ch, temp;
+	struct dchannel	*dch;
+	u_long		flags;
+
+	/* process queued resync jobs */
+	if (hc->e1_resync) {
+		/* lock, so e1_resync gets not changed */
+		spin_lock_irqsave(&HFClock, flags);
+		if (hc->e1_resync & 1) {
+			if (debug & DEBUG_HFCMULTI_PLXSD)
+				printk(KERN_DEBUG "Enable SYNC_I\n");
+			HFC_outb(hc, R_SYNC_CTRL, V_EXT_CLK_SYNC);
+			/* disable JATT, if RX_SYNC is set */
+			if (test_bit(HFC_CHIP_RX_SYNC, &hc->chip))
+				HFC_outb(hc, R_SYNC_OUT, V_SYNC_E1_RX);
+		}
+		if (hc->e1_resync & 2) {
+			if (debug & DEBUG_HFCMULTI_PLXSD)
+				printk(KERN_DEBUG "Enable jatt PLL\n");
+			HFC_outb(hc, R_SYNC_CTRL, V_SYNC_OFFS);
+		}
+		if (hc->e1_resync & 4) {
+			if (debug & DEBUG_HFCMULTI_PLXSD)
+				printk(KERN_DEBUG
+				    "Enable QUARTZ for HFC-E1\n");
+			/* set jatt to quartz */
+			HFC_outb(hc, R_SYNC_CTRL, V_EXT_CLK_SYNC
+				| V_JATT_OFF);
+			/* switch to JATT, in case it is not already */
+			HFC_outb(hc, R_SYNC_OUT, 0);
+		}
+		hc->e1_resync = 0;
+		spin_unlock_irqrestore(&HFClock, flags);
+	}
+
+	if (hc->type != 1 || hc->e1_state == 1)
+		for (ch = 0; ch <= 31; ch++) {
+			if (hc->created[hc->chan[ch].port]) {
+				hfcmulti_tx(hc, ch);
+				/* fifo is started when switching to rx-fifo */
+				hfcmulti_rx(hc, ch);
+				if (hc->chan[ch].dch &&
+				    hc->chan[ch].nt_timer > -1) {
+					dch = hc->chan[ch].dch;
+					if (!(--hc->chan[ch].nt_timer)) {
+						schedule_event(dch,
+						    FLG_PHCHANGE);
+						if (debug &
+						    DEBUG_HFCMULTI_STATE)
+							printk(KERN_DEBUG
+							    "%s: nt_timer at "
+							    "state %x\n",
+							    __func__,
+							    dch->state);
+					}
+				}
+			}
+		}
+	if (hc->type == 1 && hc->created[0]) {
+		dch = hc->chan[hc->dslot].dch;
+		if (test_bit(HFC_CFG_REPORT_LOS, &hc->chan[hc->dslot].cfg)) {
+			/* LOS */
+			temp = HFC_inb_nodebug(hc, R_SYNC_STA) & V_SIG_LOS;
+			if (!temp && hc->chan[hc->dslot].los)
+				signal_state_up(dch, L1_SIGNAL_LOS_ON,
+				    "LOS detected");
+			if (temp && !hc->chan[hc->dslot].los)
+				signal_state_up(dch, L1_SIGNAL_LOS_OFF,
+				    "LOS gone");
+			hc->chan[hc->dslot].los = temp;
+		}
+		if (test_bit(HFC_CFG_REPORT_AIS, &hc->chan[hc->dslot].cfg)) {
+			/* AIS */
+			temp = HFC_inb_nodebug(hc, R_SYNC_STA) & V_AIS;
+			if (!temp && hc->chan[hc->dslot].ais)
+				signal_state_up(dch, L1_SIGNAL_AIS_ON,
+				    "AIS detected");
+			if (temp && !hc->chan[hc->dslot].ais)
+				signal_state_up(dch, L1_SIGNAL_AIS_OFF,
+				    "AIS gone");
+			hc->chan[hc->dslot].ais = temp;
+		}
+		if (test_bit(HFC_CFG_REPORT_SLIP, &hc->chan[hc->dslot].cfg)) {
+			/* SLIP */
+			temp = HFC_inb_nodebug(hc, R_SLIP) & V_FOSLIP_RX;
+			if (!temp && hc->chan[hc->dslot].slip_rx)
+				signal_state_up(dch, L1_SIGNAL_SLIP_RX,
+				    " bit SLIP detected RX");
+			hc->chan[hc->dslot].slip_rx = temp;
+			temp = HFC_inb_nodebug(hc, R_SLIP) & V_FOSLIP_TX;
+			if (!temp && hc->chan[hc->dslot].slip_tx)
+				signal_state_up(dch, L1_SIGNAL_SLIP_TX,
+				    " bit SLIP detected TX");
+			hc->chan[hc->dslot].slip_tx = temp;
+		}
+		if (test_bit(HFC_CFG_REPORT_RDI, &hc->chan[hc->dslot].cfg)) {
+			/* RDI */
+			temp = HFC_inb_nodebug(hc, R_RX_SL0_0) & V_A;
+			if (!temp && hc->chan[hc->dslot].rdi)
+				signal_state_up(dch, L1_SIGNAL_RDI_ON,
+				    "RDI detected");
+			if (temp && !hc->chan[hc->dslot].rdi)
+				signal_state_up(dch, L1_SIGNAL_RDI_OFF,
+				    "RDI gone");
+			hc->chan[hc->dslot].rdi = temp;
+		}
+		temp = HFC_inb_nodebug(hc, R_JATT_DIR);
+		switch (hc->chan[hc->dslot].sync) {
+		case 0:
+			if ((temp & 0x60) == 0x60) {
+				if (debug & DEBUG_HFCMULTI_SYNC)
+					printk(KERN_DEBUG
+					    "%s: (id=%d) E1 now "
+					    "in clock sync\n",
+					    __func__, hc->id);
+				HFC_outb(hc, R_RX_OFF,
+				    hc->chan[hc->dslot].jitter | V_RX_INIT);
+				HFC_outb(hc, R_TX_OFF,
+				    hc->chan[hc->dslot].jitter | V_RX_INIT);
+				hc->chan[hc->dslot].sync = 1;
+				goto check_framesync;
+			}
+			break;
+		case 1:
+			if ((temp & 0x60) != 0x60) {
+				if (debug & DEBUG_HFCMULTI_SYNC)
+					printk(KERN_DEBUG
+					    "%s: (id=%d) E1 "
+					    "lost clock sync\n",
+					    __func__, hc->id);
+				hc->chan[hc->dslot].sync = 0;
+				break;
+			}
+check_framesync:
+			temp = HFC_inb_nodebug(hc, R_SYNC_STA);
+			if (temp == 0x27) {
+				if (debug & DEBUG_HFCMULTI_SYNC)
+					printk(KERN_DEBUG
+					    "%s: (id=%d) E1 "
+					    "now in frame sync\n",
+					    __func__, hc->id);
+				hc->chan[hc->dslot].sync = 2;
+			}
+			break;
+		case 2:
+			if ((temp & 0x60) != 0x60) {
+				if (debug & DEBUG_HFCMULTI_SYNC)
+					printk(KERN_DEBUG
+					    "%s: (id=%d) E1 lost "
+					    "clock & frame sync\n",
+					    __func__, hc->id);
+				hc->chan[hc->dslot].sync = 0;
+				break;
+			}
+			temp = HFC_inb_nodebug(hc, R_SYNC_STA);
+			if (temp != 0x27) {
+				if (debug & DEBUG_HFCMULTI_SYNC)
+					printk(KERN_DEBUG
+					    "%s: (id=%d) E1 "
+					    "lost frame sync\n",
+					    __func__, hc->id);
+				hc->chan[hc->dslot].sync = 1;
+			}
+			break;
+		}
+	}
+
+	if (test_bit(HFC_CHIP_WATCHDOG, &hc->chip))
+		hfcmulti_watchdog(hc);
+
+	if (hc->leds)
+		hfcmulti_leds(hc);
+}
+
+static void
+ph_state_irq(struct hfc_multi *hc, u_char r_irq_statech)
+{
+	struct dchannel	*dch;
+	int		ch;
+	int		active;
+	u_char		st_status, temp;
+
+	/* state machine */
+	for (ch = 0; ch <= 31; ch++) {
+		if (hc->chan[ch].dch) {
+			dch = hc->chan[ch].dch;
+			if (r_irq_statech & 1) {
+				HFC_outb_nodebug(hc, R_ST_SEL,
+					hc->chan[ch].port);
+				/* undocumented: delay after R_ST_SEL */
+				udelay(1);
+				/* undocumented: status changes during read */
+				st_status = HFC_inb_nodebug(hc, A_ST_RD_STATE);
+				while (st_status != (temp =
+					HFC_inb_nodebug(hc, A_ST_RD_STATE))) {
+					if (debug & DEBUG_HFCMULTI_STATE)
+						printk(KERN_DEBUG "%s: reread "
+						    "STATE because %d!=%d\n",
+						    __func__, temp,
+						    st_status);
+					st_status = temp; /* repeat */
+				}
+
+				/* Speech Design TE-sync indication */
+				if (test_bit(HFC_CHIP_PLXSD, &hc->chip) &&
+					dch->dev.D.protocol == ISDN_P_TE_S0) {
+					if (st_status & V_FR_SYNC_ST)
+						hc->syncronized |=
+						    (1 << hc->chan[ch].port);
+					else
+						hc->syncronized &=
+						   ~(1 << hc->chan[ch].port);
+				}
+				dch->state = st_status & 0x0f;
+				if (dch->dev.D.protocol == ISDN_P_NT_S0)
+					active = 3;
+				else
+					active = 7;
+				if (dch->state == active) {
+					HFC_outb_nodebug(hc, R_FIFO,
+						(ch << 1) | 1);
+					HFC_wait_nodebug(hc);
+					HFC_outb_nodebug(hc,
+						R_INC_RES_FIFO, V_RES_F);
+					HFC_wait_nodebug(hc);
+					dch->tx_idx = 0;
+				}
+				schedule_event(dch, FLG_PHCHANGE);
+				if (debug & DEBUG_HFCMULTI_STATE)
+					printk(KERN_DEBUG
+					    "%s: S/T newstate %x port %d\n",
+					    __func__, dch->state,
+					    hc->chan[ch].port);
+			}
+			r_irq_statech >>= 1;
+		}
+	}
+	if (test_bit(HFC_CHIP_PLXSD, &hc->chip))
+		plxsd_checksync(hc, 0);
+}
+
+static void
+fifo_irq(struct hfc_multi *hc, int block)
+{
+	int	ch, j;
+	struct dchannel	*dch;
+	struct bchannel	*bch;
+	u_char r_irq_fifo_bl;
+
+	r_irq_fifo_bl = HFC_inb_nodebug(hc, R_IRQ_FIFO_BL0 + block);
+	j = 0;
+	while (j < 8) {
+		ch = (block << 2) + (j >> 1);
+		dch = hc->chan[ch].dch;
+		bch = hc->chan[ch].bch;
+		if (((!dch) && (!bch)) || (!hc->created[hc->chan[ch].port])) {
+			j += 2;
+			continue;
+		}
+		if (dch && (r_irq_fifo_bl & (1 << j)) &&
+		    test_bit(FLG_ACTIVE, &dch->Flags)) {
+			hfcmulti_tx(hc, ch);
+			/* start fifo */
+			HFC_outb_nodebug(hc, R_FIFO, 0);
+			HFC_wait_nodebug(hc);
+		}
+		if (bch && (r_irq_fifo_bl & (1 << j)) &&
+		    test_bit(FLG_ACTIVE, &bch->Flags)) {
+			hfcmulti_tx(hc, ch);
+			/* start fifo */
+			HFC_outb_nodebug(hc, R_FIFO, 0);
+			HFC_wait_nodebug(hc);
+		}
+		j++;
+		if (dch && (r_irq_fifo_bl & (1 << j)) &&
+		    test_bit(FLG_ACTIVE, &dch->Flags)) {
+			hfcmulti_rx(hc, ch);
+		}
+		if (bch && (r_irq_fifo_bl & (1 << j)) &&
+		    test_bit(FLG_ACTIVE, &bch->Flags)) {
+			hfcmulti_rx(hc, ch);
+		}
+		j++;
+	}
+}
+
+#ifdef IRQ_DEBUG
+int irqsem;
+#endif
+static irqreturn_t
+hfcmulti_interrupt(int intno, void *dev_id)
+{
+#ifdef IRQCOUNT_DEBUG
+	static int iq1 = 0, iq2 = 0, iq3 = 0, iq4 = 0,
+	    iq5 = 0, iq6 = 0, iqcnt = 0;
+#endif
+	static int		count;
+	struct hfc_multi	*hc = dev_id;
+	struct dchannel		*dch;
+	u_char			r_irq_statech, status, r_irq_misc, r_irq_oview;
+	int			i;
+	u_short			*plx_acc, wval;
+	u_char			e1_syncsta, temp;
+	u_long			flags;
+
+	if (!hc) {
+		printk(KERN_ERR "HFC-multi: Spurious interrupt!\n");
+		return IRQ_NONE;
+	}
+
+	spin_lock(&hc->lock);
+
+#ifdef IRQ_DEBUG
+	if (irqsem)
+		printk(KERN_ERR "irq for card %d during irq from "
+		"card %d, this is no bug.\n", hc->id + 1, irqsem);
+	irqsem = hc->id + 1;
+#endif
+
+	if (test_bit(HFC_CHIP_PLXSD, &hc->chip)) {
+		spin_lock_irqsave(&plx_lock, flags);
+		plx_acc = (u_short *)(hc->plx_membase + PLX_INTCSR);
+		wval = readw(plx_acc);
+		spin_unlock_irqrestore(&plx_lock, flags);
+		if (!(wval & PLX_INTCSR_LINTI1_STATUS))
+			goto irq_notforus;
+	}
+
+	status = HFC_inb_nodebug(hc, R_STATUS);
+	r_irq_statech = HFC_inb_nodebug(hc, R_IRQ_STATECH);
+#ifdef IRQCOUNT_DEBUG
+	if (r_irq_statech)
+		iq1++;
+	if (status & V_DTMF_STA)
+		iq2++;
+	if (status & V_LOST_STA)
+		iq3++;
+	if (status & V_EXT_IRQSTA)
+		iq4++;
+	if (status & V_MISC_IRQSTA)
+		iq5++;
+	if (status & V_FR_IRQSTA)
+		iq6++;
+	if (iqcnt++ > 5000) {
+		printk(KERN_ERR "iq1:%x iq2:%x iq3:%x iq4:%x iq5:%x iq6:%x\n",
+		    iq1, iq2, iq3, iq4, iq5, iq6);
+		iqcnt = 0;
+	}
+#endif
+	if (!r_irq_statech &&
+	    !(status & (V_DTMF_STA | V_LOST_STA | V_EXT_IRQSTA |
+	    V_MISC_IRQSTA | V_FR_IRQSTA))) {
+		/* irq is not for us */
+		goto irq_notforus;
+	}
+	hc->irqcnt++;
+	if (r_irq_statech) {
+		if (hc->type != 1)
+			ph_state_irq(hc, r_irq_statech);
+	}
+	if (status & V_EXT_IRQSTA)
+		; /* external IRQ */
+	if (status & V_LOST_STA) {
+		/* LOST IRQ */
+		HFC_outb(hc, R_INC_RES_FIFO, V_RES_LOST); /* clear irq! */
+	}
+	if (status & V_MISC_IRQSTA) {
+		/* misc IRQ */
+		r_irq_misc = HFC_inb_nodebug(hc, R_IRQ_MISC);
+		if (r_irq_misc & V_STA_IRQ) {
+			if (hc->type == 1) {
+				/* state machine */
+				dch = hc->chan[hc->dslot].dch;
+				e1_syncsta = HFC_inb_nodebug(hc, R_SYNC_STA);
+				if (test_bit(HFC_CHIP_PLXSD, &hc->chip)
+				 && hc->e1_getclock) {
+					if (e1_syncsta & V_FR_SYNC_E1)
+						hc->syncronized = 1;
+					else
+						hc->syncronized = 0;
+				}
+				/* undocumented: status changes during read */
+				dch->state = HFC_inb_nodebug(hc, R_E1_RD_STA);
+				while (dch->state != (temp =
+					HFC_inb_nodebug(hc, R_E1_RD_STA))) {
+					if (debug & DEBUG_HFCMULTI_STATE)
+						printk(KERN_DEBUG "%s: reread "
+						    "STATE because %d!=%d\n",
+						    __func__, temp,
+						    dch->state);
+					dch->state = temp; /* repeat */
+				}
+				dch->state = HFC_inb_nodebug(hc, R_E1_RD_STA)
+					& 0x7;
+				schedule_event(dch, FLG_PHCHANGE);
+				if (debug & DEBUG_HFCMULTI_STATE)
+					printk(KERN_DEBUG
+					    "%s: E1 (id=%d) newstate %x\n",
+					    __func__, hc->id, dch->state);
+				if (test_bit(HFC_CHIP_PLXSD, &hc->chip))
+					plxsd_checksync(hc, 0);
+			}
+		}
+		if (r_irq_misc & V_TI_IRQ)
+			handle_timer_irq(hc);
+
+		if (r_irq_misc & V_DTMF_IRQ) {
+			/* -> DTMF IRQ */
+			hfcmulti_dtmf(hc);
+		}
+		/* TODO: REPLACE !!!! 125 us Interrupts are not acceptable  */
+		if (r_irq_misc & V_IRQ_PROC) {
+			/* IRQ every 125us */
+			count++;
+			/* generate 1kHz signal */
+			if (count == 8) {
+				if (hfc_interrupt)
+					hfc_interrupt();
+				count = 0;
+			}
+		}
+
+	}
+	if (status & V_FR_IRQSTA) {
+		/* FIFO IRQ */
+		r_irq_oview = HFC_inb_nodebug(hc, R_IRQ_OVIEW);
+		for (i = 0; i < 8; i++) {
+			if (r_irq_oview & (1 << i))
+				fifo_irq(hc, i);
+		}
+	}
+
+#ifdef IRQ_DEBUG
+	irqsem = 0;
+#endif
+	spin_unlock(&hc->lock);
+	return IRQ_HANDLED;
+
+irq_notforus:
+#ifdef IRQ_DEBUG
+	irqsem = 0;
+#endif
+	spin_unlock(&hc->lock);
+	return IRQ_NONE;
+}
+
+
+/*
+ * timer callback for D-chan busy resolution. Currently no function
+ */
+
+static void
+hfcmulti_dbusy_timer(struct hfc_multi *hc)
+{
+}
+
+
+/*
+ * activate/deactivate hardware for selected channels and mode
+ *
+ * configure B-channel with the given protocol
+ * ch eqals to the HFC-channel (0-31)
+ * ch is the number of channel (0-4,4-7,8-11,12-15,16-19,20-23,24-27,28-31
+ * for S/T, 1-31 for E1)
+ * the hdlc interrupts will be set/unset
+ */
+static int
+mode_hfcmulti(struct hfc_multi *hc, int ch, int protocol, int slot_tx,
+    int bank_tx, int slot_rx, int bank_rx)
+{
+	int flow_tx = 0, flow_rx = 0, routing = 0;
+	int oslot_tx, oslot_rx;
+	int conf;
+
+	if (ch < 0 || ch > 31)
+		return EINVAL;
+	oslot_tx = hc->chan[ch].slot_tx;
+	oslot_rx = hc->chan[ch].slot_rx;
+	conf = hc->chan[ch].conf;
+
+	if (debug & DEBUG_HFCMULTI_MODE)
+		printk(KERN_DEBUG
+		    "%s: card %d channel %d protocol %x slot old=%d new=%d "
+		    "bank new=%d (TX) slot old=%d new=%d bank new=%d (RX)\n",
+		    __func__, hc->id, ch, protocol, oslot_tx, slot_tx,
+		    bank_tx, oslot_rx, slot_rx, bank_rx);
+
+	if (oslot_tx >= 0 && slot_tx != oslot_tx) {
+		/* remove from slot */
+		if (debug & DEBUG_HFCMULTI_MODE)
+			printk(KERN_DEBUG "%s: remove from slot %d (TX)\n",
+			    __func__, oslot_tx);
+		if (hc->slot_owner[oslot_tx<<1] == ch) {
+			HFC_outb(hc, R_SLOT, oslot_tx << 1);
+			HFC_outb(hc, A_SL_CFG, 0);
+			HFC_outb(hc, A_CONF, 0);
+			hc->slot_owner[oslot_tx<<1] = -1;
+		} else {
+			if (debug & DEBUG_HFCMULTI_MODE)
+				printk(KERN_DEBUG
+				    "%s: we are not owner of this tx slot "
+				    "anymore, channel %d is.\n",
+				    __func__, hc->slot_owner[oslot_tx<<1]);
+		}
+	}
+
+	if (oslot_rx >= 0 && slot_rx != oslot_rx) {
+		/* remove from slot */
+		if (debug & DEBUG_HFCMULTI_MODE)
+			printk(KERN_DEBUG
+			    "%s: remove from slot %d (RX)\n",
+			    __func__, oslot_rx);
+		if (hc->slot_owner[(oslot_rx << 1) | 1] == ch) {
+			HFC_outb(hc, R_SLOT, (oslot_rx << 1) | V_SL_DIR);
+			HFC_outb(hc, A_SL_CFG, 0);
+			hc->slot_owner[(oslot_rx << 1) | 1] = -1;
+		} else {
+			if (debug & DEBUG_HFCMULTI_MODE)
+				printk(KERN_DEBUG
+				    "%s: we are not owner of this rx slot "
+				    "anymore, channel %d is.\n",
+				    __func__,
+				    hc->slot_owner[(oslot_rx << 1) | 1]);
+		}
+	}
+
+	if (slot_tx < 0) {
+		flow_tx = 0x80; /* FIFO->ST */
+		/* disable pcm slot */
+		hc->chan[ch].slot_tx = -1;
+		hc->chan[ch].bank_tx = 0;
+	} else {
+		/* set pcm slot */
+		if (hc->chan[ch].txpending)
+			flow_tx = 0x80; /* FIFO->ST */
+		else
+			flow_tx = 0xc0; /* PCM->ST */
+		/* put on slot */
+		routing = bank_tx ? 0xc0 : 0x80;
+		if (conf >= 0 || bank_tx > 1)
+			routing = 0x40; /* loop */
+		if (debug & DEBUG_HFCMULTI_MODE)
+			printk(KERN_DEBUG "%s: put channel %d to slot %d bank"
+			    " %d flow %02x routing %02x conf %d (TX)\n",
+			    __func__, ch, slot_tx, bank_tx,
+			    flow_tx, routing, conf);
+		HFC_outb(hc, R_SLOT, slot_tx << 1);
+		HFC_outb(hc, A_SL_CFG, (ch<<1) | routing);
+		HFC_outb(hc, A_CONF, (conf < 0) ? 0 : (conf | V_CONF_SL));
+		hc->slot_owner[slot_tx << 1] = ch;
+		hc->chan[ch].slot_tx = slot_tx;
+		hc->chan[ch].bank_tx = bank_tx;
+	}
+	if (slot_rx < 0) {
+		/* disable pcm slot */
+		flow_rx = 0x80; /* ST->FIFO */
+		hc->chan[ch].slot_rx = -1;
+		hc->chan[ch].bank_rx = 0;
+	} else {
+		/* set pcm slot */
+		if (hc->chan[ch].txpending)
+			flow_rx = 0x80; /* ST->FIFO */
+		else
+			flow_rx = 0xc0; /* ST->(FIFO,PCM) */
+		/* put on slot */
+		routing = bank_rx?0x80:0xc0; /* reversed */
+		if (conf >= 0 || bank_rx > 1)
+			routing = 0x40; /* loop */
+		if (debug & DEBUG_HFCMULTI_MODE)
+			printk(KERN_DEBUG "%s: put channel %d to slot %d bank"
+			    " %d flow %02x routing %02x conf %d (RX)\n",
+			    __func__, ch, slot_rx, bank_rx,
+			    flow_rx, routing, conf);
+		HFC_outb(hc, R_SLOT, (slot_rx<<1) | V_SL_DIR);
+		HFC_outb(hc, A_SL_CFG, (ch<<1) | V_CH_DIR | routing);
+		hc->slot_owner[(slot_rx<<1)|1] = ch;
+		hc->chan[ch].slot_rx = slot_rx;
+		hc->chan[ch].bank_rx = bank_rx;
+	}
+
+	switch (protocol) {
+	case (ISDN_P_NONE):
+		/* disable TX fifo */
+		HFC_outb(hc, R_FIFO, ch << 1);
+		HFC_wait(hc);
+		HFC_outb(hc, A_CON_HDLC, flow_tx | 0x00 | V_IFF);
+		HFC_outb(hc, A_SUBCH_CFG, 0);
+		HFC_outb(hc, A_IRQ_MSK, 0);
+		HFC_outb(hc, R_INC_RES_FIFO, V_RES_F);
+		HFC_wait(hc);
+		/* disable RX fifo */
+		HFC_outb(hc, R_FIFO, (ch<<1)|1);
+		HFC_wait(hc);
+		HFC_outb(hc, A_CON_HDLC, flow_rx | 0x00);
+		HFC_outb(hc, A_SUBCH_CFG, 0);
+		HFC_outb(hc, A_IRQ_MSK, 0);
+		HFC_outb(hc, R_INC_RES_FIFO, V_RES_F);
+		HFC_wait(hc);
+		if (hc->chan[ch].bch && hc->type != 1) {
+			hc->hw.a_st_ctrl0[hc->chan[ch].port] &=
+			    ((ch & 0x3) == 0)? ~V_B1_EN: ~V_B2_EN;
+			HFC_outb(hc, R_ST_SEL, hc->chan[ch].port);
+			/* undocumented: delay after R_ST_SEL */
+			udelay(1);
+			HFC_outb(hc, A_ST_CTRL0,
+			    hc->hw.a_st_ctrl0[hc->chan[ch].port]);
+		}
+		if (hc->chan[ch].bch) {
+			test_and_clear_bit(FLG_HDLC, &hc->chan[ch].bch->Flags);
+			test_and_clear_bit(FLG_TRANSPARENT,
+			    &hc->chan[ch].bch->Flags);
+		}
+		break;
+	case (ISDN_P_B_RAW): /* B-channel */
+
+		if (test_bit(HFC_CHIP_B410P, &hc->chip) &&
+		    (hc->chan[ch].slot_rx < 0) &&
+		    (hc->chan[ch].slot_tx < 0)) {
+
+			printk(KERN_DEBUG
+			    "Setting B-channel %d to echo cancelable "
+			    "state on PCM slot %d\n", ch,
+			    ((ch / 4) * 8) + ((ch % 4) * 4) + 1);
+			printk(KERN_DEBUG
+			    "Enabling pass through for channel\n");
+			vpm_out(hc, ch, ((ch / 4) * 8) +
+			    ((ch % 4) * 4) + 1, 0x01);
+			/* rx path */
+			/* S/T -> PCM */
+			HFC_outb(hc, R_FIFO, (ch << 1));
+			HFC_wait(hc);
+			HFC_outb(hc, A_CON_HDLC, 0xc0 | V_HDLC_TRP | V_IFF);
+			HFC_outb(hc, R_SLOT, (((ch / 4) * 8) +
+			    ((ch % 4) * 4) + 1) << 1);
+			HFC_outb(hc, A_SL_CFG, 0x80 | (ch << 1));
+
+			/* PCM -> FIFO */
+			HFC_outb(hc, R_FIFO, 0x20 | (ch << 1) | 1);
+			HFC_wait(hc);
+			HFC_outb(hc, A_CON_HDLC, 0x20 | V_HDLC_TRP | V_IFF);
+			HFC_outb(hc, A_SUBCH_CFG, 0);
+			HFC_outb(hc, A_IRQ_MSK, 0);
+			HFC_outb(hc, R_INC_RES_FIFO, V_RES_F);
+			HFC_wait(hc);
+			HFC_outb(hc, R_SLOT, ((((ch / 4) * 8) +
+			    ((ch % 4) * 4) + 1) << 1) | 1);
+			HFC_outb(hc, A_SL_CFG, 0x80 | 0x20 | (ch << 1) | 1);
+
+			/* tx path */
+			/* PCM -> S/T */
+			HFC_outb(hc, R_FIFO, (ch << 1) | 1);
+			HFC_wait(hc);
+			HFC_outb(hc, A_CON_HDLC, 0xc0 | V_HDLC_TRP | V_IFF);
+			HFC_outb(hc, R_SLOT, ((((ch / 4) * 8) +
+			    ((ch % 4) * 4)) << 1) | 1);
+			HFC_outb(hc, A_SL_CFG, 0x80 | 0x40 | (ch << 1) | 1);
+
+			/* FIFO -> PCM */
+			HFC_outb(hc, R_FIFO, 0x20 | (ch << 1));
+			HFC_wait(hc);
+			HFC_outb(hc, A_CON_HDLC, 0x20 | V_HDLC_TRP | V_IFF);
+			HFC_outb(hc, A_SUBCH_CFG, 0);
+			HFC_outb(hc, A_IRQ_MSK, 0);
+			HFC_outb(hc, R_INC_RES_FIFO, V_RES_F);
+			HFC_wait(hc);
+			/* tx silence */
+			HFC_outb_nodebug(hc, A_FIFO_DATA0_NOINC, silence);
+			HFC_outb(hc, R_SLOT, (((ch / 4) * 8) +
+			    ((ch % 4) * 4)) << 1);
+			HFC_outb(hc, A_SL_CFG, 0x80 | 0x20 | (ch << 1));
+		} else {
+			/* enable TX fifo */
+			HFC_outb(hc, R_FIFO, ch << 1);
+			HFC_wait(hc);
+			HFC_outb(hc, A_CON_HDLC, flow_tx | 0x00 |
+			    V_HDLC_TRP | V_IFF);
+			HFC_outb(hc, A_SUBCH_CFG, 0);
+			HFC_outb(hc, A_IRQ_MSK, 0);
+			HFC_outb(hc, R_INC_RES_FIFO, V_RES_F);
+			HFC_wait(hc);
+			/* tx silence */
+			HFC_outb_nodebug(hc, A_FIFO_DATA0_NOINC, silence);
+			/* enable RX fifo */
+			HFC_outb(hc, R_FIFO, (ch<<1)|1);
+			HFC_wait(hc);
+			HFC_outb(hc, A_CON_HDLC, flow_rx | 0x00 | V_HDLC_TRP);
+			HFC_outb(hc, A_SUBCH_CFG, 0);
+			HFC_outb(hc, A_IRQ_MSK, 0);
+			HFC_outb(hc, R_INC_RES_FIFO, V_RES_F);
+			HFC_wait(hc);
+		}
+		if (hc->type != 1) {
+			hc->hw.a_st_ctrl0[hc->chan[ch].port] |=
+			    ((ch & 0x3) == 0) ? V_B1_EN : V_B2_EN;
+			HFC_outb(hc, R_ST_SEL, hc->chan[ch].port);
+			/* undocumented: delay after R_ST_SEL */
+			udelay(1);
+			HFC_outb(hc, A_ST_CTRL0,
+			    hc->hw.a_st_ctrl0[hc->chan[ch].port]);
+		}
+		if (hc->chan[ch].bch)
+			test_and_set_bit(FLG_TRANSPARENT,
+			    &hc->chan[ch].bch->Flags);
+		break;
+	case (ISDN_P_B_HDLC): /* B-channel */
+	case (ISDN_P_TE_S0): /* D-channel */
+	case (ISDN_P_NT_S0):
+	case (ISDN_P_TE_E1):
+	case (ISDN_P_NT_E1):
+		/* enable TX fifo */
+		HFC_outb(hc, R_FIFO, ch<<1);
+		HFC_wait(hc);
+		if (hc->type == 1 || hc->chan[ch].bch) {
+			/* E1 or B-channel */
+			HFC_outb(hc, A_CON_HDLC, flow_tx | 0x04);
+			HFC_outb(hc, A_SUBCH_CFG, 0);
+		} else {
+			/* D-Channel without HDLC fill flags */
+			HFC_outb(hc, A_CON_HDLC, flow_tx | 0x04 | V_IFF);
+			HFC_outb(hc, A_SUBCH_CFG, 2);
+		}
+		HFC_outb(hc, A_IRQ_MSK, V_IRQ);
+		HFC_outb(hc, R_INC_RES_FIFO, V_RES_F);
+		HFC_wait(hc);
+		/* enable RX fifo */
+		HFC_outb(hc, R_FIFO, (ch<<1)|1);
+		HFC_wait(hc);
+		HFC_outb(hc, A_CON_HDLC, flow_rx | 0x04);
+		if (hc->type == 1 || hc->chan[ch].bch)
+			HFC_outb(hc, A_SUBCH_CFG, 0); /* full 8 bits */
+		else
+			HFC_outb(hc, A_SUBCH_CFG, 2); /* 2 bits dchannel */
+		HFC_outb(hc, A_IRQ_MSK, V_IRQ);
+		HFC_outb(hc, R_INC_RES_FIFO, V_RES_F);
+		HFC_wait(hc);
+		if (hc->chan[ch].bch) {
+			test_and_set_bit(FLG_HDLC, &hc->chan[ch].bch->Flags);
+			if (hc->type != 1) {
+				hc->hw.a_st_ctrl0[hc->chan[ch].port] |=
+				  ((ch&0x3) == 0) ? V_B1_EN : V_B2_EN;
+				HFC_outb(hc, R_ST_SEL, hc->chan[ch].port);
+				/* undocumented: delay after R_ST_SEL */
+				udelay(1);
+				HFC_outb(hc, A_ST_CTRL0,
+				  hc->hw.a_st_ctrl0[hc->chan[ch].port]);
+			}
+		}
+		break;
+	default:
+		printk(KERN_DEBUG "%s: protocol not known %x\n",
+		    __func__, protocol);
+		hc->chan[ch].protocol = ISDN_P_NONE;
+		return -ENOPROTOOPT;
+	}
+	hc->chan[ch].protocol = protocol;
+	return 0;
+}
+
+
+/*
+ * connect/disconnect PCM
+ */
+
+static void
+hfcmulti_pcm(struct hfc_multi *hc, int ch, int slot_tx, int bank_tx,
+    int slot_rx, int bank_rx)
+{
+	if (slot_rx < 0 || slot_rx < 0 || bank_tx < 0 || bank_rx < 0) {
+		/* disable PCM */
+		mode_hfcmulti(hc, ch, hc->chan[ch].protocol, -1, 0, -1, 0);
+		return;
+	}
+
+	/* enable pcm */
+	mode_hfcmulti(hc, ch, hc->chan[ch].protocol, slot_tx, bank_tx,
+		slot_rx, bank_rx);
+}
+
+/*
+ * set/disable conference
+ */
+
+static void
+hfcmulti_conf(struct hfc_multi *hc, int ch, int num)
+{
+	if (num >= 0 && num <= 7)
+		hc->chan[ch].conf = num;
+	else
+		hc->chan[ch].conf = -1;
+	mode_hfcmulti(hc, ch, hc->chan[ch].protocol, hc->chan[ch].slot_tx,
+	    hc->chan[ch].bank_tx, hc->chan[ch].slot_rx,
+	    hc->chan[ch].bank_rx);
+}
+
+
+/*
+ * set/disable sample loop
+ */
+
+/* NOTE: this function is experimental and therefore disabled */
+
+/*
+ * Layer 1 callback function
+ */
+static int
+hfcm_l1callback(struct dchannel *dch, u_int cmd)
+{
+	struct hfc_multi	*hc = dch->hw;
+	u_long	flags;
+
+	switch (cmd) {
+	case INFO3_P8:
+	case INFO3_P10:
+		break;
+	case HW_RESET_REQ:
+		/* start activation */
+		spin_lock_irqsave(&hc->lock, flags);
+		if (hc->type == 1) {
+			if (debug & DEBUG_HFCMULTI_MSG)
+				printk(KERN_DEBUG
+				    "%s: HW_RESET_REQ no BRI\n",
+				    __func__);
+		} else {
+			HFC_outb(hc, R_ST_SEL, hc->chan[dch->slot].port);
+			/* undocumented: delay after R_ST_SEL */
+			udelay(1);
+			HFC_outb(hc, A_ST_WR_STATE, V_ST_LD_STA | 3); /* F3 */
+			udelay(6); /* wait at least 5,21us */
+			HFC_outb(hc, A_ST_WR_STATE, 3);
+			HFC_outb(hc, A_ST_WR_STATE, 3 | (V_ST_ACT*3));
+				/* activate */
+		}
+		spin_unlock_irqrestore(&hc->lock, flags);
+		l1_event(dch->l1, HW_POWERUP_IND);
+		break;
+	case HW_DEACT_REQ:
+		/* start deactivation */
+		spin_lock_irqsave(&hc->lock, flags);
+		if (hc->type == 1) {
+			if (debug & DEBUG_HFCMULTI_MSG)
+				printk(KERN_DEBUG
+				    "%s: HW_DEACT_REQ no BRI\n",
+				    __func__);
+		} else {
+			HFC_outb(hc, R_ST_SEL, hc->chan[dch->slot].port);
+			/* undocumented: delay after R_ST_SEL */
+			udelay(1);
+			HFC_outb(hc, A_ST_WR_STATE, V_ST_ACT*2);
+				/* deactivate */
+			if (test_bit(HFC_CHIP_PLXSD, &hc->chip)) {
+				hc->syncronized &=
+				   ~(1 << hc->chan[dch->slot].port);
+				plxsd_checksync(hc, 0);
+			}
+		}
+		skb_queue_purge(&dch->squeue);
+		if (dch->tx_skb) {
+			dev_kfree_skb(dch->tx_skb);
+			dch->tx_skb = NULL;
+		}
+		dch->tx_idx = 0;
+		if (dch->rx_skb) {
+			dev_kfree_skb(dch->rx_skb);
+			dch->rx_skb = NULL;
+		}
+		test_and_clear_bit(FLG_TX_BUSY, &dch->Flags);
+		if (test_and_clear_bit(FLG_BUSY_TIMER, &dch->Flags))
+			del_timer(&dch->timer);
+		spin_unlock_irqrestore(&hc->lock, flags);
+		break;
+	case HW_POWERUP_REQ:
+		spin_lock_irqsave(&hc->lock, flags);
+		if (hc->type == 1) {
+			if (debug & DEBUG_HFCMULTI_MSG)
+				printk(KERN_DEBUG
+				    "%s: HW_POWERUP_REQ no BRI\n",
+				    __func__);
+		} else {
+			HFC_outb(hc, R_ST_SEL, hc->chan[dch->slot].port);
+			/* undocumented: delay after R_ST_SEL */
+			udelay(1);
+			HFC_outb(hc, A_ST_WR_STATE, 3 | 0x10); /* activate */
+			udelay(6); /* wait at least 5,21us */
+			HFC_outb(hc, A_ST_WR_STATE, 3); /* activate */
+		}
+		spin_unlock_irqrestore(&hc->lock, flags);
+		break;
+	case PH_ACTIVATE_IND:
+		test_and_set_bit(FLG_ACTIVE, &dch->Flags);
+		_queue_data(&dch->dev.D, cmd, MISDN_ID_ANY, 0, NULL,
+			GFP_ATOMIC);
+		break;
+	case PH_DEACTIVATE_IND:
+		test_and_clear_bit(FLG_ACTIVE, &dch->Flags);
+		_queue_data(&dch->dev.D, cmd, MISDN_ID_ANY, 0, NULL,
+			GFP_ATOMIC);
+		break;
+	default:
+		if (dch->debug & DEBUG_HW)
+			printk(KERN_DEBUG "%s: unknown command %x\n",
+			    __func__, cmd);
+		return -1;
+	}
+	return 0;
+}
+
+/*
+ * Layer2 -> Layer 1 Transfer
+ */
+
+static int
+handle_dmsg(struct mISDNchannel *ch, struct sk_buff *skb)
+{
+	struct mISDNdevice	*dev = container_of(ch, struct mISDNdevice, D);
+	struct dchannel		*dch = container_of(dev, struct dchannel, dev);
+	struct hfc_multi	*hc = dch->hw;
+	struct mISDNhead	*hh = mISDN_HEAD_P(skb);
+	int			ret = -EINVAL;
+	unsigned int		id;
+	u_long			flags;
+
+	switch (hh->prim) {
+	case PH_DATA_REQ:
+		if (skb->len < 1)
+			break;
+		spin_lock_irqsave(&hc->lock, flags);
+		ret = dchannel_senddata(dch, skb);
+		if (ret > 0) { /* direct TX */
+			id = hh->id; /* skb can be freed */
+			hfcmulti_tx(hc, dch->slot);
+			ret = 0;
+			/* start fifo */
+			HFC_outb(hc, R_FIFO, 0);
+			HFC_wait(hc);
+			spin_unlock_irqrestore(&hc->lock, flags);
+			queue_ch_frame(ch, PH_DATA_CNF, id, NULL);
+		} else
+			spin_unlock_irqrestore(&hc->lock, flags);
+		return ret;
+	case PH_ACTIVATE_REQ:
+		if (dch->dev.D.protocol != ISDN_P_TE_S0) {
+			spin_lock_irqsave(&hc->lock, flags);
+			ret = 0;
+			if (debug & DEBUG_HFCMULTI_MSG)
+				printk(KERN_DEBUG
+				    "%s: PH_ACTIVATE port %d (0..%d)\n",
+				    __func__, hc->chan[dch->slot].port,
+				    hc->ports-1);
+			/* start activation */
+			if (hc->type == 1) {
+				ph_state_change(dch);
+				if (debug & DEBUG_HFCMULTI_STATE)
+					printk(KERN_DEBUG
+					    "%s: E1 report state %x \n",
+					    __func__, dch->state);
+			} else {
+				HFC_outb(hc, R_ST_SEL,
+				    hc->chan[dch->slot].port);
+				/* undocumented: delay after R_ST_SEL */
+				udelay(1);
+				HFC_outb(hc, A_ST_WR_STATE, V_ST_LD_STA | 1);
+				    /* G1 */
+				udelay(6); /* wait at least 5,21us */
+				HFC_outb(hc, A_ST_WR_STATE, 1);
+				HFC_outb(hc, A_ST_WR_STATE, 1 |
+				    (V_ST_ACT*3)); /* activate */
+				dch->state = 1;
+			}
+			spin_unlock_irqrestore(&hc->lock, flags);
+		} else
+			ret = l1_event(dch->l1, hh->prim);
+		break;
+	case PH_DEACTIVATE_REQ:
+		test_and_clear_bit(FLG_L2_ACTIVATED, &dch->Flags);
+		if (dch->dev.D.protocol != ISDN_P_TE_S0) {
+			spin_lock_irqsave(&hc->lock, flags);
+			if (debug & DEBUG_HFCMULTI_MSG)
+				printk(KERN_DEBUG
+				    "%s: PH_DEACTIVATE port %d (0..%d)\n",
+				    __func__, hc->chan[dch->slot].port,
+				    hc->ports-1);
+			/* start deactivation */
+			if (hc->type == 1) {
+				if (debug & DEBUG_HFCMULTI_MSG)
+					printk(KERN_DEBUG
+					    "%s: PH_DEACTIVATE no BRI\n",
+					    __func__);
+			} else {
+				HFC_outb(hc, R_ST_SEL,
+				    hc->chan[dch->slot].port);
+				/* undocumented: delay after R_ST_SEL */
+				udelay(1);
+				HFC_outb(hc, A_ST_WR_STATE, V_ST_ACT * 2);
+				    /* deactivate */
+				dch->state = 1;
+			}
+			skb_queue_purge(&dch->squeue);
+			if (dch->tx_skb) {
+				dev_kfree_skb(dch->tx_skb);
+				dch->tx_skb = NULL;
+			}
+			dch->tx_idx = 0;
+			if (dch->rx_skb) {
+				dev_kfree_skb(dch->rx_skb);
+				dch->rx_skb = NULL;
+			}
+			test_and_clear_bit(FLG_TX_BUSY, &dch->Flags);
+			if (test_and_clear_bit(FLG_BUSY_TIMER, &dch->Flags))
+				del_timer(&dch->timer);
+#ifdef FIXME
+			if (test_and_clear_bit(FLG_L1_BUSY, &dch->Flags))
+				dchannel_sched_event(&hc->dch, D_CLEARBUSY);
+#endif
+			ret = 0;
+			spin_unlock_irqrestore(&hc->lock, flags);
+		} else
+			ret = l1_event(dch->l1, hh->prim);
+		break;
+	}
+	if (!ret)
+		dev_kfree_skb(skb);
+	return ret;
+}
+
+static void
+deactivate_bchannel(struct bchannel *bch)
+{
+	struct hfc_multi	*hc = bch->hw;
+	u_long			flags;
+
+	spin_lock_irqsave(&hc->lock, flags);
+	if (test_and_clear_bit(FLG_TX_NEXT, &bch->Flags)) {
+		dev_kfree_skb(bch->next_skb);
+		bch->next_skb = NULL;
+	}
+	if (bch->tx_skb) {
+		dev_kfree_skb(bch->tx_skb);
+		bch->tx_skb = NULL;
+	}
+	bch->tx_idx = 0;
+	if (bch->rx_skb) {
+		dev_kfree_skb(bch->rx_skb);
+		bch->rx_skb = NULL;
+	}
+	hc->chan[bch->slot].coeff_count = 0;
+	test_and_clear_bit(FLG_ACTIVE, &bch->Flags);
+	test_and_clear_bit(FLG_TX_BUSY, &bch->Flags);
+	hc->chan[bch->slot].rx_off = 0;
+	hc->chan[bch->slot].conf = -1;
+	mode_hfcmulti(hc, bch->slot, ISDN_P_NONE, -1, 0, -1, 0);
+	spin_unlock_irqrestore(&hc->lock, flags);
+}
+
+static int
+handle_bmsg(struct mISDNchannel *ch, struct sk_buff *skb)
+{
+	struct bchannel		*bch = container_of(ch, struct bchannel, ch);
+	struct hfc_multi	*hc = bch->hw;
+	int			ret = -EINVAL;
+	struct mISDNhead	*hh = mISDN_HEAD_P(skb);
+	unsigned int		id;
+	u_long			flags;
+
+	switch (hh->prim) {
+	case PH_DATA_REQ:
+		if (!skb->len)
+			break;
+		spin_lock_irqsave(&hc->lock, flags);
+		ret = bchannel_senddata(bch, skb);
+		if (ret > 0) { /* direct TX */
+			id = hh->id; /* skb can be freed */
+			hfcmulti_tx(hc, bch->slot);
+			ret = 0;
+			/* start fifo */
+			HFC_outb_nodebug(hc, R_FIFO, 0);
+			HFC_wait_nodebug(hc);
+			if (!test_bit(FLG_TRANSPARENT, &bch->Flags)) {
+				spin_unlock_irqrestore(&hc->lock, flags);
+				queue_ch_frame(ch, PH_DATA_CNF, id, NULL);
+			} else
+				spin_unlock_irqrestore(&hc->lock, flags);
+		} else
+			spin_unlock_irqrestore(&hc->lock, flags);
+		return ret;
+	case PH_ACTIVATE_REQ:
+		if (debug & DEBUG_HFCMULTI_MSG)
+			printk(KERN_DEBUG "%s: PH_ACTIVATE ch %d (0..32)\n",
+				__func__, bch->slot);
+		spin_lock_irqsave(&hc->lock, flags);
+		/* activate B-channel if not already activated */
+		if (!test_and_set_bit(FLG_ACTIVE, &bch->Flags)) {
+			hc->chan[bch->slot].txpending = 0;
+			ret = mode_hfcmulti(hc, bch->slot,
+				ch->protocol,
+				hc->chan[bch->slot].slot_tx,
+				hc->chan[bch->slot].bank_tx,
+				hc->chan[bch->slot].slot_rx,
+				hc->chan[bch->slot].bank_rx);
+			if (!ret) {
+				if (ch->protocol == ISDN_P_B_RAW && !hc->dtmf
+					&& test_bit(HFC_CHIP_DTMF, &hc->chip)) {
+					/* start decoder */
+					hc->dtmf = 1;
+					if (debug & DEBUG_HFCMULTI_DTMF)
+						printk(KERN_DEBUG
+						    "%s: start dtmf decoder\n",
+							__func__);
+					HFC_outb(hc, R_DTMF, hc->hw.r_dtmf |
+					    V_RST_DTMF);
+				}
+			}
+		} else
+			ret = 0;
+		spin_unlock_irqrestore(&hc->lock, flags);
+		if (!ret)
+			_queue_data(ch, PH_ACTIVATE_IND, MISDN_ID_ANY, 0, NULL,
+				GFP_KERNEL);
+		break;
+	case PH_CONTROL_REQ:
+		spin_lock_irqsave(&hc->lock, flags);
+		switch (hh->id) {
+		case HFC_SPL_LOOP_ON: /* set sample loop */
+			if (debug & DEBUG_HFCMULTI_MSG)
+			printk(KERN_DEBUG
+			    "%s: HFC_SPL_LOOP_ON (len = %d)\n",
+			    __func__, skb->len);
+			ret = 0;
+			break;
+		case HFC_SPL_LOOP_OFF: /* set silence */
+			if (debug & DEBUG_HFCMULTI_MSG)
+				printk(KERN_DEBUG "%s: HFC_SPL_LOOP_OFF\n",
+				    __func__);
+			ret = 0;
+			break;
+		default:
+			printk(KERN_ERR
+			     "%s: unknown PH_CONTROL_REQ info %x\n",
+			     __func__, hh->id);
+			ret = -EINVAL;
+		}
+		spin_unlock_irqrestore(&hc->lock, flags);
+		break;
+	case PH_DEACTIVATE_REQ:
+		deactivate_bchannel(bch); /* locked there */
+		_queue_data(ch, PH_DEACTIVATE_IND, MISDN_ID_ANY, 0, NULL,
+			GFP_KERNEL);
+		ret = 0;
+		break;
+	}
+	if (!ret)
+		dev_kfree_skb(skb);
+	return ret;
+}
+
+/*
+ * bchannel control function
+ */
+static int
+channel_bctrl(struct bchannel *bch, struct mISDN_ctrl_req *cq)
+{
+	int			ret = 0;
+	struct dsp_features	*features =
+		(struct dsp_features *)(*((u_long *)&cq->p1));
+	struct hfc_multi	*hc = bch->hw;
+	int			slot_tx;
+	int			bank_tx;
+	int			slot_rx;
+	int			bank_rx;
+	int			num;
+
+	switch (cq->op) {
+	case MISDN_CTRL_GETOP:
+		cq->op = MISDN_CTRL_HFC_OP | MISDN_CTRL_HW_FEATURES_OP
+			| MISDN_CTRL_RX_OFF;
+		break;
+	case MISDN_CTRL_RX_OFF: /* turn off / on rx stream */
+		hc->chan[bch->slot].rx_off = !!cq->p1;
+		if (!hc->chan[bch->slot].rx_off) {
+			/* reset fifo on rx on */
+			HFC_outb_nodebug(hc, R_FIFO, (bch->slot << 1) | 1);
+			HFC_wait_nodebug(hc);
+			HFC_outb_nodebug(hc, R_INC_RES_FIFO, V_RES_F);
+			HFC_wait_nodebug(hc);
+		}
+		if (debug & DEBUG_HFCMULTI_MSG)
+			printk(KERN_DEBUG "%s: RX_OFF request (nr=%d off=%d)\n",
+			    __func__, bch->nr, hc->chan[bch->slot].rx_off);
+		break;
+	case MISDN_CTRL_HW_FEATURES: /* fill features structure */
+		if (debug & DEBUG_HFCMULTI_MSG)
+			printk(KERN_DEBUG "%s: HW_FEATURE request\n",
+			    __func__);
+		/* create confirm */
+		features->hfc_id = hc->id;
+		if (test_bit(HFC_CHIP_DTMF, &hc->chip))
+			features->hfc_dtmf = 1;
+		features->hfc_loops = 0;
+		if (test_bit(HFC_CHIP_B410P, &hc->chip)) {
+			features->hfc_echocanhw = 1;
+		} else {
+			features->pcm_id = hc->pcm;
+			features->pcm_slots = hc->slots;
+			features->pcm_banks = 2;
+		}
+		break;
+	case MISDN_CTRL_HFC_PCM_CONN: /* connect to pcm timeslot (0..N) */
+		slot_tx = cq->p1 & 0xff;
+		bank_tx = cq->p1 >> 8;
+		slot_rx = cq->p2 & 0xff;
+		bank_rx = cq->p2 >> 8;
+		if (debug & DEBUG_HFCMULTI_MSG)
+			printk(KERN_DEBUG
+			    "%s: HFC_PCM_CONN slot %d bank %d (TX) "
+			    "slot %d bank %d (RX)\n",
+			    __func__, slot_tx, bank_tx,
+			    slot_rx, bank_rx);
+		if (slot_tx < hc->slots && bank_tx <= 2 &&
+		    slot_rx < hc->slots && bank_rx <= 2)
+			hfcmulti_pcm(hc, bch->slot,
+			    slot_tx, bank_tx, slot_rx, bank_rx);
+		else {
+			printk(KERN_WARNING
+			    "%s: HFC_PCM_CONN slot %d bank %d (TX) "
+			    "slot %d bank %d (RX) out of range\n",
+			    __func__, slot_tx, bank_tx,
+			    slot_rx, bank_rx);
+			ret = -EINVAL;
+		}
+		break;
+	case MISDN_CTRL_HFC_PCM_DISC: /* release interface from pcm timeslot */
+		if (debug & DEBUG_HFCMULTI_MSG)
+			printk(KERN_DEBUG "%s: HFC_PCM_DISC\n",
+			    __func__);
+		hfcmulti_pcm(hc, bch->slot, -1, 0, -1, 0);
+		break;
+	case MISDN_CTRL_HFC_CONF_JOIN: /* join conference (0..7) */
+		num = cq->p1 & 0xff;
+		if (debug & DEBUG_HFCMULTI_MSG)
+			printk(KERN_DEBUG "%s: HFC_CONF_JOIN conf %d\n",
+			    __func__, num);
+		if (num <= 7)
+			hfcmulti_conf(hc, bch->slot, num);
+		else {
+			printk(KERN_WARNING
+			    "%s: HW_CONF_JOIN conf %d out of range\n",
+			    __func__, num);
+			ret = -EINVAL;
+		}
+		break;
+	case MISDN_CTRL_HFC_CONF_SPLIT: /* split conference */
+		if (debug & DEBUG_HFCMULTI_MSG)
+			printk(KERN_DEBUG "%s: HFC_CONF_SPLIT\n", __func__);
+		hfcmulti_conf(hc, bch->slot, -1);
+		break;
+	case MISDN_CTRL_HFC_ECHOCAN_ON:
+		if (debug & DEBUG_HFCMULTI_MSG)
+			printk(KERN_DEBUG "%s: HFC_ECHOCAN_ON\n", __func__);
+		if (test_bit(HFC_CHIP_B410P, &hc->chip))
+			vpm_echocan_on(hc, bch->slot, cq->p1);
+		else
+			ret = -EINVAL;
+		break;
+
+	case MISDN_CTRL_HFC_ECHOCAN_OFF:
+		if (debug & DEBUG_HFCMULTI_MSG)
+			printk(KERN_DEBUG "%s: HFC_ECHOCAN_OFF\n",
+				__func__);
+		if (test_bit(HFC_CHIP_B410P, &hc->chip))
+			vpm_echocan_off(hc, bch->slot);
+		else
+			ret = -EINVAL;
+		break;
+	default:
+		printk(KERN_WARNING "%s: unknown Op %x\n",
+		    __func__, cq->op);
+		ret = -EINVAL;
+		break;
+	}
+	return ret;
+}
+
+static int
+hfcm_bctrl(struct mISDNchannel *ch, u_int cmd, void *arg)
+{
+	struct bchannel		*bch = container_of(ch, struct bchannel, ch);
+	struct hfc_multi	*hc = bch->hw;
+	int			err = -EINVAL;
+	u_long	flags;
+
+	if (bch->debug & DEBUG_HW)
+		printk(KERN_DEBUG "%s: cmd:%x %p\n",
+		    __func__, cmd, arg);
+	switch (cmd) {
+	case CLOSE_CHANNEL:
+		test_and_clear_bit(FLG_OPEN, &bch->Flags);
+		if (test_bit(FLG_ACTIVE, &bch->Flags))
+			deactivate_bchannel(bch); /* locked there */
+		ch->protocol = ISDN_P_NONE;
+		ch->peer = NULL;
+		module_put(THIS_MODULE);
+		err = 0;
+		break;
+	case CONTROL_CHANNEL:
+		spin_lock_irqsave(&hc->lock, flags);
+		err = channel_bctrl(bch, arg);
+		spin_unlock_irqrestore(&hc->lock, flags);
+		break;
+	default:
+		printk(KERN_WARNING "%s: unknown prim(%x)\n",
+			__func__, cmd);
+	}
+	return err;
+}
+
+/*
+ * handle D-channel events
+ *
+ * handle state change event
+ */
+static void
+ph_state_change(struct dchannel *dch)
+{
+	struct hfc_multi *hc = dch->hw;
+	int ch, i;
+
+	if (!dch) {
+		printk(KERN_WARNING "%s: ERROR given dch is NULL\n",
+		    __func__);
+		return;
+	}
+	ch = dch->slot;
+
+	if (hc->type == 1) {
+		if (dch->dev.D.protocol == ISDN_P_TE_E1) {
+			if (debug & DEBUG_HFCMULTI_STATE)
+				printk(KERN_DEBUG
+				    "%s: E1 TE (id=%d) newstate %x\n",
+				    __func__, hc->id, dch->state);
+		} else {
+			if (debug & DEBUG_HFCMULTI_STATE)
+				printk(KERN_DEBUG
+				    "%s: E1 NT (id=%d) newstate %x\n",
+				    __func__, hc->id, dch->state);
+		}
+		switch (dch->state) {
+		case (1):
+			if (hc->e1_state != 1) {
+			    for (i = 1; i <= 31; i++) {
+				/* reset fifos on e1 activation */
+				HFC_outb_nodebug(hc, R_FIFO, (i << 1) | 1);
+				HFC_wait_nodebug(hc);
+				HFC_outb_nodebug(hc,
+					R_INC_RES_FIFO, V_RES_F);
+				HFC_wait_nodebug(hc);
+			    }
+			}
+			test_and_set_bit(FLG_ACTIVE, &dch->Flags);
+			_queue_data(&dch->dev.D, PH_ACTIVATE_IND,
+			    MISDN_ID_ANY, 0, NULL, GFP_ATOMIC);
+			break;
+
+		default:
+			if (hc->e1_state != 1)
+				return;
+			test_and_clear_bit(FLG_ACTIVE, &dch->Flags);
+			_queue_data(&dch->dev.D, PH_DEACTIVATE_IND,
+			    MISDN_ID_ANY, 0, NULL, GFP_ATOMIC);
+		}
+		hc->e1_state = dch->state;
+	} else {
+		if (dch->dev.D.protocol == ISDN_P_TE_S0) {
+			if (debug & DEBUG_HFCMULTI_STATE)
+				printk(KERN_DEBUG
+				    "%s: S/T TE newstate %x\n",
+				    __func__, dch->state);
+			switch (dch->state) {
+			case (0):
+				l1_event(dch->l1, HW_RESET_IND);
+				break;
+			case (3):
+				l1_event(dch->l1, HW_DEACT_IND);
+				break;
+			case (5):
+			case (8):
+				l1_event(dch->l1, ANYSIGNAL);
+				break;
+			case (6):
+				l1_event(dch->l1, INFO2);
+				break;
+			case (7):
+				l1_event(dch->l1, INFO4_P8);
+				break;
+			}
+		} else {
+			if (debug & DEBUG_HFCMULTI_STATE)
+				printk(KERN_DEBUG "%s: S/T NT newstate %x\n",
+				    __func__, dch->state);
+			switch (dch->state) {
+			case (2):
+				if (hc->chan[ch].nt_timer == 0) {
+					hc->chan[ch].nt_timer = -1;
+					HFC_outb(hc, R_ST_SEL,
+					    hc->chan[ch].port);
+					/* undocumented: delay after R_ST_SEL */
+					udelay(1);
+					HFC_outb(hc, A_ST_WR_STATE, 4 |
+					    V_ST_LD_STA); /* G4 */
+					udelay(6); /* wait at least 5,21us */
+					HFC_outb(hc, A_ST_WR_STATE, 4);
+					dch->state = 4;
+				} else {
+					/* one extra count for the next event */
+					hc->chan[ch].nt_timer =
+					    nt_t1_count[poll_timer] + 1;
+					HFC_outb(hc, R_ST_SEL,
+					    hc->chan[ch].port);
+					/* undocumented: delay after R_ST_SEL */
+					udelay(1);
+					/* allow G2 -> G3 transition */
+					HFC_outb(hc, A_ST_WR_STATE, 2 |
+					    V_SET_G2_G3);
+				}
+				break;
+			case (1):
+				hc->chan[ch].nt_timer = -1;
+				test_and_clear_bit(FLG_ACTIVE, &dch->Flags);
+				_queue_data(&dch->dev.D, PH_DEACTIVATE_IND,
+				    MISDN_ID_ANY, 0, NULL, GFP_ATOMIC);
+				break;
+			case (4):
+				hc->chan[ch].nt_timer = -1;
+				break;
+			case (3):
+				hc->chan[ch].nt_timer = -1;
+				test_and_set_bit(FLG_ACTIVE, &dch->Flags);
+				_queue_data(&dch->dev.D, PH_ACTIVATE_IND,
+				    MISDN_ID_ANY, 0, NULL, GFP_ATOMIC);
+				break;
+			}
+		}
+	}
+}
+
+/*
+ * called for card mode init message
+ */
+
+static void
+hfcmulti_initmode(struct dchannel *dch)
+{
+	struct hfc_multi *hc = dch->hw;
+	u_char		a_st_wr_state, r_e1_wr_sta;
+	int		i, pt;
+
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_DEBUG "%s: entered\n", __func__);
+
+	if (hc->type == 1) {
+		hc->chan[hc->dslot].slot_tx = -1;
+		hc->chan[hc->dslot].slot_rx = -1;
+		hc->chan[hc->dslot].conf = -1;
+		if (hc->dslot) {
+			mode_hfcmulti(hc, hc->dslot, dch->dev.D.protocol,
+				-1, 0, -1, 0);
+			dch->timer.function = (void *) hfcmulti_dbusy_timer;
+			dch->timer.data = (long) dch;
+			init_timer(&dch->timer);
+		}
+		for (i = 1; i <= 31; i++) {
+			if (i == hc->dslot)
+				continue;
+			hc->chan[i].slot_tx = -1;
+			hc->chan[i].slot_rx = -1;
+			hc->chan[i].conf = -1;
+			mode_hfcmulti(hc, i, ISDN_P_NONE, -1, 0, -1, 0);
+		}
+		/* E1 */
+		if (test_bit(HFC_CFG_REPORT_LOS, &hc->chan[hc->dslot].cfg)) {
+			HFC_outb(hc, R_LOS0, 255); /* 2 ms */
+			HFC_outb(hc, R_LOS1, 255); /* 512 ms */
+		}
+		if (test_bit(HFC_CFG_OPTICAL, &hc->chan[hc->dslot].cfg)) {
+			HFC_outb(hc, R_RX0, 0);
+			hc->hw.r_tx0 = 0 | V_OUT_EN;
+		} else {
+			HFC_outb(hc, R_RX0, 1);
+			hc->hw.r_tx0 = 1 | V_OUT_EN;
+		}
+		hc->hw.r_tx1 = V_ATX | V_NTRI;
+		HFC_outb(hc, R_TX0, hc->hw.r_tx0);
+		HFC_outb(hc, R_TX1, hc->hw.r_tx1);
+		HFC_outb(hc, R_TX_FR0, 0x00);
+		HFC_outb(hc, R_TX_FR1, 0xf8);
+
+		if (test_bit(HFC_CFG_CRC4, &hc->chan[hc->dslot].cfg))
+			HFC_outb(hc, R_TX_FR2, V_TX_MF | V_TX_E | V_NEG_E);
+
+		HFC_outb(hc, R_RX_FR0, V_AUTO_RESYNC | V_AUTO_RECO | 0);
+
+		if (test_bit(HFC_CFG_CRC4, &hc->chan[hc->dslot].cfg))
+			HFC_outb(hc, R_RX_FR1, V_RX_MF | V_RX_MF_SYNC);
+
+		if (dch->dev.D.protocol == ISDN_P_NT_E1) {
+			if (debug & DEBUG_HFCMULTI_INIT)
+				printk(KERN_DEBUG "%s: E1 port is NT-mode\n",
+				    __func__);
+			r_e1_wr_sta = 0; /* G0 */
+			hc->e1_getclock = 0;
+		} else {
+			if (debug & DEBUG_HFCMULTI_INIT)
+				printk(KERN_DEBUG "%s: E1 port is TE-mode\n",
+				    __func__);
+			r_e1_wr_sta = 0; /* F0 */
+			hc->e1_getclock = 1;
+		}
+		if (test_bit(HFC_CHIP_RX_SYNC, &hc->chip))
+			HFC_outb(hc, R_SYNC_OUT, V_SYNC_E1_RX);
+		else
+			HFC_outb(hc, R_SYNC_OUT, 0);
+		if (test_bit(HFC_CHIP_E1CLOCK_GET, &hc->chip))
+			hc->e1_getclock = 1;
+		if (test_bit(HFC_CHIP_E1CLOCK_PUT, &hc->chip))
+			hc->e1_getclock = 0;
+		if (test_bit(HFC_CHIP_PCM_SLAVE, &hc->chip)) {
+			/* SLAVE (clock master) */
+			if (debug & DEBUG_HFCMULTI_INIT)
+				printk(KERN_DEBUG
+				    "%s: E1 port is clock master "
+				    "(clock from PCM)\n", __func__);
+			HFC_outb(hc, R_SYNC_CTRL, V_EXT_CLK_SYNC | V_PCM_SYNC);
+		} else {
+			if (hc->e1_getclock) {
+				/* MASTER (clock slave) */
+				if (debug & DEBUG_HFCMULTI_INIT)
+					printk(KERN_DEBUG
+					    "%s: E1 port is clock slave "
+					    "(clock to PCM)\n", __func__);
+				HFC_outb(hc, R_SYNC_CTRL, V_SYNC_OFFS);
+			} else {
+				/* MASTER (clock master) */
+				if (debug & DEBUG_HFCMULTI_INIT)
+					printk(KERN_DEBUG "%s: E1 port is "
+					    "clock master "
+					    "(clock from QUARTZ)\n",
+					    __func__);
+				HFC_outb(hc, R_SYNC_CTRL, V_EXT_CLK_SYNC |
+				    V_PCM_SYNC | V_JATT_OFF);
+				HFC_outb(hc, R_SYNC_OUT, 0);
+			}
+		}
+		HFC_outb(hc, R_JATT_ATT, 0x9c); /* undoc register */
+		HFC_outb(hc, R_PWM_MD, V_PWM0_MD);
+		HFC_outb(hc, R_PWM0, 0x50);
+		HFC_outb(hc, R_PWM1, 0xff);
+		/* state machine setup */
+		HFC_outb(hc, R_E1_WR_STA, r_e1_wr_sta | V_E1_LD_STA);
+		udelay(6); /* wait at least 5,21us */
+		HFC_outb(hc, R_E1_WR_STA, r_e1_wr_sta);
+		if (test_bit(HFC_CHIP_PLXSD, &hc->chip)) {
+			hc->syncronized = 0;
+			plxsd_checksync(hc, 0);
+		}
+	} else {
+		i = dch->slot;
+		hc->chan[i].slot_tx = -1;
+		hc->chan[i].slot_rx = -1;
+		hc->chan[i].conf = -1;
+		mode_hfcmulti(hc, i, dch->dev.D.protocol, -1, 0, -1, 0);
+		dch->timer.function = (void *)hfcmulti_dbusy_timer;
+		dch->timer.data = (long) dch;
+		init_timer(&dch->timer);
+		hc->chan[i - 2].slot_tx = -1;
+		hc->chan[i - 2].slot_rx = -1;
+		hc->chan[i - 2].conf = -1;
+		mode_hfcmulti(hc, i - 2, ISDN_P_NONE, -1, 0, -1, 0);
+		hc->chan[i - 1].slot_tx = -1;
+		hc->chan[i - 1].slot_rx = -1;
+		hc->chan[i - 1].conf = -1;
+		mode_hfcmulti(hc, i - 1, ISDN_P_NONE, -1, 0, -1, 0);
+		/* ST */
+		pt = hc->chan[i].port;
+		/* select interface */
+		HFC_outb(hc, R_ST_SEL, pt);
+		/* undocumented: delay after R_ST_SEL */
+		udelay(1);
+		if (dch->dev.D.protocol == ISDN_P_NT_S0) {
+			if (debug & DEBUG_HFCMULTI_INIT)
+				printk(KERN_DEBUG
+				    "%s: ST port %d is NT-mode\n",
+				    __func__, pt);
+			/* clock delay */
+			HFC_outb(hc, A_ST_CLK_DLY, clockdelay_nt);
+			a_st_wr_state = 1; /* G1 */
+			hc->hw.a_st_ctrl0[pt] = V_ST_MD;
+		} else {
+			if (debug & DEBUG_HFCMULTI_INIT)
+				printk(KERN_DEBUG
+				    "%s: ST port %d is TE-mode\n",
+				    __func__, pt);
+			/* clock delay */
+			HFC_outb(hc, A_ST_CLK_DLY, clockdelay_te);
+			a_st_wr_state = 2; /* F2 */
+			hc->hw.a_st_ctrl0[pt] = 0;
+		}
+		if (!test_bit(HFC_CFG_NONCAP_TX, &hc->chan[i].cfg))
+			hc->hw.a_st_ctrl0[pt] |= V_TX_LI;
+		/* line setup */
+		HFC_outb(hc, A_ST_CTRL0,  hc->hw.a_st_ctrl0[pt]);
+		/* disable E-channel */
+		if ((dch->dev.D.protocol == ISDN_P_NT_S0) ||
+		    test_bit(HFC_CFG_DIS_ECHANNEL, &hc->chan[i].cfg))
+			HFC_outb(hc, A_ST_CTRL1, V_E_IGNO);
+		else
+			HFC_outb(hc, A_ST_CTRL1, 0);
+		/* enable B-channel receive */
+		HFC_outb(hc, A_ST_CTRL2,  V_B1_RX_EN | V_B2_RX_EN);
+		/* state machine setup */
+		HFC_outb(hc, A_ST_WR_STATE, a_st_wr_state | V_ST_LD_STA);
+		udelay(6); /* wait at least 5,21us */
+		HFC_outb(hc, A_ST_WR_STATE, a_st_wr_state);
+		hc->hw.r_sci_msk |= 1 << pt;
+		/* state machine interrupts */
+		HFC_outb(hc, R_SCI_MSK, hc->hw.r_sci_msk);
+		/* unset sync on port */
+		if (test_bit(HFC_CHIP_PLXSD, &hc->chip)) {
+			hc->syncronized &=
+			   ~(1 << hc->chan[dch->slot].port);
+			plxsd_checksync(hc, 0);
+		}
+	}
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk("%s: done\n", __func__);
+}
+
+
+static int
+open_dchannel(struct hfc_multi *hc, struct dchannel *dch,
+    struct channel_req *rq)
+{
+	int	err = 0;
+	u_long	flags;
+
+	if (debug & DEBUG_HW_OPEN)
+		printk(KERN_DEBUG "%s: dev(%d) open from %p\n", __func__,
+		    dch->dev.id, __builtin_return_address(0));
+	if (rq->protocol == ISDN_P_NONE)
+		return -EINVAL;
+	if ((dch->dev.D.protocol != ISDN_P_NONE) &&
+	    (dch->dev.D.protocol != rq->protocol)) {
+	    if (debug & DEBUG_HFCMULTI_MODE)
+		printk(KERN_WARNING "%s: change protocol %x to %x\n",
+		    __func__, dch->dev.D.protocol, rq->protocol);
+	}
+	if ((dch->dev.D.protocol == ISDN_P_TE_S0)
+	 && (rq->protocol != ISDN_P_TE_S0))
+		l1_event(dch->l1, CLOSE_CHANNEL);
+	if (dch->dev.D.protocol != rq->protocol) {
+		if (rq->protocol == ISDN_P_TE_S0) {
+			err = create_l1(dch, hfcm_l1callback);
+			if (err)
+				return err;
+		}
+		dch->dev.D.protocol = rq->protocol;
+		spin_lock_irqsave(&hc->lock, flags);
+		hfcmulti_initmode(dch);
+		spin_unlock_irqrestore(&hc->lock, flags);
+	}
+
+	if (((rq->protocol == ISDN_P_NT_S0) && (dch->state == 3)) ||
+	    ((rq->protocol == ISDN_P_TE_S0) && (dch->state == 7)) ||
+	    ((rq->protocol == ISDN_P_NT_E1) && (dch->state == 1)) ||
+	    ((rq->protocol == ISDN_P_TE_E1) && (dch->state == 1))) {
+		_queue_data(&dch->dev.D, PH_ACTIVATE_IND, MISDN_ID_ANY,
+		    0, NULL, GFP_KERNEL);
+	}
+	rq->ch = &dch->dev.D;
+	if (!try_module_get(THIS_MODULE))
+		printk(KERN_WARNING "%s:cannot get module\n", __func__);
+	return 0;
+}
+
+static int
+open_bchannel(struct hfc_multi *hc, struct dchannel *dch,
+    struct channel_req *rq)
+{
+	struct bchannel	*bch;
+	int		ch;
+
+	if (!test_bit(rq->adr.channel, &dch->dev.channelmap[0]))
+		return -EINVAL;
+	if (rq->protocol == ISDN_P_NONE)
+		return -EINVAL;
+	if (hc->type == 1)
+		ch = rq->adr.channel;
+	else
+		ch = (rq->adr.channel - 1) + (dch->slot - 2);
+	bch = hc->chan[ch].bch;
+	if (!bch) {
+		printk(KERN_ERR "%s:internal error ch %d has no bch\n",
+		    __func__, ch);
+		return -EINVAL;
+	}
+	if (test_and_set_bit(FLG_OPEN, &bch->Flags))
+		return -EBUSY; /* b-channel can be only open once */
+	bch->ch.protocol = rq->protocol;
+	hc->chan[ch].rx_off = 0;
+	rq->ch = &bch->ch;
+	if (!try_module_get(THIS_MODULE))
+		printk(KERN_WARNING "%s:cannot get module\n", __func__);
+	return 0;
+}
+
+/*
+ * device control function
+ */
+static int
+channel_dctrl(struct dchannel *dch, struct mISDN_ctrl_req *cq)
+{
+	int	ret = 0;
+
+	switch (cq->op) {
+	case MISDN_CTRL_GETOP:
+		cq->op = 0;
+		break;
+	default:
+		printk(KERN_WARNING "%s: unknown Op %x\n",
+		    __func__, cq->op);
+		ret = -EINVAL;
+		break;
+	}
+	return ret;
+}
+
+static int
+hfcm_dctrl(struct mISDNchannel *ch, u_int cmd, void *arg)
+{
+	struct mISDNdevice	*dev = container_of(ch, struct mISDNdevice, D);
+	struct dchannel		*dch = container_of(dev, struct dchannel, dev);
+	struct hfc_multi	*hc = dch->hw;
+	struct channel_req	*rq;
+	int			err = 0;
+	u_long			flags;
+
+	if (dch->debug & DEBUG_HW)
+		printk(KERN_DEBUG "%s: cmd:%x %p\n",
+		    __func__, cmd, arg);
+	switch (cmd) {
+	case OPEN_CHANNEL:
+		rq = arg;
+		switch (rq->protocol) {
+		case ISDN_P_TE_S0:
+		case ISDN_P_NT_S0:
+			if (hc->type == 1) {
+				err = -EINVAL;
+				break;
+			}
+			err = open_dchannel(hc, dch, rq); /* locked there */
+			break;
+		case ISDN_P_TE_E1:
+		case ISDN_P_NT_E1:
+			if (hc->type != 1) {
+				err = -EINVAL;
+				break;
+			}
+			err = open_dchannel(hc, dch, rq); /* locked there */
+			break;
+		default:
+			spin_lock_irqsave(&hc->lock, flags);
+			err = open_bchannel(hc, dch, rq);
+			spin_unlock_irqrestore(&hc->lock, flags);
+		}
+		break;
+	case CLOSE_CHANNEL:
+		if (debug & DEBUG_HW_OPEN)
+			printk(KERN_DEBUG "%s: dev(%d) close from %p\n",
+			    __func__, dch->dev.id,
+			    __builtin_return_address(0));
+		module_put(THIS_MODULE);
+		break;
+	case CONTROL_CHANNEL:
+		spin_lock_irqsave(&hc->lock, flags);
+		err = channel_dctrl(dch, arg);
+		spin_unlock_irqrestore(&hc->lock, flags);
+		break;
+	default:
+		if (dch->debug & DEBUG_HW)
+			printk(KERN_DEBUG "%s: unknown command %x\n",
+			    __func__, cmd);
+		err = -EINVAL;
+	}
+	return err;
+}
+
+/*
+ * initialize the card
+ */
+
+/*
+ * start timer irq, wait some time and check if we have interrupts.
+ * if not, reset chip and try again.
+ */
+static int
+init_card(struct hfc_multi *hc)
+{
+	int	err = -EIO;
+	u_long	flags;
+	u_short	*plx_acc;
+	u_long	plx_flags;
+
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_DEBUG "%s: entered\n", __func__);
+
+	spin_lock_irqsave(&hc->lock, flags);
+	/* set interrupts but leave global interrupt disabled */
+	hc->hw.r_irq_ctrl = V_FIFO_IRQ;
+	disable_hwirq(hc);
+	spin_unlock_irqrestore(&hc->lock, flags);
+
+	if (request_irq(hc->pci_dev->irq, hfcmulti_interrupt, IRQF_SHARED,
+	    "HFC-multi", hc)) {
+		printk(KERN_WARNING "mISDN: Could not get interrupt %d.\n",
+		    hc->pci_dev->irq);
+		return -EIO;
+	}
+	hc->irq = hc->pci_dev->irq;
+
+	if (test_bit(HFC_CHIP_PLXSD, &hc->chip)) {
+		spin_lock_irqsave(&plx_lock, plx_flags);
+		plx_acc = (u_short *)(hc->plx_membase+PLX_INTCSR);
+		writew((PLX_INTCSR_PCIINT_ENABLE | PLX_INTCSR_LINTI1_ENABLE),
+			plx_acc); /* enable PCI & LINT1 irq */
+		spin_unlock_irqrestore(&plx_lock, plx_flags);
+	}
+
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_DEBUG "%s: IRQ %d count %d\n",
+		    __func__, hc->irq, hc->irqcnt);
+	err = init_chip(hc);
+	if (err)
+		goto error;
+	/*
+	 * Finally enable IRQ output
+	 * this is only allowed, if an IRQ routine is allready
+	 * established for this HFC, so don't do that earlier
+	 */
+	spin_lock_irqsave(&hc->lock, flags);
+	enable_hwirq(hc);
+	spin_unlock_irqrestore(&hc->lock, flags);
+	/* printk(KERN_DEBUG "no master irq set!!!\n"); */
+	set_current_state(TASK_UNINTERRUPTIBLE);
+	schedule_timeout((100*HZ)/1000); /* Timeout 100ms */
+	/* turn IRQ off until chip is completely initialized */
+	spin_lock_irqsave(&hc->lock, flags);
+	disable_hwirq(hc);
+	spin_unlock_irqrestore(&hc->lock, flags);
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_DEBUG "%s: IRQ %d count %d\n",
+		    __func__, hc->irq, hc->irqcnt);
+	if (hc->irqcnt) {
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG "%s: done\n", __func__);
+
+		return 0;
+	}
+	if (test_bit(HFC_CHIP_PCM_SLAVE, &hc->chip)) {
+		printk(KERN_INFO "ignoring missing interrupts\n");
+		return 0;
+	}
+
+	printk(KERN_ERR "HFC PCI: IRQ(%d) getting no interrupts during init.\n",
+		hc->irq);
+
+	err = -EIO;
+
+error:
+	if (test_bit(HFC_CHIP_PLXSD, &hc->chip)) {
+		spin_lock_irqsave(&plx_lock, plx_flags);
+		plx_acc = (u_short *)(hc->plx_membase+PLX_INTCSR);
+		writew(0x00, plx_acc); /*disable IRQs*/
+		spin_unlock_irqrestore(&plx_lock, plx_flags);
+	}
+
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_WARNING "%s: free irq %d\n", __func__, hc->irq);
+	if (hc->irq) {
+		free_irq(hc->irq, hc);
+		hc->irq = 0;
+	}
+
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_DEBUG "%s: done (err=%d)\n", __func__, err);
+	return err;
+}
+
+/*
+ * find pci device and set it up
+ */
+
+static int
+setup_pci(struct hfc_multi *hc, struct pci_dev *pdev,
+		const struct pci_device_id *ent)
+{
+	struct hm_map	*m = (struct hm_map *)ent->driver_data;
+
+	printk(KERN_INFO
+	    "HFC-multi: card manufacturer: '%s' card name: '%s' clock: %s\n",
+	    m->vendor_name, m->card_name, m->clock2 ? "double" : "normal");
+
+	hc->pci_dev = pdev;
+	if (m->clock2)
+		test_and_set_bit(HFC_CHIP_CLOCK2, &hc->chip);
+
+	if (ent->device == 0xB410) {
+		test_and_set_bit(HFC_CHIP_B410P, &hc->chip);
+		test_and_set_bit(HFC_CHIP_PCM_MASTER, &hc->chip);
+		test_and_clear_bit(HFC_CHIP_PCM_SLAVE, &hc->chip);
+		hc->slots = 32;
+	}
+
+	if (hc->pci_dev->irq <= 0) {
+		printk(KERN_WARNING "HFC-multi: No IRQ for PCI card found.\n");
+		return -EIO;
+	}
+	if (pci_enable_device(hc->pci_dev)) {
+		printk(KERN_WARNING "HFC-multi: Error enabling PCI card.\n");
+		return -EIO;
+	}
+	hc->leds = m->leds;
+	hc->ledstate = 0xAFFEAFFE;
+	hc->opticalsupport = m->opticalsupport;
+
+	/* set memory access methods */
+	if (m->io_mode) /* use mode from card config */
+		hc->io_mode = m->io_mode;
+	switch (hc->io_mode) {
+	case HFC_IO_MODE_PLXSD:
+		test_and_set_bit(HFC_CHIP_PLXSD, &hc->chip);
+		hc->slots = 128; /* required */
+		/* fall through */
+	case HFC_IO_MODE_PCIMEM:
+		hc->HFC_outb = HFC_outb_pcimem;
+		hc->HFC_inb = HFC_inb_pcimem;
+		hc->HFC_inw = HFC_inw_pcimem;
+		hc->HFC_wait = HFC_wait_pcimem;
+		hc->read_fifo = read_fifo_pcimem;
+		hc->write_fifo = write_fifo_pcimem;
+		break;
+	case HFC_IO_MODE_REGIO:
+		hc->HFC_outb = HFC_outb_regio;
+		hc->HFC_inb = HFC_inb_regio;
+		hc->HFC_inw = HFC_inw_regio;
+		hc->HFC_wait = HFC_wait_regio;
+		hc->read_fifo = read_fifo_regio;
+		hc->write_fifo = write_fifo_regio;
+		break;
+	default:
+		printk(KERN_WARNING "HFC-multi: Invalid IO mode.\n");
+		pci_disable_device(hc->pci_dev);
+		return -EIO;
+	}
+	hc->HFC_outb_nodebug = hc->HFC_outb;
+	hc->HFC_inb_nodebug = hc->HFC_inb;
+	hc->HFC_inw_nodebug = hc->HFC_inw;
+	hc->HFC_wait_nodebug = hc->HFC_wait;
+#ifdef HFC_REGISTER_DEBUG
+	hc->HFC_outb = HFC_outb_debug;
+	hc->HFC_inb = HFC_inb_debug;
+	hc->HFC_inw = HFC_inw_debug;
+	hc->HFC_wait = HFC_wait_debug;
+#endif
+	hc->pci_iobase = 0;
+	hc->pci_membase = NULL;
+	hc->plx_membase = NULL;
+
+	switch (hc->io_mode) {
+	case HFC_IO_MODE_PLXSD:
+		hc->plx_origmembase =  hc->pci_dev->resource[0].start;
+		/* MEMBASE 1 is PLX PCI Bridge */
+
+		if (!hc->plx_origmembase) {
+			printk(KERN_WARNING
+			  "HFC-multi: No IO-Memory for PCI PLX bridge found\n");
+			pci_disable_device(hc->pci_dev);
+			return -EIO;
+		}
+
+		hc->plx_membase = ioremap(hc->plx_origmembase, 0x80);
+		if (!hc->plx_membase) {
+			printk(KERN_WARNING
+			    "HFC-multi: failed to remap plx address space. "
+			    "(internal error)\n");
+			pci_disable_device(hc->pci_dev);
+			return -EIO;
+		}
+		printk(KERN_INFO
+		    "HFC-multi: plx_membase:%#lx plx_origmembase:%#lx\n",
+		    (u_long)hc->plx_membase, hc->plx_origmembase);
+
+		hc->pci_origmembase =  hc->pci_dev->resource[2].start;
+		    /* MEMBASE 1 is PLX PCI Bridge */
+		if (!hc->pci_origmembase) {
+			printk(KERN_WARNING
+			    "HFC-multi: No IO-Memory for PCI card found\n");
+			pci_disable_device(hc->pci_dev);
+			return -EIO;
+		}
+
+		hc->pci_membase = ioremap(hc->pci_origmembase, 0x400);
+		if (!hc->pci_membase) {
+			printk(KERN_WARNING "HFC-multi: failed to remap io "
+			    "address space. (internal error)\n");
+			pci_disable_device(hc->pci_dev);
+			return -EIO;
+		}
+
+		printk(KERN_INFO
+		    "card %d: defined at MEMBASE %#lx (%#lx) IRQ %d HZ %d "
+		    "leds-type %d\n",
+		    hc->id, (u_long)hc->pci_membase, hc->pci_origmembase,
+		    hc->pci_dev->irq, HZ, hc->leds);
+		pci_write_config_word(hc->pci_dev, PCI_COMMAND, PCI_ENA_MEMIO);
+		break;
+	case HFC_IO_MODE_PCIMEM:
+		hc->pci_origmembase = hc->pci_dev->resource[1].start;
+		if (!hc->pci_origmembase) {
+			printk(KERN_WARNING
+			    "HFC-multi: No IO-Memory for PCI card found\n");
+			pci_disable_device(hc->pci_dev);
+			return -EIO;
+		}
+
+		hc->pci_membase = ioremap(hc->pci_origmembase, 256);
+		if (!hc->pci_membase) {
+			printk(KERN_WARNING
+			    "HFC-multi: failed to remap io address space. "
+			    "(internal error)\n");
+			pci_disable_device(hc->pci_dev);
+			return -EIO;
+		}
+		printk(KERN_INFO "card %d: defined at MEMBASE %#lx (%#lx) IRQ %d "
+		    "HZ %d leds-type %d\n", hc->id, (u_long)hc->pci_membase,
+		    hc->pci_origmembase, hc->pci_dev->irq, HZ, hc->leds);
+		pci_write_config_word(hc->pci_dev, PCI_COMMAND, PCI_ENA_MEMIO);
+		break;
+	case HFC_IO_MODE_REGIO:
+		hc->pci_iobase = (u_int) hc->pci_dev->resource[0].start;
+		if (!hc->pci_iobase) {
+			printk(KERN_WARNING
+				"HFC-multi: No IO for PCI card found\n");
+			pci_disable_device(hc->pci_dev);
+			return -EIO;
+		}
+
+		if (!request_region(hc->pci_iobase, 8, "hfcmulti")) {
+			printk(KERN_WARNING "HFC-multi: failed to request "
+			    "address space at 0x%08lx (internal error)\n",
+			    hc->pci_iobase);
+			pci_disable_device(hc->pci_dev);
+			return -EIO;
+		}
+
+		printk(KERN_INFO
+		    "%s %s: defined at IOBASE %#x IRQ %d HZ %d leds-type %d\n",
+		    m->vendor_name, m->card_name, (u_int) hc->pci_iobase,
+		    hc->pci_dev->irq, HZ, hc->leds);
+		pci_write_config_word(hc->pci_dev, PCI_COMMAND, PCI_ENA_REGIO);
+		break;
+	default:
+		printk(KERN_WARNING "HFC-multi: Invalid IO mode.\n");
+		pci_disable_device(hc->pci_dev);
+		return -EIO;
+	}
+
+	pci_set_drvdata(hc->pci_dev, hc);
+
+	/* At this point the needed PCI config is done */
+	/* fifos are still not enabled */
+	return 0;
+}
+
+
+/*
+ * remove port
+ */
+
+static void
+release_port(struct hfc_multi *hc, struct dchannel *dch)
+{
+	int	pt, ci, i = 0;
+	u_long	flags;
+	struct bchannel *pb;
+
+	ci = dch->slot;
+	pt = hc->chan[ci].port;
+
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_DEBUG "%s: entered for port %d\n",
+			__func__, pt + 1);
+
+	if (pt >= hc->ports) {
+		printk(KERN_WARNING "%s: ERROR port out of range (%d).\n",
+		     __func__, pt + 1);
+		return;
+	}
+
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_DEBUG "%s: releasing port=%d\n",
+		    __func__, pt + 1);
+
+	if (dch->dev.D.protocol == ISDN_P_TE_S0)
+		l1_event(dch->l1, CLOSE_CHANNEL);
+
+	hc->chan[ci].dch = NULL;
+
+	if (hc->created[pt]) {
+		hc->created[pt] = 0;
+		mISDN_unregister_device(&dch->dev);
+	}
+
+	spin_lock_irqsave(&hc->lock, flags);
+
+	if (dch->timer.function) {
+		del_timer(&dch->timer);
+		dch->timer.function = NULL;
+	}
+
+	if (hc->type == 1) { /* E1 */
+		/* remove sync */
+		if (test_bit(HFC_CHIP_PLXSD, &hc->chip)) {
+			hc->syncronized = 0;
+			plxsd_checksync(hc, 1);
+		}
+		/* free channels */
+		for (i = 0; i <= 31; i++) {
+			if (hc->chan[i].bch) {
+				if (debug & DEBUG_HFCMULTI_INIT)
+					printk(KERN_DEBUG
+					    "%s: free port %d channel %d\n",
+					    __func__, hc->chan[i].port+1, i);
+				pb = hc->chan[i].bch;
+				hc->chan[i].bch = NULL;
+				spin_unlock_irqrestore(&hc->lock, flags);
+				mISDN_freebchannel(pb);
+				kfree(pb);
+				kfree(hc->chan[i].coeff);
+				spin_lock_irqsave(&hc->lock, flags);
+			}
+		}
+	} else {
+		/* remove sync */
+		if (test_bit(HFC_CHIP_PLXSD, &hc->chip)) {
+			hc->syncronized &=
+			   ~(1 << hc->chan[ci].port);
+			plxsd_checksync(hc, 1);
+		}
+		/* free channels */
+		if (hc->chan[ci - 2].bch) {
+			if (debug & DEBUG_HFCMULTI_INIT)
+				printk(KERN_DEBUG
+				    "%s: free port %d channel %d\n",
+				    __func__, hc->chan[ci - 2].port+1,
+				    ci - 2);
+			pb = hc->chan[ci - 2].bch;
+			hc->chan[ci - 2].bch = NULL;
+			spin_unlock_irqrestore(&hc->lock, flags);
+			mISDN_freebchannel(pb);
+			kfree(pb);
+			kfree(hc->chan[ci - 2].coeff);
+			spin_lock_irqsave(&hc->lock, flags);
+		}
+		if (hc->chan[ci - 1].bch) {
+			if (debug & DEBUG_HFCMULTI_INIT)
+				printk(KERN_DEBUG
+				    "%s: free port %d channel %d\n",
+				    __func__, hc->chan[ci - 1].port+1,
+				    ci - 1);
+			pb = hc->chan[ci - 1].bch;
+			hc->chan[ci - 1].bch = NULL;
+			spin_unlock_irqrestore(&hc->lock, flags);
+			mISDN_freebchannel(pb);
+			kfree(pb);
+			kfree(hc->chan[ci - 1].coeff);
+			spin_lock_irqsave(&hc->lock, flags);
+		}
+	}
+
+	spin_unlock_irqrestore(&hc->lock, flags);
+
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_DEBUG "%s: free port %d channel D\n", __func__, pt);
+	mISDN_freedchannel(dch);
+	kfree(dch);
+
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_DEBUG "%s: done!\n", __func__);
+}
+
+static void
+release_card(struct hfc_multi *hc)
+{
+	u_long	flags;
+	int	ch;
+
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_WARNING "%s: release card (%d) entered\n",
+		    __func__, hc->id);
+
+	spin_lock_irqsave(&hc->lock, flags);
+	disable_hwirq(hc);
+	spin_unlock_irqrestore(&hc->lock, flags);
+
+	udelay(1000);
+
+	/* dimm leds */
+	if (hc->leds)
+		hfcmulti_leds(hc);
+
+	/* disable D-channels & B-channels */
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_DEBUG "%s: disable all channels (d and b)\n",
+		    __func__);
+	for (ch = 0; ch <= 31; ch++) {
+		if (hc->chan[ch].dch)
+			release_port(hc, hc->chan[ch].dch);
+	}
+
+	/* release hardware & irq */
+	if (hc->irq) {
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_WARNING "%s: free irq %d\n",
+			    __func__, hc->irq);
+		free_irq(hc->irq, hc);
+		hc->irq = 0;
+
+	}
+	release_io_hfcmulti(hc);
+
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_WARNING "%s: remove instance from list\n",
+		     __func__);
+	list_del(&hc->list);
+
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_WARNING "%s: delete instance\n", __func__);
+	if (hc == syncmaster)
+		syncmaster = NULL;
+	kfree(hc);
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_WARNING "%s: card successfully removed\n",
+		    __func__);
+}
+
+static int
+init_e1_port(struct hfc_multi *hc, struct hm_map *m)
+{
+	struct dchannel	*dch;
+	struct bchannel	*bch;
+	int		ch, ret = 0;
+	char		name[MISDN_MAX_IDLEN];
+
+	dch = kzalloc(sizeof(struct dchannel), GFP_KERNEL);
+	if (!dch)
+		return -ENOMEM;
+	dch->debug = debug;
+	mISDN_initdchannel(dch, MAX_DFRAME_LEN_L1, ph_state_change);
+	dch->hw = hc;
+	dch->dev.Dprotocols = (1 << ISDN_P_TE_E1) | (1 << ISDN_P_NT_E1);
+	dch->dev.Bprotocols = (1 << (ISDN_P_B_RAW & ISDN_P_B_MASK)) |
+	    (1 << (ISDN_P_B_HDLC & ISDN_P_B_MASK));
+	dch->dev.D.send = handle_dmsg;
+	dch->dev.D.ctrl = hfcm_dctrl;
+	dch->dev.nrbchan = (hc->dslot)?30:31;
+	dch->slot = hc->dslot;
+	hc->chan[hc->dslot].dch = dch;
+	hc->chan[hc->dslot].port = 0;
+	hc->chan[hc->dslot].nt_timer = -1;
+	for (ch = 1; ch <= 31; ch++) {
+		if (ch == hc->dslot) /* skip dchannel */
+			continue;
+		bch = kzalloc(sizeof(struct bchannel), GFP_KERNEL);
+		if (!bch) {
+			printk(KERN_ERR "%s: no memory for bchannel\n",
+			    __func__);
+			ret = -ENOMEM;
+			goto free_chan;
+		}
+		hc->chan[ch].coeff = kzalloc(512, GFP_KERNEL);
+		if (!hc->chan[ch].coeff) {
+			printk(KERN_ERR "%s: no memory for coeffs\n",
+			    __func__);
+			ret = -ENOMEM;
+			goto free_chan;
+		}
+		bch->nr = ch;
+		bch->slot = ch;
+		bch->debug = debug;
+		mISDN_initbchannel(bch, MAX_DATA_MEM);
+		bch->hw = hc;
+		bch->ch.send = handle_bmsg;
+		bch->ch.ctrl = hfcm_bctrl;
+		bch->ch.nr = ch;
+		list_add(&bch->ch.list, &dch->dev.bchannels);
+		hc->chan[ch].bch = bch;
+		hc->chan[ch].port = 0;
+		test_and_set_bit(bch->nr, &dch->dev.channelmap[0]);
+	}
+	/* set optical line type */
+	if (port[Port_cnt] & 0x001) {
+		if (!m->opticalsupport)  {
+			printk(KERN_INFO
+			    "This board has no optical "
+			    "support\n");
+		} else {
+			if (debug & DEBUG_HFCMULTI_INIT)
+				printk(KERN_DEBUG
+				    "%s: PORT set optical "
+				    "interfacs: card(%d) "
+				    "port(%d)\n",
+				    __func__,
+				    HFC_cnt + 1, 1);
+			test_and_set_bit(HFC_CFG_OPTICAL,
+			    &hc->chan[hc->dslot].cfg);
+		}
+	}
+	/* set LOS report */
+	if (port[Port_cnt] & 0x004) {
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG "%s: PORT set "
+			    "LOS report: card(%d) port(%d)\n",
+			    __func__, HFC_cnt + 1, 1);
+		test_and_set_bit(HFC_CFG_REPORT_LOS,
+		    &hc->chan[hc->dslot].cfg);
+	}
+	/* set AIS report */
+	if (port[Port_cnt] & 0x008) {
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG "%s: PORT set "
+			    "AIS report: card(%d) port(%d)\n",
+			    __func__, HFC_cnt + 1, 1);
+		test_and_set_bit(HFC_CFG_REPORT_AIS,
+		    &hc->chan[hc->dslot].cfg);
+	}
+	/* set SLIP report */
+	if (port[Port_cnt] & 0x010) {
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG
+			    "%s: PORT set SLIP report: "
+			    "card(%d) port(%d)\n",
+			    __func__, HFC_cnt + 1, 1);
+		test_and_set_bit(HFC_CFG_REPORT_SLIP,
+		    &hc->chan[hc->dslot].cfg);
+	}
+	/* set RDI report */
+	if (port[Port_cnt] & 0x020) {
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG
+			    "%s: PORT set RDI report: "
+			    "card(%d) port(%d)\n",
+			    __func__, HFC_cnt + 1, 1);
+		test_and_set_bit(HFC_CFG_REPORT_RDI,
+		    &hc->chan[hc->dslot].cfg);
+	}
+	/* set CRC-4 Mode */
+	if (!(port[Port_cnt] & 0x100)) {
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG "%s: PORT turn on CRC4 report:"
+				" card(%d) port(%d)\n",
+				__func__, HFC_cnt + 1, 1);
+		test_and_set_bit(HFC_CFG_CRC4,
+		    &hc->chan[hc->dslot].cfg);
+	} else {
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG "%s: PORT turn off CRC4"
+				" report: card(%d) port(%d)\n",
+				__func__, HFC_cnt + 1, 1);
+	}
+	/* set forced clock */
+	if (port[Port_cnt] & 0x0200) {
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG "%s: PORT force getting clock from "
+				"E1: card(%d) port(%d)\n",
+				__func__, HFC_cnt + 1, 1);
+		test_and_set_bit(HFC_CHIP_E1CLOCK_GET, &hc->chip);
+	} else
+	if (port[Port_cnt] & 0x0400) {
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG "%s: PORT force putting clock to "
+				"E1: card(%d) port(%d)\n",
+				__func__, HFC_cnt + 1, 1);
+		test_and_set_bit(HFC_CHIP_E1CLOCK_PUT, &hc->chip);
+	}
+	/* set JATT PLL */
+	if (port[Port_cnt] & 0x0800) {
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG "%s: PORT disable JATT PLL on "
+				"E1: card(%d) port(%d)\n",
+				__func__, HFC_cnt + 1, 1);
+		test_and_set_bit(HFC_CHIP_RX_SYNC, &hc->chip);
+	}
+	/* set elastic jitter buffer */
+	if (port[Port_cnt] & 0x3000) {
+		hc->chan[hc->dslot].jitter = (port[Port_cnt]>>12) & 0x3;
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG
+			    "%s: PORT set elastic "
+			    "buffer to %d: card(%d) port(%d)\n",
+			    __func__, hc->chan[hc->dslot].jitter,
+			    HFC_cnt + 1, 1);
+	} else
+		hc->chan[hc->dslot].jitter = 2; /* default */
+	snprintf(name, MISDN_MAX_IDLEN - 1, "hfc-e1.%d", HFC_cnt + 1);
+	ret = mISDN_register_device(&dch->dev, name);
+	if (ret)
+		goto free_chan;
+	hc->created[0] = 1;
+	return ret;
+free_chan:
+	release_port(hc, dch);
+	return ret;
+}
+
+static int
+init_multi_port(struct hfc_multi *hc, int pt)
+{
+	struct dchannel	*dch;
+	struct bchannel	*bch;
+	int		ch, i, ret = 0;
+	char		name[MISDN_MAX_IDLEN];
+
+	dch = kzalloc(sizeof(struct dchannel), GFP_KERNEL);
+	if (!dch)
+		return -ENOMEM;
+	dch->debug = debug;
+	mISDN_initdchannel(dch, MAX_DFRAME_LEN_L1, ph_state_change);
+	dch->hw = hc;
+	dch->dev.Dprotocols = (1 << ISDN_P_TE_S0) | (1 << ISDN_P_NT_S0);
+	dch->dev.Bprotocols = (1 << (ISDN_P_B_RAW & ISDN_P_B_MASK)) |
+	    (1 << (ISDN_P_B_HDLC & ISDN_P_B_MASK));
+	dch->dev.D.send = handle_dmsg;
+	dch->dev.D.ctrl = hfcm_dctrl;
+	dch->dev.nrbchan = 2;
+	i = pt << 2;
+	dch->slot = i + 2;
+	hc->chan[i + 2].dch = dch;
+	hc->chan[i + 2].port = pt;
+	hc->chan[i + 2].nt_timer = -1;
+	for (ch = 0; ch < dch->dev.nrbchan; ch++) {
+		bch = kzalloc(sizeof(struct bchannel), GFP_KERNEL);
+		if (!bch) {
+			printk(KERN_ERR "%s: no memory for bchannel\n",
+			    __func__);
+			ret = -ENOMEM;
+			goto free_chan;
+		}
+		hc->chan[i + ch].coeff = kzalloc(512, GFP_KERNEL);
+		if (!hc->chan[i + ch].coeff) {
+			printk(KERN_ERR "%s: no memory for coeffs\n",
+			    __func__);
+			ret = -ENOMEM;
+			goto free_chan;
+		}
+		bch->nr = ch + 1;
+		bch->slot = i + ch;
+		bch->debug = debug;
+		mISDN_initbchannel(bch, MAX_DATA_MEM);
+		bch->hw = hc;
+		bch->ch.send = handle_bmsg;
+		bch->ch.ctrl = hfcm_bctrl;
+		bch->ch.nr = ch + 1;
+		list_add(&bch->ch.list, &dch->dev.bchannels);
+		hc->chan[i + ch].bch = bch;
+		hc->chan[i + ch].port = pt;
+		test_and_set_bit(bch->nr, &dch->dev.channelmap[0]);
+	}
+	/* set master clock */
+	if (port[Port_cnt] & 0x001) {
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG
+			    "%s: PROTOCOL set master clock: "
+			    "card(%d) port(%d)\n",
+			    __func__, HFC_cnt + 1, pt + 1);
+		if (dch->dev.D.protocol != ISDN_P_TE_S0) {
+			printk(KERN_ERR "Error: Master clock "
+			    "for port(%d) of card(%d) is only"
+			    " possible with TE-mode\n",
+			    pt + 1, HFC_cnt + 1);
+			ret = -EINVAL;
+			goto free_chan;
+		}
+		if (hc->masterclk >= 0) {
+			printk(KERN_ERR "Error: Master clock "
+			    "for port(%d) of card(%d) already "
+			    "defined for port(%d)\n",
+			    pt + 1, HFC_cnt + 1, hc->masterclk+1);
+			ret = -EINVAL;
+			goto free_chan;
+		}
+		hc->masterclk = pt;
+	}
+	/* set transmitter line to non capacitive */
+	if (port[Port_cnt] & 0x002) {
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG
+			    "%s: PROTOCOL set non capacitive "
+			    "transmitter: card(%d) port(%d)\n",
+			    __func__, HFC_cnt + 1, pt + 1);
+		test_and_set_bit(HFC_CFG_NONCAP_TX,
+		    &hc->chan[i + 2].cfg);
+	}
+	/* disable E-channel */
+	if (port[Port_cnt] & 0x004) {
+	if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG
+			    "%s: PROTOCOL disable E-channel: "
+			    "card(%d) port(%d)\n",
+			    __func__, HFC_cnt + 1, pt + 1);
+		test_and_set_bit(HFC_CFG_DIS_ECHANNEL,
+		    &hc->chan[i + 2].cfg);
+	}
+	snprintf(name, MISDN_MAX_IDLEN - 1, "hfc-%ds.%d/%d",
+		hc->type, HFC_cnt + 1, pt + 1);
+	ret = mISDN_register_device(&dch->dev, name);
+	if (ret)
+		goto free_chan;
+	hc->created[pt] = 1;
+	return ret;
+free_chan:
+	release_port(hc, dch);
+	return ret;
+}
+
+static int
+hfcmulti_init(struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+	struct hm_map	*m = (struct hm_map *)ent->driver_data;
+	int		ret_err = 0;
+	int		pt;
+	struct hfc_multi	*hc;
+	u_long		flags;
+	u_char		dips = 0, pmj = 0; /* dip settings, port mode Jumpers */
+
+	if (HFC_cnt >= MAX_CARDS) {
+		printk(KERN_ERR "too many cards (max=%d).\n",
+			MAX_CARDS);
+		return -EINVAL;
+	}
+	if ((type[HFC_cnt] & 0xff) && (type[HFC_cnt] & 0xff) != m->type) {
+		printk(KERN_WARNING "HFC-MULTI: Card '%s:%s' type %d found but "
+		    "type[%d] %d was supplied as module parameter\n",
+		    m->vendor_name, m->card_name, m->type, HFC_cnt,
+		    type[HFC_cnt] & 0xff);
+		printk(KERN_WARNING "HFC-MULTI: Load module without parameters "
+			"first, to see cards and their types.");
+		return -EINVAL;
+	}
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_DEBUG "%s: Registering %s:%s chip type %d (0x%x)\n",
+		    __func__, m->vendor_name, m->card_name, m->type,
+		    type[HFC_cnt]);
+
+	/* allocate card+fifo structure */
+	hc = kzalloc(sizeof(struct hfc_multi), GFP_KERNEL);
+	if (!hc) {
+		printk(KERN_ERR "No kmem for HFC-Multi card\n");
+		return -ENOMEM;
+	}
+	spin_lock_init(&hc->lock);
+	hc->mtyp = m;
+	hc->type =  m->type;
+	hc->ports = m->ports;
+	hc->id = HFC_cnt;
+	hc->pcm = pcm[HFC_cnt];
+	hc->io_mode = iomode[HFC_cnt];
+	if (dslot[HFC_cnt] < 0) {
+		hc->dslot = 0;
+		printk(KERN_INFO "HFC-E1 card has disabled D-channel, but "
+			"31 B-channels\n");
+	} if (dslot[HFC_cnt] > 0 && dslot[HFC_cnt] < 32) {
+		hc->dslot = dslot[HFC_cnt];
+		printk(KERN_INFO "HFC-E1 card has alternating D-channel on "
+			"time slot %d\n", dslot[HFC_cnt]);
+	} else
+		hc->dslot = 16;
+
+	/* set chip specific features */
+	hc->masterclk = -1;
+	if (type[HFC_cnt] & 0x100) {
+		test_and_set_bit(HFC_CHIP_ULAW, &hc->chip);
+		silence = 0xff; /* ulaw silence */
+	} else
+		silence = 0x2a; /* alaw silence */
+	if (!(type[HFC_cnt] & 0x200))
+		test_and_set_bit(HFC_CHIP_DTMF, &hc->chip);
+
+	if (type[HFC_cnt] & 0x800)
+		test_and_set_bit(HFC_CHIP_PCM_SLAVE, &hc->chip);
+	if (type[HFC_cnt] & 0x1000) {
+		test_and_set_bit(HFC_CHIP_PCM_MASTER, &hc->chip);
+		test_and_clear_bit(HFC_CHIP_PCM_SLAVE, &hc->chip);
+	}
+	if (type[HFC_cnt] & 0x4000)
+		test_and_set_bit(HFC_CHIP_EXRAM_128, &hc->chip);
+	if (type[HFC_cnt] & 0x8000)
+		test_and_set_bit(HFC_CHIP_EXRAM_512, &hc->chip);
+	hc->slots = 32;
+	if (type[HFC_cnt] & 0x10000)
+		hc->slots = 64;
+	if (type[HFC_cnt] & 0x20000)
+		hc->slots = 128;
+	if (type[HFC_cnt] & 0x80000) {
+		test_and_set_bit(HFC_CHIP_WATCHDOG, &hc->chip);
+		hc->wdcount = 0;
+		hc->wdbyte = V_GPIO_OUT2;
+		printk(KERN_NOTICE "Watchdog enabled\n");
+	}
+
+	/* setup pci, hc->slots may change due to PLXSD */
+	ret_err = setup_pci(hc, pdev, ent);
+	if (ret_err) {
+		if (hc == syncmaster)
+			syncmaster = NULL;
+		kfree(hc);
+		return ret_err;
+	}
+
+	/* crate channels */
+	for (pt = 0; pt < hc->ports; pt++) {
+		if (Port_cnt >= MAX_PORTS) {
+			printk(KERN_ERR "too many ports (max=%d).\n",
+				MAX_PORTS);
+			ret_err = -EINVAL;
+			goto free_card;
+		}
+		if (hc->type == 1)
+			ret_err = init_e1_port(hc, m);
+		else
+			ret_err = init_multi_port(hc, pt);
+		if (debug & DEBUG_HFCMULTI_INIT)
+			printk(KERN_DEBUG
+			    "%s: Registering D-channel, card(%d) port(%d)"
+			    "result %d\n",
+			    __func__, HFC_cnt + 1, pt, ret_err);
+
+		if (ret_err) {
+			while (pt) { /* release already registered ports */
+				pt--;
+				release_port(hc, hc->chan[(pt << 2) + 2].dch);
+			}
+			goto free_card;
+		}
+		Port_cnt++;
+	}
+
+	/* disp switches */
+	switch (m->dip_type) {
+	case DIP_4S:
+		/*
+		 * get DIP Setting for beroNet 1S/2S/4S cards
+		 *  check if Port Jumper config matches
+		 * module param 'protocol'
+		 * DIP Setting: (collect GPIO 13/14/15 (R_GPIO_IN1) +
+		 * GPI 19/23 (R_GPI_IN2))
+		 */
+		dips = ((~HFC_inb(hc, R_GPIO_IN1) & 0xE0) >> 5) |
+			((~HFC_inb(hc, R_GPI_IN2) & 0x80) >> 3) |
+			(~HFC_inb(hc, R_GPI_IN2) & 0x08);
+
+		/* Port mode (TE/NT) jumpers */
+		pmj = ((HFC_inb(hc, R_GPI_IN3) >> 4)  & 0xf);
+
+		if (test_bit(HFC_CHIP_B410P, &hc->chip))
+			pmj = ~pmj & 0xf;
+
+		printk(KERN_INFO "%s: %s DIPs(0x%x) jumpers(0x%x)\n",
+			m->vendor_name, m->card_name, dips, pmj);
+		break;
+	case DIP_8S:
+		/*
+		 * get DIP Setting for beroNet 8S0+ cards
+		 *
+		 * enable PCI auxbridge function
+		 */
+		HFC_outb(hc, R_BRG_PCM_CFG, 1 | V_PCM_CLK);
+		/* prepare access to auxport */
+		outw(0x4000, hc->pci_iobase + 4);
+		/*
+		 * some dummy reads are required to
+		 * read valid DIP switch data
+		 */
+		dips = inb(hc->pci_iobase);
+		dips = inb(hc->pci_iobase);
+		dips = inb(hc->pci_iobase);
+		dips = ~inb(hc->pci_iobase) & 0x3F;
+		outw(0x0, hc->pci_iobase + 4);
+		/* disable PCI auxbridge function */
+		HFC_outb(hc, R_BRG_PCM_CFG, V_PCM_CLK);
+		printk(KERN_INFO "%s: %s DIPs(0x%x)\n",
+		    m->vendor_name, m->card_name, dips);
+		break;
+	case DIP_E1:
+		/*
+		 * get DIP Setting for beroNet E1 cards
+		 * DIP Setting: collect GPI 4/5/6/7 (R_GPI_IN0)
+		 */
+		dips = (~HFC_inb(hc, R_GPI_IN0) & 0xF0)>>4;
+		printk(KERN_INFO "%s: %s DIPs(0x%x)\n",
+		    m->vendor_name, m->card_name, dips);
+		break;
+	}
+
+	/* add to list */
+	spin_lock_irqsave(&HFClock, flags);
+	list_add_tail(&hc->list, &HFClist);
+	spin_unlock_irqrestore(&HFClock, flags);
+
+	/* initialize hardware */
+	ret_err = init_card(hc);
+	if (ret_err) {
+		printk(KERN_ERR "init card returns %d\n", ret_err);
+		release_card(hc);
+		return ret_err;
+	}
+
+	/* start IRQ and return */
+	spin_lock_irqsave(&hc->lock, flags);
+	enable_hwirq(hc);
+	spin_unlock_irqrestore(&hc->lock, flags);
+	return 0;
+
+free_card:
+	release_io_hfcmulti(hc);
+	if (hc == syncmaster)
+		syncmaster = NULL;
+	kfree(hc);
+	return ret_err;
+}
+
+static void __devexit hfc_remove_pci(struct pci_dev *pdev)
+{
+	struct hfc_multi	*card = pci_get_drvdata(pdev);
+	u_long			flags;
+
+	if (debug)
+		printk(KERN_INFO "removing hfc_multi card vendor:%x "
+		    "device:%x subvendor:%x subdevice:%x\n",
+		    pdev->vendor, pdev->device,
+		    pdev->subsystem_vendor, pdev->subsystem_device);
+
+	if (card) {
+		spin_lock_irqsave(&HFClock, flags);
+		release_card(card);
+		spin_unlock_irqrestore(&HFClock, flags);
+	}  else {
+		if (debug)
+			printk(KERN_WARNING "%s: drvdata allready removed\n",
+			    __func__);
+	}
+}
+
+#define	VENDOR_CCD	"Cologne Chip AG"
+#define	VENDOR_BN	"beroNet GmbH"
+#define	VENDOR_DIG	"Digium Inc."
+#define VENDOR_JH	"Junghanns.NET GmbH"
+#define VENDOR_PRIM	"PrimuX"
+
+static const struct hm_map hfcm_map[] = {
+/*0*/	{VENDOR_BN, "HFC-1S Card (mini PCI)", 4, 1, 1, 3, 0, DIP_4S, 0},
+/*1*/	{VENDOR_BN, "HFC-2S Card", 4, 2, 1, 3, 0, DIP_4S},
+/*2*/	{VENDOR_BN, "HFC-2S Card (mini PCI)", 4, 2, 1, 3, 0, DIP_4S, 0},
+/*3*/	{VENDOR_BN, "HFC-4S Card", 4, 4, 1, 2, 0, DIP_4S, 0},
+/*4*/	{VENDOR_BN, "HFC-4S Card (mini PCI)", 4, 4, 1, 2, 0, 0, 0},
+/*5*/	{VENDOR_CCD, "HFC-4S Eval (old)", 4, 4, 0, 0, 0, 0, 0},
+/*6*/	{VENDOR_CCD, "HFC-4S IOB4ST", 4, 4, 1, 2, 0, 0, 0},
+/*7*/	{VENDOR_CCD, "HFC-4S", 4, 4, 1, 2, 0, 0, 0},
+/*8*/	{VENDOR_DIG, "HFC-4S Card", 4, 4, 0, 2, 0, 0, HFC_IO_MODE_REGIO},
+/*9*/	{VENDOR_CCD, "HFC-4S Swyx 4xS0 SX2 QuadBri", 4, 4, 1, 2, 0, 0, 0},
+/*10*/	{VENDOR_JH, "HFC-4S (junghanns 2.0)", 4, 4, 1, 2, 0, 0, 0},
+/*11*/	{VENDOR_PRIM, "HFC-2S Primux Card", 4, 2, 0, 0, 0, 0, 0},
+
+/*12*/	{VENDOR_BN, "HFC-8S Card", 8, 8, 1, 0, 0, 0, 0},
+/*13*/	{VENDOR_BN, "HFC-8S Card (+)", 8, 8, 1, 8, 0, DIP_8S,
+		HFC_IO_MODE_REGIO},
+/*14*/	{VENDOR_CCD, "HFC-8S Eval (old)", 8, 8, 0, 0, 0, 0, 0},
+/*15*/	{VENDOR_CCD, "HFC-8S IOB4ST Recording", 8, 8, 1, 0, 0, 0, 0},
+
+/*16*/	{VENDOR_CCD, "HFC-8S IOB8ST", 8, 8, 1, 0, 0, 0, 0},
+/*17*/	{VENDOR_CCD, "HFC-8S", 8, 8, 1, 0, 0, 0, 0},
+/*18*/	{VENDOR_CCD, "HFC-8S", 8, 8, 1, 0, 0, 0, 0},
+
+/*19*/	{VENDOR_BN, "HFC-E1 Card", 1, 1, 0, 1, 0, DIP_E1, 0},
+/*20*/	{VENDOR_BN, "HFC-E1 Card (mini PCI)", 1, 1, 0, 1, 0, 0, 0},
+/*21*/	{VENDOR_BN, "HFC-E1+ Card (Dual)", 1, 1, 0, 1, 0, DIP_E1, 0},
+/*22*/	{VENDOR_BN, "HFC-E1 Card (Dual)", 1, 1, 0, 1, 0, DIP_E1, 0},
+
+/*23*/	{VENDOR_CCD, "HFC-E1 Eval (old)", 1, 1, 0, 0, 0, 0, 0},
+/*24*/	{VENDOR_CCD, "HFC-E1 IOB1E1", 1, 1, 0, 1, 0, 0, 0},
+/*25*/	{VENDOR_CCD, "HFC-E1", 1, 1, 0, 1, 0, 0, 0},
+
+/*26*/	{VENDOR_CCD, "HFC-4S Speech Design", 4, 4, 0, 0, 0, 0,
+		HFC_IO_MODE_PLXSD},
+/*27*/	{VENDOR_CCD, "HFC-E1 Speech Design", 1, 1, 0, 0, 0, 0,
+		HFC_IO_MODE_PLXSD},
+/*28*/	{VENDOR_CCD, "HFC-4S OpenVox", 4, 4, 1, 0, 0, 0, 0},
+/*29*/	{VENDOR_CCD, "HFC-2S OpenVox", 4, 2, 1, 0, 0, 0, 0},
+/*30*/	{VENDOR_CCD, "HFC-8S OpenVox", 8, 8, 1, 0, 0, 0, 0},
+};
+
+#undef H
+#define H(x)	((unsigned long)&hfcm_map[x])
+static struct pci_device_id hfmultipci_ids[] __devinitdata = {
+
+	/* Cards with HFC-4S Chip */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC4S, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_BN1SM, 0, 0, H(0)}, /* BN1S mini PCI */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC4S, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_BN2S, 0, 0, H(1)}, /* BN2S */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC4S, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_BN2SM, 0, 0, H(2)}, /* BN2S mini PCI */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC4S, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_BN4S, 0, 0, H(3)}, /* BN4S */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC4S, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_BN4SM, 0, 0, H(4)}, /* BN4S mini PCI */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC4S, PCI_VENDOR_ID_CCD,
+		PCI_DEVICE_ID_CCD_HFC4S, 0, 0, H(5)}, /* Old Eval */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC4S, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_IOB4ST, 0, 0, H(6)}, /* IOB4ST */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC4S, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_HFC4S, 0, 0, H(7)}, /* 4S */
+	{ PCI_VENDOR_ID_DIGIUM, PCI_DEVICE_ID_DIGIUM_HFC4S,
+		PCI_VENDOR_ID_DIGIUM, PCI_DEVICE_ID_DIGIUM_HFC4S, 0, 0, H(8)},
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC4S, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_SWYX4S, 0, 0, H(9)}, /* 4S Swyx */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC4S, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_JH4S20, 0, 0, H(10)},
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC4S, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_PMX2S, 0, 0, H(11)}, /* Primux */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC4S, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_OV4S, 0, 0, H(28)}, /* OpenVox 4 */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC4S, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_OV2S, 0, 0, H(29)}, /* OpenVox 2 */
+
+	/* Cards with HFC-8S Chip */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC8S, PCI_VENDOR_ID_CCD,
+	PCI_SUBDEVICE_ID_CCD_BN8S, 0, 0, H(12)}, /* BN8S */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC8S, PCI_VENDOR_ID_CCD,
+	PCI_SUBDEVICE_ID_CCD_BN8SP, 0, 0, H(13)}, /* BN8S+ */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC8S, PCI_VENDOR_ID_CCD,
+	PCI_DEVICE_ID_CCD_HFC8S, 0, 0, H(14)}, /* old Eval */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC8S, PCI_VENDOR_ID_CCD,
+	PCI_SUBDEVICE_ID_CCD_IOB8STR, 0, 0, H(15)},
+	    /* IOB8ST Recording */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC8S, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_IOB8ST, 0, 0, H(16)}, /* IOB8ST  */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC8S, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_IOB8ST_1, 0, 0, H(17)}, /* IOB8ST  */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC8S, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_HFC8S, 0, 0, H(18)}, /* 8S */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC8S, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_OV8S, 0, 0, H(30)}, /* OpenVox 8 */
+
+
+	/* Cards with HFC-E1 Chip */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFCE1, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_BNE1, 0, 0, H(19)}, /* BNE1 */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFCE1, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_BNE1M, 0, 0, H(20)}, /* BNE1 mini PCI */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFCE1, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_BNE1DP, 0, 0, H(21)}, /* BNE1 + (Dual) */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFCE1, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_BNE1D, 0, 0, H(22)}, /* BNE1 (Dual) */
+
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFCE1, PCI_VENDOR_ID_CCD,
+		PCI_DEVICE_ID_CCD_HFCE1, 0, 0, H(23)}, /* Old Eval */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFCE1, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_IOB1E1, 0, 0, H(24)}, /* IOB1E1 */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFCE1, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_HFCE1, 0, 0, H(25)}, /* E1 */
+
+	{ PCI_VENDOR_ID_PLX, PCI_DEVICE_ID_PLX_9030, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_SPD4S, 0, 0, H(26)}, /* PLX PCI Bridge */
+	{ PCI_VENDOR_ID_PLX, PCI_DEVICE_ID_PLX_9030, PCI_VENDOR_ID_CCD,
+		PCI_SUBDEVICE_ID_CCD_SPDE1, 0, 0, H(27)}, /* PLX PCI Bridge */
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC4S, PCI_ANY_ID, PCI_ANY_ID,
+		0, 0, 0},
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFC8S, PCI_ANY_ID, PCI_ANY_ID,
+		0, 0, 0},
+	{ PCI_VENDOR_ID_CCD, PCI_DEVICE_ID_CCD_HFCE1, PCI_ANY_ID, PCI_ANY_ID,
+		0, 0, 0},
+	{0, }
+};
+#undef H
+
+MODULE_DEVICE_TABLE(pci, hfmultipci_ids);
+
+static int
+hfcmulti_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
+{
+	struct hm_map	*m = (struct hm_map *)ent->driver_data;
+	int		ret;
+
+	if (m == NULL) {
+		if (ent->vendor == PCI_VENDOR_ID_CCD)
+			if (ent->device == PCI_DEVICE_ID_CCD_HFC4S ||
+			    ent->device == PCI_DEVICE_ID_CCD_HFC8S ||
+			    ent->device == PCI_DEVICE_ID_CCD_HFCE1)
+				printk(KERN_ERR
+				    "unknown HFC multiport controller "
+				    "(vendor:%x device:%x subvendor:%x "
+				    "subdevice:%x) Please contact the "
+				    "driver maintainer for support.\n",
+				    ent->vendor, ent->device,
+				    ent->subvendor, ent->subdevice);
+		return -ENODEV;
+	}
+	ret = hfcmulti_init(pdev, ent);
+	if (ret)
+		return ret;
+	HFC_cnt++;
+	printk(KERN_INFO "%d devices registered\n", HFC_cnt);
+	return 0;
+}
+
+static struct pci_driver hfcmultipci_driver = {
+	.name		= "hfc_multi",
+	.probe		= hfcmulti_probe,
+	.remove		= __devexit_p(hfc_remove_pci),
+	.id_table	= hfmultipci_ids,
+};
+
+static void __exit
+HFCmulti_cleanup(void)
+{
+	struct hfc_multi *card, *next;
+
+	/* unload interrupt function symbol */
+	if (hfc_interrupt)
+		symbol_put(ztdummy_extern_interrupt);
+	if (register_interrupt)
+		symbol_put(ztdummy_register_interrupt);
+	if (unregister_interrupt) {
+		if (interrupt_registered) {
+			interrupt_registered = 0;
+			unregister_interrupt();
+		}
+		symbol_put(ztdummy_unregister_interrupt);
+	}
+
+	list_for_each_entry_safe(card, next, &HFClist, list)
+		release_card(card);
+	/* get rid of all devices of this driver */
+	pci_unregister_driver(&hfcmultipci_driver);
+}
+
+static int __init
+HFCmulti_init(void)
+{
+	int err;
+
+#ifdef IRQ_DEBUG
+	printk(KERN_ERR "%s: IRQ_DEBUG IS ENABLED!\n", __func__);
+#endif
+
+	spin_lock_init(&HFClock);
+	spin_lock_init(&plx_lock);
+
+	if (debug & DEBUG_HFCMULTI_INIT)
+		printk(KERN_DEBUG "%s: init entered\n", __func__);
+
+#ifdef __BIG_ENDIAN
+#error "not running on big endian machines now"
+#endif
+	hfc_interrupt = symbol_get(ztdummy_extern_interrupt);
+	register_interrupt = symbol_get(ztdummy_register_interrupt);
+	unregister_interrupt = symbol_get(ztdummy_unregister_interrupt);
+	printk(KERN_INFO "mISDN: HFC-multi driver %s\n",
+	    hfcmulti_revision);
+
+	switch (poll) {
+	case 0:
+		poll_timer = 6;
+		poll = 128;
+		break;
+		/*
+		 * wenn dieses break nochmal verschwindet,
+		 * gibt es heisse ohren :-)
+		 * "without the break you will get hot ears ???"
+		 */
+	case 8:
+		poll_timer = 2;
+		break;
+	case 16:
+		poll_timer = 3;
+		break;
+	case 32:
+		poll_timer = 4;
+		break;
+	case 64:
+		poll_timer = 5;
+		break;
+	case 128:
+		poll_timer = 6;
+		break;
+	case 256:
+		poll_timer = 7;
+		break;
+	default:
+		printk(KERN_ERR
+		    "%s: Wrong poll value (%d).\n", __func__, poll);
+		err = -EINVAL;
+		return err;
+
+	}
+
+	err = pci_register_driver(&hfcmultipci_driver);
+	if (err < 0) {
+		printk(KERN_ERR "error registering pci driver: %x\n", err);
+		if (hfc_interrupt)
+			symbol_put(ztdummy_extern_interrupt);
+		if (register_interrupt)
+			symbol_put(ztdummy_register_interrupt);
+		if (unregister_interrupt) {
+			if (interrupt_registered) {
+				interrupt_registered = 0;
+				unregister_interrupt();
+			}
+			symbol_put(ztdummy_unregister_interrupt);
+		}
+		return err;
+	}
+	return 0;
+}
+
+
+module_init(HFCmulti_init);
+module_exit(HFCmulti_cleanup);
diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h
index c3b1761aba26..ffe479ba0779 100644
--- a/include/linux/pci_ids.h
+++ b/include/linux/pci_ids.h
@@ -1832,7 +1832,13 @@
 #define PCI_DEVICE_ID_MOXA_C320		0x3200
 
 #define PCI_VENDOR_ID_CCD		0x1397
+#define PCI_DEVICE_ID_CCD_HFC4S		0x08B4
+#define PCI_SUBDEVICE_ID_CCD_PMX2S	0x1234
+#define PCI_DEVICE_ID_CCD_HFC8S		0x16B8
 #define PCI_DEVICE_ID_CCD_2BD0		0x2bd0
+#define PCI_DEVICE_ID_CCD_HFCE1		0x30B1
+#define PCI_SUBDEVICE_ID_CCD_SPD4S	0x3136
+#define PCI_SUBDEVICE_ID_CCD_SPDE1	0x3137
 #define PCI_DEVICE_ID_CCD_B000		0xb000
 #define PCI_DEVICE_ID_CCD_B006		0xb006
 #define PCI_DEVICE_ID_CCD_B007		0xb007
@@ -1842,8 +1848,32 @@
 #define PCI_DEVICE_ID_CCD_B00B		0xb00b
 #define PCI_DEVICE_ID_CCD_B00C		0xb00c
 #define PCI_DEVICE_ID_CCD_B100		0xb100
+#define PCI_SUBDEVICE_ID_CCD_IOB4ST	0xB520
+#define PCI_SUBDEVICE_ID_CCD_IOB8STR	0xB521
+#define PCI_SUBDEVICE_ID_CCD_IOB8ST	0xB522
+#define PCI_SUBDEVICE_ID_CCD_IOB1E1	0xB523
+#define PCI_SUBDEVICE_ID_CCD_SWYX4S	0xB540
+#define PCI_SUBDEVICE_ID_CCD_JH4S20	0xB550
+#define PCI_SUBDEVICE_ID_CCD_IOB8ST_1	0xB552
+#define PCI_SUBDEVICE_ID_CCD_BN4S	0xB560
+#define PCI_SUBDEVICE_ID_CCD_BN8S	0xB562
+#define PCI_SUBDEVICE_ID_CCD_BNE1	0xB563
+#define PCI_SUBDEVICE_ID_CCD_BNE1D	0xB564
+#define PCI_SUBDEVICE_ID_CCD_BNE1DP	0xB565
+#define PCI_SUBDEVICE_ID_CCD_BN2S	0xB566
+#define PCI_SUBDEVICE_ID_CCD_BN1SM	0xB567
+#define PCI_SUBDEVICE_ID_CCD_BN4SM	0xB568
+#define PCI_SUBDEVICE_ID_CCD_BN2SM	0xB569
+#define PCI_SUBDEVICE_ID_CCD_BNE1M	0xB56A
+#define PCI_SUBDEVICE_ID_CCD_BN8SP	0xB56B
+#define PCI_SUBDEVICE_ID_CCD_HFC4S	0xB620
+#define PCI_SUBDEVICE_ID_CCD_HFC8S	0xB622
 #define PCI_DEVICE_ID_CCD_B700		0xb700
 #define PCI_DEVICE_ID_CCD_B701		0xb701
+#define PCI_SUBDEVICE_ID_CCD_HFCE1	0xC523
+#define PCI_SUBDEVICE_ID_CCD_OV2S	0xE884
+#define PCI_SUBDEVICE_ID_CCD_OV4S	0xE888
+#define PCI_SUBDEVICE_ID_CCD_OV8S	0xE998
 
 #define PCI_VENDOR_ID_EXAR		0x13a8
 #define PCI_DEVICE_ID_EXAR_XR17C152	0x0152
@@ -2523,6 +2553,9 @@
 
 #define PCI_VENDOR_ID_3COM_2		0xa727
 
+#define PCI_VENDOR_ID_DIGIUM		0xd161
+#define PCI_DEVICE_ID_DIGIUM_HFC4S	0xb410
+
 #define PCI_SUBVENDOR_ID_EXSYS		0xd84d
 #define PCI_SUBDEVICE_ID_EXSYS_4014	0x4014
 #define PCI_SUBDEVICE_ID_EXSYS_4055	0x4055

From 3712b42d4b1bec29a4232a6673bf2e6dcc5faa68 Mon Sep 17 00:00:00 2001
From: Karsten Keil <kkeil@suse.de>
Date: Sun, 27 Jul 2008 02:02:10 +0200
Subject: [PATCH 1118/1171] Add layer1 over IP support

Implement a ISDN over IP tunnel to use mISDN hardware on
remote locations.

Signed-off-by: Karsten Keil <kkeil@suse.de>
---
 drivers/isdn/mISDN/Kconfig       |   19 +-
 drivers/isdn/mISDN/Makefile      |    2 +
 drivers/isdn/mISDN/l1oip.h       |   91 ++
 drivers/isdn/mISDN/l1oip_codec.c |  374 ++++++++
 drivers/isdn/mISDN/l1oip_core.c  | 1518 ++++++++++++++++++++++++++++++
 5 files changed, 2003 insertions(+), 1 deletion(-)
 create mode 100644 drivers/isdn/mISDN/l1oip.h
 create mode 100644 drivers/isdn/mISDN/l1oip_codec.c
 create mode 100644 drivers/isdn/mISDN/l1oip_core.c

diff --git a/drivers/isdn/mISDN/Kconfig b/drivers/isdn/mISDN/Kconfig
index 6a97e86e7f21..4938355c4072 100644
--- a/drivers/isdn/mISDN/Kconfig
+++ b/drivers/isdn/mISDN/Kconfig
@@ -23,5 +23,22 @@ config MISDN_DSP
 	  and get more informations about this module and it's usage.
 	  If unsure, say 'N'.
 
-	source "drivers/isdn/hardware/mISDN/Kconfig"
+config MISDN_L1OIP
+	tristate "ISDN over IP tunnel"
+	depends on MISDN
+	help
+	  Enable support for ISDN over IP tunnel.
+
+	  It features:
+	    - dynamic IP exchange, if one or both peers have dynamic IPs
+	    - BRI (S0) and PRI (S2M) interface
+	    - layer 1 control via network keepalive frames
+	    - direct tunneling of physical interface via IP
+
+	  NOTE: This protocol is called 'Layer 1 over IP' and is not
+	  compatible with ISDNoIP (Agfeo) or TDMoIP. Protocol description is
+	  provided in the source code.
+
+source "drivers/isdn/hardware/mISDN/Kconfig"
+
 endif #MISDN
diff --git a/drivers/isdn/mISDN/Makefile b/drivers/isdn/mISDN/Makefile
index 7f1a21804208..1cb5e633cf75 100644
--- a/drivers/isdn/mISDN/Makefile
+++ b/drivers/isdn/mISDN/Makefile
@@ -4,8 +4,10 @@
 
 obj-$(CONFIG_MISDN) += mISDN_core.o
 obj-$(CONFIG_MISDN_DSP) += mISDN_dsp.o
+obj-$(CONFIG_MISDN_L1OIP) += l1oip.o
 
 # multi objects
 
 mISDN_core-objs := core.o fsm.o socket.o hwchannel.o stack.o layer1.o layer2.o tei.o timerdev.o
 mISDN_dsp-objs := dsp_core.o dsp_cmx.o dsp_tones.o dsp_dtmf.o dsp_audio.o dsp_blowfish.o dsp_pipeline.o dsp_hwec.o
+l1oip-objs := l1oip_core.o l1oip_codec.o
diff --git a/drivers/isdn/mISDN/l1oip.h b/drivers/isdn/mISDN/l1oip.h
new file mode 100644
index 000000000000..a23d575449f6
--- /dev/null
+++ b/drivers/isdn/mISDN/l1oip.h
@@ -0,0 +1,91 @@
+/*
+ * see notice in l1oip.c
+ */
+
+/* debugging */
+#define DEBUG_L1OIP_INIT	0x00010000
+#define DEBUG_L1OIP_SOCKET	0x00020000
+#define DEBUG_L1OIP_MGR		0x00040000
+#define DEBUG_L1OIP_MSG		0x00080000
+
+/* enable to disorder received bchannels by sequence 2143658798... */
+/*
+#define REORDER_DEBUG
+*/
+
+/* frames */
+#define L1OIP_MAX_LEN		2048		/* max packet size form l2 */
+#define L1OIP_MAX_PERFRAME	1400		/* max data size in one frame */
+
+
+/* timers */
+#define L1OIP_KEEPALIVE		15
+#define L1OIP_TIMEOUT		65
+
+
+/* socket */
+#define L1OIP_DEFAULTPORT	931
+
+
+/* channel structure */
+struct l1oip_chan {
+	struct dchannel       	*dch;
+	struct bchannel       	*bch;
+	u32			tx_counter;	/* counts xmit bytes/packets */
+	u32			rx_counter;	/* counts recv bytes/packets */
+	u32			codecstate;	/* used by codec to save data */
+#ifdef REORDER_DEBUG
+	int			disorder_flag;
+	struct sk_buff		*disorder_skb;
+	u32			disorder_cnt;
+#endif
+};
+
+
+/* card structure */
+struct l1oip {
+	struct list_head        list;
+
+	/* card */
+	int			registered;	/* if registered with mISDN */
+	char			name[MISDN_MAX_IDLEN];
+	int			idx;		/* card index */
+	int			pri;		/* 1=pri, 0=bri */
+	int			d_idx;		/* current dchannel number */
+	int			b_num;		/* number of bchannels */
+	u32			id;		/* id of connection */
+	int			ondemand;	/* if transmis. is on demand */
+	int			bundle;		/* bundle channels in one frm */
+	int			codec;		/* codec to use for transmis. */
+	int			limit;		/* limit number of bchannels */
+
+	/* timer */
+	struct timer_list 	keep_tl;
+	struct timer_list 	timeout_tl;
+	int			timeout_on;
+	struct work_struct	workq;
+
+	/* socket */
+	struct socket 		*socket;	/* if set, socket is created */
+	struct completion 	socket_complete;/* completion of sock thread */
+	struct task_struct	*socket_thread;
+	spinlock_t 		socket_lock;	/* access sock outside thread */
+	u32			remoteip;	/* if all set, ip is assigned */
+	u16	 		localport;	/* must always be set */
+	u16	 		remoteport;	/* must always be set */
+	struct sockaddr_in	sin_local;	/* local socket name */
+	struct sockaddr_in	sin_remote;	/* remote socket name */
+	struct msghdr		sendmsg;	/* ip message to send */
+	struct iovec		sendiov;	/* iov for message */
+
+	/* frame */
+	struct l1oip_chan	chan[128];	/* channel instances */
+};
+
+extern int l1oip_law_to_4bit(u8 *data, int len, u8 *result, u32 *state);
+extern int l1oip_4bit_to_law(u8 *data, int len, u8 *result);
+extern int l1oip_alaw_to_ulaw(u8 *data, int len, u8 *result);
+extern int l1oip_ulaw_to_alaw(u8 *data, int len, u8 *result);
+extern void l1oip_4bit_free(void);
+extern int l1oip_4bit_alloc(int ulaw);
+
diff --git a/drivers/isdn/mISDN/l1oip_codec.c b/drivers/isdn/mISDN/l1oip_codec.c
new file mode 100644
index 000000000000..a2dc4570ef43
--- /dev/null
+++ b/drivers/isdn/mISDN/l1oip_codec.c
@@ -0,0 +1,374 @@
+/*
+
+ * l1oip_codec.c  generic codec using lookup table
+ *  -> conversion from a-Law to u-Law
+ *  -> conversion from u-Law to a-Law
+ *  -> compression by reducing the number of sample resolution to 4
+ *
+ * NOTE: It is not compatible with any standard codec like ADPCM.
+ *
+ * Author	Andreas Eversberg (jolly@eversberg.eu)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2, or (at your option)
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+
+ */
+
+/*
+
+How the codec works:
+--------------------
+
+The volume is increased to increase the dynamic range of the audio signal.
+Each sample is converted to a-LAW with only 16 steps of level resolution.
+A pair of two samples are stored in one byte.
+
+The first byte is stored in the upper bits, the second byte is stored in the
+lower bits.
+
+To speed up compression and decompression, two lookup tables are formed:
+
+- 16 bits index for two samples (law encoded) with 8 bit compressed result.
+- 8 bits index for one compressed data with 16 bits decompressed result.
+
+NOTE: The bytes are handled as they are law-encoded.
+
+*/
+
+#include <linux/vmalloc.h>
+#include <linux/mISDNif.h>
+#include "core.h"
+
+/* definitions of codec. don't use calculations, code may run slower. */
+
+static u8 *table_com;
+static u16 *table_dec;
+
+
+/* alaw -> ulaw */
+static u8 alaw_to_ulaw[256] =
+{
+	0xab, 0x2b, 0xe3, 0x63, 0x8b, 0x0b, 0xc9, 0x49,
+	0xba, 0x3a, 0xf6, 0x76, 0x9b, 0x1b, 0xd7, 0x57,
+	0xa3, 0x23, 0xdd, 0x5d, 0x83, 0x03, 0xc1, 0x41,
+	0xb2, 0x32, 0xeb, 0x6b, 0x93, 0x13, 0xcf, 0x4f,
+	0xaf, 0x2f, 0xe7, 0x67, 0x8f, 0x0f, 0xcd, 0x4d,
+	0xbe, 0x3e, 0xfe, 0x7e, 0x9f, 0x1f, 0xdb, 0x5b,
+	0xa7, 0x27, 0xdf, 0x5f, 0x87, 0x07, 0xc5, 0x45,
+	0xb6, 0x36, 0xef, 0x6f, 0x97, 0x17, 0xd3, 0x53,
+	0xa9, 0x29, 0xe1, 0x61, 0x89, 0x09, 0xc7, 0x47,
+	0xb8, 0x38, 0xf2, 0x72, 0x99, 0x19, 0xd5, 0x55,
+	0xa1, 0x21, 0xdc, 0x5c, 0x81, 0x01, 0xbf, 0x3f,
+	0xb0, 0x30, 0xe9, 0x69, 0x91, 0x11, 0xce, 0x4e,
+	0xad, 0x2d, 0xe5, 0x65, 0x8d, 0x0d, 0xcb, 0x4b,
+	0xbc, 0x3c, 0xfa, 0x7a, 0x9d, 0x1d, 0xd9, 0x59,
+	0xa5, 0x25, 0xde, 0x5e, 0x85, 0x05, 0xc3, 0x43,
+	0xb4, 0x34, 0xed, 0x6d, 0x95, 0x15, 0xd1, 0x51,
+	0xac, 0x2c, 0xe4, 0x64, 0x8c, 0x0c, 0xca, 0x4a,
+	0xbb, 0x3b, 0xf8, 0x78, 0x9c, 0x1c, 0xd8, 0x58,
+	0xa4, 0x24, 0xde, 0x5e, 0x84, 0x04, 0xc2, 0x42,
+	0xb3, 0x33, 0xec, 0x6c, 0x94, 0x14, 0xd0, 0x50,
+	0xb0, 0x30, 0xe8, 0x68, 0x90, 0x10, 0xce, 0x4e,
+	0xbf, 0x3f, 0xfe, 0x7e, 0xa0, 0x20, 0xdc, 0x5c,
+	0xa8, 0x28, 0xe0, 0x60, 0x88, 0x08, 0xc6, 0x46,
+	0xb7, 0x37, 0xf0, 0x70, 0x98, 0x18, 0xd4, 0x54,
+	0xaa, 0x2a, 0xe2, 0x62, 0x8a, 0x0a, 0xc8, 0x48,
+	0xb9, 0x39, 0xf4, 0x74, 0x9a, 0x1a, 0xd6, 0x56,
+	0xa2, 0x22, 0xdd, 0x5d, 0x82, 0x02, 0xc0, 0x40,
+	0xb1, 0x31, 0xea, 0x6a, 0x92, 0x12, 0xcf, 0x4f,
+	0xae, 0x2e, 0xe6, 0x66, 0x8e, 0x0e, 0xcc, 0x4c,
+	0xbd, 0x3d, 0xfc, 0x7c, 0x9e, 0x1e, 0xda, 0x5a,
+	0xa6, 0x26, 0xdf, 0x5f, 0x86, 0x06, 0xc4, 0x44,
+	0xb5, 0x35, 0xee, 0x6e, 0x96, 0x16, 0xd2, 0x52
+};
+
+/* ulaw -> alaw */
+static u8 ulaw_to_alaw[256] =
+{
+	0xab, 0x55, 0xd5, 0x15, 0x95, 0x75, 0xf5, 0x35,
+	0xb5, 0x45, 0xc5, 0x05, 0x85, 0x65, 0xe5, 0x25,
+	0xa5, 0x5d, 0xdd, 0x1d, 0x9d, 0x7d, 0xfd, 0x3d,
+	0xbd, 0x4d, 0xcd, 0x0d, 0x8d, 0x6d, 0xed, 0x2d,
+	0xad, 0x51, 0xd1, 0x11, 0x91, 0x71, 0xf1, 0x31,
+	0xb1, 0x41, 0xc1, 0x01, 0x81, 0x61, 0xe1, 0x21,
+	0x59, 0xd9, 0x19, 0x99, 0x79, 0xf9, 0x39, 0xb9,
+	0x49, 0xc9, 0x09, 0x89, 0x69, 0xe9, 0x29, 0xa9,
+	0xd7, 0x17, 0x97, 0x77, 0xf7, 0x37, 0xb7, 0x47,
+	0xc7, 0x07, 0x87, 0x67, 0xe7, 0x27, 0xa7, 0xdf,
+	0x9f, 0x7f, 0xff, 0x3f, 0xbf, 0x4f, 0xcf, 0x0f,
+	0x8f, 0x6f, 0xef, 0x2f, 0x53, 0x13, 0x73, 0x33,
+	0xb3, 0x43, 0xc3, 0x03, 0x83, 0x63, 0xe3, 0x23,
+	0xa3, 0x5b, 0xdb, 0x1b, 0x9b, 0x7b, 0xfb, 0x3b,
+	0xbb, 0xbb, 0x4b, 0x4b, 0xcb, 0xcb, 0x0b, 0x0b,
+	0x8b, 0x8b, 0x6b, 0x6b, 0xeb, 0xeb, 0x2b, 0x2b,
+	0xab, 0x54, 0xd4, 0x14, 0x94, 0x74, 0xf4, 0x34,
+	0xb4, 0x44, 0xc4, 0x04, 0x84, 0x64, 0xe4, 0x24,
+	0xa4, 0x5c, 0xdc, 0x1c, 0x9c, 0x7c, 0xfc, 0x3c,
+	0xbc, 0x4c, 0xcc, 0x0c, 0x8c, 0x6c, 0xec, 0x2c,
+	0xac, 0x50, 0xd0, 0x10, 0x90, 0x70, 0xf0, 0x30,
+	0xb0, 0x40, 0xc0, 0x00, 0x80, 0x60, 0xe0, 0x20,
+	0x58, 0xd8, 0x18, 0x98, 0x78, 0xf8, 0x38, 0xb8,
+	0x48, 0xc8, 0x08, 0x88, 0x68, 0xe8, 0x28, 0xa8,
+	0xd6, 0x16, 0x96, 0x76, 0xf6, 0x36, 0xb6, 0x46,
+	0xc6, 0x06, 0x86, 0x66, 0xe6, 0x26, 0xa6, 0xde,
+	0x9e, 0x7e, 0xfe, 0x3e, 0xbe, 0x4e, 0xce, 0x0e,
+	0x8e, 0x6e, 0xee, 0x2e, 0x52, 0x12, 0x72, 0x32,
+	0xb2, 0x42, 0xc2, 0x02, 0x82, 0x62, 0xe2, 0x22,
+	0xa2, 0x5a, 0xda, 0x1a, 0x9a, 0x7a, 0xfa, 0x3a,
+	0xba, 0xba, 0x4a, 0x4a, 0xca, 0xca, 0x0a, 0x0a,
+	0x8a, 0x8a, 0x6a, 0x6a, 0xea, 0xea, 0x2a, 0x2a
+};
+
+/* alaw -> 4bit compression */
+static u8 alaw_to_4bit[256] = {
+	0x0e, 0x01, 0x0a, 0x05, 0x0f, 0x00, 0x0c, 0x03,
+	0x0d, 0x02, 0x08, 0x07, 0x0f, 0x00, 0x0b, 0x04,
+	0x0e, 0x01, 0x0a, 0x05, 0x0f, 0x00, 0x0c, 0x03,
+	0x0d, 0x02, 0x09, 0x06, 0x0f, 0x00, 0x0b, 0x04,
+	0x0e, 0x01, 0x0a, 0x05, 0x0f, 0x00, 0x0c, 0x03,
+	0x0d, 0x02, 0x08, 0x07, 0x0f, 0x00, 0x0b, 0x04,
+	0x0e, 0x01, 0x0a, 0x05, 0x0f, 0x00, 0x0c, 0x03,
+	0x0d, 0x02, 0x09, 0x06, 0x0f, 0x00, 0x0b, 0x04,
+	0x0e, 0x01, 0x0a, 0x05, 0x0f, 0x00, 0x0c, 0x03,
+	0x0d, 0x02, 0x08, 0x07, 0x0f, 0x00, 0x0b, 0x04,
+	0x0e, 0x01, 0x0a, 0x05, 0x0f, 0x00, 0x0d, 0x02,
+	0x0e, 0x02, 0x09, 0x06, 0x0f, 0x00, 0x0b, 0x04,
+	0x0e, 0x01, 0x0a, 0x05, 0x0f, 0x00, 0x0c, 0x03,
+	0x0d, 0x02, 0x08, 0x07, 0x0f, 0x00, 0x0b, 0x04,
+	0x0e, 0x01, 0x0a, 0x05, 0x0f, 0x00, 0x0c, 0x03,
+	0x0d, 0x02, 0x09, 0x06, 0x0f, 0x00, 0x0b, 0x04,
+	0x0e, 0x01, 0x0a, 0x05, 0x0f, 0x00, 0x0c, 0x03,
+	0x0d, 0x02, 0x08, 0x07, 0x0f, 0x00, 0x0b, 0x04,
+	0x0e, 0x01, 0x0a, 0x05, 0x0f, 0x00, 0x0c, 0x03,
+	0x0d, 0x02, 0x09, 0x06, 0x0f, 0x00, 0x0b, 0x04,
+	0x0e, 0x02, 0x09, 0x06, 0x0f, 0x00, 0x0b, 0x04,
+	0x0d, 0x02, 0x08, 0x07, 0x0f, 0x01, 0x0a, 0x05,
+	0x0e, 0x01, 0x0a, 0x05, 0x0f, 0x00, 0x0c, 0x03,
+	0x0d, 0x02, 0x09, 0x07, 0x0f, 0x00, 0x0b, 0x04,
+	0x0e, 0x01, 0x0a, 0x05, 0x0f, 0x00, 0x0c, 0x03,
+	0x0d, 0x02, 0x08, 0x07, 0x0f, 0x00, 0x0b, 0x04,
+	0x0e, 0x01, 0x0a, 0x05, 0x0f, 0x00, 0x0c, 0x03,
+	0x0d, 0x02, 0x09, 0x06, 0x0f, 0x00, 0x0b, 0x04,
+	0x0e, 0x01, 0x0a, 0x05, 0x0f, 0x00, 0x0c, 0x03,
+	0x0d, 0x02, 0x08, 0x07, 0x0f, 0x00, 0x0b, 0x04,
+	0x0e, 0x01, 0x0a, 0x05, 0x0f, 0x00, 0x0c, 0x03,
+	0x0d, 0x02, 0x09, 0x06, 0x0f, 0x00, 0x0b, 0x04,
+};
+
+/* 4bit -> alaw decompression */
+static u8 _4bit_to_alaw[16] = {
+	0x5d, 0x51, 0xd9, 0xd7, 0x5f, 0x53, 0xa3, 0x4b,
+	0x2a, 0x3a, 0x22, 0x2e, 0x26, 0x56, 0x20, 0x2c,
+};
+
+/* ulaw -> 4bit compression */
+static u8 ulaw_to_4bit[256] = {
+	0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+	0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+	0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+	0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
+	0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+	0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01, 0x01,
+	0x02, 0x02, 0x02, 0x02, 0x02, 0x02, 0x02, 0x02,
+	0x02, 0x02, 0x02, 0x02, 0x02, 0x02, 0x02, 0x02,
+	0x03, 0x03, 0x03, 0x03, 0x03, 0x03, 0x03, 0x03,
+	0x03, 0x03, 0x03, 0x03, 0x03, 0x03, 0x04, 0x04,
+	0x04, 0x04, 0x04, 0x04, 0x04, 0x04, 0x04, 0x04,
+	0x04, 0x04, 0x04, 0x04, 0x05, 0x05, 0x05, 0x05,
+	0x05, 0x05, 0x05, 0x05, 0x05, 0x05, 0x05, 0x05,
+	0x06, 0x06, 0x06, 0x06, 0x06, 0x06, 0x06, 0x06,
+	0x07, 0x07, 0x07, 0x07, 0x07, 0x07, 0x07, 0x07,
+	0x07, 0x07, 0x07, 0x07, 0x07, 0x07, 0x07, 0x08,
+	0x0f, 0x0f, 0x0f, 0x0f, 0x0f, 0x0f, 0x0f, 0x0f,
+	0x0f, 0x0f, 0x0f, 0x0f, 0x0f, 0x0f, 0x0f, 0x0f,
+	0x0f, 0x0f, 0x0f, 0x0f, 0x0f, 0x0f, 0x0f, 0x0f,
+	0x0f, 0x0f, 0x0f, 0x0f, 0x0f, 0x0f, 0x0f, 0x0f,
+	0x0f, 0x0e, 0x0e, 0x0e, 0x0e, 0x0e, 0x0e, 0x0e,
+	0x0e, 0x0e, 0x0e, 0x0e, 0x0e, 0x0e, 0x0e, 0x0e,
+	0x0e, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d,
+	0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d, 0x0d,
+	0x0c, 0x0c, 0x0c, 0x0c, 0x0c, 0x0c, 0x0c, 0x0c,
+	0x0c, 0x0c, 0x0c, 0x0c, 0x0c, 0x0c, 0x0b, 0x0b,
+	0x0b, 0x0b, 0x0b, 0x0b, 0x0b, 0x0b, 0x0b, 0x0b,
+	0x0b, 0x0b, 0x0b, 0x0b, 0x0a, 0x0a, 0x0a, 0x0a,
+	0x0a, 0x0a, 0x0a, 0x0a, 0x0a, 0x0a, 0x0a, 0x0a,
+	0x09, 0x09, 0x09, 0x09, 0x09, 0x09, 0x09, 0x09,
+	0x09, 0x08, 0x08, 0x08, 0x08, 0x08, 0x08, 0x08,
+	0x08, 0x08, 0x08, 0x08, 0x08, 0x08, 0x08, 0x08,
+};
+
+/* 4bit -> ulaw decompression */
+static u8 _4bit_to_ulaw[16] = {
+	0x11, 0x21, 0x31, 0x40, 0x4e, 0x5c, 0x68, 0x71,
+	0xfe, 0xef, 0xe7, 0xdb, 0xcd, 0xbf, 0xaf, 0x9f,
+};
+
+
+/*
+ * Compresses data to the result buffer
+ * The result size must be at least half of the input buffer.
+ * The number of samples also must be even!
+ */
+int
+l1oip_law_to_4bit(u8 *data, int len, u8 *result, u32 *state)
+{
+	int ii, i = 0, o = 0;
+
+	if (!len)
+		return 0;
+
+	/* send saved byte and first input byte */
+	if (*state) {
+		*result++ = table_com[(((*state)<<8)&0xff00) | (*data++)];
+		len--;
+		o++;
+	}
+
+	ii = len >> 1;
+
+	while (i < ii) {
+		*result++ = table_com[(data[0]<<8) | (data[1])];
+		data += 2;
+		i++;
+		o++;
+	}
+
+	/* if len has an odd number, we save byte for next call */
+	if (len & 1)
+		*state = 0x100 + *data;
+	else
+		*state = 0;
+
+	return o;
+}
+
+/* Decompress data to the result buffer
+ * The result size must be the number of sample in packet. (2 * input data)
+ * The number of samples in the result are even!
+ */
+int
+l1oip_4bit_to_law(u8 *data, int len, u8 *result)
+{
+	int i = 0;
+	u16 r;
+
+	while (i < len) {
+		r = table_dec[*data++];
+		*result++ = r>>8;
+		*result++ = r;
+		i++;
+	}
+
+	return len << 1;
+}
+
+
+/*
+ * law conversion
+ */
+int
+l1oip_alaw_to_ulaw(u8 *data, int len, u8 *result)
+{
+	int i = 0;
+
+	while (i < len) {
+		*result++ = alaw_to_ulaw[*data++];
+		i++;
+	}
+
+	return len;
+}
+
+int
+l1oip_ulaw_to_alaw(u8 *data, int len, u8 *result)
+{
+	int i = 0;
+
+	while (i < len) {
+		*result++ = ulaw_to_alaw[*data++];
+		i++;
+	}
+
+	return len;
+}
+
+
+/*
+ * generate/free compression and decompression table
+ */
+void
+l1oip_4bit_free(void)
+{
+	if (table_dec)
+		vfree(table_dec);
+	if (table_com)
+		vfree(table_com);
+	table_com = NULL;
+	table_dec = NULL;
+}
+
+int
+l1oip_4bit_alloc(int ulaw)
+{
+	int i1, i2, c, sample;
+
+	/* in case, it is called again */
+	if (table_dec)
+		return 0;
+
+	/* alloc conversion tables */
+	table_com = vmalloc(65536);
+	table_dec = vmalloc(512);
+	if (!table_com | !table_dec) {
+		l1oip_4bit_free();
+		return -ENOMEM;
+	}
+	memset(table_com, 0, 65536);
+	memset(table_dec, 0, 512);
+	/* generate compression table */
+	i1 = 0;
+	while (i1 < 256) {
+		if (ulaw)
+			c = ulaw_to_4bit[i1];
+		else
+			c = alaw_to_4bit[i1];
+		i2 = 0;
+		while (i2 < 256) {
+			table_com[(i1<<8) | i2] |= (c<<4);
+			table_com[(i2<<8) | i1] |= c;
+			i2++;
+		}
+		i1++;
+	}
+
+	/* generate decompression table */
+	i1 = 0;
+	while (i1 < 16) {
+		if (ulaw)
+			sample = _4bit_to_ulaw[i1];
+		else
+			sample = _4bit_to_alaw[i1];
+		i2 = 0;
+		while (i2 < 16) {
+			table_dec[(i1<<4) | i2] |= (sample<<8);
+			table_dec[(i2<<4) | i1] |= sample;
+			i2++;
+		}
+		i1++;
+	}
+
+	return 0;
+}
+
+
diff --git a/drivers/isdn/mISDN/l1oip_core.c b/drivers/isdn/mISDN/l1oip_core.c
new file mode 100644
index 000000000000..155b99780c4f
--- /dev/null
+++ b/drivers/isdn/mISDN/l1oip_core.c
@@ -0,0 +1,1518 @@
+/*
+
+ * l1oip.c  low level driver for tunneling layer 1 over IP
+ *
+ * NOTE: It is not compatible with TDMoIP nor "ISDN over IP".
+ *
+ * Author	Andreas Eversberg (jolly@eversberg.eu)
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2, or (at your option)
+ * any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ *
+ */
+
+/* module parameters:
+ * type:
+	Value 1	= BRI
+	Value 2	= PRI
+	Value 3 = BRI (multi channel frame, not supported yet)
+	Value 4 = PRI (multi channel frame, not supported yet)
+	A multi channel frame reduces overhead to a single frame for all
+	b-channels, but increases delay.
+	(NOTE: Multi channel frames are not implemented yet.)
+
+ * codec:
+	Value 0 = transparent (default)
+	Value 1 = transfer ALAW
+	Value 2 = transfer ULAW
+	Value 3 = transfer generic 4 bit compression.
+
+ * ulaw:
+	0 = we use a-Law (default)
+	1 = we use u-Law
+
+ * limit:
+	limitation of B-channels to control bandwidth (1...126)
+	BRI: 1 or 2
+	PRI: 1-30, 31-126 (126, because dchannel ist not counted here)
+	Also limited ressources are used for stack, resulting in less channels.
+	It is possible to have more channels than 30 in PRI mode, this must
+	be supported by the application.
+
+ * ip:
+	byte representation of remote ip address (127.0.0.1 -> 127,0,0,1)
+	If not given or four 0, no remote address is set.
+	For multiple interfaces, concat ip addresses. (127,0,0,1,127,0,0,1)
+
+ * port:
+	port number (local interface)
+	If not given or 0, port 931 is used for fist instance, 932 for next...
+	For multiple interfaces, different ports must be given.
+
+ * remoteport:
+	port number (remote interface)
+	If not given or 0, remote port equals local port
+	For multiple interfaces on equal sites, different ports must be given.
+
+ * ondemand:
+	0 = fixed (always transmit packets, even when remote side timed out)
+	1 = on demand (only transmit packets, when remote side is detected)
+	the default is 0
+	NOTE: ID must also be set for on demand.
+
+ * id:
+	optional value to identify frames. This value must be equal on both
+	peers and should be random. If omitted or 0, no ID is transmitted.
+
+ * debug:
+	NOTE: only one debug value must be given for all cards
+	enable debugging (see l1oip.h for debug options)
+
+
+Special mISDN controls:
+
+ op = MISDN_CTRL_SETPEER*
+ p1 = bytes 0-3 : remote IP address in network order (left element first)
+ p2 = bytes 1-2 : remote port in network order (high byte first)
+ optional:
+ p2 = bytes 3-4 : local port in network order (high byte first)
+
+ op = MISDN_CTRL_UNSETPEER*
+
+ * Use l1oipctrl for comfortable setting or removing ip address.
+   (Layer 1 Over IP CTRL)
+
+
+L1oIP-Protocol
+--------------
+
+Frame Header:
+
+ 7 6 5 4 3 2 1 0
++---------------+
+|Ver|T|I|Coding |
++---------------+
+|  ID byte 3 *  |
++---------------+
+|  ID byte 2 *  |
++---------------+
+|  ID byte 1 *  |
++---------------+
+|  ID byte 0 *  |
++---------------+
+|M|   Channel   |
++---------------+
+|    Length *   |
++---------------+
+| Time Base MSB |
++---------------+
+| Time Base LSB |
++---------------+
+| Data....	|
+
+...
+
+|               |
++---------------+
+|M|   Channel   |
++---------------+
+|    Length *   |
++---------------+
+| Time Base MSB |
++---------------+
+| Time Base LSB |
++---------------+
+| Data....	|
+
+...
+
+
+* Only included in some cases.
+
+- Ver = Version
+If version is missmatch, the frame must be ignored.
+
+- T = Type of interface
+Must be 0 for S0 or 1 for E1.
+
+- I = Id present
+If bit is set, four ID bytes are included in frame.
+
+- ID = Connection ID
+Additional ID to prevent Denial of Service attacs. Also it prevents hijacking
+connections with dynamic IP. The ID should be random and must not be 0.
+
+- Coding = Type of codec
+Must be 0 for no transcoding. Also for D-channel and other HDLC frames.
+ 1 and 2 are reserved for explicitly use of a-LAW or u-LAW codec.
+ 3 is used for generic table compressor.
+
+- M = More channels to come. If this flag is 1, the following byte contains
+the length of the channel data. After the data block, the next channel will
+be defined. The flag for the last channel block (or if only one channel is
+transmitted), must be 0 and no length is given.
+
+- Channel = Channel number
+0 reserved
+1-3 channel data for S0 (3 is D-channel)
+1-31 channel data for E1 (16 is D-channel)
+32-127 channel data for extended E1 (16 is D-channel)
+
+- The length is used if the M-flag is 1. It is used to find the next channel
+inside frame.
+NOTE: A value of 0 equals 256 bytes of data.
+ -> For larger data blocks, a single frame must be used.
+ -> For larger streams, a single frame or multiple blocks with same channel ID
+   must be used.
+
+- Time Base = Timestamp of first sample in frame
+The "Time Base" is used to rearange packets and to detect packet loss.
+The 16 bits are sent in network order (MSB first) and count 1/8000 th of a
+second. This causes a wrap arround each 8,192 seconds. There is no requirement
+for the initial "Time Base", but 0 should be used for the first packet.
+In case of HDLC data, this timestamp counts the packet or byte number.
+
+
+Two Timers:
+
+After initialisation, a timer of 15 seconds is started. Whenever a packet is
+transmitted, the timer is reset to 15 seconds again. If the timer expires, an
+empty packet is transmitted. This keep the connection alive.
+
+When a valid packet is received, a timer 65 seconds is started. The interface
+become ACTIVE. If the timer expires, the interface becomes INACTIVE.
+
+
+Dynamic IP handling:
+
+To allow dynamic IP, the ID must be non 0. In this case, any packet with the
+correct port number and ID will be accepted. If the remote side changes its IP
+the new IP is used for all transmitted packets until it changes again.
+
+
+On Demand:
+
+If the ondemand parameter is given, the remote IP is set to 0 on timeout.
+This will stop keepalive traffic to remote. If the remote is online again,
+traffic will continue to the remote address. This is usefull for road warriors.
+This feature only works with ID set, otherwhise it is highly unsecure.
+
+
+Socket and Thread
+-----------------
+
+The complete socket opening and closing is done by a thread.
+When the thread opened a socket, the hc->socket descriptor is set. Whenever a
+packet shall be sent to the socket, the hc->socket must be checked wheter not
+NULL. To prevent change in socket descriptor, the hc->socket_lock must be used.
+To change the socket, a recall of l1oip_socket_open() will safely kill the
+socket process and create a new one.
+
+*/
+
+#define L1OIP_VERSION	0	/* 0...3 */
+
+#include <linux/module.h>
+#include <linux/delay.h>
+#include <linux/mISDNif.h>
+#include <linux/mISDNhw.h>
+#include <linux/mISDNdsp.h>
+#include <linux/init.h>
+#include <linux/in.h>
+#include <linux/inet.h>
+#include <linux/workqueue.h>
+#include <linux/kthread.h>
+#include <net/sock.h>
+#include "core.h"
+#include "l1oip.h"
+
+static const char *l1oip_revision = "2.00";
+
+static int l1oip_cnt;
+static spinlock_t l1oip_lock;
+static struct list_head l1oip_ilist;
+
+#define MAX_CARDS	16
+static u_int type[MAX_CARDS];
+static u_int codec[MAX_CARDS];
+static u_int ip[MAX_CARDS*4];
+static u_int port[MAX_CARDS];
+static u_int remoteport[MAX_CARDS];
+static u_int ondemand[MAX_CARDS];
+static u_int limit[MAX_CARDS];
+static u_int id[MAX_CARDS];
+static int debug;
+static int ulaw;
+
+MODULE_AUTHOR("Andreas Eversberg");
+MODULE_LICENSE("GPL");
+module_param_array(type, uint, NULL, S_IRUGO | S_IWUSR);
+module_param_array(codec, uint, NULL, S_IRUGO | S_IWUSR);
+module_param_array(ip, uint, NULL, S_IRUGO | S_IWUSR);
+module_param_array(port, uint, NULL, S_IRUGO | S_IWUSR);
+module_param_array(remoteport, uint, NULL, S_IRUGO | S_IWUSR);
+module_param_array(ondemand, uint, NULL, S_IRUGO | S_IWUSR);
+module_param_array(limit, uint, NULL, S_IRUGO | S_IWUSR);
+module_param_array(id, uint, NULL, S_IRUGO | S_IWUSR);
+module_param(ulaw, uint, S_IRUGO | S_IWUSR);
+module_param(debug, uint, S_IRUGO | S_IWUSR);
+
+/*
+ * send a frame via socket, if open and restart timer
+ */
+static int
+l1oip_socket_send(struct l1oip *hc, u8 localcodec, u8 channel, u32 chanmask,
+	u16 timebase, u8 *buf, int len)
+{
+	u8 *p;
+	int multi = 0;
+	u8 frame[len+32];
+	struct socket *socket = NULL;
+	mm_segment_t oldfs;
+
+	if (debug & DEBUG_L1OIP_MSG)
+		printk(KERN_DEBUG "%s: sending data to socket (len = %d)\n",
+			__func__, len);
+
+	p = frame;
+
+	/* restart timer */
+	if ((int)(hc->keep_tl.expires-jiffies) < 5*HZ) {
+		del_timer(&hc->keep_tl);
+		hc->keep_tl.expires = jiffies + L1OIP_KEEPALIVE*HZ;
+		add_timer(&hc->keep_tl);
+	} else
+		hc->keep_tl.expires = jiffies + L1OIP_KEEPALIVE*HZ;
+
+	if (debug & DEBUG_L1OIP_MSG)
+		printk(KERN_DEBUG "%s: resetting timer\n", __func__);
+
+	/* drop if we have no remote ip or port */
+	if (!hc->sin_remote.sin_addr.s_addr || !hc->sin_remote.sin_port) {
+		if (debug & DEBUG_L1OIP_MSG)
+			printk(KERN_DEBUG "%s: dropping frame, because remote "
+				"IP is not set.\n", __func__);
+		return len;
+	}
+
+	/* assemble frame */
+	*p++ = (L1OIP_VERSION<<6) /* version and coding */
+	     | (hc->pri?0x20:0x00) /* type */
+	     | (hc->id?0x10:0x00) /* id */
+	     | localcodec;
+	if (hc->id) {
+		*p++ = hc->id>>24; /* id */
+		*p++ = hc->id>>16;
+		*p++ = hc->id>>8;
+		*p++ = hc->id;
+	}
+	*p++ = (multi == 1)?0x80:0x00 + channel; /* m-flag, channel */
+	if (multi == 1)
+		*p++ = len; /* length */
+	*p++ = timebase>>8; /* time base */
+	*p++ = timebase;
+
+	if (buf && len) { /* add data to frame */
+		if (localcodec == 1 && ulaw)
+			l1oip_ulaw_to_alaw(buf, len, p);
+		else if (localcodec == 2 && !ulaw)
+			l1oip_alaw_to_ulaw(buf, len, p);
+		else if (localcodec == 3)
+			len = l1oip_law_to_4bit(buf, len, p,
+				&hc->chan[channel].codecstate);
+		else
+			memcpy(p, buf, len);
+	}
+	len += p - frame;
+
+	/* check for socket in safe condition */
+	spin_lock(&hc->socket_lock);
+	if (!hc->socket) {
+		spin_unlock(&hc->socket_lock);
+		return 0;
+	}
+	/* seize socket */
+	socket = hc->socket;
+	hc->socket = NULL;
+	spin_unlock(&hc->socket_lock);
+	/* send packet */
+	if (debug & DEBUG_L1OIP_MSG)
+		printk(KERN_DEBUG "%s: sending packet to socket (len "
+			"= %d)\n", __func__, len);
+	hc->sendiov.iov_base = frame;
+	hc->sendiov.iov_len  = len;
+	oldfs = get_fs();
+	set_fs(KERNEL_DS);
+	len = sock_sendmsg(socket, &hc->sendmsg, len);
+	set_fs(oldfs);
+	/* give socket back */
+	hc->socket = socket; /* no locking required */
+
+	return len;
+}
+
+
+/*
+ * receive channel data from socket
+ */
+static void
+l1oip_socket_recv(struct l1oip *hc, u8 remotecodec, u8 channel, u16 timebase,
+	u8 *buf, int len)
+{
+	struct sk_buff *nskb;
+	struct bchannel *bch;
+	struct dchannel *dch;
+	u8 *p;
+	u32 rx_counter;
+
+	if (len == 0) {
+		if (debug & DEBUG_L1OIP_MSG)
+			printk(KERN_DEBUG "%s: received empty keepalive data, "
+				"ignoring\n", __func__);
+		return;
+	}
+
+	if (debug & DEBUG_L1OIP_MSG)
+		printk(KERN_DEBUG "%s: received data, sending to mISDN (%d)\n",
+			__func__, len);
+
+	if (channel < 1 || channel > 127) {
+		printk(KERN_WARNING "%s: packet error - channel %d out of "
+			"range\n", __func__, channel);
+		return;
+	}
+	dch = hc->chan[channel].dch;
+	bch = hc->chan[channel].bch;
+	if (!dch && !bch) {
+		printk(KERN_WARNING "%s: packet error - channel %d not in "
+			"stack\n", __func__, channel);
+		return;
+	}
+
+	/* prepare message */
+	nskb = mI_alloc_skb((remotecodec == 3)?(len<<1):len, GFP_ATOMIC);
+	if (!nskb) {
+		printk(KERN_ERR "%s: No mem for skb.\n", __func__);
+		return;
+	}
+	p = skb_put(nskb, (remotecodec == 3)?(len<<1):len);
+
+	if (remotecodec == 1 && ulaw)
+		l1oip_alaw_to_ulaw(buf, len, p);
+	else if (remotecodec == 2 && !ulaw)
+		l1oip_ulaw_to_alaw(buf, len, p);
+	else if (remotecodec == 3)
+		len = l1oip_4bit_to_law(buf, len, p);
+	else
+		memcpy(p, buf, len);
+
+	/* send message up */
+	if (dch && len >= 2) {
+		dch->rx_skb = nskb;
+		recv_Dchannel(dch);
+	}
+	if (bch) {
+		/* expand 16 bit sequence number to 32 bit sequence number */
+		rx_counter = hc->chan[channel].rx_counter;
+		if (((s16)(timebase - rx_counter)) >= 0) {
+			/* time has changed forward */
+			if (timebase >= (rx_counter & 0xffff))
+				rx_counter =
+					(rx_counter & 0xffff0000) | timebase;
+			else
+				rx_counter = ((rx_counter & 0xffff0000)+0x10000)
+					| timebase;
+		} else {
+			/* time has changed backwards */
+			if (timebase < (rx_counter & 0xffff))
+				rx_counter =
+					(rx_counter & 0xffff0000) | timebase;
+			else
+				rx_counter = ((rx_counter & 0xffff0000)-0x10000)
+					| timebase;
+		}
+		hc->chan[channel].rx_counter = rx_counter;
+
+#ifdef REORDER_DEBUG
+		if (hc->chan[channel].disorder_flag) {
+			struct sk_buff *skb;
+			int cnt;
+			skb = hc->chan[channel].disorder_skb;
+			hc->chan[channel].disorder_skb = nskb;
+			nskb = skb;
+			cnt = hc->chan[channel].disorder_cnt;
+			hc->chan[channel].disorder_cnt = rx_counter;
+			rx_counter = cnt;
+		}
+		hc->chan[channel].disorder_flag ^= 1;
+		if (nskb)
+#endif
+		queue_ch_frame(&bch->ch, PH_DATA_IND, rx_counter, nskb);
+	}
+}
+
+
+/*
+ * parse frame and extract channel data
+ */
+static void
+l1oip_socket_parse(struct l1oip *hc, struct sockaddr_in *sin, u8 *buf, int len)
+{
+	u32			id;
+	u8			channel;
+	u8			remotecodec;
+	u16			timebase;
+	int			m, mlen;
+	int			len_start = len; /* initial frame length */
+	struct dchannel		*dch = hc->chan[hc->d_idx].dch;
+
+	if (debug & DEBUG_L1OIP_MSG)
+		printk(KERN_DEBUG "%s: received frame, parsing... (%d)\n",
+			__func__, len);
+
+	/* check lenght */
+	if (len < 1+1+2) {
+		printk(KERN_WARNING "%s: packet error - length %d below "
+			"4 bytes\n", __func__, len);
+		return;
+	}
+
+	/* check version */
+	if (((*buf)>>6) != L1OIP_VERSION) {
+		printk(KERN_WARNING "%s: packet error - unknown version %d\n",
+			__func__, buf[0]>>6);
+		return;
+	}
+
+	/* check type */
+	if (((*buf)&0x20) && !hc->pri) {
+		printk(KERN_WARNING "%s: packet error - received E1 packet "
+			"on S0 interface\n", __func__);
+		return;
+	}
+	if (!((*buf)&0x20) && hc->pri) {
+		printk(KERN_WARNING "%s: packet error - received S0 packet "
+			"on E1 interface\n", __func__);
+		return;
+	}
+
+	/* get id flag */
+	id = (*buf>>4)&1;
+
+	/* check coding */
+	remotecodec = (*buf) & 0x0f;
+	if (remotecodec > 3) {
+		printk(KERN_WARNING "%s: packet error - remotecodec %d "
+			"unsupported\n", __func__, remotecodec);
+		return;
+	}
+	buf++;
+	len--;
+
+	/* check id */
+	if (id) {
+		if (!hc->id) {
+			printk(KERN_WARNING "%s: packet error - packet has id "
+				"0x%x, but we have not\n", __func__, id);
+			return;
+		}
+		if (len < 4) {
+			printk(KERN_WARNING "%s: packet error - packet too "
+				"short for ID value\n", __func__);
+			return;
+		}
+		id = (*buf++) << 24;
+		id += (*buf++) << 16;
+		id += (*buf++) << 8;
+		id += (*buf++);
+		len -= 4;
+
+		if (id != hc->id) {
+			printk(KERN_WARNING "%s: packet error - ID mismatch, "
+				"got 0x%x, we 0x%x\n",
+				__func__, id, hc->id);
+			return;
+		}
+	} else {
+		if (hc->id) {
+			printk(KERN_WARNING "%s: packet error - packet has no "
+				"ID, but we have\n", __func__);
+			return;
+		}
+	}
+
+multiframe:
+	if (len < 1) {
+		printk(KERN_WARNING "%s: packet error - packet too short, "
+			"channel expected at position %d.\n",
+			__func__, len-len_start+1);
+		return;
+	}
+
+	/* get channel and multiframe flag */
+	channel = *buf&0x7f;
+	m = *buf >> 7;
+	buf++;
+	len--;
+
+	/* check length on multiframe */
+	if (m) {
+		if (len < 1) {
+			printk(KERN_WARNING "%s: packet error - packet too "
+				"short, length expected at position %d.\n",
+				__func__, len_start-len-1);
+			return;
+		}
+
+		mlen = *buf++;
+		len--;
+		if (mlen == 0)
+			mlen = 256;
+		if (len < mlen+3) {
+			printk(KERN_WARNING "%s: packet error - length %d at "
+				"position %d exceeds total length %d.\n",
+				__func__, mlen, len_start-len-1, len_start);
+			return;
+		}
+		if (len == mlen+3) {
+			printk(KERN_WARNING "%s: packet error - length %d at "
+				"position %d will not allow additional "
+				"packet.\n",
+				__func__, mlen, len_start-len+1);
+			return;
+		}
+	} else
+		mlen = len-2; /* single frame, substract timebase */
+
+	if (len < 2) {
+		printk(KERN_WARNING "%s: packet error - packet too short, time "
+			"base expected at position %d.\n",
+			__func__, len-len_start+1);
+		return;
+	}
+
+	/* get time base */
+	timebase = (*buf++) << 8;
+	timebase |= (*buf++);
+	len -= 2;
+
+	/* if inactive, we send up a PH_ACTIVATE and activate */
+	if (!test_bit(FLG_ACTIVE, &dch->Flags)) {
+		if (debug & (DEBUG_L1OIP_MSG|DEBUG_L1OIP_SOCKET))
+			printk(KERN_DEBUG "%s: interface become active due to "
+				"received packet\n", __func__);
+		test_and_set_bit(FLG_ACTIVE, &dch->Flags);
+		_queue_data(&dch->dev.D, PH_ACTIVATE_IND, MISDN_ID_ANY, 0,
+			NULL, GFP_ATOMIC);
+	}
+
+	/* distribute packet */
+	l1oip_socket_recv(hc, remotecodec, channel, timebase, buf, mlen);
+	buf += mlen;
+	len -= mlen;
+
+	/* multiframe */
+	if (m)
+		goto multiframe;
+
+	/* restart timer */
+	if ((int)(hc->timeout_tl.expires-jiffies) < 5*HZ || !hc->timeout_on) {
+		hc->timeout_on = 1;
+		del_timer(&hc->timeout_tl);
+		hc->timeout_tl.expires = jiffies + L1OIP_TIMEOUT*HZ;
+		add_timer(&hc->timeout_tl);
+	} else /* only adjust timer */
+		hc->timeout_tl.expires = jiffies + L1OIP_TIMEOUT*HZ;
+
+	/* if ip or source port changes */
+	if ((hc->sin_remote.sin_addr.s_addr != sin->sin_addr.s_addr)
+	 || (hc->sin_remote.sin_port != sin->sin_port)) {
+		if (debug & DEBUG_L1OIP_SOCKET)
+			printk(KERN_DEBUG "%s: remote address changes from "
+				"0x%08x to 0x%08x (port %d to %d)\n", __func__,
+				ntohl(hc->sin_remote.sin_addr.s_addr),
+				ntohl(sin->sin_addr.s_addr),
+				ntohs(hc->sin_remote.sin_port),
+				ntohs(sin->sin_port));
+		hc->sin_remote.sin_addr.s_addr = sin->sin_addr.s_addr;
+		hc->sin_remote.sin_port = sin->sin_port;
+	}
+}
+
+
+/*
+ * socket stuff
+ */
+static int
+l1oip_socket_thread(void *data)
+{
+	struct l1oip *hc = (struct l1oip *)data;
+	int ret = 0;
+	struct msghdr msg;
+	struct iovec iov;
+	mm_segment_t oldfs;
+	struct sockaddr_in sin_rx;
+	unsigned char recvbuf[1500];
+	int recvlen;
+	struct socket *socket = NULL;
+	DECLARE_COMPLETION(wait);
+
+	/* make daemon */
+	allow_signal(SIGTERM);
+
+	/* create socket */
+	if (sock_create(PF_INET, SOCK_DGRAM, IPPROTO_UDP, &socket)) {
+		printk(KERN_ERR "%s: Failed to create socket.\n", __func__);
+		return -EIO;
+	}
+
+	/* set incoming address */
+	hc->sin_local.sin_family = AF_INET;
+	hc->sin_local.sin_addr.s_addr = INADDR_ANY;
+	hc->sin_local.sin_port = htons((unsigned short)hc->localport);
+
+	/* set outgoing address */
+	hc->sin_remote.sin_family = AF_INET;
+	hc->sin_remote.sin_addr.s_addr = htonl(hc->remoteip);
+	hc->sin_remote.sin_port = htons((unsigned short)hc->remoteport);
+
+	/* bind to incomming port */
+	if (socket->ops->bind(socket, (struct sockaddr *)&hc->sin_local,
+	    sizeof(hc->sin_local))) {
+		printk(KERN_ERR "%s: Failed to bind socket to port %d.\n",
+			__func__, hc->localport);
+		ret = -EINVAL;
+		goto fail;
+	}
+
+	/* check sk */
+	if (socket->sk == NULL) {
+		printk(KERN_ERR "%s: socket->sk == NULL\n", __func__);
+		ret = -EIO;
+		goto fail;
+	}
+
+	/* build receive message */
+	msg.msg_name = &sin_rx;
+	msg.msg_namelen = sizeof(sin_rx);
+	msg.msg_control = NULL;
+	msg.msg_controllen = 0;
+	msg.msg_iov = &iov;
+	msg.msg_iovlen = 1;
+
+	/* build send message */
+	hc->sendmsg.msg_name = &hc->sin_remote;
+	hc->sendmsg.msg_namelen = sizeof(hc->sin_remote);
+	hc->sendmsg.msg_control = NULL;
+	hc->sendmsg.msg_controllen = 0;
+	hc->sendmsg.msg_iov    = &hc->sendiov;
+	hc->sendmsg.msg_iovlen = 1;
+
+	/* give away socket */
+	spin_lock(&hc->socket_lock);
+	hc->socket = socket;
+	spin_unlock(&hc->socket_lock);
+
+	/* read loop */
+	if (debug & DEBUG_L1OIP_SOCKET)
+		printk(KERN_DEBUG "%s: socket created and open\n",
+			__func__);
+	while (!signal_pending(current)) {
+		iov.iov_base = recvbuf;
+		iov.iov_len = sizeof(recvbuf);
+		oldfs = get_fs();
+		set_fs(KERNEL_DS);
+		recvlen = sock_recvmsg(socket, &msg, sizeof(recvbuf), 0);
+		set_fs(oldfs);
+		if (recvlen > 0) {
+			l1oip_socket_parse(hc, &sin_rx, recvbuf, recvlen);
+		} else {
+			if (debug & DEBUG_L1OIP_SOCKET)
+			    printk(KERN_WARNING "%s: broken pipe on socket\n",
+				__func__);
+		}
+	}
+
+	/* get socket back, check first if in use, maybe by send function */
+	spin_lock(&hc->socket_lock);
+	/* if hc->socket is NULL, it is in use until it is given back */
+	while (!hc->socket) {
+		spin_unlock(&hc->socket_lock);
+		schedule_timeout(HZ/10);
+		spin_lock(&hc->socket_lock);
+	}
+	hc->socket = NULL;
+	spin_unlock(&hc->socket_lock);
+
+	if (debug & DEBUG_L1OIP_SOCKET)
+		printk(KERN_DEBUG "%s: socket thread terminating\n",
+			__func__);
+
+fail:
+	/* close socket */
+	if (socket)
+		sock_release(socket);
+
+	/* if we got killed, signal completion */
+	complete(&hc->socket_complete);
+	hc->socket_thread = NULL; /* show termination of thread */
+
+	if (debug & DEBUG_L1OIP_SOCKET)
+		printk(KERN_DEBUG "%s: socket thread terminated\n",
+			__func__);
+	return ret;
+}
+
+static void
+l1oip_socket_close(struct l1oip *hc)
+{
+	/* kill thread */
+	if (hc->socket_thread) {
+		if (debug & DEBUG_L1OIP_SOCKET)
+			printk(KERN_DEBUG "%s: socket thread exists, "
+				"killing...\n", __func__);
+		send_sig(SIGTERM, hc->socket_thread, 0);
+		wait_for_completion(&hc->socket_complete);
+	}
+}
+
+static int
+l1oip_socket_open(struct l1oip *hc)
+{
+	/* in case of reopen, we need to close first */
+	l1oip_socket_close(hc);
+
+	init_completion(&hc->socket_complete);
+
+	/* create receive process */
+	hc->socket_thread = kthread_run(l1oip_socket_thread, hc, "l1oip_%s",
+		hc->name);
+	if (IS_ERR(hc->socket_thread)) {
+		int err = PTR_ERR(hc->socket_thread);
+		printk(KERN_ERR "%s: Failed (%d) to create socket process.\n",
+			__func__, err);
+		hc->socket_thread = NULL;
+		sock_release(hc->socket);
+		return err;
+	}
+	if (debug & DEBUG_L1OIP_SOCKET)
+		printk(KERN_DEBUG "%s: socket thread created\n", __func__);
+
+	return 0;
+}
+
+
+static void
+l1oip_send_bh(struct work_struct *work)
+{
+	struct l1oip *hc = container_of(work, struct l1oip, workq);
+
+	if (debug & (DEBUG_L1OIP_MSG|DEBUG_L1OIP_SOCKET))
+		printk(KERN_DEBUG "%s: keepalive timer expired, sending empty "
+			"frame on dchannel\n", __func__);
+
+	/* send an empty l1oip frame at D-channel */
+	l1oip_socket_send(hc, 0, hc->d_idx, 0, 0, NULL, 0);
+}
+
+
+/*
+ * timer stuff
+ */
+static void
+l1oip_keepalive(void *data)
+{
+	struct l1oip *hc = (struct l1oip *)data;
+
+	schedule_work(&hc->workq);
+}
+
+static void
+l1oip_timeout(void *data)
+{
+	struct l1oip			*hc = (struct l1oip *)data;
+	struct dchannel		*dch = hc->chan[hc->d_idx].dch;
+
+	if (debug & DEBUG_L1OIP_MSG)
+		printk(KERN_DEBUG "%s: timeout timer expired, turn layer one "
+			"down.\n", __func__);
+
+	hc->timeout_on = 0; /* state that timer must be initialized next time */
+
+	/* if timeout, we send up a PH_DEACTIVATE and deactivate */
+	if (test_bit(FLG_ACTIVE, &dch->Flags)) {
+		if (debug & (DEBUG_L1OIP_MSG|DEBUG_L1OIP_SOCKET))
+			printk(KERN_DEBUG "%s: interface become deactivated "
+				"due to timeout\n", __func__);
+		test_and_clear_bit(FLG_ACTIVE, &dch->Flags);
+		_queue_data(&dch->dev.D, PH_DEACTIVATE_IND, MISDN_ID_ANY, 0,
+			NULL, GFP_ATOMIC);
+	}
+
+	/* if we have ondemand set, we remove ip address */
+	if (hc->ondemand) {
+		if (debug & DEBUG_L1OIP_MSG)
+			printk(KERN_DEBUG "%s: on demand causes ip address to "
+				"be removed\n", __func__);
+		hc->sin_remote.sin_addr.s_addr = 0;
+	}
+}
+
+
+/*
+ * message handling
+ */
+static int
+handle_dmsg(struct mISDNchannel *ch, struct sk_buff *skb)
+{
+	struct mISDNdevice	*dev = container_of(ch, struct mISDNdevice, D);
+	struct dchannel		*dch = container_of(dev, struct dchannel, dev);
+	struct l1oip			*hc = dch->hw;
+	struct mISDNhead	*hh = mISDN_HEAD_P(skb);
+	int			ret = -EINVAL;
+	int			l, ll;
+	unsigned char		*p;
+
+	switch (hh->prim) {
+	case PH_DATA_REQ:
+		if (skb->len < 1) {
+			printk(KERN_WARNING "%s: skb too small\n",
+				__func__);
+			break;
+		}
+		if (skb->len > MAX_DFRAME_LEN_L1 || skb->len > L1OIP_MAX_LEN) {
+			printk(KERN_WARNING "%s: skb too large\n",
+				__func__);
+			break;
+		}
+		/* send frame */
+		p = skb->data;
+		l = skb->len;
+		while (l) {
+			ll = (l < L1OIP_MAX_PERFRAME)?l:L1OIP_MAX_PERFRAME;
+			l1oip_socket_send(hc, 0, dch->slot, 0,
+				hc->chan[dch->slot].tx_counter++, p, ll);
+			p += ll;
+			l -= ll;
+		}
+		skb_trim(skb, 0);
+		queue_ch_frame(ch, PH_DATA_CNF, hh->id, skb);
+		return 0;
+	case PH_ACTIVATE_REQ:
+		if (debug & (DEBUG_L1OIP_MSG|DEBUG_L1OIP_SOCKET))
+			printk(KERN_DEBUG "%s: PH_ACTIVATE channel %d (1..%d)\n"
+				, __func__, dch->slot, hc->b_num+1);
+		skb_trim(skb, 0);
+		if (test_bit(FLG_ACTIVE, &dch->Flags))
+			queue_ch_frame(ch, PH_ACTIVATE_IND, hh->id, skb);
+		else
+			queue_ch_frame(ch, PH_DEACTIVATE_IND, hh->id, skb);
+		return 0;
+	case PH_DEACTIVATE_REQ:
+		if (debug & (DEBUG_L1OIP_MSG|DEBUG_L1OIP_SOCKET))
+			printk(KERN_DEBUG "%s: PH_DEACTIVATE channel %d "
+				"(1..%d)\n", __func__, dch->slot,
+				hc->b_num+1);
+		skb_trim(skb, 0);
+		if (test_bit(FLG_ACTIVE, &dch->Flags))
+			queue_ch_frame(ch, PH_ACTIVATE_IND, hh->id, skb);
+		else
+			queue_ch_frame(ch, PH_DEACTIVATE_IND, hh->id, skb);
+		return 0;
+	}
+	if (!ret)
+		dev_kfree_skb(skb);
+	return ret;
+}
+
+static int
+channel_dctrl(struct dchannel *dch, struct mISDN_ctrl_req *cq)
+{
+	int	ret = 0;
+	struct l1oip	*hc = dch->hw;
+
+	switch (cq->op) {
+	case MISDN_CTRL_GETOP:
+		cq->op = MISDN_CTRL_SETPEER | MISDN_CTRL_UNSETPEER;
+		break;
+	case MISDN_CTRL_SETPEER:
+		hc->remoteip = (u32)cq->p1;
+		hc->remoteport = cq->p2 & 0xffff;
+		hc->localport = cq->p2 >> 16;
+		if (!hc->remoteport)
+			hc->remoteport = hc->localport;
+		if (debug & DEBUG_L1OIP_SOCKET)
+			printk(KERN_DEBUG "%s: got new ip address from user "
+				"space.\n", __func__);
+			l1oip_socket_open(hc);
+		break;
+	case MISDN_CTRL_UNSETPEER:
+		if (debug & DEBUG_L1OIP_SOCKET)
+			printk(KERN_DEBUG "%s: removing ip address.\n",
+				__func__);
+		hc->remoteip = 0;
+		l1oip_socket_open(hc);
+		break;
+	default:
+		printk(KERN_WARNING "%s: unknown Op %x\n",
+		    __func__, cq->op);
+		ret = -EINVAL;
+		break;
+	}
+	return ret;
+}
+
+static int
+open_dchannel(struct l1oip *hc, struct dchannel *dch, struct channel_req *rq)
+{
+	if (debug & DEBUG_HW_OPEN)
+		printk(KERN_DEBUG "%s: dev(%d) open from %p\n", __func__,
+		    dch->dev.id, __builtin_return_address(0));
+	if (rq->protocol == ISDN_P_NONE)
+		return -EINVAL;
+	if ((dch->dev.D.protocol != ISDN_P_NONE) &&
+	    (dch->dev.D.protocol != rq->protocol)) {
+		if (debug & DEBUG_HW_OPEN)
+			printk(KERN_WARNING "%s: change protocol %x to %x\n",
+			__func__, dch->dev.D.protocol, rq->protocol);
+	}
+	if (dch->dev.D.protocol != rq->protocol)
+		dch->dev.D.protocol = rq->protocol;
+
+	if (test_bit(FLG_ACTIVE, &dch->Flags)) {
+		_queue_data(&dch->dev.D, PH_ACTIVATE_IND, MISDN_ID_ANY,
+		    0, NULL, GFP_KERNEL);
+	}
+	rq->ch = &dch->dev.D;
+	if (!try_module_get(THIS_MODULE))
+		printk(KERN_WARNING "%s:cannot get module\n", __func__);
+	return 0;
+}
+
+static int
+open_bchannel(struct l1oip *hc, struct dchannel *dch, struct channel_req *rq)
+{
+	struct bchannel	*bch;
+	int		ch;
+
+	if (!test_bit(rq->adr.channel & 0x1f,
+		&dch->dev.channelmap[rq->adr.channel >> 5]))
+		return -EINVAL;
+	if (rq->protocol == ISDN_P_NONE)
+		return -EINVAL;
+	ch = rq->adr.channel; /* BRI: 1=B1 2=B2  PRI: 1..15,17.. */
+	bch = hc->chan[ch].bch;
+	if (!bch) {
+		printk(KERN_ERR "%s:internal error ch %d has no bch\n",
+		    __func__, ch);
+		return -EINVAL;
+	}
+	if (test_and_set_bit(FLG_OPEN, &bch->Flags))
+		return -EBUSY; /* b-channel can be only open once */
+	bch->ch.protocol = rq->protocol;
+	rq->ch = &bch->ch;
+	if (!try_module_get(THIS_MODULE))
+		printk(KERN_WARNING "%s:cannot get module\n", __func__);
+	return 0;
+}
+
+static int
+l1oip_dctrl(struct mISDNchannel *ch, u_int cmd, void *arg)
+{
+	struct mISDNdevice	*dev = container_of(ch, struct mISDNdevice, D);
+	struct dchannel		*dch = container_of(dev, struct dchannel, dev);
+	struct l1oip			*hc = dch->hw;
+	struct channel_req	*rq;
+	int			err = 0;
+
+	if (dch->debug & DEBUG_HW)
+		printk(KERN_DEBUG "%s: cmd:%x %p\n",
+		    __func__, cmd, arg);
+	switch (cmd) {
+	case OPEN_CHANNEL:
+		rq = arg;
+		switch (rq->protocol) {
+		case ISDN_P_TE_S0:
+		case ISDN_P_NT_S0:
+			if (hc->pri) {
+				err = -EINVAL;
+				break;
+			}
+			err = open_dchannel(hc, dch, rq);
+			break;
+		case ISDN_P_TE_E1:
+		case ISDN_P_NT_E1:
+			if (!hc->pri) {
+				err = -EINVAL;
+				break;
+			}
+			err = open_dchannel(hc, dch, rq);
+			break;
+		default:
+			err = open_bchannel(hc, dch, rq);
+		}
+		break;
+	case CLOSE_CHANNEL:
+		if (debug & DEBUG_HW_OPEN)
+			printk(KERN_DEBUG "%s: dev(%d) close from %p\n",
+			    __func__, dch->dev.id,
+			    __builtin_return_address(0));
+		module_put(THIS_MODULE);
+		break;
+	case CONTROL_CHANNEL:
+		err = channel_dctrl(dch, arg);
+		break;
+	default:
+		if (dch->debug & DEBUG_HW)
+			printk(KERN_DEBUG "%s: unknown command %x\n",
+			    __func__, cmd);
+		err = -EINVAL;
+	}
+	return err;
+}
+
+static int
+handle_bmsg(struct mISDNchannel *ch, struct sk_buff *skb)
+{
+	struct bchannel		*bch = container_of(ch, struct bchannel, ch);
+	struct l1oip			*hc = bch->hw;
+	int			ret = -EINVAL;
+	struct mISDNhead	*hh = mISDN_HEAD_P(skb);
+	int			l, ll, i;
+	unsigned char		*p;
+
+	switch (hh->prim) {
+	case PH_DATA_REQ:
+		if (skb->len <= 0) {
+			printk(KERN_WARNING "%s: skb too small\n",
+				__func__);
+			break;
+		}
+		if (skb->len > MAX_DFRAME_LEN_L1 || skb->len > L1OIP_MAX_LEN) {
+			printk(KERN_WARNING "%s: skb too large\n",
+				__func__);
+			break;
+		}
+		/* check for AIS / ulaw-silence */
+		p = skb->data;
+		l = skb->len;
+		for (i = 0; i < l; i++) {
+			if (*p++ != 0xff)
+				break;
+		}
+		if (i == l) {
+			if (debug & DEBUG_L1OIP_MSG)
+				printk(KERN_DEBUG "%s: got AIS, not sending, "
+					"but counting\n", __func__);
+			hc->chan[bch->slot].tx_counter += l;
+			skb_trim(skb, 0);
+			queue_ch_frame(ch, PH_DATA_CNF, hh->id, skb);
+			return 0;
+		}
+		/* check for silence */
+		p = skb->data;
+		l = skb->len;
+		for (i = 0; i < l; i++) {
+			if (*p++ != 0x2a)
+				break;
+		}
+		if (i == l) {
+			if (debug & DEBUG_L1OIP_MSG)
+				printk(KERN_DEBUG "%s: got silence, not sending"
+					", but counting\n", __func__);
+			hc->chan[bch->slot].tx_counter += l;
+			skb_trim(skb, 0);
+			queue_ch_frame(ch, PH_DATA_CNF, hh->id, skb);
+			return 0;
+		}
+
+		/* send frame */
+		p = skb->data;
+		l = skb->len;
+		while (l) {
+			ll = (l < L1OIP_MAX_PERFRAME)?l:L1OIP_MAX_PERFRAME;
+			l1oip_socket_send(hc, hc->codec, bch->slot, 0,
+				hc->chan[bch->slot].tx_counter, p, ll);
+			hc->chan[bch->slot].tx_counter += ll;
+			p += ll;
+			l -= ll;
+		}
+		skb_trim(skb, 0);
+		queue_ch_frame(ch, PH_DATA_CNF, hh->id, skb);
+		return 0;
+	case PH_ACTIVATE_REQ:
+		if (debug & (DEBUG_L1OIP_MSG|DEBUG_L1OIP_SOCKET))
+			printk(KERN_DEBUG "%s: PH_ACTIVATE channel %d (1..%d)\n"
+				, __func__, bch->slot, hc->b_num+1);
+		hc->chan[bch->slot].codecstate = 0;
+		test_and_set_bit(FLG_ACTIVE, &bch->Flags);
+		skb_trim(skb, 0);
+		queue_ch_frame(ch, PH_ACTIVATE_IND, hh->id, skb);
+		return 0;
+	case PH_DEACTIVATE_REQ:
+		if (debug & (DEBUG_L1OIP_MSG|DEBUG_L1OIP_SOCKET))
+			printk(KERN_DEBUG "%s: PH_DEACTIVATE channel %d "
+				"(1..%d)\n", __func__, bch->slot,
+				hc->b_num+1);
+		test_and_clear_bit(FLG_ACTIVE, &bch->Flags);
+		skb_trim(skb, 0);
+		queue_ch_frame(ch, PH_DEACTIVATE_IND, hh->id, skb);
+		return 0;
+	}
+	if (!ret)
+		dev_kfree_skb(skb);
+	return ret;
+}
+
+static int
+channel_bctrl(struct bchannel *bch, struct mISDN_ctrl_req *cq)
+{
+	int			ret = 0;
+	struct dsp_features	*features =
+		(struct dsp_features *)(*((u_long *)&cq->p1));
+
+	switch (cq->op) {
+	case MISDN_CTRL_GETOP:
+		cq->op = MISDN_CTRL_HW_FEATURES_OP;
+		break;
+	case MISDN_CTRL_HW_FEATURES: /* fill features structure */
+		if (debug & DEBUG_L1OIP_MSG)
+			printk(KERN_DEBUG "%s: HW_FEATURE request\n",
+			    __func__);
+		/* create confirm */
+		features->unclocked = 1;
+		features->unordered = 1;
+		break;
+	default:
+		printk(KERN_WARNING "%s: unknown Op %x\n",
+		    __func__, cq->op);
+		ret = -EINVAL;
+		break;
+	}
+	return ret;
+}
+
+static int
+l1oip_bctrl(struct mISDNchannel *ch, u_int cmd, void *arg)
+{
+	struct bchannel	*bch = container_of(ch, struct bchannel, ch);
+	int		err = -EINVAL;
+
+	if (bch->debug & DEBUG_HW)
+		printk(KERN_DEBUG "%s: cmd:%x %p\n",
+		    __func__, cmd, arg);
+	switch (cmd) {
+	case CLOSE_CHANNEL:
+		test_and_clear_bit(FLG_OPEN, &bch->Flags);
+		test_and_clear_bit(FLG_ACTIVE, &bch->Flags);
+		ch->protocol = ISDN_P_NONE;
+		ch->peer = NULL;
+		module_put(THIS_MODULE);
+		err = 0;
+		break;
+	case CONTROL_CHANNEL:
+		err = channel_bctrl(bch, arg);
+		break;
+	default:
+		printk(KERN_WARNING "%s: unknown prim(%x)\n",
+			__func__, cmd);
+	}
+	return err;
+}
+
+
+/*
+ * cleanup module and stack
+ */
+static void
+release_card(struct l1oip *hc)
+{
+	int	ch;
+
+	if (timer_pending(&hc->keep_tl))
+		del_timer(&hc->keep_tl);
+
+	if (timer_pending(&hc->timeout_tl))
+		del_timer(&hc->timeout_tl);
+
+	if (hc->socket_thread)
+		l1oip_socket_close(hc);
+
+	if (hc->registered && hc->chan[hc->d_idx].dch)
+		mISDN_unregister_device(&hc->chan[hc->d_idx].dch->dev);
+	for (ch = 0; ch < 128; ch++) {
+		if (hc->chan[ch].dch) {
+			mISDN_freedchannel(hc->chan[ch].dch);
+			kfree(hc->chan[ch].dch);
+		}
+		if (hc->chan[ch].bch) {
+			mISDN_freebchannel(hc->chan[ch].bch);
+			kfree(hc->chan[ch].bch);
+#ifdef REORDER_DEBUG
+			if (hc->chan[ch].disorder_skb)
+				dev_kfree_skb(hc->chan[ch].disorder_skb);
+#endif
+		}
+	}
+
+	spin_lock(&l1oip_lock);
+	list_del(&hc->list);
+	spin_unlock(&l1oip_lock);
+
+	kfree(hc);
+}
+
+static void
+l1oip_cleanup(void)
+{
+	struct l1oip *hc, *next;
+
+	list_for_each_entry_safe(hc, next, &l1oip_ilist, list)
+		release_card(hc);
+
+	l1oip_4bit_free();
+}
+
+
+/*
+ * module and stack init
+ */
+static int
+init_card(struct l1oip *hc, int pri, int bundle)
+{
+	struct dchannel	*dch;
+	struct bchannel	*bch;
+	int		ret;
+	int		i, ch;
+
+	spin_lock_init(&hc->socket_lock);
+	hc->idx = l1oip_cnt;
+	hc->pri = pri;
+	hc->d_idx = pri?16:3;
+	hc->b_num = pri?30:2;
+	hc->bundle = bundle;
+	if (hc->pri)
+		sprintf(hc->name, "l1oip-e1.%d", l1oip_cnt + 1);
+	else
+		sprintf(hc->name, "l1oip-s0.%d", l1oip_cnt + 1);
+
+	switch (codec[l1oip_cnt]) {
+	case 0: /* as is */
+	case 1: /* alaw */
+	case 2: /* ulaw */
+	case 3: /* 4bit */
+		break;
+	default:
+		printk(KERN_ERR "Codec(%d) not supported.\n",
+			codec[l1oip_cnt]);
+		return -EINVAL;
+	}
+	hc->codec = codec[l1oip_cnt];
+	if (debug & DEBUG_L1OIP_INIT)
+		printk(KERN_DEBUG "%s: using codec %d\n",
+			__func__, hc->codec);
+
+	if (id[l1oip_cnt] == 0) {
+		printk(KERN_WARNING "Warning: No 'id' value given or "
+			"0, this is highly unsecure. Please use 32 "
+			"bit randmom number 0x...\n");
+	}
+	hc->id = id[l1oip_cnt];
+	if (debug & DEBUG_L1OIP_INIT)
+		printk(KERN_DEBUG "%s: using id 0x%x\n", __func__, hc->id);
+
+	hc->ondemand = ondemand[l1oip_cnt];
+	if (hc->ondemand && !hc->id) {
+		printk(KERN_ERR "%s: ondemand option only allowed in "
+			"conjunction with non 0 ID\n", __func__);
+		return -EINVAL;
+	}
+
+	if (limit[l1oip_cnt])
+		hc->b_num = limit[l1oip_cnt];
+	if (!pri && hc->b_num > 2) {
+		printk(KERN_ERR "Maximum limit for BRI interface is 2 "
+			"channels.\n");
+		return -EINVAL;
+	}
+	if (pri && hc->b_num > 126) {
+		printk(KERN_ERR "Maximum limit for PRI interface is 126 "
+			"channels.\n");
+		return -EINVAL;
+	}
+	if (pri && hc->b_num > 30) {
+		printk(KERN_WARNING "Maximum limit for BRI interface is 30 "
+			"channels.\n");
+		printk(KERN_WARNING "Your selection of %d channels must be "
+			"supported by application.\n", hc->limit);
+	}
+
+	hc->remoteip = ip[l1oip_cnt<<2] << 24
+		     | ip[(l1oip_cnt<<2)+1] << 16
+		     | ip[(l1oip_cnt<<2)+2] << 8
+		     | ip[(l1oip_cnt<<2)+3];
+	hc->localport = port[l1oip_cnt]?:(L1OIP_DEFAULTPORT+l1oip_cnt);
+	if (remoteport[l1oip_cnt])
+		hc->remoteport = remoteport[l1oip_cnt];
+	else
+		hc->remoteport = hc->localport;
+	if (debug & DEBUG_L1OIP_INIT)
+		printk(KERN_DEBUG "%s: using local port %d remote ip "
+			"%d.%d.%d.%d port %d ondemand %d\n", __func__,
+			hc->localport, hc->remoteip >> 24,
+			(hc->remoteip >> 16) & 0xff,
+			(hc->remoteip >> 8) & 0xff, hc->remoteip & 0xff,
+			hc->remoteport, hc->ondemand);
+
+	dch = kzalloc(sizeof(struct dchannel), GFP_KERNEL);
+	if (!dch)
+		return -ENOMEM;
+	dch->debug = debug;
+	mISDN_initdchannel(dch, MAX_DFRAME_LEN_L1, NULL);
+	dch->hw = hc;
+	if (pri)
+		dch->dev.Dprotocols = (1 << ISDN_P_TE_E1) | (1 << ISDN_P_NT_E1);
+	else
+		dch->dev.Dprotocols = (1 << ISDN_P_TE_S0) | (1 << ISDN_P_NT_S0);
+	dch->dev.Bprotocols = (1 << (ISDN_P_B_RAW & ISDN_P_B_MASK)) |
+	    (1 << (ISDN_P_B_HDLC & ISDN_P_B_MASK));
+	dch->dev.D.send = handle_dmsg;
+	dch->dev.D.ctrl = l1oip_dctrl;
+	dch->dev.nrbchan = hc->b_num;
+	dch->slot = hc->d_idx;
+	hc->chan[hc->d_idx].dch = dch;
+	i = 1;
+	for (ch = 0; ch < dch->dev.nrbchan; ch++) {
+		if (ch == 15)
+			i++;
+		bch = kzalloc(sizeof(struct bchannel), GFP_KERNEL);
+		if (!bch) {
+			printk(KERN_ERR "%s: no memory for bchannel\n",
+			    __func__);
+			return -ENOMEM;
+		}
+		bch->nr = i + ch;
+		bch->slot = i + ch;
+		bch->debug = debug;
+		mISDN_initbchannel(bch, MAX_DATA_MEM);
+		bch->hw = hc;
+		bch->ch.send = handle_bmsg;
+		bch->ch.ctrl = l1oip_bctrl;
+		bch->ch.nr = i + ch;
+		list_add(&bch->ch.list, &dch->dev.bchannels);
+		hc->chan[i + ch].bch = bch;
+		test_and_set_bit(bch->nr & 0x1f,
+			&dch->dev.channelmap[bch->nr >> 5]);
+	}
+	ret = mISDN_register_device(&dch->dev, hc->name);
+	if (ret)
+		return ret;
+	hc->registered = 1;
+
+	if (debug & DEBUG_L1OIP_INIT)
+		printk(KERN_DEBUG "%s: Setting up network card(%d)\n",
+			__func__, l1oip_cnt + 1);
+	ret = l1oip_socket_open(hc);
+	if (ret)
+		return ret;
+
+	hc->keep_tl.function = (void *)l1oip_keepalive;
+	hc->keep_tl.data = (ulong)hc;
+	init_timer(&hc->keep_tl);
+	hc->keep_tl.expires = jiffies + 2*HZ; /* two seconds first time */
+	add_timer(&hc->keep_tl);
+
+	hc->timeout_tl.function = (void *)l1oip_timeout;
+	hc->timeout_tl.data = (ulong)hc;
+	init_timer(&hc->timeout_tl);
+	hc->timeout_on = 0; /* state that we have timer off */
+
+	return 0;
+}
+
+static int __init
+l1oip_init(void)
+{
+	int		pri, bundle;
+	struct l1oip		*hc;
+	int		ret;
+
+	printk(KERN_INFO "mISDN: Layer-1-over-IP driver Rev. %s\n",
+		l1oip_revision);
+
+	INIT_LIST_HEAD(&l1oip_ilist);
+	spin_lock_init(&l1oip_lock);
+
+	if (l1oip_4bit_alloc(ulaw))
+		return -ENOMEM;
+
+	l1oip_cnt = 0;
+	while (type[l1oip_cnt] && l1oip_cnt < MAX_CARDS) {
+		switch (type[l1oip_cnt] & 0xff) {
+		case 1:
+			pri = 0;
+			bundle = 0;
+			break;
+		case 2:
+			pri = 1;
+			bundle = 0;
+			break;
+		case 3:
+			pri = 0;
+			bundle = 1;
+			break;
+		case 4:
+			pri = 1;
+			bundle = 1;
+			break;
+		default:
+			printk(KERN_ERR "Card type(%d) not supported.\n",
+				type[l1oip_cnt] & 0xff);
+			l1oip_cleanup();
+			return -EINVAL;
+		}
+
+		if (debug & DEBUG_L1OIP_INIT)
+			printk(KERN_DEBUG "%s: interface %d is %s with %s.\n",
+				__func__, l1oip_cnt, pri?"PRI":"BRI",
+				bundle?"bundled IP packet for all B-channels"
+				 :"seperate IP packets for every B-channel");
+
+		hc = kzalloc(sizeof(struct l1oip), GFP_ATOMIC);
+		if (!hc) {
+			printk(KERN_ERR "No kmem for L1-over-IP driver.\n");
+			l1oip_cleanup();
+			return -ENOMEM;
+		}
+		INIT_WORK(&hc->workq, (void *)l1oip_send_bh);
+
+		spin_lock(&l1oip_lock);
+		list_add_tail(&hc->list, &l1oip_ilist);
+		spin_unlock(&l1oip_lock);
+
+		ret = init_card(hc, pri, bundle);
+		if (ret) {
+			l1oip_cleanup();
+			return ret;
+		}
+
+		l1oip_cnt++;
+	}
+	printk(KERN_INFO "%d virtual devices registered\n", l1oip_cnt);
+	return 0;
+}
+
+module_init(l1oip_init);
+module_exit(l1oip_cleanup);
+

From e40f51a36a6ca718e829c0933ab1e79333ac932e Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Sat, 26 Jul 2008 17:47:53 -0700
Subject: [PATCH 1119/1171] netfilter: ebtables: use nf_register_hooks()

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/bridge/netfilter/ebtable_filter.c | 18 +++++-------------
 net/bridge/netfilter/ebtable_nat.c    | 18 +++++-------------
 2 files changed, 10 insertions(+), 26 deletions(-)

diff --git a/net/bridge/netfilter/ebtable_filter.c b/net/bridge/netfilter/ebtable_filter.c
index 690bc3ab186c..1a58af51a2e2 100644
--- a/net/bridge/netfilter/ebtable_filter.c
+++ b/net/bridge/netfilter/ebtable_filter.c
@@ -93,28 +93,20 @@ static struct nf_hook_ops ebt_ops_filter[] __read_mostly = {
 
 static int __init ebtable_filter_init(void)
 {
-	int i, j, ret;
+	int ret;
 
 	ret = ebt_register_table(&frame_filter);
 	if (ret < 0)
 		return ret;
-	for (i = 0; i < ARRAY_SIZE(ebt_ops_filter); i++)
-		if ((ret = nf_register_hook(&ebt_ops_filter[i])) < 0)
-			goto cleanup;
-	return ret;
-cleanup:
-	for (j = 0; j < i; j++)
-		nf_unregister_hook(&ebt_ops_filter[j]);
-	ebt_unregister_table(&frame_filter);
+	ret = nf_register_hooks(ebt_ops_filter, ARRAY_SIZE(ebt_ops_filter));
+	if (ret < 0)
+		ebt_unregister_table(&frame_filter);
 	return ret;
 }
 
 static void __exit ebtable_filter_fini(void)
 {
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(ebt_ops_filter); i++)
-		nf_unregister_hook(&ebt_ops_filter[i]);
+	nf_unregister_hooks(ebt_ops_filter, ARRAY_SIZE(ebt_ops_filter));
 	ebt_unregister_table(&frame_filter);
 }
 
diff --git a/net/bridge/netfilter/ebtable_nat.c b/net/bridge/netfilter/ebtable_nat.c
index 5b495fe2d0b6..f60c1e78e575 100644
--- a/net/bridge/netfilter/ebtable_nat.c
+++ b/net/bridge/netfilter/ebtable_nat.c
@@ -100,28 +100,20 @@ static struct nf_hook_ops ebt_ops_nat[] __read_mostly = {
 
 static int __init ebtable_nat_init(void)
 {
-	int i, ret, j;
+	int ret;
 
 	ret = ebt_register_table(&frame_nat);
 	if (ret < 0)
 		return ret;
-	for (i = 0; i < ARRAY_SIZE(ebt_ops_nat); i++)
-		if ((ret = nf_register_hook(&ebt_ops_nat[i])) < 0)
-			goto cleanup;
-	return ret;
-cleanup:
-	for (j = 0; j < i; j++)
-		nf_unregister_hook(&ebt_ops_nat[j]);
-	ebt_unregister_table(&frame_nat);
+	ret = nf_register_hooks(ebt_ops_nat, ARRAY_SIZE(ebt_ops_nat));
+	if (ret < 0)
+		ebt_unregister_table(&frame_nat);
 	return ret;
 }
 
 static void __exit ebtable_nat_fini(void)
 {
-	int i;
-
-	for (i = 0; i < ARRAY_SIZE(ebt_ops_nat); i++)
-		nf_unregister_hook(&ebt_ops_nat[i]);
+	nf_unregister_hooks(ebt_ops_nat, ARRAY_SIZE(ebt_ops_nat));
 	ebt_unregister_table(&frame_nat);
 }
 

From 6c5a9d2e1599a099b0e47235a1c1502162b14310 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Sat, 26 Jul 2008 17:48:15 -0700
Subject: [PATCH 1120/1171] selinux: use nf_register_hooks()

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Acked-by: James Morris <jmorris@namei.org>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 security/selinux/hooks.c | 27 ++++++++-------------------
 1 file changed, 8 insertions(+), 19 deletions(-)

diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 63f131fc42e4..df0515dd4d12 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -5670,27 +5670,20 @@ static struct nf_hook_ops selinux_ipv6_ops[] = {
 static int __init selinux_nf_ip_init(void)
 {
 	int err = 0;
-	u32 iter;
 
 	if (!selinux_enabled)
 		goto out;
 
 	printk(KERN_DEBUG "SELinux:  Registering netfilter hooks\n");
 
-	for (iter = 0; iter < ARRAY_SIZE(selinux_ipv4_ops); iter++) {
-		err = nf_register_hook(&selinux_ipv4_ops[iter]);
-		if (err)
-			panic("SELinux: nf_register_hook for IPv4: error %d\n",
-			      err);
-	}
+	err = nf_register_hooks(selinux_ipv4_ops, ARRAY_SIZE(selinux_ipv4_ops));
+	if (err)
+		panic("SELinux: nf_register_hooks for IPv4: error %d\n", err);
 
 #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
-	for (iter = 0; iter < ARRAY_SIZE(selinux_ipv6_ops); iter++) {
-		err = nf_register_hook(&selinux_ipv6_ops[iter]);
-		if (err)
-			panic("SELinux: nf_register_hook for IPv6: error %d\n",
-			      err);
-	}
+	err = nf_register_hooks(selinux_ipv6_ops, ARRAY_SIZE(selinux_ipv6_ops));
+	if (err)
+		panic("SELinux: nf_register_hooks for IPv6: error %d\n", err);
 #endif	/* IPV6 */
 
 out:
@@ -5702,15 +5695,11 @@ __initcall(selinux_nf_ip_init);
 #ifdef CONFIG_SECURITY_SELINUX_DISABLE
 static void selinux_nf_ip_exit(void)
 {
-	u32 iter;
-
 	printk(KERN_DEBUG "SELinux:  Unregistering netfilter hooks\n");
 
-	for (iter = 0; iter < ARRAY_SIZE(selinux_ipv4_ops); iter++)
-		nf_unregister_hook(&selinux_ipv4_ops[iter]);
+	nf_unregister_hooks(selinux_ipv4_ops, ARRAY_SIZE(selinux_ipv4_ops));
 #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
-	for (iter = 0; iter < ARRAY_SIZE(selinux_ipv6_ops); iter++)
-		nf_unregister_hook(&selinux_ipv6_ops[iter]);
+	nf_unregister_hooks(selinux_ipv6_ops, ARRAY_SIZE(selinux_ipv6_ops));
 #endif	/* IPV6 */
 }
 #endif

From f858b4869a9136dd28cc2ab37f8b89268cc99462 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Sat, 26 Jul 2008 17:48:38 -0700
Subject: [PATCH 1121/1171] netfilter: ip{,6}tables_security: fix future
 section mismatch

Currently not visible, because NET_NS is mutually exclusive with SYSFS
which is required by SECURITY.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/netfilter/iptable_security.c  | 2 +-
 net/ipv6/netfilter/ip6table_security.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/net/ipv4/netfilter/iptable_security.c b/net/ipv4/netfilter/iptable_security.c
index 2b472ac2263a..db6d312128e1 100644
--- a/net/ipv4/netfilter/iptable_security.c
+++ b/net/ipv4/netfilter/iptable_security.c
@@ -32,7 +32,7 @@ static struct
 	struct ipt_replace repl;
 	struct ipt_standard entries[3];
 	struct ipt_error term;
-} initial_table __initdata = {
+} initial_table __net_initdata = {
 	.repl = {
 		.name = "security",
 		.valid_hooks = SECURITY_VALID_HOOKS,
diff --git a/net/ipv6/netfilter/ip6table_security.c b/net/ipv6/netfilter/ip6table_security.c
index a07abee30497..6e7131036bc6 100644
--- a/net/ipv6/netfilter/ip6table_security.c
+++ b/net/ipv6/netfilter/ip6table_security.c
@@ -31,7 +31,7 @@ static struct
 	struct ip6t_replace repl;
 	struct ip6t_standard entries[3];
 	struct ip6t_error term;
-} initial_table __initdata = {
+} initial_table __net_initdata = {
 	.repl = {
 		.name = "security",
 		.valid_hooks = SECURITY_VALID_HOOKS,

From 3918fed5f31213067c1c345bd904e1ea369e6819 Mon Sep 17 00:00:00 2001
From: Alexey Dobriyan <adobriyan@gmail.com>
Date: Sat, 26 Jul 2008 17:48:59 -0700
Subject: [PATCH 1122/1171] netfilter: arptables in netns for real

IN, FORWARD -- grab netns from in device, OUT -- from out device.

Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/netfilter/arptable_filter.c | 39 +++++++++++++++++++++-------
 1 file changed, 30 insertions(+), 9 deletions(-)

diff --git a/net/ipv4/netfilter/arptable_filter.c b/net/ipv4/netfilter/arptable_filter.c
index 3be4d07e7ed9..082f5dd3156c 100644
--- a/net/ipv4/netfilter/arptable_filter.c
+++ b/net/ipv4/netfilter/arptable_filter.c
@@ -55,32 +55,53 @@ static struct xt_table packet_filter = {
 };
 
 /* The work comes in here from netfilter.c */
-static unsigned int arpt_hook(unsigned int hook,
-			      struct sk_buff *skb,
-			      const struct net_device *in,
-			      const struct net_device *out,
-			      int (*okfn)(struct sk_buff *))
+static unsigned int arpt_in_hook(unsigned int hook,
+				 struct sk_buff *skb,
+				 const struct net_device *in,
+				 const struct net_device *out,
+				 int (*okfn)(struct sk_buff *))
 {
-	return arpt_do_table(skb, hook, in, out, init_net.ipv4.arptable_filter);
+	return arpt_do_table(skb, hook, in, out,
+			     dev_net(in)->ipv4.arptable_filter);
+}
+
+static unsigned int arpt_out_hook(unsigned int hook,
+				  struct sk_buff *skb,
+				  const struct net_device *in,
+				  const struct net_device *out,
+				  int (*okfn)(struct sk_buff *))
+{
+	return arpt_do_table(skb, hook, in, out,
+			     dev_net(out)->ipv4.arptable_filter);
+}
+
+static unsigned int arpt_forward_hook(unsigned int hook,
+				      struct sk_buff *skb,
+				      const struct net_device *in,
+				      const struct net_device *out,
+				      int (*okfn)(struct sk_buff *))
+{
+	return arpt_do_table(skb, hook, in, out,
+			     dev_net(in)->ipv4.arptable_filter);
 }
 
 static struct nf_hook_ops arpt_ops[] __read_mostly = {
 	{
-		.hook		= arpt_hook,
+		.hook		= arpt_in_hook,
 		.owner		= THIS_MODULE,
 		.pf		= NF_ARP,
 		.hooknum	= NF_ARP_IN,
 		.priority	= NF_IP_PRI_FILTER,
 	},
 	{
-		.hook		= arpt_hook,
+		.hook		= arpt_out_hook,
 		.owner		= THIS_MODULE,
 		.pf		= NF_ARP,
 		.hooknum	= NF_ARP_OUT,
 		.priority	= NF_IP_PRI_FILTER,
 	},
 	{
-		.hook		= arpt_hook,
+		.hook		= arpt_forward_hook,
 		.owner		= THIS_MODULE,
 		.pf		= NF_ARP,
 		.hooknum	= NF_ARP_FORWARD,

From 93bc4e89c260d91576840c4881d1066d84ccd422 Mon Sep 17 00:00:00 2001
From: Pekka Enberg <penberg@cs.helsinki.fi>
Date: Sat, 26 Jul 2008 17:49:33 -0700
Subject: [PATCH 1123/1171] netfilter: fix double-free and use-after free

As suggested by Patrick McHardy, introduce a __krealloc() that doesn't
free the original buffer to fix a double-free and use-after-free bug
introduced by me in netfilter that uses RCU.

Reported-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Tested-by: Dieter Ries <clip2@gmx.de>
Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 include/linux/slab.h                |  1 +
 mm/util.c                           | 46 ++++++++++++++++++++++-------
 net/netfilter/nf_conntrack_extend.c |  2 +-
 3 files changed, 37 insertions(+), 12 deletions(-)

diff --git a/include/linux/slab.h b/include/linux/slab.h
index 9aa90a6f20e0..be6f1d40b66a 100644
--- a/include/linux/slab.h
+++ b/include/linux/slab.h
@@ -96,6 +96,7 @@ int kmem_ptr_validate(struct kmem_cache *cachep, const void *ptr);
 /*
  * Common kmalloc functions provided by all allocators
  */
+void * __must_check __krealloc(const void *, size_t, gfp_t);
 void * __must_check krealloc(const void *, size_t, gfp_t);
 void kfree(const void *);
 size_t ksize(const void *);
diff --git a/mm/util.c b/mm/util.c
index 8f18683825bc..6ef9e9943f62 100644
--- a/mm/util.c
+++ b/mm/util.c
@@ -67,6 +67,38 @@ void *kmemdup(const void *src, size_t len, gfp_t gfp)
 }
 EXPORT_SYMBOL(kmemdup);
 
+/**
+ * __krealloc - like krealloc() but don't free @p.
+ * @p: object to reallocate memory for.
+ * @new_size: how many bytes of memory are required.
+ * @flags: the type of memory to allocate.
+ *
+ * This function is like krealloc() except it never frees the originally
+ * allocated buffer. Use this if you don't want to free the buffer immediately
+ * like, for example, with RCU.
+ */
+void *__krealloc(const void *p, size_t new_size, gfp_t flags)
+{
+	void *ret;
+	size_t ks = 0;
+
+	if (unlikely(!new_size))
+		return ZERO_SIZE_PTR;
+
+	if (p)
+		ks = ksize(p);
+
+	if (ks >= new_size)
+		return (void *)p;
+
+	ret = kmalloc_track_caller(new_size, flags);
+	if (ret && p)
+		memcpy(ret, p, ks);
+
+	return ret;
+}
+EXPORT_SYMBOL(__krealloc);
+
 /**
  * krealloc - reallocate memory. The contents will remain unchanged.
  * @p: object to reallocate memory for.
@@ -81,24 +113,16 @@ EXPORT_SYMBOL(kmemdup);
 void *krealloc(const void *p, size_t new_size, gfp_t flags)
 {
 	void *ret;
-	size_t ks = 0;
 
 	if (unlikely(!new_size)) {
 		kfree(p);
 		return ZERO_SIZE_PTR;
 	}
 
-	if (p)
-		ks = ksize(p);
-
-	if (ks >= new_size)
-		return (void *)p;
-
-	ret = kmalloc_track_caller(new_size, flags);
-	if (ret && p) {
-		memcpy(ret, p, ks);
+	ret = __krealloc(p, new_size, flags);
+	if (ret && p != ret)
 		kfree(p);
-	}
+
 	return ret;
 }
 EXPORT_SYMBOL(krealloc);
diff --git a/net/netfilter/nf_conntrack_extend.c b/net/netfilter/nf_conntrack_extend.c
index 3469bc71a385..c956ef7eeecb 100644
--- a/net/netfilter/nf_conntrack_extend.c
+++ b/net/netfilter/nf_conntrack_extend.c
@@ -95,7 +95,7 @@ void *__nf_ct_ext_add(struct nf_conn *ct, enum nf_ct_ext_id id, gfp_t gfp)
 	newlen = newoff + t->len;
 	rcu_read_unlock();
 
-	new = krealloc(ct->ext, newlen, gfp);
+	new = __krealloc(ct->ext, newlen, gfp);
 	if (!new)
 		return NULL;
 

From 6c64825bf40ecc1b01610762ca736b18c8a9db92 Mon Sep 17 00:00:00 2001
From: Patrick McHardy <kaber@trash.net>
Date: Sat, 26 Jul 2008 17:50:05 -0700
Subject: [PATCH 1124/1171] netfilter: nf_conntrack_extend: avoid unnecessary
 "ct->ext" dereferences

As Linus points out, "ct->ext" and "new" are always equal, avoid unnecessary
dereferences and use "new" directly.

Signed-off-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/netfilter/nf_conntrack_extend.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/net/netfilter/nf_conntrack_extend.c b/net/netfilter/nf_conntrack_extend.c
index c956ef7eeecb..4b2c769d555f 100644
--- a/net/netfilter/nf_conntrack_extend.c
+++ b/net/netfilter/nf_conntrack_extend.c
@@ -115,10 +115,10 @@ void *__nf_ct_ext_add(struct nf_conn *ct, enum nf_ct_ext_id id, gfp_t gfp)
 		ct->ext = new;
 	}
 
-	ct->ext->offset[id] = newoff;
-	ct->ext->len = newlen;
-	memset((void *)ct->ext + newoff, 0, newlen - newoff);
-	return (void *)ct->ext + newoff;
+	new->offset[id] = newoff;
+	new->len = newlen;
+	memset((void *)new + newoff, 0, newlen - newoff);
+	return (void *)new + newoff;
 }
 EXPORT_SYMBOL(__nf_ct_ext_add);
 

From 6c3b8fc618905d7599dcc514c99ce4293d476f39 Mon Sep 17 00:00:00 2001
From: Hugh Dickins <hugh@veritas.com>
Date: Sat, 26 Jul 2008 17:51:06 -0700
Subject: [PATCH 1125/1171] netns: fix ip_rt_frag_needed rt_is_expired

Running recent kernels, and using a particular vpn gateway, I've been
having to edit my mails down to get them accepted by the smtp server.

Git bisect led to commit e84f84f276473dcc673f360e8ff3203148bdf0e2 -
netns: place rt_genid into struct net.  The conversion from a != test
to rt_is_expired() put one negative too many: and now my mail works.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Acked-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 net/ipv4/route.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index e4ab0ac94f92..a507c5e27d0e 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1502,7 +1502,7 @@ unsigned short ip_rt_frag_needed(struct net *net, struct iphdr *iph,
 				    rth->fl.iif != 0 ||
 				    dst_metric_locked(&rth->u.dst, RTAX_MTU) ||
 				    !net_eq(dev_net(rth->u.dst.dev), net) ||
-				    !rt_is_expired(rth))
+				    rt_is_expired(rth))
 					continue;
 
 				if (new_mtu < 68 || new_mtu >= old_mtu) {

From d70b67c8bc72ee23b55381bd6a884f4796692f77 Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Wed, 2 Jul 2008 21:30:15 +0200
Subject: [PATCH 1126/1171] [patch] vfs: fix lookup on deleted directory

Lookup can install a child dentry for a deleted directory.  This keeps
the directory dentry alive, and the inode pinned in the cache and on
disk, even after all external references have gone away.

This isn't a big problem normally, since memory pressure or umount
will clear out the directory dentry and its children, releasing the
inode.  But for UBIFS this causes problems because its orphan area can
overflow.

Fix this by returning ENOENT for all lookups on a S_DEAD directory
before creating a child dentry.

Thanks to Zoltan Sogor for noticing this while testing UBIFS, and
Artem for the excellent analysis of the problem and testing.

Reported-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Tested-by: Artem Bityutskiy <Artem.Bityutskiy@nokia.com>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/namei.c | 19 +++++++++++++++++--
 1 file changed, 17 insertions(+), 2 deletions(-)

diff --git a/fs/namei.c b/fs/namei.c
index 01e67dddcc3d..3b26a240ade9 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -519,7 +519,14 @@ static struct dentry * real_lookup(struct dentry * parent, struct qstr * name, s
 	 */
 	result = d_lookup(parent, name);
 	if (!result) {
-		struct dentry * dentry = d_alloc(parent, name);
+		struct dentry *dentry;
+
+		/* Don't create child dentry for a dead directory. */
+		result = ERR_PTR(-ENOENT);
+		if (IS_DEADDIR(dir))
+			goto out_unlock;
+
+		dentry = d_alloc(parent, name);
 		result = ERR_PTR(-ENOMEM);
 		if (dentry) {
 			result = dir->i_op->lookup(dir, dentry, nd);
@@ -528,6 +535,7 @@ static struct dentry * real_lookup(struct dentry * parent, struct qstr * name, s
 			else
 				result = dentry;
 		}
+out_unlock:
 		mutex_unlock(&dir->i_mutex);
 		return result;
 	}
@@ -1317,7 +1325,14 @@ static struct dentry *__lookup_hash(struct qstr *name,
 
 	dentry = cached_lookup(base, name, nd);
 	if (!dentry) {
-		struct dentry *new = d_alloc(base, name);
+		struct dentry *new;
+
+		/* Don't create child dentry for a dead directory. */
+		dentry = ERR_PTR(-ENOENT);
+		if (IS_DEADDIR(inode))
+			goto out;
+
+		new = d_alloc(base, name);
 		dentry = ERR_PTR(-ENOMEM);
 		if (!new)
 			goto out;

From d2d9648ec6858e19d16a0b16da62534e85888653 Mon Sep 17 00:00:00 2001
From: Denys Vlasenko <vda.linux@googlemail.com>
Date: Tue, 1 Jul 2008 14:16:09 +0200
Subject: [PATCH 1127/1171] [PATCH] reuse xxx_fifo_fops for xxx_pipe_fops

Merge fifo and pipe file_operations.

Signed-off-by: Denys Vlasenko <vda.linux@googlemail.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/fifo.c          |  8 ++++----
 fs/pipe.c          | 51 ++++++++--------------------------------------
 include/linux/fs.h |  6 +++---
 3 files changed, 15 insertions(+), 50 deletions(-)

diff --git a/fs/fifo.c b/fs/fifo.c
index 9785e36f81e7..987bf9411495 100644
--- a/fs/fifo.c
+++ b/fs/fifo.c
@@ -57,7 +57,7 @@ static int fifo_open(struct inode *inode, struct file *filp)
 	 *  POSIX.1 says that O_NONBLOCK means return with the FIFO
 	 *  opened, even when there is no process writing the FIFO.
 	 */
-		filp->f_op = &read_fifo_fops;
+		filp->f_op = &read_pipefifo_fops;
 		pipe->r_counter++;
 		if (pipe->readers++ == 0)
 			wake_up_partner(inode);
@@ -86,7 +86,7 @@ static int fifo_open(struct inode *inode, struct file *filp)
 		if ((filp->f_flags & O_NONBLOCK) && !pipe->readers)
 			goto err;
 
-		filp->f_op = &write_fifo_fops;
+		filp->f_op = &write_pipefifo_fops;
 		pipe->w_counter++;
 		if (!pipe->writers++)
 			wake_up_partner(inode);
@@ -105,7 +105,7 @@ static int fifo_open(struct inode *inode, struct file *filp)
 	 *  This implementation will NEVER block on a O_RDWR open, since
 	 *  the process can at least talk to itself.
 	 */
-		filp->f_op = &rdwr_fifo_fops;
+		filp->f_op = &rdwr_pipefifo_fops;
 
 		pipe->readers++;
 		pipe->writers++;
@@ -151,5 +151,5 @@ err_nocleanup:
  * depending on the access mode of the file...
  */
 const struct file_operations def_fifo_fops = {
-	.open		= fifo_open,	/* will set read or write pipe_fops */
+	.open		= fifo_open,	/* will set read_ or write_pipefifo_fops */
 };
diff --git a/fs/pipe.c b/fs/pipe.c
index 10c4e9aa5c49..fcba6542b8d0 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -777,8 +777,10 @@ pipe_rdwr_open(struct inode *inode, struct file *filp)
 /*
  * The file_operations structs are not static because they
  * are also used in linux/fs/fifo.c to do operations on FIFOs.
+ *
+ * Pipes reuse fifos' file_operations structs.
  */
-const struct file_operations read_fifo_fops = {
+const struct file_operations read_pipefifo_fops = {
 	.llseek		= no_llseek,
 	.read		= do_sync_read,
 	.aio_read	= pipe_read,
@@ -790,7 +792,7 @@ const struct file_operations read_fifo_fops = {
 	.fasync		= pipe_read_fasync,
 };
 
-const struct file_operations write_fifo_fops = {
+const struct file_operations write_pipefifo_fops = {
 	.llseek		= no_llseek,
 	.read		= bad_pipe_r,
 	.write		= do_sync_write,
@@ -802,44 +804,7 @@ const struct file_operations write_fifo_fops = {
 	.fasync		= pipe_write_fasync,
 };
 
-const struct file_operations rdwr_fifo_fops = {
-	.llseek		= no_llseek,
-	.read		= do_sync_read,
-	.aio_read	= pipe_read,
-	.write		= do_sync_write,
-	.aio_write	= pipe_write,
-	.poll		= pipe_poll,
-	.unlocked_ioctl	= pipe_ioctl,
-	.open		= pipe_rdwr_open,
-	.release	= pipe_rdwr_release,
-	.fasync		= pipe_rdwr_fasync,
-};
-
-static const struct file_operations read_pipe_fops = {
-	.llseek		= no_llseek,
-	.read		= do_sync_read,
-	.aio_read	= pipe_read,
-	.write		= bad_pipe_w,
-	.poll		= pipe_poll,
-	.unlocked_ioctl	= pipe_ioctl,
-	.open		= pipe_read_open,
-	.release	= pipe_read_release,
-	.fasync		= pipe_read_fasync,
-};
-
-static const struct file_operations write_pipe_fops = {
-	.llseek		= no_llseek,
-	.read		= bad_pipe_r,
-	.write		= do_sync_write,
-	.aio_write	= pipe_write,
-	.poll		= pipe_poll,
-	.unlocked_ioctl	= pipe_ioctl,
-	.open		= pipe_write_open,
-	.release	= pipe_write_release,
-	.fasync		= pipe_write_fasync,
-};
-
-static const struct file_operations rdwr_pipe_fops = {
+const struct file_operations rdwr_pipefifo_fops = {
 	.llseek		= no_llseek,
 	.read		= do_sync_read,
 	.aio_read	= pipe_read,
@@ -927,7 +892,7 @@ static struct inode * get_pipe_inode(void)
 	inode->i_pipe = pipe;
 
 	pipe->readers = pipe->writers = 1;
-	inode->i_fop = &rdwr_pipe_fops;
+	inode->i_fop = &rdwr_pipefifo_fops;
 
 	/*
 	 * Mark the inode dirty from the very beginning,
@@ -978,7 +943,7 @@ struct file *create_write_pipe(int flags)
 	d_instantiate(dentry, inode);
 
 	err = -ENFILE;
-	f = alloc_file(pipe_mnt, dentry, FMODE_WRITE, &write_pipe_fops);
+	f = alloc_file(pipe_mnt, dentry, FMODE_WRITE, &write_pipefifo_fops);
 	if (!f)
 		goto err_dentry;
 	f->f_mapping = inode->i_mapping;
@@ -1020,7 +985,7 @@ struct file *create_read_pipe(struct file *wrf, int flags)
 
 	f->f_pos = 0;
 	f->f_flags = O_RDONLY | (flags & O_NONBLOCK);
-	f->f_op = &read_pipe_fops;
+	f->f_op = &read_pipefifo_fops;
 	f->f_mode = FMODE_READ;
 	f->f_version = 0;
 
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 53d2edb709b3..7721a2ac9c0e 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1696,9 +1696,9 @@ extern void init_special_inode(struct inode *, umode_t, dev_t);
 extern void make_bad_inode(struct inode *);
 extern int is_bad_inode(struct inode *);
 
-extern const struct file_operations read_fifo_fops;
-extern const struct file_operations write_fifo_fops;
-extern const struct file_operations rdwr_fifo_fops;
+extern const struct file_operations read_pipefifo_fops;
+extern const struct file_operations write_pipefifo_fops;
+extern const struct file_operations rdwr_pipefifo_fops;
 
 extern int fs_may_remount_ro(struct super_block *);
 

From 7ac6cd653d7c31ad6b7bb5b88c549c4ebf628c34 Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Tue, 1 Jul 2008 23:07:54 +0200
Subject: [PATCH 1128/1171] [patch] hppfs: remove hppfs_permission

hppfs_permission() is equivalent to the '.permission == NULL' case.

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/hppfs/hppfs.c | 7 -------
 1 file changed, 7 deletions(-)

diff --git a/fs/hppfs/hppfs.c b/fs/hppfs/hppfs.c
index 65077aa90f0a..2b3d1828db99 100644
--- a/fs/hppfs/hppfs.c
+++ b/fs/hppfs/hppfs.c
@@ -655,20 +655,13 @@ static void *hppfs_follow_link(struct dentry *dentry, struct nameidata *nd)
 	return proc_dentry->d_inode->i_op->follow_link(proc_dentry, nd);
 }
 
-int hppfs_permission(struct inode *inode, int mask, struct nameidata *nd)
-{
-	return generic_permission(inode, mask, NULL);
-}
-
 static const struct inode_operations hppfs_dir_iops = {
 	.lookup		= hppfs_lookup,
-	.permission	= hppfs_permission,
 };
 
 static const struct inode_operations hppfs_link_iops = {
 	.readlink	= hppfs_readlink,
 	.follow_link	= hppfs_follow_link,
-	.permission	= hppfs_permission,
 };
 
 static struct inode *get_inode(struct super_block *sb, struct dentry *dentry)

From 734550921e9b7ab924a43aa3d0bd4239dac4fbf1 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 14 Jul 2008 21:22:20 -0400
Subject: [PATCH 1129/1171] [PATCH] beginning of sysctl cleanup - ctl_table_set

New object: set of sysctls [currently - root and per-net-ns].
Contains: pointer to parent set, list of tables and "should I see this set?"
method (->is_seen(set)).
Current lists of tables are subsumed by that; net-ns contains such a beast.
->lookup() for ctl_table_root returns pointer to ctl_table_set instead of
that to ->list of that ctl_table_set.

[folded compile fixes by rdd for configs without sysctl]

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/sysctl.h      | 15 ++++++++++++--
 include/net/net_namespace.h |  4 +++-
 kernel/sysctl.c             | 41 ++++++++++++++++++++++++++++---------
 net/sysctl_net.c            | 22 +++++++++-----------
 4 files changed, 57 insertions(+), 25 deletions(-)

diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 24141b4d1a11..c1e0cf408af9 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -947,6 +947,16 @@ struct ctl_table;
 struct nsproxy;
 struct ctl_table_root;
 
+struct ctl_table_set {
+	struct list_head list;
+	struct ctl_table_set *parent;
+	int (*is_seen)(struct ctl_table_set *);
+};
+
+extern void setup_sysctl_set(struct ctl_table_set *p,
+	struct ctl_table_set *parent,
+	int (*is_seen)(struct ctl_table_set *));
+
 extern struct ctl_table_header *sysctl_head_next(struct ctl_table_header *prev);
 extern struct ctl_table_header *__sysctl_head_next(struct nsproxy *namespaces,
 						struct ctl_table_header *prev);
@@ -1049,8 +1059,8 @@ struct ctl_table
 
 struct ctl_table_root {
 	struct list_head root_list;
-	struct list_head header_list;
-	struct list_head *(*lookup)(struct ctl_table_root *root,
+	struct ctl_table_set default_set;
+	struct ctl_table_set *(*lookup)(struct ctl_table_root *root,
 					   struct nsproxy *namespaces);
 	int (*permissions)(struct ctl_table_root *root,
 			struct nsproxy *namespaces, struct ctl_table *table);
@@ -1066,6 +1076,7 @@ struct ctl_table_header
 	struct completion *unregistering;
 	struct ctl_table *ctl_table_arg;
 	struct ctl_table_root *root;
+	struct ctl_table_set *set;
 };
 
 /* struct ctl_path describes where in the hierarchy a table is added */
diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index 3855620b78a9..a8eb43cf0c7e 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -38,7 +38,9 @@ struct net {
 	struct proc_dir_entry 	*proc_net;
 	struct proc_dir_entry 	*proc_net_stat;
 
-	struct list_head	sysctl_table_headers;
+#ifdef CONFIG_SYSCTL
+	struct ctl_table_set	sysctls;
+#endif
 
 	struct net_device       *loopback_dev;          /* The loopback */
 
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 35a50db9b6ce..8ee4a0619fbb 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -160,12 +160,13 @@ static struct ctl_table root_table[];
 static struct ctl_table_root sysctl_table_root;
 static struct ctl_table_header root_table_header = {
 	.ctl_table = root_table,
-	.ctl_entry = LIST_HEAD_INIT(sysctl_table_root.header_list),
+	.ctl_entry = LIST_HEAD_INIT(sysctl_table_root.default_set.list),
 	.root = &sysctl_table_root,
+	.set = &sysctl_table_root.default_set,
 };
 static struct ctl_table_root sysctl_table_root = {
 	.root_list = LIST_HEAD_INIT(sysctl_table_root.root_list),
-	.header_list = LIST_HEAD_INIT(root_table_header.ctl_entry),
+	.default_set.list = LIST_HEAD_INIT(root_table_header.ctl_entry),
 };
 
 static struct ctl_table kern_table[];
@@ -1403,14 +1404,20 @@ void sysctl_head_finish(struct ctl_table_header *head)
 	spin_unlock(&sysctl_lock);
 }
 
+static struct ctl_table_set *
+lookup_header_set(struct ctl_table_root *root, struct nsproxy *namespaces)
+{
+	struct ctl_table_set *set = &root->default_set;
+	if (root->lookup)
+		set = root->lookup(root, namespaces);
+	return set;
+}
+
 static struct list_head *
 lookup_header_list(struct ctl_table_root *root, struct nsproxy *namespaces)
 {
-	struct list_head *header_list;
-	header_list = &root->header_list;
-	if (root->lookup)
-		header_list = root->lookup(root, namespaces);
-	return header_list;
+	struct ctl_table_set *set = lookup_header_set(root, namespaces);
+	return &set->list;
 }
 
 struct ctl_table_header *__sysctl_head_next(struct nsproxy *namespaces,
@@ -1720,7 +1727,6 @@ struct ctl_table_header *__register_sysctl_paths(
 	struct nsproxy *namespaces,
 	const struct ctl_path *path, struct ctl_table *table)
 {
-	struct list_head *header_list;
 	struct ctl_table_header *header;
 	struct ctl_table *new, **prevp;
 	unsigned int n, npath;
@@ -1772,8 +1778,8 @@ struct ctl_table_header *__register_sysctl_paths(
 	}
 #endif
 	spin_lock(&sysctl_lock);
-	header_list = lookup_header_list(root, namespaces);
-	list_add_tail(&header->ctl_entry, header_list);
+	header->set = lookup_header_set(root, namespaces);
+	list_add_tail(&header->ctl_entry, &header->set->list);
 	spin_unlock(&sysctl_lock);
 
 	return header;
@@ -1832,6 +1838,15 @@ void unregister_sysctl_table(struct ctl_table_header * header)
 	kfree(header);
 }
 
+void setup_sysctl_set(struct ctl_table_set *p,
+	struct ctl_table_set *parent,
+	int (*is_seen)(struct ctl_table_set *))
+{
+	INIT_LIST_HEAD(&p->list);
+	p->parent = parent ? parent : &sysctl_table_root.default_set;
+	p->is_seen = is_seen;
+}
+
 #else /* !CONFIG_SYSCTL */
 struct ctl_table_header *register_sysctl_table(struct ctl_table * table)
 {
@@ -1848,6 +1863,12 @@ void unregister_sysctl_table(struct ctl_table_header * table)
 {
 }
 
+void setup_sysctl_set(struct ctl_table_set *p,
+	struct ctl_table_set *parent,
+	int (*is_seen)(struct ctl_table_set *))
+{
+}
+
 #endif /* CONFIG_SYSCTL */
 
 /*
diff --git a/net/sysctl_net.c b/net/sysctl_net.c
index 63ada437fc2f..cefbc367d8be 100644
--- a/net/sysctl_net.c
+++ b/net/sysctl_net.c
@@ -29,10 +29,15 @@
 #include <linux/if_tr.h>
 #endif
 
-static struct list_head *
+static struct ctl_table_set *
 net_ctl_header_lookup(struct ctl_table_root *root, struct nsproxy *namespaces)
 {
-	return &namespaces->net_ns->sysctl_table_headers;
+	return &namespaces->net_ns->sysctls;
+}
+
+static int is_seen(struct ctl_table_set *set)
+{
+	return &current->nsproxy->net_ns->sysctls == set;
 }
 
 /* Return standard mode bits for table entry. */
@@ -53,13 +58,6 @@ static struct ctl_table_root net_sysctl_root = {
 	.permissions = net_ctl_permissions,
 };
 
-static LIST_HEAD(net_sysctl_ro_tables);
-static struct list_head *net_ctl_ro_header_lookup(struct ctl_table_root *root,
-		struct nsproxy *namespaces)
-{
-	return &net_sysctl_ro_tables;
-}
-
 static int net_ctl_ro_header_perms(struct ctl_table_root *root,
 		struct nsproxy *namespaces, struct ctl_table *table)
 {
@@ -70,19 +68,18 @@ static int net_ctl_ro_header_perms(struct ctl_table_root *root,
 }
 
 static struct ctl_table_root net_sysctl_ro_root = {
-	.lookup = net_ctl_ro_header_lookup,
 	.permissions = net_ctl_ro_header_perms,
 };
 
 static int sysctl_net_init(struct net *net)
 {
-	INIT_LIST_HEAD(&net->sysctl_table_headers);
+	setup_sysctl_set(&net->sysctls, NULL, is_seen);
 	return 0;
 }
 
 static void sysctl_net_exit(struct net *net)
 {
-	WARN_ON(!list_empty(&net->sysctl_table_headers));
+	WARN_ON(!list_empty(&net->sysctls.list));
 	return;
 }
 
@@ -98,6 +95,7 @@ static __init int sysctl_init(void)
 	if (ret)
 		goto out;
 	register_sysctl_root(&net_sysctl_root);
+	setup_sysctl_set(&net_sysctl_ro_root.default_set, NULL, NULL);
 	register_sysctl_root(&net_sysctl_ro_root);
 out:
 	return ret;

From f7e6ced4061da509f737541ca4dbd44d83a6e82f Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 15 Jul 2008 01:44:23 -0400
Subject: [PATCH 1130/1171] [PATCH] allow delayed freeing of ctl_table_header

Refcount the sucker; instead of freeing it by the end of unregistration
just drop the refcount and free only when it hits zero.  Make sure that
we _always_ make ->unregistering non-NULL in start_unregistering().

That allows anybody to get a reference to such puppy, preventing its
freeing and reuse.  It does *not* block unregistration.  Anybody who
holds such a reference can
	* try to grab a "use" reference (ctl_head_grab()); that will
succeeds if and only if it hadn't entered unregistration yet.  If it
succeeds, we can use it in all normal ways until we release the "use"
reference (with ctl_head_finish()).  Note that this relies on having
->unregistering become non-NULL in all cases when one starts to unregister
the sucker.
	* keep pointers to ctl_table entries; they *can* be freed if
the entire thing is unregistered.  However, if ctl_head_grab() succeeds,
we know that unregistration had not happened (and will not happen until
ctl_head_finish()) and such pointers can be used safely.

IOW, now we can have inodes under /proc/sys keep references to ctl_table
entries, protecting them with references to ctl_table_header and
grabbing the latter for the duration of operations that require access
to ctl_table.  That won't cause deadlocks, since unregistration will not
be stopped by mere keeping a reference to ctl_table_header.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/sysctl.h |  6 ++++++
 kernel/sysctl.c        | 37 ++++++++++++++++++++++++++++++++++++-
 2 files changed, 42 insertions(+), 1 deletion(-)

diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index c1e0cf408af9..956264d09ba0 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -957,6 +957,11 @@ extern void setup_sysctl_set(struct ctl_table_set *p,
 	struct ctl_table_set *parent,
 	int (*is_seen)(struct ctl_table_set *));
 
+struct ctl_table_header;
+
+extern void sysctl_head_get(struct ctl_table_header *);
+extern void sysctl_head_put(struct ctl_table_header *);
+extern struct ctl_table_header *sysctl_head_grab(struct ctl_table_header *);
 extern struct ctl_table_header *sysctl_head_next(struct ctl_table_header *prev);
 extern struct ctl_table_header *__sysctl_head_next(struct nsproxy *namespaces,
 						struct ctl_table_header *prev);
@@ -1073,6 +1078,7 @@ struct ctl_table_header
 	struct ctl_table *ctl_table;
 	struct list_head ctl_entry;
 	int used;
+	int count;
 	struct completion *unregistering;
 	struct ctl_table *ctl_table_arg;
 	struct ctl_table_root *root;
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 8ee4a0619fbb..60d9357e7172 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -1387,6 +1387,9 @@ static void start_unregistering(struct ctl_table_header *p)
 		spin_unlock(&sysctl_lock);
 		wait_for_completion(&wait);
 		spin_lock(&sysctl_lock);
+	} else {
+		/* anything non-NULL; we'll never dereference it */
+		p->unregistering = ERR_PTR(-EINVAL);
 	}
 	/*
 	 * do not remove from the list until nobody holds it; walking the
@@ -1395,6 +1398,32 @@ static void start_unregistering(struct ctl_table_header *p)
 	list_del_init(&p->ctl_entry);
 }
 
+void sysctl_head_get(struct ctl_table_header *head)
+{
+	spin_lock(&sysctl_lock);
+	head->count++;
+	spin_unlock(&sysctl_lock);
+}
+
+void sysctl_head_put(struct ctl_table_header *head)
+{
+	spin_lock(&sysctl_lock);
+	if (!--head->count)
+		kfree(head);
+	spin_unlock(&sysctl_lock);
+}
+
+struct ctl_table_header *sysctl_head_grab(struct ctl_table_header *head)
+{
+	if (!head)
+		BUG();
+	spin_lock(&sysctl_lock);
+	if (!use_table(head))
+		head = ERR_PTR(-ENOENT);
+	spin_unlock(&sysctl_lock);
+	return head;
+}
+
 void sysctl_head_finish(struct ctl_table_header *head)
 {
 	if (!head)
@@ -1771,6 +1800,7 @@ struct ctl_table_header *__register_sysctl_paths(
 	header->unregistering = NULL;
 	header->root = root;
 	sysctl_set_parent(NULL, header->ctl_table);
+	header->count = 1;
 #ifdef CONFIG_SYSCTL_SYSCALL_CHECK
 	if (sysctl_check_table(namespaces, header->ctl_table)) {
 		kfree(header);
@@ -1834,8 +1864,9 @@ void unregister_sysctl_table(struct ctl_table_header * header)
 
 	spin_lock(&sysctl_lock);
 	start_unregistering(header);
+	if (!--header->count)
+		kfree(header);
 	spin_unlock(&sysctl_lock);
-	kfree(header);
 }
 
 void setup_sysctl_set(struct ctl_table_set *p,
@@ -1869,6 +1900,10 @@ void setup_sysctl_set(struct ctl_table_set *p,
 {
 }
 
+void sysctl_head_put(struct ctl_table_header *head)
+{
+}
+
 #endif /* CONFIG_SYSCTL */
 
 /*

From bd7b1533cd6a68c734062aa69394bec7e2b1718e Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 15 Jul 2008 16:00:59 -0400
Subject: [PATCH 1131/1171] [PATCH] sysctl: make sure that /proc/sys/net/ipv4
 appears before per-ns ones

Massage ipv4 initialization - make sure that net.ipv4 appears as
non-per-net-namespace before it shows up in per-net-namespace sysctls.
That's the only change outside of sysctl.c needed to get sane ordering
rules and data structures for sysctls (esp. for procfs side of that
mess).

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/net/ip.h           | 2 ++
 net/ipv4/af_inet.c         | 4 ++++
 net/ipv4/sysctl_net_ipv4.c | 7 +++++++
 3 files changed, 13 insertions(+)

diff --git a/include/net/ip.h b/include/net/ip.h
index b5862b975207..250e6ef025a4 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -188,6 +188,8 @@ extern int sysctl_ip_dynaddr;
 
 extern void ipfrag_init(void);
 
+extern void ip_static_sysctl_init(void);
+
 #ifdef CONFIG_INET
 #include <net/dst.h>
 
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index f440a9f54924..354f6b54e492 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1439,6 +1439,10 @@ static int __init inet_init(void)
 
 	(void)sock_register(&inet_family_ops);
 
+#ifdef CONFIG_SYSCTL
+	ip_static_sysctl_init();
+#endif
+
 	/*
 	 *	Add all the base protocols.
 	 */
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 14ef202a2254..d63e9388d92d 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -882,4 +882,11 @@ static __init int sysctl_ipv4_init(void)
 	return 0;
 }
 
+/* set enough of tree skeleton to get rid of ordering problems */
+void __init ip_static_sysctl_init(void)
+{
+	static ctl_table table[1];
+	register_sysctl_paths(net_ipv4_ctl_path, table);
+}
+
 __initcall(sysctl_ipv4_init);

From ae7edecc9b8810770a8e5cb9a466ea4bdcfa8401 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 15 Jul 2008 06:33:31 -0400
Subject: [PATCH 1132/1171] [PATCH] sysctl: keep track of tree relationships

In a sense, that's the heart of the series.  It's based on the following
property of the trees we are actually asked to add: they can be split into
stem that is already covered by registered trees and crown that is entirely
new.  IOW, if a/b and a/c/d are introduced by our tree, then a/c is also
introduced by it.

That allows to associate tree and table entry with each node in the union;
while directory nodes might be covered by many trees, only one will cover
the node by its crown.  And that will allow much saner logics for /proc/sys
in the next patches.  This patch introduces the data structures needed to
keep track of that.

When adding a sysctl table, we find a "parent" one.  Which is to say,
find the deepest node on its stem that already is present in one of the
tables from our table set or its ancestor sets.  That table will be our
parent and that node in it - attachment point.  Add our table to list
anchored in parent, have it refer the parent and contents of attachment
point.  Also remember where its crown lives.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/linux/sysctl.h |  3 ++
 kernel/sysctl.c        | 63 ++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 66 insertions(+)

diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 956264d09ba0..3f6599aeb0db 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -1083,6 +1083,9 @@ struct ctl_table_header
 	struct ctl_table *ctl_table_arg;
 	struct ctl_table_root *root;
 	struct ctl_table_set *set;
+	struct ctl_table *attached_by;
+	struct ctl_table *attached_to;
+	struct ctl_table_header *parent;
 };
 
 /* struct ctl_path describes where in the hierarchy a table is added */
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 60d9357e7172..c9a0af887033 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -1680,6 +1680,52 @@ static __init int sysctl_init(void)
 
 core_initcall(sysctl_init);
 
+static int is_branch_in(struct ctl_table *branch, struct ctl_table *table)
+{
+	struct ctl_table *p;
+	const char *s = branch->procname;
+
+	/* branch should have named subdirectory as its first element */
+	if (!s || !branch->child)
+		return 0;
+
+	/* ... and nothing else */
+	if (branch[1].procname || branch[1].ctl_name)
+		return 0;
+
+	/* table should contain subdirectory with the same name */
+	for (p = table; p->procname || p->ctl_name; p++) {
+		if (!p->child)
+			continue;
+		if (p->procname && strcmp(p->procname, s) == 0)
+			return 1;
+	}
+	return 0;
+}
+
+/* see if attaching q to p would be an improvement */
+static void try_attach(struct ctl_table_header *p, struct ctl_table_header *q)
+{
+	struct ctl_table *to = p->ctl_table, *by = q->ctl_table;
+	int is_better = 0;
+	int not_in_parent = !p->attached_by;
+
+	while (is_branch_in(by, to)) {
+		if (by == q->attached_by)
+			is_better = 1;
+		if (to == p->attached_by)
+			not_in_parent = 1;
+		by = by->child;
+		to = to->child;
+	}
+
+	if (is_better && not_in_parent) {
+		q->attached_by = by;
+		q->attached_to = to;
+		q->parent = p;
+	}
+}
+
 /**
  * __register_sysctl_paths - register a sysctl hierarchy
  * @root: List of sysctl headers to register on
@@ -1759,6 +1805,7 @@ struct ctl_table_header *__register_sysctl_paths(
 	struct ctl_table_header *header;
 	struct ctl_table *new, **prevp;
 	unsigned int n, npath;
+	struct ctl_table_set *set;
 
 	/* Count the path components */
 	for (npath = 0; path[npath].ctl_name || path[npath].procname; ++npath)
@@ -1809,6 +1856,18 @@ struct ctl_table_header *__register_sysctl_paths(
 #endif
 	spin_lock(&sysctl_lock);
 	header->set = lookup_header_set(root, namespaces);
+	header->attached_by = header->ctl_table;
+	header->attached_to = root_table;
+	header->parent = &root_table_header;
+	for (set = header->set; set; set = set->parent) {
+		struct ctl_table_header *p;
+		list_for_each_entry(p, &set->list, ctl_entry) {
+			if (p->unregistering)
+				continue;
+			try_attach(p, header);
+		}
+	}
+	header->parent->count++;
 	list_add_tail(&header->ctl_entry, &header->set->list);
 	spin_unlock(&sysctl_lock);
 
@@ -1864,6 +1923,10 @@ void unregister_sysctl_table(struct ctl_table_header * header)
 
 	spin_lock(&sysctl_lock);
 	start_unregistering(header);
+	if (!--header->parent->count) {
+		WARN_ON(1);
+		kfree(header->parent);
+	}
 	if (!--header->count)
 		kfree(header);
 	spin_unlock(&sysctl_lock);

From 9043476f726802f4b00c96d0c4f418dde48d1304 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 15 Jul 2008 08:54:06 -0400
Subject: [PATCH 1133/1171] [PATCH] sanitize proc_sysctl

* keep references to ctl_table_head and ctl_table in /proc/sys inodes
* grab the former during operations, use the latter for access to
  entry if that succeeds
* have ->d_compare() check if table should be seen for one who does lookup;
  that allows us to avoid flipping inodes - if we have the same name resolve
  to different things, we'll just keep several dentries and ->d_compare()
  will reject the wrong ones.
* have ->lookup() and ->readdir() scan the table of our inode first, then
  walk all ctl_table_header and scan ->attached_by for those that are
  attached to our directory.
* implement ->getattr().
* get rid of insane amounts of tree-walking
* get rid of the need to know dentry in ->permission() and of the contortions
  induced by that.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/proc/inode.c         |   5 +
 fs/proc/proc_sysctl.c   | 429 ++++++++++++++++++----------------------
 include/linux/proc_fs.h |   5 +
 include/linux/sysctl.h  |   1 +
 kernel/sysctl.c         |  15 ++
 5 files changed, 219 insertions(+), 236 deletions(-)

diff --git a/fs/proc/inode.c b/fs/proc/inode.c
index b37f25dc45a5..8bb03f056c28 100644
--- a/fs/proc/inode.c
+++ b/fs/proc/inode.c
@@ -17,6 +17,7 @@
 #include <linux/init.h>
 #include <linux/module.h>
 #include <linux/smp_lock.h>
+#include <linux/sysctl.h>
 
 #include <asm/system.h>
 #include <asm/uaccess.h>
@@ -65,6 +66,8 @@ static void proc_delete_inode(struct inode *inode)
 			module_put(de->owner);
 		de_put(de);
 	}
+	if (PROC_I(inode)->sysctl)
+		sysctl_head_put(PROC_I(inode)->sysctl);
 	clear_inode(inode);
 }
 
@@ -84,6 +87,8 @@ static struct inode *proc_alloc_inode(struct super_block *sb)
 	ei->fd = 0;
 	ei->op.proc_get_link = NULL;
 	ei->pde = NULL;
+	ei->sysctl = NULL;
+	ei->sysctl_entry = NULL;
 	inode = &ei->vfs_inode;
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
 	return inode;
diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c
index 5acc001d49f6..fa1ec2433e44 100644
--- a/fs/proc/proc_sysctl.c
+++ b/fs/proc/proc_sysctl.c
@@ -10,149 +10,110 @@
 static struct dentry_operations proc_sys_dentry_operations;
 static const struct file_operations proc_sys_file_operations;
 static const struct inode_operations proc_sys_inode_operations;
+static const struct file_operations proc_sys_dir_file_operations;
+static const struct inode_operations proc_sys_dir_operations;
 
-static void proc_sys_refresh_inode(struct inode *inode, struct ctl_table *table)
-{
-	/* Refresh the cached information bits in the inode */
-	if (table) {
-		inode->i_uid = 0;
-		inode->i_gid = 0;
-		inode->i_mode = table->mode;
-		if (table->proc_handler) {
-			inode->i_mode |= S_IFREG;
-			inode->i_nlink = 1;
-		} else {
-			inode->i_mode |= S_IFDIR;
-			inode->i_nlink = 0;	/* It is too hard to figure out */
-		}
-	}
-}
-
-static struct inode *proc_sys_make_inode(struct inode *dir, struct ctl_table *table)
+static struct inode *proc_sys_make_inode(struct super_block *sb,
+		struct ctl_table_header *head, struct ctl_table *table)
 {
 	struct inode *inode;
-	struct proc_inode *dir_ei, *ei;
-	int depth;
+	struct proc_inode *ei;
 
-	inode = new_inode(dir->i_sb);
+	inode = new_inode(sb);
 	if (!inode)
 		goto out;
 
-	/* A directory is always one deeper than it's parent */
-	dir_ei = PROC_I(dir);
-	depth = dir_ei->fd + 1;
-
+	sysctl_head_get(head);
 	ei = PROC_I(inode);
-	ei->fd = depth;
+	ei->sysctl = head;
+	ei->sysctl_entry = table;
+
 	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
-	inode->i_op = &proc_sys_inode_operations;
-	inode->i_fop = &proc_sys_file_operations;
 	inode->i_flags |= S_PRIVATE; /* tell selinux to ignore this inode */
-	proc_sys_refresh_inode(inode, table);
+	inode->i_mode = table->mode;
+	if (!table->child) {
+		inode->i_mode |= S_IFREG;
+		inode->i_op = &proc_sys_inode_operations;
+		inode->i_fop = &proc_sys_file_operations;
+	} else {
+		inode->i_mode |= S_IFDIR;
+		inode->i_nlink = 0;
+		inode->i_op = &proc_sys_dir_operations;
+		inode->i_fop = &proc_sys_dir_file_operations;
+	}
 out:
 	return inode;
 }
 
-static struct dentry *proc_sys_ancestor(struct dentry *dentry, int depth)
-{
-	for (;;) {
-		struct proc_inode *ei;
-
-		ei = PROC_I(dentry->d_inode);
-		if (ei->fd == depth)
-			break; /* found */
-
-		dentry = dentry->d_parent;
-	}
-	return dentry;
-}
-
-static struct ctl_table *proc_sys_lookup_table_one(struct ctl_table *table,
-							struct qstr *name)
+static struct ctl_table *find_in_table(struct ctl_table *p, struct qstr *name)
 {
 	int len;
-	for ( ; table->ctl_name || table->procname; table++) {
+	for ( ; p->ctl_name || p->procname; p++) {
 
-		if (!table->procname)
+		if (!p->procname)
 			continue;
 
-		len = strlen(table->procname);
+		len = strlen(p->procname);
 		if (len != name->len)
 			continue;
 
-		if (memcmp(table->procname, name->name, len) != 0)
+		if (memcmp(p->procname, name->name, len) != 0)
 			continue;
 
 		/* I have a match */
-		return table;
+		return p;
 	}
 	return NULL;
 }
 
-static struct ctl_table *proc_sys_lookup_table(struct dentry *dentry,
-						struct ctl_table *table)
+struct ctl_table_header *grab_header(struct inode *inode)
 {
-	struct dentry *ancestor;
-	struct proc_inode *ei;
-	int depth, i;
-
-	ei = PROC_I(dentry->d_inode);
-	depth = ei->fd;
-
-	if (depth == 0)
-		return table;
-
-	for (i = 1; table && (i <= depth); i++) {
-		ancestor = proc_sys_ancestor(dentry, i);
-		table = proc_sys_lookup_table_one(table, &ancestor->d_name);
-		if (table)
-			table = table->child;
-	}
-	return table;
-
-}
-static struct ctl_table *proc_sys_lookup_entry(struct dentry *dparent,
-						struct qstr *name,
-						struct ctl_table *table)
-{
-	table = proc_sys_lookup_table(dparent, table);
-	if (table)
-		table = proc_sys_lookup_table_one(table, name);
-	return table;
-}
-
-static struct ctl_table *do_proc_sys_lookup(struct dentry *parent,
-						struct qstr *name,
-						struct ctl_table_header **ptr)
-{
-	struct ctl_table_header *head;
-	struct ctl_table *table = NULL;
-
-	for (head = sysctl_head_next(NULL); head;
-			head = sysctl_head_next(head)) {
-		table = proc_sys_lookup_entry(parent, name, head->ctl_table);
-		if (table)
-			break;
-	}
-	*ptr = head;
-	return table;
+	if (PROC_I(inode)->sysctl)
+		return sysctl_head_grab(PROC_I(inode)->sysctl);
+	else
+		return sysctl_head_next(NULL);
 }
 
 static struct dentry *proc_sys_lookup(struct inode *dir, struct dentry *dentry,
 					struct nameidata *nd)
 {
-	struct ctl_table_header *head;
+	struct ctl_table_header *head = grab_header(dir);
+	struct ctl_table *table = PROC_I(dir)->sysctl_entry;
+	struct ctl_table_header *h = NULL;
+	struct qstr *name = &dentry->d_name;
+	struct ctl_table *p;
 	struct inode *inode;
-	struct dentry *err;
-	struct ctl_table *table;
+	struct dentry *err = ERR_PTR(-ENOENT);
 
-	err = ERR_PTR(-ENOENT);
-	table = do_proc_sys_lookup(dentry->d_parent, &dentry->d_name, &head);
-	if (!table)
+	if (IS_ERR(head))
+		return ERR_CAST(head);
+
+	if (table && !table->child) {
+		WARN_ON(1);
+		goto out;
+	}
+
+	table = table ? table->child : head->ctl_table;
+
+	p = find_in_table(table, name);
+	if (!p) {
+		for (h = sysctl_head_next(NULL); h; h = sysctl_head_next(h)) {
+			if (h->attached_to != table)
+				continue;
+			p = find_in_table(h->attached_by, name);
+			if (p)
+				break;
+		}
+	}
+
+	if (!p)
 		goto out;
 
 	err = ERR_PTR(-ENOMEM);
-	inode = proc_sys_make_inode(dir, table);
+	inode = proc_sys_make_inode(dir->i_sb, h ? h : head, p);
+	if (h)
+		sysctl_head_finish(h);
+
 	if (!inode)
 		goto out;
 
@@ -168,22 +129,14 @@ out:
 static ssize_t proc_sys_call_handler(struct file *filp, void __user *buf,
 		size_t count, loff_t *ppos, int write)
 {
-	struct dentry *dentry = filp->f_dentry;
-	struct ctl_table_header *head;
-	struct ctl_table *table;
+	struct inode *inode = filp->f_path.dentry->d_inode;
+	struct ctl_table_header *head = grab_header(inode);
+	struct ctl_table *table = PROC_I(inode)->sysctl_entry;
 	ssize_t error;
 	size_t res;
 
-	table = do_proc_sys_lookup(dentry->d_parent, &dentry->d_name, &head);
-	/* Has the sysctl entry disappeared on us? */
-	error = -ENOENT;
-	if (!table)
-		goto out;
-
-	/* Has the sysctl entry been replaced by a directory? */
-	error = -EISDIR;
-	if (!table->proc_handler)
-		goto out;
+	if (IS_ERR(head))
+		return PTR_ERR(head);
 
 	/*
 	 * At this point we know that the sysctl was not unregistered
@@ -193,6 +146,11 @@ static ssize_t proc_sys_call_handler(struct file *filp, void __user *buf,
 	if (sysctl_perm(head->root, table, write ? MAY_WRITE : MAY_READ))
 		goto out;
 
+	/* if that can happen at all, it should be -EINVAL, not -EISDIR */
+	error = -EINVAL;
+	if (!table->proc_handler)
+		goto out;
+
 	/* careful: calling conventions are nasty here */
 	res = count;
 	error = table->proc_handler(table, write, filp, buf, &res, ppos);
@@ -218,82 +176,86 @@ static ssize_t proc_sys_write(struct file *filp, const char __user *buf,
 
 
 static int proc_sys_fill_cache(struct file *filp, void *dirent,
-				filldir_t filldir, struct ctl_table *table)
+				filldir_t filldir,
+				struct ctl_table_header *head,
+				struct ctl_table *table)
 {
-	struct ctl_table_header *head;
-	struct ctl_table *child_table = NULL;
 	struct dentry *child, *dir = filp->f_path.dentry;
 	struct inode *inode;
 	struct qstr qname;
 	ino_t ino = 0;
 	unsigned type = DT_UNKNOWN;
-	int ret;
 
 	qname.name = table->procname;
 	qname.len  = strlen(table->procname);
 	qname.hash = full_name_hash(qname.name, qname.len);
 
-	/* Suppress duplicates.
-	 * Only fill a directory entry if it is the value that
-	 * an ordinary lookup of that name returns.  Hide all
-	 * others.
-	 *
-	 * If we ever cache this translation in the dcache
-	 * I should do a dcache lookup first.  But for now
-	 * it is just simpler not to.
-	 */
-	ret = 0;
-	child_table = do_proc_sys_lookup(dir, &qname, &head);
-	sysctl_head_finish(head);
-	if (child_table != table)
-		return 0;
-
 	child = d_lookup(dir, &qname);
 	if (!child) {
-		struct dentry *new;
-		new = d_alloc(dir, &qname);
-		if (new) {
-			inode = proc_sys_make_inode(dir->d_inode, table);
-			if (!inode)
-				child = ERR_PTR(-ENOMEM);
-			else {
-				new->d_op = &proc_sys_dentry_operations;
-				d_add(new, inode);
+		child = d_alloc(dir, &qname);
+		if (child) {
+			inode = proc_sys_make_inode(dir->d_sb, head, table);
+			if (!inode) {
+				dput(child);
+				return -ENOMEM;
+			} else {
+				child->d_op = &proc_sys_dentry_operations;
+				d_add(child, inode);
 			}
-			if (child)
-				dput(new);
-			else
-				child = new;
+		} else {
+			return -ENOMEM;
 		}
 	}
-	if (!child || IS_ERR(child) || !child->d_inode)
-		goto end_instantiate;
 	inode = child->d_inode;
-	if (inode) {
-		ino  = inode->i_ino;
-		type = inode->i_mode >> 12;
-	}
+	ino  = inode->i_ino;
+	type = inode->i_mode >> 12;
 	dput(child);
-end_instantiate:
-	if (!ino)
-		ino= find_inode_number(dir, &qname);
-	if (!ino)
-		ino = 1;
-	return filldir(dirent, qname.name, qname.len, filp->f_pos, ino, type);
+	return !!filldir(dirent, qname.name, qname.len, filp->f_pos, ino, type);
+}
+
+static int scan(struct ctl_table_header *head, ctl_table *table,
+		unsigned long *pos, struct file *file,
+		void *dirent, filldir_t filldir)
+{
+
+	for (; table->ctl_name || table->procname; table++, (*pos)++) {
+		int res;
+
+		/* Can't do anything without a proc name */
+		if (!table->procname)
+			continue;
+
+		if (*pos < file->f_pos)
+			continue;
+
+		res = proc_sys_fill_cache(file, dirent, filldir, head, table);
+		if (res)
+			return res;
+
+		file->f_pos = *pos + 1;
+	}
+	return 0;
 }
 
 static int proc_sys_readdir(struct file *filp, void *dirent, filldir_t filldir)
 {
-	struct dentry *dentry = filp->f_dentry;
+	struct dentry *dentry = filp->f_path.dentry;
 	struct inode *inode = dentry->d_inode;
-	struct ctl_table_header *head = NULL;
-	struct ctl_table *table;
+	struct ctl_table_header *head = grab_header(inode);
+	struct ctl_table *table = PROC_I(inode)->sysctl_entry;
+	struct ctl_table_header *h = NULL;
 	unsigned long pos;
-	int ret;
+	int ret = -EINVAL;
 
-	ret = -ENOTDIR;
-	if (!S_ISDIR(inode->i_mode))
+	if (IS_ERR(head))
+		return PTR_ERR(head);
+
+	if (table && !table->child) {
+		WARN_ON(1);
 		goto out;
+	}
+
+	table = table ? table->child : head->ctl_table;
 
 	ret = 0;
 	/* Avoid a switch here: arm builds fail with missing __cmpdi2 */
@@ -311,30 +273,17 @@ static int proc_sys_readdir(struct file *filp, void *dirent, filldir_t filldir)
 	}
 	pos = 2;
 
-	/* - Find each instance of the directory
-	 * - Read all entries in each instance
-	 * - Before returning an entry to user space lookup the entry
-	 *   by name and if I find a different entry don't return
-	 *   this one because it means it is a buried dup.
-	 * For sysctl this should only happen for directory entries.
-	 */
-	for (head = sysctl_head_next(NULL); head; head = sysctl_head_next(head)) {
-		table = proc_sys_lookup_table(dentry, head->ctl_table);
+	ret = scan(head, table, &pos, filp, dirent, filldir);
+	if (ret)
+		goto out;
 
-		if (!table)
+	for (h = sysctl_head_next(NULL); h; h = sysctl_head_next(h)) {
+		if (h->attached_to != table)
 			continue;
-
-		for (; table->ctl_name || table->procname; table++, pos++) {
-			/* Can't do anything without a proc name */
-			if (!table->procname)
-				continue;
-
-			if (pos < filp->f_pos)
-				continue;
-
-			if (proc_sys_fill_cache(filp, dirent, filldir, table) < 0)
-				goto out;
-			filp->f_pos = pos + 1;
+		ret = scan(h, h->attached_by, &pos, filp, dirent, filldir);
+		if (ret) {
+			sysctl_head_finish(h);
+			break;
 		}
 	}
 	ret = 1;
@@ -349,47 +298,18 @@ static int proc_sys_permission(struct inode *inode, int mask, struct nameidata *
 	 * sysctl entries that are not writeable,
 	 * are _NOT_ writeable, capabilities or not.
 	 */
-	struct ctl_table_header *head;
-	struct ctl_table *table;
-	struct dentry *dentry;
-	int mode;
-	int depth;
+	struct ctl_table_header *head = grab_header(inode);
+	struct ctl_table *table = PROC_I(inode)->sysctl_entry;
 	int error;
 
-	head = NULL;
-	depth = PROC_I(inode)->fd;
+	if (IS_ERR(head))
+		return PTR_ERR(head);
 
-	/* First check the cached permissions, in case we don't have
-	 * enough information to lookup the sysctl table entry.
-	 */
-	error = -EACCES;
-	mode = inode->i_mode;
+	if (!table) /* global root - r-xr-xr-x */
+		error = mask & MAY_WRITE ? -EACCES : 0;
+	else /* Use the permissions on the sysctl table entry */
+		error = sysctl_perm(head->root, table, mask);
 
-	if (current->euid == 0)
-		mode >>= 6;
-	else if (in_group_p(0))
-		mode >>= 3;
-
-	if ((mode & mask & (MAY_READ|MAY_WRITE|MAY_EXEC)) == mask)
-		error = 0;
-
-	/* If we can't get a sysctl table entry the permission
-	 * checks on the cached mode will have to be enough.
-	 */
-	if (!nd || !depth)
-		goto out;
-
-	dentry = nd->path.dentry;
-	table = do_proc_sys_lookup(dentry->d_parent, &dentry->d_name, &head);
-
-	/* If the entry does not exist deny permission */
-	error = -EACCES;
-	if (!table)
-		goto out;
-
-	/* Use the permissions on the sysctl table entry */
-	error = sysctl_perm(head->root, table, mask);
-out:
 	sysctl_head_finish(head);
 	return error;
 }
@@ -409,33 +329,70 @@ static int proc_sys_setattr(struct dentry *dentry, struct iattr *attr)
 	return error;
 }
 
-/* I'm lazy and don't distinguish between files and directories,
- * until access time.
- */
+static int proc_sys_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat)
+{
+	struct inode *inode = dentry->d_inode;
+	struct ctl_table_header *head = grab_header(inode);
+	struct ctl_table *table = PROC_I(inode)->sysctl_entry;
+
+	if (IS_ERR(head))
+		return PTR_ERR(head);
+
+	generic_fillattr(inode, stat);
+	if (table)
+		stat->mode = (stat->mode & S_IFMT) | table->mode;
+
+	sysctl_head_finish(head);
+	return 0;
+}
+
 static const struct file_operations proc_sys_file_operations = {
 	.read		= proc_sys_read,
 	.write		= proc_sys_write,
+};
+
+static const struct file_operations proc_sys_dir_file_operations = {
 	.readdir	= proc_sys_readdir,
 };
 
 static const struct inode_operations proc_sys_inode_operations = {
+	.permission	= proc_sys_permission,
+	.setattr	= proc_sys_setattr,
+	.getattr	= proc_sys_getattr,
+};
+
+static const struct inode_operations proc_sys_dir_operations = {
 	.lookup		= proc_sys_lookup,
 	.permission	= proc_sys_permission,
 	.setattr	= proc_sys_setattr,
+	.getattr	= proc_sys_getattr,
 };
 
 static int proc_sys_revalidate(struct dentry *dentry, struct nameidata *nd)
 {
-	struct ctl_table_header *head;
-	struct ctl_table *table;
-	table = do_proc_sys_lookup(dentry->d_parent, &dentry->d_name, &head);
-	proc_sys_refresh_inode(dentry->d_inode, table);
-	sysctl_head_finish(head);
-	return !!table;
+	return !PROC_I(dentry->d_inode)->sysctl->unregistering;
+}
+
+static int proc_sys_delete(struct dentry *dentry)
+{
+	return !!PROC_I(dentry->d_inode)->sysctl->unregistering;
+}
+
+static int proc_sys_compare(struct dentry *dir, struct qstr *qstr,
+			    struct qstr *name)
+{
+	struct dentry *dentry = container_of(qstr, struct dentry, d_name);
+	if (qstr->len != name->len)
+		return 1;
+	if (memcmp(qstr->name, name->name, name->len))
+		return 1;
+	return !sysctl_is_seen(PROC_I(dentry->d_inode)->sysctl);
 }
 
 static struct dentry_operations proc_sys_dentry_operations = {
 	.d_revalidate	= proc_sys_revalidate,
+	.d_delete	= proc_sys_delete,
+	.d_compare	= proc_sys_compare,
 };
 
 static struct proc_dir_entry *proc_sys_root;
@@ -443,8 +400,8 @@ static struct proc_dir_entry *proc_sys_root;
 int proc_sys_init(void)
 {
 	proc_sys_root = proc_mkdir("sys", NULL);
-	proc_sys_root->proc_iops = &proc_sys_inode_operations;
-	proc_sys_root->proc_fops = &proc_sys_file_operations;
+	proc_sys_root->proc_iops = &proc_sys_dir_operations;
+	proc_sys_root->proc_fops = &proc_sys_dir_file_operations;
 	proc_sys_root->nlink = 0;
 	return 0;
 }
diff --git a/include/linux/proc_fs.h b/include/linux/proc_fs.h
index f560d1705afe..fb61850d1cfc 100644
--- a/include/linux/proc_fs.h
+++ b/include/linux/proc_fs.h
@@ -282,11 +282,16 @@ union proc_op {
 		struct task_struct *task);
 };
 
+struct ctl_table_header;
+struct ctl_table;
+
 struct proc_inode {
 	struct pid *pid;
 	int fd;
 	union proc_op op;
 	struct proc_dir_entry *pde;
+	struct ctl_table_header *sysctl;
+	struct ctl_table *sysctl_entry;
 	struct inode vfs_inode;
 };
 
diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 3f6599aeb0db..d0437f36921f 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -961,6 +961,7 @@ struct ctl_table_header;
 
 extern void sysctl_head_get(struct ctl_table_header *);
 extern void sysctl_head_put(struct ctl_table_header *);
+extern int sysctl_is_seen(struct ctl_table_header *);
 extern struct ctl_table_header *sysctl_head_grab(struct ctl_table_header *);
 extern struct ctl_table_header *sysctl_head_next(struct ctl_table_header *prev);
 extern struct ctl_table_header *__sysctl_head_next(struct nsproxy *namespaces,
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index c9a0af887033..ff5abcca5ddf 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -1932,6 +1932,21 @@ void unregister_sysctl_table(struct ctl_table_header * header)
 	spin_unlock(&sysctl_lock);
 }
 
+int sysctl_is_seen(struct ctl_table_header *p)
+{
+	struct ctl_table_set *set = p->set;
+	int res;
+	spin_lock(&sysctl_lock);
+	if (p->unregistering)
+		res = 0;
+	else if (!set->is_seen)
+		res = 1;
+	else
+		res = set->is_seen(set);
+	spin_unlock(&sysctl_lock);
+	return res;
+}
+
 void setup_sysctl_set(struct ctl_table_set *p,
 	struct ctl_table_set *parent,
 	int (*is_seen)(struct ctl_table_set *))

From 1bd5191d9f5d1928c4efdf604c4164b04bb88dbe Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Wed, 21 May 2008 19:15:03 +0200
Subject: [PATCH 1134/1171] [patch 05/14] hpfs: dont call permission()

hpfs_unlink() calls permission() prior to truncating the file.  HPFS
doesn't define a .permission method, so replace with explicit call to
generic_permission().

This is equivalent, except that devcgroup_inode_permission() and
security_inode_permission() are not called.

The truncation is just an implementation detail of the unlink, so
these security checks are unnecessary.

I suspect that even calling generic_permission() is unnecessary, since
we shouldn't mind if the file isn't writable.  But I leave that to the
maintainer to decide.

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
CC: Mikulas Patocka <mikulas@artax.karlin.mff.cuni.cz>
---
 fs/hpfs/namei.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/hpfs/namei.c b/fs/hpfs/namei.c
index d256559b4104..d9c59a775449 100644
--- a/fs/hpfs/namei.c
+++ b/fs/hpfs/namei.c
@@ -415,7 +415,7 @@ again:
 		d_drop(dentry);
 		spin_lock(&dentry->d_lock);
 		if (atomic_read(&dentry->d_count) > 1 ||
-		    permission(inode, MAY_WRITE, NULL) ||
+		    generic_permission(inode, MAY_WRITE, NULL) ||
 		    !S_ISREG(inode->i_mode) ||
 		    get_write_access(inode)) {
 			spin_unlock(&dentry->d_lock);

From e6305c43eda10ebfd2ad9e35d6e172ccc7bb3695 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 15 Jul 2008 21:03:57 -0400
Subject: [PATCH 1135/1171] [PATCH] sanitize ->permission() prototype

* kill nameidata * argument; map the 3 bits in ->flags anybody cares
  about to new MAY_... ones and pass with the mask.
* kill redundant gfs2_iop_permission()
* sanitize ecryptfs_permission()
* fix remaining places where ->permission() instances might barf on new
  MAY_... found in mask.

The obvious next target in that direction is permission(9)

folded fix for nfs_permission() breakage from Miklos Szeredi <mszeredi@suse.cz>

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/afs/internal.h              |  4 +---
 fs/afs/security.c              |  2 +-
 fs/bad_inode.c                 |  3 +--
 fs/cifs/cifsfs.c               |  2 +-
 fs/coda/dir.c                  |  4 +++-
 fs/coda/pioctl.c               |  6 ++----
 fs/ecryptfs/inode.c            | 17 ++---------------
 fs/ext2/acl.c                  |  2 +-
 fs/ext2/acl.h                  |  2 +-
 fs/ext3/acl.c                  |  2 +-
 fs/ext3/acl.h                  |  2 +-
 fs/ext4/acl.c                  |  2 +-
 fs/ext4/acl.h                  |  2 +-
 fs/fuse/dir.c                  |  6 +++---
 fs/gfs2/ops_inode.c            | 12 +++---------
 fs/hfs/inode.c                 |  3 +--
 fs/hfsplus/inode.c             |  2 +-
 fs/hostfs/hostfs_kern.c        |  2 +-
 fs/jffs2/acl.c                 |  2 +-
 fs/jffs2/acl.h                 |  2 +-
 fs/jfs/acl.c                   |  2 +-
 fs/jfs/jfs_acl.h               |  2 +-
 fs/namei.c                     | 23 +++++++++++++++++------
 fs/nfs/dir.c                   | 11 +++++------
 fs/ocfs2/file.c                |  2 +-
 fs/ocfs2/file.h                |  3 +--
 fs/proc/base.c                 |  3 +--
 fs/proc/proc_sysctl.c          |  2 +-
 fs/reiserfs/xattr.c            |  2 +-
 fs/smbfs/file.c                |  4 ++--
 fs/xfs/linux-2.6/xfs_iops.c    |  3 +--
 include/linux/coda_linux.h     |  2 +-
 include/linux/fs.h             |  5 ++++-
 include/linux/nfs_fs.h         |  2 +-
 include/linux/reiserfs_xattr.h |  2 +-
 include/linux/shmem_fs.h       |  2 +-
 kernel/sysctl.c                | 10 +++++-----
 mm/shmem_acl.c                 |  2 +-
 38 files changed, 74 insertions(+), 87 deletions(-)

diff --git a/fs/afs/internal.h b/fs/afs/internal.h
index 7102824ba847..3cb6920ff30b 100644
--- a/fs/afs/internal.h
+++ b/fs/afs/internal.h
@@ -469,8 +469,6 @@ extern bool afs_cm_incoming_call(struct afs_call *);
 extern const struct inode_operations afs_dir_inode_operations;
 extern const struct file_operations afs_dir_file_operations;
 
-extern int afs_permission(struct inode *, int, struct nameidata *);
-
 /*
  * file.c
  */
@@ -605,7 +603,7 @@ extern void afs_clear_permits(struct afs_vnode *);
 extern void afs_cache_permit(struct afs_vnode *, struct key *, long);
 extern void afs_zap_permits(struct rcu_head *);
 extern struct key *afs_request_key(struct afs_cell *);
-extern int afs_permission(struct inode *, int, struct nameidata *);
+extern int afs_permission(struct inode *, int);
 
 /*
  * server.c
diff --git a/fs/afs/security.c b/fs/afs/security.c
index 3bcbeceba1bb..3ef504370034 100644
--- a/fs/afs/security.c
+++ b/fs/afs/security.c
@@ -284,7 +284,7 @@ static int afs_check_permit(struct afs_vnode *vnode, struct key *key,
  * - AFS ACLs are attached to directories only, and a file is controlled by its
  *   parent directory's ACL
  */
-int afs_permission(struct inode *inode, int mask, struct nameidata *nd)
+int afs_permission(struct inode *inode, int mask)
 {
 	struct afs_vnode *vnode = AFS_FS_I(inode);
 	afs_access_t uninitialized_var(access);
diff --git a/fs/bad_inode.c b/fs/bad_inode.c
index f1c2ea8342f5..5f1538c03b1b 100644
--- a/fs/bad_inode.c
+++ b/fs/bad_inode.c
@@ -243,8 +243,7 @@ static int bad_inode_readlink(struct dentry *dentry, char __user *buffer,
 	return -EIO;
 }
 
-static int bad_inode_permission(struct inode *inode, int mask,
-			struct nameidata *nd)
+static int bad_inode_permission(struct inode *inode, int mask)
 {
 	return -EIO;
 }
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index fe5f6809cba6..1ec7076f7b24 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -267,7 +267,7 @@ cifs_statfs(struct dentry *dentry, struct kstatfs *buf)
 	return 0;
 }
 
-static int cifs_permission(struct inode *inode, int mask, struct nameidata *nd)
+static int cifs_permission(struct inode *inode, int mask)
 {
 	struct cifs_sb_info *cifs_sb;
 
diff --git a/fs/coda/dir.c b/fs/coda/dir.c
index 3d2580e00a3e..c5916228243c 100644
--- a/fs/coda/dir.c
+++ b/fs/coda/dir.c
@@ -137,9 +137,11 @@ exit:
 }
 
 
-int coda_permission(struct inode *inode, int mask, struct nameidata *nd)
+int coda_permission(struct inode *inode, int mask)
 {
         int error = 0;
+
+	mask &= MAY_READ | MAY_WRITE | MAY_EXEC;
  
 	if (!mask)
 		return 0; 
diff --git a/fs/coda/pioctl.c b/fs/coda/pioctl.c
index c21a1f552a63..c38a98974fb0 100644
--- a/fs/coda/pioctl.c
+++ b/fs/coda/pioctl.c
@@ -24,8 +24,7 @@
 #include <linux/coda_psdev.h>
 
 /* pioctl ops */
-static int coda_ioctl_permission(struct inode *inode, int mask,
-				 struct nameidata *nd);
+static int coda_ioctl_permission(struct inode *inode, int mask);
 static int coda_pioctl(struct inode * inode, struct file * filp, 
                        unsigned int cmd, unsigned long user_data);
 
@@ -42,8 +41,7 @@ const struct file_operations coda_ioctl_operations = {
 };
 
 /* the coda pioctl inode ops */
-static int coda_ioctl_permission(struct inode *inode, int mask,
-				 struct nameidata *nd)
+static int coda_ioctl_permission(struct inode *inode, int mask)
 {
         return 0;
 }
diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c
index d755455e3bff..32f4228efcd5 100644
--- a/fs/ecryptfs/inode.c
+++ b/fs/ecryptfs/inode.c
@@ -830,22 +830,9 @@ out:
 }
 
 static int
-ecryptfs_permission(struct inode *inode, int mask, struct nameidata *nd)
+ecryptfs_permission(struct inode *inode, int mask)
 {
-	int rc;
-
-        if (nd) {
-		struct vfsmount *vfsmnt_save = nd->path.mnt;
-		struct dentry *dentry_save = nd->path.dentry;
-
-		nd->path.mnt = ecryptfs_dentry_to_lower_mnt(nd->path.dentry);
-		nd->path.dentry = ecryptfs_dentry_to_lower(nd->path.dentry);
-		rc = permission(ecryptfs_inode_to_lower(inode), mask, nd);
-		nd->path.mnt = vfsmnt_save;
-		nd->path.dentry = dentry_save;
-        } else
-		rc = permission(ecryptfs_inode_to_lower(inode), mask, NULL);
-        return rc;
+	return permission(ecryptfs_inode_to_lower(inode), mask, NULL);
 }
 
 /**
diff --git a/fs/ext2/acl.c b/fs/ext2/acl.c
index e58669e1b87c..ae8c4f850b27 100644
--- a/fs/ext2/acl.c
+++ b/fs/ext2/acl.c
@@ -294,7 +294,7 @@ ext2_check_acl(struct inode *inode, int mask)
 }
 
 int
-ext2_permission(struct inode *inode, int mask, struct nameidata *nd)
+ext2_permission(struct inode *inode, int mask)
 {
 	return generic_permission(inode, mask, ext2_check_acl);
 }
diff --git a/fs/ext2/acl.h b/fs/ext2/acl.h
index 0bde85bafe38..b42cf578554b 100644
--- a/fs/ext2/acl.h
+++ b/fs/ext2/acl.h
@@ -58,7 +58,7 @@ static inline int ext2_acl_count(size_t size)
 #define EXT2_ACL_NOT_CACHED ((void *)-1)
 
 /* acl.c */
-extern int ext2_permission (struct inode *, int, struct nameidata *);
+extern int ext2_permission (struct inode *, int);
 extern int ext2_acl_chmod (struct inode *);
 extern int ext2_init_acl (struct inode *, struct inode *);
 
diff --git a/fs/ext3/acl.c b/fs/ext3/acl.c
index a754d1848173..b60bb241880c 100644
--- a/fs/ext3/acl.c
+++ b/fs/ext3/acl.c
@@ -299,7 +299,7 @@ ext3_check_acl(struct inode *inode, int mask)
 }
 
 int
-ext3_permission(struct inode *inode, int mask, struct nameidata *nd)
+ext3_permission(struct inode *inode, int mask)
 {
 	return generic_permission(inode, mask, ext3_check_acl);
 }
diff --git a/fs/ext3/acl.h b/fs/ext3/acl.h
index 0d1e6279cbfd..42da16b8cac0 100644
--- a/fs/ext3/acl.h
+++ b/fs/ext3/acl.h
@@ -58,7 +58,7 @@ static inline int ext3_acl_count(size_t size)
 #define EXT3_ACL_NOT_CACHED ((void *)-1)
 
 /* acl.c */
-extern int ext3_permission (struct inode *, int, struct nameidata *);
+extern int ext3_permission (struct inode *, int);
 extern int ext3_acl_chmod (struct inode *);
 extern int ext3_init_acl (handle_t *, struct inode *, struct inode *);
 
diff --git a/fs/ext4/acl.c b/fs/ext4/acl.c
index 3c8dab880d91..c7d04e165446 100644
--- a/fs/ext4/acl.c
+++ b/fs/ext4/acl.c
@@ -299,7 +299,7 @@ ext4_check_acl(struct inode *inode, int mask)
 }
 
 int
-ext4_permission(struct inode *inode, int mask, struct nameidata *nd)
+ext4_permission(struct inode *inode, int mask)
 {
 	return generic_permission(inode, mask, ext4_check_acl);
 }
diff --git a/fs/ext4/acl.h b/fs/ext4/acl.h
index 26a5c1abf147..cd2b855a07d6 100644
--- a/fs/ext4/acl.h
+++ b/fs/ext4/acl.h
@@ -58,7 +58,7 @@ static inline int ext4_acl_count(size_t size)
 #define EXT4_ACL_NOT_CACHED ((void *)-1)
 
 /* acl.c */
-extern int ext4_permission (struct inode *, int, struct nameidata *);
+extern int ext4_permission (struct inode *, int);
 extern int ext4_acl_chmod (struct inode *);
 extern int ext4_init_acl (handle_t *, struct inode *, struct inode *);
 
diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index 51d0035ff07e..48a7934cb950 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -898,7 +898,7 @@ static int fuse_access(struct inode *inode, int mask)
 		return PTR_ERR(req);
 
 	memset(&inarg, 0, sizeof(inarg));
-	inarg.mask = mask;
+	inarg.mask = mask & (MAY_READ | MAY_WRITE | MAY_EXEC);
 	req->in.h.opcode = FUSE_ACCESS;
 	req->in.h.nodeid = get_node_id(inode);
 	req->in.numargs = 1;
@@ -927,7 +927,7 @@ static int fuse_access(struct inode *inode, int mask)
  * access request is sent.  Execute permission is still checked
  * locally based on file mode.
  */
-static int fuse_permission(struct inode *inode, int mask, struct nameidata *nd)
+static int fuse_permission(struct inode *inode, int mask)
 {
 	struct fuse_conn *fc = get_fuse_conn(inode);
 	bool refreshed = false;
@@ -962,7 +962,7 @@ static int fuse_permission(struct inode *inode, int mask, struct nameidata *nd)
 		   exist.  So if permissions are revoked this won't be
 		   noticed immediately, only after the attribute
 		   timeout has expired */
-	} else if (nd && (nd->flags & (LOOKUP_ACCESS | LOOKUP_CHDIR))) {
+	} else if (mask & (MAY_ACCESS | MAY_CHDIR)) {
 		err = fuse_access(inode, mask);
 	} else if ((mask & MAY_EXEC) && S_ISREG(inode->i_mode)) {
 		if (!(inode->i_mode & S_IXUGO)) {
diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c
index 1e252dfc5294..4e982532f085 100644
--- a/fs/gfs2/ops_inode.c
+++ b/fs/gfs2/ops_inode.c
@@ -915,12 +915,6 @@ int gfs2_permission(struct inode *inode, int mask)
 	return error;
 }
 
-static int gfs2_iop_permission(struct inode *inode, int mask,
-			       struct nameidata *nd)
-{
-	return gfs2_permission(inode, mask);
-}
-
 static int setattr_size(struct inode *inode, struct iattr *attr)
 {
 	struct gfs2_inode *ip = GFS2_I(inode);
@@ -1150,7 +1144,7 @@ static int gfs2_removexattr(struct dentry *dentry, const char *name)
 }
 
 const struct inode_operations gfs2_file_iops = {
-	.permission = gfs2_iop_permission,
+	.permission = gfs2_permission,
 	.setattr = gfs2_setattr,
 	.getattr = gfs2_getattr,
 	.setxattr = gfs2_setxattr,
@@ -1169,7 +1163,7 @@ const struct inode_operations gfs2_dir_iops = {
 	.rmdir = gfs2_rmdir,
 	.mknod = gfs2_mknod,
 	.rename = gfs2_rename,
-	.permission = gfs2_iop_permission,
+	.permission = gfs2_permission,
 	.setattr = gfs2_setattr,
 	.getattr = gfs2_getattr,
 	.setxattr = gfs2_setxattr,
@@ -1181,7 +1175,7 @@ const struct inode_operations gfs2_dir_iops = {
 const struct inode_operations gfs2_symlink_iops = {
 	.readlink = gfs2_readlink,
 	.follow_link = gfs2_follow_link,
-	.permission = gfs2_iop_permission,
+	.permission = gfs2_permission,
 	.setattr = gfs2_setattr,
 	.getattr = gfs2_getattr,
 	.setxattr = gfs2_setxattr,
diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c
index dc4ec640e875..aa73f3fd5dd9 100644
--- a/fs/hfs/inode.c
+++ b/fs/hfs/inode.c
@@ -511,8 +511,7 @@ void hfs_clear_inode(struct inode *inode)
 	}
 }
 
-static int hfs_permission(struct inode *inode, int mask,
-			  struct nameidata *nd)
+static int hfs_permission(struct inode *inode, int mask)
 {
 	if (S_ISREG(inode->i_mode) && mask & MAY_EXEC)
 		return 0;
diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c
index cc3b5e24339b..d4014e3044d2 100644
--- a/fs/hfsplus/inode.c
+++ b/fs/hfsplus/inode.c
@@ -238,7 +238,7 @@ static void hfsplus_set_perms(struct inode *inode, struct hfsplus_perm *perms)
 	perms->dev = cpu_to_be32(HFSPLUS_I(inode).dev);
 }
 
-static int hfsplus_permission(struct inode *inode, int mask, struct nameidata *nd)
+static int hfsplus_permission(struct inode *inode, int mask)
 {
 	/* MAY_EXEC is also used for lookup, if no x bit is set allow lookup,
 	 * open_exec has the same test, so it's still not executable, if a x bit
diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c
index 5222345ddccf..d6ecabf4d231 100644
--- a/fs/hostfs/hostfs_kern.c
+++ b/fs/hostfs/hostfs_kern.c
@@ -822,7 +822,7 @@ int hostfs_rename(struct inode *from_ino, struct dentry *from,
 	return err;
 }
 
-int hostfs_permission(struct inode *ino, int desired, struct nameidata *nd)
+int hostfs_permission(struct inode *ino, int desired)
 {
 	char *name;
 	int r = 0, w = 0, x = 0, err;
diff --git a/fs/jffs2/acl.c b/fs/jffs2/acl.c
index 4c80404a9aba..d98713777a1b 100644
--- a/fs/jffs2/acl.c
+++ b/fs/jffs2/acl.c
@@ -314,7 +314,7 @@ static int jffs2_check_acl(struct inode *inode, int mask)
 	return -EAGAIN;
 }
 
-int jffs2_permission(struct inode *inode, int mask, struct nameidata *nd)
+int jffs2_permission(struct inode *inode, int mask)
 {
 	return generic_permission(inode, mask, jffs2_check_acl);
 }
diff --git a/fs/jffs2/acl.h b/fs/jffs2/acl.h
index 0bb7f003fd80..8ca058aed384 100644
--- a/fs/jffs2/acl.h
+++ b/fs/jffs2/acl.h
@@ -28,7 +28,7 @@ struct jffs2_acl_header {
 
 #define JFFS2_ACL_NOT_CACHED ((void *)-1)
 
-extern int jffs2_permission(struct inode *, int, struct nameidata *);
+extern int jffs2_permission(struct inode *, int);
 extern int jffs2_acl_chmod(struct inode *);
 extern int jffs2_init_acl_pre(struct inode *, struct inode *, int *);
 extern int jffs2_init_acl_post(struct inode *);
diff --git a/fs/jfs/acl.c b/fs/jfs/acl.c
index 4d84bdc88299..d3e5c33665de 100644
--- a/fs/jfs/acl.c
+++ b/fs/jfs/acl.c
@@ -140,7 +140,7 @@ static int jfs_check_acl(struct inode *inode, int mask)
 	return -EAGAIN;
 }
 
-int jfs_permission(struct inode *inode, int mask, struct nameidata *nd)
+int jfs_permission(struct inode *inode, int mask)
 {
 	return generic_permission(inode, mask, jfs_check_acl);
 }
diff --git a/fs/jfs/jfs_acl.h b/fs/jfs/jfs_acl.h
index 455fa4292045..88475f10a389 100644
--- a/fs/jfs/jfs_acl.h
+++ b/fs/jfs/jfs_acl.h
@@ -20,7 +20,7 @@
 
 #ifdef CONFIG_JFS_POSIX_ACL
 
-int jfs_permission(struct inode *, int, struct nameidata *);
+int jfs_permission(struct inode *, int);
 int jfs_init_acl(tid_t, struct inode *, struct inode *);
 int jfs_setattr(struct dentry *, struct iattr *);
 
diff --git a/fs/namei.c b/fs/namei.c
index 3b26a240ade9..46af98ed136b 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -185,6 +185,8 @@ int generic_permission(struct inode *inode, int mask,
 {
 	umode_t			mode = inode->i_mode;
 
+	mask &= MAY_READ | MAY_WRITE | MAY_EXEC;
+
 	if (current->fsuid == inode->i_uid)
 		mode >>= 6;
 	else {
@@ -203,7 +205,7 @@ int generic_permission(struct inode *inode, int mask,
 	/*
 	 * If the DACs are ok we don't need any capability check.
 	 */
-	if (((mode & mask & (MAY_READ|MAY_WRITE|MAY_EXEC)) == mask))
+	if ((mask & ~mode) == 0)
 		return 0;
 
  check_capabilities:
@@ -228,7 +230,7 @@ int generic_permission(struct inode *inode, int mask,
 
 int permission(struct inode *inode, int mask, struct nameidata *nd)
 {
-	int retval, submask;
+	int retval;
 	struct vfsmount *mnt = NULL;
 
 	if (nd)
@@ -261,9 +263,17 @@ int permission(struct inode *inode, int mask, struct nameidata *nd)
 	}
 
 	/* Ordinary permission routines do not understand MAY_APPEND. */
-	submask = mask & ~MAY_APPEND;
 	if (inode->i_op && inode->i_op->permission) {
-		retval = inode->i_op->permission(inode, submask, nd);
+		int extra = 0;
+		if (nd) {
+			if (nd->flags & LOOKUP_ACCESS)
+				extra |= MAY_ACCESS;
+			if (nd->flags & LOOKUP_CHDIR)
+				extra |= MAY_CHDIR;
+			if (nd->flags & LOOKUP_OPEN)
+				extra |= MAY_OPEN;
+		}
+		retval = inode->i_op->permission(inode, mask | extra);
 		if (!retval) {
 			/*
 			 * Exec permission on a regular file is denied if none
@@ -277,7 +287,7 @@ int permission(struct inode *inode, int mask, struct nameidata *nd)
 				return -EACCES;
 		}
 	} else {
-		retval = generic_permission(inode, submask, NULL);
+		retval = generic_permission(inode, mask, NULL);
 	}
 	if (retval)
 		return retval;
@@ -286,7 +296,8 @@ int permission(struct inode *inode, int mask, struct nameidata *nd)
 	if (retval)
 		return retval;
 
-	return security_inode_permission(inode, mask, nd);
+	return security_inode_permission(inode,
+			mask & (MAY_READ|MAY_WRITE|MAY_EXEC), nd);
 }
 
 /**
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 28a238dab23a..74f92b717f78 100644
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -1884,7 +1884,7 @@ static int nfs_do_access(struct inode *inode, struct rpc_cred *cred, int mask)
 		return status;
 	nfs_access_add_cache(inode, &cache);
 out:
-	if ((cache.mask & mask) == mask)
+	if ((mask & ~cache.mask & (MAY_READ | MAY_WRITE | MAY_EXEC)) == 0)
 		return 0;
 	return -EACCES;
 }
@@ -1907,17 +1907,17 @@ int nfs_may_open(struct inode *inode, struct rpc_cred *cred, int openflags)
 	return nfs_do_access(inode, cred, nfs_open_permission_mask(openflags));
 }
 
-int nfs_permission(struct inode *inode, int mask, struct nameidata *nd)
+int nfs_permission(struct inode *inode, int mask)
 {
 	struct rpc_cred *cred;
 	int res = 0;
 
 	nfs_inc_stats(inode, NFSIOS_VFSACCESS);
 
-	if (mask == 0)
+	if ((mask & (MAY_READ | MAY_WRITE | MAY_EXEC)) == 0)
 		goto out;
 	/* Is this sys_access() ? */
-	if (nd != NULL && (nd->flags & LOOKUP_ACCESS))
+	if (mask & MAY_ACCESS)
 		goto force_lookup;
 
 	switch (inode->i_mode & S_IFMT) {
@@ -1926,8 +1926,7 @@ int nfs_permission(struct inode *inode, int mask, struct nameidata *nd)
 		case S_IFREG:
 			/* NFSv4 has atomic_open... */
 			if (nfs_server_capable(inode, NFS_CAP_ATOMIC_OPEN)
-					&& nd != NULL
-					&& (nd->flags & LOOKUP_OPEN))
+					&& (mask & MAY_OPEN))
 				goto out;
 			break;
 		case S_IFDIR:
diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c
index e8514e8b6ce8..be2dd95d3a1d 100644
--- a/fs/ocfs2/file.c
+++ b/fs/ocfs2/file.c
@@ -1176,7 +1176,7 @@ bail:
 	return err;
 }
 
-int ocfs2_permission(struct inode *inode, int mask, struct nameidata *nd)
+int ocfs2_permission(struct inode *inode, int mask)
 {
 	int ret;
 
diff --git a/fs/ocfs2/file.h b/fs/ocfs2/file.h
index 048ddcaf5c80..1e27b4d017ea 100644
--- a/fs/ocfs2/file.h
+++ b/fs/ocfs2/file.h
@@ -62,8 +62,7 @@ int ocfs2_lock_allocators(struct inode *inode, struct ocfs2_dinode *di,
 int ocfs2_setattr(struct dentry *dentry, struct iattr *attr);
 int ocfs2_getattr(struct vfsmount *mnt, struct dentry *dentry,
 		  struct kstat *stat);
-int ocfs2_permission(struct inode *inode, int mask,
-		     struct nameidata *nd);
+int ocfs2_permission(struct inode *inode, int mask);
 
 int ocfs2_should_update_atime(struct inode *inode,
 			      struct vfsmount *vfsmnt);
diff --git a/fs/proc/base.c b/fs/proc/base.c
index 81bce6791bfc..d82d800389f6 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -1859,8 +1859,7 @@ static const struct file_operations proc_fd_operations = {
  * /proc/pid/fd needs a special permission handler so that a process can still
  * access /proc/self/fd after it has executed a setuid().
  */
-static int proc_fd_permission(struct inode *inode, int mask,
-				struct nameidata *nd)
+static int proc_fd_permission(struct inode *inode, int mask)
 {
 	int rv;
 
diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c
index fa1ec2433e44..f9a8b892718f 100644
--- a/fs/proc/proc_sysctl.c
+++ b/fs/proc/proc_sysctl.c
@@ -292,7 +292,7 @@ out:
 	return ret;
 }
 
-static int proc_sys_permission(struct inode *inode, int mask, struct nameidata *nd)
+static int proc_sys_permission(struct inode *inode, int mask)
 {
 	/*
 	 * sysctl entries that are not writeable,
diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c
index d7c4935c1034..bb3cb5b7cdb2 100644
--- a/fs/reiserfs/xattr.c
+++ b/fs/reiserfs/xattr.c
@@ -1250,7 +1250,7 @@ static int reiserfs_check_acl(struct inode *inode, int mask)
 	return error;
 }
 
-int reiserfs_permission(struct inode *inode, int mask, struct nameidata *nd)
+int reiserfs_permission(struct inode *inode, int mask)
 {
 	/*
 	 * We don't do permission checks on the internal objects.
diff --git a/fs/smbfs/file.c b/fs/smbfs/file.c
index 2294783320cb..e4f8d51a5553 100644
--- a/fs/smbfs/file.c
+++ b/fs/smbfs/file.c
@@ -408,7 +408,7 @@ smb_file_release(struct inode *inode, struct file * file)
  * privileges, so we need our own check for this.
  */
 static int
-smb_file_permission(struct inode *inode, int mask, struct nameidata *nd)
+smb_file_permission(struct inode *inode, int mask)
 {
 	int mode = inode->i_mode;
 	int error = 0;
@@ -417,7 +417,7 @@ smb_file_permission(struct inode *inode, int mask, struct nameidata *nd)
 
 	/* Look at user permissions */
 	mode >>= 6;
-	if ((mode & 7 & mask) != mask)
+	if (mask & ~mode & (MAY_READ | MAY_WRITE | MAY_EXEC))
 		error = -EACCES;
 	return error;
 }
diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c
index 2bf287ef5489..5fc61c824bb9 100644
--- a/fs/xfs/linux-2.6/xfs_iops.c
+++ b/fs/xfs/linux-2.6/xfs_iops.c
@@ -589,8 +589,7 @@ xfs_check_acl(
 STATIC int
 xfs_vn_permission(
 	struct inode		*inode,
-	int			mask,
-	struct nameidata	*nd)
+	int			mask)
 {
 	return generic_permission(inode, mask, xfs_check_acl);
 }
diff --git a/include/linux/coda_linux.h b/include/linux/coda_linux.h
index 31b75311e2ca..dcc228aa335a 100644
--- a/include/linux/coda_linux.h
+++ b/include/linux/coda_linux.h
@@ -37,7 +37,7 @@ extern const struct file_operations coda_ioctl_operations;
 /* operations shared over more than one file */
 int coda_open(struct inode *i, struct file *f);
 int coda_release(struct inode *i, struct file *f);
-int coda_permission(struct inode *inode, int mask, struct nameidata *nd);
+int coda_permission(struct inode *inode, int mask);
 int coda_revalidate_inode(struct dentry *);
 int coda_getattr(struct vfsmount *, struct dentry *, struct kstat *);
 int coda_setattr(struct dentry *, struct iattr *);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 7721a2ac9c0e..6c923c9b79bc 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -60,6 +60,9 @@ extern int dir_notify_enable;
 #define MAY_WRITE 2
 #define MAY_READ 4
 #define MAY_APPEND 8
+#define MAY_ACCESS 16
+#define MAY_CHDIR 32
+#define MAY_OPEN 64
 
 #define FMODE_READ 1
 #define FMODE_WRITE 2
@@ -1272,7 +1275,7 @@ struct inode_operations {
 	void * (*follow_link) (struct dentry *, struct nameidata *);
 	void (*put_link) (struct dentry *, struct nameidata *, void *);
 	void (*truncate) (struct inode *);
-	int (*permission) (struct inode *, int, struct nameidata *);
+	int (*permission) (struct inode *, int);
 	int (*setattr) (struct dentry *, struct iattr *);
 	int (*getattr) (struct vfsmount *mnt, struct dentry *, struct kstat *);
 	int (*setxattr) (struct dentry *, const char *,const void *,size_t,int);
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 29d261918734..f08f9ca602af 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -332,7 +332,7 @@ extern int nfs_refresh_inode(struct inode *, struct nfs_fattr *);
 extern int nfs_post_op_update_inode(struct inode *inode, struct nfs_fattr *fattr);
 extern int nfs_post_op_update_inode_force_wcc(struct inode *inode, struct nfs_fattr *fattr);
 extern int nfs_getattr(struct vfsmount *, struct dentry *, struct kstat *);
-extern int nfs_permission(struct inode *, int, struct nameidata *);
+extern int nfs_permission(struct inode *, int);
 extern int nfs_open(struct inode *, struct file *);
 extern int nfs_release(struct inode *, struct file *);
 extern int nfs_attribute_timeout(struct inode *inode);
diff --git a/include/linux/reiserfs_xattr.h b/include/linux/reiserfs_xattr.h
index 66a96814d614..af135ae895db 100644
--- a/include/linux/reiserfs_xattr.h
+++ b/include/linux/reiserfs_xattr.h
@@ -55,7 +55,7 @@ int reiserfs_removexattr(struct dentry *dentry, const char *name);
 int reiserfs_delete_xattrs(struct inode *inode);
 int reiserfs_chown_xattrs(struct inode *inode, struct iattr *attrs);
 int reiserfs_xattr_init(struct super_block *sb, int mount_flags);
-int reiserfs_permission(struct inode *inode, int mask, struct nameidata *nd);
+int reiserfs_permission(struct inode *inode, int mask);
 
 int reiserfs_xattr_del(struct inode *, const char *);
 int reiserfs_xattr_get(const struct inode *, const char *, void *, size_t);
diff --git a/include/linux/shmem_fs.h b/include/linux/shmem_fs.h
index f2d12d5a21b8..fd83f2584b15 100644
--- a/include/linux/shmem_fs.h
+++ b/include/linux/shmem_fs.h
@@ -43,7 +43,7 @@ static inline struct shmem_inode_info *SHMEM_I(struct inode *inode)
 }
 
 #ifdef CONFIG_TMPFS_POSIX_ACL
-int shmem_permission(struct inode *, int, struct nameidata *);
+int shmem_permission(struct inode *, int);
 int shmem_acl_init(struct inode *, struct inode *);
 void shmem_acl_destroy_inode(struct inode *);
 
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index ff5abcca5ddf..911d846f0503 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -1516,9 +1516,9 @@ static int do_sysctl_strategy(struct ctl_table_root *root,
 	int op = 0, rc;
 
 	if (oldval)
-		op |= 004;
+		op |= MAY_READ;
 	if (newval)
-		op |= 002;
+		op |= MAY_WRITE;
 	if (sysctl_perm(root, table, op))
 		return -EPERM;
 
@@ -1560,7 +1560,7 @@ repeat:
 		if (n == table->ctl_name) {
 			int error;
 			if (table->child) {
-				if (sysctl_perm(root, table, 001))
+				if (sysctl_perm(root, table, MAY_EXEC))
 					return -EPERM;
 				name++;
 				nlen--;
@@ -1635,7 +1635,7 @@ static int test_perm(int mode, int op)
 		mode >>= 6;
 	else if (in_egroup_p(0))
 		mode >>= 3;
-	if ((mode & op & 0007) == op)
+	if ((op & ~mode & (MAY_READ|MAY_WRITE|MAY_EXEC)) == 0)
 		return 0;
 	return -EACCES;
 }
@@ -1645,7 +1645,7 @@ int sysctl_perm(struct ctl_table_root *root, struct ctl_table *table, int op)
 	int error;
 	int mode;
 
-	error = security_sysctl(table, op);
+	error = security_sysctl(table, op & (MAY_READ | MAY_WRITE | MAY_EXEC));
 	if (error)
 		return error;
 
diff --git a/mm/shmem_acl.c b/mm/shmem_acl.c
index f5664c5b9eb1..8e5aadd7dcd6 100644
--- a/mm/shmem_acl.c
+++ b/mm/shmem_acl.c
@@ -191,7 +191,7 @@ shmem_check_acl(struct inode *inode, int mask)
  * shmem_permission  -  permission() inode operation
  */
 int
-shmem_permission(struct inode *inode, int mask, struct nameidata *nd)
+shmem_permission(struct inode *inode, int mask)
 {
 	return generic_permission(inode, mask, shmem_check_acl);
 }

From c82e42da8a6b2f3a85dc4d4278cb8238702f8f64 Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Tue, 24 Jun 2008 16:50:12 +0200
Subject: [PATCH 1136/1171] [patch 1/5] vfs: truncate: dont check immutable
 twice

vfs_permission(MAY_WRITE) already checked for the inode being
immutable, so no need to repeat it.

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Acked-by: Christoph Hellwig <hch@infradead.org>
---
 fs/open.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/open.c b/fs/open.c
index bb98d2fe809f..b2e4c93aed03 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -256,7 +256,7 @@ static long do_sys_truncate(const char __user * path, loff_t length)
 		goto mnt_drop_write_and_out;
 
 	error = -EPERM;
-	if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
+	if (IS_APPEND(inode))
 		goto mnt_drop_write_and_out;
 
 	error = get_write_access(inode);

From 2f1936b87783a3a56c9441b27b9ba7a747f11e8e Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Tue, 24 Jun 2008 16:50:14 +0200
Subject: [PATCH 1137/1171] [patch 3/5] vfs: change remove_suid() to
 file_remove_suid()

All calls to remove_suid() are made with a file pointer, because
(similarly to file_update_time) it is called when the file is written.

Clean up callers by passing in a file instead of a dentry.

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
---
 fs/fuse/file.c             | 2 +-
 fs/ntfs/file.c             | 2 +-
 fs/splice.c                | 4 ++--
 fs/xfs/linux-2.6/xfs_lrw.c | 2 +-
 include/linux/fs.h         | 2 +-
 mm/filemap.c               | 7 ++++---
 mm/filemap_xip.c           | 2 +-
 7 files changed, 11 insertions(+), 10 deletions(-)

diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index 67ff2c6a8f63..2bada6bbc317 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -893,7 +893,7 @@ static ssize_t fuse_file_aio_write(struct kiocb *iocb, const struct iovec *iov,
 	if (count == 0)
 		goto out;
 
-	err = remove_suid(file->f_path.dentry);
+	err = file_remove_suid(file);
 	if (err)
 		goto out;
 
diff --git a/fs/ntfs/file.c b/fs/ntfs/file.c
index 3c5550cd11d6..d020866d4232 100644
--- a/fs/ntfs/file.c
+++ b/fs/ntfs/file.c
@@ -2118,7 +2118,7 @@ static ssize_t ntfs_file_aio_write_nolock(struct kiocb *iocb,
 		goto out;
 	if (!count)
 		goto out;
-	err = remove_suid(file->f_path.dentry);
+	err = file_remove_suid(file);
 	if (err)
 		goto out;
 	file_update_time(file);
diff --git a/fs/splice.c b/fs/splice.c
index 47dc1a445d1f..b30311ba8af6 100644
--- a/fs/splice.c
+++ b/fs/splice.c
@@ -772,7 +772,7 @@ generic_file_splice_write_nolock(struct pipe_inode_info *pipe, struct file *out,
 	ssize_t ret;
 	int err;
 
-	err = remove_suid(out->f_path.dentry);
+	err = file_remove_suid(out);
 	if (unlikely(err))
 		return err;
 
@@ -830,7 +830,7 @@ generic_file_splice_write(struct pipe_inode_info *pipe, struct file *out,
 	ssize_t ret;
 
 	inode_double_lock(inode, pipe->inode);
-	ret = remove_suid(out->f_path.dentry);
+	ret = file_remove_suid(out);
 	if (likely(!ret))
 		ret = __splice_from_pipe(pipe, &sd, pipe_to_file);
 	inode_double_unlock(inode, pipe->inode);
diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c
index 5e3b57516ec7..82333b3e118e 100644
--- a/fs/xfs/linux-2.6/xfs_lrw.c
+++ b/fs/xfs/linux-2.6/xfs_lrw.c
@@ -711,7 +711,7 @@ start:
 	     !capable(CAP_FSETID)) {
 		error = xfs_write_clear_setuid(xip);
 		if (likely(!error))
-			error = -remove_suid(file->f_path.dentry);
+			error = -file_remove_suid(file);
 		if (unlikely(error)) {
 			goto out_unlock_internal;
 		}
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 6c923c9b79bc..1a3546e69f9e 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1834,7 +1834,7 @@ extern void clear_inode(struct inode *);
 extern void destroy_inode(struct inode *);
 extern struct inode *new_inode(struct super_block *);
 extern int should_remove_suid(struct dentry *);
-extern int remove_suid(struct dentry *);
+extern int file_remove_suid(struct file *);
 
 extern void __insert_inode_hash(struct inode *, unsigned long hashval);
 extern void remove_inode_hash(struct inode *);
diff --git a/mm/filemap.c b/mm/filemap.c
index 2ed8b0389c51..5de7633e1dbe 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -1758,8 +1758,9 @@ static int __remove_suid(struct dentry *dentry, int kill)
 	return notify_change(dentry, &newattrs);
 }
 
-int remove_suid(struct dentry *dentry)
+int file_remove_suid(struct file *file)
 {
+	struct dentry *dentry = file->f_path.dentry;
 	int killsuid = should_remove_suid(dentry);
 	int killpriv = security_inode_need_killpriv(dentry);
 	int error = 0;
@@ -1773,7 +1774,7 @@ int remove_suid(struct dentry *dentry)
 
 	return error;
 }
-EXPORT_SYMBOL(remove_suid);
+EXPORT_SYMBOL(file_remove_suid);
 
 static size_t __iovec_copy_from_user_inatomic(char *vaddr,
 			const struct iovec *iov, size_t base, size_t bytes)
@@ -2529,7 +2530,7 @@ __generic_file_aio_write_nolock(struct kiocb *iocb, const struct iovec *iov,
 	if (count == 0)
 		goto out;
 
-	err = remove_suid(file->f_path.dentry);
+	err = file_remove_suid(file);
 	if (err)
 		goto out;
 
diff --git a/mm/filemap_xip.c b/mm/filemap_xip.c
index 3e744abcce9d..98a3f31ccd6a 100644
--- a/mm/filemap_xip.c
+++ b/mm/filemap_xip.c
@@ -380,7 +380,7 @@ xip_file_write(struct file *filp, const char __user *buf, size_t len,
 	if (count == 0)
 		goto out_backing;
 
-	ret = remove_suid(filp->f_path.dentry);
+	ret = file_remove_suid(filp);
 	if (ret)
 		goto out_backing;
 

From 7e79eedb3b22200cc8b774baea3a7bf28d766101 Mon Sep 17 00:00:00 2001
From: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Date: Tue, 24 Jun 2008 16:50:15 +0200
Subject: [PATCH 1138/1171] [patch 4/5] vfs: reuse local variable in vfs_link()

Why not reuse "inode" which is assigned as

  struct inode *inode = old_dentry->d_inode;

in the beginning of vfs_link() ?

Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
---
 fs/namei.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/fs/namei.c b/fs/namei.c
index 46af98ed136b..3b67be7631dc 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -2524,19 +2524,19 @@ int vfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *new_de
 		return -EPERM;
 	if (!dir->i_op || !dir->i_op->link)
 		return -EPERM;
-	if (S_ISDIR(old_dentry->d_inode->i_mode))
+	if (S_ISDIR(inode->i_mode))
 		return -EPERM;
 
 	error = security_inode_link(old_dentry, dir, new_dentry);
 	if (error)
 		return error;
 
-	mutex_lock(&old_dentry->d_inode->i_mutex);
+	mutex_lock(&inode->i_mutex);
 	DQUOT_INIT(dir);
 	error = dir->i_op->link(old_dentry, dir, new_dentry);
-	mutex_unlock(&old_dentry->d_inode->i_mutex);
+	mutex_unlock(&inode->i_mutex);
 	if (!error)
-		fsnotify_link(dir, old_dentry->d_inode, new_dentry);
+		fsnotify_link(dir, inode, new_dentry);
 	return error;
 }
 

From db2e747b14991a4c6a5c98b0e5f552a193237c03 Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Tue, 24 Jun 2008 16:50:16 +0200
Subject: [PATCH 1139/1171] [patch 5/5] vfs: remove mode parameter from
 vfs_symlink()

Remove the unused mode parameter from vfs_symlink and callers.

Thanks to Tetsuo Handa for noticing.

CC: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
---
 fs/ecryptfs/inode.c |  4 +---
 fs/namei.c          |  4 ++--
 fs/nfsd/vfs.c       | 10 ++--------
 include/linux/fs.h  |  2 +-
 4 files changed, 6 insertions(+), 14 deletions(-)

diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c
index 32f4228efcd5..f25caf2b0887 100644
--- a/fs/ecryptfs/inode.c
+++ b/fs/ecryptfs/inode.c
@@ -465,7 +465,6 @@ static int ecryptfs_symlink(struct inode *dir, struct dentry *dentry,
 	int rc;
 	struct dentry *lower_dentry;
 	struct dentry *lower_dir_dentry;
-	umode_t mode;
 	char *encoded_symname;
 	int encoded_symlen;
 	struct ecryptfs_crypt_stat *crypt_stat = NULL;
@@ -473,7 +472,6 @@ static int ecryptfs_symlink(struct inode *dir, struct dentry *dentry,
 	lower_dentry = ecryptfs_dentry_to_lower(dentry);
 	dget(lower_dentry);
 	lower_dir_dentry = lock_parent(lower_dentry);
-	mode = S_IALLUGO;
 	encoded_symlen = ecryptfs_encode_filename(crypt_stat, symname,
 						  strlen(symname),
 						  &encoded_symname);
@@ -482,7 +480,7 @@ static int ecryptfs_symlink(struct inode *dir, struct dentry *dentry,
 		goto out_lock;
 	}
 	rc = vfs_symlink(lower_dir_dentry->d_inode, lower_dentry,
-			 encoded_symname, mode);
+			 encoded_symname);
 	kfree(encoded_symname);
 	if (rc || !lower_dentry->d_inode)
 		goto out_lock;
diff --git a/fs/namei.c b/fs/namei.c
index 3b67be7631dc..ae0e56fdb742 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -2434,7 +2434,7 @@ asmlinkage long sys_unlink(const char __user *pathname)
 	return do_unlinkat(AT_FDCWD, pathname);
 }
 
-int vfs_symlink(struct inode *dir, struct dentry *dentry, const char *oldname, int mode)
+int vfs_symlink(struct inode *dir, struct dentry *dentry, const char *oldname)
 {
 	int error = may_create(dir, dentry, NULL);
 
@@ -2483,7 +2483,7 @@ asmlinkage long sys_symlinkat(const char __user *oldname,
 	error = mnt_want_write(nd.path.mnt);
 	if (error)
 		goto out_dput;
-	error = vfs_symlink(nd.path.dentry->d_inode, dentry, from, S_IALLUGO);
+	error = vfs_symlink(nd.path.dentry->d_inode, dentry, from);
 	mnt_drop_write(nd.path.mnt);
 out_dput:
 	dput(dentry);
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index 0f4481e0502d..ad1ad59e3742 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -1516,7 +1516,6 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	struct dentry	*dentry, *dnew;
 	__be32		err, cerr;
 	int		host_err;
-	umode_t		mode;
 
 	err = nfserr_noent;
 	if (!flen || !plen)
@@ -1535,11 +1534,6 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp,
 	if (IS_ERR(dnew))
 		goto out_nfserr;
 
-	mode = S_IALLUGO;
-	/* Only the MODE ATTRibute is even vaguely meaningful */
-	if (iap && (iap->ia_valid & ATTR_MODE))
-		mode = iap->ia_mode & S_IALLUGO;
-
 	host_err = mnt_want_write(fhp->fh_export->ex_path.mnt);
 	if (host_err)
 		goto out_nfserr;
@@ -1551,11 +1545,11 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp,
 		else {
 			strncpy(path_alloced, path, plen);
 			path_alloced[plen] = 0;
-			host_err = vfs_symlink(dentry->d_inode, dnew, path_alloced, mode);
+			host_err = vfs_symlink(dentry->d_inode, dnew, path_alloced);
 			kfree(path_alloced);
 		}
 	} else
-		host_err = vfs_symlink(dentry->d_inode, dnew, path, mode);
+		host_err = vfs_symlink(dentry->d_inode, dnew, path);
 
 	if (!host_err) {
 		if (EX_ISSYNC(fhp->fh_export))
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 1a3546e69f9e..25998e803fc2 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1139,7 +1139,7 @@ extern int vfs_permission(struct nameidata *, int);
 extern int vfs_create(struct inode *, struct dentry *, int, struct nameidata *);
 extern int vfs_mkdir(struct inode *, struct dentry *, int);
 extern int vfs_mknod(struct inode *, struct dentry *, int, dev_t);
-extern int vfs_symlink(struct inode *, struct dentry *, const char *, int);
+extern int vfs_symlink(struct inode *, struct dentry *, const char *);
 extern int vfs_link(struct dentry *, struct inode *, struct dentry *);
 extern int vfs_rmdir(struct inode *, struct dentry *);
 extern int vfs_unlink(struct inode *, struct dentry *);

From 8bb79224b87aab92071e94d46e70bd160d89bf34 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 16 Jul 2008 09:51:03 -0400
Subject: [PATCH 1140/1171] [PATCH] permission checks for chdir need special
 treatment only on the last step

... so we ought to pass MAY_CHDIR to vfs_permission() instead of having
it triggered on every step of preceding pathname resolution.  LOOKUP_CHDIR
is killed by that.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/namei.c            | 2 --
 fs/open.c             | 5 ++---
 include/linux/namei.h | 1 -
 3 files changed, 2 insertions(+), 6 deletions(-)

diff --git a/fs/namei.c b/fs/namei.c
index ae0e56fdb742..6c76e1ee9c45 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -268,8 +268,6 @@ int permission(struct inode *inode, int mask, struct nameidata *nd)
 		if (nd) {
 			if (nd->flags & LOOKUP_ACCESS)
 				extra |= MAY_ACCESS;
-			if (nd->flags & LOOKUP_CHDIR)
-				extra |= MAY_CHDIR;
 			if (nd->flags & LOOKUP_OPEN)
 				extra |= MAY_OPEN;
 		}
diff --git a/fs/open.c b/fs/open.c
index b2e4c93aed03..8e02d42bfe44 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -501,12 +501,11 @@ asmlinkage long sys_chdir(const char __user * filename)
 	struct nameidata nd;
 	int error;
 
-	error = __user_walk(filename,
-			    LOOKUP_FOLLOW|LOOKUP_DIRECTORY|LOOKUP_CHDIR, &nd);
+	error = __user_walk(filename, LOOKUP_FOLLOW|LOOKUP_DIRECTORY, &nd);
 	if (error)
 		goto out;
 
-	error = vfs_permission(&nd, MAY_EXEC);
+	error = vfs_permission(&nd, MAY_EXEC | MAY_CHDIR);
 	if (error)
 		goto dput_and_out;
 
diff --git a/include/linux/namei.h b/include/linux/namei.h
index 24d88e98a626..3cf62d26d493 100644
--- a/include/linux/namei.h
+++ b/include/linux/namei.h
@@ -55,7 +55,6 @@ enum {LAST_NORM, LAST_ROOT, LAST_DOT, LAST_DOTDOT, LAST_BIND};
 #define LOOKUP_OPEN		(0x0100)
 #define LOOKUP_CREATE		(0x0200)
 #define LOOKUP_ACCESS		(0x0400)
-#define LOOKUP_CHDIR		(0x0800)
 
 extern int __user_walk(const char __user *, unsigned, struct nameidata *);
 extern int __user_walk_fd(int dfd, const char __user *, unsigned, struct nameidata *);

From 7f2da1e7d0330395e5e9e350b879b98a1ea495df Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sat, 10 May 2008 20:44:54 -0400
Subject: [PATCH 1141/1171] [PATCH] kill altroot

long overdue...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/namei.c                    | 89 +----------------------------------
 fs/namespace.c                |  8 +---
 fs/open.c                     |  3 +-
 include/asm-alpha/namei.h     | 17 -------
 include/asm-arm/namei.h       | 25 ----------
 include/asm-avr32/namei.h     |  7 ---
 include/asm-blackfin/namei.h  | 19 --------
 include/asm-cris/namei.h      | 17 -------
 include/asm-frv/namei.h       | 18 -------
 include/asm-h8300/namei.h     | 17 -------
 include/asm-ia64/namei.h      | 25 ----------
 include/asm-m32r/namei.h      | 17 -------
 include/asm-m68k/namei.h      | 17 -------
 include/asm-m68knommu/namei.h |  1 -
 include/asm-mips/namei.h      | 11 -----
 include/asm-mn10300/namei.h   | 22 ---------
 include/asm-parisc/namei.h    | 17 -------
 include/asm-powerpc/namei.h   | 20 --------
 include/asm-s390/namei.h      | 21 ---------
 include/asm-sh/namei.h        | 17 -------
 include/asm-sparc/namei.h     |  8 ----
 include/asm-sparc64/namei.h   |  1 -
 include/asm-um/namei.h        |  6 ---
 include/asm-v850/namei.h      | 17 -------
 include/asm-x86/namei.h       | 11 -----
 include/asm-xtensa/namei.h    | 26 ----------
 include/linux/fs_struct.h     |  3 +-
 include/linux/namei.h         |  1 -
 kernel/exec_domain.c          |  1 -
 kernel/exit.c                 |  2 -
 kernel/fork.c                 |  7 ---
 31 files changed, 5 insertions(+), 466 deletions(-)
 delete mode 100644 include/asm-alpha/namei.h
 delete mode 100644 include/asm-arm/namei.h
 delete mode 100644 include/asm-avr32/namei.h
 delete mode 100644 include/asm-blackfin/namei.h
 delete mode 100644 include/asm-cris/namei.h
 delete mode 100644 include/asm-frv/namei.h
 delete mode 100644 include/asm-h8300/namei.h
 delete mode 100644 include/asm-ia64/namei.h
 delete mode 100644 include/asm-m32r/namei.h
 delete mode 100644 include/asm-m68k/namei.h
 delete mode 100644 include/asm-m68knommu/namei.h
 delete mode 100644 include/asm-mips/namei.h
 delete mode 100644 include/asm-mn10300/namei.h
 delete mode 100644 include/asm-parisc/namei.h
 delete mode 100644 include/asm-powerpc/namei.h
 delete mode 100644 include/asm-s390/namei.h
 delete mode 100644 include/asm-sh/namei.h
 delete mode 100644 include/asm-sparc/namei.h
 delete mode 100644 include/asm-sparc64/namei.h
 delete mode 100644 include/asm-um/namei.h
 delete mode 100644 include/asm-v850/namei.h
 delete mode 100644 include/asm-x86/namei.h
 delete mode 100644 include/asm-xtensa/namei.h

diff --git a/fs/namei.c b/fs/namei.c
index 6c76e1ee9c45..095818089ac1 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -31,7 +31,6 @@
 #include <linux/file.h>
 #include <linux/fcntl.h>
 #include <linux/device_cgroup.h>
-#include <asm/namei.h>
 #include <asm/uaccess.h>
 
 #define ACC_MODE(x) ("\000\004\002\006"[(x)&O_ACCMODE])
@@ -562,27 +561,16 @@ out_unlock:
 	return result;
 }
 
-static int __emul_lookup_dentry(const char *, struct nameidata *);
-
 /* SMP-safe */
-static __always_inline int
+static __always_inline void
 walk_init_root(const char *name, struct nameidata *nd)
 {
 	struct fs_struct *fs = current->fs;
 
 	read_lock(&fs->lock);
-	if (fs->altroot.dentry && !(nd->flags & LOOKUP_NOALT)) {
-		nd->path = fs->altroot;
-		path_get(&fs->altroot);
-		read_unlock(&fs->lock);
-		if (__emul_lookup_dentry(name,nd))
-			return 0;
-		read_lock(&fs->lock);
-	}
 	nd->path = fs->root;
 	path_get(&fs->root);
 	read_unlock(&fs->lock);
-	return 1;
 }
 
 /*
@@ -623,12 +611,9 @@ static __always_inline int __vfs_follow_link(struct nameidata *nd, const char *l
 
 	if (*link == '/') {
 		path_put(&nd->path);
-		if (!walk_init_root(link, nd))
-			/* weird __emul_prefix() stuff did it */
-			goto out;
+		walk_init_root(link, nd);
 	}
 	res = link_path_walk(link, nd);
-out:
 	if (nd->depth || res || nd->last_type!=LAST_NORM)
 		return res;
 	/*
@@ -1077,67 +1062,6 @@ static int path_walk(const char *name, struct nameidata *nd)
 	return link_path_walk(name, nd);
 }
 
-/* 
- * SMP-safe: Returns 1 and nd will have valid dentry and mnt, if
- * everything is done. Returns 0 and drops input nd, if lookup failed;
- */
-static int __emul_lookup_dentry(const char *name, struct nameidata *nd)
-{
-	if (path_walk(name, nd))
-		return 0;		/* something went wrong... */
-
-	if (!nd->path.dentry->d_inode ||
-	    S_ISDIR(nd->path.dentry->d_inode->i_mode)) {
-		struct path old_path = nd->path;
-		struct qstr last = nd->last;
-		int last_type = nd->last_type;
-		struct fs_struct *fs = current->fs;
-
-		/*
-		 * NAME was not found in alternate root or it's a directory.
-		 * Try to find it in the normal root:
-		 */
-		nd->last_type = LAST_ROOT;
-		read_lock(&fs->lock);
-		nd->path = fs->root;
-		path_get(&fs->root);
-		read_unlock(&fs->lock);
-		if (path_walk(name, nd) == 0) {
-			if (nd->path.dentry->d_inode) {
-				path_put(&old_path);
-				return 1;
-			}
-			path_put(&nd->path);
-		}
-		nd->path = old_path;
-		nd->last = last;
-		nd->last_type = last_type;
-	}
-	return 1;
-}
-
-void set_fs_altroot(void)
-{
-	char *emul = __emul_prefix();
-	struct nameidata nd;
-	struct path path = {}, old_path;
-	int err;
-	struct fs_struct *fs = current->fs;
-
-	if (!emul)
-		goto set_it;
-	err = path_lookup(emul, LOOKUP_FOLLOW|LOOKUP_DIRECTORY|LOOKUP_NOALT, &nd);
-	if (!err)
-		path = nd.path;
-set_it:
-	write_lock(&fs->lock);
-	old_path = fs->altroot;
-	fs->altroot = path;
-	write_unlock(&fs->lock);
-	if (old_path.dentry)
-		path_put(&old_path);
-}
-
 /* Returns 0 and nd will be valid on success; Retuns error, otherwise. */
 static int do_path_lookup(int dfd, const char *name,
 				unsigned int flags, struct nameidata *nd)
@@ -1153,14 +1077,6 @@ static int do_path_lookup(int dfd, const char *name,
 
 	if (*name=='/') {
 		read_lock(&fs->lock);
-		if (fs->altroot.dentry && !(nd->flags & LOOKUP_NOALT)) {
-			nd->path = fs->altroot;
-			path_get(&fs->altroot);
-			read_unlock(&fs->lock);
-			if (__emul_lookup_dentry(name,nd))
-				goto out; /* found in altroot */
-			read_lock(&fs->lock);
-		}
 		nd->path = fs->root;
 		path_get(&fs->root);
 		read_unlock(&fs->lock);
@@ -1194,7 +1110,6 @@ static int do_path_lookup(int dfd, const char *name,
 	}
 
 	retval = path_walk(name, nd);
-out:
 	if (unlikely(!retval && !audit_dummy_context() && nd->path.dentry &&
 				nd->path.dentry->d_inode))
 		audit_inode(name, nd->path.dentry);
diff --git a/fs/namespace.c b/fs/namespace.c
index f30b11e2240e..c4fcf48acef8 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -1972,7 +1972,7 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns,
 		struct fs_struct *fs)
 {
 	struct mnt_namespace *new_ns;
-	struct vfsmount *rootmnt = NULL, *pwdmnt = NULL, *altrootmnt = NULL;
+	struct vfsmount *rootmnt = NULL, *pwdmnt = NULL;
 	struct vfsmount *p, *q;
 
 	new_ns = kmalloc(sizeof(struct mnt_namespace), GFP_KERNEL);
@@ -2015,10 +2015,6 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns,
 				pwdmnt = p;
 				fs->pwd.mnt = mntget(q);
 			}
-			if (p == fs->altroot.mnt) {
-				altrootmnt = p;
-				fs->altroot.mnt = mntget(q);
-			}
 		}
 		p = next_mnt(p, mnt_ns->root);
 		q = next_mnt(q, new_ns->root);
@@ -2029,8 +2025,6 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns,
 		mntput(rootmnt);
 	if (pwdmnt)
 		mntput(pwdmnt);
-	if (altrootmnt)
-		mntput(altrootmnt);
 
 	return new_ns;
 }
diff --git a/fs/open.c b/fs/open.c
index 8e02d42bfe44..d3a2a00f52dc 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -548,7 +548,7 @@ asmlinkage long sys_chroot(const char __user * filename)
 	struct nameidata nd;
 	int error;
 
-	error = __user_walk(filename, LOOKUP_FOLLOW | LOOKUP_DIRECTORY | LOOKUP_NOALT, &nd);
+	error = __user_walk(filename, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, &nd);
 	if (error)
 		goto out;
 
@@ -561,7 +561,6 @@ asmlinkage long sys_chroot(const char __user * filename)
 		goto dput_and_out;
 
 	set_fs_root(current->fs, &nd.path);
-	set_fs_altroot();
 	error = 0;
 dput_and_out:
 	path_put(&nd.path);
diff --git a/include/asm-alpha/namei.h b/include/asm-alpha/namei.h
deleted file mode 100644
index 5cc9bb39499d..000000000000
--- a/include/asm-alpha/namei.h
+++ /dev/null
@@ -1,17 +0,0 @@
-/* $Id: namei.h,v 1.1 1996/12/13 14:48:21 jj Exp $
- * linux/include/asm-alpha/namei.h
- *
- * Included from linux/fs/namei.c
- */
-
-#ifndef __ALPHA_NAMEI_H
-#define __ALPHA_NAMEI_H
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif /* __ALPHA_NAMEI_H */
diff --git a/include/asm-arm/namei.h b/include/asm-arm/namei.h
deleted file mode 100644
index a402d3b9d0f7..000000000000
--- a/include/asm-arm/namei.h
+++ /dev/null
@@ -1,25 +0,0 @@
-/* 
- * linux/include/asm-arm/namei.h
- *
- * Routines to handle famous /usr/gnemul
- * Derived from the Sparc version of this file
- *
- * Included from linux/fs/namei.c
- */
-
-#ifndef __ASMARM_NAMEI_H
-#define __ASMARM_NAMEI_H
-
-#define ARM_BSD_EMUL "usr/gnemul/bsd/"
-
-static inline char *__emul_prefix(void)
-{
-	switch (current->personality) {
-	case PER_BSD:
-		return ARM_BSD_EMUL;
-	default:
-		return NULL;
-	}
-}
-
-#endif /* __ASMARM_NAMEI_H */
diff --git a/include/asm-avr32/namei.h b/include/asm-avr32/namei.h
deleted file mode 100644
index f0a26de06cab..000000000000
--- a/include/asm-avr32/namei.h
+++ /dev/null
@@ -1,7 +0,0 @@
-#ifndef __ASM_AVR32_NAMEI_H
-#define __ASM_AVR32_NAMEI_H
-
-/* This dummy routine may be changed to something useful */
-#define __emul_prefix() NULL
-
-#endif /* __ASM_AVR32_NAMEI_H */
diff --git a/include/asm-blackfin/namei.h b/include/asm-blackfin/namei.h
deleted file mode 100644
index 8b89a2d65cb4..000000000000
--- a/include/asm-blackfin/namei.h
+++ /dev/null
@@ -1,19 +0,0 @@
-/*
- * linux/include/asm/namei.h
- *
- * Included from linux/fs/namei.c
- *
- * Changes made by Lineo Inc.    May 2001
- */
-
-#ifndef __BFIN_NAMEI_H
-#define __BFIN_NAMEI_H
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif
diff --git a/include/asm-cris/namei.h b/include/asm-cris/namei.h
deleted file mode 100644
index 8a3be7a6d9f6..000000000000
--- a/include/asm-cris/namei.h
+++ /dev/null
@@ -1,17 +0,0 @@
-/* $Id: namei.h,v 1.1 2000/07/10 16:32:31 bjornw Exp $
- * linux/include/asm-cris/namei.h
- *
- * Included from linux/fs/namei.c
- */
-
-#ifndef __CRIS_NAMEI_H
-#define __CRIS_NAMEI_H
-
-/* used to find file-system prefixes for doing emulations
- * see for example asm-sparc/namei.h
- * we don't use it...
- */
-
-#define __emul_prefix() NULL
-
-#endif /* __CRIS_NAMEI_H */
diff --git a/include/asm-frv/namei.h b/include/asm-frv/namei.h
deleted file mode 100644
index 4ea57171d951..000000000000
--- a/include/asm-frv/namei.h
+++ /dev/null
@@ -1,18 +0,0 @@
-/*
- * include/asm-frv/namei.h
- *
- * Included from linux/fs/namei.c
- */
-
-#ifndef __ASM_NAMEI_H
-#define __ASM_NAMEI_H
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif
-
diff --git a/include/asm-h8300/namei.h b/include/asm-h8300/namei.h
deleted file mode 100644
index ab6f196db6e0..000000000000
--- a/include/asm-h8300/namei.h
+++ /dev/null
@@ -1,17 +0,0 @@
-/*
- * linux/include/asm-h8300/namei.h
- *
- * Included from linux/fs/namei.c
- */
-
-#ifndef __H8300_NAMEI_H
-#define __H8300_NAMEI_H
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif
diff --git a/include/asm-ia64/namei.h b/include/asm-ia64/namei.h
deleted file mode 100644
index 78e768079083..000000000000
--- a/include/asm-ia64/namei.h
+++ /dev/null
@@ -1,25 +0,0 @@
-#ifndef _ASM_IA64_NAMEI_H
-#define _ASM_IA64_NAMEI_H
-
-/*
- * Modified 1998, 1999, 2001
- *	David Mosberger-Tang <davidm@hpl.hp.com>, Hewlett-Packard Co
- */
-
-#include <asm/ptrace.h>
-#include <asm/system.h>
-
-#define EMUL_PREFIX_LINUX_IA32 "/emul/ia32-linux/"
-
-static inline char *
-__emul_prefix (void)
-{
-	switch (current->personality) {
-	      case PER_LINUX32:
-		return EMUL_PREFIX_LINUX_IA32;
-	      default:
-		return NULL;
-	}
-}
-
-#endif /* _ASM_IA64_NAMEI_H */
diff --git a/include/asm-m32r/namei.h b/include/asm-m32r/namei.h
deleted file mode 100644
index 210f8056b805..000000000000
--- a/include/asm-m32r/namei.h
+++ /dev/null
@@ -1,17 +0,0 @@
-#ifndef _ASM_M32R_NAMEI_H
-#define _ASM_M32R_NAMEI_H
-
-/*
- * linux/include/asm-m32r/namei.h
- *
- * Included from linux/fs/namei.c
- */
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif /* _ASM_M32R_NAMEI_H */
diff --git a/include/asm-m68k/namei.h b/include/asm-m68k/namei.h
deleted file mode 100644
index f33f243b644a..000000000000
--- a/include/asm-m68k/namei.h
+++ /dev/null
@@ -1,17 +0,0 @@
-/*
- * linux/include/asm-m68k/namei.h
- *
- * Included from linux/fs/namei.c
- */
-
-#ifndef __M68K_NAMEI_H
-#define __M68K_NAMEI_H
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif
diff --git a/include/asm-m68knommu/namei.h b/include/asm-m68knommu/namei.h
deleted file mode 100644
index 31a85d27b931..000000000000
--- a/include/asm-m68knommu/namei.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-m68k/namei.h>
diff --git a/include/asm-mips/namei.h b/include/asm-mips/namei.h
deleted file mode 100644
index a6605a752469..000000000000
--- a/include/asm-mips/namei.h
+++ /dev/null
@@ -1,11 +0,0 @@
-#ifndef _ASM_NAMEI_H
-#define _ASM_NAMEI_H
-
-/*
- * This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- */
-
-#define __emul_prefix() NULL
-
-#endif /* _ASM_NAMEI_H */
diff --git a/include/asm-mn10300/namei.h b/include/asm-mn10300/namei.h
deleted file mode 100644
index bd9ce94aeb65..000000000000
--- a/include/asm-mn10300/namei.h
+++ /dev/null
@@ -1,22 +0,0 @@
-/* Emulation stuff
- *
- * Copyright (C) 2007 Red Hat, Inc. All Rights Reserved.
- * Written by David Howells (dhowells@redhat.com)
- *
- * This program is free software; you can redistribute it and/or
- * modify it under the terms of the GNU General Public Licence
- * as published by the Free Software Foundation; either version
- * 2 of the Licence, or (at your option) any later version.
- */
-
-#ifndef _ASM_NAMEI_H
-#define _ASM_NAMEI_H
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif /* _ASM_NAMEI_H */
diff --git a/include/asm-parisc/namei.h b/include/asm-parisc/namei.h
deleted file mode 100644
index 8d29b3d9fb33..000000000000
--- a/include/asm-parisc/namei.h
+++ /dev/null
@@ -1,17 +0,0 @@
-/* $Id: namei.h,v 1.1 1996/12/13 14:48:21 jj Exp $
- * linux/include/asm-parisc/namei.h
- *
- * Included from linux/fs/namei.c
- */
-
-#ifndef __PARISC_NAMEI_H
-#define __PARISC_NAMEI_H
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif /* __PARISC_NAMEI_H */
diff --git a/include/asm-powerpc/namei.h b/include/asm-powerpc/namei.h
deleted file mode 100644
index 657443474a6a..000000000000
--- a/include/asm-powerpc/namei.h
+++ /dev/null
@@ -1,20 +0,0 @@
-#ifndef _ASM_POWERPC_NAMEI_H
-#define _ASM_POWERPC_NAMEI_H
-
-#ifdef __KERNEL__
-
-/*
- * Adapted from include/asm-alpha/namei.h
- *
- * Included from fs/namei.c
- */
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif	/* __KERNEL__ */
-#endif	/* _ASM_POWERPC_NAMEI_H */
diff --git a/include/asm-s390/namei.h b/include/asm-s390/namei.h
deleted file mode 100644
index 3e286bdde4b0..000000000000
--- a/include/asm-s390/namei.h
+++ /dev/null
@@ -1,21 +0,0 @@
-/*
- *  include/asm-s390/namei.h
- *
- *  S390 version
- *
- *  Derived from "include/asm-i386/namei.h"
- *
- *  Included from linux/fs/namei.c
- */
-
-#ifndef __S390_NAMEI_H
-#define __S390_NAMEI_H
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif /* __S390_NAMEI_H */
diff --git a/include/asm-sh/namei.h b/include/asm-sh/namei.h
deleted file mode 100644
index 338a5d947143..000000000000
--- a/include/asm-sh/namei.h
+++ /dev/null
@@ -1,17 +0,0 @@
-/* $Id: namei.h,v 1.3 2000/07/04 06:24:49 gniibe Exp $
- * linux/include/asm-sh/namei.h
- *
- * Included from linux/fs/namei.c
- */
-
-#ifndef __ASM_SH_NAMEI_H
-#define __ASM_SH_NAMEI_H
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif /* __ASM_SH_NAMEI_H */
diff --git a/include/asm-sparc/namei.h b/include/asm-sparc/namei.h
deleted file mode 100644
index eff944b8e321..000000000000
--- a/include/asm-sparc/namei.h
+++ /dev/null
@@ -1,8 +0,0 @@
-#ifndef ___ASM_SPARC_NAMEI_H
-#define ___ASM_SPARC_NAMEI_H
-#if defined(__sparc__) && defined(__arch64__)
-#include <asm-sparc/namei_64.h>
-#else
-#include <asm-sparc/namei_32.h>
-#endif
-#endif
diff --git a/include/asm-sparc64/namei.h b/include/asm-sparc64/namei.h
deleted file mode 100644
index 1344a910ba2f..000000000000
--- a/include/asm-sparc64/namei.h
+++ /dev/null
@@ -1 +0,0 @@
-#include <asm-sparc/namei.h>
diff --git a/include/asm-um/namei.h b/include/asm-um/namei.h
deleted file mode 100644
index 002984d5bc85..000000000000
--- a/include/asm-um/namei.h
+++ /dev/null
@@ -1,6 +0,0 @@
-#ifndef __UM_NAMEI_H
-#define __UM_NAMEI_H
-
-#include "asm/arch/namei.h"
-
-#endif
diff --git a/include/asm-v850/namei.h b/include/asm-v850/namei.h
deleted file mode 100644
index ee8339b23843..000000000000
--- a/include/asm-v850/namei.h
+++ /dev/null
@@ -1,17 +0,0 @@
-/*
- * linux/include/asm-v850/namei.h
- *
- * Included from linux/fs/namei.c
- */
-
-#ifndef __V850_NAMEI_H__
-#define __V850_NAMEI_H__
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif /* __V850_NAMEI_H__ */
diff --git a/include/asm-x86/namei.h b/include/asm-x86/namei.h
deleted file mode 100644
index 415ef5d9550e..000000000000
--- a/include/asm-x86/namei.h
+++ /dev/null
@@ -1,11 +0,0 @@
-#ifndef _ASM_X86_NAMEI_H
-#define _ASM_X86_NAMEI_H
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif /* _ASM_X86_NAMEI_H */
diff --git a/include/asm-xtensa/namei.h b/include/asm-xtensa/namei.h
deleted file mode 100644
index 3fdff039d27d..000000000000
--- a/include/asm-xtensa/namei.h
+++ /dev/null
@@ -1,26 +0,0 @@
-/*
- * include/asm-xtensa/namei.h
- *
- * Included from linux/fs/namei.c
- *
- * This file is subject to the terms and conditions of the GNU General Public
- * License.  See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 2001 - 2005 Tensilica Inc.
- */
-
-#ifndef _XTENSA_NAMEI_H
-#define _XTENSA_NAMEI_H
-
-#ifdef __KERNEL__
-
-/* This dummy routine maybe changed to something useful
- * for /usr/gnemul/ emulation stuff.
- * Look at asm-sparc/namei.h for details.
- */
-
-#define __emul_prefix() NULL
-
-#endif	/* __KERNEL__ */
-#endif	/* _XTENSA_NAMEI_H */
diff --git a/include/linux/fs_struct.h b/include/linux/fs_struct.h
index 282f54219129..9e5a06e78d02 100644
--- a/include/linux/fs_struct.h
+++ b/include/linux/fs_struct.h
@@ -7,7 +7,7 @@ struct fs_struct {
 	atomic_t count;
 	rwlock_t lock;
 	int umask;
-	struct path root, pwd, altroot;
+	struct path root, pwd;
 };
 
 #define INIT_FS {				\
@@ -19,7 +19,6 @@ struct fs_struct {
 extern struct kmem_cache *fs_cachep;
 
 extern void exit_fs(struct task_struct *);
-extern void set_fs_altroot(void);
 extern void set_fs_root(struct fs_struct *, struct path *);
 extern void set_fs_pwd(struct fs_struct *, struct path *);
 extern struct fs_struct *copy_fs_struct(struct fs_struct *);
diff --git a/include/linux/namei.h b/include/linux/namei.h
index 3cf62d26d493..768773d57857 100644
--- a/include/linux/namei.h
+++ b/include/linux/namei.h
@@ -47,7 +47,6 @@ enum {LAST_NORM, LAST_ROOT, LAST_DOT, LAST_DOTDOT, LAST_BIND};
 #define LOOKUP_DIRECTORY	 2
 #define LOOKUP_CONTINUE		 4
 #define LOOKUP_PARENT		16
-#define LOOKUP_NOALT		32
 #define LOOKUP_REVAL		64
 /*
  * Intent data
diff --git a/kernel/exec_domain.c b/kernel/exec_domain.c
index c1ef192aa655..0d407e886735 100644
--- a/kernel/exec_domain.c
+++ b/kernel/exec_domain.c
@@ -168,7 +168,6 @@ __set_personality(u_long personality)
 	current->personality = personality;
 	oep = current_thread_info()->exec_domain;
 	current_thread_info()->exec_domain = ep;
-	set_fs_altroot();
 
 	module_put(oep->module);
 	return 0;
diff --git a/kernel/exit.c b/kernel/exit.c
index 6cdf60712bd2..0caf590548a0 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -565,8 +565,6 @@ void put_fs_struct(struct fs_struct *fs)
 	if (atomic_dec_and_test(&fs->count)) {
 		path_put(&fs->root);
 		path_put(&fs->pwd);
-		if (fs->altroot.dentry)
-			path_put(&fs->altroot);
 		kmem_cache_free(fs_cachep, fs);
 	}
 }
diff --git a/kernel/fork.c b/kernel/fork.c
index abb3ed6298f6..5e050c1317c4 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -657,13 +657,6 @@ static struct fs_struct *__copy_fs_struct(struct fs_struct *old)
 		path_get(&old->root);
 		fs->pwd = old->pwd;
 		path_get(&old->pwd);
-		if (old->altroot.dentry) {
-			fs->altroot = old->altroot;
-			path_get(&old->altroot);
-		} else {
-			fs->altroot.mnt = NULL;
-			fs->altroot.dentry = NULL;
-		}
 		read_unlock(&old->lock);
 	}
 	return fs;

From a110343f0d6d41f68b7cf8c00b57a3172c67f816 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Thu, 17 Jul 2008 09:19:08 -0400
Subject: [PATCH 1142/1171] [PATCH] fix MAY_CHDIR/MAY_ACCESS/LOOKUP_ACCESS mess

* MAY_CHDIR is redundant - it's an equivalent of MAY_ACCESS
* MAY_ACCESS on fuse should affect only the last step of pathname resolution
* fchdir() and chroot() should pass MAY_ACCESS, for the same reason why
  chdir() needs that.
* now that we pass MAY_ACCESS explicitly in all cases, LOOKUP_ACCESS can be
  removed; it has no business being in nameidata.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/fuse/dir.c         |  2 +-
 fs/namei.c            |  2 --
 fs/open.c             | 10 +++++-----
 include/linux/fs.h    |  3 +--
 include/linux/namei.h |  1 -
 5 files changed, 7 insertions(+), 11 deletions(-)

diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index 48a7934cb950..fd03330cadeb 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -962,7 +962,7 @@ static int fuse_permission(struct inode *inode, int mask)
 		   exist.  So if permissions are revoked this won't be
 		   noticed immediately, only after the attribute
 		   timeout has expired */
-	} else if (mask & (MAY_ACCESS | MAY_CHDIR)) {
+	} else if (mask & MAY_ACCESS) {
 		err = fuse_access(inode, mask);
 	} else if ((mask & MAY_EXEC) && S_ISREG(inode->i_mode)) {
 		if (!(inode->i_mode & S_IXUGO)) {
diff --git a/fs/namei.c b/fs/namei.c
index 095818089ac1..33dcaf025c49 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -265,8 +265,6 @@ int permission(struct inode *inode, int mask, struct nameidata *nd)
 	if (inode->i_op && inode->i_op->permission) {
 		int extra = 0;
 		if (nd) {
-			if (nd->flags & LOOKUP_ACCESS)
-				extra |= MAY_ACCESS;
 			if (nd->flags & LOOKUP_OPEN)
 				extra |= MAY_OPEN;
 		}
diff --git a/fs/open.c b/fs/open.c
index d3a2a00f52dc..3317e1909b2c 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -457,11 +457,11 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode)
 			old_cap = cap_set_effective(current->cap_permitted);
 	}
 
-	res = __user_walk_fd(dfd, filename, LOOKUP_FOLLOW|LOOKUP_ACCESS, &nd);
+	res = __user_walk_fd(dfd, filename, LOOKUP_FOLLOW, &nd);
 	if (res)
 		goto out;
 
-	res = vfs_permission(&nd, mode);
+	res = vfs_permission(&nd, mode | MAY_ACCESS);
 	/* SuS v2 requires we report a read only fs too */
 	if(res || !(mode & S_IWOTH) ||
 	   special_file(nd.path.dentry->d_inode->i_mode))
@@ -505,7 +505,7 @@ asmlinkage long sys_chdir(const char __user * filename)
 	if (error)
 		goto out;
 
-	error = vfs_permission(&nd, MAY_EXEC | MAY_CHDIR);
+	error = vfs_permission(&nd, MAY_EXEC | MAY_ACCESS);
 	if (error)
 		goto dput_and_out;
 
@@ -534,7 +534,7 @@ asmlinkage long sys_fchdir(unsigned int fd)
 	if (!S_ISDIR(inode->i_mode))
 		goto out_putf;
 
-	error = file_permission(file, MAY_EXEC);
+	error = file_permission(file, MAY_EXEC | MAY_ACCESS);
 	if (!error)
 		set_fs_pwd(current->fs, &file->f_path);
 out_putf:
@@ -552,7 +552,7 @@ asmlinkage long sys_chroot(const char __user * filename)
 	if (error)
 		goto out;
 
-	error = vfs_permission(&nd, MAY_EXEC);
+	error = vfs_permission(&nd, MAY_EXEC | MAY_ACCESS);
 	if (error)
 		goto dput_and_out;
 
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 25998e803fc2..d8721e818b45 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -61,8 +61,7 @@ extern int dir_notify_enable;
 #define MAY_READ 4
 #define MAY_APPEND 8
 #define MAY_ACCESS 16
-#define MAY_CHDIR 32
-#define MAY_OPEN 64
+#define MAY_OPEN 32
 
 #define FMODE_READ 1
 #define FMODE_WRITE 2
diff --git a/include/linux/namei.h b/include/linux/namei.h
index 768773d57857..60e35a02f6cb 100644
--- a/include/linux/namei.h
+++ b/include/linux/namei.h
@@ -53,7 +53,6 @@ enum {LAST_NORM, LAST_ROOT, LAST_DOT, LAST_DOTDOT, LAST_BIND};
  */
 #define LOOKUP_OPEN		(0x0100)
 #define LOOKUP_CREATE		(0x0200)
-#define LOOKUP_ACCESS		(0x0400)
 
 extern int __user_walk(const char __user *, unsigned, struct nameidata *);
 extern int __user_walk_fd(int dfd, const char __user *, unsigned, struct nameidata *);

From b77b0646ef4efe31a7449bb3d9360fd00f95433d Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Thu, 17 Jul 2008 09:37:02 -0400
Subject: [PATCH 1143/1171] [PATCH] pass MAY_OPEN to vfs_permission()
 explicitly

... and get rid of the last "let's deduce mask from nameidata->flags"
bit.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/exec.c                  |  4 ++--
 fs/namei.c                 | 13 ++++---------
 include/linux/security.h   |  7 +++----
 security/capability.c      |  3 +--
 security/security.c        |  4 ++--
 security/selinux/hooks.c   |  5 ++---
 security/smack/smack_lsm.c |  3 +--
 7 files changed, 15 insertions(+), 24 deletions(-)

diff --git a/fs/exec.c b/fs/exec.c
index b8792a131533..0ba5d355c5a1 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -118,7 +118,7 @@ asmlinkage long sys_uselib(const char __user * library)
 	if (!S_ISREG(nd.path.dentry->d_inode->i_mode))
 		goto exit;
 
-	error = vfs_permission(&nd, MAY_READ | MAY_EXEC);
+	error = vfs_permission(&nd, MAY_READ | MAY_EXEC | MAY_OPEN);
 	if (error)
 		goto exit;
 
@@ -666,7 +666,7 @@ struct file *open_exec(const char *name)
 		struct inode *inode = nd.path.dentry->d_inode;
 		file = ERR_PTR(-EACCES);
 		if (S_ISREG(inode->i_mode)) {
-			int err = vfs_permission(&nd, MAY_EXEC);
+			int err = vfs_permission(&nd, MAY_EXEC | MAY_OPEN);
 			file = ERR_PTR(err);
 			if (!err) {
 				file = nameidata_to_filp(&nd,
diff --git a/fs/namei.c b/fs/namei.c
index 33dcaf025c49..6b0e8e5e079e 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -263,12 +263,7 @@ int permission(struct inode *inode, int mask, struct nameidata *nd)
 
 	/* Ordinary permission routines do not understand MAY_APPEND. */
 	if (inode->i_op && inode->i_op->permission) {
-		int extra = 0;
-		if (nd) {
-			if (nd->flags & LOOKUP_OPEN)
-				extra |= MAY_OPEN;
-		}
-		retval = inode->i_op->permission(inode, mask | extra);
+		retval = inode->i_op->permission(inode, mask);
 		if (!retval) {
 			/*
 			 * Exec permission on a regular file is denied if none
@@ -292,7 +287,7 @@ int permission(struct inode *inode, int mask, struct nameidata *nd)
 		return retval;
 
 	return security_inode_permission(inode,
-			mask & (MAY_READ|MAY_WRITE|MAY_EXEC), nd);
+			mask & (MAY_READ|MAY_WRITE|MAY_EXEC));
 }
 
 /**
@@ -492,7 +487,7 @@ static int exec_permission_lite(struct inode *inode,
 
 	return -EACCES;
 ok:
-	return security_inode_permission(inode, MAY_EXEC, nd);
+	return security_inode_permission(inode, MAY_EXEC);
 }
 
 /*
@@ -1692,7 +1687,7 @@ struct file *do_filp_open(int dfd, const char *pathname,
 	int will_write;
 	int flag = open_to_namei_flags(open_flag);
 
-	acc_mode = ACC_MODE(flag);
+	acc_mode = MAY_OPEN | ACC_MODE(flag);
 
 	/* O_TRUNC implies we need access checks for write permissions */
 	if (flag & O_TRUNC)
diff --git a/include/linux/security.h b/include/linux/security.h
index f0e9adb22ac2..fd96e7f8a6f9 100644
--- a/include/linux/security.h
+++ b/include/linux/security.h
@@ -1362,7 +1362,7 @@ struct security_operations {
 			     struct inode *new_dir, struct dentry *new_dentry);
 	int (*inode_readlink) (struct dentry *dentry);
 	int (*inode_follow_link) (struct dentry *dentry, struct nameidata *nd);
-	int (*inode_permission) (struct inode *inode, int mask, struct nameidata *nd);
+	int (*inode_permission) (struct inode *inode, int mask);
 	int (*inode_setattr)	(struct dentry *dentry, struct iattr *attr);
 	int (*inode_getattr) (struct vfsmount *mnt, struct dentry *dentry);
 	void (*inode_delete) (struct inode *inode);
@@ -1628,7 +1628,7 @@ int security_inode_rename(struct inode *old_dir, struct dentry *old_dentry,
 			  struct inode *new_dir, struct dentry *new_dentry);
 int security_inode_readlink(struct dentry *dentry);
 int security_inode_follow_link(struct dentry *dentry, struct nameidata *nd);
-int security_inode_permission(struct inode *inode, int mask, struct nameidata *nd);
+int security_inode_permission(struct inode *inode, int mask);
 int security_inode_setattr(struct dentry *dentry, struct iattr *attr);
 int security_inode_getattr(struct vfsmount *mnt, struct dentry *dentry);
 void security_inode_delete(struct inode *inode);
@@ -2021,8 +2021,7 @@ static inline int security_inode_follow_link(struct dentry *dentry,
 	return 0;
 }
 
-static inline int security_inode_permission(struct inode *inode, int mask,
-					     struct nameidata *nd)
+static inline int security_inode_permission(struct inode *inode, int mask)
 {
 	return 0;
 }
diff --git a/security/capability.c b/security/capability.c
index 5b01c0b02422..63d10da515a5 100644
--- a/security/capability.c
+++ b/security/capability.c
@@ -211,8 +211,7 @@ static int cap_inode_follow_link(struct dentry *dentry,
 	return 0;
 }
 
-static int cap_inode_permission(struct inode *inode, int mask,
-				struct nameidata *nd)
+static int cap_inode_permission(struct inode *inode, int mask)
 {
 	return 0;
 }
diff --git a/security/security.c b/security/security.c
index 59f23b5918b3..78ed3ffde242 100644
--- a/security/security.c
+++ b/security/security.c
@@ -429,11 +429,11 @@ int security_inode_follow_link(struct dentry *dentry, struct nameidata *nd)
 	return security_ops->inode_follow_link(dentry, nd);
 }
 
-int security_inode_permission(struct inode *inode, int mask, struct nameidata *nd)
+int security_inode_permission(struct inode *inode, int mask)
 {
 	if (unlikely(IS_PRIVATE(inode)))
 		return 0;
-	return security_ops->inode_permission(inode, mask, nd);
+	return security_ops->inode_permission(inode, mask);
 }
 
 int security_inode_setattr(struct dentry *dentry, struct iattr *attr)
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 3481cde5bf15..5ba13908b5b4 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -2624,12 +2624,11 @@ static int selinux_inode_follow_link(struct dentry *dentry, struct nameidata *na
 	return dentry_has_perm(current, NULL, dentry, FILE__READ);
 }
 
-static int selinux_inode_permission(struct inode *inode, int mask,
-				    struct nameidata *nd)
+static int selinux_inode_permission(struct inode *inode, int mask)
 {
 	int rc;
 
-	rc = secondary_ops->inode_permission(inode, mask, nd);
+	rc = secondary_ops->inode_permission(inode, mask);
 	if (rc)
 		return rc;
 
diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c
index ee5a51cbc5eb..1b40e558f983 100644
--- a/security/smack/smack_lsm.c
+++ b/security/smack/smack_lsm.c
@@ -522,8 +522,7 @@ static int smack_inode_rename(struct inode *old_inode,
  *
  * Returns 0 if access is permitted, -EACCES otherwise
  */
-static int smack_inode_permission(struct inode *inode, int mask,
-				  struct nameidata *nd)
+static int smack_inode_permission(struct inode *inode, int mask)
 {
 	/*
 	 * No permission to check. Existence test. Yup, it's there.

From 672b16b2f66c149888bd876a4f92342112205fe1 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Thu, 17 Jul 2008 09:45:01 -0400
Subject: [PATCH 1144/1171] [PATCH] more nameidata removal:
 exec_permission_lite() doesn't need it

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/namei.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/fs/namei.c b/fs/namei.c
index 6b0e8e5e079e..6d75430358ac 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -460,8 +460,7 @@ static struct dentry * cached_lookup(struct dentry * parent, struct qstr * name,
  * short-cut DAC fails, then call permission() to do more
  * complete permission check.
  */
-static int exec_permission_lite(struct inode *inode,
-				       struct nameidata *nd)
+static int exec_permission_lite(struct inode *inode)
 {
 	umode_t	mode = inode->i_mode;
 
@@ -884,7 +883,7 @@ static int __link_path_walk(const char *name, struct nameidata *nd)
 		unsigned int c;
 
 		nd->flags |= LOOKUP_CONTINUE;
-		err = exec_permission_lite(inode, nd);
+		err = exec_permission_lite(inode);
 		if (err == -EAGAIN)
 			err = vfs_permission(nd, MAY_EXEC);
  		if (err)

From 88b387824fdaecb6ba0f471acf0aadf7d24739fd Mon Sep 17 00:00:00 2001
From: Li Zefan <lizf@cn.fujitsu.com>
Date: Mon, 21 Jul 2008 18:06:36 +0800
Subject: [PATCH 1145/1171] [PATCH] vfs: use kstrdup() and check failing
 allocation

- use kstrdup() instead of kmalloc() + memcpy()
- return NULL if allocating ->mnt_devname failed
- mnt_devname should be const

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Acked-by: Cyrill Gorcunov <gorcunov@gmail.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/namespace.c        | 24 +++++++++++++-----------
 include/linux/mount.h |  2 +-
 2 files changed, 14 insertions(+), 12 deletions(-)

diff --git a/fs/namespace.c b/fs/namespace.c
index c4fcf48acef8..26380f599534 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -112,9 +112,13 @@ struct vfsmount *alloc_vfsmnt(const char *name)
 		int err;
 
 		err = mnt_alloc_id(mnt);
-		if (err) {
-			kmem_cache_free(mnt_cache, mnt);
-			return NULL;
+		if (err)
+			goto out_free_cache;
+
+		if (name) {
+			mnt->mnt_devname = kstrdup(name, GFP_KERNEL);
+			if (!mnt->mnt_devname)
+				goto out_free_id;
 		}
 
 		atomic_set(&mnt->mnt_count, 1);
@@ -127,16 +131,14 @@ struct vfsmount *alloc_vfsmnt(const char *name)
 		INIT_LIST_HEAD(&mnt->mnt_slave_list);
 		INIT_LIST_HEAD(&mnt->mnt_slave);
 		atomic_set(&mnt->__mnt_writers, 0);
-		if (name) {
-			int size = strlen(name) + 1;
-			char *newname = kmalloc(size, GFP_KERNEL);
-			if (newname) {
-				memcpy(newname, name, size);
-				mnt->mnt_devname = newname;
-			}
-		}
 	}
 	return mnt;
+
+out_free_id:
+	mnt_free_id(mnt);
+out_free_cache:
+	kmem_cache_free(mnt_cache, mnt);
+	return NULL;
 }
 
 /*
diff --git a/include/linux/mount.h b/include/linux/mount.h
index 4374d1adeb4b..b5efaa2132ab 100644
--- a/include/linux/mount.h
+++ b/include/linux/mount.h
@@ -47,7 +47,7 @@ struct vfsmount {
 	struct list_head mnt_child;	/* and going through their mnt_child */
 	int mnt_flags;
 	/* 4 bytes hole on 64bits arches */
-	char *mnt_devname;		/* Name of device e.g. /dev/dsk/hda1 */
+	const char *mnt_devname;	/* Name of device e.g. /dev/dsk/hda1 */
 	struct list_head mnt_list;
 	struct list_head mnt_expire;	/* link in fs-specific expiry list */
 	struct list_head mnt_share;	/* circular list of shared mounts */

From 9767d74957450da6365c363d69e3d02d605d7375 Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Tue, 1 Jul 2008 15:01:26 +0200
Subject: [PATCH 1146/1171] [patch 1/4] vfs: utimes: move owner check into
 inode_change_ok()

Add a new ia_valid flag: ATTR_TIMES_SET, to handle the
UTIMES_OMIT/UTIMES_NOW and UTIMES_NOW/UTIMES_OMIT cases.  In these
cases neither ATTR_MTIME_SET nor ATTR_ATIME_SET is in the flags, yet
the POSIX draft specifies that permission checking is performed the
same way as if one or both of the times was explicitly set to a
timestamp.

See the path "vfs: utimensat(): fix error checking for
{UTIME_NOW,UTIME_OMIT} case" by Michael Kerrisk for the patch
introducing this behavior.

This is a cleanup, as well as allowing filesystems (NFS/fuse/...) to
perform their own permission checking instead of the default.

CC: Ulrich Drepper <drepper@redhat.com>
CC: Michael Kerrisk <mtk.manpages@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/attr.c          |  2 +-
 fs/utimes.c        | 17 ++++-------------
 include/linux/fs.h | 33 +++++++++++++++++----------------
 3 files changed, 22 insertions(+), 30 deletions(-)

diff --git a/fs/attr.c b/fs/attr.c
index 966b73e25f82..765fc75fab3b 100644
--- a/fs/attr.c
+++ b/fs/attr.c
@@ -51,7 +51,7 @@ int inode_change_ok(struct inode *inode, struct iattr *attr)
 	}
 
 	/* Check for setting the inode time. */
-	if (ia_valid & (ATTR_MTIME_SET | ATTR_ATIME_SET)) {
+	if (ia_valid & (ATTR_MTIME_SET | ATTR_ATIME_SET | ATTR_TIMES_SET)) {
 		if (!is_owner_or_cap(inode))
 			goto error;
 	}
diff --git a/fs/utimes.c b/fs/utimes.c
index b6b664e7145e..ecf8941ba34a 100644
--- a/fs/utimes.c
+++ b/fs/utimes.c
@@ -101,7 +101,6 @@ long do_utimes(int dfd, char __user *filename, struct timespec *times, int flags
 		     times[1].tv_nsec == UTIME_NOW)
 		times = NULL;
 
-	/* In most cases, the checks are done in inode_change_ok() */
 	newattrs.ia_valid = ATTR_CTIME | ATTR_MTIME | ATTR_ATIME;
 	if (times) {
 		error = -EPERM;
@@ -123,21 +122,13 @@ long do_utimes(int dfd, char __user *filename, struct timespec *times, int flags
 			newattrs.ia_mtime.tv_nsec = times[1].tv_nsec;
 			newattrs.ia_valid |= ATTR_MTIME_SET;
 		}
-
 		/*
-		 * For the UTIME_OMIT/UTIME_NOW and UTIME_NOW/UTIME_OMIT
-		 * cases, we need to make an extra check that is not done by
-		 * inode_change_ok().
+		 * Tell inode_change_ok(), that this is an explicit time
+		 * update, even if neither ATTR_ATIME_SET nor ATTR_MTIME_SET
+		 * were used.
 		 */
-		if (((times[0].tv_nsec == UTIME_NOW &&
-			    times[1].tv_nsec == UTIME_OMIT)
-		     ||
-		     (times[0].tv_nsec == UTIME_OMIT &&
-			    times[1].tv_nsec == UTIME_NOW))
-		    && !is_owner_or_cap(inode))
-			goto mnt_drop_write_and_out;
+		newattrs.ia_valid |= ATTR_TIMES_SET;
 	} else {
-
 		/*
 		 * If times is NULL (or both times are UTIME_NOW),
 		 * then we need to check permissions, because
diff --git a/include/linux/fs.h b/include/linux/fs.h
index d8721e818b45..527b9e482f99 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -320,22 +320,23 @@ typedef void (dio_iodone_t)(struct kiocb *iocb, loff_t offset,
  * Attribute flags.  These should be or-ed together to figure out what
  * has been changed!
  */
-#define ATTR_MODE	1
-#define ATTR_UID	2
-#define ATTR_GID	4
-#define ATTR_SIZE	8
-#define ATTR_ATIME	16
-#define ATTR_MTIME	32
-#define ATTR_CTIME	64
-#define ATTR_ATIME_SET	128
-#define ATTR_MTIME_SET	256
-#define ATTR_FORCE	512	/* Not a change, but a change it */
-#define ATTR_ATTR_FLAG	1024
-#define ATTR_KILL_SUID	2048
-#define ATTR_KILL_SGID	4096
-#define ATTR_FILE	8192
-#define ATTR_KILL_PRIV	16384
-#define ATTR_OPEN	32768	/* Truncating from open(O_TRUNC) */
+#define ATTR_MODE	(1 << 0)
+#define ATTR_UID	(1 << 1)
+#define ATTR_GID	(1 << 2)
+#define ATTR_SIZE	(1 << 3)
+#define ATTR_ATIME	(1 << 4)
+#define ATTR_MTIME	(1 << 5)
+#define ATTR_CTIME	(1 << 6)
+#define ATTR_ATIME_SET	(1 << 7)
+#define ATTR_MTIME_SET	(1 << 8)
+#define ATTR_FORCE	(1 << 9) /* Not a change, but a change it */
+#define ATTR_ATTR_FLAG	(1 << 10)
+#define ATTR_KILL_SUID	(1 << 11)
+#define ATTR_KILL_SGID	(1 << 12)
+#define ATTR_FILE	(1 << 13)
+#define ATTR_KILL_PRIV	(1 << 14)
+#define ATTR_OPEN	(1 << 15) /* Truncating from open(O_TRUNC) */
+#define ATTR_TIMES_SET	(1 << 16)
 
 /*
  * This is the Inode Attributes structure, used for notify_change().  It

From e9b76fedc61235da80b6b7f81dfd67ec224dfb49 Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Tue, 1 Jul 2008 15:01:27 +0200
Subject: [PATCH 1147/1171] [patch 2/4] vfs: utimes cleanup

Untange the mess that is do_utimes().  Add kerneldoc comment to
do_utimes().

CC: Ulrich Drepper <drepper@redhat.com>
CC: Michael Kerrisk <mtk.manpages@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/utimes.c | 114 ++++++++++++++++++++++++++++++----------------------
 1 file changed, 65 insertions(+), 49 deletions(-)

diff --git a/fs/utimes.c b/fs/utimes.c
index ecf8941ba34a..8e09dbdfd7f5 100644
--- a/fs/utimes.c
+++ b/fs/utimes.c
@@ -48,54 +48,15 @@ static bool nsec_valid(long nsec)
 	return nsec >= 0 && nsec <= 999999999;
 }
 
-/* If times==NULL, set access and modification to current time,
- * must be owner or have write permission.
- * Else, update from *times, must be owner or super user.
- */
-long do_utimes(int dfd, char __user *filename, struct timespec *times, int flags)
+static int utimes_common(struct path *path, struct timespec *times)
 {
 	int error;
-	struct nameidata nd;
-	struct dentry *dentry;
-	struct inode *inode;
 	struct iattr newattrs;
-	struct file *f = NULL;
-	struct vfsmount *mnt;
+	struct inode *inode = path->dentry->d_inode;
 
-	error = -EINVAL;
-	if (times && (!nsec_valid(times[0].tv_nsec) ||
-		      !nsec_valid(times[1].tv_nsec))) {
-		goto out;
-	}
-
-	if (flags & ~AT_SYMLINK_NOFOLLOW)
-		goto out;
-
-	if (filename == NULL && dfd != AT_FDCWD) {
-		error = -EINVAL;
-		if (flags & AT_SYMLINK_NOFOLLOW)
-			goto out;
-
-		error = -EBADF;
-		f = fget(dfd);
-		if (!f)
-			goto out;
-		dentry = f->f_path.dentry;
-		mnt = f->f_path.mnt;
-	} else {
-		error = __user_walk_fd(dfd, filename, (flags & AT_SYMLINK_NOFOLLOW) ? 0 : LOOKUP_FOLLOW, &nd);
-		if (error)
-			goto out;
-
-		dentry = nd.path.dentry;
-		mnt = nd.path.mnt;
-	}
-
-	inode = dentry->d_inode;
-
-	error = mnt_want_write(mnt);
+	error = mnt_want_write(path->mnt);
 	if (error)
-		goto dput_and_out;
+		goto out;
 
 	if (times && times[0].tv_nsec == UTIME_NOW &&
 		     times[1].tv_nsec == UTIME_NOW)
@@ -145,15 +106,70 @@ long do_utimes(int dfd, char __user *filename, struct timespec *times, int flags
 		}
 	}
 	mutex_lock(&inode->i_mutex);
-	error = notify_change(dentry, &newattrs);
+	error = notify_change(path->dentry, &newattrs);
 	mutex_unlock(&inode->i_mutex);
+
 mnt_drop_write_and_out:
-	mnt_drop_write(mnt);
-dput_and_out:
-	if (f)
-		fput(f);
-	else
+	mnt_drop_write(path->mnt);
+out:
+	return error;
+}
+
+/*
+ * do_utimes - change times on filename or file descriptor
+ * @dfd: open file descriptor, -1 or AT_FDCWD
+ * @filename: path name or NULL
+ * @times: new times or NULL
+ * @flags: zero or more flags (only AT_SYMLINK_NOFOLLOW for the moment)
+ *
+ * If filename is NULL and dfd refers to an open file, then operate on
+ * the file.  Otherwise look up filename, possibly using dfd as a
+ * starting point.
+ *
+ * If times==NULL, set access and modification to current time,
+ * must be owner or have write permission.
+ * Else, update from *times, must be owner or super user.
+ */
+long do_utimes(int dfd, char __user *filename, struct timespec *times, int flags)
+{
+	int error = -EINVAL;
+
+	if (times && (!nsec_valid(times[0].tv_nsec) ||
+		      !nsec_valid(times[1].tv_nsec))) {
+		goto out;
+	}
+
+	if (flags & ~AT_SYMLINK_NOFOLLOW)
+		goto out;
+
+	if (filename == NULL && dfd != AT_FDCWD) {
+		struct file *file;
+
+		if (flags & AT_SYMLINK_NOFOLLOW)
+			goto out;
+
+		file = fget(dfd);
+		error = -EBADF;
+		if (!file)
+			goto out;
+
+		error = utimes_common(&file->f_path, times);
+		fput(file);
+	} else {
+		struct nameidata nd;
+		int lookup_flags = 0;
+
+		if (!(flags & AT_SYMLINK_NOFOLLOW))
+			lookup_flags |= LOOKUP_FOLLOW;
+
+		error = __user_walk_fd(dfd, filename, lookup_flags, &nd);
+		if (error)
+			goto out;
+
+		error = utimes_common(&nd.path, times);
 		path_put(&nd.path);
+	}
+
 out:
 	return error;
 }

From b1da47e29e467f1ec36dc78d009bfb109fd533c7 Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Tue, 1 Jul 2008 15:01:28 +0200
Subject: [PATCH 1148/1171] [patch 3/4] fat: dont call notify_change

The FAT_IOCTL_SET_ATTRIBUTES ioctl() calls notify_change() to change
the file mode before changing the inode attributes.  Replace with
explicit calls to security_inode_setattr(), fat_setattr() and
fsnotify_change().

This is equivalent to the original.  The reason it is needed, is that
later in the series we move the immutable check into notify_change().
That would break the FAT_IOCTL_SET_ATTRIBUTES ioctl, as it needs to
perform the mode change regardless of the immutability of the file.

[Fix error if fat is built as a module.  Thanks to OGAWA Hirofumi for
noticing.]

Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Acked-by: OGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/fat/file.c       | 17 +++++++++++++++--
 security/security.c |  1 +
 2 files changed, 16 insertions(+), 2 deletions(-)

diff --git a/fs/fat/file.c b/fs/fat/file.c
index c672df4036e9..8707a8cfa02c 100644
--- a/fs/fat/file.c
+++ b/fs/fat/file.c
@@ -15,6 +15,8 @@
 #include <linux/writeback.h>
 #include <linux/backing-dev.h>
 #include <linux/blkdev.h>
+#include <linux/fsnotify.h>
+#include <linux/security.h>
 
 int fat_generic_ioctl(struct inode *inode, struct file *filp,
 		      unsigned int cmd, unsigned long arg)
@@ -64,6 +66,7 @@ int fat_generic_ioctl(struct inode *inode, struct file *filp,
 
 		/* Equivalent to a chmod() */
 		ia.ia_valid = ATTR_MODE | ATTR_CTIME;
+		ia.ia_ctime = current_fs_time(inode->i_sb);
 		if (is_dir) {
 			ia.ia_mode = MSDOS_MKMODE(attr,
 				S_IRWXUGO & ~sbi->options.fs_dmask)
@@ -90,11 +93,21 @@ int fat_generic_ioctl(struct inode *inode, struct file *filp,
 			}
 		}
 
-		/* This MUST be done before doing anything irreversible... */
-		err = notify_change(filp->f_path.dentry, &ia);
+		/*
+		 * The security check is questionable...  We single
+		 * out the RO attribute for checking by the security
+		 * module, just because it maps to a file mode.
+		 */
+		err = security_inode_setattr(filp->f_path.dentry, &ia);
 		if (err)
 			goto up;
 
+		/* This MUST be done before doing anything irreversible... */
+		err = fat_setattr(filp->f_path.dentry, &ia);
+		if (err)
+			goto up;
+
+		fsnotify_change(filp->f_path.dentry, ia.ia_valid);
 		if (sbi->options.sys_immutable) {
 			if (attr & ATTR_SYS)
 				inode->i_flags |= S_IMMUTABLE;
diff --git a/security/security.c b/security/security.c
index 78ed3ffde242..ff7068727757 100644
--- a/security/security.c
+++ b/security/security.c
@@ -442,6 +442,7 @@ int security_inode_setattr(struct dentry *dentry, struct iattr *attr)
 		return 0;
 	return security_ops->inode_setattr(dentry, attr);
 }
+EXPORT_SYMBOL_GPL(security_inode_setattr);
 
 int security_inode_getattr(struct vfsmount *mnt, struct dentry *dentry)
 {

From beb29e058c35ab69e96e455a12ccf7505f6de425 Mon Sep 17 00:00:00 2001
From: Miklos Szeredi <mszeredi@suse.cz>
Date: Tue, 1 Jul 2008 15:01:29 +0200
Subject: [PATCH 1149/1171] [patch 4/4] vfs: immutable inode checking cleanup

Move the immutable and append-only checks from chmod, chown and utimes
into notify_change().  Checks for immutable and append-only files are
always performed by the VFS and not by the filesystem (see
permission() and may_...() in namei.c), so these belong in
notify_change(), and not in inode_change_ok().

This should be completely equivalent.

CC: Ulrich Drepper <drepper@redhat.com>
CC: Michael Kerrisk <mtk.manpages@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/attr.c   |  5 +++++
 fs/open.c   | 24 ++----------------------
 fs/utimes.c |  4 ----
 3 files changed, 7 insertions(+), 26 deletions(-)

diff --git a/fs/attr.c b/fs/attr.c
index 765fc75fab3b..26c71ba1eed4 100644
--- a/fs/attr.c
+++ b/fs/attr.c
@@ -108,6 +108,11 @@ int notify_change(struct dentry * dentry, struct iattr * attr)
 	struct timespec now;
 	unsigned int ia_valid = attr->ia_valid;
 
+	if (ia_valid & (ATTR_MODE | ATTR_UID | ATTR_GID | ATTR_TIMES_SET)) {
+		if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
+			return -EPERM;
+	}
+
 	now = current_fs_time(inode->i_sb);
 
 	attr->ia_ctime = now;
diff --git a/fs/open.c b/fs/open.c
index 3317e1909b2c..3b3c43674be3 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -588,9 +588,6 @@ asmlinkage long sys_fchmod(unsigned int fd, mode_t mode)
 	err = mnt_want_write(file->f_path.mnt);
 	if (err)
 		goto out_putf;
-	err = -EPERM;
-	if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
-		goto out_drop_write;
 	mutex_lock(&inode->i_mutex);
 	if (mode == (mode_t) -1)
 		mode = inode->i_mode;
@@ -598,8 +595,6 @@ asmlinkage long sys_fchmod(unsigned int fd, mode_t mode)
 	newattrs.ia_valid = ATTR_MODE | ATTR_CTIME;
 	err = notify_change(dentry, &newattrs);
 	mutex_unlock(&inode->i_mutex);
-
-out_drop_write:
 	mnt_drop_write(file->f_path.mnt);
 out_putf:
 	fput(file);
@@ -623,11 +618,6 @@ asmlinkage long sys_fchmodat(int dfd, const char __user *filename,
 	error = mnt_want_write(nd.path.mnt);
 	if (error)
 		goto dput_and_out;
-
-	error = -EPERM;
-	if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
-		goto out_drop_write;
-
 	mutex_lock(&inode->i_mutex);
 	if (mode == (mode_t) -1)
 		mode = inode->i_mode;
@@ -635,8 +625,6 @@ asmlinkage long sys_fchmodat(int dfd, const char __user *filename,
 	newattrs.ia_valid = ATTR_MODE | ATTR_CTIME;
 	error = notify_change(nd.path.dentry, &newattrs);
 	mutex_unlock(&inode->i_mutex);
-
-out_drop_write:
 	mnt_drop_write(nd.path.mnt);
 dput_and_out:
 	path_put(&nd.path);
@@ -651,18 +639,10 @@ asmlinkage long sys_chmod(const char __user *filename, mode_t mode)
 
 static int chown_common(struct dentry * dentry, uid_t user, gid_t group)
 {
-	struct inode * inode;
+	struct inode *inode = dentry->d_inode;
 	int error;
 	struct iattr newattrs;
 
-	error = -ENOENT;
-	if (!(inode = dentry->d_inode)) {
-		printk(KERN_ERR "chown_common: NULL inode\n");
-		goto out;
-	}
-	error = -EPERM;
-	if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
-		goto out;
 	newattrs.ia_valid =  ATTR_CTIME;
 	if (user != (uid_t) -1) {
 		newattrs.ia_valid |= ATTR_UID;
@@ -678,7 +658,7 @@ static int chown_common(struct dentry * dentry, uid_t user, gid_t group)
 	mutex_lock(&inode->i_mutex);
 	error = notify_change(dentry, &newattrs);
 	mutex_unlock(&inode->i_mutex);
-out:
+
 	return error;
 }
 
diff --git a/fs/utimes.c b/fs/utimes.c
index 8e09dbdfd7f5..dad679d3a158 100644
--- a/fs/utimes.c
+++ b/fs/utimes.c
@@ -64,10 +64,6 @@ static int utimes_common(struct path *path, struct timespec *times)
 
 	newattrs.ia_valid = ATTR_CTIME | ATTR_MTIME | ATTR_ATIME;
 	if (times) {
-		error = -EPERM;
-                if (IS_APPEND(inode) || IS_IMMUTABLE(inode))
-			goto mnt_drop_write_and_out;
-
 		if (times[0].tv_nsec == UTIME_OMIT)
 			newattrs.ia_valid &= ~ATTR_ATIME;
 		else if (times[0].tv_nsec != UTIME_NOW) {

From e56b6a5dda1a36ffaa532df6f975ea324298fa4d Mon Sep 17 00:00:00 2001
From: Christoph Hellwig <hch@lst.de>
Date: Mon, 19 May 2008 07:53:34 +0200
Subject: [PATCH 1150/1171] Re: [PATCH 3/6] vfs: open_exec cleanup

On Mon, May 19, 2008 at 12:01:49AM +0200, Marcin Slusarz wrote:
> open_exec is needlessly indented, calls ERR_PTR with 0 argument
> (which is not valid errno) and jumps into middle of function
> just to return value.
> So clean it up a bit.

Still looks rather messy.  See below for a better version.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/exec.c | 56 ++++++++++++++++++++++++++++---------------------------
 1 file changed, 29 insertions(+), 27 deletions(-)

diff --git a/fs/exec.c b/fs/exec.c
index 0ba5d355c5a1..346e3f69c6e0 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -656,38 +656,40 @@ EXPORT_SYMBOL(setup_arg_pages);
 struct file *open_exec(const char *name)
 {
 	struct nameidata nd;
-	int err;
 	struct file *file;
+	int err;
 
-	err = path_lookup_open(AT_FDCWD, name, LOOKUP_FOLLOW, &nd, FMODE_READ|FMODE_EXEC);
-	file = ERR_PTR(err);
+	err = path_lookup_open(AT_FDCWD, name, LOOKUP_FOLLOW, &nd,
+				FMODE_READ|FMODE_EXEC);
+	if (err)
+		goto out;
 
-	if (!err) {
-		struct inode *inode = nd.path.dentry->d_inode;
-		file = ERR_PTR(-EACCES);
-		if (S_ISREG(inode->i_mode)) {
-			int err = vfs_permission(&nd, MAY_EXEC | MAY_OPEN);
-			file = ERR_PTR(err);
-			if (!err) {
-				file = nameidata_to_filp(&nd,
-							O_RDONLY|O_LARGEFILE);
-				if (!IS_ERR(file)) {
-					err = deny_write_access(file);
-					if (err) {
-						fput(file);
-						file = ERR_PTR(err);
-					}
-				}
-out:
-				return file;
-			}
-		}
-		release_open_intent(&nd);
-		path_put(&nd.path);
+	err = -EACCES;
+	if (!S_ISREG(nd.path.dentry->d_inode->i_mode))
+		goto out_path_put;
+
+	err = vfs_permission(&nd, MAY_EXEC | MAY_OPEN);
+	if (err)
+		goto out_path_put;
+
+	file = nameidata_to_filp(&nd, O_RDONLY|O_LARGEFILE);
+	if (IS_ERR(file))
+		return file;
+
+	err = deny_write_access(file);
+	if (err) {
+		fput(file);
+		goto out;
 	}
-	goto out;
-}
 
+	return file;
+
+ out_path_put:
+	release_open_intent(&nd);
+	path_put(&nd.path);
+ out:
+	return ERR_PTR(err);
+}
 EXPORT_SYMBOL(open_exec);
 
 int kernel_read(struct file *file, unsigned long offset,

From 30524472c2f728c20d6bf35191042a5d455c0a64 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 22 Jul 2008 00:02:33 -0400
Subject: [PATCH 1151/1171] [PATCH] take noexec checks to very few callers that
 care

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/exec.c  |  7 +++++++
 fs/namei.c |  9 ---------
 fs/open.c  | 10 ++++++++++
 3 files changed, 17 insertions(+), 9 deletions(-)

diff --git a/fs/exec.c b/fs/exec.c
index 346e3f69c6e0..eca58c29eded 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -118,6 +118,10 @@ asmlinkage long sys_uselib(const char __user * library)
 	if (!S_ISREG(nd.path.dentry->d_inode->i_mode))
 		goto exit;
 
+	error = -EACCES;
+	if (nd.path.mnt->mnt_flags & MNT_NOEXEC)
+		goto exit;
+
 	error = vfs_permission(&nd, MAY_READ | MAY_EXEC | MAY_OPEN);
 	if (error)
 		goto exit;
@@ -668,6 +672,9 @@ struct file *open_exec(const char *name)
 	if (!S_ISREG(nd.path.dentry->d_inode->i_mode))
 		goto out_path_put;
 
+	if (nd.path.mnt->mnt_flags & MNT_NOEXEC)
+		goto out_path_put;
+
 	err = vfs_permission(&nd, MAY_EXEC | MAY_OPEN);
 	if (err)
 		goto out_path_put;
diff --git a/fs/namei.c b/fs/namei.c
index 6d75430358ac..396cb3e5c364 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -252,15 +252,6 @@ int permission(struct inode *inode, int mask, struct nameidata *nd)
 			return -EACCES;
 	}
 
-	if ((mask & MAY_EXEC) && S_ISREG(inode->i_mode)) {
-		/*
-		 * MAY_EXEC on regular files is denied if the fs is mounted
-		 * with the "noexec" flag.
-		 */
-		if (mnt && (mnt->mnt_flags & MNT_NOEXEC))
-			return -EACCES;
-	}
-
 	/* Ordinary permission routines do not understand MAY_APPEND. */
 	if (inode->i_op && inode->i_op->permission) {
 		retval = inode->i_op->permission(inode, mask);
diff --git a/fs/open.c b/fs/open.c
index 3b3c43674be3..d5e421ad0cf6 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -461,6 +461,16 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode)
 	if (res)
 		goto out;
 
+	if ((mode & MAY_EXEC) && S_ISREG(nd.path.dentry->d_inode->i_mode)) {
+		/*
+		 * MAY_EXEC on regular files is denied if the fs is mounted
+		 * with the "noexec" flag.
+		 */
+		res = -EACCES;
+		if (nd.path.mnt->mnt_flags & MNT_NOEXEC)
+			goto out_path_release;
+	}
+
 	res = vfs_permission(&nd, mode | MAY_ACCESS);
 	/* SuS v2 requires we report a read only fs too */
 	if(res || !(mode & S_IWOTH) ||

From f419a2e3b64def707e1384ee38abb77f99af5f6d Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 22 Jul 2008 00:07:17 -0400
Subject: [PATCH 1152/1171] [PATCH] kill nameidata passing to permission(),
 rename to inode_permission()

Incidentally, the name that gives hundreds of false positives on grep
is not a good idea...

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/ecryptfs/inode.c |  2 +-
 fs/namei.c          | 22 +++++++++-------------
 fs/nfsd/nfsfh.c     |  2 +-
 fs/nfsd/vfs.c       |  4 ++--
 fs/utimes.c         |  2 +-
 fs/xattr.c          |  2 +-
 include/linux/fs.h  |  2 +-
 ipc/mqueue.c        |  2 +-
 8 files changed, 17 insertions(+), 21 deletions(-)

diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c
index f25caf2b0887..89209f00f9c7 100644
--- a/fs/ecryptfs/inode.c
+++ b/fs/ecryptfs/inode.c
@@ -830,7 +830,7 @@ out:
 static int
 ecryptfs_permission(struct inode *inode, int mask)
 {
-	return permission(ecryptfs_inode_to_lower(inode), mask, NULL);
+	return inode_permission(ecryptfs_inode_to_lower(inode), mask);
 }
 
 /**
diff --git a/fs/namei.c b/fs/namei.c
index 396cb3e5c364..5029b93ebbd5 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -227,13 +227,9 @@ int generic_permission(struct inode *inode, int mask,
 	return -EACCES;
 }
 
-int permission(struct inode *inode, int mask, struct nameidata *nd)
+int inode_permission(struct inode *inode, int mask)
 {
 	int retval;
-	struct vfsmount *mnt = NULL;
-
-	if (nd)
-		mnt = nd->path.mnt;
 
 	if (mask & MAY_WRITE) {
 		umode_t mode = inode->i_mode;
@@ -293,7 +289,7 @@ int permission(struct inode *inode, int mask, struct nameidata *nd)
  */
 int vfs_permission(struct nameidata *nd, int mask)
 {
-	return permission(nd->path.dentry->d_inode, mask, nd);
+	return inode_permission(nd->path.dentry->d_inode, mask);
 }
 
 /**
@@ -310,7 +306,7 @@ int vfs_permission(struct nameidata *nd, int mask)
  */
 int file_permission(struct file *file, int mask)
 {
-	return permission(file->f_path.dentry->d_inode, mask, NULL);
+	return inode_permission(file->f_path.dentry->d_inode, mask);
 }
 
 /*
@@ -1262,7 +1258,7 @@ static struct dentry *lookup_hash(struct nameidata *nd)
 {
 	int err;
 
-	err = permission(nd->path.dentry->d_inode, MAY_EXEC, nd);
+	err = inode_permission(nd->path.dentry->d_inode, MAY_EXEC);
 	if (err)
 		return ERR_PTR(err);
 	return __lookup_hash(&nd->last, nd->path.dentry, nd);
@@ -1310,7 +1306,7 @@ struct dentry *lookup_one_len(const char *name, struct dentry *base, int len)
 	if (err)
 		return ERR_PTR(err);
 
-	err = permission(base->d_inode, MAY_EXEC, NULL);
+	err = inode_permission(base->d_inode, MAY_EXEC);
 	if (err)
 		return ERR_PTR(err);
 	return __lookup_hash(&this, base, NULL);
@@ -1400,7 +1396,7 @@ static int may_delete(struct inode *dir,struct dentry *victim,int isdir)
 	BUG_ON(victim->d_parent->d_inode != dir);
 	audit_inode_child(victim->d_name.name, victim, dir);
 
-	error = permission(dir,MAY_WRITE | MAY_EXEC, NULL);
+	error = inode_permission(dir, MAY_WRITE | MAY_EXEC);
 	if (error)
 		return error;
 	if (IS_APPEND(dir))
@@ -1437,7 +1433,7 @@ static inline int may_create(struct inode *dir, struct dentry *child,
 		return -EEXIST;
 	if (IS_DEADDIR(dir))
 		return -ENOENT;
-	return permission(dir,MAY_WRITE | MAY_EXEC, nd);
+	return inode_permission(dir, MAY_WRITE | MAY_EXEC);
 }
 
 /* 
@@ -2543,7 +2539,7 @@ static int vfs_rename_dir(struct inode *old_dir, struct dentry *old_dentry,
 	 * we'll need to flip '..'.
 	 */
 	if (new_dir != old_dir) {
-		error = permission(old_dentry->d_inode, MAY_WRITE, NULL);
+		error = inode_permission(old_dentry->d_inode, MAY_WRITE);
 		if (error)
 			return error;
 	}
@@ -2897,7 +2893,7 @@ EXPORT_SYMBOL(page_symlink);
 EXPORT_SYMBOL(page_symlink_inode_operations);
 EXPORT_SYMBOL(path_lookup);
 EXPORT_SYMBOL(vfs_path_lookup);
-EXPORT_SYMBOL(permission);
+EXPORT_SYMBOL(inode_permission);
 EXPORT_SYMBOL(vfs_permission);
 EXPORT_SYMBOL(file_permission);
 EXPORT_SYMBOL(unlock_rename);
diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c
index f45451eb1e38..ea37c96f0445 100644
--- a/fs/nfsd/nfsfh.c
+++ b/fs/nfsd/nfsfh.c
@@ -51,7 +51,7 @@ static int nfsd_acceptable(void *expv, struct dentry *dentry)
 		/* make sure parents give x permission to user */
 		int err;
 		parent = dget_parent(tdentry);
-		err = permission(parent->d_inode, MAY_EXEC, NULL);
+		err = inode_permission(parent->d_inode, MAY_EXEC);
 		if (err < 0) {
 			dput(parent);
 			break;
diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c
index ad1ad59e3742..18060bed5267 100644
--- a/fs/nfsd/vfs.c
+++ b/fs/nfsd/vfs.c
@@ -1953,12 +1953,12 @@ nfsd_permission(struct svc_rqst *rqstp, struct svc_export *exp,
 		return 0;
 
 	/* This assumes  NFSD_MAY_{READ,WRITE,EXEC} == MAY_{READ,WRITE,EXEC} */
-	err = permission(inode, acc & (MAY_READ|MAY_WRITE|MAY_EXEC), NULL);
+	err = inode_permission(inode, acc & (MAY_READ|MAY_WRITE|MAY_EXEC));
 
 	/* Allow read access to binaries even when mode 111 */
 	if (err == -EACCES && S_ISREG(inode->i_mode) &&
 	    acc == (NFSD_MAY_READ | NFSD_MAY_OWNER_OVERRIDE))
-		err = permission(inode, MAY_EXEC, NULL);
+		err = inode_permission(inode, MAY_EXEC);
 
 	return err? nfserrno(err) : 0;
 }
diff --git a/fs/utimes.c b/fs/utimes.c
index dad679d3a158..dc28b7826259 100644
--- a/fs/utimes.c
+++ b/fs/utimes.c
@@ -96,7 +96,7 @@ static int utimes_common(struct path *path, struct timespec *times)
 			goto mnt_drop_write_and_out;
 
 		if (!is_owner_or_cap(inode)) {
-			error = permission(inode, MAY_WRITE, NULL);
+			error = inode_permission(inode, MAY_WRITE);
 			if (error)
 				goto mnt_drop_write_and_out;
 		}
diff --git a/fs/xattr.c b/fs/xattr.c
index 4706a8b1f495..b96222e05ba0 100644
--- a/fs/xattr.c
+++ b/fs/xattr.c
@@ -63,7 +63,7 @@ xattr_permission(struct inode *inode, const char *name, int mask)
 			return -EPERM;
 	}
 
-	return permission(inode, mask, NULL);
+	return inode_permission(inode, mask);
 }
 
 int
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 527b9e482f99..9d2de4cadabd 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -1770,7 +1770,7 @@ extern int do_remount_sb(struct super_block *sb, int flags,
 extern sector_t bmap(struct inode *, sector_t);
 #endif
 extern int notify_change(struct dentry *, struct iattr *);
-extern int permission(struct inode *, int, struct nameidata *);
+extern int inode_permission(struct inode *, int);
 extern int generic_permission(struct inode *, int,
 		int (*check_acl)(struct inode *, int));
 
diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index 474984f9e032..96fb36cd9874 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -638,7 +638,7 @@ static int oflag2acc[O_ACCMODE] = { MAY_READ, MAY_WRITE,
 		return ERR_PTR(-EINVAL);
 	}
 
-	if (permission(dentry->d_inode, oflag2acc[oflag & O_ACCMODE], NULL)) {
+	if (inode_permission(dentry->d_inode, oflag2acc[oflag & O_ACCMODE])) {
 		dput(dentry);
 		mntput(mqueue_mnt);
 		return ERR_PTR(-EACCES);

From 256984a83880ff7ac78055cb87baea48137f0b77 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 22 Jul 2008 08:09:30 -0400
Subject: [PATCH 1153/1171] [PATCH] preparation to __user_walk_fd cleanup

Almost all users __user_walk_fd() and friends care only about struct path.
Get rid of the few that do not.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/inotify_user.c |  2 +-
 fs/open.c         | 18 ++++++++++--------
 2 files changed, 11 insertions(+), 9 deletions(-)

diff --git a/fs/inotify_user.c b/fs/inotify_user.c
index fe79c25d95dc..9b99ebf28884 100644
--- a/fs/inotify_user.c
+++ b/fs/inotify_user.c
@@ -365,7 +365,7 @@ static int find_inode(const char __user *dirname, struct nameidata *nd,
 	if (error)
 		return error;
 	/* you can only watch an inode if you have read permissions on it */
-	error = vfs_permission(nd, MAY_READ);
+	error = inode_permission(nd->path.dentry->d_inode, MAY_READ);
 	if (error)
 		path_put(&nd->path);
 	return error;
diff --git a/fs/open.c b/fs/open.c
index d5e421ad0cf6..e94266700eda 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -251,7 +251,7 @@ static long do_sys_truncate(const char __user * path, loff_t length)
 	if (error)
 		goto dput_and_out;
 
-	error = vfs_permission(&nd, MAY_WRITE);
+	error = inode_permission(inode, MAY_WRITE);
 	if (error)
 		goto mnt_drop_write_and_out;
 
@@ -426,6 +426,7 @@ out:
 asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode)
 {
 	struct nameidata nd;
+	struct inode *inode;
 	int old_fsuid, old_fsgid;
 	kernel_cap_t uninitialized_var(old_cap);  /* !SECURE_NO_SETUID_FIXUP */
 	int res;
@@ -461,7 +462,9 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode)
 	if (res)
 		goto out;
 
-	if ((mode & MAY_EXEC) && S_ISREG(nd.path.dentry->d_inode->i_mode)) {
+	inode = nd.path.dentry->d_inode;
+
+	if ((mode & MAY_EXEC) && S_ISREG(inode->i_mode)) {
 		/*
 		 * MAY_EXEC on regular files is denied if the fs is mounted
 		 * with the "noexec" flag.
@@ -471,10 +474,9 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode)
 			goto out_path_release;
 	}
 
-	res = vfs_permission(&nd, mode | MAY_ACCESS);
+	res = inode_permission(inode, mode | MAY_ACCESS);
 	/* SuS v2 requires we report a read only fs too */
-	if(res || !(mode & S_IWOTH) ||
-	   special_file(nd.path.dentry->d_inode->i_mode))
+	if (res || !(mode & S_IWOTH) || special_file(inode->i_mode))
 		goto out_path_release;
 	/*
 	 * This is a rare case where using __mnt_is_readonly()
@@ -515,7 +517,7 @@ asmlinkage long sys_chdir(const char __user * filename)
 	if (error)
 		goto out;
 
-	error = vfs_permission(&nd, MAY_EXEC | MAY_ACCESS);
+	error = inode_permission(nd.path.dentry->d_inode, MAY_EXEC | MAY_ACCESS);
 	if (error)
 		goto dput_and_out;
 
@@ -544,7 +546,7 @@ asmlinkage long sys_fchdir(unsigned int fd)
 	if (!S_ISDIR(inode->i_mode))
 		goto out_putf;
 
-	error = file_permission(file, MAY_EXEC | MAY_ACCESS);
+	error = inode_permission(inode, MAY_EXEC | MAY_ACCESS);
 	if (!error)
 		set_fs_pwd(current->fs, &file->f_path);
 out_putf:
@@ -562,7 +564,7 @@ asmlinkage long sys_chroot(const char __user * filename)
 	if (error)
 		goto out;
 
-	error = vfs_permission(&nd, MAY_EXEC | MAY_ACCESS);
+	error = inode_permission(nd.path.dentry->d_inode, MAY_EXEC | MAY_ACCESS);
 	if (error)
 		goto dput_and_out;
 

From 2d8f30380ab8c706f4e0a8f1aaa22b5886e9ac8a Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Tue, 22 Jul 2008 09:59:21 -0400
Subject: [PATCH 1154/1171] [PATCH] sanitize __user_walk_fd() et.al.

* do not pass nameidata; struct path is all the callers want.
* switch to new helpers:
	user_path_at(dfd, pathname, flags, &path)
	user_path(pathname, &path)
	user_lpath(pathname, &path)
	user_path_dir(pathname, &path)  (fail if not a directory)
  The last 3 are trivial macro wrappers for the first one.
* remove nameidata in callers.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 arch/alpha/kernel/osf_sys.c  |  10 +--
 arch/parisc/hpux/sys_hpux.c  |  10 +--
 fs/coda/pioctl.c             |  14 ++--
 fs/compat.c                  |  20 +++---
 fs/inotify_user.c            |  22 +++----
 fs/namei.c                   |  36 +++++-----
 fs/namespace.c               |  74 ++++++++++-----------
 fs/open.c                    | 124 +++++++++++++++++------------------
 fs/stat.c                    |  32 ++++-----
 fs/utimes.c                  |   8 +--
 fs/xattr.c                   |  96 +++++++++++++--------------
 fs/xfs/linux-2.6/xfs_ioctl.c |  14 ++--
 include/linux/namei.h        |  13 ++--
 13 files changed, 235 insertions(+), 238 deletions(-)

diff --git a/arch/alpha/kernel/osf_sys.c b/arch/alpha/kernel/osf_sys.c
index 32ca1b927307..6e943135f0e0 100644
--- a/arch/alpha/kernel/osf_sys.c
+++ b/arch/alpha/kernel/osf_sys.c
@@ -253,15 +253,15 @@ do_osf_statfs(struct dentry * dentry, struct osf_statfs __user *buffer,
 }
 
 asmlinkage int
-osf_statfs(char __user *path, struct osf_statfs __user *buffer, unsigned long bufsiz)
+osf_statfs(char __user *pathname, struct osf_statfs __user *buffer, unsigned long bufsiz)
 {
-	struct nameidata nd;
+	struct path path;
 	int retval;
 
-	retval = user_path_walk(path, &nd);
+	retval = user_path(pathname, &path);
 	if (!retval) {
-		retval = do_osf_statfs(nd.path.dentry, buffer, bufsiz);
-		path_put(&nd.path);
+		retval = do_osf_statfs(path.dentry, buffer, bufsiz);
+		path_put(&path);
 	}
 	return retval;
 }
diff --git a/arch/parisc/hpux/sys_hpux.c b/arch/parisc/hpux/sys_hpux.c
index be255ebb609c..18072e03a019 100644
--- a/arch/parisc/hpux/sys_hpux.c
+++ b/arch/parisc/hpux/sys_hpux.c
@@ -210,19 +210,19 @@ static int vfs_statfs_hpux(struct dentry *dentry, struct hpux_statfs *buf)
 }
 
 /* hpux statfs */
-asmlinkage long hpux_statfs(const char __user *path,
+asmlinkage long hpux_statfs(const char __user *pathname,
 						struct hpux_statfs __user *buf)
 {
-	struct nameidata nd;
+	struct path path;
 	int error;
 
-	error = user_path_walk(path, &nd);
+	error = user_path(pathname, &path);
 	if (!error) {
 		struct hpux_statfs tmp;
-		error = vfs_statfs_hpux(nd.path.dentry, &tmp);
+		error = vfs_statfs_hpux(path.dentry, &tmp);
 		if (!error && copy_to_user(buf, &tmp, sizeof(tmp)))
 			error = -EFAULT;
-		path_put(&nd.path);
+		path_put(&path);
 	}
 	return error;
 }
diff --git a/fs/coda/pioctl.c b/fs/coda/pioctl.c
index c38a98974fb0..c51365422aa8 100644
--- a/fs/coda/pioctl.c
+++ b/fs/coda/pioctl.c
@@ -49,7 +49,7 @@ static int coda_ioctl_permission(struct inode *inode, int mask)
 static int coda_pioctl(struct inode * inode, struct file * filp, 
                        unsigned int cmd, unsigned long user_data)
 {
-	struct nameidata nd;
+	struct path path;
         int error;
 	struct PioctlData data;
         struct inode *target_inode = NULL;
@@ -64,21 +64,21 @@ static int coda_pioctl(struct inode * inode, struct file * filp,
          * Look up the pathname. Note that the pathname is in 
          * user memory, and namei takes care of this
          */
-        if ( data.follow ) {
-                error = user_path_walk(data.path, &nd);
+        if (data.follow) {
+                error = user_path(data.path, &path);
 	} else {
-	        error = user_path_walk_link(data.path, &nd);
+	        error = user_lpath(data.path, &path);
 	}
 		
 	if ( error ) {
 		return error;
         } else {
-		target_inode = nd.path.dentry->d_inode;
+		target_inode = path.dentry->d_inode;
 	}
 	
 	/* return if it is not a Coda inode */
 	if ( target_inode->i_sb != inode->i_sb ) {
-		path_put(&nd.path);
+		path_put(&path);
 	        return  -EINVAL;
 	}
 
@@ -87,7 +87,7 @@ static int coda_pioctl(struct inode * inode, struct file * filp,
 
 	error = venus_pioctl(inode->i_sb, &(cnp->c_fid), cmd, &data);
 
-	path_put(&nd.path);
+	path_put(&path);
         return error;
 }
 
diff --git a/fs/compat.c b/fs/compat.c
index 106eba28ec5a..c9d1472e65c5 100644
--- a/fs/compat.c
+++ b/fs/compat.c
@@ -234,18 +234,18 @@ static int put_compat_statfs(struct compat_statfs __user *ubuf, struct kstatfs *
  * The following statfs calls are copies of code from fs/open.c and
  * should be checked against those from time to time
  */
-asmlinkage long compat_sys_statfs(const char __user *path, struct compat_statfs __user *buf)
+asmlinkage long compat_sys_statfs(const char __user *pathname, struct compat_statfs __user *buf)
 {
-	struct nameidata nd;
+	struct path path;
 	int error;
 
-	error = user_path_walk(path, &nd);
+	error = user_path(pathname, &path);
 	if (!error) {
 		struct kstatfs tmp;
-		error = vfs_statfs(nd.path.dentry, &tmp);
+		error = vfs_statfs(path.dentry, &tmp);
 		if (!error)
 			error = put_compat_statfs(buf, &tmp);
-		path_put(&nd.path);
+		path_put(&path);
 	}
 	return error;
 }
@@ -299,21 +299,21 @@ static int put_compat_statfs64(struct compat_statfs64 __user *ubuf, struct kstat
 	return 0;
 }
 
-asmlinkage long compat_sys_statfs64(const char __user *path, compat_size_t sz, struct compat_statfs64 __user *buf)
+asmlinkage long compat_sys_statfs64(const char __user *pathname, compat_size_t sz, struct compat_statfs64 __user *buf)
 {
-	struct nameidata nd;
+	struct path path;
 	int error;
 
 	if (sz != sizeof(*buf))
 		return -EINVAL;
 
-	error = user_path_walk(path, &nd);
+	error = user_path(pathname, &path);
 	if (!error) {
 		struct kstatfs tmp;
-		error = vfs_statfs(nd.path.dentry, &tmp);
+		error = vfs_statfs(path.dentry, &tmp);
 		if (!error)
 			error = put_compat_statfs64(buf, &tmp);
-		path_put(&nd.path);
+		path_put(&path);
 	}
 	return error;
 }
diff --git a/fs/inotify_user.c b/fs/inotify_user.c
index 9b99ebf28884..60249429a253 100644
--- a/fs/inotify_user.c
+++ b/fs/inotify_user.c
@@ -354,20 +354,20 @@ static void inotify_dev_event_dequeue(struct inotify_device *dev)
 }
 
 /*
- * find_inode - resolve a user-given path to a specific inode and return a nd
+ * find_inode - resolve a user-given path to a specific inode
  */
-static int find_inode(const char __user *dirname, struct nameidata *nd,
+static int find_inode(const char __user *dirname, struct path *path,
 		      unsigned flags)
 {
 	int error;
 
-	error = __user_walk(dirname, flags, nd);
+	error = user_path_at(AT_FDCWD, dirname, flags, path);
 	if (error)
 		return error;
 	/* you can only watch an inode if you have read permissions on it */
-	error = inode_permission(nd->path.dentry->d_inode, MAY_READ);
+	error = inode_permission(path->dentry->d_inode, MAY_READ);
 	if (error)
-		path_put(&nd->path);
+		path_put(path);
 	return error;
 }
 
@@ -650,11 +650,11 @@ asmlinkage long sys_inotify_init(void)
 	return sys_inotify_init1(0);
 }
 
-asmlinkage long sys_inotify_add_watch(int fd, const char __user *path, u32 mask)
+asmlinkage long sys_inotify_add_watch(int fd, const char __user *pathname, u32 mask)
 {
 	struct inode *inode;
 	struct inotify_device *dev;
-	struct nameidata nd;
+	struct path path;
 	struct file *filp;
 	int ret, fput_needed;
 	unsigned flags = 0;
@@ -674,12 +674,12 @@ asmlinkage long sys_inotify_add_watch(int fd, const char __user *path, u32 mask)
 	if (mask & IN_ONLYDIR)
 		flags |= LOOKUP_DIRECTORY;
 
-	ret = find_inode(path, &nd, flags);
+	ret = find_inode(pathname, &path, flags);
 	if (unlikely(ret))
 		goto fput_and_out;
 
-	/* inode held in place by reference to nd; dev by fget on fd */
-	inode = nd.path.dentry->d_inode;
+	/* inode held in place by reference to path; dev by fget on fd */
+	inode = path.dentry->d_inode;
 	dev = filp->private_data;
 
 	mutex_lock(&dev->up_mutex);
@@ -688,7 +688,7 @@ asmlinkage long sys_inotify_add_watch(int fd, const char __user *path, u32 mask)
 		ret = create_watch(dev, inode, mask);
 	mutex_unlock(&dev->up_mutex);
 
-	path_put(&nd.path);
+	path_put(&path);
 fput_and_out:
 	fput_light(filp, fput_needed);
 	return ret;
diff --git a/fs/namei.c b/fs/namei.c
index 5029b93ebbd5..edb5e973f9b3 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -1334,24 +1334,24 @@ struct dentry *lookup_one_noperm(const char *name, struct dentry *base)
 	return __lookup_hash(&this, base, NULL);
 }
 
-int __user_walk_fd(int dfd, const char __user *name, unsigned flags,
-			    struct nameidata *nd)
+int user_path_at(int dfd, const char __user *name, unsigned flags,
+		 struct path *path)
 {
+	struct nameidata nd;
 	char *tmp = getname(name);
 	int err = PTR_ERR(tmp);
-
 	if (!IS_ERR(tmp)) {
-		err = do_path_lookup(dfd, tmp, flags, nd);
+
+		BUG_ON(flags & LOOKUP_PARENT);
+
+		err = do_path_lookup(dfd, tmp, flags, &nd);
 		putname(tmp);
+		if (!err)
+			*path = nd.path;
 	}
 	return err;
 }
 
-int __user_walk(const char __user *name, unsigned flags, struct nameidata *nd)
-{
-	return __user_walk_fd(AT_FDCWD, name, flags, nd);
-}
-
 /*
  * It's inline, so penalty for filesystems that don't use sticky bit is
  * minimal.
@@ -2446,7 +2446,8 @@ asmlinkage long sys_linkat(int olddfd, const char __user *oldname,
 			   int flags)
 {
 	struct dentry *new_dentry;
-	struct nameidata nd, old_nd;
+	struct nameidata nd;
+	struct path old_path;
 	int error;
 	char * to;
 
@@ -2457,16 +2458,16 @@ asmlinkage long sys_linkat(int olddfd, const char __user *oldname,
 	if (IS_ERR(to))
 		return PTR_ERR(to);
 
-	error = __user_walk_fd(olddfd, oldname,
-			       flags & AT_SYMLINK_FOLLOW ? LOOKUP_FOLLOW : 0,
-			       &old_nd);
+	error = user_path_at(olddfd, oldname,
+			     flags & AT_SYMLINK_FOLLOW ? LOOKUP_FOLLOW : 0,
+			     &old_path);
 	if (error)
 		goto exit;
 	error = do_path_lookup(newdfd, to, LOOKUP_PARENT, &nd);
 	if (error)
 		goto out;
 	error = -EXDEV;
-	if (old_nd.path.mnt != nd.path.mnt)
+	if (old_path.mnt != nd.path.mnt)
 		goto out_release;
 	new_dentry = lookup_create(&nd, 0);
 	error = PTR_ERR(new_dentry);
@@ -2475,7 +2476,7 @@ asmlinkage long sys_linkat(int olddfd, const char __user *oldname,
 	error = mnt_want_write(nd.path.mnt);
 	if (error)
 		goto out_dput;
-	error = vfs_link(old_nd.path.dentry, nd.path.dentry->d_inode, new_dentry);
+	error = vfs_link(old_path.dentry, nd.path.dentry->d_inode, new_dentry);
 	mnt_drop_write(nd.path.mnt);
 out_dput:
 	dput(new_dentry);
@@ -2484,7 +2485,7 @@ out_unlock:
 out_release:
 	path_put(&nd.path);
 out:
-	path_put(&old_nd.path);
+	path_put(&old_path);
 exit:
 	putname(to);
 
@@ -2877,8 +2878,7 @@ const struct inode_operations page_symlink_inode_operations = {
 	.put_link	= page_put_link,
 };
 
-EXPORT_SYMBOL(__user_walk);
-EXPORT_SYMBOL(__user_walk_fd);
+EXPORT_SYMBOL(user_path_at);
 EXPORT_SYMBOL(follow_down);
 EXPORT_SYMBOL(follow_up);
 EXPORT_SYMBOL(get_write_access); /* binfmt_aout */
diff --git a/fs/namespace.c b/fs/namespace.c
index 26380f599534..411728c0c8bb 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -1130,27 +1130,27 @@ static int do_umount(struct vfsmount *mnt, int flags)
 
 asmlinkage long sys_umount(char __user * name, int flags)
 {
-	struct nameidata nd;
+	struct path path;
 	int retval;
 
-	retval = __user_walk(name, LOOKUP_FOLLOW, &nd);
+	retval = user_path(name, &path);
 	if (retval)
 		goto out;
 	retval = -EINVAL;
-	if (nd.path.dentry != nd.path.mnt->mnt_root)
+	if (path.dentry != path.mnt->mnt_root)
 		goto dput_and_out;
-	if (!check_mnt(nd.path.mnt))
+	if (!check_mnt(path.mnt))
 		goto dput_and_out;
 
 	retval = -EPERM;
 	if (!capable(CAP_SYS_ADMIN))
 		goto dput_and_out;
 
-	retval = do_umount(nd.path.mnt, flags);
+	retval = do_umount(path.mnt, flags);
 dput_and_out:
 	/* we mustn't call path_put() as that would clear mnt_expiry_mark */
-	dput(nd.path.dentry);
-	mntput_no_expire(nd.path.mnt);
+	dput(path.dentry);
+	mntput_no_expire(path.mnt);
 out:
 	return retval;
 }
@@ -2179,28 +2179,26 @@ asmlinkage long sys_pivot_root(const char __user * new_root,
 			       const char __user * put_old)
 {
 	struct vfsmount *tmp;
-	struct nameidata new_nd, old_nd;
-	struct path parent_path, root_parent, root;
+	struct path new, old, parent_path, root_parent, root;
 	int error;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
 
-	error = __user_walk(new_root, LOOKUP_FOLLOW | LOOKUP_DIRECTORY,
-			    &new_nd);
+	error = user_path_dir(new_root, &new);
 	if (error)
 		goto out0;
 	error = -EINVAL;
-	if (!check_mnt(new_nd.path.mnt))
+	if (!check_mnt(new.mnt))
 		goto out1;
 
-	error = __user_walk(put_old, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, &old_nd);
+	error = user_path_dir(put_old, &old);
 	if (error)
 		goto out1;
 
-	error = security_sb_pivotroot(&old_nd.path, &new_nd.path);
+	error = security_sb_pivotroot(&old, &new);
 	if (error) {
-		path_put(&old_nd.path);
+		path_put(&old);
 		goto out1;
 	}
 
@@ -2209,69 +2207,69 @@ asmlinkage long sys_pivot_root(const char __user * new_root,
 	path_get(&current->fs->root);
 	read_unlock(&current->fs->lock);
 	down_write(&namespace_sem);
-	mutex_lock(&old_nd.path.dentry->d_inode->i_mutex);
+	mutex_lock(&old.dentry->d_inode->i_mutex);
 	error = -EINVAL;
-	if (IS_MNT_SHARED(old_nd.path.mnt) ||
-		IS_MNT_SHARED(new_nd.path.mnt->mnt_parent) ||
+	if (IS_MNT_SHARED(old.mnt) ||
+		IS_MNT_SHARED(new.mnt->mnt_parent) ||
 		IS_MNT_SHARED(root.mnt->mnt_parent))
 		goto out2;
 	if (!check_mnt(root.mnt))
 		goto out2;
 	error = -ENOENT;
-	if (IS_DEADDIR(new_nd.path.dentry->d_inode))
+	if (IS_DEADDIR(new.dentry->d_inode))
 		goto out2;
-	if (d_unhashed(new_nd.path.dentry) && !IS_ROOT(new_nd.path.dentry))
+	if (d_unhashed(new.dentry) && !IS_ROOT(new.dentry))
 		goto out2;
-	if (d_unhashed(old_nd.path.dentry) && !IS_ROOT(old_nd.path.dentry))
+	if (d_unhashed(old.dentry) && !IS_ROOT(old.dentry))
 		goto out2;
 	error = -EBUSY;
-	if (new_nd.path.mnt == root.mnt ||
-	    old_nd.path.mnt == root.mnt)
+	if (new.mnt == root.mnt ||
+	    old.mnt == root.mnt)
 		goto out2; /* loop, on the same file system  */
 	error = -EINVAL;
 	if (root.mnt->mnt_root != root.dentry)
 		goto out2; /* not a mountpoint */
 	if (root.mnt->mnt_parent == root.mnt)
 		goto out2; /* not attached */
-	if (new_nd.path.mnt->mnt_root != new_nd.path.dentry)
+	if (new.mnt->mnt_root != new.dentry)
 		goto out2; /* not a mountpoint */
-	if (new_nd.path.mnt->mnt_parent == new_nd.path.mnt)
+	if (new.mnt->mnt_parent == new.mnt)
 		goto out2; /* not attached */
 	/* make sure we can reach put_old from new_root */
-	tmp = old_nd.path.mnt;
+	tmp = old.mnt;
 	spin_lock(&vfsmount_lock);
-	if (tmp != new_nd.path.mnt) {
+	if (tmp != new.mnt) {
 		for (;;) {
 			if (tmp->mnt_parent == tmp)
 				goto out3; /* already mounted on put_old */
-			if (tmp->mnt_parent == new_nd.path.mnt)
+			if (tmp->mnt_parent == new.mnt)
 				break;
 			tmp = tmp->mnt_parent;
 		}
-		if (!is_subdir(tmp->mnt_mountpoint, new_nd.path.dentry))
+		if (!is_subdir(tmp->mnt_mountpoint, new.dentry))
 			goto out3;
-	} else if (!is_subdir(old_nd.path.dentry, new_nd.path.dentry))
+	} else if (!is_subdir(old.dentry, new.dentry))
 		goto out3;
-	detach_mnt(new_nd.path.mnt, &parent_path);
+	detach_mnt(new.mnt, &parent_path);
 	detach_mnt(root.mnt, &root_parent);
 	/* mount old root on put_old */
-	attach_mnt(root.mnt, &old_nd.path);
+	attach_mnt(root.mnt, &old);
 	/* mount new_root on / */
-	attach_mnt(new_nd.path.mnt, &root_parent);
+	attach_mnt(new.mnt, &root_parent);
 	touch_mnt_namespace(current->nsproxy->mnt_ns);
 	spin_unlock(&vfsmount_lock);
-	chroot_fs_refs(&root, &new_nd.path);
-	security_sb_post_pivotroot(&root, &new_nd.path);
+	chroot_fs_refs(&root, &new);
+	security_sb_post_pivotroot(&root, &new);
 	error = 0;
 	path_put(&root_parent);
 	path_put(&parent_path);
 out2:
-	mutex_unlock(&old_nd.path.dentry->d_inode->i_mutex);
+	mutex_unlock(&old.dentry->d_inode->i_mutex);
 	up_write(&namespace_sem);
 	path_put(&root);
-	path_put(&old_nd.path);
+	path_put(&old);
 out1:
-	path_put(&new_nd.path);
+	path_put(&new);
 out0:
 	return error;
 out3:
diff --git a/fs/open.c b/fs/open.c
index e94266700eda..3fe1a6857c75 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -122,37 +122,37 @@ static int vfs_statfs64(struct dentry *dentry, struct statfs64 *buf)
 	return 0;
 }
 
-asmlinkage long sys_statfs(const char __user * path, struct statfs __user * buf)
+asmlinkage long sys_statfs(const char __user *pathname, struct statfs __user * buf)
 {
-	struct nameidata nd;
+	struct path path;
 	int error;
 
-	error = user_path_walk(path, &nd);
+	error = user_path(pathname, &path);
 	if (!error) {
 		struct statfs tmp;
-		error = vfs_statfs_native(nd.path.dentry, &tmp);
+		error = vfs_statfs_native(path.dentry, &tmp);
 		if (!error && copy_to_user(buf, &tmp, sizeof(tmp)))
 			error = -EFAULT;
-		path_put(&nd.path);
+		path_put(&path);
 	}
 	return error;
 }
 
 
-asmlinkage long sys_statfs64(const char __user *path, size_t sz, struct statfs64 __user *buf)
+asmlinkage long sys_statfs64(const char __user *pathname, size_t sz, struct statfs64 __user *buf)
 {
-	struct nameidata nd;
+	struct path path;
 	long error;
 
 	if (sz != sizeof(*buf))
 		return -EINVAL;
-	error = user_path_walk(path, &nd);
+	error = user_path(pathname, &path);
 	if (!error) {
 		struct statfs64 tmp;
-		error = vfs_statfs64(nd.path.dentry, &tmp);
+		error = vfs_statfs64(path.dentry, &tmp);
 		if (!error && copy_to_user(buf, &tmp, sizeof(tmp)))
 			error = -EFAULT;
-		path_put(&nd.path);
+		path_put(&path);
 	}
 	return error;
 }
@@ -223,20 +223,20 @@ int do_truncate(struct dentry *dentry, loff_t length, unsigned int time_attrs,
 	return err;
 }
 
-static long do_sys_truncate(const char __user * path, loff_t length)
+static long do_sys_truncate(const char __user *pathname, loff_t length)
 {
-	struct nameidata nd;
-	struct inode * inode;
+	struct path path;
+	struct inode *inode;
 	int error;
 
 	error = -EINVAL;
 	if (length < 0)	/* sorry, but loff_t says... */
 		goto out;
 
-	error = user_path_walk(path, &nd);
+	error = user_path(pathname, &path);
 	if (error)
 		goto out;
-	inode = nd.path.dentry->d_inode;
+	inode = path.dentry->d_inode;
 
 	/* For directories it's -EISDIR, for other non-regulars - -EINVAL */
 	error = -EISDIR;
@@ -247,7 +247,7 @@ static long do_sys_truncate(const char __user * path, loff_t length)
 	if (!S_ISREG(inode->i_mode))
 		goto dput_and_out;
 
-	error = mnt_want_write(nd.path.mnt);
+	error = mnt_want_write(path.mnt);
 	if (error)
 		goto dput_and_out;
 
@@ -274,15 +274,15 @@ static long do_sys_truncate(const char __user * path, loff_t length)
 	error = locks_verify_truncate(inode, NULL, length);
 	if (!error) {
 		DQUOT_INIT(inode);
-		error = do_truncate(nd.path.dentry, length, 0, NULL);
+		error = do_truncate(path.dentry, length, 0, NULL);
 	}
 
 put_write_and_out:
 	put_write_access(inode);
 mnt_drop_write_and_out:
-	mnt_drop_write(nd.path.mnt);
+	mnt_drop_write(path.mnt);
 dput_and_out:
-	path_put(&nd.path);
+	path_put(&path);
 out:
 	return error;
 }
@@ -425,7 +425,7 @@ out:
  */
 asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode)
 {
-	struct nameidata nd;
+	struct path path;
 	struct inode *inode;
 	int old_fsuid, old_fsgid;
 	kernel_cap_t uninitialized_var(old_cap);  /* !SECURE_NO_SETUID_FIXUP */
@@ -449,7 +449,7 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode)
 		 * FIXME: There is a race here against sys_capset.  The
 		 * capabilities can change yet we will restore the old
 		 * value below.  We should hold task_capabilities_lock,
-		 * but we cannot because user_path_walk can sleep.
+		 * but we cannot because user_path_at can sleep.
 		 */
 #endif /* ndef CONFIG_SECURITY_FILE_CAPABILITIES */
 		if (current->uid)
@@ -458,11 +458,11 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode)
 			old_cap = cap_set_effective(current->cap_permitted);
 	}
 
-	res = __user_walk_fd(dfd, filename, LOOKUP_FOLLOW, &nd);
+	res = user_path_at(dfd, filename, LOOKUP_FOLLOW, &path);
 	if (res)
 		goto out;
 
-	inode = nd.path.dentry->d_inode;
+	inode = path.dentry->d_inode;
 
 	if ((mode & MAY_EXEC) && S_ISREG(inode->i_mode)) {
 		/*
@@ -470,7 +470,7 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode)
 		 * with the "noexec" flag.
 		 */
 		res = -EACCES;
-		if (nd.path.mnt->mnt_flags & MNT_NOEXEC)
+		if (path.mnt->mnt_flags & MNT_NOEXEC)
 			goto out_path_release;
 	}
 
@@ -488,11 +488,11 @@ asmlinkage long sys_faccessat(int dfd, const char __user *filename, int mode)
 	 * inherently racy and know that the fs may change
 	 * state before we even see this result.
 	 */
-	if (__mnt_is_readonly(nd.path.mnt))
+	if (__mnt_is_readonly(path.mnt))
 		res = -EROFS;
 
 out_path_release:
-	path_put(&nd.path);
+	path_put(&path);
 out:
 	current->fsuid = old_fsuid;
 	current->fsgid = old_fsgid;
@@ -510,21 +510,21 @@ asmlinkage long sys_access(const char __user *filename, int mode)
 
 asmlinkage long sys_chdir(const char __user * filename)
 {
-	struct nameidata nd;
+	struct path path;
 	int error;
 
-	error = __user_walk(filename, LOOKUP_FOLLOW|LOOKUP_DIRECTORY, &nd);
+	error = user_path_dir(filename, &path);
 	if (error)
 		goto out;
 
-	error = inode_permission(nd.path.dentry->d_inode, MAY_EXEC | MAY_ACCESS);
+	error = inode_permission(path.dentry->d_inode, MAY_EXEC | MAY_ACCESS);
 	if (error)
 		goto dput_and_out;
 
-	set_fs_pwd(current->fs, &nd.path);
+	set_fs_pwd(current->fs, &path);
 
 dput_and_out:
-	path_put(&nd.path);
+	path_put(&path);
 out:
 	return error;
 }
@@ -557,14 +557,14 @@ out:
 
 asmlinkage long sys_chroot(const char __user * filename)
 {
-	struct nameidata nd;
+	struct path path;
 	int error;
 
-	error = __user_walk(filename, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, &nd);
+	error = user_path_dir(filename, &path);
 	if (error)
 		goto out;
 
-	error = inode_permission(nd.path.dentry->d_inode, MAY_EXEC | MAY_ACCESS);
+	error = inode_permission(path.dentry->d_inode, MAY_EXEC | MAY_ACCESS);
 	if (error)
 		goto dput_and_out;
 
@@ -572,10 +572,10 @@ asmlinkage long sys_chroot(const char __user * filename)
 	if (!capable(CAP_SYS_CHROOT))
 		goto dput_and_out;
 
-	set_fs_root(current->fs, &nd.path);
+	set_fs_root(current->fs, &path);
 	error = 0;
 dput_and_out:
-	path_put(&nd.path);
+	path_put(&path);
 out:
 	return error;
 }
@@ -617,17 +617,17 @@ out:
 asmlinkage long sys_fchmodat(int dfd, const char __user *filename,
 			     mode_t mode)
 {
-	struct nameidata nd;
-	struct inode * inode;
+	struct path path;
+	struct inode *inode;
 	int error;
 	struct iattr newattrs;
 
-	error = __user_walk_fd(dfd, filename, LOOKUP_FOLLOW, &nd);
+	error = user_path_at(dfd, filename, LOOKUP_FOLLOW, &path);
 	if (error)
 		goto out;
-	inode = nd.path.dentry->d_inode;
+	inode = path.dentry->d_inode;
 
-	error = mnt_want_write(nd.path.mnt);
+	error = mnt_want_write(path.mnt);
 	if (error)
 		goto dput_and_out;
 	mutex_lock(&inode->i_mutex);
@@ -635,11 +635,11 @@ asmlinkage long sys_fchmodat(int dfd, const char __user *filename,
 		mode = inode->i_mode;
 	newattrs.ia_mode = (mode & S_IALLUGO) | (inode->i_mode & ~S_IALLUGO);
 	newattrs.ia_valid = ATTR_MODE | ATTR_CTIME;
-	error = notify_change(nd.path.dentry, &newattrs);
+	error = notify_change(path.dentry, &newattrs);
 	mutex_unlock(&inode->i_mutex);
-	mnt_drop_write(nd.path.mnt);
+	mnt_drop_write(path.mnt);
 dput_and_out:
-	path_put(&nd.path);
+	path_put(&path);
 out:
 	return error;
 }
@@ -676,19 +676,19 @@ static int chown_common(struct dentry * dentry, uid_t user, gid_t group)
 
 asmlinkage long sys_chown(const char __user * filename, uid_t user, gid_t group)
 {
-	struct nameidata nd;
+	struct path path;
 	int error;
 
-	error = user_path_walk(filename, &nd);
+	error = user_path(filename, &path);
 	if (error)
 		goto out;
-	error = mnt_want_write(nd.path.mnt);
+	error = mnt_want_write(path.mnt);
 	if (error)
 		goto out_release;
-	error = chown_common(nd.path.dentry, user, group);
-	mnt_drop_write(nd.path.mnt);
+	error = chown_common(path.dentry, user, group);
+	mnt_drop_write(path.mnt);
 out_release:
-	path_put(&nd.path);
+	path_put(&path);
 out:
 	return error;
 }
@@ -696,7 +696,7 @@ out:
 asmlinkage long sys_fchownat(int dfd, const char __user *filename, uid_t user,
 			     gid_t group, int flag)
 {
-	struct nameidata nd;
+	struct path path;
 	int error = -EINVAL;
 	int follow;
 
@@ -704,35 +704,35 @@ asmlinkage long sys_fchownat(int dfd, const char __user *filename, uid_t user,
 		goto out;
 
 	follow = (flag & AT_SYMLINK_NOFOLLOW) ? 0 : LOOKUP_FOLLOW;
-	error = __user_walk_fd(dfd, filename, follow, &nd);
+	error = user_path_at(dfd, filename, follow, &path);
 	if (error)
 		goto out;
-	error = mnt_want_write(nd.path.mnt);
+	error = mnt_want_write(path.mnt);
 	if (error)
 		goto out_release;
-	error = chown_common(nd.path.dentry, user, group);
-	mnt_drop_write(nd.path.mnt);
+	error = chown_common(path.dentry, user, group);
+	mnt_drop_write(path.mnt);
 out_release:
-	path_put(&nd.path);
+	path_put(&path);
 out:
 	return error;
 }
 
 asmlinkage long sys_lchown(const char __user * filename, uid_t user, gid_t group)
 {
-	struct nameidata nd;
+	struct path path;
 	int error;
 
-	error = user_path_walk_link(filename, &nd);
+	error = user_lpath(filename, &path);
 	if (error)
 		goto out;
-	error = mnt_want_write(nd.path.mnt);
+	error = mnt_want_write(path.mnt);
 	if (error)
 		goto out_release;
-	error = chown_common(nd.path.dentry, user, group);
-	mnt_drop_write(nd.path.mnt);
+	error = chown_common(path.dentry, user, group);
+	mnt_drop_write(path.mnt);
 out_release:
-	path_put(&nd.path);
+	path_put(&path);
 out:
 	return error;
 }
diff --git a/fs/stat.c b/fs/stat.c
index 9cf41f719d50..7c46fbeb8b76 100644
--- a/fs/stat.c
+++ b/fs/stat.c
@@ -57,13 +57,13 @@ EXPORT_SYMBOL(vfs_getattr);
 
 int vfs_stat_fd(int dfd, char __user *name, struct kstat *stat)
 {
-	struct nameidata nd;
+	struct path path;
 	int error;
 
-	error = __user_walk_fd(dfd, name, LOOKUP_FOLLOW, &nd);
+	error = user_path_at(dfd, name, LOOKUP_FOLLOW, &path);
 	if (!error) {
-		error = vfs_getattr(nd.path.mnt, nd.path.dentry, stat);
-		path_put(&nd.path);
+		error = vfs_getattr(path.mnt, path.dentry, stat);
+		path_put(&path);
 	}
 	return error;
 }
@@ -77,13 +77,13 @@ EXPORT_SYMBOL(vfs_stat);
 
 int vfs_lstat_fd(int dfd, char __user *name, struct kstat *stat)
 {
-	struct nameidata nd;
+	struct path path;
 	int error;
 
-	error = __user_walk_fd(dfd, name, 0, &nd);
+	error = user_path_at(dfd, name, 0, &path);
 	if (!error) {
-		error = vfs_getattr(nd.path.mnt, nd.path.dentry, stat);
-		path_put(&nd.path);
+		error = vfs_getattr(path.mnt, path.dentry, stat);
+		path_put(&path);
 	}
 	return error;
 }
@@ -291,29 +291,29 @@ asmlinkage long sys_newfstat(unsigned int fd, struct stat __user *statbuf)
 	return error;
 }
 
-asmlinkage long sys_readlinkat(int dfd, const char __user *path,
+asmlinkage long sys_readlinkat(int dfd, const char __user *pathname,
 				char __user *buf, int bufsiz)
 {
-	struct nameidata nd;
+	struct path path;
 	int error;
 
 	if (bufsiz <= 0)
 		return -EINVAL;
 
-	error = __user_walk_fd(dfd, path, 0, &nd);
+	error = user_path_at(dfd, pathname, 0, &path);
 	if (!error) {
-		struct inode *inode = nd.path.dentry->d_inode;
+		struct inode *inode = path.dentry->d_inode;
 
 		error = -EINVAL;
 		if (inode->i_op && inode->i_op->readlink) {
-			error = security_inode_readlink(nd.path.dentry);
+			error = security_inode_readlink(path.dentry);
 			if (!error) {
-				touch_atime(nd.path.mnt, nd.path.dentry);
-				error = inode->i_op->readlink(nd.path.dentry,
+				touch_atime(path.mnt, path.dentry);
+				error = inode->i_op->readlink(path.dentry,
 							      buf, bufsiz);
 			}
 		}
-		path_put(&nd.path);
+		path_put(&path);
 	}
 	return error;
 }
diff --git a/fs/utimes.c b/fs/utimes.c
index dc28b7826259..6929e3e91d05 100644
--- a/fs/utimes.c
+++ b/fs/utimes.c
@@ -152,18 +152,18 @@ long do_utimes(int dfd, char __user *filename, struct timespec *times, int flags
 		error = utimes_common(&file->f_path, times);
 		fput(file);
 	} else {
-		struct nameidata nd;
+		struct path path;
 		int lookup_flags = 0;
 
 		if (!(flags & AT_SYMLINK_NOFOLLOW))
 			lookup_flags |= LOOKUP_FOLLOW;
 
-		error = __user_walk_fd(dfd, filename, lookup_flags, &nd);
+		error = user_path_at(dfd, filename, lookup_flags, &path);
 		if (error)
 			goto out;
 
-		error = utimes_common(&nd.path, times);
-		path_put(&nd.path);
+		error = utimes_common(&path, times);
+		path_put(&path);
 	}
 
 out:
diff --git a/fs/xattr.c b/fs/xattr.c
index b96222e05ba0..468377e66531 100644
--- a/fs/xattr.c
+++ b/fs/xattr.c
@@ -252,40 +252,40 @@ setxattr(struct dentry *d, const char __user *name, const void __user *value,
 }
 
 asmlinkage long
-sys_setxattr(const char __user *path, const char __user *name,
+sys_setxattr(const char __user *pathname, const char __user *name,
 	     const void __user *value, size_t size, int flags)
 {
-	struct nameidata nd;
+	struct path path;
 	int error;
 
-	error = user_path_walk(path, &nd);
+	error = user_path(pathname, &path);
 	if (error)
 		return error;
-	error = mnt_want_write(nd.path.mnt);
+	error = mnt_want_write(path.mnt);
 	if (!error) {
-		error = setxattr(nd.path.dentry, name, value, size, flags);
-		mnt_drop_write(nd.path.mnt);
+		error = setxattr(path.dentry, name, value, size, flags);
+		mnt_drop_write(path.mnt);
 	}
-	path_put(&nd.path);
+	path_put(&path);
 	return error;
 }
 
 asmlinkage long
-sys_lsetxattr(const char __user *path, const char __user *name,
+sys_lsetxattr(const char __user *pathname, const char __user *name,
 	      const void __user *value, size_t size, int flags)
 {
-	struct nameidata nd;
+	struct path path;
 	int error;
 
-	error = user_path_walk_link(path, &nd);
+	error = user_lpath(pathname, &path);
 	if (error)
 		return error;
-	error = mnt_want_write(nd.path.mnt);
+	error = mnt_want_write(path.mnt);
 	if (!error) {
-		error = setxattr(nd.path.dentry, name, value, size, flags);
-		mnt_drop_write(nd.path.mnt);
+		error = setxattr(path.dentry, name, value, size, flags);
+		mnt_drop_write(path.mnt);
 	}
-	path_put(&nd.path);
+	path_put(&path);
 	return error;
 }
 
@@ -350,32 +350,32 @@ getxattr(struct dentry *d, const char __user *name, void __user *value,
 }
 
 asmlinkage ssize_t
-sys_getxattr(const char __user *path, const char __user *name,
+sys_getxattr(const char __user *pathname, const char __user *name,
 	     void __user *value, size_t size)
 {
-	struct nameidata nd;
+	struct path path;
 	ssize_t error;
 
-	error = user_path_walk(path, &nd);
+	error = user_path(pathname, &path);
 	if (error)
 		return error;
-	error = getxattr(nd.path.dentry, name, value, size);
-	path_put(&nd.path);
+	error = getxattr(path.dentry, name, value, size);
+	path_put(&path);
 	return error;
 }
 
 asmlinkage ssize_t
-sys_lgetxattr(const char __user *path, const char __user *name, void __user *value,
+sys_lgetxattr(const char __user *pathname, const char __user *name, void __user *value,
 	      size_t size)
 {
-	struct nameidata nd;
+	struct path path;
 	ssize_t error;
 
-	error = user_path_walk_link(path, &nd);
+	error = user_lpath(pathname, &path);
 	if (error)
 		return error;
-	error = getxattr(nd.path.dentry, name, value, size);
-	path_put(&nd.path);
+	error = getxattr(path.dentry, name, value, size);
+	path_put(&path);
 	return error;
 }
 
@@ -425,30 +425,30 @@ listxattr(struct dentry *d, char __user *list, size_t size)
 }
 
 asmlinkage ssize_t
-sys_listxattr(const char __user *path, char __user *list, size_t size)
+sys_listxattr(const char __user *pathname, char __user *list, size_t size)
 {
-	struct nameidata nd;
+	struct path path;
 	ssize_t error;
 
-	error = user_path_walk(path, &nd);
+	error = user_path(pathname, &path);
 	if (error)
 		return error;
-	error = listxattr(nd.path.dentry, list, size);
-	path_put(&nd.path);
+	error = listxattr(path.dentry, list, size);
+	path_put(&path);
 	return error;
 }
 
 asmlinkage ssize_t
-sys_llistxattr(const char __user *path, char __user *list, size_t size)
+sys_llistxattr(const char __user *pathname, char __user *list, size_t size)
 {
-	struct nameidata nd;
+	struct path path;
 	ssize_t error;
 
-	error = user_path_walk_link(path, &nd);
+	error = user_lpath(pathname, &path);
 	if (error)
 		return error;
-	error = listxattr(nd.path.dentry, list, size);
-	path_put(&nd.path);
+	error = listxattr(path.dentry, list, size);
+	path_put(&path);
 	return error;
 }
 
@@ -486,38 +486,38 @@ removexattr(struct dentry *d, const char __user *name)
 }
 
 asmlinkage long
-sys_removexattr(const char __user *path, const char __user *name)
+sys_removexattr(const char __user *pathname, const char __user *name)
 {
-	struct nameidata nd;
+	struct path path;
 	int error;
 
-	error = user_path_walk(path, &nd);
+	error = user_path(pathname, &path);
 	if (error)
 		return error;
-	error = mnt_want_write(nd.path.mnt);
+	error = mnt_want_write(path.mnt);
 	if (!error) {
-		error = removexattr(nd.path.dentry, name);
-		mnt_drop_write(nd.path.mnt);
+		error = removexattr(path.dentry, name);
+		mnt_drop_write(path.mnt);
 	}
-	path_put(&nd.path);
+	path_put(&path);
 	return error;
 }
 
 asmlinkage long
-sys_lremovexattr(const char __user *path, const char __user *name)
+sys_lremovexattr(const char __user *pathname, const char __user *name)
 {
-	struct nameidata nd;
+	struct path path;
 	int error;
 
-	error = user_path_walk_link(path, &nd);
+	error = user_lpath(pathname, &path);
 	if (error)
 		return error;
-	error = mnt_want_write(nd.path.mnt);
+	error = mnt_want_write(path.mnt);
 	if (!error) {
-		error = removexattr(nd.path.dentry, name);
-		mnt_drop_write(nd.path.mnt);
+		error = removexattr(path.dentry, name);
+		mnt_drop_write(path.mnt);
 	}
-	path_put(&nd.path);
+	path_put(&path);
 	return error;
 }
 
diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c
index a42ba9d71156..01939ba2d8de 100644
--- a/fs/xfs/linux-2.6/xfs_ioctl.c
+++ b/fs/xfs/linux-2.6/xfs_ioctl.c
@@ -84,17 +84,15 @@ xfs_find_handle(
 	switch (cmd) {
 	case XFS_IOC_PATH_TO_FSHANDLE:
 	case XFS_IOC_PATH_TO_HANDLE: {
-		struct nameidata	nd;
-		int			error;
-
-		error = user_path_walk_link((const char __user *)hreq.path, &nd);
+		struct path path;
+		int error = user_lpath((const char __user *)hreq.path, &path);
 		if (error)
 			return error;
 
-		ASSERT(nd.path.dentry);
-		ASSERT(nd.path.dentry->d_inode);
-		inode = igrab(nd.path.dentry->d_inode);
-		path_put(&nd.path);
+		ASSERT(path.dentry);
+		ASSERT(path.dentry->d_inode);
+		inode = igrab(path.dentry->d_inode);
+		path_put(&path);
 		break;
 	}
 
diff --git a/include/linux/namei.h b/include/linux/namei.h
index 60e35a02f6cb..00888ff69504 100644
--- a/include/linux/namei.h
+++ b/include/linux/namei.h
@@ -54,12 +54,13 @@ enum {LAST_NORM, LAST_ROOT, LAST_DOT, LAST_DOTDOT, LAST_BIND};
 #define LOOKUP_OPEN		(0x0100)
 #define LOOKUP_CREATE		(0x0200)
 
-extern int __user_walk(const char __user *, unsigned, struct nameidata *);
-extern int __user_walk_fd(int dfd, const char __user *, unsigned, struct nameidata *);
-#define user_path_walk(name,nd) \
-	__user_walk_fd(AT_FDCWD, name, LOOKUP_FOLLOW, nd)
-#define user_path_walk_link(name,nd) \
-	__user_walk_fd(AT_FDCWD, name, 0, nd)
+extern int user_path_at(int, const char __user *, unsigned, struct path *);
+
+#define user_path(name, path) user_path_at(AT_FDCWD, name, LOOKUP_FOLLOW, path)
+#define user_lpath(name, path) user_path_at(AT_FDCWD, name, 0, path)
+#define user_path_dir(name, path) \
+	user_path_at(AT_FDCWD, name, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, path)
+
 extern int path_lookup(const char *, unsigned, struct nameidata *);
 extern int vfs_path_lookup(struct dentry *, struct vfsmount *,
 			   const char *, unsigned int, struct nameidata *);

From 2ad94ae654f5eb72fd3260b706aea645cf4a7791 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Mon, 21 Jul 2008 09:32:51 -0400
Subject: [PATCH 1155/1171] [PATCH] new (local) helper: user_path_parent()

Preparation to untangling intents mess: reduce the number of do_path_lookup()
callers.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/namei.c | 137 ++++++++++++++++++++++-------------------------------
 1 file changed, 57 insertions(+), 80 deletions(-)

diff --git a/fs/namei.c b/fs/namei.c
index edb5e973f9b3..38ceb6e06eba 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -1352,6 +1352,24 @@ int user_path_at(int dfd, const char __user *name, unsigned flags,
 	return err;
 }
 
+static int user_path_parent(int dfd, const char __user *path,
+			struct nameidata *nd, char **name)
+{
+	char *s = getname(path);
+	int error;
+
+	if (IS_ERR(s))
+		return PTR_ERR(s);
+
+	error = do_path_lookup(dfd, s, LOOKUP_PARENT, nd);
+	if (error)
+		putname(s);
+	else
+		*name = s;
+
+	return error;
+}
+
 /*
  * It's inline, so penalty for filesystems that don't use sticky bit is
  * minimal.
@@ -1989,20 +2007,18 @@ static int may_mknod(mode_t mode)
 asmlinkage long sys_mknodat(int dfd, const char __user *filename, int mode,
 				unsigned dev)
 {
-	int error = 0;
-	char * tmp;
-	struct dentry * dentry;
+	int error;
+	char *tmp;
+	struct dentry *dentry;
 	struct nameidata nd;
 
 	if (S_ISDIR(mode))
 		return -EPERM;
-	tmp = getname(filename);
-	if (IS_ERR(tmp))
-		return PTR_ERR(tmp);
 
-	error = do_path_lookup(dfd, tmp, LOOKUP_PARENT, &nd);
+	error = user_path_parent(dfd, filename, &nd, &tmp);
 	if (error)
-		goto out;
+		return error;
+
 	dentry = lookup_create(&nd, 0);
 	if (IS_ERR(dentry)) {
 		error = PTR_ERR(dentry);
@@ -2034,7 +2050,6 @@ out_dput:
 out_unlock:
 	mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
 	path_put(&nd.path);
-out:
 	putname(tmp);
 
 	return error;
@@ -2074,14 +2089,10 @@ asmlinkage long sys_mkdirat(int dfd, const char __user *pathname, int mode)
 	struct dentry *dentry;
 	struct nameidata nd;
 
-	tmp = getname(pathname);
-	error = PTR_ERR(tmp);
-	if (IS_ERR(tmp))
+	error = user_path_parent(dfd, pathname, &nd, &tmp);
+	if (error)
 		goto out_err;
 
-	error = do_path_lookup(dfd, tmp, LOOKUP_PARENT, &nd);
-	if (error)
-		goto out;
 	dentry = lookup_create(&nd, 1);
 	error = PTR_ERR(dentry);
 	if (IS_ERR(dentry))
@@ -2099,7 +2110,6 @@ out_dput:
 out_unlock:
 	mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
 	path_put(&nd.path);
-out:
 	putname(tmp);
 out_err:
 	return error;
@@ -2177,13 +2187,9 @@ static long do_rmdir(int dfd, const char __user *pathname)
 	struct dentry *dentry;
 	struct nameidata nd;
 
-	name = getname(pathname);
-	if(IS_ERR(name))
-		return PTR_ERR(name);
-
-	error = do_path_lookup(dfd, name, LOOKUP_PARENT, &nd);
+	error = user_path_parent(dfd, pathname, &nd, &name);
 	if (error)
-		goto exit;
+		return error;
 
 	switch(nd.last_type) {
 		case LAST_DOTDOT:
@@ -2212,7 +2218,6 @@ exit2:
 	mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
 exit1:
 	path_put(&nd.path);
-exit:
 	putname(name);
 	return error;
 }
@@ -2261,19 +2266,16 @@ int vfs_unlink(struct inode *dir, struct dentry *dentry)
  */
 static long do_unlinkat(int dfd, const char __user *pathname)
 {
-	int error = 0;
-	char * name;
+	int error;
+	char *name;
 	struct dentry *dentry;
 	struct nameidata nd;
 	struct inode *inode = NULL;
 
-	name = getname(pathname);
-	if(IS_ERR(name))
-		return PTR_ERR(name);
-
-	error = do_path_lookup(dfd, name, LOOKUP_PARENT, &nd);
+	error = user_path_parent(dfd, pathname, &nd, &name);
 	if (error)
-		goto exit;
+		return error;
+
 	error = -EISDIR;
 	if (nd.last_type != LAST_NORM)
 		goto exit1;
@@ -2300,7 +2302,6 @@ static long do_unlinkat(int dfd, const char __user *pathname)
 		iput(inode);	/* truncate the inode here */
 exit1:
 	path_put(&nd.path);
-exit:
 	putname(name);
 	return error;
 
@@ -2350,23 +2351,20 @@ int vfs_symlink(struct inode *dir, struct dentry *dentry, const char *oldname)
 asmlinkage long sys_symlinkat(const char __user *oldname,
 			      int newdfd, const char __user *newname)
 {
-	int error = 0;
-	char * from;
-	char * to;
+	int error;
+	char *from;
+	char *to;
 	struct dentry *dentry;
 	struct nameidata nd;
 
 	from = getname(oldname);
-	if(IS_ERR(from))
+	if (IS_ERR(from))
 		return PTR_ERR(from);
-	to = getname(newname);
-	error = PTR_ERR(to);
-	if (IS_ERR(to))
+
+	error = user_path_parent(newdfd, newname, &nd, &to);
+	if (error)
 		goto out_putname;
 
-	error = do_path_lookup(newdfd, to, LOOKUP_PARENT, &nd);
-	if (error)
-		goto out;
 	dentry = lookup_create(&nd, 0);
 	error = PTR_ERR(dentry);
 	if (IS_ERR(dentry))
@@ -2382,7 +2380,6 @@ out_dput:
 out_unlock:
 	mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
 	path_put(&nd.path);
-out:
 	putname(to);
 out_putname:
 	putname(from);
@@ -2449,21 +2446,18 @@ asmlinkage long sys_linkat(int olddfd, const char __user *oldname,
 	struct nameidata nd;
 	struct path old_path;
 	int error;
-	char * to;
+	char *to;
 
 	if ((flags & ~AT_SYMLINK_FOLLOW) != 0)
 		return -EINVAL;
 
-	to = getname(newname);
-	if (IS_ERR(to))
-		return PTR_ERR(to);
-
 	error = user_path_at(olddfd, oldname,
 			     flags & AT_SYMLINK_FOLLOW ? LOOKUP_FOLLOW : 0,
 			     &old_path);
 	if (error)
-		goto exit;
-	error = do_path_lookup(newdfd, to, LOOKUP_PARENT, &nd);
+		return error;
+
+	error = user_path_parent(newdfd, newname, &nd, &to);
 	if (error)
 		goto out;
 	error = -EXDEV;
@@ -2484,10 +2478,9 @@ out_unlock:
 	mutex_unlock(&nd.path.dentry->d_inode->i_mutex);
 out_release:
 	path_put(&nd.path);
+	putname(to);
 out:
 	path_put(&old_path);
-exit:
-	putname(to);
 
 	return error;
 }
@@ -2643,20 +2636,22 @@ int vfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	return error;
 }
 
-static int do_rename(int olddfd, const char *oldname,
-			int newdfd, const char *newname)
+asmlinkage long sys_renameat(int olddfd, const char __user *oldname,
+			     int newdfd, const char __user *newname)
 {
-	int error = 0;
-	struct dentry * old_dir, * new_dir;
-	struct dentry * old_dentry, *new_dentry;
-	struct dentry * trap;
+	struct dentry *old_dir, *new_dir;
+	struct dentry *old_dentry, *new_dentry;
+	struct dentry *trap;
 	struct nameidata oldnd, newnd;
+	char *from;
+	char *to;
+	int error;
 
-	error = do_path_lookup(olddfd, oldname, LOOKUP_PARENT, &oldnd);
+	error = user_path_parent(olddfd, oldname, &oldnd, &from);
 	if (error)
 		goto exit;
 
-	error = do_path_lookup(newdfd, newname, LOOKUP_PARENT, &newnd);
+	error = user_path_parent(newdfd, newname, &newnd, &to);
 	if (error)
 		goto exit1;
 
@@ -2718,29 +2713,11 @@ exit3:
 	unlock_rename(new_dir, old_dir);
 exit2:
 	path_put(&newnd.path);
+	putname(to);
 exit1:
 	path_put(&oldnd.path);
-exit:
-	return error;
-}
-
-asmlinkage long sys_renameat(int olddfd, const char __user *oldname,
-			     int newdfd, const char __user *newname)
-{
-	int error;
-	char * from;
-	char * to;
-
-	from = getname(oldname);
-	if(IS_ERR(from))
-		return PTR_ERR(from);
-	to = getname(newname);
-	error = PTR_ERR(to);
-	if (!IS_ERR(to)) {
-		error = do_rename(olddfd, from, newdfd, to);
-		putname(to);
-	}
 	putname(from);
+exit:
 	return error;
 }
 

From a569c711f63995ad80c23918525111e0cdb0bc73 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 23 Jul 2008 14:42:05 -0400
Subject: [PATCH 1156/1171] [PATCH] don't pass nameidata to gfs2_lookupi()

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/gfs2/inode.c      | 6 +++---
 fs/gfs2/inode.h      | 2 +-
 fs/gfs2/ops_export.c | 2 +-
 fs/gfs2/ops_inode.c  | 4 ++--
 fs/gfs2/super.c      | 2 +-
 5 files changed, 8 insertions(+), 8 deletions(-)

diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c
index 6da0ab355b8a..8b0806a32948 100644
--- a/fs/gfs2/inode.c
+++ b/fs/gfs2/inode.c
@@ -448,7 +448,7 @@ struct inode *gfs2_lookup_simple(struct inode *dip, const char *name)
 	struct qstr qstr;
 	struct inode *inode;
 	gfs2_str2qstr(&qstr, name);
-	inode = gfs2_lookupi(dip, &qstr, 1, NULL);
+	inode = gfs2_lookupi(dip, &qstr, 1);
 	/* gfs2_lookupi has inconsistent callers: vfs
 	 * related routines expect NULL for no entry found,
 	 * gfs2_lookup_simple callers expect ENOENT
@@ -477,7 +477,7 @@ struct inode *gfs2_lookup_simple(struct inode *dip, const char *name)
  */
 
 struct inode *gfs2_lookupi(struct inode *dir, const struct qstr *name,
-			   int is_root, struct nameidata *nd)
+			   int is_root)
 {
 	struct super_block *sb = dir->i_sb;
 	struct gfs2_inode *dip = GFS2_I(dir);
@@ -1173,7 +1173,7 @@ int gfs2_ok_to_move(struct gfs2_inode *this, struct gfs2_inode *to)
 			break;
 		}
 
-		tmp = gfs2_lookupi(dir, &dotdot, 1, NULL);
+		tmp = gfs2_lookupi(dir, &dotdot, 1);
 		if (IS_ERR(tmp)) {
 			error = PTR_ERR(tmp);
 			break;
diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h
index 6074c2506f75..58f9607d6a86 100644
--- a/fs/gfs2/inode.h
+++ b/fs/gfs2/inode.h
@@ -83,7 +83,7 @@ int gfs2_inode_refresh(struct gfs2_inode *ip);
 int gfs2_dinode_dealloc(struct gfs2_inode *inode);
 int gfs2_change_nlink(struct gfs2_inode *ip, int diff);
 struct inode *gfs2_lookupi(struct inode *dir, const struct qstr *name,
-			   int is_root, struct nameidata *nd);
+			   int is_root);
 struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name,
 			   unsigned int mode, dev_t dev);
 int gfs2_rmdiri(struct gfs2_inode *dip, const struct qstr *name,
diff --git a/fs/gfs2/ops_export.c b/fs/gfs2/ops_export.c
index 990d9f4bc463..9cda8536530c 100644
--- a/fs/gfs2/ops_export.c
+++ b/fs/gfs2/ops_export.c
@@ -134,7 +134,7 @@ static struct dentry *gfs2_get_parent(struct dentry *child)
 	struct dentry *dentry;
 
 	gfs2_str2qstr(&dotdot, "..");
-	inode = gfs2_lookupi(child->d_inode, &dotdot, 1, NULL);
+	inode = gfs2_lookupi(child->d_inode, &dotdot, 1);
 
 	if (!inode)
 		return ERR_PTR(-ENOENT);
diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c
index 4e982532f085..e2c62f73a778 100644
--- a/fs/gfs2/ops_inode.c
+++ b/fs/gfs2/ops_inode.c
@@ -74,7 +74,7 @@ static int gfs2_create(struct inode *dir, struct dentry *dentry,
 			return PTR_ERR(inode);
 		}
 
-		inode = gfs2_lookupi(dir, &dentry->d_name, 0, nd);
+		inode = gfs2_lookupi(dir, &dentry->d_name, 0);
 		if (inode) {
 			if (!IS_ERR(inode)) {
 				gfs2_holder_uninit(ghs);
@@ -109,7 +109,7 @@ static struct dentry *gfs2_lookup(struct inode *dir, struct dentry *dentry,
 
 	dentry->d_op = &gfs2_dops;
 
-	inode = gfs2_lookupi(dir, &dentry->d_name, 0, nd);
+	inode = gfs2_lookupi(dir, &dentry->d_name, 0);
 	if (inode && IS_ERR(inode))
 		return ERR_CAST(inode);
 
diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c
index 63a8a902d9db..ca831991cbc2 100644
--- a/fs/gfs2/super.c
+++ b/fs/gfs2/super.c
@@ -389,7 +389,7 @@ int gfs2_jindex_hold(struct gfs2_sbd *sdp, struct gfs2_holder *ji_gh)
 			break;
 
 		INIT_LIST_HEAD(&jd->extent_list);
-		jd->jd_inode = gfs2_lookupi(sdp->sd_jindex, &name, 1, NULL);
+		jd->jd_inode = gfs2_lookupi(sdp->sd_jindex, &name, 1);
 		if (!jd->jd_inode || IS_ERR(jd->jd_inode)) {
 			if (!jd->jd_inode)
 				error = -ENOENT;

From 58ec42b061bf5dad8fa0370a19966cfd96eaf80c Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Wed, 23 Jul 2008 14:45:55 -0400
Subject: [PATCH 1157/1171] [PATCH] don't pass nameidata to
 __ncp_lookup_validate()

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/ncpfs/dir.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/fs/ncpfs/dir.c b/fs/ncpfs/dir.c
index 011ef0b6d2d4..07e9715b8658 100644
--- a/fs/ncpfs/dir.c
+++ b/fs/ncpfs/dir.c
@@ -266,7 +266,7 @@ leave_me:;
 
 
 static int
-__ncp_lookup_validate(struct dentry * dentry, struct nameidata *nd)
+__ncp_lookup_validate(struct dentry *dentry)
 {
 	struct ncp_server *server;
 	struct dentry *parent;
@@ -340,7 +340,7 @@ ncp_lookup_validate(struct dentry * dentry, struct nameidata *nd)
 {
 	int res;
 	lock_kernel();
-	res = __ncp_lookup_validate(dentry, nd);
+	res = __ncp_lookup_validate(dentry);
 	unlock_kernel();
 	return res;
 }

From 3c333937ee3be114b181c4861188cfe8f6a59697 Mon Sep 17 00:00:00 2001
From: Ulrich Drepper <drepper@redhat.com>
Date: Fri, 25 Jul 2008 22:32:13 -0400
Subject: [PATCH 1158/1171] [PATCH] dup3 fix

Al Viro notice one cornercase that the new dup3() code.  The dup2()
function, as a special case, handles dup-ing to the same file
descriptor.  In this case the current dup3() code does nothing at
all.  I.e., it ingnores the flags parameter.  This shouldn't happen,
the close-on-exec flag should be set if requested.

In case the O_CLOEXEC bit in the flags parameter is not set the
dup3() function should behave in this respect identical to dup2().
This means dup3(fd, fd, 0) should not actively reset the c-o-e
flag.

The patch below implements this minor change.

[AV: credits to Artur Grabowski for bringing that up as potential subtle point
in dup2() behaviour]

Signed-off-by: Ulrich Drepper <drepper@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/fcntl.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/fs/fcntl.c b/fs/fcntl.c
index 9679fcbdeaa0..ce12a6885115 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
@@ -139,8 +139,13 @@ asmlinkage long sys_dup3(unsigned int oldfd, unsigned int newfd, int flags)
 	if (!(file = fcheck(oldfd)))
 		goto out_unlock;
 	err = newfd;
-	if (newfd == oldfd)
+	if (unlikely(newfd == oldfd)) {
+		if (flags & O_CLOEXEC) {
+			fdt = files_fdtable(files);
+			FD_SET(newfd, fdt->close_on_exec);
+		}
 		goto out_unlock;
+	}
 	err = -EBADF;
 	if (newfd >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur)
 		goto out_unlock;

From 516e0cc5646f377ab80fcc2ee639892eccb99853 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sat, 26 Jul 2008 00:39:17 -0400
Subject: [PATCH 1159/1171] [PATCH] f_count may wrap around

make it atomic_long_t; while we are at it, get rid of useless checks in affs,
hfs and hpfs - ->open() always has it equal to 1, ->release() - to 0.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 drivers/net/ppp_generic.c |  6 +++---
 fs/affs/file.c            |  4 ----
 fs/aio.c                  |  6 +++---
 fs/file_table.c           | 10 +++++-----
 fs/hfs/inode.c            |  4 ----
 fs/hfsplus/inode.c        |  4 ----
 include/linux/fs.h        |  6 +++---
 include/net/af_unix.h     |  2 +-
 net/sched/sch_atm.c       |  4 ++--
 net/unix/af_unix.c        |  2 +-
 net/unix/garbage.c        | 18 +++++++++---------
 11 files changed, 27 insertions(+), 39 deletions(-)

diff --git a/drivers/net/ppp_generic.c b/drivers/net/ppp_generic.c
index 739b3ab7bccc..ddccc074a76a 100644
--- a/drivers/net/ppp_generic.c
+++ b/drivers/net/ppp_generic.c
@@ -581,12 +581,12 @@ static long ppp_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 			if (file == ppp->owner)
 				ppp_shutdown_interface(ppp);
 		}
-		if (atomic_read(&file->f_count) <= 2) {
+		if (atomic_long_read(&file->f_count) <= 2) {
 			ppp_release(NULL, file);
 			err = 0;
 		} else
-			printk(KERN_DEBUG "PPPIOCDETACH file->f_count=%d\n",
-			       atomic_read(&file->f_count));
+			printk(KERN_DEBUG "PPPIOCDETACH file->f_count=%ld\n",
+			       atomic_long_read(&file->f_count));
 		unlock_kernel();
 		return err;
 	}
diff --git a/fs/affs/file.c b/fs/affs/file.c
index 6eac7bdeec94..1377b1240b6e 100644
--- a/fs/affs/file.c
+++ b/fs/affs/file.c
@@ -46,8 +46,6 @@ const struct inode_operations affs_file_inode_operations = {
 static int
 affs_file_open(struct inode *inode, struct file *filp)
 {
-	if (atomic_read(&filp->f_count) != 1)
-		return 0;
 	pr_debug("AFFS: open(%lu,%d)\n",
 		 inode->i_ino, atomic_read(&AFFS_I(inode)->i_opencnt));
 	atomic_inc(&AFFS_I(inode)->i_opencnt);
@@ -57,8 +55,6 @@ affs_file_open(struct inode *inode, struct file *filp)
 static int
 affs_file_release(struct inode *inode, struct file *filp)
 {
-	if (atomic_read(&filp->f_count) != 0)
-		return 0;
 	pr_debug("AFFS: release(%lu, %d)\n",
 		 inode->i_ino, atomic_read(&AFFS_I(inode)->i_opencnt));
 
diff --git a/fs/aio.c b/fs/aio.c
index 0051fd94b44e..f658441d5666 100644
--- a/fs/aio.c
+++ b/fs/aio.c
@@ -512,8 +512,8 @@ static void aio_fput_routine(struct work_struct *data)
  */
 static int __aio_put_req(struct kioctx *ctx, struct kiocb *req)
 {
-	dprintk(KERN_DEBUG "aio_put(%p): f_count=%d\n",
-		req, atomic_read(&req->ki_filp->f_count));
+	dprintk(KERN_DEBUG "aio_put(%p): f_count=%ld\n",
+		req, atomic_long_read(&req->ki_filp->f_count));
 
 	assert_spin_locked(&ctx->ctx_lock);
 
@@ -528,7 +528,7 @@ static int __aio_put_req(struct kioctx *ctx, struct kiocb *req)
 	/* Must be done under the lock to serialise against cancellation.
 	 * Call this aio_fput as it duplicates fput via the fput_work.
 	 */
-	if (unlikely(atomic_dec_and_test(&req->ki_filp->f_count))) {
+	if (unlikely(atomic_long_dec_and_test(&req->ki_filp->f_count))) {
 		get_ioctx(ctx);
 		spin_lock(&fput_lock);
 		list_add(&req->ki_list, &fput_head);
diff --git a/fs/file_table.c b/fs/file_table.c
index 83084225b4c3..f45a4493f9e7 100644
--- a/fs/file_table.c
+++ b/fs/file_table.c
@@ -120,7 +120,7 @@ struct file *get_empty_filp(void)
 
 	tsk = current;
 	INIT_LIST_HEAD(&f->f_u.fu_list);
-	atomic_set(&f->f_count, 1);
+	atomic_long_set(&f->f_count, 1);
 	rwlock_init(&f->f_owner.lock);
 	f->f_uid = tsk->fsuid;
 	f->f_gid = tsk->fsgid;
@@ -219,7 +219,7 @@ EXPORT_SYMBOL(init_file);
 
 void fput(struct file *file)
 {
-	if (atomic_dec_and_test(&file->f_count))
+	if (atomic_long_dec_and_test(&file->f_count))
 		__fput(file);
 }
 
@@ -294,7 +294,7 @@ struct file *fget(unsigned int fd)
 	rcu_read_lock();
 	file = fcheck_files(files, fd);
 	if (file) {
-		if (!atomic_inc_not_zero(&file->f_count)) {
+		if (!atomic_long_inc_not_zero(&file->f_count)) {
 			/* File object ref couldn't be taken */
 			rcu_read_unlock();
 			return NULL;
@@ -326,7 +326,7 @@ struct file *fget_light(unsigned int fd, int *fput_needed)
 		rcu_read_lock();
 		file = fcheck_files(files, fd);
 		if (file) {
-			if (atomic_inc_not_zero(&file->f_count))
+			if (atomic_long_inc_not_zero(&file->f_count))
 				*fput_needed = 1;
 			else
 				/* Didn't get the reference, someone's freed */
@@ -341,7 +341,7 @@ struct file *fget_light(unsigned int fd, int *fput_needed)
 
 void put_filp(struct file *file)
 {
-	if (atomic_dec_and_test(&file->f_count)) {
+	if (atomic_long_dec_and_test(&file->f_count)) {
 		security_file_free(file);
 		file_kill(file);
 		file_free(file);
diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c
index aa73f3fd5dd9..7e19835efa2e 100644
--- a/fs/hfs/inode.c
+++ b/fs/hfs/inode.c
@@ -522,8 +522,6 @@ static int hfs_file_open(struct inode *inode, struct file *file)
 {
 	if (HFS_IS_RSRC(inode))
 		inode = HFS_I(inode)->rsrc_inode;
-	if (atomic_read(&file->f_count) != 1)
-		return 0;
 	atomic_inc(&HFS_I(inode)->opencnt);
 	return 0;
 }
@@ -534,8 +532,6 @@ static int hfs_file_release(struct inode *inode, struct file *file)
 
 	if (HFS_IS_RSRC(inode))
 		inode = HFS_I(inode)->rsrc_inode;
-	if (atomic_read(&file->f_count) != 0)
-		return 0;
 	if (atomic_dec_and_test(&HFS_I(inode)->opencnt)) {
 		mutex_lock(&inode->i_mutex);
 		hfs_file_truncate(inode);
diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c
index d4014e3044d2..b085d64a2b67 100644
--- a/fs/hfsplus/inode.c
+++ b/fs/hfsplus/inode.c
@@ -254,8 +254,6 @@ static int hfsplus_file_open(struct inode *inode, struct file *file)
 {
 	if (HFSPLUS_IS_RSRC(inode))
 		inode = HFSPLUS_I(inode).rsrc_inode;
-	if (atomic_read(&file->f_count) != 1)
-		return 0;
 	atomic_inc(&HFSPLUS_I(inode).opencnt);
 	return 0;
 }
@@ -266,8 +264,6 @@ static int hfsplus_file_release(struct inode *inode, struct file *file)
 
 	if (HFSPLUS_IS_RSRC(inode))
 		inode = HFSPLUS_I(inode).rsrc_inode;
-	if (atomic_read(&file->f_count) != 0)
-		return 0;
 	if (atomic_dec_and_test(&HFSPLUS_I(inode).opencnt)) {
 		mutex_lock(&inode->i_mutex);
 		hfsplus_file_truncate(inode);
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 9d2de4cadabd..7676fa1c20ae 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -795,7 +795,7 @@ struct file {
 #define f_dentry	f_path.dentry
 #define f_vfsmnt	f_path.mnt
 	const struct file_operations	*f_op;
-	atomic_t		f_count;
+	atomic_long_t		f_count;
 	unsigned int 		f_flags;
 	mode_t			f_mode;
 	loff_t			f_pos;
@@ -824,8 +824,8 @@ extern spinlock_t files_lock;
 #define file_list_lock() spin_lock(&files_lock);
 #define file_list_unlock() spin_unlock(&files_lock);
 
-#define get_file(x)	atomic_inc(&(x)->f_count)
-#define file_count(x)	atomic_read(&(x)->f_count)
+#define get_file(x)	atomic_long_inc(&(x)->f_count)
+#define file_count(x)	atomic_long_read(&(x)->f_count)
 
 #ifdef CONFIG_DEBUG_WRITECOUNT
 static inline void file_take_write(struct file *f)
diff --git a/include/net/af_unix.h b/include/net/af_unix.h
index 2dfa96b0575e..7dd29b7e461d 100644
--- a/include/net/af_unix.h
+++ b/include/net/af_unix.h
@@ -51,7 +51,7 @@ struct unix_sock {
         struct sock		*peer;
         struct sock		*other;
 	struct list_head	link;
-        atomic_t                inflight;
+        atomic_long_t           inflight;
         spinlock_t		lock;
 	unsigned int		gc_candidate : 1;
         wait_queue_head_t       peer_wait;
diff --git a/net/sched/sch_atm.c b/net/sched/sch_atm.c
index 04faa835be17..6b517b9dac5b 100644
--- a/net/sched/sch_atm.c
+++ b/net/sched/sch_atm.c
@@ -162,7 +162,7 @@ static void atm_tc_put(struct Qdisc *sch, unsigned long cl)
 	qdisc_destroy(flow->q);
 	tcf_destroy_chain(&flow->filter_list);
 	if (flow->sock) {
-		pr_debug("atm_tc_put: f_count %d\n",
+		pr_debug("atm_tc_put: f_count %ld\n",
 			file_count(flow->sock->file));
 		flow->vcc->pop = flow->old_pop;
 		sockfd_put(flow->sock);
@@ -259,7 +259,7 @@ static int atm_tc_change(struct Qdisc *sch, u32 classid, u32 parent,
 	sock = sockfd_lookup(fd, &error);
 	if (!sock)
 		return error;	/* f_count++ */
-	pr_debug("atm_tc_change: f_count %d\n", file_count(sock->file));
+	pr_debug("atm_tc_change: f_count %ld\n", file_count(sock->file));
 	if (sock->ops->family != PF_ATMSVC && sock->ops->family != PF_ATMPVC) {
 		error = -EPROTOTYPE;
 		goto err_out;
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 70ceb1604ad8..6e7fec74bdb3 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -603,7 +603,7 @@ static struct sock * unix_create1(struct net *net, struct socket *sock)
 	u->dentry = NULL;
 	u->mnt	  = NULL;
 	spin_lock_init(&u->lock);
-	atomic_set(&u->inflight, 0);
+	atomic_long_set(&u->inflight, 0);
 	INIT_LIST_HEAD(&u->link);
 	mutex_init(&u->readlock); /* single task reading lock */
 	init_waitqueue_head(&u->peer_wait);
diff --git a/net/unix/garbage.c b/net/unix/garbage.c
index ebdff3d877a1..2a27b84f740b 100644
--- a/net/unix/garbage.c
+++ b/net/unix/garbage.c
@@ -127,7 +127,7 @@ void unix_inflight(struct file *fp)
 	if(s) {
 		struct unix_sock *u = unix_sk(s);
 		spin_lock(&unix_gc_lock);
-		if (atomic_inc_return(&u->inflight) == 1) {
+		if (atomic_long_inc_return(&u->inflight) == 1) {
 			BUG_ON(!list_empty(&u->link));
 			list_add_tail(&u->link, &gc_inflight_list);
 		} else {
@@ -145,7 +145,7 @@ void unix_notinflight(struct file *fp)
 		struct unix_sock *u = unix_sk(s);
 		spin_lock(&unix_gc_lock);
 		BUG_ON(list_empty(&u->link));
-		if (atomic_dec_and_test(&u->inflight))
+		if (atomic_long_dec_and_test(&u->inflight))
 			list_del_init(&u->link);
 		unix_tot_inflight--;
 		spin_unlock(&unix_gc_lock);
@@ -237,17 +237,17 @@ static void scan_children(struct sock *x, void (*func)(struct unix_sock *),
 
 static void dec_inflight(struct unix_sock *usk)
 {
-	atomic_dec(&usk->inflight);
+	atomic_long_dec(&usk->inflight);
 }
 
 static void inc_inflight(struct unix_sock *usk)
 {
-	atomic_inc(&usk->inflight);
+	atomic_long_inc(&usk->inflight);
 }
 
 static void inc_inflight_move_tail(struct unix_sock *u)
 {
-	atomic_inc(&u->inflight);
+	atomic_long_inc(&u->inflight);
 	/*
 	 * If this is still a candidate, move it to the end of the
 	 * list, so that it's checked even if it was already passed
@@ -288,11 +288,11 @@ void unix_gc(void)
 	 * before the detach without atomicity guarantees.
 	 */
 	list_for_each_entry_safe(u, next, &gc_inflight_list, link) {
-		int total_refs;
-		int inflight_refs;
+		long total_refs;
+		long inflight_refs;
 
 		total_refs = file_count(u->sk.sk_socket->file);
-		inflight_refs = atomic_read(&u->inflight);
+		inflight_refs = atomic_long_read(&u->inflight);
 
 		BUG_ON(inflight_refs < 1);
 		BUG_ON(total_refs < inflight_refs);
@@ -324,7 +324,7 @@ void unix_gc(void)
 		/* Move cursor to after the current position. */
 		list_move(&cursor, &u->link);
 
-		if (atomic_read(&u->inflight) > 0) {
+		if (atomic_long_read(&u->inflight) > 0) {
 			list_move_tail(&u->link, &gc_inflight_list);
 			u->gc_candidate = 0;
 			scan_children(&u->sk, inc_inflight_move_tail, NULL);

From 964bd183624c03680796b63b4ab97ee3905a806a Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sat, 26 Jul 2008 03:33:14 -0400
Subject: [PATCH 1160/1171] [PATCH] get rid of __user_path_lookup_open

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/exec.c             | 12 +++++++++---
 fs/namei.c            | 13 -------------
 include/linux/namei.h |  1 -
 3 files changed, 9 insertions(+), 17 deletions(-)

diff --git a/fs/exec.c b/fs/exec.c
index eca58c29eded..9696bbf0f0b1 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -106,11 +106,17 @@ static inline void put_binfmt(struct linux_binfmt * fmt)
  */
 asmlinkage long sys_uselib(const char __user * library)
 {
-	struct file * file;
+	struct file *file;
 	struct nameidata nd;
-	int error;
+	char *tmp = getname(library);
+	int error = PTR_ERR(tmp);
 
-	error = __user_path_lookup_open(library, LOOKUP_FOLLOW, &nd, FMODE_READ|FMODE_EXEC);
+	if (!IS_ERR(tmp)) {
+		error = path_lookup_open(AT_FDCWD, tmp,
+					 LOOKUP_FOLLOW, &nd,
+					 FMODE_READ|FMODE_EXEC);
+		putname(tmp);
+	}
 	if (error)
 		goto out;
 
diff --git a/fs/namei.c b/fs/namei.c
index 38ceb6e06eba..a7b0a0b80128 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -1193,19 +1193,6 @@ static int path_lookup_create(int dfd, const char *name,
 			nd, open_flags, create_mode);
 }
 
-int __user_path_lookup_open(const char __user *name, unsigned int lookup_flags,
-		struct nameidata *nd, int open_flags)
-{
-	char *tmp = getname(name);
-	int err = PTR_ERR(tmp);
-
-	if (!IS_ERR(tmp)) {
-		err = __path_lookup_intent_open(AT_FDCWD, tmp, lookup_flags, nd, open_flags, 0);
-		putname(tmp);
-	}
-	return err;
-}
-
 static struct dentry *__lookup_hash(struct qstr *name,
 		struct dentry *base, struct nameidata *nd)
 {
diff --git a/include/linux/namei.h b/include/linux/namei.h
index 00888ff69504..68f8c3203c89 100644
--- a/include/linux/namei.h
+++ b/include/linux/namei.h
@@ -65,7 +65,6 @@ extern int path_lookup(const char *, unsigned, struct nameidata *);
 extern int vfs_path_lookup(struct dentry *, struct vfsmount *,
 			   const char *, unsigned int, struct nameidata *);
 
-extern int __user_path_lookup_open(const char __user *, unsigned lookup_flags, struct nameidata *nd, int open_flags);
 extern int path_lookup_open(int dfd, const char *name, unsigned lookup_flags, struct nameidata *, int open_flags);
 extern struct file *lookup_instantiate_filp(struct nameidata *nd, struct dentry *dentry,
 		int (*open)(struct inode *, struct file *));

From 3f8206d496e9e9495afb1d4e70d29712b4d403c9 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sat, 26 Jul 2008 03:46:43 -0400
Subject: [PATCH 1161/1171] [PATCH] get rid of indirect users of namei.h

fs.h needs path.h, not namei.h; nfs_fs.h doesn't need it at all.
Several places in the tree needed direct include.

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/nfsd/nfsctl.c       | 1 +
 fs/ubifs/file.c        | 1 +
 include/linux/fs.h     | 2 +-
 include/linux/nfs_fs.h | 1 -
 kernel/cgroup.c        | 1 +
 5 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c
index 1955a2702e60..c53e65f8f3a2 100644
--- a/fs/nfsd/nfsctl.c
+++ b/fs/nfsd/nfsctl.c
@@ -12,6 +12,7 @@
 #include <linux/time.h>
 #include <linux/errno.h>
 #include <linux/fs.h>
+#include <linux/namei.h>
 #include <linux/fcntl.h>
 #include <linux/net.h>
 #include <linux/in.h>
diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c
index 005a3b854d96..8565e586e533 100644
--- a/fs/ubifs/file.c
+++ b/fs/ubifs/file.c
@@ -53,6 +53,7 @@
 
 #include "ubifs.h"
 #include <linux/mount.h>
+#include <linux/namei.h>
 
 static int read_block(struct inode *inode, void *addr, unsigned int block,
 		      struct ubifs_data_node *dn)
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 7676fa1c20ae..8252b045e624 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -279,7 +279,7 @@ extern int dir_notify_enable;
 #include <linux/types.h>
 #include <linux/kdev_t.h>
 #include <linux/dcache.h>
-#include <linux/namei.h>
+#include <linux/path.h>
 #include <linux/stat.h>
 #include <linux/cache.h>
 #include <linux/kobject.h>
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index f08f9ca602af..78a5922a2f11 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -42,7 +42,6 @@
 #include <linux/in.h>
 #include <linux/kref.h>
 #include <linux/mm.h>
-#include <linux/namei.h>
 #include <linux/pagemap.h>
 #include <linux/rbtree.h>
 #include <linux/rwsem.h>
diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index 89bd6fb7894f..657f8f8d93a5 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -45,6 +45,7 @@
 #include <linux/delayacct.h>
 #include <linux/cgroupstats.h>
 #include <linux/hash.h>
+#include <linux/namei.h>
 
 #include <asm/atomic.h>
 

From 4cc38a1b383f0c6c65a3fef4ff8144e8000e4ec3 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sat, 26 Jul 2008 03:48:26 -0400
Subject: [PATCH 1162/1171] [PATCH] remove remaining namei_{32,64}.h crap

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 include/asm-sparc/namei_32.h | 13 -------------
 include/asm-sparc/namei_64.h | 13 -------------
 2 files changed, 26 deletions(-)
 delete mode 100644 include/asm-sparc/namei_32.h
 delete mode 100644 include/asm-sparc/namei_64.h

diff --git a/include/asm-sparc/namei_32.h b/include/asm-sparc/namei_32.h
deleted file mode 100644
index 0646102fb020..000000000000
--- a/include/asm-sparc/namei_32.h
+++ /dev/null
@@ -1,13 +0,0 @@
-/*
- * linux/include/asm-sparc/namei.h
- *
- * Routines to handle famous /usr/gnemul/s*.
- * Included from linux/fs/namei.c
- */
-
-#ifndef __SPARC_NAMEI_H
-#define __SPARC_NAMEI_H
-
-#define __emul_prefix() NULL
-
-#endif /* __SPARC_NAMEI_H */
diff --git a/include/asm-sparc/namei_64.h b/include/asm-sparc/namei_64.h
deleted file mode 100644
index cbc1b4c06891..000000000000
--- a/include/asm-sparc/namei_64.h
+++ /dev/null
@@ -1,13 +0,0 @@
-/*
- * linux/include/asm-sparc64/namei.h
- *
- * Routines to handle famous /usr/gnemul/s*.
- * Included from linux/fs/namei.c
- */
-
-#ifndef __SPARC64_NAMEI_H
-#define __SPARC64_NAMEI_H
-
-#define __emul_prefix() NULL
-
-#endif /* __SPARC64_NAMEI_H */

From 6c5d0512a091480c9f981162227fdb1c9d70e555 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sat, 26 Jul 2008 13:38:19 -0400
Subject: [PATCH 1163/1171] [PATCH] get rid of corner case in dup3() entirely

Since Ulrich is OK with getting rid of dup3(fd, fd, flags) completely,
to hell the damn thing goes.  Corner case for dup2() is handled in
sys_dup2() (complete with -EBADF if dup2(fd, fd) is called with fd
that is not open), the rest is done in dup3().

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/fcntl.c | 20 +++++++++++---------
 1 file changed, 11 insertions(+), 9 deletions(-)

diff --git a/fs/fcntl.c b/fs/fcntl.c
index ce12a6885115..3deec9887089 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
@@ -135,18 +135,12 @@ asmlinkage long sys_dup3(unsigned int oldfd, unsigned int newfd, int flags)
 	if ((flags & ~O_CLOEXEC) != 0)
 		return -EINVAL;
 
+	if (unlikely(oldfd == newfd))
+		return -EINVAL;
+
 	spin_lock(&files->file_lock);
 	if (!(file = fcheck(oldfd)))
 		goto out_unlock;
-	err = newfd;
-	if (unlikely(newfd == oldfd)) {
-		if (flags & O_CLOEXEC) {
-			fdt = files_fdtable(files);
-			FD_SET(newfd, fdt->close_on_exec);
-		}
-		goto out_unlock;
-	}
-	err = -EBADF;
 	if (newfd >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur)
 		goto out_unlock;
 	get_file(file);			/* We are now finished with oldfd */
@@ -194,6 +188,14 @@ out_fput:
 
 asmlinkage long sys_dup2(unsigned int oldfd, unsigned int newfd)
 {
+	if (unlikely(newfd == oldfd)) { /* corner case */
+		struct files_struct *files = current->files;
+		rcu_read_lock();
+		if (!fcheck_files(files, oldfd))
+			oldfd = -EBADF;
+		rcu_read_unlock();
+		return oldfd;
+	}
 	return sys_dup3(oldfd, newfd, 0);
 }
 

From 4e1e018ecc6f7bfd10fc75b3ff9715cc8164e0a2 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Sat, 26 Jul 2008 16:01:20 -0400
Subject: [PATCH 1164/1171] [PATCH] fix RLIM_NOFILE handling

* dup2() should return -EBADF on exceeded sysctl_nr_open
* dup() should *not* return -EINVAL even if you have rlimit set to 0;
  it should get -EMFILE instead.

Check for orig_start exceeding rlimit taken to sys_fcntl().
Failing expand_files() in dup{2,3}() now gets -EMFILE remapped to -EBADF.
Consequently, remaining checks for rlimit are taken to expand_files().

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/fcntl.c | 18 ++++++------------
 fs/file.c  |  9 +++++++++
 fs/open.c  |  9 ---------
 3 files changed, 15 insertions(+), 21 deletions(-)

diff --git a/fs/fcntl.c b/fs/fcntl.c
index 3deec9887089..61d625136813 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
@@ -64,11 +64,6 @@ static int locate_fd(unsigned int orig_start, int cloexec)
 	struct fdtable *fdt;
 
 	spin_lock(&files->file_lock);
-
-	error = -EINVAL;
-	if (orig_start >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur)
-		goto out;
-
 repeat:
 	fdt = files_fdtable(files);
 	/*
@@ -83,10 +78,6 @@ repeat:
 	if (start < fdt->max_fds)
 		newfd = find_next_zero_bit(fdt->open_fds->fds_bits,
 					   fdt->max_fds, start);
-	
-	error = -EMFILE;
-	if (newfd >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur)
-		goto out;
 
 	error = expand_files(files, newfd);
 	if (error < 0)
@@ -141,13 +132,14 @@ asmlinkage long sys_dup3(unsigned int oldfd, unsigned int newfd, int flags)
 	spin_lock(&files->file_lock);
 	if (!(file = fcheck(oldfd)))
 		goto out_unlock;
-	if (newfd >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur)
-		goto out_unlock;
 	get_file(file);			/* We are now finished with oldfd */
 
 	err = expand_files(files, newfd);
-	if (err < 0)
+	if (unlikely(err < 0)) {
+		if (err == -EMFILE)
+			err = -EBADF;
 		goto out_fput;
+	}
 
 	/* To avoid races with open() and dup(), we will mark the fd as
 	 * in-use in the open-file bitmap throughout the entire dup2()
@@ -328,6 +320,8 @@ static long do_fcntl(int fd, unsigned int cmd, unsigned long arg,
 	switch (cmd) {
 	case F_DUPFD:
 	case F_DUPFD_CLOEXEC:
+		if (arg >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur)
+			break;
 		get_file(filp);
 		err = dupfd(filp, arg, cmd == F_DUPFD_CLOEXEC);
 		break;
diff --git a/fs/file.c b/fs/file.c
index 7b3887e054d0..d8773b19fe47 100644
--- a/fs/file.c
+++ b/fs/file.c
@@ -250,9 +250,18 @@ int expand_files(struct files_struct *files, int nr)
 	struct fdtable *fdt;
 
 	fdt = files_fdtable(files);
+
+	/*
+	 * N.B. For clone tasks sharing a files structure, this test
+	 * will limit the total number of files that can be opened.
+	 */
+	if (nr >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur)
+		return -EMFILE;
+
 	/* Do we need to expand? */
 	if (nr < fdt->max_fds)
 		return 0;
+
 	/* Can we expand? */
 	if (nr >= sysctl_nr_open)
 		return -EMFILE;
diff --git a/fs/open.c b/fs/open.c
index 3fe1a6857c75..52647be277a2 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -972,7 +972,6 @@ int get_unused_fd_flags(int flags)
 	int fd, error;
 	struct fdtable *fdt;
 
-  	error = -EMFILE;
 	spin_lock(&files->file_lock);
 
 repeat:
@@ -980,13 +979,6 @@ repeat:
 	fd = find_next_zero_bit(fdt->open_fds->fds_bits, fdt->max_fds,
 				files->next_fd);
 
-	/*
-	 * N.B. For clone tasks sharing a files structure, this test
-	 * will limit the total number of files that can be opened.
-	 */
-	if (fd >= current->signal->rlim[RLIMIT_NOFILE].rlim_cur)
-		goto out;
-
 	/* Do we need to expand the fd array or fd set?  */
 	error = expand_files(files, fd);
 	if (error < 0)
@@ -997,7 +989,6 @@ repeat:
 	 	 * If we needed to expand the fs array we
 		 * might have blocked - try again.
 		 */
-		error = -EMFILE;
 		goto repeat;
 	}
 

From 6a9436d0c3cbe8941b1acd5b0736d355295cad98 Mon Sep 17 00:00:00 2001
From: Michael Buesch <mb@bu3sch.de>
Date: Sat, 26 Jul 2008 15:22:26 -0700
Subject: [PATCH 1165/1171] gpiolib: fix typo in comment

This fixes an off-by-one error in a comment.

Signed-off-by: Michael Buesch <mb@bu3sch.de>
Cc: David Brownell <david-b@pacbell.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/asm-generic/gpio.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/asm-generic/gpio.h b/include/asm-generic/gpio.h
index a3034d20ebd5..c764a8fcb058 100644
--- a/include/asm-generic/gpio.h
+++ b/include/asm-generic/gpio.h
@@ -13,7 +13,7 @@
  *
  * While the GPIO programming interface defines valid GPIO numbers
  * to be in the range 0..MAX_INT, this library restricts them to the
- * smaller range 0..ARCH_NR_GPIOS.
+ * smaller range 0..ARCH_NR_GPIOS-1.
  */
 
 #ifndef ARCH_NR_GPIOS

From b2d002dba5a8a4c0c3ec96fd1ff3c9def6bd71a1 Mon Sep 17 00:00:00 2001
From: Andrea Righi <righi.andrea@gmail.com>
Date: Sat, 26 Jul 2008 15:22:27 -0700
Subject: [PATCH 1166/1171] task IO accounting: correctly account threads IO
 statistics

Oleg Nesterov points out that we should check that the task is still alive
before we iterate over the threads.  This patch includes a fixup for this.

Also simplify do_io_accounting() implementation.

Signed-off-by: Andrea Righi <righi.andrea@gmail.com>
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 fs/proc/base.c | 54 ++++++++++++++++++++------------------------------
 1 file changed, 21 insertions(+), 33 deletions(-)

diff --git a/fs/proc/base.c b/fs/proc/base.c
index 81bce6791bfc..d744aa3c9f74 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -2406,35 +2406,18 @@ static int do_io_accounting(struct task_struct *task, char *buffer, int whole)
 	u64 rchar, wchar, syscr, syscw;
 	struct task_io_accounting ioac;
 
-	if (!whole) {
-		rchar = task->rchar;
-		wchar = task->wchar;
-		syscr = task->syscr;
-		syscw = task->syscw;
-		memcpy(&ioac, &task->ioac, sizeof(ioac));
-	} else {
+	rchar = task->rchar;
+	wchar = task->wchar;
+	syscr = task->syscr;
+	syscw = task->syscw;
+	memcpy(&ioac, &task->ioac, sizeof(ioac));
+
+	if (whole) {
 		unsigned long flags;
-		struct task_struct *t = task;
-		rchar = wchar = syscr = syscw = 0;
-		memset(&ioac, 0, sizeof(ioac));
-
-		rcu_read_lock();
-		do {
-			rchar += t->rchar;
-			wchar += t->wchar;
-			syscr += t->syscr;
-			syscw += t->syscw;
-
-			ioac.read_bytes += t->ioac.read_bytes;
-			ioac.write_bytes += t->ioac.write_bytes;
-			ioac.cancelled_write_bytes +=
-					t->ioac.cancelled_write_bytes;
-			t = next_thread(t);
-		} while (t != task);
-		rcu_read_unlock();
 
 		if (lock_task_sighand(task, &flags)) {
 			struct signal_struct *sig = task->signal;
+			struct task_struct *t = task;
 
 			rchar += sig->rchar;
 			wchar += sig->wchar;
@@ -2445,11 +2428,20 @@ static int do_io_accounting(struct task_struct *task, char *buffer, int whole)
 			ioac.write_bytes += sig->ioac.write_bytes;
 			ioac.cancelled_write_bytes +=
 					sig->ioac.cancelled_write_bytes;
+			while_each_thread(task, t) {
+				rchar += t->rchar;
+				wchar += t->wchar;
+				syscr += t->syscr;
+				syscw += t->syscw;
 
+				ioac.read_bytes += t->ioac.read_bytes;
+				ioac.write_bytes += t->ioac.write_bytes;
+				ioac.cancelled_write_bytes +=
+					t->ioac.cancelled_write_bytes;
+			}
 			unlock_task_sighand(task, &flags);
 		}
 	}
-
 	return sprintf(buffer,
 			"rchar: %llu\n"
 			"wchar: %llu\n"
@@ -2458,13 +2450,9 @@ static int do_io_accounting(struct task_struct *task, char *buffer, int whole)
 			"read_bytes: %llu\n"
 			"write_bytes: %llu\n"
 			"cancelled_write_bytes: %llu\n",
-			(unsigned long long)rchar,
-			(unsigned long long)wchar,
-			(unsigned long long)syscr,
-			(unsigned long long)syscw,
-			(unsigned long long)ioac.read_bytes,
-			(unsigned long long)ioac.write_bytes,
-			(unsigned long long)ioac.cancelled_write_bytes);
+			rchar, wchar, syscr, syscw,
+			ioac.read_bytes, ioac.write_bytes,
+			ioac.cancelled_write_bytes);
 }
 
 static int proc_tid_io_accounting(struct task_struct *task, char *buffer)

From 510a35d4a47802f4a0028aa6bd2ca2170da5e32f Mon Sep 17 00:00:00 2001
From: Andrea Righi <righi.andrea@gmail.com>
Date: Sat, 26 Jul 2008 15:22:27 -0700
Subject: [PATCH 1167/1171] hugetlb: remove unused variable warning

Remove the following warning when CONFIG_HUGETLB_PAGE is not set:

	ipc/shm.c: In function `shm_get_stat':
	ipc/shm.c:565: warning: unused variable `h'

[akpm@linux-foundation.org: use tabs, not spaces]
Signed-off-by: Andrea Righi <righi.andrea@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 include/linux/hugetlb.h | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index 9a71d4cc88c8..32e0ef0f6e1f 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -273,7 +273,10 @@ struct hstate {};
 #define huge_page_mask(h) PAGE_MASK
 #define huge_page_order(h) 0
 #define huge_page_shift(h) PAGE_SHIFT
-#define pages_per_huge_page(h) 1
+static inline unsigned int pages_per_huge_page(struct hstate *h)
+{
+	return 1;
+}
 #endif
 
 #endif /* _LINUX_HUGETLB_H */

From 3b8f14b41026fb7d7e9a4af2a4128a702d07ad26 Mon Sep 17 00:00:00 2001
From: Adrian Bunk <bunk@kernel.org>
Date: Sat, 26 Jul 2008 15:22:28 -0700
Subject: [PATCH 1168/1171] mm/util.c must #include <linux/sched.h>

mm/util.c: In function 'arch_pick_mmap_layout':
  mm/util.c:144: error: dereferencing pointer to incomplete type
  mm/util.c:145: error: 'arch_get_unmapped_area' undeclared (first use in this function)
  mm/util.c:145: error: (Each undeclared identifier is reported only once
  mm/util.c:145: error: for each function it appears in.)
  mm/util.c:146: error: 'arch_unmap_area' undeclared (first use in this function)

Signed-off-by: Adrian Bunk <bunk@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 mm/util.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/mm/util.c b/mm/util.c
index 0efd83097ecf..d8d02210e06c 100644
--- a/mm/util.c
+++ b/mm/util.c
@@ -3,6 +3,7 @@
 #include <linux/string.h>
 #include <linux/module.h>
 #include <linux/err.h>
+#include <linux/sched.h>
 #include <asm/uaccess.h>
 
 /**

From 78681ac08a611313595d13cafabae1183b71ef48 Mon Sep 17 00:00:00 2001
From: Randy Dunlap <randy.dunlap@oracle.com>
Date: Sat, 26 Jul 2008 15:22:28 -0700
Subject: [PATCH 1169/1171] firmware: fix memmap printk format warnings

Fix firmware/memmap printk format warnings:

  drivers/firmware/memmap.c:156: warning: format '%llx' expects type 'long long unsigned int', but argument 4 has type 'resource_size_t'
  drivers/firmware/memmap.c:161: warning: format '%llx' expects type 'long long unsigned int', but argument 4 has type 'resource_size_t'

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Cc: Bernhard Walle <bwalle@suse.de>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/firmware/memmap.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/drivers/firmware/memmap.c b/drivers/firmware/memmap.c
index e23399c7f773..001622eb86f9 100644
--- a/drivers/firmware/memmap.c
+++ b/drivers/firmware/memmap.c
@@ -153,12 +153,14 @@ int __init firmware_map_add_early(resource_size_t start, resource_size_t end,
 
 static ssize_t start_show(struct firmware_map_entry *entry, char *buf)
 {
-	return snprintf(buf, PAGE_SIZE, "0x%llx\n", entry->start);
+	return snprintf(buf, PAGE_SIZE, "0x%llx\n",
+		(unsigned long long)entry->start);
 }
 
 static ssize_t end_show(struct firmware_map_entry *entry, char *buf)
 {
-	return snprintf(buf, PAGE_SIZE, "0x%llx\n", entry->end);
+	return snprintf(buf, PAGE_SIZE, "0x%llx\n",
+		(unsigned long long)entry->end);
 }
 
 static ssize_t type_show(struct firmware_map_entry *entry, char *buf)

From 732730d48dc777f6577a6e0fece42b860324998e Mon Sep 17 00:00:00 2001
From: Geert Uytterhoeven <geert@linux-m68k.org>
Date: Sun, 27 Jul 2008 01:39:52 +0200
Subject: [PATCH 1170/1171] m68k: gs: use tty_port fixes

commit b5391e29f428d11755ca2c91074c6db6f5c69d7c ("gs: use tty_port")
forgot to update the m68k gs serial drivers.

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/char/ser_a2232.c | 52 +++++++++++++++++------------------
 drivers/char/vme_scc.c   | 59 ++++++++++++++++++++--------------------
 2 files changed, 56 insertions(+), 55 deletions(-)

diff --git a/drivers/char/ser_a2232.c b/drivers/char/ser_a2232.c
index 4ba3aec9e1cd..7b0c35207d9b 100644
--- a/drivers/char/ser_a2232.c
+++ b/drivers/char/ser_a2232.c
@@ -192,7 +192,7 @@ static inline void a2232_receive_char(struct a2232_port *port, int ch, int err)
 	Maybe one could implement a more efficient version by not only
 	transferring one character at a time.
 */
-	struct tty_struct *tty = port->gs.tty;
+	struct tty_struct *tty = port->gs.port.tty;
 
 #if 0
 	switch(err) {
@@ -226,7 +226,7 @@ static void a2232_disable_tx_interrupts(void *ptr)
 
 	/* Does this here really have to be? */
 	local_irq_save(flags);
-	port->gs.flags &= ~GS_TX_INTEN;
+	port->gs.port.flags &= ~GS_TX_INTEN;
 	local_irq_restore(flags);
 }
 
@@ -242,7 +242,7 @@ static void a2232_enable_tx_interrupts(void *ptr)
 
 	/* Does this here really have to be? */
 	local_irq_save(flags);
-	port->gs.flags |= GS_TX_INTEN;
+	port->gs.port.flags |= GS_TX_INTEN;
 	local_irq_restore(flags);
 }
 
@@ -276,9 +276,9 @@ static void a2232_shutdown_port(void *ptr)
 
 	local_irq_save(flags);
 
-	port->gs.flags &= ~GS_ACTIVE;
+	port->gs.port.flags &= ~GS_ACTIVE;
 	
-	if (port->gs.tty && port->gs.tty->termios->c_cflag & HUPCL) {
+	if (port->gs.port.tty && port->gs.port.tty->termios->c_cflag & HUPCL) {
 		/* Set DTR and RTS to Low, flush output.
 		   The NetBSD driver "msc.c" does it this way. */
 		stat->Command = (	(stat->Command & ~A2232CMD_CMask) | 
@@ -309,7 +309,7 @@ static int  a2232_set_real_termios(void *ptr)
 	volatile struct a2232status *status;
 	volatile struct a2232memory *mem;
 
-	if (!port->gs.tty || !port->gs.tty->termios) return 0;
+	if (!port->gs.port.tty || !port->gs.port.tty->termios) return 0;
 
 	status = a2232stat(port->which_a2232, port->which_port_on_a2232);
 	mem = a2232mem(port->which_a2232);
@@ -345,7 +345,7 @@ static int  a2232_set_real_termios(void *ptr)
 	}
 	a2232_param |= rate;
 
-	cflag  = port->gs.tty->termios->c_cflag;
+	cflag  = port->gs.port.tty->termios->c_cflag;
 
 	// get character size
 	chsize = cflag & CSIZE;
@@ -382,7 +382,7 @@ static int  a2232_set_real_termios(void *ptr)
 		the conventional way of inserting START/STOP characters
 		by hand in throttle()/unthrottle().
 	*/
-	softflow = !!( port->gs.tty->termios->c_iflag & IXOFF );
+	softflow = !!( port->gs.port.tty->termios->c_iflag & IXOFF );
 
 	// get Parity (Enabled/Disabled? If Enabled, Odd or Even?)
 	parity = cflag & (PARENB | PARODD);
@@ -400,9 +400,9 @@ static int  a2232_set_real_termios(void *ptr)
 	/*	Hmm. Maybe an own a2232_port structure
 		member would be cleaner?	*/
 	if (cflag & CLOCAL)
-		port->gs.flags &= ~ASYNC_CHECK_CD;
+		port->gs.port.flags &= ~ASYNC_CHECK_CD;
 	else
-		port->gs.flags |= ASYNC_CHECK_CD;
+		port->gs.port.flags |= ASYNC_CHECK_CD;
 
 
 	/* Now we have all parameters and can go to set them: */
@@ -482,18 +482,18 @@ static int  a2232_open(struct tty_struct * tty, struct file * filp)
 	port = &a2232_ports[line];
 	
 	tty->driver_data = port;
-	port->gs.tty = tty;
-	port->gs.count++;
+	port->gs.port.tty = tty;
+	port->gs.port.count++;
 	retval = gs_init_port(&port->gs);
 	if (retval) {
-		port->gs.count--;
+		port->gs.port.count--;
 		return retval;
 	}
-	port->gs.flags |= GS_ACTIVE;
+	port->gs.port.flags |= GS_ACTIVE;
 	retval = gs_block_til_ready(port, filp);
 
 	if (retval) {
-		port->gs.count--;
+		port->gs.port.count--;
 		return retval;
 	}
 
@@ -522,7 +522,7 @@ int ch, err, n, p;
 		for (p = 0; p < NUMLINES; p++){	/* for every port on this board */
 			err = 0;
 			port = &a2232_ports[n*NUMLINES+p];
-			if ( port->gs.flags & GS_ACTIVE ){ /* if the port is used */
+			if ( port->gs.port.flags & GS_ACTIVE ){ /* if the port is used */
 
 				status = a2232stat(n,p);
 
@@ -577,8 +577,8 @@ int ch, err, n, p;
 				obuf = mem->OutBuf[p];
 				bufpos = status->OutHead;
 				while ( (port->gs.xmit_cnt > 0)		&&
-					(!port->gs.tty->stopped)	&&
-					(!port->gs.tty->hw_stopped) ){	/* While there are chars to transmit */
+					(!port->gs.port.tty->stopped)	&&
+					(!port->gs.port.tty->hw_stopped) ){	/* While there are chars to transmit */
 					if (((bufpos+1) & A2232_IOBUFLENMASK) != status->OutTail) { /* If the A2232 buffer is not full */
 						ch = port->gs.xmit_buf[port->gs.xmit_tail];					/* get the next char to transmit */
 						port->gs.xmit_tail = (port->gs.xmit_tail+1) & (SERIAL_XMIT_SIZE-1); /* modulo-addition for the gs.xmit_buf ring-buffer */
@@ -592,8 +592,8 @@ int ch, err, n, p;
 				status->OutHead = bufpos;
 					
 				/* WakeUp if output buffer runs low */
-				if ((port->gs.xmit_cnt <= port->gs.wakeup_chars) && port->gs.tty) {
-					tty_wakeup(port->gs.tty);
+				if ((port->gs.xmit_cnt <= port->gs.wakeup_chars) && port->gs.port.tty) {
+					tty_wakeup(port->gs.port.tty);
 				}
 			} // if the port is used
 		} // for every port on the board
@@ -613,16 +613,16 @@ int ch, err, n, p;
 						struct a2232_port *port = &a2232_ports[n*7+p];
 						port->cd_status = !(ncd & 1); /* ncd&1 <=> CD is now off */
 
-						if (!(port->gs.flags & ASYNC_CHECK_CD))
+						if (!(port->gs.port.flags & ASYNC_CHECK_CD))
 							;	/* Don't report DCD changes */
 						else if (port->cd_status) { // if DCD on: DCD went UP!
 							
 							/* Are we blocking in open?*/
-							wake_up_interruptible(&port->gs.open_wait);
+							wake_up_interruptible(&port->gs.port.open_wait);
 						}
 						else { // if DCD off: DCD went DOWN!
-							if (port->gs.tty)
-								tty_hangup (port->gs.tty);
+							if (port->gs.port.tty)
+								tty_hangup (port->gs.port.tty);
 						}
 						
 					} // if CD changed for this port
@@ -655,8 +655,8 @@ static void a2232_init_portstructs(void)
 #ifdef NEW_WRITE_LOCKING
 		mutex_init(&(port->gs.port_write_mutex));
 #endif
-		init_waitqueue_head(&port->gs.open_wait);
-		init_waitqueue_head(&port->gs.close_wait);
+		init_waitqueue_head(&port->gs.port.open_wait);
+		init_waitqueue_head(&port->gs.port.close_wait);
 	}
 }
 
diff --git a/drivers/char/vme_scc.c b/drivers/char/vme_scc.c
index 69c5afe97f19..1718b3c481db 100644
--- a/drivers/char/vme_scc.c
+++ b/drivers/char/vme_scc.c
@@ -183,8 +183,8 @@ static void scc_init_portstructs(void)
 #ifdef NEW_WRITE_LOCKING
 		port->gs.port_write_mutex = MUTEX;
 #endif
-		init_waitqueue_head(&port->gs.open_wait);
-		init_waitqueue_head(&port->gs.close_wait);
+		init_waitqueue_head(&port->gs.port.open_wait);
+		init_waitqueue_head(&port->gs.port.close_wait);
 	}
 }
 
@@ -422,7 +422,7 @@ static irqreturn_t scc_rx_int(int irq, void *data)
 {
 	unsigned char	ch;
 	struct scc_port *port = data;
-	struct tty_struct *tty = port->gs.tty;
+	struct tty_struct *tty = port->gs.port.tty;
 	SCC_ACCESS_INIT(port);
 
 	ch = SCCread_NB(RX_DATA_REG);
@@ -453,7 +453,7 @@ static irqreturn_t scc_rx_int(int irq, void *data)
 static irqreturn_t scc_spcond_int(int irq, void *data)
 {
 	struct scc_port *port = data;
-	struct tty_struct *tty = port->gs.tty;
+	struct tty_struct *tty = port->gs.port.tty;
 	unsigned char	stat, ch, err;
 	int		int_pending_mask = port->channel == CHANNEL_A ?
 			                   IPR_A_RX : IPR_B_RX;
@@ -500,7 +500,7 @@ static irqreturn_t scc_tx_int(int irq, void *data)
 	struct scc_port *port = data;
 	SCC_ACCESS_INIT(port);
 
-	if (!port->gs.tty) {
+	if (!port->gs.port.tty) {
 		printk(KERN_WARNING "scc_tx_int with NULL tty!\n");
 		SCCmod (INT_AND_DMA_REG, ~IDR_TX_INT_ENAB, 0);
 		SCCwrite(COMMAND_REG, CR_TX_PENDING_RESET);
@@ -512,8 +512,9 @@ static irqreturn_t scc_tx_int(int irq, void *data)
 			SCCwrite(TX_DATA_REG, port->x_char);
 			port->x_char = 0;
 		}
-		else if ((port->gs.xmit_cnt <= 0) || port->gs.tty->stopped ||
-				port->gs.tty->hw_stopped)
+		else if ((port->gs.xmit_cnt <= 0) ||
+			 port->gs.port.tty->stopped ||
+			 port->gs.port.tty->hw_stopped)
 			break;
 		else {
 			SCCwrite(TX_DATA_REG, port->gs.xmit_buf[port->gs.xmit_tail++]);
@@ -522,15 +523,15 @@ static irqreturn_t scc_tx_int(int irq, void *data)
 				break;
 		}
 	}
-	if ((port->gs.xmit_cnt <= 0) || port->gs.tty->stopped ||
-			port->gs.tty->hw_stopped) {
+	if ((port->gs.xmit_cnt <= 0) || port->gs.port.tty->stopped ||
+	    port->gs.port.tty->hw_stopped) {
 		/* disable tx interrupts */
 		SCCmod (INT_AND_DMA_REG, ~IDR_TX_INT_ENAB, 0);
 		SCCwrite(COMMAND_REG, CR_TX_PENDING_RESET);   /* disable tx_int on next tx underrun? */
-		port->gs.flags &= ~GS_TX_INTEN;
+		port->gs.port.flags &= ~GS_TX_INTEN;
 	}
-	if (port->gs.tty && port->gs.xmit_cnt <= port->gs.wakeup_chars)
-		tty_wakeup(port->gs.tty);
+	if (port->gs.port.tty && port->gs.xmit_cnt <= port->gs.wakeup_chars)
+		tty_wakeup(port->gs.port.tty);
 
 	SCCwrite_NB(COMMAND_REG, CR_HIGHEST_IUS_RESET);
 	return IRQ_HANDLED;
@@ -550,14 +551,14 @@ static irqreturn_t scc_stat_int(int irq, void *data)
 
 	if (changed & SR_DCD) {
 		port->c_dcd = !!(sr & SR_DCD);
-		if (!(port->gs.flags & ASYNC_CHECK_CD))
+		if (!(port->gs.port.flags & ASYNC_CHECK_CD))
 			;	/* Don't report DCD changes */
 		else if (port->c_dcd) {
-			wake_up_interruptible(&port->gs.open_wait);
+			wake_up_interruptible(&port->gs.port.open_wait);
 		}
 		else {
-			if (port->gs.tty)
-				tty_hangup (port->gs.tty);
+			if (port->gs.port.tty)
+				tty_hangup (port->gs.port.tty);
 		}
 	}
 	SCCwrite(COMMAND_REG, CR_EXTSTAT_RESET);
@@ -578,7 +579,7 @@ static void scc_disable_tx_interrupts(void *ptr)
 
 	local_irq_save(flags);
 	SCCmod(INT_AND_DMA_REG, ~IDR_TX_INT_ENAB, 0);
-	port->gs.flags &= ~GS_TX_INTEN;
+	port->gs.port.flags &= ~GS_TX_INTEN;
 	local_irq_restore(flags);
 }
 
@@ -636,8 +637,8 @@ static void scc_shutdown_port(void *ptr)
 {
 	struct scc_port *port = ptr;
 
-	port->gs.flags &= ~ GS_ACTIVE;
-	if (port->gs.tty && port->gs.tty->termios->c_cflag & HUPCL) {
+	port->gs.port.flags &= ~ GS_ACTIVE;
+	if (port->gs.port.tty && port->gs.port.tty->termios->c_cflag & HUPCL) {
 		scc_setsignals (port, 0, 0);
 	}
 }
@@ -652,14 +653,14 @@ static int scc_set_real_termios (void *ptr)
 	struct scc_port *port = ptr;
 	SCC_ACCESS_INIT(port);
 
-	if (!port->gs.tty || !port->gs.tty->termios) return 0;
+	if (!port->gs.port.tty || !port->gs.port.tty->termios) return 0;
 
 	channel = port->channel;
 
 	if (channel == CHANNEL_A)
 		return 0;		/* Settings controlled by boot PROM */
 
-	cflag  = port->gs.tty->termios->c_cflag;
+	cflag  = port->gs.port.tty->termios->c_cflag;
 	baud = port->gs.baud;
 	chsize = (cflag & CSIZE) >> 4;
 
@@ -678,9 +679,9 @@ static int scc_set_real_termios (void *ptr)
 	}
 
 	if (cflag & CLOCAL)
-		port->gs.flags &= ~ASYNC_CHECK_CD;
+		port->gs.port.flags &= ~ASYNC_CHECK_CD;
 	else
-		port->gs.flags |= ASYNC_CHECK_CD;
+		port->gs.port.flags |= ASYNC_CHECK_CD;
 
 #ifdef CONFIG_MVME147_SCC
 	if (MACH_IS_MVME147)
@@ -856,7 +857,7 @@ static int scc_open (struct tty_struct * tty, struct file * filp)
 		{ COMMAND_REG, CR_EXTSTAT_RESET },
 	};
 #endif
-	if (!(port->gs.flags & ASYNC_INITIALIZED)) {
+	if (!(port->gs.port.flags & ASYNC_INITIALIZED)) {
 		local_irq_save(flags);
 #if defined(CONFIG_MVME147_SCC) || defined(CONFIG_MVME162_SCC)
 		if (MACH_IS_MVME147 || MACH_IS_MVME16x) {
@@ -880,18 +881,18 @@ static int scc_open (struct tty_struct * tty, struct file * filp)
 	}
 
 	tty->driver_data = port;
-	port->gs.tty = tty;
-	port->gs.count++;
+	port->gs.port.tty = tty;
+	port->gs.port.count++;
 	retval = gs_init_port(&port->gs);
 	if (retval) {
-		port->gs.count--;
+		port->gs.port.count--;
 		return retval;
 	}
-	port->gs.flags |= GS_ACTIVE;
+	port->gs.port.flags |= GS_ACTIVE;
 	retval = gs_block_til_ready(port, filp);
 
 	if (retval) {
-		port->gs.count--;
+		port->gs.port.count--;
 		return retval;
 	}
 

From 852fef69c0d9510a28a70221cfddd004efa02552 Mon Sep 17 00:00:00 2001
From: Oliver Neukum <oliver@neukum.org>
Date: Sat, 26 Jul 2008 22:42:42 +0200
Subject: [PATCH 1171/1171] fix for a memory leak in an error case introduced
 by fix for double free

The fix NULLed a pointer without freeing it.

Signed-off-by: Oliver Neukum <oneukum@suse.de>
Reported-by: Juha Motorsportcom <juha_motorsportcom@luukku.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
 drivers/usb/serial/ipaq.c | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)

diff --git a/drivers/usb/serial/ipaq.c b/drivers/usb/serial/ipaq.c
index 832a5a4f3cb3..cd9a2e138c8b 100644
--- a/drivers/usb/serial/ipaq.c
+++ b/drivers/usb/serial/ipaq.c
@@ -651,15 +651,17 @@ static int ipaq_open(struct tty_struct *tty,
 	 */
 
 	kfree(port->bulk_in_buffer);
-	port->bulk_in_buffer = kmalloc(URBDATA_SIZE, GFP_KERNEL);
-	if (port->bulk_in_buffer == NULL) {
-		port->bulk_out_buffer = NULL; /* prevent double free */
-		goto enomem;
-	}
-
 	kfree(port->bulk_out_buffer);
+	/* make sure the generic serial code knows */
+	port->bulk_out_buffer = NULL;
+
+	port->bulk_in_buffer = kmalloc(URBDATA_SIZE, GFP_KERNEL);
+	if (port->bulk_in_buffer == NULL)
+		goto enomem;
+
 	port->bulk_out_buffer = kmalloc(URBDATA_SIZE, GFP_KERNEL);
 	if (port->bulk_out_buffer == NULL) {
+		/* the buffer is useless, free it */
 		kfree(port->bulk_in_buffer);
 		port->bulk_in_buffer = NULL;
 		goto enomem;